evolution of the white mountain magnia series · 2007-08-29 · evolution of the white mountain...
TRANSCRIPT
EVOLUTION OF THE WHITE MOUNTAINMAGNIA SERIES
RaNoor.pn W. CnapuaN, Vassar College
CnenrBs R. wrr";;: , Cambri.d,ge, Mass.
PART I. DATA
Pnosr.BM
In recent years, a number of intensive field and laboratorystudies of the rocks of the White Mountain district in New Hamp-shire have been carried out. One result of these investigations isto show that there exists in this area a group of rocks with markedalkaline affinities (3)* to which the name White Mountain magmaseries has been applied (5, p.56). The various rock types of thisgroup form a definite series, and wherever found in the area theypossess the same relative ages. Such a sequence is of greatest im-portance to petrology and necessitates an explanation. Accord-ingly, the writers have undertaken a study of this problem, theresults of which are presented in this paper. ft is not pretendedthat this work is complete or that the problem has been entirelysolved. Certain definite conclusions have been reached, however,and it is hoped that these may lead to a more complete under-standing of the evolution of the White Mountain magma series.
The writers are especially indebted to Professor Marland Billingsof Harvard University for his valuable assistance in preparing thispaper. Several of the major ideas presented here were first sug-gested by Professor Billings, and these have led to a clearer under-standing of many of the intricate problems encountered in thecourse of the work. The writers also wish to thank Professor EsperS. Larsen, Jr., and Professor R. A. Daly for their many helpfulsuggestions and criticisms. The new mineral analyses quoted herewere made by F. A. Gonyer of Harvard University.
fn the course of these investigations the writers have limitedthemselves entirely to those areas with which they are personallyfamiliar, either from a field study or from a study of rock collec-tions. These areas include the North Conway quadrangle, theOssipee Mountains, the Belknap Mountains, Pleasant Mountainin Maine, the Franconia quadrangle, the Percy quadrangle, Tri-pyramid Mountain, and Red Hill. The rock types of each of these
* Numbers in parentheses refer to bibliography at end of paper
Tesro 1. Dera oN rnn PruroNrc Rocrs or rne Wnrrn Mounrl.rn Macue Sanrrs
NAME GABBRO N O R I T E D I O R I T E QUARTZ OIOR ITE M O N Z O D I O R I T E MONZON I TE NEPHELITE SYENITE ANALCITE SYENITE
D I S T R I B U T I O N
T R I P Y R A M I D M T N
BELKNAP MTNS:
T R I P Y R A M I D M T N . NORTH CONWAY QUAD.
P L E A S A N T M T N .
F R A N C O N I A R E G I O N
F R A N C O N I A R E G I O N B E L X N A P M T N S T R I P Y R A M I O M T N ,
PLEASANT MTN.
R E O A I L L PLEASANI MTN
S I Z E O F B O D I E S FEW SQ.FT. TO O.25 SQ. MI , NOT XNOWN F E w S Q . F T . T O O . 2 S q . M t . t . 5 s Q . M t . o . 2 5 T O 2 S O . M r .R t N 6 - D | X E - 2 . 5 M t .
OI AMETER
SMALL MSS- O,OI SQ. MI ,
NOT KNOWN o . o 3 s q . M r .
SHAPE OF BODIESI R R E G U L A R NOT XNOWN SMALL STOCre AND
D I K E SR O U G H L Y C I R C U L A R I N C O M P L E T E R I N 6 - D I K E S I R R E G U L A R A N D R I N 6 -
D I ( ENOT XNOWN SMALL PERIPBERAL
BAND ON MAIN STOCK
TOTAL AREA ANDPER CENT OFM A G M A S E R I E S
o . r o s Q . M t .
o . o z Y "YERY SMALL
o . 2 5 S Q . M t .
o,oss %
r . 5 s Q . M l .
o . 3 3 7 "
4 . O S Q . M l
o.s1 7"2 . 6 S q - M lg . s 1 Y "
V E R Y S M A L Lo - o 3 s Q . M t .
o . o o 7 7 "
T E X T U R EHYPIDIOMORPHICTGRANULARB E L X N A P S - O I A E A S I C
H Y P t O I O M O R P H r C ,
I N E Q U I G R A N U L A R
H Y P t D t O M O R P H T C ,
I N EqU I GRANU LAR
H Y P t D T O M O R P H t C I
I N E Q U I G R A N U L A R
HYPIDIOMORPHIC. EOUI6RAN-
ULAR TO SUBPORPHYRITIC
H Y P T D t O M O R P H T C T
I N E Q U I G R A N U L A R
HYPI O IOMORPH IC ,
G R A N U L A R
ALLOTRIOMORPH I C ,
I N E Q U I G R A N U L A R
D I A M E T E RO F G R A I N S
5 - r o M M 3 M M e - 3 M M . M R K M I N E M L S - O , 5 M MR E S T 2 - S M M ,
5 M M 2 - 5 M M ,A V G . 2 - 3 M M ,
o . 5 - 5 M M
COLOR DARK GRAY TO BLACX DARK MOTTLED OARK MOTTLED LIGHT 6RAY MOTTLED L I G H T G R A Y 6RAY LIGHT oRAY L I G H T G R A Y
M O D E
T R I P Y . S E L .
L A B . 6 3 . 0 4 9 , O
H B D _ 2 Z , O
A U G . 1 4 . O I A . O
f L , 6 . 7 5 . O
B r . 5 . O
o R _ . 5 -
a c c . o . a 6 . 0
r o o . o t o o . o
A C C . - P Y , T I - A P , C H L . H Y
L A B . 5 5 . S
o R . 3 . O
H Y . t O . O
D r _ 5 . 6
B r . 6 . 0
A P _ 4 . 4
M T . I t . O
A C C . O . 5
t o o . o
P . M , F .
A N D . 6 2 . 0 6 5 . 0
a f . 1 7 . 5 r 2 . o
a u 6 . r 5 . s 8 . o
A P . I . 5 T R ,
M T . 4 . O T R
T I . T R . T R
c H L . r 4 . O
A C C . j t _ O
r o o . 5 t o o o
A C C . - H B D . P Y . q T Z
A N D . 6 3 . 0
o R . 2 . O
B t t 5 . o
H A D . 9 . Oq T z . t o . o
A C C . r . O
m
AND. 72 .O
o L t G . 6 t . o 6 r . o
o R . z t . o 2 0 . o r 2 . o
A U G . 7 . O 5 . O
a t . 4 . o r o . o 3 . o
H B D . 3 . O 2 . O 7 . O
Q T Z . 2 . O 4 . O
A C C 2 - O 2 . O 2 . O
J6oJ rom- rm-
A C C . - D I A P M T . Z R .C H L ,
T R I P Y . P , M ,
A N O . 4 0 . O 4 7 . O
M P . - 4 0 . O
o R . 4 0 . o
H B D . t 2 . O
D t . 3 . O
A U G . I , 5
B t . a - o
A C C . 5 - O 3 . O
r o o . o 9 9 . 5
A C C . - A P I L . M T T I
o R . 3 6 . 0
A B . 3 7 . O
N E P H , I 4 . O
s o o . 5 . o
H B o . 7 . O
A C C . | - O
m
a c c . - A P . z R . D t . B r .A E G . M T -CANCRINITE. WOHLERITE
M P . A O . 5 8 2 . O a 4 . O
o L r 6 . 3 . 5 2 . s
A N A L . 6 . 5 6 . s 7 . 5
H B O . 7 . 5 5 . 5 4 . 5
T t . 4 . 5 2 . 5 r . 5
M T . 0 6 0 . 5 r . 5
A F , T R , O . 7 T R .
Z R . T R . T R . T R .
M 6 0 2 m
C H E M I C A LcoMPosrTroN T R I P Y . S E L M O O E I M O D E I
si o2 4 8 . O 4 4 3 - 9 4 4 6 6 7 5 6 6 6 6 0 . 9 6
T i o ? 2 . 5 6 4 . t 3 . 5 5 l o 6 0
A l z 0 3 2 0 . r o 1 6 . t 7 t 6 . a 6 r 9 . 0 3 2 t . 3 6 l 9 2 0
Fea Ot 2 . 3 2 3 . 9 6 9 A t . 0 5
FsO 6 . 1 3 t o . o 6 6 . 3 7 4 . a a 2 . 0 4
M n O r 0 9 T R . t a
M9o 4 . 6 a 5 . 0 5 4 . 6 2 a 6 . 5 5
r . s z 9 . 5 9 s . 6 3 9 5
Na2O 3 . O r 2 . 9 3 3 . 6 5 4 . 5 2 4 . 6 6 1 . O 7
K a O 7 g . 5 1 t . 2 6 6 . O 6 5 , 5 4
H z o t 4 2 4 2 t 2 6 0 A 5
H a o - t 3 o 2 o
ZrO2 N D N D N D N O o 2 o 7
6 9 N D o 4
c l N D C O z . 0 9 N D N D B a O o 6
S oa N D N D o 7 . o 6 N D
TOTALS r 0 0 . 0 6 9 9 . 6 7 I 9 . 5 o 0 . 3 7 t o o . o 5 9 9 . 4 0
ABBREV I AT IONS
A g - A L B I T EACC - ACCESSORTESA E G - A E 6 I R I N EA L L - A L L A N I T EA N A L - A N A L C I T E
A N D
ASTA U G
A N D E S I N E C H L - C H L O R I T EA P A T I T E D I - D I O P S I D EA S T R O P H Y L L I T E F A Y - F A Y A L I T EA U G I T E F L - F L U O R I T EB I O T I T E H A S T - H A S T I N G S I T E
H B DH E DH Y
L A A
H O R N B L E N D EHEDEN BERG I TEH Y P E R S T A E N EI L M E N I T EL A g R A D O R I T E
N E P HO L
M I C R O P E R T H I T E . O L I GM A G N E T I T E O R
MAY INCLUDE IL PYN E P H E L I N E P Y R O XO L I V I N E Q T Z
OLIGOCLASEORTHOC LAS EP Y R I T EP Y R O X E N EQ U A R T Z
R I E g - R I E g E C K I T ES O D - S O D A L I T ET I - T I T A N I T EzA - z rRcoN
N A M E S Y E N I T E Q U A R T ZS Y E N I T E
Q U A R T Z S Y E N I T EALBANY TYPE
GRAN ITE PORPHYRY H A S T I N G S I T EG R A N I T E
R I E B E C K I T EGRAN ITE B I O T I T E G R A N I T E
D I S T R I B U T I O N
NORTH CONWAY FRANCONIA REG
S E L K N A P M T N S . R E O H I L L
PLEASANT MTN. PERCY OUAO.
T R I P Y R A M I D M T N .
NORTH CONWAY
BELKNAP MTNS
NORTH CONWAY QUAD.
B E L K N A P M T N S .
O S S I P E E M T N S ,
F R A N C O N I A R E G I O N
F R A N C O N I A R E G I O N NORTH CONWAY QUAD
P E R C Y Q U A D ,
NORTH CONWAY QUAD,
P E R C Y Q U A O .
NORTH CONWAY qUAO, PERCY qUAO.
O S S I P E E M T N S . F R A N C O N I A R E G I O NB E L X N A P M T N S . T R I P Y R A M I O
SIZE OF BODIES o . r o - 2 0 . o s Q . M l r . 5 - 6 . 0 S Q . M r .w r D T H - O . 2 5 - t . S M t .O I A M E T E R - 8 - 9 M I ,
W I D T H I - 3 M I
4 3 : 2 S O M L
S T O C K S - I . 5 T O 4 . O M I L E S I N D I A M E T E RR | N G - D T K E S - W r O T H O . 2 5 - L O M t .
L E N G T H I - 4 M I .
S T O C K S : I - 5 M I , I N D I A M E T E RA F E W S M A L L R I N G - D I K E S
SHAPE OF BODIES R I N 6 - D I K E S A N D I R R E G U L A R S T O C X S R I N G - D I K E S I R R E G U L A R R I N 6 - D I K E N . C - I R R E G U L A RP - INCOMPLETE RIN6-OIXES
N . C . - S M A L L I R R E G . M D I E SP, - INCOMPLETE RING-DIKES
L A R G E S T O C X S O F I R R E G U L A R S H A P E , A N D V A R Y -I N G S I Z E . A F E W S M A L L R I N G - D I K E S ,
TOTAL AREA ANDP E R C E N T O FM A G M A S E R I E S
4 3 . r S Q . M9 . 4 1 7 "
7 . 5 S Q . M I
| . 6 4 y "3 9 . 9 S Q . M I
6 . 7 t %
4 3 2 S Q _ M
9 . 4 3 Y "
A L L G R A N I T E S 3 I 5 , 5 S Q . M I
6 A . 9 %
T E X T U R EH Y P I D I O M O R P H I C , G R A N U L A R .S O M E P O R P H Y R I T I CS O M E T R A C H Y T I C
PORPHYRITIC,GrcUND MASSPANALLOTRIOMORPHIC TOHYPIDIOMORPHIC GRANULAR
P O R P H Y R T T t C ,G R O U N D M A S SPA NALLOTR IOMORPH I C
H Y P I D I O M O R P H T C ,G R A N U L A R
H Y P t D t O M O R P H t C ,G R A N U L A R T OM I C R O P E G M A T I T I C
A Y P I D I O M O R P H I C . G R A N U L A R ,S O M E P H A S E S P O R P H Y R I T I C
G R A I N S I Z E o . 5 - 5 . O M M MRK MINERALS: O,5 - I ,2 MMR E S T : r - O - L O M M .
GROUNO MAS:0 .06 - O.3 MMP H E N O C m T S : 5 . O - g O M M .D A R K M I N E R A S : 0 . 6 - 1 . 4 M M
o . 5 - 3 0 M M o . 5 - 3 . O M M . o _ 5 - t o . o M M .P H E N O C R Y S T S O F T E N U P T O I . 5 C M .
C O L O R G R A Y . G R E E N , B L U I S H DARK @Y, GEEN OR PINK D A R K G R A Y A N D P I N K L I G H T G R A Y T O W H I T E SNOWY WH ITE. MOTTLED P I N K T O G R E E N . S r c T T E D W I T H D A R K M I N E R A L S
M O D E
B E L , P E R C Y P . M . T R I P Y
a T z . t . 6 3 . o L oo R 4 7 . O
M P 4 6 . 6 a 3 . O a O . Oo l f G 4 2 . O 7 . O
L A g 6 . 5
8 t 2 . O r . 5 r . o
H g D 7 . O 3 . a T R 6 . 0H E D 3 . 4w R o x t . 5 r . o
F A Y I . 9
A C C 2 . O 2 . 9 3 . 3 4 . Sr o o . o 9 9 . 6 9 9 . 3 r O O . O
B E L . N , C .
QTZ a O 6.3o R a o o - lo L r 6 3 9 . o J
- -
B t 2 . 0 O . 4H A D 7 . O
HAST - 12.6H E D 3 , 6acc _!:g __!l
r o o . o t o o . o
N C , B E L . F
Q T Z t 2 . 6 t 5 O r 3 . aM P I s e . l
Io R 1 7 4 . 1 7 r . O t 7 |o r r e J r z . o z . jH A S T t ? . 3 t z . O 3 . aH E O 2 . 4FAy 0.6a t o . 5a c c o . s 2 . o o . 5
r o o o t o o . o t o o . o
A C C : M T , A P , Z R . A L . F L
A V 6 . M O D E
Q T Z 2 6 . 5
M P 2 3 . 5o R 3 4 9
o L t G | 4
HAST 6 I
a E o 5 . I
F A Y I . Og t r . 2
a c c 0 _ 6
t o o . o
A C C : F L , A L L , A P . Z R , M T
N . C , P E R C Y
Q T Z t 1 4 Z 4 . O
M P I s e so , u J
t ' " o . sH A S T A . 3 A . O
H E D O . 4
B t o aa c c r . r 0 . 6
r o o . o r o o . o
A C C : A L . A P . Z R , F L ,M T . A E G
N . C .Q r z 3 9 . 4
M P I^ , . - l s 4 . o
a s T t . l
W
P E R C Y
o . 75 . 8
o . 7
o 5
m
A C C : Z R . M T - I L . T I ,
N . C . B E L . P E R C Y F .Q T Z 2 6 . 7 3 0 . O 2 3 0 3 9 . O ' 3 4 . 2 3 5 . r " 2 s . 9 2 6 . 5
" " I - . - 4 4 . 4 s e z s s . s l s e . e
o R > 6 9 - 4 6 4 . 4 3 7 . 0 1 - | -
o . ' c . J * o J t u o ' " . . " , . "
' u . . JB r 3 . o r . 6 I 3 . o 4 . 2 2 . 6 2 . s t . 4H A s r o , - J o o r . 6A C C O . 7 - 2 0 2 . O O . I O . 9 0 . 6 0 . 6
roo .oroo .o roo .o roo .o roo_o roo .o roo .o roo .o
A C C I F L . A L L . M T . I L . A P H A D
C H E M I C A LcoMPos rT I oN
B E L . T R I P Y N . C N . CN O R I H C O N W A Y
R€o BALDFACE BLAaK?FF------ RF6-PHASE PHASE PH6F PHASF
s i o ? 6 0 . 7 5 6 2 . t 2 6 3 . 3 2 6 2 . 2 4 G 4 . 4 0 6 6 7 | . A A 7 0 . 4 3 7 3 . 6 5 7 2 . 3 5 7 3 . 0 1
0 . 6 3 0 . 6 4 0 . 5 2 o . 6 ? o . a a o . 5 o - 2 5 0 . 6 9 0 . 2 2 0 . 3 9 N DA l 2 o 3 1 9 . 5 5 1 5 . 5 7 t A . 7 5 r 5 . 4 2 | 6 . 7 5 1 7 . o 4 3 . 5 4 t 4 . 4 7 | 3 . 7 3 t 2 . 4 9 t 3 . 7 3
F e . 0 3 f . 5 4 2 . t 6 t . 2 4 t . 9 4 3 t | . 2 4 | . 3 6 1 . O 7 O . 2 3 t . 2 Z O . 4 4F e O 2 . 9 a 2 . 5 9 | . 7 7 4 . 6 9 3 . 4 4 2 . I O r . 9 9 t . 5 4 0 . e 5 t . 2 3 t . 4 aM n O NO T R o . l l o . 2 4 o . 0 3 T R O . O 3 0 . O 9
M g o o . a I o . a 6 o . 7 | o . o 7 0 . 6 2 0 . 2 | o o . 3 5 0 . | | o . 4 ? o . o l
2 - 2 9 2 _ 3 7 2 . 0 3 2 . 6 s . a 5 . 6 2 o . 9 0 2 A O . 9 0 r . t g 0 9 4
Nb2o 4 . 5 9 6 . 7 6 6 . 0 4 4 . g o 4 . 4 4 4 . 6 A 3 . 7 3 4 . 2 4 4 . 4 7 3 . 7 6 3 - 5 O
K z o s . 9 0 4 . 7 9 4 . A 6 5 . 2 6 4 _ 9 6 5 . 5 4 5 . 6 1 4 . | 6 4 . 2 7 5 . 9 A 5 . A 2
H a o ' o . o a 0 . 4 6 0 . 3 6 o . 5 6 f . o 5 0 . 4 o . 2 9 0 . 6 5 0 . 7 7 0 . A o o . I a
H a o - o . 2 4 0 . o 9 0 . 0 6 o . o 7 o . 5 0 0 . 5 0 0 _ r 7 0 . o 5
Zroz N D N D O . O 5 N D N D N D N O N O N O N O N D
P - O . o . r 3 0 . 2 3 0 . t 6 o . r 4 o . | | o -oa o . o 4 T R O . 2 7 N O
c l N D N D S . O O . O 3 N O N O N D N D N D N D N D N D
N O N O B a O O . r l N O N O N D o . o 2TOTALS 9 9 . 7 9 1 0 0 . 6 6 t O O . 2 4 r o o . 3 5 t o o . o a t o o . o a 9 9 . 4 7 9 9 . 6 0 9 9 . C O r O O . r | 9 9 . 0 5
JOURNAL MINERALOGICAL SOCIETY OF AMERICA 503
regions is shown in some detail in Figure 1. It is entirely possible,
and in fact likely, that such areas as Mount Ascutney in Vermont,
the Burnt Meadow Mountains in Maine, and the Pawtuckaway
Mountains of New Hampshire also contain rocks belonging to this
magma series, but due to the fact that they have not been studied
by the present investigators, they will not be considered in this
discussion.Briefly, the method of attack consisted of both field and labora-
tory studies. The field studies served to determine the sizes of rock
bodies and the age relations of the various rock types of the series.
In the laboratory the mineralogy and chemistry of the rocks, and
the optics and chemistry of the minerals were determined. Par-
ticular emphasis was given to the evolution of the mineral species,
for the writers feel that such a method of attack must be adopted
if problems in the evolulion of igneous rocks are to be solved.
Rocrs
PBrnocnepny AND Cnpursrnv. The rocks of the White Moun-
tain magma series may be classified into three main groups: vol-
canic rocks, plutonic rocks, and dike rocks. The volcanic rocks
consist of flows, tuffs, and breccias of basalt, andesite, trachyte,
and rhyolite. The plutonic rocks are gabbro, norite, diorite, quartz
diorite, monzodiorite, monzonite, nephelite syenite, analcite sye-
nite, syenite , qtrartz syenite, quartz syenite (Albany type), granite
porphyry, hastingsite granite, riebeckite granite, and biotite gran-
ite. The dike rocks are listed below. The distribution of the volcanic
and plutonic rocks is shown in Figure 1. For the sake of simplicity,
all the granites have been indicated with one pattern, and the
monzodiorites and monzonites with another pattern. AII the other
rocks have been mapped with separate patterns. The data have
been obtained from the published works of Billings (3), Kingsley
(26), Jenks (23), Pirsson and Rice (33), Pirsson and Washington(31 and 32), and the unpublished works of Modell (28), Williams('10), Chapman (9), Bil l ings (6), and Page (29). Mount Monad-
nock, Vermont, and a number of bodies in which the different rock
species have not been distinguished are shown on the map, but
they were not considered in the presenL study' Since the map was
prepared, Dr. A. W. Quinn of Brown University has completed
a study of Red Hill.Lack of space prohibits a complete description of each r:ci typ:,
504 THE A M ERICAN MINERA LOGIST
accordingly, most of the data have been listed in tabular form.Table 1, which is self-explanatory, gives the most important char-acteristics of the plutonic phases, together with their modes andchemical analyses. The rocks are listed according to age as nearlyas possible, except that the quartz syenite (Albany type) has beenplaced before the granite porphyry. This was done so that thequartz syenite and the quartz syenite (Albany type) might becompared more easily. Many of the data are new and are pub-lished here for the first time. Those for the Belknaps, Percy, andFranconia regions are from the unpublished works of Modell,Chapman, and Williams respectively. The following chemical anal-yses are also new: gabbro from Tripyramid Mountain, monzo-diorite from the Belknaps, Albany qtartz syenite from the Belk-naps, and granite porphyry from the Franconia quadrangle.
The chief characteristics of the volcanic rocks, which cover about30 square miles, are shown in Table 2. The dikes belonging to themagma series are very abundant and consist of olivine diabase,kersantite, diorite, camptonite, spessartite, augite syenite por-phyry, porphyritic syenite, syenite porphyry, bostonite, hasting-site sijlvsbergite, paisanite, hastingsite granite, biotite granite,aplite, and quartz porphyry. A list of their chemical analyses isgiven in Table 3.
Tl.nr,n 2
Vor,cexrc Rocxs or run Wnrrn MouNrarN Macua Snnres+Neun Basal,t Anilesite Trae hyte Rhyolite
Drsrnrnutrox Ossipee Mtns. Ossipee Mtns. North Conway North ConwayFranconia Pleasant NItn. Ossipee Mtns.
Quad. BerknapMtns. i:l|::l#r.
Tnxrune Porphyritic Trachytic and Trachytic and *ffinr.porphyritic porphyritic
Gnerw Srzn Dense.ground- Dense ground- Dense ground- Dense ground-
Cor-on Black Black Gray, black, red, Red, gray, yel-purple low, brown,
pink
MrNrnar.s Labradorite Andesine Orthoclase MicroperthiteAugite-diopside AlkalifeldsparOligoclase Soda-orthoclaseAmphibole Amphibole -A.lbite Albite
IOURNAL MINERALOGICAL SOCIETY OF AMERICA 505
MagnetiteTitanitePyriteEpidoteCalciteChlorite
Trsr;r.2 (Continued.)
Apatite Soda-orthoclase Oligoclase
Magnetite Microperthite QraftzEpidote Quartz Hastingsite
Chlorite Zircon Biotite
Sericite Magnetite Fayalite
Leucoxene Biotite Fluorite
Apatite Zftcon
Clinozoisite Magnetite
Augite Riebeckite
Hornblende Aegirine-augite
Titanite Hornblende
Epidote Apatite IChlorite Ti:'"1""
Allanrte
ChloriteEpidote
Cnnurcar- Awalvsns or VolcaNrc Rocrs oF WHrrE MouNrarN Mlcue Samrs
r 2 3 4 5 6 7Sio: 49 .04 57 .85 65 .05T ioz 3 .74 2 .34 .25Al2o3 18 . 10 16 .30 16 . 80FerOa 2 .31 3 .16 4 .97F e O 6 . 6 9 5 . 1 0 1 . 1 2MnO .08 .10 trMgO 3 .68 2 .68 .20CaO l0 .72 5 .50 1 .68NazO 2 53 3.00 3.94KzO 1 .09 2 .92 5 .22HzO* 1 .72 .61 .30H:O- .12 .10 .45PzO; .08 .13 tr
75 .38 73 .s3 72 .25 72 .05nd nd nd .23
11.85 12 .95 13 .40 14 .72r . 7 8 . 9 8 1 . 1 0 r . 0 2
. 8 8 r . 6 6 1 . 5 3 1 . 4 6
. 1 0 1 . 1 3 . l l t rnone none none .15
. 3 3 . 9 8 . 7 4 . 7 93 . 6 8 3 . 4 6 4 . 2 7 4 . 4 25 . 3 7 5 . 6 1 5 . 5 6 4 . 0 0
.50 .30 .31 . s .s
. 1 5 . 1 1 . 1 0 . 5 5nd nd nd nd
Totals 99.90 99.80 99.98 100.02 99.51 99.37 99.94
1. Basalt-Bald Knob, Ossipee Mountains (26, p. 159).
2. Andesite-Cold Rrook, Ossipee Mountains (26, p. 159).
3. Trachyte-South Moat Mountain, North Conway quadrangle (3, p. 97)
4. Quartz porphyry-south slope Mt. Pequawket, North Conway quadrangle
(30, p.408).5. Quartz porphyry-east slope Mt. Pequawket, North Conway quadrangle (30,
p.409).
6. Qrafiz porphyry-summit Mt. Pequawket, North Conway quadrangle (30,
p 409).
7. Riebeckite quartz porphyry-Dry Brook, Albany, North Conway quadrangle
(3, p. es)* These rocks occur in part as tufis and breccias.
THE AMERICAN MINERALOGIST506
3
N
r
c
- K N 5 3 9 ; 9 9 B e e R e= :3 - = - - .
€ F < f 4 ; E + i
+ D \ O r O \ n d r D o a " \ " . N . .o . o . r a " j T o . l ? 1 . : a 6 t F F 1 F- * o s o . n O n . o H H s :
* \qBqqqq\sr F H ss g* S i N l . i ; ' o ' - j - - 6 i ? g - E
H O \ O n D 4 - N @ 6 -4 N C ) N \ O O s O * N e * ! o
i ' ^ . - = Y 7q 4 i H O o
sre3sggsBs$gs-ss g
4
+r. - . i^ s
$ Ne J + is a
. . i 6 i r i + r , . o i l d o i o
€ 3 S s S \ ; 3 $ S g g g R g. t t l
T ^ o a s o \ r c o N s E E - E
.qqqqqi i ,s iK333=soc . r < r < r < 1 r - o o . i
" d ' ; ;
4 0 \ o < a ) \ o \ o 4 + n o \ € D . , N- n n q a ? r n \ ? 9 A q - e E Q
$ - : N r - o \ o < a d E
€ 3 4 $ 5 o R $ s 3 : 8= g ' d ' J t r ' ; ; . E E E E
S g S i l S . S t s K e B g \. : : : :' B - : . l r - . o € < , N c c l t r
R B E S : S F J E B g g - s -: ' : : : :" B - : N a o r o \ o + N 3
. 8 "
E
zFrz
tsH
_F
- < i l
n
a
|l
zd
O
E
(,
@"
o B e r l i n
4e'r91-
Pr",/
flV
Gr
MI MOOSILAUKE
A S r.03 9q.ml\
:rukp/\-ft a
,ffi :\
o 5 r p l F t u a 5
Sca ld o f M i les
G
% un-ffi o""u"Effilc" Granite and Gnanite Ponphyny
Iffiqt Quartz Syenite (Albany type)
l'ffi|" syenite and Quantz SYenite
l-- les Analcite Syenite
l$iiTlr.,t Monzonite and Monzodionite
l-- loo Quartz Dionite
[ ' l D D i o r i t e
l - l c 6abbro
Frr.--.-- u Volcanics
Frc. 1. Map showing distribution of the White Mountain magrna series'
JOURNAL MINERALOGICAL SOCIETY OF AMERICA .507
Due to the magnitude of the task, the writers have deemed it ad-visable to confine their studies to the plutonic phases of the WhiteMountain magma series, and have been forced to neglect the ex-trusive and dike rocks. A complete theory, however, must ul-timately consider these too.
Annel ExrBur or PruroNrc PuRsBs. The writers have cal-culated the number of square miles of area underlain by each ofthe rock types. They have also determined the percentage of eachtype with respect to the magma series as a whole. The results areshown in Table 4. The figures are given to several decirnals in orderto represent the rarer types. For the more common types thedecimals have no significance. It should be clearly understood thatthese figures are based only upon those areas studied in recentyears, and do not include areas such as the Crawford Notch andPlymouth quadrangles for which suffi.cient data are not available.
T.qrr,e 4
Tarr,n Snowrxo Anp,qr, ExrnNr or Pr.uroNrc Rocr Tvpos awD PenceNr.a.ceor Eecu Coulanan ro Toral Mecue Snnms
Naur
GraniteGranite porphyry
Quartz syenite (Albany type)
Quartz syeniteSyeniteAnalcite syeniteNephelite syeniteMonzoniteIVlonzodiorite
Quartz dioriteDiorite
NoriteGabbro
Total
Seu.tnr Mrr,Bs Prn CnNr ol Mlclrl
The relative proportions of the various rock types as they areseen today in the White Mountain region may not necessarilyrepresent the proportions of those types as magmas. They arerather a function of specific gravities. Granite, being the lightestof the magmas in the region has a greater tendency to rise, andthus, at the present level, is more abundant than any of the othertypes, whereas gabbro forms only a very minor proportion of thoseobserved at the surface.
315 .5043 .3039.907 . 5 0
43 .100 .03
?
2 6 04.001 .50o -25?
0 1 0
Sanrns68.90009 .43008. 71001.5400L41000.0066
?0. 57000 .87000.33000.0546
?0.0218
457 .78
THE AMERICAN MINERALOGIST
AcB RBrarroNs. The most important data furnished by fieldrelations as to the evolution of the magma series pertains to therelative ages of the difierent rock types. Although the data fromany one area are by no means complete, the facts collected fromthe whole region indicate that there is a definite succession ofrocks beginning with the volcanics.
There is ample evidence that the so-called Moat volcanics areolder than most of the plutonic rocks of the White Mountainmagma series. In the North Conway quadrangle, all the plutonicphases of the series present, except the diorite, cut the volcanics.In the Ossipee Mountains, the Albany quartz syenite cuts thevolcanics and, furthermore, contains inclusions of the latter. OnPleasant Mountain, the volcanics are definitely older than theaugite syenites and syenites. Thus, the Moat volcanics are clearlyolder than the syenites and more siliceous rocks, but, unfortu-nately, there is no direct evidence as to their age relative to thegabbro, diorite, and monzonite. However, fragments of the plu-tonics have never been found in the Moat volcanics.
ft is certain that the gabbro and diorite of the series are olderthan the biotite granite since they are cut by the latter in the Belk-nap Mountains. fn the Belknap Mountains, also, some of the dio-rite occurs as inclusions in the monzodiorite. There are no data onthe age of either the norite or quartz diorite. At Pleasant Mountain,the analcite syenite cuts the monzonite and is, in turn, cut by thesyenite. The relative age of the nephelite syenite at Red Hill isnot known, but until evidence to the contrary is found, it seemsprobable that it has the same relative age as the analcite syeniteof Pleasant Mountain. For the most part, the quartz syenite isyounger than the syenite. The Albany quattz syenite is probablyessentially contemporaneous with the other qtrartz syenite. fn theNorth Conway and Percy quadrangles, the latter is found locallygrading into the hastingsite granite. In the Percy quadrangle, thehastingsite granite, in turn, grades into the riebeckite granite.Evidence from the Franconia region shows that the granite por-phyry is older than the q:uartz syenite (Albany type) since it isintruded by the latter. fn all areas where the Conway biotitegranite occurs it is definitely the youngest plutonic present sinceit cuts most of the other plutonic phases. In most areas, the dikesare younger than the biotite granite, but at Pleasant Mountain,some are older than the diorite.
TOURNAL MINERALOGICAL SOCIETY OF AMERICA 50s
In summary, the rocks of the White Mountain magma seriesmay be l isted chronologically (oldest at bottom) as shown inTable 5. The relative ages of those types shown in brackets arenot well known.
Tenr,n 5Rocxs ol rrm Wrurn Nloulqrerx M,rcua Snruos Lrstnl
According to Age (Oldest at Bottom)
Vent agglomerate of the BelknapsBiotite granite
Riebeckite granite
Hastingsite granite
Quartz syenite (Albany type)
Granite porphyry
Quartz syeniteSyenite
INephelite syenite and analcite syenite]MonzoniteMonzodiorite
IQuartz diorite]Diorite
lNoritelGabbroVolcanics-definitely older than syenite and younger rocks Age relative to
analcite syenite and older rocks unknown.
Inspection of this table shows features which are characteristicof many comagmatic regions. As evolution proceeds, there is agradual decrease in the percentage of ferromagnesian minerals andanorthite, and a corresponding increase in the alkali feldspars, and,in the final stages, qtrartz. Chemically this means a decrease iniron, magnesium, and lime and an increase in soda, potash andsilica.
Locus ol ORTGTN ol RocK TvpBs. ft is clear to those who aremost familiar with the White Mountain magma series that thedifferent rock types did not origind,te at the levels where we nowsee them. They formed at some depth below the present surfaceof the earth and were then injected into higher levels. Little differ-entiation took place in the bodies now exposed to view. The evi-dence for this is that the contacts between different members ofthe White Mountain magma series are relatively sharp-in manyinstances they show a knife-edge sharpness. ft is true that in someIocalities gradational contacts exist, but these are usually confinedto zones a few feet wide and are insignifi.cant when compared tothe size o{ the rock-bodies as a whole.
o
510 THE AMERICAN MINERALOGIST
Mncnaxrcs ol hqrnusroN. It is not the purpose of this paperto discuss the mechanics of intrusion of the White Mountainmagma series, but a brief statement is desirable. Evidence fromthe Ossipee Mountains, North Conway quadrangle, Percy quad-rangle, Belknap Mountains, Franconia quadrangle, and Tripyra-mid Mountain shows that the plutonic rocks of the series occur asstocks and irregular ring-dikes, and that cauldron subsidence hasbeen the dominant mechanism of intrusion.
MrNrnals
INrnorucroRy STATEMENT. Practically all discussions of mag-
matic differentiation are concerned merely with rock types. The
writers are convinced, however, that a far more f undamental study
is essential, and that the evolution of the individual minerals must
be thoroughly investigated. In this belief, an attempt has been
made to determine by chemical and optical methods the composi-
Tesln 6
MTNERALS ol rne Wstrr MouNrerlr }Lr.clre Srnrns
Or,rvrNn Gnoup Pr,lcrocr-esn GnouP, Olivine (hyalosiderite; Albite
Fayalite OligoclaseAndesine
PvnoxpNr Gnoul LabradoriteDiopside-hedenbergite seriesAegirine series Onruocr,asnAugiteEnstatite hvoersthene series ANontRocr,lsn
AMptrsorn GnoupHomblendeHastingsiteRiebeckite
Brotm Gnoup
Allanite
AstrophylliteFluoriteIlmenite
MrcnoPBntsrrr
Quanrz
FrI,ospatqorosAnalcite
CancriniteNePheliteSodalite
AccESsoRrljs
MagnetitePyriteTitaniteWoehleriteZircon
JOURNAL MINERALOGICAL SOCIETY OF AMEMCA 511
tion of the essential minerals in all the plutonic phases of the WhiteNlountain magma series. Obviously, to do this by chemical meansentirely would be very expensive and laborious. Consequently,certain key minerals have been chosen for chemical analysis andtheir optical properties determined. From these data, supplementedby information contained in the textbooks of optical mineralogy,it has been possible to determine the chemical composition of mostof the minerals in the White Mountain magma series.
A list of the minerals of this magma series is shown in Table 6.The essential minerals are the feldspars, olivines, pyroxenes,
amphiboles, biotites, and quartz. Quartz first becomes importantin the quartz syenites and granites, although it first appears in theqrartz diorite, which is of very limited extent.
Frrlspnns. The usual relations were found in the feldspars-calcic and intermediate plagioclases in the gabbros, diorites, andmonzonites ; alkali feldspars (albite-oligoclase, orthoclase, anortho-clase, and microperthite) in the monzonites, syenites, and granites.
BrorrrB. Three new chemical analyses have been made of bio-tites from the White Mountain magma series. One of these is fromthe gabbro of Tripyramid Mountain, a second is from the monzo-diorite of the Belknap Mountains, and a third from the Conwaygranite of the Percy quadrangle. The chemical compositions andoptical data are given in Tables 7 and 8.
It will be observed that in passing from the gabbro to thegranite, there is a systematic decrease in silica, titania, alumina,and magnesia, and a systematic increase in ferric iron, ferrous iron,manganese, and lime. The potash and minor oxides are not soregular in their variation. The most striking change is the increasein the iron-magnesia ratio. The available data suggest a contin-uous reaction series with no sharp discontinuity.
OlrvrNB. The olivine (hyalosiderite) in the gabbro of TripyramidMountain has been isolated and analyzed. The results are givenin Table 7 and the optical properties in Table 8. It is about 40/sfayalite and 60/6 forsterite. Unfortunately, the fayalite from themore siliceous rocks could not be separated for analysis. Opticaldata had to be employed to determine the chemical composition,and typical data on fayalite from a syenite are given in Table 8.They indicate fayalite, with l07o forsterite and 90/6 fayalite. Asin the biotites, there is a great increase in iron relative to magnesiain going from the basic end of the series to the siliceous. Olivine
512 THE AMERICAN MINERALOGIST
Tasln 7
Csnurcer, Couposr:rroN ol' tun MTNInALS oF TrrE Wnrre MouNrartMacu,r. SrnrBs
sioz 37 . 74 35 .88T ioz 5 .22 4 t7A1203 16.31 t4.96Fezos none 2.33FeO 15.52 20 .88MnO .06 .18MgO 14.23 10.04CaO .04 .12NaeO .44 .36KzO 8 .88 9 .20IJ"O+: : ' : ' r 1 7 . 8 0tIrU-F . 4 5 1 . 5 8
100.06 100.50LessOforF .19 .66Total
-gg.87 gg.U
3 4 5 633.48 47 .58 46 56 37 .402 .94 .37 1 .83 3 .20
t3 .64 l .16 2 .48 12 .348 . 0 0 2 . 6 0 6 . 2 4 4 . 1 6
23 .54 24 .21 27 .27 25 .84r . o 2 . 5 9 1 . 0 8 1 . 2 44 . 9 7 3 . 3 4 1 . 3 5 2 . 2 0
. 5 6 1 8 . 8 0 6 . 1 5 9 . 7 2
. 5 3 . 4 7 4 . r 7 1 . 8 07 . 8 0 . 2 1 t . 3 7 1 . 3 6
2.63 .34 t t ! t rnd .60
.95
7 8 946.98 34.96 none1.49 none 48.90|.29 none none
11.93 none 9 .8023.38 36 .77 38 .97
. 2 4 . 5 2 . 3 6
. 1 3 2 7 . 0 4 2 . 0 3l .9 l none nd8.90 nd nd2.74 nd nd1.10 nd ndnone
100.06.40
99.66 99.67 99.77 99.86 100.09 100.58 100.06
1. Biotite from gabbro (7052)*, Black Cascade, Tripyramid Mountain.2. Biotite from monzodiorite (609), one-third mile northwest of Ames station,
Belknap Mountains.3. Biotite lrom granite (19287), granite quarry one mile south of Beech Hill, Percy
quadrangle.
4. Hedenbergite from syenite (19258), east slope of knoll south of Burnside Brook
at elevation 2 160 feet, Percy quadrangle.5. Hornblende from syenite (19247), in Moore Brook at elevation 2000 feet, Percy
quadrangle.
6. Hastingsite from quartz syenite (18299), Jackson Falls, Jackson, North Conwayquadrangle (J, p. 110).
7. Riebeckite from granite (19274), I mile N.E. of small swamp on Mill Mt.
Percy quadrangle.
8. Hyalosiderite from gabbro (7052), Black Cascade, Tripyramid Mountain.
9. Ilmenite from gabbro (7052), Black Cascade, Tripyramid Mountain.
* Numbers in parentheses refer to Harvard University rock collections. F' A.
Gonyer, analyst of all except specimen 6.
(hyalosiderite) occurs in the gabbro, fayalite in the syenites andgranites. Members of the olivine series have not been observedin the diorites or monzonites. There is, apparently, a definite stagecorresponding to the intermediate types, in which members of theolivine series could not exist in equilibrium.
PvnoxBNBs. The pyroxenes of the White Mountain magmaseries belong to four groups: (1) enstatite-hypersthene, (2) aege-
I a o : l' * ' / \ l ^ .
\
. 2 ; { [ ^ N * -l ? i E i l G- i - n 1' X / \ i ' " < / d t At r \ - r - - :
t\
, o ^ o * : ? s lI v \ / ; i [ , , , ,
I . F S X I I
. t c l l ' dI o - F d i l ! r ?-f
I A !, l l </<1 S' i = ! , { - F" ;I
. , v ^ i l - o 5bo+ L i i . l r >Y E > ,; F > \ d e
N r i
" : PNJ r . V d
r - " F d| % Ag ^ ' b so o r P : _
TOIIRNAL MINERALOGICAL SOCIETY OF AMERICA
d a a F
O
z
. . i + r \ Os € @ €
* I : r 9 9j v t v
r r Q^ , @ o \ o' ' : \ O € r
i € @
= q q \
o n \ o6 6 i -J . J J
i O nN 4 \ o\ \ \
e < l O \qqq
@ O \ NO \ : N
9 \ \
m o , $@ a o9 9 \
o \ o D
\ \ \
\ o i No N r6 6 0
€
r
<l
o
a
p.
a
azFz
@ L
EF< r'lt s F
arl
z
3
o \ 4 4
N 4 \ O \ O
J J J
a n D- a 9 9
514 THE AMERICAN MINERALOGIS:T
rine, (3) augite, and (4) diopside-hedenbergite. Hypersthene isrelatively rare, being found only in the norite of TripyramidMountain. Aegerite is also rare, having been observed in a riebeck-ite granite on Bartlett Haystack in the Crawford Notch quad-rangle. Available data indicate that most of the so-called augitein the White Mountain magma series probably belongs to thediopside-hedenbergite series. The problem thus resolves itself intoa study of the diopside-hedenbergite series. One new analysis wasmade for the present paper, the pyroxene from syenite in the percyquadrangle, which was found to be hedenbergite. This analysis isgiven in Table 7, the optical properties in Table 8. It is 90/6heden-bergite, 1016 diopside
Attempts to obtain clean material for an analysis of the pyroxeneof the gabbro from Tripyramid Mountain were unsuccessful. Theoptics were obtained, however, and are shown in Table 8. Win-chell 's curve (41, p. 18a) indicates that it is diopside-hedenbergitewith 6870 diopside, 32/6 hedenbergite.
Optical data were also obtained on pyroxenes from the noritefrom Tripyramid Mountain, with the results shown in Table 8.According to Winchell 's curve (41, p. 177),they indicate a hyper-sthene with about 24/6 FeSiOt and 76/6 MgSiOr.
In a syenite porphyry from the Cape Horn ring-dike of the percyquadrangle two pyroxenes were observed. One is very rare andoccurs as phenocrysts, the other is in the groundmass. The opticalproperties of the pyroxene occurring as phenocrysts are given inTable 8. It is very similar to the pyroxene from the Tripyramidgabbro. The groundmass pyroxene, however, is a hedenbergite.
The above facts indicate that the pyroxenes of the White Moun-tain magma series belong largely to the diopside-hedenbergiteseries, and that in the gabbro the pyroxene is about 6816 diopside,32/6hedenbergite; in the syenite (except for the rare phenocrysts)and granite the pyroxene is about 10lp diopside, gO/6 hedenber-gite. There is thus a great increase in iron relative to magnesia ingoing from the basic to the siliceous end of the series. Since, how-ever, pyroxenes intermediate in composition between those of thegabbro (6816 diopside) and the syenite (1016 diopside) have notbeen observed, and since both occur in a syenite porphyry, theformer as phenocrysts, the latter as the groundmass, we are ledto believe that a discontinuity in the series may exist.
AupurnolBs. Three analyses of amphiboles from the White
JOURNAL MINERALOGICAL SOCIETY OF AMEKICA 515
Mountain magma series are available, two of which are new. Oneis a soda hornblende from a syenite in the Percy quadrangle andthe second is a riebeckite from a riebeckite granite in the same aiea.A third analysis is of a hastingsite from the North Conway quad-rangle. It will be observed that all three are very rich in iron rela-tive to magnesia. Unfortunately, amphiboles in the more basicrocks could not be successfully separated for chemical analysis.Bil l ings (4,p.293) has shown, however, from a study of the wholeNew England-Quebec alkaline province that the amphiboles inthe basic rocks have a much lower iron-magnesia ratio than thosefrom the siliceous rocks.
The following facts may be noted: (1) Riebeckite is confined tothe granites. (2) Hastingsite is confined to the syenites, quartzsyenites, and granites. (3) The range of the sodic hornblende isunknown. Riebeckite is younger than hastingsite, for in somespecimens the former has been observed as a shell about the latter.
fr-uBNrre. The opaque mineral ilmenite from the gabbro fromTripyramid Mountain was studied in polished section by thewriters, and was found to be a homogeneous ilmenite. This min-eral was also separated by the writers and analyzed by F. A. Gon-yer; the analysis is given in Table 7.
Quenrz. Consideration of a number of important facts in regardto the occurrence of quartz may be helpful in explaining later themeans whereby it has originated.
In the first place, silica-deficient femags are found mainly inthe siliceous rocks. Hastingsite and biotite-both minerals low insilica-are the principal dark minerals of the granites, the rocksin which qrtartz is most abundant. Such an association stronglysuggests that the origin of quartz is in some way related to thecrystallization of the silica-poor femags.
Secondly, with the exception of a small body of quartz dioritein the Franconia region, quartz is not present in the White Moun-tain magma series in any quantity until the quartz syenite stage.It is of the greatest importance to remember that it is also in thesyenite stage that hastingsite and fayalite first appear, and pyrox-ene becomes unimportant.
In the third place, the rocks of the so-called New Hampshiremagma series which occur in the same general area as the WhiteMountain magma series, but are somewhat older, represent a casein point. In these rocks, which include q:uartz diorite, granodiorite,
516 THE AMEMCAN MINERALOGIST
quartz monzonite, and granite, biotite is the only ferromagnesianmineral present in any amount, and it occurs in all types. In con-trast to the White Mountain magma series, quartz appears veryearly in these rocks. In other words, the appearance of biotite isaccompanied by the simultaneous appearance of quartz. This factalone is strong evidence that qtartz is, in part, the result of thecrystall ization of biotite, a sil ica-poor femag.
Work by F. S. Miller (27) on the basic rocks of the San Luis Reyquadrangle, California, shows a very similar relation betweenbiotite and quartz. Biotite, which appears in these rocks when theanorthite content of the plagioclase reaches about 56/6, is closelyaccompanied by the appearance of quartz. Apparently, the twoare closely related in origin. Miller has further found that manyof the gabbroid rocks contain nodular masses composed of pyrox-ene and calcic plagioclase. These basic nodules grade out into azone composed of plagioclase, pyroxene, biotite, and quartz. Far-ther from the center of the nodules the pyroxene has changed tohornblende, and quartz and biotite have become more abundant.The quartz and biotite appear together and increase in amountproportionally.
PART II. THEORY
TsBonres or OnrcrN or RocK Tvpns
INrnonucrony Srl.rnunNr. Many theories have been developedto explain the great diversity of igneous rocks, particularly in sucha consanguineous group as the White Mountain magma series.Some theories are based primarily on differentiation-that is, aprimary magma of uniform composition is supposed to break upinto subsidary fractions of varying composition. Other theoriesinvolve the mixing of two difierent magmas, or the assimilationof solid rock by magma. Still other theories are based on the melt-ing up of rock in place, either in whole or in part, the latter beingknown as selective fusion.
TneonrBs of DrrrBnBNTrATroN. fn recent years, a number oftheories of differentiation have been presented by petrologists.These include principally liquid immiscibility, gaseous transfer,and fractional crystallization. These theories have been summar-ized by Daly in his "Igneous Rocks and the Depths of the Earth"(12, pp. 319-332).
JOURNAL MINERALOGICAL SOCIETY OF AMEKICA 5t7
According to the theory of liquid immiscibility, difierentiationtakes place in the magmatic state by the separation of liquidphases of different compositions. The crystallization of thesephases gives rise to difierent rock types. Bowen (7, pp. 7-19) andGrieg (18) have shown, however, that the evidence is decidedlyagainst limited miscibility in natural magmas. Accordingly, wecan dismiss this method of difierentiation without further comment.
Fenner (14, p. 743) is one of the chief advocates of gaseous trans-fer. He considers that the gases responsible for differentiation form
a separate bubble phase and rise from lower to higher levels in
the magma, making selective transfer of materials from lower tohigher levels. Bowen (7, pp. 7-19) and others, although realizingthat gases may play an important role during the cooling of amagma, believe it very unlikely that notable differentiation of themolten material would accompany this action. The present writ-ers are of the opinion that this process has certainly not beenoperative to any great extent in the formation of the White Moun-tain magma series.
Fractional crystallization is the theory of differentiation mostwidely accepted at present. According to this theory, differentrock types are formed by the early crystallization of mineralswhich are richer in lime, iron, and magnesia than the magma fromwhich they separate. If these minerals are isolated from the magmaby gravitative settling or by the squeezing out of the residualmagma itself, difierent rock types are produced. For a more de-tailed discussion of this theory, the reader is referred to Bowen's
"Evolution of the Igneous Rocks" (7).Tunonrns or AssrururroN AND PunB-uBr,rrNc. DaIy (12, p.
287) has strongly emphasized the importance of both assimilationand pure-melting in the formation of igneous rocks. He definesmagmatic assimilation as "the mutual solution of magma withsolid rock, with other magma, or with foreign gas, so that a newmagma with at least some approach to chemical homogeneityresults." In other words, the assimilation of granitic material bygabbroic magma would produce a magma of some intermediatecomposition. Since differentiation of the syntectic magma accom-panies and follows assimilation, the resulting rocks are not thereadily identifiable products of addition. Magmatic assimilationmay be either high-level or abyssal (intracrustal or in the sub-stratum).
518 THE AMERICAN MINERALOGIST
By pure melting the writers mean melting up of older rockswithout contamination by magmas, solutions or gases. It may becomplete or partial and in the latter case may be called selectivefusion. The last minerals to crystallize in the original formationof rocks (quartz and feldspars) would, in general, be the first to gointo mutual solution when blocks submerged in a magma reach aIevel where sufficiently high temperatures are available. The liquidthus formed might either rise as an independent mass because ofits low specific gravity or it might become incorporated in thesurrounding magma.
TnBonrBs ol Onrcru oF ALKAr-rNE RocKS. A great manytheories have been proposed to explain the origin of the alkalinerocks. Among the writers responsible for these are Becke (2),
Jensen (24) , Evans (13) , Goldschmidt (17) , Bar th (1) , Foye (15) ,Harker (19) , Shand (34, 35) , Holmes (21,22) , Smyth (36,37) ,Gil lson (16), Kennedy (25), Daly (11), and Bowen (7). Due to a lackof space none of these will be considered here. For a complete discus-sion of any of the theories the reader is referred to the above authors.
Ourr,rNB or TnBonv AccnprBr. The writers propose to showthat none of the processes, outlined above, operating alone is ca-pable of producing the complete White Mountain magma series,but that a combination of fractional crystallization, pure melting,and assimilation is necessary. They will show that, although it isqualitatively possible to produce the White Mountain magmaseries by fractional crystallizalion, it is quantitatively impossible.
The writers will first consider the evolution of the minerals ofthe White Mountain magma series and then the mechanics ofdifferentiation through fractional crystallization. They wiII thendiscuss the quantitative difficulties involved in deriving all therocks by this process alone, and, after a consideration of assimila-tion and pure-melting, will describe in greater detail the theoryaccepted.
EvorurroN ol rHE MrNpner-s oF THE Wnrrr MouNrarNMacua Ssnrrs
FBrpspens. The general variation of feldspar composition is awell-known fact. In the White Mountain magma series, the feld-spars follow the general rule, with calcic plagioclase appearing inthe gabbro and becoming increasingly richer in alkalies as therock becomes more siliceous. Alkali feldspars (orthoclase, micro-
JOURNAL MINERALOGICAL SOCIETY OF AMERICA 519
perthite, and anorthoclase) fi.rst appear in the monzonites, butbecome abundant only in the syenites and granites. The liquidus-solidus diagram of the plagioclase feldspars is well known, wherebythe plagioclase crystallizing from a magma is more calcic than themagma from which it crystallized. The importance of this fact isdiscussed further on a later page.
Brorrrr, Or,rvrNn, PvnoxnNns, ANrpnreolns. On earlier pagesit has been pointed out that the iron-magnesia ratio in the ferro-magnesian minerals is much higher in the siliceous rocks of theWhite Mountain magma series than in the basic. J. H. L. Vogt hasmade similar observations: "The low-melling components of theferro-magnesian silicates which are rich in iron and the low-meltingalbite component of the plagioclase are all enriched in the rocksof the rest-magma direction" (39, p. 123). fn an even earlier paper(38, pp. 151-152) he gave reasons for believing that the l iquidus-solidus curve of the olivine series is similar to that of the plagio-clase series. This deduction has been confirmed by recent experi-mental work by Bowen and Schairer (8, p. 15.1).
A diagrammatic curve for olivine, pyroxene, and biotite, similarto the liquidus-solidus curve for plagioclase feldspar is shown inFigure 2. It is intended to show that the magnesia-iron ratio ismuch higher in the solid phase than in the liquid phase with whichit is in equilibrium. If complete reaction occurs, the final solidphase will have the same composition as the original liquid. If,however, reaction is incomplete, the final liquid will have a lowermagnesia-iron ratio than the original liquid. Difierentiation willproceed, therefore, in the direction of rocks with a low magnesia-iron ratio. The amphiboles will follow the same general rules.
Quanrz. It has been pointed out that (1) the silica-defi.cientfemags of the White Mountain magma series are found mainly inthe most siliceous rocks, (2) qtartz does not appear in any quantityin the White Mountain magma series until the quartz syenitestage, (3) in the older New Hampshire magma series, biotite andqvaftz appear very early in the evolution of the series, and (4)in the San Luis Rey quadrangle of California biotite and quartzappear simultaneously.
These facts indicate that the formation of free silica in the WhiteMountain magma series is, at least in part, related to the evolu-tion of silica-poor dark minerah. ft is, apparently, a process takingplace in the late stages of differentiation. The writers believe that
UEfF
tUq
:F
oz
UEoz
THE AM ERICAN MI N ERA LOGIST
in the late stages of the evolution of the White Mountain magmaseries potential metasilicates, such as members of the diopside-hedenbergite series and common hornblende broke down to giveorthosilicates such as fayalite, hastingsite, and biotite, with the
Mgo €ND OFSER I ES
FORSTERITED I O P S I D E
M 9 B I O T I T E
Fro. 2. Diagrammatic liquidus-solidus cure for forsterite-fayalite, diopside-
heden.bergite, and biotite series. Modified after Vogt. Since this figure was pre-
pared Bowen and Schairer have presented a similar diagram (8).
release of free silica to form quartz. The alkali feldspars were alsoinvolved in this process in certain reactions also with the releaseof free silica. By further fractional crystallization, a relativelysmall amount of quartz might be concentrated to form granites.
F e O E N D O FS E R I E S
FAYAL I TEHEDENBERG ITEF e B I O T I T E
TOARNAL MINERALOGICAL SOCIETY OF AMEMCA 521
TnponBrrcal EvolurroN oF THE WnrrB Mounrerx MacueSBnrBs sy Fn.acrroNar- Cnvstar,LrzATroN
Pnocnss. Let us summarize, in the light of the above facts, thetheoretical evolution, by fractional crystallization, of a gabbroicmagma such as that which gave rise to the White Mountain magmaseries. To begin with, we may assume a great chamber of gabbro,some distance below the surf ace, which is in the process o{ cooling.As cooling proceeds, olivine, pyroxene, and calcic plagioclase beginto crystallize first. fn this early stage of differentiation it is obviousthat the ferromagnesian minerals separate out in far greater pro-portion than in the later stages, since the amount of dark mineralsin each succeeding rock type has become less and less. Whetheror not there is opportunity for reaction between the magma andthese early-formed crystals is of the greatest importance. Sincecrystallization, with perfect reaction, of a gabbroic magma wouldresult in the formation of a gabbro, it is evident that in the pro-duction of the observed series, reaction has not been complete.
The minerals which crystallize early, having a specific gravitywhich is generally greater than the liquid, tend to settle into thedepths. The residual liquor left behind now has a different com-position than at first. Intrusions of the magma into the countryrock, which would originally have resulted in gabbro, now producesome such rock as diorite. Further crystallization and settling ofminerals causes the magma to become progressively less mafic untilit has finally attained the composition of a granite. fntrusion ofthe magma at appropriate intervals produces the various rocktypes of the series.
The chemical changes which take place in the magma duringthe course of difierentiation are of fundamental importance. Theseinclude: (1) a decrease in the amount of lime and an increase inthe amount of alkalies in the magma, (2) the enrichment of themagma in iron relative to magnesia, and (3) the absolute enrich-ment of the magma in silica, especially in the late stages of differ-entiation.
The decrease in the amount of lime is due, as we have seen, tothe early separation of calcic plagioclase and pyroxene. Calcicplagioclase and sodic plagioclase form a continuous reaction seriesand the early crystallization of the former causes the magma tobecome enriched in the latter. The amount of potash is also in-creased in the magma as differentiation proceeds. The more calcic
522 THE AMERICAN MINERALOGIST
plagioclase is found, accordingly, in the less siliceous rocks, whereasthe soda and potash feldspars are concentrated in the late differ-entiates.
Accompanying this process is the steady increase in the amountof iron relative to magnesia from the more mafic rocks to thegranites. We have already considered the means by which thechange takes place. It is the result of the early separation of min-erals of the biotite, olivine, and pyroxene series which are high inmagnesia relative to iron (magnesia-rich biotite, hyalosiderite,and diopside). This causes a relative increase in the iron-magnesiaratio of the magma itself . As crystallization proceeds, the mineralsformed become relatively poorer in magnesia and relatively richerin iron, until in the late stages (syenites and granites) they arehigh in iron compared to magnesia (iron-rich biotite, fayalite, andhedenbergite). There is evidence that there is a similar iron-mag-nesia-ratio change in the hornblende system.
The rather sudden increase in the silica content of a magma bydifferentiation has already been pointed out. As a rule, quartzfirst appears in the White Mountain magma series in the syenitestage. ft is also in this stage that hedenbergite and silica-deficienthastingsite first appear, and the possible origin of quartz by reac-tions involving these minerals has been considered. fn the gran-ites, quartz is present in great abundance. The fact that it isaccompanied by biotite and hastingsite (minerals low in silica)suggests that it is, in part, closely related in origin to these twominerals.
By thus emphasizing the simultaneous appearance of silica-deficient femags with quartz in the later stages of the evolution,the writers do not wish to imply that silica-deficient femags areconfined to the siliceous end of the series. Olivine is found in thegabbro, and biotite appears all through the series.
DnnrvarroN or Rocr Tvrns. It is possible to derive the seriesfrom gabbro through monzonite, syenite, and granite by the sub-traction of crystals as shown below in Table 9. There are severalprimary assumptions on which the development of this series isbased. In the first place, in order to simplify the method, biotitewas omitted all the way through the calculations, largely becauseit is in general a late mineral to crystallize. Secondly, as shown inthe liquidus-solidus diagram for the plagioclases, the feldsparwhich crystallized in a given magma will be richer in anorthite than
JOURNAL MINERALOGICAL SOCIETY OF AMERICA 523
the magma. For this reason, a feldspar more calcic than the normalplagioclase of the magma was subtracted in each case. A similarrelation holds for the iron-magnesia ratio of the ferromagnesianminerals. This is brought out in the curve shown in Figure 2. Thus,in each case, a ferromagnesian mineral was subtracted which hadrelatively more magnesia to iron than the normal mineral for therock type. Otherwise, the minerals subtracted had the composi-tions of the analyzed minerals of the various rocks, as shown above.Ferric iron has been converted to ferrous iron so that the iron couldbe handled as a total in the ferrous state. Minor constituents havebeen omitted from the calculations.
Tesln 9
Dnnrv,tnox ol Rocx Typos rnou e Gesrno Mecua lvFn.qcrrowar, Cnvstar,r,rzanox
(For details see text)
SiOzTio:Al2o3FeOMsoCaONazOKzO
OrthoclaseAlbiteAnorthitePyroxene
OlivineIlmeniteHastingsite
Total
SvnNrrn63 ffi70
. 6 1t 6 . 2 04 . 0 5
803 . 1 36 . 0 4.) oo
GnnNrrn7 t .n%o
. 5 514.322 t 0
. 2 12 . r 0J . J I
5 . 7 8
ee.77%
SvnNrrrro GneNrrn
20 027019.915 . 0 0
13.00
Glssno MoNzoxrrp48.0470 58.00%2 . 5 6 1 . 3 8
2 0 . 1 0 1 6 . 1 08 . 2 t 6 . 4 54.68 2 03
t1 .52 7 .823 .01 4 .60
.79 3 69
Total 98.9170 r0o.0770 r00 0970
MtNrnar,s Surrnectlo ro Crrewcn:
Gasgno loMoxzorvrrn
- %17 0036.0010.00
MoxzoNrtero Svrwrre
- %5 2 4
1 1 . 9 516.00
10.004 50 2.00
77 5O7o 35 t97o s7 .937
The development of this series was accomplished by startingwith the analyzed gabbro and subtracting minerals present in thegabbro, with the qualifications given above. The resulting rockwas a typical monzonite. By repeating the process, a syenite and agranite were derived. All of the calculations were made on the
THE AMERICAN MI NERALOGIST
basis of molecular proporlions. The following table shows theanalyses of the rocks and the minerals subtracted to form the rocknext in the series. Under the gabbro, the minerals subtracted toform the monzonite are givenl the derived monzonite analysis isshown and below that, the minerals subtracted to form the syeniteet cetera.
On the way through this series the diorite and monzodioritewould be developed, and by a similar process the biotite graniteswould be formed from the hastingsite granite.
The analcite and nephelite syenites are definitely off the mainIine of descent shown in the above table. Special conditions arenecessary to explain their presence. Both of these rocks form smallbodies and there is no available field evidence which offers a clueas to their origin. Daly (12, p. 510) has suggested the possibil i tyof the assimilation of lime-bearing rocks in the region, but does notexclude the possibility of the role of gaseous transfer. A third methodby which these rocks might be formed, which would be consistentwith fractional crystallizatlol, is based upon Bowen's theory ofthe incongruent melting of orthoclase to form felds- pathoids.
QuaNrrrarrve Drnlrcur-rrEs. An objection to the derivation ofthe various rock types of the White Mountain magma series byfractional crystallization is based on the belief that tremendousquantities of gabbro would be necessary to produce the syenitesand granites, and, thus, there should be large quantities of basicrocks beneath the White Mountain area. According to isostasy,such a region should be a lowland and not a mountainous area. Itis possible to compute roughly the necessary quantity of gabbroto produce the known rocks of the White Mountain magma series.This may be done in several ways, one of which is by direct applica-tion of the figures in Table 9. There can be l itt le doubt that theplutonic bodies observed at the surface extend down at least a mile,and probably much farther. We may thus readily calculate thevolume of the various plutonic bodies in the uppermost mile ofthe earth's crust, for the areal extent in square miles in thal caseis equivalent to the volume in cubic miles.
Since the known areal extent of the granite and granite por-phyry is 358.8 square miles, there would be 358.8 cubic miles ofgranite and granite porphyry. Referring to Table 9, i l 77.5/e oIthe gabbro is removed, the remaining 22.5/6 is monzonite; if35.1970 of that is removed, 14.97a of syenite remains; 57.93/6 of.
JOURNAL MINERALOGICAL SOCIETY OF AMERICA .)z J
the constituents of the syenite removed leaves 6.3/6 oI granite.Thus, to produce 358.8 cubic miles of granite, 5700 cubic miles ofgabbro would be necessary. Similarly, 90.5 cubic miles of syeniteand quartz syenite would require 607 cubic miles of gabbro, and2.6 cubic miles of monzonite would require 11.5 cubic miles ofgabbro. Thus, to produce all of the rock types by fractional crys-tall izalion of a gabbro, 6318 cubic miles of gabbro would be re-quired. If this gabbro were spread out under the whole area shownin Figure 1 about 9000 square miles-it would have a thicknessof 0.7 miles.
These figures, however, give a minimum. They assume an effi-ciency which undoubtedly does not exist. In the second place, thesefigures are based on the assumption that the plutonic rocks of themagma series are only one mile in thickness. The differentiatesare probably actually more than a mile thick. It is doubtful, never-theless, if they ever extended very high above their present highestpoint. Some of the coarse plutonic phases hold up mountains whichare only about 1000 feet lower than the highest point in the region(Mount Washington, 6290 feet). Bil l ings (3, p. 128) has shownthat the surface on which the volcanic members of the magmaseries were deposited was at least a l itt le higher than Mount Wash-ington and possibly much higher. Assuming that the ancient sur-face was a mile or more above Mount Washington, the plutonicrocks could not have extended very far above their present highestposition since coarsely crystalline rocks cool at some depth. Thereis no means, however, of knowing the depth to which the plutonicdifferentiates extend. ft seems logical to assume, however, thatthese rocks probably extend at least 3 or 4 miles below the presentsurface. If the mechanics of ring-dike intrusion as outlined byAnderson (10) are correct, the vertical dimension of a ring-dikewould be approximately the same as the diameter of the ring itself .In view of this fact, the plutonics of the ring-dikes and associatedstocks might extend down at least 8 or 10 miles.
If the above facts are true, a tremendous quantity of parentgabbroic magma would be required. The total area of the map (Fig-ure 1) showing the White Mountain magma series is 9000 squaremiles. If the magma were confined entirely beneath the area cov-ered by this map, it would extend all the way to the basaltic sub-stratum. It is diff icult to understand how so much gabbroic rockcould now underlie this part of New England and yet allow the
526 THE AMERICAN MINERALOGIST
country to stand so high isostatically. Furthermore, there is somequestion as to whether the relatively thin crust would have beenstable over such a large molten mass.
AssrlrruuoN AND Punp-lrBr-uNc. Consideration of the effectsof assimilation in producing rock types involves several factors.The depth at which the process takes place, the character of thematerial to be assimilated, the chemical composition of the magma,and the temperature of the magma all play important parts in theprocess. Conditions become more favorable for assimilation withincreasing depth, higher initial temperature and lower meltingtemperature of the assimilated material, and increasing tempera-ture of the magma. Thus, with sufficient depth and temperature,there can be no doubt as to the effectiveness of assimilation. As-similation alone will not explain the character of the rocks ob-served in the White Mountain magma series. By the assimilationof sil iceous sediments and granites it would be possible to accountfor a series from gabbro through possibly monzodiorite, but toproduce the syenites and granites, it would be necessary to assim-ilate prohibitive amounts of material. Moreover, mixing an oldergranite with gabbro could never produce a granite.
Many of the objections to assimilation, such as the heat prob-lem, also apply to pure-melting as the sole means of deriving theWhite Mountain magma series. Pure-melting may be completeor partial. Complete melting of older rocks can not explain theseries. There are no known older rocks in the region with the samecomposition as the White Mountain magma series. Moreover, incomplete melting granite would be generated first, gabbro lastand the sequence would be precisely the reverse of that observed.Selective fusion of older rocks as the sole process is also eliminatedfor the last reason given above.
Trrponv ol Rocr Evor-urroN AccBpreo
From a consideration of all the foregoing facts, it appears thatalthough fractional crystallization alone is qualitatively capable ofproducing all the rock types of the White Mountain magma series,quantitatively it is not adequate. Too great an amount of theparent magma is required to produce all the siliceous differentiates.It has also been shown that assimilation alone cannot possibly ex-plain the origin of certain of the more sil iceous rocks oI the series,and that pure-melting as the sole process is open to the same ob-
JOT]RNAL MINERALOGICAL SOCIETY OF AMERICA 527
jections as assimilation and to several others. In view oI these facts,
the writers have concluded that fractional crystallization and
assimilation have gone on together with selective fusion, and that
each has helped to produce the rock sequence of the magma series.
Qualifications as to the depth at which differentiation by frac-
tional crystall ization took place are necessary. According to Stokes'Law, and assuming that the viscosities given by Daly (I2, p. 193)for moderate depths in the earth are of the right order of magni-tude, the rate of settl ing of an olivine crystal 0.5 mm. in diameter,at depths of from 30 to 40 miles in the crust, is practically negligiblein the time available. Thus, fractional crystallization becomes im-portant only at some distance above the substratum. It is impossi-ble to calculate this distance because there are no available dataon viscosities at various depths in the crust.
On lhe basis of the facts and assumptions discussed above, theevolution of the White Mountain magma series is believed to haveresulted from the invasion of the crust by basaltic magma, largelyby stoping; melting and assimilation of the stoped blocks; andfractional crystallization of the original magma and the liquidsformed by melting and assimilation. The evolution was complexin that these processes were operating essentially simultaneouslyat different levels in the crust. Al the deeper levels, pure-meltingand assimilation were important, because there sufficient heat wasavailable lor those processes. Higher in the crust, fractional crys-tallization was going on-at first in the basaltic magma alone, andIater in the liquids formed by selective fusion and assimilation,which moved upward continuously as they formed. The gabbrosand diorites were probablyderived directlyfrom the original basalt-ic magma, but the other rock types resulted from fractional crys-tallization of the original magma contaminated by assimilated andmelted older rocks. The sequence, however, was definitely con-trolled by fractional crystallization.
The syntectic magma underwent fractional crystallization anddifferentiation in the following way. Due to early separation ofcalcic feldspars and a failure of reaction, the alkali feldspars con-centrated in the late differentiates. Similarly, due to the separationof magnesia-rich ferromagnesian minerals, iron increased relativeto magnesia in the later differentiates. Quartz had a dual origin.In part it appeared due to the separation of silica-deficient femags.Much of the quartz, however, is merely older quartz which was
528 THE AMERICAN TIINERALOGIST
present in the pre-existing granites and schists, was incorporatedin the syntectic magma and then concentrated in the later differ-entiates.
NerunB oF rrrD Nlacue CuaMsnn
The exact size, shape and location of the chamber or chambersin which differentiation occurred is problematical. One hypothesispictures a single, Iarge chamber, roughly circular in ground-planand underlying practically aII of central and northern New Hamp-shire and part of eastern Vermont and western Maine. Locally,however, there were cupolas extending up from the main chamberto within a few miles of the surface. These cupolas, which existeddirectly beneath the areas now containing rocks of the magmaseries, were the chief agents localizing the cauldron subsidenceand igneous intrusion. An alternative theory is possible. ft may beassumed that instead of one large magma chamber at depth withcupolas extending up from it, there was a separate, narrow, cylin-drical chamber underlying each local area of intrusion.
FurunB Wonr
The problem of the evolution of the White Mountain magmaseries is, of course, intimately tied up with similar problems inother areas. Information gathered elsewhere will shed light on ourproblem in the White Mountains. The work of the Geophysi-cal Laboratory at Washington is also of prime importance, andgives us much of our most significant data. More specifically in theWhite Mountain district the following additional data should begathered:
(1) Analyses of more minerals, particularly those ends of thepyroxene, amphibole, and olivine series which are as yet notanalyzed. The writers are of the opinion that additional mineralanalyses are of much more importance than rock analyses.
(2) Additional field data on the age relations of the plutonicrocks. The chronological position of some of the intrusive rocks isstill unknown.
(3) Additional data on the size, shape, and mechanics of intrusionof the rock bodies. Large areas of the White Mountain magmaseries are exposed in the Crawford Notch, Plymouth, and MountWashington quadrangles, but they have not been studied sincethe Hitchcock Survey sixty years ago. The Division of GeologicalSciences of Harvard University will probably undertake this studyduring the next few years.
JOURNAL MINERALOGICAL SOCIETY OF AMERICA 529
( ) The importance of deuteric and hydrothermal processes
should be considered much more fully.(5) The genetic relations of the volcanic and dike rocks to the
plutonics must be considered.(6) Seimological and other data may give more information as
to the depths to which the granite and syenite masses extend.
Physical data of importance include:(1) Viscosity of magmas under high pressure, in order to deter-
mine the rate of crystal settling under the influence of gravity.(2) Investigation in the laboratory of the presumed liquidus-
solidus relation existing between iron and magnesia in the ferro-
magnesian minerals as proposed in this paper.
Brrr,rocnnpuv1. Barth, T. F. W., Die Pegmatitgange der taledonischen Intrusivgesteine im
Seiland-Gebiete: Skrift. Norsk. Videnskaps Aka.d., no.8, Oslo, 7927.
2. Becke, F., Die Eruptivgebiete des Bohm. Mittlegebirges und der Amerik'
Andes: T.M.P.M., vol. 22, p. 247, 1903.
3. Billings, M. P., The Petrology of the North Conway Quadrangle in the White
Mountains of New Hampshire: Proc. Am. Acad'. Arls Sci, vol.63, no. 3, pp.
67-137, map,1928.
4. Billings, M. P., The chemistry, optics and genesis of the hastingsite group of
amphiboles: Am Mi,neral., vol. 13, pp. 287-296, 1928.
5. Billings, M. P., Paleozoic Age of the Rocks of Central New Hampshire: Science,
vol .79, pp. 55-56, 1s34.
6. Billings, M. P., Geology of the Littleton and Moosilauke Quadrangles, New
Hampshire, i n fre paro lion.
7. Bowen, N. L , The Evolution of the Igneous Rocks, Princeton,1928.
8. Bowen, N. L., and Schairer, J. F., The System MgO-FeO-SiOz: Am f ow. Sci.,
5th. Ser. , vol .29, pp. 151-217, 1935.
9. Chapman, R. W., Geology of the Percy Quadrangle, New Hampshire: Doctorate
thesis, Harvard University, 1934,10. Clough, C. T., Bailey, E. B., Wright, W. B., and Others, Tertiary and Post-
Tertiary Geology of MulI: Mem. GeoI. Soc. Scotland, 1924.
ll. Daly, R. A., Origin of the Alkaline Rocks: Bull. Geol'. Soc. Am.,vol.2l,pp.
8 7 - 1 1 8 , 1 9 1 0 .72. Daly, R. A., Igneous Rocks and the Depths of the Earth. McGraw-Hill Book
Co.. New Yorh. 1933.13. Evans,J.W.,TheOrig inof theAlkal ineRocks:GeoLMag.,vol .57,p.559,192O.14. Fenner, C. N., The Katmai Magmatic Province: Jour. Geol'., vol. 34, no. 7,
pp. 673-772, 1926.15. Foye, W. G , Nephelite Syenites of Haliburton County, Ontaio: Am' Jour. Sci.,
p . 413 , 1915 .16. Gillson, J. L., On the Origin of Alkaline F.ocks: Jour. GeoI., vol.36, no. 5, pp.
47t-474,1928.17. Goldschmidt, V. M., Stammentypen der Eruptivgesteine: Vi.dens. Shrifi. I,
Mat.-Natun. Kl,asse, no.l0, pp. l-11,1922.
530 TH E AM ERICAN ],1 I NERALOGI ST
18. Grieg, J. W., Immiscibility in Silicate lVlelts: Am. Jour. Sci, 5th. ser., vol 13,pp. 1-14, 133-154, 1927.
19. Harker, A., The Natural History of Igneous Rocks, I[ea, Yorh. 19A9.20 Howes, G. W., On a group of dissimilar eruptive rocks in Campton, New Hamp-
shfte: Am. Iour. Sci.3rd. ser. vol. 17, pp. 117-151,1879.21. Holmes, A., The Tertiary Volcanic Rocks of the District of Mozambique.
Quart. Jour. Geol Soc., p. 252, 191622. Holmes, A., and Harwood, Voicanic Rocks Near Ruwenzori: Quart. four.
Geol. Soc., Lond,on, pp. 370 442, 1932.23. Jenks, W. F., Petrology of the Alkaline Stock at Pleasant Mountain, Maine,
Am. Jour. Sci., vol 28,pp.327-340, 1934.24. Jensen, H. f., The Distribution, Origin, and Relationships of the Alkaline
Rocks: Proc. Linn Soc. N. S. Wales, vol. 33, pp. 535-586, 1908.25. Kennedy, W. Q., Trends of Differentiation in Basaltic Magmas: Am.Jour.Sci.,
vol. 25, no. 147 , pp. 239-256, 1933.26. Kingsley, L., Cauldron-Subsidence of the Ossipee Mountains: Am. Jour. Sci.
5th. ser. , vol 22, pp. 139-168, 1931.27. Miller, F S., Petrology of the San Marcos Mountain Gabbro, San Luis Rey
Quadrangle, California : Doctorate thesis, Harv ard University, 1934.28. Modell, D., The Geology and Petrology of the Relknap Mountains Area, New
Hampshire, Doctorate thesis, Harvard Universitl', 1933.29. Page, A., Geology of the Sandwich Range, New Hampshire, i.n preparotion.30. Perry, J. H., Notes of the Geology of Mt. Kearsarge : J our. Geol., voI. ll, pp.
403472, 1903.31. Pirsson, L. V., and Washington, H. S , Contributions to the Geology of New
Hampshire, I, Geology of the Belknap Mountains: Am Jour. Sci.,4th. ser ) vol.2O, pp. 344-3 52, 1905 ; vol. 22, pp. 439-457, +93-514, 1906.
32. Pirsson, L. V., and Washington, H. S., Contributions to the Geology of NewIlampshire, III; On Red Hill, M otlltolboto: Aln. Jota. Sci., vol. 23, pp. 257-276,43s-447, 1907.
33. Pirsson, L. V., and Rice, W. N , Contributions to the Geology of New Hamp-shire, IV, Geology of Tripyramid Mountain: Am. Iour. Sci.4th. ser., vol.3l,pp. 269-291, 405-43r, 1911.
34. Shand, S. J., The Problem of the Alkaline Rocks: Proc Geol. Soc. So. AJrico,pp. 19*32, 1922.
35. Shand, S. J., Limestone and the Origin of Feldspathoidal Rocks: Geol. Mog.,vol .67, no 795, Sept. 1930
36. Smyth, C. H., Chemical Composition of Alkaline Rocks, and its Significance-asto their Origin: Am. Jour. Sci.,4th. ser., vol. 36, pp. 34-36, 1913.
37. Smyth, C. H., The Origin of Alkaline Rocks: Am. Phit. Soc Proc. vol.66, 192738. Vogt, J. H. L., Die Silikatschmelzlasungen, I, Uber die Mineralbildung in
Silikalschmelzlasungen: Vid.ens. Shrift, I,Math. Naltra. Klasse, no. 8, 1903.39. Vogt, J. H. L, The Physical Chemistry of the Magmatic Differentiation of
Igneous Rocks: Videns. Shrift., I, Math. Nottal. Kl,asse, no 15, pp. 123-128,1924.
40. Williams, C. R, Geology of the Franconia Region, New Hampshire: Doctorotethesis, Hatvard University, I 934.
41. Winchell, A. N., Elements of Optical Mineralogy, Pt. II, pp. 424, John Wileyand Sons, Neut Yorh, 1927.