a re-examination and attempted replication of. the
TRANSCRIPT
1
A RE-EXAMINATION AND ATTEMPTED REPLICATION OF. THE STRUCTURE OF CATTELL'S
SIXTEEN PERSONALITY FACTOR QUESTIONNAIRE
STEPHEN FREDERICK BLINKHORN
Ph. D. Institute of Education University of London
2
ABSTRACT
A RE-EXAMINATION AND ATTEMPTED REPLICATION OF THE STRUCTURE OF CATTELL'S SIXTEEN PERSONALITY FACTOR
QUESTIONAIRE
Stephen Frederick Blinkhorn
Cattell's Sixteen Personality Factor Questionnaire is amongst the most widely used of psychometric measures of personality, yet little support for its structure is to be found in the critical literature. Consideration of what might constitute a decisive analysis leads to the view that previous studies have been deficient in concentrating unduly on technical criteria in the conduct of factor analysis to the exclusion of an analysis of psychometric and statistical features which could give rise to observed results.
Using a large sample of undergraduate subjects, both within-scale and between scale analyses were performed. The claim that the scales of the questionnaire are heterogeneous, accepted by Cattell, is shown to be invalid in statistical terms, with a few exceptions. Evidence for a limited but substantial degree of support for the structure proposed by Cattell is adduced from both traditional forms of factor analysis and the newer technique of structural analysis of covariance matrices. In the process doubt is cast on the relevance of certain technical disputes regarding the conduct of factor analysis, and the value of Cronbach's coefficient alpha as an index of psychometric adequacy is supported. The importance of reliability of variables in factor analysis is emphasised, and attention is drawn to the problems associated with the use of heterogeneous samples and ad libitum selection of variables.
Consideration of the results of the analyses suggests a reinterpretation of Cattell's scales in terms of homogeneity of content and possibly a facet analysis of the questionnaire synthesising a variety of viewpoints often presented as conflicting.
3
CONTENTS
Page No.
Abstract 2
List of Tables 4
List of Figures 6
Acknowledgements
Introduction 8
Chapter 1 11
Chapter 2 19
Chapter 3 30
Chapter 4 36 The Treatment of Error 39 Sampling of Subjects 48
Sampling of Variables 50 Choice of Matrix 54 Variable Type 57 Concluding Comments 63
Chapter 5 The Data Base 64
1. airripling and Administration 2. Size and Match of Sample 67 3. Internal Evidence of Quality 69 4. Initial Analyses 75
Chapter 6 80
Chapter 7 Introduction to Factor Analyses 89 Unrestricted Factor Analyses 91
Chapter 8 111
Chapter 9 134
Chapter 10 161
Appendix of Tables 168
References 199
List of the Factor Scales of the 16PF foldout inside
back cover
4
LIST OF TABLES
Table Page No.
1 Numbers of protocols returned 68
2 Percentages of university population of second- 68 year undergraduates and of sample by region and size/type classification
3 Chi-squared tests for fit of sample to population 70
4 Standard deviations of raw scale scores by sex 72 in three standardisation samples, for forms A and B of the 16 PF combined
5 Means and standard deviations of forms A-+B 77
scale scores
6 Indices of internal consistency and reliability for 85
the primary factor scales
7 Factor-pattern elements for the seven factor 92
solution
8 Test statistics for maximum likelihood factor 104
analyses of data for four sex x discipline groups
9 Test statistics for maximum likelihood factoring 105
of the pooled within-groups correlation matrix
10 Factor-pattern elements for the seventeen-factor 106 solution
11 The ten groups and combinations of groups, with 114
sample sizes, for the restricted analyses
12 Estimated factor-pattern coefficients for the ten 118 SIFASP analyses
13 Factor covariance (lower triangle) and correlation 120 (upper triangle) matrix for male arts group
14 Test statistics for the ten SIFASP analyses 123
15 Tucker and Lewis "reliability" coefficients•for 125 the restricted analyses
16 Residual covariances from pooled within-group 127 restricted analysis
17 Unique factor loadings from the ten restricted 130 factor analyses
18 Reordered factor correlation matrix, with 0, Q2 135 and Q4 reversed in sign, for the total group
19 Correlation matrix for the total group after 138 multivariate correction for attenuation
5
LIST OF TABLES
(font' d. )
Table Pale No.
Al Covariance matrix for male arts group 169
AZ Covariance matrix for male science group 171
A3 Covariance matrix for female arts group 173
A4 Covariance matrix for female science group 175
AS Covariance matrix for male total group 177
A6 Covariance matrix for female total group 179
A7 Covariance matrix for arts total group 181
A8 Covariance matrix for science total group 183
A9 Pooled within-group covariance matrix 185
A10 Covariance matrix for total group 187
All Factor covariance (lower triangle) and 189 correlation (upper triangle) matrix for male arts group
Al2 Factor covariance (lower triangle) and 190 correlation (upper triangle) matrix for male science group
Al3 Factor covariance (lower triangle) and 191 correlation (upper triangle) matrix for female arts group
A14 Factor covariance (lower triangle) and 192 correlation (upper triangle) matrix for female science group
Al5 Factor covariance (lower triangle) and 193 correlation (upper triangle) matrix for male total group
A16 Factor covariance (lower triangle) and 194 correlation (upper triangle) matrix for female total group
All Factor covariance (lower triangle) and 195 correlation (upper triangle) matrix for arts total group
A18 Factor covariance (lower triangle) and 196 correlation (upper triangle) matrix for science total group
Al9 Factor covariance (lower triangle) and 197 correlation (upper triangle) matrix for pooled within-group analysis
A20 Factor covariance (lower triangle) and 198 correlation (upper triangle) matrix for total group
6
LIST OF FIGURES
Figure Page No.
1 Eigenvalue plot for the total group 93
2 Eigenvalue plot for the male arts group 96
3 Eigenvalue plot for the male science group 97
4 Eigenvalue plot for the female arts group 98
5 Eigenvalue plot for the female science group 99
7
ACKNOWLEDGEMENTS
This thesis could not have been written without the
cooperation of rather more than 1,000 anonymous individuals
who began undergraduate studies in 1971, and of the careers
services of most of the universities in the United Kingdom.
The staff of the Hatfield Polytechnic Computer Centre showed
remarkable patience and tolerance in allowing me on frequent
occasions to monopolise their facilities and bore them with
idiosyncratic problems. Professor Harvey Goldstein and
Dr. Mike Berger never failed to encourage and, respectively,
served to remind me that psychologists have no monopoly on
statistics, and that there is more to psychology than statistics.
Dr. Peter Saville has been a valued adversary and colleague,
and Dr. Tony Gibson has contributed perhaps more than he will
recognise to the development of the ideas which structured this
thesis. Particular thanks are due to Linda Brassington for
undertaking the typing with remarkable accuracy and despatch.
Finally, my wife Jenny has been truly biblical in her support:
"She openeth her mouth with wisdom; And in her tongue is the law of kindness. "
(Ecclesiastes 31, xxvi)
8
INTRODUCTION
Cattell's Sixteen Personality Factor Questionnaire is
amongst the most widely used of all psychometric techniques for
the assessment of personality, yet its structure has consistently
failed to find support in the results of independent research.
This thesis is an attempt to test that structure directly, and
results from the author's view that the crucial analyses have not
elsewhere been carried out, nor has sufficient consideration been
given to the problems of sample homogeneity and the reliability of
variables in the use of factor analysis on personality test data.
The primary emphasis of the thesis is psychometric and
confined to the questionnaire and its interpretation, and the order
of discussion has been arrived at in an effort to minimise
repetition and the discussion of methodological issues and
empirical results simultaneously. As this order is not entirely
conventional a brief guide to it may help the reader follow the
line of argument.
Chapters 1 and 2 deal respectively with the historical
context of the development of the questionnaire and some of the
issues involved in the use of trait terms in the description of
personality and the explanation of behaviour. Chapter 3 outlines
the commonest criticisms that have been made of the questionnaire
9
and its construction. Chapter 4 is largely concerned with
methodological issues in factor analysis, pointing out the problems
which may arise from sample heterogeneity and unreliability of
variables, and developing the approach to analysis adopted for this
study. Chapter 5 contains an account of the fieldwork. In
Chapter 6 the question of scale homogeneity is discussed in the
light of the general claim that Cattell's scales are notably
heterogeneous: with some exceptions this claim is shown to be
ill-founded. Chapters 7 and 8 contain accounts of a number of
factor analyses, both unrestricted and restricted, which tended
to produce mutually consistent results confirming those of
Chapter 6. Chapter 9 begins with a discussion of certain
features of the factor correlation matrices from the analyses of
Chapter 8, shows that similar results arise from a multivariate
correction for attenuation of the scale inter correlaions, and then
proceeds to an inspection of the item content of certain scales,
followed by a sketch of a possible account of the structure of the
16PF in terms of content analysis of items. Chapter 10 then
summarises and reviews the empirical results.
What is presented is not therefore an account which claims
to be definitive confirmation or disconfirmation of the notional
structure of the questionnaire. Rather it is shown that with a
fairly homogeneous sample a moderately good approximation to
10
that structure can be found, but that the departures from the
notional structure may in themselves be informative and lead to
a positive reappraisal of personality questionnaires.
II
CHAPTER 1
In many respects attempts to construct questionnaire
measures of personality have been strongly influenced by develop-
ments in the measurement of mental ability. By the mid-1930's,
intelligence tests had taken a form substantially similar to that of
those currently in use. Development in this area had been aided
by the emergence of correlational techniques, and perhaps most
notably of factor analysis, which had been applied to the study of
differential aptitudes with some success.
Attempts to pursue mental testing into the area of
personality were perhaps inevitable, and indeed sufficient
development had taken place for Cattell (1933) to publish a review
article, and Guilford and Guilford (1934) to undertake one of the
first factor-analytic studies of extraversion. However, one
difference between ability and personality tests has particular
implications for the application of psychometric methods, in that
in general the former, being tests of maximum performance, may
be expected to correlate with measures of external criterion
performances where there is general agreement as to what
constitutes a better or more correct level of performance. Thus
the construction of the original Binet test could proceed without
any guiding theory as to the fine structure of abilities, and later
12
work could build from the notion of undifferentiated ability to
finer-grained structures.
In the case of personality tests, however, the question of
the dimensionality of personality is more pressing. It is by no
means clear that a single scale, say ranging from inadequate to
adequate personality, has the same kind of utility, even as a
starting point, for the development of measures of personality.
Indeed, articulated accounts of the structure of personality existed
before the development of psychometric measures, most notably
perhaps Galen's typology, adopted also by Kant, and the theories
of psychoanalytically and psychiatrically oriented authors. Given
the lack of clear criterion performances, the development of
personality questionnaires was therefore more dependent on the
theoretical orientation of test authors, and indeed between the two
world wars they proliferated.
Cattell's (1946) initial attempts to formulate a factor-based
theory of personality may be seen as an attempt to bring order
into a confused field in which the fact that two scales shared a
name was no guarantee that they measured related traits.
Furthermore, in taking as his starting-point the list of trait words
extracted from the dictionary by Allport and Odbert (1936), he
implicitly rejected all theoretical approaches other than those
embedded in the language. His stand was to assume that the
13
English language contained terms for all the distinctions that had
been found valuable in the description of personality, and that
appropriate methods of analysis applied to scores on rating scales,
questionnaires and objective performance measures would yield a
widely, or even universally, applicable framework for the
development of a definitive theory of human personality. It is
more than clear that he continues to hold this view, and also the
opinion that his methods of analysis have at any given time been
the most appropriate then available for this purpose (Cattell 1973).
Essentially, then, the only important theoretical
assumption made by Cattell at the outset was that personality is
properly conceived of in terms of traits, although, to anticipate
the discussion in the next chapter, the precise meaning he
attaches to the term is not altogether clear. Such an assumption
was at the very least quite in keeping with the historical context
in which it was made, especially given the apparent success of
Thur stone and the London school, in their different ways, in
giving a dimensional, trait-like account of the structure of mental
abilities. Amongst those who accept this assumption as
reasonable, then as now criticism of Cattell's work has centred on
detailed matters of techniques of analysis and indices of
psychometric adequacy, and it is with these matters that this
thesis is principally concerned.
14
Before passing on to an outline account of the development
and present state of Cattell's theory, there are two points which
give rise to occasional misunderstanding, and are best settled in
advance. First, although Cattell is perhaps best known to
psychologists in general for his work on the development of
personality questionnaires, originally he did not regard them as
having a long-term future as first-rank measures. Recognising
the imperfections of the questionnaire method, such as its
susceptibility to faking, he expected the technique to fall out of use
as more objective measuring devices were developed. Secondly,
there is a matter of terminology: Cattell uses the terms 'primary
factor' and 'source trait' interchangeably. Source traits are
seen as the fundamental variables of personality, discovered by
factor analysis of 'surface traits', which are clusters of observed
behaviour, somewhat analogous to pathological syndromes in
medicine. According to Cattell, source traits cause surface
traits, and bear the burden of explanation insofar as individual
differences in personality are responsible for differences in
behaviour. Thus the term 'primary factor' has ontological
rather than merely operational significance. What is a first-
order factor from the point of view of the conduct of a factor-
analytic investigation is not necessarily a primary factor in this
sense. Cattell has attempted to popularise a distinction between
factor orders (operational significance) and factor strata
15
(ontological significance), but with little success to date.
The systematic account of the 'primary factors of
personality' appearing in Cattell (1946) is based on behaviour
rating scales. Working with a reduced list of trait words drawn
from Allport and Odbert (1936), 100 subjects were rated on 171
bipolar scales, and clusters were identified. On the basis of
this analysis, a smaller set of scales was factor analysed to yield
12 'primary factors' or 'source traits'. Cattell then proceeded
to match these factors with those found in the analysis of
questionnaire responses, and made tentative identifications in
the domain of physiological and performance measures.
Howarth (1976) has presented a critical reanalysis of the
original behaviour rating study, and raises the possibility that
inaccuracies and inadequacies in this study had a detrimental
effect on the eventual published questionnaires Cattell produced.
However the link between this first exposition and the current
versions of the questionnaires is not as rigid as might be
supposed. In particular certain factors, e. g. D, J and K, are
not represented in the standard adult versions, and as
successive revisions of the scales have appeared there has been
a certain amount of reinterpretation of the meanings of factors
as well as claimed psychometric improvements of the scales
which represent them.
16
The original 12 factors proposed by Cattell were labelled
by letters, A to L, in descending order of the proportion of
variance accounted for by each. From the first, Cattell has
preferred oblique factor solutions on the grounds that the real
causal influences he sought to identify by factor analysis were
unlikely to be uncorrelated, even though 'functionally independent'.
Subsequently, (Cattell 1950), he extended his investigation of the
factors to be found in questionnaire response data, to yield the
structure now represented in his Sixteen Personality Factor
Questionnaire (16PF). As has been mentioned, three of the
original factors are missing, but three others have been added
which are claimed to have matches in rating data, and four
additional factors are included which have only been found reliably
in questionnaire data, and hence are labelled Q1 to Q4 .
Since the factors are correlated, factor analysis of their
intercorrelations is possible, and hence second-order factors
may be derived. Four of these Cattell claims to be well-
established, and four or five others tentatively identified
(Cattell, Eber and Tatsuoka 1970). Of these, the first two have
received particular attention, since they appear to align well with
Eysenck's (1947) dimensions of Extraversion and Neuroticism
(Cattell's labels being Exvia and Anxiety). Again, correlated
second-order factors may themselves be factored to yield
17
third-order factors: an example of such an analysis being
presented by Cattell, Eber and Tatsuoka (1970).
For convenience of reference, and to avoid repetitive
descriptions of the interpretation of Cattell's factors, a list of
factor labels, 'technical names' and descriptive adjectives is
bound as a fold-out at the back of this thesis.
Revised verions of the 16 PF appeared successively up to
the definitive 1967/68 editions, since when a seven-factor
supplement has appeared, and downward extensions to high-
school age and below have also been produced. A long series of
publications by Cattell and his collaborators from 1940 to the
present day has detailed the elaboration of his theory: apart
from the improvement of the questionnaires, this research has
extended the scope of Cattell's theory to include an account of
motivation and moods, has pursued factor matching between
different measurement media, has contributed technical advances
in factor-analytic methods, and has attempted to establish a
central role for personality, and for factor analysis, in many
areas of psychology.
Despite the ambitious scope of Cattell's work and the
wide-ranging claims he has made for his theory, critical treat-
ment of his approach has tended to concentrate on the validity of
18
his original claim to have discovered the fundamental variables
of personality and produced a questionnaire which measures
them. Two areas of controversy are particularly relevant to
this attempt to replicate Cattell's structure: the status of
Cattell's concept of trait, and the psychometric criticisms which
have been levelled at the 16 PF, to which questions we now turn.
19
CHAPTER 2
Only rarely has Cattell offered definitions of the word
'trait'. Two are, however, forthcoming in Cattell (1965):
"Trait: a unitary configuration in behaviour such that when one part is present in a certain degree, we can infer that a person will show the other parts in a certain degree." (p. 375)
"By a trait, therefore, we obviously mean some relatively permanent and broad reaction tendency."
(1)- 28)
These scarcely capture the distinctive characteristics of
his use of the term, but more can be gleaned from a careful
reading of his more recent writings, in particular Cattell (1973).
One distinction recurs constantly, that between 'surface
traits', observable consistencies in behaviour, which are or may
be determined by more than one 'source trait', and source traits
themselves, which are not directly observable, but are revealed
by factor analysis of surface traits. Source traits are seen as
the fundamental variables of personality, substantially invariant
across social and cultural groups (Cattell 1973, p. 330ff),
although not necessarily invariant either in terms of their
expression in behaviour or in terms of which items or tests are
most appropriate for their measurement.
20
The overriding importance which Cattell attaches to source
traits is exemplified by his account of the origins of the second-
order Exvia factor, which, as has been mentioned, seems to be
closely comparable with Eysenck's dimension of Extraversion.
Whereas Eysenck (e. g. Eysenck and Eysenck 1969) treats this
factor as the source of lower-order factors, which in his view are
far from reliably identified, and as related to the functioning of the
ascending reticular activating system in the brain stem, for
Cattell it is the result of a complex positive-feedback interaction
amongst source traits mediated by social mechanisms (Cattell
1957), and not a fundamental variable in its own right.
Next, there is the unusually strong emphasis placed by
Cattell on the use of factor analysis in the discovery of source
traits. Source traits are 'factor-dimensions', variations along
which are determined by single unitary influences (Cattell 1965,
p. 374). That is to say, he appears to view factor analysis not
merely as a convenient technical device but as the sine qua non of
research in this area, and he has devoted some effort to the
demonstration of the efficiency of factor analysis in revealing
known structures, e. g. the dimensions and contents of coffee cups
(Cattell and Sullivan 1962). Along with this emphasis has gone an
interest in the development and refinement of factor-analytic
techniques.
21
Likewise, despite hints (Cattell 1952, 1973 p. 22) of
possible nonlinearities in the regression of source traits on overt
behaviour, and of interactions between traits, states and situations,
multiple linear regression is the method of choice for prediction
purposes. Thus a comprehensive attempt to predict overt
behaviour would involve a specification of weights to be applied to
source trait scores, to state variables and to situational variables,
all of which may be classified by factor-analytic means.
Finally, the traits proposed are identified and named
a posteriori. Cattell's initial work was an attempt to define the
universe of discourse by reference to the dictionary, rather than
to build on previous work. Perhaps in consequence, the list of
traits measured by the 16PF reads as a rather eclectic collection
with no guiding theoretical orientation. Some traits, e. g. factor
C, Ego-strength, and factor G, Super-ego strength, have at least
a nominal link with psychodynamic theory; a group of five
(A, E, F, H, Q2) reflect different aspects of extraversion, a notion
with a long and complex history; one, factor I, seems not unlike
William James' notion of tender-mindedness, and others, e.g.
L and N, seem to be more or less peculiar to Cattell. From the
point of view of ontogenetic development the source traits are
equally heterogeneous. Factors A and H, he claims, have a
strong hereditary, physiological basis, whilst at the other extreme
22
factors I and M are more or less exclusively the result of
learning in childhood.
These considerations make an overall assessment of the
conceptual status of the source traits impossible, since all they
have in common is their purported causal status and their discovery
by a particular statistical/mathematical technique. The emphasis,
to borrow the title of Cattell's early book, is on the description and
measurement of personality rather than on a unified account of the
sources and causes of personality differences.
This point has a bearing on one common criticism of trait-
oriented personality theory, succinctly stated by Mischel (1968,
p.42):
"Nothing is explained, however, if the state that we have attributed to the person from his behavior (the has a trait or state of anxiety') is now invoked as the cause of the behavior from which it was inferred. We quickly emerge with the tautology: 'He behaves anxiously because he has a trait of anxiety'. "
Thus Mischel claims that the use of trait names as
explanatory devices is based on a confusion between the
construction (or interpretation) of behaviour and the causes of
behaviour. Insofar as trait names are merely summary terms
for observed and interpreted behaviour, and so long as the only
basis for the adoption of traits is descriptive in this sense, the
23
criticism holds. However, in the longer term the enterprise on
which Cattell embarked is not merely one of efficient description,
although it is at least that. Rather there is an eventual aim of
linking some traits to neural and endocrine events and structures
in such a way as to bring elements of social behaviour into the
same general nomological net as physical events. Given success
in this enterprise the methods yielding traits which do map onto
such physical variables have presumptive utility for the discovery
of traits whose sources are in, for example, child-rearing
practices.
In this respect, Cattell's work is closely modelled on the
natural sciences in which he received his early academic training.
For at various stages in the development of physics, Mischel's
criticism might well have been levelled at concepts such as mass,
valence or magnetic flux, concepts whose value derived at least
for a time from their place in a systematic account of empirical
phenomena, before their basis in the atomic and subatomic
structure of matter had been elaborated.
The extent to which links between Cattell's source traits
and neural and endocrine phenomena have actually been forged is
debatable. Royce (1973) and Fahrenberg (1977) review some of
the common ground occupied by various trait-oriented theorists,
and Cattell and Warburton (1967) present a compendium of
24
physiological and performance tests which, it is claimed, relate
to certain of Cattell's second-order factors. However, for the
moment such developments are at best promising rather than
conclusive.
On the other hand, the converse argument is powerful: if
it can be shown that traits, or some set of traits, cannot be
consistently discovered or derived, then there is little chance of
finding physical phenomena to which they relate, whereas if they
can be replicated then some explanation is called for. Such an
explanation might simply be in terms of response sets, or it might
involve either the original theory or some reworking of it, such as
Gray's (1973) reinterpretation of Eysenck's three personality
dimensions.
In recent years there have been developments in thinking
concerning the sources of a given behaviour in a given situation
converging from several authors which have a bearing on the
interpretation of Cattell's traits. For instance Mischel (1973) in
a modification of his earlier view has argued that what is needed
for adequate prediction of behaviour is information on a wide range
of variables, some of which relate to the behaviour in question only
through the conscious anticipation of the subject as being possible
consequences of his actions. Bern and Allen (1974) in a similar
vein argue for idiographic assumptions about the sources of
25
individual behaviour, even when that behaviour shows consistency
across situations and subjects. Argyle (1976) has moved away
from a notion of personality dominated by the situation to a
structuralist position in which behaviour is generated from a 'deep
structure' of rules, which may at once be idiosyncratic and yet
have elements of commonality across individuals.
Perhaps this move is best summed up by Harr and
Secord (1972) and in Harry's (1976) collection of articles
suggesting a cognitive approach to the understanding of social
behaviour. Essentially it involves a rejection of the 'classical'
model of physical science as applied to the explanation of social
behaviour, in particular insofar as this permits only of inert
entities and efficient causality. Harri and Secord state quite
explicitly (op. cit. p.256) that "A man's nature is a psycho-
physiological mix", the psychological part of the mix involving
such functions as anticipating the consequences of actions, giving
accounts of the sources of past actions and the like. In other
words they propose a formal reintroduction of the Aristotelian
notion of final causality, at least insofar as a person may
represent to himself future states of affairs and attempt to bring
them about, into the explanation of behaviour.
26
As it happens, they cite Cattell's factors C and A as
instances of the kind of powers and complementary liabilities
which satisfy, at least in outline, their criteria for explanatory
notions, and remark especially with regard to factor A that it is
a notable instance of the kind of psychophysiological mix they
regard as desirable in a full account of the sources of social
behaviour.
However, Cattell's own position on this point is either
unclear or simply so eclectic as to permit a variety of inter-
pretations. The definitions of 'trait' quoted at the beginning of
this chapter fall under Alston's (1976) notion of 'T-concept' : we
think of different persons having different degrees of the
disposition, where the degree is a function of the frequency of
response in a representative set of situations along with the
average magnitude of the responses" (p. 67, abbreviations
expanded). Alston contrasts the T -concept' quite sharply with
purposive-cognitive concepts. T-concepts, like co-operative-
ness, sociability, persistence, refer to overt, observable
reaction tendencies, such that a given level of each is associated
with a characteristic frequency and strength of response in a
given situation. Purposive-cognitive concepts, on the other
hand, such as need for social contact, or desire for control over
others, may not give rise to overt behaviour in the form of actual
27
social contact or actual exercise of control over others, yet there
is no contradiction in ascribing these characteristics to
individuals who do not involve themselves in social contact and
who do not exercise control over others.
It is at least possible that Cattell sees some of his source
traits as purposive-cognitive concepts, related causally to
T-concepts, for which read surface traits. Alternatively, one
might class some of his source traits as purposive-cognitive and
others as T-concepts. The distinction is not made explicitly by
Cattell himself, and inspection of his questionnaire items is of
little help. Some items, e. g. "As a teenager, I joined in school
sports: a. occasionally, b. fairly often, c. a great deal", seem
to demand a factual report of the respondent's past or present
behaviour, whilst others, e.g. "I would rather mix with polite
people than with rough, rebellious individuals", seem designed
to elicit a statement of preference between two hypothetical states
of affairs.
The status of questionnaire responses is far from clear:
they may be read as factual statements, including as facts needs,
preferences, values, interests and the like, as attempts to
express a self-concept in terms of item content, or as responses,
in the full-blown behaviourist sense, to stimuli, viz. the items,
whose content will not stand common-sense analysis. None of
28
these approaches is wholly satisfactory, the first because one is
potentially overrating the individual's self-knowledge and access
to accurate information about his own behaviour, the second
because there is as yet little research to show that this in fact is
what subjects think they are doing, and the third because
respondents do seem to believe that they are indeed conveying
information about themselves, and often take pains to be as
accurate as possible in their responses - a phenomenon which is
difficult to quantify but which is very apparent in the course of
supervising a test session.
The question of what actually takes place between the
reading of an item andthe recording of a response is amongst the
least researched aspects of psychometrics; however, for present
purposes we may note the problem and leave it to one side. For
there is a distinction to be made between two stages in the
development of a scientific theory: in the first phase, one collects
as accurately as possible measurements and observations
appropriate to the phenomena under study, in an attempt to
discover and formalise regularities; in the second, theories are
constructed to account for the observed regularities and tested
against them. (Of course, theories may suggest as yet
unnoticed regularities whose presence or absence may result in
modifications to or rejection of a theory, in the Popperian
29
hypothetico-deductive pattern, but here we are concerned with the
elements of theory-building, rather than the logic of scientific
progress).
The classic example of this procedure is the development
of the kinetic theory of gases as an account of the regularities
formalised in Boyle's and Charles' gas laws, these gas laws being
characterised by Harr(and Secord (1972) as a matter of 'critical
description' or 'proto-science' rather than of developed science.
Since a great deal of controversy has been fuelled by the
question of what regularities are to be found in Cattell's personality
scales, however, it seems that there is a need for some careful
critical description to discover just what regularities are present
to require a theoretical account. That is to say, the question of
whether there is a good approximation to Cattell's structure to be
found is prior to the question of why it should be present.
30
CHAPTER 3
The 16PF has received a good deal of attention in terms of
its adequacy as a psychometric device, and in terms of the fit
between questionnaire factors and factors found in rating studies.
We are here concerned exclusively with the first of these.
Three lines of attack have been adopted by Cattell's critics.
Firstly, the value of traits in predicting behaviour has been
questioned. Mischel (1968) reviews a large number of studies
relating trait measures, not specifically Cattell's, to external
criteria, and concludes that their utility has been much exaggerated.
This is a line of argument which will not be pursued here, since the
problem of how best to incorporate trait measures into the
prediction of behaviour would need a thesis to itself, and since
past failures need not imply that the exercise is futile.
Secondly, the 16PF has been specifically criticised on
account of the apparently low homogeneity of the scales it
contains. Levonian (1961), Timm (1968), Eysenck and Eysenck
(1969), Greif (1970) and Howarth, Browne and Marceau (1971)
have taken this line. With attainment and ability tests, one is
accustomed to find that the items in reliable and valid scales
correlate moderately highly, and indices of reliability based on
these correlations have a long history. Cattell has never
31
challenged the empirical results of these authors; instead he has
argued consistently that homogeneity is a red herring in the
construction of personality tests with high transfer reliability
(Cattell and Tsujioka 1964, Cattell 1973). Howarth and Browne
(1971a) attempt a rebuttal of this argument in terms of the require-
ments of simple structure as a criterion for rotation in factor
analysis. However, as will become clear when the results of
homogeneity analyses in the present study are presented, this
controversy is itself something of a red herring. It will be
argued that, with a small number of exceptions, Cattell's scales
meet accepted standards of internal consistency, length for length
being about as homogeneous as Eysenck's own scales in terms of
Cronbach's coefficient alpha (Cronbach 1951).
Thirdly and most controversially of all, the claim has been
made repeatedly that when items and/or scales from the 16PF are
factor analysed, Cattell's structure either cannot be identified, or
is not the most parsimonious solution. Becker (1961), Borgatta
(1962), Timm (1968), Eysenck and Eysenck (1969), Sells, Demaree
and Will (1970, 1971), Howarth and Browne (1971a) and Vagg and
Hammond (1976) all associate themselves with this point of view.
In other words, there is almost total agreement in published
studies arising outside Cattell's circle of collaborators that he is
simply mistaken (Guilford 1975).
32
Two specific procedural points recur: his critics claim
that Cattell consistently rotates more factors than are indicated
by various criteria for the number of common factors present,
and that his use of 'parcels' of items as the basic variables in
his analyses, rather than single items, distorts the solution by
imposing his structure a priori. A third consideration is that
Cattell typically identifies items as being significantly loaded by
a given factor on the basis of unconventionally low factor pattern
coefficients.
The focus of attention throughout this long-standing
controversy is on the replicability of Cattell's primary factors.
There is little dispute about at least two and possibly four or more
second-order factors. Indeed Eysenck (1972) makes the straight-
forward claim that the primary-factor scales of the 16PF yield no
more reliable information than the second order, thus calling into
question one of Cattell's fundamental arguments for the use of
primary factor scales.
In response, Cattell (1972) has depended largely on
technical arguments as to the adequacy of the procedures and
criteria actually used by his critics in conducting their factor
analyses. Cattell (1972), Cattell, Eber and Delhees (1973), and
Burdsal and Vaughn (1974) have also presented factor pattern
matrices purporting to show that the nominal structure of the 16PF
33
is indeed to be found as the outcome of an item-factor analysis.
However, in these studies many items have near-zero loadings on
the factors to which they supposedly contribute, and large factor-
pattern coefficients are few. This Cattell explains as being due
to 'suppressor action' - inclusion of items with low or even
negative loadings on a given factor but relatively high loadings
elsewhere purifies the questionnaire scale by cancelling out the
effects of unwanted variance associated with higher-loading items.
Put otherwise this may simply mean that factor-score coefficients
do not necessarily match factor-pattern coefficients, i. e. the best
set of items for achieving a good estimate of a factor score is not
always that set having the highest loadings. Certainly, as
Guilford (1975) points out, Cattell has never explicitly stated
which items are suppressing what.
On the question of the use of item parcels, Cattell is not
alone. Nunnally (1967) and Comrey (1970) both advocate parcelling
on the grounds that single items are too unreliable a basis for
factor analysis. More detailed consideration of this point follows
in a later chapter, as it is particularly relevant to the methods
adopted in this study.
Overall, the present status of Cattell's scales is rather
confused. There is controversy concerning both the appropriate-
ness of various technical procedure in assessing the adequacy of
34
the scales and factor-analytic studies of them, and the inter-
pretation of published analyses. Little progress has been made
over a period of nearly twenty years towards a rapprochement
between Cattell and his critics, and no agreement has been
reached as to what would count as an experimentum crucis.
Meanwhile in commercial terms the 16PF has thrived alongside
the products of other laboratories, and currently vies with
Eysenck's questionnaires as the most widely used personality
questionnaire in the United Kingdom, ranking about fifth in the
USA. Whatever the psychometric arguments that have appeared
so far, there is clearly a body of users who actually find the
scales useful or are woefully blind to their inadequacies.
This thesis is therefore aimed at reconciling three points
of view, Cattell's, his critics' and his customers'. It takes a
fresh look at the controversy surrounding the 16PF from the point
of view of the prospective user of the test, asking to what extent
Cattell's structure can be replicated, and what steps are
necessary to yield such a replication, in the hope that in the process
light will be shed on the sources of dispute. The procedures
adopted take as their fundamental variables the scales of the
published test, since if the purported structure cannot be found at
this level, there is little chance of it being found at the level of
items (and even if it were, there would be little point in using
the 16PF in its present form).
35
The next chapter therefore considers at length the problems
involved in attempts to replicate an hypothesised factor model with
a view to establishing the utility of the methods to be adopted.
36
CHAPTER 4
The use of factor analysis in personality research has not,
in general, been directed simply at a reduction in the number of
variables to be considered in a given data set. Rather it has been
used as a strategic tool, with the aim of identifying replicable
factors which will serve as fundamental elements in a rounded
theory of personality. A great part of the controversy concerning
rival accounts of the structure of personality has been attributed to
differences in the technical procedures adopted by various
theorists. In particular, three issues can be identified as
important in the eyes of Cattell:
1. Method of decomposing the matrix to be analysed
(principal components, principal factors, image
factors, alpha factors, maximum likelihood etc.)
along with corresponding methods of estimating
communalities.
2. Basis of choice for number of factors to retain
and rotate.
3. Method of rotation and criteria for acceptance of
a solution as representing 'simple structure'.
In point of fact, none of these issues is capable of resolution by
strict proof in the context in which it arises, since the funda-
mental question being asked is not "which solution gives the best
reproduction of the original data?" but rather "which solution is
37
most useful in the elaboration of a theory of personality?"
Nonetheless, they are debated at length in a highly proprietary
brand of disputation which is generally inaccessible to the non-
specialist.
Rather less thought appears to have been given to certain
other issues, notably:
4. Whether items, subscales ('parcels') or scales
should be the unit of analysis?
5. What kind of prior scaling of variables is most
appropriate for a given analysis?
6. How variables should be chosen, in terms both of
their number and of what is already known of
their interrelations.
And finally almost no discussion at all is to be found as to:
7. What the desirable characteristics of the sample
providing the data base might be.
8. Whether influences other than the personality
variables nominally under investigation may be
reflected in the data base.
It is the contention of the present author that these issues
should be tackled in reverse order, that is to say in the order,
more or less, in which they arise in the practical business of
choosing a sample, testing participants and analysing their
scores.
38
From the point of view of the present work, factor analysis
is seen as a means of confirming or disconfirming a proposed
solution or model, itself derived by factor analysis. Thus we are
not concerned with the question of whether the common-factor
model is an appropriate aid in developing a personality theory,
rather we shall be investigating the fit of a particular structure
within the scope of that model.
The primary problem with factor analysis as a research
technique is its indeterminacy. When it is used as an exploratory
tool in the development of an hypothesis, this indeterminacy need
not be a great handicap. In attempts to test the adequacy or
replicability of a factor theory, however, it is a major problem
(Pawlik 1973). To make matters worse, in factoring personality
tests we are dealing with highly imperfect variables. Even with
perfectly reliable variables, there would still be indeterminacy
arising from the lack of definition of relevant common-factor
variance, the lack of a guaranteed foolproof method of rotation,
and sampling error in the matrix to be analysed. With unreliable
variables a further source of variation enters.
Thus we now turn to a discussion of the problem of error,
then to questions of sampling of both subjects and variables, and
finally to the matter of the choice of matrix to analyse and the
choice of items or scales as the units of analysis.
39
THE TREATMENT OF ERROR
The common-factor model involves assumptions concerning
the structure of error variance in a correlation or covariance
matrix which are of particular importance in the present context.
Briefly stated, common-factor analysis decomposes an m x m
matrix into k common factors, where k is less than m, and m
'unique', 'specific' or 'error' factors ('error' in the sense of
making no contribution to the common-factor structure, rather
than implying unreliability). On the assumption that the unique
factors are orthogonal to each other and to the common factors,
the analysis then yields common factors which are formally
uncontaminated by measurement error.
In vector notation, the common-factor model may be
expressed thus:
X = p. + +
where X is the vector of observed scores for a given individual
is the constant vector
A is the m x k matrix of factor pattern coefficients
▪ is the vector, of length k, of common-factor scores for that individual
▪ is the vector, of length m, of specific factor scores
This formulation is general, in that no modification is required to
take account of correlations amongst the common factors, i. e. the
40
factor-pattern coefficients are partial regression coefficients in
the above equations. On the assumption that all specific factors
are mutually uncorrelated, and uncorrelated with all the common
factors, the correlations amongst the observed variables are
accounted for entirely in terms of .,,- .
In matrix notation, the variance-covariance matrix of the
observed variables, denoted here by E , may be expressed as:
Z = A ON “.12
where A is again the matrix of factor-pattern coefficients or loadings
co is the variance-covariance matrix of .,
q)2 is the diagonal matrix of specific factor variances, or error variances (Lord and Novick 1968 p. 533)
Thus in terms of the formalisation of the common-factor
model, the only term which may be called error relates to the
uncorrelated parts of observed variables. There are two
components of the specific factor variances. The first is errors
of measurement, i. e. the unreliability of observed scores. The
second is the extent to which the reliable parts of observed
variables are only partly to be accounted for by the common
factors in a given factor model. For instance, when a real
influence on a given variable is represented only in scores on
that variable it will not of course appear in the common-factor
41
structure, even though the contribution it makes to the variance of
that variable is consistent, and may be identified in a different
analysis involving more or different variables.
Thus, when it comes to the interpretation of an obtained
factor solution, even supposing (which is unlikely) there to be no
non-zero off-diagonal residuals, one cannot without further
information unpack the specific factor variances into measurement
error and reliable variances associated with single variables. A
further problem arises from certain inadequacies in the kinds of
measures psychologists typically use. Interest usually centres
on factors representing influences in some domain of theoretical
interest, in the present case personality. However, what are
nominally personality tests may well reflect variations in
intelligence, or social class, or test response sets. Thus a factor
analysis of personality tests may result in a structure which
accurately describes the relationships among the tests, but not
among the variables with which the investigator is primarily
concerned, i. e. the personality traits, only imperfectly
represented in test scores.
This kind of error, error with respect to some substantive
theoretical model, may arise from any of three sources. In the
first place, a trait relevant to the domain under investigation may
be reflected in scores on tests intended to measure other traits.
42
Thus anxiety might influence scores on measures of sociability -
in other words the sociability tests are to this extent inadequate -
with the result that the factor pattern matrix shows a loading of
the anxiety factor on the sociability tests. In the second place,
unwanted influences may affect scores on two or more measures,
in which case they may be identified, and rotated into separate
factors, eliminated by careful test construction, or deliberately
measured and thus accounted for. Attempts to deal with
acquiescence and social desirability response sets and the like
(Edwards 1957) illustrate this problem. Finally, variables which
are not formally represented in the tests to be factored, but which
can be measured by other means, may affect the obtained factor
structure, in particular through their influence on the mean
vectors of test variables.
To a great extent personality tests are composed of items
which may be expected to vary in significance across social and
cultural groups. That is to say, the same response to the same
item may signify different levels of the same trait, or even a
different trait altogether, in different groups. If the aim of
personality testing were merely to outline dimensions of (reported)
social behaviour, rather than such sources of differing behaviour
as may be ascribed to consistencies in the individual, this point
would be irrelevant, but we have it from Cattell (1973) that his
43
scales are intended to measure traits with characteristically
different expressions in different social, cultural and national
groups.
Thus one must ask to what extent, for instance,
differences in mean scale scores between identifiable social
groups represent real differences in personality rather than a
tendency for one group to respond more in the higher-scoring
direction on account of the cultural loading of specific items.
The consequences of spurious differences generated in this way
will be to reduce correlations between variables when single
scales are affected, and inflate them when both scales are
involved. Thus within-group correlations may vary considerably
from correlations computed over a total sample, and the more
pervasive the influence of the moderator variable, the greater the
distortion resulting.
Note that this problem is not identifiable with elements of
either of the formulations of the common-factor model presented
earlier. The formal model is concerned with accounting for the
observed variates, not the underlying theoretically important
variables. What is being suggested here is that parallel to the
effects of unreliability in increasing the size of the specific-
factor loadings is an effect of low validity on the common-factor
loadings. Specificially, when there are pervasive moderator
44
effects, one would expect increased values for correlations
computed over the total sample when the algebraic product of the
signs of the moderator-induced shift in mean scores is the same
as the sign of the within-group correlation. Depending on the
extent of the moderator effect, this may well result in either a
reduction in the number of factors retained for rotation, when the
criterion for retention is based on the size of the eigenvalues, or
a spurious increase in correlations between common factors in an
oblique solution, or the appearance of factors which are not readily
interpretable in terms of the variables nominally entered into the
analysis.
Since traditional methods of factor analysis do not recognise
these effects as error (and nor are they from the point of view of
the formal model), from the point of view of the personality
theorist distorted solutions are likely to result.
Given the heuristic nature of exploratory factor analysis as
typically used, there is little that can be done about error arising
from contamination of one test by factors hypothetically relevant
only to others, apart from attempts to purify measures by what
Cattell has called 'progressive rectification', i. e. improving the
tests in the light of successive studies - which can be something of
a circular activity. Where the result has been that structure
remains in the residual matrix, one should doubt one's criterion
45
for the retention of factors for rotation, but this is only likely to
be acceptable when nominally equivalent measures have
appreciable residual covariance. (Inspection of residuals has,
of course, been automated in the maximum likelihood and minres
methods of analysis).
The correction of moderator-induced error is rather more
feasible: likely moderators may be partialled out in advance of
factor analysis, or analysis may be performed on homogeneous
groups selected on appropriate measures of the hypothesised
moderator. Simply entering likely moderators into the analysis
will not do in general, however, since any criterion for the
number of factors to extract which is based on eigenvalues may
be too severe in consequence, the more so the greater the effect
of the moderator.
To sum up, the problem of error when using factor
analysis as an aid in theory building and testing is not exhausted
by considerations of reliability. Factor analysis deals not in
variables which have a place in some theoretical scheme of
things, but in correlations or covariances between observed
scores in a sample of variates which purport to reflect
theoretically important variables, hence one must consider
whether there may be influences affecting the validity of testsused
for the sample involved.
46
Thus far we have considered errors in factor analysis
arising from the imperfect reliability and validity of measures, a
topic which has received relatively little consideration in
published studies, although occasionally (e. g. Pawlik 1973,
Cattell 1966 p. 112) theoretical writers have devoted some space
to them. Such error has usually been lumped together with
sampling error, and no distinction has been made for purposes
of reporting the 'proportion of variance accounted for' by the
factors retained for rotation. In general, however, factor
analysts envisage the possibility of inference from the analysis
of covariance matrices derived from sample data to some larger
and usually ill-defined population. Thus limitations on the method
may be imposed by the size and representativeness of the sample,
both of which will affect the adequacy of the sample matrix as an
estimate of the population matrix, the first by way of the standard
error of the elements, the second in cases of unintentional
selection on the variables included.
Note that the second of these is not identical with the
question of moderator effects discussed earlier: moderator
effects reduce to the intrusion of unwanted variables which should
be eliminated; unintentional selection may result in a poor
estimate of the population matrix, but by way of deficient sampling
of subjects, possibly leading to restriction of range, rather than
47
from different scaling properties of the measures within the
sample.
Questions of sample adequacy can only be settled outside
the computational devices of factor analysis; the matter of the
effects of sample size on the procedure is internal to it, and
attention has been focussed on this aspect with the emergence of
maximum-likelihood methods. The smaller the sample size, the
fewer common factors are needed for a statistically acceptable
solution. That is to say, the goodness-of-fit test used may not,
with small samples, be powerful enough to reject the hypothesis
of a number of factors which, from the personality theorist's
point of view, is smaller than optimum. Conversely, with a large
sample many more factors may be indicated by the significance
test associated with maximum likelihood procedures than are
interpretable, or than would have been retained using other methods
and criteria (see esp. Kaiser 1976), particularly when large
numbers of variables are involved. Prior to the introduction of
maximum likelihood methods, the adequacy of the size of a sample
was judged by reference to rule of thumb, 'five times as many
subjects as variables' being one common criterion. However, the
apparent rigour of the chi-squared test for number of factors does
not really bear on the matter of the number of relevant factors from
the point of view of the investigator, since when error due to poor
48
test validity is present in the matrix a sufficiently large sample
size will ensure that the test will suggest a poor fit even when all
factors of theoretical interest have been extracted.
A factor analysis of personality scales may reasonably be
undertaken for either of two purposes: a) to assess the adequacy
of or aid in the development of a factor theory of personality; or
b) to investigate the adequacy of the scales representing such a
theory, given that it is accepted. One must distinguish between a
good theory poorly represented psychometrically and a bad theory
pure and simple. Bad tests may support a theoretical model, even
though they are quite unsuited for application to real-world problems.
SAMPLING OF SUBJECTS
There seems to be no general agreement as to what
constitutes an appropriate sample for the purposes of conducting
a factor-analytic investigation - indeed there appears to be very
little discussion of the topic at all. Most textbooks of factor
analysis simply ignore the matter, although Fruchter (1954)
devotes a brief paragraph to the intrusion of unwanted variance
due to deficient sampling.
Cattell (1966a) states that " ... preliminary taxonomic
work is necessary to make clear when one is sampling within
species and when across them," but continues by suggesting that
49
this presents no problem, since we "know what we are doing and
do not put men and monkeys in the same sample" (p. 113).
Eysenck (1966) takes a firmer line: "Ultimately what is
being asserted here is that correlations between variables a and b
may differ significantly in size and direction when total groups are
subdivided according to typological principles, and correlations run
over these subgroups." In the absence of such typological analyses
"correlational analysis ... is liable to gross errors ... " Yet
when it comes to one of the most often cited of his own factor
analytic studies (Eysenck and Eysenck 1969), not only is the sample
not divided, except by sex, on typological grounds; not only is no
typological analysis presented to justify the procedures adopted;
but (p. 182) apologetic remarks are made as to the inadequacy of
the subject pool as representative of the population of the country
as a whole.
Nonetheless, in principle at least, both seem to agree that
homogeneity of subjects with regard to taxonomic considerations
outside the personality domain is desirable. But at this point the
problematic nature of the concept of personality intrudes. For
the interaction between taxonomic groupings and test scores is in
a sense what personality measurement is about. Trivially,
supposing there to be extraverts and introverts as exlusively
defined categories, one seeks an interaction between test scores
50
and group membership as the basis of measurement.
Less trivially, females typically show higher average
scores than males on measures of anxiety and neuroticism. This
may of course be attributable to a real difference between the
sexes in levels of anxiety, but on the other hand may result from
a greater willingness on the part of females to admit to minor
symptoms which in fact they suffer no more than males. In the
first case the factor analyst would be justified in not analysing
results separately for the two sexes, in the second he should split
by sex. In general, given adequate sample size, there is nothing
to be lost and possibly something to be gained by splitting a sample
on taxonomic grounds or by seeking homogeneity on external
criteria, in terms of achieving a better fit to the assumptions of
the common-factor model.
SAMPLING OF VARIABLES
On the matter of sampling of variables, one couldhardly do
better than to quote Guilford (1975):
"There must be a favorable pattern or combination. of experimental variables. Each factor must be equally well represented in terms of the number of variables per factor, and each set must be equally strong with respect to the amount of variance. These require-ments for selection are important because in connection with rotation of axes, there is a general principle that strong (better represented) factors tend to rob weak ones of their variances."
51
Such requirements assume a good deal of a priori
information about the variables relevant to a given domain, and
since factor analysis has been used mainly to generate
hypotheses rather than test them, Guilford's argument appears
to suggest that in the absence of a fortunate combination of
circumstances most factor analyses will be inconclusive.
Guilford's point may be taken even further, for not only
may weak factors be robbed of their variances in rotation, they
may not be included amongst the factors retained for rotation
if better-represented factors account for the greater part of the
variance in the matrix, a problem which will be particularly
relevant when the criterion for retention of factors is based on
the absolute sizes of eigenvalues.
This question is of importance for a critical examination
of studies which include sets of scales from different authors in
the same analysis, e. g. Eysenck and Eysenck (1969) and Sells,
Demaree and Wills (1970). Both studies are concerned, not to
establish regression weights for one set of scales onto the others,
but to determine which author's scales give the 'best' picture, if
any. Both studies used principal components analysis, and had
they rotated all components there would have been little of
interest in these analyses. (This follows from the fact that
all positive definite matrices of a given order form a group
52
under non-singular transformation, which is itself simply a
reflection of the fact that such matrices have an inverse. Thus
the full set of components may always be rotated to a specified
pattern, with consequences only for the matrix of correlations
between rotated components.)
However, since in each case only the largest components
were retained, one must ask what characteristics would lead to
the representation of variables in the components rotated.
Eysenck and Eysenck in their joint factorial study of the Guilford,
Cattell and Eysenck scales extracted the first twenty principal
components of correlation matrices for each sex, made up as
follows:
From Eysenck's laboratory:
10 subscales representing Extraversion and 96 items Neuroticism
2 lie scales 18 items
From Cattell's laboratory:
15 short scales, each representing one source trait 99 items
From Guilford's laboratory:
13 short scales, each representing one primary 109 items factor
In addition three acquiescence scales were scored, one
from the items contributed by each of the three authors. On the
not unreasonable assumption that the various scales best represent,
53
length for length, their authors' thinking, we thus have for the
hypothesised personality factors:
48 each for the two Eysenck factors, in two groups of five scales.
6-7 each for the Cattell factors in fifteen scales
8-9 each for the Guilford factors in thirteen scales
Now consider three points: 1. the reliability of a test tends
to increase with length; 2. correlations are attenuated as a
function of decreasing reliability of the variables; 3. there is
general agreement that Cattell's and Guilford's scales relate to
Eysenck's as lower-order to higher order factors, hence Eysenck's
factors represent common ground. It is hardly surprising that of
the ten components extracted, after rotation the first two carried
the largest portion of variance and were "unambiguously identified
as neuroticism and extraversion". Simply on the grounds that
Eysenck's factors are very overrepresented it is clear that the
conclusion drawn, that Cattell's factors are not well replicated,
is more a comment on the procedures adopted than on the adequacy
of Cattell's model. One would need a great deal of faith in the
capacity of factor analysis to reveal the true structure of
personality traits regardless of the configuration of input
variables to accept this analysis as a reasonable test of the
adequacy of Cattell's primary factor scales.
54
This study is perhaps the clearest example of how sampling
of variables can predetermine the outcome of an analysis, and
illustrate how factor analysis favours areas of overlap between
different systems at the expense of factors which may be
represented in only one set of scales. It is likely, though more
difficult to show clearly, that the same effect is present in other
studies relating different sets of scales and items, such as Sells,
Demaree and Wills (1970) and Howarth and Browne (1971b).
CHOICE OF MATRIX
Traditionally, factor analysis has been conducted on
correlation matrix. Rarely are grounds for the choice
specifically of correlations forthcoming from authors who use the
technique, but in principle there is at least one other appropriate
index, namely covariance. Perhaps the strongest reason for
using correlations rather than covariances is that in general the
units of measurement of psychometric tests are arbitrary, so that
for a single sample study neither raw score means nor raw score
variances have any absolute significance. Hence the removal of
mean and variance effects which is implicit in the calculation of
product-moment correlations seems to be a logical choice in the
absence of strong indications to the contrary. That is to say,
whilst there are no particular grounds for setting all variables to
the same scale, psychometricians rarely have enough confidence
55
to choose scaling factors for each variable separately, hence the
use of correlations is a straightforward admission of ignorance.
Statements as to the proportion of variance extracted in an analysis
have therefore to be read as referring to weighted raw score
variance and covariance, rather than variance in personality or
whatever.
However, from time to time studies, e. g. Eysenck and
Eysenck (1969), appear in which attempts are made to relate
results from the factor analysis of two separate groups, often sex
groups, and factor-pattern comparisons are made across samples
and studies. Leaving aside the question of different rotational
principles and methods which may lead to apparent lack of
factorial invariance, the exclusion of mean and variance effects
results in different metrics for the variables in different analyses.
In confirmatory factor analysis one is concerned to discover
whether a given factor solution, in terms of high and low loadings
in the factor pattern matrix, is replicable on a different sample
of variables and/or subjects. In neither case may one safely
assume that factor variances will be identical across studies, or
between groups within a study. Consequently in the process of
selecting the strongest factors on the basis of the size of their
associated eigenvalues and then rotating them blindly there is a
risk that underextraction of factors may lead to quite different
56
solutions for different samples, or that particular criteria for
rotation may result in different positions for the axes where a
common pattern, at least in terms of which factors have high and
low loadings on which variables, could be established.
As a first step away from the totally arbitrary allocation
of unit variance to each variable, the use of a common scale for
the same variables in each sample followed by the factoring of
variance-coveriance matrices seems prudent. This does not
solve the overall problem of the relative size of variances
between variables in the sample yielding the common scale, but
is at least some improvement over the blind use of correlations.
The common scale for each variable might be obtained by
standardisation of variances in the pooled sample, and then
proceeding to analyse within-group coveriance matrices. There
are doubtless other possible approaches to this problem, and as
with so many other aspects of psychometric methods a strict
resolution of the problem will only come, if at all, with the
establishment of non-arbitrary scaling factors for particular tests.
Cattell (1944) tackles the problem of comparing factor
solutions across groups and samples, suggesting "parallel
proportional profiles" as the criterion for an acceptable solution
when more than one sample is involved. In effect, the results
57
of adopting his approach are similar to the method just outlined,
so long as an appropriate number of factors is retained for
rotation. Eysenck and Eysenck (1969) adopted Cattell's
suggestion as to the criterion for acceptability, and claimed poor
matches between factors in their samples of males and females.
However, they did not transform their factors to be as similar as
possible before attempting to match. Otherwise the critics of
Cattell's model have been notable for their omission of any
consideration of the problem.
VARIABLE TYPE
Among Cattell's practices in his use of factor analysis one
of the most criticised is his tactic of analysing correlations
between parcels, that is to say subscales of small groups of
homogeneous items, rather than between single items. Howarth
(1976) among others has claimed that in effect this rigs a factor
analysis to achieve the investigator's desired results. Eysenck
(personal communication) has suggested that a definitive
resolution of the debate between himself and Cattell must stem
from the factor analysis of items. Comrey (1970) and Nunnally
(1967) express themselves strongly against the use of factor
analysis of items, largely on the grounds that items are highly
unreliable, whilst Cattell (1972) and Burdsal and Vaughn (1974)
58
imply that parcel factoring is both equivalent in terms of results
and simply more convenient from a practical point of view. A
further consideration, rarely if ever mentioned in the psychometric
literature, is that items such as Cattell's, which have only three
score categories, do not satisfy the linear model and hence
without rescaling are not appropriate variables for factor analysis.
Serious arguments against item factoring tend to concentrate
on the intrinsic unreliability of single items. The effect of test
length on reliability is too well known to need extensive discussion;
quite simply a single-item test will be very considerably less
reliable than a test composed of a dozen or more similar items.
Unreliability of variables in a factor analysis will tend (a) to
reduce the absolute size of off-diagonal elements in a covariance
matrix (the attenuation effect); (b) to increase the amount of
correlated error through the intrusion of unwanted, specific
covariance common to only two or three items; and consequently
(c) to make rotation of the resultant factor-pattern both more
difficult and less clear in outcome, since common-factor loadings
on items will tend to be small, making the identification of
salients problematic.
A further complication arises when factoring large numbers
of items, in particular when their unreliability is taken into account.
As a rule of thumb, a sample size of from five to ten times the
59
number of variables is recommended so as to counteract the
accumulation of sampling error in the off-diagonal elements of
the matrix. Given the 368 scored items in forms A and B of the
16PF, a sample size for a single analysis of well over 1500 is
indicated. Attempts to circumvent this particular problem, and
the associated problem of arranging computer facilities to handle
a matrix of the required size, have included the analysis of
'marker' items, that is to say items which are thought on the basis
of previous analyses to be particularly efficient at delineating the
structure of the factor space. Such a strategy, however, involves
a trade-off: smaller samples are needed, but the consequent
reduction in the number of hypothesised salients for each factor
may make rotation difficult. A priori, there are only slight
grounds for assuming that the best marker items for an American
general population sample will also be best for a British under-
graduate sample.
For the purposes of confirmatory analysis, it is hard to
see why Cattell's critics have not attempted to analyse scale scores
on two or more forms of the 16PF. There is after all a clear
hypothesis implicit in the publication of the test, viz. that named
scales in each form will be loaded by corresponding factors, and
not by others. Since the rotational indeterminacy of factor
analysis, exacerbated by the low loadings characteristic of item
60
factoring, seems to preclude any agreement between Cattell and
his critics at the item level, analysing scales offers certain
distinct advantages:
1. The rank of a reduced (by insertion of communalities in
the diagonal) covariance matrix based on scales will be a lower
bound to the rank of a matrix based on the items combined in the
scales. That is, since scale scores are the simple sums of
mutually exclusive sets of item scores, at least as many factors
should be found in an item factor analysis as in a scale factor
analysis.
2. Given the higher reliability of scale scores, minor
covariances between small numbers of items resulting from
specific cultural or social influence local to the sample will be
reduced in their influence (e. g. "I can always enjoy myself at
a gay party" and "I am always straight in my dealings with my
friends" are hypothetical items with local double meanings).
3. If the hypothesised model holds, factor salients will be
larger, and hence rotation facilitated.
4. The argument that items which do not have large loadings
on factors corresponding to the scales on which they are scored
are nonetheless contributing to the purity of measurement by
suppression of unwanted covariance between other items in the
scale and other factors becomes irrelevant.
61
In other words, by analysing at the level of scale scores, the
question as to whether Cattell's model is reasonably represented
in his scales can be settled. If it does hold, then it may be worth
investigating why, against evidence from other researchers, it
does so. If it does not hold, then the way is open to develop
notions of what the structure may be at the item level.
Thus the analysis of scale scores is convenient in several
respects, requiring smaller samples, improving the chances of
clear rotation, reducing demands on computing facilities (and
hence allowing the use of more sophisticated techniques), and
yielding the general advantage of dealing with more reliable
variables.
Finally, against the use of item factoring, there is the
possibility that scales consisting of items with different difficulty
levels will suggest more or different factors in an item factor
analysis as a consequence of their difficulties. Guttman's (1954)
simplex model is but one example of how a matrix of correlations
between measures of a single attribute, each measure differing
from the others in terms of its difficulty or complexity, may have
factors which fail to reflect the underlying unidimensionality.
Lord and Novick (1968), Pawlik (1973) and Levy (1973) all point
out this difficulty in equating the dimensionality of the factor space
62
with the dimensionality of the latent space. Whilst in practice
they may be uncommon, it is possible to envisage scales whose
items differ in terms only of their difficulty. The item
inter correlation matrix for a perfect scale of this type would have
values decreasing progressively with distance from the diagonal.
Such a matrix will normally be of full rank, and will yield
principal components following an oscillatory pattern, and whilst
the scale may be unidimensional in the sense that all items are
validly measuring the same attribute, the least and most difficult
items may have near-zero correlation, be loaded by different
factors in a factor analysis and give the impression of a hetero-
geneous scale. The compensatory model of scale construction
is thus not the only possible model implicit in personality tests:
conjunctive, disjunctive and mixed models may equally claim
consideration.
Cattell is not explicit on this point, nor, given the
unreliability of items, is it a straightforward matter to decide
which model is implicit in his scales, or the extent to which they
are mixed if that be the case. However the possibility of a
progressive ordering of difficulty among the items of a scale is
yet another counterindication for the use of factor analysis of
items as a basis for evaluating the goodness of fit of Cattell's
model.
CONCLUDING COMMENTS
The considerations outlined in this chapter were not
developed in isolation from the practical problems involved in the
design and conduct of the research project to which we shall turn
shortly. Given the almost unanimously critical research
literature, the author's expectations at the outset were that
Cattell was likely to come out of a large-scale study rather badly.
To avoid the possibility of predetermined negative results,
especially given the acknowledged indeterminacy of factor analysis,
it seemed prudent to allow Cattell's hypothesised model the benefit
of the doubt and seek sources of contradictory findings other than
in the technicalities of the arbitrary decisions which have to be
made. In the event these were not hard to find. The author has
no favourite criteria or techniques, rather his aim was to discover
which techniques if any supported Cattell's model. A further
gesture in the same direction is the use of an undergraduate subject
pool. Cattell, Eber and Delhees (1973) state that most of the
development work on the 16PF was done on samples of around 200,
with a strong undergraduate or young adult representation. This
study involves a sample five times as large, and the factor analyses
reported were conducted with the specific aim of attempting to
identify Cattell's model, rather than attempting to reject it.
63
64
CHAPTER 5
THE DATA BASE:
1. Sampling and Administration
The collection of the data on which this study is based has
been described in Saville and Blinkhorn (1976). Since perhaps the
most intensive use of Cattell's scales in this country is in higher
education, in research, careers and personal counselling and by
employers of graduates in recruitment procedures, there was, at
the time the project was begun, a need for representative under-
graduate norms in Britain. A literature search revealed that no
systematic sampling of undergraduates for this purpose had been
attempted even in the USA. Saville (1973) had used random
location sampling techniques with repeated follow-up to collect
general population standardisation data for the 16PF, but despite
the undoubted success of this method it had proved expensive,
involving commercial sponsorship and a large grant from a charit-
able trust to cover the services of the British Market Research
Bureau.
In addition to the cost involved, ues of the same method for
sampling of undergraduates was impractical, largely due to the
difficulty of determining the place of residence of individuals in
the target population. Some compromise between random and
65
volunteer sampling had to be found, and eventually the cooperation
of the careers advisory services of almost all the universities in
the UK was enlisted, polytechnics and other colleges being
excluded for lack of detailed statistics of undergraduate enrolment.
Based on the latest statistics then available (1969-70 entry)
each university was asked early in 1973 to provide a sample
proportionate to its annual undergraduate intake. At the time,
students' sensitivity concerning the disclosure of personal
information led to the universities being unwilling to release lists
of names for random selection, hence selection of participants was
delegated to the careers services and could not be supervised.
Where possible the careers services arranged for random lists of
names to be generated by computer; elsewhere every n'th name
was taken from the list of registered students in their second year
of undergraduate study. Each participant was written to
individually and invited to take part.
In practice, the sample size obtained from each university
was determined by a number of factors. In some cases the
careers service was prepared to write repeatedly to the same
individuals; in others a larger sample than requested was invited
to participate and no follow-up was attempted. In the
circumstances, such matters had to be left to the discretion of
the collaborators. One university (Manchester) had no difficulty
66
in recruiting most of its original sample without follow-up, and had
to turn away uninvited volunteers, whilst at the other extreme
Sussex had a very poor response. In an attempt to offset these
differences in response rate, the universities had been stratified
by location, using arbitrarily defined regions of the UK, and by a
hybrid size/type classification (London, Oxbridge, provincial large,
provincial medium and provincial small). Shortfall in one
university was partially remedied by persuading fieldworkers in
universities in the same stratum to increase their sample.
The sample consists entirely of undergraduates in their
second (i. e. not necessarily penultimate) year of study, tested
between March and June 1973. Second year students were chosen
as being less under examination pressure than finalists, more
accultured to university life than freshmen, and likely in any case
to be about to make contact with the careers service. The sample
was randomly divided in two within each university, half the students
completing forms A and B of the 16PF, half forms C and D. All
students also completed the APU Occupational Interest Guide
(Advanced Version), and those taking the shorter forms C and D
of the 16PF also took two short Cattell scales (IPAT Anxiety Scale
and the Neuroticism Scale Questionnaire), plus form A of the
Eysenck Personality Inventory. A further balancing was introduced
by varying the administration order of forms of the 16PF, so that
67
half of each subsample completed them in the order B-A (D-C)
rather than A-B (C-D). This was achieved by manipulation of the
page order in the specially printed booklets before stitching. The
original print of the 1968 Anglicised versions of the 16PF was
reproduced lithographically in the booklets used: answer sheets
were dispensed with by instructing participants to ring their
answers in the booklet - a procedure which speeds completion and
eliminates transcription errors. Reports from the fieldworkers
suggested that about 90 minutes had been adequate time for the
majority of the participants, a figure suggested on the basis of a
small pilot study conducted at Glamorgan Polytechnic.
2. Size and match of sample
The sample quotas negotiated with each university were
aimed at providing a minimum total sample of 2,000, split equally
by sex and across pairs of forms of the 16PF; the print run
allowed for a maximum total sample of 3,000. Table 1 gives the
actual numbers of returned protocols in each subgroup. The total
of 2584 represents approximately 3% of the total undergraduate
intake in British universities for the year in question, although
clearly and designedly females are overrepresented. For the
purposes of the analyses reported later, only complete protocols,
including all biographical information, were included, to ensure
68
Forms A and B
Order A-B B-A
Forms C and D
C-D D-C
Male 309 304 305 300
Female 339 339 3)44 344
Total 648 643 649 644
Total A+B 1291 Total C+D 1293
Grand total 2584
Table 1: Numbers of protocols returned
% Pop.
Region
Males %
Sample
8
5
10
9
9
Females
% % Pop. Sample
7 9
6 6
12 9
12 11
4 5
% Pop
6
5
13
13
6
Total %
Sample
8
6
10
10
7
Wales 6
W. England 5 S.E. England 14
London 12
E.Anglia 7
Midlands 11 11 12 11 11 11
N.England 26 25 26 28 26 26
Scotland 15 17 17 15 16 16
N.Ireland 4 6 4 6 4 6
Size and type
Large Prov 32 33 37 34 33 33
Med. Prov. 29 31 29 31 29 31
Small Prov 18 21 18 22 18 21
London University 8 11 8 12 8 12
Oxbridge 10 9 4 6 8 8
Table 2: Percentages of university population of second-
year undergraduates and of sample by region
and size/type classification
69
that all analyses were based on identical samples. This reduced
the sample for forms A and B of the 16PF to 545 males, 599 females
and hence a total of 1144, a move which had no appreciable effect
either on raw score distributions or on the match of the sample to
published statistics of university entrants for the intake from which
the same was drawn (University Grants Committee 1975). Table 2
gives percentage figures for population and sample classified
according to the sampling frame mentioned above for the forms
A and B sample.
Counting Oxford, Cambridge and London universities as one
each and the colleges of the University of Wales as separate
institutions, forty-three universities participated. Of the remainder,
three refused, one did not reply and three were prevented from taking
part as a result of their materials being strikebound at Heathrow
airport. Table 3 gives the results of chi-squared goodness of fit
tests between the sample and population figures of Table 2; these
show no evidence of a significant mismatch in distribution of
respondents across these (admittedly arbitrary) sampling
categories.
3. Internal evidence of quality
All protocols were inspected individually to identify those
with missing data and to ensure that incorrectly completed documents
were not processed. In addition each script was checked for signs of
70
(8 by region
.2
by size/type
R
degrees of freedom)
,.2 A—
(4 degrees of freedom)
,-)(...2
Males 4.53 >0.80 1.85 >0.70
Females 3.12 70.90 3.60 >0.30
Total 3.41 >0.90 2.00 >0.70
Critical value for significance at 0.05 level 15.51 9. 49
Table 3: Chi-squared tests for fit of sample to population
71
blatant frivolity. In the event, only a dozen or so scripts were
rejected, all but one because certain identifying detail, e. g. sex,
had not been recorded - the exception being one respondent who had
systematically added extra response categories to certain test
items. As mentioned above, other scripts were later rejected for
present purposes due to incompleteness in other respects.
Since participation in the fieldwork was voluntary, there
was a possibility that the project would attract respondents of a
particular persoia.lity type, e. g. more anxious, more dependent
or whatever. This might be expected to result in restriction of
range of scores on one or more of the personality scales. Since
population statistics are not available the crucial comparison is of
course not possible. However, comparison of scale variances
from this sample with those from Saville's general population
sample and from the American college student standardisation
sample suggests that if such an effect is present it is shared with
other studies, in the case of Saville's with a study employing
superior sampling techniques. Table 4 gives the standard
deviations of scale scores for forms A and B combined for the
three studies, divided by sex. With the notable exception of
factor B (intelligence), there is an overall tendency for the standard
deviations of scores for the present sample to be at least as large
as for the other samples. Thus it does not seem that the problems
72
This study 0
0 -1
.. H co cd al 0 to is -P a) ;-1 1-1 rd -r-I 04 a) H F4 a) o
ci P-1 fl 8 6..3
-*a This study
Factor Males Females
A 6.92 5.85 6.92 6.58 5.02 6.04
B 2.94 3.47 2.94 2.75 3.44 2.94
C 7.60 6.99 7.19 7.11 6.67 7.10
E 7.58 7.30 7.11 7.72 6.69 7.30
F 9.39 9.23 8.28 8.56 8.61 7.96
G 6.69 5.90 '6.25 6.51 5.46 5.99
H 11.95 10.30 10.96 11.39 10.36 11.07
I 6.87 5.76 6.46 5.89 4.58 5.18
5.40 4.95 5.19 5.30 4.94 5.22
M 6.54 5.63 6.79 6.09 5.92 6.49
N 4.52 4.80 4.26 4.65 4.71 4.32
0 8.69 7.79 7.91 7.94 7.58 7.67
Q1 5.58 4.97 5.26 5.91 4.71 5.11
Q2 6.22 5.76 6.26 6.08 5.59 5.98
Q3
6.69 6.00 5.6o 6.58 5.73 5.96
Q4
9.59 9.20 8.63 9.32 8.16 8.65
Table 4: Standard deviations of raw scale scores by sex
in three standardisation samples, for forms A
and B of the 16-IF combined.
73
involved in recruiting the sample in this case had any consequences
for the range of scores on individual scales that might lead one to
suppose that self-selection had been a more serious problem than
elsewhere.
In the circumstances, willing cooperation of participants
was perhaps more important than obtaining a perfectly random
sample by obliging selected individuals to take part in the project,
which might have led to poor quality data simply because responses
are entirely within the control of the individual. Had random
responding taken place on a large scale in protest against enforced
participation, the value of the exercise would have been considerably
reduced.
The only explicit internal checks on the quality of the data
were in the lie scale of the Eysenck Personality Inventory, and the
Motivational Distortion Scales of forms C and D of the 16PF (i. e
in tests which were not administered to the half of the sample with
which this thesis is principally concerned). These scales were
designed to detect deliberate distortion of responses. The mean
scale scores for this study were almost exactly identical with those
from Eysenck's standardisation in the case of the EPI, and about
half a standard deviation lower than those from Cattell's college
student standardisation of the 16PF. It is debatable to what extent
these scales should be taken seriously as indicating level of faking,
74
since the items in question can often be identified by the suspicious
student. However, where the present author actually administered
the questionnaires, the only suspicions voiced about the detection of
faking and the like were when some participants suspected that
items were repeated verbatim as a check on consistency (which is
not the case).
The author's own observations of participants, in sessions
at Brunel University and several of the London colleges, left him
with the impression that there was considerable enthusiasm for the
project amongst those who actually took part. Each of these
sessions was, at the request of participants, followed by an
extended discussion of the aims of the project in relation to the
problems of assessing non-cognitive individual differences.
Reports from fieldworkers in other universities corroborated the
impression that motivation and interest was high.
No doubt the enthusiasm was in part due to the promise of
feedback of scores through the careers services, demand for which
continued throughout the following year. In practice, all
participants' scores on the occupational interest guide were made
available to those who requested them, and scores on the
personality tests were sent where an appropriately qualified
psychologist was available to interpret them. Anonymity was
maintained by keeping records of participants names against index
numbers only in the offices of the careers services.
4. Initial analyses
Saville and Blinkhorn (1976) report a large number of
univariate and bivariate statistical analyses, with the aim of
clarifying the degree and nature of variation within the samples.
Relevant to this thesis in particular are the following findings.
(1) Sex differences in mean scale scores on the 16PF for
several factors are both statistically significant and in some cases
large in absolute terms. Furthermore there is overall con-
sistency, with only minor exceptions, both across forms and in
comparison with corresponding findings in the general population.
(Saville and Blinkhorn 1976, tables 3. 1 to 3. 6).
(2) Mean scale score differences are statistically significant
also in several cases across discipline categories, as coded in
terms of the nine groupings used by the University Grants
Committee. In particular there are large differences on Factor I,
which also shows large sex differences. These differences
remain when discipline groups are divided by sex. The size of
these differences had not been anticipated - indeed one researcher
with considerable experience in dealing with large undergraduate
samples advised that there was little point in even collecting
discipline information, since in his experience no groupings had
ever been found to yield significant differences. In the event,
75
76
discipline differences appeared to be about as large as sex
differences overall, although mostly they were concentrated in
the division between Arts and Social Science students on the one
hand and Science and Engineering students on the other, rather
than between groups within these categories. In comparison with
the approximately comparable age group in Saville's general
population sample (mean age 20, range 16-24 years, of whom about
10% were university students) the undergraduates (mean age 20,
96% being in the range 17-24 years) show a complex pattern of
differences. Table 5 gives means and standard deviations for the
first-order scales of forms A and B combined of the 16PF for the
undergraduate sample divided by sex and discipline, and the general
population young adult group divided by sex. On some of the
scales undergraduates as a whole tend to differ in the same direction
from their general population coevals, whereas on others the young
adult mean scores are in between undergraduate discipline groups.
(3) Reliability, as estimated by alternate form correlations,
was generally higher than reported by Cattell, Eber and Tatsuoka
(1970), but in general not impressive. For forms A and B
coefficients ranged from .23 to .81, with median approximately
. 58. Fuller details and a comparison with other estimates of
reliability are given in the next Chapter.
77
ril a) a)
cd
• a)
a) 0 0
1-1
E c13 r- (1) a g 0 0 cd 0 rd a) 0 7:1
*z r=4
A mean 21.12 15.57 21.39 18.19 18.04 21.88
s.d. 6.57 6.20 6.23 6.74 5.70 5.16
B mean 18.00 18.40 18.62 18.56 15.02 14.68
s.d. 3.04 2.84 2.70 2.84 3.02 3.27
C mean 30.19 31.83 28.95 31-17 30.48 27.82
s.d. 8.04 7.14 7.13 6.85 7.13 6.65
E mean 28.28 26.44 22.23 21.30 26.77 20.87
s.d. 7.28 7.70 7.87 7.39 7.35 6.50
F mean 31.06 28.56 30.24 29.93 32.67 31-07
s.d. 9.83 8.93 8.39 8.90 8.43 8.18
G mean 21.51 21.71 21.63 22.94 21.07 21.98
s.d. 6.72 6.68 6.44 6.59 6.14 5.47
H mean 26.58 22.82 22.81 22.66 28.56 23.14
s.d. 12.26 11.45 11.46 11.29 10.21 10.49
I mean 23.35 18.10 28.64 23.22 17.02 24.14
s.d. 6.37 6.39 5.01 5.86 6.11 4.49
L mean 18.72 16.96 15.96 15.02 18.89 17.29
s.d. 5.28 5.39 5.41 5.04 5.32 5.11
M mean 28.36 26.90 28.10 25.93 23.32 21.34
s.d. 6.63 6.41 6.09 5.86 5.83 6.04
N mean 17.34 17.82 18.86 18.54 18.78 20.69
s.d. 4.85 4.24 4.69 4.59 4.63 4.66
0 mean 21.34 20.91 26.17 25.04 20.47 26.34
s.d. 8.92 8.49 8.03 7.73 8.32 7.09
Qi mean 22.32 20.99 20.27 19.29 21-57 18-13
s.d. 5.61 5.49 6.03 5.6o 5.03 4.81
Q2 mean 20.18 20.92 20.09 19.33 18.85 18.97
s.d. 6.61 5.89 6.14 5.92 5.84 5.40
23 mean 21.37 22.82 19.98 21.50 21.57 20.02
s.d. 6.97 6.42 6.77 6.09 6.09 5.67
Q4 mean 25.11 23.67 29.25 27.85 25.54 30.71
s.d. 9.98 9.28 9.41 9.10 9.24 7.38
Table 5: Means and standard deviations of forms A-f-B scale scores
78
(4) Factors C, 0, and Q4 on all forms and for all groups had
inter correlations approximately equal to their reliabilities, and
factors L and N for some groups and forms had correlations with
other scales equal to or greater than their reliabilities,
suggesting that their factorial independence was seriously in doubt.
(5) Eysenck's Extraversion and Neuroticism scales, followed
by Cattell's scales representing factors B, I and Q1 were
successively partialled out of a correlation matrix including the
two Eysenck scales and the scales of forms C and D of the 16PF,
calculated both for the total sample and for the sexes separately.
Results indicated that Eysenck's (1972) claim that the reliable
variance in Cattell's first order scales is almost totally
accounted for by Neuroticism and Extraversion is exaggerated,
and suggested a minimum of seven factors to account for these
scales.
The results summarised above form the background and
source of the present study. As far as can be ascertained, the
discovery of large discipline differences in personality scale
scores has been made elsewhere only by Barton and Cattell (1972)
in a smaller study conducted in a single college. Certainly there
appears to be a tacit assumption underlying both the conduct of
psychological research on undergraduates and criticism of such
79
research that undergraduates form a relatively homogeneous group
selected from a heterogeneous population. In the case of the
personality scales under discussion, this does not appear to be the
case. It may well be that in the course of development of the 16PF,
the use of preponderately undergraduate subjects resulted in its
being peculiarly applicable to undergraduates, but on the present
evidence it seems only prudent, in the light of the discussion in the
last chapter, to be prepared to treat sex x discipline groups
separately for the purposes of factor analysis.
The monograph containing the various initial analyses
(Saville and Blinkhorn. 1976) deliberately did not set out to tackle in
detail the points of difference between Cattell and his critics,
however certain of the results suggested that the problems of the
reliability, homogeneity and factor structure of Cattell's scales had
been to some extent exaggerated in the critical literature. The
treatment of the question of homogeneity in particular seemed
inadequate and yet there are simple widely used procedures for
arriving at some kind of an agreed conclusion, which topic provides
the material for the next Chapter.
80
CHAPTER 6
Several authors have criticised the 16PF on account of the
apparently low homogeneity of the scales it contains. The use of
homogeneity indices as a measure of the reliability of a test has,
of course, become well established in the development of measures
of attainment and ability. Commonly split-half coefficients,
estimates from the Kuder-Richardson formulae and Cronbach's
coefficient alpha (Cronbach 1951) are the indices adopted.
Perhaps the two most influential papers tackling this
question are by Levonian (1961) and Howarth, Browne and Marceau
(1971). Neither reports any overall homogeneity index for the
scales of the 16PF, rather inter-item correlations are presented
in summary form, the criterion used in each case being the
statistical significance of correlations between items within and
across scales. Howarth et al. report that of 3,267 significant
correlations, only 348 were intra-factor correlations. Further-
more, on average an item correlated significantly with 1.89 items
within its own (10- 13 item) scale, and with 15.86 items of the
remaining 171-174. Levonian's results were similar, although
as his sample was considerably smaller, absolute numbers of
significant correlations were correspondingly reduced.
81
Presentation of results in this form is highly misleading:
let us consider the same data in another form. A single form of
the 16PF has 184 scored items, implying a total of 16,836 unique
off-diagonal entries in a correlation matrix, of which 984 are
intra-factor and 15,852 extra-factor. Thus we may rewrite the
Howarth et al. results as follows: rather more than one in three
of all intra-factor correlations are significant whilst only about
one in five of all extra-factor correlations meet the criterion, and
on average an item correlates significantly with one in five of the
items in its own scale, but with one in eleven of the items outside
its scale. Given that the scales are designed to be correlated, it
is of course hardly surprising that items in different scales should
themselves be correlated.
It is in any case not at all clear that the approach adopted
by Levonian and Howarth et al. is particularly appropriate to the
problem. In Chapter 4 it was pointed out that items validly
measuring the same attribute might have low intercorrelations on
account of differences in their 'difficulty' levels leading to
different univariate and bivariate score distributions. Since the
product moment correlation coefficient essentially measures the
linear component of the relationship, it is to be expected that items
in a scale will show varying intercorrelations partly on account of
their difficulty levels.
82
For instance, items 76 and 176 in form A of the 16PF
correlate, in the total group for this study, 0.043. Univariate
score distributions for these items, expressed as percentages,
are as follows:
Score 0 1 2
Item 76 73.01 6.67 20.32
Item 176 17.67 19.44 62.89
Despite the apparently low intercorrelation of the items,
correlations between them and a composite formed of the
remaining 19 items in factor A for forms A and B combined, equally
weighted, are 0.394 for item 76 and 0.230 for item 176, very much
the order of item-partial scale correlation one expects in a fairly
homogeneous scale.
Since the instructions for the test suggest use of the middle
'uncertain' response category for no more than one item in five,
one would expect item score distributions to tend to be bimodal.
In fact, although overall this instruction appears to have been
followed, 'uncertain' responses vary widely from item to item,
from as few as 2% to over 50%, with the result that item score
distributions vary enormously. The effects of this are not
handled with any great sophistication by Levonian, who simply
combines middle response categories with the less frequently
chosen extreme response, or by Howarth, Browne and Marceau,
83
who report only the distribution of frequency of use of the middle
response without drawing attention to possible implications for
item intercorrelations.
Cattell, in response to critics on the matter of homogeneity,
has consistently agreed with them that his scales are heterogeneous,
but equally consistently has argued that this is a desirable property
(e. g. Tsujioka and Cattell 1964, Cattell 1972, Cattell 1973). Items
which have low intercorrelations but which correlate well with the
factor to be measured will, he claims, yield scales which are
factor-true, in that their error variances will be a combination of
unique variances and small components of many unwanted factors,
rather than large components of a few.
However, actual indices of scale homogeneity are almost
impossible to find in the literature where Cattell's scales are
concerned. Cattell confines himself to presenting samples of
interitem correlations, together with multiple correlations
between items and factors in a factor analysis (e. g. Cattell 1972).
In the case of the multiple correlations it is not clear whether
these are based on equal item weighting, or on the usual
maximisation procedure for multiple correlation (in which case
the inference from items and their validities to scales and their
relationship to factors is not necessarily reasonable). It seems
likely that the latter is the method adopted, since such multiple
84
correlations are easily obtainable from the matrices used in
computing factor scores, and the presetting of weights to be equal
is sufficiently unusual to warrant specific mention. If so, then
these multiple correlations probably represent an overestimate of
the validity of the questionnaire scales, which are scored by
applying equal integer weights to items.
Cattell adopts (Cattell, Eber and Tatsuoka 1970) the notion
of 'structured homogeneity', by which is meant that scales should
be internally heterogeneous, but correlate highly with alternate
forms and with the factor to be measured. Such scales are, he
suggests, the most useful and generalisable given likely variations
in item characteristics across social and cultural groups. Thus
one might expect to find homogeneity coefficients for each of the
scales of the 16PF rather lower than the corresponding alternate
form reliabilities.
Coefficient alpha was therefore computed for each of forms
A and B, and for forms A and B combined, for each sex and for the
total group. Since variations across sex groups were very small,
Table 6 presents these results for the whole group, together with
the corresponding alternate form correlations and values for
coefficient alpha taken from Burdsal and Vaughn (1974). These
last are, to the best of the present author's knowledge, the only
published values for the homogeneity of the 16PF scales.
Cro
nbac
h's
Alp
ha
0
co
eff
icie
nts
A
ltern
ate
C
ronb
ac
h's
A
lpha
7:1
C14-6
•
E-I
Cn • v-I
rd
0 M 7711
gq
.10
to
cn • r•I
,ct4 C-
P:1
CD
0
0
0
0
P:1
0
N if
0 0
'70 •0 cn
• - 0 0
a, .0 In cn
0 0
0 714 N
0 0
N 0
in .14 •
0 0
ce) •7t4 N N CO
0 0 0
'70 ON 0 U1 tf1 N
• • • 0 0 0
co cn '0 t11 Ln
0 0 0
C, .0 ,0 If1 Ln
0 0 0
N N IC) .0 '0
• • • 0 0 0
0 N C'-
0 0
0 co
• • 0 0
.0 Ln Ln co
0 0
CO 7i4 CO
0 0
ON rg
Lf) CO
. • 0 0
'70 N
0
0 .0
• 0
N 'C
0
to
0
<NI
• 0
0
en •
0
N •cti
0
•cti
0
•14
• 0
rg L(
0
cn N .
0
N .cti
0
N d
0
O N
• 0
C'
0
• 0
CO 0
0
N
0
N N .
0
N
0
If)
. 0
,ct, in
0
0
CO Ifl .
0
N
0
N
• C
M eel
0
NI
0
.0 di
• 0
re)
0
CO Tt,
• 0
N
0
0 in
0
0 7r1
• 0
N 0
CO
• 0
M
0
.0 U1
0
0 Lc)
• 0
CO 0
0
• 0
N
0
(NI N
0
di N
• 0
a) 7:1U
1
Cr) ).1 0 4-a U
F-1 073
;-4 1:14 a)
-6a
0 4-4
›.• 4.4 • f--1 • I-I
Cd
a)
a) .-4 cr)
0
;-4
4.4 0 cn
85
N <12:1 (.)4zi01--4 4 zoactaa
O 0 U
86
Clearly, these results lend no support to Cattell's claim
for 'structured homogeneity'. On the contrary, the close
correspondence between alternate form and homogeneity
coefficients suggests that items have not been combined into scales
on any principle of balanced heterogeneity, but rather that the
scales are collections of items from moderately homogeneous
domains. Since coefficient alpha may be interpreted as the
average of all possible split-half correlations, corrected for
test length, there is really nothing to choose between alpha and
alternate form correlations as a measure of scale reliability.
Lest it be thought that the close correspondence of these
two sets of indices results from cultural homogeneity in the sample
of undergraduates on which they are based, Saville (personal
communication) has found equally close correspondence between
them for his general population sample using the same scales.
Variations in the value of alpha for different forms, as, e. g.
in the cases of factors B, M and 0, are traceable to specific items
having negative or very low positive correlations with the rest of the
items formed into a composite, i. e. item-partial scale correlations.
Whilst these results fail to support Cattell's specific claim
for structured homogeneity, however, nor do they show the
disastrous lack of homogeneity that Levonian and Howarth, Browne
87
and Marceau seem to suggest. Coefficients alpha for the
Extraversion and Neuroticism scales of the Eysenck Personality
Inventory form A, based on the half of the sample which completed
forms C and D of the 16PF, which is in all respects comparable
with the forms A and B sample, are respectively 0.77 and 0.84.
Since the EPI has 24 items per scale, these figures are comparable
with the 20-26 item scales of forms A and B combined in the 16PF,
that is to say with the last column in Table 6. On this basis,
Cattell's scales for factors A, C, E, F, G, H, I, 0, Q3 and Q4 are of
approximately the same order of homogeneity as Eysenck's scales,
with Q1 and Q2 not far adrift.
In the context of ability and attainment tests, Nunnally
(1967) states that about 20 items are typically needed to reach a
value for coefficient alpha of about 0.8. Although several of
Cattell's scales fall below this level, even for two-form length, it
would appear that the discrepancy is not nearly as great as both
Cattell and his critics would have us believe. Dotnain-sampling
theory (Nunnally 1967) suggests that the 'domain validity' for a
scale should approach the square-root of coefficient alpha. In
terms of factor analysis, this may be translated into an expected
value of the factor-structure coefficient for any scale and the factor
it represents, i. e. its factor validity. Comparison of the square-
roots of the coefficients in Table 6 with factor validities quoted in
88
Tables 5.4 and 5.5 of Cattell, Eber and Tatsuoka (1970) supports
this interpretation: domain validities approach, but are in general
somewhat lower than, Cattell's claimed factor validities.
Of course within-scale statistics reveal nothing of the
dimensionality of the 16PF as a whole. It is quite conceivable that
several of the scales are drawn from the same domain of items.
However as a check on the adequacy of a factor analysis, co-
efficient alpha provides an independent source of information. As
will become clear, the results of the factor analyses which follow
bear out the value of internal consistency as a measure of scale
adequacy in the case of Cattell's scales.
89
CHAPTER 7
INTRODUCTION TO FACTOR ANALYSES
The remaining Chapters are concerned with a description
and interpretation of factor analyses conducted on forms A and B
of the 16PF. Most of these analyses are summarised rather than
described in detail, for two reasons: first, many of the analyses
gave interpretatively similar or identical results, and second
presentation of the tables associated with all the analyses would
add greatly to the bulk, but not the substance, of the thesis.
A number of strategic decisions were made in advance, in
the light of the considerations discussed in Chapter 4. Firstly, it
was decided to work with scale scores rather than items; apart
from the technical advantages of this decision already discussed,
the problems of factoring a square matrix of order 368 by any but
the crudest methods were insuperable from a data-processing
point of view. Secondly, the sex and discipline differences
mentioned in Chaper 5 seemed large enough to demand at least
some analyses based on within-group or pooled within-group
matrices. The question of variable and factor variances was
settled by standardising all scores on a common metric, viz.
n-sten normed scores, with a mean of 5.5 and a nominal standard
deviation of 2, which had been calculated on the present sample,
90
imposing not only a common metric but also as close an approxima-
tion to a normal distribution as is possible. For convenience in
interpretation the scale was further changed to a mean of 0 and
standard deviation of 1, again on the total group.
Thirdly, the choice of computational methods used was
deliberately eclectic. A variety of factoring methods, involving
different definitions of and means of estimating communalities,
and different rotational procedures, some automatic, some under
human control, were adopted by way of exploratory analysis, in
an order partly determined by availability of computer software
and time at any given moment. The utility of such an exploratory
sequence was seen as in either the partial or total replication of
Cattell's model by means of traditional factoring methods as a
prelude to the use of more expensive confirmatory methods, or
alternatively in the absence of such replication, the identification
of a likely alternative model for testing.
Certain desirable features of analysis proved not to be
possible for lack of appropriate software. Most notably, off the
shelf programs for unrestricted analysis of covariance matrices
and Harris-Kaiser orthoblique rotation could not be implemented
successfully within available resources, but in the event their lack
did not detract from the emergence of a relatively stable and highly
interpretable solution, linking Cattell's model to previous psycho-
91
metric studies, the homogeneity analysis reported in Chapter 6
and even psychobiological notions concerning the roots of
individual differences in personality.
UNRESTRICTED FACTOR ANALYSES
Perhaps the most obvious and most widely used first step
in factor analysis is quite simply to correlate all variables in the
total sample, iterate for communalities and rotate principal axes
corresponding to principal components associated with eigenvalues
greater than 1. This was accordingly done for the 32 scales in
forms A and B of the 16PF. Seven factors were rotated using
direct oblimin with the delta parameter set at zero. The resultant
factor pattern matrix is presented in Table 7, and'a plot of eigen-
values in Figure 1. This solution is rather close to Cattell's
second-order factor model:
Factor 1 is clearly very close to the second-order Anxiety
factor, with major loadings on scales C, L, 0, Q3, Q4.
Factors 2 and 3 together involve most of the scales
contributing to the second-order Exvia factor, A, E, F, H
and Q2 , with an additional loading on N.
Factor 4 bears a strong resemblance to the second order
'prodigal subjectivity' factor, with loadings on I, M and
possibly L.
Factor 5 loads G and Q3, i. e. the second order superego
factor.
, 1,3 L.) U ) .1
92
0 • 1-4
/-I 0
co N Lfi ^O +- N 'IT e4 CO N ro Co co m 0, 'AD CO N m 'AD 4' C4 C4 r. CO PI N VD 0
LID V, C'- CA -40 N) N) CN N N) Co hD +- Co lca VD OD M 04, CO ..cr Co N .qr N +- CA M
IN Co Co Co v Co Co Co m Co Co N Co m 14-71 Co Co Co C4 lr N LID VD +- • • . • • • • • • • • • • • •••••••••••••••• u.
MA Co tn M Mt M tai Co Co Co Co Co Co Ma Co Co Leto Co nmtslmmmmmmmmmm tt-
! 1 1 I I I ! ! 1111
1'0 147I -.r. ..ci -0 -- -ot ro ...1. "0 cl- N ro Co -- "a' N) 0, N) 0, -0 h, OD ..r %!' -0 VI V, -0 Co CA Lil 0.4
PO N) ro ..4o CO Co h, Co Co PIO PO h7 h0 -.0 c.4 -- Co ,r V7 h., M h0 CA PO V7 .- CA la' +- LO C4 CA D.
-40 SA '0 M Co Co Co Co M +- Co Co Co Co tO Co tO Co 47 Co Co +- tO Co o Co Co M M +- Cl Mt Mt CI • ••••• ••••••• • • • • • • •• • •••••••• ••• 13I
M M tO MI M Mt Cr tO Co Co to to MI to tar Co tO Co to to to to to Ma MA to to M M to n Co 1 I I I 1 1 1 I 1 ! I I I I I 1 I 1 a,
-17
to r.... O. CA 00 -0 CO -.1- -./- r, -4o n to to CO h., No 0, ,- 0, ,r 0, to in 1.,7 to m 0, ro c4 C to 4, N M LIO tai CO 0, r... to -- IN c, -0 to -- N 1.17 M 4- -.0 4.-- ....- ,r 01 ...- Lo r, 40 ,r r-, c.4 0, ,r
L7 LTA to SA to +- N. .,.. M M +- SA Co CA in LO M SA to MA M CA NO +- +- ,-- M +- Co +- M4 V7 +- S- • • • • • • • • • • • a • a • • a a • • • • • • • • •••••• 0
to Co to M to to M Co Co to MI M M to MI Co MA to to to 'Si to tn Co Sa M to SA Co n MA Co 4-
1 I I 1 1 1 1 ! I I 1 1111 1 1 ul
CA +- .4" 'r to -- N '‘i- OD N) V7 ro co OD NO CN LID LID N m tO cr CO 4-- +- M 'C LID CO CO VD LID -t- '
4Cr ...... IN Co .1. SO N N. ..0 CO to +- CO CO CID ..- Kt 'sr 1", to h, +4" LI h-, to -47 CO to ,0 -Lr -a to 6=
..qr to i to m m to M NO MI m tO Co Co Co to Co -- Co 'Si MI to +- M NO N LID to +- +- M 'Si Co al • ••••• ■ ••• ••••• • •• a • I. •• ••••••••• E
MI 'Si to 'Si tar M to M M MI '0 tar 'Si to 'Si Co Co to MA to 'Si M Co 'Si 'Si Co 'Si tO M to MA Co w S. I 1 1 I 1 1 ! 1 1 ! 1 ! 1 1 1 ! 1 r-I O al
..r., !Jo -o -- CA N. sr.) -4- ...r -- r, -- -o "0 1'0 .cr Co cr. cr. ..- -- CA Ul LID N 01". 0, CA N 0, *AD M +- O) N +- M MA M VA V, to cl CO r, n -o to ro ro in m CO +- 0, CO LO sO h0 0, h, hl -0 •Kt- -4- .Lr
m N1 V, Co M +- CA M +- CA M sn +.- to to h7 M M h, M4 M +- +- M +- M MA CA +- M M PO M M f- 11 • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • 0.1
MI M Co n M M M Co '0 MI M M tO tO Co to to to Co M 'Si to M to 'Si M Co to Co 'Si Si tO +1 I 1 1 I ! 1 I 1 I 1 I I
61
-ro IN CO CO +- 07 +- 'slr +4' LO +- -ID h, NO 1.0 CA M !Kr CO PO Lil CO +- to ro cr co -.1:1 -a LID M III Ci
141 M CA N 00 CA CA h, 4.-- hI7 +- CA +- 1.0 ...o cq sO N) ro co ..1- M-- r!...1 n CA -sr '4" CA sO +- hD I
04 isl L., ...- In ...0 I a N M N Ma .4- +- SA h7 'Si 'Si M MA 'Si CA -O '0 -0 to M N h0 +- to -.1- to -- r- • ••••••••••• ••• • •••• • •••••••• • •• 0
Co to 'Si MI Co Co M MA '0 Co to MI tO M tO to tO 'Si to to 'Si 'Si 'Si to MA '0 M to Co to '0 Co 1-, I 1 1 1 1 1 I 1 1 I 1 1 1 I U
f45
+- -0 CO -.0 0, h, ON CO +- Ni CO sO +- +:1" V7 Os +- OD N +- f.0 IN N. ...... 0-. VI -0 LID CO MI +- M L.L.
Co +- N CA +- .4- CO 'Si N to CO +- 0, 4- f"0 CA CA N) Os 0, to '47 1+1 N. CO R. .- ..tr Si OD N. CO
'Si Co r, to 'Si Co 4,- N .4' CA tO h, tO tO M CO Co M -Cr CA tO t5, M ..-- CO tO N IN to '0 M r, .. • ••1•16••••• •• • • • • • • ••• ■ ••• • • • • • • . • r,.
Co Co Co 'Si tO M Co tae MI to Co 'Si M to MA Co SA M 'Si M Co 'Si Co '0 to M Co to MI 'Si 'Si IZA
I I I I I I I I I I I I I I I 0.1 1-....i -In 173 I--
CC'
+- VD "qr +- CA VI ..tr E 13.4 C.) 1.4.1 1.1. CD DC 4-4 -J 2= 2.** CD CD CD at r iC 04 (...1 L.L.. CD DC -J 1: CD CD CD cn cn 0
(c) c;11:.7. r- T.: C. iii c in
0 CO
;3
CJ
d nos
D ivioi
aqg zoo void enTeAuebT2
94
Factor 6 loads B, i. e. the first and second order
intelligence factor.
Factor 7 is not unlike the second order Independence
factor, with major loadings on E and Q1.
However, there is relatively little dispute about second-
order factors, our main interest is in reproducing the first-
order structure, i. e. with single factors loading more or less
exclusively pairs of scales. The eigenvalue plot of Figure 1
was therefore inspected, as suggested by Cattell (1966b) for his
'scree' test for number of factors. The two most likely choices
are for 8 and 18 factors, implying extraction of 9 and 19
respectively to allow for one 'junk' factor into which odd scraps
of correlated error may be rotated to allow for a cleaner solution.
However, eight factors would give little scope for a closer
approximation to Cattell's model, and the choice of 18 looks on
the face of it fairly tenuous. It seemed more sensible to try other
factoring methods involving different definitions of communalities,
namely alpha factoring and image factoring. The first of these
resulted in a solution virtually indistinguishable from principal
axes, but image factoring gave a quite different result.
Thirteen factors were indicated, and again rotated by direct
oblimin. ., From this procedure a much clearer pattern emerged.
Single factors loaded A, B, F, H, I, Q1 and Q2 in both forms, another
95
G and Q3, but loadings on C, E, L, M, N, 0 and Q4 were complex
and uninterpretable. With hindsight, it is possible to say that the
scales in this last group were the source of all difficulties in
subsequent analyses. At this stage, given the quite different
results suggested by different methods, it seemed that the wisest
course was to experiment with different factoring and rotational
methods, and to look at matrices computed within sex and
discipline groups.
Initially, attempts were made to factor within-group
covariance matrices, but the proprietary software used turned out
to have traps for values over unity in the matrix, and since source
code was unavilable could not be rewritten. By way of experiment,
therefore, within group correlation matrices were factored both by
principal axes and the image method. Eigenvalue plots for the
iterated principal axes solutions are presented in Figures 2-5, but
here the scree test is of very little help: there are no obvious
choices for the number of factors. Image factoring of the same
matrices gives 18 factors for Male Arts, 18 for Male Science,
16 for Female Arts and 17 for Female Science.
Rotation of the image factors by direct oblimin yields
similar results overall. Clear factors corresponding to A, B, I, Q1
and Q2 appear in every analysis, F and H separate sometimes, E
only in the Female Science group, G and Q3 are loaded by the same
conar.D.cznini-im
0 13 1) W .th 4. 01 I
t.•03 (JR I\) Cli (4
96
cr DC 2: I/101A Uri
W 4). 4 CA 01 G1 IV 0) 0 Ch so 0 • ti,i'sliKIMMIlan.1.--mi.t,..,pn,vkvstr..z...-2^. Q •
97
•O
• © O
M
4-0
CJ
•P• ••
tj
(1)
P.)
CD
0
Fh0
CD
lii
CD
1:10
ct In
iR 0
tti
H-" co • tr)
0
IL) •• 0
1 C-)
0
4
•
9 8
ce) m zr- rn rti
1).• Po% I') (..A) f.,) 41% CA G'1
to co 1.7-1 co 0
• Cr1
(.1
9 9
co I'll r- z m n c-1 rn
O ad.
•
N W 4.1 .. cji
.• 01N C0 C9 N
Ei genval ue pl o
t f or th e f emal e sci ence grou p
33N3139 3-1t4W33
4.
eft
0
100
factor in every analysis, C, 0 and Q4 are invariably loaded by the
same factor or factors, and the remainder show no clear pattern.
Interpretatively similar results emerge from Varimax rotation of
16-factor principal axes analyses of the same matrices.
Thus a pattern begins to emerge. Factor B, as might be
expected since it is not a personality factor in the same sense as
the others, is invariably located; factors I and Q i, which Saville
and Blinkhorn (1976) showed to be clearly independent of extra-
version and neuroticism in forms C and D, also appear clearly.
Of the extraversion factors, A and Q2 are most easily located,
followed by F and H, whilst E is difficult to place. G and Q3 are
consistent but virtually inseparable, and when C, 0 and Q4 split it
is not along the lines suggested by their pairing across forms.
This emerging consistency suggested that perhaps with
different rotational procedures, allowing for deliberate targeting
of factors, Cattell's solution might be obtained. Analytic rotation
programs, after all, are known to have idiosyncracies. Varimax
for instance tends not to rotate much variance into the last few
factors when relatively many are extracted. Furthermore, based
as they are on particular operational definitions of simple structure,
they are not well-equipped to cope with the situation where one or
more variables may be complex, and not amenable to treatment in
terms of simple structure.
101
Attention was thus directed to the search for a procedure
which could rotate to a specified target. Graphical rotation was one
obvious candidate, and work was set in hand to produce an on-line
interactive graphical rotation program. Procrustean rotations of
one kind or another had undergone development to the point where
only a partly specified target was needed (Browne 1972 a and b), but
nonetheless required specification of fixed numerical values of at
least some elements in the factor pattern.
Hakstian's (1972) Optimal Resolution method seemed to offer
a better prospect. This involves specification of which factors have
high loadings on which variables, which low loadings, and which are
to remain unspecified. The difference between high and low loadings
as specified is then maximised, in the oblique case via a reference-
factor solution, thus avoiding the need for a numerically defined
target matrix.
The sixteen-factor principal axes factor patterns were
analysed by this means, with target matrices set up in three ways:
1. With the pair of scales hypothesised to be loaded by a
single factor specified as high, all others specified
as low.
Z. With only high loadings specified.
3. With only low loadings specified.
102
For all conditions, values of 1, 50, 100 and 1000 were tried for
Hakstian's weighting constant 'r'. In every case the reference-
factor correlation matrix proved to be singular, i. e. uninvertable,
and the reference-factor structure matrix could not be interpreted
with confidence.
However, a general interpretation of the failure of this
procedure is possible. For one or more pairs of Cattell's scales,
either one or both of a pair is loaded by another factor to the extent
that two factors collapse into each other in the analysis, or one or
more factors can be expressed as a linear composite of the others.
Factors C, 0 and Q4 were the most likely culprits, as noted
earlier, their intercorrelations being more or less identical with
their re/L' 61,6ie;,t,ei , and since analytic rotation had failed to
separate them. Fourteen-factor matrices were therefore entered
into optimal resolution rotation with these scales all targeted on one
factor, but again the result was the collapse of the factor inter-
correlation matrix.
At this stage, through the courtesy of Professor R.
MacDonald at the University of Alberta, a program for unrestricted
maximum likelihood factor analysis (COFA) was obtained, and at
the same time work on the author's own graphical rotation program
(ROTOG) was completed. COFA proved to have two major
103
disadvantages: firstly it aborted consistently when covariance
rather than correlation matrices were entered, and hence was not
ideal for providing comparable results across groups; secondly
it consumed large amounts of machine time, requiring a separate
run for each hypothesised number of factors rather than
successively estimating factors until an acceptable solution was
obtained. For these reasons, only two runs were performed
separately for each of the four sub-groups, one for seven or eight
factors as suggested by the Kaiser-Guttman criterion, one for
sixteen as suggested by Cattell's model. Analyses for 13-18
factors were run on the pooled within-group correlation matrix.
The Kaiser-Guttman model proved unacceptable for all
except the Female Science group, whilst the 16-factor model was
acceptable for all groups. For the pooled within-group matrix,
the 13-factor model was just acceptable. The number of factors
was then increased until the difference between chi-squared values
for successive numbers of factors was no longer significant at the
.05 level, which occurred after 17 factors had been extracted.
Test statistics are presented in Tables 8 and 9.
Graphical rotation of the 17-factor solution was attempted
with a view to targeting factors on to pairs of scales as before.
For some factors this was very straightforward. Factors
B, F, H, I, Q i and Q2 were rapidly isolated. Factors G and Q3
104
Group No. of Chi-squared Degrees of 2 factors freedom
Male Arts 7 395.14 293 0.00
Male Science 7 377.01 293 0.00
Female Arts 7 450.05 293 0.00
Female Science 8 280.24 268 0.29
Male Arts 16 75.34 104 0.98
Male Science 16 72.45 104 0.99
Female Arts 16 86.05 104 0.91
Female Science 16 59.80 104 1.00
Table 8: Test statistics for maximum likelihood factor analyses of data for four sex x discipline groups
105
No. of Chi- squared Degrees of E factors freedom
13 193.26 158 0.04
14 159.61 139 0.11
15 135.31 121 0.18
16 102.82 104 0.53
17 72.66 88 0.88
18 60.22 73 0.86
Table 9: Test statistics for maximum likelihood factoring of the pooled within-groups correlation matrix
106
r--4 0 1.11
V7 117 n ,-.. CD 'St- ,- '..0 Cr-, -o er co c, CO ro -- ro 0, CO Ci. T.) t'0 0 !"0 er ro cl u) n er (-4 v.)
u- r, co (-1 ill C4 ro C4 b7 V7 !", ,-- VO !", er n ,-• u-) b-) --t. n L....! Ls) lin L-1 -- C..1 t::zt ;7.1 'kr 171 ,-1- .....
)... CS! 1.,,-4 1,74 ei4 iSI C4 n e.'ii eiI C4 C4 tS! rs. n n r, n e, c) • • . ••••••• • • •• • 71 la •• ■ •• le....,...• .1., n n n n n n
r t n r V' n n. St CA eA e;:i i:!,:l n C, C, n n n tS4 tn n t.S.1 C, n n n LI
1 ! 1 I I I 1 ! I ! 1 01 q-
Cl N.. b7 0" 127 0- CA , - CD 1.-, b7 *44 0" ,0 ..-
CO i'l i-, CA KA "0 SI n er -- ro ro n c, er bl ts) ro -1) f,.. b7 CA ,-- VD V7 SI b7 VO n N) tS4 er r-:
co ts! n - St C, n f4 *-- C, =- ,- *-- n CA vS: C, C, M CA SA M CA VS: S4 =-- VIZi Ci Ci Ci ,..- "-- v- ru • • • • .... • • .1 II • Is • • • • • • • V. a. • • • • ••••• e)
m Ls) m! n n n ts! ts! ts! n ts. n m m n r-,1 n CS! ,-- n M S4 '<A SI n SA M n S4 ES! SA SI .4-.) 1 I I I ! I I I I 1 I I I I I I I 5:: •
C.:
IN N) V7 OD 0, VO 0" 1.-, "CI CA b7 M VO U7
n n es) CD SI c,:::" ,-- VO CA ,---. CA CA "0 "0 CA PT1 0
r'w Ci• ei) e:I eili e'.1 '5I q.:: C, St V) eA e=4 er m n C, n ts! S4 C, M r, n co - n ts) r, n Cr, er) m ul
vs • ■ ••■■ •• Is • • • ft Is •• • • • • • • 0 • • • • • • • • •
n n n n C, n n "Si C, n C, n n n n n C, n n n'n Ci S4 *A S4 Ci n Ci =- SI n SI 0
! 1 ! ! I I I II! I ! 1 I
,- ,- l'", .:!' n =- VD er CO SA U7 N' =1" b7 ._!-) ,-- +- Ci b7 CO b7 r, C IN -- n .0 -0 "0 CA 0" ..,.e.
C4 ,-- n U7 "0 "0 !.', V7 -CI "0 =- C. ,-- CA -- ,- .4- CA V7 . - C4 CA 0" *.1- <A ,--- "0 S ,--- . - OD -0 !"-- -40 =-- S4 +- mt co C n n =- C, C, ts! Ci ts: r.:;,! C, ts! if, •-• ts-I 0.- n - !C, Ls! ks! 'N riz, n in • • • •••••• ■ ••• • • • .4.....
Ci S4 Si n n St SI VS! r' V KA tii,I eil e> el) C, n n M n n n n n n KA S4 SI n n S4 St ! I I I I I I . I I I I I I U)
,c) OD IN ,D V7 CO V) I'', VO -0 CO CA CO "0 ,D .C.: e)
b7 SI S4 +- SI S! n MA n ...- MA n 0.4 MA MA n U,I ,.-::f MA MA t'.4 n ,..... n ..- 1'2! n tn n n C,- 4::
•••11•••• • ■ ••%, 'Ma • • ••••••••• •• ss••
Ci- Ci
Co Ki4 e:,) Ci e'il e.-4i ei) eiI tr... eil C, tSi eti C! e1 eg.:. C, tEl e.: Ci elI LSI tS) eS) CS) Co-: ..- M C, n SA ,--1
1 ! I I I I I I I I I I I 0
er ct- 1.-) ro -o r.,•r"-! =- SI -CP =- -40 =-- V7 f". CO CO CA i', CA KA ^0 C4 C4 n ^0 n er oo -- CD cq C 0, CO ,0 Er.1 ,0 0, e,i t!!;:i t", r', ...- 17'.: ..-- r, er
5- er v-... C, n tel es. r, - et) te cl -- ro - CH:, CO -- et. CS) n N) n n m n m CS) SA tn.) n m co r'" Ili 0 . • • •• ■ •• • • • • • A • • • • ■ • • ••••••••••• 'I..'
1-, C! n MA n CS) SI n n CS) MA n MA M n M n '"! n n n n CS) n n n M M M CS) CS) CS) n -t-, LI 1 1 I I ! 1 . I ! I I ! 03 fi, ea. L.L. CO r4 =- r--... m Cq -0 .-- 0" CA 0, N7 ,- V7 ,- )-- tn ,0 N. 00 0.1 N U7 KA t-, 0" ,-- .1: b7 "0 er cA I
CO ,-- ..-- =- CO .1- 0.'- l', ,-- b7 "7 0- )- 1.1 CA IN co n cq er OD V7 Ci •cl- n 0.0) er
N) CS) n SI b7 .- C) +- n S4 ^0 ,-- SI 'N <A KA +- CU S., St SI SI C, n clA n .-- SI n n CS) 15) SI ,D 0
... . . • . • . ••• • • •••• • • is••••••••• • s e• ,...1
• IS) n MA M CS) M M ,ts! n to CS) Se C5) SA SA v54 n vs! n Ci n n n n CS) n CS) n C5) QA St U ! I I I I ! I I I I I I I I I tr.,
1.2...
41' C.0 r-, N. 1.0 Ct.- "4" tn +- in 0' `.0 0.• Cl r, - r, n e- co ,D IfI. U7 CA ,-- CO CA er ts1 -o - rq uo ro - r, CA c-t ro 1:7 "0 er er ro ,5.4 .0 (-.4 __- er ez! cr. Ls! Ls! - es! uo un Ls. IN - er ,--- -
cq in CS) CS) n n n es! n CS) CA CA St S:, S SI flA CS) c. 15) ,- C.::: CS) c5) n ..-- SI n CS) M ,- ml Cl 0 ••• ••••• %V •••• • • • • fals••• •••••■ •• •• ,
CS) C. in MA n n tn t.,A n St IS) CSI ES) e:.) CSC C, n S4 SI M M n CS) M n ..- MA n .i) Cs! CS) C5:, Ci I I I I I I I I I I I I I I ! I
.1-1
If) 140 ro ^0 IN ^0 V, CO CA N7 n n .1' b7 VO N, U7 U7 V) ,-- .1- CS) SA 1.17 M ,-- ^0 CO "1" to 0. C-
r, r, 44- ^0 er c.-4 - IN If) -,-/- .4- ^0 f'0 N. N1 -•:!. ..qr r.... C5) cg ro IN - CA CA cq 'C CS) CS) CS) IN Cl 671
..- NI MA MA n m n t... Ls) es PS) n Lsz: to m ts) n C, n n - CS) Ls:: es! n PS) e1 C" n Ls! CS) n est • • . • • • . • • • • . • . • • • . • . • • • 4•••••• ••
St ,SI el..? CS) MA n CS) M MI MIA MA MA CS) MA Co, ..,- MA n t5) MA MA n MA MA to n n in CS) n C, M 1 I 1 I 1 1 I 1 1 I I
- er
N) er cl ro e- r 0) Ci.)Lt.! CD 2 !--4 -I 1: 2: CD CO CO CO CO 1.174 Ci 1.4.1 CL. CO I-I .qr. C) CO cr. Ci Ci
0 LL CL
•
107
•
0
r.... CI 1-, !...1 ra
c!--
r, ro -o :--.1 uo ,- c- IN r, C4 NO N, .-- 0, ,0 !", CO m! NI N) a CA Cl "1" ND .,-- C4 C4 •-• C4 ,-- C4 a !--.
. . • . • • • • • • • I: al... • a et • a • It II 4 n • a 0
to t *4 n t' n IS! n n n *4 *4 to
I I I I I ! I I I ! I ! 1 I != ID
C4 tr) "{) *-1-. 0... TN n o% — co r, -a c-%! a ND NI I", h-.I a a !"-, NO ..,- CA CO CA tn r•, CD NO 0, In :: .
NO -.0 k!-.1 '4" -0 c...! -o c..! c%! co r% r, — -4- ''.- '0 N7, n In tn •,?' OD F") h", n ..- •-- IN r',..! !,--. ,-- 1.1) tS! si
II a • •
••••••• • • •••••• ••• • • • • •••
n *4 *4 IS! t) n n m! *4 n *4 m! n n n *4 IS! *4 *4 n 0
! I I ! ! 1 I I ! I ELI .1..)
uo ro
▪
r, -.0 co o- P... r•O Cl 0, CO ,- ,-- -,-- IN Cl
m! 0,4 -- PO tn 1'0 'Kt' tn tn tn n n tn =- ,-- tn tn .-- n n tS1 tn.; t • • • • • . • . . • • • .t • el • • • • • C!....
t'2. ,.. E.... 174 ES! m. ts! m! m! *4 *4 m! S! *.! n ISA C.4 n n n n m! m-4 ES! ES! r.,! ES! ES! L I I I I I I ! II ! I' ! I ! ! 01
t-1 ',I" t0 a!
• - • • • • . • . • . • • • • • • • • • • 0 n K. *4 n n n n n n n n *4 n n n m4 n m!
1 ! ! ! ! ! i I 1 ! ! 1 I ! III ,---
vo r, Pr, 0, V) N;3 0....z! ro C4 N, ,- 1.!') ro . -- 1:"..1 ro ro llo ts4 !--o — crzz! 0, co 5-
:- ,--- ES2 t'.1.A .. - En "4:1 tiz! ES! *-- C..! ..:1 CA tn trn n n ,-- tn n t'Ir:4 n 0,-1. ES! MI ,•-• tS! ES! ..:1 1S1 ,- •!-- al 0 • ! • • • . • • . . " • . . • 4 •• la • • • w • • -!...'
n n n 0,4 m! 0,4 n n n n n n n n n n 0.-4 014 04 04 n n 0,4 m! 0,4 %:-,! n mo -,!--,, LI 1 I 1 ! ! 1 1 i ! ! 1 ! ! I It. 65 cl„ D.- CN ,- IA? OD -4) CO C..! h71 c4 0,4 o% ro -ID -o 0- m! I
0.34 n ro n m4 ro 04 n n m! n 04 ro ro uo
— ml cr- n n n 04 m! n n n n n n n n t n n• n n m! n m! n n n n n n n 0.74 0!:-.1 o . . . • • • • • • • 1. . 4 a •• '".
•I■ •• •• • .1.-:,
*4 n *4 n m) *4 n IS4 mA n IS) n n *4 I-I
I
▪
I I I I I ! ! ! II I It. ...1-1
n v.:I ,- m. cl !!,:.4 n c!!. ,- *A ,- ,- *4 *4 to n tS! tEZ! ,- m! n t''' m4 m! n n m.4 *4 M. n m! C!
. . . • • 11••16 •••••• am 4111•• ft 44 • • •• • 3
1 ! 1 I I I ! I I I I I I ! ! I .P r-.
n ea ,- n -- -- n n Si n oo n ci -- n cq n m! n 0,4 n n -- 0.4 n 014 0,4 — mi n n n • • . aa II •• ■•••el•Yli ••••• Is If•••••••••
I I I ! I I I ! I I I I I I I I I ,-
Pc! (.._) DJ Lt- CD
0 U-
y2-4 !ET:
c 0.1 01 01 r•-• 0
cq
U-
r) PCA LLI U. CD _J 1: CI ci cn co ct!
108
split into one triplet and one singlet factor, with factor G form B
loaded by the singlet. C, L, 0 and Q4 split into a 'general' factor
and three singlets. Factor A scales were loaded jointly by two
factors, and the form A scale by a singlet in addition. Factor M
form B is loaded by a singlet, and the rest of the scales have
complex loadings. The same matrix was also rotated by direct
oblimin, with such strikingly similar results that the factor pattern
in Table 10 is representative of both solutions, being in fact the
oblimin results. But once again the optimal resolution method
resulted in a collapsed factor space.
14-, 15- and 16-factor solutions for this matrix were also
rotated graphically: factor A consolidated into a single factor, but
otherwise little improvement resulted. However, in the long and
complex process of performing these graphical rotations, certain
features not immediately apparent from rotated factor pattern
matrices concerning the configuration of projections of test
variables onto the factor space became clear. Firstly in each
COFA solution the communality of one variable, different in each
run, tended to be estimated as noticeably higher than the others
with the result that a corresponding singlet factor was found.
Secondly, it proved quite impossible to steer factors to load
separately the C, L, 0 and Q4 scales according to Cattell's model.
Thirdly, the only way to separate N from E was to allow the E
scales to be loaded by factors corresponding to H and the anxiety
109
scales. In passing, it is worth commenting that since many hours
were spent trying to force Cattell's structure unsuccessfully onto
these results, corresponding perhaps to many months of labour
without computer assistance, either the possibilities of rigging
results by this means are exaggerated by its critics (e. g. Eysenck
and Eysenck 1969 p. 327), or the structure of the present data is
uncommonly well marked.
Essentially the use of maximum likelihood factoring and
graphical rotation have provided no information over and above the
13-factor image solution for the total group. The unrestricted
analyses here reported fall essentially into two groups: those using
the Kaiser-Guttman criterion for number of factors, including
principal axes, alpha factoring and COFA 7 and 8 factor solutions,
all of which yielded similar approximations to the Cattell second-
order model, and the remainder, using either the image-factor
criterion, the goodness-of-fit test in COFA or simply an arbitrary
choice of 16 factors, all of which agree more or less on which pairs
of scales are clearly loaded by single factors and which are not.
The factor patterns of Tables 7 and 10 respectively represent these
two types of solution.
Thus there appears to be partial support from these analyses
for Cattell's model. It is not suggested that any of the solutions
discussed should be regarded as definitive, as their purpose was
110
simply to facilitate a decision as to whether the expense of
computer resources involved in a more adequate test of the model
was worthwhile and to gain some idea of whether splitting the sample
by sex and discipline would have any serious consequences. While
it is difficult to demonstrate concisely, the impression was gained
from a study of all the exploratory analyses conducted, of which
there were over fifty, that splitting by subgroups tended to result in
less correlated, more identifiable factors, without noticeable
shrinking of the size of factors associated with scales on which there
were large mean differences across the groups. However, this
tendency was not so strong as to prevent the 13-factor image
solution for the total group giving results very similar indeed to
subgroup analyses.
We now turn to what was intended as the definitive procedure
for determing the fit of Cattell's model to the data.
111
CHAPTER 8
Use of unrestricted methods of factor analysis in hypothesis
testing may, as has been indicated, lead to the confirmation of an
hypothesis, but it is difficult to make an incontrovertible case for
disconfirmation, as the history of critical studies of Cattell's work
indicates. It is always possible to claim that particular matrices
are not suitable for a particular method of factor extraction and/or
rotation, since communalities can be estimated but not defined,
criteria for rotation are to some extent arbitrary, and there is no
way of arranging computational routines to ensure that correlated
measurement error is not extracted along with common-factor
variance. What is needed is a method whereby an hypothesised
model may be specified beforehand and formally tested against the
data.
Such a method has been provided by JOreskog (1970) in the
form of the technique of analysis of covariance structures (ACOVS).
Recalling the matrix formulation of the common factor model in
Chapter 4:
= A cPA'-i-W2
112
where A is the usual matrix of factor-pattern coefficients, (1)
is the matrix of factor covariances and is the (diagonal)
matrix of unique-factor loadings, elements of any or all of these
matrices are directly estimated by maximum-likelihood, subject
to whatever constraints are imposed on the model. The ACOVS
model, which specifies only the structure of the covariance
matrix, may be extended in the usual way to impose a structure on
the means, and in particular to the testing of a model in more than
one population simultaneously. The analyses reported in this
Chapter were carried out using Jbreskog and Van Thillo's (1970)
programs for simultaneous factor analysis in several populations
(SIFASP), which are simply revised versions of the original ACOVS
programs allowing easier specification of certain parameters.
The ACOVS method satisfies almost entirely the require-
ments for a decisive analysis. Firstly, covariance matrices are
acceptable (indeed preferable), thus allowing for different scale
variances across groups. Secondly, Cattell's model may be
expressed to varying degrees of strictness in terms of the para-
meters of the model. Thirdly, a test of the goodness of fit of the
hypothesised structure is provided, allowing at least a comparison
betWeen alternatives. Fourthly, a structure may be specified in
such a way as to exclude correlated error fromthe factor pattern
matrix. However, indeterminancy still remains to the extent that
113
a well-fitting factor model is not necessarily the only model that
will fit.
The SIFASP program was obtained and modified to suit the
size of the problem. Since a 16-factor solution for 32 scales
implies 512 independent elements in the factor pattern matrix, 136
in the factor covariance matrix and 32 unique factor variances,
there are 680 elements altogether for a single population. However,
most of the factor pattern coefficients are implied to be zero, and
the major problem was to expand the program to actually allow
simultaneous analysis of more than one matrix. Expansion up to
the estimation of 300 parameters proved possible, at which size the
program occupies 88K words of core storage, rather more than
half of which is accounted for by the matrix of approximations to
second derivatives in the minimisation procedure used. Since
this matrix grows approximately as the square of the number of
elements to be estimated, the modified Fletcher-Powell algorithm
used in SIFASP is clearly a constraint on the usefulness of the
program for large problems.
Variance-covariance matrices were computed for ten groups
and combinations of groups as shown in Table 11. These matrices
are to be found as Tables Al to A10 in the Appendix.
114
N
Male Arts (MA) 233
Male Science (MS) 312
Female Arts (FA) 396
Female Science (FS) 203
Male Total (MT) 545
Female Total (FT) 599
Arts Total (AT) 629
Science Total (ST) 515
Pooled Within Groups (WG) 1144
Total Group (TG) 1144
Table 11: The ten groups and combinations of groups, with sample sizes, for the restricted analyses
115
Initially, a strict form of Cattell's model was tested. The
four basic sex x discipline groups were analysed in a single run,
with each scale specified to be loaded by a single factor, the factor
pattern coefficients being constrained to equality across groups but
not across forms. The factor covariance matrices were free, but
constrained to equality across groups, and unique variances were
left entirely free. Starting values for the minimisation were:
Factor pattern coefficients 0. 6
Factor covariance matrix: identity matrix
Unique variances: 0. 5
There were thus 32 elements to be estimated in a common
factor pattern for all groups, and 136 in the factor covariance
matrix, again common to all groups; estimated separately for each
group were 32 unique variances. The total number of parameters
estimated for any group was therefore 200, of which 168 were jointly
estimated over all groups.
Since the Fletcher-Powell procedure converges approximately
quadratically, it is possible to judge whether a given model is likely
to prove satisfactory before the minimisation procedure has actually
converged. In the case of this analysis, it was clear that an
acceptable value of chi-squared was very unlikely to result, since
at the point at which the difference between successive-function
evaluations was 5% the corresponding chi-squared value was still
approximately six times the number of degrees of freedom.
116
The ideal procedure to follow at this stage would have been
to estimate a unique but unrestricted model, and then successively
impose constraints as recommended by Mreskog (1971), however
the size of such a problem exceeds core storage on any computer
available to the author in this country, and furthermore it involves
the estimation of many more parameters than there are subjects in
each subgroup. Therefore the next step was to test the model out-
lined above on each of the ten groups in Table 11, again specifying
two scales per factor and leaving factor covariance matrices and
unique variances free, since pilot runs suggested that removal of
constraints across groups would lead to a substantial reduction in
chi-squared relative to degrees of freedom.
Accordingly, when sufficient computer time became
available, this analysis was run to convergence on each of the 10
matrices mentioned above, each run requiring 2-3 hours of
processor time on the Hatfield DEC-system 10. The results for
each group were strikingly similar, and will be discussed in detail
shortly. However, the goodness-of-fit test for the model as
estimated still gave highly significant results on each run, and
furthermore in every case the program converged to a solution in
which the factor covariance matrix was not positive definite.
A number of experimental runs were then tried, changing
starting values for the minimisation, and gradually combining
117
specified loadings of factors on scales towards Cattell's second-
order factor model. Changes in starting values had no effect on
the solution obtained, although using the factor correlation matrices
in Table 10.2 of Cattell, Eber and Tatsuoka (1970) as starting values
for a sixteen-factor model speeded minimisation noticeably.
Reducing the number of factors to thirteen, by specifying loadings
of single factors on C, 0 and Q4 on the one hand and G and Q3 on
the other, whilst allowing L and N to be completely free, did not
improve the goodness of fit, whilst attempts to fit the complete
second-order factor model were abortive, resulting in negative
variance estimates for one or more factors. Allowing the number
of parameters to exceed the sample size did improve the goodness
of fit, but not surprisingly the results of such runs were entirely
uninterpretable.
It thus remains to interpret the ten separate analyses by
subgroups and groups with their apparently poor fit and singular
factor covariance matrices. Table 12 gives estimated factor
pattern values for each of the groups, all but two loadings per scale
having been set to zero by hypothesis - since each scale is loaded
only by a single factor, these are also factor-structure coefficients.
Table 13 gives an example of a factor covariance matrix (lower
triangle) with the off-diagonal elements in the upper triangle
rescaled to unit factor variance (i. e. correlations). Since these
118
G4 0 .t.i
0 E-4
L.,
E-I CO
E-4 (si
E--i
CO
tip
z 2 o
N N
0
en N
0
0 N
0
CO N
•
711 N
6
s.o N
0
N
O
CO N
.0
.0
0
CO
CO N
0
,J0 CO
11.) N
0
00 N
CO co 0
.0 N
0;
Lt1
0
N co
in .0
0
-14 N
'11 N
0
TV N
CO .0
0
N N
'II N
Li-) N
0
tin .0
0
it) N
0 N
0
co N
ON N
0
■0 N
0 CO
0
C. N
N N .
t4 N
0
N CO
t
N N
O
.0 N
0
1 N
.--1 co
• O
.-I CO
0 CO
0
0 co
N N .
7t, CO
0
CO
0
.0 N
o
0 CO
0
•
•ti N
0
IC) N
ce) N
0
714 N
cn N
CO N
6
ON .0
0
CO N
o
LU CO
0
tn N
0
cil as
• O
NI ON
.-1 ON
0
'74H Cr.
--4 as
't 0",
0;'
oN
0
ON co o
co CO
0
N Crs
0
00000000000 00000
N co
• 0
tr) in
(;) N
0
.-i .0
N .0
•,J0 N
• 0
L.rt
0
.0 rn o
cn .0
0
N
• 0
0000000000 00000
..-I ...o 0
0 ..o
0. in
0
,..4 .0
,..-1 .0
0 ,0
• 0
•714 in
0
s-O .0
o
,--1 .0
0
714 t.11
C0
00000000000 000
.fo -14 0
rel .14
•11 7ti .
0
oft n61
,--1 .11
N L.C1
0
tf) en
0
Cr, on
o
c:r, 7ti
0
N
0
LU mot' 0
cil ,711
LC) cn .
0
N In
N 7t4
Qs (-11 0
Lf)
0
CO LU
o
N cr)
0
cr.)
CD
in N
• 0
N N
N N
C:;
.0 N
.0
.0
N N o
11-1
0
ON .0
LI') N
0
N CO
O
°do
cr, ,..co 0
I-C1 .0
as in
N N
cil .0
,S) .0 .
0
CO cr
0
cr. .0
co if)
0
if) N
0
0000
a- in
0
0 .0
r...1 .0
CO LU
N in
'70 •-.0
0
CO if)
0
co Lf1
■D .0
0
M •.0
O
on N
• 0
N N
.0
.0
CO N
,-.4 N
in N.
0
N
0
co N
o
0 N
0
N 00
O
----co
• 0
CO N
ON N
cn co
.0 N
0 co .
N
0
o co o
0 CO
0
•zt4 CO
0
N re)
r:C1 w w 0 x aZ 0 a aa (r)
Esti
mate
d fa
cto
r
C)
›-• cd 0
Table
12
part
1
:
119
N 0 O N u1 CO .0 in .0 e•••4
N -44 N N co .0 co N to N Ln CO
▪
s-SO
&.4 •
CO CO CO 0 CO CI CD CO CD CI (0) 0 0 C
CD .0 Cs in d, ON lf1 N N cf) dl .0 CO .0 d4 cn .0 .0 co .0 co .0 u1 N ul N N r-
0
•
0 0 0 0 0 0 0 0 0 0 0 0 0 0
•
0
't CO 714 CY' dt dt • r-4 4-1
-
Cn
▪ N c.r) .0 .0 co N co .0 in N u1 N .0 .0 N E-
o o o o O O
•
o c0 O cr O c0
•
o
•::14 N dt tf1 In as ON CO d, (41 N N ce) •::14 N N co .0 CO .0 Lr) N in co .0 co N co
. . . . . . . . . • 0 0 0 o 0 0 0 0 0 0 0 0 0 0 0 0
.0 -44 N 0" CO en ce) L.r) M cc) o co cn (-€) .0 N .0 co .0 7r1 .0 In N co N N N
{1:1 LT-t co
•
o° do o o o oo
•
oc000
C:4 0 E-I co ,--I In in ,--I Cr, as N N .0 In o co -14 -14 N N -71-1 N .0 oN .0 CO .0 in N in co ul N N r-
C.1.1 O d o o o o o o o o o o o o o o I
0
Cn co (NI N cr, N 0 71-i ce) N C' CO 't Ce.) In ri -i-I
Cr) CO Cr/ in N N CO .0 "714 .0 Lc) .0 CO N N N (..) ct3
• 0 0 0 o 0 0 o c; o 0 0 0 0 0 0 0 7:3 U
N --I co NI N -t Li1 c■I cr) in co co -7ti on I.-I Cd
...• JD ,Z44 ....0 N N NO CO ,-.0 t11 N LSI N N CO N CO ri-i
C. -70 co co in o's •ti ,--1 0 .71, •--4 0 in o --I N
CrA .0 d, .0 .0 co .0 co .0 .0 N in co ul .0 N r-- c•1
0
•
C o 0 0 0 0 0 0 0 0 0 0 0 0 0 .. +4 F-■
Saf cd O P-4 O ...? CO 0 •0 c's in ON '14 N 'TO CO N d, 0 N 0 d, P --, in in CO u-.) 0" sr) 0" ',CO to CO `-.0 CO •-0 ON CO CO
N
000
▪
000000 00000• 00 CU
• cn as u C=1 f.T. "..71
-
1-; z oa
•
ctaa U
e te
n SIF
ASP
att
ern
coeff
icie
nts
fo
r t
0 0 0 0 0 0 0 0 O 0 O 0 0 0 0 O. +4
cc
120
NO a r. -.0 T... r4 h, MM a MI OD ,0 CA 47 4 a uo -c1 r I M IN to cr. co a cc r '7 a cr. u, CN 'Si Cl- 'Si 'Si CA .t.4- 'Si 'Si L-..1 No 0.. • • • • • • • • • • • • • • • • m CP M CP CP CP CA CP CP M CA CP CP CP CP
I 1 I 1 I 1 1
CO CA NO CA AT 47 CP CA CA a- co op pa op ,4-
rn -40 a m IN .,c1 m r, N. or a- -,c1 - N7 NI NO P0 G.
CN CA CP NO -- -- 00 CA CA CA CA .tr .0 ..- -- p., ,_ .7..i • • . • • • • •••• ••••• CI
CD 'Si (Si'Si 'Si CA CD 'Si Si 'Si 'Si CP 'Si CP ..-
I 1 1 I I 1 1 1 I 1 CM
or oo a to m oo or -- CA CP PM MI 1.0 47 P., ui
CJ -- 4 CO r, -- or AT ,O P, CO CA CA ,0 4= CN 47 CP N7 -- 43 ,0 C,1 M1 M
••• • • •• • •••• n••• tT m rsi m mmm m mmmmmmmmm
11111 1 al
• m r. CA 0, 47 47 OD 0, Pa. CO 47 47 M a M.1 MI C4 CO 0, P0 m o- o- -- co r, r,
CV CA PM CA ro to a CA -- CP CP • . • • • •• ••••• • • ••
CP CA 'Si (Si 'Si 'Si 'Si CA 'Si 'Si WA M 'Si CP WO 0 1 I I I
0, PM PM Cr- N ND PM .qr WM CA M- NO N. r4 OD 0, VO r4 N At CP 00 Or. NO CA 0, cc 0, 00 41
O CJ m 'Si M N- ,0 NI a - m 'IZr CA CA OD WM • • • • • . • • • • • • • • 43
MMMM M M M 1Z 1 I I I 1 1
NO r., ON V, I', CO Cr` CO NO OD a m PI NI C..1 cc .... Cr- a- or oo -- ro - m r. -- N. ro or a 0-
r. at
m CP CP CP 00 r, .v ,r ._ ,r is, ,_ 'Si ,r __ b, (...4 r•-• • • • • • • • • • ••••••• 7.r.,
I 1 I I 1 1 I i IS •••/
a PM CA 47 Pa AT P0 47 NO op CJ pri m ..- m 1..,
a 0, CA 0, 'Si CN OD CA 40 -0 Cr- N. r, r uo to 4= 31: CA .-- CP .4- -- ro -- -4- m is, .4- -- uo m CA CP
. • • . • • • . • • 1 b • • • • • r_
WA (Si 'Si CA M 'Si CP CP Si (Si CP CP M M Si (Si al 1 I 1 i i I 0.
is WA CO 47 ,0 CA AT NM ND -a - 47 43 CP CP N 1.17
Pa 0, 41 M -0 ...- CA .-- Pa 47 PM CA 47 h, 47 0, -- J CP CA 43 ,0 PM .-- CP 'Si CP CP 41 AD .4 -- ro o, • • • • • • . • • • • • • • • • I.-
CP m CA CP CP CP CP CA .- CP CP CA M CP CP MI 11 I I I I I I 1 1 ..4
C..1 ...... CN 10 T... era '0 47 N. 0, 43 .- CO -.0 'Si M
CA CP AT CO 0• CP 0, 4D ..-- a- m P0 .4- CP 0, .- ...-4
1-4 .- NM a- CA CA .-. CA OD CP NM '..- AT CA .-- CA 47 a, • •• • •• •••• • • •• • 5-
CP M M.! (Si 'Si CP CP CA CP CP CP CP M M M M S. I 11111 1 1 I 0
U CA 0, N N 47 NO 03 CO 03 ND N. UM M CO, NO CA ,-- PM 0- CA NM OD a- - r4 co r4 to N. 47 ,O N)
7C 47 CP 47 Pa P, M PM P0 CP .-- 40 CO .-- 47 P0 N. = • . . • . . • . • • . • • . • • M
CP M CP CP CP CA .-- WA CP CP CP CA M CP CA CP
I 1 I I I I 1 cu
.....
4 CO CO •-- CA r, 0t] 0, -0 47 vo o, ts,r000 oN ,-.4
-- -- No uo (NI -- -- -- .4. CA CO .-- PM OD 47 CO CM
CO PM -- - CA N 41 .- .- ,-- 4 47 CA 47 -- .- MM C
. • • • . . • • .. • • .. • . M
M CP MI CA CP .- M CP CP VA qA MI Mt M .-- M '1'4
1 1 1 I 1 1 I I I I 5-
CP 0, At 47 ,0 N. MM 47 0, CA .-- OD h, C, CP -
47 PM 47 AV VI CA r4 N a- MI 0, r, -o MO 40 5-
U. MI - CA P, a- ro m NM a- - o- ro ro uo CJ 'Si Cr
. . • • ••••••••• ■ II • 3
CP FA CP m- M ,-- M CA CP M CP CP CP CP CP 0
1 1 1 I 1 1 I 1 .....!
N. N PO Pa PM .4- as uo cs c...! so N. as ro CA C4 MI Cr- a ro as cs CO ND AT a uo to uo Qr
uu m r4 r4 0- VM Cr- C..1 40 a m AT N CP • • • • • • • • • • • • •
M M M ■- MMMMM CP CP CP M CP M
1 1 1 1 i I I 17-
N. R. PM 0, NO CA OD MI PM to a 471 Cr, -41 cr m (5, -40 CCI to a 'Si Cr' cr r4 r4 a- co a- C,1
t.,J P0 CP PM CA P0 CA 00 a N. m a m P0 03 47 CI • • •••••••• • • • •
M CP CA CP M CP CP CP CP -- 1 I
a -o -.0 a a- OD 4 Cr- M '0 CO (5! -0 OD N. -0 4= a- to m CA CP NI CP P0 CA N) MM CA MO 47
04 CP 47 'Si CA 'Si CA CA IS, M CP 'Si (Si CP 01 . • ••• • ••• •11••••• Li
'Si CP 'Si MCPCPCP CP CP CP 'Si 'Si CP M CA 'Si 1 1111 1 1 1 • •
MM 0, CA N. P, P, WM ND 'Si Cp p, oq m 00 N1 PM AT VI m r, r, N. m 'Si to a cr. m
.Cr 0, CP PM ro uo m CA M7 m m • • • • • • • • • • • • &IONS
CP CP 'Si 'Si W CD 'Si CA CP (Si CP 'Si 'Si m 'SiM rs i1
1 I I I 1 I I 1 IV I-
CJND a CC 04 CD UJ U. G0 VC r..r J3= 2C CD CV CV CV CS
121
matrices vary only slightly from group to group, the rest are
presented in the Appendix.
Table 13 suggests several sources of the collapse of the
factor space: note particularly the correlations between factors
corresponding to C, 0 and Q4, between E and N and between G and
Q3. To locate the source of the problem, the matrix was first
tested for rank by the method of triangular decomposition, with the
result that it appeared to be of full rank. Next the eigenstructure
of the matrix was investigated, first by the Jacobi method. This
suggested eight positive and eight negative eigenvalues, of
approximately matching magnitudes. Since this seemed rather
implausible other methods were tried, notably the Householder-
Wilkinson tridiagonalisation approach. These indicated 13 positive
and three small negative (= -0. 1) eigenvalues. The triangular
decomposition analysis was then repeated with a higher tolerance
criterion, and gave the rank of the matrix as 13. Finally the
matrix was reconstituted from the 13 positive eigenvalues and their
associated eigenvectors from the Householder-Wilkinson analysis,
and found to be identical with the original to two decimal places.
Since in general with ill-conditioned or near-singular
matrices numerical analysis methods are particularly sensitive to
the numerical accuracy of the computer used, different methods
may give different results. However, corroborative evidence is
122
available for a rank of 13 for this matrix. The triangular
decomposition analysis proceeds by exhausting sequentially the row
and column with the largest residual entries at any stage. Rank is
determined by a present criterion level, such that when no entries
remain above this value, the number of rows (and columns)
exhausted is the rank, and the remaining rows and columns are
considered 'non-basic', i. e. they may be expressed as linear
composites of the 'basic' vectors. In this case, the vectors
corresponding to factors L, N and 0 are non-basic, and the two
smallest (i. e. least independent) basic vectors correspond to Q3
and Q4.
In this respect, the results of the SIFASP runs are very
like the results of the more traditional factor analyses discussed
in the previous Chapter. The same scales consistently fail to be
separable into independent factors. However, as has been noted,
chi-square values for these analyses are highly significant,
suggesting a poor fit of model to data. Table 14 gives the results
of the goodness of fit tests. McGaw and J8reskog (1971), however,
suggest that with substantial real data bases, chi-square values may
be very much inflated by a combination of slight deviations from
multivariate normality and small amounts of correlated measure-
ment error. Since JUreskog's function for minimisation, whose
value at the minimum is the basis of the test of goodness-of-fit,
123
Group Chi-squared Degrees of 2 freedom
MA 549.11 328 0.00
MS 619.43 328 0.00
FA 695.15 328 0.00
FS 475.05 328 0.00
MT 825.58 328 0.00
FT 839.33 328 0.00
AT 965.76 328 0.00
ST 778.87 328 0.00
WG 1248.85 328 0.00
TG 1269.48 328 0.00
Table 14: Test statistics for the ten SIFASP analyses
124
involves the sample size as a multiplying factor, whereas the
degrees of freedom for the chi-squared test are determined by the
numbers of fixed and free parameters in the model estimated, taking
no account of sample size, large samples in combination with small
correlated errors may increase the value of chi-squared out of
proportion to the size of the absolute discrepancy.
To overcome this problem, Tucker and Lewis (1971) suggest
a 'reliability coefficient' for assessing the acceptability of a model
under such circumstances:
p M. mo —
- 1 0
where M. X.2
for an i-factor model
v. L
The rationale for this statistic lies in the fact that the ratio
of chi-squared to degrees of freedom is an estimate of the residual
variance under the null hypothesis, hence 15 is an estimate of the
proportion of, residual variance under a zero-common-factor model
accounted for by the k-factor model. Table 15 gives values of
for each of the ten analyses, together with values of chi-squared
for the zero-factor model. Note that with the exception of the FS
group (which was the only group for which an 8-factor unrestricted
model was acceptable, and whose sample size was close to the
number of parameters to be estimated) all values of "p are very
125
Zero factor model 16 factor model x2
P MA 22541.3 0.9838
MS 28457.25 0.9832
MT 50330.92 0.9839
FA 36671.72 0.9836
FS 18846.62 0.9999762
FT 54622.68 0.9848
AT 58860.1 0.9824
ST 46692.0 0.9854
WG 103228.02 0.9854
TG 105097.7 0.9840
(528 degrees of freedom) (328 degrees of freedom)
Table 15: Tucker and Lewis "reliability" coefficients for the restricted analyses
126
close to each other, and all values are high, suggesting an accept-
able fit; there is of course no guarantee that only the particular
model tested will yield such favourable results, but since the only
a priori model available - the second-order factor model - had
already been tried and proved abortive, there seems little point in
proceeding to a posteriori model-building at this stage.
Table 16 consists of the residual matrix for the WG run, as
an example of the size of the residual covariances remaining after
the fitting of Cattell's model; it is in all respects representative of
the other residual matrices. Note that residual covariances between
corresponding scales in the two forms are effectively zero, that
most residuals are well below 0.1 (absolute values), and that 16 of
the 22 residuals above 0.1 are associated with factors L, M and N,
which had poor internal consistency as shown in Chapter 6.
Such a residual matrix is, in effect, an indication of the
correlated error between scales under the model fitted. Since
the factor covariance matrices were left free in the specification
of the model, the fact that certain factors were estimated as being
very highly correlated needs to be borne in mind, but with this
proviso it is clear that by and large Cattell's claim that the error
variance of each scale is made up of many small components rather
than a few large components is justified, or at least that such error
components as are present are in common to both scales in a pair
to reasonably close limits.
12.7
CA CO CA A- A- A, A, Ar A- Lil AD AO A, CD r, CO N CA A- CO AO Co CA N) uo so ro C'- co in 0, CO
a — — Al Al r, ro -- — -- — sr N) DA +- Cl CO AI AA A- r, CO 111 CA AS Al CO a CO CA N) ,- CO
cn CO CO CO CO CO CO CO rn CO — — CO CO rn CO CO rn CO m rn CO CO — CO rn CO rn CO CO CO CO CO • • . • • . • • • • . • • . • • • ••••••• ••••••••
CO M CO CO CO CO CO CO CO CO CO CO CO AA CO CO CO AD AA 'CO AA AA CO CO AA CO CO AA AA AA CO AA 1 I I 1 1 I I 1 1 I
CO Cl CO Cl ro ro 4 r- 1,0 '-Cl sr co (...4 r, CO N -- Cl 1.0 t..-4 a 03 so ro CO CO AO A- "0 A- CO CO
Al CA A- Ul CO A- AO 1M MO AO CO a — sr ro CO CA a ro CO rl ro — CO a- Cl ro co ro N) CA AA A-
CV CO CO CO AA CO CO CO CO CO CO CO CA 'CO AA CO CO CO CO CO CO CO CO CO CO AA CO rn CO *4 CO CO CO
••••••••• •••••• • • •••••••••• •••••
CO CO CO Cl CO CO CO CO AA CO CO CO AA CO CO CO CO AA CO AA CO CO CO AA AA CO CO CO CO CO AA CO
1 1 1 I I 1 1 1 1 1 1
AO CO CD Cl N) a '0 CO CO CO WO -O CO CO A, CO A, Ar CA DA CA A- Cl CO Cl N. to N. cr. 'CO so cr.
CA NA CO U7 Ul a CO CO ro N. -- so N) AA CO AO A- Al AA 03 1M CA CO 41 CO AA A- CO sr — CO N im
CA AA CO CO CO CO CO CO CO AA CO AA AA CO AA CO CO CO CO CO AA CO AA CO CO CO AA CO AA CO AA AA CO
. • . . . • . . • . • • • . • . • • • • • • • . • • • . • • • • 111
CO CO CO CO AA CO CO to CO AA CO CO CO CO CO CO CO CO CO to CO AA CA CO CO CO CO CO CO AA AA CO .A1 1 I I I I I I I I I I I I U)
DA sr AO AA CO sr cr.. r, — ,-- CA CO AA CO CA A, CO In vo IN os o- co r, AA CO A, -CI CA sr — DA AA
A- Ill ro so — DO A- AD A- AO A- a a m CA Ar DA PO Al CO AO C..1 F.1 CO A- CO 47, Cl CA sr -- sr ro m
CI CO CO CO CO AA CO CO AA CO CO CO CO CO CO CO CO CO AA AA CO AA AA CO A- CO AA AA CO A- CO CO CO 5.: • . • ••••••••• • ••• • ••••• • ••• •••• • • ill
M CO CO CO CO CO CO CO CO AA AA AA CO CO AA CO AA AA to CO AA CO CO CO CO AA CO CO CO CO CO AA 1 I I I I I I I I I I I I I -r,
at
1...N r.r.I AI DA r, .-- ro CO AV A, A- A- CO AO CO sO OA CA A- 0, AA CA DA -0 A, CO DA A- ro to IN CA 4,-.'
..... Cl CO , - N Al er CA ,-- •-- •-• CO Ar Al .-- PO CA N) CO -,- Al pA C4 r0 LAI ..--. A„ CO C-.1 CA .1- .1- LI
CD CA CA CO CO CO CO CO CO CO A- A- CO CO CO AA CO AA CO CO AA CO AA AA CO AA CO AA AA CO CO CO AA AA • • . •••••••• • • •• • • ••• •• • ••••••••• ....■
MI CO CO CO M rn CO CO CO CO CO (CO CO CO Cl CO CO CO CO CO CO M CO to M CO 'CO CO CO CO AA AA .1A
1 1 I I I I I I I I I I I I IP N
ro -O AO CA AA A- A, CO AO A- CO A- CA DA sr so a ro CO CO CA A, AO a CO sr CO cis sr a r-t sr r•
AACA DA ro CO Al A- Ar AA A- AA A- sr '-0 sr sr r, -- rn — a rn Cr. Zr to- ro CO — — sr CO r,
2.: m rn -- rn m m -- m rn *4 m — *4 rn rn .--. rn rn ..- rn CA S4 M AA tn n M ,-- M CA ..,-. n I:2 • • • • • • • • • • • • • • • • • • • • . • • • 0 N• II • • • • 7:i
CO AA CO CO AA CO AA AA AA AA CO AA CO AA CO CA CO AA AA AA AA CO CO CO AA CO CO CO AA CO CA AA Al
1 1 1 1 I 1 1 I 1 I I I I r-- Cr,
M M 141 N. CO MO CO N 1...... M ...... N. A- OD AO LA CO CA OA sr CA DA OA N ro (5) sr — CO -- ro co I
so ro ,-- DA Al CA (-4 sr PI AA A- A- A- A- AA A- CA AA r, ,-- rn r-. CA r4 rn CA 1-, DA Cr. -0 a -0 ,-.. • AA CO A- AA CO m CO CO CO CO CO ' m rn CO — — rn CO m CO CO *4 CO -- CO CO *4 (5) m CO rn .,-4
. . • •••• •••••••• • • ••••••••••••• ■ • .0
M CO CO CO CO AA AA CO AA S4 CO (Si CO CO M CO CO SSA CO M M n CO CO to SSA AA CO rn CO CO CO .1.....
1 1 1 I I 1 I I I 1 t 1 1 I I • Al
N) Al- Al CO CO N) so -0 rn r, --Cl a CO AA A- AA CA Cl LA A, AO Os "0 CO N) AA A- N) r, so A-AA CO CO AA AA CO AA CO CO CO AA CO AA AA AA CO • • • • • • •••••• I•• •
CO COCA M CO t5A MIMnnn SA to CO CO CA 1 1 I I I I I I I
N. ..0 CO r, -- r, — os tn uo rn CO DA CO A- ro so CA A- rn cr. un CA CA AA CO -0 DA AA DA Al un -,71 AA CA AA CO AA AA AA CA AA CA AA CO CO m rn (5) cu • • • • . • • • • • . • • ■■ •• F-4
CO CO CO M M M AA CO CA AA CO rn CO CO CO CO 0 I I I I I 0
Al < rn Cr so CO AO AA CO AA AO CA AA CO A, AA A- a a OD VI CA A- A- A- AA A- ,0 sr a co a cl sr-
CA A4 AA CA DA CO CA AA AO sr sr rl — ro ro — -a ro to n sr CA AA AA A- A- a to -- -- ro uo r... £ 1.4 M CO CO AA CO AA CO CO AA CO AA AA (Si <A AA AA AA CO CA CO CO AA AA CO (Si AA AA CO A- CO AA CO CI A. . • • • • • . • • . • . • . • • .• • • . • • • . . . • • . . • 5_ O CO CO CO rn rn CO CO *4 rn CO *4 CO rn CO AA AA CAA KA AA (5) CO AA CO AA AA CO CO AA AA AA AA AA A- LL I 1 1 1 I I I ! I I I ! I I I I I I
IP
to sr CO -- CO In CO CO AO CO N. ro CI., so a r, DO '0 DO DA N. A- AA A- CA AI Ar CA Cr, CA A- A, 0)
C4 A- A- sr CA Al AA AA CA C.4 ,- sr ro tn vo — r4 -- AD A- AA PA AA CI Ar A- Ul ro m — cut m 1..1
:a rn rn CO AA AA AA AA CO CO AA A- CA CO CA CO '5) AA AA CO AA CO to CO CO CO AA CO CO CO AA AA A- !-- . • • • • . • . • • • • . • • • • • • . . . • • • • n • It • • • .15
CO M CO CO (Si M CO AA CO AS CO CO CO CO CO CO CO AA CO AA CO AA AA CO AA CO CO 'Si CO AA CO CO .A1 ! ! I I I I ! I I I ! ! I I I I r.-
m
uo uo r,.. CA AO AA An AA A? PA A- A- AS sr ro r, r, LC) ro 17-4 so m to sr ro c4 -- rs. sr IN cl R. :..-
sr -- DA CA AO CO VI CO AO CA AA N) A- AA N) ro so (Si CO AO A- CA AA CO sO a — -- -- CO rn uo 0
CO CO CO AA AA CO AA AA AA AA AA CO CO CO 'Si CO CO AA CO AA (Si AA 'Si CO AA CO AA CO CO AA AA AA CO L1 • • • . • • ••••••••• • • ••• •• ■ •••• •• •••
AA CO AA AA AA AA AA AA CO AA AA CO AA CO CO CO AA AA CO AA AA AA CO CI AA KA CO CO CO 'Si 'CO CO 1-.1 I i 1 I I i I i 1 i I 1 I I I III
1
CI, rn CO ,... CO ,0 CO AD CO CO CO N. CO ro ro -- -4- sr 01 — 'Si los CO CO AA PO AD (Si AA OD DA e- AD
CA AA DA Al AA ro CA DA N AA CO CA DA Ar A- PA AA CA AD 00 AA A- AA Al A- AA A, AA P71 IN r.-.1 0, .,-4
LL CO CO AA AA CO CO AA CO CO AA AA CO CO CO CO AA CO CO CO CO CO CO 'CO CO A- CO AA AA SiA (5) 'Si AA A i . • • • • • ••• •• • • • • • •• II • • • • • • • • • • II a ID
M M CO AA CO CO AA AA AA AA AA AA AA CO AA CO CO AA CO AA AA CO AA CO AA AA AA CO CO 'CO 'CO AA CA: 1 1 I IIII I I I I I !
AS AA CA A- e- CA A- OD CO A, CA DA Ul Al CA A- AO -0 A, e- ,0 r.., CO MI CO to C..4 T... 'T. n a n ct
N? CA A- AA Al CA AS CA IM Ul MI A- e- DA CO Al 1...... -- ro CO uo 4 c.1 CO '-0 — CO -- so if) sr n r......
LU CO CA CO CO CO AA CO AA AA AA CO CO 'Si CO (5) CO CO CO AA CA CO AA CO AA AA MI CO CO CO CI CA A- • • • • . • • •••••••• • • • ••••• •••••••••1
CA AA CO CO '$1 AA CA CO CO AA AA AA CO CO MI AA CI (Si CO AA AA AA to AA AA to CO CA CO CO CO 'Si -1-) I I I I I I I I I I I I I r.-
ro
N A- ..-- OA AA Is, n so ro 1.,7 so ro so co tst CR cr- co — ro co CA CA to AD -.0 sr co n ,0 A- DA Q.
CA sr rn — hl Ul A- AA CA A- 1.)1 CA sO DA LA Al CA AA CA AO DA DA OD A- CA AA AO A- AA Al rl a
A..) AA AA AA CO AA AA CA CO CO A- e- CO CO CA m m CO 'CO CO AA AA AA AA CO CO CO *4 *4 CO CO 'CO M .40 • • • • ••••• •••.• • • • • • • • • ••••• 1.114 • • .......
M CO M CO CO CA AA CO (Si CO CO AA CO CO CA CO M MI CA CO AA AA CO AA AA CO An CO AA CO CO M I I 1 I I I I I I I I I I el
1.4
sr AA A- VA AO Ul AS CA AS CO -Cl PI sr Co c-.4 CO a '— 'Si Cr- OD AA sr — uo a (Si cr. N) 4 — CA iA
A- (5) a rn rn — ,- CO Cl 1A) CA IN PO AA e- A- A- CO er CO AA CA (N CA A) CO CA A- ro n CO — M
PA CA CO AA AA AA CO CA CI 15 CO CO CO AA AA CA CA AA AA CO CO AA CO CA CO CO 'CO CO CA AA (Si AA AA A- • • • • • • • . • • . • . • . . . • • . • . • • • • • • . • n •
Ilii CO AA CO CO CO AP CA AA CO 'CO CA CO AA AA AA CA CO AA n M CO CO CO CO AA CO AA AA CO AA AA I I I 11 1 1 1 1 IIIIII
CO 4 IN a Os 1M Al CO Al Al AA N DA sO CA los CO ro a N n N DA AO -0 n co a cr- co tr, a
CO A- N AI (-1 a rq CO ro -0 *4 — uo ro CA A- AA CA CA CO CA a ,--- sr so co a AA sr CO uo -- ‹ CO CO CO CO AA AA CO MA CO AA AA AA CA AA AA CO AA AA CA CO CA CO CO CO to M4 CO AA CO CO CO M
1 • 4 • ••••••••• ••• • ••••• 0•••11 •• ••11
CO CA CO CA CA CO CO CA CO 15) CO CA CO MA CO CA CA CO CO CO CO CO MA CO CO CO CA MA MA CA CA CO I I I I I I I I I I I I I
AL 'Si
A- VA AA er A- cl AA er AL OR L) LU LL CO 7C 1.4 -J AC 2t 1=t 1=3 CA CA CA C'L OR LI LW LL. CO MC H -J 3C .11C CD rI cn CV Cb
O 0
128
a r4 1.19 Co ,- 0" r, a ro co -a cq in or- co rn co Co CO cq No 4- CO r, CO sO sO if ,- MI 0, Co .gr ,- -- Co4- cr un Co uo uo ..0 r, a to If) '- Co 4- -- Co Co Co uo Co Co -- N. N) -- r, Co ro Co
CO Co Co CA ,- Co Co ,- Co Co CA Co Co Co Co Co CA Co Co Co CA ,- CA Co CA CA CA '- CA Co Co Co CA • ••••••••••••••• •••••••••0 ••••••
M Co Co CA M Co Co CA CA CA Co Co Co Co Co CA Co Co Co Co MA Co Co Co Co MI Co Co Co Co Co Co
I I / I I I I I I / I I / I I
Ill ..-- ..--. a Lin r4 ,- CA ,- ro CA CA ,- sO CA 0, PO un 4- a 0, 0, ,- Co -CI O. rl (N r, Co Co 0,
M uo en rA a ro Co r4 ro (....1 a n a a CJ to .,-- 4- ro sO OD CA MI ,0 ,-- r, ,- cl -- ro ,- in ro
Ca Co Co Co Co Co Co M Co Co Co 1.- Co Co M Co Co M Co Co Co Co Co Co Co Co M M Co Co CA (Si Co • • • • . . • • • . • . . • . • 11•••••••••••••••
Co M n CA CA CA MA CA CA CA Co Si Co Co Co Co Co Co M Co t CA Co Co Co Co Co CA Co Co 'Si Co
I I 1111 1 1 1 1 1111 I 1 1 I I 1
co a ,0 In CO rq c.4 a co --- a MI a n -- wo ro ,- cq -o Co rq -- co r, c.4 cr 0, rl Co CA MI
CA Co M rl 149 CA Co ,- ...- U9 .,0 a rl -- Co r.4 N) r4 n a Co a Co ,-. CA Co PO 1.19 ,- ,-. CA =-. Co
Cl si Co Co Co Co 'Si Co n Co Co Co Co Co Co Co Co Co Co 'Si .- Co 'Si Co M ,-. M Co Co MA Co Co MA
e • a ft v. • • I so • • • • • . • • • • • • e . • • . . • • ■ • in
Co Co M Co Co CA Co CA Co Co CA Co CA CA CA CA M Co Co Co Co CA CZ Co CA Co Co Co Co CA M CA •1-1
1 1 I 1 I I I 1 1 1 I I 1 I IP >.
0, N, 1'0 .-4- PC1 a 0, 00 Lil CA 4-4- PO CA 0, ...0 CO 129 ,- CO 03 In PO LICI 0-. ..q. CO bc4 PO CA CA r, -- -1
...- 4- ro n sO MI ,- CA ,- CA 0, ,-. rq a -- ro rA 1..... r4 1.19 ,- uo ro 4- c.4 ro No -- a- m -- ro r, RI
C2ScA (Si Co Co M Co Co •-••• CoESI MISSI •"•• Co Co Co Co Co tsinnmslmits! Co MI Co (54 (Si (St (5) r
••••••1••••••••• •••••••••••••••• III
M Co Co Co Co Co CA CA Co CA 'Si CA CA Co Co CA Co Co Co Co CA Co Co CA Co CA CA CA CA Co Co Co I I I I I I I I I I I I I 1 I I I "0
a)
a o- co -- Cu r, cr a co -- cr -- No r, -- lo- cl -- uo co -- uo sO ID ,0 CA rn -- ro LT- cl uo -1-,,
CA -- -- -- n -- to ro b9 Li) ,-. CA CA '1' MI CA ,-. ..-- ,- ,-. '1' P9 Cr, =- ,- bc, 0, CA 41- ,- ...... ,-- LI
C) Co 'Si CA Co Co Co CA Si CA CA ..-- Co Co Co Co Si Co CA Co Co Co Co Co Co Co Co M Co Co Co M CA •r-1
•••••••••••••••• •••••••••••••• •• l:•-•
Co CA CA Co Co CA CA CA CA CA CA CA CA CA Co CA Co CA CA CA M Co Co Co M M CA CoCo Co Co (5) C:1 4-'
IIII 1 1 I 1 I I ! 1 1 I 1 1 0 a,
CO CA 4- c.4 "0 ,- 4-4-' a n a- Co uo r, ro -0 N9 CA ,- CA r, PO ,0 U9 rA '4' rl Co MI uo 4- rA ,0 5- •a CA SO CA r, ,- Ul a No N. ts/ r, c..4 Co co 4- Co Co to PO ,- r4 r4 CA 'Si CA CA CP, ,- U9 r4 N)
Z Co Co CA Co Co CA Co Co CA Co CA Co Co Co Co CA Co Co ,- CA CA CA ,- Co Co CA CA CA CA CA CA ,-. CL!. • . • • • . • . . . • . • • • • • . • • • • • •••••••••• ..71
M M (5) CA CA CA CA Co CA CA CA CA CA CA CA CA CA Co CA CA CA CA CA CA CA Co Co CA CA CA Co Co CI I I I f I 1 I 1 I IIIIII 1 1
Cr, Cr,
41.1 4- No m ro CA h0 -40 In CA 'q- CA CO 1'... CA -o a r, co IN CO •-0 rA sO 149 CA CA rl OD rl CY, ,C)
CO CA CA =- CA a- -- -- CO Co VI ,- 0, ,- PO Co CA P9 '43 149 CA b9 Co ,- 19 CA Co b9 -0 V) -- r, .- • Co m Co CA Co CA CA CA CA Co CA CA CA CA Co Co CA CA CA SiC CA CA Co CA CA CA CA CA Cy Co m (5) ..-1
• ••••••••••• • ••• •••••••••••••••• !::
Co CSI Co Co M Co M Co M Co CA CA Co (Si CA Si CA CA CA Co CA 'Si CA CA CA CA CA Si CA Co Co MI 1-.-, I I 1 I I I I I ! I I • -1
-4:
401( bl sO CO VI MO CA ,- CA MO CO r, PO rl CO U9 MO ',CI cl- CA MI CO =... b9 CA b9 .1" .,0 .1- tN -cl co
sO PO rl sO ...... sO a -- m n or un n (...4 CA r4 U9 CA CA M4 0, CA CA 1" CA bl Co ,-. N9 CA r, ,- 7,
..J Co Co CA Co ,- CA CA Co (Si ....- CA Co CA Co CA CA Co Co Co CA CA CA ,- Co Co CA Co (Si CA ,- Si CA III • • li• •• • ••• 110 •••• 111111111 • 10%11 1111• • ion.. rl
Co Co Co Co Co Co rA Co Co Co Co CA Co Co Co M Co Co Co CA 'Si CA Co CA Co CA Co CA CA CA CA CA CI 1 1 11111 I ! I I I 0
PG In =- CA r0 CO *1- ...- CA 0, f`, a No 1.**N CO 1'0 I'l rl T.. 4- uo cr a t., -- uo -0 IN b9 0, CO CA r,
Tr Co ,- (Si MI CA Co CA CA r4 1- MI ,- m a a .-- -- a to vo co 45) n a --- cl -- r..4 cl -- in 1-.. E I-I Co CA CA C:4 Co Co CA Co Cr CA CA CA ,-. Co CA Co Co Co Co Co 151 M M4 M4 Si '5) MI Co M MI (Si Co C)
5-• . • • • • • • • • . . • . . • . • • . • • • • • . • • • • • . r O m Cl Co m Co Co m Co Co CA MA Co (Si CA 'Si CA Co Co Co Co rA Co Co Co Co CA CA Co CA CI Co CA .'1-
141.. I 1 I 1 1 1 1 I 1 ! I u)
L) '1" r4 CO SA MI Co ,- =-. Cr- ,0 U9 CO r4 -0 0, 0, Co ro No 0- cl -- r, -- c...1 uo -0 uo --- -- co al
-- CA 03 CA MI CA CA CA CA CA 0, cl m uo n n -- 1:-..4 a -4 -- cq ts/ n m 45/ c.., o- a -- -47 CA LI
=C MA CA CA SiC Co MA CA CA Co CA CA CA 'Si (Si CA ....- CA Co Co Co CA CA CA CA ..-- Co ..-- CA Co Co t5. CA • • • • • • • • . • . • • . • • • • •••••••• ■ ••••• 17.1
Co (Si Co Co Co M Co Co Co (5) CA Co CA Co 'Si MA (Si Co (Si m mi MA CA Co Co CA CA 'Si M4 M to MI ,-.! I I 1 1 I 1 1 I I 1:-
ITI
cl ro rA N. 0- 'M T-. T.. N ID N. 0... 01' .1-- OD CO CO MI (Si r, =- Co EN 1 CO ,0 -(1 In MI CA Cc .1- >
'Cc r4 bl a -- n vo CA U9 R. CA PI MI CA ,-. U9 ---- =-. ,-- PO r., Co cl CO CA U9 IN to ro in P-.1 In C)
CO Co Co Co m Co m m Co Co n Co CA Co. (Si Co CSI CA Co vA (Ci Co U,1 CA CA Co CA Co cA Co rA rA CA U • • • • • • • . • • • • • • • • • • • • • • ••••••••••
M rA CA CA CA CA CA Co CA Co CA CA Co Co Co CA Co Co CA Co 'Si (Si CA CA CZ CA CA CA CA 'Si 'Si CA ...A
I 1 I I 1 I 1 1 1 1 1 1 1 I I) :1
CA CO CO ,0 CA ,0 IN 4..... , ''... CA CA Co Cc Cc .1- sD ,-. =- U9 =.- CA ,- 0, 0, ro co ro -- !Jo n 0, -o -r,
CA Co In In Co ,- 'Si -4- 0, CA ..,,M PO CA rl PO CO r4 Co a -0 in N. -- PO 0, CA =- -4- In '4- Co CSI .-.1
LL Co Co m M (Si M Co Co M Co M Co M CA (A CA Co (Si (Si CA 'Si CA CA Co Co Co CA M4 (Si Co M4 .,..- 1.11
. • . • • • a • • • • • • • • • • • •••••••••••••• Ill
MM MI Co Co Co CI (54 (Si Co Co C., Co Co t.Si (51 MI (54 to (Si Co Co (Si (Si (Si (Si (Si CA 'Si Co ,SI 1:51 CC:: 1 1 1 11111 I 1 1 1 1 1 ! 1
a a
r..1 0" In ...-.. T.. Cl Ul Cl- N .1.4. CO 0.. Cl U9 rA CO Si CA ,0 =-. T... N. '0 uo cl CA r, CO CO ,0 '1" CA 0
CO CA sr) CA OD PI ,-. CA CP ,-. ,- =- "0 U9 ri r, NI CA PO CA ,0 N9 4- ro n bI NI ,- ,- CA CO CA I: W CZ CA Co CA CA CA Co CA 'Si m 45-4 Co ts! Co 'Si Co 'Si Co Co Co Co CA in Co 'Si Co 43/ Co Co -- Co m -1-)
. ••••••••••••••11 *•••••••• • • • • N114
(Si CA Co CA CA CA CZ CA (Si CA CA rA CA CZ CA CA Co CA CA CA CA 'Si 'Si CA Co Co CA 'Si 'Si Co Co CA I I I I I 1 1 1 1 1 1 1 1 1 r....
IV
1' MA =- I-, CA ro In In CO 0, CO =-. MI CA In ..-- IN N. .,..- -o uo in ro 1- .4- co m ul co CA a co a
rl 4. in ro No co ,o ro ,.- r, to m m co m -- ,- c...1 ts., to a -- a a r4 SD PO *-- b71 '4" "0 CA
(-1 CA (Si M t54 Co M (Si Co Co Co ..-- Co (Si 'Si Co Co Co (Si CA 'Si Co CA CA M STA Si .1.... M4 STA r5:. it1:4 Co ...0 • • . • • • • • 1.11•••••• •••••••••1111 • • • a •
Co (Si CA Co M Co rA CA CA Co M M MI M M4 Co M) Co Si Co rA ISA Co Si Si Co CA M CA CA CA Co
1 1 I 1 - I I 1 ! I I i 1 1 1 I I C.I
MI ,-. CO No a uo -47 CO ,0 Cr. NI CA if) a CA rl '4' SA r, rl ,- MI CA ,- No r, -- -- -- -- uo 'Si
CA Co cq ..-- CJ 'Si ,-- N1 CA CA ..-- ro r4 CA ro n ro Co cl in m -- cl -- Co ro Co ..-- r4 (Si No -- P4 Co CA (Si CA 'Si Co M CI Co n Co (Si CA CA 'Si 'Si CA M4 (Si Co M MI In Si M Si 'Si to '51 m CD CA
• %IV •••• •1111 /1111%1 •••••• ■ ••••••• • .
Co CA Co Co CA Co Co Co CA Co Co M M Co CA Co CA 'Si CA CA Co CA (5) CA CA 'Si '5) CA CA (Si Co Co 1 1 1 1 1 1 1 1 1 1
CA 4- 0- -0 a r, uo 4- r, 07 1* 0, CO R. .- CA CA 4- cq Co ..-- co ON rq V) t (N rq 1.0 1'0 N) CO
CA .- Co IN h-4 -0 CA '1- -0 CA IN Co to (-1 4- ro n ro -- ro rq --- -- -- L') C4 Co ".•.. IN rq 4- a
,X CA CA (Si Co CA Co Co Co Co =.- CA MA Co M M M Co MI M Co (Si Co CA Co Co Co Co CA CA CA CA 'Si
. • • • • • • • . • 11••••• ••••••••••••••• •
CA Co Co CA Co Co CA CA 5) CA CA CA 64 CA MA (5) (Si Co Co Co rs4 CA CA CA MI CM (Si Co rA MA (Si CA I I I mi I I I I I I i I I 1 I
P4
4'4 M .cr r4 Mi .=1-
C <Ci7O C.) W LI. =C ...I 1= Z cn cn CL CI CJ C Q OP G.) 14./ 4. CO = 0-1 -J 3= 2C CD cn Cl Cl CN 0 0
••••4
lu 1-
129
Table 17 gives the unique factor loadings for each group:
note that there is a tendency for differences in these values between
paired scales to reflect differences in homogeneity coefficients
(Table 6), the more homogeneous of a pair tending to have a lower
unique factor loading, the tendency being more marked in the case
of larger differences (e. g. factors B and M).
The results obtained from SIFASP are generally consistent
both with analyses discussed in previous chapters and with certain
of Cattell's own published findings. For instance, Cattell, Eber and
Delhees (1973) reported a factor analysis of scale scores from four
forms of the 16PF, on a sample of about 600. This analysis showed
rather poor identification of factors E,M,N and 0. It has already
been remarked that factors E and N were difficult if not impossible
to separate by rotation of unrestricted solutions; further in the
restricted analyses the factor corresponding to N was not linearly
independent of the others, and elements in the residual matrices
tended to be larger for N than for other scales. In both forms,
alpha coefficients for these scales are very low. Factor M also
proved troublesome in rotation, perhaps partly due to the very low
homogeneity of this scale in form A, and also has some of the larger
residual covariances. The problem of separating C, O, Q4 and L,
the major components of the second-order Anxiety factor, has been
constant, and reappears in the restricted analyses; the problem
with L has no parallel in Cattell's own work, but Cattell, Eber and
130
di
0
N N 0 CO 7t1 C7, N • N ON M N M CO M N • .0 N L l Lf) Ul t*, N CO CO N N ,S) LC) • . .
0 0 0 0 0
•
0 0 0 0 0 0 0 0 0 0 0
O 1--I O CO t-r) CO N N N CO M N M CO ce) N
Lc) N Ul tn in 711 N. 00 co .0 N N Lc-)
• . . . 0 0 0 0 0 0 0 0 0 0 0 • 0 0 0 0 0
▪ N re) ON N M L.C1 N .14 N ON ON ,C) N Ui in in .70 t.r) N CO CO .0 N N Lt1
. . . . 0 C: 0 0 0 0 0 0 0 0 0 0 0 0 0
rn 0 ON LC) r•■4 .0 0 O N .4D co -1 en Ul s..0 N is) Ul No N co as co .0 N u-)
oo o o o o o o o o o o o o o o
CO Ul c.) as as as as o o cn CO N N E-1 N Ul .14 in cel N N ON CO s.0 N N S.0 Ul
CLI•
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4-1
E-4 N N N .0 co N 7t4 o co N in cn .0 N d4 cn „.„ .0 .0 No in in in d4 .0 r- co co .0 N N .0 in
0 0 0 0 O 0 0 0 0 0 0 • 0 0 0 0 0 Ti rzt 0 ,-,
N 0 N .0 r- 0 0, .0 cn c0 N c0 (/) .0 --I r-- In .7f1 .0 cil In N 00 ON .0 r-- N .0 in 0
O o
•
o o o o o o o o o o
•
o c o o u ai
't CO ON r'r) 7t1 C' in CO 0 0 •14 en 0 Ul N a) ▪ in .0 .0 in .0 cc*, N CO N CO .0 In
▪ 0 0 0 O. 0 0 0 0 0 0 0 0 0 0 0 0
co co ,c) o co con co co In N co ,t o o N '4) u> s0 Ul .7r di N co co .0 co co r Ul -
O o o o
•
o o o oo o o o o o o o
ON co .0 c0 co 0 0 CO 714 N r- s0 in in
•
N CO CO C7, .sp Ul N Ul in
0 0 0 0 0 0 • 0 O 0 0 O 0 0 0 0 0
N N m71,
7 L.1 z o aaaa U
he t
en r
est
ric
ted
facto
r
CO
2
NI M .c) cy, rs ce) O .0 Co -.0 1.11 N • • •
coo° cpc0 O
N CO C/) .0 ON 44 • • 0 0
N N in in N CO •zt, N Cs N to .0 in .0 co .0 CO in .0 .0 in in
. . . . . • O o
▪
o o
•
o o
▪
oo O o o oo o T
able
17 p
art
2:
131
N In
• O
N .0
• c0
0 co (0)
CO .0
coo
0 co
•11 in r-
c0
in .0
• O
co in
• o
;.4 NI in
ti .0
0 co
.0
.0 •
Os N
• Ln in
•41 N
-I .0
in .0
co in
0 4•J t.) 0 d o o o o o o 0 o yea
'0
in .0 co cOOM.00.000.0N.0
.0 N in N .0 on .0
ON in •i■,
U • . ..-1
0000000000 000000 4;-1.. cn N
,--4 .14 c. In N en .0 N N .0 F-I in .0
o N
• o .0
• cD co
• o in
o N o
in o
.0 o
in o cu -,-
O
ON •14 M ON 0 N O 7r4 co di cr• 7r1 r• in CO -10 in N Ln -0 co r- N in Ln Ln
. . • O o o o o o o o o o odo ocio
• 7,4 o •44 -0 0 •11
•
7t4 "44 NO cr. .0 co N it) .0 co .0 co in N .0
•.0 Ln
. • • oc o o o
ON N st) CO N as 714 o 0-- in co N • to co .0 .0 in N W in CC N co in .0 in .0 in
(
•
Soo° coo o
•
do do o 000
N CO CO ce) 111 .0 0 CO ON CO .0 0 .0 •--1 .0
•
cf. N In N in .0 co .0 N in N N .0
O do O o o O o o
•
oo o o O o
•
o p., 2 C\31---Crstf)N.0 ceIMONNNN.ONNN
• N N .0 .0 to N in -.0 N to CO in N in .0 in 0 . . .
O o o o o o o o o o o o d o o o
(1) N re) ,7t4 al° wrilox.4 zociacta
CC
facto
r lo
ath
132
Delhees had very similar difficulties with C, 0 and Q4. They
rotated their axes in such a way that corresponding factors were
linearly independent but had low loadings on the appropriate scales.
Overall, the results do not unequivocally either confirm or
disconfirm Cattell's model, rather they lend partial support. In
Cattell's favour we have the following points:
1. The majority of the analyses support about 13 common
factors .
2. Many of these factors can be rotated, in the unrestricted
analyses, to positions where they load principally
nominally equivalent scales.
3. The restricted analyses give a tolerably good fit to
Cattell's model, although only, in effect, for 13 factors.
4. The residual variances in the restricted analyses tend to
reflect differences in the reliability of scales, estimated
either by alpha or by alternate form coefficients.
5. It is at least conceivable that the closeness of the C, 0 and
Q4 factors is sample specific (and the same argument
might also be used for L and N), either because the items
in these scales do not efficiently discriminate between
the traits with British undergraduates, or because in the
present sample responses reflect state rather than trait
aspects of anxiety, which may be less differentiated, as
argued by Cattell, Eber and Delhees (1973).
133
Against Cattell we have these considerations:
1. There is a consistent finding that C, 0 and Q4 are virtually
inseparable, both in this study and in Eysenck's (1972)
reanalysis of some of Cattell's data, not to mention the
Cattell, Eber and Delhees study.
2. Factor N is so poorly represented, and has such low
internal consistency, as to be worthless.
3. The restricted analyses cannot provide a guarantee that
the model as fitted is the best or most parsimonious
possible.
4. Rotated solutions invariably collapsed when attempts were
made to force them to the complete Cattell model.
134
CHAPTER 9
The results of the restricted analyses in the last chapter are
consistent both with the homogeneity analysis of Chapter 6 and with
the results of the Cattell, Eber and Delhees study. However, in
one respect they differ from many of Cattell's published results,
namely in the factor correlation matrices. There is a strong
tendency for the correlations between factors to be rather larger
than in Cattell's own studies, a feature which has been noted in
particular in the cases of factors C, 0 and Q4 but which is wide-
spread throughout the matrices. To take but a single instance, the
correlation between A and H in the restricted analysis for the male
total group was 0.556, whereas the corresponding figure in Table
10.2 of Cattell, Eber and Tatsuoka (1970) is only 0.26. Now one
does not expect to find identical factor correlation matrices in
separate studies, but a further feature is present in the results of
the restricted analyses which is novel, or at least previously
unnoticed, namely that if the rows and columns of the factor
correlation matrices are suitably reordered, a pattern appears.
Table 18 gives one possible reordering of the matrix for the total
group, but essentially the same pattern is to be found in all ten
analyses. By and large the size of a coefficient is inversely
related to its distance from the principal diagonal. Were this
pattern to hold throughout we should have here an example of what
FOR THE TOTAL GROUP
CO
CO
0-4
CO CC LU
LU CC
CI cr
CJ cn • CD
I-
=1
1 g: REORDERED FACTOR CORRELATION MATRIX,
135
r. ro PI *- MA ON Oh Oh VO Oh M •■■ ,■ .0 Q CO cr ro m m Os CO CA *- PO in M Ps .0 Co MA
2 MA ,- cr ,- ,- cr un sO 1.11 CO ,- MA MA MA • • • • • • • • • • a • • •
14A MA MA MA CA MA MA CA MA MA Q MA MA MA CSi ,- 1 1 1 1 I I 1
Ps MA ...- CO 1.- of NI CO *- CV Q CA Ps N. Q CO ....1 CO CA CO .0 .0 ro -- CO cr ,to r4 *... ro
NI .Kr 1.0
UO UO .0 PI ..-- rA CO *- ON NO ,■ 04
• • • • •••••• It • • •
M M M MA MA MA MA MA MA MA Q MA MA MA MA 1 I 1111 1 1 I 1
cr to co cr NI .0 cr PO .0 CO cr m r. co cr r. ,o CO O. 0 cr co ,o un CO CO CO cr
CD *- CA 0, 0, Ul ISM NI CA CO CO PO • • • Is • • • • • •
MA MA MA MA MA MD MA CO CO MA MA Cal CO MA
C •4 CO Ps 0, NO CO Ul P0 cr un ro v m cr r. C4 0, ...- PI cr C4 ...- CO MA 0, CO MA PO
0114 MA Q MA MAMA MA MA **, MA ...... ... MA MA *- .. . • a • . . • • • • . •
MA Q MA MA MA MA MA MA MA MA MA Q *... MA MA 1 I I 1 I 1
ro M m cr Vra cr PO CA cr CO cr er PA *. CO CSy ro ct, m CO .7 CO *- N
h4 CA C4 C4 C4 MA PO PI MA rn r4 *- • • ••••••• ••••••
*A CA MA MA MA MA GA MA MA CA MA MA MA MA IA
11111 1 1 1 1 1
Ps *- cr CO CO CO CA .4" PO CO CO PO L7 CI. N1 Ps 471 .0 *- MA *A CP 1.0 CO N1
LI. .4" CO C4 PO *- N? sO CO **. r4 CA VD • • • • • It • • II • • • • • • •
Cal Cal 1,5;) Mi MA *A MA MA *A MA Cal *-4 IS/ rSi
1 1 I 1 1 I
CJ CO 1.0 - 0, N Ul Ps Ps CO CO r4 r4 O- 0- Li? *.zr LIMO, CO cr 4i -4c) Ul
UJ MA MA .0 Mk PO 1.0 MA sO PO PO .0 CO • • • • ••••••• • • • •
M CAM MA to MA MA MA MA MA MA CA MA MA MA
1 I 1 1 I
Ps CA LO .0 *- CO 0, CO MA Ps ON MI v "0 ■•■ NI -o CO MA MA P0 CA MA Os *- .0 CIJ CO 0, PI
Cr MA CO CA CO COCA cr ,o CO cr • • • • • • • • • • • • • *
MA MA MA MA MA MA MA MA MA MA MA Q MA MA MA 1 1111 I 1
N cr ro CO un CO co CO cr ro vo CO CI-
CJ PO -Cl od- r, LA C4 MA .0 0, cr *- AC C4 CO CO r4 NI CO .0 /JO ** CO .0
. • • • • • • • . • • • • • •
MA MA MA MA MA MA MA MA MA MA MA MA <A MA I I 1 1
N 0, cr CJ CO 0, CO PI If) cr ro 0, C4 ' N0 CO CO Ps PO U) r4 0, CO C4
CD *- CO *- CA N CO PO cr VO P0 CO CO CO *- Ul • • • •• • • •••• • •
MMMMM CO MA <A MA 4A MA MA MA MA
I III II 1
Cr` cr ,o cr CO -Cl CO Ps *- CO .0 LI 0,
PI CA VI CO SiM CO CO 00 Ps .D • CI. Zr C4 CO CN MA CO .0 .0 CO Ps r4 ** r4 CA MA 1.0 PO cr • • • . • • • • •• • II • •
M MA MA MA MA MA CSI MA MA MA MA MA 4A KA MA 1111
CA Ps CA cr m cr CA Cal 0, cr cr ,o ro m
cr 1.11 cr un cr CO co NI CO cr oN M CO 0, CN CA-- 0, SiM .0 CA CO CO CO CA CO CP -4) • • • • • • .1 • • • • • • • •
M 4.A MA MA MA MP MA MA MA MA MA MA MA MA
1 1 1 I
CA cr co m cr ,o cr ro ,o c. cr m -- CO -- co m cr m cr NO m -- ON ON 4... .0 PO MA
LA P0 cr m c, ,o MA MA *- C4MA 0, .0 *- • • • • • • . • . • • • • • •• •
MA MA MA MA MA MA MA MA MA MA MA MA MA MA 11A
1 I 1
,o cr CO co rq cr cr un L7 Ill CO Ps CO *- NI 0, .0 CO CO CO NI CO Ps NI P.0 ON N. .0 MA
mc cr m cr ro -- MA rA CA .0 Ps CO Ul CO cr • • • 0 • • • • • • • • • • • •
CO MA *- MA MA MA CO MA MA MA m MA MA MA MA MA
1 I 1 1
CV CO Sr cr N. -- Ps CA N 00 N NI 00 UO CO PO CA PO MA .0 .- cr In CA (N Ul Os MA .0 cr li.1 NI CD .0 MA in PO ....... CO ....- ....... CO un CO CO r4 CA
• • • • a • • • • • • a • • • MA ...* MA MA MA MA MA MA CLl MA IA MA MA MA MA MA
I I I I 1 1
MA CA .0 CO (N 0, cr cr r. cw r, In rA cr r, PS CO PI Os 00 In CA 6.0 Ps M ON U, CO CA CO UO a-
4 MA .0 ‘7" ...- MA MA *- CO MA cr CA CO ...- MA MA • • • • . • a • • • • • • • •
.... ID MA MD MA MR Mk MA Q MA MA MA MA MA MA CA
1 1 1
("4 cr ro 4 C3 2 CA CD CV CO AC CD LU li■ 04 CD
0.061 -0.086
0
136
Guttman has termed a 'simplex', a pattern characteristic of
correlations between items arranged in order of difficulty, and tests
in order of complexity (Guttman 1966). The order of Table 18,
however, ignoring B and I which play no part in the pattern and
L, N and 0 which have been shown not to be linearly independent
factors, is not a perfect simplex, particularly at the lower left and
upper right corners of the matrix. Nor is the pattern a straight-
forward circumplex, i. e. with the coefficients first falling and then
rising with increasing distance from the diagonal. There is however
an approximate circumplex in the part of the matrix between H and F.
The novelty of this finding might lead one to suppose that the
pattern is an artefact either of the way the restricted model was
specified or indeed of the SIFASP program used. Inspection of the
correlations between scale scores reveals some aspects of the
pattern, but it is much more clearly apparent when the matrix is
corrected for attenuation. The bivariate correction of a correlation
for attenuation due to unreliability has been in use for many years,
and consists of simply dividing a correlation by the geometric mean
of the reliabilities of the variables. Bock and Petersen (1975) have
extended this correction to the multivariate case, showing that a
maximum likelihood estimate of the correlation matrix of true scores
may be obtained by first correcting the coefficients in the usual
bivariate way, and then reconstructing the matrix from the positive
137
eigenvalues and associated eigenvectors of the matrix after
bivariate correction.
This procedure was applied to the correlations among scale
scores for the total group, using as estimates of reliability the alpha
coefficients discussed in Chapter 6. The resultant matrix is
presented in Table 19. Theoretically, this matrix ought to be
rather close to the factor inter correlation matrix, since both
procedures purport to remove error due to unreliability, and so it
proves to be. Moreover, it displays most of the features discussed
with regard to the factor analyses: note that L, M and N have much
the lowest corrected alternate form correlations - these were
troublesome in the rotation of unrestricted factor solutions, and had
the largest residuals in the restricted analyses; C, 0 and Q4 have
corrected intercorrelations almost identical with their corrected
alternate form correlations; correlations between E and N are close
to the alternate form correlation for N; G form A is nearly as
highly correlated with Q3 in both forms as with G form B; and
finally the pattern of correlations found in the factor correlation
matrices from the restricted analyses is present in very similar
form. The similarity of results obtained by these different
procedures suggests that we are dealing not with computational
artefacts, but with real features of the data. It also suggests that
the Bock and Petersen method would be a useful procedure to adopt
earlier in the sequence of analysis.
138
h,.- N) if, ..-. No if) No •-• -O ON rn rn cr Q- m No cr No o. CS CO PI tr to N. en PC 41 N) 47N
cr r. CO CA P. CO CO Ul P. tr ..-- C4 .-- CA CA P. CA CA M .- OD CA P. sO ON M sO C4 CO CA CA rC r,
CO CA CA Cs CA CA ,-- NI C4 40 CA CA CO =- .- 47 CA =- CA CO .- .--- Co tr =- tr Co N) CO CA CA sO CO • • • • • ••••••••••• ••••••••••••••••
Co CA CA CA CA Co Co Co CA CA CA CA M Co Mi ..-- CA CA CA CA CA Co MI CA CA CA CA to CA CA CA MA
1111111 I I 1 1 1 1 1 1 I 1 1 I I I
Q' Os =.- PC r4 P. cr r, ON cr un OD Co ,-- CA Cs Cs .-- CO P. ,-- 01-. CA 47 1.- T- P. OD Ps CA Cs P.
rC P, N) P, -43 P. PO Co Co r4 ,-- Ps Ps sO .- Co P. 47 sO ill CA ,- CA -43 SO Os CA CO .- ON Os 143 N) =
CJ CA CA tr CA CA OD ..-- N) cr C4 CA .4. C4 Co MA un M M in ,- 1..., .0 ,-. ,- .- tr sO 43 CA Co ON NO 0 • a • • • a • a • ft • .1 • • • • Isasswas• • • • • • • •• 4.4
Co M M M M CA CA CA M4 CA CA CA CA CA =- CA Co CA CA CA CA CA CA M Ma CA CA CA CA CA CA M 1 1 I I 1 1 ! 1 1 1 1 1 1 1 1 1 1 1 M
=.1
=- 141 =-. Ul Ps P. CA CA ,- 'V ON ,0 h... M T.. ." M h, rs M N) N) CO ..0 tr C4 =4' 0, .-- h. Os MA =
CA CO CO -0 CO sO CA CO CO r4 PC Ps cr -- m .- m r, -- -- -4- -o -- P. cr (•- -- rn -- r, m PO CU
C5 40 CA =- =- Ul =- cr n m M Cl .-- M CA CA =- 143 CA =-, CA 'T =- rn an r4 ...... T-. TT. M 00 CA CA al.-,
.1 • ■ a • • a • • • ■ a a • • a a a • • • • • • a • • • • • •• .1-,
CA CA CA Co CA CA (Si CA Co Co CA CA CO .- CA CA CA CA CA MA CA CA Co (Si CA CA CA (Si CA to Co CA 17.1
I 11111 1 1 1 1 I 1 I 5-
CA NO h. sO CO -0 -40 CA 143 r0 CA N M r... Co PO rs CA tr CP P, CO Ill CA CA 0. P, =- CA 00 OD =- 0
=- Os =4" CA Os P, N) cr PO .- CD Cs cr to -- ,o C4 .- tr =- tr -40 Ps -40 CA N) N) P. PN PN M MA tr 's-
CV CA Co .- -0 N1 tr P0 CA 43 N1 U, CA CA CA CA =- .-- (Si CA P. N) cr N cr cr 147 cr 04 Co =- •-• CA • . • • • . • . . • . . • . . . . • • • • • • . • . • • . • • • r
to m m CA CA CA CA CA M CA Co Co ....... M CA (Si Co CA CA CA CA CA CA CA CA CA CA (Si CA Co WA M 0
I I I I I I I I I I I I I I •.-4 ..1•'
*40 MA CA CA sO ,.. CA CO CO Os PO CA rs. .40 CO N1 ,- PC PC ON Os P1 P. CO 43 un (.4 r, 43 CO Cis rs U
N1 N. =-. CO Ps 143 -40 CO CA ,0 -40 CA tr tr Ps w- CO PC CO C4 0, .- =- <1' N1 143 tr Ps CA sO Os Ps Q1
C3 =-. CA CO CA =- =- !kr C4 PO CA N) Co C4 ..-- tr CO .- to P, N) ...- CA 47 CA N1 ..- =- 00 =- C4 143 -40 f...
•••••••••••••••• ••••••••••• ■ ••••• S-
(50 M M Co Co CA CA CA CA CA CA ,- CA CA Co CA CA CA CA CA CA CA CA CA Co CA CA M4 CA CA CA (Si 0
1 1 1 I I I 1 1 1 I 1 1 1 1 1 1 1 I 1 1 U
0-, Os .0 =- -40 Ps Co CA tr Cs CA NI CA Cl- ID Os cr No si r, un rs 00 N1 CA C4 cr PI 1.0 sO plo, cr al
-- ..-- 0. r, -- siD OD Cs CO CN MA ,0 0, r-, N. CA -0 CA Os CA cr CO MA Os =- cr r. -- an C4 =- cr 4-,
at =- CA rA OD sO rO 1.0 C4 N) N Co NO VI IN rq c..4 n to -- r, -0 N) 47 r4 .-- -Cl 1", .17. 47 C4 .- (Si M • . • • . . • • • . . • . • • • • • • • • • • • • • • • • • • • .1.4
M CA CA CA Co CA CA to CA CA ...- CA CA CA CO CA Co (Si CA CA Co 'Si (Si CA CA CA CA CA CA CA Co (Si r- i i 1 1 I 1 i 1 1 1 1 1 1 1 1 M
>
CA Cls cr CA CO CA CA tr WI CA Os Os N) cr tr "0 tr N) CO 143 Ps VI N. =- =.-. Os N P. Cr- .4] ,-- N) "m
CO CA Os N) tr ,-- CA CA tr MA Os -0 CO NI ...- +- cr m N) r, to so Os CO N1 =- rn SD MA =- .- N) 4-, • M Og ...- r4 ..-- N) rl cr n si (N c-.4 rn n c..4 rl r4 CA .-- r4 =- (54 ..-- CA CA P. N .- -0 CA..- .-- ...-I
• • • • • • • • • • • • • • • • • • • • • • •••••••••• :7,
M CA (Si (Si MI M M M M ..-- M MI M M M (Si M MM (Si (Si MI M (0 M M M M M M M M M E
I 1 I I 1 ! I I I I I ! I
cr cr PN .-- Co CO sO P. CA 4? tr CO 111 ..-- 0, -- cr r, co cr n -0 N 0. Os 43 00 rs CA tr PO CA 13.1
(0 cr 04 sO C.Is r4 CA tP Co cr co IN -- c.4 cg -r -- un co -- 1'0 CO CA CO 4 f4,-.1 10 .0 Os PC 143 C4 .4-'
-J CA (N Ul -o N) CA CA t5 to CA Ni PC 143 Co cr un -- -- -r un tr CA CA =- N C4 sO C4 PC .- N) sO 4-
. • • • • • . • •••••••• •••••••••••••••• to
M (Si CA CA CA CA CA MA =- CA CA (0 M M (Si M M M M M M (0 M M MA M 151 M (FA CA (5) CA 1 1 1 1 1 ! ! 1 1 1 1 1 1 CI
...! CC CA sO 143 Os Co CO Ul CA Ps cr r4 CO CA CA P. sO .0 r4 1.0 143 CA --10 sO N) Os Co .-- CA Ps r. r, cs o
N. so cr- N) 'r in- so '0 cr rl Cs CO N) CO CA N. un No -- CO P., ..... ...... PO in sO tr cr cr n cr un 5-
t 1a4 IN Cl CA r4 CA =- =- CA CA cr cl c...! cq n rn IN rn m rn CA CA =- =- 0-. C4 f.0 Ma cr -- CO rn rte) t 5.. •• ■ ••••••••••••• •••••••••••••••• C1 (Si m n n Co n n -- m so 151 m n to m (0 n to 1 to MA Co CA CA CA CA CA CA 15, CA CA (Si -.4
1.1... I I 1 1 I 1 I I I i I I i I 1 1 Ivo -t-
' O CO 00 =- Ps CA CA 40 sO CA 'Si CA sp. CA cr un Qs -- cr N N) Co PO C4 ..-- .43 =- CA cr cr 43 cr o
un PN .4' ... tr CA CA (Si CO CA CO -Cl cr co m un cr ON '0 N. 40 VI N) N) .- 40 Si 123 47 NI CA CO ..1--
MC cr -- cr N. r, so so -- C4 C4 41 cr N) cr -- rn un m N) un N. CO os -- n C4 PC 43 CA sO CA =-
•••••••••••••••• •••••••••••••••• W
M CA CA CA CA CA =- CA M CA CA CA CA MA CA CA CA 'Si CA MA CA CA CA Co CA CA Co Co MA Co Co MA J.= 1 ! ! ! ! ! ! 1 1 1 1 1 I ! -s,
Cs N) tr ON N) Co Co Co CO Co P, .-- -0 Ps Ps sO .- CA Co Ps CO Ps r. -- cr cr qn OD =- CO -40 Co 5-
.-- .-- 00 M 0, CA rA CI-. Cl T.- .0 Ul N) MI N) CO N) w- 00 Ill tr sO CA 143 ,-- tr Os sO c.-4 N) .-- cr o
CD CA CA CA N) CA CA CA =- C4 N) N) .- cr -- co -- n n r4 CA N) 0) Co Co CA 43 VO =- tr =- CO V? 4- • • • • • • . • • • • . • • • • . • • . • • ••••••••••
M CA CA CA CA =- CA CA CA CA CA CA CA CA CA Co Co CA Co Co Co CA CA CA CA CA (Si CA CA (Si CA CA X I 1 1 1 ! 1 1 1 1 1 I 1 I 1 I I I 1 I ..-1
5-
tr CA N) =- Co N) N CA CA OD sO sO CO P. CA =- N) N) sill P. C4 N) N. M M Os rs CO ON cr s0 47 4J
en 0, ,0 cr to Os '1- tr Os tr .- Is, N. -0 5., CO CA cr co un un -- (N N. ... CA OD 43 N) CO =- tr M
U.... cr rq CA .0 CA CA r. n rn .- No -- N) un r4 CA 43 =- Co NI 0, N) sO =- CA CA tr CA N sO CA =- E
. . . • • • • • • • • . • . • • • . . • . . • • . • • . • • • •
m Co m m -- CA CA CA CA CA (Si CA CA CA CA CA CA (Si CA CA Co CA 64 CA CA CA CA CA CA CA CA CA 1 1 I 1 I I I 1 I I I I I I 0
•tA
rl Ps Os CA ,-- Cs .-- 0, ..-- CA .-- CA sO 40 NI 40 143 CA CA OD .-- -40 sO PN 43 OD C4 CA PI rs CA Ps 4.:.
N) -Cl =- CA tr Co .- N) sO N) Ps 00 Cs CO .40 rs ,-- =- tr CO =- CA CO (Si .- -0 Cl) cr un cr ON T.-. M
ILI CA =- =- CA sO rn r, CA -40 CA 00 CA sO =- CA CA CA -.- CA 03 N cr un rn cr un N. N) 47 =- CA =- .-4
•••••••••••••••• •••••••••••••••• al
CA to Co -- Co to n (Si 10 M M Co CO n M n M M M MI M M M M M M Co M CA Co CA (Si 5- 1 1 I I I I I I I 1 1 I 1 I I 5-
O
cr -0 m CA N) cr CO 141 N. cr No CJ N. -- .- N) -O m CA Ps Ul Os rl 0- CA 00 00 sO N) ON N. Ul 5....)
PI ID M ....... sp 00 cr cs rq cs o. -- n No PN M U3 sO 43 sO SO CA sO cr cr cr No co CA h. CA sO
C-71 =- Co CA =- C4 CA tr CA Ul =- C4 CO =- .-- cr as -- CA OD =- CA CA 47 C4 Ul CA CA 00 CA N) li? PN ••
•••••••••••••••• •••••••••••••••• al
CA CA .-- CA CA Co (Si CA CA CA CA CA CA (0 CA CA CA CA CA CA CA Co CA Co Co Co CA CA CA M M M I ! I I ! ! I I I I I I I I 0
..-- m --.0 h. CA N) CO sO tr CN CN Co N) in ON .- ON r. un os N. co N) N) -- cr OD 143 in CO N -- ,J
cs Co un No ON T.-. N. to cr CA =-. r. cr co N) co CA CA =- CA CO ON co cr cr. cs -- N) cr N) r. Ll f-
pca m CA cA ..-- 04 Co -- 04 c...1 rl CA CA M CA CA CA CA Os =- .- IN CA CA CA CA .- .-- Si (N .- CA Co M • • . • . • • • . • . . . • • • ••• ■ ..•••••••••• CL
CA =- Co CA CA CA CA CA CA CA CA CA CA CA to CA CA CA CA CA ,:.. -7.. ..•.. CA CA CA CA CA Co M M M
I 1111 I I I I I I I 1 I cr
(Si =- tr N) cr cs ..00 m cr 0.1 ON sO C4 =-. .4" N. l'n IN 43 ON un cr cr No r, CO rs ON r, Co WI =-
CA Cs N) N) CA ,-- 43 Ps CA 00 =- N) ON 00 N. N un CJ m co -- un -- No r, un N -0 N. .- CA =- 111
'Z Co CA ,--- CA tr CA cr CA CA CA =- =- CA 43 CA CA CO- CA CA CA tr r4 tr =-- .- N .- .- CA Ps .- Co .-1
. • • . • • . ••••1•••• •••••••••••••••• An
T.. M to M Co M M M M M M M M M M (Si CA CA CA CA CA CA 'Si CA Co CA CA Co CA CA CA CA M I I I 1 1 I I 1 I I I 1 I P-
CV 41. CV N1 cr
<C LJ CL.1 1..1_ CD 7C t~ -J 2.C. CO CN CO CN CN .ocr i74 GJ La Li_ CD MC F-4 -IM.7. :it CD CD CN CN CN
• L f- 5- 5- 5- f-. 5- 5- f- f- r- r- 5- 5- C.. b1:1 C- 5- C- Y C- L L L C- C- L C- t- C- C- 00000000000000 00 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
V .0 4., 40 .0 4., 4-, -I-, .4., .0 4-, 4-, 4-, .1.) .4., .) .4.., .0 4-, -0 4., .1-, 4., -0 4-, 4-, 4., 40 U U U U U U U U U U U Li Li LJ Li u U u U U U U Li U u U U U
ID CO It CO Ca CO CO CCI MI CO CO MI CO C1:1 Ct CO CU CI 11:1 CO CO CO CO CO It It CO jO MI CO It 03 CO 11:1 • LL L 1J- 1.4.- Li- U_ U- U- Li- Li_ LL L U.- UL U- L L U- LL Li_ Li- IL Li- LL Li- L. LI- Li- Li- LL
139
.y.- 10 V, DO CA a- or. (-4 rn a r, -- m r, r4 No ri NCI NO 'Cr M T... in T... T... a c• co m N
a- -- un No .r- a Y41- CO U') (N VI ..q.. N. v- 1,0 mi N. 5-4 ro m un un rn C,4 N) ...- r ON el, rn rn R Co
cn Co Co r. ,- ,- rl -- rn No -- Co No Co Co -O OD Co Co OD Co r- r4 m N .7' ..-- a- r, -- -- m • . . • . . • • . • • . • • . • • .••••••••••• • • m m m m CA CA CA CA CA CA CA CA CA CA MA MA CA CA CA CA CA CA CA CA CA CA CA m m m
I 1 1 1 1 I I 1 I I 1 1
rn m N. (....1 ..-- (-4 a 10 Y- .- ON M M Ul in (Si In IM ID la (Si -10 Y.- a un No
cr. L1 r, N. N LO N. VI T.-. ON Cr, CO ON ON un o, r, r, -CI -o un Cl". ro PO
ON r- MA 10 CO CA VI ra -- ,- DO ....- CA 0, sO M M "0 (Si Oq sO CA rA LO ,0 Y- 00 ro 1'N
0
••• • • • • • • • ••• •••••••• 4 •• • ..-I
CA CA Co CA CA (Si CA (Si CA CA CA CA CA CA CA (Si CA CA CA (Si CA (5 CA CA CA Co CA CA ..P
1 I 1 1 I 1 1 I 1 1 1 M O
CA CO 0, CO V- r, ,,r -0 ,C1 CO Co r, m ra PO 'AD CO ,0 DO rp rn cr• N. m ct, m r-
CSA ?...2 !:.2
ii.
N -- VO N. VI VI CA PO r- N ,C1 (Si N. CN 0.4 f.., rl DO PO Cr' N) CO Yr rn -- rn
cn r, ,- rn -- No m ,- m CA 04 ..-- OD CA r4 -40 CA PI ....- DO un m m PO r-. 4=
••• • • • • • • • • • • hi • • • • IS • • • • • • , 4=
CA CA CA CA Cal CA (Si (Si CA CA Si CA CA CA CA CA CA (Si CA CA m CA CA CA Cal CSa CA CA CA Co CA m
1 I i 1 1 I 1 I I 1 1 1111 I 1 r-
V, bl ra rn 0'. -- a- N. CA ON 10 ,D r4 Y-- N. VI CA 1.11 0, 0, N') a- a a rp --- un m m No co o
-- r, a CA DO PO CA DO V.' Cf.. CA CA CA r. -- Cl- CA r, --• m cr.. CO CA a No CO ^O CA PO 1...., VI 4-
C) CA CA CA DO CA V' r4 .r. PO ',ID DO Yr CO CA r4 CA Y- CA ...... -.0 ....- CA Y.- CA C4 V, 'V- s1 CA Y- ...- Y-
•••• ••• • • • •••••• •••• ••• ••• • •• • • ,--
CA Co (Si CA (Si CA CA CA CA CA (Si Co CA CA CA CA CA CA CA CA CA CA CA MA CA Co (Si ....- CA Co Co Cl
I I I I I 1 1 1 I I I I I 1 .1-1 4=
In .0 NO:1M r4 CAN N PON .- CI, CO PO N) '•O to co -O ,o co ...- VI N CA 'AD DO 0, LI
PO CO a DO ,0 bO V- ,0 -0 r- N. r. rn -- co CSI ,r 0... n ...- in ON r, L1 oo -- rn cr. Q.
CD rMCOMIO4 ••-• 1.0 V' 04 •-• .7' CO IN .-- U1 O? r4 stalra ra CSI -o N) 04 ,- pp "0 5- • • ••• • • • • • •• ••• •••• ••• • • •• • r■
M M CSI M M M M M M CA CA CA CA CA CA CA CA MA MA CA CA CA (54 CA Cat CA CA CA CA 0
11111 I I I 1 I I 1 1 1 I I 1-1
CO CO r4 m -- co r4 a- 0.1 V... a- r, m co a- cl- o -a (.4 -o 0, TO r CA MA U1 N. a a cu
No Co Co C, CA 1.0 ro N "--1- N •-• CON Sif CO fM N IDNIO ..-- -.0 CCI CA CO m la co or. 4= = T.- C4 a- LO pa CS7 "0 IN r. -- a. ...... No rn CA CA a No un un -- Co rn m -- ill
• • • • • • •••••••. • • • • ••• • • • • C51! 10. a
•
m m CA (Si (Si Co Com ....I
m st m m Com (Sim m m m (Sim (Sim m m Com m m CA Y.-. CA 5-
I 1 1 I 1 1 1 1 1 I 1 I 1 1 I I I I I M >
Yr OD CO C, -.10 CS1 bl 0, CA DO 0, rl Y- V, a- a- a- ,-. ..40 DO CA V/ 111 CSI a N. ..- V) Y--. ,- •TI
ION at3. .0 Co DO VD .r. a- un N) -- N No m -- cr. N, .- ra N m -- Co DO 'AD N. CA CA 4= 1= T.. M DO r4 U1 CA ro r4 N -40 Y- UO ...- V• CA CA CA CA V) (.4 ra (..4 rn m 17 No m .-- ra -- --1
• • . • • . . • • • • • • • • • • • • • • • • •• :..-i
CA m Co m Cat (Si CSt Co Co m (Si m m Co m m m m m m m (Si m m Co m Co (Si so Co E 1 I I I I I I I I I 1
e....
r4 VI CA Y- 0, 0, ...... CA DO CA a- -- un -- CA DO 10 CA V... PO -4- m un rn DO VO CA CA r- CD
-J 1CA LO a- m m CA N. CA ..... VO a- (.4 -- a- -- m a- a- m ....- -- ra m m rn r4 (74 1..:1.-j 22 .7,T ,..-.
Cl'r. -4- ,- ,- -- in ,r m -- rn PO Co 0, CA VD CO r. Cr' co -- ra a (Si -- No 4=
• s • • • • 4 • s ■ • • • . a • • • • • • • • a a • • • • al
CA CA CA CA CA (Si CA CA Co CA CA CA CA CA CA (Si CA CA CA CA CA CA CA ...... CA Co Co CA CA (Si (Si
1 1 1 I 1 1 I 1 1 1 1 1 1 1 1 1
P4 LI PI Cl- Co r CA ND Y- VO Co CA ,10 DO a cq IN -.0 -- r4 Y- Co a to cr. ,- a C. ra rn ..:7 CI
a- a- CSt r, U1 ra N) C6 00 0, a- rq a- No cr. N CO MA '40 Y.- PI CA a- (Si ,- cr. CA ....- .- N) E ,--1 r4 rl M ...- CSi r- 0, r CA CA CA V' CS 4- ....- CA (Si CA N) CA Y-- CA PO VO CA VI CA CA IN r4 C1) .5- • • • • • • • • • • • • • • • • • • • • • • • •
O CSa Co M CSt M C5a (Si CA CSt (Si CA Co (Si Co CA CA Co Co CA CA CA CA r- CA (Si CA Co (Si (Si CA CA ..-4 U.. I I I I I I I I I I I ) I I I I Ill
a M ra r...... rl No r, r, DO OD CA NO DO 00 n ON Cr n CG ,- CI
..--- -0 CA N) ,- or. --- -o N. --1.1 ,C1 Cl. CA 0, OD DO (Si PI P: F*1 5; CO CO -.0 CA a- ro o-
..1..,
:a -a C5a un L1 -0 0, Yr ...... V) CA rte) Y-. a- a- (1.54 'Cr -a (Si m -- -- N --- -- un CA ro
. . . . • • • • • • • • • • • • • • • • • . • ti
Co C5a m CA m C51 m Co CG Co CSt (Si (Si (Si m m (Si (Si m m (Si Y.... CA (Si CA (Sito 1 1 I ! 1 I I 1 1
(Si CA Co (Si .--
1 1 1.-.'
.4- cr. -0 VO r, OD -0 D1 NO 00 PO 0, OD CO V' r..... -o -o CA m CA r-, un CA V' VO LO 0'11 r-
un m CA ....- 10 Y- .40 ..--- V, Y.- r4 r, -- a -- un m m bn -- -- Cat CA 0, PD m! VD C1
CD CA CSa CA a- rn CO m -- N m m a- -- No CSI ,- m --- la a- si m CA Y-. ra DO rl Y.- ,0 IN 4- • • • • • • . • • • • • • . . • • • • • • • • • •
CA CA Co Co CA CSa Co Co Co CA Co CA Co (Si MA CA Co (Si CA Y.-. t54 Co (Si Co (Si CA CA CA (Si (Si ›..:
I I I 1 1 1 I I I I I I 1 1 I • rf
VI r. un ,- N 00 VI CA CSi V, DO C, N. M Y- 0, a a- N) -- Co No CO a- rA ,0 4" ,0 h0 Ull •O .4"
...... -CO SO . If) '1" LM I", Co a Cr. No .- m -- -- (Si un (Si m 41 C, OD ...- 10 CA R. 0, VD U1 113
Li. a (.4 c4 r, cr. ra V, CA a ..-- -10 Y- VO a PO .,-- 10 ..-- Y.-. V- CA a- No -- m CA DO VD ..-- DO r..., -- T.- • • • • • • • . • • • • • • ••••••••••• 44 • ••
Co Co Co CA (Si Co Co CA CA (Si Co CA CA CSA CA CA CA CA Levi CA .r. CA Co CA CA Co CA (Si CA CA CA CA ..-
I I I I I I I I I 1 I I I I I 0 •1-1
0, 0, V, CO V, V, 0.1 1.0 a- un r, ON 00 Lit i'.. Cl. 00 n un n --- -0 0, Y.- 1.1 Y- -40 OD 0, -.0 -.40 .40 .1.,
CO fS! -4D CO LO 1.0 N. CO •--. N. (Si C.'S .0" CA CO 0, SifV- Co Li) 1.0 CO -0 0, N. rA N. (Si N) In In iri
LU Co ...... Y- CO NO N bn (....1 DO CA VI T.....Ca ....... ...... V,..! r- M (Si 'I" P.7 a- rn a un No ra Np -- Co CA ..-1 • . • • • • • • • • • • • • 1.• • • • • • • • • • • el
CA Co Co CA CA Co CA CA CA Co (Si (Si (Si CSI Co CA Si Co CA Y- Co Co CA CA CA CA CA CA CA (Si Co (Si 5- 1 1 1 I 1 I ! I 1 I 1 I 1 I -
ci
DO DO 04 N 'AD Co V" DO OD CO CA NO 'V T..... 00 ,0 ,- ON 14 IM l'O N. M -.0 10 a- cr. rp 0, CO -tp Li
CA Y- DO a CO oo No -- CO rn o... CO -- -- til -- CO un m V' CA Y- 0, CA V, ON n CN T- DO 10 ml
L) (N ..-- CO CA (Si 0.1 VO NI ..tr ,- ,- N. CA Y.- LO CO — (Si CA Co Y- ..-- Yr CA a- (Si a- co -- N) -.0 CO ... • • • • • . • • • • • • • • • •••• •••• • • • • • • in
Co CA CA CA (Si CA CA CA CA CA CA CA CA (Si Co Co CA CA ..-- (Si Co Co CA CA (Si CA Co Co CA CA CA Co 31 1 1 1 1 1 1 I 1 1 1 1 1 4=
rl V, Co CA VI CA ...... CA r. ND ,1 N) IN N -- a- ra m or. 7 a- a. CCI N m a- a- No un No r, ('I Si-'
Cq M 40 T- 'ne. T.- ON '40 U0 (Si (Si In a. -- No (Si a- m un Ca ,- a- CO Co Y.- OD Yr N r4 to ra 5-
OA M CI, M ..... r- (Si (Si M T- MI Co M M Co M M M M M T- M CAIS4 M (Si Co MI (Si (54 Co m RI • • • . • . . • • • • • • • • • ••• • • • • • •• • • •• CL
Co CA Co (Si CA Co Co Co CA Co CA CA Co Co CA Co Co YY CA C5a CA Co Co (Si CA CD CA CA CA CA (Si I 1 1 I 1 I 1 1 I I 1 I 1 ar
ND ON sO Ill C.) ,.... ON sO a- a- a- ,- V, Co Cr'. '40 CA VI r- CO a- CO IN -- a- co N.., (Si NI 0' SO
If) Co 10 ..... Co NI a- to ,- a No oo .r- r, un m Co a- CO IT -- ,- (-4 ra m (Si Co No N, Co rl
.0 ON CP ...... 0.1 UO CA 1.11 PI r- CA CA qr. Y- Ul CA ..... CA CA ...- 41 r- rA ....- (Si CA CA 4- ND CA Co ,....f W
• • a • • • • a . • • • • • •• •• • •• • • • • ••• •• ..0
M IA Co Co m Co M Co Co Co CA M M Co Co M ... (Si CA CoM Co CA Co Co M M Co M M M M M
I I I I I I I I I I I-
¢ OD L.) 4J 4- CD 1-4 Co CI CD CI ./Z C:0 C..) LLJ 4. T -J Cl
• = L L L L L L L L L L L L L C- il4 L L L = L = L L = 5- =
O 0 O 0 CP 0 0 0 0 0 CD 0 CI 0 0 CI CD 00 0 C) CI O 0 0 CI 0 00
/C • .,.a 4-, 4-4 4-, 4.2 .1-2 4-, 4-3 4-3 4.2 4..2 4-2 -I-2 4-, 4.2 4., 4-3 4-2 4.2 4.2 4_a .4.2 4.3 4.3 4-2 4.2 LS u U U U U L) U U U U C.I U u LJ U L U LJ LJ u U LS u U Li U U U U
0 Co (0 (0 RI (0 (0 CO CO (0 (U IL (I3 IV (0 (0 (0 0 03 03 An IV (0 (0 IV f0 IV CO (0 RI CO LL L.L. U. U. U. U- LL. U. U. L. U.. U. U. 1.1. U.. U. U. U.. U. U. Li. U. LL. Li.. L.L. 1.1. U. Li. Li. 1J- LL.
0
Fac
tor
n?
140
Since in the analysis of scale scores we are dealing with
composites formed from scores on mutually exclusive sets of items,
let us pursue the questions raised by these analyses to the item
level. Remembering that Cattell has consistently claimed that
items loaded by a given factor are designedly hetergeneous, a
claim vigorously endorsed but otherwise interpreted by his critics,
the correlations between items and scales for the items making up
scales A, E, F, H, Q1 and Q2 were inspected in the light of the inter-
pretations proposed by Cattell for these scales. These six were
chosen partly because none of them had been called into question
in the analyses so far, and partly because of their consistent place
in the pattern of Table 18.
Firstly, an interpretation for each scale was formulated on
the basis of its most distinctive item content in the light of Cattellts
own interpretation. Then each item in these scales was checked to
compare its correlation with the composite formed from all other
items in both forms of the same scale with its correlation with all
other scales (again at two-form length), excluding L and N on
account of their uniformly poor showing throughout. The items
were then sorted into four categories:
1. Where the proposed interpretation clearly fitted item
content and the item-partial scale correlation was higher
than the correlation with any other scale.
141
2. Where the item-partial scale correlation was higher
than the correlation with any other scale, but the
interpretation did not clearly match.
3. Where the interpretation matched, but the correlation
with another scale was within + 0.03 of the item-partial
correlation.
4. Where the interpretation did not match, and the item-
partial was exceeded by the correlation with another
scale.
In a small number of cases the interpretation matched, but the best
correlation was with another scale. These items are discussed
individually. In what follows, a typical 'matching' item is given
for each scale, and where an item correlates about as well or better
with another scale, this scale is given in brackets.
Factor A
"If I had to choose, I would rather be
a. a forester
b. uncertain
c. a secondary school teacher" Item 51 form A
Interpretation: preference for occupations and leisure activities
which essentially involve social contact.
142
1. Form A:
Form B:
3,26,51,76,101,176
3,16,51,52,101,126,151,176
2. Form B: 76 (would prefer to speak to a stranger in a railway carriage rather than stare out of the window)
3. Form A: 126(I)
4. Form A: 27(H), 52(Q2)
Form B: 27(I)
Item 151 in form B correlates more highly with I: the choice
here is between being an artist and being a secretary running a
club. I and M typically show strongly when artistic or abstract
intellectual content is involved.
Factor E
"It is more important to:
a. get along smoothly with people
b. in between
c. get your own ideas put into practice"
Item 57 form B
Interpretation: socially assertive
1. Form A: 6,7,56,156,181
Form B: 6,31,57,106,131
2. Form B: 56
3. Form A: 131(F, Q3), 180(H)
Form B: 7(Q1)
143
4. Form A: 31(Q1), 32(H), 57(H), 81(G), 155(F)
Form B: 32(0), 81(H), 155(Q1), 156(Q4), 180(I), 181(H)
Item 106 in form A fits the interpretation but correlates better
with H; the interpretation of items 180 and 181 in form B is
dubious, but on balance they have been placed under 4.
Factor F
"I am well described as a happy-go--lucky, nonchalant person
a. yes
b. in between
c. no "
Item 183 form B
Interpretation: Socially active, carefree and optimistic
1. Form A: 58,83,108,132,133,157,158,183
Form B: 8,33,58,83,132,158,183
2. Form A: 8
3. Form A: 33(H), 82(H, Q2), 107(H, Q2 )
Form B: 182(E, G, Q3)
4. Form B: 107(H), 108(H), 133(A, I),157(A)
Item 182 in form A, "I am considered a very enthusiastic person",
correlates better with H, but in terms of content is rather more
like the interpretation offered. Item 82 in form B is difficult to
place between F and H, and in fact correlates better with H.
144
Factor H
"'Stage-fright' in various social situations is something I have
experienced:
a. quite often
b. occasionally
c. hardly ever "
Item 85 form A
Interpretation (for low score): feels threatened when the focus of
attention in social situations.
1. Form A: 10,35,61,85,110,161
Form B: 35,61,85,86,110,136,161,186
2. Form A: 86,111,135
Form B: 10,36,135
Note: these items are very typical general extraversion items, e. g.
item 135 form A: "I consider myself a very sociable, outgoing
person. "
3. Form A: 60(E), 186(Q3)
Form B: 60(0)
4. Form A: 36(F), 136(F)
Form B: 111(F)
Factor Q1
"I think society should let reason lead it to new customs and throw
away old habits or mere traditions:
145
a. yes
b. in between
c. no"
Interpretation:
Item 169 form A
socially and politically radical
1. Form A: 46,120,169
Form B: 70,120,145
2. Form A: 70
3. Form B: 95(G), 169(E)
4. Form A: 20(I), 21(Q3), 95(E), 145(E), 170(G)
Form B: 20(H), 21(M), 45(Q3), 46(M)
Factor Q2
"To keep informed I like:
a. to discuss issues with people
b. in between
c. to rely on the actual news reports "
Item 96 form A
Interpretation:
and behaviour.
does not look,to social group for norms of opinion
1. Form A: 71,96,122,146,171
Form B: 72,121,171
2. Form B: 47
3. None
4. Form A: 47(H), 72(E), 97(A, H), 121(E, H, M)
Form B: 22(H), 71(H), 96(C), 97(H), 122(F), 146(Q3)
146
Item 22 in form A: "Most people would be happier if they lived
more with their fellows and did the same things as others", has
very low (less than 0. 1) correlations throughout.
In general, when an item did not fit well the proposed inter-
pretation for the scale on which it is scored, its highest correlation
could be located by consideration of its content in the light of the
interpretations of the other scales. For instance, item 31 in
form A, which is scored on E:
"An outdated law should be changed:
a. only after considerable discussion
b. in between
c. promptly"
was clearly more like the typical items of Q1, and indeed proved to
be more highly correlated with Q1. When such a relationship was
not apparent, H tended to be the scale giving the highest correlation.
Items with an artistic content tended to correlate well with I, those
with abstract intellectual content with M, and those with moral or
ethical content with G and/or Q2. The majority of items
categorised under 4 above fit the interpretation proposed for the
scale with which they correlate best.
This, admittedly subjectively based, analysis allows the
drawing of finer distinctions amongst these scales, as follows:
147
(1) A, E, F, H and Q2 all refer to the behaviour of the individual
in social contexts, whereas Q1 refers to attitudes to social and
political ideas and conventions.
(2) A, H and Q2 are in a sense "afferent" factors: they refer to
the response of an individual to or in a situation, casting him in a
relatively passive role, whereas E, F and are more "efferent",
referring to the extent that he initiates activity or change, i. e.
casting him in a more active role.
(3) A refers to a preference for situations involving social
contact, Q2 to a certain dependence on socially defined reality;
H contrasts confidence and shyness but does not imply a tendency
to initiate social contact or activity, whereas F implies a readiness
to initiate and innovate, but not necessarily to dominate or aggressively
maintain one's position, which is the major burden of E. Q1 refers
more to ideas and opinions than to actual behaviour.
Thus interpretations modified only in detail from those
provided by Cattell fit rather well 67 of the items scored on these
scales, only 11 items which correlate best within the scale on which
they are scored do not fit the interpretations, 13 fit the interpreta-
tions but correlate about as well with another scale, only 3 fit the
interpretation but correlate better with another scale, 3 are doubtful,
and the remaining 41 correlate better with other scales and by and
large fit the interpretation of the scale with which they correlate
best rather than that of the scale on which they are scored.
148
Now it may be that what is at work here is Cattell's (1972)
claimed suppressor action, whereby items not strictly loaded by
a given factor are nonetheless scored on the corresponding scale
to balance out unwanted variance introduced by other items. On
the other hand Cattell has nowhere indicated just which items are
supposed to be suppressing what. It seems likely that rescoring
of the items in line with the kind of content analysis just attempted
could lead to a clearer structure, increase the statistical homo-
geneity of the scales, which as has been shown relates rather well
to their factor validity, reduce the intercorrelations of the scales
and maybe even improve the pattern of Table 18. However, such
a reanalysis would require a separate study to avoid the possibility
that the above results are adventitious.
What can be said is that such an approach to the scales via
content analysis and item-scale correlations strips them of any
mystery which may be implied by their origins in factor analytic
research. There is for each of these scales a core of similar
items which tend to be those with the highest correlations with the
scale. Eysenck's (1972) comment that Cattell's scales are complex
in content, again a point with which Cattell has often implied
agreement, is true only to some extent, and the extent to which a
scale seems complex is 'dependent on one's interpretation of item
content. For instance, the items of factor A are by and large
149
concerned with occupational preferences, but the overall contrast
between each pair is clearly related to the amount of social contact
they involve.
For these six scales, then, there appears to be a clear but
imperfect link between item content, item-scale correlations and
slightly modified versions of the interpretations offered by Cattell.
The remaining scales all have similar clusters of core items closely
related to their interpretations, but their poorer homogeneity and/or
their high correlations with other scales make the comparison of
item-scale correlations less promising.
Beyond item content, there is the further question of item
format, and in particular Cattell's provision of three response
possibilities for each item. The second of these takes various
forms: often "in between" is used, or "uncertain". The instructions
for the test refer to these second alternatives as "middle, 'uncertain'
answers", and exhort the subject to choose, wherever possible,
either the first or third, falling back on the second only about once
every four or five items.
Percentage figures of responses to each item show that by and
large participants in this study did as they were asked, but there
were large differences in the use of the middle category. For
instance, 51.41% of answers to "I make clever, sarcastic remarks
to people if I think they deserve it" (item 7, form A) were
150
"sometimes" rather than "generally" or "never", whereas only
9. 56% answered "I occasionally tell strangers things that seem
to me important, regardless of whether they ask about them"
(item 131, form A) with "in between" rather than "yes" or "no",
with the remainder about equally split. Yet the former has a
higher correlation with its scale than the latter.
One type of item seems particularly prone to large numbers
of middle responses, viz. those with alternatives "often; sometimes;
never", "always; occasionally; rarely" and the like. Twenty-
eight such items are distributed across the two forms: of these 6
are scored on factor C, 1 on E, 1 on G, 4 on H, 1 on I, 1 on N,
5 on 0, 2 on Q2 , 2 on Q3, and 5 on Q4. Twenty-three of the
twenty-eight have response percentages above 30 for the middle
answer, rising as high as 63. 39% for item 123 in form A. These
items are with only two exceptions the most heavily endorsed in the
middle category in each scale; item 124 in form B is one of these
exceptions, only 13.01 per cent using the second alternative, but
here the choice is "occasionally; hardly ever; never" and over
70% choose "never"; the other exception occurs in factor C form B
where item 104 (26. 59%) is beaten into third place by item 5 (which
itself has "occasionally" as the middle response), and item 4 (a
mere 20.96%) is fifth after an item with "in between" as the second
alternative.
151
However, these items show no consistent deficit in their
item-scale correlations, nor do items with "often" or "occasionally"
in the body of the item show any special statistical features. Such
terms are of course eminently open to interpretation by the
respondent, and even "never" and "always" may well be subject to
fine distinctions: we are of course dealing with self report data,
which carries no warranty as to its accuracy, via multiple-choice
items which may not allow the individual to specify his most
preferred response.
This brings us to the troubled subject of the relevance of
questionnaire items to individual respondents. Item content which
touches on issues of central concern to one individual may seem
hopelessly irrelevant to another on account of its specificity, yet
if items are written at a more general level they may fail to capture
the distinctions which are most relevant in any given case by being
too all-embracing and thus impossible to answer decisively. In
other words, not only may "often", "rarely"and the like be
interpreted differently, whole items may be relevant in different
ways or not at all and be subject to different interpretations
accordingly. Bem and Allen (1974) have had some success in
improving the performance of personality test scores by the simple
means of asking subjects whether or not they regard themselves as
consistent in particular respects and whether they find various sorts
152
of item content sufficiently central to their lives to be interpretable
as test material. One might read the middle response categories
in the 16PF items as providing something of the same kind of
information, although it is not treated separately in the scoring
procedure, which also does not make provision for the omission of
items.
If one accepts that the process of answering a questionnaire
item engages conscious processes of thought, whether they be
attempts to remember past behaviour, predict future behaviour or
articulate a self-concept within the framework provided by the item,
then scores must reflect not only patterns of overt behaviour (indeed
they may not even do so), but also the processes of interpretation,
recall and self-presentation. Cattell shows some appreciation of
this fact by referring to questionnaire scores as reflecting the
"mental interiors" of factors, and it is of course a major focus of
personal construct theory in the tradition begun by George Kelly.
Similarly, any attempt to detect faking of response and social
desirability response sets implicitly recognises the scope for
interpretation of items and self-presentation through responses.
If the involvement of conscious processes of interpretation and the
like is seen not as an aberration but as part of the normal business
of answering a questionnaire, then one would expect consistency in
the individual to emerge through the sorting of items into groups
having subjectively similar content. The form in which responses
153
are recorded is far from ideal for the detection of such consistency,
but the evidence of the analysis of item-scale correlations above
suggests that the participants in this study overall sorted in a way
more consistent with the author's interpretations of the items than
with the scoring key provided by Cattell. Perhaps the fact that
many of the analyses reported herein are amongst the most favour-
able to Cattell in the history of research on his scales outside his
own laboratory reflects the homogeneity of the sample, a sample
with, one might assume, a fairly uniform life-style and considerable
commonality of outlook and experience.
To develop this point of view further, consider first of all
the fact that in scoring a questionnaire much of the specificity of
item responses is discarded, items being grouped according to the
author's notions of similarity, however derived. Next, assuming
for the moment the validity of such groupings, the further grouping
of scales into second-order factors implies a further loss of
specificity. Should this specificity be regarded as unreliability in
the sense of random error, or information which is valuable for
some purposes but irrelevant for others? Eysenck (1972) sees the
specificity of Cattell's first-order factors as unreliability but hints
that in the case of those scales which are loaded by the second-order
Exvia factor there may be reliable specificity, quoting Frenkel-
Brunswik (1942):
154
"Different classes of behavioural expressions were often related to one drive as alternative manifestations of that drive ... One drive variable may circumscribe a family of alternative manifestations unrelated to each other: the meaning of the drive concept emerges in terms of families of divergent manifestations held together dynamically or genotypically, though often not pheno-typically. "
For comparison with Eysenck's Table 2 we have Table 19, which
shows rather more clearly that twelve of Cattell's scales have
sufficient specificity replicable across forms to suggest they may
have some utility.
However, if one is interested less in the prediction of
individual behaviour and more in the physiological correlates of
personality, it would seem reasonable to look to a level of greater
generality rather than a level which is sensitive to the cognitive
processes of the subject, and of course it has been at the level of
second and third order factors in Cattell's terms that the greatest
success in finding physiological correlates has been achieved
(Cattell and Warburton 1967).
What then is the psychometric status of Cattell's second
order factors? Table 18 shows rather little evidence of a cluster
structure amongst the first order factors, although this might
change after rescoring along content-analytic lines. The largest
steps between adjacent coefficients are between H and C and
between G and M, which might be read as marking the boundaries
between an extraversion-like and a neuroticism-like factor in the
155
first case, and Cattell's second order superego and independence
factors in the second. However the evidence is far from compelling,
and attempts to rotate principal components of the matrix in Table 18
showed no clear hyperplanes using graphical techniques. In any case
the step sizes are not all that large and their presence could as well
indicate missing factors as distinct factors at a higher order.
More in line with the interpretation being developed would
be Gray's (1973) attempt at a reinterpretation of Eysenck's factors
in terms of three major emotional subsystems, represented by
distinct neuroanatomical pathways and emerging in conditioning
studies and social and psychiatric aspects of behaviour.
Particularly if the rescoring suggested above were carried out, one
might imagine that H would form the core of Gray's dimension of
"susceptibility to punishment", occupying as it does a place between
extraversion and neuroticism type factors. Certainly the central
item content of H reads very much as susceptibility to perceived
threat in social situations. Most likely candidates for his second
dimension of "susceptibility to reward" would be E and F, with
their connotations of approach, initiation of activity etc. For Gray's
third speculative dimension, which he tentatively identified with
Eysenck and Eysenck's (1968) dimension of psychoticism, the
obvious candidates by exclusion would be G and Q3, and indeed
Montag (personal communication) has carried out a factor analysis
of Form A of the 16 PF together with the three Eysenckian scales
156
and found that these two are loaded by the same factor as
psychoticism.
It is not being suggested that Gray's three proposed
dimensions can be conclusively shown to have a clear place such
that they will naturally emerge from questionnaire data. Rather,
Table 18 suggests that second order factors could be placed more
or less at will. If at the physiological level there are three (or
however many) distinct subsystems which have an important role in
generating the variety of human personality, the evidence from
this study is that they do not appear in questionnaire data as the
dominant features. However, the item-scale correlations do
suggest that participants in this study effectively sorted the items
by content, and put for instance item 157 in form B ("It would be
more interesting to live the life of a master printer rather than that
of an advertising man and promoter") in factor A (correlation of
0. 352) rather than F (correlation of 0. 239) where it is scored.
In the item content of the proposed candidates amongst Cattell's
scales for Gray's dimensions there is a good deal of similarity
with his descriptions of behaviour under the control of the three
subsystems.
The similarity for the Flight/Fight subsystem is at best
slight, but many of the items of H and C read plausibly as referring
to behaviour in response to conditioned stimuli for punishment and
157
nonreward (the Stop subsystem), and of F and E to behaviour in
response to conditioned stimuli for reward and nonpunishment, the
Approach subsystem.
Thus what is being suggested is that there is a role for both
the cognitive processes of the individual and biological variables in
a comprehensive account of those aspects of personality which are
captured by the scales of the 16PF, at the cost of denying Cattell's
primary factors their ontological primacy. The first order scales
may be viewed as attempts to measure interactions between
relatively primitive emotional mechanisms and hypothetical
situations, mediated by the cognitive processes of the individual to
a greater or lesser extent, on the assumption that a majority of
individuals in a given cultural or social group will construe each
item in a similar way. Such an account draws together aspects of
the determinism implicit in learning theory, physiological approaches
to psychology and the trait tradition in the study of personality on the
one hand, and the emphasis on individuality, purpose and the central
role of each person's cognitive apparatus and strategies
characteristic of the work of Kelly and his followers, and more
recently Harre, laying the ultimately inevitable imperfections of
psychometric techniques at the door of the irreducible individuality
of men and women.
158
The question "how many factors?" then becomes not an
ontological question of great scientific significance, but a practical
question: how many factors are needed to preserve as far as
possible the greatest amount of information about individuals whilst
still preserving comparability of scores across individuals? The
answer to this will depend not on some fixed rule but rather on the
purpose for which a scale is employed and the homogeneity of the
subjects involved.
Such a scheme has at least a superficial similarity to
Guilford's (1956) structure of intellect model, with three major
classificatory factors: situations, interpretations and responses,
with the first a function of the forms of social life embedded in a
given culture, the second classifiable only insofar as individuals
share common perceptions of the possible range of significance of
situations within that culture, and the third possibly corresponding
to a scheme such as Gray's.
The greater utility which has been claimed for Cattell's
first order factors will then be a consequence not of their peculiarly
preeminent causal status but of the extent to which they embody trait
x situation interactions in their item content in a way which maps on
to the constructs of the subjects. More precisely, what is proposed
is a trait x situation x interpretation interaction model not unlike
Argyle's (1976) account and consonant with Cronbach's (1975)
discussion of aptitude-treatment interactions.
159
The evidence of this study is that to a moderate extent
Cattell's scales capture sufficient commonality in such interactions
to make them potentially useful instruments at least with under-
graduate subjects. In such circumstances, it is not circular to
ascribe explanatory status to scores: to say that an individual avoids
public speaking because he is low on factor H may be no more than
shorthand for "he is rather more than averagely averse to perceived
threat in social situations, and he perceives public speaking as a
potentially threatening situation", but it is useful shorthand insofar
as it brings a variety of possible situations under a general rule, and
distinguishes this individual from others who avoid public speaking
out of lack of interest or a preference for asserting their importance
by way of political intrigue.
However, to the extent to which a person is idiosyncratic in
his perceptions and to which a given situation is novel, scores
derived from a questionnaire will lose both explanatory value and
predictive power, since ultimately the method depends on eliciting
self-report of typical behaviour in classes of similar situations.
The scope for change in an individuals score on such a test will
therefore be twofold: he may become more or less sensitive, say,
to threat, or he may change his ideas as to what counts as threatening.
160
To summarise this speculative account, based tenuously as
it is on psychometric evidence from a survey, it has been suggested
that trait-situation-interpretation interactions are the stuff of
personality assessment. Accurate prediction will involve knowing
the situation, the general class into which an individual would place
such a situation, and his typical behaviour in situations belonging to
that class, and even then one cannot discount the possibility of
dynamic change in any of these once the situation is real and he
must act rather than predict what his action would be. The
remarkable thing about personality questionnaires is not that they
work well, which on occasion they have been known to do, but that
they work at all.
161
CHAPTER 10
This study has yielded results some of which were not
anticipated at its inception. From a technical, psychometric
point of view, the computation of measures of internal consistency
and their comparison with alternate form correlations suggested
that the scales of the 16PF were a good deal more homogeneous
than either Cattell or his critics had led one to suppose. With a
small number of exceptions, they were quite as homogeneous as
one would expect for scales of the same length, and the close
comparability of coefficient alpha and alternate form correlations
goes contrary to Cattell's repeated claim that the items are arranged
in scales which are parallel across forms but internally hetero-
geneous. The equally close correspondence between coefficient
alpha and the factor pattern coefficients in the restricted factor
analysis (correlating 0.92 in the total group) adds to the evidence
for the value of internal consistency as a measure of scale adequacy.
In other words, both within-scale and between-forms measures
agree as to which scales are best. Given these results, the strong
similarity between the factor correlation matrices from the
restricted analyses and the scale inter correlation matrix after
multivariate correction for attenuation is perhaps more interesting
as confiimation of the utility of the latter procedure.
162
The unrestricted factor analyses proved in the end rather
unproductive. As expected, splitting the sample by sex and
discipline to decrease the effects of correlated error made rotation
easier and yielded a somewhat better approximation to the model
under investigation, but the location of the source of collapse of the
factor intercorrelation matrices when targeted rotation was
attempted proved troublesome. However, the strong similarity of
results from a variety of procedures suggested that given relatively
reliable variables the controversy over factoring methods, criteria
for retention of factors and rotational techniques is sterile.
The use of restricted factor analysis was more informative,
and so far as is known this was the first time the method had been
applied to Cattell's scales. The extravagance in computer resources
of the program used prevented an ideal sequence of analysis, but
nonetheless a consistent pattern of results emerged from all ten
analyses. Investigation of the rank of the factor correlation
matrices from these analyses pointed up some of the problems of
the numerical accuracy and idiosyncracies of particular
computational algorithms for matrix manipulation, even at 72-bit
precision. Having settled on a rank of 13, however, reanalysis of
the results of unrestricted analyses using targeted rotational methods
gave a structure close to that obtained from the restricted analyses.
163
The restricted analyses proved rather insensitive to the
presence of correlated error variance, which was so problematic
in the unrestricted analyses. By far the largest source of residual
variance was the unreliability of individual scales; off-diagonal
residual covariances were generally small, with a very slight
concentration in some of the least reliable scales. The results of
these analyses were consistent and consistently interpretable in
very similar terms to those of the other analyses.
Perhaps the least expected finding was the pattern
demonstrated in Table 18, suggesting that the correlations between
factors are systematic, but not necessarily in a sense that makes
higher-order factoring an apposite move. There is a possible link
here with the failure of attempts to fit Cattell's second-order factor
model to the scale covariance matrices using J8reskog's program.
This pattern calls for investigation and replication on another
sample.
The results with regard to individual scales are better
treated out of their conventional order.
L and N find no support in this study. Their homogeneity
is poor, their alternate form reliability equally so, and independent
factors to.represent them consistently failed to emerge. From the
factor correlation matrices, L seems to be a mixture of E and C,
164
whilst N is largely E, with in each case a large component of error
variance.
B can be largely ignored: as a short intelligence scale it
performed badly, as might be expected on such a selected sample.
M is another scale which is poor in terms of reliability,
although the form B scale is better than the form A in terms of
internal consistency. The restricted factor analyses placed an
independent factor on this pair of scales, but clearly they are
unsatisfactory and need more development.
C, 0 and Q4 have been mentioned together time after time.
At best only two linearly independent common factors could be found,
and even they are very highly correlated. A number of possible
reasons for this are available. Firstly, it may simply be that in
this sample the subjects really did have closely matched levels
of three separate traits. Possibly the distinctions implied in the
differing item content were not relevant to them. Again, these
scales may find their place in the diagnosis of disorders not
represented in the sample, or in the interpretation of scores of a
minority of individuals who score inconsistently on the three.
Perhaps the scores in fact reflect state rather than trait aspects of
anxiety as a consequence of the conditions of testing. Finally, if
one is unwilling to relinquish Cattell's attractive distinction between
165
ego-strength, guilt-proneness and ergic tension, one may look to
improvements in the item content to reinstate them as separate
factors. The evidence of this study is, however, that they are all
virtually measures of the same thing.
G and Q3 may also be treated as a pair. They are distinct,
but factor G form A is rather closer to factor Q3 , which made for
problems in the rotation of unrestricted factor solutions. Internally,
the scales are fairly homogeneous, and there is no real problem in
accepting them as potentially useful scales which doubtless could
benefit from more development.
I was the least troublesome of all the factors: it appeared
consistently in all the factor analyses and is adequately reliable
for its length. It has been little discussed because there is little
to say about it. From the item content one might suspect it to be
a measure of aesthetic interest, but otherwise it is unremarkable.
A is remarkable for the homogeneity of its content. All
but four of its items are similar in format, and a factor
corresponding to the scales is apparent in every factor analysis.
The distinction implicit in its content appears to be between seeking
and avoiding situations which necessitate social contact.
The isolation of E in the unrestricted factor analyses was
made difficult by the closeness of the N scales. Once this problem
166
had been eliminated it emerged clearly. The scales are adequately
reliable, although a large number of their items correlate better
with other scales.
F is perhaps closest to traditional ideas of extraversion in
content. In the unrestricted analyses it was quite a small factor,
but this may well be due to the presence of many similar items in
H. At 26-item length its coefficient alpha is 0.81, which is far
from poor.
H must rank as the most remarkable scale of all. Its total
of 26 items reach an alpha of 0.90, which is in excess of many
cognitive tests of similar length, yet within the scale are two
subsets of items which are distinct in their content.
Q1 has only 7 items which clearly correlate best with their
own scale, of which 6 fit the notion of social and political
radicalism. Whilst not one of the most reliable of scales, it
figured consistently in all the analyses.
Qz was most nearly related to A, was only moderately
reliable and only half its items correlated best with their own scale.
Nonetheless again it had a consistent and predictable place in all
the correlational analyses.
167
The title of this thesis refers to re-examination and
attempted replication of the structure of the 16PF. With regard to
replication, there has been partial success, but an examination of
internal scale statistics and of both factor correlation matrices and
the matrix of correlations between the scales after correction for
attenuation suggested that neither Cattell's claims for the internal
structure of his scales nor his account of their interrelationships
were well supported. Since in over thirty years development he
has failed to produce scales which are internally heterogeneous yet
reliable, and since the internal consistency of the scales proved to
be a good measure of their performance throughout, the way now
seems clear for the further development of the scales with the
emphasis on homogeneity of content. At the same time, there is
scope for the incorporation of more recent theoretical notions as
to the sources of differences in observed behaviour across and
within situations without discarding the insights Cattell has provided.
168
APPENDIX
OF
TABLES
169
...... Ln In T- '- NO Tr so r.., so rp - N. uo so CI-. ..-- CO r, ,- r... co co Tr r, bo co o- CN PO PI Cs
Tr rn -- r, so 07 PO CO CO s- M, sO PO rl -- so ro -- DO Cr. OD s- ro co co co Tr co so 'N -.4- r, CO
CD s- M sO Cr s- s- PI s- PO s- s- v0 CP s- sr CP s- CP sO CP s- CO sr s- CA CP s- v0 CP CA Tr r,
. • . • • • • • • • • . . . • . • • • • . • . • • • • . . . • • CA n n CP Cr Cr CA CP CA CP CP CA ISA MCA .0. MCA M M M M n M M M n In M CA M M
1111111 1 1 1 I 1 1 I 1 1 t I I I
CA sr 0- s- CA CO Tr N v3 CA "4" Cs Cs PO CP -C PO M, -0 M7 -0 '- Co 1"... D7 -C! CO MI 4 Tr O. OD
MI CO PO CA PO CO r.., Tr CO r, Tr -4- N. CA CA s- sO PO PO PO CP CP PO CO v0 CA CP sO s- CP CO Tr 'AD
cn -- CO N) .- CO til s- s- s- CP CO PO s- CO CP sr CO CO Tr -- r.4 Tr N CO s- CA N) Tr -- -- vD sr • • •• • . • • • • • . • • • • ••••••••••••••••
CA CP M M M CP CA CP CP CP CP CP CA CP .,- CA CP CP CA CP CA CP CP CO CO CO M CA CO CA M IS, I 1 1 I I 1 I 1 1 1 I I 1 1 1 I 1 I 1 I
CA v0 CO s- sO s- CO s- CO CO Tr CA -0 N 1.10 PO CO CD r, uo NO Cs CA PO MO s- v0 s- N. M .”44. ro
CA s- PO N. 0, CA CO CA MI sr PO C4 M CS! ND C.4 ..- M, sr OD CA -0 ,0 s- sr sO. St n Is.. 4 CO sr PO
CA PO CO CO CP Tr Cr ro -- m CO -- -- CO or CO -- CJ CO CO CO PO CO MO CP CP s- CO CO CP PO CP CO ••••••••..•.• • • • • • • • • • II • •
•• • • • • • •
M CO CO CO CO CA CP CO CO CO CO CO CA MI CP CO CO CA CA CO CO CO CO CO CO CO CO CO CO CO CP CO
1 11111 I 1 I I I I I I 1 I
r, PO N. PO v0 CO Cs s- IN N. CA s- Cs sO Cs M, CO CA co Tr rpPO Tr rP ro -- bo Tr cr. rq bo --
-- -- -- CO PO CA M, bl CA CN 'qt. M r, CO CP CA CA s- MO ,0 s- CA NO -0 PO v0 N. 0, Tr Tr co rp Tr
CA s- CO CO NI CA s- s- s- s- CA •- s- Os Cr s- (54 CP CO CO PO CA (N CP CA s- CA ,- s- Tr rp m CO • • • • . . • • • •• • 4 •• • •• • •••••••••• •••
M M M MI CA CP CA CA Cr CA M4 CO (5 M M CA CA n CP CP CP CP CA CP CO CP CO CA CO CP CO CP
I 1 I I I 1 I I I I I I I I
vD OD PO CA .-- DO N. Os PO Os v7 R. s- CA Cs s- M... PO N. 147 1.11 CO Cs Os PO v7 CO s- M, 4 Cs PO
PO CA CA sr CP PO PO CP CA s- 140 N. CO N Ln ON CO CA sO M, PO s- CO Tr Tr ,- 0, 01, IS! in 1,....
CD CA CP -0 CA CA s- sr s- CA c-.4 CA CA s- s- PO sO s- CO r, .-. -- CO 1,-.1 .._. c.4 .... „..- -O CO in ,r in • • . • • • • . • • • • 4 •• • • • • •••••••••••••
CO C4 n CP CP CP Cr CP CP CA CP s- CP Cr CP CP (54 CP Cr CO CO CP CO CO CO CO n Cr CO CO CO CO :I
1 1 1 1 I 1 1 1 I I 1 1 1 1 1 1 1 1 1 1 I CI ,
Cs PO CA C.1 M, Cs M, 4 CP Tr v7 vD CA Tr Tr rp so rq cr r, rp cr. -- -- rq ..-- r. cr. c, -- c, or. Cr'
r.4 CA CD Tr Tr bo so rP cr. Cr cr. -- CO rp Tr so so m ro co r, PO s- CP s- 00 "CI Cs CP r, Cr co
z m rP -- vo Tr rP PO s- CP CP Cs CA s- s- CP s- CP CP s- s- MO s- ro -- rp ,-- IN ..-- rP ..-- rp CS? u-1 • • • • • •• • • •• • • •• • ••• • • ••••••••••• I..)
M M M M M In en In n M M In M M M CO CO CO CO CO M CO IS, CP CO CO CP CO CO CO Cr CP
I I I I 1 1 I I I I I I I I 1 I Ili;
v0 CO sr vD CO s- -0 0- M1 sO 4 0, M, CP CA sO sO ,-- CO Cs IJO .- M, N. CO CO N. v0 sO CO N) Tr al
rp Ul CA M7 s- sr CO Cs (N Tr rP c..1 4 ro Tr N. -- -Cl co .,-- Tr bo co CO CO s- sr CA Tr -- so Cr- --1
2: CO CO CP CP CA CA CO CO s- .-- CO m .,--- Cyr m --- ..-- I:n Tr IS, -- ro -- rP rP !Ti • . • • • • • • • • • • • •• • •• 0 .1.111 a m • • , • • . • .
(St 16.4 03:1 CP CA n CO CP CO .- CP CA m (54 rp CO CSI Cr CA CO CO Cr CO m M M In CO CO CO IS, M
1 I 1 I I 1 I I 1 I I 1 1 5.. 0
CO CA MI PI r. rr m on uo m Cr NI r, co so IN -- uo ,-- 'IP os CA CP CA s- CP PO Cr sO CP M, sC1 4-
4 co ...- 1.0 PO Tr CO cr bo CA 0- rp 0, 4 r, -- CO CO CO Tr -- 0- uo -- cr. so -- cr CA C.7 PO CO
-.I CO .,- CA CA CA CO CO CO Cs CO CO CA s- CO s- PO M s- PO .- PI CO Cr CP Cl CO Cl ..-- .- CO ...-- ro :...; • • . • • • • • • . • . • • • • •• •• • ••••••••••• 4,4
CO CO CO CO MI CA CP CO n CP CP 15) CP CA CO SA (5) CO In MI CA CP CA CP CP CA m m CS 'Si rp CO r.- 1 I I i I 1 1 1 I 1 i I I I
al Sr sr r. so r, PO PO PO s- Cs Cs Tr os ..... ,--. CA vD 4 r, M, CO s- Cs v0 41 ro uo c, ro rP Tr cr ro r
co co cr. CA s- -0 N) ...- Os Cs CO N) CO CO co co -- so so CA N) sr s- s- '4' Tr CA s- PO Os Cs CA
Z 1-4 CP 0, CS! Is, CA pA ,- ,-- CO (N Tr CO CA CO CA s- '5, .-• CA W
I-. ••••••••••••••• •• • ••••••••••••• u 0 M C4 CP CP CP CP CP CP CP CP CP n n n m m m m M CA CP CP Sr CP CCP CP CA Cr n M M Lt. I 1 I 1 I 1 I I I I I 1 I I 1 I it
•...4
PI CA -.0 CP CO Ul CO PO CO sO N. N. Cs CO sr sr s- CO Cs M, -0 s- N bo Tr ro v0 CA s- N. Ols sr
0, sr PI sO M, Cs "4" NO CO CO sO Ul 121 CA '4' OD PO r, CP Tr cr. CA CP -AD PI PO r, ,-- -.0 M, cl Tr Ir., 2 NI CP • • r bl vD CP s • CP s- rte) 7 s- ro D N C T s v7 CP Cs - CP s- ' C PO A CA > • •• • •• • • •••••••• • • •ft 0
CO CA CO CO CP CA s- CP CO CO CO CO CP CO CP CO CP CO CP CO CA CO Si CO CO CO CO CO rP CO CO rp L.) I 1 I I I I 1 I 1 ! ! ! I
Cs Cs vD CP CA 141 UO ro r. -- cr. uo co -- rp so r, cr. ro r, -- rP -- vD CP CO PO CP M, Tr rp ro el
CN In t10 ..... CO CA Cs -01 4 sr PO CP M, CO CA PO v7 CO CO CO CA r.4 ro Tr -- cr. No -- -- bo co co
CD s- CP CO "- "- Cs CA ACP s- in s- s- CP L7— CO s- CA s- ,- 117 CP C4 CA CA CA s- c...! .,-. Tr -- 0 • . • • . • • • • • • • • •• • ••••••••••••••••
CO CO CO CO CO CO CO CO CO M m n Cr M CO CO CO Cr CO CO CP CA CO CI CO CO CO CO CP CP CO M .0 I 1 I 11 I I 1 1 I I 1 I 1 I I 1 1 I I
(!..15
0, r, ro L")--CS CO bo r. CO r. .-- ,C1 vD CA s- PI CA M, CP Cs PO s- OD Cs CA sr r, Tr CO so --- cl
so co Tr co r\ CO rs. s- PO s- Tr Tr c.4 CA CO 00 .- CA Cs s- -0 .- s- s- sr PO Ul PO CA Cs Tr N) GL. CA CO CA Tr -- -- -0 .- c..! -- Tr IN CA sr CO s- VO CO s- ro CO CA v0 CA CO CO CA CA s- PO Cr CO s-
••••••••••••• ••• •••••••••••••• • ■ el'
M Cl CO CA s- CO CA CP CO CO CO CO CO CO CP CO CO CO CO CO CO CO Cl CO CO CO CO CP CO CO CO CO I I ! I 1 I I 1 I 1 I 1 I 1 1 W
,-1
sr v0 v0 CO PO CP n M, PO v0 ro rp bo -- -- -- so rp ..-- uo CA OD PO PO OD N. CO CP% CA M, DO Tr z
r. rp Tr Tr co ..-- so CA PO PI Tr CA !JO Cs PO sO Cr- PO s- ro co uo Tr -- -- N 1:1°' 0". '40 MI 4'.. N 12
GU CO CO s- CO Tr ,..- uo -- CA -- ro r..1 ro CO -- rp CO m -r- Tr ;Jo --- Tr -- CA CA N) -- -- -- t54 rp I.- • • • • • • •• • . • • • •• • • • • •••••••••• • ••
CP CA CP CO CO CO CO CP CO CO CP CO CO CO CP CO CO CO CO CO CO CO CO CO CO CO CO CP CO CO CO CO I 1 ! I I I I ! I ! I I I I
(N sr Os v0 PI v0 v0 -0 PI Tr c..1 PO r, CO Cs Ul CO Tr co CO Os CA sr PO on 00 Tr r4 N. Tr uo CO
O- ro uo Tr Tr ro ro cr. -- (N co CA co r, CJ r, .,--- Cl- No 'C Cr- (N -.-- so -- -- r...! uo CA CO PO PI
Lj ,- CO CO ,- o...1 CO 4 ,-- N ,- ,- sO CO CO M7 sO ..-- CO -0 4,- s-- CO PI CA 1n CO -- -o CO ro r0 ,CI • • • a e • oe • • • • • a I • • • • I. a le v a • • • • a • ■ • •
CO CO .- CO CO CA CO CO CO CO CO CO CO Cr CP CP CA CO CO Cr CO CO CP CO CO CO CO CO CO M CO M 1 I 1 1 1 I I I I 1 1 I I I
CA s- sr -0 M, Cs CA M, CA CO PI OD PO v0 Tr bo so -- -- -- CA sO CN IN '10 IN rte) Tr Tr bo PO CO
IJO sr N) CO CO PO Tr co co uo CA PO ,-- PO N) s- CO M, CO s- PO CO s- CA -0 s- CA CA MI MI CA PO
04 CO CO Cr CO CP CO CO s- s- s- '5) CO Cl CP CA CO CO PO Cr CO s- CO CO s- CO s- CO CP CO CP CP CP • • •••• •• • • • • •• • •• • • • • •• ••• •• • ••
CO s- CO CO CO CP CO CO CO Cr CO CA CO CP CO CO CO CO CO CO CO CO CO CO CO CO Cr CP CP CP CO M
I 11111 1 I I I I I I
NI 04 CA sr Os ON N) Tr CU -0 Os -0 N CA CA s- CO '54 CO PO r. ro No CO CO CO NI CO s- 1.11 N. rte)
CO 1.11 Cs M, v0 Cs Cs CO sr CO CA s- s- ,-CO PI ON CO In 1..0 1'0 Tr rq co rp VO PO CP Cs v0 v0 CO
SE CO CO s- CO CA s- PO CO CO CA CO CA s- PO s- s- Tr CO r.4 CO c...1 r.4 Tr CO -- (N -- N CO ro (N Cr
. • • . • • • . • . . • • . . • . • . • • • • • . • • • • • • •
s- CA CO CO CA CA CA CO CO CO CO CO CO CO CO M CO CO CO CO CO CO CO CO CO CO MA CA CO CO CO CO I I 1 1 1 1 I I I I 1 1 I 1 I I
SE 04
1: s- CA PI Tr c CA ro sr 5- SE Co GA WU WI. CO AC .J AC At CA Cy cb CA CU G. .cr Co GA UW CO m ...I AM Mt CA CA CI CA CA 0 0 Li Ca. •
170
MO CA CA 4- ,- MO 4- MI sO 0- If) ,- MO CO Os Tr -.0 r, co ro N Tr i co co Ch Ch N. N. TV .r....
Tr co ro ro r, ro co Tr m co co r, TV VO 'CI M C..1 CO il CA NO N N. NO TV TV TV MI r, r, ro
cn n ' so m m -- IN rq PO CA if) CA CA Tr r, is, el n -- ,- ro rl r.., is, CA sO CA +- Tr m • • • • • . . • . • • . . • . • 11 •••••1111110••••
M M M CP CA n CA CA CA CA CA CA CP CA CA CA CA CA CA CA CA CA CA CA CA CA CA CA CA ,-- I I I I I I I I I I I I 1
r, vo in In sO CA OS OS N. CS Ch In ,- OS l'•0 'CI NO M C-.1 1.0 N. Tr N. r, ,- N. Tr 0' Tr
PI sO CA In ,-. Tr OD CA OS 10 M In CA 4 4- M, MO ro to -- ,- 0, No CA In CA M, PI M, r,
a CA CA ro mi m Tr N -- ,- CA m Tr is, n ...4o Tr m n n --..Tr N -- IS, -- c...1 Tr n -- n Tr • . • • . . . . • • • • • • • le • • • • • • • • • • • la a
tS7 *A CA CA CA 'Si'SiCA CA CP CA CA CA CA CA CA 'Si CA CA CA 'Si CA S7 Cl CA CA Co-! ,- is, I 1 1 I 1 1 1 I 1 1 I I I I
in In Tr r, n Tr N. Tr CA -- Tr rq n Tr I17 so -- so 00 •-- cs .,-- rl m m ro m -- Tr r,
CA -.0 PI M ro cr. In r, los to r, n co OD CA Tr Tr m in co so cr. Tr OD -0 Tr Tr cl cq ro r,
CA m) is, mo ..- 10 .,- in is, in .-- mo CA In ,- c...1 Ni m N1 CA Tr m In CA SSA CA CA CA ..-- ..- ,--
. . • • . . . . • . • • . • • •••••••• 11••••••
M M M M is CS! M MI M CS M MI M M is, M MI MI MI M n MI is, MI iSA MI MI IS) MI .h. M MI I I I I I I I I I I I I I I I I
MO Tr CO 4- In Cs LID M, N. CO In s) CO CO Co CA U7 MO CA MO CA VI 10Os M -0 Cs CA M,
MI CA Cs In PO Tr so -- cr. n so co ,-- m CA CO CA CA CO ,- Os CA Tr -- so Tr in
z -- iA.4.- NO CA CA CA CA St CA v... ...- CA MO .,.... is, +- CA C4 4" CA C7 is, ,- PI - ..- CA rl (N
' • . . . • • • • • • • • • • • . . . • • . • • • • • • .
is, 5A is, Is, n Co rsl is, n n is, m is is, in is, Co Co n Co i is, is, is, is, is, -- m is, is, n i 1 1 I I 1 1 1 I I I I I I I I I I
r34 n. CO I''', r..1 in n co -- so -- -- so In CO CA sO Cr Co ro Cr- +- s0 CA Cr, LID MI CA N.
ro -- is, in Cr M Tr so -- co Tr r, m is, Tr so MO NI Cr- 0- ,r co n ,- -.0 is, mo (N so r..1
C CA is, CA CA CA CA Tr ,-- ,... (N .r.- N m CA is, CA CA CA .-- 15) CA CA 0- MCA Tr is . II • • ft• 0 • • • • • 0 ••••••••• •10 •• • •
,-
Co M M M Cal is, CP CA CP Co CA CA CA CA CA CA CA CA CA CA CA Co CP Co CA Co CA Co Co
I I I I I I is,
1 1 I I
CA sO Os CO Cr- toTr M—OOrAMMM In M, N U7 MO In In CO Cr- MI ,--- -40 141 MO NO Co CS CO
'Si NO ,... T... Tr ro Tr 0-. CO -- Tr so so N co so Tr CA U7 Cr- In sO 4- ,- ,--- Cr- LID In CA MO -0
....I ,- isl rel (MI CA n M N CA Co CA ,- CA *- CA in CA MO CA CA CA CA Cl Os CA ,- CA CA CA Co CA
. • . • • . • . • • • • • • • • . . • • . • ••••••••
M M n CA CA CA CA CA CA CA CA 's CA CA CA CA CA CA CA CA is, CA CA CA CA CA CP CP CA CA CA
I I I I I I I I I I I I I I 1 I
P4 'Si M, MO in CO in In CA =- Cr- is, ro r, Tr Tr 0) MO CA Cr. N N CA PO +- MO n ,- N Tr rt _1
co N so -- — so ...- -- CTt m 0) in Tr so co IN cr. N Tr Co OD CA MO Tr is, -- h.) T... CO NO N
V. hi M — CA ,- CA -,- Tr n T..... T. r..1 M M r.... M n r..1 .r. c..! m.,: rl co m; CA CA CA S4 S4 ,- CA 5- • • • • • . • • • • . . . 11••• " ••• si • 11
CI CP CA CA CA CA CSA CA CP CA CA S4 CP in CA in CA is, CA is, is,, CA CA in CA M is, CA in CA 54 CA CA 14. I I I I i I I I I I i I I I I I
-0 cr. Tr ro ,- ,- r, so m r, ,- cr. Tr n co co n so R. ,-. in r, Tr O. OS M CO ,... .1.--.. (N.. 4"
CA ,-- ,- Tr -- ro LT.4 -- ho co ...... -- so -- Co. CO CA CA CA -0 -- sO -0 CA sO CD CO -0 ,-- Tr os r, S Tr Cl If) 4 -O CA Cs C4 CA ,- N) in is ro IN Tr ro m U1 MD -0 CA ,-. N CA CA Co In Co In CA PO • • • • • • • • • • • • • • • • • • . • • • • . • • II • •
M M M M M M is, n CO M M In CA CA M n M S7 CA CA CA CA ,-- S4 In is, S7 n CA Cal M n
! I I I I I ! 1 1 1 ! 1 !
M sO CA CO In CP -- Cs CA -=-- Cr- C.74 MO Cs ,- 00 rq CO In n MO N. M, r.... co PO ro CO CA Cr- In CA
Tr is, rl in -- N rq Tr Cr- in ro ro so -CI ro ro -- so Cr- uo rl Tr -0 co 10 .1- N so Lo ....- ,0
CD CA CA Co ,-. CA in S4 CA CA CA .-- CA CA Co 4- CA .=.- CA Co ,-- CA CA CA CA CA ,- CA is, ,- is 4- =-
•••••••• • . • • . • • • • • . • • • . . . • . • • . •
CA 'Si CP CA CA CP CA n CA M CA CA CA M CA CA CA S4 CA r4 is, -r- m CP CA CA CA CA S4 CA CA In I I I I I I I I I I I I I I I I I I
N CA Cr- N 0, — -0 T... Cr- 1.0 MI 10 is, l'71 NO N. MI Tr Cl C..1 .■-• N') If) 0, In i54 N Cm. int ..-- CA MO
MI MO Cr- CO• sO CA Cr- MO — Tr N N. Cl '0 M IT Ir..1 TV NO MI N N T- rA Cr- Cl- CA CA CA sO ,- 04
1.1... CA ,- .=-- in CO ,- sO ,- ro n PO .-- CA ro N -- N n -- Tr -- N so CA CA CA 4- CA -- Tr -- ,- • . • • . . • . • • • • • 11 • •men• •••••• •••••
M M M M M n CP CA M CA CA M in CA CA SA CP CA CA CP +- CA CA CA CA CA CA CP CP MI 'Si Co I I I I I I I I I ! I I I 1 I
LID ,,- OD U7 is, N. r, co m Cr- r, ro Tr h7I rei T-. T.'. n Ch CO CA n — is, In 4- N) 0- 4 CO Co CO
MI -- sO MI ,- Co Tr rl Tr. -- Co sO ,- CA n CO C"- KA r, N Co LO '0 4- if) Cr- OD Tr uo co N-.1 In
La CA CA ...- Tr ro ,-- Tr ro Co -- M Cg, Co CA M, Tr — ro -- CA 04 CA ,-- CA Cc:! CA CA • • ••••••• • 1.• •all• ••••••••••••• I. • •
CA CA CA 15) CA CA CA CA CA M CA is, CP CA CA CA CA is, Co Co CA 15 CA CA CA Co is. Cn CA is, is, CV
I I I i I 1 I I I I I I !
M ...- CO ..- N VI Cr- M, .-- CO OS N 1.0 R. -0 r, r, ro n Cr- n in r..,.. ro ro so uo r, CO so Cr. r,
uo co so -- cr. Co Co -0 Co CO MI Cl sO CO In Cl- 4 CA CA N '0 0,.. CA CA CA N) C.4 Ig
If) 7.1E
(-1 CA CA sO ,- ,- CA 4 N ro n -- r, is, m Tr so -- Co c..1 Co .r- is, in N ro m C•.1 Tt
CO sO CA sO MO r, Tr so so r, CO ro ro m r, n Tr co co co uo so No Tr Cr- Co Cr- so Co +- sO Cs M, MO Tr OS is, CA NO NO ro Tr ro •• q" Co PI Co M CA CP +- Co 04 CA +- CO Co Co +- ro is, Co n m is, ro 'Si n •••••• •• • • • 11•• • • • • . • • • •
/Si MI 131 /S1 tSi CSl CA 'SI VA CSI ISI CS! CAC in ISt SI In f In in in in Cl in CA MI
I I 1 1 1 1 1
1.14
RA MO 4 C CA MI Tr 04 GJ CO MC hi -J MIC Mt CD CZ CM CM CM <r 04 WU CD MC I-4 -J MC NC CD CN CM CM CM
0 0 U- U-
6.0
48 -0
.249
• . • • ••• •• •••••• • • •••••••• •ft • ••
Co Co CP CA CA Co Co CA CA Co CA CA CA CA CA Co CA Co +- Co 'Si Co CP CA CA CA CA CA Co CA I I I I I I III I I
eA ,-. Tr Co N Cr- CA N LID ,- CA In N CO M, CA Co CA MI CA Tr CO -0 CO In CA CA ,-- ,0 ,ct
Co M, Cr- MO CA CA r, sO M sO CA *A M 4 In In +- .1- CA Co Tr -Cu M Cs 4 M) CA CA MO CO
04 CA MO Co Co Co ,- CG is. — m Co Co CA m is, it m CO m Co is. m m Co n el ,- n m n • • • • • • ■ . . . . . • • • • . • • ••• • • • • ••
M Co M M. M M M MI CA CP CA CA CA CP Co CA CD CA CA CP CA CA Co CA CA CP CA I 1 1 ! ! I 1 I I I I I I I I
- 4 M, CA 4- M, +- CA sO Os N. Cr- M, 4- Cs In Cs 0) Tr is, N ..-- -- so ro Cr. M uo m r, r,
-- Cr. ro cl sO CA +- s0 MI CA Tt S4 Cs Tr Tr m N ro N Tr In CA hD - .,-- In CA sO CA rl r, Co
CN CA CA CA ,- +- CA CA w- .- M CA MI 4 CP •- CA M Co CA CA — +- is, CP 'Si Tr CA Cr- CA M4 is, • . . . . • • . . • • • . . • . • • • • • • • . • . • • • •
CA CA CA CA CA CP CA CP CP Co CP CA M CP CP CA CA Co MI CA CA CA CP CA CA CA CA CA CA CA I I 1 I I I I I I I I I I
OD Tr 'N Cr- r, Co C4 MO CA -40 Cr- +- Tr -- vo cr. oo Tr r, cr. cr. co co m ro In sO MO MO =-- 0-.
CA CA U7 CS r.0 .,.. T... T-.. ID- CA Os Cs Tr r, -- so -- cr. Tr Tr n so so uo in ro -- Tr Tr r.., Tr
C) el CA -0 ..-- CA - Tr CA ,... ,- -- CA Tr so -- is r, -- r..1 n In N r...1 n -- Cn is, N Tr so • • • . . . • • • • • . • • . . . . • . • • . • • . • • • • • a
CA CA CA S4 CA CA CA CA CP CA CA CP CA CA CA CP CA 'S, CA CA Co Co CA CP CA M CA CA is, CA CA _, 1 I I I I I I I I I I I I I I I I I I 0
Cr,
171
CA Lil -0 In 0- 1.11 SO 111 CA a- No ro No cs os n cl z In ..40 ,- ,- CN 04 ,r ,_ ,r ,_ -0 ,_ (N ..40
a co -- NO Cs ,- sO 4.- CS SO CI sO PI IN r, L....14 Do CO CA sO r, ,- CA CO M PO r4 ,- ,- ,- IN a a
CN M M 10 M 'Si +- NO 'Si CI .-- n In M 'Si M CO 'Si cc, "zr r- ,- ,-- SI M (-1 ,-. ,- Do n n P0 In
•••••••••• ■ •••.• • • • • . • • • . • . . • • . •
M 'Si M M 'Si M Si M M M CD CD CD CD CD CD CD cc. CD CD CD CD CD CD cc, CD 'Si CD CD CD CD CD
I I I I I I I I I I I 1IIIIIII I 1 I I
CO M ON ON PO ,- n CO r, a- -- so ,- co CN ON M *qt. ,....': CO 00 M N. In a r, PO M 0-, Cr.. ,-. ro
PI -0 NO CA CO3 Cs CO sO *- a- co ro m cs ,-- 'Si n Do r, 'Si CA N 'Si co so cs m ,- CA -0 In IN CA
cn n m r..1 m m a- .- ...- ,- 'Si m ro m m cis ro 'Si m CA cc. ,- 4 *- 'Si 'Si ,-- CA r.".1 n M *4 NI
1111•••••••••••••• •••• • • ••• • ••• • ft 1111
CA 'Si 'Si 'Si CD CD CD CD CD 'Si r CD 'Si CD CD CD CD CA CA CA M CA M CA MI CA M 'Si fA n 'Si 'Si I 1 1 i 1 I 1 I 1 1 i 1 i 1 i i I
-gr r, 0- a- c...1 a- co ro so m co co uo r, co cs a co r, 0- -0 ro a m NI a Cs a- In '- N) -0
CA CO 4 'Si CA C4 -0 M ,-. sO NI a- Si CA NI ..-- N. r, a a CA In 'i" 'Si r, 'Si 'Si a n CA CO CA ts0 Cr, ro m m ,- PI ,- PO M MI M ,- M M M M 'Si NI M cc, n CA ,- CA M M 'Si *- M M NO M n • • •• • • • • • ••• 11••• ••• • ••• ••••• ••••
M MI fA CA M 'Si M M CA M CA M M +- CA M M CA M M 'S M SI M M M CA M M M M M
I 1 I 1 I 1 1 1 I 1 1 1 i i 1 1 1
a- CA f', LO LO -0 -0 N +- +- a- n r, uo ,- -0 cs '0 so -0 NI 0- ro PO r4 ,-. N. n -0 cs. co ro
.-- m vo so so a In CA CO ON M r, -0 a- CJ cs r, 10 CA NO In OS CD 'Si -0 Cul In NO NO ,-- T- CA D1
CD CA M CA PO rl CA CA M ,- *- *- *- M M SI M M 'Si M NI ,- *- CA ,- c...1 ,-. --- ,- PO CA .- M
. • • . • • • • • • • . • • • • . • • • • . • • • • • . . • • . M M M M M M M M M M CA M +- CA M M M M M SI M SI M CA CD CD CZ) 'Si CA M la M
1 I I I I i I i i I I i i I I
N) — LO 0- M T... CO In In a- uo r, m CO -0 NI CO r, 43 (N tom) a- -- so CO In -a Do 'Si cs. Do O- ✓- a- cs. ro CA IN CA P.1 '4- CO In In siD 'Si SI NI *- Ni IN CO OD 'AD sO M 'Si NI a- -0 a- PO 1-0 In
CO +- M a CA +- +- •q- +- +- +- +-- ON ..-. M in In T... M a- CA ,- M 4 *- CA ,- M In M ,- ND a ca • • . • . . • • • • • . • • • • • • . • . ••••••••• :1
'Si CD CD CD CD CD CA CA M MI M M M M MI M CA MI M M SI CA fA M CD M M M M *4 fA ST CI
1 11111 1 I 1 I 1! I I 1 1 I I I 1 r_ T,
-0 ..01, -40 0, NI -40 +- -40 ro a Cr- Do -a CD ,- sO ..-- NO Cr- M CO ' CA sO NI In 'Si NI CD CD In CN
-010 +- Ps PO .:4- 0, +- NO bl CO 10 IN -q- MI ,0 CO CO M 0- CO ‹r 0 0 M 1'0 r-, NO 0, 0, -I- sl al
Z Si CD +- CA l'n MI CA M +- +- IN +- +- +- MT MI +- M +- +- CA cc, CA CA Sc, .-- +- +- M Si fA M LI • • • • ••• • • ••••• •• •••• ••••• •• • • • • • r - n IS MI M 151 'Si M M 'Si CA *a la 'Si 'SI 'Si 'Si 'Si 151 M M MI M CA CA CA fA CA CA SSA SI 'S M 21
I I I 1 I I I I 1 I I I 1 1 I I •H Li
a- a- a as N. CA .- 4* ND N. a- a- ,-. n a- a- os IN so co ro cs (N rs CO ,- NO sO PO M CO sO In
rs r, m CA -.ID CA 10 '4- CO -.0 In CO la ro CO CA N, CA CA PO PO In CA N., s3 -0 In 0:1 1-0) -40 1.0 CO
1: CD CD .- =- CD ..-- ,- +- CD 13'.. ,-- ....- ..-- ra MI +- CA CA +- ,- CA M ,- 'Si ,- t*0 M n CA CA MT S4 al • • . • • • • • • . • • . . . • • • . . • • • • • . . • . . . • .-1 CA m n m m n m n 'Si n m n n n m m m n n 'Si 'Si 'Si m n m n n m 151 'Si 15) M M
I I I I I I I I I I I E
OD n 0, CS V' a- a c...1 ro N) ro Do ,-. so r, CA ,- IN sD M PI CI n CA ,- s0 .cr a, 4 -0 0, CO f••••
CO CA +- ro so os a Do Do co ro a- es so .nr so ro r, r, c...1 -- CA n M In n sO CA OD n 0- In 0
--I 'Si +- CA In N M +- M Cl- Si .4-- +- +- CA +- CA +- SA +- CA CA +.- CA +- ro cc, rq 151 ,- m m cl 4- • • • . . . • • . • . • . • • • • • e • • • • • e • • • • • a a
M CA CA CA M 'Si ra M 'Si CA n M cc, CA M M M M 'Si CA MI 'Si M 'Si M M M MI CA M MI M I I I i 1 I i i I i 1 1 I ! 1 •'-i
!..- <Y: MI -40 P.0 Os IN IN +- +- CA a- so NI CA V') 00 in SO OD MA NO N. .4D V.5.1 Ll PO -4.- CA CD sO Os IN CD 4-,
CA N -0 sO sO CA M 0- In a -- NO CO ..- ,-. 0-. NO N ,- In N) a- 0- 151 -0 CO if) a- PI In M -.0 M ✓ 1.4 M T.. T.. 'Si M T- T- NO MI T- M T- M M T.. M M M T- M M M MI 'q- r- r4 'Si c..4 n cc -- ,.... z ,.--. • . • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • O CA M M CA CA CA CA CA M MI MI 'Si MT M SI M 'Si 'Si 'Si n 'S. CA n MI CA 'Si MI CA M 151 M M a
1.1. I I 1 I 1 I I 1 1 1 1 I 1 1 1 I i u r-
a- co In ND IN N. CA ,- 4 ,- ,- CO --0 OD M sO M
a- a- SO Mi a cs cs m a- 10 0- CA CA *A sO +- T, Pa r411 r K r3 ri.';Pi q! ';,7 ;17. N. '43 8:", gr'. E-,!. •ri
=C In ...... ro 10 In M ON ..... T'. T- Cl a rq ro ,- ro ro -- CA a If) 'Si r-, ,- 'Si •-• cl ,r ,- N) ,- ,.... r- • • • — • • •••• •II• II • ••• •••• • ••••• • a 177.
M M 'Si M M 'Si M 'Si n M MA 'Si MI M MI SI 'Si M n M 'Si M n n 'Si M MA 151 M MI 'Si Sc, > I 1 1 I I 1 1 I 1 1 I I 0
ILI
CO +- -0 -0 O. 0, IN rs a N -0 -- so a -- 10 In -- a a- CA CO Si ...- NI ..-- If) rl sO ,- -0 sO
SO M 0, NO C.4 '4"' Os CA Cs. c...1 a- 1--, Do -0 co -o -- ro a- r, ,- Do N m r, co Do a- ro cs co ro • .
CO +- CD 'Si -.-- ,- 0, M +- 'Si +- M +- Cq +- '71- +- M M +- M CA in 'Si 'Si M ...... *- ,- ,- SI a CA al
. . • • • • • • • . • . . . • . • • • • • • •• • •• 11• • • • r-
M M :Si M lala 'Si 'Si 'Si M MI MA 'Si 'Si 'Si CA 'Si la M 'SI m m m m m M 151 ,SA 'Si 'Si M.! m F1 1 1 1 1 1 1 1 1 1 I I I I 1 i I i i I
CO 'Si ON T... CN ON N. N. a- N r.1 Mi r., N 1•10 CS SI r..., Ct.. ....., 0... 10.1 'Si NO r40 in ,r 00 0, Is, In ,:-.4 f-
0-. so ro a- ro N '4' -0 sO NO MCA a- (-1 cs ,- 0- ro 'Si co No r.., CA Si CA -0 Cs U0 CA In ,- 0- N
I.L. +- =- +- *4- 0- +- In n CA MA V, ....- CA PO M M CA *- M CA -0 ,- 4 CA ,- M ,-. ,- ,- ro ,- 'Si a • . • • • • • • • • • • • 4 • • • • • • • • • • • • • • • • • •
'S. 'Si M 'Si M 'Si CA CA CA 'Si M M M M M 'Si M 'Si M M 'Si 'Si M 'Si M M 'Si SI la la 5:, MI CA I 1 I 1 i i i i I I I 1 1 1 ! <1....
cs rl a- n ,- so ro os Cr- cs cs os Do a- OT Do In 10 IN *- CO NI In NI N 'T T.... rl .q" ..1' Cr. so al
r, ro c-1 co a- ro n so ro CA IN NI sO CA ON CN In M LO M In MA +- a- co r/ -- n so m n ro r-I
1.1.1 CD ...- ,- co a- ,- Do Salts *- CA CA PO ,- n MI N,- M 10 4* CA a- ,- .-- (-1 ro c-1 rl -- Si n 2D
. • . • . • • • • • • • • • . • . • • • • • . . . . • • • . • • ,-
'Si CD CD 'Si CD 'Si 'Si 'Si CD 'Si CA 'Si 'Si 'Si CA M 'Si 454 SSA 'Si M CA CA MI M M 'Si 'Si M SA SI SSA I-- I I I I I I I I I I I I I I 1
CA fA NO .V. Os NO In In cs a- so Do N 0.,.. 0" ..C., N. MI 454 PO ,... a- ro Do os so r, r, -- ,- so a
ro a -CI CA NI 0- -0 sO ,- 'Si *- Cs sO M c...1 ro a a r, cs ,-. co N) rl n '5, N) a ,- In 'Si +-
C.) +- CD 0.- +- +- cD In +- eq ,- .- a- m m CA In ,- m ,1- ,- cl n ,r ,- cA m ,- 0 'Si ,- rn ,r • . • • • • . . • • • • • • • • ••••••••••••• • ••
M 'Si M 'Si M 'Si 'Si 'Si CD CD 'Si CD 'Si 151 CD CD CD '51 'Si CD 'Si 'Si 'Si 'Si CD CD 'Si 'Si CD 'Si 'Si 151 I I I I I I I I I I I I I I
SSA a- 'Si rq m ,-. co so n a ,0 ,- CA N. MI In in CO rl CO ....... sO n MI CA a co -0 so 0- sO 'Si
r, ,- a- PO -40 ro .gr CA CA IN in a- ro -a M ....- Cs In .zr CO a- CA M M sO M CA M a- Do a- c-.4
0.4 CD CD CD ..-- +- CD =- ,-- ,..- M M M M M M M n CA M M ,- MI *- ,- M *- M ,-. M ..-- M M
11•••••••••••11•1111 Il• ••• •• •• •• It ••••
M T.... M M Si CA M CA CA M MI CA M M M M M M M CD CD CD CD CD CD CD CD 'Si CD CD CD CD
I I I I I I I I i i I 1 i I I I
— m CI ON 03 co a N) CO a- so ro a- a- co c..1 ro cs r, co ON ON 1.0 T.. CA -0 CA M CA sO -.0 MI
CO N hl IN 0- 43 a- rq OD rs -0 ,- m co 43 co a- m Si In ,ID r.,, Do ,-. -0 co tO 0 'Si ,C1 -.0 CO ¢ op 'Si -- m .— ,— ro m m 'Si 'Si -- m rl m m a- ,-. .- m c...1 ,- cq 'Si M 'Si '51 +- 'Si Cl 'Si 'Si
. • . • • • • •••••••q• 11•••••••••••••••
CR CD CD CD MD CD CD CN CA CD CD CD CD CD CD CD CD CD Si CD CD CD CD CD C4 CD CD CD CD CD Si CA I I I I I I I 1 1 1 1 1 1 i I
04
C4 NO .gt C CA PO r, 4C 04 C-1 W La- CO CC '-4 J AC 2C CD CD CD C11 CD C. .cr 04 Li W 1.11. CD CC 0.-4 —I AC 2C CD Cb CD CD CZ
W u.
.A. PQ .,.... (Si if) CO -O ,- (Si LO N N. NO 'C' ro m ro c...1 Cr- (Si N -- CA -O .,--. CO CA cr un co cr N. u.-.1 -t-,
,-- m CA cr KA m cr (Si (Si r, un (Si No r. No (Si N 1 ,- (Si <V .,..-. ,- CA CO (A to CA NO CA NO 117. E I-I M 4.- .1.... ,.... N 'Si .1.- cr ,-. (Si 'Si ..... ,... M 'Si (Si M (Si M .,.... r.- 4.... Lei CO CA Cl (Si rl 'Si (Si (Si 'Si E S- • • • • • • • • • • • • • • • • • • • • • • • • • • •
O (Si M (Si (Si M (Si M M 'Si (Si (Si (Si (Si (Si M (A (Si (Si (Si (Si (Si (A to to rA (51 (Si (Si iii (Si 'Si (A W 1.4. I I I I I I I I I I I I I I I I 1J
C:
V) M ro In (Si (Si (Si (A (Si CA N CO cr N Cr- -0 .0 0.. TT .40 CA -0 Si( ,- In (A CA CO CA CA 65
10 'Si ro -- CA r. un Cr- m N un n m co co or co un (A cr cr CSI cr. -7 r, ,-- cr u) if) rA ;JO •,-1
2 CA ..-.. cr cr cr m r, KA m .--- IN 4 rl CA ,-- N) Ci 'Si ro cr if) m t5i m KA -- ,-- un .--- N) (N CA r- • • • • • . . • • • • • . • II • • • • • • • • • • • • • • fi7I
M 'Si (Si (A 'Si M 'Si (A Si 'Si (A to (A (Si (A to (Si 'Si (Si (Si (A (Si 'Si (Si (Si (Si (Si (Si (Si (A to > I I I I I I I I I I I I I I Cl
13
ON -O cr ro Cr- Co NO -0 PO ON <I- ON CO (A ,- <V ON ,-'T cr co CA -0 N) 1r (Si In (A CA ON CA
r. c...1 co m CA 10 NO cr rl un cr CO cr m CA (A CO Vs. NO NO (A *It ...- MI ..---. sO LI? N. ,- N.) r, • .
CD *- CA M rl ,-- un KA KA ,-- n ItT4 *- 1- cr .-- *- m M ,- CA M (A •-• ..-- MA v- M (A ,-- NO ,.. CI
. . • • . • • • • • . • • • • • V • • • 1 • • • • • • • • • • M VA IA M M M In M M M fA M M M M M MA M (A (A .- M M m m m M (A M (A (A ..I..'
I i I I I i I I I I I 11 I I
ON •-• CO Os ON CO CO co ro ro No co ,-- r, (Si m r..1 KA cr ND CA CO ...- ,- ....- In Os (A ,- 5-
sO ..-- LO -0 r, ,- NO CO Co Cr- Li) Ci ..- (N N. co cl ma -0 it' cr (-1 Cr- Cr- r, 0.„ (Si Mi -0 113
LI- N rq cr NO N If) N (A N ..--- .,..- CA ..- ,- P1 'Si (Si 1", -,..- CA LO .,---. ..--- (Si CA CA (A Ni -- 'Si Ci • . • . • • 1 • • 4 1 • • • • V • • II • • • • • • • • • •
(Si L5q (Si 'Si (Si 6i M M (Si (Si M (Si (Si (Si (Si (Si M M (Si ,- (Si 'Si M (Si (Si Si M M M n (Si (N 1 1 i 1 t I I I I I I I ,x
co C0 ro '- -- bl M UZi CO SI CA ND Cl-. CO -0 rA In ND CO CA Cr- ,-- ,- ..-- Cl-. CO CA CO In In N. al
LC) CP (Si CO N. CA Ki 0, CO ID rl CA r, -o If) os m N No (Si si (Si Mi Mi ND LO cr co c, 8-1
LU (Si to ....- In CA rya <V to CA .,-- -- CA PO (Si (Si ,- (A (A Cr- CO ...- cr -- CA CA CA CA CA (A 'Si (A .i0 • • • . • • • . • 4 • • • • 4..484 ... ••••ft• • •
it
M CA M M M to M tSi M M M M M M M M M M M (A to CA (A M M (A to (A (A M to to II- I I I I I I I I I I I
N. (A N. Cr- P1 -0 -0 Cr- NO sO N (Si In TT If) •-• ND (Si -- ON CA CO CA (Si .7 ,- ,- 4-4- m1
(Si N V71 (Si 44- r, CA (A N. r'0 cr KA No M4 (Si ,- Cr- CO r, Lr.! ...- ,.... 1.0 MI Cr- A0 CA ,- 10
C.) ..-- *It CA M N ...... ...- *- mei' m (Si N cr -- KA Cr- 'Si (Si m PO 'Si CA (1 ..-- '1' 'Si ••-- Mi <1-
172
(A -0 CA NO 00 OD NO ON ON PO PO -0 CI C.4 N. ,- CA CA LO 4,-- cr. cr O. cr (Si Cl-. (Si
cr Co CA PO ON PO ND In 00 'Si ro un M r4 cr to CA ON NO P, 10 ND .- ,- (A ND Os If) ON Ill
CN CA (A (A (A CA ,- CA 'Si 'Si <' 'Si a NI If) Si to to (A ..-- CA (A CA (Si CA </- (Si (Si PO CO
. . • . . • . . • • • II • • 0 • • II a • • • II • • V •
M to M M (Si a Si (Si M (Si (A 'Si to (A (A (Si (Si (Si (Si Si (Si (A 'Si 'Si (A (A (A to (A (A
I I I 1 I I I 1 I I I I I
sO 0. 0 No r, or co un --. IN No I", In (A ON CA r, OD h, NO CA NI cr o-. c, PO </- to Si ,.-- CO (A Cl-. If) cr N. <1' CA ,-- 00 CI NI to CA NO (A ND i- un rn (Si O. CN CAM (A ...- cr .--. KA (A (A cr ro m M (A .,-- NI CA (A (A (A ,- CO (A (A Os ro
. • • • II • • • • • • II • • • • • • • • • II • • II
M M (Si (A to Si( to (A 'Si to to CA (A (A (A 'Si Cii (Si (Si (Si (Si 'Si 'Si to (Si (A (Si (Si 'Si (Si
I I I I I I I I I I I I I I I
..0 ON .-- .cr M ,- OD Cr- `40 CO .1-■ 0' ''- CO O 6'7 ,.... 1.17 Os CA CI cr CO ro r, Cr- cr oN cr m..! N -O 10 If) Si( 10 ON NI In to NO Cr- OD If) N NO CA <1- (Si ..-- 10 NO CO CI cr 17.4 N. -- ro un
CN CA -..- ,-- ..-- PO (A ro m m m to PO (A (A to -,-- (A NI -- NO to (A ...-- to ..-- (A OD rA M • • • . • • • . • • . • • • . . . • • • • • . . . . • • .
(A M to (A to 'Si (A (Si (A 'Si (Si to to 'Si 'Si to Crai (A M (A to 'Si (A Si 'Si to 'Si (Si M 'Si (A to
I I I I I I I I I IIIII I
CA -O cr Cr- -Ci No No cr ro co un Cr- so rs7 'Si co un (Si IN co r, No (Si N r, cr ro cr
to cr VO ON I'D co ro IN Nc CSi No to Cr' Ps. LO N CO (A PO r, R.. N. 117 Cr,
C5 to (Si (A CA •-• (Si -- CA (Si (Si (A (Si to CA (Si (Si -- 'Si CO to CO (A (A 'Si
. • • • • • • . . • • • •
•
• • • • • • • • • • • • •
M (Si to to MfAMMfAMM fAMMfAMMfAnfAMM (Si
I I I I I I I I
to -O h, N CO CA NO CO ON NO N) (A <r Si .,...- ON ..414 1 Cl T... 10 to In ro un c, un CA ON CA ON
If)Col4 (Si LO cr -- cr rq co PO ro M CA .•-• NO CA NO r, (A <V (Si 'Si PO NO ,-- h, CA <V -0
CD -- ...- L17 r..1 -- .- cr CA (A ( -- ,- (Si CO LO ...- 5i CA CA (Si LO CA ..-- (Si 'Si ON (Si ..-- N) ,r G. • • . • . • • • . • • • . . • • • • . . • • . • . • . • ..-5
(5, m KA n m m m m m m m (Si m m SiK m KA m m KA KA KA m M (A (A (A 'Si (A 0
I 1 I I I I I I I I I I I I I I I I I 5- Cr.
ri 00 .r.-. cr In r. CA cr ro KA cr r. cr. ro cr -0,- tS co -- KA un cr CA NO (A ON (A h, NO <V
CI CA ,- Cr- In -,.- if) -0 10 N cr vo cr cr KA M un Cr- -0 -- (51 cr r, -- -0 N) cr so n al
z m m ro -- ,- rq m CA M -,- to ,- ...- c..1 ,- m m rl rq -- -- to -- -- co (A ,-- M ...- CA U
. . . • • • • . • • . • • • II • • • • • • II • • • • • • • • r-
M Si to M n M (Si M (Si M M M MI to M (Si 'Si 'Si to M M (Si 'Si (Si (5 MI (Si (Si (Si (Si M 'Si al
! 1 1 1 1 I I t i I 1 1 1 1 1 I 1 1 1 I } •r--1
cr No cr ro -- -- cr -0 -- un 1.1 -- mot' N ,- N. Cr- CA ON -- cr -- N -- co No 10 CO N. 0'
(A to CA NO OD <V CO (A sO ro ro un m m IN •-- z
ter' MI'Si (A (A (A CA (A is -- CA ,-- ON -- (A PO ..-- uA (A W
• • • • •••••••••• • • • • • • • • • • • • • • II • r-1
M M (Si 'Si (Si (Si M (Si n (Si (Si (Si M n (Si M M (Si ':Si M (Si M (Si to 'Si 'Si to (A (Si M M 177,
i i I I I I I I I E
CA ON ro No PO CO PO CO CA PO <V <V 'Si 1- CO+-MME CO co r, CA ro r, co co
-0 tSI co CA N. P, NO LO '0 (S 'Si CA m Mt
Ao (Si s.4 (5, (-1 ,r ,r(Si CO o, s,
J Cal m 04 ,- (A (Si CO -.- to N CA 'Si 'Si CA 'Si to N CA CA ON •-• •-- (A (A IN 4- • • • • • • • • • • • • • • • • • • • • • • • • • • • • •
(SI 'Si (Si (A to (A (A (Si (Si (A 'Si (Si to (Si (A (Si (A to (Si to (A eSi (A (A (Si (A (Si (Si (Si (Si Si
I I I I I I I I I I I I I
0 -0
. • • • • .. • • . • . • . • .
(Si to 'Si 'Si m CA (Si t5i (Si (Si (Si (Si (Si (Si (Si to (Si (A 'Si 'Si (Si 'SI M m 'Si 'Si (Si m m KA (Si KA I I I I I I I I I I I I
ON Co ro PO P,.. In CO P, CA CI r, 00 cr m r, ro un un m cN No m cr os -- ND M Lil IA, CA
to If) "I" (A CI M CA CA CO In .1- N CA CO In 'Si in r, ro CO -r N) ro 'Si CA -0 ro -- r..!
04 ..-- N (Si ....-- ,-..- en .-. 'Si M 'Si (Si 'Si M M) (Si OD (Si (Si M (Si M Si 'Si 'Si IS) Si 151 (Si M (A • • • • • • . • . • • • . . • . • • • ••••• • • ••••
(Si M M 'Si M (Si (Si M M M M 'Si (Si to (Si (Si (Si (A (A (Si M 'Si (Si M (Si (A (Si (Si (Si 'Si
I I I I 1 I I I I I I I I I I I I
NI If) N. In to M ,- ND ,-- Os •-• CO ',1" (Si r..! OD N. -,- (Si N. cr No 0, Si r, so Os (Si CO NO N)
.1- Cr- cr ro CN r, ro ro r, co -- r. In CO CA CO 'Si ND CA (Si Cr- CA .- CA cr so m No CA PO
<C•1-151 *--NN C5i POM ,-- CA ,- ..- Cii VI (Si (Si CO (Si ,- ,- PO .- CA (Si to 'Si M ...- ,- CA 151 (A • • • • • • . . • . • • • • • . • . • • • . • • • • .111.11
MI (Si M Mi MI M Si 'Si M 'Si M 6a 'Si 'Si In to 'Si In In (Si (Si (Si In 'Si fil M M (Si M (Si l' I I I I I I I I I I
• • • • • • • • • • •
po <r S-
•
<r 04 (-3 WU WI- CD MC 1-4 J AC Z CD Ci Cb Ci CD C- <E OA GJ WW LL CD 2 O 0
-J AC NC CD CN CZ CN CD
173
un c-.4 N O. O. In M CO NI r, co -01 .,--. ro -- uo ,..-.. ro N) r, c...1 CO -- ON ON Cl. C.4 C,, r, rq -4- *-
*I" CA ON r4 CO T.- .4- -- c...1 ON N0 M M T- M CCI 1.17 147 PO V, cl ro b7 r, to 41 CA CO N PO -4- ,0 ....-
a ..-- m No m m m c...1 m PI ,- M b7 M .... VI 00 .4-- MA 1- tn M4 M CA MI rl M .4.- LO CD =,- ro 40
•••••••••••••••• 18••• • ••••••••••11 tn in M M M M n en IS4 In Ma M tn IS M tn M1 tn In In MI M In MI tS4 MI to IS4 tn MI, M tS4
I 1 1 1 1 1 I 1 1 1 1 1 1 1111 1 1
os r, r, n sO 41 VI 1.11 174 M CO PO •.4-- .q. c...1 ,-.. Cl' r, ,c1 IN ,o -- ON N rl -o .v.- C4 Cr. .t -1- N)
ro N. c1.-. r. so Lin so ,- t (Si .44.- sO ...CI CO b7 bl 00 ...4. N. in CT. PO .nr ro N) PI PI ro N. -0 to -1- to
C t! m N V4 rl .zt- m M4 V7.-- C4 CA Cl tn 0, M MA M ro -- cl PO M SI .4-- r4 V7 CA IN tn uo 1- • . . . • • • • . • • •1111•• •••• ■ •• •••• • ••••
M M M M M M M M M MI M M M M MI M MI M M M M MI tn M M M MI M Mi MA n n t 1 1 1 1 1 1 1 1 1 1 I 1 i 1 i
m ro .4- b7 M .cr N a- r, uo co -- c..1 Cr. ..4- ro on to 117 W1 Cr- '''' (171 bO b7 tn b7 .4-- VI -0 s0 -,4.-
N M ,..- .4.- CO N. VI sO PO V, MA N. n -4- ro uo n N. ro --- .a.. ro ,- Cr- so PO .71' M rq ro -o -- ro
a ro m ...- n C4 Mi r4 M Si M M .4.- tn Cr- MA .4.- r4 M .4...... M CA M .-- M ,- tn n -- tn -tt m n • • . • • • . • • • • • • . . . • . • • • • • • • . . • • . • •
m m m m MI tn M M M M M M M M M M M M M M MA MI M In tn MI n MA MI M tn M
11111 1 I 1 I 1 1 1 1 1 1
PO VI N .4.- .4- co -4.. n -- 1.41 Cr, 110 'T CJ .,-- ,- .4- m -- to si m No CO ON ON T- .q- r, .q- uo m
.q. VO VD 1.17 PO r, m CO uo -- uo .q. ..-- -4. o0 -- r, PI VI .4-- CA r, ro m PD tn b7 sO .q. c...1 cr CA
CI M M CD PO r..1 r4 r4 tn CA N 104 tn M MI Cl M M MA M ..t r4 N 44.- CA -4.-- CA .4-- MI In v- • • • • • • • • • • • • • • • • • • • • • • • • • • • • • •••
M M M M M M M M M M M M .4.- MI M M M MA MI M M tn MA Si MI t51 MI M M M4 M M
1 1 1 1 I 1 1 1 1 1 1 1 1 1
OD tn MI sO 0- MA ("4 VI CO 0, 1.47 b7 Ll ,- VI sO r4 sO 04 b7 .4.-. r, cl .40 CO V, sO -0 IN 0, b7 CA
mm 1' 141 CO -.4. cs tn uo sO . t. rl -4- m -.4o n -40 (\I m r, OD .4-- PO M4 VO 1- sO M n VO rl VI
CO ....- .,-- lil SiC CD CD Cl M CA .4-- M CO M -- CJ 1.37 .4-. M .q- Si n n ro n c., m tn uo n -- vo .tr a • • • • . • . • • • • . • . • • • • • • • . • • ••• ••• •• 7.; M M n M n tn n M M MI M M M M MI MA M M MA MI MI MA tn In Mt M M M MMA n MI 0
11111111 I I 1 11111 1 I 1 1 I r- IT,
m m C, r4 sO -.0 -0 .4.- uo N. m vo Cr- on co CO CA ..t 1.0 1- r..1 ON ,0 tn PO .4- -s ...o ro co c...! uo
a- .4- uo .nr ro 127 -- N ifl VD -0 1" b7 V, sO M r4 r4 MI CO r4 M V, Cr- sO C4 CO 0, 1.1.4 -4--. ,0 CCI VI
7st ..-.. CD CD VI CA r4 04 ...- ,-- .4-- OD MA C4 M Cl M M M MA CA r4 r4 .4-- MI In PO C4 M rl M -- tn .A.=
.••••••••••••••. .••••••••••••••• s-
Si Si n tn In tn Si tn Sit M Mi SiI MI M SD M CD CD CD M CD CD CD CD CD CD CD In In M M M 113
1111 1 1 1 1 1 1 1 1 1 1 1 1 1 a,
0, b7 M V, VI Ul sO r..1 .- r..1 V, 10, 127 bl M N. CO r4 V, C4 V7 tn Cr- C..1 -- 0, r..I ON -st" In ....- NO ..---1
PO rA 0, 0-- '4- r, 0, CO r, uo ro ..0 -t- MA .4.- PI .Kr. -- r, -4- uo -- r, co r, ,0 ,0 Cr- m m m ,... !1:1
17. M M ..- M M .4.- MA M M C, .4-- .4-- Cl tn ..- ..- M MA GI .4-- MI -4-.. MI MI MI CA tn MI VI In tn .4.- E • . • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • al
M MI MA CD M CD SI CD CD M M CD CD CD MA t M M MI In MI m m m MA Mt MA 'Si tn Si 'Si M Cr-.
1 I I 1 I I 1 I I 1 I I -
In =-- CO r, CO CA 'Si NI CO -,-- Ill CO .4-- V, IN VI CA .4-- .q.. ro ,o -- co CJ No ro r, tn ...o n in, ro in
-- ro PO Cr- in -.0 0./... -10 'Si V, b7 147 1.0 V, t 0, r..1 MA 0, VD Si r4 Si uo ro bl MA tn 0, 1- CA sO 4-
J .,-- CD PO NO =- ...-- CD CD mm ....- CJ CA CD NI NO M CD rl NO CA ....- CD ,- NI ..-- CA CA .4--.4 .4-- r4 .g.. • • • • • . . • • • • . . . • . . • . . • • • • • VI • • • • •
M tn M M M M M M T... M M M M tn M M 'Si M M M In M M MA M MI MA M n Si M MI
1 1 1 1 I 1 1 I 1 1 I 1 I I 1 5-
SE ....- In CO CJ r, 0, In CO NI r4 ,- In n 1- b7 CO -4-. 0, V, CO .4-.. 0, V, s0 N rl N. ,.... ... OD V, sO ill
P., IN c...1 .1. .4- -- IN bl ,CI CO C...1 n CO VI M Cl sO PO b7 C4 .... V, Sr VI ro r, uo r, -- m m ..s. r E 1-4 M M M ,- CD CD CD r. n tn .,-- n n CD CD SO 'S CD CD =-.. -,-- 'Si CD CA CD CD CD CD CD CD CD 'Si
• • 11 • • • • • 11 II • • • • • • • • •••••••••••• • • ai
O 'Si *4 t54 *4 M M M M MI tn MA MA M In M) Si (Si tn tn on In IS1 MI M en MI In m m m tn m U 14. 1 1 1 I I 1 1 I liti I 1 I .-
co
ro CA ........ r, In -0 -0 In CD ,0 .0 c.-.1 .a.. cl ro n N. c...1 .vi- r, os 00 oo r, .1- CO 111 CD CO .1- ,0 CO .4-1
In NO CA CD 0, N. r, r, Cr Cl'- .- Cr- Si 40 T.. T- Cr- 'q- If, N. T- ON -40 -ci r, n -- Cr- .q. uo Cr. ..,-- !:-
mc rq m ro uo .4- m 0, CD CD M CA CA CA r..1 SO rl CA (Si CA NO In SO r, ton m -- MA VI .4-- VD 'S .4.- 115 • • • • • • • • . . • • • . • • . •• •••• • • •• • •• • • >
'Si M 'Si 'Si Si CA CD CD CD (Si 'Si SiC 'Si 'Si 'Si CD 'Si (Si 'Si 'Si 'Si 'Si (Si 'Si 'Si 'Si CD (Si (Si 'Si CD 'Si ) I I I I I ! I I I I I I I I (3
N .4. so -- r, Cr, so Cr, rq uo so m CO !cr uo bl 04 V, c...1 r, ..- N) -4- ...- -,tr -- 'Si CJ os N CO 0.` ..
cq oN -0 so in ,-- r, ,- so N. In -4- r, ro ,0a 'Si os No ro r, r, c..1 ro N) 'Si No CA MI .4-- tn 0- CI
CO ,... 'Si (Si r4 CA 0, 'Si St ,-- — CA CA CACD .4- (Si 'Si 'Si ,-- rq cs.1 .4. -,- 'Si (Si VD C4 'Si VD 'Si -4- .,- r-
ME• ••• ••• 411•11% • N • • • • • • 4 • • • • • • • • • 0
'Si MI n M CA CA CD 'Si CD CD M M MA ISA M MA M CD eD CD CD CD M M CD CD CD SI CD SO 'Si CD I I I I I I I I I I I I I I 1 I I 4-5
r
ro ro -- Cr. a. r, uo r, CO ro NO Cl'- .4- Si -0 Cr- Cr- L47 r. uo .i. vo Cr, CA VI -,,r .- CO ro ul N. ,0 T
,-- m 4:1 m m LI...) Cr a to -4- N') co ro r, uo -- rq CJ 'Si r, N uo ,-- oo .4- m uo .4- ro 'Si C'- NO La
14. N -- --- .4. 00 CA v m ,-- m rui CA CA CA CA M (N m m r..1 In (....! .,s m n -- -- -- -- ro -- -- • • • • . . • . • • • ••••• V • • • • * •Illt• 0 • • ••• 1.41
MA M CD CA CD CD M M CA MA Mt Mt M (Si CD CD (Si CD CDCD CD CD CD M CD CD CD CD Si I'D CD CP SE I I I I I I I I I I I I I I I I I
a)
-- so ro.,- oN -- r, CA IN r, r-.1 so --- L7 MI ON rl OD L7 MI MA ISA 1.0 VI CO .4.- V, 1.17 -CI r, m Cr. 4.-4 N sr ,- PO 'Si -Cl M St Ok Cr- .4- uo uo CO -0 CO V, 'Si U7 1- sO 0 r4 v CO M 'Si ro ,- sO VD Si J...5
uu 'Si m m oN .tr CA In ...- NO CD NO CD MI CD CA CD CD CD CD In .q. r.4 v .,-- -- ro ro ..-Mm -- ol M I--
SE
c-1 PO E r.1 sr 5_ .gc cw CD mc -J ra Mt CD CV CN CV CD G.. SE 04 LJ LW Li_ CD MC F4 -J NC Mt CD CN CN CV ,I 0
Li-
• ••• 4 %• •••••••• ••• 4 IRNMO 4 UN • •••••
Si Si Si Si SA S! ta Si n n n Si VA SI Szi Si n M tS4 M S4 'Si (Si 'Si (Si 'Si (Si m 'Sim m (5 I I I I I I I I I I I I I I I
PO 0.. SO NI ...- -.0 ,- 00 CO CD 0, CD r4 1. V, N. ON. .,... S! cq o, n ....r N3 00 In '4' ..-.. PO ln In CO
0, .,-- 03 =.- ..0 ,0 CA CA NI Cr- L7 .4- to -- r, CJ uo cl 'Si uo -4-. ,o N. In VI In 0, .1- (Si CO co co
CJ CD CD CD M =-. CD PO CD NI -,.... CD In CD ..--. CA ,I3 =-. CD UO CD ...- M NO CD NO eD CD In CD =-. CA ,r . • • • • • ...... • ••• • • •• • it ■ • • • • • • • • ••
'Si 'Si *-- 'Si m m Sit St t'M In Mi (5) Mt *4 St M (S4 S4 tsi 'Si 'Si (S) (Si (Si t!A St M MI 154 P.54 151 S I I I 1 1 1 1 1 1 I t I I
CO ,..-- Cr- -.0 ro .4.. c...4 un -- uo VI MA 1.1 VD r. CJ b7 VI MA M .4-- Cr- 1- bl 1- .4-- ....- CJ MI r, Cr- --
No Cr- ,- .4- m Cl- ro rq PO C4 1- (Si PO .4.- Cr. 0" -0 M '0 1.0 CO Cl .4-- b7 -4-- 04 (Si b7 bl -0 'Si Cr.
P4 CD 0, CD M ,- M 15, CD CD M MI ,..... M (Si M M Mi VD 1$) n l) -4-- (5) to M M -- M •-'5; .4.-- tn ••• • • 1111••••• •••• ••• Ili •••• 0 •• • ••• •
MI In *4 (Si m 1s, m 1St Sa ,S, 151 S4 Si VA Si (Si (Si 1:54 I:54 St (Si m '5) (Si 'Si M M (Si Si Si m 'Si I I I I I I I I I I I I I I
PO CO NI ..- PO CA N) ...-- Ul 0, N) CO NO rn Cr, 121 CO r4 b7 1" CO 111 CO V7 Lil Cl Cr. .4-- CO Cr. CO V1
r, ro 0- CJ --- rl In r, -- ro .4- m .1- m r, r4 MA V, 03 MI C4 r,...4.- -- ,- .4- m .0 co ,_ r.., ,D
Ct CO CD CD M CA ..-- CA OD =.- CD =,.. =-. CA NO S4 =-. In CD ...- ...- CA CD CA CD =-- CA ...-- ,-- CD ..r. m n • . • • • • • • • • 1111•••• ••• ••••• • •• • •• 4 •
MI tn MI M MI MI MI MI MI MI trA Si S4 VA S 1 VA SI SI 1:54 VA tA ISI Si VA St VA VA MI VA SI VA St
I I I I I I I I I I I I I I I
174
ro co cr• so cs co ..aci ro PO In CA CA Ar Al A- At AO 0, CO Co PI 0, A- At ,0 A-- OD Al CA ro PO A- eN Cl Co -4- Cl A- A- A- Co 4- Co -4- Co No
L11 .4 4 r, 1.0 NO L7 un Co N) cr. 4- in CO 4- uo CSA ,- to Co oo Co Cl r, co Cs 41- c.4 C4 Al Co CA In A- A- CA A- Co r0 Co CA 4-
••••••• 2 .2••• • /I X • • • • •••••11211••• • •
MI Co Co rA Co Co CA CA CA CA Co CA Co Co Co Co CA Co Co Co CA CA Cl CA Co Co Co Co CA Co Co Co 1 1 1 1 1 1 1 I I 1 1 I 1 1 1
CO Cs 1.11 CA N. CO AD r-, m- c..4 wo in -0 A" -zr to n ro or. ,r .4- 4- Co Co .4- CO 5-, At r Co co
N) N GI CO N) Cs Co 0, CA CA Co ,0 CA At -- 4- AD Co CA A, CO 0, In Co 0 -0 N) In NO AD A- CO CA
Cl Co ,- (-A .-- A- At CA Co CA Co A- N) A- CA AA N1 rA A- PO Co A- N) ,- CA CA CA CA Al A- CA 0, ,r • •••••••••••••• • • •• so••• • . . . • . . •
Co CA Co CA M CA CA Co CA Co CA Co Co CA M Co CA Co Co CA Co Co (Si Co Co Co Co Co Co Co M Co I I I I I I I I I I I I I I I
Cl' r*, In N. In N. 4- co n un co cs .4- ...0 ,r C4 '-C) If) Co -- ro un el cr. or to ..r Op ,- ..4-. U')
Cl ..-- ,o CO -0 (Si A- AO CA Ar Co A- AO CA ,0 NO At NA ND ND Co N. Ar Co At Al N) "0 A- (N r, ,- CA
Cl 4- Co -- Co ro Co ro Co -- Co Co .-- -- 4- Co A- ND Co CA CA CA Co Al -- CA A- Co CA -- 0- m -- 1. 111R ••• 1121I• ...In • • • • 1• • ••••• •••••• •
M Co Co Co Co CA Co Co Co (Si Co Co CA Co Co Co Co (Si CA rA Co Co Co CA Co Co Co CA Co (Si Co m 1 I 1 11111 1 I 1 1 1 1 I 1 1
CO CA N) NO ND 0 0) A-. -.0 At N) N. A, VI Cs N. 4- A3 N r, CO 0- -T Cl co Cl. Cr. co '0 co ...r .4.
A-- Co In Co A- N) CA 4- ,- Cr- Co c, m ,r N) -0 ND 4- r, ro Co (N N. ro r, 4- Co -0 CA Cr. Cl UD Ar
eN CAA- Co PO ..-- ro -- n ,-. ro CA CA In CA CA CA CA Co CA .4. ...- (A ,- is, A- ..r. -- Co AA ,- ,- --
..• • • • •••••••••• 11••••••••••••• X M
'Si Co Co Co Co Co Co Co Co Co Co CA CA Co Co Co Co Co Co Co Co Co Co tS, Co Co Co CA A- CA Co CA 1 I 1 1 I 1 I 1 1 I
A- CA A- If) CO Cl CA A- Co 0, AD -0 4- -- Cl. cs A3 (-4 In Cl CS AD Co At In CA CO -0 CO At N. Cs
AD 10 At N) At CA 0, A, CI Cs Cs CA AD CA N. CA A- Al OD .4. -- CA OD In ro uo r.„ un el -- to 0,
co ...- Co un -- -- M PO Co CA Co 'Si In Co A- CA In Cl Co 4- .- el Co 4- Co Cl CA Co CO CA Cl VI 4- CI
••••••••••••• 4.1 • • • • •••••••••••• • 21
Co (Si Co Co MI M Co Co Si Cl M Co Co Co Co M Co Co MI SI M Co (Si Co Co Co CA CA CA (Si CA CA ID 1 1 1 1 1 1 1 I 1 I 1 1 I 1 I 1 1 1 1 1 1 r-
cr,
0, -- AT r, A- Co AA A, A, Cl 4- -o -- Al ,r (N co co r, 4- .- to -- Ni -- N. Cl co Cr. un co tO
m Co CT- M In AO A- 1271 Co AD CO '0 AO Co N) co ro so Cl un --- ,- to 4- ro -o ro N. AD -0 in CO 1.11
At A- A- Co NI A- Cl CA CA Cl CA CA Co A- Co Al A- Co CA CA CA Cl CA Co Co A- CA Co Co A- Co Cl CA -s) • . • • • • • . • • . • . • • • • • • . • • . • • . • • • • • •
Co Co Co Co Co Co Co Co CA Cl Co CA CA CA (Si (Si Co Co Co CO CA Co Co Co Co Co Co CA Co Co CA Co 115 1 1 1 1 1 1 1 1 1 1 1 I 1 I I 1 t
al N -- un ,- 4- A- Co CA tO CS Ar A, 0, CA -0 Ch .47 N. CO or. ,r 4- Co At AD If) N. r4 rq cr. 4- Co ,...
At CA In CA CA rA CA A, In AD CA At CA At AO CA NO At =- A, CO OD CA CA Al CA NO AD CA VO tO A, Al • N Co m PO A- NO A- Co .- CA ND (Si CA Si CA CA A- Co •A- PO Co A- A- -- Co Co (N Co .4- -- Cl Co 1-. -
•••••••••••••120 4 •••••• •••••••• • • al
Co Co M CA Co Co (Si Co Co Co Co Co Co M 'Si Co CA Co Co CA CA (Si Co Co CA A- Co CA Co (Si 'Si Co 4- 1 1 1 1 I 1 1 I I I I 1 I
,:...
In At CO CO NI .4. .,....r r,... -0 -- N) CO 0, VI Cl a -0 -- Ll -0 0) ro Cl 4- CA 0 .- In CO Cs CA At CI
A- A- ro co 4. to r, to ro N AD to ro to PO un 1- un co IN rA A- CA r4 to, ro AD Al .4- un -0 (N 4-
J A- Co N) A- Co CA (Si Co N) Cl Co CA A- A- A- CA A- Co Cl (N CA Co A- A- CO Co A- CA A- CA Co N)
. • • . • . • • • . • . • . • • • • . • . • • . . . • . • . • .
CA CA CA CA CA CA CA M CA CA m m M m m CA CA CA CA CA CA CA CA CA CA CA CA CA CA CA CA CA 44 i I I i I I I i I 1 i IIII I 1 I 5-
.1.) PG ro un ro ro Cl A- N -0 CA Cl CA -O CO In CA Cs A- Co 4. co Co co r, Cr- 4- 1- to 4- Cl CA CA AO !Z •-U') In At CO NO *0 ND In CO Cs CA CA AO AO Co ND A- CA CO AO AD A, Al CA CA At In A, At ,0 CA E
E hi la CP Co A- (Si Co CA N -' (Si (Si (Si Cl Co CA CA CA Co Co A- ,- (Si Co r.„. ,- ..-. tsi 'Si 'Si ,- Co m 5- • . • • • • ••••••• •• • • • • • ••••••• 2 n • • • al 0 Co Co (Si Co Co m Co Co Co Co CA CA Co Co Co CA CA (Si Co Co Co (Si CA '5) Co CA AA CA CA CA (5) CA AI U.. I I I I I I I I I I I I I I I I I I I I r-
tr,
co 1r 41- un cs 4- co N co CN AD CA -0 CA 0, .- CA AO Os In ,- In (Si N. N f;.! ,- Co .4- 4,-, .4.- IJA •!..., ,r --- 1-.., el ,-. ert -0 ,r Cl N. N ro ro lo- N) A, A- NO cs .4- ND PI AD A, CA CA ND CO NO Co AA CO '...-
:C CA CA ro -4- .4- -- A, CI CA CA A- tO A- ,- AA CA ro m eq ro 44- ..- o, m .- ,- m -r -- ro -- ,-. III
. • . • • • •••••••••• • • ••••• ••••••••• >
Co (Si Co CA CA CA CA rA CA CA (Si CA CA CA CA CA CA Co Co M (Si rA Co MI 'SA Co Mi t'S! m m (5) m a i I I I I I I I I I 1 I 1 5_3
1.11 0, CA CA AO AI CO Os A- CA Cs. N. CA A- A- CO Cr- CO AA r, r, .4- In co ro ,r. ro -0 0, -0 4. -0 ..
N. CA A3 In LO N Cl'. N. CA ....' M .I...' r‘.. ...- 4- Al A- A, A, AO N. N1 ND AO A- CO A- CA A, AV In CA G
0 mi A- GI cq N ..T. m m -- -- rq m N m to m Co CA 'Si CA CA Cs A- CA CA A- CA CA CA CA ro el .
. . • . • . • • . . • • • . . • . • • • • • • . • • • . • • • • -r.,
m in m m Co m m m m m Co m CP CA CA CA CA CA CA CA CA CA CI Co (Si CA (Si 'Si CA CA (5) Co I 1 1 1 I 1 1 1 1 1 I I I I 1 1 I I 10
r.-
co .2... CP-. Co .4" ,- CN T- -0 1'0 rq T.... M Cr- ND CA Cr- CO Cs At At N -- co CO V ,- 0- CO ,-- 4- un !.1
Cl CO At sO N rr.. ,-. ..- Co vo Cl co CA tO Ni N) AO Ar Cl (Si N) N. ,0 Li) CA CO -- A- Cl A, Cr- N) CI
U. Cl Co ....... .4- un Cl in .,-- CA Co CA Co Cl CA Cl Co CA Co CA N) Cr- CA AP A- CA CA CA CA A- CA -- A- ▪ • • •••• • • • • •..• • •• •••••• • • •• MI
Ma MI Co MI CP Co Co Co Co Co (Si CA Co Co (Si 'Si Co Co 'Si CA CA Co CP rA CA 'Si (Si CA CA Co CA (5) ANC_) I I I I 1 II! I I I I I I I I
at
,r 'Si Cl 'Si Al A, N. CO N) Cl Ar In AD In Cl A, N M IN 4' 'T N n CO '0 0-' .q.. Ol N n 0... N ,..1
Co 0 n 4" N N) N. CA ND At CO N ..- 4- CN Cl In A- 0, r"., CA Al At CO (N r, un 4- (Sr Co OD A- i
LU A- 'Si Co In Cl Cl Al A- N) A- Cl Co ,r Co -- 'Si Co Co Co to... N) C..1 N) ,- Cl. N) CA ,- ....4- Cl 'Si ,- • . • • • • • • • . . • • . • • • • . • • ••••••••• • • j•••••
M Co Co 'Si Co Co 'Si M M (Si (Si CA Co 'Si (Si (Si Co Co Co CA Co Co Co Co (Si (Si 'Si Co CA AA CA CA 1 I 1 I I I I i I I 1 I I
In to Co AO N. CA At r, 4- r, uo r4 ...- un -CI N) vo N) N) N. 0, I..0 0-. 4- un co r, un r, un -0 4-
CO -0 Co AO CD AD AO AO Cr- A, (Si CA N) A- AO IN r, ro ro or. N N. cr• 'Si CO ,- cr./ co -o No r, ,-
L) ...- Co LC) Co Co .- Cl Co N Co Co 4- Co ..- N) 4- ,- Co OD CA CA Co Cl CA CA A- (N At CA Cl in 1.1") • • • • • • • . • • • . . • . • . . • • • • . • • . . • ••• •
Co Co Co Mt Co Co Co CA Co Co Co M (Si Co Co M Co Co CA CA CA CA CA Co (Si Cl Co 'Si Co Co Co Co I I I I I I I I I I I I I 1 1 1
CA N) A- CO AO r, N cr. ,-. c...1 4- -0 m N) r, ro ...o r.... N) CA CO CCI AA Co A- r, CO CA "0 Co Cr At
N Cl CA CA Cl 0, At N) Co A- Cl CA Ni AO r", ND 0 CO ND A- 4- r, -0 -- 'Jo 4. -o N) r, N) r4 n
04 Co PO (Si (Si CA Co Co Co Co Co Co M M Co 'Si M Co r, Co Si 'Si Co n (Si Cr m m CA m m -- Co • • . . • . . . • • . . • • • • . . . . . — • • • ••• • •
Co (Si Co Co M Co 'Si M M Co Co 'Si Si Co Co 'Si CA Co CA M CA Co 'Si Co Co CA Co Co Co Co CA Co I I I I I I I I I I I I
CO in Ch C.4 ON N N .h.. Cl CO Cl. Cl. At CO CN ..... 03 ,0 VD CA 0, 0, CA A- AD A3 CO NO 4- ..- N) lin
Co AD 1.11 r., c-4 Co 0, AD Cl 4- CA A3 r.... r, Ar AA N. ,0 N. UO N) A- A- N1 At N) A3 A- ,r ro Co In
<X In CA A- Co Cl Co Cl 'Si M Co Co ....... M Cl Co ,- N. Co ..-- (Si Cl Co P7 Co A- A- Co CA (Si N) Co M • a • • • 1. • • • • • •0••• • • • • ••••••••• • • •
Co Co Co CA CA Co Co CSI CA Co Co Co CA Co Co Co Co Co Co Co Co Co Co Co Co Co Co Co CA Co Co rA I 1 1 I 1 1 1 I I
AE i.
E ...- c.4 Ni .7- E — N I,r) V- A. .4C PA 5-.) LW U. Co MC 1--1 ...I AC NC CD CM Cl Cl Cl C. *X Oq 6) UJ U. Co = 1-1 -I AC At Cl CA Cl Cl CM 0 0 U. U..
175
r, No (-1 r, No ro V1 n co No -- No r, r- sr un 0, CO CO 4- sO Cs 44- r, r" ro
.tr ,c1 t tSi N) EN CO r ,0 4- Si N N r -CI CA N7 CCI r.r I IS; 1.0 - CN r- (5) N t51 1'0 m EN co ta m 4 IS) r m rl m N) a m -4- m CA Ul
masa.. • • •••• • • • • 11•••••• ■■ ••1111••••
M CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD Si Si fAntantamtamm I 1 I 1 I I 1 I 1 1 1 I I I 1 1 1
• r, 4'0 Cs 00 r, M In M r, ....-- PI CA 4- CA PI NO ..... Cl-. -Cl r, 0- sO CO N) CA r- r0 ,-- -o -o ,---
PI SI 4- M N. CO 'gi- NI PO ND CA In -o PO sO NI CA Si ro .q- 4- co c..1 rg .1- clop N) ..4- Cl r...1 If) ,1.- 0.
C5 CD CD CA r- ,-- PI M ,--- CA r- CM r- M to N. Ni ,-. M .- ,- ...- po m m m ,- r..1 ,- ,- r 44- Cl. • • . • • . . • • . • . • . • • • . . • ... • . • • •••• II
M Cl M M M Si. M M M n Co M M Si M M Si S4 fili M M (5, M M M n M SI M fA M M
1 1 1 I 1111 1111 I 1 1 1 I I I 1 I I
CI M CO CA sO PO r- r- N - m If) co m Tr N. CO In rl N) Cs N. sO CA PO PO -0 N. 0, 0, If) ,--.
CA M NO -0 CS If) ,..-- CO N) a- -- r, r, m - -Cl No .cr Co uo ,-- 4- c..1 - If) -- - c..1 If) r, n .4- CO
CN rl I:A ,...- SI N) tA N) .r.- Si Si Co ,--. IS) CP M ..---- CA (Si 51 Si CA V.:A PO ,--- CA M M ,--.. M .tr m m
. • • • • • . • • . . . • • . • . • • • • • • . • . . • • . • •
CSI Co Co M Co Co Co Co Co IS M tA (Si M IS) Co Co Co Si Co M Co (Si Co M Co M M SI (IA n n
I 1 1 1 1 1 I I I I I I I 1 1 1
N .q- CO Os -'Cl N VI If) CA Co If) SI r- CO CA sO .q- N) m In SO 4- N. 4-0 N) r, CO PO CO CO On ,-
.- N SI ,ON Li) ,..- N Li) PI `AD 'V Co '0 M N) ..--- C.4 SA ,-- N 'V CO CA ,-- ,0 Li) ,-- 'f- 15) 0, If) .-0
0 Co Co M CA .r.- .1-. ,-- Co I- Co r" ,-- N Co Co Co SI SI SA N I- .1- T. •■•• SA ,- ,- M tr n M M • 4 •• • • II•1•• •••• ••••• 4 ••• ■ • • IR
IS)•
Co CD CD CD CD Co Co Co (Si M M M Co Co M M CA M M Co 15) tA M Si Co M M Si M SI M
I I I I 1 I I 1 I I I I I I
(N M In r- ,--- In r-. r-- M -Cl In Os Co If) r.1 ,- N) uo N) CO CO '4" Os P., CA CA ,0 -T. CO rd ND ,-
-0 M M ill 44- ..-- Cl Co 4- 4.- Cs PO Si r, -Cl N. sO CO r, 00 uo 4- .a- ..-- -q- ro ro CA CO 0- CO N) 0
CO ...- Co .q. Co ," M CA r- CA Si M N. T... .,.. .,-- .q. Co-- N) Co ...- M CA r- CA Co M TV tA ,...- CA 4.0 • • ••••• • •• • • •• •• ••••• • •• ••• •• • • • 0
M n M M M n M M M M IS) M 15) M M CM Si M Si (IA M M M Co M St M I5 SI Co Si (Si r-.
1 1 1 1 1 I I 1 1 I i 1 I 1 I I I I 1 I Co
MI Os N. r- ,..- Cs PO 00 fA r- Si If) In M .- NO M T." N. -0 N) ,- n PO r- 4'0 N. M CO OD Cs 4.0 Cl
,C1 In VI 0, CP ro .,-- m co r, sD 0, '4' r, n 4,, In M ,0 OD r... M 4.0 0 PO ro CO M CO CA CA SA U
Z 'Si Cl ..-- CA ,- ,-. CA M M M CO M r- M M Si M r- Co ,-- CA CA CA CA Si CA ,--- r- r- r- M M r- • • ••••••• ••• • • WV ••••• • ••••• • ••• • al
Si fA SI M CA M Co M M n M Co Co M (Si Si (5) Co Co M Si M Si M (S Si M M Si M 154 CD •.
III)! I 1 1 1 1 1 1 1 1 1 1 1 I Li u'i
*A 0, Cs ta 4- r, ro CO NI M ,...- '40 M r--. h, 07 In r- CA If) sD .q" tA N. N CO SA .q- -- co N)
-40 In In M MO Os ,-. Si CA Os r, 4- -c -- CA M CA CD CO CA Si M ---0 0, Cs CA PO 0, Cs r, 0, ..4. el
Co M Co ,-- ,-- Co M CA Co CO M M M M r- r- r- M M T- ,^ M M M M CA -- SI ,--- M SI SI 1-4 • • • ••• • • ••••• • •• •••••••••• • •• •••
Co M Co Co M M Co M M M Si M M M M M KA (Si Co M (Si M M M M (5 15) M SI n CA (5) :-- I I I I I I I I I I 1 I I Cl
5-
..,:r In CO CA I5 CP 'cl- If) -0 CA fA M CA CA M 4- N. uo ,0 4- - 4- - n .4. Cl s0 CO T- N) r, CA
N) CA ,-- CA Co r- N) If) SI CA CO 4- N) a- ro a- -A- uo OD r, rl '5) ta 4' CA CO CO r- CO N) if) N) 5..
--I CD Mt N) ro cl ..-- - 15. CO M M CA ,-- M CA 4.1 M M CA r- CA (N Si M CA M r" CA ,--- M CA sr CI
. • • • • • • . . . . . • • . . • . a * • * • • • a • • • le • • r.!..- MI M M M M M M M M M M M M M M M M M M MI Co M St S4 M M ta m M SI M Si
1 I I I 1 I I 1 1 1 I 1 I >: •Ti
Z 1'0 ..-- 4. No cl., NI No CO If) CO CO r- n ,--- if) M CA r, N. sO 0, CA CO fA --0 CO CA 4- 4' 4- -0 CA 1.
'0 sD M ...-- sO n CA n In M M M 1.0 PO ro -o rl to r, .,-/- n M ,-- sD Tr ,--- •4.- 00 s7 In .CI Si .1-, E N4 Co M M M M M M ,0 M Cl M T- MI ,.... T-. M T... T." M 15) M M M N) M CA Co M M (S ,-- ,--- i-C, S.- • • • • •••• • • • ••• •Il •••••••• • • ••• ••• P.
O Co M M Si 'Si M M SI SI M M Si fA M M Si 'Si Si SI KA M M M M M M M St n M KA M Li.. I 1 1 1 I I I I I 1 I I Cl
u
CO If) CO M M ...- sO -0 'cl" VO N) N PI ,-- Si n TI" M Cs N. rl CO CO r, Co -0 Cl- Os ...." CO M (5 1.---- ▪ - CA CO If) KA SA (Si NI 4'0 r- ,..- CA N. CO PO r- .q- No O. OD 4' ,-- N) 4- cl,.• r, CO N. (-I No n -0 6.1
S PO r- r- If) -AO M M 'S ...-- M CA CA r- 4.0 KA CA 4. m (S (N uo - r, n n - ....- Cl rq .,..1- - ta •,1
. • • • • • • • • • • • • • a • • • • • • • • • • • • • • • • • . r.- M M M CA M CA .-- M M M M M M M M M SI M Co M M M M M M M M M M Co SI 'Si n
I I I 1 I I i I I I 1 I 1 1 I ..-> 0
If) N sO CO CO ..-- r- CA ON N 0, if) N. PI N. MI PO FO M 4" '47 Cs -Cl r, Os N. ,--- '0 '4" 0, IN PO L)
M r, PO M n sO M In -,-. CS PO ...- ..-- r..., .4' CO sO r- Cr. PO '4" M CA r, ...- a- .4- c.A Cl uo r, a-
CD ..-- CD Co CA ...- N. SI 1:5/ r..- M ,-- M r- M PD M M Co M ,..- ,-- In M 15) M ..-- CA 15) ,- M 4-9 r- •11
• • •••••• • • • ••••• ••••• • • • • • • • • I.• • a,
Co M M Cal M M Co Si M M M M M SI Si M M M M Si KA Co SI M M (S M fA St Co M m I 1 1 I i 1 i 1 1 1 I I i I 11 I I C'-.1
Co .4- CO ,-- N. CO M CP Si 4- ..- -- -0 ,13 CO -0 PO 4' CA If) Ul 4- -- if) ijo ,-. rl 0, ro - 4- (-1 .4-,
n -o co -- to Li) m 4D M ro O. -A- If) If) CO CA .,-.r -gr c- r, r... Tr CA CA CA ,- CA CO OD Cl '0 IN r-
11.. PO r..-. r- ,lt CP ,-- -0 Co CA ,-- ..-- r- ,^ N) ,-- ...-- PO r- M ....- -0 CA In r- ..--. ,-- CA ,--- -,--- 4- M SA ill • • • • • . • • • • • • • • • • • ••••• ••••• • ••• • Ci. C4 M to M M M to 'Si Co Co M Si 'Si m Co m CD CD CD Co CD ta n M SI M M Co M M SI M
I I I I 1 1 1 I 1 I 1 1 1 I 1 4-
C?<4:
r- r- .tr r... CO Co SD CA Co -..- r- Cl- CA 0, N. fA n N -q- Cr. OD r, PO CO ,-. r, oN . 1.- N) CA o,
..-- in - .,-A- - m uo -- CA M CP In Ns 0, N. PO ,0 In ,0 In M 1.0 0, PO CP NI PO n r, CS 'CI IN Cl
LU ...- Co CD CN ..4. CA 141 SI N) ..-- CA Co NI M ...... 15) CA (Si Co .a- 4- N) PO ,..- Co N) .q. m ro -- n ..-- ,--1 • . • • • • . CA
S • . • • • . • • • • • . • • • • • • • • . z
CA CD CD M M M M M M SA M M M M M M SA M Cil M M M St M M M St M M St 65 I 1 I 1 1 I I I 1 I 1 1 I I I I-
.4- a- r. m 00 sO CO 4" CO 0, N. In CO CO PO CA r- In PO CP CA uo P. t'... 41 CD C) 4- CO r, ,0 CO
00 M r, - OD HO OD M ,..- n 441 M SD ,0 M M CO fA Os SI CA n sO ,- r, .P r0 ,0 r- 1.0 4.- 4-
U Co ...-- 0.-. CD ,- Co ,- Co PO 'Si ...- 4' Co ,-. CA In SA Co N) M CA M N 'Si CA Co r- N) Co CA CA Tr • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • CA M M Co Co Co M M M Co Co M C11 M Co IS) Co M Co M tA M M fA M M Co Si 'Si SI Co M
I I I I 1 I I I I I I 1 1 I I
CA rte} rt" r- .q.• Cl n ,-- n Cs Cs Co .nr Co r... No uo 4- 1.0 SO In -0 ...r ro c-. N. No CO Cs sO rA .q.
N ,•-• Si n ,CI N C.4 ,0 CA If) V) Co -Cl -.0 4- Co ,--- NI ..-- ..- (f) CI, ,II •-• NI MI N CA CO ,- TI- ,-
P4 CD ON. ,-- CD ,..- CD ..... CD CD CD CD CD M M Mt M M MI M Co r.- Cl fill ,---. (Si Co Co M M ,-- Co Co • • • . • a • • • • 4 • • ••• ••• •IIII•• ••••• • • •
Co M Co Co Co ta to Co m Co m (Si m Co n m m to M M CD CD CD CD M M M fA M M M M
I 11111 I 1 I 1 I 1111 I I
C'..I C..1 ct ,- Co n 00 ro cr to N) N r, Co '- r, MI N -- n VI CO r- sO 00 Co ,-- N) 1.0 ,-- CA VO
CA N. CO r- Co (Si 'V' sO N) NO sO sO CA SI M sO .q. m ...- ..-. ..-- '40 -0 PO Cl--. r0 r0 'ql" N. Cs ,-. .q" <C CO CD Co ..-- 4.1 ..r. N) m to m Co ....- m ro m Co In r... r- M ro Co N) IS) to m m ....- Co ro - Co • . • • • • • • •• ••• ••• ••• •• • ••• •• ••• 1111
Co Co CD Co Co Co CD Co Co Co Co C4 Co CD CD Co Co CD Co Cal Co Co C.4 CA Cl CD to CD Co CD CD CD I I I 1 1 I I I 1 1 1 I I I
04 MI I".7 sr sr 04 GJ Ul U.. CD MC -.I AC .7.M CD CV CV CO CO c_ 04 C. W 1.4.. CD MC AC Wit CD CO CO CO CO
0
176
PO a- co a Cl ro m Cl CA +- PI +- +- +- N) m uo n m CO Cr-. ,4 ON In in T-• in 54 .1- a r, r, No n ro n ro No Co a N co n CA KA In h, A- r, m a, a CA VA on m m m m n m N) KA KA CA In KA KA m m m m m Co
• • • • • • • • • • • • • • • • • • • • • • II ••••••••• MMMMMMMMMMMMIMM M M M trig MMMM M M M MMnM I I I I I I i 1 1 1 1 I 1
CA CA NO CA 'gr h, KA NO h, NI a ro Cl-- DO NO A- Ai +- +- CA NO CO In +- N. a- ts, r, -- n uo n
PO +- '1- a- No No r, m No In 0, CA OD In 'gr "gr 0, +- NO KA KA In a- No a n a- uo ro m a- -o .,71..
cr. — n cq m n N) -- -- CA KA KA CA KA KA 'gr CA KA KA N) KA +- NI +- KA A- +- CA CAI KA KA 00 N1
/••21••••1111•• ■ ••• • • V •12•••••••••• a
M M M M Mi KA KA KA KA KA ISA KA Si SA KA M KA KA KA KA KA KA SA KA Can KA KA SA KA lx) KA I 1 1 I I I 1 1 1 1 1 I I I I I I I I
+- .-4D Pa N) +- Cr-- co a NI CO CO CA CO Cr. '<I In PO CP NO — a- -- r..! No -O uo -- a N. No KA In
CA 0- -- uo ON CA In NO In ro N IN0' ON KA VI a- N MI NO 0-. SI M T". MI a CA ro No co uo a uo cv ro -- CA +- 'q- m a- m n n -- -- r a ts, .- ro n -- m a n a- -- CA m n -- -- a n n
. • . • . • . • . . • • • • • • • • • • • • • • • • • • II • • • M M M MII M M MI M M M MI M n M MI Si M n M M M M MI M M M MI M Si M MI 151 1 1 1 1 1 1 1 I I I 1 1 1 I 1 1 1
In CP co a ro a- -- a -- ,-. ro OD OD ON TT. +- CA a- No N1 AO a !Jo -- ro 0, No a in r, -- ....-
-- r, co ..-- r, co n CA NO CO ON CO CO M N. Cl N) — Cl CO ON -CI CA CA CO NO CO CA If) A-- CO KA CA
CI KA KA KA NI ,-- +- CA KA +- +- +- KA 'gr KA +- KA CA KA KA CA ,--- +- +- KA KA PO +- KA ..-- +- KA +- • • • • • . • • . • ••••• II • •a• •• ■ ••••••• a a
M M M M IS, 'Si KA KA KA KA M in M M m KA 151 CA Mt CA M IS) M MI CA CA MA 151 +- 151 CA M I I I I I I I I I I I I I I
NI CO 'cr CN CN -CI a a- co a- n ,- ro r, ro N. a- uo -- ro -..- IS -- a NI NO V1 CO 't a r, a,
a- CA No uo CO CA N. Co T.". 0. M CA a 10 CA N. a- 1 ro a CO Ar CA KA MI +- N. AO 10 --0 PO ON Q.
OD +- M N) M — M CA KA CAI KA A- a n -- -- a -- m a n ,- m 1,0 ,- CA m m N 15 ,_ CA m :-..!
. • . • • • • • • • • . • 11•• • • • ••• • Il••••• •• • ri
M M M M 151 15) 151 151 1151 M M CA CA CA CA MA MA 15) CA CA MI to 15, CA Cal MA MA 151 MA M CA f- IIIIIII I I I IIIII I I I Cr,
--O M N. C..I ,-.. 0.- C1 NO M N NO CO -0 ,-. a- -- r..! co No a co m -- a r, a- In -4D ,- KA In 0
NI CA NO NO CA a- co a- co ro co ro ,-.. CA a- ON CA 10 +- ND KA CA h, +- 1.0 CA r0 IN CA ro uo 0, Li
74 'Si MA +- 1 N N — 157 .- +- +- KA +- KA CA KA KA M ,- CA CA NI KA KA KA CA 0, KA A- KA CA ,- r- .- • • • • • • , • • • • 114 ••• • 1.•• • • • • ••••••••
Iii
CA 15) CA 151 CA CA 15, CA CA MA CA 15 KA KA KA KA IS, KA KA KA n M Si 151 KA KA KA KA 151 ts) m m •m—I
I 1 1111 I I 1 I I 1 1 1 1 I I Li U)
SI r'.. CA +- +- r, No co c...! co PI CA h, NO CA OP a r, AO +- CO In PO CA h, 00 1N. .'40 0.. uo a- uo
PO ro a- ro — a- r... ..--. co CA PO NO In ,-- NI In +- a- uo No uo -Cl No co uo -'1 rA -- co c..! a n al
CA CA CA NO +- +- +- N M N Cl MA +- CA +- SI Mt M M! N fA ,- ,- ,- Si Ct. I N n ro T.! — m r—I • • • • • in • • • st • • • • • • • • • • • m • • ••••••• • IT;
CA CA CA 15, CA CA MA CA 15) CA MA MA 15, MA MI KA 'Si KA KA 15) 15) M M '5 15) KA KA KA KA KA KA KA F• I 1 1 1 1 1 I 1 I I I I ID
CO ON 10 00 10 C► 154.0aCA ,- (A ro ro PO N In KA NO AO +- P 'gr CP h, 0 N N AO N 0,4-
CP NI N a CA +- 0, "4" CA ON In '71- NO ,-- CO 10 N. .q.. N) '10 "4- NO +- VI 10 DO NI NO a- n N. *--
-.I KA KA CA MI +- CA CA CAIN m m c..! m rq M PO M M Cl ,- — M ,- M OD 'Si 'Si CA KA Cl +- CA Cr
•••••••••••••••• • ••••••••••••• • • 4-
CA CA CA CA CA KA KA KA is KA CA CA CA CA CA CA KA IS! MA CA MA SI M CA CA Mt CA KA KA KA KA KA 1 1 I I 1 1 1 1 I I I I I I I
AA al NO N) 1."N 1'..0 uo r, N. KA KA N. PO N. N) CA CO In CA 1...... Cr-- M Ul 00 1.. n .I. 01 '". 0'. — '0 T... ..40
N) T... .I..... PO N ra a No a- Cl-- uo ,- ,- If) a m No m m No -- NI CA PO +- 00 +" KA CO KA 0, +A AA £ 1"..1 M "..— M ,... ....— M IN) M 15 157— ,..." —5715.1 15) MI 15) ....... ".... "-- M CO M ,— M ...... M "-- M 15) al C. • . . . • • . . • . • • . • • . . • • • • •••••••••bq £ O 15, KA 15) ISi 15) 15) KA KA KA KA IS) IS) KA ISi 15) KA M ni 15) M M 151 Si SI SA S4 S! Sf Si Si ts, ts, U. I I I I I I 1 i 1 1 I I 1 1 1 0.1
u
,- .q.. N. N. ,- NO CO CO ,-. M 1,1 0.- N. NO -0 a- CO CP N. ON CA +- CO a- a- ro ts) -- In CA In 0,
^.10 .0 -40 CI- CA Cl N) +- KA NO NI At CA +- CA N. 0, N) IN T'. CA 1.0 CO CA ,0 NO IN CA CA +- -0 h, !li
MC IsO n CA ND til KA P, KA KA KA CA CA A- PO KA CA N1 KA +- c....! a- m co m -- -- ts) ro — a -- -- •-, • • • • • •••41.1.111.••• 1I•• • • • •• ••••••••
5)-
ts) M 15) 15) IS) IS M m n n n 15:i n 137 I5 n n n IS n 1:34 n m KA IS) 'Si 15! KA KA IS) KA KA Ai I 1 1 I 1 I 1 I I I I 1 I 1 1 A.
0
Co NO In CO 'gr CP CO CA a- a -- a a- r, Cl- co ,- a. ,- m a- co -- CD CP In CO KA a -- co 07.1 C.:1
-0 0.- N) uo a KA +- KA KA 15 IS) 'gr CO CA CA In N. CA A71 KA CA N 10 PO a- -o cl a- CA In a uo
(Si CA MA IS) ro CA III +- KA CA KA CA IS) +- KA PO KA KA KA KA CA NO OP KA +- KA +- N) m -- ts) ro -- ... • • • . . • • • •••• ■•••• ••• • • • •••••••••• 0
M 'Si '5 MA CA KA 'Si KA KA KA 'Si KA KA KA KA KA KA KA KA 151 KAKA KA KA CA IS) 'Si M KA 15) 1) is 'I I I 1 I I I I I I I I I I I AA
NO PO CA CP LO NO VO ON T- ..0 N) CO ND 0, h, CO a- cA PI r..... TT a rq uo -- co a -- No a ro l5) .-P
-- uo CA m r, -1- a uo c.., IS) -- uo a a CO -- CA -O r.... .'T N) c..1 CA -- KA AO KA CO NO KA In KA 5.-
Lt.. NO +- CA a -O— in 'Si Cg +- CA +- +- CA +- +- PO KA KA +- CP N) 1 -- -- m c...! -- -- a— u
ts) m tsa Si St ,$) n ISA n 151 tom n n 15t 13.1 n 151 n m IS m M 15) 151 M M M M M n n I I I I I I I I I I I I I I I a
.(1:
CL its
z CA ro a E CA ro r- CC OA C.) W (.L. Co MC H ....I 3= Mt CD cn CN Co Cb 6.. -'Cl 04 C.) W L. Co DC t..•i MC Mt CD Co Co ci ci 0 0
LL W6.
In NO O.1- uo a- N.. "0 'gr LO NO CO In NO NO OP 0, NO N In N. KA CP KA In A- AO NA I'A.1 A- (-1 A-
,-- ,- KA. (r) IN tO CO "1- IN C.4 CO CO CA — ..zr ,-- Cr, "-ID V.1 C'4 'I' n ,--- -0 -O "4:1 "0'1' 0-- Cr, n C4 al
LU CA KA KA 'gr +- +- CA KA +- +- +- KA CA KA +- KA to KA KA CO +- c4 C..1 ,- ,- C..1 C.4 in cl M m ,- ,-.4 • • • • • • • • • • • • . • • . • • . . • . • • . • • • • • • •
CA CA CA CA CA M M MA MA M to KA KA KA KA 154 KA KA K KA KA 15) 'Si KA I5 KA 15, '5) KA IS) 'S to ..i17.
1 I I I 1 I I i 1 I I I 1 I h-
+- VI ro r, CA KA OP N ".10 0..1 r"... 1"0 M I'0 ON In .t. a ro r, ro -- r, a ,o uo CO +- NO ,C1 A- KA
+- 4--- OP NO CP OP OP r, co co No r, ...-- if) a- n a- a ro ro r, uo h, KA AAJ 10 +- PO CO NO KA KA
C.).— CA ro m m m m IS, rA m m to m n ,- a .-. n r, ts., m ts) -- m c...! m -- a n -- ro a- • • . • • • • . • • • • • . • • • • • • • • • . • • • • • • • It
M M M M M M 151 KA KA KA KA KA KA KA KA KA KA KA KA IS) KA KA KA KA KA KA KA KA KA KA KA KA I Il• IIII I I 1 1 1 I 1 I
r., a- uo uo a- ro KA r,. uo -- .-- If) ro If) +- a n co a- No 0.4 a Cr, r, I5 N CA UO Al- CA A- Lo
NI N) KA In a- .,..- No ro In CO KA CO KA KA N) a a- N) Cr- NO ,10 C'.4 1...) Cil a a- uo a- c...! m PI CO
aci -- NI m m -- m M — M S4 +- KA KA 15) 154 KA KA Cr, KA 'Si KA to KA KA KA KA KA KA KA 15, KA KA • • • • 1. • • • • • • • • • • • • • • • • • • • • • . • • . • •
CA Mt CD CA CA CA CA MA MA MA CA CA CA CA MA M CA CA CA CA 15) M MA MA CA MA MA MA KA KA KA KA
I 1 I 1 1 1 1 1 I 1 I 1 I III! I 1 1 1 I
N) In +- KA N) ro a- CA N. In 'Si ro a- co No — CN M a- Cr, a- ...- co CA uo a -.... a C.4 N) a m
C ..... co No a '0 a- rA a- CA In ,0 CA a- m co Ar a- a- Cl- c..! N. Cr. No r, -- CA a ...- N ,- IN
‹t VI CD CA N PI Mi a ,-. m ...-. M +- MA N +- CA CA 5i— MI NI CA ro es) m m m -- rq ro m m . • SI • • •••••• ••••••••
CA CA CA CA CA CA CA CA CA CA CA CD CA CA M MA +.- CA Mt CA CA CA MA CA CA CA CA CA CA CA M to I I I 1 1 I I I I
177
r, D1 SO OS '40 N) +- +- In PO -o VI CA CA V, M In CO In VO M U7 CO PO Os r, -o co si a Cr- No
a Cr- -- Cr, r, co uo 1' 1' CO 1- M CO U7 CA r, PO CO CO so ro -- CO N Co up CO ,.r v, oq !ct- os --
cn n CO bl CA CO +- N) +- CA ,- +- In CO CO PO Cr- r CO in -.-- -,-- CO 1- Cc, Cl M ,-- U7 CA .- ro No
. • . • • • . . • • • • • • . • • . . . • . • . • . • • • • • .
CO n CA CA M CA CO CA CO CO CA CO Cl CA M CO CO CO CA CO CO CO CO CA MI CA MI fA CA CA M CA
I I 1 1 1 1 1 I I I 1 11111 1 ; I I
co .- CO ci, r, m r, — CO co Cr. to,-. CO co CO r, r4 N) b7 V.. CO CO 1- Os CA sO CA -,-- Cr. U7 PI CA
N) co N) CO r..1 cr CO N) co N N N) PO N CO ,0N to VON CO CO CO CA in Co ,0 CO N In Cr- -
C5 CO CO CA .- M U7 +- ,-- -!-- CA CA ro — CO Cr N) n CO N) CO +- 1- -- CA -- +- CA ro CO CO D7 N) • • . • . . • . • • . • • • .. • • • . • • • • • • • • I • • •
M M CA CO CO M CA M CO M CA CA M CA CA M M M M M CO CO CA M M M CA CA t CO MI M I 1 I 1 1 I I 1 I 1 I I I I I I I I
CO ro CO p, .q- If) -- ,- co co ,- uo -- -- co Cl C4 CA -q- a CO U7 Cc) D7 ,-- Os sO .- M CC) ro a
CA 00 •ct PO ,-- V, CA PO M ,0 +- PO 1- CO M M fA r, uo m — — -- No ro -- ro co ro n No n +- i to CA n — VI +- N) M M M - M CO M QA CA N) fA M M ro — c.4 n m m CO m m a M CO
• ••• • ••••••••••• •••••••• • •••••••
CO M n Si n M M CSI t M M M Cr ....... CO CO CAM SI M t n Si CO M CO CO M CO CO CO CO
i 11111 I I I I I I I I I
r, ro r4 Cl Cs sO CO CS CO M '40 ,... 0 T.. M CA Cr, sO VO M CO a- Cr- P7 N. ,-- V, 1- fA CA +- 1-
+- M ,- sO V, 1- CA CA VO M CO 1- sO M M CA 117 Cr, VO M U7 CA •-• Ul sO -,- +- sO ro co r4 CO r4
cn m m m ro c.4 (..4 CA CA CA +- .-- +- CA M ,- CO CA Cl MI ro r4 r4 -- — (-4 CA ,-- +- VO CA CA M
••••••.••••••••. •••••••••.•..••• CO CA CA CA M 15) M SI CA M M M .— M M M 15) M MI M M CA IS) M MI CO n CA M CO M M
1 1 1 I I I 1 1 1 I 1 I I 1 1 I
m r, Cl 0% Cr- VI sD 1" CA M -- M +- In Cr- V1 +- .4- -4- OD U7 OD 174 r, No OD V, OD PO CO Cr, OD
In CO U7 CA sO a ro 'q' N M OD OS SO a PI OD a CA CO CA V, 1"CO a- cl ro N. ....- No n oo cr
CD ,- M In r4 -- — a- -- -- CS -- Cs ,- M PO b7 ...- M In CA +- M '‘r- 4.- CA ,- CA sO M CA PO 'Tr • • • • • ••••••••••• • • • • • • • • • • • • • • • • u.
CO CO CO CO CO n CO M CO Cl CO M CO Si CO Car Si M CO Cl CO Cl M CO CO M Si CO CO CO M Cr 1-.
I 1 1 1 1 1 I I I 1111111 1 I i 1 to
r, (-4 CA CD VI CA V, In N. CA sO +- sO +- Os sO +- CO M M N. 'Cr M 00.. n SO rl Cc .,.. n '''''' T''' !il'i:
bo a a m CCD uo CA a — Cr- N. OD a N) PO IS) a a cl Cr- CA CO CO sO MI U7 '— sO sO N') N) a
z CO m — ro ro CO N) CO ..- CO CO +- -!-- ,-- Cl +- +- CO +- .- N) m 04 CO m .- 04 .- CO — CO ts, e-, • • • •• •
• • . • . • • • • • • • • • • • • • • • • • • • 11.;
Cl M M M KA CA CA M CA M M M M CA CA M MI M MI CA M M M 15) M Cl M CA MI M 15) 15) 4.,
I I I I I I I I I 1 1 1 I I I , 0 4-1
CO OD CO Cl CA N) V, sO U7 U7 CA CA CA OD CO VI CA CA MI CA +- V, Ul V7 N) N) in ,- OD M Os Cs
N. M M lr CO NI 00 .- CA +- Cr- 15) CO +- V, a uo ro CO a No — No (-4 — Cl-- uo IS) co r4 No r. al
A: CO 04 M M 151 Cl r- CO 15) r- r- M r- r- n M r- r- +- VD CO +- CA M M 15) ,--1
•••••••••••••••. • • • • • • • • • • • • • A • • al
M M CO CO CO CO CO M M +- M CA 15) 15) M CA 15) 15) M M M M M M M 15) Cl MI CA CA CA CO E I I 1 1 1 1 I 1 1 1 1 I
N) r,r, -qo -- M N a c.4 ro uo r, r4 CO co CO uo n a- r, uo r, In CO sO +- N. V, OS 0 N) ,.. ,^ 0
Ca? a cl ..-- r. r, ro in No c.. 4 — r, n No r, CO CA Cs V7 Cr, r... M qti ro a ro a or No — ro r4 4-
J .-- ..-- Cl N) CA CO .- M 0, M +- ,-- CA CO ...- Cl ...... CS? CA +- Cl T..... Si M In IS! CA CO '— m -- N) • . . • • . • . • • . • . • • . • • . • • . . • • • •••••• X
n M M M M M fA CA CA M M M M M M M CO M M CA CA CA M M tA CA M M CA M MM •■••4 I I I I 1 1 ; 1 1 1 1 1 I I
..Cr .q- co 04 Cl ro N) rq to 04 -0 uo Cr OS ... •-r CO Cr a Cr-. or r4 N SO I'l MR OD CO ro a In IT!
No a .,-.. No uo uo uo r4 In +- 1" a- rot„ OD 1' CO In fA a- n cl or n vo or uo ro oN ,0 r, ,- c E 1,-I CO r- 04 M n .- CO Cr, CO CA t5 +- M CO +- +- +- CA CA M Cc, CO M b7 15, CA Cci CA 15) CO +- CA I- ................................0 O M CA CA Cl M M MA m MA M rA in n m n CA M M CA CA M M M M M M CA M M CA fA Cl U
L.L. 1 i 1 1 1 i I ! 1 1 I I I 1 r- f,
.gi- Cr +- N +- 1" CA a r. r, No CO ,-- r. — uo CO CA CO OD sO sO CO CA Os sO ,-- CO D7 -0 Cl-- ..-1
+- 'Si N a ..-- Cr- CO U7 N) 00 CA N) CA N) V7 a CA +- CO V, 1.0 V7 VI CO U7 -/- N' ,- !gp ,.„1- Cr- v, r-
....:c a — ro In sO CO M CO +- ,.... V7 1' CA N) +- VO •,,r +- VO a '0 m CO +- CA +- CA 1' +- 1- ,-- ,- n • • • ••••••••••••• • • • • • • • •• •• • • • • • >
CO CO CO M M M ...... n M CO M CO M M CO M 15, CO M 'Si Cl CO KA fA CO M CO CA CO M CA M CI 1 I I 1 I I 1 I I I ! I 0
CA 1'40 h0 f'... Cs CO ,- VI r. ro CA NO sO In M NO a PO +- +- OD Cs M sO M VO MI Os (-4 r, a uo • .
r. a N Cl +- V7 Cr- uo r\ N) uo -a c.4 c.4 n uo (-4 No r. Cr, N. !..0 'JO CA 147 N) Cr- CA V, ,- CO +- al
CD +- M M +- +- Cs fA ,- M +- MI +- CA +- U7 +- M M +- fi? +- D7 M M M CA +- +- ,- .,- !q- cA r • • • •• • • • • • 4 • • • • • •••••••• le •• • • • • • ip
CA CO CO CO CA CO CO CO Cl M CO CO 15) CO CO CO CO MI CO 'Si CA 'Si Cl 'Si M CO CO CO CO 15) 'Si CO I I I I I I I I I I I I I I I I I I I I ▪ 5-
D7 Cr- a r„ a- or r. N) M fA VO Cs Cr, a r, No CO N) VO CA Os +- CA N. In 1- VO IS) r4 uo r4 N) ai
U7 CA r. No a -- — Uri r, m CO sO 1- V, Cr. CO VO Cr. CO M sO sO -- OD Cr sO CA OS ro uo cr r, Cr
IA. 04 -- !!.- 1- 'Si -,- sO CO CA ,-- ro — r4 ro 15 'Si N) I5 CO N) r, -- uo ,- ,- m (..4 -- -- .4- CO CO
•••••••••••••••. • • • • • . • • • • • • • • • • DO
CO CO M M +- CO CO CA CA M Cal fA M M CA 'Si CA CA CO M IS, MI CO CO 'Si 15) CA CO CO MI M Cl <C. I 1 I 1 1 I 1 1 1 I 1 I I I I
a!
CA cis up N v, v„ +- CA +- CA OD 0, CA V, 0, 0% +- %CI PO M D7 M ,0 Cs ro r4 cr ho uo
Cal r, CA N. SD CA a sD a m CA V, +- CA V, -- CO CA OD VI M 117 D7 0, CA CA M U7 .0
W +- Cl +- CO a uo CO ro ro N) CO CA M M 1- U7 -- CA CA ro r4 n t5) 17 h-
=I E ...... IN N) ...,r Z ...- (-4 ro Ar r- (C 04 La w U. CD 22 I-1 -J NC lt CD CV CV CV ci I. c 04 LI LW U. CD 2C 0-1 -.I DC DC CD CV CD CV CD 0 0 U. UL.
• • A • • • • • • • • • • • • • • • • • • 4 m • • • • • • • • •
M M 'Si CO M M M CO 'Si M M M M Cc M M CO CO CO M M MI CO CO CO M CO CO 'Si f'A M M I I I 1 I 1 I 1 1 I I I 1 1•
I
ct, r, ro uo a N) Cr- CA -0 CA CA CA CA M CO ,0 M CA M CA Cci N) U7 CO M a uo Cl a to N) a
(-4 ro -- cl N. r, r, — c.4 'St a uo No ro 03 Cr- 0% sD sD r, co uo h7 M r, ts) ro oN r4 — to ,--
C.) +- CO CO ,-- +- M VO CA CA ,-- .!-- U7 15) 15) CA U7 M CO Ul ,- -- CA 1- CA r4 n -- uo CO (-4 ro uo
. . • . • . • • • • • • . • . • . • • . • • • • • ••••4 am
Si Cl ,-. Cl Si SI CO CA n CO Si CO CO CA M CO CO CO MI MI M CO 154 15) 15) CO CO CO CO Is, Is, Is, I I I I I I I ! I I I I I I
un a r, 0- Cs N, a CO V, CO CA V, N) ro -- uo r, r, Cl-- r, Cr, 1., IS! CR ill CO M1 15) 151 ,0 in 1.0
SO 0.1 NO r, r4 a MI 1- •1- M a n — a ro -- uo m uo 1- VI ,-- Ul M ,0 M -- V, a to — (-4
gi m CO CO Cl "-- 'Si -- -,-- — — CO m CO 15. CO Ls, m ro CO CO ...-. 'Si 15) r.. M •-.. CO M IS) ',.. M CA • • a • • ••••••••••• ••••••••• EN • •11 NT•
CO .-MMIMIMIMCAMISIMMMMMM CO COCO CO CO CO M tS1 tSri Mt Si CA CO ts, ts, CO CO 1 1 1 1 1 1 I I I I I I I I I 1
+- U7 Cr-. CA bl CA a a N) CO N Cl N. M CO N. a N. SO N 0" Cr- 'T 00 CA N '0 M c.4 r, M %0
CS sD CS CO U7 V, '— -0 M V, U7 U7 CA CO CO 0, 0, N. a r, Cl -- No ro ro CA CA V, CA .!-- V7 L7
CO M M +- +- CA •-• .4' M +- M CO +- CA VP CA CA U7 CO +- M VI CA VI M CA •-• CA ,- CA V1 ,- CA ••••••• •••••••• ••••••• • ••0 • 111.9 • .--MAMMAMISIMMMMSIMMISIMM Si154 M M MASIMMMIMMMMIMMM
I 1 1 1 1 I I 1 1 1 1 1 1
178
sO N) 4" CS NI In Cr. In T... Ch ....... co 4- 4- c..4 No r, 'Si CO CA C4 .1- ro =4- co r, No -ci No ,- r, CA •q" DO rA — VD N. ,- h, ,- CA N. 4- Cr- CA ,- Ch. ..-- 'Si CA 4- r, n No cr. r, ,:- PO CA -4- uo 'Si ro ro
Cl CO CO uo CO CO CA -- CA ro CO 'Si =4- CO CO ro No m CO In PA +- +- Cl +- CA MD CA In MA +.- 4- CN • • a • ■ • • • • • a • • • • • • • • • • a a • • 11••••••
CO CO CO CD CD CO CO CO CO CO CO CO CO CO CO CD CD 'Si CD CD CO CO Cl 'Si CO CD 'Si CD t 'Si CD CO
I I I I I I I I I I I I I
CO blr.1121 CA 4" sCi 4" +- Cr. +- Os +- Mr./ Cs +- C..I CD -.-..!" If) N. MI ND N. DO N. -0 CO •q• un N
PO NO +- 1,1 n Cl, CO ON I'N ro ,o PO CO CO CD DO ON CD CD CO DO ro -o PO h, NO In P, NO IN OD ro
CN ,- CD MO CA CD =r ,- ,- ,- m m ro m CD V1 PO CD CD PO CD ,- ro CA n mi n CA ro n Si O. -4-
•••••••••••••••• • • Olin% •11•1111•••••
M CO CO CO CO CO CO CO CO CO CO CO MA CO CO CO CO CO CO CO MA 'Si CO MD CO CO CO CO CO CO 'Si PA I I I I I 1 I I 1 1 I I I 1 I I I 1 I
P. sO PO ...- In N. WI In PO CO 'Si CO CO CO vo =4- 4- ,- 0, -0 Ch. Co -o =4- lo, CD DO N. cr. r... =r =-
C4 ..-- CD ,-- CA In +- +t SO +- CA NO 'Si CA '0 I', 4- -- CA ,-. NO h, CO ro -0 NI CO ,r N. =4- =4- r, CO G M +- CA +- =r ,- .4- m m m ,- (-.4 n =1.- m ,- ro mi r4 n PO +- 'T PA +- +- m ,- m o- m ,-
••••••• • . . • . • • • • • • • • • • • • . • • • • • . .
CO CO CO M CO MA MA CO CO 'Si CO CO M M MI CO CO CO CO CO CO CO PA MA CO CO 'Si CO CO M CO CO I I I 1 1 1 I 1 1 11111 1
C4 t!D =4- un CA CA CO CO VI CO •-- VI CO CO Cr. 'Si N. ,r CO ,- No CA Ch. 4- (Si QD ...- Cl- Ch. Ch. CO NO
+- CA 4" CA Cl I'l r". 4" 0, -40 CO ,0 ,C1 CO (Si CO IN 0, 4' ,C1 ,0 ..- CO CO Si =4. -0 1> uo lo- =r -0 un
Cl CO CO CO CA +- ,- .- CD ,... r..1 CO CO ro 'Si 'Si 'Si Si 'Si (Si (N ,- ,- CO 'Si -- PO +- MI CO CO PA PA
. • • • • • • • • • • • • • • • • • • • • • • • • • • • a . • •
CO CO CO PA CO CO CO CO CO (Si MD CO CO (Si CO CO CO M CO CO (Si Si M CO 'Si CO CO 'Si Cl CO CO 'Si
I I I I I I 1 I I I I I
M CO CA Cr- CO CN. +- CO Os +- CO co 4- -- ,- co ,- 4- CA CA ..--- ..-- Cl- CO ON 'rr (Si N. 0,- N. NO ND N N 0N ON ON (N — NI Cr- CO NO — ro ro No N. =4- Li) CO — 4- PO 4- ri -o ro IT N) un r, N. 4-
CO =- MA In +- +- +- ,:- (-1 m ,- ,-. No -- m PO In .=-. r.ID NO CA rd CD bl CA ,- CD CD CP. CD +.- ro 10 • • • • • • • • • ••••••• •••••••••••••••• ca
Ts:Isis-Inns, 'Si CO tSi nmin CO CO MtS1 ("Si 'Si CO sins: nmsins, (Si sins, CO .--..-.
I 1 i 1 1 1 1 I I I I 1 1 I 1 I I I I 0 5-
-40 n n ,- MI CO NO IN 1'... n CA IN rs... NO CA NO ,- CA 140 CA In n Mn IN =..... n =- n .-.. 0 IN -0 Cr,
CA ,- No un CA ON CD DO 4. mo ,- N sO OD CO *nl if) =1- N. N. =4- co CO n N. ro in Cr- Cr- ,r 'Si (N
z CD CD ,- MI CA +- CA MA CA MD CA M +- M CA +- 'Si 'Si +- CA ro ,- -- n ,- r.4 cs n -- m rq c..4 r-I • • . • • •• 1.• •••••• • ai• ••• • 1.• • a •••••• Ir.
M CO CO (Si 'Si CO tSi 'SiCD CO CO CD CO M CO CO CO M Si Si Si 'Si CO M 'Si CO CO CO CO CO CO (Si Si 4= I I I I I I 1 I I 1 I I I 11 I 1 I I I 0
4., N CO ."41" (N ,1- PO Ch- PO h, PO NO OD ,-- ON NO N. NO CA ON OD PO NO =r ND I.40 SiL uo ,r 'Si 'Si in r,
CA CO CO In sO ro 1- cr. PO ON DO PO +- PO LO CO +- Ni 'Si NO CO NO 'q" PI NO CO NO MI NO 'Si 1.41 N) I)) • ,- ,- CO (-.4 (Si 'N ,- CA CO PO — +..- CA CO ,- CO CD 'Si CD CA — 'Si ...... CA CD ON 'N CD PO ,- 'Si 'Si r.4
■ • • • • • •• ■ ••••••• • •• ■ ••••• ••••••• ill
M CO CO CO CO CO CO CO CO MA 'Si (Si CO CO CO CO 'Si CO (Si CO 'Si CO CO 'Si CO CO 'Si 'Si 'Si 'Si M CO E 1 1 1 I I 1 1 I I I I
!:-
CA VD CO PI VD 'Si CA PO ,-. NO Ci -0 N. 4,-. CA 0, CA h, PO CO OD h, N) Cr. CO N) ,-- O. (Si 0-.. N. 0) 0
PO NO N ....". M Ul In In =4- ,- n CA ,-. ..-- CA VD PO 4- N PO PO h, PO CD -0 -0 h, NO 4- PO ND ,r re_
.....1 'Si CO CA CA •-• CO CO 'Si r•0 =.- CO CA CA 'Si =-. CA MI M CA CA ,-- qD CD ,-. ON 'Si ,- ,-. ,-- ,- CD CA • . . • • . • • • . . • • . • • • • • • • • •ot ■ a I. ••• ■ 11
--
CO CO CO CO CO M CO 'Si M M M 'Si 'Si 'Si M 'Si SID 'Si 'Si 'Si 'Si CO CO 'Si CO 'Si CD (Si CO 'Si CD CD •,-f I I 1 I I i 1 I 1 1 1 I 1 I 1 1 5-
4.-' PP CO Cs CO Cs r, -0 CO ,C1 -0 N) CI-. N. If) DO Cr. N) N. N (Si ND 4- 4- 'Si -0 Ch. -0 N. CO '1' -4- N) .1- 113
M m n n CO rl m n ro IN No =I- No ro CO 00 =.- n CA Cr. CA .r.-. CA CO n n n CA n ,0 N. IN E E ,-4 M +- CA .h.- ,-. CD — DO 'Si ,-. CD ,- ,- CO CD (Si ,- CD — Si ,-. ,- ,..- CO — CA CD CA 'Si 'Si CD ..-- r- • • • • ••• • • ••••••• •s• • ••••• ■ •••••• a., O CO MD 'Si CO CO 'Si 'Si 'Si 'Si CO MI CO 'Si CO CO 'Si (Si CD Si CO CO CO 'Si CO CO 'Si Si (Si CD (Si 'Si CO lj IL 11 1 I 1 I I 1 1 1 i 1 I 1 I ■-•
'15
=r CO uo NO CA 'Si NO N. CO DO CD CA 0- CD 41' CO Cr- PO ,- DO NO CO NO (Si PO 4* Do 0, 0, ,0 ho N) .r.1
,0 DO 111 PI ,- VD N) Cr- CD NO CO CO 1.0 NO CO CA 4- un C../ ON Cl. Li) CO CA PO '1' 'Si =r Co ro PO 0-. 5- S MI CO =r- 4- un CO co CO CO — 'N 1- ,- CA ,- 4- ro n 4- N) DO 'Si 'Si ,- CO ,-- ,-.. VD 'Si -1- c-.4 rq (F.-. • • • • • • • • • • • . . • • . . • . • . . . • • • . . • • . • .7:-
CO CO CO CO CO CO 'Si CO CO (Si CO CO CO CO CO 'Si CO 'Si (Si 'Si CO 'Si ,- CO Cl 'Si CO n CO (Si (Si n C i i i i I 1 I 1 i i I I I 1 CI
Os r, In MD ..-- Cs sO CA in IN 4- co 4- DO CO DO -0 0., OD N. Cl, un co 4- r, -0 DO ,- CA CO 1-, ,r ..
,-. ,- un co No ro ro rq (Si ,- co =4- -- ,- CO CO CA Si N. DO PO CA DO ,-- N. -0 Co P.".1 tSi CD NO ND ci
co cs4 m m ,- ,- uo n m ,- mn 'Si CA +- =I- m —'S; CO ,- c.4 n n ,- n n ,- m ,- ,- PO +- 3 • • • • • • • • • • • • • ••• 11•••••• • ••••••• • 4J
CO CO CD CO CO CO CO 'Si CO CO CO CO CO 'Si CD CO CO CD CO 'Si CO ...- CO M CO 'Si 'Si 'Si tai (Si CO M I I 1 I I I I I I I I I I I I I 4-.'
..:-.
Os 0" CO un o co co cr. r,, ,- r, un CO CO COO CO CO ...- ro ,- cl- ••O 4- CO ro un ,- -0 0, DO CA !T1
CO ro co CA NO IN In 'Si IN NO CA IN Cr — N. =.-. N. -0 CO r, No ro 0, r„, 1.,0 (Si ,r cr ,- N. 1-4,) 'Si C).
U- NI +- +- In N. -- -0 CO CA (Si In +- CA In +- +- NI CO +- ro ,- rq uo ,- ,- ,- N) (-.4 -- r6 -- -- • • • . . • • • . . • • • • . • . . • . • • . • . • • • • . • • In
PA CO CO CO CO PA CO CO CO CO CO CO (Si CO CO M CO CO CO tSi +... CO CO 'Si 'Si CO 'Si 'Si 'Si CO 'Si (Si tZ I 1 I I 1 I 1 1 I I I I 1 I 1
al
N. N Cl 'Si CA +.- CO Ch. DO CA CD CO (Si =1- N N) ,- -43 CA CD NO N. un ro CO co CA CA +- '0 4" CA +-I
N. ...1- N CO CO 0... r..., 4" Cl, -4- Cr- CA DO ,- PO PO N. CD CO '0 N In Cs CN tO s0 N — "0 '0 If) N .0
UJ CO 'Si ,-- "4- PO CD 4" (Si ..--- ..-- ,- CA PO CO (Si ,- - 'Si (Si co ro ,- ro 'Si CA CA CA CA CA 'Si CD CO ill • . . • • • • • • . • • • • • • •••••••••••••••• I..
M CO M 'Si CO CO '5, CO (Si CO CO PA 'Si MD MD MD CO CO CO CO Si PA CO 'Si 'Si CO 'Si CO CO CO (Si CO I I I I I I 1 I I I I I 1 I I
• •
sO Cs CO PO ro ,- CA =4- r, m M ."4" PO .1" In In Cs 1.- sCi IN ,- co ,- ts4 ro a- ND CA CD CP.. C-../ co
=4- un No r, co N M M ro m CA CO PD PA +- sO OD +- 11- CO CD N Cl Cl N CO N CO '0 '... co =4- O — CO DO CO Cl ,- ro c.4 CA ,-- =-. In (Si CO PO In 'Si (Si CO MD +- PA =4- ,- rq CO ..-- No 'Si CA In In • • • • 11•• ••••• 4 • II • • •• • I.• • • • • •
M M M M M M M M M M M M MI M M M n M .-- M M LTD MI M M M M n M M ITD M I 1 1 I 1 1 I I I 1 I I 1
IN IN CO NO 1'0 n M 4- 'ql- CA co =4- -0 r.4 ro co =4- 0, ,- ND co 0, r0 CA N. CA CA ,r ,r ,- IN 'Si
N. CO No r, cs No ,- In ON ND =4- (N ro DO CA 'Si 'Si ,r ,- ro -o (Si un DO 'r$." N) =I- uo =4- IN CO CA
04 CO MO CO CD CO CO — CO (Si CO CA CA CO CD 'Si CD 'Si CO CD CO CD (Si CO (Si (Si CO CO CO 'Si Si Si CD
. . • • • • • 114•••••• • • •0••••• • •• •••••
CO MA CO CO CO CO CO CO CO CO (Si CO M CO 'Si CO CO (Si CO M M CO (Si M CO 'Si Si CO M CO 'Si i5,
I 1 I I I I I I I 11 1 I" I I I
4" N CD +- CO =4- un co CO CO ,- ,- Cr- c-.4 ri Lin (Si 4 cr. ,- (Si -o cr. N. CA -0 ,- +- N ,r ,- h{.,
O` In Oh ...... In CA CA CO CA DO .4- ,r- cs N. CO co ri CO co PL IN CA ,4" +.... NO +- In .1" Cs +- (Si PA
4X In CA (Si CA PO M .4- CO I'l M M CO Si 'Si ,...' CO +.- PO +- CO Si CO +- (Si In PA PA
a . • • * • •••••••••• • • • ••••• • •••••••
CO CO CO CA CO CO CD CA CO (Si CO CO CO CO CD Co-! ..-- (Si (Si CO CO CO (Si CO CO 'Si CO CO (Si CO CO (Si I I I I I I I I I
<E gc.4
1: -- 04 rt".1 4" :C .- r..1 N) .nr C. <C 40 C...1 im U_ CO MC D-4 ...J AC Nt Cl Cl Cl Cl Cl .."-. <C 44 C.) I.J.J U. CO MC 1-4 -4 AC NC CD Cl Cl Cl Cl 0 o U.. 14.
179
r, PO CA ,t ,4" CA CD CD ,- 10 -7 Do CO ON M CO
r, No PO -0 N) m N) CA CO t r, ,T r4 7 -0 Cr r, 10 ,0 +- -0 CO +- CA 0, CD CA ro • ron c...! -0 o-. m If) +- +4" +4' h. CN CD CD 1.0 CD M CD CA CD +- CD 7 m -- PO OD +- CD m m m CA CD rl CD +- LID CD ro 10
•••••11•11•111111•11•11 111••••••••11111111• • •
M CD CD CD M CD M CD CD CD CD M I) CD C4 CD CD CD CD CD M CD CD M CD I CD CD CD CD CD CD
1 f 1 1 I I 1 1 1 I I I 1 I 1 1 1
PO CD CD CO ,0 CA 10 r. Cl CI AD CI ,0 PO CD PO r. CD "4' ,- ON OD CO r, CO IN '4- CO r, "4- r. 0,
ro N) 10 C. PO CI CD CD Cn. C, -7 co ro m m 0- PO CA -0 0, CO +- 1- N) N -- m m c.4 N -...1- -- CO
C5 CD Si CA 04 CA 7 M CD CA +- +- N 04 CD CO PO CD CD CA ,- rl rte) m m -- CA rte) N IN n Do ro • • • . . . . . . . • • • . . • • • • • • • • • . . • . • • . .
tst ta m m ta cst ta CA 1 Cad m 15) 15) 15) 154 CI ID ID IS) 154 ID C4 ID CI ID 15) 15) ID ID M ISI CD 1 1 1 1 11 1 1 1 I I 1 1 1 1 I 1 1 I
,T +- r, Cl 'AD rl CD 10 10 +- +- CD CD sO PO On 0, PO +- r, r4 r, rA CD PO -.0 7 0- (-4 co — ro
CA CO +- '4' Co 0, +- M CD C. 04 CO rte) cq ro m (N -7 CA n N) '4" CD ro to ND N) (5) N) 15) ,r ,_ If) O CA CD +- CD (N CD CD ..-- CD 15) 15) +- 15) C?- CD +- C.1 M *- M Cl M 0.1 M *- M M *- IS) .4" M M • • . • . • • . • . • • • • • • • • • • . • . • . • • • • • • •
IISI *A IfS4 m m m m 15) 154 m m m m is, m 154 si is, m CI 15) ID 15) MI 15) 154 IS) 15) 15) (Si IS) SI
I I I I I I I I I I I I I
a CA +- 0, IS) C?- 10 CO On +- a m ro sO CO +- CA ,1" CI CA +- CA C?- N) op NO CO CA CD CD CZ M 15) M PO CA (N +- CD CA +- CA 15 CI, CO CA CO
• CO ro ro ro uo No rq rq Do m vo r, ( PO if) 0, 7 ro Do -- C?- LP CD If) CO CO
MID PO N CD 10 CD CC +-
1111•••••••••••••• ••••111•111111111•1111••
m m m m m m m CO m (Si m m m m m m CO m m Is, m m est m s4 m cs, (54 m m s! si. 1 . 1 I I 1 I I i i I I 1 1 1
m C. CD CD 10 10 0, 00 ,0 CO DP NO CD CI ML? 7 uo 7 CO C, CA VD ,0 CD 0, C. 00 r N. N) CO
(N ,0 CD 10 CD PO ,0 '4" If) .!-- -so 0, -4) PO PO 0, 10 r C?- r, CD CA CO t 7 CD 1.0 C, r4 LP ,- 0,
CD ..-- m In M .,...- CO CA CD CA +- CD r, CD +- CA 7 ,--- m ro m -- m ro m CA CP CD "4- CO +- CD PO CD.
. • • • • • . • • • • • • • • . . • . • • • • • • • . . • • • • 1-,
ID ID m CO s! s! m CO cs! n CO CO ISI MI IS) M M4 (Si IS) MI CO M 1St M M M (Si 151 154 15) n (Si 0 1 1 1 1 1 I I I I I I I I I I I I I I I
r.r.'
a ON IN CA 0, +- CD CA CA On CO 10 ,t ,- SD -7 IN m 7- oo or C, UP NO CD CD 0, ,- CD sD to r4
CO J CC- (N — ,-- — ,-- (N o... No No — CO Co CA +- if) CA 7 ,- C?- or C, PO Co 7 m Do Do CN 1171 r-1
7gt M M M 1.40 C..1 Cl CA +- +- CD OD CD CA CD +- CD CD CD M CA CA +- +- CD CD CA CA ,- CA CD CD CD M . • . • • • • . • . • • . • •• 11114••••••••••11.11 ..f...1
M M CO CA CD CD CD CD CO CD CD CD M CD CO CD CO CD CO CD M CD CD CD CD 15) CD (Si CD CD CD CO ri 1 1 i• I I 1 t 1 i I I I I I I I I -,
N a co co m Lo a CO 11.1 a a, CO -- -- rte) Do m r, N ii, -- LI-) if) 15) (N CN CN .tr -- ON If) n Ill
m m m m oo No r, c-.4 -- OD C, +- CO CA 7 -- r4 -- No 7- r, 0., r, 1.0 c, so co Go 00 ,.. ,r co ri
AC (Si CO ..-- ...- 15) ,-..- M CA CD 0, 15) ,- ,- CD ,- ,- +- CD CO +- CO CO CO +- CO 04 (Si CO CA 15) M CD M
. • • • • • . • • .. • • • • . • • • • • • • • • • . • • . • • r m m m m 15! 151 m m m m 15! m m m 15) m m m m CO m 15) m CO m m CI s) m m m CO W
1 i 1 1 1 1 I I 1 1 i 1 t I It-
CO *- r", if) DP If) r. Cn On. 10 CA SO CN if) CD +- CA CO CO N) +- 0-. NO ND CD UP N CA CO ,!" if) 10 r-
V' CO V' C. r, 10 CD ,-. 7 -- IN 10 +- r. ON 00 Cl T.... ON 00 T." Ul MI 00 M.) CD CD +- On co -.41- u) CI
--I M M CO VI +- +- +- CD C, CD +- CA CA CD r4 ro CO M rq rl IN -- m m CD +- IN N — m r4 a 4-
. • • . • . • • • • . • • • . . • ••••••••••• ■ ••••
M CO M M CD CD CD CD CD CD CD Can, CD M CO CD 151 CO CO 15) CD CD CD CD CD CD CO CD CD CD CD CD >: 1 1 I 1 I 1 I I I I 1 1 1 • r-I
5-. .o.: CO 04 NO ,- 7- Do CI ,-- 0, m (....4 OD 00 Ul r. CD CD r. CA ,T CO CA CO CO +- ND 10 CO CO ,- r, CD 4-,
CI CO 00 C. M CD 7 0, ,-- rA +- ..q. ro m o-. -113 -.0 -4o C, OD 10 +- CI CO CD .c1- +- On bp '71 I-, co I7
E 1-4 T- M (Si CO CD CD M OD CD CA +- CD CO +- CO CD +- CO CD CO CD CD CD N) CO ,- CO CD 15) CO CD m r r- • • . • • • • . • • • . • • • . a Via M•Ig• • ••• • ■ •••
0 M M M CO M M CO CO 151 M !Si CD (Si (Si CD (Si CO CD CD CD CD CD CD CO CD (Si Cl CD CD CD CD 'Si 11! 11. I 1 1 I I 1 1 1 1 I 1 1 1 u
r
CO 7 N — n (..4 DP OD C. ..1-- MI C?- DD CD DD m — 00 CO C, NO N r, PO 0- rq 7 IS .c1- +- NO OD M
CO -0 IN CA PO DD CO 7- m r, -- ..0 o- m (5., -- Do m m 7- c..4 m Do Do r, ro r, Do r, O. 0- 0, •r 1
:c c...4 m CA 10 If) CD ON WA ,- SI r4 N -- N) m N ro m (N PO II) +- C, CD M +- CD ro — ro n n (-
CO
▪
. . • • • • . . • • • — • • . — • . • • •••• • ••• • r5
M M M M CO M M M M M CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CO CO CD CD CD CO > I 1 1 I 1 1 1 1 1 1 1 I I ! I CI
LI
'41" CO CD -0 DP On CA DD DP DD +- Ill 0, CA CA CA OD Os CA UP 0, +- 7 IN r..., I."0 I.r.) OD .,:r ro -- r..!
tA CD CA 7- N NO DD CI 10 "CI *- !"0 C..1 *- N) ".10 M In *q" CO C..! Os r"... CA CA if) 10 CA "4" N) CD CO . .
CD ..- CO CO Cl C.4 CO 'Si CD +- ,- CA CO CA CD 7- CO m m ,-- N r4 7- m m m cl N CO r4 m 7 -- 0! • a • It a • •••••••••• ••••••••••••••• • r m m m m m m m m m m m m m m CO m m 151 m m CO m CO m CD CD CO CD CO CD 15) CO CI
II! I I I I I I I I 1 1 ! 1 1 I I 4-5
No 7- Do m 00 if) CO •q" 1.0 CD 0- 10 CD NO NO '/" CD 10 a rl r, 7 ro CO CD CO UP CD CA CD N) DP I-
7 CA NO +- CD N ro m r,. CI — m — 0, ro Do r, No rl 7- n Do Cr) oo r, m r, -o Do 7 LI) ,- 115 1.16. Cl ....- ....- CO'gr N ID M ,- M C.4 ,- C..I (-4 r4 m rq CO CO c4 No rq 7 CO m ca. • . • . • • . • • . • • • • • • . • • • • • • . • . • . . . • •
--------
CO M M M CO CO CO M M M CO 'Si M CO M M M (Si M CO 15) M MI M CD CD CD CD CD CO CD CD NO I I I I 1 1 I 1 I 1 1 1 I 1 1 <3.:
+- CO CA NO CD NO ,- +- DP 00 CA PO C, CD 00 7 7- ro (..4 ,..- CA CD DP 00 OD CI +- r, CO 00 m n al
N) 7- m N) — 7 CA C. N M IN LO rq CO PO N .q.- -- -CI -- -,..!- ot-. -- CA 1.0 +- LID CD CO CD ,- CD I-4 W CO M M or, 7 EN Do m ro — ro m ro m IN m — m m Do .1- IN 7- -- -- ro ro — ro — ...- rq .2:! • • •••••••••••••• ••••••••••••• • •• 117
CO CO CO 15) M CO MI SI CO CO M M CO CD CO CD CO CD CD CO CD 15) CO CD CO CD CD CD CO M CD CD F- 1 I I I 1 I 1 1 1 I I I I I I 1
a m ol, CA 10 CD CA ,0 C. CD r, CD +- r, CD CA +- 7 7 10 CO 1.0 NO 7- m m o- 0- NO CD ,- Cin
-0 CA ,0 M -0 CA r, CO "4" CD 0, CD N1 7 r, 0, n — r, (...4 r, n 1...0 CO 04 ..4- m CIO ,- ,- co !,,
C3 M M M M ,-- M CA M CD +- CD 10 CI +- rq Do ..-- m 7- m ,- m m M m m •-.. .4.. 154 N (N -,.4.. • • • • • • • • • • • • • • • • •••••••••11••••••
M ti) ...- M M M M fA M MI M M M M M M MI MI M M M n M MA M M M CO M M M M I I I 1 1- 1 1 I 1 I I 1 1 1 1 1
cl N) m CO 'qt CO 7- cl ,-- 7 or r, r4 -- m N) Do CD UP 0, if) 7 -- -- CA sO -O 7.- 001 Do On -Cl
CD NO CA ,t 04 CO NO CD CD CD ,4- NO 7- -- Do -0 7 — 7 No m Do ro r,. IN r4 r, 7- CA CD 1.0 10
04 CO CO- M M ..". M CO M CO MI M CO M CO 15) CO MI N) CD CD +- CO CD CO CD CD CO M +- CD CD (5:,
11••••••••••••••• ••••••••••• • • • • •
CO CO CO CO M M M 15) M CO CO CO M M CD M CD CO CD CD 15) CD CD CO CD CO CD CD CO CD CD (Si I I I 1 I I 1 I I I 1 I I I I
OS CA a .,-- ,0 -a CO N) CO CA a m a a m r, m ro 0' N '0 N. N C.4 ...10 1.54.1 C?- ,0 (....1 .0 On +- N CO NO 1"0 '7 M CO CI a ro CO rq c4 OD N1 0, 10 CD 7- 10 If) UD CO CA CO -0 N 7 (N m r, LID
.CE CO CD CO CD 04 +- (N +- CD CD CD +- (Si rA CO CO UP CD +- CO N CO CA CD +- ,- CO ,- CO cl. cD CD . . • • . • • . • • • . • . • • • • • • 4 II • • • •••411*•
CO CO CO CO MI CO M CO CO CO CO CO CO MI CO CO CO CO CO CO CO CO CO CO CO CO CO CO CA CA CO CO I I I I 1 I I I I I I I I
.eCC !MCI
E ...- c4 rya ..:r c -- C.4 NI ,:r r- <C aQ C.J UJ U. Co MC 1.4 ...1 A: = C) CD 07 Cl CD L <r OA LJ LW I.A. CD MG I-4 -A AC DC CD CN Cl CN CV O 0 Li. LL.
180
,-- '-0 Os CO L) N CO MA In CO CA CO NO PO ON Os CA NO IN CO r ro cl ,-- cr. CO a- so a, CJ n ro •Zr In 1.171 r..m,- co CN CO in CO in amuo CON NN N.NNON CCI CO CO In in so romm a
I CO CO si- CA ,- ,- CO CO sr CO CA ro -- CO PO If) CO CO sr ,-- -- .,-- CO-- PO CO CA a •--. — a- co
•••••••••1•••••• •••••••■ • ■■ •••••••
CO CA CO CO MA MA CA MA MA CO CO CO CO CO CO CO CO CO CO MA CO CO CO CO CO CO CO CO CO ISA CO CO I I 1 I 1 I 1 I 1 I I 1 I I I
0, CI, ...... CA rl r- NO r, in In 10 po ,- ,- p.... c..) Ao Cs sr CA CA N ,4.- r, r, P, -0 AD AO sr ro m
po r, in co ,- in CO CN r, ..4- ,4.- Cl. — co -- — a- CO CO In NO OD JO CA CO N CO in CO CO CA -.T. In CD Cl CA CA .!-- ,- a- n n rq <5.4 m ro <34 m uo ro m n ro n — ro — m m CA rq ro -- m a a • • • • • • • • . • • . • . • . • • • . . . • . • . . . . . . .
CO MI CO CA CO CO CO CA Cl CO CO CO CO CO CO CA CO CO CO CO CO CO MA MA CO CO CO CO CO CO CO CO I I 1 1 I 1 I I I I I I I I I I I I I
AO VI PI OD PO NI -- w- a- Cr. so r, m co a. a- — r, NI NO NO CA MA CA sr CA ro m ro In .,-.1- cl
N CO CO .--M-4- InCIN NI OD ,--• in in In .ffi- A-- sr a -- ,- -.0 — No sr If) in m r.0 CO 0.1 -.0 N! CA
OD sr CO CA ,- PO CO PO CO CO CO CO +- CA sr CO •-• NO CO CA ,-.- NI CO N) CO CA -- CO CA CO Os CO -- •II•••••••••••••• •••••••••••• ■■ •••
CO CO CO CO CO CO CO CO CO MA CA CO CO CO MA CO CO MA MA CO CO CO CO CO MA CA CO CO CO MA MA CO 111111 I I 1 I MI I 1
CA OD AO CO CA a- a- CO CO ,-- CA PO in CA IN N. In NO CO -- CO PO CA NO ,-- -- NO PO In PO AD 0...
..-- CA CA r- ro in a- N. In Cs CO In CA MAMA N ,-- ,- a- r, N. a- N) NO in CA CA In (' <31 r..! n ro
cr. n — n N.) r- CA .-- CA -,-. CA CA MA LOMA CA CA ,-CA m po ,- c,) ,- m ,- .-4- — ts, — r..7..,,, — -- ■ ••••• ■■ •Issu ••••• •••••••• ■••••••••
M M M CO CO CO CO M CO CO CO CO CO CO M CO CO CO CO CO CO CA CO CO CA CA CO CO -!-- CO n CO I I I I I I I I I 1 I
•
NO 'A' ON N CO CO CO co N sr +.- CO NO, 0) sr CO In CO -0 NO 0- In CO In so co in N) CO -o -so
-q. .q.- co m AO CA UO Os +- CO MA V, In NO CA -- CO ro r, n m m rq co ro CA N CA MA v.:A MA -0
CD +- CO a- -- — CO N) CO CA CO - a- CO -- CA in +- CO -a- -- rq CO a- CO rq CO CO CO CA CA N) a- G.
.. • . . . • • . . • . • . . • • • • • • • • • • . • • • • • • . 11
Mt MA Cl CA MA MA MA MA CA CA CA M CA n MA MA MA CA CA MA MA CO CA CA mA CA MA MA MA MA M CA lo I I I I I I I I I I I I I I I I I I I I r-
Cr,
Os AD Os +- VI In at uo c....1 a a N. rq a- at PO OD CA -,-- Cr. MA PO ro -o r.„ in uo CO so ro so a-
N r, m uo N UO N ,--. MA OD a- uo a m CO uo a- AD Os In ..-- VI CA CA CA LI-A OD r, In VI In In ,--4
Z CO CO ...- NI +- CA CO CO CA CO CA CO +- CO NI +- CO CO +- CA CA CA CO CO ,-- CA CO M +- CO N CA M ▪ •••••••••• •• ■••• ••••••••••• ■ •••• ..I.'
CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO in I I I t 1 I I I I 1 I t i I I I I .-!-:.
CO AD CO VI CO VI CA AO In Os CO Os AD NO IN CA PO AO +- ,- UD NO In Os CA V, In AD +- rq N. n ri)
AD CA a- — CO In PO a ro AO OD NI Os NO CO CO N .:1- CO a- r,. OD VI In 'tr 'AO In rq CO CO CO uo .--1
1: — CO (Si NO +- CA +- +- ,- CA CA CO +- CO CA CO MA MA ,-- NO CO ,- ,-- — (Si Cr, CA Si a- — V.,1 CO M ■ ••••••••••••••• ••••••••••••••114 P.
CO CO Si CO SiC CO MA MA MA CO CA CA CO CO CO CO MA CO Si CO CO CO CO CO CO CO MA CO CO CO CO CO W
I I I I I I I I I I I I I 4-
•O CA MA OD MA N Os +- MA CA MA MA In PO CO NO N MA ND NO Os AD CA MA MA rq r, uo ,-. a- r, a -
CA CA CA In N, CA V, ro n r, PO ar ,- so ,- 0, T... Cl N. MA PO MA CA OD CO sr CA NO CA In N MA 0
....1 +- MA PO +- CO CO CO CO NO CO CO N +- +- +- CA +- CA CA CA CO CO - MA CO CO +- CA - EN CO ro 4- • • • • • • • • • • • • • . . • • • ••••••••••• ■ ••
M M M M M M MI M M M MI CA MA MA MA MA MA MA MA MA MA MA MA MA MA MA MA M M MA MA CA X I I I 1 I I I I I I 1 I I 1 1 I I • r.4
5- 04 CA .).- *r co co r.... NO CO NO CA NO AD PO PO N MA a-- .,--. OD OD NO OD N MA MA Os AD CO AO CA N *-- -1-:,
CA N, CO CA OD N In OD CO VI OS a- in No rl ,-- 0-. CO +.- NO at In In Sr CO In CA CO In In CO CO M
E 1.4 CO CO CO — CO CO CO In Si — CO CO — CO CO CO CA CO CO ,-... •-• MA CO 00 CO -,-- MA Si CO CO MA CO r • • • • • • . • • ••••••• ••••••••••••••• ■• 0 M CO CO (Si CO (Si CO CO CO CO M CO CO MA CO (Si CO CO CO CO CO (Si CO MA CA CO CO CO CO n m n CI IA. I I I . 1 I I I 1 I I 1 I I I I I LI
r
N. 1..... AO NO NO a- r,. ro so uo uo ro ro IN ro — ro m r, c-,1 r, OD a- r, rq in ro uo 17.-.! n -a rq ri).
co No ro ,- in r, in PO CO N Cs (Si N) PO PO N a- ro uo CO a <34 ro uo rq ro <31 CA NO sr CA CCD •H
7C N (Si NO a- .4.. CO r, CO n m -- ND -- c..1 n cq ro m CA NO a- — cr. CO — — m a- *-- ro -- -- r-
. ••••••••••••••• •••••••••••••••11 IT"
CO (Si CO CO CO CO CO CO CO CO CO CO CO CO (Si CO CO CO MA CO CO CO SiC CO CO CO CO (Si CO CO MA CO I I I 1 I I I 1 I I 1 I I I 0
r, a. uo m I'— N N Os In ON C Os Ncarte C
-a O C
V, In MA Os NO Os CA ,- In Os Os CA sr MA a so — a r, a- a uo n uo m co uo n No ro vo o- ..
Co CO CO Si rq N' ,-- m — m ,-- m r-.1 m ro n CO m (Si rq rq o- — n m ,- CA MA CA CA po -- 0
•••••••••••••••• • • • ••••••••••••• 3
CO CO CO CO CO CO (Si CO CA MA Si CA CO CO CO CO CO MA CA CO CO CA (Si MA CO CO CO CO (Si (Si CO CO 1..' I I I I I I I I I I I I I I I I I
-I--'
*47 1:71 NI CA P, CIA NO PO -.- ,- Cs V, -a cl 0, ,-. r, NO -a — — CA V, ro a in CA AD MA AO MA PO
L7 MA MA N V CO CA CA In +- P, .-- MA Os .q4.- -- AO AO In sr Li ro Cr, a a- ro r, -- CO a- -- CO CA M
Ia. CA -,-- +-- sr AO CA In CA CA MA CA ,-- +.- CA CA MA CA MA MA CA OS rq a — m MA CA 1:-.4 — ro — — La
. • . . • ••••••••••• amtles•••••••••••
CO CO CO CO CO CO CO CO CO CO CO MI CO CA CO Si CO MA MA CO CO CO CO MA MI CA CO CO MA CO MA MA Ao I I I 1 I I I I I I I I I ! I I SE
r, a uo — rq uo r, a- ro In OD OD NO N ,- AO NO UO Os -a -- rl rl co -o -- a AD ,-- NI CA fA W
UO NO CA -- sr MA a co CO a a r, un ro co r..! N. ..-- N rq uo a- m -o <2, a- uo <5.4 r., so so rq --,
LLI MA MI CI In rl rq ro n CA -!-- CA CA PO MA - MA MA MA MA Os CA CA PO -,--- CA NO CA =- NO ,-- MA -!-- ..!D
•••••••••••••••• •••••••••••••••• IT.I
CO CO (Si CO CO CO Si CO CO CO CO CO CO CO CO MA CO MA CO CO CO CO CO CO CO MA CO CO CO CO MA CA V- I I I I I I I I i I I I I I
Or in a N a- N CO r%1 co cl a 'r CO ,- ,r. N) -- c..1 17.4 0, ,r a N. CO -O ,- ,- CO co N) ,r r,
a- a r, NO CA a <3.4 CN CS '0 C.4 CS MI S4 OS U7I L171 I.0 n r, .-4. r, U7 T.- r, n CS N IN ,-- In r,
C.) — MA a- CO CO — (-1 m rq CO CO ro CO — rq a — n oo CO CO CO N CO CA +- -,-- a- CO N ro a
. • • • • • • • • . • • • . • • . • • • .... • • ••• ■ ••
CO CO MA CA CO MA CO CO (Si MA CO CO MA CO CO CO MA (Si MA CA CO MA CO CO CO MA CO MA CO CO MA CA I I I III ! I I I I I I I I I
NO ro a- NI in Os OD r'N CO r.... MA In =- PO MA AD CA AO CA In NO ,---. MA .-- MA AO CA In ro r-, a so
CA — ,-- — so uo m so -- — uo a rd CA NO PO NO NO In -- In a ro n IN a -o ro a- — co c-.4
04 CO VI CA CO CO CO CO CO CO CO CO MA CO CO CO CO CO CO MA MA (Si (Si CO CO CO MA CO MA CO CO CO CO
•••••••••••••••• IlMom••■ ••••• ■ ••••
CO CO CO CO CO CO MA (5) MA (Si CO CO CO Si< CO CA (Si CO CO CO CO CO CO Si MA (5) MA CA CO CO CO fa I I 11111 I I 1 I I I I I I
CO VI +- sr NI OD .,-. NO CA CO N -a -1.. c, r, N. N. c.-.4 — so r, r, ro a N. NO CO MA In ,- No co
In a m a !...,. m uo so rq cl — uo N. a- rq --- m ro in r, so — a a ,..- r, a oo — a- m rq
<E li, M ..- ,- EN M In -.-- M ..- M ..-- M c...1 m •-• a n — n rq m ro n ,-. m m — ,--. ro CO CO
•••••••••••••••• ••••••••••••••••
CO CO MA MA CO CO CO M CO CO MA CO CO CO 'Si CO CO MA CO (Si M CO CO CO CA (Si CO CO CA CO CO CO I I I I I I I I I I
SE 04
E +- NI NO sr c +- CA Po ,r. 5- <C 04 C.) WW WI. Co MC I-, -.) AC AC cD C CA CA CA C. <r 04 C.) fA) I.A. CO MC 0.-4 ...) C MC CD CA CA CA Cu 0 CI W6 WL
181
C51 ,- r, .cs- co N uo r, N co 4- ,o CR No PO -,-.. N. ,- CO SI NO ,..- Is. 'q" 0' MI 0. M4 el uo co -0
4- *-- -0 in CA N 0- n -0 N7 't C. ' -43 N rn N 4- Cn.I N 0, CO 4" N. M MI Mt N 4- MI ,CI N Co
cn -- m so rn m m ro ,- ro -- m NO MA *- 4- os -- m uo rn m rn ro -- 0.1 rn -- so m -- .v- so • • • • • • ••••••• •• • ••• • •••• ••••••• •
M M M M M M n M to n M t CR CR CR CR CR CR CR CR CR CR CR CR CR CR CR CR CR CR CR M 1111111 I 1 1 1 11111 1 I 1
M N. CA D7 N C) LO 1.11 PO NO r. '0 N CO ft- MO CO 00 Cl 'AO 0 CR VI 0 VO ro Ni C) ro ,r 0.1 CA
MI ,-- 'T M ON 0, CN r., IS4 .,.r co ,0 rq m ,- op po 4. .,..- m rq ,- r, ,- ro rn CA 't D1 MN MO 0, -0
Cr fl.- m vl ,- ,- v m ,- Cl m .- N, 04 C) os 4- rn m 4- ,..- 04 vo .-- m -- CA NO ro c- m If 4' 1111•••••• •••••• • • ••• •••••••••• • • •
CR CR CR CR CR CR CR CR CR CR CR CR CR CR CR n M M M M M M Ma MN MI MA MN M M MA M MA
1 1 1 1 1 1 1 I 1 1 I I I I 1 1 f
CO ND D1 ND r, CA ,o .q. -- uo rn r, uo 0.1 co ro -- ..,..r CA VO CA Cs MI rq c, -- ro el ,f1- CA NO ND
Cl Os M M M CA — Os CO CA -- n ,-- 'qt .7:- -- N co rn ,- ,- as CA .t3- n oo so rn 04 vo Cs) -- -..r
cn N m -- ,- ro C) Cl m rn rn — -- m cs rn ,- 04 rn ,- fiff CA M rl CS) m m rn -- rn uo rn rn • • • • • ••••••••••• ••••• ■ •••••••• • •
CR CR CR CR CR CR CR CR CR CR CR CR 154 CR CR CR CR CR CR CR CR CR CR 15) CR CR CR CR CR CR CR CR I 1 1 1 1 1 I I I 1 1 1 1 1 I
co .nr so ro CS) N (5) CA CA VI CI CA Os If) (N ON 1.11 CO 00 Os N7 N. r, OD M NO cr r„-- .q- 4- 0-
- CO r4 NO .-- ro IN el r, so CA 4- cr 04 -,r rn so -0 -- N vo .cr r, ro 0- ,- Ni r, uo 0.1 cl -0 — O CR CR CR 4” CA el CA ,- Cl CA CA .,- CR CR CA CR CR CR CR -cr 04 N ,- CA CA CA ,-. ff--. If) .f.- m rn • . • . . • •••••••••• ••• •0 I ••••••• II • •
M MA n M M M MA M M4 M MN MI ,-- M4 M4 M M M M M M M MA M M4 MI M M CA) n M
I I I I I I I 1 I I I I 1 1
CA 0 4- CA 4- r, NI N. CN OD N. CO 04 r, so so ho 4- ro r4 OD Os NO 121 CA CO OD ,- rl ND r, el
M r, co r, 4- r, co rl m co ,r 4- 4- ,- CA MN CO MN "ft CO MD M4 rl PO CO OD r, NI 'cr OD 0- CA O ,- CA 0 ,- ,- CR VO ,- CA ,- ,- ON ,- ,- tO ,10 ,-- CR 0 ,- ,- CR 4' ,- ,- CR CR -40 CS! ...... ro uo • . . • . • • . • . • . • • • • . • . • • • • • • ••••• • • CI
M M 15) St n M M M M (Si M M M M MI 45) MN M n Mt CR CR m m M., 15) M M rn In rn rn 7-., i i l i Iii I I I I I I I 1 I 1 I I 0
O4 r, -0 0- ro ro r, m uo uo N P. VO MI r., 4- c, N N. TT if) ,-- r. (N .q• .cr N N) so IN so r..! crr
0- -- ,- OD ,--- N. CN. CO In CO -nr .a- -4- m so 0, ,- rn so ,- CC- 0-- -i- LO 00 N. CO CO Cs ND CO --
= CR CR ,- V0 VO ,- CA ,- ,- CR ON ,- CA ,- ,- CR CR CR CR NI CA ,-- CA ,- M CA CA ,- -f-- M MN M4 f--, • • . • • • • • . • • • • . • • • • • • • • • • • • • • . • .• fil
M M MA CA MN M M M M M M M MI M M M MI M4 MN n M M MN M MN n MN MN M M n M4 H.:, 1111 1 1 1 1 1 I i I I I I I 0
O 4 1.11 N rn ro N. rn cs uo uo co N) uo so co CA .-- ON ON CO CO N. In M 1.0 NO D7 CO CA ,r .1-
CA N NO M r. NO CA NO MO CO 00 CO CA ,-- OD -4- r, ro r, CA -4- co -- cll., CO 0.1 uo rn -- .1- 04 rn u) • CR CR ,- ,- CR ,- ,.. ,-- C4 0, CR ,- CA CR MI .,.-- n n M -f--- MA M4 ,-- MA M MI M -- in m4 M ,— ..f...1 • • • . • • • . . • • . • . • • . • • • . • • • 1. • • • • • • 4
M M M M M M M M M M CA M M MN MN MI M MA M MI MI 15) M MI MN 15'4 15) M MN M Mt M 11", 1 1 I 1 1 I 1 1 1 I I I 1 1
::-
4- co vo rs .q. CA 1.0 ,-- 1.11 Cs 1.17 Cr CA ,-- NO CA CO MO CA r, ,- NO CA MO M7 ,.... -.:r rn so 0) CD MA ID
NO CO CO r. ON ,- CR 0 CR PO 111 MI -0 04 .q. ro rn uo CO M -0 ,-- CS) •••fr 1.11 CA ,-- NO r--. MN CO -- cf-
-J CR CR CA 1'0 ,- ,- ,- CR CR CR ,- CA CA M CA rte) M4 M4 CA MO CA ,-- M ,-- ro -- 04 -- -- ,- -- ,r • . • • • • . • • • • . • • • • • • • . • • . • • • • • • •• • X
CR CR CR M M M 45) to ,- 15) (5) M M4 (5) n M M M MA M MA MN M4 n (5) M n MN MN Mt M4 MN •.--1 I I I 1 1 1 1 I 1 1 I 1 1 1 1 :..-
14 ¢ rn e, IN m -- cs N. -4- ,- mf mf r, CA .q- 1.11 r. ND r, ,..- NI 1.11 CO CA CO NO CO r, ,-- CO 1.17 r, Cs fr.',
-4- 00 ,-- VO r, NO .7r 4' Ll PO CO CA l', CO CR -0 04 ro r, vo uo ,1.. uo -- to 04 ro rl pp n ts4 ,0 £
E 1-.1 CR CR ,- CA CR CR ,- 0, CR ,- ,- ,- ,- M4 ,- ,- CSI CR ,- CA ,- CR ,- -a- ,- -- rn 04 m rn ,-- ,- 5- • • • • • • • • • • • • • • • • • • • • •••• • ••••• • • al
0 MI MI M M M MI M MI M MN M MA CR CR CR CR CR CR CR CR M4 t4 M MA M SI C) M M n MA f4 U UL I I 1 1 1 I I I 1 1 I I 1 I 5--
ai
4- r, -- 04 -- uo r, r..... in N. N. 1'0 M NO b1 111 No .i- uo -- 00 m uo 00 0.1 m CS Lo (.4 ,0 ts, 0, .,-,
0, V0 R. -0 -0 MN MO 4- rn 04 cs Cr! IN lo- r, m T.; ;;, P '2' :*. ;i2 ;.1 "I' cq cl 6 ',1:). Cl 4 .4-1 Q7.1
f....
:C CA MA MO Ll L7 M M4 •-• ,-- ,..- CA NO CA CA M ro 175
111••••••••• ••• • • • •••• ft •••••••• • • >
MA M M M M M T- M in M M MA M M M M M n M n (5) M) C5) M fiff M M M M (5) MA M CO 1 I i I I I I I I I I I I I I C:1
-0 ,0 4' M Cs ro MO 0, CA VD MO Ns r., 04 . IN 0, LI VD 0- 0, r, MI SO 1.0 CO 0-. n Ol .1. M rn ..
-4- rn 01 0, 0, CA M ,0 ,- ,0 l',.. r, CA .s- 0, OS 01 TT CO ,0 CD CO ,0 ,- Vl 0, 4- r, r, -0 ..,.1- CR Ill
CD *- M M — ,-- CI, n M .-- — ,-- M04 rn -zr C) CS) rn — -- 04 4- m rn Si 04 elrn 04 is..! 4- IN r- • • • • • a • • ••••••• • • •• •••••••• • •• Me b Mt Mt M M M M MI M CR CR M M M (Si M M MN M4 M M CA M (5) MN M Cif M MA M MA MI CS)
1 1 1 tIll 1 I 1 I I 1 I I I 1 I I 1 44
PP) N. .-- 0 CO ON ,- ,- .T r N) .q. rn r, rq co 0, ND CO No .,-.. m 0, -- m 44- co 0, C'- m -- co ',7;
M 0- 0- ro ro CC- so r, 0- r, ,- ...r. ro CA Os N. U1 CA NO CO CO 'q• CO MO ,-- CO co N. CA ,- 4.- IT. O
U. 04 m — 4- ON ,- 0 MP .-- MA MO ,- CA MO ,-- MN CA M M CA ND CA -i- -- m rn -- -- -- .,t. -- rn • . • • • . . • • • . • . • • • • • . . • • . • ••• •• • v.
CO CR CA CP CR CR CR M CR CR CR m m m m m CA CR CR CR MI M M M M M M MI M n M MN -tr I I I I I I I I I I I I I I I I
CU
Ll .g• OD -0 1.0 M CA St N. r, Cs rl ro ND 111 *q" MO NO NO MN ,- r.... 0.1 so co N. CO CO TT NO CT, !-r) ,-.1
M MO r, ,0 VO 0, ,0 N1 r, MN CO r, -- m CN C.4 CI, 1"...1 NI CN MO CA NO ,-- -0 0, .1- 04 r, r, 4- - W CR CR C5) O. 4- — uo CA NO ,-- VD — '4" ,- ,- MA M M MN 121 U1 CA 4- N 04 01 ro (N el rn rn -- 115 • • • • . • . • • • •••• • • ••• V • ••••••• •• • • 1...
CA CR CR CP CR C4 CR CR CR CR CR CR CR CR CR CR CR CR CP C4 CR CR CR CA CR CI CR CR CR CR CA CR I I I I I I I I I I I 1 I 1 1 1
0 ,- 4- co ,- -I- -- CA NO 1.11 so 4- ,o vo 04 N. 4- co co so ro 04 ro ml uo Ni ro r, MO Mo r, Cs.,
04 rn r, r, cs 04 r, -- co so ,- OD ,0 MI 114 L7 .qr. (N -0— r, ro ro U1 ,-- ,1- rn 0, n ri —L!
GJ ,- CR CR CA ,- CR ro ,- CA •-• *-- U1 SI •-• tO ,0 .-- M D1 .- --- MN 'T ,-- PO M ,-. 141 M4 CA ro uo
••• •••• •••• •••• • ••• • ••••• ••••• • •
M4 MA ,-- M M Mt m m rn m rn rn CR CR CR n MI M MI M MI CR C4 CR CR CR C4 CR CR CR CR CR I I I I I I I I I I I I I I I I
iri CO ,... 4.- N so r, 0, CO .q- r, Ll. 4.. -0 r, ,- (N Cr- so -- ,- .q. ro uo CO so ro ,- so 15 r, ,r
C51 rn rn ro 0, ro NO CO CO r"... T.. N. C..I n .q- so 'T CA NO ro rn co rn co -- uo so 04 rn to r, <I-
x's; es! m m 1:5) n MI Is) M M CS) in MI Mt Mt Mt 15) PO Mt C) ,- n n Mt n Mt Mt n — n n CS) • • • • • • • • • • • • • • • • • • • • • • • • • • • • ORO •
MI - M M M MI M M MI MA MI M MN CM MI M M 154 MN MN St MI ES) M M MN MI MN MN M M4 n 1 1 1 I I 1 1 1 1 1 I I 1 I I
U1 1.11 U7 Ul VI -0 4- n 4- LI CA CA CO CD M CA 0- MO ,-- 0, V0 CR L7 4- cis Ci- C) 00 ro 4- cl, r,
CA M N M N) 4' CC- ‘-r- so 04 Cl- ro co Cl- — ..... CN In M r, 0.1 4- m ,-- 04 ro 04 uo os m ro uo ¢ 0, m -- m cl ,- CA CR (Si CR CR ,- CR CA ,- ,- 4' CR CA CR CA ,- MO M ,-- CA ,-- ,- MA -r -- m
. • • • • . . • • • • • . • • • • • • . • • • • • • . • • • • •
CR CD CM CA CR C4 CR (Si CR CR CR CA CR CR CR C4 C) M CA CR CR CR CR CR CR CR C4 CR CA (5) CR CR
I 1 I 1 I 1 I I I I I I I I I
Cl, IN11 •Zr
..
•
4CKt ILJ 11.1 1..1 J i 1:11 0i •Cr LU CO Of C23 0 0
182
rN .73- cil NI Co NO ON 0, M 'r CA CA ON -o CA -o CA NO 00 0, CA CA In N) In OD In r- n C4 1- MI
c' L7 ,-4- kr) ,- 0, m co -O „-- es, ,- IN ,-- cr -0 co Lo In co Ls, - al N In In in -0 t Cr, PO 'AD M CN CA CA
In r- CA CA r- r- 1" .- to In Co Co cr '0 Co Co un Co ,-- - CA ,- Cl Co IN ,0 m -- .-4. m
•••• • • • ••• • ••••• • • 0 • 4 0 • • • • • II • • • • M M M M Co M M M Co M M M CA CA CA CA CA CA CA CA CA CA CA CA CA CA to M MCA M ,--
I 1 1 f I I I I I I I I I I
ON N. N, ON r- M M N, OD cr sO IN cr ro cq co m - r, co cq - co ro ..t CA cr so n ...- ,c1 cr
M r0 r, . .7 cr cr so Co CO c.1 co oN so "- Cr, CA c.J un - Co D7 r..... CO NI CA C'. In EN CA 1.0 Cl ..40 G .... Cl NO Co ..-- cr - - - Cl m ro Co Co In '4' Co Co cr Co ,- ro ,-. - Co - rq PO +- n C cr • • • • ••• • It • •••••• • • •••••• •• • II• • . •
CA CA CA CA CA CA CA CA CA CA CA CA CA CA CA Co Co Co CA Co Co CA CA Co CA CA CA CA CA CA r- CA I I 1 1 I 1 I I I I I 1 I 1 1 1 I I
1- Co b1 -0 Co 1' -0 In 03 CA CA sO cr CJ cr un ro CA ro so cr cr. cr Co cl - CO so -- C'. - IN
IN Co ro cl r, - sD CA Co Co .74• sO 00 rq Co D7 '0 NI ,- N. - ro so 00 CO "0 ,- CA Co 0, CA In ro
C5 'qt Co (.4 Co cr Co cr Co ,- Co Co - - un Co - ro Co IN - ro Co PO M IN -- n cl ts! n Co r- • . ••••• • ••• ••••• ••••••••••••• • . •
CA CA Co CA Co Co CA Co CA CA Co CA CA Co Co Co m Co Co M Co ta ta Co M Co n Cal to ....- ta M
[Mill I I i 1 I I I I I 1
NI -0 In r- 0- IN cq CC' sO 00 sO CA r- cr ro c..) PO cr to so 0-. 'q- SO "40 SO Co N r, N) ,- IS, Co
r- 0, M M N. CA r. C') N) N. r- cl... cr IN ro Co Co cr un L1 sO CA PO OD U7 CA r- MI 1' PO C'- CA CO
CN CA r- CA CA r- CA r- CA r- NO r- CA VD CA CA CA CA CA CA N1 r- CA CA CA ,- cr cl n n ts! - ts! • • • • • • . • • • . ••••• • ••••••••• ••••• •
CA CA CA CA CA CA CA MA M M M CA CA Co LA Co Co Co CA CA Co CA CA M4 Co Co M M r- IS MI Co
I I I 1 I I I i I I I I
CO r- N, CO 0, M U7 r- Co D7 NO r,„ CA n Co • CJ cr. CA IN P, N) CA -O Co CO P0 uo CA un cr
CD r- CA In CA r- CA 1- CA r- r- r- r- r- 141
PO NO ON CO NI Co cr ro ch. (-1 NO M N SO SO ,- Cr. cr ts, ro cr ro cr so in OD N cr n R.. n
Co so CA CA in CA r- CA CA Ce, CA CA r0 .40 • • •••••••• • •nle• M •01••••• • ••• ••• • • C.-1
Co M to MI M M Co Co CA Co Co M M M 15) M 13) M M M n 134 M 15) M M 15) t5) M M 15, MI LI 1111111 1 I I I I I I I I I I I I CI
r--
Co N) 141 CO CO 0, 0, IN cr -0 CA CO "4" ro ro r, U7 CO CA D7 D7 CO 1- r- CA N) cr N) N. co -4- un cr,
CA sO M cr co cr r... Pte) ,-. b7 CO N, N Co cr r, ro ts! CA N CO =- r- cr CO OD ,0 CO MI CA In ,0
X r- Co r- NA r- CA CA Co CA Co C4 t5.1 ...- M NO ,-- Mit! Co CA CA IN (N Co n - cq ch. 15) CA tst cl cl ri
m • • a . • • • • • • a • • w . . ••••••••••••••• 17,
CA CA CA to CA CA CA CA CA CA Co Co Co Co Co Co Co M Co Co M M to Co to Mt M Co IS, MI 154 ts! 1 1 1 1 1 1 1 t I i 1 1 1 I 1 I 1 1 0
„!..:,
cr. -0 N) r, cr CO 'Si CO ,-- PO cr CO sO ,- NO PO Co M -0 cr oo cr. '- un -- n ro cq Co - IN CC,
ND In cr CN CO 0, 04 CA CA CA N OD PO sO CA n =-. CA OD U7 CO sD ,-- cr - ts! CO U7 r- .- ON LO 414
X CA CA CA CA Mt CA +- ,-- r- PO CA M CA M CA MI - CA C, .1-:' • • • • • • • • • • • • • a . • . • m w • a m • • • • • • • • •
Co Co CA Co Co Co CA Co Co Co Co Co CA Co C) CA CA CA CA CA CA CI M M n r- to M M4 M M4 M iii I 1 I I 1 I I 1 1 1 I 1 1 1
-
0, OD In CO VA ro rd sD U7 M 1' CA M CN PO 0, NO ,0 ,- 0, 00 ,- CO M 0, ,-- CA 0, sD CA cr un n
CA r- r- ,0 r- NI CA NO U7 CO CO CO r- CO M MI Co N N CA U7 1" U7 CO -O r- CO -O CA ,0 CA U7 g-
...I r- Co NI CA Co Co Co r- N1 CA CA r- CA Co r- N r- CA CA 141 CA CA CA r- 0, CA .-- r- r- C.! CS, (-1 • • . • • • • .... • • •11.1 •••••• •••• • ••• • 1 X
CA CA Mt Mt CA Co Mt CA M to to M CA CA CA CA CA M M M M n M Co Co Co Co M Co Co n M •H 1 I 1 I I I I I I 1 I I I I 1 I I I r-
i-3 gA V' In VA ,0 r- ,0 CO CO tO In CA In CO CA un cr un ro co rq M r41 r- In M D7 ,- PO ,0 M PO PO rt-J
r- CO Ul r- PO ,- cr ,-. -4- c-. In N1 0, CA P.') 141 CA -0 r- CO ,- CO N, r'.. CO 1' cr - un co ro un 7. E 1-4 CA CA r- CA ..- CA r- cr - n ,-. ..-- CA MI M r- M M ,...- IN CA M 1.- CO r- r- M rq n n - -- r- • • • . • • • • . • • • •••• • •••••••••••• ••• aJ O Co Co Co Co Co m n Co 15, Co n to to GA CA CA Co Co Co CA CA ra ta Co to CA n M Co M MI Co. U LL. I IIIII I I 1 I I I I I I 1 I I I r
un PO ro cq cr. n In rq c.d r, r, -.0 P, NI ro r,.. ro - un rq is! sO OD r- CO ,- 1- cr sD cr co un •-,
C, m ro ,0 CO ,0 N1 In CA r- cr rq to 1' +- 1-, .,- ro ch. 0- ro -0 cr r, bn - ,- cr co co CO N r-
X PO CA 1- cr cr n CO r- to r- CA .1- r- CA r- PO PO Co PO NO U7 MI M4 r- M4 r- Co Ul 114 PO r- CI 10 • • • • • • . • • • • • • . • . •••••••••••••••• >
Co Co Co M Co M Co Co Co Co to to Co Co Co Co Co Co Co Co M to r- Co Co Co Co Co Co Co Co Co 0 1 I I I I I 1 ! 1 1 I 1 ! I 1...1
Co 1- CA IN, Co IN Co CO ,0 CO r- 0, N. 0, Co r- r- 0, CO PO r- ,0 -O ro - oh. OD Co cr Cr. - CA ..
cr CO NI CA cr CO 'CI cr - CO 0, MI r, c..1 N, cr un cq IN In 4O N SO co cr -0 - NO ro ,0 N OD CI
GO ,- MI MICA CA 1- MI MI +- M! r- MI CAC :MICA CM CA CS, r4 rq cr, m m GA r- CA CA rl n ro ,- -2: • I. • • • a • so 11 • . 1 • • • • • • •1•••••• •••• • 1. ../..''
Co Co Co ts1 Co CA Co CA Co Co CA CA Co CA CA Co Co Co CA Co Co Co CA CA to Co m Co Co Co Co Co i i 1 i I I I I I I 1 1 1 I I I I I 4.,
!....
NI r- 141 r- .- CI, CO In r- CO In co rn c...1 un -o r, r, op so .- ,-. IS, Co CO CO U7 ro 0-. cr IN CA i1
CA Co N tO CO CA CN In mD cr 0, N) cr 0, ,- Co 1' DO 03 N, NI ,0 PO =-. D7 CO CO ,./' cq ro uo - a.
Lt. N ..- ..-• in -o CA Li.1 4.-. (N Co N .-• CA CA CA Co CA Co MI PO Co CA In CA MI Co CA CA r- rte) r- r- • • • • • • • • . • . • • . • . • • • • • • • • •• • ••• • • r,
Co Co Co Co CP CA Co CA Co Co CA CA Co Co Co CA Co Co Co MI ..-- Co CA Co 'Si Co CA Co CA CA CA CA <r 1 I I I I 1 1 I I I I 1 ! I I
I,
CN ...- "0 n sD 0- ,- NO N 0.. ...... NI 0.. in "0 Co rn C'- cr n -.0 PO rl cq cr, cr un op ,0 '0 OD 0, ..-1 N rn ....... 0.• CO SO 120 V1 M CA ,- CO D7 ,-- CA Cr. CO IN n cl IN rn 0... 00 C..1 1.0 r..... ro sO -,-- Mt Co ..c!
LU Co Co .- U7 CA ,- cr cq ro - ro .- cr Co - n Co ts! Co ts! ro IN ro CA ro PO CA CA PO +- MI 'S fll . . . • • • • • . • • • • • • • • • • • . • . • .11.• ki•••• F-
VA CA CA Co Co Co CA to CA CA Co CA CA Co CA CA Co Co Co r- Co Co Co Co Co Co M n Co Co Co MI 1 1 I I I I I I I I I I 1 1 f
r- mD CO 141 CO NO In r- CA 0, N, PO OD cl cq CO ON 0, U7 cr CO CO U7 CO ,- ,0 (-1 cr, un vs! r, co
C so so ro -0 CO CA N CO !N. -o cr cq .- n N sD ,- OD Cr! CO N, 0, ,- N CO CA n D7 N .0.... CO
(e) rq is! un ts! n - ro ,-- cq Co m un m - cr un U • • • • .. • • • • • . • . • . • •••••••••••••••
CA Co CA CA Co CA CA Co CA CA CA CA M4 Co Co Co to MA Co Co Co M M 154 M4 Co Co Co IS, Co ts! ts! 1 1 1 I I 1 1 I 1 1 1 i 1
ro CN CO Prl NO In cr IN ro - cq cr co cr co - n -0 cn c, r, Cr, - ro -o Co CO ro 'T CA r- ro
U7 CA CA NO CA +- ,-- ro un MCA n - M r- CA 1" ,- ,-- CA U7 c..1 ro so r.... IN n cr un - un un
04 Co PO Co Co M Co 15, 154 M 114 MI Co M Co Co M Co CO Co MI Co Co MI Co Co M4 15) n Co M n 154 • • • . • . • • • . • ••••• • op •••• a •••••• ■ ••
Co M Co Co 'Si Co M 154 n Co Co Co Co Co M Co Co Co M Co Co Co Co Cl M n MI Mt Co M n M4 I I I I I I I I I I
Cl- CA cr ro cr• ch. so ,0 CO CA 0, bl D7 ,- CO N. -0 ts! CI" ro r, - ro un C.) Co un ro ro ro ts! cq
c0, cr cr ch un CA +- CA MI P, r- OD ,0 CO cr cr m cr -01 co -4- un - CA M r- NO CN cr ro (-1 un
.‹C 1' CA ..-- CA CA M PO M4 MI Mt MI r- M4 CA M ,- CO M r- 1 CA M NI Co +- r- M r- M NI MI MI . • ••• ■ •11•10 • •M••• •••••• • •••••• • • •
CA CA CA CA CA CA CA CA CA CA Co CA Co CA CA CA m m to Co m to to to CA t& m n mi m m n 1 I I I I I I I I I I
ex
•-•. C4 MI "4" Z •■-• Mu .1-
001 C.) W L.. CD MC I-1 .. AC NC CD CV CZ CZ C. ‹C W U. CD MC AC At CD CD Cb CV CI 0 0
183
,10 CN OiN ^0 Ch In DCI CA CO r, r, No cr. Tr co co Ch c.4 ,o vs., in os Tr os m ,_ ,r 1,... m L.c.,
Tr ro m ,- Co Tr N1 N. -4- DO PO CO Tr cci CD CO r, 4 uo m Tr DO D1 Cr. CD Tr ro N. Tr r,
CD CD CD In CD CD ,- CA ,- CA +- CD In CD CD CA CO CD CD ,- ,- ro m N CD ,- In CD CD PC1 1.10 • • . . • • . . • • • . • • • • • ••••II0••••
M M M 'Si M M M CD 'Si CD CD CD 'Si 'Si 'Si CD 'Si 'Si CI CD CD CSI 'Si 'Si CD 'Si CD CD M C4 'Si CD
1 I I I I I I I I I I I 1 I I 1 1 I I I
NO CO N, In If) CO DO CD 1- PO ro In NO PO CO PO Tr tsl -o cc Tr co n No IN cci In CD CO In ND CA M +- ..- PO CA CO Ns ON ND In to Tr cci N un CCI cq In CA ND CO ,- CA In CO in DI CA
Cri 'Si CA +- +- Tr m - m 'Si CA 'Si 'Si CO CA CD 'Si • 'Si +- ,- CD CD ..-- CA rl CD 'Si Tr PO . . . • • . . • • • • • • •11 l• ••••••••••
M 'Si CD CD 'Si CD CS, CD CD 'Si 'Si CD CD CD CD 'Si 'Si C a1 CD CD CI CA CD CD CD 'Si (Si CD 'Si CD 'Si I I 11 I 1 1 I I 1 I I I I I I I I I
NI DO CA PO -o r, CD In IN M 1' CO DO 1' rl cis N ,- r-, r, N. oo cr Cs CO DO CD NO CM IN CO Tr m ro e,ry n N Si If) PO Cr CD CD CA ,-- CA ,0 0% 1' NO 'Si Cs PO MI tSri a CSI m ro - N1 CD CD +.- CD CD Cs CD CD CD CD CD CA CD CA M CD CD CO M CD Tr m4 M
• . • • • • . . • • • •9 ••••••••••a••••
(Si M n 'Si n t5aM (Si 'Si CD M 15, CD CD t5I CD CD CD CD CD CD CD (Si CD CD CD CD CD 'Si CD 11111 I I 1 I 11 1 I I
PO In PO IN CO If) +- PO Tr ON Tr Cr Tr if) 0, M ro -0 n tsi ro Tr - No No c., 0N PO CO CO Tr
CI No IN Cr Cr Si MI Cr. CD IN N. ND ro LI) CO CA CD sO NO N ND CD CO In ro ro No EN vo -- C 'Si m ro - ...- CA Cs CD CD 'Si 'Si CD ND ,- CA +- CA ,- ,- ,- ,- NO 'Si 'Si m! • • • . • . •••• ■ ••• •111111•••••••••••
CD CD CD M CD CD t5W CD CD M M 'Si MCI CD CD CD 'Si CD CD CD CD CD CD CD CD CD 'Si CD CD I I 1 1 i 1 1 1 I I 1 I I I
IN ,C1 CD In CM Cr DO CD Cs CD 1- M ND DO +- N, DO In Cr sO ON n rl IN OD PO +- DO ND M
CA In D- ,- ..-- Tr CCI In DO Tr - is.... bl DO +4- CD Tr un -.0 un - r, r, lo, cr ro un r, PO ND Tr O. CD M Tr IN - - ro - - - Cr - m IN UPI - n Tr CA +- M PO ..-- •-• M CD DO CD +- PO Tr 7.; . . . • • • • • • • . . • . • • ••••••• ■ •••••• 0
M CD CD CD CD CD CI CD 'Si CD '5, CD 'Si CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD Cr. 1 I I I I I 1 I I I I 1 I 1 I 1 I r...
CI PO N) +- CD +- Tr ts! NO Cs Cr ND PO ND Tr Tr EN Tr N 17.4 DD +- MI In CO CD ND 0, -o -o r".1 D.-. CA CD OD DD Tr cl r) IN Tr n ro OD ,- ..-- CD D.- 11.1 CA 1- CO CD ND IN Ti' ND ON n o..i it
Z m - ro N m N ts! ,- ,- co - - CD M +- CD +.- CM CM ..--- IDA CD CD D.- DD. ,- 0.-- n trA 1-7', .p.) • •• ••••••• II• •••••••••••••••• 0
M n 'Si CA n CD CD CD 'Si CD CD to CD CD M CD CD CD CD CD CD CD CD 'Si CD CD CD CD CD CD CD 4-, I i I I 1 i I 1 I 1 I I I 1
Cl
DO In M N +- ND N co un ts! tT4 4 un 4 N. CI M Cr PD Tr CO OD Tr ro oN Ti- EN c..! r. CD ,-. DI
NO CO 1- ..-- Cr ND CA Tr PO DO N 0, ro co - No un - cr uo m! CO ,- IN C 1: m m..., -- ..-- C4 ..-- CD Cr +- ,- CD CD D- n'Si.- ..-- (C! C4 CD M +- PO CD +- CA M 1:71 CD ;ID ▪ • • • • • • • . • • • •••• • • •m• ••• • • ....4
Si n n CS1 M m m m m m M M n M M M M M M M n n (5i M Si CSI n SI n LI I I I I I I I I I I I I VI
00 4 AON C..1 Cr CA CD PO rl CO Tr un MI co n co vo r un N -0 - - cr 0, co un N - In 1.11 r-
NI CO In NO CA M 4 DO +- CA CM In 0, m! N. ,0 ND Cr +4- CI CO mD +- D- ro ro r, 0, ts! -4- c, o -..1 M CD N PI N +- +- M Cr CD ..-- ,- D.- +- CM CD CA CA Cl D- CD ..-- tO CD CA '5, +- CD +- CA .... • • . • • • • • . • • • • . • • • • • • • . . . • . • . . •
CD M Cu CD CD CA CD CD CA CD CD (5) CD CD CD CD CD C5a CD CD CD CD CD CD CD CD CD CD CD CD n n ...< 1 ) I I I I I I I 1 I I I •.--1
r Dr DO In DO NO N, CA In NO N +- rte) ,-- N -Cl C4 r., o- Tr No os Tr N .- N) MI DO r, 'S M:i CA ..p
R. CD Cr n ND r-, uo MITr co ro tra Tr Tr N Ca N Tr EN co No co No MI Tr un rl c..! No ,- m E '-I n T- (5) ....- n 'Si 1-, n ,-- n ,- ,- n ,- ,- ,- ,- ,- (Si tr,...! n Tr - C4 tS! (-1 m.! CD ..-- CA s- !D • . . • • • • ••••••• ••••••• ••• ••
O CD CD CD CD CD (5) CD CD CD CD CD 'Si CD CD CD CD M CD CD CD 'Si CD CD CD CD CD CD CD (Si CD Ill LL I i 1 I I I I I I I I I LI
4 !JO "'I'. r% VII 0,1 n 1..... C4 DO sO In NO OD eD DO .,- NO In PO 1.- OD Tr No cl -0 co No oN m
Tr M !Ds N No Cr r, 1- Cs NO r CD PO CO N, Cr CD CO CD DO IN Tr n m! In CD NO CD CO 10 +- •,•-■
2 ND CA Ul In 0% m -- m C.4 PO Cr CA ND .,-- +- Tr If) CD ND +- MI ,- C.! ND CA PO D- ,- r- • • •• ••••• •••• •••••••••••••••• III
CSI M M CD CA CD Si CD CD CD a CD M CD CD CD 'Si CD CD M CD (S CD CD CD CD M CD M CD CD 1 1 I I ! 1 I I I ! ! 1 0
(....)
CAN CA CA +- PO Cs ,- CA .-- D- CO PO CO '0 ,... U7I Tr 4 r, uo ro 1.-,. DO ND CO Cs CA PO CA
Tr 4 r, No 1- N, DO CD M +- CO Cr CD CA NI ND CA CA Cs CO PO DO CA PO NO CO 0, CA -.1- ta.! ..
CO +- CD +- +- CO CD +- +- +- CD ,-. +- 1- +- CD CD +- CD +- DO 'Si CD CD +- +-- M ,- CD .1- CA al • 1611••••••• 11.1•• 11••••••••11•••••• IT
M M M n 'Si Si CD CD M CD CD CD 'SiCD CD CD CD 'Si M CD SiC CD CD CD CD CD CI M CD CD 3 I I I I I I i I I I I I 1 II! I D.'
+- CO Tr ON ..." 1..... 1"... Cs ND +- In r, EN Leo Cr -- r, un 1.), un c.4 co Tr ,0 ...-- Tr un No uo N co r....
DO DO +- CA cr -o m rq co r, ..- oN Tr ro 4 CA PO ND D- ND ,0 In DO D- CO CD DO "4- OD CD Cs D5
U. CA D- Tr Cs +- U CD CM 'Si CA +- .- PO D- SiC ND D- 'Si ol so T.... Tr ,- 'Si m Cl ,- ,- ro ,- 'Si Ct • • • . . •••••••••• •••••••••••0 •••
CD CSI M (Si CD CD CD CD CD CD CD CD CD M CD (5) CD CD CD CD 'Si CD CD CD 'Si 'Si (Si 'Si m n n t54 co 1 1 i I 1 i I I I 1 1 1 1 wr
Ul M CO ro Tr CA +- ,0 CA M PO CD NI sO DO DO ND 1" +.-- CO ,- No al
Tr No un ,- No cq n DO Tr tra ,- . CD ,- CO CD 0- PO Tr CA CD CD CD 12 % 2 6:::: ',.74 r..1 4 ;:! ,-.1
1..i.1 CD M 0- Tr ,- un ,- ro - NI CA NO CD .- CD CA CD CD DO Tr to Tr Cl - C.! ro ,- N) ,- ts! CD • . • • • • • • . • • •••Il •••••••••••••••• M
M CA CP M CD M CD CD CD M M CD CD M CD CD M MCP CD M CD CD '5, CD CD CD 'Si CD CD CD c.a I- I I I II! I I I I I 1 I
In If) 0, CO CO N In ,0 0, In M ,0 in ro r, Cr Ml rte) r, Cl- -C, Tr r, CO Do op
,- DO DO DO DO r, cs c, In CO DO DO N ,- - CA CA N) m ,- r.... -0 r, ro - 1
T-7. n .1- R Cl;
C.) +- CD Cr CD +- M CA CD CA CD Tr m n cq un ,- m .•..1- -- CA CD PO M CA CD +- 1- 'Si +- CA ..1- • •
▪
• • • • • • • ••••••••••••••••
M M 'Si Si (Si CD Ira m tsi m CA 'Si V31 C54 V34 CD 'Si CD CD CD M eD CD CD CD M MCI M n M M I I I I I 1 I 1 I 1 I 1 I
CA DO ro m N Cr un Tr No r, rte) ,- co Cr. In co Cr .ct. If) CD DO Cr Tr r, co No IN n No -
r, No m No 4 ro m co NO CA M In PO CD DO CO el DO 4 CA OD Cu CA ND CA DO DD Tr Tr CA
AP rra Cr CD +- +- +- +- CD M CD CD CD CD CD CD CA CD +- CD CD ....- CD CD CD CD CD +- CD CD • • •• le•IIII •411•• ••••••••411••••••
M CD CD CD to CA CD CD 'Si IS (Si CD (Si CD Si CD CD CD CD CD CD CD 'Si CD CD CD '5) CD CD CD 'Si 1 1 1 1 1 I 1 I 1 1 1 1 1 I I
CO In In +- Tr No co N. '0 ... co Tr -o No No - r T T Tr -4 No r, T cr No n n cr
-0 ,- r un 4 4 r, r T T Cr Tr N. N ro ,- OD
Cl M ND 0, Tr (Si DD M CD +- ..q.- DD +-
4 OO CD D. CD CA PO CD CD CD 'Si CD CD PO CD CD DO ,- CD CD N) +- Cl m .- m m m m N) M M • • • • • • . • • • . • • • • •••••••••••••••
CP CD CD CD CS1 '5) CD CD CD M 'Si CD M M M M (Si M (54 M M Si 'Si CA CD M M m m M 'Si I I I I I I I I I I I I I I
P4
,-.. CJ M Tr E Cl N) CE .5. C-1 W Li. CD MC 1-4 MC Mt CM CC cn CM CC .. <r 04 LU LI- CD MC s.-.1 J MC MC C) cm cr cr cr
LA.. LA..
0 0
0
184
ON ,- t CA 00 CA ON IN to ,..- ....o (Si r ,o uo uo No co PO CO CA Tr vo ,.... vo 'Si N. rq ,-. uo -- Tr
Tr -- c-.4 CA Tr Cr- S -- -- Cr- Cr. CA T ,- ,- CA N. CA VO In r.., cs r..1 rl m m -- 0- c-..! m CO
el m m Tr n m CA *- CA CA KA (Si Tr n m PO In (Si (5, T (Si KA ,- CA ,- CA KA ,- Tr m m4 Tr co
111111•••• • • • • • • •• • •••••• • • • II• ••• •• M (Si M M (Si M MA MI M M M M M M M In M (5, M (5, M M M M n (Si (Si (Si 'Si to KA MA
I I I I I I I I I I I 1 I I
M sO M Tr r..1 ro so sO In KA sO P, PO KA CO PO -.-- CO MA In N. M co Tr P, Mt In PO ro --
ro -o -4- co m m Tr uo ,o Tr so n ro Pr! ,- ,0 Tr n -- r4 co Tr CA CO In sO 1.17 Cr, CA CA CA KA KA Tr-
., . • . . . • . . . • • • • • • 11,1••••• • 11013•11• II•
M (Si KA MA KA to KA to MA MA to 'Si KA MA KA MA KA (5, to KA M4 MM M M (5, M M MI M Si In I I I I I I 1 i I I I I I I I I I
(Si m Tr CA U1 IN OD MA Cr. In CO In KA N. CA sO Tr uo 'Si vo Tr N. CA Cr, N. PO In
rq Tr Tr Cr- r, co r..1 cs ro CA KA bl r, r4 Sr In P, Pr! Cl CA
CN .- r- r- ro mnnn n Tr n m-4 ro n ro n PO KA ,- MA KA rn 00 MA KA • • • • • • • • ••• ••• • • ••• ••• •• • le•••••
KA KA MA KA KA MA KA MA MA KA KA KA MA to MA KA 'Si 'Si MA KA M KA MA KA KA KA KA n KA KA MA
I I 1 I I 1 I I I 1 I I
sat CA In *- CA sO N. ,... r.4 N. ,- PO ON In Tr Tr Tr uo n oo ro CA P, ND ON ON Is, PO ,- Lo ,.....
,- sO ,- CA r CA KA CA ON CA PO P, sO N) CO NO PO CA -0 Cts ,- KA .4' sO T PO CA P7 r.. CA CA Cr'.
(3 n M MI NI r- CA KA =- CA ,- KA ND KA KA KA *- MI MA CA ,- ,- .- MI •- V) .- n Os MI KA (5) • • • • • • . . • • . . • • . . • •••• • ••• • • • II•
tn KA KA KA CA M KA KA KA KA MI M M M M MI M M M MI KA MA K KA KA CA KA KA KA KA
I I I I I I I I I I I I I I
ON sO ,- P, b7 Cs CA sO CA CA m! ro CN CO CO .,.." n CO r..1 Tr co -o vo r, r4 co uo r, N. -o IN
ON In h. Cr. In Os '0 LIO Is, ,- Tr uo ro ro uo Tr r.4 KA CN 'q- T.- In If) PO CA ..0 n ro -- r..1 o, CI
CD M MI .4" r- r- MI NO CA KA ,- ,- If) ,- KA CA Ln r- Mt Tr CA CA KA Tr r4 -- n o• n -- ro Tr • • • • •••••• • 14.1•• • •••••••• • •••••• 9
II
M tn 'Si M (Si M M n M M M (5) M M M M M MA (5, KA KA KA KA MA KA KA MI MA n m n
1 mil i 1 i I , iii i 11 1 I Cr.
OD Tr ro co so ro uo Tr co CO CA ON CA PO P, KA T In P, CO P, r, CA P, CO Os Os Is, P, ..--,
CA PI sO 03 CA T PO CO P., tom) Pte) Cl- (N n -- r.4 CA so uo (---A Cr- M r- In '.0 CA P7 0, Os a; M MA ,- P7 ,- CA CA CA MI ,- MA ,- KA CA ,- MI MI •-- CA CA N M M ,- CO MA ,- , KA -
• • • • • . • • • • • • • • . . . • • • • • . . . • . • • C)
'Si KA MA Ca KA M KA KA KA KA 'Si M KA M (Si KA KA KA KA KA MI KA KA KA m! r,...! ml M M MA -r=
I I IIII I I I I I 1 1 I I I I I
r, cs cs r-. Tr ro co uo r, No CN SO CS .r..... sO CO CA n n ID LI
w- '0 ..40 in in P7 CA IN Cr-. In MA ,- Os M. In ,- ,-- %CI sO OD P, -0 CA MA ,- If) N N. sO KA MA NO 0% ,- N) Cr. In -
N M St CA r- r- r..1 m ro -' -- n -- (Si (Si 'Si n CA MA KA *- CA *- ON ,- MI N) (Si n M 0.1 • • . • • • • • • • •• • • •••••••••••••••• .1-1
MI M (Si M M M M M M 'Si (Si MI M (Si (Si (Si (Si M.4 M4 M (Si (5, MI M4 MI n n M n M I-I i I I I I I I I I I 1,11
sO a- so m so In CO Os NO 1'0 CA sO Cr, KA Tr los -- CO oo Tr m r, Ii 1 r, so c...1 (-1 so N. r, uo f-
KA CA ro co .-- ro n -- CA (5, ON 00 Cr- OD 1.17 In (N P, co OU ro m, -- r..; -4- so so n ro
...I r- n CA r- (Si (Si 'S NO r- to ,-- r- MI n CA MI M CA CA M (Si 'Si 0, n (N Kt- II• •• • •• • • • • •12 • • • •••••• II •••••• •
M M (Si M M 'Si (Si M St M M 'Si (Si (Si (Si MI M (Si (Si (Si in n LSA M 'Si M (Si (Si (Si 'Si M
I I I I I I 1 I I I I I I I I I I • H 5-
ix, CA Cr- CO IN '4r. IN a CA Cr, a -- n -- N. co ch. m C'- If) Tr cs CA n in P. P, IN r.. Tr Tr ,- ..P
Tr M IN M In CA M r- In CO IN M NO IN Un CP, I,7 NI *- MA 00 00 KA MA NI -CI T In n n E 1'4 MI T^ 'S CA ,- (5, - - KA (Si r- CA to KA to (Si 'Si KA CA - KA *- CA KA CA KA 'S MA *- IC r- . • . • • • . • • • • • • • • . . • • 11.• • 11•••••• O M M (S I (S MI M 151 M Si (Si n M (Si n SA M M M CA M (5) rz! CSa Kn 'Si In (Si KS! In KS! (Si n_! U. I I I I I I 1 I i I ! I 1 I Li
c
Tr so IN ON 00 NO NI r- 00 sO Os Tr r, Tr vo Tr co co ...- in PO P, MA P, CO P. In CA In CO In IT.;
ON CO sO KA In in a m CR Tr r, r. Tr CA Tr ro -0 co ro cr• m-4 uo r, co No cr. uo Tr IN CO
c.. •r =CAMN aa M. KA KA CA PO-CAr10 NOM CA NOTKA NK A-KA Tr- m -- CA 5-
. . •• • M• • • • • • • • • •••••••• • • •• ■ •• al
(Si M 'Si 'Si M M MI (Si M M M M (51 M M M M MI M m M MI (5, KA KA (5 KA (5) KA KA 'S ›. I I I I I I I I I I I I I CI
L.) Q rn a sO CA V) In Cr. r- 00 CA sO ro r. co n ro n ro CO T •-• N) CA 15) ,0 CO ,0 1..0 n In Q-
r, CA r. m so ND CA CO 00 ..- IN - Cl 0, 111 sO CA CA Tr ro r, ro n co r, r, r.4 uo n Cr- cl CA • •
CO r- M M M— ul n -- m -- m (1 (S;! NI M ,.... M MI CA CA M M r- M M CA n ..-- n M ,-- C) • . • • • . • • • • . . • • • • • • • • • • • • • . • • • • • ,r.
(5) M M M M M MI (5 M n (Si M (Si (5) SA M M M 'Si In r- M M '51 (Si M SA M (Si m in .4.; I i 1 I I 1 1 1 1 1 1 1 i 1 I
,...' Ul "40 T' VI r. ro vo so Tr Cl CN n n Tr ,- CA
vp n Tr n .r.-. so a u--.1 o.. a so r. OD 00 In IN rq
CO sO CA MI bl In In so wo in (1,;., N7 sO P, NO NO KA CA N In =- ,-
U. NI r- CA 'q- sO r- In M CA MI CA r- r- CA ,- .r- ro m m rq m IN Tr -- in m rl (--1 -- P :1-7. Cr.
11 o C).
. • . • • • • • . • • • • • • I. •••••■ ••••• • • n e •
M (Si M M M MI M M (Si M M M (Si M M MI SA M M M r- (Si MI M (Si (Si SA MI (Si n (Si n co I I I I I I I I I I I I I I .1.:
r- Tr Cr, co c.r. Tr -- Tr IN ro cr U') r::! IN ml sO In CA -q.. cl — CO ,- T 00 ,- If) .=1... 'S .,zy. Li) CO al CA sO KA Tr ,- Cr n Tr Tr ,o so so IN CA vo OD IN Is. .. .`0 NO NI PO ,- CA In sD .,-.1-. ci- .-.1-. co !...... .--,
La MI M r- Ll CA M 'q- r- CA ,- CA CA NO KA MA ,- Mt KA KA Cs CA CA PO CA CA CA CA CI CA n MI Mt .in • • • • • . • • . • . • • . . • . . • • • • • • • 117
MA 'S KA KA KA KA (5) KA KA KA KA MA KA KA KA KA M KA (Si KA KA MI MA KA (Si KA (5, 'SI M In M m I-
I I I I I I I I I I I I ! I
IN 0, IN T.... uo Tr C - In CT m! Tr m ro co vo co -- Tr CA In sO n h0 CO U In CA in N, r. Tr CC! MI CA Cr- CA CA 10
C.1 M M "4- M M r- r- • • • . . • • • . • • • • . • .
r- CA r- ,-- 4- St M r- Tr -- •—• • • •—• KA OD MA n MA (N MA CA 'S ,- T to ,-
l'r.: -1-.
KA (Si n (51 m-4 (Si m m n KA MA KA (Si KA (Si KA KA KA KA (Si n (Si (Si KA KA (Si MA MA (5 (Si (5)
I I I I I I I I I I I I
,- CO rte) Tr N. uo n so Tr ro Tr N. N) r.....! ,1- rq Sr' -.0 Cr- CA 'T MA CO Os ,- in MA co Tr CA CO CO
,- 00 CA ON NO CA KA KA P. CA ,- Tr IN r..1 uo CA ,- OD CA P, s0 CA sO PI ,-. n CA M CA r- , - ro
174 r- CA (Si M .... (Si T.. tn to t54 n M M (5) 'S (Si KA OD (Si 'S 15) KA KA to KA M.! 'S KA MA m! n in • • • . • • • • •••••••• •• • •••••• • ••• NI. •
M (Si M (Si M M M M (Si MA n M (Si (Si M M M KA KA KA MA (Si 'Si (5 (5 In M (Si M In rs! KA
I I I I I I I I I I I I I I I I
sO M NO IN 4..... T... 00 NO In KA Tr -- NO T. NO 0% rl Tr n In Is, NO Tr PO Os n p, LS,/ )' ,:r---aO ,- In CA KA CA ro os (N N. co ,- n -- Tr c..1 cr Tr ,- m co ro CA N) Os In CA ,- CA NO CA KA CA
<E li) r- CA ro m ro -- m KA ,- ,- M M M n ON M T'" M4 PO ,^ V') (Si M M (Si ,-. — In n MI • • • • • • • • • • ft • • • • • 14 • • • • a • • • • ■ ■ • a • • M M M (Si (Si M M M M 'Si M M M M KA KA KA MA KA KA KA (Si Mt Mt M M M to M M M (Si
I I I I I I I I I I
E r- IN ro Tr £ ..- (N MI -43-, f.... 4:c IS) L) LO U- CO 2 0-4 J = nt CD CZ CN CN CN C. ..x 04 C.:I LU LI- CD 2 I...4 -I 3C nt CD CN CN CV CV O 0 Li- La_
0
185
'1- CA Cs ,..- NI bl -4- CA ,i1 ".- N. ro r, „- ..-..4 if) -4- CA -0 ,-- CO UD U-.1 ri 0, N) ro .4- r, -O ✓ n v co n r, m r, co ro ro so PO I'l VO In CO ....... r- n In CO 1'... -4- PO r, .1- -4- v n .4- so cs
C — m uo m m — (-1 n ro — ts. LID CA M PO CO .-- CA LID CA t) CA PO CA N CA -- LiD MA ,-- N) VA ▪ . • • • • • ••••••••• •••• • • ••• ••• ••• •
M M M M M M M M M4 M4 M IS t5 t M4 MI 15) M4 M4 M MI 15) M4 M4 n M n M n M M n
1111111 1 1 1 1111111 1 1 1 1
In M ..-- Ch v. m CA U, CA CA Os -!-- -4- r, ro -- cr -- r, so PO -- CO ..- -0 -- r.. .1- Cr. OD LID
NO CD ,-- sO OD sO sO r, rq os -- co uo n — uo ,- 44- cr CA Os r, -- CA ,--- r, los co uo hD CA CD
C m m rq -- -- v m m CA CA ,-- r4 — m cr. ro -- ro -- m — — CA CA ,-- CA UD ND CA CA L-.4 -- ..• . . . . • • . • • • • • • • • . • • • ■■ ••• •011•1111
M M M M M M M M M M M4 M4 M M CA CA CA CA CA CA CA CA CA CA CA CA Si CA In CA CA CA I I I I I I 1 1 I I I I 1 1 I I 1
LID r, r, OD ,- ,--, Cs CA sO v — cs n so cr ,-- -4- uo -,-- IN ..- CJ r, co c.,. LID CA -0 °I- NI
v... t--%. CA CO CN PO LID CA !JD CA .-- CA Cr_, -...- ,40 15, "CI 0, ....- Ill +.-- r, 1.11 .4- N) C1 to v. co n uo n c..!
C PO Si CO CA N1 CA N) CA CA CA ,-- CA CA Cs CA CA CA CA CA 15) 1..-.4 15; N CA CA 15) 15) CA CO J. SI n • • •••••• •••• •••• •••••••• • •••••••
ta CO M M CO CA CA CA CA CA CA CA CA CA 15) CA 15) CA CA CA CO CO CA CA CA CA CA Cl CA CA M4 CA 1 11111 I I I 1 I I 1
VI r, Cs PO rq ro to to ro ro v cq r, n -s ND . cis
.1- — uo v CA CA CN CO M NO CO T... In ,-- LID PO CA N1 0, 1A1 CrCs. ,-- co r, rl ,0 s. .,.„4- rs... uo m CA .1- Cs NO PO CO LID Os LID Cs ..P PO r„. IAD
G m m m vo rq IN -- m cl ..- ,-- — os n -- n m m m ro ...- cq ts. .4- in m q54 • • . . . • . • • . • • • • • • ••••• • •••••• •Il • •
n 15) M M M 15; M ta to M M M 15, M 15) MI 154 MI M Mt 154 M 15) M 15; to 15) 154 IS! Si I I 1 1 1 1 1 1 I I I I I I I
Os CO IJD ..- -- ND N, 0, Cs cq CP- -- N. r, 1.0 NI ,-- CA CA ,-- so (.4 Cr. In DD ..1" CA CO 12D
-1- ro CA VO M) CO ID CO ..-- '0 ...- CO ,- CO CO in V) In CO uo v. -- ON OD ND OD sO v v co -1- .1-
cD — m uo ,-- — n PO CA CA ..- ..- OD ,-- CA CA VI ,-- CA cr -- — m ro n CA CA In LID IS! ..- ro .4-
. • • • . • • • • . • • • • . . . • • . • . • • . . . • • • • • :.:-. n m n n n n n n n 15, 15, m n 15) ts. 15) 15) n 15; n ts) 15; 15) ts. is. ts. 15, 15) 15, CA CA CA •s
1111111 1 I I IIIIIII I I I I 5- 4=
-4- uo .1- so co co r, cs IN ro CO CN .4- "-- cs -- v 1 co — cs co uo No uo N. IN co cq -- LID CO fr.■
..- MI .- ..- Cs PO sO sO r4 CA sO ,- CO CA ..- s0 so IS, so cl so v ro r, „-- csi ro cq so cs so 0)
CO rte:? n S.
.74' CA CA ,-- N) CA ,-- CA CA *- ..- CO '— r- ,- *-- CO CA CA CA cl N ..-- IN n m c.%! ca -- -- • • • • . . • • • . • .. • . • • • • .. • ••••••••• ■ • 0
CA CA CA CA CA CA CA CA CA CA CA CA CA Cl CA CA CA MA CA CA CA CA CA CA CA CA CA CA 15. CA In CA U
1111 I 1 1 I 11111 I I I I I I= 0.1
sO -- CA .qr -- CO LID LO 0- v N) ro to v IN -O .4- co N) ro uo N ON CO v r, co tal ro is) vo Cr, •,-I
CO CO Os PO IAD PI CA CO CO ...- sO CO N. IN v co r- IN N) — co v — v ro so n so so ,- 0, Ml
1= ts. m -- — m ...- — -- ts. cs ,- ,- ,... m n — CA CA CA ,-- CA CA ,- V.:4 r- VO CA ,-- CA CA MA In f75
. • • • . • • . • • • . . • • • • • • • • . . . . • • • • • [Si MI M M M M CO M 15) M CO CO M M M M M 15, CO M 15) M M M n MI MI 15) 15, MI M n
•
G 1 I I 1 1 I I 1 I I 1 1 1 U
M 0 sO Os Li? N) ,-. 0 -0 Ch. rl In ro --10 ri CA sD CA CA LO .q. n OD CA .-- CA ..- .1- s0
OD CA !JD ,- '1. s0 CI
CA CO N NI ,- ,-- ,- CA v ro cl n ts. IN ro ro uo so ro ro ro n cs -- "7, CO OD LI
-J CA CA CA NO CA -- ,- CA 0- CO T.- Cl CA CA CA NO 15; CA CA rA CA ...- CA CO NO CA CA ,- ..- IS ,- r.4 0 • • • • • • • . . • . . • . • • • • • • • . • • . • . • • f
CA CA CA CA CA CA MA CA CA 15) CA 15, CA CA MA CA CA CA mi CA CA CA CO CA CA CA CA 15) CA CA CA MA Cr, I I I 1 I I I I I I 1 1 1 I
r-
5 CA IN LID ,-- sO CA h.-) ro sCi VI ON T'.. P.75:4 rl LO .1- uo ro co to tio n OD C. .4- CO ..- CA CA r, CA •1-1
...- 0, 0, Cr. v. r, CO !JD n -4- so CO OD !JD Cr. CO in ts. — Cr. so to co v r, los n LID LID T1" Cr, IN .1.:
Z 1-4 CO M S4 M M CO M 1",.. n .— CA n CA CA 15) In MA CA ,-- CA CA CA n MI 1:74 ,- :.'.! ,- CO CO t!.;:! ,-- 4=
a et a •
f- • • • • • • •• •• • ••• •• •• •• • • • Nov • ,....I
0 M CA CA CA CO 15) CA CA CA CA CA to CA CA CA CO CA CO CO CA CO IS) CA CA n MA CA CA CA CA n n 3;
LL. I I I I 1 I I I Illi I I I I 1,
ro N PO vr uo 0- — uo ,- in N ,-- IJD Cr, CA UD CA 1.0 !AD rl -0 ,- sD ND 0- -4- v cq so uo CO 9- a,
cq co ro cq so -- c-q co -- — so uo los n r, r.% IJD .4- in n N -4- co s) CA NI NI CO UD ...- b-.! •..1- r.---1 = NI CA NI LID Ul CA CA Ch • • • . . • • •••••• •1•0
T T.... Cl 1..0 -- ro n c.4 t.~ ts! c.! v If! m r..,- n -- — ro -4- -- ,- 0 • . . . ,. • • • " • 0 t61 ts, m m ts. m — n ts. m m n n n 15, n ts. 154 n IS, ts, CO n n CO CO m n m) m4 CA n CL
I 1 1 ! 1 I 1 I I 1 I I I I
r, N. N. PO n Cs in NA CO 61 N) .- CO NI Cs. ,- N) CO CA CA N) -0 CO PO PO -0 NI CO sD 111
v• -- CO IN M *- 1.,.. '41' In CO CA In -0 M CA M In '4' M ,- +- n .4- cs4 r, ts. r, to os c
CO ..- CA '— ,- Os CA CA .-- .- CA CA CO 'q- ,- MA CA ,- -..- CA VI CA CA CA CA EN n cq n v — CI
. . • • . • • • • • • . •• • • •• • • • • It •• • • • •
In Cal CA CO CA 15) to CA CA MI M1 154 MA CA S4 M SI M M S4 CO M n M 154 M SI n CO MA In C14 -!;, I I I I I 1 I I 1 I 1 I I I 1 1 I I f-
17
LID ^O sO .4- .4- ro uo ,0 ..- CO CO CA ..- --4- -- m ,o -0 sD CO .-- -0 -- "0 N, ,0 sD Os ^0 n ...7.!
In IN rn h0 IN so ..4.-
,- .q- Cp, ...- In M COO, ND IN ro so N. co N. "71 NO r, r, IAD CA CO 0- N. P.1 0- CA 0,
1.L. CA M CA ,- CA PO ,-- CA CA CO CA CA sD CA .1- MA CA •,-- ,- ,-0-
In CA NA MA SI CA CA CA CA CA MA MI CA CA
7.1 ,- 15,
. • • • IN • • ■ •• ■ •• ••• N• •• • • •• • • .1"
COM COM M M MI M 15, n M/ COCA CA COCA COMi CA CA CO I I I I 1 1 I I I I I 1 I ! 1:14
L....!
0, sD NO -4- r., .11- ..- CN '1" ,0 Os NO CO CA ,-- 1f.1 ,- r,. PO UI CO ,.... sO CO ,-- CA N, ..-- .1" sD 0, ✓- so CO MCC, c4 cr N) — — ro -4. 0- co n uo 't.:1- MA 0, N N) CA PO N OD if) LID CO ...... VD CA it
LU CA tS, CO 0, v. LiD CA NO ,-- ro ,-. N) n — n — 15, n vt- v rq v. .,-- -- CA ro — c.4 — n -- r-
. • • • • . . • • . • . • . • • . . • • • • • •4 NI •••
M M M M 154 M M M M M M M n 154 M M M to CO CO CA CA IS) M mA CA In CA MI CA CA I 1 I I I 1 I 1 I 1 1 1 1
CA CSA ,0 ,0 "0 r, NO Ul sD M 1- ..- 0, N. ,- C CA PO r., ro Cr, so m. cs co rq co .4- — to CO MI
CA rn CA sO N. Ml Os IN PO ...- rl un co ND CO MI rl M n CO T.. n M OD ro ,- N) — — 0-
C1 '-CO.CA In ,- Cr1 PO M CA .-- ,-- IAD CA CA CA IAD ,- CO .4' .,... cq — uo n CA CA .1:
. • . . • • . • • • • • • • 7 % b: 74•1111•4•0. • • •
M CA .-- CO M to M M MI CA CO CA CO CO CO CA CA CA CO CA CA CO CO CO CA CA In MA ISA CA CA M I I 1 1 I I I 1 I I I f I
0, fn CA -0 ,- CA CA Ill -,-- LID ,- r, N. CA CO CA MA ..- CA ,- -0 CO C..1 Os "0 -,..- N. CO N1 N rq O^ ro "0 0,1 M CO CS CO N. n to cu. ts. .-- uo -s cq ro os so v ,- Cr! -4- cq --
CP M CA 0) ,- 15) MA CA ISA CA In MA M CA CA MI CA ro CA 15) -- 5) 154 Pu4 l5 m Isi m. ts) m. ks. is. ▪ •••••••••••••• ■ ••• • • • 0 •• •••••
CO CS4 CA *A CA 154 CA MI CA to MA MI CA M MA 0) CA M MA M4 CO CA Q CO ISA In ISA 1S4 M SI M n I 1 1 1 1 I I 1 1 I I 1 I I I
CA 0, N CN 1.11 r, NI CA CA -0 1- ,-- NO LAD PO .4- 41- Cr. v m so so ^O CO uo 0- ,- .4- so cq cr 0)
0". Cl c-1 v ro ..v. cq -- CO ,- -- .tr v CA CO CA 0, CA ,0 ..- IAD NI tr.M PO Cs IAD ,0 0 In '0 ..-- -0 • CO M ...- CA N ..- PO CO CO CO Mi .- MI NI CO ,-- .4- m ,- ta N .,-- is. 15) -- CO ,- M MI 151 ,-
•••••••••••••••• • . • . • • • • •• • • •• •
CO CO CO CO CO M *A MA MI CA MI CO CO CO M M M M In CO CA CO In CO CO 15) CA CA CA CO CO I I I 1 I I I I I 1 1 I I 1
ex 4:4 .r. ....- 01 r.0 .4" Z ...- 01 N) .4- 04 (...) LO Li_ C.N 5 P-4 ...I A: C CD CN CA Com:, CA 1- CE OP L) LU L. CD CC 6-4 --I A= At c:A cti cli cti Cr o 0 LA- U-
186
CO IN Co 0% ,... -CI .71" rl No cs co 10 Os N) 1r -0 Cp. CA r, Tr -40 ON 1.- Co ri M 0, Tr -0 1,0 TV CI-.
Tr No — cs CA Os CP .4- CA Co CO Co Tr If) CA III Cr- C..I Cl Co IN Co IN Tr -O r, Tr N) Co cp.. Co ,- Co
0 Co Co Tr ,-. Co ,- -- ,- ro Co n Tr Co Co N) 0 Co CD In CD +- +- CA Co rl Co Cl In Co +- .4- Cr.
a • ••••••••••••• ONN•11•••••••••••
M n Co Co Co Co Co Co Co Co Co M Co Co Co Co Co Co Co Cu Co Co CD CD Co Co Co CD CD Co m Co
1 1 ! 1 I I I I I I I I I I I
0, PO CO '.0 CD CO Co r, Tr n In CO Tr Tr co r, 03^:r N) IN IN +-,- N) -.0 +-+-• CO ro co co Tr
ro -- rq cs un CA PO WO CS CO Tr No Tr r, m rl No — V- -at n un No co No No PO PO PO co Tr No ,-
CN ,- *A CA CD +- Tr ,-- n ,... Cl m ro m tSa un ro m tsa ro n — to — n n ,- rl to In MI oN Tr
••••••• ■ •••••••• • a * • • ■■ •011•0•••••
Co CD Co Cl CD CD CO CD Co Co Co Co Co CD Co-! Co CD Co Co Co Co Co Co Co CD CD Co CO CD CD Co Co
I I I I I I I I I I I I I I I I I I
rq co 1-0 Tr ,0 1,0 In cl •- tSi ,-- EN r, No I:, -4-. Tr OD r, In '4- r, In Tr rl — Tr o, un r, co to
rq -0 Tr -- ..- Cr- N ....... .1.. In CA CS Co N) If) If) Tr IN — -- un Tr No co Co Co M CD IX) PO If) "1- Co
Cl N) Co CA +- PO Co Sr' Co Cl Co Co — Co Tr Co — N) n rl m ro Co N) Co CA .”-- Co ..-.. CS! Cr, m +-
••• • . • . . • • . • . . . . . • • . . • ••••••••••
M Si Co Co Co Si CD Co Co Co Co Co Co Co Co Co Co Co Co CD Co Co Co Co Co Co M Co CD CD Co Co
I IIIII I I I I I I I I
%CI r, +- +- CP NO NO CA NO PO CA Tr Co Co Tr to co Tr un un Tr IT co Tr In CO .r.- Os 1171 0 ro ,o
...-. ro 00 +- CO PO CD 0 In r, r, ...ci Tr Tr CS! In CD CO Tr No — CA NO CD Tr rl r, r, CA CD PO co Os
CN n Co Co (N ,- Ci 4-- Co ,- 04 ,- Co "It Co ,- M Co Si Co ND ,- ,- ,- Si ,- I-0 ,- Co SI M ISI n • •••••••••••1111•• •••••• ■ •••••••••
Co Co Co Co Co Co Co Co Co Si Co Co Co Co Co Co Co Co CD Co CD Co Co Co Co Co Co Co +- CD Co CD
I I I I I I I I I I I I I
Tr — Tr ro No to IN -,..- .1- IS) CD In Tr In r, PO Co r, -- Co Tr r.... No in ro IN ro co cis c, co Tr
No — ro uo r, r, co uo In CD CA ,r Os CO OD Tr r, ts! to in CA +-.. CO ,r CD to co r, r.a co ro in
CD ,- n In -- ..-- CD to — — — — in Co M 04 in ,- M trl ,- N M Tr — rq ts! usa co n — PO 0
1, •••••••••••••••• • ■••••••••• ■ ••••• X
Co Co Co Co Co Co Co Co Co Co CD Co Co Co Co CD Co Co Co Co Co Co Co Co Co Co Co CD Co CD Co Co .,1
I I I I I I I I I I I I I I I I I I I !...- 4-,
T- IN CO CD NO PO Tr co — co IN uo n IN ,-- o- — rq -- to — -- CI- co uo Tr r, ro — Tr — IT p5
No Tr — uo cs c-..! ro n IN No ro No uo Tr cs Tr -- — co -ci r, 0-4 Tr — Tr Tr — co r, n ro ro r
Z Co Co .--. PO ,- N =-. Co N Co C..1 rS, ..- Co CA +.... CD CD +- Cl CA CA Co Co +- (N Cr.. CD .— CD CA CA • • • • • • • • • . • • • • . . • • • • •••••••••••• ID
M M Co M Co Co Co Co Co Co Co IS) IS) Co Co n Co Co Co CD Co Co Co Co Co CD Co Co Co Co Co Co U
I I 1 I I 1 I I I I I I I I I I I iTi
0- -O N 4-- VO Co .qr ,-- N IN N 0'.. -40 CO '0 Cl T.. CO ',I" C.4 C..' No uo cs co — .1- CI CO ,- ,- la ..-I
In -0 VI 00 CO ,4- VI O. OD +.... CA CO Cs PO N ',1" in n '''.r M 00 CA PO r, In r, "1- PO r„ ts! ro Tr r-
u: -- Co Co rq Co rl -- -- IS) N) N Co +- Co .— Cr.! Co Co Co In ta ,- ,-- ,-- n Cr. rl 15 to -- ,-- es, it • • . . . • • • •••••••• 1■ ••••••••••• ■•••• >
Co M Co M Co n Co S4 M Co Co CD CD Co Co Co Co CD Co Co CD Co Co Co CD Is, Co Co CD Co Co CD Cl
I I I I I I I I I 1 I I I 1 I.)
In 0, 03 CO -.0 IN Cl- Tr .-- Tr un IN ro r, — un m cs CI- Tr c...1 No Cr. -.0 op OD n PO 10 CA NO ro cl
CI- T- CO N. Co Co C..! IN .,.. M T.. I', ID CI N. N CO Cl '40 T- r'0 !'0 CO .- IS, if) 1' Co C.4 n -C N. -
J Cl n C4 ,- Co ta M Co In ,- *A N .-- Co ,- Cl Co Co N C.4 Co IS) Co ,- CI, n ,- CA =- Cl SI (N n
. • • • . . • • • • • . • . • • .. • • • . • • .. • • • .. • -
St Co Co Co la Co Co Co Co Co Si SI SI M Co Co Co Co SI Co Co Co Co Co Si Co Si IS4 Co Co ta Co CI -
I I I I I I I I i / I 1 I 1 I I I I I 1- .... iii M N 0, ,I0 ,-- ,ID PO CD CD CO -.0 NO r, CA OD UO Tr Cr. co n No Tr to -4- No cs CO n Tr Tr to co •,-1
M CI- Co ro uo ,-. Co Tr I:, 03 r, co co N) N PO CA CA .4. If) 10 Cr- Co Cr- ,- r, +- Tr Tr ts! -40 -0 3-
r 1.-1 M Co T- T- CoT- T- ro IS) n Co Co — tst Co m Co Co Co +- =-. Co -- r, +- +- Co +- Co Co n Co 1-, r- • • • . • • . . . • • • . • • . • . . . • . • • • • . . . • • • • r- 1
0 mmts!msinnnts4ntsitsinntsa Co Co Co IS! Co Co Co Co Co Co Co In Co Co Co Co Co .-. 1.1. I I I I I I I I I I I I I I 77
-it Co Co T- -0 N) -0 co co os if) ,...- co — .- co n r, Co — Cl- .- — N) cs n 0- -0 CO In .,-- a) .-
Co Tr m N r, .,-- co co Co Co N) oN to r -- Tr IN Co Tr Tr Co ro Cr! Co co N) Tr OD Co CO CO Tr 1...-1
:c ro n Tr Tr Tr n r... m is! — rl PO ...- CA +- PO PO CD PO PO In CD CD +- CD +- M Tr — PO .,-. rq o • . • • • • . . • • .1 • • • • • • • • • • • • • • • 1. • • • • • CI
Co Co Co Si IS) M Co n Co M Co Co Co *A Co Co CD CD Co IS) Co Co .r.... Co Co CD CD CD M CD CD 15 0_ I I I I I I I I I I I I I I I I
011
'0 "10 -0 CO T.. C..1 T." SI Co N CO Co r-- N N) ,- Co Cl- "Il- ,-- ,-- if) ,-- Tr -0 -0 -- r... Cr- r... =- Cr, Cl
VI NI ,-- VI .-- ..-- 'Ir VI ND ro Tr — ro un r, r, r, -- r, n r, co ro Cr, ro IN c..1 -- No -o No r,
0 ,- SA ta 04 N If) Co Co ,- M ,-- Si N Ca ND Co Co Co Co (N CA 0-. Co Co Co +- CA Co +- CD t":1 4=4
. . . . •• • IN••••••• 4NO•NOONN•1000•••• ,-
Co Co MA M Co Co SII Co Co Co Co Co Co Co Co Co M Co *A *A M Co Co Co Co Co Co Co Co Co Co CD -0
I I I I I I I I I I I I I I I I I I I-
-0,-- 0-- Si) CO N -O In •Izt- If) 0,N Cr-'— C-4 .4.1 Co -0 CO N. 0, ,0 IN CI ,-- -4- .CI- .1- N -.0
un 0,1 co r, r, n r-, No ro uo -o Tr IT r, CT- co co un r, n rq r, ts! un ro op r, CA rq Tr uo CD
Li- EN ,.... ..-- Tr ...ci EN un m C4 Co N ,- ,- N ,- Co C.4 Co Co N) St N In ,- M ta Cl Cl ,-- NI ,-- ,- ..• 1411•21•41/14•111111• • • • ••••••• •NMUMM cr
Co eD CD Co Co Co Co Co Co Co IS, Co ta M Co St SI Co Co *A ,- Co *A Co Ca n Ca Co Co ta 15, SI I I I I I I I I I I I I I I I a!
,-.4
*A Ci VI NI -0 00 CA r-0 In VI ,-- ,-In LO ,ID N CO VO C.4 0, ,- ,--. ,-- La Tr IN to OD in in IN Tr z!
,-- Tr s! cr. -0 Tr Co c, to PO CA In .4" +- Co n Cr, CA CD OD Co Co Tr uo ,- m ...0 1.0 .,.... If) Co C.., 111
LIJ MI M T- Tr c...1 .,..- 1- n IN ,-- rl ,- ro n ....- Ls.! ts! m Ca al ro rq ro -- r..1 to rq ....- ro m n CD I-- • • • • ••• ■ ••441411•• •••••••••••01•4111
M Co Co Co Co Co *A PS) Si Co Co Co Co M M Co Ca Co Co Co Co m tsi ts! Co ts.i m n n m n Co
I I I I I I I I I I I I I I
IC -- r, r-, -0 N) n CCI CA N) CO PO Cs ,r r, "4- NO +- n CA co Tr n 03 Cr. Tr — -- uo r, ro r,
,0 uo m m uo ID If) '- -O 0- .0 co n If) Cr m IC N) =- CD N r, Tr -4- No Tr co N) No -- No n
C..) ..-- Co LO Co Co ,- CA ,- Cl Co MI IC Co ts! rl un -- Co Cr- Co CD CD N) Co IN us, -- uo m cl ro If)
. ••• • •••..-...• • • toN4 • ION•NMNOUNN•W
Co Co IS, M Co Co Co Co Co Co Co Co Co Co Co Co M Co Si Co Co M M Co Co n Co n In 'Si n Co I I - 1 I I I I I I I 1 i 1 1
Cr- rs! to — -o — In to c...1 CO Tr I.r.I (,) ,- ,-- iiTi CO MI ,... r.O '<ION. r, oN Cr, co N r-, '1" CO -1- N
CA ....... Cl Tr r, ts! Tr Co un m Co N) Co -- 'Cr — N -IC N) CA In +- I'D IN IN t54 .,.. Co Tr ,- Tr m
Ix! m N) ts! Co ts4 Co. Co IS Co n n Co CD CD CD CD Co CO CD CD CD CD I5, IS) 15, 03 CD CD I5 to rrr I5
•••••••••••••••• N•11111•••••••••••
Co IS) Co Co Co M 'Si Co Co Co n m m Co Co Co Co Co n Co Co Co CA Co 15. CD Co M CD CD CD CD
I I I I I I I I I I I I I I I I I
IC n rq If) N) Cr m If) -0 Tr Tr IN m c, 0- cq uo co to co n Ca m IC 15) -- ,- m co Tr CO 0,
0- m N) un co rq In In N) CO -CI If) In 0- =-.. +- In rl Tr c, co r, rq CJ co If) — r, co r..! -- IN
[E IC Co .- -- rd n N) Co Co 15, M .— IS) Cl n T- CO Co T- Si C.4 Co N) Co Co n n T- In ro ts! ta
. . • • . 0•••••••••• IIMON1102•••••••••
CA Co Co Co Si 154 M M ta M Co Si LS4 n Co Co M sa Co Co Co *A Is) Co n n n CA CD 03 CD CD I I I I I I I I I
PC4
0,1 ,r z rA po .= 04 GJ W U. CO Z 1.-1 -J 1= MV CO CO CO CO CO t•- <r ix. GJ W LA. CO CC H -J 3= Mt CO CO CO CO CO
0 CI 146.
187
00 P., 11-.... Cr- NO in CO S.1 N. NO NO PO -O Cr- '0 cr - ro -- mo n co ro -- No ro Cr, ro co If)
cr L7 ro m uo uo - co 0, - CA 0, OD r, No r, ro r, n ro c-.1 OD DO DO rn m cr cr n n 1.-..! a, '4
cn t51 m so m m - cl -- ro - n in n n PO Cr. M M in .n.... M M NI ,- CA t ,- ^O CD ,-- r NO • • • N. • •1111••••■ •• •••••••••••••••0
I M M M CD CD M M M Si M CD M ED M M CD Cl CD n M t n CD CD t CD ED Si 'Si M
111111 1 1 I 1 11111 1 1 1 1
M ,- CS Ln Cl CS NO n 1'0 OS CO PI r, DO Os r, cr r, CN In ,- CA CO IN ,- ,- r.. OD CO IN
PO M ,-- OD Ul r, r, No co CA - Cr. M cr m NO r, CA PO ,- 0-. NO -,-- GI Cr. 00 0, CA UO cn LS IN -- ,- cr n - IN - n ro ..--. Si OS ro CD Cl ro t! - rj ,- ,- m ,_ IN in
. • . • • • . . . •1111111311 • •••• 11•••••• 00 M MI n M M M MI MI ID M CD M M M GI CD GI CD M M CD GA GI CD CD Cl tai CD CD G
I I 1 1 1 1 1 1 I 1 I 1 1 1 1 1 1
CA -0 GI r,. PO NO PO M CO CD PO r, ED -0 r, ,0 ON CA PO r, co n cr a PO r, ND cr -- OD l'O CO
CA NO CA Os Cs ro uo - DO M CA Cl OD ,- O ED "0 ED ,-- DO ,- r, No cr csol 0.o• po cr co rc., uo n
cn ro n m n ro n ND CD CD CD +- M GI Cr- GI CD ND ID I M CA CD CA CD ml ED M CD CD Ezt" n Q.! • s • • • • • ••• 0116114 • • • • 11 0 4 II n • • a • • •
M M M Of M4 M4 M MI M4 M 'Si M n 15, M M M4 M En M En M4 &5, M MI M M CD M CD CD CD
1 11111 1 I 1 i 1 1 I
cr cr r, vo 0, co 00 DO DO GI
ED r, NO PO NO NO r, PO No cr .0 ro
in uo Crcr n ro ro
rq ro CC, -- -- N. ,0 CA 1.0 DO PO 'Si CD DO cr DO DO in CA
cn n 5, m vo c4 CJ CA CA CD CA CD SA SI CA cr CA CA ,-- CA CA CA +- y CD CD CD • • • •
• •
• • • • • •
▪
• • • • • • • • M CA CA CA CA CA CA CA CA CA CA CA CA CA CA CA CA CA CA CA CA CA CD SA CA CA CA CA CD CD CD
1 I 1 I 1 I I I I 1 1 I
CD N. CO U7 CO ,-. CA ,- NO ND r, +- R. CO NO n Os -- CD CD CA CD CS n r, (-4 cr - ,- co cr
a ro ro co CA Os -.0 OS CS 1.17 b7 ND DO OD n 00 CA Cs! 'r.1. -,-- n r„ 0, OD -CI tr....1 ID ',CI Cr
CD CD CD Ul ,-.. ,-. CA 147 ,- ,- ,-. ,-. 0, ,--. n ro uo - n uo CA ,- CD cr - -- m n ^O m -- ro cr
. • . . • . • . . • . • . • • . . . • . • • • 11 • • • • •
MI M CA CA ml Cl CA CA CA CA '5! CA CIA CD '5 CD MI CD M ES) CD CD CD GI GI CD CD n M CD CD
1111111 1 1 1 1 1 1 1 1 1 1 I 1 1
sO cr r, m n co No No so in - PO ,- PO Cs ,0 CA CO CO OD CD M cr -- DO sO cr a No r, cr DO C
'71- CD CA IlD EN CA r. cr NO Cs 0, DO ,-- M Cs 0, 'Er CD CO NO r'', NO cr cr cr c.1 PO NO P. 'El CA ....
GI M ,-- in CA ,-- c4 - - m! co .-- IN .-. m n n n n IN EN ..-- CA - M c.1 CA ,- ,- n CD CD Cr,
. • • • . • • . . • • • • • • • • • • • • • • II • • • • • •
M M CA CA M CD CD MI CS! M n M 15, Cl CD CD '5 CA CD CD CD ED CID '5 CD n CD n CD '5 CD CD ,-1
1111 I 1 1 I 1 I 1 i 1 I 1 I 117, .4.=
In c, cr vo m! n r, n - CO DO NO CD M NO r, DO r, n c..1 No - r, Cr, in co ro - cr n OD 0
1.11 40 CA CA 0, 1.17 CA CA CA Os Cr, DO CO CA +- CA NO n CO 'nor NO cr -- PO Cr, CA N.. a CO (-4 uo 4...'
CA CD ,...- ...- CA ,-.. ,- CA CA 0, CD ,-.. ,- CD ,-. ,-- ,.... c;:! itA ,- tlIA CA .,-.. ,- CD vo m m cl n CD CD • • ■ • ••• • • ••••••• • •• •• ••••••• • • • !■
CA CA CA CD CA ED CA CA CA CA CD CA CD CA CA CA M CA CD CD ED Cl '5 CA CD CA CD CD CA CA CD CD 0 1 1 I 1 1 1 I I I 1 1 I I 4-
CA N. ro ..-- uo o4 DO CO NO ,- DO 00 n CD cr CA in N CA In in '.0 ,- CO In CO IN T.- NO X .1--
M CO co r, CA -- No CA ro a- ro n c.1 - DO -0 DO cr -4- n m cr co IN CO -0 1...... 'CI
-J CA CD CA 147 ri ,- ,- M 0, SA ,- ,-. CA CA IN 147 CD CD CA CA CA ,-- CD ,- ro n IN - .-- t5.4 - ro 5- • • NIR• 11•1111•11•11, • • • N • • • • • • • • • • • • +1
M M MI M CD CA CD CA CA CA CA CD 15, ES) CA CA CA CA CA CD CA CD CA 15) CA CA CD 'Si CD CA Ca CA I 1 1 I I I I I 1 I I I 1 E
tE ,0 ,-. CA ,- rq cr so cr DO n NO CA DO n NO CO PO Cr CO ro cr co uo co uo - cr 0, ,- n ,- 0
CO Cr, r. P.. CA ,-. N. m.! ,- CA *V Cr, CA in OD Cr rn CA Cr, ED ro so co -- c-.4 - - co ho n co cr "•,
E I--I ,- cD ,-. ,- Cl .,- CD ,-. CA CA ,-- ,- ,- M ,- -- CA CD +.- CA 15) M CD sO +- CA CD CA '5 n c...! r- . • • • • • • . • • . • • • . • . • ••••• ••• • ••• M O M4 M M '5 M '5 En TT MI M 15) CD CD M CD 15, CD D CD CD G M CD C? n 15) n•
CD CD '5 l 15) ..--1 LL 11111 1 1 I I I I I 1 1 I 1 r-
!•!.;
I", 40 V CO c, cr so r, r, No - co ro Cl, uo co ro uo r, co uo r, - ho m CO cr cr ..-- - Cr, -2..
NO CA ''',1- sO ,-- NO N. cq cl r, No -- so co C -q. Li
:IC 147 140 117 1.17 CA Cr CD +- +- CA NO CA PO M CA PO GI CA cr 111 CD r. ,- ED PO .,-- '4-,-. CD
. • • . • . . • • • • • • • • •••••••••• • • .
1$) 151 M Cl CA CA ,- M CA 15) CD CA CA SA CD CD CD CA M M M CD ES) CD n n CD CD CD 15) n n .. 1 1 ! 1 1 1 I 1 1 1 1 1 1 0
!::
N. CO r, N N.:1 Cr. cr CA OD CO CO ND CD ,0 Cr, NO CA -4- cr co c4 co n PO PD Cr. r. CO CN C)
CJ taa (-4 r, r, m ,-- - ,-u7 0.4 Cr- -- ho 1", -,- CD GI sD in n n - -- m cr cl co v,,, -o m
CD ,..- CA ,-. ,-. Cr, ES) ..,- ,- ,-- CD CA CD V - CA CD ,- ,- CA Id7 CD CA CD CA CA ES) CA ES) cr cq
. • • • • • . • . • . . • • .
15, M En M 154 'Si CA CD C CD C 15) CD CD CD CD M t
•
ES) M GI M CD '5 5 C5) '5
▪
C•4 11 !.... n n N
1 1 1 1 I 1 I 1 I 1 I I 1 I I 1 I I I CI.
0, ,0 40 CO N CO CA - 1.(7 C^A CA I.17 CI*. PO DO CS n In sO hl cr uo co r, cr rq n n n IN n
DO sO CA VO r, -0 CD ..-- Cr- CA -- PO r, DO ,- IN CO CD r, -- n OD 0 NO '',4- n NO CD
LL CA 'q- Cr, ,-- DO CD CA CA ..-- CA ND ,- M PO M CD CA NO CA cr -- m CD ,- ,- ..-- cr -- <I:
. • • . . • • . . • . . . • . • • . . . • • • . . • . •
M M M CD CA M CA M tal CD M GI CD 15) CD '5 CD CD M 15) '5 CA CD CD GI '5 n n 15) n n Ill
I I I I I I I I I I I I I I rs-1
Cr' in CA M NO N cr - - ro n co DO r. Os ,- sO CD ,- -o--- CA CA 00 Cs cD DO ISD ro NI CA n CA 115
NO r, sO CA r, cr r, r, C4 DO CO 0, Os DO !SD cr No ro r, co so ro CD NO 00 DO ,- Cr- Cr. P:O OD I-
CJ CD CA n Cr- cr - ho - ro - in - in 15) - 15, ,- CD n DO cr cq cr c..! CA CA in CA CA CD n CD • • • . • . • • • • • • • • . . • • • • • • • 4 • • • • '5 M 15) M M NI) M M M MI n Cl CD CD n CD M tD 15) M M ED M CD CD M M CD n 15) ED CD
1 I I 1 1 1 1 1 I 1 I I 1 I 1
V CO DO CA NO 00 NO V4 in cr r, N. n CO NO CA Ca sO -0 NO DO NO 'Si .1- N. in co
CO CA cr n. No c.4 CA r, co n CA in uo a. co n co c..! CA .- CO -- Cr -o co IN cl vo n roz!
(....) CA CA CA M ,- CA 141 ,-. CA ,- DO M M CA NO D C CN • • • • . • • • • • • 4 • • • ••• •• • ••••••••
M M T... M M M En M M M MI M4 M MI M tA GI ED M CA M CA ID Cl Cl CD CD CD ED CD
1 I 1 I 1 1 I I 1 1 1 I 1 !
CA ,-. 00 DO ,0 ,- in ,-- r. Cr. n cr No n CO cq n m No -- IN T.. cq - uo r, PO -0 rn No
ro Cr- CA -0 CA CD OJ Cs CO ,0 NO ,-- CA ,- NO n +.- DO cr 'N ro cr Cr, CA NCO'E - ,- Co 'E CA ,--
PCe CA 0, ES) CD ,- CA M GI CD GA 15) CD MM M NO n M +- D C '5 n D
GI ID GI CD CD ID • • • • • • • • • • • • • • ••• •••••• ••• • •••
CA '5 CD CD CA CD CA 15) 15, C CA CD CA CD CA m m m n ES) m m m n 15) n m m n n n
I 1 1 1 I 1 I I I I I I I I I I
cr sa cr 0- 0- r, r, sD CA DO +-- 'I-4- CA 00 PO no - cr uo ,-- cr in =.1.. co 1.17 N In' n ,-. CO ,.-
CO in CO - DO CI in CO M un Cr uo in PO In Cs. CA c...1 vo CO 1.0 - CA ts1 cr uo ESA in co IN „-
cc Cr- m to m c..4 ,-- r7 .-- m ts! t5a m ESA In M ES1 In 1.S1' ,-- M CA •••• ro - - • • • • • • • • • • • • • II 11 0 4 • • • II • • • 1.S.1 • CSI
P.".•I 71 151
,Si CA CA CA CA CA CA CA CA CD CA CA CA CD CD CA CA ED CA CA CA CA CA CD CA CA CA CA CD CD
1 1 1 1 I I 1 1 I 1 1 I I
.4r "III
E - CA 147 .4- C ,- c".1 m cr cr co G.) UJ U. CO = I-, --I NC NM CO CU CO CO CO r_ eZ 1.74 LJ 1JJ u- C7 MC I-41 -J 3C Dt CO CO CO CO CO
0 0
0
U..
188
r- No CO rA uo c.1.T. No co If) sr to CO In LIO sr ra co ra ra 4 sr -0 r, CA Cr, Cd sr CV CA CA U1 CA rl r' ra ra sr ra ra 4 -o
Al cA r, N. Co r, ro N. cl ra -O Cr. — N) T — in 4 -o to N') cr,
Ci L1 Co C.1 If) Co Co • • • • 4 • • • • 11••• • • • •US1111.1••••*11111111M
CA CA CA GA CA CA Co I) CA CA CA MA CA CA CA I CA Co I) n Co Co Co n Co Co Co Co Co ra ra Co 1 1 1 I I I 1 I 1 1 1 I I
OD PO V7 CA CA V,. -- n -.0 N 4 co ...o N7I CO CO M 'CI cl sr OD OD ro ra r, sr ra "I- CA -.ID 0, -0
VI V, CA -,- PO NO '4' 1.171 OD V, In 4 V, In CA '4" oN cA to co P9 In sr co to -4- to to r, op P7 r„ sr
CD GA CA CO CA ,-- '4' ..- ,--. ,- CA CA V7 CA CA In NI MA CA N, CA .,- N7 ,-.. -- CA ...- CA 19 In CA Cr. ".1* . • • ... • • ..... • • . • . . 2 0R11••••• •••• 11
M Co M Co Co Co Co Co Co Co Co Co Co Co Co Co Co CA Co Co Co Co Co Co CA CA CA CA CA CA GI CA I I I I 1 I 1 1 I I I 1 1 1 III 1
.,--: IN, r-, CA to [CI ..-- -,--: •,..: GI I`, ,--- 4 CO CO PO Cr, CA .zr .."4- In CO .- 10 CO ,- CA ---r o, -4- ,o —
CA co sr ra 0, M -4D ,- CA ,0 CA V, -CI 1.19 PO sr IN ro -- CA OD sr r.. OD 19 ,--- Co ra -..o ro -(1 to co
a ro ra cA ra sr ra sr ra m CA n r- M .zi' M ,- I'''') M Cq M N7 CA PO CA CA -- Co ,--- CA 0, GA 15:
. • • • . • • • ..... • 0 • 11. •••••••••••• • 11
Co Co Co Co Co CA GA Co Co Co Co Co Co CA CA GA CA CA CA CA CA CA GA CA CA CA CA CA t CA CA CA 1 111111 I I I 1 1111 1
GI -0 V, PA CA N. sr Cr, r, sr No — N1 -,-- V, 0, ,-- Co In ,- 0, LIO '1' N1 N. ,--... CO r, CO 0- CA Co
,-- N) CO ,--. 0, r GI 43 in Co CO ,1) In In CA in Co CA '4- -.10 PO CA IN ,- sr ro OD r, PO CA PA CO 0-
C5 CA CA CA CA .-- CA .-- CA ,- CA .--- CA "4- CA .c- CA ,-C, CA N7 ,- ,- ,--. GA ,- po .,... m ts, ts, t, m • • • • • . . • 11411•• 11M • 111111 11111•••••11111111011
SIGISIMMSIGISIGIGI Co Co m Co n Co Co m m 14 Co Co Co Co Co n m n ..-- Co Co Co I I 1 I I I I I 1 I i 1 1
L,) NA sr to cA a, sr sr co CS sr ro sr po
cr, CO IN Co '4' CO N n CD CA In CA CA to IN ra CA PO ,0
co r, PO CO CO V, CA Cc Oc V, "P sr CA N CA -0 •q- m Cr, !f) L1 NI r, 4 N7 -0
Cl C4 6.1 cr 64 Co Co N) in ••• •••• •••••••• • • • • le • •• •••••• ft • •
CoCA CA Co Co Co cA CA CA Co Co Co Cr Co CA CA Co Co Co Co CA Co Co Co Co Co Cr GA CA Co Co Co 11111 I I I I I i I I I I
r, r, ra uo r, N) CO ,-- PO NA sr N N. to -- -0 PA sr Co Co sr N N) r, — if) -o a- N) N Co r... Q
in sr CA In 0, CA V7 -- CA V, Ni ,0 1.0 41- o, sr ra -- co r, r... cA bo 7- uo sr -- r, N M ri F*0
= n 64 ..-- PO .,-. rA -- Co rq Co rq Co -,--- ra cA -- is, Co ....- cA cl cA Co Co ,--. CA C CA ,-- t5 cl cA F •••••• •••• ••• • • • • 111111•••••••••• • •
M M M M M M 64 n n n ISA n CA CA CA CA CA GA is, M n is, is, CA CA CA CA CA CA CA CA GA ,..„, I I 1111 I I I 1 I I I I I ! I
If) n PO if) V, N) Co In CO OD -.0 N N) N. cl — ,- ci- OD N CO N. 0- In ,--- IN n Co -- — sr to n
sr -o cA co co sr sr — OD CA CA CO CA N7 CO sr ro ts! .!'• m co CA PO Cr- ID N '4' P, CO CIA N71 sr 4.-'
*--. is, Co CA GI CA ,- rl Co N) N Co (-A Co .- is, Co 15) Co to Cr — — -- ra cr, cA ra n ,..... 4' n • • ••••••••••••• • • 11X 11•••••• •••• • •
CA CA GA Co n Co CA CA CA CA CA 15) CA CA Co CA CA CA CA CA Co Co CA CA Co Co CA Co CA Co CA CA I I 1 1 I ) I I 1 1 I I I I cp..
OD *-- In CA sr m in CO -- to uo ra — to CO Po ,--- V... ,- po No sr lo, ci- ho — — — r, co r„, r,
is, CA Co 140 Co Co Co rA sr cr, sr cr, Co 0- 0, PO V.„ "i- in CO sr If) No ^0 PA t) In in r9 --- "1- N) •ci
_.1 ,- m c..1 N m Co n ,- po i5 Co -t-- CA Co Co CA Co CA CA CA Co CA CA .,--- 0-- CA ,-- -)--.. -- CA IS:, cA r- • •s• • • • ilm• ft ••••• a • •••• • •••• • ••• ■ ..!..,
S) CA GA GA CA CA Co CA CA CA CA CA CA CA CA CA Co n Co n CA CA CA CA CA CA CA cA CA CA CA CA fI5 - I 1 1 1 1 1 1 1 1 I I I I I I 1 E
Pc.) sr cl No Cr- r, co — uo -o a, — LT. to 0, to ..-- sr cA sr sr vo No co m Cr. uo r, r, to to ra co ID
15) Cr, No ts! — — Co ,- ra Ni 4 i'-.. to C.,1 n N) In In ..--- 1.).9 N) cA Co ,-- -.0 LT. ,- if) cr N) PA ,0 LI E N4 •N" M T- Cl T- Co ,- -.0 ,- ," ,^ 4.-. C-I CIA ,-- .,-- -,- Co ,-- CA ,-- -- ,-- Cr .-- .c.- CA Cl n IS! ,- ,- r- f- ••••• ■••••• • •14 • • • • • • • •••••• ••••• • iti
O Co M M M SI n n M n n M M n n n Co Co Co i-4 tTrA tS4 t! ,-- v.:14 CA CA CA CA CA CA M •,-.1 LI_ 11111 ! ) I ! I 1 I I III r_
A'
N) ,- -0 CO 0) CAN CO IS) N "z1- n-o -4- — co CO .1-. — .,q- 0) LOCO CO 0-. 0-, MI CO -.4- r- N) N
,- sr cl- to r, — a, co ra ,- sr . uo sr ,- uo to — sr No — to n n -.ci N-3 n o, -- co co -,1.- o
= hO n ff, sr sr ra r, ISA n ,- C.,/ .7- ,- C.4 ,- NO NO CA ro m n n 6! ,- 64 ..--. 64 sr — ro — CA L) • • . • • • • . • . • • • • • • • • •••••• ••••111• ■ •■
M In M Co Co Co Co Co CA Co Co CA CA CA Co Co Co CA Co Co Co Co ,- is, Co Co n Co M Co Co n . •
I I I i I I I I I 1 I I i I 1 CI 4, L 0.4 in to uo r, Co in ,0 sr r, cl op PO CO co -c1 .....,
In ro — -o ra Co sr -o sr sr -o — uo -o No if) OD CA -0 NI ,0 CA PA CA in CA CA CA r, r, sr uo
a -- Co n CA CA n M n .,-- Co ,- Co cA ra to Co Co Co ra cA CA is, Co ,- CA ,-- CA CA ,-- Co CO . . • . • • • • •• • •• •• • • • • ••••••• •••• IR • r.-•
Co CA CA CA CA CA CA cA Co CA CA CA CA Co CA CA Co CA CA to Co :c- CA CA CA Co t5I CA CA CA Co Co 111 1 1 I I I I I I I I I I I I 1 I 0
,- ,- Co rA sr -4- co to CA AD CA CA .c... CO 0, In -0 Co *-- Cv CA V, CO In ,0 OD "T" r9 0- PO CO CA mA
CO CA CO CO OD CA CO r.-.1 sr -o r, sr v- N. Cr- CO ....- -.0 N Si sr -ID ..-- ro -4- OD r... C1 C.4 '4-- in .1.'
1.1). CA ,-- .,-. "A- ,0 CA In GA CA CA CA ,-- CA CA .-- Co N) CA Co N) CA CA In ,--. Co Co CA CA ,- co ,-. ,- wr ••••...• ••••••• • • ON• 1111•••••••14 • • •
Co M Co Co CA CA Co CA Co to Co CA CA CA CA Co CA Co Co Co ,-. Co Co Co is, Co Co M Co Co Co Co All
1 I I I I I I I I I I I I I I ri J.1
In -0 sr ,- In 1" r".. OD V, CA OD n N7 IN N. ....... 1...". 00 '0 r.0 0". n sr se r . - ----Co ,-- .,-- sr sr — ill
uo 4 — r, In NI rA to r, sr -0 CA ,-- ,- V, CA co sr PO 0, CA NI ,10 V7 co n p, sr 1,0 co ro sr 1..._
1..urats! ..-- IrIN •■•-• •Ir CAN r- N N 4 Is, Si ,- Co CA C.,A 0, rocN r.r.) CAN PONNPO tA Co tA
. • . • • • • . • • • •••• • • • •••••••••••• • I/ CA CA CA CA CA CA Ca to Co m m n n m m m m m m m m n n M Co PSI M n M M M n I I I I I I I I I I I I 1 I
er Co cA -- -43 CA In Cv If) CA CO ,-- IN N) r,.. N) -0 4 to -0 — ro — sr __ co Co u-s, u--., ,r c,.., N. N if) CA N) .1. ,0 If) Cr, In CO Cr) ,-- CA In ,--. CO CA cA N) NA V, ,0 4 *-. PO sr CO -0 -0 r.s, co ro
L.) ,- Co In CA CA ,- CA ,- CA CA Ca In CA Co PO In ..-- CA Cr. CA CA GA to -- cl is, — uo Co cA N) IC' • • • • • • • • • • , . • Y • • • • • • • •11•••••• • /1 •
M CA CA CA CA CA Co Co CA GA Co CA Co CA CA CA Co CA CA CA CA GA CA CA CA CA CA GA CA CA Co Co 1 I I 1 I 1 I 1 1 III I
• CA -.0 CA In CA ro ro CA r, oo Cr. m cA sr ,- op r, sr co Cr In sr cl r, o- sr r, ra rd -o CA
CA ,-. CA ,0 V, CA In CA ,0 CA CA -- CA .c- NA CA CA "1" CA sr No cA — uo sr m -- Cq sr ,- ro
04 CA CO GA Co CA CA Co is, CA CA CA Co CA CA CA Co GA CC) ra 15) Co Co ra Co ra is, ra ra Co Co ts! ra ••• ••X •••• • •••• • • • • 111111••••••••••
Co CA CA GA Co Co CA t.St GA CA CA M M M CA CP Co CA Co cA CA CA CA CA GA CA n M M n in Co I I I I I I I I I I I I I I
If) CA CO ^0 CA 0, CO *-- .1- IC) CA Co ..-- 0-- IN cr ,0 CO --Cl ,-. ,Ci CA CO sr ,- ,- N) 00 — Cr, Co 0-
0- n CO sr ,-- Co N. to uo N) "4" CA ,10 CA CA N. V, CA Co CO *-.. CO PO PO N NA Co CA Co NI cA --
CE In CA Co ,-. PO CA NI CA CA -/- CA ,- Co CA CA Co 0, Co +-. Co NA Co Ni ....- Co Co 15) ,- ...- ro n Co • • • . • . • . • • . • • . • • 0 • • ••••••••••111/ •
M Co Co M n 64 Co Co Co Co Co Si Co mA Co CA CA Co CA Co CA CA GA Co CA Co Co CA CA GA Co Co
I I I I I I I I I I I
SA
,- m r- CA 1,1 -4-
5- • Cr Cra C) W L. CD MC P--# -J AC MC Co CN Co Co S. ‹C 04 Lu L CO MC H --I 3= Co Co Co Co Co O 0 LL. U-
189
so Tr r, Tr so so r, NO Tr ro co so r4 un .- Tr Ii) ro ro 174 un cis co co Tr Tr un cn n cl Tr PO -0 ,-- SA CA -o .•••••• •• • ••••••
KA KM SA n SA n Sti M n CC. CA M
11111 1 I CO Cl sO r.4 Tr un n Cd CA CO co r, 0) "r n
M -o Tr n 174 -o n r,, r, os Tr -o ND ro so PO CD CA n sO CO CA CA CA CA .4- sO PO ••••••••••••••••
n CC, fAtAfAn SA n tM n KM n tM n 1 I 1 I I I I I
0, CO Tr u' -- to os n n ro ro In N. c4 .tttr -4- N NI
cn n ro -- so -o n ro n so -- CA ••••••• • • • MR ••••
MMMMMMMMMMMMM CA SA MI
L7 ty N N Cr^ in in co 0- N. 0) if) VD tsi Cr re) M Ni co ow, PO ro co N.
cn n -0 CA NO Cl N) V) Tr N, ts4 M ••••• •••• •••••• MMUMMMMMMMMMM M M M 1 I I I I I I ON ro ro os CA -0 ro Tr un c-4 to -- so N. CO D' ro c4 r, Tr n co sO CA 0, 00 0, CO In
C) r4 IS, VO CA sO NO Tr n Tr c-4 r--.1 0) uo • • • • . • •• NII•••• • M MMMMMM Mi n n
1 11111
O Tj
..-..
-0 r, 0, In N. CO 0, CO sO 00 Tr- n ro ro c4 co al
cis Tr- 0- CI) *-- 1'0 ,-- SA N. ,---. N. N) 0, .4. 0, N .-A
Z n M M CO r, Tr Tr -- Tr un -- m -4- -- In IN 17. • • • . • • • • • 4 N • • • • • ,.
M M M Si Ma n n MI MI rA ,-- la KM SA MI la 115 I I I I I I I I • ■-1
r--
Tr ro CA In r, Tr to uo so to CA ND SA -- M PO 4-D Tr C) cl 0. m cr. co c-4 un so 0, r, r, Tr iv) v-.1
5- • • . • • • • • . • • • .... ID
eM la KA Ma n MI M 0) MI Cl Ca. MI M la fa M CI
I 1 I 1 I I CL 7.,
M CO 1.11 ,10 CA Tr N) so Tr ,- If) ,0 CC, M ND In ....... N. 0, in n so ,- CA -- N In 1..) N In r, In IT.
.....1 la CA sO ,0 ro --- m KM eM SA In sD *I- +- ND 0- i.." • • • • . •411•1 ••11•4 II 0
M M M M tM M SA M ,-- la MI la MI M M Si •.-t 1 I I I I I I I -iD
m
IN -- los un -- Tr Tr so LO r, 0, ,0 ,-- OD ,10 M r-1 c..4 n '4- CO 0, Si CN ,C1 ,-- -4- mr to -4- m ,- iii
1-4 ,-- rn Tr cl cq .,-- rq oo n ro ..-- Tr IN -- c.4 un !."'
11 2 N • . • . 4 4 • V V •
la la la n la la Si n M MA la M MI Si n S4 8 I 1 i 1 1 1 I 1 1 LI
CA 0- CA N. VD ND on co co so r, ,eo m oo so cl -rn
= Ul t
•
IZ! in N N S4 MI tO SI r- in CO ..-- i.r) MI r, Ii' • 4 • • • 0 II • • • • • . •
IM n la CI M M ,- KA n MI M IS, St IS, IS, SA .... 1 I 1 I 1 I 1 ai
1-1 °,1tr OD CO ,-- CA r, co ch so un un cr..
-- ,-. so un (NI -- „- -- -4- c-,) co .- Ni riLl 2 co ,--
CD ND w- ,-- N CA In ,-- ,-- ...- •••V In r' un -- -- ro 115 • • • . • . . . • • . • . • • •,-1
la n M M MI ,-- SI la Mt SI fA kM MS1 n .1...- SI 5- I I I I I I I I I I 4-'
n 0, '4- Lf-.I ,0 r, V) 0-, CA ...- CO N. 0, KM ,-- 5-
VI ND if) .4. NO CA CA CA .4- -- ND 0, N. *0 PO so al
L. N) CA N. Tr ro n ro Tr ,- cis ro NO In CA SI Mt • • • • • • . • . . • • • • • 0
n 1M KM SI ,-- CC) ,-- M IS, M SA fa 0) St 0) n .---, 1 I I I I I I I ......
N N r.0 N r''', .1- cis if) 0- Ni so IN ....... T... Cr. to ai
CA ,- CA ND 0, .,..I- ND Ce. 0- OD •..0 -4- '1- in UO In u
LU -- M CA CA C'-. N) 0, CA -0 '4- KM .,4' ,- ,--- SA
••••••• • •••• • 4111, .;
M f 74 M T... M M M M n St ,--• V-14 KM Si trid- n • e-I
i I I I I I i I r.-
r-, PO 0- -0 ,-- CA OD NI N:' uo -,4. In Cr, -0 0- n ml -0 OD UD ,4 la CO 0- CA cl -4- OD ,I- CA r 0
CD ND la NO CA 40 CA CO •.i" r, la *-- Tr m ro Cr' Lin u • • • • N • • • N • • • • • • M M ..T. M M M M M n IS, IZA ,-- KM IS, la
I I 1 1 I 0 -i-.)
.1- -0 ,0 '‘). .1" CO 'kl- 0, n sO CO m) -0 CO r, ,0 Li
'1' If! KM - CA SA ro ts, ro c4 to ro - CA ND V) it.
Xi MA In SI KM •-• ,- la CA CA ,- Si KM M la Mi la LL. • • R 1 • • • • • • n . • a
MI MI MI KA KM CC eM MI n SA 0) fA CM n n 15' .• I III! I I I T^
CA N n Si N. CA N) CI ,z
M ro Tr ro m N r,. M •q- 0, 15' ON
ri
••••• 4 4 ..• • •21.•
to CC) M M CC, n tM fA IS, IS, la 0) eM I I I I I I I I fro I-
,-- CA Po .4- otr ,Z CD U4 L CD 2 1-4 ...i 3: DK CD CU CU CU CU
and correlation
190
CA 0, CA r- M r OD NI N. ro ON NO T... ...... ro un
=4- 4- -- CJ No rl PO CO 4- -O rq r. MI r- -O N)
CV r- M 0, r- M -4- -- un -- (....4 co m m No on • • • . . . . • • • • . . .
C m m m m m m m m m m el m Cl -- mil I 1 1 1 1
CE T-. NO M C.4 NO 1 0. r, 0, r, co -- ro un E) In
PO r.- rl No -- m M IN -4- -- ND NO r- ND r- NO
Cl -- m 1.0 M CA ND r- CA r- .q. -40 ci m ...- co ...-. • • • •
•
elle ••••• rr, M M M M M rsl M M M M M M M M r- M
1 1 1 1 1 I 1 1 1 1 al LI
ilD NO CO CA 121 NI LID PO r- Mr =3- -- ro 0, -- -- ,-- ._
IN IN =r 141 0, =- ND N. SI c..4 No CI-. =1- F.-.1 -,-. W
a N. CA r- r- NO CA 1.11 r- M r- PO ,- M -(I M M .,-1 • • • • ••••••••••••
MM'U
MI MI M M M MI M4 M Ma M M M M U) 1 11111 1
W
CO Cr- Ma NO Ma M DO CA ND CA CO CA r- CA r- ..-4
,- CA PO 0.1 NO 1.11 0, M CA OD CO ...r. ro .17.; ro CT- m tZ
CN r- Ma M IN VD V1 1- ,- 1471 LID 1.0 CA SI r- Mt E . . • • • . . • . . • • • • •
M M M M M M M M M M M M M M M M !:- 1 1 I I 1 1 11
4- 1471 ..4- -4- ro co ,1-. co CO ro MI NO ..4. co co m ,-
OD CA N. -4. m ip, CO CO 4- co uo t -- =1- ro c, :..:
co cl -- c, .=1.- r0 r..... ND NO CA r- Ma CA CA r- N ,...... .,i • • • • • • • ft • • • • • • r■
SI M n M M M M M M SI Si .e... Si n M 4... 4....' I !fill I I I ID
E CA Cr- N. Cr- VA N) 1.21 n CO MD CA M M ND LD
NO 11-) CA N. U-.1 ,r r, ro n CO M NO IN ND '0 CO Z CA r- +- CO N N) 1.11 M NO .4" Mt M .nr ro -4- 'N • . . • • • • • • . • • • . . . m m I m 4,-... m n m m 'Si ..-- n m m m el. i 1 1 1 1 i i I I
Cr- co co -4- r, 0, ,- 04 CA
,- 4-- CO .4- '71 i'.4-5; Pt 117- rl 12 ;:?, r L„, „
• „, ,...... ...... „.. ....... „ „ ,,, „ N. 1" r- I- ...- -- -- • • • ••••■ • •• 11•1.••
M M 'Si 'Si M 'Si M MA MI MI St n M M M4 M 1 1 1 1 1 I
1D, ,- CA !JD ND .4- M r- ...- 1- -- =1- CO N. m N.
S =I- m r.. N m c.,1 N) -- un r, -- in .1- J -- IN 1471 Si) NI CA r- ro Cr- -- un r..1 4- m c... No
• • • le • • • • • le • • • • •
MSIMMmrSIMMMMMMMisiMMA 1 1 1 1 I I
M No m -0 m in -- N N. 0, CA N. un OD M ND CA N... NO CO CO NO Cr- CA CO 'Si ND CO Cl- N. IN M
1-4 'SI el Cl r- ,-- 'Si ."- 00 CA =1- m ro m n -- IN • . • . . . . . • . • 4 • el • •
M S'S SI M M Si Si Ma Ma M M MI MI M M 1111 1 1 1 1 1
,0 -0 CA M -0 Lo ro C4 Ma ,- ,- 0- CA N) M IN ,0 un r.. =1- 4- 'SI 0, OD CA N) Ma CO ,0 Ma ,r -0
:c un CA 1.0 N. N. r- M r- r- rd ^O N ro in ro uo . . . • . . • ••• • ■• ■ ••• Si SI M Si 'S SI ..... M M n 'Si M SI n M r31
I 1 I I I 1
NI 141 n. 0) N ro N) -- ,o c, in CO ND '0 LID -0
0, CA Cr- 4- co No N) l',.. ND r 10 ma NO N. YD Co r- M r- CA CA -4. -- 'Si rl IN 4- IN 4.. cl m ,F. • • . . • • • . • . • 4 • • •
'Si M M M 'Si ...... M Mt MI Ma Ma M Ma M ,- M 1 1 1 1 1 1 1 1 1 1
CO iiD NO r- Ma .q. ro .4. ml No ,- n O. 'Si No 'S rq un n 4- N) ro -- ND N) -Cl N)
Lt. 071 CA 04 -CI LO .4- Cr, IN ,r -- O. ,r ro ,0 N CS4 • • • • • • • a et . • • • • *
M 'S M4 M ,- Ma M M M M M M4 Ma Ma Ma Ma 1 1 1 1 1 1 I 1
'S. NO 00 =1- =r- n ..zr op -,-- h7I N. CO 11'.1 I...., ID F.0 ....- NO MI NO En '0 n o- =.r. IN IN NO NO CO ,.* N''.1
W N CA CA ro 0, ro 0, -- No -Zr m un N. -- M 04 • • . • . • • . • • . . • • • 12
CSI M M r- M n M M Ma 'S r- M Ma Mia M Ma 1 1 1 I I 1 I 1
M r- r- 0, NO Ma r- CO CA rl Ma IdD N... NO ND 0, IN ro -- N. un =4- Si ro co m ro N. -- ro 0, Zr,
(...1 CA M Ma 04 CA CA NO IN -4- -- ..- m m uo .-- . • • . . • . • • 4 • VIO0•• M M r- M MI M M IMI MI Ma 'S r- M4 'Si M4 =-
I 1 1 1 1 1 1
N) r .,-- ch.. =4- ,-- (-4 ..... N. ,o 4- ro ro 'Si Cr, -0 IN NO 04 M4 =1- cl 00 N. In r- n Cr- L.... =1- ,-
Pq r- a- m CA CI 'S ...- ,- ,- .....- ,... 4$4 'Si ,- M Ma . . • . • • . • . • • • • . • •
'$4 M M 'Si M M MI 'Si m m m el m m 'Si M I 1 1 I I 1 1 1 I
r.. -- ro --• ro n CCI OD 0, .c.. uo co .1- No n ,- ro m PO r- N-.1 Ma n r- OD r- CA '0 ON U7I U^a 0
<7 N ..... 04 NI 14-1 CA In M M 'Si EN CA Ma UD r- . . • • 0 II • • • • 0 • • • • •
M M 'Si Ma M M M M M M M Mt MI MI M M I 1 1 I 1
04 ro
<X 04 C) UJ LA- CD MC 3: 2t C2 CM CV CD CV
191
M n r.1 a -- r, m ,- -- a No PI CO ,.... CA co .7r CO 1"1 In ON 00 ..-- CO ,- Cr- n ,- CCI ....- C 0, Cr,
ci ....- ..-- in- -.-- N (-1 m r, 6 MI CO ,- rq NO N • • • • . • • • . • . • • • . 4 M M M M M M GA M M M M M M M4 M ...- I I I I I I I
Ca
Cr- IN .I. C..4 CO 0' .,.. C.4 '0 a c4 cy, r4 r, in r, -.1
N, MI Co EN r4 cl, r, Cr. rA .4- ..-- in 117 0, ..-- •40 N. 0
CV ....- ..-- ,C1 I.) In IN t...4.4 *4 .4- a- -ci in ro m co *4 ii... . . • 4 • II • • • • • • • • • • Cr, M *4 *4 M *4 M M M M GI MA n M M ...- ..--
I I I I I I I I I 0 4)
a- ro OD 0, 00 .....-. C4 a r, -- co a- or -o or m 5-
r4 tr) r, -4- .4- CO rei -0 -- N, co .,.- m ,0 CO ,- 0, M
CN NO M t :';'4 a m 4- rl i- ....- G4 MI ..- NO GA -....-. • • • • • • 110••• • • 11•• 4.1
M M Sr! t M GA n M M4 G4 n n G4 MI G4 M .--4 I I i I I I I I 1 M ..7
... N. ...0 04 '0 CO n *4 117 Cr. l', rA to N. in rA a,
-- ro No ro No ca a ..r 4i) N 47 trj ,D ..40 m rq m (f- cr. m -- *4 ,o vo In CA CA a -ci If) GA Cr. -- v., ,- •..•• . .. • • • . . • • . . 5-
M M et, M M M M M M M4 M4 GA GA M4 MI SI 0 I I I I I I I 4-
..:
-
•
-- --0 G. ND .4" Cr- 0) MI 1.0 -- *ch NO OD 14-4 c.4 •ri
- t I IN 4! V) t5.1 471 ,- m r4 m cl 00 ....- 5- • • •••••• • • • ••• 4.3
M M M M m m m -- m m m -- m
i 1 I I i i I z
o- -- -- in r- Cr- co ro vo uo -- c-4 r, a cl. r, ....,
ro --o r, No m ro -- a- r, -q- ro -- r4 ,- in ,1-. 13.1
.7tt Cl GI ..- N In NO Cq c4 a. r, ON t-S! 10 M OD 1.'4 ..--1 • • • . • • • • • .1 • • • • It • Cr,
M M GI GA M GI M Gt M M M GI M rit M M r
I I I I I I I I al •,-1
in or CA P.... r, co in r,1 -- -- r4 uo 00 .r In 10 5- CA CA CA M IN CA Cr, M Cr- *q" ro m a r, in m • • • • . • . • • • . • • • . • ,..
m m m m M M M M M .-- M M GI M M GI al I I I 1 i !a
L
.-- cr IN -- r4 co ro c.4 co ,- No ,o m r, n _,
ro ro -4- co m co *-4 in o- co co o- a -ip la uo -- ....I rl M N. NO c- tr... M VO NO ci r0 .4- PO rA In IN . • • • . • . . • . • • • • • .
M M SI M M M M4 M4 M GA M GI M4 GA M4 M 0 I I I I I I I I .1-4
.1..,
M No r4 a Cr.. tsl co co co -- rq Cr-. N op Cr- N m
I-4 M M M NO CA ...- ...... 00 CA 1.0 rl M CA ....- GA G4 04 • ••••••••• • • • • • • r■••
1 I I I 1 I I 1 o 1..1 NO NO C, IN, 0, -- NO NO CO If) -.0 ,o in co uo a
r4 -- co in ,- -o IN in n No a r, r, a a N T1
x a la a No r, -- ro -- m N N r, C4 a ,- ro r- • • • • • • • . . • . . . . • • m I I 1 I 1 I I I
a!
No -- a- r4 in IN CO m r, r, a co a n uo co ,4
CA M IN OD Cr- IN 10 NO M NO a -o re) r0 in re) Cr. CD .--. ri ts! a a- co (-4 -- r4 co co is! r, m a- ro
• • ••••••••••••• • M
M M M *4 M .I. M M M MI M M M M4 ...- M4 ..-1
I I I I I I I I I 5- CA CO M 40 in If) GI .3" 0, 40 M a ..qr v CO 1,-,1
14. ."4. .I.. .... .40 ID op m r4 r4 (.4 No ro ro uo -o -- 6 • • • • • • • • • • • • • • • • ",.
M M M M .- M ....- M M M M MI M M M M 0 I I I I I I I .-1
.....,
IV •ft •• ••• •••• • IOU It
*4 *4 M ..... M *4 *4 M M m m t la m m m m
I I I I I I I I •.-t ,:-
r, c-4 -0 co a m r, in -- co m -- cr, in co r, m o- r, Cr, uo m -- IN op co 04 OD OD rl ..-- CA XI
C.) Cl GI ,-- M N ,- ,0 GI ,0 GA ..- ,.... M ri 0, ..- C1 • • . • • • • . • . • • . . • • U
GI M .-- St M M M M GA M M -,- M M M ,- I I I I I I i S-
CI
GA 0- 00 OD 0- r, -0 M rA 0, -- .1- -- CA 0, CA -,,
04 ,- N I:13-4 n ,-- CA MI GA M M M4 ,-- ..-- GI CA -- M • . • • • • • • . • • 154 33 M i M I$ GA G14 14 37 M M GA
i I
a a- m ro ro ro -o cr n No r, co r, ▪ N. co a -4- N) a -c! c4 GA rA -o CA N, • M CA M a -- I) ro .1- • . • • . • . • • •••• NION
GIMMMMGIMMIGISIM GA Mi ra GA I I I I I
cl IN) .4- ¢ 0.4 C.) UU L. CD MC I-1 NC z CD ci CN ci CN
ro No a a ro in -o a m Cr,
-0 a a ON cr r4 r4 co a cp r, co .4- r4 No 14!
Lu m m m a cr r, or a No co co IN m in r4 I1
Table
A13
:
192
If) :q• CA CO CO ON NO N, CA CA CO N. NO CO CA r, r, un CO r, co c-4 ON NT- NO PO CO ND
4-- CO CJ Csa l57 CSC CO CA ON N Vl CO
cq ro
..4C AO (-7 LU U- CD MC -J A: Z CD I= C7 CN CN cq ro
..4C AO (-7 LU U- CD MC -J A: Z CD I= C7 CN CN
CO ,-- CO CA CO CA CA CO =-
I I 1 1 I I 1 1 0
.4- '0 Cr. '0 ON CO r, CO PO In V, If) CO N.., CO Cl
ND N -0 .4-• CO N) ,- ,-- N -.0 CJ In CO CA IJO 5:-.
CII CO CO In ,-- CA V, 4-.. CA In Ni ..r -4- n -- ..4. CN Cr, • • • • • .1 • • X • • 0 • • II
CO CO n CO CO CO CO CO CO CO CO CO CA CA =-. CO ai I 1 1 1 1 I I 1 I I LI
1.171 CO CO in NO .4- CA ...1. N) ..-- N, LO CO CO In CO 64
CA CO In CO CO CO ND =- If) NO ....... ..a0 CO If) 0-. ...- CO ....1
NO ..-- -.4- -- -o m N C.4 C..1 SI -.:1- CA N, -CA 1-.1 • 4 • • 14 • • • • • X V • • • Lit
CO CA CO CA CA CA CO CO CA CA CA CO CO CA CO CO
1 11111 1 1 111 ...--1
!,-.3
• • • • • • • • 4 4 • • 4 • ft • al
in CO M M CO CO CO CO CO CO CO CA CO CO CA CO ,--1 I 1 1 1111 1 I I OD
C.:
0-. VD V) IN OD OD CO N If) CO IN +- ,- +- -0 0, M
ro ,- -- o- -- r, to -ip CJ cr. r, ro -- -,:r .,-4- .1-1
X M M SI -0 +- ro IN in m V.) NO M N SI .4- ,- r-
. • • . • . . . • • • . • 4 • • 1..:.
M CO CO M M CO M CO CO +- In CO CO CO CO Cl I I I I I
al
cr. ro so r, PO MI r0 0, CO ,0 OD OD 0% CO CA IN 9- -- CO ro CO .- un N '- c...! N CO n 44- PO CO CO
.J CO MI 01-. •‘40 CA N CO CO CO CO OD CO OD CA -0 ..r ..., • • • • • • • . . • • w • . • ..... CO CO CO CO M CO CO CO ,^ CO CO M CO CO M 1--
I I I I I I I !=.• CI
0, M PO M ,- CA IN M 0, V') C..1 .q. In ON CA -0 ••-t CA Li NO ,0 V) 0, h71 PO 0, 4.,
1-.1 .- ..- CO..-- COCOCO0, COUl COgl CO ;F.4 P! ,- ai
„„ • • • . • • • • • • r—I
CO CO CO CO CO CO CO CA CO CO CS, CO CO CO CO CO W I I I I I I I I -
r-
co -r ro ro co NO ND V, CO Cr, Cl
ON CA CO 4:1 -4- -.0 CO NO CA CO ',lc tom,
R *1-7, F;s1 t, :c n -- 1-0 ,0 N CO !J71 CO CO N) 4 ,0 MO r... ,-- or • • • . • • • . • • . . . • -1-1
CA CO CO CO CO CO =.- Mt CO CO CO CO CO CO M t-- I I I I 1 I I M
L .-- ..r CA h0 N -0 c-.1 -- CA .-4- CO CA ro -,.r un co al
CD ....... ....- .,..- .tr N) co -- -- N) -.1- CO CO ro M ..-- PO I—I • a • * • • • . • • • 4 • • • Cr,
CO CO CO CO CO +- CO CO CO CO CO CO M CO ..- M != I I I I I I I I I iri
• H ....... ..q. -.0 r, .1- -- cl co fir) cr. cq uz! -4- r, -- ...-
CJ m NO if) Cl- CO CO In ID -CI N) Cl CA CA ro co 4-, LL VI CA CA VD N) VD M M CA CA If) .1- OD IN rte) n • . • • • • . • • • • • • • 5-.•
"S., CO CO CO T— CO =-. CO CA KA CO Cr,! CO CO CO CO al 1 I ; I I I I I 2:
CO co CO o- r) v..; CA CO ..-- LO CA ON CA CA ==. ,.r 1 n-4
CO VD CO CO c:'4 CO ON OD In NO == -0 CA 1.'7:: N OD .... La CA M M +4- OD N 0,.. TT r... CO M -- r, c.-4 cq CJ
. • • ••••11.1•R 11 4 4 ID
CO CO CO -. CO CO CO CO CO CO ,- CO CA CO CA M ki 1 1 I 1 i 1 1 I if=
-4- N) co CJ -- -4.) N CO CO CO un .=;- N CO ND CO •,-1
N) -o -- ,-- -o CO c..; r, cr. N NO -0 N, == In ,-- 5.-
C..) CA CA ND ,- PO CA -,1- m m t.,...! -- c.4 m -,4- r, Vi 11) • • • • 4 • • • • • • 4 • • ::-
CA n =-. CA M M M CA ..- M CA =-. M CA CA =- 0 1 1 i I 1 1 1 I LI
ND NO In .1- NO ,0 VD Mt CA VD CA +- IN CA CO.. ,T ,.:-
VI V, In 1.0 CO CA +- .,-- ts! ,o Ls! IN 00 m ,0 m 0
Po CO n CO CO -- -- -- '- CO M CO CO CO =-- CO CO .-1- • • • • 11.1.1.11•• •4••• LI
CO CO CO CO Si CO CO CO CO CO CO CO CO CO CO CO 61
I 1 I I 1 I I I I I LL
,-4-. Vi in N N. 0- .,-r ro o, .r CO. ,- .cr <1.- co N) On
N. C..1 "0 r..., n CO C..I IN ....... .q. SI In 0.) n C..1 ..... ....1-
cr.• CO c...1 rl -a -- N ..-- CO CO CO po ...- -.0 CO ..- ,- • . • • • • • • • . • . . • • • CL SISIMMMIntn SISIMMSISISIMM
I I I I I a! r—ri -CI ii, h-
193
CT- 00 r, PO ,r m PO CO In<A CP -0 r-. CO C4 CA CO ,r cr. r, No CO CO CO ON CO N) CO NO
cn CO CO cr. CO CO cq ,r CA NO +- CA CO CO NO -0 II • • • • •1111 41111 1411•0%
CO CO CO CO CO CO <A CO CO CO CO CO CO CO CO ,- I 11111 I 1 I 1
CO CO .7- vD In Ck -4- ,0 1...... 'IT C4 .4- N. vD 0, .:1- ca. .
M c?.. cq CO P, C?- ,- ,0 DO CO 0- In CA 1.3, CO r, cl, LI
Cl CO CO In CO r- CO Cl C4 C4 r- ,r No ,-- m CA CP Cl •••••••••• • • el• • • r••••
M M M M MM <A <A <A <A <A <A <A <A .-- <A Cl', I 1 1 1 1 1 I 1 1 1
......4
II) Ok CA In CA CA -0 CO N r cr. — m — cr• uo m
cq cq ,r r, CO rn CA CO 0, 00 -- In ,r 1') No p, ...1.. ..!..:,
C5 NO r- C4 r- NO CA NO <A <A ,- CA CA Cr, NO <A +- 0 • . •••••••• • • •• • • 4-, CO CO CO CO CO CO n <A <A <A <A <A <A <A <A <A I 11111 1 ill
.--1 PI +.- N In N. ,-- DO OD ,- .r- .....- M In CA ID, PO 0,
,-- Ok CA <A cr. ro m co ,r. — CA M.! -- PO +- <A z
a <A <A <A NO ro ,r- CA <A In DO CA 0, <A CA <A
.. • • • • • • • . • • • • • • !:-
<A <A <A <A <A <A <A <A Ms! <A <A KA ta <A CACI 0 I I I 1 I I 1 '+-
-O CA m ,r -lo ro -- -o r, N) ,-- CO C4 ,- CA +-
'q- N. CP. CO 03 N ro ,r ,t- r, ,-.1- co ,4- ro ro No ..-1 C0 C4 CO Ok N) Cl ,--- NO PO N) ,- <A N) CA Cl CO PO
• • • • • . • • • • • . • • 1-)
CO CO CO CO CO CO CO CO <A <A <A ....- KA I) <A ...-. M I I II I I I 1 i Z
Ok CA .7- -- 'q- ,-- rA ta NO .,-- rA <A r' 1,O 00 DO ..-..
CA DO .-- P, PO 00 <A KA PO <A <A .1. rA N CA CA al
.7-e.. ,-- CO -- co r, N) DO CO DO DO P, CO 'Zr — 'Zr N r-I • • • • • • • • • • 11••••• 17'
M M M CO M is, is, is, is, is, is, is, <A is, is, ta 0-- I I I I I I I I M
• --I
CA ,- ,-- PO DO CO PO <A ,r r, cr. m No ro m r r
— cr, --o -4) MI 03 .1- CP -0 — CA -0 -0 N. N CO 4,-, <A +- <A '1' ,-- CA CA "1' <A NO ro — ,r m --
. • . • . • • • • • • • • • . . <I CSI MI <A <A <A <A <A <A <A <A <A <A <A I:5 is, <A el 1 1 I I I 1 I CL
Ca.
<A 1-0 r.1 1'. r- 0, vD OD c‘.1 ,r ro — ND vD <A +- ...1
ro uo r, 0, ci- -0 o- -o -to uo co ,r CJ r, uo ,r ,... _..1 r- CA DO In 1'0 ,-- <A <A ,--- <A ,.1- ,1-- DO <A In CO • . • ••••••••••• • 1.• -
CO <A <A <A <A <A <A <A +..- <A CO <A <A ta <A ta 0 1 I 1 I I I I •,-1
-r-
,0 N CA CO CO N) — CO -0 CO CO CA N. -4:1 CO N. 05
CP PO In DO PO OD PO P, I.N, cr. m ci ,r r, m! h71 .--1
I-1 ...... CA PO +- ,--• <A ,-- <A <A ro m ,1- m m ro ro al • • • • •••••• • • • • ••
CA CA is, CO is, CO is, is, •-• <A is, is, is, is, is, is, is, 5.. 1 1 1 1 1 I I 1 0
,CI Ch CO OD CO ro DO .-- If) PO N. CpI- COC•!1-0 CO0:,
In VO ,r ro uo Cr ,r uo — vD CA <A DO PO <A I
. • • . . . • • • • . • • • • . m CO is, CO is, m CO r- is, is, <A is, <A CO KA <A CO
I I 1 1 I I .... iii
LO Ul r- ,- ,r m cl ci, cl co ,o cl, CO CI CP ro 1-4
CA Ul CO In 1.41 CA PO <A PO CO <A In CO PO r, ,r cl-,
C7 CA CO +-- CA CA NO +- .-- CA CA 'Zr cq Zr c%) — ,r r- • ••••••••• IS • •• •• M
M M M CO is, +- CO is, <A <A <A is, is, is, +- 's, •,-1 I I I 1 1 I 1 1 1 I r-
4.) N) bo ,-- to "Zr to C-I CI- CO No co Cs r- Cr- r- ,17.1
1'... <A r- <A <A CO Ck In <A CA CA Ck CO <A NO PO ,....
Li. 'Zr C4 Cl r, ,r ro cl, — 1.0 ....... 1,, 1.■0 1......1 ,o 04 is, ID • • ••••11.1 ••• • •11•11 ...4.
<N <A <A <A ....- <A <A <A <A CSI <A <A <A <A CA <A C1 I 1 I I 1 1 I 1 r-I
....0 O. OD C?-DO cr C,,. ,-4- m No cr. m N. Cr- If) ,-1- --
No ro m m -- uo CA N <A Cr, MA 0- CA vD Cr.. NO al
Li! C4 r- C4 Cl 0, MI 0, '- 1'... ro co 'Zr r, — CO — l..J
. . • • . . • . . . . . . . • • .7.: CO <A CO +- C: <A CO <A CO <A <A <A CO CO <A M
I I I I 1 1 I 1 ••••1
M .4" r- r- l',.. 0, co .1- r, — ,r r, DO CO N CA ,0 M
ID. <A In DO N. DO 00 <A CP DO <A <A <A ,:t. ro No :.---
L) r- CO C4 Cl C4 C4 No ,r No CO — ro CO CA N N) CI • • . • • • • . • • • . • . • • u <A CO ,-- CO <A <A <A CO CO is, <A +- CO CO CO *-
I 1 I 1 1 I 5- CI
1.0 ON. ...... ..:1- -.0 0D No .c -!-:, PO <A CO -4" PO CA N.
<A NO <A la NO Zr- m -ci co is, ro in ,--. CO CA <A LI
04 ,.... -'Zr CO -- -- CO ,..- — 11.•• T... CO CO CO is, CO is, M • • . . . • • . • ...... • 14L.. is, CO CO is, is, CO is, <A is, is, is, KA <A <A is, is, 1 1 I I I I I • •
In N. OD NO CO ,-- +- CO CA N) 0, CA DO 'Zr -- NO ,- Cr- vD CO CO NO In CO 0, PO CO CO N. OD CO CA 4
4 CO CO C4 C4 Lil Cl 1.0 ..-- ,- CO r- 04 CO ,4- CO T. •••••••.111.1 • • • • • • al
<A <A <A <A <A <A <A <A <A <A <A <A <A <A M4 <A ,-1 i I I 1 1 I .o
IV r-
ro 'Zr ¢ CO 4.) LW Li- CD MC r~ -.I NM MC cn CN CI CV CN
194
ro ro i i Os CO In -4- ON 0.4 IJ 04 ▪ ro uo 0 to IN N. N. uo 04 os c, so n CI n ts! rq rq m OD n 01 CO 01 ,10
•
•
• • . • . • • • • • • •
tn I 44 M M M n 44 n n
I I I I I I I Li.
ro .4- os m us uo r, co In 0, ,- 44 mo m LID Os os Cl
CILI 44 4:1 sO 01 ro r, m .-- '73- -4. so uo MO n sO 44 I-
.• •••••••••••••• 17! 40 ra tn 41 44 M tn 44 40 ta 44 44
I I III III 11.1
P, In Os M1 01 In PO In OD LO CA N. Os n -!.. IN la r',.. 0i In .mt -4- -- In PO .cr ro 1-. o. If) .cr el cn m ro n uo m! uo m ro — m PO M r,.ts1 r.4 .4-,
. • 111.•• 4 •••0110 4 •
SI n SI SI SI Si SI M SI n n M n 44 n CI IIIII
ID Cl •-• M -40 01 cr 04 44 '4- sO If) 44 0- 44 PO ,- 4-
• 4 V • • • • • II n • • 4 •
M n Si Si 'S n M n M SI ta 44 44 44
I I I i I I 0 4-
.tr ,- CO .1- If) so r, CP- .q.- CO N.
cr- los so r., r-, to ..4. so ct rui n rq so .c.r n n x
CO CA M 0- -- CA M If) ,-- In ,..- t3 PO M CO CO T.0 •,.., • . • • . . • • • • . • • • • II r....
M n SI n M M SI IS M SI n ..... M M
IIIIIII I I I al
Z
ro -gr —N. c.,4 so r, m uo co -o t... .1- so c4 ,- In -4- -o ro IN uo r., cis so -4- o, r.14 rl Isp bo ,-
al
••••••••••• .•••• r.-4
44 44 4.4 n 4:I! M 44 44 n to 44 tn tn 44 44 41 Cr,
I 1 1 I I I I !-'
c..4 -4- c...1 so In IN VO ,0 l',.. r,.. ,- o-. ,1- .•.:- so ro •,...,
• ••••••...•••.•••
m ts4 4! Si 44 tn 44 44 44 44 44 44 44 44 44 44 5-
I I I i I I CI
• • • • • • a • • • • • • • •
SI n n SI n n M M M SI n SI SI MI n SI ..:::
I i I I i 1 1 C) •,...
LO N 00 CO In T- L) -.0 M . r ro cp.. i,--.1 .tr ,- 1....4
• ft • • • • 0 • • • II NI II • • W
Si SI SI tri t SI in P — ts, l M tSI1 MI M M M r-
I I I I I I 0
co so IN -Cl 'N 04 co M M -.1.. -40 .,-- CO 11- Os
- so N. .--N ,-- MI NI (-ANN Ir.) T- r0 In
. • • 11 • • • • • • • . . s r=
44 4.4 tn 44 44 44 ,- 44 0:i M n 44 4! 44 15) n N
I I I I I I I
NO 0- .4- ts! to un n co co to so vs, ..4- op “-1 or, 0
0- N T- lxi .gr t) N. ro cl uo uo In -...1- -4- n to ...A
•••••••• b9111.0 • • • :::::
M M MI M M ..- 44 la la 44 44 tn 44 n .- 44 IC
I I I I I I I i I • ,_, c-
MO IN 00 CA sO Cs. PO ,0 Is) 0:i 0, If) Cl ,0 4111•••••11•11,•••••
1
Ill
15)
SS . n 15) 5) S S S S 15) 15) 0:,
I I I I c
-O1-corrruo
▪
rqmo tuto
!
„ m! -0 -o -4- co .q- ro OD -o ,0 CO CA ,- CA MO N.
0 • • • • ••••••••••••
Li
SI M M .... tn n 44 M tn 41 44 44 44 44 44 tn I'''
I I I I I I I I N . ,-,
LN los -- os uo ,- co LI, rq IN co os -,:r rq OD r-
ro IN IN r, rq uo IN r, cs ts! ...4- rq to m • • • R •11W••••• U• NI 0
44 la ,- la 44 40 n la tn la 44 ...... 44 tS) IISII ,- LI
I I I I I I I
.4- ,0 sO PO Cl-. P, CO PI ,0 .4- n uo If) 44 CA If) 0
N) -Cl CA -0 in N. in P, ...- PO MO CO ,- 415) Cr.. -1-5 P4 M) --0 n 15) ,-- M Mt M n n 0:i to =.- 0:! ,- M tj • • • . • . . . • . IIIIIIIIn • N
44 44 la tn 44 44 M M n M M n M 15 n n L!...
I I I I I I
n N) Co I.11 ro N CO If) Os N rl ro n co 04 ON T. Cr- If) CO Cr- uo .q4
tt M CA M m m „- „- rq •- • • • • • • • • • . MMMM)MMIMM)
•
Mi M tn 44 n I
-0
ct
13
I—
1.41 ‹C Co C..) LW LL, CD MC P.M cn CN cn an cn
0
195
▪ ro tt Cr, Cr. CO Co Cl hl CA CA N) MI CO Cr'.
uo uo bl t Co r- ^ O CO r- ,- Co Co CI Os ro
Co Co r4 ro r4 n 0- Co (-4 No so
. • .
▪
. ••• • • ••••• MMMMMS/MMI t52 M M C2 to M
I I I I I I I
co -- m oo -- r, so No .:...r .a- uo m4 No cl r, ,- o.
M in Co IN r, co Co r. r, -- ro ro cr. -o No -4- ,- .7., G -- -- so ,- IN r, -- -- ND NO IJ7 1:7 CI Ms IJ7 ,.... 0 • . • . • • • • 11 • II • • • • •
Co M Co M Co Co M Co Co M tM Co In Co ,- ...... 0%
I I I I I I I I I 1-1
.t4- un m r. Cr, 4- ro b7 CA V, sO n OD CO Co N, 111
CA ,- (-4 so ND Cr, ,- b7 CA C,1 b7 OD N7 CA N) -1-0
CV sO Co rouo Co b7 Co PO ,- Co Cl r- CO Co Cl 0 • . • . . • . . • . . • . . • 1-1 Cl Co Co C2 M Co Co Si M Co Co M Co Co MI M
I I I I IIII I iii 4.)
C7, r., ID, C4 n PO CA ID r.... 00 In oo -4- .cl- ro
ID co CA r, CA CO NO PO sO CO Cl CO CO Co N, Co M
CV Co CM Co ro 4- (-4 ro ..4- b7 b7 ..-- sO ...- Cl Co • • • • • • • • • .1 • • • • 5.■
Co Co Co 6r M Co Cl Co Co Co n M M Co Co Co 0
I I I I i I I q- 4' so -- r, so co to so ,- Co so Co Cr, N. so ..,74
••=1
co -q- r, to r, co n 4- 1..... U7 1 s0 N, Cr0, b-
CD Cl CI Cr- N) Cl n -.0 ro ro -- 15) ro -- cl co N) e.- • . • . •. . . • ••• • IS 4-1
C2 m m m Co SA MI MI in M M r- MI n MI ....... 65
I I I I I I I I 1 I E
so ro -- IN r, ro -0 Cr- N. N. uo r4 Cr-- -- r, N.
r, uo n IN ...- uo ro co n ro n -- ro In No .4- 0)
z ..--. Co ..-- co so b7 N) Cl b7 .0 M ,- .4- (5) No r4 r-1 • 4•••• • • • • ••• ••• Cr,
Co Co Co M MI M (Si Co M MI ,-- M Co M M MI r^ I I I I I I I III
•ri
sO CA -0 n CA CA CO 1.17 -q- b7 MI CO CA CO ..-- .0 f-
CO Cr. ..-- CA .4- -0 co 0- N. -- 4- ..4-. -- Cr- uo -- 4-) • rq M MV, ..-- 1" ,- rl M r, b7 *- 1- Co V7 (5) • • • • • • •••••••••• 5•-•
Co (Si Co MI Co MI M M M Co Co Co Co Co Co Co Ci
I I I I I I o o
r... r, ,-. tcr.! ,- lo- r74 -40 In 0., ...- C?.. Ln r, M T.. ....' ....1
00 VO 1- Co ,0 CA MI .4- cr uo co rz) ,0 IN N. -q1" ,... -.I *--,- .0 N. (-.4r- n ro co n -4- .4- PO CA V7 N, • • • • • •••••••• •• • f•-•
(1.4..1 Co Co Co Co M M Co Co Co Cu Co M Co Co Co CI
I I I I I I I .m.1
-0 C.4 N. C2 N. OD sO ND ...:r uo ,ID rA PO -O M
-- rs, n uo .tr MI 1/7 0, r- CO U7 CA SD r...1
F4 n r.."^ r(:,1 C.4 n c.-4 Co ro CA CA 1i' rl n CA ND 13.! • • • . • • • • • • It • • • • r_
Co Co IS, C2 M Cr M r- Co MI MI MI M Co Co Co ;:.. I I i 1 I I ! i b
Li
,0 CO N, ,- CO ND r, cr OD ,- OD 117
U7 CA PO CO rA 0, M PO CO 0, CA CA 1.17 117 127 "-I
= -,,r m Lio r, ro ro n -- ro co CA b7 CA b7 r • • • . • • • • • ••••• 115
'5) M M Co M .... M Co MI Co Co Co Co Co Co I I I 1 i 1 1 ......
el
In ro ro r4 0- 4 CO -- ro ro ro N, w- .0 ,--1
C'- 0- ,-. .0 V, Cr, r- ,0 (N .<:1- PO MI ..-- r- N.
CO r- M =- ro ro m..! n -- uo r, ....- b7 *- PO ro !F.--.
. • • . • 110•/1111111 ••• M q74 n n M M r M 454 M M M MI M M r- M
•:1 I I I I I I I I I
• • • • • • • • • • I • • •
MIMMM MIMMan tS1 tat n CSI C2 M I I I I I
113
..- CA ND 1- • CC1 C.L LU 4. CD MC 1..4 -.I NC DC CM C11 CN CV CM
41 N. Cr- IJ7 r- .q- Co co -o (-4 r, -0 .t.l. No lo... -q-
0, /.- n sO Os CA 1. .cr ts: .1- uo 0, r4 or. IN -- r-
I.J... NO r- Cl s0 .q- s0 Si N7 VO NO L7 -q- 1., al • • • • ••••• ••••
Co! n n M ...... M T.-. n M M M M M M n n 0 I I I I I I I ..-4
,...
1" CA b7 CO U7 CO Co ..4- ND r, PO -0 N) N. *0 h7 CA CA Cs VI Cs "1- PI CO ,- NO CA ,- 1- urI ID
Ud M MI n CA CO b7 CO U7 r, ..4- Cr. -4- o n rq lz1 LI • • • • • •••••• 119 • r-
m n Co -- Co Co m n Co Co Co Co (5) n Cu.Co M I I 1 1 1 ' I .--1
1., N, PO N, M -CI U7 -0 1.0 ro ro r, uo m n R:i
0, ..4- v- -- co -o m -- -- -- sO CA -0 -0 '40 0"
0 CA n CA r- CA .,..... PO sO m! =- IN M VO CO ro ID • • • • • • 4 • • • • • 0 • u m m -- m m m m m LTA m m -- MI Mi M r-
I I I I I r- o
.gt rUs 0... T.... M In T` CO try n '4. r.... .4. N 4."'
V7I V-.1 r'.I L171 M CIN. (-4 ro 0, uo .,4- .1- uo -- Cr. wo U
ing q;1 lil M Co M M r- MI SI MI M M M In MI 65 • • • Id • • •••• •• • L.
Co Co MI M C2 MI M M Co Co Co Co Co M Mi Co
I I I I I I I •• r,
-cl r.. -0 uo CA OD b7 ..,4- 9- ro o uo ro ,-
uo ro n r..1 q .q. Cr. -- o N -0 (5) .,r *-- CO CO .r.Z <E CO C2 VD Co c4 .q.- n -- rq -- ro m! uo --
196
ro N ON n -ci sr ro ro r, uo ro es ,-. ro sr sr sr m CA r- M N) '.0 mo cr. sr (y, (y„ 1,0 '.0 sr (N Cl M MI Cr'. ,- M IN N) CA D7 ,..- ,- CO M S4 ^0 "0
••..n•g 1111••11011••
M M M CA n M M M n CA M tn n n ca ,- 1111111 I 1 1 1 Ci.
.....
N) 'q' CA CA N r, so mo mo cs r.., '1- ,I0 in 0
N) ro bo so sr N) CO CA bo m oo sr co -- r C.4 n ....
CD M M b7 tat CA N CA CA ro ,- sr ki7 ,-. M CA Q", ..• ..•
.• . • • • 11711119 M M MA M n M MI MI MA M o; tn n tn ,-
Si ,--1 I I I I I I I I I I M
.i...
CA CN r- OD sr 0, 1.17 r, sr m n sr sr IN ro OD 0
cq sr es mo n in co r..1 -to r, (N N. Cr, sr EN -,-,r -0 4-1 Cn T- Cl T- '0 r- NO V..;:l T- Cl r- r.S1 r., n n . . • . . . . . .. • • • 11 9 11 at
MI M CO CO MI CO CO CO CO CO Mt M CO CO MI CO LI 1 11111 1 I f=
al
mo U7 M Lea sr es OD 1%0 N in cq -0 r, Co CO M •,-i
r, .0 r‘i .,:r ,- 1.17 b7 0, CA CA CA ND PO CA sr Li
Cl CO CO CO C, ro N) tov3 bo mo r..1 CO n -- n IA . • • • • • • • • 11••••• CA CO CO CO CO CO CO M CO CO MA CO .1.... tn CO CO
1 I I I I i I I 0 C- 4 sr ..-N CA ,- n ,- ml OD U7 sr CA CD sr os so uo es so OD CO rl n mo uo -- uo co m so
O ,-. M 0. r0 rqM S: s to - n rq C••1 r, ri •.-1 • • • . . • • ■ • • •
MI M MI M M M CO CO CSI M CO 9..... CO CO CO 9.... 4.'
1 IIIII I I I M Z
CO Ca ,-• -0 M r-, Cl =- -0 a.. ,- in M r.".; .:I- r'..!
r- N ND t5.:1 NCI tS) T- Cl ND so sr PO tn bl ▪ m -- es -o mouo m uo mo n m 127 CA 1./7 r.1 Iii
. . • . • • • • • ••••••• r...1 n CO CO MI MI CO MI MI CO CO T.... CO CO M CO M 07, i i 1 i i i i i E
P5
,- 'cl- Cr- CO r, „,r -0 r0 -0 ,10 OD ,10 ,:t CO sO CO •,-f n ro CO OD .4- 1-0 U7 U7 M N ,- sr ro ON N. 'AD f- • • . • • • • • ••• 1111•••
MI M M (5 (5) CO tar) M CJ'.! CA MI ml M M CO CO f-
I I I I I I 0 Imo.
sr uo .,.- r, -o ro so sr 1.2. N 0' V.'...:.! al ,q- t,-.4
-J CO ,-. -CI -0 (N CA CO ND CO (5) in in In ..— PO N. ...... • • • • • • • • • • a • • •
M M M CO CO M CO M M CO CO M CO M M M r” I I I I I I I I Cl
,-1
UP N ON .4" sr sr n m n -- n rl r, sr os co -o r.... oo cr. sr NO +- N. OD CA 'I' CO hP ,ID n
1..4 V".. ,... 9.... Cl CO M .1.. 0' CA N) n sr -- m c.4 ro ,---1 • • • . • 1111•1111011•11•• W
MI 15) MI CO CO M SI P5.1 (Sra M n tn tn CSI n St f- 1 1 1 1 I i I !:-
Ct
,- ,- ,- • M ...- -0 ..40 b71 CO LI Cr- Mt ..0 Cr, IN ,.. In 0, -40 -0 .q• sr m sr co If) Cl" N) -- n OD CA
:C 1.0 •-• sr -..ci r, CO MI ,- CO (N in N 'nr *-10 Cq LO /-1 . • . ••••••••••••• r"..
M CO M M CSa M M (5) CO CO I I I I „.,
CO cs CO r, co "C mo r.., CO CO 1.**, L7 OD CO CO (11
M U7 N U7 Si CO '<I U7 CO -.0 '0 ...- N MI Cl CO ,I
O C..1 CO .,- N) 1'0 -0 CO CO C.4 Cq *0 =- sr IN -- N) Cr • . . . • • • • . • 4 11 11 ••11 r-
tn in CO tn n ,- M CO CO n (5) Si n CO +- Si M I I I I I I I I I .,--1
5-
ND CO CA Ch. 1"0 CO tn U7 N sr ro r,. ,-- oo rq 9..... +"
9..... n 9..... n .1... N, sr cr, -4- N. M 0... to sr uo ,-
Lt. U7 CA CA U7 CO Ul +- tn NO ,- 0-. l'.".1 '4" N N) ts! ,- . . . . • . • . 111.11*M MVO
Ill
MI CO n CO .,.. n ..... CO CO n CO MA tn CO n CO 2: 1 I 1 1 1 1 i 1 CI
.--1 Lo 'r O. -o sr 117 U7 CO CA CO ND C' .4.- sr IN -, r, r, CA co U7 CA -0 +- if) CO ,1- CO in M 117 -CI
LU +... +- CA CO U7 OD ro -ci -r m sr co -- CS) ,--- at • . • . • . • •11111111M • n• LI
tn CO MI ,...... CO CO CO 5) tn MA .r..... MA MI tn in MA v=":
I 1 1 1 I 1 I I I Ti
• 3-4
ND .4' CA a) m sr OD sr rl cr. ...lo -- mo r- CA CA CO U7 Cr' VD if) Cr' CI" CO NO CA CA -,:r ro czi
C...1 CA Si M +- CA CA U7 ,- L7 Mi ,- in tn CA ,C1 CA U R 111•11 •••• • • • • • 11 CI
CO MI 9..... CO CO (Si CO CO CO MA CO ,- CO .rm CO ,- LI 1 i i i i i 1
5-
ID --r.:1 CA CA ,- on mi 0, 117 Sr ro tm -o a ,- ,r r+7 ro un -- o, ,r ,r ,r -- .,--,- 1:,..:.! .1-;,
Pci ,- sr -- CA n ,- tn tc in M n CO CO LI R • • • • 0 0 • • 0 Ill
CO MI tn n Si Cal Si in tS4 (5) ISI in Si CO CO Lt.- I I I I I I I I
• • Cl ,0 ,.... In r`:1 Cn .:-. ,- ro ,"4"0 1'0 ,- r, Co -I- -0 r, o- ro sr n m co -o sr ro sr ,-
..4r r-, CO ,-. In Cl L7 ,- M n ,-. ,- MI U7 CO (Si ,z • % 4 9 •• %a 1.119••
CA CA CO CO CO CO CSa M CO CO Si M CO CO M Si CI I 1 I I I i r--1
ED N
1--
r."J 4:r
+a-
w C 1 LIJ LI. CD 2 I-4 3= 2.: CD CV CD C1 CN
197
G1 to -71- -- m n cci m .rr ..4- IN 7- co C•.I -- -o .qr LD ..T. un In M 7- un so to NO 1.0 CP, 7" CO In ti!
cn ,- m o., m m c,.1 PO Y- r,. St CA OD M ..-- NO 7" ,...1 • •••• • ••••••• .1•• 1.11 M SI M M M MI M M M M M M M SI SI ,- I 1 I I I I 1 I ,:-.1
M
r..I tri cr, si -.- =4" ,-- ,-- Y-- CO 0, ,- CA ON CSI LID f- PO CA ro co CO -Cl 0- os No r, N. Cl CO 7- -- m
• . . • • . . ••••••••• 0
M M M MI M M Si M SI CO CO M CO CO ..-- -- -.:, I 1 I I I I I I I 0
CO r, CA Cl N 4- 4 CO CO In ,-- CA NO In CO N. :7i..,
CA "CI NO PO M CIN CA N '4) ro CN 7.- cs. -5.. m r, Ct. I
or, No m to -- un ..- un n IN -- ,- c4 n OD SA !"..: • . . • • • .• • • . • • . • •-t
CO M SI M CO CO CO M CO CO CO M M M CO M 4: I I I I I I I 4=
• r-i
N. CA In SI PO -T to -Cl -0 r, os rq r„-- (N -.. -- IN os -- r, -- ro No N. -,..r c...1 n PO SI PO ! e.:1
cn n m n NO r.) .1” Cl -- -tr so un -- op n r..1 n -1-1 . • • • • . • • • • 11•• ••
Ill
CO CO n CO M M SI CO M CO M CO CO CO M SI — I I 1 I I I Cl
N. N 0
In CA 'r/- ro ,o e.:1 ON N. OD M NO SI CO ,- a SA ,- CO OD CO OP. 0, N NO In CA PO PO 0, CO Cr`
CD PO M ON CA rl SI In CA 7" ,-. SI CA =- CA CO .,-- r- • ••••• •• • • •••• CI
CO M M M CO CO CO M CO CO CO ..-- M CO M ,-- 4- 1111111 I I I
In N VI CA NO N CO ,- =-. ro No uo r, co un IT. •,...f ro vo ro !I-) CA CA N CA n ..-- CA CA CO M NO In Z.:: M SI ,-- CO ND in ro -- if' ND N CO If) if) CA 3 • . • . . • . • . • • . • • •M
St M M M CO M M CO CO CO CO CO n M CO CO E I 1 I I I I 1
,-,
,- CA CO .cr In CA OP. *- CO PO CA r. c..! op, co co al
In ,- 4- Ni 7- No -- ro n co -Cl •q" Cr. Cr, M NO .-...1
1: ,-- ,-SI In ,-- PO CA 7- M N .14- -- ..q- CO ro CO cr., • • • • ON •••• ••II• •• r-
M M M M CO CO M M SI CO M M Si CO CO CO M I I I 1 I 1 I •.•I
!-- NO CO ,-. ,-. .q.- SI N M 0, N NO NO In N (N CO
CA OD CA CO CO CA CA CA CO PO CA n ,- Cl-.
-J M .-- NO NO CA ,- CO CA CO CO .q-. If) -i- (N -.4- CO r-. • • • • . . • ••••••••• SI
M M M M M M M M T'. MI M M M M M M El. 1 1 1 1 1 1 1 I EL
:1 ON NO OD .nr r, 7- ro No un -- un PO In NO In 00
CA N .......
,- OD CA CA CO "4' SA In ON OD ,cr up oo r...
1+4 (Si +- CA 04 CA Si =-.. CO CA if) CO CA =-. CO ,- ,-. .- • . • • • • • • • • • • • • • • Cl
m m CO CO CO m (Si M CO MI St M M (Si (Si Si •,...1 I I I I I I I -1-,
M CO In LID In In SI In CA -q- ro os PO NO N t'=! .--1
CO Si CP. PI ,-- 0- MI IN N. so r, Cr, N. m al
:c In St In NO N. M PO ,- St CA PO N 0.1 In CA LO r- • • • • • • • . • • • • • • • • r••••
M CO CO CO CO CO ,-- CO CO 'Si CO (Si Si CO M CO 0
I I I 1 I I ! LI
Cr- CA If) CO N 0-. ,-- N. CO N rq ro 0. 7- c...) CO .7,
co rq ro ,o ',,z- !In IN IN ol op -,.r to so Kt M V, r:
CD ..- M .!-- ro r) ..q. M CO CA PO If) =-- v ,-- ol ro m • • • •••1 •••• ••• 11••
M CO M M M Y-- M M CO CO (Si CO CO SI ,- CO .--. I I I I I I I I I GI
M CO CA ON 7- CO ,- CO ON N CO, CA SI CA PI St Cr,
vl r. -- Lo CO N. N. r....1 CA mt SI NO CA =.- N. n r. It. 4gt. --
IN No ro v os CA NO ,- .0 M in ..40 M. ;.., ro . • • • • . • • • . . • • ••• •,..1
MI M M M ,.... M M M M M M n M M n M r-
I I I I I I I i-,
if) CN OD NO =- N -O CO N ON CO PO ..zr Cr- .,11. ND et LLI .-- M SI
. • .,- OD 1• OD CA No 4- r, N ,.CO SI C SI tl • • . • ■ •••••••• 0
M SI M .!-- SiI CO M M M M M CI . SI M SI .---o
I I I I I I I
P.7..I CO NO CO PO r, .1- n, Cr- Np wp ro LID PO n CO 0
OD ,- P., ON NO N. ro ,- Ni LI (...) CA SI ,- SI CA ,- NO CA N SI ,- ..-- irl'T! rq
CS'! 71 • • • • ••••••••• a;
M M ,- SI M M SI M SI n MI ,- n SI M ,- ...-1
I I I I I I r- 03
'q- N ND CO CA CA ro --
-- n os r, os cr, ,- so co 'so ro >
so rq CO CA N-.1 =- NO 7" NO --.1- c,
04 St NO tSI SI =- Mt n ....- M M M M M M n St U • • • • It • ••••••••••
M 'Si M M M M CO M n( M CO CO MI CO SO r- I I I I I I I o 7- so ro No n n IN 7. -- un c...! n o., r, -- ClL
ON T- ID 1'0 h0 0, n CA NO ,- CA CO rl 0, rl In M
<E NO CO CA ,- '4" ,-- In M (SI ,- SI CA CO .t.t. -- LI_ • . • • •••••• ••• • • •
M 'Si M M M M M M CO CO M M St M Si St ..
I t I I I I Cr,
4:c
S1
..tr 04 U W IA- CO D.: 0-.1 ....I 3 CD CZ CI CS CS h-
198
IN '.0 U7 Os +4' ,- +A" •-- Mt KA Cr ,- sD Tr cl, Tr in rn O. — ,o cl, n m Tr co .4' +I" ,- rn G m m MI M 0.1 r7 C-.1 ,0 M 4.- 0, M .1.- ,o r, • • ••• •11•••• ■ •11
CA to KA KA KA KA M MI KA t A CA CI KA ,- I I I I I I I 1 I
Os KA sD +- -o tr. sO ' Os -0 o •-• N. CO N) r1 '54" ,- N. 03
KA ,..- rl N •A :t CA Os CO Os ,-.- 117 t7 U7
Cl CA N N) rA +4" U7 CA T4 -.CI ,-. O. • • • s • 117•••••• .;
CA KA n KA M KA CA KA KA KA KA KA ,-.. ,- 0 I I I I I I I I I ..*-..
CP CA CO Ul CO N. r, 117 N) ,-- rq 1 un ,- ,- Tr —
c., rn -ci ..-- I.17 Os CA sO n M") CA P7 Tr -0 a- if) ,ci .-1
a ,c1 m rn m U.1 +- U7 KA CA +- ,-. CA KA sO KA ,-. 011 • • • •I• ••• ••• • •• • .I.:'
M M M t M n n n n n M KA KA KA KA n 0 I 11111 I
„.... pp, Ig t,,„ IN 0... Os 1)7 PI +- ON sO sD 'ti" U7 CA IN c.-
+- I-7 OD sD (7, CA ro OD N. U7 KA KA 0 G KA QA QA sD ND +4" CA r, .4" sD 67 ,-. ,- KA rn CM (.1- • • a • • • • • • • • • • • • •
M M M MI M M M M M M M4 M T- MI t M I I I I I 1 • 2-1
Tr Tr Tr —
N. Tr co Tr r, CA sO N. r, cl, CO la7 ..'!-;
C• ,-.. Ma Os ro cq si Lin rn rn ,--m cl c,i CA CO ND E • • • • • • • • . . . • . •
Ma QA KA KA QA KA KA KA KA n n ..-- n n n ...- --- 1111111 I 1 I Ci
....1 r', ,- -- Cr, rn in, In Tr -ci m Lin rn c..! rzi r, or!
r..1 — co m e ...2 . .%
— in
z m, m -- OD -0 M7 tgi b7 sO M .D+--ti CA (6 • • • • . • • • . • . . • • • 4.i
M n n n n iS CA KA MI t +- CA KA QA QA KA r.... I I I I i I I -I-,
NI t7 rn r, m h7 Tr Tr co Tr Lin ,o c..1 ,o cq -. r„ — ro rs! -ci r, ro cl, r.-.4 co — c, rn ai
3= M -- MI 471 ,-. rl CA NI KA CO U7 +- sO KA NI KA ctL
4 ••• ••• • • ••••■ -L.
KA KA KA Ma M4 QA KA Cu KA KA KA CA KA CA KA iS, :3 I I I I I I ---
r., sD KA IN m Tr 75t cq c-.1 co un ,- rz.4 mi ..o .c. un rn rn ,c1 co CO sD ,- CA n CA r, — cs, VZ! 0
--I KA .r. sD -.0 CA ,-- KA CA Os KA U7 NI U7 c-4 Tr r, •r-I • • • • a • • • • • I • • • • 4..,
n M M M MI M M n not ti4 ml n n n KA KA III I I I I I I I ,....1
ci
rn Tr -- c..1 Lin rn o, co IN .-ci rn rn — rn rsi rn r...
OD Tr ci-.. Tr mi -- cq cr• rn iS, m O. m m co Tr !....
1-4 cl — IN rn ,- m — cl rq Tr cq Tr ci CA ND '4" CI
••• ■ ••41•11•• •••• • LI KA KA QA QA QA CA iS, ,- KA KA KA KA KA KA KA MI
I I I i I I I I 7-1 r
-O sD CO U7 U7 Os Tr -o C. N) — r, co m — un ir, R- cl, co r, rn -- C) r„ -CI un CS! r, -o sO OD U7
T Tr m +T..' sCI N. KA Tr — Km C.4 M7 r„ rn in 1:-..1 un ..-.
. . • . • • . . • • • . . • • . al
iS, m m iS m mi -- m m m KA QA KA n iS, CS r..1 1 I 1 1 1 1 I cr:,
r
Tr un Tr 1.7 KA Tr NI CA U7 .r. CO P7 CO ND un rte) r
U7 CA Tr un un o, ro c..1 un O. ci, in co un un Tr •r•4
CO +..- to +- rn rn m m iS Cl Tr r, — ,o — Tr +4" 5- • . • • ••• ■ • •• •• ••• H.A M IS, M m m tea mn iS, KA KA KA KA KA KA .— KA
I I 1111 I I I I 5- CM
CA +- KA sO KA rm r, r, ND 1.":1 4:1 Lri ,0 sO sD :3
In Cr- 0, tt-- ,0 -,.- ro -o N M CA KA CO Os CO IN 0
4- 'V .- +- -.0 CP, N. rl +- N) CA Os Tr Tr -O Tr Is! ,H • • I. • • • a a • • • • • • a •
n n KA KA +- KA +- KA Cal CA KA KA KA KA KA KA I I I I I I I al
LI
Cl "-J .”- r- ON. cl r, N. Tr P7 ON 0:I l'''') M "CI ,- r
CA ..- ,0 ,CI Tr n OD Ch N. ,- CI '54" sD CO Cl"! IT.
14.1 KA U7 KA sO CA Tr r, un — • ,-1
M4 KA Ma .a.-- ,- iS, ,..- C4 KA CA ,-- UK".:i t:21 P: ...K.; :;).1 515 • . • . • ••• 4 •• 5.4
I I I I i I I I
La .1... cl, r, 00 U7 CA ,- Tr un IN CA N. Ca, N CA CO +- NI Tr co cl cq if Tr rn •— un iS c-A 1"."
1-1
C-1 ,-. M T- T- 01 C.4 ..10 ND sz.1 m..! ,..- KA C.4 CO Pr' r- • • • • • • • • • • • • • • • C)
KA CI +- KA KA KA KA QA KA KA C4 ,-. KA KA KA ,- -1.-, I I I i I I U
513
c..i CA Tr n Lin un ci- rn KA KA Tr rn re) la, un ro u_ c..! r, -- o, co CA 0, KA ,CI PC1 ND I'D
134 KA Tr Ca m -- qm t.T.4 --- n m rm m n qm m n .. • 4•1411••• ■■ •••••• KA
M KA IS, KA KA IS, KA KA KA IS, QA IS, KA 15, KA IS, CA I I I I I 1 I ex
+4" 1.11 Os In Os Os -,--. N. +4" M7 CO ND IS, -.0 .4:1 CO -CI ,- CO +-- IN
CI
CO ,- If) -CI Tr 1:,.:4 Tr ,- N) ,ci ,-, <lc cl, m ,o cq un N) m IS, m c..! MI U7 KA M .f.)
• •••••••••••••• a 115 KA KA KA KA KA KA KA KA CA KA KA KA IS, IS, KA h-
1 I 1 I - I I
C') N-J 04 C.) UJ li CD MC 1-4 J riC 2: CD CD CD CD CN
199
REFERENCES
200
Allport, G. W. & Odbert, H. S. (1936). Trait-names, a Psycho-lexical Study. Psychological Monographs 47, Whole No. 211.
Alston, W. P. (1976). Traits, Consistency and Conceputal Alternatives for Personality Theory. Chapter 4 in Harrg, R. (ed). Personality. Oxford: Basil Blackwell
Argyle, M. (1976). Personality and Social Behaviour. Chapter 6 in Harre", R. (ed). Personality. Oxford: Basil Blackwell.
Barton, K. & Cattell, R.B. (1972). Personality Characteristics of Female Psychology, Science and Art Majors. Psychological Reports 31, 807-813.
Becker, W. C. (1961). A Comparison of the Factor Structure and Other Properties of the 16PF and the Guilford-Martin Personality Inventories. Educational & Psychological Measurement 21, 393-404.
Bern, D.J. & Allen, A. (1974). On Predicting Some of the People Some of the Time: The Search for Cross-situational Consistencies in Behaviour. Psychological Review 81, 506-520.
Bock, R. D. & Petersen, A. C. (1975). A Multivariate Correction for Attenuation. Biometrika 62, 673-678.
Borgatta, E. F. (1962). The Coincidence of Subtests in Four Personality Inventories. Journal of Social Psychology 56, 227-244.
Browne, M. W. (1972a). Orthogonal Rotation to a Partially Specified Target. British Journal of Mathematical and Statistical Psychology 25, 115-120.
Browne, M. W. (1972b). Oblique Rotation to a Partially Specified Target. British Journal of Mathematical and Statistical Psychology 25, 207-212.
Burdsal, C.A. & Vaughn, D. S. (1974). A Contrast of the Personality Structure of College Students found in the Questionnaire Medium by Items as Compared to Parcels. Journal of Genetic Psychology 125, 219-224.
Cattell, R. B. (1933). Temperament Tests: II. Tests. British Journal of Psychology 23, 308-329.
Cattell, R. B. (1944). Parallel Proportional Profiles and Other Principles for Determining the Choice of Factors by Rotation. Psychometrika 9, 267-283.
Cattell, R. B. (1946). The Description and Measurement of Personality. New York: World.
201
Cattell, R. B. (1950). The Main Questionnaire Factors in Questionnaire Self-Estimate Material. Journal of Social Psychology 31, 3-38.
Cattell, R. B. (1952). The Three Basic Factor Analytic Research Designs - Their Interrelations and Derivatives. Psychological Bulletin 49, 499-520.
Cattell, R. B. (1957). Personality and Motivation Structure and Measurement. New York: World.
Cattell, R. B. (1965). The Scientific Analysis of Personality. London: Penguin.
Cattell, R. B. (1966a). Handbook of Multivariate Experimental Psychology. Chicago: Rand McNally.
Cattell, R. B. (1966b). The Scree Test for the Number of Factors. Multivariate Behavioural Research 1, 140-161.
Cattell, R. B. (1972). The 16PF and Basic Personality Structure: a Reply to Eysenck. Journal of Behavioural Science 1, 169-187.
Cattell, R. B. (1973). Personality and Mood by Questionnaire. San Fransisco: Jossey-Bass.
Cattell, R. B. , Eber, H. W. , & Delhees, K. H. (1973). A Large Sample Cross Validation of the Personality Trait Structure of the 16PF with Some Clinical Implications. Multivariate Behavioural Research, Special Issue, 107-132.
Cattell, R. B. , Eber, H. W. , & Tatsuoka, M. M. (1970). Handbook for the Sixteen Personality Factor Questionnaire. Champaign, Illinois: IPAT.
Cattell, R. B. & Sullivan, W. (1962). The Scientific Nature of Factors: A Demonstration by Cups of Coffee. Behavioural Science, 7, 184-193.
Cattell, R. B. , & Tsujioka, B. (1964). The Importance of Factor Trueness and Validity versus Homogeneity and Orthogonality, in Test Scales. Educational and Psychological Measurement 24, 3-30
Cattell, R. B. , & Warburton, F. W. (1967). Objective Personality and Motivation Tests: A Theoretical Introduction and Practical Compendium. Urbana: University of Illinois Press.
Comrey, A. L. (1970). Manual for Comrey Personality Scales. San Diego: Educational and Industrial Testing Service.
Cronbach, L. J. (1951). Coefficient Alpha and the Internal Structure of Tests. Psychometrika 16, 297-334.
202
Cronbach, L. J. (1975). Beyond the Two Disciplines of Scientific Psychology. American Psychologist 30, 116-127.
Edwards, A. L. (1957). The Social Desirability Variable in Personality Assessment and Research. New York: Dryden.
Eysenck, H. J. (1947). Dimensions of Personality. London: Routledge and Kegan Paul.
Eysenck, H. J. (1966). Personality and Experimental Psychology. Bulletin of the British Psychological Society 19, 1-28.
Eysenck, H. J. (1972). Primaries or Second-order Factors: a Critical Consideration of Cattell's 16PF Battery. British Journal of Social and Clinical Psychology 11, 265-269.
Eysenck, S. B. G. , & Eysenck, H. J. (1968). The Measurement of Psychoticism. British Journal of Social and Clinical Psychology 7, 286-294.
Eysenck, H. J. , & Eysenck, S. B. G. (1969). Personality Structure and Measurement. London: Routledge and Kegal Paul.
Fahrenberg, J. (1977). Physiological Concepts in Personality Research, Chapter 25 in Cattell, R. B. , & Dreger, R. M. (eds. Handbook of Modern Personality Theory. New York: Halsted Press.
Frenkel-Brunswik, E. (1942). Motivation and Behaviour. Genetic Psychology Monographs 26, 121-265.
Fruchter, B. (1954). Introduction to Factor Analysis. Princeton, N. J.: Van Nostrand.
Gray, J. (1973). Causal Theories of Personality and How to Test Them. In Royce, J. R. (ed). Multivariate Analysis and Psychological Theory. London: Academic Press.
Greif, S. (1970). Untersuchungen zur Deutschen tbersetzung des 16PF Fragebogens. Psychologische Beitrage 12, 186-213.
Guilford, J. P. (1956). The Structure of Intellect. Pscyhological Bulletin, 53, 267-293.
Guilford, J. P. (1975). Factors and Factors of Personality. Psychological Bulletin, 82, 802-814.
Guilford, J. P. , & Guilford, R. B. (1934). An Analysis of the Factors in a Typical Test of Introversion-extraversion. Journal of Abnormal and Social Psychology, 28, 377-399.
Guttman, L. (1966). Order Analysis of Correlation Matrices. Chapter 14 in Cattell, R. B. (ed.) Handbook of Multivariate Experimental Psychology. Chicago: Rand McNally.
203
Guttman, L. (1954). A New Approach to Factor Analysis: The Radex, in Lazarsfeld, P. F. (ed.) Mathematical Thinking in the Social Sciences, Glencoe, Free Press.
Hakstian, A. R. (1972). Optimising the Resolution Between Salient and Non-salient Factor Pattern Coefficients. British Journal of Mathematical and Statistical Psychology, 25, 229-245.
Harre", R. (ed.) (1976). Personality. Oxford: Basil Blackwell.
Harre-, R. , & Secord, P. F. (1972). The Explanation of Social Behaviour. Oxford: Basil Blackwell.
Howarth, E. (1976). Were Cattell's "Personality Sphere" Factors Correctly Identified in the First Instance? British Journal of Psychology, 67, 213-230.
Howarth, E. , & Browne, J.A. (1971a). An Item-factor Analysis of the 16PF Personality Questionnaire. Personality: An International Journal, 2, 117-139.
Howarth, E. , & Browne, J.A. (1971b). Investigation of Personality Factors in a Canadian Context. I. Marker Structure in Personality Questionnaire Items. Canadian Journal of Behavioural Science, 3, 161-173.
Howarth, E. , Browne, J.A. , & Marceau, R. (1971). An Item Analysis of Cattell's 16PF. Canadian Journal of Behavioural Science, 4, 85-90.
JOreskog, K. G. (1970). A General Method of Analysis of Covariance Structures. Biometrika, 57, 239-251.
JOreskog, K. G. (1971). Simultaneous Factor Analysis in Several Populations. Psychometrika, 36, 409-426.
Kaiser, H. F. (1976). Review of Lawley and Maxwell's "Factor Analysis as a Statistical Method". Educational and Psychological Measurement, 36, 586-589.
Levonian, E. (1961). A Statistical Analysis of the 16 Personality Factor Questionnaire. Education and Psychological Measurement, 21, 589-596.
Levy, P. (1973). On the Relation Between Test Theory and Psychology. Chapter 1 in Kline, P. (ed.) New Approaches in Psychological Measurement. London: Wiley.
Lord, F. M. , & Novick, M. R. (1968). Statistical Theories of Mental Test Scores. Reading, Mass.: Addison-Wesley.
McGaw, B. , & JOreskog, K. G. (1971). Factorial Invariance of Ability Measures in Groups Differing in Intelligence and Socio-economic Status. British Journal of Mathematical and Statistical Psychology, 24, 154-168.
204
Mischel, W. (1968). Personality and Assessment. New York: Wiley.
Mischel, W. (1973). Towards a Cognitive Social Learning Reconceptualization of Personality. Psychological Review, 80, 252-283.
Nunnally, J. (1967). Psychometric Theory. New York: McGraw Hill.
Pawlik, K. (1973). Right Answers to the Wrong Questions? A Re-examination of Factor Analytic Personality Research and its Contribution to Personality Theory, in Royce, J. R. (ed. ) Multivariate Analysis and Psychological Theory. London: Academic Press.
Royce, J. R. (1973). The Conceptual Framework for a Multi-factor Theory of Individuality, in Royce, J. R. (ed. ) Multivariate Analysis and Psychological Theory. London: Academic Press.
Saville, P. (1973). The Standardisation of an Adult Personality Inventory on the British Population. Bulletin of the British Psychological Society, 26, 25-29.
Saville, P. , & Blinkhorn, S. (1976). Undergraduate Personality by Factored Scales. Windsor: NFER.
Sells, S. B. , Demaree, R. G., & Will, D. P. (1970). Dimensions of Personality: I Conjoint Factor Structure of Cattell and Guilford Trait Markers. Multivariate Behavioural Research, 5, 391-422.
Sells, S. B. , Demaree, R. G. , & Will, D. P. (1971). Dimensions of Personality: II Separate Factor Structures in Guilford and Cattell Trait Markers. Multivariate Behavioural Research, 6, 135-185.
Timm, U. (1968). Reliabilitat und Faktorenstruktur von Cattells 16PF test bei Einer Deutschen Stichprobe. Zeitschrift fur Experimentelle und Angewante_ Psychologie,15, 354-373.
Tucker, L. R. , & Lewis, C. (1971). A Reliability Coefficient for Maximum Likelihood Factor Analysis. Topics in Factor Analysis II. Technical Report, Office of Naval Research.
University Grants Committee (1975). Statistics in Education: Universities Vol. 6. London: HMSO.
Vagg, P. R. , & Hammond, S. B. (1976). The Number and Kind of Invariant Personality (Q) Factors: a Partial Replication of Eysenck and Eysenck. British Journal of Social and Clinical Psychology, 15, 121-129.
205
Van Thillo, M. , & JOreskog, K. G. (1970). SIFASP - A General Computer Program for Simultaneous Factor Analysis in Several Populations. Res. Bull. 70-62. Princeton, N. J.: Educational Testing Service.
cc 0
ta
▪
.
HIGH SCORE UZSCRIPTION
A OUTGOING, WARMHEARTED, EASY-GOING, PARTICIPATING (Affectothymia, iormerly cyclothymia)
C
E
F
G
H
I
THE PRIMARY FACTOR SCALES OF THE 16PF
LOW SCORE' DESCRifri ION
RESERVED, DETACHED, CRITICAL, ALOOF
(Sizothymio)
LESS INTELLIGENT, CONCRETE- THINKING
(Lower schoicstic me-Eal capacity)
AFFECTED BY FEELINGS, EMOTIONAL-LY LESS STABLE, EAE1LY UPSET
(Lower ego strength)
HUMBLE, MILD, ACCOMMODATING, CONFORMING
(Suornissiveness)
SOBER, PRUDENT, SERIOUS, TACITURN (Deskirgenc.y)
EXPEDIENT, DISREGARDS RULES, FEELS FEW 03LIGATIONS
(Weaker superego strength) I
SHY, RESTRAINED, TIMID, THREAT-SENSITIVE
(Threctia)
TOUGH-MINDED, SELF-RELIANT, • REALISTIC, NO-NONSENSE
(Horria)
TRUSTING, ADAPTABLE, FREE OF JEALOUSY, EASY TO GET ALONG WITH (Alakia)
PRACTICAL, CAREFUL, CONVENTION-AL, REGULATED BY EXTERNAL REALITIES, PROPER (Praxemio)
FORTHRIGHT, NATURA_, ARTLESS. UNPRETENTIOUS
(Arrlessness)
SELF-ASSURED, CONFIDENT, SERENE
(Untroubled adequacy)
CONSERVATIVE, RESPECTING ESTAB- LISHED IDEAS, TOLERANT OF TRADI- TIONAL DIFFICULTIES (Conservatism)
GROUP-DEPENDENT, A "JOINER" AND SOUND FOLLONER
(Group adherence)
UNDISCIPLINED SELF-CONFLICT, FOL-LOWS OWN URGES, CARELESS OF PROTOCOL (Low integrction1
RELAXED, TRANQUIL, UNFRUSTRA. rED
((-ow ergic tersior.)
MORE INTELLIGENT, ABSTRACT- B THINIKING, BR!!".34T
(-IicEer rnento: capacity)
EMOTIONALLY STABLE, FACES REALITY, CALM MATURE (Higher ego strength)
ASSERT:VE, AGGRESSIVE, STUBBORN. COMPETITIVE (Dominance)
HAPPY-GO-LUCKY, IMPULSIVELY LIVELY, GAY, ENTHUSIASTIC (Surgency)
CONSCIENTIOUS, PERSEVERING, STAID, MORALISTIC (Stronger superego strength)
VENTURESOME, SOCIALLY BOLD, UNINHIBITED, SPONTANEOUS (Formio)
TENDER-MINDED, CLINGING, OVER-PROTECTED, SENSITIVE ;Prernsio)
SUSPICIOUS, SELF-OPINIONATED, L HARD TO FOOL
(Protension)
M
N
0
Qi
Q2
IMAGINATIVE, WRAPPED UP IN INNER _.P.GENCIES, CAE:EL ESS OF PRACTICAL (Autia) ■.'ATTEIRS, BOHEMIAN
SHREWD, CALCULATING, WORLDLY, PENETRATING (Shrewdness)
APPREHENSIVE, SELF-REPROACHING, WORRYING, TRCALED (Guilt proneness)
EXPERIMENTING, LIBERAI LNALYTICAL, FREE-THINKING
set)
SELF-SUFFICIENT, PREFERS OWN DECISICI4S, RESDURCEELIL (Self-sufiiciency)
CONTROLLED, SOCIALLY PRECISE, OLLOr. ■ NO SELF-IMAGE
ah se;(-conceE.• contra')
TENSE, 7RUSTRETED, DRIVEN, -_VERWPOLIGHT ;I-kgh ery,c tensi:n)