scientific journal experiments with microform

3
pressing numbered keys on the top of the device. As to possible general users of Data phone service and the new device, Mr. Sevebeck mentioned salesmen, credit card users, and businessmen as among those who might eventually benefit directly by the convenience of the service. It would connect them to data processing centers in stores, main offices, and warehouses where orders would be recorded, merchan- dise routed, checks written, and bills made out—all automatically. Whether data fed into the computers consisted of a lengthy order for goods originating hundreds of miles from the home office, or the time an employee re- ports for work and leaves, or details of a sale across the counter, the transaction would be processed to completion auto- matically and error free. Mr. Sevebeck and his Western Electric assistants, P. J. Grunfelder and M. V. Dilorio, were aided by Bell Telephone Laboratories and the American Telephone & Telegraph Company in the development of this new device to utilize Dataphone service. Computer with Special Generator for Accurate Missile Firing A compact computer, utilizing a unique adjustable nonlinear function generator, has been developed by the U. S. Army Signal Research and Development Lab- oratory, Fort Monmouth, N. J., to over- come the effects of near-surface winds on missile accuracy. Because of the relatively low velocity of a missile immediately after firing, near-surface winds have a considerable effect upon its trajectory. To insure that the missile hits a predicted impact point, low-altitude winds must be determined accurately. Pilot balloon runs are made shortly before a firing to measure these winds, and to provide data from which their effect can be computed. A special computer, which can deter- mine ballistic wind effects rapidly and accurately direct from pilot balloon data, iliil^SH^Ki «l "VERNISTAT" special function generator can be adjusted quickly and easily to provide any mathematical or empirical function, in- cluding those with multiple slope reversals. Two of these devices, developed by the Vernistat Division of Perkin-Elmer Corp., are incorporated in a special computer developed by the U. S. Army Signal Research and De- velopment Laboratory, to overcome the effects of near-surface winds on missile accuracy. incorporates two adjustable nonlinear function generators developed by the Vernistat Division of the Perkin-Elmer Corporation, Norwalk, Conn. These de- vices make possible adjustments in non- linear inputs to the computer with a minimum of time and calculation. Prior to the computer's development, the necessary computations were done manually. Often, by the time the com- putations were completed, a new set of weather conditions had developed. The computer displays the data in the form of an X and Y co-ordinate change (north-south and east-west). It provides for an automatic data input from a theodolite, which enables continuous computations. These calculations are com- pleted simultaneously with the comple- tion of each pilot balloon run. The theodolite data consist of two signals, an angle of elevation and an azimuth read- ing. Other input data are an assumed rate of rise of the test balloon, and a function which relates the effect of the wind on the balloon to the effect of the wind on the missile at various altitudes. This latter function is the rho factor. The balloon's position from the launch- ing site is determined along the north- south and east-west axes from the ground range and azimuth. The rho factor con- tinuously converts this to a correction displacement. The computation then is integrated to give net displacement from the "no wind" impact point. The computer automatically reads out how many miles north-south and east- west the point of impact of the missile will be offset by the ballistic wind. This information is superimposed upon other data characteristics of the particular mis- sile's trajectory, enabling launching tilt adjustments. Different test balloons have various rates of rise, and these ballistic wind effects upon the rho factor vary with different missiles. Thus, it is necessary to have an easy and quick method of changing the input signals in the com- puter. In the new Army computer, this is accomplished through the use of two "Vernistat" Adjustable Non-Linear Func- tion Generators. These devices enable adjustments of nonlinear inputs to the computer with a minimum of time and calculation. Each of the function genera- tors has a shaft position input propor- tional to time. The output of the first function generator is a voltage repre- senting the balloon's altitude. The second furnishes the rho factor. The Vernistat takes only a few seconds to change one function to another, giving the computer an extreme amount of adaptability. The high resolution and low quadrature of the Vernistat precision potentiometers em- ployed with the function generators are consistent with the high performance requirements of the computer. Scientific Journal Experiments with Microform Experimental publication of a scientific journal exclusively in microform—a gen- eral term referring to various methods of microfacsimile presentation including mi- crofilm, Microcards, Microprint, micro- fiches, etc.—is being conducted for the next 3 years by the American Institute of Bio- logical Sciences (AIBS), Washington, D.C. Under assistance grants from the Coun- cil on Library Resources of Washington, D. C, and the National Science Founda- tion, the journal, Wildlife Disease, the official publication of the Wildlife Disease Association, began publication as a quar- terly in January 1959. The purpose of the experiment is to determine (1) whether a small specialist group, unable to afford a journal in letter- press, can do so with microform; (2) whether a microform journal will serve the purpose of scientific communication for author, reader, and library; (3) whether this technique will expedite pub- lication of the results of research; (4) whether, through cost reduction, less abridgement of important data will be necessary; (5) whether photographic re- production will lend itself to superior presentation of photographic data over half-tone reproduction; and (6) what op- tima can be found in terms of microtext medium, page size and arrangement, etc. The journal is published on 5- by 3- inch Microcards, manufactured by the Microcard Corporation, West Salem, Wis., each issue comprising about 4 cards. Each card contains a single article of up to 47 pages in microtext, but bears in full-size type the citation of author, title, and issue number. A leaflet in full-size type accom- panies each issue, containing abstracts of the articles; these are reported to Bio- logical Abstracts, and it is anticipated, in consequence, that they will not need to be retained permanently. Optical devices are needed to read the journal, but these are nonportable and expensive. One of the objectives of the experiment, therefore, is to test the ap- plicability of a small, portable, and inex- pensive hand-viewer, which is being pro- vided to the original members of the As- sociation at a nomnal charge, and will be available to later members at a cost ex- pected to be less than $10. Although the experiment is being conducted initially with Microcards, it is anticipated that other forms of microtext may later be compared. Announcement on the International Yard and Pound The Directors of the following standards laboratories: Applied Physics Division, National Research Council, Ottawa, Can- ada; Dominion Physical Laboratory, Lower Hutt, New Zealand; National Bu- reau of Standards, Washington, U. S.; Na- tional Physical Laboratory, Teddington, U. K.; National Physical Research Labo- ratory, Pretoria, South Africa; and Na- tional Standards Laboratory, Sydney, Australia, have discussed the existing dif- ferences between the values assigned to the yard and to the pound in different countries. To secure identical values for each of these units in precise measure- ments for science and technology, it has been agreed to adopt an international yard 400 Of Current Interest ELECTRICAL ENGINEERING

Upload: leminh

Post on 09-Mar-2017

219 views

Category:

Documents


6 download

TRANSCRIPT

Page 1: Scientific journal experiments with microform

pressing numbered keys on the top of the device.

As to possible general users of Data phone service and the new device, Mr. Sevebeck mentioned salesmen, credit card users, and businessmen as among those who might eventually benefit directly by the convenience of the service. It would connect them to data processing centers in stores, main offices, and warehouses where orders would be recorded, merchan­dise routed, checks written, and bills made out—all automatically.

Whether data fed into the computers consisted of a lengthy order for goods originating hundreds of miles from the home office, or the time an employee re­ports for work and leaves, or details of a sale across the counter, the transaction would be processed to completion auto­matically and error free.

Mr. Sevebeck and his Western Electric assistants, P. J. Grunfelder and M. V. Dilorio, were aided by Bell Te lephone Laboratories and the American Telephone & Telegraph Company in the development of this new device to utilize Dataphone service.

Computer with Special Generator for Accurate Missile Firing

A compact computer, utilizing a unique adjustable nonlinear function generator, has been developed by the U. S. Army Signal Research and Development Lab­oratory, Fort Monmouth, N. J., to over­come the effects of near-surface winds on missile accuracy.

Because of the relatively low velocity of a missile immediately after firing, near-surface winds have a considerable effect upon its trajectory. T o insure that the missile hits a predicted impact point, low-altitude winds must be determined accurately. Pilot balloon runs are made shortly before a firing to measure these winds, and to provide data from which their effect can be computed.

A special computer, which can deter­mine ballistic wind effects rapidly and accurately direct from pilot balloon data,

iliil SH Ki «l "VERNISTAT" special function generator can

be adjusted quickly and easily to provide

any mathematical or empirical function, in­

cluding those with multiple slope reversals.

Two of these devices, developed by the

Vernistat Division of Perkin-Elmer Corp. , are

incorporated in a special computer developed

by the U. S. A rmy Signal Research and De ­

velopment Laboratory, to overcome the effects

of near-surface winds on missile accuracy.

incorporates two adjustable nonlinear function generators developed by the Vernistat Division of the Perkin-Elmer Corporation, Norwalk, Conn. These de­vices make possible adjustments in non­linear inputs to the computer with a m i n i m u m of time and calculation.

Prior to the computer's development, the necessary computations were done manually. Often, by the time the com­putations were completed, a new set of weather conditions had developed.

T h e computer displays the data in the form of an X and Y co-ordinate change (north-south and east-west). It provides for an automatic data input from a theodolite, which enables continuous computations. These calculations are com­pleted simultaneously with the comple­tion of each pilot balloon run. T h e theodolite data consist of two signals, an angle of elevation and an azimuth read­ing. Other input data are an assumed rate of rise of the test balloon, and a function which relates the effect of the wind on the balloon to the effect of the wind on the missile at various altitudes. This latter function is the rho factor.

T h e balloon's position from the launch­ing site is determined along the north-south and east-west axes from the ground range and azimuth. T h e rho factor con­tinuously converts this to a correction displacement. T h e computation then is integrated to give net displacement from the "no wind" impact point.

T h e computer automatically reads out how many miles north-south and east-west the point of impact of the missile will be offset by the ballistic wind. Thi s information is superimposed upon other data characteristics of the particular mis­sile's trajectory, enabling launching tilt adjustments.

Different test balloons have various rates of rise, and these ballistic wind effects upon the rho factor vary with different missiles. Thus , it is necessary to have an easy and quick method of changing the input signals in the com­puter. In the new Army computer, this is accomplished through the use of two "Vernistat" Adjustable Non-Linear Func­tion Generators. These devices enable adjustments of nonlinear inputs to the computer with a minimum of time and calculation. Each of the function genera­tors has a shaft position input propor­tional to time. T h e output of the first function generator is a voltage repre­senting the balloon's altitude. T h e second furnishes the rho factor. T h e Vernistat takes only a few seconds to change one function to another, giving the computer an extreme amount of adaptability. T h e high resolution and low quadrature of the Vernistat precision potentiometers em­ployed with the function generators are consistent with the high performance requirements of the computer.

Scientific Journal Experiments with Microform

Experimental publication of a scientific journal exclusively in microform—a gen­eral term referring to various methods of

microfacsimile presentation including mi­crofilm, Microcards, Microprint, micro­fiches, etc.—is being conducted for the next 3 years by the American Institute of Bio­logical Sciences (AIBS), Washington, D.C.

Under assistance grants from the Coun­cil on Library Resources of Washington, D. C , and the National Science Founda­tion, the journal, Wildlife Disease, the official publication of the Wildlife Disease Association, began publication as a quar­terly in January 1959.

The purpose of the experiment is to determine (1) whether a small specialist group, unable to afford a journal in letter­press, can do so with microform; (2) whether a microform journal will serve the purpose of scientific communication for author, reader, and library; (3) whether this technique will expedite pub­lication of the results of research; (4) whether, through cost reduction, less abridgement of important data will be necessary; (5) whether photographic re­production will lend itself to superior presentation of photographic data over half-tone reproduction; and (6) what op­tima can be found in terms of microtext medium, page size and arrangement, etc.

T h e journal is published on 5- by 3-inch Microcards, manufactured by the Microcard Corporation, West Salem, Wis., each issue comprising about 4 cards. Each card contains a single article of up to 47 pages in microtext, but bears in full-size type the citation of author, title, and issue number. A leaflet in full-size type accom­panies each issue, containing abstracts of the articles; these are reported to Bio­logical Abstracts, and it is anticipated, in consequence, that they will not need to be retained permanently.

Optical devices are needed to read the journal, but these are nonportable and expensive. One of the objectives of the experiment, therefore, is to test the ap­plicability of a small, portable, and inex­pensive hand-viewer, which is being pro­vided to the original members of the As­sociation at a nomnal charge, and will be available to later members at a cost ex­pected to be less than $10. Although the experiment is being conducted initially with Microcards, it is anticipated that other forms of microtext may later be compared.

Announcement on the International Yard and Pound

T h e Directors of the following standards laboratories: Applied Physics Division, National Research Council, Ottawa, Can­ada; Dominion Physical Laboratory, Lower Hutt , New Zealand; National Bu­reau of Standards, Washington, U. S.; Na­tional Physical Laboratory, Teddington, U. K.; National Physical Research Labo­ratory, Pretoria, South Africa; and Na­tional Standards Laboratory, Sydney, Australia, have discussed the existing dif­ferences between the values assigned to the yard and to the pound in different countries. T o secure identical values for each of these units in precise measure­ments for science and technology, it has been agreed to adopt an international yard

400 Of Current Interest ELECTRICAL ENGINEERING

Page 2: Scientific journal experiments with microform

and an international pound having the following definition: the international yard equals 0.9144 meter; the international pound equals 0.45359237 kilogram.

It has also been agreed that, unless otherwise required, all nonmetric cali­brations carried out by these laboratories for science and technology on and after July 1, 1959, will be made in terms of the international units as previously de­fined or their multiples or submultiples.

T h e preceding announcement is being made concurrently by all the laboratories listed above. T h e follow­ing paragraphs provide explanatory material concerning the background and use of the new units within the United States.

T h e international inch, derived from the international yard, is exactly equal to 25.4 millimeters. Thi s value for the inch has been legally adopted by Canada. Also this value was approved by the American Standards Association for "Inch-mill imeter conversion for industrial use" in 1933 (American Standard B48 .1-1933), was adopted by the National Ad­visory Committee for Aeronautics in 1952, and has been adopted by many standardiz­ing organizations in other countries.

At present, for the calibration of l ine standards and end gauges having nominal lengths expressed in inches, the National Bureau of Standards is using the inch defined by the Mendenhall order (Fun­damental Standards of Length and Mass, Bulletin No. 26, United States Coast and Geodetic Survey by T . C. Mendenhall) published in 1893. T h e values correspond­ing to this order are approximately: 1 yard equals 0.91440183 meter; and 1 inch equals 25.4000508 millimeters. These are derived from the exact relation: 1 yard equals 3,600/3,937 meter. T h e inch used by the National Physical Laboratory of the United Kingdom for its calibrations is defined by the equation 1 inch equals 25.399956 mm. It will be noted that the International Inch is approximately 2 parts per mil l ion shorter than the inch presently used by the National Bureau of Standards, and somewhat less than 2 parts per mil l ion longer than the inch now used by National Physical Laboratory. T o avoid possible confusion, during the transition period National Bureau of Standards calibrations of length or mass expressed in English units will embody a statement indicating clearly the unit which has been used if the choice intro­duces a significant difference in the cali­bration values. Furthermore, if the accu­racy of the calibration is such that the certified values would be the same in either "International" units or the older units, the qualifying adjective "Interna­tional" will not be used, i.e., the values will be expressed, for example, as so many inches or pounds.

T h e Coast and Geodetic Survey has re­quested the following exception with which the National Bureau of Standards concurs.

"Any data expressed in feet, de­rived from and published as a result of geodetic surveys, shall tacitly bear

the relationship: One foot equals 1,200/3,937 international meter. Th i s relationship shall continue in being, for the purpose given herein, until such a time as it becomes desirable and expedient to readjust the basic geodetic survey networks in the United States, after which the ratio, as implied by the international yard, shall apply."

Thi s unit shall be referred to as the American Survey Foot. Inasmuch as there is little or no interchange of survey data, where the foot measurements are used, with industrial and scientific data, where the international units will be used, it is anticipated that no confusion will result from this dual usage. For example, base line surveys which might enter into a velocity of light determination would in­variably be made in terms of meters.

T h e values of the pounds currently in use in the United States, United Kingdom, and Canada are as follows:

1 U. S. pound = 0.453 592 4277 kg 1 British pound = 0.453 592 338 kg I Canadian pound = 0.453 592 43 kg 1 International

pound = 0.453 592 37 kg

T h e relative differences in the various pounds are substantially less than in the yards but since masses can be measured with greater accuracy than lengths, the differences can be significant. T h e present British pound is about one part in ten mill ion smaller than the international pound, whereas the U. S. and Canadian pounds are about one and one-half parts in ten mil l ion larger.

T h e conversion factor for the interna­tional pound was selected so as to be ex­actly divisible by 7 to give the following value for the grain: 1 International grain = 0.06479891 gram.

T h e grain is the common unit in avoir­dupois, apothecary, and troy pounds. There are 7,000 grains in the avoirdupois pound, and 5,760 grains in both the apothecary pound and the troy pound.

T h e standard United States gallon and the Imperial gallon are so substantially different that a compromise international gallon was not practicable. T h e United States gallon is defined as equal to 231 cubic inches. On the other hand the Im­

perial gallon is defined as the volume of 10 pounds of water under specified standard conditions. A fairly exact relationship is 1 Imperial gallon = 1.20094 U. S. gallons, or less exactly 1 Imperial gallon = 6 /5 U. S. gallons.

Electronic Control of Coating and Combining Unit

T h e fabrication of pressure-sensitive tapes, adhesives, and allied rubber prod­ucts has been put into a production basis at the modern facilities of the new Johns-Manville Dutch Brand plant in Chicago, 111., according to company engineers.

T h e new facilities utilize the latest mixing, coating, slitting, and combining equipment now available in the rubber and chemical industries in an electroni­cally controlled, automatic system of man­ufacture.

Increasing former production capacity by more than 50 per cent, the new plant has already made possible the fabrication of a line of entirely new and improved products, including paper masking tapes, strapping tapes, paper electrical thermo­setting and nonthermosetting tapes, My­lar (polyester) electrical tapes, cloth ther­mosetting electrical tapes, transparent acetate electrical tape and cloth non­thermosetting electrical tapes.

It is based on a single-story, straight-line layout with the flow of materials going directly from a mixing room on the north side to a shipping area on the south side.

T h e coating equipment, built to Johns-Manville specifications by the Frank Egan Company, is completely electrically con­trolled and operated with application set­tings as fine 3/10,000th of an inch.

T h e release coater, capable of coating 48-inch and 60-inch wide webs at a speed of 40 yards per minute, is of the print roll, two-temperature-zone type with an over-all oven length of 70 feet.

T h e anchor coater is also of the print roll type capable of coating 48-inch and 60-inch wide webs at a speed of 40 yards a minute. A 40-foot oven is provided for drying in a single temperature zone.

A new adhesive coater is supplied witb two coating heads, one of the reverse roll type, a second of the knife over-roll type. Both heads are electronically controlled

E L E C T R O N I C C O N ­

TROLS govern mecha­

nized manufactur ing

progress at the Johns-

Manv i l le new Dutch

Brand Division plant

in Chicago, III. The

system of electronic

controls, built to speci­

fication by the Frank

Egan Company , per­

mit coating appl ica­

tions as fine as 3 /

10,000ths of an inch

at speeds as h igh as

4 0 feet per minute.

APRIL 1 9 5 9 Of Current Interest 401

Page 3: Scientific journal experiments with microform

and are designed for interchangeable op­eration. Three temperature oven zones provide 120 feet of drying area. In addi­tion, the adhesive coater is supplied with a 20-foot refrigeration zone for cooling purposes.

All coating lines are equipped with safety controlled motor drives synchro­nized to the main electronic control pan­els to provide for efficient operation.

Slitting of the coated base materials is accomplished on both rotary-knife slitters and slicing-cutters depending upon the type and specifications of the tape man­ufactured.

Specially designed and custom-engi­neered tapes are cut and wound on an electronically controlled Dusenbery slitter. This equipment allows for special tension settings and speeds and is expected to play an important role in the development of many new tapes which could not be handled on other types of slitting equip­ment.

One of the important new processes at the new plant is a unique "combining" unit. It is the first equipment of its type ever designed from the factory floor up to the specific job of laminating parallel strands of organic or synthetic materials on flexible sheets required in fabricating new and improved products.

Essentially the laminating of glass fiber or other synthetic or organic yarns to other materials on flexible flat sheets to give them added strength and new uses, in the process, various materials such as paper, plastic films, or metal foils are bonded under pressure to fine strands of yarn ranging from as few as 12 or as many as 100 parallel strands per inch, depend­ing on the strength desired or other fac­tors.

Typical of Dutch Brand materials de­signed to meet customer requirements in the nation's rapidly changing technology is a broad range of insulating tapes for manufacturers of electrical equipment and instruments; gasket materials and mask­ing tapes for the automotive industry; tapes for sealing duct insulations; and adhesives for the construction industry.

Reinforced Mylar sheets have been pro­duced on the combining equipment for use as insulators and supports for mica splits used by industry. Metal foil, rein­forced by parallel plastic strands and laminated to asbestos paper, has been fabricated as an exceptionally effective,

strong and dimensionally stable pipe wrap. Rubber sheeting, reinforced with parallel plastic strands, has been made experimentally on the unique equipment as a protection for service entrance cables.

One new material developed on the combining machine was originally created to protect U.S. Navy shipboard cable. In this case, fire-retarded asbestos paper as thin as cigaret paper, and having unique dielectric properties, was lami­nated with parallel strands of glass fibers, 15 strands to the inch. It made an effective cable covering and, with slight variations, can be utilized in producing other new materials.

Powder Pattern Technique Shows Ferroelectric Domains

A new method for delineating the domain structure at the surfaces of ferro­electric crystals was described recently by G. L. Pearson and W. I,. Feldman of Bell Te lephone Laboratories. It has been used to delineate domain structures in great detail on a wide variety of crystals in which domains have never been observed before, according to a paper presented at the American Physcial Society meeting in Chicago, 111.

According to its discoverers, the method should be applicable to the study of piezoelectric and pyroelectric charged sur­faces as well as ferroelectric materials.

T h e new technique uses colloidal sus­pensions of electrostatically charged powders in an insulating organic l iquid. When a few drops of this suspension are applied to the face of the crystal, the charged powder is immediately attracted to the ferroelectric domains carrying an opposite charge, and covers their entire area.

T h e most effective materials for this work are commercial spray-grade sulfur and red lead oxide, each suspended in hexane. T h e sulfur deposits on negatively charged domains, while the lead oxide deposits on the positively charged do­mains. Each of the suspensions is applied separately; the second is not applied until the hexane in the first has evaporated. If the two suspensions mix on the crystal surface, each may lose its charge. In this event, no pattern is formed. T h e powders

are fixed in place indefinitely by their electrostatic charge after the hexane evaporates.

T h e yellow sulfur and red lead oxide provide brightly colored delineation of the positive and negative domains, and the pattern shows great detail.

A dispersion of a cross-linked polymer derived from polystyrene can also be used as the negatively charged colloid, in place of the lead oxide. This can be dyed any desired color with an oil-soluble dye.

T h e basis for the success of the new technique lies in the fact that although the colloid as a whole is electrically neu­tral, individual particles acquire a diffuse, double-layer charge when brought in con­tact with the liquid. Under the influence of the "built-in" electric field, the colloidal particles are attracted either to the posi­tive or negative domains depending on the orientation of their dipole layers. For these experiments, an insulating l iquid of low viscosity and low dielectric constant such as hexane is desirable, so that the charged particles are free to move toward the ferroelectric domains under maximum electrostatic attraction.

Powder patterns of an entirely different nature result if a colloid of finely divided barium titanate in hexane is used instead of sulfur and lead oxide. T h e BaTiO^ particles carry no net electric charge, but they do have a very high dielectric con­stant. Thus , when they are placed on the surface of a ferroelectric crystal contain­ing positive and negative domains, the particles polarize in the external electro­static fields and are deposited at domain boundaries where large field gradients exist. T h e resulting powder patterns out­l ine the domain boundaries well, but do not differentiate between positively and negatively charged domains.

Th i s new technique has provided the first information available on the domain structure of a number of crystalline ferro­electric materials, including triglycine sul­fate and guanidinium gall ium selenate hexahydrate. Previous optical determina­tions which have been made on Rochelle salt, and domain etching procedures on barium titanate have been confirmed by the new method also.

Thus , a simple, easily employed tech­nique joins the group of tools available to the physical chemist and physicist in their continuing search for knowledge of the ultmate structure of matter.

Sandwich Aluminum Used for New Refrigerators

A method of manufacturing refrigerator cabinets, which requires very little tooling cost and permits profitable production of small quantities of special models, has been developed recently by Westinghouse Electric Corporation, Columbus, Ohio.

T h e new method involves the use of a "sandwich" of a luminum sheets bonded to expanded polystyrene bead insulation. Both built-in and free-standing refriger­ators are being manufactured by the new process.

T h e panels or sandwiches, supplied by the Aluminum Company of America, are

CLOSE-UP of triglycine

sulfate crystal, show­

ing ferroelectric do ­

main patterns, del ine­

ated by new powder

technique developed

by scientists at Bell

Telephone Laboratories.

Crystals in foreground

include other samples

of triglycine sulfate

a n d a hexagonal crys­

tal of gadol in ium a lu ­

minum sulfate hexa­

hydrate (GASH) .

402 Of Current Interest ELECTRICAL ENGINEERING