pani
fe ett iN
Pita Ue Vad fl ie

tf

AS Z

ae 2

\ ee wy Loa
‘ ue) oa

ae
i
Bs
road
Mh ON ab FY
nA

es Le
or

cae
1

——

u

f
He
mm

Bape)

THE

AMERICAN
JOURNAL OF SCLENCE.

Epitrorn: EDWARD S. DANA.

ASSOCIATE EDITORS

Proressors WILLIAM M. DAVIS ann REGINALD A. DALY,
oF CAMBRIDGE,

ProFessors HORACE L. WELLS, CHARLES SCHUCHERT,
HERBERT EH. GREGORY, WESLEY R. COE anp
FREDERICK.E. BEACH, or New Haven,

Proressor EDWARD W. BERRY, or BALtTImMorgE,
Drs. FREDERICK L. RANSOME anp WILLIAM BOWIE,
oF WASHINGTON.

FIFTH SERIES
Vor t= (WHOLE NUMBER, CCl} +

WITH THREE PLATES

oe Gods. &. Fe

NEW HAVEN, CONNECTICUT.

I eae

CONTENTS TO VOLUME IIL.

Number 13.

Page
Arr. I.—Genetic Features of Alnoitic Rocks at Isle Cadieux,
Bebee-by oN. Is DowEN. —(Wath Plate ll) .....2..5..- 1
Arr. IJ.—On Some Natural and Synthetic Melilites; by A. F.
Nee IC HONE ere he i a ena Ae etre pete ax hel oe 5» 30
Art. IIJ.—Carex Notes; by I. W. Croxry. Carex apado sp.
Prem Hbsbiateuld nj ein Se. Ss s be es oe 88

SCLENOPEFIC INTELLIGENCE.

Chemistry and Physics.—New Process for Determining Fluorine, M. TRAVERs :
The Separation of the Element Chlorine into Isotopes, W. D. Harkins
and A. Hayes, 92.—Qualitative Chemical Analysis, O. F. Tower: A
Course of Instruction in Quantitative Chemical Analysis, G. McP. Smita,
93.—American Chemistry, H. Hate: Report on Atomic Structure: Veloc-
ity of Sound at High Temperatures, 94.—Traité de Dynamique, J. D’ ALEM-
BERT: Heat and Light, S. E. Brown, 95.—Physik und Erkenntnisstheorie,
KE. GEHRCKE: Séries Trigonométrique, M.Lrecat: Die Grundlagen der Geo-
metric, L. HEFFTER, 96.

Geology.— A prebistoric Human Skull from Rhodesia, A. S. Woopwarp, 96.—
A new Generic Name for Pliceyon Marshi, M. R. THorPE: Publications of
the United States Geological Survey, G. O. Smiru, 97.

Miscellaneous Scientific Intelligence.— Bibliotheca Zoologica, II, O.TASCHENBERG:
Publications from the Ronald Press Company, 98.—The ‘‘Electrician”’
Diamond Jubilee Number, 99.

PY. 23 CONTENTS

Number 44.

Page
Art. IV.—Gastropod Trails in Pennsylvanian Sandstones
in Dexas; by {2 POWERS =e. 802.62 ne 101
Arr. V.—Seaside Notes; by P. E. Rav noup aber oS 108
Art. VI.—The ‘‘Varve Shales” of Australia; by T. W.
DAW. oie ee See ee See, Sr 115
Arr. VII.—Mineralogy: Augite of Haleakala, Maui, Ha-
wailan Islands; by H. 8. Wasuineron and H. E.
MBER WIN. 2.) cde kee eet aren oi one Seek 117
Art. VIII.—Oligocene Rodents of the Genus Ischyromys;
by. Hi. L. WROMmnE Ss, oe gee ee 123
Art. IX.—Glaciation of the Mountains of Japan; by N.
VAMASAKI. fia 2008 04. (ae oe ioe ee 131
Art. X.—Studies in the Cyperacee; by T. Horm.
XXXIII. Carices aeorastachyz: Macrochetze nob.
and Nesophile nob. (With 12 eeuNes drawn from
nature by the author): 2% sie Se 138

SCTENTIFIC INTE LIGNE.

Chemistry and Physics.—A new Cyanide, W. S. Lanpis: A Very Sensitive
Reaction for Copper, P. THomas and G. CaRprentiER, 145.—Lehrbuch der
Hisenhiittenkunde, B. Osann: Mining Physics and Chemistry, Z. W.
WHITAKER, 146. —The Adsorption of Gas by Charcoal and Silica, 147.—
Tables Annuelles de Constants et Donnés Numériques, 148. —Report of the
National Physical Laboratory for 1920: Die Naturwissenschaften, Zweiter
Band, F. Dannemann, 149.

Geology and Mineralogy.—Geological Survey of Canada, 150. — A Review of
the Evideuce of the Taconic Revolution, T. H. CLARK: Fossilium Cata-
logus, K. LAMBRECHT, etc.: Report of the State Geologist on the Mineral
Industries and Geology of Vermont, 1919-1920: Bernice Pauahi Bishop
Museum of Polynesian Ethnology and Natural History, Honolulu, Hawaii,
151.—Mikroskopische Physiographie der Mineralien und Gesteine, H. Ro-
SENBUSCH, 152.— Morphogenese der Oetscherlandschaft, K. DiwaLp: New
Meteorites, G. P. MERRILL, 153. — Feldspar Studies: Optical, Chemical ©
and Thermal Properties of Moonstone from Korea, S. Kozu and M. Suzuxk1:
A Manual of Determinative Mineralogy with Tables, J. V. Lewis, 164.—A
Text-Book of Mineralogy, E, 8S. Dana and W. E. Forp, 15d.

Miscellaneous Scientific Intelligence.—Dairy Bacteriology, ORLA-JENSEN, 190.
—Report of the Secretary of the Smithsonian Institutlon, C. D. WaucoTt,
for the year ending June 30, 1921, 156.—Publications of the Carnegie In-
stitution of Washington, yo Report of the Librarian of Congress, H.
PutTNaAM, and Report of the Superintendent of the Library Buildings and
Grounds, F. L. AvERILL, for the year ending June 380, 1921, 158.

CONTENTS Vv

ING rea Oxe nea bas

Arr. XI.—Restoration of Blastomeryx marshi; by R. S. Lutz
UNV TLD: Peni TTD NS ga aie OV ie an ea 159

Art. XII.—Oregon Tertiary Canide, with Descriptions of
INowatiornis: bye oR HORE 22 oc. . o20)) tes ee 162

Arr. XIII.—The Crystallographic and Atomic Symmetries
of Ammonium Chloride; by R. W. G. Wyckorr...... Lee

Art. XIV.—The Crystal Structure of Silver Oxide (Ag,O);
Rerum eos x og DVN OOH CAG cre eo SG cee «Seb epee a8 184

Arr. XV.—Carboniferous Plantsfrom Peru; by E. W. Berry 189

Art. X VI. —Melanovanadite, a New Mineral from Mina Ragra,
Pasco, Peru; by W. Linperen, L. F. Haminton and

CUS LE BUCURESTI 195
Art. XVII.—The Ceratopyge .Fauna in Western North
PRaeiie2 wiht ee bik PUA MOND 99252 tor onesies ea gy oe 204

Art. XVIII.—The Pitts Meteorite; by 8S. W. McCatiize... 211

SCIENTIFIC INTELLIGENCE.

Chemistry and Physics.—The Reduction of Ferric Salts with Mercury, L. W.
McCay and W. T. ANDERSON, JR.: Synthetic Gasoline, 216.—The Precipi-
tation of Arsenic as Sulphide from Solutions of Arsenates, J. H. REEpy:
Colorimetric Analysis, F. D. SNELL: Fluid Resistance, C. WIESELSBERGER,
217.—Les Fondaments de la Geometric, H. PoIncars#, 219.

Geology and Mineralogy.—James Hall of Albany, Geologist and Palzontolo-
gist, 1811-1898, J. M. CuarKker, 220.— The Problem of the St. Peter
Sandstone, C. L. Dake: A Geological Reconnaissance of the Dominican
Republic, T. W. VauGuHan and others, 221.— A Manual of Seismology, C.
Davison, 222.—Earth Evolution and its Facial Expression, W. H. Hopss:
West Virginia Geological Survey, I. C. Wuite, 228 —Virginia Geological
Survey. T. L. Watson: Wisconsin Geological and Natural History Sur-
vey, W. O. Horcuxiss: Dana’s Textbook of Mineralogy, Third Edition,
W. i. Forp, 224.—A new Meteoric Iron, 225.

Miscellaneous Scientific Intelligence.—Carneyie Foundation for the Advance-
ment of Teaching, Sixteenth Annual Report of the President, H. S.
PRitcHETT, and Treasurer, R. A. Franks, for the Year ending June 30,
1921, 225.—A Monograph of the Existing Crinoids, A. H. CLarkK, 226.—
Source Book for the Economic Geography of North America, C. C. CoLBy;
Ueber die Vorstellungenen der Tiere: ein Beitrag zur Eutwicklungspsy-
chologie, H.VoLKLetT: Le Movement scientifique contemporain en France :
1, Les sciences naturelles, G. MATISSE, 227.

VI CONTENTS

Number a6:
Page
Arr. XIX.—Minor Faulting in the Cayuga Lake Region;
by EH. DP SLONGy 1). 0 s5 52 2. oe ee 229
Art. XX.—Description of a new Species of Fossil Herring,
Quisque bakeri, from the Texas Miocene; by D.S. Jorpan, 249
Art. XXI.—A New Genus of Fossil Fruit; by E. W. Berry, 251
Art. XXII.—The Relations Between the Purcell Range and
the Rocky Mountains in British Columbia, Canada; by
LL.D. BURLING, . 3.5.2 eee 204
Art. XXIII.—Some Complex Chlorides containing Gold. I.
Pollard’s Ammonium-Silver-Auric Chloride; by H. L.
| Wid, 5. ee a 3 eRe een nce ee 257
ArT. X XITV.—Studies in the Cyperacee; by TuHEro. Horm, 260
ArT. XX V.—Collophane, a Ae Neglected Mineral; by

A. PF. Rocurs, 2.93. 20h ee oe Pee 269
Art. XX VI.—A New Gens of Oligocene Hyzenodontidex; by
MOOR. THORPE, Wee ep eee See oe, ee 270
ART. XXVII.—The Status of Homogalax, with Two New

Species; by Hi. Ii Wrexaninic Vesey er ee ee 288
Art. XX VIII.—A Tantalate and Some Columbites fom Custer

County, South Dakota; by W. i) HuAppEN, ..- .- 2-2 293

SCIENTIFIC INTELLIGENCE.

Chemistry and Physics.—The Separation of the Isotopes of Mercury, G. von
Hrvesy: The Color of Ferric Ammoninm Alum, J. BONNELL and E. P.
PaRMAN, 300.—The Persulphides of Hydrogen, J. H. Watton and L. B.
Parton: A Separation of Germanium from Arsenic, J. H. MUuusr, 301.
—Asymmetry of the Gaseous Molecule, R. Gans, 502.—L’Atome, Dr.
ACHALME: Calculus and Graphs, L. M. Passano: The Manufacture of
Optical Glass and of Optical Systems, F. HE. Wricut, 303.

Geology.— Notes on Artic Ordovician and Silurian Cephalopods, A. F. Forrsts,
304,—Stratigraphy of the Pennsylvanian Formations of North-Central
Texas, F. B. PLUMMER and R. C. Moore: Recent Mollusca of the Gulf of
Mexico and Pleistocene and Pliocene Species from the Gulf States: Hand-
buch der Relionalen Geologie: The Topographic and Geological Survey of
Pennsylvania, G. H. Asuutey, 300.—United States Bureau of Mines, H. F.
Bain, 306.—Metamorphism in Meteorites, 307.

Miscellaneous Scientific Intelligence.—Carnegie Institution of Washington Year
Book, No. 20, 1921, 8307.—Proceedings of the First Pan-Pacific Scientific
Conference: A Laboratory Manual for Comparative Vertebrate Anatomy,
309.—The Vitamins: Publications of British Museum of Natural History,
London, 1921, 510.—Rcegister zum Zoologischer Anzeiger: George Weber’s
Lehr- und Handbuch der Weltgeschichte, 311.—Observatory Publications:
Tables and other Data for Engineers and other Business Men: Bibliotheca
Zoologica II: Mentally Deficient Children, Treatment and Training, 312.

Obituary.—J. C. BRANNER: J. F. BottomtEy: M. VERworRN: G. L. CIMICIAN:
B. W. McFarRuanp: C. W. WAIDNER, 314.

ghd aviesuadonae

CONTENTS WEE

Num ber ir.
Art. XXIX.—Some Complex Chlorides Containing Gold;
DBs due Ai SST Se ee 315
Arr. XXX.—An American Spirulirostra; by E. W. Burry, 327
Art. XXXIJ.—Meteoric Iron from Odessa, Ector Co., Texas;

DT Sat E UTS SIPS oa ce a 335
ArT. XXXII.—Some Sharks’ Teeth from the California
Paineem ean PORDAN. 9.5. 5 a. eos ne eo ease 338

Arr. XXXIII.—An Upper Cambrian Fauna of Pacific Type
in the European Arctic Region; by O. Hotrrpant,.... 343

Art. XXXIV.—The Great Dustfall of March 19, 1920; by
Rene WEN crim andeby: thos MIELER, 0... ..604+...... 349

Arr. XXXV.—Helaletes Redefined; by E. L. TROXELL, ea GD

Art. XXXVI.—Areocyon, a Probable Old World Migrant;
D0 us ine TRE DIRT eS aceite Fe el aaa ee 371

SCIENTIFIC INTELLIGENCE.

Chemistry and Physics.—Hydrated Oxalic Acid as an Oxidimetrie Standard, A.
K. Hitt and J. M. Smita: The Atomic Weight of Beryllium (Glucinum),
~H6niescomip and BIRKENBACH, 9378.—An Introduction to the Physics
and Chemistry of Colloids, E. HatscueK: Distillation Principles and
Processes, S. YounG, 379.—Separation of Isotopes, F. W. Aston, 380.—
Newcomb-Engelmann Popuiare Astronomie, H. LupENpoRFF, 381.—The
Two Orbit Theory of Radiation, F. H. BricELow, 382.

Geology and Mineralogy.—Ueber das Becken, den Schultergiirtel und einige
andere Teile der Londoner Archaeopteryx, B. PETRONIEVICS, 382.—Die
Antike Tierwelt, O. KELLER: Origin and Eyolution of the Human Race,
A. CHURCHWARD, 383.—The Topographic and Geological Survey of Pennsyl-
vania, G. H. Asaiey: Carboniferous Glaciation of South Africa, A. L. pu
Torr: South Australia Geological Report for 1920, L. K. Warp: Mineral
Production in the United States and elsewhere, 384.—The future of the
Comstock Lode, 385.—A new Meteoric Iron, G. P. MERRILL, 386.

Miscellaneous Scientific Intelligence.—National Academy of Sciences: A Text-
book of Zoology, 386.—Reptiles of the World, R. L. Dirmars, 387.,—Mono-
graphia delle Cocciniglie italiane, G. Leonarpi: Fauna Hawaiiensis, D.
SHarp, 388.—World Atlas of Commercial Geology, 389.—The Friendly
Arctic, V. Steransson, 390.—The Evolution of Climates, M. MANSON: An
Essay on the Physiclogy of Mind, F. X. Dercum, 391.—Royal Natural
History Museum at Brussels, 392.

Obituary.—C. JornDAN: B. Moore: A. D. WALKER: T. LigpiscH, 392.

VII CONTENTS

Number 18.

Page
Arr, XX X VII.—A Critical Review of Chamberlin’s Ground-
work forthe Study of Megadiastrophism; by W. F. Jonzs, 393

Arr. XX XVIII.—Some Complex Chlorides Containing Gold.
III. A New Cesium-Auric Chloride; by H. L. Wutts, 414

Art. XX XIX.—A Chromophore Grouping of Atoms in In-
organic Triple Salts, and a General Theory for the Cause
of the Colors of Substances; by H. L. Wexuts,........ 417

Arr. XL.—Some Tertiary Carnivora in the Marsh Collection,
with Descriptions of New Forms; by M. R. Tuorps, .. 423

Art. XLI.—The Dilemma of the Paleoclimatologists; by R.
W.. Saves, Harvard Winiversityoss... =... 2). eae 456

SCIENTIFIC INTELLIGENCE.

Chemistry and Physics.—Heats of Neutralization of Potassium Sodium and
Lithium Hydroxides, etc,, T. W. RicHarps and A. W. Rowe: Rapid
Iodometric Estimation of Copper and Iron in Mixtures of their Salts, I. W.
Wark, 474.—Introduction to Physical Chemistry, J. WALKER: Colloid
Chemistry of the Proteins, W. Pautt, 475.—The Aurora Line of the Night
Sky: The Color of the Sea, Raman, 476.—Proportionality of Mass and
Weight, C. F. Brusu, 477.—The Teaching of General Science, W. L.
EIKENBERG, 478.

Geology and Mineralogy.—Hesperopithecus, the first anthropoid primate found
in America, H. F. OsBorn, 478.—Shallow-water Foraminifera of the Tortugas
Region, J. A. CusHmMan: Fossil Echini of the West Indies, R. T. JAckKson:
Triassic Fishes from Spitzbergen, EH. A. S. Srensid: The Miocene of
Northern Costa Rica, A. A. Ousson, 479.—The Geology of the Corocoro
Copper District of Bolivia, J. T. SINGEWALD, JR., and E. W. Berry: A
Guide to the Fossil Remains of Man, etc.: Illinois Geological Survey, F.
W. DEWo tr, 480.—Illinois State Water Survey: Geological Survey of
Western Australia, A. Grpp MarrLanp, 481.

Miscellaneous Scientific Intelligence.—Washington meeting of the National
Academy of Sciences: Considérations sur l’Etre vivant, 482.—Ftinf Reden
von Ewald Hering: Board of Scientific Advice for India, 483.

Obituary.—P. A. Guve: H. N. Dickson: G. B. MatuEws, 483.
INDEX: 484.

_ Error: EDWARD s. DANA.

se te ai <, ; 2 Me 2
2

: ee me _ ASSOCIATE EDITORS _

BERT EE. GREGORY, WESLEY R. COR AND
sae = ‘BEACH, ‘OF Naw ase

p
\

2 FIRTH SERIES |
| vou. “W{WHOLE NUMBER, CONTI},
No. 1B -JANUARY, £929.

: a ABe _ WITH TWO PLATES.

ey = US i

NEW HAVEN, CONNECTICUT.

ge em sae ae 5

rh ey 3 < a a ‘ -
MOREHOUSE & TAYLOR CO., PRINTERS, 125 TEMPLE STREET, )

‘Six sAgtiasd- per year, in advance, $6.40 to countries in the
‘anada, Single numbers 50 cents; No, 271, one dollar.
; Seto at the Post Office at N ew Haven, Conn,, under the Act

Sn & ~

$3.50.

“WILEY BOOKS»
Ready Januars 15th.

THIRD EDITION,

REVISED

A TEXT-BOOK OF MINERALOGY —

With an Extended Treatise on Crystallography and Physical Mineralogy
By EDWARD SALISBURY DANA, Professor of ee and Curator OEs:

pa Oey Yale University.

THIRD EDITION,

REVISED

By WILLIAM E. FORD, ra of Minerelosy, Sheffield Scientific

Schoal of Yale University.

A new edition of this famous book, brought abreast of present-day prac-
tice, and presenting in a clear, concise way, all the information needed by the
elementar y and advanced student of this i important science.

720 pages. 6 by 9.

Profusely illustrated.

Cloth $5.00.

The Other Dana Mineralogies

_ Minerals and How to
Study Them.
A Book for Beginners in Mineralogy.
- By Epwarp 8, Dana,
- 880 pages. dO by 7. 318 figures.
Cloth; $2.00.-

Dana’s Manual of Mineralogy.

Thirteenth Edition by WILLIAM E.
ForD.

460 pages. 45 by 714. 307 figures,
10 plates. Cloth $3.00—Flexibly

87 pages.

System of Mineralogy |
of JAMES DwicuTt DANA,

With Appendices I and II,complet-
ing the work to 1909. are

P Sixth Edition by Epwarbp S. Dana.

132 pages. 634 by 10. Over 1400
figures. Half leather, $15.00. .

Third Appendix to the Sixth |
Edition.
Completing the work to 1915.—
By- WiLuiamM E. Forp.
63/-by 10. . Cloth $2.00.

298 pages. 514 by 8.

}

Have you seen this hew Edition?
A MANUAL OF DETERMINATIVE MINERALOGY.
THIRD, REVISED AND ENLARGED EDITION.

By Prof. J. Votney Lewis, Rutgers College.
dl double pages.

81 figures. Cloth = 00.

Send TO-DAY for copies on Free Examination terms.

JOHN WILEY & SONS, Inc.

432 Fourth Avenue, New York

AIS 1.22

Am. Jour. Sci. Vol. Ill, 1922.

bh ae

THE

AMERICAN JOURNAL OF SCIENCE

[FIFTH SERIES.]

ART. L.—Genetic Features of Alnoitic Rocks at Isle
Cadieux, Quebec; by N. L. Bowen. With Plate I.

Introduction.

In the vicinity of Montreal, Canada, several intrusive
masses of alnoitic character have been found that are
genetically related to the main Mount Royal intrusion.
It was, indeed, from this area that Dr.. F. D. Adams
described the first alnoite found on this continent These
alnoites have aremarkable tendency to form breccias with
the invaded pre-Cambrian and Paleozoic rocks and an
excellent description of them, with particular reference
to the breccia, has been given by Dr. Robert Harvie.?

During the past summer my attention was called by Dr.
Harvie to the fact that a new occurrence had been found
at Isle Cadieux in the course of a road-materials survey,’
but that no detailed work had been done on it. Accord-
ingly the locality was visited and the material collected
has been found to be of such interest as to warrant special
description. I am greatly indebted to Dr. Harvie and to
the Director of the Geological Survey of Canada for the
opportunity of making this study.

Location.

Isle Cadieux is a flag station at mileage 26.8 from
Montreal on the Canadian Pacific short line between
Ottawa and Montreal. The outcrop of alnoitic rocks is

*F. D. Adams, this Journal, (3) 43, 269-279, 1892.

? On the origin and relations of the Palaeozoic breccia of the vicinity of

Montreal, Trans. Roy. Soc. Can., 3d Ser., vol. 3, sec. IV, pp. 249-299, 1910.
*Geol. Surv. Can., Summary Report for 1916, pp. 198-206.

Am, Jour. Sc1.—Firts Seriss, Vou. III, No. 13.—January, 1922.
1

2 N. L. Bowen—Genetic Features of

a rather inconspicuous knoll upwards of 200 yards in ©
diameter lying immediately adjacent to and south of the
railway track about half a mile west of the station. The ~
rocks are well exposed in the knoll itself, but their
relations to surrounding rocks are not visible. The simi-
larity to the alnoites of the Paleozoic intrusives makes it
impossible to doubt that it belongs, with them, to the
alkalie rocks of the petrographic province to which Dr.
Adams has given the name, Monteregian Hills.*

Rock Types. :

The principal rock of the outcrop is dark gray, mostly
fine-grained but mottled by large poikilitic biotites about
1 cm. in diameter which are the only erystals determinable
in the hand specimen. This type constitutes practically
the whole mass but there are streaks and indefinite
patches of a coarser-grained type in which biotite and a
light-colored constituent are readily distinguishable, and
also, in one place, a fine-grained minette-like type as an
irregular dike.

Mineral Constituents.

Under the microscope the principal type, constituting
the main mass, is found to consist of biotite, olivine,
augite, melilite, perovskite, an opaque ore mineral,
apatite, marialite, and alteration products of the above
minerals, principally carbonates. The rock therefore
has the typical composition of an alnoite, that is, a mica
peridotite with melilite. In one important particular
this rock is different from all other alnoites as described.
It contains two olivines, the ordinary olivine, chrysolite,°
and monticellite, the lime-magnesia olivine, a mineral
hitherto unrecognized except as a product of contact
metamorphism.

Biotite is decidedly the most abundant constituent of
the rock. It occurs in large grains up to 1 em. in diame- |
ter that are full of corroded inclusions of chrysolite and
augite, visible even in the hand specimen. ‘The biotite

*F. D. Adams, Jour. Geol., 11, 239, 1903.

° Throughout this paper it will be necessary to refer to the ordinary, mag-
nesia-iron olivine specifically as chrysolite in order to avoid the confusion
that would be involved in the use of the group name, olivine.

Alnoitic Rocks at Isle Cadieux, Que. 3

is not highly colored, its strongest absorption being a
rather pale ereenish brown. ‘Tn this respect it is,
however, markedly variable in a single crystal and in
particular tends to be less strongly colored where it abuts
against enclosed augite and more strongly colored where
it is against chrysolite. Its refractive indices are
appr oximately y = 1.61, ¢= 1.96 and it is nearly uniaxial.

Monticellite 1s, on the whole, the second constituent
in point of abundance. It occurs in fairly large grains
up to 1.5 mm. in diameter. These likewise include
corroded remnants of chrysolite and augite crystals,
but since the monticellite is not in such large grains
as the biotite, the potkilitic nature is not so obvious.
Occasionally the monticellite is in optical continuity
with an enclosed chrysolite grain. The optical prop-
erties of the lime-magnesia olivine, monticellite, agree
with those of the same mineral as described by Pen-
field and Forbes from Magnet Cove. The birefrin-
gence is moderate, the colors in thin section rising only
to an orange-red of the first order and usually falling far
short of that color. The optic axial angle 2V = 70° + 5°
and the sign is negative." In a section normal to the
optic axis there is a marked bending of the bar and the
sign is readily determined. In these respects the mineral
is distinct from the other olivine, chrysolite, which fre-
quently shows the brilliant colors of higher orders and
for which the axial bar remains so nearly straight that
one cannot be sure of the direction of its curvature or of
the optical sign. ‘The difference of birefringence is
apparent in this interference figure also, for the chrysolite
shows the inner colored lemniscates, whereas these are
beyond the aperture of the objective in monticellite.

It is, however, particularly in their quite different
relations to the other minerals that one is led in the first
instance to suspect the existence of two distinct olivines
in the rock, the monticellite occurring as oikocrysts, the
chrysolite as chadacrysts.. These relations will be
described in detail later, but advantage was taken of the

® This Journal, (4) 1, 135, 1896.

7 The value of 2V for monticellite from Magnet Cove has been incorrectly
transcribed into the texts of Iddings, Winchell and Rosenbusch, where it is
given as 37°31’. This is really the value of V as determined by Penfield and
Forbes.

®Iddings, Igneous Rocks, I, p. 202, 1909.

cs N. L. Bowen—Genetic Features of

difference between the two olivines in this particular to
permit the determination of the refractive indices. If
the rock is powdered and the measurement of indices by
the immersion method is undertaken, one usually cannot
be sure to which olivine an individual grain belongs. In’
thin section, however, there is no confusing them, on
account of the structural difference mentioned, so a slide
was uncovered, washed free from balsam and the rock
slice itself used in immersion liquids.. Thus the indices
of the monticellite were found to be y 1.668; a= 1.653
+ 002. Direct measurement of the maximum birefrin-
gence gave 0.015, so that while measurements so made are
not of the highest accuracy, it appears that the birefrin-
gence is slightly less than that in the Magnet Cove
mineral. The Magnet Cove mineral contains nearly 5
per cent FeO, and in view of the close approach in indices
the present mineral must have nearly the same com-
position. Pure artificial monticellite has distinctly lower
indices.®

The chrysolite has already been described in part,
while contrasting it with monticellite. Its refractive
indices, measured as above, are y—1.700; «—1.665
+ .003 and 2V close to 90°. It probably contains in the
neighborhood of 10 per cent FeO, though the effect of a
possible TiO, content is not known.

The augite needs but brief mention. Its extinction
‘angle mounts to about 405° and the refractive index
y=1.715. ‘This combination of properties rather excludes
any significant alkali content and fixes it as ordinary
augite. Itis slightly brownish to greenish in thin section
and the typical cleavage is well developed.

Lhe melilite is in comparatively small grains that show
the typical tabular development parallel to the base,
though they are not idiomorphic. The basal cleavage
is sometimes marked only by the development of minute
seams of an alteration product of high birefringence.
Occasionally it may be well developed and accompanied
by prismatic cleavage of nearly equal strength. The
melilite itself always has very low birefringence but is
distinctly variable in optical properties. This variation
appears to be connected with the amount of melilite in
the rock. Where present in small amount it is optically

* Ferguson and Merwin, this Journal, (4) 48, 92, 1919.

Alnoitic Rocks at Isle Cadieux, Que. 5

positive; in somewhat greater amount it is about iso-
tropic, and where present in considerable amount it is
zoned, with the inner zones nearly isotropic and the outer
zones distinctly negative.

The perovskite is quite abundant for a constituent that
usually occurs in such minor amounts. It is in sharp
octahedra of a wine yellow color in thin section and is
apparently isotropic, the twinning phenomena not being
visible in these grains.

The opaque ore mineral has a bluish reflection and is
presumably titaniferous magnetite; at any rate it is
strongly magnetic. It occurs rather abundantly as dis-
seminated grains and is particularly strongly developed
as a rim about corroded chrysolite grains where they
abut against biotite, monticellite or melilite.

The apatite is present in unusually large amount and
in relatively large grains. It is distinguished from meli-
lite by its higher birefringence and by the hexagonal
shape of its basal sections.

The marialite occurs as a few small grains that were
detected in only some of the slides where they were
closely associated with monticellite. Its refringence is
comparable with that of quartz, that is, only shghtly
higher than that of balsam, and in this particular it is in
marked contrast with all the other constituents of the
rock which are uniformly minerals of high relief. The
birefringence is likewise comparable with that of quartz,
but is appreciably higher. It is uniaxial and negative.
These properties agree with those of a scapolite close to
the sodic end of the series, that is, marialite. This
mineral has been observed in trachytes of the Phlegrean
Fields, near Naples.

The alteration products were not given very careful
study. They are mainly carbonates formed by the
alteration of monticellite and melilite. The carbonate
patches are frequently crowded with minute needles of
apatite which suggest that the alteration of the minerals
is not to be assigned entirely to weathering but is in part
a post-magmatic phenomenon intimately connected with
the consolidation of the rock. It is noteworthy that
neither olivine shows the serpentine type of alteration.

While the rock as a whole is considerably altered, the
alteration is unevenly distributed, and in many parts of

6 N. L. Bowen—Genetic Features of

a section the minerals monticellite and melilite may be
found without even incipient change. The monticellite
is apparently considerably more susceptible than melilite.

Variation in the Principal Type.

The greater part of the mass appears to be essentially
uniform in hand specimens and in order to gain an idea
of such variability as might exist it was necessary to
resort to collection of specimens regularly distributed
over the outcrop. Some thirty specimens were collected
and they show considerable variation in the proportions
of the minerals. Usually biotite is the most abundant
mineral and is always an important mineral, but it is not
infrequently exceeded in amount by monticellite and
sometimes by melilite or chrysolite. Melilite may be
present to the extent of about 30 per cent and again may
occur in vanishingly small quantity. Augite may amount
to about 20 per cent but is usually less and there is a
general tendency for monticellite to occur in greater
quantity where augite is more abundant. Chrysolite
may rise to 35 per cent or again sink to 10 per cent.
Pyroxene, monticellite and melilite were practically
absent in only one specimen, giving essentially an ordi-
nary mica peridotite (see Plate I (c) ).

Chemical Composition of the Principal Type.

In view of the variability of the rock it was desirable
to choose for analysis a specimen which appeared under
the microscope to represent the general average. At the
same time it was desirable to choose material which was
as unaltered as possible, and this latter restriction made
it necessary to take material differing somewhat from the
average. The actual specimen chosen contains the
constituent minerals in approximately the following
proportions in weight per cent: chrysolite 30, melilite
20, biotite 20, monticellite 10, augite 6, all others 9. No
great accuracy is claimed for these figures because the
rock as a whole is rather unfavorable for the estimation
of proportions of the minerals. The general average of
the rock would be decidedly higher in monticellite, con-
siderably lower in melilite and chrysolite and somewhat
higher in biotite and augite.

Alnoitic Rocks at Isle Cadieux, Que. il

For the analysis (I, Table I) and for the other analyses
in this paper I am indebted to Dr. Washington and take
this opportunity of expressing my thanks.

TABLE I.
I DE Norm of I*
Si0, 33.26 30.85 An 8.06
AIO; 5.90 8.21 Kp i-93
Fe,O, . 5.30 3.33 Ne 5.68
FeO 6.54 6.52 Ol 49.56
MgO 26.41 23.16 Cs 15.82
CaO 14.47 * 16.46 Mt 7.66
Na,O 1.23 1.01 Il 4.10
K,O 0.82 1.43 Ap 2.02
H,O + 1.91 1.22 Symbol
H,O — 0.09 0.05 Vis SGI) 225s 2a
CO, 1.10 3.04
HO; 215 2.87
ZrO, none n.d. :
PS Ox 0.76 1.90
MnO 0.15 0.21
CresO3 0.05 n.d.
BaO 0.08 n.d.
SO, 0.22 Gls
100.44 100.26

I. Monticellite alnoite (melilite-rich) Isle Cadieux, Quebec.
H.S. Washington analyst.

II. Monticellite alnoite (melilite-poor) Isle Cadieux, Quebec.
H.S. Washington analyst.

* Calculated by Dr. Washington. The CO, indicates 2.50 per cent calcite.

Under II in Table I is given an analysis of a variety
rather rich in monticellite and very poor in melilite.
This specimen was chosen for analysis partly in order to
have a chemical check upon the optical identification of
the monticellite. The microscope shows the minerals to
be present in roughly the following proportions in weight
per cent: biotite 30, monticellite 25, chrysolite 15, augite
10, melilite 3, apatite 4, perovskite and ore 7, carbonates 6.
Though there are lime minerals other than monticellite in
considerable amount, they fail entirely to account for
the 16.5 per cent of lime shown in the analysis. About
one half the lime must be assigned to the monticellite
which checks approximately with the amount of monti-

8 N. L. Bowen—Genetic Features of

cellite present. Hixcept in its unusually low melilite,
specimen II is fairly typical of the average.

Comparison of the two analyses shows no very marked |
differences in spite of the considerable contrast in pro-
portions of the constituents. Lime is even higher in the
variety poor in melilite because it is correspondingly
richer in monticellite (and in minor lime-bearing min-
erals) and the lime content of monticellites and melilites
is about the same. Magnesia‘is higher in I on account
of the greater amount of chrysolite. The lower biotite
content of I shows up in the lower potash and alumina.
The biotite must contain much soda. ‘The lower silica of
Il can not be assigned any special significance on account
of its greater alteration (note CO,).

Comparison of the analyses with those of other alnoites
shows the present types to be higher in magnesia. Soda
in excess of potash as in I is exceptional in alnoites which
are usually like II in this respect in spite of their common
association with sodic rocks. ‘This is connected with the
low biotite content of I; in fact, this type, in virtue of
decreased biotite, leans toward melilite basalt.

Specimen I is apparently the freshest alnoite yet
analyzed and even II compares favorably with others in
this respect.

No new names are proposed for these rocks nor for any
of the types described in this paper. It is believed that
the descriptive terms employed have a distinct advantage
over the meaningless names based on locality that are
being coined daily and with which the literature is already
overburdened. |

‘Melilite-Biotite Rock.

Before going on to discuss the relations of one mineral
to another in the principal type whose mineralogy was
given above, the streaks and patches that have been
mentioned as occurring in it will be described. The
streaks are from 1 inch to 2 feet wide and traverse the
main type after the manner of dikes or veins with
indefinite boundaries. They consist almost entirely of
biotite and melilite with subordinate amounts of apatite,
chrysolite, monticellite, perovskite, and an opaque ore
mineral.

Alnoitic Rocks at Isle Cadieux, Que. 9

The biotite is often in idiomorphic hexagonal-shaped
plates and is not obviously poikilitic, but under the micro-
scope occasional corroded inclusions of chrysolite, augite
(?) and monticellite are to be found. It is much more
strongly colored than the biotite of the main rock.

The melilite shows the typical tabular development
and the plates may be as much as 5 mm. across. ‘The
erystals are conspicuously zoned, the zones being marked
by a difference in birefringence. The center of the
erystal is usually isotropic, though occasionally it is
barely over the border and in the positive series. The
outer zones are plainly negative and increasingly so as
one passes toward the rim. (See Plate I (a).) Hven
here, however, the birefringence is still very weak, much
less, for example, than that of apatite. The measured
index of the ordinary ray is somewhat variable, but is
close to 1.636. The melilite, like the biotite, contains cor-
roded inclusions of chrysolite and monticellite.

The apatite is present in such amount and in such large
prisms as to raise it somewhat above the class of minor
constituents. It is characterized by its freshness, lack
of cleavage and high birefringence as compared with
melilite. z |

The presence of relatively small amounts of chrysolite,
augite, and monticellite as corroded inclusions in biotite
and melilite has already been mentioned. An analysis of
the melilite-biotite rock is given in Table IJ under LI.

Fine-grained Alnovte.

The fine-grained type that has been called minette-
like in the hand specimen occurs at the southern side of
the knoll as a dikelike intrusive in the main mass. It
‘consists essentially of chrysolite crystals, partly resorbed,
lying in a matrix of biotite and melilite with the usual
minor constituents of the principal type of which it is
merely a textural variant. Augite and monticellite are
practically if not entirely absent, a condition which was
noted in one specimen of the main mass. This fine-
grained type is considerably fresher than the coarser
facies usually is.

10 N. L. Bowen—Genetic Features of

TABLE II.
I II Norm of II.

Sid, 31.10 83.31 Q 61.08
Al,O; 10.08 4.46 Or 23.91
Fe,O, 3.64 AAO) Ab 0.52
FeO B30 0.45 Ae OAs,
MgO 12225 0.28 Ns 1.83
CaO Oat 2.87 Di ow
Na,O 1.95 1.42 Wo 5.80
K,O 3.56 4.02 fl 1.06
H,O + 0.76 1.02
H,O — 0.46 0.18
CO, 4.94. TaLOl
TiO, le 0.64 Symbol
12). 22: 0ili n.d. (1) Wis2 sie
MnO 0.09 Ey Tiles

99.69 99.75

I. Melilite-biotite rock, Isle Cadieux, Quebec.
H. 8S. Washington analyst.

II. Aplitic inclusion in monticellite alnoite, Isle Cadieux,
Quebec. H. S. Washington analyst.

Distribution of Similar Rock Types.

About a mile northwestward from the outcrop to
which particular attention is here given, there is a cut on
the railway track and the material taken from this cut is
almost entirely of rock types similar to those in the
outerop studied. Here are found various facies of
monticellite alnoite and biotite-melilite rock exactly lke
the types already described. From the present condition
of the cut it is not possible to determine definitely whether
this material had been in place, though its uniform
nature strongly suggests that such was the case. Kven
if it was of the nature of drift it could not have been
derived from the outcrop described, for it has not the
appropriate position relative to the direction of motion
of the glacier. The existence of this material proves
definitely, then, that the types described are not confined
to the outcrop studied and may be rather widespread in
this area.

Alnoitic Rocks at Isle Cadieux, Que. lot

Inclusions.

Inclusions in the igneous mass are decidedly rare; not
even locally was any facies observed that remotely
approached a breccia.t®? A few inclusions were seen,
however, and it was possible to obtain some of these for
sectioning. They were found to be of three kinds: (a)
large augite crystals, (b) an aplitic rock, (c) a sodic
syenite.

The augite crystals are fairly numerous at the north-
ern margin of the mass as exposed, and stand out as
knobs about an inch in diameter on the weathered surface.
Peripheral alteration of the crystals by the magma is
often to be seen even in the hand specimens, the core being
of black glassy appearance and the border dull. The
optical properties of the crystals show them to be iden-
tical with the augite of the rock itself. This fact sug-
gests that they may be phenocrysts rather than inclusions
but their general appearance is rather that of cognate
xenocrysts.

The inclusion of an aplitic rock that was broken out for
study was found to be of a highly siliceous type, probably
a fragment of a fine-grained dike. It is very rich in
quartz; the feldspar, evidently an alkaline one, is much
kaolinized, and these make up practically the whole rock.
There is, however, a little egiritic pyroxene ‘‘in minute
prisms, frequently forming matted masses,’’" with green
to yellow pleochroic colors, small extinction angle, and
negative elongation. A few grains of zircon were
observed also. The quartz of this rock shows under the
microscope a marked development of an imperfect cleav-
age that breaks it up into roughly rectangular pieces.
The presence of this cleavage or cracking is probably to
be taken as evidence that the inclusion has been heated
above 975° by immersion in the alnoite material, the
quartz therefore passing through the a-8 quartz inversion
- point.” Presumably it was not heated to 870°, for there
is no suggestion of a change to tridymite. In the course

10The interior portions of some of the masses described by Harvie (op.
cit. p. 256) are often nearly free from inclusions.

11 The egirite of the tinguaite of this province is so described by O’Neill,
Geol. Surv. Can. Mem., 43, p. 60, 1914.

42 Wright and Larsen, Quartz as a geologic thermometer, this Journal, 27,
437, 1909.

12 N. L. Bowen—Genetic Features of

of the analysis’ of this rock Dr. Washington noted a
resistance of its powder to wetting unusual in his expe-
rience. This resistance is characteristic of very fine
powders and may be connected with the ease of reduction
of the cracked quartz of this rock to such a fine state.

The analysis is given in Table II under II and the eal-
culation of the norm, also by Dr. Washington, is given
with it. It proves to be that of a very unusual rock,
highly siliceous and showing high potash with rather high
lime, in which respects it resembles moldavites that are
believed to be of meteoric origin.t* In its content of
egiritic pyroxene it would appear to be related to the
alkalic rocks of the Monteregian province and thus to
be a cognate inclusion. The high potash would seem
to contradict this and the amount of silica is quite
unmatched in any described rocks of this petrographic
province.

The other type of inclusion examined consists almost
exclusively of oligoclase with a little microcline. These
are cut by veinlets of hydronephelite. A little mica and
a greenish pyroxene are the only other constituents.

This is apparently a rock of the alkalic series also, and
possibly all of the inclusions examined are to be regarded
as cognate or belonging genetically with the containing
rock. :

Relations of the Mimerals.

We shall now return to a discussion of the relations of
one mineral to another in the principal type or, better
perhaps, in the mass as a whole.

In order that the details may be more intelligently fol-
lowed, the conclusion arrived at from them will be
anticipated and may be stated as follows. The mass
originally consisted of chrysolite and augite, nearly com-
pletely consolidated as such. It was then acted upon
with falling temperature by an alkalic liquid (magma)
which was, in part at least, its own interstitial liquid, and
as a result of this action the other minerals, biotite, mon-
ticellite, meliite, perovskite, apatite and marialite were
formed. The minerals may therefore be divided into two
classes, the original minerals augite and chrysolite and

U.S. Geol. Survey, Prof. Paper 99, p. 53.

Alnoitic Rocks at

| fonal Muse
the replacing minerals principally , monticellite
and melilite.

There is no very obvious tendency for any one replac-
ing mineral to be substituted for any particular original
mineral. The replacement is rather of the mass as a
whole. Occasionally, however, a more or less definite
pseudomorph of monticellite after augite may be noted
and frequently a monticellite band is interposed between
mica and augite. Moreover, in facies of the rock where
augite was absent monticellite is not developed. The
evidence is good that augite is the principal if not the
sole source of the monticellite. (See Plate I(d).)

The original augites were large crystals approaching
1 cm. in diameter but the corroded remnants are seldom
one-tenth as large. Several adjacent individual remnants
embedded in the replacing minerals are frequently seen
to be similarly oriented, with their cleavages and extinc-
tions coincident, and it is not difficult to picture the
original augites now largely replaced. (See Plate I(b).)
Where the replacing mineral is biotite it is practically
always paler in color adjacent to augite.

The original chrysolites were rather less than one-
half as large as the augites. Three or four adjacent rem-
nants can frequently be connected up to construct the
original crystals, though this is not so conspicuous as
in the case of the augites. Whether biotite, monticellite
or melilite is the replacing mineral the chrysolites nearly
always show a rim about them of particularly strong
concentration of ore minerals. (See Plate I(c).) The
augites occasionally show such a rim but it is usually
within the border rather than at the border. There is a
development of very fine crystals of perovskite in these
rims and as one passes outward coarser and coarser
crystals are encountered. ‘They are plentiful in the mon-
ticellite, melilite and biotite, but the central portions of
chrysolite and augite crystals are free from them. There
is a strong suggestion, then, that the perovskites were
formed as a result of reaction upon the original minerals.
As has been noted before, where monticellite resorbs and
replaces chrysolite the two are frequently in optical con-
tinuity, their boundary being marked by the dark band
of ore minerals.

When one thus separates the minerals into the two

14 N. L. Bowen—Genetic Features of

classes of original and replacing minerals and pictures
the original minerals as they were before replacement
began, it becomes apparent that there could have been no
ereat amount of any material in much of the rock other
than augite and chrysolite. It is probable that there was
a moderate amount of interstitial liquid which gave oppor-
tunity for such uniform action at all points in the mass
and upon each individual mineral grain. It is probable,
too, that this interstitial liquid was itself of such a nature
as to enter into reaction, as the temperature changed, with
the crystals that had already separated from it, in such
a manner as to produce the new minerals actually found.
It is not probable, however, that the interstitial liquid
was of sufficient quantity to produce so much change.
Rather is it to be supposed that there was a movement
of liquid of the same kind through the interstices of the
mass, that the liquid reacted with the solid phases and
passed on (or was crowded out), carrying with it some
of the products of the reaction until finally the interstices
were filled up.

However these details may fit the fact, it is certain that
a liquid permeated the rock and partly replaced augite
and chrysolite by monticellite, biotite and melilite. Now
it is believed that the seams of melilite-biotite rock
represent simply streaks along which the freedom of pas-
sage of liquid has been greater and where the replacement
is more advanced. In these monticellite is corroded by
biotite and melilite in much the same manner as are the
original minerals in the main mass. ‘The monticellite
is, then, to be regarded as an intermediate step in the
replacement of augite and chrysolite by melilite and
biotite.

Nature of the Reacting Inquid.

The question naturally arises as to the nature of the
liquid (magma) which could produce results of the kind
noted. ‘The formation of such minerals as monticellite
and melilite might on first thought be considered to
require a lime-rich liquid. On the other hand the for-
mation of biotite points definitely to an alkalic liquid
and it will be found that there are good reasons for
believing that an alkalic liquid is capable of producing
the lime-rich minerals as well. The alkalic liquid 1s,

Alnoitic Rocks at Isle Cadieux, Que. 15

moreover, a decidedly appropriate one in this particular
province; indeed nephelite is frequently an important
eroundmass constituent in some of the alnoites of
Harvie.

Petrologists will have no difficulty in accepting the
formation of biotite from other femic minerals (in this
case principally chrysolite) because it is an action with
which they are already acquainted. The formation of
monticellite from other femic materials (in this case prin-
cipally augite) has not been noted elsewhere. Assuming
that it is desirable to describe the action in a few words,
this may be done roughly by stating that the monticellite
is formed as a result of desilication of augite by the
alkale liquid. We thus find an explanation of the fact
that where no augite was present in the original material

no monticellite was formed.

The liquid that accomplished this action was in all
probability a nephelite-rich liquid closely related to that
which formed the nephelite aplites of this province, and
in order to throw some light on equilibrium in such liquids
the results of experimental study of related liquids will
now be presented.

Experimental Studies of Related Mixtures.

The system CaO-MgO-Si0, has been investigated by
Ferguson and Merwin and their results are expressed
in the equilibrum diagram, fig. 1. <A liquid made up of
00 per cent forsterite and 50 per cent akermanite begins
to erystallize with deposition of the olivine, forsterite,
an end member of the chrysolites; when the temperature
falls to 1450° monticellite begins to crystallize; and at
1456° akermanite, an end member of the melilites, begins
to crystallize and monticellite redissolves until finally
monticellite is entirely replaced by akermanite and the
whole mass has completely solidified. In this case, then,
we have the succession, forsterite (chrysolite)-monticel-
. lite-akermanite (melilite), the akermanite having a replac-
ing relation to monticellite. This liquid and the adjacent
liquids which show the same effects are, of course, them-
selves very rich in lime and magnesia, but when other
components are added the same relations must persist at
first though they will be modified and may finally disap-

(C776)
2570

16 N. L. Bowen—Genetic Features of —

pear as more and more of other components are added.
The writer has investigated the effects of the addition
of nephelite to some of these mixtures. The complete
quantitative solution of the problem can not be expressed
in terms of less than five components, and this is, of

ING) de
SxO
NS 1710

Crastobakte

— Bounaares ; —J/sotherms determined; —--/sotherms by extrapolation; © Compounds ;
+++,GZZA Solid solutions

Fic. 1.—Equilibrium diagram of the system: CaO-MgO-Si0,
(after Ferguson and Merwin).

course, not possible without a vast amount of preliminary ~

work on the simpler systems. However, an approximate
statement is possible; in other words, a statement of the
kinds of mineral phases present at various temperatures
may be made without giving the precise composition of
the phases of variable composition or the exact relative
proportions of the phases.

gO

2800

~Alnoitic Rocks at Isle Cadieux, Que. 1g

The results of the investigation of equilibrium in mix-
tures of diopside, CaMgSi,O,, and nephelite, NaAlSiO,
are given in Table III and in fig. 2, which, it must be
realized, is not a two-component diagram, though having
the same general form. It gives the names of the phases
present at various temperatures in mixtures whose total
composition can be expressed in varying proportions of

TABLE III.
Composition
Nephelite Diopside Temp. Phases present
10 90 1350 Glass and diopside
10 90 1355 Glass only
20 80 1100 Glass, diopside and melilite
20 80 AGO ae Glass, diopside and melilite
20 80 1200 Glass and diopside
20 80 1285 Glass and diopside
20 80 1290 Glass and forsterite
20 80 1295 Glass only .
35 65 1100 - Glass, diopside and melilite
35 65 1175 Glass, diopside and melilite
35 65 1180 Glass, diopside and olivine
35 65 1220 Glass, diopside and olivine
35 65 1230 Glass and olivine
35 65 1255 Glass and olivine
35 65 ~ 1260 Glass only
50 50 1050 Glass, melilite and olivine
50 50 1175 Glass, melilite and olivine
50 50 1180 Glass and olivine
50 50 1215 Glass and olivine
50 50 1220 Glass only
60 40 1050 Glass, melilite and olivine
60 40 1140 Glass, melilite and olivine
60 40 1150 Glass, nephelite, melilite and olivine
60 40 1175 Glass, nephelite, melilite and olivine
60 40 1180 Glass and olivine
60 40 1190 Glass only
70 30 1150 Glass, nephelite, melilite and olivine
70 30 1175 Glass, nephelite, melilite and olivine
70 30 1180 Glass and nephelite
70 30 1230 Glass and nephelite
70 30 1235 Glass only
80 20 1150 Glass(?)nephelite, melilite and olivine
80 20 1180 Glass and nephelite
80 20 1245 Glass and nephelite
80 20 1255 Glass and carnegieite

Am. Jour. Sct.—FirtH Srrizs, Vou. III, No. 18.—January, 1922.
2

18 N. L. Bowen—Genetic Features of

diopside and nephelite, but the composition of some of
the individual phases can not be so expressed.'#

IDG, 2

WiAhi 51 Oy

GO WNephelie

GO

Ne. Fel. 0. 811g.

00 LG

AO
Wt per cert
Fig. 2.—Pseudo-binary diagram of equilibrium in mixtures of nephelite and
diopside.

Jemperatores of complete comsolhdarion unknown.
40

JO

/Tel, Di. & liq.
20

Ol. Dia lig
/0

SS
VS
eS
SJ SS
S S S S S SS
oO s SS N S

“In Table III and fig. 2 the group name, olivine, is used because the
olivine formed at lower temperatures may be monticellite, though it is cer-
tainly forsterite at higher temperatures.

Alnoitic Rocks at Isle Cadieux, Que. 19

Since both components melt congruently there is, at
either end, a portion which behaves at higher tempera-
tures as an ordinary two-component mix but towards the
middle the relations are much more complicated. In
particular it is to be noted that the early crystals are
olivine and melilite, that is, nephelite reacts with diop-
side in such a way as to produce these very basic
molecules which are then precipitated. The separation
of these necessarily renders the liquid with which
they are in equilibrium correspondingly more siliceous.
Another important general feature is the very low
temperature at which there is still some liquid in the
intermediate mixtures. Thus at temperatures in the
neighborhood of 1000° some liquid remains and in many
mixtures it is impossible to tell at what temperature com-
plete solidification occurs, on account of the slowness
with which crystallization proceeds at such temperatures.

The behavior of any particular mixture is given by the
diagram, but will be put in words for certain of these
that are of special significance. A mixture of 30 per cent
nephelite, 70 per cent diopside begins to crystallize at
1270° with separation of forsterite. This continues to
separate until a temperature of 1240° is reached, when
diopside begins to separate and forsterite to redissolve.
The re-solution of forsterite continues until the tem-
perature has fallen to about 1180° and at about the same
temperature melilite begins to separate. At tem-
peratures below this and as far as 1100° the mass
consists of diopside, melilite and hquid. Below 1100°
the erystals formed are so small and the attainment of
equilibrium so slow that further changes can not be
definitely described, the temperature of final consolidation
being also unknown. No nephelite is formed at any tem-
perature where the identification of phases is certain;
indeed the reaction which forms melilite presumably
destroys the nephelite and part of the diopside.

A mixture of 50 per cent nephelite and 50 per cent
diopside begins to erystallize at 1220° with separation
of forsterite which continues down to about 1180°. At
this temperature melilite begins to crystallize, and below
this temperature the mass consists of melilite, forsterite
and liquid down to temperatures where the crystals
formed are no longer identifiable. Here again the tem-

20 N. L. Bowen—Genetic Features of

perature of final consolidation is unknown. Neither
nephelite nor diopside appears, at least not at any tem-
perature where the identification of phases is possible.
They mutually destroy each other. in the reaction which
produces melilite.

A mixture of 70 per cent nephelite and 30 per cent
diopside begins to crystallize at 1235° with separation of
nephelite. At about 1180° the nephelite is jomed by
olivine and melilite, and down to temperatures where the
crystals formed are no longer identifiable the mass
consists of nephelite, melilite, forsterite and liquid. No
diopside appears, it being used up by reaction with
nephelite to form melilite.

Nature of the Melilites.

The melilites formed from the mixtures studied do not
correspond with the end member, akermanite, whose
equilibrium relations are given in fie. 1, but are strictly
analogous to natural igneous melilites. Some of the
crystals, particularly those formed at higher temper-
atures, are positive like akermanite but of much weaker
birefringence. With increasing richness in nephelite in
the total mixture the melilites separated become isotropic
and finally pass over into negative melilites. The decrease
of birefringence is accompanied by a slight decrease
of refringence which is about 1.630 in the sensibly
isotropic member. The series is quite different, there-
fore, from the akermanite-gehlenite series of Ferguson
and Buddington in which the refringence increases as the
birefringence decreases on passing away from aker-
manite. The present artificial melilites are, however,
readily explained and correlated with natural melilites
in the light of the recent work of Buddington.1® The
relation to Buddington’s mixtures is brought out clearly
by adding nephelite directly to akermanite and crys-
tallzing the fused mixture at a low temperature. When
10 per cent nephelite is added to akermanite the product
is nearly homogenous melilite of somewhat lower refrin-
gence and distinctly lower birefringence than akermanite.
There is, however, a very little forsterite in excess and

* These melilites fall under the subdivision humboldtilite as used by
Buddington. Following paper, this number, p. 75.

Alnoitic Rocks at Isle Cadieux, Que. 21

also minute stringers of glass. As more nephelite is
added the forsterite and glass increase somewhat and the
birefringence of the melilite further decreases. When 25
per cent nephelite is added the melilite is sensibly iso-
tropic and its index about 1.630, and further addition of
nephelite produces negative melilite. The series of meli-
lites is evidently strictly analogous to the akermanite:
sarcolite: soda-sareolite series of Buddington (p. 956).
An equation can be written showing the formation of these
molecules by reaction of akermanite and nephelite (the
akermanite being in excess) with the formation of some
forsterite and a more siliceous alkalic liquid that remains
as glass. The equation is as follows:

nephelite akermanite sarcolite
15(NaAlISi0,) + 6(2CaO.MgO.2Si0,) = 4(8CaO. Al,O,.3810,) +
soda-sarcolite olivine

2(3Na,0.Al,0,.3Si0,) + 3Mg,SiO, + 3NaAlSi,O,

The melilites formed in the nephelite-diopside mixtures
are exactly like those formed in the nephelite-akermanite
mixtures except that they do not show as great range of
composition. As noted, they may be either positive or
negative but are never far removed from the isotropic
member; in other words they never approach the pure
akermanite end member but appear to center around the
low melting mixture which Buddington has decided is
the case in natural melilite (p. 87). The reactions taking
place when nephelite is added to diopside may be pictured
by imagining that the first step is the production of for-
sterite and akermanite from diopside by desilication as
follows:
4CaMeSi,0, + 3NaAlSiO, —

2(2CaO.Mg0O.28i0,) + Mg.Si0, + 3NaAl18i,0,
The akermanite then further reacts with nephelite after
the manner indicated in the first equation to produce the
other molecules of the typical melilites. The forsterite
and the melilites resulting are the first erystals precip-
itated from the intermediate mixtures.

The correspondence with the melilites of the rocks
described is quite marked for it will be recalled that in the
portion of the rock where reaction with the alkalic liquid
has not proceded far the melilite is positive (akermanite-
rich) whereas in the biotite-melilite seams where reaction

2Y N. L. Bowen—Genetic Features of

has gone farther the melilites have isotropic cores and
negative outer portions. A zoning of this particular
nature has been noted in melilite from San Venanzo.1® In
the natural rock ordinary olivine was not one of the
products of the reaction for the reason that the reacting
liquid differed from the experimental liquid, notably in
the potash content, with the result that mica was formed
instead. :

Space Relations of the Equilibrium Fields.

The separation of melilite and olivine from liquids
whose total composition can be expressed as a mixture of
nephelite and diopside has already been discussed in
some detail and pictured in fig. 2. Asan aid to the under-
standing of equilibrium relations in the rock we have
described, it 1s perhaps desirable to present the results
in a somewhat different manner, that is, in terms of a
composition tetrahedron having as its base the CaO-
MgO-Si0, triangle and nephelite at the other corner.
The compounds and their equilibria with liquids for all
compositions lying on the base have been studied by
Ferguson and Merwin, and our present purpose is to
describe how their fields are affected as we pass out into
the tetrahedron, that is, as nephelite is added. It is
apparent from the equations that have been given above
that even four components are not sufficient to express
accurately the composition of all the phases present but
we could, as in the pseudo-binary diagram (fig. 2), mark
the tetrahedron out into fields showing the nature of the
solid phase in equilibrium with liquid, the total com-
position of the mixture being given by a point in the
tetrahedron. |

No attempt has been made to map out the fields
throughout the whole tetrahedron, which would be a labor
of many years. Attention has been concentrated upon
the position of the melilite, olivine and pyroxene fields as
having particular interest in the present connection.

The fields that have been delimited for the MgO-CaO-
SiO, mixtures constituting the base will pass upward
into the tetrahedron, the boundary curves becoming
boundary surfaces. The position of these surfaces as

* Rosenbusch, Mikroskopische Physiographie, I. 2, p. 71, 1905.

Alnoitic Rocks at Isle Cadieux, Que. 23

they are encountered by the lnes indicating the changing
composition of any crystallizing liquid mixture may be
shown approximately by projecting upon the base the sur-
faces where thus encountered. The position of the point
on the nephelite-diopside join (A) serves as a reference
or indicating point.

In the CaO-MgO-Si0, system itself as shown in fig. 1
all of the liquids in equilibrium with monticellite are more
siliceous than monticellite itself. On the other hand
diopside lies in the pyroxene field, that is, the hquids with
which it is in equilibrium surround it and are both more
siliceous and less siliceous. When nephelite is added the
pyroxene field is pulled over toward the silica corner and
with about 19 per cent nephelite the boundary surface
pyroxene-forsterite passes through the indicating point
(A) which thereafter lies in the forsterite field. At 20
per cent nephelite the section through the tetrahedron
shows the projected fields in the relative positions of
fig. 3, the arrows indicating falling temperature. This
shows that forsterite crystallizes first, pyroxene second,
and then melilite from a mixture given by the indicating
point.

With 40 per cent nephelite the relative positions of the
projected fields are as shown in fig. 4. Thus forsterite
crystallizes first and is joined simultaneously by both
pyroxene and melilite from a mixture of diopside and
nephelite as given by the indicating point. With 55 per
cent nephelite the fields as projected are shown in fig. 5.

The manner in which the fields of forsterite and melilite
are drawn farther over toward the silica corner as more
and more nephelite is added is well shown by this series
of figures (1, 3,4 and 5).*” A simple, if only qualitative
explanation of this is that equilibrium in the liquid is such
that the nephelite in the liquid takes silica to itself and
leaves relatively little for the lime and magnesia, thereby
increasing the concentration in the liquid of the low-silica,
lime and magnesia compounds and causing their pre-
cipitation at an early stage. Indeed the series of figures
is but a graphical and qualitative expression of the
equations given on page 21. This affords an example
of the manner in which both affinity and solubility combine

“In fig. 1 the point corresponding to (A) is the diopside point itself.

24 N. L. Bowen—Genetic Features of

Fig. 3.

Fyrosene
<i

/felilite \O Wiviae
aX
4 \
Mons
LO picellite
OB

Rye. 4:

Py oKe7e

Meliiite : 4 wine

7\

Yor -
COxtceliite
o8

Fies. 3, 4, 5 and 6.—Projections of fields of stability in nephelite-diop-
side mixtures upon the CaO-MgO-SiO, plane. Same orientation as fig. 1.
Projected point for mixtures of nephelite with diopside, A; with monti-
cellite, B; with akermanite, C; with enstatite, D; with forsterite, E.

Fie. 3, 20 per cent nephelite; Fig. 4, 40 per cent; Fig. 5, 55 per cent;
Fig. 6, deduced relations in magma.

Alnoitic Rocks at Isle Cadieux, Que.

Free5.

Fyroxeeé

A
Creliite ,\ Olivine.
SEN

oe ‘
OC fop-

TEs IG

Fyroxren
Mehr, O4
| Wivine

0)
Vio 6” ve
Cc

o8

25

26 N. L. Bowen—Genetic Features of

to determine the solid phases that shall separate from a
liquid, the affinity of nephelite for silica being unques-
tionably an important factor in the present instance.

On account of the impossibility of distinguishing
between forsterite and monticellite in the small crystals
obtained at the relatively low temperatures, I am unable
to speak with certainty regarding the behavior of the
monticellite field. It is apparently pulled over in the
same manner as the other fields, indeed even in the ter-
nary system itself (fig. 1) it is displaced in that direc-
tion, so the addition of nephelite presumably emphasizes
this natural tendency. Monticellite may therefore sep-
arate from liquids not particularly rich in lime and
magnesia, providing they are alkalic. The manner in
which the monticellite field separates a part of the for-
sterite field from part of the akermanite field, with a
consequent reaction relation, has already been pointed
out. This separation is apparently more marked in the
alkalic liquids and in the case of the natural rock
described it appears that, for the system it represents,
the fields must be considered to have the relative positions
shown in fig. 6, which gives their projection upon the
CaOQ-MgO-Si0, plane in a manner analogous to that of
figs. 3,4, and 5. Such a figure will, at any rate, afford an
explanation of the paragenesis of minerals in the rock
and is but a further modification of figs. 3, 4, and 5 in the
direction in which modification has been demonstrated in
the presence of alkali-rich liquid. The figure (6) shows
that the crystallization of a liquid given by the indicating
point may begin with the formation of olivine, which is
followed by pyroxene, then monticellite, then melilite, the
monticellite and melilite having a reaction relation to
pyroxene and olivine; that is, the latter are resorbed and
replaced by monticellite and this in turn by melilite.
This is the relation that the actual minerals of the rock
show, with biotite as an additional replacing mineral, the
investigation of which would require the addition of
several other components. The occurrence of melilite as
reaction rims about augite or olivine has been noted in
other rocks of this area!* and indeed a great many of the
occurrences of melilite suggest its origin by reaction of
pyroxene with alkalie liquid.

** Robert Harvie, op. cit., p. 261.

Alnoitic Rocks at Isle Cadieux, Que. oi

Replacing or Reaction Relation of the Minerals of the Rock.

The use of the word replacing in the foregoing discus-
sion and the division of the minerals of the rock into the
two classes, original and replacing, should not mislead the
reader into believing that the two sets of minerals are
considered to have formed under markedly different con-
ditions separated by an abrupt change. The chrysolite
and augite may very well have formed from a somewhat
alkalic liquid, but their separation, with decreasing tem-
perature, rendered the liquid so much more alkalic that
finally they became unstable in contact with such hquid
and were replaced by melilite and biotite with monticellite
as an intermediate step. It seems necessary to suppose,
however, that the crystals of chrysolite and augite had
accumulated locally and that there was only a relatively
small amount of alkalic liquid, so that the composition of
the mass as a whole was dominated by the richness
in chrysolite and augite and the new minerals formed
were determined mainly by this factor. In facies of the
rock where chrysolite and augite had not accumulated
in such large amounts the composition of the whole mass
was characterized by the greater richness in the alkalic
hquid and the new minerals formed from the augite and
chrysolite are alkalic pyroxenes, amphiboles and mica,
changes which have been described by those who have
studied the essexites and other rocks of this particular
provinee.!®

- The essexites, alkalic pyroxenites and other ultrabasic
rocks of this province, that have received a host of locality
names, have considerable interest in the present connec-
tion. The experimental results show that the fields of
the basic silicates, olivine, melilite, etce., extend over
beyond the join nephelite-diopside (metasilicate). The
fields of the metasilicates themselves are, of course,
crowded over in front of them to still more siliceous
mixtures. The reaction relations of the alnoite minerals
indicate similar conditions in the natural magmas. The
suggestion is strong, therefore, that some of the basic
rocks of this province, in so far as they consist of augitic
pyroxene with some nephelite, could not have formed
from liquids of their own composition but must have

“In particular see G. A. Young, The Geology and Petrography of Mt.
Yamaska, Geol. Surv. Can. 16, p. 25H, 1906.

Gao. N. L. Bowen—Genetic Features of

formed by accumulation of their crystals from much more
salic liquids (nephelite syenites, etc.). ‘The remarkable
development of flow structure in these rocks, which is ©
summarized for the whole province by O'Neill, 20 might
therefore be regarded as a natural result of their mode
of origin. If, as is suggested, they do not come into
existence except as a result of a considerable degree of
erystal accumulation of the appropriate kind, the devel-
opment of flow structure, when such a mass is moved
1s inevitable.

Formation of Inme-rich Minerals in Alkalic Rocks.

The series of figures (1, 3, 4, and 5) serves to emphasize
the fact that in the presence of considerable nephelite
the very basic silicates forsterite, melilite and possibly
monticellite separate from liquids well over on the silica
side of the nephelite-diopside join. The presence in a
rock of the lime-rich mineral melilite need not, therefore,
be considered as indicating the existence of a magma par-
ticularly rich in lime but rather of an alkali-rich magma
from which melilite-bearing facies might be formed by
differentiation (crystal accumulation). Moreover, the
intimate association of melilite and nephelite rocks in
nature is to be regarded as further evidence of the impor-
tance of crystallization-differentiation, melilite being
a normal cafemic constituent of nephelite-bearing rocks.

Kiven such minerals as garnet and vesuvianite, closely
related as they are to melilite,2* may, perhaps, be capable
of formation as normal constituents from alkalic magmas
and are not necessarily an indication of contamination of
the magma by lime-rich rocks such as limestone. These
very basic silicates may conveniently be regarded as the
normal basic minerals of rocks (mainly metasilicates)
desilicated in the presence of alkalic molecules such as
nephelite. It will be noted that this action, which has
been demonstrated for the artificial mixtures, is exactly
the opposite of that which Daly supposes to take place in
magmas, viz., the desilication of feldspathic molecules by
lime with formation of feldspathoids. This fact should
not, however, in itself, be urged as necessarily disproving

** Geol. Surv. Can. Memoir, 43, p. 22, 1914.
** The sarcolite molecule of melilite is also a garnet molecule.

Alnoitic Rocks at Isle Cadieux, Que. 29

Daly’s assumption, for it is conceivable that we are
dealing with a condition of equilibrium whose position
may be varied under different conditions. :

The occurrence of monticellite in alnoite is another
example of the same sort of thing. It is a mineral which,
like vesuvianite and garnet, is found in metamorphosed
limestones, but it is commonly believed to be confined
to such a mode of occurrence. However, there is in the
collections of Dr. H. S. Washington a series of slides
from the original Alno locality,?? only one of which is of
alnoite (labeled fine-grained alnoite, Aldersnaset, Alno)
and in this slide I find that many of the chrysolite grains
are surrounded by rims of monticellite, always in optical
continuity with the chrysolite. Monticellite embays
chrysolite and occasionally fingers through it. (See
Plate I(f).) The reaction relation is very plain. In fact,
while one cannot speak with confidence from the evidenee
of a single slide, there is much in this alnoite of the type
locality suggestive of the same reactions (replacements )
that have been detailed here for the Quebec occurrence.
It seems not improbable that monticellite may occur in
other examples but has been overlooked. It would be
rather easy to do so in the presence of ordinary olivine
(chrysolite), particularly in cases where their manner
of occurrence was not so distinctive as in the present case.
Indeed, O’Neill deseribes ‘‘an unknown colorless min-
eral,’’ occurring in a much more salic type of rock in this
same Monteregian province, whose properties as given
agree excellently with those of monticellite. It occurs
in a nephelite-sodalite syenite close to its contact only.”
If it is monticellite its occurrence in the quickly chilled
contact variety of the rock and its absence elsewhere is as
one might expect, for in the more slowly cooled part its
place is taken by other minerals just as ordinary olivine
may occur in the contact facies of, say, a diabase which is
elsewhere free from olivine.

On the whole, then, it seems not unlikely that mon-
ticellite may be of reasonably frequent occurrence in
alkalic rocks. Indeed this is the more probable since the

~The specimens were obtained from Dr. Hégbom. A duplicate of this
collection is in U. 8S. Geological Survey Petrographic Reference Collection,
where this rock is No. 1044.

* St. Hilaire and Rougemont Mountains, Quebec, Geol. Survey. Can.,
Memoir 43, p. 42, 1914.

30 N. L. Bowen—Genetic Features of

pure lime olivine itself (8Ca,SiO,) is known to, occur in
such rocks, though it has frequently been stated that
this compound is unknown as arock mineral. F. P. Paul
deseribes an unknown mineral in a nephelite-eudialite
basalt from Shannon Tier, Tasmania.** His chemical
determinations are sufficient to prove that the mineral
is a lime-silica compound high in lime and free from
magnesia, and he himself decides that it is probably
Ca,Si0,. That he is certainly correct is demonstrated by
the correspondence of properties with those of artificial
BCa,8i0,,°° which is shown below.

KS a 2V Sign

Cleavage yt
BCa,Si0, poor 1.735 1.717 large +
‘Tasmanian mineral 2.olivine-like 1.746 1.718 64° +

Petrographers appear to have entirely overlooked this
occurrence of lime olivine.2° Its nearest relative among
recognized minerals is hillebrandite, Ca,SiO,.H,0.% The
occurrence of Ca,SiO, as a mineral in an igneous rock
may be regarded as further justification for the use of
this molecule in calculating the norm.”®

The compound Ca,S8iO, occurs in three forms. The
B-form with which the natural mineral corresponds is
stable only between 1420° and 675°, and when pure it
changes at room temperature (or higher) to the y-form,
the change being accompanied by such increase of volume
that the whole mass is reduced to powder. When embed-
ded in a mass of glass or other minerals this change is
prevented and in such a case it will persist and its proper-
ties can be studied at room temperature. Its occurrence
as grains in the natural rock is, therefore, not contrary to
laboratory experience. It may be noted that the artificial
mineral is strongly attacked by water with solution of
hme and it would be of interest to test the Tasmanian
rock in this particular. According to Paul the natural
mineral is readily attacked by very dilute HCl.

The rock in which this mineral occurs is a local varia-
tion of melilite-nephelite basalt. A reference to fg. 1
shows that the field of Ca,SiO, borders against the aker-

*T. M. P. M., 25, 309, 1906.

* Day, Shepherd and Wright, this Journal, 22, 295, 1906; also Rankin and
Wright, ibid., 39, 75, 1915.

* For example, in Rosenbusch, Mikroskopische Physiographie (1908), p.
1449, it is still described as an unknown mineral. ;

“ 'H. K. Wright, this Journal, (4) 26, 551, 1908.

“ H. S. Washington, J. Wash. Acad. Sci., 5, 345, 1915.

Alnoitic Rocks at Isle Cadieux, Que. Sul

manite (melilite) field and the composition of akermanite
is quite close to the border. In view of the demonstrated
effect of nephelite in pulling the fields over towards the
silica corner the occurrence of Ca,SiO, in such a rock is
perhaps not surprising.

Analcite Dikes.

The formation of very basic minerals by a process
which may be briefly, if somewhat inadequately, described
as desilication has been demonstrated for the case of
melilite in the artificial mixtures. The equations that
have been written to show the formation of the charac-
teristic melilite molecules from diopside show also, of
necessity, a molecule that is formed by silication of the
nephelite molecule; and the actual experimental results
show the formation of a liquid which refuses to crystal-
lize under dry melt conditions and has, as a glass, the
very low refractive index appropriate to a more sili-
ceous sodium aluminum silicate. This has been written
NaA1Si,O,. Now alnoite is a monchiquite, and as such
often contains analcite as a base or a glassy base of the
composition of analcite, viz., NaAlSi,O,.H,O. The Isle
Cadieux alnoite shows none of this, but Harvie has found
it, or alteration products that may be referred to it, in
alnoites of this vicinity.*® In this analcitic material we
may, then, have the other product of the reaction of
nephelite-rich liquid on augite to produce melilite. It
appears here as a hydrous mineral, for the whole system
was evidently quite rich in water, as witness, the abun-
dance of biotite.

In discussing the replacement of the original minerals
of the alnoite, it has been stated as probable that this
was accomplished by a nephelite-rich liquid that passed
‘through the partly consolidated rock filling the interstices
with reaction minerals and carrying on some of the reac-
tion products.*° These latter were presumably largely
analcite and it may be to such liquids that the analcite
dikes of this vicinity are to be referred.*!

” Op. cit., p. 274. |

*° In this connection the arrangement of the small melilites about the older
erystals is significant. They have a ‘‘stream-like disposition,’’ as Rogers
and Du Toit describe a similar structure in African melilite basalts. Geol.
Com., Cape of Good Hope, Ann. Rep., 1903, p. 47.

7 F. D. Adams, Twelfth International Geologic Congress, Canada, Guide
Book No. 3, p. 45.

32 N. L. Bowen—Genetic Features of

Such straining or filtering of liquid from erystals I
have elsewhere suggested as a dominant factor in the
formation of those unusual types that occur only as com-
plementary or diaschistic dike rocks.** A change in the
nature of a dike along its strike is a feature that might |
be expected on this supposition, and Shannon has recently ~
found such changes in lamprophyres of Idaho.??

For a discussion of certain reaction effects in analcitic
rocks the reader is referred to papers by Tyrrell?* and
by Scott.®?

Summary.

Alnoitie rocks at Isle Cadieux near Montreal, Canada,
are found to consist principally of monticellite alnoite,
a hewly recognized (not a new) rock type. Besides this
type there is a variety consisting almost exclusively of
melilite and biotite, which is apparently new.

A study of the relations of the minerals indicates that
the rock originally consisted of augite and chrysolite and
was nearly completely consolidated as such. These min-
erals were then attacked, with lowering temperature,
probably by their own interstitial liquid as it changed in
composition, and were replaced by monticellite, melilite
and biotite with marialite, perovskite and titaniferous
magnetite as minor produets of the reaction. The monti-
cellite is itself replaced by melilite and biotite and the
melilite-biotite rock is the end product of the replacement.

The replacement was accomplished by an _ alkalic
liquid (magma) which formed monticellite from augite
by desilicating it and later gave rise to melilite and the
more definitely alkalic mineral biotite.

The melilites are both positive and negative and are
sometimes conspicuously zoned, with a positive core, an
isotropic intermediate zone, and a negative rim. 7

In an experimental part of the paper equilibrium in
mixtures of nephelite and. diopside is determined. It is
found that from intermediate mixtures forsterite and
melilite are the first products to crystallize and the

* Proe. Nat. Acad. Sci., 6, 161, 1920:

* Proc. U. 8S. Nat. Museum, 57, 1920, p. 480.

*“G. W. Tyrrell, Quart. Jour. Geol. Soc., 72, pt. 2, 84-131, 1917.

*° Alex. Scott, Trans. Geol. Soc. Glasgow, 16, pt. 1, 34-45, 1915-16.

Alnoitic Rocks at Isle Cadieux, Que. 35

melilites so formed are analogous to natural melilites in
composition and optical properties.

It is thus proved experimentally that nephelite reacts
with diopside to form melilite, a reaction analogous to
that which is considered to have taken place between
augite and alkalic liquid in the natural rocks.

This reaction is of the nature of a desilication of the
diopside, and while the formation of monticellite in this
manner has not been demonstrated in the relatively sim-
ple experimental mixtures, the demonstration of desilica-
tion of diopside in these mixtures is believed to give
support to the idea that monticellite is so formed from
augite in the more complex natural mixture.

An outstanding difference between the natural, replac-
ing (reacting) liquid and the artificial mixtures is the
presence of potash and water in the former. Asa result
of this difference the reaction products instead of being
forsterite and melilite were, in the natural rock, biotite
and melilite with monticellite as an intermediate step.
In fact chrysolite, originally present, was itself a prin-
cipal source of biotite. The liquid, as modified, by the
reaction, passed on and possibly gave rise to analcite
dikes. © . |

It is probable that monticellite occurs fairly frequently
aS an igneous rock mineral in rocks related to those
described. Re-examination of the original alnoite shows
its presence there with, apparently, a similar relationship.
Hiven the pure lime olivine itself (8Ca,SiO,) has been
found in a related rock from Tasma, ‘ia, a fact to which
attention is here directed because it appears to have been
overlooked in general petrologic literature.

Such lme-rich minerals, including also melilite and
the related minerals garnet, vesuvianite and others, may
be considered to have formed in alkalic magmas as the
result of desilication of the more normal cafemic
molecules (metasilicates). Their presence in alkalic
_ rocks may therefore be the result of normal equilibrium
processes in the magma and not of addition to the magma
of lime-rich rocks (limestone, etc.).

Geophysical Laboratory,

Carnegie Institution of Washington,
May, 1921.

Am. JOUR. Se Saw Series, VOL, III, No. 13.—Janvuary, 1922.

on N. L. Bowen—Alnoitic Rocks.

EXPLANATION OF PLATE I.

a. Zoned melilite with isotropic core and birefracting (negative) rim ;
x20. Melilite-biotite rock, Isle Cadieux.

b. Remnants of a single pyroxene crystal (P, P) in matrix principally
of biotite; x67. Monticellite alnoite, Isle Cadieux.

c. Resorbed chrysolite crystals (chadacrysts) in matrix (a single
oikokryst) of biotite showing rims of ore at borders of chrysolites; x 33.
Alnoite (approaching mica peridotite), Isle Cadieux.

d. Resorbed pyroxene crystal (P) in matrix principally of monticellite;
x20. Monticellite alnoite, Isle Cadieux.

e. Blocky replacement of a pyroxene crystal by biotite; x20. Monticel-
lite alnoite, Isle Cadieux.

f. Resorbed chrysolites (c) surrounded by similarly oriented monticel-
lite (M); x33. Alnoite from Aldersniset, Alné.

Buddington—Natural and Synthetic Melilites. 35

ART. 86); Some Natural and Synthetic Melilites; by
A. F’. BuppiIneTon.

INTRODUCTION.

The melilite group is usually considered to include the
minerals akermanite, gehlenite, humboldtilite, sarcolite,
fuggerite, and the ferric iron-rich melilte originally
called melilite. Many different theories have been pro-
posed to explain their composition. One of the most
recent is suggested by Schaller,| who has reconsidered
previous theories and all the available analyses of the
natural minerals and has arrived at a new theory of
their composition. His statement, in brief, is that all the
suitably described members of the melilite and gehlenite
group can be satisfactorily explained as isomorphous
mixtures of the primary compounds, sarcolite (3Ca0O.
A1,0;.5810,), a hypothetical soda-sarcolite (8Na,0.A1,O,.
3810,), velardenite (2CaO.Al1,0;.Si0.), and akermanite
(4M¢0.8CaO.9Si0.). He further suggests the possibil-
ity that a ferric sarcolite (8CaO.Fe,0,.38Si0,) may be
another component of the melilites. 3

The present work was undertaken to test this hypothe-
sis, and it is a pleasure to state that the results of the
laboratory study on pure mixtures confirm the more
essential features of Schaller’s theory. The present
work, however, indicates that the melilites are too complex
to be wholly explained in this way and that several modi-
fications are necessary. These concern, in part, the com-
position of the mineral akermanite, and the polymorphic
character of the 3RO.R,0,;.3Si0, compounds and certain
solid solutions into which they enter.

Schaller states that the composition of akermanite is
4M¢0.8Ca0.9Si0,. Ferguson and Merwin? in their
study of the ternary system CaO-MgO-SiO, were unable
to prepare a compound or solid solution having this for-
mula. On the other hand, they prepared a compound
2CaQ.Mg0O.2810, whose crystal system and essential
optical properties corresponded to those of the mineral

*W. T. Schaller, U. S. Geol. Surv., Bull. 610, pp. 106-128, 1916. When-
ever the work of Schaller is referred to in this paper, the reference is that
given here.

* J. B. Ferguson and H. E. Merwin, this Journal, 48, 118-122, 1919.

36 Buddington—Natural and Synthetic Melilites.

akermanite. They therefore suggested that the compound
2Ca0.Mg0.2Si0, found by them is the synthetic equival-
ent of akermanite. Later, Ferguson and Buddington?
found that the compounds 2CaO.Mg0O.2S8i0, and 2Ca0O.
Al1,O,.8i0, form a complete isomorphous series of solid
solutions. This confirmed the suggestion of Vogt* that
akermanite and gehlenite form a complete series of solid
solutions, and also serves to indicate the correctness of
the formula assigned to akermanite by Ferguson and
Merwin. The same series of analyses of natural min-
erals used by Schaller in his study has been recalculated
on the assumption that 2CaO.MgO.28i0, is the correct
formula for akermanite, and the results have been found
to check equally well with those of Schaller for the hum-
boldtilite and gehlenite varieties, but not for the ferric
iron rich melilites or for akermanite itself. In this
paper the term akermanite will be applied to the com-
pound 2CaO.Mg0O.2Si10,, since the balance of the evidence
seems to favor this interpretation.

- Schaller has proposed that the compound 2CaQO.Al1,0O3.
Si0, be called velardenite, since the natural gehlenites are
not definite compounds but variable solid solutions, and
since the gehlenite of Velardena, Mexico, approaches
more nearly than any so far analyzed to the pure com-
pound, 2Ca0.Al,0,.Si0,. The term gehlenite, however,
has been used by Vogt® to designate one of the indepen-
dent primary members of the melilite group, and this
usage 1s common in the literature. The pure compound
2CaQO.Al,0;.810, was first prepared by Shepherd and
Rankin® and was called gehlenite.

The term albite is used both for the pure end member
and also for any plagioclase between Ab and Ab,An,.
Similarly, the term gehlenite may be used for the pure
end member and for solid solutions within certain limits.
The compound 2CaO.AI,0;.Si0, is therefore called gehle-

nite in this paper, and is the equivalent of the velardenite
of Schaller.

* J. B. Ferguson and A. F. Buddington, this Journal, 50, 131-140, 1920.

*J. H. L. Vogt, Mineralbildung in Schmelzmassen, p. 96, 1892. Vogt gives
the formula (Ca,R;);.(Al,Fe,)Si,0,. for gehlenite and (Ca,R),Si,0, for
akermanite.

tel pets Voet, doe. eit,

*E. S. Shepherd and G. A. Rankin, Preliminary report on the ternary
system CaQ-Al,0,-SiO.; a study of the constitution of portland cement
clinker, with optical study by F. E. Wright, J. Ind. Eng. Chem., 3, 224, 1911.

Buddington—Natural and Synthetic Melilites. 37

A statement showing the names of the minerals
equivalent in composition to the compounds used in this
investigation is given in Table LI.

PABEH L.

Formule of compounds and names of the mineral or synthetic
equivalent.

Compound Mineral equivalent
2CaO.Mg0.2Si0, Akermanite
2Ca0.Al1,0,.S810, Gehlenite
3Ca0.Al1,0,.3810, Grossularite
3CaO.Fe,0,.3810, Andradite
3Na,0.Al,0,.38i0, Lagoriolite*
(90 (3Ca0. Al, O,.38i0,) ) :

110 (3Na,0.A1,0,.38i0,) § eercorite

*Prepared and named by Morozewiez. Known only in synthetic prepara-
tions.

CHOICE OF SoLip SOLUTIONS FOR INVESTIGATION.

Akermanite and gehlenite had previously been found
to form a true binary system, but each of the three com-
pounds 3Na_,0.A1,0;.3810, ; 3CaO.AI,0;.3810,; and 3Ca0O.
Fe,0,.3810, is unstable at or below its melting point, and
most of the solid solutions of these compounds with
akermanite and gehlenite which were investigated are
likewise unstable at their melting point; so that these
solutions as a whole, from the standpoint of the phase
rule, must be treated as a part of a six-component system
embracing their constituent oxides. No attempt has
been made in this work to study the solid solutions of
2CaO.Mg0.2810., 2CaO.Al,0,.810,, and the 3RO.R,0,.
3810, compounds as a problem in itself. This can be
done only after the many simpler systems embraced
within the six-component system of which these solid
solutions form a portion shall have been worked out.
Only a few of the possible solid solutions of the com-
pounds under consideration have been studied. These,
numbering in all about 100, were chosen with the objeci
of obtaining data to explain the melilite group of min-
erals as now known from recorded analyses. No effort
has been made to determine the nature of the dissocia-
tion products of the unstable solid solutions, nor of the
other phases present in those mixtures exhibiting
‘inhomogeneity.

38 Buddington—Natural and Synthetic Melilites.

Schaller, from a study of the natural gehlenites, showed
that the major components are gehlenite, akermanite,
and 3CaO.A1,0,.3810,. Accordingly, these three com-
pounds were chosen as end members and their relations
were studied. They were found to have a limited field
of solid solution. The composition 90% 3Ca0.Al1,03.
38Si0 + 10% 3Na,0.A1,0,.38810,, corresponding to the
composition of sarcolite, was then taken as another
end member, and its relations in mixtures of 2CaO.MgO.
28i0, and 2Ca0.Al1,0,.Si0, were studied. A field of
complete solid solution was found for these mixtures,
with the exception of a doubtful zone close to the 2CaO.
Mg0.2810, end.

Schaller, from a study of the humboldtilites and the
original melilites of Capo di Bove, had shown that the
major components of these minerals were the compounds
3Ca0.A1,0;.3810,, 2CaO.MgO.2810., and 3CaQ0.Fe,Q;.
38i0,, with minor amounts of 3Na,0.A1,0,.3810, and
2CaO.Al1,0,.8i10,. The soda compound varies from 8 to
12 percent, averaging about 9 percent. Accordingly,
a system was studied in which the percentage of 3Na,0.
Al,O,.38810, was maintained constant at 10 percent, and
the percentages of the compounds 2CaO.Mg0O.28i0,,
2CaO.Al,0;.810,, and 3CaO.Fe.0;.88i0, were varied.

Another system was studied in which all five com-
pounds were introduced, the percentage of the soda com-
pound being maintained constant at 10 percent and 2CaO.
Al,O;.8i10, at 20 percent, while the percentages of the
other three compounds were varied.

The compositions of all the analyzed natural minerals
of the melilite group, except one, approximate to com-
positions embraced within or between the systems thus
studied. A synthetic preparation corresponding in com-
position to the mineral fuggerite, which is classed with
the melilites but falls outside these compositions, was
studied separately. 3

Three analyses of natural humboldtilites made for the.
writer by Dr. H. 8. Washington showed that a ferrous
iron compound was an accessory component constituting
up to 15 percent or more of this mineral. Apparatus
for dealing with ferrous iron compounds must be devel-
oped before this component can be studied; it is there-
fore not considered in this work.

Buddington—Natural and Synthetic Melilites. 39

MeEtTHODS.

Six constituents, CaO, Na,O, MgO, Al,O;, Fe,O;, and
‘$i0., enter into the composition of the solutions dealt
with. Chemically pure calcium carbonate, anhydrous
sodium carbonate, silica, and ferric oxide were used
in making up charges. Pure magnesium oxide was
obtained by heating basic magnesium carbonate in a
Fletcher gas furnace, leaching to separate alkalic impuri-
ties, and reheating. ‘The aluminum oxide was purified by
boiling with a solution of ammonium chloride followed
by heating in the Fletcher furnace.

Charges of a definite composition were made up by
mixing together the constituent oxides or their equiva-
lents in proper weights to give a ten gram portion.
These were then ground for one hour in a mechanical
grinder.

All charges containing ferric oxide were melted in a
platinum crucible suspended in a platinum resistance fur-
nace at temperatures of from 1400° to 1450° and were then
quenched to form glasses. Charges were melted three
times with intervening, grindings to a fineness sufficient
to pass a sixty-five mesh sieve.

Charges free from ferric oxide were made up by using
the Fletcher blast furnace in place of the electric furnace.
This treatment gave homogeneous glasses that were used
in subsequent experiments.

In their work on the system lime-ferric oxide, Sosman
and Merwin‘ found that a certain amount of ferrous iron
is formed in all melted mixtures of CaO and Fe,O.,.
When the molecular percentage of Fe.O, was less than 50
percent in mixtures heated to 1400°-1500°, they found
the percentage of ferrous oxide was small. In order to
ascertain the possible extent of dissociation in the ferric
oxide mixtures used in the course of this work, two
charges of prepared glasses were analyzed for ferrous
iron by Dr. H.S. Washington. One preparation contained
0.14 percent of Fe,O, and showed 0.36 percent of FeO;
the other contained 14.1 percent of Fe,O, and 1.19 percent
of FeO. The optical and thermal data for mixtures rich
in ferric iron, therefore, need a slight correction for
dissociation.

*R. B. Sosman and H. E. Merwin, J. Wash. Acad. Sci., 6, 532-7, 1916.

40 Buddington—Natural and Synthetic Melilites.

The quenching method was employed throughout the
research. Details of this method may be found in many
of the previous publications of this laboratory. A dia-
gram of the type of furnace used in this work may be
found in the paper by Ferguson and Merwin® on the
ternary system CaO-MgO-Si0,. The melting points
used in the calibration of the thermoelements were those
of: gold 1062.4°, diopside 1391.5°, akermanite 1458°,
anorthite 1550°, and gehlenite 1590°. | :

In most cases, crystals used for the determination of
their optical properties were prepared by holding the
proper glass over night at a temperature just below the
dissociation point, or the beginning of melting. A very
great difference was found in the character of the crystal-
lization. Mixtures with 50 percent or more of the 3CaO.
Al,O,.8810, compound crystallized in fibrous aggregates
which did not grow appreciably with continued heating.
If these charges are heated above the dissociation point,
so that a little liquid is obtained to facilitate recrystal-
lization, the resulting aggregates are mottled throughout
with the dissociation products and are unsatisfactory,
unless the material is crystallized-so far above the solidus
that a change in their composition results and the opti- -
cal properties of the crystals obtained are materially
changed. The best crystallized preparations of such com-
positions were obtained by heating the glasses 20 hours
or longer somewhat above the lowest temperature at
which the glass will reerystallize. The optical determi-
nations of the indices of refraction made on mixtures
rich in the 8CaO.A1,0,.38i10, compound may be in error
as much as + 0.005. Glasses with less than 50 percent of
the 3CaO.Al1,0..3Si0, compound and as little as 20 per-
cent of the 3CaO.Fe,0,.3810, compound give very good
crystalline material if held for fifteen minutes to an hour
at a temperature a little below the solidus. The glass of
composition containing 90 molecular percent of 3Ca0O.
Al,O;.38i0, and 10 percent of 3Na,0.A1,0,.38i0, was
held for one week at a point just above the solidus to
obtain crystals free from air for the determination of the
specific gravity. Some glass and dissociation products
were present in this material. :

“J. B. Ferguson and H. E. Merwin, op. cit., p. 88.

Buddington—Natural and Synthetic Melilites: 41

The succeeding tables saeenh only selected data which
are considered significant for the problem. The indices
of refraction are ‘all for sodium light. Where indices of
- yefraction are given for mixtures “which are slightly dis-
sociated, the observations apply to the dominant erystal
phase.

MIXTURES OF 3Ca0.Al,0,.38i0, ; 2CaO.Mg0O.28i0,, AND
2CaO0.Al1,0,.S10,.

A preliminary series of experiments made on mixtures
of 2CaO.Mg0.2Si0., 2CaO.Al1,0;.8i0., and 3Ca0.Al,0;.
3810, gave unsatisfactory inhomogeneous crystalline
ageregates at the temperatures of the experiments.
Work on these mixtures, accordingly, was temporarily
abandoned. The constant presence of a sodium com-
pound in the natural melilites and the existence of the
natural mineral sarcolite, containing about 10 percent of
the compound 3Na,0.A1,0,.38S10,, suggested that mate-
rial of the composition of sarcolite be tried in place of the
ealclum compound 3Ca0.Al,0;.38i0,. In the progress
of this later work it was discovered that certain mixtures
which were heterogeneous at temperatures just below
the solidus were homogeneous at lower temperatures.
With this information as a guide, work on mixtures of
2CaO.MzeO0.28i0,, 2CaQ.Al,0,.810,, and 3CaQ.Al,Q..
3810., was resumed and significant results were obtained.

Mixtures of 2CaO.MgO.2S810, and 3CaO.A1,0,.3S810,,
which contained 70 percent or more of the former com-
ponent, are homogeneous below the dissociation point
except for a few grains with a much higher index of
refraction. Whether these grains actually mean an
incomplete solid solution or are the result of slight
impurities in the melt, or whether they are compounds
formed at lower temperatures and existing in unstable
equilibrium at higher temperatures, was not ascertained.

It has been previously shown® that the compounds
2CaO.Al1,0;.Si0, and 2CaO.Mg0O.2Si0, are isomorphous
and form a complete series of solid solutions with a
minimum melting point. In table II the optical charac-
ters and thermal data for mixtures of these two compo-
nents are given, and in fig. 1 the indices of refraction are
plotted against the composition in weight percent. The

° J. B. Ferguson and A. F. Buddington, loc. cit.

42 “Buddington—Natural and Synthetic Melihites.

TaBLeE II.

Optical and thermal data for mixtures of 2CaO.A1,0,.Si0.; and 2Ca0.MgO.
2Si0,. Compositions expressed in weight percent,
temperatures in degrees Centigrade.

2CaO. 2CaO. Crystal Glass
ATO}. MgO. All - trace =" trace
Si0,. 2 810, Nw Ne erystal Glass erystal All glass
100 0 1.669 1.658 1590°
80 20 1.664 1.657 1498° 1502° 15oo 15a
60 40) 1.657 1.654 1441° 1445° 1506° 1512°
50 50 1.653 ~ 1.652 1475° 1478°
40 60 1.648 1.649 1400° 1404° 1430° 1432°
30 70 1.645 1.648 1390° 1392° 1394° 1397°
25 75 1.643 1.647 1385° 1388° 1390°
10 90 1.637 1.643 1416° 1420° 1424° 1427°
0 100 1.632 1.639 1458°

optical properties are a continuous function of the com-
position, and crystals of a certain intermediate mixture
are isotropic for sodium light, constituting a transition
phase from the positive akermanite to the negative
gehlenite.

Fig. 1

w 1670 , “O70
S =

> 4660 1-660
NN) ,

tC

> O50 1-650
Ne

S

» 640) 1-040
Ny

WS

S 7630 1-630.

0 40. 20° W3O "AORPEO GSCO) 170 RNeOrn Commas
2CAO-MGO25i0Q WT-PERCENT 2C€2@0-Alz03-St Op

1.—Curves for refractive indices plotted against composition in weight
percent for mixtures crystallized just below the solidus.

From the data given by Rankin and Wright?® in their
study of the ternary system CaO-Al,0;-SiO,, it is found
that the point of beginning of melting for all mixtures of
2Ca0.Al1,0;.8i0. and 3CaQ.Al,0;.38i0, is the eutectic
melting at 1265° and that no solid solution exists above
this point. Mixtures of these two compounds are the
only ones investigated in the present work which failed
to show evidence of some solid solution above the begin-
ning of melting, either accompanied 6r unaccompanied by

*G. A. Rankin and F. E. Wright, this Journal, 39, 1-79, 1915.

Buddington—Natural and Synthetic Melilites. +48

dissociation. Below the solidus for mixtures of these
two components, however, solid solution does exist for
limited ranges of temperature and of composition.

Ap ey Cl

Optical and thermal data for mixtures of 2CaO.Al1,0;.8i0, and 3Ca0.A1,0;.
3S8i0,. Compositions expressed in weight percent and
temperature in degrees Centigrade.

2CaO. 3CaO. One Trace of Glass
A1,0; Al,O3.
Si0, 3810, Ne ne phase ciation crystals ll glass
FOOe= 0 1.669 1.658 1590°
75 25 1.658 1.650 TAS 2 1 02 1566° 1574°
@ishra)je-4@l hire)
50 50 1.648 1.642 ate? 1150° 1524° 1530°
Givhrs) ss: Gehrs)
35 65 1.643 11.635 Trace dissoc. 1484° 1490°
(16 hrs. at 980°)
20 80 1.642 11.633 Inhomogeneous 1428° 1433°
(16 hrs. at 980°)
0 100 Inhomogeneous 1340° 1347°

(16 hrs. at 980°)

*The dots thought to represent dissociated products are so small in
amount, and so minute, that this determination is donb and this
mixture may be homogeneous up to the eutectic.

t Determined on major crystal phase.

In Table III are given the optical characters and
thermal data for the mixtures of these two compounds
which were investigated. Owing to the very finely fibrous
character of the preparations crystallized some distance
below the eutectic, it is very difficult in many cases to
determine the homogeneity or inhomogeneity of the
material under examination. For this reason no impor-
tance is attached to the temperatures given as marking
the beginning of dissociation, and the limits of solid solu-
tion are likewise only approximations. The end compo-
nent, 3CaO.Al,0,3810, when held for 16 hours at 980°
gave a fibrous aggregate which was conspicuously inho-
mogeneous. Filaments of a mineral, presumed to be
wollastonite, are prominent. A mixture containing 20
percent of 2Ca0.Al,0,.Si0,, heated for 16 hours at 980°,
likewise gave a conspicuously inhomogeneous fibrous
aggregate, but not to such an extent as the end member
alone. With an increasing percentage of 2CaQ.A1,O3.
SiO, the evidence of inhomogeneity decreases and in a
tans of the composition 35 percent 2CaQO.A1,0,,.Si0,-65

44 Buddington—Natural and Synthetic Melilites.

percent 3CaO.A1,0,.3810,. these is a mere trace of inho-
mogeneity in material held for 16 hours at 980°; and a
mixture of equal parts of the two components, treated at
the same temperature for the same length of time, appears
perfectly homogeneous. The upper range of homo-
geneity for this latter mixture, however, is limited, and
dissociation products appear in preparations heated for
one hour at 1150°, or over 100° below the eutectic. Ina
mixture of 75 percent 2CaO.Al,O,.Si0, and 25 percent
3CaQO.Al,O,.38i0, the beginning of dissociation is not
well marked, and the material may be homogeneous up
to the eutectic. In summary, then, mixtures of 2Ca0O.
Al,O;.810, and 3CaO.AIL,0;.3810, exhibit a limited range

Fig. 2.

lndex of Refraction

2CaO0-AlyO;- Si O> WT PERCENT SCAOAlz 03-351 Op

2.—Curves for refractive indices plotted against composition in weight
percent for mixtures which show complete homogeneity when crystallized at
about 1000°. Indices of refraction for dominant phase of inhomogeneous
mixture are indicated where determined.

for complete homogeneity extending from the 2Ca0O.
Al,O;.SiO, component to a little beyond a 50-50 mixture,
and within a temperature range that is limited upwards
by dissociation or the eutectic point, and of unknown
lower limits. Partial solid solution is found in mixtures
containing so much as 80 per cent 3CaO.A1,0,.3Si0., as
is indicated by the indices of refraction determined on
the major constituent of preparations crystallized at
980°, and may extend further. But a break in the series
of completely homogeneous mixtures towards the 3Ca0O.
Al,O;.38810, component beyond a 50-50 mixture is indi-
cated by a break in the curves for indices of refraction.
In fig. 2 the indices of refraction are plotted against com-

Buddington—Natural and Synthetic Melilites. 45

position in weight percent, and the curves connect those —
mixtures which show complete homogeneity for a tem-
perature range above 980° (the lower working limit
under the conditions of the experiment). The indices of
refraction for the major constituent of inhomogeneous
mixtures crystallized for 16 hours at 980° are also
plotted. | |

Since dissociation takes place at or before the eutectic,
thermal data for the liquidus and solidus cannot be used
as evidence of solid solution. |

None of the crystallized mixtures of 2CaO.Mg0O.28i0,
and 3Ca0.Al1.0;.3Si0, was found to be absolutely homo-
geneous within the temperature range explored. Mix-
tures containing 70, 80, and 90 percent of the 2CaO.Mg0O.
28i0, compound, however, showed only a trace of some
included, minute, more highly refracting, low-birefringent
grains, and are essentially homogeneous for a range below
the dissociation point, or the point of beginning of
- melting, as the case may be. The lowest temperature at
which experiments were made was 980°, so that nothing
is known of the relations below this point. A mixture
containing only 10 percent of the 3CaO.Al,0;.3810, com-
pound appears homogeneous up to the point of beginning
of melting (1417°+6°) except for the qualification noted
above. Dissociation (in addition to the trace of grains
noted above) in mixtures richer in this compound, how-
ever, takes place much below this, and those containing
do percent 3CaQ0.Al,0;.38i0, show a trace of interstitial
inhomogeneity at 980°; those containing a greater per-
centage of this compound show still greater inhomogene-
ity at the same temperature. In Table IV the optical
characters and the thermal data for mixtures of these
two compounds are given. Owing to the lack of positive
evidence of the complete homogeneity of any of the pre-
parations examined, no curves have been drawn to show’
the relation of the optical characters to the composition.
It is believed, however, that the optical characters given
are those of material essentially the same in chemical com-
position as would be indicated by the mixture designated.

Mixtures of the three components 2Ca0.Al,0.,.Si0,,
38CaQ.Al,0;.38i0, and 2CaO.MgO.2Si0O, form an incom-
plete series of solid solutions, completely homogeneous
in the field along the 2CaO.A1,0,.810,-2CaO.Mg0.2Si0,

46 Buddington—Natural and Synthetic Melilites.

TABLE LV.

Optical and thermal data for mixtures of 2CaO.Mg0O.2Si0, and 3Ca0.A10;.
3Si0,. Compositions expressed in weight percent and
| temperatures in degrees Centigrade.

3CaO. 2CaO. *Hssentially *Trace of Glass

Al,O,,. MgO. one crystal disso- +
* 3810, 2810, No Ne phase ciation crystals All glass
0 100 1.632 1.639 1458°
10 90 1408° 1420° . 1525°
20 80 1.634 1.639 1215° 1240° 1403° 1409°
(1 hr.) (% kr.)
30 70 1.629 1.632 1200° 1240° 1380° 1386°
(Ghalires)) (% hr.)
45 55 1.629+ 1.630T 980°
(16 hrs.)
65 35 Inhomogeneous 1317° 1324°
(980° nG hrs:
80 20 | Inhomogeneous 1317° 1323°
(980° 16 hrs.)
100 0 Inhomogeneous . 1340° 1347°

(980° 16 hrs.)

* Trace of higher refracting birefringent dots present, and preparations
are therefore not completely homogeneous.
+t Determined on major crystal phase.

side, and incomplete towards the 3CaO.A1,0;.38i0, com-
ponent, at temperatures above 980° (the lower limit at
which experiments were performed) and below the tem-
perature of dissociation.

The intermediate mixtures of the three end mem-
bers containing 40 percent 3CaO.A1,0,.38i10, proved
extremely unsatisfactory. Held for16 hours at 980°, they
appeared as homogeneous fibrous aggregates, whereas
at 1100°, for the same length of time, they are so very
cloudy, with areas of brownish dots, as to make it impos-
sible to judge whether they are simply full of included
alr or, as seems more probable, are mineralogically inho-
mogeneous. At 1170° the preparations are relatively
‘clear again and show distinct interstitial birefringent
material. It is possible but not probable that even in the
preparations at 980° the material is actually inhomogene-
ous but appears to be homogeneous because of the finely
fibrous character. The general rule is that the prepara-
tions rich in the 3CaO.A1,0,.38i0, component, both in
this system and in those containing the 3Na,0.Al,0O..
3810, compound, with slow nucleation and crystallization
(about 1000°), are clear and quite free of air; whereas

Buddington—Natural and Synthetic Melilites. 47

with more rapid nucleation and crystallization at higher
temperatures (about 1100°+50°), much more air is
ineluded and the crystalline material is very cloudy. As
soon as dissociation begins, the air is expelled and the
preparations are clear but minutely heterogeneous with
the dissociated products. .

The optical characters and thermal data for mixtures
of the three components are given in Table V. In fig. 3

TABLE Y.
Optical and thermal data for mixtures of 2Ca0.Al1.0..Si0.,;:2Ca0.Mg0O.
2Si0,:3Ca0.A1.0,.3Si0.. Compositions expressed in weight
percent and temperatures in degrees Centigrade.

3CaQO. 2CaO. 2CaO. One Trace of | Glass
Al,O,. MgO. ALO, erystal disso- -} traceof
3810, 2Si0, S10. ne he phase ciation crystals All glass.

20 30 50 1.652 1.648 1170° 1190° i494° 1501°
(16 hrs.) (16 hrs.)

20 60 20 1.637 1.639 1100° 1140° 1378° 1383°
(16 hrs.) (16 hrs.)

40 20 40 1.647 1.642 1170° +=«1480° ~=—_:1486°
(980° 16 hrs.) (16 hrs.)

60 20 20 *1.641 *1.634 9go°«1408° = «14159
(16 hrs.)

80 10 10 *1.638 *1.633 Inhomogeneous
x (16 hrs. at 980°)
40 40 20 1.640 1.637 1180° 1410° 1414°
‘ s.) (20 hrs.)
15 15 70 1.657 1.652 1180° 1200° 1541° 1547°
(16 hrs.) (16 hrs.)

* Determined on major crystal phase.

the compositions in weight percent of the mixtures
investigated are plotted in triangular coordinates with
isotherms for the temperatures of complete melting. A
point of possible significance bearing on the origin of the
natural melilites is that the mixtures which are closest in
composition to the natural minerals lie in the zone with
lowest temperatures of complete melting. In fig. 4,
limes connecting similar indices of refraction of the ordi-
nary ray and the extraordinary ray have been plotted
against the composition in weight percent for those mix-
tures which show complete solid solution within a definite
temperature range above 980° (the approximate lower
working limit under the conditions of the experiment).
Lines connecting those mixtures which show the same
birefringence have also been plotted. Both 2Ca0O.Al1,0 .

48 Buddington—Natural and Synthetic Melilites.

SiO, and the 3Ca0.A1,0,.3810, compound have negative
optical characters, whereas 2CaO.MgO.28i0, is positive,
so that the combination of the three end members gives a
certain definite series of mixtures which are isotropic to
a definite wave length of light. The lines connecting com-

Fie. 3.
2 C:0-A,0,-S.0,

(\e.
cae 40 80
2C:0°-M:0-2 S.0, WT PERCENT 3 €:0°A1,0,-3 $0,

3.—Triangular concentration diagram showing compositions of prepara-
tions investigated and the isotherms for the temperatures of complete
melting.

positions whose birefringence is 0.000 and 0.005 are more
or less concentric to the 2CaO:MgO.2S8i0, component.

Mixtures or 2Ca0.Mg0.2S8i0,, 2Ca0.A1,0,.Si0,, anp
[(90 83CaO.A1,0;.3810,) (10 3Na,0.A1,0;.3810.) ]

Mixtures of 2CaO.Mg0.2Si0, and the solid solution,
equivalent in composition to the natural mineral sarcolite,
which contain 80 percent or more of the 2CaO.Mg0O.2Si0.,

Buddington—Natural and Synthetic Melihtes. 49

molecule, show a trace of grains with a much higher index
of refraction at temperatures below the dissociation
point. The same phenomenon has already been referred
to in the case of the mixtures of 2CaO.MgO.2Si0, and
3Ca0.Al,0,.3810., and its meaning is not known. With

Fia. 4.
2 €10°A:,0,°S10,
CS

20 40
2C:0-Mc:0°-2S10, WT PERCENT 3 €,:0-A.,0,°3S10,

4.—Triangular concentration diagram showing compositions of prepara-
tions investigated. Lines connecting similar indices of refraction have been
plotted for those mixtures which show complete homogeneity within a
definite temperature range above 980° (the lowest temperature of experi-
ments). Lines for equal birefringence also plotted.

this possible exception, the mixtures of the compounds
2CaO.MgO.2S8i0,, 2CaO.Al1,0,.Si0,, and the solid solu-
tion with the composition of sareolite [(90 3Ca0.Al1,0,.
39810.) (10 3Na,0.A1,0,.3Si0.)] form a complete series
of solid solutions within a temperature range above
980°—the approximate lower working limit under the
conditions of the experiments—and below the solidus or
the dissociation point, as the case may be.

Am. Jour. Sci.—Firta Serirs, Vou. III, No. 13.—January, 1922.
4

50 Buddington—Natural and Synthetic Melilites.

TABLE VI.

Optical and thermal data for mixtures of 2CaO.MgO.2Si0,; and [(90
3Ca0.A1,0;.3810,) (10 3Na,0.A1,0; 3810.) ] Compositions expressed
in weight percent, and temperatures in degrees Centigrade.
90 2Ca0.MgO.28i0, — 10 : 90 3Ca0.A1,0;.38i0,
10 3Na,0.A1,0;38i0,

Temperature Time ! Phases

1180° 16 hrs. Essentially homogeneous very fine grained
fibrous aggregate, with sutured texture.
Trace of minute grains of much higher index
of refraction. ©

Nw = 1.6384 ite == 1-639
1200° ie lines Trace of dissociation in some grains. —
1428° VY hr. Glass plus trace of crystals.
1430° 1/3 hr. All glass.

80 2Ca0.Mg0O.28i0, — 20[ (90 3C0a0.A1,0,.38i0,) (10 3Na,0.A1,0;.3810,) ]

Temperature Time Phases
900° 16 hrs. Essentially homogeneoug fibrous granular aggre-
gate. Trace of minute grains of much higher
index of refraction and low birefringence.

1100° aby lane Same as above.

1150° 1 hr: Same as above. Mw»—1.632 ne —1.637

lilac 2 hrs. Trace of dissociation.

1300° 45 hrs. Conspicuous poikilitic texture. Second phase

occurring as minute inclusions in the major
erystal phase.

1320° 1% hr. Single crystal phase plus trace of glass.

1409° 1/3 hr. Glass plus trace crystals; crystal habit tabular
parallel to base.

1414° 1S. All glass.

60 20a0.Mg0.28i0, — 40[(90 3Ca0.Al,0;.3$i0,) (10 3Na,0.A1,0;.38i0,) ]

Temperature Time Phases
1100° 2 hrs. Homogeneous bladed aggregate, full of minute
air bubbles.
1130° 3 hrs. Minute trace dissociation. Fine grained, ribbed

and bladed aggregate. Anomalous interfer-
ence colors.

Nw = 1.632 Ne = 1.632
1361° ijasvhr. Glass plus trace crystals.
1366° 1/3 hr. All glass.

50 2Ca0.Mg0.28i0, — 50[(90 3CaO.A1,0,.38i0,) (10 3Na,0.A1,0,.3Si0.) ]

Temperature Time Phases

1100° 16 hrs. Homogeneous bladed aggregate full of minute
air bubbles. Nw = 1.631 Ne = 1.630

1130° 3 hrs. Trace of dissociation; interstitial strongly

birefringent material.

Oi 40 hrs. Distinct dissociation; 2d phase interstitial.

1200° 40 hrs. 2d phase in well-formed rods.

1341° 1/3 hr. Glass plus trace erystals; crystal habit thick

tabular to pseudo-cubie with and without
modification by 2d order prism.
1346° 1/3 hr. All glass.

Buddington—Natural and Synthetic Melilites. 51

40 2Ca0.Mg0.2Si0, — 60[(90 3Ca0.A1,0;.38i0,) (10 3Na,0.A1,0;.3S810,) ]

Temperature Time Phases
1075° 40 hrs. Homogeneous fibrous aggregate.
1100° 2 hrs. Essentially homogeneous; trace of intersti-

tial second crystal phase in some grains.
Much air; fibrous aggregate.

Nw = 1.630 Ne = 1.628
TSM 3 hrs. More distinct trace of interstitial higher bire-
fringent material. Clear of air.
1210° 16 hrs. Two crystal phases; approximately half and
half.
1328° 1/3 hr. Glass plus trace crystals.
1331° 1/3 hr. All glass.

20 2CaO0.Mg0.28i0, — 80[ (90 3Ca0.Al,0,.38i0,) (10 3Na,0.A1,0,.38i0.) ]

Temperature Time Phases
980° 16 hrs. Fibrous and bladed aggregate. Homogeneous.
Nw = 1.632 Ne — 1.622.
1100° yn Essentially homogeneous; trace of dissociation
in some aggregates.
1130° 3 hrs. Trace of interstitial second erystal phase.
1320° 1/3 hr. Glass plus trace crystals.
1325° 1/3 hr. All glass.

10 2CaO.Mg0.28i0, — 90[(90 3Ca0.Al,0,.38i0,) (10 3Na,0.A1,0,.38i0,) ]

Temperature Time Phases
1000° 16 hrs. Homogeneous, fibrous, and bladed aggregate.
Nw = 1.633 te— ele Og
1125° 40 hrs. Practically all homogeneous; trace of dissocia-
tion in some grains.
1150° Phr: Trace of dissociation and interstitial second

crystal phase.

[ (90 3Ca0.AI,0,.39i0,) (10 3Na,0.A1,0;.3Si0,) ]

Temperature Time Phases
700° 144 hrs. Still all glass. No trace of erystallization.
980° 16 hrs, Very fine grained fibrous aggregate. Trace of

wollastonite fibers believed to be due to im-
purity of melt or to exceptionally strong crys-
tallizing power of wollastonite. Otherwise
homogeneous.

1000° 16 hrs, Very well crystallized material. Ribbed and
bladed aggregates and coarse laths. Trace of
feather-like crystals, probably wollastonite.

Ne = 1:63] Ne — 1.615
1100° 2 hrs. Trace of dissociation in some grains, mostly
homogeneous fine-grained fibrous aggregate.
1125° 1 hr. Very fine grained fibrous granular aggregate

with interstitial strong birefringent material
forming second crystal phase.

1140° 2 hrs. Trace of dissociation very distinct.

1165° 168 hrs. Coarse blades and laths, with interstitial disso-
ciated material and a trace of glass.
Nw — 1.640 Ne = 1.616

52 Buddington—Natural and Synthetic Melshites.

Temperature Time Phases
1260° 24 hrs. EKuhedral tetragonal crystal plus glass. Crystal
habit is thick tabular, pseudo-cubic, and short
prismatic; combination of 1st order prism
and base alone, or lst and 2d order prism and

base.
: Nw = 1.651 ne = 1.636
1320° 72 hrs. Glass plus euhedral crystals, similar to pre-
ceding.
Nw — 1.656 ne = 1.645
1327° 1 hr. Glass plus trace of erystals.
1334° 1 hr. All glass.

TABLE VII.

Optical and thermal data for mixtures of 2Ca0O.Al,0;.Si0O, and [(90
3Ca0.A1,0,.38i0,) (10 3Na.0.A1,0;.38i0,) | Compositions expressed
in weight percent, and temperatures in degrees Centigrade.

90 2Ca0.Al,0,.8i0,—10[(90 3Ca0.A1,0,.38i0,)(10 3Na,0.A1,0;.38i0,) ]

Temperature Time ‘Phases

1200° 18 hrs. Fibrous granular aggregate with sutured tex-
ture. Clear and homogeneous single erystal
phase.

1415° 1 hr. Coarse crystalline aggregate with faintest trace
of glass. Nw = 1.664 Ne = 1.654

1490° 1 hr. Essentially all crystal. Barest trace of glass.

No evidence of dissociation.
1500° hrs Crystals plus distinct trace of glass.

80 2Ca0.Al,0,.8i0, — 20[(90 3Ca0.A1,0,.38i0,) (10 3Na,0.A1,0,.38i0,) ]

Temperature Time Phases

1100° 2 hrs. Very good, clear crystalline material with large
coarse laths and bladed spherulites.
Nw = 1.660 ne = 1.650

1220° 1 hr. Little dissociation shown by interstitial strong
birefringent material. Fine-grained aggre-
gate.

1270° 1 hr. Crystals plus included minute drops of glass.

1435° 1 hr. Coarse grained crystalline aggregate plus a little
interstitial glass. nw — 1.661 Ne = 1.653

1562° 1% hr. Glass plus trace of crystals.

1568° 1 hr. All glass.

60 2Ca0.Al,0,.8i0, — 40[(90 3Ca0.Al,0,.38i0,) (10 3Na,0.A1,0;.3Si0,) ]

Temperature Time Phases

1200° ee loa Spherulitic fibers and blades, and fibrous and
ribbed structured grains. Homogeneous single
crystal phase.
Nw = 1.653 Ne = 1.642
1225° 1 hr. Most grains with a trace of glass, some grains
apparently homogeneous.

Buddington—Natural and Synthetic Melilites. 53

Temperature Time Phases
1250° Lhe, Crystals plus trace of glass, the glass occurring
as minute droplets in the crystals.
1524° Y% hr. Glass plus trace of erystals. Crystal habit is

medium tabular parallel to base; combination
of 1st order prism and base, usually modified
by the 2d order prism. 2d order pyramid
rare.

1533° Y% hr. All glass.

40 2Ca0.A1,0;.Si0, — 60[(90 3Ca0.A1,0,.3Si0,) (10 3Na,0.A1,0;.38i0.) ]

Temperature Time Phases
1085° 2 hrs. Perfectly homogeneous radiating fibrous aggre-
gate with some plates and ribbed structure.
1110° i ive Perfectly homogeneous.
Nw = 1.645 Ne = 1.633
1125° ihr: Beginning of trace of dissociation.
1150° fd hr Trace of dissociation.
1415° 16 hrs. Glass plus few crystals. Crystal habit predomi-

nantly tabular parallel to base; combination
of base and Ist order prism modified by 2d
order prism. 2d-order pyramid and ditetra-
; gonal noted on a few ecrystais. A few equidi-
- mensional crystals present.
1485° 1/3 hr. Glass plus trace crystals.

1490° 1/3 hr. All glass.
20 2Ca0.A1,0;.Si0, — 80[(90 3Ca0O.A1,0;.38i0,) (10 3Na,0.A1,0;.3Si0,) ]
Temperature Time ~ Phases

1000° 16 hrs. Homogeneous ribbed and bladed aggregate.

Much coarser grain than 16 hrs. at 1175°.
Ne — 1.637 ne — 1.625
1150° 16 hrs. Minute trace of dissociation. Some aggregates

seem homogeneous.
1418° 1 hr. Glass plus trace crystals.
1422° 1% hr. All glass.

TABLE VIII.

Optical and thermal data for mixtures of 2CaO.Mg0O.2Si0,; 2Ca0.A1,0;.Si0,;
and [(90 3Ca0.A1,0,.3Si0.) (10 3Na.0.A1,0,38i0,) ] Composition
in weight percent and. temperatures in degrees Centigrade.

Prise f 77 ar § 90 3Ca0.A10,.38i0
70 2Ca0.Al1,0;.8i0, —15 2CaO.Mg0O.2Si0, 10 3Na.0. A1,0,.38i0,

Temperature Time Phases
1200° iL. hr; Homogeeous single crystal phase.
Nw — 1.658 Ne = 1.652
1205° 16 hrs. Trace of beginning of dissociation. Poikilitic
texture.
1537° 1/3 hr. Glass plus trace of erystals.

1542° 1/3 hr. All glass.

54. Buddington—Natural and Synthetic Melilites.

50 2Ca0.Al,0,.810, — 30 2Ca0.Mg0.28i0, — 20 10 oN OAC
-fAl,U3. 2

Temperature Time Phases

1100° 16 hrs. Homogeneous fibrous granular and radiating
_ sperulitic aggregate. nw= 1.650 ne= 1.645

12003) = Lonr. Homogeneous fine fibrous aggregate.

1210° don: Coarse laths and trace of glass.

1280° 16 hrs. Crystals plus little glass.

1478 V/o alin: Glass plus trace crystals.

1484° 1/3 hr. All glass.

50 2Ca0.A1,0;.Si0,

90 3Ca0.A1,0,.38i0
DU CBI ORE OSE Sac oil : 10 3Na,0.A1,0,.39i0;

Temperature Time Phases
1150° 1 hr. Very fine fibrous homogeneous aggregate.
1175° 1 hr. Trace of beginning of dissociation.
No — 1.650 Ne = 1.644
1290° 40 hrs. Crystals plus little glass.

Nw — 1.655 Ne — 1.649

40 2Ca0.A1,0,.810, — 40 2Ca0.Mg0.28i0, — 20 : aco a eae
; 5 Q2\*e 273° 2

Temperature Time Phases
1175° Phrs Homogeneous single crystal phase.
Na —=1.648 ne— 1.645
1185° 1 chr. Trace of beginning of dissociation.
1446° 1/3 hr. Glass plus trace of crystals.
1450° 1/3 hr. All glass.

40 2Ca0.A1,0,.8i0, — 20 20a0.Mg0.28i0.— 40 } 59 eee
2VU-fAloUs. 2

Temperature Time Phases
IDOLOS 1 hr. Homogeneous single crystal phase.
Nw — 1.645 Ne — 1.640
1140° are Distinct trace of dissociation.

30 2Ca0.Al,0,.8i0, — 40 20a0.Mg0.28i0,— 80 } 19 a ee
2. tA13. 2

Temperature Time Phases
900° 18 hrs. Homogeneous single crystal phase.

1175° sb oishe- Homogeneous aggregate of coarse plates and
laths.
No — 1.644 ne = 1.641

1220° 1 lei Trace of dissociation.

1250° ich: Little dissociation. Minor phase as strong bire-

fringent rods and interstitial.
1419° 1/3 hr. Glass plus trace of crystals.
1423° Ib/Be lines All glass.

20 2Ca0.A1,0,.8i0, — 60 20a0.Mg0.28i0, — 20 } a SG ee
ge 23e 2

Temperature Time Phases

1200° 2 hrs. Homogeneous. w= 1.641 ne= 1.641
1205° 40 hrs. Trace of dissociation.

Buddington—Natural and Synthetic Melilites. 55
Temperature Time Phases
1360° 16 hrs. Glass plus few crystals. Crystal habit, thin to

thick tabular parallel base, combination of 1st
order prism and base modified by 2d order

prism.
1375° Y hr. Glass plus trace of crystals.
1378° Y% hr. All glass.

20 2Ca0.A1,0;.8i0,

40 2CaO.Mg0.28i0, — 40 10 ae

Temperature Time Pnases
900° 16 hrs. Very fine grained fibrous granular aggregate.

1130° iB is Homogeneous single erystal phase.
Nw = 1.640 ne— 1.638

1150° ihr: Trace of dissociation. 2d crystal phase occurs
as interstitial material.

1388° Y% hr. Glass plus trace of crystals.

1395° 1% hr. All glass.

59 {90 3Ca0.A1,0,.38i0,
99410 3Na,0.A1,0;.38i0,

20 2Ca0.Al1,0;.Si0,

Temperature Time Phases
1100° 1 hr. Homogeneous single crystal phase.
Ne = 1.639 Ne = 1.635
1150° Ay, hr. Bare trace of dissociation.
1390° 1/3 hr. Glass plus trace of crystals.
1400° 1/3 hr. All glass.

Sse wa, §90 3Ca0:Al,0,.38i0,
20 2Ca0.A1,0,.Si0, — 20 2Ca0.Mg0.28i0.— 60) 19 31Va,0.A1,0,.38i0,

Temperature Time Phases

1125° a fe Fig Homogeneous crystal aggregate with ribbed
; structure and coarse laths.

Nw — 1.639 Me = 1-632

1150° Thr: Trace of beginning of dissociation.

1360° 16 hrs. Glass plus few crystals. Crystal habit; tabular
to pseudo-cubiec; combination of base and Ist
order prism usually modified by 2d order
prism. 2d order pyramid present on an occa-
sional crystal.

1389° 1% hr. Glass plus trace of erystals.

1395° 1 hr. All glass.

(90 3Ca0.Al,0,.3Si0,

pea BL Oe SiO, 110 3Na,0.A1,0,.38i0,

80 2Ca0.Mg0O.2S8i0, — 10

Temperature Time Phases

1160° 16 hrs. Large clear crystal plates obtained by first melt-
ing material and undercooling rapidly.
Nw — 1.636 ne — 1.641

1200° dae ie Homogeneous fibrous granular aggregate.
1225° Ahr; Trace of dissociation; poikilitic texture.
1397° 1 hr. Glass plus trace of crystals.

1405° 1% hr. All glass.

56 Buddington—Natural and Synthetic Melihtes.

0 .A1,0,.38i0
10 2Ca0.A1,0,.8i0, — 50 2Ca0.Mg0.28i0, — 40 ‘40 aL Ou oe

Temperature Time Phases
ELLOS “hr, Homogeneous erystal aggregate.
1125° i hr: Trace of beginning of dissociation.
Nw = 1.634 Ne — 1.635
1358° 1/3 hr. Glass plus trace of crystal.
1363° 1/3 hr. All glass.
nee ee span en) 90 3CaO.Al0, 3s1@.
10 2Ca0.A1,0;.810, — 10 2Ca0.Mg0.2Si0, O07 10 3Na,0.A1,0,.38i0,
Temperature Time Phases
980° 10 hrs. Homogeneous medium grained fibrous aggregate
and lath shaped crystals in a finer ground
mass.
Nw — 1.634 ne — 1.623
1050° 4 hrs. Homogeneous fine grained fibrous granular ag-
gregate.
1100° 40 hrs. Homogeneous crystal phase.
1150° iPe lanes A little dissociation.

Tables II, VI, VII, VIII give the optical characters
and thermal data for the preparations examined. Fig-
ures 1, 5, and 6 show the indices of refraction plotted
against composition in weight percent. In fig. 7 the com-
positions of the mixtures investigated are plotted in

Fig. 5:

Index of fiffraction

O 20 ALO GO SO /OO
2CaAO-MGO:-2SiQ WT PERCENT 90[9Ca0-AlzO3-9St Op]
/O[BNGz O-Ale O3-FS¢ Qn]

5.—Curves for refractive indices plotted against composition in weight
percent for mixtures crystallized above 1000° and .below the point of

dissociation.

_triangular coordinates together with the isotherms for the
temperatures of complete melting. As in the case of
fig. 4, a significant point is that those compositions which
approximate most closely to the natural minerals he in
the zone of mixtures with the lowest temperatures of com-
plete melting. In fig. 8 the indices of refraction are

=I

Or

Buddington—Natural and Synthetic Melilites.

Index of Kefract fon

O 20 +O OO FO ae IOC
2C2O-AbOsSiQ — WT-PERCENT 902CAO-AbOs35¢ 02]
lOf3 Naz O-Al> Q3-F5é O2]

6.—Curves for refractive indices plotted against composition in weight
percent for mixtures crystallized above 1000° and below the point of
dissociation or the beginning of melting.

> eae 5 90 (3 €.0-AL,2,°3$10,]
r ee ee ft Anh 2
20:0°:0°2 $19, PECTRRSEM 10 [3 N.,0°A1,0;°3 $10)

7.—Triangular concentration diagram showing compositions of prepara-
tions investigated and the isotherms for the temperatures of complete

melting.

58 Buddington—Natural and Synthetic Melilites. —

plotted against the compositions, and curves are drawn
connecting mixtures which show equal birefringence.
The results are similar to those obtained in the case of
mixtures of 2CaO.Mg0O.2810,, 2CaQ.Al,0,.S10,, and
3CaO.A1,03.38105.

Fig. 8.
2 C.0°A1,0,°S10,

20 8
2 Cr0>Ms0-2510, of wean ” 90 [36:0-A1,0,°3 $10,]
10 [3 N.,0-A1,0,°3 $10,

8.—Triangular concentration diagram showing the compositions of

preparations investigated. Lines connecting similar indices of refraction

have been plotted for those mixtures which show complete homogeneity

_ within a definite temperature range above 980° (the lowest temperature of
experiments). Lines of equal birefringence also plotted.

Mixtures oF 10 Percent 3Na,0.A1,0,.38i0, WITH VARY-
ING Amounts OF 8CaQO.Fe,0,.3810,, 3CaO.A1,0,.3810,, AND
2CaO.Mg0.28i10,.

Mixtures made up by maintaining the compound
3Na,O0.Al,0,.3810, at a constant ratio of 10 percent and
varying the percentages of 3CaO.Fe,0;.38i0., 3Ca0.
Al,O;.38810,, and 2CaO.MgO.2S8i0O, gave a very limited

Buddington—Natural and Synthetic Melilites. 59

field of solid solution lying along the line between 2CaO.
Mg0.2Si0, and 3CaO.Al,0;.3810, and restricted to the
3CaO.Al,0,.3810, end. However, if the compositions are
plotted upon a triangular diagram, there is a zone extend-
ing from the component 3CaO.Al,0,.3Si0, to about the
center of the line connecting the components 2CaO.Mg0O.
28i0, and 3CaO.Fe,03.38i10, in which the preparations
lack but a trifle of being completely homogeneous. Tables
[X-XIT give the optical and thermal data for the mixtures
investigated.

Index of Fiefraction

9012 CaO-MgO-2.5i OQ] W7-PERCENT 9OL38C2O -Fez Oy FSi OZ]

JS

/0[3 Naz O-Al>Q;°3Séi Oz] (OL37Na2Al,033Si Oz |]

9.—Curves for refractive indices plotted against composition in weight
percent for mixtures crystallized just below the solidus which show complete
homogeneity. Indices of refraction for dominant phase of inhomogeneous
mixtures are indicated when determined.

Fig. 9 shows the refractive indices plotted against the
composition in weight percent for those compositions
that lie along the line connecting the components 2Ca0O.
Mg0.2Si0, and 3CaO.Fe,0,.3S8i0., the amount of 3Na.0.
Al,0,;.3Si0, being kept constant at 10 percent. The
curves connect only those compositions which were found
to form completely homogeneous mixtures within a tem-
perature range lying above 1000°, the lowest temperature
at which observations of these mixtures were made. In
the case of those mixtures which showed inhomogeneity,
the indices of refraction for the dominant phase are

60

3Na,0. 2CaO. 3CaO.
MgO. Fe.0;.
2810, 3810,

A1,03.

3810,
10
10
10
10
10
10
10
10

10

90
82
70
67.5
54
45
36
18
00

00
8
75

TABLE IX.
Optical and thermal data for mixtures of 10 percent 3Na,0.A1,0,.3Si0, with

Compositions in weight percent.

Nw Ne

*1.632 *1.635
*1.639 *1.639

22.5 1.654 1.644

36
45
54
72

305

1.675 1.653
1.688 1.658
*1.699 *1.662

One Crystals

-erystal -t trace

phase of glass

Inhomogeneous
(40 hrs. at 1000°)
Inhomogeneous
(40 hrs. at 1000°)
Inhomogeeous
(40 hrs. at 1000°)
1164° LEO

(% hr.) (% hr.)
1145° 1160°
(40 hrs. ) (% hr.)
1145° 1162°
(15 hrs.) (1% hr.)
Inhomogeneous
(40 hrs. at 1000°)
Inhomogeneous
(2 hrs. at 1070°)
Inhomogeneous
(2 hrs. at 1070°)

* Determined on dominant crystal phase.

3Na.0. 2CaO. 3CaO.
.MgoO. A1,0;.
8810, 2S8i0,. 3S8i0..

A1,0;.
10
10
10
10
10

10

90
63
50
45
27

9

00
27
40
45
63

81

TABLE X,
Optical and thermal data for mixtures of 10 percent 3Na,0.A1,0;.3Si0, with

Compositions in weight percent.

Nw ne
*1.632 *1.635

*1.632 *1.630
1.630 1.628
1.630 1.628
1.631 1.625

1.633 1.621

Glass --
trace of
erystals

1391°

1352°
1325°
1305°
1284°
1235°
1235°

One Traceof Glass +
erystal disso- trace
phase ciation. crystals
Inhomogeneous 13912
Inhomogeneous 1360°
130° 1150°

(ishr:) Gir)
1140° 1160° 1334°
Gbbr.)(Cl hrs)
1165° 1185° 1317°
(1 hr.) (1 hr.)
1125° 1327°
(16 hrs.)

* Determined on dominant crystal phase.

Buddington—Natural and Synthetic Melilites.

varying percentages of 2CaO.Mg0.28i0, and 3Ca0.Fe,0,. ABO:

All glass
1358°

1357°
1330°
1309°

—-1290° |
1240°
1244°

varying percentages of 2CaO.MgO.2Si0, and 3CaO.A1,0,.3Si0,.

All glass.
1398°

1367°
1340°
1324°
1334°

Buddington—Natural and Synthetic Melilites. 61

TABLE XI.

Optical and thermal data for mixtures of 10 percent 3Na,0.A1,0,.3Si0, with
varying percentages of 3Ca0O.A1,0;.3Si0, and 3CaO.Fe,0;.3Si0,.
Compositions in weight percent.

3Na.0. 3CaO. 3Ca0O. Glass --
Al1,0;. Al,0;. Fe,0;. traceof All
38i0, 3S8i0, 3S8i0, nw Ne crystals glass
10 905 2005-2 63h 1.65 1327° 1334°
Inhomogeneous
10 81 9 *1.642 *1.632 (40 hrs. at 1000°)
10 67.5 22.5 *1.655 *1.637 1270° 1275° Inhomogeneous
(16 hrs. at 1125°)
3 Inhomogeneous
10 54 936 (16 hrs. at 1160°)

* Determined on dominant crystal phase.

TABLE XII.

Optical and thermal data for mixtures of 10 percent 3Na,0.A1,0;.3Si0, with
varying percentages of 2Ca0.Mg0.2Si0.;3Ca0.A1,0;.38i0.; and
3Ca0.Fe,0;.38i0,. Compositions expressed
in weight percent.

* Essential-
3Na,0. 2CaO. 3CaO 3Ca0. ly one Traceof Glass +-
Al1,0;. MgO. FeO, A1,0;. erystal disso- Traceof All
38810, 2810, 38i0,3S8i0, nw Ne phase. ciation. crystals. glass.
10 54 27 9 1.663 1.648 1140° 1170° a
(20 hrs.)

tie toe. ot. 18 - 1.663 1.648 1160° 1170°
(1 hr.) (16 hrs.)
PU) 1S. AS et Gol 1.641 1150° 1170° 13292 1335°
: (1 hr.) (16 hrs.)
10 40 18 32 1.650 1.640 1150° LEO?
(16 hrs.) (1 hr.)
#0 38 14 - 38 1.645. 1.638 1140° TAO?
(20 hrs.) (16 hrs.)
10° 400-110 40° 1.639 1.635 1140° 1165° todo 1316°
(16 hrs.) (1 hr.)
10-27 9 54 1640 1.632 1140° 1160° 1308° 13815°
@ohr>): (1. hr;)
i}: 18 9 638 1.642 1.633 -+1000° 1125°
(40 hrs.) (16 hrs.)
HO; - 47 33 10 41.671 41.653 Inhomogeneous TS1G6ce 1320?
(16 hrs. at 1110°)
10 63 13.5 13.5 21.643 21.640 Inhomogeneous
(16 hrs. at 1000°)

£02. 36) 367: <8 Inhomogeneous
(16 hrs. at 1170°)
LO 30°" 30 30 Inhomogeneous 1285° 1291°

(16 hrs. at 1150°)
10 34 22 34 41.655 41.645 Inhomogeneous
(4 hrs. at 1040°)

* These preparations are not completely homogeneous as they contain a
trace of minute grains of much higher indices of refraction. It is believed,
however, that the inhomogeneity is so slight that the crystals are of essen-
tially the same composition as the melt, except in the case of those mixtures
_ marked inhomogeneous.

t Completely homogeneous. a Determined on dominant crystal phase.

62 Buddington—Natural and Synthetic Melhilites.

Taste XIII.

Optical characters and thermal data of mixtures contaiming 10 percent
3Na,0.A1,0;.38i0, and 20 percent 2Ca0.A1,0,.8i0, with varying
percentages of 2CaO.Mg0O.28i0, and 3CaO.Fe,0,.3S8i0,.
Compositions expressed in weight percent.

3Na,0. 2Ca0. 2Ca0. 3Ca0. | One Crystals Glass +
Al,O;. Al,O;. MgO. FeO. ‘erystal -_ Trace Trace All
8810, SiO, 2810, 3Si0, nw ~ Me phase. of glass. crystals. glass.
Oi 220 70 00 Inhomogeneous
(2 hrs at 1050°)

10 3920 56 14 Inhomogeneous

(1 hbr. at 1176*)
10 3820 46 24 Inhomogeneous

(1 hr. at 1151°)
1020 35-3 35. IOs 1006 Trace of 1298° 1305°

! (16 hrs. at 1150°) Inhomogeneity
10: 20-28 | 42-1689 =) 669 1177° A9iS 1278° . 1284°
(2 hrs.) (2 hrs.)
10 20 17.5 52.5 1.694 *1.673 Inhomogeneous
(40 hrs. at 1160°)

10 =. 20 10.5 59.5 *1.711 *1.684 Inhomogeneous ~
- (16 hrs. at 1125°)

* Determined on dominant erystal phase.

TABLE XIV.

Optical characters and thermal data of mixtures containing 10 percent
3Na,0.A1,0,;.38i0, and 20 percent 2Ca0O-Al,0;.Si0, with varying
percentages of 2CaO.MgO.2Si0, and 3Ca0.A1,0,.3S8i0,.

3Na.O. 2CaO. 2CaO. 3CaO. One Traceof Glass +
Al,O;. Al,0;. MgO. A1,O;. erystal disso- Traceof All
388i0, SiO, 28i0, 3810, nw Ne phase. ciation. crystals glass.
10 820 70 00 Inhomogeneous <4 Or ne
(2 hrs. at 1050°)
10 820 35 385 Inhomogeneous
(2 hrs. at 1050°)
10 20 28 42 1.638 1.633 1200° 1220° 1380° 1386°
_ (Lhr.) (%Y% hr.)
10 20 17.5 52.5 1.641 1.634 1130° 1145°
@iG hres) Gleb.)
10 20 00.0 70 1.6388 1.624 1140° 1165°
(2 hrs.) (1 hr.)

Buddington—Natural and Synthetic Melilites.

63
TABLE XV.

Optical characters and thermal data of mixtures containing 10 percent
3Na.0.A1,0;.38i0, and 20 percent 2CaO.A1,0;.8i0, with varying
percentages of 3Ca0.Al.0,.3Si0, and 3Ca0.Fe,0;3S810..
Compositions expressed in weight percent.

3Na.O. 2CaO. 3CaO..3CaO. One Traceof Glass +

ALO. AGO: AlLO,. Fe.0;. erystal disso- Trace All

38i0, SiO, 3810, 38i0O, nw Ne phase. ciation. crystals glass.

10. +20 7 > 00S £638 = 12624 1140° 1160° 1412° 1417°
(2 hrs.) (16 hrs.)

ie 20. OG 14 ~ 1.655"* 1.640 1142° 1160° 1380° 13886°
(20 hrs.) (40 hrs.)

105 20.49" 21° (1.657 (1.645 1145° =: 1160° 1360° 1366°
(2 hrs.) (1 hr.)

10 820 42 28 Inhomogeneous 1344° 1350°

(16 hrs. at 1125°)

TABLE XVI.

Optical characters and thermal data of mixtures containing 10 percent 3Na,0.A1,0;.3Si0,

and 20 percent 2CaO.Al1,0;.8i0, with varying percentages of 2Ca0.Mg0.2Si0,;

3Ca0.A1,0;.38i0,; and 3CaO.Fe,0,.38Si0, Compositions in
weight percent.

. Crystals Glass
3Na,0. 2CaO. 2CaO. 3CaO. 3CaO. One ‘Trace plus plus
Al,O;. Al,0;. MgO. Fe,O;. Al.Os. crystal disso- trace trace
3810, SiO, 2Si0,3Si0O, 3S8i0, nw Ne phase. ciation. glass. crystals.

10 20 10 10 50 — 1.648 1.638 155° ne GeyP 1376°
(1 hr.) (1 hr.)
10 20 30 10 30 81.648 1.642 1225° Sats LOAQ Ce, PAO SS
(ihr!) (1 hr.)
10 20 21 21 28 1.660 1.649 1200° 1220° 1328°
(1 hr.) (4% hr.)
10 20 28 28 ioe LOE 1.657 12208 ae a 12352° --AS102
(1 hr.) (1 hr.)
10 20 7 28 aon Pee Gk, G56 Bare trace inhomog. 1334°

(1 hr. at 1147°)

10 20 21 35 it 71-680 223-663 Bare trace inhomogeneity

(16 hrs. at 1150°)
10 20 35 Pht AAI FL GGa' SY.653 Trace of inhomogeneity

*Determined on major crystal phase, which lacks but little of complete homogeneity.

plotted. Fig. 10 similarly shows the indices of refraction
plotted against composition in weight percent for the
solid solutions consisting of 10 percent 3Na,0.AI1,0,.
38si0, and variable amounts of 2CaO.Mg0O.2Si0, and
8CaQ.Al,0,.388i0,, within a temperature range above
1000°. Mixtures of 3Ca0.Al1,0,.3Si0, and 3CaO.Fe,.O3.
2810, with as little as 10 percent of the latter constituent,
showed some inhomogeneity, as they did also at all higher
percentages investigated.

64 Buddington—Natural and Synthetic Melihtes.

Fig. 10.

Index of Kefraction

90CaO-MgO-2S10,] WTPERCENT 90[3Ca0-Al Qs 3Sé Oz]
10{3Naz0A |p Q3°-3Sié Op | (Of3 NazO-Al> 03°3Sé Op }

10.—Curves for refractive indices plotted against composition in weight
percent for mixtures crystallized between 980° and the dissociation point.

Mixtures oF 10 Percent 3Na,0.A1,0,.3810, AND 20 PERCENT
2Ca0.Al,0,.8i10, WITH VARYING PERCENTAGES OF 3Ca0.Fe,0,.
88i0,, 3Ca0.A1,0,.388i0,, anp 2CaO.Mg0O.28i10,.

Mixtures of 10 percent 3Na,0.A1,0;.38810, with vary-
ing percentages of 3CaO.A1,0,.3810,, 3CaO.F'e,03.38i10.
and 2Ca0.Mg¢0.2Si0, were found to form inhomogeneous
- mixtures under the conditions of experiment, except for
a very small range of compositions near the 3Ca0.AI,Qs3.
3810,-2CaO.Mg0.2Si0, line at the 83CaO0.A1,0,.3810, end.
Natural melilites have a wider range of composition, and —
some have a considerable percentage of the gehlenite
molecule. A series of mixtures was therefore prepared,
each of which contained 10 percent 3Na,0.A1,0,.3S810,
and 20 percent 2CaO.Al,0,.Si0,, but with varying
amounts of 3CaQ0.Fe,03.3810,, 2CaO.MgO.2Si0O, and
3CaQ0.Al,0,.3810,. The results (fig. 11) show that the
presence of 20 percent 2CaQ.Al,0;.S10, has greatly
extended the area of solid solution (for a temperature
range above 980°), and includes mixtures which. are simi-
lar in composition to the natural melilites. ‘Tables
XIIT-XVI.

The compositions of the mixtures experimented with
are relatively far apart, and the plotted limits of solid
solution are therefore only approximate.

COMPARISON OF MINERALS WITH SYNTHETIC PREPARATIONS.

Introduction.—The data obtained from a study of some
synthetic mixtures of compounds, believed to be the more
common components of the natural melilites, have been
set forth in the preceding pages. It now remains to

Buddington—Natural and Synthetic Melilites. 65

attempt to correlate these data with those derived from
studies of the minerals. The optical data suggest them-
selves as being the best for this purpose, but unfortu-
nately, with respect to this group of minerals, such data
are meager and in only a few cases have both the optical

Fie. 11.
70 3 CsO°F:,0,°3 Si0,
20 2C:0°A.,0,°S:0,
10 3 .N.,0°A1,0,-3 S10,

@ COMPLETE SOLID SoLUTION
O INCOMPLETE SOLID SOLUTION

70 2C:0-M:0-2 S10,” ae 70 3 C,0-Ai,0,°3 S10,
20 2C.:0°A:,0,° S:0, WI PERCENT 20 2 C:0-A.,0,- $0,
10 3 .Ns,0°A.,0,°3 $10, 10 3 Ns,0°A.,0,°3 $10,

11.—Triangular concentration diagram showing compositions of prepara-
tions investigated. Lines connecting similar indices of refraction have been
plotted for those mixtures which show complete homogeneity just below
the beginning of melting, or the point of dissociation, and above 1100° (the
lowest temperature at which experiments were made).

data and the chemical analysis been given for the same
specimen. This is strikingly true of the best known
melilites, those of Italy.

Partially to remedy this defect, specimens of Italian
ferric iron-rich melilites, and of gehlenites, humboldt-
ilites, sarcolite, and fuggerite were obtained for purposes
of study from the U. S. National Museum and from

Am. Jour. Sct.—FirtuH Series, Vou. III, No. 13.—January, 1922.
5 :

66 Buddington—Natural and Synthetic Melilites.

various dealers in the United States. Dr. H. S. Wash-
ington of this laboratory also obtained for the writer sev-_
eral specimens of the Capo di Bove melilites from Pro-
fessor F. Millosevich of the University of Rome. These
specimens proved of exceptional interest and afforded a
melilite of composition different from any hitherto
reported from this locality. The writer is greatly
indebted to Professor Millosevich for his courtesy and
generosity in sending these specimens.

Four samples of the minerals of this group were
obtained in sufficient purity for analytical purposes, and
Dr. H. S. Washington very generously made the chemical
analyses. The writer wishes to express his appreciation
of this service.

Separation of material—The samples of humboldtilite
studied were obtained by crushing the rock containing
them as an accessory constituent to a sufficient fineness to
pass a 35 mesh sieve. The lighter components, mainly
nephelite and leucite, were separated with bromoform,
and the pyroxene by means of an electro-magnet. After
this treatment the samples were examined under the
microscope and found to be sufficiently pure.

The densities given for the minerals represent the
density at 25° of the solutions in which the heavier por-
tion of the purified materials sank.

' AKERMANITE.

Akermanite was first named and described by Vogt
from its occurrence in artificial slags long before its dis-
covery innature. The mineral akermanite was identified
by Zambonini"' at Vesuvius and its properties are given
by him as follows: tetragonal, cleavages 001 and 110;
density 3.12; uniaxial, positive; for sodium light,
NM == 1.6332, ne—n. =0.006. The compound 2CaO.MgO.
2810, was obtained by Ferguson and Merwin” and cor-
related by them with the mineral akermanite as a result
of a study of its optical characters, which are given by
them as tetragonal, uniaxial positive, for sodium hght
mo == 1.031 andve = 1.638. The density was determined

* EF. Zambonini, Mineralogia Vesuviana, p. 255, 1910.
* Op. cit., pp. 118 and- 122:

Buddington—Natural and Synthetic Melilites. 67

by Ferguson and Buddington' as 2.944. The explana-
tion for the much lower density of the artificial akerman-
ite is not apparent, but the natural mineral is not pure
and the synthetic material may be a trifle hght owing to
the presence of included air. Material crystallized from
- akermanite glass at 700° has the same indices of refrac-
tion as that crystallized at 1450° so that there is no
evidence of polymorphism between these temperatures.
The compound 2CaO.MgO.28i0, enters into complete
solid solution with 2CaO.A1,O0,.Si0, and into limited solid
solution with 3RO.R,O,.8810, compounds. In this prop-
erty it is analogous to the mineral akermanite. The two
analyses of natural akermanites, however, show silica in
excess of that called for by the formula for the synthetic
compound 2CaO.MgO.2Si0,. This discrepancy remains
to be explained but the balance of evidence favors the
essential identity of the synthetic preparation and the
mineral.

TABLE XVII.

Akermanite from Vesuvius.
Caleulation from Freda’s analysis.

2CaO.Mg0.2Si0,
Original Analysis an
analysis reduced Mol. 3Ca0O.Al,0;.3Si0, Difference
molecules
Sere. 5S 46.70 46.47 7764 6902 —862
Al.O 1.09 .96 94 94
(CANO ee 39.62 39.23 7054 6902 —152
MSOs 13.38 13.34 3310 3310
100.79 100.00

Table XVII gives the analysis by Freda of akermanite
from Vesuvius, stated in terms of its constituent mole-
cules and compared with the total number of molecules
which can be recomputed to form 2CaO.Mg0O.2Si0, (90.4
percent) and 3Ca0.Al1,0,.3Si0, (4.2 percent).

Table XVIII gives the analysis by Zambonini of aker-
manite from Vesuvius, stated in terms of its constituent
molecules and compared with the total number of mole-
cules which can be recomputed to form 2CaO.Mg0O.2Si0,
(90 percent) and 3CaO.A1,0,.3S8i0, (4.8 percent).

= Op: cit., p.. 132.

68 Buddington—Natural and Synthetic Melilites.

TABLE XVIII.

Akermanite from Vesuvius.
Caleulation from Zambonini’s analysis.

2Ca0.Mg0O.2Si0,
Original Analysis and
analysis reduced Mol. 3Ca0O.A1,0;.3Si0, Difference
molecules

SiO; 22 46205 46.33 7683 6908 5
INT Os eT ie .96 1.08 106 106
Ca@ivce 212 39.30 39.31 7020 6908 —112
MigOr. car 13.30 13.28 3295 3295
HeEO! we oe ay

100.23 100.00

GEHLENITE.

Gehlenite of Velardena—The gehlenite described by
Wright't from Velardena, Mexico, approximates more
closely in composition the compound 2CaQ.A1,0,.Si0,
than any of the other gehlenites described. According
to Wright it is associated with colorless pyroxene, yellow
garnet, magnetite, rutile, and spurrite, and is formed at
the contact of a basic intrusive diorite and limestone.

TABLE XIX.
Gehlenite from Velardena, Mexico.

Original Analysis Calculated

analysis reduced Mol. molecules Difference
Sil Osea ced PASS) 26.76 4438 4438
INOS 2 ones 28.27 BT ETAL 2779 +8
CAOw nme. 39.55 40.19 Ces 7158 —19
Iii eal © aac ogee 2.44 2.48 615 623 sa:
OS cep. fe 03 03 4 — 4
HejsOek .& os el 1.45 81 80 —i1
MEOer yt... 50 750) 70 72 a,
MERION ce a 01 01 if
NBO i. ei SP: yl 34 La ;
NEON es 10 10 10
BAS OMe 1.85

100.27 100.00

In Table XIX is given the analysis by Allen of this
gehlenite from Velardena, Mexico, reduced, with elimina-

“FE, K. Wright, this Journal, 26, 545-7, 1908.

Buddington—Natural and Synthetic Melilites. 69

tion of water, stated in terms of its constituent molecules,
and compared with the calculated molecules in a mixture
composed of 76 percent 2CaQO.Al,O0,.8i0,, 17 percent
2CaO.MgO.28i0., 4 percent 3CaO.Fe,0,.3810,, 2.3 per-
eent 2CaO.FeO.2S810, and 0.7 percent 3(Na,K),0.A1,O3.
3810,.

When recalculated in terms of the components already
deseribed it is found to consist roughly of 76 percent
gehlenite, 17 percent akermanite, and 7 percent ferric and
ferrous compounds. A comparison of its properties with
those of artificial gehlenite having the same ratio of aker-
manite to gehlenite, but free of iron compounds, is given
in Table XX. It will be seen that the agreement is very

TABLE X_X.

Comparison of properties of natural Gehlenites with those of similar artificial
preparations.

Natural Artificial Natural Artificial Natural Artificial

Locality Valardefia, Monzoni, Tulare Co.,
Mexico. Italy. Calif.

J K. T. Allen. Rammelsberg. E. V. Shannon.
Gehlenite*. 2.5... 76 82 58 67 66 70
Akermanite ...... 17 18 28 33 28 30
Iron Compounds .. 7 14 6

SS ae ee 1.63 1.639 1.662 1.659 1.660 1.660

Lp ORE ER oe 1.633 1.632 1.657 1.655 1.657 1.656
Birefringence ..... .006 007 005 004 .003 004
BEUSIGY. faf0c)- (o-oo 3.039 3.024 3.008 3.02 3.01
STG Ca aa ies 1475° 20° 1504°-+5° 1875° 25° 1462°-+5°
Enquadus::.. ./ 2.545 1555°20° 1556°==5° 1500°=20° 1530°+-5°

Remarks: The data.on the Velardefia gehlenite were determined by F. EH.
Wright on the same sample as that analyzed. The optical data and thermal
data for the Monzoni gehlenite were determined by the writer on a sample
different from that analyzed by Rammelsberg. The optical data for the
California gehlenite were determined by the writer on the same sample as
that analyzed by Shannon.

close, and would be even closer if the values of the min-
eral were corrected for the presence of iron compounds
which tend to raise the indices of refraction, raise the
density, and lower the solidus and liquidus. In order
to ascertain whether the mineral gehlenite would invert
to some different form at higher temperatures, a charge
was heated for 16 hours at 1450° without showing any
change in indices of refraction. A second charge was

70 Buddington—Natural and Synthetic Melihites.

completely melted and quenched to form glass. This
glass was then recrystallized at 1450° and the resulting
crystals had the same indices as the original material.

California Gehlenites—A gehlenite from Crestmore,
near Riverside, California, has been described by W. F.
Foshag.’®> It occurs associated with spurrite and other
minerals. It is rather striking that such a rare mineral
as spurrite should occur associated with the relatively
rare mineral gehlenite at two such widely separated
localities. A specimen of the gehlenite loaned by the
U. S. National Museum (No. 87275) had the following
indices: 2» 1.662 + 0.008, e == 1.658 = 0.003.

A specimen of gehlenite from Tulare County, Califor-
nia, also loaned by the U. S. National Museum, had the
following indices: m» —1.660 + 0.003, 7 = 1.657 + 0.003.
This material was associated with garnet (index 1.815 +=
0.006), magnetite, and pyroxene, and had a density of
about 3.02. It has been analyzed by EK. V. Shannon, and
the recalculated analysis shows roughly about 66 percent
gehlenite, 28 percent akermanite, and 6 percent ferric and
ferrous compounds. A comparison of its properties
with those of artificial gehlenite having the same ratio
of gehlenite to akermanite, but free of iron compounds,
is given in Table XX. The agreement is very close.

The analysis by Shannon, reduced to 100, stated in
terms of its constituent molecules and compared with the
calculated molecules in a mixture of assumed components
most closely approximating the natural mineral in com-
position is given in Table XXI.

TABLE X XI.
Gehlenite from Tulare County, California.

Original Analysis Calculated Differ- Mixture

analysis reduced. Mol. molecules ence weight percent.
SiOx <2 88 27.75 4602 4814 1212 66 2Ca0.A1,0;.Si0,
AO ek eee 20.40 2490 2400 — 90 28 2CaO.MgO.2Si0,
MesOy ee weks 0.59 1.58 99 88 —11 4.5 3Ca0O.Fe.0,.38i0,
Me Oneness 43 43 60 49 —]11 1.5 2Ca0.Fe0.2Si0,
WAM st oes 4.18 A Noe O42 eel 026 = eG
(OB CP ee Paros 40.86 A067 F250 (SG
1gEOo 30
H,O— ° 04

100.80 100.00
* Am. Mineralogist, 5, 80-81, 1920.

Buddington—Natural and Synthetic Melilites. 71

Monzoni gehlenite—A specimen of limestone, labeled
Monzoni (Fassathal), with minute disseminated crystals
of gehlenite, was obtained by Dr. Wright from the U.S.
National Museum. The indices of refraction were deter-
mined on this mineral, and rough tests made for the loca-
tion of the solidus and liquidus.. If we assume this
material to be essentially the same as that analyzed by
Rammelsberg from this locality, and recalculate the
analysis in terms of constituent compounds, we find that
the mineral consists of about 58 percent gehlenite, 28 per-
eent akermanite, and 14 percent ferric and ferrous iron
compounds. The results of the study of the natural
mineral and a comparison of its properties with those of
an artificial preparation having the same ratio of gehlen-
ite to akermanite, but free of iron compounds, are given
in Table XX. If the relatively large percentage of iron
compounds in the mineral is taken into account, the
agreement is good.

TaBLE XXII.
Gehlenite from Monzont.

Original Analysis Calculated

analysis — reduced Mol. molecules Difference
Sie ee POETS 30.20 5008 5218 210
AOS 2.2. < 22.02 22.33 | 2155 2109 — 46
Lo a 37.90 38.43 6850 7508 +342
2.125, 9 eee 3.88 3.93 975 1025 S55
PEO; : 3.22 3.27 205 181 — 24
PEOIS. 22.32 1.63 1.65 230 ) aoe r. eyons
MnO ...... 19 19 27 § noe fa”
Ign. by dif-

eerenge ts 138
100.00 100.00

Table XXII gives the analysis by Rammelsberg of
gehlenite from Monzoni, stated in terms of its constituent
molecules and compared with the caleulated molecules
in a mixture composed of 58 percent 2CaO.Al,0;.Si0,,
28 percent 2CaO.MgO.2Si0,, 8 percent 2CaO.FeO0.2Si0,,
and 6 percent 2CaO.Fe.0,.Si0,.

The agreement between the properties of the artificial
preparations and the corresponding minerals is so close
that we feel warranted in concluding that the artificial
series of 2CaO.MgO.2Si0,-2Ca0O.A1,0,.S10, solid solu-
tions are pure synthetic analogues of the akermanite-

72 Buddington—N atural and Synthetic Melilites.

gehlenite mineral series. Reference is also made here
to the fact that Vogt'® studied the relations of akerman-
ite, gehlenite, and their mixtures in artificial slags, and
concluded that they form an isomorphous series. His
materials necessarily contained iron, manganese, soda,
and other impurities,.and differed to that extent from the
artificial preparations studied in this work.

GROSSULARITE.

The pure compound 3CaQO.Al,0.,.3810, 1s known in
nature only as the mineral grossularite, one of the garnet
eroup, crystallizing in the isometric system. As a com- ~
ponent of other minerals, however, it forms 90 percent of
sarcolite, up to 50 percent of some humboldtilites, and
isomorphous mixtures with other members of the garnet
_ group. ,

The mineral sarcolite is positive, uniaxial; erystalliz-
ing in the tetragonal system, with indices of refraction
NM —= 1.6035 and ne —1.6147. Since sarcolite contains 90
percent of 3CaO.A1,0;.38810,, this compound may have
similar, though somewhat modified, optical characters in
the form in which it exists there.

A series of solid solutions of 3CaQ.Al1,0,.8810, with
2CaO.MgO.2810,, 2CaQ.Al,03.810., 3Na,0.A1,0;.3810,
and 3CaQO.Fe,0,.388i0,, formed within a temperature
range above 1000°, have been described in this paper.
These solid solutions crystallize in the tetragonal system
and the 3CaO.A1,0,.8810, compound has the effect of a
negative uniaxial material with a moderate birefringence.
If we project the curves of the indices of refraction for
complete solid solutions of 3CaO.Al,0,.3810, and 3Na,0.
Al,O,.38810., we find that the former component will have
the =r gliaee No == 1.6383 22 0.000; 2x2 = 1624 = 000s ae
that it is uniaxial negative “with a birefringence of
0.011 + 0.005.

We therefore have evidence of at least three distinct
forms in which the compound 3CaO.A1;0;.3810, may exist
in solid solutions.

Further evidence of the polymorphism of this com-
pound is indicated by the birefringence observed in many

16 Toc. cit.

Buddington—Natural and Synthetic Melilites. 73

garnets. Many attempts have been made to synthesize
erossularite, usually without success. Shepherd and
Rankin,'* however, describe the formation of grossularite
by the action of aluminum chloride upon calcium ortho-
silicate heated with water under pressure in a steel bomb
at a temperature of 400° + 50°.

Artificial glass of the composition of grossularite was
erystallized for 16 hours at 980° and gave an unidentified
fibrous aggregate of moderate birefringence in which at
least two different compounds are present. At the
eutectic or melting point (1265°) it forms a mixture of
gehlenite, anorthite, and pseudo-wollastonite.

Grossularite held for possible inversion or dissociation
for 166 hours at 800° and for 16 hours at 1100° was
unchanged.

SARCOLITE.

The composition of the mineral sarecolite is wey
expressed as 3(9Ca.Naz)0.Al,0;.38i02, and the crystal
system is tetragonal. The indices of refraction given by
Zambonini'® are %» 1.6035 and ~ —1.6147, and the
density 2.92. The mineral is optically positive.

Bladed erystals formed by crystallizing a synthetic
mixture of the same composition for 16 hours at 1000°
have the indices 2 —1.631+0.005 and n» —1.615 +
0.005 ; the optical character being negative and the crystal
system tetragonal. Thus the indices of refraction of the
artificial erystals do not agree with those of the natural
sarcolite, and the optical character is different so that
the artificial crystals must be a polymorphic form, though
crystallizing in the same system.

To test for possible inversion phenomena, a specimen
of sarcolite from Monte Somma, obtained from the U.S.
National Museum, was subjected to a series of heat treat-
ments. The mineral was held for 156 hours at 700°, 166
hours at 800°, 96 hours at 1000°, and 4 hours at 1100°
without showing a trace of alteration except in an occa-
sional grain. In a few grains the alteration observed
might be interpreted as indicating inversion, but the
natural mineral could not be obtained in absolute purity

7 E. S. Shepherd and G. A. Rankin, this Journal, 28, 305, 1909.
$F. Zambonini, op. cit., p. 247.

74 Buddington—Natural and Synthetic Melilites.

and reactions with impurities may explain the changes
observed.

When the sareolite is held for 40 hours at 1150°, how-
ever, it breaks down and all the grains are locally so full
of minute, much higher refracting dots that they appear
cloudy and brownish in color. Some of the grains invert
to material of negative optical character, whereas others
remain unaltered except for numerous minute dots.
Some of the inverted grains show a trace of glass. A few
orains of the sarcolite contain smai! pseudo-cubes as the
result of recrystallization. Many of the clouded brown-
ish grains have a clear border which has higher indices
than the original sareolite. Similarly, when held for 2
hours at 1170°, the sarcolite grains are partly changed
to minute pseudo-cubes of higher indices of refraction
with here and there traces of glass. Positive evidence of
inversion without trace of melting is lacking and the
experiments do not prove whether such may or may not
take place, as the rate of reaction may be so slow that a
longer time interval or the presence of a mineralizer
would be necessary.

EH’ UGGERITE.

Weinschenk'® has described a mineral near Lake Selle
in the Monzonithal which resembles in some of its proper-
ties some of the gehlenite-akermanite series. It occurs
in limestone at the contact with monzonite and is in thick
four-sided tabular crystals, probably tetragonal. The
color is light apple green and the density is 3.18. Indices
of refraction for yellow light are essentially the same:
No == Ne == 1.691, with anomalous interference colors. It
is decomposed by dilute acids with separation of pulveru-
lent silica.

A chemical analysis of the material was made by
Hi. Mayr, and this on recaleulation gives about 40 percent
2CaO.AL,O0;.810,, 35 percent 2CaO.Mg0.2S8i0,, 10 percent
3dCaO0.Fe,0,;.3810,, 10 percent 3CaO.A1,0,.38i0., and 5
percent 3Na,O0.A1,0,.3810,. An artificial mixture of
this composition crystallized just below the solidus
(1845° + 10°) had the following indices for the erystals:

HH. Weinschenk, Z. Kryst., 27, 577, 1896.

Buddington—Natural and Synthetic Melilites. 75

NM» = 1.658 and ne — 1.654, tetragonal system. Crystals
formed at 1150° give the same results. It forms a
complete solid solution with a melting interval extend-
ing from the solidus at 1345° + 10° to the liquidus at
222 eed Ae

The optical properties of the mineral and of artificial
erystals equivalent in composition but formed at about
1345°, therefore, do not agree by a wide margin. The
difference is far greater than the limits of error involved.
The synthetic preparation, therefore, is a polymorphic
form involving the same compounds as the mineral
fuggerite.

Many specimens labeled ‘‘fuggerite’’ were obtained
by the writer from dealers and museums in the United
States, but all were too completely replaced by grossular-
ite or were too much altered to be satisfactory for experi-
mentation. One specimen alone still showed a minute
core of the original material, only partially replaced by
garnet. Its indices were similar to those given by Wein-
schenk and it showed conspicuous anomalous interference
colors. Such garnet pseudomorphs were described by
Weinschenk in his original paper.

HUMBOLDTILITE.

The term melilite was originally (1796) applied to the
reddish-brown to yellow erystals found in the leucito-
phyre at Capo di Bove, Italy, in allusion to the honey
yellow color. The name humboldtilite was subsequently
| (1882) given to the white to greenish erystals of similar
erystallog graphic habit found in the metamorphosed blocks
of limestone at Vesuvius. As a result of a study of the
chemical and crystallographic characters of these two
minerals by Des Cloizeaux and Damour (1843) the min-
eral humboldtilite was classified as melilite. Mierisch,2°
however, has given sufficient data as to differences in
manner of occurrence, crystallographic habit, color, min-
eral association, and cleavage between the or iginal meli-
lite of Capo di. Bove and humboldtilite of Vesuvius to
warrant the retention of the latter name to designate in
general a series of minerals of the melilite group similar

° Tscherm. Mitth., N. F., pp. 149-153, 1887.

76 Buddington—Natural and Synthetic Melilites.

in composition and character to that first described from
Vesuvius. They differ essentially from the mineral orig-
inally called melilite in having relatively less ferric iron
and in some cases a relatively greater percentage of
ferrous iron, and a larger percentage of the 3CaO.A1,O3.
3810, molecule.

TABLE XXIII.

Chemical analyses of Humboldtilite from Monte Somma, Vesuvius.

A B GH D E
SiO; ee eee ae 39.86 41.69 40.36 40.69 41.09
AMOgcih st eae 11.37 9.59 12.04 10.88 10.93
FEO, occ: spe ee 50 76 15 4.43 3.40
FEO. eee 1.78 3.75 1.53
MeQiie.escesoe nee 7.63 5.32 6.56 5.75 5.87
CAOI; ec eee 35.58 32.82 34.71 31.81 34.78
Na,0 Se eee 2.13 4.76 3.34 4,43 3.40
KO ACT eee 82 68 30 36 68
HO: hea 0.49 0.29 0.79
HO Ne ies h eee 0.10 0.05 0.13 24
Total... cee 100.26 99.71 100.51 98.35 100.39
Density 2). ee 2.975 2.925 2.92 to
2.95

A. White crystals, thickly tabular parallel to base, occurring in vugs and
as interstitial material in Vesbite, a lava composed of about 65% leucite,
18% humboldtilite, 20% pyroxene, 2% magnetite. See Jour. Wash. Ac. Sei,
vol. 10, No. 9, 272, 1920. Washington, analyst.

B. Crystals and interstitial material in vesbite. Crystal habit is tabular
parallel to base, combination of 1st order prism and base modified by 2d
order prism. Washington, analyst.

C. Crystals occurring in pockets in limestone masses included in leucitite.
Crystal habit thick tabular parallel to base; combination of 1st order prism
modified by ditetragonal prism and base. Colorless to transparent. Wash-
ington, analyst.

D. Analysis by Damour, Dana’s System of Mineralogy, 6th edit., p. 475.

EK. Analysis by Bodlaender, Neues Jahrb., 1893, 1, p. 17.

Monte Somma Humboldtiulite—tIn Table XXIII are
given the chemical analyses of three specimens of hum-
boldtilite selected by the writer and analyzed by Dr. H. S.
Washington. Two analyses made by other analysts are
included for comparison. A noteworthy difference lies
in the fact that the iron as found by Dr. Washington is
for the most part in the ferrous state, whereas the condi-
tion of the iron as reported by most of the other analysts
is the ferric oxide, the two oxides not having been sepa-
rately determined. |

Buddington—Natural and Synthetic Meliites.

TABLE XXIV.

TT

Comparison of Humboldtilite analyses with calculated composition of isomor-

Original
analysis
SUL eee 39.86
ICO. ee Se
2G: 0 ee 50
HeOe oi as Levis)
NERO) 2 sreress 7.63
CA a ee 35.58
IN aLO ve cece 2.13
ROO Ce sed 82
Original
analysis
SiOst. This: 41.69
ATO ss. 9.59
(oY Cee .76
MeO? 8554 4 Sis
MeOsii? 22: 5.32
OF\0 See ene 32.82
ONCE @ paar 4.76
LG 0 /Saeeaaae .68
Original
analysis
STO Eee ee rie 40.36
9 a ea 12.04
HOw si. . ka
PEO Fee se xyes eae
MEOE. 3.5. 6.56
CAO. f..27>. 34.71
es OM sais: 7 3.34 -
TO CER corse 30

Analysis

reduced.
39.98
11.41
00
1.79
7.65
35.69
2.15
83

Analysis

reduced.
41.94
9.65
AE
3.78
5.36
33.02
4.79
.69

Analysis
reduced.
40.52 -

12.09

LS
1.54
6.60
34.85
Soe
30

In Table XXIV
analyses of the humboldtilites, the analyses reduced to
.100 per cent, the constituent molecules, and for com-
parison the calculated molecules of the mixture of
assumed components most closely approximating the
natural mineral in composition.

In Table XXV are given the results of recalculating the
first three analyses of Table XXIII in terms of their
assumed components, together with the optical characters.
of the minerals analyzed.

phous matures.

Humboldtilite A.

Calculated Differ-
Mol. molecules ence
6630 6634 + 4
PIATGS S104 ee 75

ol SO 28
249 230 — 19
1898 1831 — 67
6362 6563 +201
347 347

88 88

Humboldtilite B.

Calculated Differ-
Mol. molecules ence
6958 6844 -—114

944 969, == 25

48 SO) shed

527 525 — 2
1330 13855 + 25
5886 5998 +112

TOTS) FETS

73 73
Humboldtilite C.

Caleulated Differ-
Mol. molecules ence
6719 6653 — 66
1185y,= 1191, 6

AT 40 — 7
214 197 — 17
1688 1575 63
6223 6373 +150
540 540

32 32

are

Mixture
weight percent.
2CaO.MgO.28i0,
3Ca0O.A1,0;.3810,.
2Ca0.A1.,0,;.Si0,
7 2CaO.FeO.2S8i10, |
7 3(Na,K).O.AI,0O;.

3810,
2 3CaO.Fe,0;.3810,

50

Mixture
weight percent.

2Ca0.Mg0.28i0,

3C0a0.A10;.38i0,

2Ca0.Fe0.28i0,

3(Na,K).0.A1,0.
3810,

3 3Ca0.Fe,0,.38i0,

37
dl
16
13

Mixture
weight percent.

2CaO.MgO.2Si0,

3CaO.A1,0,.38i10,

9 3(Na,K),.ALO;.
3S8i0,

6 2CaO.Fe0.2S8i0,

2 3CaO.Fe,0,.3810,

given the original chemical

78 Buddington—Natural and Synthetic Melilites.

TABLE XOCy..

Comparison of properties of natural Humboldtilites with those of similar
synthetic preparations.

A B C

Components Natural Artif. Natural Artif. Natural Artif.
Gehillenite ce jae oe 10 10 iia ss 8 10
Akermanite .,.. ck. ei 50 54 37 48 43 45
3Ca0.A1,0;.8810, .... 24 26 31 39 32 33
3Na,0.A1,0;.3810, =
3K,0.A1,0,.39i0, t Re ith 10 13 10 9 10
3CaO.Ge,0;.3810, .... 2 3 5 ee 2
Ferrous minerals .... 7 16 6

ORME TERRE HOES te cy ON 1.639 1.635 UWAGSi7/ 1.633 1.637 1.635

LO a wiPE rote ee he! mtv ertacs 1.633 1.633 ik63u) 1.630 1.631 1.632
Birefringence ....... .003 002 .006 003 .006 .003

The nature of the ferrous iron compound is unknown,
but for purposes of recalculation the writer follows
Schaller in assuming it to exist as a ferrous iron aker-
manite (2CaO.FeO.2S8i0,). |

For purposes of comparison, the optical data for arti-
ficial mixtures of similar composition, but lacking the fer-
rous iron compound, have been computed and are included
in Table XXV.

From an inspection of this table, it is evident that the
effect of the ferrous iron compound is to increase both
the indices of refraction and the birefringence.

The comparison of the natural minerals and the arti-
ficial mixtures shows an agreement of the optical charac-
ters well within the limits of the possible errors involved.
As a further check on this matter, several experiments
(Table X XVI) were made on the natural minerals, and
the results obtained are all confirmatory of the essential
identity of the natural minerals and the synthetic
preparations.

This interpretation, however, is conditioned by the
facts that the nature and properties of the ferrous iron
compound are unknown, and that accurate analyses and
indices of refraction of humboldtilites showing a wider
range of composition than those now available are very
much needed. ‘The amount and effect of ferric iron in the
humboldtilites also needs to be studied.

Table XX VI shows that all three humboldtilites when
_ held for 16 hours at 1150° showed no noticeable change.
All three, when completely melted and quenched to form

Buddington—Natural and Synthetic Melilites. 79

GTi OOO EE

Thermal data for Humboldtilites from Monte Somma.

Humoboldtilite A.

1150° 16 hrs. Clear and unaltered.

LEO? 40 hrs. Trace of dissociation. Poikilitic rods in some
grains, but most grains unchanged. Altered
grains are mottled in appearance, unaltered
grains. are fresh and clear.

1200° 16 hrs. Part of material is dissociated, and traces of
glass are present; part is unaltered and with-
out traces of glass. Indices of refraction of
altered material are higher.

1225° 16 hrs. Crystals plus a small amount of interstitial
glass.
1320° ihr: Glass plus few erystals. The crystals exhibit

anomalous bluish and brownish yellow inter-
ference colors.

1343° har All glass.
Humboldtilite B.
1150° 16 hrs. Clear and unaltered.
e702 20 hrs. Dissociated. For the major crystalline material

Nw — 1.642 and ne = 1.635.

Humboldtilite C.

1150° 16 hrs. Glass formed by melting at higher temperature
and quenching was recrystallized and the
erystals had the following indices
Nw = 1.639 nme = 1.633.

1170° 20 hrs. Unaltered; clear and fresh.
1200° 16 hrs. Most grains are clear without a trace of glass.
A few grains with impurities show traces of
glass.
1235° 16 hrs. Interstitial glass noticeable.
13138° ihr: Crystals with gray interference colors and glass.
1335° Thr. All glass.

glass, and when subsequently recrystallized at 1150° for
16 hours, exhibited indices of refraction which are within
the limits of error for those of the original material.

A and B are also similar to the synthetic preparations
inasmuch as they dissociate at similar temperatures, most
of the material having higher indices of refraction after
dissociation. No trace of dissociation was noted in
experiments with C.

The indices of refraction alone were determined on
two other Italian Specimens with the following results.
The first specimen is a limestone fragment, with hum-
boldtilite crystals occurring in pockets with biotite and
pyroxene. The crystals are short prismatic, a combina-

80 Buddington—Natural and Synthetic Melilites.

tion of the base and first order prism modified by the
ditetragonal prism. ». —1.636 ne =1.629. The other
specimen is from Monte Albano, Latium, Italy, and is
a leucite lava with medium sized tabular crystals of hum-
boldtilite partly enclosing leucite. ~» ==1.639 ne = 1.637.

Humboldtite from Latwwm.—tIn a recent article on
melilites F. Muillosevich?! describes melilite (humboldt-
ilite) obtained from blocks in peperino associated with
pyroxene, leucite, hauyne, and yellow garnet. In Table
XXVII are given the original chemical analysis, the

TABLE XXVII.
Humboldtilite from Latium.

Original Analysis Calculated Differ- Mixture

analysis reduced. Mol. molecules ence weight percent.
DIOS meee eae 41.07 41.24 6839 6802 —&37 40 2CaO.Mg0O.2Si0,
EAN O Bpeneaenes © 10.47. 10.51 1030 1055 425 38 3Ca0O.AI1,0;.38i0,
FeO; feo 3.80 3.82 239 236 —83 £412 3Ca0O.Fe,0;.38i0,
1 ENG) O Jenga tg None 10 (Na,K),0.A1,0;.
CAO ew ce okaae 33.92 34.06 6071 . -6157-- +96 3810,
MioOns tee 6.02 6.05 1501 1465 —36
IN ARON ke teres By PAS) Beil PAT 527
J GOA ence 5 1.04 1.05 111 aleliale

99.57 100.00

analysis reduced to 100, the constituent molecules, and for
comparison the calculated molecules for a mixture of
assumed components most closely approximating the
natural mineral in composition.

The indices of refraction for the natural mineral are
_given by Millosevich as 7» — 1.633 and ve — 1.629 with a
birefringence of 0.004. The indices of refraction for a
synthetic preparation of similar composition are %o,==
1.639 and: == 1.635 with a birefringence of 0.004. The
difference is greater than might be expected, but the
agreement may still be considered good.

Colorado humboldtilite. A humboldtilite from Gun-
nison County, Colorado, has been analyzed by Schaller
and described by Larsen and Hunter.22. It forms two-
thirds of a rock called uncompahgrite, and is associated

* Rend. Accad. Lincei, vol. 30, pp. 80-84, 1921.
*E. S. Larsen and J. F. Hunter, J. Wash. Acad. Sci, vol. 4, p. 473, 1914.

Buddington—Natural and Synthetic Melilites. 81

TABLE XXVIII.
Humboldtilite from Colorado.

Original Analysis Calculated Differ- Mixture
analysis reduced. Mol. molecules ence weight percent.
S28 es oe 42.07 43.80 7263 6830 —433 45 3Ca0.AI1,0,.3Si0,
BS hs .20 .20 25 — 25 34 2CaO.Mg0O.2Si0,
i; Ut) Ree 10.30 10.52 1052 1189 1137 10 2Ca0.Fe0.28i0,
2S eee 35.41 34.64 6174 6254 + 80 9 3(Na,K).0.A1,0;.
1 re 4.15 4,32 1072 1245 +173 3810,
iG) Saree 50 52 33 39 + 6 2 3CaO.Fe.0,.3Si0,
MeO het kA 2.18 2.27 315
oC —— igen sbiggto oo ooo! aie
LG 8 0 apa 3.24 ok 543 576 + 33
on eee Trace
fk foe a 82
| eee 90
ci) Ae 47

100.40 100.00

with pyroxene, perofskite, and apatite. Its density is
2.98. Table XXVIII gives a statement of the original
analysis, the analysis reduced to 100 with elimination of
water, apatite, and calcium carbonate, the analysis stated
in terms of its constituent molecules, and for comparison
the calculated molecules of a mixture of assumed com-
ponents most closely approximating the natural mineral
in composition. This analysis does not lend itself
readily to recalculation in terms of the components
considered.

A comparison of its optical properties with those
of a similar synthetic preparation is given in Table
XXIX. The agreement here, again, is as close as ae
be expected.

LA Bin LX.

Comparison of properties of Colorado Humboldtilites with those of similar
synthetic preparations.

Components. Natural. Artificial.
iieorina Witenes) ory ts br hrs 5 . 8. S 34 40
SOO Se ST 1 ee 45 50
3Na.0.A1,0,. 38i0, l 9 10
ae OALO Ronis § SS
SBOE GOB? hes. ore ak ross 2,

Perreus: noneriiss 3. ft 10
Hig ae Pea, oe Ede eRe 1.632 1.630
Uf: RUAN Ae kee 282) SO ee 1.626— 1.627
Bere PrIM CN CR Eo ia yo acncs Le" le wie’ 007 003

Am. Jour. Sci.—F irra Series, Vou. III, No. 13.—Janvary, 1922,
6

82 Buddington—Natural and Synthetic Melilites.

A specimen of the mineral supplied by Dr. Larsen was
heated for 20 hours at 1150°C. and exhibited no altera-
tion; heated for 20 hours at 1175°C., it dissociated. The
synthetic material dissociates at about the same tem-
perature. The mineral, called simply melilite by Larsen,
is here more specifically regarded as the humboldtilite
variety.

The agreement between the optical properties of the
humboldtilite minerals and homogeneous preparations
of similar composition formed above 1000°C. is very close
and within the limits of error involved. Assuming that
the pure solid solutions are essentially true equivalents
of the corresponding humboldtilites, it follows by analogy
that the compound 3CaO.Al1,0,.3810, enters into the hum-
boldtilites with the effect of a negative, moderately
birefringent, uniaxial compound, for which a mineral
equivalent as a separate entity is unknown.

The present explanation obviates the difficulty of
conceiving the formation of a negative uniaxial mineral
from an isomorphous mixture of two positive uniaxial
end members, as suggested by Schaller.

The humboldtilites, according to the interpretation
offered here, are essentially isomorphous mixtures of
positive uniaxial akermanite (2CaO.MgO.2Si0,) and a
negative uniaxial, moderately birefringent form of 3CaO.
Al,O;.38810, with minor amounts of gehlenite, a ferrous
iron compound, and 3RO.R,O0,.388i0, compounds.

FERRIC IRON-RICH MELILITE.

Among the specimens of melilite from Capo di Bove,
obtained by Dr. H. S. Washington from Professor F.
Millosevich, was one which proved to be a new member
of the group. It occurs in pockets, about 114 inch in -
diameter, of massive crystalline granular material asso-
ciated with nephelite and pyroxene, and as minute erys-
tals coating druses in a melilite-leucitite lava. The
crystals are yellowish-brown and coated with a needle-
like unidentified mineral. Their habit is tabular to
pseudo-cubic, or a combination of the first order prism
and base modified by the second order prism.

Under the microscope they are found to show a marked
zonal growth. The cores are isotropic for Na light and

Buddington—Natural and Synthetic Melilites. 88

give an anomalous berlin-blue interference color for day-
light. Surrounding the cores are zones showing yellow-
ish-brown interference colors grading outward into
material with normal yellow interference colors. The
index of refraction of the isotropic cores is about 1.654
and the indices of the material exhibiting the maximum
interference colors are” —1.666 and ~- = 1.661 with a
birefringence of about 0.005-0.006. Peg structure is
conspicuous.

The material is so intimately associated with nephel-
ite and pyroxene that it was necessary to grind it to a
fineness sufficient to pass a 150 mesh sereen. Bromoform
was then used to separate the lighter constituents, Klein’s
solution to effect a first separation of the melilite from
the pyroxene, and an electro-magnet to complete the puri-
fication. The melilite after this treatment was found,
when examined under the microscope, to contain about
4 percent pyroxene and 2 percent nephelite.

iW: @.O:E
Capo di Bove Melilite.
. i 10

Si@eees esc ok 40.56 40.03
ENG OSs ce Sect 6.18 5.66
Gs Os ost at 7.44 7.76
MeO eit aise il 40
IMM Oi S sovsnpect oy: none none
iW @ ae cts 9.53 9.43
CAO ae. 31.07 32.17
INasOe.. Sarees. 2.96 2.83
EO) es ita Se 1.76 172,
MONS Sue a, none none

100.01 100.00

I. Analysis of Capo di Bove melilite with 2 percent nephelite and 4
percent pyroxene. Washington, analyst.
II. Analysis recalculated for pure melilite.

Only a small amount of material was obtained from
a large amount of the granular nodule. <A chemical
analysis was made of this by Dr. Washington. Owing
to the small quantity of material available, only 0.2936
gram in all, water was not determined. In Table XXX
are given the original analysis and the analysis recalcu-
lated to eliminate the effect of the impurity of 2 percent

84 Buddington—Natural and Synthetic Melilites,

of nephelite and 4 percent of pyroxene. The analysis,
recalculated in terms of the assumed components, shows
considerable discrepancy.

The analysis differs from the others in that ferric oxide
is high and alumina low, the former evidently replacing
the latter. The indices of refraction also differ from
those of other specimens obtained by the writer from
Capo di Bove. To test for possible inversion phenomena,
a sample was heated for 5 days at 1100° + 25°. A few
grains are completely inverted but most are cloudy with
clear material working in from the edges. Held for 16
hours at 1140° + 20° the material is completely inverted
to an aggregate of clear bladed grains with lower indices
of refraction m» 1.647 and vm: 1.639. Held for 2
hours at 1165° there is a little glass present.

The composition of this material is such that it does
not fall within the limits of the compositions of the syn-
thetic mixtures studied, and no comparison can be made.

Another specimen from the same locality, among those —
received from Professor Millosevich, has the following
indices: o =-1.645 and nm 1.638. This material was
held for one week at 1050° and was then found to consist
of an unidentifiable cloudy brown aggregate.

Another specimen of melilite from Capo di Bove
obtained from the U.S. National Museum had the follow-
ing indices of refraction: “1.644 and ne 1.637.
The mineral occurs as reddish-brown pseudo-cubic to ~
tabular crystals in a crevice in leucitophyre. Thermal
data determined on this specimen are given in Table
XXXII.

The ferric iron rich melilites cannot be recalculated in
terms of the components used in the synthetic mixtures

TABLE XXX].
Thermal data for Capo di Bove Melilite. nw—1.644 and ne— 1.637.
700° 170 hrs. . Clear and altered.
850° 140 hrs. Completely changed to a fibrous aggregate of

mottled grains with irregular extinction.
Indices of refraction 1.648-1.656.

1050° 20 hrs. Dense aggregate of birefringent fibers with
minute dots of high refracting material.

Av. n= 1.661.
1135° 15 hrs. Similar to preceding, but more highly refracting

grains more conspicuous.

Buddington—Natural and Synthetic Melilites. 85

without discrepancies of some magnitude. The chemical
analyses by Gambonini of two ferric iron rich melilites
from Capo di Bove are given in Table XXXII. The

HU AP Tun) GXONGXCLT,

Comparison of analyses of ferric iron rich melilites by Zambonini with
calculated composition of synthetic miature.

1th ae. ae ITI.

STKOS Rin CeeeSe es Deieanet <9 6 40.14 39.20 38.38
INILLOs: SS ey BRP Cac ne CREO 6.47 7.56 5.90
CaOne ote eee see: wi 32.98 43.18 34.13
VIR ONG. AERP oP aan cae 6.33 6.41 6.64
EO ih ee haa se evens wy yor 9.95 11.34 10.98
MOOG Se eral act hier e acs: Scie 3
Ns OR accra re oe eas 2.18 2.20 3.97
Ol std inctercene iSie Tae tailor oe 1.49 1.45
EE Otto « exe cee oe, & en he 27 21
TO hall Sas a aio paemiets & Trace Trace

100.34 100.56 100.00

I. Yellow melilite from Capo di Bove.
II. Brown melilite from Capo di Bove.
III. Calculated composition of mixture of 45 percent 2CaO.Mg0O.2Si0,,
35 percent 3CaO.Fe,0;.38i0., 10 percent 3Na,0.A1,0;.3Si0, and 10 percent
2Ca0.A1,0;.Si0..

composition of a synthetic mixture composed of 45 per-
cent akermanite, 35 percent andradite, 10 percent 3Na,0.
Al,O;.38S10,, and 10 percent gehlenite is given for com-
parison. The erystallized synthetic mixture of this
composition has the following indices of refraction:
N» —= 1.676 and 7 1.658. These indices are conspicu-
ously higher than those of the natural melilites of similar
composition, if we may assume that our mineral was
similar to those analyzed by Zambonini. The natural
mineral, when heated at 850°, changes to a fibrous aggre-
gate of higher indices of refraction and the actual inver-
sion or dissociation point may be much lower. ‘The syn-
. thetic preparation and the natural mineral therefore do
not correspond.

The following results show that the synthetic prepara-
tions may have different phases. A glass consisting of
10 percent 3Na,0.A1,0,.38810.,, 67.5 percent 2CaO.MgO.
2810,, and 22.5 percent 3CaO.Fe,0,.38i0, was held for
1 month at 700° + 25°. The result was an aggregate
of minute spherulites with blue gray fibers with an aver-

86 Buddington—Natural and Synthetic Melilites. —

age index of refraction of about 1.695 and more strongly
birefringent fibers with index about 1.680. The indices
for the same preparation erystallized at the solidus are
Nw —1.654 andre —1.644. A mixture composed of 10 per-
cent 3Na,0.A1,0,.3810,, 45 percent 2CaO.MgO.28i0.,,
and 45 percent of 3CaO.Fe,0,.3810,, heated for 1 month
at about 700°, gave an aggregate of low birefringent
spherulites with an average index of about 1.730 + 0.010
and more strongly birefringent fibres. This is con-
trasted with the indices ”w = 1.688 and ”: = 1.658 for the
same preparation crystallized at the solidus. Several
experiments with other mixtures at the same tempera-
ture for the same length of time gave similar results,
with varying high indices of refraction.

SUMMARY

The character of the work upon which this paper is
based and the conclusions reached are briefly summarized
here. :

Over 100 synthetic crystalline mixtures of 2CaO.MgO.
28i0,, 2CaO.Al,03.8i0,, and 3RO.R.0,.3810, compounds
were prepared from appropriate glasses at temperatures
above 1000° (the approximate lower limit of experi-
ments); and their homogeneity, the optical characters
of those forming homogeneous mixtures and of the dom-
inant phase of some of those forming inhomogeneous
mixtures, and the temperatures of complete melting, were
determined.

The compounds 2CaO.Mg0O.2810, and 2Ca0Q.Al1,O3.
SiO, have previously been shown to form a complete
series of solid solutions at the solidus, and are here shown
to agree very closely in their properties with the aker-
manite-gehlenite series of minerals. 3

In certain homogeneous solid solutions with the other
compounds at certain temperatures 3CaO.A1,0,.3Si0, is
shown to have the properties of a negative, uniaxial,
moderately birefringent crystalline compound.

Mixtures of 2CaO.Mg0O.28i0,, 2Ca0.A1,0,.Si0,, and
a mixture of 90 percent 3CaO.AI,0,.3Si0, + 10 percent
3dNa,O.A1,0;.8810, form a complete series of solid solu-
tions, except for a trace of inhomogeneity in some pre-
parations high in 2CaO.Mg0O.2Si0O,. Mixtures -of these

Buddington—Natural and Synthetic Melilites. 87

compounds yield erystallized products essentially similar
in composition and properties to the minerals of the
humboldtilite series, varietal members of the melilite
group relatively poor in ferric iron.

The humboldtilites are interpreted as essentially isomor-
phous mixtures of positive uniaxial akermanite (2Ca0O.
Mg0.2Si0.,) and a negative, tetragonal, uniaxial, moder-
ately birefringent form of 3CaO.A1,0;.3810, with minor
amounts of gehlenite, a ferrous iron compound, and
3RO.R,0,.38S10, compounds. The compositions of the
humboldtilites he in a zone which exhibits the lowest
temperatures of complete melting for the components
involved.

Mixtures of 3CaO0.Fe,0,.38i10,, 2CaO.Mg0O.28i10,,
3Na,0.Al,0,.3810,, 2CaO.Al,0;.810., and 3CaQ.Al,QOg.
~3810,, similar in composition to some of the ferric iron
rich melilites, were studied, and the properties of the
synthetic crystalline material were found to be quite
different from those of natural minerais of similar
composition. -

The natural ferric iron rich melilites probably form at
temperatures lower than those of the present experi-
ments since some of such melilites investigated decom-
pose or invert at temperatures at least as low as 850° C.

Three new analyses of humboldtilites and one of ferric
iron rich melilite by Dr. H. S. Washington are given.
The ferric iron rich melilite differs from any hitherto
analyzed.

This study serves to eanlotnc the great complexity
of this group of minerals and the necessity for further
data on their composition and properties, and for further
experiment in synthesis. The extent to which such known
and hypothetical compounds as 3M¢0O.Al1,0,.3Si0. ;
dHeO.Al,0,.8810,; 3K,0.A1,03.38i0. ; 2C0a0.FeO. 28i0.;
2CaO.Fe,0,;.810,, etc., may enter into the melilite min-
erals, and their corresponding effect, is yet to be studied.

In conclusion, the writer desires to express his thanks
to Doctors N. L. Bowen, H. EK. Merwin, and H. S. Wash-
ington for their friendly criticism and assistance in this
work.

Geophysical Laboratory,

Carnegie Institution of Washington,
Washington, D. C., June 20, 1921.

88 I. W. Clokey—Carex Notes.

Arr, IIJ.—Carex Notes; by Ira W. Croxzry. Carex
apoda sp. nov. With Plate IT.

Culms 7-S8dm. tall, erect, sharply angled, smooth,
phyllopodic, dark brown at base. Leaves shorter than
the culm, 3.5-5mm. wide, flat, rough on the margins,
situated on the lowest third of the culm. Sheaths over-
lapping, loose; ventral band hyaline, white or tawny,
easily ruptured. Spikes 4-6, contiguous but distinet or
the lowest somewhat separated, erect, upper sessile or
subsessile, lower one or two short peduncled. Terminal
spike gynaecandrous, lateral spikes pistillate, all many
flowered; 10-20mm. long, 5-7mm. wide. Lowest
bract or bracts leaf-like, shorter to about the length of
the inflorescence, upper bracts scale-like. Scales lanceo-
late, acute to short cuspidate, about as wide as but shorter
than the perigynia, brown, somewhat shining; midrib
obsolete or narrow and a little hghter than the balance of
the scale. Perigynia nerveless, compressed, narrowly
oblong or narrowly elliptic, sessile, white or tinged with
light brown, darkest over the achene, 4-4.2mm. long,
1.2mm. wide, membranous, smooth, with short, cylindric,
dark brown, emarginate beak. Achene triangular, ellip-
tic, hght brown, dull, 1.3-1.4 mm. long, .7-.75 mm. wide, on
a slender stipe 1 mm. long raising the achene to the center
of the perigyma. Stigmas three.

Rhizoma verisimiliter cespitosum. Culmus 70-80 cm. altus,
strictus, ad basin vaginis aphyllis purpureis obtectus. Folia.
culmo breviora plana angusta fere glabra. Spiculae 4-6
oblongae 1-2 em. longae densiflorae obtusae contigue sessiles, ima
remotiuscula brevissime pedunculata, terminalis gynaecandra
clavata, laterales 2 longiones. Bracteae superiores setacez ima
foliacea evaginans spicam terminalem attingens. Squameze ®&
oblongo-ovatae acutae vel brevissime cuspidatae atro-purpureae
nervo medio obscuro. Utriculi squamas subaequantes sed lationes
erecti elliptice 4-4.2 mm. longi glagri albidi leviter pupreo macu-
lati plurinervosi haud stipitati, rostro brevi bidentulo apiculati.
Nux elliptica stipitata trigona. Stigmata 3.

The above description is based on specimens collected
by Prof. Aven Nelson and J. F. Macbride in ponds at an
elevation of 1794 meters, at Mackay, Custer Co., Idaho,
Aug. 1, 1911. Their number 1533, deposited in my her-
barium, is designated as the type. Their number 1535,

Am

jours Sci Vale Ml, W922:

Plate II.

I. W. Clokey—Carexz Notes. 89

collected at the same locality, belongs here. Both num-
bers were distributed as Carex atrata L. from the Rocky
Mountain Herbarium.

Carex apoda belongs to the Atrare Kunth. or Melonan-
_ the Drej. and is probably most closely related to Carex
atrata lh. and Carex epapilosa Mackenzie. It can be dis-
tinguished from these two species as well as from Carex
atratiforms Britton (Carex atrata L. var. ovata (Rudge)
Boott) with which it might be confused, as follows:

The margins of the perigynia much narrower than the
achene
Perigynia 4mm. long, 1.2mm. wide. Upper lateral spikes ses-
sile or subsessile, lower spikes short peduncled
Carex apoda
Perigynia 2.5-3mm. long, 1. mm. wide. Lateral spikes on
slender peduncles |
Carex atratiformis Britton
The margins of the perigynia as wide as the achene
Perigynia smooth, achene long stipitate
Carex epapillosa Mackenzie
Perigynia ee Oran achene short stipitate
Carex atrata L.

In addition to the above, Carex apoda differs from Carex —
atrata L. and Carex epapillosa Mackenzie, in having the
leaves scattered over the lowest third of the culm, not
clustered at the base; the scales are about as broad and
noticeably shorter, not narrower and as long or longer
than the perigynia. Further Carex apoda differs from
Carex atrata lL. in having the perigynia smooth and the
achene long stipitate. Finally the perigynia of Carex
apoda, having the length more than three times the width,
differs from that of any other species in the group.

Carex Paysonis sp. nov.

Grows in clumps; rootstock slender, ight brown, cov-
eréd with dull, ight brown scales which are not fibrolous.
Culms hght green, slender, rough below the head, 3-5 dm.
high, about twice the length of the leaves, phyllopodic,
with base surrounded by the non fibrolous remains of the
leaves of the previous year, dull brown not purplish
tinged at base. Leaves 3-6 to the fertile culm, situated
close to the bottom of the culm, light green, attenuated

90 — I. W. Clokey—Carex Notes.

1.5-3 mm. wide, flat, slightly rough on the edges. Culm
otherwise naked except for a slender green leaf or bract
4-12 em. below and not reaching the top of the inflores-
cence, with or without a pistillate spike. Sheaths over-
lapping, ventral band hyaline, white or tawny, easily rup-
tured. Terminal spike staminate, sessile or subsessile,
narrowly elliptic, 11-18 mm. long, 3-5 mm. wide. Fre-
quently one or two very small staminate or androgynous,
sessile spikes at the base of the terminal spike. Lateral
pistillate spikes 2-3, separated, upper sessile, the other
one or two short peduneled, ascending, elliptic, acute or
acutish, 8-20mm. long, 4-8mm. wide. All spikes dark
brown or black. Upper bracts scale-like, brown; lower
bracts slender, brownish, shorter than the subtended
spike. Scales acute or acutish, dull, dark brown, uni-
form, midrib obsolete; staminate scales broadly lanceo-
late; pistillate scales lanceolate, much narrower than and
one half to almost as long as the perigynia, hidden or
almost hidden by the perigynia. Perigynia strongly com-
pressed, sessile, glabrous, dull, broadly elliptic, 4-5 mm.
long, 2-3 mm. wide, texture thin, lower half white, upper
half and beak dark brown or black, lateral nerves usually
well defined; beak cylindric, one third millimeter long,
emarginate. Achene situated low in the perigynia, short
stipitate, oblanceolate, pointed at both ends, smooth, tri-
angular, tan colored, 2 mm. long, 8.-9. mm. wide. Stigmas
three.

Rhizoma laxe cxspitosum et stoloniferum. Culmus 30-40 cm.
altus versus apicem leviter curvatus. Folia culmo breviora ear-
inata angusta fere glabra. Spiculae 3-5 remotae + peduncu-
latae, ima longe-peduneculata, terminalis 4, altera ¢ brevis
adjecta, laterales ¢ ellipticae fere turgidi. Bractea infirma
brevis evaginans spicam-terminalem haud attigens. Squamae 2
lanceolatae acutae nigricantes. Utriculi squamas subaequantes
sed multo lationes patentes papyracei late ovales compressi
4-5 mm. longi stramineo-virides grosse purpereo maculati glabri
tenuiter plurinervii, haud stipitati in rostrum breve ore integro
vel bidentulo contracti. Nux gees subsessilis trigona.
Stigmata 3.

The above description is based upon plants collected by
Dr. and Mrs. HE. B. Payson, No. 2224, on dry rock slides on
the eastern slope of Grand Teton, J ackson’s Hole, Wyom-
ing. All the material of this number has been examined.
Named for Mrs. Lois B. Payson. The type is deposited
in my herbarium.

I. W. Clokey—Carex Notes. 9t

Carex Paysonis belongs to the Atrate Kunth or Mel-
onanthe Drej. Though phyllopodic it is most closely
related to the aphyllopodic Carex spectabilis Dewey.
Mr. Theodore Holm, who has done so much to differ-
entiate the species in this part of the troublesome Atrate
Kunth. or Melonanthe Drej., has kindly examined some
of this material and places it ‘‘not far from Carex spec-
tabilis Dewey, but very distinct from this.’’ Carex Pay-
sons can be separated from the most closely related
species by the following key:

Culms aphyllopodie
Carex spectabilis Dewey
Carex montanensis Bailey
Carex venustula Holm
Carex macrochaeta C. A. Mey
Carex Tolmier Boott

(As defined by: Mr. Theodore Holm)
Culms phyllopodic
Pistillate spikes 3-6, perigynia 3 mm. long

Carex Tolmier Boott

(As deseribed in Rydberg’s Rocky Mountain Flora)
Pistillate spikes 2-3
Pistillate spikes filiform stalked, drooping
Carex atrofusca Schk.
(Carex ustulata Wahl.)

Pistillate spikes erect, upper sessile, lower short peduncled
Seales with pale midvein extended into a short awn, longer
than the stipitate perigynia

Carex microcheta Holm

Seales at most acute, uniform dark brown, midrib obso-
lete, shorter than the perigynia which is not stipitate
Carex Paysonis

EXPLANATION OF PLATE II.

Fig. 1—Carex apoda.

Fig. 2—Idem pistillate scale.

Fig. 3—Id. pistillate scale.

Fig. 4—Id. perigynia.

Fig. 5—Id. achene.

Fig. 6—Id. perigynia, longitudinal section.
Fig. 7—Carex Paysonis.

Fig. 8—Idem staminate scale.

Fig. 9—Id. pistillate scale.

Fig. 10—Id. _ perigynia.

Fig. 11—Id. —achene.

Fig. 12—Id. _perigynia, longitudinal section.

Drawings by Miss Rena Duthie.

92 Scientific Intellugence.

SCIENTIFIC INTELLIGENCE

I. Curmistry AND Puysics.

1. A New Process for Determining Fluorine.—Since the
methods that have been employed for this determination are dif-
ficult and often inaccurate, the new process devised by M.
TRAVERS appears to be very important. He precipitates the
fluorine in cold solution in the form of K,SiF, and titrates the
washed precipitate by means of 1/5 normal K O H in the pres-
ence of boiling water, using penolphthalein as an indicator. The
whole process can be carried out in glass vessels.

The fluorine is brought into solution in the form of an alkaline
fluoride, then a solution of potassium silicate of known strength
is added in sufficient amount to furnish about double the theo-
retically required quantity of silica. The solution is then neu-
tralized with ordinary hydrochloric acid, with the use of heli-
anthine as an indicator, and a slight excess, about 2 ec, of the
acid is added. Then solid potassium chloride is added to the
extent of about 20% of the liquid.. The precipitate is filtered
upon a dense filter, washed with 20% KCl solution or with
50% alcohol until the washings are neutral to helianthine, and
then the precipitate with the filter paper is boiled with water and
titrated as mentioned above. Of course, the volumetric KOH
solution should be free from carbonate. Test analyses showed
excellent results with HK F,, and the author states that the
presence of boron does not interfere. It is expected that the
method will be useful, not only in the ordinary determination of
fluorine, but also in the study of compounds in which a part or
the whole of this element is disguised in the form of complex
radicals.

It should be stated that the author has used this method suc-
cessfully, with a reversed precipitation, for the determination of
silica. This process was carried out in silver vessels, and, while
this process is designed for general use in the determination of
silica, it is evident that it would be particularly useful for silica
in the presence of fluorides—Comptes Rendus, 178, 714, 836.

H. L. W.

2. The Separation of the Klement Chlorine into Isotopes.—
Wiuu1AmM D. Harkins and ANson HaAygs, by the fractional dif-
fusion of hydrogen chloride gas through clay pipe-stems in a
very elaborate apparatus, have obtained heavier fractions of the
acid which gave for the atomic weight of chlorine values of about
39.49 in place of 35.46, the constant of ordinary chlorine. The
atomic weights were compared by getting the purified acids in
solution to practically the same density by dilution and careful
employment of the pycnometer, and then titrating the solutions
by the use of sodium hydroxide in weight burettes.

Chemistry and Physics. 93

The increase in the atomic weight seems very small when it
is considered that Harkins supposes the heavier isotope to have
the atomic weight 37.00, but it appears that the small increase is
a definite one, and the authors say that this seems to be the first
separation of isotopes reported from which there is any definite
evidence.

The work on the diffusion of chlorine and hydrogen chloride
was begun in 1915, and Harkins has been aided by several
co-workers in the investigation. Several previous announce-
ments of successful results have also been made previously.—
Jour. Amer. Chem. Soc., 48, 1808. H. L. W.

3. Qualitative Chemical Analysis; by OLIN FREEMAN TOWER.
8vo, pp. 89. Philadelphia, 1921 (P. Blakiston’s Son & Co.
Price, $1.50 net ).—This is the fourth edition, revised, of a labora-
tory manual for use in inorganic qualitative analysis. It gives
unusually clear and full directions, together with numerous
_ explanatory notes, in connection with the analytical processes.
While many teachers of qualitative analysis use the course as an
opportunity to give a thorough drill in the writing of chemical
equations, the author of this book does not appear to favor such
a plan, for, besides a few equations given in the theoretical intro-
duction, scarcely any others are given, although the chemical
formulas of products formed in the operations are mentioned.
In connection with descriptive matter the student is frequently
given references to text-books of general chemistry.

In conclusion it may be said that the book gives a very good
course of analysis, with well-selected methods very satisfactorily
presented. H. L. W.

4. A Course of Instruction in Quantitative Chemical Analy-
sis; by Grorce McPHat SmitH. 8vo, pp. 218. New York,
- 1921 (The Macmillan Company).—This is a revised edition of
a work which first appeared about two years ago. It presents a
course of laboratory work for beginners in the subject, with very
full and clear directions, copious explanatory notes, and an excel-
lent list of questions. There are also 100 analytical problems
without answers, equally divided between gravimetric and vol-
umetric analysis.

The book in general gives a very good impression, but it is
unfortunate that it teaches a slow and difficult method of weigh-
ing, including the finding of the ‘‘rest-point’’ of the balance by
the use of long swings. This is likely to discourage the beginner
and to handicap him greatly-in his subsequent analytical work.
It is far easier and quicker, as well as decidedly more accurate to
employ the shortest swings that can be seen readily, and to use
the center of the graduated scale as the basis, since any deviation
from equilibrium in the balance is eliminated in the practically
invariable practice of weighing by difference. H. L. W.

94 3 Scientific Intelligence.

5. American Chemistry; by Harrison Hae. 8vo, pp. 215.
New York, 1921 (D. Van Nostrand Company).—We find in this
book a series of essays in popular style giving very interesting
accounts of various achievements in applied chemistry and their
importance in some of the large industries. There are 63 good
illustrations, many of them of full-page size; there are many
striking statistical statements and comparisons; there are numer-
ous references to the literature, and the topics discussed are of
great general interest. The book is, therefore, a useful one for
the general reader with scientific inclinations, and it appears to
be well suited for arousing the interest of students. H. L. W.

6. Report on Atomic Structure.—The report of a committee
on Atomic Structure appointed by the National Research Coun-
eil and consisting of Professors David L. Webster of Stan-
ford University, and Leigh Page of Yale University, has
been published by the above named body. In the first part,
entitled The Present Conception of Atomic Structure, Professor
Webster has discussed the evidence as to the existence of electrons
and nuclei; the number of electrons in the atom; the specula-
tions of Lewis and Langmuir as to the position of the electrons as
deduced from chemical statics; Bohr’s theory of energy levels,
and Parson’s hypothesis as to the forces holding electrons in their
places. Various questions as to the structure and dynamics of
the nucleus raised by radioactive data, and the significance of
the data of radiation and chemical dynamics as bearing on the
structure and dynamics of the electron, are also reviewed. The
author inclines to the radical view that in this field the law of
the conservation of energy is only statistical.

The second and longer part of the report, by Professor Page,
is devoted to The Dynamical Theory of Atomic Structure. The
author shows how far the postulate of electrons revolving about
a central nucleus, and the quantum theory, are able to account
quantitatively for the spectroscopic data. The various topics
treated are the spectra of hydrogen and similar elements, X-ray
spectra, the Kossel relations, the Stark effect, the Zeeman effect
and the Ritz formula— Bull. Nat. Research Council 2, 336, 1921.

F, E. B.

7. Velocity of Sound at High Temperatures.—A research
published under the above title by Messrs. H. B. Dixon, C. Camp-
bell, and A. Parker, was undertaken with the object of arriving
at formule for the variation of the specific heat of gases with the
temperature, which might be applicable to the extremely high
temperatures reached in explosions. The result forms an
important contribution to the knowledge of the gradients of
these curves. The sound method was chosen chiefly for the rea-
son that the velocity of sound in a heated gas gives the ratio of
the specific heats at the temperature of the experiment, and not

Chemistry and Physics. 95

as in the method of mixtures at a mean Lemperaunre between that
of the heated and of the cooled gas.

In the experiments as carried out, direct measurements were
made of the time taken by a sound wave to travel a known dis-
tance through the gas which was contained in a coiled tube so
that it could be raised to the desired temperature. Each end of
the tube was closed by a steel diaphragm which might be struck
by a hammer co} ntrolled by an electromagnet. As the compres-
sion travelled along the tube it successively raised two disks
placed about 18 meters apart, breaking a circuit, and recording
the interval of time elapsed upon a pendulum chronograph. As
certain small portions of the tube at either end projected beyond
the bath, or furnace, and were cooled by a water jacket, it was
necessary to introduce a proper correction for these ends.

The different gases studied were air, nitrogen, carbon dioxide,
nitrous oxide, methane, ethane and argon. Tubes of lead, steel
and silica were employed. The lead tube was conveniently used
for temperatures up to 100° C. and the steel tube for the higher
temperatures up to 1000° C. Exception had to be made however
in the case of air and carbon dioxide as they attacked the metal.
For these gases a tube of silica heated electrically by a wire coiled
around it was employed. The relation between the velocity of
~ sound in a tube and that in a free gas is discussed and a satis-
factory correction formula adopted.

The means of the velocities as found from the different tubes
were then plotted and tabulated for the different temperatures
and the various gases. The formulas both for the ratio of the
specific heats and their difference are introduced and the values
calculated. The final result sought, namely, the value of the
specific heat expressed as a function of the temperature is now
calculated, and discussed with reference to the determinations of
other investigators. For the detailed results the reader will
desire to consult the original memoir.—Proc. Roy. Soc. 100, 1,
L921. FE. E. B.

8. Traté de Dynanique; by JEAN D’AtEMBERT. Vol. I, pp.
XL, 102. Vol. il, Pp. 187. Paris, 1921. (Gauthier-Villars et
Cie). This is areprint in the collection, Les Matres de la Pensée
Scientifique, of the work of D’Alembert edited by Maurice Sol-
orine. As these memoirs are most complete they may, in the
words of the editor’s notice, be regarded as indispensable docu-
ments to historians of science and civilization. They also offer to
the student an easy and inexpensive means of meeting at the
source the experimental methods and ingenious processes of the
great investigators. To many these concrete methods will be far
more fertile and suggestive than the schematic rules of the
manuals. F. BE. B.

9. Heat and Inght; by 8S. E. Brown. Pp. numbered 170-

96 Scientific Intelligence.

428, Cambridge, 1921 (Cambridge University Press). This
volume contains sections IV and V of the author’s Experimental
Physics and may be described as a laboratory manual of high-
school, or first year college grade. It contains directions for a
large number of experiments, together with numerous problems
to test the student’s understanding of the principles involved.
It is self-contained in that it does not require the study of any
other text. The press work is inferior. F. E. B.
10. Physik und Erkenntnisstheorie; by E. Grurcxs, Pp. II,
119. Leipzig, 1921 (B. G. Teubner).—This brochure is an exam-
ination of the fundamental postulates and principles of physics
considered as a system of philosophy. Its scope is about that
of the order of a doctor’s dissertation. The author does not
appear to be particularly sympathetic toward the Einstein
hypothesis. F. E. B.
11. Séries Trigonométrique; by Maurice Lecat. Pp. VIII,
168. Louvain, 1921 (Chez l’Auteur). This is a very complete
bibliography of the subject indicated. The author index occu-
pies 133 pages, and a list of the journal and collections cited fills.
16 pages more. Supplementary tables show the distribution of
the articles according to geographical location, language, and
university. The appendix contains a list of papers on the Cal-
culus of Variations which have appeared during the period 1915
to 1920. BE Bt
12. Die Grundlagen der Geometric; by LotTHarR HEFFTER,
Pp. II, 27. Leipzig, 1921 (B. G. Teubner).—A discussion of
the fundamental postulates of geometry. The treatment pos-
sesses such rigor as has been deemed necessary that they may
serve for the foundation of analytical and projective geometry
as developed in the two volume Lehrbuch der Analytische
Geometrie of the author. | F. E. B.

Il. Gro.ogey.

1. A prehistoric Human Skull from Rhodesita—Nature of
November 24 states that the new skull from Rhodesia was exhib-
ited by Dr. A. SmiraH Woopwarp at a meeting of the Zoological
Society on November 22. The skull, which was found in the
Broken Hill Mine at a depth of 60 ft. below the water-level and
90 ft. below ground level, is in a remarkably fresh state of preser-
vation. It is much broken on the right side and the lower jaw
is missing. The brain-case is of modern human type, and the
bone not thicker than that of the ordinary Huropean; the
capacity, though not yet accurately determined, is clearly above
the lower human limit. The orbits are large and square, with
pronounced overhanging ridges much extended laterally. The
forward position of the foramen magnum indicates that the skull

Geology. 97

was poised on an upright trunk. The palate is large, but typi-
cally human, and adapted to perfect speech. It is remarkable
that the teeth are much affected by caries. The lower jaw must
have been massive and larger than the Heidelberg jaw. The
appearance of flatness of the frontal area suggests a comparison
with Pithecanthropus erectus. Dr. Smith Woodward ‘was
inclined to find the nearest approach to the Rhodesian skull in
the Neanderthal type from La Chapelle aux Saints in France.
Though markedly modern in regard to the brain-case, in its facial
characters, while it is essentially human, it appears to hold a
position between the gorilla and the Neanderthal man. Frag-
ments of the long bones, both femur and tibia, which have been
found indicate that, unlike Neanderthal man, Rhodesian man
~walked in a perfectly upright posture. Dr. Smith Woodward
regarded Rhodesian man as possibly a later development than
Neanderthal man, but Prof. Elliot Smith suggested that he might
represent a primitive type of which Neanderthal man might be
a highly specialised form.

2. A new Generic Name for Pliocyon Marsh; by M. RB.
THORPE (Communicated ).—At the time of the publication of my
paper on ‘‘Two new Carnivora’’ (this Journal (5), 1, June,
1921), I was not aware of the fact that the generic name Pliocyon
was preoccupied. Since such was the case, I now propose the
name Arwocyon (dapais, slender) to supplant my Pliocyon, and
the type of the genus will therefore be Arwocyon marsht.

In April, 1918, W. D. Matthew described a new genus of dogs
under the name of Pliocyon. The type of this genus was col-
lected in the Snake Creek beds (Lower Pliocene) of western
Nebraska, and represents a genus totally distinct from Arwocyon,
which is from the Rattlesnake beds (Middle Pliocene) of Oregon.

3. Publications of the Umted States Geological Survey,
GEORGE OTIS SMITH, director.—The well-understood difficulties
connected with the printing industry the past year have made it
impracticable to present promptly and in full the publications of
the U. S. Geological Survey as has been the custom in this Jour-
nal (see earlier, 50, 469-70). The most important issues received
are the following:

TopocrAPHic ATLAS: Fifty-one sheets.

GroLocic Fouios: No. 211, Elkton-Wilmineton Folio; by
F. Bascom and B. L. Minurr. No. 212, Syracuse-Lakin Folio;
by N. H. Darton. |

PROFESSIONAL Papers: No. 121, Helium-bearing natural gas;
byo 5. hocers, No. 12s. Part... .No-129. Parts A, B.

Forty-first Annual Report of the Director for the year ending
June 30, 1920.

BULLETINS :—No. 679, Microscopic determination of non- .
opaque minerals; by E. S. Larsen. No. 682, Marble resources
of southeastern Alaska; by E. F. BurcHarp and T. CHApin. No.

Am. Jour. Sct.—Firtu Seriszs, Vou. III, No. 18.—Janvuary, 1922.
|

98 ‘ Scientific Intelligence.

702, Oil possiblities, Baxter Basin, Wyoming; by A. R. SHuutTz.
No. 703, Iron and other Industries of the Sarre district; by
A. H. Brooxs and M. F. LAaCrorx. No. 704, Igneous rocks,
Essex Co., Mass.; by C. H. Cuapp. No. 706, Iron Resources of
Europe; by M. Rorster. No. 713, Resources of Fort Hall Res-
ervation, Idaho; by G. R. Mansrienp. No. 714, Mineral
Resources of Alaska; by A. H. Brooxs et al. No. 719, Petroleum
in Alaska; G. C. Martin. No. 721, Petroleum Resources of
Kern Co., ‘Calif. ; by W. A. ENGLISH. Also parts of Nos. 715,
716, 725, 726.

Minerat RESOURCES OF THE eee States; 1918, 1919, 1920,
separate chapters and Summaries.

WATER SUPPLY Papers: Nos. 447, 449, 453, 456, 459, 460, 462,
466, 467, 468, 470, 471, 475, 476, 481, 491. Also parts of 490,
500.

III. Misce,LANrEous ScreNTIFIC INTELLIGENCE.

1. Bibliotheca Zoologica ll. Verzeichnis der Schriften weber
Zoologie welche in den periodischen Werken enthalten und vom
Jahre 1861-1800 selbstandig erschienen sind; bearbeitet von Dr.
O. TascHENBERG. Lieferung 24. Leipzig (Wilhelm Engel-
mann. )—Parts 21 to 23 of this important work were noticed in
the number for June, 1921 (vol. 1, p. 519) and part 20 in the
following number (vol. 2, p. 60.) Part 24 has since been issued,
embracing pages 6313-6392 and signatures 785-794 of the Sup-
plements. It is put out at the cost of 36 marks.

It is interesting to note that the earlier issue of the Bibliotheca
Zoologica, edited by Professor V. Carus, contained a catalogue of -
zoological papers published from 1846 to 1860. The present
work, begun in 1886, was planned to cover the period from 1861
to 1880 (see the full notice in this Journal (8), 33, p. 246, 1887).
Twelve parts of some 320 pages each were contemplated and the
year 1888 set for the completion of the work. Now, however,
thirty-five years after the first part was published such has been
the thoroughness and care with which the undertaking has been
conducted that the whole up to the present time covers very
nearly 6,400 pages. This is a monumental work, indispensable
to all zoologists who can hardly fail to appreciate the magnitude
of the debt they owe to the self-sacrificing labors of the author.

2. Publications from the Ronald Press Company (20 Vesey
St., New York City).—The following volumes from the Ronald
Press have been noticed in this Journal: Practical Bank Opera-
tion, by L. H. Langston (1, 378; New York State Income Tax
Procedure, by R. H. Montgomery (1, 468); Personal Relations
in Industry, by A. M. Simons (2, 238) ; Elements of Bond Invest-
ment, by A. M. Sakolski (2, 308). Volumes recently received
are:

Miscellaneous Scientific Intelligence. 99

Wall Street Accounting; by F. 8. TopmMan. Pp. xv, 352,
1921 (Price six dollars).—This book gives a thorough, detailed
description of the accounting methods developed by leading
brokerage houses in handling their highly specialized business.
It gives, in addition, in the course of the discussion of the special
problems involved, a full and illuminating analysis of the tech-
nique of trading in the stock and commodity markets. It repre-
sents an enlargement and fundamental revision of ‘‘ Brokerage
Accounts’’ by the same author (1916).

Time Study and Job Analysis;- by W. O. LicHtNER. Pp. xvii,
397; 81 figures. 1921 (Price six dollars).—The standardiza-
tion of working methods is the subject here discussed in detail.
After presenting the methods followed in accomplishing the
result aimed at, explicit directions are given for training the
engineering staff. These are followed by charts and illustrations
giving the accepted engineering procedure for installing and
maintaining standardized methods, followed by a discussion of
standardization in its relation to production, stability of labor, ete.

The Ronald Press also conducts a monthly periodical entitled
Adminstration ($5 per year); this is described as a Journal of
Business Analysis and Control. The editors are R. B. KEstsEr,
J. M. Lee and E. H. Garpner. The scope of the periodical is
broad, the subjects treated are of vital interest to the community
as well as to the capital invested in individual enterprises, and
the writers are men of experience in practical business.

3. The ‘‘Electrician’? Diamond Jubilee Number.—The Elec-
trician, which appeared on November 9, 1861, celebrated its
Diamond Jubilee on November 11. A great change has taken
place in electrical engineering in these sixty years. Many inven-
tions and designs have been perfected making electric lighting
possible from 1880 on; later a network of supply mains has
spread through all large cities. In 1890 came the genesis of elec-
trical traction as on the City and South London Railway. This
period too has seen the invention of the telephone and the wide-
spread adoption of the electric motor for industrial purposes. The
pioneer work of Marconi, Hertz, Lodge and Fleming has culmi-
nated in the invention of commercial wireless telegraphy. Thus
the pages of the Electrician are a chronicle of the transformation
of electricity from a little understood science not only into a
ereat industry but into a means of improving our present
civilization.

The Jubilee issue gives congratulatory messages from many
eminent engineers and scientists; it also has historical and
technical accounts of the development of telegraphic and other
electrical applications since 1861.

iS Ae
>

. .
Sas

| AV aRb’s Naturat Science Estas isHMent
A Supply-House for Scientific Material.
Founded 1862. : Incorporated 1890.

A few of our recent circulars in the various
departments:

Geology : J-32. Descriptive Catalogue of a Petrographic Col- —
lection of American Rocks. J-188 and supplement.
Price-List of Rocks. |

Mineralogy: J-220. Collections. J-225. Minerals by Weight.
J-224. Autumnal Announcements.

Paleontology: J-201. Evolution of the Horse. J-199. Pale-
ozoic index fossils. J-115, Collections of Fossils.

Entomology: J-38. Supplies. J-229. Life Histories. J-230.
Live Pupe.

Zoology: J-223. Material for dissection. J-207, Dissections
of Typical Animals, etc. J-38. Models. ;

Microscope Slides: J-189. Slides of Parasites, J-29. Cata-

~. logue of Slides. :

Taxidermy: J-22. North American Birdskins. Z-31. General
Taxidermy.

Human Anatomy: -J-37. Skeletons & Models.

General; J-228. List of Circulars & Catalogues.

Ward’s Natural Science Establishment
84-102 College Ave., Rochester, N. Y., U.S. A.

6é : 99
“<SCIENTIA
INTERNATIONAL REVIEW OF SCIENTIFIC SYNTHESIS. Issued monthly (each
number consisting of 100-to 120. pages). Editor: EUGENIO RIGNANO.

This is the only review which has a really international collaboration ; which is

of world-wide circulation and occupies itself with the synthesis and unification of
knowledge, in the history of the sciences, mathematics, astronomy, geology, physics,
- ¢hemistry, biology,.psychology and sociology.
It is the only review which; by inquiries among eminent selentists and writers
~ (on: The philosophical principles of the sciences; Fundamental astronomical and physical
questions of current interest; Contributions given by the various countries to different
branches of knowledge; Question of vitalism;Social question; International questions
raised by the world war), makes a study of the problems interesting scientific and intel-
lectual cireles throughout the world. ;

It has published articles by Messrs.: Abbot-Arrhenius =Ashley = Bayliss - Beichman-
Bigourdan - Bohlin - Bohn= Bonnesen = Borel = Bouty = Bragg = Bruni = Burdick = Carver =
Caullery = Chamberlin = Charlier = Claparede = Clark = Costantin = Crommelin = Crowter =
Darwin = Delage = De Vries = Durkheim = Eddington = Edgeworth - Emery = Enriques -
Fabry - - Findlay - Fisher = Fowler = Golgi = Gregory = Harper = Hartog = Heiberg - Hinks -
Hopkins-Inigues-Innes-Janet- Kaptein-Kaye-Kidd-Langevin-Lebedew-Lloyd Morgan =
Lodge = Loisy = Lorentz - Loria -Lowell - MacBride =-"Meillet =Moret-Muir-Peano =Picard =
Poincare - Puiseux - Rabaud = Rey Pastor = Righi=-Rignano-Russell-Rutherford-Sagnac =
Sarton = Schiaparelli=- Scott = See = Sherrington = Soddy = Starling = Syedberg = Thomson =
Thorndike-Turner -Volterra-Webb-Weiss = Zeeman and more than a hundred others.

** SCIENTIA’’ publishes its articles in the language of its authors, and joins to the
principal text a supplement containing the French translations of all the articles that
are not in French. (Write for a Specimen Number to the General Secretary of
“Scientia’’, Milan.) Annual subscription : 40 sh., or 19 dollars post free.

Office : 43 Foro Bonaparte, Milan, Italy.

Publishers: WILLIAMS & NORGATE-London; FELIX ALCAN -Paris
NICOLA ZANICHELLI-Bologna; RUIZ HERMANOS-Madrid;
WILLIAMS & WILKINS CO-Baltimore.

y
‘nov. “OW ith Plate a

SOTEN TIFIO INTELLIGENCE.

Chemistry ad Physies. —New Poses for Determining Hikeny M. f
~ The Separation’ oe the Element Chlorine into. Isotopes, W. D. a
~>-and A.) HAYES; = —Qualitative Chemical Analysis, QO. F, Towne:
Course of Bee ae in Quantitative Chemical Analysis, G. Mc :
‘93.—American Chemistry, H. Hane: Report on Atomic Structure:
_ ity of Sound at High Temperatures, 94.—Traité de Dynamique, J. D’
BERT: Heat and Light, S. E. Brown, 95.—Physik und Erkenntnissthe
_ E. Grarcxe: Séries Trigonométrique, M. Lucat: Die Grundiag
: metric, L. HEFFTER, 96. Seg Sad eee

Geswou = Sak prehistoric Human Skull fronn eh wee x . Woon .D, 06. E
A new Generic Name for Pliocyon Marshi, M. R. THorpe: Publi of-
_ the United States Geological Survey, G- O; SMITH, UT eg

: ig + Miscellaneous Scientific Tnteligenen < eaetiecs deatasica, I, O. TASCHEN
_ Publications from the Ronald Press COREA ses —The Sik rician”
Prationd J ubeee Sede 99. ee tag eee ern eu x:

= % f

a Fe Fea ae SR So a SE ere
re Saat en SS Em en 9 Seon ae
RS ag ait

eCopt S

» cme U. S. Nat. Museum.

FEBRUARY, 1922

AMERICAN,
| J OURNAL OF SCIENCE,

_ Eprror: EDWARD S. DANA.

ee eS VT eae =, nt PE Se ee tae

ASSOCIATE EDITORS

-Prorassons WILLIAM M. DAVIS anp REGINALD A. oie
OF CAMBRIDGE,

|| Prormssors HORACE L. WELLS, CHARLES caer |
|| HERBERT E. GREGORY, WESLEY R. COE anp
a, _ FREDERICK E. BEACH, or New Haven,

- PROFESSOR EDWARD W. BERRY, or Bautimorgs,
- Drs. thes L. RANSOME anp WILLIAM BOWIE,
OF WasHINGTON.

ss FIFTH SERIES
8 Cs VOL. II—[WHOLE NUMBER. CCID).
xe : > * No. 14 FEBRUARY; 1929.

| |
| NEW HAVEN, CONNECTICUT.
|

1922.

THE. TUTTLE, MOREHOUSE & TAYLOR CO., PRINTERS, 123 TEMPLE STREET. |

y

> Published monthly, Six dollars per year, in’advance. $6.40 to countries in the
aa e Baaial- Union ; $6.25 to Canada. Sin; gle numbers 50 cents; No, 271, one dollar.
-_Entered as second- class matter at the Post Office at New Haven, Gonn., under the Act

Pesan,

S fi wah 6 18%.

“Just Published.

Economics of Petroleum.

By JOSEPH E. POGUE, Consulting Engineer. —

A new book of live interest to everyone in any way aoucerned with Oil_by wea" | F

a writer of wide experience in this field.
out, and offers a wide range of information of vast importance to the petroleum —
industry. The Trend of Oil-field Development and Oil Production, Eeonomie
Organization, Petroleum Prices, Petroleum By-Products, Motor Fuel Problem,
-and Mexico as a Source of Petroleum, are a few of the subjects discussed.

6 by 9. 151 figures, 150 charts, and over 100 tables.

3870 pages.
| Cloth, $6.00.

Ready early in March ,
Coal.

By ELwoop S. Moore, M.A., Ph. D,, Denti ar the School of Mines, ‘The
Pennsylvania State College.
Brings together the main facts relating to coal, for the practical man, the

It is thorough and interesting through- —

scientist and the layman.

About 450 pages. 6 by 9.

HANDBOOK FOR THE FIELD
GEOLOGIST.

By the late C. W. Hayes. Third
Edition, Thoroughly Revised and Re-
arranged by SIDNEY PAIGE.

Outlines. methods of procedure in
field work found by experience to be
best.

166 pages. 44by 63. 20 figures.

Flexibly bound, $2.50.

142 figures, 19 plates.

Cloth binding.

FIELD MAPPING FOR THE OIL
GEOLOGIST.

“By C. A. WARNER.

A handbook of field methods of
value to those geologists who have had
littleexperience with the methods com- -
monly used in examining undr lee
territory.

148 pages. 44 by 7. 38 figures.
Flexibly Ge $2.90.

Any Wiley book may be had on Free Kxamination—write TO-DAY,

JOHN WILEY & SONS, Inc.

432 Fourth Avenue, New York

AJS 2.92

THE

AMERICAN JOURNAL OF SCIENCE

eee Sen ES:

oe

Art. 1V.—Gastropod Trails im Pennsylvaman Sand-
stones in Texas; by Sipney Powers.

During a brief examination of the Pennsylvanian
section near Shafter, western Texas, in 1919, the tracks
here described were observed on the eastern side of Cibolo
Creek. Dr. J. A. Udden reported on the geology of the
region in 1904 and described the tracks as those of trilo-
bites (?).1 In the same report he described conglomerates
which he believed to be of:glacial origin. The drawings
of the tracks (figs. 1-3) have been made with great care
by Mr. G. S. Barkentin, of Albany, N. Y. The following
analysis of their origin has been made possible through
the careful study of Professor P. H. Raymond who is con-
tributing an accompanying paper on trails. The speci-
mens were given to the Museum of Comparative Geology.

Shafter, a silver mining camp situated in the canyon of
Cibolo Creek, 19 miles north of Presidio on the Mexican
border, is shown on the Shafter topographic sheet of the
U.S. Geological Survey. It is in the midst of the Trans-
Pecos volcanic plateau and the exposures of Cretaceous
and Pennsylvanian sedimentary rocks are due to block-
faulting followed by deep dissection. The silver is largely
chloride and occurs as fissure and contact deposits in the
Cibolo limestone of Pennsylvanian age. The country
rock runs 1 to 3 ounees to the ton, average ore 10 0z., good
ore 80 to 90 oz.

The geologic column is as follows:

Phcintoeene ee = oa. AS os SS Bolson deposits.

EN EU REIT a2 oe Ad? SS eee Volcanics.

Upper Cretaceous............. Voleanics, granitic intrusions ( ?)
Lower Cretaceous ............ Limestones.

* The geology of the Shafter silver mines district, Presidio County, Texas,
University of Texas Mineral Survey, Bull. 8, p. 60, 1904.

Am. Jour. Sct.—Firta Serizs, Vou. IIJ, No. 14.—Frsruary, 1922.
8

102 §. Powers—Gastropod Trails in Sandstones.

Permian Ana te eco ee ee ee Cibolo limestone.
Transition beds.
Alta sandstone.

) Alta shale.

| Cieneguita beds.

(Intrusive contact)

The correlation of the Pennsylvanian section is with
the Tesnus, Dimple, Haymond, and Gaptank formations
of the Marathon uplift, with the Strawn formation (?)
of Northern-central Texas, and with part of the Hueco
limestone of western Texas. The Cibolo limestone is
correlated with the Capitan limestone on the New Mexico
line.?

Exposures of Pennsylvanian rocks are limited to six
localities, but further investigation will doubtless reveal
others. Two of these have not been verified. The prin-
cipal exposures are north of Shafter, one ‘‘east and
southeast of the Ojo Bonito and the Cieneguita ranches,’’
and one along Sierra Alta creek east of Cibolo ranch. An
exposure only a few hundred feet long is found on the
southernmost Morita creek one mile southwest of the
Morita road, and a short distance northeast of a mine
tunnel. Two exposures are reported below the box
canyon south of the junction of Cibolo and Morita creeks,
one above the site of the Spunk ranch, and above the site
of the Dutchover ranch. Another exposure is mapped
by B. F. Hill on the western side of the Chinati Moun-
tains. Pennsylvanian rocks also outcrop southeast of
Shafter in the Solhtario dome and in Mexico where they
have been found by Dr. Udden.

Northeast of the present Cieneguita ranch and east
of the voleanics there is an area of black shales and
quartz-pebble conglomerates striking north-south and eut
off on the northeast by fine-grained, porphyritic, gray
granite, the femic constituents of which are pyroxene
and biotite. There is little metamorphism at the contact,
but the igneous rock has invaded the shales in the form
of tongues and fragments of the shale are included in the
granite. The granite at the contact is of moderate grain
with pink feldspar and colorless quartz crystals of equal

2 J. A. Udden et al., Review of the geology of pesas; Univ. of Texas, Bull.
44, p. 43, 1916.

S. Powers—Gastropod Trails mm Sandstones. 108

size. Some distance east of this outcrop Dr. Udden
found exposures of the same Cieneguita series with
pebbles of limestone and quartz ranging ‘‘in sizes up to
three inches in diameter... . faceted in a manner similar
to the faceting of small boulders and pebbles in glacial
deposits’’ (loc. cit. p. 14). Where observed by the writer
in another locality the conglomerate showed well rounded.
quartz pebbles 14 to 14 inch in diameter which indicated
long transportation by running water. Numerous fos-
sils are reported from this formation by Dr. Udden.

In Sierra Alta creek the upper beds of the series are
well exposed. An additional outcrop was recently noted
by Dr. Udden where the new road to Cieneguita ranch
climbs out of Cibolo creek. East of the creek along this
road there is an excellent exposure of Alta shales with
interbedded sandstone strata one inch to four inches
thick. The upper surfaces of the sandstones are covered
with trails like those here illustrated and described.

Trails were found by Dr. Udden in a small limestone
ledge on the northern side of Sierra Alta hill (a conspic-
uous voleanic butte unnamed on topographic sheet)
associated with a Productus (near cora), a Chonetes,
and a Pleurotomaria. They were found by the writer at
the Morita creek exposure not far east of an old prospect
tunnel. No shells are associated with the tracks, but fine
hmestone conglomerates about 20 feet lower in the section
at the last locality contain numerous bryozoa and small
tabulate corals.

Description of the Tratls.

There are two distinct kinds of trails: a simple type
about 14 inch wide consisting of crescentic indentations 6
to 10 in a linear inch, and 1/20 inch deep (fig. 2); a com-
pound type 34 inch to 114 inches wide consisting of a
central groove from 14 to 5/16 inch wide with elevated
looped concentric ridges 1/20 inch high and 14 to 1% inch
wide on either side (figs. 1 and 3). The central groove
in the second type appears to be similar to the simple
tracks where it is clearly shown. These two types never
appear on the same surface and only the first type was
observed at Morita creek. The compound type is con-
fined to sandstones whose surface is covered by minute

104 S. Powers—Gastropod Trails im Sandstones.

phe ae
N : : :
i Aly
{ BS y
} en lM te
aS : Wii “ i « 2 ee
iN ee 3 <

NG a
~ ce lag Ss

)
:
.

, =<;
= \
soe

Fie. 1.—A slab of sandstone about 14 inches long and 8.5 inches wide,
showing the characteristics of the wide trails, and their frequency. A little
less than one half natural size. The total length of the scale is one inch.

S. Powers—Gastropod Trails in Sandstones. 105

flakes of mica. Therefore, it is believed that both types
of trails were made by the same animal and that the pre-
servation of the complete markings is dependent upon the
character of the surface over which the animal travelled.
The essential means of locomotion was a single foot
which pressed deeply into soft mud and lightly into hard,
mica-covered sand. ‘The concentric ridges on either side
of the central groove in the compound trails are evidently

Fig. 2.—Some trails of the narrow type. The depressions between the
septa-like ridges are very deep. Compare with the central part of the trail
shown in fig. 1. Natural size.

the impression of the shell or body of the animal on either
side of the foot and the preservation of these delicate
markings was dependent upon a relatively firm surface.
In the case of mud these markings were elevated above
the central groove and the slightest amount of erosion
would remove them while burying the groove. If this is
not the correct explanation for the two ‘kinds of trails the
fact that certain micaceous surfaces show only the com-

106 S. Powers—Gastropod Trails im Sandstones.

pound type while all other surfaces within a few inches
both above and below in the section show simple trails is _
- difficult to account for.

The frequency and curvature of the trails are best
understood by referring to the illustration. One slab
shows 26 distinct trails in 56 sq. inches or one trail to 2 sq.
inches. The trails bend sharply, cross, and occasionally
combine, but there is no indication that the animals had
any common trail or that they crawled over a former

Fic. 3—Short portions of two very well preserved trails showing the
lobate character of the sand which was pushed aside during the advance of
the animal. One-fourth less than natural size.

trail except by accident. The frequent turns im the trails
indicate that the animals were not in the habit of pro-
gressing in a straight line and therefore it is inferred
that their power of vision was extremely limited. Be-
cause there is no indication of common centers from
which tracks lead and because the tracks end suddenly
it is evident that the animals burrowed into the sand or
mud and that the wet mud concealed the places where
they disappeared and reappeared.

Professor Raymond has examined trails on beaches
near Boston with reference to the origin of these tracks
and he finds that the common gastropod Littorina prac-
tically duplicates them except that the convex ridges
along center are not so steep as in some of the fossil

S. Powers—Gastropod Trails i Sandstones. 107

trails. The ridges in the modern trails point forward
and may be present or absent according to the firmness of
the sand. Professor Raymond finds that both the direc-
- tion and the preservation of the modern trails is depend-
ent upon the condition of moisture on the beach as well as
upon the composition of the beach surface. The gastro-
pods make the trails when the beach is out of water and
early drying is necessary for preservation. In addition
to the positive evidence of the gastropodous origin Pro-
fessor Raymond notes that the eccentric winding course
is very different from the straight trails made by
annelids. It is therefore concluded that both types of
the Shafter trails were made by gastropods and that the
difference between these trails was dependent upon physi-
eal conditions rather than upon different species of
gastropods.

108-© P. E. Raymond—Seaside Notes.

_ Arr. V.—Seaside Notes; by Purcy HK. Raymonp.

The logical place for the student of invertebrate fossils
during the summer is at the seashore, but unfortunately
most of us find ourselves occupied in the interior of the
country when we leave our laboratories. Many things
which have a geological or paleontological bearing are
happening at the present moment, and it is the purpose ot
this note to record two observations which seem to have a
certain significance.

Preservation of Jelly-fishes.

I happened in the early summer of 1919'to be on the
sandy outer beach at Duxbury, Mass., on a windy day
shortly after a storm. <A large number of the common
jelly-fish, Staurophora, had been driven inshore and when
I arrived had been left stranded for some hours by the
retreating tide. A stiff land breeze was drifting sand
down the beach in considerable quantities, and much of
it found lodgment in the jelly-fishes. The sand so
caught adhered to the jelly-fish, which had lost so much
water by evaporation that its thickness was less than one
millimeter. Although the specimens had shrunk con-
siderably, their diameters averaging from 90 to 100 mm., —
most of them retained a nearly circular outline. The
sand varied from fine to coarse, and one specimen which
I preserved has in it a flat pebble 14 mm. across.

I expected to find these specimens very fragile but had
no difficulty in lifting them off the sand by taking hold of
one edge and peeling them off slowly. J spread four of
them flat in a newspaper and placed them under the
back seat of the car. Three weeks later, on removing
them, they had not changed in any respect. . They were
then laid on a laboratory table, still in the newspaper,
and a book placed on them. At the end of three months
they were in the same condition as when collected. They
were still limp and flexible and could be lifted by the edge
without pulling apart.

At this time I treated them lightly with shellac, which
caused a certain amount of wrinkling and contraction,
and has made them brittle. The specimens with the finest
sand are of course the best preserved, but all still show

P. E. Raymond—Seaside Notes. 109

more or less prominently the characteristic cross and the
outline of the velum.

These observations suggest a method by which jelly-
fish and other soft-bodied animals may at times have been
preserved as fossils. The gelatinous material of the
body appears to cement the grains of sand together, and
at the same time, the mass of the latter checks the shrink-
age. If particles of clay instead of sand covered the ani-
mal, the preservation would be much more perfect, and
the extreme slowness of decomposition shown by the
present specimens would indicate ample time for burial.

Agassiz' long ago made this same observation in regard
to the possible preservation of jelly-fish in sand. It is
not, however, probable that permanent fossils could be
formed in this way under the conditions which prevail on
modern sandy beaches. In early Paleozoic times there
were, however, extensive flats with aluminous or ecalea-
reous muds which seem to have been inundated periodi-
cally, perhaps in periods of persistent onshore winds, and
then subjected to long periods of desiccation. These muds
became thoroughly dried out and cemented between
inundations, and conditions were optimum for preserva-
tion in the manner indicated above.

Trails.

Among the more common fossils are the little under-
stood trails which are so frequently seen on the surfaces
of layers of sandstone and shale. It has long been the
custom to refer to them as ‘‘worm trails,’’ the idea having
originated apparently from the fact that the early inves-
tigators (see particularly MacLeay in Murchison’s Silu-
rian System, 1859, p. 699 et seq.) considered them to be
the actual impressions of Annelida. Nathorst long ago
pointed out that many of them must have been made
imnorganically, or by Crustacea and Mollusca (KX. Svensk.
Vetensk. Akad. Handl., vol. 18, p. 66, 1881) and a visit
to the seashore should be sufficient to convince any pale-
ontologist that he was correct.

Winding, irregular trails are ascribed to worms,
because they look like elongate chetopods, or masses of
their castings. A little reflection, however, will show
that it is just these trails, with their sharp turnings,

* Contributions to Natural History of U.S. of America, 1862, vol. 4, p. 63.

110 P. E. Raymond—Seaside Notes.

which must have been made by short and not elongated
animals. Investigations on the sandy beaches during the
last few years have shown that whenever a trail-making
animal can be identified, it invariably turns out to be
something other thana worm. One of the most tortuous
of the worm-like trails, seen usually in the troughs of rip-
ple marks, is made by a small amphipod which travels
about just ‘beneath the surface of the sand.

One frequently sees on the mud by the roadside trails
of the common earthworms. These are usually nearly
straight, or in curves of long radius. Doubtless marine
chetopods may come to the surface and crawl along, but
I have not seen such trails on the beaches near Boston.

The most common trail seen in this vicinity is that
made by the omnipresent gastropod, Littorina ltorea.
These animals live on those parts of the beach where there
are rocks or other substances to which to cling, and
during the time of ebbing of the tide, many of them crawl
across the wet sand. The trail which they leave is
generally a little wider than the transverse diameter of
the aperture of the shell, is bounded by a pair of lateral
ridges, and marked transversely by ridges which are
bowed forward in the direction in which the animal
travels. In creeping along, the outer lower part of the
lip is directed forward and the spire points backward to
the right. When moving forward the sand is pushed up
in a little ridge ahead of the animal, then the shell is
raised somewhat and pushed over the mound in front, so
that the ridge is preserved even after the Littorina has
crawled over it.

The trails vary a great deal according to the firmness
of the sand. When the animal travels through water the
lateral ridges are usually high and broad and the trans-
verse ones are not seen at all. On very firm sand neither
set of ridges is made, so that there is much change in the
character of the trail from one point to another as it
leads over the various inequalities of the rocky and
rippled beach.

The course in the case of Littorina is usually quite
straight except for deviations due to the various obstacles
encountered, but the animals do sometimes wander rather
aimlessly and make a considerably curved trail. One
marking made by a Littorina was 13 feet long, and, to the

P. E. Raymond—Seaside Notes. stet

outside of the lateral ridges, averaged from 11 to 12 mm.
in width, except on firm sand, where it was 10 mm. The
shell of the animal which made it measured 10 mm. across
the aperture. There were six transverse bars in 33 mm.
Another trail was 14 mm. wide, 2.50 mm. deep, and had
14 transverse ridges in 80 mm. A good photograph by
Miss Dickerson of Littorina litorea and its trails may be
seen in the American Museum Journal, vol. 16, p. 374,
1916.

Lunatia heros is another gastropod which makes con-
spicuous trails on the sandy beaches. Being a large shell
it makes a broad track very similar to that of Littorina,
but with much more obscure transverse ridges. The
latter characteristic is probably due to the enormous size
of the foot. This animal is in the habit of burying itself
in the sand, and plowing along just beneath the surface.
A trail ean often be seen ending abruptly, with a little
heap of sand at the end. Digging away the sand will
reveal the Lunatia just beneath the surface. Such mark-
ings remind one very much of Climacticnites, and support
Woodworth’s contention that the famous Cambrian trail
was made by a large Mollusk.? Climacticnites is essen-
tially the same sort of trail as that made by Littorina,
and since the transverse ridges point forward in the
latter, Woodworth and Walcott were probably right in
their interpretations of the direction of motion of the
animal which made the former. The oval depressions
at the end of the trail of Climactienites seem to me
not merely the place where it rested for a time. The
specimen from the Potsdam at Moers, N. Y., now in the
Museum of Comparative Zoology shows that the terminal
impression was not a simple one, but there are two or
three depressions at the end of each trail, all more or less
irregular, and in one ease the sand is piled up at one end,
in another it has slumped at one side just as the sand
appears after a Lunatia has buried itself,

The specimens of Climacticmtes young: figured by
Walcott? also support the interpretation as the trail of a
gastropod. These beautifully preserved specimens not
only show the forward-directed transverse ridges pro-
duced by the foot, but also a series of very regular, closely

7\N. Y. State Mus., Bull. 69, 1903, p. 959.
* Smithson. Misel. Colls., 1912, vol. 57, p. 259, pl. 38, fig. 1, pl. 39, fig. 2.

112 P. H. Raymond—Seaside Notes.

spaced curved lines such as might have been made by the
frontal margin of the anterior part of the foot (propo-
dium) which is most strongly developed in those gas-
tropods which crawl on wet sand, and is apparently
pushed ahead to smooth out the path of the animal.

Nicholson and Etheridge* figured the trail made by a
recent Purpura, and called attention to the fact that it
-greatly resembled some of the markings ascribed to
annelids. In spite of the evident similarity of this trail
to those generally called Nereites, and which are so com-
mon in the early Paleozoic rocks of EKurope and North
~ America, Nicholson still referred the latter to the Anne-
lida. Nereites is a trail with exceedingly numerous and
abrupt turnings, and how an elongate animal like a che-
topod could drag itself along and leave so sharp and clean
cut an impression is difficult to understand. A short gas-
tropod, hitching itself along, and wandering back and
forth in search of food, would make just such a trail.

Hanecock® was among the first to doubt the vermiceous
origin of markings of this nature, and was led by observa-
tions on the beaches of Durham, Eing., to ascribe them to
the activities of Amphipoda. Trails like that shown on
pl. 14, fig. 1 of Hancock’s paper are very common in the
hollows between ripples on the beaches about Boston, and
are made by small Amphipods. On plate 17 of the same
article is represented a trail very like the smaller form
described by Dr. Powers in the article accompanying this
one. This nodular type of trail was explained by Han-
cock as due to the work of an amphipod or other crusta-
cean burrowing its way, with frequent periods of rest,
through a rather tenacious mud which was strong enough
to prevent collapse of the tunnel after the animal had
passed through. The nodular character he conceived of
as being produced by a deviation of the animal from the
horizontal plane, rising and sinking a little at each
advance. ‘T'he specimens which he described seem quite
perfect, and do not show any trace of lateral ridges, and
hence differ from the trails of gastropods.

Mon. Silurian Foss. Girvan Dist., fase. III; reproduced in Nicholson &

Lydekker, Manual of Palaeontology, ed. 3, 1889, vol. 1, p. 489, fig. 351.
> Am. Mag. Nat. Hist., ser. 3, vol. 2, p. 443, 1858.

P. E. Raymond—Seaside Notes. fel

Trails in the Pre-Cambrian.

The proper recognition of the origin of trails is of par-
ticular importance in the case of Pre-Cambrian strata.

The following have been described as trails of worms
by Waleott.6 All were obtained from the Beltian of
Montana.

Helminthordichmtes? nethartensis Walcott was de-
scribed as the trail of a slender worm, a minute mollusk,
oracrustacean. Walcott grouped this specimen with the
worms, but evidently rather favored its interpretation
as a mollusk, for he stated: ‘‘From the convolutions
shown on the upper portions of figures 2 and 4 (pl. 24)
the impression is given that the animal moved very much
as a small mollusk does when wandering about on the
mud at low tide.’’ This view is the one shared by the
present writer, and since mollusea other than gastropoda
do not appear till very late in the Cambrian, it seems
probable that this trail was made by some kind of a snail.

Helnunthoidichnites? spiralis Walcott has the spiral
form of a watch-spring and while it reveals no very con-
elusive evidence of its origin, is very likely organic, and
if so, was probably formed by a small gastropod, since
it les entirely upon the surface of the layer, and is too
smoothly coiled to have been made by an annelid in
motion. |
_ Helminthoidichmutes meeki Walcott is from 1.50 to
2.00 mm. in width, and is notable for its remarkably
symmetrical curves. It is not a furrow or ridge, but a
plane marking such'as might have been made by a gas-
tropod crawling along and leaving a path of mucous
behind it.

Planolites corrugatus Waleott was described as the
east of a burrowing worm, but is more likely a burrow.
It appears to be 4 or 5 mm. in diameter, and along a part
of it the cast is annulated, suggesting the former occu-
panecy of a segmented worm. Such annulations are not,
however, due directly to the segmentation of the animal
which made them, but as an annelid eats its way through
the mud, the material passes through the body and fills
the tunnel behind, so that each segment of the fossil cor-
responds to a hitch forward, rather than to a portion of
the body.

° Bull. Geol. Soc. Am., vol. 10, pp. 236, 237, pl. 24, 1899.

114 P. EK. Raymond—Seaside Notes.

Planolites superbus Walcott may or may not be
organic. It has somewhat the appearance of the cast of
a tunnel parallel to the bedding and if so, may have been
made by a gastropod, a crustacean, or possibly, an
annelid.

Only one of these Pre-Cambrian trails presents any
appearance which would seem to entitle it to be called the
burrow of aworm. ‘The others are more probably either
of a gastropodean origin, or are entirely inorganic.

The Preservation of Trails.

After studying the trails on the beaches it becomes evi-
dent that to allow their preservation somewhat unusual
conditions must prevail. ‘T'rails made during the ebbing ©
of the tide are completely obliterated in the flow,
and those made in shallow water are destroyed by the
motions of currents and waves. ‘To insure preservation,
it appears to be necessary that the mud shall contain suf-
ficient cement, in the way of calcium carbonate, oxides of
iron, hydrous silica, or finely divided clay, to consolidate
the surface quickly. An example of the latter process
was noted recently. A fresh bird-track was noted in mud
which was not yet dry at noon on Monday. Tuesday
noon it was dry and firm. ‘Tuesday night there was a
hard rain, and Wednesday noon the track was still visible,
covered with about two inches of water and partially
filled with fine mud. It had not lost its clear outlines and
enough cementation had taken place during the drying
process to allow preservation.

Although it is possible that such action may take place
between tides, as indicated by Agassiz’s oft-quoted obser-
vations on the calcareous mud of Florida, it does not
occur in ordinary sediments. The flood plains of deltas,
and the playa deposits of arid regions present ideal con-
ditions, but such localities are not inhabited by marine
animals. ‘T’o preserve the trails of the latter requires
the postulation of some such processes as are outlined
above in connection with the jelly-fishes, and the abun-
dance of trails in formations of all ages would seem to
- indicate the presence of marine playas throughout

geologic time. 7

Geological Museum,
Cambridge, Mass.

T. W. E. David—‘Varve Shales’’ of Australia. 115

Anr. Vi—The “‘Varve Shales’? of Australia*; -by
T. W. EpvcewortH Davin, Professor of Geology and
Physical Geography, University of Sydney.

1. Varve Shales of Lower Carboniferous Age.

Varve shales interstratified with beds containing the
Carboniferous fossil ferns Cardiopteris and Rhacopteris
have recently been described by Mr. Sussmilch and the
author.t They occur chiefly at two horizons separated by
100 feet to over 600 feet of strata.

The specimen here exhibited is a laminated shale—
lamine of deep brick-red alternating with thinner lamine
of pink to hght grey tint. It exhibits beautiful examples
of minute contemporaneous contortions, aifecting inter-
mediate laminz, but leaving those above and below the
contorted zone quite undisturbed. ‘The shales are inter-
stratified with tillites, and glacial conglomerates and
glaciated pebbles occur in turn. It is considered, there-
fore, that they are genuine Paleozoic ‘‘varves’’ as
defined by Professor Gerard de Geer,” and described by
Robert W. Sayles.*

The paired lamine are very conspicuous and on the
assumption that each pair of lamine has an average thick-
ness of two-tenths of an inch and that each pair repre-
sents one year’s deposition of sediment, the authors have
roughly counted the number of pairs in such sections as
are available and estimated that the aggregate of 220 feet
of varve shale required a period for deposition of at least
4,000 years. Thisis a minimum estimate. For example,
in the specimen exhibited, which is 84 millimeters thick,
there are twelve pairs of lamine. If this were the aver-
age thickness of the paired lamine the time needed for
their formation would be nearer 10,000 years.

The authors ascribe the contemporaneous contortions
in these ‘‘varves’’ to the movement of ice in some form
tending to develop horizontal gliding planes.

* Read at the First Pan-Pacifie Scientific Conference at Honolulu, August,
1920.

* Sussmilch, C. A. and David, T. W. Edgeworth, Carboniferous and Permo-
Carboniferous rocks, New South Wales, Journ. Roy. Soc., New South Wales,
vol. 53, pp. 270-273, 1919. ;

7 de Geer, Gerard, A geochronology of the last 12,000 years, Compt. Rend.
Cong. Geol. Intern. Sess. II, pp. 241-253, 2 plates, 1910, 1912.

* Sayles, Robert W., The Squantum tillite member of the Roxbury con-
glomerate series. Memoirs, Mus. Comp. Zool. Harvard, vol. 47, No 1, 1919.

116 T. W. E. David—‘Varve Shales’’ of Australia.

A fragment of the middle Carboniferous plant Cardi-
opteris is preserved at the top of the specimen exhibited
and it is hoped that when these varves are examined in
detail they will throw some light on middle Carboniferous
chronology.

2. Varve shales of late Proterozoic or Lower Cambrian Age.

The specimens of laminated rock exhibited were
obtained by Mr. EK. C. Andrews and Mr. W. H. Browne at
Campbell’s Creek near Poolamacea, north of the Barrier
mines at Broken Hill, New South Wales. They belong to
the horizon of ‘Tapley’s Hill Shale,’’ so named from the
type locality near-Adelaide, South Australia.t These
shales immediately overlie the masses of tilliite described
by Professor Howchin as aggregating some 1,000 feet in
thickness. There can be little doubt that they are
‘‘varve’’ shales, and that the lamine indicate seasonal
deposition. One specimen shows marked contempora-
neous contortion. The paired bands have a maximum
thickness of about one inch and their total thickness is
about 1,000 feet. This gives a minimum of 12,000 years
as the time needed for the accumulation of the Tapley
Hill Shales.

* Howchin, Walter, The geology of South Australia, pp. 344, 362.

Washington and Merwin—Hawauan Augite. 117

Arr. VII—Mineralogy. Augite of Haleakala, Mau,
Hawaiian Islands;| by .Hmnry S. Wasuineton and
H. HE. Merwin, Geophysical Laboratory, Carnegie Insti-
tution of Washington.

The Hawaiian lavas are, on the whole, of very simple
mineral composition,” but we know little of the chemical
characters of the minerals that compose them. Augite
and olivine constitute almost the only mafic minerals; the
orthorhombic pyroxenes, amphiboles, and micas being
seldom present in the lavas. Of the olivines we have one
analysis,® that of one from a flow of Mauna Loa, Hawaii;
and M. Aurousseau, of this Laboratory, is now studying
others. But no analysis has been published of any augite
from the islands, although this mineral is more constant
and abundant in the lavas than is olivine, which fre-
quently occurs as large phenocrysts and is therefore
more prominent. Some knowledge of the general compo-
sition of the Hawaiian augite is consequently of consider-
able importance for the study of the petrology of the
Hawaiian Islands, so it is a pleasure to have the oppor-
tunity to determine the optical and chemical data for a
Hawaiian augite. The augite crystals described here
were collected in September, 1920, by Dr. J. Allan Thom-
son, of Wellington, New Zealand, who very kindly placed
them at our disposal for study, a courtesy for which we
extend our hearty thanks.

The augite crystals were found ‘‘along the trail from
the Rest House to Red Hill,’’? a small parasitic cone at
the southwest corner of the rim of the great crater of
Haleakala, on the island of Maui. The crystals are
unquestionably some of those that are mentioned by
Cross* and other writers as abundant in this locality, and
are probably derived from a lava that is perhaps identical
with one which is called picritic basalt by Cross, who
gives an analysis by Steiger.

The crystals vary from about one half to one centi-
meter in length, are of a shining jet-black color, and of

? Received November, 1921.

* Cf. W. Cross, U. 8. Geol. Survey, Prof. Paper 88, 1915.

* Analysis by G. Steiger, in Daly, Jour. Geol., 19, 295, 1911.
*'W. Cross, U. 8. Geol. Survey, Prof. Paper 88, p. 28, 1915.

Am. Jour. Sci.—FirtH Srriss, Vou. III, No. 14.—Frprvuary, 1922.
]

118 H. 8. Washington é& H. EK. Merwin—Mineralogy.

the simple habit which is usually shown by such loose
crystals of augite from basaltic rocks. The planes pres-
ent are: a(100), b(010), m(110), and s(111). Some of
the crystals are twinned on the front pinacoid (100).
The crystals were measured goniometrically, but subse-
quent sectioning showed the presence of a surface film,
about 0.05 mm. thick, having a much higher refractive
index, stronger pleochroism, and other properties, indi-
eating a higher ferric iron content than the rest of the
erystal, so that the goniometric measurements cannot be
of definite value, and they are therefore not given.

In thin section the color is a very pale gray, almost col-
orless, but with a faint tinge of green; the pleochroism is
so slight as to be scarcely noticeable. Extinction angles
were measured on two sections parallel to the side pina-
eoid (010), cut centrally through two crystals that were
twinned on a(100). The angle y c= 47°-48° for red
(630 wz), and 49° for blue (480 wz). For the same wave
lengths 2V, measured 61°-62° and 2V, =—958°-60°. The
birefringence y—a on these sections was .024; B—a (caleu-
lated) is .006.

Measurements of refractive indices were made on a
sample of the powder prepared for analysis. The lowest
value found was 1.695, the highest was 1.727; 8 was
1.704-1.709. Thus, the following indices represent the
material used for the chemical analysis; «1.700, 6B =
HUG. 7 —— ee i

The density was kindly determined by Dr. L. H.
Adams, using a pycnometer and thermostat, on the mate-
rial used for the analysis. The value obtained was 3.358.

A chemical analysis was made of material carefully
separated by heavy solutions and the electromagnet, that
- was practically free from the outer film and from glass
and other inclusions. The powder was dried at 110°.
The results are shown in the table of analyses, with
analyses of other augites from basalts for comparison.
The analysis does not differ materially from those of,
other augites from basaltic lavas, but the presence of
about one-quarter of one per cent of chromic oxide is
of interest. Steiger found 0.18 per cent of Cr,O, in the
olivine of Mauna Loa, and several analyses of various
Hawaiian lavas show that this constituent is present in
many of them in readily determinable amounts.

il 2 reat
IQR es .. 47.70 47.06 50.0 Matin 50.94. Se re
SS 6.82 yin a1 3.89
PE Oro «.-siraies 2 3.36 1.30 1.47 2.05
1:29 Sanaa 4.43 8.15 4.96 7.41
Oy oo se Fe 13.34 13.52 14.01 14.59
Ga =. 64.0 21.30 19.33 22.48 20.34
Nas Oo: oer 0.65 0.33 0.73 0.61
(SG 6 erate OPES 0.03 0.11 0.01 0.18
EOS re. 0.15 0.20 0.22 0.08
nO FANS. 1.89 1.82 2.11 0.96
DESO: BLE 584, 0.23 trace n. d. Oe
RO $9). 4579s 0.16 0.20 0.21 0.10
100.11 99.85 100.00 100.71
Density .... 3.398 3.366 3.236
1. Augite, near Red Hill, Haleakala, Maui, Hawaiian Islands, Washington

analyst.

2. Augite, near Grant’s, Mount Taylor Region, New Mexico, Chatard
analyst. J.S. Diller, U.S. Geol. Survey, Bull. 591, p. 149, 1915.

3. Augite, Monti Rossi, Eruption of 1669, Etna, Sicily, Washington analyst.
Washington and Merwin, this Journal, 1, 29, 1921. Corrected for 4
per cent of magnetite.

4. Augite, Il Liscione, Stromboli. Washington analyst, Kozu and Wash-
ington, this Journal, 45, 467, 1918. Contains 0.08 SrO.

When we attempt the interpretation of the analysis in
terms of mineral molecules we are confronted with the
question of the disposition of the sesquioxides, that is,
the alumina and ferric oxide above the amounts needed to
form acmite or possibly jadeite molecules. This is the
erux in the interpretation of all aluminous augites, and it
is usually met, it is scarcely necessary to say, by the
assumption of the existence of the so-called Tschermak
molecule (Mg, Fe)O.(Al, Fe),O3.Si0,. This is not the
place for a proper discussion of this vexed matter, which
must be postponed to a later occasion when the general
subject of the composition of the pyroxenes is taken up.
It may, however, be said here that our studies so far have
led us to disbelief in the existence of the Tschermak mole-
cule, a conclusion in which we are at one with Boeke’ and
Zambonini.® We cannot, however, agree with Boeke’s
conclusion that ‘‘the aluminous monoclinic augite is

5H. E. Boeke, Zs. Kryst., 53, 445, 1914.
°F. Zambonini, Atti Acc. Sci. Napoli, 16, No. 2, p. 9, 1914.

120 H. 8S. Washington € H. FE. Merwan—Mineralogy.

essentially a mix-crystal of the components Si0,, CaO,
(Mg, Fe)O, and (Al, Fe),0,’’ whose saturation bounda-
ries are defined according to his tetrahedral projection.
According to this view the aluminous augites would differ
from the other pyroxenes in the lack of stoichiometric
relations or the presence of definite molecules, and would
be regarded as indefinite mixtures. Zambonini assumes
the presence in the aluminous augites of three general
molecules, diopside-hedenbergite (with their compo-
nents), acmite-jadeite, and spinel. While we agree with
him that the augites are best regarded as made up for the
most part or almost wholly of the first two mineral mole-
cule groups, yet there are serious difficulties in the way
of assuming the presence of a spinel molecule to account
for the presence of the sesquioxides. The most important
of these, and the only one to be mentioned here, is that
combination of the basic RO needed for the spinel would
subtract just that amount from the bases needed to sat-
isfy the silica in order to conform to the metasilicate
ratio and would leave unsatisfied the equivalent amount
of silica. This is especially true of the best analyses of
the augites,’ and would seem to be an insuperable objec-
tion to Gambonini’s view of the composition of the
aluminous augites.

For various reasons, which it is not necessary to dis-
cuss fully here, we assume that, in generai, the alumina
and ferric oxide (above that needed for acmite-jadeite
molecules) are present as such in solid solution with
diopside-hedenbergite (with or without clinoenstatite)
and acmite-jadeite. As was pointed out many years ago
by Piccini,® the T’schermak molecule is equivalent to one
of (Mg, Fe)O.S10, plus one of alumina or ferric oxide,
that is, RSi0, + B.O;. It should furthermore be noted
that the assumption of the T'schermak molecule binds an
amount of (Mg, F’e)O, equal to that of the (Al, Fe).O.,
which would otherwise enter into the diopside molecule;
this tends to the formation of molecules of wollastonite,

“It may be pointed out that a very considerable number of the analyses of
augite used by Boeke ‘(and also by Zambonini), are of very poor quality,
either because of imcompleteness (as regards titanium and soda especially),
or because of inaccuracy in the execution (such as Doelter’s analyses of
Cape Verde augites). A much more critical and exacting selection of
analyses is necessary for study of the problem.

* A. Piccini, Trans. Acad. Lineei, (3), 4, 224, 1880. Cf. Zambonini, op.
Cit. 9p: 0:

Augite of Haleakala, Hawauan Islands. 121

a mineral which must be regarded as but doubtfully
pyroxenic, and which does not appear in the recasting of
the best analyses of the non-aluminous pyroxenes.

The molecular composition of the Haleakala augite, on
the basis assumed by us, and with titanium dioxide reck-
oned with silica, is as follows:

i

Ca Ma On a Sith x. cil doc bal 3 69.12
COUINENSTL Or, Cat ont Sn rear a ream se A5eks
INE SELEIS TAG ie ae eer me 5.08
UM ees) eee ty cone re ey fo Se ene 1.90
FesiO, So Rae Ar UALS SO iG. 0.40
Cae Oo c. ce oe ya eee 8.65
100.28

The 8i0, + TiO, demanded by the bases on this inter-
pretation is (molecularly) 0.828 or 49.68 per cent (reck-
oned as silica), while there was actually found 0.819 or
47.70 SiO, + 1.89 TiO,, that is 49.59 per cent, a fairly sat-
isfactory agreement. The augite is thus essentially a
hedenbergite- diopside, with small amounts of acmite,
clinoenstatite, and alumina in solid solution.

It may be instructive to give the molecular composition
as caleulated on the basis of Zambonini’s assumption,
that the Al,O, is present as a spinel (RO.R,O,) in solid
solution. The composition on this basis is as follows:

CTS OF erect tte A as re, 58.97
(ORD DYESS 12 09a ge ei: Seed 9 eee 12.65
NaHesi One. fare wick out tees 5.08
SEIS TN EREEGS Cae Rg Sian ht ein ea eGo!
(Me, Fe) 0. GAPE eC) J Oe es 12°23
Me GEXCESG ers tc ok eed, & 4.20

This interpretation is clearly not as satisfactory as the
preceding one, chiefly because of the presence of excess
silica, and partly because of the presence of the doubt-
fully pyroxeniec wollastonite molecule instead of the
pur ely pyroxenic clinoenstatite molecule. The presence
of both of these somewhat exotic molecules is directly
brought about by the assumption of a spinel molecule
rather than free alumina and ferric oxide in solid solution.

In calculating the mode from the norm in rock classi-
fication, it has been found to be very convenient to caleu-
late the norms of the mafic minerals present in the rock
(augite, hornblende, or biotite), the process being the

122 H. 8. Washington & H. E. Merwin—Mweralogy.

same as that for the calculation of the norm of a rock,
and to use the standard mineral molecules thus obtained
for making the necessary readjustments.? The recal-
culation of norm into mode, or mode into norm, is thus
much simpler than if the ratios of the various con-
stituents of the mineral are used, as was advocated in the
original publication of the quantitative classification.’°
The norm of the Haleakala augite is given here, as it will
be of use in calculating the modes of many Hawaiian, and
probably other, lavas.

Norm of Haleakala Augite.

ANOrthitie: Weis assert Bee 15.57
INemieliter: .. 20tit wie ee rege 3.12
DD)VOp Stes | AG Season ee eee 70.68
Olivine ss oe eee econ ana UepPh)
Macmetite: oii. eer ee 4.87
Tamenite sa. 0.5 eee ein en eae 3.65
Cir iiibe sere ieee ae ee 0.37

Attention may be called to the presence of the outer
film of more highly ferric material, which was noted on a
previous page. This points to a state of more highly
oxidizing conditions in the magma during the last stages
of crystal growth. Possibly a similar relation might be
detected in other zonally built augite phenocrysts’ by
observation of the differences in extinction angle between
the border and the interior, if a definite relation between
the extinction angle and the ferric oxide content could be
established. The case of our augite seems to be analo-
gous with that of the acmite-aegirite group, in which it
has been often noted that mixed crystals of acmite and
aegirite generally show a border of acmite with an
interior of aegirite. This is readily observable because
of the very pale yellow color of acmite and the bright
green, pleochroic color of aegirite. This subject of the
higher state of oxidation of the iron in the outer parts of
pyroxene crystals suggests some interesting lines of
thought, but discussion of them must await another
occasion.

Geophysical Laboratory,

Washington, D. C.

° Cf. H. S. Washington, The Roman Comagmatic Region, Carnegie Publica-
tion No. 57, p. 134, 1906.

* Cross, Iddings, Pirsson, and Washington, A Quantitative Classification of
Igneous Rocks. Chicago, 1903, p. 211.

Troxell—Rodents of Genus Ischyromys. 123

Art. VIII—Oligocene Rodents of the Genus Ischyromys;
by Epwarp L. Troxett.

[Contributions from the Othniel Charles Marsh Publication Fund, Peabody
Museum, Yale University, New Haven, Conn. |

A great diversity of form is shown in the rodents of the
Early Tertiary period; some of these have lived even to
the present time, others became extinct long ago and are
only remotely connected with existing genera.

Leidy in 1856! first described the squirrel-like rodent
which he called [schyromys typus, and in 1889? published
a detailed morphology of the skull and teeth, having not
only a full appreciation of the relationship to the modern
Sciurus but also a realization that the one was not
ancestral to the other.

Cope in 1873* and 1881* added new points of interest
concerning the skeleton, and further emphasized the rela-
tionship of this genus to the squirrels in the nature of the
teeth, but to other genera, Arctomys, Castor, etc., in skull
and skeletal features, and he agreed with Alston that
there should be a separate family, the Ischyromyide.
Several new genera and species were made by Cope,
which he afterward concluded were synonyms of [schy-
romys typus Leidy or belonged to entirely different
groups.’ Colotaxis cristatus and Gymnoptychus chry-
sodon are equivalent to Leidy’s genus and species; they
are based on what one finds to be moderate-sized speci-
mens with or without the variable tubercles between the
lobes of the lower molars.

Matthew® has worked out the interrelation of the early
rodents and discusses in full the family Ischyromyide.
His new species and subgenus [schyromys (Titanothe-
riomys) veterior was based on the narrow heel of M., the
narrow incisors, and the earlier geological age, 1. e. Titan-
otherium beds. He places the following genera under
Ischyromyide Alston: Ischyromys, Paramys, Sciuravus,

1 Joseph Leidy, Proc. Acad. Nat. Sci. Phila., 8, 89.

? Joseph Leidy, Jour. Acad. Nat. Sci. Phila. (2), 7, 335, pl. 26, figs. 1-6.

*K.D. Cope, Pal. Bull. No. 15, p. 1; ibid., No. 16, p. 5.

*E. D. Cope, Bull. U. 8. Geol. and Geog. Survey Terr., vol. 6, 366-368.

°See O. P. Hay, Science (2), 10, p. 253, 1899.
°*W. D. Matthew, Bull. Amer. Mus. Nat. Hist., vol. 28, 43-72.

124 Troxell—Rodents of Genus Ischyromys.

Mysops, and Prosciurus, and comes to the conclusion that
the genus Ischyromys reached extinction in Oligocene
time. In the present paper, however, the suggestion is
offered that they may have developed into the modern
prairie-dog of the genus Cynomys. .

DESCRIPTION OF NEW SPECIES.

Ischyromys pliacus, sp. Nov.
(Fie. 1.)

Holotype, Cat. No. 12511, Y. P. M. Middle Oligocene, Cherry Creek,
Colorado.

Fig. 1.—Ischyromys pliacus, sp. nov. Lower right dentition of the largest
species of the genus. Note the many small cusps on teeth. Holotype.
Cat. No. Zoli Ye Pee G2:

This specimen, consisting of the right lower jaw, is
notable for its large size and for the great number of
tubercles and deep pits on both premolar and molars. -
The posterior cross crests do not arise directly from the
external tubercle but from its union with the central
longitudinal ridge, thus forming a ‘‘Y’’. This cross
crest is made up of two distinct minute cusps, but only on
P, and M, are they stillunworn. Small additional tuber-
cles are to be seen on the posterior side of the anterior
cross crests on all the teeth and conspicuous cusps occupy
the wide openings of the external grooves.

Measurements are given in a table further on.

Ischyromys typus nanus, subsp. nov.
(ies. 2, 3.)

Holotype, Cat. No. 12519, Y. P. M. Oligocene (lower Oreodon beds),
Warbonnet ranch, 12 miles north of Harrison, Nebraska. Paratype, Cat.
No. 12555. Gerry’s Ranch, Weld Co., Colo.

The specimens here described consist of the small lower
jaws with the molar teeth. The holotype was a part of a
young animal in which the last molar was just being cut.
Besides the fact that this is the smallest subspecies of
ischyromyds, it is further distinguished by the very nar-

Troxell—Rodents of Genus Ischyromys. — 125

row M, and by the general absence of secondary tubercles,
external basal cusps, ete. In the ratios given in the table
beyond, it is seen that J. pliacus is over 50 per cent larger
in certain dimensions and averages about 30 per cent
greater.

Fie. 3

Fic. 2.—Ischyromys typus nanus, subsp. nov. Holotype. Cat. No. 12519,
Y. P.M. ‘Three molars from right lower jaw. M’-? very narrow. 2.

Fig. 3.—Ischyromys typus nanus, subsp. nov. Paratype. Cat. No. 12555,
Y. P. M. Left lower molars, to show results of wear. x 2.

Ischyromys typus lloydt, subsp. nov.

(Fes. 4, 5, 7.)

Holotype, Cat. No. 12521, Y. P. M. Oligocene (lower Oreodon beds),
from the ranch of Mr. Paul. Zerbst, 10 miles north of Harrison, Nebraska.

This new subspecies is based upon an unusually com-
plete skull and jaws, together with certain portions of the
skeleton. These were found in 1915 by the writer who
takes pleasure in adding them to the collections of Pea-
body Museum. The new name is given in honor of Dean
Lloyd of the University of Michigan, in eS ofa
delightful friendship.

Fic. 4.—Ischyromys typus lloydi, subsp. nov. Holotype. Cat. No. 12521,
Y. P. M. x2. A, molars and premolar of lower right side, and single
deciduous premolar, Dp,, from opposite side. J, left upper molars and pre-
molars; the latter were drawn from a vertical position, and the skull was
then rotated inward in order to get the crown view of the molars.

The specimen is moderate in size and corresponds in
general to the dimensions of I. typus Leidy. The upper
molars, unlike those in most specimens, do not decrease in
size toward the rear, for the last molar is actually greater
than the premolar or M'. The third permanent premolar,
the first of the cheek series, is just coming into view; it
was actually hidden beneath the conical deciduous prede-
cessor when first worked out of the matrix. A single
tubercle with a crescentic ridge on the postero-internal

126 Troxell—Rodents of Genus Ischyromys.

corner forms the crown of this tooth, which in shape and
size resembles the lead of a pencil, an elongated cylinder.

P* is distinctly molariform, but small and nearly round.
Unlike the molars, it has no internal groove and the cross
erests extend more than halfway across the crown. This
tooth now stands vertical in the maxillary, but had it
grown to its normal position, it would have turned out-
ward, following the curve of the large interior root, to
face more nearly like the molars.

M' and M? are alike in general form. In each, the pos-
terior groove, after slight wear, becomes a pit near the
center of the posterior side. M*® seems to be shghtly
rotated backward and inward, giving the cross crests an
oblique direction; it is elongated fore and aft. The tri-
angular area about the posterior pit is small.

The lower molars have the two simple outer cusps with
the two main, parallel cross crests leading from them.
The crowns are higher and the teeth are larger and wider
than in J. typus nanus just described. The last molar is
the widest; the first is narrowest and the total length of
the three molars is 10 per cent greater than that of
I. nanus. In all the subspecies of I. typus there is a
notable absence of small secondary cusps.

The first premolar was just appearing in the lower jaw
to take the place of the single deciduous tooth which had
been serving the young animal. Dp, is elongated but
narrow and has alow crown. The anterior cusps are not
paired transversely, for the imner one is placed far
anterior. (See fig. 4 4.)

The lower jaws are short and relatively deep; the
ventral border forms a broad curve backward to a point
beneath M,; here there begins a reverse curve and the
thin edge is folded inward to lend support to the pointed
angle. The coronoid extends as far above the condyle as
the angle does below; the two together give a vertical
measurement of about 26 mm. to the posterior part of the
ramus. On the upper side of the ridge leading to the
condyle opens the mandibular foramen; this passage then
leads down and forward, following along the upper sur-
face of the incisor root within the ramus, and finally
emerges in the mental foramen just in front of P,.
Except for the greater height of the ascending ramus, the
extended angle, and the lack of a distinct muscle scar on

Troxell—Rodents of Genus Ischyromys. 127

the outer surface, this mandible resembles that of
Sciurus, ef. S. carolinensis. -

Additional distinctive features in the present sub-
species may be noted in the vascular impressions on the
supra-occipital and the slight overhanging of this bone,
the large otic bulle, shghtly crushed down, the flat tri-
angle formed by the sagittal crest where it joins the inion,
and the absence of a deep groove on the zygomatic pedicle
near the maxillary.

SSS

S S hy AS MY .
} rh QA

Fie. 5.—Ischyromys typus lloydi, subsp. nov. Holotype. Cat. No. 12521,
Y. P. M. Side view of skull and jaws. Premolars just coming into view.
Skull slightly bent downward by crushing. Nat. size.

Measurements of Holotypes.

. I.nanus Ratio I. pliacus Ratio I. lloydi
mm. mm. mm.
Length of three lower molars.... 9.8 124 12.2 113 1k!
Sea hyOh WE cheery eis Sa7oers's ale Ss 3.2 116 Ball 113 3.6
ARC TEEOE 1 Rs RRC eee ee 2.7 LBSz Sal 119 3.2
MASHER ORS rk A ces bic. Be wk ios 3.2 125 4.0 116 Bae
WMG ee Visor eho tals 2 os 2.7 152 4.1 126 3.4
GH MAO pee ae es SA sats Ose 131 4,2 119 3.8
Wirohiia es Meets 2 heat is oS fie a os 3.0 123 Sel) 110 3133

INTERPRETATION OF ISCHYROMYS.

Wear of the teeth—The long transverse ridges or
crests and the corresponding grooves led Leidy to observe
that the upper teeth were like the lower ones reversed;
this is true, however, only ina general sense. In addition
to the transverse crests of enamel, the teeth show series

128 Troxell—Rodents of Genus Ischyromys.

of ridges clearly the result of lateral wear; furthermore,
the upper teeth are often worn to the roots on the inner
side while the outer borders still stand high. This gives
evidence that the movement of the jaws was transverse
rather than longitudinal as in the case of many rodents
where the grinding surface as a unit is worn to
one smooth plane (cf. the water hog, Hydrocherus
capybara). .

Fig. 6.—Upper milk teeth of Ischyromys typus Leidy, Cat. No. 12508,
Y. P. M. Dp’ is very much like the premolar which succeeds it. Dp* is
short-crowned, but elongated fore and aft. Note that the first molar faces
outward. X 2.

The third superior molar is set in the maxillary
facing downward and outward at an angle of more than
40°, but the deviation from the vertical is progressively
less toward the anterior teeth; the lower teeth correspond-
ingly face upward and inward. In the action of chewing,
therefore, the first contact is formed by the vertically
placed premolars, then, as the lower teeth move inward,
the molars from front to rear successively come into use.

Taking each tooth separately, its action may be ana-
lyzed into a chopping process with the first contact of the
sharp grooves and crests; this becomes secondarily a
tearing or grinding where two teeth are fully opposed; at
that phase the inner tubercles of the upper tooth and the
outer cusps of the lower tooth resist further lateral
movement, thus causing additional ermidine of the food
particles.

The milk teeth, like the premolars in ee early stages,
stand vertical both above and below, but the wear of these
teeth indicates that the immature individual has already
learned the transverse movement of the jaws. There are
two milk teeth above and one below in the young animal.

Relationship and adaptations——A most interesting
comparison may be made between Ischyromys, the fossil,
and Cynomys, the prairie dog, and although certain fea-
tures stand out sharply separating the two genera, yet
they show a fundamental similarity which tells of close
relationship. The skull of Cynomys (Cat. No. 01228,

Troxell—Rodents of Genus Ischyromys. 129

Y. P. M.) is about the same length but broader, especially
across the temporal portion of the zygomatic arch; there
are strong postorbital processes which are entirely absent
in the fossil. Further distinctions of Cynomys are the
narrower antero-posterior dimension of the teeth, the
large P? and M+, the anterior position of the infraorbital
foramen resulting from the large groove anterior to and
beneath the zygomatic projection of the maxillary, the
large angle of the lower jaw with numerous additional
projections for muscle insertions, the decrease in the size
of the coronoid, the longer segments of the hind limbs
with the ecnemial crest of the tibia and the third trochan-
ter of the femur far up on the shaft, the single groove
(instead of double as in the fossil) on the distal end of the
tibia for the tendon of the deep digital flexor running to
the sole of the foot, and the large, longer claws.

Important similarities, on the other hand, may be noted
in the structure of the teeth with the cross crests of
enamel, the shape of the occipital region of the skull and
the sagittal crest, the form of the nasals and premaxil-
laries, the peculiar supra-auditory foramen, and the same
dental formula. There is a remarkable agreement in the
size and form of the humerus and of the distal end of the
femur in each genus, both being distinct from those of the
squirrel, Sciwrus. In the humerus there is a strong del-
toid crest, a wide condyloid crest on the outer side, and a
supracondylar foramen on the inner side distally. This
foramen, according to Flower, is a departure from the
general rodent humerus, but is seen in the wombat and in
certain carnivores. Itis thought probable that the spiral
condylar grooves on the humerus permit a freer move-
ment of the ulna and allow its rotation for swimming or
burrowing as in the beaver, while the large deltoid prom-
inence and the lateral condyloid crest, again like those of
the beaver, tell of the strong muscles which manipulated
the front limb.

An objection which may be offered to any theory
attempting to derive Cynomys from Ischyromys is the
presence of the very small third premolar in the latter.
This resolves itself into the question whether P* of the
fossil is reduced to a vestigial state and therefore much
advanced in evolution, or whether it is small and yet unde-
veloped but has within it the potential molariform tooth
of Cynomys.

130 Troxell—Rodents of Genus Ischyromys.

The prairie dog now occupies the same Great Plains
region where once scurried the small ischyromyds, but
the conditions are greatly changed: the former is a bur-
rowing animal in a semi-arid climate; the latter lived
where the Oligocene streams from the young Rocky

Fie. 7.—Ischyromys typus lloydi, subsp. nov. Holotype. Cat. No. 12521,
Y. P.M. Nat. size. <A, front and back views of humerus; note broad distal
end, supracondylar foramen on posterior side, spiral condylar grooves, and
large deltoid crest on one side. 8, front and rear of distal end of the heavy
femur. C, right tibia viewed from inner side, showing deep double groove
for tendons to the foot. (See also figs. 4 and 5.)

Mountains spread widely their flood waters, just as the
rivers from the east slope of the Andes do to-day. This -
probably necessitated burrowing in the dry seasons and

at other times called for the ability to cope with the rising
flood.

Yamasaki—Glaciation of Japan Mountains. 181

Art. 1X —Glaciation of the Mountains of Japan*; by
N. YAMASAKI.

Glaciation, which played so important a role in the
development of the topography of parts of Kurope. and
America, is as yet known only in a few places in eastern
Asia. In China proper there are only a few evidences of
glaciation. Loezy notes its existence on the great snowy
mountains on the western boundary of the Province of
Szechuan. He also observed moraines on an eroded
plateau in the same mountains, but at an elevation below
the extremities of the present glaciers. We have no fur-
ther information as to the existence of glaciers in any of
the other high mountains of that mighty country, save
the discovery of tillite of great geologic age near the
Yangtze Gorge by Willis and Blackwelder. There is
searcely a report of the existence of glaciation farther
north, in the Amur Region, the Littoral Province, Kamt-
schatka or in Siberia.

It is only recently that the question of elaciation in
Japan has attr acted the attention of scientists and been
discussed by them. Forty years ago the Hida Mountains,
a lofty chain in Middle Japan, had already been visited by
a score of foreign alpinists. Among them, Atkinson
reports that the snow of the snowfield on Mount Hakusan
and Mount Tateyama near the coast of the Japan Sea is
hard enough to be called ice. inch, who crossed the
famous Harinoki Pass, described three or four snowbeds
and miniature glaciers. Divers also observed in the same
region some snowfields with the characteristic appearance
of glaciers. John Milne, then professor at Tokyo, com-
piled their observations and came to the conclusion that
at an elevation of about 5,000 feet in middle Japan and at
4,000 feet a little farther north, near the coast of the
Japan Sea, there are fields of permanent snow, which are
sometimes so firmly consolidated as to form small
glaciers. He further proved the possibility of Pleisto-
cene glaciation in Japan on the ground of climatological
and biological data. He also reported the presence of
‘‘roches moutonnées’’ on Gassan, a high, dormant volcano

* Read at the First Pan-Pacifie Scientific Conference at Honolulu, August,
1920.

132 Yamasaki—Glaciation of Japan Mountaims.

in North Japan, but later observers are doubtful of their
character.

Since that time no further investigations were reported,
not even from Milne, until recently some of our geogra-
phers took up the subject again. I have always been of
the opinion that there may be some traces of glaciers in
our mountains, especially the lofty ones of Middle Japan,
for there are several peaks 3,000 meters or more in height,
which have snow during nearly all of the year in conse-
quence of the heavy precipitation that results from the
oceanic climate. The winter snowfall is greater than that
of the European Alps. ‘Moreover the temperature does
not differ greatly from that of the Alps. So believing the
possibility of the existence of glacial features in our lofty
mountains, I have visited them several times to investi-
gate the question more thoroughly. ' Fortunately I made
the discovery of topographic forms characteristic of
glaciated mountains with the supporting evidence of
moraines and strie which show the former existence of
an ice covering in these mountains. |

Now let us consider the topography of Middle Japan.
Tn the center of the main island of Japan, where the two
principal mountain ares of North Japan and South Japan
meet, the terrestrial configuration attains its greatest
complexity. Several mountains, both tectonic and vol-
canic, stand side by side and upon one another. Indeed
the renowned volcanic chain of Fuji, with its numerous
gigantic cones, traverses the island and occupies the
depression zone between the precipitous scarps of the
two principal ares. Especially esteemed for their prom-
inent features are the mountains that lie on the western
side of the Fuji chain and form the northern extremity of
the southern are of the cordillera of our archipelago.
Their culminating peaks are 3,000 meters in. elevation, or
even more. The ridges are sharp and ragged, forming
sometimes pointed needles, pinnacles, or steep rocky walls
facing gorges a thousand meters deep. These scenes are
of truly Alpine character, and some years ago these moun-
tains were given the popular name of ‘‘The Japanese
Alps.’’ There are three principal mountains, namely
Akaishi, Kiso and Hida, which run parallel to each other
and trend southwest to north-northeast. Of these, Hidais
the highest and most characteristic. It forms the eastern

Yamasaki—Glaciation of Japan Mountains. 133

margin of Hida Plateau and falls abruptly to the rift val-
leys of Matsumotodairai and Himekawa with precipitous
fault searps. The most gigantic and imposing view that
it presents attracts many alpinists every summer from all
parts of the country. Yarigataka or Spear Peak, often
ealled the Japanese Matterhorn, Shiroumadake, Hodak-
ayama, Kurodake, Yakushidake, and T'ateyama are the
best known peaks. Besides them, Ontake, Norikuradake
and Yuodake are eminent volcanic cones and lie in the
southern part of this mountain. Hari-nokitoge is noted
as the highest mountain pass in Japan.

Climatologically treated, there is at present no moun- |
tain in Japan from the central chain of the Island of Sak-
halin on the north to the backbone of Formosa in the south
(nearly 4,000 meters high) which rises above the snow-
line. Even the beautiful snow-clad cone of Fuji, 3,778
meters high, loses its white cap in midsummer, leaving
only small patches or streaks of snow on the shaded side
of the crater or in the radial valleys. The same is true of
the northern part of the Japanese Alps. The annual pre-
cipitation in this region is as high as 2,500 millimeters,
owing to the great winter snowfall. Not on the mountain
itself, but on the plains or hills at the foot of its northern
slopes, the snow is so deep that small village houses are
entirely buried in it. On the ridges and flanks of the
mountains there are many patches of snow even in mid-
summer. ‘These, however, gradually melt away before
the next snowfall—by the end of October at the latest.
But it is not infrequently observed that rather large
snowfields, often a few meters deep, last through the year
in shaded valleys or below cliffs. Most of these snow-
fields consist merely of snow, but in some cases, as I have
observed in Shiroumadake, for example, the snow is con-
solidated partly to translucent ice near its base. One
might give the names ‘‘puny glaciers’’ or ‘‘miniature
glaciers’’ to such ice masses.

The climatic conditions in these mountains are not far
different from those of certain other regions where
glaciers are found at present. We have no meteorolog-
ical stations on any peak of the Japanese Alps, but if we
compare the temperatures of three meteorological sta-
tions in the towns of Takayama, Matsumoto, and Nagano
near the mountains, with that of Vienna at the foot of

Am. Jour. Sci.—FirtH Series, Vou. III, No. 14.—Frsrvuary, 1922,
10

18! Yamasaki—Glaciation of Japan Mountaims.

the Kuropean Alps, we can prepare the following table in
which temperatures are reduced to sea-level.

Elevation, above Temperature, January July
Station sea-level, meters Annual, C. Temperature Temperature
Takayama.... 561.4 10.0° —2.3° DAS
Matsumoto ... 582.1 LOM —2,.1° 22..0°
Naecano: i... : 420.4 OLOe —1.5° 22.9
Average ... 921.3 10.3° —2.()° 22.2°
Wienma, ane ce 170.0 10.0° —1.6° 22.9°

The temperatures at the four stations are nearly equal
to each other, while the elevations of the Japanese stations
are greater than the elevation of the Vienna station.
Dachstein, near Vienna, in the northern Limestone Alps,
has its snow-line below 3,000 meters elevation, and Hohe
Tauern in the Central Alps is covered with permanent
snow at an elevation of 2,860 meters. Of course, snow-
fall and glaciation are influenced by not only temperature
but also by other controlling conditions such as the
amount of precipitation, the humidity, the direction of
the prevailing winds, and topographic features. If we
consider only temperature, we find that the snow-line for
the Japanese Alps would be only a few hundred meters
above their summits, so that with a fall of only a few
degrees in temperature the snow-line would be depressed
to these peaks. ‘The precipitation would be sufficient to
make snowfields and glaciers.

In the course of a close examination of the topography
of the mountains we found various characteristic forms
of ridges and valleys due to glacial processes. The most
distinctive feature is the occurrence of cirques or kar at
the valley heads near the crests. Along the crest of the
central range of the Japanese Alps proper or the Hida
Mountain, from the gigantic needle of Yarigatake to the
renowned peak of Shiroumadake, we find this particular
type of topography. Again, nearly a dozen cirques lie
side by side on the ridges of Yakushidake and Tateyama,
to the west of the former, and between them is the deep
gorge of Kurobe. If we stand on a prominent peak, such
as 'Tateyama, it is not difficult to find a score of cirques on
the surrounding mountains. Some of them, and the val-
leys below them as well, are often covered with snow
patches in midsummer days. Weathering and erosion

Yamasaki—Glaciation of Japan Mountans. 135

have not as yet altered the forms greatly, and they are
still fresh and well preserved. Where many cirques lie
close together, as in JXurotake and Yakushidake, charac-
teristic comb-ridges are formed between them.

The most important and peculiar feature of these cirques
is that their floors are all of about the same height above
sea-level, as shown in the following table:

Meters.
CHOTET OY: Sak sae ar ean Me ae yASYI\ 0)
IMEGUAR ear ore eee aoe. S 2600
Ee rinOkae en 2500
INelsamomnatad is. ae: aie. 2550
iGumotalkentens st Sa ees 2500-2600
Wiakushier te es aa 2.600
J ETEUTOROC Sas Ream eee oe eae IR 2500
MAG Csyrailiiele so ee ees case Sic se tes, 3 2500-2600

In short, the cirque floors range from 2,500 to 2,600
meters in elevation, and average 2,550 meters. What is
the significance of this regularity? Perhaps there is no
better explanation than a former lower altitude of the
snow-line. Above this level the lofty ridges of the Japan-
ese Alps were once covered with permanent snow, and
just below there were hanging glaciers here and there
along the flanks of the ranges. Of course these glaciers
were not so large or extensive as those which existed in
the European Alps during the last ice age. We have not
yet found any morainal landscape at the foot of the moun-
tains. There are neither terminal moraines nor glacial
lakes at the lower extremities of the mountain valleys. The
glaciers must have terminated at a little higher elevation.
Perhaps certain exceedingly deep incisions in the upper
parts of the valleys are due to glacial erosion. Other
indications of glaciation that have been noted are the
striations on the rock blocks of Shirouma and the
moraines in several places, especially the peculiar moraine
mounds in a high valley basin of Kamikochi near the
famous peaks of Yarigatake and Hodakayama, which
were recently discovered by the late Professor Ozeki.

There is no question at present as to the glacial origin
of the peculiar features of our lofty mountains, and it
remains only to discuss when this glaciation took place.
We have not come to any satisfactory conclusions on this

186 =Yamasaki—Glaciation of Japan Mountains.

question as yet. The correlation of that period with the
ice age of HKurope and America is too difficult with our
present knowledge. We have not yet found any glacial
or glacio-fluviatile strata of diluvial age of undoubted
extension. Most of the Tertiary and post-Tertiary
strata that have been studied to date consist of marine or
brackish water formations, and our paleontologists are
eagerly endeavoring to determine the climatic conditions
of that epoch with especial reference to their fossil
faunas. Their opinions are as yet in disagreement.
According to Prof. Yokoyama of Tokyo University, who
has studied especially the formations of Tertiary and
later ages in the approaches of the Bay of Tokyo, the
climate of Middle Japan in the Pliocene was colder and
then became warmer in the diluvium than at present. So
he denied the evidence of an ice age in our diluvium. On
the other hand Professor Yabe, of the University of
Sendai, holds that at the end of the Tertiary the climate of
Japan became gradually colder and reached the minimum
temperature in the ice age, after which it became somewhat
warmer but again colder until it attained the temperature
of the present.

Summing up briefly what I have mentioned here, the
lofty mountains in Middle Japan in 36° north latitude
were once covered by permanent snow and bore glaciers.
Their existence is clearly proved only by. the character-
istic cirque topography with the corroborating evidence
of moraines and striae. The floor of many cirques are
nearly at a uniform elevation, 2,550 meters, which marks
the former snow line. The glaciers were hanging glaciers
and not of great extent and were located near the summits
of the ridges. No glacial or glacio-fluviatile deposits
have been discovered as yet that would allow of exact
determination of the date of the glaciation. In short,
the glaciated area of Japan is limited to a small district of
high elevation, in great contrast to the vast extent of the
glaciated areas of America and Kurope. The lack of
traces of glaciation in eastern Asia, even in high latitudes,
will also deserve further research. How and when did
the ice age occur in our country? How is it related to that
of other regions? Other equally important subjects for
investigation are the direct cause of the change of climate,
and its influences on topography and the distribution of

Yamasaki—Glaciation of Japan Mountains. 137

fauna and flora. Of course it is not our aim to attack the
problem of glaciation and subordinate questions from
only the geographical point of view. Help from geology,
paleontology, astronomy, climatology, biology and all
allied sciences must be cordially accepted. Especially
necessary are actual observations and comparative study
of the data. ‘T'’o such research, I am sure, this Conference
will give the best opportunities, and I beg you, gentlemen,
to kindly he!p me with your valuable suggestions as to the
present subject.

1388 T. Holm—Studies wm the Cyperacee.

Arr. X.—Studies in the Cyperace; by Tuo. Hou.
XXXIII. Carices aeorastachye: Macrocheteé nob.
and Nesophile nob. (With 12 figures drawn from
nature by the author.)

Among the tristigmatic sections of the Aeorastachye
the Macrocheteé comprise only a few species, viz. C.
macrocheta C. A. Mey., C. flavocuspis Franch. et Sav.,
C. gansuensis Franch., and C. scita Maxim.

C. macrocheta:

For many years this species has passed for C. podo-
carpa R. Br., as stated in a previously published paper.
C. A. Meyer described the species,? and has given an
excellent figure of it; some parts of the diagnosis may
be reprinted here inasmuch as, after all, the original
diagnosis is the most important:

‘‘Carex spica mascula solitaria, femineis binis distantibus
pedunculatis fructiferis subnutantibus, stigmatibus tribus,
perigynius laevibus compressis oblongo-lanceolatis acutis ore
integerrimis, gluma aristata brevioribus, bracteis foliaceis amplec-

tentibus. ... Folia floralia basi auriculato-amplectentia (nee
vaginata) ; infimum culmi partem superiorem excedens, superius
setaceo-filiforme spica duplo triplove longius.... Glumae

masculae et femineae oblongae, acutae, carina setaque longissima
scabra flava instructae; caeterum piceae et margine omnino con-
colores... . Perigynium obsolete nervosum, basi flavescens,
apice piceum, glumarum seta multo brevius.... Stigmata tria,
rarissime duo.”’

The species is aphyllopodic, and remarkably constant
in its habit, inflorescence and flowers. The terminal
spike is usually staminate, but in a few specimens from
Unalaska this was either androgynous, gynecandrous or
even entirely pistillate. The number of lateral spikes
varies from one to four; our material, representing
sixty one specimens, showed:

* This Journal, vol. 48, p. 17, 1919.
* Cyperaceze nove descriptionibus et iconibus illustrate. (Mém. Savants
EKtrangers, St. Petersbourg; vol. 1, p. 30, 1831.)

T. Holm—Studies in the Cyperacee. 139

1 staminate spike in 50 specimens.
9 ce ce c¢ EE ee a
Ss pistillaie so “9 4/33 s

9 ee ee ee 26 a4

1 4 ce ¢e¢ 3, 4

The relative length of the staminate and pistillate scale,
and the perigynium varies considerably even in the same
spike, when examined from base to apex. For instance,
in the typical plant we measured these organs, and the
length of the staminate scale was 10 mm., including the
arista, 3 mm.; in the same specimens the length of the
pistillate scale was 8.5 mm. including the arista, 3.5
mm. ; the length of the mature perigynium was 5 mm., and
the width 2.5 mm. But in some specimens from
Unalaska of which the spikes were exceptionally thick
and heavy, the following measures were observed: The
staminate scale was 19 mm. in length, including the
arista, 8mm.; the length of the pistillate scale, from
near the base of the spike, was 23 mm., including the
arista, 10 mm.; at the middle of the same spike the pis-
tillate scale measured 20 mm., including the arista,
10 mm. E

We have described two varieties: emarginata (figs.
4-6) from Alaska: Kukak Bay, and macrochlena (fig.
7-9) from St. Paul Island, Bering Sea.* Of these the
latter possesses remarkably large perigynia, measuring
7 mm. in length, and 3 mm. in width, while the subtending
Squama measured only 6 mm. in length, including the
arista, 1.5 mm.

Typical C. macrocheta is widely distributed on the
coast of Alaska, extending to Yukon, Washington and
Oregon; it occurs on the Asiatic coast: St. Lawrence
Bay, Kamtschatka and Sachalin. A near ally of C.
macrocheta is:

C. flavocuspis.

Franchet has redescribed the speciest and the diagnosis
reads as follows:

* This Journal, vol. 17, p. 314, 1904.
* Les Carex de 1’Asie orientale (Nouv Arch. Mus. d’hist. naturelle, ser. IIT,
vol. 9, p. 143, Paris, 1897).

140 T. Holm—Studies im the Cyperacee.

‘Rhizoma stoloniferum, stolonibus gracilibus, horizontalibus;
eculmi debiles 20-30 em. alti, acute trianguli, etiam superne leves,
inferne tantum foliati; vaginae infimae fulvae; folia culmos
aequantia vel superantia, mollia, 4 mm. lati; bracteae, evaginatae,
inferior foliacea inflorescentiam superans; spiculae 3-4 inferiores
pedunculatae, pedunculis ochrea fusca 5 mm. longa, cinctis;
spicula mascula unica, late oblonga vel oblonga, circiter 2 cm.
longa, squamis obovatis, fuscis, dorso lutescentibus, nervo in
mucronem levem excurrente; spiculae femineae 2-5 em. longae,
superior sessilis, squamis intense fuscis oblongis, masculis homo-
morphis sed longius acuminatis; utriculus 4 mm. longus squamas
subaequans, laete virescens, apice fuscescens, compressus, ellip-
ticus, margine et faciebus levis, in rostrum breve paulo emargina-
tum sensim desinens; stylus trifidus, achaenium sessile. Hab.
Le Japon: ile de Nippon, etc.’’

C. gansuensis.”

‘‘Repens; culmi 40-70 cent., acutanguli, etiam apice leves;
folia 4-6 mm. lata, culmis breviora; bractea inferior foliacea,
inflorescentia brevior; spiculae 3-4, omnes pedunculatae, suprema
tota mascula, rarius feminea superior subsessilis; spicula mascula
linearis, squamis rubro-fuscis, oblongis, mucronatis; spiculae
femineae cylindraceae, inferne et superne attenuatae acutae,
inferior paulo remota, pendens 3-4 cent. longa; squamez fuscae,
ovato-lanceolate, mucronate vel breviter cuspidatae utriculos vix
superantes, utriculi ovati, nervosi, leves, apice attenuati, in
rostrum tubulosum ore truncatum 1 mm. longum desinens;
stylus trifidus.

Hab. Le Japon.

Rappelle assez bien le C. Gmelin1, mais les feuilles sont plus
larges, les écailles au moins aussi longues que les utricules et ter-
minées par une pointe lisse; le bec des utricules est aussi sensible-
ment plus long.”’

These two species are so well defined by Franchet that
there seems no doubt of their validity as distinct species.
The fact that Kukenthal (1. c. p. 414) has disposed of
them as mere synonyms is of no importance, and espe-
cially not, when we examine his compilation of the so-
called ‘‘Scite’’; in this section, which is characterized
as being composed of ‘‘Aphyllopode,’’ we meet with
several species, which nevertheless are true ‘‘Phyl-
lopode,’’ namely C. rariflora Sm. (C. podocarpa R. Br.),
Sn ae nob., C. nesophila nob., and C. littoralis

chw.

* Nouv, Archives ]. ¢., p. 152.

T. Holm—Studies im the Cyperacee. 141

C. scita®

‘‘Rhizomate brevi obliquo squamis castaneis tecto stolonifero,
eulmo fertili 3-pedali gracili trigquetro laevi ac fasciculo sterili
per bina appositis basi squamis castaneis superioribus laminigeris
vestitis, foliis linearibus acuminatis carinatis margine reflexo
remote scaberulis culmo breviaribus, inflorescentia brevi e spicu-
lis 4-6 pedunculatis multifloris: terminali ¢ oblongo-lineari
utrinque attenuata erecta, reliquis apice masculis, superioribus 2
approximatis breve pedunculatis erectiusculis, inferioribus dis-
tantibus pedunculos teretiusculos laeves superantibus nutantibus,
fructiferis oblongo-ellipticis v. oblongo-lanceolatis pollicaribus v.
brevioribus, bracteis foliaceis non vaginantibus, infima culmum
superante, sequente subaequante, tertia spiculam quartaque
-pedunculum subsuperante fere setaceis, squamis (infimis vacuis)
lanceolato-linearibus atris utriculo brevioribus et angustioribus
costa viridi sensim in aristam scabram utriculo et in inferioribus
squama longiorem abeunte, utriculis erectis triquetris membrana-
eels viridibus secus angulos superne atrodenticulatis quavis facie
tenue 2-3-nervius oblongo-linearibus ore parvo quasi immerso
brevissime bidentatis, stigmatibus 3 (2) late villosis stylum
inclusum parum excedentibus, caryopsi in medio utriculo laxe
nidulante in stipitem stylumque sesquibreviores sensim attenuata
lanceolato-elliptica acute triquetra laevi straminea faciebus
convexiusculis.

Nippon media, ad rivulos, verosimiliter in montibus Hakone.

In serie mere artificiali androgynarum prope C. scitulam Boott.
(Ill. IV. 177. t. 600) ponenda, sed ab omnibus hisce jam culmis
basi squamatis abhorrens. Quo ultimo signo et gracilitate appro-
pinquat speciebus distigmaticis C. cryptocarpae C. A. Mey. et C.
Middendorfiu F. Schmidt., signis numerosis tamen adhuc
diversis.’’

In the work of Franchet, cited above, some species are
described which evidently are close allies of C. scita, viz.
C. rushirensis, C. scabrinervia, C. xanthathera and C. uro-
lems, all from Japan, but of which we have no material
for comparison. Moreover Meinshausen’ has described
a species C. koraginensis from Koraginsk, Kamtschatka,
which evidently also belongs to this section.

The Macrochaete represent a very natural section; the
Species show the same general habit, and are aphyllopo-
dic. The distribution of the sexes varies but slightly; we

* Maximovicz, C. I. Diagnoses plantarun navarum asiaticarum VI. (Bull.
de 1’Acad. Imp. d. se. de St. Petersbourg, vol. 31, p. 115), 1887.
"Die Cyperaceen der Flora Russlands St. Petersburg, 1901, p. 351.

142
‘s f yy ay

WZ [| | .f

| Ne

Y

CEEZSE

We / |

: \x

eS
WAS
WS
ZZ
SS
Cee

SS

—<$<————

9)

T. Holm—Studies in the Cyperacee. 148

have seen that in C. macrocheta the terminal spike may
sometimes be androgynous, gynecandrous or even purely
pistillate, but such structures are merely exceptions; im
C. koraginensis the terminal spike is often androgynous.
Very characteristic are the mucronate to aristate squa-
mae, and the membranaceous, more or less elliptic
perigynia with a short beak, though tubular in C. gan-
suensis; the perigynium varies from glabrous to densely
spinulose (C. scita, ete.) The species are confined to the
northwestern corner of this continent, and to the Asiatic
coast from St. Lawrence Bay to Japan.

Nesophile.

Carex nesophila nob. (figs. 10-12) is a species of such
peculiar habit and structure, that we feel unable to
place it in any other way than in a section of its own.
The culm is phyllopodic; the bracts are not sheathing;
the pistillate spikes are most often pedunculate, and
sometimes even drooping; the squamae are acute,
amplectent, and the midrib is thick, consisting of three
veins; the perigynium varies from oval to elliptical
oblong; it is distinctly several-nerved, and the short
beak is mostly entire; the number of stigmata is three,
or sometimes two.* In other words the habit reminds of
that of C. salina Wahlenb. or C. hyperborea Drej., while
the perigynium corresponds well with that of C. macro-
cheta. ‘he species was found abundant on uplands (St.
Paul Island and Popoff Island, Bering Sea) associated
with Sieversia, Potentilla and Artemisia, but not with
any other species of Carex; it seems to be a type charac-
teristic of these islands.

It is thus somewhat related to C. macrocheta, but can-
not be regarded as a variety of this, since it is phyllopo-
dic; moreover as stated above, the structure of the
Squama is different. When C. macrocheta ascends the
mountains it does resemble C. nesophila to some extent
by the coarser structure of culm, leaves and _ spikes.
Such specimens we have received from Dr. Walter H.
Eivans, who collected them ‘‘on mountain top, 1,200 ft.
(Kodiak, Alaska)’’; but in these specimens the culms are

* This Journal, vol. 17, p. 315, 1904.

144 T. Holu—Studies m the Cyperacee.

aphyllopodic, the staminate and pistillate scales are
mucronate to aristate.

It is also interesting to notice that the Macrochete-
alliance in the Japanese islands, so excellently described
by Franchet, contains no type comparable to C. neso-
phila; the members of this alliance are aphyllopodic and
of the same graceful habit as the typical C. macrocheta.

Clinton, Md., August, 1921.

EXPLANATION OF FIGURES.

Fie. 1. Carex macrochaeta C. A. Mey., the typical plant; scale of the
staminate flower; enlarged.

Fig. 2. Same species; scale of the pistillate flower; enlarged.

Fic. 3. Same species; utriculus; enlarged.
Fie. 4. C.macrochaeta var. emarginata nob.; the inflorescence; natural
size.

Fie. 5. Scale of pistillate flower of same; enlarged.

Fie. 6. Utriculus of same; enlarged.

Fie. 7. C..macrochaeta var. macrochlaena nob.; the inflorescence; nat-
ural size.

Fig. 8. Scale of pistillate flower of same; enlarged.

Fic. 9. Utriculus of same; enlarged.

Fie. 10. C. nesophila nob.; the inflorescence; natural size.
Fie. 11. Scale of pistillate flower of same; enlarged.

Fie. 12. Utrieulus of same; enlarged.

Chemistry and Physics. 145

MON tate ENTRAR LETGERNC EH
I. CHEmistRY AND Puysics.

1. A New Cyanide—wW. S. Lanpis has given an interesting
account of the recent development of the commercial prepara-
tion of cyanide from calcium cyanamide at Niagara Falls.

After the successful fixation of atmospheric nitrogen by bring-
ing this gas into contact with hot caleum carbide, CaC,, with the
formation of the cyanamide, CaNCN, and free carbon, work
was carried on in Germany for about 15 years in the effort to
manufacture cyanide from the cyanamide. The most promising
method appeared to be fusion with sodium chioride, but the
process was not a commercial success on account of difficulties in
carrying it out and the low yields of cyanide obtained.

In 1916 the supply of sodium cyanide in this country became
very small and its price very high, so that an attempt was made
at Niagara Falls, where a cyanamide plant was in successful
operation, to manufacture cyanide from this material by fusing
it with sodium chloride. By means of improvement in the
process, such as the addition of a little calcium carbide to prevent
frothing, and rapid heating to a higher temperature for a shorter
time, cyanide was produced on a larger scale of a much higher
erade than the Germans had ever hoped for. Within a year the
process was further improved and the product made still better.
Since that time cyanide has been manufactured on a very large
scale, equivalent to the production of 4,000,000 lbs. of sodium
cyanide per year, and of a grade corresponding to about 35% of
sodium cyanide. Even richer products than this have been
obtained, and it is expected that the quality will be considerably
improved in the future.

The product has proved itself very satisfactory for use in gold
extraction, for the preparation of sodium ferro-cyanide, and for
other purposes, so this development may be regarded as a remark-
able triumph of the American electrochemist.—Trans. Amer.
Electro-chem. Soc. 38 (1920). H. L. W.

2. A Very Sensitive Reaction for Copper.—Pirnrre THOMAS.
and GEORGES CARPENTIER have found that the alkaline solution
phenolphthalin (the colorless reduction-product of the well-
known indicator phenolphthalein) which was used by Kastle for
the detection of vegetable oxydases and afterwards recommended
by Meyer as a test for blood, is, under proper conditions, an
exceedingly delicate reagent for copper. The reagent was pre-
pared by dissolving 2g of phenolphthalein and 20 of pure potash
in 100 ce of water and boiling the solution with 10¢ of zinc-dust
until colorless. To 10 ce of very dilute solutions of copper four
drops of the reagent were added and then a drop of hydrogen
peroxide solution of a strength corresponding to 5 or 6 volumes:

146 Scientific Intelligence.

of oxygen. <A solution containing 10-°, that is, one millionth of
copper, gave immediately a rose tint changing to bright red in a
few seconds. At a dilution of 10~ the color did not appear
until after 10 or 20 seconds, while at 10-°, the limit, a very pale
color appeared after a few minutes which could be easily seen by
comparison with a blank containing no copper.

The authors have compared this test with others, including the
one with hematoxylin described in this journal in 1906 by Brad-
ley, and they conclude that it is the most delicate of all. They
give warning that certain supplies of distilled water and many
so-called chemically pure reagents contain appreciable amounts
of copper when subjected to this test.—Comptes Rendus, 173,
1082. H. as We

3. Lehrbuch der Eisenhittenkunde; von BERNHARD OSANN.
Zweiter Band: Erzugung und Higenschaften des schmiedbaren
Hisens. 8vo, pp. 794. Leipsie, 1921 (Verlag von Wilhelm Engel-
mann).—This text-book on the production and properties of
wrought iron and steel appears to be a most excellent treatise
that can be highly recommended to those who read German and
are interested in the subject. It presents many tables of statis-
ties, explains the historical development of the important pro-
cesses, and give a very complete and satisfactory account of the
present condition of the art from both the theoretical and prac-
tical points of view.

The book is copiously supplied with no less than 651 illustra-
tions ‘‘in the text’’ and 10 larger tables or plates. A large pro-
portion of the figures are admirable sectional drawings of
apparatus, showing the details of construction.

It should be noticed that the operations which give the final
form to the metal, such as rolling, drawing, hammering and
casting, are not treated in the book on account of the necessity
for limiting its size. Theoretical metallography is only very
briefly discussed, although the practical subjects of hardening,
cementing, ignition, etc., are thoroughly treated. H. Tews

4. Mining Physics and Chemistry; by Z. W. WHITAKER.
12mo, pp. XII and 268. New York, 1921 (Longmans, Green &
Co. Price, $3.00).—The author of this text-book is a Certified
Colliery Manager, and Lecturer in the Mining Department, Uni-
versity College, Nottingham, England. The book presents the
simpler fundamental principles and facts of physics and chem-
istry, with particular attention to topics of special interest in con-
nection with coal-mining. For instance, considerable attention
is paid to pumps, and the subject of mine gases is rather elabor-
ately treated in connection with the detection of explosive mix-
tures and the physiological effects of CO, and CO,. As another
example the discussion of explosives and their employment in
the mining of coal is quite elaborate.

Chenustry and Physics. 147

The book gives an excellent impression in regard to its clear
presentation of the subject in an elementary way. Without
doubt it might be used to advantage by many of those engaged in
coal mining in the United States, for the differences in technical
terms, mining regulations, and weights and measures in England
and this country world be of little importance in a book devoted
chiefly to scientific instruction. H. L. W.

5. The Adsorption of Gas by Charcoal and Silica.—The exten-
sive application of the phenomenon of adsorption of gases by
charcoal in the construction of masks for protection from noxious
gases, and for the reduction of pressure at liquid air temperatures
in evacuated vessels has greatly stimulated research upon the
nature of the process. An investigation by Professor Henry
Briggs recently published had for its main object the discovery of
a non-inflammable substitute for activated charcoal but the results
are also important from his consideration of the theoretical
aspects of the phenomenon.

The gases used were nitrogen and hydrogen and out of a large
number of non-inflammable substances tried a dried and granular
silica gel prepared from water glass was found to be the best.
Its capacity for nitrogen at 190° C. and at atmospheric pressure
was found, on the volumetric basis, to be greater than that of the
best charcoal by 66 per cent. Both of these substances showed
preferential adsorption for nitrogen and hydrogen and exemplify
the influence of chemical composition upon the phenomenon.

In any sample of granulated charcoal or silica the author
regards the gross volume to be made up of: (a) the volume of the
solid matter, (b) the interstitial volume 1. e. of the open spaces or
voids between the grains, (c) the internal gaseous volume which
is made up of the volume of the capillaries or polymeral inter-
stices within the granules themselves. A rough estimate indicates
that these volumes constitute 27, 31, and 42 per cent of the whole
in activated charcoal or silica but have the relation 51, 6, 48 per
cent in de-activated silica. The author is of the opinion that in
considering the intrinsic nature of adsorption we have not to do
with the ultimate molecule of the solid but with elaborate poly-
mers. The degree of canalization of the substance both on the
microscopic and the molar scale is important but these passages
are of use only when they traverse a medium which is in proper
physical condition. For a high capacity silica when de-activated
has no adsorbing power although it still retains one-fifth of its
original internal gaseous volume, and on the other hand pow-
dered graphite which has no capillary passages shows consider-
able adsorbing power. It is to be inferred that the connecting
passages serve to distribute the gas but that it is held chiefly on
the surfaces of polymers in the openings of molar dimensions.
On this view any treatment which would disrupt the elaborate

148 Scientific Intelligence.

polymers of a substance would render the material more active
and this is apparently confirmed. A high capacity silica may be
de-activated by a heating which does not destroy the capillaries
since 20 per cent of the original porosity remained. Here a gran-
ule which was originally an aggregate of small polymers riddled
with openings has been converted into a completely polymerized
unit by partial vitrification.

The behavior of carbon may also be explained on the above
assumption. It is known that a dense charcoal subjected to a
prolonged heating in the presence of a small amount of oxygen
or when treated for a shorter time with superheated steam has
its adsorptive power greatly increased. Many writers have
thought that this activation resulted mainly from the expulsion of
the residual hydrocarbons of the raw charcoal, but this assump-
tion is open to the objections, (1), that the small amount of vola-
tile hydrocarbon which is left to be expelled is disproportionate to
the result produced; (2), that the presence of oxygen though
helping to break up the hydrocarbons would not be essential to
their ejection, and, (3), the fact that a temperature exceeding
1000°C. is needed before they would be all expelled does not
square with the theory, if as is usually stated, a temperature from
350° to 450° C. is the optimum for air activation. The more prob-
able explanation is that a highly polymerized substance like ecar-
bon is likely to be simplified by heating. It is found, however,
that heating to 1000° alone has no permanent effect; the simplifi-
cation resulting from the raising of temperature is: followed by
a reunion on cooling. But if a little oxygen either from air or
from steam is available when the polymers are split, a portion of
the carbon is removed as CO,, and their cleavage is perpetuated.

Obviously oxygen would not serve the same useful purpose in
the case of silica, but here the preparation of the silica from the
hydrogel by heating involves the driving off of water which is
for the most part chemically combined with the silica and the
process will tend to break up the polymers.—Proc. Roy, Soc. 100,
88, 1921). FR. EE

6. Tables Annuelles de Constantes et Données Numériques ; by
the Comité international de l’Unon de Chimie pure et appliquée.
Vol. IV, Part I. Pp. XXII, 626. Paris, 1921 (Gauthier-Villars
et Cie.).—The present volume contains the determinations of phy-
sical constants as abstracted from researches which appeared dur-
ing the years 1913-1914-1915-1916, and is published with finan-
cial aid from a large number of scientific societies throughout the
world.

Without sacrificing anything of their purpose to make the
tables complete, the editors have wisely exercised their discretion
in making useful condensations and omitting results which were
obtained under uncertain or ill-defined conditions, but including

Chemistry and Physics. 149

reference to them in the list of memoirs. In certain cases the
printing of innumerable figures has been avoided by drawing
curves; in others, such as the study of fusion in binary and
tenary systems, only the properly characteristic values have been
retained. .

The material is tabulated first under the twenty chapter heads,
compressibility, elasticity, density, viscosity, surface tension,
expansion, specific heats, thermal conductivity, thermodynamics,
melting points, vapor pressures, gas laws, acoustics, photometry,
radiation, infra-red spectrum, absorption, refraction and dis-
persion, spectroscopy, rotatory power. These are printed in each
of the four languages, French, German, English, and Italian,
arranged in parallel columns. Each chapter head is followed
with a detailed list of topics and the pages upon which the tables
are to be found. The text of the work, however, where any
explanation is necessary is In French. The source from which
the values were taken is indicated in every instance and a con-
siderable bibliography is frequently appended.

This relatively large work will be found extremely valuable in
every technical or university laboratory. The distribution in
this country is to be made through the University of Chicago
Press from whom it may be obtained at $13.25 net.

The second part which is promised for 1922 will treat of electri-
cal constants, solubilities, colloids, and other matters pertaining
more particularly to chemistry. | Ey eB.

7. Report of the National Physical Laboratory for 1920. Pp.
132. London, 1921 (His Majesty’s Stationery Office).—The
activities of this (Teddington) laboratory will be of interest to
anyone engaged in research in Physical Science, and attention
may be called to the list of papers, pp. 37-42, published by the
laboratory or communicated by members of its staff to scientific
societies or to technical journals. F. E. B.

8. Die Naturwissenschaften: Zweiter Band; by FRrIEepRICH
DANNEMANN. Pp. X, 508. Leipzig, 1921 (Wilhelm Engel-
mann ).—This is the second edition of the volume in which the
author has traced with great care the progress of scientific
thought from Galileo to the middle of the Highteenth Century.
It appears to be historically accurate and is entertainingly
written. It was the author’s aim to supply a book which should
exhibit not only the development of the principles of physical
science but also to show their relation to such other branches as
philosophy, mathematics, therapeutics and technology. The
volume is illustrated with 132 figures drawn from the original
memoirs and gives their sources. The work is well indexed and
contains extensive notes to the literature of the subject. It will
be appreciated both by students and by teachers. F. E. B.

Am. Jour. Sct.—Firta Series, Vou. III, No. 14.—Frsrvuary, 1922.
11

150 Scientific Intelligence.

Il. Grotogy anp MINERALOGY.

1. Geological Survey of Canada.—Bulletin 33 of the Canadian
Geological Survey (109 pp., 12 pls., October, 1921) is made up
of the following papers: Faunal and Sediment Variation in the
Anticosti Sequence, by W. H. Twenhofel; New Species of Devo-
nian Crinoidea from Northern Canada, by Frank Sprmger; The
Range of certain Lower Ordovician Faunas of the Ottawa Valley,
with Descriptions of some New Species, by Alice EH. Wilson; The
Fossil Molluscan Faunas of the Marl Deposits of the Ottawa Dis-
trict, by E. J. Whittaker; and Two New North American
Cycadeoids, by G. R. Wieland.

Professor T’wenhofel sets forth in his paper very important
conclusions regarding the Anticosti Island Silurian and Ordo-
vician sediments and faunas. He says that ‘‘lateral gradation
of sediments and faunas may so develop that one type of sediment
with its fauna may overlap another—the conditions responsible
for one type of deposition migrating laterally with respect to the
other. The common interpretation would be ‘overlap’ of the
one by the other, a withdrawal of the sea, a land interval, and the
development of an unconformity.’’ Twenhofel does not accept
this current interpretation of lateral changes, but states it to be
‘the purpose of this article to describe examples of sediment and
faunal variation in the shallow Ordovician and Silurian seas in
which were deposited the sediments which now constitute the
rocks of the Anticosti sequence.’’

Doctor Springer’s paper describes and figures two new crinoids
belonging to the genus Melocrinus, from the Mackenzie basin.

The paper by Miss Wilson materially increases our knowledge
of the geological range of the species comprising the Black
River and Trenton faunas in the Ottawa valley, and adds some
new forms to these faunas. The author shows the range of
species by means of a series of tables. These indicate at just what
point in the section each species makes its first appearance, and
just where it disappears from the section. This paper is a good
example of the sort of precise work in collecting and studying
fossil faunas which is very much needed.

Mr. Whittaker’s contribution deals with a fresh-water fossil
fauna found in the marls of the Ottawa valley, which, in its time
relations, lies between the latest marine Pleistocene interval and
the living mollusean fauna. Students wishing to study the fresh-
water fossils of the Ottawa valley Pleistocene wil! find the plates
and keys of this paper most useful. The illustrations include an
aeroplane photograph showing the relation of the fossil marl
deposits to the present water-level at McKay Lake. So far as the
reviewer is aware, this is the first aeroplane photograph to be
used in illustrating a paleontological paper.

Geology and Mineralogy. 151

- Doctor Wieland describes the first eyead ever recorded from
Canadian rocks, found in the Belly River beds in Alberta, and a
second new species from the Trinity of Texas. EH. M. KINDLE.

2. A Review of the Evidence for the Tacomc Revolution; by
Tuomas H. Cuark; Proc. Boston Soe. Nat. Hist., 36, 135-163,
1921.—The author finds but little orogenic evidence on which the
hypothesis of the Taconic Revolution of Dana (1863) can be
based. This evidence is restricted to the southwestern margin of
the New England mass from Otisville, New Jersey, north to
Becraft Mt., New York. ‘‘Beyond this region there is no evi-
dence of the Taconic Revolution.’’ He does not deny epeiro-
genic movement and high land areas toward the ciose of the Ordo-
vician, resulting in thick formations of clastics of Richmondian
and early Silurian times. Cais:

3. Fossilium Catalogus, 1: Animalia. Pars 12, Aves, by K.
LampBrecHt, Pars 18, Cmdana triadica, by C. DrENER. Pars 14,
Ammonoidea permiana, by C. Diener. 1921.—Part 12 of this
valuable work gives the more important literature on the known
fossil birds of the world, nearly 700 in number; also the material
on which the species are based, the geologic horizon, the geo-
graphic locality, and the museum in which the material is kept.
The classification followed is in the main that of Furbringer.
Part 13 cites the literature of 185 forms (in 33 genera) of Hexa-
eoralla, 7(3) of Tetracoralla, 11(5) of Tabulata, 13(6) of Monti-
eulipora, 1 of Aleyonaria, and 15(12) of Hydrozoa. Part 14 gives
a bibliography for 198 forms of Permian ammonites, in 41 genera
or subgenera. Seven of these genera also occur in the Carboni-
ferous and the same number pass into the Triassic. The author
also gives valuable remarks on their geologic and geographic
distribution. Cus:

4. Report of the State Geologist on the Mineral Industries
and Geology of Vermont, 1919-1920, 332 pp., 47 pls., 24 text figs.,
1921.—This report consists of the following ten papers: Struc-
tural and metamorphic geology of the Hanover district, New
Hampshire, by J. W. Merritt; A contribution to the geology of
Essex County, by R. A. Schroeder; Notes on areal geology of the
western flank of the Green Mountains, by N. C. Dale; Geology
and mineralogy of Braintree, by C. H. Richardson and C. K.
Cabeen; Detailed study of the Trenton beds of Grand Isle, by
State Geologist G. H. Perkins; Report on Trenton fossils from
Grand Isle, by R. Ruedemann; Progress in tale production, by
EK. C. Jacobs; Studies in the geology of western Vermont, by
C. E. Gordon; The geology of Lake Willoughby, by E. C. Jacobs;
and Mineral resources, by G. H. Perkins.

Dd. Bernice Pauahi Bishop Museum of Polynesian Ethnology
and Natural History, Honolulu, Hawaiu.—Recent publications of
this Museum include the following titles. From the Occasional

152 Scientific Intelligence.

Papers, vol. 7, (1) Director’s Report for 1918 (pp. 1-12);
(2) The Languages of the Pacific, by J. M. Brown (15-29) ; (3)
New Hawaiian Plants, VII, by C. N. Forbes (33-39) ; (4) A New
Cyanea from Lanai, Hawaii, by C. N. Forbes and G. C. Munro
(43); (5) Notes on Marsilea villosa Kaulf, by C. N. Forbes (47-
49); (6) a New Variety of Partulina hornert, by J. J. and
A. Gouveia (53); (7) New Species of Sierola with Explanatory
Notes by D. T. Fullaway (57-158); (8) Director’s Report for
1919 (168-176) ; (9) Edible Mollusca of the Oregon Coast, by
C. H. Edmondson (179-201); (10) Fish Poisoning in the
Hawaiian Islands, by J. F. G. Stokes (219-282); (11) An
Archeological Survey of Haleakala, by K. P. Emory (237-259) ;
vol. 8, (1) Report of the Director for 1920 (8-28). In the series
of Memoirs: vol. I, (1) A Monographie Study of the Genus
Pritchardia, by O. Beccari and J. F. Rock (77 pp., 23 pls.) ; (2)
A Contribution to Samoan Somatology, by L. R. Sullivan (17 pp.,
if plsa)s

6. Mikroskopische Physiographie der Minerahen und Ges-
teine; by H. RosenpuscH. Band 1, Erste Halfte: Dve
Petrographische wichtigen Minerahen und die Methoden ihrer
Untersuchung. Fifth edition, fully revised by EH. A. WULFING.
First installment, 252 pages, with 192 text figures and one colored
plate (an interference chart). Stuttgart, 1921 (EK. Schweizer-
bart’sche Verlagsbuchhandlung)—From the preface we learn
that the new edition of the Physiographie, whose appearance has
been delayed by the war, is practically a new work, so great has
been the addition, selection, and reworking of the material. In
order to make room for the large amount of matter available and
to keep the new edition from growing unwieldy, some matter that
appeared in the previous edition has been omitted, such as the
treatment of stereographic projections, some of the graphical
methods, and certain demonstrations in geometric optics.

Following the Introduction, which is shortened and much
changed from that in the fourth edition, is a section on methods
of preparing material for optical examination, the main portion
of which is, of course, devoted to describing the methods of
making of thin sections. Incidentally, Wilfing shows that he
does not view with sympathy the substitution of kollolith for the
time-honored Canada balsam! The new position of this section
in the book is obviously an improvement over that in the old
edition, where it was inserted in the midst of the theory of opties.

Much new material appears in the book. Some of this may
be briefly mentioned. There is an interesting computation on the
requisite size (or number) of thin sections necessary to obtain
certain definitely oriented sections of a mineral, say a basal sec-
tion of a uniaxial mineral or a section cut parallel to the c-axis.
The treatment of the dispersion of birefringence is newly added
material.

Geology and Mineralogy. toe

Becke’s point that the thin sections generally used in micro-
scopical work are so thin that most interference figures do not
show isochromatic curves and that therefore the isogyres are the
characteristic phenomena is given proper emphasis. The inter-
ference figures shown in most manuals are such as we never see:
in everyday work. On the whole, the treatment of interference
figures is inadequate, as it is in all manuals. The only adequate
presentation of this difficult subject is Becke’s masterly exposi-
tion, which should be consulted in the original.t Probably the
second installment, which will deal with practical applications,
will remedy this inadequacy.

The treatment of dispersion has been improved and elaieed
and is illustrated by excellent figures. The new account of pleo-
~chroic halos shows more clearly than any other revision in the
book one of the notable lines of progress since the fourth edition.
The description of the various.kinds of luminescence is new and
has been introduced in the hope that it is anticipatory of future
advances.

The deseriptions of the various prisms for obtaining polarized
light, occupying pages 222-239, seem unnecessarily detailed, in
view of the professed purpose of the book; and they might well
be condensed or omitted. The installment ends with the section
on the preparation of monochromatic leght.

In conclusion, we can say that the thoroughness of the revision
and the eare shown in the first installment of the new edition to
bring the subject abreast of the times leads us to look forward
with anticipation to the succeeding installments of the great
source book in microscopic petrography. ADOLPH KNOPF.

7. Morphogenese der Oecetscherlandschaft. Vienna, 1921,
Karu Drwap (published by the Author). Large 8vo, 404 pages
with a large folded map, 1:25,000, and a 21-page supplement
containing 70 block diagrams.—This volume is a serious but seem-
ingly venturesome effort to interpret the complicated forms of
the Oetscher district, near the northern margin of the Alps, about
80 kil. west-southwest of Vienna, in terms of successive episodes
of erosion prompted by many partial uplifts of the region.
Twelve systems of forms are thought to be identified. The book
is difficult reading because the text is a veritable ticket of details,
without summaries. W. M. D.

8. New Meteorites; by Grorcr P. Mzerrinu (communicated ).—
The National Museum reports the receipt of a fragment of a
heretofore unknown meteorite (pallasite of the Rockiky type)
from Cold Bay, western Alaska. The entire mass as found was
in form of a badly oxidized mass of but a few pounds weight,
which was at once broken up and in large part lost. The find is
of interest, being the second from Alaska proper, the first being
that of Chilkat (an iron). Two finds, Skookum Gulch and Gay

* Becke, F., Tschermak’s Min. Petr. Mitt., vol. 24, pp. 1-34, 1905.

154 Scientific Intelligence.

Gulch, have been reported, however, from the Yukon District,
Canada.

Samples of the recently reported ‘‘2-ton’’ mass of meteoric
iron found near Navajo, Arizona, which have reached the
‘ National Museum show it to belong to the dense, nearly structure-
less varieties carrying but 5.81 per cent nickel, and provisionally
referred to the Nedagolla group. This puts to rest the early
supposition that the same might be a straggler from the Canyon
Diablo fall.

A 54-pound mass of meteoric iron belonging to the octahedrite
group has been reported to the National Museum from Mount
Tabby, 20 miles northwest of Hannah P. O., Utah. G. P. M.

9. Feldspar Studies—From Science Reports of the Tohoku
Imperial University, Series ITI, vol. 1, No. 1, pp. 1-32. Sendai,
Japan, 1921. X-ray analysis of Adularia and Moonstone and the
influence of temperature on the atomic arrangement of these min-
erals; by SHUKUSUKE Kozu and YosurrosaH1 Enpo.—The two
feldspars investigated had the following compositions:

Orthoclase Albite Anorthite
Adularia, St, Gotthard i... 88.3 (23.1 2.9
Moonstone, Ceylon ......... 74.4 23.1 | 24

The adularia was shown to have a single space-lattice which
showed no variation over a wide range of temperature. The
moonstone showed a combination of two space-lattices which were
stable at 700° C. and lower temperatures, but as the temperature
was raised above this point the two lattices approached each other
until they became identical. The eight plates contain many
Laue-spot photographs.

Optical, Chenncal and Thermal Properties of Moonstone from
Korea; by SHuUKUSUKE Kozu and MasatosHr SuzuK1.—This
material corresponded to the composition,

Orthoclase: ).- 61.0. Albite} 3.3 33-1) Amorthites a9 see

The Laue-photograph showed also in this case double spots
indicating two intergrown space-lattices, although here they
appeared in the picture taken on the side pinacoid while with the
Ceylon moonstone they appeared on the base. The two lattices
merge into one on raising the temperature but at a lower point
than was the case with the Ceylon mineral. At the same time
the schillerization shown by the mineral disappeared, leading the
authors to assume that this property is IN some way due to the
presence of two molecular lattices in the mineral.

10. A Manual of Determinative Mineralogy with Tables; by
J. Voungy Lewis. Third, Revised and Enlarged Edition. 298
pp. New York, 1921 (John Wiley and Sons).—The present
edition of this very useful book is at least twice the original size.

Miscellaneous Scientific Intelligence. 155

Complete Determinative Tables based upon physical characters
have been added which occupy nearly one-half of the book.
These together with the older tables based upon blowpipe and
chemical tests make a very complete series of determinative tables
- which should serve for all ordinary purposes. The whole book
has been recast and revised. W. E. F.
Sanidine from the Eifel; by SHuKusUKE Kozu and KUuNI-
KATsu Seto.—The paper is a continuation of a previous thermal
study made of this material by the senior author. W. E. F.
11. A Text-Book of Mineralogy; by Epwarp S. Dana. Third
edition revised and enlarged by Wmu1AmM HE. Forp. Pp. ix, 720.
New York, 1922 (John Wiley & Sons).—A notice of the new
edition of this well-known work, just issued, will be given later.

III. Misce,ttanrous Sctentiric INTELLIGENCE,

1. Dary Bacteriology; by ORLA-JENSEN. Pp. 180, Philadel-
phia, 1921 (P. Blakiston’s Son & Co.)—This, the English transla-
tion by Paul S. Arup of the second Danish edition of 1916,
brought up to date by recent corrections and additions by the
author, is a text-book for use of students of dairying, and also
others who have to deal with milk and dairy products. It is essen-
tially a practical, understandable book on dairy bacteriology, and
not a scientific treatise. The three chapters given to general
microbiology are very elementary, concise and to the point. The
remainder of the book is a study of the relation which various
microorganisms have to the changes taking place in dairy prod-
ucts. The normal and abnormal microbial flora of milk, butter,
and the cheeses are very ably discussed. Other defects in these
products which are not caused by bacteria, yeasts, or molds are
described and remedies for each suggested. The best methods
for properly handling and preserving milk are outlined. There
is a rather full discussion of the various sour milks, the methods
of preparation, characteristics of the particular organisms used,
ete. The chapter on cheese-making takes up in detail the methods
of controlling the processes of ripening. One chapter is devoted
to a consideration of the various tests used in the grading of
market milk.

Considering the ability of the author in the field of systematic
bacteriology, it is rather a disappointment not to find a detailed
scheme of classification for at least the milk bacteria. However,
since the book is intended for elementary students and dairymen,
the brief classification of the lactic acid bacteria given will proba-
bly serve the purpose quite as well as a more elaborate one. The
book is of exceptional value because it is built up on the experi-
ence of the author covering a period of twenty-five years.

GEORGE F. REDDISH.

156 Scientific Intelligence.

2. Report of the Secretary of the Smithsoman Institution,
CHARLES D. Waucort, for the year ending June 30, 1921.—The
Smithsonian Institution ever since its foundation has made most
commendable use of the income at its disposal, but it has suffered,
with other institutions having an original considerable endow- .
ment, from the fact that additions to its funds have been slow to
come by gift or bequest. The need arising from this is felt
peculiarly by the Smithsonian since the scope of its work has been
continually enlarged in directions not contemplated at the begin-
ning. The situation is aggravated at present by the general
increase in cost of materials and labor. In this connection it is
eratifying to learn that most of the bequest ($50,000) of Mrs.
Virginia P. Bacon has been paid during the past year. It is also
announced: that, by her will, the entire estate of Miss Caroline
Henry, daughter of the first secretary, Joseph Henry, will finally
come to the Institution; she has in addition made special gifts
immediately available. .

Among the numerous varied lines of scientific exploration
being carried on, none is more important than the work of
Dr. Walcott that has already yielded such valuable results. His
recent field work in the Canadian Rockies had as its main objects:
‘‘(1) the determination of the character and extent of the great
interval of nondeposition of sedimentary rock-forming material
along the Front Range of the Rocky Mountains west of Calgary,
Alberta; (2) the clearing up of the relations of the summit and
base of the great Glacier Lake section of 1919 to the geological
formation above and below.”’

Dr. Walcott describes the solution of the two problems as fol-
lows: “*The Rocky Mountain front is formed of masses of evenly
bedded limestone that have been pushed eastward over the softer
rocks of the Cretaceous plains-forming rocks. This overthrust is
many miles in extent and occurred long before the Devils Gap,
Ghost River Gap, and other openings were cut through the cliffs
by running water and rivers of ice. Great headlands and high
buttes have been formed by the silent forces of water and frost,
many of which stand out against the western sky as seen from the
distant foothills and plains.

It was among these cliffs that we found that the first great cliff
was of lower Middle Cambrian age, and that resting on its upper
surface there were 285 feet of a yellowish weathering magne-
sian limestone, named the Ghost River formation, which repre-
sents the great interval between the Cambrian below and the
Devonian above. Sixty miles to the west, over 4 miles in thick-
ness of limestone, shales and sandstones occur in the break in
sedimentation of Ghost River cliffs. Returning to Bow Valley, -
the party left the Canadian Pacific Railroad at Lake Louise and
went north over Pipestone Pass to the Siffleur River, which is

Miscellaneous Scientific Intellugence. 157

tributary to the Saskatchewan. In the northward-facing cliffs,
25 miles east of the Glacier Lake section of 1919, and 40 miles
north of Lake Louise, a geological section was studied that tied
in the base of the Glacier Lake section of 1919 with the Middle
and Lower Cambrian formations. Returning up the canyon

valley of the Siffleur River to the wide upper valley of the Clear-
water River, a most perfectly exposed series of limestones, shales,
and sandstones of Upper Cambrian and later formations was
found, which cleared up the relations of the upper portion of the
Glacier Lake section to the Ordovician above.’

The year’s work of the National Museum, now in charge of
W. ve C. RAVENEL, has been particularly important in that the
National Gallery of Art has now been separated from the museum
and made an independent unit. The Aircraft Building has also
been opened to the public, and much work has been done in reor-
ganization, so that the museum at present has a total of 49
recognized subdivisions. A total of nearly 340,000 specimens
have been added to the museum during the year; many of these
are mentioned in detail.

The International Exchange service has been much enlarged
owing to the resumption of relations with Germany; an increase
of more than 82,000 packages over the preceding year is noted.

The work of the Astrophysical Observatory is summarized by
the director, Dr. C. G. ABgort, as follows:

‘*The year has been marked by the transfer of the solar radia-
tion measurements from Mount Wilson, California, to Mount
Harqua Hala, Arizona, to secure more perfect weather conditions.
It is intended to continue solar constant observations there daily
when possible throughout the entire year for several years.
Similar duplicate observations are to be carried on at Monte-
zuma, Chile, at the private station of the Smithsonian. Thus it
is hoped to provide an excellent basis of solar radiation measur-
ments to compare with weather phenomena. This may lead to
advance in methods of weather forecasting. Volume IV of the
annals, covering the years 1912 to 1920, is practically ready for
the press.’’ 7

3. Publications of the Carnegie Institution of Washington.—
Recent publications of the Carnegie Institution are given in the
following list (see earlier, vol. 1, p. 376, 1921):

No. 185. Index to United States Documents relating to for-
eign affairs, 1828 to 1861; by Apruame R. Hasse. In three
parts; Part Ill, pp: 1333; 1980 - quarto.

No. 273. Contributions to-Embryology, Nos. 47 and 48. Vol.
X, quarto, pp. 11, 103; 16 plates.

Nos. 275, 276. Contributions to Embryology, No. 56. Vol.
XII. Studies on abortuses; by F. P. Maun and A. M. Meyer.

158 Scientific Intelligence.

Quarto, pp. 364; 25 plates, 6 figs. Vol. XIII, Nos. 57-64. Pp.
146; 20 plates, 25 fies.

No. 284. Distribution of V egetation in the United States as
related to climatic conditions; by B. E. Livineston and ForR=EstT
SHREVE. Pp. xvi, 590; 73 plates, 74 figs., 152 tables.

No. 298. Leodicide of the West Indian Region; by A. L.
TREADWELL. Quarto, pp. iv, 131; 9 plates, 467. fies.

No. 303. Tables, Factors and Formulas for computing respira-
tory exchange and biological transformation of energy; prepared
by DME CARPENTER. Pp. 123; 33 tables with introduction
explaining description and use.

No. 305. Selection in Cladocera on the basis of a physiological
character; by ArtHuUR M. Banta. Pp. 170; 18 figs., 49 tables.

No. 307. Growth in Trees; by D. T. MacDouesu. Pp. 41;
16 figs.

No. 310. Displacement Interferometry applied to acoustics
and to gravitation; by Carn Barus. Pp. vii, 149; 183 figs.
No. 313. Rubber-Content of North American Plants; by
Harvey M. Hauty and Frances E. Lone. Pp. 60; 3 plates.

No. 315. Aeration and Air-Content: The Role of Oxygen in
Root Activity; by Freprric HK. CLeEMEnts. Pp. 183. ;

-Also a Classified list (with prices) of publications of the Car--
negie Institution, dated February 15, 1921.

4. Report of the Inbrarian of Congress, HERBERT PUTNAM,
and Report of the Superintendent of the Inbrary Building and
Grounds, F. L. Averituu, for the year ending June 30, 1921.
Pp. 207, with 6 full page illustrations. Washington, D. C.—
The striking point made by Mr. Putnam is one already urged,
namely a thorough readjustment of the salary schedule. The
lack of this is causing a serious loss in the staff through resigna-
tions. A small number of additional assistants and clerks is also
needed to make the organization more perfect. The sum called
for by these needs is $170,930; deducting present bonuses, the
net increase is $83,330. The library continues to gain in num-
bers, the number of books added for the year being 86,923.
Besides these, manuscripts, maps and charts, music and prints
have been obtained by gift or purchase. Many of these are
described in detail.

aig ain Sate

ARDS. carne Scene epee

AD Supply-House for Scientific Material.
Founded 1862. : Incorporated 1890:
. a few of our recent circulars in the various
ae _ departments:

Geology: J-32. Descriptive Catalogue of a Petrographic Col-
lection of American Rocks. - J-188 and Se pee

Se 2 ‘ Price-List of Rocks.

Mineralogy: J-220. Collections. J-238. Minerals by. Weight.
J-224. Autumnal Announcements.
Paleontology: J-201. Evolution of the Horse. J-199. Palx-
ozoic index fossils. J-115. Collections of Fossils.
Ve: Entomology: J-33. Supplies. J-229. Life Histories. J-230. .
Live Pupe.
Zoology: J-223. Material for dissection. J-207. Didseationk
of Typical Animals, ete. J-38. Models.
Microscope Slides: J- 189. Slides of Parasites. J-29. Cata-
logue of Slides.
Taxidermy : J-22. North American Birdskins. Z-31. General
Taxidermy.
Human apts J-37. Skeletons & Models.
General; -228. List of Circulars & Catalogues.

‘Ward’s Natural Science Establishment
84-102 College Ave., Rochester, N. Y., U.S. “A.

“SCIENTIA”

“INTERNATIONAL REVIEW OF SCIENTIFIC SYNTHESIS. Issued monthly (each

number consisting of 100 to 120 pages). Editor: EUGENIO RIGNANO.

This is the only review which has a really international collaboration; which is
of world-wide circulation and occupies itself with the synthesis and unification of
knowledge, in the history of the sciences, mathematics, astronomy, geology, physics,
chemistry, biology, psychology and sociology.

It is the only review which, by inquiries among eminent scientists and writers
(on: The philosophical principles of the sciences; Fundamental astronomical and physical
questions of current interest; Contributions given by the various countries to different
branches of knowledge; Question of vitalism; Social question; International questions
raised by the world war), makes a study of the problems interesting scientific and intel-
lectual circles throughout the world.

It has published articles by Messrs, : Ahho. Nichenkis =Ashley = Bayliss = Beichman-
Bigourdan - Bohlin - Bohn= Bonnesen = Borel = Bounty = Bragg = Brani = Burdick = Carver -
Caullery = Chamberlin - Charlier = Claparede - Clark - Costantin = Crommelin = Crowter=
Darwin = Delage = De Vries = Durkheim = Eddington = Edgeworth = Emery = Enriques =
Fabry - Findlay = Fisher = Fowler = Golgi = Gregory = Harper - Hartog = Heiberg - Hinks-=
Hopkins-Inigues-Innes-Janet-Kaptein-Kaye-Kidd-Langevin-Lebedew-Lloyd Morgan-

_ Lodge = Loisy - Lorentz - Loria -Lowell - MacBride= Meillet -Moret-Muir-Peano -Picard -

Poincare = Puiseux - Rabaud- Rey Pastor = Righi=Rignano-Russell-Rutherford-Sagnac-=
Sarton= Schiaparelli-= Scott = See - Sherrington = Soddy = Starling = Svedberg- Thomson =
Thorndike-Turner -Volterra=Webb-Weiss = Zeeman and more thana hundred others.

**SCIENTIA’’ publishes its articles in the language of its authors, and joins to the
principal text a supplement containing the French translations of all the articles that
are not in French. (Write for a Specimen Number to the General Secretary of
“Scientia”, Milan.) Annual Be peri pon: 40 sh., or 19 dollars post free.

Office : 43 Foro Bonaparte, Milan, Italy.

Publishers: WILLIAMS & NORGATE-London; FELIX ALCAN -Paris
~ NICOLA ZANICHELLI-Bologna; RUIZ HERMANOS-Madrid;
WILLIAMS & WILKINS CO-Baltimore.

= z L % = x = i

Art. TV-— —Gastropod Praila: in Sour. Sandstones i
*Ap-Texas; ‘by SaPOwERS. 22. 2-o1o es Sates es oe :

Arr. V. —Seaside Notes; by P. E. Raa oe

i oon Arr. VI, —The “Varve: Shales” of Australi; by a8 w.
eon : Davip.......-- Fe ate Pits AP APE Taine ace eC 3 Ge

: Arr. VIL- Mineralogy: hae of Haleakala, Maui, ie :
"--waiian Islands; ae i. S. “Wasuryerox: and i fh sam
MER WIN: Se aceon oe, See ea ee a eee

| Arr, -V If, ““Oligbeane Bodentd of the Genus eehyeomyss 2 =
sis Ps by E, L. TROXELL.... AE es st acl eat: eee ;
oe ABT, TX: —Glaciation of the Mountains of Japan ; = ty a ioe

i MN AMASAIEO Misco 75ers a aa ries ~ 188 ee
a _ Arr. X:—Studies in the Cyperacen; by pT Howe? “ip
XXXIII. Carices aeorastachye: Macrochxtz nob. ay
- and Nesophile. nob. (With 12 figures drawn. from 3! Raa eae
ay es Py. the author). 32 25533 P50. eda Er oe PM ea 138 ae ig em
ae - SCIENTIFIC INTELLIGENCE, fl Ag Re

Chemistry and. Physics.—A new Cyanide, W. S. Lanpis: A Very ‘Sensitive: %
Reaction for Copper, P. Tuomas and G. Carpentier, 145.—Lehrbuch der —
-Eisenhiittenkunde, B. Osann: Mining Physics. and Chemistry, Z. W
“WuitakER, 146.—The Adsorption of Gas by Charcoal and Silica, 147.— }
Tables Annuelles de Constants et Donnés Numériques, 148. —Report ofthe |
| ~ National Physical Laboratory for 1920: Die a Delieryiseenes eas Zweiter

Band, F. DaNNEMANN, 149.

Geology and Mineraieee. — Geological Siiriey of Cannas. 150. —- at Review" of |)
- ‘the Hvideuce of the Taconic Revolution, T. H. CLARK: Fossilium Cata-
logus, K. LAMBRECHT, etc.: Report of the State Geologist on the Mineral
‘Industries and Geology of Vermont, 1919-1920: Bernice Pauahi Bishop
Museum of Polynesian Ethnology and Natural History, Honolulu, Hawaii,
-151.—Mikroskopische Physiographie der Mineralien und Gesteine, H. Ro- |
_-SENBUSCH, 152.— Morphogenese der Oetscherlandschaft, K. Drwaxp: New.
Meteorites, G. P. MerRILL, 153. — Feldspar Studies: ‘Optical, Chemical
and Thermal Properties of Moonstone from Korea, S. Kozu and M. Suzuki: }|
“A Manual of Determinatiye.Mineralogy with Tables, J. V. eae 154, fot: cae fe

_ Text-Book of Mineralogy, KE, 5, Dana and W. KE. Forp, 155. = es

Miscellaneous Scientific Intelligence.—Dairy Bacteriology, pare 155.
—Report of the Secretary of the Smithsonian Institutlon, C. D, WaLcorT,
for the year ending June 30, 1921, 156.—Publications of the Carnegie In
stitution of Washington, 157. — Report of the Librarian of Congress, Bree

~ Putnam, and Report of the Superintendent of the Library Buildings ‘and. ;
Grounds, F. L. AVERILL, for the year ending June 30, 1921, 158. [a

‘Library, U. S. Nat. Museum. ae ee = pe xs .

= ¢ ~

|

2

THE | = |
|

|

AMERICAN

* Bieviie: eu aR S. DANA.

ASSOCIATE EDITORS

7 FESSOR WILLIAM M. DAVIS anp REGINALD A. DALY,
or CAMBRIDGE,

Boies HORACE LL. WELLS, CHARLES SCHUCHERT :
5 HERBERT E. GREGORY, WESLEY R. COE anp
ie FREDERICK E. BEACH, or New Haven,

" PRoFEsson EDWARD W. BERRY, oF BALTIMORE,

>
OF ae :
Rae es Ze aR FIFTH SERIES |

ape VOL. Iil{WHOLE NUMBER, CIID) |
Ee No. 15 MARCH, 1922. |
eee oe : 2 Aoi PLATE lt. f
__NEW HAVEN, CONNECTICUT.

*
‘

eee FE 492 2.

e or : xs : = : a
MOnsEOUSE & TAYLOR CO., PRINTERS, 123 TEMPLE STREET. |

‘ ana:

; Six dollars per year, in advance. $6.40 to countries in the
ada. Single numbers 50 cents; No. 271, one dollar.

ose ashe Post Office at New Haven, Comn., under the Act

re oe ¢ % @ Without a Peer ee? ¢ @ Be |

DANA-FORD
TEXT - BOOK OF

MINERALOGY

BY EDWARD SALISBURY DANA

Third Edition, Revised.
BY WILLIAM E. FORD

PROMINENT Mineralogist refers to this new edition as ‘‘a worthy
A successor of the book that has so long stood without a peer among text_
books of mineralogy.”
An extended treatise on Crystallography and Physical Mineralogy is included.
720 pages. 6 by 9. Profusely illustrated. Cloth, $5.00.

USEFUL MINERALS AND RARE ORES —

By ALEXANDER McLEop

BOOK of ‘‘Practical Instructions in the Search for, and the Determina-
tion of, the Useful Minerals, including the ‘‘Rare Ores,” to quote from
the complete title of this popular work. The Second Edition has been very
well received. . yee
204 pages. 4 by 64. Flexible binding, $2.50.

Ready late in May!

A HAND-BOOK OF THE PETROLEUM INDUSTRY

By Davin T. Day, Ph.D., Editor-in-Chief, in Collaboration with
Twenty Contributors and Associate Editors.

A complete, authoritative treatise.

JOHN WILEY & SONS, Inc.

432 Fourth Avenue, New York

AJS 3.22 :

-

Plate Ill

Naits Jou Seto Wol, Wh, W222,

"1/1 ynoqe X

‘yoodse

TePeLPAS

TOT wysupw xhsawoys

De HO

m0T}B

104Sey—

.

(ea)

JUL

I 981d

db dsb ad:

AMERICAN JOURNAL OF SCIENCE

PEERS Bek RSs

0. —— —

Art. XI—Restoration of Blastomeryx marsh; by
Ricuarp 8. Luni. With Plate III.

[Contributions from the Othniel Charles Marsh Publication Fund, Pea-
body Museum, Yale University, New Haven, Conn. |

In this Journal for August, 1920, I described the skull
and jaws of a new species of Blastomeryx which was col-
lected by Professor Marsh in 1873 along the Niobrara
River not far from the mouth of Antelope Creek,
Nebraska. Further material pertaining to this species
has come to ght in the skull and a considerable portion
of the skeleton of a second individual collected at the same
time and place and designated by Professor Marsh
“‘Rum[inant|(XX),’’ the holotype, Cat. No. 10937,
Wee... being ‘*Rum.(X)-”’

The second skull, Cat. No. 10756, Y. P. M., differs from
the type in its shghtly smaller size, the teeth being in
about the same degree of wear in the relatively smaller
canine alveolus, and in the position of the horn, which is
not only smaller, but is set back from the rim of the orbit
as in the female prong-buck. One is therefore justified
in the assumption that this second specimen represents
the female of the species of which the holotype is the male.

The skeletal material was so much more complete that a
mount of the animal was attempted, utilizing the male
skull and jaws and the female skeleton. Thus the mount
is composite, and embraces both Nos. 10937 and 10756,
difference in size of the skeletal parts present in each ani-
mal being almost negligible.

Distally the limbs are beautifully preserved, including
the entire manus with complete lateral metacarpals and
digits. Whether the pes possessed lateral digits is not
clear. Proximally the posterior limb bones are_repre-
sented by articular Se only, ne :

"MOISTA JO o[oUv 0} ONP T[ “Oly YIM poreduod sv squity pury Jo eimgsod zo vduvyo yuorrddy *)/T
qnoqe X ‘TUT ‘sy Aq popepoyr “qoodse ysey~ “TUT wsiow chsawojisyy,g Fo uoye10ysey-—'Z “DIT

Lull—Restoration of Blastomeryx marsha.

160

Lull—Restoration of Blastomeryx marsh. 161

fore limb were approximately complete. The humerus
could have been.no shorter than restored, as there is bone
throughout; it may have been longer, but this 1s extremely
doubtful. Of the femoral and tibial length I am not so
sure. Doctor Matthew’s restoration of Blastomeryax!
was used, but in no two hmb segments was there corre-
spondence where the length was actually preserved. The
two restorations differ in many details, some of which
may well be due to interpretation of material. As usual
with these small mounts, I have restored the flesh on the
right side, with carefully wrought muscular interpreta-
tion and a minimum of conjectural surface detail. The

mounting of the skeleton was very largely the work of
Mr. Hugh Gibb.

*W. D. Matthew, Bull. Amer. Mus. Nat. Hist., vol. 24, fig. 6, 1908.

162 M. R. Thorpe—Oregon Tertiary Cande

Arr. XII.—Oregon Tertiary Camde, with Descriptions
of New Forms; by Maucotm RuTrHERFORD T'HORPE.

[Contributions from the Othniel Charles Marsh Publication Fund, Pea-
body Museum, Yale University, New Haven, Conn. |

CONTENTS.

Introduction.
Description of species.

Cynodictis oregonensis Merriam.

C. angustidens (Marsh).

Nothocyon geismarianus (Cope).

N.latidens (Cope).

N.lemur (Cope).

Philotrox condoni Merriam.

Paradaphenus transversus (Wortman and Matthew).

Temnocyon altigenis Cope.

Mesocyon corypheus (Cope).

M. brachyops Merriam.

M. josephi josephi (Cope).

M. josephi secundus, subsp. nov.

Pericyon socialis, gen. et sp. nov.

Enhydrocyon oregonensis, Sp. Nov.

Tephrocyon rurestris (Condon).
References.

INTRODUCTION.

During the course of a study of the canid and allied
- material in the Marsh Collection at. Yale, it was found that
Amplicyon angustidens Marsh and Canis gregarius Cope
were specifically identical. Evidence of this will be pre-
sented in a forthcoming paper on the Great Plains
Canide. Marsh’s description antedates Cope’s by two
years, and since both species belong in the genus Cyno-
dictis, the C. angustidens of Marsh (1871) has precedence
over Cope’s name proposed in 1878. The type, Cat. No.
11762, Y. P. M., consists of the anterior part of a right
ramus with teeth.

The taxonomic position and affinities of the new genus,
Pericyon, herein proposed, are not at present discussed,
other than to say that it is probably a derivative of
Mesocyon. It apparently stands closer to that genus
than it does to Temnocyon, which it approaches in size.

DESCRIPTION OF SPECIES.
Cynodictis oregonensis Merriam.

This specific name was applied to the John Day forms of

with Descriptions of New Forms. 163

C. angustidens (Marsh) (Galecynus gregarius (Cope) ).
The species is distinguished from the latter by the ‘‘con-
stant presence of a posterior cusp in addition to the ante-
rior and posterior basal tubercles on P,, the larger M?,
larger brain case, less pronounced postorbital constric-
tion, and other characters’’ (Merriam 1906, p. 11).. The
presence of the additional cusp on P, was not noted by
Cope nor is it shown in the figures accompanying his
deseription of the John Day specimens of C. angustidens.
This character appears on some of the Yale specimens of
the White River C. angustidens and C. paterculus, and is
apparently an individual variation which can not be used
for specific determination. The other characters, mainly
dependent upon larger size, are what we should expect
from the John Day beds. This increase in size over the
- White River forms is seemingly a very constant charac-
ter in various genera, having representatives, in both that
area and the Great Plains region, of approximately the
same geologic age. )

This species is represented in the Marsh Collection by
several skulls, all from the middle John Day. The infra-
orbital foramen above the interval between P? and P*, size
and shape of the bulle, posteriorly pointed nasals, flat
palate, and moderate slope of basicranial axis are all sim-
uar to those in C. angustidens. Specimen No. 127338,
Y. P. M., from the Fossil Horse beds on Cottonwood
Creek, has three upper and lower premolars on the right
side and four on the left. It is upper John Day in age.

A variant from the typical C. oregonensis is found in
skull No. 12700, Y. P. M., collected near Logan Butte,
Crook County, Oregon, in middle John Day beds. The
face and jaws are unusually long; cranium high; diame-
ter of postorbital constriction small; postorbital pro-
cesses prominent; angle on ramus prominent; length of
inferior molar-premolar series 48 mm.

Measurements.
(Chiefly from Cat. No. 12679, Y. P. M.)

mm.
LOVES USF ke se 103
LDU SUieie air VRE Tee Sn Ie BH
PAMCheEr Or POSLOrOIpal CONSELICTION ....6506.50.04- 55. 14

164 M. R. Thorpe—Oregon Tertiary Cande

Length of superior tooth-row, with canine............... 45.5
heneth of superior premolawseries= +44 see noes 25
Lenethy oi superior molar Seriesi ser ae nee eee 112
Ant-post. diameter of Me go Ny a eis ey eae ee )
Transverse «diameter of Mes 5 oa, oe ee eee 8
Length of inferior premolar series (No. 12681, Y.P.M.).. 20
Leneth of inferior molar series (No. 12681, Y. P. M.)..... 19

Cynodictis angustidens (Marsh).

Parts of rami with teeth, from the John Day formation,
have been tentatively identified with this species. The
majority of these were collected at Bridge Creek. They
are very close to the type and much smaller than C. ore-
gonensis.

Two specimens from the Fossil Horse beds on Cotton-
wood Creek, Cat. No. 12696, Y. P. M., are more slender —
than the type, and the teeth are more crowded. These
may represent a new form, more closely allied to C.
angustidens than is C. oregonensis, but with material so
fragmentary it seems best to identify them with the for-
mer species for the present.

Nothocyon geismarianus (Cope).

Remains of this species were collected at Turtle Cove
and Haystack Valley from middle John Day strata. The
specimens consist of parts of skulls and associated jaws,
with both superior and inferior dentition, but unfortu-
nately are not well preserved.

Nothocyon latidens (Cope).

In the Marsh Collection, a pair of mandibles with teeth,
Cat. No. 12794, Y. P. M., represent this species. Both
carnassials exhibit ‘‘a narrow tubercle at the external
base of the principal cusp’’ which Cope considered diag-
nostic of latidens. In addition there is also present a
quite small tubercle just anterior to the base of the
entoconid. The specimen was found at Turtle Cove in
middle John Day strata.

with Descriptions of New Forms. 165

Measurements.

Cat. No. Holotype
12794 of N.
WéGdes IM latidens
mm. mm.
Eeeranit-POSt. diameter .) re oe.e so. oes 6 5.0
Merah =POSteGtameter oo. si. ce ks Sate 8.4 8
M,, ant.-post. diameter of heel ......... 3.0 3.0
Deprhok ramus at sectorial ... 2.32... .. 10.5 10.5

Nothocyon lemur (Cope).

Parts of a right and of a left ramus with teeth are
referred to this species. They bear the catalogue number
12797, Y. P. M. The ramus is considerably more slender
than that of N. latidens, and the measurements of this
specimen accord well with those of the type. However,
the sectorial possesses the small tubercle on the external
base of the protoconid, a characteristic of latidens. Mer-
riam has described mandibles possessing this character,
but otherwise agreeing with N. lemwr, and he is inclined to
think that this feature is common to both species. J am
inclined to concur in this opinion.

Measurements.

Cat. No
eo Holotype
Yer Me of N. lemur
mm, mm.
reeeectnit- WOSts CI aMeter . <a. ge 2. st a cid: 0.0
MepramE = POSt. CIAIMeberN. 002 sess oe ea es 8 8
Me ant:—post. diameter of heel’s... 25... 2.9 3
Depinvor ramus at sectoriall. 4. 253 oe. ff 8

Philotrox condom Merriam.

Several parts of rami with teeth represent this genus
and species, but they furnish no additional data to Mer-
riam’s descriptions (1906) of the form. They were col-
lected at both Bridge Creek and Turtle Cove, from middle
John Day strata. Cat. No. 12725 is the left ramus of a
submature individual, showing partly deciduous and
partly permanent dentition, including the incisors.

166 M. R. Thorpe—Oregon Tertiary Cande

Paradaphenus transversus (Wortman and Matthew).

This genus was defined by Wortman and Matthew in
1899 (p. 129) as follows: ‘‘Upper molars much extended
and symmetrical transversely; M? aligned with outer
cusps of anterior molars. Heels of lower molars wide
and deep, basin-shaped; M, and M, with two anterior
cusps and basin-heel.’’ The type species of the genus is
P. cuspigerus ee which is based on specimen No.
6852, A. M. N. H

The species P. tr ansversus was defined at the same time
(p. 180) as ‘‘Size one third larger (lineal). Upper pre-
molars compressed.’’ The type is No. 6851, A. M. N. H.
(referred to Amphicyon hartshormanus by Cope). For
amplification of this genus and species, Cat. No. 10064,
Y. P. M., may be used. It is a skull which lacks the pos-
terior part of the cranium, as well as the crowns of part
of the dental series, ete the roots distinct, except
those of the incisors.

In addition to the characters mentioned by Wortman
and Matthew, we are enabled to add the following: infra-
orbital foramen above posterior part of P*; face short;
diameter of postorbital constriction large; cranium low
and full; sagittal crest barely marked; frontal ridges
unite just posterior to coronal suture; palatonarial bor-
der opposite M?; palate flat, descending abruptly at alveo-
lar parapet; P' and P? spaced; deuterocone of P* prom-
inent; M* very small; zygomata not widely expanded.

A pair of lower jaws, Cat. No. 12714, Y. P. M., collected
near Camp Watson on the John Day River, are identified
with this species. In many respects these jaws are sim-
ilar to those of Daphenus. M, is a very small convex
nub, considerably worn, and may be smaller than indi-
cated by Wortman and Matthew. P, differs from that
of Daphenus in having a very small incipient posterior
cusp. The incisors are small and laterally compressed.
Both Nos. 10064 and 12714 are middle John Day.

It is well to note the spelling of this generic name, Para-
daphenus, while Leidy’s generic name is Daphenus.

with Descriptions of New Forms. 167

Measurements.
(Cat. No. 10064, Y. P. M.)

mm
tora ckullsleneth: partly estimated :..)..........ss 0... 157
Bizygomatic diameter, partly estimated ................ 77
rameter or: postorbital comstriction .......2.0..... 6.65 os 30
Meneihiol smperior premolar series ..........%-..008-0- +4
Menon Omsuperior MOlar SeTICS <2. ag. seek ke et ee ee ee 20

(Cat. No. 12714, Y. P. M.)

Ben itoiaimiertor dental series, C-MEA IMG. . ) 02 see. (2
Sen phe owunterior premolar Series 2.5. 2.55 66k oe ee 34
Men spheoimiheriOr. MOlArcSETIES. oo. ow ices dis eee ee ews 21.5
Eee PO eC et emo Mit. Sst Shaw ee wh eee 3 ties 14.5
Pee BO SieeO TAME bers Oil yin e Sect. y Ae SSS alesse ss ode bos ke 8
EeeereO Sime CITI HERAO Ey VE hn wah ce es wale Sure one 4

Temnocyon altigens Cope.

This comparatively rare species is represented in the
Marsh Collection by a fully adult skull, Cat. No. 10065,
Y. P. M., and other material. This specimen corresponds
with Cope’s type, and possesses the zygomata and cranium
which were lacking in his specimen and which have been
heretofore undescribed. It varies somewhat from the
individuals described by Merriam (1906, pp. 21-29), as
will be seen from a comparison of measurements.

The condylar foramen is small, while the foramen
lacerum posterius is large. The eustachian opening and
the external carotid foramen are separate. The optic
foramen is large. The infra-orbital foramen is larger
than in any specimen of Canis familiaris of equivalent or
even of somewhat larger size.

The palatonarial border lies 8 mm. posterior to M?’.
The zygomata are slender in proportion to the size of the
skull, absolutely weaker than in C. familiaris of equiva-
lent size, and less expanded. The cranium is large and
full, while the postorbital constriction is well marked.
The sagittal crest is well defined. The very marked
medial frontal depression, seen in C. familiaris, is totally
lacking in this species.

The deuterocone of P* is very large and heavy and is
separated by a deep valley from the protocone. Nor-
mally P*? has no posterior tubercle, but this specimen

168 M. R. Thorpe—Oregon Tertiary Camde

shows an exceedingly small incipient one, while the poste-
rior basal tubercle is much smaller than in 7. ferox, but
practically the same asin the type. Merriam 1906 (p. 23)
says that 7. altigenis is characterized by the absence of a
posterior basal tubercle, but in the absence of other diver-
gent characters, I do not consider that this tubercle on
P? invalidates the identification with this species.

Measurements.
(CatsINo, 10065 Yo RM)

mm
Skull length, prosthion to occip. condyles, inc............ 205*
Diameter:ot brain2 Case wr. et eae ee ee eee 62
IDB MANS Ee OL OOSMOMIOCAL COMSGUCUICTN 2455 .cnccecousctace: Oona
Bizyeomatic- diameter s5 <a, ee ee ee 120
Length of molar ‘senies gue ce ee ee 20
Lemethe ot oP 4 sic 3 2g ae ae 19
Transverse diameter across deuterocone of P* ........... 14.5

* Approximate.

Mesocyon corypheus (Cope).

This species was very abundant in the John Day Basin
and is represented in the Marsh Collection by some excel-
lently preserved skulls with jaws. Cat. No. 12702, Y.P.M.,
collected at Turtle Cove, differs in some respects from the
type as described and figured by Cope. Itis about 10 mm.
shorter than the type, but the postorbital constriction is
one-third greater than that of Cope’s specimen, while the
latter shows a little greater bizygomatic diameter. The
muzzle is 24.5 mm. in minimum transverse diameter; in
the type it is 30.2 mm. The superior canine of the Yale
specimen is much more slender and recurved. ‘The supe-
rior tooth measurements correspond more closely to those
of No. 1888, Univ. Calif. Coll. Vert. Pal. (Merriam 1906,
p. 19) than to the type. The junction of the temporal
ridges lies 16 mm. posterior to the postorbital constric-
tion and in the type it is nearly above the constriction.

Cat. No. 12703, Y. P. M., also found at Turtle Cove, is a
skull and jaws, which, in the main, lack only the zygomata
and some of the incisors. This is a typical example of the
species. P,1is very much reduced; the posterior tubercle
of P, is fairly prominent, but the one on P, is very heavy
and prominent; the masseteric fossa very large and
deep.

with Descriptions of New Forms. 169

Measurements.
(Cat2Now 12703 "Y~ Pe M:)

mm
Length of ramus, angle to incisor border, inc. ........... 122
Length of inferior dental series with canine............. 78
Meneheor mierior premolar Series ¢ ...<. 55-266 ce ce ee ee 37
Men hivOnamerrOr molar Seles .. |. 2... sls sees cee es Dal
Pepiumonramus belowemiddle of Mi se.5. es. eee Wieo

Specimen No. 12704, Y. P. M., from Turtle Cove, is the
anterior part of a skull with some teeth. It is an old indi-
vidual, with the teeth much worn. These teeth are more
robust and heavier than in the type or in the other Yale
specimens of this genus. M? is nearly a third wider
antero-posteriorly and 1 mm. greater in transverse diam-
eter than the type M?. The individual tooth measure-
ments are close to those of M. brachyops Merriam, but
the total length from the posterior edge of M? to the poste-
rior edge of the canine is 2 mm. greater than in the type of
M. corypheus. Merriam (1906, p. 17) found a consider-
able variation in size within this species and the Yale
specimens apparently confirm this opinion. I suspect
that this latter specimen, Cat. No. 12704, represents a
male.

Mesocyon brachyops Merriam.

Cat. Nos. 12710 and 12719 of the Marsh Collection, from
Bridge Creek and Turtle Cove, respectively, consist of
maxille with parts of premolar and molar series. They
are from upper John Day strata.

Mesocyon josepht josepli (Cope).

A nearly complete skull with mandibles attached, Cat.
No. 12705, Y. P. M., is very close to the type. It was col-
lected in middle John Day strata in the Fossil Horse beds
on Cottonwood Creek. This species has been known here-
tofore only from an anterior part of a skull with teeth.
The description may be amplified as follows:

Specific characters About 12 per cent shorter in total
length than M. corypheus; muzzle short; infra-orbital
foramen above the interval between P? and P*; postor-

170 M. R. Thorpe—Oregon Tertiary Camde

bital constriction narrow; brain-case moderately large;
junction of temporal ridges slightly beyond postorbital
constriction; zygoma expanded; sagittal crest low; bulle
narrow and ovate; masseteric fossa large and deep, with
long axis nearly horizontal; coronoid low and angle
heavy. In the Yale specimen, P' is quite close to the
canine, a condition analogous to that in M. brachyops
Merriam. An individual variation of No. 12705 is the
possession of five inferior premolars in the right ramus.
The accessory premolar is of nearly equal size with P,.
The posterior basal heel of P, is robust, as well as the
posterior tubercle of the same tooth. P, has a small
posterior cusp and small basal tubercle.

Measurements.
(CCritE ING, IZ, WG, IB. INL)

mm.
Length of skull, prosthion to occip. condyles, inc. ........ 144
Bizyeomatic diameter (pce een ee eee 98
Diameter of postorbitaltconmsumicyionee seen ere 20
Diameterot braim-caseNmaxces. ape en eee eee 47
Beneth: of ramus, max... Seis See eee ee 105.5
Depthsor ramus below middlleson Wiser eee 19.3
Length of inferior dental series, with canine ............ 66
Bene thor interior premOlackserlcs meta ene 29
Leneth of inferior premolar series, with accessory premolar 32
ene thot aatkerion aml arise aie ses a eee 26

Mesocyon josephi secundus, subsp. nov.
(Fies. 1 and 2.)

Holotype, Cat. No. 10063 (skull), Y. P. M. Upper Oligocene (middle
John Day), John Day Basin, Oregon.

Distinctive characters.—Premolars crowded and no
diastemata between the molars and canine; P? obliquely
placed, much the same as the corresponding tooth in
Cynodesmus thooides Scott; P® has a small posterior
tubercle in addition to the posterior basal one; canine
more ovate in cross-section; molars and P* increased in
size and P? decreased.

In comparison with M. joseph, sensu stricto, the new
subspecies differs in the characters noted above, and in
the fact that the molars of the latter are about one third

with Descriptions of New Forms. JE
larger in both transverse and antero-posterior diameters ;
P+ is one fourth larger, P*? the same size, P? about one
fourth smaller, and P' alike in both. The total length of

dentition from M2? to I® inclusive is the same in both,
67 mm.

NY
\ ’ Ny \ NY
NV

. 2S

10063 TYPE
Fic. 1—Mesocyon josephi secundus, subsp. nov. Holotype. x 2/3.
Cat. No. 10063 is approximately the same length and
has the same bizygomatic diameter as M. drummondanus
Douglass, but it differs from the latter in the following
respects: (1) larger I?; (2) premolars crowded; (38) P*
longer but not wider; (4) P? obliquely piaced, and P?
smaller and ovate in cross-section; (5) transverse part
of maxillo-palatine suture lies on a line back of the deu-
terocone of P+, while in 47. drummondanus it is on this
line; (6) bullze more ovate and their long diameter less

oblique to the sagittal plane in M. drummondanus; (7)
smaller brain-case diameter.

\ Sa : JOOG3 TYPE
wi gwen 5) ee
: . \ lin \ : A i wi AQ 2 y) EE >»
eZ ee Wf Seer

Fic. 2—Mesocyon josephi secundus, subsp. nov. Holotype. x 2/3.
M. josephi secundus, M. josepli, and M. drummondanus
are closely allied and yet show clearly recognizable differ-
ences. No. 10063 and M. josephi josephi are middle John
Day and from the John Day Basin, while the type of the
other species was collected a little east of Drummond,

172 M. R. Thorpe—Oregon Tertiary Cande

near the Hellgate River, Montana, and is probably like-
wise of Upper Oligocene age.

Measurements of Holotype.

mm.
Length of skull, occip. condyles to prosthion ............ 143.5
Diameter or postorbinalaconstrichiony eee ae PAL,
Diameter of -brain=case,pimlax.. cea ee ee 46
lenethror superior molar Series ate ene See 15e5
leneth of superior: premolarsenies: eer ern ee 37

Pericyon socialis, gen. et sp. nov.
(Fie. 3.)

Holotype, Cat. No. 12715, Y. P. M. Both rami, permanent dentition,
unworn. Upper Oligocene (upper John Day), Haystack Valley, John Day
Valley, Oregon. Paratype, Cat. No. 12737, Y. P. M., from Turtle Cove.

= (RS
ONS

WP ints»

SS > SS
SSS Ses
as

Fic. 3.—Pericyon socialis, gen. et sp. nov. Holotype. x 2/3.

Distinctive characters—Slhightly smaller than Daphe-
nus vetus Leidy; dental formula (lower jaw) I8, Cl, P4,
M3: inferior premolars compressed; P, has posterior
tubercle but very small posterior cingulum; very prom-
inent posterior tubercle and large basal heel on P,; M,
has large and well developed metaconid and prominent |
hypoconid; M, with two anterior cusps, a low hypo-
conid and a basin-shaped entoconid; M, nearly three
quarters as long as M,, with a small paraconid from
which extends inward, backward, and finally outward
a faint semi-lunar ridge; masseteric fossa deep; inferior
border of ramus more curved antero-posteriorly than
that of D. vetus, but about the same depth below the tooth-
row in each.

with Descriptions of New Forms. 173

The major distinctions between this form and Para-
daphenus are: (1) much larger size than any species of
the latter genus; (2) M, absolutely very much larger,
for in Paradaphenus and Daphenus this tooth is a small
convex nub; and (3) presence of posterior cusp on P,.
From Temnocyon it differs in being smaller, with the heels
of M, and M, very much less trenchant; with both pro-
toconid and metaconid of M, on anterior half, and with
posterior cusp on P,. M, is not medially constricted as
in 7. altigenis, nor is the posterior cusp of P, so exter-
nally situated in this new form. MJesocyon has anterior
basal tubercles on P;, and P,,which are lacking in Pericyon.
The heels of P, and of M, are much larger than in Meso-
cyon, and the posterior cusp of P, is higher and much more
prominent. Mesocyon is smaller, with the premolars less
compressed.

From Tephrocyon, Pericyon is differentiated by the
much larger M,;; the different arrangement of the conids
and heel of M,; the much more marked lateral compres-
sion of the premolars, their greater degree of hypsodonty,
and the smaller posterior cusp on Ps, as well as the larger
size; the greater area occupied by the masseteric fossa
and the much less steeply inclined postero-inferior border
of the ramus below the coronoid process, the anterior bor-
der of which rises more nearly vertically in T’ephrocyon
than in the new genus. Tephrocyon is Middle Miocene;
Pericyon, Upper Oligocene.

Measurements of Holotype.

mm.
Mewetneor dentaluseries with Camine=., . cos os. eS le 85.5
UIST de OLE TNOENS EATS See nee eae mee a 35)
ewer OF premolar Genes... hss ee clots tee rac ek 40)
Wepimor canmis pelowenuddlesol Mey x. ia. Sees. es ce: 22
Pee OSs OU AMICG ETT Ol Me, ae Sa 8 i fst tie Mente os, te os we It

ei pO Sih olen rere ony Ws Seem te ree ee 11
Ant.-post. diameter of M,

Enhydrocyon oregonensis, sp. nov.
(Fies. 4 and 5.)

Holotype, Cat. No. 12730, Y. P. M. Skull lacking posterior and _ basi-
cranial areas of cranium. Upper Oligocene (middle John Day), Turtle Cove,
John Day Valley, Oregon.

This species is much smaller than EF. basilatus and E.

174 M. R. Thorpe—Oregon Tertiary Canide

sectorwus. I am inclined to view the latter two as prac-
tically the same or at most as being one a subspecies of
the other. In comparison with E. stenocephalus Cope,
the new form differs in having less expanded zygomata;
nasal bones differently shaped and pointed posteriorly,
terminating just anterior to the postfrontal processes;
frontal ridges uniting just beyond the postorbital con-
striction; muzzle 20 per cent longer; less interorbital
width; shorter length of superior dental series; longer
sectorial; considerably smaller M?; longer and wider M';
less axial length.

Fie. 4.—Enhydrocyon oregonensis, sp. nov. Holotype. x 2/3.

\

Ye eeaiViE

Sl
\ Zi

Word Za a y
SAY.

Lo i Cy
| il

) WN MONS ml
Fig. 5.—Enhydrocyon oregonensis, sp. nov.- Holotype. x 2/3.

vit

I ZX

The incisors of this new species are peculiar in that
they have accessory cusps on their internal and lateral
surfaces, with the exception of T°. I) has an internal
inferiorly bifurcated cusp, as well as a small external
lateral one near the tip, while I’, in addition to the inter-
nal bifurcated cusp, has a lateral cusp on each side, the
outside one being the larger and nearer the base. PP? is

with Descriptions of New Forms. 175

rotated inward, being placed at an angle of about 45°
from the sagittal plane. It has both anterior and poste-
rior basal tubercles and posterior cusp. P® is nearly
parallel to the sagittal plane, has a prominent posterior
cusp but small basal tubercles. P* is oblique to the sagit-
tal plane, its metacone is deep and wide, and the deutero-
cone is small. M' is typical of the genus. M? is much
smaller than that of E. stenocephalus and but slightly
larger than that of EF. crassidens. The infra-orbital fora-
men is above the extreme posterior edge of P? and is ver-
tically oval. 3

This species differs from EF. crassidens chiefly in
smaller size; wider P*; larger M?; different outline of
superior dental series; different incisor forms; smaller
deuterocone of P*; more posterior position of infra-orbi-
tal foramen; larger orbit; more robust malar below orbit;
and slightly more elongate face. ©

Measurements of Holotype.

mm.
Rene o1 skull partly estimated ... 1... - ue obec clase: 150
Midr OF-PoOslorbisal constriction... ..a 255.0622. 02k eo. 28
Keneih or superior dentition ....<......5.5. 5% ee oe 15
Wibmiameeshe IMCISOlS 7, a's ose ka cs heads dee wees s Bay, sce 2.9
Diameter of canine, ant.-post. 11 mm.;transverse....... 8.2
Diameter of P?, ant.-post. 8.7 mine= transverse). 6... 5)
Diameter of P*, ant.-post. 11.6 mm. transverse ....... ff
Diameter of P*, ant.-post. £9 2mm: transverse-s....... 11.6
Diameter of M’, ant.-post. 10.1 mm.; transverse ....... 15.5
Diameter of M?, ant.-post. Ae Mas GANS Verse? ..5 <2. « 4)

Tephrocyon rurestris (Condon).

This genus and species is represented at Yale mainly
by superior and inferior teeth of several individuals, all
from the Mascall formation of the John Day Valley.

REFERENCES.

Condon, T. 1902: The two islands and what came of them. Portland, Ore.

Cope, E. D. 1879A. On some characters of the Miocene fauna of Oregon.
Proc. Amer. Philos. Soc., 18, 63-78.

—1879B. Second contribution to a knowledge of the Miocene fauna of
Oregon. Ibid., 18, 370-376.

—1879C. Observations on the faune of the Miocene Tertiaries of Oregon.
Bull. U. S. Geol. and Geog. Survey Terr., vol. 5, 55-69.

Am. Jour. Sct.—Firtu Series, Vou. III, No. 15.—Manrca, 1922.

176 M. R. Thorpe—Oregon Tertiary Canide.

Cope, E. D. 1879D. On the genera of Felide and Canide. Proc.. Acad.
Nat. Sci., Phila., 30, 168-194.

—1881A. On the Nimravide and Canidz of the Miocene period. Bull.
U.S. Geol. and Geog. Survey Terr., vol. 6, 165-181.

—1881B. Miocene dogs. Amer. Nat., 15, 497.

—1883. On the extinct dogs of North America. Ihbid., 17, 235-249.

—1884. Tertiary Vertebrata, Book I. Rept. U. 8. Geol. Survey Terr., 3.

Douglass, E. 1903. New vertebrates from the Montana Tertiary. Ann.
Carnegie Mus., 2, 145-200.

HKyerman, J. 1894. Preliminary notice of a new species of Temnocyon, and
a new genus from the John Day Miocene of Oregon. Amer. Geol., 14,
320-321.

—1896. The genus Temnocyon and a new species thereof and the new genus
Hypotemnodon, from the John Day Miocene of Oregon. Ibid., 17, 267-
287.

Marsh, O. C. 1871. Notice of some new fossil mammals and birds from
the Tertiary formations of the West. This Journal (3), 2, 120-127.
Matthew, W. D. 1903. The fauna of the Titanotherium beds at Pipestone
Springs, Montana. Bull. Amer. Mus. Nat. Hist., vol. 19, 197-226.
—1907. A Lower Miocene fauna from South Dakota. Ibid., vol. 23, 169-

aU).

Merriam, J. C. 1903. The Pliocene and Quaternary Canide of the Great
Valley of California. Univ. Calif., Bull., Dept. Geology, vol. 3, 277-
290.

—1906. Carnivora from the Tertiary formations of the John Day region.

_ Ibid., vol. 5, 1-64.

—1913. Notes on the canid genus Tephrocyon. Ibid., vol. 7, 359-372.

—and Sinclair, W. J. 1907. Tertiary faunas of the John Day region.
Ibid., vol. 5, 171-205.

Scott, W. B. 1890. The dogs of the American Miocene. Bull. Princeton
Coll., 2, 37-39.

—1898. Notes on the Canide of the White River Oligocene. Trans. Amer.
Philos. Soe., 19, 327-415.

Wortman, J. L., and Matthew, W. D. 1899. The ancestry of certain mem-
bers of the Canide, the Viverride, and Procyonide. Bull. Amer. Mus.
Nat. Hist., vol. 12, 109-138.

R. W. G. Wyckoff-—Ammonium Chloride. 177

Art.- XIII.—The Crystallographic and Atomic Sym-
metries of Ammonum Chloride; by Ratpo W. G.
WyckorF.

Introduction.—lt was early observed that the maximum
symmetry of the arrangement of the atoms of certain
crystals, as determined from a study of their X-ray dif-
fraction patterns, was not in accord with the symmetry
assigned to these crystals as a result of the purely erystal-
lographic observations of the occurrences of faces and of
etch-figure formation. Typical of these apparent dis-
crepancies between the results of the formal crystal-
lography and of the X-ray investigations were potassium
chloride (sylvine) and cuprous oxide (cuprite). In both
of these instances, however, the symmetry of the arrange-
ment of the atoms was greater than that arising from the
erystallographic observations, so that it was always pos-
sible to avoid a direct conflict between the results of the
two methods by assuming that one or more of the kinds of
atoms entering into these compounds had about them
fields of force of such a shape as to impart to the atomic
ageregate the necessary lower degree of symmetry. In
view of the absence of any information concerning the
shapes of atoms such observations clearly had a certain
measure of justification.

The determination of the probable structure of ammo-
nium chloride, however, brings forward another case of
discrepancy which cannot be dismissed by similar assump-
tions. It is the purpose of this paper to show that not
only is the structure which has been assigned to this
crystal incompatible with its described symmetry; but
that there is no possible structure, possessing the sym-
metry required by the recorded erystallographic observa-
tions and at the same time permissible from the stand-
point of the chemistry of ammonium chloride, which is in
even approximate agreement with these X-ray data. As
a consequence it seems necessary to conclude either (1)
that the very considerable data relating to the symmetry
of ammonium chloride are incorrect, or else (2) that
information concerning the occurrences of forms upon
crystals and more especially the symmetry of etch-figures
does not in all cases furnish an indication of the internal
symmetry of the crystal. The second of these conclusions

178 =R. W. G. Wyckoff—Crystallographic and

is so difficult to practical crystallography that a further
investigation of the symmetry of ammonium chloride
seems imperative.

The observed Symmetry of Ammonium Chloride.—
Ammonium chloride is assigned to the enantiomorphiec
hemihedry of the cubic system both upon the basis of
observed face development and because of the symmetry
of etch-figures. Pentagonal icositetrahedral forms such
as (875) and (943) have been reported on erystals from
different sources. The type of etch-figure described as
occurring upon such faces as (211) points definitely to the
absence of any planes of symmetry.+

The X-ray Data upon Ammomum Chloride—Both
spectrometer measurements” and powder reflections? con-
clusively indicate that the ratio n?/m—1, where n is the
order of the reflection and m is the number of chemical
molecules within the unit cell. On this basis a ‘‘body-
centered arrangement’’ has been assigned to the nitro-
gen and chlorine atoms of the crystal. From the results
of the theory of space groups it is readily shown that if
there is thus one molecule in the unit, the placing of the
four hydrogen atoms associated with each nitrogen atom
requires that the symmetry of this arrangement be that
of the space group Ta’. According to this structure the
co-ordinate positions of the atoms in a unit cube are:

Che Cle ike
N: 000,

fs: Wee: tun; ua; tnu.

This structure is tetrahedral cubic, a degree of symmetry
incompatible with its recorded crystallographic charac-
teristics.

The diffraction data concerning ammonium chloride
which have been collected have been used for no other
purpose towards determining its crystal structure than to
evaluate the ratio of n?/m. From these data there is no

*A summary of the crystallographic information is given in Groth, Chem-
ische Krystallographie, I, 182 (Leipzig, 1906).

2 We SE and W. L. Bragg, X-rays and Crystal Structure, p. 110 (London,
1918).

7G. Bartlett and I. Langmuir, J. Am. Chem. Soc., 43, 84, 1921.

*It is readily shown by approximate calculations of intensity carried out
in a manner previously outlined (Ralph W. G. Wyckoff, this Journal, 2,
239, 1921) that all of the diffraction data are in accord with this simple

Atomic Symmetries of Ammonium Chloride. 179

immediate method of selecting between values of =
1, 2, 3, or 4, with corresponding values of m1, 8, 27, and
64, and it is natural to look to a more complicated struc-
ture arising from one of these for the apparently
necessary reconciliation between the symmetry of atom
arrangement and face development.

It has, however, already been intimated that in this par-
ticular instance no such more complicated arrangement is
available. This can be shown as follows. There is a
maximum of 96 equivalent positions in the unit for any
space group having an enantiomorphic cubic hemihedry.
This means that if we make the apparently chemically
necessary assumption that all of the nitrogen and all of
the chlorine atoms in ammonium chloride are alike, there
may be as many as 96 nitrogen and 96 chlorine atoms in
the unit. If furthermore these atoms lie upon elements of
symmetry their number within the unit will be some sub-

TABLE 1,

Special cases of the enantiomorphic cubic space groups.
_ Equivalent positions in unit cell

Group Twenty- Thirty-
One Four Eight Twelve Sixteen four two
re ESET TG i eae See =a
: ea) 1(1) eR. et ECS) oi
Bees Yo 1 sk sees Petal): P11)
(2 See eee, os 9 aa 1(1)
Ni SS Sa ee thE. 1 Ce kL) ae
Gress oD 1(1) GIG et CS) ==
eee 2 1(1) tela svi 3 teal (3) a
1 nee, 9 die G) aa

Explanatory Note: If a number in the preceding table is not followed
by a parenthesis, the positions in the corresponding arrangement are com-
pletely determined by the symmetry considerations; the parenthesized
values give the number of variable parameters possessed by the various
arrangements to which they refer.

structure for ammonium chloride. Thus these calculations yield for the
strongest diffractions that are recorded the following calculated intensities:

Intensities
Indices of Plane Order Estimated Caleulated
110 ie 10 3,050 arbitrary units
111 1 i 220
100 2 2 679
210 1 15 359
ZAL 1 3 1,690

180 R. W. G. Wyckof—Crystallograpme and

multiple of 96. In other words, the number of atoms of
nitrogen and of chlorine ina unit cube must be 96 or some
submultiple thereof. Neither 64 nor 27 are such submul-
tiples so that it must be concluded that either one or eight
molecules of ammonium chloride are associated with the
unit cube. These two groups of possible arrangements
are most readily studied with the aid of table 1.5 It will
be observed that there is no space group isomorphous
with O (the enantiomorphic hemihedry) having special -
eases which permit the placing of one molecule of ammo-
nium chloride within the unit. All possible arrangements
for the atoms of ammonium chloride that will have the
desired symmetry are as follows:

(2) From O*:

iL il 3b SB@Bito 6113} ik.ey83

N: 000; 5303 2 Ghar gee? wae? oe? 444

e 1

ils eos lta 4 Ue $00; 4449 Se GbaE 2) 4449 as
H: Thirty-two points one of which is we.

(3) From O°:
Ni: 000 ; 0243 £08 5 Z
Cl: LH; 440; 044; 4 F
H: Two groups of 16 eoteraler positions obtained by
assigning two different values of «to groups of points one of which
Is UUU.

(4) From O':
Hight equivalent positions:
uuu; uuu, Hud, Niu; “HU, udu; Huu, Ui.

we
wwe we

ING 22h O) == Wi,
Cle AG OF == Wen
H: At w= w,, and at 24 other positions which may
be chosen in various ways.
(5) From 0”:

Eight equivalent positions:
wu; uun; uuu; tnu; 4—u,4—u,t—Uu;
Jal 1. 1 1 ms
ULF F—UjU+F, F—-UjUTZ,U+F;
1 JE
ULF UL —U.
ING ue
Cl: At u = u,,
H: At wu =u, and at 24 other points which can be
chosen in several ways.

° This table and the discussion which follows is based upon material con-
tained in a book entitled ‘‘An Analytical Expression of the Theory of
Space Groups,’’ as yet unpublished. A similar table will be found in
P. Niggli, Geometrische Krystallographie des Discontinuums, p. 410.

Atomic Symmetries of Ammonwm Chloride. 181

(6) From O°:
Hight equivalent positions:

UU; UA- 55 = Ung Ub; a) Mate 1S Ee ee Ul tea G
UUU; UE, 9—U,UTE; pees i mas tea ut U4t-3 5 Ue

N: At u= wu,
@l: At v= wu
H: At w= wu,, and 24 other points which can be

chosen in different ways.

(7) An arrangement can be deduced from 0’ which is enan-
tiomorphie with that from (6). It does not require a more de-
tailed treatment.

Comparison of possible Structures with the Diffraction
Data—tThe diffraction effects to be expected from each
of these erystallographically possible structures® can be
calculated in the usual fashion’ and then compared with
the observed reflections from various planes. The scat-
tering power of the hydrogen atom is negligible compared
with the seatterings of the nitrogen and chlorine atoms, so
that they may be omitted from these calculations of
intensities. Such a comparison leads to the following
results.

Possibility (2): For planes whose indices are two even
and one odd there should be no reflection until the fourth
order. Planes such as (100) and (210) give second order
reflections, however, so that this structure must be ruled
out as impossible.

Possibility (3): This arrangement must be excluded
for the same reason.

All of the remaining arrangements lead to structures
which are chemically highly improbable. Particularly
(3) and (4) yield structures that are from the standpoint
of the chemist utterly inconceivable in that they group
together about one point in the unit all of the ammonium
groups and about some other point all of the chlorine
atoms. In view of the close agreement that exists
between the observed diffraction effects and those caleu-

* An arrangement of atoms having holohedral symmetry, but to which an
enantiomorphism might be ascribed by assigning to its atoms certain dis-
symmetries, could be developed from the space group 0* which would produce
the desired diffraction effects. In such a structure, however, though all of
the ammonium radicals were alike, four of the chlorine atoms would be dif-
ferent from the other four. From the standpoint of its chemistry such a
structure is scarcely permissible and may be eliminated upon such grounds.

“Ralph W. G. Wyckoff, this Journal, 50, 317, 1920; ete.

182 R. W. G. Wyckoff—Crystallographic and -

lated for the simple body-centered arrangement, (1), it
would be anticipated that the correct structure approxi-
mates it quite closely. Inspection, however, shows that
none of the arrangements can be made to approach this
erouping of atoms. ‘To discuss satisfactorily the diffrac-
tion effects from each of these possibilities it is necessary
to assign to each of.the variable parameters a range of
values covering the entire unit cell and to calculate there-
from the intensities of reflection from various planes.
These rather laborious calculations can be carried out
with the aid of a graphical method which is an obvious
extension of that one previously used in discussing the
various possible arrangements for the atoms in periclase
(MgO).® <A rigorous treatment of these arrangements is
scarcely attainable because of the somewhat qualitative
character of diffraction data and of the assumptions upon
which their discussion is based. It can, however, be
shown that none of these arrangements will furnish rea-
sonable structures which can account for the observed
diffractions.

Such, then, are the results which lead to the impasse
outlined in the introduction to this paper. The simple
structure which is in agreement with the X-ray data is
possessed of a tetrahedral hemihedral symmetry while the
crystallographic information points to an enantiomorphie
hemihedry. ‘There is, furthermore, no chemically per- —
missible arrangement of atoms, no matter how compli-
cated we may wish to choose it, which will account for
these diffraction data and will be capable of possessing in
any way an enantiomorphic hemihedry.?®

The issue thus appears to be sharply drawn. If pre-
vious observations upon crystals of ammonium chloride
are correct in giving it pentagonal icositetrahedral faces,
there seems to arise some uncertainty as to the value of
the symmetry assignments commonly made by erystal-
lographers upon their interpretations of face develop-
ment. The more or less obvious interpretation of the
sides of etch-figures as solution faces upon a crystal leads,
in its application to the present case, to the same con-
clusion. An attempt to account for this discrepancy =

* Ralph W. G. Wyckoff, this Journal, 1, 138, 1921.
®° This statement is true no matter what symmetry characteristics may be
assigned to the constituent atoms.

Atomic Symmetries of Ammonium Chloride. “183

attacking the assumed mechanism of X-ray diffraction is
not likely to yield results of value for the following rea-
sons. The datum’? which has been used here to establish
the impossibility of an enantiomorphic hemihedry for
ammonium chloride is simply the value of the ratio n*/m.
The correctness of this determination rests simply upon
the idea of atoms as diffracting centers for X-rays, upon
the correctness of the determination of the wave-lengths
of X-rays, which itself is a consequence of the application
of the quantum principle, and upon an approximately cor-
rect determination of the number of molecules in a gram-
molecular weight of any substance. As a consequence the
denial of the determined value of n?/m must involve the
denial of one of these fundamental factors, the general
validity of any one of which is reinforced from many
directions in physical science.

It is of great interest to all students of general crystal-
lography to know whether there are any discrepancies
between the external symmetry and the structure of erys-
tals. To this end a further study of the external sym-
metry of ammonium chloride crystals is now needed.

Summary.—lt is shown that not only is the symmetry
of the structure that has been assigned to ammonium
chloride in conflict with its observed symmetry but that
there is no other possible structure which will possess the
requisite symmetry. Asa result the necessity of further
study of the erystallography of this salt is urged.

Geophysical Laboratory,
Carnegie Institution of Washington,
November, 1921.

#0 The chemically requisite assumption that in a crystal chemically identical
atoms must be related in the same way to other atoms is of course likewise
involved.

184 Wyckoff—Crystal Structure of Silver Oxide.

Art. XIV.—The Crystal Structure of Silver Onde
(Ag,0) ; by Rate W. G. Wycxkorr.

Introduction.—Little is known of the crystallography
of silver oxide (Ag,O) other than that it crystallizes as a
powder in microscopic isotropic crystals and can be grown
as small octahedrons.! It is thus impossible to assign
silver oxide crystals to any particular one of the classes
of cubic symmetry.

Diffraction Data and the Number of Molecules within
the Unit Cell—Because of the difficulty of producing
other than microscopic crystals the available diffraction
data have been limited to those to be had from powder
reflections. Powder photographs obtained in the custom-
ary manner showed lines of marked intensity. Measure-
ments (upon two separate films) of the first four of them
—all that are required to establish the structure of this
erystal—are given in table 1.

The density of silver oxide? has been determined as
7.02. Because of the impossibility of immediately identify-
ing the different lines as reflections from particular planes
of atoms, the indices of the planes causing these reflections
and the number of chemical molecules to be associated
with the unit cell (or more precisely, the value of m/n?*)
can be obtained only by assuming different values for m,
calculating therefrom the spacings of simple planes in the
corresponding unit and seeing whether the observed
reflections can be identified in positions with some of these
possible planes. Since as m is varied the resultant spac-
ings change as the cube root, while the relative spacings of
different planes stand in the ratio of the square roots, it
should theoretically always be possible to make such an
identification uniquely.? A determination of m for silver
oxide by this procedure showed that if m==2 the reflec-
tions observed in the photographs correspond with simple
planes (see table 1). Since, furthermore this agreement
would be equally well obtained from any unit having the

1'Vogel, Poggendorffs Ann. 118, 145, 1863.

2 Biedermann, Chem. Kalender, 1, 53, 1914.

® The actual spectrum lines are so wide, however, that such a manner of
determining the number of chemical molecules in the unit need be entirely
satisfactory only in case the correct indices of the planes actually producing
the reflections are simple.

Wyckof—Crystal Structure of Silver Oxide. 185

same value of m/n?, where n may be chosen as any integer,
it is evident that m for Ag,O can just as well be equal
to 16.

TABLE 1. Measurements upon two Silver Oxide Films.

Distanee of line from

central image dnia/nin A. U. hkl digo As We
Film 1. Film 2. Rimi. Film 2. Filmi. Film 2.
2.86 em. 2.85 em. 275° eat ie 111(1) 4.775 4.779
3.30° 5.H0° 2.38° 2.38" 100(2) 4.770 4.770
4,72° AT a® 6 1672 110(2) 4.758 4.752
5.55 5.54" 1.428 ily eis 1NSGD)) 4.768 4.770

Mean 4.768 4.768

The value dj. is of course the length of the side of the unit cube.

The Possible Structures for Silver Oade—If m=2:
There is but one way of arranging two molecules of Ag,O
within a unit cell so that the resulting structure will have
any form of cubic symmetry.* It is as follows:

Arrangement (a) : Oxygen atoms at

OUR 22x

Silver atoms at
SMA 5 JER Sh 5) SS 10355 esishil
4445) 4449 44) 444°

This grouping can be deduced from the space groups
i ©? or Oi?:

If m= 16: If the two silver atoms in the molecule of
Ag,O are chemically (and hence erystallographically)
alike, there are three ways of arranging sixteen molecules
of Ag,O in a unit cube so that the resulting structure has
cubic symmetry.® They are:

Arrangement (b): In this arrangement, which can be
deduced from the space groups T:*,0* and On’, the six-
teen oxygen atoms lie in unique positions; the thirty-two
silver atoms possess one variable parameter and are upon
trigonal axes.

*These results are taken from tables which form a portion of a book
entitled ‘‘An Analytical Representation of the Theory of Space Groups’’
which is shortly to be published by the Carnegie Institution of Washington.
Some of them are also given in P. Niggli, Geometrische Krystallographie des
Discontinuums (Leipzig, 1919).

* If it were assumed that the two silver atoms could be different, then the
several space groups having as special cases sixteen equivalent positions with
one variable parameter would introduce additional possibilities. Such an
assumption finds no justification from the chemistry of the compound and
would run counter to the general trend of crystal structure determinations
thus far made.

186 Wyckoff—Crystal Structure of Silver Oxide.

Arrangement (c): This grouping, deducible from the
space group O °, is simply a twice scale enlargement of
arrangement (@). |

Arrangement (d): In this arrangement, derived from
021°, the oxygen atoms lie in unique positions, the silver
atoms upon trigonal axes.

The Choice of the Correct Structure.—An approximate
calculation of the intensity of reflection from any plane
for any of these possible structures can be made by the
method already outlined. For this purpose use was
made of the expression

Loe (Ae Baa) ) nee
where [is the intensity of a line in the powder photograph,
p is the number of planes of the reflecting form, d/n is the
ratio of the spacing of the reflecting planes to the order of

the reflection and (A? + B?) has the meaning prewea
given to it.

Arrangement (a): For this grouping
If nis odd:

When the indices of the reflecting plane are 2 odd
and 1 even,
A= 205 RB ==.08
Indices: 2 even, 1 odd,
Ac = 05. B = 0K
Indices: all odd, mS.
A=0,B=++4 Ag.

If m is even:

Indices: 2 odd, 1 even,
A 20 aA eB 1 0r
Indices: 2 even, 1 odd or all odd,
A = 204 4 Ag, — when n = 2, + when
=A abi <O.
In these computations O and Ag, the scattering powers
of oxygen and ailver, respectively, are taken equal to the
atomic numbers. A comparison between the estimated
intensities of the four principal lines of the photograph
and the intensities of these lines calculated for this
arrangement is given in table 2; the other planes appear-

-° Ralph W. G. Wyckoff, this Journal, 2, 239, 1921; Jour. Wash. Acad. Sci.
11, 429, 1921.

Wyckof—Crystal Structure of Silver Oxide. 187

ing in the photograph fit about equally well. The agree-
ment is as good as could be expected in view of the uncer- °
tainties involved in these calculations.

TABLE 2.
Intensity (arbitrary units)
Indices Estimated Calculated for arrangement(a)
111 (1) 10 77,800
100 (2) | ‘h 34,900
110 (2) 7 43,200
113 (1) 6 50,500

Arrangements (b), (c) and (d).—Because of its iden-
tity with (a), grouping (c) requires no further treatment.
Because of lack of knowledge of the laws of scattering,
an entirely satisfying elimination of (b) and (d) can
searcely be made with these data. It can, however, be
shown that they are both improbable.

These observations then have shown that there is but
one simple structure, (a), the ‘‘cuprous oxide structure,’’
which is in agreement with powder measurements
obtained from silver oxide. Furthermore they have
shown that neither of the two more complicated group-
ings which are possible structures from the determination
of the value of the ratio m/n? are probable. In all likell-
hood, then, the atoms of silver oxide are arranged as
defined by (a).

It is of interest that silver as oxide agrees with its
analogue, monovalent copper, in departing from the per-
haps more common cubie arrangement of two atoms of
one kind and one of another, the ‘‘caleium fluoride struc-
ture.’’ Since arrangement (a) can be deduced from space
groups possessing either paramorphic or enantiomorphic
hemihedry, or holohedry, this determination of the struc-
ture of silver oxide has not succeeded in defining uniquely
its crystallographic symmetry. On the basis of recorded
face development it has been customary to assign to
cuprous oxide, with the same structure, an enantiomor-
phic hemihedry. It has been suggested’ that this enantio-
morphism could be explained if one or both of the kinds
of atoms occupied positions slightly displaced from that
of (a). Since in the arrangement of their atom positions,

*W. H. and W. L. Bragg, X-rays and Crystal Structure, p. 157 (London,
1916).

188 Wyckoff—Crystal Structure of Silver Oxide.

not only the simple grouping (a) but likewise all of the
‘ more complicated groupings possess holohedral sym-
metry, it is evident that, assuming as we have that the two
silver atoms of Ag.O are equivalent, this explanation is
untenable. If a real enantiomorphism exists in the case
of cuprous oxide, it must be traced to some other cause
than symmetry of the arrangement of the atoms.

Summary.— Employing the method of powders, it is
shown that silver oxide has the same structure as that
assigned to cuprous oxide. The length of the side of the.
unit cube is determined to 4.768 A. U.

Geophysical Laboratory,

Carnegie Institution of Washington.
December, 1921.

E. W. Berry—Carboniferous Plants from Peru. 189

ART. XV.—Carboniferous Plants from Peru; by Epwarp
W. Berry.

Just south of the port of Pisco the peninsula of
Paracas, celebrated in the War of Independence,
juts out into the Pacific, forming a bold wind-and-wave-
swept headland (Lat. 13° 55’ S., Long. 76° 33’ W.). It is
about 220 km. south of Callao and 25 km. southwest of the
port of Pisco, and is of great geological interest since it is
largely made up of continental Carboniferous sediments
and constitutes one of the very few deposits of this char-
acter in South America, and the only known occurrence of
rocks of this age on the West Coast of South Africa.

The outerop of coal-bearing rocks on Paracas was dis-
eovered by F.. C. Fuchs, who published a brief account?
of it in 1900. Fuchs made a considerable collection of the
fossil plants, which are now in the Museum at Lima,
where I had the privilege of seeing them. He identified the
following forms: Calamites suckowu, Sphenopteris har-
tle beni, Lepidodendron sternbergu, Sigillaria tessellata,
Stigmaria ficoides and Baiera pluripartita, and consid-
ered the deposit to be of Upper Carboniferous age.

The true Sphenopteris hartlebena of Dunker, which has
since been referred to Ruffordia goepperti, is a character-
istic species of the Wealden, and the Paracas form which
Fuchs thought represented this species is Palmatopteris
furcata, a rather widespread Carboniferous fernlike plant
of the Sphenopterid group. Fuchs’s Bacera pluripartita
is not a Baiera but a species of the genus Eremopteris, his
Sigillaria I was unable to identify from an inspection of
the material, and as there are no true Sigillarias in my col-
lections [ cannot say whether his specific name is correct
or not. His Lepidodendron is not Lepidodendron stern-
bergu but represents both of the species recorded from
this locality in the present paper. His Calamites and
Stigmaria appear to be correctly determined.

The searcity of coal at tidewater on the West Coast
aroused a great local interest in Fuchs’s discovery, per-
tenencias were quickly taken out and much money was

1A fully illustrated account of this flora will be published at a later date.
George Huntington Williams Publication, No. 9.

?Fuchs, F. C., Nota sobre el terreno carbonifero de la peninsula de Para-
eas, Bol. de Minas Industria y Construcciones, tomo 16, No. 7, Lima, 1900.

-

190 EZ. W. Berry—Carboniferous Plants from Peru.

spent in putting down concrete-lined shafts and in dia-
mond drilling, although the complete section is admirably
exposed along the south shore of the peninsula. Pros-
pecting has not resulted in economic development, since
the coal seams are thin and the coal contains a prohibitive
percentage of ash.

Subsequently Fuchs gave a brief account of the geology
of this general region in his paper on the copper deposits
around Ica and Nazca.* In this report he considers that
the Carboniferous at Paracas is continuous to the east-
ward beneath the Pleistocene and recent deposits that
form most of the surface of the country in the belt lying
between the igneous rocks of the Western Range of the
Andes and the present coast, 1. e. the Pampa de Condor,
Pampa de Chunchanga, Pampa de Pisco and Tablazo de
Iea. Another account of the Paracas Carboniferous with-
out, however, adding anything to our knowledge concern-
ing it, was published by Dorea in 1909.4

Steinmann, who did not visit Paracas, saw Fuchs’s col-
lection at the Cuerpo de Ingenieros de Minas in Lima. In
a short note published in 1910 he makes the interesting
generalization that the plant-bearing Carboniferous in
South America is Lower and the invertebrate-bearing
Carboniferous is Upper in age.’ He states, with his usual
assurance, that the Paracas forms represent:

Archaeocalamites radiatus.
Lepidodendron ef. Veltheimi.
Lepidodendron ef. Volkmanni.
Sphenopteris affinis = S. furcata.
Rhodea filifera.

Rhabdocarpus.

None of these forms is represented in the large collec-
tions which I made at Paracas, nor did I see any of them
~in the Fuchs collection.

Finally in 1917 in Lisson’s admirable compilation of
Peruvian fossils® the two species Lepidodendron rimosum
and Lepidodendron obovatum are listed from Paracas,
the determinations being by the late Professor Zeiller,

° Fuchs, F. G., La Regidn Cuprifera de los Alrededores de Ica y Nazca;
Cuerpo de Ingenieros de Minas, Bol. 29, Lima, 1905.

*Dorea, I. R., Estudio sobre los Yacimientos Carboniferos de Paracas,
Bol. Soe. "Ingenieros, vol. 11, pp. 104-130, Lima, 1909.

5 Geol. Rundschau, Bd. 1, p. 50, 1910.

°Lisson, C. I., Edad de los Fésiles Peruanos y Distribucién de sus
Depésitos en Toda la Republica, pp. 20-21, Lima, 1917.

E. W. Berry—Carbomferous Plants from Peru. 191

and the age is given as Westphalian. Both of these
species are represented in my collections.

The question of the age of the Paracas deposits and
their relation to the widespread Carboniferous limestones
of the Andes is one of great importance. it was appar-
ently some conversation with Senor Bravo, the Director
of the Cuerpo de Ingenieros de Minas regarding the Para-
eas continental Carboniferous that was the basis for the
beautiful diagrams of the relations of land and sea during
Carboniferous time published by Bowman,’ which he
unfortunately located at Pacasmayo which is 805 km.
north of Paracas in a region of crystalline rocks.

There was not time for a detailed study of the Carbon-
iferous of Paracas during my visit, necessarily short since
the peninsula is a practically uninhabited desert. The
wide and desert coastal plain, interrupted only by the
irrigated valley of Ica, that extends from the igneous
foothills of the Western Range to the ocean, consists of
wind-blown sands, desert pavement gravel, and paper
shales. Similar deposits form the neck of the Paracas
peninsula, which is thus the result of the block-faulted
Carboniferous and apparently bears no relation to west-
erly spurs from the Andes or igneous intrusions in the
Coastal Plain.

Following is a measured section of the easternmost
fault block and was repeated in the next block to the south-
west... The horizons from which fossil plants were col-
lected are indicated and there is no chronologic change in
the flora from top to bottom although fossil plants are
more varied in the lowermost horizon.

Thin to heavy bedded rather coarse greenish-gray sandstone, 13
Greenish-gray massive and crossbedded sandstone with vary- -
ing amounts of shaly interealations with Lepidodendron.
(le LLC ZAC EOE Bae pa aS one nee aries aa nse eco Fm (5
Dark shale with sandstone layers less than a foot in thick-

VESEY BPS a le, CEL Ie eke et co Sam art Ort is ea ea D0
Sandstone with a shale parting in the middle.............. 30
Sandy and somewhat carbonaceous fossiliferous shale...... 80
“SATE SIV R ees Sgt 1 ee eee ec a ety ol Sok pee 2D
Massive sandstone passing into thin-bedded sandstone along

CLE RIA SURES UTC i eae “nl Dib PA aya Did) eae ean Re ee 30

‘Bowman, I., The Andes of South America, p. 19.
* Measured by Professor Joseph T. Singewald, Jr.

Am. Jour. Sct.—Firra Serres, Vor. III, No. 15.—Manrcu, 1922.
14

192 H. W. Berry—Carbomferous Plants from Peru.

Massive greenish-grey somewhat arkosic sandstone......... 15
Thin-bedded sandstome 52 is ea oe ek ele 3 eee 10
Thin-bedded sandstone with dark shale (no coal) toward the

top Gunnel am-shaile)—. a4 cee Saher Oe etek? ets eee on

Alternating sandy and carbonaceous shale with 6 inch coal

at top; abundanthy tossiliterous s(iummel) a... cone 38
Massive greenish-eray somewhat arkosic sandstone........ 20
Shale carbonaceous above, sandy below (N. 20 H, 25 E) (tun-

WEL). 9 epee als acs ee Re eg ce ee ee 10
Greenish-eray sandstone, thin-bedded above.............. 5
Gray shale ae Er Ce RE ei oa ee 2
Interbedded thin sandstones and shales .................. 3
Shale . iid. GV ee OS ee ee ee t
Gray ‘arkosic Sandstone cnc aek) 8 ees. Se eG eee 5)
Sandy shalle < us saccades Ries eee ee eee 18
Thin-bedded sandstone and sandy shale with more massive

sandstone. at ase + <acue onl: seegeebcese babes sR ao < Aoeo oe sale ee eee 70
Dark carbonaceous shale with 2 to 3 inches of coal at top,

abundantly fossiliterous (Ni o , 25 1) (iamarel ) 0. eee 18
Massive greenish sandstone 0%... «eo... 2 oe 15
Sandy Shale ic NSS i 5 ee en ee On eee 13
Pleistocene o85

Section repeated by faulting to the southwest.

The materials are relatively coarse throughout and
would seem to indicate rapid deposition. Between 53 and
04 per cent of the total thickness is described as sand-
stone, which 1s often coarse and arkosic.. Of the 273 feet
described as shale 192 feet are distinctly sandy, so that less
than 14% of the total thickness, ineluding the so-called
coal seams, is fine-grained shale and even the coal con-
tains much silty impurities. No underclays with rootlets,
or upright stems were observed and the coaiy layers have
every appearance of having been formed of drift material,
which also appears to have been the case in the probably
contemporaneous and very similar continental Carbon-
iferous examined on the Copocabanya Peninsula on Lake
Titicaca in Bolivia.

The Flora.

The flora comprises the following forms:.
Pteridophytes or Pteridospermophytes.

1 Palmatopteris furcata (Brongn. )
2 Eremopteris whitei Berry.
3 Eremopteris peruianus Berry.

E. W. Berry—Carbomferous Plants from Peru, 193

Arthrophyta—Calamariales—Calamariaceae.

4 Calamites suckowii Brongn.
5 Calamostachys sp.

eR a eminem arageae
Lepidodendron rimosum Sternb.
: Lepidodendron obovatum Sternb.
8 Lepidophyllum sp.
9 Lepidostrobus sp.
10 Stigmaria sp.
11 Knorria sp.

It is thus extremely hmited, although some of the ele-
ments are exceedingly common, and this is especially true
of Palmatopteris furcata, Eremopteris whiter, Calamites
suckowu and Lepidodendron rumosum. I imagine that
the coarseness of the sediments and the apparent drifting
of the material is mainly responsible for the absence of a
more representative flora. Thus the present collections
contain no traces of Sigillaria, Cordaites, Sphenophyllum,
calamite foliage, nor of any Neuropterids, Pecopterids,
Alethopterids or Lonchopterids. It is this feature of the
flora which is undoubtedly responsible for the opinon of
Steinmann, quoted on a preceding page, that the Paracas
flora is of Lower Carboniferous age.

A somewhat similar situation is furnished by the flora
of the Kuttung series of New South Waies, where the
so-called Rhacopteris flora appears to be wholly lacking
in Neuropteris, Alethopteris and Pecopteris. Walkom,
however, and quite rightly I believe, correlates this Kut-
tung flora with the nee stage of the Kuropean
section.

The nearest known occurrence of marine Carboniferous
in Peru is at Huanta, Dept. of Ayacucho, 277 km. N.E. of
Paracas. The only known outcrop of the marine Carbon-
iferous west of the Western Range is at Cotahuasi, Dept.
of Arequipa. This is about 150 km. inland from the pres-
ent coast and about 395 km. S.K. of Paracas.

Continental Carboniferous, mostly of unknown age, is
more widely distributed in South America than has been
suspected. ‘Thus there are some continental sediments in
the lower part of the section on the Copacabanya peninsula,
and Titicaca Island, at the Cerro de Lacctacucho, Sicuani,
Dept. of Cuzco, at the Pongo Mainique on the Urubamba,

194 EF. W. Berry—Carbomferous Plants from Peru.

Dept. of Cuzco. Steimmann (Geol. Rundschau, Bd. 2, pp.
50-51, 1910) records Lepidodendron and Rhacopteris
inaequilatera Goppert from near Huichaycota on the Rio
Huallaga in the Kastern Range 1 km. south of Huanuco in
the Department of that name. . In the same publication
similar deposits, said to contain Archaeocalamites radia-—
tus and Lepidodendron cf. Volkmanni are recorded from
Retamito, which is between San Juan and Mendoaz in the
Argentine Cordillera.

1 think that there can be no doubt that the Paracas
Carboniferous is younger than the Dinantian stage of
the Kuropean section and that it corresponds to the West-
phalian stage. The marine Carboniferous of the Andes
is usually considered to be of Uralian age, that is to say,
Stephanian in terms of the continental section. Whether
or not the marine series represents more than Uralian has
not yet been definitely determined. The finding of one at
least of the Paracas plants in the lower part of the section
on the Copacabanya peninsula, Bolivia, several hundred
feet below the fossiliferous limestones, would lead to the
inference that the Paracas plant-bearing Carboniferous
is older than the bulk of the marine Andean Carbonifer-
ous and especially the highly fossiliferous portion of the
latter.

Johns Hopkins University,
Baltimore, Md.

W. Inndgren, etc—A New Mineral. 195

Arr. XVI.—WMelanovanadite, a New Mineral from Mina
Ragra, Pasco, Peru; by Watprmar Linperen, L. F.
Haminton and CHar es PaLacHE.

Introduction.—Late in 1920 Mr. W. Spencer Hutch-
inson, consulting engineer for the Vanadium Corpora-
tion of America, brought to the attention of the senior
author three specimens of a mineral collected by him at
Mina Ragra, Peru. He suspected that it was a new min-
eral, and this opinion was proved correct by chemical and
optical examination. The formula appears to be 2Ca0O.
2V.0,.3V.O0, and the name of Melanovanadite is proposed
for it, in allusion to it being practically the only vanadium
mineral of a deep black color.,

DescriptionThe mineral was collected from No. 1
tunnel in the lower part of the great patronite deposit
at Mina Ragra, below the strongly oxidized zone and
90 feet below the surface. This tunnel extended out
into the surrounding black shale from the so-called
Veta Madre, or main, lower section of the deposit. Mr.
Hutchinson believes that the ore is not conformable with
the dip of the slates but forms a crosscutting body. The
mineral occurred along fractures in the shale, not far
from the main body, and there was enough of it to be
mined as rather high-grade vanadium ore. One of the
specimens showed a compact brownish gray shale rich in
gypsum. On a joint plane were flat radial or divergent
aggregates of a black lustrous mineral of prismatic habit
and near these the shale was blackened.

The best specimen, about 12 by 8 by 6 em., is composed
of a soft porous black material evidently altered shale
traversed by brecciated openings or fractures coated by
closely massed divergent and velvety bunches of small
acicular black crystals, with a maximum length of 3 or 4
mm. These bunches are easily removed and a quantity
of about 2 grams was obtained which, for purposes of
analysis, was hand-picked under the binocular micro-
scope. ‘The product was almost pure, containing only
a few particles of altered shale and of pascoite. Reg gar d-
ing the paragenesis and the nature of the altered shale
see below.

*The ending ‘‘vanadite’’ is an obsolete form of ‘‘vanadinite,’’ but
there can scarcely be any objection to using this form in the present case.

196 Lindgren, Hamilton, Palache—Melanovanadite

Crystal Form.—tThe greater thickness of the needles is
about 0.5 mm. ranging down to 0.1 and 0.01 mm. The
color is black, luster almost submetallic, streak very dark
reddish brown, the hardness is 2.5, the specific gravity
3.477 at 20° C. The habit of the crystals is prismatic,
parallel to c, with monoclinic symmetry. The principal
faces consist of a flat striated prism, the longer diagonal
being parallel to the b-axis, minor pinacoidal faces and
usually well-developed terminal faces of pyramids and
smaller domes. There is a perfect cleavage parallel to
the clinopinacoid. Although small the crystals can be
measured.

Optical Properties —Under the microscope the crystals
remain black except in very thin prisms which are trans-
lucent with brown color. Flat cleavage pieces parallel
to the clinopinacoid only become translucent when the
thickness is about 0.003 mm. and then show maximum
extinction of about 15°. Resting on the prism the erys-
tals become brown translucent with a thickness of about
0.03 mm. and then show lower extinctions of 12° to 13°,
while these resting more nearly on the orthopinacoid
extinguish at lower angles ranging down to 0°. The
perfect cleavage being perpendicular to the poorly devel-
oped orthopinacoid extinctions of 0° are rarely seen.
Exact optical measurements are difficult on account of
the deep color. Obscure hyperbole show on the prism
faces and it is probable that the plane of the optie
axes 1s perpendicular to the perfect cleavage (010). The
absorption is very strong. The a ray is visible with dark
yellowish brown color through the prism faces and the
orthopinacoid and has according to a determination
kindly made by Professor C. H. Warren a coefficient
of refraction of a little less than 1.74; while the 8 and y
rays are somewhat higher but cannot be measured exactly
on account of the strong absorption. The @ and y rays
therefore lie in the 010 plane of perfect cleavage and
their absorption is so strong that such cleavage pieces
only become translucent in extremely thin plates, with
dark reddish brown color, @ and y differing slightly in
depth of tint. The double refraction is strong.

Composition.—An analysis by L. F. Hamilton gave the
following composition:

a New Mineral from Peru. £97,

ECT S.Ct ee oe 52.61
Wiha Lec Boi! Se Beltey top eae ir ty ae an Dono
OSLO ST a a oa ee a 9.89
Gees eee eT oe MEN MOE 27
See ONS Ue Oe ee en ae 1 89
SEUSS ogee De he ahd anette antec en Bese eee 1.66

99.66

The calcium was determined by fusion with sodium
carbonate and leaching with sodium carbonate water, leav-
ing CaCO,;, MgCO, and Fe(OH), insoluble. The residue
was dissolved in HCl; Fe(OH), was removed with
ammonia, calcium precipitated as CaC,O,, filtered, dis-
solved and titrated with standard permanganate. MgO
was determined in filtrate by usual process.

For total vanadium the fusion and titration method
outlined by A. H. Low in his ‘‘ Technical Methods of Ore
Analysis,’’ 8th ed., 1919, p. 279, was followed. The van-
adyl was determined by direct. solution of mineral in
H,SO, and in atmosphere of CO, followed by titration
with standard permanganate. V.O; was obtained by
difference.

The mineral had been iret for several months in a dry
warm atmosphere, the analysis being completed in Jan-
uary, 1921. All determinations were made in duplicate
and some repeated. The determinations gave an actual
figure of 0.42% for total water on a sample of about 0.12
gram, which is only a gain in weight of a chloride tube of
0.0005 grams or an allowable error in weighing a tube of
10 to 12 grams capacity.

Assuming the alumina and iron oxides and magnesium
and silica to be impurities from the admixed shale, the
analysis may be recalculated to:.

Wee Reeteh satel dees 55. 54.90
STC Sie Se ten ea 34.78
EUS So oO a ee ee ae ee 10.32
100.00
From this we ealeulate
z page
8 IES Ae eee nT Ta) = 0-302
bas Re pe Wi ee ee ee er BeOS js = a209}
166
(
Ec i Mare od pe eae Ss een
D6

Nosy Oy ee ae. —— 303 2209 3 184

198 Lindgren, Hamilton, Palache—Meclanovanadite

or approximately 3:2: 2. The formula would, therefore,
be
2 Ca0.3 V,0,.2 V,O,

The calculated composition of this would be

W Oe, 25 Se ica eed ra ioe ror tee ern ata 59.16
Vi Oe Sel Seo age ee nn Sa a genie era 30. oF
(OP © perrmmmnin UDbrant nm rit rar ok eh ay SINE SL Dy Gao Sal 11.30

100.00

Before the blowpipe the mineral fuses very easily to a
brown liquid but gives no flame reaction for calcium.
It gives strong bead reaction for vanadium. It is
easily soluble in dilute HNO,, HCl, and H,SO,, with
apple-green color. In concentrated HNO, it dissolves
with evolution of red fumes to a dark brown solution.
Upon evaporation it gives a red or brown residue. It is
also easily decomposed by KOH with brownish color.

In May, 1921, a small sample of the mineral was sent to
Dr. W. T. Schaller of Washington. The material had
then for some time been exposed to a warm damp atmos-
phere. Dr. Schaller stated in a letter that the mineral
appeared to contain much water. In view of this, the
following re-examination was undertaken. A dry tube
test showed at once that this was now true, an abundance
of water being given off. The larger part was driven off
upon drying at 105°, but equally large amounts were lost
in desiccator above sulphuric acid. A sample in desiccator
lost 10.26% H,O in twelve days; the weight was taken for
six more days, but remained constant. The dried mate-
rial took up water rapidly after being removed from the
desiccator, even while it was being weighed.

The most representative results showed a loss of 10.7%
H,O upon drying at 105° C. About 5.9% was held above
105°, making a total water content of 16.6. The exact
point at which this last water was given off was not deter-
mined on account of the very low fusion point of the
mineral. |

A sample carefully cleaned and finely ground and dried
for two days at 105° C. gave the following figures:

a New Mineral from Peru. 199

Total vanadium as V,O. 86.07 .
5 33.48
Vv. 50) 49.38
CaO 10.65
He,O; =-- Al,O, 1839)
H,O-- 5.90
100.80

Si0, and MgO not determined.

It is certain that this mineral has hygroscopic qualities
very similar to the uranium micas and is strongly influ-
enced by atmospheric moisture or dryness.

In regard to the 5.9% held above 105° C. the results
seem difficult to harmonize, but we have the definite asser-
tion of the chemist that when first analyzed in the winter
of 1921, the mineral was practically anhydrous. This is
supported by the senior author’s statement that the min-
eral then gave no water in closed tube. Just what the
condition of the material was when formed in the deposit
is not easy to say. In view of the uncertainty still
attached to the part played by water in this mineral we
have decided to let the formula stand as based upon the
first and complete analysis.

Paragenesis.—In places the black acicular crystals are
coated with a yellow powder, and sometimes larger
masses of an orange-yellow substance cover the terminal
edges or are embedded between the black prisms. This
orange mineral, which is easily soluble in water, was
identified as pascoite (Ca,V,O,-.11H.O) and agrees in
its optical character with that substance. It is clearly
a product of hydration of melanovanadite.—[The above
by W. Lindgren. |

It remained to examine more closely the dark altered
shale upon which the bunches of melanovanadite were
attached.

Quite generally, this material gave a stong vanadium
reaction and also gave off more or less SO, in open tube.
It also contained organic matter. To test for the pres-
ence of native sulphur the powdered material was di-
gested with CS,. Upon evaporation the solution yielded
a resinous product but no sulphur. The residue was
washed with alcohol and leached with water. It gave a
yellowish solution containing vanadium doubtless from

200 Lindgren, Hamilton, Palache—Melanovanadite

finely divided pascoite. The residue gave a_ strong
sulphur reaction in open tube and then dissolved in
HNO, giving a green color and strong vanadium reaction.
It was “assumed, therefore, that a sulphide was present
and to determine it if possible many polished sections
were made.

The polished seprods showed in a dark gray gangue
one or two minute grains of pyrite, several minute grains
of hard greyish w ite color, not definitely determined. ~
Some bright scales of native copper were also observed.
The vanadium was present in the form of two minerals.
Minute prisms of melanovanadite were distributed
through the mass, clearly replacing the shale material
(lar oely eypsum). There were also veinlets of the same
substance. In oblique reflected light the melanovanadite
shows a blackish, submetallic luster, but in vertical illu-
mination it is rather bright grevish white with a tinge of
purple. It is very soft and gives a brown streak and is
instantly blackened by HNO,;, HCl, and KOH. Included
-in veinlets and masses of melanovanadite are irregular
masses of a metallic soft mineral of a brighter gray color
which undoubtedly is a vanadium sulphide though I am
not at all sure that it is patronite. The minute size and
fine intergrowth of the minerals tend to make a definite
identification difficult. The original vanadium mineral
was then a sulphide. Oxidation converted this to a blue
vanadyl sulphate (Minasragrite) which even now is abun-
dantly present in the mine water. Reaction with CaCO,
in the water and further oxidation produced the calcium
vanadyl vanadate, melanovanadite, and the hydration of
the latter resulted in the orange-colored pascoite. The
process was carried out in the lower part of the oxidized
zone.

The real patronite is a black mineral of fine grain and
metallic luster. In polished section it is shown to consist
of a very fine-grained mixture of three minerals. The
first and most abundant constituent is purplish gray and
is sometimes developed in rudely prismatic, thick forms.
The other two constituents are respectively light gray and
darker gray in color, and are developed in irregular
grains. All three minerals are doubtless sulphides.

The description given by Davy and Farnham (Micro-
scopical Determination of Opaque Minerals) does not

a New Mineral from Peru. 201

apply to the pure patronite but to some intermediate
product of alteration. It is believed that Hillebrand’s
analysis of patronite was made on partly altered material.

Crystat Form or MELANOVANADITE (by C. Palache).
System Monoclhine.
Amal Ratio a:b : c=0.4737 :1:0.5815 B—88° 371,’
Wo NAAT gg, = 0.581 w=_'88° 3714"

d

Four crystals were measured on the two-cirele goni-
ometer. They proved little suited to accurate measure-
ment, the prism zone being badly striated and the
terminal faces, while apparently bright and plane, giving

202 Lindgren, Hamilton, Palache—Melanovanadate

faint and often multiple signals. The figure reproduces
the habit of most of the crystals.

Description of the forms:

b (010) is always present, generally as a narrow,
bright face. The cleavage parallel to it is so perfect that
it is difficult to detach or even handle a crystal without de-
veloping cleavage surfaces.

a (100) is represented on every crystal by a rather
broad curved surface which gives a train of reflections,
none of which is from a true face. :

c (O01) is not present as a face. The inclination of
the clino-axis, to which it would be parallel, was calculated
from the other terminal faces present.

h (230) forms narrow but distinct faces on every
erystal. Its faces are far more perfect than other forms
of the prism zone except b.

(530*) is the dominant prism form; its faces are
broad but badly striated and their position-angles show
wide variations. .

Poor reflections were obtained from numerous other
prism faces but were too uncertain to be regarded as
indicating definite forms. None were found at all near
to the position of the unit prism which would naturally
be expected to occur rather than the two prisms with
rather complex indices given above.

p (111) is the dominant terminal form. Its faces
are generally bright and often developed to such an ex-
tent as to reduce other terminal faces to mere lines.

s (121) was observed but once as a minute face; it
has therefore been omitted from the figure.

d (032) is always present and its faces have a better
quality than any others on the crystals.

g (012) was present on each crystal measured but:
with faces so small as scarcely to be measureable. It
was. not figured therefore.

w (101) is quite large on most crystals but its faces are
always curved so that the position-angles are widely
variable.

The table contains the observed angles with their limits
of variation and the angles calculated from the derived
elements.

a New Mineral from Peru. 208

No. of Measured Limits Calculated
faces
® p ? p p p
SS ee SSS
b (010) 6 00°34’ 90°00’ 00°00’— 1°12" 00°00’ 90°00
h (230) 10 955 00 90 00 54 12 — 5d 49 54 3% 90 00
Ptaad) C74 30-90 0073 32 — 75°40 74 08390 00

w (101) 4 —90 35 50 17 87 05 — 95 35 49°56’ —50°33’—90 00 50 17
PeGieyes 2233516) 255 0-144 441595 —17 104 13 16 16
wes yro 1 53 41 82 0-36 =. 4 23 41 24 —41 42. 1 34 41 064
p (111) % 65 34 54 00 64 51 -— 66 51 53 28 —54 55 65 05 54 05
s (121) 1 —45 00 59 33 — 45 59 59 09

* This prism should probably be replaced by the prism with the simpler
indices (210). The latter gives excellent zonal relations with the terminal
planes which is not true of (530) The crystal has been drawn as though
the form were (210) which requires an angle ¢ of 76° 41’. The observed
angles showed none as large as this, however, and the exact symbol of this
form must remain in doubt.

204 P. FE. Raymond—Ceratopyge Fauna

Arr. XVII.—The Ceratopyge Fauna m Western North
America; by Percy Ki. Raymonp.

The determination of the boundary between the Cam-
brian and Ordovician systems in the West has proven as
difficult a problem as it is in the Kast. The following
contribution to the discussion of this question is based
primarily upon a collection made by Mr. Francis P. Shep-
ard, a graduate of Harvard, in the course of the investiga-
tion of the structure of the Rocky Mountain Trench.

The determinable fossils in his collections, from four
localities, in the Columbia River Valley, B. C., are:

Iingulella moosensis Walcott, Hystricurus tuberculatus (Wal-
L. ? allam Walcott, eott),

Obolus mollisonensis Walcott, Cyphaspis ? brevimarginata
Eoorthis desmopleura (Meek), Walcott,

Eoorthis sp. ind., Symphysurus cleorus (Walcott),
Dalmanella hamburgensis(Wal- 8S. elongatus (Walcott),

eott), Hemigyraspis meconnella Ray-
Syntrophia nundina Walcott, mond,
Raphistoma nason Walcott, IH. carvbouensis (Walcott),
Arthrorachis sp. ind., Megalaspis shepards sp. nov.

Menocephalus sp. ind.

This is a good Ceratopyge fauna, comparable to that
which is found at the base of the Ordovician at many local-
ities in Europe. It lacks both Euloma and Niobe, but as
is well known, there is great variation in the composition
of this fauna in the different regions in which it is found.
Broegger long ago? called attention to its presence in the
Kureka District, in the lower part of the Pogonip. More
recently Walcott* has noted the same zone in British Col-
umbia, but listed only a very few species. The importance
of this fauna in the determination of the boundary
between the Cambrian and Ordovician is so great as to
justify a review of its distribution in western North
America.

_ * Not to burden this paper with detailed descriptions of fossils, Megalaspis
shepardi may be characterized briefly as follows: Known from pygidia only.
That shield elongate, sub-triangular, pointed behind. Axial lobe not prom-
inent, contracted near the middle, with about six faintly defined rings.
Pleural lobes nearly smooth, with one pair of obscure ribs. Shell strongly
punctate. Holotype collected by Mrs. Francis P. Shepard and now in the
Museum of Comparative Zoology.

* Nyt. Mag. for Naturvidensk, vol. 35, p. 229, 1896.

* Smithson. Misel. Colls., vol. 57, No. 7, p. 229, 1912.

in Western North America. 205

Eureka District, Nevada.

The faunas of the lower part of the Pogonip are but
unperfectly known, yet there is more evidence of the Cera-
topyge zone than was noted by Broegger. Lllenurus eure-
kensis is a Symphysurus, and Dikelocephalus finalis an
Apatokephalus, as pointed out by Broegger. Symphy-
surus? goldfussi Walcott belongs to the genus Ceratopyge
itself, and Asaphus caribouensis is a Hemigyraspis (cran-
idium and free cheek only; the pygidium associated is that
of Symphysurus eurekensis). Ptychoparia (Huloma)
affinis is, however, probably not a Huloma.

The ranges of the fossils in the Pogonip are not very
definitely established. Hague gives the thickness of the
formation as about 3,000 feet near Hureka, and at White
Pine over 5,000 feet. Near the base are such important
species as Syntrophia nundina, Hemigyraspis caribouen-
sis, and Symphysurus eur ekensis.

‘‘Several hundred feet higher,’’ the fauna contains,
among others:

Schizambon typicalis, Dalmanella hamburgensis, Syn-
trophia nundina and Apatokephalus finalis.

From Hague’s descriptions, it may be judged that the
lower several hundred feet, and probably more than a
thousand, hold the Ceratopyge fauna. Receptaculites,
Maclurites and other genera suggestive of Beekmantown
appear about midway in the Pogonip, according to him.
In one list, Receptaculites, Syntrophia nundina and Sym-
physurus are found together and other lists indicate a
mingling of Ceratopyge and Beekmantown faunas to
within 800 feet of the top. This indicates a much greater
vertical range for this assemblage than is known any-
where in Kurope. In the lower part of the Pogonip, there
is a Strong admixture of Upper Cambrian species.

Utah.

Walcott has described a section in the House Range,
Millard County,’ western Utah. The highest formation
of the Upper Cambrian is named the Notch Peak. No
fossils were found in place, but a drift-boulder supposed

* Walcott, Mon. 8, 1884; Hague, Mon. 20, 1892, U.S. G. S.
* Smithson. Miscl. Colls., vol. 53, p. 173, 1908.

206 P. E. Raymond—Ceratopyge Fauna

to have been derived from strata in the upper 640 feet,
contained the following species (names corrected to agree
with later publications by Walcott) :

Eoorthis desmopleura (Meek), Schizambon typicalis Walcott,
Agraulos sp. ind., Solenoplewra ? sp. ind., Tsinama cleora
Walcott.

The last species is really a Symphysurus. Walcott®
proposed the genus 7'sinania for trilobites of the type of
Illaenurus canens Walcott, a Chinese species which has a
cranidium similar to that of Symphysurus, but more
nearly hemispherical, and a pygidium which is nearly as
long as wide. This latter feature is the most distinguish-
ing one for the genus, and is not shared by any American
form. Tsinama cleora, as shown by Walcott’s figures,
has the short pygidium of the typical Symphysurus.

The Blacksmith Fork section in the Wasatch Mountains
of northern Utah was also described by Walcott.’ Both
the Lower Ordovician, and the upper 190 feet of the
St. Charles formation, referred to the Upper Cambrian,
contain the Ceratopyge fauna, although some of the more
characteristic trilobites are lacking. Important species
in the St. Charles are: Hoorthis desmopleura (Meek),
Syntrophia nundina Walcott, Schizambon typicalis Wal-
cott, Menocephalus sp. ind., and Illaenurus |[Symphy-
surus| sp. md.

According to Tomlinson,? Beekmantown fossils are
present in strata 1,260 feet above the top of the St. Charles
formation.

Richardson’ has deseribed briefly a section in the Ran-
dolph quadrangle in Northern Utah. This is east of the
section on Blacksmith Fork, and the Cambrian sequence
is the same.

Resting on the St. Charles is the Garden City limestone,
with a thickness of 1,000 feet,'which contains a fairly large
fauna, partially identified by Kirk. Although the aspect
is in general that of the Beekmantown, one can find sug-
gestions of the Ceratopyge fauna.

Above the Garden City is the Swan Peak quartzite, 500
feet thick, from which some fossils have been obtained.
Among them Kirk has identified Symphysurus goldfussi

° Smithson, Misel. Colls., vol. 64, No. 1, p. 43, 1914.
7 Smithson. Miscl. Colls., vol. 53, p. 191, 1908.
Sour Geol wole 25,1917.

°* This Journal, vol. 36, p. 406, 1913.

in Western North America. 207

Walcott. Since this is a Ceratopyge, the identification,
if correct, would be of the utmost importance.

British Columbia.

I first suspected the presence of the Ceratopyge fauna
in British Columbia when I described from isolated speci-
mens which had long been in the collection of the Cana-
dian Geological Survey, the trilobite Hemigyraspis
meconnelli.t® It was collected three miles east of Golden,
and it is interesting to find it in Mr. Shepard’s collection.

Professor Daly found, 2 miles west of Donald Station,
B. C., two trilobites of this fauna which have been
described by Walcott'! as Dikelocephalus ? dalyi and
Tsinama elongata. The latter is a Symphysurus, and
was found also by Mr. Shepard.

Professor J. A. Allan has given the name Goodsir to a
formation, 6,000 feet in thickness, which overlies the
Upper Cambrian m the Ottertail Range, and occurs on
both sides of the Beaverfoot Range. The strata which
are found on the eastern side of the Columbia Valley at
Golden belong to this formation.'”

From the lower portion of it, at a locality 10 miles
southeast of Leanchoil, B. C., Allan collected a small
fauna from which Walcott described’* the following new
species: Obolus mollisonensis, Lingulella moosensis, L.
(?) allani, and Ceratopyge canadensis. While there can
be no doubt that this is a faunule of the Ceratopyge zone,
the trilobite is hardly sufficiently like the European repre-
sentatives of the genus to be included in Ceratopyge.
Walcott states that ‘‘This species differs from Cerato-
pyge forficula Sars in the greater length of the frontal
limb of the cranidium, longer palpebral lobes, and nar-
rower fixed cheeks. The pygidium differs most in having
a short median lobe, broader border, and the side spine
springing from the first instead of the second segment.’’

All this is true, and it should also be added that the
olabella is not definitely outlined, and tapers forward in
C. canadensis, while it is strongly outlined, truncated in

* Geol. Sur. Canada, Mus. Bull. vol. 1, 1913, p. 41.

4 Smithson. Miscel. Colls., vol. 57, 1914, p. 367 and ibidem, vol. 64, p. 228,
1916. :

Sum. Rept. Geol. Sur. Canada for 1912, 1914, p. 171.

** Smithson. Miscl. Colls., vol. 57, p. 229, 1912.

Am. Jour. Sci.—F irra Series, Vou. ITI, No. 15.—MarcuH, 1922.
15

208 P. EB. Raymond—Ceratopyge Fauna

front, and expands forward in. the typical Ceratopyge.
The Canadian form has no glabellar furrows, while C. for-
ficula always shows at least the posterior pair. A most
important difference is in the course of the facial sutures
in front of the eyes. In C. forficula they diverge and
quickly intersect the margin, whereas in C. canadensis
they turn but little outward, then inward, so that the
anterior part of the cranidium is pointed, as in Isotelus.

The position and origin of the spines on the pygidium
are fundamentally different from those of Ceratopyge. In
C. forficula these spines are, as Dr. Walcott has said, the
continuation of the second segment, but they are firmly
welded into the pygidium, so that they appear as projec-
tions of the margin. In C. canadensis the spine-bearing
first segment of the pygidium is so nearly severed from
it, that it would be almost as true to say that the last seg-
ment of the thorax was laterally extended into spines, as
that the first one of the pygidium was spinose.

It is difficult to place this trilobite. The general aspect
is that of the Asaphidae, and were there not 10 (or 11)
segments in the thorax, one would be inclined to refer it
to Asaphellus. The most nearly allied form which I have
seen described seems to me to be Dolichometopus (Housia)
varro Waleott, from the Upper Cambrian of the House
Range in Utah.1* This trilobite is, unfortunately, itself
imperfectly known and of uncertain affinities.

In order to avoid giving a new generic name, however,
I would suggest that the trilobite from the Goodsir be
called Housia canadensis, unless it can be shown to be
generically distinct from the type of Housia:—neither is
very suggestive of Dolichometopus.

In the Robson Peak district on the border between Brit-
ish Columbia and Alberta, Walcott'® has referred the
Robson limestone to the Ordovician and found at the base
an assemblage in which ‘‘there is a commingling of the
lower Ordovician and Upper Cambrian faunas.’’ From
Billings’ Butte, north of Mount Resplendent, he has
identified, among others:

Eoorthis desmopleura (Meek), Syntrophia nundina Walcott, Agnostus Sp.,

Triarthrus sp. ind., Peltwra (pygidia), Apatokephalus, and two species of
Illaenurus [Symph ysurus).

“ Smithson, Miscl. Colls., vol. 64, p. 374, 1916.
45 Smithson, Misel. Colls., vol. 57, p. 336, 1913.

in Western North America. 20

This is a very good Ceratopyge fauna, and the presence
of Peltura is especially interesting.

SUMMARY.

It appears that with the exception of Symphysurus, the
trilobites characteristic of the Ceratopyge fauna, such as
Ceratopyge, Apatokephalus and Hemigyraspis are not
sufficiently abundant to be very useful as guide fossils,
although diagnostic when present. Among the brachio-
pods, Syntrophia nundina and Eoorthis desmopleura are
ubiquitous and easily identified, so that an association of
Symphysurus with these may be used as a guide to the
fauna. On this basis, it seems that the Ceratopyge fauna
occupies a position at the base of the Ordovician from
Robson Peak in Canada southward to the Eureka District
in Nevada, on the western side of the Rocky Mountains. It
is somewhat doubtful if it will be found in that part of
Montana east of the Continental Divide, but further south
it may extend across New Mexico and Texas to the Appa-
lachians, although no trace of it has yet been reported.
In the east it is known only in Pennsylvania, New Jersey
and New York, until the northeastern Atiantic coast is
‘reached. There seem more possibilities of a connection
across the southern than the northern part of the conti-
nent, however, judging from the distribution of the Beek-
mantown. :

One very important point is the notable admixture of
Upper Cambrian fossils found in the Ceratopyge zone,
especially in Nevada and the Mount Robson district. In
no case is there either physical or faunal evidence of emer-
gence between Upper Cambrian and Lower Ordovician.
Richardson has outlined some physical evidence of an
unconformity between the St. Charles and Garden City
formations in the Randolph quadrangle in Northern
Utah, but since the upper part of the St. Charles has a
Ceratopyge fauna, and the lower part of the Garden City
avery similar one, the unconformity is in the wrong place.

Walcott'® has tentatively suggested that the lower
Pogonip contains a representative of the ‘‘Ozarkian,”’
and has described Calvinella tenuisculpta, a member of
the Dikelocephalidae, stating that it was probably from

* Smithson. Miscl. Colls., vol. 57, p. 359, 1914.

210 P. E. Raymond—Ceratopyge Fauna

about the same horizon as C. ozarkensis.. C. tenmsculpta
was found associated with Apatokephalus finalis, Sym-
physurus eurekensis, and other typical fossils of the Cera-
topyge zone. Saukia marica (Waleott) and Osceola
osceola (Hall), the only other described Dikelocephalids
in the Kureka district, are found in the Dunderberg shale,
beneath the Pogonip, associated with what appear to be
Upper Cambrian faunas.

Kither the Ozarkian (Ulrich, non Ee Coateye 1s not repre-
sented in this area, or else it is represented by the zone
with Ceratopyge. The latter alternative would be in
accordance with my view of twelve years ago,'? when I
urged that the Ceratopyge zone of the Kastern states
might be joined with the Ozarkian. Later studies in the
Appalachians have however convinced me that so far as
Eastern North America is concerned, there is no Ozarkian ©
System, and the evidence of the Ceratopyge fauna
indicates the same for the West.

Museum of Comparative Zoology,
Cambridge, Mass.

7 This Journal, vol. 30, p. 344, 1910.

S. W. McCalltie—The Pitts Meteorite. 211

Art. XVIII.—The Pitts Meteornte; by S. W. McCatuiz,
State Geologist of Georgia.

The fall of the Pitts meteorite here described was the
most striking phenomenon of its kind heretofore observed.
in the State of Georgia. The meteorite fell in a negro
settlement in the western part of Wilcox County near the
town of Pitts about 9 o’clock a. m. (Hastern time) April
20, 1921. At the time of the fall no clouds were in view
and the sun was shining brightly. It was seen as far
north as Sunny Side, in Henry County, 36 miles south of
Atlanta, and as far south as Moultrie, in Colquit County.
In addition to the above towns that appear to mark the
north and south limits of its visibility, it was also seen at
Camilla, Albany, Seville, Cordele, Hawkinsville, Perry,
Macon, and Alma. It was no doubt plainly visible over
an area of several thousand square miles, and could have
been distinctly seen by fully a quarter of a million people
had they been looking in the proper direction.

My attention was first called to the occurrence by press
notices on April 21 and on the 22d I received a specimen
of the meteorite from Colonel W. H. Dorris, of Cordele,
Ga., accompanied by a short description of the phe-
nomena. On April 24 I visited Pitts, with a view of
securing at first hand all data possible concerning the
exact locality, the attending phenomena, ete. The citi-
zens of the town rendered invaluable service in securing
the information desired, and also obtained for me for
examination and study all of the fragments of the meteor-
ite except one, which specimen was later secured from the
owner by personal request of Governor Dorsey. Several
hours were spent in the vicinity of the fall interviewing
eyewitnesses of the phenomena and in making a diagram
showing the relative position at which the fragments
struck the ground.

The Pia

The attendant phenomena witnessed by the observers of
the meteorite were similar to those noted in meteorite
falls in general and were as follows:

(1) The rapidly moving fire ball was the first phenom-
enon that attracted the attention of the observers. It was
described by eyewitnesses at Albany as a rapidly moving

212 S. W. McCallie—The Pitts Meteorite.

body about the size of a man’s head, appearing in the sky
in a northeasterly direction. At Moultrie it was referred
to as a brilliant body moving downward in zigzag course,
looking as if it might fall in the northern part of the city.
At Sunny Side, more than a hundred miles from the place
where it fell, it was seen ina southeasterly direction, fall-
ing nearly perpendicular at a rapid rate. :

(2) The dense smoke in the wake of the flaming fire ball
was referred to by the Albany and Moultrie witnesses as
a luminous trail following the flaming ball. Colonel Dor-
ris, who was in the vicinity of Pitts, speaks of the smoke
as a zigzag trail, hLngering for some minutes and assum-
ing various shapes. These shapes were thought by some
to be in the form of letters. Several persons in the imme-
diate vicinity of the fall described the smoke as white or
gray in color, and in the form of puffs and very dense.

(3) The first sound heard was compared to that of
thunder, and to many it was the first warning that any
unusual occurrence was taking place in the sky above.
At Cordele, 15 miles west of Pitts, the sound resembled
that of a heavy explosion, distinctly heard by several peo-
ple on the street. In the country, four miles east of
Cordele, two terrific explosions were noted, louder than
thunder, which so terrified the farm hands that they ran
frightened to their homes. At Hawkinsville it was
thought that an aeroplane had exploded above the city.
In the immediate vicinity of Pitts the sound was described
as several loud explosions, causing the earth to trem-
ble, followed in quick succession Py a number of lesser
explosions.

(4) The roaring and whizzing noise and the impact of
the falling fragments were heard only in the immediate
vicinity of the fall.

Description of Individual Fragments.

The location and relative distribution of the points at
which the fragments (four in number) of the Pitts meteor-
ite fell are shown on the accompanying diagram. The
largest piece weighing 57 ounces fell within less than 75
feet of Nancy Brinson’s house where it was dug up a few
minutes later still warm, but not red hot as first reported.
The fragments entered the freshly plowed sandy soil to a

S. W. McCallue—The Pitts Meteorite. 913

depth of about 16 inches forming an inconspicuous hole
less than 18 inches in diameter and scarcely half so deep.
The fragment was irregular, rhomboidal in shape, the
three greater dimensions being 5.7, 3.2, and 2.3 inches
respectively. More than two-thirds of the surface showed
the natural pitted characteristics of an iron meteor-
ite coated with black iron oxide through which in places
were to be seen shining patches of silvery white nickel-
iron. The remaining parts of the surface were rough and
angular with more or less sharp projecting points show-
ing evidence of recent rupture from other fragments.
This part of the surface was more or less smoked but it
had not the thick coating of the other part of the surface.
-The fragment was made up almost entirely of nickel-iron
throughout which, in irregular masses or veins, occurred
the stony material. The latter consisted mainly of gray
minerals interspersed with an occasional greenish gran-
ule. Polished surfaces of the iron portion of the frag-
ment when treated with dilute nitric acid showed the
typical Widmanstatten figures.

A second fragment fell by the roadside within a hun-
dred feet of Jim Harden’s house which is 700 feet south-
east of the Brinson house. This specimen buried itself
about 8 inches in the ground. It weighed 4214 ounces and
differs from the fragment above described mainly in
showing more stony material and in being more irregular
in shape. It also showed less of the naturaily pitted sur-
face but correspondingly more of the freshly fractured
surface. This specimen fell within three feet of a negro
boy walking along the road, who was covered with flying
earth. |

The third fragment fell about 4,000 feet southwest
of the second fragment and within 100 feet of where a
negro man and a boy were working in a cotton field. Only
a part of this specimen was seen by the writer as it had
been cut in pieces. However, judging from the fragments
it probably weighed less than 30 ounces. It entered the
ground only about seven inches and like the other frag-
ments was warm when dug up.

The fourth fragment was picked up in a public road
approximately 5,000 feet southwest of fragment No. 1.
No one saw this fragment fall, nevertheless it was at once
recognized by the finder who had seen other fragments.

914 S..W. McCalie—The Pitts Meteorite.

This specimen was irregular, pear-shaped, and weighed
less than two ounces. The proportion of stony material
in this fragment seems to be greater than in any of the
others but it is otherwise similar. This was the only
fragment that did not bury itself in the ground which is
accounted for by its falling on the hard road surface.

An attempt was made to fit the different pieces together
but without success indicating that only a part of the
meteorite had at that time been found.

Fic. 1. Diagram showing locations of the falls of the different fragments
of the Pitts meteorite. 1. Nancy Brinson’s house. 2. Jim Harden’s house.
3. King’s field. 4. Slater’s house.

By examining the diagram, it will be seen that four
fragments which had been found at the time of my visit,
were scattered over an area approximately a mile long
and possibly a quarter of a mile wide. It will further be
noted that the heavier fragments travel at a greater dis-
tance than the smaller ones; this indicates that the meteor-
ite was moving in a northeasterly direction, which fact
was confirmed by several observers.

S. W. McCallie—The Pitts Meteorite. 915

Mineral Composition.

The composition of the meteorite, as shown by analyses
made by Dr. Edgar Everhart, acting chemist, of the State
Geological Survey is as follows:

Analyses of Pitts Meteorite.

Constituents Stony Part Metallic Part
We tre Se ert Sat 09 91.50
“LIE E SSS 2s pease Sieg ann ae ee ee 32 6.67
umes pn kph eens OO Sy 20 04
( DESTE eS igncee re eeer 00 trace
PRM MEE EE Sich les Rl thee 4 00 1.40
ST Sh AE ures De eae ee 28.30 02
emia Bettie ee Sta eS trace A5
PMC SEN pyre Pete OGG SIS]. . He ek OO 05
Manganese oxide (MnO) ............ ea
2 LE oS RSS Re are na eee ese lial
Uy DEES. CATR Ea peeeae tears siete Oemlaear 32
“22 E ERS OO ae tones PP Se eee 5.96
_: JUUEDINSGY Bel Alpe phe aeg ey se gee eee ee 30
Merrous Oxide. 2. se a 65.52
Phosphorus: pentoxide ...... 0.0.0.0... OT
eee hrs NI Eee Aes babs 8.46
Carbon by difference ............... 2.32

Reckoning the sulphur and iron as pyrrhotite (eS) the
stony material contained 77.62 per cent of pyrrhotite, the
rest, omitting carbon, corresponding most nearly to
hypersthene. }

Dr. George P. Merrill of the National Museum, who
made an examination of the stony part of one of the
fragments, advised me that by optical and chemical
methods he was able to make out the following minerals:
olivine, diopside, and a plagioclase feldspar.

Fully 90 per cent of the four fragments was metallic,
specific gravity 7.23.

Atlanta, Georgia.

216 Scientific Intelligence.

SCIENTIFIC INTELLIGENCE
I. CuHermistry AND Puysics.

1. The Reduction of Ferric Salts with Mercury —ULrRoy W.
McCay and Wiuuiam T. ANDERSON, JR., of Princeton University,
have applied this method of reduction to the volumetric deter-
mination of iron with very satisfactory results in regard to con-
venience and accuracy. It appears that the fact that metallic
mercury will reduce ferric chloride to the ferrous salt has been
known since 1842, and that Bovar suggested its analytical appli-
eation in 1911, but the method seems to have attracted little
attention.

The present investigators carried out the reductions by shaking
solutions of ferric chloride containing hydrochioric. acid with
about 20 ce. of pure mercury in a long, narrow, glass-stoppered
bottle of about 800 ec. capacity. The shaking should be vigorous,
so as to break up the mercury into tiny globules, and it should be
continued for about 5 minutes. After filtering the liquid and
washing the residual mercury and mercurous chloride the iron
was titrated, either by the Zimmermann-Reinhardt modification
of the permanganate method or by the dichromate method, with
excellent results as shown by the test-analyses.

The authors found that solutions of ferric sulphate containing
sulphuric acid were only partially reduced by the operation, but
that when a little more than the theoretical amount of sodium
chloride was present the reaction was complete with the forma-
tion of mereurous chloride.

It appears that this rapid and accurate method is worthy of
extensive practical application, for it has the advantage over the
use of zine, for, instance in the Jones reductor, in the fact that
titanic acid is not reduced by mercury, and aiso because the
mercury process seems to be freer from accidental sources of
error.

The authors have found that when shaken with mereury in
the presence of hydrochloric acid reduction takes place with fer-
ricyanides, chromates, molybdates, vanadates, and antimonates,
and they are investigating some of these reactions.—Jour. Amer.
Chem. Soc., 43, 2372. H.. lis We

2. Synthetic Gasoline —A German process for making gaso-
line, called the Burgess process, has been developed by the orig-
inators of the Haber process for the fixation of atmospheric nitro-
gen by combining it with hydrogen. In carrying out this process
hydrogen is passed over carbon at 200 atmospheres pressure and
at a temperature of 700° C., whereby hydrocarbons are formed.
It is stated that the process has been examined by an American
engineer, who has reported that the method is in actual operation

Chemistry and Physics. DMG

and is producing motor fuel on a small scale. If this process
should succeed in competing with gasoline from petroleum it
would be of great industrial importance.—Jour. Indus. Eng.
Chem., 14, 165. H. lL. W.
3. The Precipitation of Arsenic as Sulphide from Solutions
of Arsenates.—It has been found by J. H. Rerpy that this pre-
cipitation is greatly hastened by the presence of a small quantity
of a soluble iodide in the acid solution. This catalytic effect is
doubtless due to the reduction of arsenic acid to arsenious acid or
ehloride by hydriodie acid and the continuous conversion of the
iodine thus set free to hydriodie acid by the hydrogen sulphide.
This observation has been made use of in connection with work
in qualitative analysis by adding one or two cubic centimeters of
normal ammonium iodide solution just before passing in hydro-
een sulphide in the regular course of analysis. The complete
precipitation of pentavalent arsenic is thus facilitated, and the
chances are much lessened that the student wiil fail to detect
arsenic because of non-precipitation at this point. The apparent
complications that mercury and copper may be precipitated as
iodides are rectified by the digestion with ammonium polysul-
phide which changes the iodides into the higher sulphides.—
Jour. Amer. Chem. Soc., 48, 2419. eels Wis
4. Colorimetric Analysis; by F. D. SNELL. 12mo, pp. 150.
New York, 1921 (D. Van Nostrand Company ).—In this book an
attempt has been made to collect for ready reference all the color-
imetric methods which experience has shown to be practical. The
first three chapters deal very satisfactorily with a classification of
the methods, the apparatus used and the ecaleulation of the
results; then the processes are described individually, according
to the substances to be determined. The book is a useful one for
reference use by analytical chemists, and while it appears that
some of the methods are not described with sufficient clearness
and fulness of detail, it may be said that references to the litera-
ture are usually supplied. As an example of lack of detail in
the description of a process it may be pointed out that the direc-
tions given for the determination of small amounts of potassium,
involving the precipitation of the platiniec chloride, would not
lead to a proper precipitation and collection of this salt, but of
course, an experienced analyst would known how to modify the
directions. H. lL. W.
Dd. Fluid Resistance.—The phenomena which occur in the
motion of a real fluid are so complicated that theory alone is
capable of dealing with none but the simplest cases, and progress
in practical applications, as for example in the law of resistance to
the motion of a body through a fluid, can only be made by careful
experimentation. The following is a report of a study by
C. WIESELSBERGER of the resistance experienced by a cylinder

218 Scientific Intelligence.

when struck by a current of air directed at right angles to its
axis. The expression which has been commonly used to formu-
late the resistance to motion in a fluid is F = 1/2 cApv? where v
represents the velocity of the body, A the area of the outline of
the body projected on a plane normal to this direction, p the
density of the fluid and c is a coefficient to be experimentally
determined, and presumably depending only on the nature of
the medium and the geometric form A, but independent of
its absolute magnitude and of the velocity v, in which case the
formula once completed would be applicable to geometrically sim-
ilar bodies and different velocities in the same medium. It is
however now known that ¢ is constant only for similar flow lines
and these may differ even with geometrically similar figures.

The condition for geometrically similar flow lines was first
determined by Osborne Reynolds and requires that the expression
R = vpd/p» shall be the same for cases to be compared. Here p
is the viscosity, d, some similar linear dimension of the body and
v and p have the previous signification. It is thus evident that c
will not in general be a constant but will vary with v, and the
size and shape of the body. The smallest departure is found to
occur in the case of bodies with sharp edges which stand at right
angles to the direction of the current, as for example, flat disks
met perpendicularly by an air stream. Here, for a considerable
range in the values of F the coefficient ¢ remains constant at
about 1.1. On the other hand bodies with rounded convex sur-
faces show notable variation. When & is small compared to
unity the resistance, as in Stokes’ law for falling spheres,
increases linearly with the velocity. ,

The aim of the author’s experiments was to trace the connec-
tion between c and F# throughout a large range of the Reynolds’
number. They were carried out upon cylinders of circular cross-
section which were suspended transversely to a plane current of
air, 1. e., the stream lines were the same in all planes at right
angles to the axis of the cylinder. The air velocities ranged from
1.2 to 36 meters per second which is from about 4 to 80 miles an
hour. Nine eylinders were used varying from .05 to 300 mm. in
diameter so that the range of the Reynolds’ number investigated.
was from 4.2 to 800,000. The resistance for the small cylinders
was measured by weighting them and observing. the pendulum
displacement from the vertical. In the case of the larger ones
the force was determined by the aid of a system of threads con-
nected to a balance. Any disturbance due to the effect of the
ends was corrected by a proper experimental arrangement. The
results show a number of interesting things: Ist, that the
formula of Lamb derived on the theory of ffuid friction, is appli-
cable only when F& is small compared with unity. 2d, cis a
decreasing function of & for values of the latter up to about 2,000

Chemistry and Physics. 219

where a notable minimum occurs. 3d, From R=15,000 to
180,000 the quadratic law of resistance is sensibly followed as c
remains at 1.2. At R= 200,000 a rapid fall of c begins which -
does not stop till it reaches 0.8. A similar phenomenon has been
noted in experiments on spheres and the Reynolds number is said
to have a critical value at the beginning of this interval. The
decrease of the coefficient of resistance in this region is so great
that the absolute value of the resistance experienced by a cylinder
fell off considerably as the air velocity increased. To indicate
how great a departure from the quadratic law occurs, it was
found that witi a cylinder 30 em. in diameter the resistance fell
from 4 to 2.5 kilograms as the velocity was increased from 15 to
20 meters per sec.

Considerable light is thrown upon the preceding results by a
study of the actual stream line forms under different circum-
stances. With small values of R and where the viscosity is the
predominant factor, the stream lines, though less concentrated on
the downstream side, are smooth. In the region of R = 100 an
oscillation of the stream lines behind the body begins and with
higher numbers a well developed turbulence sets in. The exist-
ence of this condition was detected by acoustic methods up to
R = 100,000.

The stream lines which separate the region of turbulence from
the irrotational region may be called contact lines and the points
where the latter leave the cylinder contact points. The breadth
of the turbulent region in a way corresponds to the magnitude of
the resistance. As the critical value of R is passed it is noted
that the contact points approach on the downstream side, thus
indicating a decrease of the resistance.

It was also found that the roughness of the surface has a
marked effect on this phenomenon as does also the geometrical
form of the body. A discussion of some of the results in this
category is also given in the paper.—Physik. Zeitschr. 22, 321,
192¢. Bebe B.

6. Les Fondaments de la Geometrie; by Henri Porncars,
Paris, undated (Etienne Chiron).—This essay first appeared in
The Momst in January 1898. As the French original was not
preserved, and the fundamental concepts of geometry are very
much in evidence at present, the time has been thought oppor-
tune for the publication of a translation of the English text.
The author’s thesis is that the Euclidean metric is simply con-
venient and its axioms nothing more than conventions. This
after all seems more than a play upon words than a matter of
debate, for if with Poincaré we define geometry as the study of
the formal properties of a certain continuous group, then since
there may be various mathematical groups which are continuous
there will be various geometries, all of which are true. The

220 | Scientific Intelligence.

physicist naturally chooses the one which most nearly corresponds
to the more or less rough group in a material body and that has
been up to the present the Euclidean geometry. Whether physi-
cal space is warped by one or two parts in a million may be left
for the future to decide.

In the author’s view geometry is not an experimental science
but merely the result of reflection upon our experience, and to
ask whether the geometry of Euclid is true and that of
Lobatchewsky false is as absurd as to ask if the metric system 1s
true and that of the foot and inch is false. 2 Es ieee

Il. Gkrotogy anp MINERALOGY.

1. James Hall of Albany, Geologist and Palwontologist, 1811-
1898; by JoHN M. Cuarkn. Pp. 565, portraits and charts,
Albany (S. C. Bishop), 1921 ($3.70 net).—This is truly a
remarkably interesting study of the great state geologist of New
York and international man of science, James Hail, by a versatile
son of the same state. No geologist will fail to read it, and all
well-informed men of science should do the same. It is not only
a faithful portrayal of an extraordinary man, who turned out a
prodigious amount of geologic work, but as well a setting forth
of North American geology as it was emerging from the Wer-
nerian stage into fuil manhood. It is a true picture of the
pioneer days of the earth sciences in this country.

Hall was a tall, well-set-up, blue-eyed man, decidedly dynamic,
sensitive to an extraordinary degree, irascible, and with sur-
passing ambition. He was rarely ever sick, and yet in his own
mind, his physician tells us, he was ‘‘a dying man for fifty
years.’’ Feverish, with unrestrained impulses, he appears to be
always in trouble with most of his associates, and yet often misled
by his confidence in those less close to him. It seems a curious
commentary that he lost much money in mining.

In 1837 Hall began his survey of the wonderful Fourth Dis-
trict, at the western end of New York State, and spent five years
in the study of its geology, ‘‘the most excellent piece of field
work he ever did.’’ From this work he gradually rose to be the
master mind in the State Survey, and for sixty-two years he
dominated the Paleozoic geology, and especially the paleontology,
of North America, leaving as his monument thirteen great quarto
volumes besides many other works in paleontology. The Palwon-
tology of New York was the dominant note of his long life.
Curiously, though he was an evolutionist thirteen years before
the appearance of The Origin of Species, yet he never wrote any-
thing on this subject. ‘‘It was ever conditions not theories that
confronted him and he went on heaping up new facts to the end.’’

As one reads the book, it becomes apparent that New York has

Geology and Mineralogy. 221

been the greatest mother of geologists, the training ground for
most American workers in the science between 1843 and 1890.
Moreover, due to Hall’s great influence, interwoven with the
growth of geologic knowledge in New York is the development of
the geology of the entire Mississippi Valley. Hall ‘‘ guided official
geologic movements in every state where they were inaugurated
and in many his own hand took a helmsman’s part.’’ Not the
least valuable feature of the book are the many letters quoted
wholly or in part, with their interesting side-lights on the emi-
nent writers and the times in which they lived. .

Doctor Clarke has made full use of his unique opportunity to
know Professor Hall and his history, and he is to be heartily
congratulated on the results he has attained. Cams

2. The Problem of the St. Peter Sandstone; by CHaruxs L.
Daxke. Univ. of Missouri School of Mines and Metallurgy, Tech-
nical Series, Bull., vol. 6, No. 1, 225 pp., 30 pls., August, 1921.—
This thorough and clearly presented study of the environment
under which the St. Peter was laid down is based on the very
wide distribution of this formation throughout the Mississippi
valley. Starting out with the prevalent idea in modern times, the
author ‘‘was biased in favor of the desert origin of the sand-
stone,’’ but finally became convinced ‘‘that the formation is in
fact of marine origin.’’ The rounded and frosted nature of the
sand grains is found to be of antecedent origin and was not pro-
duced during the time of the St. Peter deposition. This character
of the sands is inherent in those of the Upper Cambrian (Croix-
ian) and it is from this northern formation, once of far wider
distribution than at present, that the sands of the St. Peter were
derived. Similar sands occur in other Ordovician formations.
Finally, the St. Peter is a well bedded formation of cleanly
washed sands (almost 99 per cent silica), and even though fossils
are rare, a considerable fauna was obtained by Sardeson in the
midst of this series in Minnesota, which is farthest away from the
source of the marine transgression. In the south, the equivalent
formations of the St. Peter more often yield marine fossils.

Incidentally the report records much information as to the
formations beneath the St. Peter, and more especially those above
it. Two valuable correlation tables and many excellent views of
outcropping formations accompany the book. The author is to
be congratulated on his excellent work, which constituted his
dissertation for the doctorate at Columbia University. G.4S.

3. A Geological Reconnaissance of the Dominican Republic;
by T. W. VauaHan, WytTHE Cooks, D. D. Conprt, C. P. Ross,
W. P. Wooprine, and F. C. Cankrns. Geol. Survey Dominican
Republic, Memoirs, vol. 1, 268 pp., 23 pls., 1921.—This important
publication, the jomt work of the Dominican Republic and the
U.S. Geological Survey, describes the geography, geology, strati-

999 Scientific Intelligence.

graphic paleontology, water resources, and economic geology of
the Dominican Republic. The geologic sequence begins with the
‘basal complex’’ of unknown age, followed by Upper Creta-
ceous and Cenozoic marine formations.

4. A Manual of Seismology; by CHARLES Davison. Pp. xu, -
256, with 100 text figures. Cambridge University Press, 1921.—
In writing this text book the author has purposely sought to
present the subject of seismology in its geological aspect. As he
points out, ‘‘until near the close of the last century, seismology
was regarded as a department of geology, but while its growth in
that direction has by no means ceased, the more recent advances
have been largely the work of mathematicians and physicists.’’
But inasmuch as the present volume was written to constitute
one of the Cambridge Geological Series, the mathematical and
physical sides of the subject have in consequence been almost
completely omitted.

The book gives an admirable outline of the present status of
the science. It deals with the phenomena of earthquakes in gen-
eral, and therefore it gives no detailed descriptions of individual
earthquakes. The presentation of the subject is well-balanced,
concise, and lucid, and consequently makes easy reading. The
more speculative matters have been eschewed and the presenta-
tion generally sticks to the reasonably assured ground of seismol-
ogy. Most of the chapters are preceded by a list of sources,
which ean be profitably consulted by the reader who desires fuller
information.

It may perhaps be suggested that the practical deductions of
seismologic study might have been more fully treated. The
author’s concern is essentially with the purely scientific side of
the subject. The practical deductions are indeed mentioned,
rather casually on the whole, but they would seem worthy of
inclusion in a separate chapter, both in view of their importance
and in order to make them more readily available. Probably the
author’s attitude is unconsciously revealed by his placing in a
foot-note an interesting suggestion as to the probable immunity
of the Panama Canal cuts to earthquake damage. The facts
and practical deductions of seismology are of unusually wide
popular interest, more so probably than those of any other branch
of geology, and some cognizance of this interest might well be
taken. By providing the chapter suggested, the author would
aid geologists in meeting this demand.

In conclusion, this excellent book gives us an epitome of the
science of seismology in its geological bearings, and by its suc-
einctness and its admirable simplicity of treatment makes attrac-
tive much of what would otherwise be abstruse matter. It is a
welcome addition to the Cambridge Geological series.

: ADOLPH KNOPF.

Geology and Mineralogy. 223

5. Earth Evolution and its Facial Expression; by WILLIAM
Hersert Hosss. 8vo, pp. 178, 84 illustrations. New York,
1921. (The MeMillan Company).—Professor Hobbs states that
the purpose of this volume is to assemble and make available for
the non-technical reader the author’s ideas on certain questions of
theoretical geology. He favors the planetesimal hypothesis of
Chamberlin, but suggests an order of accretion which will
account for the peculiar distribution of densities in the earth.
He believes that the present figure of the earth resembles a
tetrahedron, but that this figure has developed siowly as a result
of secular cooling, the form at the end of the Paleozoic era sug-
‘gesting a huge twinned crystal. Igneous magma is formed by
the fusion of shale at a moderate depth on relief of pressure
during folding or faulting. Thus laccolites develop essentially
in place, and are the result rather than the cause of the doming
in overlying strata. All arcuate mountains have been made by
underthrusting of sediments against continental masses, the
active force resulting from subsidence of ocean floors. Some
regions in and bordering the Pacific ocean are at present suffer-
ing rapid change, as indicated by earthquakes, volcanic activity,
terraces, and fore-deeps. The book closes with the suggestion
that investigations in the Pacific will cause a great reduction in
present estimates of geologic time. Cassie

6. West Virginia Geological Survey; I. C. Wurtz, State
Geologist. Morgantown, W. Va.—Recent publications include
the following:

Detaled Report on Nicholas County; by Davi B. ReceEr.
1921. Pp. xx, 847; with 34 half tone plates and 22 text figures.
Accompanied by a separate case of topographic and geologic
maps. In the preparation of this exhaustive report Mr. Reger,
assistant geologist, has had the assistance in the field of W. Arm-
strong Price, paleontologist, and in the office work, of R. C.
Tucker and J. D. Sisler. Mr. Tucker has had charge of the
preparation of the maps and.the calculation of the areas and
tonnage of the various coal beds.

The coal development of Nicholas County has only just begun
because of partial lack of rail facilities. It contains three
important coal formations extending across the county, from
northeast to southwest. These are the New River Coal Group,
the Kanawha Group, and the lower members of the Allegheny
Series. The report contains a chapter on the paleontology of
Nicholas County and a short description of the chert deposits of
West Virginia. The price, including case of maps (delivery
charges paid), is $3.00. Extra copies of topographic map, 75
cents; of the geologic map, $1.00.

New Edition of Coal, Oil, Gas, Limestone, and Iron Ore Map.—
This map has been thoroughly revised, showing oil and gas pools,

Am. JOUR. aa aah SERIES, Vou. III, No. 15.—Manrcu, 1922.

994 Scientific Intelligence.

many anticlinal lines not heretofore shown. It includes also
booklet giving the names and post-office addresses of all the
principal coal mining operators in West Virginia up to July Ist,
1921. Scale, 8 miles to the inch. Price, folded in strong enve-
lope and delivered by mail, $1.00; 6 copies for $5.00.

A Soil Survey of Fayette County has also been issued by the
Geological Survey in co-operation with the U. 8. Department of
Agriculture; this is by J. A. Kurr, and contains a large Soil
map on a seale of 1/62,500; contour interval 50 feet.

7. Virginia Geological Survey; THomas L. Watson, Direc-
tor. Charlottesville, 1921.—Bulletin XXI (pp. 224; with 18
plates and 21 text figures). This is by ALBERT W. GILEs and dis- °
cusses the geology and coal resources of Dickenson County.

8. Wasconsin Geological and Natural History Survey; W. O.
Horcukiss, Director. Madison, Wisconsin, 1921.—Bulletin 58
(pp. 252, with 2 maps and 93 text figures) is by R. H. Wrrseck
and describes the geopraphy and economic development of South-
eastern Wisconsin.

9. Dana’s Textbook of Mineralogy. Third Edition; by
Wiuuiam EB. Forp. Pp. ix, 720, with 1,050 text figures. New
York, 1922 (John Wiley and Sons).—After an interval of nearly
twenty-four years a third edition of the Textbook has been pub-
lished, the task of revision and of incorporating the advances in
mineralogy during the past two decades having been undertaken
by Professor Ford. The general character of the book has been
retained, but the addition of the new matter required to bring the
book abreast of the times has necessitated lengthening it by 127
pages. In spite of this increase in the number: of pages, the
actual bulk of the new volume does not exceed that of the former
edition, owing to the thinner and better paper employed.

In the section on Crystallography the methods of stereographic
and gnomonic projection have been introduced. Much of the
section on Optical Characters of Minerals, long recognized as the
best concise, adequate presentation of this subject in Enelish, has
been rewritten, expanded, and brought into closer relation with
the actual working methods of microscopical technique. The
many excellent figures added here should do much to render the
subject matter more easily assimilable by the student.

In the part on Deseriptive Mineralogy the reactions in polished
sections of many of the opaque minerals are given. All species
newly recognized since the previous edition are briefly mentioned
in their proper places. The largely increased amount of optical
data given for the various minerals reflects the greatly enhanced
importance that the immersion method of determining minerals .
has assumed in recent years. The great wealth of information
concisely presented in the new edition makes it an exceedingly
valuable compendium of mineralogy. ADOLPH KNOPF.

Miscellaneous Scientific Intelligence. 225

10. A new Meteoric Iron—A beautifully fluted 425 Ib.
(193.17 kilogr.) meteoric iron found in Owens Valley, California
in 1913 and not yet described has recently been added to the
National Collections through the generosity of Mr. Lincoln Ells-
worth of New York City. The iron, an octahedrite, has been
analyzed in great detail by Professor 8. R. Brinkley of Yale Uni-
versity and a description of the same by the writer will appear in
a forthcoming number of the Memoirs of the National Academy
of Sciences. G. P. MERRILL.

Il]. Misce,tnanrous Screntiric INTELLIGENCE,

1. Carnegie Foundation for the Advancement of Teaching.
Sixteenth Annual Report of the President, HmNRY 8S. PRITCHETT,
and Treasurer, RopertT A. FRANKS, for the Year ending June 30,
1921.—The total resources of the Carnegie Foundation now
amount to $25,513,000, of which $15,192,000 belong to the per-
manent general endowment, $8,206,000 to a reserve fund to be
spent in the retirement, during the next sixty years, of teachers
now in associated institutions, $1,250,000 to the endowment of
the Division of Educational Enquiry, and $570,000 to a reserve
fund to be expended in aiding universities and colleges to adopt
the new plan of contractual annuities. In the sixteen years of
the existence of the Foundation, a total sum of nearly nine mil-
lion dollars has been distributed in retiring allowances and
pensions to 999 individuals. Of this sum former teachers of Har-
vard have received $705,000, of Yale, $609,000, of Columbia,
$525,000, of Cornell, $407,000; the remainder has gone to eighty-
five different institutions. There are now operative 379 retiring
allowances and 230 widows’ pensions, 90 of which were granted
during the last year, entailing an annual expenditure of $959,690.
The average allowance paid is $1,575.

It is particularly interesting to note that the Teachers Insur-
anee and Annuity Association of America, which was established
by the Foundation through a gift of $1,000,000 to provide insur-
ance and annuity protection for college teachers without overhead
charges, has written for teachers in 300 different institutions 982
insurance policies covering $4,973,000 of insurance and 776
annuity contracts providing $917,000 annual income at retire-
ment. A plan has been provided for the participation of
policy-holders in the management of the Association. Sixty-two
institutions have formally adopted the new plan of contributing
toward contractual retiring allowances for their teachers.

The work of the Foundation, however, extends far beyond the
distribution of pensions to retiring teachers. Some of the prom-
inent subjects studied and discussed involve the critical examina-
tion of the medical schools of the country; of the training for

226 Scientific Intelligence.

the legal profession; of College entrance requirements; and
especially important of the various pension systems introduced
by institutions of learning, by the different states and by large
business corporations. Of the subjects mentioned, to the first
have been devoted the special Bulletins IV and VI; it is further
discussed in the present report. The legal matter is the subject.
of Bulletin VIII, and XV by Alfred Z. Reed, the latter published
the past year (see this Journal, vol. 2, p. 236). The development
of pension systems has been ably handled in many annual reports
and in none more so than the present (pp. 115-157) ; ‘‘a very slow
growth in the critical attitude on the subject’’ by the authorities
involved is a conclusion much to be regretted.

2. A Monograph of the Huxsting Crinoids; by AvSsTIN
’ Hopart CuarK. Vol. 1, The Comatulids. Quarto. Part 1, pp.
1-406, with 513 text-figures and 17 plates; part 2, pp. 1-795, with
949 text-figures and 57 plates. Bulletin 82, U. 8. National
Museum. Washington, 1915 and 1921 (Government Printing
Office ).—This is a complete monograph of all that is known about
living erimoids, based upon a study of the material contained in
all the principal museums of the world as well as that collected
by the author himself on various expeditions. That the crinoids
are in no sense the rare creatures that they are popularly 1mag-
ined is indicated by an enlightening paragraph in the introdue-
tion. ‘‘During the 1906 cruise of the Albatross I handled tens of
thousands of specimens; several times I saw the forward deck of
the steamer literally buried under several tons of individuals
belonging to a species exceeding any fossil form in size; every-
where we went we found ecrinoids; we dredged them at all
depths. My ideas of the comparative importance of the recent
forms underwent a total change; surely a group so abundant,
even though very local and very unevenly distributed over the
sea floor, can not be considered as decadent or degenerate. From
my observations at sea I became convinced that the recent crinoids
are In every way as much of a factor in the present day marine
biology, and play fully as important a part, as the echinoids, the
holothurians, or the asteroids; ecologically they are more inter-
esting than any of these because of their sessile mode of life and
curlously specialized method of procuring food.”’

The first part treats exhaustively the relationships of the
crinoids, the general features of their organization and a detailed
comparison of certain of the calcareous structures of all the
known species. ;

The second part completes the account of the calcareous struc-
tures and gives a full description of each of the other organ sys-
tems of the body, including the digestive, cireuiatory, nervous,
and reproductive systems, sexual differentiation, spawning,
embryology, larval development, metamorphosis and growth.
This part also ineludes all that is at present known regarding the

Miscellaneous Scientific Intelligence. 227

habits and reactions to stimuli of the various species, with an
account of their very abundant parasites and commensals and a
detailed discussion of their coloration. .

So exhaustively is the subject treated and so well executed and
elucidative are the illustrations that the crinoids may now be
removed from their former association with the little-known
eroups of animals and placed among those most thoroughly
investigated. Were Cs

3. Source Book for the Economic Geography of North
America; by CHaruses C. Conpy. Pp. xi, 418; with 11 maps in
black and white. Chicago, 1921 (The University of Chicago
Press).—This volume is a compilation of 119 articles, from nearly
as many different authors and statistical sources, assembled here
in 14 chapters. The first chapter, entitled Regional Concepts of
Canada, contains six articles dealing successively with the size,
agricultural regions, land and water areas, ciimate, mineral
resources of the major physiographic provinees, and the popula-
tion of Canada. Chapter II deals with the resources and indus-
tries of Canada, and is followed by five chapters dealing with the
separate geographic subdivisions: Maritime Canada (and New-
foundland), St. Lawrence Lowland, Laurentian Upland, Prairie
Provinees, and Pacific Canada. A similar procedure is followed
for the United States and Mexico. |

The purpose of the book is to furnish source material for an
introductory study of the geography of North America and to
introduce students to the literature of the subject. <A large
amount of interesting, valuable, and highly diversified informa-
tion has been assembled, ranging from the apple industry in
Nova Scotia to the genesis of the ores at Butte, Montana. It will
be seen that ample scope is left for the ingenuity and resourceful-
ness of the instructor to unify so broad a body of information.

ADOLPH KNOPF.

4. Ueber die Vorstellungenen der Tiere: ein Beitrag zur
Entwicklungspsychologie; von Hans VouKuET. 126 pp. Leip-
zig und Berlin, 1914 (Wilhelm Engelmann).—This is the second
part of a series of studies on the origin and evolution of mental
processes and the psychology of human actions. The present
volume on animal intelligence analyses the presumable percep-
tions and conceptions of various types of animals with regard to
the hypothesis of total environmental complexes. Wee Foe Cs

5. Le Movement scientifique contemporan en France: 1, Les
sciences naturelles; par GrorGES Matisse. Pp. 160. Paris,
1921 (Payot & Cie.) —This little book indicates the recent devel-
opment of the natural sciences in France by summarising the
work of such eminent biologists as Delage on heredity and par-
thenogenesis, of Bataillon on artificial parthenogenesis, of Hous-
say on dynamic morphology, of Cuénot on preadaptation, of
Bohn on tropisms, and of other French investigators who are
leaders in various branches of biology. WwW. R. C.

arp’s Natural Science EstaBuisHMent
A. Supply-House for Scientific Material.
Founded 1862. Incorporated 1890.

A few of our recent circulars in the various
departments:

Geology: 5-32, Descriptive Catalogue of a Petrographic Col-
lection of American Rocks. J-188 and supplement.
Price-List of Rocks.
Mineralogy: J-220. Collections. J-238. Minerals by Weight.
_ J-224. Autumnal Announcements.
Paleontology: J-201. Evolution of the Horse. J-199. Pale-
ozoic index fossils. J-115. Collections of Fossils.
Entomology: J-38. Hie Be J-229. Life Histories. J-230.
. Live Pupe.
Zoology: J-223. Material for dissection. J-207. Dissections
of Typical Animals, etc. J-38. Models.
Microscope Slides: J-189. Slides of Parasites. J-29. Cata-
logue of Slides.
Taxidermy: J-22. North American Birdskins. Z-31. General
Taxidermy.
Human Anatomy: J-37. Skeletons & Models.
General: J-228. List of Circulars & Catalogues.

Ward's Natural Science Establishment
84-102 College Ave., Rochester, N. Y., U. S. A.

“SCIENTIA”

INTERNATIONAL REVIEW OF SCIENTIFIC SYNTHESIS. Jsswed monthly (each
number consisting of 100 to 120 pages). Editor: EUGENIO RIGNANO..

_ This is the only review which has a really international collaboration; which is
of world-wide circulation and occupies itself with the synthesis and unification of
knowledge, in the history of the sciences, mathematics, astronomy, geology, physics,
chemistry, biology, psychology and sociology.

It is the only review which, by*inquiries among eminent scientists and writers
(on; The philosophical principles of the sciences; Fundamental astronomical and physical
questions of current interest; Contributions given by the various countries to different
branches of knowledge; Question of vitalism; Social question; International questions
raised by the world war), makes a study of the problems interesting scientific and intel-
lectual circles throughout the world.

It has published articles by Messrs, : Abbot=-Arrhenius -Ashley - Bayliss - Beichman-
Bigourdan - Bohlin - Bohn-= Bonnesen = Borel = Bouty - Bragg = Bruni = Burdick =- Carver =
Caullery - Chamberlin = Charlier = Claparede = Clark = Costantin - Crommelin = Crowter =
Darwin - Delage - De Vries - Durkheim = Eddington - Edgeworth = Emery = Enriques -

Fabry - Findlay = Fisher = Fowler = Golgi = Gregory = Harper = Hartog = Heiberg - Hinks =

Hopkins-Inigues-Innes-Janet-K aptein-Kaye-Kidd-Langevin-Lebedew-Lloyd Morgan-
Lodge = Loisy =- Lorentz = Loria =Lowell - MacBride - Meillet =Moret-Muir-=Peano -Picard -
Poincare - Puiseux = Rabaud = Rey Pastor = Righi-Rignano-Russell-Rutherford-Sagnac=
Sarton = Schiaparelli= Scott = See = Sherrington = Soddy = Starling - Svedberg = Thomson =
Thorndike-Turner -Volterra=-Webb-Weiss = Zeeman and more thana hundred others.

** SCIENTIA’’ publishes its articles in the language of its authors, and joins to the
principal text a supplement containing the French translations of all the articles that
are not in French. (Write for a Specimen Number to the General Secretary of
“Scientia”, Milan.) Annual subscription: 40 sh., or 19 dollars post free.

Office : 43 Foro Bonaparte, Milan, Italy.

Publishers: WILLIAMS & NORGATE-London; FELIX ALCAN -Paris
NICOLA ZANICHELLI-Bologna; RUIZ HERMANOS-Madrid;
WILLIAMS & WILKINS CO-Baltimore.

ARTO Owls fesioee of Blastomerys » marshi ; by? R. s.
(With Plate III)... -

Azr. XII. —Oregon Tertiary Canide, with Descriptions of
New Forms; by M. R. Tuorpe ..... .

| Arr. XIII.—The Crystallographic and Atomic Symmetries |
of Ammonium Chloride; by R. W. G. WYcKorF.. ae re

Arr. XIV.—The Crystal Structure of Silver pete (Ag.0)5 3
by R. W. G. Wrcxorr

Art. XVI. —Melanovanadite, s Naw Mineral 6 ae ee oh
Pasco, Peru; by W. LINDGREN, L. F. Hamirron and ‘
O. Pazacue.

Arr. XVII.—The Ceratopyge Fauna in Western North
America; by P. E. Raymonp . ae es

Arr. XVIII.—The Pitts oeeeas e S. We MeO _ gu

SCIENTIFIC INTELLIGENCE. 7%
Bae and Physics. — ee Reduction = ss tic Salts with a Ey

217.—Les Opener dé la eae H- Porncarh, 219.

Geology and Mineralogy.—James Hall of Albany, Geologist and Patneated
gist, 1811-1898, J. M. Ciarnxe, 220.— The Problem of the St. Peter
Sandstone, C. L. DAKE: A Geological Reconnaissance of the Dominica
Republic, T. W. VaucHan and others, 221.— A Manual of Seismology,
Davison, 222.—Earth Evolution and its Facial Expression, W. H. Hops
West Virginia Geological Survey, I. C. WuHITE, 225.—Virginia Geological |
Survey. T. L. Watson: Wisconsin Geological "and Natural History Sur-
vey, W. O. Horcuxiss: Dana’s Textbook “of Mineralogy, Third Editi
W. E. Forp, 224.—A new Meteoric Iron, 2265.

Midocilancous Scientific Intelligence.—Carnegie Foundation for the Ady

ment of Teaching, Sixteenth Annual Report’ of the President, H. S._
PRITCHETT, and Treasurer, R. A. FRANKS, for the Year ending June 30, *
1921, 225.—A Monograph of the Existing Crinoids, A. H, CuarK, 226.—
Source Book for the Economic Geography of North America, GEEE Comte z
Ueber die Vorstellungenen der Tiere: ein Beitrag zur Entwicklungspsy-
chelogie, H. VoLKLET: Le Movement scent ius contemporain en France:
1, Les sciences natur elles, G. Matisse, 227 i

APRIL, 1922

Established by BENJAMIN SILLIMAN in 1818.

THE

AMERICAN
JOURNAL OF SCIENCE.

DANA.

Epitor: EDWARD S&S.

ASSOCIATE EDITORS

Prorressors WILLIAM M. DAVIS anp REGINALD A. DALY,
oF CAMBRIDGE,

Prorressors HORACE L. WELLS, CHARLES SCHUCHERT,
HERBERT E. GREGORY, WESLEY R. COE anp
FREDERICK E. BEACH, or New Haven,

PROFESSOR EDWARD W. BERRY, or BALTIMORE,
Drs. FREDERICK L. RANSOME aAnp ee BOWIE,
OF WASHINGTON. fw

i

FIFTH SERIES \ ,
VOL. IIJ—[WHOLE NUMBER, COTII).
No. 16—APRIL, 1922.

aes, NEW HAVEN, CONNECTICUT.

PO Res

THE TUTTLE, MOREHOUSE & TAYLOR CO., PRINTERS, 123 TEMPLE STREET.

bo Published monthly, Six dollars per year, in advance. $6.40 to countries in the
2a aes Postal Union ; $6.25 to Canada. Single numbers 50 cents; No. 271, one dollar,
its Entered as second-class matter atthe Post ( fice at New Hay en, Conn,, under the Act

Wil a

Published March 22nd.

COAL.

Its Properties, Analysis, Classification, Geology, ‘Extraction, Uses —
and Distribution
By ELwoop S. MOoRE
Dean of the School of Mines,
The Pennsylvania State College.

The only recent book on the market which treats in one handy volume and
in a comprehensive way everything concerning coal, including its origin,
geology, physical and chemical properties, geographic and geologic distribution,
etc,

It is an authoritative book of reference for geologists, coal men generally,
scientists, chemists (particularly organic and analytical), and geographers.

Botanists, also, will find it of value, as it takes up the question of coal flora
of the coal-forming periods. ;

462 pages. 6by9. 142 figures and several folding charts. Cloth, $5.00.

Every Geologist Should Have These Books

EcoNoMIc GEOLOGY.

Fourth Edition. By H. RIEs, Ph.D.
856 pages. 6by9. 291 figures, 75 plates. $5.00.

_ HANDBOOK FOR FIELD GEOLOGISTS.

Third Edition, Thoroughly Revised and Rearranged.
By the late C. W. HAYES, Ph.D., and SIDNEY PAIGE.
166 pages. 4 by 62. 20 figures. Flexible, $2.50.

We: are always g glad to send any Wiley books on our 10 days Free Examination — >
plan (teachers allowed 60 days).

JOHN WILEY & SONS, Inc.
432 Fourth Avenue, New York

AJS 4.22

THE

AMERICAN JOURNAL OF SCIENCE

[FIFTH SERIES. |

+0

Arr. X1X.—Mmor Faulting in the Cayuga Lake Region,*
by E. Tatum Lone.

INTRODUCTION.

The drought of the summer of 1921 caused many of the
ereeks in the vicinity of Cayuga Lake to run almost or
entirely dry. Advantage of this condition was taken to
make a rather detailed study of rock exposed in their beds
and on their banks, with a view to adding if possible some
information as to the nature of the numerous small faults
of the region, and in particular to a very unusual mani-
festation of faulting shown on a small scale in Salmon
Cr. at Ludlowville, N. Y. (See map, fig. 1.) For some
time these faults had caused considerable speculation but
weather conditions made field investigation impossible.

As it soon became evident that knowledge of the general
structure of the region would give little clue to the par-
ticular features here exposed, the beds of some thirty odd
creeks covering a distance about 20 miles to the north of
Ludlowville and 4 miles to the south, were traversed, in
addition to the whole of the east shore of Cayuga Lake,
except the lower nine miles. The Lehigh Valley Railway
tracks for this same distance and about six continuous
miles of the west shore above Taughannock, with a few
additional shorter stretches to the north, were also cov-
ered on foot so that all details could be closely observed
and contact readings made of the dip of the strata. A
launch was employed in order that sight readings could be
made on both shores,a distance of about 30 miles being cov-
ered in this way. ‘These readings proved more accurate
and more representative, as several feet instead of inches
could in this way be covered by the clinometer of a Brun-

* Grateful acknowledgment is herewith given Prof. A. J. Eames and
Prof. V. E. Monnett for helpful criticism of the manuscript.

Am. Jour. Sci.—Firts Series, Vou. III, No. 16.—Aprix, 1922.
ies

j. I. Long—Minor Fa

ulting
Ud
Fox

! cWaN

m the Cayuga Lake Region. 231

ton which was used for most readings. An average dip
of a little less than 1° (about 60’) was thus established over
by far the greater part of the lake shore. The dip being
so gentle it is almost impossible to get accurate readings
even by sighting with a level clinometer on a telescope,
hence the impracticability of contact readings.

The presence or absence of cliffs, as well as that of the
terraces, in some places so conspicuous, is due to the

Fic. 2.—The Tully Limestone between Portland Pt. and Idlewood, east
shore of Cayuga Lake. The Tully is here shown capping the Hamilton with
the usual southerly dip. Though but 20’ thick it forms a conspicuous terrace
with a corresponding dip to the south.

nature of the formation exposed at a given locality,
whether hard or soft. The height of the cliffs and ter-
races, sometimes over 100’ above the lake-level, is due to
the position of the two dominatingly hard layers of the
district—the Tully limestone immediately overlying the
Hamilton shale, and the Encrinal bed of the Hamilton,
somewhat over 100’ below. These two resistant layers
form waterfalls in all creeks which cross their outcrops,
the height of the fall being dependent upon the elevation
of the hard ledge.

232 E. T. Long—Minor Faulting

Major Structural Features.

The Watkins Glen.—Catatonk folio deals with the
region south of a point called Esty’s Glen which is located
about four miles north of Ithaca. ‘To the north of this
point little work has been done, but one would expect to
find a continuance of the large features as shown, by
Hi. M. Kindle, to exist in the region to the south, including
not only south central New York but northern Pennsyl-
vania as well. This structure is considered by him to
be the northward dying out of the results of the Appa-
lachian Mountain making whose trend and characteristics
it shares. ‘The sinuous axes of the broad low folds of the
Chemung and Portage of this area are given in a sketch
map on page one hundred of the folio, as well as on the
Areal Geology sheet. The average trend is slightly N-E
to S-W of a true east-west line and would imply pressure
from a direction 8.SE resistance being encountered in a
_N.NW direction. Hastward they swing to a more nearly
E-W course. From the south to the north end of Cayuga
Lake successively older rock are continuously met, the
general dip being to the south, though it is not constantly
so. The order of succession is: Portage at Ithaca and to
the north, until the Genesee appears in the bottom of the
cliffs about half a mile south of Esty’s. Something over
a mile to the north of this the Tully emerges from the
Lake and though seldom over 20’ thick, dominates the cliff
face of Cayuga Lake (see fig. 2) with but one break at the
mouth of Salmon-Cr., for a distance of 14 miles to the
north, after which for three more miles it makes falls
heading deep ravines in the Hamilton shale beneath. The
steepness and occasional great height of the cliffs of
Cayuga Lake are therefore all the more remarkable since
they are cut in the soft shale of the Hamilton for over half
of the 40 miles of its shore line. »

The quarry in the Tully limestone at Portland Pt.
seems to be located on the crest of the major anticline of
the whole Cayuga area at an elevation of 640’. Beds dip
away from this point to the south for a short distance at
the rate of 6°, but after 14 mile assume the usual dip of
about 1°. This continues with but slight variation to the
south end of the lake. ‘To the north, the beds dip a little

1 Watkins Glen-Catatonk folio 169, U. 8. G. S., pp. 98-111.

in the Cayuga Lake Region. 933

less than 1° northward for about a mile, when they
assume approximately a horizontal position for another
mile or so, before again dipping to the north at a rather
higher angle than before. This is again followed by a
horizontal area at Ludlowville. A short distance north-
west of the boundary of the accompanying map the dip
is reversed and the beds continue a southerly dip to the
northern end of the lake, some thirty miles distant.

The major fold will here be spoken of as the Portland
Point anticline, from the present name of the point, though
it was formerly known as Shurger’s Point and is so
named on the topographic sheet. Its axis across Cayuga
Lake trends nearly 5° south of E-W with a westerly pitch
of about 44’ per mile, that is, about %°. This westward
pitch of the anticline combined with the general southerly
dip of the strata serves to give an ever increasing height
to the strata in a northeasterly direction.

Faults in the Encrinal lumestone.

Faulting, like the other dynamic movements of the
region, is on an unobtrusive scale and would deserve little
or no attention were it not for the possibility of throwing
some light on the history of the region, indicating some-
thing of the nature and direction of the forces applied,
and illustrating in nature, even though in miniature, some
of the principles involved in rock movement. All of the
faults under observation occur in the Hamilton shale.
The difference in physical properties between the Enclinal
bed of the Hamilton and the vastly greater shaly portion
throw the faults therein occurring into two obvious
classes, though both are of the ‘‘low-angle’’ type. Those
in the Encrinal are conspicuous enough to one on the look-
out for them, but undoubtedly there are distributed
through the shale countless thousands, the existence of
which one will never guess. Vast thicknesses of the shale
have no visible bedding planes and even in situ are so
broken up into small flat lenses, often less than an inch in
either direction, that only by virtue of an offset in some of
the numerous joint planes is the presence of a fault
detected. But little is therefore known of the movements
within the shale, for its tendency to split up into this mul-
titude of lamelle at once destroys any record which might

234 _ £. TT, Long—Minor Faulting

have been preserved on the fault plane. It is almost
impossible to get data for the third dimension and impres-
sions are correspondingly vague.

Quite the reverse is the case in the Encrinal layer. In
the locality of Ludlowville and on the opposite shore,
faulting is more frequent than farther north, and the
Encrinal bed here consists of nearly two feet of coarsely
comminuted crinoid fragments containing well-preserved
fossils of good size. It is hard and compact, though

Fig. 3.—Striated columnal structure due to movement under compression
along fault planes in the Encrinal layer between Crowbar and Willow Cr.,
west shore Cayuga Lake.

coarsely crystalline, forming ledges or waterfalls as the
conditions determine; a typical ‘‘competent’’ bed, in
which the faults conform to the laws of shearing under
horizontal compression, where the direction of least
resistance is both upward and downward into the com-
aratively mobile shale. The fault planes dip both to
the north and the south at angles varying from 20°-30°,
and on both sides of the plane show striated columnar
structure produced by movement under pressure (see
fig. 3). The most numerous collection of these faults is
found on the west shore of the lake midway between
Crowbar and Willow Cr. and in a second strip of Enerinal

im the Cayuga Lake Region. 235

north of Willow Cr., the first on the south limb of the anti-
cline, the second on the north limb, the dip on both being
shehtly less than 1°.

Fig. 4.—Intersecting faults in the Encrinal layer between Crowbar and
Willow Cr. west shore Cayuga Lake. Notice the change in angle of dip as
the faults enter the underlying shale, as well as the wedge-shaped block lifted
up by the horizontal pressure.

Fig. 4 will bear repetition here as giving, from nature,
an example of faulting strikingly similar to the results of
one of Daubrée’s famous experiments, as shown in fig. 5.”
The block subjected to direct pressure from both ends
was a mixture of plaster, beeswax and resin, made to
approach natural conditions as nearly as possible and in
which were developed fractures formed at angles of about
45° to the direction of compression. The wedge. shape
of the fault block produced in the Encrinal, though of
such small size as to almost forbid comparison, seems to
illustrate a movement similar to that suggested by
Mr. R. T. Chamberlin in his article on ‘‘The Appalachian
Folds of Central Pennsylvania.’”

* Fig. 5 is a photograph of plate II, fig. 3 accompanying a discussion on
p- 316, taken from ‘‘ Etudes synthétiques de géologie expérimentale, vol. I,
by A. Daubrée.

* Chamberlin, R. T., Jour. Geol., vol. 18, No. 3, 1910 (especially pp. 246-
259).

236 E. T. Long—Minor Faulting

In the Encrinal bed the dip of the fault plane is 22°
to the north and 30° to the south, but as soon as it enters
the shale below it swings around and within a couple
of feet has assumed a 20° dip to the north instead
of 30° and a 10°- dip to the north: instead ome22a
At the intersection of the faults some of the rock is
broken away, which may give a false impression, but by
projecting the fault planes across the intersection it would
appear that the plane C D was the older, though older
possibly only by the time it takes to make a fault, for it
seems to be cut by the plane A B with a slight offset of
not more than1”. The wedge has been raised about three

Fie. 5.—Results of pressure applied to a block of plaster, wax and resin,
in an experiment of A. Daubrée. Notice the wedge block so similar to Fig.
4 above.

inches and the striated columnar structure indicates
movement in a direction about N 15° K, the strike of the
planes being approximately N 75° W. Owing to the posi-
tion of this and several nearby faults accurate measure-
ments were impossible. A comparison of fig. 4 with
fig. 10 in the article on ‘‘Low-Angle Faulting’’ by
Mr. R. T. Chamberlin and Mr. W. Z. Miller* will at once
prove interesting and instructive. Here in nature is a
striking demonstration of the trustworthiness of their
experiments. Of course one of the intersecting faults
must be eliminated in one’s mind to make the analogy
really good.

But this little fault goes even farther. Traced back and

*Chamberlin, R. T., and Miller, W. Z., Jour. Geol., vol. 26, p. 27, 1918.

in the Cayuga Lake Region. 237

upward into the overhanging vegetation on the south side
for a few inches there is evidence that after leaving the
hard Enerinal bed below, the fault again lowers its dip in
passing into the shale above. This is a very unusual
exposure as the shale in other observed cases has been cut
from under the hard limestone layer and all trace of
faulting obliterated.

Fic. 6—Outerop of Encrinal layer at the beginning of the high dip just
south of Portland Pt., east shore of Cayuga Lake.

Just opposite Crowbar and 14, mile south of Portland Pt.
the Encrinal appears emerging from the lake with a dip
rather steeper than usual. This 5° dip to the south cor-
responds with the increased dip of the Tully above while
approaching the nearby crest of the anticline at Portland
Pt., 4% mile north (fig. 6). In it one fault was seen with
a dip of 30° north and striations trending about N 8° E.
The position was such as to make an accurate reading of
the striated columnar structure impossible.

On the north limb of the anticline about one mile north
of its crest and just north of the bridge over Salmon Cr.
near its mouth at Myers, the Encrinal appears in the west
bank some feet above the village street with a dip of 1° N.
Joints and what appear to be faults occur, but the cliff is
too steep to make this part of the bed accessible. Around
the second turn up stream it again appears in the other

238 E. T. Long—Minor Faulting

bank with a cliff facing N 48° EK. This is most accessible
and where it crosses the bed of the creek gives a contact
reading for dip of nearly 2° N. This is undoubtedly too
high. Of all observed, here was found the best one for
measurements, along with several imperfect ones. Just
before the Encrinal layer crosses the creek a fault with a
dip of 25° Nis exposed. Only the foot wall remains, thus
greatly facilitating the reading of both dip and striated
structure. The latter is N 4° KH, the only positively
accurate reading for the direction of movement as shown
by the columnar structure on the fault planes. This read-
ing as all others is corrected for a magnetic declination of
8°, according to the U. 8. G. 8S. quadrangle for the area.

The following table will give some idea of an average in
the faults:

Faults dipping south Faults dipping north

angle of dip direction of striz angle of dip direction of strize
IL, AOS? SE about N. 6° E. 20° Ni about N. 6° E.
Bg PASS Sh 238° IN. N. 8° E.
35 GOO NSE ING ip OeRE 221° INI Neat oes

strike about strike about

INEST We Ne oce
4, 2eceNe N. 4° E.
strike N. 86° W.

ys 30° N. N. 8° E.

strike N. 72° W.

Nos. 1, 2 and 3 occur just north of Crowbar where cliff and shore line =
N. 40° W.

No. 3 is shown in fig. 5.

No. 4 is in Salmon Cr.; readings accurate within a fraction of a degree.

No. 5 is at Portland Pt.

For a discussion of the origin and results of rotational
strain, to which type of faulting the above appear to
belong, the reader is referred to C. K. Leith’s ‘‘Strue-
tural Geology’’ and to the above mentioned article on
‘‘Low-Angle Faulting.’? An attempted summary would
not do the subject justice, but two points may be empha-
sized. It is known that the great forces, which raised
thousands of feet of sedimentaries into the Appalachian
Mts., were applied in a practically horizontal direction.
‘“‘The planes of greatest tangential stress should there-
fore dip at angles somewhere in the neighborhood of 45°
and may plunge downward or upward.’ ‘This angle of

> Chamberlin, R. T., Appalachian Folds of Central Pa., Jour. Geol., vol. 18,
p. 247, 1910.

in the Cayuga Lake Region. 239

45°, however, may be modified and greatly reduced by a
number of conditions. Two in particular apply to the
ease in hand. As brought out in the article on Low-Angle
Faulting, 1. to lower the angle of the fault plane means,
to decrease resistance by friction as produced by normal
compressive stress; hence the avenue of least resistance
will be taken. 2. ‘‘Rotational strain may be developed
from horizontal compressive stresses, in heterogeneous
material by bedding or similar structure, which present
differences in competency.’’®

Deductions.

From the above facts as shown by the faults of the
Enerinal layer of the Cayuga Lake region it may be
assumed that the forces which produced them worked
approximately horizontally and came from a direction
between 4° and 15° west of south. This is a rather sur-
prising development in view of the preceding statement
that the region belonged to the outskirts of the Appalach-
lan province and had shared its history even though in
less active form than the mountainous tract. Some inti-
mation of the fact might have been discerned in the pro-
gressively changing curve of the axes of the anticlines as
shown in folio No. 169 page 100 as they turn from a gen-
eral SSW-NNE to an easterly direction. So ingrained,
however, is the idea that all pressure as applied to the
Appalachians must come from the southeast that the
warning passed unnoticed.

In northwestern central Pennsylvania, as well shown
on the U.S. topographic map, there occurs a sharp turn in
the trend of the Appalachians, with convexity N-W,
swinging them across northern Pennsylvania and south-
central New York in a nearly east-west course. This
soon again turns and they pass northward through New
Kingland in the usual NNE direction. Lines drawn at
right angles to the trend of the second curve which occurs
in south-eastern New York would converge not far east of
Cayuga Lake, which means that this region is practically
due north of the east-west segment. The fact that the
axis of the Portland Pt. anticline crosses Cayuga Lake

* Chamberlin, R. T. and Miller, W. Z., Low-Angle Faulting, Jour. Geol.,
vol. 26, p. 44, 1918.

240: E. T. Long—Minor Faulting

a few degrees NW-SE of a true E-W line is, therefore, in
keeping with the other surprising fact of the direction
whence came the pressure developing the faults of the
Einerinal layer, and in reality is what should be looked for,
if truly related to the Appalachian Chain.

A dike which gives some collateral evidence.

A feature which, on first sight, might in some respects
seem at variance with the previously expressed facts, is
the presence of a peridotite dike on the crest of the major
anticline. It was exposed five or six years ago in the east-
ern end of the quarry operated by the Portland Cement
Company at the point given its name. Rising from the

Fig. 7.—Inclusions of Tully limestone in a weathered peridotite (ser-
pentine) dike at Portland Point. Notice the fine bedding planes of the
limestone.

Hamilton, which forms the floor of the quarry and cutting
entirely through the Tully limestone it intrudes several
feet into the Genesee shale above. For the region it is
quite sizable, varying from 12” to 18”. The strike of the
dike, N 3°-6° W, is not in conformity with the postulated
direction of movement, which however is to be accounted
for by the fact that the dike is obviously here following the
course of one of the numerous joints belonging to the
N-S system. Its contact with the country rock is often
very close, there being no evidence of filling, but on the
other hand there are many places where the contact is
ragged, streamers of the magma having entered the hme-
stone and in some eases pieces of the Tully being broken
off and incorporated in the peridotite. (See fig. 7.) Con-

m the Cayuga Lake Region. 241

tact metamorphism is not conspicuous but a gradation of
texture is frequently noticed and the limestone is often
indurated for an inch or more. Calcite has very nearly
filled all the cavities opened up and on this soft medium
is recorded the usual N-S horizontal movement of the
region. So evidence other than faulting falls in line.
The north and south walls of the quarry very nearly par-

Fic. 8— Horizontal faulting in the Hamilton shale at the big Falls, Lud-
lowville, N. Y. The fault plane is just above the surface of the pool.

allel the axis of the fold so that the curved lines of dip
shown in these two walls as well as the eastern, together
with discrepancies in dip indicate a torsional movement,
elevation at no two corners corresponding.

Faults in the Hamilton Shale.

Quite another type of faulting from that already dis-
cussed is found in the shaly members of the Hamilton.
Here faults occur which are either movements along bed-
ding planes or parallel with them, on through various

249 E. T. Long—Minor Faulting

angles up to those very nearly vertical. One good exam-
ple of the horizontal type is shown at the water level of
the pool at the foot of the big Falls at Ludlowville, north
of the village, and just below the dam. (See fig. 8.) The
displacement is probably six inches. As the pool does
not dry up even in time of drought and no boat is pro-
vided for the convenience of visitors it was impossible to
get nearer the fault than a hundred feet or so. Itis about
30° below the Tully, the movement offsetting joints of the
N-S system, so that the upper mass was moved eastward
relative to the lower mass. This offset of the joints is
only the apparent displacement. The cliff faces very
close to due south, its trend being E-W, with no dip in
either direction which could be detected with a Brunton
clinometer or a telescope clinometer even by distant sight-
ing. As the streams always hug the north sides of their
courses it is to be inferred that a slight northerly dip
must be present. Taking the direction of pressure as
previously calculated at an average of about 10° west of
south, the chief movement will be in a general N-S direc-
tion, the E-W component representing only the very small
part played by the 10° deviation from N-S. As the fault
does not persist for any very great distance it is safe to
infer that the maximum displacement is nowhere great.
This locality is about two miles north of the crest of the
anticline at Portland Pt. and hence on the northern limb,
the dip of which is very far from constant. It appears
to be located on a second horizontal area, another level
district occurring about 100’ higher and half a mile nearer
it. Instead of being due to pressure this fault may possi-
bly, though not at all probably, be due to local relaxation
and hence belong to a type intimately connected with val-
leys and which Will be discussed in connection with the
wedge fault block to follow.

Mr. G.-C. Matson in an article on ‘‘Peridotite Dikes
near Ithaca, N. Y.’” calls attention to several dikes not
here considered. One group of four is above the high
fall over the Genesee shale, in a tributary to Salmon Cr.
just northeast of Ludlowville. ‘‘Three of these have been
faulted; the fourth does not reach up to the fault plane.
The amount of displacement is about two feet.’’> From

‘Jour. Geol., vol. 13, p. 265, 1905.

*Tbid

m the Cayuga Lake Region. Y43

this reference a horizontal fault is to be inferred and judg-
ing from the small offset a displacement similar to the
fault at the big Falls from which this is but a half mile
distant in a direction nearly due east.

Immediately east of the bridge crossing Salmon Cr. on
the road entering Ludlowville from the south is a fault of
most unusual appearance. The face of the bluff in which
it occurs trends N 73° E. that is, it faces in a general
northerly direction. Running from a point somewhere
under the bridge for over 100’ to the east there is another
horizontal fault which ends abruptly against a joint plane.
This too is at low-water level but over 20’ below the first

ee
2

Fic. 9.—Wedge-shaped block between a horizontal fault just above
water-level, and an oblique fault dipping toward the left margin of the
photograph. LHasily detected by the offset of the joints.

one mentioned. Just across the sluice to the east of the
bridge and 9 above the horizontal fault another fault
plane has been developed which dips down at an angle of
8° and meets the lower one just where it encounters the
joint plane. Cutting the face of this bluff are a number
of well-marked joints belonging to the N-S system. The
two faults form a wedge-shaped block between them,
which as seen in fig. 9 is thrust inward, that is to the east,
but exhibits no evidence of crushing at its point. The

244 E. T. Long—Minor Faulting

amount of offset, as measured in this case, is quite uni-
formly 6” on both faults. The junction of the two faults
with the joint plane has made a point of weakness in
which the creek has cut a miniature cave of a few inches
but in the absence of a brecciated zone and a decrease in
the offset of the joints in approaching the point of the
wedge it seems that the force applhed must have been
oblique to the section exposed rather than parallel with it.
At the intersection of some of the joints with the upper
fault plane a curved surface has been developed on the
joint planes, suggesting a drag movement so often seen in
connection with faulting. The direction and amount of
offset is easily seen, owing to the presence of the joints,
and corresponds quite closely with the fault at the Falls.

The face of the bluff trending as it does N 73° Ei gives a
very small angle of about 11° between it and a section
along the strike of the structure of the region, the axis of
the main fold being taken at N 85° W. One would, there-
fore, expect as in the previous case that by far the largest -
part of the movement would not be disclosed. As the
sluice ran at right angles to the bluff it was hoped that
some information in the third dimension might be gained
but the broken character of the shale had erased any evi-
dence if such ever existed. It was finally noticed that at
the intersection of one of the east-west and one of the
N 8° E joints where crossed by the dipping fault plane
about two square inches of the surface of the fault were
exposed; this showed an unmistakable dip into the bank,
that is to the south, but the surface was too small and
poorly situated to get any accurate readings. However
the dip south was apparently not less than 8° at the line of
outcrop. Presuming that this dip continues, regardless
of what its angle may be, it doubtless ultimately joins the
horizontal fault, from which it seems likely that it is a
branch. This being the case we have here a transverse
section of the two faults rather than a longitudinal sec-
tion as at first appeared. The movement of the wedge
block was therefore not primarily in and eastward, but
was northward. ‘The deviation to the west, of the forces
coming from the south which produced the faulting, plus
a slight torsional movement, would swing the joints out of
alignment to the extent of six inches and the compari-
tively mobile shale would accommodate itself to the

im the Cayuga Lake Region. QA5

oblique pressure when it could hardly have done so in the
face of direct compression.

The occurrence of the three above faults in such close
association with a large sized stream of extensive cutting
power brings up a question which may be raised by some.
Along the banks of many of the streams in the vicinity of
Cincinnati, Ohio, there occur in the Hiden shale small over-
turned folds, sometimes advanced to the stage of thrust
faulting. (Fig. 10.) These are usually found on the

Fic. 10—Fold in Eden shale, West Fork Cr., Cincinnati, O. Supposed
to be due to relaxation resulting from stream erosion.

under-cut side of terraces and are beginning to be recog-
nized as produced by relaxation due to rock removal by
stream erosion, rather than by direct pressure. ‘These
streams are all post-glacial whereas Salmon Cr., N. Y.,
runs in a valley recognized as pre-glacial. The horizon-
tal fault at the Falls has certainly no connection with a
terrace though the one by the bridge has some such rela-
tion and to a post-glacial terrace at that. However, they
appear to be so in accord with the rest of the evidence
concerning the diastrophic history of the region that it
hardly seems necessary to call in outside evidence for
their explanation.

A second type of faulting as shown in the shale mem-

Am. Jour. Sct.—FirtH Series, Vou. III, No. 16.—Apri., 1922.
18

246 EL. T. Long—Minor Faulting

bers of the Hamilton is a compound or slice fault devel-
oped on the south side of Stony Point some 25 miles down
Cayuga Lake. At this point a hard layer of limy shale,
much more compact and resisting than the surrounding
shale is sharply turned up at a 10° angle with dip to the
south. The faults occur just at the start of this high dip,
the greatest measured. For 200’ along the shore south of
the point, the N 12° W joints show horizontal faulting
with displacement of 1” to 114” of the joints of the H-W
system. ‘The movement has pushed forward each succes-
sive block between two faults from west to east. That is,
the western part of the zone has not moved so far north
as the eastern part. The N-S joints dip west a few
degrees from the vertical but the angle was not measured.
It probably does not exceed three degrees as other joints
of the region which are not vertical show a dip very gen-
erally around 2°. The faulted area covers the entire
width of the shore as exposed at this point and may exist
over an even greater width and length. The rocks of the
whole point and near to it are broken up by an excessive
number of subsidiary joints at all angles and often much
curved.

Mr. W. H. Bucher published an article on ‘‘The
Mechanical Interpretation of Joints,’’ in the Journal of
Geology for December of 1920. 'T'o one not acquainted
with the laws of mechanics the ‘‘interpretations’’ sound
quite convincing. His main thesis is, that in brittle sub-
stances the lines of greatest pressure will bisect the acute
angle formed by the shearing planes. He then applies
this law to three areas, one of which happens to be the
Cayuga Lake region. Not being acquainted in person
with the area, he relies for his calculations upon the data
given in an article by Miss Sheldon.? His conclusions,
put into figures, being that the compressive forces, must
he in a general N 35° EK direction since this line must
bisect the angle between the two major joint systems of
the region, the average of the greater part of whose
angles will run at about N 74° Hand N 5° W. This appli-
cation is here made to only the northern part of the area
which he discusses, but the results as stated are only a few

° Sheldon, P.; Some Observations and Experiments on Joint Planes, Jour.
Geol., vol. 20, No. 1, 1912.

in the Cayuga Lake Region. 247

degrees at variance with those for the larger area. As
proof of the applicability of his theory, he cites the pres-
ence of buckling as shown by a raised area in the northeast
part of the Catatonk quadrangle, and of a depression in
the central portion, near Jenksville. These are unques-
tionably present and quite possibly bear some relation to
the joints of the region, but they are not the only domes
and depressions which occur in the two quadrangles.’°
Following the axis of the Watkins anticline as he sug-
gests, the Portage-Chemung contact, starting at about
1520’ near the western boundary of the map, dips toward
Seneca Lake valley, and at Texas Hollow, to the east, has
risen to an elevation of 1480’ whence the beds continue
horizontal to a point two miles southwest of Enfield, the
last exposure west of Cayuga valley, along this line. To
the north near Reynoldsville, on the crest of the Firtree
anticline this contact is 1600’ as it is on the crest of the
Alpine anticline to the south and near Cayuga L. The
first contact shown east of Cayuga valley following the
axis of the Watkins anticline as stated, is at Pleasant Hill,
six miles east of Ithaca and 14 miles east of Enfield where
the contact is 1720’ or 240’ above the contact at Enfield.
Were the secondary synclinal fold in which Cayuta Lake
lies disregarded, there would be in this 14 miles a rise of
but 17’ to the mile, whereas the rise to the north from the
depression at Jenksville mentioned by him is over 43’ per
mile for 16 miles. The average of a number of southerly
dips north of the big bend of Cayuga Lake is about 45’ per
mile, steeper dips developing as the datum plane of the
Tully limestone is followed north. Owing to the lack of
geological and especially structural knowledge of the area
to the north of the Watkins and Catatonk quadrangles, it
seems hardly possible to form a very reliable opinion of a
district so near to the unknown. Work to the north along
Cayuga Lake this past autumn, together with observations

of the mapped quadrangles, would seem to indicate that
the greater southerly dips, even discounting what is due
to the folding process, are entitled to equal, if not greater,

**In a personal communication Mr. Bucher claims two sets of movements,
that producing the joints and buckling being prior to the one which devel-
oped the major folds and their accompanying faults. The horizontal frac-
ture planes he thinks may have originated during the first activities, only
the movement taking place later.

248 E. T. Long—Minor Faulting

significance than the pitch of the anticlines and synclines
and dips associated with a buckling process.

Judging from field evidence and acquaintance with the
region, Mr. Bucher’s results imply a large latitude in the
angle of his bisectrix as well as the presence of some
modifying or secondary forces not yet understood. Nev-
ertheless the old idea of compressive forces acting from
the southeast will apparently have to be revised in favor
of a southwesterly direction. This is probably due to the
lack of recognition accorded the short east-west segment
and may be considered a local exception. Here, two
entirely independent arguments, aiming to demonstrate
different ideas, have developed conclusions more closely
allied with each other than is either one with the usual
conception of the movements of Appalachian folding.

Ithaca, N. Y.

D. S. Jordan—New Species of Fossil Herring. 249

Arr. XX.—Description of a new Species of Fossil Her-
ring, Quisque bakeri, from the Texas Miocene; by
Davip Stare JORDAN.

Quisque bakeri, new species.

Fanily Clupeide.—Body oblong, compressed, rather short and
deep. Head heavy, about equal to depth, 314 times in length to
base of caudal, the snout bluntish; eye large, nearly as long
as snout, 3 3/5 in head; mouth large, oblique; maxillary about
reaching to below middle of eye; the lower jaw projecting, its
tip entering profile which is nearly straight. Vertebre rather
strong, striate, hour-glass shaped, about as deep as long, with

Fies. 1, 2.— Quisque bakeri Jordan.

strong ribs, neurals and hemals; the number of 13 + 13 = 26
with perhaps three additional vertebre crushed into the base
of the caudal; interneurals and interhemals obscure, the first
interhemal evident, rather well developed. Pectoral rather
short, placed low, eight rays traceable. Dorsal fin short, appar-
ently median, the rays obliterated; anal short, nearly obliterated ;

250 D. S. Jordan—New Species of Fossil Herrwg.

caudal broken, only the base preserved, the lower half with about
twelve rays; ventral fins lost; no trace of scales and no indica-
tion as to whether ventral or dorsal outline is serrated.

The type of this species, Catalogue number 401, Yale
University Museum, is a tiny fossil herring, with broken
fins, a little over an inch long, and about 1 1/3 inches to
tip of caudal, if complete. It was obtained by Mr. R. F.
Baker, from an oil well at a depth of 3,098 feet, at West
Columbia, Brazoria County, on the Gulf Coast of Texas.
It is in bluish clay shale, of the Lower Miocene, either the
Fleming or the De Witt formation, and is part of a boring
of the Texas Company Hoge Well No. 63. The species is
named after Mr. R. F. Baker, geologist of the Texas Com-
pany. |

The specimen was sent to the writer for study by Pro-
fessor Charles Schuchert. With itis a smailer fragment,
the reverse of the best preserved side, showing the poste-
rior part of the head, a pectoral fin and the vertebral
column to about the middle of the anal fin.

The obliteration of scales and scutes makes it impossi-
ble to locate the genus with accuracy. Among the small
herrings of the Miocene, it seems to come nearest to the
genus Quisque of the Southern California Miocene; the
large head, stout bones and large oblique mouth agreeing
with Quisque gilberts Jordan from Hl Modena, California,
the type of the genus.

E. W. Berry—New Genus of Fossil Fruit. 251

Art. XXI.—A New Genus of Fossil Frut; by Epwarp W.
BERRY.

Several families of the order Sapindales, which con-
tains about 20 families and over 3,000 existing species, are
abundantly represented in the geological record. This is
notably true in the case of the families Sapindacee, Ilica-
cee, Celastracee, and Anacardiacex, which are also the
largest existing families of this order.

A family that has not heretofore been recognized in the
fossil state is the Icacinacew. In the existing flora this
family consists of about 40 genera and 150 species, of
which only 8 genera with less than 30 species are found in
the Western Hemisphere, where they are, for the most
part, confined to the tropics. None of these American
genera except Mappia is found in any other continental
region, and in this genus the American forms are grouped
in the sub-genus Humappia and confined to the Antillean
region, and the Asiatic forms are segregated in the sub-
genus Trichocrater and confined to the region between
Ceylon and Farther India. It seems very probable that
these two sub-genera are not directly filiated but exhibit
convergent characters resulting from similar modifica-
tions of unlike ancestors.

The family is distinctly oriental at the present time,
and makes its greatest display around the borders of the
Indian Ocean, and its present representation in Africa on
the West and Australia on the East suggests an Asiatic
ancestry, with migrations from that region southwest-
ward over the now submerged Gondwana bank, and south-
eastward through the Kast Indian region. For example,
although there is only a single monotypic genus in addi-
tion to Trichocrater confined to Asia, there are 15 genera
with over 60 species found in the region extending from
southeastern Asia to Australia, and 3 genera with 20
species common to Asia and Africa. Africa has 10 gen-
era and about 35 species confined to that continent, and
Australia has 3 endemic genera with about 6 species.

There is only a single monotypic genus in northern
South America, and there are at least nine such in the Old
World, Three are African, one is Asian, two are Hast
Indian, one is confined to New Guinea, and two are con-

952 E,W. Berry—New Genus of Fossil Fruit.

fined to the hmited and otherwise peculiar region of New
Caledonia.

Some years ago I received casts of an unknown fruit
from a sandstone in the Wilcox Hocene exposed at the
Butler Salt dome in Freestone County, Texas. These
defied determination for several years, and I could find no
recent fruits at all like them until 1919. At that time, in
collecting fruits on the Pacific coast of Panama I obtained
a recent form that was very close to the fossil fruit. In
seeking to determine the former at the National Herba-

Fig. 1—Calatoloides eocenicum Berry.
Fig. 2—Calatola fruit from Panama, both natural size.

rium I found that it was identical with material from Mex-
ico and Costa Rica which Messrs. Standley and Stafford
were describing as a new genus, and referring it, with some
hesitation, to the family Icacinacee. This has since been
described as the genus Calatola, and three species—all
trees, are recognized. Two of these are Mexican and the
third is Costa Rican and Panamanian.

The fossil form from the Eocene of Texas is clearly
allied to this recent Central American genus, but in view
of the incompleteness of the material, and the impossibil-
ity of verifying the identity in all particulars, it has
seemed best to propose the new genus Calatoloides for the
reception of the fossil form—the name chosen serving to
suggest an ancestral relationship to the existing forms,

EB. W. Berry—New Genus of Fossil Fruit. 253

further borne out by the smaller size of the Hocene form.
It may be described as follows:

Calatoloides eocenicum gen. et sp. nov.

Based on fruit, which, as restored from casts, is a mod-
erately prolate spheroid in form, 2.5 to 3 centimeters in
length and about 2.25 centimeters in diameter, broadly
rounded proximad and bluntly pointed distad. The
sclerotesta, which in life appears to have been covered by
a thin sarcotesta, is igneous, and shows a characteristic
ornamentation similar to that of the fruits of the three
existing species of Calatola. Itis marked by a somewhat
irregular, prominent, branching and anastomosing series
of longitudinal ridges and these are connected by low,
subordinate, irregularly transverse, ridges. If it was
like the existing Calatola it contained a single large seed.
The Panama specimen was found in the sea drift on the
shore of Panama Bay, but the seed was dead and we know
nothing of the ability of the fruits of this genus to with-
stand journeys by sea, although the dry fruits appear to
be buoyant and impervious to sea water, so that ocean
currents may have been a factor in the distribution of the
Eocene ancestor.

In shape and ornamentation there is no difference
between the Hocene and the existing species except that
the fruits of the extinct genus are only one-half the size
of those of the existing genus, and the transverse ridges
between the longitudinal ridges are less prominent in the
fossil.

Any consideration of the origin and past distribution of
the members of this family must await a knowledge of its
fossil representatives, but it is not without great interest
that the exclusively Central American genus Calatola, not
recognized until 1921, should be represented by a closely
related form in the lower Hocene of northeastern T'exas
in a flora, a large proportion of which I have regarded
as having spread northward into southeastern North
America from equatorial America during the emergent
interval which followed the withdrawal of the Upper
Cretaceous sea from that area.

Johns Hopkins University,
Baltimore, Md.

254 Burling—Purcell Range and Rocky Mountains.

Arr. XXII.—The Relations Between the Purcell Range
“and the Rocky Mountains in British Columbia, Canada;
by Lancastrer D. Buruine.

During a trip along the Canadian Pacific Railway from
Field to Glacier, British Columbia, in 1915, the writer
formulated a working hypothesis to explain the strati-
graphic and structural interrelations of the Rocky Moun-
tain and Purcell Range sections. This differs from those
which have been proposed but succeeding observations
have more and more convinced him that the hypothesis
has elements of plausibility which warrant its presenta-
tion, together with such corroborative evidence as he is
able to reeall.

In order that we may understand the general relations
it may be stated briefly that the Rocky Mountains and the
Purcell Range, two mountain systems lying respectively
east and west of the Columbia River Valley at Golden,
differ markedly in the hthologic character of the rocks of
which they are composed. The western rocks are marked
by igneous intrusions, the sandstones are not so clean, and
there has been considerable metamorphism. In the
Rocky Mountains to the east there is little metamorphism
and igneous activity appears to be confined to single small
centers such as the Ice River valley occurrence south of
Field.

Speaking generally the Rocky Mountains section (to
the east) is dominantly calcareous; the Purcell Range
section (to the west and nearer to the source of the sedi-
ments) is dominantly arenaceous. And between the two
there is a relatively narrow belt of crumpled shales in
which the Rocky Mountain (Kootenay-Columbia River
valley) trench has been cut.

In the western portion of this intervening shale zone is
a fault-contact which represents the trace of the plane
along which the shales have been overthrust by the clas-
tics of the Purcell Range. (Students will recall that
Termier has here postulated thrust faulting of large mag-
nitude to account for the relations of the rocks of the
Purcell Range to those farther west.) Between the shale
zone and the Purcell clastics the change in lithology is ©
abrupt; eastward the change from shale to limestone is
eradual and progressive.

For a long time the rocks in the Purcell Range were
believed to be unfossiliferous, but there has been grad-

Burling—Purcell Range and Rocky Mountains. 255

ually accumulated by the members of the Canadian Sur-
vey considerable evidence that these rocks are at least in
part of Upper Ordovician (?), Devonian, and Upper
Paleozoic age. Paleozoic fossils younger than the Cam-
brian have thus been collected near Laurie, Lardo, Ward-
ner, ete., but so far as the writer is aware no Cambrian
fossils have been found west of the fault contact which we
have deseribed as occurring between the clastics of the
Purcell Range and the shale series trenched by the Colum-
bia River. This contact lies at a considerable elevation
above and to the west of the actual river, however, and
in the shaly series outcropping upon the slope between the
fault contact and the river are strata which have yielded
both Middle and Upper Cambrian fossils.

These two Cambrian fossil horizons were discovered by
Ami and Daly, respectively, in the section exposed by the
Canadian Pacific Railway, west of Donald, in the gorge
by which the Columbia River leaves its trench and crosses
over to the west some ten or fifteen miles north of Golden.
The Middle Cambrian fossils found by Ami are in the
collections of the Canadian Geological Survey. ‘They are
preserved as whole specimens and are curiously similar
to if not identical with the Ptychoparia kingz occurring in
the Middle Cambrian Wheeler formation of the House
Range in Utah (Smithsonian Mise. Coll., vol. 53, No. 5)
even to the cone-in-cone structure of the matrix surround-
ing the fossils. The Upper Cambrian fossils found by
Daly are also to be found at Ottawa and include forms
described by Walcott as Dicellocephalus dalyi and Illae-
nurus elongatus (Smithsonian Mise. Coll., vol. 64, No. 13,
pp. 367-368). They were collected from an outcrop occu-
pying unknown relations to the rest of the section and are
correctly referred to the Upper Cambrian by Walcott.
But the horizon of Daly’s Upper Cambrian fossils has
been found by the writer in place in the Goodsir forma-
tion which Walcott has referred to the Ordovician (Smith-
sonian Mise. Coll., vol. 57, No. 7, 1912, pp. 233-234, pl. 35).
This does not mean, however, that the reference is to be
changed from Upper Cambrian to Ordovician, for a large
part of the 6,000-foot Goodsir formation, including the
beds carrying the so-called ‘‘Ceratopyge fauna’’ of Wal-
cott, is of Upper Cambrian age. The Upper Cambrian
fossils found by Daly west of Donald are to be correlated,
together with the ‘‘Ceratopyge fauna,’’ with the Orr
formation of the House Range section in Utah (Burling,

256 Burling—Purcell Range and Rocky Mountains.

Summary Rept. Geol. Survey Canada for 1915, pp. 98-99,
1916). This places them fairly well down in the Upper
Cambrian instead of in the Ordovician. Less than one
mile east of Golden, in the gorge cut by the Kicking Horse
River, the writer has collected Normanskill graptolites
similar to those discovered by McConnell at Glenogle, 15
miles farther to the east.

We have, therefore, close together in the Rocky Moun-
tain (Columbia River) trench at Golden, and within the
single shale series which we have described as separating
the clastics of the Purcell Range from the calcareous
rocks of the Rocky Mountains, indisputable evidence of
the presence of strata representing the Middle Cambrian,
the Upper Cambrian, and the Ordovician.

This Cambro-Ordovician shale series is closely folded
and micaceous in places, and we have no evidence regard-
ing the thickness of strata here intervening between
the Middle Cambrian and Ordovician horizons above
described, but we do know that the same faunal horizons
are, in the Rocky Mountains less than 30 miles to the
east, separated by many thousands of feet of massive
limestones.

The conclusion we are about to draw must already be
evident to the student but it will be necessary first to
speak of Daly’s correlations of the rocks in the Purcell
Range with those in the Rocky Mountains. By him the
thick clasties of the Purcell Range are correlated with the
clastic basal or lower Cambrian portion only of the Rocky
Mountains section, a correlation based upon so much
study, both in the field and office, that we have hesitated to
question it. But the only gap in the evidence we have in
favor of a different conception of the relations is the fail-
ure, so far, to find Middle or Upper Cambrian fossils in
the Purcell Range clastics. And the other evidence is so
strong that we would suggest the view that the deposition
of the clastics of the Purcell Range (western), the shales
of the Rocky Mountain trench (central), and the dom-
inantly caleareous rocks of the Rocky Mountains (east-
ern) was essentially contemporaneous and that the
stresses to which the central shales have been subjected
has been relieved by crumpling in the shales themselves
and by the overthrusting of the more competent but
equivalent strata to the west. |

London, England.

Wells—Complex Chlorides contaaming Gold. 257

Art. XXIII.—Some Complex Chlorides contaamng Gold.
I. Pollard’s Ammonum-Silver-Auric Chloride; by
Horace L. WELLS.

[Contribution from the Sheffield Chemical Laboratory of Yale University. ]

Since numerous investigations upon double and triple
salts have been carried out in this laboratory by the writer
and his associates, attention was attracted to the descrip-
tion by William Branch Pollard' of a new triple chloride
to which he gave the formula (NH,),Ag,Au,Cl.,.. Since
this formula appeared to be a rather complex one, and
because Pollard used apparently drastic methods in pre-
paring the salt for analysis, including extracting it with
- ether and heating it sufficiently to volatilize ammonium
chloride, a new investigation of 1t was undertaken.

The salt was readily obtained, by following Pollard’s
directions, from a solution of 25 g. of gold in 50 ce. of
nitric and 150 ce. of hydrochloric acid, to which were
added 30-35 g. of ammonium chloride and 3 g. of silver
nitrate dissolved in 10 cc. of water. When a few more
crystals of ammonium chloride were added the silver
chloride precipitate that was present was gradually trans-
formed into the dark triple salt. Then upon heating the
salt with its mother-liquor there was a separation of sil-
ver chloride which, upon cooling, was replaced by the
triple salt.

This method of preparing the salt for analysis did not
appear to be very satisfactory on account of the concen-
trated condition of the solution and also on account of the
possible danger of the entanglement of silver chloride in
the product. It was found in the present investigation
that when solutions somewhat similar to the one recom-
mended by Pollard were diluted largely, perhaps with an
equal volume or more of 1:1, or stronger, hydrochloric
acid, it was easy to obtain clear, boiling solutions from
which the triple salt could be obtained either by cooling
or evaporation to crystaliization on the steam-bath, and
both methods of preparation, under wide variations of
conditions, have been used in obtaining the products for
analysis. The crops of crystals were usually washed to
some extent by pouring off most of the mother-liquor,
diluting the remainder rather largely with hydrochloric

* Jour. Chem. Soc., 117, 99, 1920.

258 Wells—Complex Chlorides contaonmng Gold.

acid, agitating and draining. In one ease (analysis VI)
where the crop had been deposited from a solution diluted
to an unusual extent with concentrated hydrochloric acid
no washing was done, for the sake of comparison. In all
cases the products were rapidly pressed between smooth
filter papers until the latter were no longer moistened by
the operation and then they were dried in the air. There
was practically no further loss in weight upon drying
ane IO.

The crystals obtained were always very small, usually
not over 1 or 2 mm. in diameter. The larger ones were
very dark brownish-red in color, but evidently transpar-
ent, while the smaller ones were not so dark. Pollard has
called the color purplish-brown. The crystals are beauti-
ful and brilliant, and a description of their orthorhombic
form is given in Pollard’s article.

The salt is quickly decomposed by water, but it appears
to be stable in the presence of strong hydrochloric acid, as
was observed by Pollard.

The following analyses of separate crops were made:

Prepared by Prepared by
hot evaporation cooling ;
—— _—“Galeulated for
1 iD, asl: IV. V. VI. (NH,) Ag,Au;Cl,;
IN Ee BT O22 ROO eee ee 7.13

Ae ...1431 1410 19161 1403 away 46 aoe
Au ... 38.89 38.80 39.67 38.84 38.54 38.62 38.96

Che £223 OG OO One aimee eee aoe 39.70
99.69 JOO LIES, 100.00
a Calculated from the excess of chlorine.
Calculated for Pollard’s formula
Pollard found (NH,) ,Ag,Au,Cl,3 differs from new one
NIECE 6.88 6.97 0.16
Di Coes cine Ree ae Laos 15.62 + 1.41
UL Seas cork DIY) 38.06 —— 0.90
Cee ere 39.44 BS) OD — 0.35
99.89 100.00

The analyses in the present investigation were made by
decomposing the salt with a liberal amount of water by
the aid of heat, cooling, collecting the silver chloride in a
Gooch crucible and weighing it, precipitating the gold in

Wells—Complex Chlorides containing Gold. 259

the hot filtrate by means of ammonium oxalate, and, after
hot digestion until the liquid was perfectly clear, collect-
ing and weighing the gold in a Gooch crucible, then acidi-
fying the last filtrate strongly with nitric acid, precipi-
tating silver chloride by means of silver nitrate, and
collecting and weighing the precipitate in the usual way.
Of course the chlorine in the first precipitate of silver
chloride was added to this.

From the evidence presented here it is believed that
the new formula is the correct one. It is true that
Pollard’s results agree very closely with his formula,
but he gives only one analysis, which may possibly rep-
resent selected results, while his method of prepar-
ing the salt for analysis, already alluded to, would
probably not remove any silver chloride from the salt, but
might remove small quantities of the other constituents.
In fact, it was his object to remove any gold chloride and
ammonium chloride that adhered to the crystals, so that
his results may reasonably be expected to show too much
silver chloride.

The new formula varies but shghtly from Pollard’s,
and it has little advantage over his in regard to simplicity.
The new investigation, therefore, confirms the fact that
this ammonium-silver-auric chloride has a composition
corresponding to a rather complex formula which appears
without doubt to be (NH,),Ag,Au;Cl,,;.

In view of the frequent similarity of ammonium and
potassium salts, an attempt was made to prepare a potas-
sium-silver-auric chloride, but after many experiments
under wide differences of conditions no evidence of the
existence of such a triple salt was obtained.

When cesium chloride was used, however, a triple salt
was easily obtained, and several other triple chlorides
containing cesium and gold, as well as a new double salt of
these two metals, have been prepared. This work will be
described in subsequent articles in this Journal.

New Haven, Conn.,
February, 1922.

260 T. Holm—Studies m the Cyperacee.

Art. XXIV.—Studies in the Cyperacee; by Toro. Hoto.
XXXIV. Carices aeorastachye: Glaucescentes nob.,
and Limoseé nob. (With 14 figures drawn from nature
by the author.)

Glaucescentes.
C’. glaucesens Kill. (Figs. 1-3).

The section seems, so far, to be monotypic, and C. glau-
cescens Ell. is confined to the southern States: Virginia
west to Missouri, and south to Florida. Regarding the
affinity of the species Drejer* writes:

‘* Affinitatem mihi accuratius inquirenti mox apparuit, ad
eregem Caricum aeorastachyarum. Habet enim spicam mascu-
lam solitarium; bracteas foliaceas (saltem. infimas), evaginatas,
subauriculatas; spicas femineas cylindricas, densifioras, pedun-
culatas, arrectas demum pendulas; perigynia membranacea,
nervata, arcte caryopsin includentia; rostrum brevissimum sube-
marginatum. C. glaucescens a ceteris sui gregis satis distincta
est squamarum insigni forma, quamquam nisus in eam in
squamis C. rarvflorae apparet. Inter americanas nullam scio,
quam huic proxime affinem posuerim, neque inter europaeas; in
India orientali autem plures affines videntur habitare.”’

By Kukenthal? the species is a member of his Paludose
(minume Fries), and placed between C. brasiliensis St.
Hil. and C. paludosa Good; nevertheless this same author
considers C.. Jooru Bailey to belong to the remote section
Maxime, although C. Jooru is identical with C. macro-
kolea Steud. and this C. macrokolea Steud. is by Kuken-
thal reduced to a mere form of C. glaucescens—the inevi-
table result of compilation without material for com-
parison.

Drejer gives an excellent illustration of C. glaucescens,
and from his diagnosis the points as follow may be
quoted:

‘*Perigynia matura squamis latiora et longiora, ovato-trigona,
erassa, caryopsin arcte includentia, membranacea, laevia, tenuis-
sime granulata, rostro brevissimo subemarginato. Caryopsis
brevis, crassa, obovata, trigona, lateribus excavata, stylo longo,
robusto, laevi, continuo, semper exserto terminata.’’

1Symbole Caricologice. Copenhagen, 1844, p. 14.
In Engler: Das Pflanzenreich, Leipzig, 1909. 733.

T. Holm—Studies wm the Cyperacee. 261

Am. Jour. Sci1.—FirtH Sertiss, Vou. III, No. 16.—Aprix, 1922.

262 T. Holm—Studies wm the Cyperacee.

According to Gray’s New Manual of Botany (1908)
C. macrokolea Steud. is considered specifically distinct
from C. glaucescens by the squamae being short-awned,
and the perigynium strongly ribbed; moreover the fioure
(1. c. p. 248) shows the pistillate spikes to be always ses-
sile; the geographical distribution is the same, but
extends to Texas. The section is thus confined to the
Southern States, and shows in several respects some affin-
ity to the Lamosae as regards the phyllopodie culms, the
bracts being evaginate, ete., as already pointed out by
Drejer (1. c.); but the habit is much more robust, similar
to large specimens of C. cryptocarpa.

Inmose.

In this section we have C. littoralis Schw., C. limosa L.,
C. laxa Wahlenb., C. rariflora Sm., C. stygia Fr., and C.
Magellamca Lam. They all are phyllopode, and bog
plants. With regard to their geographical distribution
C. littoralis is a very local plant in the Atlantic States
from Connecticut and southward, Maryland for instance.
C.. lumosa is widely distributed in the northwestern cor-
ner of this continent: Yukon, British Columbia, Wash-
ington, Idaho and Oregon; it occurs also in the Atlantic
States, in arctic Europe (Finmark and Russia) and
Siberia; farther south the species reaches Great Britain,
the Alps and Pyrenees, Altai and Baikal Mountains,
Korea, ete. C. lava, on the other hand, is a very rare
species in Northern Hurope: Finmark, Lapmark south to
Jamtland, arctic Russia, eastern Asia from Amur-district
south to J apan.

C. rariflora is circumpolar, and rare outside the arctic
regions; it has, however, been found in Quebec, Maine
(Mt. Katahdin), and in eastern Asia it extends as far
south as northern Japan. Its near ally C. stygia is not
rare in Alaska, and has recently been found in British
Columbia (Queen Charlotte Island); in HKurope it is
known from a few stations in Finmark and arctic Russia.
A very extensive distribution is shown by C. Magellanica,
throughout the northern hemisphere, including the arctic
coast of Finmark and Russia; as indicated by the name
the species occurs also in the most southern part of
South America.

T. Holm—Studies in the Cyperacee. 263

Thus with the only exception of C. littoralis all the
other species are known from the arctic coasts, and as
stated above, C. rarzflora is even circumpolar; it evi-
dently originated in the polar regions. The remarkably
scattered distribution of C. stygia with no mtermediate
stations between Alaska and arctic Scandinavia is an
excellent example of homologous endemism, where the
same species seems to have originated at two points,
extremely remote, but, nevertheless, associated with sev-
eral, closely related types; and these types are identical
at both stations. C. laxa shows a similar disconnected
distribution, being absent from northern Asia, except the
northeastern corner, being also absent from this continent
and Greenland; the occurrence of the species in Japan
may indicate the possibility of its existence also on this
continent; it closely resembles C. lumosa, and may have
been overlooked.

With reference to C. lumosa and C. Magellanica these
are undoubtedly of southern origin, having reached the
arctic region during the glacial epoch. Regarding C. lit-
toralis we have in this species a genuine American type of
very local distribution, and developed singularly distinct
from the other species, yet inseparable from these, to the
best of our judgment.

By examining these types more closely, we shall see that
the general habit and structure is very uniform; indeed
the only characters of prominence, which appear as really
important from a taxonomic point of view, depend upon
the more dense-flowered spikes in C. littoralis, the sheath-
ing bracts in C. laxa, and the gynaecandrous spikes in C.
Magellanica.

C. littoralis Schweinitz (Figs. 4-7).

According to Schweinitz and Torrey® Carex Barratti
Schw. et Torr. is the plant named C. littoralis by Schwein-
itz in 1824 (ibidem). It is also the same species which by
Carey* was referred to C. fiacca Schreb. (C. glauca Scop).

A very complete diagnosis and figure is given by Boott,’
from which the quotation as follows:

*A monograph of the North American species of Carex (Ann. of New
York Lyceum of Nat. Hist., vol. 1, p. 361, 1825).

*Gray’s Manual of Botany, 1857, p. 519.
° Til. genus Carex, vol. 1, p. 69, tab. CLXXXIX.

264 T. Holm—Studies in the Cyperacee.

‘*Spicis 4-5 rarius 3-6 eylindricis pedunculatis subapproximatis,
mascula saepius 1 purpurea demum ferruginea elongata erecta vel
2, reliquis foemineis apice masculis pendulis vel nutantibus rarius
erectis purpureis vel ferrugineis viridi pictis simplicibus vel
inaequalibus geminatis; bracteis evaginatis, superioribus breve
euspidatis, infima angustissima spica breviori, vel rarius foliacea
plus minus vaginata; stigm. 3; perigyniis ovalibus vel ovali-
lanceolatis obtuse triquetris saepe oblique divergentibus erostratis
obtusis vel abrupte vel sensim rostellatis, ore integro vel sube-
marginato, granulatis glabris leviter nervatis demum lutescenti-
bus apice ferrugineis, squama ovata obtusa vel subacuta mutica
fusco-purpurea vel ferruginea margine pallido nervo concolori °
longioribus.’’

The rhizome is very slender, stoloniferous; the phyllo-
podic culms measure a height of about 30-50 cm., and are
longer than the very narrow, glaucous leaves. We found
this interesting species in a cane-brake near Stony Run,
Md., where it was associated with C. stricta, folliculata
and canescens; it was quite abundant, and showed some
variation with respect to the size of the perigynia, some
specimens having the pistillate spikes heavier than
others; the habit was the same, however.

It is interesting to notice, as stated by Boott (1. ¢.), that
the pistillate spikes sometimes occur in pairs from the
same leaf-axil: ‘‘vel simplices, vel una vel altera vel
omnes geminate, rarius ternatae, quarum una abbreviata,
sessilis.’’

C. limosa L.

- The subterranean stem is ascending, relatively slender,
rooting at the nodes; there is generally one or two long
and thick brown roots at each node, beside some thinner
ones, more amply ramified; the subterranean leaves are
scale-like, and of a dark color. The length of the subter-
ranean stem may average about 40 cm. There are two
types of shoots: some, that develop a fascicle of long,
green leaves surrounding the terminal bud, which devel-
ops into a floral stem during the next year ; the other type
of shoots develops mostly scale-like leaves, and one or two
green ones, beside a flower-bearing stem. The latter type
6f shoots thus resembles the aphyllopodic, but differs
from this by the culms being actually central instead of
axillary. We may thus observe the same rhizome to bear
purely vegetative shoots (the first season), vegetative

T. Holm—Studies mm the Cyperacee. 265

shoots with a central inflorescence (the second season),
and finally floral shoots with only a very few, one or two,
green leaves, but with the base covered with several
brown, seale-like.

Regarding the inflorescence there is only one staminate
spike, and one or two lateral, pistillate; the subtending
bracts are not sheathing. Some deviations have been
observed, namely: the terminal spike may be androgy-
nous or gynaecandrous; the terminal spike may be the
only one developed, and is then purely staminate; the
lateral spikes are sometimes androgynous. While the pis-
tillate squamae are of a reddish-brown color in the typical
plant, there is also a form in which the squamae are very
pale, light-brown or yellowish. This pale form is men-
tioned by Anderson® as occurring in Seandinavia, and it
has also been found in New York, near Frankfort, Herki-
mer County, by Dr. Joseph V. Haberer.

C. lawa Wahlenb. (Figs. 12-14).

In his Flora Lapponica (1812) Wahlenberg gives an
excellent characterization of this species: ‘‘Spicis pendu-
lis subdensifloris remotis, bracteis vaginantibus foliatis,
capsulis oblongis obtusiusculis depressis squamas obtusas
aequantibus.’’ .. . ‘‘totum gramen flavescens est atque
valde molle et laxum fere ut in C. pallescente. Culmus
adeo tenuis et flaccidus ut fere decumbit.’’ The rhizome
is very slender; it is horizontal and stoloniferous; the
very thin stolons bear small, scale-like leaves, shorter than
the internodes, and they grow, close to the surface, in a
horizontal direction. With regard to the root-system, the
old rhizome bears many long, very thin roots. The floral
shoot shows the typical phyllopodic structure, none being
apparently aphyllopodie, as is the case of C. lamosa.

According to Hartman’ the terminal spike is sometimes
gynaecandrous, and in specimens from Lapmark we
notice also this spike to be androgynous; in other speci-
mens there may be from one to three single, pistillate
flowers in some distance below the terminal, staminate
spike. The oblong-elliptic perigynia are faintly nerved,
and a little longer than the oblong-ovate, obtuse or mucro-

° Skandinaviens Cyperaceer. Stockholm, 1849, p. 87.
* Handbok i Skandinaviens Flora. Stockholm, 1879.

266 T. Holy—Studies in the Cyperacee.

nate squamae. In C.lmosa the pistillate scales are a lit-
tle longer than the perigynia, and frequently cuspidate.

C. rariflora (Wahlenb.) Sm.

In this species the horizontally creeping rhizome is
more robust, and prominently stoloniferous; several very
long, thick, brownish roots proceed from the rhizome.
From the same rhizome several vegetative and flower-
bearing shoots may be developed, and the culms are typi-
cally phyllopodic. There is only one staminate spike, and
from one to four pistillate, but two are the most frequent.
The pistillate squamae are broadly ovate, obtuse to
mucronate, almost black, enclosing the broadly ovate,
faintly nerved perigynia.

Several forms have been described; according to Hart-
man (1. ¢.) the terminal spike is sometimes androgynous,
and Norman® describes the forms as follows:

‘fa. var firmior: Culmus robustior, elatior, usque ad 31 em. v.
ultra longus. Spicae femineae ovales v. oblongae, sat dense 10-17
florae, squamis saturatius rufofuscis (non piceis). Spica mascula
dilute flavo-fusea.’’

‘‘h. forma rufescens: Squamae spicae femineae rufo-fuscae
(non piceae).’’

‘fe, forma expallida: Spicae pallide flavo-fuscae, feminea
solitaria.’’

‘‘d. forma baeostachya: Spicae femineae obovato-rotundatae,
minimae vix 7: mm. superantes, 2-4 florae.’’

Finally Meinshausen® mentions a variety ‘‘brevi-
pedunculata,’’ in which the three to four pistillate spikes
are short-peduncled and contiguous.

C. stygia Fr. (Figs. 8-11).

In describing this species F'ries’® suggested that C. rart-
flora might represent a depauperate form; some years
afterwards Fries enumerated both as distinct species,"
and his diagnosis of C. stygia reads:

§ Norman, J. M. Florae arcticae Norvegiae species et formae nonnullae
novae vel minus cognitae. (Christiania Vidensk. Selsk. Forhdlgr. for 1893.
No. 16.)

® Die Cyperaceen der Flora Russlands. St. Petersbourg, 1901, p. 131.

© Novitie Flore Suecice, Mantissa III. Upsala, 1842, p. 141.

“Summa vegetabilium Scandinavie. Upsala, 1846, p. 234.

T. Holm—Studies in the Cyperacee. 267

‘‘Spica mascula solitaria erecta, femineis tristigmaticis 2-3
oblongis compactis longe pedunculatis pendulis, bracteis brevis-
sime vaginantibus subaphyllis, fructibus globoso-ovatis marginato-
ancipitibus turgidis nervosis obtusis, rostello tereti apiculatis,
squamis late ovatis convexis obtusis mucronatis obvolutis, culmo
acutangulo, foliis linearibus planis.’?’ And under C. rariflora
Fries states the reason why he keeps them separate: ‘‘ Praecedentis
quasi forma reducta, vix spithamam alta, nulli tamen adsunt
transitus et facie mox distincta. Squamae in utraque conformes,
piceae.’’

The species shows exactly the same habit as C. rar-
flora, but is remarkably robust, some culms from Queen
Charlotte Islands measuring a height of 65 em. The
number of pistillate spikes is mostly two, in nineteen
specimens out of twenty seven, all from Alaska and Queen
Charlotte Islands. It is remarkably constant wherever
it occurs, and the very local occurrence in contrast with its
near ally C. rariflora, not speaking of the total absence of
intermediate forms, seems to speak in favor of the opinion
of Fries that it is a species distinct from C. rarzflora.

C. Magellanica Lam.

That C. paupercula Michx., and C. irrigua Sm. are
merely synonyms we have discussed in a previously pub-
lished paper!?; we have also called attention to the fact
that the lateral spikes are gynaecandrous, as mentioned
already by Schkuhr, Boott and nearly all other caricog-
raphers. The Scandinavian authors Anderson (l. c.),
Hartman (1. c.), Blytt,° and Hjelt't do also point out
that the terminal spike is not always purely staminate, but
frequently gynaecandrous.

Besides by the distribution of the sexes the species dif-
fers from the others of this section by the more compact,
almost caespitose growth; by the leaves being broader;
the lowest bract reaching above the inflorescence; by the
pistillate squamae being spreading at maturity, and lan-
ceolate, with a long point. Moreover the perigynia are
broadly oval, much shorter than the squamae, and faintly
nerved.

“ Types of Canadian Carices. (Canad. Field. Nat., vol. 33, p. 75, 1919.)

* Norges Flora, Christiania, 1861.

*Conspectus Flore Fennice. Helsingfors, 1895, p. 298. (Acta Soe.
Fauna et Flora Fennica, vol. 5.)

268 T. Holm—Studies in the Cyperacee.

A variety pallens has been described by Fernald, with
scales green with pale brown or yellowish margins, and
this has also been reported from Finland by Hjelt (1. ¢.) ;
some specimens from Quebec, Ontario and Michigan
belong also to this variety.

It is thus characteristic of C. Magellamca that all the
spikes, at least the lateral, are gynaecandrous, and at the
same time being tristigmatic the species must be consid-
ered as the most evolute of the grex: Aeorastachye.

Clinton, Md., August, 1921.

EXPLANATION OF FIGURES.

Fic. 1. Inflorescence of Carex glaucescens Ell. natural size.

Fic. 2. Pistillate scale of same; enlarged.

Fie. 3. Utriculus of same; enlarged.

Fie. 4. Carex littoralis Schw.; the inflorescence; natural size.

Fig. 5. Pistillate scale of same, taken from the lower part of the spike;
enlarged.

Fie. 6. Pistillate scale of same, from the upper part of the spike;
enlarged.

Fig. 7. Utriculus of same; enlarged.

Fie. 8. Infiorescence of Carex stygia Fr.; natural size.
Fie. 9. Staminate scale of same; enlarged.

Fig. 10. Pistillate scale of same; enlarged.

Fig. 11. Utriculus of same; enlarged.

Fig. 12. Inflorescence of Carex laxa Wahlenb.; natural size.
Fig. 13. Pistillate scale of same; enlarged.

Fic. 14. Utriculus of same; enlarged.

A. F.. Rogers—Collophane. 269

Art. XX V.—Collophane, a Much Neglected Mineral; by
Austin F’. Rocsrs.

Collophane, amorphous calcium carbonate-phosphate
or carbonophosphate, must rank as one of the important
minerals, for it is the main constituent of phosphorite or
so-called phosphate rock, the production of which in the
United States for 1920 amounted to a little over 4 million
long tons, valued at about 25 million dollars. That it is
also a common and widely distributed mineral results
from the fact that fossil bones consist of collophane, as
was announced by the writer? in 1917.

The principal constituent of phosphorite? is usually
regarded as an impure massive variety of apatite, but
those that so treat it fail to realize that a given mineral
name connotes certain physical properties as well as a
given chemical composition. Apatite is the name used
for a hexagonal calcium fluophosphate mineral with a
specific gravity of about 3.2 and indices of refraction of
1.63-1.65 or in a wider sense for a group of minerals
including, in addition to the above, the corresponding:
chlorophosphate (chlorapatite), carbonophosphate (dahl-
lite), and oxyphosphate (voelekerite), all crystalline and
with about the same physical properties.

Now the chief constituent of most of the phosphorites is
amorphous and not crystalline. Both its specific gravity
and index of refraction are too low for apatite. It always
contains an appreciable amount of water which is
lacking in apatite? and besides, its carbonate content is
much higher than that of apatite. It is clear, then, that
the name apatite, either as a species name or a group
name, can not be used for the substance under. considera-
tion.

Some authors treat the amorphous calcium carbono-
phosphate in an appendix to apatite, and do not recognize
it as a definite mineral. Dana, for example, in the sixth

* Jour. Geol., 25, 531. A paper on the mineralogy of fossil bone is soon to
appear in the Williston Memorial Volume.

* Hereafter the term phosphorite is used as a rock name instead of the
more cumbersome term phosphate rock.

* As shown by the writer (Mineral. Mag., 27, 155, 1914), oxygen and not
hydroxl replaces fluorine and chlorine in some specimens of the apatite
group. To a mineral in which the compound 3Ca;(PO,)..CaO predominates,
the name voelckerite is used.

270 A. F. Rogers—Collophane.

edition of the System of Mineralogy,* says: ‘‘ Besides the
definte mineral phosphates, including normal apatite,
phosphorite [used here as a variety of apatite and not as
rock name], etec., there are also extensive deposits of amor-
phous phosphates, consisting largely of bone phosphate
(Ca,P.O,), of great economic importance, though not
being a definite chemical composition and hence not
strictly belonging to pure mineralogy. Here belong the
phosphatic nodules, coprolites, bone beds, guano, ete.’’

Does the amorphous substance that makes up the bulk
of the phosphorites deserve recognition as a distinctive
mineral? A mineral may be defined as a naturally occur-
ring, homogeneous, inorganic substance of definite or
fairly definite chemical composition and with character-
istic physical properties. Since the time that Dana’s Sys-
tem of Mineralogy was written, it has come to be recog-
nized that all minerals do not have a definite chemical
composition and that some variability must be allowed.
Among prominent examples may be mentioned pyrrhotite,
chalcocite, tetrahedrite, and nephelite. The main con-
stituent of phosphorite like all amorphous minerals is
admittedly variable in composition but it is definite within
certain limits as can be seen from the tabulated analyses
ona later page. Its properties are so characteristic that
it may be recognized by physical tests alone. The fact
that the phosphorites gerade into phosphatic limestones
and shales is no argument against the recognition of collo-
phane as a distinetive mineral, for opal grades into opal
shale and psilomelane grades ‘into indefinite manganese
dioxides.

The amorphous equivalents of crystalline minerals,
with the single exception of opal, have not been given
recognition as distinctive minerals until Cornu’s work in
1909. Recently the writer? has urged the adoption of
Cornu’s plan, though he believes that distinctive names
should be used instead of some of the names proposed by
Cornu. If such names as opal, psilomelane, limonite, and
halloysite are used for amorphous substances of variable
composition, then a distinctive name also should be used
for the amorphous calcium carbonophosphate. It bears
the same relation to crystalline dahllite (8Ca,P.Osg.
CaCO,) that opal does to chalcedony or quartz.

P69, 20892.
° Jour. Geol., 25, 515-541, 1917.

bo
of
posit

A. F. Rogers—Collophane.

The name, Collophane.

In 1870 Sandberger gave the name Kollophan to an
amorphous mineral containing calcium phosphate, cal-
cium carbonate, and water from the island of Sombrero
(West Indies). The calcium carbonate was believed to
be an impurity and so was deducted from the analysis
which on recalculation gave the formula, Ca,P,O;.H,0.

As Lacroix’ first pointed out, the calcium carbonate is
an integral part of the mineral. Specimens free from
admixed calcite effervesce vigorously when treated with
hot nitric acid. On calculation the Sombrero mineral
yields the formula 3Ca,(P.,0,).CaCO,3H,O, but it is sim-
ply a coincidence that the amount of water is 3 molecules.
Analyses of specimens from other localities prove that
the water is variable. The ratio of calcium phosphate to
calcium carbonate is also variable and some specimens
contain fluorine and the sulphate radical.

Dana changed Sandberger’s name Kollophan to collo-
phanite. The writer prefers collophane, which is simpler
and more euphonious. As Cole’ has well said: ‘‘... the
terminology of minerals formerly possessed, for the
founders of the science, as agreeable a variety as that of
other branches of natural history. There seems no need
to make technical language har sh by the undue repetition
of sounds that have no historic warrant.’’

Synonyms of Collophane.

Kollophan, Sandberger, 1870.
Apatite (in part). Most authors.
Collophanite, Dana, 1892.
Colophanite, Lacroix, 1910.
Fluocolophanite, Lacroix, 1910.
Collophane, Rogers, 1917.
Pyroclasite, Shepard, 1856.
Pyroguanite, Shepard, 1856.
Glaubapatite, Shepard, 1856.
Sombrerite, Phipson, 1862.
Monite, Shepard, 1882.
Floridite, Cox, 1890.

Quercyite, Lacroix, 1910.
Odontolite (Bone turquois)
Nauruite, Elschner, 1913.

® Neues Jahrb. Min., 308, 1870.
7 Comptes Rendu, 150, 1213, 1910.
® Outlines of Mineralogy for Geological Students, p. 169, London, 1913.

272 A. F. Rogers—Collophane.

It is reasonably certain that all of the above listed min-
erals are one and the same. Some of the names have
priority over collophane (Kollophan), but one of the limi-
tations of the law of priority set forth by Dana? is that a
name may properly be set aside when it ‘‘has been lost
sight of and has found no one to assert its claim for a
period of more than fifty years. ...’’ Collophane, or
rather its equivalent, collophanite, on the other hand, has —
been given prominence by Lacroix in his excellent trea-
tise?® on the minerals of France.

Pyroclasite and pyrogaunite are simply varieties of
hard guano. As they contain about 80 per cent of trical-
clum phosphate, they are doubtless collophane. In a
later paper Shepard says that glaubapatite is the same as
pyroclasite.

Monite is an earthy variety of collophane. Shepard’s
analysis does not show any carbon dioxide but specimens
free from admixed calcite examined by the writer show
decided effervescence in nitricacid. The sulphur trioxide,
which was deducted by Shepard, is probably an integral
part of the mineral, for specimens free from admixed
gypsum give a good microchemical gypsum test. The
collophane from other localities sometimes contains the
sulphate radical and so does apatite also for that matter.

Sombrerite is from the type locality for collophane, but
it is a phosphatic replacement of coral, while the original
collophane is an opal-like substance not due to direct
replacement. A specimen of sombrerite in which coral
structure is evident has been examined by the writer. It
agrees in all essential respects with collophane. For
example, it is soluble with effervescence in nitric acid,
but is entirely free from calcite.

Q@uerecyite is undoubtedly a variety of collophane and
not a mixture of collophane with a crystalline substance. —
A specimen in my possession agrees exactly with
Lacroix’s'! description and figures, but the apparently
crystalline layers, though doubly refracting, are really
amorphous. The analyses of quercyite together with
the microscopic examination prove its identity with
collophane.

° System of Mineralogy, 6th edition, p. xliii, 1892.

10 Mineralogie de la France et de ses Colonies, vol. 4, pp. 561-586, 1910.
11'Mineralogie de la France, pp. 579-581, figs. 1-2, 1910.

A. F, Rogers—Collophane. 273

Odontolite, also known as bone turquois or occidental
turquois, is an aluminous variety of collophane, according
to Lacroix.?”

The nauruite of Elschner? is typical collophane. The
formula 3CaP,0,.Ca(OH,F'),. is assigned to it, but speci-
mens presented to Stanford University by Dr. Elschner
contain about five per cent carbon dioxide and are entirely
free from calcite, aragonite, or dolomite. This beautiful
brown, banded, resinous mineral from the island of Nauru
is the best and most typical form of collophane that it has
been my privilege to see.

The Chemical Composition of Collophane.

There are very few complete analyses of phosphorites
on record; most of the analyses are evidently commercial
ones in which insufficient care was made in selecting the
material. The writer has with some difficulty succeeded
in collecting five fairly complete analyses of collophane,
which were probably .almost free from mechanical
impurities.

Analyses of Collophane.

Fe,0Os,
CaO MgO Al,O3 jee Os CO. F, SO, H.O Insol. Total
1. 50.70 0.80 39.10 3.96 5.02 99.58
2. 92.47 0.21 - 0.53 38.72 T.88 1.60 0.22 5.00 99.65
3. 49.73 0.50 387.40 3.75 0.88 7.05 99.24
4, 51.85 37.60 4.00 1.50 4.80 99.12
5. 50.97 0.2 0:76. 36505. “L.72 0.40 2.98 1.05 1.82 99.04*

1, Sombrero. - Nauru. 98, Pouzillac, France. 4, Mouillac, France. 5,
Crawford Mts., Utah.

*Includes 2.00 Na.O, 0.47 K,0 and 0.30 SiO..

Ratios from Collophane Analyses.

CaO MgO Fe.O; J Ub P20; CO. F3 SO, HO
1. 904 20 -- — 215 90 — — 27
2. 936 5) a — 272 43 84 3 278
3. 886 a 2 3 263 85 46 _ 392
4. 924 — 2 3 264 91 39 a 266
5. 908 5 2 6 256 39 21 37 58 ;

Na.O = 32 ; ioe &

7” Mineralogie de la France, p. 577, 1910.
* Corallogene Phosphat-Insel Austral Oceanien und Thre Produkte, Liibeck,
1913.

O74 A. F. Rogers—Collophane.

On combining these, they yield the following :+

3 Ca3(PO,4)e° 0.98 CaCO>: 0.01 CaO: 3.02 H.O. 2

3 Cas(PO.)o° 0.47 CaCO;: 0.46 CaF.° 0.03 CaSO,° 0.87 CaO: 3.05 H.O.

3 Ca3(PO.)s° 0.97 CaCO;: 0.26 CaF. 4.45 HO:

3 Ca3(PO.)o° 1.08 CaCOs3° 0.21 CaF. -0.25 CaO: 3.02 H;0. :

3 Cas(POx.)2° 0.46 CaCO;° 0.13 CaF." 0.45 CaSO.: 0.94 CaO: 0.68 H.O
+In computing these ratios magnesium, iron, aluminum, sodium, and potas-

sium have been combined with calcium.

SH 99 10

The above may be written in another form:

3 Cas3(PO.)2° 0:99 Ca(COs)° 3.02 Heo:

3 Ca3(POx)3° 1.338 Ca(CO3,F2,SO.): 3.05 H.O.

3 Ca3(PO,)e° 1.28 Ca(COs, Fs): 4.45 H.O.

3 Casa( ROWs: 1.29 Ca(COs3, F2.O0): 3.02 HO:

3 Cas(POx.)s° 1.98 Ca(COs3, F2,S04,0)- 0.68 H,.O.

The ratios vary from 3 Ca,(PO,),.CaX to 3 Ca,(PO,)>.
2 CaxX.

That these formulae represent the chemical composition is con-
firmed by three analyses of collophane of fossil bone with only
0.1 to 0.2 per cent of impurities recently made by K. 8. Boynton
under the writer’s direction. On ealculation the following
formulae are obtained : |

3 Ca, (PO,), .0.99 Ca(CO,,80,,0). 3.08H,0
3 Ca, (PO,), 1.48 Ca(CO,,F,). 2.18H,0
3 Ca, (PO,), 1.78 Ca(CO,,F.). 1.88H,0

Cee Oe

It seems clear that we have here a case of solid solu-
tions of calcium carbonate, ete. in tricalcium phosphate.
In one of these analyses (No. 5) calcium oxide predomi-
nates over calcium carbonate, and Artini'® has described
a ‘‘fluocollophanite’’ from Palestine which is near fluor-
apatite in composition. The sulphate radical is present
in some specimens of collophane and it is probable that a
crystalline mineral with the formula 3 Ca,(PO,),..CaSO,
will be found some time. Such names as fluocollophane
are not advisable. The fine distinctions made with crys-
talline minerals, for example, voelckerite, dahllite, etce.,
are hardly warranted in the case of amorphous minerals.
So, then, collophane may be used for the amorphous
equivalent of any member of the apatite group, in which
the compound 3 Ca,(PO,). predominates.

The Sombrero collophane is almost exactly 3 Ca,(PO,)>.
CaCO, and several of the others approach 3 Ca,(PO,)..Ca

16 Abstract in Zeit. f. Kryst. u. Min. lv, 320, 1915.

A. F. Rogers—Collophane. 275

(CO,,F.,S0,,0). Dahllite, which is crystalline, has the
formula 3 Ca;(PO,)..CaCO;. The fact that most amor-
phous minerals approach in chemical composition their
erystalline equivalents was brought out by Cornu.** This —
principle he called the law of ‘‘homoisochemite’’, and the
minerals themselves, ‘‘pseudostochiolite.’’

The water evidently is also variable. Water is varia-
ble in all amorphous minerals, on account of their col-
loidal origin. When the colloid is first formed the water
may be absorbed but during the hardening of the gel the
water probably becomes diffused through the dispersed
material and thus in time a solid solution is formed.

Collophane, then, may be regarded as a solid solution of
calcium carbonate, calcium fluoride, calcium sulphate, and
water in tricalcium phosphate.

Physical Properties of Collophane.

Collophane is usually massive, but sometimes has an
oolitic or concretionary structure. As is the case with
other amorphous minerals, colloform'* crusts may be
present in cavities.

The specific gravity of collophane varies from 2.6 to
2.9; the variation is due to variation in porosity as well as
to difference in chemical composition.

The hardness varies from 3 to 5. The index of refrac-
tion varies about 1.57 to 1.63, but is usually 1.59 to 1.61.
The determination of the index of refraction is one of the
best means of identifying the mineral, for, with the excep-
tion of the very rare silicate, eudialyte, it is practically
the only amorphous or weakly doubly-refracting mineral
within these limits for its index of refraction. Like other
amorphous minerals of colloidal origin collophane often
shows double refraction due to strains set up in the hard-
ening of the gel. In colloform and odlitic forms, the
double refraction causes the mineral to take a spherulitic
appearance. These pseudo-spherulites, however, lack
the fibrous structure of true spherulites.

™ Zeit. f. Chem. u. Ind. d. Kolloide, 4, 15, 89, 1909.

8 This term was coined by the writer to designate the rounded, more or
less spherical surfaces assumed by amorphous and microcrystalline substances
in open spaces. It is a general term to cover mammillary, botryoidal, etc.

276 A. F. Rogers—Collophane.

Pyrognostic and Chemical Tests.

_ Before the blowpipe, collophane fuses with difficulty on
the edges, glows, and turns white. In the closed tube it
turns dark (on account of organic matter) and gives a
fair amount of water.

It is soluble in cold nitric acid with fair effervescence
and in hot acid there is vigorous effervescence.

Summary.

The principal constituent of phosphorites or so-called
phosphate rocks and also of fossil bones is an amorphous
substance with properties sufficiently characteristic to be
entitled to recognition as a distinct mineral. It may be
called collophane.

Collophane consists largely of tricalcium phosphate
(Ca,P,O0,) with smaller amounts of calcium carbonate,
calcium fluoride, calcium sulphate, and calcium oxide. It
may be regarded as a solid solution of the latter-named
substances in tricalcium phosphate. Like most other
amorphous minerals it is of colloidal origin and contains
an indefinite amount of water. The formula of collo-
phane may be written 3Ca,(PO,),.1 Ca(CO;,F.,S0,,0).
(H,O)x, where ” is an indefinite number varying from 1
to 2. The carbonate radical usually predominates over
the other minor constituents and thus it often approaches
crystalline dahllite (83Ca,P,0,.CaCO;) in composition.

Sp. Gr. = 2,642.9; H==3 to 5; #157 to 6s70ren
with weak double refraction.

Collophane is soluble in nitric acid with effervescence.

Stanford University, September, 1921.

Thorpe—New Genus of Oligocene Hyenodontide. 277

Arr. XX VI.—A New Genus of Oligocene Hyenodontide;
by Matcotm RUTHERFORD 'T'HORPE.

| Contributions from the Othniel Charles Marsh Publication Fund, Peabody
Museum, Yale University, New Haven, Conn. |]

TABLE OF CONTENTS.

Introduction.
Description of species.
Neohyenodon, gen. nov.
N. horridus (Leidy).
Hyenodon Laizer and Parieu.
H. cruentus Leidy.
H. montanus Douglass.
H. crucians Leidy.
H. leptocephalus Scott.
Synoptic table.
References.
INTRODUCTION.

All of the North American species of hyznodonts are of
Middle Oligocene age. This group embraces diverse
forms, one of which it is now proposed to place in a new
genus, herein designated as Neohyenodon. The Yale
specimen, Cat. No. 12766 (see figs. 1 and 2), and Leidy’s
‘““Hyenodon horridus,’’ figured in 1869 (pl. III), are
selected as cotypes.

Three skull forms are represented among the American
Hyenodontide: (1) the strongly dolichocephalic form,
which is typical of Hyenodon, found in species seldom
exceeding the wolf in size; (2) the mesaticephalic torm,
typified in the New World by H. paucidens; and (3) the
large dolichocephalic type with a great range of verti-
cal jaw movement and a considerable development of
the canines, with skull modifications, similar in many
respects to those of the sabre-tooth cat Smilodon. This
third group is the one now termed Neohyenodon by the
author. The first group might well be subdivided, sep-
arating out the species with the extreme posterior open-
ing of the choane, the group that is represented by
Hf. leptocephalus. |

The smallest species of Hyenodon now known,
Hf. mustelinus, is apparently not represented in the Yale
Collection. The short-faced group, of which H. pauci-
dens is the only American example, is by no means plenti-
ful in this Museum, whereas there are many individuals
of the long and slender-jawed type, representing all but

Am. Jour. Sci.—FirtH Serizs, Vou. III, No. 16.—APRIL, 1922.

278 Thorpe—New Genus of Olugocene Hyenodontide.

the smallest species. H. minutus Douglass is omitted
from the synoptic table, for the type consists of only a
right M, and the taxonomic position of the species is
uncertain. Matthew has referred it to Pseudopterodon,
and yet Scott believes that this genus was established on
the milk dentition of Hyenodon. This M, probably
belongs to a creodont, but its reference to Hyenodon
seems to me to be doubtful.

DESCRIPTION OF SPECIES.
Neohyenodon, gen. nov.

Distinguishing Characters.—Larger than Hyenodon,
dolichocephalic, glenoids far below basicranial plane,
basicranial region foreshortened, dentition similar to
Hyenodon, except for the antero-external buttress on the
paraconid of M.. :

Neohyenodon horridus (leidy) 1853.
(Fies. 1, 2.)

Y
rw ; =
‘ \

: is, Yj WM
HH), WH
/ ‘

Yj
MMi Y pS pal
y Rea Wak

EN NS

12766 TYPE =A)
Y. P.M.

Fic. 1—Neohyenodon horridus (Ueidy). Left lateral view of skull.
< about 1/3.

T'wo very excellently preserved skulls with jaws repre-
sent this genus and species in the Marsh Collection. Cat.
No. 12766 has been selected as a cotype of the genus. It
was collected in South Dakota, and, while fully mature,
yet its sutures are clearly shown in detail. The other
specimen, Cat. No. 10074, collected near Red Clond,
Nebraska, is an old individual and is unique in that it is
the largest complete skull of this species so far reported.

Thorpe—New Genus of Oligocene Hyenodontide. 279

The total skull length is 345 mm., and the length from the
prosthion to the occipital condyles inclusive is 318 mm.
The bone is very heavy and massive, and in places quite
rugose, as, for example, on the postorbital processes of
the frontals.

The condition of wear of the teeth is noteworthy in that
it shows such marked differential abrasion. One of the
most interesting conditions is shown by the left P* which

Yoke oli: = |
Uy = iG ESS
; ta AGS ; \2 Zi A if | . jee " ct
mn \ : SS x NS SX < p> aM Lad e. Ss
aot ° Awe nN ~S : > a) =
Z Sa eS Ne Ws YY = f,

Fic.- 2—wNeohyenodon horridus (Leidy). Left half, palatal view.
x< about 1/3.

-now consists of but three rounded stumps, the entire
crown having been worn away, while the left M', although
considerably worn, yet descends at least 15 mm. below the
surface of P#. On the same side, the P? shows about twice
as much wear on the anterior half, or it is worn so that the
crown has a decidedly oblique surface sloping forward
and upward. Both P® are placed obliquely in the maxil-
lary, with the anterior part inward. A|ll of the superior
teeth are crowded, except both P!, which are isolated, the
minimum diastema being posterior, with a diameter of
4mm.

All of the inferior teeth are crowded, except P,. The
third lower molar has a very effective shearing surface,
although the crown is considerably shortened vertically
by wear, so that all trace of the fore and aft divisions is
obliterated and the remaining surface is one long, nearly
level blade. The second lower molar is low and rounded,
with no trace of a shearing surface, while M, consists of
two rounded stumps nearly level with the alveolar parapet.
The contrast is quite striking between this stump of M, and
the almost unworn contiguous right P,, extending upward
21 mm. higher than M,. The length of the symphysial
suture is 116mm. This great length is a compensatory
mechanical device for overcoming the slenderness of the

280 Thorpe—New Genus of Oligocene Hyenodontide.

rami. The posterior position of I, is typical of the
creodont carnivorous type. It is quite sige that —
skull and jaws belonged to a male.

Measurements of Dentition.
(Cat. No. 10074, Y. P. M.)

mm.
Superior molar-premolar series, length ................. 137
Superior molar series, lemoth. 222s ee 50.5
Inferior molar-premolar series, length ................. 141
Inferior molar seriesslenetheeeie 2, ee 66

A single right M,, in part of the jaw No. 12765, Y. P. M.,
from White River, Nebraska, indicates an individual
approximately as large as the one described last above.
It measures 35 mm. fore and aft, with a maximum trans-
verse diameter of 13.5 mm.

The cotype, No. 12766, Y. P. M., shows very clearly the
cranial foramina. The auditory bulle are lacking and
I am not sure whether or not any were ever present. If
this genus did have them, they were circular in basal out-
line, with a diameter of approximately 18.5 mm. The
maxillo-palatine suture is obtuse at its anterior margin,
which les opposite the middle of P*; the postorbital con-
striction is at the coronal suture; the posterior nasal
bones are acute, extending nearly to a line through the
middle of the orbits; the lacrymal is relatively small,
extending somewhat in front of and below the orbit, and
bears a slight depression, which appears greater than it
actually is due to the elevation of the orbital margin on
which is a prominent lacrymal tubercle; the infra-orbital
is quite large; and the hamular process is represented by
a distinct rugosity, approaching very close to a tubercle
in character.

Comparison with Hyenodon.—Neohyenodon difters
from Hyenodon in several distinct ways, among others in
the development of an external buttress-like ridge on the
paraconid of the last lower molar, but the most important
difference lies in the structure of the basicranial and con-
tiguous regions. ‘The superior part of the glenoid articu-
lar surface is more than 15 mm. below the basioccipital,
and in part hes below the bulla, while the postglenoid
tubercles are on a line only 14 mm. anterior to the basion.

Thorpe—New Genus of Oligocene Hyenodontide. 281

In other words, the basicranial area is foreshortened and
the glenoid surfaces are far below the basioccipital. In
Hyenodon, these articular surfaces are practically in a
plane with the basisphenoid, that is, they are much higher
and more anterior, and the relations are much more typ-
ically mammalian and carnivoroid than in Neohyenodon.

Comparison with Smilodon.—This new genus exhibits
many details of structure like those in Smilodon, but at
present we shall consider only similarities and differences
which are due to the mechanics of the jaw movements.
For comparison I have used a fully adult skull, Cat. No.
10204, Y. P. M., of Smilodon califormcus Bovard, from
the Rancho La Brea, California. The glenoid articula-
tions lie much below the level of the basicranial plane, and
the zygomatic arches are short and relatively weak in both
Neohyenodon and Smilodon. The lowering of the
glenoid permits a wide gape of the jaws before the pro-
jection of the angle impinges upon the posterior surface of
the postglenoid tubercle. This tubercle in both genera
has the same form below the glenoid surface, 1. e., it is
moderately long internally, and its inferior margin slopes
upward externally to the level of the articular surface. —
Moreover, there is no inflection to the mandibular angle in
Neohyenodon, and it does hot turn outward to the same
extent as in Smulodon, but the former genus did not need
to open the jaws so widely as did the latter.

The coronoid process is rather small and not high,
a feature which is correlated with the other changes tend-
ing to allow the required great gape. The temporalis
muscle, attached to the coronoid, and originating on the
occipital and sagittal crests and on the parietal and
Squamosal bones, was of great length and therefore prob-
ably of small leverage, but compensation for this is in part
accomplished by the posterior position of the cutting
teeth. The masseteric fossa is moderately large and
rather shallow, while the zygomatic arch is short and
weak, from which it would seem that the masseter muscle
had similar power and functions both in the creodonts and
in the true Carnivora. The pterygoid muscles, acting
mainly with the masseter, originated slightly more ante-
riorly than in Smilodon. Of the three sets of muscles,
temporalis, masseter, and internal pterygoid, each alone
was relatively weak in the power of closing the jaw, but

282 Thorpe—New Genus of Oligocene Hyenodontide.

the three acting in unison were powerful. They are all
situate near the glenoid, and afford a great are with com-
paratively small muscular contraction and expansion,
except in the case of the temporalis.

The paroccipital process lies considerably posteres to
the glenoid, and furnishes the digastric muscle with an
origin well adapted for wide jaw gape. ‘The mastoid in
Neohyenodon is very small and located on a line with the
anterior part of the occipital condyles. In contrast to
Smilodon, its small size must indicate a rather weak
development of the cleido-mastoid and sterno-mastoid
muscles, the chief function of which is to pull the head
downward upon the neck.

Comparison with Dinoceras.—Although Neohyenodon
and Smilodon are of widely separated geologic horizons
and of different orders, yet they were both carnivorous in
habit and had developed skull characters of great sim-
larity. Now let us examine a representative of another
group, totally distinct in genetic relationships and
habits from either of the above, namely, Dinoceras, sev-
eral skulls of which in the Marsh Collection have served
for comparison. This genus possessed long superior
canines and was therefore compelled to have a wide jaw
gape. In this form the peculiar glenoid articulation is
shown, although it has not: descended below the basicran-
ial plane. ‘The temporalis muscle was long and of small
leverage, while the zygomatic arch is apparently rela-
tively smaller than normal in herbivorous animals. The
condyle of the ramus shows a peculiar modification, as
well as the outward curvature of the whole ramus just in
advance of the condyle, to permit of a wide gape. An
interesting development of the ramus is the hoplopho-
neoid character shown in the long decurved processes for
the protection of the superior canines when the mouth is
closed. Protoceras is analogous to Dimoceras in these
jaw mechanics.

The primal cause for the necessity of a wide jaw gape
is great canine length. Smilodon and Dinoceras pos-
sessed very short inferior canines relative to the great
length of the superior, whereas the upper and lower
canines of Neohyenodon were of more nearly equal
length. In Dinoceras mrabile the combined length of the
upper and lower canines is 195.5 mm., the superior being

Thorpe—New Genus of Oligocene Hyenodontide. 288

176 mm. long; in Smilodon califormcus the total length 1s
174 mm., with the superior canine 154 mm., while the total
length i in N eohyenodon horridus is 100 mm. , about equally
divided between superior and inferior canine. Another
point of comparison is in the mesaticephaly of Smilodon
as compared with the dolichocephaly of Dinoceras and
Neohyenodon. Still another point to consider is that the
condyle of the ramus is situated a little below the superior
edge of the alveolar parapet in Smilodon and in Neohy-
enodon, but on a line with the tooth-row in Dinoceras.
In short, we see in these genera how a certain specific
end-result—the accomplishment of great jaw gape—was
attained by three different modes of skull and jaw
mechanical modifications. In the three genera there are
similar modifications, but the divergences are equally well —
marked. In Neohyenodon the basicranial area is fore-
shortened, the glenoids are lowered and the mandibular
condyle is but slightly above the angle. In Smilodon the
basicranial region is not foreshortened, but the cranium
elevated, the glenoids lowered, relative to the basicranial
plane but not to the palatal, while the mandibular condyle
is likewise low. Smlodon evidently was the only one of
these genera under consideration that struck downward,
hence the form of the skull. In Dinoceras the basicranial
region is somewhat foreshortened, to a lesser extent than
in Neohyenodon, while the olenoid has not lowered, but
the mandibular condyle has become elevated with respect
to the tooth-row.

Hyenodon Laizer and Parieu 1839.
Hyenodon cruentus Leidy 1853.

A skull and anterior part of both rami (Cat. No. 12762,
Y. P. M.) of this species were collected near Red Cloud,
Nebraska. It is a young individual, with the permanent
canines just beginning to appear. The lower incisors
show that the first was not more than half the size of the
second, which was nearly as large as the third. The two
second incisors have their usual position posterior to the
line of the others and are separated from each other by a
diastema of only 1.5 mm.

Another specimen, Cat. No. 12764, Y. P. M., from
Chadron, Nebraska, has. but slightly worn dentition and

284 Thorpe—New Genus of Oligocene Hyenodontide.

the lower third molar shows a very small incipient exter-
nal buttress on the anterior lobe, but it is not of the same
character as in N. horridus.

Hyenodon montanus Douglass 1901.

Both jaws of two individuals are referable to this spe-
cles. One pair of rami, Cat. No. 12767, Y. P. M., was col-
lected at White River, Nebraska, and the other pair, Cat.
No. 12768, Y. P. M., in Gerry’s canyon, Gerry’s ranch,
Colorado. The latter has nearly all the teeth present,
and shows the characters mentioned by Douglass as dif-
ferentiating this species. ‘The second incisor is as large
as the third. Its anterior part is wedged between the
posterior parts of I, and I,, that is, it is not situated so
far posteriorly as in N. horridus and H. cruentus. The
premolars especially, and the molars to a less extent, are
relatively higher crowned than in the other species of this
genus.

Hyenodon crucians Leidy 1853.

Part of a left ramus with P,, M,, and M,, Cat. No. 12770,
Y. P. M., collected at Gerry’s ranch, Colorado, is referable
to this genus. -

Specimen No. 10076 Y. P. M., collected near Hermosa,
South Dakota, consists of a skull, Jaws, many vertebre,
and many parts of the appendicular skeleton. These
bones do not show any marked deviations from the
description, given by Scott in 1894, of the osteology of
this genus.

Hyenodon leptocephalus Scott 1887.

A well preserved skull with jaws, together with an atlas
and one lateral metatarsal, Cat. No. 10075, Y. P. M., are
referred to this species, as the skull exhibits the long nar-
row cranium and the sutural union between the pterygoid
processes of the alisphenoid bone. Measurements of the
type of this species are lacking, and for these we must
derive our information from the statement that ‘‘in size
it slightly exceeds the H. crucians of Leidy’’ (Scott
1887, p. 152). |

Nearly all sutures in the Yale specimen are plainly visi-
ble, including the diagnostic character of the union of the

/

Thorpe—New Genus of Oligocene Hyenodontide. 285

pterygoids posterior to the suture joining the alisphenoid
with the palatine bones. The hamular process is repre-
sented by a prominent rugosity. The lacrymal bone is
more triangular in outline and relatively of somewhat
smaller extent than in N. horridus. The external wall of
the infra-orbital foramen bears a forward-projecting lip
which is not present in any other species, so far as I am
aware, and certainly not in any examples of them which
I have seen. The postorbital processes of the frontals,
and the temporal ridges nearly to their junction, are very
rugose. The anterior orbital margin is but very slightly
elevated above the general surface of the lacrymal bone.
The posterior part of the nasal bones is more obtuse than
in NV. horridus. A peculiar character of these nasals in
this specimen lies in the peninsula of these bones extend-
ing into the frontals near the maxillary suture, so that a
traverse on either side just posterior to the divergence of
the nasal and maxillary bones would pass in succession
toward the sagittal plane through maxillary, frontal,
nasal, frontal, and nasal bones. This may be an individ-
ual variation, but if so, it is an interesting one.

Measurements.
(CatNo 10075, ¥. P.M)

Axial length, ant. of canine to post. of glenoid process.... 151.5
Wiekeneca soul at pOSte Or NI. coc ee icc ch oa ke ees 87
Superior molar-premolar series, length................. 74
Spanier premolar series, lenethy i 6.56... 04. eke. ol oss. 49
Siamunaelenohh oF rants. . 0s. Sea. Jee ek BE) La 8 3-5
imrerior: arolar-premolar series) length’... 00.0. 82
iierion premolarseries” length... 5%. > - «caress cm os OL

SYNOPTIC TABLE.
Neohyenodon.

Glenoid articular surface far below basicranial plane; superior
premolars four; dolichocephalic; size very large; antero-
external buttress on paraconid of M,.

Length of superior molar-premolar series 127-137 mm.
N. horridus

286 Thorpe—New Genus of Oligocene Hyenodontide.

Hycenodon.

Glenoid articular surface on level with basicranial plane. No
external buttress on M,.
.A. Superior premolars three. Face short.
1. Length of superior molar-premolar series 62 mm.
H. paucidens
B. Superior premolars four. Dolichocephalic. 3
1. Palatines in contact throughout; pterygoid pro-
cesses suturally joined. Length of superior
molar-premolar series 74 mm..H. leptocephalus
2. Posterior nares opening between palatines.
a. Postorbital constriction in advance of coro-
nal suture.
(1) Length of superior molar-premolar
| series 72 mm. ...... H. crucians
b. Postorbital constriction at or behind coronal
suture.
(1) Size large. Length of superior
molar-premolar series 106 mm.
HT. cruentus
(2) Size moderate. Complete skull un-
known. Length of superior
molar-premolar series 83 mm.
H. montanus
(3) Size small. Length of superior
molar-premolar series 58 mm.
H. mustelinus

REFERENCES.

Andrews, C. W. 1906. A descriptive catalogue of the Tertiary Vertebrata
of the Fayim, Egypt. London.

— 1907. The recently discovered Tertiary Vertebrata of Egypt. Ann.
Rept. Smithsonian Inst., for 1906, 295-307.

Douglass, HE. 1901. Fossil Mammalia of the White River beds of Mon-
tana. Trans. Amer. Philos. Soce., n. s., 20, 237-279.

Filhol, H. 1876. Recherches sur les phosphorites du Quercy. Etude des
fossiles qu’on y rencontre et spécialement des mammiféres. Ann. Sci.
Géol., ser. 4, 7. Art. 7.

Laizer and Parieu. 1839. Note sur la machoire d’un carnassier fossile,
nommé Hyénodon leptorhynchus. Ann. Sci. Nat., (2), 11, 27-31, pl. 2.

Leidy, J. 1853. [Remarks on a collection of fossil Mammalia from
Nebraska.| Proc. Acad. Nat. Sci., Philadelphia, 6, 392-394.

— 1869. The extinct mammalian fauna of Dakota and Nebraska. Jour.
Acad. Nat. Sci., Philadelphia, (2), 7.

Lydekker, R. 1885. Catalogue of the fossil Mammalia in the British
Museum (Natural History), pt. I. London.

Marsh, O. C. 1884. Dinocerata. U.S. Geol. Survey, Mon. 10.

Matthew, W. D. 1901. Additional observations on the Creodonta. Bull.
Amer. Mus. Nat. Hist., vol. 14, 1-38.

Thorpe—New Genus of Oligocene Hyenodontide. 287

— 1903. The fauna of the Titanotherium beds at Pipestone Springs,
Montana. Ibid., vol. 19, 197-226.

— and Granger, W. 1915. A revision of the Lower Eocene Wasatch and
Wind River faunas. Ibid., vol. 34, 1-103, 311-328, 329-361, 429-483.
Osborn, H. F. 1909. New carnivorous mammals from the Faytim Oligo-
cene, Egypt. Ibid., vol. 26, 415-424. :

— and Wortman, J. L. 1894. Fossil mammals of the Lower Miocene
White River beds. Collection of 1892. Ibid., vol. 6, 199-228.

Scott, W. B. 1887. On some new and little known creodonts. Jour. Acad.
Nat. Sci., Philadelphia, (2), 9, 155-185.

— 1892. A revision of the North American Creodonta, with notes on some
genera which have been referred to that group. Proe. Acad. Nat. Sei.,
Philadelphia, 44, 291-323.

— 1894. The osteology of Hyenodon. Jour. Acad. Nat. Sci., Philadelphia,
(2), 9, 499-535.

— and Osborn, H. F. 1887. Preliminary account of the fossil mammals
from the White River formation, contained in the Museum of Com-
parative Zoology. Bull. Mus. Comp. Zoology, vol. 13, 151-i71.

Wortman, J. L. 1901-1902. Studies of Eocene Mammalia in the Marsh
Collection, Peabody Museum, pt. I. This Journal (4), 11-14.

288 E. L. Troxell—Status of Homogalax

Art. XXVIIL—The Status of Homogalax, with Two New
Species; by Enpwarp L. TrRoxELu.

[Contributions from the Othniel Charles Marsh Publication Fund, Peabody ~
Museum, Yale University, New Haven, Conn. |

Homogalax Hay constitutes a genus of odd-toed animals
whose relationship to the early horses is both near and
confusing. The older genus Systemodon Cope has been
broken up, part going to form this newer genus, and part,
including the genoholotype, being put under Hyracother-
aum Owen, one of the earliest of the Equide.

In 1875,' Cope made the species Orohippus tapirinus,
which in 1877? he referred to Hyracothervwm; in 1881? he
made the distinct genus Systemodon for it, because he
concluded from a study of additional material that the
species showed the lack of a diastema, a new feature.

It is necessary to disregard Cope’s later descriptions
and figured specimens in attempting to determine what
Systemodon and its type, S. taprrinus, really are, and the
fragmentary character and hmited amount of the holo-
type make accurate comparison impossible. Strangely
enough, neither the original type nor any of the later
referred specimens exhibit the feature on which the genus
was supposed to be based.

In 18964 Wortman placed the species S. tapirimus again
under Hyracotherium, which he considered a synonym of
Eohippus Marsh. Hay therefore in 1899° proposed the
name Homogalaxz, type ‘“‘S.’’ primevus Wortman, and
intended to have it include the species S: semihians Cope.
There is ample justification for the erection of this new
genus because of the imperfection of the holotype of
S. tapirinus and because of fundamental differences,
especially in the form of M, as judged from other mate-
rial; but on account of the presence of diastemata, the
small heel on M,, and for other reasons, ‘‘S.’’ semihans
van not be included in Homogalax Hay. It appears, more-
over, that this species belongs to no known genus.

*H. D. Cope, Systematic catalogue of Vertebrata of the Hocene of New
Mexico. U.S. Geog. Surv. W. 100th, Merid., p. 20.

7 Da Cope, |W: S. Geog. Surv. W. 100th Merid., 4, Paleontology, p. _ 263,
pl. 66, figs. 12-16.

oH De Cope, -amer, iNateetomi0ls,

4J. L. Wortman, Bull. Amer. Mus. Nat. Hist., vol. 8, 94-95, 1896.

°O. P. Hay, Science, new ser., 9, 593, 1899.

with Two New Species. 289
Cope’s referred specimen, ‘‘S. tapirinus’’® may belong
‘to Homogalax cf. H. primevus. It is, in my opinion, not
of Systemodon. Likewise the specimen referred to H.
tapirinus by Wortman is clearly not of that species.

There is no doubt a close relationship between all these
genera, including Hyracotherium and the early horses.
Homogalax is separated from these allies by the total
absence of a diastema, the lack of distinct tubercles on the
eross lophs of the molars, and the general peculiar shape
of the molars themselves.

Homogalax primevus (Wortman) is the genoholotype,
from the Wasatch beds. The two new species made in the
pages following are from the Bridger and Uinta, thus the
three forms represent the lower, middle and upper Kocene
respectively.

Homogalax bridgeresis, sp. nov.

Holotype, Cat. No. 12563, Y. P. M. Eocene (Bridger), Twin Buttes,
Wyoming.

(Fies. 1-2.)

Fic. 1—Homogalax bridgerensis, sp. nov. Holotype. Right maxillary
with molars, premolars, and canine. Note especially the absence of a dias-
tema, the oblong M*, subquadrate M*.?, and the narrower inner side of P*.
Nat. size.

This interesting type came to the Peabody Museum in
separate shipments in 1874. It bears two distinct labels
.and was thought to be two individuals. Now it is found
that the mandibular rami, accession No. 610, fit the maxil-
laries, No. 655, and furthermore, a small fragment of one
of the latter was actually associated with the rami.

This new species is nearest H. primevus (Wortman),
but differs from it in the narrower first upper molar, in

°E. D. Cope, Rept. U. S. Geol. Surv. Terr., pp. 618-624, pl. 56, fig. 1, 1884-

290 E. L. Troxell—Status of Homogalax

the presence of a distinct metaloph on P? and the triangu-
lar form of this tooth, in the inner cusps of P?, in the small
upper canine, and in the presence of strong cingula on the
outer side of the upper molars.

‘ tee a: RSME x
E : . Wey )
72563 ipdez S =<GEesG
UK. ~

Fie. 2.—Homogalax bridgerensis, sp. nov. Holotype. Part of the lower
jaws. Note the alveoli of the canines and incisors, and the close crowding
of the teeth. Nat. size.

Both the upper and lower series of teeth, except the
upper incisors which we do not know, are continuous
throughout. We judge the lower canines and incisors from
their alveoli. In the lower series the first and second pre-
molars each consist of the simple protocone with ridges
running to it. On P, there is a double central cone, the
true protoconid with the deuteroconid isolated from it.
The fourth premolar still has the single ridge on its broad
posterior heel, but the two anterior cones have separated
to form a transverse crest as in the molars.

The first and second molars have the two ordinary cross
crests with the exterior extensions forward as in tapirs
and rhinoceroses generally. On M, there is a strong
antero-exterior cingulum. ‘The teeth*gradually increase
in size and in the separation of the cross ridges back to
the third molar.

The prominent heel of the third molar amounts almost
to a cross ridge in the arrangement of its two cusps. This
double cusp is distinctive of Homogalax, as contrasted.
with the single one of the original Systemodon.

So little remains of the skull of this rare specimen that
it offers few criteria for its interpretation. The orbit lies
over the molars, extending to a point above the first.
Unlike that of Helaletes, the antorbital foramen is sit-
uated forward, over the third premolar, and there is other
evidence that no facial vacuity existed as in that genus.

with Two New Species. 291

Homogalax wintensis, sp. Nov.
Holotype, Cat. No. 12561, Y. P. M. THocene (Uinta), mouth of the White
River, Utah.
(Fies. 3-4.)
72567 TYPE

Fie. 3.—Homogalax uintensis, sp. nov. Holotype. Upper molars and
fourth premolar of an Upper Eocene tapiroid. Nat. size.

SMW”

\

\\

Le

72967 YATE = Ve TENGE

Fie. 4—Homogalax uintensis, sp. nov. Holotype. Lower dentition and
part of right ramus. Nat. size.

Another specimen in the Peabody Museum demands
consideration, for it not only represents a new species of
Homogalax Hay, but occurs in a totally different geologi-
eal horizon from that of H. bridgerensis described above.

H. wintensis, as compared to H. bridgerensis, is much
larger. The third upper molar is not so perfect a paral-
lelogram in form, although it too is greatly elongated
along the diagonal through the paracone and hypocone.
The antero-posterior diameter is large compared to that
of H. bridgerensis. The metaloph forms a continuous
straight ridge with that part of the ectoloph leading to the
paracone (antero-exterior). This cone is prominent,
especially on M?, while the double parastyle is very small.
On M? this style is much larger. Hach upper molar has a
minute crista and all are notable for their short posterior
side. |

In the lower dentition, one sees that the heel of the last
molar, although broad, is not definitely divided into two
cusps. The third premolar shows a distinct separation
of the metaconid (or deuteroconid) and sharp grooves
separating the elements of the tooth.

299 E. L. Troxell—Status of Homogalax.

Summary.

Systemodon Cope, an indefinite genus concerning which
there is some confusion, is considered for the present, fol-
lowing Wortman, as a synonym of Hyracotherwuwm Owen,
but it is evident from the present study that both are dis-
tinct from HKohippus Marsh.

Homogalax primevus (Wortman) Hay is a species of
tapiroids about the size of Hyrachyus agrarius Leidy;
but it may be easily separated from this and from the
early horses by the absence of diastemata, by the exist-
ence of a third lobe on the last lower molar, and by other
generalized features. There is no evidence of a facial
depression like that of Helaletes.

Two new species are here proposed: Homogalaz
bridgerensis and H. wintensis; these, together with H.
primevus, the genoholotype, represent three localities
distinct geologically and geographically. H. bridgerensis
may be distinguished by the triangular P? with distinct
metaloph, the small canine, the absence of diastema and
the geological age. H. wntensis is notable for its larger
rounded teeth, double parastyle and late Hocene age.

Measurements.

H. bridgerensis H.wintensis
12563, Y.P.M. 12561, Y.P.M.

mm. mm.
Upper dentition
Length of molar-premolar series .... 68
Henethyvotgmolarrsemicse,:ssssre i yer 38 . 44.5
Lower dentition
Wene Ga 100 SP. a pO in C ae eaee ne eee 08.3 68.3

Benoth ot molag series aa. a oe 41 47.3

W. P. Headden—Tantalate from So. Dakota. 293

Art. XXVIII.—4A Tantalate and Some Columbites from
Custer County, South Dakota; by Wittt1am P.
HEADDEN.

The minerals presented in these notes are from Custer
Co., South Dakota, with the exception of one columbite
which was found near Harney City, Pennington Co. They
were all collected by Mr. George B. Grant with the excep-
tion of the tapiolite which was collected by the writer
many years ago.

The columbites have the habit and physical properties
of this mineral as it occurs in other districts of the Black
Hills, but there were very pronounced differences among
them in specific gravity, color, and habit. They all
showed the tabular form with striations. The fracture
was uneven, inclining to granular; the color was brown
or grayish black, with shghtly shining sub-metallic luster
but wholly free from iridescence.

The tantalate differed from these columbites in all of its
physical properties. It is a pure black in color, with
shining metallic to adamantine luster and a strong irides-
cence; it is brittle and harder than the columbites and has
a higher specific gravity. .

The fragments received were small, irregular in form,
sharp angled, and showed no satisfactory cleavage sur-
faces. It is doubtful whether they actually show any
cleavage surfaces at all. The fragments were freshly
broken and clean, but there was a white, triclinic feldspar,
apparently microcline, adhering to some of them. These
fragments appear to have been originally parts of masses
that had been cracked and the surfaces thus produced
coated with a thin film of silica. Some of them contain
small pieces of quartz, but only occasionally a flake of mica.
The tantalate is all from one locality, the Gld Mike Mica
mine in the Minnehaha Gulch, 6% miles north of Custer
City, Custer Co., South Dakota. The columbite is from
three localities, two of them in Custer Co.; these are the
Old Mike Mica mine and Tin Mountain on Warren’s Gulch;
the remaining one is from Harney City, in Pennington Co.
This last locality is not far from the Etta mine in the
Harney Peak district which formerly furnished many fine
specimens of columbite, some of which have been
described in earlier numbers of this Journal. I never

Am. Jour. Scit.—FirtH Serises, Vou. III, No. 16.—ApRIL, 1922.
21

294. W. P. Headden—Tantalate from So. Dakota.

found nor learned that any sample of a tantalate was
found at this locality, but the columbites occurring there
were all of high specific gravity as were ali of the
columbites at other localities in the district. The sample
from Harney City, as will be seen from a later statement,
presents a striking case of a columbite with a high specific
gravity but not so extreme as the sample from Tin Moun-
tain in Custer Co.

The columbites from the Old Mike Mica mine differ
from all the samples of this mineral that I have seen from
the Black Hills, in that they are of low specific gravity.
One sample has the lowest specific gravity that I have
found for this mineral from any locality, viz. 5.201 at
4° C, :

The tantalate was found at the Old Mike Mica mine
and is, so far as my knowledge goes, the second instance
of the finding of this mineral in place in the Black Hills.
In 1889, or 1890, I found a single group of crystals at a
prospect hole, also in Minnehaha Gulch, 314 or more miles
north of Custer City. The form of these crystals was
subsequently determined by Prof. Penfield and found to
be that of tapiolite. These localities are not more than
three miles apart and in the same gulch. I found a tan-
talate of iron very sparingly present in stream-tin from
the Grizzly Bear Gulch and also in that from Mitchell’s
bar, but the mineral found in these placer products may
really have been from the same source, as the placers
were only different parts of the same gulch. I supposed
that the source of this tantalate was the Tin Queen lode,
which lay above the placer working in the Grizzly Bear
Gulch. The Minnehaha, Sunday, and Grizzly Bear
gulches take their rise in the same general section and
the tantalate previously found may have had its source
farther up the gulch than I supposed, and may not have
had the same origin as the tin-stone, though associated
with it in the bed of the gulch.

The tantalate from the Old Mike Mica mine varies
greatly in its specific gravity. The pieces sent were quite
small; the largest of them weighed between 40 and 90
erams, but these were not clean. The clean pieces usually
weighed less than 5 grams and seldom reached 10 grams.
Determinations of the specific gravity, made with the
Jolly balance, showed a variation from 6.9 to 8.0, and

W. P. Headden—Tantalate from So. Dakota. 295

there was no distinguishable differences between the frag-
ments. After observing this fact, I took individual pieces,
broke them up and picked out enough of the clean mineral
to make the determination of the specific gravity, using a
pyenometer; the results varied as greatly as when I used
the Jolly balance. There were a few pieces, weighing
from 15 to 25 grams, that appeared to be clean and uni-
form. J determined the specific gravity of these by sus-
pending them in water with the same results as before.
In a few instances I found a line through the sample that
suggested the possibility of a difference in the parts of the
piece. I broke these samples on this line and determined
the specific gravity of the separate parts. The greatest
difference that | found in this way was in the case of a
piece that weighed 15.4 grams. This piece showed a line
dividing it into two parts which differed shghtly in tex-
ture, color, luster and iridescence. On placing a sharp
chisel on this line and giving it a quick tap, the piece split
in two almost equal parts which were free from mica, ete.
One piece was not so bright and clear a black as the other,
hada slightly granular fracture and a weaker iridescence.
The other piece had a pure, bright black color, conchoidal
fracture, metallic adamantine luster with a strong irides-
cence. The specific gravity of the former piece was 7.019
and of the latter 7.878 at 4° C. While this is the extreme
difference found in the specific gravity of two parts of a
comparatively small piece of mineral, it is not the only
instance of it. At first, I attributed this variation in the
ease of small individual pieces to a mixture of samples,
that is, | supposed that several small masses had been
broken out and put together as a general sample of the
mineral. This may have been done but it does not follow
from the variation in the specific gravity of the individual
pieces. The columbites from the Etta mine showed
clearly that this mineral, from the same locality, varies
greatly in this respect. Even the individual crystals of a
group may vary both in specific gravity and composition.
The crystals at the Etta were large, weighing up to 10 or
more pounds, and were distinct individuals which may
have made a difference. In this case, the pieces are small
and there is no distinction of individual or even groups
of crystals. I have been unable to learn whether crystals
of this tantalate occur at the Old Mike Mica mine or not.

296 W. P. Headden—Tantalate from So. Dakota.

Under these conditions one is compelled to use the identi-
cal piece for analysis that was used in determining the
specific gravity. The description of the samples analyzed
has been given in sufficient detail:

Analysis of a Tantalate from the Old Mike Mica Mine.

Sp. er.

at 4° C.

1a 6.954
2.6 7.019
3 7.180
4, 7.468
5. 7.794
6 7.878
a 7.975

Sx iloO

a Recaleulated after deducting 1.52% tin-stone.
b Recaleulated after deducting 1.56% tin-stone.

al@r

69.55
76.08
17.24.
78.28

79.50-

81.40
83.57

Tapiolite from Prospect in Minnehaha Gulch.

77.23

ChiOp Ze:
8.63 1.50
5.49 2.84
6.97 0.81
DOs be lea
4.32 0:92
BLO Os LOLS
1.97 trace

5.18

1.38

Sn0O,4

5.29
0.22
Ueki)
0.28
0.32
0.12
0.10

0.32

FeO

10.84
14.16
13.60
13.35
13.42
12.55
13.28

14.84

¢ Recaleulated after deducting 0.88% tin-stone.
d SnO, includes traces of WO.

MnO

4.19
1.2]
1.02
1.22
1.56
1.73
UE)

0.42

Total

100.00
100.00
100.82
100.62
100.04
100.07
100.10

100.00

At. Ratio
Fe to Ta

Hee eee

4

22.06
21.96
21.95
22.05
72.02
72.03
21.99

21.95

Analyses 2 and 6 are parts of the piece weighing 15.4
grams previously mentioned and show that they differ
greatly in composition as well as in specific gravity.
Analyses 3 and 4 are parts of another piece, that weighed
25 grams and they too show that they vary in the same

Manner.

Analysis 8 is a portion broken off of the crystals of
-tapiolite found not very far from this Old Mike Mica
The specific gravity of this group of crys-

mine locality.

tals and of the pieces broken off of the base of them was
determined several times at 18° and the deviation found

was from 7:15 tow#2h.

The specific gravity of the whole

group was found to be 7.20. The average of all deter-
minations made reduced to 4° C. is 7.190, which I have
taken as that of the material analyzed. Analyses 3 and 8
have very nearly the same specific gravity and are almost.
the same in composition. These samples are not from the
same locality, but the localities are not more than three
miles from one another and are in the same gulch. The
agreement in composition is what we should expect if the
relation between the specific gravity and composition in
these minerals is as intimate as we suppose it to be;
it does not prove the mineralogical identity of the two
samples though it may reasonably be taken to suggest

this.

but

W. P. Headden—Tantalate from So. Dakota. 297

Methods of Analysis——There is no need to discuss these
as they were well known, conventional ones, except, per-
haps, the determination of titanic acid which was effected
by fusion with potassic bisulphate powdering the melt and
dissolving it in a mixture of sulphuric acid and hydrogen
peroxide. I used 5 parts of sulphuric acid, made by
diluting 3 parts of concentrated acid with 2 parts of
water; and 4 parts of 3% hydrogen peroxide. This
effects a perfect solution of the mineral which can be used
in the colorimeter without filtering after standing to allow
any bubbles of hydrogen peroxide to escape. This solu-
tion will remain clear for some hours, but eventually
tantalic acid begins to separate. I tried to separate the
acids from the salts introduced by this fusion by boiling
as usual and dissolving the separated acids in hydro-
fluoric acid, precipitating the hot solution by ammonia
and dissolving this precipitate in the acid hydrogen-per-
oxide mixture, but the acids did not go into solution.
Both of the acids, tantalic and columbic, are contam-
inated by titanic acid as separated in this analysis and the
titanic acid must be determined and deducted. I may
remark that the platinum that goes into solution in the
bisulphate fusions LOUEBEE I) in these analyses is an
annoyance.

The determination of any cassiterite or quartz that may
be present is most easily made by fusion with bisulphate
and dissolving the melt in acid hydrogen peroxide. This
is much more satisfactory than taking the residue from
the solution of the mixed acids in hydrofluoric acid, which
invariably contains a little platinum.

The analytical procedure was the same in the case of
the columbites.

Columbites.

The columbites presented in these notes have not been
described heretofore, and are from three localities; two
of them are in Custer Co., the Old Mike Mica mine, and
Tin Mountain; and one, Harney City, is in Pennington Co.
The minerals are typical columbites in habit, cleavage,
luster, etc. They are not quite so hard as the tantalates,
and are not brittle like them. They vary in color from
a brownish to a grayish black, in luster from submetallic,
- shining in the ease of the Old Mike samples, to dull sub-

298 W. P. Headden—Tantalate from So. Dakota.

metallic in those from Tin Mountain. These latter con-
sisted of small fragments of plates, and one larger piece
that weighed 17.33 grams. The specific gravity of these
differed; that of the larger piece was 6.725 and that of
the plates was 6.845 at 4° C. The characteristics of these
pieces were those of typical columbites except that the
fracture at the edge of the larger piece was distinctly con-
choidal. The inner side and the surfaces exposed by
cleaving this piece in two parts showed a fine granular
fracture. Ido not know the details of these occurrences.
The specific gravity has a wider range than I have met
with in other samples, namely from 5.201 to 6.845. The
former is the lowest and the latter the highest that I have
found for a columbite from any locality. The lowest that
I have found for a tantalate is 6.954. The sample with
this specific gravity differed as widely in its physical
properties from the columbites as the tantalate having
a specific gravity of 7.975 and was a fragment picked out
of the same lot. Its composition is that of a tantalate
having the formula 5FeT'a,O,.FeCb,O, compared with
8FeT'a,O,.7FeCb,O, for the columbite having the specific
gravity of 6.845.

In the following table of analyses of columbites num-
bers 1, 2, and 3 are from the Old Mike locality, number 4
is from Harney City and numbers 5 and 6 are from Tin
Mountain. Number 4 1s from Pennington Co., the others
are from Custer Co.

Analyses of Columbites.
Sp. gr. Ratio
at 4° C. ‘Cb,0, Ta,0;- TiO, SnO,a~ FeO, MnO Total) He:0p
5.201" 68.00 = 9338 ~ 053" 0:88 5.45 14.79 99.53)" 1 eg
5.421 67.20 10.10. 0.92 0.96 5.96..-15.08). 100.22 Saisie
5.496 63.90 13.74 0.80 0.54 5.92 14.95 99.85, sel
6.444 34.60 46.02 1:52 0.388 13:32 4.31 100.15> 9
2 6.725 (27.22) 53.47 > 1.30 9044 191 2" 5:66 = 100.005 eerie
6.845 = 287815 253.6763 7 Uto Se) 2 00s 55220) 0 O26 ea

D> OUR 62 bo

a The SnO, contains WO; in quantity up to 50% of the SnO, present.
b Lime equal to 0.48% deducted.

These analyses show, if those of the columbites are
placed first, a gradual increase in the specific gravity from
5.2 to 7.94 with a regular increase in the amount of tan-
talic acid. This is still true even when we pass the divid-

W. P. Headden—Tantalate from So. Dakota. 299

ing line between distinct mineralogical individuals, for
the tapiolite with a specific gravity of 7.19 is tretragonal
and well erystallized.

Another thing that is shown is that the tantalate is
an iron salt whereas the columbites with a tendency to
a high manganese content are mixed. In the tantalate
the ratio of Fe to Mn increases from 95.2 in the sample
having a specific gravity of 6.954 to 11:1 in those having a
specific gravity around 7.9, and to 35:1 in the tapiolite.
This ratio in the columbites is reversed, being a little less
than 3:1 in the Harney City columbite and 1:3 in those
from the Old Mike locality. The columbite from Tin
Mountain is a different mixture in which the Fe:Mn is
as 4:2, but at the same time the tantalum present has
risen to an amount equal in atomic equivalence to the
eolumbium. The columbites from the Harney Peak dis-
trict that I have studied are all intermediate in respect
to these ratios, ranging for Cb:Ta from 1:1 to 6:1 while
the Fe:Mn ratio is from 8:1 to1:1. There is one instance
in which it is reversed and becomes 1:2 in a sample that
represents a single mass of columbite in the Sarah mine.
These relations hold for such of the Black Hills colum-
bites as I have examined. This mineral from different
districts has a different, but for the district in general, a
characteristic composition. This does not apply in com-
paring the mineral from the Black Hills with samples
from other localities, even in cases in which the specific
eravities are approximately the same. In the ease of a
eolumbite from Morrison, Colo., with a specific gravity
of 5.383, we have practically no tantalic acid and an
Fe:Mn ratio of 8:7 whereas a columbite from the Old
Mike with a specific gravity of 5.421 and very little tan-
talic acid has an Fe:Mn ratio of approximately 1:8.
It would be inadmissible to compare minerals of different
specific gravities.

Colorado Agricultural College,

Fort Collins, Colo.

300 Scientific Intelligence. |

SCIENTIFIC INTELLIGENCE

I. CHEMISTRY AND PHysIcs.

1. The Separation of the Isotopes of Mercury.—The idea that
isotopes exist not only in radioactive elements, but that their
occurrence is general among the elements appears to have been
well established by a number of recent investigations. J. N.
BroOnsteD and G. von Hrvesy, of Copenhagen, have previously
made preliminary announcements of their partial separation of
metallic mercury into its isotopes by means of fractional distilla-
tion, and their observations have been confirmed by W. D.
Harkins in this country. They have now published a full
account of their work. Their method of evaporating the mer-
eury, which they eall ‘‘ideal distillation,’’ is interesting, since it
is carried out in the exhausted space between double flasks at a
temperature of about 40°, while the evaporated mercury is con-
densed in the solid condition by means of liquid air in the inner
flask. This operation was repeated with the distillates and the
residues, about 17 or 18 times in the last case, until the frac-
tions became very small. The resulting fractions were submitted
to specific gravity determinations with remarkably satisfactory
results. The extreme specific gravities found were 1.00023 and
0.99974, and in connection with the description of the work are
very convincing as to the existence of isotopes of mercury.

The authors describe also some experiments upon the diffusion
of mercury vapor-through small openings, and obtained an indi-
cation of a separation in this way.—Zeitschr. Physikal. Chem.,
9939) H. L. W.

2. Lhe Color of Ferric Ammomum Alum.—lt is well known
to chemists, especially those who have supervised its preparation
by students, that this salt usually has a violet color, but is some-
times colorless or nearly so. Ostwald advanced the theory that
the colorless product is the pure form, while the color of the other
is due to the presence of manganese. JANE BONNELL and EpGAR
Puitip PERMAN have now carefully investigated this matter.
They were unable to detect manganese in a sample of the colored
variety, and they carried out the operation of separating any
manganese in this salt by precipitating the iron as basic acetate,
converting the precipitated iron into the alum, and even after
repeating the purification a second time in the same way, they
obtained the violet product in the presence of a considerable
excess of sulphurie acid.

It was then shown experimentally that the colorless crystals
were not due to the presence of ferrous sulphate which had been
found in a certain sample of the colorless crystals, but it was
shown conclusively that the lack of color was due to the presence

Chemistry and Physics. 301

of ferric hydroxide produced by hydrolysis upon boiling solu-
tions containing but little free sulphuric acid and producing a
brown color practically complementary to the violet one and thus
hiding it—Jour. Chem. Soc., 119, 1944. H. L. W.

3. The Persulphides of Hydrogen—JamEs H. WaAutTon and
LLEWELLYN B. Parsons, of the University of Wisconsin, have
made an interesting study of hydrogen disulphide, H,S,, and the
tri-sulphide, H,S., which are constituents of the oily liquid pro-
duced by the action of hydrochloric acid upon alkali or alkali-
earth polysulphides. The existence of these two compounds had
been established, so that the results of the present extensive inves-
tigation have been improvements in the method for the prepara-
tion of the pure compounds and a more extended knowledge of
their properties and reactions. Only a few of the results can be
referred to here.

It was found that while hydrochloric acid gives the well-
known yellow oily liquid with sodium polysulphide solution, the
other common acids, acetic, phosphoric and sulphuric acids, do
not give any of the oil, but a complete decomposition into hydro-
gen sulphide and sulphur. It was found that the crude oil could
be dried most satisfactorily by the use of phosphorus pentoxide
which has no action upon it. The crude oil upon analysis gave
results corresponding closely to the formula H.S8., but it is
believed to be really a solution of sulphur in the sulphides of
hydrogen. The nearly pure disulphide and trisulphide of hydro-
gen were obtained by a single distillation under low pressure by
the use of two successive receivers, the first cooled by running
water, the second by ice and salt. Quartz vessels were used for
the distillation as well as for containers of the persulphides, since
it was found that quartz decomposed them far less rapidly than
glass. A very satisfactory method was devised for the analysis
of the persulphides. It consisted in dissolving the sample in ecar-
bon disulphide, adding acetone which caused a catalytic decom-
position into hydrogen sulphide and sulphur, evaporating to dry-
ness at 90° and weighing the sulphur.—Jour. Amer. Chem. Soc.,
43, 2539. H. L. W.

4. A Separation of Germanum from Arsenic.—JoHn H.
Muuuer, of the University of Pennsylvania, who has recently
determined the atomic weight of germanium and in connection
with that work has called attention to the difficulty of removing
the last traces of arsenic from germanium, has now solved this
problem very satisfactorily.

The distillation process from aqueous hydrochloric acid in the
presence of chlorine, which has been recommended, did not give
satisfactory results, as it was found that minute quantities of
arsenic passed into the distillate when the conditions were suita-
ble for volatilizing the germanium chloride, although this method

302 Scientific Intelligence.

separates germanium from all the other metals and semi-metals.

It was found, however, that arsenic and germanium can be
quantitatively separated by the action of hydrogen sulphide upon ~
solutions of their oxides in the presence of a large excess of hydro-
fluorie acid, for under these conditions the arsenic is precipitated
while the germanium is not affected. A series of test analyses
made with known quantities of the two elements gave most excel-
lent results by this method.—Jour. Amer. Chem. Soc., 43, 2549.

H. L. W.

5. Asymmetry of the Gaseous Molecule——According to the
theory developed by the late Lord Rayleigh, the color of the blue
sky may be explained by light which is scattered by the gaseous
molecules of the air, and, according to the simple theory, if the
molecules were spherical this scattered light should be completely
polarized. It was suggested by R. J. Strutt, the present Lord
Rayleigh, that any departure from complete polarization would
indicate that the molecule had certain preferential directions of
vibration. Experiments designed to test the degree of polariza-
tion were devised by Strutt in 1919 and showed that none of the
seventeen gases and vapors examined exhibited complete polariza-
tion of the scattered heht. This was what might have been
anticipated, for the modern view of the structure of atoms would
not lead us to expect that the atom would behave as if possessing
spherical symmetry, even apart from the grouping Of the atoms in
the molecule.

The gas under examination was placed in a metal tube with
erossed arms which were properly blackened on the inside. A
strong beam of light was sent through one of the arms and the
scattered heht was viewed through a double image prism in the
end of the transverse arm. One of these images contained the
seattered light which had been polarized and whose vibrations
were perpendicular to the traversing beam, while the other con-
tained light belonging to the unpolarized portion and on emer-
gence possessed vibrations parallel to the original beam. To com-
pare the intensities of these images, which were widely different,
a set of absorbing films of varying opacity was prepared. These
were interposed in the path of the stronger beam and the two
images photographed side by side. When the result showed two
images of sensibly the same intensity it was possible from the
calibrated scale of the absorbing diaphragms used to estimate the
relative intensities of the polarized and the unpolarized light.

These investigations of Strutt have now been repeated and
considerably extended, with various improvements in the form of
the apparatus, by R. Gans. The degree of asymmetry may be
represented to the eye by constructing an ellipsoid of revolution,
but with the understanding, of course, that this 1s in no sense to
be identified with the actual contour of the molecule. The author

Chemistry and Physics. 303

has caleulated and in a few cases drawn, not only the form of but
the absolute sizes of these ellipsoids. In confirmation ‘of the
earlier work by Strutt the ellipsoid for helium is an exceptionally
elongated, or spindle-shaped ellipsoid, indicating that this mole-
cule appr oximates a linear resonator.

Both the Kerr effect and the Faraday rotation of the plane of
polarization are theoretically dependent upon the asymmetry of
vibration in the molecule and the author has calculated the
amount of each of these effects which would be predicted from his
measurements on the polarization of scattered light. The
observed Kerr effect in different gases is in satisfactory accord
with the predicted amount but the agreement of the Faraday
effect is only fair.—Annal. der Phys. 65, One OZ. F. E. B.

6. L’Atome; by Dr. AcHALME. Pp. 24 AA. Paris, 1921 (Payot
et Cie.).—The ‘author, who is not otherwise designated than as
the Director of the labor atory of l’Ecole des Hautes Etudes, has
attempted with much ingenuity to explain the structure and form
~ of atoms. Although his book exhibits. considerable acquaintance
with the writings of many physicists, his speculations are, unfor-
tunately, based too largely upon the phenomena of chemical
statics to be convincing. The trend of current thought is all in
the direction of the Rutherford-Bohr atom, which hypothesis,
together with the spectroscopic evidence and its intimate con-
nection with atomic structure, are completely ignored.

It is to be apprehended that the author has run into a stereo-
chemie cul de sac. F. E. B.

7. Calculus and Graphs; by Ll. M. Passano. Pp. vu, 167.
New York, 1921 (The Macmillan Company).—The purpose of
the author has been to write a brief and elementary course on the
ealeulus which would make this branch of analysis available to
students of physics, of chemistry and of other sciences where some
knowledge of mathematics is required. As is indicated in the
title, the presentation has been made chiefly from the standpoint
of coordinate geometry, but it does not presume any knowledge
of formal analytical geometry or the properties of functions.
Although as a text, it is one which would not be selected by those
who desire more than an introduction to the subject, its mastery
would connote the full equivalent of. the ordinary scientific
student’s knowledge of analysis. Ps Hee Bs

8. The Manufacture of Optical Glass and of Optical Systems ;
by Lieut. Col. F. E. Wricht. Pp. 309. Washington, 1921
(Government Printing Office).—At the outbreak of the war the
production of optical glass in this country was nil, the small
amount required by instrument makers being procured from
abroad and chiefly from Germany. With the declaration of war
by the United States in 1917 the embarrassment of the govern-
ment became so acute that it was necessary to requisition both the

304 Scientific Intelligence.

scientific and the technical resources of the country to supply the
needs of the naval and military forces. This volume has been
prepared by the Ordnance Division of the War Department and
is designed to present a record of the investigations made
and the results obtained at that time. The introductory chap-
ter outlines the personnel of the work undertaken, the fac-
tories engaged and the production achieved. Chapter 2 is
devoted to the composition of optical glasses and their various
physical characteristics with extensive tables and diagrams show-
ing the chemical analysis and optical constants. The third
chapter describes the technical processes involved in mixing,
melting, and annealing the glass, and is illustrated with numer-
ous photographs of these operations taken in the factory. In
chapter 4 the methods used for the inspection and detection of
defects are explained. Chapter 5 describes the processes by
which the lenses and the prisms are shaped, ground and polished.
The sixth chapter explains the ways in which the completed opti- -
cal train is tested for satisfactory performance. The last chapter
contains a review of the optical instrument situation during the
war and the success which was attained in the production of these
instruments.

The book will be a highly informing one for any person who
is at all interested in optics and contains much data probably not
accessible elsewhere. F, E. B.

Jk. SG Ronegye

1. Notes on Arctic Ordovician and Silurian Cephalopods; by
Aug. F. Forrste. Bull. Denison Univ., vol. 19, pp. 247-306, pls.
27-35, 1921. Revolution vs. Evolution: The Paleontologist Renders
his Verdict ; by Kirtuny F. Matuer. Ibid., pp. 307-323.—The first
paper is a detailed study of twenty-eight nautilids, of which sev-
enteen occur in Arctic America, Bear Island or Spitzbergen.
There are ten new forms and two new genera, EHllesmeroceras and
Leurorthoceras. This work is a praiseworthy beginning toward
a revision of all American Ordovician and Silurian cephalopods,
which the author has in contemplation. The second paper is
philosophical in nature, dealing with the course animal life has
taken in its evolution throughout the geological ages. The
author’s conclusions are in part: 7

‘‘The erisis recorded in the rocks of latest Paleozoic Age, was
forced upon the land animals by changes in climate and environ-
ment; it was successfully passed by creatures who specialize in
the adaptation of their bodies to cold and drought, and who
escaped from the crowded confines of the sea to the almost unin-
habited silences of the land. The ‘revolution’ involved in the
dethronement of the reptiles and the exaltation of the mammals

Geology. 305

at the close of the Mesozoic Era was likewise precipitated by
external changes over which the mammals themselves had no con-
_ trol; it bettered the condition of creatures who specialized in the
care of their young and the use of their brains. Each group
which prospered had for some time displayed the very character-
istics which proved efficacious in the time of stress; the ‘revolu-
tion’ afforded the opportunity for the testing and the rewarding
of the products of progressive evolution. ... As in the past, so
in the twentieth century, the impelling forces of progress are
inherent in the environment; the response must be dependent
upon virtues intrinsic in the creatures who are to be thus tested.
Some will undoubtedly be found wanting; for them the penalty
has always been either extinction or stagnation. Others—and in
the past it has generally been a minority—will respond with
habits that will prove to be their salvation; they, and they alone,
will profit by the revolution.’’ Cis:

2. Stratigraphy of the Pennsylvanian Formations of North-
Central Texas; by FREDERICK B. PLUMMER and RAyMoND C.
Moors. Univ. of Texas Bull. No. 2132. Pp. 287, 27 pls., 19 text
fies., 1921 (1922).—It is highly pleasing to note that this careful
and detailed work on the Pennsylvanian strata of north-central
Texas was done by the geologists of the Roxana Petroleum Com-
pany, and donated to science through the Texas Bureau of
Economic Geology and Technology. It is codperation of this
kind that is to the advantage of humanity in practical and intel-
lectual ways. We congratulate all concerned.

The maximum thickness of the Pennsylvanian here is about
6,800 feet, embracing the Bend, Strawn, Canyon, and Cisco
groups of formations. The report contains a large geologic map,
eleven photogravure plates of typical fossils, many other illustra-
tions, and a complete list of the known Texas Pennsylvanian
faunas, totalling about 354 forms. No new species are described.
Of cephalopods there are forty-three species and of these no fewer
than twenty-seven are goniatites and ammonites. Cys!

3. Recent Mollusca of the Gulf of Mexico and Pleistocene and
Pliocene Species from the Gulf States. Part 2, Scaphopoda, Gas-
tropoda, Amphineura, Cephalopoda; by Caruotta J. Maury.
Bull. Amer. Paleontology, No. 38, 142 pp., 1922.—This is a list of ©
the Scaphopoda, Gastropoda, Amphineura, and Cephalopoda of
the regions and geologic times mentioned in the title. The bibli-
ography of each species is given, along with the distribution, and
there are also occasional notes on the forms.

4. Handbuch der Regionalen Geologie, 23. Heft, Aegypten,
by Max BLANCKENHORN; 24. Heft, Die Nordatlantischen Polar-
inseln, by Otto NORDENSKJOLD. 1921.

d. The Topographic and Geological Survey of Pennsylvania;
GrorcE H. AsHuEy, State Geologist—The following bulletins,

306 Scientific Intelligence.

chiefly of economic character have been received; the director
of the Survey is the author unless otherwise stated :

No. 1. Effect of the war on the price of coal in Pennsylvania.
No. 2. Oil and gas in Southeast Pennsylvania.
No. 3. Development and probable life of gas pool at McKees-
port, Penn.
No. 4. Decline of McKeesport oil pool.
No. 7. A high-grade building stone in Greene Co., Penn.
No. 12. Gas wells on Pollock Run, Westmoreland Co., Penn. ;
by J. FRENcH Rosinson.
No. 14. Future Sources of power.
No. 16. Geology of oil and gas in relation to coal.

No. 23. Coal beds in Cambrian Co., Penn.; ‘by J. Ds Sicuur:
No. 24. Coal beds in Greene Co.; by J. D. S1suEr.

No. 25. Coal reserves in Greene Co.; by JoHN F. REESE.

6. United States Bureau of Mines; H. Foster Barn, Director.

—In addition to the usual bulletins, technical papers and miners’
circulars (See, 50, 470, 1920; 1, 288, 1921). The following con-
tribution especially deserves notice:

Summarized Reports of principal Investigations being con-
ducted by the Bureau of Mines for the year beginning July 1,
1921; compiled by J. D. Stcrest, Washington, September, 1921
(a manifolded edition of 185 pages, including list of investi-
gators). The fact that the alphabetized list of these investiga-
tions covers thirteen pages in this edition gives some idea as to
the variety and extent of the work of the Bureau. This embraces
the whole field, from the subject of mine disasters, their causes
and means of prevention, to researches about radium and the
rare gases. The substance of each of the papers here included is
given in clear, condensed form.

Recent Bulletins issued are as follows:

No. 117. Structure in Paleozoic bituminous coals; by RHEIN-
HARDT THIESSEN. 250 pp., 160 pls. (80 cents.)

No. 183. Abstracts of current decisions on mines and mining,
reported from May to August, 1919; by J. W. THoMmpson.
HON Spy.

No. 184. The manufacture of sulphuric acid in the United
States; by A. E. Wetus and D. E. Foae. 216 pp., 18 pls. 36 figs.
(40 cents. )

No. 185. Pennsylvania mining statutes, annotated; by J. W.
THOMPSON. 1921. 1221 pp. ($1.00.)

No. 189. Bibliography of petroleum and allied substances in
1918; by EK. H. Burroveus. 180 pp.

No. 191. Quality of gasoline marketed in the United States;
by H. H. Hint and E. W. Dean. 270 pp., 22 figs.

No. 194. Some principles governing the production of oil
wells; by Cart H. Beau and J O. Lewis. 58 pp., 2 pls., 8 figs.

Geology. 307

No. 195. Underground conditions in oil fields; by A. W.
Amprose. 296 pp., 11 pls., 52 figs.

No. 198. Regulation of explosives in the United States, with
especial reference to the administration of the Explosives Act of
October 6, 1917, by the Bureau of Mines; by C. KE. Munroe.
45 pp.

No. 205. Flotation tests of Idaho ores, by C. T. Wricut, J. G.
PARMELEE, and J. T. Norton. 70 pp., 8 pls., 1 fig.

No. 206. Petroleum laws of all America; by J. W. THOMPSON.
448 pp. (40 cents.)

Prices are given for the last received bulletins.

7. Metamorphism in Meteorites—A recent number of the
Bulletin of the Geological Society of America (volume 32, page
395), contains an article by Dr. G. P. Merrill of the National
Museum on the origin and structure of chondritic meteorites
which is worthy of note here. He thinks to show that the
chondritic meteorites are all of a voleaniec and tufaceous origin
and their varying textural peculiarities due to metamorphism in
which both heat and pressure have had a part. It is pointed out
that the most perfect chondroidal forms are found in those
meteorites the fragmental nature of which is the most apparent,
and that they are less perfect in the crystalline forms. The clear
interstitial glassy material, sometimes isotropic and sometimes
doubly refracting, he considers, as have others before him, to be
feldspathic, but argues that it is due to metamorphism and to
have been the last mineral to congeal, representing the closing act
in the series of changes through which the stone has passed. The
dark glassy material sometimes surrounding the chondrules, in
stones of the Parnallee type, is considered secondary, the result
of a partial refusion of a fine interstitial material. The metal is
also of secondary origin, this conclusion being based upon its dis-
tribution and manner of enwrapping certain of the fragments
and chondrules, as would occur in the case of secondary precipi-
tation, and also from the fact that in stones of the Cumberland
Falls type metal of two distinct generations can be readily traced.

The idea of the author seems to be that this study having refer-
ence to the original structures of the stony meteorites and the
secondary changes which they have undergone may throw some
light upon the sources from which they have been derived and
their subsequent wanderings.

IU]. Misce,tutanrous Screntiric INTELLIGENCE,

1. Carnegie Institution of Washington. Year Book, No. 20,
1921. Pp. xxu, 475. Washington, February, 1922.—The twen-
tieth year of the lines of research conducted by the Carnegie Insti-
tution is an important point in its remarkable history. Hardly

308 Scientific Intelligence.

less noteworthy is the fact that Dr. Robert S. Woodward, who has
ouided the Institution so wisely for many years, has retired and
his place has been taken by Dr. John C. Merriam, until recently
of the University of California. An adequate knowledge of the
work accomplished the past year can only be gained by a study of
the summaries given by the directors of the different depart-
ments, ten in number, which fill pages 43 to 357 of this Year
Book. ‘To these large grants nearly $1,000,000 of income have
been devoted In addition to these are the minor grants, calling
for an expenditure of about $140,000; the results of which are
noted in pp. 359 to 464.

The number of volumes issued during the year is 18, with an
ageregate of over 4,000 octavo and nearly 1,400 quarto pages. A
total of 442 volumes have been published since the beginning
with a total of over 124,000 pages. Of the many lines of original
work discussed in this report, of especial importance is the first
measurement of the diameter of a fixed star, accomplished at the
Mount Wilson Observatory; for this Drs. Michelson and Pease
with the Director, Dr. Hale, deserve great credit. The star
first measured is the well-known Betelguese and its linear
diameter was found to be 215,000,000 miles. The diameters
of Arcturus and Antares were found to be about 21,000,000
and 400,000,000 miles respectively. Some uncertainty as to these
values arises from the fact that the parallaxes of these stars are
not absolutely known, but the epoch-making character of the
work can be in part appreciated even by the layman. It is also
to be noted that the magnetic survey of the oceans has been prac-
tically completed by the non-magnetic ship Carnegie, which since
1909 has voyaged nearly 300,000 miles.

Recent publications of the Carnegie Institution are noted in
the following list (continued from vol. 3, p. 157, February, 1922) :

No. 175. Bauer, L. A., in collaboration with J. A. Fleming,
H. W. Fisk and W. J. Peters. Land Magnetic Observations,
1914-1920, and special reports by J. A. Fleming, H. W. Fisk,
and 8. J. Barnett. (Researches of the Department of Terrestial
Magnetism, vol. IV.) Quarto. Pp. vi, 475, 9 pls., 17 text figs.
($7.25).—This volume presents, in continuation of the previous
volumes of researches (No. 175, vols. I, II and III), the results
of magnetic observations made by the Department of Terrestrial
Magnetism, 1914-1920, and four special reports:

No. 306. Contributions to the Geology and Palaeontology of
the West Indies; by T. W. VauGHAN and R. T. Jackson. Octavo.
Pp. iv, 122 pp., 18 pls., 6 text-figs —R. T. Jackson’s paper gives an
account of all the species of Echini which have been so far found
occurring as fossils in the West Indies, with keys for the identifi-
cation of genera and species showing their geographical and geol-
ogical distribution. T. W. Vaughan’s paper gives in succinct

Miscellaneous Scientific Intelligence. 309

form, for the use of geologists, what is definitely known of the
stratigraphic occurrence of the species of Echini described in
Doctor Jackson’s paper.

No. 308. Plant Habits and Habitats in the Arid Portions of
South Australia; by W. A. CaANNon. Octavo. Pp. viii, 139, 32
pls., 31 figs. ($2.75) —This is a careful investigation of the
physical environment of the vegetation of different sections of
South Australia, with details as to rainfall, evaporation, relative
humidity, ete. Dr. Cannon has previously issued a somewhat
similar work on the Algerian Sahara (publication No. 178).

No. 311. Shallow-water Foraminifera of the Tortugas Region;
by J. A. CusHMAN. (Papers from the Department of Marine
Biology, Vol. XVII.) Octavo. Pp. 85, 14 pls. ($1.50).—This
work gives the results of the study of some of the living foramini-
fera which have been very rarely studied in tropical waters. The
relationships of the fauna to other regions is discussed and a gen-
eral systematic treatment of the local foraminifera is given in
detail. The new and rare species are illustrated by numerous
plates. |

No. 314. The Behavior of Stomata; by J. V. G. LortTrie.p.
Octavo. Pp. 104, 16 pls., 54 figs. ($1.50).—This treatment of
stomatal behavior falls into three divisions. The first is descrip-
tive, and deals with the hourly stomatal movement for a 24-hour
day of a considerable number of species, including trees, shrubs,
and a wide variety of herbs, both cultivated and native and of
different ecological character. The second deals with causes of
changes in stomatal movement from day to day, as well as during
the daily march. The third deals with the effect of stomatal
movement upon transpiration.

2. Proceedings of the First Pan-Pacific Scientific Conference,
held under auspices of the Pan-Pacific Union. Three volumes,
1921.—It is well known that the first Pan-Pacific scientific con-
ference, held in the summer of 1920 at the Bishop Museum in
Honolulu, was a far-reaching and stimulating success, and some-
thing of the results there attained are here set forth in printed
form. The three volumes contain about 180 papers, totalling
949 pages, relating to almost every phase of science in the realm
of the Pacific and its bounding continents and continental islands.
Most of them are suggestive of things yet to be done, but many
record valuable observations. The feeling of the chairman of the
Conference, Herbert E. Gregory, is one of optimism that men and
means will be found to unravel, if not all, at least most of the
major problems connected with the natural history of the Pacific
‘region. Cais

3. A Laboratory Manual for Comparative Vertebrate
Anatomy; by Linpre H. Hyman. Pp. xv, 380. Chicago, 1921
(The University of Chicago Press).—In nearly all laboratories

Am. Jour. Sci.—Firta Series, Vou. III, No, 16.—Aprin, 1922.
22

310 Scientific Intelligence.

elving a course in vertebrate anatomy the comparative study of
the organ systems has in large measure supplanted the older
study of the entire anatomy of selected types. By the newer
method the student more readily grasps the significance of mor-
phological details and secures a more substantial conception of
the course of evolution.

To accompany such a laboratory study this manual is eminently
fitted. One of the primary objects of the book is to compel the
student to become self-reliant in his work, and the directions and
descriptions are so explicit that he no longer has an excuse for
calling on the instructor for an unreasonable amount of assist-
ance. This is certain to be of benefit both to student and
instructor ; to the former because of the superior mental discipline
involved, and to the latter because of a relief from constant
appeals for help. The book has already met the test successfully.

2 W. R. C.

4. The Vitamins; by H. C. SHz=rMAN and §. L. SmirH. Pp.
ii, 2738 pp. New York, 1922 (The Chemical Catalog Company.
Price, $4.00).—This is an excellent review of a subject which has
“acquired pronounced interest for the biochemist, the physiologist,
the biologist, the bacteriologist, the physician, the agriculturalist,
and the food manufacturer during the past decade. It is not an
easy task to prepare a critical summary of the history and status
of a new chapter in science that already encompasses an enormous
literature which is growing at a rapid rate; but Sherman and
Smith have succeeded. The volume is certain to become an
almost indispensable reference book on the subject of vitamins.
It presents encyclopedic information in a_ readable style.
Although the book includes elaborate compilations of data about
vitamins, there is scarcely a page that does not help to awaken an
interest in the newer aspects of food values or point to the prob-
lems raised thereby. L. B. M.

5. Publications of British Museum of Natural History, Lon-
don, 1921.—Recently issued are the following: -

Catalogue of the fossil Bryozoa (Polyzoa) m the Department
of Geology. The Cretaceous Bryozoa. Vol. III, the Cribri-
morphs, part 1; by W. D. Lane. Pp. ex, 269, with tognen
figures and 8 plates—This work has been carried along the lines
laid out by Dr. J. W. Gregory. The volume is divided into two
sections: Part I includes the Introduction, which is (A) Biologi-
eal; (B) Terminological; (C) Historical. Part II is Systematic,
eiving (1) the characters of the families; (2) doubtful species;
(3) the systematic account (pp. 16-255). Two indexes close the
volume. The excellent drawings for the plates have been made
by Miss Gertrude M. Woodward; those of the text figures mostly
by the author.

Miscellaneous Scientific Intelligence. 311

Catalogue of the Selous Collection of Big Game; by J. G.
DoLuMAN. Pp. vil, 112; with a frontispiece portrait (1906)
by Leo Weinthal of Frederick Courteney Selous.

Economic Sertes.—No. 2.—The Louse as a menace to man.
Its life-history and methods for its destruction; by JAMES
WATERSTON. Pp. 20, with one plate and 2 text figures. 1921.

No. 12. The Cockroach: its life history and how to deal with
it; by FRepERICK Laine. Pp. 18, with 2 text figures and 3
figures on the frontispiece plate.

6. Register zum Zoologischer Anzeiger, begriindet von
J. Victor Carus. Herausgegeben von Prof. Kug—EN KorscHeEtur.
Band xxxvi-xl und Bibliographia Zoologica, vol. xvili-xxil. Pp.
605. Leipzig, 1922 (Wilhelm Engelmann. Price 280 marks,
with addition for foreign countries ).—This monumental work fol-
lows essentially the lines laid down in earlier indexes already
published. The six hundred pages, closely printed and in clear
but small type, give in one alphabetical series both the name of
the author and his various articles with the names of the individ-
ual species. The extent of this work and, at the same time, its
value to the zoologist, may be in a measure understood from the
rough estimate that the Index includes something like 60,000
separate entries.

7. Georg Weber’s Lehr- und Handbuch der Weltgeschichte.
Twenty-third edition, volume I, Altertum; bearbeitet von PRor.
Dr. E. ScHWABE. Pp. xv, 793. Leipzig, 1921 (Wilhelm Engel-
mann ).—Weber’s well-known and widely used Weltgeschichte
was first issued in 1846, seventy-six years ago. Since then many
new editions have been published, and the interesting history of
the development of this remarkable work is given in the preface to
the 21st edition issued in 1902 by Professor A. Baldamus. The
22d and 23d editions have been carried through by Professor E.
Schwabe; the latter bears the date of April, 1921. It is not to
be wondered at that the subject, once covered in a single volume,
has now extended to two -volumes of which the first is now before
us. This embraces the period from the earliest records of pre-
historic man and his work down to the Imperial period of Rome,
about 400 A.D. It is to this wide range of history that the pres-
-ent volume is devoted.

The opening 25 pages give a concise summary of prehistoric
man, his development, speech, religious and political forms so far
as they can be learned from the imperfect records in existence.
Subsequent chapters of part I (I to VIL) deal with the people of
the East: the Chinese, East Indians, Babylonian-Assyrians. the
Semitic people in general and the Israelites in particular, and
finally the Persian. The second part, in four chapters, discusses
in detail Grecian history from 1500 B. C.to 190 B.C. Part third

312 Scientific Intelligence.

takes up Roman history from its beginning to 366 B. C. (chapters
ITand II). The third chapter (366 to 133 B. C.) carries Rome to
its position as the controlling power of the Mediterranean, while
the fourth and fifth chapters extend to the close of the Republic
(133-80 B. C.) and on to the Empire (30 B.C. to 395 A. D.).

The closely printed pages of this remarkable volume contain an
almost bewildering mass of information, systematically arranged ;
the scope of this can only be appreciated by a detailed study of
the whole. It is truly noteworthy that so soon after the end of
. the World War it has been possible to bring out in Germany so
exhaustive a work.

8. Observatory Publications.—Publications of the Leander
McCornuck Observatory of the Unwersity of Virguma; S. A.
MircHenL, Director. 1921. Vol. Il, part 5. Pp. 157-164.—
First stellar parallax measures (reprinted from the Astrophysical
Journal, 42, 268-270, 1915). Vol. II, part 7, pp. 201-268. This
includes 349 parabolic orbits of meteor streams and other results.

Publications of the Washburn Observatory of the Unwersity
of Wisconsin; GrorcE C. Comstock, Director. Vol. X, part 4.
Pp. 167. Observations of double stars, 1907-1919. Madison,
Wis. ~ 1921.

9. Tables and other Data for Engineers and Business Men.
Compiled by CHARLES E. Frrris, University of Tennessee.
Twenty-fourth edition; 160 pp. Knoxville, Tenn. (University
Press; price seventy-five cents).—This volume of engineering
tables, though small in size and modest in price, contains a large
amount of information of great value to the practical man. First
issued in 1905, and repeatedly noticed in these pages, it has now
reached its 24th edition. This fact alone testifies to the value
set upon it by those who use it.

10. Bibliotheca Zoologica ll. Verzeichnis der Schriften weber
Zoologie welche in den wperiodischen Werken enthalten und
vom Jahre 1861-1880 selbstandig erschienen sind; bearbeitet
von Dr. O. TascHENBERG. Lieferung 25. Leipzig (Wilhelm
Engelmann. )—Parts 21 to 23 of this important work were noticed
in June 21,1920; part 20 in July, 1921, and part 24 in January,
1922. Part 25 is now issued under the date of February, 1922
(price 82 marks). It includes signatures 795-804, or pp. 6393-
6472. See No. 6 above.

11. Mentally Deficient Children: Ther Treatment and
Tranng; by G. E. SouttLEworTH and W. A. Ports. Pp. 320.
Philadelphia, 1922 (P. Blakiston’s Son and Company.)—This is
the 5th edition of an old standard which was first published in
1895. The authors have had impressive experience in the medieal,
educational and administrative aspects of the problem of mental
deficiency, and they speak with authority. To American readers
the conerete references to the administration of the Mental

Obituary. 313

Deficiency Act of 1913 will be particularly interesting and help-
ful. There is generous reference to much that has been done in
America, in both the psychological and administrative fields, but
the authors do not give notice to the New York system of pro-
viding colony and institutional care for the feeble-minded which
is one of the most contributive developments in this country.

The chapter on diagnosis is strongest on the medical side.
There is more than the usual emphasis on syphilis as an etiologi-
eal factor.

The volume is compact, convenient, well illustrated and well
arranged, and useful alike to teachers, physicians and lay
students of the problem.

OBITUARY.

Dr. JOHN CASPER BRANNER, president emeritus of Leland
Stanford University, died on March 1 at the age of seventy-one
years. Professor Branner was active in several lines of geologi-
eal work. He was especially interested in Brazil, beginning his
work there in connection with the Geological Commission in 1875,
and as special botanist in 1880-81. He returned to Brazil in
1899, and again in 1907 and spent much of his time there till
1911. His other connections were also varied. He was topogra-
phie geologist of the survey of Pennsylvania (1883-85) and later
state geologist of Arkansas (1887-93). He was professor of.
geology in Indiana University from 1885 to 1892; from 1892 on
he was connected with the Leland Stanford University: first as
professor, later as acting president and finally president, becom-
ing emeritus January 1, 1916. Branner’s wide experience in
Brazil and in the United States enabled him to make many con-
tributions to science; of these a considerable number are to be
found in the pages of this journal. It is not strange that his
labors won for him recognition by election to a number of scien-
tific societies at home and abroad.

Dr. JAMES FRANCIS BoTToMLeEY, distinguished for his original
work in chemistry and physics, died on January 16 at the early
age of forty-seven years. His death is a serious loss to science
and to certain lines of administration, in which he was of great
value to his country. Probably his most important work was
that of silica fusion for which he received gold medals at Brussels
(1910) and Turin (1911). Few men, as noted by Nature (Feb.
16, p. 212), have their gifts and tastes so definitely determined
by heredity. He was the great-grandson of Dr. James Thomson,
professor of mathematics at Glasgow; Lord Kelvin, and James
Thomson (of Queens College, Belfast and Glasgow University)
were his great-uncles. Further his father, Dr. James Thomson
Bottomley, is now on the staff of Glasgow University.

314 Scentific Intelligence.

ProFessor Max VERWORN, the distinguished physiologist, died
at Bonn on November 23, at the age of fifty-eight years. He
was Silliman lecturer at Yale University in 1911 and his lectures
on the physiological relations of irritability were of special value;
these were later published in book form. |

Dr. Giacomo Luter Crmictan, professor of chemistry -in the
University of Bologna, died on January 2. His work was of
oreat value and chiefly in organic chemistry.

Dr. BoyNTON WELLS McFArvanp, assistant professor of Chem-
istry in Yale University, died in New Haven on March 13.

Dr. CHARLES W. WarpNneR, chief physicist of the Bureau of ~
Standards, died in Washington on March 11.

: e Supply-House for ‘Soiontific Material.
Founded 1862. Incorporated 1890.

Nee “A few of our recent circulars in the various
departments:

Ganiiciy: J-32, Descriptive Catalogue of a Petrographic Col-
lection of American Rocks. J-188 and supplement. —
Price-List of Rocks.

Mineralogy: J-220. Collections. J-258. Minerals by Weight.
J-224, Autumnal Announcements.

Paleontology: J- 201. Evolution of the Horse. J-199. Pale-
ozoic index fossils. J-115. Colleciions of Fossils.

Entomology: J-33. Supplies. J-229. Life Histories. J-230.
Live Pupe.

Zoology : J-223. Material for dissection. J-207. Dissections
of Typical Animals, etc. J-38. Models.

Microscope Slides: J- 189. Slides of Parasites. J-29. Cata-
logue of Slides.

Taxidermy: J-22. North American Birdskins. Z-31. General
Taxidermy.

Human Anatomy: J-37. Skeletons & Models.

General: J-228. List of Circulars & Catalogues.

Ward's Natural Science Establishment
84-102 College Ave., Rochester, N. Y., U.S. A.

S “&SCIENTIA”

INTERNATIONAL REVIEW OF SCIENTIFIC SYNTHESIS. Issued monthly (each

number consisting of 100 to 120 pages). Editor: EUGENIO RIGNANO.

This is the only review which has a really international collaboration ; which is
of world-wide circulation and occupies itself with the synthesis and unification of
knowledge, in the history of the sciences, mathematies, astronomy, geology, physics,
chemistry, biology, psychology and sociology.

It is the only review Which, by inquiries among eminent scientists and writers
(on: The philosophical principles of the sciences; Fundamental astronomical and physical
questions of current interest; Contributions given by the various countries to different

-pranches of knowledge; Question of vitalism; Social question; International questions

raised by the world war), makes a study of the problems interesting scientific and intel-
lectual circles throughout the world.

it has published articles by Messrs. : Abbot =Arrhenius -Ashley = Bayliss = Beichman-
Bigourdan = Bohlin - Bohn= Bonnesen = Borel = Bouty = Bragg = Bruni = Burdick = Carver =
Caullery = Chamberlin - Charlier = Claparede = Clark = Costantin - Crommelin = Crowter =
Darwin = Delage = De Vries = Durkheim = Eddington = Edgeworth = Emery = Enriques =
Fabry - Findlay - Pisher = Fowler = Golgi = Gregory = Harper = Hartog - Heiberg - Hinks-=
Hopkins-Inigues-Innes-Janet-Kaptein-Kaye-Kidd-Langevin-Lebedew- =Lloyd Morgan-
Lodge - Loisy = Lorentz = Loria=-Lowell = MacBride = Meillet =Moret=-Muir-Peano -Picard =
Poincare = Puiseux - Rabaud = Rey Pastor = Righi-Rignano-Russell-Rutherford-Sagnac =
Sarton= Schiaparelli=- Scott = See - Sherrington = Soddy = Starling - Svedberg - Thomson =
Thorndike-Turner -Volterra-Webb-Weiss = Zeeman and more than a hundred others.

** SCIENTIA’’ publishes its articles in the language of its authors, and joins to the
principal text a supplement containing the French translations of all the articles that
are not in French. (Write for a Specimen Number to the General Secretary of
“Scientia”, Milan, sending,-to defray postal and other expenses,-2 Francs in
stamps of your country.) Annual subscription: 40 sh., or 19 dollars post free,

Office : 43 Foro Bonaparte, Milan, Italy

_ Publishers: WILLIAMS & NORGATE-London; FELIX ALCAN -Paris
rs NICOLA ZANICHELLI-Bologna; RUIZ HERMANOS-Madrid;

Sal eed a & WILKINS CO-Baltimore.

CONTENTS.

Arr. XIX.—Minor Faulting in the Cayuga Lake Region; a
by E. T. Lone..... sn ta oie eee ee Nace 0G, Gals OLN ea
Arr. XX.—Description of a new Species of Fossil Herring, es
Quisque bakeri, from the Texas Miocene; by D. 8. Jorpan, 249
Arr, XXI.—A New Genus of Fossil Fruit; by E. W. Burry, 251 |
Art, XXITI.—The Relations Between the Purcell Range and ae
the Rocky Mountains in oe Columbia, Canada; by . = hye
Li De BURTING, «hyo y Seas 6 cries ee 254 | h
Arr, XXIII.—Some Complex Chlorides containing Gold. I. |
Pollard’s Ammonium-Silver-Auric Chloride; by H. L. |f
WiBlasg, Nh ce ead See ee eee oS ea Bee la
Art. XXIV.—Studies in the Cyperacee; by Tuo. Horm, 260 |
Art. XX V.—Collophane, a Much Neglected Mineral; oY ai
AB ROGERS cl Cire eee ee 269°

Art. XX VI.—A New Genus of Oligocene Hyenodontide; by
MM. .Re-PeORPR, (0 Gey ou woe eae eee nd ba Be aa QU
Art. XXVII.—The Status of Homogalax, with Two New
Species by Er. Lie TROx Mitre 955 oe lees see oe 288
Art, XX VIII.—A Tantalate and Some Columbites fom Custer ‘
County, South Dakota; by W. P. Heappen, ......... 293 | 7

SCIENTIFIC INTELLIGENCE,

Chemistry ‘aud Physics.—The Separation of the Isotopes of Mercury, G. von
Hevesy: The Color of Ferric Ammonium Alum, J. BonNELL and H. P. |
ParmMan, 300.—The Persulphides of Hydrogen, J. H. Wanton and L. B,
Parton: <A Separation of Germanium from Arsenic, J. H. MULumr, 301.
—Asymmetry of the Gaseous Molecule, R. Gans, "309 _—L’Atome, ERA Sa
ACHALME: Calculus and Graphs, L. M. Passano: The Manufacture Of abs,
Optical Glass and of Optical Systems, F. E. Wricur, 303.

Geology.—Notes on Artic Ordovician and Silurian Cephalopods, A. F. ForRsts,
304.—Stratigraphy of the Pennsylvanian Formations: of North-Central
Texas, F. B. PLummer and R.C. Moorz: Recent Mollusca of the Gulf of |
Mexico and Pleistocene and Pliocene Species from the Gulf States: Hand- —
buch der Relionalen Geologie: The Topographic and Geological Survey of |
Pennsylvania, G. H. AsHiey, 305.—United States Bureau of Mines, H. F, |
Bain, 806.—Metamorphism in Meteorites, 307.

Miscellaneous Scientific Intelligence.—Carnegie Institution of Washington Year ‘.
Book, No. 20, 1921, 307.—Proceedings of the First Pan-Pacific Scientific”
Conference: A Laboratory Manual for Comparative Vertebrate Anatomy,
309.—The Vitamins: Publications of British Museum of Natural History,
London, 1921, 310.—Register zum Zoologischer Anzeiger: George Weber’s —
Lehr- und Handbuch der Weltgeschichte, 311.—Observatory Publications:
Tables and other Data for Engineers and other Business Men: Bibliotheca
Zoologica II: Mentally Deficient Children, Treatment and Training, 312.

Obituary.—J.C. BRaNNER: J. F. Borromtey: M. Vernworw: G. L. Crmictan; |
B. W. McFaruanp: ©. W. WAIDNER, 814. : it

ocmenth U. bi wks ia k= piety Sm hss
ee ate nl or. ie Ss a Sew a

MAY, 1922

Established by BENJAMIN SILLIMAN in 1818.

THE

AMERICAN
| J OURNAL OF SCIENCE. .

Epiror: EDWARD S. DANA. P:

ASSOCIATE EDITORS

Pat saors WILLIAM M. DAVIS anp REGINALD A. ae
OF CAMBRIDGE,

Prormssors HORACE L. WELLS, CHARLES SCHUCHERT,
HERBERT E. GREGORY, WESLEY R. COE anp
FREDERICK E. BEACH, or New Haven,

g Proressor EDWARD W. BERRY, or BALTIMORE, |
Drs. FREDERICK L. RANSOME anp WILLIAM BOWIE, |
_ OF WASHINGTON.

ef. se:

VOL. II—[WHOLE NUMBER, CCIITI).

|
ae FIFTH SERIES
No. 17—MAY, 1922.

NEW HAVEN, CONNECTICUT.

Wee e:

THE TUTTLE, MOREHOUSE & TAYLOR .CO., PRINTERS, 125 TEMPLE STREET.

i

eg Published’ ngoihty: Six dollars per year, in advance. $6.40 to countries in the
ane Postal. Union ; $6.25 to Canada. Single numbers 50 cents; No. 271, one dollar,
Entered as 'secopd- -class matter at the Post Office at New Haven, Conn., under the Act

@ . of March 3, hot a a8

=

INE W WILEY BOOKS

for Geologists

Just Published!

COAL

Its Properties, Analysis, Classification, Geology, Extraction, Uses —
and Distribution
By ELwoop S. MOORE
Dean of the School of Mines, Pennsylvania State College.

*

This book gives, in compact form, the salient facts and theories relating to
coal—for the sv ientist, the practical man and thelayman, Many valuable maps
are included, showing the distribution of various types of coal in pen
every country of the world.

462 pages. 6 by 9. 142 figures. Cloth, $5.00.

Ready June 10th.

HANDBOOK OF THE PETROLEUM INDUSTRY.

By Davip T. Day, Ph.D., Editor-in-Chief, and a Staff of 16 Foremost
Petroleum Experts. :

A monumental work—the only book covering every phase of Petroleum.
Of especial interest to the geologist whose work takes him into oil-bearing lands
are the descriptions of every oil field known in the world, including all the —

rocks with which oil has ever been associated, how to recognize them, Riau dh

where they occur.
Oil shales are also thoroughly treated.

In 2 volumes—1990 pages-profusely illustrated. Cloth, $10.00.

We are always glad to send any Wiley Book on Free Examination, subject to
return if not satisfactory. Our General Catalog is also at your disposal.

JOHN WILEY & SONS, Inc.

432 Fourth Avenue, New York

AJS 5.22.

itoge ly)

AMERICAN JOURNAL OF SCIENCE

PEGI ies hanks 4
———_+0#

Art. XXIX.—Some Complex Chlorides contaimng Gold;
by H. L. WE Ls.

TI. Ceswm Triple Salts.
[Contribution from the Sheffield Chemical Laboratory of Yale University. |

In a preceding article’ a re-investigation of Pollard’s
ammonium-silver-auric chloride was described, and the
formula (NH,),Ag,Au,Cl,, was ascribed to it. Upon
extending the investigation to the employment of cesium
chloride in the place of the ammonium salt a triple com-
pound, not corresponding to the ammonium salt, but with
a simpler formula, Cs,AgAuCl,, has been obtained, while
several other triple salts apparently isomorphous with
this and forming an analogous series with it have been
prepared.

The formulas of these five new triple chlorides, with
their colors, are as follows:

Cs,Ag,Au’”,CL,. Very black opaque, powder black.
Cs,ZnAu’”,Cl,,. Yellow,* transparent, powder pale yellow.
Cs,HgAu’”,Cl,,. Orange, transparent, powder yellow..
Cs,CuAu’’,Cl,,. Crystals black, powder pale brown.
Cs,Au’,Au’’,CL.. Very black, opaque, powder black.

a Sometimes red.

The simplest formulas for the silver and aurous salts
have been doubled in order to make them correspond to
the others, for it has been found that the silver and zine
salts evidently crystallize isomorphously, since the crys-
tals of their mixtures are uniformly black in color, even
when they contain a large proportion of the yellow zine
compound, and there is similar isomorphism with the mer-
euric and aurous salts. This series of salts, therefore,

* This Journal, April number, 1922.

Am. Jour. Sci.—FirtTH Series, Vou. III, No. 17.—Mavy, 1922.
23

~

316 Wells—Complex Chlorides containing Gold.

shows interesting cases of isomorphous replacement
between two univalent atoms and one atom that is
bivalent.

The very intense black color of the two salts contain-
ing univalent silver and aurous atoms is very remarkable.
They form precipitates as black as precipitated lead sul-
phide, while the crystals of the aurous salt sometimes
show a coppery luster similar to that often displayed by
solid dyestuffs.

An attempt was made to prepare a cesium-caleium-
auric chloride, but no such compound was found under a
wide range of conditions. There is little doubt, however,
that other bivalent metals besides those that have been
used in the present investigation would yield triple chlor-
ides with cesium and auric gold.

In preparing these triple chlorides by crystallization a
large excess of cesium chloride, over the theoretical quan-
tity, 1s usually desirable, but the solutions should gener-
ally be very dilute in respect to gold in order to avoid the
deposition of the sparingly soluble double salt CsAuCl,.
Although some of the compounds have been obtained from
practically neutral solutions, the presence of much hydro-
chlorie acid, even up to the full strength of the concen-
trated liquid, is favorable to their formation, apparently
by making them more stable and less soluble. In several
instances, where a solution deposited a triple salt mixed
with CsAuCl,, it was only necessary to dilute considerably
with concentrated hydrochloric acid in order to obtain a
pure crop of the triple salt after dissolving the mixture by
heating and then cooling.

It is characteristic of these triple salts, and also of Pol-
lard’s ammonium compound, that they form very minute
crystals, usually less than 1 or 2 mm. in diameter, whether
they are formed by the slow cooling of the solutions or by ~
evaporation on the steam-bath. The crystals of the dou-
ble salt CsAuCl, that are frequently deposited with them
are invariably of far greater size. The triple salts are
evidently very sparingly soluble in their cold mother-
liquors, for the latter usually show by their pale yellow
colors that very little gold is left in solution in the usual,
preferable cases where the gold is subordinate to the
other constituents in its proportion.

All of the analyzed products were shown to be evidently

Wells—Complex Chlorides contaming Gold. 317

homogeneous by careful microscopic examination. In
some cases the crops of crystals were washed to some
extent by diluting the last part of the mother-liquor with
strong hydrochloric acid, but usually this was considered
unnecessary and was not done. In all cases the products
were carefully dried with filter paper and then in the air.
Further drying at 100° in all cases gave insignificant
losses, amounting usually to about 0.01 to 0.05%.

In all cases experiments were made under widely vary-
ing conditions, but no evidence was obtained of the exist-
ence of more than a single compound of the same three
simple salts, such as occur, as the writer has shown,? to
the number of three or four with the thiocyanates of
cesium, zine and silver, and also those of cesium, cadmium
and silver.

Ceswum-Silver-Auric Chloride, Cs,Ag,Au,Cl,,.—The
very minute, black erystals of this salt frequently show
brilhant faces under the microscope, or they sparkle when
viewed with the naked eye, but sometimes a crop was
found to consist chiefly of rounded particles, probably
crystal-aggregates, showing no distinct faces.

The compound appears to be stable in contact with
strong hydrochloric acid, but it is slowly decomposed by
cold water, more rapidly by the hot liquid, with separa-
tion of silver chloride, and frequently also of yellow
CsAuCl,.

The following analyses of separate crops were made:

Caleulated for
I II EE IV V Vie Wie Vi Cs Ac Aw, Cl,
eset 0.0. BR Teh 0 oe 34.39 33.90

As ..13.60 13.11 13.65 13.54 13.47 13.40 13.49 13.11: 13.77
Au ..25.17 24.73 25.15 24.84 24.93 24:73 25.05 25.74 25.17
eros. ee. 2725 Ee RPO ES ara! ~97 1G

a By difference.

The first three analyses represent crops produced by
cooling from solutions containing 20, 160 and 10 g.,
respectively, of cesium chloride, 3 g. each of gold as
HAuCL,, 1.5 g. each of silver nitrate, and each in a volume
of about 1000 cc., made up of about equal volumes of con-
centrated hydrochloric acid and water. The other crops
were prepared under widely varying conditions, some of

? Amer. Chem. Jour., 28, 278; 30, 144.

.818 Wells—Complex Chlorides contaumng Gold.

them by cooling hot solutions and others by evaporation
to crystallization on the steam-bath.

The method of analysis for silver, gold and chlorine
was the same as that described for Pollard’s salt in the
previous article? The cesium was determined by weigh-
ing the chloride after removing the other metals and evap-
orating to dryness.

Cestum-Zinc-Auric Chloride, Cs,ZnAu,Cl,,.—This salt
gave somewhat larger crystals than the other members of
the series. The crystals were prismatic in habit, and
sometimes as much as 3 or 4 mm. in length. Some of the
crops were yellow in color, but other products showed a
reddish or even a dark red color, and the red crops fre-
quently showed variations in the individual crystals from
yellow through various shades of red. There was no
appreciable difference in the composition of the yellow
and red products. It appeared that solutions which gave
red crops when rather dilute gave yellow ones upon con-
centration by evaporation. The cause of the changes in
color is not definitely known, but it is suspected that the
red color was caused by a slight contamination, due to the
reducing action of filter-paper, with the very strongly col-
ored salt Cs,Au’,Au’”,Cl,., which has been found to give
strong color-effects with the mercuric triple salt, as will
be explained further on in this article.

It was found that this zine salt can be prepared in the
presence of a considerable excess of zine chloride, as well
as of cesium chloride. This is due to the fact that the
double chlorides Cs,ZnCl, and Cs,ZnCl,, described? in this
laboratory long ago, are so readily soluble as not to inter-
fere.

The salt is decomposed by cold water with: the precipi-
tation of CsAuCl,, but the latter dissolves in the presence
of sufficient water.

The following analyses of separate crops under varying
conditions were made:

I II Ine IV V-Caleulated for

Yellow Yellow Reddish Red Red. Cs,ZnAu.Cl,,
CStecaas pays ee nae Beh: eight oor
Ui ihe = es MN Se ee 4.83 4,72 5.82 4.84 4.61
Ave eee 27.46 27.68 27.48 27.80 27.88 27.84
GI Bras Berne fae 5 ane oe meh Pete sane 30.04

* Loe. cit.
* Wells and Campbell, This Journal, 46, 431, 1893.

Wells—Complex Chlorides containing Gold. 319

The analyses were made by dissolving in water, precipi-
tating gold with ammonium oxalate in the presence of a
little oxalic acid, and precipitating zine in the boiling fil-
trate with sodium carbonate. Both precipitates were col-
lected, ignited and weighed in Gooch crucibles.

Mixtures of Cs,Ag,Au,Cl,, and Cs,ZnAu,Cl,.—When
the investigation of the first of these compounds was
begun it happened that the gold employed in the work was
accidentally and unsuspectedly contaminated with a little
zine. This led to perplexing results, for in most cases the
zine compound erystallized with the silver salt, evidently
isomorphously, for all the crystals were very black and
usually had a perfectly uniform appearance under the
microscope. A great many analyses of such crops were
attempted, and these showed wide variations in silver
which made it appear that several triple salts existed,
while attempts to determine cesium in them, either by
weighing the chloride or the sulphate, were mysteriously
unsatisfactory. Since the pure zine salt contains only
4.61% of that metal there was only a very small amount
of it in many of the mixtures, and hence much work was
done before it was detected.

It may be said that a few crops from this impure mate-
rial, especially such as were obtained by hot evaporation
or by partial cooling, were practically free from zine and
led to the correct formula for the silver salt before the
impurity was discovered. |

A few partial analyses of the mixtures, arranged
according to their percentages of silver, are given here to
show the wide range reached by the experiments, where
fractionation was freely used:

Caleulated for Calculated for

I Il eta IV Cs,Ag,Au.Cl,, Cs,ZnAu,Cl,,
sre. rs ae Aa Leis ae 33.90 37.51
eres et Ease 8.80 5.78 1.28 US Pf TE None
ities. 3% SSF a5 Re Di 32? RE None 4.61
Pete ee 0L6c, 26.44 28.15 27.50 Deyeiley. 27.84
les Se bere es Ba) Ach 3 oles Se i 27.16 30.04

When certain crops containing very little silver, like
IV, were obtained some pale crystals were noticed, and
then a careful qualitative analysis revealed the presence
of the zine.

320 Wells—Complex Chlorides contaamng Gold.

Cestum-Cupric-Auric Chloride, Cs,CuAu,Cl,,—_Start-
ing with the beautiful garnet-red double salt CsCuCl,, a
large sample of which has been preserved in this labora-
tory since its preparation by L. C. Dupee,® experiments
were made at first with cupric chloride and with auric
chloride in excess in various proportions, but the triple
salt did not appear until a rather large excess of cesium
chloride was used. In such solutions, when they are cold,
it forms a pale brown precipitate upon the addition of
concentrated hydrochloric acid, but the minute crystals
that form upon cooling hot solutions are very black,
although their streak is pale brown.

Four distinct crops prepared by cooling were analyzed.
The first two were made from rather concentrated,
slightly acid solutions the first of which contained much
more copper than the second and gave somewhat variable
results, while the others were obtained from the same
solutions after dilution with about equal volumes of con-
centrated hydrochloric acid:

Caleulated for

a iit III ini Ve Cs,CuAu.ClL,
Gare IRS S Tern me gent rates 37.20a 37.40a 37.55
Gish R iad) = Se 4.16 3.88 4.68 4.92 4.49
AGT Ee ihe eae. ee 26.89 29.91 27.94 27.66 27.88
01 Ried a EEE Lae Ge ANE So Lee. 30.18 30.02 30.08

a By difference.

In making the analyses the gold was precipitated by
means of ammonium oxalate in the presence of a little
free oxalic acid, and the copper was precipitated in the
filtrate with hydrogen sulphide and weighed as Cu.S.
After boiling off the H.S the chlorine was determined by
the usual gravimetric method. ,

Cesium-Mercuric-Auric Chloride, Cs,HgAu,Cl,..—It is
easy to prepare this salt in a pure condition by cooling
strong hydrochloric acid solutions that are very dilute in
respect to auric chloride and especially to mercuric chlor-
ide, but which contain a considerable excess over the theo-
retical amount of cesium chloride. It is preferable also
to have a very little nitric acid present in order to avoid
the reduction of auric chloride to the aurous condition by
the action of filter-paper or dust.

° Wells and Dupee, This Journal, 47, 91, 1894.

Wells—Complex Chlorides contaming Gold. 321

It forms minute, transparent, orange-red crystals, three
crops of which gave the following analyses:

Caleulated for

I II iNOE Cs,HgAu,Cl,,
SG severest ew ett. 43.45 43.41 42.75 43.37
J BLA GIES Go 8 gee ea ee 18.14 17.66 19.72 17.50
IL es 3 en 38.41 38.93 Bs 39.13
VSG Ce ae Se ee ee is nea Bart BARRE 26.65 27.42

The principal constituents. were very conveniently
determined by heating the substance in a Rose crucible in
a stream of hydrogen, gradually, and finally considerably
below a red heat. The loss in weight gave the mercuric
chloride and the chlorine combined with the gold, while
the residue consisted of metallic gold and cesium chloride.
This last mixture was treated with water and the gold was
collected and weighed, so that the three chlorides could
then be ealeulated. The total chlorine was determined in
a separate portion by the usual gravimetric method after
the removal of the gold and mercury by precipitation.

The pure salt, however, was not obtained at the first
attempts to prepare it. ‘Three successive crops were
obtained from a solution, the composition of which varied
from step to step on account of the removal of portions
of the crops and the effect of additions of mercuric chlor-
ide, and were approximately as follows:

A B C
@estmMmige MOTO: aren en. . nt ey 15g. 14.5¢. 14g.
Mercurie chloride Ws sees ee 2g. 2.528. 3.028.
Golds (as TeAuCly) ee att Baio Ah ys0,: lg. 0.9¢. 0.82.
Fiearochloric eter, ogre es apie fabio a.e a little a little a little
Drolurme sO UO ye oie ees ores ose Geo thie ss 200 ee. 250 ee. 300 ce.

It is to be observed that, in addition to increased dilu-
tion, the principal variation of the solutions is an
increased proportion of mercuric choride, which, even in
the first case, is largely in excess of the amount required
to form the triple salt with the gold that is present.

The crops obtained were in the form of small, prismatic
crystals which varied in color from a light reddish-brown
in A to a pale yellowish-brown in C, with B intermediate,
but the individual crystals in each crop appeared to be
exactly alike in color when examined microscopically.
These crops gave the following results upon analysis:

822 Wells—Complex Chlorides containing Gold.

Differences Differences

A B C A—B A—C
CSCIR Oe, Gees 40.51 39.43 38.37 + 1.08 + 2.14
FeCl, eee 41.59 52.58 60.40 —10.99 —18.81
ATG] ioe rica: 17.90 (09 1.23 + 9.91 + 16.67

It is evident from these results that these products
probably represent mixtures of a compound containing
gold with another containing none of that metal, and it is
not difficult to caleulate from the differences the composi-
tion of the compound without gold. The results of this
calculation are as follows:

Calculated for
From A-B From A-—C CsHg(Cl,
CsOl rs Ht eae eter oh 38.21 38.26
5 od C) yea Bae SB 98 61.43 61.79 61.74

The agreements with the calculation for CsHgCl, are
remarkably close, and there can be no doubt about this
constituent of the mixture. This double salt was
described® in this laboratory long ago, together with
Cs,HgCl,, CsHgCl, and CsHg;Cl,,, all of which yield this
salt upon recrystallization from water. It was found to
be soluble only to the extent of 1.426 parts in 100 of water
at about 17°, and since it is very probably still less soluble
in cesium chloride solutions it is evident that any consid-
erable excess of mercuric chloride should be avoided in
the preparation of the triple salt.

The evident isomorphism of the double salt CsHgCl,
with the triple salt Cs,HgAu,Cl,., as shown by the uni-
form habit and colors of the crystals of their mixtures,
makes it appear possible, unless this isomorphism is con-
sidered as merely accidental, that a multiple formula,
Cs,Hg,Cl,., should be given to the double salt, from which
the triple salt may be derived by the replacement of Hg”,
by Au 12 |

Another complication was encountered in the further
study of this triple salt, for with various experimental
solutions the products obtained varied remarkably in
color. Some of them were bright orange-red, some were
dark brown or nearly black, although transparent, while
others were very black and opaque. These differently
colored products showed little or no variation in composi-
tion, and the cause of the changes in color was not under-

°H. L. Wells, This Journal, 44, 225, 1892.

Wells—Complez Chlorides containing Gold. 328

stood for a considerable time. It was finally found, how-
ever, that the dark color was due to the reducing action of
filter-paper and the formation of aurous chloride. The
‘solutions had been filtered frequently and extracts of the
filter-papers that had been used for drying the crystals
had been added to them. When a little nitric acid, left
after dissolving the gold, together with free hydrochloric
acid were present in the solutions the reduction was pre-
vented, but otherwise a single filtration of a hot solution
that had previously given an orange-red product caused
the formation of avery dark one. The following analyses
of some of these products were made:

i II 10g: IV V VI
Dark Nearly Calculated for
brown, black, Cs,HgAu,Cl,,
trans- trans- - Black, Black, Black, Black, Orange-
parent parent opaque opaque opaque opaque red
Petes. | 3425 33.39 33.99 34.09 a Eh 33.89 34.23
PRG AS tars. 23 5 ae Wey Re 14.81 14.24 12.93
Aw. s2 24.99 26.19 25.07 25.15 24.32 24.52 25.42
LS aeee ~ Fag ances afee 26.48 26.72 27.42

The first four analyses were made by the method of
heating in hydrogen, already mentioned, and on account
of the presence of aurous chloride reliable results for
mercury were not obtained. The determinations of mer-
cury and chlorine were made by weighing mercuric sul-
phide and silver chloride. -

The results show that the dark products differ but

shghtly from the pure compound in composition. The
high percentages of mercury in the last two products indi-
cate that too large a proportion of mercuric chloride was
used in their preparation, resulting in the crystallization
with them of a little of the isomorphous double salt
CsHeCl..

Efforts were made to prepare the pure black compound,
or to obtain some light upon its composition, by preparing
products from solutions that had been subjected to long,
hot digestion with filter-paper, but, on account of the
instability of aurous chloride, metallic gold was usually
precipitated during the digestion, and products of a con-
stant composition were not obtained in this way. After
the pure black triple salt had been prepared in a pure con-
dition by another method, to be described in the next sec-
tion of this paper, and its composition was known to cor-

324 Wells—Complex Chlorides contaimng Gold.

respond to the formula Cs,;Au”,Au’”,Cl,. it was possible
to calculate the composition of the mixtures obtained by
reduction with filter-paper. In the case where the prod-
uct showed the greatest variation from the orange-red
salt in its composition, the original solution was prepared
from 6g. of the orange-red compound, 10 g. of cesium
chloride and about 300 cc. of water. Then a single 10 cm.
filter-paper was added and the whole was heated in a cov-
ered beaker upon the steam-bath for 24 hours. The filter-
paper and a considerable quantity of gold that had pre-
cipitated were then removed by filtration and a crop of
very black and opaque, apparently homogeneous erystals
was obtained by cooling. Itis evident, from the composi-
tion of the original solution and the fact that gold was
removed by precipitation, that the product was formed in
the presence of an excess of mercuric chloride, and conse-
quently that it was likely to contain the salt CsHgCl, as
an isomorphous constituent. The results of the analysis
and the calculation are as follows:

Caleulated for

10 % 20 % 70 %
Analysis Cs,Au’,Au’’’,Cl, CsHgCl, Cs,HgAu’’;Cl» Total
OSU creosote oie Sees 3.04 6.04 23.96 . 8804
IB Cais ae Aone abo SS) SESS 9.12 9.05 18.17
JAC as een cee 22.22 4.52 oes: 17.79 22.31
OURS As eee ees 26.29 2.44 4.84 19.19 26.47

The amount of the aurous compound was calculated
from the deficiency of chlorine, 0.78%, shown by the
analysis, below the amount required to give AuCl, with
the gold. Then the percentages of CsHgCl, and Cs,Hg
Au,Cl,, were calculated by finding the -proportions of
these that would bring the gold to the amount found. In
this calculation the required proportions of the constitu-
ent salts are given only approximately, but the fair agree-
ment of the total with the analysis is a satisfactory one.
The assumption, therefore, that this product was com-
posed of three isomorphous salts appears to be a plausible
one.

_Cesium-Aurous-Auric Chloride, Cs,Au’,Au’’’,Cl,>—
When a concentrated solution of cesium chloride, best in
1:1 or stronger hydrochloric acid, is mixed with solid aur-
ous chloride the result is very striking, for an intensely

Wells—Complex Chlorides containing Gold. 325

black precipitate is instantly formed, while metallic gold
is also deposited. The black compound is the triple salt
under consideration. It is not easy to obtain it in a pure
condition, free from metallic gold and the yellow double
salt CsAuCl,, but Lee anne it can be recrystallized sat-
isfactorily.

A good crop was eeuned by treating 5g. of aurous
chloride, made by heating HAuCl,, with 30 g. of cesium
chloride dissolved in a little 1:1 hydrochloric acid, then
diluting with the same acid to about 400 cc., heating to
boiling, filtering and cooling. Another satisfactory crop
Was prepared by evaporating on the steam-bath, until
erystallization took place, a similar, but somewhat more
concentrated solution. The first product was simply
dried by pressing on paper, but the second one was
washed to a considerable extent, before drying, by largely
diluting the last part of the mother- liquor ‘with hydro-
chloric acid.

The salt forms very minute black crystals which are
rapidly decomposed by water with the formation of
cesium-auric chloride and metallic gold, but they appear
to be very stable with strong hydrochloric acid.

The results of the analyses of the two crops that have

been mentioned are as follows:
Calculated for

I EE Cs,Au’,Au’”.Cl,>
SCH Pha: i oer hcn 2. OOo 37.89 38.56
Psat Bie Abe 5 Mads 44.86 45.92 ; 45.19
(GLEE PEPE ns nee, 3 ae 16.31 16.19 16.25

The analyses were made by heating the substance
(dried at 100°) ina Rose crucible in a stream of hydrogen.
The loss in weight gave the chlorine, while the residue
consisting of gold and cesium chloride, was treated with
water and the gold was collected and weighed.

An attempt was made to prepare this triple salt, the
simplest formula for which is CsAu”Cl,, by heating the
double salt CsAu’’Cl,. The latter became intensely black
at about 320°, but the loss in weight at this temperature
was only 1.68% instead of the theoretical 7.51%, so that
the change, if it did take place, was only superficial or
partial. Upon heating to the melting-point without
measuring the temperature, a reddish liquid was formed
which became intensely black upon solidification, but the

3826 Wells—Complex Chlorides contamng Gold.

loss was then 9.56%, while chlorine was still being given
off when the heating was stopped. The pure triple salt,
therefore, was not prepared in this way, but the intense
black color of the product makes it appear that it is prob-
ably formed to a considerable extent at least.

Some potassium auric chloride, KAuCl,, was heated in
a similar manner, but, although the product became very
dark brownish-red, no intense black color was observed,
as in the case of the cesium salt, and hence it seems doubt-
ful that a potassium-aurous-auric chloride can be pre-
pared in this way.

Summary.—lt has been shown in this and the preceding
article on Pollard’s salt:

That the ammonium triple chloride (NH,),Ag,Au;Cl,,
does not correspond in its type of formula to the cesium
salt Cs,Ag,Au,Cl,».

That no corresponding potassium triple chloride could
be prepared.

That a series of cesium triple chlorides, Cs,Ag,Au,Cl..,
Cs,Au’,Au’”’.Cl., CsisZnAu,Cl,, Cs,HgAu.Cl,, and Cs,
CuAu.Cl, can be prepared, which show evident isomor-
phism where two univalent silver or aurous atoms are
replaced by a bivalent atom.

That two of these new salts, Cs,Ag,Au,Cl,, and Cs,
Au’,Au’”’Cl,. possess an astonishingly black color in com-
parison with the salts containing bivalent metals.

That the double salt CsHgCl, is isomorphous with
Cs,HgAu.Cl, and with Cs,Au’,Au’”’,ClL,., thus indicating,
perhaps that the multiple formula Cs,Hg,Cl.. should be
ascribed to the former.

That no cesium-calcium-auric salt could be prepared.

Nore: The next article in this series will describe a new
eesium-aurie chloride, and this will be followed by one dealing -
with a general discussion of triple salts, as well as another on a
chromophore grouping, suggested by the black salts described
here, and a consequent theory of the cause of the colors of
substances.

- New Haven, Conn., March, 1922.

E. W. Berry—An American Spirulirostra, 327

Arr. XXX.—An American Spirulirostra; by Epwarp
W. Barry.

In 1906 Emil Bose! described certain Tertiary mollus-
ean faunas from the Isthmus of Tehuantepec in southern
Mexico. These faunas as described comprised less than
fifty species, and included no Cephalopoda, and the erro-
neous conclusion was reached that they indicated a lower
Pliocene age.

Through the courtesy of the Transcontinental Oil Com-
pany I have received large collections of invertebrates
from Tehuantepec, and it may be definitely stated, that
this fauna, imperfectly described by Bose, is a rich trop-
ical shallow-water fauna of several hundred species, and
is clearly of Miocene age, as was recognized in the field
by Dr. Bruce Wade, the paleontologist of the Company.
It is hoped, that in the course of time a fully elaborated
account of this interesting fauna will be published by one
of my students in paleontology. Meanwhile I wish to
eall attention to one of the more spectacular elements of
this fauna. 3

I refer to unusually good material of a new species rep-
resenting the genus Spirulirostra which constitutes the
first record of this interesting genus in the Western
Hemisphere. The material is more complete even than
the celebrated Spirulirostra bellardi described by d’Or-
bigny from the upper Miocene of Superga, near Turin,
Italy—a reproduction of d’Orbigny’s figures of which
have embellished nearly every textbook of conchology and
paleontology that has been published between 1842 and
the present time.

This new species from Mexico may be ealled Spiruli-
rostra americana. Upwards of a dozen specimens have
been found. These usually represent merely the more
resistant guard or rostrum with a part of the included
phragmocone, and they are generally much worn by wave
action. The best specimen, which is also slightly above
the average in size, is scarcely worn, and shows a nearly
complete guard with its contained phragmocone, and a
large part of a proostracum; and it is this specimen from
which the accompanying illustrations have been drawn.

* Bose, H., Bol. Inst. Geol. Mexico, No. 22, 1906.

328 = E. W. Berry—An American Spirulirostra.

The guard or rostrum terminates behind as a smooth
conical point. As it expands forward the rounded ventral
face develops into a prominent boss, above which it is
excavated below, and widened on either side into wide
thick flanges, which form a sweeping forward curve to
where they unite with the proostracum in front of the
phragmocone. The boss of the rostrum and the margins
of its wings are prominently mammilated, the surface
reflecting the vascular structure of the enclosing mantle
of the adult animal.

On the dorsal side the rostrum commences to fattea at
a point about opposite its ventral boss, resulting in
diverging rounded shoulders, flat top, ‘and flattened
sloping sides. The surface of the top becomes smooth
as it is continued forward but the sides show subordinate
vascular markings. The forward continuation of the
rostrum forms a thick wall about the phragmocone, the
earlier chambers of which form a subordinate boss, cen-
trally located in the ventral concavity of the rostrum,
some distance in front of the prominent ventral boss.
The former is usually broken in the fossils, and the earlier
chambers of the camerated shell can be seen through the
resulting opening.

The protoconch is bulbous like those of the ammonites, |
and is not embedded in the ventral boss of the rostrum
as it is in the other known species of Spirulirostra. The
conch for the first five or six chambers is tightly coiled,
endogastrically, but there is no impressed zone. The
diameter of the chambers increases very rapidly as they
curve upward away from the nepionic coiled portion, and
then forward, as shown in the accompanying diagram-.
matic longitudinal section. In all there appear to have
been 15 or 16 septa, which were probably transverse and
nearly straight in profile, although they are broken away
except near the sutures in every specimen seen. The
siphon was small and followed the ventral wall. The
phragmocone is circular throughout in cross section.

The proostracum is a thick, spatulate forward projec-
tion from the dorsal part of the rostrum, nearly flat
above, with rounded edges, and mammilated surface,
which becomes more conspicuously so toward its anterior
rounded margin. Viewed from below it shows two flat
lateral wings, each occupying about one fourth of its

E. W. Berry—An American Spirulirostra. 329

total width, and with the usual vascular markings. The
central half of its ventral face gradually rises from the
general surface in front of the phragmocone until at its
anterior end it projects downward a distance twice as
great as the proostracum is thick. This broad central
keel is excavated to form a shallow arch, curving slightly
forward and downward as a continuation of the arched
dorsal part. of the phragmocone, but there is no trace of
ventral or side walls or any vestiges of there ever having
been any septa. The anterior end of this squarish keel
is broken away below the anterior margin of the proostra-
cum as though in life it had continued forward a greater
or less distance beyond the dorsal rounded margin of the
proostracum as a sort of gladius or pen.
Dimensions:

Total length, 5.1 em.

Distance from tip to center of ventral boss, 1.15 em.

Distance from the tip to the boss marking the initial chambers
of the phragmocone, 1.8 em.

Length of the phragmocone, 1.7 em.

Greatest diameter of the phragmocone, 6 mm.

Greatest distance between the later formed septa, 1.5 mm.

Length of proostracum, 2.6 em.

Maximum width of the proostracum, 1.1 em.

Maximum height of the rostrum in the region of the ventral
boss, 1.0 em.

The various features that I have attempted to describe
in the foregoing paragraphs are well brought out in the
accompanying dorsal, ventral, and lateral views?, and in
the two diagrammatical, partly sectional drawings.

This species is clearly distinct from previously known
forms. From the best known of these, Spirulirostra bel-
lardi, it differs in being slightly larger, with less sharply
conical guard, its coiled early chambers, more flattened
and much larger proostracum, and in having the phrag-
mocone iniated as an internal protoconch some distance
in front of the ventral boss, and not ventrally from a
protoconch imbedded in this ventral boss. I am writing
in terms of the finished product, and it should be under-
stood that ontogenetically, the phragmocone preceded the
guard and proostracum in order of development.

* The two best of these I owe to the skill of G. S. Barkentine.

330 EH. W. Berry—An American Spirulirostra.

In addition to the type of the genus, Spirulirostra bel-
lardi,? I know of four additional records. These are
Spirulirostra hoernest, described by von Koenen* from
the Miocene of northern Germany; Spirulirostra szaj-
nochae, described by Wojcick® from the Clavulina szabot
beds (Oligocene) of Galicia. This last is represented by
a much worn rostrum showing traces of 7 septa of the
phragmocone; there 1s no ventral boss to the rostrum and
the specimen is smaller than the American form and much
like Spirulirostra hoernest. It may represent the same
species as Roemer’s record from the Oligocene of West-
phalia. The third is Spirulirostra curta Tate® repre-
sented by rare, much worn specimens from the marly lime-
stone of Janjukian age (Miocene) of Victoria, Australia.
I fail to see any material differences between Tate’s
species and Spirulirostra bellardy d’Orbigny. ‘The
fourth is an obscure specimen from the Oligocene of
Westphalia, named Spirulirostra sp., by Roemer’, and of
undetermined affinity. i
Spirulirostra hoernest was based upon two specimens
from Dingdan, in western Westphalia. This species is
shorter and stouter than the American form, with a more
prominent and forward projecting ventral boss, with the
phragmocone starting in the ventral boss as in Spirulr-
rostra bellardi, and with a suggestion of a proostracum
in the short, narrowed and truncated dorsal forward
extension of the rostrum.

These occurrences suggest something of the probable
habits of the animal. If we consider merely the Miocene
records, which are more or less nearly synchronous, the
unusual and rare occurrence of these fossils, at such
widely removed localities as the Roman Mediterranean,
North Germany, the Caribbean region, and Australia,
appear to indicate that the animal was a pelagic form.
Whether it had the habit of resorting to shallow water
seasonally to deposit its eggs, as do some of the existing

*d’Orbigny, A., Ann. Sci. Nat., 2 sér., tome 17, p. 374, pl. 11, figs. 1-6,
1842. I have counted 28 reproductions of d’Orbigny’s figures im various
text-books, and this probably does not exhaust the record.

‘Koenen, A. von, Palaeont., Band 16, p. 145, pl. 14, fig. 6a-h, 1869.

5 Wojcick, M. K., Bull. intern. acad. Sci. Cracovie, No. 10, 1903, p. 802, pl.
17, fig. 32, 1904.

‘ Tate, R., Proe. Roy. Soc. N. S. Wales, vol. 27, p. 170, pl. 1, fig. 1, 1893.

" Roemer, F., Neues Jahrb., 1851, p. 576.

E. W. Berry—An American Spirulirostra. 3381

Figs. 1-3.—Ventral, dorsal and lateral views of Spirulirostra americana
Berry, <2

sonlan ingsz,

A \
Am. JOUR. SCI.—FIFTH SERIES, Vo-. IIT, ; Out. —May, 1922.
2,

24

332 EH. W. Berry—An American Spirulirostra.

cuttles that habitually pass their time in deeper water,
or whether the shells were postmortem additions to the
littoral faunas in which they are found, cannot certainly
be determined. We can dismiss the idea advanced by
one writer on Spirula, that air in the shells is responsible
for the occurrence of that form on tropical sea beaches,
as rather far fetched. The common European Sepias,
which normally live in from 10 to 40 fathoms, come into
shallow water during the summer to deposit their eggs,
but it would seem that if such a habit were invoked to
account for the presence of the Spirulirostra shells they
ought to be discovered more frequently. On the other
hand if chance drifted specimens from the open sea
account for their occurrence in the fossil record, it is
difficult to see why a single collection lke that from
Tehuantepec should contain the remains of a dozen
individuals.

The habits of the existing Spirula might throw some
light on the habits of Spirulirostra, but unfortunately
little is known of the former, and the number of spirulas
that have been taken could be enumerated on one’s fingers,
although the shells are common enough in the warmer
parts of the Atlantic, Pacific and Indian oceans. The
single animal taken during the extended cruise of the
Challenger came from between 300 and 400 fathoms, but
appeared to have been partially digested in a fish
stomach, so that it sheds no light on the problem. | It
leads, however, to the suggestion of another possible
means of transportation from deeper water, namely, in
fish stomachs. Fishes are fond, of the existing cuttles,
and the habit of voiding undigestible hard parts is a
common enough one on the part of fishes.

Without reaching any conclusion as to the source of
the Spirulirostra remains in the deposits where they
have been found, I think the conclusion is warranted that
the animals were normally active pelagic types. The
existing Spirula is unknown in the fossil record, and its
openly coiled phragmocone shows no traces of having
ever had a guard or proostracum. Some students have
emphasized this point as though it were evidence of more
direct relationship with the original ammonite stock than
with the belemnoids. Such a line of reasoning, assuming
that Spirulirostra is related to Spirula, which seems

E. W. Berry—An American Spirulirostra. 338

Figs. 4, 5.

Fies. 4, 5.—Diagrammatic, partly sectional ventral and lateral views of
Spirulirostra americana Berry, X 2.

(a)

334 HEH. W. Berry—An American Sprrulirosira.

probable, would necessitate considering Spirulirostra as
a more evolved type than Spirula which had secondarily
acquired a rostrum and proostracum. ‘This is an absurd-
ity for which the evidence is so rare, if indeed there is any
instance, that it has led to the formulation of what is
known as Dollo’s law of the irreversibility of evolution.
I think it may be concluded that both Spirulirostra and
Spirula show vestiges of their more remote ammonite
ancestors in their protoconch and ventral siphon, but
that their more immediate ancestors were some unknown
belemnoid forms, and that Spirulirostra might well serve
as a prototype of the existing Spirula which subsequently
lost all traces of the guard and proostracum, features
whose loss may possibly be correlated with the modifica-
tion of body form from more gracefr.t lines in the diree-
tion of the short, stout bodied and tcuneated hind end of
the existing Spirula.

G. P. Merrill—Meteoric Iron from Texas. 335

Arr. XXXI—Meteoric Iron from Odessa, Ector Co.,
Texas;* by Getorce P. Merri.

The fragment of an iron meteorite described below was
brought to the writer’s attention by Dr. A. B. Bibbins
of Baltimore who states that it was found by a ranchman
at the west side of a ‘‘blow out’’ about nine miles south-
west of Odessa, a little east of Section 8, Block 45, Twnp.
3, S. Eetor Co., Texas, and placed in his hands as a
possible sample of iron ore. As received the fragment
weighed 1,120 grains and was stated to have been cut
from a larger mass—size not given. [Eixteriorly it was
much weathered and oxidized and gave little indication
of the customary pittings. The accompanying figure of
a slice cut parallel with the greater dimensions is natural
size. As will be noted, the structure is octahedral and
of coarse erystallization (Og.). The etched surface is
dull and lusterless and abundantly sprinkled with small,
angular areas of schreibersite.

As the light-is reflected from the etched surface at
varying angles, the interior field gives the effect of having
been cut parallel with the broad kamacite plates, while
in the outer marginal portion they are cut more nearly
at right angles. (See fig. 1.) The suggestion is that of
an intergrowth of two portions of unlike orientation.
Tenite plates are thin and inconspicuous as are plessite
areas, the entire mass being composed mainly of the
broad kamacite plates and the included schreibersite.

A slice of the iron freed from all crust and oxidation
products and containing no visible troilite was turned
over to Mr. HK. V. Shannon of the Museum for analysis.
He reports as follows:

‘‘A portion of the iron weighing 13.3577 grams was
dissolved in aqua regia and the nitric acid expelled by
repeated evapor ation. with hydrochloric acid. The chlo-
ride solution was filtered on a Gooch filter, the residue
being tabulated below as carbon. It possibly includes
some other extraneous substances, as dust, ete. This resi-
due was ignited after weighing ‘and was practically all
destroyed “by the ignition. After ignition the asbestos
mat was fused with potassium pyrosulphate, leached with

* Printed by permission of the Secretary of the Smithsonian Institution.

336 G. P. Merrill—Meteoric Iron from Texas.

water, the solution acidified and precipitated with ammo-
nia, ignited and fused with sodium carbonate and potas-
sium nitrate and leached with hot water. Only a doubtful
and exceedingly faint trace of chromium was found.
The chloride solution of the entire 13.36 grams was
precipitated with hydrogen sulphide, the precipitate being

Fiq. 1.

Fie. 1—Meteoric Iron, Odessa, Texas. Natural size.

examined for copper and platinum. No platinum could
be detected. The copper was thoroughly proven. The
solution, after removal of hydrogen sulphide, was made
up to 1 liter and portions of 50 cc. (.6679 gm.) were used
for determination of iron, nickel, phosphorus and man-
ganese, also soluble chromium. No manganese could be
detected by the colorimetric method, nor could any soluble
chromium be found. Cobalt was determined by the nitro-

G. P. Merrill—Meteoric Iron from Texas. 337

sobetanapthol method on a 200 ee. portion of the solution
(2.6716 gms.). Its identity was thoroughly established.

Sulphur was determined on a second portion of the iron
weighing 13.4440 grams. This was dissolved in nitric
acid, the sulphur separating in part as such in visible
floating particles. This was oxidized with fuming nitric
acid and determined as barium sulphate.’’ The results
of the analysis are as follows:

[SIROTA oan ane tere ey oe eee Ae ae 90.69
ING ee Ree, he Ae A ee 1:25
Cobia: aa te oe ities ee 14
(CODINCIE NS, Ria ene rene Oy 02
eer GMT relic ake Ps ale oes none
Ghent = oe pe ek es trace
Wipe AINE SCout eae eee ie oes ae none
CarepOme tec ce Sin hs apace oa BD
J EA UKC SI OMNOTEW Sie ata © en ea oe eC 20
SHUT OLN UE Seen netee tek a ee ett ns Se O 03

Morelli gyts oe clay am ee Oe Se 9) Bll

These figures fall well within the limits of other anal-
yses of iron of the coarse octahedrite group and need
httle comment. The small amount of material available
for analyses is probably the cause of the apparent absence
of platinum.

National Museum, Washington, D. C.

338 Jordan—Sharks’ Teeth from Califorma.

Arr. XX XIi.—Some Sharks’ Teeth from the California
Pliocene; by Davin Srarr JoRDAN.

To Mr. H. Maus Purple, general manager of the Tor-
rance Lime and Fertilizer Company of Los Angeles, I am
indebted for the opportunity to examine a number of
shark teeth of the man-eater type (Lammideé), surprising
in the fact of their coming together in one place. These
teeth were found in deposits of bones and shells of Pleisto-
cene age, composing hills at Torrance and Lomita,
suburbs of Los Angeles, between the city and the ocean.
The shark teeth may be described in detail.

1. CaRCHARODON BRANNEBRI Jordan.

Two specimens—both of extraordinary size, as large as
the largest Carcharodon megalodon of deposits along the
Atlantic.

The largest of these has the crown three inches in
height, the oblique length of its distal margin six inches.
It is somewhat oblique, the interior more convex and rela-
tively vertical; tip rather blunt. Edges of the tooth
somewhat irregular with obsolete serrations—but no well-
defined serrations except near the base.

A second specimen shows about half the tooth, split
lengthwise. It shows the long exterior margin about four
inches long, the crown 234 inches high. ‘This like the
other is flat or a bit concave on the interior side, but the
sharp edge is obviously but very finely serrated, the ser-
re blunt, 120 to 125 in number along the side.

These teeth may be provisionally identified with Car-
charodon branneri described by me! from Bolinas Bay,
California. These specimens are larger than even this
giant species, and the serre much finer. The type of
Carcharodon brannert may however have been a median
tooth of a smaller example, the teeth perhaps less worn.

In the collection from Torrance, there is another large
tooth which corresponds almost exactly to Carcharodon
branneri. The crown is 234 inches high, the long margin
somewhat over three inches. The tooth is more erect

*The Fossil Fishes of California, Univ. of California, Publ., V. no. 7,
105 JUL, OR

Jordan—Sharks’ Teeth from Califorma. 339

4 Fig. 1, a to g.

Fig. 1.—a. Carcharodon branneri, Jordan. b. Carcharodon riversi, Jordan.
¢. Isurus planus, submedian tooth, Agassiz. d. Carcharodon carcharias. I.
e. Isurus (?) glaucus, submedian tooth, M. and H. ff. Isurus (?) glaucus,
lateral tooth, M. and H. g. Isurus (?) glaucus, median tooth, M. and H.

340 Jordan—Sharks’ Teeth from California.

than the others, the interior edge quite flat and the exte-
rior quite convex. The tooth is serrulate to the tip, the
serre coarser than in the larger examples, all bluntish
and about eighty to be counted, twenty or more appar-
ently broken off.

This tooth is plainly identical with the type of Carcha-
rodon bannerz but it may be different from the two larger
examples. Perhaps the distinction is due to its being less
worn and from a different part of the mouth.

From Carcharodon megalodon Charlesworth, the giant
species of deposits along the Atlantic Coast, the present
form is plainly different as in C. megalodon, the serre are
far larger and coarser.

In life, the present species must have reached a length
of more than one hundred feet, as Carcharodon car-
charias, the living species of ‘‘Man-Hater,’’ reaching a
length of thirty-five feet, has teeth barely an inch in
height.

2. CARCHARODON ARNOLDI Jordan (Carchorodon riwverst
Jordan).

In the same collection from Torrance is a specimen
which corresponds perfectly to the type of Carcharodon
riverst Jordan, described by me on page 115, of the same
paper, from Pliocene deposits near Santa Monica. In
this example the crown is nearly two inches high, the
tooth narrowly triangular, nearly flat, and about as high
as broad at base. The denticles are coarse and blunt, the
total number being about 45 on each side. This species is
however very doubtfully distinct from Carcharias arnoldi
described by me (p. 118) from the Phocene of Pescadero,
and since found in different deposits of the California
Miocene at Lompoe, in Kern County, and elsewhere.

3. CARCHARODON CARCHARIAS (L.).

From the same deposits at Torrance, but possibly at a
higher level I have a small tooth which must belong to the
living ‘‘Man-Hater,’’ still extant on the California Coast.
It is an inch in elevation, 114 in slant height, narrower
than C. arnoldi and with the edges more flexuous. The

Jordan—Sharks’ Teeth from Califorma. 341

NIGS) 2eand 3.

PIGS. 25, 73:
California.

Carcharodon branneri, Jordan, Pleistocene: Torrance,

Serre are strong and sharp, much larger and sharper than
in the extinct species found in California, about 32 om each
side, besides eight or ten much smaller ones on the lower
angle of the tooth, the median serre much larger than

342 Jordan—Sharks’ Teeth from California.

those at the base and tip. Tip of the tooth sharper than
in any of the others, the root less concave in outline, the
crown narrower, its base 114 in its height.

4. Isurus puanus (Agassiz).

(Oxyrhina tumulus Agassiz: Isurus snuthi Jordan).

From the same deposit, I have the tooth of an Jsurus,
the crown of which is 114 inches high, the tooth narrow
and nearly erect, corresponding fairly to the figures of
Tsurus smithi Jordan (p. 11), but broader at the base than
any of those figured, the base of the crown being two-
thirds its height.

As in this genus, the teeth are very differently formed
in different parts of the mouth. I do not lay stress on
these differences, and I have no doubt of the identity
of Isurus planus, twmulus and smith, all from the
Kkern County beds. And these differ but slightly from
Isurus hastalis (Agassiz) of Kurope, with which Jordan
and Beal, in a later paper (op. cit., vol. ViI, 251, 1913),
following Maurree Leriche, have identified them. In
most of the specimens figured by Agassiz, however, as
hastalis, the inner face of the tooth has an obsolete ridge.
This appears on one or two only of our multitude of speci-
mens of [surus planus. In view of the fact that all these
California fossils have been described as distinct species,
it is better to retain these names until adequate com-
parison can be made with related species in Huropean
deposits. The living sharks along the Pacific Coast have
vet to be eritically compared with their Atlantic relatives,
and in most cases where comparison has been made, the |
species are found to be distinct.

From Torrance, I have also three much smaller teeth
which may be merely the young of the same species as
they seem too large for the living Mackerel Shark, /surus
glaucus (Muller & Henle), still extant in the Pacifie.
These come from different parts of the mouth, one, a slen-
der median tooth, being an inch high.

The genus [swropsis Gill, represented in both oceans,
has the teeth essentially as in /surus proper (Oxyrhima
Agassiz) but the lateral teeth are more slender. IJswrop-
sis 18 probably not tenable as a distinct genus.

Stanford University, California.

O. Holtedahl—U pper Cambrian Fauna. 343

Art. XXXIII.—An Upper Cambrian Fauna of Pacific
Type in the European Arctic Region; by Ovar Houte-
DAHL.

During the recent Norwegian scientific expedition to
Novaya Zemlya, the long, arched islands north of the Ural
mountains, the present author, who was also the leader of
the expedition, tried to clear up as much as possible of the
stratigraphy and tectonics of that vast country.

I shall not here go into the question of the general geo-
logical structure of this old mountain range, folded in
Permian time, nor into the stratigraphy of those Paleo-

zoic formations which have been previously known to

exist in these islands, viz., the Devonian and the Carbon-
iferous, but will only bring to the attention of geologists
the occurrence here of strata as old as the uppermost
Cambrian. As our knowledge of Cambrian rocks beyond
the 70th degree of latitude is exceedingly searee, the find
is of considerable interest. It may be useful, therefore,
to give a short preliminary note of this discovery, as the
working out of all of the fossils gathered during the expe-
dition will take considerable time.

The fossils were found near the west coast of the south-
ern isiand, on the peninsula between Bessimyanni and
Gribovii fjords. The first fossils were found in the moun-
tains 7 kilometers northwest of the head of Bessimyanni
Fjord (about 73° N. L.) im a rather dark grey, fine-
erained caleareous sandstone. By following the general
strike of the rocks (here N.-S.) I found further north a
fossiliferous sandstone of a somewhat different character,
lighter colored and with thin layers of somewhat crystal-
line limestone. This rock continues down to the Gribovii
Fjord. The preservation of the fossils is generally
rather bad and their original form has commonly been
altered by the tectonic pressure.

At the southern locality were found the following forms:
a species of brachiopod that is by far the most common
fossil and that belongs to Huenella Walcott, a genus
known from the Gambian of North America, China, and
Australia. The form, of which illustrations are given in
fioure 1, in its general exter ior features is very suggestive
of fai, texana W aleott from the Upper Cambrian of Texas.
Indeed, there exist in my material many valves that do not

344 O. Holtedahl—U pper Cambrian Fauna

in any respect differ from those shown in Walcott’s illus-
trations (see fig. 1, p. 103 of Cambrian Brachiopoda, U. 8.
Geol. Surv. Mon. 51, 1912). Rather often, however, the
Arctic specimens show more acute cardinal angles (com-
pare the cast of the dorsal valve figured here) and more
numerous plications, these occurring rather evenly dis-
tributed both over sinus and fold and the lateral parts.
This species reminds one strikingly of a small Platystro-
plua biforata, the dorsal valves being also Spirifer-like.
The rather broad and evenly plicated type of this
extremely variable Arctic species is very near to another
North American species of the same genus, H. lesleyi
Walcott, from the Upper Cambrian of Utah (see op. cit.,
text fig. 75). In the natural casts of the interior such as
are commonly found (the interior characters can also
easily be shown artificially by using acid), we notice the
typical characters of the Syntrophiide. As to whether
the interior of this Arctic species corresponds in every
detail to that of the American forms mentioned, I can not
tell, as I have seen no illustrations of the interiors of the
latter.

Besides the Huenella two orthoid brachiopods occur,
one of which may be identical with Hoorthis? melita Hall
and Whitfield, while the other is of the type of Hoorthis
wichitaensis Walcott so far as the dorsal valves are con-
cerned; the ventral valves in the latter, however, do not
have the even, gentle convexity of Hoorthis, but are flat,
with elevated umbones reaching rather far beyond the
hinge-line. In addition, there are fragments of inarticu-
late brachiopods, a Hyolithus, and traces of gastropods
(of the type of Pelagiella pagoda Walcott). Fragments
of trilobites also occur, commonly the central part of
cephalons. At least four species are represented: one is
probably a Ptychoparia, another a Solenopleura of a
rather extraordinary type, with the frontal rim very
broad and very prominent. In addition occur fragments
of a Ptychaspis sp. and of a small /llenus-like form, with
the dorsal furrows widely spaced and developed only in
the posterior part of the head; the occipital furrow is
well developed, differing in this respect from congeneric
forms. ;

From the lighter colored fossiliferous caleareous sand-
stone a very large amount of material was collected,

in the European Arctic Region. 345

sinee the rock is exposed at the shore. The preliminary
study shows, however, that there are only a few species
represented. Exceedingly common is an orthoid that
may be nearly related to Billingsella coloradoensis
(Shumard). Very rarely represented is another species
of orthoid, a reversed type, the general form of which is
rather like that of the reverse shell mentioned from the
southern locality. The surface characters of the former,
however, are different, showing rather prominent, well
rounded striz, of fairly equal strength, set relatively far
apart, and crossed by strongly marked and beautiful con-

TG seale

Fic. 1—Huenella ef. texana Walcott, « 3. From Nova Zembla.
Above: exterior of a ventral valve.
Below: natural casts of interior of a dorsal and a ventral valve.

centric lines of growth. It suggests a Hebertella. Of
inarticulate brachiopods, there are fragments of an
Obolus belonging to the subgenus Westonia, showing the
characteristic transverse parallel lines of ornamentation.
Of trilobites, one species is common, which for the present
I will refer to Anomocarella, since the characters of both
cephalon and pygidium correspond with types that Wal-
cott describes in his paper on the Cambrian faunas of

346 O. Holtedahl—U pper Cambrian Fauna

China (Research in China, 3, Washington, 1913). Yet the
size of the pygidia seems to be somewhat larger in com-
parison with the associated cephalons than is general in
Anomocarella, in this respect resembling Asaphiscus
Meek. Besides these, there are two large fragmented
cephalons that do not seem to differ from Illenus, a type
of trilobite that we do not expect in the Cambrian. A
nearly smooth pygidium with axis only faintly indicated
may belong to the genus Symphysurus Goldfuss, or to
Tsinama Walcott.

As to the age of the fossils from the southern locality,
there seems to be no reason to doubt that the presence of
Huenella indicates Upper Cambrian time. The genus is
especially characteristic of this epoch and is not known to
occur in the Ordovician. The two American forms that
come nearest to the species from Novaya Zemlya both
occur in the Upper Cambrian, and there is nothing in the
character of the rest of the fauna that nullifies such a con-
clusion. That we are dealing with a time very close to
the base of the Ordovician is indicated by the types of
orthoids, especially the reversed type pointing to Ordo-
vician relations. .

In the fossils found in the light colored sandstone, this
faunal relation to post-Cambrian (Ordovician) strata is
still more emphasized by the occurrence of a species of
Illenus. Yet the dominating trilobites are such that we
should not expect the time to be younger than Ozarkian.

Even this preliminary study of the Novaya Zemlya fos-
sils shows with great distinctness that the faunas in their
eeneral expression are highly different from those of the
Upper Cambrian (and basal Ordovician) of the British-
Seandinavian region and of the North American Atlantic
area as well. They are essentially like those from the
Cordilleran and Interior regions of North America, and
from China. It seems evident that in Novaya Zemlya we
are dealing with Upper Cambrian strata and fossils that
belonged to a large, world-wide ocean, compared with
which the North European Upper Cambrian sea with its
relatively poor and monotonous trilobite fauna has a very
local dispersion. The dominance of the ‘‘Pacific’’ realm
of Upper Cambrian time, as compared with the restricted
‘¢ Atlantie’’ one, now becomes still more accentuated, since
the great Arctic region evidently belongs to the former.

in the European Arctic Region. 847

In a short article on Paleogeography, written in Nor-
wegian,' I mentioned as a probable supposition that the
Seandinavian Middle and Upper Cambrian alum shales
with their high content of bituminous and carbonaceous
matter were deposited in a relatively closed basin, with
no open connection into the Arctic Ocean. With the dis-
covery in Novaya Zemlya of the fossils here discussed, I
think this supposition has been considerably strength-

Fic. 2—Map showing distribution of Upper Cambrian seas; _ vertical
lines mean ‘‘ Pacific’’; horizontal ones ‘‘ Atlantic’’ faunal realms. Ameri-
can conditions after Schuchert 1915. Rings mark occurrence of Huevnella.

ened, since we must now assume a land barrier and not an
open connection between the Novaya Zemlya sea and the
Scandinavian one. We do not observe, in the Upper
Cambrian faunas of Scandinavia, any marked difference
when we pass from Scania northward into Jamtland,
which is halfway the total length of the peninsula.

In the map, fig. 2, I have marked the known areas of
Upper Cambrian seas, with an indication of the probable

* Naturen, p. 81, 1919.

Am. Jour. Sci.—FirtTH Series, Vou. III, No. 17.—May, 1922.
25

348 O. Holtedahl—U pper Cambrian Fauna.

trend of the dividing land in the Arctic region as it
appears to me. With what we knew before, it is now
evident that, as regards the distribution of the two pre- -
sumably separated oceanic realms, we are dealing here
with phenomena of a rather general and not a casual char-
acter. I might recall that in another Arctic region just
north of the coast of the European mainland, that is, Bear
Island (between Norway and Spitzbergen), both the
Lower and Middle Ordovician strata have an American
and not a Kuropean fauna.?

For the probable paleogeography of these younger times
I may refer to a recent article in this Journal.?

University of Kristiania, February, 1922.

7See O. Holtedahl, Notes on the Ordovician fossils from Bear Island eol-
lected during the Swedish expeditions of 1898 and 1899, and On the Paleozoic
series of Bear Island, especially on the Heclahook system, both papers
printed in the Norsk geologisk Tidsskrift, 5, 1918-1919.

*O. Holtedahl, Paleogeography and diastrophism in the Atlantic-Arctic
region during Paleozoic time. This Journal (4), 19, 1, 1920. I will again
take the opportunity to correct the bad error which was made during the
printing of this aricle. The maps illustrating Upper Devonian and Upper
Carboniferous time should be exchanged.

A. N. Winchell—Great Dustfall of 1920. 349

Arr. XX XIV.—The Great Dustfall of March 19,1920; by
Atpxanper N. WincHeE 1, Professor of Mineralogy and
Petrology, University of Wisconsin, and Hric R.
Mruuer, Meteorologist, U. 8S. Weather Bureau.

Object of the Investigation.—Since the dustfalls of
March, 1918, that we reported in this Journal! and the
Monthly Weather Review, we have collected the solid
precipitates from all storms, with the object of throwing
light on (a) the probable origin of the dust, (b) the con-
tribution of plant food to the soils of the Kastern States,
and (c) the chemical relationships between American and
HKuropean dustfalls. During the interval of more than
three and a half years there has been only one unusually
heavy deposit of atmospheric dust at Madison, Wis-
consin, namely that of March 19, 1920.

Sources of Material and Information.—The dust was
first noticed by us in the upper layers of a fall of 0.4
inch of snow and sleet, to which it gave a grayish tinge,
on the morning of March 19, while the last of the fall
was still coming down. We immediately collected
samples from measured areas in- various places, in
Madison.

Professor Charles F. Marvin, Chief of the Weather
Bureau, upon our request, very kindly sent a telegraphic
request to the officials in charge of seven offices, to collect
samples of the dust and send them to us. The places
from which these samples were received, and the officials
to whom we are indebted for material, and descriptions
of the precipitate in situ are

Charles City, Iowa. ee. Earcan:
Dubuque, Iowa. J. H. Spencer.
Green Bay, Wis. EF’. W. Conrad.
La Crosse, Wis. EK. C. Thompson.
Ludington, Mich. C. H. Eshleman.
Saginaw, Mich. F. H. Coleman.
St. Paul, Minn. J. N. Ryker.

At Charles City, Dubuque, and La Crosse the dustfall
was noticed by the observers. At Dubuque it was

* Vol. 46, p. 599, 1918, and vol. 47, p. 133, 1919.
* Vol. 46, p. 502, 1918.

350 A. N. Winchell—Great Dustfall of 1920.

observed as a gray coating that stuck to everything after
the snow melted, and was considered the heaviest dustfall
in fifteen years. At Charles City, it was yellow or straw-
colored clay.. At La Crosse, it appeared as a hight brown-
ish layer, first noticed in shoveling the snow off the walks
in the morning. The dust- bearing snow rested upon an
inch of pure white snow. At the other places the dust
did not noticeably discolor the snow.

A postal inquiry was also sent to a hundred observers,
mostly co-operative observers in small towns where the
chance of local soot and dust was small. Sixty-seven
_ replies were received, and the samples of dust accom-
panied reports from the following:

Carlisle, Pa. C. E. Miller.
Louisville, Ky. J. L. Kendall.
Westboro, Mass. Emily W. Newcomb.

At Louisville there was a pronounced brownish haze in
the sky. At Carlisle the dust was first noticed in the
melting of snow from the snow gage for measurement.
At Westboro the dust appeared on the surface of melting
snow. Other observers, who had noticed the deposit, but
did not collect samples reported as follows: C. D. Reed,
Des Moines, Iowa, says ‘‘there was an appreciable deposit
of yellowish clay on the roof of our building during this
storm. The amount was not more than one-third as
great as occurred in the storm of February 13-14, 1919.’’
Mr. Wm. F. Baker, at Decorah, lowa, reported that
‘‘there was a fall of 1.5 inches of snow, the first layer
about .5 inch was pure white, the next half inch light
brown, and the top was a layer of pure white. At Colum-
bus, Ohio, Mr. William H. Alexander reported ‘‘no dust
or mud deposit was noted at this office, but several tele-
phone calls inquiring the cause of the mud deposit came
from the southern section of the city.’’ At Wilmington,
Ohio, Mr. Erskine R. Hayes noted ‘‘ March 19, 11:35 a.m.
Muddy hail fell. When melted in a pan left quite a
deposit of brownish-yellow loam, slightly gritty. Rain
which followed was also very muddy, left quite a lot on
roofs, and spattered sides of buildings.’’ At Raquette
Lake, N. Y. Mr. R. J. Dunning entered: ‘‘March 19: 5
p.m. a round hard snow, almost like hail, 4.3 inches, was
dirty. Hittivcor tine replies were negative, the E50 ters

seas eer

!
’
)

Gre

+
)

A. N. Winchell—Great Dustfall of 1920. 351

not having noticed anything, and the traces having dis-
appeared by the time the inquiry reached them.

The Origin of the Dust.—The meteorological conditions
that produced this dustfall were quite similar to those
of the dustfall of March 8-9, 1918. The dust-bearing
winds accompanied Low V of Chart III, Monthly Weather
Review, March, 1920. The track of this storm is repro-
duced in fig. 1 of this paper. Iixcepting the secondary
center that developed in Arkansas on March 18, 1920, this

Fie. 1.—Dust-bearing storm of March 15-20, 1920, ‘‘17a,’’ ‘‘17p,’’ ete.,
show position of storm center at 8 a. m. and 8 p. m., March 17th, etc.
Shaded areas show total precipitation of rain and melted snow during the
storm, each shade corresponding to a range of half inch in depth of precipi-
tation. Dotted line is hmit of snow on ground 8 p. m., March 15, 1920.

track is practically identical with the track of the storm
that brought the dustfall of March 8-9, 1918 (Compare
fig. 1, Monthly Weather Review, November, 1918, page
002. )

The storm of March 18-19, 1920 was more severe on
the High Plains, and less severe east of the Missouri
valley than the storm of March 8-9, 1918. Throughout

352 A. N. Winchell—Great Dustfall of 1920.

the region between Denver and Cheyenne the 1920 storm
was recorded as the worst March windstorm since 1901.
Barns, windmills, and telephone and telegraph poles were
blown down by hundreds. Plate glass windows were
blown in, winter grains were blasted by the driving sand.
Many fires were started by the wind, Denver reporting
43, the greatest number in one day in the history of the
fire department. The air was so filled with drifting snow
in the mountains, and with drifting dust and sand on the
plains that trains on the Moffat Road and the Colorado
and Southern Railway were halted for hours. The dust
clouds darkened cities in northeastern Colorado in broad
sunlight on March 18, so that artificial light had to be
used.

The dust storm prevailed as far east as St. Joseph,
Kansas City and Little Rock, where it was reported that
the ‘‘air was full of dust, sifting into offices, and covering
books and papers.’’ Haze was noted as far eastward as
Nashville, Tenn., and Columbus, Ohio, where the haziness
prevailed on the forenoon of the 19th so long as the wind
was from the southwest, but disappeared with the change
of the wind. to the northwest.

This storm was competent to cause eolian erosion
throughout its track from the Rocky Mountains eastward.
ibe caused the highest winds of the month of March, 1920
at the following places:

Velocity. Direction. Date.
Denver ao. Agee oul, W 18
Cheyenne mses. Sere 74 WwW 18
Mopekaks:. 2 Bese oe a0 W 18
Tolar sie: 2. ee eee 36 W 18
St OSC pole aay ae ere a2 W 19
iKansas -@itye... 2 SE W 18
Dittle Rock ews fee A9 NW ibs)
Kanoxaviilille: eee oe see 36 SW ae)
Tiromasivallen. sy, sec 24 SW 19
Jacksonville tn es. 49 SW 19
Ashewillé.\cgsseener cee ae ao NW 20
DIS OOI sd 3 aos 8 wis o oe 06 NE 20

The state of the ground as to snow cover as last
reported before the passage of the storm is shown in
fig. 1. All of the central and southern Plains and the
southern Plateau region was exposed.

A. N. Winchell—Great Dustfall of 1920. 353

From the foregoing facts, we infer that the dust was
mainly derived from the region of intensest wind, in
northeastern Colorado and southeastern Wyoming, and
that the supply of soil material in the air was augmented
by smaller contributions from most of the southwestern
states.

Distribution of the Dustfall—The samples of dust that
have been sent us, and the reports of observation of dust,
are from far too few places to enable us to attempt to
map the distribution of dust, as was done by Hellmann
and Meinardus? and by Mill and Lempfert* for the Kuro-
pean dustfalls of 1901 and 1905.

The rain area of the storm of March 15-21, 1920, shown
in fig. 1, indicates that most of the eastern half of the
United States was a region of possible precipitation of
dust by washing down with rain or snow. It is interest-
ing to note that the region of intensest blowing between
the Rocky Mountains and the Missouri valley remained
unwetted during the storm.

Chemical Composition.—In the chemical study of the
samples of dust we have had the help of Dr. E. J. Graul,
of the Department of Soils, University of Wisconsin, who
determined the nitrogen, phosphorus pentoxide, the alka-
lies, and the water lost at 105°.

A grant of funds for additional chemical work was
made from the research fund by the Regents of the
University of Wisconsin. The analyses under the grant
were made by Mr. Martin Tosterud, working under the
direction of Professor A. R. Whitson, in the laboratories
of the Department of Soils of the University of Wis-
consin. Mr. Tosterud determined silica, alumina, iron,
magnesia, lime, titanic acid, and water lost on ignition
in three samples. By the additional determination of
the total loss on ignition we are enabled to present the
following relatively complete analyses of three samples.

In these analyses it was not practicable to determine
the state of oxidation of the iron; total iron was deter-
mined and calculated to be ferrous oxide since the dust
in mass is gray and not colored red or yellow by ferric
iron. But it is recognized that some ferric iron is
present, since its colors are distinct under the microscope.

* Der grosse Staubfall, Abh. k. Preuss. Meteor]. Inst., II, 1901.

*The great dustfall of February, 1903, Quart. Jour. Roy. Met. Soc., 30,
p- 57, 1904.

854 A. N. Winchell—Great Dustfall of 1920.

Taste I.—Analyses of three samples of dustfall at Madison,
March 19, 1920.

(Dried at 105° C.)

I II III Average
TOE: oy sien eegert te 68.61 66.382 66.66 67.20
A Oia eee err e nae 13.81 13.44 13.89 137
GOS ts Shel acne 2.24 Leal () 2.17 rl
OY Bah earae a ees he em oO 42 Al a0
MPO Oss Cr Leh eee 1.63 1.71 1.94 1.76
Ca Pie coe ig ee 1.81 1.83 1.59 1.74
INasO\. tse eee 1.64 3.04 1.64? 2.11
HGH @ Dite teet ieeeneel Coes Deo 2.11 2.56 2.30
H. >O0-+ (above 105°C) 2.55 3.47 3.63 3.22
UI ORS pest cos ee eee 533 De 52 1s
IP 5 OPS ck rine gree 14 16 16 A
IN Sco ae ters, Ce econ sahh eel bait oo
JSAUTRIOINS oo ee Ae Doe, 5.66 5.68 3.02

101.40 101.20 — 101.22-- 101228
H,O— (below 105°C) 3.28 2.91 Dg Dold

a Metallic iron determined and calculated to FeO.

b Not determined on account of lack of material; assumed to be the same
as in sample I.

¢ Includes organic matter and CO., but not H,O nor N.

I. Dustfall collected by E. R. Miller from one square meter
of surface at 2125 Van Hise Ave., Madison, Wis., 19 March, 1920.

II. Dustfall collected by W. s. Fusch ‘from. one square yard
of surface of ice about 1,000 feet north of Science Hall on Lake
Mendota at Madison, Wis., 19 March, 1920.

III. Dustfall collected.-by A. N. Winchell from one square
yard of porch roof at 200 Prospect Ave., a Wis., 19
March, 1920.

It was hoped that special tests could be made to measure
the tenor of water soluble salts, of organic matter, and of
carbon dioxide,® but scarcity of material prevented, as
it likewise prevented the determination of soda in sample
number III. It is our opinion that sample number II
is abnormally high in soda for some unknown reason;
therefore, we have assumed for purposes of mineral e¢al-
culations that sample III has the same tenor of soda
found in sample I.

*Sample I showed no visible effervescence when treated with warm HCl,

but carbonates can be recognized in small amount in all the Madison samples
under microscope.

A. N. Winchell—Great Dustfall of 1920: 355

It is remarkable that no chemical analyses of samples
of American dustfalls are onrecord. Furthermore, anal-
yses of foreign dustfalls are very uncommon and several
of those published are too incomplete to be satisfactory
for comparisons or for calculations of mineral compo-
sition. All the other chemical analyses of foreign dust-
falls are assembled in the following table in which the
average of the analyses of the Madison dustfall is
included for convenience.

TasLE I].—Analyses of foreign dustfalls compared with the
average of the Madison analyses :—

I II ee IV V VI

Swng. 6.20 53.68 45.94 45.40 41.43 36.32
PaO rn sO 13.71: 18.44 18.35 19.97 10.38 16.35
BesO a. :.. 6.54 65% 7.03 9.19 6.08
ey ee D7

MeO... 1716 ie 1.86 ali 92 Tigedt
ee es 174 95 8.64 6.50 14.10 6.24
NasOs 26:24 LAE 116 Deol: 1.66 2.59
Pe eos 20) 2.58 2.30 2.07 1.58 Pee,
fe Oe =. 38.22

COne Sas. 6.10 3.46 8.45 3.68
Os: he

mes ES Diy

MuQ .... 38

ee 9 16 16

Ignition... 5.62 14.60 6.73 oad Ey) 0.14 13.44

Total ...101.28 99.98 ~~ 100200 99.73 99.00 100.00

I. Average of three analyses of dustfall at Madison, 19
March, 1920.

II. Analysis after air drying of dustfall at Otakaia, New
Zealand, Nov. 14, 1902, which came about 1,500 miles from Aus-
tralia. P. Marshall: Nature, 68, p. 223, 1908.

Iit. Dust m ‘‘red rain’’ fall at Lamberhurst, England, 22
Feb., 1903. 2.19% of organic carbon included in total. T. E.
Thorpe, Nature, 68, p. 54, 1903.

IV. Dustfall at Naples, Italy, March 10, 1901. P. Palmeri:
Rend. Accad. sci. fis. Naples, (3), 7, p. 157, 1901.

V. Dustfall at Naples, Italy, 25 February, 1879. Analysis by
Scacchi, quoted by P. Palmeri, loc. cit. p.161. 4.16% of organic
material and 1, 39% ‘logs’? included in total.

VI. Analysis (after air drying) of dustfall at Taormina,
Sicily, 19 March, 1901. The fall amounted to 514 tons per

356 A. N. Winchell—Great Dustfall of 1920.

square mile. 1. E. Thorpe, Nature, 68, p. 222, 1903. 9.89% of
organic carbon and 0.32% of CoO included in total.

The dustfall at Madison is the most siliceous of these
analyses, but an incomplete analysis of a dust which fell
in Tunis® in 1901 revealed 70.95% of silica, and the tenor
of silica in siroccan dust from the Sahara reaches a
maximum of 73.45%,’ calculated on a water-free basis.

The dust from the Sahara is always red and produces
the ‘‘red rain’’ or ‘‘rain of blood’’ sometimes noted in
Europe. This is due to the thoroughly oxidized state of
the iron which is unlike the condition found in the Madi-
son dust. In the percentage of alkalies and alumina the
latter does not differ markedly from the samples from
the Sahara, but it contains far less lime and very little
carbonic acid as carbonate, in contrast with the Saharan
dust. In these particulars the Madison dust closely
resembles that fallen at Otakaia, New Zealand, which
was derived from the continent of Australia.

In some respects the Madison dustfall resembles, in
chemical composition, the loess of the Mississippi valley
more closely than it does the dust from the African
desert. This is shown in the following table.

Taste III. Chemical composition of Mississippi valley loess
compared with the average composition of the
Madison dustfall.

ds 2 3 +; 5 6
S103 2s. 1.64/04 67.20 (gach 70.86 72.68 74.46
AI3OR = 2a O64: acme. 14.25 8.91 12.03 12.26

HesO.ni2 ee 61 4.02 2.97 3.53 3.25
HeQ ete ol OAR te, See AQ” “296 12
MgO .... 3.69 1.76 a2 3.12 1a 1.12
CaO. 2 ose ee ie 1.53 4.13 1.59 1.69
NasOnc0) 21535 211 1.09 1.69 1.68 1.43
Ke Onc 206 2.30 2.03 1.18 oie 1.83
EL Ones 29.05 3.29 2.48 1.10 2.50 2.70
CO 6.31 oe aa 4.70 39 AQ
MOs 2. ek) 52 cares 59 72 14
RO 06 De, sae 40 23 09
MinOr 105 38 Aube 28 06 02

Total ....100.06 101.28 100.33 99.98 100.22 99.83

*H. Bertainchand, Compt. Rend., 132, p. 1153, 1901. According to
L. Cayeux this dust contained quartz, staurolite, tourmaline, rutile, zircon,
magnetite, yellow phosphate, and diatoms.

TE. HE. Free, U. S. Bur. Soils, Bull. 68, p. 95, 1911.

A. N. Winchell—Great Dustfall of 1920. 3) (I

1. Loess from a stratum overlying residual clay, 350 feet
above the Mississippi River, near Galena, Illinois. Chamberlin
and Salisbury, 6th Ann. Rep. U. S. Geol. Survey, 1885, p. 282.
Dried at 100° C. 0.13% of organic carbon, 0.11% SO, and
0.07% Cl included in total.

2. Average composition of Madison dustfall, March 19, 1920.
5.62% ignition loss and 0.39% N. included in total.

3. Loess from Kansas City, Mo. W. E. McCourt, Missouri
Bureau Geol. & Mines, vol. 14, p. 94, 1917. 3.50% ignition loss
included in total.

4. Loess from depth of 8 feet at brickyard at Mt. Vernon, Ia.
N. Knight, Am. Geol. 29, p. 189, 1902.

5. Loess from 300 feet above the Mississippi River, 314 miles
north of Dubuque, Ia. Chamberlin and Salisbury: 6th Ann.
Rep. U.S. Geol. Survey, 1885, p. 282. Dried at 100° C. 0.51%
SO,, 0.09% organic carbon and 0.01% Cl included in total.

6. Loess from Kansas City, Mo.- Chamberlin and Salisbury:
6th Ann. Rep. U.S. Geol. Survey, 1885, p. 282. Dried at 100° C.
0.12% organic carbon, 0.06% SO, and 0.05% Cl ineluded in
total.

Except as to the state of oxidation of the iron the
Madison dust is closely similar to the loess from Kansas
City studied by McCourt; it is also much like the loess
from Dubuque and that from Kansas City examined by
Chamberlin and Salisbury; it differs from the loess of
Galena, Illinois, and that from Mt. Vernon, Iowa, in the
scarcity of carbonates of calcium and magnesium.

Mineral Composition.—The components of the dust are
so extremely fine grained that it is quite difficult to make
satisfactory determinations of the mineral constituents
with the microscope and it seems to be wholly imprac-
ticable to attempt any quantitative determinations micro-
scopically because so large a proportion of the material
is too small for identification. However, samples from
Madison contain abundant quartz grains in tiny angular
forms and no other recognizable mineral in any important
amount. Feldspar may be present, but if so twinning
crossing the tiny particles is rare; biotite flakes are very
uncommon; one shows an optic angle of about 40° (2E)
and pleochroic colors with Z—brown and Y = yellow.
Very rare fragments of calcite are also present. Still
more rarely a fragment of microcline, a light-colored
amphibole and a mineral resembling staurolite may be
detected. In spite of the gray color in mass, the red and
yellow colors of hematite and limonite are common under

358 A. N. Winchell-—Great Dustfall of 1920.

the microscope. Most of the material is cloudy and
isotropic.

It is possible to calculate the approximate mineral com-
position from the gross chemical composition by making
certain assumptions similar to those made in calculating
the mineral composition of igneous rocks. Such caleula-
tions are greatly facilitated by the use of the circular slide
rule for minerals devised by W. J. Mead.® In this way
we have computed the mineral compositions of the Mad-
ison dustfall as well as that of HKuropean dustfalls and
of Mississippi Valley loess. In these calculations, it is
assumed that all the soda is in albite feldspar, all the
potassium is in orthoclase feldspar, all the phosphoric
acid is in apatite, all the titanic acid is in ilmenite, all the
magnesia is in chlorite (if there is sufficient aluamina—
otherwise in enstatite), all the lime remaining after form-
ing apatite and calcite is in anorthite, all the alumina
remaining after forming feldspars and chlorite is in
kaolinite and all the silica remaining after forming these
silicates is in quartz. It must, of course, be admitted that
these assumptions are not all true, but they are approxi-
mations and represent possible, even if not actual, com-
binations of the oxides into minerals. They furnish a
useful means of comparing analyses, especially when the
underlying assumptions are controlled as far as possible
by microscopic study of the material. The results of
these computations are given in the following table so far
as they relate to dustfalls.

TaBLE IV. Calculated mineral composition of Madison and
European dustfalls.

1 2 3 4. 5 6 7 8 9

Quartz .... 39.91 32.82 36.87 36.47 18.36 13.34 5.75 17.32 ....
Albite .... 13.87 -25.80 13.87 17.89 14.21 9.82 22.14 14°04 52053365

Orthoclase. 13.16. 12.52 15.21 13.69 15.32 13.64 12.28 9.38 16.15
Anorthite . 8.11 8.06 686 7.64 4.73 438 4.44 16.17 7.75

Kaolinite . 10.09 3.59 10:09 .. 7.94 26.20 28:85 25.22 ~ Pe eetoeo
Chiorite .. . 8.92. 9:04.2°.9:83 49.26" 4:20\ “o.3. °S: 66.5 oe
Enstatite .. Pe co Eyelets PY Penis’. ween ee tae en & role aes
Caleate: ss. och Re RD RD GIRLS Gan 7) So) ence ee
Apatite ... 30 Ot ot ESOT 6 aie. Spee AG» isn Chee
AnH YOrite o5 8 sec so os wince Rene eng s) Diatlc joes NMeRe ee iw at grace bas Oe ge
Hematite 2 na cee O04 ESO cot, Meied Uo cunt) I ener
Iimenite 77) 2012 = 1208 a) a eS OBE

Water, etc. 6.00 7.99 7.13 7.03 1042 441 441 11.43 20.89

Total ..... 101.40 101.20 101.22 101.28 99.98 100.00 100.57 .99.00 100.65
® Economic Geology, 7, p. 136, 1912.

A. N. Winchell—Great Dustfall of 1920. 359

1. Dustfall collected at 2125 Van Hise Ave., Madison, Wis.,
19 March, 1920.

2. Dustfall collected from ice on Lake Mendota, Madison,
Wis., 19 March, 1920.

3. Dustfall collected at 200 Prospect Ave., Madison, Wis., 19
March, 1920.

4, Average dustfall at Madison, Wis., 19 March, 1920.

5. Dustfall at Otakaia, New Zealand, 14 Nov., 1902, P. Mar-
shall: Nature, 68, p. 223, 1903.

6. Dust in ‘‘red rain’’ at Lamberhurst, Eng., 22 Feb., 1903.
T. EK. Thorpe: Nature, 68, p. 54, 1903.

q —Dustiall at Naples, Italy, 10 March, 1901. P. Palmeri:
Rend. Acead. sci. fis. Naples. (3), 7, p. 157, 1901.

8. Dustfall at Naples, Italy, 25 Feb., 1879. P. Palmeri: loc.
eit. p. Ol.

See Dusiiall at, Vaopmina, sicily, 19) March, 1901... To
Thorpe: Nature, 68, p. 222, 1903.

This table shows even better than the chemical analyses
that the Madison dustfall is exceptionally rich in quartz;
it contains about the same amount of total feldspar as
in the dustfalls at Otakaia and at Naples; its content of
kaolinite is much lower than in foreign dustfalls (except
one at Naples) and the tenor of calcite is far lower
than in the European dustfalls. The Madison dust con-
tains more chlorite and less hematite than the others,
but this is at least in part due to our assumption regard-
ing the state of oxidation of the iron.

In order to show how much similarity in mineral com-
position there is between the Madison dustfall and loess
of the Mississippi Valley computations of the latter made
in the same way as those of the dust are presented in
the following table.

This tabulation shows that while the Madison dustfall
is quite unlike foreign dustfalls as shown by its tenor of
quartz, of which it contains at least twice as much as the
latter, it closely resembles the loess of the Mississippi
Valley differing no more from some samples of loess than
these differ from other samples of loess. It contains
decidedly more feldspar and less kaolinite than the loess,
but the high percentage of water not included in the eal-
culated minerals suggests that these differences are more
apparent than real; that is, the feldspar molecule is
probably already hydrated though the alkalies have not
been removed as thoroughly as in the loess. The only

360 A. N. Wanchell—Great Dustfall of 1920.

TaBLE V. Calculated mineral composition of Mississippi Valley
loess compared with that of the Madison dustfail.

il 2 3), 4 5 6
Girartz- see 0 cc creenee 36.47 39.66 43.52 45.56 48.89 49.50
INMDHLGL ne een 17.89 11.42 9,24 14.22 14.30 12.10
Orthoclases eee 13.69 12.20 12.03 12.62 6.98 10.86
Amorthite- ioc. tenn eae 7.64 oie 7.60 2.19 ; bag 4.74 —
Feaolinites crests ee 7.94 15.62 17.15 14.83 12.29 14.12
Chiorites 2544 ae ae 9.26 Sete 3.65 1.55 ee: 3.10
iBnistatibe 2 ct saa cee Bey 3.24 Bas eS Sally BS, &
Calcite’ Son. & sees Meee artes 9539 Bee 88 6.43 Tot
Ma oNeSILG ss s5... see sate 4.57 piaere rene 3.56 Riggane
Apatites te -.io ea eee 35 14 Ae 53 94 al
Gay SUIT soa onan eee awa 22 eee 1.10 sat Rei
Dimenttew sete ses meres 1.01 76 Bre 1.37 1.12 26
Hematite iin. aes.cae. eas 2.17 4.02 2.01 2.97 3.25
Madonetitermi asi. Sens 64 ATA. eg Pea 3 ere
Water, Cte.7:8e84 a. 8 7.03 Flere oldies .10 5 tells 58
Totals. aie es oes 101.28 100.03 100.33 100.22 100.65 99.83

1. Average Madison dustfall of March 19, 1920.

2. Loess, Galena, Illinois. Chamberlin and Salisbury: 6th
Ann. Rep. U. S. Geol. Survey, 1885, p. 282.

3. Loess, Kansas City, Mo. W. E. McCourt: Missouri Bureau
Geol. & Mines, 14, p. 94, 1917.

4. Loess, Dubuque, Ia. Chamberlin and Salisbury: loe. eit.

5. Loess, Mt. Vernon, Ia. N. Knight: Am. Geol., 29, 1902,
pr 89,1902)

6. Loess, Kansas City, Mo. Chamberlin and Salisbury: loc.
Cit.

other difference between the dustfall and the loess that
requires comment is the scarcity of carbonate in the
former. But this scarcity is matched by the condition
found by McCourt in the loess from Kansas City, and
the other sample from that locality as well as the loess
from Dubuque contain very little carbonate. It is note-
worthy that four samples of Kuropean dustfalls are rich
in carbonate as shown by numbers 6-9 in table IV. It
seems very probable that the amount of carbonate in a
dustfall would vary through wide limits depending largely
upon the type of chief country rocks in the area of origin
of the dust.

Accordingly, we find no fundamental distinctions in
chemical or mineral composition between the Madison
dustfall and the Mississippi loess; on the contrary, the
two are more alike than Madison dust and foreign dust.
The chief differences which exist are such as would be
produced by weathering of the dust.

A. N. Winchell—Great Dustfall of 1920. 361

Physical Composition.—Dustfalls vary considerably in
the average size of the component particles as well as in
the range of size and the abundance of particles of various
sizes. The dustfall at Madison in 1920 may well be
compared in this respect with the dustfall of 1918 as
shown in the following table.

TaBLE VI.—Size of constituents of dustfalls.

1 2 3 +

Clay, less than .005mm..... 11.15 27.49 29.01 23.32
Bene silt.005: to 0102... os. 22.01 IIA

Medium silt, .010 to .025.... 56.17 66.86 44.09 70.11
Woarse silt 025) to 050. .055. joy) 11.35

Merwe sand: .0o 40 10..2.- 1.22 4.78 5.04 9.05
Pome scanc, 10 to 250i... 8... 1.04 Roll O7 ale
Meciuam-sand, 25. to 0:.... 0.58 03 05 O04
Soarse sand, .00' to 1.00...... 0.29 .O1— 03 03
Fine gravel, 1.00 to 2.00.... 1.08 00 00 OO

ee

99.53 99.98 98.91 99.31

1. Dustfall at Madison, Wis., 9 March, 1918, this Journal, 46,
p. 602-1918.

2. Sample 1 of dustfall at Madison, Wis., 19 March, 1920.
Mechanical analysis of this and two next were made by Hazel
Hankinson in Dept. Soils, Univ. Wis. .

3. Sample 2 of dustfall at Madison, Wis., 19 March, 1920.

4. Sample 3 of dustfall at Madison, Wis., 19 March, 1920.

It is evident that the dustfall of 1920 is composed of
even finer particles than that of 1918. This is probably
due to the fact that the velocity of the wind east of the
Missouri River was less in the later storm, though the
recorded velocity west of that river was greater. We
have no explanation of the fact that the 1920 dustfall
contains about four times as great a tenor of very fine
sand as that of 1918; the ‘‘fine gravel’’ of the latter
consists largely of fragments of vegetation. In all these
samples nearly 95 per cent of the material is finer than
0.05 millimeter. Unfortunately we know of no mechan-
ical analyses of Kuropean dustfalls with which these may
be compared. It is reported merely that the commonest
size of the particles in the great dustfall' of March, 1901

* Hellmann and Meinardus, loc. cit. p. 63-65.

362 A. N. Winchell—Great Dustfall of 1920.

was between .02 and .001mm.; that is, most of the mate-
rial would be classed as silt or clay. The largest parti-
cles carried long distances by the wind are thought to
be only .07 to .08 mm.

There are many mechanical analyses of loess from this
country with which these analyses. of dustfalls may be
compared, as illustrated in the following table.

TaBLE VII.—Size of constituents of dustfalls compared
with loess.
1 2 3 4 3) 6 at 8 9 10
<.005 mm. J1.15° 10.7 25.57 24:5. 26:7 28.9 26:4 320.51 7ab aoe

.005— .010 22.01 eo

.010— .025 56.17+85.0 44.09 | 66.5 64.8 63.9 56.4 63.4 67.5 42.3
.025— .050 TEND) 11.35

.05 — .10 1.22 3.2 5.04. 5.85 .3.6. 4.6 14.5. 14 adhiaas
10 — .25 1.04 0.2. 0.87 + 1.0 «1.8». 41.35.04 = dele eee
25 — .00 0.58 O12 0.05 04 18° :.0.4:< 0.5. (O!G Ores
00 —1.00 0.29. .0.0 . 0.03 67.2 1.4" 10.6% 0.055 OLS eee
1.00 —2.00 1.08- ~0.0 0:00: 0.6. -0:0—— 0:2 =0°05~ 030 ae

—————s —_ — —————_ ————_ — —____. ——_ ——

Mbp ge he 98.91 100:0° 99.9 99.9 99.8. 99:9 100n eee

1. Dustfall at Madison, 1918.

2. Loess, 6 ft. below surface at Edwards, Miss. E. W. Shaw: U.S. Geol.
Survey, Prof. Pp. 108, 135, 1918.

3. Dustfall at Madison, 1920.

4. Loess, 3 feet below surface, Muscatine Co., Ia. Field Oper. U. S.
Bur. Soils, 1914, p. 1848.

5. Loess, 3 feet below surface, Harrison Co., Mo. Field Oper. U.S. Bur.
Soils, 1914, p. 1960.

6. Loess, 3 feet below surface, Ringgold Co., Ia. Field Oper. U. S. Bur.
Soils, 1916, p. 1918.

7. Loess, 3 feet below surface, Callaway Co., Mo. Field Oper. U. S. Bur.
Soils, 1916, p. 1933.

8. Loess, 3 feet below surface, Grundy Co., Mo. Field Oper. U. S. Bur.
Soils, 1914, p. 1991. :

9. Loess, 6 feet below surface at Weeping Water, Neb. Alway & Rost:
Soil Science I, 1916, p. 407.

10. Loess, 3 feet below surface at North Platte, Neb. W. W. Burr:
Res. Bull. 5, Agr. Exp. Sta. Neb., 1914, p. 12. .

The mechanical analyses of loess show that it varies
considerably in percentage distribution of sizes present.
The analysis of the loess from Edwards, Miss. (No. 2)
is very similar to the Madison dustfall of 1918, while
analyses 4, 5, and 6 are closely like the dustfall of 1920.
These have been selected to show that loess may be of
the same types as these dustfalls, but it is also commonly
found of other types. Analyses 7, 8, and 9 illustrate loess
differing distinctly, but not greatly, from the dustfall of

A. N. Winchell—Great Dustfali of 1920. 363

1920, while analysis 10 shows that loess may even differ
very decidedly from the dustfall.

This table does not show that there is any close rela-
tionship between dustfall and loess; it merely shows
that such a relationship can not be denied on the ground
of differences in mechanical composition.

Organic Constituents.—In addition to the mineral com-
ponents fragments of vegetation are visible; these
include spores and shreds, but for more accurate data
on the subject the samples were referred to Professor
R. H. Denniston of the Department of Botany of the
University of Wisconsin who very kindly supplied the
following information: Spores of fungus are present
which are probably to be referred to Alternaria; the
hypha of a fungus is also found, as well as trichomes,
starch grains, grass cells, and bits of charcoal.

Finally, diatoms are found in the dust which seem to
be of three types including two like those found in the
dustfall of 1918. For the examination of these diatoms
samples of the dust were sent to Dr. Albert Mann, Plant
Morphologist of the United States Department of Agri-
culture, who kindly reported as follows:

‘‘T have examined the dust brought down by a snow-
storm at Madison on March 19, 1920, and find that the
diatoms contained are in general like those in the snow-
dust you collected two years ago. But there are some
differences. In the 1918 sample the most abundant
species was Nitschia (Hantschia) amphioxus (K.) W.S.,
much outnumbering Navicula borealis (H.) Ik. In this
last sample the latter is more abundant, perhaps 3 to 1.
As I then stated, both these are characteristic of cool
sphagnum bogs and the moss on shaded tree trunks.’’

‘‘T have found two other species in this material :—
Navicula pupula K. (rather frequent) and Cymbella tur-
gidula Grun (very searce). The former is often found
in cool and damp garden soil and on the under side of
leaves in damp woods. The latter is widely distributed
in cool fresh waters.’’

Quantity of Dust Transported—In the storm which
brought a heavy dustfall to Madison in 1918, we estimated
from meager data that at least one million tons of dust
were transported long distances. In 1920 the amount of
dust deposited at several places in Madison and at some
other points was determined with the following results.

Am. Jour. Sci1.—Firta Sreries, Vou. III, No. 17.—May, 1922.
26

364. A. N. Winchell—Great Dustfall of 1920.

TasBLteE VIII.—Quantity of dust in storm of March, 1920.

Weight of dust in Weight of dust in
Sample Grams per square meter. Short tons per square mile.
GCAO ING ed eos SS ose eel | 22.4
Madison Now? G45. 10.30 29.3
Madison INO. oa 9.56 ie
asCrosse Wiss oe 9.03 14.3
Dubuque, Howat oe 4.53 12.9
Charles: @ityclas sea 8.34 23.7
Carlisle, Pay a0 hyn ede cce 4.61 13.1

About twenty more samples of this dustfall from vari-
ous parts of the country were sent to us, but the preceding
table includes all those which are sufficiently pure to make
exact records of value. The amounts may be compared
with the dustfalls at other times and places as follows:

TaBLE 1X.—Quantity of dust im other storms.

Weight of dust in Weight of dust in
Places Grams per square meter. Short tons per square mile.
Madison, 1918 ........... Sag 13.6
Naples, 290M. oe ware 11. 31.3
Gérz, Austria, 1901 ...... 112 31.8
Schemnitz, Hungary, 1901 1.9 0.4
lia mal uncove lO Oil age ees eee 1.67 4.7
Taormina, Sicily: 9 Ol eae ed Lol

Except for Madison, the data are from Hellmann and
Meinardus, loe. cit.

It is evident that the quantity of dust in the storm of
1920 is entirely comparable with that of the Kuropean
fall of 1901 in all cases measured. The area of the latter
was about 160,000 square miles while the area covered
by the American dustfall of 1920 was at least as great
and probably several times greater. The total dustfall
in Europe in 1901 was at least 1,782,200 metric tons or
1,964,000 short tons; so far as the evidence goes it indi-
cates that the American dustfall of 1920 involved at least
as great a total, and probably several times as great a
total amount of material transported.

Madison, Wisconsin, January 1, 1922.

E. L. Troxell—Helaletes Redefined. 365

Art. XXXV.—Helaletes Redefined; by Epwarp L.
7 SV EROMET TS

[Contributions from, the Othniel Charles Marsh Publication Fund, Pea-
body Museum, Yale University, New Haven, Conn. |

The genus Helaletes Marsh constitutes a group of the
smallest of the tapiroids. For two reasons I am
prompted to redescribe this important genus: first,
because it is very much misunderstood, and has no pub-
lished drawings; and second, because some new features
have come to my attention which Marsh himself had
not noted.

The species Lophiodon nanus Marsh was the first one
described; the holotype consists of both maxillaries.
Helaletes boops, however, is the genoholotype and is
based on the greater part of the skull, jaws, and skeleton
of one individual. Whether or not other existing species
should be assigned to this genus is uncertain; two have ©
been referred to it, perhaps with doubtful justification.
These are Dilophodon minusculus Seott and Desmato-
thervum guyots Scott.t

Dilophodon minusculus Scott is shown to have no pos-
terior heel on M, and is therefore quite different from H.
boops. Only its small size and the absence of P, separate
it from Hyrachyus, but these may be sufficient for a sub-
genus at least.

Desmatotherium guyoti, which has been referred to
Helaletes by authors, differs from H. bodps in its much
greater size, in the relatively greater M*? and in the dis-
proportion between the large molars and the premolars.
On the other hand, the partial separation of the internal
cusps on the third and fourth premolars and the diastema
in front of C', likewise the general resemblance in the
form of the molars, indicate relationship, but Desmato-
therium should retain at least subgeneric distinction.

Hyrachyus nanus Leidy is frequently referred to Hela-
letes by writers who seem to be unaware of the species
Helaletes nanus (Marsh) which precedes. Scott has
referred Leidy’s species to his genus Dilophodon because
of the bilobed M, and the absence of P,; if this is a cor-

*W. B. Scott, Contrib. from EH. M. Mus. Geol. & Arch., Princeton Coll.,
Bull. 3, 46-53, pl. 8, 1883.

366 E. L. Troxell—Helaletes Redefined.

rect reference it saves the species. Helaletes nanus
Leidy does not exist and it is doubtful whether he
intended to create a new species in his original descrip-
tion, for he mentioned Lophiodon nanus Marsh at that
time. )

Helaletes nanus (Marsh).
(Fria. 1.)

Holotype, Cat. No. 11080, Y. P. M. Hocene (Bridger), Grizzly Buttes,
near Fort Bridger, Wyoming.

Lophiodon nanus was one of the first three lophiodonts
described by Marsh; it is based on the two weathered
maxillaries, the right one of which bears all the cheek
teeth except P!, whose root only remains (see Fig. 1).
Following is the original description in part :?

‘‘Mhe molars differ espécially from those of the two preceding
species [Hyrachyus bardianus, H. affinis], in having a much
shallower valley between the two transverse ridges, and in hay-
ing a strong basal ridge, or shelf, at the external posterior cor-
ner of the crown. The enamel of the whole series is very
smooth. The species was probably about two-thirds the size of
L. modestus.

Pres ul.

Soo

ON
A)

11080 TYPE Y. P. M.

Fic. 1.—Helaletes nanus (Marsh). Holotype. First described by Pro-
fessor Marsh over fifty years ago. Note the double internal:cusp of the
premolars. Nat. size. .

Measurements.
Leneth of upper jaw, containing seven pos-

CeViOr: Ceet Shae eT ee 26.0 lines [55.0 mm. |
Length of same, with three last molars.... 13.7 lines [29.1 mm. |
Antero-posterior diameter of last upper

males Nah eek ete ekg aerate ree 5.0 lines [10.6 mm. ]
Transverse diameter of same’ ./..:....... 5.25 lines [11.1 mm. |

‘“The remains now known to represent this species were dis-
covered by C. W. Betts, H. B. Sargent, and the writer, in the Ter-
tiary strata at Grizzly Buttes, near Fort Bridger.’’

2 Marsh, O: Cl; This Journal (3)i2) a7, lee

E. L. Troxell—Helaletes Redefined. 367

Helaletes nanus is notable for the rounded molars, the
very prominent paracone and receded metacone, the small
size, posteriorly, of the ectoloph, and especially for the
double internal cusps on the premolars which are most
separated on P?. P? is wider than it is long (7 X 6.3
mm.); there is a postero-exterior cingulum on each
molar, and a conspicuous protocone (antero-exterior) on
the upper premolars.

In this species the maxillary presents a smooth border
eurved strongly inward above the anterior premolars,
indicating a facial pit of a shghtly different character
from that of H. boops, described later.

Helaletes boops Marsh.
(Files. 2-3.)

Holotype, Cat. No. 11807, Y. P. M. Middle Hocene (Bridger), Grizzly
Buttes, Wyoming.

This genus is readily distinguished from Hyrachyus
Leidy, as the latter is generally known, by the small
size, the double internal cusps on the premolars, the long
diastema in front of the upper canine, the compressed
short incisors, the distinctive form of the rounded molars,
the great antero-posterior diameter of M?*, the presence

IG. oe

11807 TYPE AUP MI:

Fic. 2.—Helaletes bodps Marsh. Holotype. Showing the upper teeth
complete, and the last premolar and three molars of the lower dentition.
Nat. size.

of a heel on M;, the absence of connecting ridges between
the crests of the lower molars, and finally by the presence
of a great facial pit as shown by the smooth edge of the
maxillary above the diastema and the premolars.

368 E. L. Troxell—Helaletes Redefined.

As contrasted with Helaletes nanus, H. bodps, the geno-
holotype, lacks distinctly separated cones on the inner
side of the premolars, while the two outer cones are
nearly equal in size and prominence. The first premolar
is very small. In this form the side of the maxillary
rises vertically above the premolars and ends in a sharp
edge to form the antorbital depression.

This last feature is one of great interest and heralds
a peculiarity well known in fossil horses of later periods.
The presence of a facial depression, so far as I know, has
never been noted-in an Hocene form; it was obscured
in the type of Helaletes, because the skull was so crushed
and so greatly distorted.

iGerae

(ROW. 11807 TPMPE V2 Pave

ey

-—2s
-

-—~
eo wee ee
eer

ao
-

7

JS

Oca

Fi¢.- 3.—Helaletes boops Marsh. Holotype. Skull restored from its
previous crushed condition. It now shows a rising sagittal crest, long
nasals, long diastemata in front of and behind the strong canine, the fora-
men over P*, and an opening in the face over the maxillary. It is distinctly

tapiroid. >< 2/3.

Preparation of the Skull.—To develop this feature and
others as well, it seemed eminently worth while to risk
dismembering the rare skull to enable it to be restored to
a form approximately normal. This was done and the
result demonstrates these important points: the probable
slope from the occiput to the frontals, the great width

E. L. Troxell—Helaletes Redefined. 369

over the orbits, the position of the orbits far forward,
the location of the antorbital foramen over P*, the great
length of the nasal bones, the narrowness of the maxil-
laries over the premolars, and the extent of the facial
vacuity already mentioned.

Before the separate pieces were removed from the
original crushed skull, a cast was taken of the whole;
likewise a chart was made, and on both, as a double pre-
caution, the pieces were marked by numbers correspond-
ing to those on the skull itself. With this provision
against confusing the separated parts, the skull was then
almost completely dismembered; the small fragments
were lifted from their abnormal position and restored to
their right places by fitting the broken edges together. In
most instances the original contact was found, in a few
cases the matching is conjectural and only approximately
correct, but, considering the number of parts preserved,
i.e., the incompleteness of the skull, the results of the
operation are quite satisfactory.

The elongated nasals are distinctly unlike those of the
modern tapir, but the facial depression? may well have
been the beginning of the receding nasal aperture in the
tapirs and seems to link more closely these two races.
In addition to the (1) large facial depression, one should
note these features which are distinctly tapiroid: (2) the
rising sagittal crest and the depression in the forehead,
which seems to be clearly indicated by the curvature of
the bones involved; (38) the position of the infra-orbital
foramen over the fourth premolar and near the orbit;
(4) the very narrow maxillary above the premolars; and
especially (5) the form of the teeth. The skull of the
young tapir lends itself particularly well to the gen-
eral comparison, perhaps because it is more primitive
than the adult and assumes more nearly the char-
acters of the ancestor of its race. On a young tapir
skull one can see the smooth upper edge of the maxillary
bone sending a narrow projection backward and upward,

* Its meaning is discussed in connection with the horses by W. K. Gregory
(Bull. Amer. Mus. Nat. Hist., vol. 42, pp. 265-283), who concludes in his
recent paper ‘‘that the lachrymal fossa of extinct Equide did not lodge a
sebaceous gland like the ‘larmier’ of the deer and antelopes,’’ but ‘‘ prob-
ably did lodge a greatly enlarged nasal diverticulum’’ as suggested by
Osborn, or that the malar fossa (immediately above the premolars like that
of the present Helaletes) probably marks the origin of the muscle which
raises the upper lip.

370 E. L. Troaell—Helaletes Redefined.

which does not, however, join the frontals and nasals to
form an arch over the front of the face, but rather the
maxillary of the tapir recedes to a point well above the
orbit, to form the border of the anterior nares.

Authors have postulated Heptodon Cope, as well as
Helaletes Marsh, as a forerunner of the tapir. Heptodon
is the earlier but larger form and it seems impossible to
derive the small Helaletes directly from it.

SUMMARY.

A bold technique permitted the reconstruction of the
_ skull of the holotype of Helaletes boops Marsh, which,
in its more nearly normal condition, at once showed
features hitherto unknown in the genus.

A relationship to the tapir is indicated by nearly all
the important characters of Helaletes: the position of
the antorbital foramen, the rising of the sagittal crest,
the trend toward molariformity in the premolars, the
low maxillaries, and especially the presence of a pit in
front of the orbit which may have given rise to the
receded nasal aperture of the modern animal.

M. R. Thorpe—Areocyon. 371

Arr. XXXVI—Areocyon, a Probable Old World Migrant;
by Matcoum RutTHERFORD THORPE.

[Contributions from the Othniel Charles Marsh Publication Fund, Peabody
Museum, Yale University, New Haven, Conn. }

In this Journal for June, 1921, I described the lower

jaw of a carnivore, collected through the efforts of Pro-
fessor Marsh, in Oregon, in 1874. For this carnivore I
proposed the name Pliocyon marsh. Subsequently, I
substituted the name Areocyon to supplant my preoccu-
pied Pliocyon, and a note to this effect was printed in the
January, 1922, issue of this same journal.
The presence of this lower jaw in the New World is
decidedly interesting, for to my knowledge no specimen
comparable to it has been described or reported from
among the ancient fauna of North America.

There are but two similar forms in the extinct world
faunas with which we can make close comparison, and
one of these is much closer than the other. The nearest
ally of our American form is Samocyon priumigenius Roth
and Wagner, from Greece. This genus and species is
probably somewhat more specialized, although it was pre-
served in strata of older geologic age than was the Oregon
fossil. The two specimens show a curious admixture of
specialized and generalized characters, and it is rather
difficult to be certain which represents the later phase of
evolutionary development. For example, Are@ocyon has
two incisors (specialized), Simocyon, three, while the
former has two premolars and the latter but one (special-
ized). In other characters the Grecian genus is larger
and differently proportioned.

The other European species of Siwmocyon is 8. dia-
phorus Kaup, from Germany, and it very materially dif-
fers from the Grecian and American forms. In compar-
ison with the Oregon specimen, Kaup’s type possesses
P, and P, (lacking in both of the other specimens under
consideration) ; it has a larger and higher metaconid on
M,, and a smaller hypoconid; it has a shorter and lower
P,, with more prominent basal heel; M, is set in the
ascending ramus, while in A. marshi it is nearly level
with the tooth-row; and there are still other characters
which differentiate it from both of the other forms.

372 M. R. Thorpe—Areocyon.

The geologic horizon of A. marsht is Middle Pliocene.
It was collected about one mile west of Cottonwood, on
the East Fork of the John Day River, Oregon. The
enveloping matrix was soft tuff, lying between the basal
conglomerate and the capping rim rock of rhyolite, about
3 feet below the lower edge of the latter. Merriam has
designated this formation by the name Rattlesnake.

— Simocyon primigenius was found in the Pikermi beds
near Athens, Greece, while the type of S. diaphorus came
from the gravel deposits of Eppelsheim, Germany, the
deposits of both localities being of equivalent age. There
is a difference of opinion in regard to the geologic age
of these formations. The German geologist, Lepsius,
believes that these strata are unquestionably of the Lower
Pliocene epoch, and that the Eppelsheim beds present
the northern facies of the Pikermi beds. On the other
hand, the members of the French school of geologists are
equally positive that the entire series is Upper Miocene.
Whatever geologic age we ascribe to these Pikermi and
E;ppelsheim formations, they must both be placed in the
same period. I incline more to the French view of con-
sidering them Upper Miocene, but there is not sufficient
space to go into details on this point. It should be noted
that Von Zittel places Semocyon in the Upper Miocene.

According to Gaudry, Greece was most probably con-
nected with Asia by vast plains, produced by the reces-
sion of the sea which began in the Middle Miocene and
continued throughout Upper Miocene until the Mediter-
ranean Sea had become nearly dry and had reached the
stage where it was represented by only a series of brack-
ish lakes, with the consequence that Europe and Africa
were broadly connected by land masses. All of these
Upper Miocene deposits were apparently laid down in
shallow basins of fresh water, for the most part due to
recurrent torrential flooding, after the manner of our
own Oligocene deposits in the Great Plains region of
North America. |

The extent of these deposits shows how widespread.
were these conditions of deposition throughout Europe.
The beds are more or less local areas, extending from
western Portugal (Archino on the Tagus River) to
Maragha in Central Persia, and northward to Tcherni-
gow in south central Russia and to Eppelsheim, near

M. R. Thorpe—Areocyon. 373

Worms, Germany. Taking the Pikermi region as an
‘example, we must suppose that it consisted of grass-
covered lowlands which alternated with great forests of
beeches, bamboos and allied flora, in which lived rhinoc-
eroses, bears and other carnivores, monkeys, great herds
of hipparions, antelopes, chalicotheres, proboscidians and
soon. There isa notable absence of small forms in these
deposits, which may be accounted for by the conditions
of their deposition. It is probable that stream currents
sufficiently strong and swift to transport the bones of the
larger animals were too powerful for the smaller ones
and that these small bones were ground to pieces and
destroyed by attrition.

Let us now look at the faunal affinities. The affinities
of the Pikermi genera are with those of Africa rather
than of modern Europe. There is a notable absence of
wolf- or fox-like canids, both in Pikermi and in Eppels-
heim, and the family is represented by the curious short-
faced Simocyon. EHastward we find a similar faunal
phase in Samos and Maragha. In general, the fauna of
the latter approaches very closely to that of the Pliocene
of the Siwaliks in southern Asia and of China. Samos |
possesses a hornless giraffine, Samotherium, a form very
close to the present African okapi, while the aardvark is
common to both these deposits and to Africa of to-day.

We have now briefly reviewed the physiographic con-
ditions at the time of deposition of these European beds
and also have considered the faunal phases and affinities
of the animals of that time. We have seen unmistakable
evidence pointing to African rather than to European
allies of the fauna and the subsequent eastward trend or
migration of these forms through India and China, and
this leads us to the most probable explanation of the
presence in the western coast region of North America
of a Simocyon-like form.

In Pliocene times there is every reason to believe that
Asia and North America were connected by land in the
vicinity of Bering Strait and that hosts of animals passed
and repassed over this strip of land. Of course we might
argue that the theory of convergence could account for
the similar development of faunal forms on both sides
of the Pacific, but according to the laws of chance our
argument would seem to be rather ill-founded. For

374 M. R. Thorpe—Areocyon.

example, the mastodons of both Kurope and North Amer-
ica in the Pliocene were of two kinds, that is, some had
three ridges on the intermediate molars (trilophodont)
while the others were tetralophodont. |

The order Carnivora are unquestionably of Holarctic
origin, according to Matthew, and from this area they
have spread to practically all parts of the world. The
ancestors of Simocyon are obscure, but it would be reason-
able to suppose that they may have lived in Africa, owing
to the close similarity between the faunas of that conti-
nent and of Pikermi and Eppelsheim. On account of the
considerable contrast between the HKjppelsheim S. dia-
phorus and the Pikermi S. prumigenius, it would appear
that the northern species was aberrant and became extinct
before or soon after the beginning of the Pliocene, where-
as the southern form represented the stem stock and
followed the faunal migration through India and China
and across the Bering Strait land connection into North
America. It is probable that there might have been a
concentration of carnivore types in the region of this
‘‘oame trail.’’ It may have been some such concentra-
tion of mammals in a restricted area which produced the
rise of the Carnivora in the beginning.

The early Phocene is known to have been mild of cli-
mate, but during the entire epoch there was a very
gradual cooling or lowering of temperature which slowly
drove the animals southward, and may account for the
presence in Oregon in Middle Pliocene times of this
simocyonid.

It would seem to be a reasonable explanation to suppose
that the above line of migration was the one followed by
this phylum of carnivores from southern Europe to North
America. It also could account for the differences and
similarities in structure between the Old and the New
World species of this same phylum, if we take into con-
sideration the great lapse of time between Upper Miocene
and Middle Pliocene.

The systematic position of Are@ocyon and Simocyon
ought to be considered together, for I think that they both
were probably derived from the same genus. There are
several structural characters which prevent us from con-
sidering these forms as being at all closely related to

M. R. Thorpe—Areocyon. 375

the Felide, such, for instance as the possession of the
large M, which none of the post-Middle Oligocene felids
retained. The first lower molar is not at all cat-like in
structure, and the slenderness and general characters of
the ramus all point to an origin not in the felid phylum.
The evidence, however, does point unmistakably toward
the dog-like phyla.

While it is stated above that there are no closely com-
parable American forms so far known, yet we might point
out the essential differences and show wherein Are@ocyon
could have been derived from some autochthonous genus.

In the first place, the heel of the lower carnassial is
trenchant, for the hypoconid is about medially situated
on the talonid. This is also true of Sumocyon primigen-
ws, aS shown in a very good cast (Cat. No. 11649, Y.
P. M.) of the type.. Therefore we must place them in the
group which parallels the more typical canoid line of
descent, namely, the assemblage of forms to which belong
Daphenus, Temnocyon, Enhydrocyon, Ischyrocyon, and
others up to and including the recent genera, Cyon,
Icticyon, and Lycaon. Osborn has placed Simocyon also
in the Dhole-like group, and Areocyon should likewise be
classified under this head.

If Areéocyon is an autochthonous form, then Daphenus
and Temnocyon are most probably in the direct line of
ancestry, while Enhydrocyon is apparently an aberrant
side branch with which we are not at present concerned.

In the Upper Miocene (Loup Fork beds) there is a
peculiar genus, [schyrocyon, which is here provisionally
placed in this group on account of the trenchant heels
of the first two lower molars. The large size of the
molars and the absence of the metaconids upon them
show that it is, however, far removed from the large
majority of the other known genera of the Canide, and
especially so from Areocyon. It probably represents
another of the various aberrant forms in the canid
phylum. According to Doctor Matthew, the general pro-
portions suggest Amphicyon and Dinocyon.

In the later epochs, Pliocene and Pleistocene, of North
America, we find more nearly contemporaneous forms,
but if we examine them in detail we see that the similar-
ities are more apparent than real. In these later forma-
tions there are various more or less hyenoid Canide,

376 M. R. Thorpe—Areocyon.

such as Hyenognathus Merriam, Borophagus Cope, Chas-
maporthetes Hay, and so on. It is hardly probable that
any true hyenids will be found in North America and it
seems that these hyenid forms may be due to convergence.

Let us, however, compare one of these genera, for

Mason CHARACTERS.

Hyenognathus pachyodon.* Arcwocyon marsha.

1. ‘‘Mandible short and mas- 1. Mandible moderately long
sive.”’ and slender.

2. ‘‘Alveolar margins great- 2. Alveolar margins but very
ly flared below P, and slightly flared below P,.
| sa ;

3. Dentition 2,,°C;, P;, M.: 3. Dentition 1; °C), Pave

4 Pe vandsP. smalts 4. P, and P, absent.

5. ‘*P, very large, conical, 5. P, relatively smaller, com-
without accessory tu- pressed, with prominent
bercles.”’ posterior tubercle and

heel.

6. ‘‘M, massive; protoconid 6. M, very large; protoconid
and paraconid forming robust; paraconid large
a heavy shear, meta- and high; metaconid
econid absent; heel prominent; heel large
short, with reduced hy- with low hypoconid me-
poconid and entoco- dially situated on talonid.
nid.’’ (Heel basin- (Heel trenchant. )
shaped. )

7. ‘‘M, and M, small.”’ 7. M, long and stout. No Mg.

Spe leas JOSE: 8. P, present.

-9, Premolars crowded. 9. Long diastema between P,

| andebas

10. Incisors spaced and none 10. Incisors crowded and outer-
in front of canine. most one in front of ¢@a-

nine. |

11. Symphysis long. 11. Symphysis short.

12. M, and M, set in ascend- 12. M, placed nearly level with
ing ramus. respect to tooth-row.

13. Muzzle much wider. 13. Muzzle much narrower.

14. Quaternary, probably. 14. Middle Pliocene. Near Cot-
Asphalto, Kern Co., tonwood, East Fork of
Calif. John Day River, Ore.

1 Merriam, J. C., The Pliocene and Quaternary Canide of the Great Valley
of California. Univ. Calif. Pub., Bull. Dept. Geology, 3, 278 ff. with plates
and figures, 1903.

M. R. Thorpe—Areocyon. 877

example Hyenognathus, with Areocyon. From a survey
of the following lists of major characters of the compar-
able elements of the two genera, it is seen that there can
be no close affinity between the two.

In conelusion, it is my opinion that there is nothing to
preclude the possibility of Areocyon being an autoch-
thonous form, most probably having developed from the —
trenchant-heeled series, and belonging to this pseudo-
eanoid line. However, both Professor Lull and the
author incline very much more strongly to the theory of
migration for this form and its derivation from Old
World stock. If the lower jaw upon which Arcéocyon is
established had been collected in the Pliocene beds of
Europe, I should have no hesitancy about referring it to
the genus Simocyon, or at most to a subgenus under it.

If Areocyon should prove to be a derivative of purely
American ancestry, the possibility of which I doubt at
_ present, it will be one of the most remarkable cases of
convergence known to the science of vertebrate paleon-
tology.

378 Scientific Intellagence.

SCIENTIFIC INTELLIGENCE

I! CuHrEmistry awp Puysics.

1. Hydrated Oxalic Acid as an Oxidimetric Standard.
ArTHuR KE. Hint and THomas M. SmitH have rendered a
valuable service to the art of volumetric analysis by devising a
method whereby crystallized oxalic acid H,C,0,.2H,O can be
prepared in a state of purity in regard to its contents of water,
so that it contains no excess of moisture derived from the mother-
liquor, nor any deficiency of water due to efflorescence. The
principle of the method is very simple and depends upon expos-
ing the moist acid to an atmosphere in equilibrium at all ordinary
temperatures with the pure hydrated acid. Such an atmosphere
is obtained in the presence of mixtures of the hydrated and anhy-
drous acid obtained by heating the crystallized acid in a porece-
lain dish upon the steam-bath for a few hours. When a large
excess of this mixture is placed in a desiccator with the moist acid
the moisture of the latter is lost, but none of the water of crystal-
lization can go off on account of the presence of the hydrated acid
in the mixture which establishes the proper tension of aqueous
vapor to prevent this change, since the system becomes univariant
according to the phase rule of Gibbs. The authors have found
it necessary to powder the crystallized acid sufficiently to pass a
100-mesh sieve in order to expose the mother-liquor included in
the crystals to the process of drying. The time necessary for
drying in a desiccator under these conditions is about two days,
but the time may be shortened to about one hour by the use of
a current of air which is first bubbled through a saturated solu-
tion of the acid and then is passed through U-tubes containing
the desiccating agent and through a closed jar containing the
substance. |

The results given by the authors show that the acid prepared
in this way gives results that are reliable within very small limits
of error when used with potassium permanganate, and it appears
probable that it might be used very satisfactorily for standard-
izing solutions of strong alkalies by the use of phenolphthalein
in boiling solutions.—Jour. Amer. Chem. Soc., 44, 546.

H. L. W.

2. The Atomic Weight of Beryluum (Glucinuwm).—HOnic-
SCHMID and BIRKENBACH have made a new determination of this
atomic weight, apparently with great care and skill and with
satisfactorily agreeing results by the analysis of the anhydrous
chloride. They started with the commercial carbonate and puri-
fied the material, particularly by crystallizing and afterwards
volatilizing the basic acetate. The latter was converted into

Chemistry and Physics. 379

nitrate, then into oxide, from which the chloride was prepared by
ignition with sugar- char coal in a stream of chlorine. The
chloride was fused i in quartz tubes for weighing.

The results gave the atomic weight 9. 018 for Be or Gl, which
is about 1% lower than 9.1 the value accepted in the Interna-
tional Table. The result obtained by Parsons, in this country,
from the conversion into oxide of the acetylacetonate,
Be(C.H,O.),, and the basic acetate, Be,O (C,H ROn) =. was 9.105;
henee it appears that further work is needed to establish the cor-
rect value. There appears to be no evidence as yet that beryl-
lium contains isotopes, and if it does not contain them the new
value for the atomic weight, since it is very close to an integer, is ~
perhaps more plausible than the old one——Berichte, 55, 4

H. L. W.

3. An Introduction to the Physics and Chemistry of Colloids ;
by Emm HarscHex. 12mo, pp. 172. Philadelphia, 1922 (P.
Blakiston’s Son & Co.).—This is the fourth edition, entirely
re-written and enlarged, of a little book originating in London,
the first issue of which appeared in 1913. It gives an excellent
account of the facts and theories of this rapidly developing and
exceedingly important branch of physical chemistry, and it is to
be highly recommended for the use of those who wish to obtain
a clear, fundamental knowledge of this very interesting subject.

The historical part is briefly but very well presented, and men-
tion is made of the work of the American, Carey Lea, whose
papers on ‘‘ Allotropic silver’’ were published in this journal in
1889 and were evidently very important in connection with the
colloidal condition of the metals. No attempt will be made here
to discuss the presentation of the description and theoretical
topics except to say that this has been very well done. H.L. w.

4. Distillation Principles and Processes; by SIDNEY YOUNG.
8vo, pp. 509. London, 1922 (Macmillan and Co., Limited).—
This very comprehensive work is the successor to the author’s
excellent book on ‘‘Fractional Distillation’’ which appeared in
1903. The latter dealt chiefly with the details and principles of
small-scale distillations, such as are carried out in the laboratory.
A revision of this part, with the addition of a chapter on sublima-
tion, comprises about one-half of the new book, while the remain-
der of it is devoted to the following sections prepared by specialists
in various lines of distillation on a manufacturing seale: Acetone
and n-Butyl Alcohol, by Joseph Reilly and F. R. Henley; Alco-
hol, by F. R. Henley and Dr. Reilly; Petroleum, by James Kew-
ley; Coal Tar, by T. Howard Butler; Glycerine, by Lieut.-Col.
EK. Briggs; Essential Oils, by Thos. H. Durrans.

The book contains no less than 210 illustrations. Its very
elaborate treatment of the principles and apparatus of laboratory
distillation, together with the accounts of technical operations,

AM. JouR. Sct.—FirtH Series, Vou. III, No. 17.—May, 1922.

27

380 | Scientific Intelligence.

make it a very complete and satisfactory one, not only for the
use of chemists engaged in research, but also for technical
chemists. - H. L. W.

5. Separation of Isotopes——The lecture by F. W. Astron
before the Royal Institution as reported in Nature 107, 334,
1921, contains an account of all the elements with their isotopes
which have so far been examined by the positive ray spectro-
eraph. The great interest which attaches to the sweeping sim-
plications which have been made in our ideas of mass, by the now
well established view that all atoms are built of primordial atoms
of positive and negative electricity, has stimulated efforts to
secure evidence as to the existence of various isotopes, by other
lines of attack.

A very laborious series of operations on fractional diffusion
through pipe clay has shown that neon may be separated into two
components differing in density by 0.7 of one per cent.

The spectra of the three isotopes of lead have been shown tn
have small differences in the wave leneths of the principal lines
and the same is probably true of the spectra of ordinary thallium
and. that extracted from pitchblende. |

Following Aston’s announcement of the discovery that chlor-
ine consists of a mixture of two isotopes of atomic weights 35 and
37, Merton and Hartury suggested that as chlorine gas would
probably consist of three molecules in the ratio of 9:6:1, if a
beam of white light were allowed to traverse a column of chlorine
the ight might be differently absorbed by the different molecules,
so that the emergent radiations would have different intensities
in different wave lengths. It was accordingly proposed to allow
the heht which had traversed the filter to enter a vessel contain-
ing a mixture of hydrogen and chlorine which combine under the
influence of light of these wave leneths. It was calculated that
the rates of reaction for the different HCl molecules would be in
the ratio 1:(10)°:(10)**. If this were true the hydrochloric acid
formed would consist almost entirely of HCl,. provided the reac-
tion were allowed to proceed for a sufficient leneth of time.
This experiment has now been carried out with great care (Phil.
Mag. 43, 430, 1922) but it was found impossible to say whether
a real separation had been effected or not. The failure may have
been due either to secondary reactions or to the fact that the dif-
ference in the absorption spectra of the two isotopes was insuf-
ficient for the purpose.

BRoNSTED and HeEvesy have attempted the separation of mer-
cury isotopes by methods depending upon the different
molecular velocities appearing in consequence of the dif-
ferent masses of the isotopes. When the liquid is allowed to
evaporate, the rate at which the two molecules leave the liquid
should be inversely as the square roots of the masses. If a

Chemistry and Physics. 381

strongly cooled plate is placed just above the quid each mole-
eule as it reaches the plate will be condensed to a solid before
it has the opportunity of meeting the other molecules and be
returned to the liquid. A separation will thus be effected
between the slower and the faster moving molecules. Another
method proposed was to allow the molecular flow to take place
through a small opening. In this case the lighter and more
rapidly moving molecule should hit the opening more often than
the heavier one and if it is prevented from returning by trapping
it on a cooled plate a separation between the components of the
vapor should be secured in this way.

Both the evaporation method and the effusion method were
successfully tried by these authors who found that the quantities
separated were inversely proportional to the square roots of the
molecular weights of the isotopes as determined by Aston.—Phil.
Mag. 43, 31, 1922. Hohe Be

6. Newcomb-Engelmann Populdire Astrononue; Sixth Edi-
tion edited by H. Lupenporrr. Pp. XII, 889. Leipzig, 1921
(Wilhelm Engelmann ).—The favor with which this work by real
authorities has been received may be judged by the increasing
rapidity with which the successive editions have been exhausted.
First appearing in 1881 as a translation of the Popular Astron-
omy of Simon Newcomb, the later issues have preserved little
resemblance to the original work, so numerous and extensive have
been the changes. The editors have however always sought to
preserve the historical treatment which characterized the
author’s treatise.

Our knowledge of the structure of the universe and stellar
activity has so widened since the appearance of the last edition
in 1914 that the three chapters on Stellar Astronomy forming
Part IV have been entirely rewritten, and two new sections on
the Development of Mechanics since Newton and on the Principle
of Relativity have been added to Part I. In addition all the
material has been revised in the light of the newest research and
the numerical data have been corrected from the most trust-
worthy sources. Many sections are the work of specialists as, for
example, those on Meteors, the Physical Activity of the Stars, and
Star Clusters and Nebulae, by Eberhard; Stellar Parallax,
Proper Motion, Double Stars, and Variables, by Ludendorff; The
Fundamental Laws of Mechanics, the Three-Body Problem, The
Sun, and Novae, by Freundlich; the Planets, the Structure of
the Universe and Cosmogony, by Kohlschiitter. The unavoidable
increase in the size of the book has tempted the editor to omit the
appendix devoted to biographical sketches but in response to the
-desire of his colleagues they have been retained.

The title ‘‘popular’’ connotes little more than ‘‘non-mathe-
matical.’’ Outside of that it is a serious treatise for the student,

382 Scientific Intellugence.

teacher, or reader and a valuable compendium for practical work.
The typography is most satisfying and the illustrations beauti-
fully done. Their number has been increased from 228 to 240 in
this edition. As a work on general astronomy it is unequalled in
any language and on account of the present state of exchange its ©
cost is but a fraction of what would be required to produce it in
any other country. F. E, B.

7. The Two Orbit Theory of Radiation; by FRanxK H. BiGE-
Low. Pp. VII, 37.. Vienna 1921 (Austrian State Prmting
Office).—This brochure is designated as Supplement No. 2 to
Treatises on the Atmospheres of the Sun and the Earth. For
the purposes of atmospheric physics the author rejects the Bohr
theory of stationary non-radiating orbits with electrons jumping
across radial differences with a frequency v, as too artificial for
veneral application. In its place he proposes a two orbit theory
of radiation in which it is assumed that the negative electron is
revolving about the positively charged atomic nucleus while the
latter is revolving about the instantaneous line of translation.
Under these conditions the velocity of the electron is periodically
variable and generates trains of radiation.

The paper attempts to show by thermodynamic reasoning the
origin of solar radiation and the location in the solar atmosphere
where the several spectrum lines originate. FE. BE

Il. Grontogy anno: MInerALocy.

1. Ueber das Becken, den Schultergurtel und evmge andere
Teale der Londoner Archaeopteryx, by BRANISLAV PETRONIEVICS.
Pp. 31, 2 pls., Genf (Georg & Co.), 1921.—These studies are the
result of further preparation of the famous British Museum
specimen of Archwopteryx which has exposed a number of new
skeletal elements. A description and dimensions are given of
the pelvis, and it is compared with that of the Berlin specimen
and with that of other birds and reptiles. The shoulder girdle
and other elements are treated in the same manner. As a result
of these comparisons, the author arrives at the following con-
clusions: (1) that the birds are undoubtedly derived from the
reptilian stem; (2) that their ancestors are to be sought among
the Lacertilia, or that at least birds and Lacertilia came from a
common ancestry; (8) that the similarities between the birds on
the one hand and the Dinosauria and Pterosauria on the other
are due to convergence; (4) that Archewopteryx both in its pelvie
and shoulder girdles is more primitive than Archwornis; (5)
that Archwopteryx represents a generalized as well as a mixed
bird-type, since it combines primitive reptilian characters with.
advaneed bird characters; (6) that Archwopteryx either stands
near that generalized bird-type out of which developed both the
modern Carinates and Ratites, or is itself that type; (7) that as

Geology and Mineralogy. 388

early as the Jurassic the division of the bird-stem into the Car-
inate (Archewornis) and Ratite (Archwopteryx) groups had
begun; and (8) that the division became still more marked in the
Cretaceous, in that Hesperornis lies in the line toward Ratite and
Ichthyornis in that toward Carinate, evolution.

The author gives to the Berlin specimen the new generic name
Archeornis, leaving the London specimen under the original
genus, Archwopteryx. He further finds sufficient distinction
between them to postulate the former as the ancestor of the
Carinate birds through the Cretaceous Ichthyornis, and the lat-
ter as the forerunner of the Ratite through Hesperornis, reviv-
ing Marsh’s contention that Hesperornis belongs or is related to
the Ratite group, a premise which has now but little acceptance.

R. 8. L.

2. Die Antike Tierwelt, by Orro KeELurr. Gesamtregister,
by Eucen Sraicer, Leipzig (Wilhelm Engelmann), 1920.—This
“is a somewhat detailed index pertaining to the two volumes pre-
viously noticed in this Journal (Bd. I, July, 1910; Bd. II, Octo-
ber, 1913) ; it lends final completeness to a work of great interest
to the student of zoology as well as to the antiquarian. R.S. L.

3. Origin and Evolution of the Human Race; by ALBERT
CHURCHWARD. Pp. xv, 511, 78 pls., many text figs., New York
(Macmillan Company), 1922.—A large and amply illustrated —
volume based on a considerable body of detailed and apparently
accurate research, not done in the laboratory or library but
largely in actual contact with the various peoples with which it
deals. Doctor Churchward makes a great deal of totemism and
contributes much of interest to our knowledge of this rather
obseure subject. . His main thesis, however, is to prove Africa to
have been the primal home of mankind, the place of his original
evolution and from which he migrated to all parts of the earth.
He further believes that, as is the rule with other forms of life,
man’s evolution has also resulted in increase of stature, and he
looks upon the African pygmies, not as physically degenerate,
but as in a state of persistent primitiveness. These represent the
stock out of which arose the other human races, through the
‘“Masaba’’ negro to the Nilotic negro and thence to the diverse
sorts of humanity. Churchward finds no difficulty in assigning
each of the various types of European prehistoric men—Heidel-
berg, Neanderthal, etc.—to his own place in the scheme, the above
mentioned being but Nilotic negroes which have migrated into
Europe from ancient Egypt, to be replaced in time by another
wave of migration from the same source. Chapter XXXII gives
a tabulated summary of the assumed relationships, the standard
accepted Huropean chronology of the Recent and Pleistocene
being, in the author’s opinion, absurd. The book contains much
of value, but the thesis is so novel that, as Churchward says, he

384 Scientific Intelligence.

hardly anticipates its acceptance by the present generation of
selentifie men. R. S. L.

4. The Topographic and Geological Survey of Pennsylvania;
GrorcGE H. ASHLEY, State Geologist—Bulletins 1 to 25 of the
Survey were mentioned in the last number (pp. 305, 806). Nos.
26 to 85 are now issued (also mimeographed). ‘These are all
devoted to coal beds or coal reserves in various countries except
No. 28 which discusses the magnesite of the State; it is by R. W.
STONE.

Dd. Geological Survey of the Umon of South Africa; PERcY
A. Waaner, Geologist—Memoirs issued at Pretoria, are Nos. 16
and 17, both by Dr. Wagner, and liberally illustrated. No. 16
discusses the Mutue Fides—Stavoren Tinfields and is a valuable
contribution to a subject of wide interest.

No. 17, also by Dr. Wagner, is a report on the Crocodile River
iron deposits. Though at present rather inaccessible, particu-
larly in the rainy season, being 68 miles from the nearest railway »
station, they promise to be of much importance in the future.

The geology of the country surrounding Johannesburg, being
an explanation of the Johannesburg sheet (No. 52), is discussed
by Dr. E. T. M&uior in a pamphlet of 46 pages.

6. Carbomferous Glaciation of South Africa.—This is the
title of a paper by AuEex. L. pu Torr published in the Transac-
tions of the Geological Society of South Africa (vol. 24, 1921).
A map shows clearly the radiation of the Carboniferous ice. It
is stated that the Dwyka ice-sheet of the Upper Carboniferous
was formed by the ice coalescing from several distinct centers,
Namaqualand, Griqualand West, Transvaal and Natal. The gen-
eral direction was southerly or pole- ward.

7. South Australia Geological Report for 1920; L. Kerra
Warp, Director of Mines anal Government Geologist. Adelaide,
1921. Eight pages.—The report of the director recently recelved
presents concisely the geological and economical results of the
vear’s work.

The Department of Mines has also issued No. 34 (74 pp., illus-
trated) of the Mining Review for the half-year ending June 30,
1921. This has been compiled by Lionrn C. H. Grn.

8. Mineral Production in the Unted States and elsewhere.—
Under the general head of Mineral Resources of the UNITED
STATES are to be mentioned, first of all, the well-known publica-
tions of the U. S. Geological Survey, which are issued in separate
chapters and kept admirably up to date.

The very varied mineral resources of the State of New York
are presented by Davin H. Newnanp in the Museum Bulletin
(Nos. 223, 224), Jonn M. Cuarxks, Director under the University
of the State of New York. Some forty separate important min-
erals and rocks are enumerated and their occurrence described 1 im
this publication of 315 pages (illustrated).

Geology and Mineralogy. 385

Bulletin No. 23 of the Geological Survey of Alabama, EUGENE
A. SmitH, State geologist, giving statistics of the mineral
production of ALABAMA, has been compiled from the Mineral
Resources of the National Government. Further Bulletin No.
24 by Grorce H. Cuark, assistant geologist, gives an account of
the Alabama mica deposits. The state comes sixth among the
states, North Carolina furnishing about one-half that of the
country.

The annual report on’ the mineral production of CANADA in
1920 shows that the total value amounted to neariy $228,000,000,
the highest on record and an increase of 29 per cent over 1919.
The total value in 1886 was somewhat more than $10,000,000 ;
this was doubled in 1896; another decade showed further
inerease of three and a half times, while the production now
noted (1920) is about three times that of 1906. For metallic
products the quantities obtained come in the following order:
copper, nickel, zinc, lead, silver, gold; 600 crude ounces of plati-
num were produced. Among the non-metallic products coal
leads by a large amount, with gypsum and asbestos prominent.

The second annual report for AuBERTA (1920, 152 pp.) has
been prepared by JoHN A. ALLAN. First in importance comes
coal, then rock salt and petroleum (11,718 barrels in 1920) ; clays,
iron, ete., are also described.

9. The Future of the Comstock Lode.—The well-known Com-
stock lode in Nevada, which produced such vast wealth in silver
and gold especially in the decade following 1859, has been
studied anew by the U. 8S. Geological Survey. In a recent leaflet
it is remarked that the fundamental geologic problem is the per-
sistence of the ores with increase in depth. If the rich silver
ores were deposited wholly or largely from solutions that
ascended from sources far below the surface, deeper exploration
is fully warranted. If they owed their richness to the action of
descending surface waters on ores that originally contained rela-
tively little gold and silver, then there is little to encourage
deeper mining. The steady progress in metallurgy, by which
more metal is recovered from low-grade gold and silver ores, has
given large practical importance to the question of the per-
sistence in depth of ores of this class.

EK. S. Bastin has made a microscopic study of the Comstock
ores. In his report (Bulletin 735-C) entitled ‘‘Bonanza ores of
the Comstock Lode,’’ he concludes that in the ores from depths
greater than 500 feet, which include most of the bonanza ores of
the lode, the silver is practically all in primary minerals.
Descending solutions of surface origin produced a large increase
in the silver content of certain ores obtained within 500 feet or
less of the surface, yet even at those depths a part of the silver
is contained in primary minerals, and some rich ores taken from

S80 °tare | Scientific Intelligence.

slight depths showed no secondary silver minerals. Gold, so far
as observed, is primary in all the ores. A revival of the gold and
silver age of Comstock mining is not to be looked for, since the
tremendous fracturing which created the channels that made ore
deposition possible was more extensive near the surface than at
- great depths. Nevertheless, the ‘‘roots’’ of an ore deposit so
immense are by no means small, and the Comstock operators have
in recent years shown their confidence in the existence of deep-
lying bodies of workable ore by draining a large part of the lode
to and below the 2,900-foot level. Although the deeper parts of
the lode probably contain no ore bodies comparable in size and
richness to the great bonanzas of the past, yet the primary origin
of some of the rich ores encourages deeper development.

10. <A new Meteoric Iron; by Grorce P. Merri (Communi-
cated ).—A 25-lb. mass of meteoric iron has recently come to the
National Museum from Nickelsville, Scott Co., southwest Vir-
ginia. It is badly oxidized and evidently represents a very old
fall. It is, however, of interest since it shows unmistakable evi-
dence of the secondary granulation to which Berwerth has given
the name metabolism.

III. Misce,rtantous Screntiric INTELLIGENCE,

1. National Academy of Sciences—The annual meeting of the
National Academy of Sciences was held at the Natural History
building, U. 8. National Museum, in Washington, on April 24, 25
and 26. The scientific sessions were open to the public as usual.
The number of papers offered for reading was very large, num-
bering about 40 in the preliminary program issued by the Home
Secretary, Dr. C. G. Abbot. <A particularly interesting feature
of the meeting was the address by Dr. H. A. Lorentz, professor of
physics of the University of Leiden and Foreign Associate of the
National Academy, on ‘‘Problems of modern physics.’’ This
address was given under the joint auspices of the Carnegie Insti-_
tution of Washington and the National Academy of Sciences, on
Monday evening, April 24. Reception to Dr. and Mrs. Lorentz
followed in the Galleries of the U. S. National Museum. The
subscription dinner for the members was held on the evening of
April 25 at the Hotel Powhatan, and there was a ladies subserip-
tion dinner, at the same place and time.

The late date in the month of the Academy meeting makes it
impossible to give here the lists of new members elected, active
and associate. These will follow in the next number.

2. <A Text-book of Zoology; by the late T. JEFFERY PARKER
and Wiuuiam A. Haswexu. Third edition. Two volumes; vol.
I, pp. xl, 816, with 713 figures; vol. II, pp. xx, 714, with 560 fig-

~ Miscellaneous Scientific Intelligence. 387

ures. London, 1921 (Macmillan and Co.).—For nearly a quar-
ter of a century this has been a standard text and reference book
in all parts of the English-speaking world. The first volume
includes all groups of animals except the Chordates, each of the
chapters having been carefully revised from the earlier editions,
and the sections on the Nemathelminthes, Molluseoidea and Annu-
lata considerably altered and improved. Still further revis-
ion might well have been made in some of the other groups, par-
ticularly in the classes Nemertea and Arachnida, to bring them
in line with the results of recent studies, but on the whole the
work is of such excellence that it will continue to hold its place
as the leading elementary text-book of Zoology in any language.
The second volume treats of primitive chordates and the verte-
brates, together with chapters on the distribution of animals, the
philosophy of zoology, history of zoology, and literature.
Experience with the earlier editions has shown that the book,
although sufficiently extensive to meet the needs of well advanced
students of zoology, has been so written as to make it intelligible
to those with no previous knowledge of the subject. The impor-
tant groups are each introduced by a detailed study of a typical
example, and with this as a basis the modifications of the organ
systems in the other members of.the group are described. The
illustrations are widely known for their excellence and have been
copied extensively in other works. The printing of the new
edition, however, is not quite up to the high standard of the
earlier volumes. W. R. C.
3. Reptiles of the World; Tortoises and Turtles, Croco-
dilians, Iazards and Snakes of the Eastern and Western Hemis-
pheres; by RaymMonp L. Dirmars. Pp. 373, with 89 plates.
New York, 1922 (The Macmillan Company).—Probably no one
in America has a wider knowledge of the life and habits of the
reptiles than the curator of this group of animals in the New
York Zoological Park. In this popular account of his pets,
Mr. Ditmars makes available to the general public as well as to
the scientific world the most interesting things about the crea-
tures to which he has devoted his life. In these days of special-
ization on minutiae it is pleasant to find a good old-fashioned
natural history, but this is old-fashioned only in the sense that the
subject is treated broadly and with a living interest, for it is
strictly up to date from a scientific standpoint and thoroughly
reliable. The illustrations are from photographs, most of which
were taken by the author from living specimens. The book con-
tains whatever is best worth knowing about these animals and the
reader who has the good fortune of spending a few hours with it
will no longer think of the reptiles as the repulsive creatures he
has imagined them, but will long to make a more intimate and
personal acquaintance with them. WwW. R. C.

388 Scientific Intelligence.

4, Monographia delle Coccanghe italiane; by Gustavo LEon-
ARDI. Hdited and supplemented by F. Stuvestri. Pp. 555, with
375 figures. Portici, 1920 (Ernesto della Torre).—This impor-
tant work summarizes the results of a lifelong study of the
structure, habits and life histories of the scale insects (Coccidae)
found in Italy. One hundred and forty-seven species are
included, each of these being described in great detail for
all stages in the complicated life history and illustrated for the
most part by original drawings. Because of the great economic
importance of these insects in the destruction of trees, fruits, and
other crops this work will prove of service not only to science but
also to agriculture, since a complete knowledge of the life of an
insect is essential for devising practical measures for its control.

W. R. C.

5. Fauna Hawanensis; or the Zoology of the Sandwich Isles ;
beng Results of the Explorations wstituted by the Joint Com-
mittee appointed by the Royal Society of London for Promoting
Natural Knowledge and the British Association for the Advance-
ment of Science, and carried on with the assistance of those
Bodies and of the Trustees of the Bermce Pauahi Bishop Museum
at Honolulu. Three volumes, quarto; 2410 pp., with 64 plates;
edited by Davin SHarp. Cambridge, 1899-1910 (University
Press ).—The land fauna of the Hawaiian islands is of particular
interest to zoologists because of the question concerning its
origin. Determination of the closeness of its relationship with
that of other islands of the Pacific or with that of the American
continent will afford evidence as to the nature of the geological
changes which have occurred in the more recent eras. These
studies, undertaken when the zoology of the islands was almost
unknown, show a land fauna consisting of more than 3,300
species of insects, 500 forms of mollusks and 50 birds, including
the many new species deseribed in these reports. There are no
indigenous mammals, earthworms or ants. The introduction by
R. C. L. Perkins, consisting of 228 pages, with 16 plates, reviews
the salient features of all groups of the fauna and discusses the
evidence as to the origin of the latter.

In Volume I, R. C. L. Perkins and Aug. Forel describe many
new genera and species among the 178 species of aculeate
Hymenoptera now recorded from the islands; Edward Meyrick
describes 200 new Macrolepidoptera among the 292 species
found; Lord Walsingham lists 441 species of Microlepidoptera,
of which 420 are endemic and mostly new to science; W. H.
Ashmead records 128 species of parasitic Hymenoptera, inelud-
ing many new forms; while the birds are discussed by R. C. L.
Perkins.

The Orthoptera and Neuroptera and supplements to the
Hymenoptera, and Diptera by R. C. L. Perkins are included in

Miscellaneous Scientific Intelligence. 389

Volume IT, together with the Coleoptera by Perkins and Sharp,
the Mollusea by E. R. Sykes, the earthworms by F. E. Beddard,
the entozoa by A. E. Shipley, the Arachnida by Hugéne Simon,
the Isopoda by Adrien Dollfus, the Amphipoda by T. R. R. Steb-
bing, and a supplemental list of Hemiptera by G. W. Kirkaldy.

In volume III are the Diptera by P. H. Grinshaw and P.
Speiser, Hemiptera by G. W. Kirkaldy, Coleoptera by D. Sharp,
Hugh Scott and D. C. L. Perkins as well as other groups of
insects, Mallophaga and Myriopods.

With the publication of these reports the Sandwich Islands
may be considered as better known zoologically than most other
parts of the world. But the fact that some 300 new species of
insects and numerous species of other animals have been
described since their publication shows that muck still remains to
be done. Niioudive (Ch

6. World Atlas of Commercial Geology. Intwo parts. Part
I. Distribution of Mineral Production. Pp. 72, 72 plates.
Washington, 1921 (U. S. Geological Survey, price $2.00).—
The text of this atlas consists of brief accounts of thirty of-the
more important mineral commodities. For each commodity is
oiven in compact form, and stripped of all unnecessary techni-
ealities, the salient features of the uses, possibility of employment
of substitutes, geologic occurrence, geographic distribution,
technology, centers of production and consumption, statistics of
output, and the position of the United States in the industry of
each of the mineral products. Many of the tables of world out-
put by countries are pioneer compilations and have entailed an
enormous amount of research in a dozen different languages.

The maps, as the title of the publication suggests, are the main
feature. The thirty commodities are divided into nine groups,
and the output of each commodity is shown by means of nine
series of eight maps each. The first map of each series shows the
world production and consumption of the commodities of each
eroup; for example, one series shows the industrially related
group of iron ore, manganese ore, and chromic iron ore, each com-
modity being indicated by a separate color. The first map of each
series thus gives at a glance a world view of the distribution of
mineral production, and further it serves as a key to the succeed-
ing maps, which show the production by leading districts in the
countries of the several continents, including Oceania. The pro-
duction shown is that for 1913, the last normal year before the
war. The final map of each series shows the production in the
United States in 1918. A vast amount of information is graphi-
cally summarized and illuminatinely presented in these maps.

Part Il. Water Power of the World; by Herman STABLER,
B. E. Jones, O. C. Merri, and N. C. Grover. Pp. 37, 10 maps
(U.S. Geological Survey, price, $1.00).—This is an inventory of

390 Scientific Intelligence.

the world’s water power, including both the developed and the
potential resources. The present developed capacity of the
world is 23 million horsepower and the potential water power is
440 million horsepower at ordinary low-water stage. Africa
leads in undeveloped water power, with 190 million horsepower ;
Asia has 71 million; North America 62 million, South America
54 million, and HKurope 45 million. North America has devel-
oped 12 million horsepower, more than all the rest of the world;
of this amount 9,243,000 horsepower are installed in the United
States. The quantitative distribution of these resources in the
several continents and by States in the United States is shown
graphically on an excellent series of maps (plates 2-8).
ADOLPH KNOPF.

7. The Friendly Arctic; by VILHJALMUR STEFANSSON. Pp.
xxxl, 784. New York, 1921 (The Macmillan Co.).—This
volume is the narrative of five years of Arctic exploration
in the great archipelago lying north of the American con-
tinent. The author was commander-in-chief of an expedition
supported most generously by the Canadian Government—
‘‘the most comprehensive polar expedition that ever sailed.”’
It included a scientific staff of fifteen men, who were to investi-
gate the anthropology, biology (both terrestrial and marine),
geography, geology, oceanography, and terrestrial magnet-
ism of the region explored. The expedition sailed from Nome
in July, 1918, in the Karluk, accompanied by two 30-ton
auxiliary gasoline vessels. Owing to the loss of the Karluk,
which earried most of the equipment regarded as essential to
polar exploration, the commander decided to try out the method
of “‘living off the country,’’ in order to carry out his geographic
program. The weight of all evidence was against this bold idea
—‘‘oeographers, explorers, whalers, and Eskimos alike were of
the opinion that our plans were unsound and that the attempt to
earry them out would be disastrous.’’ The scientific associates
felt this way about the commander-in-chief’s project and there-
after, according to the author, became firmly insubordinate. In
justice to them, however, it must be said that they do not present
their own case in this book. The volume is largely an account of
the triumphant success of the commander, accompanied by two
sailor companions, in exploring great areas of frozen Polar seas
north of Canada and the archipelago on the east of these seas, and
it describes the important geographic discoveries that resulted
from this new method of polar exploration. The lifeless Polar
Sea, as it was termed by the foremost authorities, was found to
be stocked abundantly with food and fuel. The book makes
absorbingly interesting reading. The ‘‘friendliness’’ of the
Arctic is much overemphasized, however, and our dissent from
the overdrawn picture of its kindliness is heightened when we
read the roster of the brave men who perished on the Karluk.

Miscellaneous Scientific Intelligence. 391

The scientific results of the expedition are being published by
the Department of the Naval Service at Ottawa and may even-
tually total twenty or thirty volumes. ADOLPH KNOPF.

8. The Evolution of Climates; by Marspen Manson. Pri-
vately printed (1922). 66 pages—The author of this paper
concludes: (1) That solar control of climates did not prevail in
any period of the geologic past. Earth heat, conserved by a non-
conducting crust, was liberated slowly after crustal disturbances.
This heat warmed the oceans, which gave off vapor to form a
permanent cloud blanket. (2) After long periods of crustal
quiet the available heat decreased, the lands were cooled, and
glaciation resulted. In most cases the heat conserved in the
oceans prevented widespread glaciation and maintained warm
climates even in polar latitudes. Liberation of additional heat
from the earth, largely as a result of crustal failure under gla-
cial loading, restored universal mildness of climate. (3) In
the Pleistocene both land and ocean were greatly cooled. As a
- consequence, cloudiness was reduced, and the present zonal
climates, under solar control, resulted.

The scientific reader must feel that the author has not been able
‘‘to explain the variations and developments of geologic and
present climates without resort to assumptions or working
hypotheses of any kind.’’ For example the attempt in the paper
to link climate and diastrophism in a reversible reaction, making
olacial loading the chief cause of profound crustal disturbances,
is in itself a ‘‘workine hypothesis’’ which serious students of
structural and historical geology will hardly accept.

CHESTER R. LONGWELL.

9. An Essay on the Physiology of Mind; by Francis X.
Dercum. Pp. 150. Philadelphia, 1922 (W. B. Saunders Com-
pany ).—This book revives a title which Maudsley used when he
published his well-known volume in 1867. Dr. Dercum endeay-
ors ‘‘to apply purely physical conceptions to the interpretation
of mind.’’ The paradoxical wording of the title may be justified
by the materializing tendency of recent psychology and by the
opposite tendency of recent physics. ‘‘The modern study of
atom reveals it to be but an expression of energy, indestructible,
persistent, unknowable. Does not this cause the difference
between the old conceptions of ‘material’ and ‘immaterial’ to
disappear? Does it not make unnecessary—as it is Impossible—
a ‘dual’ conception of the universe?’’

We should not, according to Dercum, introduce an immaterial
principle, like the psyche or the spiritus, to explain behavior ; we
should frankly apply the mechanics or physiology of the amceba
and sea-anemone to higher levels. Oxidation, reduction, ame-
boidism, transmitted compacts, destruction of molecules, polar-
ization, etc., account for lower forms of behavior, why not for
higher? ‘‘The 10,000 million intercalary neurones of the cortex

392 Scientific Intelligence.

add merely to the complexity of the response; the purely physi-
eal, automatic character of the latter remains unchanged.’’
Dercum holds that the cortical neurones have the property of
amceboidism. ‘‘In sleep the neurones have their processes
retracted; in consciousness their processes are extended.’’
(Juiescent neurones occupy the unconscious field. Dr. Dercum’s
discussion is compact and readable. It assembles many facts in
an original manner. As a broad formulation of the mechanical
basis of behavior it has a suggestive value.
In an addendum on the Pathological Physiology of Mind, the

author apphes his formule to the interpretation of dementia
praecox, delirium, hysteria, etc. It is here that the explanatory
weakness of the formule is clearly revealed; and the result
approaches what Adolph Meyer once called ‘‘neurologizing tau-
tology.’’ ‘‘Dementia praecox is essentially an affection of endo-
eenous deterioration.’’ It results in ‘‘adynamia of the field of
cortical activity.’’ ‘‘The level, the intensity of the metabolic
processes, of the neurones is lowered,’’ with resulting slowness
of speech, poverty of thought, ete. Mental phenomena may in
essence be physical, as the author holds, but their further eluci-
dation depends upon dynamic, genetic studies of personality, by
psychiatric and psychological methods. Poverty of thought, for
example, may be more apparent than real, and instead of repre-
senting a quantitative reduction in the physico-chemical sense, it
may have an altogether different significance when given a fune-
tional interpretation. ARNOLD GESELL, M.D.
10. Royal Natural History Musewm at Brussels —The fourth
part of volume VIII (Memoir 31) is an extended quarto. volume
of 208 pages with 233 textfigures, devoted to the Chironomides of
Belgium and especially those of the Flanders. The author is
Dr. M. GorETGHEBUER, who was also the author of part three on
the Ceratopogonine of Belgium (1920).

OBITUARY.

M. CamILuE JorpAN, the veteran French mathematician and a
man of rare genius, died in March at the age of eighty-four
years.

Proressor BENJAMIN Moore#, who held (from 1918) the chair
of biochemistry in the University of Oxford and earlier a mem-
ber of the staff of the Yale Medical School, died on March 38 at the
age of fifty-five years.

Dr. Augustus Diistr& WaLusr, director of the physiological
laboratory and professor of physiology in the University of
London, died on March 11 in his sixty-sixth year. ,

Dr. THEODORE LiEBISscH, professor of mineralogy at the Univer-
sity of Berlin, died on February 9 in his seventieth year.

7

farp’s NaturaL SciENCE ESTABLISHMENT
A Supply-House for Scientific Material.
Founded 1862. Incorporated 1890.

A few of our recent circulars in the various
departments:

Geology: J-32. Descriptive Catalogue of a Petrographic Col-
lection of American Rocks. J-188 and snpplement.
Price-List of Rocks.

Mineralogy: J-220. Collections. J-288. Minerals by Weight.
J-224. Autumnal Announcements.

Paleontology: J-201. Evolution of the Horse. J-199. Palz-
ozoic index fossils. J-115. Collections of Fossils.

Entomology: J-33. Supplies. J-229. Life Histories. J-230.
Live Pupe.

Zoology: J-223. Material for dissection. J-207. Dissections
of Typical Animals, ete. J-38. Models.

Microscope Slides: J-189. Slides of Parasites. J-29. Cata-
logue of Slides.

Taxidermy : J-22. North American Birdskins. Z-31. General
Taxidermy.

Human Anatomy: J-37. Skeletons & Models.

General : J-228. List of Circulars & Catalogues.

Wards Natural Science Establishment
84-102 College Ave., Rochester, N. Y., U. S. A.

“SCIENTIA”

INTERNATIONAL REVIEW OF SCIENTIFIC SYNTHESIS. Jsswed monthly (each
number consisting of 100 to 120 pages). Editor: EUGENIO RIGNANO.

This is the only review which has a really international collaboration ; which is

of world-wide circulation and occupies itself with the synthesis and unification of

knowledge, in the history of the sciences, mathematics, astronomy, geology, physics,

' chemistry, biology, psychology and sociology.

It is the only review which, by inquiries among eminent scientists and writers
(on: The philosophical principles of the sciences; Fundamental astronomical and physical
questions of current interest; Contributions given by the various countries to different
branches of knowledge; Question of vitalism; Social question; International questions
raised by the world war), makes a study of the problems interesting scientific and intel-
lectual circles throughout the world.

It has published articles by Messrs.: Abbot=Arrhenius -Ashley = Bayliss - Beichman-
Bigourdan - Bohlin - Bohn= Bonnesen - Borel = Boaty = Bragg = Bruni = Burdick - Carver -
Caullery - Chamberlin = Charlier = Claparede = Clark = Costantin - Crommelin = Crowter =
Darwin = Delage = De Vries - Durkheim = Eddington = Edgeworth = Emery = Enriques =
Fabry = Findlay = Fisher = Fowler = Golgi = Gregory = Harper = Hartog = Heiberg - Hinks-
Hopkins-Inigues-Innes-Janet-Kaptein-Kaye-Kidd-Langevin-Lebedew-Lloyd Morgan-
Lodge = Loisy = Lorentz = Loria -Lowell = MacBride= Meillet =Moret-Muir-Peano -Picard =
Poincare = Puiseux = Rabaud = Rey Pastor = Righi-Rignano-Russell-Rutherford-Sagnae =
Sarton = Schiaparelli- Scott = See = Sherrington = Soddy = Starling - Svedberg = Thomson =
Thorndike-Turner -Volterra-Webb-Weiss = Zeeman and more thana hundred others.

** SCIENTIA’”’ publishes its articles in the language of its authors, and joins to the
principal text a supplement containing the French translations of all the articles that
are not in French. (Write for a Specimen Number to the General Secretary of
“Scientia”, Milan, sending,-to defray postal and other expenses,-2 Francs in
stamps of your country.) Annual subscription : 40 sh., or 19 dolars post free.

Office : 43 Foro Bonaparte, Milan, Italy
Publishers: WILLIAMS & NORGATE-London; FELIX ALCAN -Paris

NICOLA ZANICHELLI-Bologna; RUIZ HERMANOS-Madrid;
WILLIAMS & WILKINS CO-Baltimore.

“CONTENTS.

Fe ate Rae

ee XXIX, oa Complex Chlorides Containing:
by H. L. Wetts,

ae XXXII. oe ee fron from She: Botor Co., ela a
by G. P. MERRILL, ees ee a

Art, XXXII. Satine Share’ ‘Peoth. from the California
Pliocene; by D. S. JORDAN, : Pe a AES 2 = 338

Art. XX XIII.—An Upper Cambrian Fauna of ‘Packie Type :
in the European Arctic Region; by O. Hovrepaut,. Ras

Arr, XX XIV.—The Great Dustfall of March 19, 1920; by |
A. N. Wincue ty and EH, R. MitiEr,

Art. XXX V.—Helaletes Redefined; by E. L. TROXELL, .. : 36:

Arr. XXXVI. Beye iI a Lares Old World Migrants a
Dy M. R. Tuorbs, .-. 2

SCIENTIFIC IN TELLIGENOE,

Chemistry and Physics.—Hydrated Oxalic Acid as an Oxidimetric Stawdaite a
-"'B, Hinn and J. M. Smita: The Atomic Weight of Beryllium (Glucinum),
. Honrescumip and BIRKENBACH, 378.—An Introduction to the Physics

and Chemistry of Colloids, E. HatscurK: Distillation Principles and

Processes, S. Youne, 379. —Separation of Isotopes, F. W. Aston, 880.—

Newcomb- Engelmann Populare Astronomie, H. Los 381.—The

Two Orbit Theory of Radiation, F. H. BiezLow, 382. —

Geology and Mineralogy.—Ueber das Becken, den Schultergiirtel ie einige
andere Teile. der Londoner Archaeopteryx, B, PETRONIEVICcS, 382.—Die
Antike Tierwelt, O. KELLER: Origin and Evolution of the Human Race,
A. CHURCHWARD, 383.—The Topographic and Geological Survey of Pennsyl- |
vania, G. H. Asaigy: Carboniferous Glaciation of South Africa, A, L. pu
Torr: South Australia Geological Report for 1920, L. K. WaRD: ‘Mineral:
Production in the United States and Bette: 384.—The future of the
Comstock Lode, 385.—A new Meteoric Iron, G. P. Merritt, 386. ;

Miscellaneous Scientific Intelligence.—National Academy of Sciences: oe Text-
book of Zoology, 386.—Reptiles of the World, R. L.. DirmaRs, 387.—Mono-
graphia delle Cocciniglie italiane, G. Lronarpt: Fauna Hawaiiensis, D.
SHarp, 888.—World Atlas of Commercial Geology, 389.—The Friendly
Arctic, V. Steransson, 390.—The Evolution of Vlimates, M. Manson: An
Essay on the Physiology of Mind, F. X. DERcuM, 391, ae aa
History Museum at Brussels, 392. <

- Obituary.—C. Jonpan: B. Moorn: A. D. Watxer: T. Lizpison, 392,

Aubrary, U.S. Nat. Museum.

JUN E, 1922

Established by BENJAMIN SILLIMAN in 1818.

THE

‘AMERICAN |
‘JOURNAL OF SCIENCE.

Epiror: EDWARD S. DANA. |
| = : ASSOCIATE EDITORS |
ae - Prormssons WILLIAM M. DAVIS anp REGINALD A. DALY, |
ie oF CAMBRIDGE,

Bons:

a -Prorussons HORACE L. WELLS, CHARLES SCHUCHERT, | |
se ‘HERBERT E. GREGORY, WESLEY R. COE anp |
eid FREDERICK BH. BEACH, or New Haven, |
ee i: : a
a a s: PROFESSOR EDWARD W. BERRY, or Bautimore,

oR Drs. FREDERICK L. RANSOME anp WILLIAM BOWIE,
Bra oF WASHINGTON.

a

eee FIFTH SERIES

4) —s*YVOL~.«WI—[ WHOLE NUMBER, OCTII).

Mio 18-2 FUNB, 1922.

MOREHOUSE & TAYLOR CO., PRINTERS, 123 TEMPLE STREET,

Published monthly, Six tae per year, in advance. $6.40 to countries in the
osta Baas ; $6.25 to Canada. Single numbers 50 cents; No, 271, one dollar.
eT eK che as Naples matter at the Post Office at New Haven, Conn., under the Act

-“WILBY BOOKS

for All Geologists

Ready June 25th.

HANDBOOK OF THE PETROLEUM INDUSTRY.

By DAVID T. DAY, Ph.D., Editor-in-Chief, and 16 Expert Contributors.

By far the most comprehensive and authoritative book on all phases of
Petroleum ever written.

In Two Volumes—2070 pages—6 by 9— — profusely ilustrated flexibly bound,
$15.00.

Published March, 1922. ie
COAL—Its Properties, Analysis, Classification, ||

Geology, Extraction, Uses and Distribution

By ELwoop S. MOORE
Dean of the School of Mines, Pennsylvania State College.

Gives the salient facts and theories relating to coal—for the practical man,
the scientist and the layman.

462 pages—6 by 9—142 figures—cloth bound, $5.00,

Just Issued.

THE OIL ENCYCLOPEDIA

By MARCEL MITZAKIS,

Contains detailed particulars, in alphabetical Staon of every conceivable
subject connected with the oil world,

501 pages—6 by 9—cloth bound, $6.50.

FREE EXAMINATION OFFER*

Send for these books, or for any other Wiley publications,
on 10 days’ Free Examination terms. They may be returned,
postpaid, if not satisfactory.

* Applies to residents of the U.S. proper and Canada.

JOHN WILEY & SONS, Inc.

432 Fourth Avenue, New York

London: Chapman & Hall, Ltd.
Montreal, Canada: Renouf Publishing Company.

AJS 6.22

THE

AMERICAN JOURNAL OF SCIENCE

[FIFTH SERIES.]

Arr. XXXVII—A Critical Review of Chamberlin’s
Groundwork for the Study of Megadiastrophism; by
Wim EF. Jones.

Summary of Thesis.

The evidence of earth rigidity and elasticity is here
considered from the relative viewpoint and the conclusion
is reached that while the earth may be highly rigid under
tidal stress, that it is, beneath a superficial shell, in a
yield state for stresses of diastrophic dimensions.

The evidence of seismic transmission apparently indi-
eates a viscous liquid state for depths below 0.6 of the
earth’s radius for stresses of these dimensions, and the
conclusion is reached that such a state for larger stresses
will be nearer the surface.

Seismic transmission indicates an isotropically homo-
geneous state for the earth beneath a comparatively thin
heterogeneous outer shel! and this conclusion is in direct
opposition to Chamberlin’s postulated earth heteroge-
neity both in depth and laterally, and the evidence is taken
as indicating a density stratification of the earth as a
result of its having passed through a period of fusion.

The postulate of a continuously solid state for the
earth is shown to rest on certain assumptions as to size
of the earth nucleus and its rate of growth, and a solid
heterogeneous earth with vulcanism a result of selective
liquefaction is thought to be incompatible with the facts
of occurrence of the igneous rocks.

Hvidence is submitted to show that diastrophism can
be readily accounted for by adjustment or lack of adjust-
ment of a heterogeneous outer shell resting on an asthen-
osphere or yield zone, and does not require deep-seated

variations of density or deep-seated diastrophic move-
ments.

Am. Jour. Sci.—Firtu Series, Vou. III, No. 18.—Jvune, 1922.
28

~

394 Jones—Review of Chamberlin’s Groundwork

Introduction.

As a groundwork for the study of ‘‘megadiastro-
phism,’’ a term introduced as applying to diastrophism
on the large scale—continental or sub-oceanic, or even
applying to the earth as a whole, T. C. Chamberlin has
set forth in summary form in two recent papers’ the
salient deductions and conclusions he has drawn from
his studies of the formative processes.2, R. T. Chamber-
lin in a following paper? reviews the intimations of deep
deformation derived from structural cross-section studies
of two folded and deformed belts, the results of which
have been previously published.+*

It is evident that the study in megadiastrophism is to
be one of the application of wedge-dynamics to continental
areas. ‘To consider continental areas as the upper sur-
faces of deeply pointing wedges implies, of course, that
the earth lacks an asthenosphere or yield zone. This
postulate is fundamental. The issue is thus sharply
drawn between the two schools of thought. The basic
problems of diastrophism, isostasy, and vuleanism must
be attacked and interpreted in two diametrically opposite
ways by the two schools. The one school, as represented
by T. C. and R. T. Chamberlin, interpret these prob-
lems from the viewpoint of an earth which has maintained
perfect solidity through the formative eras and down to
the present; an earth, heterogeneous in its makeup both
laterally and in depth, in which, as R. T. Chamberlin
says,’ ‘‘all thought of easy movement in the interior is
to be scrupulously avoided.’’ The other school, as rep-
resented by Daly and the late Barrell, interpret these
problems, on the other hand, from the viewpoint of an

* Groundwork for the Study of Megadiastrophism, Pt. I. Summary state-
ment of the groundwork already laid. T. C. Chamberlin, Jour. Geology, vol.
29, pp. 391-416, 1921.

_ Groundwork for the Earth’s Diastrophism, T. C. Chamberlin, Bull. Geol.
Soc. America, vol. 32, pp. 197-210, 1921.

* Thirteen papers in recent volumes Journal of Geology.

* Groundwork for the Study of Megadiastrophism, Pt. II. The Intima-
tion of Shell Deformation, R. T. Chamberlin, Jour. Geology, vol. 29, pp.
416-425, 1921.

*R. T. Chamberlin: The Appalachian Folds of Central Pennsylvania, Jour.
Geology, vol. 18, pp. 228-251, 1910.

The Building of the Colorado Rockies, Jour. Geology, vol. 27, pp. 145-
164; 225-251, 1919. :

° Jour. Geology, vol. 29, p. 421, 1921.

for the Study of Megadiastrophism. 395

earth which passed through a stage of fusion resulting
in a coarse stratification by density and which, since
erustification, has possessed a yield zone beneath a com-
paratively thin outer shell. .This latter viewpoint is not,
as Daly has pointed out,’ incompatible with the basic
postulates of the planetesimal hypothesis of geogenesis.

The purpose of the present paper is to bring together
in summary form from the various sources the evidence
leading to the conclusions that the earth passed through
a molten stage and that there exists at no great depth a
zone which yields to diastrophie differential stresses.
The writer believes that this evidence, much of it direct
evidence of seismic transmission and close field studies,
eannot be lightly set aside.

The Rigidity and Elasticity of the Earth.

General Considerations.—The term ‘‘rigidity’’ has been
generally loosely interpreted. Rigidity is dependent not
only on the stress applied but upon the time the stress
is applied. A material may be rigid for one set of stress
conditions and non-rigid for another set. The earth is
spoken of as having a higher rigidity than steel. Another
equally good way of stating the same fact is to say that
the earth yields under the same conditions that steel
will yield and, as a matter of fact, steel does yield quite
readily. Kven Adams” and Bridgman’s® experiments
besides demonstrating the high strength of rocks under
cubic compression also show that these same rocks
eventually yield if the pressure is great enough and if
the time the pressure is applied is long enough. Rigidity,
then, is merely a relative term. Nothing that we know
of has absolute rigidity. And plasticity, or yielding, is
not necessarily dependent on the physical state of a
material. Crystals become plastic at sufficiently high

*R. A. Daly: The Planetesimal Hypothesis in Relation to the Earth, Sci.
Monthly, May, 1920, p. 495.

“F. D, Adams: An Experimental Contribution of the Question of Depth
7 Zone of Flow in the Harth’s Crust, Jour. Geology, vol. 20, pp. 97-118,

On.the Amount of Internal Friction developed in Rocks during Deforma-
tion and on the Relative Plasticity of Different Types of Rocks, Jour. Geol-
ogy, vol. 25, pp. 597-637, 1917.

°P. W. Bridgman: The Failure of Cavities in Crystals and Rocks under
Pressure, this Journal (4) vol. 45, pp. 243-268, 1918.

396 Jones—Review of Chamberlin’s Groundwork

pressures and so erystalline aggregates must become
plastie. 3

Temperature increase, of course, materially facilitates
yielding and so, as far as yielding by flowage is con-
cerned, and where temperature and pressure both
increase, a material will yield under a lower pressure than
it would if pressure were increased alone. The fact of
yielding implies differential pressures. True cubic com-
pression no longer exists when yielding takes place.
Certainly any terrestrial material we know of must be
in a potential yield-state under the conditions of temper-
ature and pressure at no very great depth beneath the
surface of the earth whatever its physical state as regards
homogeneity or heterogeneity, crystallinity or non-
crystallinity and yet such a yield-state would be one of
high rigidity in the ordinarily accepted sense of that
term, that is, for any short time differential stress.

The term elasticity may have the same implication. A
material may rebound under one set of stress conditions
and not rebound under another set of conditions. Both
these terms imply simply dependent states.

A material then, under any set of conditions as regards
temperature and pressure, may have properties of rigid-
ity and elasticity of a high order in so far as short-time
stress differences are concerned and yet be non-rigid and
non-elastic for stress differences of long duration and
large magnitude, and these latter are the types of stress
conditions we are concerned with in diastrophic move-
ments. ‘he statement that the earth is more rigid and
elastic than steel should be qualified, therefore, with the
statement that it possesses these properties for certain
stress conditions—for example those imposed by tidal
pulsations.

Now Chamberlin’s interpretation of the evidences of
this rigidity—that is the rigidity shown under tidal stress
and seismic vibrations, implies that this state of rigidity
and elasticity holds good for stress differences of all
magnitudes and of all time durations. Under this inter-
pretation the terms rigidity and elasticity become prac-
tically absolute instead of dependent. But such an
interpretation is essential if he is to postulate the non-
existence of a yield-state anywhere within the earth for
almost any condition of stress. Under this conception

for the Study of Megadiastrophism. 30%

yielding can only take place by shearing or recrystal-
lization. This is vital to deeply pointing wedge deforma-
tions.

Chamberlin deduces a holocrystalline state for the
earth on the basis of its rigidity, elasticity and density.
It is evident that both rigidity and elasticity are totally
independent of erystallinity. Amorphous materials, like
glass, or even paraffn wax, under sufficient pressure
become more rigid than steel under similar pressure.
Under differential stress of sufficient magnitude and dura-
tion such materials will flow because they are true liquids.
While it is true that high pressure favors crystallization,
since for most substances the crystalline form is the most
dense, this is true for conditions of normal temperature.
High pressure experiments seem to indicate that there
is a limiting set of conditions as to pressure and temper-
ature beyond which all directional molecular arrangement
is destroyed. The material then becomes a highly viscous
liquid. There is nothing to indicate that such limiting
conditions may not exist at no very great depth within
the earth and the resulting state might, and probablv
would, be one of ready yielding to long-imposed stresses
and still be a state of high rigidity and elasticity under
short-time stresses. For material to assume such fluid
properties for short-time stresses required conditions of
higher temperature and pressure. Seismic transmissions
seem to indicate that such conditions do exist at depths
below 0.6 of the earth’s radius.

The great depth of deformation deduced by Chamberlin
rests then upon conditions of rigidity and elasticity which
are not proved by the evidence at hand. And this is
vastly important, for this deduction makes it essential
that all earth body deformation be deep-seated and due
largely to tidal stress and its consequent kneading. The
work of Michelson® and others does not indicate in any
sense that the tidal stresses are sufficient in time duration
or magnitude to overcome the rigidity or elasticity attrib-
uted to the earth. Furthermore this same evidence does
not prove that stress differences of diastrophic dimen-
sions are not competent to overcome this same rigidity
and elasticity and so cause plastic yielding.

° A. A. Michelson & H. G. Gale: The Rigidity of the Earth. Jour. Geol-
ogy, vol. 22, pp. 97-130, 1914; vol. 27, pp. 585-601, 1919.

398 Jones—Review of Chamberlin’s Groundwork

Now temperature increase is so very much more effec-
tive in decreasing the density of material than pressure
increase is in increasing its density that it is unsafe to
assume that materials within the earth are in their most
dense or crystalline state, much less to assume that some
unknown compressional state exists. Again, so far as
high pressure experiments have extended, the density
' increase upon pressure increase becomes negligible. The
rigidity, however, progressively increases. ‘The increase
of density in the depths of the earth, it would seem, must
be independent of high pressure and so independent of
rigidity.

The Evidence of Seisnuc Vibrations.—Since records
of the transmission of seismic vibrations constitute the
only direct evidence we have of the state or states of the
earth’s interior an analysis of this evidence becomes
imperative. Seismic disturbances send out vibrations of
two types, compressional and distortional. These are
called the primary and secondary waves respectively.
The former waves are dependent on the elasticity or
compressibility of the transmitting medium, while the
latter waves are dependent both on the rigidity and the
elasticity of the transmitting medium, for their propa-
gation. The two types of waves travel at different veloci- _
ties but can only become distinctly separated out in a
homogeneous medium; that is, homogeneous as to stress
effects or, in other words, isotropic. The resultant vibra-
tions which travel circumferentially from the shock center
pass through what we know is a heterogeneous medium.
The wave types are not separated in the earth’s surficial
shell.

At a minimum distance of 700 miles, or 10 degrees of
arc, from the epicenter a three-phase record becomes
decipherable. ‘The first phase vibrations to be recorded
are the compressional or primary waves; next come the
second phase distortional or secondary waves, and both |
these phases come through the earth. Finally the third
phase, undifferentiated waves, reaches the recording
device via a circumferential path.

From the mere fact that the primary and secondary
waves are separated at distances above 10° of are from
the epicenter ‘‘it is clear,’’ according to Davison,’ ‘‘that -

°C. Davison: A Manual of Seismology, Cambridge, 1921, p. 149.

for the Study of M egadiastrophism. 399

the waves arriving there must, for part of their journey,
have traversed some homogeneous material situated
below a comparatively thin layer.’’ In fact, the rays of
these vibrations which emerge at the are distance of 10°
or 700 miles actually penetrate to a depth of only 100
kilometers. Here then is apparently direct proof of a
transmitting medium which is isotropically homogeneous
at a. comparatively shallow depth beneath the surface.

Now Chamberlin interprets this necessary and demon-
strated homogeneity as a medium possessing an average
homogeneity but a minute heterogeneity! Such an
interpretation becomes, of course, essential under the
hypothesis of deep crystallinity for the earth. If such
a minutely heterogeneous medium would have the identi-
eal transmitting properties as an actually isotropic one,
then one obstacle in the way of acceptance of the postulate
is removed, but are there not other obstacles?

The necessary confinement of large heterogeneity to
such a comparatively thin outer shell would seem to
present a formidable difficulty in the way of Chamberlin’s
postulated process of the introduction of tongue melts
from the deep interior to the more surficial portions of
the earth, for certainly these introduced heterogeneities
would not all insinuate themselves above such a shallow
depth as the transmitted seismic waves make necessary.

The important point deserving special emphasis is that
the splitting up of the two vibration phases at such
shallow depths as are penetrated by rays emerging at a
distance of 700 miles from the shock center indicate a
profound change in physical state at a very shallow depth
beneath the earth’s surface. The suddenness of this
change is further indicated by the apparent reflection of
transmitted waves against the surface of demarkation.
The most reasonable conclusion to be drawn is that this
sudden change is from a heterogeneous outer shell to a
non-crystalline isotropic material similar to ‘‘under-
cooled’’ liquids. Chamberlin’s statement!” that ‘‘the hot
interior of the earth is not a suitable environment for
under-cooled liquids,’’ does not take into account that
this physical state is the combined result of temperature
and pressure rather than of temperature alone.

“ T. C. Chamberlin: Jour. Geology, vol. 29, p. 395, 1921.
“T. C. Chamberlin: The Origin of the Earth, Chicago, 1916, p. 181.

400 Jones—Review of Chamberlin’s Groundwork

There is some disagreement in the interpretation of
the records of transmitted vibrations which penetrate
depths between the surface of demarkation as between
outer shell and inner homogeneous material, and a depth
of about half the earth’s radius. Knott’? in his work,
and Davison'‘ accepts his conclusions, shows an increas-
ing rate of velocity for both phases of waves down to
depths of 1,400 kilometers (.22R. R=earth’s radius) for
the secondary and 1,600 kilometers (.25R) for the primary
waves. Slight decreases in the rates of velocity increase
are indicated at a depth of about 750 kilometers. At
these greater depths the velocities reach their maximum
and from these points down slowly decline. Oldham on
the other hand, both in his paper of 1906'° and in his
discussion of 1918,'° coneludes that there is a steady
increase of the rate of transmission, with no marked break
in regularity down to a depth of about one-half the
earth’s radius. The interpretations by Knott and Davi-
son would imply a breakdown of the earth’s rigidity and
elasticity at .22R and .25R, respectively, while Oldham’s
interpretation would place the breakdown of these states
at greater depths.

All these authorities, however, agree that below a depth
of 3.825 kilometers (.6R) the earth body is non-rigid.
Below this depth the secondary waves are either blotted
out or, according to Oldham" ‘‘...... are no longer rep-
resented in their typical form, but are replaced by a
record of different character, probably not due to any
form of wave, which has followed the direct path from
the origin, and markedly delayed from the time at which
they should have arrived.’’.

The conclusions to be drawn from these facts are
obvious. On them rests the conclusion of Oldham!'® that
the material in the central nucleus has a very low degree
of rigidity even against stresses of only a few seconds

*%C. G. Knott: The Propagation of Harthquake Waves through the Earth,
Proc. Royal Soc. Edinburgh, vol. 39, pp. 157-208, 1919.

=O; Dawadsoni: 7 Ops citeyap alo

*R. D. Oldham: The Constitution of the Interior of the Earth as revealed
by Harthquakes, Geol. Soc. London, Quart. Jour., vol. 62, pp. 135-174, 1906.

*R. D. Oldham: The Constitution of the Earth’s Interior. Nature, vol.
102, pp. 235-236, 1918.

TOD Clits ypcoo.

OPC. azo

for the Study of Megadiastrophism. 401

duration, and Knott!® states that: ‘‘ Within this nucleus
of radius .4 the material of the earth has lost its elastic-
solid character and can transmit only compressional
(primary) waves’’ and ‘‘as the nucleus is approached the
material of the earth is becoming less of an elastic-solid
and more of an elastic, highly compressed liquid.’’

The evidence of a breakdown of rigidity in the earth’s
central portions is difficult to reconcile with Chamberlin’s
statement”? that rigidity increases ‘‘towards the center
faster than density........ all the way to the center.’’
‘«This evidence is thus directly opposed to the conception
of an earth core which is dense and rigid.’’ The core is
undoubtedly dense but its rigidity has vanished.

Now if rigidity breaks down no matter what the depth
for the stress differences of such short duration and
small amplitude as seismic waves impose, then the rigidity
and elasticity for stress differences of diastrophic dimen-
sions in time duration and magnitude must break down at
much more superficial depths. At least no experimental
evidence has yet been submitted which proves or even
implies that the material within the earth cannot yield,
under the long time differential stresses imposed by dias-
trophic movements, under the conditions of temperature
and pressure which exist at not a great depth.

Earth Yielding—An earth composed of erystalline
material could apparently only yield through the dynamic
influences of condensation. The adequacy of tidal
stresses is questionable. In such an earth the major
features of surface relief, such as continents and ocean
basins, could hardly be the direct result of diastrophism
but could exist only because of differences in density
extending to great depths as a result in turn of selective
growth with a less volume of dense material falling,
during growth, per unit area on the oceanic areas than
of less dense material on the continental areas.

It would seem evident that mere surficial loading or
underloading of any possible dimensions could not, in a
crystalline rigid earth, cause sinking or rising of the
earth’s surface. It is not even apparent how peneplana-
tion would be followed by uplift and yet without excep-

# Op. cit., p. 186.

*” 'T. C. Chamberlin: Jour. Geology, vol. 29, p. 395, 1921, and Bull. Geol.
Soc. Am., vol. 32, p. 224, 1921.

402 Jones—Review of Chamberlin’s Groundwork

tion such areal denudation has always been followed by
uplift of the denuded area. On the other hand, it is
probably true that accumulations of sediments do not
cause down-sagging though their weight may be additive.
In a large measure sedimentation goes on because of
down-sagging instead of down-sagging being the result
of sedimentation. The fundamental difference in the
effects of denudation and sedimentation is undoubtedly
due to the relative areas they affect. Denudation is not
constricted to narrow belts while sedimentation is, and
in spite of the probability of greater load per unit area
in depositional belts than underload per unit area in pene-
planed areas, these latter areas respond to the under-
loading and rise. It is a question then, not of underload
or overload per unit area, but of areal distribution of
load. The evidence back of this is conclusive. The
smaller, or narrower, the area the greater the load per
unit area necessary. to cause down-sinking. In a holo-
crystalline earth these vertical movements would be
entirely controlled, in so far as surficial loading could
cause movement, by load per unit area rather than areal
distribution of load. If these loads are the result of
greater or less density extending to great depths there
is no apparent reason why a broad area so loaded should
more easily sink or rise than a smaller area similarly
loaded. The mere fact of the control by areal distribu-
tion rather than by unit area load implies a solid crust
of some strength resting on a yield zone beneath.

Direct proof of ready yielding under not excessive load
per unit area but broadly distributed is to be found in
the very definite record of crustal sinking beneath the
Pleistocene ice caps and the rebound following deglacia-
tion. Furthermore, it is becoming increasingly evident
that this sinking was accompanied by a corresponding
upwarp! at some distance in front of the advancing ice.
The on-moving load, in this case, was actually ‘‘sliding
on thin ice.’’

Furthermore the universal association of down-sag and
upwarp through geologic history is difficult of explana-

*t Joseph Barrell: Movements of the Strand-line in Pleistocene and Post-
Pleistocene, this Journal (4) vol. 40, p. 138, 1914.

R. A. Daly: Oscillation of Level in Belts Peripheral to the Pleistocene
Ice Caps, Bull. Geol. Soc. Am., vol. 31, pp. 303-318, 1920.

for the Study of Megadiastrophism. 403

tion on any other assumption than that the earth is, in
a large sense, a yielding body.

Density Arrangement Within the Earth.—It is evident
that the material within the earth is more dense than
the visible material and it is also evident that the density
becomes greater with depth though whether this increase
continues to the earth’s center is problematical. If now
the earth is heterogeneous in its constitution from the
surface downward then increase in density would be
due to increasing pressure and the consequent condensa-
tion. This state, as postulated by Chamberlin, imphes
a chemical constitution throughout the earth, in general,
similar to the visible terranes. While he has admitted
the possibility of the accumulation of more dense material
near the center through the process of selective liquefac-
tion?? Chamberlin’s final conclusion is that ‘‘the metallic
theory fits only clumsily the growing evidence that the
interior elasticity and rigidity rise faster than density,’’
and again, ‘‘the tenor of accumulating evidence implies
that the conditions which affected the progressive self-
compression of the body were more important than those
which affected the original assortment of the material.’’**
The one-fifteenth volume ‘‘core’’ is given finally a density
stratification since it was concentrated, by hypothesis,
through a ‘‘gaseo-molten’’ line of descent and would
‘‘take on a more or less concentric structure similar to
that logically assigned to the whole earth by the old-
masters.’’?° The final picture is then of a central core,
one-fifteenth of the earth’s volume, stratified by density;
surrounded by a shell composed of crystalline hetero-
geneous material, the density of the whole mass being
due to self-compression.

The large percentage of metallic alloy in meteorites,
even taking into consideration the large number of stony
meteorites which are unnoticed, would imply a similarly
large percentage of metal in the earth, and if present,
and not concentrated towards the center, the surficial
earth material should contain more of it and have a

“'T. C. Chamberlin: - Origin of the Earth, Chicago, 1916, p. 181 and
p. 237.

*'T. C. Chamberlin: The Greater Earth, Bull. Geol. Soc. Am., vol. 32,
p. 224, 1921.

**Tbid., p. 223.

* Tbid., p. 224.

404. Jones—Review of Chamberlin’s Groundwork

higher density than is actually the case. As Daly has
pointed out?® the extrusive rocks should contain conspic-
uous proportions of the metallic alloys of low tempera-
ture fusion. Their absence from the visible extrusive
rocks implies their absence from whence these rocks ~
come. Hither then the magmas are derived from surficial
depths or the metals do not constitute even a small por-
tion of the earth. But this conclusion is not compatible
with meteoritic composition nor with the earth’s magnetic
properties.

Furthermore high pressure experiments indicate that
density increases so slowly with increased pressure that
even at normal temperatures the pressures within the
earth could not give to visible rocks the required density.
Again transmitted seismic waves imply a concentric
arrangement not only of physical state but of material.

Nor can deductions favoring increase of density by self
compression be safely drawn from a consideration of the
masses and densities of nearby planets. Chamberlin has
stated”" that as between the Moon, Mars, Venus, and the
earth there is ‘‘an accelerated rate of increase for each
increment of mass,’’ and that ‘‘this clearly implies that
their densities arise from their own massiveness.’’ This
implication would carry more weight if the densities of
the Moon and of Venus were more accurately known. In
so far as the densities are determined these, when plotted
against masses, do not fall on an increment curve. The
earth’s density is too low to even fall on a straight line
drawn through the Mars and Moon points, much less
showing an accelerated density increase. If the Earth,
Venus and Mars points are taken as points on an incre-
ment curve, then the Moon’s density is too low, and if the
earth, Venus and Moon points are taken then the density
of Mars is too high. The density of Venus is doubtful
so if the Karth, Mars and Moon points be taken as points
on a curve these three bodies show a decreasing rate of
increase of density for each increment of mass rather
than an increasing rate.

To say that ‘‘simple pressure experiments are incom-
petent to determine the limit of self-compression’’> and

*R. A. Daly: Igneous Rocks and Their Origin, New York, 1914, p. 161.

*7 Jour. Geology, vol. 29, p. 401, 1921.
2 'T. C. Chamberlin: Op. cit., p. 402.

for the Study of Megadiastrophism. 405

then to, on this basis, deduce a great compression reduces
the conclusion to an assumption. The reverse line of
reasoning was adopted when the determinations of earth
rigidity for seismic and tidal stress were taken to indicate
a sunilar rigidity under diastrophic stress conditions.

An earth coarsely stratified by density due to compost-
tion rather than compression more readily conforms to
the earth’s seismic and magnetic properties and can also
meet any requirements of rigidity placed upon it. Such
an arrangement implies a stage of fusion.

The Molten Stage.

Rate of Growth of the Earth—If the fundamental
postulates of the planetesimal hypothesis be accepted then
the question as to whether or not the earth passed through
a molten stage at its present size rests upon the assumed
size of the original earth nucleus and its rate of growth.
Now any conclusion as to the size of this nucleus is pure
assumption, but whatever its size at any rate, it passed
through a stage of fusion. In order then to secure and
maintain solidity through the period of growth only one
set of assumptions can be adopted and there are, first, a
nucleus of relatively small mass and, second, an exceed-
ingly slow rate of growth. These constitute the basic
foundations of the groundwork for the study of megadi-
astrophism.

The question of fusion is a question of heat generation
per unit time, or temperature. It hinges then not so
much on the size of the nucleus but on the rate of growth,
for this would largely determine the temperatures
attained. Rate of growth in turn depends largely upon
the orbital dynamics of the interplanetary dispersed
material and the ability of the earth nucleus to gather
it in.

Chamberlin pictures as the task set for the nucleus,
which he assumes to have a diameter of 6,000 miles, the
clearing up of a belt 55 by 58 by 592 million miles.?9
Gravitation is ignored and growth is assumed to be due
entirely to orbital conjunction. Can gravity be so
ignored?

» Op. cit., p. 409, 410.

406 = Jones—Review of Chamberlin’s Groundwork

The earth’s sphere of control is approximately 700,000
by 900,000 miles in cross-section area. For the nucleus
it would be smaller but still large. It would seem evident
that any smaller body passing within these limits would
either be entrapped or thrown into a satellic orbit unless
velocity gave to it a sufficient momentum to carry it
through. But by hypothesis the planetesimals were
‘(moving in the same general direction, at somewhat
similar speeds.’”° Relative velocities were thus materi-
ally reduced. It would seem that the orbital arrange-.
ments then would favor a quite rapid entrapment. The
task of the nucleus is thus materially reduced but even
approximate determination of growth rate is speculative
since the size of the nucleus is also speculative. There
is no direct evidence, however, to show that the growth
may not have been quite rapid.

Size of Planetesimals.—As Barrell has: pointed out?
the only direct evidence we have as to size of planetes-
imals is to be sought in the asteroids. These bodies range
in size and numbers from a few over several hundred
miles in diameter to many ten miles in diameter and
probably many more of still smaller size. The lunar pits
offer only questionable evidence, for the cause of these
features is not known. But if, as-Daly points out,?? the
pits are due to impact, then they constitute the record
of the last infall since they are not veneered over by
finer material.

If there was any great amount of interplanetary dis-
persed material doubtless much of it, or most of it, was
finely divided, but there is no evidence to show that there
were not, and there is some evidence to show that there
were, numerous bodies of sizes up to some hundreds of
miles in diameter. These larger bodies would avoid
entrapment the longest owing to their great momentum.
The infall may very possibly have been of a larger pro-
portion of finer material at first, with only occasional
larger bodies, and finally of a greater proportion of
larger bodies.

9° 'T, C. Chamberlin: Ibid., p. 40.

3! Joseph Barrell and Others: The Evolution of the Harth. Yale Univ.
Press, New Haven, Ct., 1920, p. 26.

2R. A. Daly: The Planetesimal Hypothesis in Relation to the Earth,
Scientific Monthly, May, 1920, p. 489.

for the Study of Megadiastrophism. 407

Heat of Impact.—The efficacy of the heat generated by
impact of planetesimals is doubted though probably the
additive effect if the larger sized bodies were pelted into
a molten or nearly molten surface would be greater than
if pelted onto a solid surface. The heat problem is more
concerned with the rate of growth rather than with size
of planetesimals and their impact effect.

Résumé.—tIn the speculative field of geogenesis there
seems to be no direct or conclusive evidence which forbids
belief in a molten stage for the earth at approximately
its present size. To assume that the original earth
nucleus was about the same size as the present planet
and was, by hypothesis, molten seems just as reasonable
as to assume that the nucleus was small and grew slowly.
But any conclusion in this field is rendered indefinite by
some pure assumption with no evidence back of it. More
direct evidence is to be sought in a consideration of the
present physical state of the earth, previously discussed
in its recorded diastrophic history, and in the field rela-
tionships of the igneous rocks. As Daly concludes :**
‘As critical facts are slowly accumulated, the greatly
appealing planetesimal hypothesis of Chamberlin and his
co-workers may well serve as a foundation for geological
philosophy, though the subsidiary doctrine of essential
crystallinity for the earth throughout geological time be
not accepted.”’

The Earth’s Diastrophism.

The evidence submitted by Chamberlin on the dias-
trophic capabilities of the earth as indicating a solid,
crystalline, and heterogeneous state throughout does not
seem conclusive and is susceptible of more than one inter-
pretation.

A “‘gaseo-molten’’ earth is pictured by Chamberlin as
wasting its dynamic energy in maintaining fluidity leaving
‘‘little possibility of shrinkage left except the meagre
amount that could arise from further cooling.’ But is
not this ‘‘meagre amount’’ sufficient to have produced
the visible diastrophic record? To attempt to trace a
pre-Archean diastrophism far greater in amount than all

*R. A. Daly: Op. cit., p. 495.

“'T. C. Chamberlin: Bull. Geol. Soc. Am., vol. 32, p- 199, 1921.

408 Jones—Review of Chamberlin’s Groundwork

of that recorded would seem to be unnecessary. But then
with the prior assumption of a crystalline heterogeneous
earth it becomes necessary, of course, to further assume
a very great and deep-seated diastrophism of long dura-
tion. Daly has clearly shown® that simple cooling may
be far less effective in volume change than that due to
changes in state, both from liquid to solid and from one
solid form to another as a result of static metamorphism.
These changes together with contraction due to secular
cooling would, it seems, be amply sufficient to cause all
the lateral deformation of which we have record.

The actual total circumferential shortening of the earth
since Proterozoic time, as recorded on any great circle,
is not over 200 miles and is probably nearer 100 miles.
If it is 150 miles the total radial shrinkage is then about
24 miles. The amount of shortening and shrinkage that
can be allotted to pre-Paleozoic time is, of course, indefi-
nite. It may not be as much as is generally supposed.
Structural deformation of at least Proterozoic terranes
is not generally excessive and the Archean rocks are
largely igneous, bespeaking great magmatic engulfment
and intrusion. Magmatic engulfment and intrusion on
such a large scale may well be potent factors in them-
selves in lateral deformation. Where these pre-Pale-
ozoic terranes do show isochinal folding they have usually
been involved in later deformation. At least their close
association with belts of later folding is significant.

If the earth has suffered any such radial shrinkage as
700 miles, as deduced by Chamberlin by the hazardous
method of density mass comparisons as between the earth
and its nearby neighbors,*® then a major part of this
shrinkage was taken up during the period of growth and
the resulting diastrophie record lies in depths forever
hidden from view.

If now diastrophism is the expression of deep-seated
inner reorganization due to condensation upon mass
increase, what is to provide the diastrophic energies after
the cessation of growth? Are we to conclude that this
inner reorganization in favor of greater density is to
continue without additional mass increase? Hither way
these questions are answered, it would seem that diastro-

” R. A. Daly: Igneous Rocks and their Origin, New York, 1914, p. 176.
* Jour. Geology, vol. 29, p. 400, 1921.

for the Study of Megadiastrophism. 409

phism should grow progressively less after cessation of
growth. But this conclusion is difficult to reconcile with
Cenozoic deformation which far transcends all other post-
Proterozoic orogeny. :

On the other hand, with a slowly thickening crust rest-
ing ona yield zone (asthenosphere), crustal yielding, both
in time and degree, is a factor of crustal strength.
Periodicity of deformations in crescendoes to a high point
indicate increasing crustal strength and so an increasing
capacity for accumulating stresses before the breaking
point is reached. Under this view the pre-Paleozoic
deformations represent the rather easy yielding of a com-
paratively thin crust and its widespread magmatic engulf-
ment. From the viewpoint of an earth whose diastrophiec
energies arise in the deep interior it is rather difficult to
explain periodic deformation, for the forces must be
continuously active and, since the surficial portions of
such an earth would be weaker than any underlying por-
tions, deformation could hardly result from accumulative
stresses.

Even if the postulated method of very slow growth by
planetesimal accretion with its consequent state of abso-
lute solidity for the earth be accepted, the existence of
any great reserve of diastrophic energy may be seriously
questioned. If this growth took three or four billion —
years it would seem that this would give ample time for
a very complete inner reorganization pari passu, leaving
an earth practically immune to deformational stress.

The Earth’s Volcanism and the Evidence of the igneous Rocks.

Volcanism, of course, using the term in its broadest
sense, finds little place in the groundwork for studies in
megadiastrophism. Itis considered as merely an ‘‘acces-
sory’’ to a ‘“‘profoundly metamorphic earth.’’ A once
molten earth would, according to Chamberlin, not only
waste its dynamic energy and become diastrophically
sterile but ‘‘there should have been developed and
brought to the surface al/** the gaseous material in the
earth substance which high heat could set free,’’ and
‘fan earth-body formed in this way should be unsuited

* The italics are mine—W. F. J.

Am. Jour. Sci.—Firta Series, Vou. III, No. 18.—Junez, 1922.
29

~

410 Jones—Review of Chamberlin’s Groundwork

to give rise to such a degree of gaseous volcanism as 1s
actually manifested.’’*®

The first statement quoted needs some qualification,
_ for the amount of gas set free is not only a question of
heat but also of pressure. Very evidently under the
pressure conditions existing at no great depth very
material quantities of gas would remain entrapped.

The second statement quoted also requires some dis-
cussion for it is doubtful if the major manifestations of
voleanism require any such great stores of gaseous
material as are implied. By far the larger part of vol-
canism has been expressed in fissure eruptions and deep-
seated intrusions. The mechanisms of both of these do
not call into use any mechanical force from gaseous
expansion. In fact, the greater lava floods are very often
free of vesicularity. Volcanic eruptivity of the central
vent type is not conspicuous in the igneous record of
pre-Tertiary time. Only in late-Tertiary and recent
times have the dying phases of an era of great vol-
canism expressed themselves in vent eruptivity on
what may possibly be called a worldwide seale.
Such is to be expected under the conception of a
slowly thickening crust. Furthermore, volcanic vent
eruptions call into use only localized concentrations of
the juvenile gases in their underlying diatremes or
cupolas. ,

The broad facts of the occurrence of igneous rock types
become extremely difficult of explanation under the con-
ception of an earth composed throughout of materia!
similar to the visible rocks from which magmas are
derived by deep-seated selective liquefaction. Chamber-
limes babes? 2s pees if magmas consist merely of partial
solutions of heterogeneous mixtures, they would quite
certainly become highly diversified in the making. The
primary problem would then lie in the generation of the
magmas; in the ascent of magmas rather than their
descent. While differentiation in the process of solidi-
fication would still remain a factor, it would be a second-
ary matter, in the sense that rt was necessarily condi-
tioned by the previous generative process.*° Each

* Bull. Geol. Soc. Am., vol. 32, p. 207, 1921.
0105 Glin, Os Zia
“© The italics are mine—W. F. J.

for the Study of Megadiastrophism. 411

particular case presents an a posteriori problem of its
own. It is wholly unembarrassed by any requirements
that its factors shall sum up into a speculative primitive
magma.”’ .

From this scheme of magma derivation two broad con-
clusions may be drawn: First, the tongue melts reaching
the surface should be, at least, within continental areas,
generally acidic in composition, since they are supposedly
derived from the selective liquefaction of the more fusible
constituents of an earth whose composition is similar to
the visible rocks and this composition is that of common
eranite; second, the acidity should decrease towards
basicity with time since the process implies the lique-
faction of less and less fusible constituents as interior
condensation progresses.

On the other hand the outstanding fact of the occur-
rence of the igneous rocks is the striking uniformity in
composition of the great extrusive flows, this similarity
extending in time from the most ancient extrusive masses
of Archean age down to the present, and in location world
wide. Diversification is absent. And the composition of
these extrusives is predominantly basaltic. Acidic extru-
sives are conspicuous by their almost complete absence.
Derivation of these great and persistent outpourings of
basaltic material from an essentially acidic earth is not
only difficult to explain by the process of selective lique-
faction but, on the other hand, if they are considered as
differentiates of a more acidic magma, then, as Daly has
pointed out :* ‘‘The acid pole of such a hypothetical split-
ting ought to be on a similarly large seale.’’ Nor can the
acid intrusive masses fill the need, in this case, of the
acidic derivative, for then a further explanation would
be necessary to account for the predominant intrusive
form of the acid and the predominant extrusive form of
the basic poles. Furthermore, the fact of discordancy
of the batholithic contacts together with the fact that the
structural delineaments of roof and wall rocks are merely
truncated at the contact surfaces rather than altered
implies that large masses of these rocks have been
engulfed and have added their high acidity to a previously
more basic magma.

RA, Daly? ‘Op. cit: p. 164:

412 Jones—Review of Chamberlin’s Groundwork

Bearings on Isostasy.

It is, of course, an evident fact as Chamberlin states*
that: “a liquid ‘earth should be an ideal example of
perfect isostasy i in the highest sense—that is, isostasy in
perfect horizontal as well as vertical adjustment, 77 ADs,
is the further statement that: ‘‘the earth’s crust, as it
began to form, should have inherited this quality in full
perfection’’** true in its implications? This statement
implies that any subsequent deformation of the crust
would merely act as an upsetting influence to this perfect
isostatic state. Now perfect isostatic adjustment in a
‘horizontal’’ direction does not mean, of course, perfect
leveling of the surface unless, of course, in the case of a
crustified earth, the crust is of uniform ‘composition and
thickness in a worldwide sense. Such an ideal state may
possibly have obtained immediately after crustification.
But this thin crust must have been very weak and readily
subject to disruption. Fragmentation would involve
engulfment, the outpouring of more basic materials and
their solidification would then introduce heterogeneities
of density in the crust and those portions so weighted
down would seek their level of adjustment. The broader
the area so weighted down, the more perfect would that
adjustment be. In fact, the suggestion has been made
by Willis and Barrell** that the ocean basins owe their
origin to such a process as this.

A state of perfect isostasy is, under this view, possible
only within the limitations imposed by the strength. of
the earth’s crust and that strength becomes small for
broadly distributed loads and great for more constricted
loads and so an almost perfect state of isostatic adjust-
ment must prevail as between oceanic depressions and -
continental areas. There is no antagonism here between
isostasy and diastrophism. The strength of the earth’s
crust merely interposes as a barrier to perfect adjust-
ment. Diastrophism is the expression of the final over-
throw of the crustal resistive power and this overthrow

(Op. Cit. py ceo: |

SiWdems p.5 209:

= Joseph Barrell & Others: The Evolution of the Earth, 1920, pp. 39-43:

Bailey Willis: The Discoidal Structure us the Lithosphere, Bull. Geol.
Soe. Am., vol. 31, p. 298, 1920. pepe aha

for the Study of Megadiastropmsm. 413

results in a more nearly perfect isostatic balance. Dias-
trophism is then the corrective. This is the view so ably
developed by Barrell in his monumental treatise on ‘‘The
Strength of the Harth’s Crust.’’#? Heterogeneities of
load whether due to erosion, deposition, or to the intro-
duction of intrusive or extrusive masses control, within
the limits imposed by crustal strength and distribution,
the degree of adjustment attained. These heterogenet-
ties are for the most part visible and determinate.
Beneath this heterogeneous layer, according to the evi-
dence both of the igneous rocks and transmitted seismic
vibrations, there les homogeneous material which is
capable of yielding under the larger differential stresses
to which it is subjected. Broad scale heterogeneities of
the surface shell must result in similarly broad scale and
generally permanent major reliefs. This view is based
upon what we can see. The opposing view—that of
Chamberlin—must be built upon an assumed asymmetric
earth core upon which was laid down, over a period
covering several billion years and through the whims of
the atmosphere, a selective rain of planetesimal dust.
This constitutes the foundation for the study of ‘‘megadi-
astrophism.’’ Is that foundation secure?

But the conception of an isostatic heterogeneous shell
resting upon a homogeneous yield zone does not, of
course, permit the application of deeply pointing wedge
deformation to areas of continental dimensions.

Massachusetts Institute of Technology,
Cambridge, Mass.

* Joseph Barrell: The Strength of the Harth’s Crust: Jour. Geology,
vols. 22, 23, 1914-1915.

414. Wells—Complex Chlorides containing Gold.

Arr. XXXVIII.—Some Complex Chlorides contauung
Gold. Ill. A New Cesitum-Auric Chloride; by Horacz
L. WELLS.

[Contribution from the Sheffield Chemical Laboratory of Yale University. ]

This salt, Cs.Au,Cl,,, was first obtained accidentally
by cooling a very concentrated solution of cesium chloride
containing comparatively little auric chloride. It may be
prepared from practically neutral solutions, but it is best
to use strong, or even concentrated, hydrochloric acid
as the solvent, since this acid makes the salt more stable,
so that, in its presence, far less concentrated solutions of
cesium chloride will give this red salt instead of the usual,
yellow double salt CsAnCl,.

The new compound forms very minute, deep red
erystals, which are quickly decomposed by water, usually
with the formation of the yellow double salt as a pre-
cipitate.

It appears to be the only known auric double chloride
not derived from hydrochloraurie acid, HAuCl,, and to

be a unique type among the double halides of the trivalent

metals in general. It was overlooked by Wells and
Wheeler,! who made an investigation of the cesium and
rubidium chloraurates and bromaurates, in this labor-
atory, and found only CsAuCl, and 2CsAuCl,.H,O as the
double cesium chlorides. Evidently they did not carry
their experiments to a sufficient excess and concentration
of cesium chloride.

Four crops of the new salt were analyzed at first.
They were prepared by cooling solutions containing in
each case about 180 g. of cesium chloride and 1 g. of gold,
as HAuCl,, in volumes of 300, 500, 700 and 500 ec., respec-
tively. The first three solutions contained a moderate
amount of hydrochloric acid, while the last one was nearly
neutral. The results of these analyses, where the crops
were not washed in any way and were dried on filter-
paper, in the air, and finally at 100°, were as follows:

Caleulated for
It, Jol; UGE IOV, Cs,Au,Cl,, Cs;AusCl,
CsCl ...49.16 48.75 48.72 heures 49 24 48.02

ANNIE =. BOLING 00.45 00.57 50.60 00.76 51.98

* This Journal, 44, 157, 1892.

Wells—Complex Chlorides containng Gold. 415

Although the agreement of the results with the first
formula is rather close, the second one is perhaps more
probable, since there is no doubt that the products were
contaminated to an appreciable extent with cesium chlo-
ride, on account of the method used in preparing them for
analysis, the high concentration of the mother-liquors,
and also because the erystals were of exceedingly small
size and thus presented very large surfaces for contam-
ination. 3

Another crop of the salt was then prepared from 200 g.
of cesium chloride, 1.5 g. of gold (as HAuCl,) and equal
volumes of concentrated hydrochloric acid and water,
making a total volume of about 900 cc. After the crop
had been formed by cooling in a counterpoised beaker, the
whole was weighed, and then, since the amount of gold
remaining in solution was comparatively very small, it
was possible to caleulate the amount of cesium chloride
corresponding to the volatile part of the mother-liquor.
The crystals were taken out, pressed with filter papers
very rapidly, so as to avoid evaporation as far as pos-
sible, and, without air-drying, the loss in weight at 100°
was determined. From this the amount of cesium chlo-
ride derived from the mother-liquor could be calculated.
The analysis of the dried product after correcting for the
cesium chloride derived from the mother-lquor is given
beyond (V), while a duplicate portion of the same crop
was analyzed also (VI). Analysis VII was made from
a crop obtained from a solution containing 1002. of
cesium chloride and 1 g. of gold (as chloride) in a volume
of 900 ce. made up of about two volumes of concentrated
hydrochloric acid to one of water. The product was
washed somewhat by diluting the last part of the mother-
liquor with an equal volume of concentrated hydrochloric
acid, without any apparent decomposition. This crop
was dried at 100° without determining the loss, but it
was assumed that a correction for cesium chloride about
one-fourth that of the previous crops should be applied
toit. The results were as follows:

Caleulated for

We: Vale: VBI Cs;Au;Cl,,
SO be a. Fie Ae Wt 47.81 47.84 47.61 48.02
dle lrtigs Bn Stes 50.98 50.96 o1.12 01.98

Corrected for CsCl, 1.13% 1.00% 0.25%

416 Wells—Complex Chlorides containing Gold.

It is to be noticed that the summations? of these three
analyses are about 1.2% low in each case. This is to be
explained by the fact that the products when dried at
100° were in a lumpy condition, and the additional cireum-
stance that cesium chloride when dissolved in hydro-
chloric acid, as experience has often shown, can be solidi-
fied and dried at 100° only with extreme slowness. Very
probably a rather stable acid cesium chloride is formed,
which is very hygroscopic, although cesium chloride itself
is not hygroscopic to any marked extent. It is believed
that the lumps of these products held a little of this liquid
even after drying for an hour or two, to practically
constant weight, at 100°. The possibility that the double
salt contains a molecule of water of crystallization, cor-
responding to 1.02%, has been considered, but this seems
very doubtful on account of the higher summations of the
other analyses where the products were pulverulent when
dried, and because of the additional circumstance that all
of the crops contained small amounts of filter-paper
fibers, which should give somewhat low summations in
any case.

From the facts that have been presented here, there
seems to be no doubt that the formula for this salt is
Cs,Au,Cl,,. This can be written 3CsAuCl,.2CsCl, but it
is entirely improbable that the yellow double salt retains
its identity in the more complex red one, on account of
the.remarkable change in color.

The new salt, in view of its very minute crystals, its
increased stability with hydrochloric acid, its striking
color, and its very sparing solubility in the solutions from
which it is deposited, appears to resemble the cesium-
auric triple salts that have been described in the preced-
ing articles of this: series, and it seems possible that it
may be a triple salt in the sense that the gold plays two
different parts in its structure, although the gold is
undoubtedly wholly in the trivalent state.

New Haven, Conn., March, 1922.

* These were unchanged by the corrections for CsCl.

7

Wells—Atoms in Inorganic Triple Salts. 417

Arr. XXXIX.—4A Chromophore Grouping of Atoms m
Inorganic Triple Salts, and a General Theory for the
' Cause of the Colors of Substances; by Horace L.
_ WELLS.

[Contribution from the Sheffield Chemical Laboratory of Yale University. ]

The intense black color and complete opacity of the
two salts

Cs,Ag.Au”.Cl,, and Cs,Au’,Au".Che
together with the less intense blackness of
CconCimAw. 3.6L.

which, although its minute crystals were very black, was
distinguished from the first two salts by yielding a pale
brown powder instead of a jet black one, while the salts

Cs,JnAu”.Cl,, and Cs,HeAu’’Cl,,!

were comparatively pale in color and were transparent,
attracted much attention, and an explanation of these
very remarkable deep colors has been sought.

In this connection attention was directed to Setter-
berg’s very intensely colored salt?

Cs,Sb’’Sb’Cl,.

which Wells and Metzger® in this laboratory showed to
be octahedral and isomorphous with Cs,Pb"Cl,, and which
has an exceedingly deep blue color, so that even very
small crystals of it appear absolutely black and opaque,
although the very fine powder, or the precipitate, shows
the deep color.

Upon comparing the two salts

Cs,Au’,Au”.CL. and Cs,Sb’”’Sb’Cl,.

it is to be observed that each of them contains atoms of
a metal in two states of valency. This is the main fea-
ture, requiring some modification on account of the colors
of the silver and cupric compounds already mentioned,
of the theory, now presented, of a chromophore grouping.

* All of these triple chlorides were described by the writer in the May,
1922, number of this Journal.
* Ofversigt K. Vetensk. Akad. Férhandl., 1882, 23.
* Amer. Chem. Jour., 26, 268.

418 Wells—Chromophore Grouping of Atoms

Another instance where such a grouping may be
regarded as the cause of a very deep color is in Ona Se
two triple bromides

KFe’Fe’”,.Br,.3H,O and RbFe”Fe’”’,Br,.3H,0.

These salts contain ferrous and ferric atoms and are
described as being dark green and quite opaque. It is
the writer’s recollection of them, which has been con-
firmed by a private communication from Professor
Walden, that the green color appeared to be merely a
luster and that the crystals seemed to be absolutely
opaque.

It appears, further, that the cuprous and cupric atoms
act as a chromophore, for when a hydrochloric acid solu-
tion of cuprous chloride, which is colorless when fully
reduced, for instance by heating with copper wire, is
mixed with the yellow or green solution of cupric chloride
in hydrochloric acid a very dark brown solution results,
which if fairly concentrated is opaque and practically
black. Little attention is paid to this dark colored com-
pound in the chemical reference-books, but in one of
them® it is stated that a triple salt, 6NaCl. Cu.Cl,.2CuCh,
has been prepared from it by Sievert. The composition
of this appears to be uncertain, since it was obtained only
in the form of an immiscible oil upon adding alcohol and
ether to a solution of the salts, and since when a larger
proportion of sodium chloride was used in its prepara-
tion a product was obtained containing twice as much of
that salt as is shown by the above formula. Perhaps the
dark hydrochloric acid solution that has been mentioned
contains a hydrogen-cuprous-cupric triple chloride, but
it may be a double salt.

Another case to which the chromophore grouping may
be applied is that of Prussian blue and the similar blue
products containing both ferrous and ferric atoms. The
salt KFe’’Fe” (CN), is recognized, so that the chromo-
phore effect may be regarded as taking place in a triple
salt in this case. Such formulas as Fe’”’,Fe”,(CN),,. and
Fe”, Me’’,(CN),. are frequently given to Prussian blue
and Turnbull’s blue, but that they are pure double salts

* This Journal, 48, 283, 1894. .
*Dammer, Handbuch anorgan. Chem., I1., 2, 671.

in Inorganic Triple Salts, etc. 419

seems doubtful, since they have been found to retain
potassium, ete., very tenaciously when washed.®

It is believed that the instances presented here, where
gold, antimony and copper as chlorides, together with
iron as bromides and cyanides, when present in two states
of valency in triple salts produce exceedingly strong
colors, are sufficient to show very conclusively that the
new theory in regard to this atomic grouping as a chromo-
phore is founded upon facts.

In regard to the very black salt Cs,Ag,Au’”’,Cl,., which
does not contain a metal in two states of oxidation, the
theory must evidently be modified by assuming that
atoms of two different metals may sometimes form this
chromophore. Possibly it may be essential that the two
metals should be closely related, as gold and silver are,
as is shown by their positions in Mendeleéff’s table. The
salt Cs,Cu” Au’”’,Cl,. may, perhaps, be regarded in pre-
cisely the same way as the one just considered, but its
color is not very intense, since it gives a pale powder, and
as our chromophore group appears to be a very powerful
one, it may be better to consider the color of this cupric
salt to be due simply to the ordinary coloring-effects of
eupriec and auric chlorides.

It is to be observed that Pollard’s salt’, (NH,),Ag.,Au,
Cl,;, has a silver-aurie grouping in a different ratio, but
that in this case the color is very dark red, without
opacity.

The discussion presented here has been given particu-
larly in connection with a chromophore grouping in triple
salts, but it is not supposed that the effect of this group-
ing is necessarily confined to these compounds. No cases
of this effect have been thought of in undoubtedly pure
double salts, but the black precipitate produced by ammo-
nia in solutions of mixed ferrous and ferric salts may be
regarded as an effect of the grouping upon a double
hydroxide or oxide, unless, indeed, this precipitate con-
tains ammonium hydroxide, or some kind of basic salt,
as a third constituent. At all events, it appears that the

* An extension summary of what is known in regard to the complex cyan-
ides of iron, with many references to the literature, is to be found in Beil-
steins Handbuch.

"This Journal, 3, 257, 1922.

420 Wells—Chromophore Grouping of Atoms

black color of the magnetic oxide of iron may be supposed
to be due to the effect of this grouping upon the color
of the combined red and so-called grey ferric and ferrous
oxides.

The chromophore grouping that has been presented
here is a very curious thing. In the cases of Au’-Au’”
and Sb’”’-Sb’ there is a difference of two units of valency,
while with Fe’-Fe” and Cu’-Cu” there is a difference of
only one unit, and, furthermore, these four pairs of valen-
cies are all different. It is particularly remarkable that
the three chlorides, CsCl, SbCl; and SbCl, which consti-
tute one of the examples of the chromophore grouping
are all of them colorless compounds.

It seems possible that the action of tais chromophore
may be explained by exchanges of negative electrons
between the atoms that differ in valency. It is supposed
that in passing from one valency to another an atom
gives up or takes on one or more electrons, and if it is
assumed that the atoms of a metal in two states of valency
in the same molecule, instead of retaining fixed individual
valencies, continually make these exchanges of electrons,
it may be supposed that light, passing through such mole-
cules, is in some way affected, so that colors or opacity
are produced.

The hypothesis of spontaneous electronic activity that
has been advanced to explain the behavior of the chromo-
phore grouping may be applied as a general cause of
color. It may be supposed that certain atoms, occurring
in such combinations that their structures are labile, may
be continually exchanging electrons with neighboring
atoms even when atoms of an element in two states of
valency are not present.

This supposition seems plausible, since it might be
expected that unless the activity of the electrons was
spontaneous it would not take place at all in bringing
about chemical combinations and changes in valency.

The common occurrence between atoms of spontaneous
electronic exchanges, of sufficient intensity to affect. the
passage of light is the theory advanced here to explain
colors. A strong argument in its favor is the fact that
a very large proportion of the elements that give at least
one colored salt, ion, or oxide are those that occur in more
than one condition of valency and whose atoms, conse-

in Inorganic Triple Salts, ete. 421

quently, may be regarded as particularly labile in their
structures and probably capable of giving off and taking
on electrons more readily than the others. There are
many of these elements, particularly among the metals
that form weak bases. It appears that there are a few
exceptions among the metals to this connection between
eolor and multiple valency, for cadmium gives a colored
oxide, but has, perhaps, no distinct second valency,
although two suboxides of it have been described, while,
on the other hand, arsenic and antimony give colorless
salts and oxides while showing two valencies in each case.

It is a very common occurrence that chromogenic atoms
give different colors, including absence of color, under
varying conditions of valency and chemical combination.
For instance, the cupric ion is blue, anhydrous cupric
chloride is brown, anhydrous cupric sulphate is colorless,
cupric oxide is black, cuprous oxide is red, and so on.

In order to explain such variations as these on the
basis of our theory it must be supposed that different
chemical combinations and different conditions of valency
modify the electronic behavior of the atoms in such a way
that their manner of exchanging electrons varies greatly.

There is a vast amount of knowledge in regard to the
molecular structures of organic coloring-matters and con-
cerning their various chromophore groups, but it appears
that the exact cause of the colors has not been satisfac-
torily explained. The present theory, however, if it is
a reliable one, will explain these colors, as well as others,
by the assumption of electronic exchanges and the result-
ing absorption of light. Nitrogen and carbon, under the
proper conditions of chemical combination may be sup-
posed to show electronic activity, and even oxygen may
do this, if our theory is correct, as is shown by the intense
blue color of liquid ozone.

According to our theory everything that is colored or
Opaque must possess spontaneous electronic activity, and
if it is supposed that the conduction of electricity is facili-
tated by, or is even entirely dependent upon, this activity,
we have a very satisfactory explanation of the well-recog-

nized generalization that opacity to light is favorable to
electric conductivity.

429 Wells—Atoms in Inorganic Triple Salts.

Summary.

The existence has been advocated of an inorganic
chromophore grouping consisting, usually, of atoms of
the same metal in different states of valency in a molecule
of a triple salt.

An explanation of the behavior of this chromophore
has been made by advancing the theory that there is a
constant exchange of negative electrons between the
atoms of different valency, and that this activity of elec-
trons affects the passage of hght, producing colors or
opacity.

An attempt has been made to extend this theory to an
explanation of colors of substances in general by assum-
ing commonly occurring spontaneous exchanges of elec-
trons, which affect the passage of lght.

The author takes pleasure in expressing his thanks to
his colleague, Professor John Zeleny, who has kindly read
this article j in its original form and has made some valu-
able suggestions, from a physicist’s point of view, which
have led to a considerable modification in the presentation
of the theoretical part of it.

New Haven, Conn., March, 1922.

Thorpe—Carnivora in Marsh Collection. 423

Art. XL.—Some Tertiary Carmvora im the Marsh
Collection, with Descriptions of New Forms; by Mat-
coum RurHERFORD THORPE.

[Contributions from the Othniel Charles Marsh Publication Fund, Peabody
Museum, Yale University, New Haven, Conn. |

TABLE OF CONTENTS.
Introduction.

Oligocene Canide.
Cynodictis angustidens (Marsh).
C. lippincottianus (Cope).
C. paterculus Matthew.
Daphenus vetus Leidy.
D. hartshornianus (Cope).

Miocene Canide.

Mesocyon robustus Matthew.
Nothocyon annectens Peterson.

N. vulpinus Matthew.

N. vulpinus coloradoénsis, mut. nov.
N. latidens multicuspis, subsp. nov.
NV. sp.

Cynodesmus cuspidatus, sp. nov.
Tephrocyon hippophagus Matthew and Cook.
T. marshi, sp. nov.

Leptocyon vafer (Leidy).
Amphicyon americanus Wortman,
A. sinapius Matthew.

Ailurodon sevus (Leidy).

A. near wheelerianus (Cope).

A. taxoides magnus, mut. nov.

Mustelide.

Brachypsalis pachycephalus Cope.
Potamotherium lycopotamicum (Cope).

Felide.

Felis augustus Leidy.

Felis sp. °
Pseudelurus intrepidus Leidy.

P. marshi, sp. nov.

Macherodus niobrarensis, sp. nov.
Felid, gen. et sp. indet.

References.

INTRODUCTION.

The present paper is mainly descriptive of certain
specimens of carnivores which furnish us with additional
information regarding the geologic fauna of North
America in the region between the Mississippi River and
the Rocky Mountains. Moreover, it adds quite materially

ADA Thorpe—Some Tertiary Carmvora im

to our knowledge of the time and space distribution of
the forms herein described. Some of the specimens form
the basis of new species, while others apparently afford
evidence of transitional stages between certain well
defined faunal landmarks. The majority of the genera
are confined either to Oligocene or Miocene strata, but
it is very probable that in some instances the Lower Pli-
ocene has contributed a small quota of forms, as will be
explained below.

The John Day carnivore material in the Maren Collec-
tion has been described separately, while nearly all of the
European specimens belong to well known genera and
species. ‘The drawings were made by Rudolf Weber.

The strata of Upper Miocene, and possibly some of
Lower Pliocene age along the Niobrara River in the
vicinity of Valentine are herein termed the Valentine
beds, a name proposed by Barbour and Cook in 1917 for
this horizon near Valentine, in Cherry County, Nebraska.
These beds are probably somewhat lower than those on
Snake Creek (not the Snake River of western Nebraska)
or in the Devil’s Gulch, and a list of the fauna, which has
been found in the Valentine beds, given by Barbour and
Cook, shows that it comes within the Procamelus-Hip-
parion zone as defined by Osborn in 1918.

There are at least fifteen other names which have been
apphed to this formation, some of which are faunal
names, while others are preoccupied. Osborn! desig-
nated these beds as the Fort Niobrara formation, and the
type locality is on the Niobrara River, near Fort Nio-
brara. Doctor W. D. Matthew has also adopted. this
usage. However, it seems to the writer that this name
is not happily chosen, for the custom in the usage of
geologic formation names is to shorten them when pos-
sible. It will be recalled that the Benton, Pierre, Bridger,
and other formations were originally termed Fort
Benton, Fort Pierre, Fort Bridger, ete. If the ‘‘Fort”’
should be dropped from Fort Niobrara, then the name
of this Upper Miocene formation would necessarily have
to be abandoned, as Niobrara formation is the name
applied to a subdivision of the Cretaceous. .

*H. F. Osborn, Equide of the Oligocene, Miocene and Pliocene of North
America, Tconographie type revision. Mem. Amer. Mus. Nat. Hist., new
ser., 2, pt..1, 23-24, 1918. .

the Marsh Collection, etc. 425

OLIGOCENE CANID2.

Cynodictis Bravard and Pomel.

With one exception, all of the American forms of this
genus are represented in the Yale Marsh Collection.
Some of these species are based on slight but usually
constant variations, although there may be some question
as to specific rank for them.

Cynodictis angustidens (Marsh).
Fie. 1.

Fie. 1—Cynodictis angustidens (Marsh). Holotype. x 3/2. A, occlusal
view of premolars; B, external view of right ramus.

In 1871, Professor Marsh described an ‘‘anterior por-
tion of a right lower jaw, containing the last three pre-
molars, and the canine’’ (p. 124), under the new specific
name Amphicyon angustidens. This species is identical
with Cope’s Canis gregarius, proposed in 1873, and,
therefore, has precedence over the latter, since both
belong within the genus Cynodictis.

The evidence, on which I base these conclusions, is
derived from the types themselves and from the deseri ip-
tions of them. Both types are lower jaws. Marsh’s C.
angustidens and Cope’s C. gregarius are considerably
smaller than the red fox; in both P, is one-rooted, P, has
median and basal lobes, "forming a cutting edge in ‘line,
the ramus is slender and deep, while the pr emolars are
low and compressed, P,_, have anterior basal tubercles,
while P,_, possess a posterior basal tubercle and cin-
gulum.

Am. Jour. Sci.—FirtH Serizs, Vou. III, No. 18.—JuneE, 1922.
2

v

496 Thorpe—Some Tertiary Carnwora in

Both the Marsh and Cope types were collected from
the same geologic horizon, the former at Scott’s Bluff,
Nebraska, and the latter in northeastern Colorado. —

The appended measurements fall well within the limits
given by Cope, Scott, and Matthew for C. gregarws.

This species was by far the most abundant of the genus
in America. It ranged from Colorado to South Dakota
inclusive, and probably its habitat extended as far west-
ward as the interior of Oregon. It is limited in time
between the base of the Middle Oligocene and that of the
Miocene. Scott (1898, p. 364) has described in detail the
osteology of the species, his descriptions being based
mainly, and probably wholly, on White River specimens.
Wortman and Matthew (1899, p. 122) amplified Scott’s
description, so that this form is quite well known. The
holotype, Cat. No. 11762, Y. P. M., was collected by Pro-
fessor Marsh in 1870.

Besides the type, this species is represented by some
very excellently preserved skulls and jaws, especially
Cat. Nos. 10067 and 10068, Y. P. M.; by a large number
of rami with dentition; and by certain skeletal elements.
These are all typical and add nothing to our knowledge
of this form.

Measurements.

C. angustidens C. gregarius
Holotype

mm. mm.
Leneth. of premolar series. eens 195 19*
Tueneth Of Pact. Fe ie eee 6 oy
Depth of ramus at sectorial........... 10 O*
Transverse diameter of crown of P,... 2.6 2.0
Heiohtsor crown Of Pin. eee ahs 3.8
Width’ ot. jaw below P..- nae ee 4.3 4.5
Depth-or jaw velow le... eee 10 9.6

* Measurements from holotype. Other measurements in this column are
from Cope’s drawings (1884) of specimens identified by him as Galecynus
gregarws.

Wortman and Matthew (1899, p. 124) give the skull
length as 76mm. and the maximum width of the brain-
case as 29 mm. Cat. No. 12678, Y. P.M., collected by
Mr. H. B. Sargent in 1870 at Scott’s Bluff, is 2 mm. longer
and 1 mm. wider in the same respective dimensions.

the Marsh Collection, etc. 427

Scott (1898) shows in his table on page 373 that his spec-
imens range in length from 86 to 92 mm. and in width
of brain-case from 31 to 35 mm. Nearly all of the Yale
specimens, as well as Cope’s, are within the limits of
Seott’s table. This shows a difference of from 13 to 21
per cent in skull length, and yet the length of the superior
and inferior tooth-rows, either premolar-molar series, or
each individually, does not differ by more than 1 mm. in
Y. P.M. Cat. No. 12678 (corresponding to Matthew’s
dimensions of No. 8774, A.M.N.H.) from the larger
specimens which fall within the lower limits of Scott’s
table. I can detect no differences, other than size,
between the smaller and the larger specimens, and the
ratios seem to be uniform. These smaller individuals
may represent a new variety, or they may be females.
YP. M No. 12678 is fully adult.

There are several Y. P. M. specimens in which M, pos-
sessed two roots, while others show but one root. Those
having the two-rooted M, do not appear to be any more
robust than the others. In fact, Y. P. M. No. 12691 has
two roots on this molar and yet it is of the same slender
proportions as the small skull and jaws bearing the
number 12678. I fail to find any other marked distinc-
tions between the two forms. Another specimen, No.
12687, from Scott’s Bluff, Nebraska, shows very distinctly
the alveolus of the double-rooted M,. A part of a ramus,
No. 12689, possesses a small posterior cusp, together with
the usual posterior and anterior basal tubercles on P,, a
character possessed by C. oregonensis Merriam.

Cynodictis lippincottianus (Cope).

Syn.: Canis lippincottianus Cope 1873B, p. 9; Galecynus lippincottianus
Cope 1884, p. 919; Amphicyon gracilis Leidy (non Pomel) 1856, p. 90 (nom.
preoc.); Daphenus gracilis Roger 1896, p. 44; C. hylactor Hay 1899, pp.
253-254.

This species, founded by Cope, was based on several
rami from: Colorado found in Middle Oligocene strata.
Cope stated (1884, p. 920) that the ‘‘Dimensions [were]
half as large again as in C. gregarius, as indicated by
Many specimens of the latter,’? while Wortman and
Matthew (1899, p. 180) found from the type that the
teeth were one fifth greater in lineal dimensions and

498 Thorpe—Some Tertiary Carnivora in

somewhat more robust. The Yale specimens agree with
Matthew’s determination and are represented by several
fragments of rami, chiefly Y. P. M. Nos. 12684 and 12689,
from Colorado. The skull is undescribed, unless Leidy’s
Amphicyon gracilis is considered synonymous. Matthew
holds this view and I thoroughly agree with it.

Cynodictis paterculus Matthew.

This species is represented in the Marsh Collection by
many specimens of rami, collected between and including
Pawnee Buttes, Colorado, and Crow Buttes, South
Dakota. The type, No. 9616, A. M. N. H., was collected
in Montana. It apparently is not represented in the John
Day fauna, but otherwise its distribution seems to be
practically the same as that of C. angustidens. The type
horizon is Lower Oligocene in the Titanotherium beds,
but I consider that its vertical distribution must include
a part of the Middle Oligocene (lower Brule). From
the matrix and locality of the Yale specimens, it seems
that the majority are from the lower Brule rather than
from the Titanotherium beds.

Y. P. M. No. 12683 has a small posterior cusp on P, as
in C. oregonensis. It corresponds to the type in other
respects. One of the specimens has a double-rooted M,
while the others have but the one. The specific char-
acters, as outlined by Matthew, are constant, and thereby
afford strong evidence for the validity of the species.
However, I can not help feeling that this form may repre-
sent the male of C. angustidens. The main distinction
between the two is that the former is somewhat more
robust. The size of both is about the same.

Daphenus vetus Leidy.

Various localities in Nebraska and Colorado yielded
remains of this species to the collectors working under
Professor Marsh’s direction. An especially well pre-
served skull and jaws (Cat. No. 10066, Y. P. M.), from
Greeley, Colorado, was figured by J. L. Wortman in 1901
in this Journal. Another specimen, collected by Doctor
Ki. L. Troxell in Sioux County, Nebraska, was purchased
by Professor Charles Schuchert and by him presented to

the Marsh Collection, ete. 429

the Peabody Museum. It consists of the major portion
of the skeleton, as well as the skull, of which the basi-
cranial area is very excellently preserved (Cat. No. 12771,
ied geal

Daphenus hartshormanus (Cope).

Mandibular rami from Pawnee Buttes, Colorado, and
from White River, Nebraska, represent this species in the
Marsh Collection, but add little to our knowledge of it.
The Nebraska form is, however, slightly smaller than the
type, while the Colorado specimens are practically
identical.

Both of the above species of Daphenus are Middle
Oligocene (lower Brule) in age.

MIOCENE CANIDAE.

Mesocyon robustus Matthew.

A part of a lower jaw with two premolars agrees well
with the type. It was collected at Gerry’s ranch, Colo-
rado, whereas the type locality is near the Rosebud reser-
vation in South Dakota.

Nothocyon annectens Peterson.

This species is represented by rami collected at Scott’s
Bluff and south of Antelope Creek, both in Nebraska.
That from the former locality is somewhat smaller than
the type; the other locality shows specimens exceedingly
close to the type. They, however, add nothing new to
our knowledge of the species.

Nothocyon vulpinus Matthew.

In 1873, Professor Marsh collected specimens along the
Niobrara River which are unquestionably referable to
Matthew’s species. The type locality is north of the
Niobrara in southern South Dakota.

430 Thorpe—Some Tertiary Carnivora in |

Nothocyon vulpmus coloradoensis, mut. nov.
(Fie. 2.)

Holotype, Cat. No. 12812, Y. P. M. Lower Miocene, Pawnee Buttes,
Colorado. .

The holotype consists. of part of a left ramus with P;,
P,, M,, and the alveolus of M,. It differs from N. vul-
pinus, its nearest ally, in having a relatively shorter jaw,
larger sectorial, relatively smaller tubercular molar, pre-
molars crowded, but with same individual antero-pos-
terior diameter as in N. vulpinus, while their anterior
cingulum is nearly obsolete. The sectorial has a small
cusp between the protoconid and hypoconid, which is, I
think, an individual variation.

Ui DoMy a
Gy Hy
= Cp OMIM nee

72812 TYPE Y. P. M.

Fic. 2.—Nothocyon vulpinus coloradoénsis, mut nov. Holotype. Nat.
size. A, occlusal view of premolars and first molar; B, external view of
left ramus.

Measurements.
N. coloradoénsis N. vulpinus
mm, mm.
Space occupied by P,P) and iki sa. 26.5 PAT ia
Diameters of M,, transverse............. D.0 4.8
Diameters of MU jant.-post. 925.25 ..2.0..).. 14 as
Maximum depth below WER 3.95.25. ..= 13.5 11

* Taken from drawing.

Nothocyon latidens multicuspis, subsp. nov.
(Fie. 3.)

Holotype, Cat. No. 12801, Y. P. M. Miocene, near Antelope Creek,
Nebraska. Collected in 1873 by Professor Marsh.

This specimen, consisting of part of a right ramus with
M., M,, alveolus of M, and part of that of P,, 1s somewhat

the Marsh Collection, ete. 431

larger than N. latidens, but possesses the narrow tubercle
on the external base of the protoconid which is a charac-
teristic of that species. It, however, lacks the small
tubercle just anterior to the base of the entoconid which
is often present in specimens from the John Day Valley,
Oregon, referred to Cope’s species. The paraconid is
the lar crest cusp of M,, with the protoconid and hypoconid
of about equal dimensions. The entoconid is small.
There is a small tubercle or cusp developed on the pos-
tero-external base of the protoconid. From this it will
be seen that the tooth patterns of both M, and M, are
quite similar in certain respects. M, was one-rooted.

FiGie3:

72807 TYPE

(Lr

AN {29

VA\ on - rN

2. P Gm
ke dpe

Fie. 3.—Nothocyon latidens multicuspis, subsp. nov. Holotype. x 3/2.
A, Occlusal view of molars; B, external view of right ramus.

In 1907, Matthew called attention to a form ‘‘approach-
ing N. latidens in size and characters’’ from the Lower
Miocene of South Dakota, which he considered would
prove to be a new species. The Yale specimen exceeds
NV. latidens in size, whereas I believe Matthew’s specimen
is smaller than the type of Cope’s species.

The type locality for N. latidens is in the John Day
Valley, Oregon, while this new subspecies is from
Nebraska. Furthermor e, Cope’s species is Upper Oligo-
cene (middle John Day) and the new form Lower or
possibly Middle Miocene. Unfortunately we can not be
positive either of its exact locality or geologic horizon.
The reason for this is that the Yale College Scientific

432 Thorpe—Some Tertiary Carnivora in

Expedition of 1873 travelled from North Platte to
Antelope Creek apparently without making any recorded
collections. J am inclined to believe that whatever mate-
rial was collected during the course of this long traverse
was boxed at Antelope Creek, where the first collecting
camp was established, and shipped east. Hence material
from lower horizons came from the Antelope Creek camp,
but was collected somewhere between there and North
Platte (city), Nebraska. :

Measurements of Holotypes.
N. multicuspis N. latidens

mm. mm,
IM; ant post. -ciameteio. excaie. in ane O38 8
M,, ant.-post. diameter of heel........... 4 3.0
M;. ant.-post. dtameter ass. sea an mi ae 5.2
Depth below middle of sectorial ......... 1 10:5

Nothocyon sp.
Cat. No. 12791, Y. P. M. Lower Miocene, near Scott’s Bluff, Nebraska.

Both rami and part of a maxilla indicate a form of
Nothocyon which does not readily fall within any of the
known species. The length of the tooth-row is very close
to that of N. vulpinus, but the depth of ramus below the
alveolar parapet is nearly equal to that of Mesocyon
robustus, that is, the mandible is considerably heavier and
deeper than that of any other species of this genus. The
depth below M, is 16 mm., the same as shown in the figure
of M. robustus, although in the table of measurements
for that species the depth is given as 6 mm., which is
undoubtedly a typographical error. The antero-pos-
terior diameter of M, is 1 mm. greater than that of N.
vulpinus.

Harold Cook (1909) records a large form of Nothocyon
discovered near the Agate Spring quarries in lower Har-
rison beds. His specimen is ‘‘somewhat larger and
heavier than N. geismarianus Cope,’’ and has a faint
cingulum encompassing the anterior part of the sectorial.
The Yale specimen also shows a slight cingulum in this
position.

the Marsh Collection, etc. 433

Measurements.
mm.

Lieneth of lower molar-premolar series................. D3
Meneih- oO dower premolar SETIGS 65.5. erie ee ee 253)
eeeaelie Oe okt Gab WE ere sks ete es Coe Se Cec eb we O2.0

Cynodesmus cuspidatus, sp. nov.
(Fies. 4-5.)

Holotype, Cat. No. 12788, Y. P. M. Upper Miocene (Valentine beds),
Niobrara River below Rapid (now Minnechaduza) Creek, Nebraska.

The holotype consists of parts of both maxille with
molars, P? and P*, the other premolars being represented
by parts of the crowns or alveoli. A fragment of a jaw
without teeth is provisionally referred to this species.
These specimens were collected by O. Harger in 1873.

Fic. 4.—Cynodesmus cuspidatus, sp. nov. Holotype. Nat. size. Palatal
view.

The species is intermediate in size between C. thooides
Scott and C. thomsoni Matthew, more closely resembling
the former in size and the latter in dental characters,
while the convexity above the anterior root of P* is very
much more prominent than in either of the described

434 Thorpe—Some Tertiary Carnwora m

species, and that above M' is weaker. The infra-orbital
foramen is above the posterior part of P® as in C. tho-
oides, the type of the genus.

This form differs from the two species previously
described in possessing an anterior tubercle on P*, as in
Atlurodon. However, it does not appear to invalidate
the generic reference, for Cams fanuliaris possesses a
well defined anterior tubercle on P*, while C. latrans has
not the least suggestion of one. P? possesses a posterior
tubercle similar to that of C. thomsom, and unlike the
corresponding tooth in C. thooides. The shear of the
carnassial is not so transverse as in C. thomsont, but
approximately the same as in the genoholotype. The
inner half of the superior molars is much broader than
that shown in the type.

WNC)
v

\ ag
i)

eS SS S&
2S EE S=
SSS S==S . S Ss
= SS = é Se = SS
als SSS Ano
: Sf = ae ) :
Ly “9 I) ) nt ) /
TT a\ Wi iy y
AW my \\: 1)

Fig. 5.—Cynodesmus cuspidatus, sp. nov. Holotype. Nat. size. Lateral
’ view of right maxillary with teeth.

This new species is from a horizon higher than that of
the others. From the similarity of tooth structure and
the individual peculiarities of the teeth exhibited by both
C. thomsoni and C. cuspidatus, sp. nov., I regard the latter
as a derivative of the former. The type of the genus
shows variations, which may be due to regional isolation,
but whatever the cause, these seem to indicate an aberrant
tendency on the part of the type. The South American
eanid, C. cancrivorus, probably shows the nearest
approach, among modern Canide, to the genus Cyno-
desmus. :

the Marsh Collection, ete. 435

Measurements.

C. cuspidatus C. thomsoni* C. thodidest

Holotype

mm. mm. mm.
Upper molar series, length...... 1g 16 ee2,
Upper premolar series, length... 38 31 40.5
meant -post. diameter ..... 2... 3.0 + 4
P* transverse diameter ........ 3 3 oe
P= ant=post. diameter ......-..: 8 feo 9
P*, transverse diameter ......... 4 + +
iP ant-=post:) diameter 2 bse. 9 8 9:5
P?, transverse diameter ......... 4.9 4.5 opal
amt. post, diameter 2)... e334 16.5 16 IES)
Po teansverse diameter ...3. 2... 29 10 al:
M?, ant.-post. diameter ......... Les 10.5 ia
M1, transverse diameter ........ 1s 13.7 16
Masami post. diameter ......... 8 5) 7
M?, transverse diameter ........ EE 6.4 10

* From Matthew 1907.
1 From Scott 1894.

Tephrocyon hippophagus Matthew and Cook.

Typical specimens of this species were collected in
Nebraska and Colorado. The skull is, however, not
known, although it must possess characters very similar
to that of 7. rurestris. On this basis an individual M}?,
Cat. No. 12789, Y. P. M., has been referred to this genus
and species, as it is nearly the same size and shows the
same characters as the John Day type of the genus. This
superior molar was collected on the Niobrara River,
between Antelope Creek and the mouth of Minnechaduza
Creek, Nebraska, by Capt. (later Brig.-Gen.) Mills, in
1873. Specimen No. 12833, Y. P. M., collected by Pro-
fessor Lull, in ‘‘Quarry D,’’ Ft. Niobrara Bird Reserva-
tion, on the Niobrara River, Nebraska, in 1914, has parts
of both maxille. The axial length of P*, M', and M? is
d3mm. There is also, among other bones, a scapholunar
which most probably belongs with the other parts. It is
very similar in size and characters to Canis latrans.

436 Thorpe—Some Tertiary Carniwora in

Tephrocyon marshi, sp. nov.
(Fie. 6.)

Holotype, Cat. No. 12787, Y. P. M. Upper Miocene (Valentine beds),
Cherry Co., Nebraska, along the Niobrara River, not far west of the mouth
of Minnechaduza Creek. Collected by Professor Marsh in 1873.

This species is represented by a nearly complete left
ramus with P,, P;, M,, and M,. It is nearly one half
_ larger lineally than 7. happophagus and in its analogous
parts seems to correspond most closely to the specimen
recorded by Matthew and Cook (1909, p. 376) from the
Snake Creek Pliocene. It differs, however, from all the
other species of the genus in that it shows the beginning
of a shortening of the ramus with a concomitant crowd-
ing, but not reduction, of the premolars, which overlap in
each case. P, 1s placed obliquely with the anterior part
outward, while P, was probably small, with a single root.
M, was set in the ascending ramus of the jaw, and in this
specimen is absent, with the alveolus partially closed,
resembling a specimen of A/lurodon haydent in the Amer-
ican Museum (No. 9744, Upper Miocene, Montana).

BiGeos
12787 IONAG28. YIPSVE

—~

=

Fie. 6—Tephrocyon marsh, sp. nov. Holotype. x 2/3. External
view of left ramus.

From T. moritfer this new species differs in being about
one quarter smaller lineally, less robust, and in the crowd-
ing of the premolars.

If, as we are disposed to believe, Tephrocyon is approx-
imately ancestral to Cams and 4/lurodon, then this new
species seems to be in the line from which A/lurodon
developed, for here we have the robust sectorial showing

the Marsh Collection, ete. ee

a slight increase in length and robustness with a jaw
ag The premolar reduction might well result
from this form in a few generations.

Measurements of Holotype.

mm.
Mier m@leminivamoe so Siac sce felt ce nets ase cee ee 81.5
Lo DISPETE TCIBEUATOY GW GSU EQS Eats asec eee es er eer 43.
LTTE LINOUIB TASES LAS (So aN ulnet e see ara er eae 41.

P,,, ant.-post. diameter 11. ; transverse diameter .......... (8

P,, ant.-post. diameter 13.2; transverse diameter .......... 8.5
P,, ant.-post. diameter 19. ; transverse diameter .......... KO!

M,, ant.-post. diameter 30.3 ; transverse diameter .......... 12.1
M.,, ant. “post. diameter 11.2; transverse diameter .......... 8.5
Depth of jaw DETECT NY Goes aaa amen ae Sele eae en is me eo cea 30.7
ispemolemawy WONG aE Piso. Gas accede ae wise ok eis oe Foace Zoek

This species may be referable to the genus T’omarctus
Cope 1873, and in fact it is possible that Tephrocyon may
be synonymous with Cope’s genus. The holotype of the
type species, Zomarctus brevirostris Cope, was collected
near Pawnee Buttes, Colorado, in the Middle Miocene
Pawnee Creek beds. It consists of ‘‘an immature jaw,
the carnassial about half emerged, and the anterior part
of the jaw so broken that it is not at all certain that the
premolars were, as Cope considered them, reduced in
number’’ (Cope-Matthew 1915, pl. CXIXc).

This type retains only the carnassial, but if the draw-
ing of this be correct, then it apparently bears a strong
resemblance to the analogous tooth in T'ephrocyon.

Leptocyon vafer (Leidy).
Several lower jaws, with teeth, in the Marsh Collection
are referred to this slender-jawed genus. The premolars
are compressed and not crowded, and the heel of M,

shows the low entoconid crest cent ely divided into two
cusps as stated by Matthew.

Amphicyon americanus Wortman.

The type of this species, Cat. No. 10061, Y. P. M., con-
sists of a palatal portion of a skull with the teeth, except
the incisors and first premolar. It was very fully
described by J. L. Wortman in 1901.

438 Thorpe—Some Tertiary Carnivora wm

Amphicyon sinapwus Matthew.
(Fies. 7, 8.)

In the collection there is a part of a right ramus with
the base of P,, nearly complete P,, and complete M, and
M., a detached crown of a lower canine and a practically
complete left superior canine. These bear the catalogue

Fig. 7.—Amphicyon sinapius Matthew. x 2/3. External view of part
of right ramus.

number 10010, Y. P. M., and were collected on the Nio-
brara River in 1875. The superior canine has the same
characters as that of the type of A. americanus.

There are three molars in the lower jaw, the first two
of which are very robust, while the premolars are rela-

Fic. 8—Amphicyon sinapius Matthew. 2/3. Occlusal view. of molars
and premolar.

tively considerably reduced as in the superior dentition.
The paraconid of M, is reduced, while the protoconid and
metaconid are robust and prominent. Of the heel, the
hypoconid is by far the more prominent element, for the
entoconid is marginal and much reduced. The robust M,
is composed of a large protoconid, a smaller paraconid

the Marsh Collection, etc. 439

and hypoeconid. The entoconid is represented simply by
a ridge.

An unworn M,, Cat. No. 12841, Y. P. M., collected on
the Niobrara River, near the mouth of Minnechaduza
Creek, exhibits nearly the same characters, except that
the entoconid is composed of two small marginal cusps
as in Leptocyon and the metaconid is a little smaller.

In size this species is probably about the same as the
grizzly bear (Ursus horribilis).

Measurements.

10010 18258*
Y¥. P.M, A.M.N.H.

mm. mm.

PeMnOPOSG. Giameter. 0. 5. e . Pees cs ek 15

Pieame Noss. Giameter ... 2.2 26. ee i es 195

PP teansverse diameter .2.:......-.... ee. 10

Wert NOSh. diameters co. ees he SOS 39

BP oeamsverse odiameter: ¢ 8s Ye i 17 17.3

Meant os. diameter. 2°... oP. oe eS 25 Bot

Meearansverse Glameter ..: ..6.... 2 oe es. 16 17.6

Depth of ramus below middle of M, ......... a2 O13

Length of inferior tooth row, P,-M,, incl. ....- 91 92.8

* The author is indebted to Doctor W. D. Matthew for his kindness in
sending these measurements to him.

Ailurodon sevus (Leidy).

A left lower jaw with milk premolars, Cat. No. 12817,
Y. P. M., is identified with this species. The first true
molar was not erupted, but the external mandibular wall
has been removed, thus exposing this tooth. The three
deciduous premolars are similar in the main characters
to those of Canis. The details of the crown of the fourth
premolar closely resemble that of the permanent first
lower molar, except in regard to size. All of the pre-
molars are two-rooted.

Ajlurodon near wheelerianus (Cope).

Cat. No. 10060, Y. P. M., consists of part of a left ramus
with P, and P, and part of M, and M, with the canine,
the alveolus of P, and part of that of I,, together with
a part of the maxilla bearing Pt and M', and two loose
teeth. It was collected on the Niobrara River, Nebraska.

The upper carnassial possesses a stout anterior tubercle

440 Thorpe—Some Tertiary Carnwora in

and the lower jaw corresponds most closely with the type
of A. wheelerianus, except that both P, and P, are set
obliquely, anterior end internal, in the mandible, and the
two premolars preserved do not possess anterior basal
cusps, as seen in No. 8307 (Cope Collection), A. M. N. H.
The crowns of all teeth in the specimen (supposedly the
type) figured by Cope in 1877 were broken away, so that
the presence or absence of anterior basal cusps on the
premolars can not be determined. Matthew and Gidley
(1904) have definitely said that all of the superior and
inferior premolars of this species had anterior basal
cusps. Both the type and the Yale specimen are of
Upper Miocene (Valentine beds) age.

Atlurodon taxoides lacks the anterior basal cusp on
P, and P,, but differs in its much larger size, in which it
approaches A. hayden.

Another lower jaw, Cat. No. 12785, Y. P. M., was also
collected on the Niobrara River. This right ramus is
somewhat smaller and much more slender, possibly being
that of afemale. The differences, however, are not suffi-
ciently great to invalidate the identification, in my
opinion.

Lilurodon taxoides magnus, mut. nov.
(Fias. 9-11.)

Holotype, Cat. No. 10057, Y. P. M. Upper Miocene (Valentine beds),
Niobrara River, a few miles east of the mouth of Antelope Creek, Nebraska.
Collected by HE. S. Lane in 1878.

The type material consists of both rami, with part of
the right side of the face, with complete superior dental
parapet, containing P', P?, P*, and M’, all others being
represented by alveoli, together with tie distal half of
the tibia and part of a cervical vertebra.

This new mutation is the nearest in size to A. taxoides
Hatcher, but differs from it in possessing prominent
anterior basal cusps on all of the premolars, both superior
and inferior, with the possible exception of P', the
anterior part of which is broken away. The presence
of these anterior basal cusps would seem to refer it to
A, wheelerianus, but such is apparently not the ease.
Both A. taxoides and this new form are approximately
20 per cent larger. Moreover, in the latter the alveolar

the Marsh Collection, ete. 441

parapet of the mandible rapidly ascends posterior to M,,
so that a line from the posterior edge of the alveolus of
M, to the base of the canine passes through the tip of the
protoconid of the sectorial; in A. tawoides and A. wheeler-
vanus, through the anterior base of the paraconid of M,.

Rig. 9.

VOGITAT CPE

oi eS

Seas
S SS 2
ASS

NS
Sy

WAY ee
SAN SS
Sy) 8 Vs

. Ss
S

SON

Fic. 9.—lurodon taxoides magnus, mut. nov. Holotype. «1/2. Right
lateral view of part of maxillary and premaxillary with teeth.

This upward trend of the parapet is analogous to that in
A. ursimus, which species has been referred to the Amphi-
cyonine group. In other respects, however, this new
individual is not hke Amplhicyon. Another differentia-
tion lies in the fact that the length of the inferior pre-
molar series of A. wheelerianus is less than that of the

» I ERS &

Fic. 10.—4lurodon taxoides magnus, mut. nov. Holotype. «1/2. Right
palatal view.

molar series, while in both A. taxoides and the Yale
specimen the premolar length exceeds the molar by one
fifth.

The superior canine was large and, in cross-section,
oval-shaped, and separated from the external incisor by
a considerable diastema, while on the other side it was in

Am. Jour. Sct.—FirtH Series, Vou. III, No. 18.—June, 1922.

449 Thorpe—Some Tertiary Carnwora in

contact with P'. From the external incisor the incisive
alveoli show a progressive reduction in transverse diam-
eter. M? is considerably reduced in size. The parastyle
on the upper carnassial is well developed.

The large infra-orbital foramen lies above the posterior
part of P?. The anterior zygomatic pedicle is very heavy,
measuring 41 mm. from the infra-orbital border to the
alveolar parapet, directly beneath.

In some respects this new mutation is similar to A.
platyrhinus Barbour and Cook, but the former differs in

JG. Wale

70057 DOPE

—
S==

iD V7) of 1 h A\\

h,
~ ‘ Hy) ' f
SS w Vy |
S < ‘ Mj Hi

‘

Gl

sl

Fig. 11.—4#lurodon taxoides magnus, mut. nov. Holotype. «1/2. A,
occlusal view of inferior dentition; B, external view of right ramus.

the following respects: (1) somewhat larger, (2) no
crowding of the premolars except that there is no
diastema between the first and the canine, (3) larger
canine, (4) carnassial over one fifth greater in transverse
diameter and M' smaller and of relatively less transverse
diameter, and in still others, the number of which would
undoubtedly be increased if we had more of the skull of
the Yale specimen or the mandible of the Nebraska indi-
vidual.

the Marsh Collection, ete. 443

Measurements of Holotypes.

A. taxoides A.magnus A. wheelerianus

mm. mm. mm.

Length of inf. premolar series .... 62 63 46
heneth of int-molar series... .... 53 oe 49
Ant.-post. diameter of sectorial .... 34 By 28
Aunepost. diameter of Ps. fons 22 2 15
Ant=post. diameter of M, ...-.... 12 14.5 12
moni post, diameter of M,> .. 0... 8 9 6
Depth of ramus below P, .22... 2. Oo” 36 29.)
Length of sup. dental series, C-M° :

DRG S558 5 Se a ae a nee 1H
Length of sup. premolar series .... 74
Leneth of sup. molar series ....... 24
Ant.-post. diameter of carnassial .. Bh PATS)

MUSTELIDZ.

Brachypsalis pachycephalus Cope.

This genus and species is represented by the posterior
part of a right ramus, except the coronoid process, pos-
sessing M, and M,, somewhat damaged. It is Cat. No.
12780, Y. P. M., and it was collected near the mouth of
Minnechaduza Creek on the Niobrara River, Nebraska.
It is somewhat more slender than the type, in this respect
approaching B. modicus Matthew, although in other
essentials it is identifiable with Cope’s species, represent-
ing a possible variety of the latter form.

Another specimen, a right ramus, Cat. No. 12803,
Y. P. M., is more typical than No. 12780. Professor B. F.
Mudge collected the jaw near Ellis, Kansas, in the
Ogalalla formation, which is not later in age than Lower
Pliocene and may well be of late Upper Miocene. This
ramus shows characters intermediate between that of
B. modicus and B. obliquidens Sinelair, although its mod-
ifications are much less extreme than those shown by
Sinclair’s species. The length of the tooth-row and of
the individual teeth is a little greater than in the type
of the genus.

444 Thorpe—Some Tertiary Carniwora wm

Measurements.
Cat. No
12803
Ne PME Holotype
mm. mm.
Molar-premolar series, leneth ............-. a8 Sys)
Premolarcseries, lemot heater od paneer re eee BZ 31
Py, ant..0st. dtanvebenze: \pcae wpucers ecetemer eet 7
Pant post, diam eb eee ee erie 9.9
Py ant--p0st= GiaMme tery sae eee eee ae be we his
IM,, ant.-DOSt? diameter @. 14a ee ue Ease 14.5
M5, ant:-post. diameter... o5 een. ey eres 9
Depth of ramus at sectortalssun, fe ae 25 Ze

* Alveolar measurements.

Potamotherium lycopotamicum (Cope).

The type of this species, from the Mascall beds of
Oregon, has unfortunately been lost. It was based on a
jaw broken away immediately behind the carnassial, and
belonged to an animal about the size of a mink. The
type was figured in 1915 (Cope-Matthew). _

Two lower jaws, collected on the Niobrara River, are
referred to this species. One, from ‘‘Quarry D,’’ Ft.
Niobrara Bird Reservation, Nebraska, Cat. No. 12834,
Y. P. M., found by Professor Lull in 1914, has the carnas-
sial unworn and intact, as well as the premolar alveolar
parapet, with the ramus below intact. It also possesses
the alveolus of M,, the presence or absence of which in
this species has not before been determined. All of the
premolars are two-rooted except the first. The other

Measurements.

Cat. No. Cat. No.
12834 12825
Y.P.M. Y. P.M. Shloletype

; mm. mm. mm.
Molar-premolar series, exclusive of M,.. 22 EPs
Ms) amt. <post. diameter: sexes nese | 6.6
M,, transverse diameter .............. 3.8 4
Mi jlenetin vot eels, akira ec fo acts 2.4 2.4
P 7, ant =p ost: diameter peeves ot .ca0- ee 4.2 4.5
P., ant.- post: diameterm sen iio ok eee ee 3.8
PZ vant.-postadtame tion asm.) mo nceenes 3.4
Depth of ramus below sectorial ....... si

the Marsh Collection, ete. 445

specimen, Cat. No. 12825, Y. P. M., is-one of the results
of the Yale College Scientific Expedition of 1873, under
the direction of Professor Marsh. It is a portion of the
left ramus with P,, P, and P,, with the alveolus of P,
and the canine. The fourth premolar has a low posterior
cusp and prominent cingulum posteriorly. There is an
alveolus for one incisor.

ELIDA.

Material referable to the cat family is extremely meagre
in the Upper Miocene and Lower Pliocene, and it is with
this in view that I desire to call attention to some of this
material in the Yale collections.

Felis augustus Leidy.

The distal end of the humerus of a large felid was
collected by Professor Lull on the Niobrara River, in
Cherry County, Nebraska. It is a well preserved frag-
ment, somewhat over 5 inches in length, bearing catalogue
number 12809. The humerus exceeds in size that of Felis
leo and is considerably more robust. Felis augustus
must have been very close in size and probably in pro-
portions to the Bengal tiger, which in turn is larger than
the average male lion.

The lower half of another humerus, better preserved
and 9 inches in its present length, was collected by F. W.
Darby in Sheridan County, Nebraska, also near the Nio-
brara River. This humerus, Cat. No. 12810, Y. P. M.,
has practically the same dimensions as that of the type,
while No. 12809 is about one twelfth smaller.

Felis sp.

A seapholunar and a median phalanx, from the Upper
Miocene on the Niobrara River, near Fort Niobrara,
Nebraska, indicate the presence of a felid or macherodont
of about the size of Daphenodon superbus (Peterson).
The scapholunar, Cat. No. 12838, is flatter and less oblong
than that of Felis leo and the articular surfaces are in
some cases slightly divergent. The resemblance, at any
rate, is much closer to the cats, than to any of the other
carnivores.

446 Thorpe—Some Tertiary Carnwora im

Pseudelurus wntrepidus Leidy.

Two specimens of this genus and species, found on the
Niobrara River in Cherry County, Nebraska, exceed the
type in size. The difference is not of specific value, but
worthy of notice, as heretofore the type of the species
has been regarded as representing the maximum size to
which this genus attained in America. The posterior
half of a left ramus, Cat. No. 12830, Y. P. M., was col-
lected by Professor Lull in 1914. The other specimen,
Cat. No. 12816, Y. P. M., consists of a third lower pre-
molar, which is 2 mm. higher, 1.5.mm. wider, and 2 mm.
greater in antero-posterior diameter than Leidy’s type.

Pseudelurus marsh, sp. nov.
(Fie. 12.)

Holotype, Cat. No. 12865, Y. P. M. Upper Miocene (Valentine beds),
Niobrara River, near.mouth of Minnechaduza Creek, Cherry Co., Nebraska.
Collected by Professor O. C. Marsh in 1873. Paratype, Cat. No. 12815,
Y. P. M. Middle or Upper Miocene, northwestern Colorado.

The holotype consists of both lower jaws, partially
restored, and the paratype of a left ramus with the
alveolar parapet complete, as well as P; and P,. The
new species is about the size of Hoplophoneus prumevus,
or one fourth smaller than P. intrepidus and hence con-
siderably smaller than P. wntrepidus sinclairi Matthew.
The mandible is very much more slender than either of
the above forms and does not exhibit so great depth below
the tooth-row, although both the holo- and paratype are
fully adult. The teeth are not crowded as in the above
mentioned variety. Both types of the new species pos-
sessed a very diminutive P., the antero-posterior diameter
of which is not over 2mm. ‘The position of this premolar
is inward from the true line of the tooth-row, as in
Matthew’s variety. The posterior basal tubercle of P;
in the paratype is exceedingly small and that of the holo-
type is much reduced. The fourth premolar, in its com-
ponent parts, most closely resembles the analogous tooth
in P. imtrepidus sinclairt, while the metaconid of M, has
not suffered so great a reduction as in the latter variety.
The paratype alone shows a distinct, but very small
alveolus for M,, its antero-posterior diameter being 3 mm.

the Marsh Collection, ete. 447

The mental foramina have approximately the same posi-
tion as in Matthew’s felid.

The Huropean species P. quadridentatus (Blainville)
is shghtly longer than P. intrepidus, but is of approxi-
mately the same depth below the tooth-row. Pseudelurus

iY
SUL iy
i, FG

;

<lbgp

AM

Fic. 12.—Pseudelurus marshi, sp. nov. Holotype. Nat. size. A, occlusal
view of inferior dentition; B, external view of right ramus.

marshi, sp. nov., n size corresponds more closely to P.
edwardsu Filhol, although the latter is somewhat smaller.
Pseudelurus intermedius Filhol possesses a minute inner
cusp on M, which is not found on any American form,
so far as | am aware.

Measurements of Holotypes.
P.marshi P.intrepidus

mm. mm.
Leneth, condyle to median incisor ......... 97 Ieee i
Beneth.ot tooth<series: (Mis P and. P.) 20%: « 36 44.5
Length of diastema between C and P;....... 8.6 14.8
Pulte OSu senaIme ter Ol Me eet oii ea ag ee 8 16 19.6
PT POSh CUIICbeL Gin bg mm isarse (AEs ase 2 s+ > 12.5 14.8
eM OSE sOAMCLeTOOia bers Megat. ee 5, 11.6
Whtdbh of jaw -alisectenldive % ces ek. 9
Wepihtatyawat Po Aerts se She Se . 17) 23.2

448 Thorpe—Some Tertiary Carmvora m

Macherodus niobrarensis, sp. nov.
(Fie. 13.)

Holotype, Cat. No. 12829, Y. P. M. Upper Miocene (Valentine beds),
Niobrara River, Cherry Co., Nebraska.

The portion of a skull, anterior to P*, collected in 1873,
is tentatively referred to Macherodus, in spite of the
fact that no undoubted specimen of this genus has been
hitherto found in North America. This specimen does
not by any means dispel doubt as to the presence of this

TCCRY ikke

Fie. 13.—Macherodus niobrarensis, sp. nov. Holotype. «3/4. A, left
lateral view of part of maxillary and premaxillary; B, palatal view of same.

genus in the New World, but it does show that there were
forms present here which were more nearly allied to
Macherodus than to any other genus of felids.

One of the outstanding and diagnostic characters of the
specimen is seen in a cross-section of the canine, this
tooth being very much compressed, and having an antero-
posterior diameter of 21.5 mm. and a maximum trans-
verse diameter of 8.8 mm., measured at the alveolar

the Marsh Collection, etc. 449

parapet. The ratio between the transverse and antero-
posterior diameters, measured at the base of the crown,
is as follows in the forms listed below: Felis leo, 1 to 1.4,
Pseudelurus, 1 to 1.7, in the true Felis line; while in the
macherodont series, W. necator (based on Cope’s figure),
1 to 2.2, M. palmidens (Blainville), 1 to 2.6, M. mobra-
rensis, sp. nov., 1 to 2.4, and Smilodon neogeus Lund
(from Burmeister’s measurements), 1 to 2.53. MW. cras-
sidens may be nearer to the felines as indicated by the
ratio of 1 to 1.65. Hence it is seen at once that the pro-
portions of the canine in the felid and macherodont series
are very different, and that this new macherodont is
much closer to M Per odus than to any form in the feline
series of equivalent or later geologic age.

Macherodus is very br achycephalie and this new
species shows the same character. The anterior part of
the alveolus of P*® is external to a plane, parallel to the
sagittal plane, passing through the outermost part of
the canine. In the Yale specimen the anterior part of this
premolar is offset approximately 6mm.; in MW. palmidens,
about 4 mm.

This new species differs from the typical Macherodus,
in so far as we have comparable parts from which to
judge, in that the incisors are a little more anterior to
the canines, thereby making the muzzle slightly more
pointed. Another divergence is seen in the incisor pro-
portions, the external being larger and the amount of
reduction from this one oreater than i in WM. palmidens.

The skull of M. palmidens is a trifle greater than one
half that of Felis leo, while this new species is probably
about two thirds the size of the lion and hence smaller
than any of the specimens hitherto referred to this genus
from North America.

Measurements.
Cat. No. 12829

M. palmidens W...P. M.
mm. mm.
Width of palate at and including canines.. 47 54
Width of palate at anterior part of P? .... 60 ies
Length of diastema between C and P® . fi 115

Length (axial) from prosthion to line
PFN FH POSt. Of CHMIMeSl: A ek ra | 38

450 Thorpe—Some Tertiary Carmvora im

Felid, gen. et sp. indet.

Another specimen, Cat. No. 12839, Y. P. M., is that of
an undoubted felid, but whether a member of the
Macherodus or Felis line is-not determined. As eat
material from this horizon is so very scanty, in fact,
largely unknown, it will not be out of place to give a
brief description of certain bones of this skeleton. No
skull is present, but several vertebra, part of the pelvis,
ribs, and the major-part of the left femur are extant. It
was discovered in 1914 by Professor Lull and collected by
him, with the assistance of F’. W. Darby, 5 miles east of
Valentine, Nebraska. For comparison, an average-sized
skeleton (Cat. No. 01050, Y. P.M.) of a male Felis leo
was used. The geologic horizon is either Upper Miocene
or possibly Lower Pliocene, as the specimen was found at
approximately the same level as a 4-tusked mastodon,
genus not yet determined, collected by the same party.
The fossil skeleton, in the anterior portion of its anatomy,
is apparently lighter than the lion, but in the posterior —
region it is fully as heavy, or possibly a little stouter and
more robust.

The transverse processes of the atlas are slightly
thicker, of less width, but greater posterior extent; the
width of the dorsal surface is considerably greater; and
the posterior opening of the passage for the vertebral
artery is absolutely smaller than in the lion. The hypa-
pophysial tubercle is apparently reduced, while the neural
Spine is very rudimentary. The alar canal is bridged
over, and is not in the form of a notch as in Felis leo.

Measurements of Atlas.
Cat. No. 12839 Cat. No. 01050
Yy PON

eae 2k, ON Te
Max. widthsacross wanes: ermriem sss one 132 137.5
Ant.-post. diameter across articular sur-
ECCS. cy sce eho attaee ee nee ee a9 64
Ant.-post. diameter of neural arch, dorsal
STO 4.7 hess shaken cee eee eM oroe a cee 43 . 29

Ant.-post. diameter of neural arch, ven-
tial (Sid @si 5 ce ae oe Rea ees or ee 17 24

the Marsh Collection, etc. 451

The fifth and sixth cervicals, transversely, have a more
nearly square outline to the neural canal; the walls of the
neural arch are heavier; the vertebral canal 1s more
elongate; the anterior and posterior faces of the centra
are much more coneave; the postzygapophyses stand at
a slightly greater vertical angle than in the hon. The
transverse processes are directed downward and out-
ward; in F’. leo outward and downward. ‘The sixth lacks
the upper transverse process of the hon. The seventh
closely resembles that of the hon except that the costal
facets are large and well defined in the fossil and the
ends of the centrum are more concave.

Measurements of Cervicals.

5th 6th 7th
12839 01050 12839 01050 128389 01050
mm. mim. mm.
Leneth of centrum ......... Done web Boe OU 30 1530

Width across prezygapophyses 43. 74 O20 200 D2.3: 66
Width across postzygapophy-
SES See eee Rg R AS tes Sle 60) 48 61

Dorsals.—The first is very similar to that of F’. leo
except that it is slightly smaller and less robust; the
fourth has a more oval (flattened from above downward)
neural canal, less robust mammillary processes and is .
in general smaller and Lighter than that of the lion. Its
posterior articular surfaces extend but slightly beyond
the centrum, while in F’. leo they extend quite prominently.
In the eleventh dorsal, the posterior articular surfaces
are more vertical, with the superior section vertical, while
in the hon the upper part bends outward, that is, has a
more horizontal position. In the lion this vertebra has
a vestigial spine, while the fossil possesses a relatively
enormous one. The spine is wider, heavier, and longer
in the twelfth; the centrum narrower, while the meta-
pophyses of the prezygapophyses extend outward, are
heavier and more robust, and the anapophyses extend to
a line through the posterior margin of the postzygapo-
physes in the fossil. In the thirteenth, the metapophyses
extend upward and outward; the anapophyses extend
slightly beyond the posterior margin of the postzygapo-
physes; and the facets for the rib heads are deeper in
this and the twelfth than in Felis leo.

452 Thorpe—Some Tertiary Carnivora m

Measurements of Dorsals.

12839 01050
mm. mm

Length of centrum

Tet) 22a. Bae a lee ete ae 28 30

Atta 2h 0p SS. a ae ee eit peter cc, Bo. 26.5 ol

PUGH A AEE Aes <r 30.5 37.0

Mths be heaies he. Sate ey eee eee 33 40

USt his js. B4UAs i cce ee A oh ee eee 30 43
Length of spine

ALS Gs Oe RS ae Oe ie er ee Mig ae bee ELA Be 65 76

Atl Paes BRL Ae Ae eee 7C 90

nd Maneater rperegr te mmm iris eee obey, 8 ches 2 a5) 16

2G ee i be Soe ce es ee ee 47.5 39

Sith uch sea aati pies ope eereetes Seas ca eee 46 34
Max. width across transverse processes -

dist; Pune l ARG e BOS See ee een ene i) 90

Ae cael LA aoe es pc oe a 68 67
Width across postzygapophyses

Nal i s Rare te ieee Rage Nel: Fan DET oh eeprom ee 26
Width across anapophyses

POH ine BR aie oh AS ogee ec erate ts a eer ete oe 44 pill

TST: cc stip Soggy meee et ae mee Gee. eos 46.2 50
Width across metapophyses

aD evap Se Hae ee Sea eee ANS eg Othe omy Pris Ek o1 44

TV tH 2.4 Rhee oe eee ee ea 50

The first and second lumbars have heavier spines, their
anapophyses extend shghtly farther back and are situated
a little farther down than in the lon.

Measurements of Lumbars.

1st 2d
12839 01050 12839 01050
mm.) MM. mms mine

Leneth cot centrum. kp ee, ee 39 40 40 48.5
Benethicot Spine" <2 52 eaeeeeuetaens os os. cares 46.1 38 90.5 43
Width across anapophyses, 222... o../ 2. 43 46 41* 45
Width across metapophyses ............ 46 952 59 02

* Approximate.

the Marsh Collection, ete. 453

The sacrum is but slightly smaller, while the spines
are higher and a great deal more robust. The anterior
articular facets lie in two planes, one outward and
upward, and the second directly upward, while in F’. leo
they only have one—outward and upward. The rudi-
ments of the zyga- and metapophyses are larger between
the first and second segments and smaller between the
second and third than in the lion. In F. leo the articular
surfaces for the pelvis and the last lumbar are solidly
united with each other and the centrum, but in the fossil
they are widely separated and have no connection with
each other. This fossil sacrum shows a pathological con-
dition on the left side where there has been much
exostosis, and the lateral mass is much closer to the
eentrum. The right side is normal. The maximum
width across the lateral masses is 86 mm. in the fossil
and 85 mm. in F’. leo, while the combined length of the
first two sacrals is the same in both, 62 mm. ,

The remaining part of the pelvis shows close similarity
to F. leo in size and character. The tubercle for attach-
ment of the rectus femoris muscle, in advance of the
acetabulum, is slightly larger and more elevated, while
the concavity above is smaller in the fossil, the ischiatic
spine is a little less prominent, the ileo-pectineal eminence
is more robust and less elevated, and situated a little more
posteriorly. The maximum diameter of the acetabulum
in both is 40 mm.

Of the femur, the great trochanter is not quite so
heavy at the top, but extends a little farther down the
shaft. The lesser trochanter is larger, the linea aspera
more prominent, the trochanteric fossa deeper, and the
bone somewhat heavier than in the lion.

Measurements of Femur.

12839 01050
é mm, mm.
Max. transverse diam. at middle of lesser
Bie EGET at tee a CP ee 50 49
Max. distance from inf. part of lesser tro-
chanter to top of greater trochanter ........ 76 69

Transverse diameter at middle of shaft........ 28.1 29

454 Thorpe—Some Tertiary Carnwora in:

REFERENCES.

Barbour, H. H., and Cook, H. J. 1917. Skull of Hlurodon platyrhinus, sp.
nov. Neb. Geol. Survey, 7, pt. 19, 173-180.

Bovard, J. F. 1907. Notes on Quaternary Felide from California. Univ.
Calif., Bull. Dept. Geology, 5, 155-170.

Cook, H. J. 1909. Some new Carnivora from the Lower Miocene beds of
western Nebraska. Neb. Geol. Survey, 3, pt. 9, 261-272.

—1914A. A new canid from the Lower Pliocene of Nebraska. Ibid., 7,
pt. 7, 49-50.

—1914B. Note on the dentition of Amphicyon amnicola, a gigantic fossil
dog. Ibid., 7,.pt. 105 7-58.

Cope, E. D. 1873A. Third notice of extinct Vertebrata from the Tertiary
of the plains. Pal. Bull. No. 16, 1-8.

—1873B. Synopsis of new Vertebrata from the Tertiary of Colorado,
obtained during the summer of 1873, pp. 1-19. Washington.

—1877. Report upon the extinct Vertebrata obtained in New Mexico by
parties of the Expedition of 1874. U. S. Geog. Surveys W. 100th
Merid., 4, Paleontology, 1-370.

—1879. Observations on the faune of the Miocene Tertiaries of Oregon.
Bull. U. 8. Geol. & Geog. Survey Terr., vol. 5, 55-69.

—1881. On the Canide of the Loup Fork epoch. Ibid., vol. 6, 387-390.

—1884. The Vertebrata of the Tertiary formations of the West. Rept.
U. S. Geol. Survey Terr.) 3.

—1890. On two new species of Mustelide from the Loup Fork Miocene of
Nebraska. Amer. Nat., 24, 950-952.

— and Matthew, W. D. 1915. Tertiary Mammalia and Permian Verte-
brata. Amer. Mus. Nat. Hist., Mon. Series, 2.

Cragin, F. W. 1892. A new sabre-toothed tiger from the Loup Fork Ter-
tiary of Kansas. Science, 19, 17.

Douglass, E. 1903. New vertebrates from the Montana Tertiary. Ann.
Carnegie Mus., 2, 145-200.

Filhol, H. 1876-1877. Recherches sur les phosphorites du Quercy, Etude
des fossiles qu’on y rencontre et spécialement des mammiféres. Ann.
S2i. Géol., 8, 1-340.

—1891. Etudes sur les mammiféres fossiles de Sansan. Ibid., 21, 1-319.

Hatcher, J. B. 1894. On a small collection of vertebrate fossils from the
Loup Fork beds: of northwestern Nebraska; with note on the geology of
the region. Amer. Nat., 28, 236-248.

—1902. Oligocene Canide. Mem. Carnegie Mus., 1, 65-108

Hay, O. P. 1899. Notes on the nomenclature of some North American
fossil vertebrates. Science (2), 10, 253-254.

Leidy, J. 1853. (Remarks on a collection of fossil Mammalia from
Nebraska.) Proc. Acad. Nat. Sei., Phila., 6, 392-394.

—1856. Notices of remains of extinct Mammalia, discovered by Dr. F. V.
Hayden in Nebraska Territory. Ibid., 8, 88-90.

—1858. Notice of remains of extinct Vertebrata, from the valley of the
Niobrara River, collected during the exploring expedition of 1857, in
Nebraska. Ihbid., 10, 20-29.

—1872. Remarks on some extinct vertebrates. Ibid., 24, 38-40.

Marsh, O. C. 1871. Notice of some new fossil mammals and birds from
the Tertiary formations of the West. This Journal (3), 2, 120-127.
Matthew, W. D. 1901. Fossil mammals of the Tertiary of northeastern

Colorado. Mem. Amer. Mus. Nat. Hist., 1, 355-447,

—1902. New Canide from the Miocene of Colorado. Bull. Amer. Mus.
Nat. Hist., vol. 16, 281-290.

—1903. The fauna of the Titanotherium beds at Pipestone Springs, Mon-
tana. Ibid., vol. 13, 197-226.

the Marsh Collection, etc. 455

—1907. A Lower Miocene fauna from South Dakota. Ibid., 23, 162-219.

—1910. The phylogeny of the Felide. Ibid., 28, 289-316.

—1918. Contributions to the Snake Creek fauna. Ibid., 38, 183-229.

and Gidley, J. W. 1904. New or little known mammals from the Mio-

cene of South Dakota. Ibid., 20, 241-268.

and Cook, H. J. 1909. A Pliocene fauna from western Nebraska, Ibid.,

26, 361-414. y

Merriam, J. C. 1913. Notes on the canid genus TZephrocyon. Univ.
Calif., Bull. Dept. Geology, vol. 7, 359-372. ,

Peterson, O. A. 1906. The Miocene beds of western Nebraska and eastern
Wyoming, and their vertebrate fauna. Ann. Carnegie Mus., 4, 21-72.

—1910. Description of new carnivores from the Miocene of western
Nebraska. Mem. Carnegie Mus., 4, 205-278.

Roger, O. 1896. Verzeichniss der bisher bekannten fossilen Saugethiere.
Bericht. Naturwiss. Ver. f. Schwaben u. Neuburg (a. V.), 32, 1-272.

Scott, W. B. 1890. Preliminary account of the fossil mammals from the
White River and Loup Fork formations, contained in the Museum of
Comparative Zoology. Pt. 2. Bull. Mus. Comp. Zool., vol. 20, 65-100.

—1$894. The Mammalia-of the Deep River beds. Trans. Amer. Philos. Soc.,
18, 55-185.

—1898. Notes on the Canide of the White River Oligocene. Ibid., 19,
327-415.

Sinclair, W. J. 1915. Additions to the fauna of the Lower Pliocene Snake
Creek beds. Proc. Amer. Philos. Soc., 54, 73-95.

Thorpe, M. R. 1922. Oregon Tertiary Canide, with descriptions of new
forms. This Journal (5), 3, 162-176.

Wortman, J. L. 1901A. A new American species of Amphicyon. This
Journal (4), 11, 200-204.

—1901B. Studies of Eocene Mammalia in the Marsh Collection, Peabody
Museum. Pt. I, Carnivora. Ibid., 11, 333-348, 437-450.

—and Matthew, W. D. 1899. The ancestry of certain members of the
Canide, the Viverride and Procyonide. Bull. Amer. Mus. Nat. Hist.,
vol. 12, 109-138.

A56 Sayles—-Dilemma of Paleoclimatologists.

Arr, XLI.—The Dilemma of the Paleoclumatologists; by
Rosrerr W. Sayues, Harvard University.

The so-called Glacial Theory was promulgated about
1835 by Agassiz and Carpentier. Since that time geolo-
gists have sought diligently to discover the cause or
causes of periods of glaciation, and one theory after
another has been propounded to account for them, but up
to the present time no one of the many explanations has
proved wholly adequate.

In 1855 A. C. Ramsey advocated the glacial origin of
certain breccias of Permian age in the Midland district of
England, but there-were very few who accepted his view.
In 1856 the Blandford Brothers discovered the Talehir
glacial conglomerate or tillite in the Cuttack district of
India. Discoveries of tillites of the same age as those of
India were soon made in South Australia and South
Africa, then in South America, North America at Boston,
then in Alaska. There are a number of conglomerates in
other parts of the world of late Carboniferous or early
Permian age which may be glacial in origin.

Tillites that occur at other points in the time seale have
now been made known. ‘Thus, in 1908, Coleman found
glacial strie on pebbles and bowlders in the Cobalt con-
glomerate of Huronian age at Cobalt, Ontario. Although
geologists had accepted the Permo-Carboniferous glacia-
tion and the late Proterozoic or early Cambrian evidence
of glaciation, it was more or less of a strain on their faith
to believe that the earth was cool enough to permit of a
period of glaciation in the Huronian. But the gradually
increasing confidence in the planetesimal hypothesis of
Chamberlin and Moulton would do away with presump-
tion of a hot globe growing progressively cooler and
cooler, which was the prevailing belief of the geologists
of the 19th century—this, of course, based on the nebular
hypothesis of La Place.

More evidence of glaciation from all parts of the geolog-
ical column has come in. In addition to the Huronian and
Permo-Carboniferous glaciations, evidence of glaciation
in the Late Proterozoic, Lower Cambrian, Lower Ordo-
vician, Silurian, Late Silurian or Early Devonian, Trias-
sic, Cretaceous, Hocene, and Miocene, has been found.!

* For references, arranged alphabetically, see the close of this paper.

Sayles—Dilemma of Paleoclimatologists. 457

The Lower Ordovician glaciation mentioned above was
found by the writer in the presence of I’. H. Clark on Sep-
tember 21st, 1921, at Levis, Quebec, in shales of Beekman-
town age. The evidence of it is a tillite of the submarine
or sublacustrine variety, varying in thickness from about
four to eight feet. The shale below and above it is very
regularly banded suggesting seasonal deposition.

Note—Since makine this discovery Professor Coleman has
called the attention of the writer to his Presidential Address
before the Royal Society of Canada, in April, 1921, in which he
states his belief in the glacial origin of the Levis conglomerates.

Three papers of great importance to paleoclimatolo-
gists have appeared in America in the last three years
The first by Knowlton? in 1919, the second and the third
respectively by Coleman? and Schuchert* in 1921. KKnowl-
ton claims on paleobotanical evidence, that before the
Pleistocene there was a non-zonal arrangement of earth
temperature, and that the surface temperature was warm-
humid from poles to equator, for the most part, through-
out the entire geological record. He explains this by the
hypothesis of Marsden Manson’ which calls for a cloud
sphere that shut out the greater part of sunlight and heat
rays, and thus surface temperatures were kept up by the
heat of the earth itself through the ocean waters. Knowl-
ton claims that glacial periods were often accidents in the
history of the earth, and that many of the tillites indicate
glaciation of a local nature only. He also claims that
Manson’s theory explains the almost tropical distribution
of Permo-Carboniferous glaciation better than any other,
this notwithstanding the fact that tillites of this age have
been found in Alaska.

Knowlton’s evidence for the general mild climates
for the earth during many periods cannot be doubted,
but when he totally eliminates seasons during these times
of warmth the geologists demand explanations of many
things. Coleman and Schuchert have shown many excel-
lent reasons for believing that the sun held sway and sea-
sons existed in the warm periods as now, although sea-
sons were less marked during the warm episodes than our
seasons today.

The writer has asked Dr. Knowlton to explain many
things which seemed incompatible with his ideas of a uni:

Am. Jour. Sci.— Firtu Seriss, Vou. III, No. 18.—Junz, 1922.
39

~~

458 Sayles—Dilemma of Paleoclimatologists.

versal warm climate without sunlight. For example, it is
difficult to explain the evolution of our flowering plants
and insects, and the feathers on birds and hair on mam- .
mals, and the desert apparatus with which the camel is
furnished, without sunlight. Dr. Knowlton cites many
cases of annual rings of growth in fossil trees from Car-
boniferous to present time. Dr. Winifred Goldring® has
recently described Cordaites of Alleghany age from Okla-
homa which show pronounced growth rings, indicating
seasons in 35° N. latitude at that time. ‘The seasons need
- not necessarily have been cold and warm but may have
been wet and dry. How can such occurrences be
explained except by seasons? The Dead Sea is today
making seasonal deposits of mud with salt and gypsum
alternating regularly. In the past the same process went
on as-evidenced by the seasonal deposits of the Stassfurt
salts, of upper Permian age in Germany, and by the
Alsace salts, of Miocene age described by Gale” and
believed by him to be of seasonal origin. In Arizona
Shimer* described beds of gypsum and shale of Triassic
age, which alternate in a regular manner, and although he
did not say so, they are possibly of seasonal origin.
Blackwelder has recently made a study of oil shales from
the Green River region. This shale has very regularly
alternating brown and black bands. Blackwelder took
the specific gravity of the material of these brown and
black layers and found the former had an average speci-
fic gravity of about 1.9 and the latter about 1.3. This, in
the opinion of Blackwelder and the writer, is a case which
strongly suggests seasonal deposition. As time goes on
and the attention of geologists is called to such ocecur-
rences many more cases will doubtless be noted and
studied. Seasonal layers in glacial clays of Pleistocene
age have been known for many years, as described in the
works of DeGeer® and a number of other geologists.t° In
1915 the writer gave a paper before the Geological Soci-
ety of America describing seasonal banding in the slate
which occurs above the Squantum tillite in Boston harbor.
A paper comparing some Pleistocene clays in the Con-
necticut Valley with the Squantum Permo-Carboniferous
banded slates was published in 1919. As noted by
Coleman® and Barrell’! and others, this comparison
leaves little doubt as to the same origin of the

— Sayles—Dilemma of Paleoclimatologists. 459

seasonally banded Pleistocene clays and the banded
Squantum slate. Banded argillites with other tillites
in various parts of the world were also noted in this
paper. Sederholm!? and Hallet® of Sweden expressed
their belief in the seasonal nature of certain banded slates
in 1913. David and Stissmilch'* published a paper in
1919-20 in which they counted the years by means of
banded slate in an upper Carboniferous giacial lake in
Australia. At the present time the writer is engaged in
a study of banding in various argillites. From the prog-
ress so far made, it is difficult to escape the conviction that
seasonal deposition existed in many places and periods.
That all this banding is of a glacial nature, it is not neces-
sary to conclude, for seasonal deposition is going on now
far from glaciers. Shaw’ believes that seasonal deposi-
tion is going on at present in the delta of the Mississippi.
It will be necessary to compare this banding with the
bandings in slates and shales of which there are many
examples.. Coleman cites the case of the middle member
of the Sudbury Series called the McKim Graywacks con-
sisting of ‘‘thin layers of interbanded graywacks and
slate.”’

From a study of a metamorphosed slate, given to the
writer by N. E. A. Hinds, which shows regular bandings
of coarse and fine materials, it is not impossible to believe
that some of the banded schists and possibly some of the
regularly banded gneisses were originally seasonally
deposited sediments. This is a matter for future investi-
gation.

The Dual Heating System of the Earth.

With the tremendously long lease of life given to the
earth by the discovery of a very much greater number of
- unconformities and disconformities than we had any idea
of a few years ago, as pointed out by Barrell, and Cham-
berlin, and the new method of estimating the age of the
earth through the study of the transformations of radium,
it is evident that during the hundreds of millions of years
estimated, there is time for all kinds of climates. In fact,
the monotonous, non-zonal arrangement advocated by
Manson would not, as noted by Coleman, appear to have
been favorable in the least to evolution as it is recorded
in the rocks. That there were many more changes of

460 Sayles—Dilemma of Paleochmatologists.

climate than we had any idea of 20 years ago, 1s becoming
evident from the discoveries of tillites all over the world,
to say nothing of the mass of evidence for arid conditions.
Ts it necessary to conelude that there were no seasons dur-
ing all this tremendous lapse of time, or to say that there
were times when seasons were not pronounced enough to
make their mark, and at other times sufficiently pro-
nounced to manifest themselves with or without glacia-
tion? If they were not marked enough to show them-
selves in any manner who can say that Manson is not right
for such periods?

In spite of all the accumulating evidence for seasons in
the immediate and distant past, Knowlton and Manson
have stimulated an inquiry about earth heat which, let us
hope, will not cease until the matter is settled beyond all
question. They have courageously stated their views and
brought a new viewpoint in this field of past climates.
Under the false idea that this earth started hot and has
- ever since been growing cold, geologists have tried to
explain the reason, or reasons, for the giacial periods.
Now that the evidence points to the idea that the earth is
a self-heating body mainly under the influence of pressure
and radioactive substances, and that there is no evidence
in the rocks that the first stages of the earth were very
hot, is it not just as important to explain such surface
temperatures as must have existed when warm weather
plants and corals thrived near the poles, as to try to find a
cause or causes for glacial periods? No efforts have been
put forth in this direction commensurable with the efforts
expended on the glacial periods. It is about time that
eeologists face about and overcome a great difficulty that
has not been fully recognized as a difficulty, so crystal-
lized has become our opinion that the surface tempera-
tures of the earth during the warm periods were explained
by the Nebular Hypothesis and the heat of the sun. Such
is not the case if the Nebular Hypothesis be rejected, and
the more thought given to this subject, the more difficult
does it become to explain the conditions of those times.
Before the solution of the glacial climates is accomplished,
the conditions of the heat supply of the warm periods
must be understood, and perhaps the glacial question will
then settle itself.

What part of our surface temperature is due to the sun
and what part is due to our own heat? This question has

Sayles—Dilemma of Paleoclimatologists. 461

been settled by the physicists of the past with a confident
answer that the heat from the earth’s interior is a negli-
gible quantity. What would our surface temperatures be
if by some magic the ice from the poles should suddenly
be melted away? It would appear that both of these
questions must be answered satisfactorily before we can
make progress with the glacial period riddle, and the
warm period riddle.

If the earth receives at the surface more heat from the
interior than we are aware of, the place to look for it is
evidently in the oceans rather than on the continents.
When we consider the tremendous mass of ocean water,
all of which is supplied with heat in excess of 493 degrees
F’. in temperature above absolute zero, it is difficult to con-
ceive of the sun as the sole agent of the heating. There is
no zonal arrangement of temperature along the ocean bot-
toms, where the temperature rarely goes below the freez-
ing point, and it looks very much as though the earth must
prevent any lowering of the heat below that point. There
are several things which point to the earth as the heating
agent for at least part of the heat of the waters. Fora
long time such heat has been suspected. The geothermal
gradients along the coasts, as shown by the records of
deep borings, are not as much depressed as they should be
if the oceans have cooled off the upper part of the crust
close to the oceans. At a depth of two miles below the
surface of the continents and with a thermal gradient of
60 feet to a degree rise, there should be a temperature of
about 680 degrees F.. above absolute zero in middle
latitudes. We do not know what the temperatures of
the rocks are in the ocean bottoms at a depth of two miles
from the bottom of the waters. Is it possible that they
have lost 187 degrees in heating the waters? Murray?’®
records some observations on this point. He says:

‘‘Some such increase of temperature towards the bottom has
long been suspected as an effect of the internal heat of the earth;
as early as about 1840 Aimé looked for it, but his methods were
not sufficiently accurate. More recently, several indications of
a rise of temperature towards the bottom have been observed.
The pressure and the internal heat having the same effect, it is
difficult—at our present stage—to determine how much is due
to the internal heat of the earth. In any case the bottom water
temperatures would be considerably lower but for the effect of
pressure on the sinking waters.’’

462 Sayles—Dilemma of Paleoclimatologists.

Then again we are not sure just how much heat the
radioactive oozes on the ocean floor generate. Joly'’
found that the red clay and radiolarian ooze exhibit much
radioactive substance, this in the case of the red clay
being over ten times as much as in average continental
rocks. Murray says again in regard to a rise in tempera-
ture of the bottom waters:

‘‘This rise of temperature has also been attributed to decom-
posing organic matter and to radioactive matter in the deposits
at the bottom. Whatever may have been the cause we certainly
found a similar slight rise in the temperature of the deepest layer
on several subsequent occasions during our cruise.’’

We do not know how thick the red clay is, but we do
know its great extent and its great radioactive content.
All the bottom oozes have high radioactivity. Is it not
possible that these self-heating deposits prevent so great
a loss of heat from the underlying rocks and thus con- -
serve considerable heat which would otherwise be lost to
the waters? If such a conservation of heat were possible,
instead of a greatly depressed thermal gradient, due to
the cooling of the waters, might we not have a tempera-
ture of about 680 degrees F’. in absolute units under the
oozes? Possibly the temperature might be higher than
680 degrees F., for according to recent observations in
many deep borings the thermal gradient often increases
markedly after passing the 5,000-foot level, although in
some cases the reverse of this is true.

Now as to the second question. What would the tem-
peratures of our earth be without the polar ice? If it is
true that the oceans, with a constant supply of ice water
from the polar regions, receive a negligible amount of
heat from the earth—and this has not been proved—the
cold water would cool off the earth under the oceans to a
considerable extent and thus lower the geotherms. Per-
haps this is the case, but there is little evidence for such a
conclusion. If the polar ice was melted away the heat
from the ocean bottoms might warm up the ocean waters
considerably. What would be the result of this combina-
tion of earth heat and sun heat on our climates?

-At the present time the oceans and the seas cover about
139,686,000 square miles and the lands are estimated at
about 67,254,000 square miles. In comparison with some of

Sayles—Dilemma of Paleoclimatologists. 463

the former periods when epicontinental seas covered much
more of the present continents than today, we are living
during a period of high relief and land emergence and of
accelerated geological processes. When the epicontinen-
tal seas covered so much of the continents much more heat
was stored in their waters and not permitted to radiate
back into space as rapidly as is the case today on the con-
tinents, and especially in the polar regions, where so great
a part of the insolation is reflected away by the ice and
snow surfaces. There was a warm water heating system
which carried warmth to every quarter of the globe. The
seas of those periods of warmth and low relief are not even
imitated to-day, except imperfectly perhaps in one case,
the Red Sea. Here the cold bottom waters of the Indian
Ocean kept out by the sill at the straits of Bab-El-Man-
deb, do not enter, and the result is a sea with warm waters
from top to bottom, that approximates the mean annual
temperature of the Red Sea region. It is possible that
the Red Sea today approaches the conditions of the deep
oceans during the warm periods. With a depth of 1,200
fathoms the bottom waters have never been known to fall
below 552 degrees F., absolute degrees. This is about
40 degrees F. warmer than water at similar depths
in the adjacent Indian Ocean. Is this 40 degrees F. due
to the stored heat of the sun or to heat imparted by the
lithosphere under the Red Sea, or to a combination of
both?

The close agreement of our air temperatures and the
temperatures of the regolith presents an interesting case.
At the equator the mean annual temperature can be found
one yard below the surface. Ground temperatures in
middle and high latitudes several yards from the surface
are usually one or two degrees F’. higher than the mean
annual air temperature of the places. Lane* believes
that the snow cover in the northern states causes the
ground temperature about 6 feet below the surface to be 1
or 2 degrees above the mean annual temperature of the
place. In fact, observations of underground tempera-
tures several hundred feet below the surface would show
that they are in large measure controlled by latitude. The
writer finds'® that the temperature of the earth between
400 and 500 feet from the surface, is higher in the south-
ern states than in the northern states by about 16.5

464 Sayles—Dilemma of Paleoclimatologists.
Wells between 400-500 feet.
Depth 400 400 420 454 420
417 400 400 400 425
Nebraska Temp. 60 58.5 59 08.5 60.5
58 59 5A 62 61
AirTemp. 46.5 46.5 46 5 46.5 46.5
46.5 46.5 48.3 48.1 48.1
Depth 396 509 450 460 500 =500
Iowa Temp. 52 56 60 60 08 53
AirTemp. 47 50.5 80.5 50.5 49.3 46.8
Depth 530 447
Indiana Temp. 56 d4
Air Temp, dd 50.9
Depth 530
Ohio Temp. 54
Air Temp. 49.3
Depth 540 407
Pennsylvania Temp. 53 63.5 53
AirTemp.. 49.3 49.8 49.3
Depth 448 420 408
New Jersey Temp. 60 60 60
Air Temp. OO Pole Oa ro ee)
Depth 403 548 500
Texas Temp. 76 73 74
AirTemp. 62.9 68.9 65.4
Depth 427 385 480 ~ 820 526
Louisiana Temp. 79 val 72 72 74
AirTemp. 68 65.6 66.8 66.8 66.8
Depth 500 400
Mississippi Temp. TROY a hi
AirTemp. 68 63°4
Depth 487 470 409 415 495
450 456 440 495 415
Alabama Temp. 68 12 67 68 (fal
ial 72.5 71 72.5 68
AirTemp. 64.8 64.8 64.8 63.7 63.5
63.5 63.5 63.5 63.5 63.6
Depth 465 455 490
Georgia Temp. 70 70 71
Air Temp. 67 661 62.4
Depth 400? 500? 500 450 498
496 408 400
Florida Temp. Lee 78 80 78 al
80 80 80
AirTemp. 12.6 72.6 68.3 68.3 68.2
73.1 73.1 Tey,
south) Average Warth Pempenapure )s S22 coe see ee eee 73.4
North at te Beek ie Sie Lae ee ees ee Bb 56.9
16.5
South, Average Ain Temperature >) 25-23 9¢ te. eee 66
North < : OE Pic 2 Se ye ee 49.7

Averages

413

Sayles—Dilemma of Paleoclimatologists. 465

Wells between 900-1200 feet.

Averages
. Depth 1125 1000 1050 900 1150 1053
Towa Temp. Well 62 65.4 60.5 59 62 61.8
Air Temp. 50.2 49.3 49.4 46.8 49.3 A9
Depth 1050 1121 950 1040
Illinois Temp. 61.5 61 70 64.1
AirTemp. 49.4 49.4 49.9 49.5
Depth 1040 1040
Indiana Temp. 62 62
Air Temp. 50.7 50.7
Depth 1165 1165
Ohio Temp. 67 67
Air Temy. 50.4 50.4
Depth 960 1200 1030 1090 1125 1081
Pennsylvania Temp. 59.5 61 61.5 59.5 06.3 59.5
AirTemp. 49.3 49.8 50 48.3 48.5 49.1
Depth 1244 1244
New Jersey Temp. 67 67
AirTemp. 93.6 03.6
Depth 1020 1010 1100 Gt 1075
Oklahoma Temp. 84 7 86 98 !* "86.5
Air Temp. 60.4 60.3 59.2 60.3 iz 60
Depth 975 1011 996 900 970
Louisiana Temp. 82 vi) 84 79 80
Air Temp. 68.7 66 65.1 68.2 7
Depth 900 1021 1100 1007
Mississippi Temp. 82.5 Th © 75.5
Air Temp. 68 64.7 64.8 65.8
Depth 1200 930 1065
Alabama Temp. 73 79 eee’ d
Air Temp. 65.2 63.7 Ms 64.4
Depth 1020 1031 900 984
Florida Temp. 81 74 7 76
Air Temp. 68.2 68.2 68 2 68.2
South Average Earth Temperature eee ea eee ee 78.8
North “ON OES EOE 6 PO or eee en 28 Se ee 63.5
15.
South Average Air Temperature DE. Oe = te 65
North i Oy + «Ee lg le Ri tt arin le eee a 50.4
14.6

degrees F. The average difference between
annual air temperature of the stations in the northern
states (Nebraska, Iowa, Indiana, Ohio, Pennsylvania and
New Jersey) and the southern states (Texas, Louisiana,

the mean

466 Sayles—Dilemma of Paleoclimatologists.

Mississippi, Alabama, Georgia and Florida) is 16.3
degrees F. The average temperature at a depth between
400 and 500 feet for the southern states for 31 wells was
73.4 degrees FE’. and for 25 wells in the northern states 56.9
degrees F’. All boring records were taken at random.
For temperatures at depths between 900 and 1,200 feet it
was found that in the northern states with sixteen records,
the average temperature was 63.5 degrees F'., and in the
southern states between the same depths with sixteen ree-
ords, the average temperature was 78.8 degrees KF. Okla-
homa was substituted for Georgia and Illinois for
Nebraska as there were no records between 900-1,200 feet
available in Georgia and only one in Nebraska, and Texas
was omitted. The average mean annual air temperature
for the northern states was 00.4 degrees I’. and 65 degrees
I’. for the southern states. The difference between the
average ground temperatures in northern and southern
states this time was 15.3 degrees and between the air tem-
peratures 14.6 degrees F. In Nebraska and westward,
and in Texas and westward, the thermal gradients are on
the whole higher than in the Appalachian states. On this
account the states in the above list east of the Mississippi
River only were used in an additional test, and the differ-
ence between the average earth temperatures of the north
and south between depths of 400 and 500 feet was found
to be 14.9 degrees and the difference of the average air
temperatures of the stations 15.8. At depths between 900
and 1,200 feet there was a difference between the average
earth temperatures of only 11.9 degrees, and between the
average air temperatures of 15.5 degrees. A map of the
United States with thermal gradients plotted, resembling
the isostatic map recently published by Bowie, is much
needed.

In Alaska and Siberia the regolith is often frozen solid
for many feet. From there southward the temperature
of the upper crust grows warmer until at the equator the
warmest temperatures, for a large average, are obtained.
Has this always been the case?

It is probable that during the Pleistocene this condition
of the temperature gradients was accentuated. With a
removal of the polar ice and prevalence of warmer condi- -
tions would the underground conditions of the north
approach the conditions found in low latitudes today? If

Sayles—Dilemma of Paleoclimatologists. 467

so, would not the earth give much more heat to the waters
in a warm period than at such a time as the present when
the earth is evidently still cooled by glacial conditions and
has not yet gained back its warm period heat?

Now let us again consider the rocks under the bottoms
of the oceans. If the temperatures of the outermost crust
are controlled to the extent indicated above by the temper-
atures of the air, the temperatures of the rocks below the
oceans should also be controlled by the bottom waters of
the oceans. In other words, these rocks should be cooled
to a temperature of about 32° F., for an unknown depth
below the bottom. The evidence gathered from geother-
mal gradients along the coasts does not point to such a
conclusion, nor does the evidence of a rise in the tempera-
ture near ocean bottoms, as discovered by Murray, indi-
eate such cold sub-oceanic rocks.

It has been said that a drop in the mean annual tempera-
ture of 10-15 degrees F’. would bring on a glacial period.
This would not apply to one of the warm periods. ‘To
change one of the warm periods into a glacial period a drop
of something between 30-40 degrees F’. might be necessary.
it would almost seem that nothing but land emergence
could cool off the earth enough to prepare the way for
glacial conditions, and furthermore, large continental
areas are necessary for large continental glaciers, such
as are recorded in some of the tillites of the past. If the
sun should change its heat or if voleanic ash filled the
upper air during a period of great ocean transgression ne
glaciers of great extent could form, so it is only during
periods of land emergence that great continental glaciers
could register their existence in the rocks. Croll?® men-
tioned many strata barren of organic remains and
inferred from this cold periods intercalated between
warm periods. These cases may be doubtful, but the
inference has not been proved wrong. ‘'T'o change the
temperature of the earth from what it was during the
warm periods to a condition cold enough for a great ice
sheet would mean a very great change.

If the waters and lands of the past were universally
warm over the earth without very marked zonal arrange-
ments of temperature, as most of the paleontologists and
paleobotanists would have it, the waters no doubt con-
tributed largely to this condition. According to Mur-

468 Sayles—Dilemma of Paleochimatologists.

ray,'® who states the case graphically and in few words,
the influence of the ocean waters on our temperatures is
as follows:

‘Tt is thus not only the surface-water that may give off heat
to the air, but a great body of water extending to several hun-
dred metres in depth, and hence the great influence of the sea on
winter climates. The capacity of water for heat is very great
compared with that of the air. Supposing that we have 1 cubic
metre of water giving off enough heat to the air to lower the
temperature of the water one degree, this heat would be sufficient
to raise the temperature of more than 3,000 cubic metres of air by
one degree. An example will show the importance of this. Sup-
pose a body of water, 700,000 square kilometres in extent and 200
metres deep, to give off enough heat to the air in winter to lower
the water-temperature one degree, then the heat given off would
be sufficient to raise the temperature of a stratum of air covering
the whole of Europe to a height of 4,000 metres on an average ten
degrees. This explains how the Gulf Stream renders the climate
of northern Europe so much milder in winter than would be
expected from its northerly latitude. We shall see later on that
the oceanographical researches of the last few years give reason
to hope that it will even be possible to predict the winter temper-
ature of northern Europe from the temperature of the sea some
time in advance.’’

It is seen from this account that the transfer of a few
degrees of heat from a large body of water means a great
increase of heat in the atmosphere.

It is certain that cooler climatic conditions accompany
and follow mountain building and vulcanism, although not
all of these diastrophic periods had world-wide con-
tinental ice sheets. There would appear to be some
~ecausal connection between diastrophism and _ cooler
climates, and to many that connection would seem to be
the emergence of the continents and thus the breaking up
to a large extent of the warm water heating system of the
earth. Did the emergent continents really effect the
cooling?

In the case of the Pleistocene ice age the cooling effect
was a very slow one. All through the warm Kocene the
continents were emergent and yet there was no great cool-
ing, except local glaciation in the early Eocene, until the
Miocene and Pliocene. It is true that no great mountains
existed in the Hocene comparable with those of the Plio-

Sayles—Dilemma of Paleoclimatologists. 469

eene. Mountain building and vulcanism went on for mil-
lions of years before the cool Pliocene and cold Pleisto-
eene. There is a lagging effect here which points to a
eradual cooling off of the earth. In this slow cooling,
mainly during the Pliocene but to some extent also during
the Miocene, the oceans cooled before the lands, as proved
by the migration of northern marine species southward
and the less rapid migration of the land flora and fauna
during the same time.” This cooling of the waters could
have been produced by the cooling of the lthosphere
under the oceans or by the cooling of the atmosphere and
‘formation of winter ice at the poles. To cool off the
oceans to the extent indicated would take a very long time.
The process was fairly steady. The effects of water
vapor in carrying heat can hardly be emphasized too
much. With the emergence of continents and the emer-
gence of barriers in the seas, during periods of mountain
building, the scope of the heating effects of water vapor
would be much restricted, with a cooler earth as the
result. However the cooling was accomplished, when the
glacial conditions finally did set in they came on rapidly.
The earth had cooled off enough for some cause, other
than its gradual cooling off, to act abruptly. The cooling
and heating, if accomplished by secular cooling, as Man-
son would have it, is much too slow a process to account
for the interglacial episodes of the Pleistocene and
Permo-Carboniferous. Changes in the carbon dioxide
content of the atmosphere is also much too slow a process
to explain these same interglacial episodes.

A great many theories for ice ages have come and gone.
No theory can stand long which does not satisfactorily
account for interglacial episodes. Only two hypotheses
so far advanced would appear to the writer to have any
_ chance: changes in the heat of the sun, and periods of vul-
canism with great amounts of voleanic dust. If it could
be proved that a change in the heat of the sun would bring
on glaciation, when the earth had been cooled off enough
by continental emergence, affected somewhat by loss of
heat through large voleanic extrusions, then a reverse
change would explain an interglacial episode. That such
a change in the heat of the sun should come just at the
time when the earth was sufficiently cooled off, would
appear to be a coincidence. How can we understand it,

470) Sayles—Dilemma of Paleoclimatologists.

unless we say that continental emergence cooled off the
earth a little, and that the major changes were due to great
changes in sun’s heat; or that the sun changed gradually
through the Piiocene and then more abruptly to bring in
the glaciers, and then changed just as abruptly in the
reverse direction to bring on more heat again? It is con-
ceivable that the sun has a regular period of change of
great length, of an oscillative nature, compared to which
the sunspot period would be the merest ripple on a great
wave. With the greater length of geological time now
becoming evident, such a change in the sun might very
well come during a time when the lands were highly
emergent and the earth cool. At other times when the
continents were submerged such changes in the sun could
not bring on great glaciation, but might cool the earth con-
siderably and produce local glaciation. As noted above,
Croll thought the barren strata indicated periods of cold.

The voleanic dust hypothesis advocated by Hum-
phreys*? has an advantage over the solar hypothesis in
that we need not go beyond the earth itself for an explana-
tion of interglacial phases. Periods of vulcanism fol-
lowed by times of quiet clear air would explain glacial
and interglacial episodes. There have been many periods
of great vuleanism without accompanying glacial periods,
but in these cases it is most probable that the earth was
either too warm to be brought to a condition of glaciation,
or that vuleanism was too spasmodic in its action to have
a prolonged cooling effect, or it is possible that evidence
exists or existed for such cooled periods but that it
remains undiscovered or lost. It would not be remark-
able if such evidence were lost. Suppose, for example,
that we have at present reached the end of the Pleistocene
and are now at the beginning of a new warmer period of
earth history. Unless the glaciated areas sink rela-_
tively soon there will be very little to preserve as evidence
of Pleistocene glaciations. The marine Pleistocene gla-
cial deposits are very limited. The chances that these
will be uplifted are very uncertain. Such thoughts as
these lead one to conclude either that the Pleistocene
glaciation was very limited or that there is much more to
come, and that we are now living in an interglacial time.
If a small remnant of till should be preserved, would the
Pleistocene be regarded as just a local glaciation? There

Sayles—Dilemma of Paleoclimatologists. 471

ought to be some evidence of the climate of the Pleistocene
in the deposits which formed outside the giaciated areas
and on these, if there were no tillites or glacial slates, the
conclusions would be based.

If the geologists can make out four episodes of great
explosive vulcanicity during the Pleistocene to correlate
with the four glacial episodes now recognized, the most
important cause for ice ages may have been found. The
occurrence of an interglacial ash bed at Des Moines, Iowa,
reported by Keyes, is interesting in this connection. In
any event any cause within the earth itself, which could
cause glacial periods, should be investigated to the utter-
most before we venture beyond the earth. If the sun is
responsible for glacial and interglacial episodes it will be
well-nigh impossible with our present knowledge to prove
it, however strong any solar theory may be.

The finding of tillites in formations which originated
during times of great land emergence have led geologists
to infer that only at such times could the earth be cool
enough for glacial conditions, and that there was a causal
connection between mountain building and glaciation.
From recent discoveries of tillites deposited during times
of great ocean transgression it is now greatly to be ques-
tioned whether a large land surface is necessary for at
least a logical glaciation. As already pointed out, a world-
wide glaciation could not be registered without great land
masses, but local glaciation appears to have occurred dur-
ing the Silurian, about Niagara time, as discovered by
kurk?? in Alaska in 1917, and in the Beekmantown at
Levis, Quebec, as discovered more recently. Kirk found
a thick bed of tillite between what would appear to be
warm water limestones. In the Beekmantown the waters
were supposed to be warm by some and cold by others.
In both cases the extent of land was comparatively small
and the extent of marine waters was great. If there had
been extensive continental areas during these times it is
probable that great continental glaciers would have
formed. Itis to be noted that the latitude of the Silurian
case is between 55°-60° north and of the Beekmantown
case at Levis about 47° north. Very regularly banded
shales of Lower Ordovician age with marked seasonal
characters, collected by Mr. A. C. Swinnerton in the
summer of 1920 in Georgia and Tennessee, would indicate

472 Sayles—Dilemma of Paleoclumatologists.

that seasons existed in the Ordovician in latitude 35°
north. It would seem necessary to invoke some cool-
ing agent if the glaciers went some distance into
the sea, as they evidently did. Blackwelder™* and
Kirk? both believe that there were seasons in Alaska
during these times. During the Beekmantown time great
vulcanism, mostly of the explosive kind, was going on in
the British Isles. Twelve thousand feet of volcanics,
mostly of an explosive nature, were deposited in the Lake
District alone. Vulcanism was also going on in other
parts of the world. In the case of the Niagara tillite of
Kirk, voleanic activity went on contemporaneously in
what is now the Penobscot Bay region of Maine. Kirk
speaks of vast thicknesses of voleanic materials in this
Paleozoic section of southeastern Alaska. In the case of
the Beekmantown, splendid banding with seasonal charac-
ters occurs in the section. Whether or not a great land
emergence is causally connected with wide-spread glacial
conditions is still an open question.

In this paper the writer does not advocate Humphrey’s
voleanic theory to the exclusion of any solar theory, or
any other theory that will explain interglacial episodes,
but to say with Schuchert that Humphrey’s theory will
not answer, because periods of glaciation or cooling are
not found to accompany every period of vuleanism, is not
enough to disprove it. The criteria for the determination
of cool periods has increased in the last few years and
many formations must be examined again. Periodic
changes in our present climate due, according to Bruckner
and Huntington, to changes in solar heat, have’ been
observed, and the solar theories must be weighed care-
fully. If Manson’s theory, or some modification of it,
can explain interglacial episodes satisfactorily, Manson
may be nearer the truth for parts of earth history than
has been supposed.

This paper has been written in a spirit of inquiry rather
than of affirmation. The theory of seasonal banding is
on trial and as an advocate of that theory the writer
would ask geologists to suspend judgment until the evi-
- dence for periods other than the Pleistocene and Permo-
Carboniferous has been presented.

Sayles—Dilemma of Paleoclimatologists. 473

Notre: Since writing this paper Dr. Knowlton has published
a reply to the papers of Coleman and Schuchert, but it is not
possible to discuss it here.

* Abbot, C. G. § Fowle, F: E., Voleanoes and Climate. Smithsonian Mise.
Coll., vol. 60, No. 29, pp. 1-24, 1913.

* Atwood, W. W., Hocene Glacial Deposits in Southwestern Colorado,
U.S. G. 8., Prof. Paper 95-B, 1915.

™ Barrell, Joseph, Bull. Geol. Soe. America, vol. 27, pp. 112-113, 1915.

a Blackwelder, Eliot, The Climatie History of Alaska from a New View-
point, Trans. TL. Acad. Sci., vol. 10.

* Coleman, A. P., Paleobotany and Earth’s Harly History, this Journal,
(9), vol. 1, pp. 315-319, April, 1921.

»” Croll, James, Climate and Time, Ist Ed., 1875, pp. 483-434.

™ Darton, N. H., Geothermal Data of the U. S., U. 8. G. S., Bull. 701,
1920.

“David and Sussmilch, Proc. Roy. Soc., New South Wales, 53, 270,
1919-20.

*De Geer, Baron Gerard, A Geochronology of the last 12,000 years.
Compte Rendu Cong. Geol. internat. sess. 11, 1910-1912, pp. 241-253, 2 pls.

"Gale, Hoyt S., The Potash Deposits of Alsace, U. S. G. S., Bull. 715-B,
pp. 47-48, 1920.

° Goldri ing, Winifred, Annual Rings of Growth in Carboniferous Wood;
Botanical Gazette, vol. 72, No. 5, Nov., OPile

8 Halle, Thore G., On the Geological Structure of the Falkland Islands;
Bull. Geol. Inst. Univ. Upsala, vol. 11, pp. 159-160, 1911.

* Humphreys, W. J., Physies of the ’Air, pp. 569- 603, 1920. Voleanic Dust
as a factor in the production of climatic changes. Wash. Acad. Sci. Jour.,
vol. 3, pp. 365-71, 1913.

* Joly, John, Radioactivity and Geology, 1909, pp. 35-69.

22° Kirk, Edwin, Paleozoic Glaciation in Sloneneaaieen Alaska. Bull. Geol.
Soe. America, vol. 29, No. 1, March, 1918, pp. 149-150.

* Knowlton, F. H., Evolution of Geologic Climates, Bull. Geol. Soc. Amer-
ica, vol. 30, pp. 499- 565, BNI).

38 Lane, He iss Michigan Geol. & Biological Survey, Keweenaw Series of
Michigan, Chap. avis [on Vol, UYODS joules Util,

° Manson, M., Geologie and Present Chane ney, 1919, p. 217. Also Evolu-
tion of Climates.

4° Murray, Sir John, Depths of the Ocean, 1912, pp. 61-62; pp. 160, 166,
radium; pp. 221-222; 229-230, water heating.

* Sayles, R. W., Bull. Geol. Soc. America, vol. 27, pp. 110-111, 1915. Proe.
Nat. Acad. Sci., vol. 2, pp. 167-170, 1916. Seasonal Deposition in Aqueo-
glacial Sediments, Mem. Mus. Comp. Zool., Harvard, vol. 47, No. 1, 1919.

* Schuchert, C., Evolution of Geologie Climates; this Journal, (5), vol. 1,
pp. 320-324, 1921. Also, Chmates of Geologic Time; Carnegie Inst. of
Wash., Pub. No. 192, pp. 263-298.

* Scott, W. B., An Introduction to Geology, 2d ed., 1909, p. 767.

% Sederholm, J. J., Subdivision of the Pre-Cambrian of Fenno-Scandia.
Compte Rendu Cong. Geol., 11, p. 683-698, 1910-1912.

® Shaw, HE. W., U.S. G. 8S. Prof. Paper 85-B, p. 17, 1917. The Mud Lumps
at the Mouths of the Mississippi.

* Shimer, H. W., Bull. Geol. Soc. America, vol. 30, No. 4, p. 480.

Am. Jour. Sci.—FirtH Serizs, Vou. III, No. 18.—June, 1922.

Bis)

AT4 Scientific Intelligence.

SCIENTIFIC INTELLIGENCE

[I Cnemistry anp Puysics.

1. The Heats of Neutralization of Potassium, Sodium and
Inthium Hydroxides with Hydrochloric, Hydrobromic, Hydrio-
dic and Nitric Acids, at Various Dilutions —THEODORE W. RicH-
ARDS and AuLAN W. Rows have made a very elaborate study of
the heats of neutralization of each of these bases with each of the
acids; the adiabatic calorimeter, which was devised in Professor
Richards’ laboratory was employed, while the refinements in the
precautions employed to secure the greatest accuracy and the
remarkable agreements obtained in the various series of deter-
minations command the highest praise. The authors had pre-
viously determined with great care the heats of dilution of solu-
tions of these bases and acids as well as of their salts, in order
that the heats of neutralization at various concentrations might
be calculated.

Solutions of uniform molecular concentration, corresponding
to a dilution with 100 molecules of water, of the four acids and
the three bases were neutralized calorimetrically in all possible
pairs at two temperatures not far apart, and the results were
interpolated to exactly 20°. The values ranged from 13,750 to
14,085 calories, sodium hydroxide giving the lowest values among
the bases, and hydriodte acid among the acids. By means of eal-
eulations to other dilutions, and extrapolations to infinite dilu-
tion, it was concluded that the heat of formation of water from
its ions at 20° is probably not over 13.69 Cal. (20°) or 57.22 kilo-
joules, and possibly not under 13.62 Cal. or 56.93 kilojoules.

H. L. W.

2. A Rapid Todometric Estimation of Copper and Iron im
Mixtures of ther Salts —IAN WiLLIAM WARK has worked out a
method for making these determinations successively in a single
‘ solution. The process appears to be fairly accurate and very
convenient, so that it should find extensive application, particu-
larly in technical work. ‘The solution, in which the iron should
be in the ferric condition and which should be as concentrated as
possible, is neutralized with ammonia. Moderate amounts of
ammonium salts do not seriously affect the end-point. Then 2 g.
of disodium phosphate, or twice as much if there is little copper,
are added for each 0.1 g. of iron, together with 5 g. of KI and
5 ec. of acetic acid (80%) for each 0.1 g. of total metal present.
The titration with tenth normal thiosulphate for copper is then
made after waiting for 5 or 10 minutes, and the mixture is
warmed to 50° and titrated further, if necessary, after 5 min-
utes. This method for the determination of copper in the pres-:
ence of iron was described in its essential features by Moser in

Chemistry and Physics. 475

1904. To the titrated solution mentioned above 10 ec. of 6 N
sulphuric acid is added for each 0.1 eg. of total metal, and after
5 or 10 minutes the iron is titrated with the thiosulphate solu- ~
tion. No indicator except the free iodine is needed in either
titration. It was found that the method gave satisfactory results
except in cases where the amounts, either of copper or iron pres-
ent, were very small.—J. Chem. Soc., 121, 358. H. L. W.

3. Introduction to Physical Chemistry; by Str JAMES
WALKER. 8vo, pp. 440. London, 1922 (Macmillan and Co.,
Limited ).—This text book appears to have been very favorably
received and extensively used, for this is the ninth edition that
has appeared since its first publication in 1899.

The book contains 36 chapters, each of which discusses an
important topic of physical chemistry very clearly and ably, with
particular attention to the practical appheations of the theories.
The use of any but the most elementary mathematics has been
avoided, except in the last chapter which deals with the thermo-
dynamical proofs, where a rudimentary knowledge of the calcu-
lus is needed. The recent developments, such as those relating
to atomic number and isotopy, and also in regard to the structure
of atoms, including the theoretical work of the Americans G. N.
Lewis and Irving Langmuir, are well presented in this book.

It may be mentioned that the author regards a recent loniza-
tion theory by Ghosh as a satisfactory one having a theoretical
basis, whereas this theory is severely criticized by James Ken-
dall of Columbia University in the April, 1922, number of the
Journal of the American Chemical Society. Further discussion
of this matter is to be awaited with interest, and even if too much
credit has been given to this theory in the book, it is a very small
matter in connection with its general reliability and excellence.

H. L. W.

4. Colloid Chemistry of the Proteins; by WouFGaNe PAUL.
Translated by P. C. L. THorne. Part I. 8vo, pp. 140. Phila-
delphia, 1922 (P. Blakiston’s Son & Co.).—This little book has
been developed from lectures delivered in Vienna a number of
years ago, and the English translation was made in Eneland. It
deals with quantitative methods of physical and colloid chemis-
try as applied to proteins, a field in which the author and his
associates have made extensive and important investigations.
It is to be observed that the book does not deal with the ordinary
chemistry of the proteins, but considers the physical behavior of
a few of them in connection with their ionization, particularly
in the presence of acids and bases. The second part will include
the relations of the proteins to neutral salts and to the salts of the
heavy metals, to colloids and to ampholytes, the properties of the
albumin gels, and, finally, the physical chemistry of the purest
albumin so far prepared. H. L. W.

476 Scientific Intelligence.

5. The Aurora Line of the Night Sky—A green line of
unknown origin corresponding to the wave length 5578, in the
spectrum of the aurora, has been reported by several observers as
present in the sky on ordinary nights, and in comparatively low
latitudes. Various questions suggested by this occurrence have
been systematically investigated by Lord Rayleigh. If the phe-
nomenon is directly connected with the polar aurora it might be
expected that a gradation of intensity between the usually very
faint effect and a bright auroral display would be observed and
that this would become more pronounced at higher latitudes. A
series of spectral photographs was made at Terling (near Lon-
don), every night from Feb. 26 to July 3. From this systematic
series estimates of the intensities of the line were prepared and
comparisons made with the amount of magnetic disturbance and
the transit of spots over the sun’s central meridian. No obvious
connection between the intensity of this green line and the ter-
restrial or solar phenomena was found.

Comparisons between photographs made in the neighborhood
of Neweastle with those taken near London, showed that the
intensity appeared to be greater at the more southerly station,
which would indicate that the cause must be different from that
of the polar aurora. This conclusion is further supported by the
fact that the aurora line is visually observable by California
observers, 15° further south, while Rayleigh has been unable to
see it at all under ordinary conditions,

Some authors have identified this aurora line with krypton
5770, but the more recent measurements clearly prove that it does
not originate with this element. Rayleigh also tested the sugges-
tion that the green line might be the fluorescent spectrum of
ozone excited by the ultra-violet light of negative oxygen bands
but nothing corresponding to it was indicated by his experi-
ments.—Proc. Roy. Soc. 100, 367, 1922. F. E. B.

6. The Color of the Sea. —The early theories of the color of
the sky assigned its blue color to the scattering of light by par-
ticles of water or dust in the air. It is now known from the work
of Cabannes and of Rayleigh on the scattering of light by dust
free gases, and from the measurements on the light from the
atmosphere, that the sky owes its colors to diffraction by the mole-
cules of the air itself.

In regard to the color of large masses of clear water consid-
erable div ergence of opinion has existed. The late Lord Ray-
leigh was inclined. to the view that the blue color of the deep sea
was simply the blue of the sky seen by reflection. Judged by
_the literature of the subject, the trend of opinion appears to have
been that in so far as there is any real effect apart from
reflected skylight, the color is to be explained by absorption im
the water, the return of light from the depths of the liquid being
due to matter suspended in it.

Chemistry and Physics. ATT

. A recent investigation by R. V. Raman has ted him to the
conclusion that the color of the sea is due to the seattering of
light by the water molecules alone. The Rayleigh law of scatter-
ing for gas molescules cannot be applied directly to the case of a
liquid where the spacing of the molecules is so close and the free-
‘dom of movement must be small. A method of attack has how-
ever been found in the ‘“‘theory of fluctuations’’ proposed by
EINSTEIN and SMOLUCHOWSKI who consider that a liquid medium
may be regarded as undergoing small local variations of density
due to the irregular movements of the molecules and that as a
result of these fluctuations of density, light is scattered.
Raman’s calculations from this theory show good agreement
with the observed intensity of ight scattered by pure water. It
is not far from 160 times that in dust free air.

Other observations which confirm the author’s new view are
that: (a) the coefficient of extinction agrees well with the theo-
retical value in the parts of the spectrum where there is no
selective absorption; (b) a sufficiently deep layer of water is
shown to exhibit by molecular scattering a deep blue color inde-
pendent of reflected sky light; (¢) apart from the fact that the
bluest waters are highly transparent and markedly free from
colloidal matter, a discussion of the effects of suspended matter
shows that the observed results are hardly consistent with the
assumption of its presence; (d) a number of interesting
phenomena of polarization and scattering are satisfactorily
explained.—Proc. Roy. Soc. 101, 64,1922. F, E. B.

7. Proportionality of Mass and Weight.—In the Proceedings
Am. Phil. Soc. 60, 1921, CHarues F’. BrusH published a kinetic
theory of gravitation and the results of certain experiments
which were thought to corroborate it. The author’s view was
that gravitation was due to an energy flux in the ether and on
this hypothesis it was thought that the very minute negative
permeability (sic) of diamagnetic substances might offer some
appreciable obstacle to this flux and thus affect the gravitational
field behind them.

To test it three sets of experiments were tried. (1) A repeti-
tion of the Cavendish experiment with the apparatus of Boys
comparing the attraction of masses of aluminium, zine, tin, lead,
silver and bismuth on a small silver ball. (2) Comparison of
the periods of two similar gravity pendulums, one having a zine
bob and the other a bismuth bob. (3) Comparison of two similar
torsion pendulums each loaded with equal masses of zine and bis-
muth. In (1) the attraction of bismuth was reported to be less
than that of zine in the ratio of 72 to 100. In (2) the bismuth
pendulum was said to show a gain of about one oscillation in
35,000 over the zine. In (3) the bismuth period was found to be
the shorter by 1 part in 1333.

-. The experiment on the gravitational pendulum has recently

478 Scientific Intelligence.

been repeated by H. H. Porrer and O. W. RicHarson at the
Research Laboratory of King’s College, London, and they cannot
confirm Brush’s result. The gravitational accéleration of bis-
muth was found to be same as that for brass to at least one part
in 50,000.—Phys. Rev. 19, 188, 1922. is eB?

8. The Teaching of General Science; by W. L. EXKENBERRY,
pp. xii, 169, Chicago, 1922 (University of Chicago Press).—
The publishers have projected a number of volumes entitled the
Nature Study Series under the editorship of Exuuiot R. Down-
ING of which the present work is the first to appear. It is a work
upon the theory of teaching, i. e. the aims, the principles of
organization, and methods of instruction in high and secondary
schools. Under the term general science are gathered such
branches as Astronomy, Agriculture, Physiology, Biology, Com-
mercial Geography, Chemistry, Meteorology, Physiography, and
Physics. As might be expected from a work on pedagogy the
author is apologetic rather than dogmatic toward the student,
as though the study of science needed to be justified. ‘Taken
from this angle the reader will find in the book a forceful state- —
ment as to the economic-vocational values of usable facts and their
bearing on science-education, conduct-controls, tastes, ideals, and
sociological aims. Pa ee

Il. Gerotogy anp MINERALOGY.

1. Hesperopithecus, the first anthropoid primate found in
America.—lIt is thus fittingly that Professor HENRY F: OsBorN
announces in Science for May 5 what may prove to be one of the
most remarkable discoveries in the history of vertebrate paleon-
tology. The type isa single water-worn tooth found by Mr. Har-
old Cook, consulting geologist, of Agate, Neb., in the upper or
Hipparion phase of the Snake Creek beds of western Nebraska.
These beds are conceded to be of Pliocene age, and the opinion is
expressed that the tooth is certainly a contemporary fossil.

In the judgement of Professor Osborn and Doctor W. D. Mat-
thew, to whom the specimen was referred, it represents the second
or third upper molar of a new genus and species of anthropoid.
Doctors W. K. Gregory and Milo Hellman are perhaps more
specific in their statement, when they say: ‘‘On the whole we
think its nearest resemblances -are with ‘Pithecanthropus’ and
with men rather than with apes.’’ Since 1908 there has been in
the American Museum collection another tooth from this same
horizon. It was so water-worn and from so aged an animal
that it has lain thus far undescribed, but comparison with the
new form seems to show genetic if not specific affinity.

The horizon is that of the Thousand Creek (Nevada) and Rat-
tlesnake (Oregon), of which the fauna contains not only Plio-

Geology and Mineralogy. 49

hippus in abundance, but Jllingoceras and other twisted-horn
antelopes of Asiatic affinity, which suggest for Hesperopithecus
or its ancestors a possible migration from the great primate cen-
ter of the Old World.

Further corroborative evidence would Be ereatly welcomed to
establish beyond doubt this most unique occurrence. Renee

2. Shallow-water Foraminifera of the Tortugas Region; by
J. A. CusHMAN. Dept. Marine Biology, Carnegie Inst. Wash.,
vol. 17, 85 pp., 14 pls., 1922.—An interesting description of the
145 forms of foraminifers found about the extreme western end
of the Florida Keys. There are also new observations on living
specimens, relating to their movements, colors, commensals, devel-
opment, and variation. The plates are from drawings by
J. Henry Blake. GS:

3. Fossil Echini of the West Indies; by Roprert T. JACKSON.
Stratigraphic Significance of the Species of West Indian Fossil
Echim; by T. W. VauGgHan. Carnegie Inst. Wash., Pub. No.
306, 122 pp., 18 pls., 5 text figs., 1922.—In this excellent memoir,
Jackson treats of all the known fossil echini of the West Indies,
89 species. Of these 8 are Cretaceous, the remainder Cenozoic
from Eocene to Plocene. Of the known forms the author
describes 57, and of these 16 are new; the 32 forms not seen by
the author are listed. There are 12 regular or endocyclic echini,
and the remainder are of the irregular or exocyclic type of struc-
ture. Vaughan describes the stratigraphic significance of the
eroup. Gees

4. Triassic Fishes from Spitzbergen; by Erik A:Son STEN-
s16. Pt. I, Introduction, Some Remarks on the Geology of the
Triassic of Spitzbergen, and Descriptions of the Families Ces-
tracionide, Celacanthidx, Paleonischide, and Catopteride. Pp.
307, 35 pls. 90 text figs. Vi ienna (Adolf Holzhausen), 1921.—In
this monograph are deseribed and illustrated in great detail
about 40 forms of fishes that occur in the Lower Triassic of the
Ice Fjord of Spitzbergen. Of elasmobranchs there are 4 genera
and 11 species; of Ceratodus, 1 species; of crossopterygians, 5
genera and 10 species; and of actinopterygians, 8 genera and 18
species. There is also a detailed statement of the stratigraphy
of the Triassic of Spitzbergen, having a total thickness of 594
meters, of which 315 meters are essentially black shales, the
remainder mainly yellow sandstones. G28.

5. The Miocene of Northern Costa Rica; by A. A. OLSSON.
Part 1, Bull. Amer. Paleontology, No. 39, 168 pp., 15 pls., 1922
(Harris Co., Ithaca, N. Y., $2.50) —A valuable report, of which
the first thirty-four pages are devoted to a general description
of the stratigraphy and correlation. The remainder contains
descriptions of the gastropods, of which there are 208 forms (110
new). : cos

480 — Scientific Intelligence.

6. The Geology of the Corocoro Copper District of Bolivia;
by JosepH T. SINGEWALD, JR., and E. W. Berry. Johns Hop-
kins Univ. Studies m Geology,-No. 1, 117 pp., 15 plsz7 1922
($1.25). —Here is described the geology of the Corocoro district
from the standpoint of the stratigrapher, paleobotanist, and
economic geologist. The district stands at an altitude of over
13,000 feet, but in Pliocene time, when these muds and sands of
fresh water origin were being deposited, with a measured thick-
ness of more than 17,000 feet, the elevation was at or below 6,500
feet. The entombed flora consists of one fern, one gymnosperm,
and twenty dicotyledons, all of genera ‘‘which still survive at
lower levels east of the Andes, where they are represented by
closely allied species.’’ Overlying the plant beds is another
series (Desaguadero), which may be of Pleistocene age. In 1919
the district yielded about nine million pounds of copper.

This paper is the first one of a new series to be issued by the
Johns Hopkins University. Four other parts on the geology of
South America are announced. This serial deserves support
because of the good geologic and paleontologic results which have
long been emanating from the Department of Geology at the
Baltimore institution. Cris

7. A Guide to the Fossil Remains of Man in the Department
of Geology and Paleontology in the British Museum of Natural
History. Third edition, 34 pp. with 6 plates and 14 text-ficures.
London, 1922.—Recent interest in the remains of fossil man has
been much stimulated by Mr. Charles Davison’s discovery of the
Piltdown skull, by that of the Rhodesia remains (this Journal,
4, p. 96) and finally by Harold Cook’s discovery of a tooth at
Agate, Nebraska, interpreted by the authorities in the American
Museum as the first’ anthropoid primate found in America
(see page 478 in this number). Hence the unusual value of this
small volume in which the whole subject (except as to the
Nebraska Hesperopithecus) is clearly but concisely presented in
all its relations.

8. Illinois Geological Survey; FRANK W. DEWotF, Chief.—
Bulletin 42 (322 pp., reclamation map of Illinois in pocket), is
devoted to the engineering and legal aspects of law drainage in
Illinois. Of the two authors, G. W. Pickens has prepared part I
on the status of drainage in September, 1920; also part II on
engineering problems, and part IV on State aid. Part ITI, deal-
ing with legal problems, is by F. B. Leonarp, Jr. The discus-
sion of the whole subject is enlightening as showing the difficul-
ties in the organization of districts under existing laws and the
obstacles met with in the effort toward the reclamation of large
areas of fertile lands in the river bottoms. The investigation
begun in July, 1919, was completed in September, 1920, most of
the field work being done in the spring and summer of 1920.

Geology and Mineralogy. | 481

The large map (28x49 inches, seale about 8 miles to 1 inch)
presents clearly the physical data obtained.

Another publication of the Survey is ace 26 of the
Co-operative Mining Series, being investigations prepared by the
co-operation of the Illinois Survey, the Engineering Experiment
Station of the University of Illinois and the U. S. Bureau of
Mines. This bulletin (247 pp., with 8 plates, 31 figures and 7
tables) is by GinpertT H. Capy and gives data as to the coal
resources of district IV. The district named is the part of the
central portion of the State yielding coal of the No. 5, or Spring-
field, bed. All of Peoria county is here included, a large part
of Fulton county, the part of Sangamon county north of
Chatham; also smaller areas in ten other counties some of which
are underlain by this coal. The No. 5 bed yielded over 11 million
tons from this district in the year ending June 30, 1920. Among
the districts of the Co-operative Investigations, this one ranks
second in area, third in order of production, and possibly first in
the amount of workable coal.

9. Illinois State Water Survey—Bulletin No. 16, Epwarp
Bartow, Chief, gives a report (280 pp., 36 figures) for the years
1918, 1919, of the State waters, both chemical and biological.

Bulletin No. 17, A. M. Busweuu, Chief, is an index (17 pp.) to
Bulletins 1 to 16.

10. Geological Survey of Western Australia; A. Gipp Mart-
LAND, Government Geologist.—Publications received include the
-Annual Progress Report for the year 1920, pp. 31. This is
accompanied by a map of Western Australia showing the series
of geological sketch maps (four miles to the inch) and other
geological maps issued since 1896. Points of special interest are
the development of water power for the generation of electricity
in the Kimberley division; petroleum prospects of the Busselton
region, in general favorable in outlook; deep borings at several
points, for example revealing the presence of coal at various
depths in the township of Collie. All of these are by Dr. Mait-
land. Other members of the staff also note works of important
character.

Four Bulletins have been issued, as follows:

No. 78. The mining geology of Kookynie Niagara and Tampa,
North Coolgardie Goldfield, J. T. Jurson, field geologist; with
petrology by R. A. FarquHarson. Pp. 98 with 5 plates and six
figures.

No. 79. Mining geology of Comet Vale and Goongarrie, North
Coolgardie Goldfield, by J. T. Jutson. Mineralogy by E. S.
Sruipson and petrology by R. A. FaARQUHARSON. Pp. 52, with 8
plates and 8 figures.

No. 80. The Mining Centres of Quinn’s and Jasper Hill,
Murchison Gold Field, by F. R. Ferprmann. Pp. 92, with 3
plates and 18 figures.

482 _ Setentific Intelligence.

Nos. 81, 88. Geology and Mineral Resources. No. 81. The
Yalgoo Goldfield, Part I, the Warriedar goldmining centre, by
PF’. R. FeELptmann. Pp. 40, with 3 plates and 5 figures. No. 83.
The Northwest, Central and Hastern Divisions (E. Long. 119°
and 122°, 8. Lat. 22°and 28°), by H. W. B. Tausot; petrology
by R. A. FarquHaRson.

Ill. Miscettanrous Screntiric INTELLIGENCE,

1. Washington meeting of the National Academy of Sciences.
—At the recent meeting of the National Academy (see p. 386)
the following gentlemen were elected to membership: Edward
W. Berry, Johns Hopkins University; George K. Burgess, U. S.
Bureau of Standards; Rufus Cole, Rockefeller Hospital, New
York; Luther P. Eisenhart, Princeton University; Herbert
Hoover, Secretary of Commerce; George A. Hulett, Princeton
University; Charles A. Kofoid, University of California; George
P. Merrill, U. S. National Museum; Carl E. Seashore, State Uni-
versity of Iowa; Charles R. Stockard, Cornell Medical School;
Ambrose Swasey, Cleveland, Ohio; William H. Wright, Lick
Observatory. : -

Further, Dr. Albert Einstein, of the University of Berlin, was
elected foreign associate.

The presentation of medals was as follows: The J. Lawrence
Smith Medal for important contributions to knowledge concern-
ing meteorites to Dr. George P. Merrill, of the U. S. National
Museum. Also the Daniel Giraud Elliot Medai was awarded
with an honorarium to Dr. O. Abel of Vienna, for his book,
‘*Methoden der parlaobiclogischen Forshung ;

Delegates were appointed to a number of University and
Scientific meetings at home and abroad; notable among these was
the recent (May 14-17) seventh centenary of Padua, and the
one hundred and fiftieth anniversary of the Académic Royale
des Sciences de Belgique at Brussels, May 24.

The Academy voted to accept the invitation of the members in-
New York City to hold its Autumn meeting there, the details to
be arranged by the President and Home Secretary in conjune-
tion with the Local Committee.

2. Considérations sur l’Etre vivant: II, L’Individu, la Sex-
ualité, la Parthénogénése et la Mort, aw point de vue orthobion-
tique; pp. 193. 2. Note préliminare sur l’Orthobionte des
Characées; pp. 18; par CHARLES JANET. Beauvais, 1921
(Dumontier et Hagué).—TIn an earlier paper the author has pos-
tulated his theory of the orthobionte as a series of reproductive
cell aggregates (merisms) which lead from the fertilized egg, or
zygote, of one generation to the same stage in the subsequent gen-

Miscellaneous Scientific Intelligence. 483

eration. Since these reproductive bodies may be homologized in all
eroups of plants and animals they are evidently of phylogenetic
significance. In many of the groups of plants these merisms are
associated with a complex alternation of generations. In these
later papers it is shown how the orthobionte is related to the
individual of the various generations and how sexuality, par-
thenogenetic reproduction and somatic death have pages aga
lished 3 in the different groups of organisms.

3. Finf Reden von Ewald Hering; published by H.. b. Hie
Inc. Pp. 140, with portrait of Ewald Hering. Leipzig, 1921
(Wilhelm Englemann).—These lectures, although delivered
many years ago by the distinguished physiologist Hering, were
so much in advance of the time in their general conceptions that
they prove of real interest at the present day. They treat of
memory as a general function of organic matter, specific energies
of the nervous system, and theories of the vital PE UGESEee, inelud-
ing nervous activity. W. R. C.

4. Board of Scientific Advice for India. Annual Report for
the Year 1919-20. Pp. 111, Caleutta.—The thirty-eighth meet-
ing of the Board was held at Simla on May 17, 1920; the thirty-
ninth at Delhi in December 20, 1920. The president of the
Board is the Hon. J. Hullah, Secretary to the Government of
India; associated with him are eleven gentlemen in the different
departments. Numerous brief statements of matters, of import-
ance, in any ease locally, are given in this publication. Worthy
of note are the researches in solar physics carried on at Kodaika-
nal under the direction of J. Evershed. Measurements of the
displacement of cyanogen bands and also of iron lines were found
to be of the same sign and magnitude as that predicted by the
Kinstein hypothesis. This displacement differs for different sub-
stances and is not proportional to the wave-length, suggesting
the existence of some modifying influence outside of that due to
the gravitational field. Displacements in the Venus spectrum
were unfavorable to the Einstein hypothesis.

OBITUARY.

PRoFEssor Puiuippe A. Guys, the eminent Swiss organic
chemist, died on March 27, in his sixtieth year.

Dr. Henry Newton Dickson, the Scotch geographer whose
special work was in meteorology and oceanography, died
recently, at the age of fifty-six years.

Dr. GEORGE BaLLARD MatHiws, professor of mathematics in
the University of North Wales at Bangor, died on March 19, at
the age of sixty-one years.

INDEX TO VOLUME IIL*

A
Academy, National, meeting in
Washington, 386, 482.

Alnoitic rocks, Isle Cadieux, Que-
bec, Bowen, I.

Achalme, L’Atome, 303.

Ammonium § chloride,

Oi O WAC ones Gly!

Analysis. Colorimetric, Snell, 217

— See CHEMICAL.

Anatomy, Comparative Vertebrate,
Lyman, 200.

Anthropoid, first, found in America,
Osborn, 478.

Arzocyon, Thorpe, 97, 371.

symmetries

Arctic, Cambrian fauna, Holtedahl,
343.
— Cephalopods, Foerste, 304.

— Friendly, Stefansson, 300.

AS TOHONnY, Newcomb-Engelmann,
BO Teens (5

Atomic Structure,
Achalme, 303.

Atoms, chromophore grouping in
inorganic triple salts, Wells, 414.

Auric chlorides. See GOLD.

Aurora line of night sky, 476.

Report on, 94.

Australia, “varve shales,” David,
Sy
B
Bacteriology, Dairy, Orla-Jensen,
155.

Berry, E. W., Carboniferous plants
from Peru, 189; new genus of
fossil fruit, Catatoloides, 251;
American Spirulirostra, 327; Cor-
ocoro Copper District, Bolivia,
480.

Bigelow, F. H., Two Orbit Theory
of Radiation, 382.
Blastomeryx marshi,

Eien. 10'S),

Bolivia, Corocoro Gonpee District,
Singewald and Berry, 480.
Bowen, N: i. “alnortic mocks,

Cadieux,; Quebee, I.

British Museum, Natural History,
publications, 310, 480.

Brown, S. E., Heat and Light, 95.

Brussels, Royal Natural History
- Museum, publications, 302.

restoration,

Isle

Buddington, A. F., Natural and
synthetic melilites, 35.

Burling, L. D., relations between
Purcell Range and Canadian
Rockies, 254.

C

Calculus and Graphs, Passano, 393.

California, shark’s teeth -from,
Jordana 33s:

Canada geol. survey, 150.

Canidz, Oregon Tertiary, Thorpe,
162.

Carex notes, Clokey, 88.

| Carnegie Foundation, 16th annual
hep Ome 225.

— Institution, publications,
208; Year Book, 1021 207

Carnivora, Tertiary, new, in Marsh
Collection, Thorpe, 423.

Cayuga Lake region, minor fault-
Mie in, oe. one. 2263

Ceratopyge fauna in West. No.
America, Raymond, 204.

Cesium triple salts, Wells, 315.

Chamberlin’s study of Megadi-

astrophism, Jones. 30s:

Chemical Analysis, Tower, Quali-
tative, 93; Quantitative, G. McP.
Smith, 93

Chemistry, American, Hale, 94.

== Mhiysicalas VWaliiket. sAvi5ee

CHEMISTRY.

Alum, ferric ammonium, Bonnell
and Perman, 300.

157;

Arsenic, precipitation as _ sul-
phide, Reedy, 217.
Auric chlorides. See gold.

Beryllium, atomic weight, Honig-
schmid and Birkenbach, 378.
Chlorides, complex, containing
gold, Wells, 257, 315, 41455)
Chlorine, separation into iso-
topes, Harkins and Hayes, 92.

Colloids, Hatschek, 370.

Copper, :reaction for, Thomas
-and Carpenter, 145. _

— and iron, iodometric estima-

tions War ks2474 ae
Cyanide, new, Landis, 145.
Ferric salts, reduction with mer-

*This Index contains the general heads, CHEMISTRY. GEOLOGY, MINERALS, OBITUARY;
under each the title of Articles referring thereto are included.

Crystal structure.

Index.

485

cury, McCay and Anderson,|Ditmars, R. L., Reptiles of the

216.
Fluorine, determination, Travers,

Q2.

Gasoline, synthetic, 216.

Germanium, separation
arsenic, Miller, 301.

Gold in complex chlorides, Wells,
257; 315, 414.

Hydrogen, persulphides, Walton
and Parsons, 301.

Mercury, isotopes, Bronsted and
von Hevesy, 300.

Oxalic acid, hydrated,
Smith, 378.

Potassium, sodium, etc., heats of
neutralization, Richards and
Rowe, 474.

Silver oxide, crystal
Wyckoff, 184.

Children, Mentally Deficient, Shut-

tleworth and Potts, 312.

Chironomides of Belgium,

ghebuer, 302.

from

Hill and

structure,

Goet-

|

| Fluid

Geachward) A. Origin of ihe Hu

man Race, 383.

Clark, A. H., Existing Crinoids, 226. |
Clark, Tf. H-; Faconic Revolution,

TE
Clarke, Jj. M:, james Hall,
1898, 220.
Climates, Evolution, Manson, 301.
Clokey, I. W., Carex notes, 88.
Cocciniglie Italiane, Leonardi, 288. |
Colby, C. C., Economic Geography
of No. America, 227.
Colors of substances,
the cause, Wells, 414.
Comstock lode, Nevada, 385.
Crinoids, Existing, Clark, 226.
See Wyckoff.
studies in, Holm,
138; XXXIV, 260.

IStI-

theory for

Cyperacee,
come MOLT

D

Dake, C. L., problem of St. Peter |
Sandstone, 221.
D’Alembert, J., Traité de
mique, 95.

Dana, E. S., Text book of Miner-|
alogy, revised, 155, 224.

Dyna-

Dannemann, F., Die Naturwissen-|

schaften, 140.
David.-T.W.Es, *
Australia, I15.
Davison, C., Seismology, 222.

varve shales” of

Dercum, F. X., Physiology of the |

Mind, 301..
Distillation Principles, Young, 379.

Diwald, K.,

Ferris,

Gastropod

| Geo graphy,

World, 387.
Morphogenese

der
Oetscherlandschaft, Toh

Dominican Republic, Geology, 221.
Dustfall, 1920, Winchell and Miller,

349.
E

Earth Evolution, Hobbs, 223.
Eikenberry, W. L., Teaching

of
General Science, 478.

| Eisenhittenkunde, Osann, 146.
Electrician, Jubilee number, go.
| Engineers’ Tables, Ferris, 312.

F

C. E., Engineers’ Tables,

312.
resistance, Wieselsberger,
Qi

The
Foerste, A. F., Arctic Cephalopods,

| 304.
| Ford, W. E., Dana’s Text Book of

Mineralogy, L555 3224

Fossilium Catalogus, I., Lambrecht,

CS] Cari 133

G

|'Gas, adsorption by charcoal and

StliCas LAz-
trails
IOI;

in sandstones,
Raymond, 108.

Economic, “ot. * No:
Source Book, Colby,

Powers,

America,
Do ye

GEOLOGICAL REPORTS.

Canada, I50.

Illinois, 48o.
Pennsylvania, 305, 384.
South Australia, 384.
United States, 97.
Vermont, I5I1.

West. Australia, 481.
West Virginia, 223.

‘Geology, Commercial, Atlas, 380.

GEOLOGY.
Arzocyon, Thorpe, 97, 371.
Archaeopteryx of British Mu-

seum, 382.
Blastomeryx marshi, Lull, 159

Cambrian, in European Arctic,
Holtedahl, 343.

atic aes Oregon Tertiary,
Thorpe, 162.

Calatoloides, Texas, 252.

Cephalopods, Arctic, Foerste,
304.

486

Echini, Fossil, of West Indies,
Jackson and Vaughan, 479.

Faulting, minor, in Cayuga Lake
region, Long, 229.

Fauna, Ceratopyge, in West. No.
America, Raymond, 204.

— Hawaiiensis, Sharp, 388.

Fishes, ‘Triassic, from Spitz-
bergen, 479.

Foraminifera of the Tortugas,
Cushman, 479.

Fruit, fossil, new, Berry, 251.

Glaciation in Japan, Yamasaki,
131° So. Arica, 3o4-

Herring, fossil, from Texas,
Jordan, 249.

Hesperopithecus, Nebraska, Os-
born, 478.

Huenella, Arctic, 343.

Ischyromys, Oligocene, ‘Troxell,
123.

Miocene of Costa Rica, Olsson,

479.
Mollusca, Gulf of Mexico, 305.
© lake oxclennze Hyznodontide,

horpe, 277:

Pennsylvanian of Texas, Plum-

mer and Moore, 305.

Plants, Carboniferous, from Peru,

Berry, 189; Texas, 252.
Purcell Range and Rocky Mts.,

relations, Burling, 254.
Sandstone, St. Peter, Dake, 221.
— gastropod trails in, Powers,

tor; Raymond, 108.

Skull, prehistoric human, from

Rhodesia, Woodward, 96.
Taconic Revolution, Clark, 151.

Geometry, Heffter, 96.

=" Poincare, 200:

Glacial climate, Sayles, 456.
Glaciation. See GEOLOGY.
Glass, Optical, Wright, 303.

Gold chlorides, Wells, 257, 315, 414.

H

Hale, H., American Chemistry, 94.

Hall, James, 1811-1808, Clarke, 220.

Hamilton, L. F., melanovanadite,
Peru, 1095.

Haswell, W. A., Zoology, 386.

Hatschek, E., Colloids, 379.

Hawaii, Bishop Museum publica-
TOMS; ES

Hawaiian, augite, Washington and,
Merwin, II7.

Headden, W. P., tantalate and col-
umbites, So. Dakota, 203.

Heat and Light, Brown, 95.

Index.

Heffter, Geometrie, 96.

Helaletes redefined, Troxell, 365.

Hering, E., Funf Reden, 483.

Hobbs, W. H., Earth Evolution,
223.

Holm, T., Cyperacex, XX XIII,
138; XXXIV, 260.

Holtedahl, O., Cambrian fauna in
European Arctic, 343.

Homogalax, Troxell, 288.

es Race, Origin, Churchward,
353.

— remains, Rhodesia, Ne-
braska, 478; 480.

Hyznodontidz, Thorpe, 277.

96;

I

Illinois geol. survey, 480.

— State Water Survey, 481.

eee Board of Scientific Advice,
453.

Isotopes, mercury, 300; separation,
Aston, 380; separation from
chlorine, 92.

Japan, glaciation, Yamasaki, fame

Janet, Orthobionte, 482.

Jones, W.F., Chamberlin’s study of
Megadiastrophism, 393.

Jordan, D. S., fossil herring from
Texas Miocene, 249; shark’s
teeth from California, 338.

L

Lambrecht, K., Fossilium Cata-
losis) alert i

Leander McCormick Observatory,
Ale:

Lecat, M., Séries Trigonométrique,

QOn

eae G., Cocciniglie Italiane,
300.

Lewis, J. V., Mineralogy, 154.

Library of Congress, report, 158.

Lindgren, W., melanovanadite,
Pert, tO5:

Long, E. °T., minor 4anlameeam
Cayuga Lake region, 229.

Lull, R. S., restoration of Blasto-
meryx marshi, 150.

Lyman, L. H., Comparative Verte-
brate Anatomy, 300.

M

Man. See Human.
Manson, M., Evolution of Climates,

301.

Index.

Marsh Collection of Vertebrates,
Ral 150;.. Thorpe,! 97,.-162; -277,
371, 423; Troxell, 123, 288, 365.

Mass and weight, proportionality,
Brush, 477.

Matisse, G., Movement
tiique, France, 227.

McCallie, S. W., Pitts meteorite,

Scien-

BET.

Megadiastrophism,
theory, Jones, 393.

Melilites, synthetic, 35.

Chamberlin’s

Merrill, G. P., new meteorites, 153,
425, 386; metamorphism in
meteorites, 307; meteoric iron,

Odessa, Texas, 335.

Merwin, H. E., Hawaiian Augite,

t17.
Meteorite, Pitts, McCallie, 211.

Meteorites, metamorphism, Merrill,
Merrill,

OF; new, L5G 225i
Odessa, Texas, 335, 386.
Miller, E. R., 1920 dustfall, 340.

Mind, Physiology, Dercum, 301.
Mineral Production, Distribution,

389; of the United States, etc.,
384.

Mineralien Physiographie, etc.,
Rosenbusch, 152.

Mineralogy, Dana’s Textbook, re-|

vised, Ford, 155, 224.
— Determinative, Lewis,

154.
MINERALS.
Adularia, 154. Augite, Hale-
akcdla. 57.
Collophane, 260. Columbites, |
South Dakota, 297.
Feldspar studies, Japan, 154.
Melanovanadite, Peru, T95. |
Melilites, 35. Moonstone, 154.

Tantalate, South Dakota, 293.

Mines, U. S. Bureau of, 306.
Molecule, asymmetry of the gas-
eous, Gans, 302.

N

Naturwissenschaften,
Dannemann, 149.
Newcomb-Engelmann, Astronomy,

381.

Vol. rl

O
OBITUARY.
Bottomley, J. F., 313. Branner,
j te Oe se 5

GCentcian, G...L.,
Dickson, H. N.,

314.
483.

Physical Constants,

| Pliocyon marshi,

487
Guye, P. A., 483.
jlordanyM. GS 302:
Liebisch, T., 302.
Mathews, G. B., 483. McFarland.
BoeW.; 32 Moore Bs 302:
Verworn, M., 314.
Waidner: ©. + W.; - 384. ] Waller,
ALD 202.

Observatory, Leander McCormick,
Bees

Oregon, Tertiary Canide, Thorpe.
162. f

cane Jensen, Dairy Bacteriology,

Bere B., Eisenhtittenkunde, 146.

Pp

'Palache, Ce Melanov anadite, Peru,

195.
Paleoclimatologists, dilemma of,

Sayles, 456.

|Pan-Pacific Scientific Conference,

300.
Parker, T. J., Zoology, 386.
Passane, Jz. - MLS Catentus
Graphs, 303.

and

| Pauli, W., Colloid Chemistry of the

Proteins, A75
Pennsylvania, zeol. survey, 305, 384.
Peru, Carboniferous plants, Berry.
£89.
148.
149.
and Chemistry,

== Kaboratory, £620;

Physics, Mining
Whitaker, 146.

Physik, Gehrcke, 96.

change of name,

Thorpe; 97.
Poincaré, H., Geometry, 219.
Powers, S., gastropod trails in
Sandstones, IOI.
Proteins, Colloid Chemistry, Pauli,
475.
Q
Quebec, alnoitic rocks, Bowen, I.
R
Radiation, Two Orbit Theory,
Bigelow, 382.

Raymond, P. E., Seaside notes, 108;
Ceratopyge fauna in West. No.
America, 204.

Reptiles of the World, Ditmars,
387.

Resistance,
Zi fs

Rhodesia, prehistoric human skull,
W oodw ard, 96.

fluid, Wieselsberger,

488

Rocks, alnoitic, Bowen, 1.

Rodents, Ischyromys, Oligocene,
Troxell, 123.

Rogers, A. F., collophane, 260.

Ronald Press Co. publications, 08.

Rosenbusch, H., Physiographie der
Mineralien, etc., 152.

S

St. Peter sandstone, Dake, 221.
Sayles, R. W., dilemma of Paleo-
climatologists, 456.

Schwabe, E., Weber’s Weltge-
schichte, 311.
Science, ‘Teaching of General,

Eikenberry, 478.

Sea, color of, Raman, 476.

Seaside notes, Raymond, 108.

Seismology, Davison, 222.

Shark’s teeth, California Pliocene,
Jiondan-.338.

Sharp, D., Fauna Hawaiiensis, 388.

Sherman, H. C., Vitamins, 310.

Silver oxide, crystal structure,
Wyckoff, 184.

Smith, G. McP., Quantitative Chem-
cal Analysis, 93.

Smith, 8S. L., Vitamins, 310.

Smithsonian Institution, report,
156.

Snell, F. D., Colorimetric Analysis,
ZN 7).

Sound, velocity at high tempera-
tures, 94.

South Africa, glaciation, 384.

South Australia geol. survey, 384.

Spirulirostra, American, Berry, 327.

Stefansson, V., The Friendly Arctic,

300.
T

Tables, Physical, 148.

Taschenberg, O., Bibliotheca Zo-
ologica II, 98, 312.

Texas fossil fruit, Berry, 251; her-
Rimes Jondanwe2A0)

Thorpe, M. R., Areocyon marshi,
O7: * Orecon. Mertiakyas amides
162%
new, Genus, 277; Arzeocyon.e70-
(henitary: Carnivoran news 42o

Tierwelt, die Antike, Keller and

Staiger, 383.
Tower, O. F., Qualitative Chemical

PASIalleyaS iSO se

Oligocene Hyenodontide,

Index.

Trigonometric Series, Lecat, 06.

Troxell, E. L., rodents of genus
Ischyromys, 123; Homogalax,
288; Helaletes redefined, 365.

U

United States Bureau of Mines, 306.
— geol. survey, 97.

V

Varve Shales, of Australia, David,
inns ‘

Vermont geol. survey, 151.

Vitamins, Sherman and Smith, 310.

W

Walker, Sir James, Physical Chem-
US tive NAG 5.

Washington, H. S,,
Augite, 117.

Water Power of the
Stabler, Jones, etc., 380.

Weber’s Weltgeschichte, Schwabe,
2m 1

Wells, H. L., Complex chlorides
containing gold; 257. ssi sue
chromophore grouping of atoms
in inorganic triple salts, 417.

Weltgeschichte, Weber’s, 311.

Western Australia geol. survey, 481.

West Virginia geol. survey, 223.

Whitaker, Z. W., Mining Physics
and Chemistry, 146.

Winchell, A. N.; dustfall of 1920,

Hawatian

World,.

340.
Wright, F. E., Optical Glass, 303.
Wyckoff, R. W. G., symmetries of

ammonium chloride, 177; crystal

structure of silver oxide, 184.

Y
Yamasaki, N., glaciation in Japan,

me
Young, S., Distillation Principles,

379. pi
Zz }
Zoologica, Bibliotheca, II, Taschen-

Po DCLS OS. 3021 ak
Zoologischer Anzeiger, Index, 311.
Zoology, Parker and Haswell, 386.

|
j
{
|
|

RARE MINERAL SPECIMENS

NEW SCANDINAVIAN MINERALTS, obtained by our Mr.

Geo. L. English during his very recent European trip.

CORNISH MINERALS

Mr. English visited Cornwall and secured the pick of one of
the best collections there, and we are now able to offer exception-
ally fine crystallized material, among which the following species
are noteworthy:

BISMUTHINITE OLIVENITE
BLISTER COPPER PHARMACOSIDERITE
BORNITE, crystallized SIDERITE (Lady-Slippers)
BOURNONITE TENNANTITE
CHALCOPHYLLITE TETRAHEDRITE
CLINOCLASITE TORBERNITE
GOTHITE, crystallized URANINITE (PITCHBLENDE)
LIROCONITE WOOD TIN

Also

A superb lot of AMETHYST specimens from Brazil, both large
and small, priced at $1.50 to $175.00.

KRANTZ CRYSTAL MODELS

Models of pear-wood, 5 and 10 cm., glass, cardboard and
papier mache. Collections to suit every purpose. Descriptive
leaflet on request.

WARD'S NATURAL SCIENCE ESTABLISHMENT
DEPARTMENT OF MINERALOGY AND PETROGRAPHY

ALFRED C. Hawkins, Ph.D., Manager

84 College Avenue, Rochester, New York

The American Journal of Science
ESTABLISHED BY BENJAMIN SILLIMAN JIN 1818.

THE LEADING SCIENTIFIC JOURNAL IN THE UNITED STATES.

Devoted to the Physical and Natural Sciences, with special refer-
ence to Physics and Chemistry on the one hand, and to Geology and
Mineralogy on the other.

Editor: EDWARD S. DANa.

Associate Editors: Professors WILLIAM M. DAVIS and REGINALD A.
DaLy, of Cambridge; Professors H. L. WELLS, C. SCHUCHERT, H.
E. GREGORY, W. R. Cok and F. E. BEACH, of New Haven: Professor
EDWARD W. Berry, of Baltimore; Drs. FREDERICK L. RANSOME and
WILLIAM Bowlkg, of Washington.

Two volumes annually, in MONTHLY NUMBERS of about 80 pages each.

This Journal ended its first series of 50 volumes as a quarterly in 1845; its
second series of 50 volumes as a two-monthly in 1870; its third series as a
monthly ended December, 1895 A FIFTH SERIES commenced in 1921.

CONTRIBUTORS are requested to revise their manuscript with special care with
reference to the use of the linotype machine ; changes involving overrunning are
expensive. In general, the publisher cannot undertake to assume the cost of more
than minor correctious in the proof. All corrections should preferably be made
in ink.

Foot notes should be numbered consecutively from 1 up.

The following is an example of the system of references employed, the coldne
being given in heavy faced type:

Am. J. Sei., 44, 249, 1917. The year should be given in every case; this
renders a series number unessential.

Thirty separate copies of each article will be furnished to the author free of
cost and without previous notice from him. They will be provided with a plain
cover (but with reference to volume and year). If the author orders separate
copies, they will be understood to be in addition to the thirty mentioned above, and
he will receive a bill for the extra expense involved, as also for that of a printed
cover (with title, ete), when this is specially ordered. These charges will be as
moderate as possible; the rates will be increased if the article is accompanied
by plates. or involves unusual expense.

Subscription price $6 per year, or 50 cents a number, postage prepaid m the
United States; $6.25 to Canada; $6 +40 to foreign subscribers of countries in the
Postal Union. A few sets on sale of the early series. Ten-volume index
numbers on hand for the second, third and fourth series.

(GS Ten-volume Index. Vols XL-L, Fourth Series, price two dollars each.

Address.
THE AMERICAN JOURNAL OF SCIENCE,
New Hlaven, Conn.

/arv’s NaturaL Science EstaBiisHMENtT
A Supply-House for mcientinic Material.
Founded 1862. Incorporated 1890.

A few of our recent circulars in the various
departments:

Geology: J-32. Descriptive Catalogue of a Petrographie Col-
lection of American Rocks. J-188 and snpplement.
Price-List of Rocks.

Mineralogy: J-220. Collections. J-238. Minerals by Weight.
J-224. Autumnal Announcements.

Paleontology : J-201. Evolution of the Horse. J-199. Pale-

: ozoic index fossils. J-115. Collections of Fossils.

Entomology: J-33. Supplies. J-229. Life Histories. J-230.
Live Pupe.

Zoology: J-223. Material for dissection. J-207, Dissections
of Typical Animals, etc. J-38. Models.

Microscope Slides: J-189. Slides of Parasites, J-29. Catia
logue of Slides:

Taxidermy: J-22. North American Birdskins. Z-31. General
Taxidermy. .

Human Anatomy: J-87. Skeletons & Models.

General: J-228. List of Circulars & Catalogues.

Wards Natural Science Establishment
84-102 College Ave., Rochester, N. Y., U.S. A.

“SCIENTIA”

INTERNATIONAL REVIEW OF SCIENTIFIC SYNTHESIS. Jssued monthly (each .
number consisting of 100 to 120 pages). Editor: EUGENIO RIGNANO.

This is the only review which has a really international collaboration ; which is
of world-wide circulation and occupies itself with the synthesis and unification of
knowledge, in the history of the sciences, mathematics, astronomy, geology, physics,
chemistry, biology, psychology and sociology. é

It is the only review which, by inquiries among eminent scientists and writers
(on: The philosophical principles of the sciences; Fundamental astronomical and physical
questions of current interest; Contributions given by the various countries to different
branches of knowledge; Question of vitalism; Social question; International questions
raised by the world war), makes a study of the problems interesting scientific and intel-
lectual circles throughout the world.

It has published articles by Messrs.: Abbot=Arrhenius =Ashley = Bayliss - Beichman-
Bigourdan - Bohlin - Bohn= Bonnesen - Borel = Bouty = Bragg - Bruni = Burdick - Carver =
Caullery = Chamberlin = Charlier = Claparede = Clark = Costantin - Crommelin = Crowter =
Darwin = Delage = De Vries = Durkheim = Eddington = Edgeworth - Emery = Enriques =
Fabry = Findlay = = Fisher = Fowler = Golgi =- Gregory = Harper = Hartog = Heiberg - Hinks=
Hopkins-Inigues-Innes-Janet-Kaptein-Kaye-Kidd- Langevin-Lebedew-Lloyd Morgan=
Lodge = Loisy = Lorentz = Loria -Lowell = MacBride = Meillet =-Moret-Muir-Peano =Picard =
Poincare - Puiseux - Rabaud = Rey Pastor = Righi-Rignano-Russell-Rutherford-Sagnac
Sarton- Schiaparelli- Scott = See - Sherrington = Soddy = Starling - Svedberg = Thomson =
Thorndike-Turner -Volterra-Webb-Weiss=Zeeman and more than a hundred others.

**SCIENTIA”’ publishes its articles in the language of its authors, and joins to the

‘principal text a supplement containing the French translations of all the articles that

are not in French. (Write for a Specimen Number to the General Secretary of
“Scientia”, Milan, sending,-to defray postal and other expenses,-2 Francs in
stamps of your country.) Annual subscription : 40sh., or 19 dollars post free.

Office: 43 Foro Bonaparte, Milan, Italy
Publishers: WILLIAMS & NORGATE-London; FELIX ALCAN -Paris

NICOLA ZANICHELLI-Bologna; RUIZ HERMANOS-Madrid;
WILLIAMS & WILKINS CO-Baltimore.

-

Arr. XXXVII.—A Critical Review of Chamberlin’s Ground- me
work for the Study of Megadiastrophism; by W. F. Jonzs, 393 |f

Art. XX XVITI.—Some Complex Chlorides Cox taining Gold.
III. A New Cesium-Auric Chloride; by H. L. Wats, 414

Art, XXXIX.—A Chromophore Grouping of Atoms in In-

CONTENTS.

organic Triple Salts, and a General Theory for the Cause

of the Colors of Substances; by H. Le W Bus! Se ty | q

Art. XL.—Some Tertiary Carnivora in the Marsh Vollection, | |
with Descriptions of New Forms; by M. R. Taorpr, .. 423 |)

| ART, XLI.—The Dilemma of the Paleoclimatologists; by R.

WwW. SAYLES, Harvard University, *- 2.05. v2.7 eee .. 456

SCIENTIFIC INTELLIGENCE.

Chemistry and Ph ysics.—Heats of Neutralization of Potassium Sodium and
Lithium Hydroxides, etc., T. W. RicHarps and A. W. Rows: Rapid

Todometric Estimation of ¢ Jopper and Tron in Mixtures of their Salts, I. W.

Wark, 474.—Introduction to Physical Chemistry, J. WALKER: Colloid

Chemistry of the Proteins, W. Pau, 475.—The Auror:. Line of the Night

Sky: The Color of the Sea, RAMAN, 476.—Proportionality of Mass and | e
Weight, C. F. Bruss, 477.—The Teaching of General Science, Ww. L. |.

EIKENBERRY, 478,

Geology and Mineralogy.—Hesperopithecus, the first anthropoid primate Found
in America, H. F. Osporn, 478.—Shallow-water Foraminifera of the Tortugas

Region, J. ‘A. Cusnman: Fossil Echini of the West Indies, R. T. JACKSON:
Triassic Fishes from Piao E, A. §. Srensié: The Miocene of
Northern Costa Rica, A. A. Otsson, 479.—The Geology of the Coroeoro

Copper District of Bolivie, BB i SINGEWALD, Jz., and KE. W. Berry: A iB e

Guide to the Fossil Remains of Man, etc.: Illinois Geological Survey, F.

W. DeWotr, 480.—Illinois State Water Survey: Geological Survey of on

Western Australia, A. GIBB Maituanp, 481.

Miscellaneous Scientific Intelligence. —Washington meeting of the Natignals 5 >
Academy of Sciences: Considérations sur PBtre vivant, 482.—Fiinf Reden .

von Ewald Hering: Board ef Scientific Advice for India, 483.

Obituary.—P. A. Guys: H. N. Dickson: G. B. Marnews, 483.
InpEXx: 484.

iy a =

5» oe
°" an
vr

’ é pS 4” 3 o.
- a a ’

oy
fay

tg s

im
4 s
‘acai!
a
}
Ul
U =
a by

or

ee wb c ff Ts om ; .
1 Bet
i =n Vp ~ neltes ly
yale ; a 4 s F |
Ee on No } fe |
cet) 7 |

sar

et

te

BP

theses

Sal

a

Kate
ek Era

eects

Oe
Fe,
eed

vi

Ty
is

Figs |
=i | g
hey ;

a3

2 Jeers
a my

te TY
Oe

tp
on
ron

Es
&
3
=
BN.
Ot Aa
z ane
etree
Se

P
Ee
oe

cy pee
Se AE BR

Gee, 4
hr

im
i hte

& Lh “he A :
4 ¢ pate
hte LEE gh |
ti ae iva See!

Sage

Naseer

oe se

ly sti Aas |
Shine an

“Sintemaeapeetre”

mire
Ble

a ets
is gd BORA

- ok
a 5
H Rog

\

Xu

mF YS
2 tA
8 +

we

Syd
Oy cg

nL oe
ANSE fi

lt A
f

TT

3 9088 01298 6063