RCOHMN NCSC
My ‘ 4 " ff fi
: at sh AY i na i
sat PLANE NNN (
Ny NANA tah A Ae AN tA RD )
SHARON DEA
BONO RNIY i
NNT RNe y

WAP AM iN SSR NN A PaRRATN
Wi At hy be % re ‘ | 1h) \
PRAM Hh EOD i NMP OHO
Nh Avs iy 7 ; - Thy ei
1h i ahebattt iy

f

\

aM Hae
oe :
Taian

Rte

vil
Nb

Na :
my hylidy Moy dad
ih .

et tee
ee $

Pacis St

SS

AHS nity 4 by Fly Anis
ANY PAA in Ne

PEON
a} f,
ay

=e

aS

aa

Bare

Se

ih j
WON i
ANALG
\ Wy
ff

SOM '
Tee pets iene bE
COL R yO) Kati
GMM OR nen

:
5

A
Ghai
eit

Soe eee

i hee ih
WSU ES gd
nals We

Se es
<2

Bory
¥i iN) a
SUNN ASP tty
y }. Cay
f Aa ORE ft At Ne 3G ui tH ‘
HM ey ech ath ath CMHC
ANNAN aS bas
i POH Le Af
May

— Sos on

=<

| '
fale
OAM ID
ies , h
RAK Ma tk i f Ay iit
ie eat i AHEM MITCH A TGA Con NA
AA POSEN RA ii} MOH ia elitiely i
) Ua PMR Si ore at i

iM
PPR A

iS ri
Whe ¢
Nc

=

f

|

| |

ii

‘I,

o "4
UM
“>

———7

7, Z
py SEZ

5
“>
A,

|

—————
=n

il

al
it
'
ery:
7 ;
ro ‘ :
i
t 5)
oy yi
‘ i i
= Le
Dik
PL
A c
WM
7 i :
i a
i
i ie ee

J
4
ive

on aK

j
p
i
i
'
|
'
t
‘
~ i!
'
1
“
.
t
Ww \
Me
‘
4
5
i
i
5
é
¥ ’
.
'
‘
:
) a
ih bi 7
® =
‘
i
1a
UJ 7 Ger
ta
t O 7
ii t a
a 4 ¥
i '
7 T ”
Ay ’
:
i
'
on «
- i
i
i i
: f
Delay)
. ‘ ies
- : ai? i i
i Ha
ro : , rhs
I ; Wl i
, : | i J i a hy Au
al i

; AN
bet | erika f

Aelia,

(OUR OF GEOLOGY

VOL NaAOUG OS 7, 1902

iit DEVELOPMENT OF SYSTEMATIC PETROGRAPHY
IN Dae NINETEENTH CENTURY.

Part IT?

Beginning of the microscopical era.— Having presented the state
of systematic petrography up to the time when the polarizing
microscope became the instrument of prime importance in the
investigation of rocks, we now proceed to the study of the more
recent schemes of classification, based upon the larger knowl-
edge. For all systems thus far reviewed, it must be recognized
that ignorance of the characters and relationships of many rocks
rendered a comprehensive and logical scheme impossible. The
frame work of system was necessarily constructed without full
knowledge of the applicability of some of the factors employed.
With the polarizing microscope and improved methods of chemi-
cal analysis this condition has now disappeared, and while the
time may not be ripe for unbiased criticism it is plain that
modern systems of classification must ultimately be judged with
regard to the greater and almost perfect knowledge of the actual
characters of rocks enjoyed by the authors of those systems.
The responsibility for the choice of factors suitable for the con-
struction of a comprehensive system and for a logical, conse-
quent, and consistent application of those factors clearly increases
with knowledge of the objects to be classified.

The microscopical study of rocks, continuing for nearly four

*Continued from p. 376.

Vol. X, No. 5. 451

452 WHITMAN CROSS

decades with ever improving facilities and methods, has added a
vast store of knowledge concerning the characters of these
objects. Revelations concerning the composition of types long
known have often been astonishing. The essence of rock struc-
tures has been made clear. The uttermost parts of the earth
have been searched, and many new and interesting varieties have
been discovered—not all in distant fields, but often close at
hand. For new structures and new types, new terms have been
proposed, and the nomenclature has thus expanded enormously.
In the light of new discoveries, old conceptions have given way
to new ideas, and the terms expressing them have yielded to
new ones, or, in too many cases, the old nomenclature has been
retained with new definitions. But while the last third of the
nineteenth century may be termed the microscopical period of
petrography, great additions to our knowledge of rocks were also
made in this period on all older lines of study, and especially
by quantitative chemical analysis and by investigations as to the
modes of occurrence and the field relations of different rocks.

This review, which deals with system, must trace the applica-
tion of new or revised principles to classification during this.
period of rapid addition to knowledge. It is perhaps quite
natural that the greater part of the systematic advance was in
partially worked out revisions of old schemes—grafting some
new idea to the old trunk. It is also natural that the greatest
work has been in the field to which the microscope has been
particularly applied, so that at times petrography has been
treated as though narrowed to the microscopical petrography of
igneous rocks.

While the flood of microscopical rock studies was at its
height, it was manifestly impossible for anyone to do more than
to present the new information in comprehensive form, without
finished attempt to apply it to the systematic arrangement of
rocks. It is with this condition in mind that the important
works upon microscopical petrography, issued in the decade
1870 to 1880, must be judged. Their actual effect upon the:
systematic science was, however, very great, from the mere fact

SYSTEMATIC PETROGRAPHY 453

that they exerted a controlling influence over the usage of a
multitude of workers.

Ferdinand Zirkel, 1873.—In 1873 appeared Die mikroskopische
Beschaffenheit der Mineralien und Gesteine, by F. Zirkel. The
systematic point of view occupied by this authority at this time
is expressed in the following tabular analysis of his larger
divisions of rocks:

A. Non-clastic rocks (‘‘ Nicht-klastische Gesteine”’ ).
I. Simple (‘‘ Einfach” ).
II. Composite ( ‘‘Gemengt”’ ).
I. Massive ( ‘‘ Massig’’).
a. Feldspathic (“ Feldspath-haltig’’ ).
6, Non-feldspathic (‘ Feldspath-frei’”’ ).
2. Schistose (‘Schieferig’’).
B. Clastic-secondary rocks (‘“ Klastische-deuterogene-Gesteine ”’ ).

Comparing this scheme with that of the Lehrbuch der Petro-
graphie, we find that the microscope has convinced Zirkel that
crystalline cannot be appropriately opposed to clastic. He also
believes that ov7gina/ cannot be used for non-clastic rocks because
they have been discovered to be at the present time in part
composed of alteration products. He therefore falls back
on a negative term, zonclastic, admitting that it is indefinite —
“‘misslich.” JZasstve is defined as not schistose, in larger part
granular — ‘nicht geschiefert, zum grossen Theil kérnig.”’

The work deals mainly with the feldspathic, massive, composite,
non-clastic rocks. These are subdivided as follows:

I. Orthoclase rocks.
I. With quartz.

2. Without quartz, with or without plagioclase.

3. Without quartz, with nephelite ( or leucite).
II. Plagioclase rocks.

With hornblende.
With augite.

With diallage.
With hypersthene.
With mica.

. With olivine.

III. Nephelite rocks.

IV. Leucite rocks.

nm & WN

454 WHITMAN CROSS

The use of the soda-lime feldspars—oligoclase, labradorite,
and anorthite, as factors in the subdivision of feldspathic rocks,
found in the Lekréuch, had been shown by the microscope to be
an error, and it disappears without comment. The mineral com-
position of rocks is here applied to their classification in a quali-
tative way almost exclusively. Zirkel remarks: ‘‘To the non-
feldspathic, non-schistose, composite rocks belong among others:
eclogite, tourmaline rock, olivine rock, eulysite and saussurite-
gabbro.”’ Of these only three are described and within a com-
pass of two pages.

The age distinction in classification of igneous rocks is freely
characterized by Zirkel as unnatural and undesirable and it is
not formally recognized in this work as it was in the Lehrbuch,
yet he could not see his way to carry out the reform necessary
to its rejection and retained in description many of the duplicate
terms based on it.

Structure and crystalline condition were not given a defined réle
in Zirkel’s new system, but in practice the granular, porphyritic,
fluidal and glassy forms were distinguished.

A. von Lasaulx, 1875.—The Elemente der Petrographie,* by A.
von Lasaulx, issued in 1875, is another attempt to utilize the
results of microscopical study of rocks in their classification and
description. Von Lasaulx believed that since there are no true
rock species, and since transitions in all directions are most
common, classification must consist in the establishment of tyes,
about which should be grouped the intermediate kinds of rocks.
He announced as his guiding principle that rocks must be classi-
fied upon the basis. of simple, definitely known and easily
recognized, morphological properties. Genetic criteria did not
seem to him applicable because always more or less hypothetical
and in some cases entirely, so; ‘He, therefore; \discardsathe
primary classification of rocks on genetic principles, advocated
by von Cotta and others, and returns to von Leonhard’s ele-
mentary division into Szmple, Composite and Clastic rocks, omitting
the Apparently simple class as no longer necessary.

tElemente der Petrographie, Bonn, 1875, pp. 486.

SYSTEMATIC PETROGRAPHY 455

The morphological characteristic of rocks chosen by von
Lasaulx as most applicable to the systematic subdivision of the
classes mentioned was the degree or distinctness of crystallinity,
which is surely the most variable of all their properties, and
hence least adapted to the formation of well-defined groups.
The major divisions formed by von Lasaulx are as follows :

Simple Rocks :
A. Non-crystalline (amorphous) or semi-crystalline.
B. Crystalline granular.
a. Really simple.
6. Rocks forming transitions to Composite group through the appear-
ance of vicariousconstituents (¢. ¢., amphibolite, serpentine, etc.).
Composite Rocks :
A. Massive.
a. Amorphous, glassy (obsidian, etc.).
6. Semi-crystalline (including vitrophyres).
c. Crystalline.
1. With abundant glassy base (basalt, etc.).
2. With microaphanitic more or less individualized groundmass,
aa. Groundmass alone (felsite, etc.)
66. True porphyries (felsite-porphyry, etc.).
3. Rocks which are almost completely crystalline, mainly pseudo-
porphyritic, etc. (phonolite, hornblende-andesite, etc.).
4. Crystalline granular.
aa. Feldspathic (granite, etc.).
66. Non-feldspathic (Greisen, eclogite, etc.).
B. Stratified Rocks:
a. Feldspathic (gneiss, etc.).
6. Non-feldspathic (mica-schist, etc.).
Clastic Rocks :
A. Semi-clastic (clay slate, kaolin, tuff, etc.).
B. Purely clastic.

Mineral composition is applied as a factor to produce the
commonly recognized rock types within these groups. It will
be noted that many of the divisions above mentioned are not
only quite indefinite but they are, also, inconsequent. Thus,
three out of four divisions of crystalline massive rocks are but
partly crystalline.

Meteorites are described by von Lasaulx in an appendix.

456 WHITMAN CROSS

Concerning this treatment he remarks that it is the first time
that cosmic rocks have been givena place in a text-book of
petrography, but that it seems useful, for purposes of comparison,
to have them described in the same work with the terrestrial
rocks.

H, Rosenbusch, 1877.—Another important summary of the
results of the microscopical investigation of rocks appeared in
1877, under the title Mikroskopische Physiographie der massigen
Gesteine, by H. Rosenbusch. There was in this work but slight
discussion of principles of classification, and the only new factor
of note in the system used is the idea expressed in the title,
which requires some explanation. All rocks were divided into
two classes:

I. Massive rocks (‘‘ Massige Gesteine”’).

II. Stratified rocks (“ Geschichtete Gesteine”’).

A class of Metamorphic rocks was not considered feasible.

‘‘Massive’”’ and ‘stratified,’ as used by Rosenbusch in this
connection, do not refer to rock textures, as one might suppose
from the historic use of the terms; for this primary division was
avowedly intended to express an idea, first brought out by
Lossen (which will be referred to more fully in a later section of
this review), that the most important relation of rocks is the
formal one to the earth sphere. Rocks may be considered as
having been formed, either at the surface under the influence of
gravity, in more or less concentric shells or strata, or, in eruptive
bodies of irregular shape and position not determined by gravity.
Under this conception all rocks are either stratified or massive.
Massive rocks in this sense are also eruptive rocks, but Rosen-
busch chose to use the former term in systematic petrography
because free from the genetic conception involved in the latter.
In view of the evolution of this master’s ideas, presented in
later works, his language will be quoted: ‘‘The former name ts
to be preferred because it refers only to an undeniable form of

“Ter erste Name ist vorzuziehen, weil er sich lediglich auf eine unlaugbare

Erscheinungsform bezieht und keinerlei irgendwie geartetes Prajudiz iiber die genet-
ischen Verhaltnisse involvirt.”

SYSTEMATIC PETROGRAPHY 457

occurrence, and involves no possible prejudice concerning
genetic relations.”

The controlling criteria in the further construction of Rosen-
busch’s scheme will be apparent from the following partial
tabular statement :

Massive rocks.
A. Orthoclase rocks.
@aOlder:
I. Quartzose.
1. Granular = Granite Family.
2. Porphyritic = Quartz porphyry Family.
3. Vitreous = Felsite-Pitchstone Family.
II. Quartz free (3 families similar to those above).
6. Younger.
I. Quartzose.
1,2. Granular or porphyritic= Liparite Family.
3. Vitreous = Family of the acid glasses.
II. Quartz free (with families as above).

The other large groups are the following :

. Orthoclase-Nephelite, or Orthoclase-Leucite rocks.
Plagioclase rocks.
. Plagioclase-Nephelite, or Plagioclase-Leucite rocks.
. Nephelite rocks.
. Leucite rocks.
. Non-Feldspathic rocks, Of the Non-feldspathic rocks Rosenbusch
remarked! that they were all rich in olivine, and might therefore be
called Olivine rocks.

Atmuas

Each of these groups has Older and Younger divisions, and,
within these, families, established in a manner similar to that
given for the orthoclase rocks.

In this arrangement mneral composition is used, as in Zirkel’s
system. The age distinction is applied without discussion.
Texture is given a prominent role, and chemical composition is
not used.

Fouqué and Michel-Lévy, 1879.—The first effects of the
microscopical study of rocks upon petrographic system in France
may be seen in the Minéralogie micrographique, by F. Fouqué

«Sie sind sammtlich reich an Olivin, daher kann man sie kurz als Olivinge-
steine bezeichnen.”

458 WHITMAN CROSS

and A. Michel-Lévy, which was published in 1879. While
presented as an “Introduction a l’étude des roches éruptives
francaises,”’ there is in this work some discussion of principles of
classification, and a tabular view of the scheme in use by the
authors. Although the system in question has not had much
influence except in France, it is of interest from certain new and
peculiar conceptions which are given classificatory value in it.

It is first to be noted that Fouqué and Michel-Lévy abandon
almost entirely the system of Cordier. Although affirming that
rocks are simply the most abundant natural associations of min-
erals (‘‘les roches ne sont autre chose que les associations minér-
ales naturelles le plus fréquentes”’), they proceed to their
arrangement under the stated principle that a rational classifica-
tion of rocks in general must be based upon the following three
fundamental characters, namely: (1) The mode of formation;
(2) the geological age; (3) the specific mineral properties.
The last named character comprises: (a) the. nature of the
integrant minerals; (6) the structure of the rock.t By the
application of the first factor they separate eruptive rocks from
those deposited as sediments or as vein filling. By the factor
of geological age (the applicability of which is not discussed)
they distinguish clearly —‘‘nettement’’—between pre-Tertiary
and Tertiary or post-Tertiary rocks. It is believed that the
same types occur in both groups, following approximately the
same order of eruption, with predominance of basic rocks and a
tendency to the vitreous condition in the most recent occur-
rences. In these respects the new French system is practically
like others which have been reviewed, but in the use of structure
and mineral composition for the main framework of their system,
Messrs. Fouqué and Michel-Lévy apply certain peculiar con-
ceptions requiring some analysis.

Considering that in practically all eruptive rocks there have
been two (or more) distinct periods of formation of the primary

«Un classement rationel des roches, en général, doit s’appuyer sur les trois
caractéres fondamentaux suivants: 1° le mode de formation; 2° l’age géologique ;
3° la spécification minéralogique. Ce dernier caracttre comprend: (a) la nature
des minéraux intégrants; (6) Jeur structure d’association (structure de la roche.)”’

SYSTEMATIC: PETROGRAPAY 459

mineral grains, these authors proceed to give a strangely arti-
ficial weight to the products of the second period, both in
definitions of structure and in classifying rocks by mineral
composition. The structures of eruptive rock applied in classi-
fication are brought under two groups: ‘‘structures granitoides”’
and ‘‘structures trachytoides.”” The essential difference between
the two is conceived to be that in the granitoid the grains of the
two periods of consolidation resemble each other, because of
similar conditions of consolidation in the two periods, while in
the trachytoid structure there isa marked difference between
the two products asa result of changed conditions in the later
period. Inthe rocks commonly called granular it is thought
that two generations of mineral grains of approximately the
same formal character may usually be recognized. Where no
distinction can be made it is rather paradoxically assumed that
the grains all belong to the second period. But it is to be noted
that without this assumption the scheme of Fouqué and Michel-
Lévy, as it stands, could not classify such a-rock. Porphyries
in which the groundmass is granular are, from that fact, classed
with perfectly granular rocks.

Under the granitoid group of structures three varieties are
recognized: (1) Granitoid proper, in which each individual
grain has approximately equal dimensions in all directions, vari-
ation in size being disregarded; (2) pegmatoid, the regular or
graphic intergrowth of two minerals of.simultaneous crystalliza-
tion; (3) ophitic, characterized by elongated feldspar crystals,
and forming a transition to the microlitic structure.

The trachytoid group of structures has likewise three varieties:
(1) Zype pétrosiliceux, characterized by bands of minute spheru-
lites and the presence of the mysterious substance petrosilex or
microfelsite; (2) type microlithique, characterized by microlites
of feldspar, and of other minerals; (3) type vitreux, character-
ized by predominance of amorphous substance.

In explaining the microlitic type the authors point out that
their synthetic experiments prove that such microlites are prod-
ucts of pure igneous fusion, indicating to them a fundamental

460 WHITMAN CROSS

difference between the trachytoid and the granitoid structures,
since they believe that certain mineralizing agents (‘‘agens
minéralisateurs’’) are necessary to the latter development.

This view of rock structures makes the shape of the min-
eral grains all important and casts aside the formal relation-
ship prominent in the porphyritic structure as of little value.
The equidimensional graim and the elongated mzcrolite are placed
in fundamental opposition to each other.

The mineralogical composition of rocks is applied for their
classification in a qualitative way, similar in some respects to
that adopted by the German petrographers, but with the all
important modification that only the minerals of the second period
of consolidation are considered. Such a principle may be desig-
nated as subjective, extremely unnatural and highly artificial.
There is in this system no attempt to express the chemical com-
position of the rock in terms of its minerals, for in some cases
all the minerals of a rock are used in its classification, where
there was no frst period of consolidation, and in other cases
only a small portion of the constituents, as in porphyries with
abundant phenocrysts and microlitic groundmasses. Since in
porphyries this portion of second consolidation bears no definite
quantitative relation to the mass as a whole it must often happen
by this system that rocks of widely different chemical composi-
tion will be brought together and, conversely, that rocks of the
same chemical character and even of the same magma will at
times be separated. For example, certain intrusive quartzose
hornblendic diorite-porphyries of the Rocky Mountain region,
in which hornblende and plagioclase are developed entirely in
phenocrysts, would fall among the microgranulites while their
granular equivalents would be quartz-diorites. It is also clear
that all but granular rocks would be classified in this scheme
by their most obsure constituents, often to the neglect of every
prominent megascopic character, and systematic petrography
would become purely a microscopical science. It is interesting
to recall at this point the principle announced by these authors
and quoted above that a rational classification of rocks must be

SYSTEMATIC PETROGRAPHY 461

based upon the “‘ fundamental characters’ whose application has
been reviewed,

In practice, Fouqué and Michel-Lévy give the first importance
to the colorless constituents, quartz, feldspars, feldspathoids, etc.,
to produce series within which the ferromagnesian minerals are
used to make subdivisions. The existence of certain phenocrysts
is recognized, in the naming of rocks, in a few cases only.

The petrographic system for eruptive rocks elaborated by
Fouqué and Michel-Lévy in 1879 has remained the system of
France to the present time, with but slight change.

In 1889 Michel-Lévy compared the results of their system
with those of the Rosenbusch system, soon to be discussed, in a
work entitled Structures et classification des roches éruptives.* This

tParis, 1889, pp. 93.
discussion presented no new propositions of note excepting a
plan of expressing the structure and mineral composition of any
given rock bya formula. The principles which must govern
the classification of eruptive rocks are concisely stated as follows:

It is neccessary to base the classification and nomenclature of rocks upon
positive facts, independent of all hypothesis. Modern petrography possesses
the means to accomplish this end, since the principal structures or modes of
association of the minerals are well known and the minerals themselves may
be determined with precision. It is, then, exclusively structure and mineral
composition which must be relied upon in the classification and nomenclature
of rocks.*

It is to be remarked, once more, that chemical composition
is not taken into account by Michel-Lévy, either directly or
indirectly, since the partial mineral composition used by him in
classification is clearly not an expression of the chemical com-
position of the magma nor of any definite part of it. The sub-
stance classified is not the rock but merely that variable portion

t™“Notre conclusion,. . . . est qu’il faut baserla classification et la nomenclature
des roches sur des faits indépendants de toute hypothése, et de nature positive.
La pétrographie moderne dispose de moyens suffisants pour atteindre ce but sans hési-
tation: on est d’accord sur les principales structures d’association des minéraux des
roches; on sait déterminer ces minéraux avec précision. C’est donc exclustvement

sur la structure d’association et sur la composition minéralogique que nous persisterons
a nous appuyer pour classer et nommer une roche.” Loc. cz¢., p. 87.

462 WHITMAN CROSS

of the rock which the investigator judges was in a fluid or pasty
condition at the beginning of the second period of consolidation
—the ‘‘ pate.”

Samuel Allport.—During this period in which Zirkel, Rosen-
busch, Fouqué and Michel-Lévy were formulating more or less
distinct advances in systematic petrography, the English students
of rocks made but slight positive contributions in this direction.
The condition of the science may be best appreciated by reference
to the various short discussions of principles of classification by
Samuel Allport. This careful investigator often pointed out the
fallacy of the age distinction, so clearly illustrated by the long-
known ancient lavas of the British Isles, and also the importance
of Judd’s discovery of the intimate relationship of coarsely crys-
talline and volcanic rocks. This was cited to disprove the idea
that a sharp line could be drawn between the Plutonic and Vol-
canic rocks. But Allport considered it premature to suggest
any great changes either in classification or nomenclature.

Clarence King, 1878—In America no original contributions
to systematic petrography were made prior to the microscopical
period. The earliest use of the knowledge gained in that period
was probably by Clarence King,” whose appreciation of its value
led to the report upon the Microscopical Petrography of the 4oth
Parallel rocks by Zirkel, and who, also, applied certain sup-
posed facts resulting from microscopical research in his own
discussion of the classification of volcanic rocks. The propo-
sition referred to has had little influence upon petrographic
system, but has a certain importance from the standpoint of this
review as illustrating again the dangers of applying genetic ideas
in the classification of igneous rocks.

King accepted the law of Bunsen and the law of succession
of volcanic rocks advocated by von Richthofen, which have been
stated. He also considered that ‘‘a sharp line is to be drawn
between the so-called Plutonic rocks and the true igneous ones.”
The microscopical studies of which he had knowledge led him

*Report of the Geological Exploration of the Fortieth Parallel, I. Systematic
Geology, pp. 705-25, Washington, 1878.

SYSTEMATIC PETROGRAPHY 463

to the strongly stated conclusion that ‘all the volcanic rocks
show abundant evidence of fusion in the presence of glass base
and glass inclusions, while the group which is typified by granite
never shows the slightest trace of the effects of fusion.” All
known characters of plutonic rocks are interpreted as proving
them to be extreme products of the metamorphism of sediments.
After discussion of the cause of generic differences of volcanic
rocks, in which certain new hypotheses concerning magmatic
differentiation are developed, King proposes the following sys-
tematic arrangement for the family of volcanic rocks, including
therein all believed to be of truly igneous origin:

Genera.—(t) Propylite; (2) Andesite; (3) Trachyte; (4) Neolite.
Expressions of time, according to von Richthofen’s law of succession, and of
depth owing to secular refrigeration.

Species.— Expressions of chemical differentiation by specific gravity of
mineral ingredients, grouping under the law of Bunsen.

Three species only were recognized under each genus, repre-
senting respectively the quartz, biotite or hornblende, and
pyroxene-bearing forms.

Varieties Expressions of range of texture according to predominance
of secreted crystals, groundmass, or base.

M. E. Wadsworth, 1884.— Under the title Lzthological Studies ;
A Description and Classification of the Rocks of the Cordilleras, M.
Ik. Wadsworth published, in 1884, the first part of a projected
comprehensive work, intended to present a new classification of
rocks. This first part was devoted toa discussion of principles
and the beginning of the descriptive portion. Wadsworth
denounced all existing systems as highly artificial, and stated, as
the basis of his own more natural system, the belief that ‘the
older rocks now classed as distinct species are rocks that were
once identical with their younger prototypes.”’ ‘The order [of
his system] will be to pass from the glassy to the most perfectly
crystalline state; from the least altered to the most altered ;
from the most basic toward the most acidic; from the non-
fragmental to the fragmental or clastic.” He plunged at once
into a description of ultrabasic rocks, without explaining the

464 WHITMAN CROSS

proposed application of his asserted principle to the construc-
tion of a system, and no further portions of the projected work
have appeared. It is plain that the basis of Wadsworth’s
conception is contrary to known facts of petrogenesis. A
direct contribution to petrographic system is afforded by the
proposition made by Wadsworth to group terrestrial and mete-
oric masses together, applying to them a single system and
nomenclature. The descriptive portion of the published work is,
in fact, mainly occupied with discussion of meteorites rich in
iron. Wadsworth thus goes a step further than von Lasaulx,
who treated meteorites in an appendix to his discussion of ter-
restrial rocks (p. 455).

It is hoped that the trend of the evolution of systematic
petrography during the earlier portion of the microscopical era
has been fairly indicated in the preceding pages. Thetendency,
most natural under the circumstances, was to overestimate the
systematic importance of some of the discoveries made through
the microscope, and to slight other, more fundamental, prop-
erties or relations of rocks. Many mew rock names were
proposed and usage became fixed and extended in directions
where it had never been well grounded. Protests against the
tendency of the times were numerous, and especially from the
geologist’s standpoint. Many of these protests were of little
influence, because based upon imperfect appreciation of the situa-
tion; others were far too conservative in spirit.

J. D. Dana, 1878.— As an example of the conservative
geologist’s view at this time may be cited the discussion of
petrographic system by J. D. Dana in an article published in the
American Journal of Science in 1878 under the title ‘‘On some
Points in Lithology.”* This article refers to the Mikroskopische
Phystographie of Rosenbusch, and to other recent works, and
may be taken as expressing the author’s view of petrography at
the stage of its development just reviewed.

Lithology, according to Dana, is charged with the descrip-

*Amer. Jour. Sci., 3d ser., Vol. XVI, p. 335, 1878.

SYSTEMATIC PETROGRAPHY 465

tion and naming of rocks. It has “to note down their distinc-
tions in such a manner as shall best contribute to the objects of
geology.” ‘From granite down they [rocks] are with very
few exceptions mixtures of minerals, as much so as the mud of
a mud bank.” ‘Strongly drawn lines exist nowhere.” ‘‘ Rocks
are therefore of different Azmds, not of different species; and only
those mixtures are to be regarded as distinct kinds of rocks which
have a sufficiently wide distribution to make a distinct name
important to the geologist.”

Dana discusses the bases of classification adopted by petrog-
raphers of this time, such as age, structure, and contents in
certain minerals, and objects to most of them as trivial or
wrongly used. He then proposes an arrangement of the ‘“‘ Crys-
talline rocks, exclusive of the calcareous and quartzose kinds,”
under the following groups :

I. Mica and potash feldspar series — Granite, gneiss, mica schist, tra-
chyte, etc.

II. Mica and soda-lime feldspar series— Kersantite, kinzigite, ditroite,
phonolite, etc.

III. Hornblende and potash feldspar series — Syenite, hornblende-schist,
foyaite, etc.

IV. Hornblende and soda-lime feldspar series — Diorite, andesite, eupho-
tide, etc.

V. Pyroxene and potash feldspar series — Amphigenite.

VI. Pyroxene and soda-lime feldspar series— Augite-andesite, norite,
dolerite, etc.

VII. Pyroxene, garnet, epidote, or chrysolite rocks, containing little or no
feldspar, lherzolite, dunite, garnetite, etc.

VIII. Hydromagnesian and aluminous rocks. Chloritic, talcose, and
other schists, serpentine, etc.

Wiles Danaydoes notirefer in’ this article touthe broader,
grouping of rocks, it appears, from various editions of his Manual
of Geology, that he uses mode of origin to distinguish the three
great classes—AIgneous, Sedimentary, and Metamorphic — in
discussing that question; but, in arranging rocks for description,
he abandons that principle, and makes another division as more
convenient. Convenience of presentation, and not expression of
natural relations, is really the object of Dana’s arrangement. It

460 WHITMAN CROSS

is, therefore, not strictly a petrographic system. In the second
edition of the Manual (1874), igneous and metamorphic groups
were separated in description; but, probably for the reason that
such a course involved the splitting up of granitic, syenitic, and
other types supposed to be metamorphic in part and eruptive in
part, Dana had, in 1878, evidently adopted the arrangement
found even in the last edition of the Manual (1895), whereby
crystalline and fragmental were opposed to each other, as by Zirkel
in 1866.

Dana was never able to adopt the modern petrographic sys-
tems, founded so largely upon erroneous assumptions, like the
age distinction among igneous rocks, or upon genetic views with
which he was not in accord, and contented himself with an
arrangement of convenience. The main features of this order of
description appearing in the Manual of Geology, 1895, are seen
from the four primary groups under which all rocks were treated :

. Limestones, not crystalline.
. Crystalline limestones.

I
2
3. Fragmental rocks, not calcareous.
4. Crystalline rocks.

The crystalline rocks were described under five heads:

I. Siliceous rocks, or those consisting mainly of silica.
II. Rocks having alkali-bearing minerals as chief constituents.

III. Saussurite rocks.

IV. Rocks without feldspar.

V. Hydrous magnesian and aluminous rocks.

The second group was divided nearly as in the proposition of
1878, above cited. This resulted in bringing together such
unlike things as granite, greisen, minette, slate, agalmatolite,
porcelain jasper, obsidian, etc., in the ‘‘ Potash Feldspar and Mica
Seniesi:

Karl A. Lossen.— Among other protests raised by geologists
against the tendency to treat rocks fom the microscopist’s
alleged narrow standpoint, one of the more philosophical discus-
sions had an acknowledged effect making it worthy of notice;
namely, that by Karl A. Lossen, himself a petrographer of dis-

SSTEMATIC PETROGRAPHY 407

tinction, yet in first degree a geologist, prominent upon the
Prussian geological survey.

The views held by Lossen were repeatedly expressed, and
were summed up ina discussion of principles entitled Uber die
Anforderungen der Geologie an die petrographische Systematik. The
“demands” here forcibly expressed are based upon the idea that,
because the rock 1s a geological body, the geological relations of rocks
must be recognized as petrographical relations. Of all the geologi-
cal relations of rocks Lossen selected that which seemed to him
the most important, and claimed that that principle should be
used as the primary factor in petrographic classification. The
relation of rock masses to the earth sphere appealed to Lossen
as all important, and on that criterion all rocks were considered
as stratified or massive. Stratified rocks were defined as those
accumulated upon the earth’s surface under the controlling influ-
ence of gravity, causing them to assume in some degree the
form of concentric strata normal to the radius of the earth.
The material of massive or eruptive rocks, on the other hand,
was viewed as having been forced from the depths directly
against the influence of gravity, consolidating like a casting,
under the control of surrounding conditions. Surface lavas,
spreading out in sheets, while controlled largely by gravity in
their formal relations, were referred by Lossen to massive rocks,
because possessing the resemblance to a casting from one pour-
ing, and not as built up by successive additions, layer upon
layer. It will be seen that the molten condition of magmas was
actually a leading factor in Lossen’s idea, although the avowed
intention in using eruptive as an alternative for massive was to
express the force opposed to gravity, and not the molten state
which rendered them susceptible to that force.

Lavas are not the only rock masses difficult of consistent
treatment under Lossen'’s principle. Pyroclastic tuffs were con-
sidered as illustrating the fact that ‘‘das Ineinandergehen zum
Wesen der Gesteinsnatur gehért.” Lossen confessed, further,
that rocks of the first crust of the earth, according to the nebu-

1 Jahrbuch der k. pr. geol. Landesanstalt, 1883, p. 486.

468 WHITMAN CROSS

lar hypothesis, would necessarily be separated from his eruptive
class, but he considered this point immaterial because he doubted
whether any such rocks were known. Metamorphic rocks are of
such diverse origin and present such difficulties to the system-
atist that Lossen considered it inadvisable to treat them as a
distinct class.

In this discussion by Lossen, as in the majority of those ema-
nating from geologists, no appropriate distinction is made
between the rock mass as a formal unit and the material within
it, the rock proper. The primary division of Lossen is one of
rock masses, not of rocks. The consideration of the form and
position of these masses with regard to the controlling influence
of gravity or the opposing eruptive force, leads to only one of
many ways in which the geologist must classify rock bodies.
Other elemental subdivisions of the same bodies are necessary.
The geologist is, indeed, obliged to make and discuss all such
fundamental distinctions. The petrographer, on the other hand,
should not only be at liberty to select, but, in order to secure
logical excellence for his system, mst choose, as his primary
principle, that one most closely connected with the factors which
he has adopted for use in the further construction of the system-
atic arrangement of rocks. It is clear that all igneous rocks,
those produced by the consolidation of molten magmas, possess,
from this origin, material properties most useful in their detailed
classification. If arrangement by a certain characteristic due to
this origin is desired, it matters not where the rocks occur in the
earth. They may belong to the primeval crust, or form injected
masses of whatever size, shape, or attitude, or appear on the sur-
face in lava streams. If it was mode of origin, not formal rela-
tions to the earth, which gave igneous rocks the common
characters used in their systematic arrangement, then mode of
origin is logically the principle to be used by the petrographer
to bring them into one grand division.

H. Rosenbusch, 1887.—As has been mentioned, the force of
Lossen’s claim, as made in earlier publications, was admitted by
H. Rosenbusch, in the first edition of his Mzkroskopische Physio-

SYSTEMATIC PETROGRAPHY 409

graphie der massigen Gesteine, ‘‘massig”’ being used in the sense
explained by Lossen, and not in that of textural condition. In
the second edition of this widely-used handbook, which appeared
in 1887, Rosenbusch repeated his approval of Lossen’s proposi-
tion, and in the revision of his petrographic system further
emphasized geological occurrence as a factor of prime classifi-
catory value for eruptive rocks. The great changes in system
found in this volume make it practically a new work, but a
detailed statement of its scheme seems unnecessary since all
petrographers may be assumed to be familiar with it.

In the decade since the appearance of the first edition, Rosen-
busch effected a complete change of systematic base in some
important respects, the reasons for which are given in the preface.
He now considers rocks as the documents in which the history
of the earth is written, and petrography, as the science which
teaches us how to decipher those documents, becomes to him @
historical and not a descriptive science. Yo quote his declaration:

The recognition of these relations made the new edition of this book

practically a new work. Its end will have been achieved if I have succeeded
in procuring for this fundamental conception a general acceptance, and have
demonstrated that rock structure affords the safest and most productive means
for the construction of a natural system of rocks.’
A natural system of rocks must therefore be historical, z. ¢.,
genetic. Plainly the logical analysis of the broad science of
rocks presented by Naumann was either forgotten or ignored by
Rosenbusch.

With this statement of controlling principles in mind, one
involuntarily recalls that in the first edition of the Mkroskop-
wsche Physiographie the rocks to be discussed were designated
massive in preference to eruptive because the former term was
considered free from expression of any genetic idea. In this
second edition, while advocating the genetic basis of classifica-

“Die Erkenntniss dieses Verhaltnisses machte die neue Auflage dieses Buches
zu einer Neubearbeitung. Der Zweck derselben wird erreicht sein, wenn es mir
gelungen ist, dieser Grundanschauung eine allgemeinere Anerkennung zu verschaffen

und darzuthun, dass die Gesteinsstructur das sicherste und ausgiebigste Mittel zum
Aufbau eines natiirlichen Systems der Gesteine an die Hand giebt.”

470 WHITMAN CROSS

tion as all important, Rosenbusch retains the group term massive,
yet in both works it is clearly the igneous origin which is first in
mind and which is recognized as of prime importance in pro-
ducing rock structure, chosen as the leading factor in construct-
ing the new system.

After having stated his belief that rock structure is the best
basis of classification of ‘‘massive rocks’’ Rosenbusch proposed
to divide them into three groups: (1) Deep-seated rocks
(‘‘Tiefengesteine”’), (2) Dike rocks (‘‘Ganggesteine”), and (3)
Effusive rocks (‘‘Ergussgesteine”’). The critic is obliged to
point out that this proposition is inconsequent, for not only is
structure not expressed in the terms chosen, but another distinct
factor is expressed, namely, mode of occurrence. The further
development of Rosenbusch’s scheme makes it clear that he did
not intend to emphasize the actual facts of geological occurrence,
plainly as he stated them, but rather to express in this way his
conception of the genesis of structure. Recognizing that dif-
ferent structures result from the consolidation of a given magma
according to the attendant conditions, Rosenbusch selected the
geological factor appearing to him of greatest influence among
many conditions and made that the expressed basis of structural
classification. Since simplicity and logical directness are surely
of utmost importance in systematic constructions the unnecessary
indirectness of this proposition may be designated a fatal weak-
ness. Furthermore, the geologist is warranted in objecting to it
because the expressed division of igneous rocks is one which he
has used in the past and must use in the future, in its literal and
appropriate sense, quite apart from the idea hidden in the terms
of Rosenbusch’s system.

To the above noted criticisms of Rosenbusch’s first applica-
tion of structure in classification must be added another, based
upon the fact that the division of Dike rocks was not in reality
provided for rocks occurring in dikes, but for a group of rocks
for which Rosenbusch assumed a certain genesis. An hypothesis
of magmatic differentiation and assumptions of limited occur-
rence and of characteristic structure are all involved in the dis-

SYSTEMATIC PETROGRAPAY 471

crimination of the group named ‘Dike rocks.” In this light
this group is certainly not co-ordinate with the other two of the
same rank as defined.

In the subdivision of the three classes of ‘ Massive rocks”
Rosenbusch applied mineral composition as a factor, producing
Families. The quantitative composition, either chemical or
mineral, received no expression, so that, for example, anorthosite
and the most highly pyroxenic gabbro or norite are found
together in the gabbro family. Moreover in the porphyritic
Dike rocks only the phenocrysts are considered in determining
the systematic position of a givenrock. Thusa porphyry having
the chemical composition of a granite is referred to syenite-
porphyry in case its excess of silica chances to be confined to
the groundmass, while had quartz phenocrysts been present it
would have been called granite-porphyry. In the Effusive
rocks, Rosenbusch hesitates to apply the same rule consistently.
The families of these rocks are defined in very general terms as
the ‘“‘equivalents”’ of certain granular rocks and described as con-
taining certain phenocrysts in a groundmass of variable appear-
ance.

As in the earlier system, all feldspar-free rocks of the deep-
seated class are united as Peridotites. The peculiar character of
the Dike rocks as a division not co-ordinate or co-extensive in
range with the Deep-seated or Effusive rocks appears in the fact
that mineralogical groups corresponding to the granites, syenites,
and diorites, only, are recognized.

Geological age is acknowledged by Rosenbusch to have been
assigned a higher value in classification than belongs to it, but it
is retained, in the Effusive class, and the use of duplicate terms
perpetuated.

H. Rosenbusch, 1896.—The third edition of the “dkro-
skopische Phystograplie der massigen Gesteine,” issued in 1896,
contains no essentially new systematic features. The principles
above set forth are reaffirmed, and, save for the elaborated dis-
cussions of magmatic differentiation, which show more plainly
than in the preceding edition the strong influence of hypo-

472 WHITMAN CROSS

thetical considerations in giving form to this system, there is
little of note to comment upon in this place. In discussing the
essential characters of the three groups of Dike rocks, Rosen-
busch brings out more forcibly than before the genetic idea
really lying at the base of the distinction of the Dike rock class.
In connection with the discussion of the Dike rocks it is sug-
gested that a further class may be necessary to include the
intrusive rocks of sheets and laccoliths which seem to him to
possess distinctive structures.

That the system of Rosenbusch does not result in a consistent
and logical classification of igneous rocks is abundantly illus-
trated by numerous instances, many of them freely acknowl-
edged by the author. The family of the diabase rocks furnishes
one of the most notable cases. In 1887 these rocks were classi-
fied with the deep-seated rocks, although many of them were
known to be effusive; in 1896 the same rocks are placed with
the effusives, with the statement that many are intrusive.
Placed in the effusive class, they are acknowledged to be partly
of older and partly of younger age, but no age distinction is
thought to be practicable.

Justus Roth, 1883.—Shortly before Rosenbusch issued the
second edition of his Phystographie, there appeared a complete
systematic discussion of rocks by Justus Roth, in the second
volume of his General and Chemical Geology.’ Petrography is
defined by Roth as the science of the mode of origin, constitu-
tion, and alteration of rocks; 7. ¢., the petrology of many
English and American writers. In introducing the systematic
descriptive part of the subject, Roth remarks: ‘“ The difficulty
in constructing a system of rocks is completely expressed in the
term aggregate, and thereby all recourse to genera and species is
prohibited.” ? From the nature of rocks and the conditions of

1 Allgemeine und chemische Geologie; Zweiter Band, “ Petrographie” (Berlin,
1883-1885), pp. X + 695.

2“Die Schwierigkeit der Systematik der Gesteine ist durch die Bezeichnung
Aggregat vollstandig ausgedriickt und damit alle Anlehnung an Gattungen und
Species ausgeschlossen,” /oc. czt., p. 41.

SYSTEMATIC PETROGRAPHY 473

their origin, he thinks that every system must so largely repre-
sent individual opinion that probably no one system can ever
receive universal recognition.

The systematic arrangement of Roth is, in its general out-
line, as follows:

A. Rocks composed essentially of minerals.

I. Plutonic (consolidation products of molten magmas). Free from
fossils, composed of minerals or substance chemically like a mineral
aggregate.

1, Eruptive. (Breaking through other rocks.)
a. Pre-Tertiary.
6. Post-Cretaceous.
Appendices to a and 4 contain rocks produced by weathering.
Tuffs.
2. Crystalline schists.
Appendix- Weathering products.
II. Neptunic.
1. Partly fossiliferous ; composed of minerals and of the products of
the decay, decomposition, and attrition of minerals.
a, Precipitates from solution.
6. Deposits from suspension.
2. Clastic, composed of rock fragments.
B. Rocks composed essentially of organic remains.
‘C, Products of contact metamorphism.

It will be observed that the ancient crystalline schists are
regarded as the primary crust of consolidation of the earth.

The geological factors of origin, relations, or age, are variously
applied in the construction of this scheme, and in constitution
the distinction between mineral and rock particles is made. All
Plutonic rocks are regarded as consisting essentially of silica or
silicates, excepting that in the crystalline schists carbonates appear.
The silica free minerals—apatite, magnetite, ilmenite, etc., are
treated as accessory constituents. This exclusion of the latter
group of minerals from a position of systematic importance is
not discussed by Roth, but its evident result is that in certain
rocks, ¢. g., those rich in magnetite, the components do not have
their natural and logical weight in classification.

474 WHITMAN CROSS

Of the pre-Tertiary eruptive rocks, Roth makes, for con-

venience, three divisions:
I. Orthoclase rocks.

II. Plagioclase rocks.

III. Peridotites.

The first two of these groups should, logically, have been
united systematically in the division of Feldspathic rocks,
including all with appreciable content in feldspar, since the
Peridotites are defined as free, or nearly free, from feldspar.
The question of recognizing the quantitative relations of mineral
constituents is not mentioned by Roth.

The silicate minerals are applied by Roth for the subdivision
of the three main groups in the usual way, and by means of
structure the granular, porphyritic, and glassy varieties are
distinguished.

In the detailed treatment of Eruptive rocks, as in the arrange-
ment of Crystalline schists, the Neptunic rocks and the Classes
B and C, mentioned above, Roth’s order of presentation and
discussion can hardly be said to be systematic. It is an arrange-
ment for convenience of description, not based upon the logical
application of principles; and it is, therefore, not desirable to
devote more space to its analysis in this review.

E.. Kalkowsky, 1856.— A condensed text-book on rocks was
published by E. Kalkowsky, in 1886, with the title Alemente der
Lithologie.* For the primary division of rocks the author formu-
lates an original criterion, and proposes two great classes:

I. ‘“Anogene’’—of which the material came to the place of rock forma-
tion from below.
II. ‘* Katogene’’— of which the material was derived from adove.

These correspond closely to the eruptive and sedimentary
divisions of other authors.

For the classification of the ‘“Anogene” rocks Kalkowsky
applies the following factors: (1) Chemical composition as
represented in mineral composition; (2) the usual age distinc-
tion; (3) structure. He rejects genetic distinctions as unsuit-

* Heidelberg, 1886, pp. 316.

SSL EMATTC PETROGRA PHY 475

able. In detail Kalkowsky’s scheme is similar in its results to
that of Zirkel, but, as he does not define his smaller rock divi-
sions, a further discussion of his arrangement seems unnecessary.
The definitions are omitted, according to the author, because the
student must learn to know the rocks by the study of named
hand specimens and will, therefore, find out what they are with-
out definitions.

J.J. Harris Teall, 1886, r888.—The most extensive treatise
on rocks thus far published in England is the descriptive work
British Petrography, by J. J. Harris Teall, issued almost simul-
taneously with the second edition otf Rosenbusch’s Massige
Gesteine. This work lays no claim to being a systematic petrog-
raphy, and describes almost exclusively the igneous class; but
from its scope a discussion of principles of rock classification
was necessary, as explanatory of the arrangement actually used.
Teall considers rocks so complex and indefinite in character
that in the existing state of knowledge no true systematic
arrangement is possible. His order of presentation is, in fact,
one of convenience, and does not express his own views of the
most natural basis of classification.

In discussion of principles, Teall points out that chemical
composition, as the constant and primary character of igneous
rocks, is the natural basis of classification and in accordance
with the Bunsen law of two magmas. He, however, does not
work out any new proposition to use chemical composition.
The arrangement under which rocks are described is a mixture
of the methods of Rosenbusch and Michel-Lévy. All igneous
rocks are placed in seven groups, as follows:

A. Rocks composed of the ferro-magnesian minerals: olivine, enstatite,
augite, hornblende, and biotite. Feldspar absent; or, if present, occurring
only as an accessory constituent.

B. Rocks in which plagioclase is the dominating feldspathic constituent.
Nepheline and leucite absent. Orthoclase is frequently present.

C. Rocks in which orthoclase is abundant. Plagioclase usually present.
Nepheline and leucite absent.

D. Rocks containing nepheline or leucite; sometimes nepheline and
leucite. |

476 WHITMAN CROSS

£. Rocks not included in any of the preceding groups.

F, Vitreous rocks.

G. Fragmental volcanic rocks.

It will be seen that this grouping is mainly mineralogical and
does not express the quantitative element in any logical way.
It practically recognizes the entrance of feldspar, in any amount
above that of the undefined ‘‘accessory” réle, as creating a
large group of feldspathic rocks. The subdivision of these groups
is first on some further mineral distinction and after that occur-
rence and texture enter combined into the system by distin-
guishing rocks of granitic from those of trachytic texture, using
these terms in the sense of Fouqué and Michel-Lévy, and con-
ceiving that the result is practically to separate plutonic or deep-
seated from volcanic or effusive rocks. Age is not introduced as
a factor.

Within the last twenty years several attempts have been
made to apply, zz extenso, the chemical composition of rocks for
their classification. These attempts have been prompted by
various motives. Some appear to have no really practical
object, viewed from the petrographer’s standpoint; others are
connected with hypotheses of magmatic differentiation; and still
others have been inspired by a realization of the complexity of
the problem of a rational arrangement of rocks on the basis of
their numerous and highly variable mineral constituents. It
appears to the writer, however, that it may be fairly said of all
these attempts that they are either not classifications cf rocks, or
that they are not actually chemzcad classifications.

Franz Schrockenstein, 1856, 1597.— Two peculiar attempts by
Franz Schrockenstein, an Austrian writer, to discuss the chemi-
cal composition of silicate rocks, irrespective of their origin, and
upon that basis to classify them, are mentioned here only for
the sake of completeness, as these attempts have no direct bear-
ing upon a logical system of rocks. The writer has seen only
the more recent of the essays in question,* depending for the

tr Silicat-Gesteine und Meteorite,” Petrographisch-chemtsche Studie auf Grund-
lage des neuesten Standes der Wissenschaft bearbettet. Prag, 1897.

SS LAVATT EC PET ROGRAPT Y 477

earlier one* upon the summary given by F. Loewinson-Lessing.’

Schréckenstein’s view of igneous rocks will, on account of
its fantastic imaginings, impress many a reader as belonging
rather to the eighteenth, or a still earlier, century, than to the
close of the nineteenth ; and, although his propositions are of
no consequence to petrography, the fact that they have been put
forth at all, in the very last decade of the period in review, has
a certain melancholy interest.

Schréckenstein considers the original crust of the earth to
have been a szlicate of alumina, probably with excess of silica.
This simple primary magma is conceived to have been first ren-
dered impure by meteoric showers, introducing lime, magnesia,
and iron. Ata later period the alkalies and water were precipi-
tated from the atmosphere. The alkalies are considered as of
very subordinate (‘‘ nebensachlich’’) importance and the chemical
problem of rocks, as the author views it, is to compare the
relative amounts of the original alumina silicate and the meteoric
impurities. That is to say, te author proposes classes according
to the degree of adulteration of the original magma and orders
according to the character of the adulterant.

The method followed by Schréckenstein in comparing analyses
of silicate rocks appears to be somewhat as follows:

First, magnetite is calculated out, as an extraneous substance,
whenever the analysis is sufficiently modern, through determi-
nations of both ferricand ferrous oxides, to give a basis for such
calculation. When the analysis is inadequate and the iron is
lumped under one or the other oxide, the result is accepted by
the author and Fe,QO, is supposed to replace alumina or FeO is
added to MgO and CaO. Not until magnetite is deducted does
Schréckenstein consider that the real rock is under discussion.
Inasmuch as he states that after deducting magnetite there is
either no iron left or but one oxide appears, it is evident that the

* Ausfltige auf das Feld der Geologie,” Geologisch-chemische Studie der Silicat-
gesteine, II Auflage, Wien, 1886.

2“ Studien ueber die Eruptivgesteine,” Compte-Rendu, VII Cong. Géol. Internat,
1899, p. 196.

478 WHITMAN CROSS

maximum possible amount of magnetite is -deducted. The
remainder is then calculated to 100. The analyses are not given
in their original form.

The systematic plan of Schréckenstein consists, in his later
publication, in establishing five classes of silicate rocks, according
to the relations of Al,O, to RO (=CaO + MgO + FeO) as
shown by the percentages of the calculated remainder after
deducting magnetite.

ROW sa RO ROM 3
TSO gee mo: ra TAO! 4 ee

RO Li RO I

ue Al, O, oe - Al, On i

Two orders appear under each class according as lime or
magnesia dominates.

Although Schréckenstein professes to use the latest infor-
mation, as stated in the title of his recent publication, his results
are based upon the discussion of 340 analyses, many of them old,
while on the other hand no single one of the hundreds of analyses
made in the laboratory of the United States Geological Survey,
within the past twenty years, is utilized. Hundreds of European
analyses of recent date are also ignored.

It seems unnecessary to give any further details concerning
Schréckenstein’s propositions. He is not actually treating rocks
and his superficial considerations of chemical composition can
have no bearing upon true petrographic system.

fl. O. Lang, 189z.— An attempted arrangement of igneous
rocks on a chemical basis by H. O. Lang,’ in 1891, is founded
on the idea that since the feldspars are the most important con-
stituents of eruptive rocks, an appropriate and practical chemical
basis of classification may be found in the relations of the bases
potash, soda, and lime, the distinctive elements of the various
species of feldspar. In one case Lang used the percentage

*“ Versuch einer Ordnung der Eruptivgesteine nach ihrem chemischen Bestande,”’
Tscher. Min. Pet. Mitth., X11, 1891, p. 199.

“Das Mengenverhaltniss von Calcium, Natrium, und Kalium als Vergleichungs-
punkt und Ordnungsmittel der Eruptivgesteine,” Bull. Soc. Belge de Géol., 1891, V,

p. 123.

SYSTEMATIC PETROGRAPHY A79

amounts of the oxides found by analysis, and in the other the
amounts of the elements potassium, sodium, and calcium. Here
again is the situation that only a part of some rocks is actually
under discussion, and the result can be of no real value to
petrography.

fF. Loewtnson-Lessing, 1890, 1897. In 1890 an attempt at a
chemical classification of igneous rocks was made by F. Loew-
inson-Lessing,* based upon the quantitative relations of silica to
the various oxides of the bases, grouped under R,O, RO, and
Rae Oi;
empyrical formule the author thought to finda way of expressing

as shown in percentages by bulk analysis. By means of

regular relationships supposed to exist between the silica con-
tents and the various oxide groups. Rocks exhibiting the
following simple relationships were designated z¢ypfes, I to V being
the principal ones, and VI to IX intermediate :
I. SiO, = 2(R,0 + RO)+R,0,4+0.
Acid |
Rocks | ae Si0,=2(R.0 MRO ROO

II. SiO, = 2(R,O + RO) +R,0, (11 - ES)

Neutral
Rocks VII. si0,= (R01 RO) RO...
i ay,
( III. Si0,=R,0 + RO +R,0, (10 = i)
Basic 2
Rocks Te lV
| Will) SiO; —= ROl-E RO -E : R.O, (vit =e >
Olen [Wess @O>— or, on ;
Basic LX SiO:—— RO.
Rocks 2

VS Ol -— Or

Since percentages instead of molecular ratios were used, the
simple relations here adopted have no real significance as
expressing a regular connection between chemical and mineral
composition. This fault was perceived by the author and cor-

™Ftude sur la composition chimique des roches éruptives,” Bzd/. Soc. Belge de
Géol., 1890, IV, Mem., p. 221.

480 WHITMAN CROSS

rected in the publication to be discussed below. These so-called
types were assumed by Loewinson-Lessing to correspond more
or less closely to certain commonly known rock groups.

The more elaborate discussion of the chemical relationships
of igneous rocks by Loewinson-Lessing, presented to the Inter-
national Geological Congress at St. Petersburg, in 1897, and
published two years later in the Compte-Rendu, deserves some-
what fuller consideration.‘ A brief statement of the author’s
point of view is desirable before explaining the system proposed.

In reviewing the applicability of various factors in producing
a rational system, Loewinson-Lessing asserts that the mineral
composition of a rock is a function of its chemical composition.
The exceptions to this rule admitted by him are of little impor-
tance. Then follows the further statement that the principle or
characteristic of mineral composition as a basis of classification
is faulty and unsatisfactory because it does not show the relative
abundance of the minerals in the various rocks. That is, how-
ever, as it appears to the writer, not the fault of the principle,
but of the manner in which it has been applied in existing sys-
tems. If mineral composition is a function of the chemical
composition, it is just as capable of expressing the constitutional
relations of rocks as the latter, if properly used. The real
objection to its application in the quantitative way, necessary to
this expression, is simply one of practicability. The problem is
too complex.

As for his own system, Loewinson-Lessing starts from the
idea that eruptive rocks may be considered as selicate rocks and
classified as such. Whatever the facts as to predominance of
silicates in these rocks may be, it seems to the writer that this
conception is not complete as to its basis of fact, and is thus
inadequate to serve as a means of classification. Further funda-
mental propositions enunciated by the author are that: (1) sili-
cate rocks should be classified by the same artificial means as
the silicates themselves; (2) while rocks are not stoichiometric

*F. LOEWINSON-LESSING, “Studien iiber die Eruptivgesteine,” Compte-Rendu de
la VII session du Congrés Géologigue International, Russie, 1897, pp. 193-467.

SYSTEMATIC PETROGRAPHY 481

compounds, they are not accidental mixtures ; (3) one must
consider the relative amounts of a@// oxides of bases to each
other and to silica; (4) as silica is the dominant constituent it
is proper to take it as the basis for the primary classification ;
(5) the next factor to be applied must be the contents in the
three oxide groups, alkalies, alkaline earths, and the sesquioxides ;
(6) various single oxides may be used for further subdivisions.
From this statement it might be supposed that a classification
created by the successive use of the chemical factors named was
to be set up by Loewinson-Lessing, and such an arrangement of
magmas would have claim to being a chemical classification.
But the author does not do that, as we shall see.

The actual system proposed by Loewinson-Lessing is to use
the silica contents for the formation of four general groups:
(1) Acid rocks; (2) Neutral rocks; (3) Basic rocks; (4) Ultra-
basic rocks.

The second division is obtained by taking a certain number
of analyses representing known rock families (established on
the unsatisfactory basis of mineral composition) and determin-
ing the mean of these analyses, which is then set up as the com-
position of a rock ¢yfe, and its formula and coefficient of acidity
are ascertained.

It is clear that the grist of this mill depends entirely upon
what is put into the hopper. It is not a chemical classification
but a chemical characterization of mineralogical rock groups
arbitrarily selected by theauthor. It will, of course, be possible
to secure means corresponding to any formula desired as a type,
and the rocks thus having typical position could be adopted as
centerpoints of groups or families. For the ordinary range of
rocks these types would often coincide with recognized rocks,
assigned certain names in existing systems, and these names
might then be given by redefinition to the families thus indi-
cated. But what would be the purpose of such ascheme? It
could not express the existing relations between the mineral
composition of the rock and the chemical constitution of the
magma.

482 WHITMAN CROSS

A. Osann, 1900, rg0r.—A further attempt to utilize chemical
composition as a factor in the classification of igneous rocks was
made by A. Osann in the closing year of the century. Under
the title ‘“Versuch einer chemischen classification der Eruptiv-
gesteine ”’* Osann essays to use chemical composition as a supple-
ment to mineral and textural characters, by establishing various
chemical types within rock families formed upon the Rosen-
busch system. The author accepts the classes of deep-seated,
dike and effusive rocks, in the Rosenbusch sense, and the
vaguely defined families established upon mineral composition.
Realizing that by this latter factor, as currently applied, the
relative abundance of the minerals is not sufficiently taken into
account, Osann attempts to bring out quantitative relations,
within the families, by establishing certain types upon the basis
of chemical composition.’

It is clear that if the quantitative element was not sufficiently
expressed in forming the families discussed by Osann, he fails
to remedy the defect. A logical subdivision of the families on
a chemical basis would principally serve to point out the defects
in this respect, and would really weaken rather than strengthen
the system as a whole.
set up by Osann are in no sense systematic

”

But the ‘‘types
divisions of the families. The “type” of this author is simply
a chosen well-analyzed rock, differing in chemical composition
from other rocks within its family, according to the adopted
method of comparison. To a type thus established are referred
other rocks of nearly identical chemical characters. But the
types bear no definite relation to each other or to the family.

t Tschermak’s Min. und petr. Mittheilungen, Bd. XIX, 1900, pp. 351-469, and Bd.
XX, I9OI, pp. 399-558.

2 The author’s standpoint may be sufficiently understood from the following scate-
ment: “Der Hauptmissstand der mineralogisch-structurellen Classification liegt
darin, dass dem relativen Mengenverhaltniss der wesentlichen Gemengtheile zu wenig
Bedeutung zuerkannt wird, und es wird gerade die Hauptaufgabe der chemischen
sein, in dieser Richtung erganzend und vertiefend zu wirken.” ‘‘So kann es sich
bei dem hier unternommenen Versuch ebenfalls nur um ein kiinstliches System han-

deln, welches in erster Linie dazu bestimmt ist, das mineralogisch-structurelle zu
erganzen.” Jbzd, Bd. XIX, pp. 351-352.

SYSTEMATIC PETROGRAPHY 483

They merely serve to show the chemical range found within the
families so far as Osann’s examination extends.

From the above statement it would appear that Osann has
not, in reality, proposed a chemical classification in the sys-
tematic sense, and hence it is not desirable to enter further into
the analysis of this elaborate discussion of the chemical varieties
represented within the families of the Rosenbusch system.

This discussion of petrographic systems proposed during the
nineteenth century will close with a review of three attempts to
bring all known rocks into orderly arrangement. Not that the
authors think to have formulated natural or logical systems, for
that is expressly disclaimed by them. Yet in presenting these
comprehensive arrangements of rocks, according to the light of
the last decade of the century, these authors define and illus-
trate in a most effective way the present condition of systematic
petrography.

Ferdinand Zivkel, 1593, 1894.— The second edition of Zirkel’s
Lehrbuch der Petrograplie is the most comprehensive and com-
plete description of all known rocks ever published, and it there-
fore represents the present status of the systematic science as a
whole, better than any other work, and hence deserves careful
consideration. But the fact staring the student in the face is
that systematic petrography is still very largely an arrangement
for convenience of description, and is not, in its entirety,a logical
expression of relationships. Within the division of igneous
rocks there is at least some attempt at system, but the other
rocks confessedly defy logical treatment by any method as yet
proposed.

The primary division of all rocks is on general geological
grounds into four groups:

I, Igneous rocks —‘‘Massige, eruptive Erstarrungsgesteine.”’

II. Crystalline schists.

III. Sedimentary crystalline rocks (not clastic.)
IV. Clastic rocks.

This division is clearly based on geological considerations,

484 WHITMAN CROSS

and is chosen in place of the primary arrangement of the first
edition because, as Zirkel points out, of inconsistencies and unnat-
ural associations which resulted, some of which have been men-
tioned in this review. Contact metamorphic rocks are treated
in connection with the igneous rocks which produced them.
Fragmental igneous rocks are placed with the clastics.

In the systematic classification of igneous rocks Zirkel uses
the bases of arrangement in the following order: (1) mineral com-
position; (2) structure; (3) age. The availability of chemical
composition, alone, or in expressed combination with mineral
constitution, is nct discussed. The method of applying mineral
composition for the classification of igneous rocks is that com-
monly used. Concerning this Zirkel remarks:

In a mineralogical arrangement of massive rocks the following considera-
tions are at present determinative: In the great majority of these rocks
feldspars and other silicates resembling feldspars (such as nephelite, leucite,
melilite) play the chief réle, and therefore it is most natural to base the clas-
sification of such rocks upon the nature of these minerals, in accordance with
existing nomenclature. *

This procedure results in placing feldspar- or feldspathord-
bearing rocks in one large group opposed to feldspar-free rocks.
Whatever the facts may be as to the relative quantitative impor-
tance of different minerals in igneous rocks, it is clearly arbitrary
to concede to any mineral the “principal réle” where it is far
subordinate to others. The result is a qualitative expression of
mineral composition, bringing chemically unlike rocks together
in many divisions.

In the descriptive portion of the Lehrbuch, igneous rocks are
grouped under seven heads :

I. Rocks with alkali feldspar and quartz or excess of silica.
II. Rocks with alkali feldspar, without quartz or excess of silica, without.
nephelite or leucite.

t¥Fiir die mineralogische Gruppirung der Massengesteine sind zur Zeit folgende-
Erwagungen maassgebend: In der weitaus allergrossten Mehrzahl derselben spielen
Feldspathe und andere feldspathahnliche Silicate (wie Nephelin, Leucit, Melilith):
eine Hauptrolle und so scheint es am natiirlichsten, die Classification der hierher
gehdrigen Gesteine auf die Natur dieser Mineralien zu begriinden, was zugleich der
bestehenden Nomenclatur entspricht. — 7etrographie, Band I, p. 832.

SYSTEMATIC PETROGRAPHY 485

III. Rocks with alkali feldspar, without quartz or excess of silica, with
nephelite (haiiynite) or leucite.

IV. Rocks with lime-soda feldspar, without nephelite or leucite.

V. Rocks with lime-soda feldspar and nephelite or leucite.

VI. Rocks without true feldspars, but with nephelite, leucite, or melilite.

VII. Rocks without either feldspars or feldspathoid.

Structure and geological age are applied by Zirkel under each
of the mineralogical groups, as follows :

Granular rocks.

(No distinctions by age.)

Porphyritic and glassy rocks.

Pre-Tertiary,
Tertiary and recent.

The structural distinction is clearly in fact between (1)
granular and (2) non-granular, the range in structure within the
second division being by no means covered by the two terms
porphyritic and glassy.

The use of age as a factor in classification of ‘ porphyritic
and glassy rocks”’ while it is not applied to the granular rocks is
apparently more a recognition of the usage of the time, by which
a duplicate set of terms has been provided for effusive rocks,
than of any definite principle. The task of reconstructing the
nomenclature of the science is still one from which the systematic
petrographer shrinks.

The group of the crystalline schists established by Zirkel is
not founded upon definitely stated principles, and is therefore
not a systematic group. It is defined by enumeration of things
belonging in it or excluded from it, and must be treated as a
convenient expedient for purposes of rock description. But
although this is true the crystalline schist group of Zirkel is no
more unsystematic than the assemblages of other petrographers
given the same name. It is then germane to the present discus-
sion to state the actual course adopted in the Lehrbuch.

Zirkel includes in his group of the crystalline schists, and as
its most important element, the pre-sedimentary gneisses, schists,
etc., which cannot be inferred from their attitude to other rocks to
be of igneous origin. Included with these are all rocks of the

486 WHITMAN CROSS

same texture and composition demonstrably derived from sedi-
ments or occurring intercalated in the sedimentary series but not
clearly of igneous origin. Excluded from the group under dis-
cussion are the primarily banded igneous rocks and the metamor-
phic derivatives of igneous rocks whenever that origin can be
established, and whatever the process of change may have been.
In other words the group includes the rocks below the oldest
known sediments so far as they are not visibly eruptive or
igneous and all later rocks of the same characters derived from
sediments or of unknown origin.

This group then has nothing in texture or composition to distin-
guish it. Neither of the elements of the name has any restrictive
significance. The group is geologically homogeneous only in
case the schists of unknown origin are actually derived from
sediments. If, as many suppose, a large proportion of the
Archean gneisses, etc., represent igneous masses, metamor-
phosed or not, the group is not only heterogeneous from the
genetic standpoint but causes the separation of identical things.

In the subdivision of the crystalline schists mineral composi-
tion is applied, the predominant constituent causing the reference
of a rock to a certain group. The terms gneiss and schist are
not defined.

The group of crystalline, or non-clastic sedimentary rocks, is
heterogeneous in constitution as is apparent from a partial list of
the rocks referred to it: ice, cryolite, limestone, opal, quartzite,
porphyroid, iron ores, coals, diatomaceous earth. That such a
group lacks the unity required in a systematic division, and that
its descriptive name by no means covers the case, is apparent at
once. It is confessedly a grouping for convenience only, and
embraces, in fact, the remaining rocks after the other three have
been established.

H. Rosenbusch, 1898.— A comprehensive discussion of rocks
was issued in 1898, by H. Rosenbusch, entitled, Dee Elemente der
Gesteinslehre Although much less detailed than the Lehrbuch of
Zirkel, this work is of much interest as expressing the views of

t Stuttgart, 1898, pp. 546 -+ 4.

SSAA TIC PETROGRA PHY, 487

one of the great German masters, almost at the close of the
century.

Rosenbusch’s primary division of rocks is into four great
classes :

I. Eruptive rocks (‘‘ Eruptivgesteine”’).
II. Stratified rocks (‘ Die schichtigen Gesteine ’’).

III. Crystalline schists.

III. Primary crust of the earth (‘‘ Erste Erstarrungskruste’’).

Concerning the first class, it is to be noted that Rosenbusch
drops the term ‘‘massig,” used for twenty-five years, as less
appropriate—‘‘ weniger passend villeicht’’—than Eruptiive. In
the further treatment of eruptive rocks Rosenbusch does not
depart from the principles and methods of the last edition of the
“ Physiographie der massigen Gesteine,’ and there is, therefore, no
occasion to repeat the analysis of that work already given.

The stratified rocks of Rosenbusch form a class under the general
idea expressed by Lossen. It is pointed out by Rosenbusch
that the character of the materials of stratified rocks is not so
intimately related to the essence (‘‘ Wesen”’) of the mass as with
eruptive rocks, and hence there is no corresponding firm basis
for their classification.

Stratified rocks are divided into seven families, as follows:

1. Precipitates — including rock salt, gypsum, anhydrite, barite, etc.

2, Psephites and Psammites or clastic rocks.

3. Szliceous rocks—not clastic, partly chemical deposits, partly organic,
partly of undetermined origin, e. g., lydite, diatomaceous earth, sinter, etc.

4. Carbonate rocks —including limestone, dolomite, and impure calcare-
ous rocks, loess, etc.

5. [ron rocks —including spathic iron, spherosiderite, brown hematite,
bog ore, etc.

6. Clay rocks — including clay, clay-slate, phyllite, etc.

7. Porphyroid.

Appendix. Coals, etc.

In this arrangement Rosenbusch attempts no logical con-
struction of anything which can be called a system. As he
frankly admits, the porphyroids are metamorphic rocks, often
associated with the crystalline schists, and as they were not

488 WHITMAN CROSS

derived from sediments it is incorrect to place them in the strati-
fied class. In the necessity for placing coals and other carbon-
aceous rocks in an appendix is further evidence that the arrange-
ment under discussion is inadequate.

The Crystalline Schists are defined by Rosenbusch as alteration
products of eruptive or sedimentary rocks. Both dynamic and
contact metamorphism are recognized as effective in producing
them. Rosenbusch further asserts that the changes have been
entirely structural and molecular, and not chemical, hence by
quantitative analysis of a metamorphic or crystalline schist one
may arrive at a knowledge of the composition of the original
rock, eruptive or sedimentary, from which that schist was derived.
The designation metamorphic rocks is acknowledged to be appro-
priate.

The Crystalline Schists are treated under the following heads :
(r\)-Gneiss,.(2)\\- Mica schist, (3) alc schist, 3(4) 7Chlonite
schist, (5) Amphibole and pyroxene rocks, (6) Serpentine,
(7) Lime series, (8) Magnesian series,,(9) Iron series, (10)
Emery.

Concerning these groups Rosenbusch remarks :

In most of these large groups of the Crystalline Schists, which are held
together mainly through mineral composition, there are united rocksof funda-
mentally different genesis. Therefore they are not natural but rather artifi-
cial series. For the replacement of these artificial groups by natural ones
there is lacking, at the present time, both necessary breadth of experience
and maturity of judgment, from which the need for reform in various direc-
tions is evident.’

Believing that the natural classification of metamorphic rocks
must develop by the historical method, with the increase of
knowledge, Rosenbusch proposes, as a step in the desired direc-
tion, to apply the prefix ortho to the names of gneisses derived

™In den meisten dieser grossen Gruppen von Krystallinischen Schiefern, welche
lediglich durch gleichen oder ahnlichen Mineralbestand zusammengehalten werden,
sind genetisch grundverschiedene Gesteine zusammengefasst. Daher sind sie nicht
natiirliche sondern kiinstliche Reihen. Zur Umgestaltung dieser kiinstlichen Gruppen
in natiirliche fehlt zur Zeit noch einerseits die erforderliche Breite der Erfahrung,
andererseits die Reife des Urtheils und damit das Bediirfniss nach Reform in weiteren
Kreisen. — Elemente, p. 461.

SYSTEMATIC PETROGRAPHY 489

from eruptive rocks, and para to those derived from sediments.
The enumerated divisions of Crystalline Schists are not defined
in a systematic manner, and even the terms gwezss and schist are
given no definite meaning.

The fourth class of rocks advocated by Rosenbusch, on
genetic grounds, ¢he original crustal rocks, is considered by him as
not certainly represented by any known rocks. But it appears
to him probable that they possess the habit of the crystalline
schists.

Johannes Walther, 1897. — An outline of a general classifica-
tion of rocks upon a logical and consequent basis was presented
to the Seventh International Congress of Geologists in St. Peters-
burg, in 1897, by Johannes Walther.* Although but an outline
of a system this proposition deserves attention as the most con-
sistent effort yet made to formulate a system of petrography
co-ordinate in method for different classes of rocks.

Walther starts from the consideration that the growth of
petrographic system in recent years has been very one-sided, a
fact recognized by all. He believes that a natural arrangement
of igneous rocks has been provided by petrographers of the
modern school, while sedimentary and metamorphic rocks are
still arranged upon old and partly incorrect bases. Aiming to
secure a logical system, Walther formulates the following prin-
ciples which he thinks should be observed in the classification
of rocks:

I. The petrogenesis of recent deposits and the direct observation of actual
processes are the fundamental principles of classification.

II. Every older rock has primary characters given It at its formation, and

secondary ones derived by diagenesis or metamorphosis.

III. The derived characters may so change the type of the rock as to
become “essential,” while the primary characters become ‘‘ accessory.”

IV. In spite of this last condition only the primary characters should
determine the principal groups of petrographic system.

V. Next to the primary lithologic characters the primary form of occur-
rence has a classificatory value. There must be distinguished, therefore,
Unstratified, Stratified, and Dike rocks.

t“ Congrés géologique international,’ Compte-Rendu de la VII session, St. Péters-
bourg, 1897, p- 9 (issued in 1899).

490

WAITMAN CROSS

VI. The characters derived by chemical diagenesis, or by contact and
pressure metamorphism, serve for distinction of lesser groups.
VII. The altered rocks are to be placed with their original types.

The system proposed by Walther, in accordance with the

stated rules, is in outline the following:

I. Mechanical Rocks. Composed of older rock fragments; divided by

N oe

Ge

form and size of the fragments into 5 subgroups:

. Breccias: (a@) Unstratified; (4) Stratified ; (¢) Dike form.
. Conglomerates: (a) Stream deposits; (4) Delta deposits; (¢) Strand

deposits.

. Moraines.
. Psammites—sands, more or less sorted: (a) Quartzose sandstone ;

(4) Arkose ; (c) Olivine sands; (@) Iron ore sands, etc.
Pelites— of minute particles: (a) Unstratified ; (6) Stratified —fresh-
water; (c) Stratified — marine.

II. Chemical rocks. Precipitates or sublimates.

I.
2,

3

Calcium carbonate.

Calcium sulphate.

Sodium chloride.

Haloid Salts (‘‘Abraumsalze’’).
Silica.

Carbon.

Ores.

Further divisions are made by occurrence—as Stratified,
Unstratified, or in Dikes.

III. Organic rocks. Formed of the remains of animals or plants.

I.

2.

Limestone: (a) Unstratified—Reef limestone; (6) Stratified —
derived from plants (alge); (¢) Stratified — derived from animals.
Silica : (2) Diatomaceous earth and land plants; (6) Diatomaceous
earth with marine fossils; (c) Radiolarian earth.

IV. Volcanic rocks — Consolidated magmas.

I.

Lavas—Compact rocks: (a) Unstratified —deep-seated rocks; (4)
Stratified — Effusive rocks ; (c) Dike rocks.

. Tuffs — Magmas consolidated in small fragments: (@) Unstratified, in

streams, not sorted; (4) Unstratified, subaqueous accumulations near
vent — water tuffs ; (c) Stratified, sorted according to specific gravity ;
originally inclined; traversed by dikes —tuffs about a land volcano ;
dry tuffs; (2) Stratified, alternating with marine deposits, without
dikes; tuffs of sedimentation ; (e) tuff dikes or chimneys.

In connection with these primary rocks the author mentions,

SYSTEMATIC PETROGRAPHY 49!

as examples, many of the forms derived from them by diagen-
esis or metamorphosis, but does not outline the system for dis-
criminating and naming these alteration products. Some of the
metamorphic rocks, such as gneiss and mica schist, may be
formed from several primary rocks.

The proposition made by Walther is manifestly rather the
work of ageologist than of a petrographer (as was pointed out by
Brégger in discussion, when it was presented to the Congress).
Like many discussions of principles concerned in the systematic
problem, it is not sufficiently worked out to show a practical
result, and does not fully test the adaptability of the chosen fac-
tors for petrographic system. But it seems to the writer that in
this renewing of effort to treat the non-igneous rocks in logical
systematic manner lies ground for hope that something more
than an arrangement for convenience may develop during the
early years of the twentieth century.

Returning to a consideration of the principles adopted by
Walther, it may be remarked that the first one would be excellent
if the processes of rock formation were all open to examination.
Unfortunately, they are not so,in all cases. Many igneous rocks
and nearly all of metamorphic origin have resulted from pro-
cesses we cannot see in operation and can only imperfectly
imitate in experiment. The fourth rule is not a necessary con-
sequence of the facts stated under II and III. It is open to
argument whether the processes which originally produced a
rock are more deserving of recognition in petrographic system
than the processes which have greatly or entirely changed the
characters and perhaps even the composition of the original mass,
making the rock now accessible to our studies.

As to rule V it can hardly be said to warrant the application
made of it, in establishing the three divisions of unstratified,
stratified, and dike forms for all kinds of rocks. Where the
relations expressed by these terms have some genetic connection
with the properties of the rocks they may perhaps be adaptable
to classificatory purposes, but there is no logical reason for
applying this principle in unqualified form.

492 WHITMAN CROSS

The system of Walther seems specially intended to express
the changes rocks undergo rather than their characters as now
seen, and it is not apparent that the’‘author had in mind the apt
and logical analysis of the broad science of rocks which we owe
to Naumann.

That the general treatment proposed by Walther for igneous
rocks, in naming them volcanic, and making the primary division
into unstratified, stratified and dike rocks, has many objections
will be sufficiently clear from the preceding discussions of this
review. The same is true of the assumption that there exists a
satisfactory system for the classification of igneous rocks. The
definition of tuffs as composed of magma consolidated in small
particles certainly applies to but a small part of the pyroclastic
deposits.

SUMMARY.

The science of petrography, the systematic and descriptive
science of rocks, was first fairly outlined by von Leonhard (1823)
and Brongniart (1827) through the distinction between the rock
and the geological terrane, and the setting up of logical classifi-
cations for the former. Neither of these masters gave the
science a name.

The systems of von Leonhard and Brongniart necessarily used
the condition of ignorance concerning the character of many rocks
as a ground for classification. With the increase of knowledge of
rocks there have been many attempts to apply new information to
systematic purposes. Since both the geological relations and the
properties of rocks are highly varied many unlike systems have
been proposed during the century, expressing individual opinions
as to the relative importance or adaptability of principles for
the end in view. Up to the present time, however, no compre-
hensive classification of rocks has been proposed which even pre-
tends to be natural or logically consistent in all its parts.

When we view past petrographic systems, to judge as to how
far they possess natural or artificial features, it is first of all to
be noted that the system of Cordier is practically the only one
starting from the conception that rock species are natural

SYSTEMATIC PETROGRAPHY 493

units and that classification consists in the grouping by more or
less artificial means of these fundamental units. Others have
sought to make the system of rocks in some degree natural by
applying geological factors of occurrence, or genesis, as bases of
classification. The view is apparently held by some that in time
there will be a comprehensive system expressing all important
relations of rocks and that until that result is achieved all
arrangements must be regarded as unsatisfactory and temporary.

It appears to the writer that those who hold this attractive
and apparently philosophical view may not have in mind the
distinction between the formal unit and the rock substance of
that- unit,.or “that distinction :- between the various cross-
classifications of petrology and the one system of petrography,
with which the nomenclature is specially connected. The belief
‘ expressed by Lossen that ‘geological relations must be recog-
and the assertion by Rosen-

”)

nized as petrographical relations
busch that ‘‘petrography is an historical science” illustrate this
point.

If the system of petrography is to be hierarchical, as the
writer believes it should be, the natural element in system is to
be provided for in the judicious selection of broad geological
factors so related to important characters of rocks that the com-
pleted system in the construction of which those characters have
been used, will have a logical and appropriate co-ordination and
sequence of parts. That this aim has not controlled in the past
is evident from the following partial list of designations given to
the rocks which are actually consolidation products of magmas:
“Composite, crystalline-granular, and porphyritic”’ (Zirkel) ;
“‘Non-clastic, composite, massive” (Zirkel) ; ‘‘Composite-simple”’
(von Lasaulx) ; Unstratified, Anogene, Massive, Plutonic, Vol-
canic, Eruptive, Igneous. Here are expressed a number of nat-
ural relations, to be recognized in the proper place, but only
the last term refers directly to the relation most appropriate for
petrographic system. It was not the fact that eruptive force was

™See “The geological versus the petrographical classification of rocks,” by
WHITMAN Cross, Jour. GEOL., Vol. VI, p. 79, 1898.

494 WHITMAN CROSS

exerted to bring molten magmas to the sites of the rocks we
study, but the fact of the molten condition which gave its stamp
of common characters to the products of consolidation.

Arbitrary steps are necessary in the classification of such
objects as rocks, exhibiting gradations in all directions. But
that fact does not justify such artificial systems as many of those
which have been reviewed. Among the most distinctly artificial
systems are those of Cordier, Senft, and von Lasaulx; but scarcely
less so, as regards igneous rocks, are those which, while using
chemical or mineral composition as the basis of arrangement,
use only a portion of the mass. For examples: some of the
chemical classifications take only certain components into
account; Fouqué and Michel-Lévy classify igneous rocks by
the character of that variable portion of the magmas consolidat-
ing during the second period; Rosenbusch uses the phenocrysts
only, in certain parts of his system.

The fundamental requirement that systems should be logical
in construction, with consistent and consequent application of
principles adopted, has been so commonly disregarded that a
summary of instances in point seems unnecessary. Some of the
most widely used systems of today are notably illogical as to
criteria, as has been pointed out.

One of the most serious defects of modern classifications of
‘igneous rocks is a matter of bad logic, and to this defect the
writer wishes to allude once more. It is commonly admitted
that the chemical composition of these rocks is their most funda-
mental characteristic, and many authors would apparently be
glad to apply this character in classification. It is generally
stated, however, that the chemical is represented by the mineral
composition, and as the minerals are so prominent it is con-
venient to use them in system. But with no further discussion
it has been the universal plan to use the minerals in so limited a
qualitative way that they do not in fact express chemical com-
position except in a most crude and inadequate manner. This
procedure is purely and simply illogical, if the intention be to
represent chemical composition by the minerals of the rock.

SMS LEMATTE PETROGRAPHY 495

That some factors have been introduced into classification in
a manner that is quite unscientific seems plain. The age dis-
tinction is one of the factors thus abused. It has long been
known that no general distinction separated pre-Tertiary and
Tertiary igneous rocks. It may be that the average chemical
composition of magmas erupted in successive ages has undergone
some change; but neither the character of the change, nor, least
of all, any special connection with the particular time limit in
question, has been established. The assumption that igneous
rock textures, such as the granular, porphyritic, or vitreous, are
functions of geological form or place of occurrence, is known to
be contrary to the facts displayed by the rocks. Both of these
assumptions have been and are now used in rock classifications.

Stability of system is certainly desirable, within the bounds
of reason. But it is also self-evident that a system of artificial
character, in which the subjective element is dominant, can be
permanent only by universal consent of petrographers, and such
consent is not to be expected. It is a matter of experience that
genetic theories have made systems into which they have been
introduced very unstable and impossible of general adoption.
The danger of using hypotheses in classification has been well
characterized by von Cotta, somewhat as follows: Geology is
a particularly alluring field for premature attempts at the expla-
nation of imperfectly understood facts ; indeed, such attempts
are almost unavoidable in the study of this science. When one
considers hypotheses simply as such, 7. é., as stimulants toward
their possible demonstration, then they are not harmful; the
danger lies therein that one may believe them already proven
and» rest contented, *

The danger pointed out by von Cotta has been illustrated in
the classification of igneous rocks by such able men as von
Richthofen, King, and Rosenbusch. As regards the interior of
the earth, whence the molten magmas come, we cannot as yet
be sure that what we regard as a law today may not be relegated
to the status of a theory or even of an hypothesis tomorrow.

™B. VON Cotta, Gesteinslehre, 2a ed., 1862, p. vi.

496 WHITMAN CROSS

The genetic theory has its proper field of great usefulness in the
department of petrology dealing with petrogenesis. Ultimately
we may hope and expect that genetic relations of igneous rocks
may be available for a more natural classification than is now
feasible.

Any system of classification should be broad and thorough
enough to include all the objects which it professes to deal with.
But the authors of many systems outlined in this review have
been obliged to resort to the expedient of appendices to bring in
rocks not otherwise provided for. Such a necessity is, at once,
evidence of the inadequacy of the criteria guiding the authors of
such systems.

Even in the class of igneous rocks, propositions for chemical
and mineral classifications do not fully recognize the systematic
importance of some of the relatively rare constituents. Chemical
systems which consider all igneous rocks as mixtures of silicates,
or reject magnetite as extraneous, are not comprehensive.
Similarly, the schemes which do not provide for the due recogni-
tion of titanium minerals, corundum, apatite, sulphides, etc., as
important constituents in some cases, are inadequate, even for
present uses, and certainly do not provide for future needs
which can be clearly foreseen.

In conclusion, the status of systematic petrography at the
close of the nineteenth century may be summarized as follows:

1. There is as yet no comprehensive and properly systematic
classification of all rocks. All so-called systems exhibit portions
in which the rocks are treated in an unsystematic manner, for
convenience of description.and discussion. The grand divisions
are not treated by similarly logical and definite methods.

2. Rocks of igneous origin have been much more thoroughly
investigated than others and they have received correspondingly
more definite and systematic classification. The factors used in
systematic construction pertain to genesis, age, and characters.

a. The origin of the great range in chemical composition
exhibited by igneous magmas, expressed in theories of magmatic
differentiation, is an underlying factor of much importance in

SVSTEMATIC PETROGRA PHY 497

the system of Rosenbusch, and is also seen in the desire to
recognize consanguinity of the magmas of petrographic provinces,
as partially worked out by Iddings and Brégger. The avail-
ability of such factors in petrographic system is doubted by
many authorities.

6. While the distinction of older and younger series of rocks
through different sets of names is still found in the German and
French systems there is practical unanimity of opinion that the
real differences between the rocks are much less fundamental
than was supposed. In America, Great Britain, and elsewhere,
this distinction is held to be unwarranted.

c. The chemical and mineral composition of igneous rocks
and their textures are characters used as means of classification
in present systems. Chemical composition fer se is used, but
only by considering a portion of some rocks, and hence fails to
provide an adequate system. The broad chemical divisions used
by some authors are vague and overlapping.

Mineral composition is commonly assumed to represent the
fundamental chemical constitution and to be, therefore, a con-
venient and practicable means of expressing the latter. In
practice the qualitative method of applying mineral composition
in existing systems destroys its effectiveness as expressing
chemical composition.

Structure is variously used in present systems. It is acknowl-
edged to be the product of conditions, and not dependent in
marked degree upon mineral development. When applied as a
primary factor in classification (as by Rosenbusch) it separates
things which are similar in more fundamental characters, and on
this ground some authorities believe that structure should be
applied in classification after the other characters named.

3. The rocks which have formed upon the surface of the
earth by the destruction of older rocks may be viewed from so
many standpoints, as regards the origin of materials, agencies of
transportation, relations to the earth or to other rocks, charac-
ters of materials, and processes of induration, that no consistent
arrangement of these objects, deserving the name of a petro-

498 WHITMAN CROSS

graphic system, has been proposed. In the existing arrange-
ments the confusion of correlating various cross-classifications
into one whole is quite evident.

4. Metamorphic rocks, including all such in which the derived
characters are more prominent than the original ones, defy sys-
tematic treatment at the present time. Since they have been
formed from all kinds of original rocks, by many different pro-
cesses, and at many sites in the earth’s mass, there are many
standpoints from which they may be considered, and their classi-
fication is a complex problem. Among the facts most difficult
to recognize in system are the close resemblance or identity of
metamorphic products from originally different rocks, and the
similar correspondence between certain secondary and primary
rocks. The proposition of Walther to classify all metamorphics
with the masses from which they were derived is thus imprac-
ticable at the present time, even if it be thought desirable. In
relation to this class of rocks systematic petrography is in the
condition that its arrangements are tentative, awaiting new
knowledge concerning the genesis and essential characters of
the objects. )

The review of the development of systematic petrography
given in the preceding pages has been mainly a discussion of
comparatively comprehensive arrangements or systems which
have been proposed. It is, of course, true that these systems
are but correlations of ideas from many sources, and a complete
history of the subject would give to important discoveries of
fact and to critical or creative suggestions their due weight in
influencing the development of systems. But such influence is
difficult to trace, and to have attempted such a history would
have involved the expenditure of much more time than the
writer could devote to the subject. For this reason a large
number of important essays, bearing upon certain features of
classification or devoted to discussion of principles, have been
left unnoticed because they were not accompanied by general

SYSTEMATIC PETROGRAPHY 499

systematic propositions. Among the essays thus disregarded
are notable ones by Rosenbusch, Brégger, Becke, Michel-Lévy,
Teall, Iddings, Spurr, Turner, and many others.

The telling effect of searching investigations touching con-
troverted points of fundamental significance and of the judicial
remarks of those who have carried out such studies is often much
greater than either author or reader is aware. In the course of
time the influence exerted by a succession of investigations
becomes evident in some proposition for the revision of classifi-
cation. Petrography has thus come to its present condition by
a steady natural evolution, and its future growth must undoubt-

edly follow the same course.
WHITMAN Cross.

AN ANALCITE-BEARING CAMPTONITE FROM NEW
MEXICO.

DurinG the summer of 1899 the writer was a member of the
field class of Professor R. D. Salisbury, of the University of Chi-
cago. The party visited the Grand Canyon of the Colorado,
stopping on the way at several localities of interest. The first
halt was made at Las Vegas, New Mexico, where Miss Inez
Rice, a member of the class, guided the party to the butte which
forms the subject of this paper. I take pleasure in acknowledg-
ing my indebtedness to Professor Salisbury for assistance in the
field, and to Miss Rice for suggestions on the general geology
of the region. The petrographic work was done in the laboratory
of Columbia University, and thanks are due to Professor Kemp for
much kindly advice and assistance. Special thanks are also due
to Dr. F. Bascom and to Dr. H. S. Washington, for reading this
paper and for suggestions concerning it. The analysis was very
kindly contributed by Mr. George A. Goodell, of the College of
Physcians and Surgeons; the photograph which forms the
accompanying illustration (Fig. 1) was obtained through the
courtesy of Mr. K. M. Chapman, of Las Vegas. Special thanks
are due to both these gentlemen.

The Las Vegas region exhibits the general geology and
physiography typical of the eastern border of the Rocky Mount-
ains. In it are represented the two great geographical provinces,
the Great Plains and the Cordilleran region. These two geo-
graphical districts are in close correspondence with the geo-
logical structure.

The Cordilleran section consists of a doubtfully Archean
floor, upon the base-leveled surface of which Carboniferous lime-
stones were deposited. This represents an overlap beyond the
Cambrian which underlies the Carboniferous farther north.
Unconformably above the Carboniferous, and situated along

500

AN ANALCITE-BEARING CAMPTONITE SOI

the line bounding the two- geographic regions, are the Red
Beds of Permo-Triassic age, with their associated gypsum
deposits. The relation of physiographic form to geologic
structure is most excellently exhibited. The hard granitic
rocks stand out as rugged peaks, with occasional gentle slopes
where the Carbonfferous limestones are left on their flanks. The
Red Beds represent the softest rock of the region, and the

position of this outcrop is marked by a valley.

Fic. 1.—View of Camptonite butte, Las Vegas, New Mexico.

The contrast between the Cordilleran region and the Great
Plains is both structural and physiographic. The latter region
is underlaid by the Dakota and Colorado formations, with occa-
sional remnants of the Montana and Laramie. Except near the
mountains, the Cretaceous formations are nearly horizontal, but
there they are upturned. Throughout Colorado and northern
New Mexico, the outcrop of the Dakota sandstone forms a series
of ridges locally known as ‘“‘hog-backs.” Las Vegas is situated
at the eastern base of one of them.

About four miles slightly north of east of Las Vegas there is
a little butte of igneous rock of exceptional interest. The struc-
ture of the rock and its occurrence indicate that it was intruded
into the Cretaceous beds at a considerable depth, though it now
stands somewhat above its surroundings.

502 I. H. OGILVIE

The rock is probably to be referred to the group described
by Professor Pirsson™ as ‘the analcite group of igneous rocks,”
and no similar rock has hitherto been described from this region.
Similar masses are mentioned in the Elmoro folio, and termed
‘“Jamprophyres,” hence it is probable that future investigation
will bring to light other related intrusions.. No elaeolite syenite
nor other rock that could be related to these lamphrophyric
intrusions has yet been found; hence the relationship of the rock
and the magmatic history of the group to which it belongs,
remain as problems to be studied in connection with the further
investigation of the region.

The rock is medium grained, of a general gray appearance,
with large crystals of hornblende and augite which can readily
be seen with the naked eye. In the predominating phase of the
rock, these crystals are about .8™™ in length, but in occa-
sional segregations they reach a length of as much as 1%.
These ferro-magnesian minerals lie in a gray groundmass
which on microscopical examination proves to be mostly plagio-
clase.

In thin section the rock is seen to be porphyritic, with phen-
ocrysts of augite and of hornblende, occurring in equal
amounts, and of rarer biotite. The groundmass consists of a
network of plagioclase, with an isotropic substance which is
probably analcite, filling the interstices between the plagioclase
laths. Magnetite, ilmenite, and apatite are also present.

The pyroxene is always idiomorphic, occurring in large phen-
ocrysts and also to a small degree in the groundmass. It is a
pale greenish-violet, normal augite, and is very faintly pleochroic.
The cleavage is well defined. Some of the porphyritic crystals
show a slight zonal structure. Twinning is very common and

; 6 cj I ies F
usually the twinning plane is « P= (100). Certain complicated

intergrowths also occur, which probably also represent twinning,
the twinning planes being —Ps (Ton), and 75122) (igeg2)e

™L. V. Pirsson, Jour. GEOL., Vol. IV, 1896, pp. 679-690.

AN ANALCITE-BEARING CAMPTONITE 503

Intergrowths of augite and hornblende are common, and so are
occasional inclusions of augite in hornblende (Fig. 3).

The amphibole is idiomorphic, occurring only as a pheno-
cryst. It is of the basaltic hornblende type. The pleochroism
is very strong, C and b = deep brown, A= pink. The terminal
faces are usually
lacking. These phen-
ocrysts exhibit the
characteristic cleav-
ages of hornblende.
In a few slides a very
small amount of
secondary horn-
blende was’ found
associated with the
augite, but the pre-
vailing hornblende is
certainly an original
constituent.

Mica occurs in

small quantity, as
irregular shred-like
phenocrysts. It isa
very pleochroic biotite, changing from brownish-black to reddish-

Fic. 2.—Twinned aggregate of augite.

brown.

The feldspar occurs in the groundmass, as lath-shaped, poly-
synthetically twinned crystals. They form an interlocking net-
work which is difficult of interpretation; a number of readings
Of extinction angles onthe P face varied fromui5 to: 357,
indicating a plagioclase rich in lime (bytownite or anorthite).
The presence of this plagioclase is further indicated by the
high lime percentage of the analysis.

Lying between the laths of feldspar is an isotropic substance
which appears to be analcite. It occurs in such small areas and
is so thoroughly mixed with the groundmass that determina-
tions of it were necessarily imperfect. In one instance it exhib-

504 Lelie nO GALE ATS

ited the cubical cleavages of analcite. The rock was fresh and
this mineral was undoubtedly of primary origin, being appar-
ently the last to crystallize and filling all interstices. The larger
grains are free from inclusions, but are sometimes surrounded
by rings of magnetite and ilmenite grains; the smaller grains
often contain a fine
black dust suggest-
ive of leucite, » This
structure appears to
be identical with that
described: by. Mig:
Cross in the case of
an analcite basalt."
Magnetite and
ilmenite are abund-
antly distributed in
large crystals. Apa-
tite olecurs7imatine
form of elongated
prisms with trunca-

ted corners, and is
common throughout

Fic. 3.—Intergrowth of hornblende and augite,
with inclusions of augite in hornblende. the groundmass.

Very little alter-
ation could be seen in any of the constituents. As already

mentioned, very small amounts of secondary hornblende occur,
derived from the augite. The plagioclase is occasionally slightly
sausuritized, and small quantities of secondary epidote are occa-
sionally found. Therock as a whole is, however, remarkably
fresh.

The most noted district for rocks of this class is the neigh-
borhood of Lake Champlain, where they have been made known
principally through the work of Professor Kemp.? The writer

WHITMAN Cross “An Analcite-basalt from Colorado,” Jour. GEOL., Vol. V,
1897.

2J. F. Kemp, ‘Trap Dikes of Lake Champlain,” Bud. 707, U. S. Geol. Surv.

AN ANALCITE-BEARING CAMPTONITE 505

has compared the slides of the Las Vegas rock with those from
the Champlain region, and also with those from Campton Falls,’
from Whitehall and Fairhaven, Vt.,? and several neighboring
localities ; and with a series from the Black Forest in Germany.
When all these camptonites were studied and compared with
the Las Vegas rock,
a marked difference
in texture was seen.
The Las Vegas rock
is coarser grained,
the phenocrysts more
abundant, and there
ismaeless) nranked
difference in size be-
tween phenocrysts
and groundmass.
This is what might
be expected since the
eastern rocks occur
in dikes, and the Las
Vegas one ina stock.

The eastern rocks are

Fic. 4.—Typical section of the Las Vegas Camp-
all considerabl Yh tonite, showing phenocrysts and groundmass.
altered, containing

calcite, serpentine, delessite, and secondary analcite. The Las

Vegas rock is remarkably fresh. This comparison gives strong
indirect evidence of the primary character and analcitic nature
of the isotropic substance; the rock is too coarse-grained for it
to be a glass and too fresh for it to be a secondary product.

dv)

The term ‘‘Camptonite”’ is commonly applied to the plagio-
clasic lamprophyres, in distinction from minette and vogesite
which are orthoclasic. Common usage has restricted the term
Camptonite to the lamprophyric plagioclase rocks with horn-

™ Described by J. W. Hawes, Amer. Jour. Scz., ser. 3, Vol. XVII, p. 148.

2 Described by J. F. Kemp, Amer. Geol., Aug., 1889.

3F. Bascom, Wineteenth Ann. Rept. U.S. Geol. Surv., Part III.

506 RAIOGTEVULE

ANALYSES.
I II Ill IV. Vv VI

SLO Mire sevan cers 44.48 38.45 41.94 45.70 45.11 42.08
IMO poacad cob Uo 20.43 19.08 15.36 20.48 19.67 20.88
In O)nepeva ace nono.0 9.72 4.01 B27) 1.99 4.32 6.77
FeO ce noiunie 2.18 II.15 9.89 4.18 8.57 Bi 1G)
Ca@ renege senile 10.35 9.37 9.47 Wits 7) 10.45 12.48
Mig @ ee htencreysretets 5.51 6.65 Os 8.50 5.65 6.85
HM @ sous tierce oemereeerslene 57 deaeas 4.15 0.21
Min Orssapenceona ss aoe trace 0.25 Soin
Kes Ose earn oe near 1.59 172 0.19 0.80 0.64 0.44
Nias Ore eerie sonata tee 3.61 277 5.15 3.56 Bor] 2337
Loss on ignition... 22 acta 2520 2.80 Aare ays
EVs Or eeu iierereneuscpeeatae ane 1.49 ASE Byrae 0.17 3.18
COR tua eciaenee is 4.82 Aaats Stier re Soar
P,0;. 0.25

otal sini sverssess 101.65 100.11 100.44 -| 99.64 100.07 99.22

I. The Las Vegas rock.

II. Camptonite from Campton Falls, N. H. Published in Bull. 748, U. S. Geol.
Survey, Pp. 67.

III. Camptonite from Campton Falls, N. H. J. W. Hawes, Am. Jour. Sci.
Ser. 3, Vol XVII, p. 150.

IV. Gabbro from Rosswein, Saxony. Landwirthsch Versuchs station 40, 1892.

V. Diorite from Lindenfels, Hesse Darmstadt. Chelies & Klemm, 27lauterung
zur Geologischen Karte von Hesse, 1896.

VI. Basalt, middle flow, Carlsbad, Bohemia. Jahrbuch der Konighch Katser-
lichen Gesellschaft Kunstaustelinng, Vol. XL, p. 345, 1890.
blende as the principal ferro-magnesian mineral; augite camp-
tonite is applied to those in which augite equals or exceeds
hornblende; analcitite to those containing primary analcite.
Rosenbusch’s latest definition (1901) includes augite as an
essential constituent of a camptonite, thereby departing from
the original type of Hawes in which hornblende predominated.

In the case of the Las Vegas rock hornblende and augite
occur in approximately equal quantities. Since the tests for
analcite were not conclusive, the rock could not be called an
analcitite. It seems most logical to regard equal amounts of
hornblende and of augite as distinctive of a camptonite, restrict-
ing augite camptonite to those with augite in excess. The Las
Vegas rock can thus most logically be termed an analcite bear-

AN ANALCITE-BEA RING CAMPTONITE 507

ing camptonite. I am indebted to Dr. Bascom for suggestions
on nomenclature.

The accurate recalculation of the analysis proved impossible,
owing to the combination of minerals containing the same
oxides. The high Na,O is a strong indication that the isotropic
constituent is analcite, and this is further indicated by the
optical character of the plagioclase ; it is near the anorthite end
of the series, hence, has low Na,O. Microscopically horn-
blende, augite, and plagioclase are present in approximately
equal amounts.

Analyses two and three are typical camptonites, and it will
readily be seen that the Las Vegas rock is close to them chem-
ically. Analyses four, five and six are of rocks of other groups
which are also similar chemically. The likeness of the last three

was kindly suggested by Dr. Washington.
T. A. Ocievir:
GEOLOGICAL DEPARTMENT,
: COLUMBIA UNIVERSITY.

HOLYOKEITE, Ay BURBLY SEELDSPATEHICS DUABASE
FROM, THE TRIAS, OF MASSACHUSEDTDS:

In the monograph of the three river counties in Massachu-
setts,? the writer described a “white trap’”’ which occurs only in
scanty fragments ina bed of agglomerate interstratified in the
sandstone, a few feet above the surface of the great Holyoke
trap sheet at the east foot of Mount Tom, and a few rods north
of the station of the electric road going up onto the mountain.
The small angular fragments of the volcanic rock are scattered
rather distantly in the calcareous red sandstone, and seem
closely like a white, horny limestone spotted with chalcopyrite.
They include fragments of the coarse sandstone below the
Holyoke trap sheet, up through which they must have come, and
these inclusions are much coarser than the sandstone in which
they are included. The weathered surfaces show the rock to be
finely amygdaloidal, and acid brings out in the interior the same
structure which can indeed be seen, by attentive study with
a lens, on a freshly broken surface. <A few grains of yellow ore
appear here and there in the body of the rock and in the round
cavities, and rarely the reflection of a twinned plagioclase lath
is visible.

The thin section shows so exactly every structure of the dia-
base except those dependent upon the presence of iron in form
of magnetite and augite that one cannot help associating it
closely with the adjacent Holyoke sheet, as I have formerly
done by calling it a ‘‘white trap.” Even the presence of much
chalcopyrite is characteristic. Under the microscope the texture
is exactly that of the trap of the large sheets minus the augite.
It is especially like the superficial portion of the Deerfield sheet
exposed at Cheapside.

There is an ophitic network of very fine plagioclase needles

t Published by permission of the director of the United States Geological Survey.

? Monograph XXTX, U.S. Geol. Surv., 1898, pp. 365-474.

508

HIOLVOKEITE, A PURELY FELDSPATHIC DIABASE 509

of two sizes, both elongate blades with ragged ends and irregu-
lar sides; the finer about 0.03™" long; the latter having some-
times quite regular crystalline outlines. Scattered in this
network are distant feathery groups of plagioclase crystals of
first consolidation which are just visible to the eye. One of
these larger crystals cut parallel to M (010) showed the optical
figure of albite.

All these feldspars are lightly dusted with blades and grains
of a secondary mineral of low polarization, probably zoisite.
And there is avery little granular limonite scattered through the
rock, but no accumulation of it, or appearance of any chloritic
mineral to indicate the former presence of magnetite and augite
or hornblende. There are no brightly polarizing blades that
could be referred to a colorless bisilicate.

As in the peculiar red diabase from Cheapside, near Green-
field, described in the monograph* above mentioned, the small
round steam cavities are lined by a secondary growth of albite in
fresh limpid crystals.

There are other fragments associated with the Holyokeite
which are black, aphanitic, and show under the microscope a simi-
lar texture, but contain large spots of a green chloritic mineral.

The chemical composition of the rock is shown by the fol-
lowing analysis I, by Mr. Hillebrand.

If we follow the calculation of the analyst and assign the
sulphur to 0.40 chalcopyrite and 0.06 per cent. pyrite, and then
calculate the phosphoric acid as apatite, the titanic acid as
ilmenite, the potash as orthoclase, and assign the carbonic acid
to the magnesia and most of the calcium, we shall account for
30 per cent. of the analysis.

It is interesting that the remaining 70 per cent. has the com-
position of an albite of exceptional purity, and the only feldspar
determined in the slide was albite.

SiO eee. 167.80
Al,O3 = = 20.87
Na,O - - = Tl .24'

100.00

* Loe. ctt., P. 431.

510 BK. EMERSON

if 1D bs Ill
SiO@parreicsced 53.83 60.13 51.78
IKON Maaco 16.36 20.47 12.79
Fe,O3;. : 1.04 3.59
Re OR tan. wee Oats) \ 72 8.25
Mig. OR erlereker Gite laity 7.63
CaOr ear 9.81 2.59 10.70
INYO) Gaco ware 7.89 9.60 2.14
Kes Oversee 1.58 1.06 217 1039
HO. nee 1 GL Se eget at he aces 168)
H,O--..... BO ceca: cumin eee amet hcg Wengen
SRO RS eos 86 trace 1.41
LY OG Musienclaees 2) ieee alee wee CECE sa A ge ren era vere
3-44
COB oh mine 7.47 (Ignition, aes
includes H,O) )
lemOen eo aon TT Seepage olen ake Savors nin: 14
anakerausrelcene evs adie aio os eiievenete
Mn@meencee a little lost trace 0.44
al OR aes ae TIO IVC eo eee eh eons oy yee yret I heed aa! enue Mama Taste cove
SrOeconriece TOMS tes eae Dit la Teck Oncaea oe Wa coetiite || Pomel ean en stares
ih) OF © a eee a MONE, oA), Wm mal AUER ie Me |e eh a mee as
GuCurarasee STARA Sta e eea, clineateteneret sini aia b Restle 20 Waele at ent
99.77 100.20 99.89

I. Analysis of. Holyokeite made by Mr. Hillebrand, of the United States Geo-
logical Survey.

Il. Analysis by Dr. H. S. Washington of the acid dyke in the Connecticut Trias,
described by Mr. E. O. Hovey and mentioned below. Specific Gravity at 11°
C263"

ILI. Analysis of the normal Triassic diabase of West Rock, New Haven, by Mr.
G. W. Hawes. Am. Jour. of Sci., 111, 1X, 186, 1875.

*The figure here given (0.89) includes not only ferric and ferrous iron, but
aiso that in combination with sulphur, as (Cu FeS,) and possibly a little FeS,.
Because of the sulphide FeO could not be estimated. (Hillebrand.)

2 Taking the Cu as a basis the sulphides figure out 0.40 percent.CuFeS,. I can-
not say that there is any FeS, present. The mass of the sulphide seems certainly to
be chalcopyrite. (Hillebrand.)

The whole rock will then contain, roughly speaking:

Apatite - - =H. 253
Dolomite - - 0.52
Galcites ie - - 16.42
Orthoclase - - 9.41
Albite” - - - 70.25
Ilmenite - - 1.63
Chalcopyrite - =) =-40
Pyrite - - - .06

100.00

TIOLVOKEILED, AY PORELY FELDSPATHIC DIABASE 511

This ignores the water which might have been calculated,
perhaps, as kaolin, since there is no visible chloritic mineral or
zeolite. It ignores alsoa third of a per cent. of CaO, which now
forms calcite, and which with the rest of the calcium was present
in the original rock, perhaps for the most part, as anorthite;
that combined with the albite formed several intermediate varie-
ties of plagioclase.

This calcium may have been present in part as a sahlite. In
the latter case it might have been called a sahlite-diabase, but
in a sense very different from that in which the word has been
used by Térnebohm for a diabase in which the sahlite is quite
subordinate to the abundant augite. The rock in which the frag-
ments are embedded is, however, so calcareous that some part of
the calcite may have been introduced into the amygdaloidal cavi-
ties from without.

The leucophyr of Gtimbel is, as compared with diabase, ‘‘a
remarkably light-colored rock with saussuritic feldspar, pale
green augite (without hornblende and rarely with red-brown
augite), with a chloritic constituent in large quantity and tabular

d

ilmenite.” This seems to me quite plainly an altered diabase,
and different from this non-ferruginous rock.

thes onlyesimilar, rock: that, has; been described irom the
American Trias is the acid dyke discovered by Mr. E. O. Hovey
in the new cut on the Shore Line Railroad, in the eastern part
of New Haven, provisionally referred by him to keratophyr. It
is distinctly different from the type described here.* It is brick-
red like the Cheapside trap, or grayish like the second variety
described above, which accompanies the white trap.

It is largely feldspathic and the ferro-magnesian minerals are
absent, but chlorite is present in considerable abundance, and the
fields of chlorite and calcite ‘‘for the most part have definite
outlines which strongly suggest the original presence of pheno-
crysts of pyroxene (augite) in the rock.” Mr. Hovey remarks
that spherocrystalline structure was observed in some places.
This may have been the structure that I have interpreted above

Am. Jour. Sci., ser. 4, Vol. III, p. 287, 1897.

512 B. K. EMERSON

as a growth of secondary albite in steam holes. This sug-
gestion depends upon a comparison with material of excep-
tional excellence from Cheapside where the whole process could
be followed and where the perfect crystals of albite could be
isolated and determined in heavy fluid. The chemical differ-
ences between the two rocks are many, as may be seen by com-
paring the analyses given above. In the matter of SiO, and
Al,O,, the Holyokeite is intermediate between the keratophyr
of Hovey and the normal diabase.

The minute quantity of iron (not enough to satisfy the S and
TiO,) and magnesia, show that there can scarcely have been a
trace of augite or sahlite present in the Holyokeite, while the
keratophyr may have contained much bisilicate. Finally the
discrepancy in the CaO is decided, even after saturating the
CO,, and would demand, considering also the smaller amount
of SiO,, a somewhat larger proportion of anorthite in the Holy-
okeite, while the larger amount of K,O would demand for it a
greater amount of orthoclase or anorthoclase.

The Holyokeite may then be looked upon as representing a
limiting form of the diabase series where the bisilicates are

wholly wanting.
B. K. EMERSON.

Lip oeNDSEIDESTOR Mir. GREYLOCK AND
BRIGGSVILLE, MASS.

A stupy of the topographic and geologic maps of Mt. Grey-
lock will show where, theoretically, landslides would be most
likely to occur. Both the large and small slides on this moun-
tain are where, in theory, they should be. These slides are
typical of one class, and the one at Briggsville of another.

The scarring of the mountain side by this agency is a very
striking feature of the landscape, as one looks west from the
Hoosac range and from the Hoosic valley for two or three
miles on either side of Adams. The eastern side of Greylock is
the steepest on the mountain, rising 1400 feet in two-fifths of a
mile. It was on this slope, the base of which is about two miles
west of Adams, that the land slipped. In all there are three
large and eight small landslides within a mile.

The facts concerning the conditions at the time at which the
landslides occurred were gathered largely from persons in the
vicinity of Adams, but especially from Mr. C. O. Gould, whose
farm is located at the foot of the main slide. The facts are as
follows: During July and August, previous to August 20, 1901,
the date on which the slides occurred, 12.92 inches of rain fell,
being 5.8 inches above normal. At about three o’clock on the
afternoon of the 20th a very heavy rain, spoken of usually as a
cloudburst, began to fall, and lasted until seven o’clock. The
record for this period at Williamstown (on the other side of the
mountain) is 2 inches; at Mt. Tom, 3.42 inches. The water in
the creek at the foot of the mountain rose very rapidly until it
was as high as it had ever been noticed. When the slide
occurred the stream became ‘‘a torrent 75 to 100 feet wide, 10
feet deep, and filled with trees.”’ Five or six acres of the farm
were covered with sand and bowlders, some of the bowlders
being 4 feet in diameter.

513

514

HE RCEEEAND

(The steepness is not well shown.)

Mt. Greylock, showing landslides.

LANDSLIDES OF MT. GREYLOCK 515

The middle slide came first, and the large one to the north
next. The time of the occurrence of the large slide to the
south was not noticed, but all must have occurred within a few
minutes of each other.

The mountain in this place is made up, above, of a mica
schist (Greylock schist), and, below, of a rotten micaceous lime-
stone (Bellowspipe limestone). The slope is covered with loose
talus and earth, upon which is a sparse growth of trees.

The north landslide occurred at the road near the top of the
mountain, where the slide is about 50 feet wide, and extends for
1500 feet down the mountain side. It gradually widens until
at the base it is about 200 feet wide.

From the above it will be seen that all of the conditions
were favorable for a landslide; the steepness of the slope, the
fissile character of the rock, the loose material of the mountain-
side wet with the unusually heavy summer rain. All that was
needed to cause a slipping of this loose material was a large
quantity of water to add to the weight of the mass, and at the
same time act as a lubricant. The destructive force of the
water was caused by the natural drainage of an unusually heavy
rain, and by the water brought down and squeezed out of the
mass which slid down the mountain-side. There is no evidence
that a dam was formed. In the ravine through which the water
flowed the water was in places 25 feet deep, as shown by the
overflow.

The removed material was not piled up with a backward
slope toward the mountain, as is usually the case, the great
quantity of water being sufficient to remove a large part of the
‘débris so that much of the rock and earth was carried to and
deposited in the valley below.

The landslide at Briggsville is more complex. Back of the
woolen mills of Strong, Hewatt & Co., and on the west bank of
Beaver Creek, one sees stratified blue clay tilted at all angles
from a 90-degree to an almost horizontal position. A pier is
raised 4 feet and tilted to the west, and the dam to the north of
the mill is raised several feet on the west side. The tilted area

516 Fin CLELAND

is about 325 feet long and 33 feet wide. To the west of this is
an excavation from which the material in the creek bed was
derived. This cavity is some 320 feet in length from east to
west, and about 30 feet lower than the original surface.

The history of the phenomena is as follows: During 1895 a
spring, called a ‘‘geyser”’ in this vicinity, broke out with con-
siderable force. The water contained a great deal of sediment,
(finely divided blue clay) which was at first “as thick as pea

Topographic map of Greylock with position of the three large landslides.

soup’’ and contained large quantities of decayed roots. The
muddy water was of sufficient volume to so contaminate the
water of Beaver Creek as to render it unfit for the use of the
Windsor Print Works. ;

The land above the spring gradually sank as the bank was
undermined and with each subsidence of the surface the spring
moved back toward the high bank to the west. Before the
sinking occurred growing trees sank beneath the surface, in an
upright position, and poles could be driven down several feet
with little effort. During a flood in March, Igol, the bridge
pier was tilted, the dam was raised, and the clay appeared in the
creek bed. This land continued to rise for several months.

The explanation is as follows: The sediment in the water
was produced by the breaking and fracturing of the clay bank,

LANDSLIDES OF MT, GREYLOCK Siy/

which fracturing so loosened the clay that it could be carried
away in suspension. This continued until a large quantity of
clay had been removed from a stratum immediately underlying
the surface sod. The most evident explanation for the slipping
of the clay and its uptilting in the stream bed is that a stratum
of blue clay rested on an inclined plane which was rendered
slippery by the water of the spring. This stratum was separated
from the surface sod by the washing away of the clay between
them. When it slipped it was stopped either by the former
rock bank of the stream or by some other obstacle, the slope
and weight being sufficient to push the clay up from the stream
bed and to raise with it the pier and part of the dam.

After the last slipping the muddiness of the water was
greatly lessened showing that the movement of the ground had
almost ceased. Since then, however, the amount of sediment
in the water has increased, indicating that a further fracturing
is in progress.

HF: Greranp:

WILLIAMSTOWN, MASss.

NIAGARA METEORITE.

In the winter of 1900 I was informed by Mr. Vennor, a young
mineral collector of Rochester, N. Y., that he had seen in the
possession of one of his friends, now living in this city, what he
supposed to be a small iron meteorite, which had been picked
up on a ranch in North Dakota, some years before. After some
persuasion Mr. Vennor succeeded in getting his friend to permit
me to examine the specimen, which I saw at a glance was a
meteorite with marked octahedral cleavage where it had been
broken, and finally purchased it for Ward’s Natural Science
establishment.

This meteorite, which belongs to the siderite class, was found
two miles southeast of Niagara, Forks county, N. D.,%in’ the
early part of August, 1879, by Mr. F. Talbot, the son of the
lady from whom Mr. Vennor obtained it for me. Mr. Talbot
was making a collection of the various rocks and minerals in his
neighborhood, and in his search for these discovered the above
meteorite on the ranch owned by his father. As Mr. Talbot has
‘since passed away, the specimen had been retained in his family
rather as a keepsake in remembrance of him than as a visitor
from outer space.

When I received the meteorite it weighed 115 grams and
was about 30X40 X 50™™ in dimensions. It was very much oxi-
dized, of a brownish-black color, and showed no trace of the
original crust whatever. In sawing it we succeeded in getting
but three sizeable pieces, as it crumbled into small fragments
from 2 to 4 or 5 grams in weight in cutting. The largest piece
obtained of the unoxidized iron suitable for etching weighed
26 grams and is now in the Ward-Coonley collection of
meteorites. The only other piece suitable for etching weighed
17 grams and is ‘im sthe) (British) Museum jcollection: =) ihe
largest fragment from the crumbled portion of the iron was a
fairly good tetrahedron, 24 X 23 X 15™™ in its greatest diame-

518

NIAGARA METEORITE 519

ters, and weighed 18 grams. This fragment showed bright
plates of taenite on two of its cleaved faces. This feature was
likewise shown on a number of the smaller fragments.

On etching the two pieces composed of the unoxidized iron,
its octahedral structure was strongly marked in the Widman-
statten figures, the kamacite plates being somewhat broad with
a second series of markings of hair-like lines upon them about
the size of the Neumann lines on the Braunau iron.

Mr. J. M. Davison, of the Reynolds Laboratory of the Uni-
versity of Rochester, kindly made the analysis of this meteorite
for me, and found it to be composed of

Fe iat 12.9 2167
Mier ech 7.37
CO - - - - =) 0.13

100.17

Specific gravity 7.12.
From its close proximity to Niagara, we will designate the
siderite the Niagara, Forks county, N. Dakota Meteorite.

Hy La RRESTON:
ROCHESTER, N, Y.

ON THE ~SKULE VOR NY ClODAGCTVEUS 2A Ne UR Prk
CREDACHOUS PIE RODACDY I.

ELSEWHERE I have published a brief description, with a
restoration,” of an unusually complete specimen of Wyctodactylus,
recently collected by my assistant, Mr. H. T. Martin, in western
Kansas. At the time this restoration was made, the skull was
lying in its original matrix, with its palatal surface uppermost.
On account of its extreme delicacy, it was thought best then not
to attempt its removal. This, however, has since been done in
a most skillful manner by Mr. Martin, and it is now almost com-
pletely removed from the chalk in which it was embedded. Its
delicacy and fragility may be suspected from its weight, scarcely
thirty grammes, the mandible weighing about nine grammes
more. The remarkable perfection and relative freedom from
distortion render the specimen especially valuable for study.
The mandible was displaced, in no wise obscuring other parts.
The skull is of course depressed, but not so much as one might
expect from the position in which it was lying, and there is a
slight lateral compression to the right, but not enough to render
the interpretation of the characters at all difficult. I give here-
with a photographic reproduction of the back part of the skull, as
it was seen before removal from the matrix. The anterior por-
tion, on a separate block of matrix, has been omitted in order to
show the hyoid bones lying partly beneath it. In the drawings
I have attempted to restore the skull as nearly as possible to its
living condition. There is yet a possibility that I have made
the skull a little too broad, but, if so, the exaggeration must be so
small as to be scarcely appreciable. The drawings were made by
myself, at least insuring a careful study of the parts.

The premaxillary doubtless comprises the whole anterior por-
tion of the beak, as in other known pterodactyls. Inthe middle
behind, it separates the nares by a rather broad, gently convex

* American Journal of Anatomy, Vol. I, p. 297.

520

ON THE SISGLLE OFY NYCTODACTYLUS, 521

or laterally flattened bridge of stout bone, gradually and nearly
uniformly decreasing in width posteriorly to an acute point
between the frontals near the middle of the orbits. Its margins
are rounded as far back as the union with the anterior pro-
cesses of the frontals, which continue the same convex borders
to the free margin of the orbits. The sutures separating the
bone from the frontal posteriorly are indicated by lines exactly
alike on the two sides, and hence evidently sutural. On the
sides the union with the maxille is indeterminate. If this union
is as in Rhamphorhynchus, as described by Woodward,’ its posi-
tion would be back of the anterior end of the nares. In Scaph-
ognathus, however, Plieninger? would locate the suture at the
anterior extremity of the nares. Its alveolar borders form
straight, sharp, smooth ridges, with the inner wall two or three
millimeters in width, meeting the palatal surface at an angle of
about 135 degrees. The middle portion above is regularly
rounded, forming an arc of about a third of a circle. This por-
tion has its walls distinctly thicker than elsewhere, and has
resisted compression; the sides from this median convexity
seem to have been gently convex, meeting the middle convexity
in a shallow groove. The surface is smooth, as in Oruithos-
toma (Pteranodon), without pits or depressions. As in that genus
the texture of the bone of the beak has a distinct fibrillation
longitudinally, sufficiently well marked to enable one to distinguish
fragments from this part of the skeleton; a fact long since
recognized by collectors of these animal remains in the field. I
am convinced there were no pits, depressions, or conspicuous
foramina on the premaxillary, and that neither in this genus nor
in Ornithostoma was there a longitudinal ridge in the middle
above. The proof of this is positive in both this genus and in
Ornithostoma. On the under side, the bone forming the palatal
surface, in part probably composed of the premaxilla, in part of
the vomers, is of extreme tenuity, of the thickness of ordinary
writing paper. It appears to have been gently concave through-
tAnn. Mag. Nat. Hist., Vol. 1X, No. 2, 1902.
2 Paleontographica, Vol. XLI, p. 203, 1894.

522 Ss WE WILLISTON

out and smooth; on the sides between this bone and the roof
the space was more or less filled in with a loose cancellated bone,
but under the convexity of the middle there was a considerable
cavity. On the under side of the premaxillary, back of the anterior
ends of the nasals, are two pits or depressions, separated by a low
ridge, evidently rhinencephalic fosse.

The frontal bone is broad, unpaired, smooth, and nearly flat.
Anteriorly it extends forward on either side as far as the posterior
end of the nares, receiving between its two pointed processes the
posterior pointed extremity of the premaxillary, as already
described. There is no indication whatever of a median crest, a
fact to which doubtless is due the position of the skull upon its
dorsal surface, a position which I have never observed in speci-
mens of Ornithostoma. Plieninger speaks of a rudimentary
frontal crest in Prterodactylus, and its entire absence in this genus
is remarkable. Posteriorly, the union with the parietal is inde-
terminable. On the sides it forms the horizontal, thin, superior
border of the orbit for a little more than its posterior half,
terminating in an angle a little before the beginning of the
attachment of the postfrontal. At the anterior angle the border
turns inward with a concavity looking forward, and then curves
forward to be continuous with the sides of the premaxillary.
The parietal, or fronto-parietal, continues the plane of the frontal
in along, U-shaped plane, forming the rounded posterior extremity
of the skull. The sides of this slender U form the inner and
front margin of the supratemporal fossa; from this slightly
convex margin posteriorly the parietal descends downward and
outward in a thin wing, articulating at the extremity with the
posterior flattened end of the squamosal. This wing-like process
narrows anteriorly to be continuous with the sides of the brain-
Case:

The nasal is a slender bar of bone, beginning on the under
side of the premaxillary,a little back of the middle of the nares,
and extending downward and outward to unite with the ascend-
ing process of the maxilla to form the posterior boundary of the
nares. Posteriorly it is a cribriform plate filling up the concave

ONVITLE TSIGOLE OF NVCTODACTVLGS 523

sinus in front of the prefronto-lachrymal bone. The bone evi-
dently continues over the outer surface of the maxillary process
to unite with the lachrymal, and perhaps also with the ascending
process of the malar. On the under side of the premaxillary
the bones of the two sides may meet in the middle, though this
seems doubtful.

The prefrontal and lachrymal cannot be distinguished from
each other, and there seems to be some disagreement among
writers as to which of the two elements should be called the pre-
frontal, were they distinguishable. The prefrontal must be that
portion articulating along the sides of the premaxillary and
frontal, while the lachrymal is that part which forms the anterior
or anterior superior border of the orbit, either articulating
directly with the frontal and the prefrontal, or with the prefrontal
only, which one should expect. In the present specimen the
compound element is sharply distinguishable from the frontal
and premaxillary, as also the nasal. It begins on the side of the
premaxilla a little back of the visible surface of the nasal, to
which it is attached, as a slender pointed anterior process, and
widens out posteriorly, forming a concave border, to which the
thin, imperfectly ossified posterior part of the nasal attaches.
Externally it widens out again into the broader horizontal plate
forming the anterior superior roof of the orbit, which anteriorly
curves downward to meet the jugal. Its orbital margin is thin.

The jugal bone extends forward on the outer and inferior
side of the maxilla to beyond the middle of the nares. How
far front it goes cannot be determined. In Ornithostoma the dis-
tinguishing suture, placed as in Wyctodactylus, seems to be con-
tinued as far forward as the anterior end of the nares, which
would seem to indicate that the premaxilla does not reach as far
back as in Rhamphorhynchus. At the front extremity of the
orbit it sends up a slender process of bone, superimposed upon
a thin expansion of the maxilla, to meet the lachrymal and nasal.
Beyond this it forms a narrow bar, four or five millimeters in
width, to the lower extremity of the quadrate, and then turns
upward and backward a little more broadly to join, by a long

524 Se WHE ELSTON.

squamosal suture, the outer side of the quadrato-jugal, forming
the anterior superior boundary of the lower temporal vacuity.
In Ornithostoma it very clearly unites above with both the
quadrato-jugal and postfrontal, and in all probability the same
relations obtain in Vyctodactylus, but of this I am not sure.

The supratemporal bar is evidently formed of two bones, as
in Ornithostoma. The superior one, joining the outer angle of
the frontal (and parietal), is clearly the postfrontal or post-
fronto-orbital. It extends backward to the head of the quad-
rate, also touching the squamosal, and probably also joining the
jugal, as in Ornithostoma. ‘The inferior outer part of the bar,
uniting with the jugal, as described, anteriorly, the head of the
quadrate, and apparently also the squamosal posteriorly, is, I
suppose, the quadrato-jugal, though it corresponds precisely in
position and relations with the bone I have called the pro-
squamosal in the mosasaurs, following Baur. This bone in these
and other lacertilians has, at various times, been called the pro-
squamosal (Baur, Williston), the quadrato-jugal (Gegenbaur,
Merriam, Baur), the squamosal (Owen, Huxley, Cope, Baur),
and the supratemporal (Cope). In this pterodactyl the bone
does not descend to separate the jugal from the quadrate.
This, according to Seeley, is sometimes the condition among the
pterodactyls. If so, the element must surely be the quadrato-
jugal, and not the prosquamosal.

The squamosal seems clearly defined. It is a triangular, curved
bone. A small process runs forward to join the post-frontal
and ‘ quadrato-jugal;’’ a broader and flat one curves down-
ward and inward to join the wing-like squamosal process of the
parietal, forming the parieto-squamosal arch, while a third joins
the opisthotic (paroccipital), and probably also the head of the
quadrate. It thus helps to form a small vacuity on the occipital
region, between the opisthotic, parietal, and squamosal —the
supraoccipital vacuity. It has nothing to do with the brain-
case, as Seeley thinks it has in the European forms.’

The occipital condyle forms something less than a hemi-

‘SEELEY, Dragons of the Air, London, 1901.

ON ALTE SIGUEL OF; NY CTODACTYLUS 525

sphere, smoothly rounded, and with a small pit in its center. It
seems to have looked downward and backward at an angle of
about 45°, though it is possible that its angle with the horizontal
plane of the skull may have been slightly greater. The basi-
occipital forms a rounded, somewhat concave and roughened
surface, clearly directed at a considerable angle backward. The
paroccipital process is a broad, flattened bar, directed outward,
backward, and somewhat upward to join the squamosal and
quadrate broadly. Above them there is a small vacuity, as
already described, while the outer extremity expands to join
closely the inferior posterior surface of the quadrate, reaching
nearly as far as their middle. The supraoccipital is directed
upward and backward, joining the parietal in forming the pos-
terior flattened prolongation of the skull. Their separation can-
not be made out, but the two together form a considerable
concavity looking downward and backward, with a slender
median crest — the occipital crest — along the outer two-thirds.

The quadrate is more narrow above, and is firmly united with
the opisthotic, squamosal, and ‘“‘ quadrato-jugal” above ; below
and to the outer side with the jugal; internally below with the
pterygoid. The articular surface looks more nearly downward.
The transverse axes of the two join each other at an angle of
about 20°. Each has a deep oblique trochlear groove running
from within outward and backward, with the stronger convexity
at the inner extremity. The basi-sphenoid is an elongated,
flattened, or slightly convex bone, somewhat wider in front
than behind, with concave sides, and situated in the same plane
with the pterygoids and palate. It joins the pterygoids at
each anterior angle broadly and flatly.

The pterygoid is firmly and indistinguishably united with
the quadrate at its lower inner extremity, close to the articular
surface, by a broad process which ascends slightly to the plane
of the palate. The two bones are separated from each other
throughout their entire length, or at most touch each other by
the slender tips of the anterior inner processes. Anteriorly
the bone divides into two unequal processes—a slender one,

526 S. W. WILLISTON

directed inwardly and anteriorly, closely approaching or touch-
ing its mate in the plane of the palate ; and a larger one, which
goes forward to beyond the middle of the narial vacuity, articu-
lating with the ectopterygoid, and distally, by a long suture,
with the palatine. Between the two there is an arrow-shaped
vacuity posteriorly, bounded behind by the basi-sphenoid, the
indentation for the head of the arrow formed by a small, thin
inner process: Dhis form jof the pterygoid) it) will be seen as
quite different from that figured and described by Woodward in
Rhamphorhynchus. The slender inner process does not turn
upward to meet a vomer, as Woodward thinks may be the case
in the skull described by him, and the processes which, in his
specimen, turn inward and upward are, in WVyctodactylus, closely
applied to the inner side of the palatines. The pterygoid does
not unite with the vomer in Wyctodactylus.

The ectopterygoids arethin, narrow bones, articulating broadly
with the outer margin of the pterygoid, and exteriorly with the
inner side of the slender posterior extremity of the maxilla. The
bone is directed forward and outward, its concave margins form-
ing the anterior and posterior borders of the pterygo-jugal and
posterior palatine vacuities.

The palatines are long, narrow, thin bones, articulating by a
long suture on the inner side with the pterygoid, and on the outer
side with the maxilla. Anteriorly the union with the maxilla or
vomers is indeterminable.

The relations of the maxilla can be made out only in its pos-
terior part. Its union with the jugal has already been indicated.
Posteriorly it sends up a thin, curved process to meet the nasal,
forming the posterior inferior boundary of the nares. Back of
the thickened portion of this process there is a thin plate reach-
ing back under the superior process of the jugal, the bottom of
an oval depression or excavation, which clearly corresponds to
the antorbital vacuity of the earlier pterodactyls. From this it
is evident that the antorbital vacuity is not united with the nares
in either this genus or Ornithostoma.

The vomer cannot be made out. If it forms a part of the

ONDE SIGOLE, OF NY CTODA CIVEOS 527

palatal surface, it is so indissolubly united with the maxille, pal-
atines, and premaxillary that no signs of its union can be detected.
One thing is certain, it does not separate the narial vacuities,
and in all probability it does not touch the pterygoid at all.

Mandibles—The mandibles are preserved nearly complete,
their tip only being wanting—the portion which led to the
detection of the specimen as it protruded from the chalk. It
was lying horizontally, with the under-surface uppermost, but has
been entirely removed from the matrix. It has suffered a slight
lateral compression to the left.

The cotylar cavity fits well on the condyles of the quadrate.
The projection back of the articulation is very slight, less so than
in Ornithostoma. The rami are slender, flattened from within
outward, and are broadest at about their middle, the upper thin
margin being gently convex from near the cotylus to as far for-
ward as the symphysis, which begins about fifty millimeters in
advance Of the anterior end of the external nares. From the
beginning of this symphysis, diminishing in width to the anterior
extremity, the under margin is convex, not carinated. About
twenty-five millimeters back of the true symphysis of the denta-
ries there is a symphysis of the thin bone which forms the floor
of the mouth. This forms a thin, deeply curved hind border
joining rami about midway between their upper and lower mar-
vins. The floor of the mouth from here onward seems to have
been, like the opposing palatine surface, smooth, and nearly
plane. The space between these surfaces, was, however, some-
what greater posteriorly than anteriorly, since the alveolar para-
pet is considerably deeper posteriorly. The concave posterior
margin of the floor is about twenty millimeters in front of the
anterior end of the internal narial vacuity. Anteriorly the mar-
gins of the mandibles are as in the upper jaws. In cross-section
through the symphysis the mandible was, I think, somewhat con-
cave or partly plane on the sides, with the under margin convex,
not convex throughout, as I have thought might be the case in
Ornithostoma. The mandible certainly could not have possessed
very great strength in seizing.

528 S. W. WILEISTON

Openings.—The external nares are large and ovate, looking _
outward and upward and forward. They are bounded above by
the premaxillary anteriorly, the nasal posteriorly, below by the
maxilla (and premaxillary anteriorly?), behind by the nasal
above and the maxilla below.

The orbits are longer than wide, looking outward and but
slightly upward and forward. ‘The superior emargination above
(doubtless roofed over in life by the integument) if not taken
into account would leave the orbit oval in shape, with the ante-
rior end rather broader than the posterior. It was bounded
above by the frontal, prefrontal, and lachrymal, by the lachrymal
and malar in front, the malar below and behind, and the post-
frontal in part behind. It had a ring of thin, large sclerotic
plates, which were preserved in displaced positions. The sep-
arate plates were not united by imbrication, as in the mosasaurs.

The supratemporal vacuity is broadly oval in shape, deep
anteriorly and shallow posteriorly, bounded on the inner side by
the parietal, on the outer side by the postfrontal and squamosal,
and behind by the margin of the squamosal and parietal process.
It opens anteriorly into the spheno-quadrate vacuity, and has
a narrow slit between the squamosal and opisthotic. The infra-
temporal vacuity is long and narrow, broader posteriorly. It
is bounded on the inner side and behind by the quadrate, exter-
nally and above by the jugal and the quadrato-jugal. The large
oval vacuity between the quadrate and the basi-sphenoid is
broadest at the middle, bounded in front by the quadrate process
of the pterygoid, behind by the paroccepital process. The pter-
ygo-jugal vacuity is irregular and elongated, bounded behind by
the articular end of the quadrate, on the outer side by the jugal
and maxilla, on the inner side by the pterygoid posteriorly and
the ectopterygoid anteriorly. The posterior palatine vacuity is
a small oval vacuity placed obliquely between the ectopterygoid
behind, and the palatine in front, barely touching, if at all, at
the extremities the pterygoid and the maxilla. The interpter-
ygoid vacuity is arrow-shaped, placed between the basi-sphenoid
posteriorly and the pterygoids on the sides. The narial vacuity

OND LAE SIGUEELNOFVNYVCTODA CLIVE OS 529

is a very large opening, oval in general shape, but with a trian-
gular projection behind made by the two inner processes of the
pterygoids, and a smaller triangular process in front extending
back from the roof of the mouth. It was not divided in the
middle and was doubtless covered over for the most part by
membrane pierced by the internal nares.

The skull of Myctodactylus, while presenting very great
resemblances to that of Ovnithostoma (fteranodon) differs in
important characters. Its quadrate is much shorter, termi-
nating below in the condyle for the mandible nearly below the
middle of the orbit, while in Ornithostoma the articulation is in
advance of the orbit. The orbit and supratemporal vacuities
are proportionally larger, and there is no crest, not even a
rudimentary one. The bar between the orbit and the nares is
narrower, the frontal bone less broad relatively in Ornzthostoma.

In Nyctodactylus we have a less highly specialized type of
pterodactyl than is seen in Ornithostoma as is evidenced by the
absence of the crest, the plane frontal bone, and the less oblique
position of the quadrate. On the other hand the genus has the
same type of skull, and toothless jaws, and the general propor-
tions of the skeleton throughout arethe same. I believe that the
resemblances between the two genera are of much greater
importance than the differenees. These differences rest in the
non-articulation of the scapula with the supraneural plate of the
vertebra, in the absence of crest, and the position of the quadrate
especially. Both forms are toothless. If much importance is
given to the scapular articulation the two genera must go into
different families at least — Ornithostoma among the Ornithochei-
ride, and Wyctodactylus with the Pterodactylide. But sucha sev-
erance is, I, believe, questionable. It is a rather curious instance
of parallel growth that has brought about the toothless forms
of Ornithostoma from the toothed Ornithocheirus, and the toothless
Nyctodactylus from the toothed Prerodactylus ; possibly this has
been the phylogeny. On the other hand, it would also seem
possible that the same toothless form has been ancestral to both

530 S. W. WILLISTON

Nyctodactylus, the lesser, and Ornithostoma, the more, specialized
type.

If my interpretation of the elements of the skull of Wycto-
dactylus is correct, I fail to see any marked avian resemblances in
it. The only character that might be considered as such is the
undivided narial vacuity, aside from the lightness of the bones
and their closely united sutures, both adaptive characters. The
supratemporal arch is purely reptilian, as is the presence of the
ectopterygoid, etc. The union of the premaxillary with the
frontal is found in the plesiosaurs. A further discussion of this
question I hope to present in a future paper, in which I shall give
amore detailed description of the remainder of the skeleton,
with illustrations.

Iam convinced that the arrangement of the ribs back of the
sternum, not only in Wyctodactylus, but in other pterodactyls, is
different from what has been generally supposed and figured.
They extended outward in the plane of the wing membrane, which
they served in a measure to support, and did not enclose the
ventral cavity. From a recent examination of the type of
Khamphorrhynchus phyllurus Marsh, I am satisfied that this was
their function in that animal, and I doubt not it was also in all
other pterodactyls. This difference in function accounts satis-
factorily for the marked contrast in slenderness between the
antericr and posterior ribs, as also for the slight curvature of
the latter.

The hyoid bones of the present specimen are quite like those
of other pterodactyls— long, slender rods. They are shown in
part in Plate]. Atthe back of the skull in the same plate is
also seen a small triangular or V-shaped bone, which I take to
be the proatlas.

It will be of interest to note that in the matrix underlying
the skull occurred several specimens of the Cirriped Stramentum,
together with a vivid impression of a Lepas-like barnacle having
a slender, flexible, naked peduncle about 4o™ in length. The
capitulum seems to have been enclosed either in chitinous or
membranous plates, evidences of which are present, and impres-

jours GEOL, Voli Xs Noss:

jours Grove Viol, OX Noy 5: Plate II.

FTG a Te KinGsn2.

ty
wae

an

sl inky

Pes, anti i)
v i a lane ty
a Vie zah Stark

pace a

ape pe aie ae
hee et
: Neth ee
ups Dee

ONDE ESIGULE OF NVCTODAGCTYLUGS 531

sions of the cirri are plainly present. Other impressions of
phyllopods or ostracods are also abundant in the matrix.

S. W. WILLISTON.

UNIVERSITY OF KANSAS,
Lawrence, April, 1902.

EXPLANATION OF PLATES.

PLATE I. Under surface of posterior part of skull, lying in the matrix
slightly enlarged. Near the anterior end are seen the two hyoid bones, and
in the occipital region, the small, triangular proatlas.

PLATE II. Fig. 1, Under view of skull ; Fig. 2, upper view of same, both
a little more than half natural size ; 40, basi-occipital ;. ef, ectopterygoid; /7,
frontal; zJv, interpterygoid vacuity; z¢v, infratemporal vacuity; 7, jugal; Z,
lachrymal; mx, maxilla; 7a, nares; was, nasal; oc, occipital condyle; Za,
parietal; Aa/, palatine; A/, prefrontal ; Zof, postfrontal ; Apv, posterior pala-
tine vacuity; fx, premaxillary; g, quadrate; gj, quadrato-jugal; sg, squa-
mosal; z, pterygo-jugal vacuity.

CROTALOCRINUS CORA (HALL).

AmonG the internal casts of crinoids which occur commonly
in the dolomitic Niagaran limestone of northeastern Illinois and
southeastern Wisconsin, is a species which was first described by
Hall in 1868 under the name Cyathocrinus cora. The cotypes of
this species, two in number, are recorded from Racine, Wis., and
are now preserved in the collections of the American Museum of
Natural History in New York. Specimens very much larger
than these types frequently occur in the dolomite near Chicago,
but they resemble them in all essential characters and all are
undoubtedly specifically identical.

The body of this crinoid is more or less subglobular in form,
with a large column, and has been but rarely found preserved in
any condition other than as internal casts. The plates compos-
ing the body are arranged essentially as in members of the
genus Cyathocrinus. There are five large underbasals which
support five large basals, four of which are hexagonal in outline,
the fifth being heptagonal by reason of the truncation of the
distal angle for the support of a single subquadrangular anal
plate. The radials are large, wider than high, and are in con-
tact laterally, except the two posterior ones, which are separated
by the large anal plate. Until the discovery of the specimen
illustrated in the present paper, the plates above the radials
have never been observed, or at least they have never been
recognized as belonging to this species.

In I900 a single imperfect radial plate of a crinoid, with
brachial plates attached, was recognized by Weller as represent-
ing the genus Crotalocrinus, and was described under the name
C. americanus. This specimen is a natural mold of the exterior,
and was found associated with Cyathocrinus cora, but there was no
evidence by which the two forms could be correlated.

There has recently come into the possession of the Walker

532

CROTALOCRINUS CORA (HALL) 533

Museum, through the generosity of Mr. G. F. Harris, of Chicago,
a specimen of the so-called Cyathocrinus cora, upon which nearly
all of the plates of the calyx are preserved, as well as some of
the brachial plates. This specimen demonstrates that the type
of Crotalocrinus americanus is nothing more than a radial plate
with brachials attached of the so-called Cyathocrinus cora, and
that the latter species must be transferred from the genus
Cyathocrinus into Crotalocrinus. This remarkable genus of
crinoids, therefore, which has hitherto been considered as such a
rarity in the Niagaran fauna of America, proves to be one of the
common members of the fauna of this age in Illinois and Wis-
consin. It is of especial interest because it is one of those
peculiar forms which relates the Niagaran faunas of the interior
of North America, so closely with the faunas of Gotland and
England.

This American species of Crotalocrinus is most closely allied
to C. rugosus Miller, var., from Gotland, as illustrated by
Angelin.* Both have the same subglobular form of calyx, con-
stricted at the top of the radial plates, and both have the same
style of surface markings, consisting of shallow furrows which
cross the sutures dividing the plates at right angles. In the
American specimens these markings are so pronounced as to be
usually discernable even on internal casts. The arrangement of
the fixed brachial plates in the two species, however, is quite
different, if Angelin’s illustrations can be depended upon.

In the arm plates of Crotalocrinus the axial canals are a con-
spicuous feature. In converging toward the base of the arm,
these canals unite into a single trunk which pierces the radial
plate. The axial canals are well shown in both the specimens
preserving the brachial plates, and the casts of the canals are
frequently more or less well preserved in the internal casts of
the species as it is ordinarily preserved.

The species which has been described as Cyathocrinus van-
hornet S. A. M. is probably another member of the genus Crotalo-
crinus. It is known only from internal casts, but these speci-

'Iconog. Crin., Plate 17, Figs. 8, 8a.

534 STUART WELLER

mens usually preserve the cast of the basal portion of the axial
canal where it enters the radial plate. The species differs from
C. cora, especially in the strong constriction of the calyx at its
middle line.

In the light of the observations here recorded, the synonomy
of Crotalocrinus cora will stand as follows :

Crotalocrinus cora (Hall).

1868. Cyathocrinus cora Hall, Twentieth Rep. N. Y. State Cab. Nat.
FiitSt., Pe 32ds bl. lo kigs: 13,414.

1870. Cyathocrinus cora Hall Twentieth Rep. N. VY. State Cab. Nat.
Hest. (Rev.3sd.), ps. 300 b 1X1, Pigsiet3, 14:

1879. Cyathocrinus cora W. & S., Rev. Paleocr., Pt. I, p. 85.

1881. Cyathocrinus cora 5S, A. M., Jour. Cinn. Soc. Nat. Hist., Vol. IV,
p.U71.

1g00. Cyathocrinus cora Weller, Bull. No. 4 Nat. Hist. Surv., Chicago
Acad Sct.,.p. 62, PIXIV, Pigs; 6—10:

1900. Crotalocrinus americanus Weller, Bull. No. 4, Nat. Hist. Surv.,
Chicavo’ Acad. Scz., p. 14:3, Pl. 2X1V;, Pigs.

EXPLANATION OF PEATE.

Fig. 1.—A nearly perfect specimen of the calyx of Cvrotalocrinus cora
showing the plates of the calyx with some of the brachial plates.

Fig. 2.—An internal cast of Crotalocrinus cora, showing the broken bases
of the casts of the axial canals where they pierce the radial plates.

Fic. 3.—An outline of the specimen which was made the type of Cvofa/o-
crinus americanus, showing the arrangement of the fixed brachial plates in
Crotalocrinus cora.

Figs. 4-5.—The casts of two radial plates of Crotalocrinus cora, preserv-
ing the internal casts of the axial canals.

STUART WELLER.

Plate ILI.

jour Gro, Vol; x] Nows:

o °
> z a
' \ .
be ‘
4 .
€
* “
.
\
:
F :
\
\
1
} .
A ‘
wa x
3 ‘

m

a 3

i a

THE CARBONIFEROUS FISH-FAUNA OF MAZON
CREP TMEEINOIS:

OF the thousands of fossiliferous ironstone nodules of Coal
Measure age, occurring at Mazon Creek, near Morris, in Grundy
county, Illinois, only a small percentage afford indications of
vertebrate remains, and these consist principally of detached
fish-scales. Occasionally, however, complete individuals of
fossil fishes, and still more rarely, amphibian skeletons have
been brought to light, but all told the number of even tol-
erably perfect specimens preserved in different museums is very
insignificant. Probably the two finest series of Mazon Creek
nodules ever brought together are the Lacoe collection, belong-
ing to the United States National Museum in Washington, and
the Strong collection, purchased by the late Professor Marsh for
the Peabody Museum, at Yale College. Shortly before the
decease of Professor Marsh, nearly all of the fossil fishes in the
Strong collection were placed by that gentleman in the hands of
the writer for study and description; and more recently some
further material has been loaned for the same purpose by Pro-
fessor C. E. Beecher, to whom grateful acknowledgments are due.

Mazon Creek fish-scales have been exhaustively studied by
E. D. Cope’? and O. P. Hay,? and the latter has also described a
nearly perfect example of a Palwoniscid fish, named by him
Elonichthys hypsilepis. A few other Paleoniscids and Platysomids
have been described by Cope’ and by Newberry and Worthen ;*
and two Acanthodian species have recently been made known by
the present writer.’ These citations complete the literature
references on Mazon Creek fishes. In the following paragraphs

t Proc. Amer. Phil. Soc., Vol. XXXVI (1897), pp. 71-82.

2 [bid., Vol. XX XIX (1900), pp. 96-120.

3 Proc. U. S. Nat. Museum, Vol. XIV (1891), p. 462.

4 Pal. Illinois, Vol. 11 (1866), and Vol. IV (1870).

5 Bull. Mus. Comp. Zool., Vol. XX XIX (1902), pp. 93, 94.

535

536 C. R. EASTMAN

©

brief descriptions are given of two species of Acanthodes, and one
each of Cwlacanthus and Elonichthys, with a list of the known ver-
tebrate fauna occurring at this locality.

GENUS ACANTHODES, AGASSIZ.

Representatives of the Acanthodii are extremely rare in the
Paleozoic rocks of North America. If we neglect the detached
spines of Machera-
canthus, and the in-
determinable mass
of scales described
by jai Clarke as
Acanthodes pristis,*
American Acan-
thodians are limited
to but three species
of Acanthodes and

Fic. 1a.—Shag-
reen granules of 4.
one of Mesacanthus. marshi. X 4,

Of these Acan-

thodesconcinnus Whiteaves and Mesa-
canthus affinis (Whiteaves) occur in
the Upper Devonian of Scaumenac
Bay, Canada, and the recently de-
scribed Acanthodes marshiand A
beecherit are from the Mazon Creek
locality, in Illinois.

Acanthodes marshi Eastman.
Fic. 1.—Acanthodes marsht

Eastm. Coal-measures, Mazon ;
Creek, Ill. Pectoral fin with being one of the largest, as, on the

associated actiontrichia and fin- other hand, A. dcecheri is one of the

Spe smallest known Acanthodians. In A.

marsha, not only are the shagreen granules much coarser than

those of A. drvonni and A. wardi, which are the largest of

European species, but the fin-spines are considerably longer

and stouter, averaging about 9° long, and from .5 to .8°™
t Bull. U. S. Geol. Surv., No. 16 (1885), p. 42.

This species is remarkable for

CARBONIFEROUS FISH-FAUNA 5137.

wide. In Fig. 1 is shown a very interesting pectoral fin pre-
served in counterpart, and retaining the actinotrichia in natural
association with the spine. The fibrous rays are quite long and
numerous as compared with those of other species, and extend
well up toward the point of insertion of the spine. There is no
trace here, unfortunately, of a basal cartilage abutting against the
proximal end of the spine, nor does this specimen display any
of the dermal granules with which the fin-membrane was
stiffened, although such are exhibited by a smaller specimen
belonging to the Yale Museum. The scales of A. marshi are in

eoeee

Fic. 2.—Acanthodes beecheri Eastm. Coal-measures, Mazon Creek, Ill. Resto-
ration showing outline of body and position of fins. > ?.

the form of shagreen granules, averaging about one square milli-
meter in size, smooth and polished externally, and gently convex
or rounded on both the outer and attached surfaces. (Fig. Ia).
The internal structure consists of fine layers of dentine arranged
in quadrate fashion about a small central pulp-cavity. The
best account of the microscopic structure of Acanthodian and
Thelodus-like scales is that given by Rohon about nine years
ago."
Acanthodes beecheri Eastman.

Description. —A very small species, attaining an extreme length of about
5.5°™. Body elongated and slender, the maximum depth being contained
about nine times in the total length. Pectoral spines not much stouter or
longer than the others; pelvic fins small, slightly nearer the pectorals,than
the anal; anal fin slightly larger than the dorsal, which is placed immediately
behind. Length of dorsal and anal spines greater than maximum depth of
trunk. Caudal lobe remarkably elongate. Scales extremely minute,

This species is represented by two nearly complete individ-
uals preserved in counterpart and belonging to the Yale Museum,
* Mem. Acad. Imp. Sci. St. Petersbram, Vol. XLI (1893), No. 5, p. 22.

538 GYR ELASTUTAN,

neither of which, however, exhibits the caudal region satisfac-
torily, nor are the heads well preserved. Only the dorsal and
anal fin-spines are displayed by the larger specimen; but in the
smaller all the fin-spines are preserved, although the dorsal is
slightly displaced and the distal ends of the pectorals are want-
ing. The accompanying
figure, based on both speci-
mens, is of composite na-
ture, and represents “the
general outline and propor-
tions of the fins, the re-

Fic.—3. Celacanthus extguus, sp. nov. stored parts being indi-
Coal-measures, Mazon Creek, Ill. Complete

individual, lacking posterior dorsal and anal cated in dotted lines.

fins. X#.
GENUS CCELACANTHUS, AGASSIZ.

J. S. Newberry records having received from Mazon Creek
‘‘a single specimen each of Eurylepis and Celacanthus, probably
not distinct from those found at Linton, Ohio.”” No examples
of the former genus have come under the writer’s observation,
but ornamented scales and head-plates referable to Celacanthus
sometimes occur in Mazon Creek nodules, and very rarely there
are found complete fishes of small size, evidently quite distinct
from other described species. In most specimens the posterior
dorsal, anal, and pectoral fins are lacking, and it seemed at first
sight as if the second dorsal had become lost through specializa-
tion. One individual, however, shows it very distinctly, and the
absence of this and the anal in the remaining examples is to be
attributed to faulty preservation.

Celacanthus exiguus, sp. nov.

Description.—A small species, attaining a maximum length of about
4.5. Trunk narrow and elongated, the head occupying about one-fourth of
the total length. First dorsal consisting of relatively few stout rays, and sit-
uated slightly in advance of the pelvic pair; second dorsal midway between
the first dorsal and principal caudal; the latter comprising nine stout rays
above and below. [Scale-structure and ornamentation of head-bones not
observed. |

CARBONIFEROUS FISH-FAUNA 539

This species is represented by ten specimens in the Yale, and
one in the Harvard Museum, most of them being only about
BiG Moyniog,
having a narrow, gradually tapering body, which terminates in

and very deficient in preservation. They agree in

an equilobate caudal fin, with indications that the axis was pro-
longed into a supplementary
caudal. The first dorsal and
caudal, owing to their
stronger attachment, are pres-
ent in nearly all specimens,

but the remaining fins have
in most cases become de- Fic. 4.—Llonichthys perpennatus, sp.
nov. Coal-measures, Mazon Creek, Ill.

stroyed. The first dorsal has

; Complete individual, the distal portions of
usually seven or eight stout

median fins not fullyshown. X?.

rays, and is situated near the

middle of the trunk. Ten long, hollow rays are to be counted
in the single specimen displaying the posterior dorsal, and nine
above and below in the symmetrical caudal. The neural and
hemal spines are very long in the abdominal and caudal
regions. The ossifications of the axial skeleton are continued
nearly to the termination of the principal caudal. The squama-
tion must have been exceedingly delicate, as no indications of
scales are to be observed in any of the specimens, nor do any of

them have the cranial elements satisfactorily preserved.

GENUS ELONICHTHYS, GIEBEL.

Two closely related species are already known from Mazon
Creek, 4. peltigerus Newberry, and EL. hypsilepis Hay. A study
of the type-specimen of Newberry and Worthen’s so-called
“ Amblypterus macropterus,’
leaves no doubt that this is only a mutilated individual of Z.

d

now preserved in the Yale museum,

peltigerus. The type of the following new species is preserved in
the Museum of Comparative Zodlogy.
Llonichthys perpennatus, sp. nov.

Description.—A very small species, having a total length of about 2.5°™,
of which the head occupies a little less than one-fourth. Fins extremely well

540 C. R. EASTMAN

developed, the pectorals unusually long, and anal much extended; fulcra
minute. Scales relatively small, obliquely striated; dorsal ridge-scales
enlarged.

Only one individual is at present known of this interesting
little form, which is shown in Fig. 4. The head is poorly pre-
served, and the distal extremities of nearly all the fins are either
broken away or obscured by matrix. Nevertheless, sufficient
characters remain for the recognition of this as a distinct spe-
cies of Elonichthys, its chief peculiarity consisting in the remark-
able development of all the fins. The pectorals are fully one-
fourth the total length, and the anal has a more extended base-
line than in any other species of the genus. The dorsal appears
to have been high and acuminate, but is largely concealed by
matrix. The caudal is also unfavorably exposed, and flexed out
parallel with the main axis; but it is plain that the upper lobe
was much prolonged, and covered with very large, striated ridge-
scales. The dorsal-fin rays appear to have been widely jointed ;
the articulations of the other fins are not clearly recognizable.
The dermal rays of the anal and lower lobe of the caudal are
directly supported by the large haemal spines, which are firmly
united with their arches. The squamation is nowhere well pre-
served, but is best indicated in the anterior part of the trunk.
The cranial structure does not admit of particular description.
Appearances suggest that the specimen here described is an
immature individual, differing however, from other known
species.

LIST OF CARBONIFEROUS VERTEBRATES OCCURRING AT MAZON
CREEK, ILLINOIS.
ELASMOBRANCHII.

1. Pleuracanthus (Diplodus) compressus Newb. (Occurs also in Ohio and
Indiana.)

2. Pleuracanthus (Diplodus) latus Newb. (Occurs also in Ohio and
Indiana.)

- 3. Pleuracanthus (Diplodus) lucast Hay.

‘4. Acanthodes beecheri Eastm.

5. Acanthodes marshi Eastm.

6. Campodus scitulus (St. J. and W.).

CARBONIFEROUS FISH-FAUNA 541

DIPNOI,

7. Ctenodus sp. indes,

8. Sagenodus foliatus Cope.

g. Sagenodus lacovianus Cope."
10. Sagenodus occidentalis (Newb. and W.).* (Occurs also at Linton,

Ohio.)

11. Sagenodus guadratus (Newb.).* (Occurs also at Linton, Ohio.)
12. Sagenodus guincunciatus Cope."

13. Sagenodus reticulatus (Newb. and W.)."
14. Sagenodus textilis Hay."
tFounded on scales.

CROSSOPTERYGII.

15. Rhizodopsts (?) mazonius Hay.t
16, Celacanthus exiguus nobis.
17. Celacanthus robustus Newb.* (Occurs also at Linton, Ohio.)

ACTINOPTERYGII.
18. Eurylepis sp.indet. (fide J. S. Newberry).
19. Rhadinichthys gracilis (Newb. and W.).
20. Elonichthys hypstlepis Hay.
21. Elonichthys peltigerus Newb.?— (Occurs also at Linton, Ohio.)
22. Elonichthys perpennatus nobis.
23. Platysomus circularis Newb. and W.
24. Platysomus lacovianus Cope.
25. Platysomus orbicularis Newb. and W.

AMPHIBIA,

26. Amphibamus grandiceps Cope.

t Founded on scales.
? Including the so-called “* Amélyplerus macropterus’? Newb. and Worthen.

C. R. Eastman.

REVIEWS.

SUMMARIES OF THE LIGERATURE, OF SPRUCTURAL
MEMEO LAN Sse tiles

EDWIN C. ECKEL.

BERKEY, C. P. Origin and Distribution of Minnesota Clays. Amer. Geol-
ogist, Vol. XXIX, pp. 171-177, March, 1902.

Residual clays, derived either from feldspathic rocks or from limestones, are of
slight importance in Minnesota. The larger clay working establishments of the state
are using shales, stream deposits, or glacial lake clays; the smaller brick plants of
local importance use till or loess.

The shales of most importance are those of the Ordovician and Cretaceous. The
Ordovician shales are found only in the southeastern portion of the state. Most beds
are too calcareous for use, but one company in Minneapolis utilizes Ordovician shales
in the manufacture of an exceptionally fine line of front and pressed brick. The
Cretaceous shales are used in the manufacture of stoneware at Red Wing.

The clays may be divided as to origin into glacial till, glacial lake clays, glacial
stream deposits, recent alluvial deposits, and wind deposits.

Locally the till includes clay deposits of workable size. These clays differ in
character according to the drift in which they are inclosed. The “ gray drift” which
has been brought by ice movements from the north and northwest carried fine-grained,
calcareous, light-burning clays, though in places weathering may have removed
enough lime to give red-burning clays. The “‘red drift” brought in from the north
and northwest carries coarser grained clays with an excess of iron, which consequently
burn red.

Glacial lake clays, laid down in quiet water in interglacial periods, are confined
to the eastern border of the state, where extensive deposits occur and are worked on
a large scale.

Glacial stream deposits include the river silts deposited during the withdrawal of
the ice. Clays of this type are worked on the Mississippi river between Minneapolis
and Little Falls, and on the Missouri river between Shakopee and New Ulm. These
clays are obtained from the terraces bordering the present river channels, and burn
cream or gray. They are the most important of the Minnesota clays, large plants
using them being located at Chaska and Minneapolis.

Recent alluvial deposits occur in the same areas as the last class and in many
cases cannot be differentiated from them.

Most of the smaller brick plants of the state work on material obtained from loess
deposits. The clayey loams of this class are widely distributed, but are nowhere of
great thickness or value.

542

REVIEWS 543

BLATCHLEY, W. S., AND ASHLEY, G. H. The Lakes of Northern Indiana
and thetr Associated Marl Deposits. Twenty-fifth Ann. Rept. Indiana
Dept. Geology and Natural Resources. Pp. 31-321, Pls. 1 and 6-12,
Figs. I-70, 1901.

The origin and uses of marl are first discussed, issue being taken with C. A. Davis
on some points connected with the importance of Characae in the formation of marl
deposits. The marl deposits of the state are then discussed separately in great detail.

Marl deposits of sufficient size to justify the erection of cement plants occur in Indi-
ana only in the three northern tiers of counties. Areally, the largest of these deposits
is in Lake Wawasee, which contains about 1.700 acres, while the maximum thickness
(45 feet) is reported from Turkey Lake, Lagrange co. A deposit of marl equal to one
covering 160 acres, ten feet thick, will supply for thirty years a plant with a capacity
of 500 barrels per day’ Deposits of such size are termed “ workable deposits” in the
present report. Thirty-two such deposits were found and are described and mapped
in detail. A number of other deposits are described which, though of sufficient size,
have the larger part of their area covered by ten feet or more of water and are there-
fore not at present workable. Improved appliances for raising marl from beneath
such depths of water would render these deposits available.

BLATCHLEY, W. S. Portland Cement. Twenty-fifth Ann. Rept. Indiana
Dept. Geology and Natural Resources. Pp. 1-30, Pls. 2-5, I1gol.

A summary of the history, uses, composition, manufacture, and testing of Portland
cement is followed by a brief history of the industry in Indiana.

Portland cement was first manufactured in Indiana at South Bend in 1877, from
marl and clay burned in dome kilns. Operated with varying fortune, this plant was
shut down finally in 1898. Two plants, also using marl and clay, but burning the
mixture in rotary kilns, commenced operations in 1900, and at the time of report three
additional plants were in prospect.

CUMMINGS, UrRIAH. [Production of | American Rock Cement |in the U.S.
during 1899|. Twenty-first Ann. Rept. U.S. Geol. Surv., Pt. VI con-
tinued, pp. 407-411, Igol.

Résumé of the condition of the American natural cement industry during 1899,
with statistics of production. Analyses and tests are quoted of a Portland cement
manufactured at Chattanooga, Tenn., by burning a natural rock without admixture.
EcKEL, E.C. The Portland Cement Industry in New York. Engineering

News, Vol. XLV, pp. 365-367, Igol.

Résumé of early history of Portland cement manufacture in New York, with
descriptions of the six plants operating in 1900, and notes on the technology.

Until recently most New York plants used a mixture of marl and clay, burned in
dome kilns. At present, however, the use of rotary kilns, operating on mixtures of
hard limestone and clay, is increasing rapidly, half the plants in operation in 1900
being of this type.

Slag Cement Manufacture in Alabama. Eng. News, Vol. XLVII, pp.

I-62, Igo2.

Description of two slag cement plants operating in Alabama, with notes on

technology of slag cement in general.

544 REVIEWS

The Classification of the Crystalline Cements. Amer. Geologist, Vol.
XXIX, pp. 146-154, March, 1902. Reprinted in Cement, Vol. III,
pp. 109-114, May, Ig02. Reprinted in part in Engineering News,
Vol. XLEViIep. 354) May 1, 1902:

A classification of cementing materials, with notes on raw materials, technology,.
and the properties of the various products. The grouping offered is as follows:

I. Simple cements; including those materials which are produced by the expulsion
of a liquid or gas from the raw material; and whose set is due to the simple reab-
sorption of the same liquid or gas and a reassumption of original composition.

Ia. Hydrate cements; set due to reabsorption of water. Plaster-of-Paris, cement
plasters, Keene’s cement, Parian cement, etc.

14. Carbonate cements, set due to reabsorption of carbon dioxide. Limes, mag-
nesian limes, etc.

Il. Complex cements; including those cementing materials whose set is due to the
formation of new compounds during manufacture or use.

Ila. Silicate cements, set due to the formation of silicates. Hydraulic limes, natural
cements, Portland cement, pozzuolanic cements.

11d. Oxychloride cements, set due to the formation of oxychlorides. Sorel stone,
etc.

FREAR, WILLIAM, The Use of Lime upon Pennsylvania Soils. Bulletin 61,

Pa. Dept. Agriculture. 170 pp., Igoo.

Principally a discussion of lime in its relation to agriculture, but contains also very
satisfactory accounts of limestones (in general) and lime-burning, with 273 analyses
of Pennsylvania limestones, mostly compiled from reports of the Second Geological
Survey of that state.

Harris, G. D., and VEatcH, A.C. General Geology (of Louisiana). Geo.
logical Survey of Louisiana, Report for 1899. Pp. 55-138, Pls. 1-11, Figs-
2-5, geological maps.

Report on the stratigraphic and economic geology of the state. The matter of
clays, limestones, sandstones, and gravels (pp. 127-132) is here summarized.

Good brick clays are common in the alluvium and yellow loam, and are also
found at several places in the hill-lands. The Eocene clays commonly lack plasticity,
though some beds occur which will make a fair quality of earthenware. Good potter’s
clay occurs in the Lignitic, near Robeline, where it has been utilized. The clays of
the Grand Gulf hills seem to be more promising than any others in the state, good
exposures occurring in Catahoula and Vernon parishes.

The sandstones of the state are of two classes; the ferruginous sandstones of the
Eocene and Lafayette, and the siliceous sandstones of the Grand Gulf. The former
are widely distributed, but are unimportant as structural materials. The siliceous
Grand Gulf sandstones are of greater value, and have been used for jetty work and
railroad ballast.

Limestones of Cretaceous age outcrop at Winnfield, Coochie Brake, and Bayou
Chicot. The first is a pure limestone, which may be used for lime, but not for building
stone. The other two are of greater value for structural purposes. Limestone con-
cretions occurring in the Tertiary are of local importance for lime or road metal.

REVIEWS 545

The gravels of the Lafayette are rather extensively used as road metal and
railroad ballast.

HILLEBRAND, W.F. Some Principles and Methods of Rock Analysis. Bull.
176, U. S: Geol. Surv. 8vo. Pp. 115, Figs. 15, 1900.

Detailed discussion of the methods to be followed in analyses of the silicate rocks.
Stress is laid on the importance of complete and thorough analyses, and the chemist
is warned against neglect to determine elements (such as strontium, barium, vanadium,
etc.), often disregarded as unimportant.

Hopkins, T.C. Clays and Clay Industries of Pennsylvania. 111, Clays of
the Great Valley and South Mountain Areas. Appendix to the Ann.
Rept. Pa. State College for 1889-1900. 8vo, pp. 45, Igoo0.

Over half the paper is taken up with a discussion of the white clays of the Great
Valley; the remainder containing short chapters on respectively the red brick, paving
brick, and tile industries; with a brief account of the economic products, other than
clays, of southeastern Pennsylvania.

The most valuable clay deposits in southeastern Pennsylvania are those occurring
in the Cambro-Silurian areas of the Great Valley and South Mountains. These clays
are typically white ; commonly bluish-white to gray on fresh exposures, but soon weath-
ering pure white. Closely associated with the white clays are others stained more or
less by iron oxide, and being therefore yellow to brown in color. The white clays are
invariably high in silica, and low in alumina. Their iron content is also low, but the
alkalies are always too high to give refractory material. Genetically, the white clays
are direct decomposition products of light-colored hydromica slates which occur
interbedded with other Cambro-Ordivician rocks. In certain openings the clays are
shown grading into the undecomposed slates. They were originally intercalated beds
in a series of quartzites, limestones, etc.; but on the weathering of the rocks the clay
derived from the disintegrated slates crept down hillsides, tilled cavities in the cther
rocks, etc.—so that now the deposits are somewhat irregular, lenticular masses of
varying extent. At present the clays are largely used in the manufacture of paper,
and to a less extent for white tiles, and white and enameled brick.

Red building brick is made, in the Great Valley, from residual limestone clays,
residual shale clays, and alluvial deposits. Of these classes the first is the most
important. The clays residual from limestone make good brick, though often carrying
numerous fragments of limestone, quartz, etc.

Vitrified brick is manufactured at six plants. Three of these use Triassic shale ;
one, Hudson River shale ; while the remaining two use shales from the Cambro-Ordo-
vician series.

Ornamental brick and tile are manufactured at several plants, various materials
being in use, including clays, slates, etc.

IHLSENG, MAGNus C. The Road-Making Materials of Pennsylvania. Bul-
letin 69, Pa. Dept. Agriculture. Pp. 104, Figs. 1-17, relief map, col-
ored geologic map, Igoo.

General discussion of road materials, their qualities, and testing; influence of
topography on road construction; methods of constructing and repairing roads, etc.

546 REVIEWS

Scattered notes on various Pennsylvania road materials, with a section on the distri-
bution, by counties, of such materials.

Lewis, F.H. Aydraulic Cement Industry in the United States in 1899.
Mineral Industry, Vol. VIII, pp. 84, 85, Igoo.

Résumé of progress in the American cement industry during 1899.

Cement Industry in the United States in r900. Mineral Industry, Vol. IX,
pp. 77-82, gol.
Résumé of the conditions of the American Portland cement industry during 1900—
a year marked by overproduction and low prices.

Lewis, F. H., NEwpBerry, S. B., and others. Zhe Cement Industry. 8vo.
Pp. 235, Figs. 152. New York, Igoo.

A collection of articles reprinted from issues of the Engineering Record. Detailed
descriptions of a number of American cement plants, with sketches of the condition of
the cement industry in various European countries; and separate chapters on the gen-
eral technology of Portland cement (NEWBERRY) and the rotary kiln process (LEWIS).

MERRILL, G. P, Guide to the Study of the Collections in the Section of
of Applied Geology, the Non-metallic Minerals. Rept. U.5. Nat. Mus.
for 1899. Pp. 155-483, Pls. 1-30, Igol.

Comment.—The material contained in this handbook cannot well be summarized.

The sections of interest in the present connection are those on quartz (p. 215); flint

(216); limestones, mortars, and cements (pp. 264-270); dolomite (p. 274); magnesite

(p. 275); feldspars (p. 281); clays (pp. 325-353); gypsum (pp. 406-411); and road-

making materials (p. 482). The treatment of these subjects is, in general, excellent,

that of clays being eminently so. The discussion of the cements, particularly Port-
land, is less satisfactory.

NEWBERRY, S. B. [Production of | Portland Cement |in the United States
during 1899]. Twenty-first Ann. Rept. U. S. Geol. Surv., Pt. VI con-
tinued. Pp. 393-406, Igol.

Résumé of the conditlon of the American Portland cement industry during 1899,
with statistics of production. The method of calculating the proportions of the ingre-
dients in a Portland cement mixture is discussed and exemplified.

Ries, HEINRICH. Clays of New York. Bulletin 35, New York State
Museum. 8vo, pp. 455, Pls. 140, map, Igoo.

Detailed discussion of the origin, properties, testing and uses of clays, and manu-
facture of clay products; with descriptions of the clay deposits of New York and of
the industries based on them. The section on the geologic distribution of clays in
New York (pp. 275-311) is here summarized, together with the sections on shales
(pp. 825-841) and feldspar and quartz (pp. 841-844).

Small deposits of residual clay, of little or no economic impor:tance, have been
found at various points in Dutchess county.

The sedimentary clays are numerous and important, representing three geologic
periods — Quaternary, Tertiary and Cretaceous. The latter two occur only on Long
and Staten Islands, all the clays of the mainland being Quaternary.

REVIEWS 547

Local basin-shaped deposits of Quaternary clays occur at many points in the
central, western and southwestern portions of the state. These deposits are doubtless
on the sites of former ponds, formed commonly by the damming of valleys, and
later filled with the sediment of streams from the retreating ice sheet. A number of
these deposits are of economic importance, and some are now worked.

The most important and extensive clay deposits in the state, however, are those
in the Hudson valley. The Quaternary deposits here are of two types (1) estuary
deposits of fine stratified sand and blue and yellow clays; (2) cross-bedded delta
deposits of coarser material. The clay is usually blue, weathered yellow where
exposed. Itis markedly stratified horizontally, the layers of clay being separated by
extremely thin laminz of sand. At many localities the clay is overlain by the delta
deposits of rivers tributary tothe Hudson. The Quaternary history of the region is
summed up as follows: During the retreat of the ice sheet from the Hudson valley
the glacial streams deposited as kames a great amount of ground up material, princi-
pally shale. Subsequent to the retreat there was a depression of the land amounting to
80 feet at New York city and 360 feet near Schenectady. During this depression a
great amount of plastic clay was deposited, produced by glacial attrition of shales and
limestones. The upper portion of the clay is more siliceous, and it is overlain by an
extensive deposit of sand, indicating a change in the nature of the material washed
into the estuary. During the period of submergence much of the siliceous matter
washed into the estuary was deposited at the mouths of tributary streams to form
deltas.

The clays of the Champlain valley are estuary deposits of the same age as the
Hudson river clays. They underlie terraces bordering Lake Champlain, and now
standing at an elevation of 393’ A. T. Extension erosion has removed much of the
clays and sands, and it is only at sheltered points that the terraces are now prominent.
The clays are worked for brick at Plattsburg and other localities.

Cretaceous clays occur on Staten Island, and have been worked extensively at
Kreischerville and other points, the material being used in the manufacture of fire-
brick, etc.

t“ There is still some doubt as to the exact conditions under which the beds of clay
and gravel which form the greater portion of Long Island were deposited, but it is
probable that the clays represent shallow water marine deposits of Cretaceous and
Tertiary age .... The age of theclays is stil] largely a matter of speculation, and
will probably remain so in many cases unless paleontologic evidence is forthcoming.
Those on Gardiner’s Island are quite recent, as shown by the contained fossils, and
the clay on Little Neck is Cretaceous. The proof of the age of the Glen Cove clay is
not absolute .... The clays at Center Island, West Neck, Fresh Pond, and Fish-
er’s Island are very similar in appearance and composition, and are very probably of
the same age, possibly Tertiary, but we lack paleontologic or stratigraphic evidence.
At West Neck, the clay underlies the yellow gravel and the latter is covered by the
drift, so that is pre: Pleistocene.”

Shales occur in New York in the Hudson River, Medina, Clinton, Niagara, Salina,
Hamilton, Portage and Chemung groups. Of these, the last three are the most

tIn regard to the Long Island clays the author has been quoted verbatim, as the
statements could not well be summarized,—F, C, E.

548 REVIEWS

important in the present connection, though Medina shales are used in Ontario in the
manufacture of pressed brick, while vitrified brick are made from Salina shales at War-
ners, Onondaga county, N. Y. Hamilton shales are utilized at Cairo, Greene county,
in the manufacture of paving brick; and at Jewettville, Erie county, for pressed brick
manufacture. Sewer pipe, drain tile and terra cotta are made from Portage shales at
Angola, Erie county. Chemung shales are utilized at Jamestown, Chautauqua county;
Alfred, Allegany county; Hornellsville, and Corning, Steuben county; and Horse-
heads, Chemung county.

Feldspar and quartz are obtained from large pegmatite veins occurring near Bed-
ford, Westchester county, and are shipped to the potteries at Trenton, N. J.

Report on Louisiana Clay Samples. Geological Survey of Louisiana,

Report for 1899. Pp. 263-275.

Discussion of origin, composition and properties of clay, followed by reports on
physical tests of six samples of clay. Allof the clays tested could be used in the manu-
facture of pressed brick, while one could also be used for paving brick and two for
earthenware.

Clay and its Manufacture into Brick and Tile. Mineral Industry, Vol.
IX, pp. 93-135, 12 figures, Igol.
Detailed discussion of the manufacture of building and paving brick, roofing and
floor tile, terra cotta and sewer pipe.

Report on the Clays of Maryland. Special Publication, Vol. IV, Pt. III,
Maryland Geological Survey. 8vo, pp. 203-507; Pls. XIX-LXIX,
including six geologic maps; Figs. 5-34, 1902.

A general discussion of the origin, properties, uses and technology of clays is fol-
lowed by detailed descriptions of the Maryland clays, and a résumé of the industries
based on them.

According to Shattuck, the Pleistocene of Maryland is divisible into three forma-
tions; the Talbot, Wicomico, and Sunderland. The newest of these, the Talbot,
does not exceed forty-five feet in thickness ; and often carries lenses of greenish-black
clay. The Wicomico, which is from forty-five to one hundred feet thick, seldom con-
tains clay deposits of economic value. The Sunderland contains clay beds which are
well shown in Calvert and St. Mary’s counties. None of the Pleistocene clays are used
for anything except common brick, though occasionally clays occur fit for tile or
terra cotta.

The Neocene is represented by the Lafayette and Chesapeake formations. The
Lafayette consists of gravels, sands and clays, very irregularly stratified, and changing
character rapidly along the deposition planes. The Chesapeake consists chiefly of
sands of marls with only local developments of clay.

Clark and Martin divide the Maryland Eocene as follows:

GROUP. FORMATION. MEMBER.
Nanj Woodstock
ee ANISM OVeeriorrscrtooehtls Potapaco.
NUdGcOU Ono Hod ddaaomodes eae Pas poranea
GUTal svavetenerewverctetercnsts Versione Piscat
ke iscataway.

Of these four members only one, the Potapaco, contains clay deposits of impor-
tance. The Potapaco consists of argillaceous (and often gypseous) green sand, with a

REVIEWS 549

lower clayey member which is shown south of South River. The clay isa fine grained
material with occasional streaks of sand, and is in places twenty feet thick. It is
fairly plastic, abundant, and exposed near tidewater; suitable for use in the manufac-
ture of pressed brick and possibly of paving brick.

The Jurassic and Cretaceous of Maryland are divisible as follows:
Rancocas

Monmouth
Matawan

Raritan? |

Patapsco
ara cel Potomac group.

mersSiCd (pater tasp tater syste excievsyelaetatay eC) «la ouc lols «ve. e's ] Patuxent. |

Wipers Cretaceousmrae lara cmscrr seins iele cries oierct on soi e%e

NGO wera Ereta ceoOusmmmcriaicte cle e es eleieren sibue ais

The Upper Cretaceous formations include no clay deposits of economic value ;
but the Potomac group is of great importance. The Raritan? consists usually of
white sands and light colored clays, and reaches a thickness in central Maryland of
one hundred? feet. Raritan clays are found in Cecil, Kent, Harford, Baltimore,
Anne Arundel, Prince George’s, Howard, and Montgomery counties. ‘They are
worked at several points for use in pressed brick manufacture; while potter's clay
occurs in places. The Patapsco includes brightly colored mottled clays, with light
sands and clays, and has a maximum thickness of about 200 feet. In Cecil county a bed
of bluish stoneware clay often underlies the variegated clays. Refractory clays, as well
as brick clays, occur, and the Patapsco clays have been exploited to a considerable
extent. The Arundel contains lenses of bluish, siliceous, plastic clays, often carrying
iron concretions. The clays are largely used in the manufacture of common and
pressed brick, terra-cotta, roofing tile, and common pottery. The Patuxent consists
largely of sands, with occasional beds of sandy clay, and is,in this regard, the least
important member of the Potomac.

Shales from the different formations of the Carboniferous, Devonian and Silurian
have been tested and found suitable for various. uses. At present only two of these
formations furnish material of economic importance, these being the Pottsville forma-
tion of the Carboniferous and the Jennings formation of the Devonian. Both flint
clays and plastic clays occur in the Pottsville formation, the well-known Mt. Savage
fire clay being an example of the former. The beds have been opened near Frost-
burg, at Mt. Savage and west of Ellerslie ; and will probably become of even greater
economic importance. Shales of the Jennings formation are extensively used in the
manufacture of paving brick at Cumberland.

Kaolin occurs at many points in the area underlain by Algonkian rocks, notably
in Cecil county. These residual clays, having been derived from feldspathic rocks low
in iron-bearing minerals, are light colored and burn white. Only one company is
actually at work mining and washing this material, but the industry will probably
increase rapidly in importance. Owing to the presence of Patuxent and Columbia
beds, the kaolin is rarely exposed at the surface, the necessary stripping varying from

two to forty feet.

On page 397 the Raritan is excluded from the Potomac, but is included in it on
pages 399 et seg.— E.C. E.

2Elsewhere in the volume stated as 500 feet.—E. C. E.

350 REVIEWS

Less-pure residual clays, derived from rocks of various kinds and ages, are found
at many points in the state, and are used in places for brick manufacture.

SIEBENTHAL, C. E. The Silver Creek Hydraulic Limestone of Southeastern
Indiana. Twenty-fifth Ann. Rept. Indiana Dept. Geology and Natural
Resources Pp. 331-389, Pl. 14, colored geologic map, Igol.

Divided into three sections relating respectively to the stratigraphy, topography
and economic geology of the Silver Creek limestones.

A historical résumé of opinions in regard to the stratigraphy of southeastern Indi-
ana is followed by descriptions of local distribution and structure. The Devonian and
sub-Carboniferous formations occurring in this area, with their New York and Missis-
sippian (paleontologic) equivalents, are :

: 3 FEN GODStOME Hai aacs cn orcheten eealsjenece ete ciapals Tage acctaetcrn ews Kinderhook.
Sub-Carboniferous Rockfordlimestone tune snc eee ere cieic Choteau.
(aNew\ Albany "black shalei cs 00. Sie ut.ue co cuclocstes « ctvamaalelters Genesee.
| Sellersburg limestone....... sah etaoad caeted soon oscar RA AEE Hamilton.
Devonian= + Silver’ Creek hydraulic: limestone > \chjenie csic= arse seye ees Hamilton.
PetersonvaillemlimeStOme)sevcesleccya ciekelor sy oxelessietey ere otelons anaes culate Corniferous.
WPendletom:sandstomela caytiece «rue ators uacstele cuecctat tn cscue eres Schoharie.

The topography is briefly mentioned. Pleistocene terraces are described and
mapped, and a possible preglacial channel of the Ohio River is pointed out.

The Silver Creek hydraulic limestone is massive, very fine grained and usually
fossiliferous. In color it varies from light to dark or bluish-drab, weathering buff.
In composition it varies between the following limits: Calcium carbonate 52 per
cent. to 62 per cent.; magnesium carbonate 16 per cent. to 35 per cent.; silica 9.5 per
cent. to 18.5 per cent.; ferric oxide I.4 per cent. to 2.0 per cent.; alumina 2 to 5 per
cent. The cement rock is obtained by quarrying or mining, according to the situa-
tion; blasted out by dynamite and sledged to proper size (6"—12") for the kilns. The
latter are continuous up-draft kilns, and require about seventy-two hours for calcina-
tion. The burned rock is sent through crushers, “re-grinders,” and rock-emery mills.
During 1900, thirteen plants, working 116 kilns, were in operation, the total product
being 2,512,000 barrels.

RECENT FUBLICATIONS.

—ASHMEAD, WILLIAM Harris. Papers from the Harriman Alaska Expedi-
tion XXVIII: Hymenoptera. [Proc, Wash. Acad. Sci., Vol. IV, pp.
117-274, May. 1902.]|

—AsuHLEY, GEORGE H. The Eastern Interior Coal Field. [Extract from
the Twenty-second Ann. Rept. of the Survey Igoo-1. Part III—
Coal, Oil, Cement, Washington, 1902.]

—BARLOW, ALFRED ERNEST. On the Nepheline Rocks of Ice River,
British Columbia.

—Bain, H. Foster. The Western Interior Coal Field. [Extract from the
Twenty-second Ann. Rept. of the Survey, 1900-1. Part II1— Coal,
Oil, Cement. Dept. of the Interior, Washington, 1g02.]|

— Bulletin of the University of Texas, No. 15. Coal, Lignite, and Asphalt
Rocks. [The Univ. of Texas Mineral Survey, May, 1902, Austin,
Texas. |

— BLAKE, WILLIAM P. Lake Quiburis, an Ancient Pliocene Lake in Arizona,
[From the Univ. of Arizona Monthly, Vol. IV, No. 4, February, 1902. ]

— BricGs, LyMAN J. and LApHAM, Macy H. Capillary Studies and Filtra-
tion of Clay from Soil Solutions. [U.S. Dept. of Agriculture, Bureau of
Soils, Bull. 19.

—CorNET, J. Note sur les Assises comprises dans le Hainaut entre la
Meule de Bacquegnies et le Tourtia de Mons. [Extrait des Annales de
la Soc. géol. de Belg. t. XXVIII, 1901.]

Note Préliminaire sur la Composition Minéralogique des Argiles et des
Limons. [Extrait des Annales de la Soc. géol. de Belg. |

La Géologique du Bassin du Congo d’apres nos Connaissances actuelles
(1897) t. XII, 1898. [Extrait des Annales de la Soc. géol. de Belg.
de Paleontologie et d’Hydrologie, Dec., 1go1.|

Notes sur des Roches du Mont Bandupoi et du Haut Uellé. [Extrait
des Annales de la Soc. géol. de Belg. de Paleontologie et d’Hydro-
logie.|

Le Quaternaire Sableux de la Vallee de la Haine. [Extrait des Annales
de la Soc. géol. de Belg. |

— Dixon, RoLAND B. The Huntington, California Expedition. Maidu
Myths. [Bull. of the Amer. Mus. of Nat. Hist., Vol. XVII, Pt. II, pp.
33-118, New York, 1902. ]

551

552 RECENT PUBLICATIONS

— EASTMAN, C. R. Some Carboniferous Cestraciont and Acanthodian
Sharks. [Bull. of the Mus. of Comp. Zodlogy, Vol. XXXIX, No. 3,
Harvard Coll., Cambridge, Mass., June, 1902. ]

—Hosss, WILLIAM HERBERT. Former Extent of the Newark System.
[ Bull. Geol. Soc. of Amer., Vol. XIII, pp. 139-148, Rochester, 1902. ]

— LAmMBE, LAwrRENCE M. On Tryonyx Foveatus, Leidy, and Tryonyx
Vagans, Cope, from the Cretaceous Rocks of Alberta. [Reprinted from
the Summary Report for the year 1901. Geol. Survey of Canada,
Ottawa, 1g02.]

—LAwson, ANDREW C. The Eparchaean Interval. A Criticism of the Use
of the Term Algonkian. [Bull. of Dept. of Geol., Univ. of Cal., Vol.
III, pp. 51-62, May, 1902.]

— MERRILL, GEORGE P. A Newly Found Meteorite from Admire, Lyon
County, Kansas. [From the Proc. of the U.S. Nat. Mus., Vol. POA
PP- 907-913, 1902.]

— Newton, R. B. and HOLLAND, R. On some Fossils from the Islands of
Formosa and Riu-Kiu (= Loo Choo). [Reprinted from the Journal of
the College of Science, Imperial University, Tokyo, Japan, Vol. XVII,
article 6, 1902. |

— NANSEN, FRIDTJOF. Some Oceanographical Results of the Expedition
with the ‘“ Michael Sars’’ headed by Dr. J. Hjort in the summer of Igoo.
[Printed by A. W. Brogger, Christiania. |
The Oceanography of the North Polar Basin. IX. The Norwegian North

Polar Expedition, 1893-96. [Pub, by the Fridtjof Nansen Fund for
the Advancement of Science. Reprinted from Vol. III, Christiania,
1902. |

—PricE, JAMES A. Observations and Exercises on the Weather. Labora-
tory Work in Physical Geography and Meteorology. [American Book
Company. |

— RATHBURN, MARy J. Papers from the Hopkins Stanford Galapagos
Expedition, 1898-g VIII: Brachyura and Macrura. | Proc. Wash.
Acad. Sci., Vol. IV, pp. 275-292, June, 1902.]

— Ries, HENRICH. Peat, [Extract from Mineral Resources of the United
States, ‘Calendar Year, s1oo01..’ Dept:-of Interior, U.S. Geols survey,
Washington, 1902. ]

—TAaFF, JosEpH A. The Southwestern Coal Field. [Extract from the
Twenty-second Ann. Rept. of the Survey 1goo-1. Part III] —Coal,
Oil, Cement. Dept. of the Interior, Washington, 1g02.]

—TurnerR, H. W. A Sketch of the Historical Geology of Esmeralda
County, Nevada. [From the Amer. Geologist, May, 1902.]

RECENT PUBLICATIONS 553

Notes on Unusual Minerals from the Pacific States. [From the Amer.
Jour. of Sci., Vol. XIII, May, 1902.]

— SPENCER, J. W. The Windward Islands of the West Indies. With eight
plates and six charts. [Trans. of the Canadian Institute, Vol. VII, pp.
351-370, December, Igo. ]

—VotH, H.R. The Oraibi Powamu Ceremony. The Stanley McCormick
Hopi Expedition. [Field Columb. Mus. Pub. 61, Vol. III, No. 2.]

— WoopworTH, JAy Backus. Pleistocene Geology of Portions of Nassau
County and Borough of Queens. [N. Y. State Mus., Bull. 48, December,
Igol.]

—WILLIsTON, S. W. On the Skeleton of Nyctodactylus with Restoration;
With one text figure. [Reprinted from the Amer. Jour. of Anat., Vol. I,
No. 3, pp. 297-305, May, 1902.]

— WILLIAMS, HENRY S._ Fossil Faunas and their Use in Correlating
Geological Formations. [From the Amer. Jour. Sci., Vol. XIII, June,
1902.]

— WRIGHT, G. FREDERICK. Origin and distribution of the Loess in North-
ern China and Central Asia. [Bull. Geol. Soc. of Amer., Vol. XIII,
pp. 127-138, Rochester, May, 1902. |

bs

LAE

NOURNAL OF GEOLOGY

SZ PEE MBM h-OCTOBL Ki IG02

Me OUANTITATIVE CHEMICO-MINERALOGICAL
CLASSIFICATION AND NOMENCLATURE OF
IGNEOUS ROCKS:

CONTENTS.
SUMMARY.

PART I. CLASSIFICATION.
Introduction.
Defects of present system.
Basis of classification.
The facts to be considered; The existence of petrographic provinces; The
principles; A brief discussion of various principles.
Chemico-mineralogical classification.
Standard mineral composition.
Construction of the system.
Outline of the system. :
Class; Subclass; Order; Suborder; Rang; Subrang; Grad; Subgrad;
Section.
The rile of actual mineral composition and texture in rock classification.
Actual mineral composition.
Complete or almost complete accord between the mode and norm; Appreci-
able difference between norm and mode; Varieties; Indeterminable modes.
Texture. ‘
Crystallinity ; Granularity; Fabric; Heterogeneous textures.

PART Il. NOMENCLATURE.
Magmatic names.
Termination; Root; Names.
Rock names.
Actual mineral composition ; Texture ; Examples of names; Type and habit.
Rock names for general field use.
Phanerites ; Aphanites; Glasses.

Vol. X, No. 6. 555

556 CROSS, IDDINGS, PIRSSON, WASHINGTON

PART III. METHODS OF CALCULATION.
Chemical relations among salic minerals.
Chemical relations among femic minerals.
Calculation of the norm.
Percentage weights of minerals; Examples of calculations.
Calculation of norm from mode.
Calculation of mode from chemical composition of a rock.
Calculation of alferric minerals.
Aluminous pyroxenes; Aluminous amphiboles; Ferromagnesian micas;
Garnet.
An example of calculation.
Calculation of the norm; Calculation of the mode; Calculation of the norm
from the mode.
EPILOGUE.
TABLES OF ALFERRIC MINERALS AND THE ROCKS IN WHICH THEY OCCUR.

SUMMARY.

WE present in this work an entirely new system for the
classification and nomenclature of igneous rocks, which repre-
sents the results obtained after some years of consideration
and discussion of the subject in all its bearings.

In uniting our efforts for this purpose we have been actuated
by a sense of the seriousness of the responsibility assumed in
such an undertaking, and by a conviction that the best results
were to be obtained by the co-operation of several workers who
agreed on fundamental principles and who were capable of har-
monious collaboration.

Originally Professor George H. Williams was associated with
us in this work. By his death we have been deprived of a
counselor whose judgment we held in the highest esteem, and
whose loss we cannot cease to lament.

Many attempts were made to modify in various ways the
existing systems of classification in the endeavor to make them
answer the demands of modern petrology, and each attempt
was successively cast aside on practical trial. It came thus
to be perceived by all of us that the main weakness of old sys-
tems lay in their fundamental principles and methods and that
any new, logical and comprehensive system based on the vast
amount of knowledge of igneous rocks acquired in these later

CLASSIFICATION OF IGNEOUS ROCKS 557

years must be built up from the very beginning on different
lines from those heretofore employed. After many endeavors
in various directions, we gradually evolved the system here
presented. Along with its evolution has gone hand in hand the
calculation of thousands of analyses by which it has been
tested and its formation in large part controlled. It is a chemtco-
mineralogical system based on its own principles, and is in nowise
an attempt to reduce any one of existing systems to a chemical
basis or to formulate one of them in a chemical way. Its con-
cepts of rocks are in a large measure new, and hence, except in
a very small degree, it demands a new nomenclature.

In brief outline, what we propose is as follows: All igneous
rocks are classified ona basis of their chemical composition; all
rocks having like chemical composition are grouped together.
The definition of the chemical composition of a rock and of a
unit of classification is expressed in terms of certain minerals
capable of crystallizing from a magma of a given chemical com-
position, and the expression is quantitative. For this purpose
the rock-making minerals are divided into two groups, consist-
ing on the one hand mainly of the more highly siliceous
alkali- and calci-aluminous ones, and on the other of the ferro-
magnesian ones. The first group is called mnemonically the
salic group, the second is the femzc group. From this category
the micas and aluminous augites and amphiboles are excluded
for reasons given in full.

For the purpose of completely classifying a rock by this
system its chemical composition must be actually known by
chemical analysis, or approximately so by physical means or by
microscopic, optical methods indicated by the authors. The
thousands of rocks of various mineral compositions and tex-
tures whose places in this system are indicated by chemical
analysis become types for comparison, analogous to type speci-
mens of zoology and botany, by which similar rocks may be
approximately classified.

Since it is known that many magmas may crystallize into
quite different mineral combinations, according to the influence

558 CROSS, IDDINGS, PIRSSON, WASHINGTON

of attendant conditions upon the union of the various chemical
constituents of the magma, it is necessary to select avicertain
set of salic and femic minerals as uniform standards of compari-
son. These standard minerals are, for the most part, species
commonly formed, but the aluminous pyroxenes, amphiboles,
and micas are necessarily excluded. In practice, the molecular
composition of a rock, obtained from its chemical analysis, is
computed by a regular method into amounts of these standard
minerals, and the place of the rock in the system is then easily
determined.

The standard mineral composition of a rock is called its norm,
and this may be quite different from its actual mineral composition,
or mode. Methods for obtaining the latter and for indicating
its relation to the former are set forth in detail in the Part III,
on Calculations.

On the relative proportions of these two groups of standard
minerals, the rocks are first divided into five Classes, accordingly
as one or the other of these two groups alone constitutes the
norm, or is extremely abundant; whether one or the other is dom-
znant; or whether the two are in about egual proportions. The
Classes thus formed are divided into Orders on the relative pro-
portions of the minerals forming the predominant group in each
case, and in the middle group on the relative proportions of the
salic minerals. Thus in the preponderantly salic Classes the
Orders are based on the relative amounts of quartz, feldspars,
and feldspathoids.

The Orders are divided into Rangs on the chemical character
of the bases in the minerals of the preponderant group in each
case ; thus, if these were feldspathic, as to whether they are alkalic,
alkalicalcic, or calcic. The lowest division or Grad obtains
only in the three intermediate Classes, and results from the con-
sideration of the relative amounts of the minerals composing the
subordinate group.

In addition to these divisions further ones are provided for
where necessary by Subclasses, Suborders, Subrangs, and Sub-
grads.

CLASSTFIGATION OF IGNEOUS ROGKS 559

Texture is considered of minor importance in classification,
and is taken into account after the chemical and mineral
composition.

Nomenclature.— The system demands an entirely new nomen-
clature, and it has been sought to introduce this according to a
definite system, the lack of which is so painfully evident in the
present nomenclature.

The nomenclature proposed consists of three parts: primarily
of substantive names for the magmatic units, implying the chem-
ical composition and the norm, or standard mineral composition ;
secondarily, two sets of adjective terms to be used to qualify the
magmatic names; one set referring to the actual mineral com-
position, or mode, and the other to the texture of the rocks.

The magmatic name consists of a root, derived from a geo-
graphical name in all cases except for the names of Classes and
Subclasses; and of a sutiix. The suffixes are so chosen as to
vary in a definite way with the division of the system to which
the magmatic name belongs. Thus for Class, Order, Rang, and
Grad, the letters 7,7, s, and ¢ in alphabetical order are used
with the vowel a, giving in English ane, are, ase, ate. For Sub-
class, Suborder, etc.,the vowel is changed to a, giving one, ore,
ose, ote.

The roots forming the names for classes are sa/ and fem,
mnemonic of the salic and femic minerals constituting their
norms, and are combined with prefixes yielding the following:
Persalane, Dosalane, Salfemane, Dofemane, and Perfemane.

The roots for the names of the divisions smaller than Class
are derived from the names of geographical localities, and as
far as possible from those at present in use for rock names,
advantage being taken of their connotations as to magmatic
character.

The authors propose a nomenclature for field use based on
purely megascopic characters.

The work concludes with a discussion of methods of calcu-
lating mineral composition from chemical composition and the
reverse, and presents tables to aid such calculations.

560 CROSS, IDDINGS, PIRSSON, WASHINGTON

PARW. Th CLASSIFICATION:
INTRODUCTION.

The commonly acknowledged unsatisfactory character of rock
classifications in present use, and the unsystematic nomenclature
of petrography, have convinced us of the necessity for a com-
plete reconstruction of both. Recognizing the magnitude of this
task, yet desiring to see it accomplished as early as possible, we
have united our efforts toward the production of a new system
of classification and the creation of a nomenclature to express it.
Many attempts at improvement of existing schemes of classifi-
cation have been made in recent years, but they have failed to
accomplish important results because they have not gone to the
root of the matter.

The discussions of petrographers associated with the Inter-
national Geological Congress have demonstrated the futility of
attempting the regulation of petrographical nomenclature with-
out first fixing the basis of rock classification, since the concep-
tions by different petrographers of the objects to be named are
diverse. And the effort to establish a systematic classification
of igneous rocks by international conference and agreement has
in like manner proved ineffective because of evident inherent
difficulties.

The recently published Compte Rendu of the Eighth Session of
the International Geological Congress, held in Paris, furnishes an
illustration of the diversity of views held by European petrogra-
phers, while presenting in definite form the convictions of a consid-
erable number of the foremost workers in this science regarding
the principles that should govern the choice of bases of a sys-
tematic classification. To avery considerable extent we find our-
selves in accord with the opinions expressed in the report of the
Russian Committee and in the discussions and report of the Paris
Commission of October, 1899. And we are encouraged to hope
that the system here presented by us will meet with a cordial
reception by those petrographers who, sharing more or less
completely our conceptions of the fundamental principles that

GEA SSTFACATION OF IGNEOUS ROCES: 561

should control the formation of a systematic classification of
igneous rocks, may approve of our method of applying them.

We wish to acknowledge our obligations to all those whose
ideas and writings have influenced us consciously or uncon-
sciously; influences which will be evident in many places in the
succeeding pages. It is not always possible to credit a particular
conception toa particular author. The science of petrography
has developed so rapidly, and so many workers have been
engaged upon kindred problems, that similar ideas have forced
themselves upon investigators. Especially, in our own case, fre-
quent interchanges of thought have so blended and modified
our ideas that is difficult for any one of us to identify his own.

In particular we acknowledge our obligations. to those
masters of petrography from whom we caught inspiration for our
petrographical careers. To Professor Ferdinand Zirkel and to
Professor Heinrich Rosenbusch, our instructors, whose work and
thought have opened the way to a host of students of rocks, we
are indebted for much that has influenced us in shaping the
system proposed, which is in fact a natural outgrowth of present
petrographical conditions.

Defects of present systems.—From the review of rock classifi-
cation in an earlier part of this volume by one of us, it appears
that all past and existing classifications of rocks have fundamental
weaknesses arising either from the use of theoretical concepts, or
from an inconsequent or illogical application of the characters
of igneous rocks as bases of their systematic arrangement.
That a new nomenclature is required for a new system can
scarcely be disputed. The present confusion in petrography is
in no small degree due to redefinition of old terms, and to this
confusion we have no desire to add. Without going into an
extensive discussion of the vitally weak points of existing systems
of petrography, upon which all of us have expressed individual
opinions, some of the most unsatisfactory features may be
briefly stated as follows:

1. There are few definite, clearly enunciated and generally
applicable guiding principles, the consequence being that the

562 CROSS, IDDINGS, PIRSSON, WASHINGTON

present systems are to a large extent arbitrary aud subjective,
and are capable of being applied differently by different individu-
als, as is evident from the numerous cases where different names
have been given to the same rock, or the same name to obvi-
ously different rocks.

2. There is a lack of uniformity in the method of application
of existing systems not only among petrographers of different
countries, but among those of the same country.

3. The present systems are to a certain extent founded on
theory or hypothesis, while classification in order to be stable
must eschew all such bases, and be founded only on ascertained
facts.

4. Present systems are to a large extent qualitative rather
than quantitative, the result being that some given character of
rocks is arbitrarily used as a criterion beyond its natural range
of application, as, for example, the use of the mere presence of
feldspar to distinguish a group of rocks, whether it be the domi-
nant constituent or only a very subordinate one.

5. The construction of modern systems is faulty in that the
groups of rocks, or of rock families, now recognized are quite
inadequate to express known relationships, and the varying
groupings used by different authors are not based upon definite
principles, nor are the principles applied uniformly. As a result
of this condition, existing systems are highly uneven and it is
often necessary to use either too specific or too general terms in
naming a given rock.

6. The nomenclature of petrography is quite inadequate to
express the relations between the various groups. The single
termination makes it impossible to indicate in the name, fer se,
whether it applies to a large or to a small classificatory division,
or whether to rocks or minerals, and gives rise to great monotony.

7. As a consequence of these conditions, there is no guide as
to when the use of a new name is necessary or justified, each
investigator being his own judge in such matters. In some rock
groups which have been recently studied, there is an abundance
of new names, while in other more common and longer known

CLASSIFICATION OF IGNEOUS ROCKS 563

groups, a few names are made to do duty for many obviously
distinct rocks.

Upon the grounds stated, a new system of petrography seems
to us necessary. This conclusion has been reached by us only
after many efforts to revise or remodel existing schemes. Our
first conferences upon this undertaking were participated in by
George Huntington Williams, and our first exchange of sugges-
tions bears the date of May 1, 1893. These efforts on our part,
extending over ten years, have brought us from somewhat
diverse views to full accord in the system here presented.

BASIS OF CLASSIFICATION.

The progress in the knowledge of the composition, texture,
and occurrence of igneous rocks, as well as in the conceptions
of the probable causes of variation among different bodies and
within one body of rock, has been such as to justify us in con-
sidering the problem of the classification of igneous rocks in the
light of certain established facts of a general nature. The
recognition of these leads to the establishment of general prin-
ciples upon which a new system of classification may be based.

The facts to be considered may be stated categorically as fol-
lows:

The once igneous or molten condition of the magmas from
which rocks of this kind have solidified.

The characters of such magmas as solutions.

The physical and chemical properties of such solutions.

The ability of complex solutions to differentiate by osmotic
diffusion, liquation, fractional crystallization, or by other pro-
cesses, under varying conditions of temperature and pressure.

The variations in chemical composition resulting from such
differentiation.

The existence of marked chemical variations within igneous
masses of small volume in some instances, and of slight varia-
tions within bodies of large volume in others.

The absence of fixed proportions of constituents in rocks
with two or more mineral components.

564 CROSS, IDDINGS, PIRSSON, WASHINGTON

The gradations in the chemical and mineral composition of
rocks, within limits.

The limited number of the important rock-making minerals.

The existence of rocks having but one mineral component.

The variable crystallization of a chemically homogeneous
magma, or of several chemically similar magmas, whereby differ-
ent combinations of minerals may be produced from chemically
similar magmas. In other words, the fact that rocks having
diverse mineral compositions may be chemically alike.

The fact that diverse textures develop from the same magma
and from cheniically similar magmas, and conversely that similar
textures develop from chemically different magmas.

The incomplete crystallization of magmas or their solidifica-
tion into glass in many instances.

The identity of rocks of different geological ages.

The identity of many rocks with different modes of geologi-
cal occurrence.

The existence of petrographical provinces in which the igneous
rocks genetically related are distinguishable from those of other
regions, when considered in connection with the occurrence
elsewhere of similar petrographical provinces and with the
gradual transitions between provinces, should be treated as a
larger phase of differentiation. Moreover it has been shown by
Broégger* that similar rocks may be differentiated from different
parent magmas in several petrographical provinces, and may
occur in two or more unlike series. That is, rocks belonging to
distinct provinces may resemble one another so closely in their
dominant characters that they would naturally be defined by the
same terms. Hence, we conclude that all igneous rocks should
be correlated and classified in one comprehensive system based
upon principles common to igneous rocks in general.

The principles which we consider applicable to the classifica-
tion of igneous rocks, as criteria by which to judge of the facts

* Ouart. Jour. Geol. Soc., Vol. L (1894); p. 36, and Die Eruptivgesteine des Kris-
tianiagebietes, Vol. III, p. 57, Christiania, 1898.

CLASSIFICATION OF IGNEOUS ROCKS 565

to be employed and their method of application, are as fol-
lows:

The classification should be free from hypothesis, and be
based only on facts or relations determinable in the rock
itself.

The classification should be quantitative as far as possible,
and each constituent, chemical or mineral, should be given
weight in proportion to the amount present in each case, irre-
spective of its rarity or unusual occurrence.

The chemical composition of the rock is its most fundamental
character, being a quality inherent in the magma before its
solidification, and is therefore of greatest importance for its cor-
relation with other rocks.

All rocks of like chemical composition should be classed
together, and degrees of similarity should be expressed by the
relative positions or values of the systematic divisions of the
classification.

The mineral and textural characters, being dependent largely
on external conditions attending rock solidification, are to be
regarded as of subsidiary importance in classification, but should
receive due recognition in the system.

Since it is the chemical composition of the magma that is the
fundamental character of igneous rocks by which they are to be
classified, only fresh, undecomposed, or unaltered rocks are to
be employed in establishing such a classification.

A BRIEF DISCUSSION OF VARIOUS PRINCIPLES that have entered
into systems of rock classification is introduced at this place in
order to make clear our reasons for selecting those employed
in creating the system proposed by us.

Rocks have always been recognized as extremely difficult of
systematic classification, because of their infinite variation or
gradation from one kind to another in many ways. Reviewing
the many factors which may be employed in their classification,
they are found to fall into two groups, namely, geological rela-
tions, and inherent characters; and the task of the systematic
petrographer is to select the characters and apply the criteria

566 CROSS, IDDINGS, PIRSSON, WASHINGTON

which will produce the most natural and stable as well as com-
prehensive and elastic arrangement.

Of geological relations, the mode of origin is now universally
recognized as the first principle to apply to the sub-kingdom of
rocks, to secure the grand divisions of which the igneous rocks
are one. The further application of geological mode of origin
in subdivision of igneous rocks involves the use of theoretical
considerations and produces instability of system.

The relations of geological occurrence have been used in the
arrangement of igneous rocks, but unsuccessfully. Geological
age cannot now be used without violating the known fact that
rocks of many ages are identical in their material qualities.

Some systems of classification now in use are based on sup-
posed relationships between the material characters of igneous
rocks and their modes of occurrence as geological bodies, which
are only partially in accord with facts, and which therefore intro-
duce serious weakness into the foundation of the systems. There
are no particular kinds of rocks that invariably characterize geo-
logical bodies of special shapes, such as stocks, laccoliths, dikes,
etc. Nor is there any specific texture that indicates the depth
beneath the surface of the earth at which a rock has crystallized.
While there is unquestionably a relation between the texture and
mineral composition developed ina given magma and its physical
environment during eruption and intrusion and at the time of its
solidification, this relationship is so intricate, and the possibili-
ties of environment so manifold, that it cannot be made a basis
for classification.

The effort to classify rocks on a basis of their genetic rela-
tionships by grouping them in such a manner as to express the
fact that all the rocks of a particular center of eruptive action
are the differentiates of some common parent magma introduces
the utmost complexity, because each group presents a particular
set of relations, and it becomes necessary to recognize almost as
many groups as there are known centers of eruption. But as
already stated, the members of several groups may resemble one
another so closely as to be capable of the same definition and

CLASSIFICATION OF IGNEOUS ROCKS 567

deserving of the same name. The consistent application of such
a system of grouping would therefore separate rocks which are
alike so far as intrinsic qualities are concerned. And this,
according to our conception of classification, is not classification.
Many of these facts are essential to the complete petrological
understanding and description of rocks, but are not applicable
to the construction of a petrological system.

The inherent characters of igneous rocks have always been
prominent in the formation of petrographic systems, and are
plainly the features it is most natural to select. This was spe-
cially pointed out by the fathers of systematic petrography, von
Leonard and Brongniart, and has been emphasized in recent dis-
cussions. Of these characters chemical and mineral composi-
tion, structure or texture, are the most important, the others
being comparatively trivial or accidental. Structure or texture
is now known to depend so largely on variable conditions attend-
ing the consolidation of magmas that it can no longer be given
the prominent réle hitherto assigned to it. Chemical and min-
eral composition then remain as those characters of igneous
rocks most available tor their classification. Of these, it is to
be noted, that while the two are most intimately related, the for-
mer is more fundamental, since it pertains to a magma which
may consolidate as a glass or become a holocrystalline rock, and
in the latter case the mineral constitution varies with attendant
conditions.

CHEMICO-MINERALOGICAL CLASSIFICATION.

While the chemical composition of igneous rocks is their
most fundamental characteristic, it is known that there is an
absence of stoichiometric proportions among the chemical ele-
ments.or components.) (It 1s further clear that »there isan
intricate interrelationship and serial variation among these com-
ponents and an absence of chemical division lines, or of groups
or clusters of similar combinations of elements. These facts show
that any subdivision on a purely chemical basis must be arbi-
trary, unpractical and unsatisfactory.

All holocrystalline, and many of the partially crystalline

568 CROSS, IDDINGS, PIRSSON, WASHINGTON

rocks derive their most obvious characters from the mineral par-
ticles composing them. The varying proportions of unlike
minerals in rocks are most striking, and other notable features
are due to the physical properties of the minerals, their color,
cleavage, hardness, etc., or to the relative or absolute size or
shape of the particles.’ ‘Itv1s by reference tor theseycharacters
that rocks may be most readily described and identified, and it
is then desirable that the systematic classification should be
constructed as far as may be by the use of mineralogical data in
one form or another.

There are, however, reasons why mineral constitution by
itself cannot be used in the principal divisions of a compre-
hensive and logical classification of all igneous rocks. The
existence of vitreous rocks forms one of these reasons, because
such rocks cannot be classified at all by purely mineralogical
criteria. The fact that a given magma may crystallize into dif-
ferent mineral combinations is another reason. Moreover, if
mineral composition were a simple function of chemical compo-
sition, and if all rocks were holocrystalline, the number of
chemically different minerals of importance would make the
task of classifying rocks by means of mineral composition alone
practically impossible. It, therefore, appears that neither chem-
ical composition nor mineral constitution can be independently
apphed to the construction of a logical and practical classifi-
cation of igneous rocks,

The primary minerals in a holocrystalline igneous rock, when
considered chemically and quantitatively, are a full expression of
the chemical composition of the magma, and their exact determi-
nation furnishes the chemical composition of the rock. To a
large extent the mineral composition may be employed as a
means of determining the chemical composition, and since the
minerals are readily determinable optically in many cases and
are a convenient means of identifying rocks, it is advisable Zo
treat the chemical composition of rocks in terms of minerals, and to
make the basis of primary subdivisions chemico-mineralogical.

While this conclusion is, in its general terms, quite com-

CLASSIFICATION OF IGNEOUS ROCKS 569

monly asserted as the purpose of existing mineralogical systems
of rock classification, it requires but casual consideration to see
that a qualitative mineralogical system cannot express chemical
composition, and a thoroughly gwantitative scheme has never
been formulated.

Before stating the method of classification to be proposed, it
is important to point out some of the chemical and mineralogical
relationships obtaining in igneous rocks. And first it is to be
noted that while the chemical composition of a magma controls
in general the kinds of minerals that may crystallize from it, so
that quartz forms in the more siliceous rocks, and olivine in
those rich in magnesium and iron, still it does not fix absolutely
the kinds or the proportions of all-of the rock-making minerals.
This is due to the fact that a number of these minerals consist
of similar elements in diverse proportions, so that two or more
different combinations of elements may be developed in chem-
ically similar magmas. Or, as is well known, some of the
minerals having a complex composition may be dissociated into
less complex ones. A familiar example is the experimental
melting of hornblende and the obtaining in its stead pyroxene
and magnetite. Another illustration of the same kind of rela-
tionship is the chemical identity of some hornblende-andesites
and some pyroxene-andesites.

A striking illustration is furnished by the hornblendite and
camptonite of Gran, Norway, described by Brégger.* The two
rocks having almost identical chemical compositions are com-
posed in the first case of somewhat alkalic, aluminous hornblende,
and in the second of less aluminous hornblende and feldspar.

The development of biotite in some rocks and its absence
from others having like chemical composition is well known; as
shown by its presence in some gabbros and its absence from
some chemically equivalent basalts; its presence in certain
diorites and its absence from equivalent andesites ;? its develop-

1 Op. cit., pp. 60, 93.

2Ippincs, J. P., ‘The Eruptive Rocks of Electric Peak and Sepulchre Moun-
tain, Yellowstone National Park,” Zwel/th Ann. Rept. U. S. Geol. Surv. (Washington,

1892), p. 653.

570 CROSS, IDDINGS, PIRSSON, WASHINGTON

ment in minettes and absence from certain basanites. And with
this difference in biotite there is a variation in olivine, hypers-
thene, magnetite and other minerals within the rocks mentioned.

Other notable illustrations of different mineral development
in chemically similar rocks are the madupite of Wyoming’? and
venanzite of Italy,” in the latter rock melilite and olivine appear-
ing instead of pyroxene and phlogopite in the madupite ; also the
nephelite-syenite of Beemerville, N. J.,3and the leucite-phonolite
of Bracciano.+ Indeed, instances of the same kind are well known
to all, and are constantly increasing in number. It is therefore
indisputable that magmas of identical chemical character may
and do solidify as very different mineral aggregates, it being
also a possibility that they form on solidification no minerals at
all, or that they crystallize only in part.

STANDARD MINERAL COMPOSITION.— Whether vitreous or crys-
talline, all igneous rocks may be correlated by considering what
mineral combinations may be developed from their magmas if
completely crystallized. But since several mineral combinations
are possible for most magmas, it is advisable to select one of
these combinations as the standard of comparison. And for
uniformity and simplicity it is necessary to select the same one
for all rocks having like chemical composition. This may be
termed the standard mineral composition, which may or may not
correspond to the actual mineral composition. —

Before presenting the reasons for selecting certain minerals
as those best adapted for a chemico-mineralogical classification
of igneous rocks, let us consider the important rock-making
minerals from the general standpoint of their chemical compo-
sition. They may be arranged in several groups chiefly

™Cross, W., “Igneous Rocks of the Leucite Hills, etc.” dm. Jour. Scz., Vol. 1V
(1897), pp. 115-141.

2SABATINI, V. I Vulcani di S. Venanzo. Rivista di Min. e Crist., Vol. XXII
Padova, 1899, pp. I-12.. Ci. RosENBUSCH. Sd. Berl. Ak, (1899) p. 113.

3 Bull. 150, U.S. Geol. Surv. (Washington, 1898), p. 209.

4WASHINGTON, H.S., “Italian Petrographical Sketches,’ Jour. GEOL., Vol. V
(1897), pp: 43, 49.

CLASSIFICATION OF IGNEOUS ROCKS 571

distinguished by chemical characters, but also by associations in
thesrocks-| Bheyeare::

a) Silica and alumina uncombined, quartz (tridymite) and
corundum, together with zircon, which, though commonly present
in very small amount, is.oftenest found in rocks rich in silica or
alumina.

b) Aluminous non-ferromagnesian minerals; orthoclase, albite,
anorthite and mixtures of these, leucite, analcite, nephelite, soda-
lite, hauynite, noselite, cancrinite, and muscovite.

c) Aluminous ferromagnesian and calcic silicates (interme-
diate between 6) and d@) ): aluminous pyroxenes and amphiboles,
biotite, garnet, tourmaline, melilite, some spinels, etc.

a) Non-aluminous ferromagnesian and calcic silicates: hyper-
sthene (including enstatite), diopside (including hedenbergite),
acmite, olivine (including fayalite and forsterite), and aker-
manite. :

€) Non-siliceous and non-aluminous minerals with titanosilicates ;
magnetite, hematite, ilmenite, apatite, titanite, perofskite, and
fluorite, together with the native metals, and certain other
metallic oxides and sulphides.

If igneous rocks are considered from the standpoint of their
mineral composition, they are found to consist of graduating
series of quantitatively different mixtures of several groups of
minerals, and since they are all necessary to an exact expression
of the chemical composition of the rock,.each mineral or group
of minerals should receive proper recognition according to its
quantitative value. Owing to the number of minerals in most
rocks, this is a very intricate problem and we have made repeated
attempts to solve it by recognizing several independent mineral
factors at one time, involving the problem of handling three or
more co-ordinate quantities. This was found impracticable as a
basis of classification, and it was seen that the most feasible pro-
cedure is to recognize such factors successively according to
certain degress of qualities or magnitudes possessed by them in
comparison with one another. This has been done by grouping
them on a basis of chemical identity or resemblance, and of

572 CROSS, IDDINGS, PIRSSON, WASHINGTON

established affinities or associations, and by successively sub-
dividing these groups by subordinate chemical differences or
quantitative values. For this purpose it is necessary to assemble
all rock-making minerals into two chemically distinguished
groups. :

These have been made by uniting quartz, corundum and
zircon with the aluminous non-ferromagnesian minerals— feld-
spars, feldspathoids and muscovite—in one group, and by
placing the non-aluminous ferromagnesian and calcic minerals
—hypersthene, diopside, acmite, olivine, and akermanite — with
the non-siliceous and non-aluminous minerals, and titanosilicates,
—magnetite, hematite, ilmenite, apatite, etc.—in the other
group.

This leaves the aluminous ferromagnesian minerals to be
treated in another manner. The reasons for this separation of
the minerals, biotite, amphibole and augite, are discussed at
length in a later part of this article, but it may be said here that
their variable composition and occurrence, together with the fact
that they may be considered as mixtures of aluminous and non-
aluminous molecules, make it advisable to defer their introduction
into the system of classification until the actual mode of crys-
tallization of the rocks is taken into account. Since it is possible
that a magma of any given chemical composition may crystallize
without the development of these minerals, and_ since the
chemico-mineralogical expression of igneous magmas is greatly
simplified by not considering these minerals until the particular
crystallization of the magma is to be expressed, we are justified
in omitting them from the two groups of minerals which are to
be employed in determining the standard mineral composition
of an igneous rock. These two groups of standard minerals are:

GROUP I: SALIC MINERALS.

Quartz, SiO,” - - - : : S = = S i Q
Zircon, ZrO, .SiO2 - - = L c 2 : 2 Uj
Corundum, Al,O3~ - - - - - 3 : é M (e
Orthoclase, K,0.A1,03.6S5i0O,_~—- - erie < == Or

Albite, Na,O.A1,03.6SiO, —- - - 2 : 2 ab} F

Anorthite, CaO. A],03.2510, : - - - - - an

CLASSIFICATION OF IGNEOUS ROCKS 573

Leucite, K,0.Al,03.4SiO, ~~ - - - - - - leq}
Nephelite, Na,O.Al,03.2SiO,_~—- - - - - - ne|
KealiophilitepKeOMAleOn25iOs, 9 = = = i kp -L
Sodalite, 3(Na,O.A1,03.2Si0,).2NaCl oie «ies pe emSOU
Noselite, 2(/Na,O.A1,03.2Si0,).Na ,SO,4 - - - no
GROUP II: FEMIC MINERALS.
Acmite, Na,O.Fe,03.4Si0O, . - - - - - ac)
Sodium metasilicate, Na,O.SiO, - - - - - ns |
Potassium metasilicate; K,0.SiO, - - - - = eeksi| p
Diopside, CaO.(Mg, Fe)O.2SiO,_~—- - - - - di
Wollastonite, CaO.SiO, - - - - - - WO
Hypersthene, (Mg, Fe)O.SiO, - = - - - hy J
Olivine, 2(Mg, Fe)O.SiO, - - - - - - - ol 0
Akermanite, 4CaO.3Si0, - - - - - - am
Magnetite, FeO. Fe,0; - - - - - - - mt
Chromite, FeO.Cr.0O; - - - - . . - cm a
Hematite, Fe,O; - - < - - - - - - hm) i
IImenite, FeO. TiO, De ee il) pe
Titanite, CaO.TiO,.SiO,~ - - - - - - Be eats | 10
Perofskite, CaO. TiO, - - - - . - - pf [.
UMMA OPM ee ea
Apatite, 3(3€aO. P,O;).CaF; - - - - - ap
Fluorite, CaF, - - - : - - - - fr |
Calcite,CaO.CO, - < A 4 2 : Z - CGE A
Pyrite, FeS, - - - - . - - - =) pr
Native metals and other metallic oxides and sulphides. J

Sal, Fem, and Alfer.—For convenience in subsequent discus-
sion, we will anticipate the question of nomenclature, and
introduce here three terms which will be frequently used. To
express concisely the two groups of standard minerals and their
chemical characters in part, the words sa/ and fem have been
adopted. The former is employed to designate Group I,
mnemonically recalling the siliceous and a/uminous character of
its minerals. Fem indicates Group II, since its minerals are
dominantly /errosmagnesian. As adjectives to express these
ideas the words salic and femic will be used. In certain formule
employed later the words sa/ and fem will be used in the sense
just explained, as indicating any one or all of the minerals of
the respective groups. Subsequently other mnemonic syllables

will be similarly treated.

574 CROSS, IDDINGS, PIRSSON, WASHINGTON

The intermediate group c) of aluminous ferromagnesian
and calcic silicates will be designated a/fer or alferric group, this
name recalling the fact that these minerals are characterized by
the presence of a/umina and ferric oxide.

Group [—The minerals of groups a) and 6) have been
united to form the salic group (I) because of the well-recog-
nized relations between the development of quartz, feldspar
and feldspathoids in rocks and the available silica in the magmas ;
these minerals forming frequent series of rocks with a regular
range of silica. It is also done because of the association of
notable amounts of zircon and corundum with these minerals in
the more quartzose and feldspathic rocks. And further it is in
accord with the stronger affinities of the bases, potassium and
sodium, for silica and alumina, which will be discussed later.

The kaliophilite molecule is recognized, since its presence is
necessary in a few magmas so low in SiO, as not to allow the
formation of both leucite and akermanite. Muscovite is omitted
from the list in order to simplify the process of calculation. It
may be considered as made up of orthoclase with corundum and
water. Analcite is also omitted for a similar reason. It may
be considered to be composed of albite, nephelite and water.
Hauynite is omitted because it may be regarded as calcic nose-
lite, and because it introduces needless complications into the
calculation. The standard SO,-bearing feldspathoid is there-
fore considered to be a purely sodic noselite. Cancrinite is
likewise omitted as being in most cases of. secondary origin, as
well as for purposes of simplification.

Group [I—TYhe minerals of groups @) and e) have been
united to form the femic group (II) because of their freedom
from alumina and their association in notable amounts in rocks
low in silica and alumina. They are in this sense antithetical to
the salic minerals.

Wollastonite (CaSiO,) is added to the list of femic minerals
in order to simplify the calculation of standard minerals in
rocks rich in calcium, which may actually enter aluminous
molecules, such as garnet.

CLASSIFICATION OF IGNEOUS ROCKS 575

The simple metasilicate molecule Na,O.SiO,, analogous to
that of wollastonite, is assumed to be present in rocks in which
tere Wiswanwexcess of alkalies: over Al,O, and Fe,O,. It
appears in the arfvedsonite molecule, which develops in such
rocks. But this mineral, being alferric, is not included among
the standard minerals for reasons already given.

In like manner the simplified akermanite molecule, 4CaO.3SiO,,
is included among the standard minerals in place of melilite,
because of the alferric character of .the actual melilite, our
uncertain knowledge of its real composition, and its complexity.
The Na,O and-Al,O, with 2SiO, of melilite are calculated as
entering into a nephelite molecule, the Fe,O, is referred to
magnetite or hematite, leaving the CaO, partly replaced by
MgO, and the remaining SiO, in approximately the aker-
manite ratio of 4:3. The introduction of this molecule into
the calculation is necessary in rocks so low in SiO, that a sub-
silicate must develop.

The molecule (Mg, Fe)O.Fe,O,.SiO,, which is character-
istic of augite, has been omitted for the reason that the ferric
oxide is largely replaced by alumina, and because its introduc-
tion complicates the problem of calculation.

Method of calculation—In order to determine the kinds and
amounts of standard minerals that may express the composition
of a given rock, to establish its place in the system, we may
proceed by a consideration of its chemical composition as given
by chemical analysis, or by a consideration of its actual mineral
composition as determined by optical investigation, or we may
compare it with known rocks whose compositions have been
previously determined.

And since it is advisable to select the same mineral combina-
tion for the basis of comparison of chemically similar rocks, as
already stated, and since there are gradual transitions among
igneous rocks, it is necessary to follow the same method of
procedure in calculating the kinds and amounts of the standard
minerals for all rocks. The method adopted by us is explained
in detail in a later part of this paper. It is based upon certain

576 CROSS, IDDINGS, PIRSSON, WASHINGTON

well-known chemico-mineralogical relations affecting the salic
and femic minerals, both as regards the proportions of their
chemical constituents, the relative affinities of the bases for
silica, and the frequent associations of certain of the rock-making
minerals.

The method has been developed by considering, first, the
chemical composition of igneous rocks, and by devising a plan
for the calculation of standard minerals from it. Then, the
actual mineral composition of holocrystalline rocks has been
taken into account, and a plan devised, after reckoning the pro-
portions of these minerals, for estimating their approximate
chemical composition and from these data calculating the stand-
ard mineral composition of the rock.

It is evident that rocks that are not holocrystalline, or those
in which the proportions of the actual minerals cannot be
determined, must be classified in the first instance by means of
chemical analysis. Subsequently similar rocks may be classified
with greater or less precision by comparison with rocks having
similar textures and the same actual mineral compositions,
which have been analyzed chemically.

In comparing the relative quantities of different minerals in
rocks, either their mass or their volume may be made the basis
of comparison. In calculating the mineral composition from
the chemical composition of the rock, the mass is the natural
unit of comparison. The same is true if mechanical separation
of the mineral constituents is undertaken. When a comparison
is made by the eye, megascopically or microscopically, the basis
of comparison is volume. This may be transformed into mass
by multiplying the volume by the specific gravity of the mineral.
The optical methods of estimation being less exact than those
first mentioned, the basis of comparison should be mass.

CONSTRUCTION OF THE SYSTEM.

It is proposed to arrange igneous rocks by a system which
shall express their quantitative chemical and mineral constitu-
tions. With the known range and degree of variation in these

CLASSIFICATION OF IGNEQUS ROCKS 577

characters it is plain that any system of classification must be
arbitrary.

As has been pointed out, the efforts to express the quantita-
tive variation in rocks by means of several variables have shown
that such a method is undesirably complicated. And we have
sought to find a way for making groups of different taxonomic
value by subdivision in a dichotomous manner based on the
comparison of successive pairs of factors, since the simplest
mode of construction is best.

The mineral constituents of igneous rocks may exist in all
proportions from oO to 100 per cent. whether considered as part
of the whole rock, or as part of the group of standard minerals
to which they may belong. For example one rock may consist
entirely of feldspar, while another has none. And there are
intermediate rocks containing all possible percentages of feld-
spar between these extremes. There are equally wide ranges
for some of the chemical constituents, considered either with
reference to rocks or to mineral groups. Thus the feldspars of
one rock may be wholly alkalic, those of another wholly calcic,
and between these extremes there are all possible gradations.
Among alkalic feldspars are those purely potassic, others wholly
sodic, besides all possible intermediate mixtures of these.

On account of the absence of natural division lines in such
series of two variable factors, it is necessary to establish arbitrary
divisions. We have accomplished this by considering certain
simple proportions as center-points, about which variations may
be allowed within limits, which limits become the boundaries
between petrographical units.

The simplest proportions are: first, those two in which one fac-
tor constitutes the whole, and the other factor is absent; second,
that in which both factors are present in equal amounts.

Other center-points should be selected with equal respect to
these, and may be placed either midway between the three just
mentioned, or at shorter intervals. It has appeared to us best to
select those midway between the first three, namely, at points
representing the proportions three to one, and one to three,
making in all five divisions.

578 CROSS, IDDINGS, PIRSSON, WASHINGTON

It will be noted that the rocks, or mineral or chemical groups,
corresponding to the center-points of these divisions seem to
have a special value as classificatory types, but we wish to point
out that this is not actually the case. Such rocks or such
groups, are not more important, considered quantitatively, than
those occurring in any other part of the system, even on the
boundary between neighboring divisions.

This fivefold method of subdivision may be expressed graph-

ically as follows:

These divisions or units may be described as:
I, One in which the center-point is where the first group is present alone,
8:0.
II. One in which the center-point is where the first group is three times the

second, 6:2.
III. One in which the center-point is where both groups are present in equal

amounts, 4:4.
IV. One in which the center-point is where the second group is three times

the first, 2:6.
V. One in which the center-point is where the second group alone is present,

0:8.

The dividing lines between these center-points will occur at
the following: ratios39 7:1, 5/1313). 5nanGi ia 7a, hes rancesmor
the five divisions are given by the expressions :

ZL NH ZA ERS ZO Bic 8
Alay 6 Gr HA seal
WV) aa aS. rae) Vi) == Sain
mg<sst, we’

The divisions may be defined in general terms as:

CLASSIFICATION OF IGNEOUS ROCKS 579

I. In which 4 is alone present or is extreme/y abundant.

Il. In which A dominates over B.
III. In which both 4 and B are present in eguva/ or nearly equal amounts.
IV. In which B dominates over A.
V. In which B is alone present or is extremely abundant.

The factor which is extreme or dominant over the other, will
be spoken of as preponderant, the other being subordinate.

To indicate whether one of a pair of factors is present in
greater or less amount than one-eighth of the combined pair of

Nee I
factors, that is, = =—, we have adopted the terms zotadble and
BG

negligible. The former means that it is present in greater
amount, and the latter that it is present in less amount, than
that proportion.

It is to be noted that the smallest amount of any factor which
we recognize as ofable in a particular division of the system is
one-eighth of the combined pair of factors, not necessarily one-
eighth of the whole rock, since the pair of factors may be a
fraction of the entire rock.

This fivefold method of subdivision will be further applied
to that factor of a pair which is extreme or dominant over the
other, or, in the case where both are present in equal amounts,
it will be applied first to that one which is deemed to be of
the greater importance, and subsequently, when necessary, to
the less important factor.

In the case of one factor which is subordinate to another,

but not negligible, (< = > =) , such a manifold subdivision is not

advisable, since one-fourth of the factor in question would be
less than one-eighth of the pair, and hence negligible. In this
case we have adopted a threefold division, by retaining the

central division of equal proportions (<25 5), and uniting the
dominant and extreme divisions. In these divisions there is one

where two factors are equal or nearly equal (<> 3), and two

where one factor dominates or is extreme (> 3). Strictly, the

580 CROSS, IDDINGS, PLRSSON, WASHINGTON

subdivision here should be into thirds, but, in view of the con-
fusion liable to arise from the change in proportions the method
adopted seems the most advisable one.

No subdivision need be made of a factor which is present in
negligible amount, the other factor being extreme, since the
whole subordinate factor falls within the allowable limit of
variation.

It may be noted in this place that certain prefixes are used
in connection with mineralogical and chemical terms to indicate
that a factor is extremely abundant, or is dominant. In the first
case the prefix is fer. In the second case it is do or dom. When
comparison is made on a threefold basis, and one factor pre-

dominates over another (> a the prefix is pre.

Since this classification is largely a chemical one, and since
in all the calculations of minerals the molecular amounts of each
constituent only are used, it must be borne in mind that all
chemical comparison is made on the basis of the relative num-
ber of molecules, that is, it is purely molecular. For this purpose
all percentages in analyses must be reduced to molecular
ratios, by dividing each percentage by the proper molecular
weight.

OUTLINE OF THE SYSTEM.
The subdivisions of igneous rocks proposed by us, based upon
the principles discussed in the preceding pages are as follows:
CLASS, SUBCLASS.
ORDER, SUBORDER.
RANG, SUBRANG.
GRAD, SUBGRAD.

The word Rang, which is an obsolete form, equivalent to
Rank, has been chosen instead of Rank to avoid confusion,
since it is desirable that the technical term should differ from
one which is in common use for other purposes. The same is
true of Grad which is an old form of Grade.

The four terms— Class, Order, Rank and Grade —were at first
selected because they are of the same category, the first two

CLASSIFICATION OF IGNEOUS ROCKS 581

being commonly used in classification. But Rank and Grade are
so frequently employed in a general sense that it is advisable to
substitute archaic forms in their stead.

These divisions, which are successively smaller, are based on
characters of the magma of less and less importance. In other
words, the highest divisions express the broadest and quantita-
tively most important magmatic characters, those next to them
less important ones, and so on. It has been our aim to select
the sequence of characters in accordance with this plan, and also
to have homologous divisions throughout the system based on
the same kinds of characters.

The broadest distinguishing chemical characteristics are
expressed by the aluminous non-ferromagnesian minerals and
their associates, quartz and zircon, the sa/ic minerals, on the one
hand; and by the ferromagnesian non-aluminous minerals and
their associates, titanite, apatite, etc., the /emizc minerals, on the
other.

Consideration of the salic and femic mineral groups shows,
however, that the former is more simple in composition than the
latter, so that certain modifications, not of the principles, but of
their application, will be necessary in places.

Thus the salic minerals are composed chiefly of SiO, as
representing the acid radical, with small amounts of Cl] and SO,
in the sodalite group, and of K,O, Na,O and CaO as represent-
ing the bases, these last being always accompanied by an equal
amount of Al,O,, except in the sodalite group, where this is
slightly less than the soda. The alumina, in most cases,
reckoned among the bases, may play the role of acid to some
extent in certain cases. On the other hand in the femic min-
erals, leaving apatite, fluorite, sulphides, etc., out of account,
SiO);, be, O, and 11©, represent the acid radicals (the last two
possibly uncombined with a base as hematite and rutile), and
K,O, Na,O, CaO, MgO and FeO represent the bases.

Consequently, to keep to our system of two factors, already
described, a more numerous subdivision of the divisions in which
the femic minerals preponderate is necessary. But, as will be

582 CROSS, IDDINGS, PIRSSON, WASHINGTON

seen, this can readily be done without transgressing the general
principles on which the system is based or affecting the homol-
ogous characters of the various divisions in different parts of
the scheme.

The principles on which the divisions enumerated above are
made may be stated as follows:

Class.—This, the broadest division, expresses the most gen-
eral chemico-mineralogical character of the magma, and is
therefore based on the relative proportions of the salic and femic
mineral groups as calculated in the standard mineral composition
of each magma. In Classes all the salic minerals calculated for
a magma are contrasted with all the femic minerals.

Subclass —This division is based on certain broad chemical
distinctions in the salic and femic groups, which make it
possible to divide each of them into two parts. These will be
explained in detail when these divisions are described at length
later.

Classes and Subclasses exhibit the broadest and most general
characters of the magma, and are based only on the salic and
femic mineral groups, and the parts into which these may be
most broadly divided on certain chemical lines. More special
chemical characters of the salic and femic minerals in each class
are next to be distinguished, and since both cannot be indicated
at once, according to our principle of dealing with only two
factors at a time, we consider first one group of characters and
then the other. In accordance with the principle of giving
importance to constituents on the basis of their relative propor-
tions, the characters of the preponderant group, salic or femic,
will be considered first until they are fully described, and then
those of the subordinate group, femic or salic, will be taken up.
The divisions from Order to Subrang are based on the chemical
characters of the preponderant standard mineral group, while
the divisions from Grad to Subgrad will be based on the chem-
ical characters of the subordinate mineral group.

Order.—The salic minerals being in the great majority of
cases silicates and quartz, and the silica being the most abundant

CLASSIFICATION OF IGNEOUS ROCKS 583

component of these silicates, it is quantitatively the most
important chemical character of the salic minerals. Moreover
the femic minerals in most rocks are silicates, and in them also
silica is quantitatively the largest factor. Less abundant femic
minerals are ferrates, titanates, and silicotitanates, besides still
less frequent minerals of other kinds.

For these reasons the chemical characters of the salic and
femic minerals of first importance quantitatively are the acid com-
ponents, SiO,, TiO,, Fe,O,, which might be expressed uncom-
bined with bases, or by mineral molecules. In order to express
as far as possible the mineral composition together with the
chemical it is advisable to express the relative amounts of these
acids by means of mineral subgroups.

Orders in Classes with preponderant salic minerals are conse-
quently based on the proportions of quartz to the feldspars, and
of feldspars to feldspathoids, since these proportions correspond
to differences of SiO, in the salic minerals. In Classes with
preponderant femic minerals Orders are based on the proportions
between silicate and non-silicate minerals in the first instance,
and the recognition of the other acid components occurs in sub-
divisions of Orders.

Suborder.—In those Orders in which there are preponderant
amounts of acid components other than silica, the most frequent
of these being Fe,O, and TiO,, we may distinguish the relative
proportions of these in Suborders. They occur in those parts
of the system where minerals of the M subgroup preponderate.

Rang.— Having thus recognized the acid components, with
the exception of the radicals Cl and SO;, which can only be
present to a subordinate extent, we have now to take up the
recognition and expression of the bases in the magma. Conse-
quently, Rangs are formed on the chemical characters of the
bases in the minerals of the preponderant salic or femic group,
according to the Class under consideration. And since there
are several of these bases generally present it is necessary to
treat them in successive groups. The first and most general
division constitutes Rangs.

584 CROSS, IDDINGS, PIRSSON, WASHINGTON

Subrang.—The comparison of those bases which are united
to form the groups of bases recognized in forming Rangs con-
stitutes Subrangs.

Grad.—The characters of the subordinate mineral group,
femic or salic, will be treated in a manner analogous to that
employed when considering the preponderant, salic or femic,
group, the only difference being that as the group under con-
sideration is subordinate fewer distinctions will be needed.

Grads will be based on the general acidic proportions in the
minerals of the subordinate group, and will follow the plan of
the divisions for Order in the preponderant group.

Subgrad.—These divisions will follow the lines of the sub-
divisions for Rang and Subrang, and will express the general
and special chemical characters of the bases of the minerals of
the subordinate mineral group.

Sections.—The application of the above principles shows,
however, that in certain points more numerous subdivisions are
needed. This necessity is met by the formation of Sections of
any of the divisions above described. These Sections will be
based on more special characters according to circumstances.
No general rule can be laid down for them, but they will be
explained in their respective places in the subsequent descrip-
tion of the various divisions.

Family and Series —TYhe grouping of rocks proposed in this
system of classification is for the purpose of bringing together
rocks that are alike chemically, and also mineralogically and
texturally. There is need, however, in the broader treatment of
igneous rocks, especially with reference to their genetic rela-
tions and to their occurrence in petrographic provinces, to group
them in other ways. For these purposes the terms Family and
Series are appropriate, and it is proposed that they be used as
follows:

The term Family may be applied to a group of rocks that
have been developed genetically from a common magma by
processes of differentiation. In its broadest sense it may be
applied to all the rocks of a petrographic province. But it is

CLASSIFICATION OF IGNEOUS ROCKS 585

evident that there may be genetic groups of several degrees. of
consanguinity, for which more specific designations will be
necessary. It is suggested that for these groups terms analog-
ous to Family be used.

The term Sevres may be used, in an extension of Brégger’s
sense," for groups of rocks characterized by similar ratios of cer-
tain constituents, as alkalies, but with varying amounts of other
constituents, as silica. These series will be of very different
characters dependent on the petrographic province, and will
have no place in the system proper, since they will traverse
it in all directions. They may be designated, as at present, by
the use of the names of the extremes, with or without that
of an intermediate member.

CiassEs.—It is proposed to divide igneous rocks into five
Classes, according to the calculated proportions of the standard
mineral groups above defined. As previously stated, these
groups are designated as sa/ and femin formule, and the descrip-
tive adjectives derived from these are given below.

(Css 2

]
ae i , persalic.

This Class contains rocks extremely rich in the salic minerals
—quartz, feldspars, feldspathoids, and corundum.

l
hi i >? , dosalic.

Class IT:
fem I

/\

In the rocks of this Class the salic minerals are dominant and
the femic minerals subordinate.

C1ASS BITTE: cal <@ 5 >= 3 , salfeniic:
fe mipe 3595

In this Class fall the rocks in which the salic and femic min-
erals are equal or nearly equal in amount.

Glass IV: al << 3 S e , dofemic.
fem Gee <7,

* Die Eruptivgesteine des Kristiantagebtetes, Vol. I (1894), p. 169.

586 CROSS, ID DINGS, PIRSSON, WASHINGTON.

The femic minerals are here dominant and those of the salic
group subordinate.
sal

Class Vs ——< e , perfemic.
fem

The rocks of this Class are extremely rich in the femic min-
erals —pyroxenes, olivine, magnetite, etc.

SuBCLASSES.— The next step in the systematic subdivision
takes account of the distinctions among the standard minerals
constituting the salic and femic groups. There being a consid-
erable number of these minerals in each group, it is necessary for
reasons already given to divide each group into two parts. Of
the salic minerals, quartz, feldspar and the feldspathoids form a
closely associated series when considered petrographically and
chemically, and may be contrasted with corundum, zircon or
other salic ‘minerals. (Group 1, therefore, falls) into Part;
quartz, feldspars and feldspathoids, indicated in certain abbrevi-
ated expressions by the letters QO, F and L, and into Part 2,
corundum, zircon, etc., indicated respectively by C and Z.

Part 1. Quartz (Q), Feldspars (orthoclase, albite and
anorthite) (F). Seldspathoids (leucite,

Group I. 4 nephelite, sodalite, noselite) (L).

|
| Part 2. Corundum (C) and Zircon (Z).

Similarly, of the femic minerals, the pyroxenes, olivine,
akermanite, magnetite, ilmenite and titanite are closely associ-
ated in rocks, and form frequent transitional series of rocks with
different proportions of these minerals. They may be grouped
together as Part 1 of Group II, and contrasted with apatite,
fluorite, pyrite, the metals and other femic minerals constituting
Part 2. In abbreviated expressions minerals of Part 1 are indi-
cated by the letters P,O and M, those of Part 2 collectively by A.

( Part 1. Pyroxenes (diopside, hypersthene, wollas-

| tonite and acmite) (P). Olvine and aker-
Group II. J manite (O). Magnetite, hematite, tlmentte,

| rutile, perofskite and ¢titanite (M).

( Part 2. Apatite, fluorite, pyrite, etc. (A).

CLASSIFICATION OF IGNEOUS ROCKS 587

Subclasses in Classes I, II and III are made on a fivefold
basis by considering the relative proportions of the two parts of
the salic group, just described. The reasons for using the salic
minerals in Class III, instead of the femic, are given in the dis-
cussion of Orders. In Classes IV and V Subclasses are formed
on the proportions of the two parts of the femic group.

In Classes I, II and III the divisions will be as follows:

Subclass 1: “= tl
Subclass 2: Sap <i> 8.
Subclass 3: a <i : > : ‘
Subclass 4: —e <i : > : :
Subclass 5: a << ; ;

In Classes IV and V the divisions will be:

Subclass 1: las > i :
A I
Subclass 2: ee 5 .
Srey Tein?
ROM 57. 33
Subclass 2: KOS SS a
; A Ban 85
POM 2 I
Subclass 4: ——<=>-.
A 5 7
Subclass 5: = Ki :

In the first three ;classes the distinctions between corundum
and zircon may be made in those subclasses in which they are
present in notable’amount, by the formation of Sectzons as follows:

Cae,
Section 1: oe ;
Section 2: Ce ee

Tieeetas 61

CROSS, IDDINGS, PIRSSON, WASHINGTON

Ul
ee)
oe)

aE aee Cig Sas
t Bun hae ee ae ta
Section 3 Pie
Cache
Sections:
Jap
Section 5: Sipe
on 5: 5 ae

In Classes IV and V, Sections may be made analogously, when
it is necessary to recognize distinctions between notable amounts
of apatite and various metals and sulphides. They need not be
given here as the principles controlling them are the same as
above.

It will be noted that the vast majority of igneous rocks of all
Classes belong to the first Subclass in each Class. There are few
rocks known belonging to most of the Subclasses here proposed,
but if ever found their classification is thus provided for and
will not disturb that of the rocks already known.

In the remainder of this article the classification set forth
pertains to Subclasses I of each of the five Classes, unless other-
wise stated.

OrpeErS. —The division of Subclasses to form Orders is made
on a basis of the relative proportions of the standard minerals in
the preponderant group.

For Classes I and II the preponderant minerals are salic; and
in Class III salic minerals are considered before femic, and since
minerals of Part I are preponderant over those of Part 2 the
former are made the basis of subdivision, which is as follows:

a) Quartz, Q.

6) Feldspars (orthoclase, albite, anorthite), F.

c) Feldspathoids (lenads) (leucite, nephelite, sodalite and nosel-
ite); le:

Owing to the fact, already discussed, that quartz and the
feldspathoids (lenads) aré in almost all cases antithetical, so that
they do not occur together, these three factors may be employed
serially and in the first three Subclasses of Classes I, II and

CLASSIFICATION OF IGNEOUS ROCKS 589

III the Orders are formed bya double application of the plan of
fivefold division, resulting in nine Orders.

Owing to the cumbersomeness of the phrases necessary to
describe Orders when the mineralogical group-names are
employed quantitatively, it is advisable to use abbreviated terms
mnemonic of these mineralogical groups. For these we suggest
the following syllables: guar, mnemonic of quartz; fe/, mne-
monic of feldspar; J/en, mnemonic of leucite and nephelite,
which is understood to include the other standard feldspathoids,
sodalite and noselite. From these syllables, with the addition
of the proper quantitative prefixes already mentioned, are
formed adjectives descriptive of the several Orders as given
below.

On account of the resemblance between the words feldspar
and feldspathoid, in such frequent use, we suggest for the latter
the term /enad.

Order) 1): = > i ; quartz extreme, perquaric.
eae S :
Order 2:: F <4 . > r , quartz dominant, doquaric.
ee Ole 543
Order 3: F < 3 > << quartz and feldspar equal, quarfelic.
Ox esa or : ‘
Order 4: F << 3 > Ao feldspar dominant over quartz, quardofelic.
L
Order 57: a << . , feldspar extreme, perfelic.
De as aaa ; :
Order 6: =< +=>-—, feldspar dominant over lenad, lendofelic.
Breasu 8
L
Onder. 7: Fr << : > ; , feldspar and lenad equal, lenfelic.
eee Figews aia? :
Order 8: FE <a S: > 3 , lenad dominant, dolenic.
Me) :
Order 9: 7 > Ae lenad extreme, perlenic.

It will be noted that in Class III, in which the salic and
femic minerals are present in equal or nearly equal proportions,

590 CROSS, IDDINGS PIRSSON, WASHINGTON

Orders are established as though the salic minerals were pre-
ponderant. This has been done for the reason that both groups
of minerals being present in nearly equal amounts, it is neces-
sary to select one group arbitrarily for the basis of division to
form Orders. Preference is given to the salic group because the
greater number of rocks belonging to this class, so far as known,
contain slightly more salic minerals than femic, within the range
of 3 and 3, Moreover, the present custom of classifying these
rocks primarily on a basis of the feldspathic constituents may
be allowed to influence the choice. The division of Class III to
form Orders is therefore the same as those in Classes I and II.

For Classes IV and V the dominant minerals are femic, and
the division to form Orders must be based on the relative pro-
portions of minerals of the first Part of this group. They may
be grouped into silicates, 1,and non-silicates with titanosilicates,
2. The silicates are further divided into a) metasilicates, and
6) lower silicates, as follows:
i [ a) Pyroxenes, diopside, wollastonite, hypersthene, and acmite, P.

13) Olivine and akermanite, O.
2 Magnetite, hematite, ilmenite, titanite, etc., M.

Orders in these Classes are formed by comparing subgroups
1 and 2, the silicates and non-silicates. For adjectives to
describe these Orders we suggest the following syllables mne-
monic of the subgroups of femic minerals: fol, to indicate pyrox-
ene and olivine with akermanite; zzz, to indicate magnetite,
ilmanite, titantite and the other minerals of this subgroup. For
adjectives to describe Sections of Orders, based on a comparison
of the two parts of the silicate subgroup we suggest the sylla-
bles: pyr, denoting the pyroxenes, and o/, denoting olivine and

akermanite.

Orders3«: 2 = 2 = i ; femic silicate extreme, ' perpolic.

Order 2: P - O z i S : , femic silicate dominant, dopolic.

Order at P+0O Z 5 = 3 MeN silicate and non- a
M 3° 5 silicate equal, polmitic.

CLASSIFICATION OF IGNEOUS ROCKS 59I

P+O 1 : A
Order’ 4: ali < : >- , femic non-silicate dominant. domitic,
M ay
PO.) 1 : ae
Orders: = <i ai femic non-silicate extreme. permitic.

Sections of Orders—The division of Orders into Sections is
needed in Classes IV and V, to express the relative proportions
of the subgroups of the preponderant mineral groups, that is,
the proportions between the pyroxene and olivine subgroups.

In Orders 1, 2, and 3 of Classes IV and V, where the polic
minerals are extreme, or dominate over, or are equal to the
mitic, this division is carried out ona fivefold basis and results
as follows:

iP
Section 1: r3) > : ; pyroxene extreme. perpyric,
Sas yet ake Ee 5 . '
ectione2': 0 << : > ae pyroxene dominant. dopyric.
Secti : i 3 3 ivi i
on 3: 6 < 5 > a) pyroxene and olivine equal. pyrolic.
; | eed baee ec Lae ‘
Section 4: v7) < ; ac , Olivine dominant. domolic.
i Piast Ha
Section 5: r6) < as olivine extreme. perolic.

SUBORDERS.—In Orders 4 and 5 of-.Classes IV and V the
non-silicate, mitic, minerals preponderate. The subgroups of
these minerals, in greatest abundance and most characteristic of
rocks belonging to these Orders, contain Fe,O, and TiO,. The
first subgroup includes magnetite and hematite, and is indicated
by the symbol H. The second subgroup includes ilmenite,
titanite, perofskite and rutile,and is indicated by T. For these
subgroups we suggest the syllables “em and #7, mnemonic of the
minerals composing then. The relative proportions of these
minerals are recognized in Suborders on a fivefold basis, as
follows :

Suborder 1: ; perhemic.

592 CROSS, IDDINGS, PIRSSON, WASHINGTON

Suborder 2: = x t = : , dohemic.
H

Suborder 3: = < 3 > 3 , tilhemic.
pees nS

Suborder 4: Hi <i 3 > x , dotilic.
Tt etd

Suborder 5: = << , pertilic.

Ranc.— The division for Rang is based on the general chem-
ical character of the bases in the minerals of the preponderant
group in each C/ass, that is, in the salic or femic minerals used
for the formation of Orders.

For salic minerals this is expressed by the terms a/kalic and
calcic, which relate to the feldspars and lenads. Divisions
are based on the proportions of molecules of K,O’ +
Na,O', to CaO”-in these minerals, K, 0"; Na, 0" “and a0,
being the parts of these rock components allotted to salic min-
erals. The divisions in Classes I, II and III are fivefold:

Rang 1: a > ; peralkalic.
Rang 2: —— <! : , domalkalic.
Rang 3: ee < : = : , alkalicalcic.
Rang 4: eee <q : > : , docalcic.
Rang 5: ead Sea < : ‘ percalcic.

In Order 6, where feldspar is dominant over the lenads,
only the first four Rangs will occur, in Order 7 only the first
three, in Order 8 only the first two, and in Order 9 only the
first one.

For femic minerals the general chemical characters of the bases
are commonly expressed by the terms ferromagnesian, calcic, and
alkalic. But three independent factors are not to be employed

CLA SSTFICA TION OF IGNEOUS ROCKS 593

in this system at one time, and it is necessary to combine two of
the three to form a dual basis fora first subdivision. This is
accomplished by uniting the first two and comparing the pro-
portions of MgO, FeO, CaO", and the alkalies, K,O", Na,O".
In this case CaO", K,O" and Na,O", are the parts of these rock
components allotted to femic minerals. For the adjective
expressing the first named quality, corresponding to alkalic, we
propose to use the word murlic, referring to the magnesium,
zron and dime. Upon a fivefold basis of comparison of these
two sets of chemical constituents we form Rangs of Classes
IV and V, although the more alkalic Rangs here provided are
not yet known. They are:

Rang 1: a EEL > i permirlic.
Rang 2: Sao < i > ; , domirlic.
Range 3) Sa < ; = : , alkalimirlic.
Rang 4: SS aa ee << : = , domalkalic.
Rang 5: Se <_or < : peralkalic.

Sections of Rangs in Classes IV and V are necessary to recog-
nize the varying proportions between the ferromagnesian and
calcic characters of femic minerals, since the magnesium and iron
are so closely associated chemically and characterize certain
femic minerals free from calcium. For the word ferromagnesian,
which is used in a somewhat loose manner in petrography, we
propose to introduce the word, muric, to indicate strictly the
magnesium and zvon content in femic minerals, and to be mne-
monic of these two metals.

These Sections of Rangs in Classes IV and V have been
made on a fivefold basis as follows :

(Mg, Fe)O

Section 1: Cad® >2 ; permiric.
a

594 CROSS, IDDINGS, PIRSSON, WASHINGTON

Section 2: (Mentoe Us 2 , domiric.
CaO aga}

Section 3: a <F > : , calcimiric.

Section 4: Ss <s > 5 » docalcic.

Section 5: we SG : ; percalcic.

SUBRANG.— Subrangss are made on a basis of the special char-
acter of the variable or compounded chemical quality used in
forming Rang. In Classes I, I] and III this variable quality is
in the ratio of alkalies, potash and soda, and the following sub-
divisions are recognized for Rangs I, II and III.

Subrang 1: ae a ; perpotassic.
Subrang 2: 0" << Z = : , dopotassic.
Subrang 3: 0" < a ; , sodipotassic.
Subrang 4: woo? ze : 5 ; anata:
Subrang 5: ney? : , persodic.

In Rangs IV and V of Classes I, II and III only three
divisions are used because of the subordinate amount of alkalies.
These Subrangs are:

; Oy oes .
Subrang 1: Na, 0” > A A prepotassic.
OF Se 8
i 4 2 J J
Subang 49 Na, 0” Ke : = ee sodipotassic.
KeO7
Subrang 3: Na,O” < 5° presodic.

In Classes IV and V the variable quality is in the miric:
content. The alkalies are so preponderantly sodic that they do
not require subdivision, and no distinction on‘a basis of the rela-

CLASSIFICATION OF IGNEOUS ROCKS 595

tive amounts of soda and potash are needed. For the present
at.least when femic minerals are said to be alkalic, it may be
assumed that they are presodic. The divisions in Sections 1 and
2 of Rangs I and II are:

Subrang 1: — >? ) permagnesic.
Subrang 2: a= < i De ; , domagnesic.
Subrang 3: = < ; > : , magnesiferrous.
Subrang 4: — =< : > 5 , doferrous.
Subrang 5: = ae F perferrous.

In Sections 3 of Rangs I and II and in other Rangs of
Classes IV and V only ¢hvee divisions are used, because of the
subordinate amount of miric component. These are:

MgO
Subrang 1: rae SF 3 premagnesic.
Subrang 2: a << : = : , magnesiferrous.
Subrang 3: - gO a 3 preferrous.
He@hay 5.

Sections of Subrangs.—In certain Subrangs a division into
Sections is necessary on account of the presence of various acid
radicals or for other reasons.

In Classes I, II and III the salic mineral groups, quartz,
feldspar and lenads, may be considered as: (a) silica and
simple silicates, and (0) silicates combined with some other
salt, or silicates containing some other acid, as Cl and SO,.
The latter are confined wholly to the lenads. This distinc-
tion is advisable here, rather than at Suborders, where it might
otherwise fall, because of the great change in the mineralogical
character of the lenad group induced by the presence of
very small amounts of Cland SO, (often less than 1 per cent.

596 CROSS, IDDINGS, PIRSSON, WASHINGTON

of Cl yielding a notable amount of sodalite), and because, so
far as has been observed, or seems @ priort possible, the minerals
of the sodalite group are only present in the sodic or sodipotas-
sic Subrangs of Classes I, II and III. it may be pointed out
that the sodalite minerals are also characterized by the presence
of more Na,O than Al,O,.

The lenads to be contrasted are the simple silicates, nephelite
and leucite,on the one hand, and the more complex silicates,
sodalite and noselite, on the other. The former may be desig-
nated by the symbols me and /c, and the latter by so and xo,
and the mnemonic syallables indicating these subgroups are ne/
and son.

The division will be needed only in Orders 6, 7, 8 and g,
which contain notable amounts of lenads, and will be five-
fold in the sodipotassic, dosodic and persodic Subrangs of the
peralkalic, domalkalic and alkalicalcic Rangs.

3 ne, le :
Section.) ae ; pernelic.
SOMnO | ait
: ne, lc :
Section 2: a < i ss 5 , donelic.
SO) MO". E93
2 ne, Ic :
Section 3: ad 5 2 , sonnelic.
SO) TOs aes
: ne; lc I :
Section 4: a ees ee , dosonic.
sSouno. 5 99
é ne, Ic I :
Section 5: <i, personic.
SOMO

In the dopotassic Subrangs of the Rangs just mentioned, and
in the sodipotassic and presodic Subrangs of the docalcic Rangs,
the division will be threefold:

: ne, lc :
Section 2 prenelic.
SO, no: | 3
; Ic, ne :
Section 2: : 5 =° , sonnelic.
SO, Or) 3.15
: ne, lc 2
Section 3: —— - presonic.
so,no 5

CLASSIFICATION OF IGNEOUS ROCKS 597

Since we have to deal here with the two acid radicals Cl and
SO,, we may distinguish between the sodalite and noselite rocks
by dividing Sections 3, 4 and 5 of the fivefold division above,
and Sections 2 and 3 of the threefold, on a threefold basis,
which appears to be a sufficient subdivision. These Subsections
will be as follows:

l
Subsection 1: < = : ; prechloric.
: Cl 5 3 :
Sub 2 a <=>, chl Iphic.
ubsection 2 Son ee chlorsulphic
: Cl 3
dee SS Iphic.
Subsection 3 SOF < : presulphic

Rocks belonging to each of the five Classes here proposed
have been fully characterized mineralogically and chemically,
as far as the preponderant salic and femic groups of minerals are
concerned. There remains the consideration of the subordinate
group, femic or salic, in each case, in order to complete the
chemico-mineralogical classification of the rocks.

In Classes I and V since the preponderant group is present
in extreme amount, the subordinate group is negligible, and no
further division is necessary.

In Class II the subordinate group is present in notable
amount, and is femic, while in Class III the femic group is given
second place, after salic, so that in both cases we have to do
with femic minerals, and further division will take place in this
group and will be in general along the lines laid down in the
preceding pages in the description of the divisions from Order
to Subrang in Classes IV and V.

In Class IV the subordinate group is salic and is present in
notable amount, so that further division will take place in this
group, following the lines of division from Order to Subrang in
@lassessl- Il and 11

It will furthermore be seen that, since in Classes II and IV
the subordinate femic and salic minerals respectively are not
equal to the preponderant salic and femic, it will not be neces-
sary in these Classes to make the divisions, at least for the pres-

598 CROSS, IDDINGS, PIRSSON, WASHINGTON

ent, more than threefold, nor will it be useful to make in all
cases the finer distinctions which obtain in the divisions based
on the preponderant group of minerals.

In Class III, on the other hand, where the femic minerals are
equal to the salic, and may in fact exceed them in some cases,
the fivefold subdivision will be employed, and, as far as possible,
the distinctions made in the first divisions of Classes IV and V
will be recognized.

It may be noted here that it appears that in actual practice
these later divisions will be used comparatively seldom in Classes
I] and IV, and it will be remembered that they do not exist in
Classes I and V, but that they become important in the inter-
mediate Class ITI.

Graps are based on the proportions of the standard minerals
of the subordinate femic and salic groups in Classes II and 1V
and of the femic group in Class III.

For Class 1] they.are:

iP
Grad 1: a > : ; femic silicate predominant, prepolic.
Ging +O y 5 S 3 femic silicate and non-sili ee
M Bude as cate equal polinitic.
Gide lee ©) oe Hoe non-silicate predom- 2
M 5 inant, premitic.
In Class III they are:
Grad 1: Z - 2 > i ; femic silicate extreme, perpolic. |
Grad 2: E 4 2 << Z = : , femic silicate dominant, dopolic.
Grew P+0O Z 5 S 3 . femic silicate and non-sili- *
M rages cate equal polmitic.
Grad 4: aa < 5 > : , femicnon-silicate dominant, domitic.
Grad 5: eee < e femic non-silicate extreme, permitic.

CLASSIFICATION OF IGNEOUS ROCKS 599

‘In Class III it is also necessary to discriminate further
between pyroxene and olivine, corresponding to the Suborders of
Orders 1, 2 and 3 of Classes IV and V. This will be necessary
only in Grads 1, 2 and 3 of Class III]. They are:

Bree 5 :
Section 1: rel > ae pyroxene predominant, prepyric.
Secti i Leas beee3 ae

ection 2: re < = Be pyroxene and olivine equal, _ pyrolic
oe eB te ;
Section 3: re Sa ; olivine predominant preolic.

In Class IV the subordinate minerals belong to the salic
group. Their division into Grads is of the same kind as that
forming Orders in Classes I, II and III, but since their amount
is considerably smaller than in rocks of these Classes, the divi-
sion is on a threefold basis, resulting in five Grads, as follows:

Grad 1: > ; 5 quartz predominant, prequaric.
ie One S18 .
Grad 2: 7 <a 3 = ae quartz and feldspar equal, quarfelic.
Grad 3: Q = Ee << : , feldspar predominant, prefelic.
Sra ob. aiS
Grad 4: 7 < 5 > oo feldspar and lenads equal, lenfelic.
Mos ‘ :
Grad 5: F ae ; lenads predominant, prelenic.

SUBGRADS are formed on the general chemical character of
the bases of the minerals employed to form Grads, and bear the
same relation to the latter that Rangs bear to Orders. On
account of the smaller amount of the minerals involved in
Classes II and IV it is not thought desirable to give the distinc-
tions as great a taxonomic value as when the preponderant
minerals are concerned, and so they are made the basis of Sub-
grads rather than of a new taxonomic division.

In Class II the subordinate minerals being femic the chemi-
cal divisions are similar to those forming Rangs in Classes IV

600 CROSS, IDDINGS, PIRSSON, WASHINGTON

and V:. but are on. a. threetold.“basis’ because of: the. smaller
amount of femic minerals, as follows:

2 (Mig hejOri CaO" 2 5 aN
Subgrad 1: K,0" + Na,0" oe ; premirlic.
Mg, Fe)O + CaO" 5 2 Rig tan
Subgrad 2: ( K.O" + Na,0" Xe ; > : , alkalimirlic.
mea Mig e) Or pea Or nas in
Subgrad 3: K,0" + Na,O” ee prealkalic.

In Class II] Subgrads are formed ona basis of the chemical
characters of the femic minerals by analogy with those of Class
II, but because of the greater proportions of these minerals the
division is on a fivefold basis, yielding :

Subgrad 1: Sore ae > : ; permirlic.
Subgrad 2: eare or < ! > : , domirlic.
Subgrad 3: Sere mor = : > : , alkalimirlic.
Subgrad 4: Se — < ae : , domalkalic.
Subgrad 5: SEsrr oR << : peralkalic.

In Class IV the minerals forming Grads are salic, and since
they are subordinate to the femic it is advisable to make only
three divisions instead of five, as follows:
K,0' +Na,O’ 5

Subgrad 1: CaO F prealkalic.
Pri Once Na, Ou 5s :
Subgrad 2: Cad’ < ; S 5 , alkalicalcic.
K5O7-- Na, O% 3
Subgrad 3: 2 Cad’ go aD precalcic.

Sections of Subgrads in Class II] are established because the
amount of femic minerals is equal or nearly equal to that of the
salic, and it is desirable to make further chemical distinctions as
in. the case of Sections of Rangs in Classes 1V and V. The
divisions are the same in both cases, and are based on the pro-

CLASSIFICATION OF IGNEOUS ROCKS 601

portions of miric and calcic constituents. They need only be
used in Rangs I, 2, and 3.

Mg, F dot:
Sectionwct: ae : 3 4 premiric.
Mg, F
Section 2: (Mg, 2° 2 , calcimiric.
CaO Bees
(Mg, Fe)O
SEectiont.2e= — : : precalcic.

Subsections of Sections I and 2 are based on the proportions
of magnesic and ferrous components as in the case of Subrangs
insGlasseslV and Vie They are’:

M
Subsection 1: fe > : ‘ premagnesic.
MgO 3
Subsection 2: oD <e : = : , magnesiferrous.
: MgO
Subsection 3: = a , preferrous.

It is understood, of course, that all the divisions just
described are not needed, as has been explained in regard to
those based on the preponderant group of minerals. Thus a
calcic Subgrad is not needed in a Grad with dominant lenads.
In such cases it is simply inactive.

THE ROLE OF ACTUAL MINERAL DEVELOPMENT AND TEXTURE IN
ROCK CLASSIFICATION.

The methods employed in forming Classes, Orders, Rangs,
Grads, and their subdivisions are distinctly chemico-minera-
logical and chemical, the minerals considered being those
defined as standard. ‘These divisions are applicable to magmas
regardless of their mode of solidification or crystallization. The
classification so far is purely magmatic.

The chief object attained up to this point by the system
proposed is the uniting of all igneous rocks having like chemi-
cal compositions into divisions or groups which conform to our
present conceptions of chemical petrographical unity. And it
has been our endeavor at the same time to prepare such a

602 CROSS, IDDINGS, PIRSSON, WASHINGTON

scheme of subdivision that should demands be made for greater
chemical discrimination by future petrographers, further subdi-
vision may be carried on as an elaboration of the classification
here suggested.

But the problem before us being the classification of igneous
rocks, there remains the method of treatment of the rocks
themselves. This involves the consideration of the minerals
actually present, including the aluminous ferromagnesian min-
erals, which occur in the great majority of igneous rocks and
often form the preponderant constituents. It also involves the
consideration of the texture of the rock, which is perhaps the
most obvious feature, as well as the legend of its physical his-
tory. Theelimination of the aluminous ferromagnesian minerals
from the system of magmatic classification just described is
frankly for its simplification. The exact expression of the
chemical magmatic units which would be required should these
minerals be introduced would be so intricate as to be practically
useless. It is conceivable that an exact mathematical expression
involving all of the variable factors or interchangeable combina-
tions, such as are represented by the compositions of biotites,
amphiboles, and augites, is possible, but it would be beyond the
reach of petrographers generally.

The actual mineral development of an igneous rock is to a
great extent so related to the texture that the two are interde-
pendent, which naturally follows from their both being controlled
to a large degree by the physical conditions attending eruption
and solidification. But sincethese relations as well as the phys-
ical conditions are far from simple, it is therefore necessary, for
the present at least, to treat the mineral and textural develop-
ment of igneous rocks separately.

In petrographical classifications in present use either the
actual composition or the texture is made the basis of subdi-
vision, and the other is used for further subdivision. And it is
recognized that there is variation in either case within the limits
of petrographical divisions, however constituted.

This may be illustrated by the following examples:

CLASSIFICATION OF IGNEOUS ROCKS 603

1. Let us consider the case of somewhat similar rocks
having like textures but somewhat different mineral composi-
tions which have been classed together under one name ona
basis of texture, and subsequently separated on a basis of min-
eral differences, the gabbros. Among several of these rocks
with phanerocrystalline granular texture there are those with
hornblende, or mica, or olivine, and they are known as horn-
blende-gabbro, mica-gabbro, and olivine-gabbro.

2. There is a case of a number of rocks having nearly
the same mineral composition (labradorite, augite, olivine, and
magnetite) and called basalt, which possess quite different tex-
tures, such as evenly granular, ophitic, intersertal, porphyritic,
or microlitic glassy.

We can see strong grounds for both of these methods of
classification, and each may serve an important purpose in petro-
graphical work. There are instances in which one method of
- correlation is more valuable than another. Freedom of choice
should be allowed if consistent with definiteness and stability of
the system of classification.

For these reasons the variable characters of each petro-
graphical unit established by the system of classification we
propose, which characters are results of the conditions attending
solidification, or crystallization, are treated in a different manner
from the chemical characters. They are treated as variable
factors in the system, although inherent and persistent qualities
of a particular rock. They are assigned the role of qualifiers
of the magmatic units wherever there is need of further subdi-
vision. They consequently appear in the production of asystem-
atic nomenclature, which is an integral part of comprehensive
classification.

Actual Mineral Composition.

Let us first consider what minerals may actually be developed
in rocks and what relations they bear to the standard minerals
which are used to express the chemical composition of a magma
unit. And let us assume that the rock is holocrystalline. Those

604 CROSS, IDDINGS, PIRSSON, WASHINGTON

cases in which more or less of the magma becomes glass will be
discussed afterward.

For convenience we introduce two terms which have sug-
gested themselves as substitutes for the cumbrous and oft-
repeated expressions, standard mineral composition (that calcu-
lated from the rock analysis) and actual mineral composition. For
the first we propose the word zorm*, and for the second the word
mode.? Vhe standard minerals which make up the norm are to be
called the xormative 3 minerals, not the zormal ones, since the lat-
ter adjective has the meaning of usual or common.

The questions arise, how far does the mode of a holocrystal-
line rock correspond to the wxorm, and what distinctions are
necessary among rocks belonging to one magma unit because of
differences among them due to their actual mineral composition ?

There are all possible grades of correspondence between the
mode and norm of rocks when the whole range of rocks is
taken into account, but when this inquiry is applied to separate
magmatic units the answer must be modified according to the
kind of magma considered.

For example, some magmas of extremely simple composition
crystallize in one manner only, so far as our present knowledge
goes. Those composed of almost pure silica form only quartz
with negligible amounts of other things. Magmas whose norm
is quartz and orthoclase will form rocks with these minerals as
the mode. Magmas whose norm is a simple rock-making min-
eral, as labradorite or olivine, will crystallize into rocks composed
of one of these minerals only.

Consequently rocks belonging to magmatic divisions char-
acterized by extreme amounts of standard quartz, orthoclase,
albite or nephelite; of the lime-soda-feldspars; of diopside or
hypersthene, or olivine, or magnetite, will have actual mineral
compositions very closely corresponding to the calculated norm.
That is, the preponderant minerals actually crystallized will be

tA rule, pattern, model, authoritative standard.—Century Dictionary.

2 The natural disposition or manner of existence of anything.—J/dzd.

3 Establishing or setting up a norm.—Jdzd.

GLASSIFICGATION OF IGNEOUS ROCKS 605

the same as those calculated from the rock analysis. Variations
may occur in the kind of subordinate minerals for reasons which
will be explained presently.

On the other hand, magmas containing nearly equal amounts
of both salic and femic molecules or dominant amounts of the
latter may develop alferric (aluminous ferromagnesian) minerals
to such an extent as to prevent the crystallization of any appre-
ciable amount of the standard minerals. A well-known example
of this is the hornblendite of Brandberg, in Gran, Norway,’ a
rock of Class III, which crystallized as almost pure hornblende.

A similar magma, however, may crystallize into minerals
corresponding closely to the norm, as in the case of the nephelite-
basanite of Colfax county, New Mexico.

It follows from this that in some divisions of the system the
actual mineral composition of holocrystalline rocks will accord
very closely with their norms, while in others it may or may not.
And in these cases there may be various degrees of correspond-
ence between the mode and norm.

In general a number of degrees of correspondence between
the actual and the standard mineral composition of igneous rocks
may be recognized. They may be expressed as follows:

I. COMPLETE OR ALMOST COMPLETE ACCORD BETWEEN THE
MODE AND NoRM.—In the first case the two correspond exactly
both qualitatively and quantitatively. By almost complete accord
is meant one in which all the principal minerals recognized in
establishing the norm are actually present in the rock in approxi-
mately the same proportions as in the norm, or else standard
minerals not calculated in the norm or non-standard minerals
are present in such small amounts as not to affect the posi-
tion of the rock in the system. In these cases of complete or
almost complete accord the rock may be said to have a normative
mode.

A normative mode may be defined as one in which the actual
mineral composition of the rock is so nearly the same as the
standard mineral composition calculated from the analysis that
either may be used to classify the rock correctly.

t Loc. cit.

606 CROSS, IDDINGS, PIRSSON, WASHINGTON

A special and important case of accord between norm and
mode is that of the highly feldspathic rocks. It will have been
noticed that in the list of standard minerals no reference was
made to the plagioclase feldspars intermediate between albite
and anorthite (oligoclase, andesine, labradorite, bytownite), nor
to the various alkali-feldspars, other than orthoclase and albite,
such as microcline and soda-microcline (anorthoclase). This
is for the following reasons :

The feldspars present in a rock are seldom of only one kind,
several distinct species being present usually, this being true
especially of the plagioclases.

The individual plagioclase crystals are frequently zonal, so
that an exact determination of the relative amounts is a matter
of difficulty if not of impossibility.

When orthoclase, albite and anorthite molecules are present,
in rocks of identical chemical composition, the possibilities of
combination of these as actual minerals are many. Thus we
may have an orthoclase-andesine rock, in which all the albite
has crystallized with the anorthite, or an anorthoclase-labradorite
rock, where some of the albite has crystallized with the orthoclase.

We therefore classify rocks, as far as the feldspars are con-
cerned, by the calculated molecules of orthoclase, albite and
anorthite, irrespective of their actual mineral combinations. The
proportions of these will give the mean composition of the feld-
spars. Any of the possible combinations will then be normative
in the sense above, provided that none or only a negligible
amount of the feldspathic molecules crystallize as lenads.

If it should be desired to indicate in the name which combi-
nation is actually present, this can be done by the use of the
appropriate feldspar name, as described later. Since, however,
the albite molecules crystallize with those of anorthite, if these
are present in any quantity, it will be necessary to distinguish
only the cases in which the albite molecule has crystallized with
that of orthoclase, as soda-orthoclase.

2. APPRECIABLE DIFFERENCE BETWEEN THE NORM AND MODE.
—Abnormative minerals (whether standard or not) are in such

CLASSIFICATION OF IGNEOUS ROCKS 607

cases present in quantities which are not negligible, so that the
rock cannot be classified directly from the minerals actually
present without consideration of the mutual changes involved
in the diverse crystallization of the various minerals possible.
Such modes may be called abnormative.

The abnormative modes may be referred to two kinds, of
which the second is by far the most important and common,
though the first naturally precedes it in the discussion.

a) The minerals present may be all standard, but differ from
those of the norm either in kind or relative proportions, or both.
This case is generally due to shifting of the SiO, and TiO,, and
may be illustrated by these instances: quartz and leucite crys-
tallize instead of orthoclase, albite instead of nephelite with
leucite, as in leucite-basanites and tephrites, quartz with olivine
as in quartz-basalts, hypersthene with nephelite, titanite instead
of perofskite or ilmenite, the silica in this case coming from
normative feldspar molecules. The normative corundum may
also form muscovite with molecules of orthoclase and water.

These cases, it will be seen, affect the mineral molecules of
oxides capable of combining in more than one proportion with
SiO,, namely K,O, Na,O,(Fe,Mg)O. All of these modifications
are within the range of the standard salic and femic minerals,
and if not accompanied by notable amounts of non-standard
minerals (as augite or hornblende), may be called respectively
salic, femic, or salfemic abnormative modes, according as the
abnormative minerals belong to one or the other or both groups.

6) The mode may differ from the norm through the presence
of alferric (aluminous ferromagnesian) minerals, augite, horn-
blende, biotite, garnet, etc.

As is explained in the Part on Calculation, and as is evident
on consideration of the subject, the formation of any of these
minerals involves changes in some, if not all, of the standard
minerals calculated as present in the norm. While minerals of
both the standard salic and femic, and non-standard alferric
groups are involved, yet it will be sufficient to characterize these
modes as aferric.

608 CROSS, IDDINGS, PIRSSON, WASHINGTON

In both the above cases a and @ it is evident that, given the
knowledge of the norm and the mutual interrelations between
the rock-making minerals which will be explained in the Part
on Calculation, it is only necessary to know the amount of
abnormative minerals, whether standard or not, to be able to
ascertain the changes from the norm which their presence
involves. These distinguishing minerals therefore may be
called critical ones.

The question of how far quantitative distinctions based on
the amounts of the critical minerals are to be introduced into the
classification 1s discussed in connection with the nomenclature.

Having thus explained the various relations possible between
norm and mode, it will be well to summarize the correspondence
observed in fact among the different Classes. It is obvious that
no one systematic method of calculation of mineral composition
from the chemical can correspond with all the varying possibili-
ties of actual mineral development. The method adopted by us,
however, being founded on the most frequently observed min-
eral relations in igneous rock and on the most generally appli-
cable principles, should yield norms which correspond with the
mode in the majority of cases.

This is a point which can only be tested by appeal to the
literature of petrography, and comparison of the calculated
norms with the modes as furnished by the descriptions. We
have, it may be stated here, tested our proposed method in
regard to this point very thoroughly upon many hundreds of
rock analyses, and especially by means of a collection made by
one of us (H. S. W.) of all analyses of igneous rocks which
have been published since 1883, amounting to over three
thousand."

This comparison shows that the accord between norm and
mode is complete or nearly so in the very great majority of

tIt is hoped that this collection, which will show the norm in every reliable case,
will appear shortly after the publication of the present paper, and it will serve as a
general illustration of the system. Although not as yet available to others, it has been
appealed to as a practical check throughout our discussions, so that the system cannot
be regarded as merely academic, but as having been tested at every possible point.

CLASSIFICATION OF IGNEOUS ROCKS 609

rocks of Class I and in the greater portion of those of Class II,
these two classes comprising over three-quarters of the known or
analyzed igneous rocks of the globe. The accord is also very
good in Class V, which includes but few rocks. Classes III and IV
show less constant accord, as is to be expected in view of their
composition. The results of our proposed method of calcula-
tion may therefore be regarded as satisfactory.

VARIETIES.— It is evident that within the range of possible
variations in each of these modes for a given magma unit, there
may be mineralogical distinctions among the quite subordinate
components, and that it may be desirable to recognize these
differences as distinguishing features of rocks. The method of
accomplishing this is by considering the several subdivisions
upon this basis as Varieties, more specifically Modal Varteties.

A Modal Variety of a petrographical unit of whatever degree,
such as Class, Order, Rang, etc., may be defined as a rock hav-
ing a mode with a slightly different development of the quite
subordinate component minerals.

For example, a rock belonging to Class I composed of an
extreme amount of feldspar (5th Order) may have to per cent.
of nephelite actually crystallized, whereas the norm contains no
nephelite, the difference having been brought about by the
development of a small amount of hypersthene or of hornblende
instead of olivine.

The development of a small amount of hornblende, or of
biotite, in a rock with a normative mode also constitutes a Modal
Variety. The presence of small (not notable) amounts of rare
or otherwise noteworthy minerals may also thus be recognized
as varieties.

Indeterminable Modes.— Rocks not completely crystallized,
containing glass base, and those which for any reason do not
permit the determination of all the component minerals, must be
classified in the first instance in each case by comparison of the
observed minerals with the norm calculated from the chemical
analysis of the rock, and subsequently, by analogy with holo-
crystalline rocks of similar chemical composition whose texture

610 CROSS, IDDINGS, PIRSSON,. WASHINGTON

is somewhat like that of the rock in question and whose actual
mineral components can all be determined.

Texture.

The terms structure and texture are commonly used in petrog-
raphy as synonyms, although efforts have been made to dis-
criminate between them. It seems to us desirable to employ
them in different senses, and we propose to limit the use of the
term structure to those large features of rock bodies which are
known as columnar structure, spheroidal parting, platy parting,
bedding, brecciation and others. And we use ¢exture for the
material features of rocks exhibited by the mineral components
and by the groundmass of dense or glassy rocks, whether they
are viewed megascopically or microscopically. These features
are the expression of the mutual relations of the mineral par-
ticles or vitreous portions of rocks.

The texture of igneous rocks is one of their most variable
characters. It depends in very large degree upon extraneous
conditions influencing the consolidation of the magma and in
far less degree upon certain relationships between the forms of
the rock-making minerals. The factors influencing it are many.
Nearly the whole range of textures may be developed in rocks
from a single magma consolidating under different conditions.
Consequently the use of texture as a primary factor in the classi-
fication of igneous rocks can only result in the wide separation
of things alike in chemical and mineral characters, which, on the
grounds discussed in the preceding pages, are believed to be of
greatest systematic importance.

Its use as a prominent factor in classifications has been in
most cases coupled with assumptions or assertions known to be
by no means strictly in accordance with facts. When such a use
of texture is to be made, it is necessary to assume that all rocks
possess some one of the few textures at present employed in
classification, such as the granular, porphyritic, trachytoid,
glassy, etc. This further requires that these textures be so
broadly defined that they are robbed of much of their natural

CLASSIFICATION OF 1GNEOUS ROCKS 611

descriptive value, or that the definitions be so restricted that they
are not appropriate to many of the rocks to be classified.

It has been common to assume a fictitious dependence of
texture upon geological occurrence, ignoring other conditions
which are plainly of much influence; as, for example, the
assumption that granular rocks are plutonic or deep-seated, or,
from the converse point of view, that effusive rocks must be
partially glassy, or trachytic, or porphyritic.

Less often definitions of texture have involved assumptions
of their origin only partially in accord with the facts.

In some cases rocks have been classified on a basis of only
part of their texture, as when in certain porphyritic rocks the
texture of the groundmass alone has been taken into considera-
tion.

It appear to us that the use of texture in previous petro-
graphic classifications has, on the one hand, weakened the
systems by rendering them unnecessarily artificial and illogical,
and, on the other hand, has prevented the natural application of
certain textural terms. The classificatory role assigned to texture
in this system is that of a qualifier of any of the various
divisions based upon chemical and mineral characters. We
present certain general considerations of rock textures which
may aid in securing for the various factors which enter into
texture a recognition which it seems to us they have not thus
far commonly secured. The discussion does not profess to be
complete as to all rock textures but simply takes up some of the
more prominent textures and discusses their essential properties
and relations.

When we consider the texture of igneous rocks we find that
it may be separated into three factors, which though not in all
cases absolutely distinct from one another are so to a very con-
siderable-extent. They are;

I. Crystallinity. The degree of crystallization.

II. Granularity. The magnitude of the crystals.
III. Fabric. The shape and arrangement of the crystalline and non-
crystalline parts.

612 CROSS, LDDINGS, PIRSSON, WASHINGTON

Various textural modifications of igneous rocks may be dis-
tinguished from one another, but it is understood that in this
respect, as in all others, there are all gradations between different
textures. Moreover, all known textures are not developed
within every known magma unit, and some textures are specially
frequent in rocks of particular compositions.

I. CrysraLuinity.— The degree of crystallization attained by
an igneous rock is measured by the relative amounts of crystal-
lized and glassy portions. | Perfectly glassy rocks are very
uncommon, since most obsidians abound in microscopic crystals,
while many lavas that are almost completely crystallized contain
small amounts of glass. Most rocks are holocrystalline, and all
gradations between the extremes exist.

Distinctions that have been based on crystallinity are of two
sorts: one an absolute distinction based on the known absence
or presence of glass base or matrix, resulting in (a) holocrystal-
line, (6) hypocrystalline, or partly crystalline, and (c) holohyaline, or
completely glassy, rocks. The other sort of distinctions are less
definite, being based on the megascopic appearance of the rock,
but they are of great practical value. They are (a) phanero-
crystalline, (6) aphanitic, and (c) vitreous.

Phanerocrystalline rocks are generally holocrystalline.

Aphanitic rocks are in some instances microscopically holo-
crystalline, in others hypocrystalline. Because in some cases it 1s
not possible without microscopical study to determine the
presence or absence of glass matrix, it is extremely useful to refer
a rock to this textural division. It is discussed again in the
next phase of the subject.

Vitreous rocks are hypohyaline or holohyaline.

II. GRANULARITY is the quality based on the absolute size of
the crystals, which, when all igneous rocks are considered, range
from microscopic sizes to megascopic ones measured in feet.
The quality derived from the relative sizes of the crystals in one
rock may be considered as a phase either of granularity or of
fabric,

Distinctions based on differences in the absolute size of the

CLASSIFICATION OF IGNEOUS ROCKS. 613

grain of rocks have never been sharply defined, because the
crystals in an igneous rock are never of uniform size. They
vary among themselves considerably in most cases, and it is the
average size which is in mind when reference is made to the
grain of arock.

The most important distinction made on this basis is for
practical purposes, namely, that between phanerocrystalline
(phaneric) rocks, whose crystals may be seen by the unaided
eye, and aphanitic or cryptocrystalline rocks, whose crystals are not
apparent megascopically. The term aphanitic is the better one
to use in most cases, because the term cryptocrystalline implies
the holocrystalline character of the rock, which may not have
been determined. This fact is well expressed by the negative
term aphanitic.

_Phaneric (phanerocrystalline) rocks are commonly divided
into coarse-grained, medium-grained, and fine-grained. But for
these terms no limits have been set except those suggested by
Zirkel,* namely, that the first include all those whose average
grain is larger than the size of peas; that the second range from
the size of peas to that of millet seeds; and that fine-grain
include those of the size of millet seeds and smaller. Translat-
ing this into millimeters we would suggest the following scale:

Fine-grained rocks, average grain I millimeter or less.

Medium-grained rocks, average grain I to 5 millimeters.

Coarse-grained rocks, grain more than 5 millimeters.

Aphanitic rocks have never been systematically divided ona
basis of the size of the microscopic crystals, though terms bor-
rowed from the megascopical nomenclature have been adopted
to fit the case. These are: microcrystalline, which is equivalent
to microphanerocrystalline, and applies to textures whose grains
can be seen with a microscope; and mucrocryptocrystalline, which
includes textures whose grains cannot be seen with a microscope,
but which are recognized as present by the exhibition of aggre-
gate polarization between crossed nicols.

In place of distinctions corresponding to coarse, medium,

t Lehrbuch der Petrographie, Vol. I, p. 456, 1893.

614 CROSS, IDDINGS, PIRSSON, WASHINGTON

and fine-grain, which are useful when megascopically applied, it
is possible to describe the average measured diameter of micro-
scopic crystals. But there appears to be no demand for system-
atic subdivisions based on the magnitude of these minute
textures:

Textural distinctions based on the relative sizes of the crys-
tals in one rock recognize (1) evenly granular or non-porphyritic
rocks, and (2) porphyritic rocks. These distinctions cannot be
sharply drawn, because, as already said, it almost never happens.
that all crystals in a rock are the same size. Moreover, there is
a great variety of ways in which differences of size in crystals
may exhibit themselves,

The difference of size which constitutes porphyritic texture
may be defined as that in which the size of some crystals is so
much larger than that of a sufficient number to form a matrix
for the first, that the larger ones appear notably distinct from
the matrix or groundmass. There are porphyritic rocks with
phaneric (phanerocrystalline), aphanitic, or glassy, ground-
masses, the phenocrysts being megascopically notable. And there
are microporphyritic aphanitic or glassy rocks, in which the
phenocrysts are not megascopically notable, although they may
be visible.

Granular and porphyritic textures may also be considered as
modifications of the fabric.

III. Fasric.—The arrangement of the crystalline and non-
crystalline parts of a rock, or its fabric, as we propose to call it,
is dependent on (1) the shapes of the crystals, and (2) their
positions with respect to one another and to the glass base when
present:

When considered with reference to all the minerals in the
rock the fabric of most rocks is so complex that no simple
expression is applicable to it, and nearly all the formal relations
which it is customary to distinguish as different rock textures
frequently exist in one rock. However, it usually happens that
certain features dominate in each case while others may be quite
insignificant though present. Hence we may again disregard

CLASSIFICATION OF IGNEOUS ROCKS 615

those which do not occur to a notable extent and use those which
predominate as the means. of discrimination, that is, as a basis
for textural subdivision.

1. As to the shape of the constituent crystals, there are two
distinctions of a general nature:

(a) Automorphic* (idiomorphic) forms, in which the crystal
form has been developed more or less perfectly.

(6) Xenomorphic* (allotriomorphic) forms, in which the proper
crystal form has not been developed owing to interference from
outside influences.

These latter forms are of two kinds: one caused by the
interference of adjacent crystals, the commonly recognized case
in holocrystalline rocks; the other is caused by outward forces
acting in the liquid from which crystallization has taken place,
which produce irregularly shaped crystals peculiar to micro-
scopic growths in the more siliceous glasses, but not confined to
them.

Other distinctions of shape depend on the particular forms
of the crystals, whether equidimensional, tabular, prismatic, or
otherwise. |

2. The arrangements or positions of crystals with respect to
one another may be grouped under several heads:

A. Fuxtaposition.—In this case, although subordinate minerals
may be enclosed in the preponderant ones, the preponderant
minerals are adjacent to one another, and the following divisions
based on the shape of the crystals are recognized:

Granular, in which adjacent crystals have nearly the same
size, so that the rock appears to be made up of more or less
uniform grains. According to the form of the individual crys-
tals the rock is said to be:

a) Xenomorphic granular, in which the crystals are xenomor-
phic and equidimensional. This is commonly called ‘‘granitic”’
fabric.

*We adopt the use of automorphic and xenomorphic instead of zdiomorphic and
allotriomorphic, as they have undoubted priority, Rohrbach having used the former in
1886 (7. M7. P. M7, Vol. VU, p. 88), while Rosenbusch introduced the latter terms in
1887. Rohrbach’s terms have the further advantage of being shorter.

616 CROSS, IDDINGS, PIRSSON, WASHINGTON

b) Hypautomorphic granular, in which some of the crystals, or
parts of some crystals, are automorphic, while others are xeno-
morphic.

c) Panautomorphic granular, in which all the crystals possess
their proper forms more or less perfectly. Of course absolute
automorphism cannot exist in a compact, continuous rock. The
precise character of this fabric depends on the kinds of minerals
present and on their particular habit. It is, therefore, different
in rocks of different compositions so far as observed.

Tabular or prismatic fabric—There is further modification of
it in case the shapes of some of the minerals are ¢adudar or pris-
matic. In such cases there may be distinctions according as there
is: no definite arrangement of plates or prisms, which stand in
all positions; or a regular arrangement in more or less parallel
directions (fluzdal or parallel fabric), or in more or less radiatng
directions (vadiate fabric).

B. Interposition (inclusion).—The arrangements of crystals
referred mainly to inclusion of one by another are of two princi-
pal kinds:

a) Graphic fabric—One in which two minerals mutually
inclose one another, by interpenetration. This is shown by the
parallel orientation of several parts of each mineral. The famil-
iar example is the grap/ic intergrowth of quartz and feldspar.

6) Potkilitic fabvic—The second kind of interposition fabric
is that in which one mineral acts as a matrix for one or more
kinds of other minerals which do not possess parallel orienta-
tion. This is exhibited by various rocks, in some of which horn-
blende crystals form the matrix, in others orthoclase, in others
quartz.

Ophitic fabric is a special case of poikilitic, in which plagio-
clases are inclosed in augite crystals, the feldspars being rela-
tively large when compared with the areas of augite.

C. Porphyritic fabric is one characterized by the presence of
crystals surrounded by a matrix noticeably distinct from them.
These crystals are known as phenocrysts. The matrix or
groundmass may be crystalline or glassy. The phenocrysts

GLASSIFICATION OF IGNEOUS ROCKS 617

are megascopic in some cases, and microscopic in others. When
they are microscopic in a glassy matrix the fabric is called scro-
litic glassy.

D. Orbicular fabrics —There are complex fabrics so frequent
and so characteristic that they have formed bases for textural
divisions. They are:

a) Spherulitic, allied to the microlitic fabric in some cases in
that it can be traced to special forms of microscopic crystalliza-
tion, consisting of radiating prisms. It is further allied to por-
phyritic fabric, in that it may be developed locally in the magma
and produce scattered spherulites of notable size, and having the
appearance of phenocrysts.

6) Spheroidal, closely like spherulitic in some cases, where
there is crude radial arrangement of crystals, but quite different
from it in others in which there are concentric granular shells. A
somewhat diverse form of aggregate crystallization occasionally
developed in phaneric rocks.

HETEROGENEOUS TEXTURES produced by more or less hetero-
geneity in the magma are distinguished by marked variability of
(a) fabric, or (0) composition.

a) Eutaxitic texture —Variability of fabric is exhibited by
some banded or streaked lavas (rhyolites) in which alternate
layers of rock exhibit different degrees of crystallinity or dif-
ferent arrangements of crystals. Variability of texture on a
large scale is often exhibited by phaneric rocks of similar com-
position (granitic pegmatites).

6) Variability of composition may not strictly belong to the
discussion of the texture of rock, but the two are intimately
blended, and the treatment of the matter in this place may be
justified. It results in danded textures, approaching gneissic, and
is exhibited in certain gabbros; also in irregularly streaked tex-
tures, known in German as Schlieren, which may be called
schheric textures.

In classifying rocks ona basis of textural differences we have
considered the fabric, or the shape and arrangement of the crys-
tals, as more fundamental than the crystallinity or granularity,

618 CROSS, IDDINGS, PIRSSON, WASHINGTON

for the reason that it is more clearly a function of the composi-
tion of the magma, since it depends upon the shapes of the
crystals which are often characteristic of particular minerals
which preponderate in rocks of certain compositions.

On the other hand it is possible for any magma to attain any
degree of crystallinity, although all degrees, from glassy to
holocrystalline, are not known for all kinds of magmas. In the
same way any magma may develop any grade of granularity,
although some grades are oftener observed in certain kinds of
magmas than in others.

The method of recognizing these differences of texture and of
introducing them into the classification is referred to the nomen-
clature, where it will be fully stated.

PART ii NOMENGCEATURE,

Having outlined the system of classification proposed by us,
it is necessary to present and discuss the nomenclature which we
have constructed for its expression and use.

As already pointed out, the present system of nomenclature
is most unsatisfactory, for the reasons that it furnishes no dis-
tinctions between names of different values (one termination
alone being regularly used), nor indication of the relations of
any of the groups.

Furthermore, since the system of classification here proposed
is based on relations and principles quite distinct from those in
present use, the old names cannot be employed for the new
divisions. This is obvious for the reason that the new divisions
do not cover the same concepts or characters as the old ones,
and because the use of an old name with a new meaning is to be
shunned as leading to grave misunderstanding and confusion.

Rock names, like those of any other system of nomenclature,
are, or should be, composed of two parts, with distinct functions,
viz., the dody or root, and the termination. The formerexpresses
either directly, or byimplication or connotation, the character of
the object or group of objects to which it is applied, and the
latter indicates the place of the object or group in the co-ordi-

CLASSIFICATION OF IGNEOUS ROCKS 619

nated system, from the broadest to the smallest division. Each
separate division, from the largest to the smallest, should have a
distinctive appellation, precluding confusion with others, and
indicating at once its relative place in this system and its char-
acter. The names should be mnemonic, if possible, expressing
the idea to be conveyed, short, euphonious, and not liable to be
mistaken for one another.

In accordance with the principles just stated, we have con-
structed names for the various magmatic divisions which have
been explained in the preceding pages on classification. In
addition to these it will be necessary to employ qualificatory
terms to express the variable characters of mineral composition
and texture, which, with the appropriate magmatic name, will
apply to the rocksthemselves. The nomenclature, then, will be
to a certain extent polynomial. These qualifiers may be applied
at any point in the system.

MAGMATIC NAMES.

TERMINATION.—It is possible to indicate the place of any
given rock name ina system by employing only one termination
and by varying the root of the word, which then alone indicates
atthe same time both the character of the group and its relative
position in the system. Thus, with the single termination -zte
the family name, foyaite, expresses the concept of rocks formed
essentially of alkali-feldspar and nephelite, while smaller divi-
sions of this are indicated by such names as laurdalite, litchfield-
ite and lujavrite.

But this method is very unsatisfactory and primitive, in that
it does not avail itself of the very powerful aid afforded by such
possible terminations as would in themselves give considerable
knowledge of the group named. Furthermore, while feasible
with a small number of names, such a method involves a great
tax on the memory, especially with the rapid growth of our
knowledge of rocks and of new rock types, and the necessity
for new names.

These serious objections to the use of but one termination

620 CROSS, IDDINGS, PIRSSON, WASHINGTON

have so far been scarcely recognized, and this great defect in
present nomenclature is one cause of much of the prevalent out-
cry against the multiplication of rock names.

To indicate the relative place of the various divisions in the
system, from Class to. Subgrad, we have adopted a set of termi-
nations which are to be used invariably with their respective
divisions. We have endeavored to select those which are at the
same time: mnemonic in suggesting the relative positions of the
divisions ; euphonious ; not in previous use to any great extent ;
and as far as possible adapted to use in all European languages.
The termination -ite is rejected because it is already in use, not
only in petrography but in mineralogy, to which latter science,
as having a prior claim, we suggest that it be restricted. The
terminations proposed are as follows:

Class, -ane. Subclass. -one.
Order, -are. Suborder, -ore.
Rang, -ase. Subrang, -ose.
Grad, -ate. Subgrad, -ote.

These terminations were selected after trial of many that were
suggested. “Phe distinctive “consonants, 7, %7,s; 7, are in their
alphabetical order, suggesting the sequence of the taxonomic
divisions to which they refer. The vowels, @ and 0, were con-
sidered best to indicate the distinctions between each principal
division and its subdivisions. And the final e¢ is added in
English to lengthen the sound of these vowels, but it may be
omitted when the words are used in other languages.

The name of a magma belonging to a Section of any given
division will receive the termination of that particular division
preceded by the letter z, since it occupies the same relative posi-
tion in the scheme as the division in question, although of some-
what different taxonomic character.

Roor.—In choosing the root of the name two methods are
available. This part may bea syllableor syllables derived from
local or personal names or chosen arbitrarily, giving only by con-
notation an idea of the character of the group named; or it
may be composed of syllables which will of themselves, by

CLASSIFICATION OF IGNEOUS ROCKS 621

proper selection on mnemonic grounds, give an idea of t h
character.

The latter method would seem at first sight to be the better,
since it apparently involves less tax on the memory. We have,
however, made many attempts at the construction of such name
roots, giving them practical trials at all points of the scheme, but
have been forced to reject this method for all divisions except
that of Class, for the following reasons :

1. The names are apt to be excessively long and cumbrous,
especially in the smaller divisions.

2. While the distinctions can be readily seen on paper, yet
close attention must be paid to each syllable and even to each
letter, and in spoken language the names are so similar as to be
confused with each other.

3. Asa consequence, the mnemonic quality, which is the great
theoretical advantage of this method, is very seriously dimin-
ished, especially in the smaller divisions.

4. The adoption of this method involves a fixation or lack of
elasticity in the nomenclature, which would be fatal to it, if the
need arose for any change either in the method of arriving at
the divisions or in their number, since this would involve a cor-
responding change in the nomenclature.

We have, therefore, except for Classes and Subclasses, adopted
roots derived from names of localities, taking advantage of the
fact that there are at present many of these in present petro-
graphical nomenclature with connoted magmatic ideas which
are readily adaptable to our purpose. Thus with the root
nordmark is connoted the idea of an alkali-feldspathic rock,
with mask that of a rock composed essentially of alkali-
feldspar and nephelite. These are, it is true, not the only con-
notations of these names, since in both of them are also
implied ideas of texture.

In the names suggested subsequently we have endeavored to
adhere to the following principles in the selection of roots.

a) They should be adjusted to the relative position of the
divisions to which they are applied. That is, the roots of names

622 CROSS, IDDINGS, PIRSSON, WASHINGTON

which are at present in use for large rock groups, or which were
proposed by the author for such groups, are used for the larger
divisions, such as Rangss, while those which have at present, or
were originally given, narrower or more specific character or
application, are used for the smaller divisions, Subrangs, Grads,
and Subgrads.

Inasmuch as few if any of the rock names at present in use,
even the broadest, cover the wide mineralogical concepts
involved in our proposed Orders, and as it will also be well to
make the names of these broad groups quite distinctive, we
have adopted as the source of the roots for Orders the names of
countries. We have endeavored to select in each case the name
of a country where the respective Order is found in especial
abundance or is most typically represented, or which is histori-
cally connected with the Order. This is, of course, not possible
in every case, as some of the most abundant Orders are repre-
sented in several countries. In such cases we have endeavored
to apportion the ordinal roots equitably among the petrograph-
ically prominent nations.

6) The selection should be based on rocks of which good
analyses exist. This is obviously just, since a poor analysis is of
little use, and is often of positive harm, to the science and to
any system of classification, and furthermore it will frequently
happen that a name based on such will prove eventually to have
been misapplied.

c) Alocality root already in use should be employed if possible.

d) As far as is consistent with the preceding principles, the
laws of priority will hold good, as usually recognized and as
formulated in Dana’s System of Mineralogy.

e) In selecting a new root from several possible localities,
the one should be chosen which is best known, has been best
described, or which furnishes the most typical material.

f) The roots should be short, if possible not more than two
syllables, and euphonious.

g) The name of a locality where the mode is normative is
to be employed if possible.

CLASSIFICATION OF IGNEOUS ROCKS 623

h) As far as possible localities whose rocks occur near the
classificatory borders of a given group are to be avoided.

Intermediate magmatic names.—A rock, whose composition is
such that its magma occurs so near the border line of any divi-
sion of the system as to be considered a transition magma,
receives a compound magma name, made by uniting the names
of the two divisions concerned, and connecting them with a
hyphen. . The name of the division in which the rock occurs is
to be preceded by that of the neighboring division.

Names.—As the Classes are few in number the strictly mne-
monic method of root-formation is applicable to them, and we
have accordingly named them on the basis of the relative
amounts of the standard salic and femic minerals present, these
being indicated by the syllables sa/ and fem, as already explained.
The relative amounts are shown by the prefixes per- and do-,
which stand respectively for extremely abundant and dominant.
The names for the five classes are as follows :

1. Persalane: Extremely salic— De
Fem I
: ; Sal Tees
2. Dosalane: Dominantly salic >
18 Sinay Fuses
Salfemane: Equally salic and femic — D2 353
3: ae Fem) > 3ng@icee
S
4. Dofemane: Dominantly femic— eae
Bema 50a
Perfemane: Extremely femic eal < :
* y Kem. 700

For other magma units names derived from geographical
localities have been used, as already explained. Those sug-
gested by us apply to the more important divisions of the sys-
tem, and to those based on the best established data, derived
from the collection of several thousand analyses already referred
to, and from older analyses where available. There are numer-
ous divisions for which we have not sufficient data to warrant
our suggesting names at this time. Owing to the large number
of names selected, they are presented in tabulated form, in such
manner that they can be understood without difficulty.

624 CROSS, IDDINGS, PIRSSON, WASHINGTON

A list of all new words proposed by us, with their definitions,
will be found in a Glossary which will appear with a reprint of
this essay, to be published in book form by the University. of
Chicago Press. The Glossary and the Tables to aid the calcula-
tion of norms are too voluminous to be printed in this JOURNAL.

ROCK NAMES.

As just explained, the names to be applied to the systematic
divisions, or magma units, are formed from names of geographi-
cal localities modified by a series of terminations, and designate
rocks so far as their chemical composition and standard mineral
composition are concerned, but do not indicate their actual min-
eral composition or their texture. The methods of expressing
these latter characters in the systematic nomenclature remain to
be stated. As already said, each is treated independently of
the other. We take up first the expression of the actual mineral
composition.

ACTUAL MINERAL COMPOSITION.—A nomenclature that will
express the mode of a rock must indicate the kinds of minerals
present and their quantity. The question then arises: How far
is it necessary to modify, or qualify, the magmatic names in
order to accomplish this, and further: What exactness is desir-
able in expressing the quantity of the actual minerals present?

When the name of the magma unit to which a rock belongs
is given, there is implied in it the standard mineral composition,
or zorm, and in order to describe the actual mineral composi-
tion, or mode, of the rock it is only necessary to state the extent
to which this differs from the norm.

And since quantitative distinctions within a petrographical
unit of the smallest magmatic division established by this system
would further subdivide these units, it does not appear desirable
in the present state of petrographical science to carry such sub-
division to any considerable extent, however desirable it may
become in the future.

If the mode be xormative, that is, if there be complete or
almost complete accord between the standard mineral composi-

aS
8, F ~~ a > 3
ONTARARE.

RANG I, PERALK,

Subrang 1, .

Subrang 2) |

Subrang 3,

Subrang 4,e

Subrang S|. a

|

RanG 2, DoMAl .
Subrang 1!
Subrang 2] |
Subrang 3, .
Subrang 4, .

Subrang 5,_.

RANG

Subrang 1). ae
Subrang al
Subrang 3...
Subrang 4,|...

Subrang 5,|...

1

3,, ALKA].

RANG 4,

Subrang 1,| _.
Subrang 2, Nes

“Subrang Bales

—|

Docar¢ |

RANG 5, PERCAL).

SN SE SE SE

Note.— X indicz

SAL x JFL 7]
CLASS I, PERSALANE, >->7. SUBCLASS I, PERSALONE, wa >t.
: ; Qv7 O Ais 5 Q U5 3 QU3\: OMe ae 8 1 AGS hy Se Lee
; ae a<tSe os = a <t ce = eS 7
ORD ERE ey ioncese tetra UE: 2 nea 3 Paes Dra Sea SW Sa Rieu Diners ics : RO.
VICTORARE. BELGARE. COLUMBARE. BRITANNARE. CANADARE. RUSSARE, TASMANARE, | ONTARARE, <2
RANG 1, PERALKALIC, REO sf OAa DOr O Dob PORE cn Deo ere cram 1. Dargase 1. Alaskase Te Liparase 1. Mordmarkase| 1. Miaskase 1. Laugenase Tae |hieeaasek Sa
4 K,0
Subrang 1, Perpotassic, mee aUaca ears i Nestea (o) 3) 5f4, 902 isla atatover once eeu | hese iere muerersretereraee jk, x 1. Lebachose Gre vets ase s Visiscastere eeeeaen Masargnen 6 ound jpaaounusnecaancenencuyacdds
7, S20 275 } : Asopacenosevcol) Conhond 2 Bae
Subrang 2, Dopotassic, NEO) Snes cb oaoo dora DDO eOO DO aap aoe son ca Ao0d| |aartcodo Bh Ae ease 2. Magdeburgose] 2. Omeose Beene DN ay aces era ew CNIS os ois Slee | cd ee
P F K,O ra 7 i 2 . 2 2 2
Subrang 3, Sodipotassic, WarOaeeae PEATE METAS er ek eu TOMAS eel REOIE EE Bee? «Wo aoe fe oll ae a¥sticwe, co¢Sae) caacevels 3. Alaskose 3. Liparose 3. Phlegrose 3. Beemerose 3: io BRA rete Aiea) tel eta 2
b K,0 I < arkos liaskos aL UU XC MLOSE SS | lira varovaretetacsaresvarercial| aruieeeuniatnrcuatntere
Subrang 4, Dosodic, Nao eee cnooba gpabonnob000D noo booCObooRbobDOolooKOODcobHb0K0e0 4. x 4. Kallerudose 4. Nordmarkose| 4. Miaskose GG WEMEOIRS <Hosnooacnencncucllaunnaaaunanc
7, 1O ; i eee :
Subrang 5, Persodic, SEO a nogoppedseoonadouomodoo Rnd o AatoeEta SE RGmnbne teers aes 5. Westphalose 5. Noyangose Bo INO NOSE |] Ga NIETSTALOSS || Gococooscousodlloacocoocancaueullonocouusenee
RANG 2, DOMALKALIC, eG Is D000 dd0d0D dUs0000 DU landaenoed sana ad 2. x 2, Alsbachase 2. Toscanase 2. Pulaskase 2 Vibes senase” ||| Se, sistancision ence | (earerensiorn tie eieye
: «,, KO L7 5 iiocouo men enrnn oe GOSS EIC Tisiatesete eve ie tai otal tisturaval sas maMeioe ll aot oerer ereevans | eters eee
Subrang 1, Perpotassic, NaOe SOO DDUTOOA EDD ICONS |e EE ane : | Oe Roe Kn
a Je{9) : 2 Ps ma .
Subrang 2, Dopotassic, EO Meee noba Duo DUR COsHOG loqoaaOED Cane 2. Mihalose 2. Dellenose 2. Vulsinose Bi cscs fara sareters sje iiaved| bela tian steele favesetebnerel | ey Me eTE cacao TS (hahace erate
i i BO * Bp “It H k ne Se (COREL 6 8c worenbus coment tansne der
' Subrang 3, Sodipotassic, NOS : snovaucacdodonaanllods 380600060008 Ze x 3. Tehamose 3. Toscanose 3. Pulaskose By Pees slecsusnneeteene| tees
i 3820. 5 4 € rvik e Jiezz SE Mieyatecapecs jess ieseAuvwae ia (oil ieloiuyeraieharatbielepuleshsel| ls feueiW ute) bier tire ack
Subrang 4, Dosodic, NONE a SOAtDOCUUD AD ad OU coc ATED ED Rte ee 4. Alsbachose 4. Lassenose 4. Laurvikose gee ViezzenoSe |)|\yscn see commie | eee
: A, E
cake, “OZ Juk . Mariposose ; Sn | NST eeanercel eens uscd Goel a tryisrreotenva cuteisal tanita ogc
Subrang 5, Persodic, Na. <7 6.000 00 SoutnE. ban ppdUHao Mia maccyaered 5. Yukonose 5 I 5 5 Be
K,04N 20) 3 7 3 * vlAcOp IDOIGDOMDO) [pHi o.oo oda laron doh GOOOmAllato bh once
RANG 3, ALKALICALCIC, ae Sens Podo.co DOT COO uo ete oe Oe o Bs x 3. Riesenase 3. Coloradase 3. x 3
Subrang 1, Perpotassic, z aogoos obo epoG0nCa pogallapiae ac-cecacn Haden pmartates-c0 6 Monson dooucsanoc MooduneoonoocollosondonsodunmeollancboocdoanH SbdlleadonoGen Doona| jounuinetmot
K.0 Wye yetecaes eaeevane
i 2. Cae Wn See percnaOc loro oom conicn ob nl aden canon Comino. Gomes oo\rc
Subrang 2, Dopotassic, NERO 906d CONd NAGOS DODGED ROAsOMnAS Poa 2. x 2, x x ;
5 5 j i K,0 x i my ia S| ys Bs fu wiafelleve: 0 a evalu cree .0 | eievc’e a nies me) s ohevele [itiny ph Be Giuheré 0:0 QU liga ie Sis 600 Bierce
Subrang 3, Sodipotassic, NOSE sag0 CUO duO OU OUGOS laatamd menace Ds Xe 3. Riesenose 3. Amiatose 3 x
] K.0 y mS ””~*~<“—*S«S ec puede nue aye ee 0) sie}licgia ale sible ee selunasereceilis ele) sie aie, 6) areas sec0il nual a 6nenanelelecelarg
Shibiene? a, Wesodlie, 23 <=, sono csso0n gobo docdollbsoosegoroonn 4. x 4. Yellowstonose | 4. x
Na,O ~5~ 7 3: x 4
Subrang 5, Persodic, ee Se SEED BRD OOD OTe CIEy ons] |ac aE 5. Vulcanose 5. Amadorose 5. Gacaneenadnent oomnmemcane tralia que soborandauani ante ctuacs
an0) ~7
RANG 4, Docaxcie, MAO REVO) 28S SOUL ORT OU DENOTE TOTS Cha CT ORE ane eS rene eee eee 4. x Aly x 4. Labradorase | 4. x Rr Te 8 a eee mB Rammer loonie HEATH
CaO Bia)
Subr fa, AMS olerrtesusieievarstsisies ers Weccpeparte pence teteroce RRO OUO mers ean races 2 He clita Don rato oerollonaiteennottesn
Subrang 1, Prepotassic, Na,0> 3 pod soundda dowsuy cotcaullagnatondtate open leo eee Eee cece ie, x Dee
y sodi ion ees 3 2. x Din SBOE OOOEORD DE RRA OR OCOS CREO ara trac om eGo cd Hel inceeroraeoned aillarerarsr es GoeaPeera
Subrang 2, Sodipotassic, NakOnsaoare SxOdOC OBR Udogh05|loAsOauE 06s Dads Coma enter eens
$ i KiO 3 3 3. Labradorose | 3. XO || brevacecsvehereteverorae cere Wihc aMutarane) cleiott loadsa) |fevertssicavasvere oii
Pres ETN NAT CTO TCTs ee To PANT TT aol foycet cco drsvay ohevst asec] ckeraiakevelaslopare) oc as oVsuai| | Basie ase ersea ee oy, x ; f
Subrang 3, Presodie, Na,0 < goiter nesses cece seeereee|ece sees 3 3 3 |
|
RANG 5, Percarcic, So < Reno bs ates iad | eGicds as eer te a eye a ie Nagau TODS ob aciel hosdaeS SERRE He Gnriine © \llSanowourectacod| |komonmboo net oS IGascioncooaarseullcack

Norr,.— X indicates that analyses are known which belong to this division, but that no name is suggested by us.

Mile sis

Cts

* ma
Pe ah nat el es

8, e<i>s Sa
CAMPANARE. LAPPARE,
RANG I, PERALKALIC, I. X I. Urtase
Subrange ry Perpd'l Tees cae cet. lis oratc: praigiadcoatn a)
Subrang 2, Dopo DENS eal ph ols PO aera te Uy co tee
Subrang 3, Sodip | 3. x 3. Arkansose
‘Subrang 4, Dosol |-4----.-.---.: 4. x
Subrang 5, Bere 5. x 5. Urtose
NANG 325 DO OMATICATIG |i 2 eszvase. ||. a. cack. os a
SUI mney Gi, eS aonlle oeoploon decal Gene veneers ded
Subrangy 2p Dopo: ||. 2+ VeSUVOSE, aes 2 susie ica:
Subrang 3, Sodip | 3-.------ aol te sea St ae Need
Subrang 4, Doso | 4- > Sols RAT ME Gree ar a AR Ree tea
Subrang SETS Gel [ee biaccierror saety esta ater eee nasser ke:
RANG 3, ALKALICALC! Pints pene ont eieetece al | Orcresat ee eRe eE SEAS
SubranenesPerpkilsie 22 2s ees (aasen ea
Subrang 2yDopiel tase. 8s). se ales eek eai ae
Sue ee gee te ee ee
Subramea! MDGSdille gests a aealeecckm eee ek cd
Soimene Eee i peieeas Womn Rranager ys oh ee
RANG 4, DOcaALcic, ly soniay eee conan eae at Dae
SubranetlaiPrepl Wine Goose -tleseasen eyecare
Subrane peo gSodip ale sacar o-oo
Subrang 3, Dee Bases, ota C WRU Seta agen

RANG 5, PERCALCIC,

Norte.— X indicates tk

cee |
;
3
it

i

él

:

Men at

; i
ot
- : SAL (QF
CLASS II, DOSALANE, 23-<! i 5 | SUBCLASS I, DOSALONE, ee
I
Q 7 Onn Oe Qu 4) L es
mu ~aZy5 Sas 3 > 7) 3 1 8) Les
ORDERE wien e ls, eu nee a Ry Og Gee DSA oP<! >; De Se ORSESs mE<S>3 pay 95>?
HISPANARE. AUSTRARE, GERMANARE. NORGARE. ITALARE. CAMPANARE, LAPPARE,
, K,0+Na,0 7 ;
RANG 1, PERALKALIC, a O ta Sie AIO GOSODGS OOOO COOOL OOO CAPT ETRE Te eee reer ees 1. Varingase 1. Pantellerase | 1. Umptekase 1. Laurdalase 1. Lujavrase No x I. Urtase
er K,O 7
Subrang 1, Perpotassic, Na,O> x Dd000 GGdo ody Addo OObS0.4l loge Gama AOS Erte | tee ate ee Teaco ci Ils SOD ao eae Tener eee Dy fee atercseteeeten ek Ticesetessiomeeyareeacip Tictepefcrexsroceetiate Toeccor eee ee
. i ee so :
Subrang 2, Dopotassic, Na,0 < Ps SUMO aR ese Eed boLetesouoved | fay save stausvaysiei's;eisicici| cscicic aiceevenele ater stars A eetals Oca neh Su Day ects a cyst) 7ehoi ee 2Highwoodose.| 2. Fergusose Visonenopnscosy|| “bosssoagenos PV Rar eicey Seat ar
ee G K,O Sree} F z: .
Subrang 3, Sedipotassic, REO Sao B Guniced d4A0o4ODe 000 |lan Beale ado BoCee Semen Eerie eteeieneae Varingose 3. Grorudose 3. Imenose 3. Judithose 3. Janeirose 3 x 3. Arkansose
A . K,O ape
Subrang 4, Dosodic, Na,0 ee scab dn 6p 45 U6 CoGaOpoddollopue Todo oe dade laeeetesn a caer 4. x 4. Pantellerose | 4. Umptekose | 4. Laurdalose 4. Lujavrose AN era Pace 4 x
R c K,0 1 e :
Subrang 5, Persodic, Na,0 <7 JU DODOS A908 OO DEROOm OOogEH Doaee daar domenellaaaaacneamoncce CRAeDi corde mee Glohce acinar Giaahe seni Me Si Serecnsmre ee Sopaencreaeces 5 x 5. Urtose
K,0+Na,O z
RANG 2, DOMALKALIC, = Le 6 SHATAODO CUD WEDD HID | ces SrA meters beer cel brea ees Mees 2. x 2. Dacase 2. Monzonase 2. Essexase 2. Vulturase 2 MEMLUASAY ||(acancrierantee
K,O 7
Subrang 1, Perpotassic, Na.O 780.0000 3H0000 dunoH0 Aébvel|padoudsobenobdullaacoanbasoos 004 HosnSoso0ccn0796 I SG oeerelgee ems LATA eae Oncaea aoe ceniercc hort imowal anne ca nik t tks
: é K,O rae
Subrang 2, Dopotassic, NOS 0005 00 uOoD CUD Dopo0 domo Ge NEn Mad clases ocobaa De otc Dp x Dc dn ooo Geneon 2. Ciminose 2 x 2. x 2 OVESUVOSCuml|hictprt reste
3 AiG é 20 =
Subrang 3, Sodipotassic, N 2,0 5 Meee eAceMct Wels fol shciaToyeNeToneya)| auspsreesne iets ts AcieutierslloeeP tvsse emer ae Br x 3. Adamellose | 3. Monzonose 3. Borolanose 3 x Hapagctod nod looduows daddGan
r K,O =
Subrang 4, Dosodic, KO as) SMR I ete ker Resets Ce nete¥ eros sl| (vce reszs)3 casas tsusre tel eaves sheiey obs ore setae Aine hers facie 4. Dacose 4. Akerose 4. Essexose 4. Vulturose 4. Xo Pilate ares a
x 0 Ke Ome
Subrang 5, Persodic, Na,O Fvod es jndhen onda nodoucond|lbanedegeadecenallpascGG0o50 bes Ca Geetinvent dois olon Gounccadoesane 5. x Sign creeciteaene Bn Rodccramnh avd) loqooons Hosos
; Na.O y
RANG 3, ALKALICALCIC, psloaaal = <A : Se 000'00 00-0 /5013.0.4.0 6 ol bao bara Meee en memento aerrnerae 3. Almerase 3. Tonalase 3. Andase 3. Salemase 3: x oncerteunDpn rand lhoosacanoogonac,
ees 7
Subrang 1, Perpotassic, Na,0> 3 naacon boom RODIN Foto loos o Geena me tol (Necoe aericineE te ie tam garcons oocanoddosood Tyensyatyoter te clate I x Mie banon Go00d lo cag GHOROUOORG ond OpidocaAgdor
sane a LSA) 75 :
Subrang 2; Dopotassic, Na,O “173 cabo coonadoolooconsoullssaposonanes on dlloocObononE Ge Abe PE oHEneD eddoaa ne 2 x 2. x 2 x onasnaadooas © WoXefeke aiciehatatsfaters|| Kintersierurciteir rele
‘ Gg. = ASf0) ~
Subrang 3, Sodipotassic, WHOSco C105 0.0 HOG COU O.CT.07 OSD orRC noc anal GeO Ce eee eecRere 3. Almerose 3. Harzose 3. Shoshonose | 3. x sh Se |roncandoudonac ncatocemorcicctt
2
Subrang 4, Dosodic, MnO <5 acbooo onde ay oon Ves bodlbeae DDAROS eC Odd GOO ROORSA RE aeE 4. Sitkose 4. Tonalose 4. Andose 4. Salemose 4. Cee || ocopenonttabed| Iforcrtinootn Coen
7 2 :
Subrang 5, Persodic, Na wo<i nov d0O 0000 OOO RT COCR BO60 Gol [Cabo co HORS ee [bebo MERE eer Bate Gioia seci= clersvedearesrs 5. Placerose 5. Beerbachose | 5. x B intiesielsxo\esore at ayaa REMaters ceo esuavare Cuneta apain Cutan era teLerea taints
= = =
RANG 4, Docaxetc, ee <i>! 6 ato BRON) OS GAIN ORE CTO | TCC Ee ene era eee Ap x 4. Bandase 4. Hessase 4. See | WI MaraSaA Ga cores o| oon GEER EE | MEO Mecreriocete
. Ks OFS .
] Pr 2 yo Con 7o O.6bnEd. 60.00 DO0O110.6 [DOU UDDIOMA ETH IF are 5 6 SEPEIMOR || Wogcenoabcaqnad GRC POOCoIOn DDO Om co oo ton fol (coccinremboordnng looboonodn caro ce
Subrang 1, Prepotassic, Na,0> 3 dcop dd CONT ROD OR Ome I x I. Sagamose I
: i Ish) 55
Subrang 2, Sodipotassic, Na Oisarae 5.000 6d 0.00.00 OSH 0.0 |OKoRO OID sec Croce | MET ERNe HeeeneRerE PAS Cae 000 2. x 2. x Adtiootdo uoddiid| bnoocmnccndas (olla ancodnadn onal iu axcoanddo dior
Subrang 3, Presodic, so) Ere eopond bG00H kao edoad bos 6 SC EOC e OeHE teeta 3. xR 3. Bandose 3. Hessose 3. Se |Innocaseddooon, «|.ccontocncnamanloncgods bonbdor
Na,O ~5
a = = ae Fh
Kane 5, PERCAL CIC, pe Bdddoocgoedh 6800000605] obor eH Aon Cotten [NOE Eee Aer res Be (GUAR? lnc nono onan Sp OLa02 llGoaobo oD aCObenG bonnOnae oct: lloanodnaxadognd landunen ation coer
= =
Nork,—X indicates that analyses are known which belong to this division, but that no name is suggested by us.
n

mt}

eh aa aerial ry AR

seal ei
nua

erat

a

. 8, g<i>8 FD!
BOHEMARE. FINNARE,
RANG I, PERALKA| I. Chotase 1. /jolase
Subrang te | Teetototepecetccereg- st | ea
Subrang 2, | 2. Chotose 2. Madupose
Subrantaign i Sitisst hone. ces BEM Cres
Suibidanyes vik || Cpa on eomisns orc 4. liwarose
Subrang 5 | aeeupencintertetaeer 5. Ijolose
RANG 2, DomaLk 2. Albanase cle aaa cee
stn Eee
Subrang 2, | ZA banose yh |leisecter apps cisternal oys
Subrang 38 3: EX aie a ata | cenerstatersy cheated sles
Subrang 4, lbaeiC@ovose youn siete. ee. ay
Subrang 5, | 5: Sb RUGS Fall) ee ane cts ra siva\ewe Pci alc
=
RANG: 3, Avcrenen ado osIoo oboe Io donna ne aoun One
shou 1 is
- Subrang 2, | Baa ey P Ras ceee stall Uiibing Aer rec hi aeAeg
Pe eo a
SUR am cael citer Mare ea dane ae
Subrang 5, | eaeae oh cuereleXoMoul-Hopohchll-Reysheiienen iieusteacicncieits
Tess Sea
RANG 4. Docatd Bo wt 6 cad ooo ole.ool ere aaisid elenee gees ey
Shiloiesaer See rere oes or cts baler a ae eas cis ae
Subrang 2, | born und op > Dame ow re Sem aaa e ede
Subrang 3, | AGdS'oCOnIoDaoOnUd|laoIs Oe djoG Noo.
RANG 5, BeRCArG aco Uden odvugodlacduooUdeooCe eG
L

Nore.—X indica

|
|

an
i
|
'

cynic

rs

ne

SAL > FL
CLASS III, SALFEMANE, wee J SUBCLASS I, SALFEMONE, OBE Syl ,
. BEM S35 CZ I
7 ORS} Q 155 ~ O23. 3 Qe 31 QL Lx Wey ie My Ga Tieazees Ts See
ORDER aaa DIES on 3 Cy a hr Sa 6 F<E>; nRSS>S 8 F<P>s a
annd sod 66h O00 E560] lag Boodond onadnaSC ATLANTARE., VAALARE, GALLARE, PORTUGARE, | KAMERUNARE, BOHEMARE, FINNARE,
K,0+Na,O ; ; f
RANG I, PERALKALIC, — ee 2 éodanagdquanbocoaduancean||gusccocdoDadead cacao eenodeaicd 1. Rockallase Tia stsrejsi'a, sates s/airoye 1. Orendase 1. Wyomingase| 1. Malignase 1, Chotase 1. Jjolase
Subrang 1, Perpotassic, ISOS sous on God pH abdbedchoncdl lpouacapadcandoulaouay cemguctana Maso voddéuvoco|lacosanoanogeaoe 1. Orendose Wem Wyny ONIN FOSE) | eLivereteterey etc referenere Liperefeseieversvaerseys Teapercrertercieretree
P Na,O~ 1 y gs
Subrang 2, Dopotassic, 0 Lk SS EP RT eS | iy Ee cia nell RE ee Poacscddooordalodaubedd SesOw Din jaVavareteyaevcterats Dh tircveparscerateyepedate||le2meelersyacaterseren tad 2. Chotose 2. Madupose
INE{O) St Gj
: P K,O 3
Subrang 3, Sodipotassic, REO Soe §dco Geran ep Abie oaamy| |Paaeoe ohn Corens| io bate nas Deaeern sh x 2. kote acoantetane Rican ooDneTce 3: x Snood asuoucodn sloqnonnontnood Gnbanpadoanne
Subrang 4, Dosodic, se) GEESE a nig acl sob SH OGENG SASS PE] ROSE Re Eee eral CRE ce eres Pa eerie oer cit (peice acto tiers 4. x 4. Malignose (aqonaeocas doc 4. liwarose
Na,O ~5~ 7
Subrang 5, Persodic, # ee. «58 Reb oca nna oon Re ate | RoGee oats CaO ean MTeoR cee eine 8, Rodlelose |ljsooccovsavendan Ciena: motion SOUnO TE Genco oroteucne Genaomouneadsd 5. Ijolose
a,0 47
RANG 2, DOMALKALIC, a <i>s pdoonG CaooDOSUDONONE||SuP0 SDD Aeoscoss||sovudccovKsdodollae aude qocccsce DIM Se fevelcfies toe 2. Kilauase 2. Monchiguase| 2. Kamerunase | 2. Albanase |... ieee veeeees
I
: K,O
Subrang 1, Perpotassic, a <- nooo dcacadoaoubdsoyandlnvoudida soos Galleapucsssdoos eullapagubosecuo00 olbconadeoxon@ueD Meat syste Li epevete sPigetscsts wees EGA OROO G a sevononEouCn lopne bade cnonoeG
Subrang 2, Dopotassic, mo <2>S SOO eGOEOaS Hose oea Udon OoecuDoealleae.ccmed Coad ots] sendin ticicld oS oth oq] (oh Gn abe eee 2, OOM |) As secon onaan oe 2. 2. vAlbanose) iiixaes creates
Subrang 3, Sodipotassic, T0< : SG es Obs aA aeons an Sern oconcde dane bo cdonaadnrtrbolld uoacodao cote e.d | taed apeemotnrnts 3. Lamarose 3. Shonkinose | 3. x 3: x
Subrang 4, Dosodic, mao <i>? spodUOHOow Ad Donoso eHOUpUoCOdanODaNoaAS>oohononnoDddobioaoaonnano coool bounoooodosan OU 4. Kilauose 4. Monchiquose Asma menunose|irav GOMOSEs | vesiaekeienrerensies
‘
Subrang 5, Persodic, mo < poouduoDoTAnUDbDUOFOUGEoGdS||aoDUDb cODdDORGDldonbosenoapoood lountoon000000CKlloocanb00 000000H 5. x GedanasOnbs Ot Gp x os ee ditondarwac couche
RANG 3, ALKALICALCIC, a <i>? reir cescop stays cauth su cher cletekes|\ctekevoussanececsysWsvey stata] |lonoe cesta une reel /sqesoes | eecnay steccrat keeconerer ae 3. Vaalase Ss (HAR || By SEARED | 3 (AHR Wenen coacoan adan|loooonnon Nias oot
$i, PO SSO oo oo od toe Ron TOE ER Ree REN Dee eA G5 hah ot. Ss Rin Ra | re nO cr ner ae ge ee ee Oe
Subrang 1, Perpotassic, Na.O7> 1 asHbbnGUs06o HUGdoUENoD loDObOUUoUED DOO ppoUdcoddapanocllognoboonpacoaDe UWosanacodpoade iteoaanacopomas Wagqonced 080bD Mognaadponoood adeno ngouoocoRo||go bono on
5 K
4 Subrang 2, Dopotassic, WO : Sd LOO UTS Clans tol | POUT COM CMEt ohn! OsitD SomnOM OO ome On EOO OSU Fs oo HORTON DeAbSarOkOse|e2irtereiesiencciee 2 nano COGOmGNnO nO prodded pUon on OaD
Subrang 3, Sodipotassic, Tao = svoodentoccngnevdelooanerndbao nhacalloscanedoteaancsldacgbuhoodoooo7 B cco DB GmeOuA 3. Kentallenose| 3. Ourose SJooguv 00 G00 UUlloone boododurbon|oovooonocosgoan
Subrang 4, Dosodic, Hao <o> 3 Hold dua. ppb UMe Sole een cllUpoome tena dn Siial Pom acmatenecaEr ae laoo moos orn e-can 4: Vaalose 4. Gamptonose! |) 4. Limburgose | 4. Etindose) fees w seen eweveul|ioesere esse seen
2
= : K a =
Subrang 5, Persodic, NEO rece Sayed Page es suentuelyee n59 gsaanol| Seas.909Hs oULb os) hang dnRtauaeDtahobgonsnecy aac: Gasosendoonand 5 OLNOSe meer Gj x Ea En lyondosonconounlmoombaancd cx
RANG’ 4, Docatcic, oe <3>? ropcodoa nade noocaaddeallauoonoo boos onoolloonnhee ondroooD goddocdo ded 4. x 4. Auvergnase | 4. x 4. SA lospo tooopticomdcl|ino aces rovaratnareteay
S ¥ ’ i K,0 5 eee eelieo eee eee eesessne
Subrang 1, Prepotassic, Na,O> 3 sonencoADaUcpDoocObocONolonHoUDD DODO wa Godage cedopouEooboaesoDAa oud fooagcundscoud Tictyete(aisketeversieteve ite eon condoodddlluoconodonaoconolanamyun0
~ 43 c K,0
Subrang 2, Sodipotassic, NaO Soe So ooEns pa co ees plod lbons ore pole acolo oo more bo eocotl Sarco an os Darah (ov etel a s)sciuel ex 2, x Pe x 2 S Wilanrb Ooooh pt oduollnauboUrcootinedn
_ ~ Ky 2
Subrang 3, Presodic, NEO Se IngpoOOyUgooO Ao ooMcebad aden doeHooaend| Toten soar seee al Sens onanecdnnooo 3. x 3. Auvergnose | 3. x denn codccocon|lonoenoenooode ud |oodecndon oconcao
RANG 5, PERCALCIC, RaG a Nard << WWOTIOSEY OOOOH IOVS SHoooOpElbeoodCODcADEbsAl>oDEbONSUDE DO DRaBooboACOpOoOOG||boogo00dcDA00gn i ACHHCOLE? WKeaaeenougnuoos|ooo0 sods sun 00n0||ocoo 0n00005nK0n|joboaodpg00 00000

Nore.—X indicates that analyses are known which belong to this division, but that no name is suggested by us.

ae

CES

hee

Eas

Sees
ak

B SAL
> FEN
=—__——-
2S gS. EEE
a SCOACKARE, ORDER s
Se sais HU3N1 Ht
aes 5 sop pS,
are. Es Xe Mes May (Nits rare pacers te tye
|
RAtse BOND e's Wa linet emi ticctoy ceaner cH fetunculbcdtae ventas pa
|
S(webhlasemase tens mer ue selncaceceun aan
|
Wiis Ps at docs Tee OS eb ety x hareeem a ire
uettose DIN maior afteteite sonal lhavepeandensicemerenerens
Saal ee Bi ales at NE eA ial Ria arner Da ASE ane TeE \
=
Sab a ose Al crevisiies e\\s}\s\4e)i0i40| lls) Il vj,0/\e\i='{e\det) sisi je ioj (eis 2)
ay}
wn
¥
Aas
oy]
fo)
ian
(o)
Z

:
a
%

|

» SAL I = POM
CLASS IV, DOFEMANE, FEM <i - > =o SUBCLASS I, DOFEMONE, x = : ‘
é P,O_7 sO) 758 OB Olswca :
ORDER 1, PXO57, HUNGARARE, ORDER 2,-4,-<; >; SCOTARE. ORDER 3,-7-<= >=) SVERIGARE. ORDERy, Be<i>? | ADIRONDACKARR, ORDER ;
Fs! Palys Pei>3 IP AS Per Wo IT RES D555 Pivgq Don Ro F ee G eae Tae — = ~-- = =
i 10>: Lire nee Dire eee "ONG Dee b 5) Sea D 2 7OS: Nope (Oe ee Ney SSS) ~ aa ~~ 7 He S H
SECTION .iecisssecse seer x O71 3 ON1>3 395375 if O<s77 3 ON; ‘i O07: Z, 06373 3 05375 } ON<s77 3 oS; NO>: 2 08:73 3 O<3-s5 4905.77 ro Suborder NP y ipsa mee inst Cece
Minnesotiare. x Hongariare. x Pyreniare, x Paoliare. Texiare. x x Bergeniare, S| sonconahsraassacas x Svetigiare® 4|||" 5) ten Reape ase Ndivondackore Champlatnors te U 7
R I, PERMIRLIC CHO SOD aE 1, Minnesotase | 1 x 1, Wehrlase 1. Cortlandtase | 1. Lherzase v4 x 1. Paolase 1. Texase 1. Casselase I x 1. Bergenase | 1 x r I Taber; | oa aie . 5 a
ANG I, i Na,O Parrees tpeeves| Te 5 é ‘ « Lherz . : e : b : cy . XK | Tee erences . x Mes LaCereaser lice vwene Soo] Koboonpieend 1. ddirondackase | 1. Champlatate | roo cc is leccs cay cane
: o MgO-+ FeO. 7 ered re a: : a ; é é, F : oe .
Section 1, Permiric, On O eae 2 Ye(atatetey s1atstatsTej oie ef cLay 1. Minnesotiase| 1............. 1. Wehrliase 1, Cortlandtiase| 1. Lherziase Maeacanoonmoo000 1 Valbonniase | 1. Marquettiase} 1. x 1. Kalteniase || 1. Bergeniase] 1............ Mogsdussaocac Gon don voobds 1. Tabergiasel]......... Ooo doadoneeonc 1, Adirondackiase) 1, Champlaininse) no... ......cdecce cece cree
Subrang 1, Permagnesic, BOIS date agoay ao aust TE RR nt ean lL. x 1, Cortlandtose | 1. Lherzose * sere eae. ae 1. Valbonnose | I......-....-. I. x Tigges aratinese Goan adanodad onacivopdodccn puodoboccs cond donbons aco acue Sosnscobunpon |sasitGubances Siralevuscerafare: sie G I I
Subrang 2, Domagnesic, MeO <i>E.. noon atHdA00 2. Cookose Aanooouomos ton 2. Wehrlose 2. Custerose 2s INFAINCEO. | Ibenanvocages condlopnencdocavcean 2. Marquettose | 2 x Poo san isooued Phan daa aoagad so doosdsqaoge0d s0o0dcOHSaRGdd4n00r 36 Abcisnnage Kors | Seateserateretereteneter] mtemrtereret creverayeye 2. 2 2
Subrang 3, Magnesiferrous, me <i>} ALUTOOOUDOGO k F Dotonguccooun aheah oocoodnten Goi ICG DDOAAS Qhesadoccdoybl|Iyoocavouoadadsnqaa WoandspidcidDot Onoscecoassoud Bipavspotentelevervebees 3. Kaltenose 3p JHOFHSNCRA ls ocoooedoncous sxc shakeeieteiessisl| alateteet ener meterarte Choonasosnon Wlasouaeonnoos 0 peers JRveuio oeona brn cHaabn sade Fie iic|| 3
3 See DO ten eee ; te ‘| -]
Subrang 4, Doferrous, TO <s27° PA irate estan vals (nonone Baooddel | Manapackeenees (lanbecohoneoge Aetat tere acie Yai (lasusoenssbdoo|lasamocecacdoaoAgellootbodobsuadeas Accoosobuaaoac Asaradeoxcosne Phanoqadaosex (eemdcepoadod sagan apoAaBenl lobeodowS acKbabladose coca ss quateyeresisiereserets || nectar stents | rete eatelete ¢ 4 x 4 x Chavon ob panto enocoosde od 5
Subrang 5, Perferrous, M&O? Seeinysstetbertters ari: Soneonnee Sorte pr eaanree Be beac ae as| latin cies sutane oat ueisiteetren epee teed Sheen: edeeen Raa sree 5 5 } &
8 5, DEO pretest stew enee | Seeeceeeceened! Severceesce ese! Sesceessseceeal Seseecesce ee] Seeeee secs eeealeceeeseeereceen[esscceesceeece] Soeeseeces sees! Seeseeeeeneeea] Seseeceeeee eel] Steceeree sees Bouroobaoembon|connos oh GobNSdlcondgo ons eo ool! Sroerianooons 3 enolsoneipae Sco! dodo Ponocusn |e idrnnccodod ol Sngemmuaeognn OM 4”
= _ EE - —— vo
———— = a) ee u
I . . U | re a 2
Section 2, Domiric, Mite <i>s eh a Seo GRR RCO Te 2 =x 2s - ix 2. x BN h rer costes Biera qelstetcisatessi fis Bi 9X 2. Uvaldiase | 2. Casseliase | 2. x Die ves eae 2. =x PE teeR at 2s Bs EX 0 ae el anny a ee a wee ra
wee eee sete eee COCO ORO aC CS i) b TTC Mens MCI Steer rd i=i
5 p sic, MBOS7 Belcherose °
Subrang 1, Permagnesic, TeO ogi role Opdhocows nyanaterd 1, Belcherose I. x Tevessesesseee 2 Degeneres eerecee Too. sees I. x I. x p SSeS icin | cic iCcI eine icin TeferetohegeretePaietets) (theretecredatetehereratere lGondsosaraddicods onaaoono ae cies RAEN RTE NCRGERY S)
Qe 4G 5¢ |) Me ooeniontipogel| Sn crenbdenoon) || )sorcconpencoune|| Monenbodouncoo | > | 8) PQ eee | Tb Goa Geno bodell Eadoooooecoonol) fon agusosonbllopooounoo GUNN! Toonococuraballdoas onaabnandd|llonnaconsuncelsonaannencup Cay ee eC ty veeee See eee eee baie
Subrang 2, Domagnesic, MET SIDS eee. REA entender | 2. x A POROSBWEINOSEH WAN ver cepacia clare | siz a(elsisicievele/= c7e1s/e Pronkvooasboonns Py 5K 2, Uvaldose 2. Casselose Pe att all nanan aavyneie piers © 9 lleccanndurenas Be) sts OI ean aee mere [lee Roose roe cece eee aa Seite
Subrang 3, Magnesiferrous, me <i>3 HOaTHIHO BAe Godot Brvre cece eeees Bev ierree sees 3. Si Wen onocedoncnan wdado odcUdebDooG LensoadanadeD Bneicehscicis)eiel= Quocagcac 5 BhocaooDLdeDhl||soGaqcaboadood Khoabaonduceal booed OS TOnoG Ghocuc conaep 4 laangopddoueedalllaccaccne a Pieters SOCOM OG areal laraininto score 4
Subrangs 4 and § not represented, eres AG . ag \aPedeetea
Section 3, Caleimiric, me <i>? Pa A alta oo nrRee deat olwontdaneomennan lopade orravivie 06 Ob POR EEO 3. Venanziase ab IE ESO! Sb IEECMTERS lb con cpoadoagnun||baspano0cn5acadllscooos se andacal|lscopncasosacos 3. Avezaciase |..... Poe Saoe acoaccudanscan MA accanaa rot | Gareheon mare at : aililectete gatvnas lees vac Mie oa wcers A Gal
- - + MrOw 7
Subrang 1, Permagnesic, PEO ects nelle sesine ne yafedtata linaiereuneereinisvelove pYydon) cOvooroEb080 AapHOMeOHeE Hoonaogegancco | ihonguosocdosoge Kocnsedopgeooa|looadocne jon g000|loobEgD00Gob Coad janudbADacDHoHallloobacdnoeS onl! Hosgadoonco0e SDOOnGebHob Anobiids odaon wo fe (aveta ein osacseisfoeal laccturearenete Gob Moan igae onllodeanABpUrAnNTGE APE TaEe «ud! obi cRo tics eRe ares cela hn
Subrang 2, Domagnesic, We<t>s MOOG CAEL hondllonnaacsodn anno DOsrBDObSRODLnO|apmre cca aeuany olltea oGonpooodste 2. Venanzose 25) Brandbergose)|;2'y\ Ra closeimmmnyltterteeeerese tote tcystor-l Cl-telelsveseretsleZsveiecsra\| wists ove octatareystetell | erat lerel ave rete lerevetorel| 2m etcetera |e SrooooAbeon|ltie LaDudouNhoS Bac clotereiesiarter
Subrang 3, Magnesiferrous, me <t>?. Mesa | eect te teietitota| hierar vteisjeiss eis Bonobo ndcco60l hp pe Seetbre eee sloade ourrea0n0 SBoaonenouecasn Glan bedooonbocdllbocconnase donaclodosdnic qgnEcdE| 6a0cen%co0opod||sonoonontar noolp kb ANECCRS IIs ond gocacesacd}faonn. ahnoongr Baler Zeinfercinraweyel | ete cone Ree LSC QOMIBOCe TusCeNOOTC RACH Gr Ren hanic Pent si inera Baise ONG
Subrangs 4 and 5 not represented, = je - Me i
7 ; —— ee ~ ——— =

ae arene nn

4
t

pn ee En

Derr

Seen at a some Ae a

CLASSIFICATION OF IGNEOUS ROCKS 625

tion and the actual, there is obviously no need of mentioning
details of the actual mineral composition, since this is fully

CEASS2V;, PER

SECTIOD

RANG I, PERMIRLIC, =a0 + Met Feo
as

Section 1, Permiric, SEvGiO eg ©
a

SubIeng 1, Permagnesic, FeO SSS
Subrang 2, Domagnesic, eos
Subrang 3, Magnesiferrous, 5 <
Subrang 4, Doferrous, Fo <i>

= MeO I
Sub me Sg
rang 5; Perferrous, FeO Se tee ||

SN .. MgO+FeO 5
Section 2, Do so Ouse eee el S
miric, CaO < : >, 3

z NV
Subrang 1, Permagnesic, ee
ag

Subrang 2, Domagnesic, <2>2...
Teas

Subrangs 3, 4, and § not represented

ae,|
yi
\ NO Ty > 1) | SAL I ‘ 10 = Py \T POM .
CLASS V, PERFEMANE, Fey; <;. SUB-CLASS I, PERFEMONE, —— > ae
ORDER 1. ft, MAORARE.
I
: PW Pes Pees P32 le gee
7 eA ep) Lye Ree aS Sore a
SECTION Se ee ee ee ee ee te ee 10°; 2, OS7- 5 3) eae 4 Once” 7 6) <a
CAROLINIARE. |MARYLANDIARE xX MAORITARE.
RANG I, PERMIRLIC, E20 ea! hte rate Man Aenea 1. Websterase SEXO OVUSE || Woacnouonccore I x 1. Dunase
2 ao I
Section 1, Permiric me 5 ENE ER eee a 1. Mariciase I. Ke NERPRPEE Ec co 0 oe I x 1. Duniase
’ +) a ii
Subrang 1, Permagnesic, —s - SoS eke PaO i eee 1. Maricose I ax IES oo 6 ae i x I. Dunose
aie, .. MeO 75 Di Ra Se se ee Bia tanh ne eas 2... a 2 x dh tad ee oe
Subrang 2, Domagnesic, FeO “7 ee sees oosaenn
: HORE MSO SS po ecee Seater pee MIR ort aa Cee To... .....o| Qearccanc mien Re, Mae ee
Subrang 3, Magnesiferrous, FeO ae e PERE ae cies: 3 3 3 3 3
MIO) pe BIE ne ee lA One SS nee ela ar Qs soshe Se a Acai goeteusod Wales eRe RON OO
Subrang 4, Doferrous, Toe es 4 4 4
a MgO I 5 Cc 5 5
SS Sa en re ele emai eorriscecr tiers Gare aeD Osan Hono ooCoOOCDO GO| Suamagoocoga sol) Secaooogod cond: 500007 b0000000
Subrang 5, Perferrous, eG < SGo 3 Sa 5
Section 2, Domiric Me OuiebeO 7 Mea eta cee 3 2. Websteriase’|- 2. Baltimoriase |--.2. 2 eee ene ale eee ee ee
oy 5 CaO - 3 aoo anon OOD
Subran, ic, 2H _| 1. Websterose | 1. Cro... ee eee ici) On en ee ae:
rang I, Permagnesic, EOLA
Sulmanceas Domagnesic, <2 >>2...........+0000+- 2. Cecilose EMOTO Sw cocoac:. .. Biigecanae aves 6 qo oooucts cor
3
Subrangs 3, 4, and 5, MON, NEORESSMICl so sno ee bh one oD Spall Ovic eta sleet ieetirnle RR rs Ba stuido oan eel gues eae ep ecg Wed oat iy chee cua eens

_-Other-orders~of~-Glass-V--not-represented~by-analyses,

dl

s

<

RS
bees

rs

v<

Neb ds

er era eS are Sy

CLASSIFICATION OF IGNEOUS ROCKS 625

tion and the actual, there is obviously no need of mentioning
details of the actual mineral composition, since this is fully
expressed by the norm. To indicate this mode it is only neces-
sary to use the word normative, before the magmatic name and
connected with a hyphen.

As has been already remarked, though all the various possi-
ble combinations of the feldspar molecules are regarded as
standard, yet it will often be found needful to indicate the
special feldspars actually crystallized. In this case, since the
albite most often crystallizes with the anorthite molecule, form-
ing a soda-lime plagioclase, and since the character of this will
condition that of the alkali-feldspar, it will in general only be
necessary to name the soda-lime plagioclase which is present.
If, however, it is desirable to indicate the presence of microline
or anorthoclase, this can also be done.

To express the extent to which the mode of a rock differs
from the norm it is necessary to state the kinds of minerals which
are different as well as the amount to which they have been
developed.

It is proposed to use the names of the minerals in question
as adjective qualifiers of the magmatic name, as has been the
custom with such names as mica-diorite, and to express their
quantitative relations by means of prefixes or suffixes, or by the
order of their arrangement when several mineral names are
employed.

Owing to the interdependence of the minerals in any rock on
each other and on the chemical composition of the magma, it is
evident that, given the norm of a rock, which is involved in its
magma name, it 1s only necessary to state the presence and
amount of certain minerals developed, which are not in accord
with the norm, in order to determine the modifications of the
standard minerals consequent upon the development of these
particular ones.

And since such modifications take place in all degrees, from
the slightest possible change to the greatest possible, it is neces-
sary to recognize in a systematic manner differences of degree.

626 CROSS, IDDINGS, PIRSSON, WASHINGTON

Those already suggested in the chapter on classification with
respect to actual mineral classification are two: (1) modifica-
tions of the norm, which are so slight as not to interfere with
the classification of the rock directly from the mode; and (2)
modifications of the norm so considerable that the rock cannot
be classified directly from the mode without readjustment of the
molecules in accordance with the method of calculation for
obtaining the norm.

In other words, we may distinguish (1) variations among the
actual minerals which fall within the limits possible in rocks with
normative modes, and (2) those variations among the minerals
which cause the rock to possess an abnormative mode.

1. Minerals of the first kind may be termed varietal minerals.
They may be defined as minerals whose presence serves to char-
acterize and distinguish different rocks of one magma unit, but
whose amount is so small that they do not affect the character
of the mode as normative or abnormative. Varietal minerals
include: standard minerals whose presence is not indicated by
the general expression of the norm, that is, the magmatic name,
though an exact statement of the norm would recognize their
presence; or they may include any of the non-standard minerals.
For example: small amounts of quartz or feldspathoid in rocks
of Order 5, in Classes I, II, II]; or small amounts of horn-
blende and biotite, etc.

It is proposed to express in the nomenclature the presence
of varietal minerals in the rock by adding to the name of the
mineral the suffix -zc. When it is necessary to use several varie-
tal qualifiers at one time the mineral names are to be connected
by hyphens and the suffix applied to the last, as hornblende-
biotitic alsbachose.

2. Minerals of the second kind, whose presence and amounts
are such as to produce abnormative modes are called critical
minerals, as already mentioned. They are for the most part
alferric, but may be salic or femic, according to the mode of the
rock.

It is proposed to express the presence of a critical mineral

CLASSIFICATION OF IGNEOUS ROCKS 627

in a rock by using the name of the mineral without modification,
compounding it with the magmatic name by means of a hyphen.

In case there are several critical minerals present, their
names are to be used in such order that the most abundant stands
nearest to the magmatic name, and the least abundant the
farthest from it.

It may become desirable when the number of mineral names
is considerable to abbreviate and compound them in the manner
suggested by Chamberlin,* and Jevons.?

It is to be remarked that it will not always be necessary to
state the exact character or species of the critical mineral. It
will often merely be necessary to mention the mineral group,
whether augite, hornblende, mica, garnet, etc. This is because
experience shows that in a magma of a given character the
augite or hornblende formed will in general be of a more or less
constant character. Thus in highly calcic magmas the augite
will in general be of one kind, in sodic magmas of another, and
in magnesic magmas of another.

The critical minerals, especially the alferric, may be devel-
oped within wide limits, which in the more femic Classes may be
illustrated by three rocks of the dosodic Subrang of the alkali-
calcic Rang of the Order portugare, of the salfemane Class. In
this we find the hornblendite of Gran, which consists entirely of
an alkalic hornblende, with no visible salic or femic minerals.
Here also belongs the olivine-gabbro-diabase of the same locality,
in which much of the feldspar has actually crystallized as such,
and in which the hornblende consequently is less alkalic, and
augite is present. We also find here a nephelite-basanite of
Colfax county, New Mexico, in which the actual composition
corresponds quite closely with the norm. The chemical analyses
of these three rocks show their chemical likeness.

In the first case all the possible alferric mineral has been
formed, as hornblende, in the second case only a part, approxi-

t Geology of Wisconsin, Vol. 1. pp. 30-40. Madison, 1883.
? Geol. Mag., Vol. VIII, p. 304, 1901.

CROSS, IDDINGS, PIRSSON, WASHINGTON

No. Ts 2, 3. No. ts 2. 3.
SiO, 37.90 43.65 42.35 P.O; tr tr -99
AOR Meaty 11.48 12.29 KONE eemmecull: -..dieoard 205
Fe,0, 8.83 6.32 3.89 SOs Sel Merson, ct alieetewiercrees tr
FeO 8.37 8.00 FOG Er sO aii scene aelategses .10
MgO 9.50 TQ 2 13.09 WADE =a 566000 ill. ocane .04
CaO 10.75 14.00 12.49 INI @)> alii eevee MMe .03
Na,O P68 2.28 2.74 Min Oye | oN eet ren tcnegedrss PON
K,O 2D 1.51 1.04 Ba@l sic cola emcee 5
H,O+ 1.40 cei TSO SEO a) | Segcece ny |earoeeees .09
lal ORAL tou orte ; 132 TST SO} Oa Mestere ele itches sere tr
COR ina is eee ety ale Seis wae —_——-|—__— -— | -——-—
AKO} 5.30 4.00 1.82 99.69 | 100.16 100.19

No. 1. Hornblendite, Brandberget, Gran, Norway. V. Schmelck, anal.; BROGGER,
Lrupt. Gest. Kryst. Geb., Vol. III (1899), p. 93.

No. 2. Olivine-gabbro-diabase, Brandberget, Gran, Norway. V. Schmelck, anal.;
BrOGGER, Quart. Jour. Geol. Soc., Vol. L (1894), p. 19.

No. 3. Nephelite-basanite, Ciruella, Colfax Co.,N.M. W.F. Hillebrand, anal.; Au//.
Z08, Cs SiGe'Ss(LO00)5 pel 71.

mately half, chiefly as augite, while in the last very little such
modification of the norm has taken place.

It is possible to devise a method of expressing in the nomen-
clature relative amounts of critical minerals for a comparatively
simple case such as the one just given. But the problem

is more complicated when several of these
hornblende,

time, which frequently happens.

minerals, as

augite, «ands jmica, are. present) at #the same
The development of one
modifies the maximum that may be attained by another, so that
the expression of the relative degrees of development of each is
a function of the others, and an exact expression in the nomen-
clature becomes extremely difficult, and is perhaps imprac-
ticable.

It will often be found useful to be able to indicate the
presence of certain minerals in rock groups, when the relative
amount which determines the Order, or lesser division, is not
Thus we may want

to speak of the persalanes or the dosalanes which carry quartz

known or needful for the purpose in view.

or nephelite, without specifying the relative amounts of these,

z. €., without making use of the ordinal divisions. In these

CLASSIFICATION OF IGNEOUS ROCKS 629

cases it is proposed that the name of the mineral be given fol-
lowed by the word -dearing. So the cases just mentioned would
be quartz-bearing or nephelite-bearing persalane and dosalane
respectively. Such names indicate only that the rocks belong
to these Classes and carry quartz or nephelite, with no implica-
tion of their other characters.

TEXTURE.

As already explained, the texture of a rock is to be expressed
in the nomenclature by a qualifying term applied to the name
of the magmatic unit, and connected with it by a hyphen.

There are now in use terms expressing some of the com-
monest and most characteristic textural features of igneous
rocks. It is proposed to use these in their present form, or to
modify them by abbreviation in some cases with the addition of
syllables indicating the degree of granularity. The syllable -o
is added to indicate that the texture is recognizable megascop-
ically ; -2 is added when it is microscopic. For example:

Granitic = xenomorphic and hypautomorphic granular ; grano = mega-
scopically granitic, megagranitic ; gvauzZ = microscopically granitic, micro-
granitic.

Trachytic = panautomorphic with tabular feldspars ; ¢vacho = mega-
scopically trachytic, megatrachytic ; ¢rachi = microscopically trachytic,
microtrachytic.

Graphic = pegmatitic = granophyre in the Rosenbusch sense ; grapho
= megagraphic ; grapht = micrographic.

Potkilitic ; potkilo = megapoikilitic ; ozk¢éz2 = micropoikilitic,

Ophitic ; ophito = megophitic ; opfzt¢ = microphitic,

Felsitic = aphanitic; /e/so = megafelsitic ; fe/s¢ = microfelsitic, micro-
scopically homogeneous, but not isotropic glass.

Vitreous , vitro = megascopically vitreous ; vztr¢ = microscopically vit-
reous.

Spherulitic; sphero = megaspherulitic ; sper¢ = microspherulitic.

Porphyritic; phyro = megaporphyritic ; AkyrZ = microporphyritic, that
is, the phenocrysts are not megascopically notable, or are quite insignificant.

With the term pyr may be combined one indicating the tex-
ture of the groundmass as follows:

Porphyritic granitic; granophyro = megagranitic, megaporphyritic ;

630 CROSS, IDDINGS, PIRSSON, WASHINGTON

graniphyro = microgranitic, megaporphyritic (granophyric in the Vogelsang
sense); gvanipfhyri = microgranitic, microporphyritic.

Porphyritic graphic; graphophyro= megagraphic, megaporphyritic ;.
graphiphyro = micrographic, megaporphyritic ; graphiphyrzt = micrographic,
microporphyritic.

Porphyritic felsitic , felsophyro = megascopically felsitic and porphy-
ritic ; felsophyr¢ = megafelsitic, microporphyritic; fe/szphyrz = microscopically
felsitic and porphyritic.

Porphyritic vitreous; vitrophyro, vitrophyri, vitriphyro, vitriphyrt.

Porphyritic potkilitic; potkilophyro, potkiliphyro, etc.

Porphyritic ophitic , ophitophyro, ophitiphyro, etc.

Aphyro = megascopically non-porphyritic, or aphyric.

Afphyri = microscopically non-porphyritic, or aphyric.

Salphyro = megascopically porphyritic with salic phenocrysts, sa/phyrtc..

femphyro = megascopically porphyritic with femic phenocrysts, /fem-

phyric.
Alferphyro = megascopically porphyritic with alferric phenocrysts,

alferphyric.
Salfemphyric, alfersalphyric, alferfemphyric etc.

From the foregoing statement we may summarize the method
of formulating the nomenclature here proposed as follows:

1. The magmatic name, of whatever division is to be indicated,
which is formed by the use of the locality root and appropriate
termination, stands as the basis of nomenclature, and is the
substantive part of the terminology. This is because the
fundamental character of igneous rocks is the chemical composi-
tion of the magma, which persists whatever be the mineral
development or the texture determined by conditions obtaining
during solidification.

2. According to the information at hand, or to be conveyed,.
the magmatic name must be selected which represents the Class,
Order, or other division to which the rock belongs. And this
name may be qualified by mineral and textural adjectives. Thus
it is possible to indicate the Class of a rock, when first observed
in the field, and to describe its chararteristic mineral compo-
nents, and its texture. If the relative proportions of its domi-
nant minerals, salic or femic, can be readily determined, the
magmatic name for the Order can be used. Subsequently more

CLASSIFICATION OF IGNEOUS ROCKS 631

specific magmatic names can be given to it. In each of these
cases the critical mineral and the texture can be indicated by
the same qualifying terms, or more precise ones if needed.

In this respect this system possesses a distinct advantage
over former ones, in which no attempt has been made to indicate
in the nomenclature the degree of exactness with which the rock
is known or is to be described.

3. To indicate the actual mineral character of a rock when
its magmatic name is given it is only necessary to express either
the fact that it is standard or the departure from the norm by
mentioning those critical minerals whose presence induces
changes in the norm, or those varietal minerals which may be
present. When such exactness is desired:

a) If the rock possesses a normative mode the actual mineral
composition is expressed by using the word normative before
the magmatic name.

6) If the rock does not possess a normative mode and it is
desired to indicate the critical minerals present, it is proposed
that such mineral qualifiers be used in some cases without intro-
ducing quantitative modifications. These mineral qualifiers
may be used as full names attached to the magmatic name by a
hyphen, as is the present practice, or they may be abbreviated
and compounded.

c) It is suggested that the presence of small amounts of
important minerals be indicated by adding the suffix-zc to the
mineral name.

4. The texture is to be indicated by adjectives expressing the
fabric, the crystallinity, and the granularity, and may be the terms
in common use, those suggested above, or abbreviations of these.

5. Either mineral or textural qualifiers may be placed next
to the magmatic name according to the emphasis to be given
them, it being generally understood that the term nearest the
magmatic name carries the strongest emphasis; the magmatic
name coming last. The same rule is to be applied to the
arrangement of several mineral qualifiers, that nearest the mag-
matic name is to be considered the most important.

632 CROSS, IDDINGS, PIRSSON, WASHINGTON

Examples —TVo illustrate the methods proposed, the follow-
ing examples may be given, especial stress being laid on the
possibility of expressing the exact amount of knowledge which
is at hand or is to be conveyed.

1. The typical monzonite of Brégger, from Monzoni—an
evenly granular, phaneric rock, composed, as seen in the field,
of dominant feldspars, with only traces of quartz, considerable
pyroxene, hornblende, and less biotite, with insignificant amounts
of magnetite and apatite.

From the analysis by V. Schmelck* the following norm is

calculated :

Orthoclase - - - - - 26

Albite - - - - - =n? OD 68.1

Anorthite’— - - - - - 15.8

Diopside - - - - - = TS. 43) |

Hypersthene - - - - Bu 3°e 20575 ||

Olivine” - - - - - a Din \ ;

Magnetite - - - - - a eral 332

Ilmenite - - - - - = 10/6 |

Apatite - - : - - T3832) ies 3 ira
100.0

The rock belongs, therefore, to Class II, the dosalanes, with
salic minerals dominant over femic, and this is evident even in
the field.

Since, among the dominant salic minerals, neither quartz nor
feldspathoids (lenads ) are present, it belongs to the fifth Order,
and is consequently, germanare. ‘his, likewise, can be deter-
mined from the megascopic examination.

On referring to the analysis it is seen that the alkalies are to
lime in normative anorthite as the ratio 0.0907:0.057=1.70. “Che
rock, therefore, belongs to the second Rang, the domalkalic, to
which we have given the name of smonzonase. ‘This, and the fol-
lowing points, could not be determined in the field.

As K,O’;Na,O” :: 0.047: 0.050, the two are in nearly equal
amount, and the Subrang is thus the third, sodipotassic, which
we have called monzonose.

We have thus characterized the rock completely, as far as
the dominant minerals are concerned. Taking up the subord1-

* W.C. BROGGER, Eruptivgest. Kristianiageb. Vol. II, p. 24, 1895.

CLASSIFICATION OF IGNEOUS ROCKS 633

nate ones, we see from the norm that the pyroxenes and olivine
are together dominant over the non-silicates, magnetite and
ilmenite, the apatite being in negligible amount. The Grad is
therefore the second, and this might be called sonzonate.

Coming to Subgrad, we have to deal with the chemical char-
acter ofethe femic minerals, \ Thisis (Mg, Fe)O:Ca@Q" :: 0.160:
O1094—1.702,,° The Subgrad is the second, domiric, and: this
might be named monzonote.

Taking up next the mode and texture, it has been seen that
the mode is normative as far as the salic minerals are concerned,
but that hornblende, with a little biotite, largely replaces the
normative femic minerals. Hornblende is therefore the critical
mineral and biotite the varzetal. The texture is simply a normal
granitic one and rather coarse, for which we use the term
grano.

In the field, then, the rock would be referred to either its
Class, as dcotetic hornblende-grano-dosalane, or its Order, as biotitic
hornblende-germanare, or as a grano-hornblende-germanare, if we
wished to disregard the sma!l amount of biotite and emphasize |
the hornblende content.

Further study with the microscope and in the laboratory
would allow us to speak of the rock as gvano-hornblende-mon-
zonase, or biotitic hornblende-grano-monzonose, according to what
information we wished to convey.

The augite-latite of the Dardanelle Flow, in Tuolumne
county, California,t which belongs to the same Subrang, would
be spoken of as normative phyro-monzonose, since the augite is
almost wholly diopside. Since, however, in the norm of this
rock pyroxene and magnetite are present in equal amounts, it
would belong to the third Grad, which might be called darda-
nellate.

Similarly, confining ourselves to the same Subrang, the
olivine-trachyte of the Arso, in Ischia, would be olvvinic-phyro-
monzonose, the gauteite of Hibsch’, a hornblende-trachiphyro-mon-

1F, L. Ransome, 4. /. S., Vol. V, p. 363, 1898.

2, HBSCH,) 7, 10.02. 07.5 Viol. XV IL, p. 84, 1897.

634 CROSS, IDDINGS, PIRSSON, WASHINGTON

gonose,and the mica-basalt of Santa Maria Basin, Arizona,’ a
felsophyro-biotite-monzonose. A glassy facies of any of these, with
hornblende phenocrysts, for example, would be hornblende-vitro-
monzonose. A pure glass of this composition would be simply a
vitro-monzonose, OY aphyrovitro-monzonose; if microlitic it would
be a phyrivitro-monzonose.

To take another example, the normal, lithoidal, micro-spher-
ulitic and porphyritic rhyolite of the Yellowstone National Park
is spheriphyro-alaskose, which is a very concise expression for a
rock that is microscopically spherulitic, megascopically porphy-
ritic, having the chemical composition of a rock whose norm
consists of extreme salic minerals, of which quartz and feldspar
are nearly equal, the feldspars extremely alkalic, and soda and
potash in nearly equal proportions.

An example of intermediate rock is to be found in the granite
of Butte, Mont., whose composition is discussed at length in
connection with the calculation of norm and mode. This rock
belongs near the border line between Classes I and II, and the

: : ]
rock from Walkerville Station, Butte, has — = 7a Ane et siSeec

persalane near dosalane, and may be called dosalane-persalane,
which contracts to do-persalane.
The Order is quardofelic,

Se
3

it is britannare (Class I) near austrare (Class II), or austrare
britannare. The Rang is alcalicalcic,

K,O’+Na,O’ _ Bemea se

j CaQ’ = 1.38, Sa
it is coloradase (Class 1) near tonalase (Class II), or tonalase-
coloradase. The Subrang is sodipotassic,

KOs! ice
Na,O sare I, < 3 > ee
it is amiatose (Class I) near harzose (Class Il) or harzose-amiatose,

BU TAS On SiGe Se Pa LOA on G.a WO) elS.O 76

CLASSIFICATION OF IGNEOUS ROCKS 635

Type AND Hapit.—It is clearly obvious that if great pre-
cision or completeness be desired, so that both mineralogical
and textural qualifiers are used, the polynomial name resulting
will be of considerable length, comparable to such present
names as quartz-hornblende-biotite-diorite-porphyry. This will
probably not be as great a practical difficulty or inconvenience
as may appear at first sight, since after a given rock has been
described and named in full in any given article, it may be
referred to subsequently by its magmatic name alone, or by this
in conjunction with a textural or modal term, according to
circumstances.

However, since the same or similar assemblages of modal and
textural characters are found in many localities it will be as well
to be able to express these concisely. This is, after all, what is
accomplished by many of the names in the present systems of
nomenclature, although it has not been done systematically.
Thus the names granite, rhyolite, tinguaite, laurvikite, etc.,
convey primarily an idea of the qualitative mineral composition
and the texture of the respective rocks, with a very rough one
of the magmatic character.

It is to be noted that there are two degrees of similarity
among rocks which can be easily recognized and made use of.
One is almost complete zdentity, the other a general resemblance
which suggests identity.

Type.—For the first of these, zdentity or almost complete
identity, we propose the use of the term “ype. Rocks of the
same type are identical in norm, mode, and texture, or almost
completely so. They are of the same grain, have the same
fabric, the same actual mineral composition, and are so much
alike that they may be mistaken for one another or might have
been parts of one rock body. Many examples of such close
similarity are familiar to all petrographers.

The particular modal and textural features which characterize
rocks of a given type are to be expressed by a single adjective
word, composed of a root derived from a geographical locality
in which a rock of the type occurs, but not already employed
to designate a magmatic unit, and the termination -a/.

636 CROSS, IDDINGS, PIRSSON, WASHINGTON

Habit.—In order to express the fact that one rock resembles
in general appearance another well-known rock by exhibiting
some of its characteristic features without being identical in
composition or mode we propose to use the term adit, formerly
in constant requisition among petrographers. The features of a
rock characterizing its habit may be both textural and mineral.
For example: one rock may be porphyritic, with a dark-colored,
aphanitic groundmass, and the phenocrysts rhombic feldspars.
Another rock may have these features, but belong to a chem-
ically different magma. The two may be said to have the same
habit. And the second may be sufficiently described by giving
its proper magmatic name qualified by an adjective indicating
the habit of the first rock.

To accomplish this we propose that the habit of a rock be
expressed by a word formed similarly to one expressing type
but with the termination -ozd. The root of the word is to be
taken from some geographical locality. It may be the same as
one used for a type, since a common type rock may be one whose
habit will often be used in describing another, less common,
rock.

Thus a particular form of rock belonging to the Order
russare, of the persalanes, may constitute the tngual type, with
definite norm, mode, and texture; while a somewhat similarly
appearing rock belonging to the Order norgare, of the dosalane
Class may possess a tinguoid habit.

The habital qualifier may be applied to a magmatic name of any
systematic division, since it does not specify the composition of
a rock. Thus we may describe a rock as a tinguoid dosalane, or
a tinguoid norgare, a tinguoid laurdalase, etc. There may be
tinguoid persalanes and salfemanes.

It is obvious that, with the use of types and habits, the
nomenclature will tend to become binomial, and hence much
more easy of application than in the present systems, or in the
full one proposed here. It will, however, differ radically from
the binomial nomenclature in use in the organic sciences, since
the habital qualifier will not correspond to the specific terms of

CLASSIFICATION OF IGNEOUS ROCKS 637

these, applicable only to the generic name, but will be applicable
to any division of the classification from Class to Subgrad.

ROCK NAMES FOR GENERAL FIELD USE.

It is obvious that a considerable part of the system of clas-
sification and nomenclature here proposed can only be applied
upon microscopical or chemical investigation. This is equally
true of a large part of that in present use. There are many
distinctions based on characters that cannot be observed with
the unaided eye, such as differences among the plagioclase feld-
spars, and the mineral composition of aphanitic rocks.

It is also clear that the system demands a more detailed
knowledge of rocks than many geologists, mining cngineers, and
Others. interested in geology, may care to. acquire. This isa
natural consequence of the advance in petrological science,
which requires a corresponding advance in specialization and in
petrographical classification and nomenclature.

For these reasons, which might be elaborated at consider-
able length, we are convinced of the need of general petro-
graphical terms which will be serviceable in the field work of
the petrologist, and which will be of use to the general geologist
and to those who may not be able to carry on microscopical and
chemical investigation.

Such general terms should be based on megascopic char-
acters of the rocks and should be limited to such characters.
Their application should be purely objective and everything
of a subjective nature should be eliminated. It follows from
this that such terms cannot be correlated with those used
in the systematic nomenclature based on chemical and micro-
scopical properties. They must be understood to have a totally
different scope, and to indicate no more than the general mega-
scopic characters with which they are connoted.

The attitude of the person using such terms is the same as
that of geologists who studied rocks before the introduction of
the microscope, and most of the distinctive features exhibited
megascopically by igneous rocks were well known and appro-

638 CROSS, IDDINGS, PIRSSON, WASHINGTON

priately named by the early geologists, and most of the terms
used by them are in use at the present time, though their appli-
cation has been variously modified by repeated redefinition.

We recommend that those terms which are needed for gen-
eral field purposes, and are to be based on purely megascopic
characters, be selected from the terms originally proposed by the
founders of geology, and be given their original significance so
far as possible, with only such modifications as a somewhat more
systematic treatment of the matter may suggest.

When all igneous rocks are considered with reference to their
megascopic characters their most generally recognizable features
appear to be their fexture and color, and in some Cases, and to
various degrees, their mzneral constituents.

If we attempt to group them according to texture we find
them falling into three large divisions:

1. Those whose mineral components can be seen with the
unaided eye.

2. Those whose mineral components cannot all be seen with
the unaided eye, and that are composed of a greater or less
amount of lithoidal material not resolvable into its component
parts.

3. Those with vitreous luster in the whole or a part of the
mass.

The first have been called by Hauy phanerogéne, and may
be termed phanerites.

The second have been called by d’Aubuisson (1819) aphant-
tes, a name which cannot be improved upon.

The third have been long known as volcanic glasses —
obsidian, pitchstone, etc.

All phanerites, whether igneous or metamorphic, massive or
schistose, were at an earlier time called granite. But at the
beginning of the last century a number of kinds of phanero-
crystalline (phaneric) rocks were recognized, the distinctions
being based upon the minerals that could be identified mega-
scopically. These minerals are: quartz, feldspar, leucite,
‘nephelite, mica, »hornblende, -augite, the. iron ores, ete: 1t

CLASSIFICATION OF IGNEOUS ROCKS 639

is to be remembered that there were no specific distinctions
among the feldspars. It is equally true at this day that no
specific distinctions can be made with the unaided eye among
the lime-soda-feldspars, and that albite, oligoclase, andesine, and
labradorite cannot be identified as such without optical or chemi-
cal tests. It follows from this that phaneric rocks cannot be
classed by purely megascopical means as having alkali-feldspars,
or more calcic feldspars. We must not attempt to subdivide
these rocks on the basis of characters not recognizable mega-
scopically and must content ourselves with employing mineral
groups, such as feldspar, mica, amphibole (hornblende), pyroxene,
etc., as bases for their field designation.

For these reasons we suggest the following use of familiar
terms when rocks are to be named on a purely megascopical
basis.

I, PHANERITES—phanerocrystalline (phaneric) rocks.

1. Granite —all granular igneous rock composed of dominant
quartz and feldspar, of any kind, with mica, hornblende, or other
minerals in subordinate amount. This is the granite of Werner,
von Leonhard, and other early geologists, and will include what
is now termed granite, granodiorite, tonalite, and most quartz-
diorites. It will embrace all light-colored, granular rocks with
dominant feldspar and a noticeable amount of quartz. It will
include the quartz- (hornblende) -syenites of earlier geologists.

2. Syenite—all granular igneous rocks composed of dominant
feldspars, of any kind, with subordinate amounts of mica, horn-
blende, pyroxene, or other minerals, and without noticeable
amount of quartz. This is the syenite of Werner, von Leonhard,
and others, with slight modification, and will include modern
syenite, anorthosite, and the more feldspathic monzonites,
diorites, and gabbros. If it is desirable to distinguish the
plagioclase rocks when recognizable, such as the coarse-grained
anorthosites, they may be called plagioclase-syenites, or anor-
thosites.

3. Diorite—all granular igneous rocks with dominant horn-
blende and subordinate feldspar of any kind. This is the

640 CROSS, IDDINGS, PIRSSON, WASHINGTON

diorite of d’Aubuisson (1819) as defined by von Leonhard
(1823). It will include the less feldspathic diorites and horn-
blende-gabbros.

4. Gabbro—all granular igneous rocks with dominant pyrox-
ene and subordinate feldspar of any kind, with or without horn-
blende and mica. Essentially the gabbro of von Leonhard. Since
it is not possible to identify pyroxene as distinct from hornblende
in many cases, megascopically, it will probably happen that all
of those rocks which can be clearly seen to contain dominant
hornblende will be called diorite, and all doubtful ones will be
grouped with the distinctly pyroxenic gabbros. These rocks
will include the less feldspathic gabbros and norites, and some
diorites.

5. Peridotite, pyroxenite, and hornblendite—all granular igneous
rocks composed almost completely of olivine, pyroxene, or
hornblende, in variable proportions, with little or no feldspar.
These names are to be applied as at present.

Other names in common use, which can be applied without
confusion upon the basis of purely megascopical determination,
may be used.

I]. ApHANITES—lithoidal, aphanitic rocks. These may be
non-porphyritic or porphyritic, the aphanitic character being
confined to the groundmass.

A. Non-porphyritic forms, having no recognizable mineral con-
stituents, must be subdivided, if at all, upon the basis of color,
luster, or other physical properties. The early distinctions were
in reality on a basis of color, and were two:

1. Felsite (Gerhard, 1814), Phonolite (Klaproth, 1801), Petro-
silex of the French geologists. /e/szte includes all aphanitic
igneous rocks that are non-porphyritic and are light-colored, in
various tones, and with various lusters other than vitreous.
They include lithoidite (von Richthoven, 1860), or lithoidal
rhyolite, non-porphyritic trachite, and phonolite, and the lighter
colored non-porphyritic andesites, latites, etc.

2. Basalt—all dark-colored, aphanitic, igneous rocks without
phenocrysts. This will include the dark-colored andesites, non-

CLASSIFICATION OF IGNEOUS ROCKS 641

porphyritic basalts, diabases, and other lavas known by a num-
ber of names, which before the classic studies of Zirkel were
grouped togther as basalt.

B. Porphyry—porphyritic forms may all be embraced within
the general term of porphyry. They may be separated on a
basis of color to correspond to the divisions above mentioned
into two groups.

1. Leucophyre (Giimbel, 1874) and

2. Melaphyre (Brongniart, 1813).

Gumbel applied the term leucophyre to certain altered dia-
bases of light color, which would not be included within the
group here proposed, but the term was applied to altered rocks
and has never been in general use, and may advantageously be
redefined.

Leucophyres would include all porphyritic, aphanitic, igneous
rocks, with light-colored groundmass, and with phenocrysts of
any kind.

Melaphyres would include all porphyritic, aphanitic igneous
rocks with dark-colored groundmass, and with phenocrysts of
any kind.

According to the kinds of phenocrysts which may be identi-
fied megascopically these rocks may be named without reference
to the color of the groundmass as:

Quartz-porphyry or quartzophyre.

Feldspar-porphyry or feldsparphyre, but not felsophyre, since
this name is in common use for a porphyry with felsitic ground-
mass.

Hornblende-porphyry or hornblendophyre.

Mica-porphyry or micaphyre.

Augite porphyry or augitophyre (von Buch, 1824).

Olivine-porphyry or olivinophyre (Vogelsang, 1872).

If it is intended to indicate the color of the groundmass as
light or dark, we may use the terms:

Quartz-leucophyre or quartz-melaphyre.

Feldspar-leucophyre or feldspar-melaphyre.

Hornblende-leucophyre or hornblende-melaphyre.

642 CROSS, TDDINGS, PIRSSON,;, WASHINGTON

It is to be remembered that these terms leucophyre and
melaphyre imply nothing as to the composition of the ground-
mass. They strictly indicate nothing but its color.

III]. THe GLAssEs—glassy rocks have been classified on a
basis of luster and texture as follows:

1. Obsidian—vitreous rock of any color, usually black, often
red, less often brown and greenish.

2. Pitchstone—resinous and less lustrous than obsidian, and
consequently lighter colored.

3. Perlite—glassy rock with perlitic texture produced by
small spheroidal fractures.

4. Pumice—highly vesicular glass, usually white or very
light-colored.

Each of these varieties may be non-porphyritic or porphyritic.
The latter may be called

Vitrophyre (Vogelsang, 1867) and may be qualified by mineral
terms indicating the character of the prominent phenocrysts,
yielding guartz-vitrophyre, feldspar-vitrophyre, etc. They may also
be called porphyritic obsidian, pitchstone, perlite, or pumice.

PART Tih, sMETHODS OF ‘CALCULATION:

In order to obtain concordant results in all cases in the
determination of the kinds and amounts of standard minerals
that correspond to a magma of any given chemical composition,
a uniform method of calculation is necessary. This calculation
may be made either from the chemical analysis of the rock or
from the quantitative estimate of the minerals actually pres-
ent inert.

The calculation of standard minerals belonging to the salic
and femic groups, rather than that of the actual minerals which
may be present in the rock, is warranted not only by the fact of
the variable possibilities of crystallization in one magma, but
because of the difficulty of determining the quantity and chem-
ical character of the minerals actually present in many rocks.
It is further warranted because of the impossibility of determin-
ing the minerals in a great number of rocks in which they are

CLASSIFICATION OF IGNEOUS ROCKS 643

too small, and because of the incomplete crystallization of all
more or less glassy rocks.

The variability in the development and chemical composi-
tion of the alferric minerals justifies us in treating them as com-
binations by readjustment of salic and femic molecules. Their
chemico-mineralogical relations and the method of calculating
their proportions will be stated later on.

The method of calculation adopted is based upon a number of
commonly observed chemico-mineralogical relations that obtain
in the rock-making minerals, which may be stated as follows:

CHEMICAL RELATIONS AMONG SALIC MINERALS.—1!I. The con-
stant relation between the molecules of Al,O, and K,O and
Na,O in orthoclase and albite, leucite and nephelite (Al,O,
Ke Ore Nan Or cap).

2. The somewhat similar relation between these constituents
in the sodalites, where the soda is slightly in excess. The ratio
imegsodalite isweAl Oo 2 Cho: Na, O --4+Na,0 2 i) is tain
Hosclite tes wl, On pe oO 0 Na, Or ---tNa,O: 2 bir,

3. The constant relation between Al,O, and CaO in the
anorthite molecule (Al O; 4), CaO; 3 1: 1).

4. The development of corundum under favorable conditions
in rocks with excess of Al,O, over K,O, Na,O, and CaO.

5. The relation between the development of alkali-feldspar
(polysilicates) and of feldspathoids (meta- and ortho-silicates )
and the available silica in the magma, so that free silica
(quartz) does not crystallize together with leucite and nephelite.

6. The stronger affinity of Al,O, for K,O and Na,O than
for CaO, Al,O, forming alkali-feldspars and feldspathoids
(lenads) in preference to anorthite.

7. sehesconstanty ratio; between ZrO, and SiO lmin zircon;
/BSO) 8 SWOAIS 2 Wate

CHEMICAL RELATIONS AMONG FEMIC MINERALS.—1I1. The con-
stant relation between Fe,O, and Na,O in the acmite molecule
(HenOne-tN a OF celas lo).

2. The general fact that this molecule is developed in mag-
mas when K,O and Na,O are in excess of Al,O,.

644 CROSS, IDDINGS, PIRSSON, WASHINGTON

3. The constant relation between Fe,O, and FeO in magne-
tite (Ke, O;oaticOr: 1.11)

44 Dhe relation between FeO and iO > in dmenite (he@::
TOR peel,

5. The relation of TiO, and CaO in titanite and in perofskite
(AiO cee € ans mica ne)):

6. The development of the titano-silicate, titanite, in the
more siliceous rocks, and of the non-siliceous perofskite in its
place in the less siliceous ones.

7. The constant relation of P,O, to CaO in apatite (P,O,
CaO : 3313)

8. “ie relation of CaO and (Mg, Fe)O i in monoclinic pyrox-
ene, -<diopsides(Ca@r.,( Moke) Ors: ih):

6: The occurrence of) (Mg,Fe)O both asi metasilicate, on
orthosilicate, in hypersthene and olivine.

10. The frequent, but not invariable, relation between(Mg,Fe)O
and available silica, whereby metasilicate, hypersthene, forms
instead of orthosilicate, olivine, with sufficient available silica.

1. The occasional occurrence of a sodium metasilicate mole-
cule, Na,O.SiO,, which enters into arfvedsonite in rocks in
which K,O and Na,O are in excess of Al,O, and Fe,;O,.

12. The fact that, apart from apatite, the only common non-
silico-aluminous, primary rock-making minerals are magnetite
and ilmenite; the iron being the only element of the important
ones which in rocks can crystallize without SiO, or Al,Q,.

13. Finally the common occurrence of SiO,, CaO and Na,O
in minerals of both the salic and femic groups, and their result-
ing interdependence.

CALCULATION OF THE NORM.

The method adopted by us of calculating the standard min-
eral composition is as follows:

1. Determine the molecular proportions of the chemical
components of a rock as expressed by the complete analysis, by
dividing the percentage weights of each component by its
molecular weight.

CLASSIFICATION OF IGNEOUS ROCKS 645

2. Before undertaking the distribution of the chemical com-
ponents as mineral molecules, small amounts of MnO and NiO
are to be united with FeO, and of BaO and SrO with CaO; of
Cr,O, with Fe,O,, unless these unusual components occur in
sufficient amounts to make their calculation as special mineral
molecules desirable. These amounts are indicated in connec-
tion with examples of calculation given later.

3. Establish the fixed molecules by allotting:

a) to Cr,O,, if present in notable amount, FeO to satisfy
GhemratioCr, On whic Or atc for chromite;

Veto Mi@ > enough Fe@™ to) satisfy the ratio TiO, : FeO™:
Ieee tormilmenite- li; there isexcess of I10,, allot. to it equal
CaO for titanite or perofskite according to available silica, to
be determined later. If there is still an excess of TiO, it is to
be calculated as rutile;

6) tov PR, ©, allot enough €aO\-to satisfy the ratio P,O, :
CaO :: 1 : 3.33 for apatite. Allot F to satisfy CaO = 0.33P,0, ;

ad) to F not used in apatite allot CaO to form fluorite,
Ca@eork 223322"

eto CleallotwNal © im thewratio Cl, : Na,(O) S2m03 rior
sodalite ;

7 ito= SO allotNa.© ine proportion SO, : Na,O Sear sn tor
_ noselite ;

g) to Sallot FeOin proportion S : Fe(O) :: 2:1 for pyrite;

h) to CO, in undecomposed rocks allot CaO in the propor-
tionyt : 1 tor calcite. (CO, may. occur in primary ‘caleite and
cancrinite. If these minerals are secondary, the CO, is to be
neglected, since it is understood that analyses of decomposed
rocks are not available for purposes of classification.

Having adjusted the minor, inflexible, molecules, there
remain the more important but variable silicate molecules, which
form the great part of the standard mineral composition, or
norm, of most rocks.

4. To Al,O, are allotted all the K,O and Na,O, not already
disposed iotjin the propertion of Al, O,)..K;O taNa Ores 1 :
1 for alkali feldspathic and feldspathoid (lenad) molecules.

646 CROSS, IDDINGS, PIRSSON, WASHINGTON

5- Withsexcess of All @. (Al O77 Ke OF Nas @)):

a) to extra: Al,.O, -allot,€aO in. proportion of Al, Ou: ‘€a®
>: I: 1 for anorthite molecules.

o) Wtthere is durther excesson Al; O. it isto ibe considered
as corundum, Al,O,.

It it is not advisable to calculate muscovite with excess of
Al,O, instead of corundum, since it requires a readjustment of
orthoclase molecules, and muscovite may not occur in the rock,
the extra Al,O, entering alferric minerals. Its calculation is
considered subsequently.

6) Withwnsufiicient, Al, @., (Al, O, << K.@ —=Na,O):

a), Extra Na, Ou1s allotted to FeO] in: proportion HexOr:
Na, O71) Tforacmite molecules:

6) If there is still extra Na,O it is set aside for a metasili-
cate molecule (Na,SiO,).

é) When: there is am excess.of K,Ovover Al, ©, itis treated
in the same manner. It is an extremely rare occurrence.

7. In working with reliable analyses in which Fe,O, and
FeO have been correctly determined :

a) To Fe,QO, is allotted excess of Na,O under conditions6,).

6) To remaining Fe,O, is allotted available FeO in equal
proportions for magnetite.

c) If there is any excess of Fe,O, it is calculated as hema-
tite.

Analyses in which all the iron has been determined in one
form of oxidation, when it occurs in two, are of little value when
considerable iron is present. When the amount of iron is very
small the analyses may still be used as a means of classifying
the rock. For this purpose all the iron, if given as ferric oxide,
is to be calculated as FeO, except that necessary to be allotted
to Na,O for acmite, and then used as below.

8. a) Extra CaO after the foregoing assignments is allotted
to (Mg, Fe)O in proportion CaO : (Mg, Fe)O :: 1 : 1 for diop-
side molecules.

In all molecules where (Mg, Fe)O is present, MgO and FeO
are to be used in the same proportions in which they are found

CLASSIFICATION OF IGNEOUS ROCKS 647

after FeO has been allotted to the molecules previously men-
tioned. That is, they are to be introduced into diopside,
hypersthene and olivine with the same ratio between them.

6) If there is still an excess of CaO it is to be set aside for
calcium metasilicate (CaSiO,) or subsilicate (¢CaO. 3 SiO,),
equivalent to wollastonite or akermanite. Such extra CaO will
in most cases actually enter garnet, an alferric mineral.

g. With insufficient CaO, (CaO < (Mg, Fe)O):

a) Extra (Mg, Fe)O is to be set aside for metasilicate or
orthosilicate, hypersthene or olivine, according to the amount
of SiO, present.

The allotment of SiO, to form silicates begins with the bases
which occur with silica in but one proportion, and is carried on
as follows:

10... Ko ZrO), allot Si@), 1 proportion of 1:: 1 for zircon:

11. To CaO and Al,O, in anorthite is allotted equal SiO, to
fonm~eCa@ Al On 2510)

12. To CaO and (Mg, Fe)O in diopside is allotted equal
Si@_-to form CaOi7( Mg, Me) O. 25103:

13. To Na,O and Fe,O, in acmite is allotted SiO, to form
Nag On Fe, On as1O7:

14. To Na,O (or K,O) set aside for metasilicate molecules
allot Si@; to:tormeNasOusiO, or Kj Orsi©,.

15. To Na,O and Al,O, in sufficient amount to form with
NaCl sodalite, or with Na,SO, noselite, is allotted SiO, to
satisfy the formulas : 3 (Na,O.AI,O, 2SiO,).2NaCl, sodalite,
Zina OP Al Or 2510), \i Na, SO) noselite:

The allotment of silica to bases which may form two or more
silicates, ortho-, meta- or polysilicate, is controlled in part by
the amount of available silica, in part by the affinities of the
bases for silica as explained below, in accordance with certain
commonly observed facts, as follows :

a) Quartz does not occur with nephelite and leucite, that is,
the feldspathic molecules will be polysilicates, orthoclase and
albite, if there is sufficient available silica.

648 CROSS, IDDINGS, PIRSSON, WASHINGTON

b) OF the feldspathoids (lenads), the metasilicates—leucite
and analcite—are rarer than the orthosilicates, nephelite and the
sodalites. Analcite is so rare as a primary mineral that it may
be omitted from the discussion, and may be regarded as com-
posed of albite and nephelite + 2 water. Nephelite is more
frequently associated with orthoclase than leucite with albite,
from which it appears that potassic feldspathic molecules become
polysilicate when sodic ones form orthosilicate. If there is not
enough available silica to form orthoclase, leucite forms.

c) The method is also based on the infrequent occurrence of
orthorhombic pyroxene with the feldspathoids, and the frequent
occurrence of olivine with these minerals. From which may be
deduced the ruie that the development of metasilicate or
orthosilicate of (Mg, Fe)O is controlled in most cases by the
available silica after satisfying the feldspathic molecules.

d) Finally, the occurrence of melilite (akermanite) in rocks
free from the polysilicate feldspars and the metasilicate, hyper-
sthene, indicates that this subsilicate mineral is produced because
of insufficient silica to form the lowest normal silicates com-
monly developed in igneous rocks.

The order of affinity of the common rock-forming oxides for
silica, which is well established by the foregoing and other facts,
as well as by such investigations as those of Lagorio* and
Morozewicz,’ is as follows, beginning with that which has most
affinity, KO} Nas©;CaO; MgO,’ BeO. ~ The oxides Al @ wand
Fe,O, to some extent are analogous to SiO, and play in certain
respects the rdle of acidic oxides.

The validity of this order is illustrated and confirmed by the
following facts: K,O and Na,O alone form polysilicates (ortho-
clase: and albite)) with the ratios (K, Na),O 2 Si@, a. 12 3G:
They also form the metasilicates (leucite, analcite and acmite)
with the ratio (K,Na),O: SiO, :: 1: 4. Na,O also forms the
orthosilicate nephelite, as well as the minerals of the sodalite
group, with the ratio Na,O: SiO, :: 1: 2. The corresponding

* TSCHENMAK’S Min. Petr. Mitth., Vol. VIII, pp. 421 ff., 1887.

? [bit., Vol. XVIII, pp. 221 ff., 1899.

CLASSIFICATION OF IGNEOUS ROCKS 649

potash compound, kaliophilite, is rare and not a rock-making
mineral, though K,O enters into the composition of nephelite to
some extent. If potash and soda are present and there is insuf-
ficient silica to form polysilicates of both, then as a general rule,
K,O takes all the silica it can get to form orthoclase, or ortho-
clase and leucite, the soda taking the rest to form nephelite,
together with albite in some cases. The occurrence of K,O in
the micas appears at first thought to be an exception to this rule,
but further consideration shows that in the muscovite molecule,
CHAS) O- Al On 2Si@F when Ti—2K the SiO), -is six times
K,O, as in orthoclase; and when H = K, the SiO, is four times
K,O, as in leucite.

CaO forms the orthosilicate anorthite with the ratio CaO:
SiO, :: 1: 2, and controls an amount of SiO, equal to itself in
the wollastonite molecule. Inakermanite the CaO: SiO, :: 4: 3.

MgO and FeO can, at most, control only their own amounts
of silica, in the hypersthene molecule, and also form the ortho-
Silicate olivine, with the, ratio (Pe, Mg) O: SiO, i: 2:1. Finally,
FeO crystallizes out in non-siliceous and non-aluminous min-
erals, magnetite and ilmenite.

It is noteworthy that the order of affinity of these oxides
for alumina is the same in relative order as that for silica. So
K,O forms muscovite with an excess of alumina over potash,
the analagous paragonite being rare and only found in meta-
morphic rocks, Na,O being slightly in excess of Al,O, in the
sodalites, CaO largely so (3:1) in garnet, and MgO and FeO
having little affinity for it.

MgO and FeO can, it is true, form the non-siliceous, alumin-
ous mineral, spinel, RO. R,Ogj.

In the calculation of the ania mineral composition the
allotment of silica to the alternative molecules is, therefore, as
follows :

16. To CaO set aside for wollastonite or akermanite is allot-
ted tentatively SiO, to form wollastonite (CaO.SiO,).

17. To extra (Mg,Fe)O is allotted SiO, to form orthosilicate,
olivine (2(Mg,Fe)O.SiO,).

650 CROSS, IDDINGS, PIRSSON, WASHINGTON

r8. TovAl, Ovand KIO Na, © is) allotted ’SiO; to: male
polysilicate, orthoclase, albite (K, Na),O.Al,O. 6 SiO;:

a) If there is excess of SiO, it is added to the orthosilicate
of (Mg, Fe)O to raise it to the metasilicate (Me, FelOrsiOn:
If SiO, is insufficient to convert all the olivine into hypersthene
it is distributed according to the following equations :

x + y = molecules of (Mg,Fe)O.
x + = available SiO,.

where — hypersthene, y = olivine molecules.

6) Further excess of SiO, is to be allotted to TiO, and CaO
to form titanite. These constituents remain as perofskite when
there is nosexcess.o0f SiOe*

c) Further excess of SiO, is reckoned as quartz.

2
1g. If there is insufficient SiO, to form polysilicate feldspar
out of all the K.O and Na,@ with Al,O.:
a) To K,O.AI1,O, is allotted tentatively enough SiO, to
form polysilicate, orthoclase (K,O.AI,0O,.6SiO,) and the
remaining SiO, is distributed between albite and nephelite

molecules by means of the equations:

x-+-+y =molecules of Na,O.
6x + 2y = available SiO,.

where z =-albite, and'y —-nephelite’ molecules.

6) If the available SiO, in case 15, 2) is insufficient to form
nephelite with the Na,O, then enough SiO, is first allotted to
the Na,O to form nephelite and the remaining SiO, is distrib-
uted between orthoclase and leucite molecules by means of the

equations :
xy —= molecules ot KO:
6x + 4y =available SiO,.
where « = orthoclase, and y = leucite molecules.

20. If there is insufficient SiO, to form leucite and nephelite
with olivine it is necessary to reduce a sufficient number of
molecules to form the subsilicate akermanite, 4CaO.3SiO,.

a) In case there is no wollastonite this is done after dis-

GLA SSIFICA TION OF IGNEOUS ROCKS 651

tributing SiO, tentatively to form leucite, nephelite and olivine
and noting the deficit of SiO, by means of the equation:

y =; of the deficit of SiO,.

y = molecules of akermanite. (4CaO.3SiO,). .
CaO is to be taken from diopside, and the MgO and FeO so
liberated are to be calculated as olivine.

6) In case an excess of CaO has been set aside for wollas-
tonite this is first converted to akermanite by means of the
equations: y= the deficit of SiO,.

y = molecules of akermanite (4CaO. 3SiO,).

c) If there are not sufficient molecules of wollastonite to
satisfy the deficit of silica, recalculate the molecules of diopside
and wollastonite so as to make olivine, diopside and akermanite
by means of the formule.

2ax+ 3y += = available SiO,.
xta4ay = molecules of CaO.
x z = molecules of MgO + FeO.
Where *=molecules of new diopside, y= molecules of aker-
manite (4CaO.3SiO,),and = molecules of olivine.

21. If there is still not enough SiO,,all the CaO of the diopside
and wollastonite must be calculated as akermanite, the (Mg, Fe)O
being reckoned as olivine and the K,O distributed between
leucite and kaliophilite by the equations:

“ty — molecules of K,O.
4x + 2y = available SiO,.
where + is K,O in leucite and y = K,O in kaliophilite.

22 In case there is insufficient SiO, and an excess of Al,O,
and (Mg,Fe)O, which might form aluminous spinel, an alferric
mineral, the excess of Al,O, is to be calculated as corundum,
and the uncombined (Mg,Fe)O is to be estimated as femic min-
erals, being placed with the nonsilicate, mitic, group, magnetite,
ilmenite, etc.

— It will be noted that as a result of the methods given above
the following minerals are not found together in the norm,
standard mineral composition, of igneous rocks; in other

652 CROSS; TD DINGS, -PIRS SON, WASHING LOM,

words, that the calculation of the former of each pair precludes
that of the latter, and conversely.

With quartz there will be no nephelite, leucite or olivine.

With hypersthene there will be no nephelite or leucite.

With corundum there will be no diopside or acmite.

With anorthite there will be no acmite.

With wollastonite there will be no hypersthene or olivine.

With leucite there will be no albite.

Percentage weights of minerals—Having estimated the rel-
ative number of molecules of the various mineral components,
their relative masses may be obtained by multiplying each by
the molecular weight. This is readily accomplished by means
of tables, both for finding the molecular proportions corre-
sponding to percentages of the chemical components given in
analyses, such as those lately published by J. F. Kemp* and
others, for finding the percentage weights of the minerals with
constant composition when their molecular proportions have
been calculated. The weights of minerals like olivine and
pyroxene in which the component (Mg, Fe) O is variable must
be calculated from the proportions of MgO and FeO present
in the rock after deduction of FeO allotted to magnetite and
ilmenite, the same ratio between these oxides being used for
each kind of molecule containing both of them. The weights
of diopside, hypersthene and olivine, in which MgO and FeO
occur in varying amounts, may be computed from the sums of
the simple molecules CaSiO,, MgSiO,, FeSiO,, and Mg,SiO,,
Keo On.

Tables for finding the molecular proportions of the constitu-
ent oxides, and those for the percentage weights of the standard
minerals will be found in the reprint of this paper, already
alluded to.

EXAMPLES OF CALCULATIONS.

It will be useful to give some examples illustrative of
the method of calculation and of the various possibilities,
selected from several thousand calculated by us. To simplify

"Kemp, J. F., “The Recalculation of the Chemical Analyses of Rocks,” Schooe
of Mines Quarterly, Vol. XXII, pp. 75-88.

CLASSIFICATION OF IGNEOUS ROCKS 653

them, the molecular weights of the minerals which have fixed
molecules, as orthoclase, anorthite, magnetite, etc., are multiplied
by the number of molecules of the unit oxide in each case to
arrive at the percentage weight.

For general purposes very small amounts of the component
oxides may be neglected. But for close work it is necessary to
take into account even small percentages of P,O,, TiO,, SO,,
etc., that is, where any one of them amounts to two units in the
scaleiot molecular proportions; when P,O, = 0.28, TiO, — 0.16,
SO; —/0.16, CO, — 0:08, Cl — 0.07, F' = 0.04 of a per cent.

The introduction of these into the calculation is important
in proportion as the amounts of the bases with which they com-
bine are small in the rock.

A check on the results is furnished by the agreement of the
sum of the calculated mineral components and the components
not included in the calculation with the sum total of the analy-
sis. This check cannot be absolutely exact because of errors
in the determination of the last decimal figure in the calcula-
tion of each component throughout the process.

The rocks chosen to illustrate the method represent several
cases. That of the toscanose (granodiorite), Table I, is the sim-
plest case. Al,O, and SiO,, being in excess, yield normative
corundum and quartz; the femic silicate being hypersthene.

The hessose (amphibole-gabbro), Table II, illustrates the
method of adjusting SiO, between hypersthene and olivine.

The nordmarkose (litchfieldite), Table III, illustrates the
method of distributing SiO, between albite and nephelite after
reckoning the femic silicate as olivine. All the Na,O is first
allotted to Al,O,, and SiO, is allotted to orthoclase, anorthite,
and olivine; the remainder, 0.680 mol., is distributed according
to the formule.

In the case of the laurdalose (laurdalite), Table IV, insuffi-
cient Al,O, necessitates the formation of acmite molecules.
After allotting SiO, to orthoclase, acmite, diopside, and olivine,
the remainder, 0.522 mol., is distributed between albite and
nephelite.

654 CROSS, IDDINGS, PIRSSON, WASHINGTON

The vesuvose-albanose (leucitite), Table V, illustrates the
case in which akermanite (melilite) is required to satisfy the
calculation. After allotting SiO, to leucite, nephelite, anorthite,
diopside, and olivine, there is a deficit of 0.069 mol. SiO,. This
is adjusted by the introduction of akermanite according to the
formula, and the recalculation of diopside and olivine. It also illus-
trates the case of an intermediate rock, between Classes IJ and II.

AWASBASS Ir

TOSCANOSE (GRANODIORITE) EL CAPITAN, YOSEMITE VALLEY, CAL.
Bull, 168,..U. S.. Geol. Surv., p. 208.

Per cent, | Mol. Ilm, Mag. Orth. Alb. An, | Cor, Hyp. | Quartz.
SHOREoe cll eee Ole) | to JUSS eente-0 see: 258 342 Oy 5 ose 27 464
IMO mal CEO) SSO Caco thane 43 57 47 9
HesOne: TOD sl eOOAl| tant: 4 pNee 5 saates hemes niche ai
He@re Dasa Oe, 3 4 ieblehs Neecs coe Resets II
MeOne: SrA Ws ta(O)) 67a ire ON se ate Lica Aor att Phe oie 14
Ca@i 7) OO} || -aoy-ty/ Passa lla ae Runde ae Nat 47 ail
Na,O.. eS Aaalb OSifnalienee che Taps dee (57) eae
Ke Ore AOS) als OA2', lsrecae Sees A eae
H,O-+ 3 Of esticrAecn elf walerercial | Maver ssp beware
20. || none
COR tex tr Boer ol Meenceae
ABOR ar D2 2a OOR 3
ep O) irs -10 | .000 ;
SOw ce MONE |i everens
Cle vseias SOD OOO pl ccssscereens| pe ieee Ni costars ill rere oa lhesepere Reach eater ie
MnO... LSS ruil eae OO 2s |erener ses ter SAI eeeet ea ae areas wae 2
Ba@ 2. .04 | .000 :
SIO ne VO Zig esOOOE Il seemacsee deere allare’|eercheia,-a, Mill weyescacean tet et cu cea | ueraine ace wall naearoees
Ts Oe LOSei pe OOO sc ceaccial Wer Gr tall, Ssete.sel|| US Syonoy ao lecteur ein | ureter eem | eee,
100.60
FORMULA, MOL. WT. NORM,
SiO, - - - - - 464 X 60 = quartz 127/04, ©) 27.84 |
K, Peace, .65i0, : - - 43 X 556 = orthoclase = 23.91
Na, 20.A op, 6SiO, - - 57 X 524 = albite = 29-87 F 66.85 | Sal 95.61
CaO. Al, 20; 2. $i0, - - - 47 X 278 = anorthite = 13.07
Al, 26) Se. a - : = 9 X 102 = corundum =" Roz C a)
Mg iO, - - = de LOO; | an wee
(Fe, Mn)O.SiO, 4 ie ; aA Se Seals hypersthene = eine 12 3-11 a
FeO. Fe.Q3 - - - 4X 232 = magnetite = 23 M 8 & 4-49
FeO. TiO, - - - - 3 X 152 = ilmenite 1) 7S t3
100,TO
CLASS I, ORDER 4. RANG 2. SUBRANG 3.
Sal = 95° Onee7. F 66.8 7 K,O’+Na,O' _ 100 5-7 K,0' _ 43 -5 3
Fem Bee Q°> 27 Teo a CaO’ w47 ae REAC! ge

Persain ne. Britan Be re. Toscanase. Toscanose.

CLASSIFICATION OF IGNEOUS ROCKS 655

TABLE II.

HESSOSE (AMPHIBOLE GABBRO) BEAVER CREEK, BIG TREES QUAD-
RANGLE, CAL.

A. J. S., Vol. VIII (1899), p. 297; Budl., 168, U.S. G.S., p. 206.

Per Cent. | Mol. | Apat.| Ilm. | Mag. | Orth,| Alb. | An. | Diop. Remy Hyp. | Oliv.
SOs. vesoll y/s27 sfsksy {I a. se Be U2 |'2604) BOG) LIS? | Soh | eszanlhe SL
SOR eee 20 so2 2 OMY Navas ate Se 2) 44 | 158 se ie a Be
WEAO)n 6 ai 1.85 -OI1 a ou II a
Meme ten Uren) (ye a | | 3, 2 hse |e ar [eee
CAO Sasa. 13.02 P2225 eT Sale eas ah ae See |iees Ko)
Nias Oe eS .044 Re as Te a 44 a as
ISO) sear 22 O02 7s Name es HE Diels
H,O+ .. Wes2y 7s
H,O— .. o8 snr a a
IMKO)S cae .92 .OI1 fe II
IO nte.65 6 .74 .005 5
Clee ata: tr Nae :
Gin tr
MnO tr
Sr@O™: tr
WoO 02
FeS» 20
99.86

x +2 = 88 SiO,
z+ y = 139 (Mg, Fe) O

Ve—T1O2
x= 37
FORMULA, MOL. WT. NORM,
K,0.A1,0,.6Si0O,_— - - - 2% 556 =orthoclae = aco
Na,O.Al,0,;.6Si0, = - - 44 X 524 = albite = 23.06{ F 68.09 Sal 68.09
CaO.Al,0,.2SiO.— - - 158 X 278 =anorthite = 43.92
CaO.SiO, - - - - 59 X 116
MgO.Si0, .- - - 48 X 100} = diopside _ 13.10}
FeO.SiO, - - - = Er X32 |
j Se Stee We aes cry ae oe S ae = hypersthene = 3.92 { P +O 24.77
2MgO.SiO, - - - 83° Jor!) ae. Dink Fem 30.74
AROHO 6 Ae RS ee = olivine ey auc r
FeO.Fe,0, - - - - II X 232 =magnetite = ae M Bs
FeO. TiO, - - - - mr X 152 =ilmenite = 1.67 o? i
ACO IPO), c - - - 5 X 310 = apatite —_ se | iN : J
FeS, - - - - - pyrite = .20 “75
etc. - 1.37
100,20 X
CLASS Il. ORDER5. | RANG 4. SUBRANG 3.
Saliva 08500 7.5005 Bis 08 ,00)na7, K,O'+ Na,O' _ 46 ey au ake KO; 2 I
Pom sega nee S Ome a an Ca0! marco 5 NaSORT ca Sa

Dosalane. Germanare, Hlessose. Flessose.

656

CROSS, [DDINGS, PIRSSON,

TABLE III.

WASHINGTON

NORDMARKOSE (LITCHFIELDITE), LITCHFIELD, ME.

Bull. 168, U.S. Geol. Surv., p. 21.

Per cent.
SOG io avon 60.39
AT Oi ees « 22.57
Fe,Og .42
He @eaeacres 2326
Mig Oren “13
CaO: 132
Nas@aeee. 8.44
eae Ae 77,
2
AO 37
Vin Oe: .08
99.95

10

Mol. Orth. | Alb. | Nep. An,
612s “768
——_,,--——
1.006 300 680
To2") |) 34
——--—“
.221 50 136
.003 ale é
.030
.003
.005 ues 2s
102 34
——_{,---—S
"126 tye 136
.050 50
.OO1

r.006—(0.300 + 0.010 + 0,016) =0.680
64 + 2y = 680 (SiO,)
z+ y= 136 (Na,O)

t=

BL eis

102

Cor. Mag. Oliv.
16

30 teh
a 2 ie
3 27
3

In the rock the extra Al,O, enters mica, combining with the
olivine and magnetite molecules and because of low magnesia
making lepidomelane.

FORMULA.

K,0.AI, O3 .6SiO, - -
Na, 20. Al,O
CaO. Al,0,. SiO wae
Na,O.Al,O,. 2Si0,
Al, On - -
FeO, Fe,0, - -
2Mg0.SiO, - -
2FeO.SiO, - -

CLASS I,

95/357
3 Bios

Persalane,

Sal’

Fem

3-6SiO, -

Bos
16

ORDER 5.
82.64. 7

heen Wa

Canadare.

‘ = olivine

NORM,
orthoclase
albite
anorthite
nephelite
corundum
magnetite

Hl dl wet

Hl tu

H,O

RANG I.
K,0O'+Na,O' _ 186
CaO’ Was
Nordmarkase.

F

L Sal 95.35

(G;

M

Oo Fem 3.81
SUBRANG 4,

K,O'

Na,O' = ee Se

Nordmarkose.

CLASSIFICATION OF IGNEOUS ROCKS 657

AB ICn, TV;

LAURDALOSE (LAUDALITE), WEST OF POLLEN, LAUGENDAL,
NORWAY.

BrOGGER, W. C., Die Eruptivgesteine des Christiania Gebietes, Vol. III, p. 19.

Per cent. Mol. | Ap. | Ilm. | Orth. | Alb. Nep. | Acm.| Mag.| Diop.| Oliv.
366 | 156
SS ee
SiO, 56.35 -939 336 522 16 62 3
61 |. 78
——-~)-— XY
Al;O, 19.85 -195 56 139 sc =
Fe,O, I.QI .O12 sr Sis é vie 4 Sol vers
GOV aelgns 2.03 .028 12 ae 8 Gi I
MgO Ti oL7, .029 me 24 5
CizOr.,. 2.60 .046 | 15 siete aye 31
61 78
Sena een)
Na,O 8.89 .143 Sie 139 4
ik Onewa 5.31 .056 56 ‘se
H,O y/o) rae ue nok
EKO Fc 1.00 sOUZ I) as 12
BOR. .67 .005 5 ae Bi
100.68
62 + 2y = 522 (SiO,)
a+ y=139 (Na,O)
a Or
778
FORMULA, MOL. WT. NORM,
K,0.AI,03.6Si0O, - - 56 X 556 = orthoclase = 31-14) p ¢
Na.O. Al, 203.6SiO, - 61 X 524 = albite -- 31.96 ee Sal 8
Na.O.Al, “ON .2510, - - 78 X 284 =nephelite = 22.15 L 22.15 a: 5-25
Na,O. Fe,0,. 4Si0. - 4% 462. '—acmite = 1.85
CaO.SiO, - - 31 X 116 P 8.77)
} Med, SiO, - - = 24 X 100 = diopside = 6.92
FeO.SiO, - - ap eaey)
SHO SiO, fue Een . a2 olivine -46 O 46 Fem 14.45
FeO. Fe, 0; - - 8 X 232 = magnetite = 1.86
FeO. TiO, - - =u t2) 52h —almenites == 1,82 M 3.68
3CaO.P.0, - - Sip 3r0=8 apatite, —— 1.55 A 1.55
iE Oe .70
100, 41
CLASS II. ORDER 6, RANG I, es 4s
Sal _ 85.25 cy K63.10 5 K,O'+Na,O' _ 195 7 K,0" _
Fem 14. wee ee 1 = Eee ree ee CaO’ ey rae Na,O' =2<3 Ss
Dosalane. Norgare. Laurdalase. Laurdalose.

658 CROSS, IDDINGS, PIRSSON, WASHINGTON

TABLE V.
VESUVOSE-ALBANOSE (LEUCITITE). CAPO DI BOVE, ITALY.
A, J. S., Vol. 1X (1900), p. 56.
Tentative. i Final.
Per cent. | Mol. | Ilm. | Leuc.] Nep. | An. | Mag eo
Diop.| Oliv.| Q | Ak.t | Diop.| Oliv.
510% || 45-99.||.767 276i v7Oe |) 7S 298| 14 | 69 | 69 | 114] 60
AIS Oat 7/2 eos OAs | PUBS Ue 2 Oily cr alhoees er [a tewee |) Sarai [aches syol lteter
Ke,0, AN STi 9 | OZ OF a sere tenetens Be CIO eA eee eee Prices ieee
FeO.. 53 On OSU iS 26 37 7 14] 30
MgO.. SesOmlMigenls. Bah Pre IN 2) | oNk ate 43 90
CaO arora ziGro7 Speer ltnnteXa) .| 148 g2 56
NasOel 7) 2218 |s02'5 Bop eet syial tanner ii Leet eat
ISO) 5c 8.97 1.094 OG Parsee
H,O . AIS So [eacetbcns| tenn cea [Pease ye
MOn: 3371x005 1 5
MnO.. tr Sorc reset fe fe
BaO.. -25 |.001 I I
Sr@ sai, none an?
100.65
: 69
‘Mol. of akermanite = y = = 122,
K,0. eS 6 MOL, WT. NORM
48iO, 94 X 436 = leucite = 40.98 | oe
Na,O. Al 1,0,2Si0. - 35 X 284 = nephelite = see P0082 ' SAGES
CaO. Al,O,. zSi@s - - 39 X 278 = anorthite = 10.84 F = 10.84 ‘7
CaO.SiO, - - 57 X 116
} MeO, SiO, - 43 X 100 ¢ = diopside = 12.70.) Py— "12.70
FeO SiO. - - 14 X 132
2MgO.Si - - QOR GIN 7O}/ (asl yemese ic
§ 2FeO. SiO. - - - = Ol GITO2 s\n olivine = 9-36 ‘ O = 18.65 > Fem 38.20
4CaO.35i0, - - 23 X 404 = akermanite = 9.29
FeO. Fe,0, - 26 X 232 = magnetite = 6.03 ‘ MoaiG
FeO. TiO, - - - - 5 X152 = ilmenite = 76 pees)
Ee Oa 5
100. 4I
CLASS Ill, ORDER 6. RANG 2. SUBRANG 2,
Sal = 6170 503 re 10:84 573 K,O'+Na,O' _ 129 -7~ 5 K,O' _ 94 5
Fen es oe 1 Gers eatge ce CaO' Fay Skee Na,O' ae Pix
Dosalane- Campanare- Vesuvase- Vesuvose
salfemane, bohemare. albanase. albanose.

The janeirose (pseudo-leucite-sodalite-tinguaite), Table VI,
illustrates the method of calculating leucite and orthoclase,
as well as sodalite and noselite. After allotting SiO, to nephelite,
acmite, diopside, and olivine, the remainder of SiO, is 0.546 mol.

CLASSIFICATION QF IGNEOUS ROCKS 659

TABLE. VI.

JANEIROSE (PSEUDO-LEUCITE-SODALITE-TINGUAITE ) BEAVER CREEK,
BEARPAW MOUNTAINS, MONT.

Bull. 168, U. S. Geol. Surv., p. 136.

*”

Per cent. Mol, | Im. |Fluor.| NaCl. Bee Cal” Orth, | Leuc.] Nep.t) Acm.| Diop.| Oliv.
4. | cites
390 | 156
—~--Y
SHOs 51.93 866 546 192,-| 88 |" 38 2
65 | 39
— Tr’
ALO: | 20.20 200 104 OO trae
Hen Or 3.59 .022 3 cs 22 ee ae
HeOh. 1.20 .018 2 14 2
MgoO.. G22. 006 ae 5 I
CaAOe. 1.65 0) 7 Ms ae 6 etal ene PA ae 17
Na,O 8.49 137 ime) Che hae seb leo 96 | 22
65 | 39
er’
IO) ge 9.81 104 104
H,O-+ IO :
H,O— -99
iO}. .20 002 2
PO: .06 000 se
STO: .07 OOI I
BaO.. .09 OOo! 6 I
SOer: 597) 009 9 as
COs £25 .006 M Boe 6
Gi. .70 .020 Fist || 20
TP 27 .O14 14
100.58 :
27 ae BPM ese “As
100.31 i | As 2
* For nephelite, sodalite, and noselite.
6x + 4y = 546 (SiO,)
x+ y= 104 (K,0)
OS,
d= 39
FORMULA. MOL, WT. NORM,
K,0.A1,03.6SiO, = - - - 65 X 556 = orthoclase = 36.14 F 36.14
K,0O.A1,0,. 48i0, - - - 39 X 436 = leucite= 17.00 Sal. 82.75
Na,O. Al, O,.25i0. - 48 X 284 = nephelite = 13.63 4 7 BRE
A(NakOvAILO2Si0,)2(NaGI to 669) = sodalite 9.69 Gk
2(Na,O. Al,O3. 2SiO,)Na,SO, 9 X 699 = noselite 6.29
Na,O. Fe,O9. 4SiO2 - 22 X 462 acmite 10.16
CaO.SiO, - - - 17 X 116 Pt r4r4s
MgO. SiO, - - - - - 5 X 100 diopside 4.32
Et eae ate Fem, 15.60
2MgQ0.51 - - co el oe Xa) e em, 15.
2FeO.SiOy - - - - aX oh olivine #4 oat
FeO.TiO, - - - - = 2X 152 ilmenite .36 M 30
CaF, 3 zs = < = 7X 78 fluorite 355 A “55
CaOxXCOs - =e ie = =" OX TCO. calcite .60
H,O 1.09

660 CROSS, IDDINGS, PIRSSON, WASHINGTON

CLASS Il. ORDER 7. RANG I, SUBRANG 3.
Sal _ 82.75 3 OSTA 5 eg K,0'-- Na,O'-. 219 7 Ke On 104
Fem I5. ae ae 16, ICE Sone CaO’ Meo Zar Na,O' are =e
Dosalane. ltalare. Lujavrase. Janetrose.
na

The calculation of the norm from the mode—— Having described
the process by which the norm may be calculated from the chem1-
cal analysis of a rock, there remains the discussion of the process
by which it may be calculated directly from the actual mineral
composition of the rock without having the chemical analysis.

The first requisite in this case is a knowledge of the actual
mineral composition of a particular rock, and it is evident that
not every rock is sufficiently well crystallized to permit even an
approximate estimate of the kinds and quantities of all the min-
erals present. Consequently there are very many rocks in
which the norm cannot be calculated directly from the rock
without recourse to a chemical analysis. These are partly glassy
rocks, and those that are so fine-grained that the individual min-
eral components cannot be identified and measured.

But it is possible with some rocks to determine very closely
the proportions of the minerals present in them. Such rocks are
holocrystalline, and the crystals are sufficiently large to permit
their individuality and outline to be recognized. With such
rocks the method of determining the quantity of all the mineral
components is as follows:

Estimate by accurate measurement the volumetric propor-
tions of all the component minerals. This may be accomplished
by measuring with a micrometer the diameters of each crystal
in lines across thin sections of a rock, care being taken to
measure a distance at least one hundred times the average grain
of the rock.’ The proportions found for the lengths of diameters
of the various components will correspond to those of their
volumes. Several other methods have been devised which are
less accurate and need not be described here.

The volumetric proportions are to be reduced to relative
masses by multiplying the volume of each mineral by its specific
gravity and reducing the total to one hundred parts.

TROsIWAL, Verh. Wien. Geol. Reichs-Anst., Vol. XXXII, pp. 143 ff., 1898.

CLASSIFICATION OF [GNEOUS ROCKS 661

Accurate quantitative determination of the mineral compo-
nents of rocks by optical methods is difficult with coarse-grained
and coarsely porphyritic rocks as well as with extremely fine-
grained ones. When the rock contains large crystals a com-
paratively large area of it must be measured to obtain correct
proportions of the component minerals. A few thin sections
are not adequate. The measurements must be made mega-
scopically. The same is true when there are large phenocrysts.
A sufficiently large area of surface must be measured to furnish
a correct estimate of the relative proportions of the several kinds
of phenocrysts and the groundmass. Subsequently the ground-
mass may be studied and measured with a microscope and the
two sets of measurements combined.

In very fine-grained rocks, where the kinds of minerals com-
posing them can all be identified, the accuracy of measurements
of the diameters of crystals with a microscope is affected by the
overlapping of crystals within the section, and it is found by
experience that the amount of the colored crystals is overesti-
mated, while that of the colorless ones is underestimated. This
is particularly the case where the thickness of crystals is a frac-
tion of the thickness of the rock section, as with microlites and
minute inclusions. It will be necessary to determine corrections
to be applied in such cases by working on microcrystalline or
microlitic rocks whose chemical composition has been deter-
mined.

If all of the minerals actually present in a holocrystalline
rock are standard minerals, salic or femic, there may still be
uncertainty as to the norm, since the proportions of the standard
minerals actually developed may not accord with those consti-
tuting the norm. In all cases it is necessary to calculate the
norm from the actual mineral composition quantitatively deter-
mined by estimating the chemical composition of the rock from
that of each of its component minerals, and from this analy-
sis deducing the norm as in the first method described.

This involves the determination of the chemical composition
of the actual minerals present in the rock. Fora certain num-

662 CROSS, IDDINGS, PIRSSON, WASHINGTON

ber this may be based on optical investigation. The com-
position of minerals with constant or comparatively simple
molecules may be taken as that of the ideal molecule, as, for
example, in such minerals as quartz, orthoclase, albite, anorthite,
leucite, nephelite, apatite, zircon, titanite, etc.

The proportions of the chemical components reckoned as
oxides belonging to each of these minerals must be multiplied
by the percentage weight of each mineral to furnish the chem-
ical components of the whole rock. Thus 35 per cent. of quartz
== 35 per cent. of SiO, 5 lO percent, of orthoclase,— 074 77.pen
cent. of SiO), 133 percent. Al O7, 1.69 per cent 1G. ©.

When the mineral has no fixed chemical molecule, as, for
example, olivine, in which Mg and Fe are variable, it is neces-
sary to consider the composition most likely to obtain for the
mineral in a rock of about the character of the one under inves-
tigation, or to observe more specifically the optical properties
of the mineral where these are characteristic of the chemical
composition.

For the plagioclase feldspars the optical properties have
been elaborately investigated and are well known. It is neces-
sary to determine as accurately as possibly by Michel-Lévy*
methods, aided by Becke’s? method, the composition of the stri-
ated feldspars, noting the variation in zones, and estimating
approximately the average composition of the crystals. The
ratios of Ab to An must be transposed into ratios between
Na,O and CaO by halving the value of Ab. For the reason
that Ab stands tor the formula NaAlSi,O, and An for the for-
mula CaAl,Si,O,- In the first there is only one, Na, hence Ab):
An::Na:Ca. When we express the composition of albite by
Na, 0. Al, O; -Osi0,).we are using 2AbP “And in’ estimating,
Na,O from the albite combined with anorthite in a plagioclase
feldspar we obtain one-half as many molecules of Na,O as we
have Ab derived from the familiar symbol of a plagioclase,
JE OSE Tae

«Etude sur la détermination des feldspaths (Paris), 1% fasicule, 1894; 2™° fasi-
cule, 1896.

? Sitzungsb. Akad. Wiss. Wien., Vol. CII, Part I, pp. 358-76, 1893.

CLASSIFICATION OF IGNEOUS ROCKS 663

A less precise method may be used, which is sufficiently
accurate considering the inaccuracy of the approximation to the
average composition of zonally built feldspars, namely, to apply
the ratio Ab / An obtained optically directly to the percentage
weights of plagioclase. Thus 50 per cent. of plagioclase, whose
average composition has been estimated at Ab,An,, may be
separated into 30 per cent. albite and 20 per cent. anorthite
with approximate correctness.

In the case of hornblende, augite, and mica, the color and
optical properties are undoubtedly equally characteristic, but
they have not received sufficient attention to permit of the same
direct application. They are, however, guides to the choice of
typical formulas or chemical analyses, which may be used tem-
porarily in place of more exact methods.

For this purpose tables have been arranged giving the
chemical analyses of certain rock-making alferric minerals, and
also the analyses of the rocks in which these minerals occur.
From these it is possible to select cases corresponding more or
less closely to those in the rock whose norm is to be deter-
mined. For it appears from data already at hand that the
chemical composition of each mineral in a rock bears sucha
relation to the chemical composition of the whole rock, that
minerals of the same kind, for example the hornblendes, when
they occur in similar rocks have very nearly the same composi-
tion. The compositions of the hornblendes and micas in the
granodiorite of the Sierra Nevada, California, and of those in
the very similar quartz-monzonite of Butte, Montana, are nearly
the same.

For minerals with variable composition, then, select from
Tables of Analyses the analysis of the mineral corresponding
most closely to the one in the rock in question, considering
both its optical characters and the general character of the rock
in which it occurs, and reduce its several chemical constituents
to the proper amount by multiplying by the percentage of the
mineral as determined from the rock.

Having reduced each mineral to its chemical constituents

664 CROSS, I(DDINGS, PIRSSON, WASHINGTON

the sum of the constituents will represent the chemical compo-
sition of the rock. From this the salic and femic minerals may
be calculated as in the first instance, and the norm ceter-
mined.

In many cases it will not be necessary to reduce all of the
minerals to their chemical constituents in order to determine
the norm. For the minerals may be largely salic or femic as
they occur in the rock. This is the case with the feldspars,
feldspathoids (lenads), and quartz, with orthorhombic pyroxenes,
olivine, diopside, magnetite, etc.

It is chiefly the aluminous ferromagnesian minerals that
require reduction. And for first approximations the proportions
of salic and femic constituents contained in these minerals are
given in Tables XII, XIII, and XIV at the end of this Part.

From these it is seen that in aluminous pyroxenes the pro-
portion of the salic component is often 0.10, and does not
exceed 0.23. The femic component is 0.90 in most cases, and
farely) 0.77.

For hornblende and closely related amphiboles the salic
component is from 0.20 to 0.34, and in the more sodic amphi-
boles it is about 0.10.

For micas the two components are nearly equal, that is, 0.50
each.

When the amounts of salic and femic components are known
the Class of the rock is established. For rocks of the first three
Classes. the next step is the determination of the relative
amounts of quartz, feldspar, and feldspathoid. For persalane
rocks this is comparatively simple, since the other minerals are
present in small amounts. But as the amount of mica, amphi-
bole, and augite increases it becomes necessary to determine
more accurately the nature of the salic component involved in
each, in order to adjust the silica before reckoning the propor-
tions of quartz, feldspar, or feldspathoids. In such cases the
first method must be followed, or the aluminous ferromagnesian
minerals must be resolved into aluminous and non-aluminous
portions and the silica adjusted to these portions and to the

CLASSIFICATION OF IGNEOUS ROCKS 665

other salic and femic components, according to the method
given for calculating the standard minerals.

The determinations of Rang and Subrang as well as of Grad
and Subgrad follow the calculation of the chemical components.

THE CALCULATION OF THE MODE FROM THE CHEMICAL ANALYSIS
OF A_LROCK.

This is often desired in order to compare the two, and to
determine the amount of the chemical composition of some of
the component minerals, not otherwise determinable.

But it is evident that the knowledge of the actual minerals
present in the rock cannot be learned directly from the chemical
analysis, since various mineral combinations may be formed
within certain limits. It is therefore necessary to determine the
presence of these minerals by a study of the rock, and this
involves microscopical investigation in nearly all cases, This
has generally meant the simple enumeration of the kinds of
minerals present, with the crudest statement of their relative
proportions.

In some instances this would furnish data enough for the
solution of the problem, but it is clear that such cases must be
those in which all the minerals present are salic and femic, with-
out intermediate, alferric, kinds, or those cases in which the alfer-
ric minerals are developed to their limit, as for example, horn-
blende present to the exclusion of diopside, or to the exclusion
of hypersthene and olivine; biotite present to the exclusion of
hypersthene and olivine, or to the exclusion of potash-feld-
spathic minerals, or to the exhaustion of available alumina.

Under such circumstances it would be possible to calculate
the composition and proportions of the actual minerals.
But even in these cases the variable chemical character of
amphiboles and micas renders the solution of the problem
untrustworthy unless the composition of the particular amphi-
bole or mica be known exactly or approximately.

This involves the separation and analysis of one or both
of these minerals, or the reasonable assumption that they have

666 CROSS, IDDINGS, PIRSSON, WASHINGTON

approximately the same composition as other amphiboles and
micas which have been analyzed.

Further consideration of the problem will convince one that
where augites, amphiboles, and micas occur with femic minerals,
such as diopside, hypersthene, and olivine, the problem cannot
be solved by simply determining the kinds of minerals present
in the rock. The algebraic equations involve too many unknown
quantities. In other words, there may be variable amounts of
the same minerals developed from chemically similar magmas.
It becomes necessary then to determine the relative amounts of
several of these minerals, according to the number of them, in
order to reduce the number of unknown quantities in the alge-
braic equations. Then it is possible, with part of the problem
solved by microscopical study, to complete it by estimating the
remaining factors from the chemical analysis of the rock. That
is to say, in most cases the microscopical and chemical methods
must supplement one another.

Thus it is possible to calculate the probable composition of
a hornblende in a given rock when all the other minerals have
comparatively simple, or fixed, molecules, and when the quantity
of the hornblende has been determined optically. In another
case, if hornblende has been separated from the rock and
analyzed, it is possible to calculate the probable composition of
a biotite present, when the proportions of these two minerals are
known, and the other minerals in the rock have fixed molecules.

The same process may be used to determine the composition
of the groundmass, when the character and percentages of the
phenocrysts have been determined.

The method of calculation, which is illustrated by the case of
the Butte granite given on a subsequent page, may be stated as
follows:

Starting with the chemical analysis of the rock, reduce it to
molecular proportions by dividing the percentage of each chemi-
cal component by its molecular weight.

Deduct from these molecules the molecules belonging to such
minerals as have been chemically and quantitatively determined.

CLASSIFICATION OF IGNEOUS ROCKS 667

The remaining oxide molecules are to be distributed among
the minerals with fixed molecules, whose quantity, however, is
undetermined, by assigning to each mineral its proper oxide
molecules in the proportions in which they occur in the ideal
mineral molecule, as, for example; for orthoclase, 1K,O.
1Al1,0,.6Si0,, and further, by indicating the number of mole-
cules of each mineral by an algebraic symbol or letter. Thus, if
there are x molecules of orthoclase, there must be assigned to
then Oz Al On6z7s510.. “The sum of all these assigned
oxide molecules must equal the total amount of a particular
chemical component in the rock after deducting that belonging
to the minerals whose composition and quantity have been deter-
mined.

If there are more variable quantities than equations, that is,
in general, more kinds of minerals than chemical components, it
is necessary to reduce the number of unknown quantities by fix-
ing the relative amounts of several minerals, or by stating their
actual amounts.

THE CALCULATION OF ALFERRIC MINERALS.

As pointed out in the discussion of the classification of
igneous rocks on a basis of salic and femic minerals, the reasons
for omitting the alferric minerals are their variable and complex
composition, our inadequate knowledge of the chemical charac-
ter of the amphiboles and micas, and the inconstancy of their
crystallization from magmas of any given chemical composition.

The questions naturally arise, How may these minerals be
introduced into the calculation of the mineral composition of
rocks from rock analyses ? And what modifications of the norm
would follow their introduction ?

These questions, though not important for the classification
of rocks according to the system here proposed, are of interest
because of their relation to the general problem of chemico-
mineralogical classification and the possibilities of its assuming
a more elaborate form than that given it by us. The discussion
of them emphasizes the relations between the aluminous ferro-

668 CROSS, IDDINGS, PIRSSON, WASHINGTON

magnesian minerals and the salic and femic minerals, and makes
evident the effect of the crystallization of the alferric minerals
in a given magma upon the proportions of the other minerals.

We shall consider the problem of converting portions of salic
and femic minerals, already calculated from the analysis of a
rock, intoalferric minerals. The minerals in question are: alumi-
nous pyroxenes, aluminous amphiboles, micas and garnets. Rarer
minerals of this kind will not be considered. As to the chemi-
cal composition of these minerals, it is known that the amounts
of alumina and ferric oxide in monoclinic pyroxenes, amphiboles,
and micas vary considerably in different cases, so that no simple
statement can be made regarding the ratios of alumina to other
constituents in these minerals. Moreover, there is no fixed rela-
tion between the amount of aluminous ferromagnesian minerals
actually crystallized in a rock and the chemical composition
of its magma. However, a study of the chemical composi-
tion of these minerals, so far as they have been analyzed, shows
that their composition bears some relation to that of the magma
from which they crystallized. In order to present this relation-
ship as clearly as possible, Tables XII, XIII, and XIV have been
arranged, as already mentioned, giving the chemical analyses of
aluminous pyroxenes, amphiboles, and micas, and those of the
rocks from which they were separated. It is evident from these
tables that there is need for much thorough chemical investiga-
tion of the rock-making pyroxenes, amphiboles, and micas before
we shall be in a position to cope successfully with the problem
before us. For the present we can explain the method of deal-
ing with the known factors in the problem and indicate that
which may be pursued in doubtful cases.

The solution of the problem involves the transfer of alumina
from salic to femic molecules, the necessity of introducing it in
proportions corresponding to the known composition of these
minerals in each case, the consequent readjustment of molecules
among the femic minerals, and the disarrangement and readjust-
ment of molecules among the salic minerals.

The process appears at first sight complex, but only involves

CLASSIFICATION OF IGNEOUS ROCKS 669

simple algebra, and when applied to a concrete example is not
very intricate. Its value consists in familiarizing the student or
investigator with the interdependence of the various mineral
molecules in an igneous rock, and with the ranges of variation
possible within rocks of the same chemical composition.

The minerals in question being characterized by a variable
content of alumina, it is of first importance to note the amount
of alumina present in different cases and to consider what trans-
fer of chemical elements from salic to femic molecules would be
necessary in order to produce such aluminous ferromagnesian
minerals without destroying the stoichiometric proportions in
the remaining salic and femic minerals.

In the case of a rock in which the calculation of the standard
mineral composition showed the presence of an excess of Al,O,
over that required to form salic minerals it is evident that Al,O,
may be introduced into the ferromagnesian minerals by trans-
ferring it from this extra Al,O,.

But in the great majority of rocks there is no excess of Al,O,
in the sense here employed, and the production of aluminous
ferromagnesian minerals affects the standard feldspathic mole-
cules, so that the transfer of Al,O, necessitates the transfer of
those chemical bases united with it in equal proportions, namely,
calcium, sodium, and potassium.

Each of these elements may enter into the composition of the
minerals to be developed, whose molecules are more complex
than those of femic minerals. A study of the analyses of the
minerals in question shows that sodium enters into aluminous
amphiboles independently of the acmite-riebeckite molecule
(Na,O. Fe,O, .4SiO,), and that potassium also does, but to so
small an extent in most instances that the error of estimating all
of the alkali molecules as Na,O is negligible. Conversely,
potassium enters largely into the micas, and the sodium present is
so small that the alkali molecules in mica may be calculated as
though wholly K,O without notable error.

Several chemico-mineralogical relations appear to control the
amount of alumina, lime, and alkalies that may be transferred

670 CROSS, IDDINGS, PIRSSON, WASHINGTON

from salic to femic minerals to form alferric minerals.. And
since the transfer of Al,O, involves the CaO, K,O, and Na,O
to various degrees, it is convenient to compare the amount of
Al,O, to be transferred with the principal one of these com-
ponents taking part in the aluminous ferromagnesian mineral.
This component becomes a unit of comparison for the other con-
stituents of the particular mineral. In the pyroxenes and
amphiboles CaO is the component next to Al,O, most involved
in the change. It becomes the unit of comparison in these min-
erals. In the micas K,O plays this rdéle.

The molecular relations which must be taken into account
are expressed by ratios in the Tables XII, XIII, XIV at the end
of this Part, and may be summed up as follows:

Aluminous pyroxenes (Table XII).—1. The ratio of Al,O,
to CaO ranges from almost nothing to 0.23. From these data
the maximum limit is

2. The nearly equal proportions between Fe,O, and Na,O
indicate that the soda is present in the acmite molecule.
3. The generally small ratio between Na,O and CaO. The

NasOre f
—.*— is less than 0.1 in most cases. Moreover, the pres-
CaO

ence of a notable amount of acmite molecule is indicated by the

ratio

optical properties of the pyroxene.

4. The nearly constant ratio between MgO+FeO and CaO,
which is approximately

Wilfed Oe TKO) :
CaO

5. The SiO, + TiO, is approximately equal to the number of
molecules of MgO + FeO + CaO + 4Na,O, corresponding to
pyroxene molecules (MgFe)O.. CaO. 2510, and Na, O. Fe, On.
AsiO., and (Me, re \On (Al, Fe),.O. > SiOx.

Aluminous amphiboles (Table XII1).—The relations are less
definite. There is a wide range in chemical composition, and it

CLASSIFICATION OF IGNEOUS ROCKS 671

will be necessary to consider special kinds of amphibole in dif-
ferent cases.

a). 1. For hornblende occurring in rocks with lime-soda-feld-
spar the ratio of Na,O to CaO ranges from 0.08 to 0.21, while
for hornblende in one alkali feldspathic rock it is 0.28, and
for somewhat similar amphiboles, barkevikite and hastingsite,
it reaches 0.35 and 0.43.

2. The ratio of Al,O, to CaO ranges from 0.32 to 0.67 in the
first mentioned cases, common hornblende, and from 0.44 to 0.85
in the second.

3. The ratio of MgO +FeO to CaO ranges trom 1.73 to 2.72,
being nearly 2.2 in most cases. It is not 3, as given in the text-
books generally.

4. It is to be noted that in hornblende in the more calcic rocks
the ratio of MgO to FeO is 2 or more (from 2 to 4), while in
those in the perfelic, alkalic rocks, this ratio is less than 1 (from
0.1 to 0.6). In other words, magnesia dominates ferrous iron in
amphiboles of the first kind, while ferrous iron largely prepon-
derates in those of the second kind,

5. The SiO, + TiO, is equal to the MgO +FeO + CaO plus
an amount sometimes equal to 4Na,O, but not always.

6. The ratio of Fe,O, to CaO is quite variable.

6) For soda-amphiboles of various kinds, excepting that occur-
ring in comendite.

1. The ratio of Na,O to CaO ranges from 1.23 to 4.6. That
is, the lime is considerably less than the soda.

22 he matio of All ©, to; Ca© ranges from 0.23) topm3:

3. The ratio of MgO + FeO to CaO ranges from 4.6 to 21.6,
and the FeO is greatly in excess of MgO.

4. The variation in the SiO, indicates that Na,O is partly
present in the riebeckite molecule, and partly replaces CaO and
(Mg, Fe)O in the RO .SiO, molecule.

Micas ( Table XIV ).—For these, as already said, the chemical
component associated with Al,O, which may be transferred from
the feldspathic minerals is K,O, and for purposes of calculation

6:72 CROSS, IDDINGS, PIRSSON, WASHINGTON

it is convenient to compare other components with K,O. The
micas included in Table XIV are of three kinds — biotite proper,
lepidomelane, and phlogopite.

a) For biotite:

Al,O, ranges from 1.3 to 1.9, and the ratio of Fe,O, to K,O
is 0.28 to 0.33, that is, nearly constant.

The ratio of MgO + FeO to K,O is 4.5 to 5.9 and magnesia
is in excess of ferrous iron. The ratio of MgO to FeO is 1.1 to
to"l:6: Average, i.5:

The SiO, nearly conforms to the theoretical molecule
(Koad) 07 (AL Be) ©. 25107 47 2(Migj He) @.s10..

b) For lepidomelane:

The ratio of Al,O, to K,O is nearly the same as in biotite,
but that of Fe,O, to K,O is higher in two cases.

(Mg, Fe)O is lower, and ferrous iron dominates over magnesia.

c) For phlogopite:

The ratio of Al,O, to K,O is nearly the same as for biotite
and lepidomelane, but that of Fe,O, to K,O is lower.

The ratio of MgO + FeO to K,O is nearly the same as in
biotite, but magnesia greatly predominates over ferrous iron.

The ratio of SiO, to K,O, although nearly the same as in
biotite, does not conform to the formula given for biotite.

From these relations we may deduce the following method of
transferring aluminous molecules from salic and non-aluminous
molecules from femic to form molecules of augite, amphibole,
and mica.

For the calculation of aluminous pyroxene: The kind of pyrox-
ene to be calculated should depend upon the kind of rock in
which it occurs.

a) If the femic minerals already calculated for the rock
include acmite molecules, sufficient Na,O and equal Fe,O, are
to be combined with diopside molecules to satisfy the ratio

Na Ou

CaO
(D standing for datum, the value given in the pyroxene or
other mineral to be calculated); CaO in the ratio being equal to

CLASSIFICATION OF IGNEOUS ROCKS 673

CaQOj,, in diopside, plus CaO,,, derived from anorthite, in conse-
quence of the transfer of Al,O, insufficient amount to satisfy

the ratio
AO _p
CaO, CaO

Moreover, the ratio
(Mg he) Oia. D
CrOne CaO 7 7

approximately I, must be maintained by transferring (Mg,Fe)O
from femic hypersthene or olivine. If the amount of these min-
eral is comparatively small a limit is set to the amount of alu-
minous pyroxene that can be produced.

6) If there are no acmite molecules estimated among the
femic minerals, sufficient Na,O must be transferred from soda-
feldspathic molecules to satisfy the ratio

Na,O
Ci@7 == CaOx

=D,

where + Ca©, is the lime transferred from, or to, anorthite, If
iransterred ‘trom: anorthite to diopside the sign is: =>), 1f to
anorthite from diopside the sign is —.

This follows from the fact that the transfer of Na,O from
soda-feldspathic molecules liberates equal molecules of Al,O
and if all of these are not needed to satisfy the ratio

ALO;
Ca@), == CaO...

3°?

—D,

as is the case in a number of pyroxenes given in Table XII, the
Al,O, not required withdraws CaO from diopside molecules to
maintain the stoichiometric proportions obtaining in the salic
minerals by making additional anorthite.

Fe,O, is to be transferred from extra Fe,O, (hematite) or
from FeO. Fe,O, (magnetite) in sufficient amount to satisfy the
ratio.

Ke, OF

CHORE CHOON ae ee

674 CROSS, IDDINGS, PIRSSON, WASHINGTON

The liberated FeO is to be added to sufficient (Mg, FeO) from
hypcrsthene or olivine to maintain the ratio
(Mg,Fe)O on
CaO, tex (CROs ire :

When the sign of CaO,, is—the transfer of (Mg,Fe)O
and the liberated FeO is to hypersthene or olivine from diop-
side.

A readjustment of SiO, is necessary according as (Mg,Fe)O
is transferred from or to hypersthene or olivine, and as Na,O is
transferred from albite or nephelite. If there is extra SiO, it is
placed first with the bases having strongest affinity for it unless
these are already satisfied, when it appears as quartz.

In case the transfer is to be made to the complete exhaustion
of one of the limiting components, the algebraic formule which
can be devised for a particular case can be solved, but it fre-
quently happens that augite occurs in rocks together with
hypersthene or olivine, lime-soda-feldspars, and magnetite. The
algebraic problem is indeterminable, since there are more
unknown quantities than equations to satisfy them, unless the
quantity of one or more of the minerals mentioned be given.
More or less aluminous pyroxene within the limits of the magma
appears to develop in different instances. The factors controlling
the amount of augite crystallized in such cases are not at present
known. The same statement may be made with reference to
other aluminous ferromagnesian minerals.

For the calculation of aluminous amphiboles: a) Hornblendes
contain more Al,O, than Na,O, and in the great majority of
cases occur in rocks not having acmite molecules among the
femic minerals.

To transform diopside molecules into those of hornblende
sufficient Na,O must be transferred from soda-feldspars, and
enough CaO from anorthite with their equivalent Al,O, to
satisfy the ratios

NO ieee, ON a
Ca@q ff eaOs CaOz +) CaO.

CLASSIFICATION OF IGNEOUS ROCKS 675

Furthermore
Re, OF

COME C204 2

must be satisfied by transferring Fe,O, from Fe,O, (hematite),
or from FeO.Fe,O, (magnetite).
The liberated FeO with (Mg,Fe)O from hypersthene or
olivine must be transferred to satisfy the ratio
(Mg,Fe)O

ClOMNGIOR, ee

SiO, must be adjusted in the manner already indicated for
the case of pyroxene.

6) The calculation of the more alkalic amphiboles which
occur in rocks rich in soda and iron and comparatively poor in
lime and magnesia is to be made in the following manner :

Sufficient acmite molecules are transferred to satisfy the
ratio

Fe, Or, ee
CaO; ==] C20

Equal Na,O from acmite plus sufficient Na,O from soda

feldspars is transferred to satisfy the ratio
Na,O =
CaOzia= CaO.

Al,O, equal to Na,O transferred from soda-feldspars is
liberated and enough of it is transferred to amphibole to satisfy

the ratio
ASO;

(CHON Se CHO aE ee

If all the liberated Al,O, is not required to satisfy the last
named ratio, CaO must be withdrawn from diopside to form
additional anorthite, and the sign of CaQO,, is minus.

If more Al,O, than that liberated from soda feldspars is
needed to satisfy the ratio in question additional Al,O, with
CaO is transferred from anorthite, and the sign of CaO in the
expression is plus.

676 CROSS, IDDINGS, PIRSSON, WASHINGTON

The ratio
(Mg,Fe)O on
CaO, + CaO,,

must be established as well as that of

MeOie

mC

since the latter is highly characteristic.
The adjustment of SiO, follows the ratio
SiO

2 hee
CaO; + CaO,, D

and the requirements of the other molecules.

For the calculation of ferromagnesian mica.—The process 1s
essentially the same for various kinds, the values of the ratios
varying for biotite, lepidomelane, and phlogopite.

In most cases Al,O, is in excess of K,O, and in transferring
K,O from feldspathic molecules the accompanying equal Al,O,
is not sufficient to satisfy the ratio

ADO fuk

K,O nee

Extra Al,O, may be derived from corundum when estimated
present, and this will be the case in numerous rocks in which
muscovite and biotite have developed.

If=there isino, extra Alo O. of this; kind .Al)@7<must be
transferred from anorthite until the ratio in question is satisfied.
The liberated CaO must be transferred to diopside molecules.

Fe,O, is tobe transterred from Me, (hematite) or Fe@:
Fe,O, (magnetite), and the liberated FeO is to be added to
(Mg,Fe)O, to satisfy the ratio

Fe,O,

co

(MgFe)O must be transferred from hypersthene or olivine
to satisfy the ratio

(Mg,Fe)O |
KOm ey

CLASSIFICATION OF IGNEOUS ROCKS 677

and the ratio must be maintained according to the kind

Mg
FeO
of mica.

Si@* and! TiO) are “to be transferred according to ‘their

respective ratios
SiO, TiO
=D 2
kO ang K,O
After which SiO, must be adjusted among the molecules
affected. In general some SiO, will be liberated and the amount

of quartz in quartz-bearing rocks will be increased. In other

==) ]0) 2

cases lower silicates will be raised.

The calculation of muscovite is simple, for the reason that it
occurs mainly in rocks in which there is an excess of Al,O,,
estimated as corundum.

Since in common muscovite H=2K the following ratios
obtain:

ERO men Oss S10, 7
LOMA LaIKNOR UO CN KO"

Such a muscovite equals orthoclase plus corundum plus water.
2H,O+K,0 + 3A1,0, + 6SiO, =K,0+Al,0, + 6SiO, +
Pe Ibs Olea oe) Les Oe
Hence to the molecules of hypothetical corundum add half as
many of the orthoclase and an equal number of H,O. The distri-
bution of SiO, remains as before, the amount of quartz has not

been affected.

From the foregoing it is seen that the development of augite
in rocks reduces the amount of anorthite molecules calculated
among the salic minerals and also reduces the amount of hyper-
sthene or olivine that appear as femic. It affects the distribution
of silica by liberating a part of that allotted to anorthite when
the (Mg,Fe)O is derived from hypersthene, or when Na,O is
transterred from albite molecules. Its development would
reduce the amount of olivine rather than hypersthene when both
are otherwise normatively present.

The crystallization of hornblende reduces the amount of
anorthite and hypersthene or olivine reckoned as standard min-

678 CROSS, IDDINGS, PIRSSON, WASHINGTON

erals at a greater rate than that of augite would, in proportion
to the amount of femic diopside molecules converted into horn-
blende, because the ratio of Al,O, to CaO is higher, and the
ratio of (Mg,Fe)O to CaO is twice as great as in augite. The
effect on the SiO, would be similar.

The reduction of anorthite molecules by both these pro-
cesses would affect the character of the plagioclase actually
developed in the rock as compared with that reckoned as
standard.

The development of mica would reduce the molecules of
potash feldspar or of leucite, and the formation of biotite
would also reduce the amount of hypersthene or olivine, and
that of magnetite. It would reduce the amount of calculated
corundum, or, in the absence of this extra Al,O,, would reduce
the molecules of anorthite and increase those of diopside or
wollastonite. The distribution of SiO, would be affected, and
in cases where there is no extra Al,O, and where (Mg,Fe)O is
derived from hypersthene SiO, would be liberated to raise
lower to higher silicates, or to form quartz.

Garnet.— In the matter of garnet it is easily seen that to form
simple lime-alumina garnet (grossularite), 3CaO.A1,O, .3SiO,,
it is necessary to combine anorthite molecules with femic wol-
lastonite, CaO.SiO,, with the liberation of silica. And it will
be found that this garnet is frequently developed in rocks whose
estimated standard minerals include wollastonite, and its oujtel:
lization reduces the amount of anorthite.

AN EXAMPLE OF CALCULATION.

As an illustration of the several processes above described
we present the case of the Butte granite, because of its fairly
uniform composition as shown by five analyses,’ its distinct
crystallization, which is of such size as to permit of accurate
microscopical measurement, and because the two aluminous fer-
romagnesian minerals, biotite and hornblende, have been
separated and analyzed. The rock from which these minerals

* WEED, W. H., Jour. GEOL., p. 739, 1899.

CLASSIFICATION OF IGNEOUS ROCKS 679

were separated has been analyzed, and this analysis forms the
basis for the calculation of the standard mineral composition.
It is from Walkerville Station, near Butte, Mont. The thin
section measured is from a specimen of very fresh rock taken
near the surface twenty feet from the outcrop of the Parrot vein
at. Butte:

TABLE VII.
i}
Mol, | Pyr.| Ilm.| Ap. | Mag.|Orth.| Alb. | An. | Diop.| Hyp. |Qu’tz.
SiO aside 6328 Se [PT.0605) 4)) cee Ae ae Pee 2 708 27.0 1a @ Onl G4) 322
IMO S50 TSSOAe [eal 515 an 8 ae ee 4See ed Sl OS al ere = a
IWGAOR sat 2S TiDen | GeO ilies sige he cacheal baer x a: Bic
1BEO)s case 2.59 | .036 2 8 ss 1 We oe an 20162
IMHO Sse 6 BU" 2053 ea a ; iis af Ae Ls ats
CAORP 2407 Oy ae a 3 A a eles O5 3
Na OF ee Drsene |e COVE |i ea ie at a Fort cis lea
K,0 sists 4.23 | .045 sie oe 8 ee 45 eyes
H,O-+... ANN||\ O58 a ae es ae ae |
H,O-... CPAP IM = ye re ae ae aA ne ne
Ona 65 | .008 Bile
POE seas o2ile WOOK I a,
MbeKONS ace .07 | .OOI I
SHOP ae. -02 | .000 |
BAO) GS 5.0 .09 | .000
Gi Over Ene | Mreresrene we
SOK conn SRY |) valoXoya 4
OU Ne cascseseie tre || Resaeus
99.82 |
FORMULA. MOL, WT. NORM,
SiO, - - - - 323 X 60 quartz 19.38 Q 19.38
K,O.AI,0,.6Si0O, - - 45 556 orthoclase 25.02
Na,O.AI],0, . 6SiO, - = 45, 605245 calbite 23.58+ F 66.67
CaO .Al,0,.2Si0, - - 65 X 278 anorthite 18.07
CaOQsiOs 3 X 116
} Mego. SiO, 2 X 100+ diopside 67
bev ao 2 - Ti Garg 2 12) 7-45
MgO.SiO, Bl OM LOO! fq.
FeO . SiO. SS | hypersthene 6.78
FeO. FeO; 13 X 232 magnetite a M :
FeO. TiO, 8 X 152 ilmenite T22)\ ass
Fes, = - 2 X 120. pyrite 2A AN
3CaOne.O- - Iu 310 ‘apatite 3 “55
etc, :99
99.27

a) The calculation of the standard minerals, or norm, is expressed
in Table VII. MnO is combined with FeO making 0.037
molecules. SO, is allotted to FeO to form FeS, becauseof the
pyrite known to be present in the rock, TiO, to ilmenite, P,O, to
apatite, Fe,O, to magnetite, K,O to orthoclase, Na,O to albite,
the remaining ERO): (0. 065 mol.) to anorthite, the remaining

680 CROSS, IDDINGS, PIRSSON, WASHINGTON

CaO to diopside, the remaining (Mg,Fe)O to hypersthene, and
the remaining SiO, to quartz. From the respective molecules
the percentage weights are found by multiplying each by the
molecular weight.

The classification of this rock from the data given in Table
VII has been described on a previous page as an example of
an intermediate rock.

6) To calculate the mode from the chemical analysis of the
rock we have to introduce biotite and hornblende into the calcu-
lation, and we have the chemical analysis of each given in Tables
XIII and XIV. We will assume that they contain all the mag-
nesia (MgO) and all the iron oxide (FeO and Fe,O,), except
that which goes into pyrite, magnetite and ilmenite. For
these are the only ferromagnesian minerals observed in thin sec-
tions of the rock except a small amount of pyroxene intergrown
with the hornblende. This was probably included in the horn-
blende material analyzed so that the analysis represents the
mixture and may for the present calculation be treated as the
composition of one mineral.

In Tables XIII and XIV are found the following ratios :

For Butte hornblende -

MgO FeOo sh). (Al,0, 0 EO ae
CiO) 2 CO CAO = ee CAO
Na,O_ SiO

———— =, seat — 0 °
CaO UO GRO) te © 79
For Butte biotite:

MgO _ eOee INO Kes Onn
Ome reuse KO: K,0 ——— a es
ELS On SOK MOF
KO KOC | KO te

If z=molecules of CaO in hornblende,.y— molecules of 1G ©
in biotite, g=molecules of Fe,O, in resulting magnetite, w=
molecules of TiO, in resulting ilmenite,

Then 2.24%-+5.08y+z2-+w=88 =(Total (Mg,Fe)O less FeO in pyrite).
I5x+ .32y-+3 =—i13(—(otal Hen On):
A4y7-w = 8—(lotal[mO;,):
32x +1.3 y =12x-+y+(x—3)=(transferred Al,O,).

CLASSIFICATION OF [IGNEOUS ROCKS

681

That is, the Al,O, in hornblende and biotite equals the trans-
ferred soda, potash and transferred lime, which latter equals the
CaO in hornblende less that previously allotted to dioside (#—3).
From these equations are derived the following values:

w = 2.5 (3).
From these values of 4, y, 2 and w the readjustments for all
the minerals are made as indicated in Table VIII.

te tOL4n((O) ane —— tee D2))",

TABLE VIII.

2—7.51((8)\,

Mol. | Pyr. | Ilm. | Mag.| Ap. | Orth, Alb. An. Biot. | Horn, | Quartz.
1 Olu 1.065 ; 198 264 120) 72 30 381
Al,O; . 155 : 33 44 60 16 2 ae
HesOs.. .013 8 sik opr sits 4 I
ImeO Gaon .036| 2 3 8 19 4
MgO. -053 : Hor 41 12
Ca@ eens o71 3 peat 60 st 8
Na,O.. 045 ; shape 44 ie I
IK nOse ae 045 B33 ae 12
FRO seh Saas: Sat 8
SO eal: re
JERKS 008 3 : 5
P.O, . ool I aie ao
MnO . OOoI I
SOR onc 004| 4
IO h ee oie noe ae
+ = 8= mol, CaO in hornblende.
y =12=mol. K,O in biotite.
z = 8=mol, Fe,O, inmagnetite.
w= 3= mol. TiO, in ilmenite.
FORMULA. MOL. WT. MODE.
SiO, - - - - - 381 X 60 = quartz 22,86
K,0.Al,0,.6SiO, - - 33 X 556 = orthoclase 18.35
Na,O.A1,0,.6SiO, - eld <5 24) —)albite 23.06
CaO.Al,0,.2Si0O, - 60 X 278 = anorthite 16.68
8H.,O Site = aie eto ae GLa T Ole)
12K.,0 - - - 12 X 94 |
| AAO s - - - - - 16 X 102 |
4Fe, - - - - 4 X 160 | _ .
4 FERS CIO) f : i t SE S28 é biotite 10.92
|19FeO - - - - - 19 X 72
eae - SMe iow Waal g ethene ib Ge GO. |
72SiOg - - - - - 72 X 60
1Na.O - - - - =| TX 62 ]
CHO) “ay oS sid ana 8X 56
12MgO - - - - - I2X 40
BINSO) toby Pau oe ose Te 5 xX 72 p= hornblende = 3.56
| 2A1,03 Metcow le Voie es 1 2x 102
[ESO - - - - - I X 160
30510, - - - - - 30 X 60 ;
FeO.Fe,0, - - - 9 8 Xi2325. = magnetite, = ‘2:86
HeOriO ts = te 9) = 3) 152" —almenite = .46
FeS, mae Nets. vise, eae sia 2X120 = pyrite = .24
3Ca@sP.OF - - - - IX31I0 = apatite = .31
etc, 85

682 CROSS, IDDINGS, PIRSSON, WASHINGTON

c) The process of calculating the norm from the mode, that is,
the standard mineral composition from a microscopical investi-
gation of rock sections, may be illustrated as follows :

Two thin sections of Butte granite were examined; one from
a surface excavation, twenty feet from the Parrott vein, very
fresh, with almost no signs of alteration, the other from similar
rock used in building in the city of Butte, the location of the
quarry not noted.

In the first section an aggregate distance of 12,740 units of
the micrometer scale was measured, embracing 604 measure-
ments. The average diameter of the plagioclase crystals was 32
units, and the total length of plagioclase diameters was 5,492.
Table IX shows the lengths, relative volumes, assumed specific
gravities of each mineral, and the resulting percentages by weight.

SIEACB EES a IeXG;
First Rock. SEconpD Rock.
Total Relative Relative
Diameters. | Volumes, Sp. gr. Weights. Volumes, Weights.
Oidlantzeyeeceen ee 2,954 22517 2.65 2ONSS =a ZOO 20.34
Orthoclasempae-seys 2,373 18.62 2a5 7 17.57 2107, 19.82
Rlagioclaserics...c. 5492 43.10 2.68 42.47 42.29 41.48
Brotiteiect yedersens 1,130 8.87 3.00 Oni 8.15 8.94
Hornblende ....... 482 Bayh) 3.20 4.44 4.84 5.67
LENA AANAS gouecb,8 0c 252 TOV, 21.30 2) 27, 2.07 2.50
Miagnetite a: mject cr 51 . 40 5. 17 Sue) -54 1.02
LEVAgMOemd ah badoot 6 .04 5.00 .07 02 .04
12,740 99-95 99.98 99-95 99.81

In the second rock section examined an aggregate distance
of 7,122 units of the scale was measured, embracing 403 measure-
ments. The average diameter of the plagioclase was 24 units,
and the total length of plagioclase diameters was 3,012. The
relative volumes and weights are given in the last two columns
ofthe Table:

A comparison of the two cases shows how closely the rocks
resemble one another, the second being slightly higher in ortho-
clase and lower in quartz than the first. Other sections from

CLASSIFICATION OF IGNEOUS ROCKS 683

Butte granite from other localities in the vicinity of Butte show
slightly more ferromagnesian minerals. Weed’ has called atten-
tion to the generally uniform composition of this great body of
granite as indicated by five chemical analyses, one of whichis given
in the tables of micaand hornblende. The specimen from near the
Parrot vein will be made the basis of the following discussion.

Optical investigation of the plagioclase by the Michel-Lévy
method with Carlsbad twins of albite twins showed that the cen-
tral portion of each crystal has approximately the composition
of Ab,An,, but the marginal zones are more sodic, having in
some crystals the composition of oligoclase as indicated by the
optical orientation and the relative index of refraction as com-
pared with quartz by the Becke method. Without making the
necessary observations for estimating the relative proportions of
the several zones, and the average composition of the plagio-
clase as a whole, it was seen that the greater part of each crys-
tal is about Ab, An,, and only a narrow margin is oligoclase,
but the optical orientation of the outer portions shifts gradually
in most cases from the central part tothe margin. The average
composition of the plagioclase is probably andesine somewhat
more sodicthan Ab, An,. If Ab,An,, it would have the compo-
sition of the mixture of albite and anorthite calculated from the
chemical analysis of the Butte granite from Walkerville Station
already cited, after developing biotite and hornblende. In this
case the 42.47 per cent. of plagioclase would yield by molecular
calculation 24.71 per cent. albite and 17.76 per cent. anorthite.
In the second rock there would be 24.29 per cent. albite and
17.18 per cent. anorthite. The orthoclase may be considered as
wholly potassic and reckoned as pure orthoclase.

In order to obtain the salic and femic components of the biotite,
hornblende, and pyroxene, since the other minerals are standard,
we may take the analysis of Butte biotite from Table XIV and
multiply it by 0.0977, the percentage weight of biotite, and take
the analysis of Butte hornblende from Table XIII and multiply
it by 0.0681, the combined percentage weight of hornblende and

* WEED, W. H., Jour. GEOL., Vol. VII, pp. 737-50.

684

pyroxene, for the reason already given.

CROSS, IDDINGS, PIRSSON, WASHING TON,

The resulting constitu-

ents reduced to molecular proportions may be distributed among

salic and femic minerals as shown in the accompanying Table Xe

From these data we deduce the standard mineral composition

of the rock from near the Parrot vein as follows:

TABLE X.
9.97 fo OE Sum. |Fluor.| Ilm. | Mag.| Orth.| Alb. | An. | Diop.| Hyp. Ou
SIO sgeertark ha Stemi Seale Lt 66 6 12. |. 12 "|, 64" |:—49
AUS Olaye craters 1.37 -46 o1s oe it I 6 ToL ag a
CnOnasone .52 34 005 5 ae
Ke Queens Lyf ait 029 6 5 2 16
Mig@ii irra: eZ .82 OH tac te 4 | 47
CaO eho: .00 .76 014 | 2 os 6 Onl aie
INGO Go ans -OI .05 ocl me I
ISO) 6556.66 gl -08 OIL 11
ies Of out) 03
H,0-- .36 .16 ge ee
AOR eGeton 235 .10 006 6
Pe Oley ieteters .O1 .02 | .000 Se
Min @zaraews 02 .04 ooI I
Ba @ ries los .O1 ie
Gl iestome .02 mt Ae at
Higa ee rcnepastis .07 .02 004 4
FORMULA, MOL, WT.
iO, - - : 49 X 60 = quartz = 2.94
K,0.A1,0,.6Si0O. - 1m X 556 = orthoclase = 6.12
a,O.Al,0,.6Si0, ToX<e24) —-albite:— 52
CaO.Al,0,.2Si0. - 6X 278 = anorthite = 1.67
CaO.SiO, - - 6 X 116
MgO.SiO, - 40x 100 = diopside = 1.36
FeO.SiO. 2X 132
FeO. 810," _ 17 5¢ aga § = hypersthene = 6.04
FeO. Fe.O; - 5 X 232 = magnetite = 1.16
FeO.TiO, = 6X52 = ilmenite= QL
CaF, - - 2X 04 = fluorite— .19
Percentages Determined from Rock. Readjustments. Standard Mineral Composition.
Quanxtzn eee DOIBIGE: Heavecarens ras Gece yen “aol |OwebdiAegoowo 5 auc 19.61
@rthoclasewerseniecn: Tee Sty leentas soak ccotoanravere eee +6.12 | Orthoclase....... 23.69
Plagioclase...... AQ a7 Allbiten2aiw7 Urrrea keto 52:5|) lll less rueeuaretelerais 25.23
Anorthite 17.76. +-1.67 | Anorthite........ 19.43
Biotite 3) scutes OR DiC psideryyipe ke 13 Oia DIOP Silesia cierer 36
Elornmblen de srcsver ec: 4.44 | Hypersthene.... 6.94 | Hypersthene..... .94
By KOXEM Clays iets: seearert 2a 7io lt MUOnIte ttc is PTOs|) Hlwoniternertry tee .19
iimmenitenramere rete OTs plllime nites essences -QI
Malonetite i-th Oo mieoant doe BooR +1.16 | Magnetite....... .92
Ieee ah oaro.o.4 odniC .07 IAAOS coG5 6.000000 07
99.35

CLASSIFICATION OF TGNEOUS ROCKS 685

The approximate chemical composition of this rock is found
by separating each of the minerals observed in the rock into its
chemical components. The results are given below:

AVAUB IGE exo,
B “4 &
3g a c o ary 8 =
Bispace a 8 m4 2
go raion) 5) R aS oO Oo N % us) c
Se ceneaieeiose ie | 2 | & | 8 E
Boge eel reson apaacl|| vt < fo) ° fe n <
SiO, BeSo) |GeaL1 7.66] 16.97] 11.37 | 22.55] 65.24] 64.43] 64.34
EMS ORS Hit eI7 I YaclO |) Goro allisa cree aya | watered DADE es ell vilaye eXsha bea He)| ants 4/X)
Gn Or doll eG SHR |s Sel (AG Mall) aaa alae | eteacncr hl (ornare Kaci Ball abo ekenlh atay7/ 1.62
Ee Ores les 37a eS lle Obl llmren dsm |lievereuciee Millevoveroela|| vee cere: « 235i 242 2.94
INTEHO) ose calf AAEM cagtstedll) ocephen bereete| Irae tee Neen eae fetes «| ele O35) 2a TO D0)
CAO, 654i) GOO! 270 BYES /allleerseeheall Loonies a[n.sees 8 4.33| 4.38 4.24
Na,O OI] .05 DEO ator, |ntaceye 2.99} 2.95 2.76
IK) 9I|] .08 Be PLOW Neocon ZOO!) rAiie 4.04
Vea) oh Rat Nh MACON eee eerie hata | acicre eal Fee ec [Pepe 15 nalts 7215
OCA lees Oalp ranOllMe es allen ep e amide wasacs Ilhroyci cues aalets 2 43 76
ALO) OCH Mee On lara eeieltnee rete al etepece alee ses eus tes Iceni ell leva’ 45 35 53
PO; OE" e2OZE| | erate ||lneronaes Loco onial saxsesetae| Mic mena lara 03 04 14
MnO. AO 2s OAM | Warrrcoetel| Wyse toregen Matyas tre | ust saeco Near exe, .06 06 12
STOR Sees | sect sPe | as eeset| Cree ome ese e teed lh cams cttres lIevatarer ho |e decdi See 03
BaO. OMe sees | pretest [eet aioe [Pete cralyetoatoeesl|D Soccajer e: Wiahere aco .O1 Ol 06
(Cilersless (P22. aS enlaces frat enroil eres ISOs sett ce eae (ea .02 02 (Obs
Hee Seca O71) MO 24 |eee Nea loncome Ret ee lly ereetane lin steis estes dina aac .09 .09 .005 2
Sane: eee saan etetaealll es Ola tannsallne choad st[iisntene ell tuw.ais j .04 .02 .033
Mer eWeh omer ey [perce ster louie seem |feratraweray imieawewee|lterecroms ll rencraiatred [ese caliatrsin\lh (ols cel spiel [jp arse. o.01) | uae euehe 024
99-93! 99.751 99.805
1CO, 2Cu 3FeS, 4ZrOn

The chemical composition of the second rock from Butte in
which the minerals were measured is given in the same table,
and also the analysis of Butte granite from the Atlantic Mine.
The similarity of the results is an indication of the success of the
process in this instance.

With the data here assembled it is possible to classify the
rock from near the Parrot vein with almost as much accuracy as
though it had been analyzed chemically. In other cases the
data may not be so nearly correct, but the error in many cases
will be quite within the limits of variation for the division of the
classification to which the rock belongs.

It is evident that there is need for more chemical work
upon the component minerals of igneous rocks and the correla-

686 CROSS, IDDINGS, PIRSSON, WASHINGTON

tion of their optical properties and chemical composition with
that of the rocks containing them. The necessity is also appa-
rent for complete analyses not only as to the main constituents,
but also as regards some of the constituents hitherto considered
as rare or of little importance, but which modern investigation
has shown to be widely distributed: and often of considerable
influence on the norm and mode. The constituents present usu-
ally in small amounts are distributed very unequally among the
various kinds of rock magmas, and it would save the careful
analyst much needless labor if the petrographer would indicate
to him those which it is advisable to look for and determine.

EPILOGUE:

Some who read this essay will, without doubt, object to the
method of classification herein proposed because it is new and
embodies new principles, since experience has shown that one’s
mental attitude toward new methods is on the whole conserva-
tive, and tends to resist their introduction. It is, also, easier to
travel along a familiar and oft-trodden path, no matter how
crooked and obstructed it may be, than to hew out a new one
and provide a broader way for the future.

Objection will doubtless be made that the system, on account
of the definite quantitative character of its divisions, throws
a greater amount of mental responsibility in classifying upon
the one who uses it. Thus, for instance, at the very outset, it will
often happen that megascopically a rock will be so close to one
of the lines separating two different classes that it will be diff-
cult to know to which one it should be assigned. And it will
sometimes happen that later chemical or microscopical investi-
gation will show the preliminary classification to have been
wrongly made. Moreover, when such examinations have been
made, it will often occur that it is largely a matter of judgment
as to which of two adjacent Orders, Rangs, or Grads a rock
properly belongs.

To some this mental responsibility is distasteful and unsatis-
factory; they desire a classification made up of a number of
simple divisions like neat pigeon-holes or compartments into

CLASSIFICATION OF IGNEOUS ROCKS 687

which each object can be easily and promptly thrust and dock-
eted, and any method which fails to achieve this and relieve
them of considerable mental effort in classifying rocks will not,
in their estimation, be a proper one.

Unfortunately the difficulty in this respect lies not in the
method, but is inherent in the subject itself. Rocks grade into
one another in all directions, chemically, minerally, and textur-
ally, and these again overlap in the different modes of geological
occurrence.

Therefore, unless the future should reveal new properties of
rocks, as yet unknown and unsuspected, which may be used as
bases of classification, every method so far devised, or which
can be devised, must have artificial lines of division, and cause
the petrographer trouble in certain cases in deciding where his
rocks belong. The erection of the monzonite group by Brégger
is a good example of this. Where rocks contained about equal
amounts of orthoclase and plagioclase it was formerly difficult
to say whether they should be classed as syenites or diorites.
The formation of the monzonite group avoided this difficulty.
but in its turn gave rise to two others whose sum total is greater
than the original one; for it is clear that it will be equally diffi-
cult now to decide whether a syenite with considerable plagio-
clase is stilla syenite or a monzonite, and the same difficulty is met
in diorites with considerable orthoclase. Thus,the formation of
new groups simply multiplies the difficulty; it does not remove it.

The system which we propose does not, therefore, meet this
trouble; it does not pretend to, for any system based on our
present knowledge of rocks which should claim this would be in
the nature of a mere nostrum, and would simply profess to do
what it could not perform. In the ultimate analysis, every
system will and must throw on the petrographer the mental
responsibility of deciding, in a considerable number of cases, where
rocks lie close upon divisional lines, into which of two divisions
they must be put. He may not like this, but it cannot be avoided.

In different systems this difficulty is met in greater or lesser
degree in different parts of the systems. In the one we propose
it occurs chiefly in doubtful cases, at the beginning, and in the

688 CROSS, IDDINGS, PIRSSON, WASHINGTON

sum total we believe that it is less, or at all events no greater,
than in any system heretofore proposed. But in regard to this
each must judge for himself.

Objection may’ also be raised to this method of classification,
that it entails a greater amount of labor upon the petrographer
than those now in use. Perhaps there are some who, recalling
the possibility of glancing through a microscope and _ noticing
the presence of striated feldpars, with one or more dark-colored
minerals, and promptly identifying the rock, regret the intro-
duction of the chemical character of the feldspars and the
quantitative determination of all the minerals. But to such we
have no apology to make. Already precise optical methods of
determining the feldspars are in use; and the confusion result-
ing from an absence of the quantitative element in rock defini-
tions is becoming intolerable. The time has come when the
petrographer should demand the exactness and sharpness of
definition which, though obtained only by patient and careful
work, add clearness to the conceptions and proportionate weight
to the results of petrographic investigation.

It may be thought that the method of obtaining the approxi-
mate chemical composition of the magma by optical study and
computation of the minerals in a rock is tedious, as it might pos-
sibly take several hours of work. But let one who considers this
seriously make a good, careful, and accurate chemical analysis and
note the amount of time consumed. We urge him not to throw
the task on someone else, especially not on some beginner in ana- |
lytical chemistry, as has unfortunately so often happened, but to_
make the analysis himself. Indeed, every petrographer should
have made enough analyses to properly appreciate what they
mean, after which we feel sure he will not object to the length
of time involved in a microscopical analysis when one is possible.

Again, we have assumed asa condition for proper and con-
venient classification that it should be* dichotomous, since not
more than two factors can be handled advantageously at one
time. To those who may not agree with this proposition we can
only recommend the trial of forming a classification, using three

CLASSIFICA TION. OF IGNEOUS ROCKS 689

or more factors at once. We have been through this phase of
belief most thoroughly, and have convinced ourselves that two
factors are all that can be handled at once without complications
which would destroy the practical usefulness of any system.
We believe that anyone who tries it will come to the same con-
clusion.

Furthermore, repeated trials have led us to think that a five-
fold division is the most practicable one and the one best suited
for the needs of petrography. We have applied this logically
throughout, except in certain cases where a threefold seemed
sufficient.

We desire also to ask the reader to bear in mind that the
scheme of classification and nomenclature herein proposed is of
much greater importance in some features than in others. The
important points to bear in mind, and on which our system rests,
are these:

a) The bringing together of rocks of s¢milar chemical compost-
tion into the same divisions.

6) The resolution of a rock, either from its chemical or
microscopical analysis, into quantitative amounts of certain
standard minerals.

c) The establishment of a strictly quantitative basis of com-
parison of rock constituents, so that all constituents are valued
according to their proportions.

da) Yhe division of the minerals into two main groups, salic
and femic, thus obtaining two factors, and the application of the
fivefold division.

e) The recognition of the relation of the mode to the norm.

Time, use, and experience, if this system gains the currency
we hope for it, may cause modifications in the smaller divisions ;
and the suggestions we have made in regard to nomenclature,
both of rocks and of descriptive terms for their textures, may
only in part be followed. But these are minor features and
their importance is relatively small compared with the main
points enumerated above. It is upon them that the system rests,
and upon their adoption or rejection that it must either stand or
fall.

690 CROSS, IDDINGS, PI[RSSON, WASHINGTON

The petrographer who believes that these fundamental points
are correct and who is willing to accept them will not wish to
reject the whole system because he may not agree with some of
its minor features, but the one who does not accept the funda-
mental points will have no use for this system, or any of its
features, great or small.

At first it will not be easy for the petrographer to think of
rocks as expressed in this system ; it will require some use and
experience. Nearly everywhere over the whole field it intro-
duces new conceptions, and we firmly believe more correct and
logical ones than have hitherto prevailed. We have made no
attempt to modify, or patch up, or define the older previous
systems, for we believe that the day for efforts of that kind has
passed ; that they have had their use and should now be replaced
by something more in accord with the results of the great
amount of recent investigations, and of our present knowledge,
and therefore better adapted to present and future needs. It is
of course easier to patch and wear old garments than to make
new ones, but to confess this fact as controlling one’s attitude
toward new propositions would be merely to acknowledge a
condition of mental supineness.

If petrographers believe this system to be an advance over
previous ones, we offer no apology for imposing a new mental
burden upon them, for it is obvious that the ever-increasing
accumulation of knowledge of the chemical and mineral proper-
ties of rocks, and of their relations to one another and to their
texture, mode of occurrence,and their origin impose a greater
task on modern petrographers than was borne by those of a
former generation. And since it is not to be assumed that the
present state of petrographical knowledge is complete, provision
should be made for expansion and adjustment along lines which
seem to be those in which future development will take place.

WHITMAN Cross,
JoserH P. IpDDINGs,
Louis V. PiRssoNn,
Henry S. WASHINGTON.

| Pyrox. | Rock. | Pyrox
————————s ;
SiO, | 45-23 | 42.78 | 44.65
754 744
Al.O 7-73 8.66 6.62
= 075 06:
Fe,O4] 2-95 5.02
| o18 031
FeO || 4-07. | 17.96 | 3.87
} 055 054
MgO }| 12.25 10.06 | 14.76
306 36g
CaO) 3] 23-37.) 12.29 } 20,32
| 437 -363
Na,O) 47 2.31 1.29
| 007 .O2T
K,0O 12 O2U tad Q)
oor | .005
H.07 53. 7iel|i == 300 as
TiO. | 4.28 “128 | 2 93
+053 037
P05) none P
Glee: vee aa
Cr,O0 oe a
NiO .05 trace
MnO SO7I 95
~OOT
BaO none | ~*~
SrO | none |
Ce,0| |
Di,O |
| 700.96 99-87 | 99.95
aes i oh
| |
Sp. gi ‘ 2.829 | 3.411
a) D, 2) Augite from nephel
Surv., p. 63.
6) Dz 0) Augite from lémburg
p) Augite from dolertte
Ou Geol, Surv., p.
g) Augtte from monchié
| 2345 235.
@) Ai r) Augite from basalt (
e) Ai s) Augtte (c:c=54°) fre
El. Gest., p. 34€
7) Ad t) Aegtrite from nephell
2) Aq Serra do Pocos ¢

|
}
|
ail

a) Dieucitophyre. . . .
6) D&nalcite-basalt . .....
c) DNephelinite se
d) ANephelite-basalt....
e) Ae Deiter penta Patt
7a) yeane Seer Se yevalese tonite
£) fon chiquite sects
hk) A@asalt ...... ane
#) AlLeucite-tephrite...... ....
J) py lccires syenite .

sae

TABLE XII. ANALYSES OF PYROXENES AND THE ROCKS CONTAINING THEM.

z B c d e i g A tz J ht Z Mm 1 2
=> — a gq r s t
Rock. | Pyrox. | Rock. | Pyrox. | Rock. | Pyrox, | Rock. | Pyrox. | Pyrox. Rock. | Pyrox. | Rock, | Pyrox.| Rock. | Pyrox,| Rock. | Pyrox | Rock, | Pyrox, | Pyrox.| Rock. | Pyrox.| Rock. | Pyrox. | Rock. | Pyrox, Rock. | Pyrox. | Rock
s yIOx, ock. | Pyrox.| Rock. | Pyrox.| Rock. Pyrox. | Rock. Pyrox.'| Rock. Bie
50.23 0.8 58.68 SOD. 61.93 Srey 42.30 See Saas sess | $0.41 oe 67.42 531297 53:73 46.73 45.59 49:20 42.46 | 44.55 39-92 | 45.23 42.78 | 44.65 48.25 | 49.10 7.964 || Kc 6
* 4 é 6 c : : .821 +742 J ; ‘ e . 47.54 7.01 . 2 s
11.22 | mone} 19.50 Be 13.18 | oie 12.63 om ara0e wile 12.30 os 15.88 | 2.38 20.35 10.05 12.98 | 6.01 18.49 | 7.27 8.60 Ae 8.66 gee Gre een 5 A) 784 RD Bi SEMIS SIG)
3 J d J J 023 vo O71 4 i e +73 95 14.82 | 7. 16. , ,
34. | 3-19 3.63 5658 3.63 | 3-33 | 15:48] 5.86 | 8.5x 9000 5.71 | 3.08 1.37 | 9.25 3.74 3-53 4.97 ae 3.35 6.06 4.40] 2 ae 7065 -078 .077 eS cee 16.09 | 9.04 | 22.55 | 1.92
oes “007 .020 1036 | .053 ,0r9 .057 S555 : ‘038 EE || coca | ee 3-99 | trace| 2.40 | 3.8r Gay} mee Bs et 1018
1.84 | 1.82 2.58 | 12.37 2.31 | 8.39 5.07 | 3.23 | none} .... | 3.06] 4.34 1.14 | 5-15 2.13 8.20 4.70) 4:23 6.31 | 5.91 Eco)! cecr || panes || apes 024 1008 pea) cose eases zen
1025 .172 i} owes O44 060 071 059 .082 1055 ae en Oe) Big 7.52 4-53 6.67] 8.15 3.62 Aes ast
7-09 | 17-42 +79 | 10.98 EE) || EELS G28 || eg || PUR) 8.69 | 14.21 1.435] (9-43 “47 9.68 8.36 | 12.40 3-64 | 10.44 20.17 | 12.25 OHS Weve 3 115 | > -063 113 “028 ca seeer
435 274 .325 -309 321 355 .236 {310 eer "306 se 5:77 | 12.37 6.98 | 12.71 6.38 | 13.52 5.53 | 12.09 X I 3
5.99 | 23-32. | 3.03 | 22.01 3.48 | 21.30 8.42 | 18.49 | 18.42 s-- | 7-08 | 22.58 3-49 | 17-81 2.72 13.22 11,09 | 2r.79 8.70 | 22.83 | 10.68 | 23.37 | x29 0) +309 318 33 302 32 leas
416 303 3.80 -330 | .329 403 | 318 389 “407 gry | 9 | 722386, | 8:32 | 22st) 0.8 | 20.84 || 8.74] x9133 | 10.68 | 21.78 | 1.85) 4/25
aay 196 5.73 | 2-14 2.67 | 1.02 5-19 | 3.80 2.56 oe: 97 67 6.42 | 2.63 7-94 1.81 4.53 9 2] 41.4 1.gt 393 = 402 .336 345 “38 PES GES
¥ “o12 034 o16, o6r O41 o10 042 : nee de ue : se CESS 280) 3.24 | trace | 3,08 1,29 2.81 ner} 4.46 qice 8.10 Bien
g.8r 42 4.50 -94 6.11 -5° 2.73 | 1.03 -78 S506 7:53 -06 2.65 38 6.05 3.76 1.04 v4t 4.58 ga 1.03 bot S 021 ‘ Sar base San . el
004 .o10 -005 .OIL .008, .000 004 004 006 Fear y yo 4.01 trace | 2,90 49 1,10 ebahe 4.05 56 7.05 1.05
1.72 31 1.01 odes L.14 3-59 bene r,80 | undet. | ( os or Ce? 1.24 3.40 | undet. 2,31 Ia. I.45 .37 3.96 poe. 005 oot 006 e
ac Seas -46 none |) °~ tote oon -5r | undet, ewes Bae +43 soe aren xe ale oe *26 EX) a 24 <31
66 1.47 +70 +09 -78 1.32 1.53 pone 1.36 2.70} 4.28 28 “86 ps poss tase sete A006
.003 009 1019 ‘O19 1053 ay 2.85 2.76) 1,82 62
see -46 07 3.51 553 eae aon ee “5st | none 68 -036 : 008
none one ans 2.44 “sires 5 A O00 x0
| trace BSG 18 105 52 trace a5
trace none ee is te 14 Oh a Sonn
04 -03 oso iele 06 . cis trace
“15, -28 trace 28 a pee ae 24 o ee gon oe 000 Sar trace Agdd
004 . 001 005 CO 2: Se race aH09 tees +38 219 -20 trace
1.23 eno agse os Car06 aon oeO8 Sate ea0 arta +23 none sean pao e500 da6 po60 dnan 4 none gbos 13 ? ae wate 06 none Peer sp BoC) +003
24 StORs 06 none pone eed dade Snd0 o0d0 oubd 236 none daa 12 06 wae 25 04 none ie rot 523)
od 3605
103 ZxO,. none none Boy whee 959 A600 aeee Soa0 S none mae 03 Kone ents ab00 Sines oan Sent eae
Glen ‘¢ 8 8 | < 68 1 a Mataree eeece
: 100. : 04 101.42 101.00 |100.90 100,42 }100,2 92 |100.2 : 0 100.97 |100.21 00, : . : : oo
99.16 | 100.99 | 99.73 | 99:04 | 99.84 4 9° 4 7 | 99.9) 3 | 99.98 | 99.74 97 1 100.21 | 99.42 | 99-87 | 99.79 | 99.92 |x00.41 |F. 07 |100.96 | 99.87 99:95 _|100.16 [00.26 |100.23 | 99.3r | x00.51 | 99.85 r00.42 |100.78 | 99.70 | 99.01
FB. .50 100.45
100.62 2038)
22 100.42
5 100,40 68 alae ———
Pp. 2r..| 2.779 3.401 ere see 2.074 3.227 3.2 ade 2. 3.45 pnd bate Bach og00 60.08 56.00 2.79 3-43 hod aane teat aes 46 mie G00 8: ;
ae ee ssa eel ee 55a coerced lena lemeneee ice oss al es sab oesee ee Seno I Seeees ee Seema een nega en [eee | eae | meee ieee Pe een | esc eee Oe CUES SEY
a) Diopside {rom mixed wyomingite and madupite; 10ck analysis of wyomzugite (felsophyro-wyomingose) , ¢:a=37° for colorless, c:a=22° for green, from augite-soda-granite (grano-lassenose) Kekequabic Lake, n) Augite from nephelrte-basalt (uvaldose), Black Mountain, Uyalde county, Texas, ui. 108 U. S. Geol
Leucite Hills, Wyo, A.J. S., Vol. IV (1897) p. 130. Minn. Am. Geol,, Vol. XI (1893), p. 385. 5 > y, Texas, » S. Geol.
93), P. 395; Surv., p. 63,
6) Diopside (A. Merian) from /aurvékeite (grano-laurvikose) Byskoven near Laurvik, Norway, BrocGEr, W.C., h) Aegirite-augite from nephelite-syentte (grano-laurdalose), Zwart Koppies, Transvaal, RosENBuscu, ZZ, Gest., 0) Augite from lzmburgite, Limburg, Kaiserstuhl, Rosenpuscn, 2/. Gest. p. 363
: 5 BON. \s SCH, Zl. Gest., p. 363.
; Rruptiogesteine Kristianiageb, Vol. 111, pp. 23 and 3. Pp. 122, 126, 5 p) Augite from dolerite (shoshonose), Valmont, Colo. Budd, 150 U. S. Geol, Surv. p. 264, and Bull. 18 U, S
c) Diopside from pyroxene-granitite, Laveline, Vosges, Rosenbusch, El. Gest., p. 73; rock analysis of azgzte- 2) Augitte from shonkintte (grano-shonkinose), Square Butte, Mont. Pirsson, Bu/?. Geol. Soc. Am., Vol. VI Geol, Surv., p. 140. re
granitite (grano-adamellose), Laveline, Vosges, RoseNpuscH (v, Werveke), Neues Jahrb. Min. und : (2895), Pp. 410, 414. } q) Augite from monchiguite (ourose), Rio do Ouro, Serra de Tingua, Rio de Janeiro, Rosennuscn, £7. Gest., pp:
Pal., etc. (1881), Vol. I, p. 235. J) Augite from tinguazte (?) (laurdalose), Two Buttes, Colo, Bull. 168 U. S, Geol. Surv., p. 165. 234, 235.
d) Augite, ce 9°30 2 Hus : 0" i fromtccareciaraincdinephelitesbasald(eranormaliencsoys hk) Acgirrteangete from leucitophyre, Burgberg, near Rieden. Rosensuscn, £2. Gest., p. 278, ry) Augite trom basalt (auvergnose), six miles N. E. of Grants, N. M. Bull. 108 U.S. Geol. Suri, p. 170,
€) Augite, c: C=34", 2 Ha=84°50' (yellow) 2) Augite ftom analerte-basalt (monchiquose), The Basin, Colo, Bl? 108 U. S. Geol. Surv, p. 146, and Jour, 5) Augite (c:C=54°) from /euctte-tephrite (monchiquose), Falkenberg, near Tetschen, Bohemia, Rosennuscu
Katzenbuckel, Odenwald. Rosrnsuscn, 27. Gest., pp. 354, 357+ Gezot., Vol. V, p. 684. El. Gest., p. 346. . ;
; T) cee from syenttre-lamprophyre (prowersose) Two Buttes, Colo. Bull. 168, U. S. Geol. Surv, p. 16s. m) Augite, yellow, c: C=39°, from nefphelinite (hauynophyre) (vulturose), Melfi, Italy. Rosenpuscu, 22, Gest., t) Aegirite from nefphelite-syenzte (grano-janeirose) , Barreros, rock analysis from tunnel between Prata and Cascada,
&) Augite, bottle green, in part colorless, with slight pleochroism, a and b bottle green, c yellowish green, a= b>. Pp. 357. Serra do Pocos de Caldas, Las Paolo, Brazil. Rosennuscu, 2? Gest., pp, 122, 126.
TABLE XIla. MOLECULAR RATIOS IN PYROXENES.
|_sio. |i on | mice ne
Al,O, | Fe,O, |(NaK).0]| MgO | FeO | SiO. | TiO. | pemic | satic Rien A1,0,,| Fe,O, | (NaK),0] MgO |_FeO_|_Si0s | TO. | pemic | ‘Satie Rast
CaO CaO CaO CaO CaO CaO CaO SiO. CaO CaO CaO CaO CaO Cad CaO SiO,
3 were sees phyro-wyomingose Wyomingite .00 ol 03 1,04 06 2.03 .0g 1.00 .00 cheer 50050 odcna|| HOLA. ..+.+++.] Leucitophyre . Eas 12 .10 14 52 49 2.25 .02 +881 +12 | —,013
Bs peaks grano-laurvikose Laurvikite ....... .00 .00 .It 69 -44 2.13 .02 1.00 .00 monchiquose , Analcite-basalt . .-... © 15 os Of 79 16 2.11 +05 .8at 18 +000
d) eae grano-adamellose - Pyroxene-granitite. .02 105 .05 85 +31 2.22 102 . 9861 Org vulturose .. Nephelinite 17 09 07 64 20 1.82 Of Bat «18 | —,053
i) nets grano-malignose .... Nephelite-basalt ... . 04 II +21 -93 14 2.66 107 -95t .0S uvaldose Nephelite-basalt 18 Of 02 723 Th 1.80 113 Br 19 —,039
nwt gite grano-malignose Nephelite-basalt 06 16 “15 +97 Or 2.64 +12 881 | 112 (poor analysis Limburgite... 18 08 .07 1.01 15 2.05 10 842 10 —,040
‘) pes prowersose ..... .., Syenitic lamprophyre ..... 07 +04 03 -88 +16 2-11 +02 grit 09 shoshonose . Dolerite |. 19 . oo 17 28 2.03 ou Bre +19 | —.036
WA iene grano-lassenose. - Augite soda granite. ... -07 .18 114 +74 +22 2.78 .00 831 +17 ourose ... Monchiquite . .- 22 07 o7 94 20 2.26 «10 Bar 18 —.067
1) iM ee grano-laurdalose.. Nephelite-syenite .. ° +09 44 “34 +26 -6r | 3.23 +00 871 “13 auvergnose,.. . Basalt ...... 4 22 02 02 98 33 | 2.27 170 1792 | .ar 1000
aN A gite...0. . 2 grano-shonkinose . Shonkinite ... .10 204 .05, 85 +19 2.06 102 .got .10 monchiquose , Leucite-tephrite.....- i 22 Ir .06 +79. Tq 1.95 02 764 24 —.028
ugite ... 500 laurdalose. . ....| Tinguaite (?).. 10 09 .06 65 124 2.05 10 ,80r ara grano-janeirose ... Eleolite-syenite . 23 2.16 2.02 38 76 | 11.34 +00 864 iy 000
—= 2 =
All femic Ca, Mg, Fe calculated as metasilicate. 2 Femic Ca, Mg, Fe calculated partly as orthosilicate and partly as metasilicate. Deficient SiO, is that required to raise part of the salic components to leucite or nephelite.

a me 7) p
a iY ee
d- Cossy- | Ainig- _ |Amphi-
Rock. | Hornbl. Re Rock. | “vite. | matite.| Rock: | bole
aa]
SiO, ...| 66.83 | 47-49 6 | 70.30 | 43.55 | 37-92 | 74.76 | 49.x0
791 I 726 632 .818
Al,O3..| 15.24 |- 17.07 q 6.32 4.96 3.23 11.60 5.50
.069 3 048 O31 054
Fe.O,.. 2.73 | 4.88 9.23 7.97 5.81 3.50 4.20
.030 : 045 ( ,036
FeO.... 1.66 |. 10.69 1.40 | 32.87 | 35.88 .Ig | 27.7u
.148 -456 .498 385
MgO... 1.63 | 13.06 .89 . 86 33 .18 17
.326 > O21 008 004
CaOMe: Bash |ert.92 | 84 2.01 1.36 .07 pane}
2253 5 036 024 002
Na,O.. 3.10 75 } 7.70 5.29 6.58 AEE iil|tO50
.O12 i 085 . 106 169
K,O 4.46 -49 2.50 B33 Lor 4.92 1.60
.005 003 005 O17
H,O+ -56 1.86 82 5 64
H.O — none ys
(CO ee See cn
TiO; 54 1.21 Bt stale 7.57 trace
.O15 . 004
PrOF 18 | none ans HS: trace
SOME Wits:
Clan .02 saoe t
Cr,0, Bays none
NioO.. 02
MnO. Io Riche 1.98 100 50
+007 027 o13 .007
BaO. ares none | i
SrO -03 | mone gaoe
ZrO. .04 | F.06 | Curso) ane ao pee ff
100.82 |V,O3.04 10g.00 | 100.21 | 100.19 | 100.21 | 99.40
100,05 3-74 3.80
22 | 3-75 | _ 3-852
100,03 lh
|

a) Hornblende from quartégrano-liparose), Quincy, Blue Hills, Mass.,
analysis from gvartz- p, 181.
Yosemite Valley, Calz#e, San Miguel, Azores, Rosenpuscu, £7.
Surv., p. 208, and A

b) Hornblende from guartzbelite-syentte, Kangerdluarsuk, Greenland,
tion, Butte, Mont. p. 122,

c) Hornblende (green) from) paxntellerite (felso-varingose) Khartibugal,
blende-dactte (bands, El. Gest., p. 257.
Gata, Spain, RoSENBe-syenzte, Julianehaab, Greenland, RosEN-

ad) Hornblende from guard,
Table Mountain, Oa felso-liparose), San Pietro, Sardinia, RosEN-

\

TABLE XIII. ANALYSES OF AMPHIBOLES AND THE ROCKS CONTAINING THEM.

b : ‘
@ G a C i & h Z J & Zz m n ° p
Rock. | Hombl.| Rock. | Hornbl.) Rock. | Hombl,| Rock. | Hornbl.| Rock. | Hornbl.| Rock. | Hornbl,| Hornbl.| Rock. | Hornbl.| !445t- | Rock, | Barke- Riebeck}] Kato- | Arfved- Cossy- | Ainig- Amphi-
ingsite, vikite. Rock. ite. | phorite.| sonite. Rock. red mative: Rock reaies
—_
SiO, ...|, 66.83] 47-49 ,| 63.88 A573 as AE ne) so.c8 oY A6:08F Se 39-62 41.35 | 59-37 | 46.22 34:18 49-65 | 45.53 | 43-85 | 70.30 74.76 | 49.10
791 . «762 -835 +7! -660 689 +779 -569 ine F : 5 :
Al,Og..| 15-24 7.9] 15.84 Su, 15.60 B8e. 12,06 7.97, -82 | 10.52 | 13.59 | 14.92 | 13.48 | 17-92 8.12 | 11.52 at (oe? AI 6.32 re oo
.069 d .0} .07' -103 1.46 132 -079 113 : E °
2.73) 4.88 2.11 | 4.94 5.26 5-32 7.81 2.69 85 2.81 5-55 | 10.28 5-14 6.77 9.33. | 12.62 uae 9 se ae 9.23 © 2054
030 .031 033 016 017 094 032 058 -079 550) ‘os8 <024 oes Ag
1.66 |. 10.69 2.59 | 10.39 1.36 11.23 5-03 6.71 .26 8.30 4.03 7.67 | 10.33 2.02 15.18 | 21.98 19.55 23.72 33-43 1.40 2
148 144 -156 093 101 - 106 143 +210 +395 -271 301 404 a
1.63 | 13.06 2.13 | 12.32 2.61 14.08 11.86 16.31 +44 14.40 1.66 11.32 | 11.44 1.83 5-20 1.35 =a 2.46 81 -89 18
.326 -308 +352 408 360 -283 . 286 130 -034 io6n 020
3.59 | 11.92 3.97 | 11.25 6.55 | 10.62 7-74 | 11.21 .02 | 12,64 5.13 | 12.65 | 10.93 4-16 | 10,08 9.87 3.16 4-89 4.65 84 07
213 201 189 - 200 226 226 195 180 176 1056 .087 083 .
3.10 “75 2.81 17 2.50 1.3 2.35 1.22 +75 1.62 5.31 t.12 | 2.10 | 1.24 2.46 3-29 7.61 6.07 | 8.15 | 7.70 4.35
O12 o12 -022 O19 026 018 034 039 053 “122 1098 13K
4.46 49 4.23 1,22 1,63 -26 1.01 46 -22 -34 4.64 2.18 .62 6.68 1.23 2.29 6 1.06 2.50 4.92
005 013 -003 005 004 023 007 +012 024 O11
-56) 1.86 -66 | 2.29 | 2.25 -85 2.44 £.40 +27 1.97 1.25 -48 48 238 1.36 +35 1.67 15 .82 64
none “ane .22 49 epee Shon .12 aoe .08 “17 Ste ar ong 680 400 Pash ee 4a
“orl ceeve || eB “% 1.43 | .6x (76 | .92 "7 119 ‘26 | 7108 | 1.53
O15 .o18 -009 -009 -002 013 -O1g
.18 | none 2 ab 08 trace +74 18 SDA -58 - - trace
-02 aut trace a es 43, :
none oe on -16 dda a
tes 102 00 o8N ooiee 205 areere ago da00 trace trace 9505 S00
10 151 .07 54 -57 +13 .49 124 63 .09 +15 trace 45
.007 008 .008 007 .003 -009 -002 .006
.t1 | none .og | none 05 none ; suas none oon
.03 | none .o2 | none trace o0U8 ab 00
.04 | F.06 55a oar Ve OFos lees sa ae
100.82 |V,0,.04| 99.82 |I". .28 | 99.97 100,31 | 100.01 | 100.46 | 99.86 | 99.99 |100.38 |100.67 | 100.84 | 101.21 | 100.26 | 99.61 100.45 | 99.91 100.91 | 100.64 | 99.96 | 100,80 | 109.00 r00.2t | 99.40
O15 oe 10
100.05 98.77 |Sp. Gr..| 3,212 2.74 3.266 3.25 | 2.710 | 3.157 | 3.433 | 100.35 | 3.437 | 2.642 3-43 3.44
~O2 .12
100,03 98.65

*Al,O, and TiO, estimated from rock analysis.

a) Hornblende from quartz-monzonite, S. E, Mt. Hoffmann, Cal.; rock
analysis from guarts-monzonite (grano-toscanose), Nevada Falls Trail,
Yosemite Valley, Cal, (average rock of region). Bull. 168 U. S, Geol.
Surv, p, 208, and A./,S., Vol. VII,(1899), p. 297-

6) Hornblende from guarts-monsontte (grano-amiatose), Walkerville Sta-
tion, Butte, Mont. Bz//. 108, U. S. Geol, Surv., p. 116.

©) Hornblende (green) from daczte, Grenatilla, rock analysis from horn-
blende-dacite (bandose), San Pedro, Sierra del Cabo, Cabo de
Gata, Spain, Rosensuscn, 77. Gest,, pp. 285, 286. -

d) Hornblende from quartz-dtorite (grano-camptonose), 46 miles south of
Table Mountain, Cal.

e) Hornblende from hornblende-gabbro (grano-hessose), Beaver Creek, Big
Tree Quadrangle, Cal. A.J, S, Vol. VII (1899), p. 297.

}. Hornblende from andesite (umptekose), Stenzelberg, Siebengebirge,
Rosensusch, £2. Gest., pp. 290, 292.

g. Hornblende from “ hornblende-diabase,”’
Rosenpuscu, ZZ. Gest., p. 306.

h) Hornblende from syentte, Biella, Piedmont, Italy, Rosenpuscu, 2, Gest.,
Ppp. 103, 106, .

7) Hastingstte (deep blue) from nefphelzte-syenzte, Dunganon, Ontario,
Rosensuscu, Z/. Gest., p. 122.

Graveneck, near Weilberg.

Al,O, determined as 17.65 including TiO,.

k) Riebechite from granite (grano-liparose), Quincy, Blue Hills, Mass.,
A. F. S., Vol. VI, 1898, p. 18x.

2) Katophortte from santdintte, San Miguel, Azores, Rosensuscn, £7.
Gest., p. 266,

m) Arfvedsonite from xnephelrtte-syentte,
Rosensuscu, £/. Gest,, p. 122.

n) Cossyrtte (ainigmatite) from fantellerzte (felso-varingose) Khartibugal,
Pantellaria, RosenpuscuH, £7. Gest., p. 257.

0) Ainigmatite from nephelite-syenite, Julianehaab, Greenland, RosEN-
BuscH, £/, Gest., p. 122.

Kangerdluarsuk, Greenland,

Bull, 108 U.S. Geol. Surv. p. 190, Jj. Barkeutkite from sodalite-syentte (grano-pulaskose), Square Butte, Mont. p) Amphibole from comendite (felso-liparose), San Pietro, Sardinia, Rosen-
A,J.S., Vol. XLV, 1893, p. 292. BuscH, Z/. Gest., p, 257:
TABLE XIIla. MOLECULAR RATIOS IN AMPHIBOLES.
Minerals Rocks Al,O5 | Fe,0; |(Na.K).0)_ MgO FeO/ | SiO5, | TiO, Femic. | Salic “ick
CaO CaO CaO. CaO CaO CaO CaO i sio,
~oo —
a Hornblende grano-toscanose....... Quartz-monzonite .. +32 14 08 1.53 69 3.71 771 23 000
é Hornblende grano-amiatose.. ....| Quartz-monzonite, +32 +15 12 1.53 “71 3-79 owes 23 +000
c Hornblende . grano-bandose .. Dacite. : +45, “17 13 1.86 82 4.03 09 .24 | —.007
d Hornblende .... grano-camptonos Quartz-diorite 39 08 12 2.04 46 4.17 6506 .22 | —.or4
e Hornblende grano-hessose Gabbro......... +: = “45, .07 13 1.60 +44 3.46 29 —.003
7 Hornblende ...... grano-umptek Andesite (?) trachyte.. 64 <4 18 1.25 47 2.92 6980 +39 | —-095
& Hornblende .. no analysis Diabasetaecet 67 16 ~21 1.46 73 3.53 -31 $43 .000,
/ Hornblende (?) analysis . Syenite. ...... -44 132 28 72 1.16 4.28 07 +34 .000
z Hastingsite no analysis Nephelite-syenite. 64 45 43 EB} 1.60 3.23 10 +33. | —-040
e grano-pulas Sodalite-syenite. .85 12 .35 19 1.73 3.40 238) ||| 2165,
i grano-liparose . Soda granite. +23 1.9 2,18 .0 4.84 14.76 06 .000
7 Katophorite A no analysis Sanidinite. . -46 66 1.23 +70 3.46 8.72 .18 | —,030
mt Arivedsonite .... no analysis... Nephelite-syenite «5 -29 1.71 24 5 59 8.80 19 .000
2 Ainigmatite felso-varingose Pantellerite, _ 1-3 1.3 2.4 +58 | 12,60 | 20.01 pons +24 -000
o Ainigmatite no analysis.. Nephelite-syenite. 1.3 1.5 4.6 33 | 20.75 | 26.33 3-9 16 000
p Amphibole .... felso-liparose. . Comendite........- «+ .| 27-0 18.0 93.0 2.0 |192. 409- Sc 29 900

1 All Pemic Ca, Mg, Fe, calculated as metasilicate.

Deficient SiO, is that required to raise part of the salic components to leucite or nephelite.

2Wemic Ca, Mg, Fe, calculated partly as orthosilicate and partly as metasilicate.

Sr
ee i
)
aan

yd
iat

ns

f
a
; PTs i

ot
a UN

7
erat

ema h)

it

TABLE XIV. ANALYSES OF MICAS AND THE ROCKS CONTAINING THEM.

6 . $
a c dad e f rg hk t } ,
Rock. | Mica. | Rock. | Mica. | Rock. | Mica. | Rock. | Mica. | Rock. | Mica. | Rock. | Mica Mica, Rock. | Mica.
63.88 Bae 66.91 | 35-62 | 77-61 | 31-96 | 60.39 | 32.09 | 54.55 | 34-37 | 56.26 | 33.24 | 36.42 $0.23
G 594 -533 -535 +574 , 554 -
15.84 13.70 15.24 11.94 | 15.93 | 22.57 18,52 19.07 6.84 | 23.59 14.90 17-92 11.22
+134 .149 -117 “1 .067 -146 <175
2.1% 5-22 : 4.69 +55 8.06 -42 | 19.49 2.4 | 24.89 85 5.92 2.83 3-34
032 .029 .050 12 -156 +037 017
2.59 13-72 é 13.67 .87 30.35 2.26 14.10 3-12 7-47 2.61 23.57 7-4 r.84
-1g0 .190 -421 196 .104 -327 .098
EXE}, |) FErEE} =. || 12.70 trace 05 13 1.01 1.98 4.05 +27 5-15 | 20.52 7.09
+303 -317 -0o1 -025 ror 129 +513
3.97 05 | 3.5% +95 -31 +23 +32 ibe 3.15 .78 “54 -40 oo 5-99
-OO1 .O17 004 .OT4 .007
2.81 “15 3-59 .5° 3-80 1.54 8.44 7.67 2.13 7-77 1.45 2.60 1.37
-002 .008 .025 034 .023 042
4.23 9.09 3-13 7-72 4.98 8.46 4:77 4.84 9.03 5.72 7-77 6.54 9.81
097 .082 -090 -096 .082 .069
66 3.64 4.36 trace 4.25 -57 72 2.27 -37 2.19 2.50 1.72
~242 .127
22 1.21 94 +23 3€ Sood ees 93
ae B50 : 4000 came, Ao 5 1.37 tel
.65 3.52 -25 1.40 4.68 “47 2.27
O44 oe a 058
2X 10 cc 1.89
34 + -74
trace 20 +03 |
on . a 10
107 119 ees ||| cae || oy | eye || ee Wack || fen || ces 105
OIL 002 -010 -003 020 034 013
BaO..... -O4 trace .It 12 .09 +13 ee a5 1.23 1.00
006
SrO)..... 02 none +03 (?) 02 none aden trace Bae tees pone \PSoNeEE!! sce aode e S10 .2 trace
Ogres .08 |F, .26 |ZrO,.04) none «. |F..76 eee GaSe sec eee wae Ota © i ea
O14 032
100,60 }ro0. 11 100,82 |F. .17 99.82 99-59 100.00 |100.54 |100.46 99.95 |100.92 99.82 98.92 99.91 |100.27
V..0,.05 Sp. er 3.084
99.90
07
99.83

a) Biotite from granodiortte (grano-toscanose), El Capitan, Yosemite Valley, Cal. A.
J. S., Vol, VII (1899), p. 294.

4) Brotite from guarts-monzontle (grano-toscanose), Nevada Falls Trail, Yosemite
Valley, Cal. A, J. S., Vol. VII (1899), p. 297, and Bull. 168, U. S. Geol.
Surv, p. 208.

c) Brotite from guarts-monsonite (grano-amiatose), Walkerville Station, Butte, Mont,
Jour. Gron., Vol, VII, and Bud? 168, U. S, Geol, Surv., p. 116.

d) Birotite from quarts-monsontte (grano(?)toscanose), Bloods Station, Alpin Co.,
Cal. A.J. S. Vol. VII (1899), p. 294.

¢) Lefpidomelane from grantte (grano-liparose), Cape Ann, Mass. A, J. S., Vol.
XXXII, 1886, p.359. Rock analysis, H. S, Wasuincton, Jour. Grot., Vol. VI,

/) Lepidomelane from nephelite-syentte (grano-nordmarkose), Litchfield, Me. Bxé?,
108, U. S, Geol, Surv, p. 21.

&£) Lefidomelane trom syenite-pegmatite, Barkevik, Langesundfjord, Norway; rock
analysis from /aurdalrte (laurdalose), north of Love, Laugenthal. Briccer,
Eruptivgest. Kristianta geb. Wl, pp. 19 and 34.

h) Lepidomelane {rom (?) nephelite-syenite, Miask, Dana, System, p. 630; rock analysis
from arascrt (grano-miaskose), Mt, Lobatchia, Siberia. A. KARPINSKI,
Guide Excurs. VII, Cong, Geol. Int, V (1897), p- 22.

4) “‘Biotite’’ from monchiquite, Horberig, Oberbergen. Rosenpuscn, 2%. Gest., pe 234.

J) Phlogopite from wyomingite (phyro-wyomingose), Boar's Tusk, Leucite Hills, Wyo.

Pp. 793+ A.J. S., Vol. IV (1897), p. 130, and Bull. 108, U. S. Geol Surv., p. 85.
TABLE XIVa, MOLECULAR RATIOS IN MICAS.
Al,O. | Fe,0, | MgO FeO SiO, TiO, Rami: Salic. evo
KyOm | mxgON | Ke ON |KGOueKGON|NKaO SiO,
@) Biotite ....................| grano-toscanose ........ Granodiorite... anaes 1.8 -33 “47 2.07 5.71 -13 “$4 | «46
4) Biotite... grano-toscanose. Quartz-monzonite 1.4 -28 3-15 2.00 5.78 38 52 -48
¢) Biotite, grano-amiatose .| Quartz-monzonite 1.3 +32 12 1.96 6.01 44 52 48
) Biotite... no analysis .... .| Quartz-monzonite 1.9 32 .86 2.31 6.60 -30 54 46
¢) Lepidomelane. grano-liparose . Granite ..... 1.0 43 0 4.07 4.63 37 g8) |||) 245
7) Lepidomelane grano-nordmarkose Nephelite-syeni 1.6 tt -29 | 2.28 4.82 ah 40 | .60
#) Lepidomelane, grano-laurdalose Nephelite-syenite . 5% 1.2 1.05 1.07 4.26 -44 02 +38
A) Lepidomelane. grano-miaskose Nephelite-syeni' rem +35 1.57 3.98 5:27 .56 | 53 47
1) “ Biotite’’. no analysis Moachiquite. 1.5 15 7-42 1.42 5.46 FY |) oe | .48
J) Phiogopite. phyro-wyorr Wyomingite -98 14 4.9% 10 5.86 ar | 149 5%

i mn 8
chy abet oy Fkok eo

eae Heavens

pe)

Ibis.

WOWRNAL OF GEOLOGY

OGROLERA_NOVE UBER, To02

CARTOGRAPHIC REPRESENTATION OF GEOLOGICAL
FORMATIONS.

AS CARTOGRAPHIC units, in the representation of geological
structure, the lithologic individual, the faunal stage and the
geological formation are commonly regarded as practically iden-
tical. In reality, they are fundamentally distinct from one
another. Moreover, these are not the only units which it is pos-
sible and practicable to map and bring out clearly the geological
structure of the area investigated.

In the recent discussions on the units of geological mapping,
one of the most important fundamental factors appears to be
entirely overlooked. The two leading phases of the subject are
admirably summed up in the recent articles of Messrs. Willis!
and Cross.* Although, at first glance, these authors seem to
present radically different views, they are not, actually, so far
apart in their contentions as they would have us believe. Mr.
Cross’ conception is the more philosophical of the two; it is
based on genetic grounds ; and it is the one which must finally
prevail, though the local criteria of discrimination may be diverse
in different cases. In actual practice, Mr. Willis’ expressed idea
has the greater force and must be the one which must neces-
sarily long be followed. But the two conceptions are not
incompatible. In the practical application of the principles, the
final results become very nearly identical.

To every one who has given the subject critical attention, it

tJour. GEOL., Vol. IX, p. 557, 1901. 2'Tbzd., Viol. XZ p: 22251902.
Vol. X, No. 7. 691

692 CHARLES Rk. KEVES

must be quite evident that no standard yet proposed for the
delimitation of geological units in cartographic representation
exactly expresses the essential features of a complete and
rational scheme. As in all classifications of natural objects, that
of geological units, formations, or terranes should be in its
highest type genetic in character. Moreover, it should be strictly
stratigraphic, depositional or sequential, using these terms in
their broadest sense to include all rock-masses, igneous, meta-
morphic and sedimentary. With this understanding of the
theme, the most obvious characteristic of a stratigraphic scheme
is not the terranes or rock-masses themselves, nor any of their
contents, but their geometric elements, their bounding or strati-
graphic planes. In order not to carry with it the usual narrow
idea, some such term as depositional or sequential planes should
, be used in place of the name stratigraphic, and these titles, will
be hereafter given preference. The sequential, depositional or
sedimentation planes have different taxonomic values according
to the general scheme of classification adopted’.

A cross-section of the cartographic units, or geological for-
mations, as they occur in nature, may be represented by the
following sketch:

* Fic. 1.—Geologic units in cross-section.

In the general, abstract or ideal instance, each lens-shaped
figure is the unit established ‘with regard to all the facts and
conditions of the case, and not upon the restricted basis of any
part of those facts. It represents as much of the geological
development of the earth recorded in the area covered as is
practicable.” The net-work of formations coincides with the

"American Geologist, Vol. XXIV, p. 294, 1899.

REPRESENTATION OF GEOLOGICAL FORMATIONS 693

lines and planes of sedimentation, or natural s€quence, to use a
more comprehensive term. It gives a foundation for deter-
mining geologic structures and deciphering geologic events.
This foundation is the same as that governing sedimentation and
the sequential arrangement of rock-masses. The scheme is,
therefore, genetic. It is of secondary importance to consider
the composition and contents of the various lens-shaped figures.
In some cases most value must be placed upon the lithologic
character of a terrane. In other instances, the contained fossils,
or minerals, are the determining factors. Under certain con-
ditions still different features must be taken into consideration.
The cardinal fact to be always recognized is that the cartographic
unit, the geologic formation, is essentially an abstract concep-
tonne Tt may also be a lithologic, or faunal, or mineralogic, or
physiographic, or some other kind of unit capable of being
represented on maps.

As a matter of fact the lithologic features, the faunal char-
acters, the mineralogic contents, as well as many other criteria
of discrimination, are so grouped genetically to the depositional
units which it is desired to represent on the map, that if the
decipherable record of each were perfect, a cartographic repre-
sentation of the one set of facts would in a general way indicate
the probable outline of each of the other sets. But the fact that
the records of all of these groups of data for the determination
of the geologic formations are at best comparatively so meager,
makes it incumbent upon the geologist to delimit his carto-
graphic units at first according to the most obvious features
presented in the several areas covered.

It so happens that in the field the most obvious and most
useful single feature in recognizing and tracing a geological for-
mation is the lithologic. Checked by other criteria, then, the
lithologic unit corresponds very closely to the ideal cartographic
unit established. For all practical purposes for which the
geologic map is constructed lithologic individuals are amply
sufficient and accurate. When more refined investigation is
taken up some slight changes in the lines of formational delimi-
tation may be necessary; but if the lithologic determinations

694 CHARLES Ks LEVEES

have been made with ordinary care it will probably be rare that
radical alteration will be demanded.

In the present stage of geologic inquiry it is neither prac-
ticable nor desirable to map all districts with uniform refinement.
Among geological formations, we shall no doubt eventually
establish and indicate on all geological maps about five degrees
of taxonomic rank. The division lines in any one area can only
be fixed after very careful comparisons with those of all the
neighboring districts. In the main, these terranal lines will be
found to correspond to, or can easily be adjusted to, the divisional
lines separating the lithologic individuals ordinarily recognized.

When the lithologic features of formations gradually merge
into those of others, or when there is a rapid alternation of dif-
ferent kinds of rock layers, fossils or minerals, other criteria
may have to be resorted to in order to properly delimit the ter-
ranes. But this fact certainly in no way invalidates the general
principles involved in the recognition of the lithologic individual
as the leading object to be represented in cartography.

When, for example, it was found upon detailed faunal exami-
nation’ that the great St. Clair limestone of Arkansas, which
had long been considered by the workers of that state as a
single lithologic unit, was in reality two great limestones of
almost identical lithologic appearance, the one Ordovician in
age, and the other Silurian, it did not render worthless the maps
upon which these two terranes had been represented as a single
cartographic unit. Nor is the principle of mapping the htho-
logic individual to be given up on this account. Early observa-
tion was merely insufficient.-

In the case of the ferruginous sandstone of southeastern
Missouri and southwestern Illinois there is not a single con-
tinuous stratum, but a large number of disconnected deposits,
lithologically indistinguishable, and lying, at least, at two very dif-
ferent geologic horizons. One horizon is below the Kaskaskia
limestone and the other above that great rock-mass. The one is
early Carboniferous; the other mid-Carboniferous. The lower
continuous terrane is known as the Aux Vases sandstone; the

Am. Jour. Sct. (3), Vol. XEVILL, p. 327, 18094.

REPRESENTATION OF GEOLOGICAL FORMATIONS 695

other is formed by the basal sandstones of the Coal-measures.
Farther southward, in Arkansas, the difference in the strati-
graphic horizons of the two is upwards of 20,000 feet. Yet,
because of peculiarities of position, the existence of an interven-
ing unconformity plane, and the nearly same level above the
sea of neighboring outcrops, of the two horizons on the Missis-
sippi river, Worthen’ and others were led to erroneously ascribe
to the Kaskaskia (Lower Carboniferous) an extensive flora of
the Coal-measures.

A different example is that of the Carboniferous of Arkansas.
There is the enormous thickness of 26,000 feet of sediments.
Sedimentation has been uninterrupted throughout the entire
sequence. In the last formed terrane of the Lower Carbonif-
erous, there begins an alternation of sandstones and shales,
with some coal seams, continuing to the top of the section.
About 24,000 feet of this section may be regarded as a litho-
logical individual quite ‘‘uniformly varied in character.” Data
obtained farther north in Missouri show that 23,000 feet of
this enormous section are unrepresented. The Arkansas sec-
tion belongs to at least three great terranes, each having a tax-
onomic rank of series. Measured in feet, the median one alone
is five times as great as all the rest of the Carboniferous repre-
sented in the Continental Interior. The conditions presented
are represented below. Viewed from Arkansas alone the lines
separating the distinct
geological formations

might forever remain
unnoticed in the great

Fic. 2.—Carboniferous Sedimentation in Missis-
sippi Valley.

lithologic individual.
It is only by a com-
parison with sections in other localities that the terranal
divisional lines may be properly drawn.

There seems to be only one answer to the question: ‘‘What
should a geological map represent?” That is Mr. Cross’ obser-
vation that ‘‘it should represent as much of the geologic develop-
ment of the earth recorded in the area covered as is practicable.” ~

"Illinois Geol. Surv., Vol. I, p. 79, 1866.

696 CHARLES Ri KE VES.

In considering every unit that is a possible basis for carto-
graphic representation, a number of conditions have to be fully
satisfied, in order that the best results may be obtained. As
nearly as possible the unit adopted should be an abstract one,
since schemes which have been elaborated, or may be in the
future proposed, may not have different factors or different
kind of factors to appose when the new facts are compared.
The unit should be practically adaptable in order that knowledge
once acquired may not have to be worked over anew in the
field with each change of ideas necessitated by the constantly
increasing use of more and more refined methods. The unit
should be elastic, because too great rigidity of plan often breaks
down the best of schemes. The unit should be easily recogni-
zable and rapidly delimitable in the field; it should be of such
character as to be readily traced from point to point, quickly
run in on the map, and easily followed on the ground by sub-
sequent investigators who may use the map.

It has been asserted that the lithologic map is a return to
the so-called geological map of a century ago. It does not
appear that the facts of the case warrant this statement. The
geological map of today based strictly upon lithologic indi-
viduals is very nearly asfundamentally distinct from the mineral-
ogical map of a hundred years ago, as is the modern map in
which so-called geological formations are depicted. In map-
ping the geological features of an extensive region, work such
as the federal government and some of the state geological
surveys are engaged upon, the lithological individual for carto-
graphic representation necessarily takes precedence over all other
features. It will be along time after the geological map based upon
lithology principally is ready to be issued, that the perfected map
of ideal geological formations can be made. In the majority of
cases the delimitation of the latter must always rest very largely
on the lithologic characters. A map of units recognizable in
the field only after about as much study as was devoted to the
terranes in the first place by the expert stratigrapher is of small

? practical use.
For a long time yet in modern areal work, the lithologic indi-

REPRESENTATION OF GEOLOGICAL FORMATIONS 697

vidual, delimited if necessary by the aid of other criteria,
appears amply comprehensive and exact for all purposes to
which the ordinary geological map is put. As an aid in the
development of the mineral resources of the country—the
primary object of work of this kind—maps in which the litho-
logic individual is the unit amply suffice. In practice in the
field, the units broadly defined by the lithologic characters, and
those indicated by the more philosophical geologic formations,
are generally near enough alike to enable future investigators to
do their work without hindrance or uncertainty.

The suggestion of a faunal map eventually following the
lithologic map as an integral part of a complete geologic
atlas appears somewhat infelicitous. There is certainly no
room whatever for such a dual planin mapping. Such ascheme
merely leads to others, maps based upon every criterion known
or which may be devised. This is a proposition for which there
is not the slightest demand. It is beyond all probability that
parallel subdivisions should ever be found that are based upon
radically different criteria. When we consider a dual scheme
with a structural phase and a time phase based entirely upon
fossils we are considering incongruous things. And there is
not necessarily any logical connection.

In Europe, there is a classification generally presented that
is dual in character, though with singular nomenclature. Thus,
all the subdivisions of terranes and of time are strictly paralleled.
The International Geological Congresses have adopted the same
plan. It must be quite evident to the practical field geologist
that there are very serious objections to this scheme. The
critical criteria in the rock-scheme and in the time-scheme are
fundamentally distinct. In fact, they have no genetic relation-
ships whatever.

In reality, we have more than a dual scheme of classification
in geology. There is a triple scheme, a quadruple scheme, and
schemes multiple according to the number of standards involved.
Each standard gives rise to a different scheme.

The principle underlying the classification of natural phe-
nomena is that different kinds of criteria give rise to different

698 CHARLES TR. KEVES

taxonomic groups. The arrangement of rock-masses affords no
exception to the rule. Ifthe life phases be used as the pre-
dominant feature in delimiting the subdivisions of one order,
prevailing lithologic character may be given greatest weight in
another; physiography or specific biotic aspect in a third.
Matters are greatly simplified by regarding the larger subdivis-
ions of the geologic scale as essentially arbitrary, abstract time
divisions, in which lithology has no place. With the smaller
subdivisions, which are best considered as essentially structural
divisions, the time element may be practically neglected.

While we have to allude to the time-interval, during which
every rock-mass was formed, we have only universal structural
units represented in the major two of the five taxonomic cate-
gories usually recognized. On the map, the expression of the
time of formation of the terrane is by a distinctive color. The
smaller of the universal time-units is thus represented. In like
manner, only one of the rock categories becomes important on
the map, and this is commonly represented by a standard pat-
tern. This, too, is the smallest unit that is of broad geologic
significance. While this plan does not always meet every case,
it only needs slight adjustment from time to time in order to make
it the most serviceable, the most practical, the most elastic, and
most nearly in accord with local facts, of any scheme yet devised.

After all, the cartographic unit, like the species in zoology
or botany, is necessarily a matter of convenience. In the exact
delimitation of both, there enters very largely an element of
personal judgment. Through the consensus of opinion we finally
arrive at a tolerably good idea of what each unit should be.

It is commonly recognized that the principal criteria followed
in delimiting the several taxonomic orders of units and in geo-
logical classification are (1) the relative progress of life in
general as compared with that now existing; (2) the prevailing
biotic type; (3) the general lithologic phase; (4) the specific
lithologic character; and (5) the specific fossil feature.

Reference is here made to our most approved ideal classi-
fication because the lithologic individual fits closely into this
scheme as the taxonomic subdivision of the fourth order. The

REPRESENTATION OF GEOLOGICAL FORMATIONS 699

case of the St. Clair limestones of Arkansas, already referred to,
and the formations of the Rico mountains, mentioned by Mr.
Cross, are good examples in which strict lithologic separation
was not obvious at first glance. In reality, Cross and Spencer’s
“Hermosa,” ‘Rico,’ and ‘‘Dolores’’ formations approach
definite geological formations only approximately. They repre-
sent no nearer the real geological formation than do purely litho-
logic individuals. Exact faunal studies in the Rico and neigh-
boring districts are likely to require the divisional lines to be
drawn at quite different horizons. When the fuller geological
history of the region shall have been made out further rectifica-
tion will doubtless be found necessary.

As a matter of fact, the lithologic individual based primarily
upon lithology and secondarily upon fossils contained, and the

)

“geological division’ based upon the fossil characters have
essentially the same kind of values as geological elements. But
Mr. Cross, in his delimitations, does not depend entirely upon
the faunal features, for he goes on to reénforce his statements
by giving other reasons for drawing his lines where he does.
Lithologic features are manifestly among the most important
criteria in tracing the ‘‘formations.”’ In the same way it is quite
apparent that the advocates of strictly lhthologic individuals in
mapping do not and cannot depend wholly upon a uniform
rock character.

This phase of the question leads to the statement of a more
general one, that the main thing is to give clearer definition,
than has usually been done, of each cartographic unit proposed.
The unit should be defined according to (1) geographic distri-
bution, (2) topographic expression, (3) lithologic nature, (4)
stratigraphic delimitation, (5) biotic definition, and (6) mineral
content.

When this shall have been done the foundation will have
been laid for the establishment of real Geological Formations
expressive of the geological history of the area mapped. The
approximate ‘‘lithologic”’ map will not have to be materially
changed, but only accompanied by a few words of additional
explanation. CHARLES R. KEYEs.

THE MISNAMED INDIANA ANTICLINE.?

Recent work by the United States geological survey in
western Pennsylvania has revealed a number of unsuspected facts
of geologic structure. The results are being published in folios
in which the lay of the rocks is shown by deformation contours,
but it is thought desirable to call attention here to the finding
of a syncline where formerly there was considered to be an
anticline.

The map of Indiana county issued by the second geological
survey of Pennsylvania shows the Indiana anticline to extend in
a straight line through the town of Indiana. This supposed
fold has been thought to be continuous on the southwest with
the Fayette anticline in Westmoreland county and on the north-
east with the anticline which is well marked near Richmond, on
Little Mahoning Creek. The name Indiana anticline, therefore,
has been applied to the entire fold. This term has passed into
geologic literature and is still being used.

In the area adjacent to the type locality of the fold, how-
ever, the structure, as indicated by the accompanying sketch
map, is quite different than. previously interpreted. The Rich-
mond and Fayette anticlines are not continuous, but the former
pitches southwestward and the latter pitches northeastward, and
the area between the Conemaugh River and Crooked Creek, along
the extension of the axes of these folds is occupied by the
Latrobe syncline. It is an odd coincidence that the axes of the
Richmond and Fayette anticlines fall in line with each other,
and it is not surprising that these folds have been thought to be
continuous, for in the intervening region surface exposures are
poor and the structure can be deciphered only by detailed work.
The present determination is fully proved by the records of
some fifty diamond drill holes lately put down by the Rochester
and Pittsburg Coal and Iron Company.

Structural details will be published in the forthcoming Indiana

* Published by permission of the director of the United States geological survey.

700

THE MISNAMED INDIANA ANTICLINE 7OI

SKETCH MAP
OF PARTS OF
INDIANA & WESTMORELAND
COUNTIES
PENNSYLVANIA

f altsburg,°" >
ey

20 7S 4a Seo MINEES

SCALE

702 GEORGE B. RICHARDSON

and Latrobe folios, so that only a few words need be added in
explanation of the map. West of the Chestnut Ridge anticline
the Latrobe syncline forms the northern extension of the Con-
nellsville basin. This syncline rises and flattens out between
Blairsville and Indiana, displacing the two westward succeeding
folds. These folds are the Fayette anticline and the Greensburg
syncline. Well developed where they cross Loyal Hanna Creek,
northward in the vicinity of the Conemaugh River they fade
away and merge into the western flank of the Latrobe syncline.
The next fold to the west is the Jacksonville anticline, which has
its maximum development near the town of Jacksonville. South-
west of Indiana there is an offset in the axis of this fold. Thence
the arch continues northeastward, as the McKee Run anticline,
and forms a low fold separated from the Chestnut Ridge anti-
cline by the extension of the Latrobe syncline. Northeast of
Indiana this syncline is divided in two by a south-plunging anti-
cline, which, passing between Decker’s Point and Marion Center,
is well marked near the town of Richmond.

GEORGE B. RICHARDSON.
WASHINGTON, D.C.

REVISE DCU ASSIMCATION OF THE UPPER PAE O@=
ZOIC FORMATIONS OF KANSAS.

INTRODUCTION.
CLASSIFICATION,
Wabaunsee stage.

Burlingame limestone.
shales.
Emporia limestone.

formation.

Americus limestone.
Elmdale formation.
Neva limestone.
Eskridge shales.
Council Grove stage.
Alma limestone.
Garrison formation.

Chase stage.

Wreford limestone.
Matfield shales.
Florence flint.

Fort Riley limestone.
Doyle shales.
Winfield formation.

CONTENTS.

Sumner stage.

Marion formation.
Wellington shales.

Table of the Upper Paleozoic forma-
tions of Kansas.
Correlation of the Cimarron series.

THE PERMIAN QUESTION.

Correlation of the Upper Paleozoic
of Kansas with the Russian
Permian.

Opinions of various geologists.
Correlation of the Kansas and
Texas beds.

Provisional correlation of the
Kansas formations.

Conclusions of Dr. Keyes.

Conclusions of Dr. Frech.

Usage of Russian geologists.

Opinions of other European
geologists.

INTRODUCTION.

In 1895 the writer published a paper on ‘The Classification
of the Upper Paleozoic Rocks of Central Kansas”’ in the Jour-
NAL OF GEoLoGy.* Additional field work and study of the Cot-
tonwood Falls quadrangle render it advisable, in compliance
with the custom of the United States Geological Survey to
designate each lithologic individual capable of representation on
the topographic map as a formation, to subdivide three of the
units which were described as formations in that article.

A few changes in classification or nomenclature have been

made which are also explained in this paper. Dr. J. W. Beede,

* Vol. III, pp. 682-706, and pp. 764-801.

793

704 CHARLES: S. PROSSER

of Indiana University, who has been associated with the writer
in this later study, has rendered most efficient service in the field
and other work necessary for this revision. Dr. George I.
Adams, of the United States Geological Survey, spent several
days with Dr. Beede in examining part of the area of the Cot-
tonwood Falls quadrangle, and Mr. F. B. Weeks has also kindly
furnished the author with references to the descriptions of forma-
tions from the United States Survey card catalogue of geologic
formation names.
CLASSIFICATION.

WABAUNSEE STAGE.’

None but the upper rocks of this stage are exposed on the
Cottonwood Falls quadrangle and the lower ones, exposed to the
eastward, have not been carefully examined by the writer.

Burlingame limestone—At the base of the Wabaunsee is a
conspicuous and persistent limestone from seven to twelve feet
in thickness, the lower limit of which is regarded as the lower
line of that stage. It was named and briefly described by Hall
in 1896? from outcrops near Burlingame and since then it has
been traced from Nebraska across the state to Oklahoma. 3

This limestone is frequently composed of two layers, gray to
brown in color, separated by shale, and forms a massive ledge.
This is apparently the division which was termed “limestone
number g”’ by Professors Haworth and Kirk, in 1894, exposed
near the junction of the Cottonwood and Neosho rivers, which
they stated ‘‘may be called the Wyckoff lmestone .... on
account of its great exposure in the vicinity of Wyckoff.t”” This
name, however, ought to be considered a synonym, for Dr, Sar-
deson had already given an almost identical one to a division of

* The word stage is used in the sense adopted by the International Congress of
Geologists. See Work Inter. Cong. Geologists, 1886, p. 50; GILBERT, in Proc. A. A.
A. S., Vol. XXXVI, 1888, p. 186; Congrés Géologique International (8° Session),
Procés-verbaux des Séances, 1901, p. 35; and zbid., Comptes Rendus, 1% Fasc., 1901,
p. 196.

2 Untv. Geol. Surv. Kansas, Vol. I, p. 105.

3 See, ‘Map of Limestone Outcroppings,” by PRoFEssoR HawormTH, Vol. III,
Univ. Geol. Surv. Kan., 1898, Pl. VII.

4 Kan, Univ. Quart., Vol. I, Jan. 1894, p. III.

Cimarron series
Big Blue series
Missourian Sel

1 (Upper part of

*(Ajuo yred

(¢) ‘wayshs ueruieg saddn) wiajysks snosayruc

e

mae ay)

v a
ar

nt

Mea

aye

ANA

CLASSIFICATION OF THE UPPER PALEOZOIC FORMATIONS OF KANSAS.

—-\\

( Kiger Stage * -

| |
Cimarron series? - Cragin, ’96. |
Salt Fork Stage -

Permian system. (?)

cn

|
|
|
|

*

Big Blue series3 - Cragin, ’96. 4
Chase Stage - -

Council Grove Stage.

Missourian Series - Keyes, '96.

(Upper part of series) 4

Wabaunsee Stage -

Carboniferous system (upper
part only)

U L

1The classification of the formations from the top of the Kiger to the Wellington
shales. inclusive, is that of Dr, Cragin, except that he termed Kiger and Salt Fork d7vzstons
of the Cimarron series (Cod. Coll. S/udies, Vol. VI, March 27, 1896, pp. 3, 16-49). The
following year Dr, Cragin revised the classification of the Cimarron series. changing the
limits and names of some of the formations and drawing the line of separation between the
Salt Fork and Kiger divisions at the of of the Dog Creek formation instead of at its base,
as in the former classification (A. Geol., Vol. XIX, May, 1897, PP. 351-64).

2Col. Coll. Studies, Vol. VI, pp. 3, 18, 48; and see additional account in Am, Geol.
Vol, XIX, May, 1897. pp. 351-64.

3Col. Coll, Studies, Vol. V1, March 27, 1896, pp. 35 5,6. In July, 1896, Dr. Keyes
proposed ‘‘ to recognize in the ‘upper > Carboniferous of the Western Interior province three
series having equal taxonomic rank,’’ the upper one of which was named the ‘‘ Oklahoman”’
(Am. Geol., Vol. XVIII, p. 25). In defining the series it was stated that “In suggesting
the name ‘Oklahoman’ as a serial geological term it is intended to apply to all those rocks
of Carboniferous age which occur north of the Canadian river in Oklahoma, and which lie
interval of the top of the Missourian series and the base of the Cretaceous. It may
be regarded as essentially covering the same succession of strata that has long been vaguely
known under the title of ‘Permian,’ The name is derived from the territory in which the
formation has its best development and in which the most complete sequence is represented ”?
(zbrd.. p. 27). In October, 1891, Dr. Keyes recognized the Cimarron series and gave the
Oklahoman and Cimarron as the two closing series of the Carboniferous (A7. Jour, Sct.,
4th ser.. Vol. XTI, pp. 306, 309), Stating that ‘* The so-called Permian of the Western Interior
basin (Oklahoman and Cimarron, the latter generally known as the Red Beds) is composed
largely of shales and shaly sandstones. ... . The conditions existing were identical with
those under which the original Permian beds of Russia were formed’’ (77d... p. 309). The
following month Dr. Keyes published a ‘General Geological Section of the Carboniferous of
the Mississippi Valley,” in which a complete list of the series and terranes of the system is
For the portion under consideration it is as follows:

SERIES.

between the

given.
TERRANES.
¢ Kiger shales.

Cimarron “2 Salt Fork shales.

Wellington shales,
Marion limestone.
Chase limestone.
Neosho shales.

Carboniferous system (upper
portion).

Oklahoman, ..---+..+++++5*

Cottonwood liinestone.

Missourian (upper part).--+- Atchison shales.

Cragin, ’96. + Day Creek dolomite - - 2 3 - = s

Cragin, ’96. +

( Summer Stage - Cragin, 96. |

Prosser, ’95.

Prosser, ’95.

Taloga formation- - - - - - - Cragin, ’97

Cragin, ’96.

( Red Bluff formation - - - - - Cragin

Cragin, '97.
Amphitheatre dolomite,

{ Chapman dolomite,5
Cragin, 97.

Shimer gypsum
j Jenkins clay -

- Cragin, 96.
Cragin, 96.

| Dog Creek formation, Cragin, 96.
j Medicine Lodge gypsum,

Cave Creek formation, Cragin, '96.
Cragin, ’96.

Cragin, ’96.

Cragin, ’97. © Cedar Hills sandstone, Cragin, ’96.

Salt Plain member - Cragin, ’96.

Harper sandstone ® - Cragin, '96.

4
| i

L
Glass Mountain formation, } Flower-pot shales -
Kingfisher formation, Cragin, ’97. |

Wellington shales? - - - - - Cragin, 96.

Marion formation - - - - - - - - Prosser, ’95-

Winfield formation = 2 = = = = -.
Doyle shales - : = - = =

Prosser, '97.
Prosser and Beede.
Swallow, 66.

Prosser, '95.
Prosser and Beede.
Hay, ’93.

Fort Riley limestone - = - - - - =

Florence flint - - - : = - : t S
Matfield shales - = E : 2 = j
Wreford limestone - - - -

Garrison formation, | Neosho member - Prosser, '95.
Prosser and Beede. ( Florena shales, Prosser and Beede.
Alma limestone - - - - - - -

Prosser, ’94.

’ Eskridge shales - - - - - - Prosser and Beede.

Neva limestone - - - - - Prosser and Beede.
Elmdale formation - - - - - - Prosser and Beede.
J Americus limestone - - - - - - - Kirk, ’96.
\

formation - - - - - -

Emporia limestone - - - - - - - - Kirk, ’96.

shales - - - - = > E
(Burlingame limestone - - - - - - Hall, ’96.
(Am. Geol. Vol. XXVIII, p. 302.) It will be seen that the Oklahoman series, as precisely
defined above is identical with the Big Blue series proposed by Dr, Cragin in 1896, and
therefore his name. which has priority, is adopted for this classification.
only ch: nge from Professor Cragin's classification of the Big Blue series is that the Neosho
member at its base is put in the Missourian series, ‘he Rrazos series proposed by Prof, Hill
in 1902 (/wenty-first Ann. Rept. U. S. Geol. Surv. Pt, VII, p x00) for the Red Beds is
apparentl. the same as the Cimarron series of Cragin.

In my revision the

4°* Missourian”? was proposed as the name of a series by Dr. Keyes, in July, 1896
(Am. Geol., Vol. XVI11. pp. 25-27). This was an outgrowth of the Missouri stage or for-
mation which was proposed by him in 1893 (/owa Geol. Surv., Vol. 1, pp. 85, 114; and
Mon. Rev. lowa Weather Service, Vol. lV. p. 3).

s Nearly three years later Professor H. S. Williams gave the name ‘* Chapman sand-
stone’ to a formation occurring in northeastern Maine (Am, Jour. Ser., 4th ser., March,
1900, Vol, IX. pp. 203, 205, which was more fully described in Budi, U. S. Geol, Surv., No.
165, 1900, p. 78).

6It is not probable that ‘* Harper sandstone’’ can be retained for this subformation,
because the very similar term of ** Harper’s shale’’ was published by Keith two years earlier
as the name of a formation occurring near Harper's Ferry (Geologie Atlas United States,
Harpers Ferry /olio (Folio 10), 1894, pp. 3. 5)»

7 The term ‘Wellington shales.’ apparently applied to this formation. first appeared in
an article by Professor Cragin, in the Kansas City Reusew o/ Science and Industry, Vol.
VIII, April, 1895. p- 679; anda little later he published it in the Budd. Washburn College
Laboratory of Natural History, Vol. I, May (?), 1885, p. 86.

An
Iie

nf i \ yd AYsneS v
AGN OANA TU MANY ete i rai hint aap cucam ; Me aE EEA

ey

a

are
intel

tH
Piling ary
Nari

UPPER PALEOZOIC FORMA TIONS OF KANSAS 705

the Cincinnati series of the Ordovician in Minnesota, for which
he proposed “‘a new name .. . . Wykoff beds — from the town
near which the best exposure known occurs.’’* and this he later
called the Wykoff formation.? The only difference is that the
name of the Minnesota town is spelled without ac. In 1895
Professor Haworth proposed the name ‘Osage City shales”’ for
the rocks included between the top of the Topeka limestone and
the base of a thin limestone overlying the Osage coal; while
about 150 feet of the superjacent rocks in the vicinity of Bur-
lingame were named the Burlingame shales.3 Later in the same
year both divisions were more fully described by Professor
Haworth ;* but the upper limit of the Burlingame shale was not
precisely defined. The following year Mr. Hall applied the
term ‘‘ Burlingame limestone ” to eight feet of limestone which
‘covers the third and last heavy bed of shales in this section”’
with a thickness of 150 or 200 feet. 5

This shale was apparently regarded by Mr. Hall as the Bur-
lingame, since he used that name in the list of subjects at the
beginning of his chapter, ° and then the heading following that
of the Burlingame limestone is the ‘Systems above the Bur-
lingame shales.” 7

In 1898 Professor Haworth stated that ‘‘ subsequent work
has shown the unimportance of” the thin limestone overlying
the Osage coal, “so that it will not do to depend upon it as a
division line marker. Neither will the Osage coal serve such a
PUEPOSE,-as) it is by<no, means ‘continuous ..... Fromithese
considerations it seems desirable to let the name Osage apply to
the entire shale bed above the Topeka limestone and below the
Burlingame limestone . . . This renders the name Burlingame
shales superfluous and therefore it will be dropped.’’®

* Bull. Minn. Acad. Nat. Sci., Vol. III, No. 3, 1891 (?), p. 326. It is stated by
Professor N. H. Winchell and E. O. Ulrich that this paper was not distributed until
April 9, 1892 (Geol. Minn., Vol. ILI, Pt. I, of the final report, 1895, p. XLVI).

2 Am. Geol., Vol. XIX, Jan., 1897, p. 24; see also zbzd., May, 1897, pp. 332, 334.
3 Kan. Univ. Quart., Vol. IL, April, 1895, p. 278. )

4 Am. Jour. Sct., 3d ser., Vol. L, December, 1895, pp. 461, 462.

5 Univ. Geol. Surv. Kan., Vol. I, 1896, p. 105. 6 Joid., p. 99. 7 Lbid., p. 105.
8 Jéid., Vol. III, p.. 105; also see p. 73.

706 CHARLES Ss) PROSSER

b

In 1898 Dr. Adams applied the name “ Eureka limestone’
to a formation exposed in the vicinity of that city,t which Pro-
fessor Haworth correlated with the Burlingame limestone.? The
name Eureka, however, was preoccupied, since Hague used it
in 1883 for the Eureka quartzite of Nevada, and later Dr.
Branner named the Eureka shale of Arkansas.

shales.—In Lyon county, succeeding the Burlingame
limestone, according to Mr. Alva J. Smith, are nearly forty-five
feet of blue to yellow shales and friable limestones, the latter com-
prising about eight feet of the total thickness of this formation,
which is limited at the top by the Emporia blue limestone.$
Dr. Adams has recognized as a formation the shales included
between the Burlingame and Emporia limestones, which he has
named in manuscript, and therefore no name is proposed for the
division in this article.

Emporia limestone.— Yhis division, as described by Mr. Smith,
is composed of three feet of hard blue limestone at the base, the
upper six-inch layer making a good flagstone, which is exten-
sively used in Emporia. Then there is four feet of shale capped
by another hard blue limestone two feet in thickness. These
limestones ‘‘pass under the Cottonwood River at Soden’s mill,
one mile south of Emporia,’® and the Neosho River at the
Rinker bridge. It was named by Kirk in 1896,’ but, according
to Mr. Smith, at some of the localities which he mentioned it
was confused with a higher limestone. The blue Emporia lme-
stone was correctly reported by Kirk in the ‘‘Chicago Mound”
near Wyckoff; but the limestones near the ‘“‘ Emporia water-
works’’ and ‘along the hilltops about four miles south and one
mile east of Emporia’’® are higher and belong in what Mr.
Smith named the ‘‘Emporia system.’”9 The lower limestone

1 [bid.,p. 67. . 2lbid., p. 73: 3 Third Ann. Rept. U.S. Geol. Surv., pp..253, 262:
4Ann. Rept. Geol. Surv. Ark., for 1888, Vol. IV, 1891, p. 13, and see description
by Professor Simonds on p. 26.

5See A Bulletin on Lyon County Geol., 1902, pp. 2,10; and Zrans. Kan. Acad,
Science, Vol. XVII, 1901, p. 193.

Bull. Lyon County Geol., pp. 2,10; and Kan. Acad. Scz., Vol. XVII, p. 193.

7 Univ. Geol. Surv. Kan., Vol. I, p. 80. 8 Toid., p. 82.

9 Letter of Mr. Smith, Jan. 20,1902. For a description of the Emporia system
see Bull. Lyon County Geol., p. 3.

UPPER PALEOZOIC FORMATIONS OF KANSAS 707

was called the ‘“‘ Emporia blue” by Mr. Smith,’ who states that
it is much more uniform than the upper, and that he has traced
it across Lyon county. He also identified it two miles east of
Harveyville,in Wabaunsee county, and four miles northwest of
Eureka, in Greenwood county. It appears to be a persistent
limestone for some considerable distance, which can be traced
and mapped, and is, therefore, probably entitled to rank as a sub-
stage of the Wabaunsee.

formation — Above the Emporia limestone, as
described by Mr. Smith, is a zone nearly seventy feet thick, com-
posed mainly of shale, but with a foot of limestone and a five-
inch stratum of coal in the lower part, and a five-foot sandstone
near the top. Then comes a zone composed of five limestone
strata from one to two feet in thickness, separated by shales
from four to ten feet thick, and with a total thickness of twenty-
four feet. Mr. Smith named this zone the Emporia system,’
and has represented its distribution three-fourths of the distance
across Lyon county.3 He writes me, however, that ‘the five
limestones which I have included in the Emporia system are not
very persistent, and the character and thickness of the stone, as
well as the intervening shale, are subject to sudden changes. I
have been unable to identify it beyond the lines of this county.” 4
It hardly appears desirable to regard this zone as entitled to the
rank of a formation. Then, according to the measurements of
Mr. Smith, there are 210 feet of rocks composed largely of sandy
shales, but also containing some thin beds of limestone, coal, and
sandstone. The most important coal stratum is ten inches in
thickness, and it occurs 75 % feet above the base of this mem-
ber. So far as I am able to judge, it would appear advisable to
put these three members in one formation, giving it a thickness
of about three hundred feet. Dr. Adams ranked the rocks
included between the Emporia and Americus limestones as a
formation, which he has named in manuscript, and, therefore, no
name is proposed for this formation.

OCA. pi 2 Bull. Lyon County Geol., 1902, p. 3.

3 Trans. Kan. Acad. Sct., Vol. XVII, 1901, p. 191; and Bull. Lyon County Geol.,
p. 8.

4 Letter of January 20, 1902.

708 CHARLES (S, PROSSER

Americus limestone.— This name was given by Kirk in 1896
to two thin layers of limestone, separated by shale, which are
quarried near Americus. It was also noted by Haworth and
Kirk in their ‘“‘ Neosho river section”’ and called “limestone sys-
tem No. 11.”? The lower stratum is buff in color, very solid
and compact, making a good building stone with a thickness of
twenty-one inches, according to Smith. Then comes six feet of
shale with a six-inch flag limestone on top,3 making a total
thickness of over eight feet, and its distribution has been
mapped entirely across Lyon county by Smith. On the Cot-
tonwood Falls quadrangle it is probable that this limestone is
only a few feet below water in the Cottonwood River east of
Elmdale.

Elmdale formation.— This formation, the succeeding ten and
the lower and middle parts of the Marion, are represented on
the Cottonwood Falls quadrangle, which has also furnished the
majority of the names, and consequently these have been more
thoroughly studied by the writer than the preceding formations.
It is about 130 feet in thickness, and composed of yellowish to
bluish shales, with thin beds of grayish alternating limestone,
including two or three thicker ones. About thirty feet above
the base of the formation is a friable limestone with a thickness
in some localities of four feet, which is composed to a large
extent of the tests of Musulina secalica Say. This stratum
weathers readily and leaves great numbers of Fusulina in the
soil. About thirty-five feet higher is another conspicuous yel-
lowish limestone, the center of which weathers to a rough face,
and from ten to fifteen feet below the top is a limestone stratum
from three to five feet in thickness. The formation is limited at
the base by the top of the Americus limestone, and at its top by
the base of the massive Neva limestone. It is well exposed on
the bluff east of Elmdale, from which town it is named.

* Univ. Geol. Surv. Kan., Vol. I, pp. 80, 81.

2 Kan. Univ. Quart., Vol. Il, January, 1894, p. 111.

3 Bull. Lyon County Geol., pp. 3, 10.

4 Trans. Kan. Acad. Sci., Vol. XVII, 1891, p. 191; and Bull. Lyon County Geol.,
p. 8.

UPPERVUPALEOZOTC FORMATIONS OF ‘KANSAS 709

Neva limestone.— This formation consists of a massive bluish-
gray limestone or of a lower and upper massive limestone, each
one a little over four feet in thickness, separated by two feet of
shales, with a total thickness of about ten feet. The limestone,
forming frequent ledges seven feet or more in thickness, breaks
off in large blocks with sharp angles and arough, jagged surface,
weathering to a color not dissimilar to that of bleached bones.
It was noted by Swallow in 1866* and represents the upper
stratum of Haworth and Kirk’s ‘‘limestone system No. 12,’’? to
which “system” Kirk later apparently applied the name “ Dunlap
limestone.”3 The limestone is finely exposed in the anticlinal
fold to the northeast of Neva, a station on the Atchison, Topeka
& Santa Fe railroad near the junction of the Diamond Creek
and Cottonwood River valleys, hence its name.

Eskridge shales— Between the Neva and the next higher
massive limestone is a mass of shales, with perhaps some thin
limestone layers, varying from thirty to forty feet in thickness.
The shales are of greenish, chocolate, and yellowish color, and
usually form covered slopes between the two conspicuous limiting
limestones. They form the upper division of the Wabaunsee
stage, and are named from the exposures in the vicinity of
Eskridge, Wabaunsee county.

COUNCIL GROVE STAGE.

In my original description of these formations the line of
separation between the Upper Coal-measures and Permian was
doubtfulty drawn between the Cottonwood and Neosho forma-
tions ;* while the Permian appeared as a series of the Carbon-
iferous, in accordance with the usage of the United States Geol-
ogical Survey.’ Since then Dr. Frech has reviewed this classifi-
cation and drawn the lower line of the lower Dyas (Permian)
at the base of the Chase stage, while it is stated that the Neosho
is a transition to the Carboniferous, and a distinct line fails.°

* Prelim. Rept. Geol. Surv. Kan., p. 16, Nos. 82-4.

2 Kan. Untv. Quart., Vol. 11, 1894, p. 112.

3 Univ. Geol. Surv, Kan., Vol. I, 1896, p. 81.

4Jour. GEOL., Vol. III, 1895, p. 800.

SISCCe/OZd..1 ps 7O0y tne 6 Lethaea palaeozoica, Bd. Il, 2 Lief., 1899, p. 378.

710 CHARLES 'S: PROSSER

On the chart showing the partial distribution of the Carbonif-
erous, the Chase is apparently given as corresponding to the
lower part of the Arta stage, which, according to his classifica-
tion, is the oldest stage of the Permian, and the Neosho is
represented as about on the dividing line between the Permian
and Upper Carboniferous, although perhaps it is intended to
include all of it in the latter division. It is stated, however,
that the older Chase and Neosho strata (as compared with the
Marion) beyond doubt can only correspond to the Artinsk stage
(which is the oldest Permian of Russia=Arta stage), and that a
sharper division is made impossible by the absence of Cephalo-
pods.? It is to be noted that in the latter part of this work
devoted to ‘‘ Die Dyas,” in the table of Kansas Dyassic forma-
tions, Dr. Frech has left the Neosho as the oldest division of the
Paleo-Dyas. In this case, however, he was reporting the
classification of Professor Cragin, for it is stated that in the
following table the Kansas strata are enumerated according to a
recent survey, and it is not thought that he intended to set aside
his earlier correlation.3

In reference to the correlation of these divisions the writer has
stated that ‘‘The appearance and the prominence of the Pseudo-
monotis fauna in the Neosho formation furnishes a strong reason
on the biologic side”’ for its correlation with the Permian. The
presence of Pseudomonotis is no longer an important argument in
favor of putting the Neosho in the Permian, because since then
Dr. Beede has identified Pseudomonotis Hawni, and described a new
variety of that species and two new species from the older forma-
tions of the Upper Coal-measures of Kansas.5

There is a marked lithologic change at the base of the Chase
stage where it begins with the Wreford limestone, which is the
lowest one of the very cherty massive limestones. It is perhaps

' [bid., “Tab. XXIV, Einige wichtige Vorkommen des Carbon.”

2/bid., p. 377, f.n.: ‘Dass die tieferen Chase -und Neosho-Schichten nur der
Artinskischen Stufe entsprechen konnen, steht ausser Zweifel; jedoch wird eine
scharfere Abgrenzung durch das Fehlen der Cephalopoden unmoglich gemacht.”

3 Jérd., 3 Lief., 1901, p. 514. 4Jour. GEOL., Vol. IIL, 1895, p. 796.

5 Kan. Univ. Quart., Vol. VILI, April, 1899, Ser. A., pp. 79-84; and Unzu. Geo/.
Surv., Vol. VI, 1900, pp. 132-35.

UPPER PALEOZOIC FORMATIONS OF KANSAS Jl

a more satisfactory classification to regard the base of the
Permian as marked by the lower limit of the Wreford limestone
and the writer is inclined to accept this line for the division as
indicated by Dr. Frech. If this be done the writer would class
the two formations succeeding the Eskridge shales (Cottonwood
limestone and Garrison) together to form a stage for which he
would propose the name of Council Grove. The upper part
of the stage is well shown in the bluffs of the Neosho River
and its tributaries in the immediate vicinity of this city, while
the Cottonwood limestone and the overlying Florena shales may
be found in the Neosho valley, about six miles below Council
Grove.

Alma limestone —This is a massive light gray to buff-colored,
foraminiferal limestone, frequently composed of two layers with
a thickness of about six feet. It contains very few fossils, with
the exception of Fusulina secalica Say, which is extremely
abundant in its upper part, and is called ‘wild rice” by the
quarrymen. It is the most important dimension stone in Kansas,
and at various localities are extensive quarries. Its constant
lithologic character, with its line of outcrop frequently marked
by a row of massive light gray rectangular blocks filled with
Fusulina, make it one of the most important stratigraphic
horizons in the Upper Paleozoic rocks for at least two-thirds
of the distance across Kansas and into Nebraska. Swallow
called the stratum the Fusulina limestone,* and for years it has
been known commercially as the Cottonwood or Cottonwood
Falls limestone, and at other localities as the Alma and Man-
hattan limestone. Haworth and Kirk, in their ‘‘ Neosho River
section,” called it ‘‘limestone system No. 13, which is consid-
ered the equivalent of the famous Cottonwood Falls limestone ;’’?
but in their description of the quarries near Cottonwood Falls,
under their ‘‘Cottonwood River section,’ simply called it
‘No. 13,3 and did not apply to it the term Cottonwood Falls
limestone. The same year Prosser proposed the name ‘ Cotton-
wood formation”’ for the limestone and superjacent fossiliferous

"Prelim. Rept. Geol. Surv. Kan., 1866, p. 16.

2 Kan. Univ. Quart., Vol. Ul, Jan. 1894, p. 112. 3 [bia., p. 113.

712 CHARLES S. PROSSER

shales, on account of the excellent outcrops of both in the
bluffs bordering the Cottonwood River below and above Cotton-
wood Falls and Strong. The lower division was named the
‘Cottonwood limestone’ and the upper the ‘Cottonwood
shales,’* while the same limestone quarried in the vicinity of
Alma was mentioned locally as the ‘‘Alma massive limestone.” ?
The following year the formation and its two members were
more fully described by Prosser The name ‘‘Cottonwood,”’
however, was apparently used for a geological division by N. F.
Drake as early as September, 1893, when he described the
‘‘Cottonwood Creek bed’’ of the Texas Carboniferous.+ It is
now proposed to limit the formation to the Cottonwood lime-
stone and on account of the prior use of the name Cottonwood
for the Texan bed, to call it the Alma limestone from the out-
crops near the town of that name in Wabaunsee county, while
the Cottonwood shales are referred to the succeeding formation.

Garrison formation.—This formation is composed of two mem-
bers, the yellowish fossiliferous shales at the base, formerly
called the Cottonwood shales, and the upper one, composed of
the alternating gray limestones and various colored shales called
the “Neosho, with a total thickness ‘of from /a40ito, 145 feet:
The lower shales have a thickness of thirteen feet near Strong,
but decrease to two or three feet in the northern part of the
state. The lower part of these shales contains immense num-_
bers of a few species of fossils and on this account may be
readily identified wherever outcrops occur. Since the geographic
name ‘‘Cottonwood”’ is preoccupied the term ‘Cottonwood
shale’’ is abandoned, and they are renamed the Florena shales
from the exposures over the Alma limestone in the quarries
near Floren, in the Big Blue valley.

The upper member of the formation is composed of green,
chocolate, and yellowish shales alternating with grayish lime-
stones, while in the Big Blue valley a bed of gypsum occurs near
the base. Certain layers of the coarser shales and limestones

* Bull. Geol. Soc. Amer., Vol. VI, Nov. 1894, p. 40. 2 [bid., p. 44.
3Jour. GEOL., Vol. III, Oct. 1895, pp. 697-705.
4 Fourth Ann. Rept. Geol. Surv. Texas, pp. 374, 382.

UPPER PALEOZOIC FORMATIONS OF KANSAS FACS)

contain an abundant Lamellibranch fauna, and the entire fauna
is thought to be a mixture of species found in the western Coal-
measures, together with others occurring in the division gener-
ally termed’ the Permian or the Permo-Carboniferous. This
member was originally termed the Neosho formation from the
excellent outcrops in the Neosho valley near Council Grove."
The Florena shales and Neosho member are now united to form
the Garrison formation, so named on account of the good expo-
sures from Garrison south in the Big Blue valley.

CHASE STAGE.

As in the case of the Wabaunsee, it has been found advisable
for mapping to divide this stage, which was formerly called the
Chase formation, into several formations, which are described in
ascending order. These subdivisions of the Wabaunsee and
Chase stages are sufficiently definite lithologic divisions to be
traced across the Cottonwood Falls quadrangle, andsome distance
to the north and south, and therefore can be mapped. It does not
appear to the writer, however, that these divisions are entitled to
the rank of a stage, and he would term them substages. If we
consider the well-known Hamilton division a stage of the New
York Devonian, then it would appear that these divisions corre-
spond more nearly to the Moscow shales, Encrinal limestone,
Ludlowville shales, and other subdivisions of that stage, than to
the entire Hamilton.

Wreford limestone.—Yhis formation is composed of limestone
and chert, or flint as it is popularly termed throughout the Flint
Hills region, and varies in thickness from thirty-five to fifty feet.
In general it is composed of three strata, a cherty limestone
below and above, separated by a heavy limestone nearly free
from ycher.| he yrock as; butt in) color, oftens;weathernic
much lighter, and forms the first conspicuous flint terrace above
the Alma limestone. It is quite extensively quarried and
used for construction stone or crushed for railroad ballast. It
was called the Strong flint in 1895,? but is now known to be
the equivalent of the Wreford limestone, which was named by

tJouR. GEOL., Vol. III, 1895, p. 764. 2Jour. GEOL,, Vol. III, p. 773.

Feel CHARLES S 2 PROSSELR

Professor Hay in 1893 from exposures near Wreford, Geary
county, south of Junction.t The preceding report of the State
Board contained the same table of Professor Hay’s ‘‘ Fort Riley
section,” except that this division was called the ‘Walford lime-
stone,” which was undoubtedly a typographical error for
Wreford.?

Matfield shales—The formation is composed principally of
variously colored shales, with some shaly buff, occasionally
cherty limestones, and a light gray limestone two feet or so in
thickness, which occurs about thirty feet below its top. The
thickness ranges from sixty to seventy feet, and it generally
forms covered slopes between two massive and conspicuous flint
ledges. It is named from Matfield township, Chase county,
where it forms the side of the steep escarpment above the
Wreford limestone.

Florence flint.—This formation is about twenty feet in thick-
ness and consists of very cherty limestone separated by definite
layers of chert, with a band of shaly or white cellular limestone
near the center. It is excellently exposed on the McPherson
branch of the Atchison, Topeka & Santa Fé Railroad, and in
the Jones quarries along that railroad, from one to two miles
northeast of Florence, and on this account in 1895 it was named
the ‘‘ Florence flint.’ 3

Fort Riley limestone — Overlying the Florence flint is a series
of massive buff limestones, changing to thin bedded and shaly
strata in the upper part of the formation, which have a total
thickness of forty feet or more. Near the center of the forma-
tion are generally one or two massive layers, which on the
weathered surface form a conspicuous ledge that may be readily
followed by the eye for miles on the bluffs of the Cottonwood
and Kansas rivers. Swallow in 1866 applied the term “ Fort
Riley limestone” to the massive ledge in the vicinity of Fort
Riley, which he described as ‘‘a buff porous magnesian rock, in
thick beds,’ with a thickness of from eight to ten feet.4 This

tBighth Bien. Rept. Kan. State Board Agri., Part II, p. 104.

2Seventh tbid., 1891, Part Il, p. 94.

3Jour. GEOL., Vol. III, p. 773. 4Prel. Rept. Geol. Surv. Kansas, p. 14.

UPPER PALEOZOIC FORMATIONS OF KANSAS 715

name is now adopted for this formation, but its limits are extended
to include the thinner bedded limestones both below and above
the massive Fort Riley main ledge. The Florence limestone’
is apparently equivalent to the Fort Riley main ledge and the
name is now abandoned.

Doyle shales —This formation is composed of variously col-
ored shales with an occasional thin stratum of soft limestone, and
has a thickness of sixty feet. About twenty feet above the base
is a thin, grayish limestone which often appears on the surface,
and at the top are yellowish shales containing a few fossils.
These shales and the rocks of the overlying formations weather
easily and form gently undulating prairies in sharp contrast with
the rough topography produced by the flint and massive lime-
stones below. The formation is shown at various places in the
Doyle Creek valley to the southwest of Florence, from which
locality it is named the Doyle shales.

Winfield formation.—This has a thickness of about twenty-five
feet, and is composed of a cherty limestone at the base with a
massive concretionary one at the top, the two separated by yel-
lowish shales. This chert and concretionary limestone form the
highest prominent chert ledge in the Kansas Permian, and make
a marked stratigraphic horizon that is of great assistance in
determining the areal geology of eastern central Kansas. The
chert is not so uniform in occurrence as in the Wreford and
Florence flints, and at some localities this horizon is repre-
sented simply by a prominent light gray limestone, nearly free
from chert. The concretions in the upper limestone are quite
persistent through long stretches of outcrop, although occasion-
ally areas are found where they are small and inconspicuous or
absent. As arule, however, they are large, and the stratum may
readily be traced across the country either from its exposure in
bluffs or streams or from the line of loose reddish-brown con-
cretions stretching across the prairie. The irregular worn upper
surface of the concretionary limestone and the appearance of
many of the concretions, as though rolled in the mud on the
sea bottom, indicate a shallowing of the sea at this time, fol-

tJour. GEOL., Vol. III, 1895, p. 773, No. 15 of the Chase section.

7.16 CHARLES S. PROSSER

owed by a subsidence of the sea bottom before the deposition
of the succeeding even thin bedded limestones. This change of
physical condition is indicated in the fauna by the nearly com-
plete disappearance of the brachiopods and the survival of a
fauna composed mainly of Permian lamellibranchs. This for-
mation constitutes the upper division of the Chase stage, and in
the preliminary description of these rocks it was called the
“Marion flint and concretionary limestone’? from the outcrops
below Marion, and regarded as a subformation. The overlying
buff limestone and shales, however, were named the Marion
formation and hence to avoid confusion the name Marion was
dropped for the lower division and it was renamed the Winfield
concretionary limestone from the outcrops in the vicinity of
Winfield, Cowley county, in southern Kansas.?-_ Fourteen months
later Dr. Keyes published the name ‘‘ Winfield limestone,” which
he applied to a Cambrian formation found in the Mississippi val-
ley near Winfield, Lincoln county, Missouri.3 In 1900 Professor
Harris and Mr. Veatch applied the very similar name of Winn-
field limestone (spelled Winn) to a Cretaceous formation of
northern Louisiana.* Clearly the Kansas usage of the name has
priority, and Winfield is adopted as the name of this formation.

SUMNER STAGE.

The two upper formations of the Big Blue series (the Marion
and Wellington shales) were classed together by Professor Cra-
gin to form the Sumner division.6 It was named after Sumner
county in southern Kansas which includes nearly the entire
breadth of its outcrop in that part of the state. This name is
retained for this division, which is considered to have the rank
of a stage.

Marion formation.— Buff thin-bedded limestones and shales
form the principal part of this formation which is the latest
Paleozoic one found on the Cottonwood Falls quadrangle. The

‘Jour. GEOL., Vol. III, 1895, p. 772.

2 Univ. Geol. Surv. Kan., Vol. il, Feb. 15, 1897, p. 64.

3 Proc. Towa Acad. Sci., Vol. V, April 28, 1898, p. 60.

4 Geol. Surv. La., Report for 1899, Sec. II, p. 56.

5 Col. Coll. Studies, Vol. VI, pp. 3, 9, 48.

UPPER PALEOZOIC FORMATIONS OF KANSAS TAGa

lower part is composed of rather soft, porous, thin-bedded lime-
stones and shaly layers to shales, containing near the base a
considerable number of silicious geodes and occasionally some
chert. Some fifty or sixty feet above the base is a buff lime-
stone containing large numbers of small lamellibranchs, as
Pleurophorus subcuneatus M. & H., Bakewellia parva M. & H.,
Yoldia subscitula M. & H., and other species, while about twenty
feet higher is another similar limestone containing large lamel-
libranchs, as Aviculopecten occidentalis (Shum) Meek, Myatina
permiana (Swallow) M.& H., and Pseudomonotis Hawni (M. &
H.). A limestone containing Pleurophorus occurs in some local-
ities near this horizon, which also contains large chert concre-
tions.

The upper portion of the formation is composed mostly of
thin buff limestones similar to those in the lower portion, alter-
nating with a greater thickness of shales and marls, andin some
localities contains beds of gypsum and salt. On Turkey Creek,
south of the Smoky Hill valley and Abilene, a conglomerate
stratum from fifteen to twenty feet in thickness occurs some 150
feet above the base of the formation, which was first described
by Meek and Hayden in 1859." To the northwest of the Cot-
tonwood Falls quadrangle two beds of gypsum occur in this
formation in Dickinson and Saline counties, both of which are
worked. The lower one was named the “Solomon gypsum” by
Dr. Grimsley,’ and various outcrops occur up Gypsum Creek to
Gypsum, as well as on Holland Creek, near Dillon. The higher
bed is found in Greeley township, southeast of Salina, which was
called the ‘Greeley gypsum” by Dr. Cragin,3 while at Hope in
the southeastern part of Dickinson county both strata of gypsum
occur, separated by one hundred feet of shales and limestones.
In southern central Kansas there are indications of a third hori-
zon, forty feet above the Greeley gypsum, while the deposit in
the southern part of the state, about four miles northwest of
Geuda Springs, Sumner county, is in the upper part of the forma-

t Proc. Acad. Nat. Sct. Philadelphia, Vol. 1X, p. 16, No. 9.

2 Univ. Geol. Surv. Kan., Vol. V, 1899, p. 61.

3 Col. Cotl. Studies, Vol. VI, 1896, p. Io.

718 CHARLES'S. PROSSER

tion. In the southern part of the state the upper portion of the
formation consists to a large extent of clay-shales of various
colors, with some beds of limestone, gypsum, and rock salt.
Its lithologic character, as it dips deeply below the surface to
the westward, is shown by the well records to change from the
variegated shales alternating with beds of limestone and gypsum
to saliferous shales of bluish-gray to slate color alternating with
massive beds of rock salt. It covers a large portion of the
eastern two-thirds of Marion county, its lower part is quite well
shown in the vicinity of the city of Marion, and for these reasons
it was given the name ‘“ Marion formation.”’?. Qn account of the
terms ‘‘Marion flint’”’ and ‘‘ Marion concretionary limestone,”
Professor Cragin,in 1896, named this formation the ‘ Geuda Salt-
measures,” from Geuda, Sumner county, which name he with-
drew during that year in favor of the ‘‘ Marion formation.” 4

Wellington shales —This formation consists largely of bluish-
gray to slate-colored shales, but contains some red ones, and in
the southern part of the state beds of impure limestone and cal-
careous shales, together with occasional beds of gypsum and
dolomite. Limited saline deposits are reported, but no rock
salt. Fossils are very rare, and, as far as the writer is informed,
none have been found in the formation. In the Smoky Hill
valley there are about two hundred feet of the Wellington shales,
but they thicken to the south, and are reported as 450 feet in
thickness in Sumner county, near the southern line of the state.
Professor Cragin named and described these shales in 1896 from
exposures in the vicinity of Wellington, the county seat of
Sumner county.$

TABLE OF THE UPPER PALEOZOIC FORMATIONS OF KANSAS.

The formations just described, together with the succeeding
ones of the Permian, have been arranged in the following table
of the Upper Paleozoic formations of Kansas.

™See Dr. GRIMSLEY’S account in Univ. Geol. Surv. Kan., Vol. V, 1899, p. 69.

2Jour. GEOL., Vol. III, 1895, p. 786.

3 Col, Coll. Studies, Vol. V1, March, 1896, pp. 3, 9-16.

4 Am. Geol., Vol. XVIII, Aug., 1896, p. 132.

5 Col, Coll. Studies, Vol. VI, pp. 3, 16.

UPPER PALEOZOIC FORMATIONS OF KANSAS 719

CORRELATION OF THE CIMARRON SERIES.

In the earlier papers of the writer, the Cimarron series was
referred provisionally to the Permian’ ; but later the discovery
of Permian fossils as high as the Red Bluff formation, together
with other data, apparently proves that at least the greater part
of the series is of Permian age. Mostof the fossils have been
found by Professors A. H. Van Vleet and Charles N. Gould ;
the latter has described the horizons from which they were col-
lected and given a list of three specific and three generic identi-
fications of Permian vertebrates by Dr. Williston, which came
from near the base of the Harper sandstone, and eleven genera
of invertebrates identified by Dr. J]. W. Beede. He states that
the highest locality, which is in the Red Bluff formation, ‘“ has
yielded some twenty species of invertebrates, several of which
are of new forms.’’?

Dr. Beede, who has also published a note concerning these
highest fossils, states that ‘“‘they are mainly pelecypods with a
species of brachiopod and a few gasteropods. . . . Aviculopecten
occidentalis (Shum) Meek, is also present, and one other species
bearing somewhat of a resemblance to it, but quite different
from it in some respects, is also present. One of the common
fossils is a biplicate terebratuloid, Dzelasma Schucherti Beede,
belonging to a group of this genus heretofore unknown in the
American Permian. Mr. Schuchert informs me that it is very
similar to a species of this genus described by Waagen from the
Betimianvon europe, vs). a.

The presence of these fossils clearly demonstrates the Per-
mian age of these rocks, coming as they do from very near to
the top of the beds.”3 The description of these invertebrate
fossils from the Red Beds by Dr. Beede bas been published as
an Advance Bulletin of the First Biennial Report of the Okla-
homa Geological Survey. The following new species are

* Univ. Geol. Surv. Kansas, Vol. Il, 1897, pp. 89-92; Kan. Univ. Quart., Vol.
VI, 1897, pp. 150, 151; Jour. GEOL., Vol. VII, 1899, pp. 354-6.

2 Jour. GEOL., Vol. IX, July, 1901, p. 339.

3 Am. Geol., Vol. XXVIII, July, 1901, pp. 46, 47.

4 April, 1902, pp. I-11, with one plate.

720 CHARLES S.“PIROS SER

described: Bakewellia Gouldi, Conocardium oklahomaensis, Avicu-
lopecten Van Vleeti and Dielasma Schucherti; Aviculopecten occt-
dentalis is identified and, generically, specimens of Maticopsis,
Pleurotomaria, Schizodus, Lima and Pleurophorus, all of which are
from White Horse Springs, Oklahoma. Dr. G. I. Adams also
states that Dr. Williston considers the vertebrate remains from
the Harper sandstone ‘‘as equivalent to Cope’s Lower Permian
fauna from the Wichita beds of Cummins in northern Texas.’’?
He further said, in discussing the age of the Red Beds of eastern
Oklahoma, that ‘‘The age of that portion of the Red Beds which
is in strike with the Permian of Kansas may confidently be
expected to be found to be of Permian age. This is in accord-
ance with the evidence already furnished by the vertebrate
fossils. Above the Permian limestones in Kansas occur the
Wellington shales, which are bluish and greenish-gray in color.
They are probably represented southwestward by formations
which are red. The succeeding formations are typical Red Beds,
and have thus far yielded only Permian fossils.”’ ?

In.a discussion of the “Relations of ‘Upper Permian: to
Triassic’? Dr. Keyes has stated that ‘‘Prosser has been led to
believe that the greater part of the Kansas ‘Red Beds’ are
Triassic.”3 The above statement is erroneous, for previous to the
publication of Dr. Keyes’ paper my discussion of the Cimarron
series in the Kansas report appeared under the heading of ‘‘ The
Upper Permian.” At that time, however, I did not consider the
evidence strong enough to justify their correlation with the
Permian without a question.* This idea was expressed near the
close of the section on ‘Correlation ”’
“On account of this dissimilarity in lithologic characters
[between the Red Beds of Texas and Kansas| and the absence
of fossils in Kansas and northern Oklahoma, together with the
fact that there is yet no account of the careful tracing of any
part of the Red-Beds across Oklahoma to Texas where their age

in the following sentence :

t Am. Jour..Sct., 4th ser., Vol. XII, Nov., 1901, p. 383.

2 [bid., p. 386. 3Jour. GEOL., Vol. VII, July, 1899, p. 339.

4 See the table of classification in Unzv. Geol. Surv. Kansas, Vol. Il, Feb., 1897,
Pp: 94.

UPPER PALEOZOIC FORMATIONS OF KANSAS 721

could be determined by comparison with the fossiliferous ter-
ranes, the correlation of these rocks with either the Triassic or
In the later paper of that

yy

Permian is a matter of uncertainty.
year I simply quoted the opinions of Dr. Williston, Professor
Grimsley and Mr. Vaughan, without expressing any opinion,
beyondthe statement that ‘‘there is uncertainty as to theiage) ac
while in my paper succeeding that of Dr. Keyes’ in the same
number of the JoURNAL OF GEOLOGY it was stated that ‘“‘ The
Paleozoic of Kansas closes with the Cimarron group or the Red
Beds ;’’3 following which was an account of the identification
by Dr. Williston of Eryops megacephalus from the lower part of
the Cimarron series, an amphibian described by Cope in the
Permian of Texas.

THE. PERMIAN QUESTION.

CORRELATION OF THE UPPER PALEOZOIC OF KANSAS WITH THE
RUSSIAN PERMIAN.

Opinions of various geologists.—There is still a difference of
opinion among American geologists in regard to the correlation
of the Upper Paleozoic formations of Kansas with the Russian
Permian. The JourNAL OF GEOLOGy published in 1898, ‘‘A sym-
posium on the classification and nomenclature of geologic time-
divisions,’ in which Dr. Williston,+ Professor Calvin5 and Dr.
Keyes® reported adversely both as to the identification of the
Permian in Kansas and to its recognition as a period codrdinate
with the Carboniferous or Devonian; while Dr. William B. Clark
stated that’ for the later divisions of the Paleozoic he should
employ the chronologic terms Carboniferous and Permian.’ Dr.
Clark wrote me later as follows regarding this subject :

I distinctly object to the abandonment of the term Permian for a major
division and can see no just grounds for it since the division is one of
importance in Europe and other portions of the world. To be sure, in
America the Permian conditions are not as prominent, but I can see no
reason on that ground for disturbing the geological column as it has come to
be generally accepted. °

1 [bid., p. 92. 2 Kan. Univ. Quart., Vol. VI, Dec. (?), 1897, p. 150.
3 Loc. cit., Vol. VII, p.354. 5 Loc. cit., p. 353. 7 Loc. cit., p. 341.
4 Loc. ctt., Vol. VI, p. 343. 6 Loc. ctt., p. 352. ® Letter of December 16, 1898.

Wie. CHARLES) S. Ap lOS SHR

Vertebrate fossils found in northern Texas led Professor E.
D. Cope to the conclusion that the rocks were of Permian age
and he stated that ‘‘The evidence now adduced is sufficient to
assign the formation, as represented in Illinois and Texas, to the
Permian.’’* The invertebrate fossils from the same beds were
regarded by Dr. Charles A. White as indicating their Permian
age.? The stratigraphy of the Upper Paleozoic formations
of Texas was fully described by Professor W. F. Cummins, who
referred them to the Permian,3

In recent years the following geologists have studied the
Upper Paleozoic rocks of Oklahoma, Kansas or Nebraska and
termed them Rermian: “Dr. James Ps Smith: who ‘stated that
‘The lower Permo-Carboniferous strata of Kansas and Nebraska
are probably also to be correlated with the Artinsk stage [ basal
Permian of Russia].”* Professor Cragin, who gave an extended
account of ‘‘the Permian system in Kansas;’’5 Professor Wilbur
C. Knight who wrote a similar paper on ‘‘The Nebraska Per-
mian’’® and showed from tables of distribution that ‘‘ Of the
forty-four genera of invertebrates known in the Kansas and
Nebraska rocks, over three-fourths of them belong to the Per-
mian of the Orient. The remainder are nearly all American
genera and are chiefly pelecypods.”? Prof. Knight has also
stated that ‘From our present knowledge it seems advisable to
refer the Red Beds of the Laramie Plains | Wyoming | to the

Permian.’ ®

Dr. J. W. Beede, in his paper on ‘‘ A Reconnaissance
in the Blue Valley Permian”? described the Lower Permian as
represented in Kansas north of the Kansas river and in southern
Nebraska. And finally Professor Charles N. Gould and Doctor

t Geol. Surv. Texas, Second Ann. Refpt., 1891, p. 414.

2 Bull. U.S. Geol. Surv., No. 77, 1891.

3 Geol. Surv, Texas, Fourth Ann. Rept., 1893, p. 212.

4Jour. GEOL., Vol, II, 1894, p. 194; and see pp. 188, 204. Also see Proc. Am.
Phil. Soc., Vol. XX XV, 18096, reprint pp. II, 12, 24.

5 Col. Coll. Studies, Vol. VI, 1896, pp. I-49 and supplemented by one in the Am.
Geol., Vol. XIX, 1897, pp. 351-64.

©Jour. GEOL., Vol. VII, 1899, pp. 357-75.
7 Lbid., p. 370. 8Jour. GEOL., Vol. X, 1902, p. 421.
° Kan. Univ. Quart., Vol. IX, July, 1900 (1901), pp. 191-203.

UPPER PALEOZOIC FORMATIONS OF KANSAS 723

Beede have shown from fossils, the Permian age of the Kansas-
Oklahoma Red Beds."

Another paleontologist, who is studying the fossils of the
western Carboniferous writes me :

I don’t believe that there is any Permian there at all, unless possibly the
Marion and superjacent beds are Permian. I express the opinion with that
qualification, (the possibility of the Marion being Permian), and another that
the Kansas area #zay have been a shut-in basin and have retained its Car-
boniferous facies into Permian time.

Dr. Erasmas Haworth wrote me as follows:

I do not hesitate to say that | am most strongly opposed to the substitu-
tion of the term Oklahoman or any other for that of Permian. It looks now
as though the whole of the Red Beds would be called Permian. Should this
be done we will have a terrane which, along the southern line of Kansas, will
be from 2,000 to 3,000 feet thick, or in other words as thick as the whole of
the Coal-measures. This mass of earth is as different in all physical aspects
from the Coal-measures as they are from any other terrane. It should be
given a prominent place, but just how prominent I am not yet ready to
express an opinion. I do not see how anybody can well settle the question
of rank of the American Permian until all these questions are worked out,
which work will necessitate an intimate examination of the territory lying
between Kansas and Texas. I favor insisting on the use of the term Permian
and let its rank stand as others have given it until somebody is ready to
tell us in detail and in a connected way what we have in Kansas, Indian
Territory, and Texas. As far as I can see the indications now are that the
Permian ultimately, with the Red Beds included, shall be entirely separated
from the Carboniferous. ?

In 1891 Dr. Th. Tschernyschew, the former able director of
the Russian Geological Survey and the authority on the middle
and upper Paleozoic of Russia, in company with Professor H. S.
Williams, examined the rocks as exposed along the Kansas
river from Manhattan to Fort Riley. Their conclusions were
reported as follows by Mr. Robert Hay: ‘While agreeing that
the lower beds [at Fort Riley | are Permo-Carboniferous, they
state that the upper beds—where the Phacoceras is—are
decidedly Permian, the Russian professor assuring me that both

‘GOULD, Jour. GEOL., Vol. IX, I901, pp. 337-41; BEEDE, Am. Geol., Vol.
XXVIII, 1901, pp. 46, 47; and Adv. Bull. First Bien. Rept. Okla. Geol. Surv., 1902,
pp. I-Tl.

2 Letter of December 16, 1898.

724 CHARLES S; PROSSER

faunal and lithologic characters can be duplicated in the Permian

of his own country.
Cephalopods from Fort Riley and Junction, Kansas, were identi-

The specimens of Phacoceras and other

fied and described by Professor Alpheus Hyatt and came from
the Fort Riley limestone. Therefore, according to the above
statement Dr. Tschernyschew correlated the Fort Riley lime-
stone and superjacent Paleozoic formations with the Permian
of Russia. ~

Correlation of the Kansas and Texas beds.— Professor Cum-
mins reported that:

The Phacoceras Dumblez, Hyatt, has been found only along a very nar-
row horizon in the Texas Permian. . . si. “Ehis fact willl assist
materially in correlating the Texas and anes beds, as that fossil has been
reported only from cne locality in the Kansas area, where it is associated
with the same fossils as in Texas. It is quite certain that the Fort Riley
horizon is the same as the Wichita division of Texas, and is at the very top
of the division.?

In ascending order the divisions of the Texas Permian as
described by Professor Cummins are the Wichita, Clear Fork
and Double Mountain;3 while the Albany division‘ was left ‘as
the top of the Coal-measures’’5 although the statement was
made that ‘It may be that the Wichita and Albany divisions
are but different facies of the same formation’”’ for the Wichita
division north of the Brazos river ‘‘ occupies the same position,
stratigraphically, as the Albany beds onthe south.”® Later Pro-
fessor Cummins proved the correctness of the latter supposition
‘and stated that he “found the fact well established that the
Wichita and the Albany divisions were the same in time of depo-
sition, and therefore the Albany must be abandoned both as to
its name and the age to which I| had previously referred it, and

t Trans. Kan. Acad. Sct., Vo XIII, 1893, p. 38.

2Trans. Texas Acad. Sci., Vol. I1, 1897, pp. 97, 98. Also see D. W. JOHNSON,
in Bull. Sct. Lab. Denison Univ., Vol. XI, 1900, p. 223.

3 Geol. Surv. Texas, Second Ann. Rept., 1891, pp. 361, 373. Fourth zbid., 1893,
Ppp: 224-32.

4 Prof. Hill has shown that the Albany division of Cummins is the same as the
one named and described at an earlier date by Prof. Tarr as the Coleman. Twenty -
first Ann. Rept. U. S. Geol. Surv., Pt. VII, 1902, pp. 96, 97).

8 [bid., p. 224. 6 Jbid., p. 223.

UPPER PALEOZOIC FORMATIONS OF KANSAS 725

the beds composing the division must be referred to the Wichita
division of the Permian. Since the Wichita division is now
made to include the area heretofore referred to as the Albany
division, it becomes at once the most important and interesting
part of the Permian in North America.”* And he concluded
with the statement that ‘it has been determined that the Albany
division, with its numerous fossils, is but another facies of the
Wichita division which is beyond question Permian.’’?

ligerotessor «Cummins, be correct in correlating the Wort
Riley limestone with the top of the Wichita division of Texas it
very decidedly supports the reference of the Upper Paleozoic
formations of Kansas to the Permian. He had stated that all
the typical localities of invertebrate fossils described by Dr. C.
A. White were included in the Wichita division, ‘the greatest
number of vertebrate fossils described by Professor Cope’’ and
the fossil flora described by Dr. I. C. White;? while the Phaco-
ceras Dumblet ‘was taken from the very top of the Albany
division.” 4

The fossil named and described by Professor Heilprin as
Ammonites Parkeri which was reported from rocks of Carbonifer-
ous age in Wise county, northern Texas,> was referred to Popano-
ceras by Professor James P. Smith® who stated, on the authority
of Professor Cummins, that the Popanoceras Parker beds are in the
Strawn | Richland] division and therefore of the age of the Lower
Coal-measures.? The occurrence of this type in the Texas beds,
however, led Karpinsky in 1889 to write as follows: Since the
Popanoceratide up to the present time have not been found in
other countries in deposits which are older than the Permo-
Carboniferous (in which the commonest Ammonites occur),

t Trans. Texas Acad. Sct., Vol. I, 1897, p. 97. Also see JAMES P. SMITH, Proc.
Am. Phil. Soc., reprint, 1896, p. 13.
2 Trans. Texas Acad. Sci., Vol. Il, 1897, p. 97.
3 Geol. Surv. Texas, Fourth Ann. Rept., p. 225. Also see Trans. Tex. Acad. Sci.,
I, 1897, pp. 94, 95.
4 Geol. Surv. Texas, Fourth Ann. Reft., p. 223.
5 Proc. Acad. Nat. Sct. Philadelphia, 1884, Vol. 36, pp. 53-5.
6Jour. GEOL., Vol. II, 1894, p. 194; and see “ Correlation Table” on p. 204.
7 Proc. Am. Phil. Soc., Vol. XXXV, 1896, reprint, p. 16 f *.

Vol.

=—

726 CHARLES. Si PROSSER

therefore in my opinion the Texas deposits must rather be
assigned to the Permo-Carboniferous.*

In 1891 came Dr. C. A. White’s description of ‘thirty-two
species of invertebrates <9). trom, (thes fexany Rermnani
of which four Cephalopods belonging to the family Asmmonoidea
were recognized as new. It was stated that two of these types,
Waagenoceras Cumminsi and Popanoceras Walcott, ‘‘are so generally
regarded as indicating the Mesozoic age of the strata containing
them that if they alone and without any statement of correlated
facts had been submitted to any paleontologist he would not have
been warranted in referring them to an earlier period than the
Trias if he had followed the usually accepted standard of refer-
ence: 4

In conclusion Dr. White stated that ‘‘ The evidence upon which
the Texan strata have been referred to the Permian is fuller than
that which has been adduced with regard to any other North
American strata: that have been so referred. That is, the evi-
dence both of the vertebrate and invertebrate fossils is in favor
of such reference, and the difference in the character of the strata
from those of the underlying Coal-measures, although not great,
is conveniently distinguishable ;’’ 3 while he was inclined to con-
sider the Texan Permian as of younger age than the Indian and
Sicilian strata containing the commingled Mesozoic and Carbon-
iferous forms which were described by Professors Waagen and
Gemmellaro.

Waagen correlated the ‘‘ Red sandstones and shales of Texas,
with many remains of Vertebrates, Amphibia and Repttla and
Gontatites Baylorensis, Hyattoceras Cumminsi, Medhcottia Coper and
Popanoceras Walcott’’ with the ‘‘Weissliegendes and marl slate”’
which he put at the base of the magnesian limestone, that formed
the upper division of his Permian system.#

Marcou stated ‘It is certain that the Wichita division belongs

tVWém. Acad. Imp. Sciences St. Pétersbourg,VI1® Sér., t. XX XVII, No. 2, 1889, p. 93.
Also see the correlation of the Texas deposits as shown in Table C, p. 94.

2 Bull U.S. Geol. Surv., No. 77, p. 31. 3 Lbid., p. 38.

4 Mem. Geol. Surv. India, Pale, India, ser. xiii, “Salt-Range Fossils,” Vol. IV,
Pt. II, “Geological Results,” 1891. Tabular View showing the relations of the Salt-
Range Upper- Paleozoic strata to the deposits of other countries, op. p. 238.

UPPER PALEOZOIC FORMATIONS OF KANSAS 727

yy

to the Dyas (Permian)’’*; while in considering the list of fossils
given by Cummins he said: ‘It isa fauna related with the Russian
fauna of the Artinsk beds, and may be considered as the American
representative of a part of the Russian Dyas (Permian).’”

Professor James P. Smith stated that “the Ammonite-bearing
beds of northern Texas, described by Dr. C. A. White .
belong above the Artinsk stage, and in the true Permian, and are
probably of the same age as the middle division of the Mzddle
Productus limestone of the Salt Range | India ].’’3

Dr. Keyes in discussing the parallelism between the Texas and
Kansas beds said, ‘‘The Double Mountain beds are, in a broad
way, manifestly approximately equivalent to Cragin’s Cimarron
series. This leaves a considerable part of the Clear Fork beds
representing the Chase and Marion of Kansas.”* Professor de
Lapparent considered that in northern Texas the Uralian with
Productus cora and Athyris subtilita is succeeded conformably by
300 meters of sandstones and shales, occasionally calcareous, in
which the red color predominates, the base of which appears to
belong in the Artinsk.s He stated that the red gypsiferous
beds with Pleurophorus which in the western part of Texas sur-
mount the Wichita formation belong in the Upper Permian.°
Finally, Dr. Frech puts the Wichita and Clear Fork beds in the
Palaeo-Dyas and the Double Mountain beds in the Neo-Dyas’
and states that the Ammonoids described by Dr. White — Medh-
cottia Copei, Popanoceras Walcotti and P. ( Hyattites) Cumminsi have
their nearest relatives in the marine Dyas of Sicily.®

Provisional correlation of the Kansas formations.—The above
statements indicate clearly enough the differences in opinion
among geologists more or less acquainted with the Upper Paleo-
zoic formations of the Great Plains, regarding their correlation.
It is to be noted, however, that there is a more general agreement
regarding the Permian age of the Texas deposits, and if Professor

t Amer. Geol., Vol. X, 1892, p. 370. 27 G70: 371 Uc

3Jour. GEOL., Vol. II, 1894, p. 194; and see ‘Correlation Table” on p. 204.
4Jour. GEOL., Vol. VII, 1899, p. 325.

5Traité de Géologie, 4th ed., 1900, p. 981. 6 Jozd., p. 994.

7 Lethaea paleozoica, Bd. Il, 3 Lief., 1901, p. 514. 8 Jbid., p. 515.

728 CHAKLES S. “PROS SH

Cummins has correctly correlated the Fort Riley limestone with
the Texas deposits it furnishes a strong argument in favor of
referring the Upper Paleozoic formations of Kansas to the
Permian.

Furthermore, the number of American geologists who believe
that these Upper Paleozoic formations should be correlated with
the Permian and given the rank of a period or system is probably
still smaller than the number of those who would retain the name
Permian but classify it as the upper series of the Carboniferous.
The writer had hoped to carefully study the fossils of these
formations and to present their complete evidence regarding
these questions, but other duties have prevented the execution of
this plan. It has appeared to me, however, that the weight of
evidence favored correlating the upper formations with the
Permian.

Whether the Permian should be assigned the rank of a sys-
tem coordinate with the Carboniferous or regarded as the upper
subdivision of it is not quite clear, and the line of division
between the Permian and the Carboniferous is in doubt, as indi-
cated on the chart p. 730. The opinions of some of the lead-
ing European students of the Upper Paleozoic, who regard the
Permian as a distinct system and correlate certain American
formations with it, has seemed to the writer sufficient authority
for provisionally regarding it as a system, which was done in
the table of classification opp. p. 704. It is probable, however, that
the U. S. Geological Survey will retain the name Permian, but
will classify it as the last series of the Carboniferous system.

Conclusions of Dr. Keyes No one has, perhaps, insisted as
strenuously as Dr. C. R. Keyes that the name Permian should
be dropped from American geology. In 1897 he attended the
sessions of the International Congress of Geologists at St. Peters-
burg and participated in the excursions to the Carboniferous
and typical Permian of Russia. Later he prepared a paper on
the ‘‘ American homotaxial equivalents of the original Permian,”
and quotations from this cannot be regarded as from one favor-
ing the retention of the name “Permian.” Regarding the litho-
logic features Dr. Keyes said :

UPPER PALEOZOIC FORMATIONS OF KANSAS 729

The original Permian strata are indistinguishable, lithologically, from the

so-called Permian of Kansas. In both there are the same gray and varie-
gated sandy shales and marls, passing locally into sandstones, that are often
copper-bearing. Occasionally there are present thin bands and beds of buff
earthy limestone. Gypsum is abundantly developed in the beds and inter-
spersed everywhere through the rocks. Saline shales are of not infrequent
occurrence. On both continents all these pass upward into ‘‘Red Beds”’ that
are almost destitute of fossils.
And in another paragraph is a striking statement that ‘In the
Russian district one finds it difficult to imagine that he is not
wandering through some part of Kansas. Only the presence of
the Russian peasant or sudden contact with a village of the
steppes dispels the illusion.”

Secondly, under the heading, ‘‘ Range of faunas,” Dr. Keyes
reported as follows regarding the fossils:

The succession of faunas appears to be essentially the same in the Russian

Carboniferous and Permian as in the Mississippi valley. The composition of
each of the faunas is also strikingly comparable. The most noteworthy
feature of the organic remains, viewed as a whole, is the gradual replace-
ment of a purely marine type by a shore and brackish water phase, as the
change from open sea to closed water conditions took place, and finally to
those in which life could not exist. The most prominent characteristic of the
biotic change from a Carboniferous phase to a Permian one seems to be the
replacement of a predominantly brachiopod fauna by one in which lamelli-
branchs formed the preponderant element.*
While in another article Dr. Keyes said: ‘In lithological and
faunal characters the rocks are so nearly alike that it is difficult
to fancy that in the Urals one is on the opposite side of the
earth from our Iowa and Kansas beds.’’?

Under the general heading, ‘‘Comparison of the Russia and
Mississippi Valley Carboniferous,’ and subheading, ‘ Strati-
graphic parallelism,” Dr. Keyes stated that ‘(In Russia and in
the Mississippi valley the general geological sections of the
Upper Paleozoic are remarkably alike. The basins occupied by
these rocks are very nearly of the same size. As already stated
in the first-mentioned area, the Permian very greatly predomi-

‘Jour. GEOL., Vol. VII, 1899, p. 334.

2“ Permian Rocks of Eastern Russia,” in Proc. Jowa Acad, Sciences, Vol. VI,
1899, p. 231.

730 CHARLES S. PROSSER

nates as the surface rock; in the last-named, the Coal-meas-
ures. While the above paragraph is followed by the following
‘Comparison of general sections”’ in Russia and the Mississippi

yy

valley, which may evidently be regarded as Dr. Keyes’s idea of
the correlation of the upper Paleozoic rocks of the central United
States and Russia.

Russia. CHARACTER OF TERRANES, MissIssipP1 VALLEY,

Tartaran, Permo-|Shales and marls, red and variegated,|Cimarron Series
Trias, or Up-| shaly sandstones; fossils rare;
per Permian, P, “Red Beds”

Middle Permian, P,|Limestones, some dolomitic, sepa-|(Marion li.) |
rated by calcareous marl

Lower Permian, P,-b|Shales (only 200 feet thick in Kama|-————? + Series
Valley) |
Upper Permo-Car-|Limestone, heavy dolomitic (Chase li.) J

boniferous (base
of original Per-

mian) CP.
Artinsk, CP Shales, sandstones, some thin lime-|/(Neosho)
stones (Cottonwood) } Series
(Wabaunsee)

Upper Carbonifer-|Limestones and shales, highly fossil-|Missourian Series?
ous. iferous

Finally, in his “‘ Recapitulation”’ it is stated ‘‘ That while we
have in America a great succession of deposits identical in all
essential respects to the original Permian of Russia, the two
great basins merely had similar histories that are not necessarily
connected and doubtless were wholly independent of each other
and unrelated.’’3

Dr. Keyes’s description and comparison of rocks and faunas
apparently support the correlation of the Upper Paleozoic of the
Great Plains with the Permian of Russia, providing one follows

tJour. GEOL., Vol. VII, 1899, pp. 331, 332.
2 [bid., p. 332; Proc. Iowa Acad. Sciences, Vol. VI, p. 230.
3Jour. GEOL., Vol. VII, p. 341; Proc. Lowa Acad. Sciences, Vol. VI, p. 231.

UPPER PALEOZOIC FORMATIONS OF KANSAS 731

the rules of correlation generally observed by geologists.* The
evidence is apparently about as conclusive as for other systems
in this country which are correlated with the Carboniferous,
Devonian, or Silurian of Europe. Apparently the main point of
Dr. Keyes’ contention is ‘That [the | Permian, as originally
proposed, applies to a provincial series, and, according to our
usual standard, has, at best, a taxonomic rank below that of sys-
tem.’’? Yet he states it is probable that its main subdivisions
will be elevated ‘to the rank of series,’’ which, instead of caus-
ing the name Permian to be dropped, as he suggests, will more
probably leave it with the rank of a system as originally defined
by Murchison. A geologist familiar with the Kansas formations
wrote as follows concerning the provincial series question :
‘Grant, as Keyes maintains, that Permian is the name of a
provincial series, then where a similar series is found with simi-
lar fossils the same name ought to be given. All our names
were names of provincial series at first. What was Devonian
but the name of a series of rocks in Devonshire, England ?
When found in New York, by this argument, they should be
called New Yorkian or some other American name.”

The conclusions of Dr. Frech.— On the other hand, the conclu-
sions of Dr. Fritz Frech, the eminent professor of geology and
paleontology in the University of Breslau, may be considered.
He has carefully studied, both in the field and laboratory, the
Permian of Germany and Russia and examined in the field the
Permian of the United States, at least sas shown in the Grand
Canyon and near Salt Lake City, Utah. Dr. Frech gives these
rocks the rank of a system, which is also the usage of Dr. Kay-
ser, of the University of Marburg,+ but instead of Permian he

* For instance, if his account be compared with the list of physical and biotic
methods of correlation given by Professor Gilbert at the Washington meeting of the
International Congress of Geologists, it will be seen that several of the methods are
fulfilled (Congrés Géologique International, Compie Rendu, 5™° session, Washington,
pp. 68, 69).

2 Loc. cit., p. 341; and p. 231.

3See Congrés Géologique International, Compte Rendu, 5™° Ses., Washington,
1891, 1893, p. 481; and Lethaea paleozoica, Bd. II, 3 Lief., 1901, p. 515.

4See Text-Book of Comparative Geology, by E. KAYSER, translated and edited by
Philip Lake, 1893, p. 164.

732 CHARLES S, PROSSER

uses the later name of Dyas proposed by Marcou on account
of the sharply marked separation of the system into two divisions
in Germany.

Dr. Frech’s classification of the Upper Paleozoic of Kansas is
as follows:

Red shales and clays.
Marion.
Lower Dyas Chase.
Transition to Carboniferous
(distinct line fails.)

Upper Dyas

Neosho.

Cottonwood beds.
Upper Carboniferous~ Wabaunsee."

Later Dr. Frech reviewed Professor Cragin’s classification of
the Permian, and termed the Cimarron series the ‘‘ Neo-Dyas,”’
and the Big Blue series the ‘“‘ Paleo-Dyas.”* He stated that the
Dyas equaled the Permo-Carboniferous plus the Permian of
many authors, and that by general agreement at the St. Peters-
burg International Congress of Geologists the names Paleodyas
(=Permo-Carboniferous) and Neodyas (=Zechstein) are em-
ployed.

Under the discussion of the boundary line between the Dyas
and Carboniferous Dr. Frech said: The dividing line between
the Carboniferous and Dyas formations cannot be drawn with
full certainty in every region, since especially in the Dyas the
development of the local flora is nearly always the rule, and
decisive differences do not exist in the Brachiopod fauna.

Yet an agreement seems to be gradually forming every-
where. .... Where the characteristic Dyas bivalves (Pleu-
vophorus, Schizodus, Bakewellia, Pseudomonotis) appear in
masses (Kansas), there cannot be any doubt about the dividing
line.* Under the description of the Dyas of the northern hem-
isphere and the Arta stage of Russia, as Dr. Frech prefers to call
the Artinsk, he said: That the animal remains of the Permo-

* Lethaea baleozoica, Bd. II, 2 Lief., 1899, p. 378, as translated above.
2 [bid., 3 Lief., 1901, p. 514. 3 [bid., p. 453 f.

4Jbid., pp. 490, 491. Iam greatly indebted to Charles W. Mesloh, associate
professor of Germanic languages and literatures in the Ohio State University, who
very kindly translated for me several pages of Dr. Frech’s description of the Dyas.

UPPER PALEOZOIC FORMATIONS OF KANSAS Yiss

Carboniferous are in general more nearly related to the Carbon-
iferous than the Zechstein, finds its explanation in the poverty
of the species of the inland seas. The Arta stage occupies a
large space on the western slope of the Ural mountains from the
Arctic ocean to the Kirghiz Steppe and the Donetz River, and
was correctly classified with the Dyas by older investigators
(Pander). The plant remains described by Schmalhausen speak
quite decidedly for a comparison with the western Rothliegende*
Murchison considered the Arta sandstone the Millstone grit,
while the modern Russian authors mostly call it an intermediary
stage from the Carboniferous to the Dyas, Permo-Carboniferous.
If the latter assumption were correct, then the Cusel and Lebach
strata would also have to be regarded as transitional from the
Carboniferous to the Zechstein, z. ¢., the most important and
best known part of the formation would becomea transition and
only the equivalent of the German Zechstein would be desig-
nated as Permian.’

Usage of Russian geologists—Among the recent Russian geolo-
gists who have described transitional deposits between the Car-
boniferous and Permian systems, may be mentioned the follow-
ing: Krotow, who in 1888 described the Permo-Carboniferous
and Permian on the western slope of the Urals in the region of
Tscherdyn and Solikamsk.3 Th. Tschernyschew, in 1889,
described the Permo-Carboniferous of the western slope of the
central Urals, which he lettered C P, and gave as composed of
the Artinsk (C Pg), and superjacent Dolomitic limestone (C Pc),
the latter forming the base of Murchison’s Permian system.‘
Krasnopolsky, the same year, described the Permo-Carboniferous
and Permian deposits of another portion of the western Urals,5
which was followed two years later by a further description.®
Stuckenberg, in 1890, described the Permo-Carboniferous of
another region, which he gave as composed in ascending order

tJbid., pp. 493, 494. 2 [bid., p. 493, f. 2.
3 Mém. Comité Géologigue, Vol. V1, pp. 553-9.
4 /bid., Vol. III, No. 4, Blatt 139, pp. 356-66.

5 [bid., Vol. XI, No. 1, Blatt 126, pp. 506-18.

5 Jbid., No. 2, 1891, pp. 28-30.

734 CHARLES S. PROSSER

of the Artinsk and Kungur stages." Sibirzev, in 1896 carefully
described the Permian deposits near Nishny-Novgorod on the
Volga, together with those of the Permo-Carboniferous farther
to the west ;” while Stuckenberg two years later described ina
similar manner the Permo-Carboniferous and Permian formations
of the Kama basin.3

Since then Dr. Keyes has very clearly summarized the Russian
classification of the Upper Paleozoic terranes of eastern Russia in
the following table :+

Terrane. Symbol, Character,

Tartaran | PT or P, | Shales and marls, “‘ Red Beds,” very few fessils.

Zechstein Be Marls, limestones, and sandstones.

(in part.)

Seaton eae Pb Sandstones, shales, and marls with nodular limestones.

Been match iets € Pe Dolomitic limestones (base of Murchison’s Permian).

Artinsk Gig Shales, shaly sandstones. This and next terrane above
are called Permo-Carboniferous.

Spo cw OF Limestones.

Dr. Keyes states that following ‘the so-called true Carbon-
iferous of the ‘Urals’... 9. are, the transition) faunas: to ithe
Permian, according to the Russians, and by them called Permo-
Carboniferous. The two members which comprise it contain, as
pointed out by Tschernyschew, very nearly the same organic
forms, consisting largely of lamellibranchs, gasteropods, and
brachiopods. The lower terrane, termed the Artinsk is notable
for the ammonites that are found in it, which the author just
mentioned compares with those lately found in the Texas
Permian.

“The bottom terranes of the Permian, as now recognized by
the members of the Russian geological survey, present a great
paucity of fossils. The forms are chiefly lamellibranchs, yet in
some layers are fragmentary plants.

“The median part of the Permian carries what has been
regarded as the typical German Zechstein fauna.

t Ibid., Vol. IV, No. 2, Blatt 138, pp. 111-14.

2 [bid., Vol. XV, No. 2, Blatt 72, pp. 242-65.

3 [oid., Vol. XVI, No. 1, 1898, pp. 309-21.

4Jour. GEOL., Vol. VII, 1899, p. 330.

UPPER PALEOZOIC FORMATIONS OF KANSAS 735

“About the upper terrane there is much dispute as to age.
The Russian geologists are about equally divided. Amalitzky
considers it Permian. By others it is regarded as Triassic.
Fossils occur rarely. Those found are chiefly lamellibranchs.”’*

OPINIONS OF OTHER EUROPEAN GEOLOGISTS.

There are other noted European geologists, however, whose
conclusions are in general accord with those of Dr. Frech.

Waagen, in his magnificent work on the geological classifi-
cation of the Upper Paleozoic rocks of the Salt-Range in northern
India, ranked the Permian as a system which he divided into the
three following groups, arranged in ascending order: Permo-
Carboniferous, Rothliegendes and Magnesian limestone. He
published a table showing the correlation of the Upper Paleozoic
strata of the Salt-Range with similar deposits of other countries
on which the ‘‘red sandstones and shales of Texas,’
vertebrates and invertebrates which have been described by Cope

5)

containing

and White, were correlated with the lower part of the Magnesian
limestone, or upper group of his Permian system. The ‘“lime-
stones and shales, with Pseudomonotis hawni (—— speluncaria)
of Kansas, red gypsum beds of Texas’ * were regarded as equiva-
alent to the remaining portion of the Magnesian limestone group
and consequently represent the upper part of the Permian system.
De Lapparent, in the last edition of his comprehensive
treatise of geology, gives the Permian the rank of a system to the
lowest stage of which, the Artinskien or Autunien, he refers the
Neosho, Chase and Marion terranes of central Kansas. It is
stated that it would be difficult to class elsewhere than in the
Artinskien, the Neosho and the Chase, although there may be a
doubt regarding the correlation of the Marion.* On his table of
‘“Synchronism of Permian assises,’ beds with Plewrophorus and
with Pseudomonotis of Kansas are given as in the Thuringien or
Upper Permian stage,> while on the following page it is stated
t Lbid., pp. 330, 331.

2 Mem. Geol. Surv. India, Pale., India. Series 13, Salt-Range Fossils, Vol. IV,
Pt. II, Geol. Results, Calcutta, 1891, op. p. 238.

3 Traité de Géologie, 4thed., Pt. II, 1900, pp. 759-963.
4 Joid., pp. 980-1. 5 [bid., p. 993.

736 GHAREES S.-PROS SPR

that in the Upper Permian is, perhaps, also the horizon of the
upper limestones and shales of Kansas with Pseudomonotts Hawnt,
which surmount 75 meters of variously colored shales and marls,
with gypsum (assise de Marion de M. Prosser) ."

It is to be noted, however, that de Lapparent is in error in
correlating the sandstones and shales of Nebraska with the lower
part of the Penjabien or Saxonien, which he classifies as the
Middle Permian.? Such a classification puts the Nebraska City
beds at least above the Neosho and Chase, as is clearly indicated
on his table of synchronism, while as a matter of fact it has been
shown by Dr. Beede and the writer that they are probably
equivalent ‘‘to the Topeka limestones and Osage shales of the
Kansas river section, which form the upper part of Professor
Haworth’s Shawnee formation of the Upper Coal-measures.”’ 3
The rocks included between the top of the Shawnee formation,
which is marked by the base of the Burlingame limestone, and the
top of the Chase stage have an approximate thickness of 950
feet in eastern central Kansas, which gives an idea of the strati-
graphic error when the Nebraska City beds are assigned to a
position above the Chase stage.

In discussing the rank of the Permian de Lapparent wrote :
The marine types of the Permian, scarcely known until recent
years, show in Asia as in the United States greater and greater
development. Finally, the well confirmed discovery of Ammon-
ites with arborescent septe gives to the pelagic fauna of the
period a special character, at the same time that by the first
appearance of true reptiles the terrestrial fauna shows a higher
order than that of the preceding period. Therefore, we, agreeing
with the excellent arguments of Neumayr in his Ardgeschichte,
raise to the rank of system this last division of Primary time.

Finally, from among the other famous European geologists
who rank the Permian as a system and have written in support

* [bid., p. 994. 2 Jbid., pp. 986-93.

3JouR. GEOL. Vol. VII, Aug., 1899, p. 346. Also, see PROSSER, 262d., Vol. V,
March, 1897, p. 148; and BEEDE, Kan. Univ. Quart., Vol. VII, Oct., 1898, Series A,
p- 231; and Zrans. Kan. Acad. Science, Vol. XVI, 1899, p. 70.

4 [bid., p. 964.

UPPER PALEOZOIC FORMATIONS OF KANSAS Loh

of this proposition may be mentioned the following: Credner,’
Prestwich,? Neumayr,3 Sir Archibald Geikie,t Ed. Suess,’ and

Karl v. Zittel.®

CHARLES S. PROSSER.
COLUMBUS, OHIO,
June, 1902.

The question as to whether the Alma limestone should be
substituted for the Cottonwood limestone on account of the earlier
name of Cottonwood Creek beds in Texas was submitted to the
U.S. Geological Survey Committee on Formation Names and
under date of October 29, 1902, Dr. C. Willard Hayes has
sent me the following report of the committee:

This committee approved the name Coftonwood limestone at its meeting
March 29, 1902, and at that time considered the priority of Cottonwood
Creek beds, Texas. It decided at the time that, although the latter name had
priority of usage, it probably was not a clearly-defined formation but merely
a bed of unmapable dimensions. Also that inasmuch as Cottonwood Creek
beds has not occurred in literature since its first usage in 1893, and whereas
Cottonwood (Cottonwood Falls) formation has been used thirteen times since
its first usage in 1894, the latter name has acquired a place in literature on
the grounds of prescription. The committee therefore decided to adhere to
its former decision in favor of Cottonwood limestone.”

In compliance with the above decision the writer withdraws
the name Alma limestone and retains Cottonwood limestone as
the name of the Kansas formation. CS) Pp:

October 31, 1902.

* Elemente der Geologie, 6th ed., 1887, pp. 382-507.

2 Geology, Chemical, Physical and Stratigraphical, Vol. Il, 1888, pp. 8-131.

3 Erdgeschichte, Bd. Il, 1890, pp. 37-199.

4 Text-Book of Geology, 3d. ed., 1893, ‘The Geological Record,” op. p. 679 and
p. 841.

5 La face de la terre, translated by EMM. DE MARGERIE. T. II, 1900, p. 407.

6 History of Geology and Paleontology, translated by MARIA M. OGILVIE-GORDON,
IgOl, p. 453.

ON SOME GLAUCOPHANE AND ASSOCIATED SCHISTS
IN, THE COAST RANGES OF CALTFORNIEAZ

Tue blue amphibole or glaucophane schists of the California
Coast Ranges, with which are genetically associated actinolite
and garnet schists, have been objects of considerable geologic
interest since they were first observed in 1877. They have been
cited by Mr. H. W. Turner,? Dr. H. W. Fairbanks,3 and by others
from many parts of the Coast Range mountains, throughout
which they are abundant.

The schists occur to a large extent as rather massive isolated
outcrops, and, in general, do not show their schistose structure
except upon a near examination. They vary in texture from
layer to layer; a hard, compact quartzose sheet being succeeded
by a wrinkled, elastic, micaceous layer, which may be followed by
a dense massive variety containing but little mica. These
schists are for the most part entirely crystalline, and are princi-
pally characterized by the abundance of blue amphibole or glau-
cophane, which they contain. Whether this blue amphibole is
mainly glaucophane, crossite or riebeckite the writers have not
determined. There appear to be at least two varieties of the
blue amphibole present in the series, one with a wide angle
between the optic axes and strong double refraction, and another
with a very narrow angie and weak double refraction. For the
sake of convenience, however, and in accordance with the gen-
eral practice, the blue amphiboles will be referred to in this
paper as glaucophane.

tThe writers are indebted to Dr. J. P. Smith, of Stanford University, for assist-
ance and advice.

2“Notes on Some Igneous, Metamorphic and Sedimentary Rocks of the Coast
Ranges of California,” H. W. TURNER, JOUR. GEOL., Vol. VI, p. 488 e¢seg., “The
Geology of Mount Diablo, California,’ H. W. TURNER, Bull. Geol. Soc. Am., Vol. I,
p- 385.

3“ The pre-Cretaceous Age of the Metamorphic Rocks of the California Coast
Ranges,” H. W. FAIRBANKS, Am. Geol., Vol. IX, p. 160; “ Notes ona Farther Study
of the pre-Cretaceous Rocks of the California Coast Ranges,’ H. W. FAIRBANKS,
Am. Geol., Vol. XI, pp. 70-73.

738

———ooe-)hMh lh

—_—

GLAUCOPHANE AND ASSOCIATED SCHISTS 739

Of less importance though widely distributed and intimately
associated with the glaucophane is the light green actinolite
schist which occurs with the glaucophane schist in irregular
layers and masses. Both of these schists have garnets abun-
dantly developed in them.

Dikes of serpentinized peridotite and also of diabase are com-
monly found in apparent association with the schists. It was on
such an association that Dr. F. L. Ransome’ based his hypothesis
that the Angel Island glaucophane schist is the result of the
contact action of fourchite and peridotite intrusions in the
Golden Gate or Franciscan sandstones. Ransome’s conclusions
are questioned, however, by Turner,? who says: “It is yet to be
demonstrated that these schists are the result of contact
metamorphism.”’

In a short note in his paper on metamorphism, Professor C.
R. Van Hise? refers to the glaucophane schists of the northern
end of Calaveras Valley, in Alameda county, as resulting from
dynamic agencies, and says they are formed from igneous rocks
by crushing.

The writers have examined several localities where the schists
occur, and where their relationship with accompanying rocks is
clear. The principal ones are four in number: one about two
miles southwest of Healdsburg, Sonoma county, one at Camp
Meeker, Sonoma county, one mentioned by Van Hise in the
northern end of Calaveras Valley, Alameda county, and one on
Tiburon Peninsula, Marin county, in which the lawsonite
described by Ransome‘ occurs. Besides these, many smaller
exposures have been studied between Healdsburg and San Luis
Obispo county, and especially in the region around the bay of
San Francisco.

™ The Geology of Angel Island,’ F. LesLre RANsoME, Bull. Dept. of Geod,
Univ. of Calif., Vol. I, No. 7, p. 211.

? Loc. cit., p. 491.

3“ Metamorphism of Rocks and Rock Flowage,” C.R. VAN Hise, Bull. Geol. Soc.
Am., Vol. IX, p. 313.

4“On Lawsonite, a New Rock-Forming Mineral from the Tiburon Peninsula,
Marin county, California,’ F. LESLIE RANSOME, Bull. Dept. Geol., Univ. of Calif,
Volo Now 10; ip.63'tr

740 EDWARD H. NUTTER, WILLIAM B. BARBER

Flealdsburg.—At Healdsburg the schist area is nearly a mile
wide and more than four miles long. The thickness of the schist
is several hundred feet, though no exact measurements have
been made. From about half a mile south of the Junction
Schoolhouse, which is two miles southwest of Healdsburg, the
schists form a range of hills which extend ina general north-
westerly direction for several miles. In the southern end of this
area the schists are in contact with serpentine, and overlie a
boss or laccolite of it. The contact is clear and unmistakable,
as the rocks stand up above the soil, and hand specimens may
be secured which show the parting between the schist and ser-
pentine in a single fragment. In the serpentine boss is an
irregular mass of gabbro—possibly a result of magmatic differ-
entiation. In addition there are at least three serpentinized
dikes in the schist area northwest of the Junction Schoolhouse,
and also a small outcrop of diabase.

With the exception of one place where it grades into shale,
the schist is entirely crystalline, and is composed mainly of
glaucophane, actinolite, garnet, epidote, and various light-colored
micas. Some layers are very quartzose and are composed
mainly of quartz, glaucophane, garnet, epidote and a little white
mica. The schists vary much in texture and mineral composi-
tion, but are easily recognized, as there are no other rocks like
them in this locality.

South of the Junction Schoolhouse a fragment of actinolite
schist was found in the serpentine, and its plane of schistosity
makes a large angle with the planes of neighboring masses of
schist. It is clearly an inclusion and points definitely to the
serpentine being intrusive in and younger than the schist.
Ransome, in his paper on the geology of Angel Island, men-
tions similar inclusions in the serpentine there."

About a quarter of a mile southwest of the Junction School-
house the schists appear to be overlain unconformably by Golden
Gate or Franciscan? sandstones. This relation is not entirely

‘Loc. cit., p. 225.

2, As the Franciscan or Golden Gate rocks are unfossiliferous, their identification
away from the type localities is based on lithologic features and field relations, and is
consequently uncertain.

a ee

GLAUCOPHANE AND ASSOCIATED SCHISTS 741

certain, however, as the rocks may have been faulted into their
present position. About two miles north of the Junction School-
house, on the eastern flanks of the hills, are jaspers intruded by a
serpentinized. dike which has apparently had but little meta-
morphic effect.

In a hillside cut on the Healdsburg road about half a mile
east of the Schoolhouse, there is an excellent gradation from
slightly altered shale to entirely crystalline glaucophane and
actinolite schist. Specimens can be collected showing every
degree of alteration. Thin sections of the least altered shale
show incipient development of glaucophane in crystalline tufts
and radiate aggregates.

Camp Meeker— At Camp Meeker, Sonoma county, about
twenty-five miles southwest of Healdsburg, glaucophane and
actinolite schists are developed over an area which appears to
be at least a mile wide and several miles long, though the limits
were not determined.

Intrusive in the schist are small dikes of a pyroxene rock which
are themselves somewhat schistose, and have glaucophane, actin-
olite, chlorite and mica developed in them. On each side of
these dikes the schist is of the normal glaucophane type and
contains glaucophane, actinolite, garnets and white mica. North-
east of the schist area is a mass of serpentine, but at no point
examined was the relation between the two rocks clearly shown.

About half a mile north of Camp Meeker there is a gradation
from schist to shale in a distance of about three hundred feet.
The shale ts hard and wrinkled, and contains some secondary mica,
but no glaucophane or actinolite.

Tiburon.—Vhe glaucophane schist area on the Tiburon penin-
sula is about two hundred yards wide, and extends more or less
continuously from Tiburon to north of Reed’s station. Northeast
of the schist is serpentine which caps the hills; and southwest are
sandstones and shales. Near Reed’s station, however, there isa
small area of the sandstone and shales, which, on the surface,
lies between the schist and the serpentine, and is in contact with
the serpentine. At the immediate contact the shale is somewhat
hardened but is unaltered at a distance of three feet.

ZAZ EDWARD H, NUTTER, WILLIAM B. BARBER

Calaveras Valley.—I(n the northern end of Calaveras Valley,
northwest of Mount Hamilton, in Alameda county, there is an
area of glaucophane and related schists, which can best be studied
in the canyon of the Arroyo Hondo. There are two series of
glaucophane-bearing rocks in this locality; one a massive crys-
talline rock, one facies of which is an eclogite containing, princi-
pally, garnet, omphacite, glaucophane and actinolite. The other
facies is a medium grained, light colored, banded rock composed
principally of quartz, glaucophane, garnet, and white mica. The
garnets are in the form of very small crystals which are included
in the quartz and glaucophane.

The other series of glaucophane-bearing rocks overlies this
massive one unconformably, and consists of thin bedded sandy
shales having a vertical dip and a northwesterly strike. They
are much contorted, however, and are hard and schistose in places.
These beds are unfossiliferous, but are overlain unconformably by
Lower Miocene sediments. They resemble toa large extent the
sandy shales of the Golden Gate or Franciscan series, and it seems
probable that they belong with those rocks. Glaucophane is
developed in them irregularly, one bed being blue with it, while
the adjoining one oneither side may contain but little glaucophane.
White mica is developed in these beds in many places, and narrow
quartz veins are common.

In the southern end of the canyon are masses of serpentine,
but the contacts between them and the adjoining rocks are not
exposed.

In contact with the massive banded rocks is a hard, heavy,
compact, greenish rock, with bands and stripes of glaucophane
plentifully distributed through it. It may be an altered serpen-
tine, but thin sections of it show no definite minerals. It is
apparently a dike intrusive in the banded rock, but it has probably
been altered since, and may have been subjected to the same
agencies that produced the banded rock.

Conclusions. — No one explanation seems to satisfactorily ac-
count for the many different aspects and occurrences of the
glaucophane and related schists. That there has been some
development of glaucophane at the contact of basic igneous

GLAUCOPHANE AND ASSOCIATED SCHISTS 743

masses seems certain. It seems improbable, however, that the
main portion of the normal glaucophane and actinolite schists is
a-result of contact action. That the schists have not resulted
from contact action by peridotite masses seems probable, for at
some points the same masses have certainly produced but slight
alteration in adjoining sandstones and shales, and the thickness
and character of the schists is such that they could only have
been produced by metamorphosing agents acting ona large scale.
It seems difficult to believe, also, that the schists could be formed
by serpentine dikes which are smaller than the schist masses
themselves. Besides, the inclusions of schist in the Healdsburg
and Angel Island serpentines render it almost certain that the
schists are the older of the two rocks. In addition, the evidence
points to the massive glaucophane rocks and normal schists being
older than the Golden Gate or Franciscan series of rocks, for the
schists are unconformably beneath what appears to be the Golden
Gate or Franciscan rocks in the Calaveras Valley, and they prob-
ably have similar relations at Healdsburg. Finally, serpentinized
dikes are frequently found intrusive in Golden Gate or Franciscan
rocks, while at Mount Diablo,’ near Gilroy,? andin San Luis Obispo
county,’ there are serpentine dikes intrusive in the Knoxville
beds. This would of course make the dikes younger than the
schists, if the schists are older than the Golden Gate or Franciscan
series,

In some cases the schists have been formed directly out of
sedimentary rocks, and probably in some cases out of tuffs or
other igneous material. The writers have observed cases in
which basic igneous dikes have had glaucophane and other sec-
ondary minerals developed in them, and have become more or
less schistose.

It can hardly be doubted that glaucophane schists have
been developed in rocks of different ages, and older than the
Knoxville. It seems probable, also, that there is a series of

tLoc. cit., p. 390.
Communicated by Dr. J. P. Smith, Stanford University.

3“ The Stratigraphy of the California Coast Ranges,’ H. W. FAIRBANKS, JOUR.
GEOL., Vol III, p. 428.

744 EDWARD H. NUTTER, WILLIAM B. BARBER

glaucophane schists older than the Golden Gate or Franciscan,
and the possibility suggests itself that these may be but isolated
outcrops of extensive masses which underlie the Coast Ranges.
The only hypothesis that seems to satisfactorily explain the
occurrences of these rocks is that they are the result of dynamic
agencies, and may or may not be the products of widespread

regional metamorphism.
Epwarp Hoir NUutTTeErR,
WILLIAM BourRTON BARBER.
STANFORD UNIVERSITY, CALIFORNIA,
May, 1902.

tee GEOVOGIG RELATIONS OF THE’ HUMAN
RELICS OF LANSING, KANSAS.

UNDER the title “A Fossil Man from Kansas,” Professor Wil-
liston announced in Sczence of August 1, the discovery of human
remains in alluvium near the mouth of a ravine opening on the
flood plain of the Missouri river near Lansing, Kansas. He gave
a careful description of the circumstances of the discovery, of
the nature and condition of the skeleton, and of the enveloping
deposit. He confidently excluded all forms of intrusion and of
burial by creeping or sliding, attested fully the true fossil nature
of the remains, and referred them to that stage of the postglacial
period when the Missouri river was running forty or fifty feet
higher than now.

Previous to this there had been references to the discovery in
the press, which had attracted the attention of Mr. M. C. Long,
curator of the museum of Kansas City, who visited the locality,
secured as many of the bones as practicable, brought the matter
to the attention of neighboring scientists, and through them to
the scientific world.

In Sceence of August 29, under the title, ‘‘Man in Kansas
During the Iowan Stage of the Glacial Period,’’ Mr. Warren
Upham gave a brief statement of his observations and conclusions
based ona visit to the locality on August 9, in company with
Professors Winchell, Williston, Haworth, Mr. Long and others.
Mr. Upham regarded the overlying deposit as loess of the Iowan
age, and concluded that the skeleton had been ‘‘entombed at the
beginning of the loess deposition, which would refer it to the
Iowan stage of the glacial period, long after the ice sheet had
receded from Missouri and Kansas, but while it still enveloped
northern Iowa and nearly all of Wisconsin and Minnesota.”

In the American Geologist for September, he presented the
subject with greater fullness under the title, ‘‘ Man in the Ice Age
at Lansing, Kansas, and Little Falls, Minnesota.” As before,
the inhumation was referred to the Iowan stage of glaciation,

745

740 TC CLLAMBLRIELN:

comparison was made with other human relics regarded as dating
from the glacial period, and estimates in years of the duration of
the several glacial stages were added.

In the same number of the American Geologist, Proicssor Win-
chell commented at length editorially upon the Lansing skeleton.
He referred with implied approval to the article of Mr. Upham,

Fic. 1.—Side view of skull and femur found in the tunnel. From a photograph
furnished by Mr. M. C. Long.

supplied additional information relative to the history of the dis-
covery, to the deposit embracing the relics, and to the nature and
condition of these. He regarded the main material penetrated
by the tunnel as common loess, and located the skeleton in the
unstratified limestone débris that lies below it. ‘It is hence pre-
loessian, but probably not much older that the loess.” He
discussed at some length the age and relations of the loess, and
concluded: “It will require, therefore, considerable further and
careful examination of the loess sheets of Iowa, and of their
relations to the till-sheets, as well as the marginal features of the
till-sheets themselves, to enable any one to fix with any certainty

HUMAN RELICS OF LANSING, KANSAS 747

the age of the Lansing skeleton more exactly than is above indi-
cated. That it dates from glacial time, at some remote point in
the complex history of that age, is about all that can be affirmed
from the present state of knowledge of the drift deposits.”’

On September 20 the locality was visited by Professors
Samuel Calvin, W. H. Holmes, Erasmus Haworth, R. D. Salis-
bury, .W: €.tioad; Dr.G: A. Dorsey, Messrs. M. C:\Long, F.R.-
Feitshaus, Martin, R. T. Chamberlin, and the writer. This visit
was made at the request of Dr. Haworth and other geologists.
A second visit was made on October 26 at the request of Pro-
fessor Holmes and Mr. Gerard Fowke to inspect the excavations
which the latter had made under the direction of the former.
Mix eone, Mr S.J: Hare,-and Dr. Haworth joined in. this
inspection. The Messrs. Concannon tendered all necessary
privileges, as well as aid and hospitality. The following dis-
cussion is based on the data collected in these visits.

PRELIMINARY CONSIDERATIONS.

While the development of the science of river action in most
of its phases is one of the gratifying achievements of recent
decades, it is still to be confessed that a certain few of its aspects
are among the laggard features of our science, and, as it happens,
these are the ones most critically involved in the interpretation
of the Lansing remains. It may not be amiss, therefore, at the
outset to consider academically these special phases of fluvial
action so far as essential to the present discussion.

1. Scour-and-fill—One of these scantily appreciated subjects
is the great depth and important function of scour-and-fill in
certain of our large rivers. In this action both erosive and
depositional work proceed szmultanecously. It is well recognized
that erosion and deposition may take place simultaneously in the
stream bed and upon the flood plain, but the great depths and
wide extent to which certain river bottoms are scoured out and
promptly refilled is not always realized, nor the quick and con.
stant reversals of this action. This is true especially of powerful
rivers that flow upon a deep bed of loose material, as is the case
with most of the large rivers whose bottoms were built up by

748 TC. CHAMBERLIN:

glacio-fluvial deposits during the ice age. The great examples
are the larger members of the upper Mississippian system, and
pre-eminent among these, the Missouri river whose bottom deposit
is mainly sand and silt of an unusually mobile type. The vain
struggle of the United States engineers to restrain the destructive
shiftings of this river within bounds amenable to navigation and
to permanent improvement on its banks, has brought out data
which amply illustrate this profound instability, but this can only
be fully appreciated by a detailed study of the reports of the
chief of engineers:s Mr: LE: Cooley, in his report tor 1570;
(p. 1066), makes the following among many other pertinent
statements:

‘To understand the difficult nature of the problem presented
here [Eastport bend, on the Missouri river much above Lans-
ing, but where the conditions are not essentially different |, it is
necessary to consider that at high-water, the banks are under
water to a depth of three or four feet, and the current velocity
is as great as seven or eight miles an hour. The erosion of the
banks for several years past has been at the rate of about 1,100
feet per annum. When this was stopped by our revetment, a
tremendous scour was set up, carrying the bed of the river thirty
or forty feet below its normal position; in fact, the scour
undoubtedly extended to the solid rock underlying the valley.”
And again (loc. cit., p. 1071), “‘In many of the borings which
have been made here, indurated clay balls with vegetable matter
covered with a coating of sand, along with a motley collection
of gravel stones, are found within a short distance of permanent
strata. A precisely similar collection containing gumbo balls in
a soft state was dredged from sixty feet depth at the works.
These balls are from cutting banks, and the proof is conclusive
that since the river has been running in silt banks as at present,
scour has occasionally, at least, reached permanent strata at
seventy to ninety feet depth.”

Mr. Concannon informed me that eleven years ago the

«Professor Todd has called attention to some of these remarkable facts in his
bulletin on the ‘‘Moraines of Southeastern Dakota and their Attendant Deposits,”
Bull. U.S. Geol. Survey, No. 158, pp. 150, 151.

FEET ABOVE SEA

HUMAN RELICS OF LANSING, KANSAS 749

engineers found a depth of water of ninety feet in the Missouri
at a point about a quarter of a mile from his house, in what was
then the channel of the river, but which is now abandoned and
filled so that water covers the spot only at the highest stages of
the river. Until about eight years ago the course of the river
lay near the mouth of the valley in question, but is now diverted
to the opposite side of the bottoms.

PIER,
PIER
PIER

a ye SU I NBS ee he mrmye PGA?

BAY Sree

oe ae lie 7
eae | yy
f / Jf

ad

=

bed rock mgs
5 Carponiferous 4 limestone

Fic. 2,—Diagram of the changes in the bottom of the Missouri river at Blair
Bridge in 1883, as recorded by Engineer E. Gerber. Figure taken from Todd’s Bu/-
letin United States Geological Survey, No. 158, p. 15].

An accurate demonstration of the extent and rapidity of
bottom changes is furnished by the accompanying diagram-
matic record of the soundings at the Blair bridge, Nebraska, at
the intervals indicated, in the year 1883, quoted by Todd on the
authontyrot Mr “Gerber, assistant engineer (h> i Ga MV.
railroad.

An inspection of this will show that a skeleton might have
been deposited on the surface of the Carboniferous rock bottom,
much as in the case of the skeleton at Lansing, on the 28th of

750 T. C. CHAMBERLIN

July, 1883, and have been buried in alluvium as deeply as the
Lansing skeleton by August 18, only twenty-one days later.
Without doubt, within a few years it would be covered by sixty
feet of alluvium through the migration of the channel of the
river.

2. The prevalence of this profound reworking.— To illustrate
how fully and effectually the whole of the bottoms of the Mis-
souri river in this region are involved in its meanders and their
shiftings, and how its bordering bluffs are being forced to retire
by the impingement of the currents at its bends, a reduced copy
of the United States Engineers’ map is here introduced (Fig. 3),
the section being about forty miles north of the locality in ques-
tion, but representative of the conditions in all this portion of
the river. It will be noted that practically the whole valley
bottom is involved in the migrating loops, and that every part of
its silt bed is liable to be disturbed again and again by scour
and redeposit; indeed, it is probable that this has happened
repeatedly to many portions, if not to most portions of the
alluvial filling. It is perhaps not greatly beyond the facts to
regard the whole bottom filling as being shifted, step by step
down stream by successive scour and fill. ‘This is more espe-
cially true of the borders of the bottom filling next the bluffs
where the arrest and turn-about of the powerful stream gives
the greatest rotatory and deep-disturbing effects.

3. The absence of the great Dakota system of terraces—In the
widening of the bottoms thus still in progress doubtless lies the
reason why so few distinct remnants of the grand systems of
glacial terraces and glacio-fluvial deposits of Dakota, described
by Todd,? are found in this lower portion of the Missouri river.
It is probable that the whole tract once occupied by these, and
more besides, is now embraced by this widened, and still widen-
ing, zone of lateral encroachment. This is the less remarkable
when we recall that the Missouri river was formed by the union
of many preglacial streams of various connections whose lower
courses were blocked up by the ice invasion so that they were

* Loc. cit. pp. 128-140. The general nature of these is given in a later portion of
the present paper.

a

“eh
SSR AzON \

CE Ki he Lae

7 ne
Miey, Wi de ZY bt COLE.

MAP

oe the

MISSOURI RIVER

in dhe vicinity of

REFERENCE. HUN'TOWN:

Peanut nuopened

tomptehed

PHRRaE kor nar

RD MONT

e

WATHENA

AM, PHOTO-LATME.CaNY (ZEDMENCY PHCTE RE

Fic. 3.—Reproduction of map of the Missouri bottoms at St. Joseph, Mo. From
the Annual Report of the Chief of Engineers for 1879.

752 T. C. CHAMBERLIN

forced to unite and flow along the ice border. At first the newly
assembled streams flowed either in the valleys of the smaller
streams that entered into the combination, or in a new trench
cut by the new stream across the cols between the united valleys.
Thus at first it would not as a rule come into possession of a
valley bottom of capacity adequate to the united floods, and in
normal adjustment to them, and hence found little opportunity
to make deposits. To the limited extent provided, its burden of
glacial detritus was thrown down in these new and inadequate
valleys, and as a natural consequence, it has been removed in the
later process of working out an adequate valley and a suitable
adjustment. The river is still engaged in making this adjustment.

4. The significance of valley adjustment.—lf a great change is
brought about in the drainage system of a region, such as the
creation of the Missouri river by the junction of numerous ante-
cedent rivers, and a newchannel is developed to fit the new river,
there at once arises the question whether the existing features of
the valleys tributary to the new channel belong to the old or
the new régime. In part they usually belong to both, and it
becomes necessary to discriminate between these parts. This
may be done by the study of their adjustments, a method espe-
cially applicable to small tributaries that have no permanent
streams, as in the present case. The tributaries of the old
system were adjusted to the old channel and cannot be presumed
to be adjusted to the new channel, except in the rare case of exact
coincidence of the old and the new channels. Inrelation to the
new system, inherited tributaries usually present either the buried
or the hanging type, or else they have become refashioned into
adjustment to the new system. Such refashioning affects espe-
cially the mouths of tributaries. It often so happens therefore
that refashioned configuration in conformity to the new system
may dominate the mouth of a tributary, while its upper portions
retain almost wholly the old configuration. These facts warn us
of the danger of assiguing great antiquity to fluvial deposits in
the znumediate mouths of tributary valleys if these valleys are
adjusted to the present river or the present bottoms; especially
is this true if the tributary is scarcely more than a ravine, and

HUMAN RELICS OF LANSING, KANSAS 75

ios)

its erosion and deposition are intimately conditioned by its rela-
tion to the river. In all such cases there is a strong presump-
tion that the erosions and depositions at the mouth of such a
tributary, such especially as have brought it into adjustment to
the present and to the recent stages of the river, were contem-
poraneous with those stages and not accidental inheritances.

5. Meandering as a cause of alternate erosion and deposition.—
A meandering river with a deep, readily-shifted, bottom-filling
of the Missouri type imposes upon its tributary valleys alternate
stages of excavation and filling. These result (1) from the action
of the aggressive bends of the river loops against the mouths of
the tributaries, and (2); the replacement of these, after a time,
by the flood-plain peninsulas that lie within the loops. More
specifically, it is the alternate cutting of the stream itself, work-
ing hard against and under the mouth of the tributary valley,
followed by the building up of the river’s higher flood-plain
across the mouth of the valley. The first causes the waters of
the adjusted tributary to erode; the second to make deposits in
the mouth of the tributary; for in the first stage the axis of the
tributary opens out on the river itself, which may be twenty or
thirty feet, or more, lower than the upper flood-plain, and hence
the tributary then has its lowest and best opportunity to discharge
its waters and their detrital burden. Besides this, the river itself,
while in this aggressive attitude, sweeps into the mouth of the
tributary in its flood stages and aids in its excavation, and the
rushing by of the river’s strong current drags out by friction, on
the principle of draught, the waters of the tributary, and, by
acceleration, aids their excavating action. It is at this stage pre-
eminently that the tributaries cut down their valleys into adjust-
ment with the main stream bed. On the other hand, when the
active impinging bend of the river has shifted elsewhere, and in
its stead a flood-plain is being built up across the mouth of the
tributary the drainage of the latter is checked, and if the tribu-
tary be small and its waters incompetent in comparison with the
flood-plain aggradation of the river, the valley mouth will be
filled to a height corresponding to thatof the highest flood-plain.
Now, the difference between low water and high water for the

754 T. C. CHAMBERLIN

Missouri river is given by Abbott as twenty feet at St. Joseph,
above Lansing, and as thirty-five feet at its mouth; its extreme
range is somewhat greater than this.

Further, if the mouth of the tributary be blocked by the upper
flood-plain beyond the time of the latter’s growth the wash from
the tributary will build a delta, or fan, upon it, and this further
growth will continue until the waters from the tributary valley
have built up a suitable gradient for themselves across the flood-
plain to the river. This only holds good in valleys of incompe-
tent drainage which cannot cut and maintain a trench for them-
selves. If the tributary valley has a large, competent stream it
will maintain a channel-way across the flood-plain to the river, and
less aggradation will result from the shifting of the meanders,
but that is not the case in hand.

If excuse for this academic statement is needed it is found
in its special application to the case in hand; for either
action of the kind just set forth is to be accepted as an elucida-
tion of the case, as in the preferred interpretation that follows,
or it is to be shown incompetent for such elucidation before we
permit ourselves to go back of this action to earlier agencies. It
is a vital principle of good practice that the agencies and phe-
nomena nearest at hand be first considered, and, if the case
requires, be eliminated, before recourse is had to more remote
agencies. This is peculiarly true when, as in this case, the
agencies closest at hand in time have quite certainly swept away
the most of a more ancient record in making their own.

THE SPECIAL CASE.

The topographic environment of the relic-bearing deposit.—The site
of the human remains is at the bottom of a small, short, rather
steep-sided valley opening out on the flood-plain of the Missouri
river. More specifically, the valley is less than a mile long, and
less than half a mile wide, measured from crest to crest, and is
about 160 feet deep at its mouth. The slopes on either hand are
rather steep and nearly meet at a rather sharp angle in the axis
of the valley, except that this is modified by the channel or dry
run which forms narrow bottoms and little bluffs near the mouth,

HUMAN RELICS OF LANSING, KANSAS 755

Fic. 4.—Topographic map of the tributary valley at Concannon’s. From a sketch
by Professor W. H. Holmes.

750 T. C. CHAMBERLIN

for the valley is not occupied by apermanent stream. The slope
on the southward side is about as steep on the average as can be
profitably cultivated; that on the north side is steeper, so that
while the upper slope is cultivated the lower slope is left to nat-
ural growth and is partially occupied by quarries. On this steeper
portion there are some small, vague, bench-like lines of uncertain
interpretation; quite likely they are structural features depend-
ent on the alternation of the more and the less resistant layers of
the underlying strata. About twenty-five feet from the base of
the slope there is an ill-defined bench that seems to be made up

Fic. 5.—Section through the mouth of the tributary valley and the ridges on the

north and south. Merely diagrammatic.

of lodgment matter adjusted toa former higher axis of the val-
ley. There is a correspondingly vague bench on the opposite side.
The ridges are composed of Carboniferous limestone, mantled by
Pleistocene deposits (Fig.5). The glacial drift is represented by
some bowlders and smaller rubbish, but it is so scant and patchy
as to be negligible as an element of the topography. The upland
surface is mantled with loess and loam, the main portion of which
is probably referable to the Iowan stage. The lower slopes are
covered by wash from the uplands and by the skeleton-enclosing
deposit which lies near the axis of the tributary valley and con-
stitutes the vague benches above mentioned.

The back country is strongly rolling, the valleys fairly sharp,
and their debouchures into the Missouri bottoms abrupt but
well adjusted, and in their adjustments they represent the sev-
eral normal types as well as several different stages. The
bottoms of the Missouri are sharply defined by bluff faces. This
is particularly so where the little valley in question joins it.
The Missouri here runs southeastward, and the ridges bounding

HUMAN RELICS OF LANSING, KANSAS 757

the tributary valley on either side have been abruptly truncated
by the waters of the Missouri and present a sharp talus face
toward the bottoms. The recency of this face is a declared
feature and is significant. Where not occupied by rock, the
slope is formed of talus marked by slides and slump terraces so
new as still to preserve their distinctive features. A very per-
sistent slide terrace runs along the base of the south ridge at

Fic. 6.—View looking northward across the mouth of the tributary valley, show-
ing Concannon’s house at the left, and the truncated slope under it, the mouth of the

valley just beyond, and in the center the north bluff with its truncated face overlook-
ing the Missouri bottoms, on the edge of which the railroad lies. The bluff is about
160 feet high.

about the horizon of the skeleton’s burial, ending nearly opposite
it, and about ten rods distant. It is not intended here to
suggest an immediate connection between this slide action and
the burial of the relics, but merely to show the recency of the
Missouri’s work across the mouth of the tributary valley and
within a few rods of the critical locality. This propinquity is
brought into greater emphasis by noting that if a line be drawn
from the crest of the talus slope of the north bluff to the crest

758 T. C. CHAMBERLIN

of the talus slope of the south bluff, it will run back of the
skeleton’s ‘site.: The significance off this close relation jes. in
the alternate depositional and aggradational work presumably
done by the Missouri river at and in the mouth of the valley
when it was truncating the adjacent bluffs on the one hand, and
forming the adjacent bottoms on the other, in accordance with
the principles of action outlined above. The accompanying
contour map and photographs (Figs. 4 and 6-11), with their
explanations, make these relations more definite.

The precise locality of the relics is more closely defined by
an additional feature. A deep ravine starts near the crest of _
the ridge bounding the tributary valley on the south, and running
nearly parallel with the truncated face overlooking the Mis-
souri bottoms, joins the axis of the valley a few rods west of
Concannon’s. house (see Fig, 7). East of this ravines there
was doubtless once a round-back ridge of the usual erosion type
with another ravine still to the eastward, but the encroachment
of the Missouri has cut away the eastern half and substituted a
steep ‘talus. slope.) There now ‘remains a sharp-edged spur
descending toward the axis of the tributary valley, with a talus
face on the side next the Missouri bottom, and a more gentle,
yet rather steep slope to the ravine on the other side. Follow-
ing down this sharp-edged spur, it is found to flatten somewhat
for a few rods at about sixty feet above the bottom of the
valley, much as though the flattened portion might be a remnant
of a small terrace, structural or otherwise. Farther on, this
breaks down, with rock exposure, for about ten feet to another
flattening for another few rods. On this lower shoulder Mr.
Concannon’s house stands, beyond which the spur ends in a
sharp descent of about thirty feet to the dry run of the valley.
On the west side of the house the surface descends more gently
to the ravine above described. It is under this westward slope,
about one hundred feet back from the edge of the talus slope
facing the Missouri bottoms, and about seventy feet southward
from the little bluff facing the dry run of the valley, that the
human remains were found buried about twenty feet deep.
These details are given with some tediousness because they bear

HUMAN RELICS OF LANSING, KANSAS 759

upon the interpretation of the time and mode of deposition of
the formation embracing the relics.

As already stated, the tributary valley is not occupied by a
constant stream, but by periodic run-off. The channel at present
is in a slightly aggraded and apparently still aggrading stage.
It opens out upon the Missouri bottoms about two hundred feet

Fic. 7.—View of Concannon’s house and environment seen from the south-south-
west. In the foreground and center is the ravine leading down from the south,
described in the text. The locality of the skeleton is nearly under the small white

spot near the dark clump of trees on the slope at the left of the house. The ravine
joins the tributary valley just at the left of this and the latter joins the Missouri bot-
toms in front of the house. The Missouri bottoms stretch across the upper part of
the view, with the river (in its new course) and the opposite bluff in the extreme back-
ground.

from the locality of the relics, with perfect adjustment, and its
recent deposits were slightly fanned out upon the bottoms of
the main valley on our first visit, but had been largely washed
onward by the rain that intervened before the second visit, illus-
trating the nature of the present adjustment. The depth of the

760 LC. “CHAU BERIEEN:

aggradation deposit is unknown to me, but it is probably not
many feet, as the aggradation stage has but recently been
inaugurated by the detour of the river. On the north side the
spur next the Missouri bottoms grades down to this lower grada-

Fic. 8.—View from near the mouth of the tunnel looking northeastward across
the bottom of the tributary valley, showing the gradation of the footslope of the north
bluff into the Missouri bottoms seen at the right.

tion plain and the combination of lower slope and present
bottom deposits is similar to that of an earlier date on the south
side which contains the human bones (Fig. 8.) |
The present aggrading washes have made a little bottom in
the lower twenty rods of the valley, with meanders and little

HUMAN RELICS OF LANSING, KANSAS 761

bluffs where the loops bear against the older deposits of the
valley. It is in the face of the little bluff on the south side,
and about four feet above the valley bottom, that the mouth of
the tunnel that disclosed the human remains is located. The
base of the tunnel at its mouth is ten or twelve feet above the

Fig. 9.—View in the mouth of the tributary valley looking out upon the Missouri
bottoms and showing the entrance to the tunnel at the extreme right. The material
from the tunnel modifies the natural bottom, as seen in the foreground.

adjacent Missouri bottoms (Fig. 9). The lower four feet of the
little bluff is formed of a thick bed of Carboniferous limestone;
above this there is shale. The tunnel was started just above this
limestone and driven back on its gently rising surface. It was
carried by the Concannons seventy-two feet back from the face

762 DAC. CLAVIER RAEN

of the bluff, and at its inner end its base is twenty-one feet five
inches below the surface, an air-shaft permitting a tape-line meas-
urement. :

The relics found in excavating the tunnel represent an adult
who had lost several teeth and a child whose teething stage,
according to Professor Williston, implies an age of about nine

Fic. 10.—Front view of the skeleton of the adult and two of the associated bones,
with the fragment of the child’s jaw in the foreground. From a photograph furnished
by Mr. M. C. Long.

years. The former is represented by a skull, femur, and other
bones; the latter only by a fragment of a jaw (Fig. 10.) The
bones of the adult are said to have been found near the inner end
of the tunnel, and between one and two feet above its base. They
were disarranged and at slightly different depths, but it is suffi-
cient for present purposes to locate them at seventy feet from
the entrance and twenty feet from the surface. The fragment
of the child’s jaw was found about sixty feet from the entrance
and within a foot of the bottom of the tunnel. These state-
ments relative to the discovery of the bones rest upon the testt-

HUMAN RELICS OF LANSING, KANSAS 763

mony of Michael T. and Joseph F. Concannon, who dug the
tunnel. There is no ground to question their authenticity.

The associated deposit. At the mouth of the tunnel the lower
three or four feet of the deposit is composed mainly of lime-
stone fragments and earthy débris, a part of the latter seeming
to come from the Carboniferous beds, a part from the glacial
drift or the loess, and a part from the river and valley wash ; in
short, a rather heterogeneous mixture. Some parts are highly
oxidized and iron-stained and some parts are relatively fresh and
calcareous. At about three feet above the floor on the western
side there is a definite layer of dark, highly calcareous clay less
than three inches thick, but it does not appear on the opposite
side. It is thinner in the inner portion of the tunnel, where the
cross cut of Mr. Fowke shows that it rises on the west side and
pinches out irregularly within a few feet. The upper part of the
deposit at the entrance is a mottled silt of loess-loam aspect,
containing occasional stony fragments. Its response to acid is
irregular, sometimes giving no obvious effervescence, sometimes
a feeble action, and sometimes a prompt and marked response.
Sometimes the action is concentrated in definite spots, as though
it came from a bit of limestone. The action is not that charac-
teristic of typical loess. Even in the top of the tunnel some
limestone fragments were seen seven or eight feet from its base.
Even in the inner end of the tunnel the silt is notably mottled,
in part irregularly, and in part in bands, more or less horizontal,
as though controlled by stratification, though the staining is
probably secondary. Acid tests indicated that calcareous matter
is present, but that it is not abundant.

These observations were made on the tunnel as seen on our
first visit. Under the direction of Professor Holmes, Mr. Gerard
Fowke later made a series of supplementary excavations in dif-
ferent directions to develop the formation further and secure
additional fossils. <A full statement of the results will doubtless
be given in Professor Holmes’s report. He has kindly permitted
me to use such of the data thus gathered as are serviceable in
the geologic determinations. Without entering upon precise
details, it will suffice here to say that the tunnel was extended

704 PNG) CHAMBERLIN

southward until the rising of the Carboniferous beds in the bot-
tom made further extension in that direction unpromising. Only
a few feet beyond the end of the original tunnel Carboniferous
shale was found overlying the heavy stratum of limestone, and
the surface of this rose as though the foot slope of the ridge
had been reached. The correctness of this inference is scarcely
open to question as the whole environment supports it so
strongly that it had been anticipated. The ease with which this
shale was eroded, compared with the underlying limestone,
readily explains the flat limestone surface on which the tunnel
was run.

In an excavation on the west side of the tunnel, a shallow
trench was found in the upper surface of the limestone running
nearly parallel with the tunnel and also parallel to the axis of
the adjacent ravine. With little doubt this trench was the axis
of the ravine in the erosion stage just preceding the filling up
of the ravine by the relic-bearing deposit. This further aids in
explaining the nearly horizontal, but slightly rising, base of the
tunnel, since it locates it alongside the axis of the ravine on a
resistant bed (see Fig. 13.)

An offset tunnel at right angles to the original tunnel was
run eastward eleven feet from the place of the adult skeleton.
It developed about four feet of disturbed shale and mixed débris
in its base, the vague structure lines of which dipped eastward
irregularly. It had the appearance of a talus slump that had
crept down the slope of the adjacent rock surface, and warped
and slightly tilted itself backwards according to a common habit
of such masses. This doubtless took place before the upper
deposit was laid upon it and while yet the ravine was open, 7. e.,
about the close of the erosion stage. In the east end of this
offset, the silty formation has been slightly fissured along a nom-
ber of lines by tensional action and the little crevices filled with
a grayish-white soft deposit that effervesced very promptly with
acid, implying calcium carbonate. The riveing tension probably
came from the tendency of the mass to creep on the underlying
rock surface, since this rises to the east and so furnishes a slop-
ing base of shale which arrests the waters descending through

HUMAN RELICS OF LANSING, KANSAS 765

the more porous mass above and which, thus becoming wet, pre-
sents an unctuous slippery surface favorable to creep.

On the west side of the tunnel the excavation was carried
from near the point where the fragment of the child’s jaw was
found westward at right angles and was met by an open cut from

Fic. 11.—View from the westward showing the trench dug by Mr. Fowke from
the ravine toward the tunnel with which it connects below. The original tunnel runs
from left to right under the two trees seen beyond the end of the cut. The child’s jaw
was found at the intersection of the cut (extended by tunnel below) with the original
tunnel just at the left of the trees. The adult skeleton was found nearly under the
second light spot to the right of the two trees.

the ravine. This open trench (Figs. 11 and 12) afforded an admi-
rable opportunity to study the constitution and structure of the
whole section from the rock floor to the surface through a depth
of about twenty feet. The definite clayey band found near the
base in the tunnel is here wanting. As noted above, it pinches

706 lf. C. CHAMBERLIN

Fic. 12.—Nearer view of the open cut shown in Fig. 11. The shadow obscures
the larger portion, but the lighted portion on the left shows the absence of definite
stratification, and indicates something of the mottled character of the deposit.

HUMAN RELICS OF LANSING, KANSAS 707

out irregularly a short distance west of the original tunnel. Put-
ting all the facts together, it would seem that this little stratum
was laid down in the axis of the ravine shortly after the stage
of aggradation began. As it scarcely reaches three inches in
depth at its thickest point—-averaging probably less than an
inch—and is very homogeneous and peculiar, as well as very
fresh and calcareous, it was probably formed at a single stage of
inundation. Aside from this there is no distinct stratification or
lamination in the whole section, nor any complete assortment of
the material, The main material is a silt somewhat closely
resembling loess, but unlike it in the particulars already pointed
out. Through this silt, at all heights from the base to the sur-
face, there are dispersed fragments of limestone, shale, and other
débris incompatible with a typical loess deposit. The limestone
fragments were sometimes several inches across. Mr. Fowke,
who gave careful attention to the distribution of this material,
affirms that it was found indifferently at all heights, and I care-
fully verified this by an examination of the walls of the deep
open cut. Small fragments of softened limestone were so abun-
dant in some parts that the walls were mottled with the white
chalky spots made by the spade in mashing and spreading them.
There were also many bits of shale ranging up to an inch in
length, not a few of which had been sufficiently weathered to be
yellowish or brownish. These also occurred high up as well as
low down in the section.

I have said that there was no distinct stratification, lamina-
tion or assortment in the section. There was some aggregation
of the silt and the fragmental material. There were spots where
the shaly and limy débris was sufficiently abundant to lend a
gravelly aspect to the mass, but close inspection showed that it
was not really assorted, laminated, or stratified. The agency of
accumulation had obviously brought relatively more fragmental
débris to these portions, or at least had left relatively more frag-
mental débris in these portions, than in average portions, but the
aggregation did not rise to the grade of typical assortment and
lamination. That it is a wash product seems to me clear, but
not a stream deposit nor a lake deposit, nor any other form of

768 T. C. CHAMBERLIN

purely subaqueous deposition. I should identify it as a typical
aggradation deposit of the ravine and basal-slope type where the
hillside environment was Carboniferous limestone and shale
mantled with loess.

A few pebbles of drift and not a few pieces of charcoal were
found in the section, the latter at different horizons. Many land
shells and some additional bones were also found by Mr. Fowke,
but no unios. These interesting features will doubtless be
described in Professor Holmes’s report.

As had been anticipated, the excavations show that the
extent of the deposit is limited, and that it was penetrated by

—<—— ——_——_ /00 Fe er.

Fic. 13.— Cross-section from ravine at the left to truncated face overlooking the
Missouri bottoms (G) on the right. The section passes through the end of the original
tunnel ( 7") at the place of the adult skeleton (.S). It shows the supposed original
trench of the ravine in the surface of the limestone (4), the shale overlying the lime-
stone, developed in the cistern and in the extension of the tunnel (C), the limestone
blocks of the upper limestone under the house, and the deposit overlying the relics
(2). Vhe line marked 100 feet represents the distance from the place of the skeleton
to the point where the truncated slope begins, not the whole length of the section.

the tunnel nearly or quite at its greatest depth. Rock comes to
the surface just back of the house, and in the excavation for
the rear end of the house, Mr. Concannon informed me that he
reached rock which he thought was of the regular quarry kind.
In sinking for acistern eighteen feet deep on the east side of the
house, he went through about four feet of dirt, then about two
feet of loose limestone blocks, and then about twelve feet of
‘soapstone,’ so hard that he had to blast it. This is undoubt-
edly the Carboniferous shale encountered by Mr. Fowke in the
extension of the tunnel. This makes it clear that the spur on
which the house stands is formed mainly of Carboniferous beds
and is merely mantled with the silt and débris formations. The
accompanying cross-section is drawn approximately to a true

HUMAN RELICS OF LANSING, KANSAS 709

scale (Fig. 13) and shows the probable limitations of the
deposit. The extension of the tunnel shows that it thins to the
southward, while the ravine intercepts it on the west.

The surface configuration is that of a combined basal-slope
and ravine-bottom deposit, z. ¢., of aggradation in the bottom of
the ravine, combined with deposits lodged on the lower slopes
in adjustment to the aggraded bottom. The structure of the
deposit is in keeping with this interpretation. The little layer
of calcareous clay in the tunnel seems to imply deposition in
standing,
probably a back-water deposit. The absence otherwise of definite

or slowly moving water, z. @., a valley-bottom deposit,

stratification or assortment of the material, and its complete resem-
blance to secondary slope accumulations derived jointly from the
loess and the underlying beds seems to require its reference to
aggradational action. Professor Williston found the cast of a
clam shell with attached valves in the angle between the wall
and the roof of the tunnel about seven feet from the base. In
the absence of satisfactory evidences of fluvial action at this
height and in the presence of human relics, this may well be
referred to human agency.

INTERPRETATIONS.

The case is perhaps not an absolutely declared one, and a
wholly unreserved interpretation may not be warranted, but a
very strong balance of evidence seems to point in a specific
direction. Certain things seem to me clear:

1. The deposit is not true original loess. It is a mixture of
loess-like silt, Carboniferous detritus, water-laid clay and other
débris. The Carboniferous detritus was obviously derived from
the adjacent strata, in part by disintegration, in part by wear,
and in part by fracture without much rounding. The loess-like
silt was probably derived in the main by wash from the loess
mantle of the adjacent hills, but in part also by winds from the
Missouri bottoms ; possibly also in part by creep. Its charac-
ter implies that some of the silt was brought to its present posi-
tion without complete leaching, while most portions show
evidences of exposure and weathering. From such differences

172 TEC (CLERANG TET IRIE IONE

of history probably arose the variations in color, texture and
effervescence in response to acid, which were observed. The
material of the one distinctly water-laid layer was probably
derived from the Carboniferous shales at some special stage of
erosion and inundation—some unusual storm and flood, perhaps
—and was deposited without weathering, and remained undis-
turbed except on its borders.

2. The truncated faces of the adjacent Missouri bluffs, and
the numerous slides on these faces, show that the Missouri river
has worked extensively and effectively across the mouth of the
tributary only a few rods from the site of the relics, and that this
has been comparatively recent.

The rather steep slopes of the tributary valley favor the view
that the present fashioning of these is recent. The main excava-
tion of the valley probably dates back to the post-Kansan
erosion interval, and this was perhaps preceded, and perhaps
determined, by a preglacial valley. But the valley, as 77s now
fashioned, is pretty closely adjusted to the Missouri river bottoms
which are features of recent origin, and ¢izs adjustment and the
slopes and deposits involved in it is, by rather strong presumption,
to be connected with the development of the adjacent Missouri
channel. The age of the original valley and of the upland
mantles does not concern us here, unless these lower deposits,
well down in the axis of the valley, and at its junction with the
great river bottoms, are surely inheritances from the older period,
and not adjustment phenomena.

3. Ihe recordof these earlier events is here very impertect.
Even the record of the more recent of the Pleistocene events is
very scant where it should be abundant and decisive if the con-
ditions of preservation had been favorable. In Dakota, where
the Missouri river came into relation with the last stages of gen-
eral glaciation within its basin, there are three great systems of
terraces as worked out by Todd,’ viz.: 1) ‘‘The higher bowldery
terraces,’ varying from 500 feet to 350 feet above the Missouri
and connected with the outer moraine of the Wisconsin stage ;
2) ‘‘The lower bowldery terraces,” varying from 350 feet to
*Topp, Bull. U. S. Geol. Surv., No. 158, pp 128-154.

HUMAN RELICS OF LANSING, KANSAS 771

various lower levels at different points, and connected with the
second moraine of the Wisconsin stage, and 3) a complex sys-

”)

tem of ‘‘silt terraces’’ ranging from 150 feet downward, three
or four of these terraces often occurring at the same locality.
These last have not been traced into physical continuity with
any of the moraines, and doubtless represent in part the very
latest stages of glaciation, and in larger part the postglacial
stages ranging down to very recent times. A reference to Todd’s
descriptions will show that these are not mere strands or slender
benches on the valley sides, but great platforms, sometimes a
mile or two broad. Now all of these three systems, so magnifi-
cently developed in Dakota, should ideally be represented in
some way at the Lansing locality, but we have only the obscure,
sloping shoulders already described, and the little deposit con-
taining the relics. There is no sign that these belong to the
first or second of the Dakota series which are directly connected
with the first and second stages of the Wisconsin glaciation. In
Dakota these terraces are formed of very coarse material, which
gives them the title ‘‘bowldery,”’ and this implies strong cur-
rents fed by glacial débris. Normally, these high bowldery
terraces should graduate down-stream into finer gravels, sands
and silts, all bearing the distinctive marks of their glacio-fluvial
origin. The relic-bearing deposit is not of this type, and is not
overlain by this type. The most natural inference then is that
the train of glacial gravels, sands and silts borne away by the
Missouri waters from the ice edge in the more vigorous stages
of Wisconsin glaciation was carried away from this part of the
Missouri channel before the relic deposits were formed. This is
the more to be supposed because the Missouri has here recently
run hard against the highlands and truncated them, and the
tributary valleys are steep and in this special case, short and
rather sharp. Remnants of the true glacio-fluvia! deposits in
this portion of the Missouri river are rare, and an experienced
Pleistocene geologist familiar with their habit would not expect
to find them in the mouth of so narrow, steep-sided, and steep-
bottomed a tributary as that at Concannon’s. The probable
reason for the scantiness of the glacio-fluvial record in this part

7/2 TC. CHAMBERLIN

of the Missouri valley has been given in the preliminary con-
siderations. If neither of the strong bowldery terraces of so
late a stage of glaciation as the Wisconsin are represented at the
site of the burial, there is but scant ground to assume that the
earlier and much feebler and much more erosible glacio-fluvial
deposits of the Iowan are preserved.

The natural conclusion is, therefore, that the little relic-bear-
ing deposit in the valley at Concannon’s belongs either to the
same class as the silt terraces of Dakota, to which it bears a
measure of resemblance, or to some later stage.

Specific views.— While, as before remarked, the case is per-
haps not a wholly declared one, and an unqualified identification
may not be entirely warranted, the range of tenable interpre-
tation seems to me to lie within narrow limits.

1. Zhe most conservative and the most probable view.— All the
essential facts known to me seem to be explicable on the follow-
ing lines which involve the minimum of action and of assump-
tion, and which appeal only to the natural order of things. The
first stage of essential action is assigned to a time when the chan-
nel of the Missouri river ran immediately past the mouth of the
tributary valley and was higher than now to such an extent as
to be in erosive adjustment with the tributary at the top of the
rather heavy limestone layer which lies just below the tunnel.
It has already been noted that where a strong stream like the
Missouri passes hard by the mouth of such a tributary, two
effective conditions of erosion are supplied. The tributary hasa
low point of discharge and hence a high gradient, and its detri-
tus is immediately swept away by the great river. During this
stage the rock surface under the relic-bearing deposit was devel-
oped by the removal of the shales above, and the lower slopes
adjacent were measurably denuded because the conditions were
favorable to erosion and the shales were easily cut away. After
a stage of erosive adjustment of this kind, a change of relations
was brought about by the diversion of the channel of the Mis-
sourl river to some other portion of the broad valley, attended
by the substitution of a flood plain at the mouth of the tribu-
tary. As the vertical range of water is now twenty feet or more,

HUMAN RELICS OF LANSING, KANSAS Vilas

the building up of a normal upper flood plain that much above
the preceding erosion plain may be assumed. This must have
been accompanied by a filling up of the lower part of the tribu-
tary in like measure. More than this, if the diverted stream in
its new course ran on the opposite side of the bottoms, two
miles away, the tributary might have also built a fan on the
surface of the flood plain, with proportional further aggra-
dation within its mouth. Now this filling up of the axis of
the valley to the amount indicated, changed the condition of
the lower sides of the valley, and these became covered with
lodgment deposits derived from the upper slopes and with silts
blown up from the Missouri bottoms, an action still in effective
operation. Such deposits are the normal result of an effort to
establish a new set of gradients adjusted to a lifted axis. The
deposit resulting from these combined agencies should be just
such a mixed nondescript one as the actual case presents, viz.,
a little clear stratification in the lower part, some suggestion of
stratification of an uncertain sort in the other portions, but no
complete ‘stratification or assortment; a general absence of
declared structure, some limestone débris, some shale débris, a
little drift, some loess wash, some soil wash, with land shells,
some stream or back-water silt, with river shells—perhaps
humanly introduced—and some wind silt; and hence, some
portions unleached and others leached, with other variations
from a typical unitarian deposit, such as true alluvium on the
one hand, or typical loess on the other. It seems to me that
the depth of the deposit is quite within the competency of this
method, while its general configuration and aspect are in close
accord with this interpretation. Under this view the burial of
the human remains took place either during the latest phases
the erosive process of the stage indicated, or in the early
phase of the building of the flood plain. The antiquity of the
burial is measured by the time occupied by the Missouri river in
lowering its bottoms, two miles more or less in width, somewhere
from fifteen to twenty-five feet, a very respectable antiquity,
but much short of the close of the glacial invasion.

2. Possible but not probable interpretations— As _ previously

774 Tl. C. CHAMBERLIN

indicated, the case is not so declared as to render a given inter-
pretation wholly certain, and to absolutely exclude all others.
While I think them quite improbable, other times and methods
of burial may be entertained as within the bare limits of possi-
bility.

1) As noted in the description, there are some small and obscure
shoulders or terraces at different heights up to sixty feet above
the upper flood plain of the Missouri river. It is not clear that
the higher of these are anything but degradational inequalities of
structural origin, but it may be worth while to recognize that
these features may possibly be of fluvial origin, and may be
genetically connected with the lower deposit containing the
human relics, though there is no clear evidence of this. In this
case the working level of the river must be placed at perhaps
sixty feet above that of the present day, and its waters must be
supposed to have invaded the mouth of the valley more exten-
sively and deeply. The site of the relics is thus placed in the
bottom of the ancient river, though not in its main channel. It
was therefore, more or less subject to the scouring action of
the river bottom, and to alternate deposition and removal, as set
forth in the preliminary considerations. At any stage during
such submersion, when the current of the river was directed
against the mouth of the tributary, it would be theoretically
possible for the pre-existing deposit to be scoured out and
replaced in the manner so constantly illustrated by the present
action of the river, and in connection with such removal and
refilling, the relics could be introduced. This would place the
time of their burial farther back, but probably not so far as even
the latest stage of the last ice invasion.

The specific character of the deposit does not seem to me to
lend support to this interpretation. It is not. distinctly and
specifically fluvial, as it might be expected to be if formed in the
bottom of the river or in deep and constant water of any kind,
It bears the aspect of a mixed combination product, such as
postulated in the previous interpretation.

2) It may be held that the relics were buried in the early
stages of the Wisconsin glaciation, when the Missouri river was

HUMAN RELICS OF LANSING, KANSAS 775

rising because of the filling of glacial wash poured into it at the
north. In this case it would be assumed that the tributary val-
ley had previously been fashioned as it is now, that with the
filling up of the Missouri valley it also became filled in the
lower part, involving the burial of the relics, and that with
the lowering of the Missouri since the glacial period, it has been
re-excavated to its present extent. In this case the filling should
have combined the characters of a glacio-fluvial deposit and a
back-water deposit. The actual deposit does not seem to me
to be of this kind. The present adjustment of the tributary to
the Missouri river must also, in this case, be regarded as an
accident, however improbable.

3) It has been held by Upham and Winchell that the loess-
like deposit covering the relics is a part of the sheet of loess
that mantles the uplands of this region generally, and is referred
to the Iowan stage of glaciation, and that the relics were buried
in the early stages of this accumulation, or earlier. This view
receives more apparent than real support from the partial resem-
blance: of the upper part of the deposit to loess. As already
stated, this does not seem to me to be true original loess, either
of the upland or of the fluvial type, but a secondary deposit, in
part, and only in part, derived from the loess. If so, its age is
that of its derivation, not that of the parent loess. Very similar
deposits seem to have been formed at all ages since the main
loess epoch, and are being formed now, and apparently must con-
tinue to be formed as long as the general loess mantle remains
the chief source of erosion and re-deposition, but these deposits
generally betray their origin by their secondary characters, as
in this case.

4) It is even possible to regard the limestone débris in which
the skeleton was found as preglacial detritus, buried first by
the Kansan drift, which was afterward eroded, and then by the
loess-like deposit ; but in the first place, the detritus is not of
the distinctive residual surface type, since it is not thoroughly
weathered and leached as such deposits usually are, and in the
second place, the hypothesis assumes that the post-Kansan ero-
sion was adjusted to the preglacial erosion with a degree of

776 T. C. CHAMBERLIN

nicety quite improbable, and in the third place, the view leaves
very little erosion and deposition to be referred to the long,
subsequent stages, and in the fourth place, it leaves the adjust,
ment of the tributary to the Missouri a matter of accident, and
two accidents of nice adjustment in one hypothesis are some-
what too many.

5) At the other extreme, it is perhaps possible to refer the
burial to very modern action of the Missouri waters at a very
exceptionally high stage, combined with deposition by the tribu-
tary, aided by slope wash and creep and wind work from the
Missouri bottoms. This seems to me, however, to be pressing
agencies to the limit of their possibilities rather than resting
with their probabilities within the limits of their more habitual
action.

Without holding it to be quite demonstrable, it seems to me
that the weight of evidence is very strong in favor of the first
and most conservative interpretation, which finds an apt and
adequate explanation in the natural order of things.

In this connection, I beg to invite the attention of archeolo-
gists to the slight grounds for hope of finding really strong evi-
dences of man’s antiquity in the fluvial deposits of the glacial
rivers, because of the lability of these deposits to deep over-
working by scour-and-fill. Onthe Ohio, for example, the floods
are today boring out deep holes in the river and shortly filling
these again, only to bore and fill somewhere else. It would
doubtless not be difficult to sow coins of this year’s mint over
the bottom of this river in such a way that a decade hence they
would be buried a score or some scores of feet in gravel and
sand; and what is more, this gravel and sand would be of the
glacio-fluvial type, since it would be only the true glacio-fluvial
material rearranged by stream action not unlike that which origi-
nally formed it. It would hence be stratified, and nearly or
quite indistinguishable in small sections from the original. The
same process has been in progress ever since the river began to
erode the glacial filling. If its early meanders covered the whole
of the original glacial flood plain, no part of it would be exempt
from the suspicion of such overworking and natural intrusion.

HUMAN RELICS OF LANSING, KANSAS WAL

It thus appears that even if the burying gravels were of glacial
aspect, and the burial were a score or two score, or perhaps even
three or four score feet deep, it would require careful circum-
spection to remove legitimate and necessary doubts arising from
this source. This might be done in special cases on geologic
grounds, and the nature of the human deposit might in other
cases help to eliminate these sources of doubt, but special and
strong evidence of this kind is required to make a good case.

So far as the glacial ages are concerned, evidence of man’s
presence should be sought rather in the interglacial than in the
equivocal fluvial deposits. With careful identification and rea-
sonable circumspection, all sources of doubt as to age could
be removed from the intercalated deposits of the interglacial
epochs, and as these carry the relics of other life, they are com-
petent to carry those of man if he really lived in the region at
thei time:

I am permitted to add the following notes by Professor Calvin
and Professor Salisbury, who examined the deposit with me, and
who have been kind enough to read and criticise my manuscript,
as prepared before my second visit. The observations of that
visit strengthened the grounds on which they have indicated
slight divergencies from my views.

T. C. CHAMBERLIN.

STATEMENT OF PROFESSOR CALVIN.

I thank you for the opportunity you have given me to read
the manuscript of your paper on ‘‘The Geologic Relations of the
Human Relics of Lansing, Kan.’ I wish to thank you further,
not for myself alone, but on behalf of all geologists engaged in
the study of problems similar to the one under discussion, for
the full and clear presentation of the behavior of rivers of the
Missouri type in connection with migrations of their meanders,
of their work in degradation and aggradation, in scour-and-fill,
while deepening and widening their valleys, and of the chang-
ing conditions which they impose on their tributaries. The
application of the principles discussed in the preliminary part

778 T. C. CHAMBERLIN

of the paper to the interpretation of the deposit in which the
human bones were found near Lansing, Kan., as given in your
Interpretation I,seems to fit the case and harmonize all the facts
in a very admirable way. If I were to dissent at all from your
conclusions as stated in Interpretation I, it would simply be to
the extent of saying that a lowering of the Missouri valley since
the bones and associated silts were deposited, through a space
somewhat less than fifteen or twenty-five feet, would probably
be amply sufficient.
SAMUEL CALVIN.

STATEMENT OF PROFESSOR SALISBURY.

With the general conclusion of- the above paper as expressed
under the heading, ‘‘ The most conservative and the most prob-
able view,” I am in perfect accord. If I have any suggestions
to add, they are the following:

1. Aside from the distinct layer of clay in one wall of the
tunnel, I saw no structure which could properly be called strati-.
fication.

2. The band of water-laid clay seemed to me to imply stagnant
or essentially stagnant water. I am disposed to refer its origin
to a time when high water in the Missouri ponded the tributary.
Since the level of the clay is but a few feet above the historic
high-water mark of the river, the stream need not have been
flowing more than a few feet above its present level when the
clay was deposited. I see no reason for supposing that the
introduction of the skeleton and the deposition of the clay were
far separated in time.

3. The unequivocal layer of water-laid clay seems to me
strong evidence against the view that the material in which it
occurs is referable to any of the recognized loess epochs. I have
seen thousands of sections of loess, but never one with such a
seam of clay.

4. I regard the presence of the unio shell as evidence that
the loess in which the tunnel is dug is not in its original posi-

HUMAN RELICS OF LANSING, KANSAS 779

tion. Iam not aware that a unio shell has ever been found in
undisturbed loess. The presence of the shell in loess talus—for
that seems to me the proper characterization of the material in
which the human relics were found —could be readily accounted
for in various ways, one of which is suggested in the preceding
pages.

Roiuin D. SALISBURY.

STUDIES FOR STUDENTS

THE MAPPING OF THE VTRYSTALEINE SCEISTS:

PART I.— METHODS.

INTRODUCTION.
Conventional geological maps a mixture of fact and theory in unknown propor -
tions.
Outcrop maps needed to display the fact.
FactT.—THE OBSERVATIONS.
Routine field observations.
Important additional observations.
THEORY.—THE DRAWING oF BOUNDARIES AND COLORING OF MAP.
The canons of geological mapping in crystalline areas.
Canons of mapping modified by basal assumptions.

INTRODUCTION.

CONVENTIONAL GEOLOGICAL MAPS A MIXTURE OF FACT AND THEORY
IN UNKNOWN PROPORTIONS.— For areas of the crystalline schists no
person save the maker or some one familiar with the area rep-
resented, can form any estimate of the value of a geological
map. In proportions dependent not alone upon the worker and
his conditions, but upon the area itself, theory has been com-
pounded with fact, until the map represents not what is, but
what the maker thinks after a study more or less extended.
Unfortunate also it is that the thoroughness and detail of the
study is not revealed. It is true that a geological map may be
so constructed as to disclose at once an impossible condition, or
one so contrary to common experience as to excite suspicion, so
that von Decken, the greatest of German map makers, once said
that there were really but two classes of geological maps—those
that may be right and those that cannot be right. But on the
other hand, a map may be prepared with but little examination
upon the ground, which is quite as plausible in aspect, or it may
be even more plausible, than one prepared by a conscientious and

780

MAPPING OF THE CRYSTALLINE SCHTSTS 781

competent worker who at great expense of labor and thought
visits each exposure and studies it with respect to all its neighbors.

In regions in which rock types are easily distinguished, and
where the tectonic structure is simple, the difficulties above
referred to are reduced to a minimum; but in belts in which the
rocks have been profoundly metamorphosed, and where oro-
graphic forces have brought about a complex structure, the
dangers which arise from compounding fact and theory may be
so great as to practically destroy the value of the map.

OUTCROP MAPS NEEDED TO DISPLAY THE FACT.— The largest
element of danger is removed and a map increased very much in
value by indicating upon the tinted areas of the different forma-
tions (the theory) the position, the nature and the observed
peculiarities of the outcroppings (the fact). Ifa map has been
made without sufficient field study this fact will then appear, and
if it must be more thoroughly worked out and republished, the
earlier work is not lost if it was intelligently and conscientiously
done. If, however, it was carelessly, or for any reason in-
efficiently done, the examination of a few of the exposures which
have been represented upon the map will supply the basis for
judgment. In acomplex area much time is consumed in making
a just estimate of the value of a conventional geological map,
and if revision is necessary the work must be taken up de novo,
even if the field work was efficient so far as it was carried.

The serious objection to the general use of outcrop maps is,
of course, the great expense which they involve because of the
large scale which must be adopted. In the complex region
bordering Lake Superior, the United States geological survey has
most wisely adopted the plan of publishing outcrop maps to
properly present the careful work of the geologists of the division.
It is safe to say that the time is not far distant when equally
precise methods must be adopted at least by government and
state institutions for allregions of equal or of greater complexity.

FACT.— THE OBSERVATIONS.

ROUTINE FIELD OBSERVATIONS.— Location of outcrops.—The
first requisite for geological mapping in a region of crystalline

782 SLODIES HOR SLODENTLS

rocks, is an accurate base map of sufficiently largescale. Unless
the relief is slight or the topography of a simple character, it is
essential that the topography be represented either by hachures
or contours, preterably the latter!, A scaleiotanincly tothe
mile (1:62,500) with a contour interval of twenty feet will no
more than suffice to express the detail that is necessary for the
best results. If roads are numerous and reasonably good, a
bicycle with cyclometer attachment will be a valuable adjunct,
and with little doubt the best way of securing base locations as
well as getting quickly from place to place.’

By the use of the aneroid and 4-inch sighting compass, and
by pacing, all locations can probably be made with sufficient
accuracy if the base map is reasonably good and the country is
one of diversified topography. Ina wilderness like that about
Lake Superior, where the monotonous uniformity of the topog-
raphy and the forest cover make outlook impossible, the matter
of location becomes exceedingly difficult. The problem has
there been solved by scouring the country for outcrops in parallel
and contiguous belts, a compassman accompanying each geologist
to keep the direction and pace the distance, so as to allow him
some freedom of movement.

Examination of rock and collection of specimens. —YVhe exami-
nation of the rock is made upon the ground after securing the
freshly broken surface of a specimen as little weathered as pos-
sible.

The specimen should be examined with the naked eye for its
general aspect and compared with the weathered surface of the
outcrop. Its manner of fracturing under the hammer may be
of much significance, and of two closely similar rocks it may
even remain the chief distinguishing difference in the field
observation. With a pocket lens of a magnification of 6-14
diameters, according to personal preference,’ a more careful
examination is made of the fresh surface to note the essential
and accessory minerals and their relative proportions, the texture

* Hoses, “ The Geologist Awheel,” Pop. Sez. Month., Vol. LVIII, pp. 515-518.

2 The best lenses for the purpose are the Hastings aplanatic triplets, manufactured
by Bausch & Lomb, Rochester, N. Y.

LAT TTEN GOL TELLG VCLRNAS ALE TINGE | S\GLL STS) 783

of the rock, the evidence of crushing, etc. A specimen of the
fresh rock, in case the type is new to the region or of doubtful
determination, is made by trimming a fragment into the shape
of an elongated rectangular pillow three by four inches and of
a maximum thickness of aninch. One or more chips for section
cutting are also collected. When determinations are extremely
difficult a series of smaller specimens untrimmed and showing
the weathered as well as the fresh surfaces will be found far
more valuable for identification and reference. These to the
number of ten or more and representing as different phases as
possible may be chipped from a single exposure or from near-
lying exposures and given a single number as a series.
Measurement of strike and dip.—It is a time-honored and
almost uniform practice to record the present direction of the
plane of bedding in terms of the strike (the direction of the
water line if the outcrop were partially submerged) and the dip
(its inclination normal to the strike). Both these terms have a
definite meaning when applied to the only slightly disturbed
sedimentary rocks, and may properly be measured by the text-
book methods, the former by leveling the compass with its
north-south edge in contact with the bedding plane and reading
the bearing, the latter by taking the steepest inclination of the
beds —the one normal to the strike.
_ In the crystalline schist areas, however, the matter is far
from simple. The rock may have no definite structure plane,
either because of igneous origin or from metamorphism of a
bedded rock. Again it may show one or more prominent
structures, but no one of them may represent the plane of sedi-
mentation. If one of them is proven to be the plane of bedding
the best determination of the strike would probably not be
obtained by the text-book method of placing the compass edge
in contact with the rock surface. A quicker and generally more
correct method is to hold the compass above or beside the
exposure and adjust it by the eye to correspond to the average
strike of the exposure: The error of the eye in adjusting to
parallelism, which should hardly exceed 5°, is inconsiderable
when the variation between different parts of an exposure or
between near-lying exposures is taken into account.

784 SLODILS HOR SUGDEN TES:

Something further will be said upon the differentiation of
original and secondary structures

IMPORTANT ADDITIONAL OBSERVATIONS.— fitch and its measure-
ment.—Hardly less important than the inclination of the bedding
plane at a locality is the pitch of the folds—the inclination of
their trough and crest lines. Vital as is this element in the
determination of the structure of an area, it is only in recent
years that it has been given much weight in field investigations.
It is safe to say that its importance is yet only half appreciated.
To Professor Raphael Pumpelly, who directed the early field
work in the New England area, is due the credit of bringing to
the front this element in the field study of crystalline rocks.
The details regarding the methods of measuring it were worked
out by the assistants in his division.’

The pitch is often consciously or unconsciously measured as
a component element of the dip, the dip being increasingly
determined by the pitch as the outcrop is near the crest or the
trough of a fold. Pitch can only be measured at the locality
when the outcrop exhibits minor folds or plications, the study
of the Green Mountains having clearly demonstrated the fact
that the minor fold is the epitome of the major fold. The pitch
is subject to the same kind of variations within an exposure as
is the dip, and it must not be recorded from a single minor fold
or plication without first noting whether other folds or plications
give similar values. Even then its inclination cannot fairly be
assumed beyond the exposure itself, and it is only by examina-
tion of a series of exposures that a persistent pitch is determined.

When not to be made out at an exposure, pitch (and here a
persistent one) is correctly inferred when the strikes of the
opposite limbs of a fold are other than parallel. A syncline
pitches in the direction in which the strikes of its opposite limbs
diverge and an anticline in the direction in which they converge.

Inferences regarding the pitch may often be drawn from the
profiles of ridges, the gradually sloping lines being formed by
the crests of folds.

tSee Monograph XXTIT, U.S. Geol. Surv.; also VAN HIsE, “ Principles of Pre-
Cambrian Geology,” Stxteenth Ann. Report U. S. Geol. Surv., Part 1, 1896, pp.603-632.

‘sasua] zjtenb yo smox patduimio ay} usamjaq
pue yer oy} ye avadde Surpuvq pue uoneyoy Arepuooas ayy, ‘aimsodxa ay} yo woyoq puv do} ayy ivau sasudy
zyzenb payduinio jo saul] ay} Aq pajvorpulse [ejuozt1oy ApIvau MOU SI Surppaq fo oueyd [vursis10 ayy yo uosamp au

.

“SSVN NOLONIYUVEA LVANOD UVAYN GVOU ALVLS NO SSIAND NI ONIGNV& GNV NOILVITOA AYVANOOAS

"I seq *L "ON “X “JOA 10Oas) “ANOf

VIP PINGROL Litt Es CRYSTALLINE SCHIS TS 785

Van Hise has shown that the crests of folds are lines of
special importance by reason of the fact that the order of super-
position of beds is there almost sure to be the normal one, and
the additional fact that crushing being there at a minimum, con-
glomeratic and other original structures are most likely to be
preserved.

Form of minor folds —As might be inferred from the last
section, the form of the minor fold or the plication is full of
meaning respecting the characteristics of the larger folds of the
region, and it should, therefore, be carefully studied and its
peculiarities sketched or recorded.

Secondary foliation and its relation to bedding.— Foliation is one
of the commonest of the structures in the crystalline schists.
It is now well known that in the recent past the strike and dip
of foliation have rather generally been measured as those of
bedding, so that in most papers, written more than a decade
since, the secondary nature of foliation was not recognized,
and recorded observations of dip and strike as frequently Keler,
to secondary foliation as to planes of sedimentation. Even with
the larger knowledge of the present, the differentiation of the
two structures is difficult and often impossible. Two, three, or
even more sets of planes of separation, with widely different
directions, may all be easily made out in the same hand speci-
men of clastic rock, and yet no one of them correspond in direc-
tion with the bedding plane.

A difference in composition characterizing alternate parallel
bands of considerable thickness is undoubtedly the best criterion
for determining the plane of sedimentation.* A striking instance
of secondary banding is represented in Plate I, where the
bedding plane is outlined by the crumpled lenses of quartz and
by even more beautiful puckered bandings not discernible in
the view. The secondary straight banding is parallel to the
foliation. (Plate XIV of the work last cited is from near this
locality.) Such a structure, or one resembling it, may, however,
be produced by the mashing of a coarse granite or conglomer-

™See, however, Hosss, “Secondary Banding in Gneiss,” Buéd/. Geol. Soc. Am.,
Vol. III, pp. 460-464, Pl. XIV.

7386 STUDIES FOR STODENTS

ate, but the characteristic structures of these rocks are usually
not obliterated unless the bands are reduced to a thickness of
a quarter of an inch or less. Again injection of granitic mate-
rial along the planes of foliation of a rock may produce an
alternation of bands of different composition. Such a structure
is, however, apt to be recognized by an experienced observer,
and unless subsequently mashed the bands are not sufficiently
uniform in thickness to simulate bedding.

Of all the crystalline rock types of clastic origin, quartzite
offers the greatest difficulties in the search for the bedding plane.
In fact, it is generally impossible to determine it except by con-
tacts with other formations. .

Other things being equal, the more crenulated or wavy a
structure, the greater the probability that it is original. Also
the more its general direction diverges from an undisturbed
plane of foliation the more likelihood is there that it is earlier
and original. When the bedding plane cannot be made out with
certainty at a particular locality it may sometimes be inferred
from a study of the form of the minor folds in their relation
to foliation at a neighboring locality. Owing to the fact that
mashing is at a minimum in the arches of folds, careful search in
an area of closely compressed folding will often reveal at regular
intervals the beautifully crenulated arches with foliation planes
bisecting them, although between such localities the foliation
planes only can be found. Such a method may even be applied
toa slate. The green and purple slates of the Taconic range of
mountains dip to the eastward persistently and quite uniformly.
The writer was able to find at fairly regular intervals across the
strike of this foliation the acute crests of easterly dipping
anticlines preserved in lenses of infiltrated silica, thus showing
that a series of moderate thickness has, by buckling and com-
pression, brought about perfect foliation parallel to the axial
planes of the folds, and thus given the impression of vast thick-
ness.

Van Hise, in his treatment of the mechanics of flexuring —
confirmed by field observation—has shown that in moder-
ately compressed folds the foliation planes of the opposite

MAPPING OF THE CRVSTALEINE “SCHTISTS 787

limbs converge downward in an anticline and upward in a syn-
cline:

Joint planes.—Joint planes more or less perfectly developed
are almost universally present in the crystalline rocks. They
are planes of separation in parallel series, of which more than
one is usually to be made out. They usually have a very steep
inclination and are probably like the folding, in most cases a
result of the compression of the area. Their number (gcesethe
number of series) and relative perfection, and the direction and
inclination of their walls, should be carefully noted, with the
relation of these to lines of displacement and to topographic
features.

Scarps and steep rock. walls.—Little attention seems to have
been accorded these features of rock exposures. In the sequel
I propose to show their importance in the geological structure of
aregion. Their direction and magnitude should be noted, and if
possible some symbol easily distinguished from the dip and
strike symbol should be entered at least upon the working map.
A short dotted line will serve to indicate the direction of the
scarps and a figure at its end its bearing to the east or west of
north.

Margin of outcrop.—Another neglected feature of rock expo-
sures is the margin of the outcropping. Every experienced
geologist must be able to recall numerous instances where out-
crops, grouped so thickly as to indicate probably but slight
covering in the intervening area, cease abruptly at a border
which is often nearly or quite rectilinear and perhaps is also for
sections of its extension the margin of individual exposures.
Such observations, it is believed, are of great significance in the
proper interpretation of the structure of a region.

Gorges and sharp straight valleys —Even the best of the topo-
graphical maps do not indicate all gorges or rectilinear valleys
with rock walls. In the interpretation of the geological struc-
ture their significance may be considerable, and it is important
to note their direction and in many cases also the height of their

™See on the general subject of foliation, PUMPELLY, WOLFF and DALE, JWono-
graph XXIII; U. S. Geol. Surv., pp. 136-158, Pt. I, 1896; and VAN HIskE, Six-
teenth Ann. Report U. S. Geol. Surv., Pt. 1, pp. 633-668.

788 SLODTTES HOR. St CDi INS:

walls. A symbol like the cut will be found useful in recording
them upon the map. The figure at the end of the trough line of
the valley indicates the bearing in degrees to the east or west of
north, as the ‘case. may be. Bhe slope lines on ‘either side
pointed toward the trough are like Dana’s dip symbols adjusted

in length to indicate the steepness of the

¢ Qo 20 slope, a large slope line indicating a gentler
ayy slope. The figures used in connection with
pe these lines indicate the height in feet of the

ES walls on either side, an estimate bein
Le fe gS
20 ie expressed by the == sign: Very large and

important valleys are often omitted from
topographic maps the topography of which is sketched, because
hidden from points at which the sketches were taken.

Nature of contact surface of formations.—Contacts of forma-
tions wherever exposed should be carefully scrutinized, not only
to discover evidence of conformity or unconformity, but for indi-
cation of slipping, thrusting or faulting. The contact plane of
formations for two reasons is likely to be a locus of displace-
ment. In the first place contacts are likely to be planes of
weakness where a maximum of movement has occurred ; and in
the second place, displacement, whether by normal faulting or
thrusting, produces new contacts. It 1s, therefore, to be noted
at the contact of sedimentary formations: first, whether the
plane of contact is the natural result of sedimentation, and
whether conformable or not; second, whether the deformation
of the contact plane, if present, is such as may be explained by
accommodation of layers in flexuring, or must be accounted for
by the more violent processes of normal faulting or thrust.
With good exposures the direction of the plane of displacement
will probably allow of a decision between thrusting and normal
faulting. If insufficiently exposed the nature of the folding and
the general character of the deformation within the area may
afford a clue. Both normal faulting and thrusting may, how-
ever, have occurred at the same contact.

MAPPING OR LHE CRY STA LEINE SCHISTLS. 789

THEORY.— THE DRAWING OF BOUNDARIES AND COLORING
OF MAP.

THE CANONS OF GEOLOGICAL MAPPING FOR THE CRYSTALLINE AREAS.
—The determination of what types of rock shall be grouped under
one formation-color may be a problem of the greatest difficulty,
but in the absence of conglomerates or clearly marked uncon-
formities, the question is a petrological one to be settled by the
best judgment of the observer on the ground checked by subse-
quent microsopical and perhaps chemical investigation in the
laboratory. Where the contacts are not sharp but exhibit
gradational facies, there may be a question as to where the divid-
ing surface should be fixed, but this is after all a matter of
comparatively small moment, and however determined, there is
little danger that much violence will be done the facts.

The drawing of boundaries where contacts are not exposed —
and contacts are generally covered—is a problem calling not
only for much judgment but often also for a large imagina-
tion. In most domains of science the worker is allowed to
express his lack of adequate data upon which to base a conclu-
sion, and to reserve his judgment pending the obtaining of fuller
information. Here, however, the demand is peremptory and the
line must be drawn with or without knowledge. It is no uncom-
mon experience to find within a region of intricate geological
structure an alluvial valley a mile or more in width bordered
by drift mantles of only lesser width before the bed rock is
exposed. However the boundary lines of the formations are
drawn, the map maker may be reasonably sure of one fact only,
—that the reality is totally different from the representation.
Convention requires that all areas shall be colored to show the
underlying rock, and it may at least be said that the result is
every whit as attractive and plausible in appearance as though
the coloring were based throughout upon observed facts.

It is perhaps a step in the right direction to leave the lakes
and rivers uncolored, as is now done by the United States
geological survey upon its atlas sheets; but to this also there
are serious objections, as, for example, in the area about New
York city, where the numerous bridges, piers and abutments, tun-

790 STUDIES FOR STUDENTS

nels, and government dredgings afford a body of knowledge
of the submarine bed rock equal to that obtainable from most
land areas.

The necessity for drawing boundaries in covered areas has
evolved a body of largely unwritten doctrine which may not
inaptly be referred to as the canons of geological mapping.
for the crystalline areas. Some of these canons are truths,
others half truths, and still others have little basis in fact.
Some of the more important of them are subjoined with brief
critical notes. They are of course to be weighed in connection
with one another and a compromise decided upon.

1. Propinquity: of outcrop.— Near-lying outcrops should be con-
nected when not separated by outcrops of a aifferent formation.
This rule is almost axiomatic and thoroughly sound in principle
provided no contrary evidence is at hand; but there is yet a possi-
bility of error even when outcrops are separated by less than
one hundred feet.

2. Contacts covered.—TIhe boundary ts not at the margin of
either of the separated outcrops of different formations, but at some
line between. This is a general rule but with many exceptions.

3. Strike.-—Direction of boundary conforms to strike of near
outcrops. A rule of much value but to be weighed with other
considerations before decision is made.

4. Slopes.— On slopes boundaries should be corrected for strike
and dip. With vertical beds if the strike is away from the ele-
vation in the general direction in which the boundary is being
extended, the boundary will run down the slope, if toward the
elevation it will run up the slope. If the slope also changes in
such manner that the contour line converges less or diverges
more from the strike the boundary will curve more rapidly
down the slope; if the contours diverge less from or converge
more with the strike in that same direction, the boundaries will
the more rapidly ascend the slope. If the dip is with the ele-
vation the boundary bends the more up the slope, if out from the
elevation the more down the slope.

5. Lopographic breaks.— Sudden changes in slope not caused
by talus or adrift are likely to correspond to contacts.

WAT ATHENG OTE TILE NCRNSTLALEINE SCHISTS 791

6. Variable hardness of rock.— Ridges are occupied by harder
rock, valleys by softer or more soluble rock. To be applied with
great caution. In many districts, in the opinion of the writer,
the valleys are nearly if not quite as often hard as soft rock.
A structural line of weakness may have produced a pre-glacial
stream bed which the ice-cap widened and deepened; an ante-
cedent stream may have produced the same result in hard
rocks, an orographic block may have suffered depression, etc.

7. Hydrography.—In a limestone and hard rock country lakes,
ponds, rivers, and swamps are generally in softer or more soluble
vock. A most pernicious doctrine because so generally accepted
and so frequently wrong. Faith in the canon also discourages
investigation. The writer has in mind a nearly circular lake in
the middle of a great valley largely underlaid by limestone, but
about one-half the lake bottom is mica schist, the other half
limestone.

8. Ridge and valley structure.— Ridges are more generally syn-
clines and valleys anticlines. A half truth. It is perhaps true in
the greater number of instances of rocks which suffer an amount
of plication and rupturing of flexures (schists, slates, and some-
times limestones). It is not true of the heavy gneisses which
so generally raise their heads in domes, the beds quite generally
dipping with the slopes.

9. Llocks.— Angular blocks when not lying on steep slopes
and not clearly ice borne are on bed rock of the same kind. The
tendency is to drive this doctrine to its limit. For great lime-
stone blocks not strengthened by silica or silicates it may per-
haps be assumed that they are not very far from the parent ledge.
With a framework of silica they have been carried by the ice
thirty miles or more.

10. Stnk-hole structure.— Superficially undrained areas when not
morainal (or in adrift) are on limestone. A pretty safe rule.

11. Sow.—In a limestone and hard rock country arable farm land
7s on limestone and woodland on hard rock. A fairly useful rule if
applied intelligently.

CANONS OF MAPPING MODIFIED BY BASAL ASSUMPTIONS.—The
above canons are to be considered in connection with one

792 STUDIES FOR STUDENTS

another, the worker giving to each what weight in his judg-
ment is best to secure a compromise for the result, They are
perhaps as good a set of rules as we can secure for the settle-
ment of matters regarding which the known facts are insufficient
to allow of a definitive conclusion.

A map constructed upon observations and rules as above
outlined might be difficult to read, as the order of superposition
of the formations might not at once be clear. The formations
do not always appear upon the map in their direct order, being
sometimes inverted, and sometimes individual formations do not
lie next to their immediate neighbors above and below. The
map maker is expected to show how each formation has been
brought into its present positions and attitudes, to do which
fundamental assumptions of far-reaching importance are made
Theory is thereby brought in very large measure into the solu-
tion of the problem, it may be so as to take precedence over the
canons of doctrine which have been outlined above. These
basal assumptions and their effect upon the map will be con-

sidered in a second paper.
WILLIAM HERBERT Hoses.

ADI ROR LAL.

THE discovery of human remains under twenty feet of débris
near Lansing, Kan., has revived interest in the antiquity of man
in America, and fortunately on more hopeful lines than hereto
fore, since the mode of occurrence at Lansing is more definitely
determinate than in most previous cases of the kind, and the
geologic elements of the problem are more declared, though, as
it happens, they belong to a much overlooked yet very common
type. The recent studies of Brower and Winchell on the
quartz chips at Little Falls have brought that case into more
definite form.

There remain about the same differences of interpretation as
heretofore, but these will pass away as the specific identification
of glacio-fluvial, alluvial, and sub-aérial adjustment deposits
becomes more familiar and precise, and as their interpretation is
at once given greater latitude and made more strictly dependent
on discriminative criteria.

In the judgment of the writer, neither of the above cases
affords any substantial ground for affirming the presence of
man in America during the glacial period; but they do afford a
strong presumption that man in this country has witnessed very
notable progress in the deepening of the channels of the
Missouri and Mississippi rivers. In time there may be found
means for estimating the rate at which these rivers are lowering
their channels, but at present these are wanting, and there is no
trustworthy method of estimating in years the time consumed
in the deepening which has taken place since the human relics
were buried.

Cee:

REVIEWS

Kakabikansing. By J. V. Brower. St. Paul, Minn.: H. L.
Collins & Co.

Unper this bizarre title Mr. Brower has described the occurrence
of quartz chippings at Little Falls, Minn., prefixing a sketch of the
previous studies of Winchell, Babbitt, Upham, Hill, Holmes, and
Hershey, and affixing a letter from Professor Winchell and a state-
ment on “ Primitive Man in the Ice Age” by Mr. Warren Upham.
The descriptions of Mr. Brower are apparently careful and candid, so
far as intention goes, but they are obviously not those of a critical
geological observer. They neglect most of the really discriminative
factors and embrace much inconsequential matter. Notably also they
have the trait, so common to the untrained worker, of incorporating
interpretation unconsciously while insisting on ‘‘ ascertained facts.”
“The glacial river’? plays a notable part in the description of the
formations, whereas the very thing to be demonstrated is the ‘ glacial”
or non-glacial character of the river at the time the formations in
question were made. None the less the excellent photographs and the
maps, together with the statement of Professor Winchell, largely sup-
ply the lacking data and make it possible to consider whether the
interpretations put upon them are the normal ones or not.

From these it appears that there overspreads the plain once occu-
pied by the Mississippi waters, but now above their reach, a surface
layer of dirty pebbly sand of the typical structureless kind which
usually covers abandoned flood plains of sand and gravel. ‘This is
about four feet thick and at places near the river contains many chips
of white vein quartz of undoubted human origin. The source of the
quartz is unquestionably the veins in the outcropping slate over which
the falls are formed. This quartz-bearing slate does not now rise as
high as the upper surface of the plain, and this fact has been urged by
Holmes and Hershey as evidence that the quartz chippings were not
taken from the parent ledge until the plain had been cut down to the
requisite depth after its original completion. Mr. Brower, while not
answering this objection by positive evidence, holds that the crest of
the quartz-bearing ledge was exposed at seasons of low water, though

794

REVIEWS 795

covered at times of flood. It is of course probable that the crest of
the ledge has been worn down where the river flows over it, but such
erosive covering by the river does not fit in well with the view that
this same portion was the source whence large quantities of vein
quartz were quarried at the same time. It is clearly urging a bare
possibility at best rather than a probable occurrence.

If, however, the case rested merely on the possibility of reaching
the source of the quartz while yet the uppermost layers of the original
plain were in the process of formation, it might be ungenerous to
refuse to entertain the utmost possibilities of the case in favor of
glacial man in America. But the facts of the case, taken just as given
in this paper, do not seem to the reviewer to afford even a plausible
ground for assigning the quartz chips to the glacial stage of the river.
The surface deposit in which they are found, as described and illus-
trated in the paper, not only does not bear the characteristics of a
glacio-fluvial deposit, but bears quite clear evidence that it is not
glacio-fluvial. The descriptions cite the fact that the surface deposit
is highest near the bank of the present bottoms, after the common
habit of existing degrading rivers. This habit is recognized and the
facts are summarized in the following quotation (p. 73): ‘‘At Little
Falls, Minn., the eastern portion of the sandy plain on the east side of
the Mississippi is several feet lower than the crest of the plain at the
easheend of the: dam.” Phat ‘fact is“important. © After the %great
glacial river which overspread the entire plain at Little Falls had with-
drawn into the narrower limits of an eroded streambed, that river,
often in freshet from the effects of the melting ice-sheet, occasionally
re-overflowed the entire plain, disturbing and overturning the sandy
surface, mixing into its materials every chipped quartz blade or spall
which had been placed by the hand of man upon the surface adjoin-
ing the newly eroded and narrower channel. ‘The higher altitudes of
the plain along the Mississippi between ‘The Notch’ and the dam
were caused by successive stages of recurring overflowage, creating
additional surface deposits upon the plain nearest to the newly formed
river bank.” This is indeed ‘‘important,’” as the author himself
naively remarks, since it shows, as the author also recognizes with
equal unconsciousness of its real meaning, that it is the characteristic
action of streams of ¢he present non-glacial régime. It is here recog-
nized, with undoubted correctness, that the quartzes were buried ‘‘by
disturbing and overturning the sandy surface” and by “additional
surface deposits.”” The reference of this, however, to g@acza/ waters is

796 REVIEWS

wholly without evidence and quite against the probabilities. Glacial
streams as a rule have the aggrading habit, and are not therefore
“‘ withdrawn into the narrower limits of an eroded streambed,’ but on
the contrary, are constantly shifting their courses from one point to
another across their whole plain. Usually they subdivide into a com-
plex plexus of numerous shallow shifting branches. ‘There is there-
fore no reason whatever to suppose that the present channel of the
Mississippi at Little Falls was in existence, even in its initial stages,
while the river remained truly a glacial stream. The fact that the
relic-bearing deposit is closely related to the present stream is evidence
that tt was postglacial. ‘Yhe deposit that carries the relics supports
the same view, for it bears the characteristics of a postglacial rather
a glacial formation. On the evidence submitted, therefore, in the
paper the inference is rather imperative that the quartz chips were
buried at some stage when postglacial rather than glacial conditions
prevailed. To make this more clear, it may be worth while to sketch
the normal succession of events and to gather from these the normal
interpretation of the time and mode of burial of the quartz chips,
assuming, as everywhere throughout this review, the complete trust-
worthiness of the evidence given in the paper, especially that afforded
by its excellent photographic illustrations.

1. During the time the glacial border lay across the sources of the
Mississippi and it was therefore truly a glacial stream, the normal
inference is that it had the aggrading habit because of its overburden
of glacial detritus; that it took the form of a plexus of numerous
branchlets, and that it occupied, by the constant shifting of these,
the whole plain which it was engaged in building up. In the nature
of the case it should normally have no fixed channel nor any perma-
nent flood plain deposit, since its whole plain was periodically covered
by the channels of the branching and shifting streams. Its typical
deposits should have been clean, fresh, well-assorted stratified sands
and gravels. Any human relics left anywhere on the plain, even on
portions not at the time occupied by the stream, should have been
worked over and incorporated more or less deeply by scour-and-fill in
the clean, stratified gravels and sands. None such are reported.

2. When the glacier had retired from the basin and no longer
overloaded the Mississippi with its detritus, a transition stage should
naturally have followed. During this the first work was to adjust the
stream to the conditions that immediately followed the glacial retreat.
It is to be presumed that the upper branches of the river were aggraded

REVIEWS 797

to different slopes, dependent on their relations to the glacial supply
of detritus thrown into them. Asa rule, the gradient is much higher
near the ice edge than at a distance from it. These high gradients are
presumably the first to be reduced, while the material so derived is
shifted to the lower gradients which continue their aggradation until
the whole becomes adjusted to the new conditions. It must also be
considered whether the fresh drift surfaces left by the recent retreat
of the ice and the numerous new trenches of the young streams engaged
in developing the new drainage system may not have kept the Missis-
sippi in an aggrading, or at least static condition for a notable period
after the direct influence of the ice sheet was withdrawn. Any human
relics left on the plain during this stage should normally have been
subject to incorporation by scour-and-fill in the clean, fresh, stratified
gravel. But none are reported.

3. After the stage of transition had passed and the Mississippi had
assumed the degradational phase, a period must probably be recog-
nized during which its shifting meanders occupied the whole of its
plain—except occasional protected embayments—and degraded it
from side to side, removing the whole surface of the previous glacio-
fluvial and transitional-adjustment plain. ‘This action is dependent
on the balance of prevailing conditions, and these vary for different
rivers and different portions of the same river. Toa large extent, the
Mississippi has continued actiun of this phase down to the present time
and has cut away the whole of the upper part of the glacio-fluvial plain.
It is only here and there in favored localities that any remnants that
can with probability be regarded as portions of the upper glacio- fluvial
plain can now be recognized at any great distance from the ice edge.
Just how long the river would continue to occupy by its shifting courses
the whole upper plain at Little Falls cannot be determined by any evi-
dence given in the paper, but the stage should be recognized in the
interpretation of the history of the region, and particularly in the
determination of the age of the surface deposits of the plain as it now
exists. Any human relics left on the plain in this stage would be liable
to be incorporated in the clean stratified gravel by scour-and-fill. But
none are reported.

4. After the stages above noted had passed, the river developed a
more restricted track and limited its erosion essentially to this, sinking
its channel gradually into the broad plain and covering the remainder
only in flood time. It would continue to flood the upper plain until
the channel reached a depth greater than the height of the flood stages.

798 REVIEWS

What this height is habitually in this part of the Mississippi is not
known to the reviewer. The range between low and high water is
given by Abbott as twenty feet at St. Paul, thirty-five feet at the mouth
of the Missouri, and fifty feet at some points below. From the data
given in the paper, it would appear that in the natural river, before
influenced by damming, the low water was from twenty to twenty-five
feet below the main plain. If, therefore, an average flood stage were
applicable to this locality, the deepening of the channel since the river
floods rose to the plain could be estimated at only a few feet, but the
barrier formed by the slate renders an estimate of the time very uncer-
tain. Ifthe slate were once much higher than now, it should have
kept the river longer within reach of the plain at flood time, so that
the hypothesis that the slate has been notably cut down by the river,
introduced to avoid the criticisms of Holmes and Hershey, is not with-
out its embarrassments in another direction.

Now, it seems clear from the evidence presented in the paper that
the quartz chips were not spread over the plain while the clean strati-
fied gravels were being formed, nor while the river was meandering
over the plain in its transitional-adjustment stage, nor in its general
degradational stage, for at all of these stages, scour-and-fill should
have incorporated the chips in the stratified sands and gravels. The
chips were quite clearly introduced after the Mississippi had “ with-
drawn into the narrower limits of an eroded stream bed” and while
only its flood stages overflowed the upper plain. This normally
occurred in the fourth stage sketched above. As the recent cutting
down of the channel has been slow on account of the slate barrier, a
very considerable period has probably elapsed since the Mississippi
last reached the upper plain even in its highest flood stages, except as
these might be made exceptional by ice jams and similar obstructions.
This gives the origin of the chips a respectable antiquity, but does not
offer any presumption that it fell within the glacial period, or even
very near its close.. This seems to the reviewer to be the normal inter-

pretation of the evidence presented in the paper.
Cae:

TURGENE “POBLICGALIONS

—AmI, H. M. Esquisse Geologique du Canada Materiaux pour servir a la
Preparation d’un Chronographe Geologique. [Extrait du ‘“ Naturaliste
Canadien ” (Publié par M. l’abbé V. A. Huard), Quebec, 1902.]

—Brooks, ALFRED HULSE. Geological Reconnaissances in Southeastern
Alaska. [Bull. of the Geol. Soc. of Amer., Vol. XIII, pp. 253-66,
Rochester, August, 1902. ]

and RICHARDSON, GEORGE B., COLLIER, ARTHUR J. and MEN-
DENHALL, WALTER C. Reconnaissance in the Cape Nome and Norton
Bay Regions, Alaska, in 1go0. [Dept. of the Interior, U. S. Geol. Surv.,
Washington, Igo!.]

—B6uM, Dr. AuGusT. Geschichte der Moranenkunde, mit 4 Tafeln und 2
Textfiguren. [Abhandlungen der K. K. Geographischen Gesellschaft
in Wein. Wein, 1901. |

—CLARK, W. B. and Bippins, A. Geology of the Potomac Group in the
Middle Atlantic Slope. [Bull. of the Geol. Soc. of Amer., Vol. XIII,
pp. 187-214, Pls. XXII-XXVIII. Rochester, July, 1902.]

— -——— and Martin, G.C. Correlation of the Coal-measures of Mary-
land: [Bull of the Geol. Soc. of Amer., Vol. XIII, pp. 215-32, Pls.
XXIX-XXXIX. Rochester, July, 1902.]

—CARDOT, J. and THERIOT,I. Papers from the Harriman-Alaska Expedi-
tion, XXIX. The Mosses of Alaska. [Proc. of the Washington Acad.
of Sci., Vol. IV, pp. 293-372, Pls. XIII-XXIII. Washington, July,
1902. |

—CUSHING, H. P. Recent Geologic Work in Franklin and St. Lawrence
Counties (New York). [Report of the Director and State Geologist,
1900. |
Pre-Cambrian Outlier at Little Falls, Herkimer County (New York),

[New York State Museum. Report of the Director and State Geolo-
gist, 1900. ]

—ECKEL, EDwInN C. The Quarry Industry in Southeastern New York.
[Reprinted from the Twentieth Report of the State Geologist, r1go00.
Albany, 1902.]

—FARRINGTON, O.C. The Action of Copper Sulphate upon Iron Meteor-
ites. Article IX. [From the American Journal of Science, Vol. X1V,
July, 1902.]

799

800 RECENT (PUBITCATIONS

Meteorite Studies I. [Field Columbian Museum, Publication 64, Geo-
logical Series, Vol. I, No. 11, May, 1902.]

—FAIRCHILD, H. L. Pleistocene Geology of Western New York. Report
of Progress for 1g00. [Reprinted from the Twentieth Report of the
New York State Geologist, 1900. Albany, 1902.]

— Geological Survey of Louisiana, Report of 1902. Director William C.
Stubbs; Gilbert D. Harris, geologist in charge. [Baton Rouge, La.,
1902. |

— Geological Survey of New Jersey, Annual Report of the State Geologist
for the year 1901. [MacCrellish and Quigley, state printers, Trenton,
Nei; 19025]

— Handbook to the Mining and Geological Museum, Sydney. By George
W. Card (Curator and Mineralogist), with special references to the min-
eralogical collections. [Department of Mines and Agriculture. _ Wil-
liam Applegate Gullick, government printer, Sydney, New South Wales,
1902. ]

— HILL, ROBERT T. The Beaumont Oil Field, with notes on other oil fields
of the Texas region. [Reprinted from the Journal of the Franklin Insti-
tute, Philadelphia, August—October, 1902. |

—KEYES, CHARLES R. Devonian Interval in Missouri. [Bull. of the Geol.
soc; of Amer; Vol. XII1, pp. 267-92, Pl. XEDV.; Rochester, August,
1902. |

—Knicut, W. C., and Stosson, E. E.. The Newcastle Oil Field. [School
of Mines, University of Wyoming. Petroleum Series, Bull. V, June,
1902. Laramie. |

—MERRIAM, JOHN C. Triassic Ichthyopterygia from California and Nevada.
[University of California Publications. Bull. Dept. of Geol., Vol. III, No.
4, pp. 63-108, June, 1902. |

— MeunteER, M. STANISLAS. Etudes Géologique sur le Terrain Quaternaire
du Canton de Vaud (suisse), [Extrait du Bulletin de la Société d’His-
toire Naturelle d’Autun. Tome Quinziéme, 1902.]|

— PEARSON, HERBERT W. (Revised and edited by William F. Phelps.)
A Nebulo-Meteoric Hypothesis of Creation. [J. J. LeTourneau & Co.,
Duluth, Minn., 1902. |

— Records of the Geological Survey of New South Wales, Vol. VII, Part I,
1902. [Department of Mines and Agriculture. William Applegate
Gullick, government printer, Sydney, New South Wales, 1902.]

—Ruies, HEinricH. The Production of Flint and Feldspar in rgot, [Extract
from Mineral Resources of the United States, calendar year Igol.
Washington, 1902. ]

RECENT, PUBLICA TIONS relep |

—SHIMER, HENRY W., and GRABAU, AMADEUS W. Hamilton Group of
Thedford, Ontario. [Bull. of the Geol. Soc. of Amer., Vol. XIII, pp.
149-86. Rochester, June, 1902.]

— SCHRADER, F. C. Geological Section of the Rocky Mountains in North-
ern Alaska. [Bull. of the Geol. Soc. of Amer., Vol. XIII, pp. 233-52,
Pls. XL-XLIII. Rochester, July, 1g02.]

—SCHUCHERT, CHARLES. On the Helderbergian Fossils near Montreal,
Canada. [From the American Geologist, April, 1902. |

— SNODGRASS, ROBERT EVANS. Papers from the Hopkins-Stanford Galapa-
gos Expedition 1898-9, VIII. Entomological Results (7). Schistocera,
Sphingonotus and Halmenus. [Proc. of the Washington Acad. of Sci.,
Vol. IV, pp. 411-54, Pls. XXVI-XXVII. Washington, August, 1902.]

—SARDESON, F. W. On the Deceptive Fossilization of certain Pelecypod
Species and on the Genus Eyrymya. [From the American Geologist,
Vol. XXX, July, 1902.]

— TorrEY, HArRry Beat. Papers from the Harriman Alaska Expedition
XXX. Anemones, with Discussion of Variation in Metridium. [ Proc.
of the Washington Acad. of Sci., Vol. IV, pp. 373-410, Pls. XXIV—
XXV. Text figures 5-21. Washington, August, 1902. |

—TAassIN, WirT. The Casa Grandes Meteorite. [From the Proc. of the
U.S. National Museum, Vol. XXV, pp. 69-74 (with Pls. I-IV), Wash-
ington, 1902. |

— ULRICH, E. O., and SCHUCHERT, CHARLES, Paleozoic Seas and Barriers
in Eastern North America. [Reprinted from the Report of the New
York State Paleontologist, t901. Albany, 1902. ]

—WuitTE, Davip. Fossil Alga from the Chemung of New York, with
Remarks on the Genus, Haliserites, Sternberg. [Reprinted from the
Report of the New York State Paleontologist, 1901. Albany, 1902.]

— —— and CAMPBELL, Marius R. and HASELTINE, ROBERT M. The
Northern Appalachian Coal Field, embracing papers on the Bituminous
Coal Fields of Pennsylvania, Maryland and Ohio. [Extract from the
Twenty-second Annual Report of the Survey, Igoo-1, Pt. III, Coal,
Oil, and Cement. Department of the Interior, Washington, 1902. |

— WoRTMAN, J. L. Studies of Eocene Mammalia in the Marsh Collection,
Peabody Museum, Pt. I, Carnivora (with eleven plates). [From the
American Journal of Science, Vols. XI-XIV, 1got-2.]

— WAINWRIGHT, J. T. The Fallacy of the Second Law of Thermodynamics
and the Feasibility of Transmuting Terrestrial Heat into Available
Energy. (Read before the Pittsburg Meeting of the “‘ Physical Section”’
of the American Association for the Advancement of Science, July,
1902.) Chicago, 1902.

THE

H@OWINN Ak OF GEOLOGY

NOVEMBER-DECEMBER, 1902

AV NATURAL “GAS” EXPLOSION NEAR WALDRON, IND:

Introduction.—On the eleventh day of August, 1890, a natural
gas explosion of much violence occurred near Waldron, in Shelby
county, Indiana. The ground was disturbed, and fractures, crevi-
ces, and craters were formed over an area of several acres.

Many newspaper accounts of the explosion were published.
In most cases these were greatly exaggerated. Bare mention
of the occurrence was made by E. T. J. Jordan, natural gas
supervisor of Indiana, in his report to the state geologist, in
I8gQI."

In discussing a violent natural gas explosion that occurred
at Coffeyville, Kan., on July 26, 1894, Haworth refers to a
similar explosion that had occurred in Indiana, giving the local-
ity as Kokomo; he says:

One similar occurrence is known in the gas field of Indiana, near

Kokomo, in which a fissure was formed in the solid limestone from which
natural gas escaped with explosive violence and caught fire from a burning

2

log heap near by.?

Haworth refers here to the Shelby county explosion, which
did not occur near Kokomo, however, but took place about one
hundred miles south and slightly east of that city. Neither was
a visible crevice formed in solid limestone, as the explosion
occurred in an area that was covered with soil and gravel, though

*Indiana Department of Geology and Natural Resources, Seventeenth Annual
Report, p. 338. Indianapolis, 1892.

? Kansas University Quarterly, Vol. IV, p.95. (Mr. Haworth’s information con-
cerning the Indiana explosion was obtained from Mr. William Moore, of Kokomo.)
NolpxG Nona) 803

804 J. F. NEWSOM

limestone crops out near by (at Z. S. Fig. 1) and also 4% mile up
stream from the area of the main explosion.

Later (1899), in a report upon the Waldron shale, made to
the state geologist of Indiana, Mr. J. A. Price states briefly such
facts concerning the Shelby county explosion as could be gleaned
from statements of the citizens of the neighborhood at that
time:

Except for the brief notices mentioned above no account of
this interesting explosion has appeared in any scientific publica-
tion, so far as the writer is aware, and it is with a desire to place
on record the salient facts concerning it that the following map,
figures, and paper are published at this time.

The writer visited the locality on August 15, 1890, four days
after the explosion occurred, while its effects were still fresh,
and while the gas escaping from the crevices and craters still
burned intermittently when ignited. From notes taken at the
time, a map was prepared and a short account of the explosion
was written, and read on November 7, 1890, before the Indiana
University Scientific Society. The map and facts of the present
paper are taken from the paper read, but not published at that
time:

The figures are reproduced from photographs obtained from
J. T. Schaub, Hope, Ind., and Everett Ayers, of Germantown,
Ind.

Locatton.—The explosion occurred near the center of section
7, 11 north, 8 east; the affected area is two anda half mules
south, slightly west of the village of Waldron, and two and a half
miles west, slightly south of the village of St. Paul. Gas wells were
reported to be producing at the time from both of these villages.

Owing to the fact that the Ogden Cemetery was at the edge
of the disturbed area, the outburst of gas was known locally as
the Ogden Cemetery, “‘earthquake,, |. “gas ; explosion, = and
‘blow-out.”’

A total ‘area, ot about ten) acres™ was -attecteds. ihe most
violent explosion was limited, however, to a space of about two
acres at the east side of the locality affected, and is shown
approximately by the Nos. 1, 3, 5, and 6, Fig. 1. The topog-

NATURAL GAS EXPLOSION NEAR WALDRON, IND. 805

raphy and relations of the area can be best understood by
neference to: Pig. 1:

The area that was shaken up lies at the south side of Big
Flat Rock River or Creek, in an almost half circle bend formed
by that stream, and is comparatively flat creek ‘bottom land”’

Second
Bottom.

100 0 re) fe.

Fic. 1.—Map of the area affected by the Ogden Cemetery gas explosion, in
Shelby county, Ind.

made up of alluvium. The south side of this area, marked
‘second bottom,” (Fig. 1) is slightly higher than the portion,
included by the Nosi74,73,.1, 8, 6, ro, and 11, in which the
explosion occurred. At the time of the explosion growing corn
covered the second bottom.

At the east and north side of the stream is a wooded bluff,
of glacial gravel and clay, about forty feet high; this bluff was

806 J. F. NEWSOM

covered with forest trees at the time of the explosion. There
were also a few trees of considerable size bordering the creek on
the side across from the bluff.

Geologic relations —Most of the area affected by the explo-
sion is covered by alluvial soil or gravel; only one exposure of
rock in: (placer Occurs: (viz. 7atrL., ive |) sock in aplace
(belonging to the Niagara group) outcrops one-fourth mile up
stream from the area of the main explosion, and in the affected
area the limestone is probably only ten or twenty feet below the
surface. The strata of this region lie approximately flat.

The rocks which underlie the region are shown by the follow-
ing section of the strata passed through in drilling the gas well
at C. J. Bickhart’s mill, about two miles east of Ogden Cemetery.

SECTION AT BICKHART’S MILL.

I, Clay - - - - - - - 4 feet

2. Limestone (Niagara, in part at least) - - 86

3. Shale (Hudson River) - - - = 774

4. Limestone (Trenton) - - - - a2?
Total - - - - - 886

Gas with a pressure of 335 lbs to the square inch was struck
at nine feet in the Trenton limestone.

An irregularly bedded, porous conglomerate which gives off
a fetid odor from a freshly broken surface is exposed for 150
feet along the east bank of the river about 150 yards above the
area in which the main explosion occurred. This conglomerate
is made up for the most part of coarse sand and small pebbles,
but it contains some bowlders at least eight inches in diameter ;
it dips 30°, N.30°E., and is overlain by soft glacial .clay and
gravel. This is probably a locally hardened glacial deposit,
and, if there is any connection between it and the gas explosion,
that connection is not apparent to the writer.

Character and results of the explosion— The explosion occurred
about 9g o’clock, on the morning of August I1; it does not
appear to have been accompanied by any violent report, for the
attention of the people living in the neighborhood was first

* Reported by C. J. Bickhart in 1902.

807

NATURAL GAS EXPLOSION NEAR WALDRON, IND.

‘06gI ‘zi ysnsny usaye} ydessojoyd ve worgq ‘uotsofdxa ay} 0} 101d povujs 19}eM aay {99I9 ayy JO
W0}}Oq Pdr}VAa][a ay} OsTe st sv ‘uMOYS oe UOISO]dxa oy} Aq UMOP UMOT $92I} PUD Ie] BY TV ‘aInjord ay Jo pus jYSI oy] Je UMOYsS
Ajaieq SIU10D UMOP UMO[G ‘I ‘S81q ‘—% ‘aINssy ULE aq} JO APIS YINOs dy] Je JNO UMOT YIAVI BY] UO Sulpuyys av punoisyoVq ayy Ur
ajdoad aq} Jo ysop «*yeay 9y Aq pauUMOIq SdAVvI] I1DY} PYY PU IBALI 9Y} JO apIs Jsva 9y} UO o1¥ 9INjoOId JY} JO WoT dy} ye S907} YL,
‘YO pouing SSM} 1a[[VUIs puv soAva| IT9Y] peYy SaAvo] JO Use Iv Jey} $9eI} BY, “aanjoId ayy Jo 19}Uad ay} JO IYSU ay} ye usV—S SI
uotsojdxa ayy Aq afqessedur apewm sea yorum peor AyuN0D ayy, ‘I “Stq ul b qurod oy} avau wor rdAtI ayy dn matA—'z ‘DIY

“SN :

wie ae 7

808 J. F. NEWSOM

attracted by a roaring or whizzing sound like the escape of steam
from a boiler, but much louder. Upon looking toward the river,
a sheet of flame was seen extending about 250 yards along the
east side of the river bend. Regarding the height of the flame,
estimates of those in sight at the time varied between 150 and
300 feet or more.

The height and intensity of the flame were sufficient to sear
the leaves on the trees standing on the gravel bluff east of the
river, and to completely burn off the leaves and small twigs from
the trees standing near the fissures through which the gas
escaped.

The vegetation around the openings (8, 9, 10, Fig. 1) at the
west side of the area showed that the gas that escaped from
those openings did not burn.

The roaring and flame continued at their height for about fif-
teen minutes, and then gradually subsided, and the gas burned
from only a few of the crevices that had been formed.

The explosion had opened up a great number of crevices in
the soil near the river bank (3-4, Fig. 1. See Fig. 2 also).
When the ground was visited by the writer, these fissures varied
in width from a few inches to four or five feet, with a like
variation in depth, the shallowness being caused by the caving
in of the soft soil at the sides and from the top.

Some of the fissures had evidently been formed by the
upheaval, or depression, of the surface which had been elevated
in some places, and in others had sunk down as much as four or
five feet. The larger fissures, however, showed that they had
been formed by the soil being blown out from below, the blown-
out material being piled up at the sides of the fissures.

The character of the fissures is shown by Fig. 2, which shows
also the ‘‘heaved up” condition of the surface. The center of
the public road is shown in the foreground at the right side of
Fig. 2. At this place (corresponding to the point 2 in Fig. 1)
the road was so twisted and fissured as to be impassable.

The river bed from 5 to 6, Fig. 1, had been raised several
feet, leaving the bottom exposed where water had stood before,
while the west bank of the stream, from 3 to 4, Fig. I, had sunk

NATURAL GAS EXPLOSION NEAR WALDRON, IND. 809

several feet. It was reported that the river flowed into the crev-
ices between 3 and 6, Fig. 1, for some hours after the explosion.
Forest trees near 3, 5, and 11, Fig. 1, had been blown up by
the roots and blown several (10-20) feet from where they had
stood. The leaves on the trees and bushes from 5 to 11, Fig.
1 (about two hundred yards) were seared by the heat. On the
west side of the creek from B’ to 4 the smaller branches (and
in some cases the bark) of the trees were entirely burned off.

: . : : Kh = ee
: so tS .

Fic. 3.—View from the east side of the stream below 7, Fig. 1, looking westward

across the river towards the point where the explosion was most violent. The trees

shown in the water were blown up by the explosion, as were also the piles of earth
along the river bank on which the figures stand.

Johnson

Many of the trees were completely plastered from bottom to
top with a fine mud, unlike anything exposed at the surface.
This mud must have been thrown out with great force, as was
shown by the manner in which it was plastered upon the trees
and corn of the affected area. Lying about the fissures and
over the surface were large quantities of this mud, which, upon
drying, hardened somewhat and was easily cut into various
designs, which were sold as souvenirs to visitors who came in
great numbers to the locality.

810 J. FF. NEWSOM

Examined under the microscope, this mud is seen to be com-
posed of an impalpable powder, such as might have been formed

by direct precipitation of its constituents from solution.

ANALYSIS OF THE

MUD BLOWN

FROM THE CREVICES.

A. J. Cox, analyst.

Per Cent
Lime (CaO) - - - - . 13.4
Magnesia (MgO) - - - eet 430
Iron oxide (Fe,O3) - - - 15.45
Silica and insoluble silicates - - - 49.75
Carbon dioxide (CO, ) - - - - PANIC)
Water - - - - - - 2 ea 5u0
Total - - - - - 100.6

When seen by the writer, the principal fissure formed by the
explosion (see I, 2, Fig. 1) was between fifty and one hundred
feet long, about five feet wide, and five feet deep. This crevice
had evidently been much deeper, but when the gas ceased escap-
ing under sufficient pressure to keep it open, loose earth had caved
into it and partially filled it. The fissure extended almost east
and west, and was along the foot of the slope, between the sec-
ond bottom and the lower ground. At its south side was a mass
of earth, #-x’, from five to ten feet high, that had been blown
from the fissure.

A field of corn was growing south of the main fissure at the
time of the explosion. About an-acre’ot thisicorm B,.b 10,6
Fig. 1, was burned brown by the heat, while half of this burned
portion, 4, B, B’, Fig. 1, was blown flat upon the ground, and
much of the blown down corn was plastered with mud. The
corn was blown over toward the south, showing that the force
came from the fissure at the north side. Some of the blown
down corn is shown at the right end of Fig. 2.

At 11, Fig. 1, a slice of the gravel bluff was shaken down by
the explosion, and some trees were blown up by the roots.

Irregular openings or craters were blown out at 8, 9, 10, Fig.
1, but the explosion was not so violent here as at the east side
of the affected area.

NATURAL GAS EXPLOSION NEAR WALDRON, IND, 811

; {iti
ma mate

A Johnson:

Fic. 4 —One of the fissures evidently formed by the earth cracking apart, owing
to upheaval rather than by the earth being blown out. The piles of earth in the
background of the picture correspond to x«—x' of Fig. 1 and are at the south side of
the main fissure, I, 2 of Fig. 1.

When visited by the writer, four days after the explosion
occurred, a little gas was escaping from many of the fissures,
and this would burn for a short time when ignited.

It was reported by the citizens of the neighborhood that a
fire had been burning in brushwood ona small island in the river
about 150 yards above the main fissure previous to the explosion,

812 J. F. NEWSOM

and it is probable that the escaping gas was ignited by this
fire.

The accompanying figures show the nature of the fissures Wael
were produced by the explosion.

Cause of the explosion—The explosion was supposed at the
time to have been caused by gas escaping below the casing from
the wells at either St. Paul or Waldron, or at both places, and
finding its way between the strata to the point where the explo-
sion took place, at which point the strata were too weak to
withstand the pressure.

Whether the gas did come from these wells, whether it came
up from below through a crevice, or whether it was generated
at no great depth in the strata, more or less directly below the
area where the explosion occurred, is not known.

It is evident, however, that gas accumulated below an
impervious layer until the pressure became sufficient to rend this
impervious bed, when the explosion followed. The pressure
required to do this cannot be known, because neither the depth
nor strength of the confining strata is known.

It is obvious that gas, under a pressure of say two or three
hundred pounds to the square inch, if confined by horizontal
strata, sufficiently near the surface so that the weight of the
overlying material would not equal the pressure of the gas,
would tend to spread laterally between the strata, and to cause
these to bend upwards. If the pressure should become great
enough, the strata would finally break and the gas would escape,
possibly with explosive violence. Such an explosion would not
necessarily indicate that a large supply of gas was involved, for
a small quantity would exert as great a pressure to the square
foot as a larger one.

The area affected by the Ogden Cemetery explosion covered
about ten acres; the pressure on this surface at 250 pounds* per
square inch would have been 36,000 pounds to each square foot
of surface.

Taking the weight of the shales and surface soil as 175

* The ‘frock pressure” of the gas in some portions of the Indiana field in 1890
was over 300 pounds to the square inch.

NATURAL GAS EXPLOSION NEAR WALDRON, IND. 813

pounds to the cubic foot, this pressure would have equaled the
weight of the overlying materials at a depth of 205 feet.

Assuming the pressure of the gas to have been 250 pounds
to the square inch, then the strata which enclosed the gas prior
to the explosion may have been anywhere between the surface
and a depth of 205 feet.

tf

Fic. 5.—View looking southwestward from the top of the bluff at 7, Fig. 1.
Prior to the explosion the water of the stream flowed over the surface of the ground
shown in the right foreground of the picture. (From a photograph.)

In closing, attention is directed to the fact that the locality
in which the Ogden Cemetery explosion occurred is near the
southwestern edge of the natural gas area of Indiana, and that

814 J. F. NEWSOM

the Trenton limestone, the great gas producing formation, is
about 850 feet below the surface at this point.’
J. F. Newsom.

STANFORD UNIVERSITY,
August 25, 1902.

« At Geneva, three and one-half miles southwest from Ogden Cemetery, Trenton
rock (the gas producing bed of the region), was struck in a gas well at a depth of
832 feet. “At Bickhart’s mill, about two miles east of the cemetery, Trenton was
struck at a depth of 864 feet, with gas at a pressure of 335 pounds to the square inch.

PNCHING: OF QUARTZ IN THE INTERIOR OF ‘CON=
GLlOMERATES.*

General statement.—Those who have studied the Coal-measure
and other highly siliceous conglomerates have frequently reported
the occurrence of distinct evidences of etching of the quartz
pebbles on certain of the exposed surfaces. The portion
removed by the etching solutions may amount to as much as
nine-tenths of the original pebbles, though one-third to one-half
of their mass are more common amounts (Fig. 1). The peb-
bles affected ex-
hibit strongly
pitted and usu-

ally very rough Fic. Meageccuon showing eianac vetis: c) Ouelines of etched
surfaces. Generalized from surface shown in Fig. 2, 4.

surfaces on the
exposed parts, while the portions remaining imbedded in the
matrix retain unaltered their rounded water-worn characters.
The position in which the etched pebbles have been most gener-
ally observed is upon the flat upper surfaces of ledges or upon
similar surfaces of detached blocks, but instances have been
noted, especially in the region about Olean, N. Y., where the
etching is also strongly developed on the under sides of over-
hanging ledges. So far as known the only published dis-
cussion of the phenomena is by Dr. C. W. Hayes? who regards
the etching as a superficial feature due to the action of azo- humic
acid under ordinary atmospheric conditions.

Character of etching of conglomerate at Blossburg, Pa.—While
examining the talus from an outcrop of Pottsville conglomerate
at Blossburg, Tioga county, Pa., in 1901, the writer noted certain
peculiarities which suggested that the quartz etching might be,
in part at least, an internal rather than a superficial feature of
the rock. The peculiarities first noted consisted of strongly

t Published by permission of the Director of the U. S. Geological Survey.

2 Solution of Silica under Atmospheric Conditions,” Geol. Soc. Am., Bull., Vol.
VIII, pp. 213-20.

815

816 MYRON L. FULLER

etched: pebbles (Mig. 2,4) upon the) flat “suntacey ona, trace
ment of conglomerate which, judging from its shape, was a part
of a much larger bowlder which had split apart along the etched
surface. This seemed to indicate an internal etching along a
bedding plane, but could not be regarded as conclusive. Further
search, however, brought to light bowlders several feet in diame-
ter in which etched bedding planes could be traced through the
mass of the rock. Because of the thickness of the bowlders,
specimens of the etched surtaces were difficult to obtain, but the
character of the etching in a sandy portion of the rock is shown
in the illustrations B and C of Fig. 2, which are from photo-
graphs of specimens taken from opposite sides of an etched
plane extending through a bowlder of some size. Not only
are the surfaces strongly pitted by the complete removal
of a portion of the grains, but under a glass the individual
grains, even those of the smallest size, exhibit distinct evidences
of etching, either by distinct pittings, or by the general “ground
glass appearance”’ of their surfaces.

Still further examination brought to light evidences of etch-
ing within quarried blocks at distances of five feet or more from
the surfaces, but in these cases the etching was confined to the
immediate vicinity of plant remains, now carbonized or changed
to coal, and, so far as could be detected, was not connected with
either etched or unetched bedding or joint planes. The etching,
though distinct, was not of the pronounced type characterizing
many of the bedding planes (Fig. 2, A), but was more of
the character shown by the illustrations & and C of the same
plate.’

The etching, so far as noted at Blossburg, was confined to the
bedding planes or to the vicinity of the vegetable remains in the

™Specimens of quartz pebbles from the Carboniferous conglomerate of Ohio
showing strong impressions of plant stems, were exhibited at the Cleveland meeting of
the American Association, in 1853, by Professor Jehu Brainerd. These impressions,
as urged by Professor J. S. Newberry, were probably due to the action of humic acids
evolved during the decay of the vegetable matter, and appear to be of the same nature
as the etching about the plant remains in the conglomerate at Blossburg. (See JEHU
BRAINERD, Origin of the Quartz Pebbles of the Sandstone Conglomerate and the For-
mation of Stratified Sand Rocks, Cleveland, 1854, p. 16; and J.S. NEWBERRY; Ofzo
Geol. Surv., Vol. U, Pt. I, 1874, p. 111.

ETCHING OF QUARTZ IN CONGLOMERATES Ouy7,

mass of the rock. The portions of the conglomerate bowlders
lying in the humus or exposed to the atmosphere, though some-
times presenting roughened surfaces suggesting etching, proved

Hies 2:

A. Etched surface of a bedding plane in Pottsville conglomerate from Blossburg,
Tioga county, Pa.

B and C. Etched surfaces from opposite sides of bedding plane in a sandy
portion of Pottsville conglomerate from Blossburg, Tioga county, Pa.

on examination to be unetched, the roughening being due to the
separation of entire grains by the ordinary processes of disinte-
gration. The deep channels and pot-like cavities observed in

818 MV RON WES OLE ER:

the conglomerate at other points in Pennsylvania and described
by Mr. Arthur Winslow’ are of similar origin, the only work of
solution apparently being the removal of the cement binding the
grains together. The disintegration of the conglomerate into
sand likewise appears to be unaccompanied by solution, except
that ofthe cement:

Where the etched surfaces are fully exposed to the action of
the weather, as on the upper surfaces of ledges, they are, so far
as known, always bleached to a gray or white color, but where
they are more or less protected, as on the under sides of over-
hanging ledges, they are almost invariably stained with iron. It
is a well-known fact that the action of the humic acids is accom-
panied by a leaching of the iron constituent, or at least a
conversion of the iron to the ferrous state, in which it is ordi-
narily inconspicuous. The oxidized condition of the overhang-
ing surface would seem, therefore, to indicate that the etching,
if due to the action of the organic acids mentioned, is not going
on at the present time. The portions of the bowlders in contact
with the humus of the soil, though not etched, are distinctly
bleached.

Summary of evidence.—The evidences that the etching prob-
ably took place in the instances described in the interior of the
conglomerate at some past time rather than at the surface under
present conditions are as follows:

1. Etching has been noted about vegetable remains at dis-
tances of several feet from the nearest exposed surface, showing
that the solution is not a superficial feature.

2. Bedding planes extending throughout bowlders of several
feet in diameter are frequently etched throughout, indicating the
penetration of the solvent to distances of at least a number of
feet from the surfaces.

3. The continuations of horizontal etched surfaces, both on
the upper and overhanging portions of ledges, frequently ex-
tend up to and disappear into the mass of the rock, the exposed

™“ Peculiarities of the Weathering in the Pottsville Conglomerate,” Sczexzce, Vol.
Li, pp: 12-14;

ETCHING OF QUARTZ IN CONGLOMERATES 819

parts, in fact, apparently being portions of much larger surfaces,
reaching well into the interior of the rock mass.

4. The etched surfaces, except those about the vegetable re-
mains, appear generally, if not always, to be confined to princi-
pal orto cross-bedding planes, or at least to joints, or other
planes of separation, such as might have afforded a passage for
solutions through the rock. The general absence of etching on
surfaces other than the types mentioned, especially on the
exposed exteriors of the masses, strongly suggests that the
etching was internal.

5. The lack of bleaching on many of the etched surfaces
shows that the etching is not the result of organic solvents act-
ing at the present time, and affords negative evidence in favor
of internal solution.

Conditions during the etching.—On examining the etched sur-
faces, it is found that only the immediate exterior is affected,
the pebbles and sand grains being pitted or planed down to a
general surface, without there being, in most cases, the slightest
evidence of solution of the cementing materials (Fig. 3). This
evidence tends to show
that the rocks were con-
solidated when the etch-
ing took place, and prob- Fic. 3.—Diagrammatic section showing charac-

ably indicates rapid ter of etched surfaces of conglomerate. (C. W.
Haves, Geol. Soc. Am., Bull., Vol. VIII., p. 217.
JY

action of the etching
solutions, for if the rocks were open and porous, as they would
have been before consolidation, or if the solutions passed
through them for considerable lengths of time, it seems almost
certain that the solutions would have penetrated for some dis-
tance into the body of the rocks and would have produced an
internal etching or disintegration of the portions adjoining the
bedding or other planes along which the solutions passed. The
existence of small open cavities about the vegetable remains is
also probably to be regarded as evidence of the consolidated
condition of the rock at the time the solution took place.

If the internal nature of the etching of the conglomerate is
admitted, the question arises as to whether it occurred while the

820 MYRON L. FULLER

latter was deeply buried or after it had become exposed at the
surface through erosion. The extent of some of the etched sur-
faces, frequently amounting to several yards, or even rods, and
the apparent continuation of these surfaces indefinitely into the
rock, though affording no positive evidence, is at least suggest-
ive of deep-seated conditions. The evidence afforded by the
bowlders, however, appears to be more to the point, for it is dif-
ficult to conceive of conditions which would permit solutions to
enter such bedding planes and to penetrate and etch the pebbles
throughout their extent, while the external portions of the same
bowlders, both where exposed to the atmosphere and to the
humus of the soil, are entirely without evidences of etching.

Source and nature of the solvent——The more sandy portions of
the conglomerate at Blossburg contain considerable numbers of
feldspathic grains, which, on weathering, it may be conceived,
might have afforded alkaline solutions of sufficient strength to
have produced an etching of the quartz. The fact, however, that
other sandstones and conglomerates higher in feldspathic constit-
uents in equally advanced stages of alteration, do not, in general,
show any evidences of etching, makes it improbable, in the
present instance, that the alkaline solutions were the active
etching agents.

There appears to be no reasonable doubt that the solutions,
which produced the etching about the fossil vegetable remains,
were derived from the vegetable matter itself, in the process of
the change from the original woody state to the present car-
bonized condition. Vegetable remains are abundant in many
parts of the conglomerate, and must have furnished considerable
quantities of organic acids. If, in fact, such was the source of
the etching solutions, the latter, in order to reach the bedding
planes or other passages, must have traversed the mass of the
rock, probably through the agency of the force of capillarity.
That the portions of the rock thus traversed are lacking in
visible etching is probably due to the wide diffusion of the
solutions and the relatively immense areas of the surfaces of
the constituent grains.

The evidence, as far as it goes, aule appear to favor the

ETCHING OF QUARTZ IN CONGLOMERATES 821

view that the solvent had its origin in the vegetable matter of
the rock itself, that it was concentrated along bedding planes or
other passages in the rock, and that it was only where so con-
centrated that visible etching was produced. The etching is not,
however, to be necessarily considered as confined to those parts
of the rock abounding in vegetable remains, the solutions prob-
ably being conducted by the bedding or other planes to portions
remote from the original source of the solvent.

Conclustons—The observations have been insufficient to war-
rant broad generalizations, but the evidence, though not so de-
cisive as might be desired, seems to favor, as far as the Blossburg
locality is concerned, the conclusions given below. It is not
probable, however, that they will hold for the occurrences at all
localities.

1. The etching has not taken place under existing conditions.
This is indicated by the unetched character of the portions of
the rock exposed to the atmosphere or to the humus of the
soil, and by the present oxidized condition of most the etched
surfaces.

2. The etching is an internal feature, as is indicated by the
etching about the plant remains, by the etched interiors of
detached blocks, and by the extension of etched surfaces into
the mass of the rock.

3. The etching took place after the rock was fully consoli-
dated, as is shown by the fact that it is confined to the immedi-
ate walls of the bedding or other planes, the etching solutions
failing to penetrate the rock or even to dissolve the cementing
material except very locally.

4. The etching is believed to have taken place while the rock
was moderately deeply buried, the depth, however, probably not
exceeding 500 feet, this being a common limit of the circulation
of fresh water below the general surface, as shown by the
numerous oil and gas drillings in Pennsylvania.

5. The active solvent was probably an organic acid and was
most likely derived from the vegetable remains within the rock

itself.

Myron L. FuLuer.
U. S. GEOLOGICAL SURVEY.

THE OIL-: AND GAS-PRODUCING ROCKS OF OHIO?

Tus state began producing oil in 1860. From that period
until 1885 the output was restricted to the eastern part of the
state, and was derived entirely from rocks of Carboniferous age.
The production during that time was relatively small, and of
little importance in a commercial sense. In 1885 the vast
repository in the Trenton limestone was discovered, and this at
once gave the industry in the state great impetus. The yield
was such that in 1895 it surpassed that of Pennsylvania and
New York combined, and since that date the state has held the
firstrank. Gas first became an important fuel in Ohio in 1884,
the year of the great discovery at Findlay. From 1888 to 1900,
inclusive, the value of this product exceeded $28,000,000.

The rocks, particularly the lower ones, producing these vast
supplies have been in part studied, especially by Orton,’ but
thus far no paper has been published which considered all the
producing strata. It is the purpose of this article to enumerate
these, to discuss briefly their composition and general charac-
ters, and to show their stratigraphical position.

THE ORDOVICIAN.

In Ohio the upper half only of this great division needs be
considered. This is divided by Orton as follows:

3. Hudson River series” - : - 300-750 feet
2. Utica shales - - - - o-300
1, Trenton limestone - - - - 50

The upper two members contain both oil and gas, but rarely
in commercial quantities. Occasionally high-pressure gas wells
are found, but these soon give out. Nowhere in Ohio can either
of these formations be regarded as a producer of natural gas.
Occasionally in the northwestern corner of the state oil is found

t Published by permission of Edward Orton, Jr., state geologist.
2 Geological Survey of Ohio, Vol. VI, and First Annual Report (1890).

822

THE OIL- AND GAS-PRODUCING ROCKS OF OHIO 823

in the upper members, and wells are known that have had a
large initial production ; but it is a general rule that such wells
rapidly decrease, falling far short of their early promise.

THE PRINCIPAL DIVISIONS OF THE GEOLOGICAL SCALE IN OHIO, AND THE
OIL- AND GAS-BEARING MEMBERS.

( Goose Run sand
Mitchell sand

{ Coal measures~ First Cow Run sand

Macksburg 500-foot sand

| Second Cow Run sand
|
|

Maxton sand

ete ae
f Mountain limestone

Logan group +

{ Keener sand

ea Bigtnion | ig Ijun sand
L { Squaw sand
| Berea grit
Devonian - - - - - - Ohio shales
Gilecian é : z : : : : Power Helderberg sand
Clinton sand
Ordovician - - - - - - Trenton limestone

The Trenton limestone—Quite the reverse, however, with the
first member, known as the ‘Trenton limestone.” Not only is
this the most important source of oil in Ohio, but probably it
is not excelled by any single formation in the world. It
forms the floor, so to speak, of the entire state, being found
wherever the drill penetrates to a sufficient depth. It is dis-
puted whether or not the formation reaches the surface any-
where, but Dr. Orton considered the exposure near Point Pleas-
ant in Clermont county as forming the top of the Trenton.*

The composition of the Trenton has been made the subject
of careful investigation by Orton, who found that the oil- and

t Geological Survey of Ohio, Vol. VII, p. 4.

824 J. A. BOWNOCKER

gas-bearing beds are magnesian. This is shown by the follow-
ing analyses:

CaCO, MegCO, | Ins. Res. | Aland FeOx.
Findlay igias rocks... ea. 53-50 43.05 He 770) 1.25
Bowling Green gas rock.... 51.78 36.80 4.89 ae
Heim axolenockerienisaccetcer: 55.90 38.85 £75 2.94

It was further found that in some places at least the mag-
nesian character of the rock changes rapidly from the surface
of the formation down. Thus an analysis of the rock lying 100
feet below the top of the Trenton at Bowling Green showed
over 88 per cent. of CaCO, and less than 7 per cent. of MgCO,;
while, as shown above, the upper portion of the same formation
has less than 52 per cent. of CaCO, and more than 36 per cent.
of MgCO,. The magnesian character is important, since it ren-
ders the rock porous, thus making it a suitable reservoir for the
oil and gas. Outside of the producing territory in Ohio the
Trenton loses its magnesian character, the CaCO, composing
usually at least 75 per cent. of the formation.*

The rock is often highly fossiliferous, and occasionally the
pieces brought up by the sand pump are little more than a
cemented mass of shells, resembling in this respect the lime-
stones of the Hudson River series. Dr. Orton referred the for-
mation, as shown in the Findlay well, to the Galena, Trenton
proper, and Birdseye divisions.? The total thickness of the
Trenton in northwestern Ohio is unknown, but it exceeds 780
feet. In the southwestern part of the state it is 650 feet, as
shown in a well drilled on the McGhee farm in Liberty town-
ship, Adams county. This thickness is similar to that reported
in Kentucky.

The relation stratigraphically of the Trenton to the next
important producer, the Clinton, is shown in the following
records:

t Joid., Vol. VI, pp. 103, 104. 2 [bid., Vol. VI, p. 116.

THE OIL- AND GAS-PRODUCING ROCKS OF OHIO 825

WELL NO. I, HAINES FARM, ROSEVILLE, SANDUSKY COUNTY

Drive pipe - - - - - - 30 feet
( Gray lime - - - - 140
White lime - - - - - 160
Niagara series, 4 Be eee ze ; c ; ae
White slate, first break - - 2
| Brown lime, hard : - - 21
| Light slate, second break ~~ <2 20
Clinton limestone 2 - - - - - 100
Medina shales (red) - - - - - - 80
Hudson River series (white shales) - - - 427
Utica shale (brown) - - - - - - 252%
Trenton limestone at - - = - - 1,252%

For comparison a record is given of a well drilled on the Rohr
farm near Groveport, Franklin county, in the central part of the
States:

Drift . - - . - - - = eg Oateet
Ohio shales = : - - 2 ! - 167
Cornifereus limestone, top at - - - - =) 3303
Niagara limestone, bottom at - - - - g8o
Clinton shales - - - - - - =i ell
Clinton sand, shells only at - - = - iG 610)
Trenton limestone at - - - - - 2,146
Bottom of well (in Trenton) - - - a 2.0755

THE SILURIAN.

The rocks of this age in Ohio are divided by Orton as fol-
lows:

4. Lower Helderberg limestone - - 50-600 feet
3. Niagara limestone and shale . - 125-380
2. Clinton group - - - - - 20-150
1. Medina shale - - - - 25-150

The Clinton.— Of these divisions the second and fourth only
call for recognition in this article. Along its line of outcrop in
southwestern Ohio the Clinton consists essentially of a highly
crystalline limestone, rich in fossils, especially crinoid stems.
Commonly it has a light color, the tint of which varies from place
to place. The rock takes a good polish, and is sometimes called
marble. Occasionally the formation contains lean hematite, and

t Jbid., Vol. VII, pp. 4, 5.

826 J. A. BOWNOCKER

one of the earliest blast furnaces in the state relied on this ore.
In composition the rock is calcareous, and at one point becomes
the purest rock of this type in the state.

Northward, under cover, the rock undergoes notable changes,
the result being that it closely resembles the overlying Niagara.
It is in the central portion of the state, however, that the greatest
change is found. Instead of a well-marked limestone, there is
found in its place shales of different colors and composition and
a conspicuous sandstone; the latter the repository for the great
reservoirs of gas in Fairfield, Hocking, Licking and Knox coun-
ties. In northern Vinton countya pool of oil also has lately been
found in it, and one well has been secured in Perry county.

The relations of these beds to the overlying Niagara are shown
in the following partial log of a well on the Bauer farm near Sugar
Grove. Samples of drillings below the Niagara were taken by

the writer.
Berea grit - - - é : : top at 620 feet
bottom at 645
Corniferous, Helderberg,

and Niagara limestones - - -

( Shales, light chocolate-colored.
| Little lime - - = E

top at 1,430
bottom at 2,132
top at 2,132

bottom at 2,168

Shales, green and chocolate-colored, ( top at 2,168

Clinton 4 the latter fossiliferous. Some lime bottom at 2,199

Shales, green and chocolate-colored.
Much lime - - ae

top at 2,199
bottom at 2,236

{| Clinton sand at - 2,236

Generally when the gas sand has been penetrated to a depth
of from Io to 15 feet the drill is stopped, and consequently data
farther downare less abundant. In western Ohio the basal portion
of the Niagara is commonly occupied by shales,’ but northward
and eastward this member contracts. In central Ohio it consists
of shales and thin beds of limestone, and the bottom of these
is regarded as the line of junction of the Niagara and the Clinton.

The following skeleton record of a well on the Spire farm near
Sugar Grove shows the position of the Medina as well as the
Clinton:

1 Jbid., pp. 11-13.

THE OIL- AND GAS-PRODUCING ROCKS OF OHIO 827

Drift - : - : - - - - i feet
Berea grit, bottom at - - - - - - 460
Corniferous, Helderberg, and Niagara limestones,

top at - - - - - - - - R235

bottom at - - - - - - = toi
Clinton sand, top at - - - - - 2,02

bottom at - . = - - 2,045

Medina (red shales) top at - - - - 2,055
Bottom of wellat - - - - - - 2,075

This makes the red shales lying from 10 to 30 feet below the
Clinton sand, the top of the Medina.

The Lower Helderberg — Rocks of this age produce gas and
oil in one locality, viz., near Jefferson, Ashtabula county. The
succession of strata there is shown by the following record of
Webb Well No. 1. All measurements were made with a steel
line.

Thickness of

formation, Totalidepth.
Drift - - : - - - - - - 230teet. 23 feet.
Ohio shales - - - - - - - 1,671 1,704
Corniferous and Lower Helderberg limestones - 288 1,992
Gas sand at - - - - - - - - - == 1,902

The sand, which is from 30 to 40 feet in thickness, is inter-
bedded in the Lower Helderberg limestone. Similar conditions
are found in Lucas and Wood counties, in the former of which
the sand has been quarried for the manufacture of glass.t In
the gas field under consideration the sand is moderately coarse,
has a light color, and is highly fossiliferous. The gas is found
in the top of the formation, and just below lies a large and
threatening reservoir of salt water. A number of towns and
villages are being supplied with fuel from this field. The oil
wells are few in number and insignificant in production.

THE DEVONIAN.

The rocks of this age found in Ohio, have been classified by
Orton? as follows:

3. Ohio shales” - - - - - 250-3,000 feet
2. Olentangy shale - - - - 25
3. Devonian limestone (Corniferous) 75

WOid ss Den ligis 2 /bid., pp. 4 and 18-26.

828 J. A. BOWNOCKER

None of these formations produce oil, and the Ohio shales
alone yield gas. The available supply of this, however is very
small, and its use has been limited to domestic purposes. The
principal counties have been those along the lake shore in the
northeastern corner of the state. These shales, which underlie
the eastern half of the state, are wedge-shaped, with the apex
reaching from the lake to the Ohio. The cities, Columbus,
Delaware, and Bucyrus, lie on or near this apex. Eastward
the formation thickens rapidly and near Wellsville on the Ohio
river has been penetrated to a depth of 2,600 feet without
reaching the base. This feature has led to much confusion on
the part of the driller, who has expected to find the interval
between the Berea grit and the Devonian limestone, the same in
the eastern part of the state that he did in the central.

The gas secured in these shales is not from any one horizon,
but varies stratigraphically from place to place. The wells are
all small. Very commonly the shales make a show of gas, but
usually the yield is so light as to be commercially valueless.
The aggregate amount contained, however, must be very large,
and if it could be collected would form one of the most valuable
supplies of fuel in the state."

THE CARBONIFEROUS.

Classifying the formations of this great division, as has been
customary, into the Lower Carboniferous, Coal-measures, and
Permian, it is found that the oil- and gas-bearing rocks are
restricted to the first two members. These will now be con-
sidered in order.

I. THE LOWER CARBONIFEROUS.

The Berea grit.—This is the most extensive sandstone of the
state. Its area above and below drainage is about 15,000 square
miles, or more than one-third of the area of the state. Its value
is commensurate with its extent. ‘‘Its economic value above
ground is great, but it is greater below. In its outcrop it is a
source of the finest building-stone and the best grindstone grit

'For a full discussion of this subject see Geological Survey of Ohio, Vol. V1. pp.
410-42.

THE OIL- AND GAS-PRODUCING ROCKS OF OHIO 829

of the country, and when it dips beneath the surface it becomes
the repository of invaluable supplies of petroleum, gas, and salt
Water.

The composition and structure are very constant. The color
is gray below drainage, but has a tinge of yellow above. The
sand is of moderate fineness, and composed almost wholly of
silica. Occasionally it contains carbonate of lime, probably as
a cement, but this never constitutes more than a very small frac-
tion of the formation. That found below the surface around the
village Berea in Cuyahoga county is undistinguishable from that
obtained at a depth of more than 2,000 feet in Washington

‘county. Sometimes, though not usually, the formation divides,

a bed of shale lying between two of sandstone; at other places
the upper portion of the formation consists of sandstone and the
lower of shales. In thickness the formation varies from 50 feet
or more in the northern part of the state to a half dozen or less
in the southeastern portion, and occasionally disappears entirely,
its place being occupied by shales. The Berea grit is succeeded
above by the Berea (Sunbury) and Cuyahoga shales, having
usually an aggregate thickness of from 500 to 600 feet, and
below by the Ohio shales, having a great and rapidly increasing
thickness eastward. The most remarkable character of the
formation, however, remains to be mentioned, viz., its persist-
ence. From its outcrop it has been followed by the drill from
county to county, and often from section to section, until the
eastern and southeastern limit of the state is reached. It is as
easily recognized below drainage as above, and this character
makes it a stratigraphical landmark of great value to both driller
and geologist. In many counties in the eastern part of the state,
especially those fronting on the river, numerous efforts have been
made to find a productive sand below the Berea, but in every
case this effort has been unsuccessful. It may now be taken as
having been demonstrated by the drill that when the Berea sand
has been passed in this territory the last hope of oil or gas has
gone.

While a trace of oil or gas has perhaps been found in every

/bi2., Vol. VII, p. 28.

830 J. A. BOWNOCKER

county where the formation exists, the production in commercial
quantities is limited to Lorain, Medina, Trumbull, Columbiana,
Stark, Jefferson, Harrison, Belmont, Guernsey, Monroe, Noble,
Vinton, Perry, Athens, Morgan, and Washington counties.

The Logan group.—The relative position of this group and the
Berea is shown by the following skeleton record of a well on the
Mead farm in the Elk Run Pool, Washington county.

Sapa ; ; : top at I, 20OMeet.
bottom at 1,280
Maxton sand = - ie ae ee
J bottom at 1,500
Logan ' Mountain limestone - { top at 1,510
group | (‘“ Big lime’’) : : i bottom at 1,545
’ ! top at 1,560
[ Big Injun series - - bottom satin a6
\ top at 2,124

Berea grit - : ; ; ) \ | bottom at,” 25138

This shows that the two formations are about 400 feet apart,
and as has already been stated the interval is occupied by the
Berea and Cuyahoga shales.

The Logan group, as classified by Orton, consists of three

members —a conglomerate, sandstone, and shale —and has a
maximum thickness of 350 feet.7 Quite recently, however, Pro-
fessor Prosser has considered the question, and he divides the
group as follows :?

2. Logan formation = Logan sandstone and shales.

1. Black Hand Formation = Logan conglomerate.

Recent work by the drill demonstrates that the maximum thick-
ness is twice that stated by Orton. The group is limited above
by the Lower Carboniferous limestone, but this is rarely recog-
nized in deep wells, and consequently the upper limit is usually
uncertain. The first member of the group that is recognized by
the driller isthe Salt sand. The relation stratigraphically of this
member to the Lower Carboniferous limestone is shown by the
following record of a well on the Longshore farm in section 15
of Brush Creek township, Muskingum county. All measure-
ments were made with a steel line.

* Ibid., Vol. VII, pp. 4, 5, 32-5. 2 Jour. GEOL., Vol. IX, pp. 205-31.

THE OIL- AND GAS-PRODUCING ROCKS OF OHIO 831

Pom top at 20 feet.
Putnam Hill limestone = bottom at 23
top at 225
Lower Carboniferous limestone bottom at 265
{ top at 450
Salt sand - - | bottom at 485
top at 919
Berea shale z + 1 bottom at 952
: top at 950
Crearenit gy on i bottom at 980
h j top at 980
Bedford shales (brick red) - | bottom at 1,010
Ohio shales, top at - = 7 Es =, Hehe

This shows that the Salt sand lies 185 feet below the Lower
Carboniferous limestone. Combining this with the record of the
Mead well previously given makes a total thickness for the
Logan group of 715 feet. Probably in the extreme southeastern
corner of the state the thickness is still greater.

Above drainage the group is well marked. Hills capped
with sandstone or conglomerate stand out in bold relief, so that
the limits of the formation are discernible from a distance. The
conglomerate is the most conspicuous member of the group, It
is essentially a quartz rock, the largest pebbles of which do not
commonly exceed one-half of an inch in diameter. It is the
best known bridge stone in the state. The sandstone member
has usually a yellow or brown color, but sometimes this becomes
strikingly variegated. It is extensively used for building pur-
poses. The members of this group are much less constant in
their physical characters than the Berea grit, and hence their
identification is correspondingly more difficult.

Under cover the Logan group undergoes important changes,
and the several formations are given different names from those
at the surface. Thus, instead of the Logan conglomerate, sand-
stone, and shale, we have the Big Injun sand, Mountain lime-
stone, Salt sand,etc. The correlation of the strata bearing these
two sets of names is as follows. This is based on their strati-
graphical succession and lithology. While the limits under-
ground cannot be sharply drawn in most cases, they are perhaps
as definite as those along the line of outcrop.

832 J. A. BOWNOCKER

Logan shales =shales (unnamed)
Logan sandstone = Salt sand

__ Mountain limestone
~ Big Injun series

Logan conglomerate
The limestone just named, and which is known also as the
‘Big lime,’’ has a maximum thickness in this state of IIo feet.
It is a light-colored, hard, massive rock, free from oil and gas
except along its margin, where the formation becomes broken
and the layers of limestone are intercalated with sandstone. The
formation is wedge-shaped with the apex to the northwest. It
is limited to the eastern half of Washington and Monroe counties,
and the southeastern corner of Belmont. To the west and north
its place is occupied by shales and sandstones. The formation
divides the Logan group into two unequal parts, and serves as a
guide-post to the driller.
The Big Injun series—This consists of the following
members:

Slate - - o- 20 feet
Keener sand@- - o- 60
Slate - - O- 25
Big Injun sand 0-175
Slate - - o-— 10
Sguaw sand - O- 30

From this it is seen that the group varies greatly. Sometimes
it is little more than one great mass of sandstone, while at other
times it is broken into a series of alternating beds of slate and
sandstone. The Keener sand occasionally lies immediately below
the Mountain limestone, but more often is separated from that
rock by a few feet of shales. It varies considerably in texture,
but is usually coarse and open, sometimes conglomeritic. The
sand was named from the Keener farm near Sistersville, West
Virginia." It is an important source of oil in Monroe and
Washington counties. The sand is separated from the under-
lying Big Injun by a bed of slate. Sometimes the latter is want-
ing, and then the two sands run together and are conjointly
called the Big Injun. The sand in question (Big Injun) is by
far the thickest member of the series, but in other respects

' West Virginia Geological Survey, Vol. I, p. 357.

THE OIL- AND GAS-PRODUCING ROCKS OF OHIO 833

resembles the Keener. It is recognized in several counties in
southeastern Ohio, but is a producer of oil or gas in commercial
quantities in Monroe and Washington counties only. To the
west and north the formation becomes too broken to be a
repository for oil or gas. Below the Big Injun, and separated
from it by a thin bed of shales, there is occasionally found
another layer of sand known as the Squaw. It is decidedly
patchy and never extends over large areas in this state. The
best records of it are reported from Monroe county, but it is of
little importance even there.

The Salt sand.—This is the highest of the sands of the Logan
group. It has a gray color, is moderately coarse, and nearly
always is charged with brine. Occasionally it contains a little
oil and gas, but it cannot be recognized as a producer of either.

Il. THE COAL-MEASURES.

While a large number of strata belonging to the Coal-meas-
ures have been or are now sources of oil, comparatively few of
these have been important ina commercial sense. Generally the
sands are local, and cannot be traced over an area of more than
a fraction of amile. Such sands usually have names, but because
of their small area and production they will not be further noticed
in this article. The most important sands of the Coal-measures
are the following:

. Goose Run sand.
Mitchell sand.

. First Cow Run sand.

. Macksburg 500-foot sand.
. Second Cow Run sand.

me NW FN

The relative positions of sands 1 and 3 are shown by the fol-
lowing data taken from Centennial Well No. 6 at Cow Run:

Thickness De gto Op
Pittsburg coal - . - - 1 foot 11 feet
First Cow Run sand - - ay feet 325
Second Cow Run sand - - 64 776

The stratigraphical relation of sands 2 and 3 1s shown by the
following record taken from Dunn Well No. 6, near Macksburg.
The well head is six feet below the Meigs Creek coal.

834 J. A. BOWNOCKER

Depth to top of
Thickness, formation.
Ames limestone, - - - 1 foot 285 feet
First Cow Run sand - - en Gefeet 385
Cambridge limestone - 4 AII
Dunkard sand (300-foot) - - 95 530
Macksburg 500-foot sand - 22 702

The Dunkard or 300-foot sand is quite generally regarded as
the equivalent of the Mahoning. That this cannot be correct is
shown by the following partial section of the Hocking Valley
coal field."

Thickness,

Cambridge limestone - - - - 2-10 feet
Mahoning sandstone, upper division, - 50
Brush Creek coal (No. 7a) - - - 2%
Brush Creek limestone - - - 4
Mahoning sandstone, lower division - 15-25
Upper Freeport coal - - - - O- 3

} Clay
Upper Freeport { Shales - = - 35

| Limestone and sandstone
Middle Kittaning coal (No. 6) - - 5-13

This shows the Mahoning sandstone lying immediately below
the Cambridge lime, but experience in the field demonstrates that
the two are not ordinarily in contact. Measuring from the base
of the lime rock to that of the Mahoning, an interval of 76 feet
is found, while, according to the records in the Dunn well, the
interval is 210 feet. A divergence of this sort cannot be
explained by assuming that the section expands eastward, for a
study of the relative positions of the Pittsburgh coal, the Ames
and Cambridge limestones farther west with the same formations
near Macksburg, shows no material change, and it is certainly
unreasonable to assume that the formations just below the Cam-
bridge limestone expand at anything like the rate that would be
necessary to make the interval which this nomenclature requires.
Further, naming the sand in question the Mahoning makes impos-
sible a rational classification of the lower formations. The position
of the sand with reference to the Cambridge lime is that of the
Upper Freeport.

* Geological Survey of Ohio, Vol. V, p. 918.

THE OIL- AND GAS-PRODUCING ROCKS OF OHIO 835

The positions of sands 3 and 4 are shown by the record of
Centennial Well No. 7, at Cow Run. The well head is at the
horizon of the Meigs Creek coal.

Depth to top of

Thickness. formation.
Pittsburg coal (No. 8) - ant 116 feet
Mitchell sand - 5 - 25 205
Red shales (“ Big Red”’) - - 80 235
First Cow Run sand - . 32 423

The relative positions of sands 4 and 5 are shown by the fol-
lowing taken from Reed Well No. 4, near Marietta:

Goose Run sand E top at 300 feet
bottom at 331

Mitchell sand - top at 525
bottom at 546

The Second Cow Run sand.—This is one of the least important
members of the group now under consideration. It has been a
small producer at Cow Run in Washington county for more than
thirty years, but rarely has been found beyond that locality. It
is reported, however, quite frequently, for the driller gives this
name to almost any sand lying from 100 to 500 feet below the
first Cow Run sand. As may be seen from the records given the
interval between the two is 400 feet. It lies 760 feet below the
Pittsburg coal, and is the thickest member of the group, sometimes
exceeding 60 feet. Occasionally the formation is divided by
few feet of slate, in which case the oil les in the lower part.
The sand possesses no qualities that serve to distinguish it from
the higher members.

The sand belongs near the base of the Coal-measures. The
partial record of the Rice well, given below, shows it to be the
first sand above the Salt sand, the two being separated by 79 feet
of shales, the latter probably the equivalent of the shales of the
Logan group. This, with the thickness of the formation, and the
fact that it is sometimes divided by a few feet of shales, makes
quite certain its identification as the Massillon sandstone.

The Macksburg 500-foot sand.— This is important at Macksburg
and vicinity only. It lies 670 feet below the Meigs Creek coal

836 J. A. BOWNOCKER

and about 580 feet below the Pittsburg, as is shown in the
following partial log of George Rice Well No. 18, at Macksburg.'

To top of
Thickness. formation.
Meigs Creek coal - - - 5 feet 10 feet
First Cow Run sand - - 35 343
Dunkard sand (300- oe Sy 7S 554
500-foot sand - - 17 685
Sand, pebbly (800- eae - =) 5 775
Slate - - - - 79 826
Salt sand - - - - - 190 905

This sand commonly ranges from I0 to 30 feet in thickness.
It is usually quite coarse, but does not become a conglomerate.
Like the First Cow Run sand, it does not form a continuous
stratum, but is decidedly patchy. Stratigraphically considered,
the formation belongs near the top of the Conglomerate Coal-
measures, and its position with reference to the Cambridge lime
and Dunkard sand strongly indicates that it is the Tionesta sand-
stone.

The Second Cow Run and the Macksburg 500-foot sand have
been regarded by many as equivalent. As has already been
suggested by Professor White, however, this cannot be the case.”
Examination of the record of the Rice well shows that the inter-
val between the First Cow Run and 500-foot sands is only 307
feet, while, as already stated, the interval should be 4oo feet.
Measuring from the Meigs Creek coal, equally conclusive figures
are secured. Thus the Second Cow Run sand lies 840 feet
below the Meigs Creek coal, while the interval between this seam
and the 500-foot sand is only 670 feet. The sand at Macks-
burg known as the 800-foot is probably the equivalent of the
Second Cow Run.

The first Cow Run sand.—This is the most important and
best-known sand of the group. As is shown in the record of the
Dunn well, its position is 100 feet below the Ames limestone. In
western Morgan county, near the outcrop of the sand, the inter-
val ranges from 70 feet to 100. The Ames limestone there lies
170 feet below the Pittsburg coal. The identification of the

*West Virginia Geol. Sur., Vol. 1, p. 298.

? West Virginia Geological Survey, Vol. 1, p. 299.

THE OIL- AND GAS-PRODUCING ROCKS OF OHIO 837

sand is made more certain by the fact that the Cambridge lime-
stone lies immediately below or is separated from the sand by a
thin bed of shales. Unfortunately these limestones are not
recognized in Monroe and the eastern part of Washington
counties, and hence the sand in question often cannot be identi-
fied there with certainty. In such localities any shallow sand,
especially if it makes a show of oil or gas, is known as the Cow
Run. The relation of the sand to the two limestones places it
about 100 feet above the base of the Conemaugh formation.

If the Berea grit be taken as the type of a persistent stratum
of sandstone, the First Cow Run sand may properly be selected
as the representative of the opposite type. It is the most patchy
of the oil or gas rocks of Ohio. The maximum thickness of
the formation may be taken at 50 feet, but even a short distance
from this the stratum may become very thin or actually dis-
appear entirely. The texture varies much, and where product-
ive is sometimes conglomeritic. Pebbles three-fourths of an inch
in diameter have been found, and those one-fourth of an inch
are common. The pebbles consist of quartz, but small grains of
other minerals are found. From this type the formation changes
to a hard compact sandstone.

The sand is an important source of oil in Morgan and Wash-
ington counties. It seems to have been first struck at Macks-
burg in 1860, where it is known as the 140-foot sand ; this being
the depth at which the sand was found in the valley at that
place. In 1861 the sand was found at Cow Run and has there
been a producer ever since. It is seen that the latter name has
not the claim of priority, but it is so much more widely used
that it is here retained.

The Mitchell sand.—This is the source of a small oil field a
few miles east of Marietta. The sand is comparable in thick-
ness with the First Cow Run, but in texture is less conglomer-
itic. As has been shown, it lies about go feet below the Pitts-
burg coal, or 200 feet above the First Cow Run sand. _ Its place
is near the summit of the Conemaugh formation. Wells in this
sand commonly begin with a relatively large production, but
decrease very rapidly.

838 J. A. BOWNOCKER

The Goose Run sand.— The formation known by this name has
supplied a score or more small wells near Marietta. The sand
is patchy, and the life of the wells very short. In fact, the rock
is of little value, and is recognized here simply to make the
record complete stratigraphically. Lying nearly 200 feet above
the Mitchell, the sand belongs 100 feet above the Pittsburg coal,
or, in other words, near the middle of the Monongahela forma-
tion.

The following table shows the great divisions of the Coal-
measures of Ohio, and also the approximate positions, at least, in
these of the several sands just discussed :

Dunkard formation, or Upper Bar-

No oil or gas rocks.
ren Coal-measures, 500 feet.

Monongahela Formation, or Upper

c Too feet.
Productive Coal-measures, 200 feet.

Pittsburg or No. 8 coal.
go feet.

Mitchell sand.

200 feet.

First Cow Run sand.

Conemaugh formation, or Lower
Barren Coal-measures, 500 feet.
Cambridge limestone.
Mahoning sandstone.
( Upper Freeport or No. 7 coal.
| Dunkard or 300-foot sand=?

) Goose Run Sand.
|
|
|
5
|
|
|
NS

Allegheny Formation, or Lower
Productive Coal-measures, 250 feet. Freeport sandstone.
50 feet.

Macksburg 500-foot sand= ? Tion-
esta sandstone.

70 feet.

Second Cow Run sand = Massillon
sandstone.
| Sharon or No. 1 Coal.

Conglomerate Coal-measures, 250
feet.

——_—__A._____, t+

J. A. BowNOocKER.
COLUMBUS,
October 10, 1902.

Hb NVAP AOE GLACIER IN 1902":

In the front range of the Rocky Mountains of Colorado are
several remnants of glaciers. One of these, the Arapahoe gla-
cier,? has an area at present of about one-half square mile. It
lies on the eastern slope of the continental divide, twenty-one
miles due west from the city of Boulder. It is inclosed by
nearly vertical walls five to seven hundred feet in height, form-
ing a deep cirque opening to the east. The valley leading east-
ward is that of a branch of North Boulder Creek. This valley
has been occupied by a glacier to a distance of at least eight
miles to the east. The bottom of the valley for four miles east-
ward from the present glacier is occupied by a series of lakes
resulting from the glacial action and the intervening topography
is; matked by roche moutonees (Fig: 1). “The tront of the
glacier has retreated up this valley to its present position, and
its length has been thereby reduced from nine miles to one mile.

The features of this glacier at the close of the summer of
1902 have a peculiar interest. The snowfall for the past three
winters has been deficient, and the melting in the ensuing sum-
mers has been excessive. The results of these climatic circum-
stances appear in a great contraction of the ice-covered area,
an unusual exposure of fresh moraines and, what is still more
important, an almost complete absence of snow below the névé.
This last condition is responsible for many fine exposures reveal-
ing the stratification of the ice, and for the complete uncovering
of crevasses, some of which stand open as much as ten feet.

Moraines—The bottom of the valley is far from being a
single trough, and the glacier is therefore by no means a single

’ The observations on which this account is based were made in the last week of

August. The photographs were taken by Judge Junius Henderson, of Boulder, Colo.
The map is constructed from a survey made by Hugh F. Watts, of Boulder, Colo.

2For a general description of this glacier, its position, dimensions, shape, and
evidences of its true glacial character, see paper by WILLIs T. LEE, this JOURNAL,
Vol. VIII, p. 647.

839

840 . N. M. FENNEMAN

ice stream. In its more vigorous days, when it filled the valley
for some miles and was some hundreds of feet deeper than at
present, the inequalities of its bed must have been unable to
control the ice movement to any great extent, and therefore
failed to subdivide the glacier into a system of branching tribu-

Fic. 1.—Looking east, down the valley, from a point one-half mile below the
present terminus ofthe glacier. The shelf seen at the left is in the main rock terrace.
The lower and narrower valley suggests uplift and considerable erosion before the

glacier was developed.

taries. It must then have appeared essentially as a single ice
stream. Indeed, its appearance, as viewed from the neighboring
mountains two years ago,’ seems to have suggested but two
branches, separated by a moraine so low that a few extra yards
of depth would have rendered the whole a unit at the surface.
The valley bottom below the present terminus is barred with
crescent-shaped terminal moraines which mark the termini of

T WILLIS TL. LER, loc. ce.

THE ARAPAHOE GLACIER 841

the small glaciers into which the larger ice mass became divided
in its dying stages. Other crescentic moraines are now just
becoming exposed at the present border of the ice, the horns of
the crescents being in the positions of medial moraines which
were formed on the valley bottom between tributary glaciers,
and are now for the first time being exposed by the wasting of

: A
E ARAPAHOE PEAK. _

OF THE

~ ARAPAHOE GLACIER —

AUGUST 1902.

Ly mite |

HiiGye2:

the ice. Several of these medial lines are distinctly terminal in
their relations to the ice on either side, as shown by the fact
that the movements of the ice are not in the direction of such
lines, but normal to their trend from both sides. (One such is
well shown on the map, Fig. 2.)

The moraines whose positions indicate that they have been
for a long time exposed are composed, not of rounded bowlders,
but of large fragments from among which all smaller stones
have been removed. This statement is made of those near
the present limit of the ice, and therefore formed since the

842 N. M. FENNEMAN

glacier became very short. Those situated some miles down
the valley contain many rounded bowlders, for the obvious rea-
son that many of these have traveled far. Between the moraines
of great angular fragments and the ice, are newer ridges whose
constitution is in striking contrast with that of the outer ridges.
These newer deposits contain fragments of the same order of
size as those contained in the older ridges, but intermixed with
them are stones of all smaller sizes. In these ridges, often cap-
ping the several fragments or lying in masses at the very sum-
mit, are quantities of mingled mud and gravel so fresh that they
bear almost no evidence of having been washed by the rains of
a single season. They resemble nothing so much as the fresh
dump from a great steam shovel. On the north side of the
glacier such a ridge rises from forty to fifty feet above the present
level of the ice, showing that the surface of the glacier has wasted
by atleastthat amount since the ridge wasformed. Theunwashed
character of its summit makes it difficult to admit that more than
a single season has passed since its first exposure to subaérial
weathering. This necessitates the conclusion that the glacier
has wasted at least forty or fifty feet during the past summer.
The recent contraction of the glacier is further evidenced by
the exposure of ice in one of the fresh morainic ridges now well
separated from the body of the slacier (Higi3)| | Super
ficially this ridge differs not at all from any other moraine of
recent formation. Except for the fortunate section which
reveals its core, it would appear to be built of a framework of
huge fragments, from among which the finer materials have been
but partly removed by surface erosion. The section, apparently
due to stream erosion, exposes a surface of perhaps twenty or
thirty square yards of ice. The position of the stratification
planes, dipping toward the head of the glacier, agrees with the
supposition that these ice masses may still be in place, having
been covered and protected by the abundant débris at the ice
front. A second ridge of similar character, only partly sepa-
rated from the glacier and several hundred feet within the first
ridge, is plainly the result of the present summer’s melting. The
front of the ice was early in the season completely covered with

THE ARAPAHOE GLACIER 843

rock débris. This covering formed such a protection that the
outer rim of the glacier wasted more slowly than the ice imme-
diately in the rear. A trough has therefore begun, which in
another season like the one just past, may grow to such an extent
as to leave the inner ridge completely detached from the glacier.

Zi

Fic. 3.—Terminal moraine; the dark bands are ice. The rocks between the
dark bands probably overlie ice. The stratification of the ice is transverse to the
bands seen here.

The third steep ice front behind the second trough is already
becoming covered with débris. A series of years of deficient
snowfall or excessive melting, or both, might in this way produce
a series of concentric ridges having ice cores, whose subsequent
melting would leave a topography very much more choppy than
that which would result from a series of ordinary recessional
moraines. It is suggested that a part of the very confused
morainic area several miles east of Silver Lake may thus be con-
nected with the wasting of the glacier at an excessive rate.

844 N. M. FENNEMAN

Crevasses.— A second great advantage from the recent defi-
ciency of snowfall and the excess of melting was found in the
complete uncovering of crevasses. No part of the glacieris free

Fic. 4.—Crevasse, open about Io feet.

from cracks, but near the middle a greater steepness of the bed
gives rise to crevassing on a large scale (Figs. 4 ands).
The slope of the ice surface for some distance at this place was

THE ARAPATIOER GLACIER 845

measured at twenty-four degrees, which is greater than the slope
at any other place excluding the névé and the immediate margin
of the glacier. Many of the crevasses in this zone are forty rods
long, and some stand open as much as ten feet. They are spaced
at intervals of a few yards. The blocks thus formed appear to
have been tilted forward. This is inferred from the fact that the

Fic. 5.—General view looking south over crevasses to the névé.

brink on the lee side of each crevasse is higher than that on the
opposite or stoss side, and the surface of each block has a steeper
slope than that of the ice as a whole in this vicinity. Lines
dropped into these crevasses did not reach depths greater than
fifty feet, but apparently water is running at much greater depths.
The appearance of the ice in these crevasses is clear and bluish,
with occasional bands which are stained as if with dust. In the
same zone are some small cracks, limited in length to seven or
eight feet, which stand open about two inches at the center.
Near the margin of the ice a system of longitudinal crevasses
is made quite prominent by the control which these exercise over

846 N: M. FENNEMAN

the surface drainage. They reach apparently no great depths.
Many of them, which are several feet in depth and contain run-
ning water, are open less than half an inch at the top. Much of

Fic. 6.—The Bergschrund, open about 15 feet. The columns at the right are ice ;
above them is snow. -

the surface drainage follows these narrow crevasses. That they
are primarily due to fracture and not to erosion is evident from
their absolute straightness for long distances.

A most striking feature of this glacier is the prominent Berg-

THE ARAPAHOE GLACIER 847

schrund, or gaping crevasse, at the line which may be taken as
the limit of the névé and the beginning of the glacier proper
(Bios 6) > The very steep slope of the névé, its whiteness,
and the solitariness of this great crack make it strikingly promi-
nent from any point from which the glacier is visible. It is
almost continuous around the cirque and stands open at many
places as muchas ten or fifteen feet. A partial descent into this
opening revealed the fact that it traverses from ten to twenty feet
of snow at the surface, below which is a colonnade of great icicles
lining a chasm at least fifty feet deep. Independent of these
icicles the walls of the chasm below the ten or twenty feet of snow,
seem to be of the same blue ice which is seen in the crevasses at
the center of the glacier.

In the examination of crevasses the excessive melting of the
past summer proved a most fortunate circumstance. Probably
every crevasse down to the most minute was uncovered. The
real significance of this appeared after a snowfall of not more than
two inches on August 28. Onthe following day many crevasses,
having widths approximating ten inches, were found completely
covered by crusts of snow. It seemed evident from this that a
snowfall of a very few inches would easily obscure crevasses in
which a man might be lost.

The complete absence of surface snow was of great advan-
tage in observing the stratification of the ice. Dark lines which
were nearly horizontal and continuous for long distances were
plainly visible for some miles. These mark the outcrops of ice
strata. The alternate strata are of clear ice, while the inter-
vening ones have a more spongy texture, the constituent beds
being for the most part not thicker than two inches and usually
less than one inch. The wasting of layers of varying hardness
produces small cliffs and platforms analogus to those observed
in canyons cut in stratified rock. The fine débris accumulates
along the lines of these outcrops and forms the dark lines which
are visible for great distances. The dip of stratification planes
near the margin of the ice is well seen in the canyons of sur-
face streams and is uniformly toward the direction from which
them ices comes (Higa 7\s 0 Dips) of thirty degrees on more

8438 NN. M. FENNEMAN

are common a hundred yards from the edge and the steepness
increases as the margin 1s approached. Englacial drift disposed
in planes was found at only one point and that was in the buried
ice. The material incorporated here contained nothing coarser
than sand. Planes showing discoloration, as if by dust accumu-
lated on the surface, are not uncommon. In the large: crevasses

Fic. 7.—Stream channel showing stratification of the ice—r1oo yards from front.

these are sometimes very prominent in contrast with the clear
bluish ice.

The drainage of the glacier has no special peculiarity, and in
this lies its chief interest, because it aids in classifying this ice
field with well-developed alpine glaciers. Streams are guided
largely by crevasses and frequently drop suddenly into the body
of the glacier. Pits are not uncommon; one of these, whose
opening at the surface was less than ten square inches, was found
toi be three teet deep:

THE ARAPAHOE GLACIER 849

Where the ice is snow-covered and on the smaller segments
which do not take on the nature of perfect glacier ice, there
often results from the surface drainage a peculiar hummocky
surface. The process by which these hummocks or knobs come
into existence is somewhat indirect. The water draining froma

very steep surface of uncompacted névé, produces at first a

Fic. 8.—Lake within the present terminal moraine. The prominent lines in the
ice at the left are shearing planes.

system of small parallel channels, which may be a yard or two
apart. The water in these channels percolates to some extent,
and in freezing produces under the channel a rib which is harder
and more ice-like than the snow between channels. Differential
melting next leaves what was once a channel exposed as a ridge.
This ridge has at no time an even crest, and soon comes to
appear as a line of small icy hummocks regularly spaced at
intervals of a few feet. The isolated patches of névé attending
the Arapahoe glacier, which are just too small to take on com-

850 N. M. FENNEMAN

plete glacial character, often show this knobby surface to a
remarkable degree.

The water issuing from below the glacier is characteristically
milky. Small pools are found whose bottoms cannot be seen
at depths greater than four inches. The turbid water escaping
from below is soon mixed with the clear drainage from the sur-
face, and the turbidity is thereby reduced; but the first few
lakes of the long chain in the valley below are notably turbid.

There are no records of the extent of this glacier in former
years. It is viewed from the rim of the cirque by occasional
visitors to Arapahoe peak, but visits to the glacier itself have
been rare. A few photographs taken from the surrounding
heights in former years would indicate that the loss of forty
or fifty feet in depth during the season just passed is not all to
be credited to the permanent wasting of the glacier. That
amount has plainly disappeared from the surface since last win-
ter; this conclusion seems necessary from the presence of
absolutely fresh mud on the top of the moraine which rises to
that height above the present ice surface. Former photographs,
however, also show this same ridge rising above the ice, though
by a less amount than at the present time. As stated above, the
past two or three years have been favorable to wasting. As the
glacier is now accurately mapped and largely photographed, a
few more years may yield the interesting determination whether
the present climate of this locality in the Rocky Mountains is
capable of supporting a permanent glacier or whether the recently
observed shrinkage is a part of its final disappearance.

A second consideration of theoretical significance is con-
nected with the rock terraces several hundred feet high on the
sides of the valley below the present glacier (Fig. 1). These
plainly indicate an uplift. It is equally plain that the glaciation
was not the result of the uplift. The valley cut below these
terraces is several hundred feet deep and five hundred to one
thousand feet wide. If the glaciation as well as the valley
trenching resulted from this uplift, then the small stream must
have cut out the above-mentioncd valley in the time required for
the ice to accumulate after the climate became suited to glaciers.

THE ARAPAHOE GLACIER 851

This assumption is evidently absurd, and it becomes necessary
to assume that a considerable interval elapsed between the
uplift and the advent of a glacial climate. The supposition may
still be made that the glaciation followed close upon another
and later uplift. The significance of the observations in this
particular terraced valley is not in excluding hypotheses of later
uplifts, but in making clear the time relations between the uplift
here exemplified and the coming on of the glacier.

N. M. FENNEMAN.
UNIVERSITY OF COLORADO.

THE FRANCEVILLE (EL PASO COUNTY, COLORADO)
METEORITE.

TuIs meteorite was bought by Professor F. W. Cragin, of
Colorado College, Colorado Springs, Colo., from Mr. David
Anderson and wife of Colorado Springs. It was found by Mr:
Anderson, as nearly as he can remember, in 1890, about one and
one-half miles southwest of the home ranch of Skinner and Ash-
ley, which is east of Franceville. According to Mr. Anderson,
it was totally above ground, and there were no signs of any other
meteorites.

From the time it was found until purchased by Professor
Cragin it had been kept in the home of Mrs. Anderson in Col-
orado Springs, half-forgotten. It was purchased from Professor
Cragin by Ward's Natural Science Establishment, Rochester,
New York, in August, 1902. From its external form it is one of
the most interesting of the mafy meteorites that have been in
the possession of that firm.

It is a decidedly flattened rhombic pyramid with a somewhat
sharp ridge extending around the center of the mass on the four
rhombic sides. The dimensions of the mass in these directions
are 21 23° with a thickness of 11.5°™. On one side of this
central axis the pyramid projects 6%™; on the opposite side,
5a5 0, aS, Semmes Miler 2:

The decidedly octahedral form of this iron seems unques-
tionably due to its separation along natural cleavage planes from
a much larger mass; but it is surprising that the form should
not have been much more distorted by the erosion due to fric-
tion in passing through the atmosphere.

The whole iron is more or less mottled, ranging in color
froma reddish-brown toa brownish-black. It is entirely covered
with pittings on all sides. Those on what may properly be
termed the upper side (Fig. 1) are much more distinct, owing
to their size and depth, than elsewhere. Just below the medium
ridge on one end there is an unusually large pitting, some 10™

852

THE FRANCEVILLE METEORITE 853

long and 2° deep. There are two or three small, but much
deeper, finger-like pittings from which troilite has undoubtedly
been weathered out.

Two small corners of the mass have been broken off and
have the appearance of very old breaks, as the surfaces are
entirely oxidized. These surfaces show markedly distinct octa-

Fic. 1.— Upper side, showing pittings. (Two-fifths actual size.)

hedral cleavage. Probably not more than thirty grams have
been taken from it since it reached terra firma. ‘Two small pro-
tuberances, one 2°, the other 1° in diameter, were sawed
off, and the faces polished and etched. One of these is shown
in Fig. 1, the other in Fig. 2. Otherwise the mass was entire
until sawed into sections.

Upon slicing the mass but one troilite nodule of any size was
found. This occurred on one end-piece and the adjoining slice,
and was 14™™ in diameter, with two small patches of nickel-

854 H. L. PRESTON

iferous iron in its center. The slices show fractures more or less
extending across their surfaces along the natural cleavage faces
which are the edges of the kamacite plates, and in some
instances the rhombic form produced by the Widmanstatten
figures, as is the case in the San Angelo meteorite,’ are strongly
outlined by these fissures.

Upon etching the iron the Widmanstatten figures are readily
brought out by acids. These are particularly sharp and clear,

Fic. 2.—Showing pyramidal form. (Two-fifths actual size.)

and of large size, as shown in Fig. 3. The kamacite plates
average from. to 1.5 27 ain diameter, with) an occasional ore
of 2™™". They are unusual in the fact that they extend -in/an
unbroken line in many instances from go to 120 ™ in length.
The taenite occurs in minute films between the kamacite plates.

The plessite patches are comparatively small for an iron of
such coarse crystallization. Some of these patches show no
structure when etched, except a slightly pitted surface, while
others are prominently made up of alternate layers of kamacite
and taenite, producing sharply the so-called Laphamite lines.

Schreibersite is not visible on the etched surfaces microscopi-
cally, not even surrounding the troilite nodules, as is usually the
case.

t American Journal of Science, Ser. 3, Vol. V, pp. 269-72.

DAE AMCAN CE VILLE ME TEORTLALS 855

Mr. John M. Davison, to whom was given 23.9 grams for
analysis, reports:

The specific gravity of this siderite is 7.87. A proximate analysis gave :

Soluble in hydrochloric acid, kamacite and taenite, 99.16 per cent.
‘Combined-carbon, not. determined. © 9... © ..,4%.
Undissolved in hydro- j senreibctsite ate On
Enigio acd graphite and silicates (trace) .003
platinum (from 23.9 grams) trace

100,

The analysis of kamacite and taenite together gave Fe. 91.92 per cent.

Ni S513

100.05 per cent.
The weight of this most interesting siderite is forty-one
pounds and six and one-half ounces, or 18.3 kilograms.
Colorado has not been prolific in supplying meteorites to the
scientific world. As far as noted, there have been but five,
including the present iron, all of which are siderites, viz. :

Russel Gulch, Gilpin county = : = found 1863
Bear Creek, near Denver - - - found 1866
Jefferson, thirty miles from Denver - - fell June, 1867
Franceville, El] Paso county - = - found 1890
Mount Ouray, Chaffee county - - - found 1894

@ne of these, the: “Jetierson (No. 81, Shepard Collection),
thirty miles from Denver, Colo.,” listed as having fallen in June,
1867, I can find no account of outside of the descriptive cata-
logue of the meteorite collection in the United States National
Museum, January, 1902. It seems apparent that a mistake has
been made in labeling this specimen, and it must be dropped asa
distinct fall for the following reasons:

The Bear Creek meteorite has been noted in most catalogues
as having been found in Denver county, Colo., which is also a
mistake, as Colorado has no county by this name. The Bear
Creek meteorite was first mentioned by Shepard’ as found upon
the eastern slope of the Sierra Madre range of the Rocky
Mountains, and Henry’ notes it as found in a deep gulch near

t American Journal of Science, Ser. 2, Vol. XLII, pp. 250, 251.

2 [bid., pp. 286, 287.

856 FL Le PRES LOIN:

Bear Creek, about twenty-five.or thirty miles from Denver.
Smith3 in describing this meteorite gave it the name of Bear
Creek. As Denver is on the boundary line between Arapahoe
and Jefferson counties, and as there is a Bear Creek extending
clear across Jefferson county from west to east, emptying into
the Platte, according to Henry’s description, the locality of the

Fic. 3.—Section showing Widmanstatten figures. (Three-fifths actual size.)

Bear Creek meteorite must lie in the west central part of Jeffer-
son county.

Therefore it seems likely that the iron noted in the Shepard
Collection as “‘ Jefferson, thirty miles from Denver,” is in reality
a portion of the Bear Creek meteorite labeled “ Jefferson,”
meaning Jefferson county, and that the date of fall, June, 1867,
is an error; particularly so, as the Bear Creek is described by
Henry’ as being ‘‘shattered on one end,” so that small pieces
could be readily detached. ‘‘ Denver county” has evidently
been substituted itor “Denver city ’ in many of ‘the meteorite
lists, as no county is given in any of the early reports of the
Bear Creek meteorite. Moreover, the Sierra Madre range is
west of Denver, and the Bear Creek meteorite is described as
having been found on its eastern slope, which in all probability

8 Zid., Vol. XLIII, pp. 66, 67.

THE FRANCEVILLE METEORITE 857

would bring it in Jefferson county. So it would seem best that
“Jefferson” should be discarded entirely as a distinct fall and
be called ‘‘ Bear Creek,’’ and that ‘Denver county”’ in all
meteorite lists should read ‘‘ Denver city.” Thus we reduce the
Colorado meteorites to four distinct falls.

H. L. PRESTON.
ROCHESTER, N. Y.

STUDIES FORD LUDENES

THE MAPPING OF THE CRYSTALLINE SCHISTS:
II. BASAL ASSUMPTIONS.

Introduction.
Deformation under compressive stresses either by flow or by rupture.
Deformation zones of the lithosphere.
Mechanics of deformation by flow.
Mechanics of deformation by rupture.

Systems of faults.

Direct genetic relations of joint systems to systems of normal faults.

Criteria for recognizing a system of folds.

Criteria for recognizing a combined system of joints and faults.

Key areas for determining manner of deformation of a region.

Nature of evidence for establishing a fault.

Nature of evidence for establishing a system of faults.

Method of mapping an area deformed by a system of folds in conjunction with a
system of faults.

INTRODUCTION.

In an earlier paper bearing this general title? the writer has
discussed the methods of mapping the crystalline schists, and
taken occasion to emphasize the necessity for a larger use of
outcrop maps for all complexly deformed areas. It was also
advocated that many peculiarities of rock exposures besides
those ordinarily observed be examined and recorded, and their
significance be sought in the final interpretation—the drawing
of boundaries and the coloring of the map. In the present
paper it is proposed to go somewhat deeper into the subject,
and to discuss the basal and perhaps unconscious assumptions
of the geologist in the making of his map.

If the writer’s contention is upheld, it will be shown that
methods of mapping have not kept pace with the advance of

«Published with the permission of the director of the United States Geological
Survey.

2Jour. GEOL., Vol. X, 1902, pp. 780-92.

858

WIA? PING OT LITLE ORM S TALE TINE SC EUS TGS: 859

the science of geology, and that, whereas the mechanics of
crustal deformation, as now understood, and observation alike,
take account of two distinct methods of rock deformation, geo-
logical mappers seem generally to have regarded but one as
possessing much significance. The subject is an intricate one
and one which does not readily lend itself to absolute demon-
stration, but in another place concrete examples will be chosen
and described in some detail to show that the subject is a much
larger one than is ordinarily understood.

DEFORMATION UNDER COMPRESSIVE STRESSES EITHER BY FLOW
OR BY RUPTURE.

The stresses within the earth’s crust which bring about suffi-
ciently important deformation to affect the tectonic structure of
an area may be assumed to be vertical compression due to the
weight of superincumbent material, and lateral compression due
to crustal shortening from whatever cause,’ but in the last
analysis a result of gravitation. Tensile stresses are locally pro-
duced when rocks assume new attitudes under compressive
stress, but their importance as a direct cause of deformation
over considerable areas is probably subordinate and largely
limited to the effect of temperature change in surface masses
of igneous rock, and hence may largely be dismissed from con-
sideration in the present discussion.

The generally accepted classification of displacements result-
ing from earth stresses has been thus expressed by Suess :?

The visible displacements3 in the earth’s superstructure of rock are the
result of movements which proceed from the diminution of volume of our

‘VAN Hise, “Estimates and Causes of Crustal Shortening,” JouR. GEOL.,
Vol. VI (1898), pp. 10-64, also, “Earth Movements,” Zvans. Wis. Acad. Sct., etc.,
Vol. XI (1898), pp. 465-516.

2Surss, Das Antlitz der Erde, 1885, Vol. I, p. 143. See also, MARGERIE ET
Heim, Les dislocations de 1’ écorce terrestre, Zurich, 1888, p. 8.

3The double meaning of the German and French words Dislocation and disloca-
tion is especially unfortunate. In continental usage “dislocation” is employed in its
more general sense for ‘‘ displacement” (as for example, in the title of Margérie and
Heim’s work), and quite generally also in a special sense for “‘disjointing,” or equiv-
alent to ‘“‘faulting.” American usage is, I believe, wholly restricted to the word
applied in the latter sense.

860 SLUDIES FOR STUDENTS

planet. Strains brought about by this process indicate an inclination to dif-
ferentiate themselves into tangential and radial strains, thereby inducing
horizontal (that is, compressive and folding), and vertical (that is, depres-
sional) movements. Displacements have therefore been separated into two
principal groups, by one of which have been produced more or less hori-
zontal, by the other more or less vertical changes of position of portions of
the crust.

Under compressive stresses rocks may suffer deformation in
one or both of two ways: by flow, when under sufficient load ;
and by rupture. when the load is insufficient to cause flow.

DEFORMATION ZONES OF THE LITHOSPHERE.

It has been shown that at a depth of 10,000 meters, more or
less, even the strongest rocks must find relief from stress by
flow, and hence below that depth there must be a zone which,
as respects its manner of deformation, may be called the zone of
flow: At very moderate depths relief from Stress gwill) be
obtained by rupture, and this uppermost belt is therefore denom-
inated the zone of fracture. Since, however, the depth at which
rocks will flow is dependent primarily upon their crushing
strength, there will be an intermediate zone or belt within which
deformation will take place partly by rupture and partly by flow
(zone of combined fracture and flow). There are many othe
conditions—such as amount of contained water, elevation of
isogeotherms, etc.— which modify the depths of the zone of flow,
but they need not be taken up here.

It is clear from these considerations that rocks. may be
deformed at one period by fracture (when under light loads),
and at another time by flow (when below a depth of 10,000
meters, more or less). During a general elevation of a region
deformation by flow within a given rock mass, will in general
precede that by fracture, and the structures which arise from
fracture will be superimposed upon those developed by flow.
During a considerable subsidence of a province structures due
to flow will be superimposed upon fracture structures and either
wholly or partially efface them. The presence in the same

<VAN Hiss, “ Principles of North American Pre-Cambrian Geology,” Szxteenth
Ann. Rept U.S. Geol. Survey, Part I (1896), pp. 589-94.

MAPPING OF THE CRYSTALLINE SCAISTS 801

rocks of well-developed flow structures and well-marked fracture
structures independent as regards their direction of the flow
structures, is strong evidence that the deformation by fracture
was subsequent to that by flow. Moreover, in the very nature
of the problem it is to be assumed that a fracture structure
will in most cases be superimposed upon a flow structure, owing
to the fact that degradational forces, as they continuously
remove the upper layers of the lithosphere, gradually relieve the
load upon the deeply buried rocks, until these are at last
exposed to our view. This, however, is not the only reason
why we should assume a fracture structure to be the later when
found in conjunction with a flow structure, for fracture struct-
ures are quickly effaced by flow, whereas flow structures, though
disturbed and distorted, are not obliterated by fracture
structures.

THE MECHANICS OF DEFORMATION BY FLOW.

The mechanics of this problem is a large subject fully treated
elsewhere* and may well be largely omitted from this discussion.
Deformation by flow produces flexuring or folding of the initially
horizontally bedded rocks. The position and the shape of the
individual flexures will be determined by many factors: as, the
position of old shore lines with their planes of weakness, and the
initial dip of the beds resting upon them; the position of masses
of igneous material (which by refusing to yield as readily as the
bedded rock may elevate a portion of the latter above the zone
of flow and cause rupture by thrust at the planes of maximum
mashing); the position of earlier planes of rupture or foliation,
etc. Flexuring is thus the method by which the heavily loaded
rocks accommodate themselves to lateral compression, and results
in a crustal shortening and thickening. In every fold there is a
local compression normal to the beds in the limbs of the fold,
and a local tension in the same relative direction in the arches,
the tendency of which is to cause a local transfer of rock mate-

* WILLIS, ‘‘ Mechanics of Appalachian Structure,” Zhzrteenth Annual Report GES:
Geol. Survey, Part II, 1891-2, pp. 211-282. WAN HIsE, Joc. cit., pp. 589-668; see

also Surss, Antlitze der Erde, Vol. I, pp. 142-189; MARGERIE ET HEIM, loc. cit,
49-87.

862 SHLODIES HOR SLOIDEN TS

rial from the limbs to the arches, the limbs becoming thinner and
the arches thicker by this process. Foliation planes, or planes
of easy separation, are also produced, which in moderately com-
pressed folds make small angles with the axial plane of the
folds,’ but in closely compressed folds they are approximately
parallel to that plane.

MECHANICS OF DEFORMATION BY RUPTURE.

The problem of deformation by lateral compression of a
crustal block has been treated from a mathematical standpoint
by Becker,? his discussion assuming the section of crust to be an
elastic solid.

This block may be regarded, then, as subjected to stresses
from which it acquires the properties of an anisotropic medium
in which the greatest and the least axes of the strain ellipsoid
lie in the horizontal plane, owing to the fact that gravity opposes
an elongation of the particles in the vertical direction. The
effect of such stress is to produce a system of vertical planes of
dislocation which in a mass of infinite resistance (perfect elasti-
city) would make 45° with the direction of pressure, but other-
wise (if more ‘or less plastic) the angle would be less. With
pressures rapidly applied it is believed that rocks behave like
highly elastic bodies. The best evidences that such is the case
seem to be furnished by the almost universal observation that the
most prominent joint planes are nearly vertical (except in subse-
quently tilted rocks) and so generally nearly perpendicular to one
another ;3 and by the experiments of Crosby,4 who found that a
system strained to a point far below that of rupture (near which
plasticity becomes for the first time an important factor) is
relieved from its stress by dislocation through shock communi-
cated to the system (for example, by an earthquake). In the

tSee this JOURNAL, Vol. X, p. 787.

2Gro. F. Becker, “ Finite Homogeneous Strain, Flow, and Rupture of Rocks,”
Bull, Geol. Soc. Am., Vol. 1V (1893), p. 50; see also VAN HIsk, /oc. cit., p. 672.

3 A. DAUBREE, Etudes synthetiques de géologie expérimentale, Part I (Paris, 1879),
pp- 300, 301; see also BECKER, Joc. c7/., p. 73.

4W.O. Crossy, “The Origin of Parallel and Intersecting Joints,” Am. Geol.,
Vol. XII (1893), pp. 368-75.

MAPPINGOF, THE CRYSTALLINE SGHISTS 863

opinion of the writer, these considerations render invalid many
of the data collected from test pieces in laboratories. Elsewhere
the writer has developed this subject more extensively.* It has
there been pointed out that the production of a system of verti-
cal and perpendicular joint planes may be, and probably would
be, but the first step in the process of relieving by rupture the
stresses within a crustal block. Bvy the formation of prismatic
joints the potential energy of the system is lowered and a read-
justment of the stresses brought about. Furthermore, a tend-
ency is set up favoring local depression, for the reason that rocks
in the zone of fracture support their load by their rigidity, and
this element of rigidity has been much reduced through the sys-
tem of dislocations. The crustal block rests upon the rocks in
the zone of flow, which are in a condition of potential fluidity,
or potential viscosity. The differences of composition and den-
sity within the zone of flow, combined with the varying perfec-
tion of the joint planes within the zone of fracture (which may
be due to lack of homogeneity and perhaps to other causes),
furnish conditions for a most irregular and unequal depression
of the component blocks produced by the dislocation. On the
basis of experiments and observation alike it is believed that the
type of tectonic structure produced, before planed down by
erosive agencies, may be well illustrated by that of a mosaic,
the backing of which has been removed and the component
blocks so disturbed as to stand at different altitudes, but with
similar axes parallel.

A resumption, or, what is more likely, a continuation, of
strong compression of an area in the same horizontal plane, or a
second shock communicated to it, would doubtless produce a
result of a somewhat different character from the first. The
resultant of the lateral compressive stress is by reason of the
ready-formed joint-planes resolved in directions parallel and
perpendicular to them, the former producing a shear along the
planes themselves, and the latter a shear along the diagonals of
the blocks of the primary system. Torsional stresses are also
set up so as to produce fracturing at the edges of the prismatic

* Twenty-first Annual Report U. S. Geol. Survey, Part II], pp. 124-33.

864 STUDIES FOR STUDENTS

blocks. If the direction of the new compression made a large
angle with the direction of the first, the directions of the new
planes of dislocation would be difficult to determine, and torsion
would be at amaximum. If, however, its direction corresponded
closely to the first, as is most likely if it occurred in the same
general period of disturbances affecting the area, the later planes
of dislocation would in their directions bear a simple relation to
the earlier. New depression of the area would quite likely fol-
low the new dislocation.

Gilbert has shown that within an area which is undergoing
depression along fault planes it is likely that earthquakes will
follow one another in succession, owing to the fact that oro-
graphic blocks are supported in part by the friction along their
vertical walls, and that starting friction is greater than sliding
friction.’ The momentum acquired by the slipping block is
effective in reducing temporarily the energy of the system, and
time is required for its recovery.

To summarize the mechanics of crustal deformation by rup-
ture, it may be assumed that its result would probably be to
produce vertical polygonal prismatic blocks bounded by joint
and fault planes in a number (large or small) of parallel series
intersecting one another, and that these blocks would stand at
relatively different altitudes.

SYSTEMS OF FAULTS.

Continental European geologists have quite generally recog-
nized the importance of systems of normal faults in bringing
about important deformations, and have described such systems in
many widely separated areas. This may, in fact, be said to consti- _
tute the most marked line of cleavage between the structural
geological investigations of Kurope and of America. The trend
of European investigation in this regard was early given by
Leopold v. Buch in Germany and Elie de Beaumont in France.
The latter correctly located along the Rhine valley’ parallel

‘GILBERT, “A Theory of the Earthquakes of the Great Basin, with a Practical
Illustration,” Am. Journ. Sct. (3), Vol. XX VII, 1884, pp. 49-53.
2 DUFRENOY ET ELIE DE BEAUMONT, Explication dela carte géologique dela France,
Voli lp: 398:

MAPPING OF THE CRYSTALLINE SCHISTS $65

lines of faults, and, following the theory of the time, supposed
the mountain masses upon either side of the Rhine valley to
have been forced up above their former level so as to produce a
deep channel. That there are many parallel faults which follow
the direction of the valley walls of the Rhine has been
abundantly proven by later investigations.‘ Today the presence
of numerous dislocations in parallel and often intersecting series?
has been recognized in many European areas extending from
the steppes of southern Russia upon the east, to the Pyr-
enees upon the west, to describe which there have been
brought into use a considerable number of special terms. Some
of these terms have no equivalent English expresssion, and such
as have are, in some cases, unnecessarily long. Among the
foreign terms are: Ger. bruchfeld, Fr. champs des -fractures
(area of dislocations); Ger. Bruchnetz, Fr. réseaux régulier des
failles (fault net-work); Ger. Senkungsfeld (sunken area); Ger.
Fforst, Fr. horst or mole (elevated area left between two or more
sunken areas); Ger. Scholle (orographic block); Ger. Graben
(long and narrow sunken block); Ger. Sriicke (long and
narrow elevated block); Ger. Zhurm (relatively high block of
nearly equal basal dimensions); Ger. Kessel (relatively low block
of nearly equal basal dimensions) ; etc.

It has been especially the work of Suess to correlate the
many scattered observations and in the first volume of his
monumental work upon Face of the Earth? to show how the
lineaments of the continent are lines of normal faulting, between
which great crustal blocks have been depressed by different
amounts. In America the description of what I have called the

IDAUBREE, Description géologique et minéralogique du departement du bas-rhin
(Strasbourg, 1852), pp. 384-406; BENECKE and COHEN, Geognostische Beschreibung
der Umgegend von Heidelberg (Strasbourg, 1881), pp. 595 ff.; AUSFELD Geologische
Shizze der Gegend von Rheinfelden, Mitth. d. Aadrg. naturf. Gesell., Vol. III (1882),
pp. 83-102 ; etc.

2In German usage the term System is generally employed to cover any grouping
whether of parallel or of intersecting faults. The author has described under the
term “series of faults”? a group of parallel faults, restricting the use of the larger
term “‘ system of faults ” to the network composed of parallel series. Twenty-first Ann.
Report U. S. Geol. Survey, Part III, pp. 98 ff.

3 Antlitz der Erde, Prag and Leipzig, 1885.

866 SLODIBS FH CRSS LO DENTS:

mosaic type of infaulted structure seems to have been restricted
to papers upon the Great Basin of the West and upon areas of
Newark rocks along the Atlantic border, in both of which
provinces only the simplest types of folding are observable. It
is highly probable that joints and ‘‘minor faults” have been
elsewhere frequently observed but thought to have throws so
small as to be negligible, the cumulative effects of the small
displacements along numerous near-lying planes being over-
looked. Faults have too frequently been regarded as isolated
phenomena. Within the crystalline areas isolated normal
faults have sometimes been supposed to stand in some relation
to folds, and because evidence of individual faults is within such
areas most difficult to secure, faults have been the last resort of
mappers in securing harmony between areal relations and basal
hypotheses. The tendency to use faults ‘to get out of tight
places”’ has probably been in direct ratio to the inadequacy
of the hypotheses assumed, and it is, therefore, not strange
that faults have been and still are in much disfavor among the
most competent and conscientious of American geologists.
There are, however, as the writer believes, methods of showing
the connection of a system of faults with the deformation of an
area, through the peculiarities of its joint system, topography,
hydrography, outcrop distribution, etc.

It will be the object of this paper to point out especially
the methods of recognizing fault systems when in conjunction
with systems of folds, because in the manner of deformation of
an area is involved the most fundamental assumptions of
geological mapping. The view that important deformation
may be brought about within a folded area of the crystalline
schists by a system of normal faults, has hardly been discussed
by American geologists, the system of folds being invariably
present and supposed to be the only important manner of
deformation exhibited.

The impression should not be gained that in the mind of the
writer the relative importance of the two methods of deforma-
tion is correctly set forth by the space here devoted to each.
Deformation by folding is comparatively well understood and

MAPPINGVNOPMTTE CRYSTALLINE (SCHISTS 867

exploited in the literature,’ whereas, as pointed out, the consid-
eration of deformations by a system of fractures is, at least for
American areas of the crystalline schists, a new departure in
geological mapping.

DIRECT GENETIC RELATIONS OF JOINT SYSTEMS TO SYSTEMS
OF NORMAL FAULTS.

As early as 1879 Daubrée brought out clearly the relation of
normal faults to planes of jointing. As in the case of so many
other important discoveries, however, little weight seems to
have been attached to the results by contemporary geologists,
and it is doubtful if one geologist in ten now records with care
the direction of joint planes. His conclusions the eminent
French geologist states as follows :

I. The constancy over large areas of the orientation of certain systems
of joints has been already confirmed by Sedgwick, de la Beche, John
Phillips, and later by other geologists. Further, it has been recognized in
Cornwall that the joints maintain their direction in passing from the granite
into the schist or £2/7as to traverse which they penetrate.

noone This permanence of orientation of the joints ..... makes them
approach faults, of whose mechanical origin there is no doubt.

It is further known that in many countries joints may be seen passing by
various intermediaries into faults properly so called.

2. The direction of joints in Yorkshire has been the subject of a great
number of observations of another eminent geologist, John Phillips,? who has
collected them into a rose of directions: it results that two directions
predominate greatly over the others and that the two directions are
perpendicular to each other.

3. [Referring to the deformation and distortion of fossils in the vicinity
of joints.|.....It is therefore an important result to have arrived at a
consideration of joints as the effect of rupture the same as faults, which
differ above all by their dimensions.

4. [Referring to the pebbles of conglomerate which have been observed
to be neatly cut through by joint planes.]..... A powerful cutting or
shearing action has operated in the formation of joints.

To summarize, the characteristic feature which manifests itself in the
innumerable fissures of the earth’s crust is a parallelism which reproduces
itself in the large and in the small fractures—in the fault as in the joint.3

tSee WILLIS, Joc czt.,; MARGERIE ET HEIM, of. czt.; VAN HISE, Joc. cit.
2PHILLIPS, //ustrations of Yorkshire, Vol. Il, 1836.
3 DAUBREE, Géologie Experimentale (Paris, 1879), pp. 304-6.

868 SLOUDIES FOR STODENTS

A joint becomes a normal fault so soon as displacement
occurs along its walls, and the essential difference, therefore,
between normal faults and joints may be said to lie in this con-
dition. An area of the earth’s crust which has been deformed
by jointing under compressive stress is thereby made especially
liable to vertical changes of level, because its rigidity, and there-
fore its competency to’ uphold its load, has been reduced.
Depression of the area, as a whole (and perhaps elevation of
certain portions of it), will take place along the joint planes
already formed. There is thus, probably, in all cases at least
an infinitesimal displacement along the joint walls, and thus no
sharp line can be drawn separating joints from faults.

Brégger, in his work upon the Langesund-Skien region of
southern Norway, says of the joint and fault system:?

“The dislocations, that is to say, the faults, have in fact cut the landscape
through and through, and not alone parallel to one system of lines, but first
chiefly parallel to two principal systems, and then also parallel to other less
prevalent directions. .

If we consider how thickly the rocks are penetrated by very small faults,
it follows, in fact, that a portion of the crust cut up in this fashion is built up
like separate rectangular blocks of masonry.

The study in much detail of the circumscribed area of the
Pomperaug valley in Connecticut and its vicinity? brought out
facts quite independent of, but altogether harmonious with those
discovered by Brégger in Norway.

In the Connecticut region it was shown that a system of
nearly vertical faults falls into a number of intersecting series of
parallel dislocations, of which four are quite prevalent, and that
the directions of the series bear relations to one another, for
which an explanation may be found in the mechanics of jointing
under compression. A relation was also disclosed between the
directions of the faults and the sizes of the included orographic
blocks. It was further observed that the system of faults is
parallel to and grades into the system of joints, and like it is

™W. C. BrROGGER, “Spaltenverwerfungen in der Gegend Langesund-Skien,” /Vy¢
Magasin for Naturvidenskaberne, Vol. XXVIII (1884), p. 401.

2 Hoss, “The Newark System of the Pomperaug Valley, Connecticut.” Twenty-
first Annual Report U.S. Geol. Survey, Part III (1901), pp. 1-160.

WAPPINGIOR TILE CRYSTALLINE SCHIST S 869

spaced with an approach to uniformity. As regards the larger
displacements of the area, it was found that they seldom occur
wholly along one plane, but are distributed over a zone of
parallel and near-lying planes — the throw is distributed.

The faults of the less common series run along the diagonals
of the larger composite blocks, and are believed to owe their
formation to the downthrow of these blocks, and not to the
original relief of the area from compression.

CRITERIA FOR RECOGNIZING A SYSTEM OF FOLDS.

While not necessarily present in all areas of disturbed rocks,
it is fair to assume that in all areas of the crystalline schists in
which bedded rocks are found, folds will be recognizable. The
possibility of unraveling their relations to one another within a
system will depend upon the intricacy of the structure, the fre-
quency of the exposures, the freshness of the rock, the perfection
of a combined joint and fault system, etc.

In general it may be said that flow structures and those pro-
duced by fracture are in contrast, the former exhibiting warped
(curving) surfaces, and the latter plane surfaces. The courses
of the fold structures across the map are, therefore, curved lines,
whereas the latter course over the map in straight lines, or in
zigzags made up of rectilinear elements. The courses of fault
structures upon the map are little disturbed by the relief of the
topography in proportion to their inclination —hade, being abso-
lutely unaffected by topography when the hade is go°. The
nature of known formation boundaries upon the map, and that of
other structures is thus one of the most important considerations
in the determination of the kind of deformation to which the
area has been subjected. When alternations of pitch of folds are
not present, folded rocks present also rectilinear boundaries, but
this consideration loses importance in most areas of intricate
geological structure, because evidences of pitch can generally be
made out. Change of pitch has been described as characteristic
of few areas, because, as already explained, it has until compara-
tively recently seldom been looked for.

Other indications of the presence of flexures within a given

870 STUDIES FOR STUDENTS

area are observed at the individual exposures. In areas of
folded rocks minor curvings of bedding planes are usually exhib-
ited, and generally in great perfection. Changes in the strike
and the dip of the beds are, when large or numerous exposures
are found, observed to be, in general, gradual rather than abrupt;
and, if abrupt, search will reveal intermediate values for the dip,
or the strike, or both, as the case may be. Foliation planes are
characteristic of, and connected with, folding, but not with
faulting.

CRITERIA FOR RECOGNIZING A SYSTEM OF FAULTS.

The characteristics of a jointed and normally faulted area of
the earth’s crust have been touched upon in the last section.
With abundant outcrops this structure may be disclosed upon
the areal map by the generally rectilinear or zigzag boundaries
of formations. If not, some indication may be afforded by the
prevalence of straight lines and sharp bends in the topography,
particularly if the lines fall into parallel and intersecting series.
Such a relief may not be apparent without careful examination
of the map, for the reason that the particular combination of
directions is not known, and it may yet be quite striking when
once discovered. Again, the drainage lines may compose a net-
work which in direction conforms both to that of the topographic
lines, and that of the formation boundaries. The prominent
joints observed at the individual exposures, it may be supposed,
will also be in conformity with the directions of the geologic and
topographic lines above enumerated. This has never been better
expressed than in the classic work of Professor Th. Kjerulf upon
the geology of southern and central Norway, which concludes
with these words:

Thus the great systems of fissures which cut up the surface produce the
primary lines in the aspect of the surfaceof Norway. The mysterious network
of these lines is stamped in indelible characters; this may indeed remain a
long time unnoticed; if, however, it has once been seen, it will never again
escape observation. Like a moss-grown inscription upon a plate of marble, it
is there and may be recognized. Here all embodied representations of

plateaus, of tilted plains, and of every sort of erosion have not prevailed to
hide the writing or to withdraw it from observation. Push these all aside and

MAPPING OF DHE: CRVSTALLINE SCAISTS 871

the eye again sees the runic characters, and it depends only upon this, that
they all be correctly understood in the future.*

The observation of these striking lineaments of the surface,
however, unless they be of great perfection, hardly furnishes an
adequate demonstration that an important deformation by fault-
ing has occurred. The proof of this must rest upon the dis-
covery of a considerable number of individual faults whose
directions relegate them to a system of parallel and intersecting
series. The methods by which individual faults may be recog-
nized will be outlined in the sequel.

KEY AREAS FOR DETERMINING THE MANNER OF DEFORMATION OF
A REGION.

In geological mapping it is almost an unwritten law that
geological sequences must be established at points where forma-
tions appear in their larger masses at the surface, it being
assumed that structural relations are in such cases the simpler.
Areas which show a considerable number of formations brought
closely together in small masses at the surface are looked upon
with suspicion as areas of local and “‘minor’’ faulting, or as con-
taining intercalated beds of unusual types due to purely local
conditions of sedimentation. Whichever of these views is
assumed, the areas are likely to receive but small attention;
first, because the problem of their structure is difficult to unravel
and not regarded as affecting the larger questions of the region ;
and, second, because if once unraveled, the scale of the map
would not allow of its representation. Such areas are, therefore,
most frequently represented upon the map in the color of the
formation which is believed to compose their greater part.

However unsuited these intricate areas may be supposed to
be for establishing the order of succession of geological forma-
tions, they are nevertheless, it is believed, in many cases the
keys (and perhaps the only ones) to unlock the secret of the
manner of deformation which has affected the area as a whole.
Complicated indeed, they often require only patience and
industry for their unraveling, whereas the larger masses by their

1 OD. cit., Pp» 334.

872 STUDIES FOR STUDENTS

very simplicity of areal distribution allow of several hypotheses,
any one of which would adequately explain them.

NATURE OF EVIDENCE FOR ESTABLISHING A FAULT.

The several ways in which individual faults may be discovered
the writer has elsewhere recited... Some of them are alone
sufficient as proof, while others must be regarded as offering
incomplete evidence to be weighed in connection with additional
observations. As modified for the problem before us these
methods will now be considered. Many of them could apply
equally to thrusting or to normal faulting, but observation of
fault hade will generally be sufficient to determine which in a
given case is present.

1. The observation of beds formed at different times in juxtapost-
tion along a plane transverse to their bedding —With conformable
formations such an observation may be regarded as absolute
proof of a fault, and is, in fact, one concerning which there is no
disagreement among geologists. It is only rarely, however, that
such a contact will be found exposed, and hence the method
serves but seldom.

2. Offsetting of formations in outcrop.—Al\most equally reliable
as a method for determining a fault is the offsetting of forma-
tions in outcrop. If in following the mutual boundaries of two
formations, the formations are found abruptly thrown to the right
or left and continued either along the old or in a new direction,
it is only necessary to show that the core of a pitching fold has
not conditioned the break of continuity to prove the presence
of accross fault. When several such faults occur in series with
throw in the same sense, either the apparent course of the
boundary as a whole is given a different rectilinear direction, or
a crescentic outline is conditioned. The crescentic ridges which
are so prominent a feature of the Connecticut valley are pro-
duced in this way. The three ways in which such ridges may be
produced by normal faulting have been elsewhere pointed out.’

3. Offsetting of outcrops as definite topographic features —With
rocks which are resistant to the erosive agencies, or where fault-

* Twenty-first Annual Report, U. S. Geol. Survey, Part III, pp. 85-93.

2? HOBBS, oc. czt., pp. 95-8.

MAPPING OF THE CRYSTALLINE SCHISTS 873

ing has been accomplished in a not too early geological time,
orographic blocks may cause a distinct offsetting of elevations
upon the general surface and display an arrangement in zigzags
or en échelon, which it is difficult to explain upon any other basis
than that of normal faulting. Fig. 1 shows by the outline of
the wooded area the serrated border of schist outcrops due to
this cause. In Fig. 2 are exhibited four distinct orographic
blocks of limestone arranged en échelon and owing their position
above the general level to a skeleton work of silica which has
fitted them to withstand the gnawing effect of erosion.

4. Dikes.—Ilt has long been recognized that dikes of igneous
or aqueo-igneous rock, and veins whose walls are plane sur-
faces, represent the positions of. either joint or fault fissures
which have been subsequently filled. Especially in the Harz
mountains of northern Germany has this method of finding the
position of old planes of dislocation been successfully employed.

5. Abrupt changes of strike and dip not indicated in folds —An
essential characteristic of every folded area is that gradational
values can be found to correspond to all important changes of
strike or dip, owing to the fact that folds appear in curves in
nearly all of their sections (with undulating crests and trough
lines in all sections, otherwise in all sections except those paral-
lel to crest and trough lines). With excessively sharp pitching
folds, changes of strike and dip by small amounts may be as
abrupt as though caused by faults, but search will generally
reveal at the apex, in the one case the core of the fold, and in
the other the evidences of rupture, although it is quite possible
that both will be found together.

6. Discovery of fault breccia — Reibungs or fault breccia is
very generally found where fault planes of considerable throw
are actually uncovered. It is also often found in other places
and especially in pits and shafts where access to the wall rock
is not possible. Even in these cases, however, it furnishes
valuable evidence that faulting has occurred. The formations
which are in contact along the fault plane are likely to be indi-
cated by the angular fragments of the breccia, which are usually
cemented together by calcareous or siliceous material.

STUDIES FOR STUDENTS

View looking north-northwest along

Fic. 1.—Normal faults indicated by offsets of areas of outcrop in the topographic features.

the western flank of the Laurel Lake Ridge, Lee, Mass.

875

VEEN Gag Oe TET CLS ATL ETN EAS GEES AS:

j[ey eB pue suo jutod ev ye 4jsva Bulyooy, MarIA

‘sseyl ‘syTeq AoTysy Jo ysamyjsou sayrur
‘0/7427 UI PEBULIIY DIUOJSIWT] PIYIII[Is Jo syo[q o1ydviB010 [[ewS—z ‘oI

pe

876 SLUDIES FOR STUDENTS

7. Observation of slickensides—While offering contributary
evidence in many instances, observation of slickensides is one of
the less important methods of locating a fault, and it could
hardly be used without other evidence.

8. Disappearance of outcrops along a rectilinear boundary.— As
already pointed out, little notice seems to have been taken of
the way in which outcrops and areas of near-lying outcrops ter-
minate, though in the opinion of the writer such boundaries are
full of meaning. When outcrops or groups of outcrops ter-
minate in right lines which are at the same time parallel to the
strike of the beds and to the topographic contour lines, no
special explanation is required. If they terminate in right lines
across the strike of the beds, but along the topographic con-
tour lines, their explanation may be found, either, first, in the
erosion history of the region with no dependence upon geologic
structure planes; second, in the valley lines as fixed by pre-
existing fault planes; or, third, in a fault scarp still in evidence
at the locality. If the outcrop boundary is a right line and
neither parallel to strike of beds nor to topographic contour
lines, it is pretty sure to represent the position of a fault plane."
In glaciated areas regard must, however, always be had to essen-
tially local and irregular accumulations of drift as affecting this
explanation.

g. Occurrence of scarps in the more resistant rock.— In certain
areas, at least, of the crystalline schists in which there is accent-
uated topographic relief, scarps are among the commonest of
the physiographic features. In the mapping of the hard-rock
geology account has, however, rarely been taken of them, it
being perhaps rather generally assumed that their interpretation
is the special field of the physiographer and glacialist. If,
however, it can be shown that their directions conform to planes
of jointing and faulting, their study will furnish some of the most
fascinating and instructive lessons for the geologist in the crystal-
line areas. It is doubtless because their secrets lie hidden in
the structure of the underlying rocks that rock scarps as topo-

*See Hoss, Joc. cit., pp. 89-91 for special cases of faults fixed by rectilinear
boundaries of outcrop.

MAPPING OF THE CRYSTALLINE SCAHISTS 877

graphic features have received so little consideration at the hands
of the modern school of physiographers.

The perfection of rock scarps will doubtless be dependent
primarily upon the resistance of the rock of which they are
formed to the agencies of erosion, and to the time that has
elapsed since their formation. In dense basalt their life is
probably the longest, and the present surfaces will most nearly
represent the original fracture planes. It is a fact of general
observation that scarps are usually interrupted in such a way as
to produce a step-like structure, which suggests in nearly all
cases that the throw was distributed over several parallel and
near-lying planes. The great cliffs of the Newark basalt of the
Connecticut valley produced in an unknown period of post-
Newark time, furnish striking illustrations of rock scarps in
great numbers and perfection.

The evidence that precipitous rock walls are connected genet-
ically with the fault and joint system of a region—an inher-
ently probable supposition—-must be drawn from a study of
the direction, not of a single cliff, but of a large number of
scarps embracing a considerable area. Accurate topographic
maps are most instructive in this regard, but they fail to tell the
full story. Photography here comes to the aid of the geologist,
not only in recording topographic peculiarities, but in the
exposition of them. To give to photographs their full sphere of
usefulness, however, it is important to indicate upon the map
from what points they were taken and the pointing of the
camera at each locality. A photograph common and uninter-
esting in itself becomes full of meaning in its relation to the
map. A character like Fig. 3 entered upon the map
and consisting of a circle which incloses the point at
which the view was taken, and proceeding from the _ Fis. 3.
circle a short arrow to show the camera pointing, will supply
the needful information. A figure within the circle refers to the
number of the photograph or to the plate of a report, as the
case may be.

Scarps may, however, be produced in other ways than by
faulting. Other agencies most potent to produce them are:

878 SAO DIES HORTSRODENTS:

first, in glaciated regions the ice mantle which has carried away
blocks from the lee side of elevations lying in its path; and,
second, in regions of severe climate, the frost, which has separ-
ated blocks and allowed them to roll down steep slopes. A
special cause of scarp formation may be the undermining by
solution of beds of calcareous rock, which has allowed the over-
lying beds to fracture and fall. In all cases, however, some
earlier structure planes must be assumed along which the block
has separated, and these planes may have been either planes of
jointing, or faulting, or foliation, or of original bedding. Con-
sidering each scarp by itself, therefore, and quite independent of
the system of the region, it is more likely to represent a fault in
proportion: first, as it does not face in the direction in which
the ice mantle moved over the area; and, second, as it is not
parallel to a plane of fissility or of bedding (not parallel to the
strike).

Study of the southern New England area has shown that
cliffs of considerable extent which adhere to a uniform trend
often for long distances usually do so, not along a single plane,
but by a series of minor zigzags the elements of which are joints
or faults of other series combined with the one which corre-
sponds in direction with the general trend of the cliff." Fig. 4

. will illustrate the manner in which it is

accomplished. Suchacomposite cliff
is shown in Fig. 5. Figs. 1 and 10
must be regarded as showing the same type of structure, but in

Fic. 4.

the next higher orders of magnitude, respectively, within a com-
posite series. This observation of the composite nature of lines
of dislocation is regarded as of much significance in showing the
genetic connection of faults and joints, as well as in reading the
geological structure of an area.

10. Fault gorges—In many places, and especially throughout -
New England, are found deep clefts in the harder rock which

ICG

are locally known as ‘‘purgatories,” ‘‘ devil’s dens, ice-glens,”

‘‘ice-gorges,” etc. The walls of these clefts are nearly or quite

‘Vide BRANNER, “Geology in its Relation to Topography,” Zrans. Am. Soc. Civ.

Engineers, Vol. XX1X (1898), p. 63.

879

MAPPINGIOL LEE CRVS TALLINE, SCHISTS

099 “MA 001 FN “M .O4 FN “A 28t
SUMS s}[NVj Iayfeus yy fayrur we Apiwou AOy"Y . BSF CN Spud] [[eM sy], “ssVyT ‘weysulsd J, ‘9Tqqod 9y} JO [[eVM ysvayyNog

FN
‘apny

-IUSvUI JO LapIO IIMOT v JO sy[Nez AuvW Jo pasoduioos puv Sulypney Aq paonpoid uoNoeAp Ul IvaUTTIJIaI AT[VIJUaSsa FI Q—'S ‘Ol

880 SLODIES FOR ISLODENTS

perpendicular, and their bottoms are occupied, not by a stream,
but by an accumulation of great angular blocks composed of
the same material as the walls. Deep below this accumulation
of blocks one may sometimes thread his way until the ice and
snow of the preceding winter are found unmelted even at the
end of summer. Such
gorges find their sim-
plest explanation in the
local depression of a
long and narrow oro-
graphic block along
vertical joint walls
(Graben), with the ac-

FIG. Geterseceecian of fault-gorge (so called companiment of much
“‘ice-gorge’’) near Hartsville, Mass. The stippled fracturing and disloca-
area represents snow and ice. tion in the block itself.
(See Bigs: ovand: 7);

11. Arrangement of surface springs in right lines.—Since the
time of Daubeny and Forbes the locations of springs have been

connected with fissures. Peale in his report upon the mineral
waters of the United States cites many regions where springs are
arranged along fault planes, and, speaking of the Great Basin,
says: ‘‘A map of the hot springs of the Great Basin would be,
to a great extent, a map of its displacements." Regions might
be multiplied where essentially the same facts have been observed.

NATURE OF EVIDENCE FOR ESTABLISHING A SYSTEM OF FAULTS.

Observed faults are in parallel and intersecting seres—Having
made out within a region the presence of a number (larger or
smaller) of normal faults, it is next to be determined whether
they stand in any relation to one another If a considerable
number have been observed, it is likely that some will be found
to be parallel, or as nearly parallel as the errors of observation
allow. Evidence of this nature is, however, valuable in propor-

'PEALE, Fourteenth Ann. Report U. S. Geol. Survey, 1894, Part I], pp. 63, 64;
see also RUSSELL, Fourth Ann. Report U. S. Geol. Survey, 1884, p. 452; HILL AND
VAUGHAN, Lighteenth Report, tbid., 1898, Part II, Plate XLVI; Hosss, 7venty-firs¢
Report, rbid., Part ILI, pp. 91-2.

MAPPING OPAL EE GRVSTALEINE, SCHISTLS

The view is from within the gorge and looks

The “Ice Gorge” near Hartsville, Mass.

Fic. 7. —A gorge produced by faults.

toward the entrance.

882 STUDIES FOR STUDENTS

tion as the number of observed faults is large. It has been
observed in the Pomperaug valley and fully confirmed by expe-
rience gained subsequent to the publication of the report upon
that area, that a regular spacing of normal faults is as character-
istic of them as it is of joints. It is, therefore, often possible to
determine not only the shape but the size of orographic blocks
included between series of normal faults within a system or
network.

2. Observed faults are parallel to the joint system.—In this
observation lies the most important evidence that the region as
a whole is affected by a system of normal faults genetically
connected with its system of joints. Not only is it to be
observed whether the two systems are parallel, but, if so, it is to
be noted whether any simple relation connects the direction of
the individual joint and fault series with the dimensions of the
orographic blocks which they have conditioned.

3. Zigzag topography.—The study of the topographic map
supplemented by observation and by photographs made in the
field may show that the physiographic features are bounded by
straight elements which coincide in direction with the network
determined by the joints and faults. Figs. 9 and 10 represent
topography of this type, the one from the intricately faulted
Newark area near Meriden, Conn., the other from the crystalline
belt of Berkshire county, Mass. A good illustration also of
this type of topography is Monument Mountain in Berkshire
county, whose main mass is composed essentially of a single
fairly uniform formation and is believed to owe its exceptionally
rugged topographic development almost solely to its joint and
fault planes. (See Fig. 8.)

In the paper which has here been so frequently cited for
illustrations of fault structures the writer has used the expression
“floating block topography” in reference to a very striking
physiographic development, by which are revealed orographic
blocks, generally of similar size and shape standing at different
altitudes and with bounding fault-planes in evidence as scarps.
Fig. 11 displays such a type of physiographic relief. It is to
be noted that the blocks of this type (for convenience ‘“‘unit”

MAPPING OF THE CRYSTALLINE SCHISTS 883

etd

Fic. 8.—Details of topography of Monument Mountain, near Stockbridge, Mass.

blocks, though subdivided by faults and joints) are sometimes
grouped into hills—composite blocks—which are spaced over
the general surface much like the black squares of a checker
board. Fig. 11 exhibits about two-thirds of such a hill in a
series of the kind described. Fig. 2 is a detail taken from
another hill of the series.

\

s
IS gets
ISSN Es. oes
PS GENS CSV = 2
Wei —=

—-—

& WSs Orc
AS 5 a 2 SS) SSS +
. . ——
ae " & so 5
b VE ~ ° \ SY \ 4 i
N ft y {' “ WS =
\ - { SS a Oo
N <5) Z BNA ( : . nq af A
. = ! Xen S sa 8
a = u g >
\y \\\q Ca) Rs =< toY) 4
~ =~ { Ss —— = oo 3
- y 52. SS . -
2 » PS\ NS WYO SS VSS SD) Bg 2
SA (@ a Y Vy AN NS | iS © Ss
AY ( Wi "i )) ~ \ \ —
y)) = WY py | : AA ney
\\ (((4 X = ZHiff Z S25: SS a g 5p
(Com y= Via —— Se
Ne SS Z) ae SSS SS BEA 1(C = = z a = 2
pe a == == \ Y ed
SS = C ) \ 4 :
) = Xd CANS

STUDIES FOR STUDENTS

MAPPING OF THE CRYSTALLINE SCHISTS 885

=

Gp
\f
Ie 7
= \ (
y ~~ Zam \ =
re =
Z ,

o NS YZ =
VAS SZ Se ee KK
/, ; nS SS nan
: sf = = —~\ Y
Cee

N

Zo

Ly) ) \ Q\} SS Ge \
Fic. 10.—Zigzag topography in the crystalline schists. Scale one inch equals one

mile. From the Pittsfield sheet of the map of Massachusetts by the U. S. Geological
Survey.

886 SLODIES FORK STRODEN IES

4. The drainage system a network of parallel and interesting
series. —The recognition of the fact that a system of joints and
faults has been responsible for the present directions of streams
within any region, aside from purely theoretical considerations
of the closet naturalist, must be credited to Professor Theodor
Kjerulf, the former director of the Geological Survey of Norway.
His study of the newer and accurate topographic maps of the
country led him to see in the courses of the valleys, lakes and
fjords, the lines of dislocations.t The occurrence in many areas of
similarly regular networks of streams in which the elements are
essentially straight lines in parallel series over considerable
distances has now long been known, and has been given an
adequate explanation by Daubrée as conditioned by the system
of fractures (dthoclases) of the region, in part by the faults
(paraclases) and in part by the joints (dzaclases).? Daubrée
cites the works of many geologists to show that such stream
networks have been observed in many widely separated areas.
Since the appearance of his work (1879), numerous other
examples have been added.3 Lowl, in his work on valleys,‘
treats the subject under three heads—fold valleys, fault valleys,
and erosion valleys. He mentions the Rhine and the Leine in
Germany and the Piave in the southern Alps, as well as other
valleys, which are of the fault type.

Daubrée appears to ascribe the orientation of the larger num-
ber of stream channels to joints rather than faults, his view

*KJERULF, Geologie des siidlichen und mtttleren Norwegens, auth. German ed.
by Ap. GuRLT (Bonn, 1880), pp. 328-34. See, however, in this connection, STUR,

“Das Isonzothal,” etc., Jahrb. d. k. k. geol. Reichsanst., Vol. 1X (1858), pp. 324-66,
and plate.

?DAUBREE, Géologie experimentale (1879), pp. 332-73.

3 SuEsS, Antlitz der Erde, Vol. 1 (1885), p. 339. BROGGER, ‘“ Spaltenverwerfungen
in der Gegend Langesund-Skien,” Wyt Magasin for Naturvidensk., Vol. XXVIII
(1884), pp. 253-419. ‘‘ Ueber die Bildungsgeschichte des Kristianiafjords,” zdzd.,
Vol. XXX (1886), pp. 99-231. Kemp, “Preliminary Report on the Geology of Essex
County,” Report of State Geologist of New York for Year 1893, Albany, 1894, pp.
438-40. VAN HIsE, “Origin of the Dells of the Wisconsin.” 7rans. Wis. Acad. Sct.,
Vol. X (1895), 556-60. Hoss, Joc. cé¢t. chap. v., also “The River System of Con-
necticut,” JouR. GEOL., Vol. 1X (1901, pp. 469-85).

4‘LOwWL, Thalbildung (Prag, 1884), pp. 19-43.

887

MAPPING OF THE CRYSTALLINE SCHISTS

‘sseq ‘streg Aatysy jo jsomyON ‘Aydvssodoy ,, Yoo[q Surywo[y ,,— ‘11 ‘DIy

888 STUDIES. FOR STUDENTS

being that the course of the stream has been conditioned by the
gaping of the joint fissure, and he finds in some instances that
the beds on opposite sides of the stream stand at the same level.
Léwl takes account of the lateral translation of streams working
in inclined strata, and argues that the bed of the stream does
not represent the position of the fault, which in many cases can
be observed high up upon the slope. Speaking of the valley of the
Oder in the Harz mountains, a stream which follows the direc-
tion of the Oder fault, he says :

The Oder valley is therefore nothing more than a performance of the

flowing water. Its course along a line of disturbance proves only that this
directed the erosion into a definite course.

The writer, by taking account of the great number of parallel
faults within a series, would ascribe the courses of the streams
within a network more largely to the presence of many long and
narrow down-thrown orographic blocks (Graben), which have
conducted the water like canals... Upon this assumption the
equal altitude of the strata upon opposite sides of a valley does
not preclude the probability of faults along the bed of the chan-
nel. Whatever explanation be offered of the directing of
streams upon a large scale by joints and faults, there can be no
doubt that streams are of greatest service to the geologist in
showing the arections rather than the positions of lines of dislo-
cation. Regular networks of streams may be assumed to be
much more common than the literature of the subject would
indicate, for the reason that comparatively little attempt has
been made to explain stream directions on the basis of structure
planes of the underlying rocks.

5. Zigzag course of coast lines.— Evidence of the same kind
as that derived from drainage systems is sometimes to be sup-
plied by coast lines, which may disclose the lineaments of a
submerged drainage system, the border of a sunken composite
orographic block, or a margin of areas of outcroppings. The
coast of Scotland between Sutherland and Ross furnishes an
illustration.’

t Twenty-first Ann. Report U.S. Geol. Survey, Part III, p. 149, Pl. XVI., Jour.
GEOL., Vol. IX, p. 483.

2Jupp, “The Secondary Strata of Scotland,” Quart. Jour. Geol. Soc. (1873), p
131, 134, Pl. VII; see also SUEssS, of. c2¢., Vol. I, p. 260.

WAIPIPING NOR GLTE(CRVSTALLINE SCHLSTS 889

It should be by a consideration of all the above outlined
methods that the presence or absence of a fault system should
be determined for any given area.

METHOD OF MAPPING A CRYSTALLINE AREA DEFORMED BY A
SYSTEM, “OF FOEDS IN CONJUNCTION WITH A SYSTEM OF
FAULTS.

The discovery that the beds within a region of crystalline
schists have assumed their present attitudes in consequence of
deformation of the area by a system of folds in conjunction with
one of normal faults, greatly increases, it must be admitted, the
difficulties in the way of accurately setting forth the geology
upon maps. The question may well be asked whether even with
patience and industry the areal distribution of formations can be
adequately represented. Where outcrops are abundant and well
distributed, and where formations are sharply differentiated
petrographically, the difficulties can be overcome; but where
gneiss, schist, and quartzite types by reason of gradational mem-
bers are with difficulty distinguished, even under favorable con-
ditions, their appearance in a fault mosaic will in many cases
defy the most skillful worker to unravel.

To fix the order of succession of formations, it will be neces-
sary, not only to exclude the possibilty of an overturn (as by
finding the arch of a cross fold), but actually to follow one
formation beneath the other.

In other ways the problem of mapping will be simplified.
The appearance of formations, first in one areal succession, and
again ina different succession not to be accounted for by overturn-
ing, will find an adequate explanation. Upon the theory that
folding alone has accomplished the deformation of an area of
crystalline rocks, geological sections have been strained literally
and figuratively beyond the point of rupture. The mechanics of
folding is now pretty well understood, and the appearance in
geological sections of synclines which have one limb several
times the thickness of the other (a not uncommon means of adjust-
ing to theory) finds, it is believed, as little warrant in mechanics
as it does in observation.

890 STUDIES FOR STUDENTS

In explaining an intricately faulted area on the basis of
deformation by folding only, it is also the tendency to make
geological formations much more heterogeneous than they
really are, until at last they have so many different phases that
all distinguishing petrographic differences are lost. With the
broader interpretation petrography will take a higher place in
geological mapping.

It is the intention of the writer in another place to present
with the aid of maps and photographs some concrete examples
of areas within the belt of crystalline schists where deformation
has been accomplished in part by folding and in part by a

system of faults.
WILLIAM HERBERT Hoss.
MADISON, WIS.
September I, 1902.

REVIEWS.

PRE-CAMBRIAN SUMMARIES FOR 1og0r.

Van Hise’ describes the geology of the Lake Superior iron
ore deposits. The general succession of formations in the iron-
bearing districts appears in the following table:

MESABI.

PENOKEE-GOGEBIC.

Keweenawan:
Great basal gabbro and
granite, intrusive in all
lower formations.

Gabbros, diabases, etc.

VERMILION,

Great Gabbro.

Upper Huronian (Mesabi
series):

Virginia slate (upper slate
formation).
Biwabik formation (iron-
bearing formation).
Pokegama formation
(quartzite and quartz-slate
formation).

(Penokee-Gogebic series):

Tyler slate (upper slate
formation).
Ironwood formation.
(iron-bearing formation.)
Palms formation (quartz-
slate formation).

Lower Huronian:

Granite intrusive in lower
formations.
Slate-graywacke—conglom-
erate formation. (Equiva-
lent to the Ogishke and
Knife formations of the
Vermilion district.)

Bad River limestone
(cherty limestone forma-
tion).

Animikie series:
Upper slate formation.
Gunflint formation (iron-
bearing formation).

Intrusives.

Knife slates.

Lower Huronian _iron-
bearing formation.
Ogishke conglomerate.

Archean:

Greenstones, hornblende
schists, and porphyries.

Granite and_ granitoid
gneiss.
Schists and fine-grained
gneiss.

Vermilion series:

Intrusive granites, por-
phyries, and greenstones.
Soudan formation (the
iron-bearing formation).
Ely greenstone, an ellip
soidally parted basic ig-
neous and largely volca-
nic rock.

*“ The Iron-Ore Deposits of the Lake Superior Region,” by C. R. VAN HIsE,
assisted in Mesabi and Vermilion sections by C. K. Leith and J. Morgan Clements
respectively, Zwenty-first Ann. Rept. U. S. Geol. Surv., Part III, 1901, pp. 305-434;

with geological maps.

891

REVIEWS

MARQUETTE.

CrvysTAL FALLs.

MENOMINEE,

Upper Marquette:

Michigamme formation (lo-
cally replaced by Clarks-
burg volcanic formation).
One might divide Michi-
gamme sedimentary forma-
tion into three parts:
(a) upper slate member,
(6) iron-bearing member,
(c) lower slate member.
Ishpeming formation, con-
sisting of two members;
the Bijiki schist (in west-
ern part of district), and
the Goodrich quartzite,
containing detrital ores at
its base.

Michigamme formation,
containing an iron-bear-
ing horizon not separated
in mapping for much of
the district,but in a south-
eastern part having as
lower formations (a) the
Groveland formation, and
(4) the Mansfield slate.

Upper Menominee:

Hanbury slate, bearing
in lower portions calca-
reous Slates, etc., contain-
ing siderite and iron ox-
ide.
Vulcan formation, con-
sisting in descending or-
der of three members: (a)
- Curry member (iron-bear-
ing); (4) Briar slate; (c)
Traders member (iron-
bearing).

Lower Marquette:

Negaunee formation (the
chief iron-bearing forma-
tion).

Siamo slate, containing in-
ter-stratified amygdaloid.
Ajibik quartzite.

Wewe slate.

Kona dolomite.

Mesnard quartzite.

Hemlock formation.
Negaunee formation (in
northeastern part of the
district).

Randville dolomite.
Sturgeon quartzite.

Lower Menominee:

Negaunee formation (in
small patches).
Randville dolomite.
Sturgeon quartzite.

Granite, syenite, peridotite.
Kitchi schist and Mona
schist, the latter banded,
and in a few places con-
taining narrow bands of
iron-bearing formation.
Palmer gneiss.

Granite.

Granites and gneisses.
Quinnesec schist.

Van Hise and otherst have discussed the geology of the

-Lake Superior region in previous reports, both general and
detailed. The present report is asummary of the earlier reports,

tSee especially ‘Principles of Pre-Cambrian Geology,” in Srxteenth Annual
‘Report, Part I, U. S. Geol. Surv.; pp. 571-874; Bulletin No. 86, U. S. Geol, Surv.,
Monographs U. S. Geol. Surv., Vol. XIX, XXVIII, XXXVI; and Folio U. S. Geol.
Surv., No. 62.

REVIEWS | 893

but it contains in addition many new features of interest. Atten-
tion will be directed only to such conclusions as are new or vary
from those given in the preceding reports.

In the Vermilion district the great Stuntz conglomerate and
equivalent rocks have been found to lie unconformably under
the Animikie series, which has been referred to the Upper
Huronian by the United States Geological Survey, and thus the
Stuntz conglomerate is correlated with the Lower Huronian
series instead of with the Upper Huronian, as in former reports.
The underlying greenstones, green schists, and iron formation,
the latter of sedimentary origin, are thus thrown into the
Archean.

In the Mesabi district the Keewatin of the Minnesota Sur-
vey, which was in large part classed as ‘‘Archean”’ by the United
States Geological Survey and recognized by all as being uncon-
formably below the Animikie, is subdivided on the basis of recent
work in the district by C. K. Leith into an upper series of gray-
wackes, slates and conglomerates, correlated with the Lower
Huronian, and a lower basement complex, consisting mainly of
greenstones and green schists, correlated with the Archean.
This correlation is based on the equivalence of the Animikie
with the Upper Huronian long maintained by geologists of the
United States Geological Survey.

In the Marquette district certain jaspers and associated
cherty and slaty rocks found intimately associated with the
basement rocks of the Archean, and previously supposed to be
Huronian rocks infolded with the Archean, are now themselves
called Archean.

In the Michipicoten district of Canada the iron formation,
and associated greenstones and green schists, are correlated
respectively with the iron formation with associated green-
stones and green schists of the Vermilion district, and are there-
fore classed as Archean.

The iron ores of the Lake Superior regionare supposed to have
originated from iron carbonate in all districts outside of the
Mesabi. In the Mesabi district the ores have mainly resulted
from the alteration of a green ferrous silicate in small granules

894 REVIEWS

as first shown by Spurr, but the ores have come also in small
part from the alteration of iron carbonate which is correlative
in origin with the green granules. The ferrous silicate granules
are believed not to be glauconite, as they were named by Spurr.

Comment.— The reference to the Archean of sedimentary iron
formation rocks in the Vermilion, Michipicoten, and Marquette
districts is a source of surpriseand comment among many inter-
ested in pre-Cambrian stratigraphy. The term ‘“‘Archean”’ has
been consistently used by the United States Geological Survey
for the “basement complex * of the Lake) Superion resion,
consisting essentially of igneous rocks, and all pre-Cambrian
sedimentary rocks have been referred to the Algonkian. In
survey reports on Lake Superior geology there have been slight
and ‘‘gneisses”’

”

variations from this usage, for certain ‘‘tuffs
(Kitchi and Palmer) classed as Archean in the Marquette dis-
trict have been referred to as partly sedimentary. In the present
paper Van Hise has gone a step farther and included sedi-
mentary formations of considerable importance in the Archean.
The basal complex rocks of igneous origin are very closely
associated with sedimentaries, and in many areas scarcely to be
discriminated in the mapping. On the other hand, both igneous
and sedimentary rocks are sharply separated by a profound and
conspicuous unconformity from sediments called Huronian or
Algonkian. In mapping in the Lake Supericr region it has been
found convenient, and in many cases necessary, to class together
all rocks, igneous and subordinately sedimentary, below this well
recognized unconformity at the base of the Huronian. Van Hise
has chosen to retain the term ‘‘Archean”’ for this structural unit.
It is thought by many geologists that it would have been prefer-
able to have extended the term ‘“ Algonkian”’ to include the sedi-
ments beneath the well recognized Huronian, thus adding one
more series to the Algonkian, keeping in the Algonkian all recog-
nizable sedimentary rocks beneath the Cambrian, and leaving
the term Archean for the true igneous basement complex (if there
be such a thing). If, because of the close association of the
basement igneous rocks and the sedimentary rocks beneath the
Huronian, it were found desirable to consider the basal igneous

REVIEWS 895

and sedimentary rocks together as a structural unit under one
term, the same geologists suggest that some local term might
have been applied to the basal complex in the Lake Superior
country instead of the general term ‘‘Archean.”’

The correlation of the upper series of the Mesabi and Penokee-
Gogebic districts (which are agreed by all to be equivalent to
the Animikie series) with the Upper Huronian of the original
Huronian area of the north shore of Lake Huron, first worked
out by Logan and Murray, is also questioned by certain Cana-
dian and Minnesota geologists, who believe that these series are
younger than the true Huronian of the original Huronian area.
They would correlate Van Hise’s Lower Huronian with the Upper
Huronian of the original Huronian area, and thus Van Hise’s
Archean would in large part fall into their Lower Huronian.

Finally, it is maintained by Canadian geologists and others
that the granites classed by Van Hise as Archean are intrusive
into the Huronian rocks. Van Hise’s position is that the
granitic complex contains rocks both older and younger than
the Huronian sedimentary rocks; that the fact that certain
granites are clearly intrusive into the sediments does not militate
against the evidence offered by basal conglomerates that another,
and perhaps larger portion, is really unconformably below the
Huronian rocks.

It would not be possible in a summary of this nature fully
to summarize the arguments and reasons which have led Van
Hise to the conclusions outlined in the above paper. He has in
preparation a final monograph on Lake Superior geology in
which these and allied questions are fully discussed in the light
of recent developments. In view of Van Hise’s. close and
exhaustive work in the Lake Superior region, which is in many
respects a type pre-Cambrian region, and the long period of
time during which this work has been carried on, his final state-
ment of his position with reference to the geology of the area
will be awaited with interest.

Coleman’ reports on an examination of the ‘‘ Lower Huronian”’

“Tron Ranges of the Lower Huronian,” by A. P. COLEMAN, Zenth Report
of the Bureau of Mines, Ontario, 1901, pp. 181-212.

896 REVIEWS

iron ranges at Shining Tree Lake, Clear Lake, Sault Ste. Marie,
Aberdeen Additional, Batchawana Bay and northward, Michipi-
coten Harbor, and the Helen mine, all in Ontario. As a result
of this work it is possible to greatly extend the areas of Lower
Huronian iron formation rocks in Ontario. Their general distri-
bution is stated by Coleman as follows :'

It has long been known that the Vermilion iron range of Minnesota
crossed the boundary into Ontario at Hunter’s Island, and jaspery iron ores
have been found at various points in that region—e. g., by W. H. C. Smith
on Jasper Lake, where banded jasper and hematite have a width of forty or
fifty feet. Brecciated jasper and iron ore are known from the Mattawin
region also, and from a number of points to the north, as noted by Dr. Bell;
but in the latter case the deposits may be of Animikie age, and therefore not
to be included here.

In the band of Huronian mapped as running eastward from Lake Nipigon
in the direction of Long Lake, a deposit of banded jasper and iron ore has
been examined by Mr. J. Watson Bain, near the mouth of Black Sturgeon
River, and an extension of it is reported from the same stretch of Huronian,
north of Long Lake. To the south of this, ten or twelve miles from the
mouth of Pic River, the other type of deposit is found, brecciated granular
silica with magnetite.

The Michipicoten range is separated from this by a wide area of Laur-
entian, and the first outcrop occurs about fifteen miles west of Iron Lake,
near the headwaters of Dog River. It is jaspery and cherty, with inter-
banded magnetite and hematite ; but on the eastward extension of the line
across Dog River one finds the granular variety mixed with magnetite near
Paint Lake. A few miles to the north what appears to be a parallel band
has been traced by Professor Willmott and has been found to include prom-
ising ore deposits. The whole extent of this range, as traced by Professor
Willmott, is about twenty-seven miles, the longest continuous stretch in
Ontario.

Farther to the east not much has been reported until Magpie River is
crossed by Speight’s east and west base line, where a range of hills with
jaspery and cherty material interbanded with ore runs two or three miles
southeasterly. Six or eight miles to the south the iron-bearing rocks are
found again near Park’s Lake, and can be followed four or five miles west
and southwest, including the promising Josephine mine now being developed.
In the same direction is Lake Eleanor, where siderite and the banded silica
are found; and two miles west is the Boyer Lake property described before,
the band having a length of about a mile and a quarter and running east and
west. Ten miles to the southwest is the Gros Cap deposit of sandy and

™Pp. 200-202.

REVIEWS 897

quartzitic sock interbanded with hematite, sunk upon many years ago, but
not developed to any great extent as a mine. Rock of the same sort has
been found near Michipicoten Post and Cape Choyé, Professor Willmott
having ascertained the latter range to be six miles long, though without ore
so far as known.

Once more a wide band of Laurentian intervenes, the next examples of
the iron formatian occurring sixty-five miles to the south, near Batchawana
Bay, where two bands run nearly parallel to one another, about east and
west, each from four and a half to five miles long, with reported finds some
miles to the east on Harmony River, and to the west near Pointe aux Mines
on Lake Superior. The two bands differ in character, the one to the south
being of jasper with hematite and that to the north cherty or quartzitic with
magnetite.

The next known locality, about nine miles northeast of the Sault, is of
the same character as the last, and is enclosed in Laurentian rocks instead
of Huronian or basic eruptives as in most other localities. The wide band
of Huronian between the ‘‘Soo”’ and Sudbury is not known to contain any
rocks of the iron formation, though the large numbers of bright red jasper
pebbles in the conglomerates of the upper Huronian must have a source
somewhere in the region.

In the considerable area of Huronian north of Woman River, on the main
line of the Canadian Pacific railway, jaspery iron ore has been reported, but
no details have been obtained regarding it. To the east of this Mr. Whitson
has examined a jasper range running several miles in a southeasterly direc-
tion, between the upper end of Onaping Lake and Meteor Lake; and about
twenty-five miles to the northeast is a range of jasper, chert and impure sid-
erite with magnetite and limonite of the Shining Tree Lake region, running
somewhat west of north and south of east for three anda half miles, and
perhaps for double that distance. A few miles farther northeast jasper is
reported from one of the chain of lakes belonging to Montreal River, and
still farther to the north at Night Hawk Lake.

In the southern portion of this Huronian region rocks of the kind
occur on the northwest shore of Lake Wahnapitae as dull red jasper and
chert with magnetite, and extend west in the Laurentian in Hutton and
Wisner townships. It may be that there are two separate bands here, the
southern one east and west and the northern one north and south, the latter
the more important, as containing what are said to be large deposits of mag-
netite.

Of the iron range to the northeast of Lake Wahnapitae, Professor Miller
reports that “starting from Lake Temiscaming on the east the first outcrop,
which is of small size and was not visited by me, is situated a short distance
east of the east end of Rabbit lake. The next outcrops occur along the
northeast shore of the east end of the northeast arm of Lake Temagami, the
band here stretching from near Snake Lake west to Tetapaga. Outcrops

898 REVIEWS

occur also near the end of Matagama Point. A belt lies parallel to these
outcrops to the northward and runs, with breaks here and there, from near
the west side of Net Lake to Kokoko Lake.

‘“Then there is an isolated belt stretching from near Cross Lake, north of
west, past the southern extremity of the south arm of Temagami to the south-
west arm of the same sheet of water, and to the westward outcrops are
found on Emerald Lake. A band, more or less broken, runs along the north
of Lake Wahnapitae northwest into and through part of the township of
Hutton.”

He adds that ‘in nearly all cases the iron ore, magnetite, is intimately
interbanded with jasper, which varies much in color in different parts of the
field. In some outcrops the magnetite is pretty massive, and if situated near
a railroad could apparently be worked profitably. The breadth of the band
of interlaminated material is sometimes 500 feet or more. It is at times
much bent and fractured, having been disturbed by igneous intrusions, and
some of these disturbed bands give evidence of being worthy of more careful
prospecting than we are able to do in the limited time at our disposal.”’

Belts of iron-bearing rocks, like those described in this report, are found
also in Quebec, though up to the present little has been done in searching
them out. Mr. McOuat has reported from the eighth portage of the Quinze,
on the headwaters of the Ottawa above Lake Temiscaming, an ore which
forms “layers from the thickness of paper to about an inch, and is interlami-
nated with similar layers of whitish-gray and dull red fine-grained quartzite.
The iron ore constitutes probably from a fourth to a third of the whole, and
as the thickness of the whole band is about thirty feet, the total thickness of
the layers of iron ore would probably not be less than eight feet.’’ This is
evidently the same type of deposit as those described from Ontario.

Mr. A. P. Low describes jaspery iron ores which he compares with those
of Michigan, and also cherty iron carbonates from various points in Labrador,
and probably some of these occurrences are Lower Huronian, though from
his description it is clear that most of them are of Animikie age.

From the statement just given it will be seen that bands of jaspery cherty,
or sandstone-like rock interbanded with magnetite, hematite, or limonite and
sometimes associated with siderite, occur from point to point across the
whole of northern Ontario, with lengths varying from a hundred feet to
twenty-seven miles. Almost all of the important areas mapped as Huronian
have more or less extensive belts of this rock, and in several cases isolated
patches or strips of it occur in the Laurentian, as if these were remnants left
when less resistant Huronian rocks had disappeared. These portions con-
tained in the granite are never red jasper, but generally cherty or quartzitic,
and the iron ore is magnetite, whereas in Huronian areas we generally find
jasper or granular silica with hematite or limonite.

Associated with the iron formation rocks above mentioned,

REVIEWS 899

and also in other areas, are conglomerates containing numerous
fragments of iron formation rocks, and therefore believed to be of
Upper Huronian age. In addition to the conglomerates whose
distribution was described in a previous report,’ conglomerates
are known in the following areas:

The Doré? river conglomerate, which contains many pebbles of sandstone
and chert, has been found to extend within a few miles of the Helen mine,
and to be about twenty-four miles in length from the mouth of Dog River on
the west where it begins. In sections of some of the pebbles siderite has
been found, proving that materials exactly like those at the mine were rolled
on the inter-Huronian beach before the conglomerate was formed. Similar
conglomerates have been found at other points in this Huronian area; for
instance, two or three miles north of Coetz Lake, not far from the Josephine
mine.

Conglomerates have not yet been found nearer the Batchawana jasper and
chert beds than atthe north end of Goulais Bay, fifteen miles to the south, where
Murray mapped jasper conglomerate many years ago. The extensive bands
of quartzitic conglomerate containing blood-red jasper pebbles in the original
Huronian region, extending from Lake George almost to Thessalon, about
thirty miles, and found in several different bands, some of them quite to the
north of those mapped by Murray, have never been accounted for, since no
jasper has been found nearer than Batchawana, more than fifty miles to the
northwest, and there the jaspers are much duller in color. The accompany-
ing black chert pebbles, which are equally common, might have been supplied
by the cherty iron ore band nine miles northeast of the Sault Ste. Marie,
mentioned in a previous part of this report, though this is about ten miles
from the nearest of the conglomerates. The region is, however, little known
beyond the few miles of settled country along the St. Mary’s River and the
north shore of Lake Huron, and future exploration may solve the problem.

Numerous conglomerates occur along the same stretch of Huronian to
Sudbury, but no jasper or chert pebbles are known in them, though they are
found in quartzites and graywackes somewhat farther east on Lake Mata-
gamashing, not far from the jasper iron ore belt north of Lake Wahnapitae.
Small amounts of jasper conglomerate have been noted northward from this,
anda graywacke conglomerate containing jasper and chert pebbles extends for
some miles parallel to the Shining Tree Lake iron range but a mile or two to
the west.

East of Lake Wahnapitae conglomerates with jasper are known at vari-
ous points to Lake Temiscaming, and also on the Quebec shore of that lake

"Bureau of Mines, 1900, pp. 180-86, and Bull. G. S. A., Vol. XI, 1900, pp.
107-14. See Summaries, JouR. GEOL., Vol. IX, 1901, pp. 447-9.

JP p203-—4.

goo REVIEWS

near Baie des Perés. That they extend still further to the east, is shown by
Low’s report on Labrador, where conglomerates with Laurentian bowlders
and jasper bowlders and pebbles seem to be common.

It is not certain, of course, that every one of these rocks containing
pebbles of jasper, chert, or sandstone is a basal conglomerate of the Upper
Huronian, but many of them undoubtedly are, and in the majority of cases
the source of their pebbles is found in adjoining bands of siliceous iron-bear-
ing rocks which may be looked on as belonging to a horizon near the top of
the Lower Huronian, Van Hise’s Mareniscan.

Willmott’ describes the geology of the Michipicoten area
northeast of Lake Superior. He makes the succession from the
base up as follows:

i. lower\ Eluronian: ‘green ‘schists. “Some of jthese jare
undoubted lava flows showing the characteristic elliptical struc-
ture described by Clements as occurring in the Hemlock forma-
tion of the Crystal Falls district. At a number of points
agglomerates are found, as at Little Gros Cap fish station, north
of Goetz Lake, east of Manitowoc, and elsewhere. Commoner
occurrences are the various green schists, chlorite, hornblende,
mica, and sericite schists. Presumably all these schists are
derived from lavas, basic and acidic. The dip of the schists is
always nearly vertical and the strike follows closely the line of
contact with the granite, to be described later.

2. Lower Huronian sediments. The most characteristic of
these is a belt of ferruginous chert, which has been found at
intervals for about sixty miles. This rock consists of banded
hematite and silica with usually some residual carbonate of iron.
The bands vary from one-tenth of an inch in thickness up to
several inches. The silica is sometimes very like loaf sugar;
again it is like quartzite, chert, or jasper. Red jasper is not
infrequent. The hill at the back of the Helen mine is a huge
mass of siliceous carbonates. The rocks, as a whole, and the
mode of occurrence of the ore, are strikingly like the Lower
Huronian iron formations of Marquette and Tower. Besides the
iron formation, beds of carbonaceous shales and limestones have
been recognized at several points. Shale occurs interstratified

™The Michipicoten Huronian Area,” by A. B. WILLMOTT, American Geologist,
Vol. XXVIII, 1901, pp. 14-19.

REVIEWS gol

with the ferruginous chert at Iron Lake. Near Paint Creek and
at Eleanor Lake it has also been found. Whether it always
underlies the iron formation is undetermined, but it probably
does not. The cherty limestone has been traced in a fairly con-
tinuous line from the Helen mine to the east of Park’s Lake, a
distance of twelve miles.

3. Upper Huronian sediments. These consist of schist-con-
glomerate. Coleman describes the bowlders as ‘granites most
frequently, then quartzites, or sandstones with pebbles generally
small, next green schists, then felsite schists and porphyroids,
and finally a few gneisses, but none of the Laurentian type.” A
thin section of one of the quartzite pebbles showed considerable
carbonate, proving that it undoubtedly came from the iron
formation. This conglomerate has been traced pretty contin-
uously for thirty-eight miles in a semicircular belt around the cen-
tral granite boss. At many points, as at Iron Lake, Dog River,
Doré River, Wawa Lake, it contains pebbles from the very char-
acteristic iron formation. This fixes its age as Upper Huronian.

4. Laurentian granite, intrusive in the Huronian rock. The
granite is in undoubted eruptive relations with the conglomerate
along the shore of Superior, for example a few miles west of
the Doré. A mileand a half up the Magpie, a boss of granite is in
eruptive contact with the conglomerate, and although it may not
be of the same age as the larger boss three miles to the north-
west, it probably is. In the opposite direction a succession of
granite gneiss bosses intrusive in the schists are found, for six
miles, after which the granitoid gneiss occurs without interrup-
tion for over a hundred miles.

Comment.—Van Hise’ has referred Willmott’s ‘‘ Lower Huron-
ian’’ green schist and iron-bearing formation to the Archean,
and has referred his ‘‘Upper Huronian”’ schist-conglomerate
series to the Lower Huronian. Indeed he would so refer most
of the conglomerates described as Upper Huronian by Coleman
in the article above summarized. The Michipicoten series show
close similarities in structure and lithology to portions of the
Archean and Lower Huronian series, respectively, of the Ver-

*See summary of Van Hise’s report on a preceding page.

902 REVIEWS

milion iron-bearing district of Minnesota. There is little dispute
as to the equivalence of the series in the two districts, but there
is dispute as to their correlation with the original Huronian series
of the north shore of Lake Huron, and thus as to their nomen-
clature. At the east end of the Vermilion district they have
been proven to underlie unconformably the Animikie series of
the Mesabi district and its eastward continuation to Thunder
Bay on Lake Superior, which has been correlated by Van Hise
and others with the Upper Huronian of the original Huronian
area, but which the Canadian geologists and others régard as
younger than the true Upper Huronian. As already noted, Van
Hise has in preparation a final monograph on Lake Superior
geology in which the position of the United States Geological
Survey with reference to the correlation of these series in the
light of recent work will be fully stated.

Bain! describes the iron belt of Lake Nipigon, devoting his
attention mainly to economic and petrographic features. The
stratigraphical features are covered in the general report of
Coleman summarized on a previous page.

Miller? describes the iron ores and associated rocks of the
area adjacent to Lake Temegami and of the Lake Wahnafutae
and Hutton areas to the west, all in the Nipissing district of
Ontario. The Temegami has been previously mapped and
reported upon by A. E. Barlow.3 Miller’s discussion of the
general geology of this area follows that of Barlow, with minor
additions and corrections.

Winchell‘ publishes a geological atlas of the state of Minne-
sota with synoptical description of plates. This volume con-
tains maps and general conclusions found in Volumes IV and V
of the Minnesota Survey, summarized in this JOURNAL, Vol. IX,
pp. 79-86. One additional map is published, a general geologi-
cal map of the state.

«The Iron Belt on Lake Nipigon,”’ by J. W. Bain, Tenth Report of Bureau of
Mines, Ontario, 1901, pp. 212-14.

2“Tron Ores of Nipissing District,” by WILLET G. MILLER, Jenth Report of the
Bureau of Mines, Ontario, 1901, pp. 160-80.

3Annual Report Geol. Survey of Canada, Vo). X, Part I, for 1897, pp. 302. Sum-

merized, JouR. GEOL., Vol. VIII, 1900, pp. 439-41.
4 The Geological and Natural History Survey of Minnesota, Vol. VI, 1900-1.

REVIEWS 903

Duparc’ describes the copper-bearing (Keweenawan) rocks
of the northwest extremity of Keweenaw Point, Michigan. The
article is throughout merely a summary of previous reports on
this area by the geologists of the Michigan and United States
Survey, and this moreover without a single reference to such
reports.

Hall? describes the Keweenawan rocks south and southwest
from Duluth, along the St. Louis and St. Croix rivers, and shows
that a series of alternating lava flows and sediments le in a
synclinal northeast-southwest trough, the western border of
which is marked by a profound fault. To the east of the fault
the Keweenawan rocks are highly tilted to the southeast, while
to the west of it the Cambrian rocks are much broken up. The
relations of the fault to the distribution of the flows and
analogy with other volcanic regions seem to show that the fault
was a plane of weakness along which most of the lavas were
originally erupted. The faulting was pre-Cambrian, Cambrian,
and post-Cambrian, as shown by the fact that in some places
the Cambrian rests horizontally upon the upturned Keweenawan
rocks, and in others is much broken up.

Hall? describes and maps the slates and associated rocks in
the vicinity of Cloquet and Carlton on the St. Louis River, and
certain hornblendic and micaceous schists associated with gran-
ite and diabase to the west along the Mississippi and Snake
rivers. He maintains that the slates and graywackes to the
east and the hornblendic schists to the west belong to one
and the same series, and that the schists have resulted from the
metamorphism of graywacke and slate by the intrusion of
granite. Still later intrusions of diabase have cut both the
granites and the slates. Accepting Spurr’s statement that
the Carlton slates are Keewatin or Lower Huronian, the con-

*“Note sur la Region Cuprefere de l’extremite Nord-Est de la peninsule de
Keweenaw (Lac superieur),” par Louis Duparc, Archives Sci. Physique et Nat.,
Tome X, 1900, p. 21.

2“ Keweenawan Area of Eastern Minnesota,” by C. W. HALL, Bzlletin of the
Geological Society of America, Vol. XII, 1901, pp. 313-42, Pls. 27-28.

3“ Keewatin Area of Eastern and Central Minnesota,” by C. W. HALL, Bulletin
of the Geological Society of America, Vol. XII, 1901, pp. 343-76, Pls. 29-32.

904 REVIEWS

clusion is reached that the schists to the southwest are Lower
Huronian, and that the intruding granites are post-Lower
Huronian. If this conclusion be correct, the granites and schists
of the central and eastern portions of Minnesota must be mapped
as Algonkian rather than Archean, as in the past.

Ami? briefly summarizes the salient features of the geology
of the principal cities of eastern Canada, including St. John,
Ottawa, Quebec, Montreal, and Toronto.

Ami? summarizes the geology of Canada, and indicates the
meaning and correlation of the principal terms employed in
Canadian geological nomenclature.

Ells3 sketches the development of geological work in the
province of Quebec.

Ells+ describes and maps the geology of the Three Rivers
map, Province of Quebec.
Archean rocks occupy most of the northwestern portion of the

”)

sheet of the ‘‘ Eastern Townships

area north of the St. Lawrence River. <A portion of this area,
including anorthosite masses, has previously been described by
Adams.5

The great mass of the rocks seen pertain to the Grenville series,
rather than to the so-called ‘‘ Fundamental’”’ gneiss. The composi-
tion of the Grenville series, with its crystalline limestones and
with rusty gneiss bands, very closely resembles that met with in
the lower Ottawa district, but the calcareous members are much
less widely developed. There are also large areas of anortho-
site, red granite, augen-gneiss and masses of green pyroxenic
diabase. Quartzite is an important component of this series,
and large areas of this rock, similar to that found along the

™On the Geology of the Principal Cities of Eastern Canada,” by HENRY M.
Ami, 7vans. Royal Soc. of Canada, Vol. VI, 1900; sec. iv, pp. 125-64.

2“ Synopsis of the Geology of Canada,” by HENRY M. AMI, 7vans. Royal Soc.
of Canada, Vol. VI, 1900, sec. iv, pp. 187-225.

3“ Problems in Quebec Geology,” by ROBERT W. ELLS, Record of Scéence, Vol.
VII, 1898, pp. 480-502.

4 Geol. Surv. of Canada, Annual Report, New Series, Vol. XI, for 1898, pp. 5 J-70
J; with geological map.

5 Geol. Surv. of Canada, Vol. VIII, Part J. Summarized, Jour. GEOL., Vol]. VII,
p. 401.

REVIEWS 905

Ottawa, are found associated with the gneiss as far north as the
northern limit of the map-sheet.

The definition of the so-called fundamental gneiss is, as a
matter of fact, not always possible in this district. If the
latter appears at all, it must be along the crests of some of the
numerous north-south anticlines, which are generally low, the
rocks over a large area being inclined at low angles. The pre-
vailing gneiss is a grayish and hornblendic variety, generally
quartzose, and with frequent bands in which garnets are
abundant.

The anorthosites are intrusive in the Grenville series. The
Grenville series are correlated with the Hastings series and both
are equivalent to the Huronian. The Fundamental gneiss is
Laurentian.

Bell? describes and maps the geology of the Baffin Land
shore of Hudson Strait.

The rocks of the northern side of Hudson Strait from North
Bay to Chorkback Inlet and inland to Lake Mingo consist of
well stratified hornblende and mica-gneiss, mostly gray in color,
but sometimes reddish, interstratified with great bands of crys-
talline limestones, parallel to one another and conformable to
the strike of the gneiss, which in a general way may be said to
be parallel to the coast in the above distance. The direction,
however, varies somewhat in different sections of the coast. All
are of Laurentian age.

The distinguishing feature in the geology of the southern
part of Baffin Land is the great abundance, thickness and regu-
larity of the limestones associated with the gneisses. At least
ten immense bands, as shown on the accompanying map, were
recognized, and it is probable that two others, discovered in
North Bay, are distinct from any of these. There would, there-
fore, appear to be twelve principal bands as far as known, to say
nothing of numerous minor ones, between Icy Cape and Chork-
back Inlet. Their total thicknesses may be 30,000 feet, or an
average of 2,500 feet for each of the principal bands. These
rocks are correlated with the Grenville.

"Geol. Surv. of Canada, Annual Report, New Series, Vol. XI, Part M, for 1898,
pp. 5M-38M; with geological map.

g06 REVIEU Ss

Low? describes and maps the geology of the south coast of
Hudson Strait and the west and south shores of Ungava Bay.
Granite and gneiss of various ages occupy three-fourths of the
coastal area. Associated with them are gabbros, diabases and
other greenstones, cherts, quartzites, shales, and slates. All are
provisionally referred to the Archean.? Flat-bedded Cambrian
rocks rest with apparent unconformity upon the crystalline
complex.

Tyrrell and Dowling? describe and map the east shore of Lake
Winnipeg. The rocks are all Archean and the great prepon-
derance of gneisses and granites of the Laurentian is the chief
feature. Small areas of Huronian greenstones and schists occur
in two localities, one on Lac du Bonnet and the other at the
mouth of Wannipegow River.

In an account of a trip from Edmonton through Yellow Dog
Pass, in the Rocky Mountains, to Canoe River a tributary of the
Columbia River, McEvoy‘ describes and maps Shuswap rocks,
of Archean’ age, occurring on Mica Mountain near the west-
ern end of the route. The series includes dark glittering
mica-schist, easily weathering, thinly foliated garnetiferous mica-
schist, with a high percentage of mica and garnet, hard garnet-
iferous mica-schist in massive beds, bands of dark fine-grained
micaceous rock apparently of eruptive origin, and layers of fine-
grained gneiss which, in some instances at least, is certainly
intrusive. The whole series, while differing somewhat from the
Shuswap series of the southern interior of British Columbia,
shows the main characteristics of that series and may be classed
as such: The age of this series as given by Dr. Dawson is
Archean.

t Geol, Surv. of Canada, Annual Report, New Series, Vol. XI, Part L, for 1898,
pp. 5 L-47 L; with geological map.

2 Pre-Cambrian.

3 Geol. Surv. of Canada, Annual Report, New Series, Vol. XI, for 1898, pp. 5 G-98
G; with geological map.

4 Geol. Surv. of Canada, Annual Report, New Series, Vol. XI, for 1898, pp. 5 D-44
D; with sketch map.

5 Pre-Cambrian.

REVIEWS 907

Dawson‘ describes the geology of the Rocky Mountain region
in Canada. The oldest rocks of the region belong to the Shus-
wap series of Archean? age. The Shuswap series characterizes
considerable areas of the Selkirk, Columbia, and adjacent ranges
in the southern part of British Columbia. It is known also in
the Cariboo mountains and near the sources of the North Thomp-
son and Fraser, about latitude 53°. It is again well developed
on the Finlay River, where the country has been geologically
examined, between the 56th and 57th parallels of latitude.
Northward to this point these rocks appear to be confined to a
belt lying to the west of the Laramide range and to come to the
surface seldom, if at all, in that range. Further north similar
rocks occur in the Yukon district in several ranges lying more
to the west, but still with nearly identical characters, in so far as
they are known. The Shuswap series includes highly meta-
morphosed sediments with perhaps the addition of contemporane-
ous bedded volcanic materials. They are grayish mica-gneisses,
with some garnetiferous and hornblendic gneisses, glittering
mica-schists, crystalline limestones and quartzites. Gneisses in
association with the last mentioned rocks often become highly
calcareous or siliceous and contain scales of graphite, which are
also often present in the limestones. These bedded materials
are, however, associated with a much greater volume of mica-
schists and gneisses of more massive appearance, most of which
are evidently foliated plutonic rocks, and are often found to pass
into unfoliated granites. The association of these different
classes of rocks is so close that it may never be possible to
separate them on the map over any considerable area. The
granites may often have been truly eruptive in origin, but the
frequent recurrence of quartzites among them in some regions
indicates that they are at least in part, the result of a further
alteration of the bedded rocks. The original bedded portions
of the series closely resemble those of the Grenville series of the

™“ Geological Record of the Rocky Mountain Region in Canada,” address by the

President, GEORGE M. Dawson, Bulletin of the Geological Society of America, Vol.
XII, 1901, pp. 57-92.

2 Pre-Cambrian.

908 REVIEWS

Province of Quebec, and the associated gneisses resemble the
Fundamental Gneiss of the same region. The greatest thick-
ness of the Shuswap rocks so far measured, where there is no
suspicion of repetition, on Kootenay Lake, is about 5,000 feet,
but even here there are doubtless included considerable inter-
calations of foliated eruptives.

Schrader* describes certain granites and schists seen in a
reconnaissance along the Chandler and Koyukuk rivers of Alaska.
These are referred to as ‘‘basal”’ and certain of the schists are
correlated with Spurr’s Birch Lake series of schists, but no
attempt at further correlation is made.

Brooks? describes the Archean of the Tanana-Yukon divide.
A broad belt of crystalline rocks extends in a northeast-south-
west direction in the region of the Tanana-Yukon divide,
embracing a series of gneisses, mica-schists, basic schists, and
various intrusives, chiefly of an acid character. Near the middle
Tanana this series bends to the west and south, and its continua-
tion is to be sought for in the region of the upper Kuskokwim.
To the southeast this belt is probably continued by the granitic
rocks on the Pelly River, described by Dawson. What evidence
we have goes to show that this is the basal series of the Yukon
Basin, and as it contains no recognizable detrital material it can
properly be assigned to the Archean. Whatever the original
character of the rocks may have been, they are now essentially
mica-schists and gneisses, with considerable intrusive material.
Their metamorphic condition is the strongest argument for con-
sidering them older than any of the sedimentary rocks.

Comstock’ reviews the stratigraphy of Arizona. Granites,
gneisses, and schists, probably of pre-Cambrian age, occur at

™“ Preliminary Report on a Reconnaissance along the Chandler and Koyukuk

Rivers, Alaska, in 1899,” by F. C. SCHRADER, Zwenty-first Ann. Rept. U. S. Geol.
Surv., Part II, t901, pp. 441-86; with sketch map.

2“ A Reconnaissance from Pyramid Harbor to Eagle City, Alaska, Including a
Description of the Copper Deposits of the Upper White and Tanana Rivers,” by
A. H. Brooks, Twenty-first Ann. Rept. U. S. Geol. Surv., Part Il, 1901, pp. 331-
91; with sketch map.

3° The Geology and Vein Phenomena of Arizona,” by T. B. Comstock, 77ans.
Am. Inst. Min. Eng., Vol. XXX, 1900, pp. 1038-1101.

REVIEWS 909

various places, but have not been thoroughly studied. The
massive granites are exposed in a limited tract surrounding
Prescott, including Granite Mountain, and farther southwest in
the Grand Canyon of the Colorado, in the valley of the Colorado
jn Yuma county, and in the Little Dragoon Mountains, Cochise
county. These massive granites form a basement for the other
rocks of the region.

In a number of places are found fissile granites which appear
to lie between the massive granites and the schistose strata
which form the floor of much of the area of western Arizona.

Schists are well exposed in northwestern Arizona, in Mohave
county, with E.-W. strike, tilted at high angles. Very similar
exposures occur in other districts, as near Oracle and Mammoth,
in Pinal county; in tracts in Pima county, and in portions of
Yuma county. In Graham and Cochise counties, and, to a less
extent, in Gila county, with occasional outcrops near the adjoin-
ing boundary of Maricopa and Yavapai counties, the same trend
is prominent. Erosion has exposed portions of the same
terrane, thrown into the N.E.-S.W. trend, in limited areas in
Graham and Yavapai counties, and, possibly, in the northeast
portion of Yuma county.

Blake’ describes the salient features of the geology of Arizona.
The Santa Catalina, Rincon, and Rillito mountains consist largely
of granitic gneisses and schistose rocks of pre-Cambrian age with
a highly complex folded structure, and exhibiting a high degree
of metamorphism. Taken together, these mountains may be
regarded as the main axis of ancient uplift, and of insular land
areas in the pre-Cambrian and Paleozoic periods, the beginning
ofthe “Arizona wands

The gneiss of the southern side of the Santa Catalina near
Tucson is regarded as Archean. It is remarkable for its regu-
larity of stratification and its great thickness, probably over 10,000
feet. It occurs in great tabular masses made up of thin layers
which, when seen laterally, give the appearance of evenly strati-

“Some Salient Features in the Geology of Arizona with Evidences of Shallow

Seas in Paleozoic Time,” by WILLIAM P. BLAKE, American Geologist, Vol. XXVII,
IQO1, pp. 160-67.

QIO REVIEWS

fied shales and sandstones. In the same range, but on the
northeastern side, facing the valley of the San Pedro, another
formation of thinly bedded and highly crumpled mica schist in
sharply defined zigzag folds is referred to the Huronian, and is ~
given the name Arizonan.

Hershey* describes the schistose rocks of the Klamath
Mountains in northwestern California. On the whole it seems
impracticable to fix upon any particular part of the time
between the Archean and the Devonian as the period of deposi-
tion of the Klamath schists, but it is believed that the evidence
favors the earlier or Algonkian portion rather than the Cambrian
or Silurian portion.

Spencer*® describes Algonkian rocks occurring in the center
of the Rico Mountains of Colorado. They consist of quartzites,
quartzitic schists, and biotite and actinolite schists. The exposed
thickness of the quartzites is over 350 feet and probably as
much as 500 feet. The relations of the quartzites to the schists
have not been ascertained. The schists and quartzites of this
area are similar in every way to the series of rocks exposed in
the upper part of the Animas Canyon and in adjacent portions
of the Quartzite or Needle Mountains, where they have been
referred to the Algonkian by Emmons and Van Hise. The
quartzite of the Rico Mountains is directly along the strike of
the great quartzite belt in the Animas Canyon and Needle Moun-
tain area.

Jaggar,3 in connection with the discussion of the laccoliths
of the northern Black Hills, incidentally refers to the structure
of the Algonkian. Its lamination abuts abruptly against the
hard basal Cambrian quartzite or the conglomerate, and has a
fairly uniform strike of north-northwest. The Algonkian surface
is seen to be warped.

t™ Metamorphic Formations of Northwestern California,” by OScAR H. HERSHEY,
American Geologist, Vol. XX VII, 1901, pp. 225-45.

2 Twenty-first Ann. Rept. U.S. Geol. Surv., Part II, 1901, pp. 37-78; with geo-
logical map.

3“ The Laccoliths of the Black Hills,’ by T. A. JAGGAR,JR., Zwenty-first Ann.
Rept. U.S. Geol. Surv., Part III, 1901, pp. 171-303.

REVIEWS gil

Hills* describes the geology of the Walsenburg Folio of
Colorado.

The principal mass of the Greenhorn Mountains consists of
coarse and fine-grained granites and gneisses, hornblende, mica,
and chlorite-schist,and subordinate masses of garnet-and epidote-
schist, and occasional vein-like bodies of coarse pegmatite. The
schistose rocks are more prominent at the southern extremity of
the mountains than elsewhere, while the granite and gneissic
rocks are more prominent in the main mass toward the culmi-
nating point.

Their origin probably dates back to the Archean period.
No further correlation is attempted.

Watson? describes the Crane rocks. of the Piedmont
Plateau of Georgia and concludes that they were not all con-
temporaneous in origin. Some of them are pre-Cambrian, while
others may possibly be later in age. However, the youngest
acid intrusives could not have been later than, if as late as, the
last great Appalachian disturbance or uplift.

Kemp, Newland and Hill? further discuss the geology of
Hamilton, Warren and Washington counties, extending the
observations noted in previous reports to the west and south.
The additional points of interest are the occurrence of anortho-
sites in Johnsburg, in Warren county, the southernmost exposure
yet known in the eastern mountains, and ‘the increasing
certainty of the existence of sedimentary gneisses in Fort Ann
and Johnsburg townships.”

™“ Walsenburg Folio, Colorado,” by R. C. HILLs, Geol. Atlas of the U. S., No.
68, 1900.

2“ The Granitic Rocks of Georgia and their Relationships,” by THoMAs LEONARD
WATSON, American Geologist, Vol. XX VII, 1901, pp. 199-225.

3“ Preliminary Report on the Geology of Hamilton, Warren and Washington
Counties,” by J. F. Kemp, D. H. NEWLAND, and B. F. HILL, Zighteenth Ann. Rept.
Stale Geologist of New York, published in Fifty-second Rept. NV. Y. State Museum, Vol.
II, 1900, pp.141-62; with geological maps.

4For general discussion of classification of geology of these counties see Refz. of

the New York State Geologist for 1897, published in Fifty-first Ann. Rept. New York
State Museum, Vol. 11, 1899, pp. 499-553. Summarized, Jour. GEOL., Vol. IX, rgor,

Pp. 444.

gi2 REVIEWS

Smyth’ maps and describes the geology of the crystalline
rocks in the vicinity of the St. Lawrence River in the towns of
Alexander, Clayton, and Theresa, in Jefferson county, together
with portions of Rossie and Hammond in St. Lawrence county.
The crystalline rocks of the district are discriminated on the
map as granite, granite-gneiss, granite-gneiss with much schist
and quartzite, schists and quartzites with much granite-gneiss,
schists and quartzites of limestone series, and crystalline lime-
stone. This classification indicates the close association of the
various rocks in the field. Under the term schists are classed a
variety of rocks, such as quartzite, hornblende and mica schists,
hornblende, pyroxene, and mica gneisses, etc., of both igneous
and sedimentary origin. The gneiss formation is not a unit, but
rather a complex, so far as age is concerned. It is the most wide-
spread of the pre-Cambrian rocks. There is abundant evidence
that the granite and granite-gneiss are for the most part younger
than the schists and quartzite. The granite has a close genetic
relation to the granite-gneiss and is identical with a part of the
latter. The granite may be the youngest member of the
granite-gneiss complex. The quartzite may have a thickness as
great as 500 feet.

Kemp and Hill? report progress of work on the pre-Cambrian
formations in parts of Warren, Saratoga, Fulton, and Mont-
gomery counties.

Cushing’ maps and describes the geology of Rand Hill and
vicinity, Clinton county. The basal rocks are gneisses of the
Dannemora formation occurring in the southwestern portion of
the area. These are of doubtful origin, but probably mostly

™ Geology of the Crystalline Rocks in the Vicinity of the St. Lawrence River,”
by C. H. SMYTH, JR., Wineteenth Ann. Rept. State Geologist of New York, published
in Fifly-third Ann. Rept. N. Y. State Museum, Vol. 1, 1901, pp. 185-204; with
geological map.

2“ Pre-Cambrian Formations in Parts of Warren, Saratoga, Fulton, and Mont-
gomery Counties,” by J. F. Kemp and B. F. HILL, Mineleenth Ann. Rept. State Geolo-
gist of New York, published in Fifty-third Ann. Rept. N. Y. State Museum, Vol. 1,
I9OI, pp. 121-135; with geological maps.

3**Geology of Rand Hill and Vicinity, Clinton County,” by H. P. CusHING,
Nineteenth Ann. Rept. State Geologist of New York, published in Fifty-third Ann. Rept.
N. Y. State Museum, Vol. I, 1901, pp. 139-182; with geological map.

REVIEWS 913

igneous. Intrusive in them are undoubted igneous rocks—
gabbro, anorthosite-gabbro, augite-syenite, syenite and dikes of
syenite and diabase.

CAKE
MADISON, WIs.

THE Summary Report of the Geological Survey Department of Canada
for the year Igor is a paper-bound volume of 270 pages, giving a
concise outline of the work done during the year. The volume is
issued in order to give to the public, without delay, the results of the
year’s investigations in different parts of the Dominion. The rapidly
increasing mining industry of the country makes it very desirable that
the information in possession of the Survey should be immediately
available, whereas the full annual reports are, as a rule, two or more
years behind the field work. During the last three years the Canadian
Survey has lost a number of its men by death and resignation. The
death of Dr. G. M. Dawson, late director, was a severe loss, and the
resignation of such men as J. B. Tyrrel, A. P. Low, J. McEvoy and
R. W. Brock leaves places which the younger appointees are not yet
able to fill. Dr. Robert Bell, formerly deputy director, has been
appointed acting director. The present staff numbers fifty-four.

During the year rgor thirty-one parties were in the field, and work
was done in each of the seven provinces and in the territories of
Alberta, Yukon and Keewatin. R.G. McConnel and Joseph Keele
continued the work in the Yukon field. Considerable searching has
been done in the hope of finding the lodes from which the gold of the
placers was derived. Auriferous quartz veins cut the igneous and
clastic schists which are so abundant in the Yukon valley. Asa rule,
the veins, though numerous, are too small and discontinuous to
warrant mining operations. The schists are also auriferous in places.
R. A. Daly acted as geologist with the Canadian commission
co-operating with the United States commission in locating the British
Columbia section of the international boundary. ‘he predominant
rocks of the Coast Range were found to be metamorphic sediments,
of which but few strata were fossiliferous. The mountain forms are
regarded as erosional rather than constructional. Erosional features
characteristic of the work of Alpine glaciers—cirques, cols, rock
basins, ampitheaters, and deep re-entrants—are abundant. The
existing glaciers are small, and there is no indication of general
glaciation having prevailed in the belt. R. W. Brock worked in the

Q14 REVIEWS

Boundary Mining District of British Columbia. The rocks of the
district are largely eruptives and intrusives, consisting of greenstones,
granites, tuffs, and lava, though older sediments, consisting of lime-
stones, argillites, quartzites, and other metamorphic rocks have been
caught up and included in the igneous flows. The region is wholly
within the area covered by the Cordilleran ice sheet. The strie
show that the movement was mainly S. 30° E. Assorted glacial
materials are very abundant, and terracing is very prominent, often
reaching a height of 2,000 feet above the present valley levels. The
ore bodies are of large size, but irregular in form and usually of low
grade. ‘They occur in all rocks except the newest, and mineralization
has extended, with gradually decreasing richness, far into the country
rock. Deposits within the limestones or the greenstones and at the
contact of the two, are’ of most frequent occurrence: The xvein
minerals are chiefly magnetite, pyrrhotite, chalcopyrite, marcasite,
arsenopyrite, and micaceous hematite, with more or less galena,
sphalerite, and molybdenite. Magnetite is never abundant where
pyrrhotite is prominent. The values are principally in copper, gold,
and silver. The smelters at Grand Forks and Greenwood have been
working steadily, but the pyritic smelter at Boundary Falls has not
yet been blown in. W. W. Leach continued the examination of the
Crow’s Nest coal fields, which lie on the eastern border of British
Columbia. The main area covers about 230 square miles, has twenty-
two distinct seams with a total thickness of 216 feet, of which too feet
is workable coal of excellent quality. The smaller, or Green Hills
area, covers about seven square miles and contains 79 feet of workable
coal in seven seams. L. M. Lambe resumed his work on the Creta-
ceous rocks of the Red Deer River, Alberta, and succeeded ‘‘in securing
a large collection of chelonian, dinosaurian, crocodilian, fish, primitive
mammalian, and other vertebrate remains of considerable value.”
Two Chelonians— 77ionyx foveatus and Trionyx vagans—are figured
and described. William McInnes worked on the pre-Cambrian for-
mations between Lakes Superior and Manitoba. ‘The primary object
of the season’s work was to trace with greater accuracy the Sturgeon
Lake gold-bearing belt and to work out the geology of an area lying
to the southeast of the eastern half of Lac Seul.” All of the impor-
tant Huronian areas of this region are now mapped. Some promising
mineral lands are being developed. A. W. G. Wilson spent the
summer in the region west of the Nipigon river and lake. ‘The rocks
of the area studied belong to the Laurentian, Huronian, and Animikie

REVIEWS O15

series. The Laurentian rocks consist of gneisses and granites with
some schists; the Huronian, of ferruginous quartzite interbanded with
hematite of good quality and varying in character from soft red to
specular. The Animikie formation is the most widespread. A deep
red dolomite, varying from a rather heavy bedded, compact, fine-
grained rock to a coarser shaly variety, is the most important litho-
logical unit. Associated with the dolomite is a highly ferruginous
dolomitic sandstone which locally carries angular masses of quartz.
Conformably overlying the red dolomite, on the Upper Spruce river,
is a fine-grained, non-fossiliferous, greenish-gray shale, which is capped
by fifteen feet of trap. Trappean flows, probably gabbroid, occasion-
ally reaching a thickness of 300 feet, overlie the sediments for the
most part, but are in places interbedded with them. In the eastern
half of the Nipigon area the Huronian schists pass into jasper and
hematite and form three iron ranges, one north and two south of
Sturgeon river. Animikie traps are abundant in this area, and show
about the same relations to the sedimentary series as in the western:
half. The explorations of D. B. Dowling in the region to the west of
James bay have shown that the Paleozoic area beginning at the
southern extremity of James bay extends uninterruptedly along the
coast to cape Churchill in Lat. 59° N. The rocks of the newly
explored area in the angle between James and Hudson bays are of
Silurian age. To the southwest of Cape Henrietta Maria is a consider-
able area of Animikie rocks consisting of ferruginous slates and jaspers
capped by trap.

Work in the vicinity of Lake Abitibi near the Ontario-Quebec
boundary resulted in the more accurate mapping of the pre-Cambrian
rocks. A. E. Barlow has added many important details to the recon-
naissance work done in Denison, Graham and Creighton townships
of the Sudbury mining region in 1890. The unaltered rock of
the nickel-copper deposits is a.quartz-hypersthene-gabbro or norite.
The pyrrhotite, chalcopyrite, and pyrite. were original constituents of
the magma. In the process of slow cooling of these norite masses a
species of differentiation resulted in the segregation of the sulphides
“so that the final solidification saw the ore-bodies under very much
the same conditions as at present obtain.”’ Associated with the norite
in the ore-bearing ranges are micropegmatites which are undoubtedly
differentiated portions of the nickel-bearing eruptive, as there is a
perfect transition from one rock type to the other. ‘There are three
distinct bands of the norite and associated micropegmatites. Of these

g16 REVIEWS

the southern, having a length of thirty-two miles and an average
width of two miles, is the most important, though the others are not
without promise. F. D. Adams returned to his work in the pre-
Cambrian area of eastern Ontario. Two maps, covering about 4,200
square miles, have been prepared. The northern half of the area
is occupied by granite-gneiss (probably equivalent to Logan’s
fundamental gneiss) while the rocks of the southern area are chiefly
ancient sediments, largely limestones, resting on the gneissic series,
but invaded, brecciated, and altered by it. Bathylithic masses of
gneiss occur in both the granite-gneiss and the ancient sedimentary
areas. In the southeastern portion of the area, associated with com-
paratively unaltered limestones, are masses of amphibolite and other
foliated rocks, with an occasional band of conglomerate. ‘The nephe-
line syenite of the southern portion of the areaseems to have some gen-
etic connection with the limestones and with the granites. Intrusions of
gabbro are found associated with the amphibolites. In southwestern
Ontario, R. Chalmers made a reconnaissance survey of the superficial
deposits, and has recorded some observations relative to recent changes
of the level of Lakes Huron and Erie. He also gathered data
bearing on the salt, oil, and gas industries of that part of the province.
In eastern Ontario R. W. Ells has made a re-survey of the Paleozoic
formations about Kingston with a view to determining the age of
certain limestones and shaly and arkose sandstones forming the basal
strata of the sedimentary series. Murray (Geological Survey of Canada,
1852-3) regarded the limestones as of Black River age and the sand-
stones as of Potsdam. The apparent conformity of the two led later
workers to suggest that the sandstones were a local development of
the Black River. Ells finds abundant proofs of the Black River age
of the limestones. As to the sandstones, though no conclusive proofs
were found, he inclines to the view of Murray. He gives brief notes
on the economic products of the area. Of these the more important
are iron, mica, feldspar, marl, and building-stone. The Survey has
planned the careful petrographical study of the rocks of the range of
volcanic hills which crosses the St. Lawrence valley in the vicinity of
Montreal. O.E. Le Roy has completed the work on Rigaud Mountain
and is at work on the rocks of Beloeil. Dr. Adams will soon publish
a description of Mount Johnson, and he and Professor Harrington are
at work on the petrography of Mount Royal. J. A. Dresser finds
Shefford Mountain to be composed of essexite, nordmarkite, and
pulaskite. Work on Anticosti island confirms the general features of

REVIEWS Q17

the report of Richardson in 1856, and the account in the Geology of
Canada, 1863. Abundant evidence shows that the island is gradually
rising. Professor Bailey continued his work on the Paleozoic rocks
of southern New Brunswick. Some areas west of St. John river,
formerly mapped as Ordovician, on lithological grounds, are found to
contain a rather meager Silurian fauna. Similar areas east of the
river are probably of the same age. A systematic examination of the
Carboniferous strata of New Brunswick, with a view to the discovery of
workable coal has not afforded satisfactory results. Some reconnais-
sance work was done in Prince Edward Island by L. W. Watson. H.
Fletcher finds that the sedimentary rocks about the Basin of Mines
include representatives of all the Paleozoic groups and the Triassic and
the Pleistocene. Igneous rocks are abundant, and include the Triassic
traps and the intrusive masses of gray granite and gray diorite. The
examination of. the outcrops of the Horton series has failed to afford
data by which its true position could be determined. It is evident,
however, that it is below strata corresponding to the Keokuk-St. Louis.
But some would place it near the base of the Devonian. E. R.
Fairbault worked on the structural geology of the Cambrian gold-
bearing slates and quartzites. The auriferous quartz veins in the area
examined follow the contact of the bedded slates and quartzites. In
places the folding has caused the development of saddle reefs, but the
larger number of veins are on the limbs of the anticlines. G. F.
Matthew presents. in tabular form the results of his study of the
Cambrian rocks of Cape Breton. He shows the completeness of the
Cambrian record of the maritime provinces and correlates the forma-
tions with those of Great Britain. Extended notes are added on the
particulars of the faunas. Dr. Hoffman reports briefly on the chemi-
cal and mineralogical work of the Survey. The substantial progress of
the mineral industries is shown in the report of E. D. Ingall. A very
valuable feature of the report is the index maps of the various
provinces showing the areas covered by the various maps published
and in preparation, and giving dates of the reports in which the areas
are described. J. F. Whiteaves reports on the work in Paleontology.

Some of the individual reports seem to indicate lack of experience
both in the prosecution of the field work and in the preparation of the
report. The smallness of the parties and the large areas covered by
some of them, make it impossible to do more than hasty reconnais-

sance work.
R. D. GEORGE.

g18 REVIEWS

Bibliography and Catalogue of the Fossil Vertebrata of North America.
By OLiverR Perry Hay. Washington: Bulletin of the
United States Geological Survey, No. 179, 1902. Pp. 868.

VERTEBRATE palzontologists, not only of America, but of the world,
are under obligations to Dr. Hay for his very useful bibliography and
catalogue of North American fossil vertebrates. Only he who has
attempted something of the sort, or who has toiled for many hours
trying to find what Cope has written upon some given subject —and
there are few subjects in vertebrate paleontology that he has not
written about—will appreciate, not only the vast amount of pains-
taking labor that has been involved in the production of the work, but
also its value as atime saver. The work is as complete and accurate
as any one could expect it to be. Very few American papers are
omitted, so far as the writer can discover, and he has discovered but
few wrong references, not as many as might be expected from the mere
mechanical execution of the book and its proofreading. ‘The work is
an indispensable necessity for every student of vertebrate paleontology,
for which he can not be too grateful ; and it is much more than the
title indicates.

There are some, however, who will not wholly agree with the author
in the many changes he has made in nomenclature—changes that are
not always consistent, but which, while detracting somewhat from the
usefulness of the work as a guide, do not affect it as a tool. Such long-
established names as /chthyosaurus, Pterodactylus, Mastodon, Dicotyles,
Oreodon, Lacertilia, Ophidia, etc., have been ruthlessly decapitated on
the score of priority, which sometimes, as in the case of Prerodactylus,
is a little strained. The older French writers persistently transformed
technical zodlogical terms into the vernacular, and doubtless Cuvier
intended Prerodactylus as the name of the flying reptiles. Laszlosaurus,
though expressing a falsehood, and contrary to the best canons of
nomenclature, he accepts in place of the well-known Zeuglodon. He
refuses to accept Dinocerata in place of the wrongly-formed Dznocerea,
though the latter term was rejected by its author, and yet changes
Toxochelyde into Toxochelyidae. He accepts Detnodon as distinct from
Dinodon, but rejects Detnosauria in favor of Dinosauria. As a purist
and priorist he goes too far, and becomes involved in difficulties, as is
always sure to be the case. Surely it is allowable to correct manifestly
wrongly spelled or wrongly Latinized names. But these opinions of
the author one need not follow unless he chooses, and he has done a
service in calling attention to these trivial but annoying matters of

REVIEWS gIg

dispute. One does not always see how the author has been guided in
the selection of papers by foreign writers. From the list of over one
hundred titles of papers published by Smith Woodward prior to 1900,
sixty-nine are recorded. We can only be thankful that so many are
given.

The geological survey is to be congratulated upon the publication
of the work ; but we cannot help wishing that the printer had left the

leaves untrimmed.
S. W. W.

Evolution of the Northern Part of the Lowlands of Southeastern
Missouri. By Proressor C. F. Marsur. The University of
Missouri Studies, Vol. 1, No. 3 (Columbia), July, 1902. viii
+ 63 pages; plates I-VII.

THE paper is of more than usual interest to physiographers in that
it presents in a comprehensive way the history of an extremely inter-
esting locality. It is divided into Parts I and II, the former treating
of the geography and geology of the region, and the latter of its
physiographic development.

The writer abandons his former ideas of the origin of Crowley’s
Ridge,’ and agrees, in the main, with the views of Dr. Branner, pub-
lished several years ago, while state geologist of Arkansas.’ It is
shown that the lowland north and west. of Crowley’s and Benton
Ridges, which is spoken of in the paper as the Advance lowland, was
formed by the Mississippi at a time when it turned westward at the
present site of Cape Girardeau and flowed past Delta, Poplar Bluff,
and Neelysville, Mo., and Pocahontas, Powhattan, and Newport, Ark.
While the Mississippi was forming the Advance lowland, the Ohio was
eroding the broad valley between the eastern edge of Crowley’s Ridge
and the uplands of western Tennessee, and which the author calls the
Cairo lowland.

While the Mississippi is the larger of the two streams, it has twice
been captured by the Ohio. ‘The capture of the larger stream by the
smaller was made possible by the latter having the lower flood plain.
The first capture was effected by a small tributary of the Ohio, working
its way headward, through what was then a continuous ridge separating
the two great rivers, along or near the present course of Little River;
the second, by another small tributary working its way northward from

' Proc. Bos. Soc. Nat. Hist., Vol. XX VI (1895), pp. 479-88.

2 Ann. Rep. Geol. Sur. Ark., Vol. IL (1889), preface, p. xiv.

920 REVIEWS

the present site of Commerce to where Gray’s Point now stands. Asa
result of this, the Mississippi abandoned the Advance lowland at its
eastern end and assumed its present course.

Other cases of capture of smaller streams are described in detail,
which make interesting reading for the student of physiography.

The time of the lowland formation was the interval between the
first and second glacial epochs.

Altogether it is an admirable piece of work. ‘There are two points
of courtesy, however, upon which the paper is open to some criticism :
The discrediting of the published views of Dr. Branner upon the sub-
ject of the paper, and the underestimating of the maps of the region,
published by the Mississippi River Commission. When in 1895 the
author advocated a different theory regarding this drainage,’ Branner’s
views were duly considered and promptly upset; now that Branner is
found to be correct, his theory of the origin of Crowley’s Ridge is
spoken of as being “ merely a statement of the popular view.” The
true theory of the ridge’s history may be the popular one in Missouri,
but the present writer has reason to believe that it is not the popular
one in Arkansas even today. Again, Professor Marbut says that Dr.
Branner’s statements “‘lay no claim to scientific completeness.” This
is true; but it should be noted that Dr. Branner published his ideas,
not in the body of the report on Crowley’s Ridge, but in the preface
to the report. The report itself was written by Professor Call, an
assistant on the Arkansas Geological Survey, with whose ideas as to
the origin of the ridge Dr. Branner did not agree ; and not agreeing,
justice to himself required him to say so, but at the same time good
taste and expediency forbade a full discussion of the matter at that
time and place. The present writer happens to know that the data
collected by the late Arkansas survey on the history of Crowley’s Ridge
would fill a good-sized volume. While Dr. Branner’s treatment of the
subject made no pretense of being exhaustive, his statements were
based upon a large amount of data at his command. The author like-
wise says that when he undertook the work on the region “no sort of
topographic map of this part of the lowlands was in existence.”’ The
map published with the Crowley’s Ridge report of the Arkansas Geo-
logical Survey, and the map of the Mississippi River Commission from
which the Arkansas map was largely compiled, so far as their por-
trayal of the physiographic history of the region goes, are in all essen-

TLOGs GUL.

LS)
i

REVIEWS G

tials relating to the questions discussed, like that published by
Professor Marbut, Plate VI of his paper.

Again, the present writer wishes to express his appreciation of
Professor Marbut’s paper as an interesting and detailed account of
physiographic changes, the main features of which were already

known.
AS He Pur DUE.
UNIVERSITY OF ARKANSAS,
Fayetteville, Ark.

Ensayo de una bibliografia historica t jeografica de Chile. Por Nico-
LAS ANRIQUE R.1I L. IGNACIO SILvA A. Santiago de Chile.
LQO2.) 1 SVO, Xix---) O70) pages.

THis work is of the first importance to all students of the history,
geography, and geology of Chile. It contains 2,561 titles, to many of
which are added brief but valuable annotations. The bulk of the
works listed are in the Spanish language, but there are many in Ger-
man, French, and English. The first 996 titles relate to the history
of Chile ; the remaining 1,565 relate to its geography, including topog-
raphy, hydrography, seismology, meteorology, travels, geology, pale-
ontology, and mineralogy. The introduction to the second part con-
tains a sketch of the physical geography of Chile. The author observes
that the number of Chilean volcanoes has been greatly exaggerated.
A list of them is given, with their latitudes, elevations, and dates of
last eruptions. This list mentions forty volcanoes, for several of which
no eruptions have been reported. The second part of this introduc-
tion devotes eleven pages to the meteorology and climate, under which
are included earthquakes, the most important of which are listed. The
third part of the introduction treats briefly of ethnographie geogra-
phy. In spite of numerous oversights and omissions this is one of
the most valuable publications made of late years in Chile and it is to
be hoped that it will be turned to abundant account by our students

of both political and natural history.

J. C. BRANNER.
STANFORD UNIVERSITY,

California, November 4, 1902.

RECENT PUBLICATIONS.

—ANDERSON, TEMPEST, AND FLETT, JOHN S. Preliminary Report on the
Recent Eruption of the Soufriére in St. Vincent, and of a visit to Mont
Pelée, in Martinique. [From the Proc. of the Royal Society, WO ILSOXs
London. |

—BLAKE, W. P. Diatom-Earth in Arizona. [Trans. of the Amer. Inst. of
Min. Eng. |

—BAKER, FRANK COLLINS. The Mollusca of the Chicago Area. The
Gastropoda. [Chicago Acad, Sci., Bull. No. 3, Part II of Nat. Hist.
Survey. |

—Bureau of Mines, Report of the. [Printed and published by L. K. Cam-
eron, Toronto, Canada, 1902. ]

—BRANNER, J. C., AND NEwsoM, J. F. The Phosphate Rocks of Arkansas.
[Arkansas Agricultural Experiment Station, Bull. No. 74. Fayetteville,
Ark. ]

—Crook, ALJA ROBINSON. The Mineralogy of the Chicago Area. [Chicago
Acad. Sci., Bull. No. 5 of Nat. Hist. Survey, July, 1902.]

—CLARK, HERBERT LYMAN. Papers from the Hopkins-Stanford Galapagos
Expedition 1898-9. XII: Echinodermata. [Proc. Wash. Acad. Sci.,
Vol. IV, pp. 521-31, September, 1902. Washington, D. C.]

—CALKINS, FRANK C. A Contribution to the Petrography of the John Day
Basin. [Univ. of Cal. Pub., Bull. Dept. Geol., Vol. III, No. 5, pp. 109-
172, Pl: a7. Berkeley, Cals he Univ, Press.

—CARTER, Oscar C.S. Death Valley. [Reprinted from the Journal of the
Franklin Institute, September, 1902.]

—Davis, W. M. River Terraces in New England. [Bull. Mus. of Comp.
Zool., Vols XX XVIII, Geol. Ser., Vol. V, No: 7.) Harvard College;
Cambridge, Mass. ]

—ECcKEL, Epwin C. Iron and Steel Slags. [Extract from the Mineral
Resources of the U. S. 1901. Dept. of the Interior, U. S. Geol. Surv.,
Washington, D. C., 1902.]

_—EasTMAN, C.R. On the Genus Peripristis, S¢. John. [Extract from the
Geol. Mag., N. S., Decade IV, Vol. IX, pp. 388-91, September, 1902,
London: Dulau & Co., 37 Soho Sq. W. |]

__FROSTERUS, BENJAMIN. Bergyggnaden i sydéstra Finland. [Bulletin de
la Commission géologique de Finlande, No. 13. Helsingfors, 1902. |

922

RECENT PUBLICATIONS 923

—GEIKIE, SIR ARCHIBALD. The Volcanic Necks of East Fife. [Extract from
the Memoir of the Geologic Survey on “The Geology of Eastern Fife.’
Glasgow, Scotland: James Hedderwick & Sons.]

—Geological Commission of Cape of Good Hope, Annual Report of the
[Cape Town, S. Africa: W. A. Richards & Sons, Government Printers. |

—GLANGEAUD, M., PH. The Volcano of Gravenoire. The Chain of Puys.
The Massif of Mt. Dore. [Reprinted from Proc. Geol. Assoc., Part
VI, Vol. XVII, 1902.]

—Geological Survey of Michigan for the Year tgo1. Report of the. ALFRED
C. LANE, State Geologist. [Lansing, Mich.: Wynkoop, Hallenbeck,
Cranford Co., State Printers. |

—Geological Survey of Georgia. W. S. YEATES, State Geologist. Roads
and Road-Building Materials of Georgia. Bulletin No. 8. [Atlanta,
Ga.: Geo. W. Harrison, State Printer. |

—GREENLY, E. Jaspers of Southeastern Anglesey. [Quart. Jour. Geol.
Soc., Vol. LVIII, pp. 425-40, Pls. XV and XVI, 1902.]

—Hovey, E.O. Martinique and St. Vincent; a Preliminary Report upon
the Eruption of 1902. [Extract from Bulletin of the Amer. Mus, Nat.
Hist., Vol. XVI, Art. XXVI, pp. 333-72. Pls. XXXIII-LI. New York,
1902. |

—HawortTH, Erasmus. Mineral Resources of Kansas, 1900 and IgOl.
[The Univ. Geol. Surv. of Kansas. Dept. of Physics, Geology, and
Mineralogy, Lawrence, Kan. |

—KROEBER, ALFRED L. The Arapahoe. The Mrs. Morris K. Jesup Expedi-
tion. [Bull. of the Amer. Mus., Nat. Hist., Vol. XVIII, Part I, pp. 1-
150. New York, September 3, 1902. |

—KEYES, CHARLES R. Origine éolienne du Loess. [Extrait du Bulletin de
la Société belge de Géologie, de Paléontologie et d’Hydrologie. Tome
XII, 1898. Bruxelles, 1got.]

—LEonarD, A.G. Geology of Wapello County. [From Iowa Geol. Surv.,
Vol. XII, Ann. Rept. 1901, pp. 439-99. Des Moines, Ia.: B. Murphy,
State Printer. ]

—LIVINGSTON, BURTON EDWARD. The Distribution of the Plant Societies
of Kent County, Michigan. [Board of Geol. Surv. Ann. Rept. for 1go1.]

—Museum of Comparative Zoélogy, Memoirs of. Reports on the Scientific
Results of the Expedition to the Tropical Pacific, in charge of Alexander
Agassiz, by the U. S. Fish Commission steamer ‘Albatross,”’ from
August, 1899, to March, 1g00._ Preliminary Report and List of Stations
by Alexander Agassiz, with Remarks on the Deep-Sea Deposits by Sir
JoHN MurRrAy (with 21 charts), [Cambridge, Mass.: Harvard College,
January, 1902.]

9024 RECENT POBLTECA TIONS

—Mareut, C. F. The Evolution of the Northern Part of the Lowlands of
Southeastern Missouri. [Pub.by the Univ. of Mo., Vol. I, No. 3, 1902.]

—OSBORN, HENRY FAIRFIELD AND LAMBE, LAWRENCE M. Contributions
to Canadian Paleontology. Vol. III, Part II. On Vertebrates of the
Mid-Cretaceous of the Northwest Territory. [Geol. Surv. of Canada.
Ottawa: Government Printing Bureau, 1902.]

—Pixtspry, H. A., AND VANATTA, E.G. Papers from the Hopkins-Stanford
Galapagos Expedition, XIII. Marine Mollusca. [Proc. Wash. Acad.
Sci., Vol. IV, pp. 549-60: Washington, D. C., September, 1902. |

—PrOSSER,C.S. The Specimen of Nematopyton in the New York State
Museum. [From the American Geologist, June, 1902. ]

—PENFIELD, S.L. On the Use of the Stereographic Projector for Geograph-
ical Maps and Sailing Charts. [From the Amer jours Sci) Vole schin;
May, 1902.]

—ReuscuH, Hans. Vore Dale og Fjelde. Hvorledes Formen af Norges
Overflade er dannet. Med 37 Figurer i Teksten. [Separataftryk af
“Naturen,’’ No. 1-5, 1902. Bergen: John Griegs Bogtrykkeri. |

—Ru1eESs, HEINRICH. Occurrence of Glass-pot Clays in the United States,
| Extract from Mineral Resources of the.U.S., 1901, Dept. of the Interior,
U.S. Geol. Surv., Washington, D. C., 1902.|

—RamsAy, WILHELM AND BorGstTROm, L. H. Der Meteorit von Bjurbéle
bei Borga. [Bull. de la Commission géologique de Finlande, No. 12.
Helsingfors, 1902. ]

—University of Texas Mineral Survey. The Terlingua Quicksilver Deposits,
Brewster County. Bull. No. 4. [University of Texas, Bull. No. 15,
Austin, Tex., 1902. ]

—WILDER, FRANK A. Geology of Webster County. [From Iowa Geol.
Survey, Vol. XII, Ann. Rept., 1901, pp. 63-233. Des Moines, Ia.]

UNDEX TO VOLUME XX.

Adephagous and Clavicorn Coleoptera, from the Tertiary Deposits of Floris-

sant, Colo.,

Atlin District, Briti

Geologica

Analcite-Bearing Camptonite from New Mexico, An.

etc. S. H. Scudder. Review by S. W.
Adirondacks and Champlain Valley, Glacial Phenomena in the.

sh Columbia, Glaciation in the. J.C. Gwillim - -
American Museum of Natural History, the Paleontological Collections of the
1 Department of. Edmund O. Hovey - -

I. H. Ogilvie

Anderson, F. M. The Physiographic Features of the Klamath Mountains

Anticline, The Mi

Georgia,

Relationship of the Corniferous Fauna in the.

Arapahoe Glacier i

Arnold, Delos and Ralph Arnold.

snamed Indiana.

George B. Richardson - -
Aplite, Pegmatite and Tourmaline Bunches in the Stone Mountain Granite of

On the Occurrence of. Thomas L. Watson
Appalachian Province of North America, The Composition, Origin and

n1g0o2, The. N.M. Fenneman

Stratigraphy of the Coast of Southern California : -

Atmosphere, The Carbonic Anhydride of the.

Review by Thomas L. Watson

Bain, H. Foster. Individuals of Stratigraphic Classification :

Review: Genesis of Ore Deposits -
Barker, William B., Edward H. Nutter and. On Some Glaucopiane and
Associated Schists in the Coast Ranges of California -
Baselevel, Grade and Peneplain. W.M. Davis -
Bibliography and Catalogue of the Fossil Vertebrata of North Puente: e

Oliver Pa

rry Hay.

Bull U. S. G. S. No. 79.

Stuart Weller

Discussion

I. H. Ogilvie

The Marine Pliocene and Pleistocene

E. A. Letts and R. F. Blake.

Review by S. W. W.

Black Hills and Adjoining Regions in South Dakota and Wyoming, Geol-
N. H.

ogy and Water Resources of the Southern Half of the.

Darton.

Blackwelder, Elliot.

Review by E. B. - : -

Review: Geology and Water Resources of the Southern

Half of the Black Hills and Adjoining Regions in South Dakota and

Wyoming.
Blake, E. A. Letts

Boston Mountain Physiography. Oscar H. Beery

N. H. Darton - : >

and R.F. The Carbonic pes of the Rameteher
Review by Thomas L. Watson -

Bownocker, J. A. The Oil- and Gas-Producing Rocks of Ohio -

Branner, J. C. Review.

Chile
Briggsville, Mass.,
British Columbia,

The Landslides of Mt. Greylock and.

Glaciation in the Atlin District.
925

J. C. Gwillim

H. F. Cleland

Ensayo de una bibliografia historica i jeografica de

PAGE

220

397
182

252
500
144
700

186

423
839

117
318

139
434

738
77

918
325

325

318
160
822

g21
513
182

926 INDEX TO VOLOME X

Brogger, W. C. Om de senglaciale og postglaciale Nivaforandringer i
Kristianiafeltet (Molluskfauna). Review by R. D. S. Y = -
Brower, J. V. Kakabikansing. Review by T.C.C. - - - - -

California, Neocene Deposits of the Klamath Region. Oscar H. Hershey
California, On Some Glaucophane and Associated Schists in the Coast Range
of. Edward H. Nutter and William B. Barber - - - -
California, The Marine Pliocene and Pleistocene Stratigraphy of the Coast
of Southern. Delos Arnold and Ralph Arnold - - - -
Calvin, Samuel]. Note on the Human Relics of Lansing, Kansas - - -
Camptonite from New Mexico, An Analcite-Bearing. I. H. Ogilvie -
Canada, The Summary Report of the Geological Survey, pects of. For
the year 1901. Review by R. D. George - e
Carbonic Anhydride of the Atmosphere, The. E. A. Letts and R. F. Blake.
Review by Thomas L. Watson - - - - - - -
Carboniferous Fish Fauna of Mazon Creek, Illinois. C. R. Eastman -
Carboniferous of the Sangre de Cristo Range, Colorado, Note on the. Wil-

liste leee - - - - - - - - - -

Cartographic Representation of Geological Formations. Charles R Keyes -
Case, E. C., Paleontological Notes - - - - - - - -
Cement Industry, The. A Reprint from the Engineering Journal. Review by
F, A. Wilder - i = - = - - . - - -
Chamberlin, T. C., Editorials: Antiquity of Man in America - - -
Recent Development of Petrography — - - - - - - -
The Geologic Relations-of the Human Relics of Ee Kansas -
Review: Kakabikausing. J. V. Brower. - 2 - - -
Champlain Valley, Glacial Phenomena in the Adirondacks and. I. H. Ogilvie
Chemico-Mineralogical Classification and Nomenclature of Igneous Rocks.
Whitman Cross, ues P. Iddings, Louis V. Pirsson, Bate Ss.
Washington - - - - - - - - -
Chile, Ensayo de una bibliografia historica i jeografica de. Review by Jn€-
Branner - - - - - - - - - - - -
Classification and Nomenclature of Igneous Rocks, A Quantitative Chemico-
Mineralogical. Whitman Cross, Joseph P. Iddings, Louis V. Pirs-
son, Henry S. Washington - - - - - - - -
Classification of the Upper Paleozoic Formations of Kansas, Revised. Charles
S. Prosser . : = - - - - - = - -
Clavicorn Coleoptera from the Tertiary Deposits of Florissant, Colo., Adepha-
gous and S. H. Scudder. Review by S.W. - - - = -
Cleland, H. F. The Landslides of Mt. Greylock and Briggsville, Mass. -
Coast of Southern California, The Marine Pliocene and Pleistocene Stratig-
raphy of the. Delos Arnold and Ralph Arnold — - - - -
Coast Ranges of California, On Some Glaucophane and Associated Schists.
Edward H. Nutter and William Bb, Barber - - - - -
Coleoptera from the Tertiary Deposits of Floressant, Colo., Adephagous and
Clavicorn. S.H. Scudder. Review by S. W. - - - -
Collections of the Geological Department of the American Museum of Natural
History, The Paleontological. Edmund O. Hovey - - - -

PAGE

323
794

377
738
100 7/

777
500

913

318
535
393
691
256
219
793
433
745

794
397

555

g21

555
703

220
513

117
738
220

252

INDEX TO VOLUME X 927

Colorado and Northern New Mexico, The Morrison Shales of Southern. ie
Walliseieers = - - - - - - - - - - 36
Colorado Meteorite, The Franceville, E] Paso County. H.L. Preston - - 852
Colorado. Note on the Carboniferous of the Sangre de Cristo Range. Wil-
lisimlceleee - - - - - - - - - - 5. 3H0)33
Composition, Origin and Relationships of the Corniferous Fauna in the Appa-
lachian Province of North America, The. Stuart Weller - - = e232
Conglomerates, Etching of Quartz in the Interior of. M. L. Fuller - - 815
Cora, Crotalocrinus, Hall. Stuart Weller - - - - : : = 85332
Corniferous Fauna in the Appalachian Province of North America, The Com-

position, Origin and Relationships of the. Stuart Weller - - = 423
Cretaceous Pterodactyl, On the Skull of aon ae An Upper. S. W.

Williston - - - - 2 - - - - - Ee 520)
Cross, Whitman, Geologic Formations verszs Titologié Individuals - = 222

The Meche of Systematic Petrography in the Nineteenth Cen-

tury - = = © = 2 = = = 331, 451
Cross, Whitman, jocephy P. Iddings, ‘Louis V. Pirsson, Henry S. Warren

A Quantitative Chemico-Mineralogical Classification and Nomen-

clature of Igneous Rocks - - - - : - - - - 555
Crotalocrinus Cora, Hall. Stuart Weller - - - - - - - 532
Crystalline Schists, The Mapping of the. William Herbert Hobbs.

Ranta - - - - - - - - - - - - 780

Part II - - - - - - - - - - . - 858
Davis, W. M., Baselevel, Grade, and Peneplain- - - - - 77
Development of Systematic Petrography in the Nineteenth Gnas The.

Whitman Cross - - - - - - - - - 331, 451
Development of the Profile of Equilibrium of the Subaqueous Shore Terrace.

N. M. Fenneman - - - - - - : - - - I
Diabase from the Trias of Massachusetts, pooper A Purely Feldspathic.

B. K. Emerson - - - - - - - - 508
Eastman, C. R. The Carboniferous Fish- Fauna of Mazon Creek, Illinois - 535
Eckel, Edwin C., Summaries of the Literature of Structural Materials - 442, 542

The Preparation of a Geologic Map - - - - - - - 59
EDITORIALS:
T. C. Chamberlin. Antiquity of Man in America - - - - =) 7.98
Recent Development of Petrography - - - - - - 433
C. R. Van Hise. Upon ‘The Nomenclature of the Lake Supe For-
mations,’ by A. B. Willmott — - - - - - eel
E] Paso County, Colorado, Meteorite, The Franceville. H.L. Preston - eee O 52
Emerson, B. K. Holyokeite, A Purely aS ewe Diabase from the Trias of

Massachusetts - - - - - - - - 508
Ensayo de una bibliografia historica i jeografica de Chile. Review oF Jel.

Branner - - - - - - - - - =2 O21
Etching of Quartz in the Interior of Conglomerates. M. L. Fuller - - == 815

Explosion Near Waldron, Indiana, A Natural Gas. J. F. Newsom - OOS

928 INDEX TO VOLUME X

Fenneman, N. M. Development of the Profile of Equilibrium of the Subaque-
ous Shore Terrace - - - - - - . - - -
The Arapahoe Glacier in 1902 - - - - - - - -
Fish-Fauna of Mazon Creek, Illinois, The Carboniferous. C. R. Eastman
Florissant, Col., Adaphagous and Clavicorn Coleoptera from the Tertiary
Deposits of. S.H. Scudder. Review by S. W. - - - -
Fossil Vertebrata of North America, Bibliography and Catalogue of the. Oli-
ver Parry Hay. Bull. U.S. G.S. No. 179. Review by S. W. W.
Franceville (E] Paso County, Colorado) Meteorite. H.L. Preston - -
Fuller, M.L. Etching of Quartz in the Interior of Conglomerates - - -

Gas Explosion Near Waldron, Indiana, A Natural. J. F. Newsom - -
Gas-Producing Rocks of Ohio, The Oil- and. J. A.Bownocker — - - :
Genesis of Ore Deposits. Review by H. Foster Bain - - . Siaodtane
Geological Department of the American Museum of Natural History. The
Paleontological Collections of the. Edmund O. Hovey - - -
Geological Formations, Cartographic Representation of. Charles R. Keyes
Geologic Formations versus Lithologic Individuals. Whitman Cross - -
Geologic Map, The Preparation of a. Edwin C. Eckel - - = - -
Geologic Relations of the Human Relics of Lansing, Kansas, The. T. C.
Chamberlin - - - - - - - - E 2 -
Geology and Water Resources of the Southern Half of the Black Hills and
Adjoining Regions in South Dakota and Wyoming. N. H. Darton.
Review by E. B. - - - - - - - - - -
George, R. D. Review: The Summary Report of the Geological Survey,
Department of Canada. Forthe Year19go1_ - - - - -
Georgia, On the Occurrence of Aplite, Pegmatite, and Tourmaline Bunches in
the Stone Mountain Granite of. Thomas L. Watson - - =
Glacial Phenomena in the Adirondacks and Champlain Valley. I. H. Ogilvie
Glaciation in the Atlin District, British Columbia. J.C. Gwillim - = =
Glacier in 1902, The Arapahoe. N. M. Fenneman - - - - -
Glaciers, The Variations of. VII. Harry Fielding Reid - - - -
Glaucophane and Associated Schists in the Coast Ranges of California, On
Some. Edward H. Nutter and William B. Barber - - - -
Grade and Peneplain, Baselevel. W.M. Davis - - - - - -
Gwillim, J.C. Glaciation in the Altin District, British Columbia - - -

Hall, Crotalocrinus Cora. Stuart Weller - - - - - - :
Hay, Oliver Parry. Bibliography and Catalogue of the Fossil Vertebrata of
North America. Bull. U. S.G.S. No. 179. Review by S. W.W. -
Hershey, Oscar H., Boston Mountain Physiography. - - - - :
Neocene Deposits of the Klamath Region, California. - - - -
Hobbs, William Herbert. Studies for Students: The Mapping of the Crystal-
line Schists. Part I - - - - - - - - -

Part II. - - - - - - - - - - - -
Holyokeite, A Purely Feldspathic Diabase from the Trias of Massachusetts.
D. K. Emerson - - - - - - - - . -

PAGE

325
913

186

397
182

839
313

738

77
182

532

918
160
377

780
858

508

INDEX TO VOLUME X

Hovey, Edmund O. The Paleontological Collections of the Geological Depart-
ment of the American Museum of Natural History - - - -
Human Relics of Lansing, Kansas, The Geological Relations of the. T.C.
Chamberlin — - : - - - - - - : : -

Ice Work in Southern Michigan. eo Nerzel = - -

Iddings, Joseph P., Louis V. om Henry 5S. Washington, Whitman Cross.
A Ouantiative Chemico-Mineralogical Classification and Nomen-
clature of Igneous Rocks - - . - : - - :

Ulinois, The Carboniferous Fish-Fauna of Mazon Creek. C. R. Eastman

Indiana, A Natural Gas Explosion near Waldron. J. F. Newsom - -

Indiana Anticline, The Misnamed. George B. Richardson - -

Individuals of Stratigraphic Classification: Discussion. H. Foster Bain

Interior of Conglomerates, Etching of Quartz inthe. M.L. Fuller - -

Kakabikansing. J. V. Brower. Review by T.C. C. - - - - -
Kansas, Revised Classification of the Upper Paleozoic Formations of. Charles
S. Prosser - - - - - - : . : :
Kansas, The Geologic Relations of the Human Relics of Lansing. T. C.
Chamberlin - - - - - - - - - -
Keyes, Charles R. Cartographic Representation of f Geological Formations
Klamath Mountains, The Physiographic Features of the. F.M. Anderson
Klamath Region, California, Neocene Deposits of the. Oscar H. Hershey -
Knight, Wilbur C., The Laramie Plains, Red Beds, and Their Age -

Lake Superior Formations, Nomenclature of. A. B. Willmott - - -
Landslides of Mt. Greylock and Briggsville, Mass. H. F. Cleland - -
Lansing, Kansas, The Geologic Relations of the Human Relics of. T.C.
Chamberlin — - - - - - - - - - -
Laramie Plains, Red Beds and Their ee The. Wilbur,C. Knight - -
Lee, Willis T., Note on the Carboniferous of the Sangre de Cristo boas
Golsrads - . - - - - - -
The Morrison Shales of Southern Colorado and Northern New Mexico
Letts E, A. and R. F. Blake, The Carbonic Anhydride of the Atmosphere.
Review. Thomas L. Watson - - - - - = - -
Lithologic Individuals, Geologic Formations versus. Whitman Cross - -
Loess With Horizontal Shearing Planes. J. A. Udden - - -
Lowlands of Southeast Missouri. Evolution of the Northern Part of the. C.
F. Marbut. Review by A. H. Purdue - - - = ;

Map, The Preparation of a Geologic. Edwin C. Eckel - - - =
Mapping of the Crystalline Schists, The. William Herbert Hobbs. Part I -
Part II - - - - - - - - : - -
Marbut, C. F., Evolution of the Northern Part of the Lowlands of Southeast
Missourl University of Missouri Studies, Vol. I, No. 3. Review

by A. H. Purdue - - - . - - - - - -

Marine Pliocene and Pleistocene Stratigraphy of the Coast of Southern Cali-
fornia, The. Delos Arnold and Ralph Arnold - - - - -

793

745
691
144
377
413

67
513

745
413

393
30

318
223
245

919

59
780
858
919

117

930 INDEX, TO VOLUME X

Massachusetts, Holyokeite, a Purely Feldspathic Diabase from the Trias of. ee
B. K. Emerson- - - - - - - - - 508
Massachusetts, The Landslides of Mt. Greylock and Briggs H.F.Cleland 513
Mazon Creek, Illinois, The Carboniferous Fish-Fauna of. .R. Eastman - 535
Meteorite, Niagara. H. L. Preston - - - - - - Bile ithe
Meteorite, The Franceville (E] Paso County, Colorado). H. L. Preston - 852
Michigan, Ice Work in Southeastern. W.H. Sherzer - - - 194
Microscope, A New Combination Wedge for Use with the ee oeaeiem
Fred Eugene Wright - - - - - - - - - 33
Mineralogical Classification and Nomenclature of Igneous Rocks, A Quanti-
tative Chemico-. Whitman Cross, ei Pe Sey Louis V. Pirs-
son, Henry S. Washington - - - - - - 555
Misnamed Indiana Anticline, The. George B. Richardson - - - = 7100)
Missouri, Evolution of the Northern Part of the Lowlands of Southeastern.
C. F. Marbut. University of Missouri Studies, Vol. I, No. 3. Review
by A. H. Purdue - - - - - - - - - 919
Morrison Shales of Southern Colorado and Northern New Mie The. Wil-
liSielleee - - - - - - - - - 36
Mt. Greylock and Briggsville, Mass., The Landslides of. H. F. Cleland Shr Gy}
Natural Gas Explosion near Waldron, Indiana, A. J. F. Newsom - - - 803
Neocene Deposits of the Klamath Region, California. Oscar H. Hershey - 377
New Combination Wedge for Use with the Petrographical Microscope. Fred
Eugene Wright - - - - - - - - - - 33
New Mexico, An Analcite-Bearing Camptonite of. I. H. Ogilvie - - - 500
New Mexico, The Morrison Shales of Southern Colorado and Northern. Willis
Wels) fe - - . - - - - - - - - 36
Newsom, J. F. A Natural Gas Explosion near Waldron, Indiana - - - 803
Niagara Meteorite. H. L. Preston : - - - - - - - 518
Nomenclature of Igneous Rocks, A Quantitative Chemico - Minerological
Classification and. Whitman Cross, Joseph P. Iddings, Louis V.
Pirsson, Henry S. Washington - - - - 555
Nomenclature of the Lake Superior Formations. A.B. Willmot - - - 67
North America, The Composition, Origin, and Relationships of the Corniferous
Fauna in the Appalachian Province of. Stuart Weller - - - 423
Note on the Carboniferous of the Sangre de Cristo Range, Colorado. Willis
A wlee en ia - - - - - - - - - - - - 393
Notes, Paleontological. E.C. Case - - - - - - - - 256
Nutter, Edward H. and William B. Barber, On some Glaucophane and Asso-
ciated Schists in the Coast Range of California - - - - 738
Nyctodactylus, An Upper Cretaceous Pterodactyl. S. W. Williston - - 520
Ogilvie, J. H. An Analcite-Bearing Camptonite from New Mexico - - 500
Glacial Phenomena in the Adirondacks and,Champlain Valley - = 30/7)
Ohio, The Oil-and Gas-Producing Rocks of. J. A. Bownocker - - - 822
Ohio, The Sunbury Shale of. Charles S. Prosser - - - - - - 262

Ohio, The Sunbury Shale of, Note on. Charles S. Prosser - - - =e 2S

INDEX TO VOLUME X 931

PAGE
Oil- and Gas-Producing Rocks of Ohio. J. A. Bownocker «= < 2822
Ore Deposits, Genesis of. Review. H. Foster Bain - - - - - 434
Paleontological Collections in the Geological Department of the American

Museum of Natural History, The. Edmund O. Hovey - - - 252
Paleontological Notes. E.C. Case - - - - - - : - 256
Paleozoic Formations of Kansas, Revised Classification of the Upper. Charles

S. Prosser - - . - - - - - . - - 703
Pegmatite and Tourmaline Bunches in the Stone Mountain Granite of Georgia,

On the Occurrence of Aplite. Thomas L. Watson - - : - 186
Peneplain, Baselevel, Grade and. W.M. Davis _ - - 77
Petrographical Microscope, A New Combination Wedge for Use with the.

Fred Eugene Wright - - - - - - 33
Petrography in the Nineteenth Century, The Tpeelonne of Seaiencae

Whitman Cross - - . - - - - - - 331, 451
Physiographic Features of the Klamath Mountains. F.M. Anderson - - 144
Physiography, Boston Mountain. Oscar H. Hershey - - - - 160
Pirsson, Louis V., Henry S. Washington, Whitman Cross, Toone P. Iddings,

A Quantitative Chemico-Mineralogical Classification and Nomen-

clature of Igneous Rocks - - - - - - - - 555
Pjettursson, Helgi, Morzener i den Islandska eat pamaaiat Review by

J. A. Udden - = - - - : : Se Aliite)
Pleistocene Stratigraphy of the Coast of Southern California, The Marine

Pliocene and. Delos Arnold and Ralph Arnold - - - Seely /
Pliocene and Pleistocene Stratigraphy of the Coast of Southern California.

Delos Arnold and Ralph Arnold - : - - - - =a ToL7,
Preparation of a Geologic Map, The. Edwin C. Eckel - - - So SYS)
Preston, H. L., The Franceville (E] Paso County, Colorado), Meteorite - =i O52
Preston, H. itp Niagara Meteorite - - - - - - - Se Gitte!
Profile of Equilibrium of the Subaqueous Shore Terrace, Development of the.

N.M. Fenneman- - - - - - - - - - - I
Prosser, Charles S., Note on the Sunbury Shale - - - - - - 328

The Sunbury Shale of Ohio - - - - : - 262
Pterodactyl, On the Skull of ead An ee Cretaceous. S. W.

Williston - - - - - - - - - - 520
Publications, Recent - - - - - I14, 221, 329, 450, 551, 799, 922
Purdue, A. H. Review: Evolution of the Northern Part of the Lowlands

of Southeastern Missouri. By C. F. Marbut - - - : - gIg
Quantitative Chemico-Mineralogical Classification and Nomenclature of

Igneous Rocks, A. Whitman Cross, Joseph P. Iddings, Louis V.

Pirsson, Henry S. Washington - - - - - - - 555
Quartz in the Interior of Conglomerates, Etching of. M. L. Fuller - - 815
Recent Publications - - - - - 114, 221, 329, 450, 551, 799, 922
Red Beds and Their Age, The Laramie Plains. Wilbur C. Knight - - 413

Reid, Harry Fielding, The Variations of Glaciers - - - - =o TK

932 INDEX TO VOLUME X

Relics of Lansing, Kansas, The Geologic Relations of the Human. T. C.
Chamberlin - 5 - = - - - - - -
Representation of Geologic Formations, Cartographic. Charles R. Keyes -
Reviews: Adephagous and Clavicorn Coleoptera from the Tertiary Deposits
of Florissant, Colorado, etc. S.H. Scudder (S. W.) : : -
Bibliography and Catalogue of the Fossil Vertebrata of North America.
Bull. U.S. G.S. No. 179. Oliver Parry Hay (S. W. W.) : -
Carbonic Anhydride of the Atmosphere, The. E. A. Letts and R. F.
Blake (Thomas L. Watson) - - - = -
Ensayo de una bibliografia historica i jecoeonce de Chile. By Nicolas
Aurique R. I. L. Ignocio Silva A. (J.C. Branner) — - - -
Evolution of the Northern Part of the Lowlands of Southeast Mieeeer
University of Missouri Studies, Vol. I, No. 3. C. F. Marbut (A. H.
Purdue) - - - . - - - - - - - -
Genesis of Ore Deposits. (H. Foster Bain) - - ~ = =
Influence of Country Rock on Mineral Veins. Walter Harvey Weed
(Frank A. Wilder) - - - - - - - - - =
Kakabikansing. J. V. Brower (T.C.C.) - - - - - =
Moreener i den Islandska Palagonitformation. Helgi Pjettursson (J. A.
Udden) - - - - - - - - - - -
Om de Senglaciale og Postglaciale Nivaforandringer I Kristianiafeltet
(Molluskfaunan), W. C. Brogger (R. D. S.) - - - - -

Summaries of the Literature of Structural Materials(Edwin C.Eckel) 442,

The Cement Industry. Engineering Journal. Reprint. (F. A.W.) -

The Summary Report of the Geological Survey, Department of Canada

for the Year 1901. (R. D. George) - - - - - -

Revised Classification of the Upper Paleozoic Formations of Kansas. Charles
S. Prosser. : - - - - - - - -

Salisbury, R. D. Review. Om de Senglaciale og Postglaciale Nivaforand-
ringer I Krestianiafeltet (Molluskfaunan). W. C. Brogger - -
Note on the Human Relics of Lansing, Kansas - - - -
Sangre de Cristo Range, Colorado, Note on the Carboniferous of the. Willis
(ele enime= - - - - - - - - - - -
Schists in the Coast Ranges of California, on Some Glaucophane and Associ-
ated. Edward H. Nutter and William B. Barber’ - - = -

Schists, The Mapping of the Crystalline. William Herbert Hobbs.
Part I - - - - - - - - - - - -
Part II - - - - - - - - - - -. =
Shearing Planes, Loess with Horizontal. J. A. Udden - - - - -
Sherzer, W. H. Ice Work in Southeastern Michigan — - e - > -
Shore Terrace, Development of the Profile of Equilibrium of the Subaqueous.
N. M. Fenneman - - - - - - - - - -
Skull of Nyctodactylus, An Upper Cretaceous Pterodactyl, On the. S. W.
Williston - - - - - - - - - - - -
South Dakota and Wyoming, Geology and Water Resources of the Southern
Half of the Black Hills and Adjoining Regions in. N. H. Darton.
Review (E. B.) - E = : : a E 2 s = is

PAGE

745
691

220
918
318

g21

919

113

325

INDEX TO VOLUME X

Stone Mountain Granite of Georgia, On the Occurrence of Aplite, Pegmatite
and Tourmaline Bunches in the. Thomas L. Watson - -
Stratigraphic Classification: Discussions, Individuals of. H. Foster Bain
Structural Material, Summaries of the Literature of. Edwin C. Eckel - 442,
STUDIES FOR STUDENTS: Baselevel Grade and Peneplain. W. M. Davis
The Mapping of the Crystalline Schists. William Herbert Hobbs.
PartI - - - - = = - - = - =
Part II - - - - - - - - - . . -
Subaqueous Shore Terrace, Development of the Profile of Equilibrium of the.
N. M. Fenneman - = - - - - - - - -
Summaries of the Literature of Structural Materials. Edwin C. Eckel - 442,
Sunbury Shale of Ohio, The. Charles 5. Prosser - - = - - -
Sunbury Shale of Ohio, The, Note on. Charles S. Prosser - - :
Systematic Petrography in the Nineteenth Century, The Devclopment of.
Whitman Cross- - - - - - - - - 22i1,
Terrace, Development of the Profile of Equilibrium of the Subaqueous Shore.
N. M. Fenneman~ - - - : - - - - - -
Tourmaline Bunches in the Stone Mountain Granite of Georgia, On the Occur-
ence of Aplite, Pegmatite and. Thomas L. Watson - - -
Trias of Massachusetts, Holyokeite, A Purely Feldspathic Diabase from the.
B. K,. Emerson - - - - - - - =
Udden, J. A. Loess with Horizontal Shearing Planes” - : - - -
Review: Morzener i den Islandska Palagonitformation. Helgi Pjettursson
Van Hise, C. R. Editorial: On the ‘‘ Nomenclature of the Lake Superior For-
mations,” by A. B. Willmott — - - - - - - .
Variations of Glaciers, The. VII. Harry Fielding Reid - - - -
Vertebrata of North America, Bibliography and Catalogue of the Fossil. Bull.
U.S. G. S.No. 179. Oliver Parry Hay. Review by S. W. W. -
Waldron, Indiana, A Natural Gas Explosion near. J. F. Newsom - - -
Washington, Henry S., Whitman Cross, Joseph P. Iddings, Louis V. Pirsson.
A Quantitative Chemico-Mineralogical Classification and Nomen-
clature of Igneous Rocks - - - - - - - -
Watson, Thomas L. On the Occurrence of Aplite, Pegmatite and Tourmaline
Bunches in the Stone Mountain Granite of Georgia - - - -
Review: The Carbonic se of the Atmosphere. E. A. Letts
and R. F. Blake . - - - - - -
Wedge for Use with the Petrographical Nimes A New Combination. Fred
Eugene Wright - - = : - - - -
Weller, Stuart, Crotalocrinus Cora, Hall - - - - - -
Review : Adephagous and Clavicorn Coleoptera from the Tertiary
Deposits of Florissant, Colorado, etc. S.W. Scudder - - -
The Composition, Origin and Relationships of the Corniferous Fauna
in the Appalachian Province of North America - - - -

245
218

112
313

918

803

934 INDEX TO VOLUME X

Wilder, Frank A. Reviews: Influence of Country Rock on Mineral Veins.
Walter Harvey Weed - - - . - - - - :

The Cement Industry. Engineering Journal. Reprint - - .
Willis, Bailey. Editorial: Reorganization of the United States Geological

Survey - - - 2 : - i : 2 :
Williston, S. W. On the Skull of Nyctodactylus, An Upper Cretaceous
Pterodactyl = - = - - - = : = : :

Review: Bibliography and Caisieees of the Fossil Vertebrata of North
America. Bull. U.S.G.S.No.179. Oliver Parry Hay - - -
Willmott, A. B., The Nomenclature of the Lake Superior Formations - s
Wright, Fred Eugene, A New Combination Wedge for Use with the Petro-
graphical Microscope = - - - - : : > -
Wyoming, Geology and Water Resources of the Southern Half of the Black
Hills and Adjoining Regions in South Dakota and. N. H. Darton.

Review by E. B.~ - - - See - - -

PAGE

113
219

217

~ 520

918
67

33

325

aa

“hy,
ty

‘
ache

Po Re
| TH x Qe : a

|
eo)
> “y
Be

Lk

&,
ag

Se

eile
Gl
7 a
Ho ith
“>

a

aR
2 é,
oe)
“by
; aN
a ~s EN
eee.
i

et i: :
ee #
ey

es EAP BS
eae

ERTS

_ %s
tay eB ey

ro Sh “te

era)
os ee
Eres ay /

Ss

wee

4a -
Shin

aia FN ant
k v
ere, en
APE

se
op,

tail

may

on slp penile
! | I ‘ i

| Ht i i
ey

Rta
arian
Eee:

Mp se
eS
al

Th geet
‘ igh

L NW

es

<
yt.
1)

ey:

ro Baa pay

Pine =
SS NSE

= all

il (i i

a> y o% Z

dil: tf
Teak, Lg

=
iets
Sn AS

Fy S,
‘geil

TE

Ler ‘ Oper :
> Ree d
o> :
fe) oO

Hie

aT a py
if wy” ey a its ale % s
Jd he Six ie:
s, Moy * S¢
> i 4H,
“no, “ft H a 6 ie | ae
ae ee

MITHSONIAN INSTITUTION LIBRARIES

WL ih

=

=

_ 390 8 01366 9

i ma ts TU HeAh
! NA oARAy MAAC ai
i rich
iy i
1

2

Sete

ih is nth rng
if hh ; uk

Ned)
eh Wk

Meni ylosl
ADA
nah d

+f
i
i
Hh
rd

Ss Mi ,

) He

Pat i Ry i Vee
CN Nini i Sei rt AAI
Ahab