HAN) fh
AY
(ie

‘
Ai

i
Ql

i

Hy

|

a
vi)
ALA

i

(i

‘lie

i \

S= ae == z
See See Se
sor == See

Sa eae tee
eee

Bee
= =: r

rae
SS

AHI HY
cc ni
aaa eS

Same
=~

——

Same

ee

LL
ary sees iy oF

wr a

= Say

Pees
ne
my Ly, Ye

YEE SR ot ey ee
A a ‘ai
ee 5, ~ ei
a eS = he 40, ya Ss
Sons

iy, |

io . a

|

Si Ve, ” Se

aes = -
Lone ica

BSF iced
EEA Oa,
zy 72, ©.

adh

rat “gl
%; [he

% or
%,
or

Whe e H
aN come le y
i!

Ar SOT Ly

by ot
ee s

—<—
Ade SS
Be
tee | ir

aS,
iL) (oe

ig 2
i o “a a
- Ky SS
jean
ZEROS. “A
(ase Si ~Y
LN ae Se ?
ah af)
& .
Sh

Sie
“a oS
% oy
"ey :
l

ait
il

4
=i
~

fat
ee)

eZ Tg: it :

“ph LAG
sage =
eo BD ‘ \ % eae
ay eet
cai ns
4, (hee
PS, MY
ts
Sewn oe
My < #
| a A fits gle aM 0 oR:
| > ies Phe ee ay
“y Y (Eas)
ESS)
Eee

Ps

Ss

_ fie
f A
a
i

<
5
=

S

= a i
Bes} fewer j tS
be ] eee) = ty, med!

A,
Se
S

fa
se
oss #350 aes

Sey

Ke
et ua
eo, *

Re
tame

ie \

i 4

i Ce

Vy

i

Re Sy
in sath li y a thy
2 J ps

aD é <

pROEN, seer

eS ane fEERI liber 4
e a.

afpetie
co fan, Lacie Wh
eG abe hd
‘ Sergi Sees ay)
“Seen?

ry iN ag on a é é a. = ee ih |
».M We a

os hi
" &

%% 4 > pty

WO) fa 73 a

i

" Sy

iy a

& a
Agi a

7 a 4

fis
rtrd os
ean
2

st.

ol

fs

EN

a
S
ts
cee

ou

i.
a
patti

(es

Sz, @
gts Mga Ta
Se

mee

a
INGE

at

a, oe
a oe
it :

hs i

:
- 7
7
7
: y ee a
6 - a
- , 4 f a - -
:
oD =
. -
-
: - 7 e -
| te i 7
7 - ‘i 7 : ‘
7 ; -
Pal -
- =
7 a i -

cm

ts

EON

ony

i

Ny,

WOURNAL OF GEOLOGY

JULY ACGUST,: 007

“

ON A POSSIBLE FUNCTION OF DISRUPTIVE APPROACH
IN LEE? FORMATION - OF METEORITES, COMETS,
AND ONE BULAL.*

ACCORDING to a familiar doctrine founded on the researches
of Roche, Maxwell, and others, a small body passing within a
certain distance (the Roche limit) of a larger dense body will be
torn into fragments by differential attraction. In reality, the
doctrine is applicable to the close approach of any two bodies
of sufficient mass and density, but, as this more familiar case of
a small body in close approach to a larger body is the one sup-
posed to be involved in the origin of comets and certain meteor-
ites, it will at first be taken as representative, and the wider
application of the doctrine will be considered later.

The sphere defined by Roche’s limit is computed on the
basis of a liquid body whose cohesion is negligible, and whose
self-gravitation alone is considered. It is obvious, therefore,
that when cohesion is a notable factor, a small body might pass
through the outermost part of this Roche sphere without suffer-
ing disruption, but that, if a nearer approach were made to the
large body, fragmentation might take place. There is, there-
fore, a sphere within the Roche limit—which may be called the

tI am greatly indebted to Dr. F. R. Moulton for suggestions and criticisms, and

for formule for certain auxiliary computations that do not appear in the paper. I
am under obligations to Mr. C. E. Siebenthal for the diagrams and other aid.

?From THE ASTROPHYSICAL JOURNAL, Vol. XIV, No. 1, July 1go1.
Vol. LX, No. 5. 369

omen
oe,

mel TATE ? Te
“ensonian ohe
ett inst; tdy

w.,

™

3

a,

a,

f fonal Museu:

370 T. C. CHAMBERLIN

sphere of disruption—which is applicable to solid bodies as
distinguished from liquid bodies.

The size of this sphere of disruption compared with the
Roche sphere depends, among other things, on the coefficient
of cohesion and the size of the body to be disrupted. The
coefficient of cohesion being the same, the sphere of disruption
is relatively smallest when small bodies are to be disrupted, and
becomes larger as the size of the body increases until it is
sensibly as large as the Roche sphere. To illustrate this con-
cretely, let disruption be supposed to take place along a diamet-
rical section normal to the gravitative pull, dividing the body
into halves. Let the bodies to be disrupted be spherical and
homogeneous. The cohesion to be overcome will then obviously
vary as the areas of the diametrical sections, and these areas
vary as the squares of the radii of the bodies. But the masses
of homogeneous spheres vary as the cubes of their radu, and
the gravitative pull varies as the masses, modified by the differ-
ential tidal pull. It follows that mutual gravitation will more
effectively disrupt large bodies than small ones. The limit at
which the fragmentation of a solid body will take place will
therefore approach more and more closely that of a fluid body
as the size of the solid body becomes larger. For solid bodies
of considerable dimensions, as asteroids, for example, the limit
of disruption approaches sufficiently near Roche’s limit to make
the difference negligible in a general discussion. This will appear
the more evident from the following numerical considerations.

Experimental data as to the tensile strength of rock are very
limited, as the material is rarely used where tensile stresses are
involved, but all the results of experimental tests given in
Johnson’s Material of Construction fall notably below 1000
pounds to the square inch, and this figure may be assumed as a
liberal representative estimate. The weight of representative
rock may be taken as ;4, pound per cubic inch. The tensile
strength of an inch cube is therefore to its weight, at the surface
of the earth, as 10,000 to 1. Using the same data, the tensile
strength of a mile-cube of rock is to its weight as 1 to 6.36,

FORMATION OF METEORITES By

while that of a 100-mile-cube is as I to 636. It will be seen,
therefore, that in a comparatively small body the cohesive
resistance to disruption bears a very small relation to the gravity
of the inass, and that for large bodies it is negligible. For such
bodies, the Roche limit may be taken as appreciably the limit
of the sphere of disruption.

These numerical considerations, however, show that frag-
mentation by differential gravity acting alone will not become
minute in any such case as that of a satellite or asteroid making
a near approach to one of the planets.

But there are additional considerations that influence the
practical result. The outer portion of the earth, and doubtless
that of the satellites, asteroids, and cold planets generally, is
deeply traversed by fissures—oblique and horizontal as well as
vertical—which render it little more than a pavement of dissev-
ered blocks which could be lifted away with little resistance
beyond that of gravity. The relief of pressure upon the less
fissured portion below, which would follow upon the removal] of
the overlying fissured portion, and the sudden exposure of this
under portion to a lower temperature resultant from this removal,
would develop new stresses; and these would doubtless give
rise to additional fissuring and further easy removal, and thus
the process would be extended. It is not improbable that the
sudden rending open of a sphere that is hot within and the
consequent exposure of the highly heated rocks in the interior
to much lower temperatures would result in sufficiently great
differential contraction to minutely disrupt the fragments irre-
spective of differential gravitation. The central portions of a
body sufficiently hot to melt at surface pressures would doubtless
pass immediately into the liquid condition on the removal of the
pressure of the overlying rock, and this passage might, not
unlikely, take on eruptive violence by reason of the included
and highly compressed gases—or substances in a potentially

gaseous state—in which case an extremely minute division
would ensue. In the case of the earth, there is good reason to
believe that if its interior gravitative stresses were suddenly

372 T. C. CHAMBERLIN

removed, its internal elasticity would disrupt its exterior with
much violence; and if the gravitative stresses were more grad-
ually removed, the disruption would still be complete and per-
vasive, though less violent. How far a similar view may be
entertained with reference to small bodies like the asteroids is
uncertain, but even in these it is not improbable that the internal
elastic factors would offset in some large part, if not entirely,
the restraining force of the general cohesion of the mass.

From these considerations it would seem that the sphere of
disruption, even in solid bodies of the nature of satellites and
asteroids, may closely approximate to the theoretical Roche
limit, while, for large bodies intensely compressed and very hot
within, the practical sphere of disruption might actually exceed
the Roche sphere. In the case of large gaseous bodies like
the sun, intensely heated and compressed in the central por-
tions, the disruptive or dispersive sphere must be much larger
than the. Roche. sphere. But of this later. For the smaller
solid bodies, and for present purposes, it may be assumed
that the sphere of disruption is practically defined by Roche’s
limit.

The size of the sphere of disruption compared with the size
of the body producing the disruption is an essential point in this
discussion. The relative magnitude of these varies for every
couplet of bodies brought under consideration, because it is
dependent on density, cohesion, internal elasticity, and other
varying factors. Roche has shown that, if the two bodies are
incompressible fluids of the same density, and without cohesion,
the limit of disruption is 2.44 times the radius of the body
producing the disruption. The cross section of this body will
therefore be to the cross section of the Roche sphere as I is to
5.95. The disk of the outer ring of Saturn, compared with that
of the planet, whose density is unusually low, is a trifle below
this ratio (1:5.29), but may be taken as a practical sanction of
the figure theoretically deduced. The disk of the Earth, a dense
body, is to the disk of the Roche limit, as computed by Darwin,
as I to 7.5. It may therefore be concluded that where planets

FORMATION OF METEORITES 373

and planet-like bodies are concerned, the sphere of disruption
has a cross section from 5 to 7.5 times as great as the central
body It follows from this, that to a passing body the sphere of
disruption exposes a disk five to seven times as great as the
central body, and hence there are from four to six times as many
chances that the passing body will invade the sphere of disrup-
tion without collision, as that it will strike the central body. In
other words, the fragmentation of a small body by near approach to
a large one of the nature of the planets will be from four to six times
as tmminent as actual collision.

That disruptions or explosions of some kind actually take
place in the heavens, and that not uncommonly, seems to be
implied by the sudden appearance of new stars, often with great
brilliancy, followed by rapid decline to obscurity or extinction.’
Five such new stars have been recorded during the last decade,
and the survey of the heavens during this period has not been
entirely exhaustive. The appearance of such new stars has been
referred to collision, but their frequency has been felt to be an
objection to this view, and other explanations, of the nature of
eruptions or explosions, have been offered, but usually without
assigning any probable cause for such extraordinary explosive
action. The numerical objection is, in some measure, removed
if the possibilities of disruptive approach be added to those of
collision; and it will be seen further on that special condi-
tions giving rise to distant approaches that are merely disturb-
ing at the outset, may ultimately give rise to large possibilities
for disruptive approaches.

That bodies pass within the disruptive sphere of other bodies
is known from the fact that at least four comets have been
observed to pass within the Roche limit of the sun, and these
would quite certainly have been torn into fragments if they had
not already been in that condition. There are, therefore, some
observational grounds for the view that instances of bodies pass-
ing through the disruptive spheres of other bodies are not so
rare as to render their results unimportant.

‘A fact which has become very familiar and impressive, since this was written,
by the appearance of ova Ferset.

374 Tl. C. CHAMBERLIN

In the considerations now set forth, there seems to be war-
rant for the proposition that solid bodies may suffer fragmentation
without actual collision with other bodies, and that the bodwes so dis-
rupted may constitute comets so long as the fragments remain clustered,
and that when these fragments become dispersed, they may constitute
one variety of meteorites. Only the first part of the proposition
is novel—if indeed that is—for the disintegration of comets
into meteorites is an accepted doctrine. The characteristics of
comets other than their fragmental structure will need to be con-
sidered, but this may best be taken up later.

The foregoing conclusion, as a purely ideal proposition, does
not appear to need discussion, unless the fundamental deduc-
tions of Roche, Maxwell, and others are questioned. Nor does
its application to the adventitious cases of wandering bodies per-
mit definite discussion, for neither the nature nor the number of
such bodies is known; nor is the likelihood of their close
approach to other bodies capable of estimation. But, on the
probable supposition that the stars are centers of systems like
our sun, there are hypothetical cases of approach of these sys-
tems to each other that by disturbance of the planetary orbits
may lead on to disruptive approach of the individual bodies, and
thus give effective application to the doctrine; and these invite
consideration. It must be confessed that these cases, likewise,
cannot be discussed with much satisfaction, since the movements
of the assumed solar systems and their relations to each other
are but very imperfectly known. Present data, however, war-
rant the assumption that the stars and their attendants are mov-
ing in various directions at various velocities, and that they are
probably not controlled by any central body; nor do they prob-
ably follow concentric orbits so adjusted to each other as to for-
bid close approaches. The conception that the movements of
the stars are somewhat analogous to those of the molecules of
an exceedingly attenuated gas in an open space, actuated by the
attraction of their common but dispersed mass, seems the most
probable that can be entertained in the present state of knowl-
edge. It may at least be made the basis for the assumptions
necessary to further discuss the doctrine in hand.

FORMATION OF METEORITES 375

Let two stars be assumed to be attended by sécondaries like
those of the sun, and to pass each other near enough to initiate
serious disturbances in the orbits of the planets and satellites of
the two systems. It is not necessary that this disturbance shall
be so great as to bring about a disruptive approach of any of
these bodies at once, but merely that this shall be the ulterior
effect, which may be long delayed. The two systems need not
necessarily invade each other’s actual limit, that is, the two
suns need not approach each other within the sum of the radii
of the orbits of their outermost planets.‘ For example, in the
ideal case of two solar systems, it is not necessary that the
orbits of the two Neptunes shall actually cut each other. If the
undisturbed orbits merely touch each other, or even closely
approach each other, it seems clear that if Meptune be at the time
coming toward the point of such ideal contact, or near approach,
the attraction of the passing sun, together with MVeptune’s own
momentum, will carry the planet far beyond the limit of its own
ideal orbit into the sphere of dominant influence of the passing
SUlha witethesame: time, the paths of ‘the immer orbits of ,both
systems will be distorted in a quite irregular way, dependent on
their various positions in their several orbits. The transfer of
an outermost planet from one system to another under these con-
ditions of general disturbance, or any other radical change in
the orbits of the outer planets, will quite certainly lead on to
other disturbances of orbit, some of which may sooner or later
lead to disruptive approach, though the result of such a compli-
cation is beyond the reach of precise prediction.

A still more remote approach between two systems in which
the only result is a pronounced elongation of the orbits of the
two systems, may ultimately result in close approaches, for, if
the orbit of any of the planets of the two systems be elongated
so that its perihelion distance is less than the aphelion distance
of the next inner planet, or its aphelion distance greater than the
petihelion: distance gof: the next outer: planet, a disruptive

‘In the illustrative examples it is assumed for convenience that the planes of the
systems are normal to the systems’ lines of movement.

376 T. C. CHAMBERLIN

approach, although it will not necessarily follow, because the
planes may not coincide, and for other reasons, may result —if
not at once, at least ultimately—as a consequence of the shift-
ings and modifications which such a disturbed condition involves.
For example, it is obvious that by a favorable conjunction with
a passing system whose sun is distant from JVeptune considerably
more than the radius of his orbit, there may be an elongation of
the orbit of Meptune so as to make it cut one or more of the
inner orbits, and that further modifications may arise out of
these relations which will either increase or decrease the eccen-
tricity. The principles applicable here are identical with those
that have been found to produce radical modifications of the
orbits of comets and that have been worked out by H. A.
Newton and others.

To embrace the full possibilities of the case, it is therefore
necessary to consider (1) the effects of systems passing each
other at distances varying from those in which the outermost
planets do not even cut each other’s orbits, down to center-on-
center collisions, and (2) to take account of the ulterior effects
of disturbed orbits, as well as the immediate effects. This last
is a consideration of no small importance in the qualitative as
well as the quantitative application of the doctrine, for it distrib-
utes the effects over an indefinite period of time, and does not
require their coincidence with the passage of the systems. The
ulterior effects, so far as the disruption of secondaries is con-
cerned, may apparently be much greater than the immediate
effects. If this is not already clear, let a specific case be taken,
as, for example, two solar systems passing each other so that
their centers shall be 500,000,000 miles apart at nearest
approach. If the planes of the systems are transverse to their
paths, the ideal undisturbed orbits of the asteroids will touch,
or closely approach, or slightly cut each other, as the individual
case may be. The ideal orbits of the /upzzers will fall but little
short of the passing sun, while the ideal orbits of Saturn, Uranus,
and Neptune will fall outside the passing sun. While the precise
results of such an event cannot be computed, it is quite certain

FORMATION OF METEORITES BY.

that the secondary systems of the two suns will be most pro-
foundly disturbed and the symmetrical and harmonious relations
of the planetary orbits be utterly broken up. While even in this
case the zmmediate contingency of a disruptive approach of one
secondary to another may not be high, there will arise a
perpetuated series of contingencies, the consequences of which will
apparently be immeasurably greater than those immediately inci-
dent to the disturbing action, and the end of this perpetuated
series of contingencies can scarcely be foreseen. Assuming that
the great planets will exercise the same kind of influence over the
small planets and asteroids that pass near them that /apiter does
over comets, the range of possible contingencies involves, on the
one hand, closer and closer approaches and even collisions with
the Sun and with other planets, and, on the other hand, the devel-
opment of extremely elliptical orbits that will carry the small
bodies into the sphere of influence of some other system. How
large a proportion of these theoretical possibilities will be rea-
lized in a given disturbed system, it is impossible to determine,
for the problem is far beyond the power of mathematical analysis,
but it seems at least probable that results of moment may
ensue.

If we may judge from the solar system, the small bodies may
be assumed to be at least fifty times as numerous as the large
ones, while not improbably they are a hundred or several hundred
times as numerous. Other things being equal, they should show
the characteristic effects of the action under discussion with
correspondingly greater frequency. But the other conditions
intensify these effects. A small body may be disrupted by a
large one, but not necessarily the reverse. So, too, a small body
may be thrown into an erratic orbit, while the orbit of the large
body may not be sensibly affected, as shown by the changes in
the orbits of comets caused by Jupiter. By far the most common
effect of the close approach of two star systems should therefore
be the fragmentation of the small bodies by being caused to
pass within the spheres of disruption of the large bodies. As
previously indicated, the contingency of acquiring at the same time

378 T, C. CHAMBERLIN

highly erratic orbits is imminent, and these are specially subject to still
further changes, and thus these fragmental clusters come to possess by
the very circumstances of their birth the second characteristic of comets,
as well as the first.

Whether they would possess at the same time, or come at
length to possess, the ¢izrd characteristic of comets, the attenu-
ated matter of which cometic tails are made, is not so clear,
since the nature of this matter and its condition are not yet fully
known. The recent discoveries relative to the extreme ionization
of matter and perhaps even its corpuscular dissociation, and the
radio-activity of certain kinds of matter are at least very sugges-
tive in this connection. Six of the elements reported by good
authority as detected in meteorites, are known to possess, or to
be habitually associated with, radio-active matter, viz., barium,
bismuth, cerium, lead, titanium, and uranium. It is not very
material here whether this radio-activity is really possessed by
all these elements themselves, or simply by substances associated
with them. If the coma and tails of comets are dependent
on rare substances of a radio-active or extremely volatile nature,
and hence permanently retensible only in the interior of bodies,
it would be difficult to imagine conditions more favorable for
setting them free in unusual volume than minute tidal disruption ;
particularly is this true if the retention of these substances is
dependent on low temperature, as seems to be the case, since
they are brought forth and driven away at a highly accelerated
rate as the sun is approached. This view seems also to be sup-
ported by the fact that comets which remain long in the vicinity
of the sun, as for example the short-period comets, lose their
tails in a brief period. |

If the attenuated cometic matter owes its essential peculiarities
to electric states, these might perhaps be derivable from the
revolutionary movements of the magnetic elements in the frag-
mental swarm, for by the hypothesis of tidal disruption the
swarm should inherit a rotatory movement, and the fragments
should contain both magnetic and magnetizable matter, variously
associated with diamagnetic matter.

FORMATION: OF METEORITES 379

That short-period comets are subject to progressive disinte-
gration, and that their scattered elements constitute one class
of meteorites, is familiar doctrine. There seems no reason for
withholding the conception from comets which have parabolic or
hyperbolic orbits, for in certain cases such comets have shown
signs of disruption and partial dispersion in their perihelion
passages. To the dispersed elements of these comets of high
velocity is assigned (in part at least) such meteorites as come to
earth from diverse directions with velocities incompatible with
an origin within the solar system.

It remains to consider whether the fragments derived from
the disruption of an asteroid, satellite, or small planet through
differential gravitative strain without collision, will satisfy the
characteristics which meteorites display. Ample data fora judg-
ment on this vital point will be found in the articles on the
structure of meteorites in the first two numbers of the Journal of
Geology for the current year, by Dr. Farrington, who, at my
request, has kindly brought together in succinct and systematic
form the essential characteristics of meteoric structure. A study
of these characteristics will show that, while they embrace a
great and very significant variety, they are all referable to the
structures that are appropriate to small planets, while it is diffi-
cult to see how all of these characteristics can be found in deriva-
tives from any of the alternative sources to which meteorites
have been assigned, namely, volcanic action of the moon or of
the planets, explosive projection from the sun, or individual
aggregation inspace. Some of the matter is fragmentary, imply-
ing surface conditions, while some of it is coarsely crystalline,
implying deep-seated conditions. Some is volatile and com-
bustible, implying the absence of high temperature throughout
its whole antecedent history, while some as distinctly implies
the presence of high temperature. In some meteorites the iron
is segregated, while in others it is disseminated. Frequent
brecciated structures imply fracturing and recementation. Faz/t-
ing and slickensides demonstrate movements attributable to the
parent body, but not to the meteorite itself. Veins imply internal

380 IR (Go (Gla CANE) KILI OM

transfer and redeposit of molecules. The absence of the oxida-
tion of the iron accords with internal conditions, and also with
the supposed absence of atmospheres from small planets,
asteroids, and satellites. In short, every feature of the meteor-
ites, save, of course, the external effects of fragmentation and
of heating during their fall through the atmosphere, is assignable
to small planetary bodies riven into fragments without great
heat and, by reason of this, retaining the varied structure attained
in the parent body.

As previously indicated, the disruption of a body like the
earth, the main mass of which has a temperature much above
the melting point of its substances at low pressures, and which
is greatly compressed within by self-gravity, would doubtless
cause it to burst forth into a luminous body with perhaps some
dispersive violence. The progressive stages of distortion which
take origin in simple tidal protuberances and grow to greater
and greater degrees of deformation and crustal fissuring, until
the final stage of disruption is reached, could scarcely fail to
bring some parts of the ocean into contact with some parts of
the heated interior, with inevitable Krakatoan consequences.
Fragments of the crust under these conditions might possibly
give origin to meteorites, but the probabilities of such fragments
being projected beyond the 640,000 miles of the earth’s domi-
nant influence, or beyond the similar spheres of influence of
other massive planets, would not seem to be great; and, if real-
ized, the fragments would doubtless be reduced to dust, as in
the case of the Krakatoan explosion, and this state of minute
division would exclude such meteorites from recognition except
as vanishing shooting stars. The probabilities are that the
matter of a disrupted earth or a similar massive planet would
be again assembled into a planetary body by its strong self-
gravity. The phenomenon would therefore be that of a tempo-
rary star. Assuming considerable dispersion, it might be rather
brilliant for a time, but would rapidly cool as the result of such
dispersion, and soon sink into invisibility. In the case of such
a body as Jupiter, accepting current doctrine as to his nature, the

FORMATION OF METEORITES 381

initial brilliancy must be much greater and the cooling to invisi-
bility much more prolonged.t | To phenomena of this class may
perhaps be tentatively assigned certain of the temporary stars.
Obviously these can only be such as had no prior visibility, and
such as sink sooner or later again into invisibility. Whether
this invisibility were due to the superficial cooling of the nucleus,
or merely to a deep enshrouding of cooled vapor, would be
immaterial.

It has already been indicated that a possible result of the
serious disturbance of one solar system by the near transit of
another would be a fall of some of the disturbed planets upon
one of the suns. This also might be an ulterior rather than
an immediate result, through the’ modifying effects of other
planets, as well as the direct effects of the primary disturb-
ance. Such a fall must be presumed to give rise to a notable
increase of heat in the central body, as well as to mechanical
effects, both of which would be conditioned by the mass and
velocity of the secondary. An outburst of greater or less bril-
liancy must be presumed to be the result. The mechanical
effects upon the sun would probably involve great changes in
temporary density and condition, as well as outshoots of hot

gases in various directions at high velocities. The effects might
— thus coincide with the phenomena of that class of the temporary
stars in which a luminous state precedes and follows the out-
burst, and in which varying densities or high velocities in
opposite directions seem to attend the temporary brilliant
stage.

The disruption of suns has been neglected thus far. While,
under the terms of the hypothesis, the disruption of these
bodies must be rarer events than the fragmentation of the
more numerous small bodies, the results must be correspond-
ingly more important, by reason of their magnitude and char-
acter. |

It has already been observed, in passing, that the internal

*Such a body as /wpzter might perhaps, under proper conditions, be dispersed
in the same manner as a sun as sketched beyond.

382 T. C. CHAMBERLIN

elasticity of large hot bodies under great self-gravitative com-
pression may so far aid in disruption, by coédperation with the
differential gravity of an adjacent body, as to cause dispersion
even before the Roche limit is reached. In the case of very
large bodies that are already gaseous, such as the sun, this phe-
nomenon gives rise to a special case of extreme interest. Under
this special case, there arise a large variety of particular instances
due to the varying sizes, velocities, paths, rotations, and consti-
tutions of the couplets of stars concerned, and also to the adven-
titious effects of their secondaries; but, for a simple illustrative
case, let it be assumed that two bodies of equal masses and
equal velocities are approaching each other on parabolic paths,
and that at periastron they will pass through each other’s spheres
of disruption, or rather, spheres of dispersion. For convenience,
let if be assumed that one of these bodies (A, Fig. 1) is gaseous,
while the other (4) has already become so cold and solid as to
act essentially as a unit, though disrupted. The history of the dis-
persion of the gaseous body may then be followed alone. Let
the rate of rotation of the gaseous body (A) be relatively low,
as in the case of our sun. It may then be neglected in a general
discussion, since as a dynamic factor it is trivial compared with
the enormous energies of momentum and of elastic dispersion
involved. This will appear clear in the outcome. Furthermore,
the direction of rotation with reference to the parabolic paths
might happen to be any one of an indefinite number, in many of
which the effect would be inconsequential, even if the energy
were large. In the close approach of these two bodies the
two great dynamic factors of special interest are (1) the
tidal distortion, and (2) the elastic expansion of the gaseous
body.

While the two bodies are yet distant from each other, they
must begin to take on elongation of the tidal type as the result
of their mutual differential attraction, this elongation being aided
by the high internal mobility and elasticity of the gaseous body.
As the bodies approach periastron, this elongation must progress
at an accelerated rate. At the moment of the entrance of the

FORMATION OF METEORITES 38

QD

gaseous body A upon the Roche sphere of the body 4S, self-
gravitation, in accordance with Roche's doctrine, will have been
completely neutralized on the lines joining the centers of A
and 4, and its restraining influence upon the elasticity of the
gaseous body on these lines will have been removed, while the
gravitative constraint in the transverse section will be increased.
The expansive energy of the compressed gaseous matter will
therefore be left to exert its full projectile force in the direction
of the axis of elongation. While I am unable to offer a numeri-
cal estimate of the magnitude of this expansional energy in such
a body as the sun, it is certainly of a high order of magnitude.
The speed at which prominences are projected from the sun
under present conditions closely approximates to the parabolic
velocity with respect to the sun, and this is accomplished in
spite of gravitation and a resisting atmosphere.

The case in hand, therefore, starts with simple tidal elonga-
tion at a distance, and increases to an explosive maximum as the
bodies approach periastron. This increase is at first gradual, but
in the last stages of approach to periastron the acceleration is
exceedingly rapid. In any attempt to follow the process in
more detail, adhering to the recognized principles of tidal action,
four particulars are of special moment: (1) the progressive elon-
gation of the body, (2) the change in the direction of tidal dis-
tortion, (3) the lag of the line of maximum elongation behind
the line of maximum attraction, and (4) the rotatory effects
arising from the gravitation of 4 on the tidal protuberances of A,
which in this case will be peculiarly effective because of the
enormous distortion of 4 and the very close proximity of 5 at
the critical stages. The principles from which these effects arise
are thoroughly demonstrated and are familiar to all students of
tidal phenomena, and it is only their special applications to this
case that need be discussed. The rotatory effects are a little
peculiar in that both of the tidally acting bodies are rapidly
approaching each other, and developing extraordinarily powerful
differential attractions, while at the same time they are swinging
about their common center of gravity. Near periastron they

384 LAG. (CHAMBERLIN,

may be regarded as performing a semi-revolution about each
other. By the terms of the special case in hand, this semi-revo-
lution must be performed in avery few hours. During these
few hours the gaseous body (A) is undergoing elongation at a
rate not much less than that represented by its full explosive

Fic. 1.— Diagram illustrating the elongating and rotatory effects ofa solid
stellar body, B, upon a gaseous sun, 4, during their mutual approach to periastron.
A 23445 indicate successive positions, changes of form, and rotation of the gaseous
star on its approach to periastron. At *3445 represent the successive positions of
a solid body of equal mass and velocity which is assumed for convenience to remain
intact. Position 41 corresponds to 41, A? to £2, etc. The lines joining their cen-
ters indicate the successive directions of mutual attraction. The arrows: indicate
direction of movement.

competency. The rotational forces are diagrammatically illus-
trated in Fig. 1, in which the lag is merely estimated and the
distortion of A is simplified while that of B is neglected.

FORMATION OF METEORITES 385

It is assumed that the lag of the axis of elongation of J is
uch that the effective path of explosive projection will be
directed to the rear‘of B. -It must be noticed that if the center
of A passes through the outer part of the Roche sphere of 4, the
nearest edge of A, if undistorted, would pass within two or three
hundred thousand miles of 4, and hence that the projective
elongation of dA must pass critically near £; but the relative
speed of the bodies A and & is so great—both being near the
parabolic velocity with respect to the other—that the projected
matter of A can only collide with 4 on the supposition that the
velocity of projection at least equals the parabolic velocity of
the body and acts instantaneously, the last of which is impossi-
ble. This is based on the assumption that the transverse com-
ponent of the attraction of & prevents the elongation of the
minor axis of A, which is true of liquid bodies tidally affected,
but might perhaps break down in a gaseous body under these
extraordinary conditions. The point, however, is not important
here, for if the edge of the projected part of A collide with 5,
it will only intensify the rotatory effects under consideration,
and such collision is contemplated as an essential feature of the
next tollowing case, but-is excluded here as. the-effects of
approach without collision is the special theme under discussion.
The very close approach of the elongated extremity of A to
& obviously gives great effectiveness to the rotatory influence of
&’s attraction upon it. If the amount of this attraction be rep-
resented by the fall of ;,57 part of the mass of A toward Sata
mean distance of 200,000 miles from #’s surface —the masses of
A and & being each equal to that of the sun—such a fall for
about two hours and a half would generate a momentum equal
to the whole revolutionary and rotatory momentum of the pres-
ent solar system. It would appear, therefore, that under the
conditions postulated a rotation of a highly effective kind must
be imparted to the elongated body. It will now appear that the
previous rotatory energy of the sun, which is only about 2 per
cent. that of the solar system, is a negligible factor.
‘Phe history ot: A then takes this form: (1) A very rapid

386 7. C. CHAMBERLIN

elongation in the hour or two preceding its entrance upon the
Roche sphere. (2) After entrance upon the Roche sphere, an
explosive elongation actuated by the elastic energy then remain-
ing in the body unrestrained by self-gravity in the axis of elon-
gation. (A portion of the original elastic energy had been
consumed in the previous elongation and a corresponding amount
of momentum had been acquired, the larger component of which
would be effective along the changed line of elongation.) (3)
After passing out of the Roche sphere, the restraints of gravity
begin again to be felt and rapidly increase as A and JB retire
from each other, but the distance to which the extremities of A
have already been projected, and the new relations thereby
assumed to the remaining mass of A, and to 4, render the
renewed gravitative influence far less effective than the original.
and the projection must continue until the momentum acquired
is overcome. (4) Coincident with this projection a constantly
increasing rotation toward 4 has been generated, which possibly
reached an effectiveness comparable to that of the solar system.
The effects of explosive projection combined with concurrent rotation
must obviously give rise to a spiral form.

It seems clear from the nature of the case that there would
be a certain brief period when the climax of projective effects
would be reached, and that a stream of material of much greater
mass and velocity than at other instants would at this time be
projected from the extremities of the elongated mass in both
directions. There should therefore be two chief arms to the
resulting spiral starting from the opposite points of the central
mass and extending outward to the limits of the spiral—indeed
constituting the most outlying portions of the spiral. These
must be curved in a common direction by the rotation of the
mass. Such predominant arms are notable features in the typi-
cal spiral nebule. They are well shown in Nos. 1, 2, 3, 4, 5, and
6, Plate I, all of which are reproductions from photographs
furnished by the late Professor Keeler.

In the illustrative case that has just been discussed the solid
body 4 was made to represent a convenient possible case but

JouR.“Gron., Vol. TX, No.5 Plate I

5 6
SPIRAL NEBULAE
t. Mes7 Can. Ven.=G. C. 3572-3574 4. H LV 76 Cephet=G. C. 4594
2. M ror Urs. Maj.=G. C. 3770—3771 5. 433 Trianguli=G. C. 352

3. M74 Pisctum=G. C. 372 6. H 753 Pegasi and her nebula

FORMATION OF METEORITES 387

one whose real frequency is quite unknown, since extinct suns
are beyond the reach of observation. If the active lives of suns
are no greater than the periods deduced from computations
founded on the Helmholtz theory of solar heat, extinct suns
should either be numerous, or the whole previous history of the
stellar system must have been short; or else, as a fertium quid,
some effective means of regeneration must be assumed. .

In the more typical case of two live suns coming into
such close relations, its seems probable that mutual dispersion
might follow without serious collision, since the analysis of the
phenomena seems to show that the mutual elongations of the
live suns would develop on essentially parallel lines whose con-
stant shiftings would be mutually consonant, as illustrated in Fig.
2. If no serious contacts were developed, the two resulting
spirals would separate and pursue the paths normal to their
parent stars, with such modifications as may have resulted
from the loss of energy involved in giving rotation to the
nebule.

If, however, the periastron approach is so close that partial
collision ensues, the analysis seems to indicate that the elongated
bodies which would be developed previous to contact would not
collide end to end centrally, but by a lateral shear, as illustrated
in Fig. 3. In this case the arrested momentum combines with
mutual attraction to give a rotatory movement of the highest
order, and the heat and the resilience from impact must combine
to intensify the dispersive competency. The arrest of momentum
may be presumed to go so far in some cases as to cause the two
bodies to unite to form a single spiral nebula of the largest and
most dispersed order, such perhaps as the well-known great
spiral nebule; or the arrest may be partial, and certain parts of
one-or the other, or both,.of the, masses may escape. “In No:
1, Plate I, we seem to have a possible example of this, in which
the escaping, or partially escaping, mass" is still associated with
the longest arm of the spiral.

«This is assumed to have been a dead sun because of the limited evidence of
explosive action.

388 TEAG CLAN BILLION:

In case the collision of two suns becomes essentially central,
a general dispersion of the most violent sort may be inferred to
follow, and this may find exemplification in the vast irregular
nebulz, which are in many cases more or less radiant, and in
some cases consist of two irregular masses which perhaps repre-

Fic. 2.— Diagram illustrating the progressive elongation and rotation of two suns,
C and D, approaching perihelion. The position C* corresponds to Dt, C? to D?, etc.;
the lines joining these indicate the successive directions of mutual gravitation, and the
arrows indicate direction of movement. The progressive elongation, the lag, and the
rotation of the bodies at successive stages are diagrammatically indicated.

sent the wrecked originals. The collision of dead suns in which
disruption shortly preceded actual impact may also play a part
in forming irregular nebule.

Speculation may perhaps go so far as to attribute ring nebule

FORMATION OF METEORITES 389

to the central penetration of a concentrated solid body through
a gaseous mass.

It is as impossible as it is unnecessary to consider here the
infinite variety of sub-cases which the hypothesis under consid-

\

wecce-”

Fic. 3.— Diagram illustrating the same phenomena as Fig. 2, save that the peri-
astron distance is so small that the bodies collide by a shearing stroke.

eration involves, but it seems advisable to note that the case of
equal suns with equal velocities, which has been used in illus-
tration, is not the most prevalent case; for inequality of mass
and momentum is quite certainly the rule, rather than equality
or sub-equality. Where one of the suns is much smaller than
the other, the dispersive influence will be most largely felt by it,

390 l. €. CHAMBERLIN

and so it seems probable that there may be a series of cases in
which the minor members of the couplets are dispersed with
different intensities into complete nebule while the major mem-
bers only suffer varying degrees of eruptive action or partial
conversion into nebule and so perhaps become stars with
nebulous adjuncts or atmospheres. Under this conception small
nebule should be much more numerous than large ones. If
large hot planets, such as Jupiter is supposed to be, are poten-
tially gaseous, and if by disturbing approaches of stellar systems
such planets are thrown out of their allegiance to their primary
suns and take on comet-like courses, they would be specially
liable to disruption and dispersion into small nebulz, and would
augment the number of the latter.

Whether the existing stellar movements and the mutual attrac-
tions of the stars are such as to give any substantial ground for
believing that close approach can be a chief agency in producing
comets, meteorites, and nebule, can only be determined when
some approximate knowledge of the dispersion, the masses, the
velocities, and the paths of the stars is gained. If the stars be-
considered simply as so many scattered bodies flying through
space in straight lines at computed rates, and all mutual attrac-
tions and systematic relations be ignored, the frequency of dis-
turbing approaches would not seem to be great and the
quantitative value of the doctrine here sketched would seem to
be questionable. The solar system has certainly never been
subjected to disturbing approach since its present organization.
But the assumptions made are certainly not the true ones and
may not be representative. Besides the mere hazard of flying
bodies, the mutual attraction of two stars after they enter upon
each other’s spheres of dominant influence —and these are very
large —increases notably the probabilities of a disturbing
approach even in the case of stars moving in opposed directions,
while in the case of stars moving in sub-parallel and gently con-
verging paths at sub-equal velocities, it may apparently become
a dominant factor. At the average computed distances of the
stars from each other, their mutual attractions are very slight, and

FORMATION OF METEORITES 391

in the central portion of the stellar system, in which the sun
seems to be placed at present, the general attractions are prob-
ably nearly balanced. Two stars, therefore, whose speeds are
sub-equal and whose paths gently converge, may be controlled
almost freely by their mutual attractions after they come within
the spheres of each other’s dominant influence. Such stars under
mutual control would describe paths relative to each other
similar to those assumed in the discussion. Their closeness of
approach at periastron would be determined by the relative
differences (not the total amounts) of their speeds and momenta-
The principle of sub-parallel movements applies here and gives
results quite at variance with those that obtain in cases of opposed
movements, where the relative sums of the velocities and
momenta are to be considered. The movements of the long-
orbit comets seem to be concrete expressions of this principle,
as their perihelia are largely clustered on the front side of the
Sun, z. e., the side toward which it is moving, and they make
close approaches to it. Such star clusters as the Pletades, the
members of which seem to have proper movements nearly the
same in amount and direction, are doubtless also expressions of
the principle of sub-parallelism, and in their remarkable neb-
ulosity they may at the same time illustrate the doctrine of dis-
turbed secondaries leading on to dispersive action, a part of the
product of which remains associated with the stars themselves,
while a part is more or less widely scattered, as the terms of the
doctrine require.

If our stellar system has a definite boundary and 1s a flat-
tened spheroidal cluster or a discoid, and if the ideal paths of the
stars are elongate orbits stretching from border to border across
the heart of the cluster (except as diverted by close approaches),
then the orbital speeds and momenta should be lowest on the
outer surface, and the paths should there be most frequently
sub-parallel, and hence the conditions for the close approach of
two suns through their reciprocal attraction be there most favor-
able. Now, visible nebula are most frequent in the regions
polar to the Milky Way, and they may be regarded as lying on

392 I. €. CHAMBERLIN

the flat sides or outer border of the stellar discoid where these
conditions of low orbital velocity and momenta and prevalent
sub-parallelism are dominant, and thus the distribution of nebule
and the doctrine of close approach seem to be, so far at least,
brought into harmony.

It may be needless to remark that the general conception
lying back of the doctrine of dispersion by close approach has
a complementary regenerative or reconstructive phase, which,
taken with the dispersive phase, makes up acyclic process. With
the disruptive action there is correlated a reciprocal concentrative
action, which is supposed to reproduce organized systems out of
the wreckage of disrupted systems. The notion is further enter-
tained that the two processes may be mutually self-adjustable,
within the limits of general conditions, and thus may give a
large degree of perpetuity to the existing phase of the stellar

system.
T. C. CHAMBERLIN.

UNIVERSITY OF CHICAGO,
June 1901.

SEU DITESEFOR STUDENTS

THE CONSTILUENTS OF METEORITES, 1

ELEMENTS

THe following elements have by good authorities been
reported as detected in meteorites by means of chemical or
spectroscopic examination :

Aluminum
Antimony
Argon
Arsenic
Barium
Bismuth
Calcium
Carbon
Cerium
Chlorine
Chromium
Cobalt
Copper
Didymium

Helium
Hydrogen
Iodine

Iron

Lead —
Lithium
Magnesium
Manganese
Molybdenum
Nickel
Nitrogen
Oxygen
Palladium
Phosphorus

Potassium
Selenium
Silicon
Strontium
Sodium
Sulphur
Thallium
Tin
Titanium
Tungsten
Uranium
Vanadium
Zine

Many of these, however, occur only as traces, while others

may possibly have been introduced by terrestrial agencies. The

following list will be therefore more satisfactory as giving the

primary and fundamental elements known to enter into the com-

position of meteorites :

Aluminum
Calcium
Carbon
Chlorine
Chromium
Cobalt
Copper

Hydrogen
Iron
Magnesium
Manganese
Nickel
Nitrogen
Oxygen

Phosphorus
Potassium
Silicon
Sodium
Sulphur

It will be of interest to compare the more important of these in

393

304 STUDIES FOR STUDENTS

the order of their relative abundance, with the eight most
important elements of the earth’s crust placed in similar order.
The list of the latter is taken from Roscoe and Schorlemmer.'

METEORIC SERIES TERRESTRIAL SERIES

1. Iron I. Oxygen

2. Oxygen 2. Silicon

3. Silicon 3. Aluminum

4. Magnesium 4. Iron

5. Nickel 5. Calcium

6. Sulphur 6. Magnesium

7. Calcium 7. Sodium

8. Aluminum 8. Potassium

It should be remembered in drawing conclusions from the
above list that the elements of cosmic matter in its entirety are
here compared with the elements of only the crust of the earth ;
further, that the meteoritic matter now known probably does not
show a true proportion of stony matter. As I have shown else-
where,’ the iron meteorites are much more likely to be known
and preserved than the stony. It is probable, therefore, that if
the average composition of meteoritic matter were known, iron
would not occupy so high a place as it does in the above table.
The relative excess of magnesium and nickel, and scarcity of
aluminum and calcium in meteoritic, as compared with terrestrial,
matter may be due to the same cause.

COMPOUNDS

The elements of meteorites chiefly occur combined. The
exceptions are iron, nickel, cobalt, and copper, all of which
occur largely in the form of alloys, carbon, and the gases,
hydrogen, and nitrogen, probably held as elements in the pores
of meteorites.

The compounds of meteorites according to the mineralogical
names by which they are generally known, and roughly in the
order of their relative abundance, are as follows, the minerals
not occurring upon the earth being printed in italics:

* Treatise on Chemistry, Vol. I. 2JouR. GEOL., Vol. V, p. 126.

THE. CONSTITUENTS OF METEORITES 395

ESSENTIAL ACCESSORY
Nickel-iron Included gases
Chrysolite Iron sulphide
Orthorhombic pyroxene Schretbersite
Monoclinic pyroxene Graphite
Plagioclase Cohenite
Maskelynite Glass

Chromite
Amorphous carbon
Diamond
Daubreelite
Tridymite

Lawrencite
Magnetite
Oldhamtite
Hydro carbons
A brief account will be given of each of these.
Nickel-iron.— This is the most widely distributed constituent
of meteorities and in quantity it exceeds all the others combined.
It makes up practically the entire mass of all the iron meteorites,
the larger part of the mass of the iron-stone meteorites and is
found in nearly all, though not all, the stone meteorites. It is
an alloy of iron and nickel in which the percentage of nickel
varies from about 6 per cent. to about 20 per cent. Some iron
masses claimed to be meteorites contain a higher percentage and
some authorities regard the nickel-iron of most stone meteorites
as generally containing from 20 to 40 per cent. of nickel, but
this is somewhat uncertain. From 0.5 to 2 per cent. of cobalt
always accompanies the nickel, as well as .006 to .02 per cent.
of copper. Traces of manganese and tin are also often found.
The terrestrial nickel-iron of the Greenland basalts differs from
that of meteorites in having a lower percentage of nickel (0.25
to 4 per cent.) and in containing a considerable amount (3 per
cent.) of carbon. The terrestrial nickel-irons known as awaruite
and josephinite contain higher percentages of nickel than the
meteoritic, the percentages being 67.7 per cent. and 30.5 per
cent. respectively. In color, meteoritic nickel-iron varies from
iron or steel-gray to silver-white, according to the percentage of

396 STUDIES FOR STUDENTS

nickel present. In hardness and tenacity the nickel-iron of differ-
ent meteorites varies greatly. That of some meteorites is harder
than steel, that of others softer than wrought iron. That of
some meteorites is so brittle as to break in pieces with a blow of
the hammer, that of others so malleable that it can be worked
into implements of various shapes. Nickel-iron is strongly mag-
netic and some iron meteorites exhibit polarity due perhaps to
induction of the magnetism of the earth. The specific gravity
of nickel-iron ranges between 7.6 and 7.9. It is dissolved at
ordinary temperatures by the common acids, by solutions of
copper sulphate, by copper chloride, by mercuric chloride, by
bromine water, by copper ammonium chloride, and by a few
other reagents. Some masses of nickel-iron when placed in
neutral solutions of copper sulphate reduce the latter, while
others do not. The former are known, according to the terms
first used by Wohler, as active, the latter as passive irons.
Nickel-iron oxidizes rapidly when. exposed to the atmosphere,
the rapidity decreasing, however, with increase in the percentage
of nickel. In regard to the manner of occurrence of the nickel-
iron it may be noted that in the iron meteorites it forms a com-
pact mass except in so far as it is interrupted by inclusions of
other minerals. In the iron-stone meteorites all gradations
occur from a continuous network to isolated grains. In the
stone meteorites it is present in the latter form. A more or
less lineal arrangement of these grains, recalling Widmanstatten
figures, is often observed in the stone meteorites. When the
substance occurs in grains, whether large or small, the shape of
these is usually very uneven, being sometimes more or less
rounded but generally irregularly branching. Sometimes regular
forms such as cubes and octahedrons may be observed. In the
Ochansk meteorite, von Siemaschko observed actual crystals
made up of a combination of the cube, octahedron, dodecahedron,
and a tetrahexahedron. Other cuboidal forms have been
observed. The two or possibly three subordinate alloys (kama-
cite, taenite, and plessite) of which nickel-iron is composed have
been described in a previous article and their composition given.

THE CONSTTILTOENTS OF METEORITES 397

Chrysolite—This is, next to nickel-iron, the chief mineral
constituent of meteorites. It is found in all the iron-stone and
nearly all the stone meteorites and makes up a large part of
their mass. It occurs as crystals and as rounded and angular
grains. In the group of iron-stone meteorites known as palla-
sites it is porphyritically developed‘in the nickel-iron; in other
iron-stone meteorites it forms together with pyroxene a granular
aggregate filling the meshes of a network of nickel-iron. In the
chondritic meteorites the manner of its occurrence has already
been described. Crystals occurring in cavities or isolated by
dissolving adjacent nickel-iron lend themselves readily to gonio-
metric measurement. A total of twenty forms, similar to those
found on terrestrial chrysolite has thus been identified. The
color of the mineral is usually the typical olive-green of terres-
trial chrysolite but may vary to honey-yellow or red. Much of
the meteoritic chrysolite is characterized by an abundance of
opaque inclusions often regularly arranged. Intergrowths with
a colorless to dark brown glass are also common, especially in
the chrysolite of chondritic meteorites. Gas pores are rare.
Alteration products so common to terrestrial chrysolite are
entirely lacking. Much of the chrysolite shows a strong ten-
dency to fissuring, especially in thin sections. Well-marked
cleavage is not common. Numerous analyses of mechanically
separated chrysolite show a composition similar to that of the
terrestrial mineral. The percentage of Fe in these analyses shows
variations from about IO per cent. to about 30 per cent. One
feature of the composition of meteoric chrysolite which seems
at first difficult to account for, is an almost entire lack of nickel
oxide. This, as is well known, is a very constant constituent of
terrestrial chrysolite. Daubrée has shown, however, that an
absence of nickel from meteoritic chrysolite should be expected,
since nickel has less affinity for oxygen than iron and would not
be attacked until the latter was completely oxidized. While ter-
restrial iron has been completely oxidized that of meteorites has
not. The correctness of this explanation has further been shown
experimentally by fusing terrestrial chrysolite with pyroxene in

398 SLUDIES FOR STRODENTS

the presence of a reducing agent. The nickel of the chryso-
lite then formed an alloy with the iron of the pyroxene. The
siliceous portion of meteorites that is soluble in hydrochloric
acid may for the most part be considered chrysolite, since
numerous analyses of this portion give results corresponding in
composition to this mineral.

Orthorhombic pyroxenes.— The minerals of this group are next
in abundance to chrysolite as a constituent of meteorites. They
form an essential part of nearly all stone meteorites and are not
lacking in the iron-stone meteorites. At least four meteorites
consist of orthorhombic pyroxenes alone. These are the meteor-
ite of Bishopville, practically composed of enstatite alone, and
those of Manegaon, Ibbenbiihren and Shalka, which consist
essentially of hypersthene. The color of the orthorhombic
pyroxenes varies from colorless through white to various shades
of green. Often the mineral has the typical color of chryso-
lite. In thin section the pyroxene is colorless to slightly
colored. Its habit is usually prismatic but it may also occur
as rounded grains. Crystais with well defined planes have
been observed in the Breitenbach, Bustee, Manegaon and other
meteorites. A total of thirty-two forms has thus been identi-
fied and the axial relations found to correspond with those of
terrestrial hypersthene. Prismatic, macrodiagonal and brachy-
diagonal cleavages are recognizable. It is especially character-
istic of the mineral to form eccentric, radiating, polysomatic
chondri, the structure of which has been described in a previous
article:

Numerous chemical analyses of mechanically separated ortho-
rhombic pyroxenes have been made. These show all grada-
tions between the compositions represented by the formulas
MgSiO, (enstatite) (Mg, Fe) SiO, (bronzite) and (Fe, Mg)
SOR (hypersthene). The portion insoluble in acids, of meteor-
ites consisting essentially of nickel-iron, chrysolite and ortho-
rhombic pyroxenes, may be considered to be essentially the
latter, as shown by numerous analyses which give results cor-
responding with the pyroxene formula. The orthorhombic

THE -GONSTITOENTS OF METEORITES 399

pyroxenes of meteorites are thus seen to be entirely compar-
able to the terrestrial minerals of the same name.

Monochnic pyroxenes— Two kinds of monoclinic pyroxenes
have been identified in meteorites, the first bearing iron and
alumina, the second free from alumina and nearly free from
iron. The first may be considered similar to terrestrial augite,
the second to terrestrial diopside. Augite has been identified
in many meteorites, diopside positively only in one. Crystals
of meteoritic augite have been measured goniometrically and
eight forms similar to those of terrestrial augite found. As a
rule, however, the augite occurs as grains or splinters. It varies
from brown to green in color, in some meteorites is pleochroic
in thin section in others not at all. Parting parallel to the base,
owing to repeated twinning, is common and characteristic. It
is sometimes regularly intergrown with orthorhombic pyroxene.
Inclusions of glass and black dust are common. Pyroxene
resembling diopside was identified by Maskelyne in the Bustee
meteorite. It occurred in grains and splinters and was of a gray
to violet color. A few goniometric measurements were possible.
Analysis showed the composition to be that of a calcium-
magnesium pyroxene. Crystals and grains from a few other
meteorites may perhaps be referred to diopside but the determi-
nation is not certain.

Plagioclase —Of the minerals of the feldspar group, anorthite
may be mentioned as forming an essential constituent of the
classes of stone meteorites known as eukrites and howardites
and as occurring in others. It forms according to Rammelsberg
about 35 per cent. of the stones of Juvinas and Stannern. Of
the other members of the plagioclase series, albite, oligoclase
and labradorite have been reported in single meteorites, but in
most cases where plagioclase has been found the species has not
been determined. Orthoclase has not yet been identified in any
meteorite. Crystals of anthortite from the Jonzac meteorite
reach a length of 1™. From the druses of the Juvinas meteor-
ite anorthite crystals were obtained which served for goniomet-
ric measurement, eight forms being thus identified. Some

400 SLUDIES FOR SLROIDENTS.

anorthite crystals show twinning according to the Carlsbad law
and in the Llano del Inca and Dona Inez meteorites twins
according to the albite and pericline laws were found. The
mineral is sometimes white and sometimes colorless and in luster
varies from dull to vitreous. Inclusions nearly always abound
and they are generally regularly arranged. The inclusions are
chiefly colorless glass, but sometimes brownish glass and opaque
grains occur. Analyses of mechanically isolated anorthite have
been made which show a composition similar to that of terres-
trial anorthite. CaO amounts to about 18 per cent. in these
analyses.

Calculating from analyses Tschermak concludes the feldspar
of the stone of Gopalpur to be oligoclase, Lindstrom that of
Hessle to be the same and Schilling that of Tennasilm to be
labradorite. The presence of plagiociase other than anorthite
has been proved by microscopical and chemical examination
of other meteorites, but the species have rarely been deter-
mined. Such feldspars occur as lath-shaped individuals and
as grains and splinters. Inclusions are much less common
than in anorthite. Rounded, elongated inclusions referred by
Tschermak to chrysolite and bronzite are, however, quite char-
acteristic. Gas inclusions seem to be more abundant in the
feldspars of meteorites than in any other constituent, though
even here they are rare. 3

Maskelynite.— This is an isotropic, colorless, though becoming
milky through alteration, transparent mineral of vitreous luster
and conchoidal fracture. Its hardness is somewhat over 6;
specific gravity 2.65. It has no cleavage but shows occasional
irregular cracks and strie similar to those of plagioclase. Inclu-
sions of magnetite and augite are arranged in apparent zones.
The mineral is slightly decomposed by hydrochloric acid. Thin
splinters fuse, but with difficulty. Lath-shaped individuals with
rectangular outlines occur, but in most meteorites the mineral is
present as minute grains. It forms 22% per cent. of the
meteorite of Shergotty, the remainder of the meteorite being
augite and magnetite. It is also an accessory constitutent in

THE CONSTITUENTS OF METEORITES 40!

the meteorites of Chateau Renard, Alfianello, Milena, Mocs, and
others. Its composition is about that of labradorite. Tschermak
regards the mineral as a fused feldspar, while Groth and Brezina
consider it a distinct species allied to leucite. Its straight,
sharply defined outlines, the existence of striz, and the absence
of any fused appearance make Tschermak’s view difficult to
accept, though the mineral resembles the feldspars in so many
other respects.

Included gases.—A\\ meteorites which have so far been tested
give off on heating one or more of the following gases: Hydro-
gen, carbon monoxide, carbon dioxide, nitrogen, and marsh gas.
Comparing the iron meteorites with the stone meteorites in
regard to the kind of gases given off it is found that the former
are characterized by a high content of H and CO, the latter by
an excess of CO,. The following table of analyses of gases
from eight iron and six stone meteorites, quoted from Cohen,
gives an idea of the relative quantity of each gas:

H C0. CO; N CH,

Iron meteorites 63200) 20:70). 26.12" 97,52-"5°0.57

Stone meteorites 417255,- 4.15 70.66; 92:20) 4:n7
The volumes of the gases obtained vary from 0.97 of a volume
given off from the iron of Shingle Springs to 47.13 volumes col-
lected from the Magura iron. The average number of volumes
obtained from the meteorites quoted in the above table is 2.82.
The gases in meteorites appear therefore to be under a somewhat
greater pressure than that of the earth’s atmosphere. It has often
been urged that the gases obtained from meteorites by the meth-
ods above mentioned may have been absorbed from our own atmos-
phere. Itis known on the one hand that terrestrial rocks give off on
treatment gases very similar in kind and quantity to those obtained
from meteorites. Thus Wright obtained from one ordinary trap
rock 34 of a volume of gas, 13 per cent. of which was CO, and
the remainder chiefly hydrogen, and from another, one volume
of gas containing 24 per cent. CO, and the remainder chiefly
hydrogen. Tilden has also recently shown that “the crystalline
rocks of the surface of the earth contain very notable quantities

402 SOGDITES FORT SILRODEN TS:

of gas, consisting of hydrogen in preponderance, carbon dioxide,
and carbon monoxide in large percentage, and nitrogen and
marsh gas in small quantities, with water vapor, but with a prac-
tical absence of oxygen. Twenty-five analyses including ancient
and modern volcanic and even some metamorphic rocks gave an
average volume of gas equal to about four and one half times
the volumes of the containing rocks.”’* Further, it is urged that
no meteorites have been analyzed as to their gases immediately
after their fall. In contrast to these facts it should be noted
that the Homestead meteorite was analyzed for gases by Wright
within three months from the time of its fall. A second analy-
sis was made a year later in order to test the influence of the
earth’s atmosphere upon the stone. It was found that very little
change had taken place except a slight Joss of carbonic acid.
Ansdell and Dewar in testing the gases of the Pultusk and Mocs
meteorites chose stones of those falls which were completely
incrusted so that the chances of absorption of gases from the
earth’s atmosphere might be reduced to a minimum. Yet the
results obtained accorded well with those from other meteoric
stones and for Pultusk the percentages were remarkably like
those derived by Wright in a previous and independent examina-
tion of stones of the same fall. There seems, therefore, good
reason to believe that the gases obtained from meteorites are
brought with them from space and that they have not been
derived from the earth’s atmosphere.

How the gases are held by the meteorites is uncertain.
Wright was inclined to believe that the pores occasionally noted
in the silicates of meteorites indicated cavities where the gas
was held. Such pores are of too rare occurrence, however, to
meet the demands of the problem. The phenomenon seems
more like the occlusion of hydrogen by platinum or zinc, and
the gases are probably held partly in the intermolecular spaces
and partly chemically united. Travers, however, regards them
as produced by heat from the non-gaseous elements of the

*T. C. CHAMBERLIN: JOUR. GEOL., Vol. VII, p.558. Quoted from Chemical
News, April 9, 1897.

THE CONSTITUENTS OF METEORITES 403

meteorites.* The magnetic and nonmagnetic or, in other words,
the metallic and stony portions of the Homestead meteorite
were tested separately by Wright in order to determine whether
these different portions exercised any selective action in holding
gases. The investigation gave the following results :

|

| Volumes H CcoO+co, N
Emtine:stone. 2.2.2.0 nicneee | Toi7 50.93 48.07 1.00
Magnetic portion,o.51.... i 48 59.38 B5.72 1.90
Non-magnetic portion, 0.97. | He

30.96 66.96 2.08

The results show no important differences in the gases held
by the different portions. By way of caution, attention should
be called to the fact that the gases in meteorites may not have
been originally present in the form and. quantities which the
analyses indicate. Thus Wright in making his analyses found
CO, rapidly reduced to CO through contact with heated iron.
Likewise, H, CO, and iron may at a moderate heat reduce the
iron oxide present in many meteorites, and thus the character of
each be changed. The percentages of the different gases obtained
by analyses may be, therefore, more indicative than absolute.

Cohen calls attention to the fact that from artificial irons
may be obtained gases corresponding both qualitatively and
quantitatively to those obtained from meteoric irons. The fol-
lowing list of analyses illustrates this.

H CO CO, N
White, carbonaceous cast iron.......... 74.07 16.76 3.59 5.58
MV GEStE SL Aireiarats nyclotie oitaisve cicesoten parte. 52.6 24.3 16.55 6.5
Ordinary gray massive charcoal iron....| 38.60 49.20 12.20
GrayiColewrO Ms ccs. ctercieei see foisvevers ene) «Gt bess 37.70 57.90 8.40
Steeler canes ce site denen wudied season ates 2On27, 63.65 227 11.36

Finally it should be noted that, according to the investiga-
tions of Vogel, Wright, and Lockyer, the spectra of the gases
obtained from meteorites show remarkable resemblances to the
spectra of comets.

* Proc. Roy. Soc., Vol. LXIV, pp. 130-142.

404 STUDIES FOR STUDENTS

Iron sulphide.—Tvrotlite-pyrrhotite—The exact form and com-
position of the iron sulphide which is a common ingredient in
meteorites is a question not yet satisfactorily answered. For
convenience, Rose’s assumption that the iron sulphide of iron
meteorites is troilite, that of stone meteorites pyrrhotite, is
usually followed, but there are many occurrences which do not
harmonize with this view.

The iron sulphide known as troilite is usually found massive,
though crystals have been observed which have been referred by
Brezina to the hexagonal and by Linck to the isometric system.
The color varies from bronze-yellow to tomback-brown. Streak
black. Hardness, 4. Specific gravity, 4.68—-4.82. Generally
found to be non-magnetic, although magnetic troilite has been
reported. Cohen suggests that the magnetism may be due to
included nickel-iron. The mineral fuses in the reducing flame
to a black, magnetic globule. Decomposed by hydrochloric
acid with evolution of hydrogen sulphide, but without separation
of sulphur. Not affected by copper sulphate or fuming nitric
acid. These reagents may be used, therefore, for its separation.

Most analyses show a composition approximating very
closely to FeS. Meunier, however, obtained results more
nearly in accord with the formula Fe,,S,,. As this is the com-
position of pyrrhotite he regards the two as identical. The
specific gravities which he obtained, however, correspond to
those observed by others for troilite, and there seems therefore,
some reason to doubt the correctness of his analysis.

Troilite is almost universally present in the iron meteorites.
It may be very unequally distributed in a single mass, however,
being abundant in some portions and lacking in others. It
usually occurs in the form of nodules, but also as plates and
lamellae. The nodules vary greatly in shape and size. Rounded
and oval forms are common, as are also lens and dumb-bell
shapes. In Carlton a star-like form occurs. Smith separated
from the Cosby’s Creek iron a nodule weighing 200 grams,
while one from the Magura iron measured 13 in diameter.
When troilite occurs as lamellae, these are often regularly

LHE CONSTITUENTS OF METEORITES. 405

arranged parallel to the planes of a cube. Lamellae having this
arrangement are known as Reichenbach lamellae. Individual
lamellae of this sort average from 0.1-0.2™" in width and
14%-3%°" in length. They cross layers of kamacite, and
hence’ must have formed before these. Troilite often occurs
intergrown with schreibersite and graphite, and these sometimes
surround it. It also often includes nickel-iron.

The fusion and dissipation of troilite nodules during the pas-
sage of a meteorite through the atmosphere is a cause of the
depressions often to be observed on the surface of both iron and
stone meteorites.

The iron sulphide of the stone meteorites oeturs chiefly as
grains, sometimes as plates, and sometimes in vein-like forms.
As mentioned in a previous article, it also occurs in chondri, fre-
quently forming their periphery, while at other times it is in the
form of grains. Crystals from the druses of the Juvinas meteor-
ite measured by Rose proved to be hexagonal and to have forms
similar to those of terrestrial pyrrhotite. It is largely on account
of these observations that the iron sulphide of stone meteorites
is considered to be pyrrhotite. On the other hand, the iron
sulphide of stone meteorites differs from pyrrhotite in being,
for the most part, non-magnetic, and in giving no free sul-
phur on decomposition with hydrochloric acid. Further, most
analyses show a ‘composition corresponding to the formula
Fess:

Schretbersite—This mineral, peculiar to meteorites (if its pos-
sible occurrence in the terrestrial iron of Greenland be excepted)
is also one of their most remarkable constituents, since it gives
proof that the meteorites in which it occurs could not have
been exposed for any long time to the action of free oxygen.
The mineral is a phosphide of iron, nickel, and cobalt, having
the general formula (Fe, Ni, Co), P, though the relative pro-
portions of the metals vary. The normal color is tin-white,
but this may readily alter to bronze-yellow or steel-gray on
exposure to the air. Hardness 6.5, specific gravity 6.3-7.28.
Strongly magnetic, and when magnetized retains its magnetism

400 STUDIES FOR STUDENTS

for a long time. Very brittle, being thus distinguished from
taenite, with which it is often confounded. Another property
which distinguishes it from taenite and from cohenite is that it
is insoluble in copper-ammonium chloride. It is soluble in
ordinary dilute acids and in acetic acid. Does not reduce cop-
per from a copper sulphate solution. Easily fusible betore the
blowpipe toa magnetic globule. It occurs as crystals, flakes,
foliae, grains, and as needles. In the latter form it was long
regarded a separate mineral, and was known under the name
of rhabdite, but the identity of rhabdite and schreibersite has
been proved by Cohen. The needles and plates often exhibit
angular outlines. Individual masses of the mineral often reach
a considerable size, one from the Carlton iron being 14 ™. in
length. The mineral also forms a considerable portion of the
mass of some meteorites, such as Bella Roca, Primitiva, and
Tombigbee River. It is the most widely distributed constituent
of iron meteorites, aside from nickel iron, and is believed to be
usually associated with the latter mineral in the stone meteorites,
though its quantity is so small that it has not often been deter-
mined. The small percentage of phosphorus usually found in
the analysis of stone meteorites is generally referred to this
mineral. Schreibersite has been reported in the terrestrial iron
of Greenland, but its presence is not proved. Phosphides
similar to schreibersite have been made in several ways arti-
ficially. The process followed has been essentially to heat iron
to a high temperature together with a phosphorus-bearing
compound.

Graphite.—This substance occurs in grains of sufficient size
for ready examination only in the meteoric irons. In these it
is usually in the form of nodules but sometimes occurs in plates
or grains. The nodules often reach considerable size. One
nodule taken from the Cosby’s Creek iron is as large as an
ordinary pear and weighs 92 grams. Even larger ones were
found in the Magura iron. Toluca, Cranbourne, Chulafinnee and
Mazapil are other irons which contain considerable graphite.
Graphite has been estimated to form 1.17 per cent. of the mass of

THE CONSTITUENTS OF METEORITES 407

Magura and 0.8 percent. of the Cosby’s Creek iron. The mineral
is usually associated with iron sulphide. With this it may be
intimately intergrown or the one may enclose the other. Its
texture is compact rather than foliated. Smith found that the
meteoric graphite oxidized much more rapidly than terrestrial
graphite on treatment with nitric acid and chlorate of potash.
This feature distinguishes it from the amorphous carbon separated
from cast iron. ©The meteoritic graphite is. also very pure.
Although occurring in nodules of the size described, which must
have segregated from the surrounding mass, the ash amounted,
in an analysis made by Smith, to only 1 per cent. By ether was
extracted a small quantity of a substance made up of sulphur and
a hydro-carbon, which constituted the only other impurity.
Emphasizing the differences between meteoritic and terrestrial
graphite Smith was inclined to believe that the graphite of
meteorites must have been formed by the action of bi-sulphide
of carbon upon incandescent iron rather than that it was analo-
gous in its origin to terrestrial graphite. Ansdell and Dewar,
however, concluded from elaborate comparisons of meteoric and
terrestrial graphite that they were similar in origin, and were
formed by the action of water, gases and other agents on
metal carbides. Whatever its mode of formation the occur-
rence of graphite in meteorites is of geological interest as
proving that graphite may be formed in nature without the
agency of life.

Cohenitte.— This is a carbide of iron, nickel and cobalt. It
has been positively identified in only a few meteorites but is
doubtless of common occurrence. Its formula is (Fe, Ni, Co),
C. The mineral is of metallic luster and tin-white color, though
readily tarnishing to bronze-yellow. Hardness, 5.5—6. Specific
gravity, 7.23-7.24. Strongly magnetic; very brittle. Insoluble
in dilute hydrochloric acid and decomposed by concentrated
hydrochloric acid only with difficulty. Easily soluble in copper-
ammonium chloride. It occurs as isolated crystals on which
several forms of the isometric system have been noted; also as
grains. Elongated crystals, reaching a length of 8™™ are

408 STUDIES FOR STUDENTS

found in the Magura meteorite. These are arranged parallel to
octahedral planes. An iron carbide similar to cohenite is formed
in cast iron when the latter is heated to a temperature of 600—
700° C. and slowly cooled. Cohenite occurs in the terrestrial
iron of Niakornak, Greenland.

OLIVER C. FARRINGTON.

(To be continued.)

Pie eA wOZ OIG, FORMATIONS. OF -ALLEGANY
COUNTY, MARYLAND:

INTRODUCTION

THE author of this paper has been engaged since the summer
of 1897 as chief of the Division of Appalachian Geology of
the Maryland Geological Survey in studying the geological for-
mations of the western counties of Maryland. He has had as
assistants in this work at different times Messrs. GG] Oo Earra,
R. B. Rowe, G. C. Martin, A. C. McLaughlin, and A. P. Romine.
Mr. Richard B. Rowe, who was already acquainted with the New
York formations, reached the conclusion, as the result of his
field work during 1897, that several of the Paleozoic forma-
tions of western Maryland could be correlated with those of
New York. The continuation of Mr. Rowe’s work, together
with that of Mr. Romine, under my direction during the
field seasons of 1898, 1899, and 1goo further confirmed these
views.

The following account of the Paleozoic formations of Allegany
county, Maryland, embraces a brief description of their charac-
ter and distribution, together with a statement regarding their
probable correlation with the New York and Pennsylvania for-
mations. The report of Dr. C. C. O’Harra on “The Geology of
Allegany County”? incorporated the revised classification of
western Maryland devised by Dr. William B. Clark, the writer;
and his assistants, and thus represents the conclusions, based on
the field work carried on during the seasons of 1897-1900. The
writer is under obligations to Professor Bailey Willis, whose
manuscript on ‘The Appalachian Region— Paleozoic Appa-
lachia, or the History of Maryland during Paleozoic Time,”’3 was

t Published by permission of Dr. William Bullock Clark, state geologist of Mary-
land.

?Md. Geol. Survey, Allegany county, 1900, pp. 57-163.

3 Maryland Geol. Survey, Vol. IV, 1900, pp. 23-93.

409

410 CHARLES 'S. PROSSER

kindly placed at his disposal, and has recently been published
under the auspices of the state survey.

THICKNESS

The strata covering Allegany county have a thickness vary-
ing from about 13,300 to 16,000 feet. This thickness ts divided
between the four geological systems represented in the county
in the following manner: Silurian, from 2200 to 2400 feet;
Devonian, 7875 to 10,200 feet; Carboniferous, 2825 to 3000
feet, and Permian (?), about 400 feet.

SILURIAN STRATA

Juniata formation.—Vhe oldest rocks outcropping in Allegany
county belong in this formation, and are shown in only one
locality. This outcrop forms the lower part of the cliffs in the
gorge known as ‘‘the Narrows”’ just northwest of Cumberland,
where Wills Creek has cut a deep and narrow trench through Wills
Mountain. This gorge presents an admirable example of a narrow
transverse valley where a stream has cut through a mountain
ridge, and this locality has a justly deserved reputation for great
natural beauty. The upper 370 feet of the formation is well
shown on the northern side of the creek; but in many places in
the gorge it is concealed for the most part by heavy talus of white
quartzite blocks from the cliff above. In the Narrows 550 feet
of the Juniata are shown, but no fossils have been found at this
locality. The formation is composed of alternating beds of deep
red shales and sandstones which have no regularity of succes-
sion, but show a much greater total thickness of shales than
sandstones. The sandstones are hard, fine-grained, cross-
bedded, and micaceous, some of the strata a foot or more in
thickness, but usually less than six inches. The shales are
micaceous, weathering readily, and the beds vary from an inch
to six feet or more in thickness. This formation was named
from the Juniata River, Pa., though in the earlier reports of that
state it was termed the Medina or Levant red sandstone (No.
IV4). It probably represents the older part of the Medina

PALEOZOLG FORMATIONS OF MARYLAND 411

stage of New York near the base of the Upper Silurian, which,
in western New York, is composed mainly of red shales.

Tuscarora formation—This quartzitic sandstone, which rests
conformably upon the Juniata formation, and forms the beautiful
flat-topped arch of Wills Mountain is admirably shown on its west-
ern side, especially at the entrance to the Narrows, and in the upper
part of the high massive cliffs bordering the gorge. It also forms
the greater part of the exposed rocks in Evitt’s and Tussey’s
mountains and two small areas along the Baltimore and Ohio
Railroad near Potomac Station. The formation is a white to
light gray, very hard quartzose sandstone composed of fairly
coarse quartz grains cemented by siliceous material. Small
pebbles of quartz and also those of yellowish-green hard clay
occasionally occur. The layers are frequently very massive and
cross-bedded structure is not infrequent. Its thickness varies
from 250 to 300 feet, and it furnishes building stone and ballast.
Arthrophycus harlani, a sea-weed, the only fossil found in this
sandstone in the county is fairly common on the upper surfaces
of the higher beds in the Narrows.

The Tuscarora formation, named from Tuscarora Mountain in
Pennsylvania, was formerly called the Medina or Levant white
sandstone (No. 1Vc), and represents the upper part of the
Medina stage of New York. The Juniata and Tuscarora forma-
tions are probably equivalent to the Medina formation of New
York, but in western New York, where it is typically represented,
especially in the Niagara region, there is not such a definite sepa-
ration into a lower divison composed of red shales and sand-
stones and an upper one composed of a white quartzose
sandstone as in Maryland. In the Niagara region the Medina
is composed of the following divisions named in ascending
order: Lower Medina, composed of red shales only about the
upper 115 feet of which is exposed in outcrop. Upper Medina,
composed of seven zones: (1) 25 feet of gray quartzose sand-
stone; (2) 25 feet of thin gray shales; (3) five feet of gray
sandstones and sandy shales; (4) 6 feet of mainly gray argilla-
ceous shales which become reddish at the top; (5) 35 to 40 feet

412 CHARLES S. PROSSER.

of mainly thin-bedded sandstone reddish in color or gray mot-
tled with red; (6) 12 to 15 feet of massive sandstone in beds
from one to several feet in thickness and varying in color from
reddish to grayish; (7) at the top 7% feet of a hard massive-
bedded and compact white quartzose sandstone similar to No. I.
In some thin sandstones in the upper part of No. 6 occurs the
characteristic Medina fossil known as Arthrophycus harlan (Con-
reeuel)) a

Clinton formation.—TVhe largest area of this formation flanks
both sides of Wills Mountain extending to the Potomac River,
while two other areas flank the southern ends of Evitt’s and
Tussey’s mountains. The formation is composed largely of
yellowish-green to reddish shales, but on weathering, the flat
surfaces frequently have a scarlet tint. There are blackish
shales and thin fossiliferous limestones in its upper part as well
as a greenish-gray to reddish sandstone. The most important
lithologic character of the formation, however, is the two beds
of iron ore, the lower occurring from 120 to 160 feet above its
base and consisting of two strata of iron ore separated by a band
of greenish-yellow shales from 6 inches to 6 feet thick. The
two ore-bearing strata have a thickness of from Io to 12 feet.
At Cumberland the upper bed of iron ore is 270 feet above the
lower one, in the midst of a brownish, calcareous sandstone,
nearly 3 feet thick which is directly above a massive 5-foot
sandstone stratum. There are g inches of quite clear, fossilifer-
ous iron ore and the overlying greenish shales and thin bands of
bluish limestone also contain fossils. The thickness of the for-
mation varies from 550 to 600 feet. Fossils are common in the
middle and upper portions, some of the species being identical
with those of the New York Clinton. The name is derived from
the exposures at Clinton, Oneida county, N. Y., where the beds
of iron ore have been mined for many years, and the stage is
identical with No. Va of the Pennsylvania survey which is

*For an excellent account of the Medina formation along the Niagara River, see

Bull. N. Y. State Museum, No. 45, 1901, pp. 87-95 and accompanying geological map
by Dr. Amadeus W. Grabau.

PALEOZOIC FORMATIONS OF MARYLAND 413

called the Rockwood formation in the Piedmont folio of the U.
S. Geological Survey.

Magara formation—This formation surrounds the three areas
mentioned under the Clinton formation. The lower part con-
sists of thin-bedded, blue limestones with thin shale partings;
but in the upper part the shales predominate and become black-
ish in color. The thickness varies from less than 250 to fully
300 feet. The thin limestones contain brachiopods and other
fossils, some of which are specifically identical with the Niagara
fossils. The formation is named from the admirable exposures
at Niagara Falls and represents No. Vé of the Pennsylvania sur-
vey and the lower part of the Lewistown formation of the Pied-
mont folio. The revised classification of the New York series by
Messrs. Clarke and Schuchert, however, drops the Niagara
formation or group and returns to the earlier classification
of Rochester shale, Lockport limestone, and Guelph dolo-
mites

There has been more or less uncertainty regarding the iden-
tification of the Niagara limestone south of New York; but
recently Dr. Weller has conclusively shown that the Decker
Ferry formation of western New Jersey and eastern Pennsyl-
vania is of the same age ‘‘as the Rochester shale and Lockport
limestone of Clarke and Schuchert, or as the Niagara formation
of most authors.’’? A small collection of fossils from the beds
near Cumberland was submitted to Dr. Weller who kindly exam-
ined them and wrote that his impression is that they are the
same as the Decker Ferry formation, and in New Jersey there
are sufficient authentic Niagaran species to definitely refer the
formation to the Niagaran. In conclusion he stated that “I
should think you would be fully justified in considering the
Cumberland fauna as of Niagaran age.”3 Mr. Schuchert has
studied these beds in the field as well as their fossils and he posi-
tively correlates them with the Niagara. His statement is that

t Science, N. S., Vol. X, 1899, p. 876.

?Geol. Surv. N. J., Ann. Rept. for 1899-1900, p. 18, and also see p. 20.

3 Letter of May 29, Igol.

414 CHARLES S. PROSSER

“there are beds at Cumberland, Md., holding a fauna of the age
between the Lockport | Rochester ?| shale and the Guelph of
New York and Ontario. This fauna has its peculiarities, but the
aspect is certainly not Clinton.’’”

Salina formation.—Vhis formation borders the Niagara but the
rocks are largely concealed over most of the areas except along
Wills Creek in Cumberland, Flintstone Creek at Flintstone, and
the Baltimore and Ohio Railroad near Potomac Station, where
one finds the best exposure. In the Potomac section there are
four cement beds which have great economic value, the lowest
one situated about twenty-five feet above the base of the forma-
tion. Fifty feet of the succeeding 140 feet of the formation 1s
composed of the four cement beds which are separated by shales
and impure limestones. Succeeding the upper cement bed are
450 feet or more of gray shales, drab and blue limestones
and sandstone, the limestones predominating.? The thickness is
about 700 feet. Fossils are not common. The formation was
named from Salina in central New York and is represented by
No, Ve of the Pennsylvania survey and the second part of the
Lewistown formation of the Piedmont folio.

The statement has been made that ‘“ the geological reader
will wonder on what basis the name Salina is applied to the rocks
so described”’ in Maryland It is perhaps sufficient to state
that Professor Lesley, the former state geologist of Pennsy]-
vania and admirable stratigraphical geologist, correlated the
corresponding rocks of Pennsylvania with the Salina. In Bed-
ford county, Pennsylvania, which lies immediately north of the
Cumberland region, Professor Lesley gave the upper and middle
divisions of the Salina as 628 feet in thickness to which is to be
added a portion of the 472 feet composing the lower Salina

™ Letter of June 26, 1901. ;

2 Mr. Schuchert has recently studied this formation in the Cumberland region and
he writes me as follows: ‘I am inclined to cut out of the Salina the lower 23’ 6” as
given on p. 93 of O’Harra’s report (Maryland Geological Survey, Allegany county).

This thickness I now refer to the rest of the Niagara. I also extend the Salina a
little higher, making the total thickness 704'.” (Letter of June 26, 1901.)

3 Am. Jour. Sci., 4th ser., Vol. XI, p. 240, 1901.

PALEOZOIC FORMATIONS OF MARYLAND 415

which included the Niagara beds at its bottom.’’ Mr. Schuchert
also fully concurs in correlating these Maryland beds with the
Salina:

DEVONIAN STRATA ”

Helderberg limestone — The largest area covered by this for-
mation is in the central part of the county which it enters to
the north of Flintstone and then runs in a zigzag manner back
and forth until it leaves the county on the western side of Evitt’s
Mountain. Another area follows Shriver Ridge and passes
through western Cumberland across the county. A third area
enters in the eastern part of Wills Creek valley and runs across
the county, keeping west of Wills Mountain, to Potomac Sta-
tion; and, finally, to the southwest is the Fort Hill area, beween
Rawlings and Dawson. The best localities for studying this
formation are the Devil’s Backbone, northwest of Cumberland ;
the cliff on the West Virginia side at Cedar Cliff and the Balti-
more and Ohio Railroad cut near Potomac Station. The lower
400 feet of the formation is composed of fairly thin-bedded
bluish-gray limestones, separate pieces of which have a metallic
ring when sharply hit. The more shaly layers contain fossils
among which are Zentaculites gyracanthus and Spirifer vanuxemi,
characteristic species of the Tentaculite limestone in New York
with which this zone is correlated.

Messrs. Clarke and Schuchert in their revised classification
revived Vanuxem’s geographical name of Manlius limestone
for the paleontological one of Tentaculite limestone.’ The
Maryland zone was put in the Helderbergian instead of the
Cayugan period because, as clearly stated by Dr. O’Harra, ‘the
lithological break between it and the Salina is very marked and
can be followed in the field . . . . while there is no lithological
break between the Tentaculite and Lower Pentamerus subforma-
tions, and the division for mapping purposes cannot be made
ere... 4

*Summary Description Geol. Pennsylvania, Vol. II, p. 839, 1892.

?'The Geological Survey of Maryland has followed Dr. Clarke and Mr. Schuchert
in referring the Helderbergian period to the Devonian system.

3Science, N.S., Vol. X, pp. 876, 877, 1899. 4 Allegany county, p. 96.

416 CHARLES, Se PROSSER

Higher in the formation are massive darker blue limestones,
about 240 feet in thickness according to Schuchert or from 50
to 150 feet as given in Rowe’s report, some of the layers of
which contain numerous specimens of Pentamerus (Steberella)
galeatus. This zone probably represents the Lower Pentamerus
limestone of New York, of which the above fossil is a character-
istic species.

In place of the paleontological name Lower Pentamerus
limestone, Messrs. Clarke and Schuchert proposed Coeymans
limestone.’

In the upper fifty feet or more of the limestones are frequent
specimens of Sprrifer macropleurus characteristic of the Delthyris
shaly limestone of New York, with which this zone is correlated.
In New York this division of the Helderberg consists of cal-
careous shales and shaly limestones with some beds a foot or
more in thickness, but in Maryland it is composed mainly of
fairly massive limestone. With the exception of the Lower
Pentamerus, all the Helderberg limestones are more massive in
Maryland than in New York. In place of the term Catskill or
Delthyris shaly limestone Messrs. Clarke and Schuchert pro-
posed the geographic name of New Scotland beds.?

The Becraft limestone, which caps the Helderberg limestone
of New York, is apparently not represented in Allegany county,
although some eighty-five feet of it occurs farther east in
Washington county, Md.

The thickness of the foundation varies from 750 to goo feet,
and fossils are common in some of the layers. The limestones
are valuable for quicklime, ballast, road-metal, and building
purposes. The formation is named from the lower limestones
of the Helderberg Mountains in eastern New York, and is the
equivalent of No. VI of the Pennsylvania survey, and the upper
part of the Lewistown formation of the Piedmont folio.

Oriskany sandstone.—The easternmost area of Oriskany sand-
stone is that of Stratford Ridge, to the northeast of Oldtown ;
then follows that of the central part of the county bordering

‘Science, N. S., Vol. X, pp. 876, 877. 2 Tbid., pp. 876, 877.

PALEOZOIC FORMATIONS OF MARYLAND AL,

the zigzag Helderberg area; then, in order named, that crossing
the county along the eastern side of Shriver Ridge; the narrow
band extending across the county to the west of the Helderberg
area west of Wills Mountain; to the southwest the band on the
western side of Fort Hill, and, lastly, the area extending from
the Twenty-first Bridge to Monster Rock at Keyser.

Fic. 1.—Devil’s Backbone, near Cumberland, showing Helderberg limestone in
the steeper part and Oriskany sandstone in the farther railroad cut.

The lower part of the formation is mainly a bluish-black
cherty limestone, 75 to 100 feet in thickness, the chert in
nodules and layers, with some dark gray arenaceous shales; and
the remainder of the formation is mostly a sandstone, fre-
quently calcareous, and varying in color from gray to white,
about 250 feet in thickness. Toward the top there are a few
bands varying from grit to conglomerate. The sandstone, on
account of its calcareous cement, weathers readily to a friable
brownish or buff, porous rock, which, when protected from
erosion eventually forms beds of sand. Its thickness varies

418 (GY AAIRIOTIS. Sy, STK OSV

from 325 to 350 feet. It furnishes railroad ballast and good
glass sand. Fossils occur abundantly in zones varying in thick-
ness from an inch to several feet. The most perfect specimens
may be obtained from the beds of sand from the disintegrated
rock, and frequently from pockets of sand in the partly
weathered rock. Sprrifer arenosus and other common species of
the formation in New York are abundant, together with species
which are restricted to its southern distribution. There are
numerous springs along the contact of the Helderberg and the
Oriskany sandstone. The formation is named from outcrops
near Oriskany Falls, in central New York; is known as No. VII
in Pennsylvania; and is the Monterey sandstone of the Piedmont
folio.

Romney formation.— This formation enters the county on the
eastern side of Iron Ore Ridge, northeast of Flintstone, crosses
it and covers a large area to the east, north, and west of Old-
town, then alternates with the Oriskany areas in the southern
central part of the county, and finally crosses in a V-shaped
area—the eastern arm west of Nicholas Mountain, the point
along Evitt’s Creek, and the western arm passing through the
eastern part of Cumberland. The western area enters the county
at Ellerslie, crosses it to the Potomac River, and then extends
southwest to the bend in the river at Keyser, W. Va.

The transition from the Oriskany sandsone to the black
shale of the Romney is very abrupt, as may be seen at various
exposures of the contact, especially east of the church on the
Williams Road, two and one half miles southeast of Cumberland,
and at Monster Rock on the W. Va. Central Railroad, near
Keyser, W. Va. The lower part of the formation is composed
of thin black shales, weathering to a rusty brown, in which are
some bands of bluish-gray limestone about 150 feet above the
base. This portion of the formation is well shown in the two
railroad cuts just north of the Twenty-first Bridge on the Balti-
more and Ohio Railroad. The black shales contain specimens
of Liorhynchus liitaris and other small fossils, and in lithological
characters agree with ‘Marcellus shale of New York or No.

PALEOZOIC FORMATIONS OF MARYLAND 419

VIIIé of Pennsylvania, which they represent. The higher rocks
are drab and bluish argillaceous to arenaceous shales and thin
sandstones, which usually weather to an olive or yellowish-gray
tint. At certain localities they are very fossiliferous, contain-
ing numerous specimens of Spirifer granulosus, S. mucronatus,
Athyris spiriferoides, Tropidoleptus carinatus, Chonetes coronata,
Phacops rana and other characteristic species of the Hamilton
formation of New York, the fauna amounting to about 150
species. The formation, which varies in thickness from 1600 to
1650 feet, is named from the exposures near Romney, in north-
eastern West Virginia, and represents the Marcellus shale and
Hamilton beds of New York, and No. VIIIé and c of Pennsyl-
vania.

In 1842 Emmons proposed the name Erie group for all the
New York rocks between the base of the Marcellus shales and
the top of the Chemung." Mr. Darton, in 1892, proposed and
defined the Romney shales, named from exposures in the vicinity
of Romney, Hampshire county, in northeastern West Virginia,’
which are now known to be equivalent to the Marcellus shales
and Hamilton beds of New York. Messrs. Clarke and Schu-
chert, in 1899, used the term Erian in their revised classification
of the New York series for the group composed of the Marcellus
shales and Hamilton beds, and stated that it represented the
‘Erie Division ”’ revived with a restricted meaning.3

It appears that the Romney formation is equivalent to the
Erian group of New York; but the writer is undecided as to
which name the laws of nomenclature entitle to recognition in
Maryland.

Jennings formation.—The eastern area crosses the eastern
part of the county from the northeast to the southwest; the
second area lies to the west of Green Ridge, and in its northern
half covers a large district to the north and east of the Romney
formation; the third covers the lower part of Evitt’s Creek
valley between the two arms of the V-shaped Romney area, and

"Geology, New York, Pt. II, pp. 100, 429.

2 Am. Geologist, Vol. X, pp. 17, 18. 3 Science, N. S.,-Vol. X, pp. 876, 877.

420 CHARLES Si PROSSER.

the fourth one is the broad band covering the lower part of the
eastern face of Alleghany Front and extending from the Penn-
sylvania line southwesterly to the Potomac River above Keyser,
W.Va. The lower part. of the formation composed of thin,
black, argillaceous shales in which a few species, such as Luchiola
speciosa, Lunulicardium fragile, and Stylolina fissurella are com-.
mon, immediately succeeds beds containing characteristic Ham-
ilton fossils, and is well shown on Flintstone Creek, a few rods
above its mouth, opposite the old Flintstone tannery; by the
side of the National Road three miles northeast of Cumberland
to the west of Evitt’s Creek at Folks Mill, and on the Williams
road east of Cumberland. This subformation corresponds in
lithologic character and stratigraphic position to the Genesee
shale of New York and No. VIIIe of Pennsylvania.

Foilowing this are olive to bluish fine argillaceous shales
alternating with thin bedded sandstones. A few fossils occur
in the more bluish layers. This division of the Jennings corre-
sponds to the Portage formation of New York, or that facies
named the Naples beds by Dr. Clarke, and No. VIIIf/ of Penn-
sylvania. Dr. J. M. Clarke, who has described the Jennings
fauna for the Maryland Geological Survey, writes me that the
fauna of these two lower divisions of the Jennings formation ‘‘is
and he states
that the lower division is considered ‘‘as an integral part of the
Naples beds, bearing the Naples fauna.”

Succeeding the Portage are ee arenaceous_ shales

)

very distinctly that of the Naples subprovince ;

weathering to a buff color alternating with thin micaceous sand-
stones. Occasional layers are fossiliferous and the character-
istic Chemung species, Spivifer disjunctus, is not uncommon. The
lithological appearance of this part of the formation is quite
similar to that of a considerable part of the Chemung in south-
western New York. Higher the sandstones predominate, and
these vary in color from yellowish-gray through brownish-gray
to dark red, and vary in texture from sandstone and grits toa
white pebble conglomerate. Some of these sandstones are quite

tLetter of June Ig, Igol.

PALEOZOIC FORMATIONS OF MARYLAND A2T

massive, and in Jennings Run about one and one half miles
above Corrigansville, a zone of grit and sandstone is thirty-five
feet thick. The red rocks increase above this horizon, but
Chemung fossils including Sfirifer disjunctus extend some 650
feet higher, and the line between the Jennings and Hampshire

Fic. 2,—Heavy sandstone and conglomerate beds in the upper part of the Jen-
nings formation, as shown by the roadside above Corriganville.

formations has been drawn at the top of this fauna. The forma-
tion is between 3800 and 4000 feet in thickness. The upper
part of the Jennings may be correlated with the Chemung of
New York or No. VIIIg of Pennsylvania.

Hlampshire formation.—This formation crosses the extreme
eastern part of the county from the northeast to the southwest ;
the next area, which is the largest, flanks each side of Town Hill
and crosses the county in the same general direction; while the
third extends along the middle part of the eastern face of
Alleghany Front. The rocks consist mainly of an alternation
of red, flaggy, and massive sandstones and arenaceous or

422 CGHAREES S PROSSER:

argillaceous shales which both laterally and vertically merge
gradually into each other and in the upper part of the formation
shales, some of which are gray, and some brown in color,
predominate. Some of the sandstones are crossbedded, and in
the lower part of the formation, massive. The thickness varies
from 1900 to 2000 feet. Fossils occur very infrequently. The
formation is named from Hampshire county in northeastern
West Virginia, a considerable area of which is underlain by it,
and it represents at least part of the Catskill formation of New
York, or No. IX of Pennsylvania.

CARBONIFEROUS STRATA

Pocono sandstone—TYhe most eastern area of this formation,
which forms the upper part of Town Hill, is mainly a massive
conglomerate, while in the western area, extending across the
county from Pennsylvania to West Virginia along the eastern
face of Alleghany Front, the lower part of the formation is a
coarse-grained, grayish-green, micaceous sandstone. Near the
middle are shales containing fragments of plants, and the upper
part is a grayish-green or reddish-green, micaceous, flaggy sand-
stone with some interbedded shales of various colors. Some of
the layers are cross-bedded and others are conglomeratic. The
thickness varies from 400 to 450 feet, and fragments of plants
are the only fossils noted with the exception of a few shells
apparently from this formation which were found nine miles
northeast of Oakland, Garrett county. The formation is named
from Pocono plateau in northeastern Pennsylvania, and is No.
X of the Pennsylvania reports.

Greenbrier limestone.— This formation crosses the county as a
narrow band along the eastern face of the Alleghany Front from
Pennsylvania to West Virginia. The best exposures are in Jen-
nings Run above the railroad water-tank ; and on the north bank
of the Potomac River below the mouth of Stony Run, as well as
in the lower part of the run, two miles below Westernport. The
lower part of the formation is composed largely of bluish-gray,
arenaceous limestone, the middle of red and olive shales and the

PALEOZOIC FORMATIONS OF MARYLAND 423

upper part of massive bluish-gray limestones and calcareous
shales. The limestone is valuable for road-metal and in Garrett
county is quarried and burned to a considerable extent for quick-
lime which is used as a fertilizer. The thickness varies from
200 to 250 feet and the upper part is quite fossiliferous. The

Fic. 3.—Greenbrier limestone on western bank of Youghiogheny River, south-
west of Oakland, Md.

formation is named from Greenbrier county in southeastern
West Virginia where it reaches a thickness of 1000 feet or more.

Mauch Chunk formation.— This, like the preceding formation,
crosses the county along the eastern face of Alleghany Front
but the band is broader, covering the upper part of the mountain
slope. The rocks are mainly red arenaceous and argillaceous
shales and sandstones, but a little above the middle of the forma-
tion is about 100 feet of reddish thin-bedded standstones. At

424 CHARLES S. PROSSER:

the top of the formation, as shown in the Cumberland and
Pennsylvania R. R. cut east of Barrelville, is a greenish zone five
feet thick composed partly of sandstone and partly of a calcare-
ous breccia containing clay pebbles. The thickness of the for-
mation is 650 feet. In the lower part of the shales and along
the Greenbrier-Mauch Chunk contact are numerous excellent
springs. The formation is named from Mauch Chunk in eastern
Pennsylvania and is the Canaan formation of the Piedmont folio.
The Greenbrier limestone and Mauch Chunk shales taken
together are known as No. XI of the Pennsylvania reports.

Pottsville formation—The preceding three formations are
usually grouped together as the sub-Carboniferous or Lower
Carboniferous, and the Pottsville is classed as the oldest of the
Carboniferous proper or Upper Carboniferous formations. It
crosses the county from Pennsylvania to the Potomac, in general
forming the crest line of Alleghany Front although in the north-
ern part it is lower, and extends up the Potomac valley to above
Westernport. At .numerous places near the crest line of Alle-
ghany Front it forms conspicuous cliffs. It also occurs at the
northwestern corner of the county. The formation is composed
of massive light gray sandstones with some conglomerate strata
and thin-bedded gray sandstones and shales. Some of the
shales are black and there are several thin beds of coal. The
most important are first, the Bloomington, which is exposed along
the Baltimore and Ohio Railroad west of Piedmont, W. Va.,
commonly known as the Railroad seam, varying in thickness
from less than two to more than three feet and occurring about
150 teet below the. top of the formation. The second is the
Westernport seam about two feet thick which occurs below the
Homewood sandstone near the top of the formation. Some of
the sandstones are suitable for building stone. The thickness is
estimated as between 450 and 500 feet and there are fragments
of fossil plants. The formation is named from the exposures of
massive conglomerate in the vicinity of Pottsville in eastern
Pennsylvania. It is the Blackwater formation of the Piedmont
folio and No. XII of the Pennsylvania reports.

PALEOZOIC FORMATIONS OF MARYLAND 425

Allegheny formation.— This is the first of the Coal-measure
formations and in general covers the western slope of Alleghany ~
Front extending from Pennsylvania to the Potomac valley and
up it into Garrett county. It also extends up George’s Creek

valley from Westernport to the vicinity of Morrison’s, one mile

Fic. 4.—Block of Pottsville conglomerate, at side of National Road on top of
- Meadow Mountain, Garrett Co.

below Barton; and occurs in the northwestern corner of the
county. The rocks consist of massive to thin-bedded grayish to
olive sandstones, gray and black shales and beds of fire-clay and
coal. At Barrelville and on the northern part of Dan’s Moun-
tain two coal beds occur near the base of the formation which
have not been found near Westernport. The lower one, called
the ‘“Bluebaugh” (Brookville) coal, varies in thickness from
24% to 4% feet and 30 feet or more above it is the ‘“ Parker”’

Depo WPI eH Oo

426 CHARLES. S.PROSSER

(Clarion) coal with a thickness of two feet which was reported
by the miners to reach 4% feet. At Westernport nearly 100
feet above the base is animpure coal, called the ‘‘ Split six-foot”’
by the miners, showing a thickness of 4 feet, and 130 feet above
the base is the most valuable coal seam of this formation, the
“Davis” (Lower Kittanning), commonly known as the ‘“Six-
foot,” with a thickness of about 5 feet in the lower George’s
Creek valley. Nearly 170 feet above the Davis seam is the
‘*Thomas ”’

formation, and from its general thickness in the George’s Creek

(Upper Freeport) coal, which forms the top of the

valley is known as the ‘“‘Three-foot’’ seam. Fossil shells have
been found in the black or bluish shales at a few localities. The
thickness of the formation is about 300 feet. It is named from
the exposures on the Allegheny River in western Pennsylvania,
is very generally called the Lower Productive-measures, is No.
XIII of the Pennsylvania survey and includes the Savage forma-
tion and lower part of the Bayard of the Piedmont folio.
Conemaugh formation.—The area south of the Pennsylvania
line to a parallel line passing through Little Alleghany and to
the west of the foot of the western slope of Little Alleghany
and Piney mountains is largely covered by this formation. Then
it extends parallel to Dan’s Mountain, to the southwestern part
of the county and covers a large portion of the steep slopes of
the hills bordering the lower George’s Creek valley, continuing
up the valley to Ocean. It also appears in the upper part ofa
number of the small valleys along the western border of the
county. The lower part of the formation, representing the
Mahoning sandstone, is frequently a massive gray sandstone with
bands of yellowish shales reaching a thickness of about 100
feet. In the upper part of this sandstone, about eighty-five feet
above the Thomas coal, is a coal seam about two feet in thick-
ness underlain by astratum of fire clay. The succeeding rocks
are grayish to brownish sandstones and yellowish to gray and
black arenaceous and arigillaceous shales with beds of coal and
fire clay. In some of the localities there are quite massive gray
to brownish-gray sandstones near the middle and top of the

PALEOZOIC FORMATIONS OF MARYLAND 427

formation. From 225 to 230 feet above the Thomas coal and base
of the formation and about 400 feet below the Elkgarden coal isa
coal seam with a general thickness of four feet, but varying from
three to nearly five feet, named the Barton (Four-foot) coal,
and worked to a considerable extent in the vicinity of that town.
A thin seam nearly one foot thick in outcrop occurs about 440
feet above the Thomas coal or 220 feet below the Elkgarden,
while approximately 500 feet above the Thomas coal and 150
feet below the Elkgarden coal is a zone composed of alternating
shales and impure coal, varying in thickness from seven to ten
feet, known locally as the ‘ Dirty Nine-foot’’ and called the
“Franklin ’’ coal. In the Lonaconing section, thirteen feet
below the base of the Franklin coal are nearly three feet of coal
and black, thin shale. There are also two or three impure lime-
stone strata and some irregular beds of iron ore. The forma-
tion is clearly defined by the top of the Thomas coal at the
base and the base of the Elkgarden (Pittsburg) coal at the top,
the thickness varying from 600 to nearly 640 feet. A few
invertebrate fossils have been found, principally on the bank of
George’s Creek at Barton, and fossil plants in the black shales.

The Conemaugh formation was named from the Conemaugh
River in western Pennsylvania, is frequently called the Lower
Barren-measures, is No. XIV of the Pennsylvania reports, is the
upper part of the Bayard and Fairtax formation of the Piedmont
folio, and the Elk River series of West Virginia.

Monongahela formation.—Vhis formation, south of an east and
west line passing through Little Alleghany, covers the larger
part of George’s Creek valley as far south as Ocean and most of
the area west to the county line. To the north of Little Alle-
ghany, two high hills are capped by it. From Ocean to Lona-
coning the upper part of the steep hills bounding the George’s
Creek valley are in the Monongahela, which also caps most of
the highest hills as far south as Hampshire to the northeast of
Westernport. The rocks consist largely of light gray to black
shales with some grayish sandstones which form occasional
massive strata. There are also several dark colored limestones,

428 CHARLES S. PROSSER.

bands of iron ore, and beds of coal. The Elkgarden (Pittsburg)
coal, the noted seam of western Maryland, and known locally as
the ‘‘ Big Vein” or “ Fourteen-foot’’ seam, occurs at the base of
this formation. The main mass of coal varies in thickness from
ten to nearly fourteen feet, above which are frequently from
three to nine feet of alternating coal and black shale which, in
the southern part of the George’s Creek field, is capped by
twenty-five feet of thin black shales in which coal occasionally
occurs.

A seam of coal two and one half feet in thickness is reported
in the Consolidation Coal Company’s new shaft 92 feet above
the> base of “the: Elkgarden,{coalt = Mrom (120) ton 140mfect
above the top of the Elkgarden is the Tyson (Sewickley)
coal varying in thickness from three to seven feet. Finally, at
the top of the formation about 255 feet above the base of the
Elkgarden coal is the Koontz (Waynesburg) coal two feet thick
and reported to reach a thickness of four and one half feet. The
top of this coal determines the upper limit of the Monongahela
formation which has a thickness of a little more than 250 feet.
Fossils are rare.

The Monongahela formation was named from the exposures
along the Monongahela River in southwestern Pennsylvania,
is popularly known as the Upper Productive-measures, is No.
XV of the Pennsylvania reports and the Elkgarden formation of
the Piedmont folio.

PERMIAN STRATA (?)

Dunkard formation.— The largest area of this formation
partly underlies the city of Frostburg and covers a considerable
tract to the east and southeast of the city. It covers the high
part of several hills to the south of Frostburg, extending as far
south as Detmold Hill on the western side of George’s Creek and
the hill south of Pekin on the eastern side. The rocks consist
largely of argillaceous shales, which when weathered are reddish-
green, with some beds of sandstone, limestone, and coal. A
stratum of coal and black shale four feet thick oceurs 120 feet
above the base of the formation and a drab limestone, five feet

PALEOZOIC FORMATIONS OF MARYLAND 429

thick, weathering buff occurs about 295 feet above it. This
limestone, some of the layers of which contain plenty of speci-
mens of Ostracods (Primatia frostburgensis Jones) and a few
other species, has been quarried to some extent toward the top
of Vale’s Hill, east of the Consolidation Coal Company’s pumping
station, and it is succeeded by thin black shales. The top of
the hill is ninety feet higher and on its slope forty feet above the
limestone ledge are loose pieces of bluish thin-bedded limestone
containing small Ostracod and Gastropod shells. Near the top
are loose blocks of coarse-grained sandstone which probably caps
the hill. This hill shows about 390 feet of the Dunkard forma-
tion, which is probably its greatest thickness in the county. In
addition to the fossils noted in the limestones, ferns were found
in the shales overlying the Koontz coal.

The formation is named from the exposures along Dunkard
Creek, near the West Virginia-Pennsylvania line, is frequently
called the Upper Barren-measures, and is No. XVI of the Penn-

sylvania report. ;
CHARLES S. PROSSER.
COLUMBUS, O.,

July 1901.

THE DEPOSIMION OR COPREER =BYo SOLUMONS: OF
PE RROUS SAEmS

INTRODUCTORY

THE genesis of the great deposits of native copper is a sub-
ject which has invited considerable speculation. From its occur-
rence in the cupriferous conglomerates and sandstones as a
cement, or replacer, of the constituent grains, the copper is
plainly of secondary origin. Its position here indicates that it
resulted as a deposition from aqueous solution.’ The metal,
doubtless, first dissolved as sulphate, an oxidation-product of an
original sulphide. Under the influence of solutions of calcium
bicarbonate, or alkaline silicates, the sulphate was speedily con-
verted into the carbonate, or silicate.? From solutions of one,
or both, of these latter salts were probably derived the several
classes of copper deposits.

In its paragenetic relations the position of the metallic copper
indicates, as Pumpelly shows, that it was deposited after the forma-
tion of the non-alkaline silicates and before that of the alkaline
silicates. As is further pointed out by the same writer, there is
an intimate connection between the native copper and such iron-
bearing minerals as delessite, epidote, and the green earth
silicates. So constant is their common occurrence that he is
led irresistibly to the conclusion that there is a close genetic
relation between the reduced copper and the ferric oxide con-
tained in the associated minerals; that, indeed, the reduction of
the copper oxide to metallic copper was produced by the oxida-
tion of ferrous derivatives. Later Irving, in support of the same
view, called attention to the fact that many particles of copper
enclosed a central core of magnetite.3

™R. D. IRviNG: Mon. U.S. Geol. Surv., No. 5 (1883), p. 420.

? RAPHAEL PUMPELLY: Am. Jour. Sci., Vol. II (1871), p. 353.

3Mon. U.S. Geol. Surv., loc. cit.

430

LHE DEPOSITION OF COPPER 431

A number of attempts has been made to discover the condi-
tions under which copper may be deposited by solutions of ferrous
salts. As early as 1861, Knop succeeded in forming cuprous
hydroxide by treating a mixture of cupric and ferrous sulphates
with alkaline carbonate.t In one instance he speaks of obtaining
traces of copper. In 1864, Wibel repeated Knop’s experiments,?
but was unable to verify the latter's statement regarding the
reduction to metallic copper. When, however, a mixture of fer-
rous and cupric hydroxides, formed by adding potassium
hydroxide to a solution of the sulphates, was heated to 210° C.,
traces of copper were obtained. Solutions of the sulphates and
coarsely powdered Wollastonite, subjected to the same treat-
ment, yielded a like result. The separation of metal, however,
was slight and the part played by the ferrous hydroxide in the
reduction was at the time somewhat questioned. In 1867, Braun
observed the partial deposition of copper from a mixture of
ferrous and cupric salts when these were dissolved in large excess
of ammonium carbonate.3 The same year Weith secured a ready
reduction in the presence of tartaric acid.+ He failed to note, how-
ever, that under the same conditions, the organic acid itself will
slowly reduce the copper salt. A mixture of calcium hydroxide
with a solution of ferrous and cupric sulphates, was allowed to stand
for several weeks. The precipitate thus obtained, when treated
with acetic acid, left a residue of cuprous oxide and copper.

Strangely enough Weith overlooked the fact that cuprous
oxide with acetic acid yields cupric acetate and metallic copper.
In 1869, Hunt stated that he had obtained metallic copper by the
action of cupric chloride on freshly precipitated ferrous hydroxide,
or carbonate. Nothing is given, however, to indicate that the
metal was actually detected by isolating it from co-precipitated
ferric hydroxide.

tA. Knop: N. Jahrb. f. Min. (1861), S. 513.

?FERD. WIBEL: “ Das Gediegen-Kupfer und das Rothkupfererz”’ (1864), S. 14.
3E. BRAUN: Zeit. fiir Chem. (1867), 569.

4W. WEITH: Zeit. fiir Chem. (1867), 623.

5STERRY Hunt: Comp. r. (1869), 1357.

432 Jil (Co JEG QYOY Ld,

THEORETICAL

It can be shown that the deposition of copper by solutions
of ferrous salts is a reversible reaction governed by the ordinary
laws of chemical equilibrium. It is to Arrhenius largely that we
owe the view that certain substances in solution are more, or less,
dissociated into electrically charged parts, or ions. This theory
has proved of the highest value in affording an insight into
the principles of chemical reacticns. Different substances differ
much in their tendency to pass into the ionic form and this
tendency is greatly influenced by external conditions, particu-
larly by the nature of the solvent.

The chief source of ions is the dissociation of electrically
neutral molecules, such as occurs in the aqueous solutions of
salts, acids, and bases. They may further be formed from elec-
trically neutral substances which enter the ionic condition by
partially, or wholly, appropriating the electric charge of ions
already present. As an example of this mode of formation
may be mentioned the reduction of ferric salts by the action of
metallic iron. _ 2Fe@ly > Fe = 3heCl The solution jor metallic
copper in ferric chloride is an action of the “same-nature,
Cu], 2heCl — CnC azine:

As is seen the deposition of copper by a ferrous salt would
be the reverse of this last reaction.

The conditions under which such reduction should occur
may be readily determined. In a system which contains a solu-
tion of iron and cupric salts in contact with metallic copper and
in which the several constituents have attained a constant value,
a condition of equilibrium subsists on the one hand between fer-
ric, cuprous, ferrous, and cupric ions (He) Cu zeae te Cua),
and on the other hand between ferric, ferrous, and cuprous ions
and the active mass of the metallic copper (Cu-++ Fe === Fe’
+ Cu). Ifa, b,c, d, respectively, represent the active masses

of the ions in the first instance, an equation of equilibrium may
ab
ca
™F. W. KUSTER: Zeit. f. Elec. Chem., Vol. IV, p. 105.
?,W. NERNST: Theoretical Chemistry, p. 358.

be thus formulated, —.— 7) Bhe active;madss ot thescoppes

THE DEPOSITION OF COPPER 433

is of constant value, hence in the second instance, retaining the
; =k . The precipita-
tion of copper would then be favored by increasing the concen-
tration of ferrous, cuprous, and cupric ions and by decreasing
that of the ferric ions. The deposition of the metal, conse-
quently, should depend on the relative active masses of the ions
present. This assumption is fully sustained by the experimental

same letters, we have the expression

evidence which follows:

The precipitation of metallic copper by solutions of ferrous salts ts
a reversible action, whose direction in any case ts determined by the
relative concentration of the ferrous, ferric, and copper (cuprous and
cupric) tons.

EXPERIMENTAL *

(a) Ina solution containing an appreciable quantity of ferric
ions, or in which these would be formed in the course of the
reaction, metallic copper will not be deposited.

This is shown in the inability of ferrous chloride, or sulphate,
to reduce corresponding copper salts, even though the mixed
solutions stand for an indefinite period. This inaction is, indeed,
to be expected when we consider that solutions of soluble ferric
salts, as the sulphate and chloride, easily dissolve metallic cop-
per with formation of a cupric salt.

From this it is readily understood why Wibel? obtained no
reduction of the copper salt on heating together to 210° C.a
solution of the mixed sulphates.

(0) Ina solution containing few ferric ions and in which the
reaction does not result in their appreciable increase, a sufficient
concentration of ferrous and copper ions will result in the depo-
sition of metallic copper.

The tendency of ferrous to reduce copper salts is shown in
the precipitation of cuprous sulphocyanate by the action of the
ammonium salt on a solution of ferrous and cupric chlorides.
The same tendency appears in the formation of cuprous chloride,

tAllthe reactions given in the following paragraph have been experimentally
determined by its author except as indicated by references.—Ed. Jour. GEOL.

2“ Das Gediegen-Kupfer und das Rothkupfrerz,” S. 20.

434 Jabs (Gs JI QILIS,

as noted by Hunt, on heating cupric oxide with a solution of ferrous
chloride.* Fromanemulsion of ferrous and cupric hydroxides, after
long standing, may be separated crystals of cuprous oxide. That
further reduction is largely determined by the concentration of
the cupric and ferrous ions, appears probable from the action of
ammonium carbonate on a solution of ferrous and cupric chlo-
rides,?- The precipitate first formed on adding the carbonate to
the mixed chlorides, dissolves in an excess of the precipitant to a
yellow liquid, from which, on standing twenty-four hours, there
is deposited, on the walls of the vessel, a slight but brilliant mir-
ror of metallic copper.

The influence of the concentration of the ions on the reduc-
tion of copper is clearly shown in the behavior of the mixed
carbonates under varying conditions.

When one adds a solution of cupric and ferrous chlorides (1
mol. CuCl,: 2 mol. FeCl, ) to a considerable excess of sodium
carbonate, there is obtained a greenish precipitate of the carbon-
ates which undergoes but slight change on standing. Such a
solution, indeed, would naturally little favor the separation of
metallic copper since the highly ionized alkaline carbonate would
greatly decrease the active masses of the ferrous and cupric
ions.

If the alkaline carbonate employed be only slightly in excess
of that required to precipitate the copper and iron salts, the con-
centration of the carbonic acid ions will be greatly diminished.
Under these more favorable conditions reduction slowly takes
place with loss of carbon dioxide. The carbonates gradually
change in color to a brick-red precipitate containing metallic
copper and basic ferric carbonate.

As is well known the acid ferrous and cupric carbonates are
more soluble than the corresponding normal or basic salts of
these metals. The influence of this greater solubility is shown
in the precipitation of small amounts of copper even in the pres-
ence of large excess of acid alkaline carbonates. If one adds

*STERRY Hunt: Comp. r., loc. cit.

2E. BRauN: Zeit. ftir Chem., loc. cit.

LHL DEPOSITION OF COPPER 435

the metallic chlorides to a saturated solution of acid potassium
carbonate and allows the mixture to stand for twenty-four hours
there is deposited on the walls of the vessel a slight film of
metallic copper mixed with basic ferric carbonate.

The solubility of the acid carbonates of iron and copper is
largely increased under pressure.t. The greater concentration
of the metallic ions thus obtained produces a ready reduction
of copper even in the presence of concentrated solutions of the
acid alkaline carbonates.

In a thick-walled flask holding a saturated solution of potas-
sium bicarbonate is placed a tube containing a solution of fer-
rous and cupric chlorides. The flask is then filled with carbon
dioxide, tightly sealed, and the contents of the tube mixed with
the alkaline bicarbonate. The precipitate formed gradually
loses carbonic acid, finally assuming the brick-red color already
noted. The supernatant ruddy liquid owes its color to the pres-
ence of some basic ferric carbonate which, when the solution is
warmed, deposits as ferric hydroxide. The precipitate contains
finely divided copper which cannot readily be freed from the inti-
mately associated ferric iron by treatment with hydrochloric acid
because of the solvent action of ferric chloride, but on digesting
the original mixture a short time a coagulum may usually be
formed from which, by repeated agitation with water, the heavier
metal is separated. The copper thus obtained is of characteristic
appearance; is insoluble in hydrochloric acid, soluble in nitric,
the solution showing the presence of copper and absence of
iron.

The reduction of copper from the cuprous condition may
be effected in the same manner as from the cupric. This is
readily shown by substituting for the cupric salt in the above
reaction cuprous chloride dissolved in a solution of chloride of
sodium. From one gram of cuprous chloride more than half
the metal may be easily isolated as pure copper free from ferric
iron.

From these results it is quite evident that the conditions

™R. WAGNER: Zeit. f. analyt. Chem. Vol. VI, p. 167.

436 Jak (Gp JEL COYOTE,

under which the oxidation of ferrous salts may result in the
deposition of copper are those which obtain in the circulation of
underground waters. The theory of Pumpelly and others based
on paragenetic relations is thus fully sustained by chemical evi-

dence.
He G2 BipprEe:
UNIVERSITY OF CHICAGO,
June 1901.

EVIDENCE OR SALOCAL SUBSIDENCE. IN. THE
INTERIOR

In the spring of 1883, I made a survey to build a levee along
the Wabash River on the west side of Parke county, Indiana, for
a length of twelve miles. I took the levels with great care, and
checked on the river water every half mile to guard against
errors. The great flood of the preceding winter had left its high
water mark very plain on the trees inthe bottoms, and I checked
on them also. I cut some sixty bench marks on the trees in
running the levels, some of which are still intact. The lower
end of the levee was built square across the narrow bottom to
the bluff and crossed a bayou through which the flood water ran
off of the bottoms into the river. We built an automatic flood-
gate across this bayou so as to shut out river, but let out inside
water from breaks above. The gates were hung to heavy brick
walls built on timber foundations three feet thick, and deeply
bedded below the bottom of the bayou. A bench mark was
cut on a bur oak tree near the walls, and the level of the
walls was taken when built. I had charge of the maintenance
and repair of this levee four years from its building, and had
frequent occasion to run the level over the top to restore breaks,
for it was built only twenty-one feet above low water, whereas
the great floods rise twenty-eight feet. I set the grade stakes
for the contractors to work to, and in doing so ran the level
over the ground again. I speak of all this to show that my
leveling was correct, as so many levelings would detect any error,
and none were found to exceed a halfinch. I can say positively
that the levels were correct in 1883.

This spring (1901) the levee was to be raised three feet, mak-
ing it twenty-four feet above low water, under a new law of
the state, but including only the lower seven miles. I leveled
the work again, and found bench marks again intact except the

437

438 JOHN T. CAMPBELL

lower (south) mile and a quarter, which showed a decline south-
ward amounting to ten inches at the lower (south) end, as shown
by the mark on the bur oak and top of the gate walls. I went
back to the C. & E. I. railroad bridge at Clinton, two and a half
miles above the south end, and started my level from a mark
known to be in tally with the level of 1883, and ran carefully
over the work again, and it varied from the one made just before
only a quarter of an inch. And the bench mark on the bur oak
and the top of the gate walls had gone down ten inches (,83, of
afoot). I was right in 1883 andIlam right now. What caused
this sink, or subsidence? I can think of nothing so likely to
cause it as the Charleston earthquake. The wave of that earth-
quake somewhere south of us changed from westward and went

northward along the Wabash.
Joun T. CAMPBELL.
ROCKVILLE, INDIANA,
July 20, 1901.

Ti GORI

WitnH the death of Dr. Joseph Le Conte there has passed
away perhaps the last distinguished American representative of
the general geologist as typified during the past century. This
passing type of the general geologist was a distinctive outgrowth
and representative of a transitional stage of intellectual pro-
cedure —a passage from the former mode in which the general-
izing and philosophical factors held precedence and the toilsome
modes of scientific verification followed as their servitors, to the
present or at least the coming method in which scientific deter-
minations are the basal factors to which generalizations and
philosophies are but dependent accessories. We owe much of
the transition itself to Dana and Le Conte, the two noblest
American representatives of the passing type, for while they
grew up under the influence of the older intellectual attitude,
they grew out of it in spirit while they steadied and guided the
transition. They were distinctively students of geology in the
special sense in which that term implies the organized doctrine
of the earth, rather than students of what might be termed gezcs,
the immediate study of the earthitself in the field and the labora-
tory. They were preéminently students of the accumulated
data and of the literature of the science, with generalization and
philosophic inference as their dominant inspiration. Neither
Dana nor Le Conte were eminently field students; much less
were they specialists in a chosen field of the broad geological
domain. Their point of view was that of the organizer and of
the philosopher, and the contribution they made in their chosen
sphere was indispensable and immeasurably valuable. How this
necessary function is to be met in the future, with the increasing
complexities and profundities into which every branch is rapidly
growing, it is difficult to foresee, further than that it must in
some way be intimately associated with extensive personal
researches in the field and the laboratory, and must be guided

439

440 EDITORIAL

by areversal of the old-time attitude of philosophy and science
toward each other. The philosophical factor must be put into
service as the active handmaid of scientific determination rather
than as its guide and leader. It may indeed go before as scout
to roughly reconnoiter the way, and it may come after to
assemble and interpret the results, but it must ever be tentative
and dependent on rigorous scientific determination. Deduction,
inference, interpretation, theory, hypothesis, and the other phil-
osophical factors must be merely initial steps and sequential
steps attendant on rigorous science as the end. None the less,
the philosophical factors and the philosophical point of view are
indispensable if the science is to make its most wholesome
progress, and we owe to Le Conte and to those he typifies an
immeasurable debt, for they have kept us in fresh touch with
the generalizations and the philosophy of the science, and have
inspired us with their own contributions to the broader concep-
tions of geology and of its relations to kindred sciences. The
writings of Le Conte are graced by the fruits of wide learning,
a lucid style, a genial attitude, and a candor that has called
forth universal love and admiration. TAGLC:

THE progress of opinion in regard to the origin of the solar
system, and incidentally of the earth, is indicated by the follow-
ing recent utterances of astronomers of high rank:

This simple hypothesis (Laplace’s nebular hypothesis) has recently been
severely attacked, and it is doubtful whether it will survive the blow. Indeed,
we may be compelled to seek the origin of stellar systems in the spiral
nebulae, which: Keeler’s photographic survey made just before his death
showed to represent a true type form. It is evident that much remains to
be done before the mystery which surrounds the genesis of stars can be
cleared away.— PROFESSOR GEORGE E. HALE, Director Yerkes Observatory,
in address to Visiting Committee, University Record, June 28, 1901, p. 141.

Though, without doubt, the system was evolved in some way from a
primitive nebula, we may say with certainty that it did not follow the orderly
course marked out for it by Laplace.-— PROFESSOR C. L. DOOLITTLE, of the
University.of Pennsylvania, in annual address delivered before the Univer-
sity of Pennsylvania chapters of the Society of Sigma Xi, June 13, Igor,
printed in Sczence, July 5, 1901, pp. 11-12.

REVIEWS

SUMMARIES OF CURRENT NORTH AMERICAN PRE-CAM-
BRIAN LITERATURE.’

WaLcotTr?’ reports on the results of an examination of Cambrian
and pre-Cambrian formations on Smith’s Sound, Newfoundland, during
the summer of 1899. At Smith point he found the Olenellus fauna
369 feet below the summit of the Etcheminian, and one of its types,
Coleoloides typicalis, in the basal bed of the Cambrian, on the south
side of Random Island. ‘This retains the Etcheminian of Newfound-
land in the Lower Cambrian.

The Random terrane, so-called from a typical section on Random
Sound, is a series of sandstones, quartzitic sandstones, and sandy shales,
resting conformably upon the Signal Hill conglomerate (which was
formerly supposed to represent the top of the Avalon or Algonkian
series) and extending up to the base of the Cambrian. The Random
terrane is thus the upper member of the Avalon series and fills a por-
tion, if not all, of the gap between the Signal Hill conglomerate and
the Cambrian. The Cambrian rests on the Random terrane with a
thin belt of conglomerate. The thickness of the terrane is probably
tooo feet. In one horizon in the terrane were found several varieties
of annelid trails, including a variety about 5 millimeters broad, a
slender form % millimeter broad, and an annulated trail 2 to 3 milli-
meters in width.

An examination of the form known as Aspidella terranovica found
in the Momable terrane of the Avalon series proved the supposed
fossil to be a spherulitic concretion, and this removes it from among
the possible pre-Cambrian forms of life.

Cushing? describes and maps the pre-Cambrian rocks of Franklin

*Continued from p. 87, Vol. 1X, this JOURNAL.

?Random, A Pre-Cambrian Upper Algonkian Terrane, by CHARLES D.,
WatcotTtT: Bulletin of the Geological Society of America, Vol. XI, 1900, pp. 3-5.

3Preliminary Report on the geology of Franklin county, Pt. III, by H. P.
CusHINnG: Eighteenth Annual Report of the State Geologist of the State of New
York, 1900, pp. 75-128. With geological map.

441

442 REVIEWS

county, New York. These are classified as Grenville (Algonkian)
rocks, igneous rocks intrusive in the Grenville, and other igneous
rocks of doubtful age, possibly in part older than the Grenville rocks.

The Grenville rocks occur in small disconnected patches sur-
rounded by intrusive igneous rocks. Some of them have such position
with reference to one another that they seem to represent remnants of
what were originally two continuous parallel N. E. to S. W. belts.
The characteristic rock of the series is the crystalline marble. ‘This is
intricately infolded with quartzose and hornblendic gneisses, and with
fine-grained granitic, syenitic, and gabbroic gneisses precisely like
gneisses which occur in other areas where no member of the Grenville
series is to be found.

The gneisses of undetermined age include granite, syenite, diorite,
and gabbro gneisses, together with intermediate varieties. They
occupy a very large area. If all these gneisses are igneous (as is
thought probable) there are three possibilities in regard to their age.

1. They may represent in whole or part a more ancient series than
the Grenville.

2. They may represent a somewhat later series intrusive in the Gren-
ville, but older than the great gabbro, syenite, and granite intrusions.

3. They may represent thoroughly foliated phases of these later
intrusions.

In Dr. Cushing’s present judgment they will be found to belong
partly under 2 and partly under 3, but more especially the former.

No rocks have been found in the northern Adirondacks which can
be shown to be older than the Grenville series, but in every case in
which the relations have been made out, the adjacent rocks show intru-
sive contacts with the Grenville rocks. On the other hand, the Gren-
ville is a sedimentary series and must have been laid down on some floor.

Younger than the Grenville rocks and for the most part younger
than the doubtful gneisses are a considerable quantity of igneous rocks
comprising gabbros (anorthosites) syenites, and granites. These again
occupy large areas.

In the northern portion of the county, Upper Cambrian rocks
overlie the pre-Cambrian rocks with unconformity.

Smyth * discusses certain features of recent work in the western

* The Crystalline Rocks of Western Adirondack Region, by C.H.SMyTH: Rept.

of the New York State Geologist for 1897, published in Fifty-first Ann. Rept. of New
York State Museum, Vol. II, 1899, pp. 469-467.

REVIEWS 443

Adirondack region. He concludes that the rock previously called
gabbro by Nason, Van Hise, Williams, and himself, south of the belt
of limestone in the Diana-Pitcairn area in Lewis and St. Lawrence
counties, is an augite syenite of igneous origin, although it passes into
a hornblende-gneiss which is unquestionably a result of dynamic action.
The origin of other gneisses has been inferred to be igneous from their
similarity to this gneiss which has been particularly studied, and it is
evidence of this kind which serves as a basis for Smyth’s conclusions,
previously published, that some of the gneisses on the western
Adirondacks are certainly, and most of them probably, of igneous
origin.

With the view of exploring the central and little known portion of
the Adirondacks, a reconnaissance was made through the area contigu-
ous to the Fulton chain of Jakes and Raquette Lake in the counties of
Hamilton and Herkimer. It was found that the heart of the Adiron-
dacks is made up essentially of gneiss, with minor quantities of crystal-
line limestone and its associated sedimentary gneisses and _ schists.
This is precisely analogous to what was found by the writer in St.
Lawrence, Jefferson, Northern Lewis, and southwestern Hamilton
counties, and by Kemp and Cushing in the eastern Adirondacks.
These facts lead to the conclusion that the Adirondack region, instead
of consisting of a great central mass of gabbro, surrounded by a nar-
row fringe of gneisses and limestones with quaquaversal dip, is essen-
tially composed of gneisses, with numerous limestone belts, having
northeast strike, and northward dip, and cut through on the east by
immense intrusions of gabbro. It is still possible, of course, that some
areas of gabbro may be found in the unexplored portions of the west-
ern half, but even should this be so, it would not materially modify the
above conclusion, as such masses must necessarily be isolated intru-
sions of no great extent, rather than parts of a large area.

Kemp,’ in connection with the description of the magnetite
deposits of the Adirondacks, briefly describes the general features of
the geology of the gabbro and gneiss of Westport, Elizabethtown, and
Newcomb townships in Essex county, New York, and presents a geo-
logical map of the former two townships.

tSee Summary Jour. GEOL., Vol. VII, p. 406.

2 The Titaniferous Iron Ores of the Adirondacks, by J. F. Kemp: Nineteenth Ann.
Rept. U. S. Geol. Surv., 1897-8, Pt. III, 1899, pp. 397-399.

MO REVIEWS

Kemp and Newland‘ make a preliminary report on the geology of
Washington, Warren, and parts of Essex and Hamilton counties, New
York. Some of the points particularly noted are:

The excessive mashing and granulation of the gneisses, giving them
in places semblance to quartzite. The greenish gneisses, consisting in
largest part of microperthite, were originally eruptive rocks. The dis-
covery is reported of quartzose gneisses or foliated quartzites which are
certainly metamorphosed sediments. They form notable areas along
the head of South Bay, Whitehall township. ‘Their presence indicates
the probable presence of a considerable series of clastic sediments.
The crystalline limestones themselves have been found in small expo-
sures over almost all of Warren county, and generally in the crystal-
line belt of Washington. They are most extensive in Newcomb and
Minerva townships of Essex, and to the south become thinner and
more scattered. So far as we have observed they are less common in
eastern Hamilton county. There is evidence to show that strati-
graphical relations can be proven and that anticlines and synclines can
be demonstrated.

Dikes of basic gabbro usually of moderate width, but lithologi-
cally like the larger masses in Essex county, have been met over a wide
area — in fact almost every township in Warren, but the basaltic traps
almost disappear.

Kemp’ summarizes the present knowledge of the pre-Cambrian
rocks of the Adirondacks. Most of the features have been covered in
previous articles. Attention is called to the distribution of the sedimen-
tary crystalline rocks, the Oswegatchie series (equivalent to the Gren-
ville series of Adams and perhaps the Huronian). These consist of
limestones, sedimentary gneisses, and quartzites. They occupy greater
area than has been supposed. The limestones are found chiefly in
the northwest, and the southeast or eastern portions of the Adirondack
area of crystalline rocks. ‘They are in small quantity, or altogether
absent in the northern portion, in the broad belt running from

‘Preliminary Rept. on the Geology of Washington, Warren, parts of Essex and
Hamilton counties, by J. F. Kemp and D. H. NEWLAND: Rept. of the New York
State Geologist for 1897, published in Fifty-first Ann. Rept., New York State
Museum, Vol. II, 1899, pp. 499-553.

?Pre-Cambrian Sediments in the Adirondacks, by J. F. Kemp: Vice Presiden-
tial address published in the Proceedings of the A. A. A. S., Vol. XLIX, 1900, pp.

157-184.

REVIEWS 445

northeast to southwest across the area, and along the southern and
southwestern border. On the northwest they are in extended and
comparatively broad belts, but in the eastern portion they appear in
many small and separated exposures, associated with some quartzites
and much greater amounts of characteristic gniesses, but greatly
broken up by igneous intrusions. ‘The quartzites thus far known are
in small quantity, but such as they are, they are found principally in
the eastern portions of the area, where the limestones are thinnest
and most scattered. From the presence of the quartzites it is inferred
that clastic sediments must have been present in larger amounts than
has heretofore been realized. On the east it has not been proven
that sediments form synclines pinched into the underlying gneissoid
rocks. On the contrary they seem to constitute low, dipping, flat
monoclines.

Comment.— The complex geology of the Adirondack crystalline
rocks is being rapidly worked out by Kemp, Smyth, Cushing, and °
others. The frequent brief papers issued by these geologists in nearly
all cases report some important advance in the solution of their prob-
lem. ‘The precise relations of these advances to the general problem
may not be clear to the average geological reader, too busily engaged
to follow the subject closely, and for such the general summary of
Adirondack geology given by Kemp ‘* will be of value.

In a previous comment? on Adirondack geology the state of
geological knowledge, as indicated by the literature on the subject
then available, was briefly summarized by the writer, and here atten-
tion will be called only to later developments. One of the most
interesting of these is the extension of the areas of pre-Cambrian
sedimentary and associated rocks, and the corresponding contraction
in the area of the great Adirondack gabbro. This was formerly
supposed to occupy the great central area of the Adirondacks with
the pre-Cambrian sediments and the associated gneisses around its
periphery. Recent work seems to show that the area is occupied by
gneisses, with narrow limestone belts, cut through on the east by a
number of immense intrusions of gabbro. Another advance is the
discovery of greater quantities of clastic sediment than have before

*Pre-Cambrian Sediments in the Adirondacks, by J. F. Kemp: Vice Presiden:
tial address published in Proceedings of the A. A. A. S., Vol. XLIX, 1900, pp.
157-184.

2JouR. GEOL., Vol. VII, pp. 410-411.

446 | REVIEWS

been realized in the eastern portion of the Adirondack area, where
the limestone is thinnest. The main problem of the region, the origin
of the gneisses, is as yet far from settlement. The tendency is, how-
ever, to ascribe to them an igneous origin, and to place them later
than the Oswegatchie series, in the areas where they have been most
closely studied.

Jones,‘ in connection with a description of Tallulah Gorge of north-
eastern Georgia, describes the crystalline rocks there occurring, and
gives a little sketch map showing their relations. They are called
pre-Cambrian.

Watson?’ describes the granitic rocks of the Piedmont plateau of
Georgia. Field and laboratory studies indicate that they are not all
contemporaneous in origin. Some of them are pre-Cambrian, while
others may possibly be later in age.

Adams 3 describes the Laurentian granitoid gneiss and granite of
the Admiralty group of the Thousand Islands, Ontario. The granitoid
gneiss is presumably derived by metamorphism from the granite. A
large exposure of crystalline limestone on Island No. 18 resembles in
all respects that of the Grenville series of the mainland adjacent.

Parks‘ describes the geology of the Moose River Basin in Canada,
including the Moose and Abitibi Rivers, tributary to James Bay. This
is an immense triangular area of which the apex is at James Bay, and
the base stretches from above Lake Abitibi to a point west of Kabina-
kagami. The southern and major portion of this triangular area con-
sists of Laurentian gneisses and granites crossed, by bands of Huronian
rocks. Along the Abitibi River, Huronian rocks, consisting of
altered diorites, pyrites, gray quartz schists, and some soft decom-
posed schists occupy the country to the south, extending as far north
as the head of the first long rapid on the Frederick House River. The
line of contact of this belt crosses the Abitibi below the Iroquois

‘The Geology of the Tallulah Gorge, by S. P. Jones: American Geologist,
Vol. XXVII, 1901, pp. 67-75.

2The Granitic Rocks of Georgia and Their Relationships, by T. L. WaTson :
American Geologist, Vol. XX VII, 1901, pp. 223-225.

3Notes on the Geology of the Admiralty Group of the Thousand Isiands, by
FRANK D. ADAMs: Can. Rec. of Sci., Vol. VII, 1897, pp. 267-272.

4PARKS, WILLIAM A.: The Nipissing-Algoma boundary, Eighth Rept. Ont. Bur.
Mines, 1899, pp. 175-204, with geological map; Niven’s base line, Ninth Rept. Ont.
Bur. Mines, 1900, pp. 125-142; The Huronian of the Moose River Basin, University
of Toronto Studies, Geol. Series No. 1, 1900, pp. 35, with sketch map.

REVIEWS 447

Falls. From this point to the Lobstick portage, Laurentian gneisses
and inica schists crop out occasionally. The narrow Huronian belt
from the Lobstick to the foot of the canyon or Long Portage, consists
mainly of augite-syenite, passing into gabbro to the north. Beyond
this portage Laurentian gneiss extends to the Devonian contact above
the Sextant rapids.

Coleman* gives a general account of a visit to all the iron and
copper regions of the Lake Superior country. For the ranges on the
United States side of the boundary no facts are given not found in the
published reports. On the Canadian side of the boundary the Michi-
picoten Range, the iron formation near Dog River, and the siliceous
iron ores of Batchawana Bay are described. In the Michipicoten
range the Helen mine in particular is referred to. In general, the rocks,
including the ore at this mine, have all the appearance of Lower
Huronian or Keewatin rocks, as in the Vermilion district, and not
those of the Upper Huronian or Animikie, as in the Mesaba.

Near Dog River are iron formation rocks similar to those extend-
ing northeast from Michipicoten bay. It is thought probable that
the two may connect.

The occurrence and relations of iron formation material northeast
from Michipicoten Bay and near Dog River are indicated on a sketch
map.

Coleman,’ as a result of an examination of the new Michipicoten
iron district, and the consideration of other iron formation areas in
Ontario, has collected facts which seem to throw some light on the
relative ages of the different areas mapped as Huronian on the north
shore. In the Michipicoten district iron-formation material, consist-
ing of banded ferruginous sandstones, cherts, and jaspers, standing
nearly vertical, extends from Little Gros Cap northeastward for twenty
miles ; then bending to the north and west it takes a westerly direction
for more than thirty miles. The width of the belt is but a few hun-
dred yards.

Sandstones of the same peculiar type occur at Little Turtle Lake,
east of Rainy Lake and near Fort Frances, on Rainy River, as well as
at the Scramble gold mine, near Rat Portage, on Lake of the Woods.

*COLEMAN, Dr. A. P.: Copper and Iron Regions of Ontario, by A. P.
COLEMAN. Report of the Ontario Bureau of Mines for 1900, pp. 143-I9QI.

? Upper and Lower Huronian in Ontario, by ARTHUR P. COLEMAN: Bull. Geol.
*Soc. Am., Vol. I, 1900, pp. 107-114.

448 REVIEWS

Thin sections of these rocks show the same polygonal shapes of the
grains of quartz, and more or less iron ore is associated with speci-
mens from each locality. It is very probable, then, that the same
horizon exists at points far to the west of Lake Superior.

Turning toward the east, specimens very like the jaspery varieties
of the Michipicoten iron range are found interbedded with ‘iron ores
near Lakes Wahnapitae and Temagami, between Sudbury and the
Ottawa River.

At Batchawana Bay at the southeast end of Lake Superior, a
siliceous rock with narrow bands of magnetite occurs, which is prob-
ably the equivalent of the Michipicoten rock.

If, as seems probable, these jaspers are the equivalents of the west-
ern Huronian sandstones, there is a definite horizon traceable from
point to point across the whole northern end of the province, a dis-
tance of more than six hundred miles. .

At anumber of places over this area conglomerates, containing
jasper, ferruginous sandstone or chert pebbles, probably derived from
the source above described, are known. Beginning at the west, some
of these conglomerates occur. as follows : on Shoal Lake, east of Rainy
Lake ; west end of Schist Lake; near Mosher Bay, at the east end of
Upper Manitou Lake; a mile east of Fort Frances on the Rainy
River; near Rat Portage; near the mouth of Doré River; in the
original Huronian area, north of Lake Huron, particularly the Thes-
salon area; on Lake Temiscaming.

It is assumed that the iron-formation material cannot be other
than Lower Huronian, and that the conglomerates must represent a
basal horizon of the Upper Huronian. The break between the Upper
and Lower Huronian thus represented is a most profound one, and
affords a good basis for the correlation of the Huronian formations. It is
further suggested that this great- unconformity may be the same as
that between the Upper and Lower Huronian formations on the south
shore of Lake Superior and in Minnesota.

Comment.—As stated by Dr. Coleman a number of the conglomer-
ates above mentioned have been regarded by Pumpelly, Irving, Van Hise,
and other United States geologists, as basal to the Lower Huronian— on
structural evidence. Dr. Coleman places them in the Upper Huronian
because they contain fragments of iron formation material which
are assumed to be Lower Huronian. According to the generally
accepted ideas of the number and relations of the pre-Cambrian

REVIEWS 449

iron bearing formations, this assumption is perfectly justified and
the conclusion follows as to the Upper Huronian age of the typical
conglomerates mentioned.

But, lately evidence has been accumulated pointing to a conclusion
of a rather radical nature. This evidence has been such that Van
Hise* in a general article on the iron bearing formations of the Lake
Superior country just published, describes ¢aree iron bearing forma-
tions, the Upper Huronian, Lower Huronian, and Archean. The
most important of the Archean iron bearing formations are the Ver-
milion and the Michipicoten.

Van Hise himself in his published articles on the pre-Cambrian
has persistently maintained the essentially non-clastic nature of the
Archean, and the post-Archean age of all the iron bearing formations
of the Lake Superior country. But new evidence on the subject,
secured principally during the past year, has been so decisive that he
has not hesitated to announce as proven the existence of an Archean
or Basement Complex iron-bearing formation.

If there is an Archean iron formation, to which the Michipicoten
and Vermilion iron formations belong, then Dr. Coleman’s argument
as to the Upper Huronian age of conglomerates containing iron for-
mation fragments is rendered ineffective, and the conclusions indicated
by the structural evidence that the great conglomerates and accom-
panying rocks above described are Lower Huronian must stand, until
decisive evidence to the contrary is found.

Grant’ describes and maps the Upper and Lower Keweenawan
copper-bearing rocks of Douglas county, Wisconsin. The Lower
Keweenawan appears in a broad belt running from northeast to south-
west across the county, widening toward the southwest, and in a small
belt cutting through the southeastern corner of the county. It con-
sists mainly of basic lava flows, associated with which, in the area in
the southeast corner of the county, are a few beds of conglomerate
composed of débris of the closely adjacent underlying rocks. The
Upper Keweenawan appears in a broad belt in the southeastern part
of the county between the two belts of Lower Keweenawan rocks. It

*The iron-ore deposits of the Lake Superior region, by C. R. VAN HIsE:
Twenty-first Ann. Rept. U. S. Geol. Surv., Pt. III, 1901, p. 322.

? Preliminary Report on Copper Bearing Rocks in Douglas county, Wisconsin, by
U.S. GRANT: Wisconsin Geological and Natural History Survey, Vol. VI, 1900, pp.
55+

450 REVIEWS

is a series of conglomerates, sandstones and shales. In a belt north
of the northern belt of Lower Keweenawan rocks, extending from these
rocks to the shore of Lake Superior, is the Lake Superior sandstone
(Cambrian). ‘This is either flat-lying or dips slightly toward Lake
Superior. The junction of the sandstone with the Lower Keweenawan
is marked by a fault, along which the Lake Superior sandstone has
been depressed, in some places probably as much as several hundred:
feet.

The Upper and Lower Keweenawan belts form a syncline, the axis
of which runs northeast and southwest through the center of the tract
underlain by Upper Keweenawan rocks.

While the Keweenawan rocks of this area are the same in kind and
age as are the productive copper-bearing rocks of Keweenaw Point,
the probable unproductive character of the Douglas county rocks is
intimated.

Alexander Winchell’ prefaces a detailed petrographical description
of certain phases of the gabbroid rocks of Minnesota with a brief
account of the general succession in structure of formations in north-
eastern Minnesota. This is essentially the same as given by N. H.
Winchell? in Volumes IV and V of the Minnesota State Survey. The
correlation of this succession with the succession determined by the
United States Geological Survey is discussed.

Comment.— Mr. Winchell’s ideas as to succession and structure
determined by the United States Geological Survey are naturally
derived mainly from Bulletin 86 of the Survey and from the “ Princi-
ples of Pre-Cambrian Geology” published in the Sixteenth Annual
Report of the Survey. However, since these reports have been issued,
the United States Geological Survey has done somewhat detailed field
work in northeastern Minnesota as a result of which the ideas of the
United States geologists on the succession and correlation have been
considerably changed. ‘The new conclusions of the Survey are briefly
outlined by Van Hise in the Twenty-first Annual Report. This paper
should be referred to by anyone reading Mr. Winchell’s discussion of
the correlation.

* Mineralogical and Petrographic study of the gabbroid rocks of Minnesota, and
more particularly of the plagioclastites, by ALEXANDER N. WINCHELL: American

Geologist, Volume XXVI, 1900, General part, pp. 153-162, with geological sketch
map of Northeastern Minnesota.

? See summaries, JouR. GEOL., Vol. IX, pp. 79-86.

REVIEWS A451

Van Hise and Bayley* describe and map the geology of a portion
of the Menominee iron district of Michigan.
The pre-Cambrian succession is as follows :

Po

( Hanbury slate.

Vulcan formation, subdivided

Upper Menominee - into the Curry ore-bearing

X

member, Brier slate, and
; Traders ore-bearing mem-

Algonkian - ~~
= | eo berm
Unconformity
-: :
Negaunee formation.
| : :
; Lower Menominee - ~ Randville dolomite.
L Sturgeon quartzite.
Unconformity
(. (Granites and gneisses, cut by
| | granite and diabase dikes.
Archean a SIH x 2 z ae
| | Quinnesec schists, cut by acid
{ | and basic dikes and veins.

In general the Algonkian rocks constitute a trough bounded on
the north by the Archean rocks.

The Archean.— The Quinnesec schists are dark green or black
basic schists and spheroidal greenstones, cut by large dikes of gabbro,
diabase, and granite, and by smaller dikes of a schistose quartz por-
phyry. These occur in two areas, one along the Menominee River to
the south of the Huronian rocks, and another in the west-central end
of the district.

Bordering the Algonkian trough on the north is a complex of
granites, gneisses, hornblende schists, and a few greenstone schists, all
cut by dikes of diabase and granite. This complex is called the
“Northern. Complex.:’ Most ofthe Archean rocks are igneous.
Although there is no evidence of this, some of the fragmental tuffs
may have been water-deposited. The Quinnesec schists and the
Northern Complex are called Archean because they resemble lithologi-
cally other areas of Archean rocks in the Lake Superior country, and
they both underlie the Algonkian series. The Northern Complex
underlies the series with unconformity. The Quinnesec schists have

*The Menominee special folio, by CHARLES R. VAN HIsE and S, W. BAYLEy:
Geological Atlas of United States, Folio No. 62, U. S. Geol. Surv., 1900.

452 REVIEWS

not been observed in contact, and hence the presence or absence of a
normal erosion unconformity cannot be inferred.

The Lower Menominee sertes.—'The formations of the Lower
Menominee series are observed only in the center and on the northern
side of the Menominee trough. The Sturgeon formation is composed
mainly of a hard white vitreous quartzite forming a continuous border
of bare hills bordering the Archean complex. At its base is a coarse
conglomerate made up of débris from the underlying Archean com-
plex. The belt is in general a southward dipping monocline with
dips varying from 25° to perpendicularity, although there are many
reverse dips to the north. Its thickness is placed at from 1000 to
1250 feet.

Above, the Sturgeon quartzite grades into the Randville formation
which is mainly a homogeneous dolomite interstratified with siliceous
or argillaceous layers. This formation appears in three belts. The
northern one is just south of the belt of the Sturgeon quartzite. The
central belt is on the north side of Lake Antoine for a portion of
its length, passes eastward between the Cuff and the Indiana mines,
and ends at the bluff known as Iron Hill in the east half of Sec. 32, T.
40 N., R. 29 W. The southern belt of dolomite extends all the way
from the western side of the sandstone bluff west of Iron Mountain to
the village of Waucedah, at the eastern end of the mapped area.
Structurally the northern belt of dolomite is a southward dipping
monocline, while the two southern belts are anticlines. The thickness
is not determined on satisfactory evidence, but is probably 1000 feet-
or more. The Randville formation is found, in a number of mines.
in contact with the basal formation of the Upper Menominee series
Here there is a coarse conglomerate in the basal part of the overlying
formation indicating unconformity.

The Negaunee formation, overlying the Randville dolomite, is
represented in the district by so few and so small outcrops that it is
mapped with the Vulcan formation. Its presence is inferred mainly
from the occurrence of abundant iron formation débris in the basal
conglomerate of the Upper Menominee formation, showing that the
Lower Menominee iron-bearing series must have been present. In
the Marquette district an iron-bearing formation (the Negaunee) occu-
pies an exactly similar stratigraphical position. :

The Upper Menominee sertes—The formations between the uncon-
formity at the top of the Lower Menominee and the unconformity at

REVIEWS 453

the base of the Lake Superior sandstone, are placed in the Upper
Menominee series. These occur in two great series, the Vulcan and
the Hanbury.

The Vulcan formation is unconformable above the upper part of
the Lower Huronian, which for most of the district is the Randville
formation, and unconformable below the Hanbury slate. For parts of
the district the Vulcan iron-bearing formation does not appear at all
between the dolomite and the slate and its absence is explained by the
unconformity between the Vulcan formation and the Hanbury slate.
The Vulcan formation embraces three members. ‘These are, from the
base up, the Traders iron-bearing member, the brier slate, and the
Curry iron-bearing member. ‘They are mapped as a single formation.
The principal area of the Vulcan iron formation is in the belt goo to
1300 feet wide, following the sinuosities of the southern border of the
southern belt of Randville dolomite. It is generally absent north of
the southern belt except at the east end where it appears at the Loretto
mine and eastward. The second important area of Vulcan iron for-
ination stretches off about five miles along the south side of the central
dolomite belt running north of Lakes Antoine and Fumee, and ending
somewhere about the east line of Range 30 West. At the east end of
the dolomite area the iron-bearing formation appears in the lean slates
at Iron Hill. The third stretch of country in which the iron-bearing
beds are to be expected is that which borders the northern dolomite
belt, but while pits have shown the existence of the formation here its
‘distribution is unknown. The other areas in which the Vulcan for-
mation may occur are those bordering the Quinnesec schists, but this
has not yet been determined.

The Traders member consists of ferruginous conglomerate, ferru-
ginous quartzite, heavily ferruginous quartzose slates and iron ore
deposits. The Brier member consists of heavy black ferruginous and
quartzose slates. ‘The Curry member consists of interbedded japsilite,
ferruginous quartzose slates and iron ore deposits. ‘The relations of
the Traders and Brier Hill members where there has been no disturb-
ance of the strata is that of gradation. Where there has been disturb-
ance, as in the vicinity of Norway, there has been a zone of differential
movement between the two, resulting in slickensides and brecciated
zones. Between the Brier slates and the Curry member there is gra-
dation.

The Vulcan formation is bent into folds of several orders of

454 REVIEWS

magnitude, the greater ones corresponding approximately to the folds
in the underlying Randville dolomite. The total thickness of the
formation is probably 600 to 700 feet.

The iron ore deposits of large size rest upon relatively impervious
formations, which are in such positions as to constitute pitching
troughs. A pitching trough may be made (a) by the dolomite forma-
tion underlying the Traders member of the Vulcan formations, (4) by
a slate constituting the lower part of the Traders member, and (c) by
the Brier slate between the Traders and Curry members of the Vulcan
formation. ‘The dolomite formation is especially likely to furnish an
impervious basement where its upper horizon has been transformed
into a talc-schist, as a eoldsagiiencs of folding and shearing between
the formations.

Unconformably above the Vulcan iron formation is the Hanbury
formation, which forms three large belts in the syncline of the older
rocks, and occupies a very large proportion of the district. The for-
mation comprises clay slates, calcareous slates, graphite slates, gray-
wackes, quartzite, ferruginous dolomite, and rare bodies of ferruginous
chert and iron oxide. The formation is much thicker than any of the
other formations of the district, but it is probably not thicker than
2000 Or 3000 feet.

Wilder* describes and maps the Sioux quartzites and quartz por-
phyries of Lyon county, lowa. No points concerning the stratigraphy
or age have been added to those already given by other writers.

Bain’ describes the geology of the Wichita Mountains. Gabbros
and porphyries of pre-Cambrian and probably of Archean age are
present. The gabbro is more prominent in the western portion of the
mountains, being especially well developed in the Raggedy Mountains,
and the porphyry is more common in the eastern part of the moun-
tains, being typically developed at Carrollton Mountain.

Matthews * gives a detailed petrographical description of the gran-
ites of the Pike’s Peak quadrangle of Colorado. ‘They are referred to
the late Algonkian period.

*Geology of Lyon county, by FRANK A. WILDER: Iowa Geol. Surv., Vol. X,
1899, pp. 96-108.

? Geology of Wichita Mountains, by H. Foster Bain: Bull. Geol. Soc. Am., Vol.
XI, 1900, pp. 127-144, Pls. XV-XVII.

3The Granite Rocks of the Pike’s Peak Quadrangle, by A. B. MATTHEWS: JOUR.
GEOL., Vol. VIII, 1900, pp. 214-240.

REVIEWS 455

Cross* maps and describes the geology of the Telluride quadrangle,
Colorado, and briefly sketches the geology of the San Juan region, of
which the Telluride quadrangle is a part.

In the Telluride quadrangle, along Canyon Creek north of Stony
Mountain, is a small body of upturned quartzites, with an intercalated
rhyolite sheet, which have been referred to the Algonkian. The
quartzites are coarse and grade into fine conglomerate.

Ancient granites, gneisses, and schists are known in the Animas
Valley and in the Uncompahgre plateau. These rocks have usually
been considered as belonging to the Archean, but some of them are
probably younger than the great series of quartzites exhibited in the
Needle Mountains to the south, and younger than the quartzites
beneath the volcanics in the canyons of the Uncompahgre, above
Ouray, which have been referred to the Algonkian. These quartzites
stand on edge or have been greatly disturbed. The relations of these
isolated exposures to contemporaneous formations elsewhere are
unknown. :

Spurr? maps and briefly describes the Archean? granite of the
Aspen district of Colorado. This is unconformably below and in
direct contact with sediments of upper Cambrian age.

Davis* in a general account of a trip through the Colorado Canyon
district briefly describes certain features of the pre-Cambrian geology.
He calls attention to the extraordinary evenness of the floor of schists
with granite dikes (Archean) upon which the Chuar and Unkar ter-
ranes (Algonkian) rest. The floor for the Paleozoic strata is somewhat
less regular than the floor for the Unkar. In two places the pre-Cam-
brian rocks rise higher than the basal Tonto (Cambrian) sandstone.
The Archean schists beneath the Unkar have a steep and regular slope,
indicating uniform resistance to erosion. Where, beneath the Tonto,
they show a bench, it is taken to indicate a softer character at
this point, probably due to a longer period of pre-Tonto weather-
ing.

* Telluride Folio, Colorado, by WHITMAN Cross: Geol. Atlas of the U. S., No.
57, 1899. .

* Geology of the Aspen Mining District, Colorado, with Atlas, by J. E. Spurr:
Mon. U.S. Geol. Sury., No. 31, 1898, pp. 1-4,

3 The term Archean is evidently used in the sense of pre-Cambrian.

4 Notes on the Colorado Canyon District, by W. M. Davis: Am. Jour. Sci., 4th
ser., Vol. X, 1900, pp. 251-259.

456 REVIEWS

Blake * refers to the Archean the thick layers of gneiss forming the
southern flank of the Santa Catalina Mountains, Arizona. The gneiss
is in flat layers representing beds. A part of it isaugen gneiss; other
layers are quartzose and seemingly quartzites.

Knight? in connection with the discussion of the artesian basins of
Wyoming gives a brief description, accompanied by a map, of the
geology of the state. Algonkian and Archean rocks are present. ‘The
Archean rocks consist mainly of granite, in places cut by dikes of
porphyry containing mineral ores, which can be seen in typical expos-
ure at Sherman, Laramie Peak, east of Whalen Canyon, along the Big
Horn, Wind River, Gros Ventre, Medicine Bow, Ferris, Seminoe, and
Owl Creek ranges, along the Sweetwater River, a few miles northwest
of Rawlins, and north of Clark’s Fork, in Big Horn county.

The Algonkian rocks are for the first time separated from the
Archean. They consist of schists in great profusion, marbles, and
quartzites, all cut with dikes of eruptive rocks.- They occur in granite
basins in unconformity with the Archean, and form important bands
in numerous localities. The strike of the series varies from north
to northeast and the dip of the strata is seldom less than 65-75”.
The thickness of the entire series has not been absolutely measured,
but including the eruptive band, which does not form an important
part, the maximum thickness in Wyoming is about 20,000 feet. Typical
areas have been found in the Black Hills in Wyoming, and occasional
outcrops from that place to the Hartville hills—one exposure being
east of Lusk, another at Rawhide Butte, and a large one in Whalen
Canyon. They also occur at Halleck Canyon, Plumbago Canyon, in
the Medicine Bow Mountains, nearly all of the Sierra Madre Moun-
tains, in the Seminoe Mountains and in the Sweetwater mining district
of the Wind River range. None of these localities have been examined
in detail; but sufficient work has been done to prove that these rocks
were at one time sedimentary, and that they have been changed by
metamorphism to schists. In the Sweetwater districts the rocks are
chiefly schists; but there are many bands of erruptive rock that form
dikes which follow the strike of the formation.

t Mining in Arizona, by WM. P. BLAKE: published in report of the Governor of
Arizona to the Secretary of the Interior, Washington, 1899, p. 142.

2A preliminary Report on the Artesian Basins of Wyoming, by WILBUR C.
KNIGHT: Wyoming Experiment Station, Bulletin No. 45, 1900. Part on pre-Cam-
brian, pp. I1I1-116. With geological map. This is the first geological map of Wyo-
ming that has appeared.

REVIEWS . 457

Weed* maps and describes the pre-Cambrian rocks in the Fort
Benton and Little Belt Mountains quadrangles of Montana.

The Archean rocks are found only in the Little Belt range in the
southwestern part of the Fort Benton quadrangle and in the north-
western part of the Little Belt Mountains quadrangle. They are
gneisses and schists of various kinds, and of somewhat uncertain
origin. They are, in part at least, of igneous origin, and none of
them show any traces of sedimentary origin. ‘Their relations to the
Algonkian rocks are those of unconformity. The Algonkian rocks
are found in the mountain tracts of the Little Belt range, in Castle
Mountain, and in the low range crossed by Sixteenmile Creek in the
southwest corner of the Little Belt Mountain quadrangle. They are
divided into the Neihart quartzite and the Belt formation,’ both of
which are parts of what Mr. Walcott has called the Belt Terrane.

The Neihart quartzite is a hard pink and gray quartzite forming
the base of the Belt Terrane for this area. It is found in the vicinity
of Neihart in the Little Belt Mountains. Its thickness is about six
hundred feet. The Belt formation consists mainly of slaty, siliceous
shales, but also contains interbedded limestone and quartzite. Fossils
found in this series (in the shales above the formation which Mr.
Walcott has named the Newland limestone member of the Belt Ter-
rane), represent the earliest forms of life yet known. Near Neihart
the Algonkian period is represented by 4000 feet of beds, while further
south and west the thickness is much greater.

Overlying the Algonkian rocks conformably are rocks containing
Middle Cambrian fossils. North of Neihart they rest directly on the
Archean.

Reconnaissance geological surveys in Alaska and adjacent portions
of British Columbia, by United States and Canadian government
parties, have shown the basal rock over considerable areas to be a
granite, which is provisionally assigned to the Archean.? Such granite

Fort Benton and Little Belt Mountains Folios, by WALTER HARVEY WEED:
Geol. Atlas of the U. S., Nos. 55 and 56, 1899.

See also Geology of the Little Belt Mountains, Montana: Twentieth Ann. Rept.
U.S. Geol Surv., 1898-9, Pt. III, 1900, pp. 278-284.

?The Belt formation includes the various lithological members of the Belt Terrane
which Mr. Walcott has named the Chamberlin shale, the Newland limestone, the
Greyson shale, the Spokane shale, and the Empire shale.

3 Usually in the sense of pre-Cambrian.

458 REVIEWS

is reported as occurring along the Pelly and Dease Rivers (Dawson *
and Hayes ;?) to the west, between the northern base of the St. Elias
Mountains on the Yukon River (Hayes?); along the Upper Tanana
River (Allen* and Brooks’), which is correlated by Spurr with the
granite along the Pelly River; along Fortymile Creek, a tributary of
the Yukon near the Canadian-Alaskan boundary (Spurr°); forming
the core of the Kaiyuh Mountains (described by Dall,’ referred to
Archean by Spurr®); possibly forming the core of the Alaska Penin-
sula and the Aleutian Islands (noted by Dall® and Purington,” referred

to Archean by Spurr").
C2keAerigiae

On Rival Theortes of Cosmogony. By the Rev. O. FIsHEr. ‘Amert-
can Journal of Sctence, June 1901, Pp. 414-422.

In this article the author has brought the current gaseo-molten
hypothesis of the origin of the earth into comparison with the hypoth-
esis of gradual accretion without a molten state recently advanced
by Chamberlin, and has endeavored to test the tenability of the newer
hypothesis by subjecting some of its fundamental postulates to mathe-
matical and physical inquiries. The author disclaims holding a brief
for either hypothesis and well sustains his claim to an impartial
attitude.

™GEOoRGE M. Dawson: Geological Natural History Survey of Canada, Vol. III,
Pt. I, 1887-8, p. 34B.

2C. WILLARD HAvES: Geographic Magazine, Vol. IV, 1892, p. 139.

SWocr cites pant 30;

4LIEUTENANT H. D. ALLEN: Expedition to the Copper, Tanana, and Koyukuk
Rivers, Senate Documents, Washington, 1897, p. 159.

5A. H. Brooks: Twentieth Ann. Rept. U. 5. Geol. Sury., Pt. VII, 1900, pp.
460-465.

6yJ. E, Spurr: Eighteenth Ann. Rept. U. S. Geol. Surv., Pt. Ill, 1898.. pp.
I 34-140.

7W. H. DALL: Seventeenth Ann. Rept. U. S. Geol. Surv., Pt. I, 1896, pp.
862, 863.

8J, E. Spurr: Twentieth Ann. Rept. U. S. Geol. Surv. Pt. VII, 1900, pp.
235 and 241.

9W. H. DALL: Seventeenth Ann. Rept. U.S. Geol. Surv., Pt. I, 1896, p. 135.

70C,..W. PURINGTON: Manuscript map referred to by Spurr.

mJy. E. Spurr: Twentieth Ann. Rept. U. S. Geol. Surv. Pt. VII, I900, pp.
233-235.

REVIEWS 459

He cites at the outset a difficulty, ‘perhaps more apparent than
real,’ encountered by the newer hypothesis in the sporadic arrange-
ment of the meteoric material which, if like known meteorites, would
differ from the existing surface rocks. This difficulty, however, loses
much, if not all,-of its force when the effects of volcanic action are
considered. The hypothesis assumes that the interior heat which
arises from compression gives rise to the melting of certain constitu-
ents of the rock mass, and that these, previous to eruption, undergo
magmatic differentiation into the well-known igneous rocks and prob-
ably into others which are but rarely ejected because of their high
specific gravity, as the iron-bearing basalt of Disco Island, Greenland,
and other extremely basic rocks of the ferro-magnesian type. Volcanic
action is assumed to have begun effectively before the growing earth
reached the size of the moon and all accretions subsequently made
would be more or less invaded and overflown by igneous intrusions
and extrusions of differentiated lava. In the closing stages of the
earth’s growth, the infall of meteoric matter declined gradually to an
inappreciable amount, while the volcanic action is thought to have
continued with relative vigor for a notable period after the essential
cessation of growth, and to have perpetuated itself in less activity
down to the present time. If the moon may be taken as an illustra-
tion of the prevalence and effectiveness of surface vulcanism in a body
one eightieth of the earth’s mass, it does not seem violent to suppose
that the original meteoric matter of the earth would be deeply buried
under surface lava flows and tuffs in the closing stages of its growth.
Recent studies in the Lake Superior region, in Scandinavia, and in
Lapland seem to concur in showing that the oldest known rocks con-
sist of such lava flows and pyroclastic layers associated with some
small amounts of ordinary clastic material, all mashed into schistos-
ity. Into these schists, the great granitic series were intruded. Under
the newer hypothesis these intrusions are to be regarded as merely
a continuation of the earlier active vulcanism which was then more
largely basic, but which had now, in the progress of magmatic differ-
entiation, attained a dominant acidic character, perhaps as the partial
complement of the earlier basic flows of the schist series or of the later
basic flows of the Algonkian. The “fundamental gneiss” does not,
therefore, appear, in the light of these recent studies, to be funda-
mental, nor does the ‘‘basement complex” appear to be dasa/. These
recent investigations seem to bring the Archean series into almost

| 6D. g>eeme

460 REVIEWS

ideal conformity with the accretion hypothesis, if under that hypoth-
esis the process of accretion is conceived as dying away gradually by
a transition into a stage of dominant vulcanism, which in turn gradu-
ally passes into the present phase of dominant aqueous activity. On
the other hand, progressive investigation seems more than ever to give
negative results in the line of the discovery of “the original crust”
of the hypothetical molten stage, and the survival of the older hypoth-
esis will perhaps require the recognition of a dominant eruptive
stage similar to that postulated by the newer hypothesis, to which all
or most of the Archean rocks are to be referred. In the light of these
late Archean investigations, the difficulties of the old hypothesis seem
at least as great as those of the new, for the old hypothesis must
account for the non-appearance or scant appearance of ‘the original
crust,” while the new must account for the non-appearance or scant
appearance of the supposed highly basic, magnesian, iron-bearing
meteoric matter. The new hypothesis has the advantage of having
theoretically postulated in advance what field studies are now bringing
into recognition in spite of prepossessions inherited from the older
view.

Passing the problem of superficial constitution as not necessarily
serious, Fisher justly regards the increase of internal density and high
internal temperature as incontestable facts of radical importance, and
inquires how these facts may be accounted for on the meteoric theory.
He assumes the average density of the meteoric matter to be nearly
that of average surface rock, 2.75, and adds that “‘if this is too low,
the arguments based upon it will not be affected in any great degree.”
Fisher feels tolerably certain that the law of internal density is fairly
represented by Laplace’s law, which is that ‘the increase of the square
of the density varies as the increase of the pressure.” In the case of
a slow growth by solid accretion, the internal density must be mainly
referred to compression. If, however, the specific gravity of the
original meteoric material be taken at some figure between 3.5 and 4,
as derived from known meteorites (Farrington’s figure in 3.69), the
amount of compression is appreciably less than on the assumption of
2.75 made in the computations. Fisher finds that at a depth of goo
miles, where by Laplace’s law the density should be 3.88 the compres-
sibility would be 1.4021 X 106. ‘*This may be looked upon asa small
compressibility, seeing that the compressibility of water similarly
measured is 4.78 X 10 5 or nearly forty times as great.” The linear

REVIEWS 461

dimensions would be reduced about one tenth. ‘At the center the
compressibility similarly measured would be very small, viz., 2.5 x
10%, while the condensation would be large, viz., 0.744.”

In the absence of direct measurements on the compressibility of
rocks, the author computes its value from the values of Young’s modu-
lus and the modulus of rigidity which have been obtained in some
instances, and compares the result with the theoretical compressibility
of surface 10cks deduced from Laplace’s law. Respecting the results,
he remarks that “it is certainly not a little remarkable how closely
this value ranges with those found by experiment. It is of the same
order of magnitude but rather smaller than the average.” He adds:

We find here a somewhat strong presumption in favor of the view that
the earth consists throughout of matter not very dissimilar from what we know
at the surface, and that the internal densities are due rather to condensation
than to the presence of heavier substances such as metals. But it is not a
proof of this.

Respecting the alternative view that the greater density toward the
center is due to heavy metals, Fisher says:

We may probably dismiss the supposition that these all fell in first, and
only regard them as segregated from a uniform mass of some kind, and hav-
ing gravitated towards the center. This implies a condition of liquidity. If
the materials were solid this separation could not have occurred. Now the
only force that we know of that could cause the denser materials to move by
a kind of convection towards the center is gravity; and in a solid gravity
would not have that effect. Moreover, it must not be forgotten that gravity
continually diminishes as we go deeper into the earth, and that at the center
bodies have actually no weight. It is greatest at the surface, and if not com-
petent to segregate downwards the heavy particles of a rock at the surface,
which we know it is not, still less could it have that effect near the earth's
center.

Neither can we attribute this segregation to pressure; for pressures act
equally upon the surface of heavy or light materials. If we had a layer of
mixed shot and sand, no steady pressure laid upon it would force the shot to
the bottom and bring the sand to the top.

It seems, therefore, that the view that the denser materials in the interior
consist of heavy metals necessitates a condition of liquidity of the whole, which
accords more readily with the nebular than with the meteoric theory of its
origin. For we may imagine that ina nebular mass cooling from the exterior,
the first change from a nebulous or gaseous state would be the formation of a
rain of condensed particles falling downwards, which would continue until
the whole mass became liquid, and thus the heavier elements would begin to

462 REVIEWS

collect towards the center. In this case the highest possible interior temper-
ature would be that at which the gaseous first assumed the liquid condition
under the pressure at the depth.

Paradoxical as it appears, it is therefore possible that the temperature in
the interior may have been rendered higher by a conglomeration of cold solid
meteorites than by the cooling of a nebula.

We have no means of judging whether the meteorites would come in
rapidly or slowly, but in either case if we take no account of the heat arising
from impact, the amount produced by condensation would be the same; the
only difference in the two cases being that it would be generated in a less or
greater time. In the meanwhile a covering of a badly conducting material
would concurrently accumulate, preventing the rapid escape of this heat, and
at the same time increasing the pressure, the compression, and the heat.

To form an idea of the temperature which would be produced by the
condensation of matter of surface density to the density now existing at any
given depth within the earth, not taking into account its diffusion by conduc-
tion or otherwise, we require to know the work which has been expended
upon it. Now we can estimate this in the following manner. Conceive the earth
to have been built up of meteorites falling in, so that shell after shell accumu-
lated until the globe attained its present size. Then, fixing the attention
upon a particular unit volume, say a cubic foot, of the substance, and omit-
ting atmospheric pressure, it would successively be subject to every degree of
pressure from zero, when the shell of which it formed a part was not covered
up, until the present pressure was reached, when it was buried to the depth
at which it now lies. If then we know the relation between the pressure and
the compression at every depth at the present moment, it will give us the
relation between the pressure and the compression which that particular
volume has obeyed during the course of ages; that is to say, we can judge
how much compression any given pressure would have produced in the sub-
stance under the conditions involved.

Laplace’s law of density being based upon the assumption that the
increase of pressure within the earth is proportional to the increase of the
square of the density, in terms of a pressure of one pound upon the square
foot, this leads to the result, that the pressure at the depth where the density
is f is equal to 5.9X 10" (f?—s*)[where s=density of surface rock and =
density of rock at the depth under consideration].

If we accept Laplace’s law, this expresses a fact, whether the increase
of density is due to condensation by pressure or to increased density in the
intrinsic nature of the matter. But if we assume that the increase of density
is caused solely by the pressure, then the above relation gives the amount of
pressure which would reduce matter of density s to matter of density # under
circumstances existing within the earth. It will therefore remain true if the

REVIEWS 403

matter changes its state from solid to liquid, and from liquid to gas. If, for
instance, we wished to apply a pressure which would reduce surface rock to
the density 3, it ought to be 5.9X10’ (g—2.75") = 8.481 X10” pounds per
square foot, supposing no heat be allowed to escape. If the experiment
could be made, it would afford a test of the truth or otherwise of the present
hypothesis.

When we know the relation between the pressure and the condensation
which it would produce, it is feasible to estimate the heat which would be
generated, and also the temperature, provided we assume the specific heat
of the substance, which for surface rock has been determined. For instance,
at the depth of 0.1 of the radius, or about 4oo miles deep, where the density
would be 3.88, the temperature produced by condensation would be 1.2608 X
10° Fahr., or 7.0044 X 10* Cent. [70,044°], while at the center the figures
would reach 2.7756 X10° Fahr., or 1.0242 X 10° Cent. [1,024,200°]. It
seems at any rate that the meteoric theory would not fall short of accounting
for temperatures as high as might be desired. It must at the same time be
remembered that much of this heat would not be called into existence until
the substance into which it was, as it were, being squeezed, had already been
deeply buried under a badly conducting covering, so that the escape of heat
would not take place as fast as it was generated, as would probably be the
case with heat generated at the suface byimpacts. Thus the hypothesis that
the present high internal temperatures are due to compression seems quite
admissible.

We may compare the above named temperatures with some that are
known. Acheson, for instance, obtains 6500° Fahr. in his Carborundum elec-
tric furnace, and 3300° Fahr. has been obtained by the oxyhydrogen flame.
These temperatures are contemptible compared with those mentioned above.
The Hon. Clarence King, prolonging Dr. Barus’ line for the melting point of
diabase (which is 1170° C, at the earth’s surface) to the earth’s center, gives
the temperature 76000° Cent., which is of the same order of magnitude as
condensation would produce at only 400 miles depth.

Fisher considers the bearing of the temperature of lava as deter-
mined by Bartoli at Etna (1to60°C. or 1932° F.) on the question, and
finds that the theoretical depth at which this lava temperature would
be produced by condensation would be about forty-three miles. The
same temperature would be reached at the accepted gradient of 1° F.
for sixty feet in about twenty-two miles.

It seems then that the hypothesis, that the internal densities are due to
the condensation of matter of surface density, will not account for a tempera-
ture gradient originally as high as at present. [The computed gradient cor-
responds pretty nearly with the low gradient found at the Calumet and Hecla

464 REVIEWS

mine.] Nevertheless the above observations upon the temperature of lava,
and the comparatively small depth, forty miles, at which condensation of
rock would be capable of producing it, together with the small amount of
condensation necessary, viz., 0.041, render it quite probable that fusion may
have ensued in the deep interior without the necessity of a greater amount of
condensation than such materials might be supposed capable of under the
enormous pressure to which they would be subjected, even allowing for the
increase of the melting point under pressure. . . . . It will be noticed
that a compression less than would be requisite of itself to produce the
necessary density would be sufficient to produce the requisite temperature for
fusion. But while any stratum was cooling by the conduction upwards of its
own heat of compression, it would be receiving heat from regions below,
where, so long as condensation was going on, the materials would grow hot-
ter and hotter. It seems therefore possible that the upper layers, forming
what we call the crust of the earth, may have received sufficient heat supplied
from below to render the temperature gradient at the present time higher
than it was originally, and that even those Archean rocks, which are by many
thought to have been once melted, do not necessarily prove that the earth
was not built of cold meteorites.

The presence of water upon the earth has to be accounted for, and the
meteoric theory does not easily lend itself for this purpose. Not only is water
present in the ocean and in the atmosphere, but also in a state of solution in
the interior, as is testified by the enormous amount of steam emitted by vol-
canoes, and by cooling lava. It does not seem possible that molten rock can
imbibe water from without, because it would be driven away instead of
attracted, since the superficial tension of a substance diminishes as the tem-
perature rises.

The problem of accounting for the vast quantities of steam emitted
by lavas is shared by both theories. Under the hypothesis of a molten
earth, steam must have been absorbed either in the original molten
state or during the later stages of segregation and ascent, neither of
which alternatives seems to be free from difficulties. Under the
meteoric hypothesis, it is assumed that hydrogen, carbon dioxide,
carbon monoxide, and nitrogen were carried into the whole body of
the earth by the infalling matter in some such degree as they are
brought to the surface now by meteorites, and that these gases, joined
with oxygen derived from the partial reduction of the oxides of the
meteoric matter when subjected to the high temperatures of the
interior, were extruded by volcanic and similar means and gave rise to
the ocean and atmosphere. Under this hypothesis the volcanic gases
are regarded as mainly original and as merely lingering expressions

REVIEWS 465

of the process that was much more intense during the later stages
of the earth’s growth. Cosmic accretions, which may be a notable
factor, would be equally functions of either hypothesis so far as the
maintenance of the atmosphere and ocean is concerned.

In submitting the newer hypothesis to the test of physical princi-
ples and mathematical computations, Fisher has done it an honor that
is sincerely appreciated. By showing that its more radical features lie
within the tenable limits of theory, he has helped to give it a place as
a genuine working hypothesis; and as such it may have some stimu-
lating value as a competitor of the gaseous and molten theory which
has practically monopolized geological opinion for the past century.

pa One OF

Glacial Sculpture of the Bighorn Mountains, Wyoming. By FRANcoIS
E. Matrues. Extract from the Twenty-first Annual Report
of the United States Geological Survey, 1899-1900. Wash-
ington, Igoo.

Glaciation affected the crest of the Bighorn Mountains for more
than thirty miles. The range was not covered by a continuous ice
cap, and glaciation was confined to valleys. The mountains abound in
well developed, elongate, valley-like cirques, which have been but little
altered by postglacial changes. The author indorses Johnson’s view
of the origin of cirques, namely, that they are due to sharply localized
and abnormally vigorous weathering, by rapid alternation of freezing
and thawing at the exposed bottoms of dergschrunds. Mr. Matthes’
studies have led him to the conclusion that the location of the derg-
schrunds in any valley is determined by the depth of the névé.

The longest glacier of the Bighorn Mountains is said to have been
eighteen miles in length, its terminus reaching down to an altitude of
less than 7000 feet. The thickness of the larger glaciers was 1000
to 1500 feet. Small glaciers still exist in the highest part of the
range, a little below 44° 30’, at an altitude of about 12,000 feet.

In addition to the account of the effects of the active valley glaciers
on topography, the author discusses the effect of inactive snow and
névé. The névé effects are described under the term ‘‘nivation,” and
the “‘nivated” valleys are distinguished from the glaciated valleys.
This, so far as we are aware, is the first attempt to analyze the effects
of inactive ice and névé on topography. The discussion even involves

466 REVIEWS

the effects of snow banks. ‘The thickness of the névé fields which did
not become glaciers is estimated to have been 100 to 150 feet, and the
conclusion is reached that, on a grade of about 12 per cent., the névé
must attain the thickness of at least 125 feet in order to have motion.

Certain phases of the problem of glacial motion are touched, and
the conclusion reached that the cause of glacial motion is to be sought
in the weight of the ice mass, and that it is independent of the tem-
perature of the air. No attempt is made, however, to decide the real
nature of glacial motion, or what processes are involved in it.

RED eS.

Annual Report of the Board of Regents of the Smithsonian Institution,
showing the Operations, Expenditures, and Condition of
the Institution for the year ending June 30, 1899.°

In the appendix accompanying the official report of the governing
bodies and the secretary of the Smithsonian Institution, a complemen-
tary number of geological articles are introduced. ‘These are, for the
most part, republications, and include ‘‘On Lord Kelvin’s Address on
the Age of the Earth as an Abode Fitted for Life,” by T. C. Chamber-
lin; ‘“‘An Estimate of the Geological Age of the Earth,” by J. Joly;
She Petrined thorests of Arizona, by eester shewW andi anesent
Conditions of the Floor of the Ocean; Evolution of the Continental
and Oceanic Areas,” by Sir John Murray; ‘The Truth About the
Mammoth,” by Frederic A. Lucas ; “‘ Mammoth Ivory,” by R. Lydek-
ker ; and “ Review of the Evidence Relating to Auriferous Gravel Man
in California,” by William H. Holmes. Several of the physical and
biological articles, and those relating to general aspects of science,
also possess points of interest to geologists. (Cr

RECENT PUBLICATIONS

—BECKER, GEORGE F. Conditions Requisite to our Success in the Philip-
pine Islands. An address delivered before the American Geographical
Society, February 20, 1901. [Reprinted from Bull. of the Am. Geog.

Soc., April 1901.|

Report on the Geology of the Philippine Islands. Followed by a ver-
sion of Ueber Tertiare Fossilien von den Philippinen (1895) by K
Martin. [Extract from the Twenty-first Annual Report of the U.S.
Geological Survey, 1899-I900. Part I1I—General Geology, Ore and
Phosphate Deposits, Philippines.] Washington, Igor.

—COoLLi£, GEORGE Lucius. Wisconsin Shore of Lake Superior. [ Bull. of
the Geol. Soc. of Am., Vol. XII, pp. 197-216.] Rochester, April 1gol.

—Davis, W. M. An Excursion in Bosnia, Hercegovina, and Dalmatia.
[From Bulletin, Vol. III, No. 2, Geog. Soc. of Philadelphia. ]

An Excursion to the Grand Canyon of the Colorado. [Bulletin of the
Museum of Comparative Zoélogy at Harvard College, Vol. XX XVIII.
Geological Series, Vol. V, No. 4; with two plates.]| Cambridge, Mass.,
May Iogol.

The Geographical Cycle. [Paper read at the Seventh International
Geographical Congress of Berlin in 1899.] Berlin, 1900.

—DorseEy, GEorGE A, Archeological Investigations on the Island of La
Plata, Ecuador. [Field Columbian Museum, Publication 56; Anthro-
pological Series, Vol. II, No. 5.] Chicago, April got.

—DorseEy, GEORGE A AND H. R. VoTH. The Oraibi Soyal Ceremony ;
The Stanley McCormick Hopi Expedition. [Field Columbian Museum,
Publication 55; Anthropological Series, Vol. III, No. 1.] Chicago,
March Igol.

—DvumBLeE, E. T. I, Cretaceous of Obispo Canyon, Sonora, Mexico.
[Reprinted from the Transactions of the Texas Academy of Sciences,
Vol. IV, Part I, t900.] II, Occurrence of Oyster Shells in Volcanic
Deposits in Sonora, Mexico.

Geology of the Beaumont Oil Field.

Natural Coke of the Santa Clara Coal-Field, Sonora, Mexico. [Transac-
tions of the American Institute of Mining Engineers, California Meet-
ing, September 1899.]

467

468 KE CENT POBLICA TIONS

The Iron Ores of East Texas.

The Oil Deposits of Texas.

—E.uiot, D.G. A List of the Land and Sea Mammals of North America
North of Mexico. [Field Columbian Museum, Publication 57; Zodlog-
ical Series, Vol. II, No. 2.] Chicago, June Igol.

A List of Mammals obtained by Thaddeus Surber, in North and South
Carolina, Georgia, and Florida. [Field Columbian Museum, Publi-
cation 58; Zodlogical Series, Vol. III, No. 4.] Chicago, June
Igol.

The Caribou of the Kenai Peninsula, Alaska. [Field Columbian
Museum, Publication 59; Zodlogical Series, Vol. III, No. 5.] Chicago,
July tgot.

—FIsHER, Rev. O. On Rival Theories of Cosmogony. [From the American
Journal of Science, Vol. XI, June 1g01.]

—India, Geological Survey of. General Report for the Year ending March
31, 1901. C. L. Griesbach, Director. Calcutta, 1gol.

—New Jersey, Geological Survey of. Annual Report of the State Geologist
for the Year 1900, ‘Prenton, N.°J., Toor.

—NutTtinG, C.C. The Hydroids of the Woods Hole Region. [Extracted
from U.S. Fish Commission Bulletin for 1899, pp. 325-386. Date of
publication, June 8, tg01.] Washington, Igor.

—OLDHAM, R. D. Beach Formation in the Thirlmere Reservoir. [ Abstract
of a paper read before the British Association, Bradford, 1900.]
—@yen, P. A. Variations of Norwegian Glaciers. [Separataftryk af “Nyt

f. Naturvidenskab,” B. 39, H. 1. Kr. ania, Igo1.|] Christiania, April

IgOl.

—RABOT, CHARLES. Les Variations des Longueur Des Glaciers dans les
Régions Artiques et Boréales. [Extrait des Archives des sciences
physiques et naturelles Années 1899 et 1900.] Genéve et Bale, Igoo.

—Rerp, H. F. Observations of Earthquakes. [From the Johns Hopkins

~ University Circular, No. 152, May 1go1.]

The Variations of Glaciers, VI. [Reprinted from the JOURNAL OF GEOL-
oGy, Vol. IX, No. 3, April-May 1go1.]

-—SHALER, N.S. Broad Valleys of the Cordilleras. [Bulletin of the Geol.
Soc. of Am., Vol. XII, pp. 271-300.] Rochester, June Igol.

—SHIMEK, B. The Loess of Iowa City and Vicinity. Iowa Pteridophyta
(Con.). Addenda to the Flora of Lyon County. [Excerpt from the
Bulletin of the Laboratories of Natural History, State University of
Iowa, Vol. V, No. 2, pp. 195-216.] May root.

Plate I

Jour. Gror., Vol. IX, No. 6

lverimetynedea dt a

RIVER MAP OF CONNECTICUT

(Reduced from the two-sheet map of Connecticut by the U, S. Geological Survey)

Vhe dotted lines indicate the prominent “trough lines,” which trend as follows
(Ah ah Bey IN FP 1, Cy Ch Aiea No 2? 1D, gy N. 48° E.
i N. 20° B. h*, h?, etc., N. 90° E.W.

a*, a®, etc.,, N. 44° W.

ot, 6°, etc,, N. 5° W. dxndavelcwIN-wrsoubs

ewe. OF -GeOLOGY

SEPTEMBER-OCTOBER, roor

THE RIVER ‘SYSi EM OF CONNECTICUT:

THE tendency of the modern school of physiographers seems
to be to ascribe little importance to geological structure planes
as a factor in determining the position and the orientation of
water courses. In an earlier period greater attention seems to
have been accorded to this condition, and quite apart from the
purely theoretical conceptions of the closet geologists many
valuable observations were placed upon record. Th. Kjerulf,
the former Director of the Geological Survey of Norway,
remarked the regularity in the arrangement of fjords, streams,
and valleys as represented upon the map of Norway, and from
this was led to believe that their directions corresponded to the
faults in a system of dislocations. Daubrée’ has furnished many
examples of regular networks of stream channels which resem-
ble the network produced by a number of intersecting series
of parallel joint planes (réseaux réguliers de cassures), and the
explanation of this correspondence he believed to be a causal
one, the planes of separation, or jointing, being locally gap-
ing where the streams adhere to the joint direction, but closed
where they have been diverted from the course of the joint.
Van Hise} has in a similiar way explained the zigzags of

* Published with the permission of the Director of the U. S. Geological Survey.

? DAUBREE: Géologie Experimentale. Paris, 1879, p. 361.

3 VAN HIsE: Trans. Wis. Acad. Sci., etc., Vol. X., pp. 556-560, 1895.

Vol. IX, No. 6. 469

470 WILLIAM HERBERT HOBBS

the smali streams which enter the Wisconsin River at the
“Dells” as due to a system of joints induced by the gentle fold-
ing of the rocks. That a more or less uniform system of joints
exists in the rocks almost throughout the state of Wisconsin has
been shown by Buckley’s observations."

So far as the writer is aware, the only detailed studies that
have been made establishing the definite relationship of the
stream channels of a region to its system of joints or faults, are
those of Broégger? in the Langesund-Skien and Christiania
regions of southern Norway, and that of the writer? in the Pom-
peraug Valley region of Connecticut. Of the elaborately faulted
region which Bréogger has studied, he states:

It is not exaggerated when to my own astonishment I must state, as the
final result of my observations in this region, that almost every valley, every
cleft, is formed along a fault fissure... .. The significance of the faults for
the formation of the valley straits is thus as profound as possible within the
stretch of land described, since almost every cliff, every vale, every bay
has been formed upon a line of dislocation; indeed the presence of clefts
was for me, at the last, the surest index for the discovery of dislocations.+
| Translation. |

Review of the geological structure of the Pomperaug Valley area
of Connecticut. — Regarding the Connecticut area to which refer-
ence is made, it will be necessary here to review the study in
order to make clear the generalizations with which this paper is
especially concerned. It is now well known that the several areas
of Newark rocks of the eastern United States are complexly
faulted, and the many common structural peculiarities of the
several isolated areas give rise to the belief that the forces which
have produced the dislocations have affected, not the Newark
rocks alone, but the larger region of which they are parts—the
Piedmont plateau of the eastern United States.5

‘BUCKLEY: Bull. 4, Geol and Nat. Hist. Surv. Wis., pp. 450-460, 1898.

?BrROGGER: Nyt Magazin for Naturvidenskaberne, Vol. XXVIII, pp. 253-419,
1884. Jbid., Vol. XXX, pp. 99-231, 1886.

3 Hoss: Twenty-first Ann. Rept. U.S. Geol. Surv., Pt. III, 1901, pp. 1-162.

4Op. cit., pp. 34, 342.

5Cf. Davis: The Triassic Formation of Connecticut, Seventh Ann. Rept. U.S.
Geol. Surv., 1888, pp. 481-490.

LEE TAVER SY SLEM OF CONNECTICOT 471

The detailed study of the Pomperaug Valley area has devel-
oped the fact that a complex system made up of intersecting series
of parallel nearly vertical joints and faults there divides the crust
into a large number of orographic blocks, the smaller of which
have dimensions of less than one hundred paces. The different
throws along the numerous faults bounding these blocks have
brought about a structure not unlike that of a mosaic from
which the supporting base has been lowered and the individual
stones been displaced by different amounts due to the inequali-
ties of their lateral support. For the area as a whole, the joints
and corresponding faults are embraced mainly in four series, the
individual members in which trend N. + 34° W.,N.+55° E., N.
ae 56 We, and N.-115-. -Series of faults.less common for the
area as a whole, but several of them numerous enough in its
southern portions, have directions N. + 33° E., N. + 44° W., N.
S160, War N. = 00° HaWerN.- 20° Hand Nx == 25° W...the
order being approximately that of frequency of occurrence. Of
the more common series, those trending N. + 55° E. and N. +
34° W. are nearly normal to one another, as would be true of a
pair of joint planes, but the larger throws within the region seem
generally to have taken place along one of these planes (N. +
55° E.) and one of the other prevailing directions (N. + 5° W.).
Aside from the regularity in direction observed to characterize
the numerous faults and bring them into a number of parallel
series, those faults of the same order of displacement are
observed to be spaced also with noteworthy regularity. The
smallest of the orographic blocks which could be measured, for
convenience called the “ unit’’ blocks, were found to be quite gen-
erally about 50 paces (150 feet) along the direction N. + 55°E.,
and 100 paces (300 feet) along the direction N. + 5° W.—the
equivalent of two rhombic prisms in contact along one side.
The larger, or ‘‘composite” blocks, which are of various orders
of magnitude, are made up of the ‘‘unit” blocks and bordered
by displacements of a higher order—greater throw. The fault
intervals in the several series, and hence the shapes of the oro-
graphic blocks, are found to be closely related to the directions

472 WILLIAM HERBERT HOBBS

of the prevailing and rarer fault series in a manner which is made
clear by the diagram of Fig. 1. In this figure the shape of the
unit orographic block—and of many composite blocks of. sev-
eral orders as well—is shown by the black and also by the
stippled area. Two of the prevailing fault directions, N. + 15° E.
and N. + 34° W. correspond to
the longer and the shorter diago-
nals respectively of this unit
block. The fault direction N.+
33° E., which is the most com-
mon of those occurring in the
southern Pomperaug area, corre-
sponds closely to the longer
diagonal of three unit blocks in
contact along their longer side.
Another of the series of faults
characteristic of the same south-
ern area and trending N. + 20°
E., corresponds to the longer

Fic. 1.—Diagram to illustrate the diagonal of six unit blocks, two

relationships existing between the shape composite blocks like those just
of the “unit” orographic blocks and the

described placed in contact along
directions of the fault series.

their longer sides. The remain-
ing directions of faults observed in the area are respectively the
shorter diagonals of composite blocks three units long and two
wide, four long and six wide, four long and seven wide, and
two long and seven wide; though clearly in these latter instances,
the relationships indicated in the diagram are less significant
owing to the more complex nature of the composite blocks of
which the fault directions are the diagonals. For a more impor-
tant indication of the same relationship, which is observed in the
actual composite blocks of the area studied, the reader must be
referred to the original report.‘
For the area in question the conclusion has been drawn that
the earth’s crust has been here subjected to compressive stresses,

INZOC ACIP PLL ja 2 0.

THE? RIVER SYSTEM OF CONNECTICOTL 473

the resultant of which acted in a direction normal to the axis of
Green Mountain folding (N. + 80° W.), stresses which were
relieved by dislocation along the planes of maximum shear,
approximately 45° to either side of the direction of pressure,
(ew Ss e= Gb w@Neta 55 hand N. = 34° W. The remain-
ing common directions of fault planes (N. + 15° E. and N. +
5° W.) could be explained by a subsequent development of
compressive stress acting along the initial direction; for in a
region containing planes of separation at 45° to the direction of
pressure, there would be a resolution of the stress into compo-
nents acting along the planes of separation and along that diag-
onal of the blocks which is nearly normal to the pressure. The
direction N. + 15° E. corresponds to this diagonal of the joint
block already formed, and the remaining direction N. + 5° W.
is the corresponding diagonal of two such blocks in contact. To
explain these latter dislocations upon the same principle, it must
be assumed that, owing to differences between the alternate joint
planes of the same series, these double joint blocks acted to
some extent as units. The fault directions characteristic espe-
cially of the southern portion of the area and disclosing such
intimate relationships to the four generally prevalent ones, might
be explained either by assuming that in the later compression of
the jointed area composite blocks of different shapes, because
composed of a different number or different arrangement of unit
blocks, acted as units (whether due to the greater perfection ot
their bounding fault planes over intermediate ones, to the par-
tial closing or healing of the intermediate faults, or to some
other cause). In this case the maximum shear would be along
the diagonal of the composite blocks, as would be the case in
the unit blocks themselves, and it is shown in the report under
review that these directions do correspond, not only in direction,
but also in position with the diagonals of important composite
blocks of the area. Under certain conditions, which may or may
not have obtained in the area, planes of dislocation having the
same directions might have been produced through the depres-
sion of the composite blocks subsequent to the jointing of the

474 WILLIAM HERBERT HOBBS

area, Since in this case also the tendency would be for blocks to
rupture along the diagonals.

The oriented drainage of the Pomperaug Valley.—With this brief
summary of the geological structure within the Pomperaug Val-
ley area we may consider its drainage. It was found in study-
ing the area that the streams, large and small, for considerable
distances adhere with great fidelity to the directions of some of
the prevailing faults, and that in many cases after being diverted
from them, it was noted that they had returned persistently to
the old direction. This correspondence of drainage lines to
geological structure planes is far too close to be accidental.
The four prevailing fault series diverge from their nearest neigh-
bors at angles of about 39°, 20°, 29°, and 92°. A difference in
angle between a fault direction and the general direction of a
stream course equivalent to 7°, or about one third the small-
est difference of angle between neighboring fault directions,
would represent a divergence of one in eight, which would hardly
be accepted by the eye as an indication of parallelism. It is
not to be expected that the actual course of a stream will now
be coincident with or even absolutely parallel to any fault direc-
tion, for there have unquestionably been many local conditions
which have produced larger or smaller migrations of the river
channels. Their general direction has, however, it would seem,
been maintained despite the minor accidents which have marked
their life-histories, and even under so revolutionary a change as
complete reversal of drainage.

It was further shown in the investigation under review, that
in the walls of crystalline rocks surrounding the Newark beds in
the Pomperaug basin, the same adhesion of the water courses to
the direction of the observed faults (extended) could be deter-
mined. It was hardly to be expected that these peculiarities
would cease to be observable so soon as the immediate vicinity
of the Newark basin was left behind, if it be indeed true that the
dislocation of the area is due to a compression of the general
region in which this area of Newark rocks and the much larger
one of the Connecticut valley are included. Owing, however,

THE RIVER SVSTEM OF CONNECTICUT 475

to the absence over the greater part of the surrounding area of
widely different rocks in thin beds, the difficulties of locating
fault planes are almost infinitely increased; and, indeed, except
under especially fortuitous conditions they elude observation.
Having established, however, a relationship in one area, the
problem is before us to determine
whether the river system of the larger
area of Connecticut exhibits any indi-
cation-of the existence of rectilinear
directions more or less persistent em-
braced in a network of parallel series
like that of the Pomperaug Valley; or,
if this be not true, whether any other
persistent and recurrent directions can ; ;
be observed. It will be further. of ee
special interest if such a system affords Cee ola SS

indication of a,regular space-interval) “2.2.22: 4... - Ny as
between such parallel lines of drainage. ONC ie ~

The ortented drainage of the Shepaug NL ue he
River.—In prosecuting our inquiry a ee LANGE aA
regarding the orientation of the drain- Noa
age lines of Connecticut, the valley of
the Shepaug River (its southern por-
tion) was first examined, since this sorn,\y
basin immediately adjoins to the west Ne
that of the Pomperaug. The diagram
of this river traced from the atlas
sheet of the map of Connecticut, pre-

3 Miles.

Fic. 2.— Diagram to show
the relationship of drainage lines
pared by the topographic corps of the in the basin of the Shepaug

United States Geological Survey River tore pre valeau ound
: : joint lines of the neighboring

(Hig: 2) affords clear evidence that Pompersue bac

the geological structure planes have

here played an important part in giving direction to the river's

channels. The dotted lines of the figure show the inferred

approximate positions of fault planes whose course the river

has adopted. The most marked adhesion of the river or its

476 WILLIAM HERBERT HOBBS

branches to any given direction is very closely the direction
N. 5° W., as shown by the upper course (in the diagram ) of the
trunk stream, as well as by the upper course of Jack Brook, its
main eastern tributary, and by the small intermediate branch of
the latter.. These three channels are not only evenly spaced,
but to the southward their extensions correspond in position
with the southerly trending elements of great elbows of the Hou-
satonic (see Plate II). After receiving the waters of Jack Brook,
the Shepaug flows for nearly three miles in the direction S. +
15° W. (N. + 15° E.), another of the four prevailing fault direc-
tions observed in the Pomperaug Valley. Later it flows S. +
34° E. (N. + 34° W), while several tributaries flow along this
direction and S. + 55° W.(N.+ 55° E.), the two remaining of
the four prevalent fault directions of the Pomperaug Valley.
The oriented drainage within the area surrounding the Pomperaug
Valley—In the map of Plate II, a larger area, of which the
Pomperaug Valley is the center, has been considered. Upon
this map, traced from the United States Geological Survey atlas
sheets, which are on a scale of one inch to the mile, the full

straight lines represent observed faults (extended), and the
dotted lines the inferred approximate position of faults. It has,
of course, been necessary to omit all but a few of the numerous
faults which were observed in the Pomperaug Valley area. The
figures on the margin of the map indicate the directions (in
degrees) either to the east or west of north of the trough or
fault lines which emerge near them. With the greater com-
plexity of this map and the larger number of fault directions
represented, relationships are not at once so apparent as in Fig.
2, which upon a reduced scale is here included. Special atten-
tion is directed, however, to the fact that the Housatonic itself,
which after flowing about three miles in a direction S. + 5° E.
(N. = 5° W.)), assumes the direction'S3- 61, (N-=— O12)
at first ina nearly straight channel for three miles, and then ina
zigzag course for eight miles more. Many interesting peculiari-
ties will be noticed in the orientation of the smaller streams
shown upon this map; as, for example, along the N.+ 55° E.

477

TITER RES SLM OF :CONNEC TICES:

Plate II

[X,2 No 6

Jour. -GErore Vol:

‘may} Ieau ssed Yor ,,SAUT] YHNOI},, 94} JO UBIPLIIUI JY} JO JSAM 10 4sva

ay} 0} saaiBep url a[Suv oy} 9ATs ursieu ay} UO samnsy oy} $(souvyd ynvz 10 yurof yo suontsod ayeurxoidde paszrayut)

Soul] Ysnolj,,

ale sauly poop

AHTIVA ONV

lot _ 6S Ed

ay} { (papuayxa) vare AaT[eA Snvisdwog ayy fo s}[Nvj paarasqo aie soul] [INF 24,

WHdWOd AHL ONIGNNOUANS VAUNV AHL HO dVW AHA

sz mm

478 WILLIAM HERBERT HOBBS

line which approximates to a diagonal of the map, along the
N. + 33° E. line which meets the lower margin of the map near
where the Housatonic meets it, etc. It will hardly escape notice,
also, that the most marked lines in each series are spaced with
considerable regularity, and if one space is wider than the
others, it is often so much wider as to suggest that the space-
interval is a multiple of the space which has been regarded
temporarily as the unit.

The river system of Connecticut— An examination of the
larger area approximately coéxtensive with that of the state of
Connecticut has been made by use of the ‘‘two-sheet map of

”)

Connecticut,” on a scale of one half inch to the mile, prepared
by the United States Geological Survey Those lines and char-
acters of this map, such as topography and culture, with which
Wwe are not now concerned and which would obscure the rela-
tionships sought, were eliminated by preparing a careful tracing
of all the streams and their minor branches. Upon this map
tracing (about six feet long by four wide) approximately recti-
linear stretches of river channel, and especially the stretches of
neighboring streams which hold approximately to the same line,
were sought. If these directions were found to agree closely
with any of the fault directions observed in the Pomperaug
basin, dotted lines were drawn following those directions and
coinciding as closely as possible with the river courses. If such
an observed direction was found not to coincide closely with any
of the fault directions determined, a direction was sought which
would approximate most closely to it, and a similar dotted line
with this direction drawn along the course of the river. The

5)

term ‘trough lines”’ used to designate these lines, need for the
present be given no further signification than lines so favored by
nature that the waters of the region have been induced to
adopt them for their channels over longer or shorter distances.
On a map of this scale the trough lines, if rectilinear, should be
slightly curved, but inasmuch as the present river courses,
because of the many accidents of their history, can only roughly

approximate to the direction initially given them, it would bean

THE RIVER SYSTEM OF CONNECTICUT 479

over-refinement to introduce a correction of this nature, and they
are, therefore, left straight. When all important trough lines
had been thus represented, the map was reduced by photogra-
phy and the etching of Plate I produced after the important
rivers had been made a little heavier in order that their course
might be apparent. These details of the study have been noted
because the dangers of introducing the personal element while
drawing the course of a river with any theory of its orientation
in mind are very great. The map conveys a wrong impression,
therefore, chiefly in its slightly exaggerating the volume of cer-
tain streams which it was necessary to draw with heavier lines in
order that their correspondences might be apparent.

The first trough lines to impress the observer are those desig-
nated @,, @,, @,,@,, upon the map, lines which trend approxi-
mately N. 44° W., and which include the lower reaches of the
Housatonic, of the Connecticut below Middletown, of the lower
Willimantic and the Shetucket, and a stretch of the Quinebaug. A
less-marked trough line between z, and a@,, would include impor-
tant bends of the Naugatuck and Quinnipiac rivers. The sharp
bend of the Connecticut River at Springfield and Patchoug River,
tributary to the Quinebaug, may indicate the course of another
trough line in this series. Most noteworthy of all lines in this
series, however, is @,, since the lower stretch of the Connecticut
(some twenty-five miles long) is continued in the zigzag Sebethe
River, so as almost to connect with the stretch of the Farming-
ton River above its sharpest bend at Farmington. The direc-
tion of the lower stretch of the Connecticut is of especial inter-
est because the river at Middletown deserts the softer Newark
sediments to flow across the crystalline uplands, a peculiarity
which has been explained by Professor Davis" through conform-
able superimposition, the stream being supposed powerful enough
to maintain an initial course along this direction during the rise
of the uplands, at which time the harder gneisses were discov-
ered. The same hypothesis has been offered by Kimmel? to

‘Davis: The Triassic Formation of Connecticut, Eighteenth Ann. Rep. U.S.
Geol. Sury., Pt. II, 1898, p. 165.

7 KUMMEL: Some Rivers of Connecticut, Jour. GEOL.,Vol. I, p. 379, 1893.

480 WILLIAM HERBERT HOBBS

explain the lower course of the Housatonic, which similarly
deserts the limestone to cross the uplands. The presence of
important structure planes in these positions would, in the view
of the writer, afford the simpler explanation. It will be noted
here that there is considerable uniformity in the spacing of the

trough lines of this series, especially a,, a3, a

37 “4:

The next most striking series of trough lines is indicated
in the course of the Connecticut from Springfield to Hartford
(c,), a distance of twenty-six miles; of the Willimantic (c, );
of the Mt. Hlope(¢.)); the Little (c.)); and ya slono: stretchyor
the Quinebaug (c,). The direction of these trough lines is
about N. 5° E., though this is not one of those observed to
characterize the faults in the Pomperaug Valley area. Again,
the spacing of these trough lines is quite regular if we regard
the space between the Connecticut and the Willimantic as a
double interval. More striking, perhaps, than any of these
trough lines is the one indicated in the series of smaller streams
which extend along the line c, of the map. These streams are
too small to deserve names upon a map of this scale, but some
of them are known as Five-Mile River, Whetstone Brook, Moosup
River, Mt. Misery Brook, etc. The direction of the series was in
fact obtained from them and applied to the other trough lines in
the series. This direction, while not an observed fault direction,
corresponds to the longer diagonal of two unit orographic blocks
of the Pomperaug Valley placed in contact along their shorter
sides (see the dotted diagonal in Fig. 1), and thus it fills an
important gap in the system of dislocations outlined.

The trough lines d,,d@,, and d,, which trend N.+ 15° E., rep-
resent a third series. The line d,, which corresponds in position
with no very important stream, is an observed fault of the
Pomperaug Valley area (extended), in which area, however, the
Pomperaug River adheres closely to its direction; ad, corre-
sponds closely for a distance of about fifteen miles with the
course of the lower Naugatuck; while to the south the lower
Housatonic flows in a nearly parallel direction some distance
farther to the east. Most striking of this series is the trough

TE TAVER SY STEM OF CONNECTICGL 481

line @,, along which are arranged stretches of the West, the
Quinnipiac with its tributary the Ten-Mile, the Farmington with
its tributary the Pequabuck (for about fifteen miles), and a minor
branch of the Farmington, which enters at its last great bend.
This line, except near New Haven, is some distance east of the
curving western boundary of the Newark area of the Connecticut
valley.

The only important fault line observed to hold to the direc-
tion of the N.+ 20° E. faults, which is marked fon the map,
follows the Quinnipiac for about fifteen miles and continued
northward coincides with the course of the Connecticut fora dis-
tance of about six miles.

Following the direction N.+ 33° E. are the trough lines ¢,,
G5, €9 €4y mn; and é,, the-latter lesstmarked than the others, and
e, being an important observed fault in the Pomperaug Valley
area. The line e, follows for a distance of fifteen miles or more,
the middle course of the Housatonic; e¢, is nearly in line with
the Croton and the Aspetuck rivers; while e, is outlined by the
Pine River and the small branch of the Connecticut, which is
continuous with the southwesterly flowing elbow of that river at
Middletown. The line e, shows no striking correspondences on
a map of this scale, though midway between it and e, the Sal-
mon River, a branch of the Connecticut, for about seven miles
follows the direction closely.

ihe: troughslines| of-the N72- 5 W. series (0,05, 05,.etc.),
while not prominently marked by the courses of any great
streams, save for short distances by the Housatonic, the Nauga-
tuck, and the Connecticut, yet all show on a larger scale map in
the minor stream branches many interesting correspondences.
As this direction agrees more or less closely with the slope of
the Cretaceous plain of erosion of southern New England, the
failure of the strong streams to follow this direction is worthy
of note.

Though less marked than some of the other series, the
trough lines which trend N.+90° E. W. seem to be clearly
outlined, especially “,, which passes through the divide of the

482 WILLIAM HERBERT HOBBS

Farmington and Quinnipiac at Plainville, and #,, which follows
the easterly course of the Connecticut for a distance below Mid-
dletown.

A trough line seems to be indicated in the Salmon River and
a branch of the Willimantic (g). This direction is N. 48° E.
and does not correspond to any observed fault direction in the
Pomperaug Valley area. Neither is it the diagonal of any very
simple composite block of that area, which otherwise might indi-
cate its relationship to this system.

Conclustons—In conclusion it may be stated that the rivers
of Connecticut seem to indicate by the orientation of their chan-
nels the existence of a regular network composed of a number
of intersecting series of parallel lines, which for lack of a better
term have been designated ¢vough lines ; and, further, that with two
exceptions the more important of these trough lines correspond
closely in direction with the directions of fault series observed
to characterize the complexly faulted area of Newark rocks in
the Pomperaug Valley. Of the two exceptions to the rule, the
more noteworthy one (N.+5° E.) fills an important gap in
the system of faults determined for that area. This study is
therefore in its bearing a confirmation of the conclusions arrived
at by Kjerulf, Daubrée, and Brégger, who have seen in the
orientation of water courses the strong directing influence of
geological structure planes.

There are obviously a number of ways in which the dislocations
of a region, like the one under consideration, might be made to
account for the orientation of stream courses. The direction of
streams by the joint or fault planes themselves may be competent
to explain the network indicated, more particularly if the streams
began their cutting in the soft Newark sediments, which easily
sustain secondary fractures near fault planes. That some voids
occur along the fault planes of the Pomperaug Valley would seem
to be indicated by the fact that these planes have conducted the
underground waters to the surface at so many places within the
area of the Newark rocks. Tension joints should, however, be
more effective than compression joints in the control of drainage

THE RIVER SYSTEM OF ‘CONNECTICUT 483

lines, if it be assumed with Daubrée that the gaping fissure
planes have directed the streams in their courses. The pre-
sumptive evidence is, the writer believes, in favor of the former
development of the Newark rocks over a much larger territory
than that which they now occupy and probably over the entire
state of Connecticut.

It is an observation of much interest that the minor twig-like
branches of the streams, which in the deeply eroded mass of
crystallines must have been adjusted after the capping of sand-
stones had been removed, show an equally strong tendency with
the master streams to follow the special directions indicated by
the system as a whole."

The study of the fault system of the Pomperaug Basin offers,
however, another rational and natural explanation of the network
of streams, provided the assumption is made that the drainage
is adjusted to that formed in the geographic cycle which suc-
ceeded the deformation of the area. The system of parallel
faults has divided the area into vertical triangular, rhombic, or
rhomboidal prisms, which stand at different relative altitudes.
These prisms are found to be grouped into composite blocks of
increasingly higher orders, the peculiar property of each of
which is that the average altitude of its component prisms
approximates (however roughly) to a fixed value—the com-
posite blocks have an average level surface, although alternate
prisms or alternate subordinate blocks for short distances pro-
ject above or stand below the general level. The initial surface
formed by these prisms would be marked by canal-like structure
trenches (Graden) which follow the directions of fault planes
and which have stronger directive power, as regards streams, at
the junction of the trenches and at the crossings with the similar
trenches of other series.

Although the numerous generally curving fault planes dis-
covered by Davis in his extensive studies of the Newark of the
Connecticut valley have here been omitted from consideration,
for the reason that no close relationship to the orientation of

* Twenty-first Ann. Rept. U. S. Geol. Surv., Pt. III, p. 145, Fig. 52.

484 WILLIAM HERBERT HOBBS

streams is apparent in their directions, there is no intention
thereby to minimize the importance of those studies. The investi-
gation of the smaller Newark area of Connecticut, which by con-
trast was almost microscopic in its detail, brought out a series of
facts which for their interpretation required a totally different
theory from the one to which Professor Davis was led by his
studies. The two hypotheses have, however, this in common, that
the primary cause of the deformation in the Newark rocks is
assumed to be the compression of the crust within the area of
southern New England by a force of compression, the resultant
of which acted in a direction W. N. W. to E.S. E.

The present writer has been led to the conclusion that the
courses of large faults within this general area, if not approxi-
mately rectilinear, are in reality zigzags, the elements of which
are essentially right lines, examples of this kind being by no
means rare in, and, in fact, generally characteristic of, the Pom-
peraug Valley. It is not impossible that many of the larger
faults described by Professor Davis, if examined in greater
detail, might show this peculiarity, and perhaps also fall into
the system which has here and elsewhere been elaborated. The
numerous broken lines which are so apparent in the boundaries
of the trap hills of the Connecticut valley, as represented on the
topographic atlas sheets (e. g., Meriden sheet), would seem to
favor this view. It is in any case important, as it would seem
to the writer, to consider in two stages the dislocations brought
about mainly by a lateral compression of a section of the earth’s
crust, inasmuch as jointing (the production of planes of separa-
tion) is in these cases a necessary prerequisite to faulting (dis-
placement along planes of separation). The modern views of
geologists concerning joint planes produced by the shear from
lateral compressive stresses are now sufficiently in accord to
assume that vertical block faulting takes place along ready-formed
planes of jointing. In his description of the faulted area of south-
ern Norway, Brégger* has been careful to make this distinction.

T Wocweit.

THE RAVER SYSEEM OF CONNECTICGT 485

With the explanation of the numerous diversions of rivers
from their rectilinear stretches this paper is not especially con-
cerned, although some explanation of such diversions could be
found in the peculiarities of a region faulted like that of the
Pomperaug Valley. The modern criteria of the physiographer
deal adequately with this matter. It is the definite orientation
of water courses which the new science seems to have neglected.

WILLIAM HERBERT Hopes.

COMPOSITES GENBESIS OR DHE ARKANSAS WANE EM
THROUGH THE OZARK HIGHLANDS

On account of its singular course through the Ozark highland
region the Arkansas River presents, at the present time, unusual
geological interest. Its location in this part of its course, has
given rise to two very different opinions regarding the geologi-
cal age of the highlands; and also regarding the question as to
whether there are two distinct uplifts, as has been advocated by
the Arkansas geologists, or only one, as has been urged by
others who have worked in the region. Recently there have
accumulated new data bearing directly upon the problem.

Topographically, the Ozark highlands comprise two imposing,
nearly equal, elevated regions, separated from each other by a
broad deep trough—the Arkansas River valley. The vast plain
surrounding the highlands is about 400 feet above sea level on
the eastern side and twice this elevation on the west side. The
Arkansas River flows along on the horizon of this general
grade-plain. On the south side of the river the highlands rise
to heights of nearly 3000 feet above the sea; and on the north
to about 1800 feet.

Diverse apparently in topographic expression, lithologic
composition, geologic structure, and geological age, the district
south of the Arkansas River has been known as the Ouachita
Mountains, and that north of the stream the Ozark plateau. On
the assumption that there are two distinct uplifts, the river of
Arkansas is regarded as forming a natural dividing line between
the two regions. At first glance, the simplest explanation for
the position of the stream is forced upon the attention. Pre-
mising a single uplift, the accounting for the waterway’s course
meets with difficulties which, from superficial consideration,
appear well-nigh unsurmountable. The present note attempts
to sum up the evidence going to show that the facts actually
sustain the second premise.

486

GENESIS OF THE ARKANSAS VALLEY 487

In comparing the two districts, it is their differences and not
their points of resemblance which are most conspicuous. In
the Ouachita region the surface relief is notably mountainous,
long ridges and isolated peaks, with wide, flat-bottomed valleys
intervening. In the northern district the country is far from
appearing mountainous; it is, for the most part, a vast undulat-
ing plain, but sharply and deeply dissected around the borders,
with the streams flowing in v-shaped valleys. In the south the
rocks are more or less indurated or metamorphosed, and cut at
intervals by eruptives. Nowhere in the north do the strata

Ouacuita Mts Boston Mts Ozark PLATEAU

Fic. 1.— Peneplains of the Ozark region.

show alteration or evidence of the presence of eruptives. The
southern district is folded to a marked degree, approaching
closely the Appalachian structure; while the northern region is
only gently bowed. Regarding the geological ages of the two
districts, the Ouachita has been thought to have been upraised
towards the close of the Carboniferous; the northern area has
been commonly considered as having been an elevated region
ever since pre-Cambrian times. The present uplift, however, is
now believed to be of very recent origin; and the upward
movement is thought to be still in progress.

The physiographical history of the region and the relations
of the graded surfaces of the Ozark highlands are best indicated
in diagram (Fig. 1).

Two distinct base levels are ieee anible in the region. They
have been called the Cretaceous and the Tertiary peneplains.
These titles will be retained for the present. The first of the
peneplains rises out of the level savannas of the Mississippi
embayment, but soon becomes deeply broken as it rises and
passes into the Ouachita region. It is there believed to be

488 CHARLES URAIGE NLS)

continued northward in the mountain summits, which are often flat-
topped.‘ The later peneplain is thought to be represented in
the intermontane flats which are about 1500 feet lower than tops
of the mountains. The floor of the Arkansas valley is coinci-
dent with the Tertiary peneplain. Beyond the stream northward
the Tertiary surface rises rapidly according to Hershey,’ and
soon in the region of the Boston Mountains the two peneplains
practically merge. In Missouri only the Tertiary plain has
been distinguished, and this is regarded as forming the general
upland surface of the uplift.

It is probable that north of the Boston Mountains it will be
exceedingly difficult to differentiate at any point the two pene-
plains. Present evidence goes to show that during the interval
between the formation of the two peneplains in the south the
erosion in the north was comparatively slight, and resulted in
merely lowering the general surface of the plain already formed
during the Cretaceous.

Some time ago it was incidentally stated that the Ozark
highlands formed a single unit bowed up from the Red River to
the Missouri. The most obvious support for this conclusion is
found in the physiographic development of the region. But the
evidence is not alone from this source.

The physiographic data would indicate that in the Ozark
region the uprising since Cretaceous times has been not only
periodical in its character, but that it has been also differential.
Lately the movement has been more marked in the north than
in the south.

But there were special conditions existing that enabled the
Arkansas River to hold its own against the great barrier which
started to rise across its course. Ina limited belt in this part of
the Ozark region an enormous mass of non-resistant clay shales
had been deposited in Carboniferous times. The thickness
attained was much greater than that of the Carboniferous

t Arkansas Geol. Sury., Ann. Rept., 1890, Vol. III.

? American Geologist, Vol. XXVII, p. 25, 1901.

3 Missouri Geol. Surv., Vol. VIII, p. 331, 1895.

GENESIS OF THE ARKANSAS VALLEY 489

sediments anywhere else on the American continent, being
upward of 20,000 feet, according to Branner.*’ The peculiarities
of this great sequence of soft shales have lately been discussed
in some detail, and the real significance of the Arkansan series,
as it is called, pointed out.’

Thus, independent of whatever geological structure the
Arkansas valley may have, the enormous column of shales was
of such character as to enable the great stream to scoop out a
trough sufficiently vast and broad to give its topographic form

Ovacuita Mrs Bostom Mts. Ozark Prateau.

————
fer Coal Hleasunres

SSS Mississippran Le

Fic. 2.— Stratigraphy of the Ozark Highlands.

the effect of a depression between two uplifts.

There is another deceptive feature connected with the valley
of the Arkansas, that must be taken into consideration. Besides
being a topographical trough, the valley is also a structural trough.
A broad and shallow syncline stretches from the crest of the
Boston Mountains to the first range of the Ouachitas. The strata
closely folded in the extreme southern part of the highland dis-
trict spread out rapidly towards the north until they form gentle
undulations that are so characteristic of other parts of the Mis-
sissippi basin. The Ozark arch in Missouri constitutes the last
great swell northward. Its southern limb passes into the broad
syncline which contains the Arkansas valley. This relationship
of structure is represented by a north and south cross section
CPig.-2)):

The operation of different geological processes may be either

‘Am. Jour. Sci., (4), Vol. II, p. 235, 1896.

? Bull. Geol. Soc. America, Vol. XII, p. 173, 1901.

490 CHARLES Kk. KEVES

all compensating, or all cumulative in their effects. Between
the two extremes the sum of the antagonistic tendencies may
have very variable values. The present valley of the Arkansas
River as it crosses the Ozark highlands is a noteworthy illustra-
tion in which the combined effects of perfectly independent
processes are curiously cumulative in character. It is on this
account chiefly that the real facts concerning the development
of the great uplift have been so largely obscured.

Summing up: The different geological conditions when the
Arkansas River initiated its course across the Ozark region, (1)
an undeformed lowland flat in which the strata had been folded
to a marked degree before being beveled and the country
reduced to the state of a peneplain, (2) a remarkable, yet nar-
row, belt, bordered on either side by resistant rocks, of soft
shales of prodigious thickness which, when a new epoch of
uprising was inaugurated, enabled the stream to easily keep its
channel down to the general base level of the country surround-
ing the uplift, and (3) a broad structural trough, which, how-
ever, was only one of many synclines nearby and parallel to it—
were highly cumulative in effect in imparting to the uplift the
present aspect of twin elevations. By this singular combination
of geological conditions the Arkansas River instead of being
forced to turn aside by the great topographic dome which, out
of the Cretaceous peneplain, arose athwart its path, was able to
saw in two the arching strata.

Topographically, the Ozark highlands form two distinct ele-
vated regions. Structurally as well as topographically the
Arkansas valley is a trough. But structurally the Ozark high-
lands, as a whole, form an immense dome bowed from the Red

River to the Missouri.
CHARLES R. KEYES.

AE SCONDMCONTTRIBUTTON: [O° THE: NATURAL
HISTORY OF MARL?

THE writer recently published a paper on the relation of
algae to marl deposits.?_ Since it appeared, continued investiga -
tion has led to the discovery of additional confirmatory evidence
that the close relationship there pointed out of the algae, espe-
cially Chara, to marl or lake lime deposits, exists to even a
greater extent than was suspected.

Experimental work has been conducted along three lines, all
of which have been fairly productive of results, and a brief
account of this work may be of interest.

First, a series of mechanical analyses of typical white marl
from different localities were made. The method of analysis
used was a simple one, a modification of the beaker method
used in soil analysis. The sample, chosen at random from a
large specimen from the deposit under investigation, was dried
in an air bath at 110° C. for sufficient time to remove any included
moisture, and weighed. It was then mixed with distilled water
in a large beaker and thoroughly stirred with a rubber-tipped
glass rod, care being taken to stir it until all lumps caused by the
adhesion of the finer particles to the coarser were broken up
Care ‘was also taken that no more crushing should take place
than was absolutely necessary. After all lumps were disinte-
grated, the water, with the finer particles suspended in it, was
poured off into another beaker and fresh water was added to
the first and the material was again stirred. This was continued
until water poured into the first beaker was nearly free from
finer matter and became clear on standing a few moments. The
coarse material left in the bottom of the beaker was dried, sorted
into various grades by a series of sieves and each grade weighed.

* Printed by permission of ALFRED C. LANE, State Geologist of Michigan,

2 Jour. GEOL., Vol. VIII, No. 6. September—October Igoo.

491

492 CHARLES A. DAVIS:

The finer material was also sorted by stirring, settling, and
decantation, and the matter of different degrees of fineness dried
and weighed. The finest matter was usually separated from the
water by filtering through a dried and weighed filter, and the
water concentrated by evaporation and again filtered to remove
any of the calcium carbonate dissolved in the various processes,
and the final residue of water was evaporated in a watch glass
and weighed. An exceedingly interesting feature of this latter
experiment was the finding of a water soluble calcium salt, in
small proportion, it is true, but still easily weighable and not to
be neglected. The results of such an analysis of a sample from
the Cedar Lake marl beds in Montcalm county, Mich., gave the
following results. The sample used was collected from a hole
made with a spade by cutting away the turf over the marl, then
taking out sufficient marl to be sure that there was no peat or
other surface matter present and the material used taken from a
spadeful thrown out from two or three feet below this level.
From this sample about thirty grams were taken and treated
as described above, and after the finer material had been sepa-
rated from the coarser by washing and drying, the latter was
passed through a set of standard gauge sieves, twenty, forty,
sixty, eighty, and one hundred meshes to the linear inch, after
which all shells and recognizable shell fragments, sand grains and
vegetable débris, up to the sixty-mesh siftings, were removed
and weighed separately. This gave the following grades: (1)
Material too coarse to pass through the twenty-mesh sieve, (2)
that held by the forty-mesh, (3) that held by the sixty-mesh,
(4) that held by the eighty-mesh, (5) that held by the hun-
dred-mesh, (6) that which passed through the hundred-mesh,
(7) that which was filtered out, (8) water soluble salts, (9)
shells, shell fragments, and miscellaneous matter. .

The following is the result of the analysis of Cedar Lade marl
made and graded as described :

THE NATURAL AISTORY OF MARL 493

Grade (1) - - - - - = 32.25°%
2) = - - - ~ 6.06
SA) iets A Ws ae cits tC ge a NS 3)
4) = - - - - 2.90
Se - - - - - 4.81
6) - - - : - 15.64
7) Bree. Ba ef ;
8) { 30 52
)

SO a ae 0.28

(
(
(
(
(
(
(
(
(

9

100.04

A second analysis was made from a specimen made up of
twenty samples taken by boring with an auger over about one-
half of the deposit at Littlefield Lake, Isabella county, most of
the samples coming from a depth of at least twenty feet, six to
eight meters below the surface of the deposit. This analysis
gave the following results:

sae - - - - - - ee
2 - - = 5 = 14.4
(3) - - - - nT 2a76

(4) = - - - - 2.56
Tete eet
Os ee ee nie ie

(8) - . - - - eee yen

(9) ‘ = S = - 1.04
99-79

A third sample, from the holdings of the Michigan Portland
Cement Company at Coldwater, Mich., a fine high grade white
marl, very powdery, gave the following: Weight of sample,
20.— grams.

Grade (1) - - . - - - 0.36%
Qe ee See ans
(6) ow te a Ee G5
ANN aa are eae 3.34
(5) - : = 3 - - 6.44
(6) ee 0 ee 20-00
Oh ea : : = = - 49.12
(8) - - Not determined.
(OV eee ees C00

Loss and soluble matter (by
difference) = - - - 1.02
100.00

‘In this case determined by drying down the residue and weighing.

2The soluble matter contains a certain undetermined amount of alkaline chlo-
rides as well as a soluble calcium salt.

4904 CHARLES A. DAVIS

These samples represent (1) central, (2) north central, and
(3) southern parts of the lower peninsula respectively, and may
be taken as typical of the marl deposits of Michigan. When it
is stated that, in general, it is easily possible to recognize with a
simple microscope the particles which are held by the one
hundred-mesh sieve, or even those which pass through it, if the
finer matter has been carefully separated-by washing, as charac-
teristic Chara incrustation or Schizothrix concretions, it will be
seen that these results show conclusively that a large part of the
marl from these three samples is identifiable as of algal origin,
and studies of the marls from other localities give similar
results.

The Coldwater sample (3) was exceedingly fine in texture,
and it was difficult to avoid loss in sorting and weighing, as
every current of air carried away some of the particles, and
some also adhered to sieves and weighing dishes in spite of all
precautions. Even this sample shows nearly 50 per cent. of
easily identified Chara incrustation. The fineness of the parti-
cles in a given marl bed varies much in different parts of the
bed, and the degree of fineness is probably chiefly dependent
upon the conditions of current and wave action under which the
bed was formed, that which was deposited where the wave or
current action was strong being coarser than that in stiller water
or that on the lee side of exposed banks. This fact was noted
at Littlefield Lake when samples of marl were collected along
exposed shores above the wave line, which were 95 per cent.
coarse fragments of Chara incrustation and Schizothrix nodules,
while in other parts of the shore line the marl was of such fine-
ness that it was like fine, white clay. The fragments of the
Chara incrustation are generally easily recognized even when of
minute size, because they preserve, usually very perfectly, the
peculiar form of the stem and branches of the plant. This
structure of the stem and branches is, in brief, a series of small
tubes, grouped about a larger central one, and is easily seen
with the unaided eye in larger fragments. Even when the tubes
have been crushed, as is the case with many of the thinner ones,

LAE NALTORAL: AISTORY OF MARL 495

it is frequently possible to recognize fragments of. them with the
compound microscope. Finally the incrustation is distinctly
crystalline in ultimate form of the constituent particles, and
when it has disintegrated the crystals and their fragments are
found to constitute a large per cent. of the finer particles of the
resulting marl. On the growing tips of the younger branches
and the leaves of Chara, numbers of isolated crystals of calcium
carbonate may be seen. Farther back on stems and branches
the crystals become more numerous, then coalesce into a thin,
fragile covering, and finally on the lower part of the plant the
covering becomes dense and thick. It is evident therefore that
the decay of the younger parts of the plants would furnish a
mass of more or less free or loosely aggregated crystals of
microscopic size, which would retain their crystalline form, in
some degree at least, for an indefinite time and be recognizable,
hence the presence of microcrystals in marl may furnish addi-
tional evidence as to the origin of the deposits.

The larger fragments of Chara incrustation as found in mar]
are frequently much thicker and heavier than any which occur
among fragments of recent origin, z. ¢., those obtained from any
part of living, vigorously growing Chara from the beds of the
plant existing in ponds from which the marl may have been
obtained. While this subject needs further investigation, it is
probable that such thickened incrustations have originated in
several ways, the principal ones being, if the writer’s notes have
any bearing on the subject, as follows:

1. On short, stunted plants that grow for a long time on
unfavorable soil, such as sand or pure marl. Such plants have
relatively very short internodes and generally thick incrusta-
tions, much thicker than those on plants growing normally.

2. From the growth of the lime-secreting, blue-green algae,
such as Schizothrix, Rivularia, etc., either upon living chara or
upon the fragments of broken incrustation, as a nucleus. Such
a growth might produce considerable thickening of the Chara
incrustation.

3. From the inclusion of the fragments within the nodules

496 CHARLES A. DAVIS

formed by the growth of the incrusting blue-green algae, in shallow
water, and the subsequent destruction of the nodules by wave or
other disintegrating action. In this case a thickened fragment .
may be left either free or attached to other material. Several
fragments may be cemented.together, and such aggregations
have been observed.

4. By the deposition of calcium carbonate on fragments of
incrustation, the source of the salt deposited being soluble
calcium-organic compounds left free in the water by the decay
of dead Chara, the precipitation being caused by the reducing
action of chemical compounds derived from the decay of
organic matter or the growth of bacteria. It may be conceived
also that the calcium carbonate thus deposited might also act as
a cementing material to fasten the finer particles of marl to the
incrustation as a nucleus.

5. By the deposition in more or less coarsely crystalline
form of calcium carbonate which is dissolved by water perco-
lating through the marl. This is probably considerable in
amount and takes place in a manner analogous to, if not identical
with, the formation of concretions in clays and shales. It is
probable that in this way the crystals may be formed, which rather
rarely are found filling the cavities in the Chara incrustations,
left by the large axial cells of the plant. The fact that in the
great majority of cases these cell cavities are entirely empty or
are simply mechanically filled with fine particles of marl, is a
most serious objection to considering that this form of chemical
precipitation is an important one in the history of marl, but that
it is occasionally operative is extremely probable.

6. It is possible that the thick incrustations may have been
formed at some earlier period in the history of the lakes when
conditions for the development of the plant forms and their
activities were greater. This is not probable however, for the
thick incrustations are often found from the surface of the marl
beds throughout the entire deposits.

In addition to the marl analyses given above a check analysis
was made of a specimen of material made up from the washings

THE NATURAL HISTORY OF MARL 497

and fragments of a mass of Chara plants collected from Cedar
Lake, and allowed to die slowly and to break up in water kept
cold and fresh by conducting a small stream from the hydrant
through it. The plants gradually died, broke up, and settled to
the bottom of the containing vessel and seemed to undergo
farther disintegration there from the growth of fungi, eventually
forming a relatively finely divided deposit which was of rather
dark color, when wet. A quantity of this was dried at 100°C.,
some of the longer and larger fragments of stems were removed,
and the residue was weighed and subjected to the same treat-
ment as the marl samples. The analysis gave the following :

Grade (1) - - - - - - 1.12%
(2) - - - - - 24.43
Gee i rags
(4) - . - - - 8.26
CN ee ek
Bp ee
(8) ey eran eee Ae afer aa ae ei03.0
(9) - - - - - 0.12

Soluble organic matter and loss” - era Os4il

100.00

It will be seen that nearly as much fine matter was present in
this material as in the finest of the marls analyzed, and that the
finer grades of sifted material are quite as well represented as
in the finer marl. The material is somewhat more bulky for a
given weight and is perhaps slightly darker in color, but not
much more so than many samples of marl.

Grade for grade it is identical in appearance and structure to
the marl samples, and the only possible difference that can be
detected is the slightly green tint and the organic matter present
in the plant residue. It is also noticeable that the larger pieces
do not show as thick an incrustation as do the larger pieces from
the marl samples and, of course, Schizothrix and other coarse
matter is not present.

It will be seen by inspecting the analyses, that shells and
recognizable shell fragments are but a very insignificant part of

498 CHARLES A. DAVIS

the total quantity of the marl. It is surprisingly small when all
thingsare taken into account. While it is probably true that not all
the minute shell fragments have been separated in any of these
analyses, it is also true that the weight of such overlooked
particles is more than counterbalanced by marl fragments which
are included within the cavities of the whole shells, and adhere
to both broken and whole shells in crevices and sculpturings in
such a way as to refuse to become separated in the processes of
washing out the marl. The whole shells are mainly small, fragile
forms, many of them immature, and it is evident that they would
be broken by any action that would crush the Chara incrusta-
tion.

A second line of investigation took into consideration the
milky appearance of the waters of some marl lakes. This’ has
been considered by some investigators as possibly due to the
presence of calcium carbonate precipitated from the water
either by the liberation of dissolved carbon dioxid from the
water and hence from the calcium bicarbonate or by change of
temperature of the water after it has reached the lakes.

The writer has not found among the marl lakes of the south-
ern peninsula of Michigan that those with turbid water were
common, even where marl banks were apparently forming with
considerable rapidity.

‘“Merl”’ or Marl Lake in Montcalm county, situated on the
same stream as Cedar Lake and a mile or more’ below it, is,
however, one of the lakes in which the water is usually of
almost milky whiteness, and has sufficient suspended matter in
it to render it nearly opaque for depths of a meter or a little
more. The conditions in this lake are widely different from
those at Cedar Lake and other marl lakes in the vicinity and are
suggestive of the cause of the turbidity. At Cedar Lake there
is a border of grassy or sedgy marsh extending around the lake
on three sides and generally underlaid by marl, and the lake
bottom slopes sharply and abruptly from the edge of the marsh
to a depth of at least ten meters. In other words the lake is
simply a deep hole, with steep sides, and perhaps represents the

THE NATURAL.HAHISTORY.OF MARL 499

deepest part of the more extensive lake which formerly occu-
pied the area included by the marsh and marl beds. This marsh
covering is general on the marl beds of the region and the lake
may be said to be a typical marl lake for the locality in which
it lies, for there are several others near by which are practically
identical in essential points of structure.

At Marl Lake, however, the filling of the lake has not reached
the same stage. There is practically no open marsh, but the
lake is shallow for seventy-five or a hundred meters from the
shore, then abruptly deepens to an undetermined depth over a
relatively small area. The bottom over the shallow area is of
pure white marl, and the water is apparently not more than sixty
or seventy centimeters deep at the margin of the central hole,
while near the shore it is scarcely one-third as deep. In brief,
here is a lake in which there is a broad platform of marl sur-
rounding a deep hole, which again is all that remains of the
deep water of a lake which is filling with marl. Boring shows
that the bed of the lake is nearly as far below the surface under
the marl platform, as where the marl has not yet been deposited.

Upon the shoreward edges of the platform and in small
areas farther out upon it, the turf-forming plants are beginning
to establish themselves, but as yet they have not made any
marked impression, seeming to have a hard struggle to get a
foothold. The conditions are then a broad area of shallow
water overlying a wide platform of marl, which, if a strong wind
should reach it, would be stirred to its depths, and with it the
lighter parts of the marl upon which it rests. The marl thus
stirred up, in turn, is carried to all parts of the lake by surface
and other currents and makes the water turbid. These facts led
to an investigation as to the rapidity with which marl, once
stirred up, would settle out of perfectly still water, and some
interesting results were obtained. The experiments were made
as follows: (1) A glass tube 1.58 meters long and 2.5 centi-
meters wide was filled with distilled water, into which a quantity
of finely divided marl was turned and the tube was shaken to
insure a thorough mixing of water and marl. The tube was then

500 CHARLES A. DAVIS

clamped into a vertical position and left perfectly still until the
marl had settled out, notes being made, daily at first, of the rate
of settling. In the beginning, the heavier particles settled
rapidly, forming, as does clay in settling from water with which
it is mixed, stratification planes, which, however, after a few days
disappeared, and only the lighter parts of the marl remained in
suspension. These were distinctly visible for five weeks, on
looking through the tube towards a window, and at the end of
six weeks, a black object lowered into the tube in a well-lighted
room, was not visible beyond ninety centimeters from the surface
of the water. (2) A glass cylinder with a foot, 38 high and 7™
wide, having a capacity of a little more than a liter was nearly
filled with distilled water and the residue from the washings of a
sample used in analysis was thoroughly mixed with it, and set
aside, notes being made as before. This material subsided rather
more slowly than the other, and at the end of ten weeks, under
daylight illumination, the bottom of the vessel was barely visible
when one looked down through the water from above.

The results obtained by Barus’ in his work on the subsidence
of solid matter into suspension, in liquids, show that settling is
much more rapid in water containing dissolved salts, even small
proportion, than in distilled water, and the foregoing experi-
ments were checked as follows: (1) A cylinder about the size
of the one used in the second experiment was filled with water
in which a small amount of calcium chlorid had been dissolved,
and ammonium carbonate was added until a precipitate was
formed. This was stirred thoroughly and left to settle. In
three days the precipitate had fully subsided and the liquid was
clear. (2) Two cylinders of equal size were filled, one with
distilled water, the other with water from a river fed, in part, by
marl lakes. Equal quantities of fine marl were shaken up with
the water and the rate of settling was compared. The marl was
not as fine as that used in the other studies and settled more
rapidly. The river water was clear in fifteen days, while the

"CARL Barus: ‘‘Subsidence of Fine Solid Particles in Liquids.” Bull. of the
U. S. Geol. Surv., No. 36.

THE NATURAL HISTORY OF MARL 501

distilled water was not clear when the experiment terminated,
but was nearly so, showing that the subsidence was not quite so
rapid in distilled water as in natural lake or river water.

These results indicate that,if for any cause the marl in a
marl lake is stirred up effectually, as it may be in those where
the beds are exposed to wave action, the water will remain
turbid for some time; even in summer time the chances are that
there will be sufficiently frequent high winds to keep the water
always turbid. It may be stated that in some of the lakes which
have been studied by the writer, the marl beds have filled the
entire lake to within a fraction of a meter of the surface of the
water, with some parts only a few centimeters deep. Until such
shallows are occupied by vegetation the water is likely to be
turbid from the mechanical action of waves upon the deposits.
At Littlefield Lake, described elsewhere,’ the water is only slightly
turbid, although there are extensive shallows and exposed banks,
but there the body of water is extensive and of considerable
depth, while the greater part of the exposed marl is granular
and the particles too coarse to be held long in suspension, and
the finer deposits too small and too well protected to be reached
by effective waves, so that the amount of suspended marl is not
great enough to produce marked turbidity in the entire body of
water.

It may be worthy of note that the residue of suspended
matter, filtered out from the sample of Chara fragments (analy-
sis (4) above) was sufficiently fine to give a marked turbidity
to distilled water for several days, and at the time of filtering
had not subsided, demonstrating the fact that very finely divided
particles may originate from the simple breaking up of the
Chara plants by ordinary decomposition of the vegetable matter.

It is difficult to account for the fact that the deeper parts of
marl lakes are generally free from any thick deposits of a cal-
careous nature. Lack of records of sufficient exploration makes
any statement purely tentative, but about seven to nine meters?

tC. A. Davis: ‘‘A Remarkable Marl Lake,” Jour. GEOL., Vol. VIII, No. 6.

_ ? WESENBERG-LUND: Lake-Lime, Pea Ore, Lake Gytje. Saertryk af Medde-
lelser fra Dansk Geologisk Forening, No. 7, p. 156.

502 CHARLES A. DAVIS

seems to be the limit of depth of recorded occurrence of Chara
plants. The remains of the plants, then, would only accumu-
late, in place, above that depth, and the material reaching
greater depths would be that held in suspension in the water,
and hence be relatively very small in quantity and accumulate
slowly. A possible additional cause of slow accumulation is
that in the greater depths, z.¢., over ten meters, the greater
abundance of dissolved carbon dioxid held in solution by pres-
sure dissolves the finer particles of marl which reach these
depths.

From these investigations it seems (1) that marl, even of
the very white pulverulent type, is really made up of a mixture
of coarser and finer matter covered up and concealed by the
finer particles, which act as the binding material. (2) That the
coarser material is present in the proportion of from 50 to 95
per cent. (3) That this coarser material is easily recognizable
with the unaided eye and hand lens, as the incrustation pro-
duced on the algae, Schizothrix and Chara, principally the latter,
to particles less than one one-hundredth of an inch in diameter.
(4) That the finer matter is largely recognizable under the com-
pound microscope as crystalline in structure, and is derived
from the algal incrustation by the breaking up, through decay
of the plants, of the thinner and more fragile parts, or by dis-
integration of the younger parts not fully covered. (5) That
some of this finer matter is capable of remaining suspended in
water a sufficiently long time, after being shaken up with it, to
make it unnecessary to advance any other hypothesis to explain
the turbidity of the waters of some marl lakes, than that it is
caused by mechanical stirring up the marl by wave or other
agency. (6) That shells and shell remains are not important
factors in the production of the marl beds which are of largest
extent. (7) That there is in marl a small amount of a water
soluble calcium salt, readily soluble in distilled water, after
complete evaporation.

2A, J. PIETERS: Plants of Lake St. Clair, Bull. Mich. Fish Commission, No. 2,
p- 6.

THE NATURAL HISTORY OF MARL 503

Studies were undertaken to determine the method of concen-
tration and precipitation of the calcium carbonate by Chara.

As has already been indicated elsewhere, the calcium car-
bonate is present on the outside of the plant as an incrustation,
and this is made up of crystals, which are rather remote and
scattered on the growing parts of the plants, and form a com-
plete covering on the older parts, which is uniformly thicker on
the basal joints of the stems than it is on the upper ones. Con-
sidering the hypothesis that the deposition of the salt was the
result of purely external chemical action as not fully capable of
satisfying all the existing conditions, the formation of the
incrustation was taken up as a biological problem, and an inves-
tigation was made upon the cell contents of Chara, first, micro-
scopically by the study of thin sections. Various parts of the
plant were sectioned while still living, and the attempt was
made to find out if the calcium carbonate was present as a part
of the cell contents in recognizable crystalline form. Various
parts of the plant were examined, but no crystals undoubtedly in
place among the contents of the cells were observed, although
numbers were to be seen on outside walls of all cells.

Next an attempt was made to determine the presence of the
calcium in soluble form in the cell contents, by the use of dilute
neutral solution of ammonium oxalate. An immediate response
to the test was received by the formation of large numbers of
minute, characteristic, octahedral crystals of calcium oxalate
on the surface and imbedded in the contracted protoplasmic
contents of the cell. The number of these crystals was so
large, and they were so evenly distributed through the cell con-
tents, that it was evident that a large amount of some soluble
calcium salt was diffused through the cell sap of the plant.
The next step was to isolate this compound and to deter-
mine its composition. A considerable quantity of the growing
tips of Chara were rubbed up in a mortar and the pulp was
thoroughly extracted with distilled water. This water extract was
filtered and tested to determine the presence of calcium and an
abundant precipitate obtained by using ammonium oxalate,

504 CHARLES A. DAVIS

which, on being separated and tested, proved to be calcium
oxalate. It was evident that the calcium salt in the plant was
stable and readily soluble in water. This latter fact was farther
demonstrated by evaporating some of the extract to dryness
and again taking it up with water. Almost the entire amount
of the calcium salt was redissolved, only a small portion of it
becoming insoluble and precipitating as the carbonate. This
ready solubility demonstrated the fact that the salt was not
derived from the incrustation on portions of the plant used, and
the same fact excluded from the list of possible compounds,
salts of some of the mere common organic acids found in plant
juices, which are insoluble.

Qualitative chemical tests were, however, made to determine,
if possible, whether any of these acids were present with nega-
tive results, and it was demonstrated by this means that there
was but a single salt present and not a mixture. Search was
then made to determine the acid present, and a result obtained
which was so unexpected that it was seriously questioned and
the work was gone over again. The second result confirmed the
the first and the work of ascertaining the correctness of these
two results was turned over to Mr. F. E. West, Instructor in
Chemistry in Alma College, who had had special training and
much practice in organic analysis. His work was done entirely
independently, with material gathered at a different season, and
by another method of analysis, but his results were identical
with my own, and show that calcium exists in the water extract
of Chara as calcium succinate. The fact that the succinate is
one of the few water soluble calcium salts and that there is a
soluble salt of the metal in the cell sap of the plant makes it
probable that this is the compound of the metal which the plant
accumulates in its cells.

It is not possible from actual investigation to explain the
method by which the calcium salt is abstracted from the water,
where it exists as the acid- or bi-carbonate or the sulphate™ in

‘It has been shown that Chara decomposes several calcium salts, the sulphate
among others.

THE NATURAL HISTORY OF MARL 505

small per cent., and is concentrated in the cells of the plant as
calcium succinate and later deposited upon the outside of the
same cells as the normal or monocarbonate in crystalline form
in considerable quantities.

Some culture experiments which were undertaken by the
writer to determine under what conditions of soil, light, and
temperature Chara thrives best, incidentally demonstrated that
the plant actually gets its lime from the water and not from the
soil. One of the soils which was used as a substratum in which
to grow plants was pure quartz sea-sand which had been washed
with acid to remove any traces of calcium salt which might be
present. The plants grew in this medium readily, and on the
newer parts developed nearly, if not quite, as many calcium
carbonate crystals as plants growing in pure marl. It should be
apparent, however, to even the casual observer that the plants
cannot take all the lime they precipitate from the soil, or even a
considerable part of it, for if they did the marl beds, being made
up principally of Chara remains, would never have accumulated,
for the material would have been used over and over again and
could not increase much-in amount, if it increased at all. In
the present state of our knowledge of the life processes of
aquatic plants, it seems hardly possible to state the probable
method of the formation of the calcium succinate, or even the
probable use of it to the plant, and no attempt will be made by
the writer in the present paper todo so. It does seem probable,
however, that this compound accumulates in the cells, until it
reaches sufficient density to begin to diffuse through the cell
walls by osmosis. Outside the cells, or in its passage through
the walls, it is decomposed directly into the carbonate, possibly
by oxidation of the succinic acid by free oxygen given off by
the plants, possibly by some substance in the cell walls, or,
more probably, by the decomposition of the acid by some of
the organic compounds in the water, such as the organic fer-
ments, due to bacterial growth in the organic débris at the
bottom of the mass of growing Chara. The water extract of
Chara rapidly changes on standing, undergoing putrefactive

506 CHARLES A. DAVIS

decomposition, becomes exceedingly offensive in odors devel-
oped, and calcium carbonate crystallizes out on the bottom and
sides of the containing vessel, while the succinic acid disappears,
gas, possibly carbon dioxid, being given off more or less abun-
dantly. Whether these changes takes place on the outside of
the living plants, in the cell walls, or in the water surrounding
the plants has not yet been determined.

Sufficient evidence is here presented, however, if the writer’s
conclusions are correct, to show that the plants under discussion
are active agents in the concentration of calcium salts in the
fresh water lakes of Michigan, and that they alone have pro-
duced a very large part of the marl which has accumulated in
these lakes. It seems probable also that the principles devel-
oped by these studies are of very wide application in working
out problems presented by formations developed under similar
conditions elsewhere.

CHARLES A. Davis.
ALMA COLLEGE,

July 1, 1901.

PERKNITE (LIME-MAGNESIA ROCKS):

THERE are sometimes associated with diorites, gabbros and
peridotites, dark rocks composed largely, or entirely, of mono-
clinic amphibole or pyroxene, or both. These rocks differ
mineralogically from diorites and gabbros, in containing little
or no feldspar, and from peridotites in containing rhombic
- pyroxine or olivine in relatively small amount, if present at all.
Chemically these rocks contain less alumina than diorites and
gabbros, and less magnesia than peridotites. They are low in
alumina and in the alkalis, moderately rich in lime, magnesia,
and the iron oxides.

The chief constituents of perknite are monoclinic amphibole
and monoclinic pyroxene; the secondary constituents rhombic
pyroxene, olivine and feldspar; the accessories biotite, iron ore,
etc., but only one of the primary constituents may be present
with none of the secondary constituents or accessories. The
existence of this group of rocks has long been recognized, but
from their occurrence usually in small masses, and from the fact
that many of them are of simple composition so that the self-
explanatory names pyroxenite and amphibolite or hornblendite
have answered, they have never been grouped together under
one name.

In the State of New York? and in California3 there are rocks
containing both monoclinic pyroxene and amphibole as principal
constituents, and doubtless this is likewise the case in many
other parts of the world. Moreover, in California such rocks
form areas of geological importance. There is, therefore, some
reason in grouping all of these lime-magnesia rocks under a
common name. It is proposed to call the group perknite from

‘Published by permission of the Director of the U. S. Geological Survey.

2G. H. WILLIAMS: Am. Jour. Sci., Vol. XX XI, 1886, p. 40.

3 TURNER: Am. Jour. Sci., Vol. V, 1898, p. 423. Turner and Ransome. Sonora
folio.

507

508 H. W. TURNER

the Greek word vepxvos, meaning dark. It will include grano-
lites of the following specific names:

Pyroxenite.

Hornblendite (Williams).

Websterite (Diallage and ortho-rhombic pyroxene) (Wil-
liams).

Diallagite.

Hornblende-hypersthene rock (Merrill).

Amphibole-pyroxene rock (Turner).

The group may be graphically represented by the method
employed by Hobbs’ and his representation of a composite
pyroxenite will approximate to that of a typical perknite. The
following table of analyses will give the reader a notion of the
composition of the rocks which may be properly included in
this group.

1. Hornblendite.— Geo. Steiger, analyst. This partial analy-
sis is here published for the first time. The rock is from a dike
cutting through the basement complex and overlying Cambrian
rocks, 2km north of Silver Peak village, in Esmeralda county,
Nev. It is composed chiefly of green hornblende with some
feldspar. The rock grades into a basic diorite.

2. Amphibole-pyroxene rock.—W. ¥.. Hillebrand, analyst. Not
before published. Rocks of this type are very abundant in
Mariposa county, Cal. Mr. F. M. Anderson, of the University
of California, has likewise collected them in northern California.
This rock in its typical development is composed of original
pyroxene and amphibole in grains of nearly equal size, with a lit-
tle quartz and pyrrhotite. Scattered through the rock are pheno-
crysts about one centimeter in diameter, of brown amphibole,
which contain in a poikilitic manner, as inclusions, the constitu-
ents of the groundmass.

3. Perknite (author’s name, fer:dotite).— Belchertown. Bull.
U. S. Geol. Survey, No. 168, p.30. L.G. Eakins, analyst. The
rock is composed of hornblende, pyroxene, biotite, olivine and
magnetite.

*Jour. GEOL., Vol. VIII, 1900, p. 14.

PERKNITE (LIME-MAGNESIA ROCKS) 509
ANALYSES OF PERKNITES; LIME-MAGNESIA ROCKS.
I II Ill IV NV; VI VII
Amphibole- .
Name er enG pyroxene Perknite | Pyroxonite | Websterite Websterite Sosa ea
SiO, 46.28 48.04 48.63 50.80 53625 53.21 52.58
INI AO) Brae ee eral be eCROeee 7.82 eee 3.40 2.80 1.94 3.69
HCE @sescneveyerall aasiere 2,01 2201 130 .69 1.44 1.90
Me Oise ctes. aval» acta « One2e 3.90 8.11 5.93 TO, 6.50
MgO 19.54 Tene Ave Yio) ora 19.91 20.78 20.86
CaO 9.91 13.01 13.04 12 ot 16,22 Tol He n22
NEMO naaod ge 2E2T .69 34 ; t .19 sau 22
KOs. 1.89 .48 28 pace trace 07 .10
He O— T1090 2Cl <n. SL .05 =04
ESOT TOC) ee 2.90 2 52 .24 .87 37
U1 O) Sretaverererses)| woe emacs. 1.16 -47 NONE yy Geeta 720 ali
Lx Olnermee ener wocsteveites Ally wescnsins. [hme eiteersn Pie rene etston PIDs rveds geexd trace
COR ares woes) dc none tEACC allt Mecoreress wnat pacts F3K(0)
Peo Ojem werencteuerei|| Meera es trace 220 taACe Tal) ane ees trace
SOpeece all eiaete BQ BSC rata TRACE, Seis ecco 1/28. 2, ne acess
SI Reta ee see's HOO MMA er rertcts Clee ae tena, 1 Fes, 03
NOOR ree cai woetie ei eae: 4 stcceae ill eras Wl) oe ener .03
Ci O Gereccgsest|h veeusser’s .90 36 2 54 20 |
NiO
INI OMe cntlbm ita UE he ceae lan ariche, 1 Ah Rie mays .07 -E .03
CoO
Min Ore rresrsca| © ideas none 1a) aaly, .09 722 aint
other
BAO Pre casa le nears none ieewe A casio oll Same e none 4 | constitu-
ents
STO Beers ericsvstee | ecctnee cr alll) cette O(a p ceric |!) aobernte none .50
eV Orm arama, sac 0) ageiesee | ostrewe Aaece y hte s trace sacs
MRO Galley stssers. | anveys cote 100.06 100.13 100.03 99.98 100.47 100.49
TRESSEO Cea. IA uceeccrtce ©45
99.61 |
4. Pyroxenite.—Johnnycake road, Baltimore. Bull. U. S.

Geol. Survey, No: 168, p. 42.

and diallage.

5. Websterite.—From Mt. Diablo.
Bull U. 5S. Geol. Survey, No. 168, p.-213, and Bull. Geol. Soc.
Am., Vol. II, p. 406; analysis No. 242.
of orthorhombic pyroxene and diallage.

Composed entirely of hypersthene

W. H. Melville, analyst.

The rock is composed

*Owing to analytical reasons this determination of ferrous iron is unsatis-
factory.—W. F. Hillebrand.

2 This determination somewhat doubtful.—W. F. H.

510 H. W. TURNER

6. Websterite— Oakwood, Cecil county, Md. Composed of
hypersthene and diallage. W.F. Hillebrand, analyst. Bull. U.
S. Geol. Survey, No. 168, p. 43.

7. Composite pyroxentte.— Hobbs, Jour. GEOL., Vol. VIII, p. 30.
This analysis is a composite from three analyses of pyroxenites,
and one analysis of a hornblende-hypersthene rock from Gallatin
county, Mont.

Professor J. S. Diller, in his bulletin on ‘The Educational
Series of Rock Specimens,’’* introduces three specimens which
would fall into the perknite group.

No. 110, a pyroxenite, is described by Professor George H.
Williams.

No. 111, feldspathic peridotite, is described by Professor
George H. Williams.

No. 113, Cortlandite (hornblende-peridotite), is described
by Williams and Iddings. The rock is composed of brown
hornblende, olivine, pyroxene, biotite, feldspar, and magne-
tite.

Some rocks that have been termed wehrlite will fall also into
this group.

EFFUSIVE AND DIKE ROCKS

Corresponding to the plutonic group of perknite there
undoubtedly occur effusive and dike rocks. Professor Rosen-
busch regards hornblende-picrites as effusive, and some of these
have the chemical and mineral composition of perknite. This
will also be true of augitites, since in these augite or monoclinic
pyroxene is the chief constituent.

Professor J. P. Iddings has been kind enough to criticise the
above paper and calls my attention to the fact that my definition
of perknite would bring into the group the kimberlite of Ken-
tucky with 9.46 per cent. of lime, and the kimberlite of South
Africa with 9.60 per cent. of lime, as wellas the amphibole-peri-
dotite of Schriesheim with 7.22 per cent. of lime. He also calls
attention to the fact that with hypersthene-enstatite rocks there

™ Bull. U. S. Geol. Surv., No. 168.

PERKNITE (LIME-MAGNESIA ROCKS) 511

may be very little lime present and yet they would not be peri-
dotite. This brings us to the re-definition of peridotite, and I
should define a peridotite chemically as a magnesia rock with
usually less than 6 per cent. of lime. This would put a rock
composed entirely of hypersthene or enstatite with the peri-
dotites, where they certainly belong chemically.

H. W. Turner.

SAN FRANCISCO, CAL.,
June 20, Igor.

THE BORDER-LINE BETWEEN PALEOZOIC AND
MESOZOIC IN WESTERN AMERICA

THERE have been in recent years in America many contro-
versies as to the Silurian-Devonian and the Devonian-Carbonif-
erous boundaries; but American geologists have. always felt
secure as to the line of demarcation between the Paleozoic and
the Mesozoic. This has always been thought to be marked by a
grand chasm, a hiatus in stratigraphy and a break in life,
accompanying a great change in physical geography, all of
which is true of the region east of the Rocky Mountains.

But later discoveriesin the Great Basin region have shown
that the gap is at least partly filled out by marine sediments,
and that the hiatus is not universal in America. Thé most
important of these discoveries was the finding of marine Permian
in northern Texas, with forms suggestive of the Mesozoic types
of life, and the finding in southeastern Idaho of marine Lower
Trias, with a fauna reminiscent of the Paleozoic.

Recently there has been evident among geologists a tendency
to revert to first conditions in determining the boundaries of
geologic systems. They have been inclined to use unconformi-
ties as the dividing lines between geologic groups, and to
look upon geologic systems as realities, and not mere names for
the convenience of stratigraphers. This must mean that they
regard the geologic events that delimited the systems as of
world-wide effectiveness, or even of cosmic origin. But almost
every trangression of the sea has been found to be balanced by
a retreat elsewhere, every uplift to have its correlative subsi-
dence. And instill other parts of the earth there may have been
neither uplift nor subsidence. Even the Appalachian revolution
did not effect the western Carboniferous Sea, and this is gener-
ally admitted to have been one of the greatest events in the
dynamic history of North America.

In defining the upper Paleozoic and the lower Mesozoic no

512

PALEOZOIC AND MESOZOIC BORDER-LINE 513

account was taken of conditions of formation, and this has led
to endless difficulties in correlation. The Upper Carboniferous
was based entirely on lacustrine deposits, and the identification
of its marine equivalent is still making trouble for geologists.
The Permian was based on a basin deposit, partly lacustrine and
partly of brackish water origin, and the recognition of its purely
marine types has caused much controversy in Europe and Amer-
ica. The Trias was based at first entirely on the deposits of
the Germanic inland sea, and only recently has any uniformity
been attained in its correlation and nomenclature.

The greater divisions are now all named, and there is no room
for new systems, all geologic time, and possibly something over,
being taken up by those already defined. But because the first
naming of geologic divisions was based on unconformities,
which represented erosion intervals and consequent gaps in the
record, further exploration must necessarily bring to light some-
where in the world passage beds between the artificial systems.
What rules then shall be followed in the correlation of these
passage-beds, or new formations? If the new formations are
shown to be the homotaxial equivalents of parts of systems
already defined, there can be no question, for faunal or lithologic
differences cannot be taken in account in different provinces or
regions.

If the new formation, however, lies between two systems, one
of which is sharply defined and the other described only vaguely,
as including the beds below or above the one with definite
boundaries, then the passage-beds must naturally be assigned to
the system with a flexible margin.

If both the bordering systems should be definitely bounded,
or if neither should be, then one or the other must be stretched
to take in the passage-beds, and the assignment will be based
on paleontologic relationship to one or the other. But if this
relationship should be no closer to one than the other, then prior-
ity of assignment would have to decide on the nomenclature of
the doubtful beds.

The geologists of India have for many years recognized that

514 JAMES PERRIN SMITH

in the Himalayas and the Salt Range there was a continuity of
life and sedimentation from the Paleozoic into the Mesozoic,
and there the line of demarcation has been arbitrary. The
stages or zones used in drawing this paleontologic line in India
have become types, and are used in interregional correlation
wherever similar beds are known. A section is given below of
the uppermost Permian and the Lower Trias of India, for con-
venience of comparison with the American section.

SECTION OF THE PERMIAN AND LOWER TRIAS OF INDIA AFTER WAAGEN, DIENER,

AND NOETLING

Middle — Upper Ceratite limestone.
Flemingites beds.
Stachella beds.
Ceratite marls.
Lower Ceratite limestone.

< f Ceratite sandstone

| Lower +

\e | Otoceras beds ,
(

(

|

|

(

Trias

Upper Productus limestone, with Cyclolobus oldhamt, and
Upper Xenaspis carbonaria.

Per- Middle Productus limestone, with Xezxasfis carbonaria.

aa Laer oe Productus limestone, with Fusulina kattaensts.

acial beds.

Waagen drew the line between Paleozoic and Mesozoic at
the base of the Ceratite formation because of asupposed uncon-
formity, and because the Permian types of brachiopods such as
Productus, Chonetes, and the Orthidae have not been found above
this line. But the unconformity has since been shown not to
exist, and several of the Permian ammonite genera range up into
the Ceratite formation, such as Otoceras, Medhcottia and Xenaspis.

The upper Productus limestone is acknowledged to be younger
than any known European marine Permian, and the Otoceras
beds are universally considered to be older than any Trias yet
known from Europe; therefore the continuity of sedimentation
and life makes it exceedingly difficult to draw a line that will
satisfy everybody. And, indeed, if either the Trias or the Per-
mian had been named first in India, there would have been no
division of this series.

The Wichita Permian of Texas, while carrying forms such as
Medhcottia, Waagenoceras, and Popanoceras, harbingers of the
varied Triassic ammonites of later times, contains in its other
members so many Paleozoic types that no one would think of

PALEOZOIC AND MESOZOIC BORDER-LINE 515

classing itin the Mesozoic. There can be no controversy as to
its age, which is approximately equivalent to that of the lower
Productus limestone of India. It lies conformably on the Coal-
measures, and contains many species that have ranged up from
the latter.

But above the Wichita Permian lies the great series of the
Red Beds. These have been assigned sometimes to the Permian
and sometimes to the Trias, and they, doubtless, belong to both.
Marine fossils have been found in these Red Beds only in three
places. The Geological Survey of Texas ' found in the Double
Mountain formation at the falls of Salt Croton Creek, Kent
county, Texas, Medlicottia, Waagenoceras, and Pleurophorus, all
akin to forms from the Wichita beds. The writer has examined
‘these forms and recognizes them to be still typically Permian
in character, although they are very near the top of the forma-
tion. Above these beds lies the fresh water ‘‘ Dockum’”’ series
referred by Cope to the Trias.

C. N. Gould? has recently found marine and brackish-water
Permian fossils in the Red Beds of Oklahoma, in the Cimmaron
series; the forms still being of Paleozoic type, such as Conocar-
dium, Aviculopecten, Schizodus, Pleurophorus,and Bakevellia. The
stratigraphic position of these is equivalent to the Double
Mountain beds of Texas, and they belong unequivocally to the
Paleozoic.

In the Red Beds of Utah, at Ft. Douglas near Salt Lake City,
F, Frech3 has found Pleurophorus imbricatus Waagen, Allorisma
conf. elegans King, Emondia aspinwallensis Meek, Schizodus
schlothemt King, and Dalmanella sp. indet., a fauna that would
pass for Paleozoic anywhere, and belongs in the same horizon as
the Double Mountain beds of Texas and the Cimmaron series of
Oklahoma and Kansas. They are the homotaxial equivalents
of the Upper Permian.

tGeological Survey of Texas, Second Annual Report, 1890. Report on the Geol-
ogy of Northwestern Texas, by W. F. CUMMINS, p. 408.

2JouR. GEOL., Vol. IX, 1901. No. 4, p. 337.

3Lethaea Geognostica, Vol. II, Lieferung 3. 1900, p. 515.

516 JAMES PERRIN SMITH

From the close of the Coal-measures the land encroached on
the basin-sea by a progressive westward uplift, and the brackish
water and basin deposits take a successively higher place in the
geologic column towards the west. Thus somewhere in the eastern
part of the Great Basin region marine intercalations of Lower
Trias may be expected to be found in the Red Beds. Indeed,
this formation in southeastern Idaho lies above and conformably
upon the marine limestones with a Lower Triassic fauna. It is,
then, in the upper part of this Red Beds formation, in the fresh
and brackish-water deposits and its rare marine facies, that we
are to seek for the transition from Paleozoic to Mesozoic.

C. A. White’ has described from the Aspen Mountains of
Idaho a marine fauna older than any Trias known up to that
time in America, and younger than any known Permian. The
Meekoceras beds in which this fauna was found he assigned to
the Lower Trias. According to A. C. Peale? this formation lies
conformably upon the Carboniferous, and below the Red Beds.
The Weekoceras fauna was found near the base of the series, which
here has a thickness approximating three thousand feet. The
Carboniferous limestone below contained Productus multistriatus
Meek, which has been found by the writer in the uppermost
Paleozoic beds of California, and has been cited as a character-
istic fossil of the Permian of northern Europe.

Since that time Professor Alpheus Hyatt and the writer have
visited the Meekoceras beds, and their joint collections yielded ;
Meekoceras gracilitatis White, M. (Gyronites) aplanatum White,
M. (Koninckites) mushbachanum White, Aspidites, Flemingites,
Pseudosageceras, Ussuria, Ophiceras, Clypites, Danubites, Nannites,
and a number of new genera.

The writer3 has previously described the Ceratite limestone
of the Lower Trias of Inyo county, California. During the past

Twelfth Ann. Rep. U. S. Geol. and Geog. Surv. Terr., Part I (1880). Triassic
Fossils from Southeastern Idaho.

2Bull. U. S. Geol. and Geog. Surv. Terr., Vol. V., No. 1 (1879). Jura-Trias
Section of Southeastern Idaho and Western Wyoming, p. 119.

3Jour. GEOL., Vol. VI, No. 8, 1898. Geographic Relations of the Trias of Cali-
fornia.

PALEOZOIC AND MESOZOIC BORDER-LINE 517

winter he has visited this locality and added a great many species
and genera to the collection from that formation. Those now
known from there are: Meekoceras gracilitatis White, MM.
(Gyronites) aplanatum White, M. (Koninckites)- mushbachanum
White, WZ. conf. vadiosum Waagen, M. conf. falcatum Waagen,
M. aff. boreale Diener, Aspidites, Prionolobus, Danubites, Proptychites,
Xenaspis, Lecanites, Nannites, Ussuria, Pseudosageceras, and Clypites,
besides a number of new genera. Several of the species, both
new and described, are identical with forms from the Meekoceras
beds of southeastern Idaho, with no greater difference in the
fauna than might be expected in localities separated by six
hundred miles.

Below the Meekoceras beds of Inyo county lie several hundred
feet of barren shales, and below these is siliceous limestone con-
taining Pusulina. Above the Meekoceras beds are eight hundred
feet of calcareous shales with impressions of ammonites, and
thena few feet of black limestone with Acrochordiceras, Hungarites,
Tivolites, Ceratites, and Xenodiscus, and Parapopanoceras, probably
belonging to the base of the Middle Trias. The entire series
appears to be conformable, from the Upper Carboniferous to the
Middle Trias.

The fauna of the Meekoceras beds of Idaho and California is
most intimately related to that of the Ceratite formation of
India, and of the homotaxial Proptychites beds of Ussuri in eastern
Siberia. Meekoceras, Gyronites, Koninckites, Danubites, Proptychites,
and Ophiceras are known both in the Ceratite formation of India,
and in the Ussuri beds of Siberia; Pseudosageceras and Ussuria
are known elsewhere only in the Ussuri formation; and the
species of all these regions are nearly related, in part apparently
identical.

It is, therefore, plain that the correlation of the Ceratite
formation of India has a direct bearing on the determination of
the age of the Meekoceras beds of America. Albert Oppel, who
in 1865 described fossils from the Ceratite formation of India,
referred them at first to the Jura, and then afterwards to the
Trias, as typical Jurassic beds were found considerably above

518 JAMES PERRIN SMITH

them. But before this, in 1863, de Koninck, who also described a
number of ammonites from the Ceratite formation, referred them
very doubtfully to the Paleozoic, because they were supposed to
have come from the Productus limestone. C. L. Griesbach? who
first described the fauna of the Ovoceras beds of India, referred
them to the Lower Trias, in which he has been followed by
Wilhelm Waagen? and Carl Diener.3 These writers consider the
lower part of the Ceratite formation to be younger than any
known Permian, and older than any fauna described from the
European Trias, but they regard the affinities of the species as
closer to known Mesozoic types, and the beds were considered
as homotaxial with the lower part of the Werfen sandstone, the
base of the Trias in the Mediterranean region. Now since this
part of the Mediterranean section is barren of fossils, this corre-
lation is based entirely on the stratigraphic position of the
Ceratite formation.

C. Diener‘, in describing the fauna of the Proptychites beds
of Ussuri Bay in eastern Siberia, correlated them with the Ofoceras
beds of India, because of the occurrence of several species com-
mon to the two regions. It is, therefore, evident that while the
correlation of the Ceratite formation of India with the Proptychites
beds of Siberia and the Weekoceras beds of western America may
be accepted as correct, the final determination of the age of these
beds depends entirely on a comparison with the type of the
Lower Trias, the Buntsandstone of Germany, and the Werfen
beds of the Alps.

In spite of the weakness of this argument, the Ceratite forma-
tion of India has until recently gone unquestioned as the type
of the strictly marine equivalent of the Lower Trias, and all

™Records Geol. Surv. of India, Vol. XIII, 1880. Paleontological Notes on
the Lower Trias of the Himalayas, p. 94.

?Mem. Geol. Surv. India, Pal. Indica, Ser. XIII. Salt Range Fossils, Vol. II.
Fossils from the Ceratite Formation.

3lbid., Ser. XV. Himalayan Fossils, Vol. II, Part I. Cephalopoda of the
Lower Trias.

4Mem. Com. Geol. St. Petersbourg, Vol. XIV, No. 3. Triadische Cephalopoden-
faunen der Ostsiberischen Kiistenprovinz.

PALEOZOIC AND MESOZOIC BORDER-LINE 519

correlations of the marine beds have been made by comparison
with it. During the last year F. Noetling,’ paleontologist of the
Geological Survey of India, has startled geologists in their fancied
security, by the statement, based on his later investigations, that
the Ozoceras beds of India belong to the upper Permian, and are
older than the Werfen beds of the Alps. As they lie conform-
ably upon the upper Productus limestone, they represent, accord-
ing to Noetling, strata of which the equivalents are lacking in
the European section. Accordingly he proposes to call the
entire Ceratite formation upper Permian, and as a name for the
stage he proposes the term Bactrian. If this should prove to
be correct it would throw the Ceratite formation of India, the
Proptychites beds of Siberia, and the Meekoceras beds of Idaho
and California into the upper Permian.

A. von Kraft,? also of the Geological Survey of India, agrees
with Noetling, but only as to the Permian age of the Ofoceras
beds; the upper part of the series where Meekoceras and flem-
ingites are so abundant he regards as certainly of lower Triassic
age. He says that Medlicottia dalatlamae Diener of the Otoceras
beds is identical with WZ. wynnez Waagen of the upper Productus
limestone, and that Cyclolobus oldhami Waagen and Xenaspis
carbonaria \Naagen, which in the Salt Range occur in the mid-
dle Productus limestone, in the Himalayas occur in the Kuling
Productus shales only twenty or thirty feet below the Ofoceras
beds. This would make the bottom of the Ofoceras beds (the
lower Ceratite limestone) the equivalent of the upper Productus
limestone of the Permian. This argument, however, might work
both ways, for it would just as well prove that the upper Pro-
ductus limestone belonged to the base of the Trias, since in any
case the division line must be arbitrary. Nor would the finding
of Productus itself in: the Ofoceras: beds prove ‘them to -be

*Geol. Surv. of India, General Rept. for 1899 (1900). Notes on the Relation-
ship between the Productus Limestone and the Ceratite Formation of the Salt Range,
and Neues Jahrbuch fiir Min. Geol. und Pal., 1900. Bd. I, p. 139, Ueber die Auffin-
dung von Otoceras sp. in der Salt Range. ;

?Centralblatt fiir Min. Geol. und Pal. Bd. II, 1901, p. 275, Ueber das Permische
Alter der Otoceras-Stufe des Himalyas. j

520 JAMES PERRIN SMITH

Paleozoic, for if Otoceras, Medlicottia, Xenodiscus, and Xenaspis can
range into the Mesozoic, there is no known reason why Productus
should not have done so. Indeed, Xenaspis and Xenodiscus
range up into the Middle Trias, associated with faunas charac-
teristic of the Muschelkalk.

C. Diener* meets Noetling’s argument by stating that if
the Ceratite formation of India is not of Lower Triassic age,
then we have nowhere in the world a strictly marine equivalent
of the Werfen beds, and that we have the anomaly of Middle
Triassic beds lying conformably on Permian. He shows that
the marine Permian of the Alps, which lies below the Werfen
beds, has not the fauna of the Osoceras beds, but one analogous
to that of the Productus limestone. Diener admits that the
Otoceras horizon is older than the fossilferous portion of the
Werfen beds, but thinks that the rule of priority of reference
must be followed, where there are no paleontologic grounds
against it and many for it. He also cites a recent paper by A.
Bittner? to show that there is found in the Proptychites beds of
Ussuri a typical pelecypod and brachiopod fauna of Werfen
character ; it must be admitted, however, that this fauna was
not found with the ammonites, and might belong considerably
above them.

The most direct comparison with the European Trias has
been made by Lukas Waagen,3 who shows that in the Ceratite
marls and the Ceratite sandstone of India he has identified a
number of pelecypods identical with forms characteristic of
middle and upper Werfen beds of the Alps. This leaves the
question about as A. von Kraft stated it. The Ceratite forma-
tion, from the Ceratite marls up, certainly belongs to the Lower
Trias, since the Werfen beds are the type, while the Ofoceras
beds of the Himalayas, or their equivalent, the lower Ceratite
limestone, may belong to the upper Permian. But the fauna of

*Centralblatt fiir Min. Geol. und Pal. Bd. I, 1900, p. 1, “‘ Veber die Grenze des
Perm- und Triassystems, etc.”

2 Mém. Com. Geol. St. Petersbourg, Vol. VII, 1899, ‘‘ Verstein. aus den Trias-Abla-
gerungen des Siid-Ussuri Gebietes in der Ostsibirischen Kiistenprovinz.

3 Centralblatt fiir Min. Geol. und Pal. Bd. I, 1900, p. 285.

PALEOZOIC AND MESOZOIC BORDER-LINE 521

the lower Ceratite limestone and that of the Ceratite marls are
so similar that a separation of the two is out of the question,
and even the genus O¢oceras is found in the latter. Now, since
these doubtful beds are younger than any accepted Permian, and
older than any authentic Trias, they might with equal propriety
be assigned to either, and we shall have to extend one or the
other system to include them. The question will have to be
decided either by paleontologic relationship, or by priority of
reference. The Ofoceras beds contain Meckoceras, Proptychites,
Ophiceras, and several other ammonite genera that have never
been found in the Paleozoic; they contain also Medlicottia and
Otoceras that are more characteristic of Upper Permian; but they
lack the Productidae and Orthidae that characterize the Permian
formation. Thus the paleontologic evidence is about equal in
favor of a reference to either Paleozoic or Mesozoic. But the
geologists that described the fauna of the doubtful beds have
almost unanimously referred them to the Lower Trias, and this
must be the final verdict.

The fauna of the Meekoceras beds of Idaho and California is
most intimately related to that of the Ceratite marls and the
lower part of the Ceratite sandstone of India, with most of the
genera in common, and several species that seem to be identical.
And although the writer has searched carefully for Ofoceras in
both places, no trace either of this genus or of Wedlicottia was
found. It seems likely, then, that even if the bottom of the
Ceratite formation should be cut off from the Trias and assigned
to the Permian, this change would not affect the nomenclature
of our American formations that are now considered as the bot-
tom of the Mesozoic series, and the Weekoceras beds will stand
as the type of the marine Lower Trias, where White and Hyatt
placed them in 1879. The real transition from Paleozoic to
Mesozoic must be sought in the conformable series below the
Meekoceras beds, and above the Fusulina limestone.

JAMES PERRIN SMITH.

STANFORD UNIVERSITY,
California.

STUDIES FOR STUDENTS

THE, CONSMIUPNESOR METEORITES. al

Glass —This is an abundant constituent of the stone meteor-
ites, few if any being entirely without it. It is variously dis-
tributed, occurring now as vein matter, now scattered through
the substance of chondri, now enclosed in the substance of a
single mineral, and now enclosing various minerals.

In the Parnallee, Mezo-Madaras, Chassigny, Farmington and
a few other meteorites it has been described as forming a net-
work in which the other minerals are imbedded. Its ocurrence
in this manner is rare, however, it playing usually a merely acces-
sory part. It chiefly abounds as inclusions and intergrowths in
chrysolite, taking in this association a great variety of forms.
Other minerals too, frequently have inclusions of glass. Itmay
occur in fragments of considerable size or the particles may be
of a dustlike minuteness. 5

Its abundance in chondri has already been mentioned. By all
these occurrences a rapid crystallization or cooling of the
meteorite substance is strongly indicated. Like the glass of
terrestrial lavas it seems to be the result of cooling so rapid as
to prevent differentiation and orderly crystallization of the
magma. The especial abundance of glass in meteoritic chrys-
olite, the least fusible and therefore the earliest cool ing ingre-
dient further favors this conclusion.

The prevailing color of the glass of meteorites is brown.
Much is however colorless and some occurs so dark as to be
opaque. Grayish and greenish tones occur but are rare.

Chromite.—Nearly all stone meteorites give on analysis a
small percentage of chromium which is usually considered as
being present “ins the form of -chromite, Bhe@ Gr, Oj whe
mineral is not so abundant in the iron and iron-stone meteorites

522

THE CONSTITUENTS OF: METEORITES 523

but has been detected in several and in the Coahuila irons occurs
in nodules of considerable size (17™™ X 12™™).

It is identical with terrestrial chromite in composition and
properties. Not being acted upon by acids, it may be readily
distinguished from daubréelite. It is generally non-magnetic,
but sometimes feebly magnetic. Where crystals occur they are
commonly octahedrons, sometimes modified by other forms.

Amorphous carbon.—Meteorites of the group known as car-
bonaceous meteorites, as well as some others, are permeated by
a dull-black pulverulent coloring matter which is usually left as
a residue on treatment of the meteorite with acid. This residue
sometimes amounts to from 2—4.5 per cent. of the mass.

A residue similar in character though smaller in amount is
likewise found after dissolving many of the iron meteorites.
These residues on being heated in air glow, usually become
lighter in color and give off carbon dioxide. They must there-
fore be considered practically pure carbon.

Berzelius and Wohler believed this carbon to have originated
so far as the carbonaceous meteorites are concerned, from the
decomposition of the hydrocarbons of the latter. In this respect
they regarded it analogous to terrestrial humus, though of very
different origin. Smith considered it similar in origin to the
graphite of iron meteorites and Weinschenk believes it similar to
one of the forms of carbon produced in the making of cast iron.
No indications that it had an organic origin have ever been dis-
covered. x

Diamond.—TVhe existence of diamonds has been definitely
proven in only two meteorites, those of Caffon Diablo and Nowo-
Urei. Diamonds have, however, also been reported from the
irons of Magura and Smithville and the stone of Carcote. The
diamonds of. the Cafon Diablo meteorites have been most
studied. Here they are found as minute particles or dust left
as a residue after dissolving the meteorite in acid. The particles
rarely exceed 1%2™" in diameter. They are usually brown -to
black in color but sometimes are colorless and transparent.
They accompany graphite, amorphous carbon and often troilite

524 STUDIES FOR STUDENTS

and schreibersite. They have a tendency to gather in little
clefts or hollows and are not regularly distributed. Their char-
acter as diamond is proven chiefly by their hardness, but analy-
ses and a study of their behavior in polarized light give
confirmatory results. Huntington found some also which showed
crystal forms of diamonds. The occurrence of diamonds in
meteorites suggests interesting analogies with their terrestrial
occurrence. Knop and Daubrée call attention to the fact that
the peridotic rocks in which terrestrial diamonds occur are the
rocks most nearly allied in composition to meteorites. In the
iron meteorites, as Moissan has proven satisfactorily by experi-
ment, diamond is to be considered a form in which, under certain
conditions of heat and pressure, carbon separates. Moissan
obtained diamonds by heating to a high temperature iron satu-
rated with carbon and allowing it to cool under pressure. The
carbon was then found to exist in three forms, graphite, foliated
carbon, and a diamond powder which latter corresponded to
that obtained from the Cafion Diablo meteorites.

A form of carbon resembling graphite but differing in hav-
ing a hardness of 2.5 and being isometric in crystallization, has
been noted in the Magura, Cosby’s Creek, Youndegin, Toluca
and a few other iron meteorites. It was first discovered by
Fletcher, who considered it a distinct species and gave it the
name cliftonite. Other authorities, however, regard cliftonite
as a pseudomorph after diamond, since its crystals closely
resemble those of diamond in form.

Daubréelite—This mineral is an iron-chromium sulphide pecu-
liar to meteorites. Its composition is Fe S, Cr, S,. It is found
in nearly all the cubic iron meteorites and has also been identi-
fied in the irons of Toluca, Nelson county, Cranbourne, Canon
Diablo and others. It has never been found in stone meteorites.
It usually accompanies troilite, either bordering nodules or cross-
ing them in veins. Sometimes, however, it occurs as thin plates
or grains. It is black in color, has a black streak, is of metallic
luster, brittle and not magnetic. It is infusible before the blow-
pipe and becomes magnetic in the reducing flame. It is not

THE CONSTITUENTS OF METEORITES 525

attacked by hot or cold hydrochloric acid, but is completely
dissolved by nitric acid without the separation of free sulphur.
This solubility distinguishes it from chromite. Its system of
crystallization is not known though it exhibits rectangular and
triangular partings which indicate one of the systems of high
symmetry. . Meunier obtained the mineral artificially by treat-
ing an alloy of iron and chromium at a red heat with hydrogen
sulphide.

Tridymite.—This mineral has been positively identified in
only one meteorite (Steinbach), but it probably also occurs in
the Vaca Muerta and Crab Orchard Mountains meteorites. These
are all ironstone meteorites. In the Steinbach meteorite it
forms from 8.5 to 33 per cent. of the non-metallic constituents
and occurs intergrown with bronzite.

Maskelyne, who first described the mineral, considered it on
account of its optically biaxial character a new orthorhombic
form of silica and gave it the name of asmanite. Since tridy-
mite is now known, however, to exhibit biaxial characters and
the minerals agree in most other respects, they are generally
considered identical.

Tridymite occurs in meteorites in the form of rounded grains
or plates, some of which reach a length of 3™™. They are
colorless to white to rusty brown in color.

Well developed crystals are rare but from facets on rounded
grains a total of twelve forms has been determined.

Analyses show a composition of practically pure silica, with
iron oxide and magnesia present as impurities.

Lawrencite.—This is a solid ferrous chloride which has been
described from the iron meteorites of Tazewell, Smith Mountain,
and Laurens county. Formula Fe Cl,. Color green to brown.
While described in the solid form from only the few meteorites
mentioned, the presence of lawrencite in many other iron meteor-
ites is generally believed to be indicated by the greenish drops
which exude on their surfaces. These drops are ferric chloride
or mixtures of ferric and nickel chloride, while occasionally
pure nickel chloride occurs.

526 STUDIES FOR STUDENTS

The meteorites from which ferric chloride exudes disinte-
grate with especial rapidity. Such meteorites are often known
as ‘‘ sweating” meteorites. The “sweating” is rarely noted in
ironstone or stone meteorites, but the small percentage of chlo-
rine found in the analysis of many of these meteorites is usually
referred to lawrencite. Some authorities hold that the substance
is not an original constituent of any meteorite, but is wholly of
terrestrial origin. This is not the general opinion however.

Ferrous chloride has been noted among the sublimation
products at Vesuvius and is reported as having been found in
the terrestrial nickel-iron of Ovifak.

Magnetite.— Several stone or ironstone meteorites have been
found to contain black, magnetic grains which dissolve in hydro-
chloric acid without effervescence to form a yellow solution.

In the meteorites of Shergotty and Dona Inez these are suf-
ficiently abundant to form an essential constituent. They con-
stitute 4.57 per cent. of the Shergotty meteorite. Similar grains
occur as inclusions in maskelynite, pyroxene, and chrysolite in
the above and other meteorites. They are regarded as mag-
netite.

No well marked crystals of meteoritic magnetite have as yet
been described.

Magnetite has been reported as a constituent of several iron
meteorites though only one analysis has been made, that of
Meunier of magnetite from the crust of one of the Toluca irons.
The composition of this agreed with that of terrestrial magnet-
ite. Several other iron meteorites show magnetite in their
crust... Here, however, the magnetite may have originated from
the oxidation of the iron of the mass since its'arrival upon the
earth: Pe
Magnetic spherules have been dredged up from ocean depths,
which Murray and Renard regard as particles of meteoric iron
oxidized to magnetite since their arrival upon the earth. __

Oldhamite.—This is a simple calcium sulphide with the for-
mula Ca S,. Grains of it were found by Maskelyne embedded
in the enstatite or augite of the Bustee meteorite. It,is light

THE CONSTITUENTS OF METEORITES S27

brown in color and transparent when pure. MHardness 4, specific
gravity 2.58. It is isotropic and has equal cleavage in three
directions, hence is doubtless isometric. In the Bustee meteor-
ite it occurs in rounded grains, coated with gypsum through
alteration.

Certain yellow grains found in the Bishopville meteorite were
also considered by Maskelyne to be this mineral. Aside from
these two occurrences it has not been positively identified in
any other meteorite.

Calcium sulphide resembling oldhamite was obtained by
Maskelyne by heating caustic lime in a glass tube, first with
hydrogen, then with hydrogen sulphide. Vogt has noted a
similar compound formed in furnace slags.

It has not been found as a terrestrial mineral.

On dissolving the oldhamite of the Bustee meteorite, Maske-
lyne found a residue constituting about 0.3 per cent. of the
weight of the former, consisting of yellow octahedrons of micro-
scopic size. These were found to be unaffected by acids or
oxygen, while qualitative tests indicated sulphur, calcium, and
titanium or zirconium. Maskelyne regarded the mineral there-
fore an oxysulphide of calcium and titanium and gave it the
name osbornite. No other occurrence of the mineral is known.

HHydrocarbons.—The hydrocarbons found in meteorites may be
divided, following Cohen,’ into three classes: (a) compounds of
carbon:and hydrogen ; (0) compounds of carbon, hydrogen and
sulphur; and (c) compounds of carbon, hydrogen and oxygen.
They especially characterize meteorites of the class known as
carbonaceous, which includes seven or eight distinct falls of mete-
orites of black color, low specific gravity and containing a sen-
sible amount of carbon. They have been obtained from some
other meteorites however, such as those of Collescipoli and Goal-
para. The hydrocarbons of the first class are obtained by treating
these meteorites with alcohol or ether. They are resinous or wax-
like bodies which completely volatilize on the application of heat.
When heated in a closed tube the resinous substances first fuse,

*Meteoritenkunde. Heft. I, p. 159.

528 STUDIES FOR STUDENTS

then are decomposed to form amorphous carbon and an oil hav-
a bituminous or fatty odor. Such substances were considered
by Wohler similar to the mineral wax ozocerite and by Shepard
they were designated meteoritic petroleum. Friedheim states
that a substance extracted by him from the meteorite of Nagaya
by means of ether had a bituminous odor, volatilized at 200°
and resembled a product of distillation of brown coal. A similar
substance extracted by Roscoe from the meteorite of Alais was
found to have a composition corresponding nearly to the formula
(Caah

Hydrocarbons of the second class were obtained by Smith
by treating the graphite of iron meteorites and some carbona-
ceous meteorites with ether. These compounds were fusible
and volatile. He regarded them as having the general compo-
sition C,H,,S,;. He obtained similar products by treating
cast iron with ether or petroleum as did also Berthelot by the
action of ether on sulphur or iron sulphide in the presence of
oxygen.

Hydrocarbons of the third class have been obtained from
the meteorites of Orgueil and Hessle. The Orgueil extract
resembles peat, humus or lignite in its composition and proper-
ties. That from Hessle has approximately the composition
nie or Or.s

The above mentioned facts make it clear that a number of
meteorites contain products of an easily destructible, volatile,
and combustible character which resemble terrestrial bitumens,
petroleum or oxygenated hydrocarbons. The quantity of these
products is relatively small, being less than 1 per cent. in the
majority of meteorites in which they occur. Yet that they
occur at all is significant. While some have urged that these
products might have arisen from the union of their elements in
the terrestrial atmosphere there seems little reason for doubting
their pre-terrestrial origin. There is no evidence that life had
anything to do with their origin. We must conclude then that
they were formed in an inorganic way by a union of their ele-
ments. The conclusion at once suggests the possibility that

LHE CONSTITOENTS OF METEORITES 529

terrestrial hydrocarbons need not always be referred to an
organic origin, but may have been formed in a purely inorganic
way.

The occurrence of hydrocarbons in meteorites further shows
that such meteorites could not have been subjected to any
high degree of heat subsequent at least to the formation of these
compounds, and that the heating of meteorites during their fall
to the earth has in many cases been only superficial.

The trails of light, sometimes enduring several minutes,
observed following in the wake of some meteors may perhaps
indicate the presence of carbonaceous matter in those bodies.
The stone shower which took place at Hessle was accompanied
by luminous effects and with the stones fell a brownish-black
powder which contained 71 per cent. carbonaceous matter.
Other carbonaceous meteorites have fallen, however, without
exhibiting any marked luminous phenomena.

Other compounds.— Besides the above well-determined com-
pounds a number of others have been reported at different times
which are (1) present in insignificant amount or (2) their
occurrence has not been confirmed, or (3) they may be of ter-
restrial origin. Among these a few may be mentioned: Quartz.
This mineral, as is well known, is remarkable for its absence
from meteorites. Yet it doubtless does occur in minute grains
in a number of iron meteorites, since on dissolving them a
residue is left, the grains of which possess the properties of
quartz. Its occurrence in any stone or ironstone meteorite has
never yet been established. Pyrite. This mineral has been
reported a number of times, but sufficient proof to establish its
identity has not been given. Von Siemaschko reported from
the meteorite of Ochansk a brass-yellow pentagonal dodecahe-
dron of which, however, he gave no measurements. Daubrée
found in the meteorite of Senhadja, bronze-yellow grains insolu-
ble in hydrochloric acid, soluble in aquaregia and altering easily
to iron sulphate. While these and other observations suggest
pyrite they are not conclusive. Salts soluble in water. Several
of the carbonaceous meteorites as well as one or two others give

530 SDLUDIES FOR STUDENTS

on evaporation of the water extract a residue of soluble salts
reaching in quantity in one case as high as Io per cent. of the
mass. These salts include nickel, calcium, magnesium, potas-
sium, sodium and ammonium sulphates and chlorides.

Since the meteorites in which they occur are very porous in
character and show other signs of alteration these compounds
are usually considered to be formed by terrestrial modification
of the meteorite and not to exist as original constituents. Dau-
brée, however, gives good reasons for regarding the sodium
chloride which he found in the Lancé meteorite an original con-
stituent. These reasons are that the meteorite had lain only
three days in a clayey bed before it was picked up and no salt
is known to have come near it. Bveunnerite. This mineral was
found in the meteorite of Orgueil occurring in the form of little
transparent crystals. The identity of the mineral was estab-
lished both by qualitative tests and by goniometric measure-
ments. It has been suggested that it was of secondary origin.
As it was found well within the interior of some masses, this,
however, hardly seems likely. This is the only carbonate known
from meteorites.

A number of other minerals have been reported from meteor-
ites without sufficient grounds, according to the writer’s view, to
support the conclusion. Cohen considers them doubtful while
Meunier accepts most of them. These are: Apatite, tolite,
wollastonite, titanite, garnet, vesuvianite, mica, aragonite, leucite, cas-
siterite, hornblende, anthophyllite and orthoclase.

Mineral aggregates —The different aggregates which the com-
pounds above described form in different meteorites are too
various to be recorded here in detail. For an account of
these, reference should be made to the elaborate classifications
of Meunier,’ Brezina* or Wilfing.3

A few general observations may be made here, however,
following the lines of the classification given by Wiulfing. The

* Revision des Pierres Météoriques. Paris, 1807.

2 Annalen des k. k. Naturhistorischen Hofmuseums. Bd. X, Heft 3u. 4. Vienna,
1896.

3Die Meteoriten in Sammlungen. ‘Tiibingen, 1897, pp. 447-460.

LHE CONSTITUENTS OF METEORITES 531

iron-meteorites, as already indicated, are made up chiefly of
nickel-iron, with schreibersite, troilite, daubréelite and a few
other minerals occurring as accessories.

Of the ironstone meteorites the largest quantity are of the
so-called pallasites, formed chiefly of chrysolite and nickel
iron. Nine falls of this group are known, having a weight of 1742
kilograms. In the group known as siderophyrs, represented by
one fall (82 kilos), of meteoritic matter, bronzite and tridymite
are associated with the nickel-iron. In the group of mesosid-
erites (grahamites) represented by ten falls (483 kilos), of
meteoritic matter, the nickel-iron is accompanied by chrysolite,
bronzite, plagioclase, and augite. In the group lodranite, com-
posed of one fall with a weight of 1 kilo, chrysolite and bronzite
are associated with nickel-iron.

Passing to the stone meteorites the following groups and
weights may be noted:

A. Stones rich in calcium and magnesium and containing little or no
nickel-iron,

1. Angrite. Chiefly augite. One fall, weight 0.4**.

2. Eukrite. Augite and anorthite. Four falls, weight gi‘.

3. Shergottite. Augite and maskelynite. One fall, weight 5*.

4. Howardite. Augite, anorthite, bronzite, and chrysolite. Ten falls,
weight 5*8,

B. Stones rich in magnesia and containing little or no nickel-iron.
Bustite. Diopside and bronzite. Two falls, weight 1.7*%.
Chassignite. Chiefly chrysolite. One fall, weight o.9*¢.

Chladnite. Chiefly orthorhombic pyroxene. Four falls, weight 9.
. Amphoterite. Chiefly chrysolite and bronzite. Three falls, weight 4o*.

C. Stones rich in magnesia and consisting essentially of chrysolite,
bronzite, nickel-iron, and iron sulphide. Here belong the great majority of
stone meteorites.

PWN

A comparison of the constituents as above described with
those of the crust of the earth brings to view some interesting
similarities and contrasts. Under semilarities may be noted the
fact that the elements of meteorites are the same as those of
the earth and that they unite according to the same chemical and
physical laws. No new element has been discovered in meteor-
ites and the chemical compounds of meteorites similar to those
of the earth agree even to the details of their crystal form.

532 STUDIES FOR STUDENTS

Under contrasts it may be noted that two agents which have
affected largely the composition of the crust of the earth have
been lacking either wholly or in part in the formation of meteor-
ites. These agents are water and oxygen. The lack of water
is proved by the fresh and unaltered character of the minerals
found in meteorites and the absence of all hydrous minerals.
Thus the chrysolite of meteorites is never found serpentinized nor
are the pyroxenes changed to chlorite nor the feldspar to kaolin.

Further, zeolites, micas, epidote, tourmaline and all other
minerals in the formation of which water and water vapor play
a part are entirely lacking from meteorites.

Similarity, oxygen, at least in excess, is lacking from the
constituents of meteorites. Such substances as_ nickel-iron,
schreibersite, and lawrencite, which make up so large a part of
the composition of meteorites would rapidly have been oxydized
had they been exposed to the action of oxygen as it occurs
upon the earth. The silicates of meteorites are however oxydized
compounds which show that oxygen is present to some degree
in space.

Again, as noted by Cohen," the important rock-forming min-
erals of the crust of the earth are either lacking or play an
insignificant part in the formation of meteorites. Such are
quartz, orthoclase, the acid plagioclases, the micas, the amphi-
boles, leucite, and nepheline. Vice versa, the chief mineral con-
stitutents of meteorites occur in but insignificant amount upon
the earth. Such are nickel-iron, the orthorhombic pyroxenes
and chrysolite, while such compounds as schreibersite, cohenite,
lawrencite, oldhamite, daubréelite and troilite rarely or never
occur terrestrially. Looked at quantitatively then it may be said
that terrestrial rocks abound in free silica, lime, alumina, and
alkalies, while meteorites abound in iron, nickel and magnesia.
Whether these quantitative differences would be maintained if
the constitution of the earth as a whole could be compared with
that of meteorites, is, as hinted at the beginning, doubtful.

OLIVER C. FARRINGTON.

“Op: cits, p. 323:

Among the innumerable phases of the
great calamity which has fallen upon our
country and the world in the tragic death
of PRESIDENT McKINLEy, the loss of a
generous friend of science is by no means
the least.” The prosperity ‘of ‘the .scientific
bureaus under his administration has been
as marked in improved organization and in
method as in extension and in generous
patronage, and the establishment of a new
bureau in the interest of scientific and com-
mercial precision is a laudable feature of

special moment.

LZ DITORIAL

In December, 1898, Mr. G. K. Gilbert presented to the Geo-
logical Society of America a paper upon ripple-marks and cross-
bedding in which he undertakes to explain the large ripples of
the Medina formation.t. Mr. Gilbert became satisfied that these
ripples differ ‘‘in no respect except size from the familiar ripple-
mark of the bathing beach and the museum slab.”’ In order to
account for the size of some of the large ripples upon this
theory he has inferred that the waves producing them were sixty
feet high and made in ‘‘a large ocean.”

In the July number of the American Geologist, Professor H.
L. Fairchild objects to the deep ocean theory of the origin of
these ripples, and brings evidence to show that they are beach
structures.”

Without going into the details of either of the articles men-
tioned the present writer wishes to call attention in this con-
nection to a paper upon the origin of beach cusps published in
this JOURNAL (September—October, 1900, Vol. VIII, pp. 481-
484), and to suggest that the explanation of the giant ripples
spoken of by Gilbert and Fairchild is to be found in the seaward
extension of beach cusps. The beach cusps are from sixty to
eighty feet apart, from a few inches to three feet in vertical
height and extend oceanward in approximately parallel lines.
They are formed by the interference of two sets of waves of
translation, and are therefore to be looked for not only on the
beach where they appear at the water’s edge, but as far out as
the waves drag upon the sea bottom, and always pointing away
from the shore. This theory appears to account readily for all

‘Ripple-marks and Cross-bedding, by GROVE KARL GILBERT. Bull. Geol. Soc.
Amer. Vol. X, pp. 135-140.

2 Beach Structures in Medina Sandstone, by H. L. FAIRCHILD. Amer. Geologist,
Vol. XXVIII, pp. 9-14.

535

536 EDITORIAL

the phenomena observed in connection with the ripples in the
Medina without doing violence to the theory of the shallow
water origin of those beds. J. C. BRANNER

THE experiment of holding the summer meeting of the
American Association for the Advancement of Science as far
west as Denver may be regarded as a success. The attendance
compared favorably with what has previously been realized at
several meetings in the interior, though for obvious reasons it
was less than the attendance at meetings held in the more
populous and accessible centers of the East. The papers and
discussions, so far as one could judge from listening to those of
a single section and from current opinion, also compared favor-
ably with those of average meetings. There was less diversion
from the specific purposes of the association by formal social
functions which were few, and there was correspondingly greater
real social intercourse between fellow scientists, because the
intersessional intervals were more largely left free for this,
a most laudable feature. The provisions for scientific excur-
sions, at least in geologic lines, were notably more ample than
usual and were arranged for the afternoons of the regular session,
the morning sessions being extended to make this possible. The
facilities for general and varied excursions at the close of the
formal sessions were exceptionally generous. Only one feature
of the general appointments and of the environment needs to be
singled out for adverse comment, and that was the dreary silli-
ness of the Denver press which, apparently recognizing its limita-
tions in reporting appreciatively and intelligently the real
scientific news, tried to make up for its inabilities by stale witti-
cisms and coarse cartoons, interspersed with extravagant personal
laudations of ‘the-greatest-scientist-on-earth” type. A few
subjects relating to the economic interests of the region and to
popular themes were, however, well reported.

The general addresses were excellent; that of retiring-
President Woodward was an incisive and discriminating discus-
sion of the progress of science, graced with an artistic marshaling

EDITORIAL 537

of lights and shadows as wholesome as it was skilful; that of
Vice-President Van Hise on the philosophy of ore deposition
was clear, strong and effective, and especially laudable, as a
popular address in a mining region, for its unhesitating advocacy
of the unpalatable as well as the acceptable phases of his
doctrine.

Previous to the meeting a ten-day excursion of geologists
was planned by Professors Van Hise and Emmons and carried
out ina most admirable manner. The selection of routes and
places from among the phenomenal possibilities of Colorado
certainly made no small demands upon the knowledge and dis-
cretion of those in charge, but no whisper of a possible improve-
ment was heard. The climax of interest was reached in the
San Juan Mountains, where the exemplification of many and
varied phases of geological phenomena from the Archean to the
Pleistocene is marvelously impressive. The aid rendered by
prominent citizens at various points visited and the generous
hospitality extended to the party were beyond all praise. It
would be a delight to acknowledge our obligations in special and
individual terms, if, beginning with the exceptional courtesies of
_ Walsh, Lay, and Freeland, it were possible to find an end of the
list. About two dozen geologists participated.

The four geological sessions were crowded with papers well
distributed over the various departments of geology and
embodying much of exceptional interest and value. The papers
read before the Geological Society of America, presented on
the first morning, were as follows: ‘Account of the Geological
Excursion,’ °C. R, Van. Hise; ‘Junction of the Lake Superior
Sandstone and the Keweenawan Traps in Wisconsin,” U. S.
Grant; ‘‘ Hydrographic History in South Dakota,” J. E. Todd ;
“The Still Rivers of Western Connecticut,’ W. H. Hobbs;
‘Geology of the Northeast Coast of Brazil,’ John C. Branner ;
“Classification of the Geological Formations of Tennessee,”
James M. Safford; “Horizons of Phosphate Rock in Ten-
nessee,”’ James M. Safford.

The following papers were presented before Section E:

538 EDITORIAL

‘The Effect of Diurnal Heat on Glacial Activity,” J. FE. Todd ;
‘On Extra Terrestrial Stresses,” E. Haworth; ‘‘On Stoping as
a Factor in the Formation of Terraces,’ T.C. Chamberlin; ‘‘On
Campodus, Helicoprion, Acanthus and other Paleozoic Sharks,”
Charles R. Eastman; “ The Oscillations of the Coast Ranges of
California,” A. C. Lawson; ‘“‘Some Features of the Geology of
Golden, Colorado,” H. B. Patton; ‘“‘ The Geological Occurrence
of Oil in Colorado,” A. Lakes; ‘‘ Report on Some Studies Rela-
tive to Primal Questions in Geology,” T. C. Chamberlin; ‘A
Plea for Greater Simplicity in the Language of Science,” T. A.

Rickard ; ‘‘Sandstone Intrusions near Santa Cruz, California,”
J. F. Newsome and J. C. Branner; ‘On the Pleistocene Deposits
ot -lowa; Samuel) Galvin: “Problems .ot] the Ouatenmany,

Deposits of the South Platte Valley,’ George L. Cannon;
‘“Physiography of the Boston Mountains, Arkansas,” A. H.
Purdue; ‘Some Problems of the Dakota Artesian System,”
James E. Todd; “A Quantitative Study of Variation in the
Fossil Brachiopod Platystrophia biforta, Schl.,” E. R. Cumings
and A. V. Mauck; ‘Interpretation of Some Drainage Changes
in Southwestern, Ohio,’ W.:G: Tight; “The: Minerals” and
Mineral Localities of Texas,” F.W. Simonds; ‘‘The Minerals
Associated with Copper in Southeastern Arizona and South-
western New Mexico,” G. H. Stone; ‘‘ The Extinct Glaziers of
New Mexico and Arizona,’ G. H. Stone; ‘‘ Note on Certain
Copper Minerals,’ A. N. Winchell ; ‘‘ The Areal Geology of the
Castle Rock’ Region,” W.T.Leée.. The officers of the section
were: C. R. Van Hise, vice president, and H. B. Patton, secre-
tary. Those elected for the ensuing year are: O. A. Derby, of
San Paulo, Brazil, vice president ; and F. P. Gulliver, of South-
boro, Mass., secretary. : G

REVIEWS

THREE PHASES OF MODERN PALEONTOLOGY

I. Usntacrinus: Its Structure and Relations. By FRANK SPRINGER.
(Mem. Mus. Comp. Zoél., Vol. XXV, pp. I-90, 8 pls.,
Cambridge, 1901.)

Il. Onskany Fauna of Becraft Mountain. By Joun M. CrarKe.
(Mem. New York State Mus., Vol. III, No. 3, 128 pp., 9
pls., Albany, 1900. )

III. Stvatigraphical Succession of the Fossil Floras of the Pottsville
‘Formation in the Southern Anthracite Coal Field. By Davip
Waite. - (U.S. Geol. Surv., Twentieth Ann. Rept., Part 11,
Pp. 749-930, 13 pls., Washington, 1901.)

Three notable contributions to our knowledge of fossil organisms
have lately appeared from the hands of the printer. They are notable
as making distinct advancements in paleontology. ‘They are notable
as typifying the three distinct phases into which the science relating to
ancient life has finally resolved itself. They are notable as model
works of their kind, each representing the general subject from a very
different viewpoint, and hence show very diverse modes of treatment
and the very diverse character of paleontological] inquiry of today.

1. The crinoids have long been an attractive theme to geologists.
Ever since the discovery by Marsh,*in 1870, of the remarkable Creta-
ceous crinoid afterward called Uintacrinus, great interest has been
taken by paleontologists in each new accession. Grinnell, Meek,
Willison & Hill, and Logan, in this country, and in Europe, Schliter
and Bather, have described carefully the known material. It has
remained for Mr. Springer, so long intimately associated with the
lamented Wachsmuth, to give us a magnificent monograph on the sub-
ject, including a large amount of new information derived from rich,
lately discovered material. And this after one would naturally think
that about all that it was possible to say had been said.

The special charm and value in this work is the strictly morpho-
logical character that it presents. In this respect it fully keeps up the

539

540 REVIEWS

same high standard of excellence that made the Worth American Crin-
oidea Camerata of Wachsmuth & Springer so acceptable to all stu-
dents of fossil organisms.

Unusual interest centers in the composition of the base of Uinta-
crinus. A feature that has long been regarded as fundamental in the
separation of the larger taxonomic groups is here found in one and
the same species. ‘Considering the apparent identity of these forms
in every other point of structure, coupled with their mode of occur-
rence and association, I do not see how such association [as made by
Mr. Bather] can possibly be made in this case. We therefore have
apparently to deal with a case of individual variation as to this sup-
posed primitive character, within the limits of a species. That is to
say, in this species, living in the same locality, having the same
environment, floating in the same mass, certain individuals matured to
represent one stage of larval development, z. ¢., with infrabasals ; and
others in another stage, 7. ¢., with basals only.

In short, there are the two supposed distinct types, Monocyclica and
Dicyclica, occurring in both young and adult of one and the same species.
It will not do to say that the species is dicyclic, but in certain individuals the
infrabasals are not developed, or are kidden by the centrale, or have dis-
appeared by atrophy, If this were so, the centrale ought to be interradial
in both cases; whereas, as already shown, its orientation is reversed from
one to the other, precisely as in the typical monocyclic and dicyclic forms.

Such a condition is believed to be unique among the crinoids.
The bearing upon certain recently proposed classifications of the
crinoids is also important. Bather and Jaekel have both severely
criticised Wachsmuth & Springer’s classification and have erected
schemes that are ‘sought by the modern biologist.”

‘There is no doubt,” says Mr. Springer, “that each author who
undertakes to express his ideas of descent in a new scheme of classi-
fication does so in the belief that his own structure is a substantial
pyramid whose base is firmly established upon the ruins of the top-
heavy contrivances of his predecessors. With regard to the crinoids,
there have appeared, since our monograph of the Camerata, two
elaborate classifications, each avowedly based upon phylogenetic prin-
ciples, viz., that of Mr. Bather, already mentioned, and one by Dr.
Jaekel, whose general researches and great works upon the crinoids
of Germany constitute a rich contribution to science. ‘The views of
the latter author are to be developed in full detail in his magnificent

REVIEWS 541

“Stammesgeschichte der Pelmatozoen,” the first part of which,
embracing the Thecoidea and Cystoidea, has just been published.
He, likewise, finds fault with Wachsmuth & Springer, because, in his
opinion, they have dealt with the morphological conditions as they
found them too much from an anatomical standpoint, and have not
sufficiently taken into account the import of the modifications due to
descent. He finds in the changes in the systematic arrangement of
the crinoids made by Wachsmuth & Springer in their successive
writings, proof that the right road to the solution of the great ques-
tions of classification had not yet been found. We have, therefore,
two new and almost simultaneous phylogenetic classifications, by two
of the most eminent living authorities, both predicated in part upon
the insufficiencies of Wachsmuth & Springer’s system, and each
believed by its author to be a new and correct reading of the race
history of the crinoids. From such sources, and following such a
preface, we should not unnaturally expect a brilliant illumination of
the road, in search of which their predecessors have floundered in
darkness. But to our dismay we find that instead of celebrating a
conclusive settlement of these questions, we are only invited to witness
fresh controversy. For these new chroniclers do not read their history
alike, and their two classifications are about as diametrically and
fundamentally opposite as anything could be.”

Uintacrinus presents a striking resemblance to the living crinoid
Actinometra in the eccentric position of its mouth, the central position
of the anus, the absence of any calcified ambulacral skeleton on disk,
arms and pinnules, the structure and distribution of the disk ambu-
lacra, the form and proportions of brachials, and distribution of
syzygies, the variable size of the anal tube, and the instability of the
base. |

The systematic position of Uintacrinus will be a matter of contro-
versy for a long time to come. As yet hardly any two authorities
agree in placing it in the same position.

Wherever it may belong, and whatever its line of descent, there is no
doubt that Uintacrinus is both a protean and convergent form more remark-
able than any we have hitherto encountered among the crinoids.. Along with
great variability in the base and interbrachial regions, it combines :

The interbrachial system and fixed pinnules of the Camerata ;

The pliant test of the Flexibilia ;

The large visceral cavity of both of these;

542 REVIEWS

The exoclyclic disk and open ambulacara, and the arms, pinnules, and
syzygies of Actinometra ;

The free-floating character of the Comatule ;

The dicyclic base of the Dicyclica;

The monocyclic base of the Monocyclica.

A noteworthy feature that should receive special mention in con-
nection with this monograph is the distribution of study material,
illustrating the points and structures discussed, to some of the prin-
cipal museums. It is a feature that could well be imitated by other
workers in paleontology. In this way the principal type specimens
have been deposited in the Museum of Comparative Zoélogy. A large
slab has been placed in the National Museum at Washington; it con-
tains specimens exhibiting most of the characters discussed. A fine
series of specimens have also been sent to the British Museum, and to
the Royal Museum of Natural History at Berlin, where they will be
accessible to European students.

2. As stated by Dr. Clarke, in his prefatory note to the Orzskany
Fauna of Becraft Mountain, the original purpose of his work was
solely to depict the character and composition of the Oriskany fauna
of Becraft Mountain, which of itself displays many features of interest.
In its progress, however, various questions have arisen which concern
the intrinsic value of the fauna and its importance in correlation. Yet
without an understanding otf the fauna itself it would be impracticable
to discuss the latter problems, and for this reason the title of the paper
is restricted to the principal argument of the work, to which the dis-
cussions of a somewhat broader scope are corollaries.

A fauna which finds its highest development at Becraft Mountain,
near Hudson, in Columbia county, N. Y., links together in the char-
acter of its species, the calcareous shales and limestones of the Lower
Helderberg and the normal Oriskany sandstone.

The interesting bearings of this assemblage of species, its new forms and
new associations and its real importance in the correlation of the Lower
Devonian are sufficient reason for presenting its characteristics in detail.

A brief account of the stratigraphy of the mountain is given, and
also the general New York section as recently revised. Accompanying
these is a small geological sketch-map of the Becraft Mountain syncline.

The greater portion of the memoir is devoted to the description of
species, which are finely illustrated by nine plates of figures. A table
of the vertical range of species occurring in the region is given.

REVIEWS 543

With our present knowledge there are thus 113 recognizable, distinct
specific forms in the fauna of the Oriskany at Becraft Mountain, and of these
94 are identifiable with species already known or are clearly new forms
peculiar to the fauna. Of the 94, 25 preceded the introduction of the
Oriskany sedimentation, having been first described from the fauna of the
Helderbergian. In the arenaceous beds of the Oriskany 23 occur; 10 range
upward into the faunas of the Upper Helderberg (Ulsterian), but a part of
these are restricted to the sandy, lower beds of this formation (Schoharie grit),
and others have been noted only in the chert beds of Ontario, Canada, where
the intermixture of Oriskany and Onondaga species is well marked and has
been recorded by Schuchert. The fauna contains 35 species which so far as
known are peculiar to it. On farther analysis of the table, it is evident in
some cases that species which range down and upward are restricted to
particular groups. Thus the alien trilobites are from the Helderbergian ; the
gasteropods are exclusively Oriskany; while the alien lamellibranchs are
mostly Helderbergian. But the leading factor of the fauna, the brachiopod,
has its derivation as freely from below as in the Oriskany invasion.

The faunal values of the different species are then summed up.
Concerning the nature of the Oriskany fauna of New York, the
author says:

The fauna of the calcareous Oriskany is in no sense a mixed assemblage,
or an intermingling of faunas of adjacent provinces. The sequence of life
has continued without interruption from the Helderbergian (Kingston beds)
into the sediments of the Oriskany and Onondaga limestone.

It is extremely probable that important variations from the fauna of the
Catskill shaly (New Scotland) limestone had already made their appearance
in the Becraft limestone, and that we first become acquainted with some of
these in the study of the calcareous Oriskany. No proof therefore could be
adduced more emphatically confirmatory of the intimate faunal relations of
the Helderbergian with the Oriskany fauna and its successors than the facts
brought forward in this paper.

The fauna discussed is that of the calcareous facies of the Oriskany for-
mation. The sedimentary deposits of this and the neighboring sections were
essentially limestones notwithstanding the silicious content, whether diffused
through the mass or segregated as cherty secondary product. In the earlier
presentation of this fauna it was regarded as of lower Oriskany horizon on
account of the presence of many Helderbergian species, but we believe it will
be more correctly construed as the representative of the proper and normal
Oriskany fauna, the true fauna of this time-unit inclosed in the sediments of
its proper habitat.

A chapter is devoted to discussing the Devonic age of the Helder-
bergian fauna and the base of the Devonic system in New York.

544 REVIEWS

The fact of the presence of numerous Helderbergian species in the fauna
of the Oriskany of Becraft Mountain, as an integral part of that fauna, not a
casual intermixture, is sufficient demonstration that the fauna of the Helder-
bergian became modified in its continued existence by the departure or
extinction of certain of its species only. A fair percentage kept the field up
to the time of and pending the incursion of species of the early Oriskany. In
this way the former became a true and proper part of this new fauna with
whose indicial species it coexisted throughout the remainder of its duration.
A modification so gradual as to permit such an uninterrupted existence can-
not sever the close relation of the one fauna in its entirety to the other. It is
therefore a natural corollary from the account given of the Oriskany fauna,
to consider briefly the relation of the organic assemblage constituting the
typical and normal Helderbergian to the Devonic type of organic life, and
that formation in its relation to the Devonic system.

Arguments are adduced from the intrinsic characters of the fauna,
from correlation, and from stratigraphy.

3. Of very different nature is Mr. White’s work on the plants of the
Pennsylvania Coal-measures. The stratigraphical interval which he
considers is occupied by the Pottsville formation, Pottsville series, or
Pottsville conglomerate. It is described as a series of largely arena-
ceous beds of variable thickness which in eastern Pennsylvania lies
between the Maunch Chunk red shale, or distinctly Lower Carbonif-
erous, and the lower productive Coal-measures, or distinctly Upper
Carboniferous.

The investigation was intended to establish three propositions: (1)
The exploitation and elaboration from, a stratigraphic standpoint, of
the fossil plants of the Pottsville formation in the type region of the
southern anthracite coal field; (2) the critical analysis and. compara-
tive study of the plant material collected, with a view to the discovery
of the existence of. any natural paleontological subdivisions, zones, or
horizons, and their paleontologic characters, or the species of strati-
graphic value ; (3) the discovery of the paleontologic limits as differ-
ing or as agreeing with the lithologic limits of the type section, and
the consequent paleontologic definition of the formation.

The main aim of the investigation is the paleontologic definition
of the terrane. |

Two other, largely concomitant, results that are either economic or sci-
entific in their nature have also been reached in the process of the elaboration
of the fossil plants of the formation in the typical region. The first, of some
economic interest, is the correlation of the groups of beds, or of individual

REVIEWS 545

coals wrought in disconnected or somewhat isolated portions of the southern
anthracite field. The other, which concerns the question of general geologi-
cal correlation, is the acquisition of data for the determination of the age of
the Pottsville formation —z. ¢., (a) the time interval represented by the type
section, and (4) the equivalents, in a broad sense, of the formation in other
basins of this province and in other parts of the world. Incidentally also,
through the discovery in the Pottsville of floras already more or less com-
pletely known from isolated and uncorrelated terranes in other regions of the
United States, the way is opened to the proper reference and correlation of
those terranes with the Pottsville, or with portions thereof.

As introductory to the consideration of the plant remains as a
means of geological classification there are presented a sketch of the
general geological structure of the southern anthracite coal field, a
description of the Pottsville formation in the typical locality, the com-
position of the formation, the coals contained, and their commercial
names as guide horizons.

A type paleobotanic section of the Pottsville is then discussed, and
the groupings of beds are enumerated. The various species and their
observed distribution within the formation and in the field is given
in detailed tabulated form, in which all the species are listed, together
with their respective ranges. The floras of the several subdivisions
indicated are discussed in some detail. The correlative comparison
of the horizons of the southern field with those of the other anthracite
fields is of special interest.

A considerable portion of the memoir is devoted to the description
of the most characteristic species of plants found in the Pottsville,
and to notes on many other species.

One of the most surprising, as well as interesting, facts observed in the
study of the Pottsville floras is the large element that is common in the
latter and to the flora described by Sir William Dawson from the supposed
Middle Devonian beds of St. John, New Brunswick. In fact, taking into
view the entire flora of the Pottsville formation in the Appalachian province,
the identities in the composition of the floras are so great, with respect to
both genera and species, as to leave little room for doubt that we have in the
“Fern Ledges” at St. John beds of nearly the same age as the Pottsville
formation in Pennsylvania. On the whole, while recognizing in the Potts-
ville formation a group of terranes equal in rank to the Lower Coal-measures,
Alleghany series, etc., 1 do not favor a classification which relegates the
entire formation hard and fast to the Upper Carboniferous, but I even antici-
pate a possible necessity for its permanent division into two groups, the lower

546 REVIEWS

of which may eventually perhaps be referred to the Lower Carboniferous.
From a paleobotanic standpoint the Pottsville formation is the beginning of
Mesocarboniferous.

Briefly stated, the following are some of the general conclusions
reached :

No evidence of a marked or general unconformity between the
Pottsville and Mauch Chunk is noticeable in this region, though at
various points within several hundred feet of the strata beds of small
bowlders or coarse conglomerates are imposed, in knife-edge contact,
on the distinctly uneven surfaces of olive-green mud-beds.

The flora in the roof of the Buck Mountain coal, or its supposed
equivalents, at the base of the Lower Coal-measures at Pottsville is a
typical Coal-measures flora, very distinct from the floras typical of the
Pottsville formation.

The fossil plants of the Pottsville formation in the type region
exhibit a rapid development and series of changes or modifications,
which, if treated with great systematic refinement, are of high strati-
graphic value. With the exception of the species from the topmost
beds of the formation, the ferns are, in general, readily distinguished
specifically from those at the base of the Lower Coal-measures, or
Alleghany series, as recognized in the northern United States, while
the floras of the lower portions of the section are found, in passing
downward, to bear still less resemblance to those of the Lower Coal-
measures. [wo principal divisions of the formation, to which com-
paratively few fern species are common, are recognized. ‘These divi-
sions, which coincide with the natural grouping of the Lykens coals,
are here termed the Lower Lykens division and the Upper Lykens
division. A portion, including about two hundred feet of the type
section between these two paleontologic divisions, contains a mixed
flora, and has been temporarily designated the Lower Intermediate
division.

Further paleontologic study of the Pottsville formation appears to
fully confirm the earlier conclusion, based on the examination of the
plants, that the thinner sections of the formation along the northern
and western borders of the Appalachian trough do. not contain beds
as old as those in the lower portion of the thick sections along the
eastern border, ¢. g., in the Schuylkill and Great Flat Top regions.
The positions of the respective floras in the sections plainly indicate
a transgression of the sea toward the north and west during Pottsville
time.

REVIEWS 547

Both lithologically and paleontologically the Pottsville formation
constitutes a division of the Carboniferous coérdinate with the Lower
Coal-measures, Alleghany series, etc. As such it forms the lower
member of what may in a broad sense, be termed the Mesocarbonifer-
ous in the Appalachian province.

The lowest beds in the thickest sections, which appear to be con-
tinuous by transition with the deposition of the Maunch Chunk red
shales, are perhaps to be regarded as coarse, coast-detrital redeposi-
tions, contemporaneous with the uppermost beds of the red shale or
marine Lower Carboniferous sediments of other regions.

The flora of the Pottsville formation is so far identical, in both its
generic and its specific composition, with that from the supposed
Middle Devonian beds at St. John, as to leave no room for a great

difference in the age of the latter.
CHARLES R. KEYEs.

Towa Geological Survey; SAMUEI, CALVIN, State Geologist ;
A. G. LEonarD, Assistant State Geologist ; Annual Report for
1900 (Vol. XI, 519 pages, 12 plates, 43 figures, g maps. Des
Moines, 1901).

In scope and style this report follows closely the previous volumes
of the series. It includes the usual administrative reports, the statis-
tical reports of the mineral production, and detailed reports on
Louisa, Marion, Pottawattamie, Cedar, Page, and Clay and O’Brien
counties, the last two being treated together. These reports are
written, by J. A. Udden, B: L. Miller; J. A. Udden, W. H. Norton;
Samuel Calvin, and T. H. Macbride respectively. They contain a
careful review in each case of the local geology and serve to put on
record facts and observations which may be used in later discussion of
the theoretical problems involved. Following the usage of the survey,
these discussions are taken up as fast as the development of the gen-
eral survey allows them to be intelligently discussed. In the case of
Louisa county, for example, the problems of the drift of that region
have already been discussed in their theoretical phases in Mr. Udden’s
report on Muscatine county, Mr. Norton’s report on Scott county, and
Mr. Leverett’s well-known monograph.* In the Louisa report Mr.
Udden gives many interesting and valuable details confirmatory of the

1U. S. Geol. Surv., Mon. XXXVIII. The Illinois Glacial Lobe.

548 REVIEWS

general conclusions already reached. In the Cedar county report, on
the other hand, Mr. Norton places clearly before the reader the facts
as, to the structure and situation of the beautiful land forms to which
McGee has given the name Zaha, and also states the various hypotheses
which have been suggested for their origin. He makes no attempt,
however, to balance the probabilities and reach a decision in the
matter. The same method is observed in the discussion of the high
dips observed in the Gower limestone of the Niagara. This is con-
fessedly a difficult problem, and its solution is wisely left until the
whole area shall have been studied. .

In the reports on Page and Pottawattamie counties the survey
takes up the southwestern portion of the state. Apparently most inter-
esting results are likely to follow this survey. Mr. Calvin’s suggestion
that the drift of Page county may be the pre-Kansan instead of the
Kansan fits in with many of the known facts of the field in Iowa and
neighboring states. It is a far-reaching suggestion, and the develop-
ment of the hypothesis will be watched with interest. In Pottawat-
tamie county the facts are possibly susceptible of either interpretation.
Mr. Udden does not, however, discuss the probabilities. His report
is particularly interesting to the general student in the large use which
he makes of mechanical analyses in the study of the drift. This isa
method which seems likely to become more and more useful in the study
of unconsolidated materials. The important results already reached
by Whitney in the study of soils seem merely the forerunners of what
is possible on a wider application of the method to the study of geo-
logical problems in general. Mr. Udden’s hypothesis of the origin of
the rivers of the county is ingenious, but the argument is not alto-
gether convincing. The suggestion that the surface of a great ice
sheet would afford a better opportunity for the development of such a
regular system of rivers than the drift surface left by the melting away.
of the ice seems open to question. The surface of the Kansan drift,
it is true, is everywhere notably even. It is impossible, however, to
suppose that so thick and widespread a drift sheet as this would be laid
down without the development of moraines and other surface irregu-
larities. Yet such irregularities, where they now occur, are very faint,
and are apparently the marks of much greater original irregularities.
The present plane surface seems likely to be a result of erosion itself.
Remembering the long period since these older drifts were deposited
and the manifestly great erosion to which they have been subjected, it

REVIEWS 549

seems a bit fanciful to believe that the streams now have a general
course originally developed on the top of the ice. It is also difficult
to imagine how streams, granting that they were developed on the ice
in a symmetrical fashion, could be transferred to the drift surface
during the melting of the ice without disturbance of their arrange-
ment. If the disappearance be supposed to be accomplished by. frontal
retreat, the edge of the ice itself will give a sufficiently marked line for
the development of the cross streams, and the major streams would in
all probability be developed normal to the ice front. This apparently
accords with the facts of the field and does not involve the difficulty
of accounting for the development of a vigorous surface drainage on
the ice and its transfer to the drift surface without notable destruction
of its arrangement. The whole problem is an interesting one, and
certainly Mr. Udden’s hypothesis, if valid, is widely applicable. Ifwe
are to be able from the present stream arrangement to infer so much
regarding the condition during glacial times a considerable step in
advance is made. ‘This fact warrants the closest skepticism in exam-
ining the hypothesis.

The volume as a whole is one which geologists will welcome.
While the long series of annual reports made up of separate county
reports is confusing to one not familiar with the region, the survey is
systematically building up a work which will prove valuable not only
to students of the local geology but to geologists in general. The
work will lend itself to final summary treatment, and in the meantime
the county reports not only serve a useful purpose locally, but enable
outsiders to keep in touch with the results. The progress map of the
drift mapping (plate II) is particularly valuable in this regard.

[ep bi ee

Beach Structures in the Medina Sandstone. By H.L. Faircuizp.
American Geologist, Vol. XXVIII, pp. 9-14, plates 2-6.

In the July number of the American Geologist Professor H. L.
Fairchild discusses the “‘ Beach Structures in the Medina Sandstone.”
The features considered are certain troughs and swells of a breadth
ranging from thirty to eighty feet. These are assumed to be of the
same nature (probably in some cases the same examples) as those which
formed the basis of Dr. Gilbert’s theory of “giant ripples.” Gilbert
had concluded (Bulletin Geol. Soc. Amer., March 1899) that with

550 REVIEWS

“certain combinations of amplitude and frequency” the largest ocean
waves might produce ripples having a length equal to half the height
of the wave. The supposed giant ripples observed at Lockport, N. Y.,
reached a length of thirty feet, implying waves sixty feet high in the
Medina sea. Fairchild explains the same features as spits or rather
beach ridges, sometimes observed in series of as many as four crests
with intervening troughs. Of the fifteen halftones with which the
article is illustrated, about one third are clear enough to show without
doubt that the features referred to are of the same nature as those dis-
cussed and figured in Gilbert’s article. Four of the cuts show sand
ridges on the present shore of Lake Ontario which are supposed to
exemplify the process by which the features in the Medina were made.
The remaining cuts and a few expressions in the article leave it in
doubt as to whether the features observed were all of the type from
which Gilbert’s conclusions were drawn. ‘The latter found no inter-
vals of more than thirty feet from crest to crest; the article under dis-
cussion refers to similar spacings of eighty feet. Gilbert’s discussion
of the features as ripples implies a symmetry or at least unity in the
trough by which its breadth could be fairly estimated when only a
portion could be seen. Fairchild speaks of the trough as only a
“negative or passive element” which may have any width varying
with the accidents which determine the location of a new beach ridge.
The observations of both, and both interpretations, involve abundant
cross bedding. Itwould seem that the two processes hypothesized would
produce different dispositions of this crossbedding, as well as differ-
ences in the symmetry of the trough. Gilbert shows by diagram the
necessary arrangement of the crossbedding under various suppositions
as to the shifting of the ripple system during deposition. It is quite
improbable that the patterns in vertical cross section thus produced
would be duplicated in a series of beach ridges. In the latter case the
lamination on the landward side of an outer ridge should be plainly
distinguishable from the lamination on the lakeward slope of the next
ridge within. Thelamination corresponding to this latter slope should
disappear only in the most extreme cases of landward migration of
the ridge, a condition quite improbable where a series of ridges is pro-
duced. Fairchild’s objection that ripple marks are not found of
intermediate sizes between eight inches and the “‘ giants”’
of interest.

is certainly

N. M. F.

REVIEWS gol

The Beaufort's Dyke, off the Coast of the Mull of Galloway. By H
G. Kinanan. Proceedings of the Royal Irish Academy,
Third Series, Vol. VI, No. 1.

In the Proceedings of the Royal Irish Academy, Third Series,
Vol. VI, No. 1, Mr. H. G. Kinahan, District Surveyor (Retired), H. M.
Geological Survey, describes ‘The Beaufort’s Dyke, off the Coast of
the Mull of Galloway.” This “dyke” is a deep trough in the bed of
the Irish Sea nearer the coast of Scotland. The plate accompanying
the article reproduced from the Admiralty charts, shows a long line of
soundings of 100 to 148 fathoms, surrounded by a comparatively level
bottom 70 or 80 fathoms deep. Attention is just now directed to
these depths by the proposition to construct a tunnel from Scotland to
Ireland. The geological interest lies in the fact that in this trough
“there are sands, gravels and their adjuncts, at depths of from 120 to
144 fathoms that are carried backward and forward similarly to those
on an ordinary sea beach.” This is much deeper than the waves and
currents even on most exposed coasts have been supposed to act. The
vigor with which this deep washing occurs may be inferred from figures
given in a table, which show that at one point the bottom was cut
down by erosion froma depth of 117 fathoms to a depth of 146 fathoms
from 1894 to 1897 or nearly 60 feet a year. At other times filling was
similarly rapid. The origin of the dyke or trough itself is left in doubt
but fauiting and glacial action are suggested. The origin of currents
capable of acting at that depth is the main subject of interest. It is
assumed that while the effect of wind waves and tide waves is reduced
to zero at moderate depths there are deep currents induced by these
superficial waves, to the depth of whose action no definite limit is
assigned. The author contrasts these currents with the ordinary tidal
races which may be supposed to erode shallow places more than deep
ones. He refers to his former publications for detailed proof that
denudation by these currents in question “‘is in ratio to the depth of
the water.” The whole article goes to show that when the movements
of water are controlled by channels or reflected from irregular shores,
the coarseness of the sediments at given depths bears little relation to
the mathematically computed power of the water for those depths.

INVES Es

RECENT PUBLICATIONS

-—American Institute of Mining Engineers, Transactions of the. Vol. XXX.
February, 1900, to September, 1900, inclusive. New York, !got.

—BAKER, FRANK COLLINS. 1. Digitations of the Mantle in Physa. 2.
Description of a New Species of Limnza. Bulletin of the Chicago
Aeadmy of Sciences, Vol. II, No. 4. July 1, Igor.

The Molluscan Fauna of the Genesee River. [Reprint from the Ameri-
can Naturalist, Vol. XXXV, No. 416, August, 1go1.] Ginn & Co.,
Boston.

—BIELER, THEODORE. Etude Préliminaire sur le Modelé Glaciaire et le
Paysage Drumlinique Dans la Plaine Vaudoise. [Bulletin de la Société
Vaudoise des Sciences Naturelles. 4° S., Vol. XXXVII, No. 139. |
Librairie F. Rouge, Rue Haldimand, Lausanne, Mars, Igol!.

—DrEAN, BASHFORD. Paleontological Notes. I. On Two New Arthro-
dires from the Cleveland Shale of Ohio. II. On the Characters of
Mylostoma Newberry. III. Further Notes on the Relationships of the
Arthrognathi. Plates III-VIII. [New York Academy of Sciences,
Memoirs, Vol. II, Part III, 1901.] Published by the Academy, New
York, Igol.

—DEWALQUE,G. Mélanges Géologiques. Hutiéme & Derniére Série. Brux-
elles et Liége, 1897—I9g00.

—ELzs, R. W. The Devonian of the Acadian Provinces. [Reprinted from
the Canadian Record of Science, Vol. VIII, No. 6, for July, tgo1, issued
August 10, 190T.|

—HamBeErG, AXEL. Geologiska och Fysiskt-Geografiska Undersékningar
Sarjekfjallen. Tryckt i Central-Tryckeriet, Stockholm, Igol.

—HOLLENDER, ARTUR. Om Sveriges Nivaférandringar efter Manniskans
Invandring. Akademisk Afhandling som med Tillstand af vidtberémda
filosofiska Fakultetens i Upsala mattatisk-naturvetenskapliga Sektion

till offtentlig Granskning framstalles. [A Larosalen N:r 7 Ons-
dagen den 29 Maj, 1901, kl.4 E. M.] P. A. Norstedt & Séner, Stock-
holm, Igol.

—HoumEs, WILLIAM H. Review of the Evidence Relating to Auriferous
Gravel Man in California. [Fromthe Smithsonian Report for 1899, pp.
419-472, with Plates I-XVI.] Washington, Igo},

552

RECENT PUBLICATIONS 555

—lIowa Geological Survey, Vol. XI, Annual Report for 1900. Samuel
Calvin, State Geologist; A. G. Leonard, Assistant State Geologist. Des
Moines, Igol.

—lowa, State University of. Bulletin from the Laboratories of Natural His-
tory, Vol. 5, No. 2. Iowa City, Iowa, May, Igol.

—JAQUET, J. B. The Iron Ore Deposits of New South Wales; with maps,
plates, and sections. [Memoirs of the Geological Survey of New South
Wales; Geology, No. 2.] Sydney, igor.

—JeEvons, H. STANLEY. A Numerical Scale of Texture for Rocks,
[Extracted from the Geological Magazine, Decade IV, Vol. V, No. 408,
p- 255, June, 1808. ]

A Systematic Nomenclature for Igneous Rocks. [Extracted from the
Geological Magazine, N.S,, Decade IV, Vol. VIII, pp. 304-316, July,
1901.] Dulau & Co., 39 Soho Square, W., London.

—KERFORNE, F. Etude de la Région Silurique Occidentale de la
Presqu ile de Crozon (Finistére). Imprimerie Fr. Simon, Successor de
slew Ovwo.. Rue des Caries, Ientics, LoOl,

—Kuautzscu, A. Bericht itber Endmoranen und Tiefbohrungen im Grund-
moranengebiete des Blattes Rastenburg (Ostpreussen). [Separatab-
druck aus dem Jahrbuch der kGnigl. preuss. geologischen Landesanstalt
fiir rg00.] Berlin, 1901.

—Lay, H.C. Recent Geological Phenomena in the “ Telluride Quadrangle”’
of the U.S. Geological Survey in Colorado. | Transactions of the Ameri-
can Institute of Mining Engineers. |

—LEONARD, ARTHUR GRAY. The Basic Rocks of Northeastern Maryland and
their Relation to the Granite. [From the American Goologist, Septem-
ber, rgot.|

—LIVERSIDGE, A. On the Crystalline Structure of some Silver and Copper
Nuggets (with Plates VII-IX). On the Crystalline Structure of some
Gold Nuggets from Victoria, New Zealand, and Klondyke (with Plates
X-XIII). [Reprinted from the Journal and Proceedings of the Royal
Society of N. S. Wales, Vol. XXXIV.]

—Mourton, MICHEL. Sur l’Etat d’Avancement du Répertoire Universel
des Travaux Concernant les Sciences Géologiques (Bibliographia
Géologica). [Extrait des Annales (Bulletins des séances) de la Société
royale Malacologique de Belgique. Tome XXXVI, 1901, pp. XVI a
XX. | P, Weissenbruch, 49 Rue du Poincoin, Bruxelles, tgo1.

—Norton, WILLIAM HARMON. Geology of Cedar County. [From Iowa
Geological Survey, Vol, XI; Annual Report, 1900, pp. 279-396.] Des
Moines, Igo1,

554 RECENT PUBLICATIONS

—Perarson, H. W., Duluth, Minn. Oscillations in the Sea-Level.
[Extracted from the Geological Magazine, N. S., Decade IV, Vol.
VIII, April, May, and June, r901.] Dulau & Co., 39 Soho Square, W.,
London.

—PuHILLIPs, WILLIAM BATTLE. Texas Petroleum. [Bulletin of the Univer-
sity of Texas, No. 5. The University of Texas Mineral Survey Bulletin
No. 1, July, 1900. ]

—RADDE, GusTAV. Museum Caucasicum. III. Tiflis, 1go1.

—RICHTHOFEN, FERDINAND VON. Geomorphologische Studien aus Ostasien
Sitzung der physikalisch-mathematischen Classe vom 18. Juli. [Zit-
ungsberichte der Kéniglich Preussischen Akademie der Wissenschaften
zu Berlin, Ig01. XXXVI.]

—SIEBENTHAL, C. E. The Silver Creek Hydraulic Limestone of South-
eastern Indiana. [From the Twenty-fifth Annual Report of the Depart-
ment of Geology and Natural Resources of Indiana, 1900.]

—Smytu, C. H., Jk. Geology of the Adirondack Region. [Reprinted from
Appalachia, Vol. IX, No, 1, 1899.]

Report on Crystalline Rocks of the Western Adirondack Regions.

Sketch of the Physiographical Development of the Adirondack League
Club Preserve. [Prepared for the Club by permission of the Director
of the United States Geological Survey. |

—Spurk, J. E. Origin and Structure of the Basin Ranges. |[Bull. of the
Geol. Soc. of Am., Vol. XII, pp. 217-270. Pls. XX-XXV.] Rochester,
April, 1901.

—STONEY, G. JOHNSTONE. Survey of that Part of the Range of Nature’s
Operations which Man is Competent to Study. [From the Smithsonian
Report for 1899, pp. 207-222, with Fig. 1.] Washington, Igol.

—UDDEN, J. A. Geology of Louisa county. [From Iowa Geological Survey,
Vol. XI; Annual Report, goo, pp. 55-126.] Des Moines, Igo1.
Geology of Pottawattamie County. [From Iowa Geological Survey, Vol.

XI; Annual Report, 1900, pp. 201-277.| Des Moines, Igol1.

—VAUGHAN, T. WAYLAND. Some Fossil Corals from the Elevated Reefs of
Curacao, Arube, and Bonaire. [From the “Sammlungen des Geologi-
schen Reichs-Museums in Leiden,” Series 3, Bd. II, Heft 1.] E. J.
Brill, Leiden, 1go1.

—Washington Academy of Sciences, Proceedings of the:

Papers from the Harriman Alaska Expedition, XXII, Entomological
Results (14): The Odonata. By Rolla P. Currie. Vol. III, pp. 217—

223, julya13loow.
Papers from the Harriman Alaska Expedition, XXIII, The Ascidians.

REGENT PUBEICA LIONS 555

By W. E. Ritter, University of California. Vol. III, pp. 225-266.
Pls. XXVII-XXX, July 13, Igol.

A New Species of Olenellus from the Lower Cambrian of York County,
Pennsylvania. By Atreus Wanner. Vol. III, pp. 267-272. Pls.
XXXI-XXXII, July 13, Igor. ;

Synopsis of the Rice Rats (Genus Oryzomys) of the United States and
Mexico. By G. Hart Merriam. Vol. III, pp. 273-295, July 26, 1901.

Papers from the Harriman Alaska Expedition, XXIV, The Willows of
Alaska.. By Frederick V. Coville. Vol. III, pp. 297-362. Pls.
XXXITI-XLII, August 23, 1901.

Papers from the Hopkins-Stanford Galapagos Expedition, 1898-9: I,
Entomological Results (1): Hemiptera. By Otto Heidemann, U. S.
Department of Agriculture. Vol. III, pp. 363-370, August 23, Igol.

—WHITFIELD, R. P. Note on a Very Fine Example of Helicoceras Steven-
soni Preserving the Outer Chamber. [Author’s Edition, extracted from

Bulletin of the American Museum of Natural History, Vol. XIV, Article

XVI, p. 219. New York, July 29, 1go1.]

Notice of a Remarkable Case of Combination between Two Different
Genera of Living Corals. [Author’s Edition, extracted from Bulletin
of the American Museum of Natural History, Vol. XIV, Article XVII,
pp. 221,222. New York, July 29, 19o!.]

Some Observations on Corals from the Bahamas, with Description of a
New Species. [Author’s Edition, extracted from Bulletin of the
American Museum of Natural History, Vol. XIV, Article XVIII, pp.
223,224. New York, July 29, 1901.]

--WINCHELL, N.H. Editorial Comment: The Archean of the Alps. [From

the American Geologist, September, 1901.]

JOURNAL OF GEOLOGY

OCGLOL ER NOVEMBER; Toor

INDIVIDUALES OF sl RATIGRAPHIC “CLASSIFICATION.

INDIVIDUALS TO BE DISTINGUISHED.

SHOULD geologists map the record of physical conditions or
the record of biological conditions—rocks or fossils? Both,
but with distinction.

When the geologist enters the field to do stratigraphic work,
one of the first problems to confront him is where he shall
divide the series of rocks he is studying; and often, when read-
ing a paper, we are perplexed to know where these lines have
been drawn and what the divisions are intended to represent.
This discussion is an effort to arrive at a better understanding of
what we classify and what may be mapped.

Since the earliest days of geologic work there has been
recognition of different kinds of rocks. For many years geolo-
gists compared the various rocks they found, and correlated
them from continent to continent on the basis of like physical
conditions represented in the similar lithologic characters of the
rocks, Identity of physical conditions was interpreted as indi-
cating the same date, but we now know that the physical charac-
teristics of rocks are repeated from time to time, and are diverse
in different provinces at the same time, and that therefore they
do not afford criteria of contemporaneous deposition. Further-
more, conditions of sedimentation are related to currents, shores,
and other moving features; the zones of deposition may migrate
VoleixesNow7: 557

558 BAILEY WILLIS

with these features, and an identical sediment accumulate in the
migrating zone successively, not simultaneously, over adjacent
areas,

After the days of William Smith a second class of divisions
arose—divisions based on the fossils which the rocks contain.
Fossils were found to occur in certain associations, which were
called faunas, and these became the basis of a classification of
rocks. On the hypothesis of special creations and destructions
each fauna, wherever occurring, represented a certain date, and
thus faunas became the significant figures expressing age. But
special creation has given way to evolution, and we recognize
migration of faunas as a fact. For instance, Walcott has stated
that it took a long time for the Olenellus fauna to move round
the globe. A greater or less time interval must elapse between
the earliest appearance of a fauna at one place and its earliest
appearance at another place remote from the first, hence, a
fauna does not indicate a precise date in the narrow sense in
which it was once taken. Professor H. S. Williams has pointed
out that in a wider sense any fauna endures a length of time,
from its initial appearance somewhere to its extinction every-
where; and this interval is an episode of evolution which has a
fixed place in geologic history. In that sense faunas have defi-
nite time values, but the discovery of that value in any case is
dependent on refined and extensive paleontologic research.

Accordingly, when studying a stratigraphic series, a geologist
may recognize distinctions of lithologic character, variations of
faunal content, and succession of physical or faunal changes.
The differences enable him to define lithologic individuals, faunal
individuals, and time intervals. Though often intimately related,
sediments and faunas are by no means necessarily bounded by
the same limits in space or in time; they constitute not identical
but unlike things. They may migrate together or independently.
Either may cease and the other continue. When each of them
has been described and discusssed in its local and general rela-
tions, the problem of correlation in terms of earth history may
be hopefully attacked; but when strata and faunas are treated

STRATIGRAPHIC CLASSIFICATION 559

under one name, confusion ensues and the conclusion becomes a
guess.

It follows that we need to recognize, define, and name three
separate things: (1) lithologic individuals; (2) faunal individ-
uals; and (3) time intervals.

LITHOLOGIC INDIVIDUALS.

Definition.— Some twelve years ago there was held a confer-
ence of the geologists of the United States Geological Survey,
and at that time the basis for the Geologic Atlas of the United
States was laid. That foundation was planned upon the simplest
lines. It was proposed that the maps should exhibit the distri-
bution of local lithologic individuals.

Referring to the Zenth Annual Report, Part I, in the account
of that conference we find the following statement (pp. 63-64):
“Among the clastic rocks there shall be recognized two classes
of divisions, viz:, structural divisions and time divisions.”
Observe that time is set off in contrast to structure, and that
structure is not defined by time. Then, defining structural units,
the report says (p. 64): ‘The structural divisions shall be the
units of cartography, and shall be designated formations. Their
discrimination shall be based upon the local sequence of rocks,
lines of separation being drawn at points in the stratigraphic
column where lithologic characters change.” Proceeding to
emphasize that, the report further says (p. 64): ‘Each forma-
tion shall contain between its upper and lower limits either rock
of uniform character or rock uniformly varied in character.” By
the latter phrase it was recognized that there might be groups of
lithologic individuals which singly could not be mapped because
too small—a difficulty which had to be met from the practical
side. Furthermore, the report says: ‘As each lithologic unit
is the result of conditions of deposition that were local as well
as temporary, it is to be assumed that each formation is limited
in horizontal extent; the formation should be recognized and
should be called by the same name as far as it can be traced and
identified by means of its lithologic characters, azded by tts

560 BAILEV WHILETS

stratigraphic association and its contained fossils.” (Italics the
present writer’s. )

Formations and fossils —YVhat the purpose of the definition is
to emphasize lithologic character, and make it the essential of
individuality of a formation, is clear, but the habit of classifying
roughly by fossils is strong, and the few words here italicized
have led many to lose sight of the distinction intended. There
is a difference between using fossils as one of several means of
identification and employing them as essential characters. In
the former case, other characters being the same, the occurrence
of a known fossil is an aid, but its non-occurrence sets no limit.
In the latter case the range of the characteristic fossils defines
the extent of the division. By the one method we may define
a lithologic formation according to all its characters. By the
second method we limit the lithologic unit by a fauna anda
fauna by the lithologic unit, when either may occur beyond the
other; and thus, combining two unlike things in one definition,
we can recognize neither.

The writer by no means advocates disregard of fossils in the
identification of formations. But abundant experience of the
ablest stratigraphers shows that classification by faunas requires
most refined investigation based upon thorough knowledge of
the rocks. To map the formations is a necessary preliminary
to determining the faunal units. We should proceed from the
simple and obvious to the complex and obscure. We should
trace out lithologic individuals, according to their constitution
and stratigraphic associations. If these leave us in doubt, fos-
sils may prove valuable ear-marks; but they should not set
limits in the discrimination of formations. When the map of
formations is made, the way is prepared for the paleontologist
to ascertain the number and bounds of the faunal units and to
draw the faunal map.

In mapping formations it is convenient to combine them in
systems—Cambrian, Cretaceous, Eocene, etc. Fossils are the
means of this preliminary rough classification, and the little
evidence required is usually obtained in studying the lithology.

SLHATIGRAPHIC CLASSIFICATION 561

But it would be a slur upon paleontology to consider such
arrangement final. Even were the dividing planes between
great systems fixed, the precise recognition might require
detailed investigation of faunas; but in many cases they are not
fixed beyond question, either by faunas or by unconformities.
Geology has not arrived at a final classification. We need still
to accumulate physical and biological facts, to keep them dis-
tinct, and to compare them from district to district, and from
country to country, before our systems can be said to be estab-
lished.

The writer is indebted to his colleague, Mr. Whitman Cross,
for a case in point—that of the Hermosa, Dolores, and Rico,
so-called formations. Quoting from Mr. Cross’s statement
before the Geological Society of Washington (as revised by him),
the case is as follows:

[A diagram exhibited] represented the problems in subdivision of the
great series of alternating shales, sandstones, conglomerates, limestones, and
strata of intermediate lithologic character represented in the Rico quad-
rangle, southwestern Colorado. This series of rocks, about 4,500 feet in

Dolores formation (Triassic) Red

Rvcosonm ationg(PernmolGarbonifenous)i a | ee) = | citaie te -rolcieiatelelcte cists ot cle

Hermosa formation (Upper Carboniferous) Gray

Paleozoic Mesozoic

thickness, extends from the base of the Upper Carboniferous to the top of
the Trias. The lower 2,000 feet of strata are characterized by Upper Car-
boniferous fossils. The intermediate three or four hundred feet by a well-
defined Permo-Carboniferous fauna, and a considerable portion of the upper
2,000 feet by Triassic fossils.

Recognizing the importance of the time divisions indicated, the Upper
Carboniferous strata have been grouped as the Hermosa formation; the
strata bearing the Permo-Carboniferous fossils as the Rico formation; the
strata containing Triassic fossils as the Dolores formation. With the Tri-
assic strata are provisionally included at the present time other strata not
known to be fossiliferous. The upper limit of the Dolores is a definite litho-
logic and structural horizon. The lower limit cannot be determined upon
lithologic grounds alone. Asa line must be drawn on arbitrary grounds, the
Dolores complex has been extended below to the uppermost stratum containing

562 BATE VWAHMEIELS:

Permo-Carboniferous fauna. This alternation of strata of different lithologic
character can be subdivided in great detail when the exigencies of mapping
require it. Zhe zmportant lines limiting the Rico formation above and below
cannot be drawn upon lithologic grounds. If lithologic character alone is
relied upon for the subdivision of this great complex embracing strata
belonging partly to the Paleozoic and partly to the Mesozoic, the formation
units would be of relatively minor importance, and the great historical sub-
divisions would not be expressed. The criterion of color applied to this
complex groups the Permo-Carboniferous with the Trias. This has been
done in previous general discussions of this complex in this region. In fact,
the Permo-Carboniferous beds should be grouped with the Carboniferous if
the larger elements of time division are to receive recognition upon the geo-
logic map of the Rico quadrangle. (Italics the present writer’s.)

There are two questions involved: First, what Mr. Cross’s
map expresses; second, what it should express. The Hermosa,
Rico, and Dolores are clearly not formations in the sense defined
in the Zenth Annual Report, for lithologic continuity is divided
at the top and bottom of the Rico on the basis of contained.
fossils. They are intended to be faunal divisions, but their
limits are very ill defined, since fossils are commonly so few in
the Red Beds that the finding of them higher or lower in the
series is a matter of accidental discovery rather than of occur-
rence. Phe map appears, then, to express indefinitely, the
arbitrary application of a time scale (Paleozoic, Mesozoic or
Carboniferous, Permo-Carboniferous, Triassic ) to a series which
is capable of division into lithologic individuals. Were it so
divided it would yield a record of physical conditions, a record
which is now obscured by this classification.

It is an important fact that conditions of erosion and deposi-
tion were continuously favorable to the accumulation of red
sediments while biologic migration or evolution modified the
organisms present from Paleozoic to Mesozoic types. But in
discussing the formations their continuity is the essential, just
as in describing the faunas their discontinuity would be. To
impose the division of the latter upon the lithologic individual
is misleading, and to call the faunal units ‘formations’ in a
publication for which that term has been accurately defined is a
misuse of the word.

SLRATIGRAPHIC CLASSIFICA TION 563

A very similar case is that of the Shenandoah limestone,
which carries Cambrian fossils in its lower portion and Calcifer-
ous fossils near the top. But, though it thus corresponds to
parts of two great periods, as we usually classify geologic time,
it is mapped as one formation because it is a lithologic unit.

Formation names. —‘‘The formation should be recognized and
should be called by the same name as. far as it can be traced
and identified by means of its lthologic characters, aided by

d

its stratigraphic association and contained fossils.” Following
this rule (Zenth Annual Report, Part I, p. 64), various cases may
arise, some of which are illustrated in the accompanying dia-
grams.

Figure 1 shows the mz shale passing into a limestone which
retains identical stratigraphic associations. Being exactly con-
tinuous stratigraphic units, they should retain the same geo-
graphic name on grounds of convenience and simplicity.

Figure 2 shows the m shale grading into a limestone with
prolonged overlap, so that the two rocks must be discriminated
in one area. Not only are they lithologically different but they
have different stratigraphic associations, and they should receive
distinct geographic names: m shale and z limestone.

Figure 3 shows the #z shale continuing as an individual with
reduced thickness into new stratigraphic associations. Individu-
ality is not dependent on thickness nor on stratigraphic associa-
tion only; it is determined by continuity, and the formation may
retain its name, # shale. But the group pms cannot then be
called the m group, because m would have two meanings, one
for a simple part, and one for a complex whole.

Figure 4 shows the mm shale replaced by five formations, two
of which, & and J, are shales like it. Individuality is here lost
in multiplicity. Neither B nor D can be distinguished as repre-
senting m, which must, accordinly, give way where they divide
into two. The complex ABCDE may be called the m group,
since it has equivalent stratigraphic associations with 7.

Figures 5 and 6 illustrate the occurrence and naming of local
lenses in a formation where such divisions are not of sufficient

564 BAILEY, WAELTS:

extent to justify a distinct series of names for the lens and the
parts of the formation above and below it. One lens may receive
the formation name, but two or more must be named distinctly.

Figure 7 presents the case of a formation which isa least
practicable lithologic individual for mapping, but which includes
fractional parts that are important in discussion. The fractional
character of the parts may be indicated by calling them mem-
bers, each member being given a distinctive geographic name.

Lithologic individual and thickness.— Thickness is not an essen-
tial character of a lithologic individual. Layers, strata, beds,
and their complexes vary so generally in this respect that indi-
viduality based upon constancy thereof must be lost in a
short distance. And if thickness be not constant for any one
individual, still less can volume be considered an element of the
definition of formations in general. Lithologic individuality
knows no such limits. Continuity of rock character is the
essential core of the definition, and this may extend through a
hundred or a thousand feet or more or less.

The writer has avoided the use of the phrase ‘‘lithologic
unit,” because discussion developed the fact that, in the minds
of some, unit means a definite quantity, and these persons think
that a thick formation should be a group or series, because of its
magnitude. Not volume, but uniformity of constitution, defines
a lithologic individual or ‘‘ formation.”

Lithologic individual and time.—Thickness of strata, at right
angles to bedding, is considered a measure of the epoch of
deposition, due account being taken of the estimated rate. Thus
for that particular place the lithologic unit is significant of defi-
nite lapse of time. It bears, however, no mark of date. If we
trace the formation a hundred or a thousand miles, and again
measure its thickness, another determination of an epoch is
obtained, again without date. Are the epochs identical? The
common assumption is that they are, and one reads of the epoch
of the Dakota formation without reference to locality. Thus
used, the phrase should mean all the time during which the
Dakota sandstone was spread from its eastern to its western

——— = = = L l oe | i
= - = = | | ]
= : = | ) ies
== = 2 I
= = — L |
m shale or formation Mm limestone or formatwun
it
= = ———
E = —— —— | n| | iE
ae eo T | es T
m shale or formation nh limestone or formation.
Ill
= = = ae IE = I
| =)
= = il Z = . je zs [ : Eas

C group

= mm = = - =

m= shale or formation

s lumeslone or formatton
m_ shale or formation
p sandstone or formation

m group

limestone or formattor
shale or formation
sandstone or formation
shale or formation
sandstone or formation

Eta!

m shale or formation

and m lense

m shale with n and p
sandstone lenses

M skale or formation including p fireclay mem-
ber, C coal member, and a tire clay member

566 PALLET VUES EES)

limit. Since that transgression took a long time, the eastern
Dakota sandstone is older than the western; and as the advanc-
ing zone of sand accumulation, following the shore, was in turn
followed by a zone of shale deposit, the eastern sands were being
buried under Benton shale while the western sands gathered.
When the transgression ceased and Niobrara chalk was laid
down, the conditions were more nearly uniform throughout the
sea. Niobrara then probably marks approximately one and the
same epoch throughout its extent. Benton may do so in less
exact degree; and Dakota corresponds at an eastern point to
the beginning, at a western to the closing, of the cycle of
deposits.

It is convenient to conceive the time scale as marked by
horizontal lines. On sucha scale the Dakota formation would
be represented by a diagonal line, and the Niobrara by a nearly
horizontal line.

The writer does not put these ideas forward as new; but in
defining a formation it is important to clear away certain mis-
leading conceptions that appear in the literature and in discus-
sion of the points at issue. Only at a particular place does a
formation belong to a definite age: when traced to another
locality it may be older or younger.

FAUNAL UNITS.

Classification by faunas.— The term faunal unit or individual
is here used as a parallel to lithologic individual, to designate a
set of strata characterized by a common fauna. The writer does
not undertake to say whether a fauna should be defined by
varieties, species, or genera; by recognition of one or many
associated organisms. It may be granted that a fauna is some-
thing which each working paleontologist will define for himself
within certain broad limits, just as lithologic character is some-
thing which each stratigrapher defines for himself within certain
limits. But classification of strata by faunas is a different thing
from classification by formations.

Professor Williams, in 1897, published a paper on dual

STRATIGRAPHIC CLASSIFICATION

567

nomenclature, in which he clearly stated the distinction which

the present writer would now
individual and a faunal unit.

emphasize between a lithologic

Following the language in which a formation is defined in
the Zenth Annual Report, a faunal unit may be defined as fol-

lows:
LITHOLOGIC INDIVIDUAL.

“The structural divisions shall be
units of cartography, and shall be
Their dis-
crimination shall be based upon the

designated formations.

local sequence of rocks, lines of

separation being drawn at those
points in the stratigraphic column

where lithologic characters change.”

Pursuing the description of
say further :

FAUNAL UNIT.

The faunal units shall be units of
correlation, and shall be designated
stages.t. Their discrimination shall
be based upon their fossil content,
lines of separation being determined
at those points in the stratigraphic

column where faunas change.

a faunal unit, or stage, we may

As a faunal unit is characterized by the life which

it contains, and as organisms are migratory, it is not to be
assumed that a stage is limited horizontally. It may be recog-
nized in diverse provinces or continents, and therefore the name
which a faunal unit receives should not be a local name, but
should be applied wherever that unit is recognized.

Distinctions between formation and stage — A formation is a set
of strata characterized by and limited to uniform constitution.
A stage is a set of strata containing and limited to a certain
fauna.

Lithologic constitution results from conditions which are
local and temporary, and which, though migratory, are rarely
more than provincial in extent. A fauna, though evolved in
adaptation to local conditions, may be capable of world-wide
migration. A formation, therefore, is geographically limited; a
fauna is not, necessarily.

Physical conditions which determine rock constitution are
But

Accordingly, a

recurrent and repeat the deposition of similar sediments.

organisms, once extinct, do not reappear.

«The term s¢age is here used provisionally, to afford a word for the purposes of
this discussion only.

568 BATEL VSVWVATETETS:

formation may belong to any geologic age; a fauna belongs
only to that age in which it was evolved and died out.

A principal object in studying formations is to read the
physical history of the earth, but the principal object of investi-
gating fossils is to get at the life history. Each is an essential
aid to the other, and the time relations of fossils are funda-
mental. But the two lines of inquiry should not be confused.
They are confused when we call a stage a formation, or vice
versa, or place the limitations of the one upon the other.

SIEPS IN ‘GEOLOGIC “SCIENCE:

The question with which this article opens may now be given
a fuller answer. The first step in the geologic study of any
locality is to ascertain the relations of the rock masses, and
those relations are most happily expressed in a map which
exhibits the distribution of formations. This map illustrates
local facts. It can be complete in itself, even though no con-
nection is established between the facts of that particular
district and those of the world at large, but if it be a correct
map it will fit into the general record when the connections are
tracede

The second step in geologic investigation is the detailed
study of faunas and their distribution in stages. To this second
step the first is an essential, as a good topographic map is to
both. Through close analysis of the faunas and comparative
study of their distribution, the data may be gathered for a map
of the stages represented in the district. This map of stages
may in some cases resemble closely that of the formations, but
in others there will be marked differences.

With the development of maps showing the distribution of
stages we may arrive at correlation of events, and thus be able
to compare physical conditions the world over, fitting into its
place with some degree of exactness the record of the forma-
tions for any locality. Thus correlation is the third step, a step
which may be facilitated through other lines of research, but
which is fundamental in broad studies of the earth’s history.

sittA TI GRAPHIC CLASSIFICATION 569

With a better knowledge of the physical geography of the
globe at successive epochs we may more hopefully attack the
great problems of mountain growth, continental development,
and earth dynamics. But our first work is to map the lithologic
individuals, while our associates, the paleontologists, distinguish

the faunal units of stratigraphy.
BaILey WILLIs.

THE DISCRIMINATION OF TIME-VALUES IN
GEOLOGY.

THE imperfection of the present systems of classification and
correlation of sedimentary rocks concerns more directly the
interpretation of the facts than the facts themselves.

The work of the geologists of the United States in mapping
and recording the stratigraphic sequence of formations was never
more exact and comprehensive. The paleontologist was never
more particular in his records of the faunal contents of each
formation and fossiliferous zone, and his comparisons were never
more full and precise. But the extension of knowledge over
vast territory has brought to light hundreds and thousands of
outcrops of the same formations, showing a diversity of faunal
composition which cannot be translated entirely into difference
in geologic age.

So long as surveys were confined to local areas separated by
spaces across which the continuity of formations could not be
traced, it was practicable to use a system of nomenclature and
classification in which lithologic formations and their strati-
graphic succession were chiefly considered. When, however,
the intervals between local areas were filled up and it was neces-
sary to correlate geological sections in which the formational
divisions are in part or wholly dissimilar, the duality of the
lithologic and biologic facts become apparent. These two
sets of facts are entirely different in nature and in origin, and
for their scientific discrimination duality of nomenclature is
essential.

The confusion of these two kinds of evidence was natural,
and has been perpetuated by the common practice of adopting
the lithologic formation as the unit of classification, making the
time divisions to apply strictly to the formations instead of to
the faunas and floras, by which alone the chronologic epochs in
which they were formed can be discriminated. This confusion

570

DISCRIMINATION OF TIME-VALUES IN GEOLOGY 57!

it seen in the discussion of classification and nomenclature in the
Tenth Annual Report of the United States Geological Survey, and
in the legends of the folio maps. For instance, take the Sewanee
Folio, Tennessee: The legend is as follows, viz.:

Walden sandstone - - |

Lookout sandstone - \ Garhoniterous

Bangor limestone - - |
Fort Payne chert - - |

Chattanooga black shale } Devonian.

Rockwood formation -
Chickamauga limestone $ SjJyrian.
Knox dolomite - - -

According to the rules in the Tenth Annual Report, the first

series of names are “structural divisions... . units of cartog-
raphy, and shall be designated formations” (p. 64).
The second series of names are ‘time divisions . . . . defined

primarily by paleontology and secondarily by structure, and
they shall be called periods” (p. 65).

Although everybody understands what is meant by the classi-
fication in the legend, the principle described in the rules is
wrong in that the legend on the map refers to a classification of
rocks: and the real fact in the case is that in the Sewanee quad-
rangle the Walden, Lookout, Bangor and Fort Payne formations
together constitute the Carboniferous system, and the map makes
no record of periods of time but only of formations of rocks.
The Devonian system of that quadrangle consists of the one
Chattanooga formation; and the Rockwood, Chickamauga, and
Knox formations are the only representatives of the Silurian
system recognized on the sheet.

The European nomenclature avoids this confusion by recog-
nizing a set of stratigraphic names and their categories; with a
corresponding set of categories for the chronologic classifica-
tion—the names of the divisions being the same in both the
stratigraphic and chronologic scales. Instead of referring all
stratigraphic divisions to one category (the formation), different

572 HENRY SHALER WILLIAMS

categories are used for formations of different relative size; mak-
ing the list of names to be group, system, series, stage, as adopted
by the international Congress. Each of these stratigraphic divi-
sions has its corresponding chronologic category, viz., era,
period, epoch, age. The European has no difficulty in express-
ing on the map, or in discussing, either the time or the structural
relations of the formation. On the map, the Walden, Lookout,
Bangor, and Fort Payne would be to him four series, together
constituting the Carboniferous system. By placing the Chat-
tanooga in the category of series he at once would indicate that
he does not regard the formation as necessarily representing the
whole Devonian system. On the other hand, when he speaks of
the Carboniferous period he is not discussing any local set of
formations but the total period of time in which lived a definite
set of plants and animals, only a few of which are discovered in
any one local formation. There can be no question that the
systems of the European geology can be recognized in this
country only by the fossils—but that does not change them from
formational aggregates into time divisions.

The implication in the Zenth Annual Keport that the divisions
which are discriminated by fossils must be chronologic, and not
structural, suggests the way in which our usage may be improved ;
but the fallacy of the principle is seen by noticing that the
smaller formations (the series and étages of the international
nomenclature), are to be discriminated by their fossil contents
as well as the larger ones (the systems). If discrimination
‘primarily by fossils” were to be the test as to whether the
division were structural or chronologic then formations would
become chronologic divisions in every newly surveyed area in
which actual lithologic continuity could not be traced to some
standard outcrop.

These two sets of facts (structural and paleontological) both
have to do with the classification of formations on a time basis ;
and those who are accustomed to frame their conceptions of
geological time on the basis of one set of facts, find difficulty in
even conceiving that there is any other basis.

DISCRIMINATION OF TIME-VALUES IN GEOLOGY 573

TIME VALUES OF FORMATIONS.

The regular sequence of stratified sediments forms a natural
geological column, which, in any particular section of the earth’s
crust, is so conspicuously subdivided by lithologic differences in
kind of sediments that the divisions form the most satisfactory
kind of natural time division for geological classification.

These natural, lithologic divisions of the crust of the earth
are technically called formations in the nomenclature of the
United States Geological Survey. And in any standard section,
such as that of New York state, the order, composition and thick-
ness of the several formations is exactly known; and for that
section, too, the fossils of each separate formation are known
accurately and in large numbers. Geologists have been accus-
tomed to use such a local column of known geological formations
as a standard time scale; and as examination has extended to
sections of the crust in other parts of the continent, the classifi-
cation and correlation of the other columns have been made to
correspond, by correlation, with such a standard column of
formations. Two methods have been used in establishing the
correlation : (1) by tracing continuity in the lithologic forma-
tion; and (2) by recognition of identity of the fossil species con-
tained in the formations. Both of these methods have rested
on an assumed interpretation of the facts, the correctness of
which may be questioned quite independently of the established
fact of continuity or of identity. |

The assumptions on which these interpretations rest are that
correlations of time relations can be established in the first case
by continuity of lithologic formation, and in the second case by
wdentity of fossils. In regard to the first case, it would be incor-
rect to say that the assumption is entirely false, for in some
cases, and to a limited extent, lithologic continuity of a forma-
tion is undoubtedly synonymous with sameness of the period of
the sedimentation represented by the formation. But the facts
are abundant, and well known to all field geologists, to prove
that formational continuity is not co-ordinate with lithologic
uniformity; and since our standard definition of a formation

574 HENRY SHALER WILLIAMS

(Tenth Annual Report, United States Geological Survey) is based
upon its lithologic uniformity, it is certain that in all cases in
which the lithologic changes affect the upper or lower limits of
a formation (which is expressed by thinning or thickening of
the formation) there must be discordance between the forma-
tional continuity and the time represented by it. It requires but
a moment’s reflection, further, to show that two sections in dif-
ferent regions may, on other evidence, be known to represent
the same interval of time but present no similarity lithologically;
this can receive only the one interpretation that formational dis-
continuity does represent time uniformity, which is the converse
of the original assumption.

In other words, while it is practicable in some cases to
assume that formations which are clearly continuous may be
deposited during the same period of time, it is clear that litho-
logic uniformity (by which the continuity of the formation is
recognized) is not a safe guide in making chronologic correla-
tions, however much value may be placed upon the lithologic
divisions of a standard geologic section, as natural divisions of
a geologic column made on a time basis. The tracing of time
equivalences by formational continuity is unsatisfactory, not
because of any failure on the part of a formation, as a litho-
logic unit, to represent a definite period of geologic time, but
because the time relations of the formation are not expressed
by any of the lithologic characters by which one formation is
distinguished from another. The confusion the geologist is apt
to fall into in discussing this point may be illustrated by the
measurement of the altitude of a rock outcropping on a moun-
tain side. The base of the Olean conglomerate, for instance, as
it appears at Olean Rock City may represent exactly the alti-
tude of 2,340 feet above the level of the sea (McKean County
Report, Second Pennsylvana Geological Survey, 8. 59), but its
altitude above the sea has no relationship whatever to any of
its lithological peculiarities. In forming an altitude scale, it
is in the region a conspicuous mark for the altitude at whicn
it lies, and if its dip. be ‘considered, the continuitysvolm the

DISCRIMINATION OF TIME-VALUES IN GEOLOGY. 575

conglomerate may be relied upon for estimates of approximate
altitude. Nevertheless, in itself the formation possesses no
altitude value, it is only a conspicuous stratum in the region
where it appears, that now lies at a definite elevation above the
sea—the amount of the elevation must be determined by other
means. It may indicate an altitude already determined, but it
has no intrinsic altitude value, and is not a measure of altitude.

In similar manner we may say that the geologic formation,
as a lithologic unit, was formed at some definite epoch of geo-
logic time which may be approximately measured by other
means; but the amount of time from any datum point to the
time when it was formed is not in any degree indicated by the
lithologic or structural characters of the formations; and in this
sense a formation may be strictly said to possess no intrinsic
time value, but to require the “ paleontologic evidence”’ to prove
to what ‘‘period” of time it belongs. This conclusion might
almost be drawn from the statements regarding periods in the
Tenth Annual Report (p. 65).

It is, however, not necessary to remind geologists that fossils
possess a value as means of determining the time relations of
formations. But this time value of fossils will be better appre-
ciated if attention be given to the nature of the evidence furn-
nished by fossils regarding the period of time in which they
lived.

TIME VALUE OF FOSSILS.

Fossils derive their time-value from the fact that the morphologic
characters presented by them are temporary in nature. The form of
a trilobite, expressed in its various morphologic characters, was
constructed by organisms at a definite period of time in the
history of the earth ; so that the presence of a fossil trilobite
imbedded in a rock formation is direct evidence of the geologic
age in which alone the trilobite lived. Hence it is that for the
whole surface of the earth fossils become marks of the time
divisions and’ the means of correlating formations on a time-
basis; this is the second method of correlation referred to above.
Identity of fossils is direct evidence of sameness of time. But

576 HENRY SHALER WILLIAMS

in applying such a statement much confusion and imperfect cor-
relation has arisen by failure to recognize wherein consists the
time-value of the fossil.

Identity of fossils may be wrongly interpreted —a trilobite
is a trilobite, it is true, but one trilobite is a Paradoxzides and
another is Phacops; and the one indicates an early period of
Paleozoic time, while the latter indicates a much later period.
Then, again, Phacops cristata appeared in the early part of the
life history of the genus Phacops, while Phacops rana is a later
species of the same genus. Thus it is evident that in speaking
of zdentity of fossils one may have in mind the same c/ass, the
same order, or the same family, genus, or species; and the time
value will differ according to the category of zodlogical classi-
fication to which the identity applies. The time represented by
the morphological character of the genus Paradoxides is much
shorter in length than that represented by the sub-class Trilobita ;
and the period of time represented by the characters of the
genus Paradoxides is diverse from that represented by the genus
Phacops. So in regard to all kinds of organisms represented by
fossils: the morphological characters expressive of the more
comprehensive zodlogical divisions are of greater antiquity than
those of less comprehensive divisions. The time-value of a
class character is, in most cases, as long as all the recorded time
of stratified rocks; while the time-value of a species is often
not over a tenth of that length.

In order to discuss these time relations it becomes necessary
to make scientific discrimination of the characters of organisms
in their time relations. First of all, what is it about the organ-
ism which is directly co-ordinate with time? An answer to this
question will be reached by considering the fundamental charac-
teristic of organisms—their growth. The form possessed by
any organism living is expressed only as the individual grows
or develops from an embryonic or formless state. The form is
acquired by each individual organism; always by individual
development requiring duration of time. This time is the living
period of the individual. If every individual developed some

DISCRIMINATION OF TIME-VALUES IN GEOLOGY 577

morphologic peculiarity by which it could be distinguished,
each such individual would represent a definite point of time,
and the succession of individuals thus distinguished would
express the history of the world in terms of the life periods of
the successive individuals. Practically it is impossible so to
distinguish the characters of individuals that they can serve as
time indicators.

The naturalist does, however, recognize characters of suffi-
cient distinctness to differentiate one species from another.
Specific characters, like individual characters, have been acquired
gradually as generation has succeeded generation in ordinary
reproduction. The process of acquiring specific characters is
called evolution; and specific evolution takes longer time than
individual development. The length of time during which in
any particular race generation continues without appreciable
disturbance of the specific characters of the offspring, varies in
different lines of descent, and specific characters are more
readily distinguished in one case than in another. But in all
cases the continuity of the specific characters without special
change represents a considerable period of geological time, and
this length of time is measured by the presence in the formations
of the same species of fossils. The period of the continuance
of the same species is therefore a definite length of time for
each species, since each species began at some definite time, and
(unless now living) became extinct at a later definite point of
time. This definiteness of the life-period of a species is inde-
pendent of our present knowledge either of its measure in years,
or of the thickness or stratigraphic position of the formation in
which the fossils are preserved.

Having noted that the morphologic characters of organisms
are temporary in nature, we may further observe that this tem-
porary quality has to do with the vitality of the organism in a
measurable way. As in the case of two living organisms we
consider the vigor of that one to be the greater, which, under
similar circumstances, lives the longer; so, in general, the length
of endurance of a species may be taken as the measure of some

578 HENRY SHALER WILLIAMS

kind of enduring power which inheres in the race itself. This
enduring power of organisms, expressed by the repetition of like
characters in successive fossil forms, is the time quality which
has already been used by geologists in making correlation of
formations by fossils, and to which we must look for the making
of a scientific time-scale.

In order to isolate this time quality I have proposed to speak
of it as the dzonzc quality or value of the organism. The bionic
guality of an organism may, then, be defined as its quality of continu-
ing, and repeating in successive generations, the same morphologic
characters. \n the case of the unit individual, it is the continuing
of the characters, since biologically the use of parts wastes them ;
and only as they are renovated may they be said to continue;
the active organs are here alone under consideration, for the
inactive hard parts are biologically dead parts whose endurance
is dependent only on absence of agencies of destruction. If,
now, we can discover some way of observing, recording, and
measuring the bionic values of fossils, we will be furnished with
a means of constructing a geological time-scale on a separate
(not to say independent) basis from the supposed time-scale
represented by the geologic column of successive formations.

In the development of such a time-scale the first point to
notice is that the characters of organisms differ in their bionic
value in direct proportion to theiy taxonomic rank. Thus, as we
have seen in the case of the Trilobites, the characters which have
ordinal or sub-class rank have persisted in the history of organ-
isms vastly longer than the characters of generic value; and
these are of greater bionic value than specific characters.

In discriminating periods in geological time we may look first
to the well-known categories of zodlogy and botany, as a basis
of determination of the order of rank in the time divisions, viz.,
the time of endurance of a specific character (or, concretely, the
length of time represented by the presence of the same species in
successive strata of the rocks) is of subordinate rank to the time
of endurance of a generic character. And if we should adopt
the name chron to apply to geological time-units in general,

DISCRIMINATION OF TIME-VALUES IN GEOLOGY 579

and dzochron to the units whose measure is the endurance of
organic characters, we have a means of constructing a system of
nomenclature which will express what is now known of geological
time relations, and (more important still), which will serve as an
aid in accumulating the necessary statistics to perfect the geo-
logical time-scale.

ORDER OF MAGNITUDE OF BIONIC UNITS.

In expanding this system of nomenclature the following
table will indicate the principle upon which the fundamental
units of time value will be discriminated and named. The time
unit of lowest rank will be based upon the life endurance of an
individual organism; the amount of organic vigor expressed by
the preservation of the individual life constitutes a bionic unit of
simplest or lowest rank; the individual, therefore, is an organic
unit of monobionic rank. How many individual lives are possi-
ble in the life-history of a species we at present do not know,
but we do know that the bionic value of the species (or, strictly
speaking, of specific characters) is of an entirely higher order
than that of the individual. To be more concrete the individual,
the species, the genus, etc., constitute organic units of consecu-
tively higher and higher order of bionic magnitude, which state-
ment may be tabulated in the following way:

THE BIONIC VALUES OF THE SEVERAL CATEGORIES OF CLASSIFI-
CATION OF ORGANISMS.

Individual - - a monobionic unit.
Species - - a diobionic unit.
Genus - - a tribionic unit.
Family - - a tetrabionic unit,
Order - - a pentabionic unit.
Class - - a sexbionic unit.

In the table the bionic order of magnitude is expressed by
the prefix mono-, di-, tri-, etc.; thus the dibionic unit is an abstract
mode of naming the order of magnitude of the organic force
expressed in preserving, by generation, the specific characters of
an organism. Comparing this with the force which preserves in
like manner generic characters, the former is seen to be of second

580 HENRY SHALER WILLIAMS

order of magnitude, while the latter is of tertiary order of mag-
nitude. So far the magnitudes are relative, and they become
concrete only when the bionic value of the characters of some
particular species or genus is considered.

Paleontologists are familiar with the very long range of the
species Atrypa reticularis ; Rhynchonella cuboides, on the other hand,
has avery short range. Inthe nomenclature proposed (so long
as both are considered to be species), the fact would be expressed
by saying that the Atrypa reticularis biochron is longer than the
Rhynchonella cuboides biochron. Nevertheless, in the familiar
categories used by the zodlogist and botanist is found a means of
expressing definiteness in, at least, the order of magnitude of
biochrons ; whereas there is no way of distinguishing the order of
magnitude of geochrons, except in feet thickness. In the case
of the biochron it is only necessary to indicate the name of the
species or genus in order to fix a definite value to the biochron.
Such values are already definitely expressed when we speak of
the “reptilian age, the “age of fishes,” the “‘olenellus zomem
The definiteness is indicated by the name of the particular group
of organisms made use of, or adopted as the measure of the

biochron.
THE DUAL NOMENCLATURE.

In proposing a dual nomenclature it is essential to indicate
this basis of measurement of the chron, and to distinguish the
geochron (expressed in terms of feet thickness of stratified sedi-
ments of uniform lithologic constitution) from the dzochron
(expressed in terms of presence in the sediments of fossils of
the same species, genus, or family).

Thus the time value of the Hamilton formation would be
spoken of as the Hamilton geochron; while the time value of the
species 7vopidoleptus carinatus would be the Tyopidoleptus biochron.
The same kind of difference in values is observed to pertain to
both biochrons and geochrons. As it is impossible to fix any
standard length for the endurance of a species, so it is impossible
to fix any standard of thickness for geological formations; they
may vary from a few inches to many hundreds of feet in thickness.

DISCRIMINATION OF TIME-VALUES IN GEOLOGY 581

Since the thickness, as well as the kind of sediment, varies
with the geographical locality, the association of a geographical
name with the lithologic character of the formation becomes a
definite and precise mode of identifying a formation; so that
the name ‘Medina sandstone”? becomes a definite formation
name, the characteristics of which may be observed, and their
definition fully elaborated at Medina, where the formation
appears with its typical characters. Its place ina geological col-
umn, its thickness, and its composition, are all made definite by
the name Medina; any sandstone outcropping elsewhere can be
classed as ‘‘ Medina sandstone” only by possessing the charac-
ters considered to be essential to formational continuity and
integrity. Of course one of those characters may be theoreti-
cally defined as the geologic time of its deposition; this may
be indicated by the contained fossils. But formational continuity
and identity may be established in the case of non-identity of
fossils, and in one section the fossils may rise to a higher strati-
graphic plane than in another; so that actual formational continuity
may disagree with the evidence of duration presented by the fossil spe-
cies. This fact is illustrated in the case of the Catskill formation,
which is known to occupy a stratigraphic position in eastern
New York and Pennsylvania continuous with rocks further west,
possessing both different lithologic characters and containing
fossils regarded as characteristic of the Chemung formation.
Such facts find easy expression when the nomenclature is fur-
nished us by which to separate a geochron from a biochron.
The Sfirifer disjunctus biochron is a different thing from the
Chemung geochron. In defining the Chemung formation it has
become necessary to distinguish it from the formation following
it. This following formation in many sections in New York and
Pennsylvania is a red sandstone and shale lacking the marine
fossils, sometimes holding fish of possibly marine, but probably
brackish-water habitat. For purposes of mapping and classifica-
cation, the Chemung formation is succeeded by the Catskill
formation; but the evidence is conclusive that the time of depo-
sition of the red sediments of the Catskill formation of eastern

582 HENRY SHALER WILLIAMS

New York was in large measure synchronous with the sedimen-
tation of gray rocks with abundant marine fossils, including
Spirifer aisjunctus, in western New York and Pennsylvania.

If the endurance of the Spurifer disjunctus fauna be made the
mark of the Sperifer disjunctus biochron, it becomes quite possible
to state and discuss the facts as they are, and to understand the
natural explanation of the facts as a gradual transgression of
the shore from which the red sediments were deposited west-
ward with the progress of time. The geochron of red beds
began earlier in the scale of biochrons, and as we expand the
idea we will discover that it is also earlier in actual geological
time, and that the biochron is the measure upon which we must
depend in constructing the standard geological time-scale.

THE CONSTRUCTION OF A BIONIC TIME SCALE.

The next question arises, how shall this time-scale be con-
structed? First, it must be constructed, primarily, on the basis
of the bionic values of the fossils. Secondly, the names of the
time divisions must be distinct from those used in designating
the formational divisions of the rocks. Thirdly, biological names
should be chosen for the divisions, and so far as practicable,
some abundant or characteristic fossil should furnish the name;
as in naming formations the locality in which the formation
appears in full force and with full characteristics is used.
Fourthly, it cannot be expected that the divisions of an accu-
rate time-scale, based upon many different sources of measure-
ment of the time intervals, will present a continuous series
without breaks and without overlapping of the divisions. If a
single means of measuring the lapse of time were used, such
breaks and lapping might be avoided; only in case the begin-
ning of one species or fauna was everywhere dependent upon
the cessation of the preceding one, could the scale be made
without liability of failure to meet at the limits of the time
intervals indicated. Fifthly, the classification and definition of
geological formations should be made, so far as practicable,
independent of the fossils or the time relations indicated by

DISCRIMINATION OF TIME-VALUES IN GEOLOGY 583

them; and the fossils should be used (as now their use is scien-
tifically most valuable) in establishing correlations, not in fur-
nishing definitions. Sixthly, the selection of the grander time
divisions should be made to conform, so far as found practicable,
with the standard divisions of the formation scale; but their
discrimination should be based strictly upon fossil evidence and
not upon the lithologic or stratigraphic characters of formations.
Seventhly, in forming the time-scale the continuance of the
fauna in its integrity, as marked by the dominance of its
characteristic species, will constitute a more satisfactory basis
of discrimination than the species alone. Slight differences in
the aggregate of species found in the successive strata of rocks
is quite consistent with the continuance of the fauna in its integ-
rity, and to distinguish the fauna as a whole from its local and
temporary expression the latter may appropriately be called a
faunule.

In constructing a time-scale on the basis of the bionic values
of fossils, it is practicable to give to the categories now in use
greater precision and scientific definition. If geologists chose
to adopt the following terms in the restricted sense designated,
such a set of terms would prove of great value.

TERMS OF THE BIONIC TIME SCALE.

Chron.—An indefinite division of geological time.

Geochron.—The time equivalent of a formation.

Biochron.—The time equivalent of a fauna or flora.

Flemera.—The technical name for a monobiochron, indicated by the
preservation of the individual characteristics of all the species of a local
faunule, as shown by the association in the rocks of the same species, in the
same relative abundance, size, and vigor. An example is the hemera of
Rhynchonella (Hypothyris) cubotdes.

Epoch.—The name of a dibiochron, indicating the time equivalent of the
endurance of a particular species and of the integrity of the fauna of which it
is the dominant characteristic. An example is the 77ofidoleptus carinatus
epoch, which corresponds closely to the limits of the Hamilton formation of
eastern New York.

Pertod.— May be defined as a tribiochron. This is the time equivalent of
the continuance of a gemus. An example is the ?’aradoxides period, which
corresponds to the Acadian formation of the Cambrian system.

584 HENRY SHALER WILLIAMS

Era.— May be used to indicate a tetrabiochron; and Olenide era would
indicate the life range of the family OJ/enzdae, corresponding in length,
approximately, to the geochron of the Cambrian system, though not
strictly so.

Eon.— May stand as the name for a pentabiochron; an example of which
is the 77zlobzte eon, the time equivalent of the continuance of the order, or
sub-class, 77z/obzta, which closely approximates the length of the Paleozoic
geochron.

CLASSIFICATION AND NOMENCLATURE OF THE TRILOBITE EON
(PALEOZOIC) CONSTRUCTED ON DHE BASIS. OF DHET BIONIC
VALUES OF FOSSILS.

Formational Equivalents

Eon Periods Epochs (Approximate)
( Cameratus - ~- Coal measures.
Rare Increbescens - Kaskaskia, St. Louis.
je leullly sien Logani - - - Keokuk, Burlington.
Marionensis - Kinderhook.
Disjunctus - - Chemung.
Mucronatus- - Hamilton.
6. Phacopsian Acuminatus- - Corniferous.
Trilobiteon Arenosus - - Oriskany.
l Macropleurus - Lower Helderberg.
Vanuxemi - - Waterlime, etc.
Be (Callyeronieat Radiatus - - Niagara, etc.
4. Asaphian P Ordovician.
3. Olenian |
2. Paradoxidean i Cambrian.
1. Olenellian J

Applying the principles set forth in this paper, a tentative
table may be constructed which will express concretely what is
meant by a biochronic classification and nomenclature. In this
table the attempt is made to find names of genera of Trilobites
for the periods of that eon in which Trilobites constitute a
characteristic feature. The genera chosen actually do lap over
each other in several places, but it is doubtful if any genera
could be chosen which, if they did not lap over in their chrono-
logic range, would leave actual gaps not represented in any divi-
sion of the scale. The genus Sfivifer is selected in the upper

DISCRIMINATION OR TIME-VALUES IN GEOLOGY 585

divisions of the scale, as furnishing a convenient set of diag-
nostic species; they, too, do not in every case avoid lapping, a
difficulty which it has already been said is unavoidable when the
means of measurement are multiple and not mutually exclusive.
The foregoing table is offered, then, as a means of illustrating
the proposed plan.

HENRY SHALER WILLIAMS.
NEW HAVEN, Conn,

VARTATIONS OR SU EXGURE sine CREA EN ise eeAsas
IGNEOUS, KOCKS OF THE GREAT BASING

INTRODUCTION.

DurinG the field work of the writer in the Great Basin, in
1899, he found in a number of localities, chiefly near the edge
of the Sierras, cases of what seemed at first extraordinary transi-
tions or intimate alternations of texture in the Tertiary igneous
rocks, which are for the most part extrusive. Microscopic and
comparative examination in the office corroborated the field
conclusions and showed also that the transitions were not as
abrupt as they seemed at first, but were more gradual ; and the
study of the conditions of crystallization which may be inferred
from the structure of the different varieties shows that similar
conditions must exist in many other places and similar transi-
tions may be looked for. It is true that there are in that por-
tion of the Great Basin especiaily under consideration (namely,
the district lying within fifty miles of Carson) exceptionally
favorable circumstances for the exposure of both the surface
portions and the originally deeply buried portions of lavas.
The region is an arid one, and therefore the general erosion is
slight; nevertheless, waters derived from the moister Sierras
reach out into this region in the form of streams, and have
accomplished much special or basal? erosion. There are also a
number of lakes, which in former times were much more exten-
sive, and probably existed in one stage or another since early
Tertiary times; and the basal erosion of these lakes has proba
bly been considerable.

Some of the special localities where the observations were
made will now be described in detail.

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

27. e., Erosion which works at the base of topographic features, undercutting
them.
586

VARIATIONS OF TEXTURE IN [TGNEOUS ROCKS 587

ANDESITIC ROCKS:
MASON’S BUTTE.

Field description.—Mason’s Butte lies close to Walker River,
in Mason Valley, about four miles due south from Wabuska. It
is thus situated midway between the northern end of the Walker
River range and the northern end of the Smith Valley and
Pinenut ranges. The rocks of the butte are related to those of
all these ranges, as will be shown later. The butte itself is
about a mile and a half long in a northeasterly direction, and
about half a mile wide. It presents from a little distance the
appearance of typical volcanic rock, being distinctly thinly
bedded, with red and gray zones. On the western face of the
butte is a scarp two or three hundred feet high, and from here
easterly there are a series of saw teeth, caused by the unequal
erosion of the bands of which it is composed; then the butte
sinks gradually into the plain again. The bands of igneous
rock dip easterly 15° at the western end of the butte, and the
dip increases to the east, so that on the eastern end they dip
30°. Here they are locally reversed and dip west, probably
from movement subsequent to the eruption, which movement is
also evidenced by shearing. The beds strike parallel with the

longest extension of the butte (see Fig. 1).

Fic. 1.—Cross section of Mason Butte, showing alternations of course and fine
textured dioritic beds. Drawn to scale. Scale, 1 inch=850 feet.

Upon examination, the rocks are found to be diorite and
andesite in alternating conformable layers. Fourteen different
layers of diorite were found in the half-mile section, with layers
of diorite porphyry and andesite between. From a structural
point of view, all these rocks grade into one another. Some-
times the gradation may be actually seen in a single bed; for
example, a highly porphyritic andesite, which in the field was

588 J. E. SPURR

taken for a diorite porphyry, was found to change gradually
into a coarsely crystalline diorite; but generally the beds are
separate. The coarse-grained rock contains some dark inclu-
sions, which appear to be like the finer-grained rocks. The
diorite is also associated with alaskite* which occurs in many
large segregated masses irregularly distributed, and with a few
quartz veins, also segregational. The alaskite and quartz veins
are found only in the diorite; never in the fine-grained rocks.
At the eastern end of the butte, also associated with the diorite,
are considerable masses of hornblendite which grade into normal
diorite.

Although a careful search was made, no intrusive phenomena
showing that the andesites are intrusive into the interbedded
diorites, or vice versa, were found. Although the whole butte
is of rugged rock and entirely free from any vegetation sufficient
to obscure exposures, yet the succession of the different beds is
as normal and regular as in the case of sediments. The ande-
sites are evidently similar to the Tertiary andesites which occur
plentifully in the whole region round about. The conclusion
reached in the field, therefore, was that Mason’s Butte repre-
sents the roots of old volcanic flows, that the apparent bedding
is a flow structure on a large scale, and that the coarse-grained
and fine-grained rocks, the diorites and andesites, are different
forms of crystallization from a single magma. Some of the
fine-grained rocks are separate layers from the coarse-grained
ones, while some are simply variations in them. The diorite
naturally incloses portions of an earlier formed crust, which is
identical with the finer-grained beds.

Microscopic evidence.—The structure and composition of
typical specimens from this butte will now be described:

Hornblende-biotite-quartz-diorite? (117 N.).—This rock in the
hand specimen is medium coarse granular and rather dark, on

See Am. Geol., April, 1900, p. 230. Alaskite is proposed as a general name for
rocks consisting of quartz and alkali feldspars without essential ferromagnesian con-
stituents.

? These numbers refer to the specimens in the writer’s collection, and are given
for the purposes of subsequent identification.

VARIATIONS OF TEXTURE IN IGNEOUS ROCKS 589

account of a liberal percentage of ferromagnesian minerals. It
has the appearance of a typical granular rock. Under the
microscope the structure is coarse. The largest grains are as
much as 8™™ in diameter, and these grains, which have hypidio-
morphic or idiomorphic outlines, are closely intergrown. Filling
the spaces left by the intergrowth of these larger grains is a
somewhat scanty mesostasis, composed of grains averaging
about 1™™ in diameter. The minerals of the larger and the
smaller grains are the same, consisting of feldspar, which is
greatly in excess, and quartz, green hornblende, and biotite.
The quartz is distinctly subordinate in amount to the feldspar.
The feldspars were tested by the Fouqué method and gave, on
sections perpendicular to both the positive and the negative
bisectrices, the extinction angles for oligoclase.

Hornblende-bhiotite-quartz-diorite (123 N.).—This does not differ
in the hand specimen to any noticeable extent from 117 N.
Under the microscope also the two are essentially alike, but with
auslight ditterence, In 123 N. the larger crystals are more
abundant, so that the structure at first sight appears to be coarse
allotriomorphic granular. On closer analysis the section is seen
to be made up of more or less idiomorphic crystals of feldspar,
green hornblende, and biotite, closely packed together, with the
spaces between filled with a scanty mesostasis of quartz, feldspar,
hornblende, and biotite, these grains being of all sizes, com-
mencing with the size of the idiomorphic grains just described.
In this section, as in 117 N., the quartz is a subordinate essen-
tial. The feldspar, tested twice by the Fouqué method, gives
the angles for andesine-oligoclase.

Hornblende-andesite (116 N.).—This is a porphyritic rock with
an abundant, greenish groundmass, and is entirely similar in
appearance to the hornblende-andesites of the Pinenut range,
just west of the butte. Under the microscope the groundmass
is found to be holocrystalline and microgranular, consisting
chiefly of allotriomorphic feldspar and hornblende. The pheno-
crysts reach a diameter of about 2.75™™, and consist of feldspar
and green hornblende. The feldspars, when tested, gave twice

590 J. E. SPURR

the extinction angles of andesine. In this section the pheno-
crysts are less in amount than the groundmass.

Flornblende-andesite (118 N.).—In the hand specimen this has
the appearance of a diorite closely verging upon diorite por-
phyry. Under the microscope, however, there is found to be,
between the closely packed phenocrysts, a micro-granular
groundmass, slightly finer grained than that of 116 N._ It con-
sists chiefly of feldspar with some hornblende. The phenocrysts
also are chiefly feldspar, with subordinate hornblende, which is
largely altered to epidote. Determinations of the feldspar show
that it is oligoclase. The larger phenocrysts have an average
diameter of about 27".

Flornblende-quartz-andesite (121 N.).—This rock is similar in
appearance to 118 N. It has also a fine holocrystalline granular
groundmass, showing no fluxional arrangement and composed of
feldspar and quartz. The phenocrysts have a maximum diameter
of about 3™™, and consist of feldspar, green hornblende, and
subordinate quartz, the last named in large rounded or corroded
crystals. The hornblende is largely altered to epidote. The
feldspar, tested twice by the Fouqué method, shows the extinc-
tion angles of andesine and andesine-oligoclase.

Ffornblende-quartz-andesite (120 N.).—This rock in the hand
specimen is dense, greenish and fine grained, having a trap-like
appearance, with small phenocrysts in an aphanitic groundmass.
Under the microscope the rock is seen to be much like those
just described. The groundmass is fine holocrystalline, consist-
ing of lath-shaped feldspars and hornblende, with subordinate
quartz and some pyrite and siderite. There is a slight fluxional
arrangement. The phenocrysts belong to a distinct generation
from the groundmass (7. ¢., there are no transitions) and consist
of feldspar, pale green hornblende, and a single large grain of
corroded quartz. The phenocrysts reach a diameter of 2™™.
The hornblende is largely altered to calcite, chlorite, and
epidote.

FTornblende-andesite (119 N.).—This is from the same bed as
120 N., and is simply a variation from it. In appearance it is

VARIATIONS-OF TEXTURE IN IGNEOUS KOCKS -591

somewhat different, on account of the more abundant con-
spicuous feldspar phenocrysts, which are, however, very small,
so that the rock resembles a miniature of 121 N., reduced three
or four times. Under the microscope the groundmass is seen to
have been originally glassy, but now is devitrified, and has a
cryptocrystalline structure, The phenocrysts attain a diameter
of about 1™™, and consist of feldspar and hornblende. The
feldspar, upon optical examination, proves to be oligoclase.
The hornblende is mostly altered to calcite and chlorite.

Alaskite (114 N.).—This in the hand specimen is a typical,
rather fine-grained, dense rock. Under the microscope the
structure is hypidiomorphic granular. The grains average 1™™
in diameter, and consist of quartz, orthoclase, and microcline in
somewhat equal proportions. The only dark minerals present
are a few grains of chlorite and pyrite.

Flornblendite (124 N.).—This is a coarse-grained rock, typical
of its kind, and also consists of green hornblende, with blotches
o1eepidote. . Under the mucroscope the structure is seen to be
allotriomorphic. No other minerals are present save those men-
tioned, and the epidote is secondary to the hornblende. In the
section studied the grains average about 3™™ in diameter.

Analysis of structure.—The series of rocks just described,
from 117 N. to 119 N., inclusive, shows a transition, which has
at one end a typical coarse granular rock and at the other a por-
phyritic rock with a glassy groundmass. In the intermediate
stages the phenocrysts increase in numbers and somewhat in
size, while the groundmass shrinks in volume and becomes also
coarser. In 123 N., which is a typical granular rock only
slightly removed from 117 N., these phenocrysts have increased
until they nearly fill the rock, while the small portion of ground-
mass which remains has coarsened so that the grains are a
millimeter in diameter; thus the rock at first sight appears a
typical allotriomorphic granular rock, from which it is moreover
not far removed.

The very slight variations of structure between any two adja-
cent members of this transition series shows that the differences

592 Woo PR SI AGES

in condition which brought about the separation of the. coarse-
grained from the fine-grained rocks were relatively small.
These differences are plainly longer cooling periods (and hence
longer crystallization intervals) in some portions of the rock, as
compared with others. The coarse granular rocks, judging from
their almost complete and uniform crystallization, have crystal-
lized entirely in their present position; the segregation from
them of alaskite on the one hand, and hornblendite on the other,
is additional evidence of this. The porphyritic forms, however,
show generally two distinct generations, and this, with the fact
that some of the phenocrysts of the first generation, especially
the quartz, show corrosion and resorbtion, indicates a change of
conditions; also the slight fluxional arrangement, sometimes
observed, shows at least local movement. Yet the contempo-
raneity of the fine-grained porphyritic rocks with the granu-
lar forms, shows that the crystallization break registered by
the former did not arise from any important change of posi-
tion.

Composition of rocks. —The general mineral composition of
rocks of all structures (leaving out the alaskite and hornblen-
dite, which are plainly segregation products from the hornblende
quartz-diorite), is the same, being a hornblende-quartz-diorite or
quartz-andesite. In the andesites the quartz, where present, was
plainly in process of resorbtion by the groundmass at the time
of solidification, so that it is probable (from the fact the horn-
blende andesites and the hornblende-quartz-andesites often occur
as variations of a single bed) that in those andesites which do
not show free quartz the composition is the same as where they
do, the quartz belonging to the first period of crystallization
having been entirely resorbed.

VARIATIONS OF TEXTURE IN IGNEOUS ROCKS 593

Analyses of the rocks are as follows (analyst, Dr. H. N.
Stokes):

No. 1, No. 117 N. No. 2, No. r20 N.
Hornblende-biotite- Hornblende-quartz-
quartz-diorite andesite
SiO, 60.25 SiO >, eee)
Al,O. 17.90 Al,O, 10.57
Fe,O% 3.08 HeLOn 3.84
FeO 2.44 FeO 2.45
MgO 2.44 MgO 5.79
CaO 5-57 CaO 6.30
Na,O 4.29 Na,O 3.40
K,0 1.89 K,O 2.55
IB IROE = .20 H,O-+ .39
H,O T24 IBEX@) 2.33
AUTO). .65 AUO}s 86
CO; none CO, 1.61
RsOe 25 P,O; 2.
MnO .06 MnO .08
SrO .07 SrO trace
100.33 99.83

ineNoswtherationol Na,© KO; C20. ita. In No,
2 the same ratio is 1:1.13. These ratios are calculated from
the quotient figures obtained by dividing the percentages of
each by the molecular weights. Similarly in No. 1 the ratio of
i OrCaQrser 4,05, in No. 2ethessame rations I) a4aie.

No. 2 contains less quartz and more hornblende than No. 1,
hence the decrease in silica and the increase in magnesia ; other-
wise the two rocks are the same.

Correlation of rocks.—As before stated, just west of Mason’s
Butte lies the northern end of the Smith Valley and Pinenut
ranges. On crossing from Waubuska through Churchill Can-
yon, andesites were found which were.recognized in the field as
similar to those in the butte. The bedding of these rocks is
nearly horizontal. They are fine-grained hornblende-andesites,
often with fine holocrystalline groundmass; occasionally they
vary to fine-grained diorite porphry. In higher portions of the
mountains (these relatively coarse-grained lavas are exposed in
the lowest cuts) the lavas are dacites and pyroxene andesites.

Andesites form the main mass of the Pinenut range at its
northern end, which is separated from the Washoe district at

594 J. E. SPURR

the southern end of the Virginia range by a comparatively nar-
row valley. The lavas of one range are evidently continuous
with those of the other.* In the Washoe district have been
found eruptions of hornblende-mica-andesite and quartz-ande-
site or dacite, corresponding exactly to that at Mason Butte.
At Washoe, also, the andesite becomes, under favorable condi-
tions, coarsely crystalline, and Messrs. Hague and Iddings ?
have noted that upon this complete crystallization quartz sepa-
rates out, producing a mica-quartz-diorite.

RHYOLITIC ROCKS.

TRANSITIONS IN TEXTURE OF THE BASAL RHYOLITES OF THE
PINENUT RANGE. ;

Field description —TVhe central core of the Pinenut range is
made up of granitic rocks. These were examined in two locali-
ties: one southeast from Dayton, and one west from Wellington.

At the first mentioned locality, granitic rocks are exposed
along an easterly facing scarp which is at the north end of Smith
Valley. The rocks are all granular, so far as observed, but
vary much in texture from fine to coarse; they are often por-
phyritic. They show a distinct banding, resembling at a dis-
tance rude stratification; this banding is due to a zonal arrange-
ment of the rocks of different textures.

In the district west from Wellington, rhyolitie and granitic
rocks are exposed in the spur just east from the Mountain
House, and: are jof “exceptional interest; dilere, jbemeathy the
andesitic rocks which cover the mountain slopes, is found a
highly indurated volcanic conglomerate and sandstone, appa-
rently waterlaid, and consisting entirely of rhyolitic material.
A short distance farther on, the rocks from which these detri-
tals are derived were found in place. These original rocks show
great variations, passing from a fine-grained, almost aphanitic,

™See “The Succession and Relation of Tertiary Igneous Rocks in the Great
Basin Region,’ Jour. GEOL., Vol. VIII, p. 621.

2Bull. 17, U. S. Geol. Surv.

VARIATIONS OF TEXTURE IN IGNEOUS ROCKS 595

rhyolite, to coarse siliceous granite and alaskite. The number-
less variations are within a few feet of one another and are
arranged in bands, recalling immediately the similar phenome-
non in the andesitic rocks of Mason Butte. In the case of the
rhyolitic rocks also there is no sign of intrusion of one into
another, nor in general does there appear to be any marked
gradations between the different bands at contact, the bound-
aries between them being fairly distinct.

It is clear in the field that the variations are chiefly textural
and that the composition of all the varieties is nearly the same
— that of siliceous granite or rhyolite.

In this case again, we have rocks which appear to represent
the roots of old volcanics, being intermediate between com-
pletely massive plutonic igneous rocks and superficial fine-
grained volcanics. They must have suffered a _ flowage
resulting in the formation of this peculiar streaky structure,
while the great variations in texture in the different bands show
crystallization at points still far removed from the surface.

Of the different rock varieties in the locality west from Wel-
lington the following will be briefly described.

Biotite-rhyolite (164 N*.).—Structure porphyritic; groundmass
cryptocrystalline, probably devitrified glass. The phenocrysts
are of all sizes, the larger ones grading down to those which
vanish in the groundmass. They are of feldspar, quartz, and
biotite, the latter decomposed. The feldspar was determined to
be largely albite and oligoclase-albite, although there is some
orthoclase, The largest phenocryst of feldspar measures
237.=> in diameter; the largest one of the quartz,1%4™™. The
rock contains angular fragments of finer grain. Some of these
seem to be devitrified rhyolitic glass, while others are fragments
of more basic lava, probably andesite — these latter show small
lath-shaped feldspar phenocrysts in a glassy semi-devitrified
groundmass.

Liotite-rhyolite (165 N*.).—In this specimen the groundmass
becomes slightly coarser than in the preceding specimen and is
very fine microgranular. It also becomes more scant than in

596 V5 1 SI AGIRI

164 N?, on account of the multiplication of phenocrysts, which
show the same great variety in point of size as the rock just
described. The same angular fragments of more basic lava also
occur. The phenocrysts are quartz, orthoclase, and a striated
feldspar; the latter tested twice by the Fouqué method proves
to be albite. The orthoclase was optically determined as
such.

Rhoylite (162 N?.).—In this rock the groundmass is fine holo-
crystalline, coarse enough to enable one to distinguish the
mosaic of quartz and feldspar. This groundmass contains the
same fragments of more basic lava that have already been
described, and also encloses broken phenocrysts of mostly
unstriated feldspar.

Granite-porphyry, fine-grained (169 N.).—In this rock the
groundmass is fine granular, sometimes granophyric, and con-
rists chiefly of quartz and orthoclase. The phenocrysts are
abundant and consist of quartz and feldspar, with chlorite and
epidote which are derived from the decomposition of ferromag-
nesian minerals. The feldspar phenocrysts are partly orthoclase
and partly a striated feldspar, which, determined by the Foqué
method, proves to be oligoclase-albite.

Biotite-granite, medium-grained (171 N.).—In this rock the
grains are of two distincts sizes, one many times larger than the
other. The largér grains have a tendency to idiomorphism, the
smaller grains to allotriomorphism. The smaller grains are-
included between the interlocking larger ones and may be con-
sidered as forming an overgrown groundmass, partly crowded
out by the multiplication and joining of phenocrysts, which are
represented by the larger grains. The minerals of the rock are
quartz, feldspar, biotite, and magnetite. The feldspar is almost
entirely orthoclase, with some microcline and albite.

Granite, coarse (175 N.).—The structure of this rock is like
that of the preceding, only coarser. It contains many perfect
idiomorphic crystals of feldspar, often touching and almost inter-
locking, and smaller crystals of bleaching biotite and ragged pale
green hornblende, the last perhaps secondary. These minerals

VARIATIONS OF TEXTURE IN IGNEOUS ROCKS 597

are in general cemented by a mesostasis of coarse allotrio-
morphic quartz, which often includes or is intergrown with, in
poikilitic fashion, smaller crystals of feldspar, pale green horn-
blende, and sphene. The feldspar included in the quartz sinks
to very small dimensions, whereas the ordinary feldspar grain is
very large. The structure may be regarded as the coarsening
of the porphyritic structure, or at least closely related to it by
reason of the two generations. The large feldspar crystals are
partly orthoclase, but are chiefly finely striated. Optical deter-
minations of the striated crystals show microcline-anorthoclase
and albite.

Granite, coarse (172 N.).—This rock is almost entirely like
175 N, and has a good deal of the peculiar structure of this
rock, but in general is more hypidiomorphic granular or truly
granitic.

Analysts of the structures of the granite-rhyolites. —TVhe analysis
of the structure of the granites. and rhyolites just described helps
toward a better understanding of their relation.

164 N (a). Here are phenocrysts of all sizes, gradually
shrinking in size to the glassy (sometimes slightly devitrified )
groundmass ; 2. é., the crystallization, instead of belonging to one
or two distinct generations, represents many generations, not
separable from one another. This is a proof of gradual and
equable hardening. It shows that the viscosity increased very
slowly and regularly to the point of final complete solidification,
the newer crystals having progressively smaller fields of crystal-
lization.

165 N(a). This is like 164 N(a) except that the groundmass
diminishes, on account of multiplication of phenocrysts. This
marks a longer cooling period than 164 N (a), so long as almost
to permit of total crystallization as relatively large crystals.

162 N(a). In this section the feldspar phenocrysts are not
connected by gradual transitions with the groundmass, which is
slightly coarser than that of 165 N(a) and is much more abun-
dant. We have, therefore, two distinct generations of crystalli-
zation, and this, together with the frequently broken character

598 J. E. SPURR

of the phenocrysts, indicates a break in the crystallization, evi-
dently resultant from a movement of the solidifying mass. The
order of events in this section was, therefore, (1) comparatively
slow crystallization of feldspar; (2) flowage, producing a change
of conditions; (3) medium rapid cooling, bringing about the
uniform moderately fine crystallization of the rest of the rock.

169 N. Here the line between the abundant phenocrysts and
the fine-grained groundmass is, in general, distinct, for although
there are transitions between the two they are not so abundant
as in rocks like 164 N(a) and 165 N(a). This connotes a
shorter period of first crystallization (when the phenocrysts were
formed) than does 165 N(a), then a more rapid cooling than
165 N(a) to acertain point, then a slower rate of crystallization,
permitting the formation of the uniform fine granular structure.

171 N. The structure of this connotes a long period of slight
viscosity, during which the crystals of quartz and feldspar could
grow until they touched and sometimes interlocked. The differ-
ence in size between these crystals and the grains of the ground-
mass or mesostasis which fills the space between them, implies
a slight break or change of conditions, after which was again a
comparatively slow uniform crystallization of the rest of the
rock, producing an even allotriomorphic granular structure.
There are then two distinct generations of crystals. The result-
ing structure is entirely similar to the ophitic structure of diaba-
ses, save that in these siliceous feldspars the forms are not so
elongated, and so the structure is not so striking. The struc-
ture of this specimen, however, differs from the typical granitic
structure in the same way that a diabase differs in structure
from a’ gabbro. A longer period for the first crystallization,
reducing the mesostasis to a still smaller percentage, would give
the aplitic structure, where the idiomorphic crystals are predom-
inant and occupy the greater portion of the section.

175 N. This is like 171 N, but in general the mesostasis of
comparatively small grains is wanting, being replaced by a filling
of coarse allotriomorphic quartz. In this case the conditions of
crystallization have evidently been gradual throughout. The

VARIATIONS OF TEXTURE IN IGNEOUS ROCKS 599

rock érystallized slowly under conditions of slight viscosity.
After the exhaustion of the feldspathic material residual quartz
crystallized in the remaining spaces.

Tabulation. —The following table shows the gradation from
fine-grained rhyolite to granite:

Specimen No. Character of groundmass.

164 N(a) - - - Glassy.

165 N (a) = - Cryptocrystalline.

162 N(a) - - - Finely microgranular.

169 N- - - - Microgranular; micropegmatic.

D7ZIIN Ses = - - Granular; grains average .5™™ diameter.
1723N = - - Granular; grains average .5™™ diameter.
175 N - - > Large quartz grains, average 2.25™™.

Analyses. —The following are analyses of the two fairly typi-
cal specimens as above described (analyst, Dr. H. N. Stokes):

(x) oe pew (2) 172 N. Siliceous granite.
SsiO% 71.40 S105 75.09
AE Ole 15.06 Al,O, 13.51
Fe,0O, i ey Fe,0, 1.03
FeO 88 FeO 08
MgO m5 MgO 18

= Fea 1.54 CaO QI
Na,O 4.19 Na,O 2258
K,O 3-39 K,O 4.71
eo 16 [BLO 17
H,0O+ .88 H,O+ 25
TiO, 20 TiO, 22
COP none Cos none
Ro Or .08 BO .04
MnO trace MnO trace
SrO trace SrO trace
99.73 99.87

It will be seen that No 2 is slightly more siliceous than No. 1;
nevertheless the two rocks are intimately related. In No. 1 the
relation of 1K @-- Na @ +CaO— 12:26." In Nor-2 the same
ratio equals 1: .15. Similarly, in No. 1 the relation of K,O:CaO
—en7 5. Ineo. 2-the same ratio equals) 12,32:
Conclusions. — Many of the bands in this granite-rhyolite
series show by their structure that they have undergone no break

600 Uh 1s SV OTE

in crystallization from beginning to end. The rocks have evi-
dently crystallized entirely in their present position, and the slow
hardening which is ordinarily indicated shows that this point of
consolidation was originally some distance from the surface.

In other specimens there have been slight breaks, bringing
about two, three, or more generations of crystals which in the
rocks near by are not distinguishable. These minor breaks were
due to slight migrations of material in certain bands, which
flowed slightly during the process of cooling, as is proved by the
angular fragments of finer-grained lava which they contain and
by the occasional broken condition of the phenocrysts.

As explanation of the difference in crystallization between
the granular bands and the intercalated fine-grained ones, it
must be remembered that those bands which were affected by
flowage must have been at the time those possessing least
viscosity and consequently those which were least crystallized.
The final crystallization of these bands, therefore, took place
at a later period than that of granular bands, at which period
the rate of solidification was very likely more rapid. It is prob-
able, moreover, that the movement of flowage brought on of
itself a more rapid crystallization than if the rock had been
undisturbed, and that thus the finer-grained groundmass
orginated. The same suggestions hold good for the similar
phenomena, already described at Mason Butte.’

t™In connection with the conclusion above arrived at, 2. ¢., that the phenocrysts
of the rocks were formed practically in place, compare the papers by Professor Pirs-
son and Professor Crosby. (Am. Jour of Sci., Vol. VIII, April, 1899, p. 271, “On
the Phenocrysts of Intrusive Igneous Rocks,” and American Geologist, Vol. XXV>
No. 5, May, 1900, “On the Origin of Phenocrysts and the Development of Porphyritic
Texture in Igneous Rocks.”’)

Professor Pirsson argues that the phenocrysts of intrusive rocks are not neces-
sarily intratelluric, and that there is no necessity of more than one period of crystailiza-
tion even for porphyritic rocks. From the fact that contact zones are often without
phenocrysts, while the rest of the rock contains them; that in a contemporaneous
complex of dikes and sheets some may have phenocrysts while others do not; from
observed cases where fluidal phenomena show that phenocrysts have developed after
the fowage; from the arrangement of the crystals of the groundmass around some
phenocrysts, showing that these crystals have been crowded and shoved during the
growth of the larger crystals; and from the fact that many granites (which have gen-
erally been considered intratelluric) contain very large phenocrysts, Professor Pirsson

VARIATIONS OF TEXTURE IN IGNEOUS ROCKS 601

TRANSITIONS OF TEXTURE IN THE GRANITE-RHYOLITES OF THE
QUINN CANYON RANGE.

Tela. description —Vhe Quinn Canyon range lies a long dis-
tance east ot all the other localities which have been described,
being almost due south from Eureka and nearly west of Pioche.
The whole southern portion of the range is buried in rhyolitic
flows... The range was examined by the writer at its northern
end, where the Paleozoic core of the mountains emerge from the
volcanic covering. On the western side of the range, near the
contact of the Paleozoics with the rhyolite, the stratified rocks
are pierced by numerous great dikes, which vary from coarse to
fine in texture. These dike rocks seem similar in composition,
and sometimes in texture, to the rather massive rhyolite which
forms the hills to the west of this locality.

Microscopic description— A specimen of the main rhyolite
examined under the microscope has the following characteristics :

Rhyolite (241 N.).—This rock has phenocrysts of all sizes,

reasons that phenocrysts are not necessarily of a distinct crystallization period as
compared with the groundmass, but may in some cases be formed in place. Professor
Pirsson advances the explanation that a comparatively rapid fall of temperature and
decrease of hydration, resulting in a viscosity augmenting in an increasing ratio, may
produce monogenetic phenocrysts (that is, phenocrysts which occur only in a single
generation). Recurrent phenocrysts (that is, those occurring in more than one gen-
eration) he explains as due to mass action, believing that minerals which are present
in very large quantity are especially active crystallizers.

Professor Crosby believes that no sudden changes of temperature, hydration, or
pressure, are necessary for the formation of phenocrysts. He believes that in a gradu-
ally consolidating rock the crystallization first established may be brought to a close
by the gradually increasing viscosity and that after passing this critical point new
zones of crystallization will be established, of much smaller field, and at this point
the groundmass begins and the phenocrysts end. If the rate of cooling is still
slower, an allotriomorphic granular texture results, while if the rate is more rapid the
texture may be glassy or nearly so.

The deductions of the writer, given above, agree with those of the authors cited
in this, that phenocrysts may be formed in place. His observations, however, go to
show that where cooling is strictly uniform there will be no distinct generations, but
a gradual transition from phenocrysts to groundmass; whereas, if there are distinct
generations, they are brought about by breaks in the conditions of consolidation, even
though these breaks be comparatively slight.

™G. K. GILBERT, Survey West of the rooth Meridian, Vol. I11, Geology, p. 122.

602 JAB SRURR.

from 1%™™ in diameter grading down into groundmass, which
is cryptocrystalline, probably a devitrified glass.

Two selected specimens of the dike rocks have the following
characteristics :

Biotite-granite-porphyry, near rhyolite (243 N.).—Like 241 N,
this rock has phenocrysts of all sizes, from 1% ™™ in diameter
down to the groundmass. There are, however, more phenocrysts
in this rock than in the one just described. As in 241 N, the
phenocrysts have no fluxional arrangements, but a divergent one.
The groundmass is fine holocrystalline allotriomorphic ganular.

Biotite-granite (242 N.).—This rock consists of grains of all
sizes from 6% ™™ in diameter down to the very minutest dimen-
sions. The smallest ones, which are very abundant, are about
.0o2 to .03™™ in diameter. There is a tendency to idiomorphism
throughout. The smaller sizes of crystals act as mesostasis LOiaa
the larger ones, and these have a mesostasis of the still smaller
ones. The essential minerals are quartz, othoclase, and biotite,
with accessory hornblende, titanite, magnetite, and a little
striated feldspar.

Analyses.—The chemical composition of these rocks is as
follows (analyst, Dr. H. N. Stokes):

(1) No. 214 N, Siliceous (2) No. 242 N, Biotite
Rhyolite. Granite.
SiO, 74.67 SiO, 71.48
Al,O3 eA Al,Oz 13.00
Fe,0, 1.06 Res OF 1.25
FeO .18 FeO 1.55
MgO trace MgO 95
CaO 120 CaO 2.60
Na,O 3.90 Na,O 2.60
K,O 4.62 K,O 4.24
H,O — 18 lat Oo 20
HO: 22 H3O-- 1.24
Ti@s .07 TOs; -43
CO; .79 C@e 30
POF 06 P.O; .09
S trace S none
MnO none MnO .09
BaO none BaO -09
SsrO none
100.35 | 100.11

VARIATIONS OF TEXTURE IN IGNEOUS ROCKS 603

InuNotat the! relation K>O-- Na,O : CaO——1 :.2. 9 In No: 2
the same relation equals 1:.53. In the same way, in No. 1,
Or Ca@;—- 1. 446, In No, 2 theisame ratio equals: 1 : 1,

Conclustons.—In the field the evident relation of the dikes to
the rhyolite led to the inference that they had been the feed-
ers of the extrusive rock. Under the microscope the composi-
tion of the three rocks is found to be the same, and the structure
shows variations indicating no great differences in the condi-
tions of cooling. The structure is identical with that of certain
specimens of rhyolite-granite just described from the Pinenut
range, and therefore need not be analyzed again. Briefly, in
all three it indicates complete crystallization in one place, with
no interrupting movement. The period of consolidation for 241
N was comparatively short, that of 243 N somewhat longer,
and that of 242 N markedly greater than 243 N.

ANALOGOUS CASES OF VARIATIONS OF TEXTURE IN OTHER PARTS
OF THE GREAT BASIN.

In the Washoe district, Nevada,-not far from the first
locality decribed by the writer, Messrs. Hague and Iddings*
found a gradual transition from pyroxene andesites with glassy
sroundmass to pyroxene-diabase with coarse granular structure.
In the Sutro Tunnel they found coarsening of the crystallization
as the tunnel nears the core of Mount Davidson, so that at one
end the rock may be called andesite and at the other end diabase.
They also discovered like transitions between andesite and
granular diorites.

Similarly they found that the earlier hornblende-andesite
passes into diorite, while the later hornblende-mica-andesite
changes into mica-diorite in such a way that the two rocks are
inseparable. They concluded that a dike of so-called diabase
is a variation of the basalt, which was one of the latest extrusions.

In short, according to these writers, the coarse holocrystalline
rocks of the Washoe district are chiefly Tertiary, and are partly
extrusive and partly closely connected with extrusives of similar

Bull. 17, U. S. Geol. Surv.

604 J. E. SPURR

composition. Another interesting conclusion is that the change
between the lava texture and the granitoid texture consists
chiefly in the coarsening of the groundmass.

In 1899, Messrs. Tower and Smith described textural transi-
tions in the Tintic range in Utah, which lies within the petro-
graphic province of the Great Basin and is situated southward
from Great Salt Lake.*- In this district is found pyroxene
andesite or perhaps more properly latite, which is an effusive
rock and is closely associated with granular monzonite. There
are all variations of texture between andesite with glassy ground-
mass to that with a holocrystalline groundmass; from this to
closely similar rocks, also with holocrystalline groundmass, which
are called monzonite porphyry ; and from these through panidio-
morphic granular phases to those of hypidiomorphic granular
structure.

CONCLUSIONS.

In the Great Basin, particularly in Nevada, we have Tertiary
extrusive rocks which show transitions from a granular structure
with glassy groundmass. The different phases are often inti-
mately associated, and structural analysis shows that the differ-
ences of crystallization which brought about these variations
were slight, a relatively small decrease of the rate of cooling
being sufficient to allow the formation of the holocrystalline
instead of the porphyritic structure.

Transitions similar to those found in the Great Basin have
been sparingly chronicled elsewhere. These appear to become
rare in proportion as the rocks become siliceous. This is so
because with a given relatively rapid rate of cooling a magma
of basic composition will consolidate with a holocrystalline
structure, while a siliceous magma will become fine-grained
and porphyritic. We have accordingly many instances of
holocrystalline diabases which are certainly extrusive, and of
similar rocks in rather fresher condition (generally due to
their being younger) which have been called dolerites. In
the more siliceous rocks such textural transitions are rare in

' Nineteenth Ann. Rept. U. S. Geol. Surv., p. 656.

VARIATIONS OF TEXTURE IN IGNEOUS ROCKS 605

effusive bodies, but further down, at the roots of volcanoes, the
conditions are such as to allow viscosity to increase with the
same slowness that it does with a more rapid cooling rate in
more basic rocks. Hence with increasing acidity we find the
coarser grained varieties further removed from the surface. In
general, however, it is plain that a granular rock is not necessarily
a deep-seated one, in the formerly accepted sense of the word.

Another conclusion which may be made from the foregoing
studies is, that the more important structures are not peculiar
to particular rocks. The porphyritic and the coarse granular
allotriomorphic or hypidiomorphic structures are already recog-
nized as characterizing all rocks, of whatever chemical composi-
tion. Also the aplitic structure, or that in which idiomorphic
minerals (which are the same as the phenocrysts of the por-
phyries) form the greater bulk of the rock, has been recognized
as universal by Rosenbusch, who has described it in granitic rocks
and in all intermediate ones down to gabbros; thus his rock
terms include syenite aplite and gabbro aplite. The ophitic
structure has been generally supposed to be characteristic of
diabases, and without question is here best exhibited, on account
of the rate of solidification which the basic composition of a
magma entails and also because the elongated forms of the basic
feldspars make the structure prominent. The foregoing studies,
however, show that this structure is intermediate between the
porphyritic and the aplitic structures, representing a stage in
crystallization when the idiomorphic crystals (or phenocrysts,
as they are called in the porphyries) have multiplied and grown
so that they interlock; and that like these other structures it
may occur in any rock. In the granites it is not so striking as
in more basic rocks, on account of the blunt form of the alkaline
feldspars which form the first generation of crystals, but it is
nevertheless present in some of the granites which have been
studied. In diorites the ophitic structure has been occasionally
described.*

TROSENBUSCH, of. cit., p. 2560; ZIRKEL, Lehrbuch der Petrographie, 2d ed.,
p. 483.

606 PL STOLL

The writer, of course, interprets the term ophitic in its broader
and not in its narrower sense, accepting it as meaning a structure
where a network of interlocking, divergent, comparatively large
feldspar crystals is filled in by grains of much smaller dimensions,
whatever the nature of these grains may be. He does not inter-
pret it as meaning that the mesostasis is necessarily augite. He
finds grounds for this broader acceptation in the writings of
Rosenbusch, Zirkel, and others. Rosenbusch applies the term
ophitic’ to diabases where the mesostasis is not augite, but an
aggregate of primary quartz and feldspar.

Therefore, the glassy, fine porphyritic, coarse porphyritic,
ophitic, aplitic and hypidiomorphic granular structures may occur
inany rocks. They pass by gradual transitions into one another
and are dependent upon relatively very slight differences in con-
ditions of cooling. All may be formed without any marked
migration of the consolidating rock.

These conclusions are important in considering rock classifi-
cation, as showing that structure cannot be made the element
of greatest importance. Granites, granite porphyries, granite
aplites and rhyolites, for example, must not be separated, but
put as closely together as possible, and the same is true of
diabases, diabase porphyrites, diabase aplites, and basalts.

J, Ee SPURR:
* Mikroskopische Physiographie, 3d ed., p. 1117.

THE_FOYAITE-IJOLITE SERIES OF. MAGNET COVE:
A CHEMICAL STUDY IN DIFFERENTIATION.

INTRODUCTORY.

SOMETIME since I published a paper’ on the ‘Igneous Com-
plex of Magnet Cove,” in which it was shown that the main types
found there were arranged in avery regular series from the center
to the periphery of the mass, and that this was an excellent
example of the differentiation of a magma in place, presenting,
however, the anomaly of being less ‘‘basic”’ at the borders than
at the center. It was also remarked that the analyses then
available ‘‘vary continuously in one direction, with scarcely a
break or abnormality of any kind.”?

Since then several considerations led to the belief that a new
and more detailed chemical examination of the main rock types
was desirable. Several of these, notably the ‘‘ leucite-porphyry ”
and ijolite, are representatives of rock groups of great theoretical
importance, complete analyses of which are highly desirable. In
this respect many of those published by Williams? are defective,
the non-determinations of the rarer constituents being largely
due to the fact that the importance of completeness in rock
analysis was not recognized at the time they were made.+

Such a reéxamination seemed to be the more desirable, since
in a recent paper Pirsson’ has shown that the rocks occurring in
the Little Belt Mountains of Montana form an extremely regular
series. By plotting the constituent oxides on an abscissal basis
of distance from the center of the mass, he arrived at the con-
clusion in this case that, ‘‘ given the percentage of one element,
the chemical composition of any rock of the series to within a

t Bull. Geol. Soc. Amer., Vol. XI, p. 389, 1900. 20D. Cit.5 Pp: 403.

3J. F. WriiiaMs, “Igneous Rocks of Arkansas.” <Azn. Rep. Geol. Surv. Ark.
Vol. II, 1890.

4 Cf. J. C. BRANNER, in Williams, of. c7¢., p. xiv.
5L. V. Pirsson, Twentieth Ann. Rep. U. S. Geol. Surv., Part UH, p. 569 ff., 1900.
607

608 HENRY S. WASHINGTON

fraction of I per cent, can be deduced from the diagram.” This
instance is the first in which the differentiation of a mass of
magma is rendered capable of exact mathematical treatment and
proof, and its great theoretical interest and importance is obvious.
The conditions of the locality were such that it was thought that
the Magnet Cove complex might furnish another favorable
example of the same kind, a hope which was justified by the
results obtained, as will be seen later.

The analyses of six of the representative plutonic rocks of
Magnet Cove, as well as two of Fourche Mountain, were there-
fore undertaken, with the determination of the rarer constituents
which might be present. For petrographic descriptions the
reader is referred to the two papers cited above.

In this connection I would express my full endorsement of
Pirsson’s remarks* on the importance of good and complete
analyses, which are absolutely essential for such mathematical
discussion, and with him deplore their comparative rarity. To
many petrographers, any collection of figures which foots up
within 98 and 102 per cent. is a usable analysis, even though the
results are at variance with the mineralogical composition, and
some of the obviously important constituents are not estimated.
It is not realized that such ‘‘analyses’’ do far more harm than
good to the science. From this point of view alone, the excel-
lent work of Dr. Hillebrand and the other chemists of the United
States Geological Survey is of inestimable value, as they have
set a standard to fall short of which in any marked degree
should be accounted a petrographic sin. _Let me be the first to
confess ‘‘ peccavt.”

ANALYSES.

Pulaskite.— The specimen of this type was collected at the
Little Rock Granite Company’s quarry, at Fourche Mountain,
near Little Rock, the type locality. In I is given my analysis,
in II that of R. N. Brackett, as quoted by Williams. The two
do not differ materially, I showing rather less Fe,O, and CaO
and more »FeO, MeO, "and alkalis, thought sthe: ratrojmor

™L. V. PIRSSON, of. czt., p. 578.

FOYVAITE-IJOLITE SERIES OF MAGNET COVE 609

I II I II
1 OS acne seer 60.20 60.03 dit Ose rane 0.14
A Ok concer 20.40 20 7709 MOLTO Ss ices. « trace oun
Fe,O, ayia “isto — |H Neva Ok serra eae 0.15 0.07
He Oe ae cence: 1.88 0.75 SIO) see! Gum One
Wek O Pb cncacecibe 1.04 0.80 (2) lis eer 0.09
CA neers 2.00 2.62 Seabee soos none avanti
Nias OR az esc 6.30 5.96 Mimi Oey. on ee trace trace
os Oerne et aa 6.07 5.48 Ba@ irra trace sarki
H,O(110°+) On23 0.53
H,O(110°—) 0.10 0.06 100.47 IO1.07
Co ORas ont: none ites

I. Pulaskite, Fourche Mountain. Washington, analyst.
II. Pulaskite, Fourche Mountain. Brackett, analyst. WILLIAMS, of. cit, p. 70.

Na,O:K,O is about the same in both, being 1.57 in I and 1.66
in II. The specimen analyzed by Brackett apparently contained
a little more acmite, though the variations are scarcely more
than are to be expected in different specimens from the same mass.

Analysis I can be calculated out as below. There is appar-
ently an excess of Al,O,, amounting to about 4 per cent. in
this, as well as in II, which it is difficult to account for. It does
not seem to be due to kaolinite or hydronephelite, as the rock
is too fresh, and there is no muscovite present. Such an excess
of Al,O, above (Na,O+K,0) — Fe,O, may be observed in
many analyses of nephelite-syenites, if the Fe,O, is calculated
as acmite, and no anorthite molecule is assumed to be present.
Among these are some by Hillebrand, and there is no reason to
suppose that this peculiarity is due to errors of analysis. It is a
feature of this group of rocks which seems to call for investiga-
tion, and may possibly be connected with the occurrence of
corundum in the nephelite-syenites.

Orthoclase - - - - - 35.1
Albite - - - - - - - 39.8
Nephelite - - - - - - 3.1
Sodalite - - - - - =, &12
Aegirite - - - - . - 5.1
Hornblende, diopside, and biotite - - II.0
Apatite - - - - - - 0.5
Extra alumina = - - - = AR2

100.0

610 HENRY S. WASHINGTON

Pulaskite (foyaite).— This is Williams’s ‘‘gray granite,” the
specimen coming from Braddock’s quarry, on Fourche Moun-
tain. My analysis is given in I below, with that of W. A. Noyes
as quoted by Williams in II. The two are closely alike in all

I II I II
SIO) corcod pad 60.13 59.70 CO sree alec none
AE Orgies sckors 20.03 18.85 AB OYE Alone 1.15
Hex Ores 36 4.85 (GO oy As 6 a eee 0.05
HE Oe Secciieeees 1h BB Hever PaOmtegnccr 0.06
Mic Oi varaedar 0.76 0.68 SOs ss s6000 0.14
Cari sane 0.87 1.34 MnO.. trace
NEYO Son adae 6.30 6.29 BaOm nus trace
IS Olona doc 5-97 5-97 ——
HS Ol(arors-) 1.41 1.88 100.72 99.56
EDS ON(uvOR—) 0.16 ania

I. Pulaskite, Braddock’s quarry, Fourche Mountain. Washington, analyst.
II. Pulaskite (‘‘ Foyaite”), same locality. Noyes, analyst. WILLIAMS, of. c7¢., p. 81.
respects except Al,O,, which is about 2 per cent. higher in I,
as about 1 per cent. ot T1@,, and a _trace: of FO mustabe
deducted trom’ the yAl, O;of Tl

The analysis (1) calculates out as follows, and it is evident
that there is no essential difference between this rock and the
last mentioned.

Orthoclase - 28.6
Albite - - = 30:0
Nephelite - - 6.2
Nosean - Sah he)
Kaolin - =) pa 78
Aegirite - OHS)
Biotite - - 7.9
Titanite - =| = 2.7:

100.0

Foyaite—What seemed to be an average specimen of the
occurrence at Diamond Jo quarry, Magnet Cove, was chosen for
analysis. No pyrite was visible, though the analysis shows a
trace of sulphur. One or two small garnets were seen in one
section, which mineral is not mentioned by Williams as an
accessory.

FOYAITE-IJOLITE SERIES OF MAGNET COVE 611

Apart from the determination of minor constituents, my
analysis (1) does not differ materially from that of Brackett

I Il I II
SHOR Gan gcen 53.09 53.38 ALTO} emerens O.I1
FAA OVS aon cree 21.16 20.22 LO ncrcrarccn 0.04
Bes Oeics. a 1.89 1.56 Pi Ole ream 0.15
Hie © Biles haere ae 2.04 1.99 DORE fii none
MgO 0.32 0.29 Mey steaciccans 0.02 mises
CaOnmeniiene 2.30 2.20 SS itedeete arent 0.08 i PP
Na,O 6.86 7.89 Wint@k eee 0.20 trace
Kes Okgnc sonra: 8.42 6n21 Ba@e eee: 0.61 Mereig
H,O(110°-++) Tents)
iS Olio s——) 0.24 3.42 100.48 100.03
COBMSt eae 0.82

I. Foyaite, Diamond Jo quarry, Magnet Cove. Washington, analyst.
II. Foyaite, same locality. Brackett and Smith, analysts. WILLIAMS, Sp. gr. 2.599
at 26°C., op. cit., p. 238.

and Smith (II), except in the presence in the latter of nearly
2 per cent. of pyrite, and in the alkalis. I gives a ratio of
Na,O: K,O of 1.3, while in II it is 1.92. The amount of BaO
in I is high, and is noteworthy since it is found only in traces
or not at all in the other rocks. It probably belongs with the
abundant orthoclase, since other cases are known of BaO par-
tially replacing K,O in the feldspar of alkaline rocks. The
analysis calculates out thus:

Orthoclase —- 51.8
Nephelite - = 2003
Cancrinite - eal
Aegirite - Been ete
Diopside - - 8.6

Titanite and pyrite - 0.5

100.0
It is seen that this bears out Williams’s remark that ‘the
orthoclase is perfectly free from isomorphous mixtures of other
feldspars.” It is indeed somewhat remarkable that the albite
molecule should be entirely lacking, or almost so, in a rock
containing so much soda. It seems to be characteristic-of the

Be oan

612 HENRY S., WASHINGT ON,

Magnet Cove rocks that albite, as well as anorthite and horn-
blende, are of very limited occurrence.

Covite (‘‘Shonkinite’”’).—The specimen of this rock, which is
Williams’s “ fine grained syenite,”
house on the western border of the Cove, and the analysis was

came from below the school-

published in my former paper. It is repeated here, with the
addition of P,O,, which has been determined since. Several
analyses of analogous rocks are also given.

I II III IV V

SiO... 49.70 44.05 47.61 48.98 46.99
ASO: 18.45 13.87 14.26 12.29 17.94
IeAOn eases 3.39 6.06 4.90 2.88 2.56
He@ re wees 4.32 2.94 4.07 5-77 7.50
IMIG @ mitoses: 2aa2 5.15 2.62 9.19 Ba22
(Cri O Reactors 7.91 9.57 VAY /al 9.65 7.85
NEROs epamee 5.33 5.67 6.70 222 6.35
AO hein aid. ao 4.95 4.49 4.08 4.96 2.62
H,O(110°-+) 1.09 2.10 1.89 0.56 0.65
lal (@){(aiiKe)—>)) 0.25 0.95 0.26 0.26 ui
(GOnaeog eee es OnE crs
Tei Olerts eeessvens 133 0.95 1.38 0.98 0.94
ZrO.. aisip 0.18 BAe
lP a Oaare om 6 i 0.40 1.50 1238 0.98 0.94
SOR es 0.61
Ol Sap See aca cade trace 0.37
1 us neain ceo ona ners trace 0.22
Sesistcneuteener pated seats 0.03 eye Rater
MIMO) Fs a oodles trace ONL, 0.30 0.08 trace
BaOreiceeat 0.76 0.41 0.43 none
Sr@Mate senna: 0.37 0.36 0.08

99-44 99-99 100.68 99-99 99.60

I. Covite, Below Schoolhouse, Magnet Cove. Washington, analyst. 4zd/. Geol.
Soc. Amer., Vol. XI, p. 399, 1900.

II. Theralite, Gordon’s Butte, Crazy Mountains, Montana. Hillebrand, analyst.
J. E. Wourr, Bull. No. 150 U.S. Geol. Surv., p. 201, 1898.

III. “ Tinguaite,” Two Buttes, Colo. Hillebrand, analyst. Bul. rg8 U. S. Geol.
Surv., p. 182, 1897.

IV. Shonkinite, Yogo Peak, Little Belt Mountains, Montana. MHillebrand, analyst.
WEED and Prirsson, Am. Jour. Sct., Vol. L, p. 474, 1895.

V. Essexite, Salem Neck, Mass. Washington, analyst. JouR. GEOL., Vol. VII,

Pp. 57, 1899.

The mineralogical composition of these rocks is such that
their calculation must, of necessity, be arbitrary and only

FAVAITE-IJOLITE SERIES OF MAGNET COVE 613

approximate, but those of I, II, 1V, and V may be very roughly
reckoned out, as below, that of IVa being Pirsson’s calculation.

la Ila IVa Va

Orthoclase ease isuesa. Deus 20.2 29 - TOR
PAMb tes fess, ceacecsie antiees aa evs 22.8 a 2 Laie
AmMorthitel sce kclc sci, store Binds a 17.2
ING PWelite top ees custo claret) 9.0 23 a 20.0
FL AUV MC is ancin sieeta tye oan sao 4 a Rates
PLO BITTON aa ccs as Giaeste ¢ eoevst eee 4.5 4 he Bay
DVO PSIG Chi relepeteters erstateues at: 9.0 25 35 Brill
Ilornblendet irre ct. sets « 18.8 : 5 he P
IBIOUItER jeg: nes hee cee etoases ; : 18 Ree
Olivine ys ern, etter yee Bore. 2 7 Oin2
Malo titers. sitio. saves 255 7 i Ave
eitamite gece teaser eestee execs ors Rot 2 4.0
JN ENDED Gag ween Oe 60 Goer 1.0 4 1.0

100.0 100 100 100.0

In my former paper I discussed briefly the position of this
rock in classification, and provisionally put it with the shonki-
nites. At that time the mineralogical composition had not been
calculated, and this position was assigned to it because it resem-
bled Pirsson’s shonkinites, except in the presence of nephelite
and of hornblende instead of biotite, and also because it came
under Rosenbusch’s definition? of these rocks, whose essential
features according to him, are the presence of abundant dark
minerals along with nephelite and orthoclase. As was also
remarked, it cannot be put with the essexites or theralites
(although chemically closely resembling these), on account of
the lack of plagioclase.

In this connection it is of great interest to note the fact that,
in his latest description? of typical theralite, J. E. Wolff states
that there is nothing which can strictly be called soda-lime feld-
spar present. Indeed this fact is evident from a consideration
of the analyses by Hillebrand, published in the same place.
The name theralite, therefore, cannot be applicable to Wolff’s
Montana rocks, or else its definition must be changed.

t ROSENBUSCH, Llemente der Gesteinslehre, p. 174, 1898.

2 Bull. U. S. Geol. Surv., No. 150, p. 197, 1898.

614 HENRY S. WASHINGTON

It will be seen that, though the covite and the theralite of
Wolff resemble each other in qualitative mineralogical compo-
sition, as both are composed essentially of alkali-feldspar, nephe-
lite and ferromagnesian minerals, and that both are distinctly
leucocratic in character, yet that in a quantitative mineralogical
way they are decidedly different. The feldspathic constituents
of the covite are very largely feldspar, with only accessory
amounts of nephelite, while the theralite shows about as much
feldspathoid as feldspar. The calculation of the latter cannot
be exact, since some of the soda goes into the feldspar, but this
must be small, and cannot affect the result to any great extent.
It is evident, then, that the name of theralite is not appropriate
for the Magnet Cove rock, though: it might be used in the
present very vague and loose method of classification, based
largely on qualitative mineralogical composition.

A comparison of Pirsson’s descriptions* with Rosenbusch’s
definition of shonkinite indicates that the latter has been appa-
rently laboring under a misapprehension of the former’s descrip-
tions, and that his definition does not cover the rocks as Pirsson
described them. Pirsson expressly states in each case that
nephelite is either entirely absent or present only in mere
traces, which does not coincide with the definition which makes
nephelite an essential constituent.

Although resembling each other in many ways, yet there are
certain striking differences between the analysis of the Magnet
Cove rock and those of shonkinite. In SiO,, iron oxides, CaO and
K,O they are closely alike, but in the Magnet Cove rock Al,O,
and Na,Oare higher and MgO lower. ‘Indeed the calculations of
the mineralogical composition, though that of Ia is only approxi-
mate, show clearly that while the ‘“‘covite”’ is distinctly leuco-
“cratic the shonkinite is as decidedly melanocratic. A similar
distinction will be pointed out between the ‘“ leucite-porphyry ”
and missourite. In this respect the rock under consideration
resembles the typical essexite, though here again there is a dis-
tinct difference in the amount of K,O, in the essexite this being

tL. V. Pirsson, Bull. Geol. Soc. Am., Vol. VI, p. 408, 1895; Am. Jour. Scz., Vol.
L, p. 474, 1805; Am. Jour. Sct., Vol. I, p. 358, 1896. -

FAVAITE-ITJOLITE SERIES OF MAGNET COVE 615

much lower, and plagioclase entering to a very considerable
extent:

For this leucocratic holocrystalline combination of ortho-
clase (alkali-feldspar) and less nephelite, with hornblende and
aegirite-augite, of granitic structure, and with a composition
like that given in the analysis above, I would propose the name
of Covite. If only the qualitative, not the quantitative, miner-
alogical composition be considered, the covites may be called
basic nephelite-syenites or foyaites. But the whole tendency
of modern petrography is, rightly, against this narrow view of
rock classification, and the use of a new name seems to be
abundantly justified. In ordinary typical foyaites the alkali-
feldspars and nephelite, eté., make up from 75 to go per cent,
of the rock, the dark minerals consequently only from 10 to
25 percent. In the covites, on the other hand, while the type
is rather leucocratic, the ight and dark minerals are. present
more nearly in the same amount, and these rocks might justly
becalled “ mesocratic.”

As a matter of fact, accepting Pirsson’s definition of shonki-
nite as the standard (viz., melanocratic combination of alkali-
feldspar with pyroxene, etc.), the covites are the rocks which
correspond to Rosenbusch’s definition of shonkinite. A simi-
lar rock, which also belongs here, is that the analysis of
which is given in III, and which Cross provisionally called a
etinoualtes ”

Arkite (‘‘Leucite-porphyry”).—The specimen which was selected
for analysis came from an exposure a little to the northeast of
and above Diamond Jo quarry. Judging from the other speci-
mens which I collected around the area, it seemed to be repre-
sentative and an average specimen of the occurrences. A good
sized hand specimen was used for the analysis, so as to obtain a
fair sample of this rather coarsely porphyritic rock.

The results, given in I, were rather surprising in comparison
with the analysis by W. A. Noyes of another specimen from the
neighborhood (II). Not only is SiO, much lower, but MgO is
a little higher, CaO much more so, and, though the total amount

616 HENRY S. WASHINGTON

of alkalis remains the same, the new analysis shows a rock
relatively richer in potash as compared with soda.

I II III I II Ill

SLO Gee. reae vena seas 44.40] 50.96] 46.06] | ZrO, taco GOB) soc oars
AO iat coactonarseercnes LORS || PLO O 71 | LOO | en Ole Wag ieeeleryat stents: Oge7 aoca|| “Osan
ISGAOn aves ooo 6-6 SoG Wao Bay SO coscaceaoes 0.06] trace 0.05
REA OO) By Uae) ers ate PAST AUN merece E SOmele RG lineenemi erie lmtnace 0.25} 0.03
Mg Ojscxawneteucerescdcns Wg Sl Wo Aa) hoy IMGONO\ So oo onc 60510 6 0.08] trace | trace
Cal eeiscsn tector Sellen Ae etel) Uo HS BNO) poo conoccs 5 OOH) Moaoalls Osa
Nan @oeiciteerue samen ROSS ea lolly bees I SiO sows macace aoc Pea Bab a| » Oc Be
a Olver nsiersitenee iets of68.14| 6.77) 5.14 (a

EES OU(TOR=>)) eel Leki 38 UN 100.76|100.01| 99.57
ILO (iio ease Oued S36 eal) none Se Meessn OR Clea lepreesers 0.06] 0.01
COR ere re ictet Olsson eu cllin tetas -——.__--~

Ti Olpediveisnaye alae TSO 5:2 | Med O's heres ae beats |FoUaenenet 99.95| 99.56

I. Arkite, Magnet Cove. Washington, analyst. Sp. gr., 2.770 at 26° C.
II. Arkite, Magnet Cove. Noyes, analyst. WILLIAMS, of. c?¢., p. 276.
III. Missourite, head of Shonkin Creek, Highwood Mountains, Montana. Hurlbut,

analyst. WEED and Pirsson, Am. Jour. Sct., Vol. Il, p. 321, 1896.

The discrepancy between the two analyses of the leucite
rock cannot be explained by the supposition that the specimen
analyzed by Noyes carried a larger proportion of pseudo-leucite,
since, although the other constituents work out well on this
basis, the amount of K,O in II is not intermediate between that
in I and in Williams’ analysis of a pseudo-leucite crystal. It
seems to be the case that Noyes’ specimen represents a slightly
different phase, possibly richer in aegirite, but poorer in diopside
and garnet. From my own observations in the field and the
specimens collected, I conclude that the specimen of I represents
the normal rock more closely than that of II.

It may be remarked that this supposition is borne out by the
fact that analysis I is, in a general way, intermediate between
that of the covite and that of the ijolite, given later, while
Noyes’ is not. This is to be expected in view of the observa-
tion noted in my former paper (p. 395), that ‘‘ while the rela-
tions of the ‘fine grained’ (shonkinitic) syenite to the leucite-
porphyry are uncertain, the former lies apparently outside or
above the latter. <

In the absence of discrimination between the two iron oxides

FAVAITE-I/OLITE SERIES OF MAGNET COVE O17

in II, it is impossible to make a satisfactory calculation of
Noyes’ analysis, but No. I works out thus, the result being only
an approximation, owing to the composition of the rock. In
ITJa is given that of the missourite, as calculated by Pirsson.

la Illa

Orthoclase castes sare ets 20 IBC UCIbC Berge aptetsus tates earn 16
IGEUCILE acter dee tetcch aomreewe eA 30.9 PVA Cubes antares tenses sete 4
INeplelitemern melts atari 2155 EOC CSurnatwace Ato as 4
ENE ITIL Ce recs ce-tey oe srafenerae cu 8.4 NGI C Maret ee xsisene Slava reete 50
DiOpSidemer ier ata.) nie cue cen 10.8 Olivine 15
Garmeteacermce ei dae oa 14.5 DIO LUCE ee in tierce 6
Iigeyeoldenmnrena nee Rio oe oor 5

100.0 100

It is evident from this table that while both rocks are alike
in being composed essentially of leucite, with subordinate
nephelite (or zeolites), and dark minerals, yet that they differ
radicaliy from each other, just as did the covite and shonkinite.
The Magnet Cove rock is distinctly leucocratic, carrying about 66
per cent. of light minerals, while the missourite is as decidedly
melanocratic, carrying only 24 per cent. of these.

It is obvious from the mineralogical composition, as well as
from the analysis, that the name ‘syenite’’ which has been
applied to this rock is not justified, if this term is to retain any
precision of meaning except that of indicating the absence of
quartz and an alkaline character. Since this is so, and since the
rock represents a most interesting and quite distinct Ly Pe; it cer
tainly should have a distinct appellation of its own.

It would seem peculiarly appropriate to honor the memory
of its first describer, J. F. Williams, by calling it Williamsite.
But since this name has been already preémpted by Shepard for
a variety of serpentine, and as it would be a solitary exception
among rock names, it will be best not to do so. I propose,
therefore, the name of ‘‘arkite”’ (from the usual abbreviation of
the state name Arkansas), the essential features being a holo-
crystalline, porphyritic, leucocratic combination of leucite (or
pseudo-leucite) and nephelite, with pyroxene and garnet.

618 HENRY S. WASHINGTON

Tjolite.—The analysis of this type, from below Dr. Thornton’s,
has been already published," but is here repeated, with the addi-
tion of several constituents which have been determined since.
In II is given the analysis of a typical ijolite from liwaara, in
Finland. ~The) twordo not dither materially, except that, Wis
higher in CaO and correspondingly lower in Na,O.

J II I Il

SHOR auanisec 41.75 43.70 COniniee an: none BERS
IMO ceo 0108 17.04 TOR 7, THO Sees 0.58 0.89
Hes Ore ciese 6.35 235 JARO Ba si -8 6 0.05 See
He Oi aereiewine 2Al 3.94 PS OP Sanaa 1.09 1.34
Ng Oe nicie Ae Git ZiQge s-lllhSmenarnemccuys none Neues
Ca@Qreye cee 14.57 10.30 IMGNKO SS oes bic trace trace
Nas Oren (oye 1047/ 9.78 BaQiincvoies ae none
Te Oe se 3.98 2.87
H,O(110°+) 0.62 8
H,O(110°—) 0.28 0.59 100.60 100.30
I. [jolite, Magnet Cove. Washington, analyst. Sz. Geol. Soc. Am., Vol. XI, p.

399, I900. Sp. Gr. 3084—26°C.

II. Ijolite, iwaara, Finland. Sahlbom, analyst.
Finl., No. 11, p. 17, 1900.

V. HaAcKMAN, Bzll. Com. Geol.

The mineralogical composition of the two is given below,
that of Ila being Hackman’s calculation. JI is almost exactly
half nephelite, while TE contains rather ‘less than shalt vot, ths
mineral, but both may reasonably be called mesocratic. Hack-

man’s specimen did not contain any garnet, but this is a very

la Ila2

INephelitesteensecr 38.7 55.00
INCRAVENE 5.65.0 6.0.0 016 0 4.6 Fe tS
Di psidlenaere vcr Biles 3322
ANUS chan Gone c 6c GOs GaSe ie aes
Mielamiterseyraiecichete Igy les: | ee aemtcmnes
ARitamitermecerenrae: aa 2.16
NORNIIWS 3 56.5) old o10:c 3.0 By eity/

100.0 100.70

*H. S. WASHINGTON, of. c7t., p. 399.

? There is a clerical error in Hackman’s results, as he gives the nephelite as.
51.02, the sum as 100.50.

FAVATTE-IJOLITE, SERIES OF MAGNET COVE 619

variable constituent in the Finland ijolites. Hackman (of. ciz.,
p. 4) notes the identity between the Magnet Cove rock and the
Finland and Alné ijolites.

Biotite-ijolite—An analysis was also made of this rock, the
specimen coming from near the Baptist church, and the results
are given in I. My specimen was, unfortunately far from fresh,
so that the figures are of little value. Williams’ analysis (II)
of the same type, undoubtedly made on fresher material, is to
be preferred. The chief feature of interest in I is the (for this
oxide) large amount of ZrO,, which may be correlated with the
neighboring ‘‘eudialyte-syenite pegmatite” described by Wil-

liams.

I II I II
SHO)leamoooes 38.11 38.93 SOV aoe eterna: none State
JIA Oe Basten 20.84 15.41 Cl erage: Saieus 0.02
ResOr 5.67 5. 10 DAC pene els 0.14 0.891
He@ ie cacs ot 1.46 4.24 Min Oreo 5- 0.14 trace
IMIG OM ees ve 3.80 5 aS SEO) Gans oo trace aed
CaO here rien 14.44 16.49 SL OMe ccmtns trace
Nia sOMe ainces: 6.65 527, Tete Oo 2S oss trace
Ke Osteo. aes 2,12 1.78
H,O(110°-+-) 4-51 5.20 100.60 100.87?
H,O(110°—) 0.57 WeSseO) erect 0.04
COne see ae 0.65 uae
AKO aoa ee 0.48 1.62
ZiOR op ate 0.18 on awa ; pe a
Bs Oe ees steer 0.84 0.35 aia eae : :

The composition of the rock is such that any calculation of
the mineralogical composition must be rather arbitrary and
unsatisfactory, but the following (Ila) represents roughly and

Ila,

Orthoclase . - - 4.8
Nephelite - - - 24.1
Biotite = - - = 6:2
Diopside - - - 30.6
Melanite - - - = = 2350
Schorlomite - . - 627.
Magnetite - - - =", 3.6
Apatite - - - - 1.0
100.0

tFeS,. 2 Williams gives 100.57.

620 HENRY S. WASHINGTON

approximately that of Williams’ specimen. It is probable that
my specimen was rather richer in nephelite and poorer in dark
minerals than Williams’. The composition is much the same
in a general way as that of the ijolite, only that it is melano-
cratic, rather than mesocratic.

Jacupirangite—A new analysis was made of the dark, coarse-
grained rock, composed largely of augite, which occurs as a
small mass northeast of the main area, on Cove Creek. This
was deemed to be advisable since the analysis of Williams
showed more CaO, or less MgO and FeO, than was necessary
to form augite or any other mineral present. The analysis of
Williams is given in I, and my results in II, with two other
analyses for comparison.

No I Il Ill IV

SiO eee macr ond epnses Bee! BOstal 38.39 Le 38.38 45.05
IA es Ora ih rasa tary ete ecrceeagre oct 8.22 7:05 o& 6.15 o 6.50
Key A OURt i Niner stan erate ced anni 8.29 o 9.07 11.70 3.83
Hie ©) aie eee eerie yee etena Brean (yc, 109) 8.14 7.09
Mig Oe rss areca cs iteueeectauete 8.19. 11.58 Il .472,9% 12.07
CAO Bear parineee eaes sere ete 18.85 3 19.01 18.60 3 18.82
INE Olina eerher sa ceaereeea Rian ate ta DralOrs as 0.74 0.78 5/4 0.94
Kes OM ceeacnae eianee orsaek 1.086 0.75 On 13) p67 0.78
EPS ORTON See, cheep ies 1.40 0.33 0.54 be 7
lal tO) UO ee bog atioadoo. 5 lds 0.14 0.18 ay
COR Baia aire spree arene eee 0. 32, none
TiO, Brelil Oe 4-54. 4.32 2.65
TsO) grercerac OGs aa tonic Crs creat a none eae

TaNGRE sectors ta wai ted Soeeheetnti aaa rete 2.10 0.24
POR iis edad un entececas a ee 0.82 0.17 0.15
CU eet ae te Hite, Spe pete eae ra
Siac Se ayers ecnet teeter ues GER — 6.03%, - 0.42
Min O08 Sareea wees Zatacaonn 0.32 004 0.16
Ba Ores muuk orsal on ieasucesneeae aoe trace
SEO Mae waco ane ere amare ero trace

99.22 99.89 100.72 100.88

Sp. gr., 3:407—26" €. ;
I. Jacupirangite, Magnet Cove. J. F. Williams, analyst. Of. ci¢., p. 227.
Il. Jacupirangite, Magnet Cove. H.S. Washington, analyst.
Ill. Jacupirangite, Jacupiranga, Sao Paolo, Brazil. H.S. Washington, analyst.
IV. Pyroxenite, Brandberget, Gran, Norway. L.Schmelck, analyst. W.C. BROGGER
OPT (GS Sey MOM IL, jos Si IO
tHe Ss.

FAVAITE-IJOLITE SERIES OF MAGNET COVE 621

While I and II are alike in a general way, yet there are
marked differences in SiO,, FeO, MgO and S. The specimen
analyzed by Williams carries considerable pyrite, while mine
only showed a few specks of it. The differences in the other
constituents named may be attributed to alteration, especially in
view of Williams’ statement that the specimen analyzed by him
was not fresh.t Analysis II calculates out readily as follows :

Nephelite - - - - 4
Diopside - - - - 64
Augite - - - 15
Biotite - - - - S245,
Magnetite - - - - 8.7
Pyrite, = - - - = On
Calcite - - = - 0.6
100.0

In my former paper this pyroxenite was referred somewhat
doubtfully to the jacupirangite of Derby. Through the kind-
ness of this gentleman, to whom I would express here my deep
acknowledgments, I have lately received numerous specimens
of the Brazilian types. A comparison of these with the Magnet
Cove specimens makes it evident that the two occurrences differ
chiefly in size of grain, the Arkansas rock being very coarse,
while those from Brazil are much finer grained. In all other
essential respects the two are closely alike.

From the microscopical examination of the specimens which
Professor Derby sent me, it is evident that the ‘‘Jacupirangites’”’
of Brazil vary from rocks rich in nephelite, and which are true
ijolites, closely analogous to those of Magnet Cove and Finland,
through rocks composed predominantly of pyroxene, with small
and varying amounts of magnetite and nephelite, to types
extremely rich in magnetite and with no nephelite or only traces
of this mineral. Accepting then the name of Jacupirangite for
the medium type, the application of this name to the Magnet
Cove rock is abundantly justified, since the only difference is the
comparatively unimportant one of size of grain, both being
holocrystalline.

* WILLIAMS, of. cit., p. 227.

622 HENRY S. WASHINGTON

That this identity of the two, based on mineralogical grounds,
is correct, is substantiated by a chemical analysis of one of
Derby’s specimens made by myself. For this purpose an
apparently medium specimen was chosen, composed largely of
a violet-brown augite, with some magnetite (more than in the
Arkansas rock) and only a little nephelite (less than in the
other). No biotite was present, and only traces of apatite.
This analysis, given in III of the table, is most remarkably close
to that of the Magnet Cove jacupirangite in all respects, except
the iron oxides. Indeed the figures for silica, magnesia, lime, soda,
water, titanic acid and manganese are close enough to belong to
duplicate analyses of the same specimen, and those for alumina
and potash do not differ greatly. The higher iron oxides are
of course connected with the more abundant magnetite, but,
apart from this, the mineralogical composition is closely similar.

The closest known analogue of these rocks is probably the
pyroxenite of Brandberget, an analysis of which is given in IV
above. The only noteworthy differences are in SiO, and Fe,QOg.
That of the former apparently conditioned the formation of
nephelite in the Magnet Cove and Brazil rocks and plagioclase
at Brandberget, while the higher ferric oxide of II and III is to
be connected partly with the more abundant magnetite in the

former.
Henry S. WASHINGTON.

[ Zo be continued. |

Tibe PRE-(PRRESERIAL HISTORY OF METEORITES:

THE completion of the studies for students relating to the
composition and structure of meteorites, which have recently
been published in this JouRNAL, furnishes an opportunity for me
to record certain deductions which seem to me warranted by the
facts there presented, but which, being largely theoretical, had
best be stated as the expression of individual opinion.

That theories of the origin and cosmic history of meteorites
have been propounded before, and that these have varied widely
in character, the present writer is well aware. These theories
may be mentioned at the outset, together with the names of
those who have given them special support, without, however,
entering into any discussion of the merits of each.

Meteorites have been declared to be (1) terrestrial matter
discharged into space by the volcanoes of the earth and returned
to it again (Sir Robert Ball); (2) matter discharged from
the volcanoes of the moon (La Place, J. Lawrence Smith);
(3) matter ejected from the sun (Sorby); (4) portions of
shattered stars (Meunier); (5) portions of a shattered planet
(Boisse); (6) portions of comets (Newton); (7) clouds of gas
or dust cemented and solidified by the action of the earth’s
atmosphere (Brezina).

All of these hypotheses have been urged by men of eminence,
each urging strong reasons for his views. These reasons can be
learned by study of the original authorities, and the discussion
of them in the present article is not a part of my purpose. I
shall endeavor here simply to present my own views and my
reasons for the same.

The study of meteorites has shown that:

1. The majority of iron meteorites are octahedral.

2. The majority of stone meteorites are chondritic, and contain consider-
able glass.

3. Between iron and stone meteorites there is every gradation—they are
formed of the same sort of matter.

623

624 OLIVER C. FARRINGTON

The above statements would probably not be questioned by
any authorities of the present day. The following, however,
might not be agreed to by all:

4. The substance of meteorites was in a solid state before the fall of these
bodies to the earth.

5. The structure of the majority of meteorites shows that their substance
has cooled from a liquid or semi-liquid condition to that of a solid.

6. The structure of the majority of iron meteorites shows that the change
from a liquid or semi-liquid to a solid state has taken place slowly.

7. The structure of the majority of stone meteorites shows that the change
from a liquid or semi-liquid to the solid state has taken place rapidly.

The four latter statements may then be briefly discussed,
and important known objections to them stated.

Concerning statement 4: It was suggested by writers in the
early part of the last century that meteorites were concretions
formed in our own atmosphere. Brezina inclines to accept this
view with the modification that the substance of meteorites was
extra-terrestrial, but that it arrived at the earth in the shape of
gas or dust and was cemented or solidified by the earth’s atmos-
phere. To my own mind, the slickensided surfaces and veins
exhibited by many meteorites afford sufficient contradiction of
such a view, and compel the conclusion that the matter in which
such structures occur had existed in a solid state for a con-
siderable length of time before it reached the earth.

Concerning statement 5: Several writers, but especially
Daubree,’ have expressed the conviction that the substance of
meteorites gives evidence of having passed directly from a
gaseous or vaporous state to that of a solid. The opinion seems
to be based chiefly on Meunier’s synthetic experiments, in which
he succeeded in reproducing mineral aggregations having the
composition of meteorites and somewhat resembling them in
structure, by the inter-action of vapors.? But, as pointed out
by Cohen,3 the absence of gas and vapor pores in meteorites

«“ Observations sur les conditions qui paraissent avoir preside a la formation des
meteorites,” Comptes Rendus, 1893, CX VI, pp. 345-7.

2Encyclopedie Chimigue, Tome II, “‘ Meteorites,” chap. v.

3 Meteoriten-kunde, Heft I, p. 327.

PRE-TERRESTRIAL HISTORY. OF METEORITES 625

argues against such an origin of their substance, and further,
Fouqué and Lévy produced by cooling from fusion, mineral
aggregates as closely resembling meteorites as those made by
Meunier from vapors. Again, the crystalline structure of the
minerals of meteorites perfectly resembles that. of terrestrial
minerals known to be produced by cooling from fusion.

Concerning statement 6: That the complete crystalline
structure possessed by the great majority of iron meteorites
indicates a lapse of time sufficient for a slow, uniform arrange-
ment of the molecules of their mass, in other words a slow cool-
ing, has rarely beendoubted. Sucha conclusion certainly accords
with all terrestrial experience and observation. It has been
suggested by Cohen,’ however, that the crystalline structure
expressed in iron meteorites by the Widmanstatten figures may
be really a sort of skeleton growth, similar to that seen when
needles of ice form over the surface of rapidly cooling water,
and that hence the Widmanstatten figures may indicate a rapid
crystallization. It is unfortunate that no attempt to reproduce
Widmanstatten figures artificially in iron has ever yet succeeded,
for if this could be done valuable evidence for judgment on this
point could be secured.

Taking the evidence as it stands, however, and especially
taking into consideration iron meteorites like that of La Caille,
whose structural features show a complete parallelism through-
out a large entire mass, the indications seem to me to point
strongly to slow crystallization. Certainly analogies between
the formation of crystals in iron and in water should be drawn
with hesitation. Iron is far more viscous than water and move-
ment in it would take place slowly. Further, the crystalline
plates of meteoritic iron differ in composition, showing that time
must have elapsed for separation of ingredients as has not taken
place in the ice formed upon water.

Concerning statement 7: This opinion is based chiefly on
the large quantity of glass found in most of the chondritic
meteorites, which, it is to be noted, make up by far the larger

* Meteoriten-kunde, Heft I, p. 326.

626 OLIVER C. FARRINGTON

quantity of known stony meteoritic matter. Glass is known to
indicate rapid cooling. Further, the character of the chondri
themselves is such as to lead many students of the subject,
notably Brezina and Wadsworth, tu believe that they are the
result of rapid and arrested crystallization. The fact that
chrysolite, the least fusible and therefore the earliest cooling
mineral, forms the most chondri, lends support to this view. It
must be confessed that the real origin of chondri is as yet very
obscure and the theory above suggested is far from accounting
for many of their peculiarities. Yet the facts above noted seem
to me to argue more strongly in favor of a rapid cooling of the
substance found in such meteorites than a slow one.

If the arguments in favor of the above statements seem
sustained, then the conclusion to which they appear to me to
point is the following: Meteorites are portions of a disrupted mass of
cosmic matter which had a spheroidal form, increased in density
toward the center, and cooled from a liquid or semi-liquid to a solid
state before disruption.

The application of this hypothesis to the subject in hand
may perhaps best be traced by applying it in a reverse order.
Given a defined quantity of liquid or semi-liquid. It will take
the form of a spheroid, since this is the only form known in
which a liquid mass would maintain itself in space. Its mate-
rials would arrange themselves according to density... ihe irony,
for example, would sink to the center, and the slag-like silicates
rise to the surface, as they may daily be seen to do in a blast
furnace, or as a centrifugal separator assorts substances accord-
ing to density. The exterior of the sphere owing to contact
with the cold of space would be cooled with comparative sud-
denness, giving the minerals of the surface a glassy, brittle
character. The protected interior would cool more slowly, giv-
ing the molecules of the metallic center an opportunity to
arrange themselves in an orderly, crystalline fashion. In
time, however, the globe becomes solidified from center to
circumference. During the process of solidification, and later,
many processes of disruption and adjustment go on as the

PRE-TERRESTRIAL HISTORY OF METEORITES 627

result of strains of various kinds, record of which is to be
found in the structure of meteorites. Fissures will be formed,
into some of which pasty metallic matter will be forced from
below, and which will, in its passage upward, enclose angular
fragments of the siliceous crust. Other fissures occurring only in
the siliceous portion of the globe will give rise to the formation of
quantities of angular fragments, which will be cemented together
again by pressure to form breccias. Such fissures would be com-
paratively large, and affect a considerable area of the globe.
Other minor fissures would form in ramifying networks, which
would be filled by adjacent substance penetrating in a more or
less liquid form. Differential movements of solid portions,
without the existence of fissures, would produce slickensided
surfaces. Finally, the progressive disruption of the body occurs.
To produce this, two or three forces may be appealed to. In the
first place, there is the familiar fissuring from shrinking and
contraction as the body passes from the liquid to the solid
state. It is perfectly evident that a certain amount of this is
taking place upon the earth.*. Meunier suggests further, that
in the moon we can see this process extended as much farther
as the moon is more fully cooled than the earth, and he regards
the well-known bright streaks of the moon as enormous fissures
(vainures) showing a progressive disruption of its mass.* While
few probably at the present day would accept this interpretation
of the bright streaks of the moon, there are numerous other
indications that the moon is considerably fissured. Meunier
also points to the asteroids as an illustration of a dismembered
heavenly body.

In the second place, strains corresponding to the tidal strains
of the earth would produce a constant disruptive effect ; and, in
the third place, the recent investigation of Professor Chamberlin,
has shown how the fragmentation of a small body may take
place by near approach to a large one.

Once the body is broken up, its fragments may be drawn out

tSee CHAMBERLIN, “On a Possible Function,” etc., JouR. GEOL., Vol. IX, No. 5.

2 Cours de Géologie Comparée, pp. 258 et seg. 3 Op. cit.

62

oe)

OLIVER C. FARRINGTON

into the form of a somewhat attenuated swarm or cluster. The
possibility of the subsequent capture of single portions of such
a cluster by the earth can hardly be denied. The character of
the portion captured, in respect to its structure, density, and
composition, then, will depend on the position it occupied in
the globe of which it formed a part.

That meteorites exist in such swarms in space seems very
probable from a recent investigation of Hégbom." Plotting the
known meteorite falls according to the days of the year, he has
discovered undoubted and significant groupings. Thus of the
nine known howardites, three fell during the first days of
August, and three during the first half of December. The
probabilities are stated to be several thousand to one against
such an occurrence being a mere coincidence. Again, of the
three known eukrites two fell June 13-15. The chances are said
to be ninety to one that these had a common origin. There
are numerous other groups brought out on the chart so made,
which seem to point to the existence in space of meteorite clus-
ters met on the same date by the earth in its annual revolution.
The probabilities seem sufficiently in favor of the existence of
such clusters to warrant placing some reliance in the constitu-
tion indicated for them. Thus with the group of August howard-
ites previously mentioned are associated one siderite and three
chondrites; with the December howardites, one siderite, one
bustite, one chladnite, and a number of chondrites. The consti-
tution indicated for these swarms resembles therefore, to a
remarkable degree, that called for by the hypothesis here
advocated, especially when it is remembered how exceedingly
fragmentary must be evidence based on the few meteorites seen
to fall. A similar constitution has been also exhibited at times
in the substance of a single meteoric shower. A notable case
is that of Estherville, which contained all gradations, from iron
to stone meteorites.

Two or three other points of evidence may be noted as tend-
ing to show that, in such a globe as that assumed, the substances

t Bull. Geol. Inst. of the University of Upsala, Vol. V., Part 1.

PRE-TERRESTIIAL HISTORY OF METEORITES 629

were arranged in order of their densities and that it had at some
time a hot interior and cold exterior.

The first is drawn from our knowledge of the crystallography
of iron. According to modern metallographists, iron occurs
in three allotropic modifications known as alpha, beta, and
gamma irons. When heated to a temperature not higher than
700° C., iron remains in what is known as the alpha state; from
700 tovso0 G.itassumes: the beta state, and ‘from 860° C. to
the melting point, the gamma state. In cooling, from the melt-
ing point, for instance, the iron does not necessarily return
through these modifications, but remains in the state which it
assumed at the highest temperature... Now gamma iron crys-
tallizes in octahedrons, while alpha and beta iron crystallize in
cubes. The majority of meteoritic iron is plainly octahedral in
structure. It is hard to escape the conclusion, therefore, that it
has been heated to a temperature at least as high as 860° C.,
and, further, that the cubic irons, some of which occur among
meteorites, have been subjected to a somewhat lower degree of
heat. The latter, it is true, must be cooled and reheated to a
somewhat lower temperature than at first to have their structure
accounted for on this theory. But this could quite reasonably
occur, and their relative quantity is so small as to make them of
minor importance.

A second corroborative fact is that the carbonaceous
meteorites are of exceptionally low specific gravity. Now the
carbonaceous meteorites are those which contain hydrocarbons
which could not exist under any high degree of heat. Such
meteorites could not have experienced any sensible heating
subsequent at least to the formation of these hydrocarbons.
But the low density of these meteorites would place them,
according to the hypothesis, on the outer surface of the sphe-
roid, where, after the first solidification from cooling, little
further heating would be encountered.

A third corroborative fact is found in the existence of
diamonds in the iron meteorite of Cafion Diablo, a meteorite

tF. OSMOND, Zhe Metallographist, July, 1900.

630 ; OLIVER C. FARRINGTON

which exists in quantity amounting to several tons. Such
diamonds have been produced by Moissan by heating to a high
temperature iron saturated with carbon and allowing it to cool
under pressure. This is exactly the process through which the
substance of an iron meteorite would pass if formed according
to the above hypothesis. The other meteorite known to contain
diamonds, Nowo-Ure}, is also one which would have formed not
far from the metallic center of such a globe, as it contains large
metallic veins and by some is classed among the iron-stone
meteorites.

The hypothesis outlined above may ask the special attention
of the geologist, on account of the suggestions it may offer
regarding the history of the earth. If it be true that meteorites
are fragments of a broken-up globe, it is not unlikely that they
show to us, to some extent, the constitution of our own globe.
Uniformity of cosmic matter. has been indicated by all studies
of meteorites, as well as by all spectroscopic inquiries into the
chemistry of space. Uniformity of cosmic history seems there-
fore probable also.

I have shown in the studies previously referred to that
meteorites chiefly differ in composition from the crustal terres-
trial rocks with which we are familiar in having an excess of iron,
nickel, and magnesium, and in being practically without free
silica, oxygen, and water. Assuming that the earth, however,
has passed through a history lke that of the globe above
hypothesized, the absence of iron, nickel, and magnesium* from
its crust is explained by the conclusion that they have been car-
ried within its interior by their density. They are therefore
removed from our observation, except as occasional outflows such
as that known in Greenland bring them to view. It is well known
that the density of the earth as a whole requires that its interior
contain matter of higher specific gravity than that with which we
are familiar upon its crust, and it has often been suggested that

*Magnesium is here referred to not as the element, which is relatively light, but

as the essential constituent of chrysolite, which is of high specific gravity and forms
some of the heaviest terrestrial eruptive rocks.

PRE-TERRESTRIAL AISTORY OF METEORITES Ogi

this matter may be iron and other metals. The alternative sup-
position is that the matter of the interior may be like that of the
crust but has become denser through condensation and pressure.

The free silica of the earth’s crust is readily accounted for
if we remember that the-rocks of the earth’s crust have been
worked over, while in meteorites they are seen in their primitive
condition. When silicates are exposed to the action of carbonic
acid for any length of time the bases change to carbonates and
silica is set free.* It seems reasonable to suppose that the vast
amount of calcium and magnesium now held as limestone was
originally in the form of silicates. If the carbonic acid of the
limestones should be withdrawn to the atmosphere, and their
bases combine with the excess of silica of the crust, rocks as
basic as those of meteorites would probably be formed.

Similarly, the lack of oxygen in meteorites may be only
relative, and because much of the matter of which they are
composed was in the interior, deep-seated and protected from
gaseous action. The superficial, lighter siliceous portions of
meteorites are found to be oxidized. It is reasonable to believe
the earth’s substance is not oxidized except for its superficial
crust. It may be urged in support of the view that oxygen
could not have been present where meteorites were formed that
little or no oxygen is found among the free gases obtained from
meteorites. But rocks do not seem to have the power of absorb-
ing and holding oxygen as they do other gases. Terrestrial
rocks do not contain it, although they hold hydrogen, carbon
dioxide, and carbon monoxide in large quantities.* Yet there is
no lack of oxygen in the earth’s atmosphere.

The absence of water from meteorites is an important gap
in the parallelism of constitution of meteorites with that of the
earth. In the gases hydrogen and oxygen, which it has been
shown meteorites possess, a cosmic body has the elements neces-
sary for the formation of water. Conditions of nascence, or
possibly of electricity, might exist in a body the size of the

*This fact is more fully stated by SiR JOHN MurRRAY, Proc. Roy. Soc. of Edin-

burgh, 1890-1, p. 229.

*

? See Studies, JouR. GEOL., Vol. IX, p. 402.

632 OLIVER C. FARRINGTON

earth, which would lead to the formation of water from these
gases, which might not prevail upon a body of smaller size.
The chief reason, however, for the absence of water from
meteorites seems to be the fact that ‘the size of the meteorite
spheroid was probably not sufficient to enable it to hold a
quantity of the free gases competent to the formation of water
or even to retain water vapor if it was once formed. The
spheroid was probably as destitute of an atmosphere as the moon.

I am well aware that an origin of meteorites from a shattered
globe has been suggested before. Boisse was perhaps the first
to do this,* but the idea was especially elaborated by Meunier,?
who reconstructed from the various types of meteorites a hypo-
thetical globe in which these types were arranged largely in
accordance with their density. But the hypotheses of these
authors were based purely on considerations of density, it being
arbitrarily assumed that in a cosmic body substances would
arrange themselves according to density.

It was when a study of the structural characters of meteorites
showed me that the substances “ad apparently arranged them-
selves in the order of their density, and exhibited corresponding
differences of structure, that I was led to form the above hypoth-
esis; and it is perhaps worth remarking that I reached this con-
clusion before I had seen Meunier’s papers on the subject.

It is not likely that the globes or spheroids, such as have
been here hypothesized, were of large size. In the solar system,
for example, there is no indication that the total disruption of a
body anything like in size to a planet has ever taken place.
Such an occurrence would produce effects more catastrophic in
their nature than could be referred to matter so small in relative
quantity as that which has within human experience reached the
earth in the form of meteorites. Such globes would, perhaps, in
their entirety, never come within human observation. But that
there exists in space a vast quantity of fragmental matter beside
that visible to us as stars, nebulz, and the bodies of the solar sys-
tem, there can be little doubt. OLIVER C. FARRINGTON.

* Memotres de la Société des lettres, sciences et arts de ’ Aveyron, Vol. VII, p. 168.

2 Cours de Géologte Comparée.

FIDE TORIAL

WitH this number we present to the readers of the JOURNAL
two able articles on geologic classification and nomenclature.
It is expected that these will be followed by others representing
different points of view. In the opening numbers of the next
volume we hope to offer a series of very carefully matured
articles on the classification and nomenclature of minerals and
rocks, by some of our foremost petrographers. It is hoped that
these discussions will receive the thoughtful consideration of
progressive geologists and petrographers. The importance of
revising our present systems of classification and nomenclature,
if systems they may be called, is equaled only by the impor-
tance of thorough preliminary scrutiny of the proposed substi-
tutes, lest we impose on ourselves new systems scarcely less
infelicitous than the old. A critical and deliberate circumspec-
tion, attended by full discussion, may well be followed by the
adoption of the improved systems by those who have become
convinced of their merits, if full liberty to follow the old
practice is unreservedly accorded to the unconvinced and to
the constitutionally conservative. It is doubtful whether we
should try to force new systems into usage by legislative pro-
cesses, but concerted adoption by those who have the courage
of their convictions will go far towards securing the desired end.

1 C..C.

633

REVIEWS.

Zinc and Lead Region of North Arkansas. By JOHN C. BRANNER
(Arkansas Geological Survey, Annual Report 1892, 396
pp.) Little; Rock, 1901.)

Tue lead and zinc deposits of the Ozark region have received
attention from the geological surveys of Arkansas, Missouri, Kansas,
and the federal government. The United States Geological Survey
and the University Geological Survey of Kansas will shortly have out
reports on the subject. Missouri, through her geological survey, has
already published an exhaustive account of the deposits in two large
volumes, by Mr. Arthur Winslow. After delays of nearly ten years,
Arkansas has at last seen fit to make appropriations for the publication
of the report on the zinc and lead deposits of the north part of the
state. It is by the former state geologist, Dr. J. C. Branner.

The publication of Dr. Branner’s report has long been looked for-
ward to by all interested in the subject of lead and zinc. In many
respects it is the most welcome contribution to our knowledge of the
geology of the Ozark region that has yet been made.

Preliminary to the consideration of the ores is a short description
of the surface relief of the region, illustrated by an excellent photo-
graphic reproduction of Branner’s Relief Model of Arkansas. ‘The
zinc and lead deposits described are located chiefly north of the
Arkansas river. ‘The region here included under the name of Ozark
plateau embraces nearly all of that part of the Ozark mountains within
the state of Arkansas. It includes almost the entire region between
the Arkansas river and the Missouri line, and between the St. Louis,
Iron Mountain & Southern railway and the Indian Territory line.
The Ozark region in Arkansas is made up of three plateaus that rise
like ragged-edged steps one above another, each with a few outliers
standing out upon the next step below.”

In order of their importance, the zinc ores of northern Arkansas
are sphalerite, smithsonite and calamine, besides several other miner-
als of zinc which do not occur in sufficient quantities to entitle them
to be looked upon as ores.

634

REVIEWS 635

The zinc ores are regarded as having been deposited by under-
ground waters. Emphasis is laid on their accumulation along
synclinal troughs and water-way breccias. ‘‘The details of the theory
of the accumulation of the Arkansas ores along synclines and other
water-ways were first suggested by field observations made in this
region in 1889, and the whole theory has been much strengthened by
subsequent work.”

According to their genetic relations there are three kinds of sul-
phide ores: (a) the bedded deposits, which are contemporaneous
with the rocks in which they occur; (4) the veins and other fracture
deposits in which the ores are of later age than the accompanying
beds, and (c) the breccia deposits not formed on fractures, but like-
wise of later age than the accompanying beds. In addition to the
sulphide ores there are carbonate and silicate ores, derived by altera-
tion from the sulphides and forming genetically a fourth class. 5

Regarding the origin of the bedded deposits, it is stated that they
“have originated for the most part where we now find them.” The
cherts are made of silica of organic origin, that is, they were deposited
over the sea bottom as silicious skeletons of diatoms or other micro-
scopic remains of plants or animals. The zinc came from the adjacent
land areas of the period in which these beds were laid down. Upon
entering the sea the zinc-bearing waters had their zinc contents pre-
cipitated in the form of sphalerite or zinc sulphide by the organic
matter that contributed the silica of the chert beds. The zinc crystal-
lized out while these silicious sediments were yet soft and yielding.
In time the sediments hardened and formed the firm, flinty rocks and
pressed closely about the zinc blende crystals. *

“The crystals of zinc blende, however, were not originally as large
as we now find them in the disseminated ores, even where these crys-
tals are no larger than a pin head. They were at first even micro-
scopic, but, as Ostwald has pointed out, there is a tendency in such
cases for the small crystals to pass into solution and to recrystallize
upon the larger ones which grew at the expense of the small ones.
In the bedded deposits this took place before the enclosing sediments
were hardened.”

The vein deposits are those occupying the spaces left by fractures
in the strata. ‘The ores are confined to the fractured zone and to its
immediate walls. When the ore is found in the walls it seldom pene-
trates them to any considerable depth, but is confined to small

636 REVIEWS

fractures that seem to be parts of the great fissures. In appearance
the fissure ores are not different from the bedded deposits. But they
are stated to have a very different origin. The ores of this class have
all been brought into their present position by solution, probably
from the Ordovician bedded deposits.

The question of the origin of the breccia ores ‘“‘has been one of
the most puzzling problems encountered in the zinc regions. The
only theory for these formations that seems tenable is that of the
apparently irregular masses of breccia, that is, the breccias not upon
fault and such like fractures, have been formed along ancient under-
ground water-courses.”’

One of the most suggestive points brought out in this considera-
tion of the zinc ores is the relation of synclines to the presence of ores.
Dr. Branner says: ‘If the hypothetical history here assigned the
north Arkansas zinc ores is thus far correct, we are forced to conclude
that the geologic structure of the region is of the utmost importance
in the determination of the present distribution of the ores. In an
elevated region of approximately horizontal or very gently folded
sediments, the waters falling upon the ground and soaking into the
earth tend to seek the bottoms of the synclinal troughs. The process
of ore accumulation in such a region would therefore tend to carry the
ores into the synclines. ‘The rocks of the zinc region, although not
far from horizontal, are gently folded. Wherever folds have been
exposed in the zinc mines the bottoms and sides of these folds have
been found richer in zinc than the adjacent portions of the same beds.
This is a rule to which I know but few exceptions. The inference
seems to be warrantéd that the synclinal troughs should be located
and examined for the richer zinc accumulations.”

Of exceptional interest at this time are the notes on the faults of
north Arkansas. For the first time in the consideration of the zinc
region something tangible regarding these structures and their char-
acter is made available. The throw of the faults, though never very
great, is sometimes four hundred feet or more. The character of the
folds found in the vicinity of the faults is shown by numerous
figures.

The illustrations are unusually good.
C. R. KEYEs:

REVIEWS 637

Texas Petroleum. By WiLuiaM BatTLe Puivuipes, Ph.D., Director.
The University of Texas Mineral Survey Bulletin No 1;
102 pp., plates, maps. ' Austin, July, 1901.

THE University of Texas Mineral Survey, organized in May, 1901,
with Dr. William B. Phillips of the university as director, establishes
a new record for expeditious work in official geologic investigation by
the timely appearance of this volume on a subject which is attracting
much attention within the state and without.

An historical account of the development of the Texas oil fields is
followed by a chapter on the nature and origin of petroleum, and
other chapters on the oil and gas-bearing formations and the utiliza-
tion of Texas oils.

The Paleozoic formations are not known to hold oil or gas in com-
mercial quantities. The Cretaceous formation, more specifically the
Corsicana field, has furnished practically all of the oil which has been
produced until the current year. This field has a well-defined extent
of from two to three miles in width by six and one-half miles in length
in a northeasterly direction. The oil is reached at a depth of 1,050
feet in a soft, gray, foraminiferal shale. In July, 1901, there were 603
producing wells, with an average daily output of about 3,000 barrels of
oil worth 70 cents per barrel. The production of oil in Texas for
1899 was 669,013 barrels, while that for 1900 was 836,039 barrels,
almost all coming from this field. The Corsicana refinery has a capac-
ity of 1,500 barrels of crude oil daily. Half the output consists of
gasoline and kerosene, the residuum being marketed as fuel.

In the Tertiary, the Nacogdoches field was the first to be discovered,
dating from 1867. The oil is found in Eocene strata at depths of 70
to 150 feet, and is a heavy lubricating oil with a high boiling point
and non-gumming qualities. No oil has been produced in this region
since the early part of 1900.

The Beaumont field has been the center of attraction since January
Io, 1901, when the famous Lucas “gusher”’ was brought in. In July,
1got, there were fourteen producing wells all within an area 1,000 by
2,000 feet on Spindle Top Heights, a low ridge lying about four miles
south of Beaumont. The ridge is about one mile wide and two miles
long in a northeasterly direction and reaches a maximum elevation
of 30 feet above the surrounding prairie. Wells outside the proven
area are dry. It is presumed that the ridge marks an anticline, though

638 REVIEWS

the structure is not yet known certainly. Concerning the fabulous
production of the gushers, no definite figures can yet be given, but the
production is unquestionably large. Pipe lines connect the field with
tide water at Sabine Pass and Port Arthur where refineries are in pro-
cess of construction. In quality the oil is a heavy fuel oil, the price
of which, in July, 1901, varied from 20 to 4o cents per barrel.

In connection with the origin of the oil, an investigation was
made of the so-called “oil ponds,” certain quiet spots in the Gulf
near Sabine Pass and popularly supposed to be caused by oil escaping
from submarine springs. ‘The areas were found to be over extensive
beds of black ooze. The examination of samples of this ooze disclosed
the presence of sulphur and of diatoms containing oil globules, but of
no free oil except such as manifestly came down from the overflow of
the Beaumont wells. Sulphur deposits occur over the Beaumont oil
horizon. The possible analogy of the present conditions to the condi-
tions prevailing when the Beaumont oil-bearing formation was depos-
ited is suggested as well as the possible connection of diatoms with
the oil production.

The report is well printed and illustrated with a number of photo-
graphs of characteristic scenes in the different fields, including several
of the “‘gushers”’ in action, and is in all respects a worthy inauguration

of the new survey.
CaaS:

Lessons in Physical Geography. By CHARLES R. Dryer. Ameri-
can Book Co.: New York, Cincinnati, and Chicago, 1901.

This text-book of high-school grade covers the field of physical
geography from the modern standpoint. It has several characteristics
for which it deserves recognition as more than a mere variation of
what has already been accomplished in other books. It is, first of all,
a very concrete presentation of the physiographic principles which
have recently come into prominence. Thirty-four pages are given to
three typical river systems, the Mississippi, the Colorado, and the St.
Lawrence, and thirteen pages to the drift sheet of North America,
beside the general treatment of glaciers. The chapters which are
devoted to more general subjects also abound in descriptive examples
and pictorial illustrations. In the second place, the book has a large
number of illustrations which are new to text-books; many of these
are drawn from Indiana and neighboring states and will be welcome

REVIEWS 639

to teachers as showing that the modern science of physiography does
not rest on a few classical examples. -The book leaves the very whole-
some impression that the United States abounds in valuable illustra-
tions which are yet unknown to text-book literature, the bringing
forward of which depends largely upon teachers who are working in
their vicinity, just as Mr. Dryer has sought out those of his own state.
The arrangement of the book is good for those who may wish to follow
the author’s own order and convenient for those who do not. The
suggestions on method, both in the “ Realistic Exercises” and in the
appendix, are helpful. The teacher is introduced to many of the val-
uable materials which are now available for this study. A good bibli-
ography is given. The author is plainly in touch with the most
recent work in his science. A few statements or suggestions in geol-
ogy do not take into account some of the recent work. A sentence
on p. 48, despite previous cautious statements, implies the belief that
isostacy alone may support the earth’s broad plateaus. On the same
page the wrinkling of the earth’s crust is ascribed solely to cooling of
the interior. A faulty impression of slate would be left by the men-
tion on p. 32. Under “Causes of Glacial Motion,” three theories are
mentioned: plastic flow, regelation, and alternate melting and freez-
ing. Processes which in recent studies have become more prominent
than these are not mentioned. ‘The too general and perhaps mislead-
ing contrast between the “older drift”? and the “newer drift” are not
due to any lack of information on the author’s part, whose familiarity
with the complexity of drift problems is shown both in this and other
writings. An unfortunate expression on p. 135 would leave the
impression that the Great Basin as a whole is actually a das¢n with a
rim. ‘The mention of “subordinate basins” serves to emphasize this
error. Small shortcomings like those mentioned here are but rare
exceptions in this very good text-book. In its characteristic qualities
the book not only meets the general demands of good science and
good teaching, but is well adapted to the particular needs of the pres-

ent time.
N. M. F.

Some Notes Regarding Vaerdal; The Great Landshp. Dr. Hans
Reuscu. JVorges geologiske Undersogelse; Aarbog for 1goo.

THis complete and well illustrated note, which Dr. Reusch has
summarized in English, recounts a very remarkable landslip. The

640 REVIEWS

level surface into which the river Vaerdal has cut a steep-sided valley
is an upland of stratified marine clays, deposited during a submergence
of the Norway coast since the Glacial Period. Within these clays was
a great mass of “quick clay”’ not constituting a definite stratum but
existing, probably, in more or less definite lenticular masses. A small
side stream, the Follo, had cut a short gorge into the quick clays, giv-
ing the latter an exit to the main valley. On the night of the 19th of
May, 1893, a volume of this semi-fluid clay, estimated at 55 million
cubic meters, escaped into the larger valley, inundating it to the extent
of eight and one half square kilometers. The collapse occupied one
half hour and the advancing front of the mud traveled five or six kilo-
meters in three quarters of an hour. Some of the inhabitants were
rescued from the roof of their house after sailing three and a half
miles on the river of mud. Overa part of the area the upper layer
of clay was firm, and, with the overlying turf, constituted a crust suf-
ficiently strong to remain intact while the quick clay flowed out from
beneath. Parts of fields bearing trees were thus dropped vertically
downward, leaving the trees standing erect at the lower level. The
vertical distance through which the surface fell is not given, but the
pictures represent it as many meters and the sides of the pit as quite
sheer in many places. The author gives a note, also, on a similar
but smaller landslip which occurred on the 16th of August of the same
year in the valley of the small stream Graaelven. The finely banded
marine clays concerned in this slip are made the basis of a time esti-
mate. Their thickness is taken at fifty meters, and they consist of
alternating dark and light layers. On the supposition that one dark
layer and an adjacent light layer were deposited in one year, the time
consumed in their deposition is estimated at 4,000 years. ‘The propor-
tion of post-glacial time which this represents is not estimated.
N. Mook:

Geological Map of West Virginia. Second edition. I. C. WuirTE,
State Geologist. Published by West Virginia Geological
Survey, Morgantown, W. Va.

THE Geological Map of West Virginia, first published in 1899, has
recently been revised and new features added. The map shows in
separate colors the three great coal formations of West Virginia, viz.,
the New River or Pocahontas, the lowest; the Allegheny-Kanawha

REVIEWS 641

series in the middle; and the Monongahela or Pittsburg (Connellsville)
at the top. ‘Two features not shown on the original map have been
added in this edition, viz., the prominent anticlinal lines, and the
locations and names of every coal mine in the state shipping coal by
rail or river, up to July 15, 1901, the approximate locations of the
mines being indicated by numbered black dots, and the corresponding
names and numbers printed on the margin of the map by counties.
The map shows also oil and gas developments of the state, and should
prove of much use to those interested in these subjects. Copies may
be purchased (50 cents) from the West Virginia Geological Survey,
Box 448, Morgantown, W. Va.

RECENT PUBLICATIONS

—American Academy of Arts and Sciences, Proceedings of the. Vol.
XXXVI, No. 29, June, Igol.

—Ami, HENRY M. Knoydart Formation of Nova Scotia. [Bull. of the Geol.
Soc. of America, Vol. XII, pp. 301-312, Pl. XXVI.]_ Rochester, 1901.

—Atti della Academia Olimpica Di Vicenza. Annate 189 9-I900, Vol.
XXXII. Vicenza, Igoo.

—BARTON, GEORGE H. Outline of Elementary Lithology. Boston, 1goo.

—BRANNER, JOHN C. The Zinc- and Lead-Ore Deposits of North Arkansas.
[A paper read before the American Institute of Mining Engineers, at its
Mexican Meeting, November, 1901. ] Author’s edition, Igol.

—BUCKLEY, ERNEST ROBERTSON. The Clays and Clay Industries of Wis-
consin. [ Bulletin No. VII (Part I), Economic Series No. 4, Wisconsin
Geological and Natural History Survey.] Madison, Wis., Igor.

—CALHOUN, F.H.H. Early Conceptions Concerning the Earth and Natural
Objects. [Reprinted from the Bulletin of the American Bureau of
Geography, Vol. II, No. 3, September, Igol.

—CUSHING, H. P. Geology of Rand Hill and Vicinity, Clinton County.
[Reprinted from the Nineteenth Annual Report of the State Geologist ;
University of the State of New York: State Museum,] Albany, Igor.

—DRYER, CHARLES R. Lessons in Physical Geography. American Book
Company, New York, Cincinnati, Chicago, Igol.

—Dupots, Euc. Les Causes Probables du Phénoméne Paléoglaciaire Permo-
Carboniférien dans les Basses Latitudes. [Extrait des Archives Teyler,
Séri II, T. VII, Quatriéme partie.] Haarlem, Igo].

—GLANGEAUD, M. Po. Monographie du Volcan de Gravenoire Pres de
Clermont-Ferrand. [Bulletin des Services de la Carte Géologique de la
France et des Topographies Souterraines. No. 82, Tome XII, 1goo-
Ig01.]| Paris, gol.

—GRANT, ULYSSES SHERMAN. Preliminary Report of the Copper-Bearing
Rocks of Douglas County, Wisconsin. [Wisconsin Geological and
Natural History Survey Bulletin No. 6 (Second Edition); Economic
Series No. 3.] Madison, Wis., {gol.

—HA.Lu, C. W. Keewatin Area of Eastern and Central Minnesota. [ Bull.
of the Geol. Soc. of America, Vol. XII, pp. 343-376, Pls. XXIX-XXXII.]
Rochester, August, IgolI.

642

RECENT PUBLICATIONS 643

—Keweenawan Area of Eastern Minnesota. [Bull. of the Geol, Soc. of
America, Vol. XII, pp. 313-342, Pls. XXVII, XXVIII.] Rochester, |
August, Igol.

—H6Gpom, A. G. Eine Meteorstatistische Studie. Hierzu eine Tabelle und
Tafel IV. [Reprinted from Bull. of the Geol. Instit. of Upsala, No. 9,
Vol. V, Part I, 1900.] Upsala, Igor.

—Kansas Academy of Sciences, Transactions of the. Vol. XVII, 1899-1900.
Topeka, Igol.

—LERoy, Osmond EpGAR. Geology of Rigaud Mountain, Canada. [Bull.
of the Geol. Soc. of Am., Vol. XII, pp. 377-394, Pls. 33-34.] Rochester,
September, Igol.

—List of the published writings of Elkanah Billings, F. G. S., Paleontologist
to the Geological cy of Canada, 1856-1876. prepared by B. E.
Walker, F. G. S., Toronto, Canada. [Reprinted from the Canadian
Record of Scierice; Vol. VIII, No. 6, for July, 1901. issued roth August,
1g9o!.|

—Monr, CHARLES. Plant Life of Alabama. [Reprint of Vol. VI, Contri-
butions from the U.S. National Herbarium, published July 31, Igo1, by
the U. S. Dept. of Agriculture. Prepared in codperation with the
Geological Survey of Alabama. Eugene Allen Smith, State Geologist. |
Alabama Edition, with portrait and biography of the author. Mont-
gomery, Ala., Igol.

—New York Academy of Sciences, Annals of the. Vol. XIV, Part I.
Charles Lane Poor, Editor,

-—PITTMAN, EpDwarRD F. The Mineral Resources of New South Wales.
Geological Survey of New South Wales; Edward F. Pittman, Govern-
ment Geologist.

—PROSSER, CHARLES S. The Paleozoic Formation of Allegany County,
Maryland. [Reprinted from the Journal of Geology. Vol. IX, No. 5,
July-August, tg01.]| The University of Chicago Press.

—READE, T. MELLARD. Sand-Blast of the Shore and its Erosive Effect
on Wood. [| Extracted from the Geological Magazine, N. S., Decade
IV, Vol. VIII, pp. 193- che May, Igo1.] Dulau & Co., 37 Sone Square,
W., London.

—READE, T. MELLARD, AND PHILIP HOLLAND. The Green Slates of the
Lake District, with a Theory of Slate Structure and Slaty Cleavage.
[Reprinted from the Proceedings of the Liverpool Geological Society,
1go0-1go1.] Liverpool, 1got.

—ReuscuH, Hans. Norges Geologiske Underségelse, No. 32. Aarbog for
1900. [With English Summary.] Kristiania, Igor.

—RUSSELL, ISRAEL CooK. Geology and Water Resources of Nez Perces
County, Idaho, Parts I and II. | Bulletins No. 53 and 54, Water-Supply
and Irrigation Papers of the United States Geological Survey.] Wash-
ington, I1gOl.

644 RECENT PUBLICATIONS

-—SHIMEK, B, I. Pyramidula Shimekii (Pils.) Shim. II. The Iowa Pteri-
dophyta. [Excerpt from the Bulletin of the Laboratories of Natural
History, State University of Iowa. Vol. V, pp. 139-170.] April, Igor.

The Loess of Iowa City ,and Vicinity. Iowa Pteridophyta (Con.).
Addenda to the Flora of Lyon County. [Excerpt from the Bulletin of
the Laboratories of Natural History, State University of Iowa. Vol.
V, No. 2, pp. 195-216.] May, Igor.

—StTokes, H. N. On Pyrite and Marcasite. [Bulletin of the United States
Geological Survey, No. 186: Series E, Chemistry and Physics, 35-]
Washington, Igol.

—TuRNER, H. W. The Esmeralda Formation, A Fresh-Water Lake
Deposit in Nevada. With a Description of the Fossil Plants, by F. H.
Knowlton, and of a Fossil Fish, by F. A. Lucas. [Extract from the
Twenty-first Annual Report of the United States Geological Survey,
1899-1900, Part II, General Geology, Economic Geology, Alaska.]|
Washington, Igoo.

—United States Department of Agriculture, Division of Soils, Milton Whit-
ney, Chief. Circular No. 8:Reclamation of Salt Marsh Lands, Thomas
H. Means, Assistant.

Bulletin No. 17: Soil Solutions, Their Nature and Functions, and the
Classification of Alkali Lands; by Frank K. Cameron, Soil Chemist,
Division of Soils (Codperating with the Division of Chemistry).
Washington, Igol.

Bulletin No. 18: Solution Studies of Salts Occurring in Alkali Soils;
by Frank K. Cameron, Lyman J. Briggs, and Atherton Seidell.
Washington, Igol.

—Washington Academy of Sciences, Proceedings of the.
Papers from the Hopkins Stanford Galapagos Expedition, 1898-1899.
II. Entomological Results (2): Diptera; by D. W. Coquillet, Custo-
dian of Diptera, United States National Museum. Vol. III, pp. 371-
379, November 7, Igol.

Papers from the Hopkins Stanford Galapagos Expedition, 1898-1899.
III. Entomological Results (3): Odonata. [Text Figures 29-34.
By Rolla P. Currie, Aid, Division of Insects, United States National
Museum. Vol. III, pp. 381-389, November 7, Igo!.

Papers from the Hopkins Stanford Galapagos Expedition, 1898-1899.
IV. Entomological results (4): Orthoptera. [Text Figures 35-44. |
By Jerome McNeill. Vol. III, pp. 487-506, November 7, Igo!.

Synonomy of the Fish Skeleton. [Plates LXIII-LXV. Text Figures
45-46.] By Edwin Chapin Starks, Leland Stanford Junior University.
Vol. III, pp. 507-539, November 7, Igol.

WEED, W. H., and L. V. Pirsson. Geology of the Shonkin Sag and
Palisade Butte Laccoliths in the Highwood Mountains of Montana.
[From the American Journal of Science, Vol. XII, July, tgot.]

ds Ud:

1OUKN AL OF GEOLOGY

NOVENBER-~DE CEM BER,..7OOL

ih HOV AITE-VOLILE SERIES OF MAGNET COVE:

A CHpMIC AD STUDY IN DIFFERENTIATION,

DISCUSSION.

Ue

In the table below are given the chief molecular ratios of

those analyses of the Magnet Cove rocks which I am led to

believe are the most reliable and representative. There are also

I II Il IV V VI VII

VIII

60.20

Si iyaaercks cs cusrerseyee en tere 38.39 | 38.93] 41.75 | 44.40 | 49.70] 53.09 | 60.13
ETO S staoueatncne otetecrerd irate Ae SAN UeO2Z On 5S) Les Sq Lie sail Onli |e eel 5
ANNO Ren edo ette oars arte aenee Ob Lhe Ala 7 sO4nl 10:95 |) 1On4 5) 20 10)||20..03
(HE ON eee ere tee cia vat aes LAA OeO S| -OnOAee 7 Ali a7 37a se7qal S45
IWS Ooi tais ausystoclnas nhereeete Were On |e See SH7leAse alee ac 7e5ul 262) Ole) eOnt 70
Cal Oran ro dis toasusle sit esete ee 1OVOm | TO.40) 14.57 8.40 | 7.91) 3.201) 007
NBS O) a cae cet aa een nets On7A e527, | Only ih Or50i) 5. 3351" 0.56)|) 76.30
REO s eerie ok Oe75: |) TRF: 632000) “Saclay |) A005) S..42 1) 507
MOLECULAR RATIO
SiO Psy sy ciemiatsctebavees te .640| .649| .696] .740| .828] .885]1.002
WG @ Ftc caterer 2 eat eet NOSSO UO O07 |e Ore, iO 5)|) MOOT) OA!
AL SO guar wens chats raat TOOOs nL S else alo Ol! Sy lots 207i. LOO
(He O)Mae cnnas secu eine 9 lOO, |panl2e yale .082] .102] .052| .048
Mig Oi Ae ye sie cannon - 2200) |) 2Oif polls |e O4d 1.058) 008)" J0LO
Ga@ Orne ce tia eer: .220| 1204) .260| .152| .141|) 3059) .015
INAS OM Not caer syrendee eee 012) |) 7085 | 1008) ~£05)|/ .086| 5117 | : 100
Ke ORF rates mca reel esOOGr ee O00) §0429|) "0571" )..0524| 8 .000)\|", 001
Na,O
es eR En anor aSOm S800 le 2283) ls2T | h.O2i) 1.301) T64
K,O
Specitic Grayitiye ss mre 2,407 | 2.760'|' 3.084 2-770 2.599 |2.557
I. Jacupirangite. V. Covite.

II. Biotite Iolite. VI. Foyaite, Diamond Jo.

III. Ijolite. VII. Pulaskite, Braddock’s quarry.

VI. Arkite. VIII. Pulaskite, Type, Fourche Mt.

‘Continued from p. 622.
Vol. IX, No. 8. 645

646 HENRY S. WASHINGTON

inserted the figures for my two analyses of the Fourche Moun-
tain rocks, which, although somewhat distant from Magnet Cove,
undoubtedly belong to the same general regional magma. To
render the diagrams less complicated the iron oxides are calcu-
lated together as (FeO).

CaO

-600 Teal III IV V VI VIII 1050
DIAGRAM I.

It has already been explained that at Magnet Cove the
arrangement of rocks from center to periphery is regularly serial,
from basic to intermediate, z. e., from II to VI. If, therefore,
the analyses are plotted according to Iddings’s method, using the
silica values for the abscissae, the diagram will represent in a
general way the variation of this particular magma in space.
This was done in my former paper’ for the earlier analyses, and
the results of the present investigation are shown in Diagram I.

1H. S. WASHINGTON, of. c2t., p. 404.

FOVATTEASOLITE SERIES OF MAGNET GOVE 647

Comparison of the two shows that the curves, or rather zig-
zags, are much alike even though the analyses of (and con-
sequently the abscissal positions of the figures for) arkite are
different in the two. At the same time the new diagram is
markedly more broken, and varies less regularly and con-
tinuously than the former. The regularly serial character of
the first is thus apparently diminished, and what it was thought
would be an excellent example of regular differentiation seems
to turn out rather the contrary.

But, as Pirsson justly remarks, the use of SiO, for the
abscissae is arbitrary, and, since this is one of the most impor-
tant rock ingredients, its variation should also be shown in a
manner directly comparable with those of the other oxides.

It would seem to be undeniable that this is a legitimate,
indeed a most logical, method if the differentiated mass has not
suffered disturbance and if circumstances permit the determi-
nation of the correct distances of the various differentiates from
the center, since the diagram then represents, not only the com-
positions of the various phases, but their actual relations in
space, both as to direction and as to relative position.

It often happens, as apparently at Magnet Cove, that the
successive differentiation products are sharply separated from
each other, transition forms being either lacking or very small
in amount as compared with the main types. To correspond
then with the actual state of affairs, the diagram should consist
of steps, z. ¢., horizontal lines of a length equal to the breadth of
each zone, at the respective ordinal positions for each oxide.

Since, however, the analyses may be assumed to represent the
average composition of each differentiation product, and we
desire to study the course and the laws of differentiation, it is
legitimate to represent the position of each constituent by a
point, and the lines connecting these will therefore express the
course of the differentiation of the mass of magma, even though,
as an actual matter of fact, all the possible gradations repre-
sented by the curves may not be present. Such a procedure is
quite in accordance with the general practice in chemical and

648 HENRY S. WASHINGTON

physical research, and this point is mentioned only because such
methods have, as yet, found little application in petrography.

In order to construct the curves two important field data are
required; the position of the center of the mass or area, and the
relative distances from this of the various types analyzed.

By center is meant that of the innermost petrographic zone
or core, not necessarily the geometrical center of the mass, as
this petrographic center may be conceivably geometrically
eccentric. Since, in many cases, as here, we have only one sec-
tion, a horizontal one, this point is not necessarily the center of
the mass, but rather its epicenter, to use the seismological term.

Since the area of Magnet Cove forms a fairly regular ellipse
with axes of about 5 and 3 kilometers, the center of the igneous
area is easily determined. Its position is marked approximately
by the Baptist church,* which lies in the small central exposure
of biotite-ijolite. Inasmuch as we do not know whether the
plane of the present exposed area cuts the mass centrally or
above the center, we cannot tell whether the central point of
this is the true center of differentiation or not. It is probably
not so. But as faras the types exposed are concerned this is of
little moment, as their mutual relations would remain the same,
or approximately so, in any Case.

Having determined the center the next point is to determine
the distances of the various types from this. It is obvious that
for the proper plotting of the curves, and hence the study of the
course of differentiation, this is of great importance, since the
points which determine the various curves will be shifted in one
direction or another according to the distances selected. This
would alter very materially the slope of the curves, and even
their. character or shape, as by the shifting of the abscissal posi-
tions a straight line will become a curve, or a simple curve of the
second degree may assume the form of an inflexed one of the third.

At Magnet Cove we cannot determine the abscissal positions
by simply measuring the distances from the center to the par-
ticular spots where the analyzed specimens were collected,

* Cf. the maps in papers already cited.

FOVAITE-IJOLITE SERIES OF MAGNET COVE 649

because, owing to the elliptical shape of the area, a specimen
(e. g., covite) from near the end of the major axis may be at an
actually greater distance from the center than one (e. g., foyaite)
from the end of the minor axis, though genetically inside the
latter.

It would, of course be best to have several analyses of each
type from different parts of the zones, both radially and circu-
larly, so as to get the mean composition of each. But as that
involves the making of very many analyses, we must be content
at present with selecting what seem to be representative speci-
mens, and assume that their analyses correspond to the average
composition of each type.

Assuming this, two courses are open to us. We can either
measure the distances from the center along a radial line on
which all occur, or average the distances of the various occur-
rences of each type. The latter has been the process adopted
here, since it seemed more likely to eliminate errors due to local
conditions.

For each type measurements were made on Williams’s map in
many directions from the Baptist church to the middle point of
each zone exposed, and the mean of these taken in each case.
On the diagrams the abscissal positions of III, IV, V and VI,
from the origin at II represent these relatively, as it is not nec-
essary that the diagrams be of the same scale as the map. The
position of the foyaite (VI) is not fixed as accurately as those of
the others, since, being at the periphery, it is in great part over-
laid by the surrounding shales. Small outcrops outside the
main area, however, allow a rough estimate of its average dis-
tance, though it undoubtedly extends farther away from the
exposed area than the few outcrops indicate.

In my former paper I assumed that the petrographic center
of the area was in the ‘‘magnetite bed,” and that this was the
result of the decomposition of underlying jacupirangite. As;
however, this is quite uncertain, it seems best for the purpose in
view to disregard this area. For the present then the analysis
(and the diagram position) of the jacupirangite may be neg-

650 HENRY S. WASHINGTON

lected. The same is true of the pulaskite analyses, these rocks
lying quite outside the Magnet Cove area. All these rocks will
be discussed subsequently.

We thus obtain the result shown in Diagram 2, where the
points are connected by straight lines. In this, and the follow-
ing, the vertical scale for SiO, begins .400 lower than the others,
so as to condense the diagram and at the same time preserve

SiO,
GaO
SiO,
Al,O5
Al,O,
MgO
(FeO) Na,O
Na,O K,0
CaO
(FeO)
K,O
MgO
II Ill IV Van nvall

DIAGRAM 2.

the relative forms of all the curves (given later), and not flatten
that of silica, as would be done if a smaller vertical scale were
used for this than for the other constituents.

When Diagram 2 is examined it is clear that, with the
exception of the values for covite (V), all the points of the
respective oxides lie along very smooth curves. Kor con-
venience in further discussion the curves formed by the figures
for II, III, IV and VI are plotted separately in Diagram 3. The
values for V (covite) are entirely omitted from this, and the
position and relationships of this rock will be discussed later on.
The curves marked Fand J will also be explained presently.
All the curves, it may be mentioned, were drawn with a spline,
so that the personal equation is eliminated as far as possible.

FOVAITE-IJOLITE SERIES OF MAGNET COVE 651

Within the limits from II to VI the curves are simple, that
of SiO, alone showing inflexion about at the center of the dia-
gram, rising sharply toward the right (acid end) and falling
gently toward the left (basic end). Most of the curves are quite
flat, especially those of Al,O,, Na,O and K,O, which approx-
imate straight lines. At the same time they are all distinctly

DIAGRAM 3.

curves and not strictly linear.* It will be remembered that
Pirsson’ says that all of the lines of his diagram should prob-
ably be drawn as very flat curves (much flatter than these), and
these observations are in accordance with the conclusion of
Harker,3 that strictly linear series of rocks are of rare occur-
rence.

With the exception of SiO,, which is inflexed and that of

*This is best seen in the large-scale drawing from which the diagram is repro-
duced.

217. Vie PIRSSON, Op C20, ps 5715
3A, HARKER, JouR. GEOL., Vol. VIII, p. 392, 1900.

652 HENRY S. WASHINGTON

Al,O,, which is almost a straight line but slightly convex,’ all
the curves are concave’ (toward the bottom).

The curves divide themselves naturally into two groups,
according to general direction. Those of SiO,, Al,O,, Na,O
and K,O ascend toward the right (the periphery), the three last
almost arithmetically. It will be noticed that the K,O curve
ascends more rapidly than that of Na,O to a point a little to the
right of 1V, when it dropsa trifle more rapidly. This is expressed
in the series of the ratios of these two oxides, already given in
Table I. The second group is that of (FeO), MgO and CaO,
which descend toward the right, and at a greater rate than those
of the other group rise, SiO, excepted.

It may be noted, by the way, that if TiO, be plotted with
SiO,, the curve of the sums of the two becomes rather more flat
(especially about IV), and the inflexion at the left is almost
overcome. For facilitating this observation I have put the
molecular ratios. of Ti@> next to those jot Si@. inethestable,
though it would complicate the diagram unnecessarily to put in
this joint curve. This seems to confirm the general belief that
TiO, plays the part of an acid radical, like SiO,, in rock magmas.

The general results can be concisely shown by plotting the
sums respectively of the ‘‘ascending” and the ‘descending”’
oxides, except silica. We then get the two curves F (that of
Al,O;, Na, O, K,O) and 17 (FeO, Mic OVE2O) on the linenor
which all the determining points fall very exactly. They are
both decidedly concave, “ascending and M/ descending, and their
smoothness and regularity are very striking.

These results are in strong harmony with those of Pirsson,
and the general characters of the curves in each diagram are very
similar, though there are some differences in detail. Thus all
his curves are much flatter, the SiO, does not show signs of
inflexion, and the Al,O, is concave and K,O convex, while in
mine these two are reversed. But these are small matters, pos-
sibly due to the abscissal distances at Yogo Peak not being as

«The terms convex and concave will be understood as referring always to the
axis, the bottom of the diagram.

HOVATTE-TJOLITE SERIES OF MAGNET COVE 653

easily ascertainable asat Magnet Cove. The general conformity
of the two is very good evidence to show that, if plotted accord-
ing to their spacial (2. ¢., genetic) relations in the mass, the
analysis of the components of an igneous complex will furnish
regular curve. This fact is almost proof positive of the view
that the variation of rocks is due to differentiation of some
sort.

Whether this differentiation is always as is now believed,
viz., that the oxides of Al, Na and K tend to segregate in one
direction, while those of Fe, Mg and Ca segregate in another, as
well as the process by which these changes are brought about,
are separate questions, which further investigation must settle.
It may be that the general course indicated by the diagrams of
Pirsson and myself are typical of all rock differentiation, or it
may be that with magmas of different character the course of
differentiation may be radically different, and that the same
oxides do not always tend to go together.

At any rate, it may be confidently expected that where a
mass of magma has been differentiated zz széu and is of approxi-
mately regular shape, has not been subjected to secondary dis-
turbing conditions, and the exposures sufficient, we can express
the relations of the differentiation products and the course of
differentiation mathematically, as has been done in these two
instances. Of course, for this purpose, it is absolutely essential
that the analyzed specimens be representative, and that the
analyses be complete and accurate. Othérwise the curves will
be misleading or else uneven zig-zags, only rough approxima-
tions to the truth, and possibly not even that. There must also
be present at least three differentiates, as otherwise only straight
lines connecting the two can be drawn.

Thefact that the: S10; curvets the only one which is inflexed,
and that it runs very sharply up toward the acid end, leads to
some interesting conclusions. Since, toward the acid end, the
curves of (Fe@); MgO and CaO drop much more rapidly than
those.et Al, OF, Na OLand KO, it. is ‘evident. that’ at.a short
distance to the right (in other words with a slight increase in

654 HENRY S. WASHINGTON

silica), they will practically disappear if the differentiation con-
tinues as indicated by the body of the diagram. Further differ-
entiation in this direction then would lead to the production of
a purely feldspathic or feldspathoidal rock. If continued still
further quartz (free silica) would appear and the rock become
aplitic in character. Finally, since the silica is increasing at a
rapid geometrical rate, while the other constituents are drop-
ping, the extreme result of differentiation in this direction would
be pure quartz.

This inference is obviously in line with the experiments of
Barus and Iddingst which indicated that in igneous magmas
SiO, plays the rdle of electrolytic solvent, analogous to that of
H,O in aqueous solutions. This result is also in harmony with
the experiments and conclusions of Lagorio? and Morozewicz,3
who come to the conclusion that the predominant magmatic sol-
vent is composed of silica and alkalis, and that it has the power of
dissolving large amounts of alumina. The results obtained above
would indicate that alumina itself is an essential constituent of
the solvent, and it would also seem that there need be no stoichio-
metrical ratio between the four constituents. As has been
indicated above, however, it will not do to push conclusions too
far from such meager data, and it is by no means necessary to
infer that a rock solvent of this character is the only possible
one. But it will be as well to defer all discussion of these topics
until more complete data are available.

The comparative: rarity of occurrences of purely or very
highly siliceous igneous rocks may presumably be ascribed to
the fact that long before this phase of differentiation has been
reached, the mass will, in most cases, have become solid (owing
to the high melting point and great viscosity of such mixtures),
and hence incapable of further change in this way.

It is of interest to note in this connection that a specimen
of ‘“‘aplite”’ has been collected by Dr. Weed, and ts now in the

™Barus and IDDINGs, Am. Jour. Sct., Vol. XLIV, p. 248, 1892.

?Lacorio, Min. Pet. Mitth., Vol. VIII, p. 508, 1887.

3 MorOzEWICZ, Min. Pet. Mitth., Vol. XVII, p. 235, 1899.

FOVAITE-IJOLITE SERIES OF MAGNET COVE 655

United States Geological Survey Reference Collection (No. 813).
It is of a small dike, about one and a quarter inches wide, cutting
the shale near Neusch’s gulley, which it has metamorphosed.
I am indebted to Dr. Ransome for the examination and descrip-
tion of this specimen which he sent at my request. ‘‘ Under the
microscope, the dikelet is seen to consist almost wholly of cloudy
alkali-feldspar, with no quartz or nephelite, and a little biotite.
With high power, the feldspar (between crossed nicols) all shows
the fine shadowy striping indicative of a soda-bearing feldspar.”
Two garnets also occur at the borders of the dike. The occur-
rence of this aplite dike is clearly corroborative of the view of
the course of differentiation which has been just expressed, and
it is probable that further search would reveal others which have
heretofore escaped notice.

Turning to the other end of the diagram, there is good
ground for the belief that there must be inflexion upwards of
one or more of the curves beyond II to the left. Ifthe curves
are extrapolated to the left, at a distance, let us say, equal to
that between II and III, the sum of the constituents reduced to
precentages amounts to only 55.6.

It is obvious therefore, either that some other component of
the magma than any of those plotted is greatly concentrated at
the basic end, or else that the curves of one or more of the
plotted constituents must run very sharply upward, thus causing
inflexion.

In the former case a probable additional constituent would
be P,O,, which would yield, with high CaO, MgO and (FeO),
an apatite-rich pyroxenite like that of Ahvenvaara in Finland,"
or with disappearance of SiO,, an apatite-magnetite rock like
that of Alné.? If TiO, should be the constituent to assume
extraordinary proportions toward the basic end, we would expect,
with disappearance of SiO,, titaniferous magnetites, or such rocks
as the magnetite-perofskite rock of Brazil, described by Derby.3

*V. HACKMAN, Bull. Com. Géol. Finl., No. XI, p. 36, 1900.

2 Cf. ROSENBUSCH, Llemente, No. 3, p. 133, 1898.

30. A. DERBY, Weues Jahrb., 1894, Vol. II, p. 297.

656 HENRY S; WASHINGTON

Such products are, however, very exceptional, and are only
to be expected in cases of very complete differentiation. In
general we would only look for sharp upward inflexions of the
(FeO) and MgO curves, which would yield, with the slowly
dropping silica and the high CaO, a pyroxenite rich in magne-
tite. This is just the character of the jacupirangite of Magnet
Cove (i); and of those of Brazil and Alné, and I have indicated
its connection with the others accordingly by the dotted lines to
thevlett:

Inasmuch as the specimens of this come from a small iso-
lated mass outside the main area, its relations to the other types
are uncertain, and its diagrammatic position has been given on
the basis of its silica content. It seems to be probable that if,
as is likely, such a rock is connected genetically with the others,
its abscissal position should be considerably more to the left.
As this rock is met with in the immediate vicinity of the main
area, and is a theoretically possible differentiation product of the
magma, it seems reasonable to assume that the section at Magnet
Cove cuts the mass some distance above the center, and that
below the biotite-ijolite is a core of jacupirangite, as previously
supposed.

It is obvious from the theoretical discussion, as well as from
observations here and at similar regions, that the relations toward
the basic end are far more complex than at the acid end. This
arises from the fact that the oxides involved here are capable of
more numerous mineralogical combinations, and also because
elements which are only present to a small extent in the body
of the magma may here assume proportions of great impor-
tance. The fact that these extreme basic differentiation prod-
ucts are far more common than the purely siliceous ones may
be ascribed to the greater fusibility of magmas of a basic charac-
ter, and the consequent possibility of differentiation among
them at temperatures when the more acid end of the series is
solid.

In this connection attention may be ‘called=to, the waet,
analogous to the segregation of TiO, and P,O, at Magnet Cove

FOYAITE-IJOLITE SERIES OF MAGNET COVE 657

and Brazil, that in large steel castings, such as those for modern
artillery, etc., there is a very marked concentration of ‘‘impuri-
ties,’ as phosphorus and sulphur, toward the center of the mass.

As all the curves are so smooth and well defined, it seems
highly probable that equations for them could be found and that
their properties as such could be discussed. In this way we
could get at an exact knowledge of the law of differentiation,
in this particular case at least. It is a matter of regret that I
am not mathematician enough to do this, but there are other
applications of the data at hand which are capable of simple
mathematical treatment.

Since the area of Magnet Cove is a fairly regular ellipse, and
the zones of the various types are concentric about the center,
by taking the average distance of each we practically reduce the
ellipses to circles, the average distances being the radii.

Now, since II is at the center, if we suppose Diagram 3 to
be revolved about the vertical line at II as an axis, it follows that
the solids of revolution so generated by each of the curves (with
the bounding lines at the sides and bottom), will represent the
amount of each oxide in the original magma, and that their sum
will represent the composition of the magma as a whole, before
differentiation.

This is not strictly true, since we are ignorant of the exact
shape and extent of the complex, but as a first approximation
and an illustration of the method, it will be of interest to calcu-
late the results which are obtained on this basis. Asa matter of
fact, the recent description of the Shonkin Sag laccolith by
Weed and Pirsson* renders it extremely probable that the foyaite
is present in far greater relative amount than the surface expo-
sures indicate. This would necessitate a very considerable cor-
rection, but, as we have no means at present of estimating this,
it will be as well to give the figures based solely on the field |
observations, leaving possible corrections for the future.

The process of calculating the various volumes is very simple
in theory, but somewhat complicated and laborious in practice.

*WEED and Pirsson, Am. Jour. Sct., Vol. XII, p. 1, 1901.

658 HENRY S. WASHINGTON

As it is a somewhat new departure in petrography it may be of
use to others to outlhne the method which I have employed.
The curves shown in Diagram 3 were plotted on paper ruled in
inches and tenths. The particular scale is simply a matter of
convenience, and it is not necessary to reduce the percentages to
100, as we are dealing with relative amounts.

The formula employed is well known, being the second of
Guldin’s theorems, viz., the volume of the solid generated by
the revolution of a closed curve or plane figure about an axis in
its plane, but\ exterior to itself, is equal to the product of the
area of the generating curve into the path described by the cen-
ter of gravity of the revolving area.

Veo77A,
where V is the volume, 7 the distance from the axis to the center
of gravity, and A the area of the plane figure.

The areas of the curve, z. ¢., of the space embraced within the
curve itself, and the limits of the diagram, are easily found,
either by counting the squares, or by calculation of the area of
the trapezoids formed by the respective chords and the limiting
lines, and addition to these of the areas embraced between the
chords and the curves.

The centers of gravity are found by dividing the trapezoids
into two triangles, and finding their centers of gravity, when the
center of gravity of the trapezoid will be at the intersection of
the line connecting the centers of the two triangles and one con-
necting the middle points of the two parallel sides. In the
case of the more curved lines a correction must be made for the
area between the chord and the curve, but this will always be
small. SiO, was regarded as composed of the large rectangle
from .400 below the bottom of the diagram to .249, and the
space between this upper boundary and the inflexed curve.

The resultant volumes, being based on the molecular ratios,
have to be multiplied by the molecular weights of the respective
oxides, in order to arrive at the percentage composition of the
whole. In this way I obtained the following figures, which are
given in full to illustrate the method.

FOVAITE-IJOLITE SERIES OF MAGNET COVE

659

Mot, Ratio
A r Vv VxXmol. wt. | Percentage
Found. | Calc.
: ( 52.0 4. 1306.2 : eo
SOs ape i 630 5-4 203.4 90576.0 47.24 2797 |. 787
EN Oats one 14.0 AZ 375-7 253214 19.99 “190))". 200
(FeO)..... 8.25 265 181.0 13032.0 6.80 .094 | .074
Mig @i acs - 5.9 2.9 107.4 4296.0 2.24 .056 | .029
CAaOiae ene 16.2 3.5 356.1 19941 .6 10.40 a irsyisy |p gata)
INI Oke mae 8.1 Aen 208.6 12933.2 6.75 «100) ||) 010
ISOs oars 8.8 ey] 134.2 12614.8 6.58 .070 | .09g0 |
100.00

Of the Magnet Cove rocks this resembles most that of arkite
(IV) especially as regards Al,O,, (MeO), MeO; and Na, O;
though it is distinctly higher in SiO, and CaO and lower in
K,O. Referring it to Diagram 3,its position established by
means of SiO, is shown at X, and the points where this vertical
istcut by the oxide curves are the “molecular ratios calc.” of
the table.
positions of the various oxides as found.

The small crosses along the vertical indicate the
They can be identified
by the values in the table.

It will be observed by reference to the diagram or to the last
two columns of the table, that in the case of oxides whose curves
are approximately straight lines,as Al,O, and Na,O, the found
and calculated values coincide, while in the case of oxides yield-
ing decided curves the value found is below that calculated.

This is in accordance with the demonstration of Harker’
that if a series be linear the admixture of two or more members
will produce a rock having the composition of a possible
member of the series, while in a curvilinear series the mixture
will not correspond to a possible member.

Another method for arriving at the composition of the
magma as a whole would seem to be furnished by the deter-
mination of the mean point of each of the curves, thus giving
the average composition. If the equations of the various curves
But for
practical purposes it can be done by determining, for each

were known, these could be calculated mathematically.

T HARKER, Jour. GEOL., Vol. VIII, p, 394, 1900.

660 HENRY S. WASHINGTON

oxide, the ordinal value for each successive tenth of an inch, and
taking the mean. The result of this process is given in II
below, that given by the previous process being given in I.

I II

SiO@e see 47.24 45.44
IN KOs s 565.610 19.99 19.06
(ANSON osoce es 6.80 7.75
Mig © errs cee 2.24 3.31
CalOpirne aire 10.40 11.81
Na,O 6.75 6.37
K,O 6.58 6.26

100.00 100.00

The two agree fairly well, and are of the same general char-
acter, though there are marked discrepancies, II being decidedly
more basic in all respects than I. What may be the explanation
of this, 1 am not mathematician enough tosay. But the general
agreement would indicate that one of the two, or their mean,
cannot. be far trom the. truth, 2. 12.,.as neamasithemdatagatanand
permit of approximation.

It is of interest to note that I have been unable to find the
analysis of any rock which agrees at all closely with either of
these two results. Those which are as high in alkalis being
lower in bivalent oxides, while those which agree in this respect
are lower in alkalis and alumina. Whether this indicates that
there are serious sources of error in the method employed, or
else that some undifferentiated magmas may possess chemical
compositions not corresponding to those of rocks as yet known,
is a question which cannot be decided here. It would seem as
if there were nothing a prior contrary to the latter hypothesis.

In this connection Harker’s* remark may be cited: ‘‘ Given
a series such that its diagram has markedly curved lines, the
result of the admixture of two members may be something not
only foreign to the series, but highly peculiar by comparison
with igneous rocks in general.” It is true that Harker was dis-
cussing the case of the mixture of two members of a series, but

HARKER, of. ctt., p. 395.

FOYVAITE-IJOLITE SERIES OF MAGNET COVE 661

differentiation and admixture (of two members of a series) may
to a certain extent be regarded as inverse processes, so that the
occurrence of a magma of this anomalous composition need
not occasion surprise. Being rich in both of the generally
antagonistic groups of oxides, it would be especially liable to
differentiation. The general lability of the monzonitic magmas
as regards the conditions controlling crystallization has been
pointed out elsewhere."

The general chemical composition can also be calculated by
the relative volumes of the various phases, which has been the
only method heretofore available. This would seem to be far
more uncertain than the new method, which is based on the
mathematical course of differentiation, since the ignorance
of certain data may affect the result very seriously. Thus we
cannot tell where the boundaries between two zones really fall,
and (beneath the hornstone ridge especially) whether there
may not be a zone of transitional material.

Assuming that the limits come half way between zones, and
that they are of uniform thickness in all directions, we can easily

Volumes Weights

rate ty : 0.14 0.15

100 ee ea 8.64 9.85

TV eaters : 38.27 - 39.15

Mal csiataceee. 52:95 50.85

100.00 100.00
SiO, - - - - - 50.02
PUN OF oo eee : - = 20,89
(FeO): - - - - - 5.87
MgO - - - - - 1.36
Ga = - - - - 6.89
Na,O - - - - - 6.86
Ke Oo = - - - - 8.10
100.00

*F. L. RANSOME, Am. Jour. Sci., Vol. V, p. 370, 1898. H.S. WASHINGTON,
Jour. GEOL., Vol. V, p. 376, 1897.

662 HENRY S. WASHINGTON

calculate the volumes of the several spherical shells, which must
also be assumed to represent the true ellipsoidal ones. The
results are given below, including the relative volumes and
weights (obtained by correction of the former for specific
gravity), and the average composition deduced from this latter.

This result is notably less basic than the former calculated
from the curves, and approaches somewhat closely to the com-
positions of the foyaite and the arkite, though in a general way
intermediate between the two. This is so, since these two form
(on this basis) 90 per cent. of the whole. It must be remem-
bered, however, that this method is not based on curves, but on
a succession of steps, and that the influence of the greater
width of the more acid phases is intensified by their greater
distance from the center. At the same time both methods
indicate a magma rich in Al,O,, CaO and alkalis, low in SiO,
and MgO, and with moderate iron.

Inasmuch as there must be a (probably rather large) cor-
rection made for the greater mass of foyaite, on the analogy of
the Shonkin Sag laccolith, all these figures can, for the present,
be regarded as only suggestive and illustrative of the method
of investigation proposed, than representing exactly the actual
state of affairs.

It is of course hazardous to theorize on such limited data as
are yet available, but the methods indicated in Pirsson’s paper
and the present one would seem to be of not uncommon appli-
cability, and well worth further trial in the investigation of other
favorable localities indeed: sas kirsson has inemanl<edr-saate
would seem as if this should be the point of departure in the
study of other series.” The methods indicated certainly put
the study of rock differentiation upon a purely mathematical
basis, which in the hands of a competently mathematical petro-
grapher, should surely lead to an exact quantitative knowledge
of the laws which control this, and very probably, with the aid
of physical chemistry, to a knowledge of the rationale of the
process.

t PIRSSON, of. cit, p. 576.

FOYAITE-IJOLITE SERIES OF MAGNET COVE 663

In my former paper I suggested as an explanation of the
exceptional character of the Magnet Cove and Umptek lac-
coliths, in having the borders more acid than the centers, that
the arrangement depended on the general chemical character of
the undifferentiated magma. The process of differentiation was
conceived to be, at least for such small bodies, in great part, a
sort of fractional crystallization, the magma being regarded as
a solution, so that, in accordance with the laws of cooling solu-
tions, the solvent (2. ¢., the portion present in excess) crystal-
lizes out first around the borders on cooling of the mass.

From what has been learned of the composition of the
magma, it is evident that, even though low in silica, it was
originally of a decidedly leucocratic character. In other words,
the potential feldspathic and feldspathoidal constituents pre-
dominated very largely over the calco-ferromagnesian. This is
seen plainly from the relative weights of the spherical shells, but
even the more basic composition derived from the curves shows
the same thing. Thus the composition with 47.24 per cent. of
SiO, may be obtained approximately by several different mix-
tures of all or some of the types analyzed, but in every case it
necessitates taking from six-tenths to eight-tenths of foyaite, or
foyaite and arkite. It seems scarcely necessary to give these
calculations, which are purely empirical. The same composition
may also be reduced to mineralogical composition in several
ways, according to the assumptions made, but here, also, we get
about two-thirds of leucocratic minerals.

The original body of magma, then, at Magnet Cove was,
notwithstanding its low silica, decidedly leucocratic, as demanded
by the theory, so that the alumina and alkalis, with the proper
amount of silica for the formation of feldspar and feldspathoids,
playing the role of solvent, would crystallize first, and hence
form the outer portion of the mass.

The latest paper by Weed and Pirrson, already cited, is of
great interest in this connection. Here it is shown conclusively
that in the well-dissected Shonkin Sag laccolith the outer mel-
anocratic shonkinite is present in enormously greater quantity

664 HENRY S. WASHINGTON

than the core of syenite, which, though basic, is distinctly
leucocratic. The composition of the whole, then, would be
melanocratic, as demanded by the theory, though with notable
amounts of alkalis and alumina. The same general relations
are assumed by analogy for the previously described Square
Butte laccolith, whose magma thus possessed a similar strongly
melanocratic character, z. ¢., with a basic ‘‘solvent’’ portion, as
was suggested.’

In this connection attention may be called to two other
examples of differentiated masses in which the borders are more
acid than the center. One is the igneous area at Aln6,? which
is of special interest, since the rocks of this locality are very
much like those of Magnet Cove. Another example is that of
the Rieserferner massif in the Tyrol as described by Becke.3
The central part of this is a typical tonalite, while the borders
are composed of what is called ‘‘ Randgranit.” Although,
unfortunately, no analyses are given, it is very evident from the
descriptions and from the separations by heavy solutions that
the border rock is decidedly higher in alkalis (especially
potash) and silica than the main central mass.

Another region which offers close analogies in many ways
with that under discussion is that of Ice River, in British
Columbia, the rocks of which, collected by Dr. G. M. Dawson,
have been briefly noticed by A. E. Barlow. As the specimens
were collected on a hasty trip, nothing is as yet known of their
mutual relations in the area, but they form an unbroken series
‘from the most basic ijolite containing 36.988 per cent. of silica,
to ordinary nepheline and sodalite syenites containing 53.638
per cent. of silica.” Through the kindness of Dr. Barlow I
have been able to examine sections of the ijolite, and it is inter-
esting to note that, while closely analogous to the ijolites of

*H. S. WASHINGTON, of. c7t., p. 411.

?HOcBom, Afh. Sver. Geol. Unders., No. 148, 1895. Map Il. This is explained
by Hogbom as due to melting and absorption of the surrounding gneiss.

3F. BEcKE, AZin. Pet. Mitth., Vol. XIII, p. 379, 1893.
4A. E. BARLow, Sczence, N. S., Vol. XI, p. 1022, 1900.

POVAITE-IJOLITE SERIES OF MAGNET COVE 665

Finland and Magnet Cove in most respects, yet that here horn-
blende replaces augite, thus differing from other known occur-
rences of this rock. It seems probable that these rocks are not
as rich in lime as those of Magnet Cove, but higher in MgO and
Fe:

In regard to the possible fourth type of laccolithic differ-
entiation, which was mentioned in my previous paper, namely,
that with a gabbroitic or peridotitic or pyroxenitic composition,
the suggestion may be advanced here that representatives of
this are to be found in the numerous sheets and dikes of diabase,
which, as is well known, seldom show marked differentiation
between the borders and the center. This is in accord with the
view that in these masses the ‘‘basic”’ solvent is present to the
almost total exclusion of the feldspathic portion.

As Pirsson has already pointed out,* the viscosity of the
magma has an important bearing on the form assumed by an
intruded mass. The highly viscous acid magmas will tend to
arch up the overlying strata and form high laccoliths, while the
more fluid, basic magmas do not possess sufficient viscosity to
do this in general, and will hence form relatively thin intruded
sheets. But, at the same time, many of these sheets are of
thickness sufficient to allow of differentiation, if that had been
possible through the composition of the magma.

Such sheets, then, may be regarded as the basic homologues
of the acid, undifferentiated laccoliths of the Mt. Henry type,
differing in form, but like them in that the solvent is largely in
excess in the magma, and hence not susceptible to differentiation.

The abnormality of covite has been briefly noted, and a few
words must be devoted to it before bringing this paper to a
close. It will be observed on reference to Diagram 2 that for
thiserock tle positions, of SiO), Al,O.,..Na,O} and ie Onate
below the corresponding abscissal points of the ‘‘normal”’ curves,
while those of (FeO), MgO, and CaO are above. In other
words, the positions of all the constituents of covite are consist-
ently inversions of what may be called the normal (for Magnet

*L. V. Pirsson, Ezghteenth Ann. Rep. U. S. Geol. Surv., Pt. III, p. 586, 1898.

666 HENRY S. WASHINGTON

Cove); where the curve actually ascends, these would cause it
to descend, and vwice versa.

Three explanations may be advanced for this. The first is,
assuming that the covite is a primary differentiation product like
II, III], 1V, and VI, that the course of differentiation was .not
regular, but subject to comparatively large variations of an
irregular character. This seems to be unreasonable @ priori, and
is also rendered untenable by the great regularity of the curves,
if this type be left out of account, by the fact that the abnor-
malities are systematic in direction, and correspond inversely to
the general characters of the respective ‘‘normal”’ curves, as
well as by the field relations of this rock.

Another explanation is that the covite isa mixed rock, pro-
duced by the combination of either two differentiates of the
magina, or one of its differentiates with foreign rock. That it
cannot have been produced by mixture of arkite and foyaite is
clear from the fact that many of its constituents are not inter-
mediate between those of these two. It might have been pro-
duced by a mixture of foyaite and either ijolite or biotite-ijolite
though its position in the complex militates strongly against
this view. That it is not due to a mixture of the magma, or
parts of it, with the country rock is evident from the composi-
tion of the latter, which is too low in MgO and CaO to form
covite from foyaite, and too poor in alkalis to form it from the
more basic members. Its position in the mass, between the
foyaite and arkite, not on the extreme border, is also adverse.

The last, and most probable, hypothesis is that the covite is
not the result of the primary differentiation which produced the
other types, but of a secondary differentiation of one of the
differentiaties of the primary process. In such a further differ-
entiation we would expect the same oxides to differentiate in
directions like those of the primary process, but in an intensified
degree. This, in the case of the more basic of the two comple-
mentary secondary differentiates, would give rise to just the
abnormalities noted in regard to covite.

Of just what particular phase of the primary differentiation

FOVAITE-IJOLITE SERIES OF MAGNET COVE 667

this secondary differentiate is a product it is a little difficult to
say, but the evidence of Williams’s descriptions, my own obser-
vations, and the mineralogical and chemical data, indicate that
it was the foyaite sub-magna (VI) which has undergone this
further change.

If this view be correct there must exist a complementary
differentiate, a rock in which the positions of the various oxides
are, as regards the ‘‘normal”’ curves, inverse to those of covite.
That is, the loci of silica, alumina, and the alkalis would be
above, and those of iron oxides, magnesia, and lime would be
below, the ‘‘normal.’’

Whether we actually find this rock or not is more or less an
accidental matter of erosion, etc. But a chemical analysis indi-
cates that such exists in the case of the ‘“foyaite’ which
occupies the small area in the northeastern part of the main
mass, which was erroneously colored as ijolite on Williams’s
map.* This is a white, coarse-grained, holocrystalline rock,

I II Ill IV Ia

DIO siarccearaetyers 53.54 53.09 49.70 fetta .892
Al,O, 23.95 Pilg ills) 18.45 23.62 .234
ERC Ole ass aiecein oes ey Lt 1.89 3.39 1.36 .007
Re Opertavevcutcecs nancies: 12 2.04 4.32 1.47 s(Oun'y/
MIG O's avec.s gases 0.08 On32 222 0.33 .002
CaO@Rrasitiocnts = 3 0.71 21.20 7.91 1.51 O12
Nae OP sie teteias savers 8.62 6.86 5.33 8.25 a iete)
Ke OMe it ere cots 8.87 8.42 4.95 8.43 .094
H,O (110°-++) 1.09 Tes 1.09 ene ae
Val KO) (aie) —)) 0.14 0.24 0.25
COP eran istanneye sey: 0.290 0.82 none
NNO inecistreeeeeerneeae trace OnE i 23}
Pi OSes sais oe Op 05 0.40 shee
Other constituents 0.95 site Tee)

99.55 100.48 99.44 100.00

I, Foyaite, northeast part of area.
II. Foyaite, Diamond Jo quarry.
III. Covite, schoolhouse, western part of area.
IV. Mixture of eight parts of I, and one of III.
Ia. Molecular ratios of I.

* Cf. H. S. WASHINGTON, of. cit., p. 394, note.

668 HENRY S. WASHINGTON

composed of alkali feldspar, with nephelite (partly altered to
cancrinite), and rare grains of aegirine-augite. The position of
this, outside the zone of arkite, is where we would look for such
a differentiate of the foyaite magma.

The analysis, made by myself, of this rock is shown in I
above, those of the primary foyaite and the covite in II and III,
the molecular ratios of 1 being given in la, for comparison with
those of the others on a preceding page. It will be seen that
the values obtained for the main Diamond Jo foyaite, which is
quite typical of this rock found around the border, are inter-
mediate in every case between those for the other two rocks.
Reference to Diagram 2 will also make evident the fact that
the positions of the various oxides would be, as regards the
normal curves, exactly the inverse of those of the covite. The
figures for the new analysis fall above or below their respective
curves, where those of covite are below or above. This is just
what would be expected in the case of secondary differentiates,
as has been explained above.

The results of a mixture of eight parts of I and one part of
III are shown in IV. It approximates fairly well to the compo-
sition of the typical foyaite, especially in SiO,, MgO, and K,O,
though Al,O,, and Na,O are considerably higher, and iron
oxides and CaO lower.

It may be of interest to give the calculated mineralogical
composition of I, which works out as follows:

Orthoclase - - - 585253
Nephelite - - - - 36.9
Cancrinite - - - - FON 2ES
Aegirite - = - - 3.2
Diopside - - = : Soe ey)
Extra alumina - - : 1.0

99-9

The possibility of a secondary differentiation taking place
in a primary differentiate is, it must be conceded, difficult to
reconcile with the hypothesis already advanced, since according
to this the differentiation takes place by a successive crystalliza-

FOVAITEJ/SOLITE SERIES OF MAGNET COVE 669

tion, 2. ¢., solidification of the magma, and any such process is,
of course, impossible in a solid mass of rock.

This difficulty, however, does not seem to be insuperable.
The evolution of heat on the solidification of molten lava is a
well-known phenomena, having been actually observed, and the
same has been experimentally verified, and the amount of heat
evolved on the solidification of diabase has been determined by
Barus.?’ Therefore the crystallization of minerals from a molten
magma is an exothermic change.

It is therefore conceivable that the solidification of a lac-
colithic mass may give rise to sufficient heat to remelt por-
tions of it, which might easily remain liquid long enough for a
secondary differentiation to take place. That this was actually
the case at Magnet Cove cannot be definitely shown, but it is at
least an explanation which has a certain degree of probability
in its favor.

In regard to problems of interpolation and extrapolation, by
which Pirsson obtained such close agreement between calculated
and observed chemical composition, we are not in a favorable
position, since the analyses which I have made exhaust the
known main types of abyssal rocks at the locality. A compari-
son of the analyses of the dike rocks given by Williams with
the curves in Diagram 3 confirms the supposition expressed in
my former paper that the tinguaites and the nephelite-porphyry
are aschistic, and that the fourchites and ouachitites like the
covite, are diaschistic.

It was expected that the analyses of the Fourche Mountain
pulaskites would fall in with the curves as laid down. But,
determining the abscissal position by silica, it is found that
Al,O,, Na,O, and K,O are below, and (FeO), MgO and CaO
are above, what would be their ‘‘normal’’ positions, as will be
seen on reference to the diagram.

Although these rocks are unquestionably derived from the
same general magma, yet, as their distance from Magnet Cove
is about forty miles, it is clear that we need not be surprised to

™C. Barus, Am. Jour. Sci., Vol. XLIII, p. 56, 1892.

670 HENRY S. WASHINGTON

find discrepancies, due to differences in the process of differ-
entiation.

At Magnet Cove we have, without any doubt, the results of
the differentiation in place of a small body of magma. This
particular mass may have had originally the composition of the
magma underlying the whole region, or it may have been itself
a differentiate of this. Differentiation in such a large body of
magma as that underlying the whole igneous region of Arkansas
would naturally be likely to give rise to diverse products at
different points, in which, however, could still be traced some
of the original general characters of the whole. We are as yet
scarcely in the position to deal with such intricate and obscure
problems, but the results of Pirsson’s investigations and of those
embodied in the preceding pages seem to furnish a promising

means of attack.
Henry S. WASHINGTON.

PHCURLARY ERPECIS., DUE, TOA LIGHTNING .DIS-
CHARGE ON LAKE CHAMPLAIN IN AUGUST Igoo.

AFTER a period of long continued drought, when the ground
was very dry, a thunder shower arose in the Adirondacks, which
passed east, crossing Lake Champlain over Westport and along
south of the crest of Split Rock Mountain. No rain fell north
of the top of the mountain, but a very severe storm passed to
the south. When the storm had nearly disappeared, a sudden
discharge of lightning passed down from the clouds, striking
about half way down the northern slope of the mountain, entirely
outside of the rain area and into the dry trees and rocks.
In about a half a minute a cloud of what appeared to be dust
could be seen rising from among the pines and juniper bushes.
This, however, in a couple of minutes proved to be smoke, and
in less than five minutes a very well developed forest fire was
under way. Fortunately, a number of persons saw the discharge
and saw the fire start, and immediately hastening to the spot
were able to extinguish the fire before it had burned over more
than a small fraction of an, acre.

The peculiarity of the discharge was immediately observed
upon coming upon the locality. An old pine tree seemed to have
received the most severe part of the discharge and was badly
split in the familiar manner. In addition to this, however, a num-
ber of places were immediately noticed where the lightning had
struck either into the rocks or into the dirt overlying the rock.
In two cases the discharge into the rock was of such force as to
split the rock, tearing up fragments weighing as much as fifty to
one hundred pounds and scattering them about. In other places
the discharge upon the rock was comparatively slight, producing
simply small fractures in the rock, and in some cases the effect
was so slight as to simply remove the dry moss, leaving a
small white spot not as large as the finger-nail. These partial

671

See ENG
Aa ost
67/2 WILLIAM HALLOCK

discharges of such varying force were scattered over an area of
perhaps thirty to forty feet square, the more violent ones being
within twelve or fifteen feet of each other, Upon examining the
point at which any one of these discharges struck, a white incrus-
tation was apparent upon the rock, as if white paint had either
been spattered about or had been spread over as a rough, branch-
ing, straggling line. These white incrustations, in some cases,
could be traced for a foot or more down into the cracks between
the rocks. In other cases, they were mere spots. These white
streaks were, undoubtedly, the paths along which the electricity
ran, and a superficial examination showed that the white was due
to an incipient fusion of the surface of the rock. Unfortunately,
it was not practicable to get satisfactory photographs of these
markings or to bring in large specimens. Small specimens, how-
ever, were brought in, and have been subjected to investigation.

The probable explanation of the scattering discharge of this
particular lightning is to be found in the extreme dryness of the
ground. The cloud charged with electricity would, of course,
induce the opposite kind in the trees and rocks immediately
beneath it. Then, when the discharge came, it was necessary
that each prominence should discharge to the cloud zxdividually,
because the ground connecting the different prominences was
too poor a conductor to rapidly collect the quantities of elec-
tricity and discharge them through a single point, as is usually

the case.
WILLIAM HALLOCK.
COLUMBIA UNIVERSITY, '
New York City.

ASSL Or Tiras uRUCTURE OF PULGURITES.

Some obscure problems in structures occurring within artifi-
cial and natural glasses (rhyolites, tachylites, etc.) receive light
from the study of fulgurites, as representatives of instantaneous
fusion, and frequently, as it will be shown, of equally rapid
devitrification.

Hitherto, however, a constitution of homogeneous glass has
been universally observed, under the microscope, in all sand-
fulgurites, without the least trace of devitrification in the largest
masses.?. Occasional cloudy stains of brown iron-oxide and
black manganese-oxide have been noted, and frequent enclosure
of remnants of quartz-grains* and of bubbles, both more abund-
ant near the outer walls of fulgurite3 In rock-fulgurites, a
single instance of devitrification has been recorded. The results
of examination of four fulgurites will now be described.

1. Fulgurite (lightning-tube) from sand, Poland, ‘a small
fragment, together with thin cross-sections, prepared by Mr.
James Walker, of the New York Microscopical Society. This
fulgurite is of small size, from 5 to 8™™ in diameter, with central
aperture or lumen usually 2.5 to 4™™ in diameter, and glass wall
varying mostly from 0.6 to 2.0™™ in thickness, roughened out-
wardly by adhering sand-grains in a continuous coating. The
photomicrographs (Figs. 1 and 2) may serve to explain certain
features as yet unrecognized in other fulgurites.

The wall presents, under low magnifying power, an appa-
rently homogeneous glass, streaked by occasional cloudy wisps

1A Perfect Glass,” ARAGO (Azz. d. Ch. et. d. Phys., Vol. XIX (1821), p. 290)
and all later investigators.

2Von GUMBEL, Zezts. d. D. geol. Ges., Vol. XXXIV (1882), pp. 647-648.

J. S. DILLER, Am. Jour. Sct., Vol. XXVIII (1884), pp. 252-258.

G. P. MERRILL, Proc. U. S. Nat. Mus. (1886), p. 84.

3Best shown in the longitudinal and cross-sections of a sand-fulgurite by A.
WICHMANN (Zeids. d. D. geol. Ges., Vol. XXXV (1883), Pl. XXVIII).

673

674 An Al OLIEN:

of brownish and reddish iron-stains, and besprinkled with
swarms of bubbles of varying size.

The inner surface of the lumen is for the most part smooth
and shining, with boundary sharply defined in cross-section

Fic. 1.—Sand-Fulgurite, Poland. X10. Photomicrograph of cross-section.

(Bie: 1), as in other fulgurites; but, here and there, a few
points, pustules and needles of glass are found to project, of a
length up to 0.17™. For the origin of the central lumen an
explanation long accepted has been that suggested by Watt for
the hollows in a fulgurite-mass, ‘the expansion of moisture
while the fusion existed.’ In this cross-section the outline is

* Phil. Trans. Roy. Soc. Lond., Vol. LXXX (1790), p. 302.

STUDY OF STRUCTURE OF FULGURITES 675

nearly circular, with maximum and minimum axes as 4: 3; in
another it is still more elliptical. This difference, common in
sand-fulgurites, has been attributed to distortion of the tube by
pressure of the surrounding sand upon the fulgurite while still
plastic. In one cross-section, one side of the tube appears
partly crushed together, with coalescence of opposite parts of
the wall into a blebby mass of coarse bubbles and partial oblit-
eration of the lumen. The characteristics of such a fulgurite
seem therefore to be naturally divided into those developed
during the sudden dilatation of the tube, and those which may
have ensued during its quick compression and in places partial
collapse.

In regard to the distribution of the bubbles or vesicles, three
vaguely marked bands may be distinguished :* a marginal tract,
next the lumen, comparatively free from vesicles and clear: a
middle portion of the wall comprising most of the larger
vesicles; and the gathering of dark swarms of the more minute
vesicles toward the outer margin.

The last (partly shown in Fig. 2) vastly predominate in
number over the larger vesicles, are almost universally spherical,
vary greatly in diameter down to Ip or less, and compose the
dark clouds, seen under low magnifying power, on inner side of
sand-grains adhering to outer side of the wall. Others are also
dispersed more irregularly in lines and bands through patches of
the glass (best shown under magnifying power of at least 300
times). A careful search was made among these dark bubbles,
particularly the most minute, by means of a tenth-inch objective
of good definition, for traces of enclosed water, but no liquid
could be distinguished. From this I conclude that any watery
vapor, derived from moisture present in the sand, has been mostly
expelled in the explosions, and also that this has probably had
far less to do, in expansion of the lumen and formation and
compression of bubbles, than the elastic force evolved by sudden
heating of the large volume of air occupying the interstices of the

tIn a fulgurite from Milton, Florida, no definite order of arrangement occurred.
(MERRILL, /oc. cit, p. 90).

676 A. A. JULIEN

sand-grains. The bubbles should therefore be more properly
denominated air-vesicles (/uft-blasen) than vapor- or steam-cavi-
ties (dampf-poren.)

The larger vesicles, which mostly occupy the middle part of

the tube-wall, seem to have been produced by crowding together

Fic. 2.— The same. X50. Part of wall.

and coalescence of smaller ones. The largest, which have been
perforated and emptied, in making the thin section, present light
outlines, up to 0.2 to 0.3™™ in diameter, mostly elliptical, but
also oval, pearshaped, lenticular, triangular, or squeezed together
in groups, with contiguous walls flattened by pressure in direc-
tion tangential to the tube-wall. Even a particular vesicle may
vary much in size and in form of cross-section at different
depths, as if distorted after formation. The longer axes of the

SHLUDYVAOP SURUCTORE OF PULEGURIFES 677

elongated vesicles are in general disposed radially’ toward the
lumen of the tube, though occasionally one may lie obliquely or
even at right angles to that direction, as if twisted around by a
sudden thrust. All the facts seem to point to strong lateral or
tangential compression of the vesicles within the surrounding
glass wall, and to their consequent extension and distortion in
iieycinection otureliet.*

The sudden expansion of air and vapor by the electric dis-
charge has thus effected the dilatation of the lumen, the genera-
tion of bubbles throughout the fused glass, and the coalescence
of those first formed into the larger vesicles. The relief of this
tension outwardly, toward the margin of the tube, first caused
the radial elongation of the larger vesicles and extension of
conical projections and wings from outer side of the plastic tube,
aided, doubtless, by lateral offshoots of the electric current.
Then, in the moments succeeding the passage of the main cur-
rent, the effects of sudden condensation, lateral contraction and
recoil are shown in the instances of partial collapse of the tube,
in the reverted distortion of vesicles, z. ¢., inwardly toward the
lumen, and in explosion of some bubbles nearest the lumen into
pustules and points of glass over its surface. Where the glass
retained its plasticity longer, the vesicles have recovered their
normal spherical form; this condition seems to have been
regained in nearly all fulgurites whose vesicles have been care-
fully examined by others. It is to be noted that elongated
vesicles in radial position have been recorded in only two cases
of rock-fulgurites, in both within partially devitrified glass, as
found in one by Rutley and in the other by Wichmann; in the
latter case (viz., fulgurite III, beyond) I have made the same

‘A position already recognized in other fulgurites by Wichmann and Rutley.

?One writer considers “this radial arrangement ... . possibly indicative of a
rudimentary crystallization in the fulgurite glass” (RICHARDSON, A@7z. Collector, Vol.
III (1896), p. 132).

3 MERRILL, /oc. cit., 87. One consequence of these offshoots, and of the ozonized
atmosphere thereby developed, is shown in the reddish stains, due to oxidation of
iron, in the sand surrounding a fulgurite, to the distance of 3 or 4 inches; e. g., in that
found near Starczynow, Poland (RoEMER, JV. Jahrd. f. Min., 1876, pp. 33-40).

678 Yl Hdl, SCI GMa

observation, in addition to fulgurites I, I], and IV. Their
exclusive association, therefore, with devitrification may be a
consequence of more sluggish movement and imperfect recovery
from distortion within the more viscous crystallite-laden glass.

Another feature, quite distinct in the vesicles of larger size,
is the separation of each from the surrounding glass by a limpid
pellicle, never exceeding 0.2m in thickness. This appears both
in cross-section, with sharply differentiated outlines, and also in
jagged remnants around the margin of an emptied vesicle, on
surfaces of the thin section, like the edges of a broken eggshell.
It may be considered a glass coating, suddenly chilled and
condensed in contact with the bubble, at first consolidation of
the tube-wall.

Around the outer margin of the wall occurs, as usual in
sand-fulgurites, a continuous row of adhering sand-grains (Fig.
1), semi-fused and to that extent rendered white and opaque.
These grains are partly rounded and from 0.2 to 0.6™™ across.
All were successively examined around the thin cross-section by
the usual optical methods, those of feldspar being generally
recognizable by traces of cleavage, cloudy alteration, oblique
extinction, lower interference-colors than those of quartz, and
more or less complete biaxial interference-figures, whose nega-
tive character could often be distinguished. Out of 35 grains
23 were identified as orthoclase, the remainder as quartz. In
the more angular grains wavy extinction testified to remaining
mechanical stress. The indications are that the original sand
was very fine and highly feldspathic, free from mica and with
few ferruginous particles.

While the outer extremity of a sand-grain, so attached to
the glass wall, often retains its translucency and color entirely
uninjured by the electric discharge, this is sharply divided from
an altered milk-white inner portion, in which the change con-
sists mainly, with quartz, in very minute fracturing and conse-
quent opacity. This passes, with feldspar, into a translucent
border, next the glass, in which minute needles or crystallites
abound, suggesting immediate devitrification after fusion. From

SLODY OF STAOCT ORE OF LOLGORI LES 679

this border very irregular milky threads descend into the glass
in a confused network, swarming with the most minute air-
bubbles, approaching I or 2H in diameter. In this vicinity, also,
fibrous threads or streaks of glass occur, which display, between
crossed nicols, feeble colors of aggregate polarization.

On reception of this fulgurite from Mr. Walker, he stated:
‘IT have noticed in several places on the outside of the tube

“somes very fine threads of fused quartz, like delicate
spider-webs, connecting the sand-grains, as if the partly fused
grains had been forced apart while still soft.” This new feature
in fulgurites I can confirm. The glass threads are colorless,
brownish, or sometimes black; generally smooth and glistening,
but occasionally roughened; more or less curved; passing from
grain to grain of the adhering sand, but sometimes projecting as
ii broken; mostly oi4 to 1.177 in length, and 0.015 to,o:047"
in width.*. In the glass between neighboring sand-grains many
minute round holes are also perceptible* under a low magnify-
ing power, apparently produced by exploded bubbles. In sev-
eral cases, in the glass behind a sand-grain, there is a limpid
band with faintly marked outline (best shown in photomicro-
graph, Fig. 2), the latter corresponding to the back contour of
the sand-grain. This must mark its original position before it
was jerked outward to a distance of half its diameter and the
space behind filled in at once with clear glass, distinct because
free from bubbles. To the force and extent of this outward
jerk, in some instances, the glass threads doubtless owe their
formation.

Between crossed nicols the thin section at first appears dark,
like a homogeneous glass, with the exception of the ring of
highly refracting sand-grains outside and occasional gleams of
reflection from scattered air-vesicles. In ordinary light, under

*A coating of sublimed silica is often found upon the carborundum crystals
manufactured in the electric furnace at Niagara Falls. The interlaced threads of
quartz of which it is composed much exceed the dimensions of the natural threads
above described. They differ also in the variation of diameter in a thread and in the
common occurrence of a peculiar beading along many threads.

2 MERRILL, Joc. cit., p. 84.
p. o4

680 A Alera (CLE TEENN:

low magnifying power, a faint granulation is discernible through-
out the glass. This is resolved under higher power ( X 300)
into an irregular, sparse to abundant distribution of crystallites
through a predominant glassy base, from the margin of the
lumen to the fusion-border of the sand-grains. No special
concentration occurs, except in: occasional richer wisps and
streaks, like diffusion-streams, across minute patches of clearer
glass. In general they are scattered in the same ,;way among
and near to the air-vesicles. In some cases, however, an elon-
gated vesicle is surrounded by a band of perfectly clear glass,
free from microlites, 5 to 10m in width, which broadens to 20 or
30 opposite the ends of the major axis of the vesicle. This
has been plainly due to compression and extension of an
original envelope of viscid glass, chilled and consolidated by
proximity of the vesicle, before devitrification could take place
within this envelope. But in general throughout the glass we
may recognize one condition that has favored crystallization, in
the influence of absorbed vapors, through the abundantly inter-
mixed bubbles.*

These microlites are straight or curved, sometimes lath-
shaped, less than Im in length in some cloudy aggregates, but
largely 3 to 134, oF evenextend into threads ofteny bent ox
_crooked, 30 or 40m in length. They he in all positions and
never display any fluidal arrangement,’ though the larger num-
ber seem to be radially disposed toward the center of the ful-
gurite-tube. Between crossed nicols the microlites exhibit very
feeble double refraction, mostly pale gray, here and there
brightening into pale greenish-white of the first order. All
these forms appear to represent the regeneration of feldspar. A
very few margarites were also distinguished, and one spherulite

« As in the pumiceous glass of the fulgurite of Little Ararat III, as well as in the
devitrified glass of Monte Viso, of which Rutley states that the vesicles are sometimes

so closely packed that the glass is quite spongy. However, the pumiceous glass of
other fulgurites has been found entirely homogeneous.

2In fulgurite-lumps from Florida, Merrill recognized a fluidal structure in the
homogeneous glass near bubbles, “as if by sudden expansion of a steam-bubble in
the plastic material” (doc. czt., p. 87).

SOV OPES LRUGLORE Of PULGORI TES 681

of about 3m diameter, made up of concentric shells, which
exhibited a faint cross, remaining fixed on rotation.

This unique occurrence of devitrification in the glass of a sand-
fulgurite is perhaps mainly connected with the high proportion
of feldspar in the original sand. Such a chemical constitution
must have approached that of the rhyolite* used in the experi-
ments of Barus and Iddings, whose viscosity and whose electric
conduction, when fused, were both found to exceed those of the
less acid rocks, and led to the conclusion that ‘‘electric conduc-
tion increases with the degree of the acidity of the magma. .. . .
Since fusibility decreases in a marked way as the composition
of the magma approaches pure silica, it follows that, in a series
of different magmas, electric conduction at any given tempera-
From this

)

ture increases in proportion as the viscosity increases.’
it must be inferred that the more quartzose sands, when once
melted by a lightning stroke, have offered the viscous medium
of readiest passage to the current.

Exception has been rightly taken, however, to the assump-
tion that the material of a fulgurite-tube was necessarily or
entirely ‘“‘quartz-glass.”? The general distribution of feldspar
in notable quantity through ordinary sands is a fact in favor of
the suggestion of Wichmann and Harting, that the minerals of
comparatively ready fusibility have served as flux and thus
ended in reducing the more refractory. Particularly then in
this fulgurite do we need to consider not only the relative fusi-
bilities of feldspar and quartz, but also the relative degrees of
their solvency in contact with the first formed portion of molten
glass, doubtless so instantaneous in formation as to be independ-
ent of all differences in fusibility, solution, conduction or any
other property of the various sand-grains.

To throw light on this question and on the variations,
recorded beyond, in devitrification of such natural glasses, a

Silica, 75.5 per cent.; alumna, 13.25; soda, 4.76; potassa, 2.85, etc. Heated in
a platinum crucible, it melted at 1500° C., was very viscous (“a stiff paste”) even at
1600°, and quite viscid at 1700°. Am. Jour. Scz., Vol. XLIV (1892), pp. 242-249.

2See v. GUMBEL, Joc. cit, and zdem. Vol. XXXVI (1884), pp. 179-180, and criti-
cism by A. WICHMANN.

682 Av Ae CLEIN:

preliminary examination was made of a specimen of artificially
vitrified gneiss, with brown glassy fracture, obtained from the
hearth of a limekiln at Tuckahoe, N. Y. This had been evi-
dently thoroughly roasted and deprived of moisture but not
fused (as the gneiss structure still perfectly survived, in the
undisturbed parallel biotite-scales), saturated with slag from the
kiln, and finally thrown out and slowly cooled upon the refuse-
dump. Ina thin section under the microscope, the results of
the action of fused glass upon the original minerals of the gneiss
were found to be as follows: a slight solvent attack upon the
biotite, rendering it in part opaque, with brown haloes and
cloudy wisps of iron oxide feathering out into the adjacent
glass, apparently by quiet molecular diffusion; a decided but
partial solution of the quartz, of which a small portion survived
in angular remnants, still in place; and complete solution and
disappearance of the feldspar. The main mass of the saturated
rock had thus been converted into pure colorless glass, with a
few coffee-colored stains, without any trace of crystallites or
new stony matter, notwithstanding its slow cooling. However,
globules of clear glass, about 0.5™™ in diameter, were abundantly
interspersed, colorless like their matrix and with little adherence,
since many had become loosened and removed from the surface
of the thin section during the process of grinding. It was con-
sequently inferred :

1. That minerals of acidic constitution, feldspar and quartz,
may be largely carried into solution, in contact with fused glass,
at a temperature below their thermal melting points.

2. That their fused products tend to cohere zm situ in sphe-
roidal globules, through possession of a different degree of
density, viscosity and contraction on cooling, from that of the
surrounding glass.

3. That the process of devitrification has been prevented in
this vitrified gneiss by the peculiar constitution, probably the
acidity, of the unsaturated glass, and by an insufficient period of
cooling.

In the development of a sand-fulgurite byalmost instantaneous

SHOUD OF STRUCTURE OF FULGURITES 683

fusion, it is unlikely that any selective power could have been
exerted among the sand-grains, through their slight variations
in fusibility, conductivity or other property. All within a radius
of a few millimeters were suddenly and completely fused, with
a limitation so sharply defined, in this fulgurite, that, in any
sand-grain, the quartz or feldspar may remain entirely unaltered
within an interval of about 0.02™" of the same material fused
and again devitrified. The molten mass of its wall became a
supersaturated solution of feldspar, with specially strong sol-
vency, however, toward any external quartz-grains within its
reach, through the tendency to form more acid silicate. Such
action probably accounts for the increased amount of silica
which has been determined in fulgurite-glass over that in the
original sand in several instances.’ In this fulgurite also the
preponderance of feldspar over quartz in the attached sand-
grains may possibly have become exaggerated by the slower
solution of the former mineral.

The view has been advanced? that a selective power of fusion
has been exerted by the electric current among the grains, but
without regard to their fusibility : the poorer conductors (quartz-
grains), offering so strong resistance as to become heated to the
point of fusion, those substances which are the best conductors
(iron-oxides and feldspars) escaping with least injury. It is
notable, however, that it is not a good conductor but quartz only
which has ever been recognized in remnants of grains inclosed
in the glass of other sand-fulgurites, and that no remnants what-
ever are enclosed within the glass of this fulgurite, apparently
because so rich in flux.

Three rock-fulgurites will now be described in the order of
increasing devitrification.

‘In a fulgurite from Union Grove, IIL, silica 91.66 per cent. in the glass to 84.83

in the sand; in another, 95.91 in the glass, to about 90 in the sand. (MEKRILL,
loc. cit., p. 85.)

2MERRILL, Joc. ci¢., p. 87. Diller notes, however, in a fulgurite on hypersthene-
basalt, that the sequence of alteration is according to degree of fusibility of the com-
ponents of the rock, greatest in the groundmass, then hypersthene, then feldspar, and
least in olivine.

684 YA, Aly, IROL

Il. Fulgurite on gneiss, near Split Rock, Lake Champlain, New
York; specimen collected by Professor William Hallock, of
Columbia University, immediately after the lightning-stroke, as
described in his accompanying article. Thin sections of this
granite-gneiss were prepared, one in plane parallel to and within
about 0.03™™ of the fulguritic crust, the other in cross-section.
The rock is found to present a surface somewhat dulled by
weathering, and to consist, to about 70 per cent. of its volume,
of feldspars, microcline, microperthitic orthoclase and a little
plagioclase, besides much quartz and a small amount of mag-
netite, biotite, garnet, and rutile, but no trace of fulguritic glass.
No undulatory extinction occurs between the nicols, as evidence
of strain.

The fulgurite-pellicle is entirely superficial,’ brittle, and with
slight adherence to the rock, from’ 0:1 to ©.3"™ im thickness;
nearly white, consisting chiefly of a minutely blebby hyaline
crust with blistered, glistening surface, whose bubbles, rarely
exceeding 0.2™™ in diameter, are nearly spherical but somewhat
elongated normally to the surface of the rock. On careful
examination of these, in a flake mounted in balsam, under a
good tenth-inch immersion objective, no water of condensation
could be detected in any vesicle.

Many short glass-fibers occur, sometimes with globular ends,
resembling some of those in ‘“‘Pele’s Hair” from the crater of
Kilauea, Hawaii. This white pumiceous slag is quite uniformly
attached to the predominant surfaces of feldspar, though pierced
by the uncovered gray grains of quartz. It is dispersed over
the rather even surface of the hard gneiss in radial, somewhat
dendritic forms, often many inches in length, thinning out into
feathéry fleeces at the margins. Wherever these pass over
minute plates and seams of the dark minerals of the Trock,-the
blebby fulgurite assume a yellow, red, brown, or black COlOr, AS
a dark enamel, or in the form of tiny spherules.?

* Similar superficial films of fulgurite have been often observed, by Humboldt in
Mexico, by Brun in the Alps, by Ramond on mica-schist, on limestone, and on phon-
olite in the Pyrenees and Auvergne.

2 Over the fulgurite-crust on hornblende-gneiss at Mount Blanc distinct white and

Sie Oe STO e LURE Of 1rULG ORL» 685

The white material, when flaked off and mounted in Canada
balsam, reveals a limpid glass, dispersed with very few and
minute straight microlites and very rarely rhombic plates; the
latter display rather bright interference-colors between crossed
nicols and parallel extinction.t Where thinnest, the glass
remains isotrope and dark, while the thicker portion, generally
toward the center of the flake, glows with the greenish-white of
the first order of Newton’s colors; in places the color reaches
the sky-blue of the second order, exactly like a thin scale of
the underlying feldspar. The indications are that devitrification
had begun, throughout the glass, in globulites too extremely
minute for distinction, even under high magnifying power, but
whose concentrated effect in depolarization becomes visible,
between the crossed nicols, in the thicker parts of the flakes.
The flakes of brown or colored glass, however, are found to be
uniformly isotrope. It should be noted that, as the greater
thickness of these flakes approaches or exceeds 0.5™™, it is
probable that the presence of crystallites might hardly have
been recognized in a ground section of the ordinary thinness,
0.02 to 0.05™™. In such an investigation plane surfaces are not
indispensable, and the examination of splinters of a glass may
serve an important office. Evidences of incipient devitrification,
in fulgurites and other glasses, have possibly escaped detection
by observers relying entirely on examination of thin sections.

The facts above described lead to the following conclusions:

1. We have here to do with different conditions from those
dark globules were dispersed: “the fused surface of each crystal solidified almost
exactly zz situ, except where sputtering of the molten matter was caused” (RUTLEY,

Quar. Jour. Geol. Soc. (1885), pp. 152-156). The inclosed globules in the slag-
saturated gneiss of Tuckahoe may represent a similar tendency to isolation.

™The homogeneity of glass which apparently prevailed in all fulgurites examined
by the early observers led naturally to the statement in 1884 that the absence of crys-
tallites ‘may be used as a means of distinguishing fulgurite from other natural
glasses” (J. S. DILLER, Am. Jour. Sci., Vol. XXVIII (1884), pp. 252-258). In 1889,
Rutley made the first and hitherto the only record of the occurrence of devitrifica-
toin in a vesicular fulgurite-glass on glaucophane-schist, at Monte Viso, Cottian Alps.
The crystalline forms consisted of globulites, margarites, and longulites, and even
symmetrical microliths (Quar. Jour. Geol. Soc., Vol. XLV (1889), pp. 60-66).

686 AA COLETTE:

which attended the sand-fulgurite I, viz., the lessening force of a
divided electric current, passing over the surface of a compact
rock-mass. The effect of the electric action on this solid sur-
face of gneiss has been more diffuse, feeble, and superficial than
in the sand-fulgurite, and, in this case, confined to the feldspar:

2. The fulgurite crust has been produced almost entirely by
fusion of surfaces of feldspar-grains rather than of quartz, and
in small degree by that of the iron-containing minerals. The
property of fusibility, rather than that of imperfect conduction,
has determined the amount of attack on each grain. Each kind
of glass, colorless and colored, is sharply confined to the min-
eral surfaces from whose fusion it originated, with little tendency
to intermixture.

3. The bubbles throughout the fulgurite-crust owe their for-
mation chiefly to expansion of air, and in part doubtless of
steam, derived from moisture in the weathered surface of the
rock. Their sputtering explosion has probably produced the
tiny fibers over the blebby surface.

4. The surprising partial devitrification, which has instanta-
neously followed fusion throughout the delicate white pellicle,
has been evidently facilitated by the supersaturated feldspar
solution of which the molten glass almost entirely consisted,
and by its consequent unstable molecular condition.

III. Aulgurite in augite-andesite, summit of Lesser Ararat,
Armenia. The specimen was one of those collected by Dr. E.
O. Hovey, of the American Museum of Natural History, New
York.

_ The fulgurite material from this peak has received repeated
study in the field or laboratory by successive observers. Abich*
detected only pure glass in the fulgurite, but in such abundance
that he suggested the name ‘fulgurite-andesite’’ for the mate-
rial on the apex of the peak. Gustav Rose? also came to the
same conclusion, and determined the difficult fusibility of the

* Switz. Akad. Wiss. Wien, Vol. LX (1870), I. Abth., pp. 153-161.
2 Zeit. d. D. geol. Ges., Vol. XXV (1873), pp. 112, 113.

SHUDY OF STRUCTURE OF FULGUN TES. 687

glass on thin edges. A. Wichmann’ made the most careful
petrographic study of this and other fulgurites and distinguished

d

this as ‘entirely homogeneous glass,” without any alteration of
the andesite at the sharply defined contact. He considered the
bubbles as vapor-cavities (dampf-poren), and gave a clear repre-
sentation of the structure of a lightning-tube in a well-drawn
cross-section. The results of my own examination, which follow,
differ in some important particulars.

The augite-andesite of this peak, though consisting of fresh
plagioclase-feldspar, with some partly decayed augite, horn-
blende, orthoclase, and magnetite, is deeply disintegrated by
weathering, so that, in some blocks, its grains crumble readily
under pressure from one’s fingers. It is also traversed, even in
the most solid material, by numerous cavities and channels, a
few millimeters in width, of most irregular form, crossing and
connecting with each other at intervals of a few centimeters.
Most of these passages, in the more decayed specimens I have
examined, in the collection of Dr. Hovey, are one or two centi-
meters in width, and often coated by the olive-green fulgurite
glass to a depth of 1 to 1.5™™; some are even solidly filled up
by the glass, with a cross-section of 5 to 8™™ in diameter or
over. Other cavities of exactly the same form and size were
noticed, adjoining but not connecting, which are now and appear
always to have been entirely free from fulgurite. Most, if not
all of these, therefore, seem to be preéxisting cavities, later
occupied in many cases by fulgurite. A confirmation of this is
found, by optical examination of the thin sections, in the absence
of undulatory extinction in the grains of minerals adjacent to
the fulgurite, z. ¢., the lack of any indication of strain likely to
have resulted from actual perforation of the rock by lightning.
On the other hand, certain other specimens in the collection
from the same peak are penetrated by cylindrical winding tubes,’

t Jdem., Vol. XXXV (1883), pp. 849-859.

2The approximately cylindrical form of the latter, and their curved windings,
have been well represented by Rutley, from the fulgurite furrows on glaucophane-
schist at Monte Viso, Cottian Alps (/oc. cz¢., plate).

688 AA GLEN

lined or filled with the dark glass, which seem clearly to owe
their perforation to the action of electric discharges upon more
disintegrated blocks of the crumbling rock.

The glass is found to be rich in bubbles, especially toward
and near the surface of the inner rock-wall, and in places even
blown up into blebs and blisters. All of the smaller vesicles and
most of the larger are approximately spherical. In none of
these could any water of condensation be discovered under a
tenth-inch objective; their contents appear to be entirely gas-
eous, 2. é., chiefly derived, in all probability, from expansion of
air rather than steam. Many of the larger vesicles present a
more or less elongated form, with major axis always arranged in
position at right angles to the adjacent rock-wall. This distor-
tion of vesicles in the glass of a rock-fulgurite may be always
attributed, in my opinion, entirely to the reaction of pressure
inward, z. e., toward the lumen, which has instantaneously fol-
lowed the passage of the electric discharge, mainly from expan-
sion of the collected bubbles, partly from increased volume of
the glass during fusion of the rock.

When crushed splinters of the glass are examined, though
found sometimes entirely amorphous, as reported by Wichmann
and others, it is also in places rich in bunches of stony matter,
isolated needles (with bright interference-colors between crossed
nicols and parallel extinction) and clusters of. microlitic fibers,
sometimes radiating around a bubble.

Again, at the line of contact between this glass and the ande-
site-wall, an intermediate stony layer occurs, 0.15 to 0.70™™ in
thickness, containing a few obscure outlines of bubbles. This
plainly consists of wholly devitrified glass, made up of a felt of
irregularly crossing needles, fibrous curved wisps, or straight
bundles, some rectangular in form, 0.08 to 0.207” in length,
whose axes lie mostly parallel to the line of contact with the
adjacent rock-wall. In the straight bundles an extinction-angle
of about 19° is uniformly obtained, often with undulatory phase,
and all offer the characteristics of orthoclase. There is a remark-
able resemblance, if not identity, of these fibrous aggregates, in

SUCDIY ROTA STROCT ORE OF FPULGOURITES 68g

form, structure, and optical character, to the artificial microlitic
bundles obtained in the experiments of J. S. Diller (loc. cit.)
by long fusion of amorphous fulgurite-glass in crucibles.

The difference of the above results from those hitherto
reported from study of the fulgurite of this peak may be
attributable to variation in the structure of the fulgurite, particu-
larly in regard to devitrification, in different parts of the surface.

IV. fulgurite from summit of Central Butte, Little Missouri
Buttes, Wyoming; specimen collected by Mr. John D. Irving.
This is a small fragment of phonolite, apparently from the edge
of intersection between two joints, down which corner the fulgu-
rite runs in a shining, cream-colored, slightly brownish crust,
about 157" wide and 3™™ thick. Ona fractured cross-section it
shows distinct lamination parallel to the surface of contact, as if
the latent structure had been developed by weathering.

The rock consists mainly of orthoclase-phenocrysts, whose
idiomorphic character is obscured by fractures, imbedded in a
smaller volume of granular, somewhat ochreous holocrystalline
groundmass, through which needles of hornblende, plagioclase,
apatite, and granules of magnetite are dispersed. Between the
crossed nicols, the transparent minerals, in a thin section, gener-
ally display a decided undulatory extinction, in evidence of con-
dition of strain.

This fulgurite exhibits under the microscope the unique char-
acter of a wholly devitrified mass, light brownish-gray, delicately
laminated in cross-section (a, Fig. 3) by a very irregular flow-
structure,’ with interruptions and intersections of lamine sug-
gesting those observed in the cross-stratification of certain sand
deposits. Some laminae, especially near the contact-line (e. g.,
just below c’), are bent and faulted, with ends slipped past each
other. The thicker lamine consist of very minute, colorless

*Compare the thin underlying coat, in which fusion is less complete, with dark

fluidal banding, and which envelops numerous crystal remnants, described by Diller

in the “mixed zone” of fulgurite-glass in immediate contact with basalt at Mt.
Thielson (Joc. ci¢.). Of this he states, it is difficult to conceive how it has been pro-

duced, ‘‘unless it is due to the repulsion of the particles among themselves.’

690 AL A UL IEN,

needles, fibers, and granules, in an irregular felted mass, through
which obscure circular and elliptical outlines (0, etc.) indicate
the position of numerous original bubbles, now flattened, freed
from gas, and even partly obliterated. With this darker
material, evidently entirely devitrified glass, there occur frequent

Fic. 3.— Rock-Fulgurite, Wyoming. X50. Cross-section of contact.

alternations (d, d, eles) of thin films of a microscopic breccia or
‘““mylonite,” shining brightly between the crossed nicols. This
is composed apparently of rock-dust,* made up chiefly of angu-
lar splinters of feldspar; a few coarser fragments of the same
are also irregularly interspersed (f). Along the line of contact
(d@’), the same angular dust is drawn out in streaks or gathered

™ Apparently related to the opaque whitish layer, between the glass and rock, in
the fulgurite of Monte Viso, and referred by Rutley to altered titanite.

SLOUDVIOR STRUCTORE OF FULGURITES. 691

in embayments. Its origin is well shown next the Carlsbad-
twin of orthoclase (c—c’), whose outer end (c’), projecting into
the fulgurite current, has been shattered into this dust to a
depth of 0.1™™, seemingly after the adjoining fulgurite lamine
had somewhat consolidated. This and other feldspar-grains
lying along the edge of the stream, have been completely
encircled by fulgurite, with an inner film of angular dust, mark-
ing the outline of the minutely-shattered surface of feldspar.
This micro-breccia is brought out between the crossed nicols in
distinct bright stripes. The phonolite groundmass generally
shows rather deeper embayments by fulgurite-attack than the
feldspar-grains ; but the hornblende-needles seem to lie entirely
unaffected, even in immediate contact (%) with the fulgurite-
stream.

It is possible that the devitrified crust which constitutes. this
fulgurite may be a remnant of an inner stony layer (as in Ful-
gurite III), from which a glass-coating may have been scaled
off during weathering; but of the latter there is no evidence,
and the stony crust now occurs coated by small lichens. Yet
the peculiar structure of this crust, with its alternations of
micro-crystalline and of micro-brecciated lamine, signifies that
devitrification has not been a secondary process due to weather-
ing of a glass.

RESUME.

Considerable variation appears, in the characteristics of these
rock-fulgurites, from two probable causes: the variations in the
electric current, in regard to volume, intensity, duration, and a
probable series of successive discharges, in some cases, within
the same fulgurite (IV); and the difference of the rock-material
in the three specimens, gneiss, andesite, and phonolite.

In regard to the duration of a lightning-flash, it is under-
stood not to have usually exceeded one hundred thousandth of
a second (Sylvanus Thompson). The discharge is known to be
oscillatory, surging back and forth, and even in many cases, like
that observed by Professor Hallock, apparently persistent and
continuous for some moments. Nevertheless, allowing for the

692 Aly Aly OTLB
@
long retention of the image of the flash upon the retina, its

duration is rarely likely to have much exceeded the above
quoted minute interval.

The results of the electric action are of two classes, super-
ficial and internal.

The superficial consist, in succession, of fracture of outer
pellicle of exposed minerals: fusion zm széu of the surface of
more fusible grains: and development of bubbles in the enamel
of molten glass from expansion of moisture and air, with fre-
quent explosions and spattering of glass globules.

The internal results comprise the penetration of preéxisting
cavities of the rock, and often the actual piercing of surface of
outcrop or blocks, if softened and disintegrated by weathering,
by lightning-tubes of symmetrical cross-section, as well as pas-
sage downward through fissures and joints.

This has been accompanied by fracture and fusion of pel-
licles over the surfaces thus traversed, especially of the more
fusible minerals and groundmass ; development of air- and vapor-
bubbles in the molten coating; and sometimes, at least, a suc-
cession of waves, sweeping downward fused glass and rock-dust,
in a series of alternating films, the oldest nearest the rock-wall
and subjected to frequent dislocation.

At the moment of cessation of the electric current, there
have followed reaction from the intense pressure, expansion of
vesicles, and regurgitation of glass upon the surface of the rock
from some filled hghtning tubes;* explosion and disappearance
of bubbles nearest the surface, distortion of many by inward
relief of pressure, or general recovery of spherical form; and
partial devitrification of the glass, occasionally complete, where
certain favorable conditions have prevailed— probably the pres-
ence of moist air in intermixed bubbles, a supersaturated neutral
glass, and an over-heated or slowly-cooling contiguous sheet of
rock.

The instantaneous regeneration of feldspar only, in the form

‘First recorded by HUMBOLDT, in rock-fulgurite on volcano of Nevada de
Toluca, Mexico (Azz. d. Ch. e¢ d. Phys, Vol. XIX (1821), p. 299.)

SLODY OF STROUCTORE OR FULGURITES. 693

of microlites, in every instance, in the series I have examined,
seems at first somewhat at variance with the conclusions of
Lagorio as to the particularly difficult saturation of a molten
silicate by the oxides represented in that mineral. But we have
here, apparently, a far different magma from that employed in
the experiments of Pelouze, artificial glass, 7. ¢., from every evi-
dence, a supersaturated solution of feldspar in its own fused
material. In such a magma-solution, a combination of silica,
alumina, and alkali molecules, must find conditions highly favor-
ing ease and rapidity in re-crystallization.
ALExIs A, JULIEN.

COLUMBIA UNIVERSITY,
New York City.

EXPLANATION OF FIGURES

fig. 1. Sand-Fulgurite, Poland.— Photomicrograph of cross-section,
showing sharply defined outline of central lumen, the glass wall with its
vesicles, and ring of dark semi-fused sand-grains, adhering to the irregular
outer boundary of the fulgurite. X1o.

fig. 2. The same.— Photomicrograph of section of part of the wall,
showing forms, radial distension, and distribution of the vesicles, the smaller
and darker ones still occupied by air. Several of the sand-grains appear
along the outer boundary, each with its inner crescent of limpid glass. The
inner dark part of a grain indicates the side which has been partly fused ;
the outer clear part, that which remains unaltered. X50.

fig. 3. Rock-Fulgurite, Central Butte, Wyoming.— Drawing of cross-
section, showing line of contact (@', #, c’, e) between phonolite, Z, and
fulgurite, a.

6. Obscure flattened vesicles.

c-c'. Carlsbad twin of orthoclase, with shattered ends, surrounded by
feldspar-dust and devitrified glass, enclosing a few bubbles.

ad, d, da. Films of rock-dust, alternating with lamine of devitrified glass.

@'. Film of rock-dust along the contact.

é. Projecting corner of feldspar-grain, shattered by the electric current.

J, etc. Feldspar fragments scattered through the fulgurite.

A. Green hornblende needle, unaffected by the fulgurite-stream. 50.

PHYSIOGRAPHY OF THE BOSTON MOUNTAINS,
IIRIKANS AS

Tue highlands of Arkansas lie in the northwestern part of
the state, and comprise about half its area. They are divided,
physiographically, into a north and a south part by the valley
of the Arkansas River. They are also divided structurally into
the same parts, the former being a region of horizontally-bedded
rocks, somewhat disturbed by faulting and folding, while the
latter, known as the Ouachita Mountains, is distinctly a folded
region.

This northern division of the Arkansas highlands, with its
westward extension into Indian Territory, constitutes the south-
ern part of the Ozark region.* In Arkansas, it is divided into a
low and a high part, the former extending northward into Mis-
sourl, and passing, along its southern border, into the latter, by
an irregular but bold escarpment from 500 to 1000 feet high.

It is this latter region that is known in Arkansas as the
Boston Mountains. Including that part which les in Indian
Territory, its total length 1s about 215 miles, about 170 of which
is in Arkansas. Its average width approximates 35 miles. On
the south, it passes into the valley of the Arkansas River by
steep slopes, though less precipitous than those on the north.

These mountains are by far the highest part of the Ozark
region as well as the most picturesque. Their highest point, so
far as determined, is some miles east of the town of Winslow,
on the St. Louis and San Francisco railway, where the altitude
is 2,250 feet> | Prom, /this region of ‘highest elevation;ethicy,

™Read before Section E of the American Association for the Advancement of
Science at the Denver meeting, August Igor.

2 The Ouachita Mountains have been included by some writers with the Ozarks ;
but because of the great structural and topographic differences in the two regions, to
say nothing of the probable historic differences, this is manifestly wrong.

3 Topographic map United States Geological Survey, Winslow quadrangle.

694

695

PHVSIOGRAPHY OF THE BOSTON MOUNTAINS

‘sesuvy1y jo dew jolla s.tguuvig wWoIj 0}OYG—'I ‘DIT

1) §

696 A. H. PURDUE

gradually fall off to the east, sinking below the Tertiary deposits
just west of the St. Louis, Iron Mountain and Southern railway
and south of White River; also from this region of highest
elevation, they fall off westward to the Grand River in Indian
Territory.* It will be seen that the east-west line along the
crest of these mountains forms a gentle arch in the middle.
Structuraily, in the western part of Arkansas, these mountains
are a broad, flat anticline, the strike of which is east and west.
According to the geologists of the Arkansas Geological Survey,
it appears that the extreme eastern part of the region is mono-
clinal in structure, with the dip to the south.”

With the exception of the Illinois River in the western part
of the state, the drainage of the region is northward and east-
ward into White River, and southward into the Arkansas. The
direction of the streams has been determined by the slopes
incident to the uplift, modified in some cases by faulting and
flexuring. The effect of the latter upon. Little Red River and
neighboring streams has already been noted by Professors New-
som and Branner.3 The westward course of the Mulberry River
has been determined by a fault. Detailed work of the region
would doubtless disclose numerous other similar examples.

‘The drainage of the region is that intermediate between
youth and maturity. The streams are vigorous, and have com-
pletely dissected the plateau by the formation of gorges from
500 to 1000 feet deep, thus producing a very rugged topog-
raphy over the whole region. Between these gorges the slopes
often meet, forming more or less rounded hills; but more fre-
quently the intervening area is occupied by flat-topped, sand-
stone-capped hills of limited extent.

The tributaries of both the Arkansas and the White rivers
have worked their way back to, and in many cases, far beyond

™Dr. N. F. DRAKE, in Proc. of the Am. Phil. Soc., Vol. XXXVI, No. 156, p. 332.

2 NEwsom and BRANNER, “ The Red River and Clinton monoclines, Arkansas,”
Am. Geologist, Vol. XX, July 1897, pp. I-13.

R. A. F. PENROSE, JR., Ark. Geolog. Surv., Vol. 1, 1890; section with pocket map.

3 Loc. cit.

PHYVSIOGRAPHY OF THE BOSTON MOUNTAINS 697

the original water divide of the plateau, making the water
divide as it now exists, a very zigzag line. In the western part
of the state, the south-flowing streams are the stronger, and as a
rule are robbing the White River basin of territory in this
locality. Further east, in the middle portion of the region, the
north-flowing streams are the stronger, and seem to be encroach-
ing upon the drainage area of the Arkansas, while in the east-
ern part, the south-flowing streams head very near the north
escarpment of the plateau.

The rocks of the region are mainly unmetamorphosed sand-
stones and shales, those at the base being of Lower Carboniferous
age, and those at the top belonging to the Coal-measure series.
These alternating hard and soft rocks have produced the ter-
races on the hill slopes, which are so characteristic of dissected
regions of horizontal strata. As these terraces are often of con-
siderable width, and are favorable horizons for springs, they are
inviting to the farmer, and can be located miles away by the
small farms on the mountain sides.

The low region to the north of the Boston Mountains is one
of great denudation. From its northeastern part, all the rocks
have been removed above the Ordovician, leaving those exposed
at the surface. West and south of this is a region from which the
Upper Carboniferous rocks have been removed, leaving those of
Lower Carboniferous age at the surface. Standing up promi-
nently on the latter are numerous hills of circumdenudation,
composed of remnants of the horizontal strata of the Boston
Mountains, and serving as living witnesses to their former
extent. The height of these outliers very closely approximates
that-of the plateau of which they were formerly a part. This
uniformity in height between the various parts of the dissected
Boston plateau and its outliers suggests a peneplain, and herein
lies the physiographic problem of the region.

In a region of folded or inclined strata the determination of
a peneplain becomes a question of comparative ease, for in
those cases denudation will have reduced both hard and soft
strata to practically the same level, the peneplain intersecting

698 TNs Jake, SOUR IOUS,

strata of all degrees of hardness. But in the case of horizontal
strata undergoing base-leveling, the conditions are quite differ-
ent, for then the peneplain conforms to the hard stratum or
strata that) happen) toy be mear seay levelly sliysweh va region
be subsequently elevated, the streams are revived, the region
dissected, and the former peneplain represented by the tops of
the hills, which would still be capped by the hard strata that
were conformable with the peneplain before the region was ele-
vated. Now this is exactly the structural and topographic con-
ditions of the Boston Mountains and their outliers. But it
happens that these are also the structural and topographic con-
ditions that would prevail in a region of horizontal strata that
has been elevated from beneath the ocean and is undergoing
the process of base-leveling for the first time. So the problem
presents itself as to which condition prevails in the Boston
Mountains, and unfortunately criteria for its solution are largely
if not wholly wanting.

Ordinarily, for the determination of a peneplain we look
to the streams. In such cases, as is well known, the streams
are winding, and flow in more or less steep-sided, symmetrical
valleys, which are themselves cut down in wider valleys. In
the Boston Mountains there is no such evidence of a peneplain.
The streams of the region are all young, with the characteristic
steep-sided gorges of such streams. So far as the writer has
been able to observe, there is nothing in the region indicating
an uplift since the present streams came into existence. Their
valleys are relatively wide at their mouths, and gradually
decrease in width back to their sources, as would be expected of
streams cutting into a plateau of horizontal strata. The slopes
are undisturbed by terraces, excepting such ‘as those mren-
tioned above, which are due to structure. Along the southern
base, the oldest of the streams have reached the temporary base-
level of the Arkansas River, and meander somewhat, but none
of them to any great extent.

It follows that evidence of a former base-leveling, if there
be such, must be looked for elsewhere than in the streams. A

PHYSIOGRAPAY OF THE BOSTON MOUNTAINS 699

recent writer’ claims that the tops of the Boston Mountains
represent a peneplain, and cites as evidence the fact that they
correspond very closely in height with the Ouachita Mountains
south of the Arkansas valley. This evidence is given on the
assumption that the rather uniform height of these mountains
represents a peneplain; but this is a hypothesis far from being
established. Mr. L. S. Griswold, in his work on the novaculite
region of Arkansas, encountered the problem of the noncon-
formance of some of the main streams of the region to the
structure and topography, to account for which he presents the
theory of a post-Carboniferous base-level, on which was subse-
quently deposited Cretaceous strata.2 If the present writer
correctly interprets Mr. Griswold, he believes the south-flowing
streams, which form water-gaps in some of the highest moun-
tains of the region, are superimposed streams, their courses
having been determined by the slope of the Cretaceous area
after elevation. Mr. Griswold does not claim that the evidence
of this is conclusive. It is the opinion of the present writer,
from somewhat limited observation, that the even crests of the
Ouachita Mountains are due to structural and lithological con-
ditions and not to base-leveling. But were it established that
they represent a peneplain, the fact that the Boston Mountains
closely agree with them in height does not argue a peneplain
for the latter. The one is a folded area, and the other an area
of horizontal rocks (Fig. 2). Erosion in the one has resulted
in wide, anticlinal valleys through which flow sluggish streams,
while erosion in the other is in its early stages. It would seem
to follow that the time of elevation of the one region is far ante-
cedent to that of the other, and consequently the correspon-
dence in height between the two only accidental.

If, however, we look to the north of the Boston Mountains,
we find conditions which seem to throw some light upon the
subject. As has already been said, this is a region of great
denudation. Its general elevation is from 700 to 1000 feet

™O. H. HERSHEY, Am. Geologist, Vol. XX VII, No. 1, pp. 25 e¢ seg.

2 Ark. Geol. Surv., 1890, Vol. III, pp. 220.

700

as

Amin

Boston

Back bone Mt.

A Sugar Loaf Mt

Frc. 2— North-south section of the Boston Mountains and adjacent regions, near Arkansas-Indian Territory line.

Alo Jats dA IOGTS,

lower than that of the Boston Mountains. Its
streams are mature, the valleys comparatively wide,
and the topography in general presents the aspect
of much greater age than that of the Boston Moun-
tains. Professor C. F. Marbut, in discussing that
part of this region which lies in Missouri,’ claims
that it was base-leveled in early Tertiary times, and
the present cycle of erosion was instituted by an
elevation which dates from middle or late Tertiary
times. Be that as it may, the question as to whether
the region to the north of the Boston Mountains
ever suffered denudation to the extent of base-level-
ing does not particularly concern us here. The fact
of interest is that the denudation of the extensive
region to the north has been very great and the
topography is old, while that of the Boston Moun-
tains is limited and the topography young.

It would appear that this difference in topogra-
phy cannot be attributed to the massive beds of sand-
stone at the top of the Boston Mountains, for these
same beds, while they have doubtless had a great
deal to do with preserving the region, formerly
extended over much if not all the denudated area to
the north. Besides, if we attribute the preservation
of these mountains to the character of the rocks com-
posing them, we are encountered by the question as
to why erosion has been so extensive to the north of
the region, removing the rocks over a large area,
leaving only here and there hills or circumdenuda-
tion, while in the southern part adjacent to the Arkan-
sas valley it has scarcely begun.

I am able to account for the great difference in
the stages of erosion in the two regions only by con-
ceiving the Boston Mountain area to have been ata
lower elevation than the area to the north during

t Mo. Geol. Surv., Vol. X, pp. 27-29.

PHYSIOGRAPAY OF THE BOSTON MOUNTAINS 7OI

the time the extensive denudation was going on over the lat-
ter. So low must it have stood that the strata now composing
their summits suffered but little erosion, while the same beds
extending northward suffered much because of their greater
height. If this be true, the actual amount of degradation
suffered by the Boston Mountain region is indeterminable;
but as there was more or less of it, and the region stood at a low
level, it would be considered a peneplain. The elevation, which
must have occurred in late Tertiary or in post-Tertiary time,
was greatest along the present east-west axis of the plateau,
gradually decreasing to the northward, and changing the region
from a low, monotonous plain to a plateau approximating 2,500
feet 11 height, greatly modifying the former drainage and insti-
tuting that of the present.

Aside from the difference in topography between the region
under discussion and the one to the north, the writer cannot at
present claim very great support for the idea herein presented.
There are, however, some other facts that seem to lend the
hypothesis support. (1) The region being on the border of the
Ozark uplift, it is probable that during the greater part of its
history it lay at a low level and consequently suffered compara-
tively little from erosion. (2) The outliers of the Boston
Mountains to the north are as a rule lower than the main
plateau, though capped by the same rocks, thus indicating an
axis of elevation to the south of the outliers. (3) The eastward
course of White River and its tributaries may be due to their
having been diverted from what would seem a more natural
southern course, at the time of the uplift.

Aare RuRDUE,

UNIVERSITY OF ARKANSAS,
Fayetteville, Ark.

DHE DISCOVERYGODVAy NEW, FOSSiii dT Arkin
OREGON

A FAIRLY complete phylogenetic series of early Miocene
tapirs has been made known to science through the researches
of Messrs. Wortman,! Earle, and, Hatcher.2 Between these
ancestral forms, referable to the genus Profapirus, and the living
species is a gap in the line of descent which has remained
unbridged until the fortunate discovery of the form presently to
be described.

Our knowledge of the tapir phylum since the White River
epoch may be summarized in a few words. In 1873, Dr. Joseph
Leidy: described under the name Lophiodon oregonensis, two imper-
fect superior molars obtained by Professor Thomas Condon at
Bridge Creek, Oregon. Two species have been described by
Professor Marsh,‘ which he refers to the genus Zapiravus: T. rarus
from the Loup Fork of the Rocky Mountains, and Japiravus
validus from the Miocene of New Jersey. From the brevity of
the description and the lack of figures, these species are prac-
tically indeterminate. Remains of tapirs belonging to the exist-
ing genus are known from the Quaternary gravels of California,®
and have been described from several localities in the eastern
states.

During the summer of 1900, Professor John C. Merriam and
Mr. V. C. Osmont, of the University of California, while collect-
ing in the fossil beds of the John Day valley, Oregon, obtained

tJ. L. WorTMAN and C. Ear.e, “Ancestors of the Tapir from the Lower
Miocene of Dakota,’”’ Bull. Am. Mus. Nat. Hist., Vol. V, p. 159.

2J. B. Hatcuer, “ Recent and Fossil Tapirs,” Am. Jour. Scz., 4th ser., Vol. I,
p- 161, 1896.
3U. S. Geol. Surv. of the Territories, Vol. I, p. 219, Pl. II, Fig. 1.
40. C. Marsu, Am. Jour. Sct., Vol. XIV, p. 252, 1877.
SJ. D. WHITNEY, ‘“ Aurif. Gravels,” Mem. Mus. Comp. Zool., Harvard, Vol. VI,
p. 250; W. P. BLaKkeE, Am. Jour. Sci. Vol. XLV, p. 381.
702

DISCOVERY OF NEW FOSSIL°TAPIR IN OREGON 793

the bones which form the subject of the following discussion.
The remains are from the Promerycochcerus horizon (Upper John
Day) exposed on the bank of the John Day river, beneath the
Columbia basalt, to the west of Spray post-office, Wheeler
county, Oregon. Ina recent paper,’ Professor Merriam named
the beds of the Upper John Day the Paracotylops beds, basing
the name on the new genus Paracotylops, proposed in the same
paper by Dr. W. D. Matthew for the typical Oreodonts of this
horizon. In the numbers of the American Journal of Science
for last December and January, a paper by Mr. E. Douglass
appeared in which these Oreodonts were provisionally named
Promerycocherus. Neither Professor Merriam nor Dr. Matthew
read this article before the publication of Professor Merriam’s
paper, and consequently did not notice the new name. It now
appears that Promerycocherus should be retained as a generic
name, and consequently, at Professor Merriam’s suggestion, the
name of the beds of the upper division has been changed from
Paracotylops to Promerycochcerus beds.

ihe-stype specimen (No. M934) Univ.;ot Cal. Pale Mus.)
comprises several superior incisors; the lower jaw lacking the
posterior portion, with representatives of all the inferior denti-
tion excepting the canines and the third molar; the proximal
portion of the left humerus; the left radius; the scaphoid, lunar,
magnum, and unciform of the right carpus; three metacarpals of
the same side, and a few phalanges. The bones are those of a
single individual of a new species of the genus Protapirus, for
which the name Protapirus robustus is proposed. It is consid-
erably larger than any of the White River species of Protapirus,
and would approximate in size the most specialized hving tapir,
Elasmognathus bard. The lower jaw is represented about one-
half natural size in Fig. 1. The symphysial region was found
in place, imbedded in a buff colored tuff so characteristic of the
Upper John Day beds that the expression ‘buff beds”’ was used
as a convenient field term for this horizon. The other bones lay
loose on the surface in the immediate vicinity.

™ A Contribution to the Geology of the John Day Basin,” Bul’. Dept. Geol., U. of
Cal., Vol. II, No. 9, p. 296; JouR. GEOL., Jan.-Feb., 1901, p. 72.

704 WhTTAM | SUN GILATER:

The dentition.—The superior incisors are larger than the cor-
responding teeth in Elasmognathus bard. Vhe inferior incisors
are slightly smaller than the superior. Both series have the
crowns somewhat cupped, especially so in the superior incisors.
The first and second inferior incisors are of equal size, while the
third is two-thirds as large as those preceding it. The crowns
of both canines are broken off, but the diameters of their roots,
measured on the alveolar borders, are greater than the corre-
sponding parts of the larger incisors. A long diastema succeeds
the canines. The premolars have their anterior cusps united
into transverse ridges, slightly notched at the summit. In all

Jk ike

except the second premolar, the ridges are perpendicular to the
long axis of the jaw. They are equally developed on the third
and fourth premolars. Posterior cross crests are not developed
on any of the premolars. In the second premolar, the proto-
conid is larger than the deuteroconid and is situated farther for-
ward than the latter. In the succeeding premolars, these cusps
are of the same size and are situated directly opposite each
other. The tetartoconid of the premolars is smaller than the
metaconid. The latter cusp is united with the inner side of the
base of the protoconid by a ridge. This structure is also found
in the molars, all of which have two cross crests. The posterior
crest in the first and second molars is somewhat oblique to the
axis of the jaw. The third molar is too imperfectly preserved
to describe. Anterior and posterior cingula are present on all
the molars and premolars. Traces of an external cingulum are
found at the outer end of the transverse valley in all except the
second premolar. In this tooth the paraconid is very large,
uniting by a ridge with the protoconid. In the remaining pre-
molars the paraconid is replaced by a style rising but little above
the level of the anterior cingulum.

DISCOVERY OF NEW FOSSIL TAPIR IN OREGON 705

The jaw.— The inferior border of the jaw is parallel with the
alveolar border. The symphysial portion rises at a low angle,
much less thanin the tapir. The flatness of this angle is perhaps
due in part to a slight amount of crushing which the specimen
has sustained. The posterior border of the mental foramen is
directly below the anterior border of the second premolar.

The fore limb.—The humerus and radius have about the
same shape as in the tapir. The deltoid ridge of the former is
broken off, so that it is impossible to say whether it was hooked
or not. The shaft of the radius is more strongly curved than
the corresponding element in Protapirus validus as figured by
Hatcher,* but a part of the curvature may be due to distortion.
The carpus does not call for special description, not differing
materially from that of Protapirus obliquidens. W.& E. The ante-
rior contact of the lunar and magnum is still small, as in the
White River species. There were four digits in the manus, the
length of metacarpals III, 1V, and V being about the same as in
E. bairdi, but less robust. In shape they correspond closely with
the metacarpals of the latter, except that the proximal portion
of the fifth is inclined at a greater angle to the shaft of the bone
than inthe living form. The phalanges, which are of the second
row, are shorter and less robust than those of the tapir.

Phylogenetic position — The remains just described indicate an
animal much larger than any of the White River species of the
same genus. The structure of the molars and premolars suggests
Protapirus validus as a probable ancestor. There are, however,
several differences. In addition to the considerable difference
in size, the third premolar of P. rvobustus has the anterior cross
crest vertical to the long axis of the jaw, while in P. validus it
is somewhat oblique. The diastema, as in P. validus, is shorter
than in Alasmognathus, while the mental foramen has moved
slightly posterior to the position it occupies in the White River
ancestor. Gradations between the two types probably occur
among the as yet unknown tapirs of the Lower and Middle John
Day.

PEOC« Gl... PLO 72 bis, 1.

700 WALEIAM.. J. SINCLATR

MEASUREMENTS OF PROTAPIRUS ROBUSTUS

Length of inferior molar-premolar series - - - 12810 1302"
Length of inferior premolar series - - - - - - - 60
Length of inferior molar series (approximate) - - - - 70
Length of diastema - - - - - - = - - 46
Length of symphysis measured on lower side - - - - ak
Depth of ramus below alveolus of Pm. 2 - - - - - - 48
Depth of ramus below alveolus of Pm. 4 - - 2 - - 49
Depth of ramus below alveolus of M. 3 - - - - - AG oe
Length of radius - - - - - - - - Seon
Breadth of proximal end of radius” - - - = : - - 45
Breadth of distal end of radius - - - - - - - 39

In this connection, may be mentioned a second specimen,
(No. M 1525 University of California Pal. Museum), representing
probably another new species, obtained by the writer in the
uppermost beds of the John Day system, on Johnson Creek,
Grant county, Oregon. The horizon is considerably higher than
that from which Protapirus robustus was obtained, and appears to
be faunally distinct. It is characterized by the remainsiot
numerous individuals of a camel belonging to the genus Pro-
tomeryx and by a rodent generically new. The tapir remains are
of a young animal and are not complete enough to characterize
specifically. They comprise fragments of a jaw with which three
incisors and the second inferior premolar are preserved. The two
large incisors, apparently the inferior median pair, are two-thirds
as large as the corresponding teeth in P. vobustus. They are
spatulate in shape and slightly cupped. The anterior face is
marked by delicate growth lines. The third incisor is an exceed-
ingly small tooth with the crown 3$™™ broad. Imperfect preser-
vation of the symphysial region renders it impossible to make
any statement regarding the canine. The second premolar of
the right side is the only one of’ the cheek teeth perfectly pre-
served. This tooth is entirely unworn, and was just appearing
through the gums at the time of the animal’s death. In this
tooth, the’tetartoconid is much larger than in P. robustus and the
junction of the metaconid with the tetartoconid is much more
complete, forming a cross crest but slightly notched. A ridge

DISCOVERVVOPINEW FOSSTL TAPIR IN OREGON — 707

joins the former cusp with the middle of the anterior cross crest.
The protoconid is considerably anterior to the deuteroconid and
as in P. robustus is united with the paraconid bya ridge. The
anterior cross crest is sharply notched, but this structure would
probably assume the character of the anterior cross crest in
P. robustus with the wearing down of the deuterocone by use.
External cingula appear at the outer margin of the median valley
and on the external side of the paraconid. A posterior cingulum
is also developed.

MEASUREMENTS
Width of the crown of first inferior incisor - - - - - (oy peste
Width of the crown of third inferior incisor - - - - =a sip
Length of the first premolar crown, antero-posteriorly — - - - 154

Wo. J. SINCLAIR.
PAL-ZEONTOLOGICAL LABORATORY,

University of California.

THE FORMATION AS THE BASIS FOR GEOLOGIC
MAPPING

In a recent number of this JournaL Mr. Bailey Willis, ina
paper on ‘Individuals of Stratigraphic Classification,” has
restated and rediscussed the problem which must be solved
before cartographic work of any magnitude can be planned.
This problem involves a careful consideration of the relative
weights to be assigned, in any system of classification to be
used on geologic maps, to faunal, lithologic, and chronologic
(successional) characters. Mr. Willis discusses the question in
its various aspects, and his final decision is that the lithologic
unit (formation) is best adapted to the requirements of the car-
tographer.

While agreeing, in the main, with the conclusions reached
by Mr. Willis, it seems desirable to call attention to certain
arguments, not specifically mentioned by him, which may be
adduced in support of those conclusions; and, further, to exam-
ine the results of the application of the proposed system of
classification to some particular cases of interest.

Before commencing the discussion of this question I wish to
acknowledge my indebtedness to Dr. F. J. H. Merrill, director
of the New York State Museum, who has greatly aided me, with
both criticism and advice, during the preparation of this paper.

THE NECESSITY FOR UNIFORMITY

Though the formation, defined primarily by lithologic char-
acters, was officially adopted in 1889 as the cartographic unit of
the United States Geological Survey, in practice it has not
entirely superseded other units of classification. Great variety —
exists in the practice of the various state geological surveys, as
is indicated by their official maps; and greater variety, as might
indeed be expected, in unofficial maps accompanying papers on

708

FORMATION THE BASIS FOR GEOLOGIC MAPPING 709

areal geology, even though the authors be connected with sur-
veys whose official practice is uniform.

In considering the representation on a map of the geologi-
cal features of a small, isolated area, the question of the taxo-
nomic unit to be employed is usually of minor importance ;
though even in this case, as shown later, there would seem to be
good reasons for the adoption of the lithologic individual as
this unit. The interest attaching to this question increases
directly with the size of the area to be mapped; and the prob-
lem becomes of paramount importance when the independent
work of several geologists is to be combined, as in the compi-
lation of a geologic map of an entire state or other large area.

In view of these facts, and of the slow but steady spread of
the geologic-folio work carried on by the United States Geologi-
cal Survey, it would seem necessary that the various local sur-
veys should adopt the same general system of classification for
map work. In the present state of our knowledge, detailed
chronologic classification, the unit for which would be the
epoch’ of earth-history, is impossible; and the problem of unifi-
cation is therefore resolved into the choice between two alterna-
tive systems, based respectively upon biologic and lithologic
characters. In the opinion of the writer a geologic survey,
whether state or national, can best accomplish the work for
which it is intended by adopting as its cartographic and taxo-
nomic unit the formation, defined as a lithologic individual.

THE RELATIVE SCIENTIFIC VALUES OF BIOLOGIC AND
LITHOLOGIC UNITS

For the purposes of the present discussion it is necessary to
point out that our knowledge of the two histories (7. ¢., of sedi-
mentation and of life), so far as that knowledge can be expressed
on maps, is decidedly different in grade. The present phase of
earth-history may be examined for a determination of the truth
of this statement. We can map without difficulty the topo-
graphic features of the earth; and the possible accuracy and

t The word epoch is here used in an entirely general sense.

710 EDWIN GC. FEKELE

precision of such mapping is limited only by the consideration
of expense. Moreover, we are able in most cases to discuss
topographic features in terms of causality.

With regard to the present biologic condition of the earth
our knowledge is much less definite. Mapping the distribution
of existing faunas or floras is not practicable in the present state
of our knowledge of biotic’ distribution, except in a very gen-
eral way; anything like a detailed map is impossible. This con-
dition is due partly to the expense of collecting sufficient data
on the range of the different species of plants and animals.
Most of the difficulty, however, arises from the fact that the
principles underlying and regulating animal and plant distribu-
tion are still far from being well understood, though of late
years great advances have been made in that direction.

If our knowledge is thus imperfect with regard to the exist-
ing biota,* our knowledge of the facts and causes of its distribu-
tion during past ages is still less definite. It is noteworthy that
two of the most successful attempts? to account for biotic
distribution and evolution in periods antedating the present
have been the work of geologists specializing in physiographic
work, a branch which necessitates lithologic rather than biologic
discriminations.

Obviously, the paleontologic record must always be more
defective than the lithologic; for fossils are not always found
where rocks are exposed. In addition to gaps occurring because
of local lack of fossils, two periods are particularly incapable of
being treated on a biologic basis; one immensely long period at
the commencement of geologic history, and one short but highly
important period immediately preceding the beginning of written
history.

With regard to the relative values of the biologic and litho-
logic units as measures of absolute time, the case is somewhat

t Biota = the sum of the fauna and flora of aregion. Cf STEJNEGER, American
Naturalist, February 1901, p. 89.

2T. C. CHAMBERLIN, “‘Systematic Source of Evolution of Provincial Faunas,”
JouRN. GEOL., VI, pp. 597-608, 1898. J. B. WoopworrTH, “ Relation between Base-
leveling and Organic Evolution,” American Geologist, XIV, pp. 209-235, 1894.

FORMATION THE BASIS FOR GEOLOGIC MAPPING 711

in favor of the latter unit; for the relation existing between
time taken in deposition and thickness of formation is much
more traceable than that between extent of variation and time
taken in evolution.

In considering the adoption of a basis for geologic mapping
the biologic unit has, therefore, no antecedent claim to special
consideration on scientific grounds; and its value can be com-
pared directly with that of the lithologic unit in regard to their
relative practicability and utility.

THE RELATIVE PRACTICABILITY OF THE TWO SYSTEMS

A lithologic unit can usually be so described and defined as
to be readily identified by any future worker in the same or an
adjoining area through which the formation extends.

It should also be noted that all geologists substantially agree
upon the use of the terms in which lithologic individuals are
defined, and will consequently agree closely in their valuation

0 66 dey

of any series. ‘‘ Limestone, sandstone,” “slate, conglom-
erate,” all have fairly definite meanings, and it is improbable
that terms such as these are used in very different senses by
different geologists. The biologic unit, on the other hand, is
rarely capable of being so described or defined as to be accepta-
ble to all paleontologists. The difference is that the lithologic
individual is a fact; the biotic individual, as commonly described,
is a fact plus an interpretation; and while there may be sub-
stantial agreement as to the facts in any given case, it is but
rarely that the interpretations will coincide.

The Hudson formation in the vicinity of Albany, N. Y., pre-
sents an excellent example of some of the practical difficulties
encountered in attempting to represent faunal distinctions ona
map; anda brief sketch of the conditions may be of interest in
the present connection.

i)

The ‘‘ Hudson River group’ of the earlier classifications —

a9,

the ‘‘ Hudson formation” of the present system of nomenclature
—comprises in the central portion of the Hudson River valley,

where it is best developed and exhibited, a thick and extensive

712 BDWIN VG. BEKELE

series of shales, with interbedded sandstones and occasional
thin beds or lenses of limestone. This series is well shown
along the banks of the Hudson River, extending almost uninter-
ruptedly from Fort Edward to Cornwall. Faunal differences
justify the paleontological division of the ‘‘group”’ into four
stages, of which the two lower carry a fauna which would corre-
late them with the Trenton (using that term in its paleonto-
logical significance), while the other two represent respectively
the Utica and Lorraine beds of the Mohawk valley. This divi-
sion into stages, however, is and must be largely theoretical, for
it cannot well be carried out in practice on a map. Fossils,
especially of species which can be regarded as of taxonomic
importance, are neither profusely nor regularly distributed
throughout the beds in question. Outcrops carrying character-
istic fossils are too few and far between to warrant mapping the
area, on a paleontological basis, on any large scale. The litho-
logic differences which occur in the group have no stratigraphic
or cartographic value, being too slight and variable to admit of
separate representation on a Map.

Modern geologic mapping, especially if the base of the map
is to be a topographic atlas sheet, in order to make adequate
returns for the expense involved, must be accurate within the
limit fixed by the scale of the map.

The production of a geologic map is necessarily accom-
plished by the exercise of two functions, observation and
inference. Observation involves the location of outcrops and
lithologic boundaries with reference to fixed points in the con-
trol of the map, and is therefore purely a matter of engineering.
The exercise of the function of inference is necessary in order
to indicate the positions of boundaries concealed by superficial
material. Inferences in relation to such matters are dependent
for their accuracy on the training and judgment of the geologist;
his appreciation of the relations existing between structure and
topography, and his knowledge of the geometric effects of dip,
pitch, etc., in determining the position, both horizontal and ver-
tical, of a concealed boundary line.

FORMATION THE BASIS FOR GEOLOGIC MAPPING 713

Other things being equal, the determination of the altitude
and geographic position of a given point is dependent upon the
engineering skill of the observer; and the cartographic unit to
be employed does not seriously affect this phase of the work of
mapping. With regard to inference, however, this is not the
case. No necessary relation exists between a biologic unit and
the topography or structure of the area in which it occurs. On
the other hand, topography, structure, and the lithologic unit
are in general closely related. Recognition of these relations
makes it possible to draw inferences concerning the position of
the concealed boundary of a lithologic unit from the geologic
structure and topography of the area discussed.

THE RELATIVE UTILITY OF LITHOLOGIC AND BIOLOGIC MAPS

A lithologic unit is normally also an economic unit, whereas
there is no necessary relation between the biologic unit and the
economic importance, or lack of importance, of the rocks con-
taining a given biota. Though this argument may be regarded
by some as of less weight than one based on purely scientific
grounds, it is nevertheless valid, as applied to the question under
discussion. It should not be forgotten that no geologic survey
has ever been instituted, save for economic reasons; that the
chief argument that can be used to obtain state support for such
surveys is the direct ‘economic return to the public; and that
simple justice demands, in return for this support, that the geo-
logic results obtained be placed in such a form as to be of the
greatest possible use to that public.

The lithologic unit is also generally related to geographic
forms in a very definite manner, as well as to geologic structure.
The effects of these relations upon the actual work of mapping
have been discussed. Here it is only necessary to point out
that both topography and structure are often of economic
importance. A map based on a unit which bears some definite
relation to them is, therefore, of greater value from an economic
standpoint than if the unit be one not so related.

714 EE PIWAIN \C. 2 GREE

THE DIFFICULTIES IN MAPPING ON THE LITHOLOGIC BASIS

Difficulties will certainly be encountered in mapping any
large area on a lithologic basis, but these difficulties will arise
rather from differences in the interpretation of the rules for the
nomenclature of the units than from any defects in the system
itself. Certain cases of interest in this connection are discussed
below:

Case 1.—The case is that cited first by Mr. Willis (p. 563).
A shale passes along the strike into a limestone which retains
identical stratigraphic associations. ‘Being exactly continuous
stratigraphic units, they should retain the same geographic name
on grounds of convenience and simplicity.” Considered as a
stratigraphic unit, the two would be discussed as the “X”
formation. In order to preserve all the advantages of the litho-
logic system of classification, however, the limestone phase
should be differentiated, in both discussion and mapping, from
thecshale: phase; theix respective names bene «then thes Ge
linfestonesand the “xX~ yshale. Phe term) <phase 4 iscem=s ato
serve a useful purpose in this connection, and the present-writer,
therefore, proposes it for use in marking variations of sedimenta-
tion or of metamorphism within the stratigraphic unit. When
used later in this paper, it will be with this restricted and
definite meaning. :

This case is illustrated excellently in New York state and
western New England, in this example the phases: representing
variations in metamorphism. ‘The Hudson shales of the central
part of New York state are progressively metamorphosed to the
eastward, becoming first slates and then mica-schists. The
schistose phase has been called, in Massachusetts, the ‘‘ Berk-
shire schist ;’’ in southeastern New York, by Merrill, the ‘‘ Man-
hattan ischist;;-:while= Dale “has described’ the intermediate

”)

)

phases as ‘‘ Hudson slates.” Throughout the entire area, though
differing thus in character because of variations in metamorphic
action, the rocks are essentially continuous stratigraphically;
and recognition of this fact has led to the recent proposition to

eliminate the names Berkshire and Manhattan, and denote the

FORMATION THE BASIS: FOR GEOLOGIC MAPPING 715

various phases of the Hudson formation by the term ‘‘ Hudson
shale,” Hudson slate,” “‘ Hudson =:schist.” This. system: of
nomenclature, while recognizing the essential stratigrapic equiva-
lence and continuity of the various phases, allows these phases
to be discriminated from each other on lithologic grounds. On
maps this discrimination can best be shown by the-use of some
of the various overprints used to denote metamorphism.

Case 2.—The second case noted by Mrs. Willis is that of a
“shale grading into a limestone with prolonged overlap, so that
the two rocks must be discriminated in one area. Not only
are they lithologically different but they have different strati-
graphic associations, and they should receive distinct ’’ forma-
tional names. As noted under the preceding case, this is the
condition which usually obtains where there is a horizontal
gradation in character of sedimentation. The writer believes
that Mr. Willis has here suggested the proper treatment of the
question, but wishes to point out an actual case of some areal
importance in which this ruling has not been followed.

The example occurs in the northwestern part of Massa-
chusetts. To. the-:west.:6f Hoosac Mountain, the Cambrian
quartzite is overlaid by Stockbridge limestone and this in turn
by Hudson (Berkshire) schist; on and east of the mountain the
limestone does not appear. It has been generally assumed that
the schist of the mountain represented both the Berkshire schist
and the Stockbridge limestone, and the name ‘‘ Hoosac schist”
has been applied to it in several. publications.

What the writer wishes to point out is that, under the pro-
posed rules, the name at present used for one of those forma-
tions would seem to be untenable. For, if the eastern (Hoosac)
schist represents both the Berkshire schist and Stockbridge lime-
stone, the case becomes the same as 2, and in that case all
the schist should be called. Berkshire .and all the limestone
Stockbridge. If the transition be considered sudden, then it
exhibits a slight modification of case 3, and still all the schist
is Berkshire schist and all the limestone Stockbridge limestone.
If the relations at Hoosac Mountain be referable to overlap,

716 SPIOW ION, (On, II CIKG BY 6,

then it is obvious that the Hoosac schist is simply an equivalent
of the Berkshire schist, while none of the Stockbridge limestone
is represented by any of the schist east of the mountain. On
any hypothesis, therefore, the formation name ‘‘Hoosac schist”
is apparently untenable.

Case 3.—The third example cited by Mr. Willis is that of a
shale which varies horizontally in such a manner as to become |
differentiated into several superposed members, one of which is
a similar shale, the others limestone or sandstone. His decision
is that in such a case all the shale can be given the same forma-
tion name, while the new members (limestone, sandstone, etc.),
are to be given distinctive names; but that, in case it be desira-
ble to refer to the group as a whole insome area where it is thus
differentiated into several members, the geographic name applied
to the shale alone cannot be applied to the entire group.

The last three cases cited by Mr. Willis present instances of
subordinate parts or local developments (of the lithologic unit)
which, though too small for mapping, are of interest in discus-
sion. To such sections of the unit the terms ‘‘member”’ or
“Jens”? may be applied. This method of treatment and termi-
nology, if followed strictly and uniformly, would seem to be
entirely satisfactory.

SUMMARY.

It will be seen that the essential character of the proposed
unit is that it is uniform, or uniformly variable, in lithologic
character. Both its upper and lower limits and its lateral boun-
daries will therefore by marked by lithologic differences. So
long as the formation is geographically continuous, no other
criterion is necessary or even admissible.

In the case of discontinuities in outcrop, caused by the pres-
ence of bodies of water, or of intervening areas of other rocks
or superficial deposits, the question will arise as to the treatment
in mapping of two distinct bodies of rock, identical lithologically.
In this case the aid of stratigraphic association or contained
fossils may be invoked. :

In regard to sharpness of definition, ease and accuracy of

FORMATION THE BASIS FOR GEOLOGIC MAPPING 717

mapping, and utility of the resulting maps, the lithologic unit
would seem to leave lttle to be desired. Difficulties arising
from diverse interpretations of the rules for its discrimina-
tion can be overcome with little trouble. A more serious diffi-
culty, however, is to be encountered in regard to the question of
the geographic names to be given the units. It is not sufficient
to say that the rules of priority will determine the name to be
used, even if those rules be re-formulated with special regard to
this particular question. The difficulty to which the writer
alludes can almost be said to be inherent in the definition of the
unit and to be avoidable only by giving a very free interpreta-
tion of that definition. It arises when two or more geologists, who
commenced work in more or less widely separated areas, have car-
ried areal mapping toa junction. Identical formations will then
bear different names on adjoining folios. In regions whose
geology is fairly well known, identity of nomenclature can gen-
erally be arranged in advance with safety. In less well-known
regions, however, the possibility that the same formation may
require treatment under different names in adjoining folios will
have to be accepted as a necessary evil, largely outweighed by
the positive advantages of adopting the lithologic unit as the

basis for mapping.
Epwin C. ECKEL.
NEw YorK STATE MUSEUM,
Albany, N. Y.

fede WORK IN THE WESTERN MOUNTAINS
LN TOON:

DurinG the season of 1901 some time was given to the
study of Pleistocene problems in the western mountains. Four
small parties, beside the writer, were in the field. The work of
three of the four parties was the somewhat detailed study of
selected areas, and was intended to make known the general
conditions of glaciation at several somewhat widely separated
localities. When a sufficient number of selected areas have
been studied similarly, the results will afford a basis for pre-
liminary conclusions concerning the course of Pleistocene his-
tory in the western mountains, and will be helpful in guiding
future work. The work of this summer was intended as the
beginning study looking to me ones
ae with the parties eee to. One par ty was in north-
western Montana, east of the Rockies; one farther west, on the
western side of the Rockies, in northwestern Montana, northern
Idaho, and eastern Washington; one in the Wasatch Mountains ;
and one in the mountains of New Mexico, a few miles northeast
of Santa Fé.

The following brief summary will give some idea of the
work done:

NORTHWESTERN MONTANA, EAST OF THE ROCKIES.

Mr. Fred) Hs mH. Calhoun, accompanied? by Mins Bruce
McLeish, spent three months in the northwestern part of Mon-
tana. The area studied, some 8000 square miles in extent, lies
just south of the 4oth parallel, and just east of the Rocky
Mountains. It is the area in the angle between the Continental
ice sheet from the northeast, on the one hand, and the Cordil-
leran ice sheet from the northwest, and the mountain glaciers
from the west, on the other.

™ Published by permission of the Director of the U. S. Geological Survey.

718

GLACIAL WORK IN WESTEKN MOUNTAINS Iie)

The moraine formed by the northeastern ice sheet of the
Wisconsin epoch was traced from Fort Benton, vza Choteau, to
the point where it crosses the 49th parallel, about longitude
112° 20’. . From-this point the moraine bends to the west and
continues in this. direction to the Watertown River, where it
turns to the north, its course being nearly parallel to the moun-
tains. .
The moraines of fourteen glaciers from the Rocky Mountains
were mapped. The recessional moraines of both the valley and
continental ice masses were studied as far as time permitted.
Study was also made of the extent of valley trains and out-
wash plains, and of the lakes which existed in front of the ice
during the glacial period.

Among the results of the work, the following may be men-
tioned:

1. The westward and southwestward extension of the Wis-
consin drift is greater than had been supposed.

2. The till of the northeastern ice sheet and that of the moun-
tain glaciers overlap just south of the 4gth parallel, between the
10’ and 113° 15’.-- At-no’other-point south
of the boundary line did the ice from opposite ‘directions

O°

meridians, of 113

occupy the same territory; but in Two Medicine Valley (lati-
tude. 48° 30’) the ice from the east advanced to within two
miles of the position occupied at an earlier time by the ice
from the mountains.

3.---The -ice:-sheetfrom.:the northeast-Creached’ its. -most
advanced position after the valley glaciers from the west had
retreated, for the drift of the former overlies the drift. of the
latter at various points, both north and south of the national
boundary. In some cases the till from the northeast overlies
the till from the west. This is true. in the locality mentioned
under 2, and farther north in the valleys of St. Mary’s River,
Belly River, and Lees Creek. In other places the till from the
east overlies fluviatile deposits connected with the western drift,
but extending down the valleys some miles beyond (east of) the
ends of the mountain glaciers. In stil other cases. lacustrine

720 TOTES GD SAVES BO Kava

and berg deposits, having their source in the northeastern ice,
overlie the till laid down at an earlier time by the mountain
glaciers.

4. The valley glaciers belonged to a recent epoch. They
were certainly not earlier than the Iowa epoch, and probably as
recent as the Wisconsin.

5. The drift of the northeastern ice sheet is of Wisconsin,
possibly late Wisconsin age.

6. So far as the evidence gathered shows, the length of time
which elapsed after the deposition of the Rocky Mountain drift,
and before the deposition of the drift of the ice sheet from the
northeast, was not great.

7. The time since the last glacial epoch, as shown by several
lines of evidence in the area studied, seems to have been geo-
logically short.

8. The Sweet Grass Hills, situated just south of the 49th par-
allel and about thirty miles back from the edge of the ice sheet
from the northeast, were not covered by the ice. The hills stood
as three great nunataks, the highest reaching about 2000 feet
above the surface of the ice.

g. The slope of the surface of the continental ice sheet
between its edge and a point 25 miles back from its edge was
estimated at about 50 feet per mile. This determination was
made in the vicinity of the Sweet Grass Hills. The slope of the
surface of the ice in various valleys was also determined.

10. The existence of a long narrow lake in front of the edge
of the Continental ice sheet, and between it and the mountains,
was proved, and its extent determined.

11. Considerable changes in drainage were effected by the
ice. Some data were gathered which throw light on the former
courses of the Missouri, the Teton, and the Two Medicine rivers,
and of Buck, Muddy, and Cut Bank creeks.

Besides these conclusions touching glacial problems, some
other observations were made:

12. The region studied shows at least three periods of ero-
sion, when the land stood at different levels.

GLACIAL WORK IN WESTERN MOUNTAINS 721

13. The high-level quartzite gravels on the plains east of the
mountains are believed to be deposits made by streams at the
close of the first epoch of base-leveling recorded in the present
topography.

14. The strata at the mouth of Sun River and Muddy Creek
canyons contain fossils of Carboniferousage. Rocks of the same
age prevail in the whole of the eastern face of the mountains,
where their thickness is several thousand feet.

Some other scattering data were collected touching the gen-
eral geology of the region studied.

BETWEEN THE ROCKIES AND THE CASCADES IN NORTHWESTERN
MONTANA, NORTHERN IDAHO, AND EASTERN WASHINGTON

Messrs. George H. Garrey and Eliot Blackwelder spent most
of August and September in the study of glacial phenomena
in northwestern Montana, northern Idaho, and northeastern
Washington, most of the time being spent in the latter state,
between Chelan Falls and Newport. The moraine of the ice
lobe which came south through the Okanogan Valley, loops
southeastward from Chelan Falls to a point about six miles west

of Coulee City, where it bends northeastward to a point five
' miles north of the city, on the bluff of the Grand Coulee.
Morainic material not aggregated into a marked terminal
moraine, continues northward along the entire western rim of
the Coulee. During glacial times the ice probably blocked the
Columbia at its junction with the Grand Coulee, forcing the
waters of the Columbia down the Coulee, as has been suggested
by Russell and others. That glacial drainage followed the
course indicated is suggested by the fact that the terraces of the
Columbia above the Coulee are continuous with terraces in the
Coulee.

Northeast of the junction of the Grand Coulee with the
Columbia, the moraine of the Okanogan or Coulee City lobe
takes the form of a well-developed ridge which, at a point about
five miles north of the junction of the Columbia and the Grand
Coulee, the most northerly point where it was seen, was trending

722 ROLLIN Die SALTSE OieN

slightly to the west, of north. After continuing. northward
for a distance, this moraine probably loops eastward around
- the mountains in the Colville Indian Reservation, connecting with
the moraine of the ice lobe which occupied the Columbia valley,
north of its junction with the Spokane. The Okanogan or
Coulee City ice lobe was about thirty-five miles wide in the lati-
tude of Chelan Falls, and nearly fifty miles wide at the upper
end of the Grand Coulee.

The moraine of the ice lobe which came down the valley of
the Columbia, just west of the meridian of 118°, was traced
southward from a point in the Colville Indian Reservation about
three miles west of Kettle Falls on the Columbia, to the point
where it crosses the Columbia seven miles southwest of Fruit-
land. Between these points, the moraine lies three to six miles
west of the Columbia. Just east of the Columbia the moraine is
lost in the terraces at the junction of the Columbia and Spokane
valleys. From this point the moraine was traced northeastward,
passing about three miles east of Fruitland, and thence along the
west face of the mountains, to the Huckleberry Mountain in the
Sue comen ot Dp 32, hk. 33). (about vlatituden 43) as5.@)\amate
this point it loops around and over the northeastern end of
this mountain, and then turns southward in the Colville River
Valley. The glacier of the Columbia Valley had a width of
about six and one-half miles in latitude 48°, and a width of
about fifteen miles, fifteen miles farther north.

As already mentioned, the moraine of the Columbia glacier
loops over and crosses Huckleberry Mountain and becomes con-
tinuous, in the Colville Valley, with a lobe of ice which descended
that valley to Springdale. The moraine of the east side of this
lobe was traced northward along the mountains east of the Col-
ville Valley, to Old Dominion Mountain, seven miles east of
Colville. “This.seems to be the latitude where, the moraine
swings across the mountains to the east, though the crossing was
not demonstrated. The Colville glacier had a width of about
thirteen miles at Valley, about eight miles north of the southern
end of the lobes. Lhe width of “thescombinedsColvillesand

GLACIAL WORK IN WESTERN MOUNTAINS 723

Columbia glaciers just north of their union (lat. 48° 15’ approxi-
mately), was about thirty-three miles.

In the Pend d’Oreille Valley there was another ice lobe, which
extended down the valley to a point three miles southwest of
Davis Lake. To the northwest this moraine is believed to con-
nect with that of the Colville lobe, north of Old Dominion
Mountain, though this connection was not established. The
moraine of this lobe crosses the Pend d’Oreille River to the
eastward at the great bend of the stream eight or ten miles
above Newport. East of the Pend d’Oreille the moraine of the
east side of this ice lobe was judged, on the basis of topography,
to turn northward.

Another ice lobe came down the Kootenai Valley, but its
southernmost limit was not determined. On the basis of data
gathered at other times by Professor Chamberlin and the writer,
the ice of the Kootenai Valley is believed to have extended to
the southern boundary of Pend d’Oreille Lake. It is possible
that the Kootenai lobe connected to the northwestward with the
Pend d’Oreille lobe; but, if so, the connection is well to the
north of Bonners Ferry, and probably north of the boundary
line.

It remains to be determined whether the Okanogan, Columbia,
Colville, Pend d’Oreille,and Kootenai glaciers were marginal lobes
of asingle continuous ice sheet, the body of which lay to the north,
or whether some of them had separate sources in the mountains.
The connection of the second and third of these lobes was estab-
lished, and data at hand make the connection of the third and
fourth seem probable. The connection of the first with second,
and of the fifth with fourth, if such connection exists, is proba-
bly near the boundary if not beyond it. In any case, the gla-
ciers partially traced out were large—quite beyond the Alpine
class.

A few data concerning glaciation farther east were also
gathered. General glaciation did not occur as far south as New-
port at the east line of Washington. Above Libby (in the
Kootenai Valley near the west line of Montana) there was gen-

724 ROLLIN D. SALISBURY

eral glaciation from the north, though a few miles south of the
river in this longitude drift from the north comes in contact with
drift from the Cabinet Mountains. Whether the glaciation from
opposite directions was contemporaneous, was not determined.
Deposits of lacustrine clay some 300 feet thick in the valley
near Libby, point to notable interruptions of drainage during
glacial time, and probably to considerable changes since.

The Flathead valley contained a great glacier which advanced
southward beyond the southern end of Flathead lake, though the
southern limit of the ice was not traced out this season. In the
latitude of Kalispell, the ice of this valley was about 3000 feet
thick, as shown by the height to which the west face of the
Kootenai Mountains east of the valley, was glaciated. The gen-
eral direction of ice movement in this valley at Kalispell, and
between this place and the lake, was south-southeast. The
main body of the ice, therefore, came down the valley from the
north northwest, and not from the mountains immediately east
or northeast, though the main glacier was reénforced to some
extent. by ice from this direction. The source of ice which
moved to the south-southeast was not determined. One lobe
of the Flathead glacier moved southwest up the valley of Ashley
Creek (the valley followed by the Great Northern Railway west
of Kalispell), and another advanced westward a short distance
beyond the end of the west arm of Flathead Lake.

The face of the Kootenai Mountains east of the Flathead
Valley is abrupt, and not deeply indented. From the slight
amount of study given to this face of the range, it would appear
that local glaciation here was not of great extent or severity, for
some of the considerable mountain ravines opening out to the
west afford little evidence of glaciation. On the other hand,
glaciation was severe in some of the valleys, as in that followed
by the Great Northern Railway east of the Flathead Valley.

THE WASATCH MOUNTAINS.

Mr. Wallace W. Atwood spent about two months in the study
of the glacial phenomena of the Wasatch Mountains. He was

GLACIAL WORK IN WESTERN NOUNTAINS 725

accompanied and assisted by William Peterson, M. J. Averett,
H. B. Atwood, and Arthur Church.

The portion of the Wasatch Mountains studied lies between
the parallels of 4o° 15’ and 41°. The maximum width of the
range in this latitude is about twenty miles. The total area
examined for glacial phenomena was about 1000 square miles.

Within this area, the positions of fifty Pleistocene glaciers
exceeding a mile in length were determined. Traces of several
smaller glaciers and of more than a dozen névé fields were also
found and mapped. Of the fifty larger glaciers, seven reached
the shore line of Lake Bonneville, and the moraines of at least
three of them are :clearly seen to- be partially buried by the
fluviatile deposits near the shore, or possibly by the shore
deposits of the lake itself.

The elevation of a catchment basin necessary to give rise to
a glacier in this region, was between 8000 and gooo feet, and,
except where the basins were very favorably situated, the latter
figure is more nearly correct.

The crest line of the Wasatch Mountains is near the eastern
border of the range. The valleys of the west slope are there-
fore much longer than those of the east, and the glaciers of the
two slopes were, in a general way, of corresponding lengths. The
number of ice tongues on the west was also much greater than on
the east. Of the fifty glaciers over one mile in length, forty-six
were west of the crest, and but four east of it. Of the ten glaciers
which reach or exceeded five miles in length, nine moved west-
ward and but one eastward. The Little Cottonwood glacier on
the west slope was twelve miles long, while the greatest length
reached by any glacier on the east slope was five miles. East
of the crest, one glacier descended to an altitude as low as 6000
feet, and two others descended to 7000 feet. On the west slope
fourteen glaciers descended to an altitude of less than 6000 feet,
and seven to 5000 feet.

The greater number and size of glaciers on the west side of
the range as compared with the east side, were determined by (1)
the larger catchment basins, and (2) the heavier snowfall. <A

720 ROLLIN D. SALISBURY

third factor, of local importance, was the accumulation of snow
in catchment basins among the lofty peaks of two east-west
divides, lying west of the main crest. These basins, in many
instances, furnished tributary glaciers to the main canyons, and
thus greatly increased the amount and strength of the work done
by the ice on the west side of the range.

Glaciation was not only more extensive, but also more vigor-
ous on the west side of the range than on the east. This is shown
by a more complete cleaning out of loose material from the
glaciated valleys of the west slope, by the more complete reduc-
tion of their asperities of surface, by the greater deepening of
the main canyons on this side, leaving their tributary valleys
“hanging” 200 to 300 feet above, and by the development of
more massive moraines. The moraines of the east-slope glaciers
are insignificant in comparison with those of the west slope
glaciers.

The work of the season removes all doubt as to the duality
of the ice age in the Wasatch Mountains. There were at least
two ice epochs separated by a long interglacial interval. Evi-
dence of more than two epochs was not found. The basis for
the above conclusion is as follows:

1. In several valleys there are outer moraines, much older
than the inner moraines of the same valleys. A considerable
tunnel exposure in one of these outer moraines showed essen-
tially all of the abundant granite bowlders, up to four feet in
diameter, so thoroughly disintegrated that they had been cut
through with picks and shovels in the excavation of the tunnel.
Most of them could be crumbled by the hand. This condition
of things was not superficial, but held to the depth of twenty
feet, the deepest point of exposure.

2. In certain glaciated canyons there are such variations in
the amounts of postglacial change (weathering, erosion, etc.)
which different portions of the drift and valley walls have suf-
fered, that it seemed necessary to postulate much longer expo-
sure for certain parts than for others. In all such cases, the
drift which appears to be older, extends beyond that which

GLACIAL WORK IN WESTERN MOUNTAINS f27

appears to be younger. The ice of the earlier epoch, therefore,
seems to have been more extensive.

3. Two distinct sheets of drift were found at several points
in the valleys of both the North and South Forks of the Ameri-
can Fork. These sheets of drift are separated by a soil twelve
to eighteen inches thick. In the same localities, the soil on the
surface of the upper drift sheet, formed since the last retreat of
the ice, is from four to six inches thick. So far as the thickness
of soil is a basis for estimating time, it would indicate that the
interval between the ice epochs was longer than the time since
the. last,

The topography of the region examined was somewhat modi-
fied by glaciation. The valleys which were occupied by ice are
usually U-shaped, and their slopes are commonly smoothed off
as far up the sides as the ice reached. Such forms are in sharp
contrast with the V-shaped canyons and rugged slopes of the
valleys not occupied by ice. In some of the glaciated valleys
massive moraines were built up. These are sometimes in the
canyons, and sometimes at their débouchures, according to the
position which the ends of the glaciers reached at the time of
their maximum extension. When tributary glaciers joined the
main, medial moraines were formed, and as the ice melted, these
moraines were left as ridges, parallel with the course of the val-
ley. Recessional moraines are frequently found crossing the
valleys as crescentic ridges, convex downstream. These ridges
often served as dams, above which lakes accumulated, rose, and
overflowed. The outlet streams of such lakes cut gorges in the
moraines. Near the heads of many valleys the ice gouged out
rock basins, 50 to 200 feet in diameter and from 5 to 20 feet in
depth. At least thirteen of the thirty-six glacial lakes mapped,
are in rock basins apparently made by the ice.

The relation of the moraines at the west base of the Wasatch,
to the fluviatile (or shore) deposits in the Bonneville basin, is
such as to indicate, as Gilbert has pointed out, that the last
advance of the ice from the mountains occurred during a late
period in the history of Lake Bonneville. The close correlation

728 Te OLENA SAVES BIOTA

of the earlier ice epoch with the earlier wide extension of the
Bonneville waters cannot as yet be asserted, although the data
at hand strongly suggest the contemporaneity of the early
expanded stage of Lake Bonneville and the early glaciation.

THE MOUNTAINS NEAR SANTA FE

Messrs. John E. Webb and William A. Averill spent
about one month in the mountains of New Mexico, a few miles
northeast of Santa Fé. The area studied was the Santa Fé
range of the Rockies, which traverses the northeastern portion
of New Mexico, running in a general north-northeast, south-
southwest direction through the adjacent parts of Santa Fé, Rio
Arraba, Mora, and San Miguel counties, between the parallels of
25° 45° and 36, and ‘the- meridians of 105.-35) sand 105" 50m
The aggregate area comprises about 250 square miles.

The height of the range in the area studied varies from 7000
feet to 13,306, feet, the latter being the altitude of the highest
of the Truchas peaks. The altitude which seems to have been
necessary for the generation of glaciers was 11,700 to 12,000
feet. The following peaks attained the requisite height: Lake
peak, 12,380 feet; Baldy peak, 12,623 feet ; Pecos Baldy (Cone
peak?),- 12,550 feet; Jrcarilla” peak, 12,944 feet; andthe
Truchas group, the highest peak of which is 13,306 feet. Clear
evidence of glaciation was not seen on lower peaks, though it
cannot be said that many of the lower peaks were studied in the
detail necessary to demenstrate the complete absence of glacia-
tion.

Heading against these peaks and their connecting ridges,
fifty cirques bearing evidences of glaciation were observed.
Névé fields, with incipient motion, existed at some points where
well-formed glaciers did not develop. Sixteen of the cirques
are in the group of mountains (Lake, Baldy, and associated
ridges) at the head of Santa Fé Creek; seven in the Pecos Baldy
group; twelve among the Truchas peaks; and fifteen about
Jicarilla peak, and the ridge between it and the Truchas peaks.
Of these cirques, twenty-five open toward the east, thirteen

GLACIAL WORK IN WESTERN MOUNTAINS 729

toward the north, seven toward the south, and five toward the
west. The cirques range from one-fourth of a mile to two miles in
width, and from one-eighth of a mile to two and one-half miles in
length, and from 150 to 1000 feet in depth. Most of them con-
tain lakes or ponds, some of them as many as six, and almost
all contain flats or bogs, indicating the former existence of
standing water, where there is none now. The total number of
ponds and lakelets seen was about eighty. Since they were seen
in the rainy season, the number of permanent bodies of stand-
ing water may be less.

The longest glacier track studied is in the valley of the Santa
Fé Creek. The ice emanating from three cirques united to form
a single glacier which Messrs. Webb and Averill think extended
some seven miles down the valley, its lower end reaching down
to an altitude of about 9200 feet, the lowest altitude, so far as
seen, to which the ice descended. The largest area of glacia-
tion from a single source is that on the east side of the Pecos
Baldy group, in the valley of Jack’s Creek, while the largest
continuous area of glaciation (to which the ice from several
sources contributed) is that along the group of mountains at the
headsor Santa Pe Creek:

The moraine matter left in the cirques, and in the valleys
below them, often has a pronounced hummock-and-hollow
topography. Its composition varies from point to point. The
chief accumulations are in the bottoms of the valleys. The
morainic matter is often disposed in ridges most commonly
roughly parallel to the axes of the valleys and to the direction
of ice movement, but often oblique or even at right angles to
this direction.

Stria were found on voches moutonneés in several places, while
smoothed and polished rock surfaces are of frequent occur-
Fence:

The higher ridges and peaks of the glaciated regions are
serrate, but serrate crests are confined strictly to the higher
parts of the range. Serration often characterizes the tops of the
mountains, on the sides of which the glaciated cirques le.

730 Ts OLLI AD SALE US 3 CG a

OTHER OBSERVATIONS

In addition to the foregoing work, some further data were
gathered at several points, touching Pleistocene geology in the
West. It was found that loess has somewhat extensive devel-
opment in eastern Washington and northeastern Oregon, though
its limits were not determined. It is widespread in Douglas,
Lincoln, Whitman, Columbia, Walla Walla and probably Spo-
kane counties, Washington. It is very generally ,distributed
over the northern part of Umatilla county, Oregon. In geo-
graphic distribution it seems to correspond, in a general way,
with the great wheat belt of the states mentioned. In topo-
graphic distribution, it has the general habit of the corre-
sponding formation in the Mississippi basin; that is, it has a
preference for considerable elevations. So far as seen, how-
ever, its disposition to follow streams is less pronounced than
in the Mississippi basin east of the Missouri. Its variations
in composition and structure, too, are similar to those of the
loess of the interior. Its thickness is very variable, and for
the most part undetermined, but locally it is at least twenty-
five to thirty feet thick, and its maximum is probably consider-
ably more. In many places, as at Bolles Junction (Walla Walla
county), it abounds in calcareous concretions of the usual loess
type. In other places, as at Alto (Columbia county), it assumes
the columnar structure, so characteristic of the loess of some
other localities where the formation is well known. In many
places, it constitutes the entire mantle rock. In some places it
is associated with sand, which may be either beneath it or inter-
bedded with it; and in other localities it contains considerable
beds of volcanic ash. The sand and volcanic ash associated
with it are sometimes distinctly stratified. In some cases, the
stratification is clearly the work of wind, but in others this was
not evident. The impression gained was that much of the loess
seen is of eolian origin.

Observations on glaciation were made at several points, aside
from those enumerated in the preceding pages. Data sufficient
for the intelligent direction of future Pleistocene work were

GLACIAL WORK IN WESTERN MOUNTAINS 73.

gathered for a number of localities where detailed work in the
near future seems desirable. In addition to preliminary studies
of this sort, glaciation was found to have occurred in some
localities from which it has not heretofore been reported. For
example, it was found that glaciers affected the north slope of
both the Spanish Peaks of Colorado. These mountains, 12,708
and 13,623 feet high respectively, appear to have represented
about the limiting conditions for glaciation, for the glaciers of
both mountains were small, and confined to their northern
slopes. One of the limiting conditions here was doubtless the
small area which could serve as a catchment basin on either
peak. Had the areas of these mountains been larger, more
considerable glaciation would doubtless have been developed at
this altitude.

It was found, among other things, that the detailed study of
glacial problems will locally have an important economic bear-
ing. An extensive placer mining plant is being established at
Breckenridge, Col., to wash the glacial moraine matter of the
Blue River valley, and the gravels of the valley train below.

Ro.uin D. SALISBURY.

REVIEWS

Preliminary Description of the Geology and Water Resources of the
Southern half of the Black Hills and Adjoining Regions in South
Dakota and Wyoming. By N. H. Darton. | Extract from
the Twenty-first Annual Report of the U. S. Geol. Surv.,
1899-1900, Pt. IV, Hydrography; pp. 489-599; Plates
LVIII-CXII; Washington, 1901. |

THE topics discussed in this paper are topography, stratigraphic-
geology, geologic structure, geologic history, water resources, minerals,
soils, climate and timber. The Black Hills rise out of the Great

Plains as a small group of forest covered mountains several thousand

feet high. Their central area, composed of mountains and parks, is of

crystalline schists and granites. Bordering the old rocks is a belt of
limestone plateaus surrounded in turn by the Red Valley, a depression
resulting from the etching out of a layer of softer rock —the Spearfish
formation of Triassic age. Beyond the valley and concentric with it
is the hogback range, due to a hard sandstone not weathered down.

Still farther outward are the Plains. Careful descriptions of the litho-

logic characters and the distribution of each formation follow; in

fact the paper is descriptive rather than theoretic, but does not fail to
call attention to broad features and to make generalized statements.
The oldest sediments are Cambrian. They are probably of later

Cambrian time and give one more item of evidence leading to the con-

clusion that Cambrian time was one of extensive submergence. Here

Cambrian sands were deposited along seashores and in estuaries with

gradual submergence and probable ultimate covering of the crystallines

with sediments of this age. The chief evidence is the presence of fine
grained deposits in the Upper Cambrian such as are not formed near
land. At present no Silurian or Devonian strata occur. The reason
is problematic. Either there was no deposition or else little deposition
and subsequent removal. Carboniferous time began with deep water
and marine conditions, and was consummated after the laying down
of 800-1000 feet of rock. ‘Triassic and Jurassic are both present and

732

REVIEWS 7238

distinguishable because an abrupt change in the character of the sedi-
ments and some erosion occur at the top of the former. Cretaceous
sediments have accumulated to the thickness of about 6000 feet and
consist largely of shales with some sandstone and a little limestone.
Tertiary rocks represented by the White River formation are mostly of
clay; fuller’s earth, sand, limestone, and grits occur. Still later
deposits are classed under earlier and later Pleistocene beds.

The structure of the region is that of an elongated low dome with
anticlinal wrinkles or spurs running off from the central area. The
maximum uplift was north of Harney’s Peak and amounted to gooo
Rees

One of the most admirable features of this paper is the abundance
of real geographic material embodied in it. The region described is
one of submaturely dissected, domed mountains in which all the sedi-
ments have been removed from the central portion leaving subdued
hills and broad parks of crystalline rocks. The concentric arrange-
ment of the different formations is due to the uplift and to the pro-
cesses of erosion which though similar for all strata, have been much
more effective in reducing some than others. Soft layers are found
etched out, hard layers standing up, but each and every one has its
peculiar forms. Thus geography follows stratigraphy.

Attempts by the author, more or less successful, to reconstruct the
geography of the locality at different periods and to follow it through
its phases, are persistently made, ‘and they yield an interesting element
in the work. In this connection there is a contour map showing the
present relief of the upper surface of the Dakota sandstone over a part
of the area. This kind of work forms a very desirable feature.

The considerable eastward dip away from the Black Hills, where
most of the formations outcrop, together with other necessary features,
gives good artesian conditions to this region. Many wells reaching
the Dakota and Lakota sandstones flow with good water. The scanty
rainfall on the plains eastward renders these wells very useful. Many
streams diminish or cease by sinkage when they cross the sands.
Along Cheyenne and Fall rivers there is some opportunity for irriga-
tion — good bottom lands and reservoir space, with fair water supply.

Soils are mostly residuary from underlying rocks. Overplaced soils
and sand dunes are found in places. ‘‘Sedentary soils” may be criti-
cised. Nosoilis sedentary in the sense thatit does not move. “ Soilin
place” or preferably “residual soil’? more adequately carries the

734 REVIEWS

meaning. The relation of soils to underlying rocks, and of crops to
soils make very instructive points. Considerable mineral wealth is
found in coal, gypsum, and lithographic limestone in large slabs.
These of sufficient size to be valuable are not obtainable elsewhere in
the United States.

The notes and diagrams on climate are good but disappointing
because they do not indicate the significance of the condition of the
meteorologic elements from a geologic point of view. There is little
utility in giving space in a geologic report to climatic notes unless
some use is made of them. The climate must have characteristic
influence on erosion, soil formation, soil transportation, etc. The posi-
tion of volcanic ash to the eastward of the volcanoes, now extinct, just
west of the Black Hills shows similarity of winds in the White River
time to those of the present time.

Why cannot the U. S. G. S. reports, in discussing such a region as
this, have a section on geography? It should embrace most if not
all of the section here devoted to topography, but should go farther
with the question, incorporating some that is said under soils, climate,
water resources, etc., and then discuss the influence of relief, soil and
water on products, transportation, distribution of plants and animals,

man, and his various occupations.
Gero. D. HUBBARD.
EASTERN ILLINOIS STATE NORMAL SCHOOL,

Charleston, Ill. 5

The High Plains and Their Utihzation. By WiLLaRD D. JOHNSON.
[Extract from the Twenty-first Annual Report of the U.S.
Geol. Surv., 1899-1900, Pt. IV, Hydrography; pp. 601-
768; Plates CXIII-CLVI. |

THe High Plains, as defined in the first chapter, correspond
approximately to the Central Plains. They form a belt extending
from central Texas northward across Oklahoma and western Kansas
into Nebraska, but are more characteristic in Texas because less
eroded. By topographic difference they constitute a topographic
unit of the great geographic unit—the Great Plains. The High
Plains practically have no drainage, hence their surface is in general
a dead level known locally as the “ Flats.”

Chapter 11 discusses the origin of these plains. They are the

REVIEWS 3%

remnants of an old débris apron built in the latter half of Tertiary
time, and preserved because of lack of sufficient precipitation to pro-
duce run-off. ‘The author takes the ground that this apron of waste
was deposited and arranged by streams, but shows that the streams
were not of the ordinary type. A careful analysis of arid region
stream work leads to the conclusion that “‘there is no such thing as
sheet flow’? in the sense of a uniform flood of water and waste, but
that the more or less perfect sheet is ‘‘one of intimately lacing threads
of current.”

The deposits of which the plains are composed are the result of
this stream work and not of still water work. As proof of the proposi-
tion, the author finds the following facts:

(1) Wide distribution, at all depths of gravels, which decrease in
size from the mountains. (2) The size of these coarse materials is
often that of cobbles. (3) Gravel courses run east and west and are
cross-bedded— marking the channel courses of eastward flowing
streams. (4) Sand bed salso occur in courses elongated east and west
in the “clays.” (5) Interlaced stream beds occur as shown where
erosion has disclosed them. (6) Even most clay beds are thin and
elongated east and west.

Mr. Johnson finds the “‘mortar-beds”’ cutting across the local
bedding of sands and gravels, and he considers the beds to have been
formed by the cementation of the coarse materials at the level of the
ground water by salts carried down in the sinking surface water. It
appears that his explanation of the distribution of water in these
gravels is similar to that of water distribution in the morainic materials
of Illinois and Indiana, as worked out by Leverett and others, and
stated by Leverett in the Eighteenth Annual Report of the U.S. Geol.
surv. Pt. 1V. |

Chapter III is a discussion of the “deficiencies of climate.” The
climate of the High Plains is described as subhumid. It varies from
humid to arid through a period of years. ‘The author very properly
states that the amount of precipitation is not the only factor in deter-
mining the climate, or even the arability of a region. The Dakota
wheat fields actually have less rainfall than the Staked Plains of Texas,
but the rainfall is more effective in Dakota. In Texas (1) rainfall is
more spasmodic; (2) temperature is higher, increasing evaporation
and decreasing relative humidity; (3) there are more hours of sun-
shine, and (4) greater wind movement. In the summary he reaches

736 REVIEWS

two conclusions which are thoroughly borne out by the facts and dis-
cussion. (a) A truer index of climate than rainfall is relative humid-
ity. (4) Observations covering a quarter of a century indicate that no
change in climate, save that of short-period oscillations, takes place.
While the High Plains area is a climatic unit, it shows evidence of
agreement in the series of climatic changes through Pleistocene, with
the Great Basin lake fluctuations, and with the advance and retreat of
the Rocky Mountain glaciers.

The fourth chapter gives a graphic picture of the boom of 1888—
1893, with the subsequent disasters, and some suggestions as to what
may be grown on the High Plains.

Chapter V proves the impossibility of general irrigation within the
High Plains. The facts are as follows: (1) The rainfall and streams
are wholly inadequate; (2) the mountain streams from the west are so
loaded with waste that ponds to retain their waters would be exceed-
ingly short-lived; (3) all mountain streams must pass through a
broad, arid, fertile strip amply sufficient to utilize all their waters; (4)
artesian water in very small quantity is found in the valleys, but the
entire supply is altogether too small for general irrigation. This
source, however, is sufficient to furnish water for a garden and ranch
headquarters at occasional intervals.

In the closing pages there is a discussion of the origin of peculiar
sink holes, and a theoretic explanation of the movements of under-
ground waters. This last brings together considerable valuable mate-
rial which formerly has been scattered and little known, but it presents
scarcely anything new.

The author thus states his purpose at the outset, and he accom-
plishes his object: ‘‘To show that the High Plains, except in insig-
nificant degree, are non-irrigable, either from streams, flowing or
stored, or from underground sources, and that, therefore, for general
agriculture, they are irreclaimable; but that, on the other hand, water
from underground is obtainable in sufficient amount for reclamation
of the entire area to other uses; that such reclamation has in fact
already begun, and is in progress of gradual, but sure development ;
and that it will be universally profitable.”

This paper will be of great service to those in the arid and sub-
arid regions, and to those who may contemplate immigration thither,
because it gives authentic information concerning fertility and water-
supply there. But the value of the paper is not alone in its descriptive

REVIEWS 73.

work. To science the able discussions of causes and effects, of climatic
relations to geology, geography, and agriculture, of water-supply, etc.,
are the most attractive features. The pages devoted to underground
water resources are specially strong. The only regret is that the paper

is not finished.
GEORGE D. HUBBARD.
CHARLESTON, ILL.

The Bauxite Deposits of Arkansas. By CHARLES WILLARD Hayes,
U. S. Geological Survey, Twenty-first Annual Report, Part
III, pp. 435-472. Maps and plates.

COMMERCIAL deposits of bauxite have been known in Arkansas
since 1890, when they were first discovered by the State Geological
Survey. The present report by Dr. Hayes represents, however, the
first detailed systematic study of these deposits, and its appearance
when developments are just beginning in the area, makes it a very
timely one.

Preliminary to the description of the Arkansas deposits proper,
the report opens with a brief summary of the “distribution of bauxite
deposits in the United States.”” The Arkansas deposits are limited
to a small area twenty miles long and five or six miles wide, lying
south and southwest of Little Rock, in the adjacent parts of Pulaski
and Saline counties. The most important deposits of the state are
grouped into two districts, with less important isolated ones between.
The general geologic and physiographic relations of the region are
recounted in some detail, derived largely, as the author says, from the
state survey reports. Three distinct groups of rocks are made out,
namely, the Paleozoic sediments in the northwest, the Tertiary and
recent sediments on the southeast, and areas of intrusive igneous
rocks in the two bauxite districts. ‘The Paleozoic rocks are similarly
folded and closely resemble those of the Alabaina-Georgia-Tennessee
area of the southern Appalachians. The region was probably several
times reduced to a nearly featureless plain, and sometime after the
folding, probably, during Cretaceous, the igneous rocks were, accord-
ing to Williams, intruded. Beginning with the Cretaceous, the region
was several times invaded by the sea and Tertiary sediments still form
a considerable part of the area.

In the detailed description of the deposits, the two districts are

738 REVIEWS

separately described in some detail, and certain peculiarities possessed
by each are pointed out. The few scattered isolated deposits are
likewise described.

In the Bryant district, which is the most southwesterly one, the
bauxite occurs in two distinct forms: (a) granitic bauxite, and (4) piso-
litic bauxite. The granitic bauxite forms the basal portions of the
beds, and in most cases rests immediately on a layer of kaolin, derived
by the ordinary processes of decay from the syenite. The bauxite is
spongy in texture with no trace of the pisolitic structure, but showing
partial traces of the granitic structure in which the individual feld-
spars are replaced by a porus skeleton of alumina. It is probable,
says Hayes, that this type of bauxite is in every case derived directly
from the syenite by the decomposition of the feldspar and the eleolite,
and the removal in solution of silica, ime and alkalis, the alumina
alone remaining of the original constituents. The pisolitic type is
more uniform than that of the Georgia-Alabama region and forms the
upper parts of the beds. The two forms of ore are not separated, as
a rule, in the same section, by any sharp and definite line.

In the Fourche Mountain district only the pisolitic type of the
ore has been found, which, when nearest the syenite margin, rests on
a layer of kaolin as in the Bryant district. Those deposits more dis-
tant from the syenite margin are probably interstratified with sedi-
mentary beds of Tertiary age.

The scattered isolated deposits found between the two districts
resemble in their mode of occurrence the Georgia-Alabama deposits.

The deposits are in beds or layers, which range in thickness from
zero to forty feet, and have a probable average thickness of ten to
fifteen feet for the two districts, and two to five feet and more in case
of the isolated bodies of ore.

In chemical composition the Arkansas bauxite varies within wide
limits. The granitic type is the purest, and in selected samples con-
tains less than 3 per cent. of silica and 1 per cent. of iron oxide, and
corresponds in composition to the formula Al,O,3H,O — the tri-
hydrate of alumina. In the white bauxitic kaolins the silica ranges as
high as 20 to 30 per cent., and in extreme cases the iron oxide reaches
50 per cent. in some of the highly ferruginous types of the ore.

Concerning the origin of the Arkansas bauxite deposits Dr. Hayes
says that they are so intimately associated with the igneous rocks of
the region that genetic relationship between the two is at once sug-

REVIEWS 739

gested. The characteristic pisolitic structure of the upper portion of
the deposits indicates chemical precipitation. The granitic bauxite
forming the lower or basal portions of the beds, and the bowlders, are
evidently of a different origin. The bauxite was probably laid down
on the syenite rather than on the kaolin, as there is no indication that
the kaolin is an intermediate product between the fresh syenite and
the bauxite.

In his report on the igneous rocks of Arkansas Dr. J. Francis Wil-
liams suggests two theories for the accumulation of the bauxite.
First, that the bauxite was formed by the decomposition of a bed of
clastic material, derived principally from the syenite. The second,
which he regarded as the most probable one, involved the action of
the waters of the Tertiary sea on the still highly heated igneous rocks,
in which, under high temperature and pressure, the constituents of the
syenite were dissolved and brought to the surface in solution, the
water emerging as hot springs. In the discussion of these theories
Dr. Hayes points out serious difficulties to both.

In the last part of the report Dr. Hayes treats of the ‘‘ Economic
‘‘Amount of Ore,’’ which includes
atabular statement of the estimate of the amount of ore in the Arkansas

?

Relations”’ under ‘“‘ Development ;’

bauxite region, and, according to the author’s calculations, shows the
total amount estimated in outcrops to be 6,608,500 long tons, and the
total amount under cover 43,711,200 long tons; “Quality of the
Ores ;”” and ‘‘ Mining and Preparation of the Ore for Market.”

THomas L. WarTSON.
DENISON UNIVERSITY,

Granville, Ohio.

RECENT FUBLICATIONS

—ADAMS, CHARLES C. Baseleveling and its Faunal Significance, with
Illustrations from Southeastern United States. [Reprint from The
American Naturalist, Vol. XXXV, No. 418, October, 1go1.] Ginn &
Company, Boston, Igol.

—BAKER, FRANK COLLINS. Some Interesting Molluscan Monstrosities.
Transactions of the Academy of Science of St. Louis, Vol. XI, No. 8.
Issued November 26, Igol.

—BLASDALE, WALTER C. Contributions to the Mineralogy of California.
[Bulletin of the Department of Geology, University of California, Vol. II,
No. 11, pp. 327-348.] Berkeley, November, 1go1.

—CoLeMAN, A. P. Glacial and Interglacial Beds Near Toronto. [Reprinted
from the JOURNAL OF GEOLOGY, Vol. 1X, No. 4, May-June, 1901.] The
University of Chicago Press.

Marine and Freshwater Beaches of Ontario. [Bulletin of the Geological
Society of America, Vol. XII, pp 129-146.] Rochester, March, Igor:

—COLLIE, GEORGE L. Physiography of Wisconsin. [Reprinted from the
Bulletin of the American Bureau of Geography, Vol. II, No. 3, Septem-
ber, 1901.]_ Winona, Minn., Igor.

—Congrés Geologique International. Comptes Rendus de la VIII® Session,
en France. Imprimerie Le Bigot Freres, Lille. Paris, 1go1.

—DarTON, NELSON HorATIO. Preliminary Description of the Geology and
Water Resources of the Southern Half of the Black Hills and Adjoining
Regions in South Dakota and Wyoming. [Extract from the Twenty-
first Annual Report, United States Geological Survey, 1899-1900. Part
I\V— Hydrography.] Washington, rgol.

—DINGEsS, J. Textbuch zum Relief der Nérdlichen Kalkalpen. 1:50000.
A. Fackler, Mindelheim (Siidbayern), 1901.

—Dupots, EUGENE. Paradoxe klimatische toestanden in het Paleozische
tijdvak, beschouwd in verband met den vroegeren aard der zonnes-
traling. [Overgedrukt uit de Handelingen van het Achtste Neder-
landsch Natuur- en Geneeskundig Congress. |

—ECcKEL, Epwin C. The Portland Cement Industry in New York. [Pub-
lished, by permission of the State Geologist of New York, in Engineering
News, May 16, rgot.]

The Snakes of New York State: An Annotated Check List. [Reprint
from the American Naturalist, Vol. XXXV, No. 410, February, tgot.]
Ginn & Company, Boston, Igol.

740

KRECENT PUBLICATIONS 741

—FINSTERWALDER, S., ET E. MureEtT. Les Variations Périodiques des
Glaciers. Sixiéme Rapport, 1g00. [Extrait des Archives des Sciences
physiques et naturelles, T. XII, tg01.] Librairie Georg & Ci*., Genéve.
Igol.

—GOULD, CHARLES NEWTON. The Dakota Cretaceous of Kansas and
Nebraska, and other papers. [From Vol. XVII, Transactions Kansas
Academy of Science.] Topeka, Igor.

—HARKER, ALFRED. Ice-Erosion in the Cuillin Hills, Skye. [Transac-
tions of the Royal Society of Edinburgh, Vol. XL, Part II (No. 12).]
(With a map.) Robert Grant & Son, 107 Princes Street, Edinburgh;
and Williams & Norgate, 14 Henrietta Street, Covent Garden, London,
IgOl.

—Hosps, WILLIAM HERBERT. The Newark System of Pomperaug Valley,
Connecticut. With a Report on Fossil Wood by F. H. Knowlton,
[Extract from the Twenty-first Annual Report U. S. Geological Survey,
1899-1900, Part I11—General Geology, Ore and Phosphate Deposits,
Philippines.]| Washington, Igol.

The River System of Connecticut. [Reprinted from the JouRNAL OF
GEOLOGY, Vol. 1X, No. 6, September—October, 190I.| The University
of Chicago Press.

—KFEILHACK, CONRAD. Geologisch-morphologische Ubersichtskarte der
Proving Pommern. Mit Benutzung der Aufnamen der Geologischen
Landesanstalt entworfen. In Vertrieb bei der Kéniglich Preussischen
Geologischen Landesanstalt u. Bergakademie, Berlin, N. 4, Invaliden-
strasse 44.

—KunZ, GEORGE F. The Production of Precious Stones in 1go0. [Extract
from Mineral Resources of the United States, Calendar Year 1900:
David T. Day, Chief of Division of Mining and Mineral Resources, U. S.
Geological Survey.] Washington, Igol.

—Lane, A.C. The Economic Geology of Michigan in its Relation to the
Business World. [Lecture delivered at the Field Columbian Museum,
Chicago, November 2, 1g01.] The Michigan Miner, Saginaw, Mich.,
December I, Igol.

—MERRILL, GEORGE P. Ona Stony Meteorite, which fell near Felix, Perry
County, Alabama, May 15, 1900. [From the Proceedings of the United
States National Museum, Vol. XXIV, pp. 193-198 (with Plates XIII,
XIV).] Washington, Igor.

—MOo.LenGRAAFF, G. A. F. Géologie de la République Sud-Africaine du
Transvaal. [Extrait du Bulletin de la Société Géologique de France.]
23, Rue serpente,. V1, Paris, 1901.

742 KE CE NITRO BLMGCATIONS,

—United States Department of Agriculture. Report No. 70. Exhaustion
and Abandonment of Soils. Testimony of Milton Whitney, Chief of
Division of Soils, before the Industrial Commission. Washington, Igor.

—United States Geological Survey. Twenty-first Annual Report, 1899—I900.
Part I, Director’s Report, including Triangulation, Primary Traverse,
and Spirit Leveling. Part VI, Mineral Resources of the United States,
1899; Metallic Products, Coal, and Coke. Part VI (continued), Mineral
Resources of the United States, 1899; Non-metallic Products, except
Coal and Coke.

—WARD, LESTER F. Geology of the Little Colorado Valley. [From the
American Journal of Science, Vol. XII, December, 1g01.]

—WARMAN, PHILIP CREVELING. Catalogue and Index of the Publications
of the United States Geological Survey, 1880 to 1901. [Bulletin No.
177, U.S. Geological Survey.] Washington, Igol.

—Washington Academy of Sciences, Proceedings of the.

Papers from the Harriman Alaska Expedition, XXV. The Algae.
[Plates XLIII-LXII.] By De Alton Saunders. Vol. III, pp. 391-
486, November 15, Igol.

Papers from the Harriman Alaska Expedition, XXVI. MHarrimanella, a
New Genus of Heathers. [Text Figures 62-66.] By Frederick V.
Coville. Vol. III, pp. 569-576, December 6, Igol!.

Papers from the Hopkins-Stanford Galapagos Expedition, 1898-9.
V. Entmological Results (5): Thysanura and Termitide. [Text
Figures 47-57.] By Nathan Banks, U.S. National Museum. Vol.
III, pp. 541-546, November 29, Igol.

Papers from the Hopkins-Stanford Galapagos Expedition, 1898-9,
VI. The Isopods. By Harriet Richardson, Collaborator, Smithsonian
Institution. Vol. III, pp. 565-568, November 29, Igol.

Descriptions of 23 New Harvest Mice (Genus Reithrodontomys). By G.
Hart Merriam. Vol. III, pp. 547-558, November 29, Igol.

Seven New Mammals from Mexico, Including a New Genus of Rodents.
By G. Hart Merriam. Vol. III, pp. 559-563, November 29, Igol.

Preliminary Revision of the Pumas (Felis Concolor Group). By C.
Hart Merriam. Vol. III, pp. 577-600, December 11, Igol.

—WoopwortTH, J. B. The History and Conditions of Mining in the Rich-
mond Coal-Basin, Virginia. [Transactions of the American Institute of
Mining Engineers (Richmond Meeting, February 1901).|

—WRIGHT, FRED. EUGENE. Die Foyaitisch-Theralthischen Eruptiveges-
teine der Insel Cabo Frio, Rio de Janeiro, Brasilien. | Tschermak’s
Mineralog. und petrographische Mittheilungen, Bd. XX, Heft 3.] Alfred
Holder, Rothenthurmstrasse 15, Wien.

[NDEX TO VotumeE LX.

Adams, Frank Dawson. Excursion to the Pyrenees in Connection with Eighth
International Geological Congress - - - - - -
Agriculture, Department of, for 1900, Year Book of the United States.
Review by C. - - - - - - - - - -
Allegany County, Maryland, The Paleozoic Formations of. Charles S. Prosser
Allgemeine Limnologie ; Handbuch der Seenkunde. F.A. Forel. Review by
IRS IDE Sh = - . - - - - - - - - -
Alpen, Die vierte Eiszeit im Bereiche der. Won Albrecht Penck. Review by
R--Dss. = - - - - . - - - - - -
American Association for the Advancement of Science, Denver Meeting.
Editorial by T. C. C. - - - - - - - - -
Ammonite-Genus from the Lower Carboniferous, Prodromites, A New. James
Perrin Smith, and Stuart Weller - - - - - - -
Anthracite Coal Field, Stratigraphical succession of the Fossil Floras of the
Pottsville Formation in the Southern. David White. Review by
Charles R. Keyes” - - - - - = - - - -
Arkansas, Physiography of the Boston Mountains. A. H. Purdue - - -
Arkansas, The Bauxite Deposits of. Charles Willard Hayes, U.S. Geological
Survey. Twenty-First Annual sues Part III. Review by Thomas
L. Watson - - - - - - - - -
Arkansas Valley, Composite Genesis of, Thro the Ozark Highlands. Charles
R. Keyes - - - : = - - - - -
Artesian Basins of Te roeinee A See maak on the. Wilbur C. Knight.
Review by R. D.S. - - - - - - - - -

Bain, H. F., Review of Annual Report for 1900. Iowa Geological Survey — -
Baldwin, Evelyn B. Observer Weather Bureau. Meteorological Observations
of the Second Wellman Expedition. Report of the Chief of the
Weather Bureau, United States Department of Agriculture, 1899-1900.
Part VIII. Review by T. C. C. - - - - - - -
Basis for Geologic Mapping, The Formation as the. Edwin C. Eckel - -
Bauxite Deposits of Arkansas. Charles Willard Hayes, U. S. Geological Sur-
vey, Twenty-First Annual Report. Part III. Review by Thomas L.
Watson - - - - = - > - > -
Beach Structures in the Medina Sandstone. H. L. Fairchild. Review by
N. M. F.-- - - - - - - - - - -
Beaufort’s Dyke off the Coast of the Mull of Galena H. G. Kinahan.
Review by N. M. F. - - : - - - - = - -

743

28

363
409

199

202

536

255

544

694

694

486

200

547

276
708
737
549

551

744 INDEX TO VOLUME IX

Becraft Mountain, Columbia County, New York. The Oriskany Fauna of the.
J. M. Clark. Review by 5. W. - - - - - - -

Review by Charles R. Keyes - - - - - - - :
Bedford as a Formational Name, The Use of. Edgar R. Cumings - - -
“Bedford Limestone,’ On the Use of the Term. C. E. Siebenthal < =
Bedford Limestone, The Term. Charles S. Prosser - - - - -
Biddle, H. C., The Deposition of Copper by Solutions of Ferrous Salts - -
Big Horn Mountains, Wyoming, Glacial Sculpture of the. Francois E.
Matthes. Review by R. D.S. - - - - - - -
Blatchley, W. S., State Geologist, Twenty-Fifth Annual Report, Department of
Geology and Natural Resources of Indiana. Review by C. E. Sieben-

thal - - - - - - - - - - - -
Border-Line between Paleozoic and Mesozoic in Western America. James
Perrin Smith - - - - - - - - - - :

Boston Mountains, Arkansas, Physiography of the. N. H. Purdue - A tc
Branner, J. C. Editorial: Ripples of the Medina Sandstone - - -
Review: A Record of the Geology of Texas for the Decade Ending
December 31, 1896. Irederic W. Simonds - - - - -

Zinc and Lead Regions of North Arkansas, Review by C. R. Keyes -

Calvin, Samuei, State Geologist, lowa Geological Survey, A. G. Leonard,
Assistant State Geologist. Annual Report for 1900. Review by H.F.B.
Campbell, John T. Evidence of a Local Subsidence in the Interior - -
Canada, Summary Report of the Geological Survey Department for the year
1900. Review by C. - - - - - - - - -

Canals, Flumes, and Pipes, The Conveyance of Water in Irrigation. Samuel
Fortier. Review by G. B. H. - - - - - - - -

Cave Earths, Nitrates in. Henry W. Nichols - - - - - -
Certain Peculiar Eskers and Esker Lakes of Northeastern Indiana. Charles
R. Dryer - - - - - - - - - - - -
Charpentier, Henri. Les Minéraux utiles et leurs gisements. Géologie et
Minéralogie Appliquées. Review by J. C. Branner - - - -
Chamberlin, T. C. On a Possible Function of Disruptive Approach in the
Formation of Meteorites, Comets, and Nebule - - - - -
Editorials: Death of George M. Dawson - - - - - =
Death of Joseph Le Conte - - - - - - - - -
Denver Meeting of the A. A.A. S. - - - - - - -
Geologic Classification and Nomenclature - - - - - -
Nomenclature in Geology - - - - - - - - -
Reviews: Annual Report of the Board of Regents of the Smithsonian
Institution - - - - - - - - - - -
Meteorological Observations of the Second Wellman Expedition by
Evelyn B. Baldwin, Observer, Weather Bureau. Report of the Chief

of the Weather Bureau, United States ee of Agriculture,
1899-1900, Part VII - - - = - - - -

On Rival Theories of Cosmogony. O. Fisher — - - = = =

PAGE

278
542
232

234
270

430

465

354

512
694
535

gI
634

547
437

357

361
236

123
198

369
267
439
536
633
267

466

276
458

INDEX TO VOLUME 1X

Summary Report of the Geological Survey Department [of Canada]

for the year 1900 - 2 - - - - = - - -

The Geological History of the Rivers of East Yorkshire. T. R. Cowper

Reed - - - = - - 2 6 - - -

The Norwegian North Polar Expedition aoa 1896. Scientific Results.

Vol. II. Edited by Fridtjof Nansen - - . - -

Year Book of the United States Department of meet for 1900 -
Chemical Study in Differentiation; Part II. The Foyaite-Ijolite Series of Mag-
net Cove. Henry T. Washington - Stee) |e - : -

Clark, J. M., The Oriskany Fauna of the Becraft Mountain, Columbia County,
New York. Review by 8. W. - - - - - - - -
Classification of the Waverly Series of Central Ohio. Charles S. Prosser -
Coleman, A. P. Glacial and Interglacial Beds near Toronto - - - -
Colorado, the Morrison Formation of Southeastern. Willis T. Lee = =
Composite-Genesis of the Arkansas Valley through the Ozark Highlands.
Charles R. Keyes” - - = - = - = - - -
Connecticut, The River System of. William Herbert Hobbs - - - -
Constituents of Meteorites.- I. Studies for Students. O. C. Farrington -
Constituents of Meteorites. II. Studies for Students. O.C. Farrington =
Copper, The Deposition of, by Solutions of Ferrous Salts. H.C. Biddle :
Correlation of the Kinderhook Formations of Southwestern Missouri. Stuart
Weller’ - - - - - - - - - - - -
Cosmogony, On the Rival Theories of. O. Fisher. Review by T.C. C. -
Cumings, Edgar R. The Use of Bedford as a Formational Name - - -

Darton, N. H. Preliminary Description of the Geology and Water Resources
of the Southern Half of the Black Hills and Adjoining Regions
in South Dakota and Wyoming. Review by George D. Hubbard — -

Davis, Charles A. A Second Contribution to the Natural History of Marl -

Dawson, George M., Death of. Editorial by T.C.C. - - - - -

Deposition of Copper by Solutions of Ferrous Salts. H.C. Biddle - -

Deposits of Arkansas, The Bauxite. Charles William Hayes, U.S. Geological
Survey, Twenty-first annual Reports, Part III. Review by Thomas
L. Watson” - - - oy ee Qe fs Se - : :

Derivation of the Terrestrial Spheroid from the Rhombic Dodecahedron.
Charles R. Keyes” - - - - - - - - - -

Discovery of a New Fossil Tapir in Oregon. Wm. J. Sinclair - - =

Disruptive Approach in the Formation of Meteorites, Comets, and Nebulez, On
a Possible Function of. T.C. Chamberlin - - - - -

Dryer, Charles R. Certain Peculiar Eskers and Esker Lakes of Northeastern
Indiana - - - - - ey) bs - - - - -

Lessons in Physical Geography. Review by N.M. F. - - - -

Eckel, Edwin C. The Formation as the Basis for Geologic Mapping — - -
EDITORIALS:

American Association for the Advancement of Science, Denver Meeting.

sliver On ge - - - - - - - - - - -

745

PAGE
357

360

273
363

645

278
205
285
343

486
469
393
522
430

130

458
239

73
491
267

430

737

740 INDEX TO VOLUME IX

Bates’ Hole. Wilbur C. Knight. (John C. Merriam) - - - =

Board of Managers of the Bureau of Geology and Mines, Missouri -
Classification and Nomenclature. T.C.C. - - = - - -
Cordilleran Section, Geological Society of America, Second Annual
Meeting of the. John C. Merriam - - Cs - - - -

Death of George M. Dawson. T.C. C. - - - - - -
Death of Joseph Le Conte. T.C.C. - - - - - - -
Drainage Features of California. Andrew C. Lawson. (John C. Mer-

riam) - - . - - - - - - - - -
Evidences of Shallow Seas in Paleozoic Time in Southern Arizona.
W.P. Blake. (John C. Merriam) - - - - - - -
Excursion, Geological and Geographical, in the North Atlantic - -
Feldspar-Corundum Rock from Plumas County, California. Andrew C.
Lawson. (John C. Merriam) - - - - - - - -
Geological Society of America, Cordilleran Section, Second Annual
Annual Meeting of the. John C. Merriam - = = : c
Geological Society of America, Thirteenth Meeting. J. P. I. - -
Geology of the Three Sisters, Oregon. H. W. Fairbanks. (John C.
Merriam) - - - - - - - - - - - -
Granites in the Klamath Mountains, On the Age of. Oscar H. Peay

(John C. Merriam) - - - - - - - -

Great Basin in Eastern California and Southwestern Neos The
Geology of the. H.W. Turner. (John C. Merriam) - - -

John Day Basin, A Geological Section through the. (John C. Merriam)
Neocene Basins of the Klamath Mountains. F.M. Anderson. (John C.
Merriam) - - = . - - - - - - - -
Nomenclature in Geology. T.C.C - - - - - - -
Origin of the Solar System - - - - - - - - -
Pedological Geology of California, A Sketch of. E. W. Hilgard.

(John C. Merriam) - - - - - - - - - -

Ripples of the Medina Sandstone. J. C. Branner - - - -
Sierra Madre near Pasadena, The. E. W. Claypole. (John C. Mer-

riam) - - - - - - - - - - - -

The Term Bedford Limestone. Charles S. Prosser - - - -
Eighth International Geological Congress, Excursion to the Pyrenees in
Connection with. Frank Dawson Adams - - - = 2

Eiszeit im Bereiche der Alpen. Die vierte, von Albrecht Penck. Review by
ReaD iS = - - - - - - - - - - -

Eskers and Esker Lakes of Northeastern Indiana, Certain Peculiar, Charles R.
Dryer - = = - - - 2 = - = - -
Evidence of a Local Subsidence in the Interior. John T. Campbell : -
Excursion to the Pyrenees in Connection with Eighth International Geological
Congress. Frank Dawson Adams - - : - S - -

Fairchild, H. L. Beach Structures in the Medina Sandstone. Review by N.
M. F. - = : = 2 Z = x Z x :

267
440

535

270

202

123
437

28

549

INDEX ‘TO VOLUME 1X

Farrington, O.C. The Constituents of Meteorites. I. Studies for Students
The Constituents of Meteorites. II. Studies for Students - . -
Structure of Meteorites. I. Studies for Students £ . - -
Structure of Meteorites. II. Studies for Students - - - -

Fenneman, N.M. Reviews: Beach Structures in the Medina Sandstone by H.

L. Fairchild = - = = 2 - = - - : - -
Lessons in Physical Geography by Charles R. Dryer - - = -
Some Notes Regarding Vaerdal, the Great Landship, by Dr. Hans

Reusch - - - - - - 7 - = = - =
The Beaufort’s Dyke off the Coast of the Mull of Galloway, by H. G.

Kinahan - - = - = = - - - - -

Ferrous Salts. The Deposition of Copper, by Solutions of. H.C. Biddle -

Fisher, O. On Rival Theories of Cosmogony. Review by T. C. C. - -

Floras of the Pottsville Formation in the Southern Anthracite Coal Field,

Stratigraphical Succession of the Fossil. David White. Review by

Charles R. Keyes” - - = - = - > - - -

Forel, F. A. Handbuch der Seenkunde; Allgemeine Limnologie. Review

by Ri Ss 0 - - - - : - - - - =

Formation as the Basis for Geologic Mapping, The. Edwin C. Eckel - =

Fossil Tapir in Oregon, The Discovery of a New. William J. Sinclair =

Foyaite-Ijolite Series of Magnet Cove. A Chemical Study in Differentia-

tion. Part II. Henry S. Washington - = > = - -

Fulgurites, A Study of the Structure of. Alexis A. Julien - - - -

Fuller, Myron L. Probable Representatives of Pre-Wisconsin Till in South-

eastern Massachusetts - = = - - - = = -

Galloway, The Beaufort’s Dyke off the Coast of the Mull of. H. G. Kinahan.
Review by N. M. F. - - - = - - = = -
Gannett, Henry. Profiles of Rivers in the United States. Review by G. B. H.
Geological Society of America, Thirteenth Annual Meeting. Editorial, J.
Bante - - - - - - . - - - : =
Geology and Water Resources of the Southern Half of the Black Hills and
Adjoining Regions in South Dakota and Wyoming, Preliminary
Description of. N.H. Darton. Review by George D. Hubbard -
Glacial and Interglacial Beds near Toronto. A. P. Coleman - - - -
Glacial Work in the Western Mountains in 1901. R. D. Salisbury = -
Glaciers, The Variations of. VI. Harry Fielding Reid - - - -
Gould, Charles Newton. Notes on the Fossils from the Kansas-Oklahoma
Red-Beds - = > > S - - - - = -

Hayes, Charles Willard. The Bauxite Deposits of Arkansas. U.S. Geologi-
cal Survey, Twenty-first Annual Report, Part III. Review by Thomas
L. Watson - - - - - - - - = - =
Heraclea, Zeiller’s Flora of the Carboniferous Basin of. An illustration of
Paleozoic Plant Distribution. Review by David White - = -

732
285
718
250

337

737

192

748 INDEX TO VOLUME IX

High Plains and Their Utilization. Willard D. Johnson. Review by George
D. Hubbard” - - a ? - - ay fe 5

Hobbs, William Herbert. The River System of Connecticut - : - -
Hochregionen des Kaukasus, Aus den. Gottfried Merzbacher. Review by J.
15 ill = - - - i - - - - 2 F -
Hubbard, George D.—Reviews: Preliminary Description of the Geology and
Water Resources of the Southern Half of the Black Hills and
Adjoining Regions of South Dakota and Wyoming. N. H. Darton

The High Plains and Their Utilization. Willard D. Johnson 7 -

Iddings, J. P.—Editorial: Thirteenth Meeting of the Geological Society of
America - - - - - - : = = e - -

Review: Aus den Hochregionen des Kaukasus. Gottfried Merzbacher -
Indiana, Department of Geology and Natural Resources of. Twenty-fifth
Annual Report. W.S. Blatchley, State Geologist. Review by C. E.
Siebenthal - = - - - = - : - 2 is
Individuals of Stratigraphic Classification. Bailey Willis : - - -
Interglacial Beds near Toronto, Glacial and. H. P. Coleman - : -
Irrigation Canals, Flumes, and Pipes, The Conveyance of Water in. Samuel
Fortier. Review by G. B. H. - = - > - = - -

Johnson, Willard D. The High Plains and Their Utilization. Review by
George D. Hubbard - - - - - = - - -
Julien, Alexis A. A Study of the Structure of Ainlaurttee - : -

Kansas-Oklahoma Red-Beds, Notes on the Fossils from the. Charles New-
ton Gould - - - - - = - S -
Kansas, The University Geological ee of, Vol. IV. Paleontology, Part II.
Samuel W. Williston. Review by Stuart Weller - - - -
Kaukasus, Aus den Hochregionen des. Gottfried Merzbacher. Review by J.
Pal: - - - = - - = - - = @
Keyes, Charles R. Composite Genesis of the Arkansas Valley ‘Heenan the
Ozark Highlands — - - - - - - - - - -
Derivation of the Terrestrial Spheroid from the Rhombic Dodecahedron
Reviews: Three Phases of Modern Paleontology: (I) Uintacrinus: Its
Structure and Relations, Frank Springer; (II) Oriskany Fauna of
Becraft Mountain, John M. Clark; (III) Stratigraphical Succession of
the Fossil Floras of the Pottsville Formation in the Southern Anthra-
cite Coal Field, David White - - - - - - - -

Zinc and Lead Regions of North Arkansas. John C. Branner~ -
Kinahan, H.G. The Beaufort’s Dyke off the Coast of the Mull of ealorer
Review by N.M.F. - - - - - - - - -
Kinderhook Formations of Southwestern ieee Correlation of the. Stuart
Weller” - - - - - - - - - - - -
Knight, Wilbur C. A Preliminary Report on the Artesian Basins of Wyoming.
Review by R. D.S. - - - - - - - - - -

PAGE

734
469

359

732
734

67
359

354
557
285
361
734
673
337
361
359

486
244

539
634

557
130

200

INDEX TO VOLUME IX 749

E
Landslip, Some Notes Regarding the Great, Vaerdal. Dr. Hans Reusch. is
Review by N. M.F. - - - : : 2 7 : 2 - 639
Lead Regions of North Arkansas, and Zinc. John C. Branner. Review byC.
R. Keyes - - - : = : 2 = 2 = - 634
Le Conte, Joseph, Death of. Editorial by T. C. C. - = = . = Ago
Lee, Willis T. The Morrison Formation of Southeastern Coloraio - - 343

Leith, C. k. Summaries of Current North American Pre-Cambrian Literature 79--441
Leonard, A. G., Assistant State Geologist. Samuel Calvin, State Geologist;
Iowa Geological Survey, Annual Report for 1900. Review by H.F.B. 547

Lime-Magnesia Rocks, Perknite. H.W. Turner - - - - - - 507
Lower Carboniferous, Prodromites,a New Ammonite Genus from the. James
Perrin Smith and Stuart Weller - . . - - . - 255

Magnet Cove, The Foyaite-Ijolite Series of; A Chemical Study in Differentia-

tion, II. Henry S. Washington - - - - - - - 645
Mapping, The Formation as the Basis for Geologic. Edwin C. Eckel - - 708
Marl, A Second Contribution to the Natural History of. Charles A. Davis - 491
McKinley, President, Memorial - - - - - - - - - 533
Medina Sandstone, Beach Structures in the. H. L. Fairchild. Review by

N. M. F. - - - - - - - - - - . - 549
Medina Sandstone, Ripple-marks. Editorial by J. C. Branner - - - 535
Memorial, President McKinley : - - - - - - - - 533
Merriam, John C. Cordilleran Section of the Geological Society of America,

Second Annual Meeting - - - - - - - . - 68
Mesozoic, The Border-Line Between the Paleozoic and the, in Western Amer-

ica. James Perrin Smith - - - - - . - - 512
Meteorites, Comets, and Nebulze, On a Possible Function of Disruptive Approach

in the Formation of. T.C.Chamberlin - - - - - - 369
Meteorites, The Constituents of. I. Studies for Students. O. C. Farring-

ton - : z : = = i = 5 5 = = 393
Meteorites, The Constituents of. II. Studies for Students. Oliver C.

Farrington - - - - - - - - - - - 522
Minéraux utiles et leurs gisements. Géologie et minéralogie appliquées. Par

Henri Charpentier. Review by J. C. Branner - - - - - 198
Miniature Overthrust Fault and Anticline, Illustrated Note ona. A. H. Pur-

due - - - - - - - - - - - - - 341
Monticuliporoidea, Problem of the. I. Frederick W. Sardeson - - - I
Monticuliporoidea, Problem of the. II. F. W. Sardeson - - - - 149
Morrison Formation of Southeastern Colorado, The. Willis T. Lee - - 343
Mountains, Glacial Work in the Western, in Igo1. R. D. Salisbury z - 718

Nansen, Fridtjof, The Norwegian North Polar Expedition, 1893 to 1896. Sci-
entific Results. Vol. I. Review by R. D. S. - - - - - 87
Violates eviews by eli. ©,.C.5 = - - - - - - re 273

Natural History of Marl, A Second Contribution tothe. Charles A. Davis - 491

750 INDEX TO VOLUME IX

PAGE
Nichols, Henry W., Nitrates in Cave Earths - : - - - - =e 230
Nitrates in Cave Earths. Henry W. Nichols - - - - - e230
Nomenclature in Geology. Editorial by T.C.C. - . - - - S, oy
North American Pre-Cambrian Literature, Summaries of Current. C. K.
Leith - - - - e 2 . = - : 79, 441
Notes on the Fossils from the Kansas-Oklahoma Red-Beds. Charles Newton
Gould - - - - - - - - - - > = 27,

Nutter, Edward Hoit, Sketch of the Geology of the Salinas Valley, California 330

Oregon, The Discovery of a New Fossil Tapir in. William J. Sinclair - = O02
Origin of the Phenocrysts in the Porphyritic Granites of Georgia. Thomas L.
Watson - - - - - - - - - - 97

Origin of the Yosemite Valley, The Pleistocene Geology of the South Central
Sierra Nevada with Especial Reference to the. Henry Ward Turner.

Review by R. D.S. - - - - - : - - - - go
Overthrust Fault and Anticline, Illustrated Note on a Miniature. A. H.

Purdue - - - - - - - = - - = Bylat
Ozark Highlands, Composite-Genesis of the Arkansas Valley ee the.

Charles R. Keyes - - - - - - - 486

Paleontology, Part II. Samuel W. Williston. The University Geological

Survey of Kansas, Vol. lV. Review by Stuart Weller - - - 361
Paleontology, Three Phases of Modern.—Review by Charles R. Keyes - - 5390
Paleozoic and Mesozoic, The Border Line between, in Western America.

James Perrin Smith - - - - - - - - - =) Bie
Paleozoic Formations of Allegany County, Maryland. Charles S. Prosser - 409
Paleozoic Plant Distribution, An Illustration of, Zeiller’s Flora of the Carbonif-

erous Basin of Heraclea. Review by David White — - - : = OR
Penck, Albrecht, Die vierte-Eiszeit im Bereiche der Alpen. Review by

IRS IDES) 3 - - - - - - - - - - > 20
Perknite (Lime-Magnesia Rocks). H.W. Turner - - - C - s. S07
Petroleum, Texas. William Battle Phillips. Review by C. E. S. - - = ORY,
Phases of Modern Paleontology, Three. Review by Charles R. Keyes - - 539
Phenocrysts in the Porphyritic Granites of Georgia, On the Origin of the.

Thomas L. Watson - - - > - - - - - 97
Phillips, William Battle. Texas Petroleum. Review by C. E.S. - - = (92%/
Physical Geography, Lessons in. Charles R. Dryer. Review by N.M.F. - 638
Physiography of the Boston Mountains, Arkansas. A. H. Purdue - - - 694
Plains and Their Utilization, The High. Willard D. Johnson. Review by

George D. Hubbard - - - - - - - - = eS 4

Polar Expedition, The Norwegian North, 1893-1896, Scientific Results.
Edited by Fridtjof Nansen:
Vol. I.. (Review by R. DoS. - - - - - - - -, 87
Vol. II. Review by T.C.C. - - - - - - - S Gh7/®
Possible Function of Disruptive Approach in the Formation of Meteorites,
Comets and Nebule, On a. T.C. Chamberlin - - - - =) 1360

INDEX TO VOLUME 1X 751
PAGE
Pottsville Formation in the Southern Anthracite Coal Field, Stratigraphical
Succession of the Fossil Floras of the. David White. Review by
Charles R. Keyes” - - - - - = - - - oP e544
Pre-Cambrian Literature, Summaries of Current North American. C. K.
Leith - - - = - - = - : - - 79, 441
Preliminary Description of the 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 George D. Hubbard - 732
Pre-Wisconsin Till in Southeastern Massachusetts, Probable Representatives
of. Myron L. Fuller - - . - - - - - Sepecvire
Probable Representatives of Pre-Wisconsin Till in Southeastern Massa-
chusetts. Myron L. Fuller - - - - - = - - 311
Problem of the Monticuliporoidea. I. Frederick W. Sardeson — - - - I
Problem cf the Monticuliporoidea. II. F. W. Sardeson - - - - 149
Prodromites, A New Ammonite-Genus from the Lower Carboniferous. James
Perrin Smith and Stuart Weller - - - - - - - 255
Prosser, Charles S. The Classification of the Waverly Series of Central Ohio 205
The Paleozoic Formations of Allegany County, Maryland - : 409
Purdue, A. H., Illustrated Note on a Miniature Overthrust Fault and Anticline 341
Physiography of the Boston Mountains - - - - - - 694
Valleys of Solution in Northern Arkansas - - - - - - 47
Pyrenees, Excursion to the, in connection with Eighth International Geolog-
ical Congress. Frank Dawson Adams - - - - - - 28
RECENT PUBLICATIONS - - - - - 92, 203, 280, 364, 467, 552, 642, 740
Red-Beds, Notes on the Fossils from the Kansas-Oklahoma. Charles Newton
Gould - - - . - - - - - - - - 337
Reed, F. R. Cowper, The Geological History of the Rivers of East Yorkshire.
Review by T. C. C. - - - - - - - - 360
Reid, Harry Fielding, The Variations of Glaciers, VI - - - - - 250
Reusch, Dr. Hans, Some Notes Regarding Vaerdal, The Great eae
Review by N. M. F. - - - - - - - 639
Reviews: Annual Report of the Board of eee of the Smithsonian Institu-
tions= 1 (G:) - - - - : - - - - - 466
A Preliminary Report on the Artesian Basins of Woes Wilbur C.
Knight. (R.D.58.)'- = - - = - - = - 200
Aus den Hochregionen des Kaukasus. Gottfried Merzbacher. (J. P. L) 359
Beach Structures in the Medina Sandstone. H. L. Fairchild. (N.
MES) = - - = - - - - - - - - 549
Conveyance of Water in Irrigation Canals, Flumes, and Pipes. Samuel
Fortier. (G. B. H.) - - - - - - - - - - 361
Department of Geology and Natural Resources of Indiana. Twenty-
Fifth Annual Report. W. S. Blatchley, State Geologist. (C. E.
Siebenthal) - - - . - - - - - - - 354
Die vierte Eiszeit im Bereiche der Alpen. Von Albrecht Penck. (R.
D..S.) - - - - - - . - - - > - 202

752 INDEX TO VOLUME IX

Geological History of the Rivers of East Yorkshire. F. R. Cowper ee
Reed EbvGiC.) Mir - - - - - -+ - 360

Geological Map of West Virginie I. C. White. - - - - - 640

Géologie et Minéralogie appliquées. Les minéraux utiles et leurs

gisements. Par Henri Charpentier. (J. C. Branner) - - - 1098
Glacial Sculpture of the Big Horn Mountains, Wyoming. Francois

Matthes. (R. D.S.) - - = - - - - - - 465
Handbuch der Seenkunde; Allgemeine Limnologie. F. A. Forel.

(R.D.'S))= - - - - 5 - - - - - - 199
Iowa Geological Survey. Samuel Calvin, State Geologist; A. G.

Leonard, Assistant State Geologist. Annual Report for 1900.

CELEB) = - > - - - - - - - S) Bayy
Lessons in Physical Geography. Charles R. Dryer. (N.M.F.) - - 638
Meteorological Observations of the Second Wellmann Expedition, by

Evelyn B, Baldwin, Observer Weather Bureau. Report of the Chief

of the Weather Bureau, United States ee se of Agriculture,

1899-1900, Part VII. (T.C.C.) - - - - 2 = 4276
Oriskany Fauna of the Becraft Mountain, Columbia oa New York.

J. M. Clark. (S. W.) - - - - - - - - 278

(Charles R. Keyes) — - - - - - - - 542
Preliminary Description of the Geology and Water Resources of the

Southern Half of the Black Hills and Adjoining Regions in South

Dakota and Wyoming. N.H. Darton. Extract from the Twenty-

first Annual Report of the United States Geological Survey, 1899-

1900, Part IV. (George D. Hubbard) - = = - - = | 732
Profiles of Rivers in the United States, Henry Gannett. (G.B.H.) - 363
Record of the Geology of Texas for the Decade ending December 31,

1896. Frederic W. Simonds. (J. C. Branner) - = - - - gi
Rival Theories of Cosmogony. O. Fisher. (T. C. C.) - - - 458
Some Notes Regarding Vaerdal; The Great Landslip. Dr. Hans

Reusch. (N. M. F.) = - - : : = 2 - a (5240)
Summaries of Current North American Pre-Cambrian Literature. C. K.

Leith - - - - - : - = - - - 79, 441
Summary Report of the Geological Survey Department [of Canada] for

the Year 1900. (C.) - - - 2 = = = - a 85%)
Texas Petroleum. William Battle Phillips. (C. E. S.) - - = 637
The Bauxite Deposits of Arkansas. Charles Willard Hayes. Twenty-

first Annual Report, United States Geological Survey, Part III.

(Thomas L. Watson) - S : : s - 2 E Ss BH
The Beauforts’ Dyke, off the Coast of the Mull of Galloway. H. G.

Kinahan. (N. M. F.) - - - - - - - - - 551

The High Plains and Their Utilization. Willard D. Johnson. Extract
from the Twenty-first Annual Report of the United States Geological
Survey, 1899-1900. Part IV. Hydrography. (George D. Hubbard) 734
The Norwegian North Polar Expedition, 1893-1896: Scientific Results.
Vol. II. Edited by Fridtjof Nansen. (T.C.C.) - - - =e 273

INDEX TO VOLUME IX

The Norwegian Polar Expedition, 1893-1806: Scientific Results.
Edited by Fridtjof Nansen. . Vol. !. (R. D.S.) - - -

The Pleistocene Geology of the South Central Sierra Nevada, with
Especial Reference to the Origin of the Yosemite Valley. Henry

Ward Turner. (R. D. 5.) - - - - - - - -

Three Phases of Modern Paleontology: (1) Uintacrinus; Its Structure

and Relations. Frank Springer. (II) Oriskany Fauna of Becraft
Mountain. John M. Clark. (III) Stratigraphical Succession of the

Fossil Floras of the Pottsville Formation in the Southern Anthracite

Coal Field. David White. (Charles R. Keyes) - . - -
University Geological Survey of Kansas, Vol. IV. Paleontology,

Part Il. Samuel W. Williston, (S. W.) - - - - -

Year Book of the United States Department of recence: for 1900.

(e:) . - - - - - - - - - - -

Zeiller’s Flora of the Carboniferous Basin of Heraclea: An Illustration

of Paleozoic Plant Distribution. (David White) - -- - -

Zinc and Lead Regions of North Arkansas. John C. Branner. (C. R.
Keyes) - - . - : - - - - - -
Rhombic Dodecahedron, Derivation of the Terrestrial Spheriod from the.
Charles R. Keyes - - - - - = - - = -

Rivers in the United States, Profile of. Henry Gannett. Review by G. B. H.
Rivers of East Yorkshire, The Geological History of the. F. R. Cowper Reed.
Review by T.C.C. - -— - - Pe

River System of Connecticut, The. William Herbert Hobbs - - - -

Salinas Valley, California, Sketch of the Geology of. Edward Hoit Nutter -
Salisbury, R. D. Glacial Work in the Western Mountains in 1901 - -
Reviews: Die vierte-Eiszert im Bereiche der Alpen. Von Albrecht
Penck - - - - - - - - - - - -

Glacial Sculpture of the Big Horn Mountains, Wyoming. Francois E.
Matthes - - - - - . - - . - . :
Handbuch der Seenkunde: Allgemeine Limnologie. F. A. Fore] -
Norwegian Polar Expedition, 1893 to 1896, Scientific Results. Edited

by Fridtjof Nansen. Vol.I - . - - - - -

Report on the Artesian Basins of ae A PaaS Wilbur C.
Knight - - - - - - - - - -

The Pleistocene Geology of the South Central Sierra Nevada, with
Especial Reference to the Origin of the Yosemite Valley. Henry

Ward Turner - - - - - - - - - -
Sardeson, Frederick W. Problem of the wen couivereices - - - -
Scientific Results. The Norwegian North Polar Expedition, 1893 to 1896.
Vol. I. Edited by Fridtjof Nansen. Review by R. D. S. - - -

Vol. II. Review by T.C.C. - - = - - - - -
Sculpture of the Big Horn Mountains, Wyoming, Glacial. Francois E.
Matthes. Review by R. D. S. - - = - - - -
Siebenthal, C. E., On the Use of the Term “ Bedford Limestone ”’ - -

j50

PAGE

87

90

754 INDEX TO VOLUME 1X

Reviews: Texas Petroleum, by William Battle Phillips - - - -
Twenty-fifth Annual Report, Department of Geology and Natural
Resources of Indiana. W.5S. Blatchley, State Geologist - - -
Simonds, Frederic W., A Record of the Geology of Texas for the Decade End-
ing December 31, 1896. Review by J.C. Branner - = e -

Sinclair, Wm. J. The Discovery of a New Fossil Tapir in Oregon - -
Sketch of the Geology of the Salinas Valley, California. Edward Hoit Nutter
Smith, James Perrin, and Stuart Weller. Prodromites, A] New Ammonite-
Genus from the Lower Carboniferous - - = - - E

Smith, James Perrin. The Border-Line Between the Paleozoic and the
Mesozoic in Western America - - - - - - = c

Smithsonian Institution, Annual Report of the Board of Regents of. the.
Review by C. - - - = = = } = = 2 =

Solar System, Origin of. Editorial - - - - - - - -
Solution, Valleys of, in Northern Arkansas. A.H. Purdue - = - =
Springer, Frank. Uintacrinus; Its Structure and Relations. Review by
Charles R. Keyes” - - - - - - - - - -
Stratigraphic Classification, Individuals of. Bailey Willis — - - . -
Stratigraphical Succession of the Fossil Floras of the Pottsville Formation in
the Southern Anthracite Coal Field. David White. Review by

Charles R. Keyes - - - - - - - : - -
Structure of Fulgurites, A Study of the. Alexis A. Julien - - - =
Structure of Meteorites. I. Studies for Students. O.C. Farrington - -
Structure of Meteorites. II. Studies for Students. O.C. Farrington - -
STUDIES FOR STUDENTS: The Constituents of Meteorites. I. O.C. Farring-
ton); = = > = = = - - - - - 5

The Constituents of Meteorites. II. Oliver C. Farrington - 2 -

The Structure of Meteorites. I. O.C. Farrington — - = : -

The Structure of Meteorites. II. O.C. Farrington’ - - - -

Study of the Structure of Fulgurites. Alexis A. Julien - - - c ;
Subsidence in the Interior, Evidence of a Local. John T. Campbell =

Tapir, in Oregon, the Discovery of a New Fossil. William J. Sinclair - -
Terrestrial Spheroid from the Rhombic Dodecahedron, Derivation of. Charles

R. Keyes - - - - : : - . - - - -
Texas Petroleum. William Battle Phillips. (C.E.S.) - - - = =
Texas, Record of the Geology of, for the Decade Ending December 31, 1896.

Frederick W. Simonds. (J.C. Branner) - - - - - -
Turner, H. W. Perknite (Lime-Magnesia Rocks) - - - - - -

Uintacrinus: Its Structure and Relations. Frank Springer. Review by
Charles R. Keyes” - - - - - - 2 Z 2 :

Vaerdal; Some Notes Regarding the Great Landslip. Dr. Hans Reusch.
Review by N. M. F. - - - = - : a - : 4

PAGE

637
354

gI
702
330

255
512

466
440
47

539
597,

544
673

51
174

393
522

51
174
673
437

702

244
637

gl
507

539

639

INDEX TO VOLUME IX WS

PAGE
Valleys of Solution in Northern Arkansas. A. H. Purdue - - - - 47
Variations of Glaciers. VI. Harry Fielding Reid - - - - 250

Washington, Henry S. The Foyaite-Ijolite Series of Magnet Cove. A Chem-
ical Study in Differentiation - 2 - - - = - 645
Water Resources of the Southern Half of the Black Hills and Adjoining
Regions in South Dakota and Wyoming, Preliminary Description of
the Geology and. N.H. Darton. Review by George D. Hubbard 732
Watson, Thomas L. On the Origin of the Phenocrysts in the Ai hae
Granites of Georgia- - - - = - - - - 97
Review of the Bauxite Deposits of Arkansas, by Charles Willard Hayes,
United States Geological Survey, Twenty-first Annual Report, Part

Ill - - - - - - - . - - - - S727
Waverly Series of Central Ohio, Classification of. Charles S. Prosser - - 205
Weller, Stuart, and James Perrin Smith; Prodromites, A New Ammontie Genus

from the Lower Carboniferous - - - - - - - - 255
Weller, Stuart, Correlation of the Kinderhook Formations of Southwestern

Missour1 - - - - - - - 2 - - - = 130

Reviews: The Oriskany Fauna of the Becraft Mountain, Columbia County,

New York, by J. M. Clark - - - - - - - - 278
The University Geological Survey of Kansas, Vol. IV. palconaleay Part

Il. Samuel W. Williston” - . - - - - : - 362

Wellman Expedition, Meteorological Observations of the Second, by Evelyn B,
Baldwin, Observer Weather Bureau. Report of the Chief of the
Weather Bureau, United States Department of Agriculture, 1899-

1900, Part VII. Review by T. C.C. - - - - - - 276
West Virginia, Geological Map of, I. C. White - - - - 640
White, David. Review of Zeiller’s Plant Flora of the Carboniferous Basin of

Heraclea. An Illustration of Paleozoic Plant Distribution - - 192

Stratigraphical Succession of the Fossil Floras of the Pottsville Forma-
tions in Southern Anthracite Coal Field. Review by Charles R.

Keyes - > - - - - - - - - 544
White, I. C., Geological Map of West Virginia - - - = - - 640
Willis, Bailey, Individuals of Stratigraphic Classification - - - - 557

Yorkshire, the Geological Set of the Rivers of East, F. R. Cowper Reed.
Review by T. C. C. - - - - . - . - 360

Zinc and Lead Regions of North Arkansas, John C. Branner. Review by C.
R. Keyes - - - = : 2 ss 3 A a =. 634

‘o) me
& il 1 I ij oe
| \ i

dh

|

by | [

I

OED,
A
8

Pay
fe
ip 3

So

\|

2 7

lL

| s O,
(@)

ye i &, |
; ik eS
ty, es aN iy

x

Sy
Zs
Ay
i Tw
a TON

nie ao

a

Pe

ee
ar
i

EI

Fy

,

uN
PN
rae fees |

SN Age

alt
rein
lie

i
! \'

; =)

% SS
te

2 ae

‘
: ri i i ;
, “a
4

XSF
eR
Ne ‘at

pis
naire
welt =
ee te &
pi eos,
7, ¢,

Ni eee |
Es
AON i
; Naren lf
st creel me
i “ “ited pest

A
>
Sia
a
ae

2S

i

if

3 ergy 0
ee :. bs

Sie gS

VA OF yi
iS, 877
NL ne

wy Mahe
A
Rr Ose
erat
‘a
es
1
\

Pal

oe

Bt

“N
wh

(

Ke |
¥ *

a
“AOL
OLD
Es Sonia
Pee
eRe

Ad.

is

Je
a eS)

oe

e
ve

ty, Ne
hi O)

‘se Pe iy
Se

ip, eas
is es
on eerie y

Se

Channa
NS

\.

ime
—

See

aA
a

{ Cea \ MOWER Aa ie t { a Ao
Whe Wed 1

TSSOP ay ea (
YA My TOA TNT |
} Wa iat Wy

{ fi a
i\ AHA AC OAL
} pag Wake ea ed CH P|
el hu {
' WW ty Wye
\\ AVIV Aa Tey AY
\( Hea a yeti { i
} i j ,
{ EN MAVAEG edie tutte aie
Ht iM Wo ayia
{ Myre Pata

PAIN AVANT SL RECS SPO OP PU th ac

oy PUTAS RT RH CUTE UR RCN

‘ WOM AE AVAE ii LARUE 4 AEH ‘
WA Van (rue ( 4 ht

UA RAGANG Ait (hy Vitali) H {

it { AUN wat f 4
ONY Bann \ (at ((:

) ( ay Hab ghnd WALL Helin sat AL
i \ FALUN CRAITATTY ERA ART ERA A ORS NA ORI LDS OH
POL ada WUMIEGATRSIP AAU ALA RUA UA A ttt (
{ Mart cust { an WAG Reed

WARGO Goalie AAEM LYG CHEK CUI iy ORAS GOR PT Ga
UAW PAA AUN UN nih, Wa \

AU A Ad HWE eal Bash i A ald

BAD AD ALN OU IACO AC IN Aen! A Niatistiadatindal

Uj) {OV ANAL WAAL OAK dd RUA ae Oe Ne Pa a ER

Wel vaW AY, Mat

WCAG

4 eG)

By AOA A

Ube (ha

VU Pie ll
OR

WALA ENA ie
mipeevietbid ras yaa tV I lait
Nghia hu Pri AW (ete Ah yf
\
Wa Up AE Wied
shh UE I Ut as Al |
‘ Vase hear tict Wiel A} AMA
i (esipwie ines AN YT TER TES TPR km (a)

“ WN BAC
ri UTA
PUT RUT GeL
HUN AL et
Halo Gob at
yi AIA
f Way
Wavy
yh ‘
Vaan at
AM A
We AL
{ Haale
AA A Maes
fi { Vliet
Spt vy
WA A

i

¥oUnk

holo Whi
Woon
i iy

VARA ARUP ARG SMITHSONIAN INSTITUTION LIBRARIES

3 9088 01366 9973

Ih