qtet,
ait

( EB ee Sea) : 39 @)sGk-o he D: ee

LIBRARY OF

R.D. LACOE.

For the Promotion of Research in

PALEOBOTANY: and PALEOZOOLOGY

RETURN TO

SMITHSONIAN INSTITUTION

WASHINGTON, D. C.

bbejadeiat

QUARTERLY JOURS AEN

OF

THE GEOLOGICAL SOCIETY OF LONDON.

VoL. XXX.

1. Description of the Sxuxt of a Sprcies of HarirnErtum (H. Can-
HAMI) from the Rep Crae of Surrorx. By Witrram Henry
Frower, Ksq., F.R.S., F.G.S., &c., Hunterian Professor of Compa-
rative Anatomy, and Conservator of the Museum of the Royal
College of Surgeons of England. (Read November 5, 1873.)

[Puate I.]

Waite looking, a few weeks ago, over the very rich collection of
Crag fossils formed by the Rev. H. Canham, of Waldringfield, near
Woodbridge, that gentleman called my attention to an unusually fine

fragment of a skull, which he had been unable to identify with any |

known form. He very obligingly allowed me to bring it to London
for the purpose of careful examination and comparison ; and I have
the pleasure of exhibiting it to the Society this evening.

The specimen was found in the so-called “ coprolite ” or bone-

bed at the base of the Red Crag at Foxhall, about two miles from
Waldringfield; and it presents the usual aspect of the mammalian

remains from that bed. It is heavily mineralized, of a rich dark.
brown colour, almost black in some parts, with the surface much —

worn and polished, and marked here and there with the characteristic
round or oval shallow pits, the supposed Pholas-borings. Unfortu-
nately, before it was extracted from the matrix in which it lay, it
was broken by the pick into several pieces, some of which were lost
by the workmen ; but all that were preserved have been skilfully
reunited by Mr. Canham.

The great interest of this skull consists in its affording the first
recorded evidence of the former existence of an animal of the remark-
able order Sirenia in this country.

The fragment consists of the anterior or facial portion of the
cranium, which has separated, probably before fossilization, from the

3. G. Ss Nos Ly: B

/
fy

y
]

“yY

%
\

oY

2 W. H. FLOWER ON A HALITHERIUM

posterior part at the fronto-parictal suture, and in a line descending
vertically therefrom. This portion has then been subjected to severe
attrition, by which the greater part of the premaxillary rostrum, the
orbital processes of the frontals, and the zygomatic processes of the
maxillaries, and other projecting parts, have been removed. In
consequence of this, what may be called the external features of the
skull, which are especially necessary to determine its closer affinities,
are greatly marred, though enough remains of its essential structure to
enable us to pronounce with confidence as to its general relationship
to known forms. Fortunately the whole of the portion of the maxille
in which the molar series of teeth are implanted is preserved ; and
though the teeth have fallen from the alveoli in the front part of the
series, and in the posterior part are ground down to mere stumps, so
that the form of the crowns cannot be ascertained in any way, many
important dental characters may still be deduced from the number,
form, size, and position of the sockets and roots that remain.

As the intensely hard ivory-like rostra of the ziphioid cetaceans,
the tympanic bones of the Baleenidee, and the teeth of terrestrial
mammals almost alone remain in these deposits to attest the former
existence of their owners, it is doubtless to the extreme massiveness
and density of the cranial bones, so characteristic of the order Sirenia,
that we owe the preservation of so large a portion of the skull, under
the very unfavourable conditions to which, in common with the
other fossils of this formation, it must have been exposed.

A comparison of the fragment with the skulls of the two existing
forms of Sirenia, the Manati (Manatus) and the Dugong (Halicore),

gives the following results.

As regards size, the skull must have been considerably larger than
that of either of those animals, as the following comparison of some
of its dimensions with those of a full-sized American Manati and

an Australian Dugong will show :—

Halitheriwm. | Manatus. Halicore.
Teng nae upper gua | from meétre. métre. metre.
poole peealranmioe | pais | ove | ele
APCLbULE ...........eeeeceee ener
surface of frontal bones to 0-153 0:120 : 0-115

Height of skull from mr

lower surface of palate-bones
Width of skull in temporal 0:070 0-050 0:070
fosse, narrowest part ......
Width inside anteorbital fora-

0-095 | 0-062 6-058
nthe ayckeprnasare eesandsuBDoe
Width between widest part of

outer border of maxillary 0-114 0-064 0:067

BES BH ss Tiel a

The posterior surface of the fragment (fig. 1) shows in its upper

~ part the concave impression of the anterior cerebral lobes, with a

* Something should be allowed for wearing of the edges of the bone to make
the comparison exact. ‘

FROM THE RED CRAG OF SUFFOLK. 3

deep median hollow; but the crista galli and the cribriform plates
are broken away, giving full view into the nasal cavities, divided as
usual posteriorly by a horizontal partition (P.) into an upper or
“ olfactory chamber ” and a lower or “ narial passage.” The lateral
portions of this partition formed by the palate-bones(PZ.) still remain,
but its central part, together with the whole of the vomer, has dis-
appeared. In the middle of the roof of the olfactory chamber the
upper part of the stout septum formed by the strongly ossified
mesethmoid (VE#.) remains; and a prominent longitudinal ridge
(£T.) on the lateral wall, but preserved on the right side, is all that
indicates the turbinals. The whole of the sphenoids and their
dependencies and the pterygoids are broken away, the fracture ex-
tending through the bodies of the palate-bones, just behind the pos-
terior molar teeth.

The upper surface (fig. 2) includes the whole extent of the frontals,
except the antero-lateral processes which form the upper margin of
the orbit, which have been worn away. It is probably also due to
attrition that the general surface is evenly convex from side to side,
instead of laterally ridged and flat or even concave in the centre as
in the existing species. The anterior edge of this surface, forming
the upper boundary of the anterior nares, is evenly arched, showing
no median process as in the Manati, and to a less extent in the
Dugong ; but this may also be due to the wearing off of the thin part
of the edges of the bones. A nearly semicircular suture an inch
behind this margin appears plainly to mark off a distinct pair of
nasal bones (Va.), joining each other in the middle line and therefore
widely differing from their rudimentary or suppressed condition in
recent Sirenians. But, as is well known, in some species of Hali-
therium they are developed as conspicuously as in the present
example*. The size and form of the anterior nasal aperture (fig. 3)
is characteristically Sirenian, and in its height compared with its
breadth more like that of Manatus than Halicore. Its boundary on
the right side by the ascending process of the preemaxilla (P Mw.) is
well seen ; but on the left this is broken away ; and, as before men-
tioned, the anterior boundary with the greater part of the rostrum
has unfortunately perished. The floor of the cavity has precisely
the same general form as in the recent Sirenians.

The lateral surface of the fragment (fig. 4) shows the inner wall
of the anterior part of the temporal fossa, and of the orbit, the
prominent margins of the last-named cavity and the zygoma being
removed, and the great anteorbital foramen thus converted into an
open groove. This region presents a striking difference of conform-
ation from both of the existing genera. In Halicore, in the dried
skull, there is a very large vacuity between the frontal and maxillary
bones, connecting the orbital cavity with the nasal fossa. In the
genus Manatus this vacuity is greatly reduced, the thin edges of the
two bones, in the form of delicate paper-like laminz, uniting more
or less completely in different skulls. In the present fossil specimen

* H. Schinzi, Kaup, ‘ Beitrage zur naheren Kenntniss der urweltlichen Sau-
gethiere,’ Heft ii. 1855, tab. ii. fig. 2.
B2

4 W. H. FLOWER ON A HALITHERIUM

the orbit is completely separated from the nasal fossa by a stout bony
wall nearly an inch in thickness, across which the suture between
the frontal (/r.) and the maxillary (Ww.) can be clearly traced.
What remains of the anterior root of the zygoma (Zy.) shows that
this process must have been considerably more massive than in either
of the existing species. ‘The foramina of this region of the skull
correspond generally with those of the Manati. Posteriorly several
grooves run horizontally forward from the broken part of the skull ;
one of these (a) conducts the optic nerve to the orbit. Further for-
ward, in the suture between the frontal and maxillary bones (at 6) is
a considerable-sized foramen (apparently accidentally enlarged on
the right side) for the passage inwards of the palatine branches of
the second division of the fifth nerve In front of the anteorbital
foramen, in the suture between the maxilla and ascending or nasal
branch of the premaxilla, is a foramen (c).which transmits the
branch of the same nerve and the vessels which supply the horny
plate, which doubtless existed on the under surface of the rostrum.
This canal is again exposed (at d) in its course between the bones,
by the destruction of the anterior extremity of the skull.

The inferior surface (fig. 5, two thirds natural size) shows the
palate and sockets of the molar teeth. The former is elongated and
narrow, as in the existing Sirenians, but rather broader posteriorly
than in front. Its surface is quite flat, and rounded at the edges, owing
to attrition. The posterior edge is concave in outline, but more evenly
rounded than in either Dugong or Manati, in both of which a deep
V-shaped notch exists in the middle line, entirely wanting in the
present fossil*. The suture between the palatine (PZ.) and maxillary
bones (Mzw.) is very distinctly seen, the extent of the former being
much the same as in the modern Sirenians. Near the middle line,
just in front of the suture, are a pair of foramina, also found in the
corresponding situation in Manatus ; and further forward are several
irregular foramina, which would indicate a considerable vascular
supply to this part of the palate, as in the recent genera.

Dentition—The number of maxillary teeth, as indicated by the
alveoli, is six on each side, all placed in close contact with each
other. The two most anterior have simple rounded alveoli about
equal in size (0-017m. in diameter), indicating teeth with single
cylindrical roots as in the Dugong, but differing in being rounded at
the apex. The third appears to have had two roots, outer and inner,
each rounded and smaller than those in front. The three following
teeth resemble each other in form, though the most anterior or
fourth of the whole series is rather smaller than the other two.
They have three roots—one inner, compressed from side to side, and
two supporting the outer border, each compressed from before back-
wards exactly as in the Manati. The general form of the tooth
appears wider in proportion to its antero-posterior diameter than in
that genus; but this is perhaps more apparent than real, as the
minor root, as in the Manati, diverges considerably from the others,
so that the deeper the section the wider the whole base of the tooth

* The notch shown in the figure appears to be due to a fracture.

FROM THE RED CRAG OF SUFFOLK, 5

appears. The teeth, on the whole, are considerably larger in pro-
portion to the size of the skull than in Manatus, as well as fewer in
number, but differ still more fundamentally in presenting a distinction
in character in the fore and hind part of the series, a division into
simple premolars and more complex molars, if these terms can be
correctly employed when we have no knowledge of a succession of
the teeth. ;

It will be seen from the foregoing description that the specimen
presents many characters common to the Manati and the Dugong,
and others by which it differs from both, the most striking of them
being the more normal development of the nasal bones and the outer
wall of the nasal fossee, and especially the dentition, in all of which
it shows a more generalized condition. From Rhytina it is removed
still further, as that genus is characterized by entire absence of
maxillary teeth.

It will be necessary now to consider its relation with the extinct
Sirenians of European Miocene and Pliocene deposits.

These are generally known collectively under the generic name
Halitherum, Kaup, though several more or or less marked subdivi-
sions have been established. The genus is characterized by its
founder, as far as the teeth are concerned, by having “two (as in
Halicore) tusks in the steeply decurved preemaxille, and six tubercular
molars with closed roots, increasing in size from before backwards ’””*.
Of the preemaxillary rostrum or tusks of our present specimen we
know nothing; but the general characters of the molars certainly
correspond with those of Halitherium. In that genus, however,
theree ar two forms—one to which Professor Capellini has given the
name of Felsinotherium, founded on a very perfect specimen from
the Pliocene beds near Bolognat+ and including the French Pliocene
species figured by Gervais as H. Serresiit. From these the present
specimen decidedly differs in the absence of the deep notch in the
posterior edge of the palate, in the position of the anterior root of
the zygoma, which rises from the maxilla opposite to the penultimate
and antepenultimate molars, whereas it is much further forward in
the above-mentioned species, and in the larger size and greater
number of molar teeth, which do not appear to exceed five in either
the Italian or French species assigned to this group.

On the other hand, it approaches more nearly to H. Schinz, Kaup,
from the Miocene of the Rhine valley, which appears to have had
six molars, the first two having single roots. Comparing it, however,
with Kavup’s figures § and with specimens in the British Museum,
distinctions certainly of specific value can be seen, particularly in
the superior size and massiveness of construction, in the form of the
nasal bones and their relation to the upper ends of the preemaxille,
the form of the hinder edge of the palate, and the greater size of

* Beitrage zur naheren Kenntniss der urweltlichen Saugethiere. Zweites Heft
(1855), p. 10.
t “Sul Felsinotherio.” Bologna: 1872.
a Paotoeie et Paléontologie Frangaises, 2nd ed. (1859), pls. iv. v. and vi.
yp. cit.

Part

6 W. H. FLOWER ON A HALITHERIUM

the molar teeth, both absolutely and relatively to the cranium
generally.

Very few remains of Sirenians have hitherto been found in Belgium.
My esteemed friend Professor Van Beneden has recently described and
fieured the much-mutilated occipital portion of a skull of an animal
apparently of this order, to which he has given the name of Crassi-
therium robustum*. It is obvious that there are no materials for
instituting a comparison between this and our present specimen.
M. le Vicomte Du Bus has given a preliminary notice of the discovery
of the remains of a species of Halitherium in the clay at Boom,
between Antwerp and Brussels. But of the head unfortunately
very little was found, and no detailed description has been published.
And the same may be said of a nearly complete skeleton discovered
by M. Lesseliers at Basel, near Rupelmonde, in the same neighbour-
hood, referred to in Professor Van Beneden’s memoir quoted above.

A very thorough examination and comparison of all remains of
extinct Sirenians scattered throughout the various continental mu-
seums must be made before their specific distinctions can be satis-
factorily ascertained. It is quite possible that the present specimen
might then be found to belong to one of the already-named species ;
but, as has just been pointed out, it presents characters in which it
does not agree with any of those which are sufficiently preserved and
well-described to admit of comparison. I think it will be most
convenient to bestow upon it a specific designation, and therefore
propose that of Canhami, to commemorate the assiduity with which
its owner has collected, and liberally made available for scientific

* examination, the paleeontological treasures of the neighbourhood in

which he resides.

The subject of the origin of the very mixed fauna of the Red Crag
bone-bed has been so frequently and ably discussed in the publica-
tions of this Society and elsewhere, that there is no occasion for me
to enter into it at present. I would only remark in passing that the
discovery of a Halitherium allied to the species found in the neigh-
bourhood of Darmstadt is an additional instance to those already
recorded of the existence in our Crag of forms characteristic of the
Miocene fauna of the Rhine valley. It is also worthy of mention
that Mr. Canham’s collection contains three fine specimens of teeth
of Squalodon, a form often found associated elsewhere with Hali-
theriumt. On the other hand, as showing the mixed nature of the
fauna, there are from the same beds some beautifully preserved
teeth of Coryphodon—one in particular, a last upper molar, very

_ like, but rather larger than that figured by Hébert as C. eocenus,

from the earliest Tertiaries of France$.

* “Un Sirénien nouveau du terrain Rupelien,” Bull. Acad. Roy. de Belgique,
2me série, t. xxii, (1871) p. 164.

+ Bull. Acad. Roy. de Belgique, 2me série, t. xvi. (1868) p- 20.

¢ One of these is mentioned by Mr. RayLankester (Quarterly Journal of the

Geological Society, vol. xxvi. (1870) p. 512). The others appear to have been
added since.

§ Ann. Se. Nat. 4me série, t. vi. p. 1.

7

of

t

xt
ay

Quart. Journ .Geol Soc. Vol XXX

1 ;
3 nat, size.

Mantern Bros imp.

¥TUM CANHAMT.

Se

nat , St

le9

nat, sixe

3

Mintern Bros. imp.

a
cn.

G.H.Ford & C.L.Griesbac

HALITHERIUM CANHAMT.

FROM THE RED CRAG OF SUFFOLK. af

DESCRIPTION OF PLATE I.

Fig. 1. Posterior surface of the fragment of skull of Halithertwm from the Crag.
2. Upper surface.
3. Anterior surface.
4, Lateral surface.
All one third the natural size.

5. Under surface of the same, two thirds the natural size.

Letters in all the figures :—F’. frontal bones; Wa. nasals; Mz. maxillary ;
PMzx. premaczillary ;* WE. mesethmoid; HT. ethmoturbinal; PL. palatine;
Zy. zygomatie process of maxillary; P. (in fig. 1) remains of horizontal parti-
tion separating olfactory chamber from posterior narial passages ; a, b, c, d (in
fig. 4), foramina for the passage of nerves and vessels; J 1, 2, 3, 4, 5, and 6
(in fig. 5), sockets for the six maxillary or molar teeth.

DIscussIon.

Mr. Prestwicu thought the specimen a most interesting addition
to the derived fossils of the lower beds of the Red Crag. It was most
likely derived from some of the Miocene beds which formerly existed,
probably on what is now the basin of the German Ocean.

Mr. H. Woopwarp mentioned that in the Woodwardian Collection
there was a skull of Halitherium from the Miocene of Darmstadt. It
was, he thought, of great interest to meet with these Miocene forms
in the Crag, most of the fossils of which appear to have been derived
from the lower beds of the London Clay.

Dr. Lerra Apams mentioned the discovery of a tooth of Halithe-
rium in a calcareous bed in Malta, where also he had discovered one
of the ear-bones.

Mr. Srerey pointed out that the skull presented some peculiari-
ties, which made him doubt whether it could rightly be ascribed to
Halitherium. He thought it might possibly belong to a new genus ;
at the same time he had noticed in the Crag deposits some vertebrae
which he thought might be attributed to Halitherium.

The Cuarrman (Prof. Ramsay) was glad to find that so many
geologists were disposed to regard the majority of the fossil bones
from the Crag as derivative. He had long regarded them as belonging
to a Miocene period, and probably a late one, and to a time when this
country was united to the continent. When at the Crag period a
portion of the surface was submerged, the neighbouring land might,
however, have been still inhabited by the old Miocene fauna.

8 H. WOODWARD ON FORMS

2. New Facts bearing on the Inquiry concerning Forms interme-
diate between Brrps and Reprites. By Henry Woopwarp, Esq.,
F.R.S., F.G.S., British Museum. (Read November 5, 1873.)

Tuere is perhaps no single point along the whole line of defences
raised by the opponents of the theory of Evolution which has been
more warmly contested than the question of the significance at-
taching to the relationship of birds to reptiles. Professor Huxley
truly remarks that “to superficial observation no two groups of
beings can appear to be more entirely dissimilar than reptiles and
birds. Placed side by side, a humming-bird and a tortoise, an
ostrich and a crocodile, offer the strongest contrast, and a stork
seems to have little but animality in common with the snake it
swallows” *.

“In perfect strictness,” writes the same authority t, “no doubt,
it is true that birds are no more modified reptiles than reptiles are
modified birds, the reptilian and ornithic types being both, in reality,
somewhat different superstructures raised upon one and the same
ground-plan ; but it is also true that some reptiles deviate so very
much less from that ground-plan than any bird does, that they
might be taken to represent that which is common to both classes
without any serious error.”

“A lizard is not very far from being the centre of the circle, the
periphery of which is occupied by Chelonia, Ichthyosauia, Plesio-
’ sauria, Pterosauria, and Aves” tf.

If we consider for a moment what are the peculiarities presented
by each of these divisions of the Savrorsrpa (under which great
class Prof. Huxley has proposed to unite these several orders), we
shall begin to perceive that the peculiar and distinctive characters
which we have hitherto assumed to be expressive of the class Aves
are not more remarkable than many which distinguish the divisions
of the Reptilia from one another.

On what grounds, then, are their wide differentiation insisted
upon by the separatists, and their combination advocated by evolu-
tionists ? In the case of those who insist on their separation, it is
urged that.we have no connecting links demonstrable between these
widely different forms of to-day ; whilst on the other handit has already
been shown by Prof. Huxley and others that there are abundant
points of anatomical structure shared even by existing Aves and
Reptilia, and that many of the desiderated forms, exhibiting inter-
mediate points of structure, have already been noticed among the
recent discoveries which have rewarded the labours of the geologist
and paleontologist, not only in Europe, but also in America.

Admitting generally that all existing birds differ definitely from
living reptiles, ‘‘ one comparatively small section, nevertheless, cer-

* Huxley's Lecture, Royal Institution, February 7th, 1868.
t Proc. Zool. Soc. Lond. 1867, p. 415. t Lbid.

INTERMEDIATE BETWEEN BIRDS AND REPTILES. >

tainly does come nearer reptiles than the others. These are the
Ratite, or Struthious birds, comprising the Ostrich, Rea, Emu,
Cassowary, Apteryx, and the but recently extinct birds of New
Zealand, the Dinornithes, which attained gigantic dimensions. Ail
these birds are remarkable for the small size of their wings, the
absence of a crest or keel upon the breast-bone, and of a complete
furcula, and, in many cases, for the late union of the bones of the
pinion, the foot, and the skull. In this last character, in the form of
the sternum, of the shoulder-girdle, and in some peculiarities of the
skull, these birds are more reptilian than the rest; but the total
amount of approximation to the reptilian type is but small, and the
gap between reptiles and birds is but very slightly narrowed by their
existence” *.

All the living, or recently extinct, wingless birds, however, de-
serve to be specially mentioned, not only on account of the more
generalized type of structure which they offer, but because they
present, in their distribution, one of the most interesting geogra-
phical problems to the consideration of naturalists and paleeonto-
logists.

Africa has its Ostrich +; Madagascar had its Mpyornis; Java
and the adjacent islands of the Indian archipelago the Cassowary ;
Australia its Emu and extinct Dromornist; New Zealand its
Dinornis ; South America its Rhea; the London Clay its Dasornis
londiniensis §. Thus, then, we have this most ancient type of wing-
less running birds presenting not merely the greatest range in time,
but also the widest geographical distribution over the globe of any
order .of its class. None of these forms, however, depart in the main
from the avian type, and we must seek still further for modifications
of even greater import than the wingless birds present.

Is there any thing in the Pterosawria which affords a missing
link ?

If we examine that remarkable group the ‘“ flying lizards,” or
Pterodactyles, extending in time from the Lias to the Chalk, we
find they exhibit many adaptive modifications of the avian type, such
as the great air-cavities in the bones, and the prolongation of the
premaxille into beaks (which were probably sheathed in horn),
although the rest of the jaw was armed with teeth. But the manus,
with its four free digits (three armed with claws, and the fourth
enormously prolonged to support the wing-membrane), the non-
avian characters of the pelvis and hind limbs, all depart most widely
from the ornithic type.

But among the various additions made to our knowledge of ex-
tinct forms, some at least may certainly be claimed as affording links

* See Huxley’s Lecture, Geol. Mag. 1868, vol. v. p. 360.

1 Formerly the geographical range of the Ostrich extended into Arabia, Persia,
and part of India, within the Historic period; but it has been exterminated by
the agency of man.

+ Dromornis australis, Owen, Trans. Zool. Soe. 1873, vol. viii. part vi., and
Geol. Mag. 1869, vol. vi. p. 383.

§ See Trans. Zool. Soc. 1872, vol. vii. p. 123, pl. 16. This specimen was de-
tected and its avian characters pointed out by Mr. Davies, of the British Museum.

10 H. WOODWARD ON FORMS

by which to unite more closely the SavropsmpA to one another as a
class. Foremost of these in importance must be mentioned :—

I. The discovery of that remarkable Mesozoic type of bird, the
Archeopteryx macrura (Owen), with its peculiar reptilian-like tail,
composed of twenty free and apparently unanchylosed cylindrical
vertebre, each supporting a pair of quill-feathers, the last fifteen
vertebre having no transverse processes, and tapering gradually to
the extremity.

Two of the digits of the wing (or manus) have curved claws,
much stronger than those of any existing bird; and the metacarpal
bones appear to be quite free and disunited.

In these particulars the Archeopteryx certainly does exhibit a
closer approximation to reptilian structure than any modern bird *.

II. The next remarkable fossil bird to be enumerated is the Ich-
thyornis dispar (Marsh), discovered by Prof. O. C. Marsh, in 1872,
in the Upper Cretaceous beds of Kansas, U.S. It possessed in both
jaws well-developed teeth, which were quite numerous, and im-
planted in distinct sockets. The teeth were small, compressed, and
pointed, and all similar in character. Those in the lower jaw
number about twenty in each ramus, and are all more or less in-
clined backward. ‘The series extends over the entire upper margin
of the dentary bone, the front tooth being very near the extremity.
The maxillary teeth appear to have been equally numerous and
essentially the same as those in the mandible. The jaws were,
apparently, not encased in a horny sheath. The bones of the wings
and legs all conform to the true ornithic type. The vertebre were
all biconcave, the concavities at each end of the centra being
distinct and nearly alike. The tail does not appear to have been
preserved. The bird was about the size of a pigeon. The bones
do not appear to have been pneumatic, although most of them are
hollow. The species, Prof. Marsh considers, was carnivorous and
probably aquatic.

A second form discovered has been named Apatornis celer (Marsh).

Prof. Marsh thinks it probable that Archwopteryw possessed teeth
and biconcave vertebree. In confirmation of the former suggestion,
it may be mentioned that part of a small detached jaw with teeth
was detected by Mr. John Evans, F.R.S., Sec. Geol. Soc., upon the
slab containing the skeleton, and possibly may have belonged to it,
although referred by Prof. Owen to the “ premaxillary bone of a
ish? 4:

In an article published by Mr. Evans in the ‘ Natural History
Review’ for 1865, vol. v. p. 415, he carefully and critically dis-
cusses this question; and he furthermore cites the opinion of the
late illnstrious Hermann von Meyer, that the teeth observable in
the jaw were unlike any heretofore known from the lithographic
stone, although they somewhat resembled those of Acrosaurus. On

* See Prof. Owen on Archeopteryx, Phil. Trans. 1863.

t Phil. Trans. 1863, p. 33. See also article by H. Woodward in ‘ Intellectual
Observer,’ Dec. 1862, vol. ii. p. 313, and plate; and by S. J. Mackie, ‘ Geologist,’
1865, vol. vi. p. 1.

INTERMEDIATE BETWEEN BIRDS AND REPTILES. 11

the whole, however, Von Meyer concludes that there can be little
doubt the jaw really belonged to Archwopterya; and he expressed
his conviction that this singular feathered fossil cannot be looked
upon as a bird with persistent embryonic characters, but rather as a
distinct type, and perfect of its kind.

In this article Mr. Evans also gives the result of his researches in
reference to a certain nodular mass haying a distinctly bilobed out-
line which he had observed upon the slab containing the Archeo-
pteryx. Mr. Evans, having prepared casts of the brain-cavity of a
great number of living birds, was enabled to institute a long series
of careful comparisons, and to demonstrate most satisfactorily the cor-
rectness of his determination, that the mass in question represented
a cast of the anterior portion of the brain-cavity of Archeopteryx.

III. Prof. Owen has added to his ornithological researches another
Eocene bird from the London Clay of Sheppey *, to which his at-
tention had been drawn by Mr. W. Davies, of the British Museum,
who also worked it out with his own hands.

This bird, which he has named Odontopteryx toliapica, is ren-
dered remarkable by the very prominent denticulation of the alveolar
margins of the jaws, to which its generic appellation refers.

The denticulations are intrinsic parts of the bone bearing them,
and are of two sizes, numerous smaller denticles occupying the spaces
between the larger ones, which are about half an inch apart.

When perfect the skull was probably 5 to 6 inches in length; but
the anterior extremity is wanting.

Prof. Owen concludes the Odontopteryx to have belonged to or
near the Anatide, and that it was web-footed and a fish-eating bird,
for which its serrated jaws would admirably adapt it.

As we have not the other parts of the bird, but the skull alone,
it would be presumptuous to surmise as to the presence or absence
of other modifications besides the pseudo-teeth with which the jaws
are armed f.

From the extreme rarity of all terrestrial animal remains pre-
served in a fossil state, it may justly be concluded that many more
such archaic birds, having reptilian modifications, actually existed in
the Mesozoic epoch, although they may possibly never be discovered
by geologists.

Passing from birds to Dinosauria, we are able to enumerate
several forms of reptilia which appear to offer points of structure
tending towards the so-called “ wingless birds,” or those birds
which are devoid of the power of flight by reason of the relatively
small size of their fore limbs and the feathers which they support.

* See Quart. Journ. Geol. Soc. for November 1873, and abstract in Geol.
- Mag. August 1873, p. 376.

+ Many living birds, and notably the Merganser serrator, have a denticulated
mandibular border which, although connected only with the horny covering,
and not with the bones of the mandibles (as in Odontopteryx), yet is sufficient
to prove that the presence of feathers can no longer be looked upon as neces-
sarily implying that the beak with which they were preened must have been
edentulous. (John Evans, op. cit. p. 421.)

12 H. WOODWARD ON FORMS

I. The first of these is a small and very singular reptile from the
Oolite of Solenhofen, which, notwithstanding its minute size, Prof.
Huxley thinks must be placed ‘with the Dinosaurs (the Compsogna-
thus longipes of Andreas Wagner), not much more than 2 feet in
length, having a small head, with toothed jaws, supported on a long
and slender neck. The iliac bones are prolonged in front of and
behind the acetabulum ; the pubes were long and slender. The bones
of the fore limb are very small, and probably furnished with two
clawed digits. The hind limb is very large, and disposed as in birds,
the femur being shorter than the tibia. The proximal division of the
tarsus is anchylosed with the tibia as in birds. The distal ends of
the tarsal bones in the foot are not united with the three long and
slender metatarsals, corresponding with the second, third, and fourth
toes. ‘There is only a rudimentary metatarsal to the fifth toe.

It is impossible to look at the structure of this strange reptile
and to doubt that it hopped or walked in an erect or semierect
position, after the manner of a bird, to which its long neck, slight
head, and small anterior limbs must have given it an extraordinary
resemblance *.

II. Much of the recently acquired knowledge of the huge Dino-
sauria of our own Secondary rocks has resulted from the earnest
labours of Professor Phillips, bestowed upon the remarkable series
of remains in the Oxford Museum.

We now know certainly that the Megalosaurus, that huge carni-
vorous lizard, perhaps 30 feet long, which ranged from the Lias to
the Wealden, had strong but not massive hind limbs, and short
reduced fore limbs, five anchylosed sacral vertebrae (Owen), the
ilium, ischium, and pubis slender and bird-lke, as in the Ostrich,
the scapula and coracoid resembling those of the Apteryx. From
all these considerations Professor Phillips agrees with Profs. Owen
and Huxley in viewing the Megalosaurus, “not as a ground-
crawler, like the alligator, but moving with free steps, chiefly, if
not solely, on the hind limbs, and claiming a curious analogy, if
not some degree of affinity, with the Ostrich” +.

What we have cited as regards the carnivorous Meqgalosaurus is
true also of the vegetable-eating lizards of the Mesozoic rocks
(Iguanodon, Scelidosarus, &c.).

The sacrum is composed of from four to six vertebrae. The pelvic
bones are bird-like in form and disposition; there is a strong crest
which passes between the head of the fibula and the tibia, as in
birds. The tibia has a great anterior or “ procnemial” crest, not
seen in other reptiles, but existing in most birds, especially the
running and swimming birds. The toes are reduced in number;
Scelidosaurus has four toes and a rudiment of a fifth ; Zguanodon
has three, with a rudimentary indication of a fourth.

There is evidence in ‘‘the manner in which the three principal
metatarsals articulate together, that they were very intimately and

* Prof. Huxley’s Lecture, Joc. cit.
+ ‘Geology of Oxford and the Valley of the Thames,’ by John Phillips, M.A.,
E.R.S., F.G.8., &., 1871, p. 196.

INTERMEDIATE BELWEEN BIRDS AND REPTILES. 13

firmly united, and that a sufficient base for the support of the body
was thus afforded by the spreading out of the phalangeal regions of
the toes.” (Huxley, loc. cit.)

Mantell long since, and more recently Leidy, have concluded, from
the great difference in the size between the fore and hind limbs,
that Iguanodon and Hadrosaurus, as well as other Dinosauria, may
have supported themselves for a time at least upon their hind legs.
But Mr. Beckle’s discovery of pairs of large three-toed foot-prints,
of such a size and at such a distance apart that it is difficult to
believe they have been made by any thing but Jguanodon, leads to
the supposition that this vast reptile, and perhaps others of its
family, must have walked temporarily or permanently upon its hind

, loc. cit. p. 364).

Many years since, Mr. Allan Cunningham (H. M. Botanical Col-
lector for Kew), who accompanied the Expedition to survey the
Intertropical and Western Coasts of Australia, from 1818 to 1822,
under Captain Philip King, R.N., F.R.S., secured a specimen of a
remarkable Frilled Lizard, which had perched itself upon the stem
of a small decayed tree, at Careening Bay, Port Nelson*. This has
been named Chlamydosaurus Kingii, by Dr. Gray. The stufted
specimen is preserved in the British-Museum collection in a semi-
erect position, its fore feet (which are very much smaller than
the hind feet) scarcely touching the ground at the extremities of
the claws. I had the advantage (in company with my esteemed
colleague, Dr. Giinther, the Assistant Keeper of the zoological
collections of the British Museum) to hear the remarks of an
Australian resident on this lizard, which is common about the
gardens in the environs of Sydney. This observer reports that the
lizard in question not merely sits up occasionally, but habitually
runs upon the ground on its hind legs, its fore paws not touching
the earth. This statement interested me immensely; and on re-
peating it to Prof. Huxley, I learnt that that acute observer,
Mr. Gerard Krefft, of the Sydney Museum, had noted the same
peculiarity. With this upright carriage, special modifications of
the sacrum and pelvic bones are necessary; and, no doubt, when a
specimen is dissected, this very interesting point will be corrobo-
rated by the form and articulation of the bones. As geologists, we
cannot but be interested in this peculiarly modified existing type
of lizard, occurring, as it does, on the continent of Australia (which
has yielded such a remarkable assemblage of Tertiary and existing
Marsupialia to be critically examined and chronicled by our great
anatomist, Professor Owen)—a land also remarkable for the pos-
session of many living Mollusca of Mesozoic types on its coasts.

It is noteworthy in connexion with these observations on bipedal
reptilia (existing and extinct), to find that the Solenhofen Lime-
stone, which has yielded both the long- and short-tailed Péterosauria,
the long-tailed Reptilian-like birdy, and the long-necked short-armed
Avian lizardt supplies a bipedal track upon one of its slabs, which

* See King’s ‘Survey of Australia,’ 8vo, 1827, vol. ii., Appendix, p. 424.
+ Archeopteryx. { Compsognathus.

14

Track of Saurian from the Solenhofen Limestone (Ichnites lithographicus, Oppel), one fourth natural size.

H. WOODWARD ON FORMS

reminded me at once of what Chla-
mydosaurus or Compsognathus might
produce under favourable condi-
tions. The slab presents a median
track formed by the tail drawn along
on the ground; the. two hind feet
with outspread toes leave their mark,
whilst the fore paws just touch the

- ground, leaving a dot-like impression

on either side of the median line
(see figure). Dr. Oppel has named
this track I[chnites lithographicus*.

In calling attention to the slab,
Dr. Oppel remarks that the foot-
steps are ranged in two parallel lines
with a middle continuous impres-
sion alternately stronger and fainter.
The footsteps go in pairs. It re-
sembles no other form of track
already noticed or described from
this or any other formation.

Dr. Oppel thinks Archeopteryx
might have formed such a track,
although he is unable to explain the
alternate deepening of the middle
track.

Remembering that the tail of Ar-
cheopteryx is bordered all the way
by feathers, it will at once be seen
that it could not leave behind a clear
and simple furrow, but a broad
smudge composed of many lines, like
that left by a flat colour-brush drawn
along upon paper.

The tail of a lizard progressing by
hops and supporting itself upon its
hind limbs and tail, would, however,
produce just such impressions, the
deepening of the furrow made by
the tail being caused by the depres-
sion of the tail in making each on-
ward hop.

Viewed by the additional lght
which our present knowledge of the
structure of the Mesozoic Dinosauria
and of the existing Chlamydosaurus

* Palaontologische Mittheilungen aus
dem Museum des Konigl. Bayer. Staates
von Dr. Albert ear Stuttgardt, 1862,
tab. 39, p. 121.

id

INTERMEDIATE BETWEEN BIRDS AND REPTILES. ° 15

affords, we need no longer be doubtful as to the origin of the many
bipedal tracks which occur in the Trias and upwards.

Some are very probably the “spoor” of Struthious birds which
may have existed fully as far back as the beginning of the Secondary
Period; but most are, no doubt, due to the bipedal habit of our
Secondary reptiles, a peculiarity still maintained by the Australian
Chlamydosaurus.

Discussion.

Mr. Sertry thought that the footprints on the slab cited in the
paper had been produced by some saurian, such as a Pterodactyle,
the fore limbs of which were wider apart than its hind limbs, rather
than by Compsognathus. If the foot-track had been due to a saurian
walking on its hind legs only, he thought that the principal im-
pressions must of necessity have been nearer together. He disputed
the correctness of the term “ adaptive modification” as applied to
the air-cavities in bones. He was inclined to regard the Pterodactyle
as more closely allied to birds than did the author of the paper.
The condition of the carpus, as well as the tarsus, in these reptiles
showed their ornithie affinities. He cited jerboas, kangaroos, and
other forms, in which the hind legs were mainly used for pro-
gression, but in which the sacrum and other bones were not modified,
as instances calculated to inspire caution in connecting the mode of
progression with structure.

Mr. Hvrxe could not regard the tracks as those of a Pterodactyle,
as the inner marks were much less distinct than the outer, and would
therefore hardly be due to the hinder limbs, on which the weight
would mainly fall.

Mr. Branrorp agreed with Mr. Seeley that the mere fact of the
Chlamydosaurus walking on its hind legs did not suffice to prove any
affinity with Dinosaurians.

Mr. Woopwarp, in reply, stated that the two points on which
he had mainly founded the paper were :—Ist, the occurrence of
footprints in the Solenhofen limestone, characteristic of a bipedal
progression of some saurian, which had, moreover, used its tail from
time to time to give it a forward impetus ; and 2ndly, the method
of walking of Chlamydosaurus. With regard to animals thus pro-
gressing, he was not prepared to accept the view that there was no
corresponding modification in structure.

16! J. W. HULKE ON A LARGE SAURIAN LIMB-BONE

3. Note on a very Lares Savetan Lims-sone adapted for Proerus-
ston upon Lanp, from the Kimmuriper Cray of Weymourn, Dorset.
By J. W. Hurks, Hsq., F.G.S., F.R.S. (Read November 5, 1873.)

[Puare II. |

Ar rare intervals there have been obtained from the Kimmeridge
Clay in several localities (¢nland and also on the Dorset coast) re-
mains of very large reptiles differing from the contemporary Enalio-
saurs (the Plio-, Plesio-, and Ichthyosaurs) by the adaptation of
their limbs to walking upon dry land. On June 23, 1869, I brought
before the Society a large humerus of such a reptile, which had been
obtained in Kimmeridge Bay by J. C. Mansel Pleydell, Esq., a
Fellow of our Society. It was afterwards presented by him to the
British Museum*; and to the Saurian indicated by it I gave the
name of Jschyrosaurus.

The subject of this note is a much larger limb-bone lately
found in the Kimmeridge-clay beds, near Weymouth, by Mr. R. I.
Smith, and intended to be added to the national collection. It was
enveloped in large septarian masses, which stuck so closely to it that
thin lamine of the surface of the bone were unavoidably detached
in stripping the matrix from it. The natural surface, where unin-
jured, is smooth; it has a close, fine grain resembling that of the
humerus of the /schyrosaurus, and quite unlike the coarse texture of
many Enaliosaurian bones. The bone has been much fissured, and
cemented together by spar; and some parts have been distorted by
squeezing ; but the general figure is well preserved.

It has a closer resemblance to the Crocodilian type of humerus
than to any other bone; and I am disposed to regard it as a humerus,
the left one. Its present length is 54 inches ; but the articular sur-
faces of both ends are wanting, and for these scarcely less than 9 addi-
tional inches can be allowed ; so that the whole length of the perfect
bone can hardly have been less than 63 inches. The middle of the
shaft is cylindroid; its girth is 21 inches; and its horizontal transverse
diameter is 7°8. Its transverse section (PI. IT. fig. 4) is a subtrigonal
figure ; and it exhibits a large coarsely cancellated core enclosed in a
stout compact cortical ring. Towards the proximal end the width
increases, chiefly by the backward sweep of the posterior border,
to a present maximum of 17 inches; but this would be increased
by the absent posterior moiety of the proximal surface, including the
articular caput and the adjoining end of the posterior border. The
outline of the dorsal or upper surface in this situation is transversely
convex, whilst longitudinally it rises in a gentle curve from the
cylindroid shaft to the proximal end, owing to the increased thick-

* A description and figures of it will be found in the Quart. Journ. Geol. Soc.
vol. xxy. p. 386.

t In the following description the bone is imagined to be placed nearly hori-
zontally, with its long axis perpendicular to the animal’s trunk.

ARUS HUMERO-CRISTATUS

Aral size

M&N Hanhart imp,
G.L Griesbach -

OSAURUS HUMERO-CRISTATUS
HUMERUS OF (eee vee

naiee

AN

ts

~

} =

ae
iis
Savi ue

Tc,

FROM THE KIMMERIDGE CLAY OF WEYMOUTH. 17

ness required here for the support of the terminal articular caput.
The distal moiety of the dorsal surface beyond the shaft expands
gradually to a maximum width of 16 inches at the distal articular
surface. A very long, wide, and rather deep intercondyloid groove
traverses it longitudinally. The ventral surface of the expanded
proximal moiety is yery hollow transversely ; and beyond the shaft
its distal moiety in the same direction is gently convex. The distal
articular surface (fig. 3) is an oblong, the long diameter of which
measures 16 inches, and the short one averages 6 inches. It is
divided into a pair of condyles by a very shallow vertical groove,
which above joins the dorsal intercondylar groove and below ends
between two low eminences at the ventral surface. The posterior
border, gradually contracting from the shaft towards the proximal
end, becomes here a relatively thin rounded edge. In the distal
moiety this border is stouter. The anterior border in its proximal half
is much wider than the corresponding part of the posterior border ;
it is flattened and produced downwards as a ventrally projecting crest
(fig. 2, dc.) which greatly increases the hollowness of the ventral
surface in this part. The distal moiety of this border, in its whole
length, has the form of a thin, rough, very prominent crest project-
ing forwards. These crests form one of the most striking features
of the humerus, which distinguish it immediately from the almost
equally large Mantellian Pelorosaurian humerus preserved in the
British Museum*, and from the almost equally huge, but rather
stouter, humeri of the Ceteosaurus oxoniensis in the Oxford Univer-
sity Museum, so admirably restored by Prof. Phillips+, as also from
the much smaller humerus of Jschyrosawrus to which I have already
referred.

A general correspondence with the humerus of Ceteosaurus oxoni-
ensis inclines me to provisionally refer this new Kimmeridge Saurian
to the genus Ceteosaurus as typified in C. ovoniensis. Its rough strong
crests suggest the specific designation humero-cristatus (Ceteosaurus
humero-cristatus).

EXPLANATION OF PLATE IL.

Fig. 1. Dorsal or posterior surface of humerus: d. distal end ; px. proximal
end; ant. anterior border; post. posterior border ; cr. crest.
2. Ventral or anterior surface of humerus: pa. proximal end; d. distal
end; ant. anterior or outer border; post. posterior or inner border ;
dc. deltoid crest.. *
3. View of distal articular end.
4. Transverse section near middle of shaft.

Discussion.

Mr. Sretzy remarked that the internal structure of the bone re-
sembled that found in Grgantosaurus, and the general form of the
humerus was such as might be expected did it belong to an animal
of that genus.

* British Fossil Reptilia of the Wealden Formation, Supplement ii. vol. xii.
p. 39.

+ Geology of Oxford, p. 272.
OO. Gass No. LL C

18 J. W. HULKE ON THE ANATOMY OF HYPSILOPHODON FOXII.

4. Scupprementat Nore on the Anatomy of HypsttopHopon Foxtr.
By J. W. Hors, Esq., F.R.S., F.G.S. (Read November 19,

1873.)
[Puate IIT.]

Ar the close of last Session I read a note upon some remains of an
immature Hypsilophodon Fou which I had shortly before obtained
in Brixton Bay, Isle of Wight, from the west end of the well-known
bed which crops out at the top of the cliff at Barnes Chine and dips
under the beach at Cowleaze Chine. ‘Their chief value consisted in
the additional light they threw upon its dentition, and the informa-
tion they afforded of the form and the proportions of the limbs. In
September I was so fortunate as to obtain in the same locality parts
of two individuals (one probably fully grown) which, as they illus-
trate some structures better than any other remains of this Dinosaur
yet before the Society, have appeared to me worthy of being made
the subject of a supplementary note.

The bones are imbedded in a block of sandy elay-stone which had
fallen from the cliff and had been washed to and fro by the sea until
some of them had become much abraded. The most important are
a skull and two chains of vertebre, each including a considerable
part of the sacrum.

Skull.—This is larger than that found by Mr. Fox, which Prof.
Huxley exhibited here in November 1869. Its upper surface was ex-
posed; and I have laid bare its right side (PI. III. fig.1). The maxillary
apparatus is broken off from the cranium proper, and twisted round so
that the dentigerous border of the maxille and the palate now look
upwards, the pterygoids resting in the lower part of the right orbit.

The upper surface of the skull is a long rhomboid (I refer now to
the part behind the front of the orbits), of which the short diameter
connects the stout postorbital processes; and the sides are lines
drawn from these to the front of the supraorbital arch and to the
extremity of a salient occipital (Pa’.) spine in which the parietal region
terminates behind instead of presenting here the entering angle
usual in lizards’ skulls. Large pieces of the parietal and of the
frontal bones have exfoliated, laying bare the matrix moulded to
the inner surface of the vault. Between the temples this pre-
sents a ridge suggestive of a parietal crest; and between the
orbits is a mesial furrow indicative of the division of the principal
frontal bone. The root of the right. parietal suspensory process
only is preserved (Sp.); its direction is nearly vertical to that of the
parietal crest.

The orbit is very capacious ; ‘4 inch below its upper border lie six
of the thin bony scales of the sclerotic coat of the eyeball (S.).

Thepreemaxille (Prmx.)want the edentulous anterior extremity seen
in Mr. Fox’s specimen; but other parts of their structure are better
displayed here, owing to their partial separation from the maxille.

J. W. HULKE ON THE ANATOMY OF HYPSILOPHODON FOXII. 19

The body of the prazemaxilla is a vertical plate, -45 inch deep from its
nasal to its dentigerous border, smooth, except quite in front, where
its surface is wrinkled. From each end rises a strong process. That
in front is a compressed trihedral blade narrowing upwards, shorter
than the posterior or outer process. Applied to its fellow of the
other side it forms the lower part of the septum between the anterior
nares. Its front edge seen on the surface of the snout is stout; the
posterior edge is thin.

The posterior or outer process, broader and longer, is closely ap-
plied to the anterior border of the maxilla, but not suturally united
with it. It overlaps the maxilla, which has a shallow groove for its
reception. The dentigerous border, nearly straight, is ‘65 inch long ;
and in this space it contains, I think in separate sockets, five mature
eylindrical teeth, of which the roots, with only small portions of the
crowns, now remain. At their inner side, between the second and
third and the fourth and fifth teeth, two immature crowns are just
visible. A large triangular palatal process, mesially united to its
fellow, completely roofs this part of the mouth. From the anterior
palatine foramen to the posterior extremity of the interpreemaxillary
suture measures ‘7 inch. This sutural margin is longer than the
free posterior border, and it forms a projecting angle to which, on
the right side, the front of a vomer is (V.) attached.

The teeth all lie behind the anterior palatine foramen; the small
portion of the edge of the right jaw in front of this is smooth and
toothless.

The maxille(Mx. Ma'.) are large subtriangular bones. The left is
very perfect. Its straight dentigerous border, 1-6 inch long, contains
an unbroken series of eleven* compressed sculptured teeth, of which
the front four are smaller than the others. The hinder margin of
the crown of each tooth slightly overlaps the front margin of that
next behind it. The crowns are obliquely worn, the thickly enamelled
outer contour being the longer. The number of premaxillary teeth
agrees with that of Mr. Fox’s skull; the maxillary teeth are one
more in myskull. The teeth themselves agree so closely with those
described in my last note as to make any further account of them
unnecessary. In front of its dentigerous part the lower border of
the maxilla and its upper border converge and send forward upon
the deep surface of the preemaxilla the thin grooved plate mentioned
as receiving the posterior ascending process of the latter.

Above this plate the anterior border of the maxilla rises in a
sinuous curve to a height of 1:1 inch above the second tooth, making
here a blunt angle with its upper border, which behind this declines
in a gentle hollow curve to a height of -45 inch above the last tooth.
Above this tooth, at the height of -35 inch, the surface of the maxilla
is angulated, and a strong triangular process, at least -6 inch long,
passes backwards. The uncertainty whether a narrow line obliquely
crossing the junction of this process and the body of the maxilla is
an accidental crack or a suture leaves it doubtful whether this process
is part of the maxilla or a separate bone.

* Perhaps one tooth is missing between the second and the third.
C2

20 J. W. HULKE ON THE ANATOMY OF HYPSILOPHODON FOXII.

In the body of the maxilla above the third to sixth tooth is a large
subtriangular gap ; it is the aperture between the orbit and external
nostril seen in Mr. Foxs skull. Below this, and extending nearly
the whole length of the bone, the outer surface of the maxilla is
pierced by a chain cf conspicuous foramina, such as are seen in the
maxille of Megalosaurus and Teratosaurus.

The divergence of the maxille posteriorly partially exposes the
palatal apparatus, the hinder part of which lies in the right orbit.
The pterygoids (Pt, Pt’), not mesially joined, but separated by a fissure,
have a remarkably stout body, the posterior border of which bears a
very large basisphenoidal process, anteriorly limited by a prominent
ridge produced downwards, and terminating angularly at the mesial
border. Theleft pterygoid (P?’) retains the root of a strong quadratic
process directed outwards and backwards, in front of which the hollow
outer border runs out in an ectopterygoid. In front of the ptery-
goids the palatals (PJ, Pl’) are partially visible, their inner borders
also separated by a fissure. The left palatal, which is best seen, is a
flat rod 35 inch wide, with (so far as it is exposed) parallel margins.
Its buccal surface is longitudinally grooved.

It is almost superfluous to remark that the skull of Hypsilophodon,
as was, indeed, shown by Mr. Fox’s specimen, is constructed after
the lacertilian and not after the crocodilian pattern. In this respect,
so far as the material allows of the comparison being made, it agrees
with the large skull from Brooke which I brought under the notice
of the Society two years ago, and provisionally referred to Jguanodon
Mantellr.

Spinal Column (fig. 2.)—Crossing the block from right to left, at a
little distance from the skull, is a continuous chain of eight consecutive
vertebrae. The ventral surface of the centra is uppermost. The
first three from the right are too much mutilated for description ;
the fourth is much abraded, the fifth less so; but the sixth, seventh,
and eighth are sufficiently preserved to exhibit all their essential
characters. These three last centra are inseparably anchylosed,
every trace of their primitive separateness (which is still evident
between each of the central (1s, 2s) to their right) has quite disap-
peared. The seventh and eighth centra are further distinguished by
the confluence of the expanded distal ends of their transverse pro-
cesses. These two marks—confluence of the vertebral centres of
the outer ends of the transverse processes—make it certain that
the seventh and eighth vertebre are part of the sacrum. The
sixth vertebra (Z) has distinct transverse processes which stand out
from the neural arch in the form of flattened, tapering blades,
‘4 inch long. Confluent with that border of the transverse processes
furthest from the sacrum, at their union with the neural arch, is a
pair of articular processes, the articulating surfaces of which have an
upward and inward aspect; this aspect and their position prove
them to be prezygapophyses (Prz.). A vertebra whose centrum has
coalesced with that of a next sacral, which yet has its own separate
transverse processes, and also whose articular processes furthest
from the sacrum bear the characters of prezygapophyses, must

J. W. HULKE ON THE ANATOMY OF HYPSILOPHODON FOXITI. mall

precede the sacrum and cannot follow it; the sixth vertebra must
in fine be the last lumbar.

The length of this centrum is rather less than ‘9 inch, the same
as that of the fifth and fourth centra. Its form is cylindric, its
contour transversely convex, and longitudinally hollow, the middle
slightly contracted and the ends swollen, particularly that which is
anchylosed to the first sacral. Its transverse diameter at its middle
is ‘6 inch, at its front end -8, and at its posterior end somewhat
more. The transverse processes of the second, third, and fourth
lumbar vertebre have slender ribs anchylosed to their extremities,
a distinct knot marks the union of dia- and pleurapophysis. They
differ in this respect from the corresponding vertebre in the Alligator
(A. luctus) and other existing crocodilians, in which the traces of
the primitive separateness of the transverse process and rib disappear
with the maturity of the individual.

The determination of the first lumbar carries with it that of the
next succeeding vertebra, it is the first sacral (1 s.); we have then
the first, and not the posterior moiety of the sacrum. The first and
second sacral centra are much smaller than the last lumbar, a
similar difference of bulk obtains in the sacrum assigned to Jguwanodon
Mantelli ; but this difference does not extend to their figure, which
has a general resemblance to that of the lumbar vertebre. It too
is cylindroid, constricted at the middle and expanded at its end,
which gives the lower contour of the chain a sinuous outline, hollow
at the middle of the centra and convex at their coalesced extremities.
The swelling which marks the junction of the coalesced centra is
not a uniformly tumid nodal ring; but it is greatest at the union of
the sides and inferior surface, forming here a pair of small elevations
similar to those in the reputed sacra of Zguanodon Mantelli and
Hyleosaurus. The transverse process of the first sacral vertebra
springs from the junction of this vertebra with the last lumbar,
standing out from here vertically to the axis of the sacrum. It is
remarkably stout, the antero-posterior diameter of its root is °6
inch; its anterior contour merges into the lateral contour of the last
lumbar centrum, greatly increasing the apparent bulk of this. At ‘5
inch distance from its origin, it bends backwards nearly at a right angle
to its first direction, and joins the dilated outer end of the second
transverse process springing from the union of the second and first
centra, and it includes with this a large subcircular loop. A third
transverse process in like manner abuts on the junction of the third
and second sacral centres, and from two of the loops with the second
and fourth transverse processes, making in all three of those loops
or nerve-foramina ; but the third and fourth centra are missing, the
third having been broken off just behind its union with the second,
Against the strong buttress formed by the confluent dilated ends of
the transverse processes on the right side lies a fragment of the
right ileum (J/.).

Below this chain of vertebre lies a second chain of seven
smaller vertebre with part of a sacrum including four centra. It
appeared so unlikely that this should be part of the spinal column

22 J. W. HULKE ON THE ANATOMY OF HYPSILOPHODON FOXII.

of a second individual, and so probable that it might be the posterior
moiety of the near-lying larger sacrum with part of the tail, that at
first I rather hastily imagined it to be such; but unwilling to leave
it doubtful, I laid bare the articular and transverse processes of the
two vertebre next the sacrum, which proved them to be lumbar.
After this I could not resist the conviction that I had investigated
in the same block of stone the remains of two distinct individuals;
the smaller sacrum repeats all the essential features of the larger one.
The third centrum (3 s.), missing in that, is here well preserved, as
is also the second ; but the first and the last lumbar centra are badly
mutilated. The third lumbar centrum is better preserved than any
other; its lateral surface is less convex and more plane vertically
than the corresponding part of the first lumbar centrum of the larger
individual.

From beneath the right side. of the sacrum, partly hidden by a
fragment of a pelvic bone, the proximal half of the right femur
projects (F¢.). Its inner trochanter is well preserved, wanting only
the thin triangular lower angle. At its inner side is a very distinct
shallow pit. Near the skull and beneath the larger chain of ver-
tebree, I found several very thin bony plates having one surface
granular, the other smooth and furrowed by a vascular net. Their
shape was irregularly polygonal; and their size varied much, some
attaining an area of about 1 square inch. I regard them as thin
scutes (fig. 1, sc.).

Prof. Owen has taken exception (Quart. Journ. Geol. Soe. vol. xxix.
p- 531) to the generic distinctness of Hypsilophodon, and maintained
its identity with the genus Jguanodon, basing his argument mainly
‘on the similarity of their compressed, ridged teeth, on the peculiar
mode in which these wear down, and on the spout-like form of the
edentulous anterior extremity of the mandible in both. Fully re-
cognizing these points of structural agreement as evidence of a very
close affinity, it appears to me that there remain so many and so
great differences as to fully justify the adoption of the separate genus
Hypsilophodon. As I stated fully in my first note what appeared
to me the chief structural differences, it is unnecessary to recapitulate
them here; they were chiefly those presented by the limbs, and had
respect to their form and proportions, and to the number of toes.
In his paper of November 10, 1869, Prof. Huxley noticed certain
vertebral differences ; but; his comparison did not extend to the
sacra, this segment of the spinal column being hidden in the Mantell-
Bowerbank fossil, the subject of the paper. I have therefore taken
the opportunity which my recent acquisitions afford, to compare
my Hypsilophodon sacra with the type specimen of the [guanodon-
Mantelli sacrum figured in the ‘ Fossil Reptilia of the Wealden
Formation’*. The result is that I find the form of the vertebral
centrum quite different, being cylindroid, rounded below in Hypsi-
lophodon, laterally compressed, so much as to be angulated or almost
keeled below, in Jguanodon ; this difference seems to me of higher
than specific value.

* Fossil Reptilia of the Wealden Formation, order ‘“‘ Dinosauria,’” p. 11.

GH.Ford & C.L..Griesbach.

HYPSILOPEM

Quart. Journ. Geol. Soc.Vol. AAA. FILL

Mintern Bros. mp

ION. FOXIL.

at

; a eal
. ad f 4
< i

Quart. Journ.Geol. Soc Vol. XXX. PLII -

GHFord & C.L.Griesbach. Mintern Bros. imp .

HYPSILOPHODON FONII.

159
eee i

e . ae

J. W. HULKE ON THE ANATOMY OF HYPSILOPHODON FOXII. 23

EXPLANATION OF PLATE ITI.
Remains of Hypsilophodon Foxit.

Fig. 1. Skull: Pa. parietal bone; Pa’. its supraoccipital spinous process ;
Fr. frontal bone; Po. postorbital process; Pro. preorbital process ;
Sp. suspensorial process; S. bony plates of sclerotic coat of eyeball
lying beneath orbital arch; Mz. right maxilla; Ma’. left maxilla;
Prinz. premaxille ; V. yomer ; Pi. right palatal bone ; Pl’. left palatal
bone; 7. right pterygoid ; Pz’. left pterygoid; Sc. scutes.
2. L. last lumbar vertebra; 1s, 2s, 3s, first, second, third sacral vertebra ;
Prz, prezygapophysis ; Psz. postazygapophysis; J/. ileum; Fe, femur.

Discussion.

Mr. Bory Dawxrrs thought there was as much distinction between
Hypsilophodon and Iquanodon as between Hipparion and Hquus, and
that this was quite sufficient to be regarded as generic rather than
specific. He was not satisfied as to the additional bone in the foot
in Mr. Beccles’s specimen, but thought it might belong to some other
part of the animal. He considered that all the teeth of Jgwanodon
were always ground flat by wear.

Mr. Szrtzy considered that the author was likely to substantiate
his opinions. He pointed out certain differences in the structure and
form of the maxillary and other bones of the skull in Hypsilophodon
and [guanodon, and especially in the maxillary. He attached great
importance to the thickening of the enamel at the base of the teeth
of Hypsilophodon, which approximated to that which was found in
some mammals. The teeth commonly reputed to be those of Jgua-
nodon might, he thought, belong to different species, if not genera,
and showed some divergence in character. The observations on the
palatal bones of Hypsilophodon were, he thought, calculated to throw
great light on the anatomy of Dinosaurs,

24 J. W. HULKE ON THE ASTRAGALUS OF IGUANODON.

5. Note on an Astracatus of Ievanopon Mantetu. By J. W.
Hoxxg, Esq., F.R.S., F.G.8. (Read November 5, 1873.)

I am indebted to the courtesy of one of our Fellows, E.P. Wilkins, Esq.,
of Newport, Isle of Wight, for the opportunity of exhibiting this re-
markable astragalus of Mantell’s Jguanodon, a bonenot contained in the
rich series of remains of this Dinosaur preserved in the British Museum,
and, so far as I can ascertain, hitherto unknown. Mr. Wilkins ac-
quired it several years ago with other reptilian fossils obtained from
the cliffs in Brixton Bay; and last September, when I paid a hasty
visit to his collection, he showed it to me as a bone which had much
puzzled him, and which he had been unable to determine. On learn-
ing its extreme interest, he anticipated my wish for a sketch of it by
suggesting that I might take it with me to London and bring it under
the notice of the Geological Society.

It has a depressed oblong figure, measuring in its long axis between
the extreme limits 94 inches, and in its short axis 53 inches. The
under surface has the regular pulley-shape characteristic of a movable
hinge-joint (fig.1). A well-marked median constriction separates the
lateral portions, the inner of which is rather more swollen.than the
outer. The upper surface (fig. 2) has an irregular unsymmetrical
shape, adapted to that of the distal end of the tibia of Mantell’s
Iguanodon. ‘The applied surfaces of astragalus and tibia must have
so interlocked as to have prevented all motion between them. A
. strong ridge passing between the anterior and posterior margins divides
the upper surface into a larger antero-internal and a lesser postero-
external moiety. The former of these is a wide trough strongly con-
cave from back to front. The outer moiety slants steeply down from
the dividing ridge, and it descends much lower than the general level of
the inner half of the same surface, than which it is also much narrower.
The inner half looks upwards, backwards, and inwards ; and the outer
half looks upwards, forwards, and outwards. The downward slope of
the outer moiety of this surface renders the outer border of the bone
very thin; it is indented at its middle by a notch ; and in its pre-
sent condition it does not show any mark of an articulation with an
os calcis. The inner border of the bone is so stout that it might be
properly termed a surface; it has a vertical depth of about 23
inches. The posterior border, less stout, rises quickly from its outer
end, where it includes a right angle with the outer border to the
ridge which divides the upper surface; and from here it curves
gently downwards, rising again at its inner end (fig. 3). The anterior
border, thinner than the posterior, rises from both ends in the form of
a blunt lip, which belongs more to the inner than to the outer moiety
of the bone (fig. 4), and was received into the entering angle and
groove present in the anterior surface of the distal end of the tibia.
The base of this lip in the present specimen has been fractured and
pressed backwards so as to make the outer edge of the lip now appear

J. W. HULKE ON THE ASTRAGALUS OF IGUANODON.

yper View.

Fig. 2.—Up

oe

Fig. 1.—Under view of Astragalus of Iguanodon.

aS

=

[e)

Sy

&,

3

S ad

SS aa

e .

S S
*~

o

= r

a 3

| a

=

faa) (eo)

ah Sy
on

e.

i

?, Inner 3

26 J. W. HULKE ON THE ASTRAGALUS OF IGUANODON.

continuous with the end of the ridge that divides the upper surface ;
but when undistorted, the ridge ended, not at the outer edge, but on
the posterior surface of the lip. This lip clearly represents the
ascending process of the avian astragalus.

To those who were present in June 1870, when Professor Huxley
made his valuable communication “‘On the Affinities between the
Dinosaurian Reptiles and Birds,” any comments upon the bone will
be superfluous; but for others who are not acquainted with that
paper, I may state that, besides its value as a distinct addition to
the anatomy of Zguanodon, it has an additional interest as constitu-
ting another link in the chain which joins these two classes. In all
extant reptiles, and, excepting the Dinosauria, in all fossil ones which
have a foot capable of being flexed and extended on the leg, this
hinge-movement occurs between the leg-bones and the astragalus; but
in the Dinosauria the tibia and the astragalus were tightly inter-
locked, and the movable ankle-joint was between the astragalus
and the next distal segment of the tarsus, which is its position in
birds. In most of these, however, the separateness of the astragalus
and the tibia disappears at an early age; only in the Ratite does it
continue to maturity. In the New-Zealand Dinornithide and their
living successors, the Apterygide, I have seen it present in fully
grown skeletons. In the Dinosauria the distinctness of these bones,
transient in birds, persisted throughout the whole life of the indi-
vidual.

READE—DRIFT-BEDS OF THE NORTH-WEST OF ENGLAND. 27

6. The Drirt-Beps of the Nortu-West or Enerann.— Part I. SHEL
of the LancasHIRE and CHESHIRE Low-LevEL BovLpER-cLay and
Sanps. By T. Metziarp Reape, Esq., C.E., F.G.S. (Read
November 19, 1873.)

ConrTENTS.

1. Introduction and Explanatory Section.
2. Description of Localities.
3. Mode of Occurrence and Condition of the Shells, constitution of the Beds,

and general inferences therefrom.
4, Analysis of the list of Shells and comparison with the present Molluscous

Fauna of the British seas.
5. Position of the Low-level Boulder-clays and sands in the Glacial series.

1. Lyrropuctrion.

BeroreE treating of the stratigraphical distribution of the Low-level
Boulder-clay and sands of Lancashire and Cheshire in which the
shell-fragments occur, I have thought it better to discuss first, by
the light of the fragments themselves, the geographical distribution
of the species represented, their mode of occurrence and condition,
and the nature of the matrix in which they are found—some of the
problems involved in an accurate interpretation of the glacial-marine
phenomena of the drift of the north-west of England.

To enable you to understand the character of the drift from which
the shells were taken, I exhibit-a section of the Bootle Lane Station
cutting (fig. 1), in which most of the beds seen about Liverpool are
typically represented. The majority of the shells have been derived
from clay answering more nearly to No. 7 than to the others. No
distinction, however, can be made on paleontological grounds between
any of the various beds, which I have reason to think are only local
developments of one system. The beds are as follows, in ascending
order :—

1. Pebble-beds of the Trias (Bunter).

2. Shattered rock.

3. Red Sand, with tightly compacted rubble and débris of the
Trias. (Ground-moraine equivalent of the Scotch till.)

4. Lowest bed of Boulder-clay (unstratified); largely composed of
the red sand, containing rounded and subangular pebbles and boulders,
mostly striated, of mountain-limestone, granite, syenite, Silurian
grits, traps, and decomposed greenstone, &c. Also shell-fragments,
as shown on the list and marked (Lower Clay), apparently quietly
deposited on the red sand. A few boulders rest on the red sand.

5. Stratified sand and shell-fragments. This bed splits in two and
is separated by the Lower Clay under the bridge to the 8.W., the
Lower Clay here resting upon yellow sand (rock débris). At the
N.E. end, at a, it is divided by a bed or tongue of the Lower Clay.
At g is a small bed of shingle, situated at the thinning out of the
Lower Clay.

(Surface-level about 120 above O. D.)

AUK
\
te

‘

\
-
AS

Fig. 1. Section at Bootle-Lane Station.

\
\

\\

8.W.

T. MELLARD READE ON THE

QS

AN

\

TUTTI TATTLE SL

YS
\

Ss

\
Wy

\\
YY
~

\\

Horizontal and Vertical Scale of Feet.

6. Bed of fine unctuous clay or
marl, laminated for a few inches
where itrests on sand below. Above
thisitis homogeneous, unlaminated,
containing a small proportion of
sand, and having boulders, similar
to bed 4, sparsely distributed
through it; contains shell-frag-
ments. It is of a bright red brown,
shining when cut with a spade. A
similar clay was got out of the
Huskisson Branch Dock. When
washed it is shown to contain small
rounded gravel.

7. The ordinary brick-clay from
which most of my shells have been
taken, contains more stones than
No. 6, and more subangular and
striated stones than No. 4. It is
in constitution intermediate be-
tween 4 and 6. No. 6 impercep-
tibly shades into No. 7.

8. Sand-bed. At c¢ containing
bands of stratified clay ; it is here
of a red and orange colour. d is
an intercalated bed of clay and
gravel. ¢ is stratified sand with
thin beds of sandy blue clay.

9. Stratified yellow sand. In
the upper part this contains thin
beds of peat, and belongs to the
‘‘ Washed Drift sand” of my
“ Postglacial Geology of Lancashire
and Cheshire.’ It in fact becomes
subaerial ; it 1s capped by surface-
soil*.

* Specimens of beds 4, 5, 6, and 7,
were kindly examined microscopically for
me by Mr. David Robertson, F.G:S., of
Glasgow. He sends me the following
list of Foraminifera, and says, ‘“‘ It does
not appear that the various beds need
be kept separate, seeing that there is no-
thing in particular in the one bed that is
not in the other..... Cytheridea papil-
losa, Bosquet, is the only ostracod that
need be noticed, which is moderately
common. There are one or two other
imperfect forms not yet made out, but
whose absence can be of little conse-
quence. Cytheridea papillosa is not

DRIFT-BEDS OF THE NORTH-WEST OF ENGLAND. 29

2. LocaLITIES IN WHICH THE SHELLS WERE FOUND.

Toxteth Park.—The whole of these specimens were taken from the
brickfields, and were distributed through the clay. There are thirty
species. Elevation about 120 feet above ordnance datum.

Kirkdale Upper Brickfields—Most of these specimens were taken
from the clay in the brickfields near to the Kirkdale Industrial Schools,
just above Bootle-Lane Station. Level about 125 feet above O. D.
(16 species).

Kirkdale Lower Brickficlds—Yhese occurred in the brickfields
between the latter place and Sandhills Station. The shells are very
sparsely distributed through the clay (9 species).

Bootle-Lane Station.—These shells, though belonging to the same
locality as the last two, were taken from the railway-cutting, where
the distinction of beds could be observed. The upper beds correspond
with the beds of the Kirkdale brickfields. The shells of the lower
bed were collected for me mostly by one of the men working at the
excavations. Surface about 120 feet above O. D. 14 species in Lower
clay, 8 in Upper clay.

Bootle Northern Outlet Sewer.—Surface-levels are between the 25
and 50 feet contours. The shells were all picked by me from the
clay thrown out of the excavations. It was pretty rich in shells.
To all appearance the bed was like the preceding brick-clays. A nest
of decomposed greenstone boulders occurred at one spot. No. of
species, 26.

Garston.—These were picked promiscuously out of the railway-
cutting and dock excavations. Those from the dock were below high-
water mark (9 species).

River Dee.—Yaken from the face of sea-cliff between Dawpool and
West Kirby, 9 species. Mr. Mackintosh has given a list of fifteen
species from the lower Boulder-clay, thirteen of which I have marked
A in this column; the other two species are Scrobicularia alba and

%
uncommon in our Scottish glacial clays, raised beaches, and in the present
surrounding seas.

In sample of bed No. 4 there were one Echinus-plate and some small pieces
of Polyzoa, Salicornaria.

FoRAMINIFERA.
1. Biloculina ringens, Lamk. 9. Bulimina pupoides, D’ Ord.
2. Quinqueloculina seminulum, Zinn. 10. Discorbina rosacea, D’ Orb.
3. Ferussacii, D’ Orb. 11. Truncatulina lobatula, Walker.
4, Lagena sulcata, W. & J. 12. Rotalia Beccarii, Linn.
5. globosa, Mont. 13. Polystomella crispa, Linn.
6. marginata, Mont. 14, striato-punctata, I. § M.
Ue squamosa, Mont. 15. Nonionina asterizans, IF’. ¢ M.
8. Polymorphina compressa, D’ Ord. 16.

depressula, W. & J.

This last species is the prevailing form; Lagena sulcata, Polystomella striato-
punctata, and Nonionina depressula are moderately common ; the others are
represented by ones and twos; none exceeds three.”

30 T, MELLARD READE ON THE

Lacuna divaricata (Q. Journal of the Geological Society, vol. xxviii.

. 39).

. ian of Edge-Hill Station.—A portion of these were obtained by
me out of the brickfields; the remainder from the railway-cutting
just above. Level about 175 feet above ordnance datum. The
clay is similar in constitution to that of all the preceding brick-
fields. Number of species in railway-cutting, 37; brickfields, 14
species.

; Warrington.—These shells were found by Mr. Pears, and are now
in the Warrington Museum. ‘They are described by Mr. Paterson
as occurring in a sand-seam on the Liverpool Extension Railway
between Bewsey and Sankey. Ihave myself examined the section,
which consists of stratified beds of sand,-included in the Boulder-
clay. Jam indebted to Mr. Darbishire for the correction of the list,
the one published by Mr. Paterson being incorrect; he informs me
that the condition of the shells bears a striking resemblance to those of
the Blackpool Middle sands. Surface-level about 50 feet above O. D.

Birkenhead tramway-road, 13 species.

Gas-works, Linacre.—About 50 feet aboveO. D. These were ob-
tained by Mr. Isaac Roberts, F.G.S., during the excavations for the
gas-holders ; all of them appear to have been taken from the clay
(see Proceedings of the Liverpool Geological Society, 1870-1, pp. 68,
69). The individuals were more perfect than most of those found
by me.

: Various other Localities about Liverpool._—These were found by Mr.
G. H. Morton, F.G.S., and others, and are given by him in the Pro-
ceedings of the Liverpool Geological Society, 1871-72, pp. 92 & 93.

The last column represents shells of the list, which are to be found
tiow on the Southport or Formby shores. I am indebted to Mr. Chas.
Brown, of Southport, who thoroughly worked out the conchology of
the district, for these particulars. Those marked a were found by
him alive.

Note.—For purposes of comparison, I have given Mr. Darbishire’s
list of Moel Tryfaen, Macclesfield, and Blackpool shells; but in
addition to these there are the following species and varieties, repre-
sentatives of which have not been found by me, viz. :—

Pholas candida. Fissurella reticulata.
Mya arenaria. Trochus cinerarius.
Corbula nucleus. Littorina rudis.
Tellina proxima. littoralis.
Mactra elliptica. Lacuna vincta.
Venus casina. Natica clausa.
Artemis lincta. monilifera.
Astarte crebricostata. Trichotropis borealis.
Cardium aculeatum. Fusus gracilis.
norvegicum. Trophon barvicensis.
fasciatum. scalariformis.
Mytilus edulis. — Gunneri.
Nucula. Mangelia nebula.
Arca lactea. pyramidalis.
Patella vulgata. Cyprza europea.

Dentalium abyssorum.

| | [To face page 30.
AY ABOUT LIVERPOO]

‘ales, as well as species found),

Extracted
SAAC from Hig
: Morron’s
No. i List. TH, RanGE, &c.
irae NT aa (Among stones and Huci, and
p-water mark of neap tides,
39, | * O a (S. Wood); C. and R. Crag |
: pms; muddy and weedy locali- {.
Chief habitat, the shallower |
great numbers in from 7-10
33. | % | crereerecees vr (00 fathoms; affects gravelly i
34, Jeers ? seeree Ind 650 fathoms; muddy sand, }
he coast of Northumberland. |
ind of wide distribution.
35, beeen O r lentiful in the Irish Sea; be- |
‘ north.
36, | * O f jlow tide-marks ; most variable |
37. | * O 7 al, sand (verge of littoral and }
an zones). E
98, frreeeee (S) vr fathoms; gravelly and stony |
j hark to 145 fathoms (Beechey).
39, | * O Ff \ter to 100 fathoms. Extended
g glacial epoch.
4Q, prrsesttr | cereeees wees ¢ Pliocene (extinct in Mediter-
8, living on various kinds of
shell-banks.
41. %* Ganoccacces ||] ‘dosacK lathoms.
49. | ¥ senveceecees fo bst generally diffused glacial |
(Stony ground in the lami- |
eep-water zones, Jeffreys.)
AQ, Jercseors |e yacsevsvsces i Essentially northern. ‘Ten- }
)-150 fathoms, Jeffreys), |
44, | * | etcoeonsc: ur ind coralline zones.

reely range south of the Bri
t most prolific in Celtic and
nd mostly increasing In num
nland and Boreal America.

cea aat ad gat ai

nie
i

ee ANE Or * mie
oes i: :

LIST OF SHELLS FROM THE BOULDER-CLAY ABOUT LIVERPOor, [To face page 30.

Comparative columns are given, showing the occurrence of the same species in other localities in the N.W. of England and Wales,

as well as species found by others in the Low-level Boulder-clays of Lancashire and Cheshire.

Low-loyel Lancashire Boulder-olay, Mr. Darnisuine’s List.
p Recent. Now]
Isaao | Extracted Low-leyel | found on
Parensox’s Rama from igh: Southport
Found by Mr. D, Metranp Reape. eaten Rosunte' inarea High: Toyel Sands and Bonldos-lay| “hore
No. Srecizs. ae Driihieall re ices Hanrrar, Derm, Rasox, ko,
| ; Brows,
: Bootle-Lane | Bootle Riyer Dee. | N.D. of Edgehill Various oth —
Toxtoth | *irkdale Station. Northern! Garston, | ararzea, Station, Birken-| Warsington.| Zinaero | localities | Mfoot | Bacclessield.
Parks eee fae aw | wet found by Me) Tatteay.| Leds Blon«) Gasworks. | about | Tryfen. Blackpool.
2 ia) padita| Jaret | Upper | Sewer. Mackintosh. | Brickiteld| Gutgrey Liverpool. Older. | Newer.
Pholas crispata, Linné... ot | [ean Eee eter | Uemreee nl Meere cat | r-raaas Sacer |) |! eats avec rere | nor | eee wa 2 |C. Omg. Burress Various kinds of rock, clay, gyp-
: b sum, and peat (Jeffreys). f

2. | Saxicava rugosa, Linné .. teres | seteee sree r he r tenses | seeese * 2 |C. ond R. Coe. Burrows in limestono, chalk, and
sandstones. Most. generally distributed glacial’ shill
(Forbes), Vertical range great, near low water to
145 fathoms.

3. |—norvo ca, Spengh. wees) cece | seers Peon ates cook || core atenee Gopstrox ||| seaceee ur sense seesaceeneee. coon tp gosthotiiu | 4 \C.and R. Crag. 30 fathoms at Durham, and 82 fathoms
at Shetland (Jeffreys). Abundant in the glacial clays
of tho Clyde, with valves togother as if in sifu. Only
onol men has been found on the Norwogian

: A few specimens havo been obtained
astof Northumberland and Durham. Fossil
; in Mediterranean.
4. | Mya troncata, Linné ... f f seen i sen f f f f f a o Decerccccerren eretrreeere f i r r +a | 1 Cand R, Crag. Mediterrancan in glacial epoch. Sand
| and mud and Bravels aes and deep-sea down to
oor 145 fathoms. Clyde beds in sifu.
Big | Essmimobin Xerxoenete;|Chemr 55/7. be cree eee ees r “ ve 2 S o Be |) cnn . 2 lo Gee Raro in glacial bats (Porte). Sand) and
; mud, 3-90 fathoms.

6: | Donax vittatus, Da Costa Ca errr eccccr eceer eer ef cee | cee . or Or | sess ur vr “2 2 |Sandy shores. Littoral, near low water to 25 fathoms.

Few of our bivalves moro uniyersally dilfused on
‘ Sa British shores.

7. | Tellina balthica, Linné a a a a a a a aa a a * o x (o) r c f c wa 2 |First appeared in the glacial epoch. Sandy bays and

inlets.“ Unerringly marks the shallow water under
REF which Pleistocene formations accumulated" (Forbes),

8. | Mactra solida, Zinné ... r 50000 r r cd r A r i . EVs | tess ores ts r r r r . 1 /C.and R. Crag. Usually littoral. Sand and gravelly
sand, near low-water mark to 15 fathoms; 39 fathouis

9. | Lutraria elliptica, Lamarck ...... iy e a f fi) cua ¥ i cra| gece é 2 Oozy sand and mud. Low water of

15 fathoms; 12 fathoms in Anglesea.
pee i Abundant round the British coast.

10, | Scrobicularia piperata, Gmelin...) of |... pan fetes ae | siti ra Ue eee ss ano. |) ei Wai 2 \R. Crag. Littoral. Estuarine mud whore fresh water

| | casionally flows over them, Beds of mud and cls
a s at low-water mark, and as deep us 4 fathoms (Jeffr

11, | Venus chione, Linn€ scene) i. | cesta |) ease on || Perce r 9 r ef oo | f 5 |. Crag. 12-20 fathoms in eiscrat Bay. “R

turned southwards during prevalence of glacial con-
— - ditions” (?) (Forbes),

12. | Tapes virgineus, Zinné ... oa eerste |Werteesre ut yetess ts |) cea) “i]|ftesae aren || eesonee: ||! ccs |} craig HMeesessstesash ll werracencess DW cet. || exuce 2 |C.and R. Crag. Littoral, to 145 fathoms. Tt appears
to bo more of a southern than-a northern species

en (Jeffreys).

13, | Venus gallina, Linné .... r ae | Wena r r r ur r ur r ea 2 |Sandy ons near low water to greatest depths (

| | thoms, Jeffreys). Appears to have commenced its
hen | | existence during “Newer Pliocene.”

14. | —exol eta, Zanné. r Wom | isenerwid [eres | Roce el || Secece wl Medes? ll iaecttewse tA) dectee- || artes | | . - ur . 2 \0. Crag. Low water to 80 fathoms, Sandy bays.

15, | —lineta, Pult. . vr | r \C.ond R. Omg. Sand, sometimes mixed with mud;

et . ™ | i low-water mark to 00 fathoms.

16. | Cyprina islandica, Linné. c c Chom |) Eee ener c c © ce . o * 2} e ° f . 1 |Crag. 5-80 fathoms (low water to 100fathoms, Jeffrey).
Extended to the Mediterranean during glacial epoch,
Very variable in form, being of wide range.

17. | Astarte sulcata, Da Costa ... ans |) orien |) aaete I} eects II) succes Chea rests AVON |fircsnset | ieee unin cron vr sor eee 1 /Rare in Crag. Sand end mud, 7-85 fathoms (145 fa-

La thoms off Mull of Galloway).

18, | —clliptica, Brown .... cae | one [ieee int, # # rs r ? les r oon i cr ae 3 |Crag (only on authority of Woodward's Norwich-Crag
list) 5-10 fathoms, mud (10-40 fathoms, on muddy
bottom, Forbes),

19. | —— compressa, Montagu ... epson Ween Ce ners ca | aveecee | Patera r . ? c sien cctent Gre || cate fs | 3 \C. and R. Crag. 7-40 fathoms; 70 fathoms, Zetland
Seas: muddy bottom. Found in fathoms, cold
area north of the Hebrides (Jeffreys). Sand, often
mixed with mud.

20. | — borealis, Chemn. e ° c c © ° c 4 e . (oy 9) eter cos c e ur 4 |Crag. Fresh single valve taken in 80 fathoms. The

[This T consider to be typ) most southern known limit of its habitation is Kiel
cal foal et the Bonlder-ley, Bay, in the Baltic (Jeffreys).
calities.]
21. | Cardium echinatum, Linné......... c ° ‘ c © cy WM cde oA c e . * r iP r r . 2 |R. Crag, 7-80 fathoms. Mud and sand, and muddy
i gravel.
22. | — tuberculatum, Liané ..., site Press ur cee eernns seuu. | S$ [Sandy bays. Low water spring tides to 12 fathoms.
23. | edule, Linné .. a a a c r ea | 2 |R. Crag. Littoral, low-water mark to a few fathoms.
| Sand. Replaced in Greenland and Boreal America
| by Cardium idandicum.

24. | Mytilus modiolus, Zinné.... Fie |\pece.. f pat | lca, a oe MP eer r f if . ve r r ano oiainceox *a 1 Crag. Low water to 60 fathoms. Small at great

depths. Frequents gravelly and muddy localities,
a most frequent from 7-30 fathoms.
25, | Leda pernuls, Miter Co) eee | ieee Wierescial ieee ur 4 vr hell cana ll nn co r 4 |Not in Crag. Mr. Jellreys dredged a young live spe
men in 80 fathoms, off the Shetland Isles. Arctic
A and Scandinavian seas, from 10-100 fathon
26, | Pectunculus glycymeris, Zinné...) or | 2 | uu. | een Sea | erase Af r ur ur eccrc ana | 2 |Dies out in upper beds of Crag. 15-60 fathoms (7-00
| fathoms, sandy and shelly gravel and Nullipore, Jé/-
freys).

27, | Pecten opercularis, Zinné ....., muh ste | yaes ieee waters r At o * r vr r fer OG |e (x ed R. Crag. 5-100 fathoms. Common to nll
sandy coasts,

28. | Ostrea edulis, Linné.. c Pea | a sscciall zee. | Paton 37° ||| Serene] Meepenerra eRe r on *% {o) ur ur crn r ea 2 |¢. Gan lower beds of Red Crag. 12-26 fathoms (low

_ i Fy water to 45 fathoms, Jeffreys).
29, | Dentalinm entalis, Zinné . Gl eerccoe tl Iicceertoell eons A litesice Gl cofecne ||) Spoor on TIN) |Visesrey o tenes ® 2 )R. Crag. Sand, or sandy mud, 3-100 fathoms. 145
4 a futhoms off tho Mull of Galloway (Jeffreys).
80. | — tarentinum, Lamarck nies or 5 |©.and R. Crag. Low wator to 25 fathoms.
81. | Littorins litores, Zinné .. r r 1 |Numerous varieties in R.Crag. Linble to great changes
~ ————}- ~ ‘inch peranch coloursy(Amongestones und uch and
on rocks below high-water mark of neap tides,
effreys.)
| Durri inné. 2 |Vory raro in the Crag (8. Wood); ©. and R.

Be ecrecou ere CONTE cette) a G a a a a a aA a @ 0 o % ° ° é £ a “CRortes). 4-100 fathoms; muddy and weedy local
ties in great abundance. Chief babitat, the shallower
sea-beds, occurring in great numbers in from 7-10
fathoms.

33. | Aporrhais pes-pelecani, Linné ... r r te SE [Wee nal pees ae “ r ee * i ur ur vr or . 2 (eS eR Me Crag. 4-100 fathoms; affects gravelly

4 ; onion 3 |North of Hebrides, 189 and 650 fathoms; muddy sand,

SA. | Notica gramlandios, Beck ss...) coor Ee |) ctrocaa. |) asrempee | | eeeeaee sore. ||| Ce || cee ? gee || one || 29 40-60 fathoms, off the coast of Northumberland,

Exeentilly northern, and of wide aisteibution
F : . 2 [5-20 fathonis. Most plentiful in the Irish Sea; be-
985, | Murex crinacous, Zinnb .....sccc00.| cesses | sseeee oat rere lieccen. || -ahes Peel lWesissvel |i anit illsrevssee a ° fr f ig f ‘aoe mearcac ae Teo RET
4 ° . 1 |R.Crag. Rarely lives below tide-marks; most variable
96. | Purpura lapillus, Linné .. PN CA ered eee AaB ll a7 SBN ROR sl yaoee | nk Pace Oo * ° ee eta he oe “i of Una TTestcea,
fi ‘ vccmmene | 2 (O.and R, Crag. Littoral, sand (verge of littoral and
87. | Nassa reticulata, Linné ............ conor || [Ieee coh r coum r snd 4 crboss r eons Oo x (e) ia Cha Z i jo. tbper part of laminarian zones), is oe
P . = . 2 |0. and R. Crag. 0-50 fathoms; gravelly and stony
38, | — incrassata, Strom .. obi: oxpip:|||erans cme, é ° ur 7 | is Sound Law-water mark to 140 fathorns Bechey)
- ¥. “a 1 |C. and R. Crag. jw water to 100 fathoms. Exteni

99. | Buceinum undatum, Linné........) f FE | ever | es i f f i . o ¥ fe) 8 : f ta Mediterranean Glring glacial poe.

. A sa 1 JR. Crag. Sicilian Nower Pliocene (extinct in Mediter-

40. | Pusus antiquus, Linné..... r ee os aacuee con ry r r cerere o *nettaxe G “4 ranean). 5-30 fathoms, on various kinds of

I ground, but preferring s| nks.
Peery 3 |C.and R. Crags. 5-80 fathoms.
ai rae eeraas eer Sonne on coro cop % sere setae 3 (Crag. One of the most generally diffused glacial
. phon truncatus, Strom .. r r f ¥ shells; 5-60 fathoms. (Stony ground in the lami-
marian, Cae ind doep-water zones rege)
2 1 rag. 3-100 fathoms, Fssentially northern. ‘en-
Pleurotoma turriculs, Montagu.) se. | cee. Buin ||| oC | seeere| Ue eee Mee eeeeA Pa Se rer o Pernt f dency to vary great (10-160 fathoms, Jeffreys),
Sand, in the latninarian and coralline zones.

Shells marked 1 are species now living throughout the Celtic Tegion in common with the North seus, and scarcely range south of the British ep j
2 now living, ranging far south into the Lusitanian and Mediterranean regions, but most prolific in Celtic and North Seas. il = ch the Arctic Circ!
& a ” still existing in the British seas, but confined to the northern portion of them, and mostly increasing in number of individuals as they approseh the Aretic Circle,

3 van =
4 a 4 mn now known living only in European seas north of Britain, or in the seas of Greenland and Boreal America,
» » Soon fossil in the Crag, but now restricted to British and South-European seas, .

DRIFT-BEDS OF THE NORTH-WEST OF ENGLAND. 31

3. Mops oF OccURRENCE AND ConDITION OF THE SHELLS, dc.

The whole of the shells in the appended list, consisting in all of
44 species, occurred distributed through the drift. In nearly every
case they were taken from the clay beds ; for though in the included
sand-seams shells are occasionally met with, they are, as a rule, rare.
Being small, fragmentary, and rolled, sparsely distributed through
the clay, and not in any case found in zones or beds, it would be very
difficult to make a collection from the faces of the clay in excava-
tions; but when the clay is “ cast ” for brick-making, and has been
subjected to heavy rain, the fragments are weathered out and are
more readily found. The bulk of my specimens have been obtained
in this way by my own hands, and are undoubtedly genuine.

Beds or zones of sands containing shells more perfect in their state
of preservation are, however, occasionally met with; more perfect
specimens than mine of some of the species have also been found in
the clay, such as those taken by Mr. Isaac Roberts, F.G.8., from the
excavations for the Linacre Gas-works*. The Warrington shells,
which are, according to Mr. Darbishire, in about the same condition
as the Blackpool specimens, were (if the locality be correctly de-
scribed by Mr. Paterson?) taken out of stratified beds of sand and
gravel, which I have myself inspected in the excavations of theLiver-
pool Extension Railway near Sankey Bridges. These stratified beds
bear a closer relationship to the sands and gravels of the so-called
middle drift at Blackpool than to the clays from which I have obtained
most of my specimens.

Strange and puzzling as is the condition of the fragments and their
distribution through the clays, it is one of the main peculiarities
which should point to a correct interpretation of the way in which
these beds of drift were laid down.

Professor Forbes, who has remarked on this in his admirable con-
tribution to the literature of the Glacial periodt, which, even now,
for comprehensiveness and grasp is unexcelled, and who is evidently
much puzzled to account for the usual fragmentary character of the
shells and their tolerably even distribution through the drift, has
imagined it to be due to the ploughing up of icebergs or the great
rush of ocean-currents§ ; but these suppositions, as I will presently
show, cannot be more than a limited and partial explanation of a
general and wide-spread phenomenon.

It is worthy of notice that the shells most generally found whole
are of a form most calculated to resist pressure, and, in some cases, of
quite minute dimensions. T'wrritella terebra, found everywhere in
the greatest abundance, is not unfrequently perfect. Trophon trun-
catus and Pleurotoma turricula are also found perfect, while the large
Fusus antiquus, and Buccinum undatum are generally represented

* «Proceedings of the Liverpool Geological Society,’ 1870-71. p. 68.

+ Procceedings of the Warrington Literary and Philosophical Society.

¢ “Fauna and Flora of the British Isles,’ Memoirs of the Geological
Survey, vol. i. p. 383.

§ Ibid. p. 384.

32 , MELLARD READE ON THE

only by well-worn fragments of the columella. Oyprina islandica
is invariably found in fragments, sometimes angular and sometimes
worn, while the hinge, being more adapted to rolling, is usually much
rounded and worn. It is quite exceptional to find a perfect valve of
a bivalve, and it is confined, as far as my experience goes, to small
and strong valves. I had the good fortune to find a perfect valve of
Leda pernula at Edge Hill*. I have also a perfect valve of Tellina
balthica, retaining part of the colouring, from the sewer in Thomas
Lane, Broad Green. Some few other species occasionally occur more
or less perfect ; but we may safely say that, as a rule, those only are
preserved which, by the peculiarity of their form, or their minuteness
combined with their form, are fitted to resist pressure or rolling about.

Again, the association of the various species, distributed entirely
without order through the clays, shows that they could not possibly
have lived together on the same bottom, some being peculiar to sand,
others to mud, some to rock, and others to shingle, some requiring
deep water, and others shallow ; so that the conclusion is irresistibly
forced upon us that they must have been to a large extent trans-
ported. Had the confusion been due to the ploughing up of icebergs, as
suggested by Forbes, the disturbance at each stage of subsidence would
not have reached beyond a certain depth below the surface of the sea,
and all below that depth would have been free. It isin fact impossible
to construct a satisfactory explanation which does not allow for a
successively varying depth of sea-bottom from zero up to 1400, and
perhaps even 2000 feet. We have, as before shown, every reason
to infer that some at least of the bottom beds of the drift, of which
I purpose giving you sections and more fully explaining in my second
paper, are those originally formed on the first subsidence of the land
after the retreat of the ice-sheet to the mountain-districts of the

north}. In several examples we see the bottom-clay reposing upon

the sand ground-moraine derived from the Triassic rocks, with the
line of division as sharply defined as could well be ; and resting upon
the sand and in the clay are boulders and pebbles, evidently transported
from the north, and having their surfaces scratched, indented, and
polishedt. This bottom-clay is largely composed of the red sand
upon which it rests, and out of which, together with mud from more
distant localities, it has been formed. ‘This is the distinguishing
difference between the inferior and the overlying clays, and varies
from a strongly marked distinction, according to the proportions of
the included sand, to a difference of appearance only minute. If,
however, we examine the constitution of the clay itself, we find no
real or appreciable distinction from that above it ; for in both the same
distribution of shell-fragments occurs, and the included boulders are

* Now in the possession of Mr. Darbishiire.

+ The existence of the ice-sheet and the proofs of the statement will be given
before the subject is completed. (See also abstract of the author’s paper on
Glacial Strie at Miller’s Bridge, ‘Proceedings of the Liverpool Geological
Society,’ Session 1872-73.)

{ These are found 162 feet below the surface at Widnes, or 140 feet below
the ordnance datum in the ancient bed of the Mersey. See ‘The Buried Valley
of the Mersey,” by the author, in the same volume.

DRIFI-BEDS OF THE NORTH-WEST OF ENGLAND. 33

of the same mineralogical type and most probably from the same pa-
rent rocks, the larger proportion being scratched and polished. Now
assuming, as we are justified in doing, that the lower beds of the
Lancashire clays are older than the Macclesfield and Moel Tryfaen
drifts*, it is quite evident that in some of the numerous sections I
shall show you we ought to find representatives of the deposits which
must have taken place at all depths between the two extremes men-
tioned, both during subsidence and after elevation. That, however,
which constitutes the great puzzle, and which led Forbes to infer the
drifts to be all shallow-water deposits, is the occurrence of littoral
shells, the absence of deep-sea corals, and the extremely similar, not
to say identical, character which they all exhibit. It is true one
part is more sandy or stony than another, one part contains gravel,
and another consists almost wholly of a fine unctuous clay; but
throughout are the broken and rolled shell-fragments and the
scratched erratic pebbles and boulders.

It is easy, however, to see that the subsidence and reelevation of
the land to a vertical extent each way of 2000 feet must, by the
deflection of tidal currents (each stage of vertical movement having
its own proper system of stream-tides), alternately bring every por-
tion of the sea-bottom under its erosive action. That these tides,
assisted by winds and storms, must act on the coast-lines and sweep
off the shells thrown up on the beach, again to distribute their
fragments over the sea-bottom, is equally evident. That tidal action
is effective at much greater depths than is generally admitted, I
have satisfied myself, and, having carefully studied the question,
shall, I hope, at a future time bring before you sufficient facts to
prove it. It also presents a probable explanation of much of the
curious bedding and stratification met with so generally in the
Drift. From the condition of many of the boulders and pebbles so
irregularly scratched, and from a consideration of the climatic con-
ditions of the period, pointed to by many facts, I infer the prevalence
of coast-ice—an additional cause for the distribution and extreme
comminution of most of the shell-specimens. It must not be lost
sight of, if my explanation of the phenomena be correct, that though
the shells are in fragments, they are (unless they are derivative, of
which there is no proof) as truly representative of the conditions
prevailing at the time of their deposition as if they had been found
on the spot where they died.

4, AnAtysis oF THE List, CoMPARISON WITH THE PRESENT MoztLuscovs
Fauna oF THE BritisH sEAs, AND INFERENCES THEREFROM f.

The most generally diffused shell in the Lancashire and Cheshire

* It is necessary to assume this for our present purpose ; but the question will
be discussed 7m extenso and proof given when I deal stratigraphically with the
beds.

t For the identification of most of the shells I am indebted to Mr. R. D. Dar-
bishire, F.G.S., who has with great care gone over from time to time some pounds

Od, Gas, Ne. 117, D

34 T, MELLARD READE ON THE

Boulder-clays, as well as the one found most frequently perfect, is
Turritella terebra. Tellina balthica comes next in the scale of fre-
quency, but is seldom found perfect. Cardiwm edule, C. echinatum,
and Cyprina islandica, always found in fragments, follow in the
order given; but the typical fossil, the representative of the climatic
conditions prevailing at the time the clays were laid down, is un-
doubtedly in my opinion Astarte borealis. Though not the most fre-
quent, itis a common fossil ;+and I have found it in all the localities
mentioned in the list, as well as in other places unrecorded. It has
also been found in the same series of clays at Blackpool, Warrington,
Chester, and other places, by independent observers. Unlike Leda
pernula, which, though occurring in the same clays, is rare, or Saai-
cava norvegica, which is represented only by two fragments, Astarte
borealis may safely be taken as indicative of the then condition of the
British seas. According to Mr. Gwyn Jeffreys, “the most southern
known limit of its habitation is Kiel Bay in the Baltic.” Astarte
compressa, A. elliptica, and Trophon truncatus, also fossils of the clay,
though still existing in the British seas, are confined to the northern
portion of them, and increase in number of individuals as they approach
the Arctic circle. A reference to the list will also show that twelve
of the remaining species are northern and Celtic shells, scarcely
ranging south of the British seas. Twenty-one range far south into
the Lusitanian and Mediterranean regions, but are most prolific in
the Celtic and northern seas ; while Venus chione (several well-marked
hinge-fragments of which as well as other fragments I found at Edge-
hill), Dentalium tarentinum, and Cardium tuberculatum (only single
fragments of which occurred to me) are southern shells.

Though all, with the exception of Astarte borealis, Leda pernula,
- and Sawicava norvegica, may be described as Irish-Sea shells, yet,
if we make a fair comparison, we shall find that the Boulder-clay
assemblage possesses a more northern facies than the present fauna
of the British seas. If I may so express it, the area of the Boulder-
clay fauna and thatof the present British seas overlap; but their centres
are considerably apart, that of the former being in a higher latitude.
To assist in the illustration of this idea, I have given a column showing
the species now found on the shore between Southport and Formby,
the nearest locality for comparison ; and reference to it will show that
the most characteristic northern forms are absent. There are in
addition also many Celtic and southern forms found on the same
beach which are absent in the Boulder-clay.

The presence of Venus chionein the low-level Boulder-clays as well
as in the Macclesfield drift where it was first discovered by Mr. Dar-
bishire, is certainly a remarkable fact ; but it is not an isolated one ;
for Cardium aculeatum and C. pygmeum, both Lusitanian forms, are
found in the Scotch drift; and in the Irish drift southern forms also

weight of fragments, and has assisted me in various ways. To Mr. Gwyn Jeffreys
I owe the identification of some of the more doubtful species, and a revision of
my list.

DRIFT-BEDS OF THE NORTH-WEST OF ENGLAND, 39

occur*, If weturn our attention to the east coast of North America,
we may find a possible explanation of this curious mixture; for at
Cape Cod, in latitude 42°, and, I believe, also in higher latitudes,
arctic and southern forms are now dredged up alive from the same
bottom.

D. Posrtion or THE LANCASHIRE AND CHESHIRE LOW-LEVEL BovuLDER-
CLAYS AND SANDS IN THE GLACIAL SERIES.

A comparison of my list with that of the shells from the Middle
drift at Blackpool will hardly, I think, establish the hypothesis often
broached, that the so-called Middle Drift represents interglacial con-
ditions. No doubt fluctuations of temperature occurred during the
laying down of the Boulder-clay series. Though I have found about
Liverpool a greater number of northern forms than occur in the
Blackpool sands (the latter being, I consider, but a local develop-
ment of the Drift), it is impossible to draw with safety wide genera-
lizations from the paleontological evidence. On the other hand the
so-called Lower Boulder-clay of Mr. Mackintosh at Dawpoolt has
not yielded him a more northern assemblage than the Brick or Upper
Boulder-clay of Lancashire has yielded me§. It is partly on these
grounds that I have ventured to group together all those beds
which are typically represented in the Bootle-Lane Station, under the
title of Low-level Boulder-clays and sands. Whether or not there
are stratigraphical grounds for their subdivision will be discussed in
my concluding paper.

For purposes of comparison I have appended a list of the Moel-
Tryfaen and Macclesfield shells||; but though, when compared with the
Liverpool shells, the former shows a more and the latter a less northern
facies, I do not think we can safely deduce therefrom the relative
ages of the Drift in which they occur. Proximity to the mountains
generating the glaciers may have been tke cause in the first case,
local circumstances in the latter. It is only by a consideration
of their structure, position, and stratigraphy, combined with the

* Forbes notes the discovery by Captain James of T'wrritella incrassata, a Crag
fossil, a southern form of Fusus, and a Mitra allied to the Spanish species in
the Wexford gravels.

t It must not be lost sight of that glacial conditions in a low latitude would
produce extremes of heat and cold greater than would occur in a higher latitude
having the same mean annual temperature, and consequently currents of variable
temperature would result. Forbes says the distribution of Mollusca depends not
on latitude but temperature.

+ Quart. Journ. Geol. Soc. vol. xxviii. p. 388.

§ See ‘‘ Notes on various Shells found in Stratified Drift near Macclesfield,”
Memoirs of the Literary and Philosophical Society of Manchester, Session
1864-65.

|| The Lower Clay of Bootle-Lane Station is an exact counterpart of the Lower
Clay at Dawpool; but that, again, yielded me the same characteristic fossils, as
will be seen on reference to the list.

D2

36 T. MELLARD READE ON THE

oscillations of level which occurred during their deposition, that the
relation of the high-level sands and gravels to. the low-level
Boulder-clays can be satisfactorily made out; and when this is
done the palzontological evidence may very properly be taken
into account.

Considering, then, all the low-level beds as a group, what position
do they hold in the Glacial series? Before answering this question it
will be necessary to glance at the condition of the base-rock upon
which they generally rest. It may be safely asserted that wherever
the Boulder-clay is removed in Lancashire and the rock exposed, if
it is fitted to retain striations and groovings, there they are sure to
be found. Where the rock is unfitted to receive them, it is, so far
as my observation extends, ground to powder, broken up into closely
compacted rubble, or otherwise displaced or shattered. Every thing
bears evidence of the passage of an ice-sheet over the country in a
north-westerly direction ; and this, it is needless to say, represents in-
tense glacial conditions. When, however, we come to the Marine Drift,
if paleontological evidence is of any value, the climate must have
been much ameliorated, being probably such as is found in the northern
parts of Labrador. Thus, then, if my premises be correct, there
must have been a great break or lapse of time between the cutting
of the strize and the deposition of the Marine Drift. If so, during this
period the rocky surface, together with the detritus upon it, must
have been subject to subaerial conditions; for I can find no evidence
of any marine clays (though I have examined the base and searched
for them in many places) older than those forming the subject of this
paper.

If, on the other hand, we turn our attention to the east coast of
England and Scotland, we find evidences of much more severe cold
prevailing during the deposition of some of the beds of Marine Drift
than any that are to be discovered in Lancashire. The Arctic shells
found by the Rev. Thos. Brown at Elie in Fifeshire and Errol in Perth-
shire were declared by Dr. Otto Torel to be identical with species now
living in front of the great glacier of Spitzbergen*. The Bridlington
beds, on the coast of Yorkshire, have also yielded a molluscous fauna
of a more arctic character than either the Lancashire clays or the Clyde
bedsf. Mr. Searles Wood, jun., whose elaborate investigations of the
East-Anglian glacial phenomena are of so much value, also considers
that he has discovered a complete series of beds commencing after the
Cromer Forest-bed with a crag-like fauna, gradually developing into
an arctic one, and thence shading down to the Celtic fauna of the
present seas. If Mr. Wood’s reasoning is correct, and be coupled with
the break I have shown existing between the period of the ice-sheet
in Lancashire and the deposition of the marine clays, it seems diffi-
cult to escape from the conclusion that the north-west lowlands were
under subaerial conditions while the east of England was depressed
below the sea-level. It would thus also follow that the low-level
Boulder-clays and sands of Lancashire belong to the later glacial

* Trans. of the Royal Society of Hdinb. vol. xxiv. p. 627.
T Quart. Journ. Geol. Soc. 1869, p. 97.

DRIFT-BEDS OF THE NORTH-WEST OF ENGLAND. 37

group; but for a full discussion of this point I must wait until I am
enabled to lay further evidence before you*.

Having now sketched, though imperfectly, some of the principal
elements in the formation of correct conclusions as to the age and
sequence of the Lancashire Drift, I shall be enabled next to approach
the question of distribution and stratigraphy, which I have investi-
gated over a large area of the N.W. of England, with greater freedom
than I otherwise could have done. As yet, however, I have not dealt
with the travelled and striated blocks, boulders, and pebbles, which
will fall more naturally to the concluding division of my paper, in
which the structure of the beds will be described and discussed.

* The speculations on the physical geography and causes producing the climatic
conditions prevailing at the time, I have omitted for further consideration and,
if necessary, combination with facts to be brought forward in my concluding paper.

[For Discussion, see page 41. |

38 R. D. DARBISHIRE ON A DEPOSIT OF

7. Note on a Deposit of Mrippre Puiristocenrs Graven in the
Worpen-Hatt Pits, near Leyntanp, LAncasutre; and on a Cot-
LECTION of SHELLS and FRaemeEnts of Suetis found there by
Miss M. H. Frarineron. By R. D. Darsisuire, Esq., B.A.,
F.G.8. (Read November 19, 1873.)

Axovr a mile from the Leyland station, on the east side of the Bolton
and Preston railway, there is an extended flat-topped hill of small
elevation, which has for many years past been the chief or only
source of supply for a considerable district, of coarse and fine road-
gravel. The pits from which the gravel is got have been worked from
the eastward. The material is screened and sorted on the ground,
the stones and gravel carried away, and the refuse left to fill up
holes. At present the exposed section of the hill is seen in two
large excavations as a face some 40 or 50 feet in height, of which
the upper 10 feet consists, under the sward, of yellow brick-clay.
The remainder of the face exhibits a thick bed, 30 to 40 feet thick, of
marine gravel (shingle) and sands, very rarely showing any clayey
admixture. The gravel consists of pebbles of a very large variety of
rocks, ranging in size from that of a cylindrical mass 8” by 18” down
to that of coarse and fine shingle and sand. ‘There are occasionally
small layers, of slight extent and thickness, of fine sea-sand; and the
whole mass rests on a bed of the same sort of sand, the depth of
_ which has not yet been ascertained.

The accumulation of stones from different rocks and from different
strata is remarkable, and will, it is hoped, be worked out hereafter.
A characteristic fossil of this bed is coral from the Mountain Lime-
stone. This occurs not unfrequently—masses, of several species,
lying amongst the gravel, and, having had the matrix dissolved
away, exhibiting the polypidom singularly clean and perfect. A
certain small proportion of the larger stones exhibit glacial striation ;
arid, much more rarely, small much-worn flints occur, of a grey or
brown colour when fractured.

Throughout the gravel there occur very sparsely small fragments
of marine shells ; but near the bottom of the bed these, with whole
shells occasionally, occur more frequently, immediately above the
underlying mass of fine sand.

The elevation of this lower shell-bearing layer is about from 240 to
250 feet above the level of the sea. For many years past Miss M. H.
Ffarington, of Worden Hall, has carefully collected shells and frag-
ments as they occurred in these gravel-pits. The collection which
she has made is remarkable for the range of species, for the abun-
dance of specimens, and for its clear exhibition, upon the result
of years of collection, of the comparative frequency of occurrence of
species.

MIDDLE PLEISTOCENE GRAVEL IN LANCASHIRE. 39

The annexed list enumerates all which have yet been identified.
The names are given according to the British Conchology of Mr.
Jeffreys.

In addition to its extent, the list exhibits several points of interest,
and, in certain characteristic respects, points of novelty in comparison
with any hitherto published lists from deposits of similar age in
Lancashire or England. These are noted in the sequel.

Except where otherwise marked, all the species are represented by
small much-worn fragments; but the occurrence of whole and of
comparatively little-worn shells is much more frequent than in the
gravel at Blackpool or the Liverpool clays.

The assemblage of species as a whole most nearly represents that
of the Wexford gravels, of which it may not be too wild to consider
this deposit a Lancashire representation, if not extension. The
present occurrence of the larger univalves (Aporrhais, Purpura,
Murex, Nassa, and Buccinum undatum) comparatively little worn
may perhaps indicate that this bed is one of deposit at a time when
the shells found in it were living at no great distance, and probably
on or near a sea-beach or at the bottom of some littoral current.
In this respect the facies of the collection differs from that of similar
representations from the Blackpool or Liverpool beds, and resembles
rather those from Moel Tryfaen and Macclesfield. One might almost
conjecture that the Blackpool and Liverpool fossils were actually
those of some Leyland or Macclesfield shores, further worn and
widely redistributed by the waves and currents of a retreating
ocean.

The list, although containing a few names of species of a northern
character (Astarte and Fusus), contains also several of shells of markedly
southern origin. Venus (Cytherea) chione, which has been identified
at Macclesfield and in several of Mr. Reade’s Liverpool localities,
may almost be called a characteristic fossil at Leyland. Cardium tuber-
culatum (rusticum, F. & H.) occurs not unfrequently. One perfect and
characteristic hinge of Mactra glauca (helvacea, F. & H.) was found.
All these are essentially southern species. The only shell which is
peculiarly arctic in character is Fusus (Trophon) craticulatus, Fabr., a
species now living in Greenland, of which one fine and (for drift) fairly °
fresh-looking shell has been found. The same species was identified
at Moel Tryfaen.

With this last-named exception the series from Leyland must be
described as very similar to that of the fauna of the present seas
along the western shores of Britain. A curious feature of it is
the occurrence in comparative frequency of Fusus antiquus, var.
contrarius, a Wexford fossil and common in the Crag. Another is
the similar occurrence of limestone pebbles burrowed by Saaicava
rugosa, and in some cases still enclosing the two valves of the
miner.

In the following Table v. r. means that 1 to 3 specimens have
occurred; r.3 to 10; f., frequent; ¢c., common; and a., abundant.

40

R. D. DARBISHIRE ON A DEPOSIT OF

4)
2

SOOO SS OU COINS) ie

10.

Species.

Pholas crispata.......--..2..0---

Saxicava rugoSa.........2--..see
NOLVELICA......ee0e ee eee ene
Solen siliqua ..........00eeeeeee-
Mya truncata... ....0..-.eseeeee
Lutraria elliptica ..........see0.
Mactra glauca .........0+...8 D8
solida vtreeenaes
Psammobia ferroensis .........
Tellina balthica ...............

Tapes pullastra .........-+2..0...
Venus Casina,........-..-0eceeeees
chione .......

gallina, var. striatula ...
Astarte arctica ..........2.2..+..
sulcata, var. elliptica ...
SULEENEY | Sonnooconsdedoconcads
Cyprina islandica...............

—— tuberculatum ............
== GIMATEATIIN ~ Gousooosdoose
Pectunculus glycymeris ......
Mytilus modiolus...............
Pecten opercularis ............
Ostrea edulis ..................
Dentalium entalis ............
Littorina litorea ...............
Obtusatale eee

Natica catena ..,,........20...-
Aporrhais pes-pelecani.........
Purpura lapillus ...............

Buccinum undatum ............
Murex erinaceus ...... ese

Trophon truncatus ............
craticulatus ...............
Fusus antiquus..................
, var. contrarius ...
Sa (HUEVOS a ashoadaapdantonodoc
propinquus ..............-
Nassa reticulata,.................

CUO inicio

eoeeeeeee

Balanus porcatus ............ ..
Cliona

Bete w meee cee esses cere eens

Occur-
rence.

v. Yr.

| Very abundant, large and small.

Remarks.

(Panopea).

One characteristic hinge.

Valves and fragments very thick.

(T. solidula). Some valves ex-
tremely thickened.

Doubtful fragments.

Small fragments.

(Cytherea). Many hinges and
other characteristic fragments.

One whole valve occurred.
Many hinges and characteristic
fragments.

Characteristic fragments.

Whole and fragments.
an . The shells
exhibiting a crag-like variation
in form.

A few comparatively unbroken
shells occurred.

Very fine and little-worn shells
occurred.

One fine shell.
\ About equally frequent. Shells
and fragments.

Many whole shells, with much
variation in form.

Large.

In shells and fragments.

MIDDLE PLEISTOCENE GRAVEL IN LANCASHIRE. 41

Discussion.

Mr. DarsisHireE was not prepared to accept the view of the shells
in the Drift having existed on the spots where now found. He
thought rather that the fragmentary remains might have been de-
rived from the destruction of earlier beds deposited under somewhat
different conditions. The occurrence at Wexford of nearly similar
beds to those at Leyland pointed to a great destruction of an old
sea-shore.

Mr. Gwyn Jerrreys thought that all the shells found in these
Lancashire beds were just such as might have been thrown up on
the shore, though the matrix in which some of them are found is
not sandy. The Zvrophon discovered was 7’. truncatus, and not
T. clathratus. Neither was he quite satisfied that Miss Ffarington’s
Fusus (or rather Trophon) was really 7’. craticulatus. The occurrence
of Fusus antiquus, monstr. contrarius, did not surprise him, though
that of Mactra glauca was very remarkable. He did not believe in
the retiring or voluntary migration of Mollusca, though they might be
transported by currents or driven away by want of food. He did
not regard any of the shells as truly Arctic, and doubted whether
any of them afforded clear evidence of climatal conditions.

Mr. Prestwicn remarked on the progress which had been made
in our knowledge of these shells since Sir P. Egerton had first called
attention to the Drift in which they occur. The number of perfect
specimens from Leyland was, he thought, very striking. He had
some difficulty in following Mr. Reade into the large theoretical
questions into which he had entered, but pointed out that the stria-
tion of the surface of the country was significant of a period of
intense cold, which any alteration in the arrangement and pro-
portions of land and water could hardly account. But in the over-
lying Boulder-clay the fragments of shells were all of species still
existing in the neighbouring seas of the present day; and he did
not think that at the time of its deposit the climate was of necessity
intensely cold.

Prof. Hucues did not think that the deposits were in any way
immediately connected with the Boulder-clay, to which they were
long subsequent. He rather correlated them with the Hessle and
Kelsey beds of the east coast. The deposits might in many cases
have been formed on the shore of a sea which was eroding a cliff of
Boulder-clay ; and by this means there would be an admixture of
the more recent shells with the redeposited boulders from the older
clay. He submitted that the shells belonged to an age succeeding
the true Glacial period. In the higher deposits there were still some
traces of the more Arctic forms, while a more southern facies came
over the fauna of the lower beds.

Mr. Cuartesworte observed on the possibility of the transport of
shells in the stomachs of fishes. As to the comminuted condition of
Cyprina islandica, he remarked that in the Crag beds these shells
are nearly always much cracked, even when delicate shells in the
neighbourhood are perfect.

42 ON A MIDDLE PLEISTOCENE GRAVEL IN LANCASHIRE.

Prof. Ramsay was glad that the old view as to the successive
elevations and submergences during the Glacial period was not
likely to be disturbed. As to the physical causes which conduced
to the extreme cold, he did not undervalue the changes in physical
geography; but if the astronomical causes, the existence of which
seemed now to be fairly established, would have produced the effects,
he did not see why they should be ignored even if the geographical
causes might suffice. These latter seemed to be at best theoretical,
whereas the former seemed mathematically necessary. He was not
inclined to detach the shells from the clay, and thought that during
the time of their deposit there were still glaciers on the higher
parts of the land. He did not agree with Prof. Hughes in regard-
ing the beds with striated pebbles in the Vale of Clwyd as post-
glacial, and could not believe that in the case of the reconstruction
of the beds the strize could be preserved and the pebbles not be-
come smooth.

Mr. Reap, in reply, stated that his observations were intended
to apply merely to the conditions under which the beds containing
the shells had been deposited, and not to the period of extreme cold,
for which he was quite willing to admit the potency of astronomical
causes. He agreed with Prof. Ramsay in regarding the clay as a
real Boulder-clay, the pebbles in it being for the most part scratched.

H. G. FORDHAM ON STRUCTURE DEVELOPED IN CHALK. 43

8. Norms on the Srrucrure sometimes developed in Cuatx. By
H. Guorce Forpuam, Esq., F.G.S. (Read December 3, 1873.)

A papER on this subject (the peculiar parallel striz found in
Chalk), read before the Society last May, brought up a discussion
in which such a great variety of opinion as to the origin of this
structure was shown, that I thought the following few facts might
perhaps be of interest.

In a pit one mile east of Ashwell, on the borders of Hertfordshire
and Cambridgeshire, a small section of the ‘“‘Chalk without flints” of
the Geological Survey, about 40 or 50 feet above its base, is exposed.
The chalk here is hard, and may be divided into layers, which differ
considerably in structure. There are at least three varieties :—(1)
layers of hard chalk, divided into slabs about an inch thick; (2) beds
much thicker and more massive than the last; and (3) a bed of a
concretionary nature. These beds are from six inches to several
feet in thickness.

The concretions are marked nearly all over by the lines of
this structure. I find that the lines are only found on the con-
eretions (which are easily removed from the parent rock) and in
their very immediate neighbourhood. The other beds are entirely
free from any trace of the structure. .

The chalk at this particular spot has not been very much faulted,
and the layers can be traced without change for some little distance;
it is hard; and the fossils in it, which at this place are compa-
ratively numerous in the concretionary bed, are invariably crushed,
as if by pressure: the structure seems to have no connexion with
them.

These facts seem to point to a relationship between the concre-
tions and this structure. I believe the peculiar strie in this case
are due to an incipient crystallization arising from the formation of
the concretions. In support of the view that they are due to crys-
tallization, I exhibit a specimen of iron pyrites from the Chalk
of Beachy Head, attached to which is a small portion of chalk, very
hard and striated. Now the chalk there is in very Jarge masses, as
much as 6 feet each way, and does not appear to be in any way
anomalous in structure. It would seem that in this specimen the
crystallization of the pyrites had induced a crystallization in the
chalk.

I believe that in some places, however, an almost identical
structure is due to slickensides, but only in very broken and faulted
strata.

Discussion.

Mr. Sretery observed that the structure was familiar to all. If
it were due to crystallization, whether incipient or otherwise, he

44 H. G. FORDHAM ON STRUCTURE DEVELOPED IN CHALK.

wished to know to what combination with lime for a base the
form of the crystals was due. He thought that a certain amount
of phosphate of lime was present in the concretions, but was absent
in similar specimens in the Upper Chalk; so that it appeared as if
the same cause could not apply in both cases. The striee were not,
he thought, due to slipping or to organic growth, but might arise
from some alteration in the character of the chalk.

Mr. Evans observed that the strie appeared to be due to two
causes—crystallization, whether incipient or destroyed, and slicken-
sides. He thought that in the Ashwell specimens much was owing
to the hard nodules resisting pressure better than the surrounding
chalk, which, in being condensed, passed over their surfaces and
produced a kind of slickensides. The same appeared to have been
the case with the nodule of pyrites.

Mr. Forsrs remarked that an inspection of the specimens on the
table convinced him that in several instances the structure was due
to slickensides, but that in others traces of a very different structure
were visible, which he imagined was due to crystallization, the car-
bonate of lime having most probably assumed the form of Aragonite,
which, owing to its instability, had lost its crystalline lustre and
assumed a mealy or chalky appearance.

Mr. Jupp thought that the difference between the conclusions of
the two papers communicated to the Society upon this subject was
mainly due to the difference between the Chalk of Yorkshire and
that of the south of England. The abundance of this structure in
Yorkshire had caused Mr. Mortimer to connect it with some
organism concerned in the formation of the Chalk; while its rarity
in the south had led Mr. Fordham to assign another and a chemical
cause. He thought that it became apparent only in those parts of
‘the Chalk through which water most readily passed, and considered
that in some cases the crystallization had been that of Aragonite,
in others that of Calcite. He commented on the dimorphism of
carbonate of lime, which under slightly different conditions assumed
different crystalline forms ; and remarked that, as was the case with
many igneous rocks, the structure in the Chalk only became apparent
after weathering.

Mr. Mryrr remarked that the striae were most common where
nodules were present, and in beds in which the fossils were crushed:
So far as he had observed, they were always vertical; and he attri-
buted them to a re-arrangement of the particles of the Chalk
under pressure.

Mr. Forpuam explained that two of the specimens had been taken
from the sides of a fissure, and were actually slickensides. He
thought that some of the nodules exhibit traces of internal struc-
ture; but it was true that the general direction of the striz was at
right angles to the lines of bedding. He was glad that his remarks
had led to so interesting a discussion.

Mr. Prestwicu suggested that an analysis should be made of the
specimens,

A. W. EDGELL ON BUDLEIGH-SALTERION LAMELLIBRANCHS. 45,

9. Notes on some LAmELLIBRANCHS of the BupLEeteH-SAaLTERTON
Pessies. By Arraur Wyatt Enpcetr, Esq., F.G.S. (Read
January 7, 1874.)

[Plates IV.-VI.]

Ir will be recollected that in the Quarterly Journal of the Geologi-
cal Society, vol. xx. p. 283, there is a paper by Mr. Vicary, of
Exeter, drawing attention to the pebble-bed of Budleigh-Salterton ;
appended to this is a description of some of the included fossils by
the late Mr. Salter. Since this a paper on the Brachiopoda from
the pebbles has been given (in vol. xxvi. p. 70) by Mr. Davidson. In
the former paper a full description of the bed is given by Mr. Vicary,
to which there is little or nothing to add. It is sufficient to say that
it consists of a bed of nearly 100 feet of pebbles of various rocks,
inclined at a low angle to the horizon, lying on and covered by the
Trias, that the pebbles contain fossils, as a rule, unknown elsewhere
in England, that the fossils in some are Silurian and in others
Devonian, and that the matrix including Silurian and Devonian
fossils is often identical. Through the kindness of Mr. Vicary I
have been allowed to figure some of the more remarkable Lamelli-
branchs in his collection which were undescribed by Mr. Salter (or
of which the descriptions and figures have been lost); and I have
added a few of my own. This task would not have been attempted
had there been any chance of a more competent person’s under-
taking it. The figures annexed have been submitted to M. Lebes-
conte, of Rennes, who has a large collection of Norman and Breton
fossils ; he recognizes several of them. M. Bayan, of the Ecole des
Mines at Paris, has also seen them, and compared them with the
collections of MM. d’Orbigny and de Verneuil ; they have also been
shown to M. Gaston de Tromelin, who is working at the Norman and
Breton rocks.

In July last I visited M. de Tromelin at Argentan, and was much
struck by the general resemblance between the fossils and rock-
specimens in his collection to those in that of Mr. Vicary. The
first question I asked was, ‘‘ whether Orthis reduaw was by far the
most abundant fossil,” which is notably the case in Devon; the
answer was that no other shell was nearly so frequent ; and this
appears to be worth noting. M. de Tromelin obligingly took me to
see some beds of quartzite near Trun (about six miles from Argen-
tan) which are used for road-metal, they are believed to be the
“Grés Armoricain.” It was easy to recognize in these beds the
material of one of our commonest Budleigh pebbles, which is
whitish hard quartzite, with a profusion of Trachyderma traver-
sing it. M. de Tromelin says he has found no other fossil in it.
Mr. Vicary tells me he has never found any thing in the same pebble
with Trachyderma ; nor have I. These beds of quartzite are nearly
vertical, as far as can be observed, the Zrachyderma-tubes being

46 A. W. EDGELL ON BUDLEIGH-SALTERTON LAMELLIBRANCHS,

almost horizontal ; the bedding is otherwise obscure. The strike
appears to be about N.W. and 8.E.

I cannot help thinking that a closer acquaintance with the beds
of Normandy and Brittany will assist persons acquainted with the
Budleigh-Salterton bed to refer each pebble to its proper origin,
though the resemblance between the Silurian and Devonian quart-
zites of France in some cases is very great indeed. The figures
given here are all of the natural size*.

Moproropsis ARMoricrI (Salter). Pl. IV. figs. 1,1a. (Mr. Vicary’s
collection.)

Syn. Modiolopsis armoricana; M. armorica, Bigsby, ‘Thesaurus
Siluricus ;’ Avicula prima, D’Orbigny, ‘ Prodrome.’

This figure has been recognized by all the three above-named
gentlemen; and M. Bayan found a specimen in M. de Verneuil’s
collection labelled “ Modiolopsis armoricana, Salter. Avicula prima,
D’Orb.” The figure given by Salter, Q.J.G. 8. vol. xx. pl. 16. fig. 1,
is, as I believe, the interior cast of this ; such interior casts are com-
mon and variable in form. I have figured one from my collection,
Pl. V. fig. 6. As it is clearly not an Avicula, I have kept Salter’s
name.

Moptororsis Lrprsconti, sp.nov. Pl. LV. figs. 2, 2a. (Mr. Vicary’s
collection.)

The outline of this, which I take to be an interior cast, differs
from that of any other in Mr. Vicary’s collection. The cast is ovate,
rather gibbous at about one third from the hinge; and the gibbosity
decreases more rapidly towards the hinge-line than towards the
yentral margin. The pallial impression is visible in places in a
strong light. Length 2 in. 7} lines; breadth (from hinge-line to
ventral margin) | in. 7 lines; thickness 4 lines. I have ventured to
name this after M. Lebesconte, of Rennes, as I am informed that,
though known in May, it has received no name.

SANGUINOLITES ? sp. (conTortts, Salter?) Pl. IV. figs.3, 3a. (Mr.
Vicary’s collection.)

This appears to be crushed; but Mr. Vicary informs me that it
was not considered by Mr. Salter to be so. The left valve is convex
along the hinge-line, while the right valve, from the beak to the anal
angle, is concave. The beak of the left valve curves over the hinge-
line, which is straight ; it is marked by irregular lines of growth.
Length of hinge-line 1| in. 2 lines; from hinge-line to ventral margin
103 lines. It answers, as well as can be observed, to the genus it is
here referred to ; it bears a label ‘allied to Sangquinolites contortus,”
which I believe to be in Mr. Salter’s writing. This form is recog-
nized by M. Lebesconte as French.

* Since the above rematks were written, I hear from M. de Tromelin that the

pebbles of Bayeux and Carentan are fossiliferous, Orthis redux being common
in them, ‘They occur in the Trias, as do the Budleigh pebbles.

A. W. EDGELL ON BUDLEIGH-SALTERTON LAMELLIBRANCHS. AT

AVICULOPECTEN TROMELINI, sp. noy. Pl. V. fig. 1.

This is the exterior of a shell in my collection ; the broken ear is
much better seen on the other half of the pebble, and shows the
shell to have been almost if not quite equilateral. The radiating
ribs are well developed, and alternate towards the centre with
smaller ones; these latter are indistinct and irregular towards the
sides. Measurement from beak to ventral margin 1} inch; breadth
across the shell 1 in. 7 lines.

PreringEa, sp.? Pl. V. figs. 2, 2a. (Mr. Vicary’s collection.)
(Compare P. lineatula.)

The surface of this is not well enough preserved to be sure of it,
I have a specimen from Gahard, which may be the same. M.
Lebesconte recognizes it.

Preriyga, sp. Pl. V. fig. 3. (Mz. Vicary’s collection.)

This is a very globose inflated form, with an acute wing on the anal
side of the beak and concentric lines of growth ; length from point of
wing to beak rather more than 6 lines. Mr. Vicary has other larger
specimens.

PreRINEA RETROFLEXA (Hisinger). Pl. V. fig. 4. (Mr. Vicary’s
collection.)

This is so like some forms of this very variable shell that I can
give it no other name. Length from buccal to anal angle 1 in.
5 lines ; from hinge-line to ventral margin 84 lines. It is recog-
nized by M. Lebesconte.

Patmarca, sp. Pl. V. figs. 5, 5a. (Mr. Vicary’s collection.)

Two specimens are figured which differ from the Palcarca secunda
of Mr. Salter ; the angle made by the anal slope with the hinge-
line is much more obtuse, and the hinge-line longer in proportion to
the shell; the teeth also appear to differ. It is recognized by M.
Lebesconte.

Avicuta, sp. Pl. VI. fig. 1. (Mr. Vicary’s collection.)
A distinct form, but very impertect.

CrrmpopHorus? PI. VI. figs. 2, 2a.

M. de Tromelin thinks this may be a Cleidophorus; the buccal
muscular impression is doubtful. It is in my collection.

Prerinea, sp. Pl. VI. fig. 3. (Mr. Vicary’s collection.)

The cast of an exterior from Gahard is before me exactly re-
sembling this in outline; M. Lebesconte recognizes it. M. Bayan
thinks it may be Aviculu matutinalis, D’'Orb. Pl. VI. fig. 5 also
represents a specimen of this species.

LunvuLacaRpium ventRIcosum. PI. VI. figs. 4, 4a. (Mr. Vicary’s
collection. )

Very like M. Barrande’s genus Stlurina ; the beak is slightly in-

1

48 A. W. EDGELL ON BUDLELGH-SALTERTON LAMELLIBRANCHS.

clined to the left; it has radiating ribs; from beak to ventral
margin 1 in. 34 lines ; thickness of the valve shown 4 lines. Named
by Mr. Etheridge.

Crenoponta, sp. Pl. VI. figs. 6, 6a, 66, 6c. (Mr. Vicary’s collec-
tion.)

Very inflated; the incurved beaks are nearly a line apart, hinge-
line short, with four teeth visible in each valve, muscular scars on
both sides of the beak strongly defined, pallial impression visible on
the buccal half. Recognized by M. Lebesconte. From beak to
ventral margin nearly 9 lines; length i in. 44 lines; thickness of
both valves nearly 8 lines.

Orrmonora? Pl. VI. fig. 7. (Mr. Vicary’s collection.)

The grain of the matrix is too rough to make much of this ; it is
very like some specimens in Jermyn Street from the Upper Silurian
beds. M. de Tromelin recognizes it.

; EXPLANATION OF PLATHS IYV.-VI.
Shells from the Budleigh-Salterton pebbles, natural size.

Prats IV.
Hee: 1, la. Modiopsis armorici (Salzer).

2, 2a. Modiolopsis Lebesconti, sp. nov.
3, 3a. Sanguinolites ? (contortus, Salter ?).

Puate V.

Fig. 1. Aviculopecten Tromelini, sp. nov.
2. Pterinea, sp. (lineatula?); 2a. portion of surface enlarged.
3. Pterinea, sp.
4, Pterinea retroflexa (Hisinger).
5, 5a. Palearca, sp.
6, 6a. Modiolopsis armorici, internal cast (?).

Puate VI.

Fig. 1. Avicula, sp.
2, 2a. Cleidophorus?
3. Pterinea, sp.
4, 4a. Lunulacardium ventricosum.
5. Pterinea, sp.
6, 6a, 66, 6c. Ctenodonta, sp.
7. Orthonota ?

Discussron.

Prof. Ramsay commented on the value of the paper. The dis-
tance of the parent rocks, the rolling of the pebbles, and their travel-
ling from south to north, threw considerable light on the physical
condition of the Triassic period. If the New Red Sandstone were,
as was now generally supposed, of lake origin, the information given
was of still higher interest.

Madlare & Macdonald Lith to *he \oeen London

New Autogruphic

BUDLEIGH-SALTERTON FOSSILS.

Quart.Journ. Geol Soc. Val. XXX.PLV.

2
=
es 2
£
A ;
Fi ]
£ i
1 s
i i
Ye 3
ee

Maclore & Macdonald. Tith tothe Qneen London
New Autnfraphie Proesas

Quart. Journ. Geol.Soc. Vol XXX. PLV1.

a

Wodlore & Macdonald Tath ta the uses Lond
Now Antosraphir |'rner

BUDLEIGH-SALTEREON FOSSILS.

A. W. EDGELL ON BUDLEIGH-SALTERTON LAMELLIBRANCHS. 49

Mr. Gopwin-Austen observed that one of the remarkable features
of the Budleigh-Salterton-beds was the presence of fine sandy
deposits, both above and below, which at once suggested a difficult
question as to what could have been the condition under which
they were deposited. He thought that they might be connected
with some glacial action, especially as blocks of porphyry, 25 tons
in weight, had been transported from the neighbourhood of Exeter
and deposited in the Triassic beds, which could hardly have been
effected otherwise than by ice. In the same manner as the shingle
of Lake Superior is carried away and redeposited by shore-ice, so he
thought it possible that some action of the same kind might during
a portion of the New Red Sandstone period have drifted materials
off from the French shore of the Triassic lake, and deposited them
in this shingle-bed at Budleigh-Salterton. In conclusion, he alluded
to the loss which the Society had sustained in the death of Professor
Louis Agassiz, its distinguished foreign member, to whom the now
generally accepted term roches moutonnées and other evidence of
glacial action, so often mentioned in the course of the evening, was
due.

Mr. Warraxer pointed out that in the slaty rocks at Mevagissey
in Southern Cornwall, there were quartz reefs of similar material to
the material of the pebbles at Budleigh-Salterton, and recommended
a careful examination of these reefs before accepting the Breton origin
of the pebbles as conclusively established. The deposit at Budleigh
Salterton was, he believed, a lenticular mass of no great extent.

Mr. Erneriper remarked that Mr. Salter had long ago been of
opinion that the fossils in the pebbles were of French origin.
Mr. Tawney, also, had examined the beds near Torbay and Babba-
combe, and found the lithological character curiously like that of
the pebbles. He did not, however, think that any of the same
fossils had been found in them; nor had they been found in the
Mevagissey beds in Cornwall.

Mr. Hicxs thought that we were going too far in search of the
original home of the fossils, which might have been in some rocks
destroyed during the formation of the Channel.

Q. J. G.S. No. 117. 2

5” A. W. SLIFFE ON THE MUD-CRATERS AND

10. On the Mup-crarers and GrotoercaL Srructure of the Muxran
Coast. By Lieut. A. W. Srirrs, late Indian Navy, F.R.A.S.
(Read December 3, 1873.)

{Communicated by Prof. Ramsay, F.R.S8., V.P.G.S.]

Tue coast of Mekran (the ancient Gedrosia) extends in an east and
west direction for about 500 miles, between the town of Sonmiani,
on the Pir Ali river near the frontier of Hindustan, and Ras Jashak,
near the entrance point of the Persian Gulf (fig. 1

It is an almost rainless district, but occasionally subject to heavy
storms of rain not of long duration, and generally occurring in the
winter months. Sometimes no rain falls during the whole year, or
even for two years successively. There are no rivers of any size;
but the large watercourses after heavy rain discharge large volumes
of water.

The appearance of the coast is singular, and the hills near the sea
are all of similar formation—table hills with abrupt, almost perpen-
dicular sides, fantastic peaks, pillars, and pinnacles rising out of ex-
tensive plains of clay being the universal type.

These clay plains are irregularly veined in places with crystalline
gypsum, and are impregnated with saliferous matter, which efflor— |
esces on the surface. It need hardly be added that nearly the whole
country near the sea is a desert.

The hills m the vicinity of the coast are of various heights above
the plain, and rise above 2000 feet in some places: they are chiefly
_ composed of beds of clay from 50 to 1500 feet in thickness above ~
the sea-level, capped and sometimes alternating with coarse,
friable, fossiliferous calcareous strata from 5 to 30 feet in thickness,
the whole either horizontal or dipping at a small angle. Beyond
Jashak to the westward, and Ras Kicheri to the eastward, the form-
ation appears similar; but the strata are highly inclined, probably at
an angle of from 40° to 60°, the ridges of the hills being separated
by great anticlinal valleys.

Fossils have not been observed in the clay; but the calcareous
beds are composed entirely of marine organic remains, in parts dis-
tinct and perfect, forming a shelly breccia, and passing into a more
or less compact hmestone.

These beds, as indicated by the fossils, are ‘referred by Mr.

Ktheridge to the Miocene epoch. He says :—

«The shells appear to me to be of Miocene age, although all are
mere casts, and therefore it is very unsafe to pronounce definitely
as to their age. I cannot determine them to belong to the modern
species of the Persian Gulf. The casts of Conus, Cyprea, and
Ostrea are certainly not those of existing species ; and the mass of
shelly breccia containing Dentalium, Cardita, Astarte, Venus, Trochus,

Arca, and Telling I believe to be of more modern date, if not receut.

‘The unctuous argillaceous deposit at Ras Farsah upon which the

GEOLOGICAL STRUCTURE OF THE MEKRAN COAST. 5)

terrace of shelly beds lies, is bored by the living Lithodomus allied
to the Indo-Pacific LZ. attenuatus. They appear to be vastly abun-
dant in the clay, which is riddled by the perforations in all directions
and at all angles.

‘The genera of shells occurring in the shelly layer, believed to
be of Miocene age, are Hburna, Ostrea, Cyprea, Conus, and Cardium ;
and all occur in the form of casts.”

The surface of these beds is in some places strewed with pebbles,
much water-worn, composed of jasper and quartz, &c., and possibly
of trappean origin.

With the possible exception of the craters to be shortly referred
to, the writer has observed no traces of volcanic action on this coast.
Within the Persian Gulf, however, a similar formation, which ex-
tends along the entire north coast, and forms all the large islands, has
been much disturbed by the protrusion of recent volcanic matter.

Near Jashak, at the western end of the coast, is found a hot spring,
ef a temperature of 128°, situated near high-water mark, and but
little above the sea-ievel. It is unimportant in volume, and has
six or seven little surface-basins close together. The water emits
a strong odour of sulphur, and forms a stalagmitic deposit round its
margin*, There are also springs of pure hot water near Karachi
beyond the eastern boundary of the region.

As might be expected from the nature of the formation, a vast
amount of denudation has taken place, and is still in progress, which
would be much accelerated but for the almost rainless nature of the
country. The clay being gradually removed by rain, when it occurs,
or in the sea-cliffs by the action of the waves, the upper stratum
falls in great masses or slabs, and forms a talus, which in some
degree protects the base of the cliffs from further disintegration.

On the inland faces of the hills, ravines and precipitous valleys
are formed in the same manner; and the surface-crust of the ground,
eracked and broken in every direction, often presents an appearance
such as might be the result of an earthquake.

A typical section at Guadur is given in the diagram (fig. 2): the
immense space between the cliffs has, it is presumed, been removed
by denudation. The several projecting headlands, and the island
of Astola, apparently all belong to the same group, and stand as
remnants showing the original extent of the formation.

In some parts, as at the great headland called the Malan, large
masses of the clay are constantly falling with a dull roar, breaking
up into clouds of dust. From one cliff the writer has watched a
succession of such landslips for an hour with hardly any inter-
mission.

The mud-craters, which have hitherto been attributed to volcanic
action, are, it is believed, peculiar to this coast. They are found
over an extent of about 200 miles of coast from Guadur to Ras
Kicheri. Probably the most characteristic group is that near the
latter place, where there are a number close together, some of which
may be termed extinct, others being in action. These craters vary
* It is regretted that the specimen for analysis has been mislaid.

E2

o2 A. W. STIFFE ON THE MUD-CRATERS AND

in height from 20 to 300 or 400 feet above the plain ; and a probable
section is shown in the diagram (fig. 3). They rise out of the clay
plain many miles from any hills, and are cones of clay of very regular
form, with truncated tops, and sides at an angle estimated at about 40°,
or whatever the limiting angle or angle of rest of such mud may be.

The largest one ascended by the writer was about 100 feet
wide on top, and nearly circular; it resembled a cup filled to the
brim with semiliquid mud somewhat thicker than treacle, which
slowly and only now and again overflowed, trickling down the outside
of the cone, but scarcely sufficient in quantity to reach the base.
From time to time an ebullition of gas or air took place from the
surface of the pool.

Mr. Ward has kindly analyzed the mud ejected, and found it to
consist chiefly of clay with a large admixture of carbonate of lime
and a small proportion of quartz sand. The aqueous part contains
much chlorine, a little sulphuric acid, a little lime, with soda and
a trace of potash. No magnesia was detected in the solution. It
would therefore probably be merely a water containing common
salt with a little sulphate of lime.

The edge or lip of the cup was very narrow, in parts barely
allowing room to walk round it, and being quite soft under foot. A
plummet on being thrown in from the side sank rapidly from 50 to 60
feet, when it stopped, apparently from the friction of the line
against the side, there being no means at hand of lowering it over
the centre. There is no heat accompanying the phenomenon; and
there is no reasonable doubt that the walls of the cones are formed
entirely by the accumulated overflow and induration of mud.

It is therefore a question whether it is necessary to resort to vol-
-canic agency, or whether hydrostatic pressure alone is not quite
sufficient to account for these mud-springs, as they may perhaps be
more appropriately called.

The thickness of the clay deposit below the level of the plains is
unknown; the veins of gypsum seem to indicate it is not merely
superficial, or derived fram the denudation of the hills. From the
form of the sea-bottom it is possible that the same clay beds which
form the hills attain a considerable thickness below the present sea-
level. The sea-bottom for some miles from the coast is of similar
nature and appearance to the clay of the plains, with a depth of water
increasing gradually to 20 or 30 fathoms and then sinking suddenly
and (in several places, at least) quite precipitously to a depth of from
300 to 400 fathoms.

There is some evidence of the existence of these curious craters
under the sea. Near Jashak I discovered a shoal, three miles off
shore, rising suddenly from 13 fathoms to a height of only 10 feet
below the surface. It is very small, not a ship’s length across, and
composed of clay which, on the shoal part, was so tenacious that the
lead could hardly be disengaged when let go on it. It is unusual
to find so small a shoal unless of rock; and it is suggested that it may
be probably of similar origin to the craters on shore.

‘The natives state that the action of these craters (the amount of

[To face p. 52.]

67

ge
opllekran Coast.

/ i |

ymarine cliff.

horizontal.

are known to exist.

leseribed.

Fig. 1.—Sketch Map e Mekran Coast. ; [Zo face p. 32]
58 60 61 62 63 G4 65 66 oy

s

!

Approximate line of submarine cliff.

= Strata approaching the horizontal.
Bes Areas in which craters sre known to exist.

Highly inclined strata.

Fig. 3.—Section of Crater called Chundra Koop.
Position of hot-springs.
@---- Supposed submarine crater.
be Line of section in fig. 2.
a. Dry mud. 6, Liquid mud.
Fig. 2.—Section across Guadur and Jebel Mehdi. - ee

Geographical Miles

Datum line

__ Sea level : 5 a ee eee

|
Y
1

‘The natives state that the action of these craters (the amour

GEOLOGICAL STRUCTURE OF THE MEKRAN COAST. 53

ebullition and discharge) is much increased at spring tides; and this
the writer has heard from several different sources.

None have been heard of more than a few miles from the shore, so

- far as the writer is aware.

As to the causes of the present configuration of the coast, it is
suggested that since the deposit of the Miocene beds the great sub-
marine cliff may at one period have been raised above the sea,
which then washed its base—that the coast then subsided, probably
to near the present level, when the beds were removed from the
edge of the great cliff to the present coast-line—and that the coast
may have been depressed below its present level sufficiently to allow
the sea to produce the present inland cliffs, since which it has been
again slightly raised.

The presence of Jithodomus in the clay considerably above the
present sea-level seems to warrant, at any rate, the latter hypothesis.

The author’s time when visiting the places on the coast has been
much taken up with his duties connected with the telegraph cable,
which is his excuse for the incompleteness of the information and of
the series of fossils.

Discussion.

Mr. Evans thought that the fact of the mud from some of the
cones having ceased to flow was in favour of the theory of their
being due to hydrostatic action, the outflow ceasing when the pres-
sure was insufficient to overcome the resistance, and a fresh vent
being found elsewhere.

Mr. Forsss had frequently observed the mud-craters of Peru,
which were undoubtedly due to volcanic action, though in many
instances the outflow was cold, or nearly so. In the cases described
by the author, however, it was possible that they might be due to
another cause.

Prof. Puitires mentioned a spring which rose through mud in
Bridlington Harbour, from which the outflow varied both with tide
and rain. The spring tides in the Persian Gulf might effect a rise
in the mud-craters by obstructing some passage by which water from
the upland was discharged into the sea. Were this so, the upward
movement in the crater would on each occasion probably occur some-
what later than the high tides.

Mr. Sretry remarked on the presence of undoubted volcanic action
in the district, in the shape of hot springs &e., and therefore ques-
tioned the propriety of bringing in hydrostatic pressure to account
for the phenomena.

Mr. Mrzter cited the mud-volcanoes of Iceland as instances of
pure volcanic action, and drew a distinction between the occurrence
of mud-craters near the sea and at a distance from it.

The AvrHor, in reply, observed that he had merely suggested the
possibility of the outflow being due to hydrostatic pressure, and did
not insist upon it. He stated that there was ample evidence of vol-
canic action at a distance of 200 or 300 miles from the district he
had described.

54 E. J. DUNN ON THE MODE OF OCCURRENCE

11. On the Move of Occurrence of Diamonps in Soutm Arnica.
By E. J. Dunn, Esq. (Read December 17, 1873.)

{Communicated by Professor Ramsay, F.R.S., V.P.G.S.]

Tur conditions under which diamonds occur in South Africa are
quite different from those of every other known locality, and are so
unusual as to deserve the earnest attention of all geologists.

It is quite certain that the present ‘“‘dry diggings,’ such as
“‘ Colesberg Kopje,” “ Du Toit’s Pan,” “ De Beer’s,” “ Bultfontein,”
and “ Jagersfontein,” are being worked in the rock which has brought
the diamonds to the present surface; they all possess certain com-
mon features of very marked character. Each area is more or less
circular in form (fig. 1, a); the boundary of each is usually horizontal,
or nearly horizontal, shale. At the junction, and back for a distance
of from one to several feet, the edges of the shale are bent sharply
upwards (fig. 1, 6).

The contents of these “ pipes” in the shale are the same in all
cases, and show distinctly that they are of igneous origin. The base
is more or less decomposed gabbro (?) or euphotide (?), through which
are scattered particles, fragments, and huge masses of shale, nodules
of dolerite, occasional fragments of chloritic schist, micaceous schist,
and gneiss. The principal foreign ingredient is the shale, which in
many places, particularly at Colesberg Kopje, is thoroughly commi-
nuted, forming a breccia, with euphotide (?) as a base. Where large
masses of shale occur, the lines of bedding, as might be expected, are
not horizontal, but lie in all directions.

From the forms, contents, and general features of the pipes, it is
reasonable to infer that they are merely the channels that con-
nected ancient volcanic craters with deep-seated reservoirs of molten
rock.

On sinking into these pipes, the following sequence, with slight
variations, occurs :—First, from a few inches to several feet of red
sand, or sandy soil, brought into its present position by the agency
of wind. Even within the last 20 years portions of the neigh-
bouring country have been stripped of soil down to the bed-rock,
and other portions covered by such means.

Then a layer of tufaceous lime, from a few inches 9 8 or 10 feet
thick, generally much harder and purer on the surface, shading
gradually underneath into the decomposed rock from the altered
pyroxenic minerals of which it has doubtless been formed. The

segregation of the lime to the surface would be materially assisted

by climatic conditions; for in this subtropical region rain usually
falls in heavy showers during the summer, No sooner is the rain
over than the intense heat causes rapid evaporation.

Lower still is the very friable decomposed euphotide (?), so soft and
crumbling that, for a few feet, it may be almost dug out with a spade.

OF DIAMONDS IN SOUTH AFRIOA. 55

It is of greenish or yellowish colour; a considerable portion of it is
altered into serpentinous material ; gradually, as the depth increases,
it becomes firmer. For a depth of from 30 to 40 feet, cracks,
joints, and irregular cavities filled with red sand from the surface
penetrate; with the sand, and showing that it has come from the
surface, are fragments of ostrich-egg-shell, small rounded grains of
chalcedony, agate, dc., identical with the same substances mixed with
the surface-soil. There are also veins of calc-spar and nodules of
iron pyrites.

At 100 feet in depth, at Colesberg Kopje, the rock is very dark
in colour, much more compact, and less altered than near the
surface. At 130 feet, the greatest depth so far attained, the rock
becomes compact, tough, and shows the original texture, though
the constituents are altered, notably the pyroxene or augite into
bronzite*.

From the surface down, the following minerals are met with
in the altered rock, viz. garnet, calc-spar, mica, bronzite, augite,
diopside, diallage, iron pyrites, &c.

The entangled blocks of shale and sandstone are frequently altered,
the latter sometimes into quartz rock.

Disseminated throughout the decomposed rock, diamonds are met
with, from the enormous size of over 150 carats down to minute ones
only the 100th of a carat in weight. Still smaller ones probably
occur, but are too minute to be observed.

Many of the diamonds are beautifully formed crystals ; but a large
percentage consists of fragments and broken crystals; and it is a
noticeable fact that the corresponding pieces, even when the original
erystal was of large size, are never found, though most carefully
looked for. This would argue very forcibly against the supposition
that the euphotide (?) rock in which the diamonds are now found is
their original matrix or “ mother rock.”

At the same time, it is marvellous that it contains such a rich
sprinkling of glittering gems if not the original matrix. Colesberg
Kopje, for instance, is but 122 acres in area, and has been dug out
to a depth of from 80 to 90 feet, yielding some millions’ worth of
diamonds. Some of the claims have produced immense numbers of
diamonds. One claim, 30 x 30 feet, worked out to a depth of 100 feet,
was recently sold for £4000.

A very well known fact on the diamond-fields, and one rather in
favour of the euphotide (?) being the mother rock, is that each of the
pipes furnishes diamonds easily distinguishable from those found in
the others. Bultfontein produces small white stones, occasionally
specked and flawed, but very rarely coloured; while Du Toit’s Pan,
within half a mile, seldom yields other than coloured stones. So
well marked are the characteristics of the diamonds from the various
diggings, that diamond-buyers can generally tell by the appearance
of a stone the locality it has come from.

Denudation has effected great alterations since the shales were so
completely penetrated by igneous matter; it is therefore not so

* Specimen recently sent by Sir H. Barkly.

56 E. J. DUNN ON THE MODE OF OCCURRENCE

surprising that not a vestige of a crater, or of any of the usual
subaerial products of volcanoes, has been found in §. Africa.

A brief sketch of the various intrusive rocks of S. Africa will help
to explain their relation to the one under notice.

The oldest are the diorite dykes of Namaqualand, running about
east and west, extremely rich in copper ores, frequently very mica-
ceous. They were formed subsequently to the “ Namaqualand Schists”
(pre-Silurian), but do not penetrate the “‘ Table Mountain Sandstone”
(Devonian ?). Penetrating the same rocks, gneiss and metamorphic,
are dykes of granite, syenite, and porphyry, probably of the same
age as the diorites.

The extensive intrusive action by which thousands of square miles
of Dicynodon-shales were penetrated by enormous dykes and lateral
sheets of dolerite, took place at a much more recent period, as the
“Stormberg beds” (containing coal and abundant fossil plant-remains,
such as Sphenopteris elongata, Pecopteris odontopterordes, Cyclopteris
cuneata, &e.) are penetrated by doleritic amygdaloids &c., while the
“‘Sundays-River beds” (Middle or Upper Oolite) are not intersected
by any intrusive rock.

The most recent and, at the same time, most interesting eruptive
rock in 8. Africa is the ‘Trap Conglomerate,” or breccia, apparently
a doleritic rock, which in places becomes a conglomerate on account
of the intermixture of pebbles and boulders of other rocks, while in
other places it is filled with angular fragments forming a friction-
breccia. It extends continuously in a horse-shoe form for several
hundred miles in length, varying from 1 to 3 miles in breadth, and
throwing off many bifurcations. Detached outliers are not uncom-
mon. It penetrates the diamond-field rocks at Pniel, and contains
fragments of amygdaloid entangled in its mass.

It is possible that the diamond-bearing rock filling the pipes
belongs to one or the other of the last-mentioned intrusive out-
flows.

Areas similar in every respect to those worked at the diamond-
fields occur near Schietfontein (see fig. 1), about 200 miles from the
present workings, while Jagersfontein is 70 miles distant in another
direction. Those near Schietfontein have not yet been proved; but
as all similar places at the fields have yielded diamonds, itis probable
that they will also when properly opened.

Near the diggings it is by no means an easy task to find the pipes,
through the accumulation of surface-soil; but at Schietfontein the
boundary can be traced to a foot by means of the upturned edges of
the shale.

The variety of rocks occurring within a very limited area sur-
rounding Pniel, on the Vaal river, is very noticeable, and would
require considerable attention to the details of their occurrence
before satisfactory conclusions could be arrived at. ‘There are
dolerites, diorites, amygdaloids, porphyries, trap-breccia, &c. Near
Du Toit’s Pan the rock, which is of doleritic character, presents very
different texture and colour within a few yards. Along the banks
of the Vaal “ almonds” of quartz that have weathered out of amyg-

OF DIAMONDS IN SOUTH AFRICA. a7

daloid cover the surface of the country, in places, several inches deep.
Where masses of shale have been entangled in the igneous rock, it
has been altered into hornstone.

Diamonds have been distributed from their original position in
the igneous rock by three distinct agencies—water, ice, and the
wind.

To the action of running water the diamond-bearing drifts of the
Vaal river are referable. They are of two dates. The older one,

Fig. 1.—Plan and Section of Pipe on Fisher's Farm, Shietfontein.
(Scale, 10 chains to 1 inch.)

a

5 ———
— i rT

=—Z

Zz lll == \ SSS = 4
il An
SITTIN

a. Plan. 6. Vertical section in line ed.

apparently Postpliocene, occurs at a higher level than the other, as
outliers along the present valley (its present position) indicate that
the Vaal has deepened its course since the old drift was deposited.

The famous Pniel Kopje was of this age, the depth of sinking,
about 30 feet, consisting of a few feet of fine gravel, composed
of agates, chalcedony, amygdaloid, dolerite, &c., with some gar-
nets, imenite, topaz, quartz, and diamonds. Below this, in places,
was a hard layer of conglomerate, the pebbles cemented with lime.
The remainder of the drift was composed of large boulders of igneous
rock, exceedingly tough and hard, with fine drift containing diamonds
filling the interstices. The.bottom is soft igneous rock, probably
dolerite.

At Gong Gong a calcareous conglomerate occurs high above
the recent drift, and appears to belong to the same age as the Pniel
drift. Wherever worked, along the course of the Vaal, diamonds

Q. 0: G. Ss Nowa iy. F

*(snodeytuog.ep) speq-uopoudorg aeddq -»

‘SOYA OLJLIOTO. *a

‘soy dp oytorg, “f

‘opIURL “0
‘(UBIMOAE(T) OUOJSpURY ULBJUNOPY [qT “¢

‘Aadydtog "y

2

"(g) otc 4D
‘(aiydaouzep_) ystgog puxyenbeueyy “9

‘ssLou “p

58

E. J. DUNN ON THE MODE OF OCCURRENCE

~ Ms Ns Naugat. -~ Schuit Klip.
i )
( Kat River -- Zwart Modder.
ae iver.
(e Buis Valley. }\\\\\_____ T’Nous.
y i Prieska.
a iii
V }}))
1/1)
| )
\\
\
a
) ) Naame Lota Rivers (/)/)// (tana Great Falls
| J) (Orange R.).
pel

——S——

"*-~~ Rhenoster Kop.

Zak River.
Kakamas.

River).

Brakfontein.

--—— Hope Town.

\if--— Orange RB. | (((@----- Cross-hammers
Kopje.
Ss
) © [eset Tourmalin Kop
(Orange R.).
)---- Hzelberg.

<--—— Modder R.
-

til
® 4 al Ni !

\ Hi} s Na
i

aH \

ZAM

| A ) yy Bogner,
oS SH 7/7, Blinkfontein.
NNN i

PEM Ree Kl e, a onaaeetn

De Nuis (Orange

Port Nolloth.

Oogarbis Hill.

Annenous.

Klipfontein Hill.

Ookiep Mine.
- Springbok.

_ No.6 Copper-lode.
, _. Kweek Fontein.

6.1aqsaj0g 02 YROP/ON 10g 3M IUD) ayQ mouf WON.I—~Z “SLT

| N’Gaun.

©
\\_. N’Gannais. oS
)__ I’Gaams.

(YOU T 0} 109} QOLOT ‘TBs [BoL}10A £ Your T 07 sertM 1% noge ‘ayeos [eyUOZTIO FT)

OF DIAMONDS IN SOUTH AFRICA. D9

were found in this drift. Some localities have yielded immense
quantities, as Pniel, Klipdrift, &e.

The newer one, or “recent” drift, occupies the present river-
bed. It differs in composition from the older drift, in containing a
larger percentage of sandstone pebbles and pebbles of soft rocks.
its general colour is greyish, while the older drift is of a reddish
colour. The newer drift contains the same well-rounded polished
pebbles of agate &c. as the older one.

Diamond-workings have been successfully prosecuted for a conside-
rable distance along the Vaal river. In the newer drift ‘‘ Cawood’s
Hope” was an unusually rich digging; while Waldeck’s Plant yielded
the largest diamond yet found in 8. Africa; its weight is 288 carats.

The gems found in the drift-workings are of finer quality, and
invariably command ea higher price than those obtained from the
various “ dry diggings.”

A very extensive development of “ glacial conglomerate ” spreads
over the country along the course of the Orange River, west of
Hope Town (lat. 29° 8.). Further west it leaves the river to the
north, and spreads as a belt through Bushmanland. Isolated tracts
oceur near Beaufort West.

At Prieska, on the Orange river, diamonds were found in this
conglomerate. Several have been found on its surface between
Prieska and Hope Town, notably the “ Star of 8. Africa.”

Systematic search is not likely to meet with much reward in the
case of this conglomerate, as the very nature of its formation forbids
the hope of diamonds being thickiy distributed in it.

Many of the hard boulders are scored and scratched in such a
manner as to leave no doubt concerning its origin.

That the wind has acted as an agent in the distribution of dia-
monds, though perhaps of small ones only, is clearly proved by their
occurrence in the drifting sandy soil at Du Toit’s Pan. It has, no
doubt, been the means by which the streaks of sand found tv a con-
siderable depth in the pipes were supplied.

A complete series of the diamond-field rocks has been placed in the
British Museum. Their arrangement in the country is shown by the
accompanying section from Port Nolloth to Colesberg Kopje (fig. 2).

Discussion.

Mr. Masketyne complimented the author on the contribution
which he had made to geological science, by showing that the rock
in which the diamonds occur is confined to the pipes such as he had
described. He was able, from specimens which he had examined,
to state positively that this certainly igneous rock differed essen-
tially from the dolerites and other igneous rocks in the neighbour-
hood. He would not attempt to give it a name, whether gabbro,
euphotide, or any other designation. The materials, so far as they
could be determined in their present altered condition, were such as
would not build up any one of the known rocks. There were
garnets, and minerals resembling clinochlore, smaragdite, phlogopite,
diallage, and calcite, and apparently another mineral entirely ser-

60 E. J. DUNN ON SOUTH-AFRICAN DIAMONDS.

pentinized. The diallage was the true diallage, viz. a mineral of
augitic type. The mineral resembling clinochlore was, he had
proved, vermiculite ; and the phlogopite was not improbably Jeffrey-
site. The calcite had been thought to have been derived from
the decomposition of augitic silicates; but this he doubted. One
specimen, from a depth of 130 feet, exhibited a completely serpen-
tinous metamorphosis, which had involved one mineral as yet
undetermined, but probably enstatite. The absence of felspars
in all the rocks but one was singular; in one specimen from Bult-
fontein, however, there was an abundance of kaolinite, proving the
original presence of a felspar. He thought that the fracture of the
diamonds might be due to the breaking up of the rocks in which
they are now found. <A fact pointed out by the late Prof. Rose,
viz. that the octahedral faces of diamonds became grooved after
burning by triangular striations, was of great importance in con-
sidering the changes these rock-fragments had undergone, as
many of the diamonds from the upper beds had their surface
striated in this manner. It would be a question of interest to
observe whether those from the lower depths were similarly marked
and equally shattered.

Mr. Forzss observed that, owing to his not having had the op-
portunity of examining the rocks in question, he could not speak
as to their nature ; but as he looked upon the rock gabbro as being
in the main composed of triclinic felspar with diallage, he could not
regard any of the rock-specimens shown as entitled to this name,
especially as they, as a rule, were even devoid of felspar. The black
rock, found deepest, appeared to him to resemble one of the altered
basalts, in which Zirkel had shown that the olivine was changed into
a peculiar greenish serpentine-like mineral.

Mr. Duny, in reply, stated that he had merely made use of the
term gabbro as a provisional name. There could, he thought, be
no doubt of the volcanic origin of the rock; and, moreover, near
Bultfontein and elsewhere, there were sheets and dykes of dolerite,
but quite different from the rocks in the pipes. Some of the fragments
in the matrix were fragments of shale, mica-schist, and hornblende,
which had probably been brought up into the pipes from below.

Prof. Ramsay remarked on the wide application of the term
gabbro, which could hardly be said to imply any distinct mine-
ralogical character. Looking at the question broadly, the facts
seemed to be, that among nearly horizontal shales there were
patches of rock of approximately circular form, with the shale turned
up around their edges, and altered for a certain distance. The
appearances were therefore those of the extrusion of a heated or
molten body by expansion from below ; and possibly, as the author
suggested, other traces of eruption had been removed by subse-
quent denudation. He could not regard the rocks in the pipes as
in any way metamorphic in the usual acceptation of the term.
Though the voleanic origin of those rocks might be accepted, he
thought it more probable that the diamonds were brought up from
some subterranean metamorphosed rock, than generated in the lava.

A. B. WYNNE ON THE PHYSICAL GEOLOGY OF THE UPPER PUNJAB. 61

12. Observations on some Fratunus in the Prystcat, Guoroey of the
Ovurer Himarayan Reeron of the Upper Punsis, Inpra. By
A. B. Wrwnz, F.G.S. &e. (Read December 17, 1873.)

[Puate VII. ]

ContTENTS,

1. Geographical position of Upper Punjab.

2-5. Geological structure; different zones; isolated appearance df the rocks
underlying the outer zone at the Salt range.

6. Previously published geological information about the Upper Punjab.

7. Comparative table of local and other Himalayan series; remarks thereon.

8. Physical conditions of Upper-Punjab rocks: alteration in strike of outer
ranges of Himalaya.

9. Himalayan and extra-Himalayan features, resembling Alpine and extra-
Alpine features of continental geology.

10-12. Description of the local Newer Tertiary Series.

13. Observations regarding the Simla Himalayan region as bearing upon the
geology of the Upper Punjab.

14, 15. Junction of the outer Tertiary Zone with the “ Hill-rocks,” and locai
relations of contact.

16. Difficulty of applying the explanation of the Simla area here.

17. Occurrence of Simla relations at Dundelee near this district.

18. Dr. Verchere’s representation of structural relations of outer Tertiary zone
in Upper Punjab, and basal contact at Oori, in Kashmere.

19-21. Consideration of local features with reference to abnormal contact of
outer Tertiary Zone and “ Hill-rocks.”

22. Conclusion.

1. Tue geographical position of the Upper Punjab is too well
known to require detailed description ; and it will be sufficient to say
that the country to which the following observations mainly refer is
that part of ‘the land of the five rivers ” in which the Jhilam *
(Hydaspes of the ancients) and the Indus or Abba Sin (* Father
of Waters”) flow rapidly through rocky channels before finally
leaving the hills. (See Map, Pl. VII.)

The principal British stations in the country are those of Jhilam,
Rawul Pindi, Abbottabad, Fort Attock, and the Hill Station of
Murree f.

2. The geological structure of the ground with reference to the
points to be noticed may be generally stated thus :—This section of
the Outer Himalayas consists of a yaried assemblage of rocks. The
crystalline, granitoid, syenitic, and schistose rocks, with abundance
of greenstone dykes, found far in among the hills, are succeeded by
slates and limestones (of possibly, Silurian age) unconformably over-
lain by Triassic and perhaps older rocks, which may be provisionally

* Always pronounced by natives Jhelum.

+ Koh Mari, or the Mountain of Mari, from an old Mahomedan shrine, or
Zyarut, now a Hindoo temple. The station and locality are very commonly
called by the natives Musheeri, a village of that name being situated in the glen
below.

Q.J.G.8. No. 118. G

62 A. B. WYNNE ON SOME FEATURES IN THE

called infra-Triassic. The Triassic beds are, again, unconformably
succeeded by Jurassic limestones, Cretaceous limestones, and a mass
of Nummulitic limestone and shaly strata, all three resting conform-
ably, the newer on the older.

From the Trias upwards and inclusive, the rocks consist so largely
of limestones that the country formed of these Secondary and older
Tertiary strata may be comprehensively spoken of as a zone of
limestone hills or “ Hill-limestones.” Exterior to this zone is
another of hills (occasionally high) and broken plains composed of
a mass of sandstones and clays, with conglomerates, the whole of
enormous thickness and of Tertiary age. These sandstones, clays,
and conglomerates have been subdivided in this and in other dis-
tricts into groups, among which EKocene (Nummulitic) and Miocene
(Sivalik) stages have been recognized.

3. The relation of this outer zone to that of the hill-limestones, is
to form one of the principal subjects of the present communication.

4, So far as is known to the writer, the belt of Tertiary sand-
stones, clays, and conglomerates just now mentioned passes along
the whole southern foot of the Himalayan Mountains, beneath the
thick alluvial accumulations of the plains, from Assam to Afghanis-
tan, with but one exception, where an outer edge to the belt is
seen brought up on the northern flanks of the Salt range in the
Punjab and along its tortuous extension trans-Indus. The width
of the belt in this part of the Punjaéb is thus fixed at from 50 to
60 miles; but it decreases westward and turns to the south along
the Indian flanks of the Suliman Mountains.

Beneath the outer edge of the Tertiary belt the Salt range partly
repeats the systems found in the Himalayan hills, but with very con-
siderable differences as to the number and extent of the formations,
and also as to their lithological or petrological and paleontological
characters*.

Beyond the Salt range to the southward stretch the flat deserts
of the Punjab and Sind, broken only by a small cluster of elevations
called the Korana hills, the rocks of which, according to Dr.
Fleming (Journal Asiatic Soc. Beng. vol. xxii. 1853, p. 444) appear
to belong to one of the older groups, perhaps older than any of the
Salt-range rocks, and not far removed from the Metamorphic or
Crystalline series.

5. Geological information in a published form respecting the dis-
trict under notice is rather scanty and uncertain; but the whole region
lies within the area of Dr. Albert Verchere’s map appended to his
voluminous paper read before the Asiatic Society of Bengal (1866—
67). Dr. Fleming’s reports, previously alluded to, treat of the Salt
range only; and a small memoir on the Geology of Sir Ban Mountain,
near Abbottabad (Mem. Geological Surv. Ind. vol. viii.) deals chiefly
with the local geology of that mountain. Some official correspond-

* The comparison has not been as yet fully worked out; but enough of evi-
dence has been collected to warrant this statement, the paleontological differ-
ences noticed having been confirmed by Dr. W. Waagen’s inspection of the
rocks, in company with the writer.

\ mh ny Me apn
: a Wane ge : wi Oe wk } = - ty oe . f ey Ty, ity
~ eh + me re Mr a ny ANA is ‘ aly 7 we hg! i ina ibe

/ Te A { iM vi Hit : ff , i my ¢

A
me
a

i byt ae ra A i‘ ' vee
‘piel a Ni he r oi 1
geist a by soli pat a i

Pip De,"

iis 7 it i

ed min

i bho That
ra

6 walle, LA yi

= Oe

<b,
nee

oll i ae
ay “) moat Ate
Pine 0)

ae A ia

i

. inet
ik va
kite ,
ii iii Pi |

To face page 63. |

SALT RANGE, Upper PunsAB.
ies a i slieeinad
Post-Tertiary ...... Kunjure post-tertiary or drift. Ku
- Unconformity
Tertiary sandstone-and-clay zone, viz. :— Te
Conglomerate sandstone and clays: bones &. &c.
| LIBR H rac05 3-9: Soft grey sandstones and red clays...........-10+...
Thick red clays and grey sandstone.
| Greenish sandstone: crocodilian remains and
apeeene fossil wood.
| aa ela 4 | Nummulitic limestone.
(That of the hills absent in the Salt range?) ...
CsSABED TS cosa 1o2s Olive sandstone: few fossils.
INR earl WE peak tues ee an
Jurassic..........5. 1 Sandstone, limestone, and shale (fossils).
| sees fama
Green shales, limestones, &c. (fossils), and red
AUEWAESO sooasqn05002 sandstones and shale, Salt pseudomorphs.
|
| eee Lend
| Infra-Triassic ......
i" 5 |
| Carboniferous ... { Limestones, dolomite, &c.: fossils.
| Upper sandstone, age unknown.
aa Ebene
| Silurian ? ............| Obolus- or Siphonotreta-zone, dark shale.
Lower or purple sandstone.
Red gypseous marl and rock salt, precise age
unknown.
| Metamorphic ......
| Crystalline .........
*

Nors.—It appears from a paper by

PHYSICAL GEOLOGY OF THE UPPER PUNJAB. 63

ence with the local government, and a Report by Mr. H. B. Medli-
cott, in 1859, on the subject of coal said to exist in the Murree hills,
have been printed ; and there are other scattered geological allusions
to portions of the district ; but in none of these, so far as the writer is
aware, are the structural relations between the different rock-groups
discussed at length or with any accuracy of detail*.

The nearest places to the Upper Punjab which have been systemati-
cally explored by officers of the Geological Survey of India, are, first,
the Himalayan regions for many miles north-west and south-west of
Simla, lying between the rivers Ganges and Ravee (which may be
called the Simla area), examined and ably reported upon by Mr. H.
B. Medlicott (Mem. Geol. Surv. Ind. vol. ii. pt. 2); and, secondly,
a region situated further in the interior of the N.W. Himalaya,
explored by Dr. Stoliczkay (see Mem. Geol. Surv. Ind. vols. ii. & v.).

These two regions, however, are separated by a wide gap, including
the whole of the valley of Kashmere, from the Upper Punjab, which
the researches of Mr. Drew (for some years geologically employed by
the Kashmere government) would doubtless, if published, go far to
fill up in a satisfactory manner. Something is already known of this
intervening country from the papers of Major Godwin-Austen and
others published by the Geological Society, from Dr. Verchere’s
‘paper already mentioned, and a few other sources, but not sufficient
to enable the formations and groups on each side to be connected
across so great an interval, though intermediate links undoubtedly
exist.

6. In order to present a concise view of the rocks entering into
the structure of the Upper Punjib and N.W. Himalaya, so far as
yet definitely ascertained, the local series of the former are placed,
in the table opposite, side by side with the classifications of Mr.
Medlicott and Dr. Stoliczka, as given in the memoirs referred to.

This table will show that in the outer Himalaya of the Upper Pun-
jab, as well as in the Salt range, several geological formations are
represented. Five of these likewise occur in Dr. Stoliczka’s classifi-
cation of the Western Central Himalayan sections ; but in the Simla
area, from want of paleontological evidence, while there are eleven
groups distinguished, the Eocene and Miocene formations only are
identified. There is a suggestion doubtfully put forward by Mr.
Medlicott, partly on the authority of the MM. Schlagintweit, that
the Krol rocks may be Nummulitic, though his observations tended
to negative this ; and it has been also thought not unlikely that they
may be Triassic ; but of the ages of the groups below nothing is
known.

Major Godwin-Austen, Messrs. Davidson, Etheridge, 8. P. Wood-
ward, Dr. Verchere, and M. de Verneuil have shown that Carboni-
ferous Limestone occurs in Kashmere, and, according to Drs. Ver-

* Since the above was written, a short paper by the writer on the geology of
this region has appeared in the Records Geol. Sury. Ind. vol. vi. part 5, p. 59.

t Dr. Stoliczka’s deputation with the present expedition to Yarkand, will, it
is expected, much extend his former valuable researches, with which his new
observations will form an interesting and connected series.

EZ

To face page 63

J

SAL? RANGE, Upper Punsas.

Ovurer Himanaya, Uprrr Pongas.
Mount Sir-ban * &e.

Ovrer Himanaya, Sia country.
Ganges to Rayee. (Medlicott.)

N.W. Cenrran Himarava, Surues to Urrer Inpvs.
(Stoliczka.)

| Post-Tertiary

| Miocene...........-

Kunjure post-tertiary or drift.

Unconformity ———
Tertiary sandstone-and-clay zone, viz. :—
Conglomerate sandstone and clays: bones &e. &e.

Soft grey sandstones and red clays

Kunjure post-tertiary or drift.
— —Unconformity———

Tertiary sandstone-and-clay zone, viz. :—
Conglomerates and clays and sandstone: bones ......
Soft grey sandstones and red clays: bones and obscure
plants.

Indo-Gangetic-plains deposits.

| Sub-Himalayan Series.

| Tertiary sandstone-and-clay zone, viz. :—

Sivalik conglomerates, sandstones, clays: bones.

Unconformity
‘Nahun lignite sandstones and clays.

Thick red clays and grey sandstone.

Gray and purple sandstone and purple clays: bones

SSS Urconformity.
[ Kasaoli grey and purple sandstones.

Mammaliferous deposits of Tibet.

and obscure plants. (Murree beds.) Subathu: Dugshai purple sandstones and red clays.
Greenish sandstone: crocodilian remains and Variegated and gypseous clays and limestone: fossils, abatow) Subathu fine silty clays with limestones.
E fossil wood. as bones and Nummulites &c. (Nummulites.)
Dore ree 4 | Nummulitic limestone. |
| — Abnormal junction or fault——— | - Unconformity
| | Himalayan Series.
| (| (That of the hills absent in the Salt range?) ... Nummulitic limestone and shale of hills. Krol Limestone. Nummulitic t ? Trias?
Infra Krol. Carbonaceous shales or slates, age) { oy
act : unknown. Chikkim shale,
OR toe Olive sandstone: few fossils. Thin-bedded limestone; no fossils known. Chikkim limestone, Cretaceous.
innipureterrupmious Limestonels; fossils! .cscsec.ss-eeseee=|[ f) repre
ee Blini. Limestone and conglomerate, age unknown. | —_ — ——
: : Pn hl Brea crwauccostcovsawiuccceccecsscsecceressy 1 NG ie IR ST Stl Ss HL Hit ocean Gieumal series. :
| Jurassic............ { Sandstone, limestone, and shale (fossils). Limestone (fossils) and Spiti shale (fossils). Spiti shale. | Jurassic.
Unconformity — —- ———$——————
Green shales, limestones, &e. (fossils), and red ae oe
| abarastc #..--..-..5 sandstones and shale, Salt pseudomorphs. Thin-bedded limestone and slaty shale: fossils. Ee Triassic.
; Dolomite and Megalodon-limestone. | :
Infra-Triassic ...... Siliceous dolomite over red sandstone and shale. | | | | Liling series ?
Unconformity- —| —
, Carboniferous ... { Limestones, dolomite, &c.: fossils. (As yet undiscovered.) Kuling series ? } Carboniferous.
| Upper sandstone, age unknown.
! Silurian? ...........- Oholus- or Siphonotreta-zone, dark shale. Muth series ? \ Silurian
Dark Attock slates &c.; no fossils known. Infra Blini. States, age unknown. | Babeh series ? ;
Lower or purple sandstone.
Red gypseous marl and rock salt, precise age
unknown.
| Metamorphic ...... Metamorphic. Metamorphie.
Crystalline ......... a Syenite and Trap. , Crystalline and suberystalline, &e.

* Pronounced Seer-bun.

+ Doubtfully referred by Mr. Medlicott to this age.

Nore.—It appears from a paper by Mr. Medlicott (Records Geol. Sury. yol. vi. part 5, p 52), that the Sivalik rocks may be either Older Pliocene or Upper Miocene.

a |

64 A. B. WYNNE ON SOME FEATURES IN THE

chere and Stoliczka, very possibly Triassic rocks also. In this
territory, too, the presence of the granitoid and schistose rocks of
Hazara in the Upper Punjéb has been ascertained, and likewise
that of the outer Tertiary belt of sandstones, clays, &c., while the
Krol limestone has been conjecturally identified in the feudatory
state of Poonch, to the south-west of the Kashmere valley, by Mr.
Medlicott.

7. There can be scarcely a doubt that several of Mr. Medlicott’s
divisions of his sub-Himalayan series are present in the Upper
Punjéb; indeed we have the advantage of his personal identifica-
tion of some of them in this district, and also, among the next older
rocks, his provisional recognition of the Krol limestone near Murree.
To attempt a further comparison of the lower rocks of the Upper
Punjab with those of the Simla area, until organic remains have
been found, would be little better than guess-work; they have
therefore been bracketed in the table; but the Infra-Blini rocks
seem, from description, to resemble those we have called the Attock
slates. If this could be established, resemblances might also be
found among the Metamorphic and Crystalline groups.

With respect to the part of the series older than Tertiary, it
appears strange that the more central Himalayan regions traversed
by Dr. Stoliczka should present a closer analogy to the outer
Himalaya of the district under notice than the Simla area does ; but
from the table given it will be seen that such is the case.

8. The physical conditions of the Upper-Punjab rocks, both in the
Salt range and in the outer Himalayan hills, include contortion
and faulting; but the disturbance, in part at least, of the Salt range
has been much less than in the latter hills, and differs somewhat in
kind. For instance, in the Salt range most of the complexities are
produced by extensive land-slippage of ground previously more or
less the scene of disturbance, while in “the hills” intense contor-
tion, inversion, and faulting have taken place. A feature common to
both localities is the prevalence of dips towards the more elevated
masses of the Himalaya and Hindoo Koosh Mountains *.

The intervening Tertiary sandstone, clay, and conglomerate zone,
where the ground has a plain or plateau form, frequently presents
nearly as much contorted disturbance as either the Salt range or
the outer Himalayan hills.

In this Upper Punjab country too, the normal Himalayan strike
of the rocks and ranges alters abruptly in the Jhilam valley (as
noticed by Mr. Medlicott) from a bearing of E. 35° S. to N.E. and
S.W., marking a change in the direction of the disturbing forces.
North-easterly, or more east and westerly, lines of strike prevail
elsewhere over the district, except towards Kohat and the Bunnoo

* During a recent visit to Kashmere this dip towards instead of from the
great mountain-axes of the Himalaya, so common on the outer flanks, was found
to occur also in the chain running along the N.H. side of the valley, where the
glens of the Sind and Siddur and other rivers cross the strike, exhibiting strong
dips to the north-eastward, and some fine escarpments of Carboniferous Lime-

stone, slate, and other rocks presented the opposite way. Local curves over axes
parallel to the range also occur.

PHYSICAL GEOLOGY OF THE UPPER PUNJAB. 65

frontier, where the westerly run of the rocky ridges is, so to speak,
crushed against that of the stronger nearly north and south run of
the Suliman range.

The hills forming the outworks of the Himalaya in this district,
may be said roughly to vary from below 5000 to over 9000 feet
above sea-level, the higher elevations of 14,000 and 15,000 feet
belonging to lofty chains forming integral portions of the above
mountain-system. The most lofty peak, perhaps, in view from the
higher hills between Murree and Abbottabad is that of Nanga
Parbat (Nunga Parbut), 26,000 and odd feet, situated near Astor,
north of the Kashmere valley. The Salt-range elevations reach
from 2000 to 5000 feet above sea-level; and the intervening plateau
to the north may be taken at 1700 feet, the altitude of Rawul Pindi
Station.

9. One peculiarity to be mentioned before passing on is, that a
decided difference exists between the physical geology of the outer
Himalayan regions and that of the Salt range, akin to the differ-
ence between the Alpine and extra-Alpine characters of European
rock-groups*. Taking the Obolus- (or Stphonotreta-) zone of the
Salt range and the Attock slates of the Himalayan hills to be both
Siluriant, they are marked by characters petrographically entirely
distinct, and from the Silurian upwards to the Miocene there is no
formation common to both localities which does not present in each
strong dissimilarity both in lithological aspect and in the assem-
blage of contained organic remains, though the latter prove the
rocks to be Triassic, Jurassic, Cretaceous, or Eocene, as the case
may be, with the same certainty in one locality as in the other.

There appears, then, to be a parallel between the Alpine and
extra-Alpine and the Himalayan and extra-Himalayan features of
these two distant regions; but the diversity of character in this
district has been contended for no further in the ascending series
than above the Eocene, because the Miocene forms a part of the
great Tertiary sandstone, conglomerate and clay belt, which is in
contact on different sides with both Himalayan and extra-Hima-
layan rockst. Even in the Miocene formation, however, some

* The verification of this is due to the Palzontological officers of the Geo-
logical Survey of India.

t The reasons for provisionally assuming the Attock slates to be of Silurian
age are their similarity of position to those of Dr. Stoliczka’s Himalayan see-
tions, and the recorded fact of Dr. Falconer and Major Vicary having discovered
the existence of Silurian fossils in the mountains drained by the Cabul river
(Quart. Journ. Geol. Soc. vol. vil. p. 38), mountains which may be fairly supposed
to be formed by an extension of the slates known to project westward into the
Peshawur valley beyond Attock.

t Pending the possible discovery of Carboniferous rocks in the outer Hima-
laya of the Upper Punjab, it may be as well to state that among the Kashmere
Carboniferous fossils occurring within the Himalayan region, species probably
identical with those of the Salt range have been determined ; but it remains
to be seen whether the whole Carboniferous fauna of one locality is or is not
identical with that of the other. The rocks are certainiy different: limestones
occur in both; but the dark slaty fossiliferous Carboniferous rocks of the Siddur
valley in Cashmere have no similarity to any of the Salt-range Carboniferous
beds with which the writer is acquainted.

66 A. B. WYNNE ON SOME FEATURES IN THE

dissimilarity may be traced, inasmuch as the sandstones and clays
which rest with apparent entire conformity upon the Nummulitic
limestones of the Salt range, differ in character from those beds of
the outer Tertiary belt which are brought into violently discordant
junction with the hill-limestones of the outer Himalaya *.

The recurrence of these mountain and extra-mountain features,
so to speak, at such great distances may indicate a connexion
between the former conditions of deposition and the early history of
the great chains themselves ; at least causes which would affect the
distribution of land and water appear most competent to have
originated the diversity f.

10. In the table given at p. 63 an effort has been made to corre-
late the newer Tertiary series of this part of the Punjab with the suc-
cession in the Simla area, as gathered from Mr. Medlicott’s descrip-
tion ; but it may be as well to give a slightly more comprehensive
account of these Upper-Punjaéb newer Tertiary beds, for the sake of
comparison with the series recorded by that gentleman (see Report
previously cited).

From the sections exposed in the river Indus, and numerous
observations over the Pot’war or Sind Saugor Doab (which has been
called also the Rawul Pindi plateau), it appears that the only
natural base of these rocks of any extent to be met with is along
their upturned edge flanking the northern slopes of and rising on to
the Salt range. Here, notwithstanding the perfect parallelism of
the sandstone and clay beds to the strong underlying Nummulitic
limestone, the absence of subjacent erosion of the latter, and the
great length of the apparent conformable contact, there is reason to
suspect that the conformity may be but local and not general, or
else that the Salt-range Nummulitic limestone is not a strictly
synchronous extension of the Nummulitic limestone to be found
in other parts of the district.

A. The bottom rocks of this region are soft or harder grey or
greenish sandstones, in places slightly calcareous, alternating with
red, or sometimes dull grey or greenish olive clays. The contained
fossils are blocks of petrified timber, crocodilian and chelonian bones
and plates, with rarely (derived, or perhaps sometimes contempora-
neous ?) Nummulites in the lower part.

B. Where the Nummulitic limestone has thinned out, these beds
rest on some of the older groups of supposed Cretaceous or Triassic
age, and they are overlain by a strongly marked zone (B) of red

* The rocks of this belt have been stated to possess more or less affinity with
those of the Simla area; but they differ en masse and in detail from the large
Tertiary deposits newer than Eocene (Nummulitic) of the distant province of
Kutch. It remains to be seen whether and where a transition between the two
types may be found, one of these types belonging to the Himalaya region, the
other more nearly associated with the southern extension of the Suliman chain,
two great mountain-ranges which inosculate near the frontier of the Punjab,
where the whole of the Tertiary rocks resemble in character those of the Salt
range and outer Himalaya.

+ It is perhaps hardly necessary to note that the features referred to in this

section as bearing relation to the mountain-ranges include, in consequence of
this relation, more than the common mutability of formations.

=

PHYSICAL GEOLOGY OF THE UPPER PUNJAB. 67

clays containing harder slightly calcareous layers interstratified with
grey sandstones. The upper and lower boundaries of this group are
quite indefinite ; and the only fossils found in it besides obscure
plant-markings are small fragments of bones. So far as position
goes, this zone seems to correspond more or less with another to be
presently noticed under the letter E.

C. The upper part of the red band just mentioned passes gradu-
ally into a great thickness of soft pulverulent grey or light bluish
sandstones, alternating with red clays and occasional bands of light-
coloured slightly ferruginous conglomerate, which sometimes con-
tain bones, and pebbles of Nummulitic limestone enclosing small
Nummulites. The strong light-coloured sandstones have here and
there thin layers of lignite, the carbonized remains of fossil plants
or trees ; and the zone has been called the “ red and grey group.”

D. Above these red and grey beds the sandstones continue of
much the same character; but the alternating clays are fre-
quently of an orange-drab or pink colour, and towards the top
the series gradually passes into a thick set of incoherent boulder-
and pebble-beds, sometimes cemented so as to form conglomerates,
the pebbles of which are a most heterogeneous assemblage derived
from the crystalline, igneous, and limestone rocks of the Himalaya.
In the sandy beds associated with this conglomerate portion of the
group fossil bones also occur; and from the waste of the softer
strata the gold of the Pot’war country is apparently obtained.

Throughout the whole of the foregoing and among the rocks
next to be mentioned are numerous beds of a fine pseudo-conglo-
meratic aspect, composed of a sandy calcareous matrix charged
with small clay concretions and sometimes ferruginous nodules.
These beds not unfrequently contain fragments of bones.

E. So far the succession has been traced from the Salt-range
bottom beds upwards; but on coming to the northern side of the
Tertiary sandstone, clay, and ‘conglomerate belt, we find a change
in the formation. Here the series may be followed downwards
from the group D through C ; but in the place of the red zone B is
a very great thickness of purple and grey and purplish-grey or blue
sandstones, of much harder nature than any of those previously
described, with polished metallic-looking films, as of iron or man-
ganese, on joint-surfaces, alternating with purple or reddish clays
not unfrequently having a nodular structure. The sandstones are
often veined with carbonate of lime and penetrated by obscure
plant-impressions; and more rarely they enclose imperfect bone-
fragments.

Along the southern edge of this band its beds pass without any
abrupt or marked change into the “red and grey series” C; but
on the northern side, at the base of the limestone hills (or where all
is hilly ground, then contiguous to the hill-limestones), there are
many alternations of purple and red beds, with grey and yellow
muddy calcareous strata, associated with dark olive, blackish purple,
or greenish-grey shaly and sandstone bands, in the vicinity of
which a zone of very red clay, including layers, veins, and masses

68 A. B. WYNNE ON SOME FEATURES IN THE

of white and reddish gypsum, may be frequently observed. The light-
yellow earthy calcareous beds are crowded with Nummulites and
large Rotaline, Ostree and other Bivalves, with a few Gasteropods,
while more solid limestone, of a lavender-grey colour, found nearer
to the gypsum, contains a few thin-shelled Gasteropoda resembling
some species of Planorbis, but no Nummulites that could be observed.
Near these latter bands, and generally not far removed from the
gypseous zone, the beds (most frequently those which are calcareous
or limestone), are in places impregnated with petroleum, affording a
scanty supply, for some time utilized in lighting part of the station
of Rawul Pindi with gas.

11. These varied strata along the inner or mountain margin of the
zone H, are subject to some amount of change both as to quantity and
composition. At certain localities the dark olive-coloured beds are
more largely developed, containing some whitish quartzose gravelly
bands ; in some parts bright purple clays, in places highly ferruginous,
predominate, associated with massive dark bituminous limestones
enclosing very small Foraminifera, and closely resembling in char-
acter the Nummulitic “‘ Hill-limestones.” At other localities the
ordinary purple beds have interstratified with them but a few lime-
stone layers, or sandstone, or marly bands, containing, as usual,
quantities of Nummulites, Rotaline, Orbitolites ?, and some fragmen-
tary bones, while in yet other places the dark olive and greenish
beds are locally absent. The layers just described form the oldest
portion of the outer Himalayan Tertiary sandstone and clay zone
along its junction with the “ Hill-limestones,” though to an observer
approaching the foot of the hills the almost constant inversion of
the strata would make them appear newer than the main mass of
the subdivision E, which has been provisionally called the “ Murree
beds ” from its prevalence at the hill station of that name.

This group EH, from many of its characters, appears closely to
correspond with Mr. Medlicott’s Nahun group; but the lower portion
would seem to present most of the features of his Subathu division
(Mem. Geol. Surv. Ind. vol. i1. pt. 2, pp. 11, 12, &e.) *.

12. The whole of the Outer Himalayan Tertiary sandstone and clay
belt, from the limestone hills on the north to the Salt range in a
southerly direction, has presented, so far as yet known, no single
instance of a clear unconformity or overlap within its limits, while
its entire aspect, notwithstanding much contortion, is that of parallel
sequence throughout. No erosion of any of its beds prior to the
deposition of others upon them has been observed; and yet it must
be remembered that some of the layers, and these on different
horizons, contain rolled pebbles of fossiliferous Nummulitic lime-
stone, proving breaks to have occurred subsequently to the deposition
of this rock. It cannot be asserted with accuracy whence these

* The description of the Nahun group in this Report would answer in a
general way for all the Tertiary sandstones &. below the conglomerates of our
group D; and these conglomerates would answer well to the Sivalik beds, with
which some of them have been identified by Theobald and Falconer (Journ.
Asiatic Soc. Beng. 1854, p. 651).

PHYSICAL GEOLOGY OF THE UPPER PUNJAB. 69

rolled pebbles were derived ; but other fragments associated with
them appear to have come from a northerly direction rather than
from the Salt range, where apparent conformity has been stated to
exist*,

13. This state of things seems widely different from that of the
Simla area, where Mr. Medlicott, who speaks of the Murree rocks as
Subathu, and the beds next succeeding the Salt-range Nummulitic
as Sivalik (Memoir previously referred to), has found two strong
unconformities or overlaps—namely, between his Sivalik and Nahun,
and between his Nahun and Subathu groups ; further, Mr. Medlicott
has found the whole of the outer Tertiary detrital zone of the Simla
area from the base of his Subathu group upwards discordant to the
Himalayan and Hill series, and discordant to each other, all three
of his divisions transgressively reaching and resting upon the older
rocks.

Referring to the Report of the Simla area, it appears impossible
to disagree with the author’s conclusions as to the relations of the
newer rocks in that country. Here, however, the structural features
differ from those he describes, and the breaks may not have existed to
the same extent; but this supposition interferes with the connexion he
would establish between their existence and the mode of elevation of
the Himalaya—uniless a difference may have existed also in the kind
of formative agency which produced the Himalayan extension to-
wards the Upper Punjab.

14, The junction of these newer Tertiary rocks with those
forming the higher of the Outer Himalayan hills, both in the Simla
area and in the Upper Punjab, is one apparently marked by much
the same features of disturbance, distortion, and inversion, or abnor-
mal superposition in the Tertiary strata along the contact (characters
also known to prevail in the Alps, see Medlicott’s “‘The Alps and the
Himalayas,” Quart. Journ. Geol. Soc. vol. xxiv. p. 34 &.), the
chief resemblance being that in both regions a disturbed and abnor-
mal junction of these Tertiaries with the hill rocks occupies the same
geographical and orographical position among the outer hills or at
their foot, and the chief difference lying in the fact that in the
Upper Punjab there is but one prominent zone of such abnormal
junction (that coincident with the boundary between the Hill-lime-
stones and the outer Tertiary belt), while in the Simla area this
feature is said to be repeated at least three times (Report cited).

In the district under notice this junction follows a line describing
an arc of a circle having a radius of 60 miles, and its centre some
72 miles N. 20° W. from Rawul Pindi, the western continuation of
the line passing with less curvature just north of the station of
Kohat, whence it enters the Affreedi and other Afghan hills. (See

* Mr. Medlicott states that the sandstones and clays of the upper sub-
Himalayan groups [7. e. Sivalik?] rest upon a denuded surface of the Nummu-
litic limestone of the Salt range (Memoir quoted, p. 91). Nothing of the kind
* has been observed from end to end of the Salt range by the writer, but the
contrary, as above stated. This, however, is no reason for assuming a foregone
conclusion on the part of Mr. Medlicott, based upon his observations in the
Simla area.

70 A. B. WYNNE ON SOME FEATURES IN THE

Map, Pl. VII.) Its eastern extension, from the point where the are
touches the Jhilam river, turns abruptly northwards along the right
bank of that stream, then, crossing the sharp angle formed by this
river near Mozufferabad, bends to the south-westward, lying gene-
rally high up on the flanks of the Kyjnag range in Kashmere. At
the village of Oori in this territory it crosses the Jhilam (here called
the Vedusta) with a bearing which would take the line along the
outer flanks of the Peer Punjal chain.

The rocks of the outer Tertiary belt in the Upper Punjab, and
as far as they have been seen in Kashmere, which are in contact
with the limestones and other rocks of the hills, unlike those of the
Simla area, belong to one subdivision only, namely the Murree beds
(E), z. e. Subathu or Nahun, as the case may be. On the inner or
«‘ Hill” side of the line, however, different groups of the hill-rocks
come into contact with these Murree beds. Strong Nummulitic
limestone (a portion of which was conjecturally identified with the
Krol limestone of the Simla area by Mr. Medlicott, 1. ¢. pp. 91,
92) occurs commonly in this position ; but sometimes the rocks in
junction are of Cretaceous or Jurassic age, as shown by the Am-
monites, Belemnites, and T'rigonie which they contain, while in the
Kashmere territory, if any dependence is to be placed on a somewhat
distant view, the Murree beds are sometimes in junction with meta-
morphic or semimetamorphic slates, with crystalline rocks and with
hardened and altered-looking limestones, in which, what appeared
to be the distorted forms of small Rotalinw drawn by cleavage were
recognized, giving ground for the conjecture that near Oori, where
the observation was made, some representative of the Hill Nummu-
litic beds was present *.

15. The local relations of the contact ¢ are tolerably uniform in
their discordance, the hill-rocks being crushed into most compli-
cated folds, while the Murree beds are either less contorted or
vertical or inclined, frequently underlying towards, seldom away from
the hills, and then only at high angles. How closely similar these
features of the contact may be to those of the Simla area it is not
easy to assert without having inspected both; but from the descrip-
tions given they seem to have a general resemblance; Mr. Medli-
cott’s study of the circumstances led to the belief that the junction
with the hill-rocks marks a limit of deposition of the Tertiary beds
from the base of the Subathu upwards, these having been deposited
on a denuded surface of the hill-rocks, and the whole subsequently
reduced by disturbance (of both simultaneously) to their present
complex condition. In the Upper Punjab country, evidence of any
denudation of the Hill-limestones preceding the deposition of the
Murree beds (considered as equivalents of the Simla-area Subathu

* Here, as usual, inversion appeared to be the rule, these limestone rocks
overlying bright red and green clays and purple sandstones, clays, &c. of the
Murree type. i

+ It should be stated that actual contact is seldom seen along this line of
junction, though the rocks im situ on different sides frequently approach each
other very closely.

PHYSICAL GEOLOGY OF THE UPPER PUNJAB. ak

or Nahun) has never been detected ; and the features which expose
Jurassic or other rocks beneath the Eocene limestones and shales of
the hills at the line of contact, are those produced by the same action
of denudation which has affected the Murree beds themselves. Nor
is the line of contact in the Upper Punjab strictly a limit of deposi-
tion confining the appearance of either the hill-rocks on one side or

e. Hill- Nummulitic Limestone.
red and greenish-grey shales.

fe

a
at
a=)

:
Oe)
»

=I

S
“e)

oH

9°

So

nD
4
eo)
oH

°

o

‘S|
“4
el

t Shahka-Noorpoor, N.K. by KE. from Rawul Pindi.
d. Hill- Nummulitic Limestone.

d:. Jurassic, Cretaceous, or Triassic unfossiliferous Limestone.
(Length about 3 miles; highest hills 4000 feet.)

6. Limestone bands (Nummulitic), gypseous

1.—Section across the Chita range at Lumbidund, near the Indus.
(Length about 6 miles ; hills 1500-2000 feet.)
b. Limestone bands, with Nummulites &c.

o

qi

oO

~~

: 8 3S

E Hl

rg Ss 4

Q eS ©

S oo

fi S B

4 QR S

Ms 3

; XN -

aS} = S

2 =

a & S oe
Fe S q
if 3 S o
fet 8) 2 uw) | =e 2
een y i 4
et 2 a = |) 2
= 4 4 ——~ =
e im) : e
3 bey Ss 3

the Murree beds on the other. For instance, a mass of grey slaty
shales with associated red rocks, both containing Nummulites and
possessing the aspect of the Subathu portion of the Murree group,
occurs in the hills crossed from the Peshawur valley southwards by
the Meer-Kulan Pass, and lying far to the north of the abnormal
contact-line situated under the southern foot of Nilab Gash moun-

72 A. B. WYNNE ON SOME FEATURES IN THE

tain, on the Indus below Attock. Red and variegated gypseous
rocks resembling those in the Murree group previously mentioned,
occur far down in the ravine of the Haro river near its source, and at
Doongagully on the new mountain-road from Murree to Abbottabad,
while the possibility of the extension of the hill type of Nummu-
litic limestone southwards beneath the Pot’war country is supported
by its reappearance at the hill of Khairee Moorut, ten miles south
of the junction-line referred to, and west by south from Rawul
Pindi.

Although the line of junction presents ordinarily the appearance
of a single line of contact (often concealed owing to its position at
the foot of hills), this is not 2 constant condition ; for it is accompa-
nied in places by somewhat parallel lines of the same kind situated
within very short distances, and traceable into close convergence, if
not into absolute unity, with the main one (see section, fig.3). It
also throws off a branch north of Rawul Pindi, having all the
characters by which it is itself distinguished, and bringing the same
sort of Lower Murree rocks into junction with the Hill-limestones.
This branch, diverging at an acute angle from the north side of the
line, passes westward along the outer foot of the ridge behind
which are the ruins of the ancient Taxila at Shah-ka-deri.

The whole aspect of the contact along this line presents none of
the features or irregular outlines which might be expected to result
from unconformity, overlap, or limit of deposition; and, further,
it might be urged that nowhere along the junction are the newer
rocks on one side of the line or its branches made up of any recog-
nizable detritus of the older beds on the other side; nor, indeed,
are any prominent coarse detrital deposits, such as might indicate
an adjacent shore, to be found in the lower strata of the Murree
group, a few minor beds of coarse sandstone at one part of the
junction only (in section, fig. 3, 6") having been met with along the
whole region of contact from Oori to Kohat, a distance of fully 200
miles. Nor does it appear that the basal beds of the known Tertiary
deposits in the Simla area along their inner (and even more ex-
tended) boundary, differ in these respects from the junction beds of
the Tertiary belt with the “ Hill-rocks” in this region.

The contact here, on the contrary, presents straight or but slightly
curved lines, like those of fractured dislocation, without sinuosities
resembling deep bays, creeks, or promontories ; nor are there visible
signs of islands lying off the supposed ancient coast, the detached
elongated hill of Khairee Moorut being apparently enclosed between
lines of fault.

In a case of ordinary unconformity or overlap subjected to denu-
dation, discordant patches of the Murree beds might be expected to
occur resting on the older rocks ; but these have not been met with,
any detached portions which have been found being apparently let in
among the older rocks by faults, rather than folded into contortions.

16. Under all these circumstances it is difficult to recognize in
this and the Simla region the same relations among formations or
groups, several of which may be fairly considered identical and in

q Shah-durrah Valley, N.N.E. from Rawul Pindi.

ton crossin
(Length about 5 miles; highest point about 5000 feet.)

Fig. 3.—Sect

PHYSICAL GEOLOGY OF THE UPPER PUNJAB. 73

the same orographical position. Even in the Simla area a fault-like
character is attributed to the abnormal junction; and had we here
the transgressive overlap of different portions of the Tertiary series
on to the hill-rocks, it might be easier to assimilate the features of
the two areas and to admit for this district Mr. Medlicott’s view
that the junction-line is not one of fault but of unconformity.

N.N.W.

b'. Limestone bands in red gypseous and grey clays and shales,
N.N.W.

Probable faults of Shah-durrah.
6'. Gypseous and other clays and limestones

c. Hill- Nummulitic Limestone.

les; highest point about 5000 feet.)

=
S
>
<
SE
=
>
=~
S
Ss
—
=
=
. “SS
5 : f
3 A Cy
3 s Zz y
S ae iH
PI q @:
B 3 os 8 i
EI 5 S33 2.
NS °
3 E = so q
3 gee om
Ss BS Oo = |
iS a ce = oS
B a s S |
S) gq = =]
1 3 g
ce] q ~~ cas a Bee We
3 Ss Oa cs
ai ak
2 = = °
5 ane So g
= ~ q
2 aes 3
ze 2 8 oS ~)
es So 48
a) cS si BS
ee Z ne a ‘
EE, £3 | oa ¢ Z
ED oo - oN
ae ag ~~ a
OE 32) S| _ &)
Ze poli! pl =
pe Zz ae z La, d ae
< aff a eS eR ere
Ke aE = 3 8: — Sy
2?) J 8% a Bll eeeacetacvncotctate ad Wee
SS 3 2 ;
Ea a SPERM A NRE) SOE ieee
E S 3 j
2 5 ca]
pee [ea] u
faa = a
es

Placing the more northern contact-lines as indicated in the fore-
going observations, and those of the Simla area also, upon a map of
small scale, they seem, thus reduced, to possess considerable irregu-
larity, favouring the idea of unconformity, as in the case of the deep
sinus coinciding with the bend of the Jhilam at Mozufferabad ; but

ce. Hill- Nummulitic Limestone.

6. Nummulitic-Limestone bands.
(Nummulitic).

a. Murree beds.

74 A. B. WYNNE ON SOME FEATURES IN THE

it should be remembered that the depth of this bay is twenty miles
on its shortest side, lying within British territory, on the right
bank of the Jhilam, that this side is there nearly a straight line, and
as strongly resembles a fault separating highly contorted rocks as any
other portion of the contact. Examined in the same way, other parts
of the junction lose much of their apparent sinuosity.

17. The sudden change in the direction of the ranges and strike of
the rocks in the Jhilam valley has been already mentioned ; and in this
connexion it is somewhat singular to mark that on the further side
of that river from our district, in the tributary valley of the Poonch
(Kashmere), occurs the nearest instance of the unconformity con-
tended for by Mr. Medlicott, at a place called Dundelee. The strike
of the rocks (EK. 35°S.) belongs to the same system as that of the
Himalaya east of the Jhilam valley ; and the section as given in Mr.
Medlicott’s report is extracted (fig. 5) for comparison with those

Fig. 5.—Section at Dundelee, Poonch Valley.
(From ‘Mem. Geol. Surv. India,’ vol. ii. pl. 2, p. 96.)

c'. Kral group. c?. Infra-Krol. d. Subathu group.

within our district (figs. 2, 3, and 4) and lying about 45 miles to the
west-north-west.

The “hard blue limestone,” c', in this section, taken by Mr. Med-
licott to represent his Krol subdivision, if it is the same as that of
Mochipoora Mountain (as may be gathered from a footnote at p. 92
of the Report), is most probably Nummulitic, or might also include
some of the Triassic limestones of that range; at any rate it would
appear to belong to the Hill-limestone rocks. c’ is said to be com-
posed of “ thin carbonaceous slaty shales,” which, if they are the same
as those of Shah-durrah, mentioned in the succeeding passage, pro-
bably belong to the zone marked d* in the section fig.5; indeed
this might be inferred from the parallelism to the latter which
Mr. Medlicott mentions. d*, red sandstone and clays, and d*, num-
mulitic clays (and limestones ?), are evidently the same as those of
the Murree beds previously described under letter E. If the slaty
shales c* pass conformably below the limestone c', they probably
represent some of the many hard splintery shaly beds of the Mochi-
poora nummulitic rocks.

Of course, without having seen the Dundelee section, these remarks
upon it are given with reserve; still it is much nearer to our district
than to the Simla area, and this must be the excuse for an attempt
to interpret it according to Upper-Punjab experience.

rod

PHYSICAL GEOLOGY OF THE UPPER PUNJAB. 79

Independently of the question whether the beds c? in this section
belong to the central ridge or to the flanking rocks, those marked
@ and d° (the description of which so exactly coincides with the
Murree group) still appear unconformable to c’, the central limestone ;
and yet there is only wanting an apparently very probable line of fault
on each side to make this section resemble a modification of the same
kind of feature as is represented in our Shah-durrah and Chuttur
sections (figs. 3 and 4), where the discordance has no similarity to
unconformity.

It may also be instructive to compare this Dundelee section with
another in the Pot’war district, some 60 miles to the 8.W., taken
N.W. and §.E. across Diljubba Mountain, part of a chain of hills
which runs from the Salt range almost directly towards Dundelee,
but vanishes before reaching the Jhilam river. The strike of the
beds is parallel to that of the ridge; and the section is given from
memory.

Fig. 6.—Section across Diljubba Mountain.

N.W.

1. Salt-range series, including Cretaceous ?, Triassic, Silurian, and Salt marl &.
2. Nummulitic Limestone. 3. Grey Sandstones, Red Clays, &e. A (bones).
4. Red zone B. 5. Fault.

The Nummulitic limestone in this section has not its full develop-
ment; but it shows well the parallelism between this and the suc-
ceeding strata. The fault shown is the probable extension of a
long fracture, most strikingly seen further to the west; and no sign
of unconformity appears in the section.

Whether the movements connected with the origin of the Hima-
laya (as out-thrust produced by settlement, &c.), to some of which
Mr. Medlicott ascribes the succession of abnormal junctions of
the outer Tertiary zones, may have differed in kind, intensity, or in
number as well as in direction here, is as yet mere matter of specu-
lation ; but it is at least possible thaf in the neighbourhood of this
region the greater variety in the bearings of the vast mountain masses
may be accompanied by greater complexity in their structure.

18. The only attempt to explain the structure of the Tertiary belt
in this Upper Punjab region with which the writer is acquainted is
that given in Dr. Verchere’s paper before the Asiatic Society, where
a diagrammatic section occurs at p. 104,

76 A. B. WYNNE ON SOME FEATURES IN THE

In this the contorted lower Miocene (Murree) beds are made to
thin out towards the south, where they are overlapped by the Sivalik
or upper rocks, these also thinning out towards the bulk of the
Murree series, which latter is shown to rest unconformably upon
porphyry, felstone, &c. In support of this section there seems to be
nothing to advance. No trace has been found of the overlap of the
two series as shown; and the rocks against which the lower group
rests are not porphyry and felstone, but limestone and shale. This
is the case from Murree westward to Kohat; and even at Oori, in
Kashmere, the Murree group was found in junction with limestones,
as already mentioned, and the latter with metamorphosed schistose
rocks ; nor could the representation of the locality given in another of
the same author’s sections be recognized.

19. The general features of the outer Tertiary junction with the
hill-rocks will have been gathered from the foregoing remarks ; but
the difficulty remains of asserting the non-existence of unconformity
along a complex boundary, only because nothing of the kind can be
observed. On the other hand it is equally difficult to rest satisfied
from the evidence obtainable that the discordant junction marks a
limit of unconformable deposition. Reverting to the sections of this
district appended, there is nothing in them contrary to the exist-
ence of faulting ; the crucial-test section of Mr. Medlicott near the
Markunda river has no parallel among our observations ; and though
the unconformity which it exhibits (Report cited, p. 108) be unques-
tionable, there seems to be room left to doubt its applicability beyond
the boundary line on which it occurs, which is also that of the newest
subdivision beneath the alluvium as well as of the coarsest beds, most
palpably formed from detritus of the neighbouring rocks, among the
three groups of the Simla Tertiary series.

Nor does it appear reasonable to suppose that the basal contact of
these outer Himalayan Tertiary beds must have everywhere occurred
along a denuded cliff-line, which is a part of the supposition advanced
for the Simla area. If so, the continuous-cliff-line conditions would
in all probability have to be extended from the Simla area to the
Punjab, a distance of some 600 miles, or, for all we can tell, even to
the whole Indian frontage of the Himalaya range, or from this to the
Alps. It may fairly be asked, does any coast-line, ancient or modern,
present an unbroken cliff-line of this length, or one so little marked
by indentations ?

The circumstance that the line of abnormal contact is also a geo-
logical boundary limiting the main development of this part of the
Tertiary formation, at first sight greatly favours the idea of
unconformity. Supposing the Upper-Punjab contact, however, to
have been caused by displacement, there is nothing in this demanding
that the newer beds should continue persistently far beyond their
present general limits, or if they did even to a greater extent than
is known, the denudation of this country would have been amply
sufficient to have removed them from all prominent positions*. In

* Notwithstanding this, the writer’s impression must be stated that these beds
of the Tertiary formation had no widely extended.range over the Outer Hima-

PHYSICAL GEOLOGY OF THE UPPER PUNJAB. 77
the hilly and contorted limestone-and-slate country lying just north-
ward of the line of contact, there are two distinct unconformities (see
table) observable without difficulty, notwithstanding local contortion
and faulting ; and this being the case, it is not easy to conceive why
unconformity should be invisible along the sharply defined contact
of the hill-lhmestones with the outer Tertiary zone.

20. Having thus failed to find in this line of contact proof of either
conformity or unconformity of the adjacent rocks, it only remains to
glance at the general structure of the country, in order to see if there
are any points observable likely to aid in arriving at a fairly satis-
factory explanation of the facts.

A feature common to the structure of the whole of the Upper
Punjab is prevalence of lateral changes in the series of which it is
formed. Thus the great Nummulitic deposits of the Salt range die
out to the east, as do also the Carboniferous, the fossiliferous por-
tion of the Trias, and some other groups of that region ; while others,
again, disappear westward ; so that each end of the range presents
an almost entirely different section. In the hills to the north also
the Cretaceous rocks have a local development, the Trias of different
regions differs, and the character of the Jurassic beds is inconstant.

These changes are not, as a rule, connected with perceptible un-
conformity, the northern Jurassic group being the ‘“‘ exception which
proves the rule,” but seem to have resulted from gradual thinning
out of the beds, their original partial deposition in detached patches,
or difference of the rock-forming materials.

The outer Tertiary belt presents a sort of gradation or transition
towards the hill-character ; for we find among the rocks of the Murree
zone harder beds than elsewhere, limestones also occasionally appear,
and some of these are dark in colour, as the hill-beds usually are,
and contain in places the same kind of minute organisms, chiefly un-
determined Foraminifera. The prevalent reddish or purple colour of
the group changes too ; and grey shaly bands appear of quite the same
aspect as those of the hill-rocks, in the part of the series nearest to
the latter. None of these shales nor any alternations of the beds with
limestones characterize the basal portion of these newer Tertiary
layers in the vicinity of the Salt range.

The Nummulitic limestones of the Salt range, with their large
Bivalyes and Gasteropoda, have the appearance of shallow-water
origin, while the diminutive organisms of the hill- Nummulitic Lime-
stone would seem to have inhabited greater depths.

Another common feature of the country is contortion of its strata,
this disturbance having affected some of the very newest Tertiary
beds, so as to place them in a vertical position, and having almost
everywhere thrown the rocks into more or less complicated folds,

layan hill-country. Their great extent along the base of the Himalaya might
favour the supposition that they stretched also in other directions; but in a
country where we have hills 7000 ft. in altitude, entirely composed of these
newer Tertiary rocks, considerable masses of them might be expected to occur in
the lower and contorted limestone hilly ground north of the main boundary, if;
they had ever covered that region. Such masses are almost entirely absent..

Q.J.G.8. No. 118. H

18 A, B. WYNNE ON SOME FEATURES IN THE

which have in many cases been pushed over from the direction of the
mountains so as to produce inversion of the strata. As instances of
this, may be mentioned the complication near the mouth of the
Chichalee pass, westward of Kalabagh (trans-Indus)*, the convolu-
tions along the escarpment or slopes of the Affreedi hills overlook-
ing Kohat, and the well-marked case on the hill road from Shaal’-
ditta to Khanpoor, northwards from Rawul Pindi, where the Jurassic
series behind the outer ridge, with its characteristic band, one mass
of a large Trigoma (like T. ventricosa, Krauss), almost horizontally
overlies the Nummulitic limestone.

It must be admitted that, were these numerous foldings and con-
tortions expanded and restored to their original horizontality, or any ©
thing like it, the area occupied by the now contorted rocks would be
much greater than it is at present.

21. All geological formations being subject to lateral variability,
it would seem from the foregoing remarks that those of this district
are no exceptions to the rule, if, indeed, the variability is not un-
usually pronounced throughout the local formations. Hence it does
not seem improbable that the red Tertiary zone of the Salt range
may correspond to the Murree group of the northern side of the
Tertiary belt, so that, if we could eliminate the line of abnormal con-
tact bounding the latter, we should have an ascending sequence from
the Salt-range Nummulitic to the newest rocks of the formation on
_ that side of the belt, and to the north another aseending sequence
from the hill- Nummulitic Limestone, but one formed under different
conditions of depth of water, contemporaneous with the former, yet
only in a general way, neither succession being wholly without
evidence of interruption, in the small rolled pebbles of fossiliferous
Nummulitic limestone previously mentioned.

It is possible that the hill- Nummulitic Limestones alternated some-
what in their upper portion with the purple and grey Murree beds,
in favour of which the similarity of certain of the basal limestones
and shales to those of the hill- Nummulitic rocks, may be again
noticed as indicating a slow change of condition, although the two
groups are now separated by the most abrupt discordance, disturbance,
and displacement. This change, too, may have taken place over a
much wider area than that now occupied by the rocks, if the lateral
expansion caused by restoring the contorted beds to their original
horizontality be granted, thus enlarging the space within which
thinning out might have occurred.

With regard to a separation from an early Tertiary period between
the Salt range and northern hill-regions, this seems possible. The
nature of the outer Tertiary belt suggests a shallow-water origin; and
yet in the Murree beds alone a thickness (which no reason exists to
doubt) of a mile and a half, or nearly 8000 feet, has been observed,
while 10,000 or 15,000 feet does not appear at all too much for the
whole series: but this means 2500 fathoms; and the impossibility of

* Noticed by Dr. Fleming, who gives a somewhat imperfect illustrative section
across the inversion. See his report to Gov. of India, Journ. Asiat. Soc. Beng.
1853, pp, 266, 267.

- Vol. XXX. P).VIL

>

4

~

= Peshawuro
IC
= Line of an
2 °Kohat Sect 1
a i
x T 4

a
Egy ANION |
Wy

pss

M
THE NEWER

NORTH

F. Dangerfield lith.

lod

PHYSICAL GEOLOGY OF THE UPPER PUNJAB. 19

associating such a depth with the idea of a shallow sea or lake con-
strains one to call in the aid of subsidence to account for the accumu-
lation. Subsidence, however, can hardly be universal any more than
elevation ; so that an idea of limitation is suggested for the area re-
ceiving the deposits and consequent denudation round the border.
The first clear evidence of this in the case before us dates from a
considerable height in the post-Nummulitic series.

22. In conclusion it may be inferred from the foregoing observa-
tions that the physical relation between the outer Himalayan lime-
stones, &c., and the adjacent great Tertiary belt are obscure in the
Upper Punjaéb, and that the local facts hardly warrant the close
application of arguments used to explain some similar relations in
the Simla area. Palpable unconformity has not been found along
the region of contact, nor any positive evidence leading to the con-
viction that it exists.

Inversion of the lower strata of the outer Tertiary band is common
close to the line of contact; but this is not surprising, the violent
disturbance which has affected all the rocks having inverted some of
the hill-limestones within a comparatively short distance of the
junction. It would also appear that the Simla area possesses a struc-
ture more easily comprehended from the fact that the outer Tertiary
zone has a definite Nummulitic (Subathu) base resting on rocks of
uncertain age, while here Nummulitic rocks certainly occur on both
sides of the same outer Tertiary abnormal boundary. This abnormal
junction-line is evidently a feature inseparably connected with the
causation of the great mountain-chains, preserving, as it does, a
parallelism to the axes of the outer ranges; and whether it marks
an invisible break in the series or not, the conviction is strongly
impressed on the observer that it is chiefly due to intensity of distur-
bance amounting to fracture and dislocation, combined with more or
less complicated inversion, the result of lateral pressure along a
region which has suffered, and from the same cause, a similar thrust
to that affecting the place known as the spring of any ordinary
arch.

The detailed examination of the upper Punjab has not been com-
pleted ; and only the general facts bearing upon the peculiar junction-
boundary of a large and important group of the local rocks have been
considered, with the impression that it is only by collecting such facts
and comparing them with others similarly obtained that the nature
and value of obscure points in physical geology can be arrived at.

EXPLANATION OF PLATE VII.
Sketch Map showing the Newer Tertiary boundary in North-western India.

Discussion.

Mr. Drew gave some further explanution of the author’s views

and of the geology of the district.
Prof. Ramsay agreed with the author as to the analogy between
H 2

Quart . Journ. Geol. Soc. Vol. XXX. Pl.VIL

MAP SHOWING

THE NEWER TERTIARY BOUNDARY

in

NORTH-WESTERN INDIA

80 A.B. WYNNE ON THE PHYSICAL GEOLOGY OF THE UPPER PUNJAB.

the outer ridges of the Himalayas and those of the Alps, and also
as to the difficulty in such cases of distinguishing between an in-
version and a fault. From the different conditions as to metamor-
phism of the Miocene and Kocene rocks in the Alps, he had been
led to question the fact of the inversion, and now found that most
Swiss geologists had come round to the opinion that the present
position of the beds was not due to mere inversion but to actual
dislocation. He thought this might also prove to be the case in
the Himalayas.

A. CG. RAMSAY ON THE VALLEY OF THE RHINE. 81

13. The Puystcat History of the Vauiny of the Ruins. By A.C.
Ramsay, LL.D., V.P.R.S. (Read February 4, 1874.)

[Puate VITI.]

Havine on several journeys, extending over more than twenty
years, had occasion to traverse the valley of the Rhine, from the
plains between the sea and Cologne to its sources in the Alps, I
have long desired to explain, if possible, the physical origin of the
valley, and especially of the part that lies between the Drachenfels
and Basel. Other occupations for the last thirteen years prevented
me from revisiting the greater part of the ground; and I had there-
fore no opportunity of testing the truth of such hypotheses as had
occurred to me.

Last summer I revisited the country, spending nearly six weeks
on the investigation of the subject; and the following paper em-
bodies the broader conclusions at which I arrived. First, however,
it may be useful briefly to sketch the great features of the valley of
the Rhine, from its sources to its mouth.

The Rhine is usually stated to have two principal sources—that
of the Vorder-Rhein (in a glacier region), 7689 feet, and that of
the Hinter-Rhein, where it springs from the Rheinwald glacier,
7268 feet above the level of the sea. Both are in the Canton of
Glarus. The accumulating waters, swelled by many tributary
streams, pass through the Lake of Constance, 1305 feet above
the sea; and thence, hemmed in between the lower slopes of the
Schwarzwald and the eastern end of the Jura, the river flows
westerly as far as Basel, where it is 803 feet above the sea. A
little above that city it escapes from the immediate neighbourhood
of the hills, and, suddenly bending, flows northerly through the
great gravelly inclined plain that les between the mountains of the
Vosges and the Hardt and those of the Schwarzwald and the
Odenwald. ‘This plain extends north for about 170 miles to Mainz
and Bingen, where it is about 272 feet above the sea, showing a fall
between Basel and Mainz of 531 feet. The average slope of the
plain is therefore 0° 2’ 3”, or about 3 feet 14 inch per mile. In
its broader parts the plain is from twenty-five to thirty miles in
width.

At Bingen, the Rhine suddenly enters the gorge, which, inclusive
of the larger windings of the river, is about sixty miles in length,
as far as Rolandseck, near which a second great plain of the Rhine
begins, at first narrow, but below the Siebengebirge broadening
rapidly, to form still lower down the modern delta which merges
in the vast plains that bound the German Ocean between Calais
and the Elbe.

The chief questions which I now attempt to decide are, what is
the origin of the present features of the plain that extends be-
tween Basel and Mainz, and also of the gorge between Bingen and

82 A.C. RAMSAY ON THE PHYSICAL HISTORY

Rolandseck, through which the waters of the Rhine escape into the
lower plain ?

Looking south from the hills above Bingen, or from the northern
slopes of the Taunus, the impression readily rises in the mind that
the vast plain, bounded on all sides by hills, must have been a lake
at a very recent geological period, and that the opening of the gorge
was the cause of the drainage of this supposed body of water 170
miles in length. Several observers have entertained this idea, at
least in conversation with me; and this popular notion may be
found printed in Baedeker’s Guide-book. The opening of the gorge
was formerly sometimes attributed to the disturbance and wide frac-
ture of the Devonian strata through which the river runs, though it
is unlikely that this view is now at all generally held.

For many years I also inclined to the hypothesis that the long
plain above the gorge might have been a lake in geological times
comparatively recent. I also thought it not unlikely that the action
of glaciers might at least have assisted in the work of excavating
the hollow, bordered as it is on three sides by well-known old
glacier-regions—Switzerland on the south, the Schwarzwald on the
east, and the Vosges on the west; and this year I determined to put
the whole question to the proof.

As I proceeded with the work, I could find no signs of moraine
matter on the slopes near Bingen and Mainz, no evidence that
the glaciers of the Schwarzwald and the Vosges had ever been
large enough to descend into the plains of the Rhine, or that the
great old glaciers of Switzerland had ever reached the Rhine valley
as far as Basel; and therefore glaciers seem to have had nothing to —
do with the excavation of the hollow. Point by point, all the other
circumstances that might have helped to support a post-Miocene
lake-theory also gave way; and beginning anew with fresher and
sounder data, founded on more accurate knowledge of the physical
geography and geology of the area than I previously possessed, I at
length arrived at certain definite conclusions.

In connexion with the subject, it is of importance to endeavour to
ascertain approximately the geological date at which the waters of the
Rhine began to flow in their present course ; and to do this it is neces-
sary to state briefly some of the later events relating to the history of
the Alps, and of the Miocene strata of Switzerland and the Rhine.

It is well known that the Alps, as a mountain-range, existed in
some form before and during the deposition of the Miocene strata
(see Map, fig. 2); and considering the length of time that has elapsed
since the beginning of that epoch in those regions, and the amount
of waste that all the preexisting Alpine rocks have undergone
during and since that period, it may perhaps be assumed, notwith-
standing the subsequent elevation that attended the disturbance of
the Miocene rocks, that relatively to the Miocene deposits the Alps
may have been during Miocene times as high above the level of the
sea as they are now.

One of the most striking proofs of the waste of these older Alps
during Miocene times is found in the well-known conglomerates or

OF THE VALLEY OF THE RHINE. 83

Nagelfiuh of the Righi, the Rossberg, and the other subalpine hills
that flank the older mountains, continuously over great spaces, and
at intervals elsewhere, all the way from the Rhine near its entrance
into the Lake of Constance to the Lake of Thun, and which again
appear on the north side of the Lake of Geneva, near Vevey
(Map, Pl. VII1). Nor are these conglomerates confined to one precise
geological horizon.

The statement also is often made that, during the deposition of
the Miocene strata, the Jura as a mountain-range had no existence.
This is a slight overstatement of the case; but, without going
into details, it is sufficient for my present purpose to menticn
that in the opinion of Mr. Merian, of Basel, and, I believe, of other
Swiss geologists, the range of the Jura in a very rudimentary stage
preceded the deposition of the Miocene strata—so far, that the pre-
Miocene formations had been somewhat disturbed and considerably
denuded, so as to form a kind of low undulating plain, afterwards
entirely buried in the Miocene waters.

During the deposition of the Miocene strata the whole of the region
now occupied by the Jura, and of the area that lies between that
range and the Oberland, and far to the east into Bavaria and the
dominions of Austria, was generally but little raised above the level
of the sea and frequently depressed below it, as witnessed by the
occurrence of marine beds among the Miocene freshwater strata,
not in one middle band or subformation, as is sometimes stated, but
in several occasional interstratifications, by no means universally
spread throughout all the Swiss area (Map, fig. 2).

There is also every reason for the belief that the patches of
Miocene rocks in the Jura (Map, Pl. VIII), both freshwater and
marine, were originally united to the larger mass that forms the low-
lands of Switzerland. On the elevation of the whole territory, and the
more special disturbance and denudations consequent on the uprising
of the Jura, the Molasse that lay on the Mesozoic formations of that
range was converted into a series of outliers still existing in the
large synclinal basin-shaped hollows, which, as broad mosses and
green meadows, so frequently form the upper valleys of the hilly
land of the area, and which have been so well described by Professor
Desor. These outliers of one original stretch of Molasse thus easily
carry us to the confines of Basel. If this be true, it establishes
a direct original continuation of the Miocene strata with those of
the basin of Mainz, between Basel and Mainz, including in the
word Miocene (for the sake of brevity) all the subformations classed
by some geologists, both in Switzerland and Germany, under the
terms Oligocene and Miocene (Map, Pl. VIII).

These Miocene strata form much of the country bordering the
plain of the Rhine, near Basel and thence to Mulhausen and
still further west, rising in high tabular masses of flat-lying strata
which overlook the broad flats of alluvial gravel that lie between
the Schwarzwald and the southern end of the Vosges. Thence
they skirt the plain on the right bank of the river as far north as
Freiburg ; and on the left they appear at intervals to the west of

84 A. C. RAMSAY ON THE PHYSICAL HISTORY

the alluvium, all the way from Mulhausen to Mainz and the neigh-
bourhood of Bingen. Between the two last-named towns and the
country west of Worms they form broad tablelands, in great part
consisting of flat, or nearly flat, strata of limestone, rising in places
from 300 to about 450 feet above the plain of the Rhine (as shown
in fig. 3, marked a). An isolated patch, surrounded by alluvium,
according to Von Dechen’s map, rises in the very middle of the
great plain west of Carlsruhe, rather more than half way between
Basel and Mainz.

Taking all these circumstances into account, there is good reason
for the belief that the whole of that part of Germany which is now
the valley of the Rhine, between Basel and Mainz, was once filled
with Miocene strata (Map, fig. 2), the precise thickness of which
is to me unknown. If we consider what is now concealed by the
alluvium (Map, Pl. VIII.), and the thickness of the exposed tabular
- masses irrespectively of the removal of any overlying beds by denu-
dation, then the original plain must have formed the surface of a
set of strata from 300 to 500 feet thick, and probably more.

During their deposition, and when completed, these strata, to-
gether with those of Switzerland, were bounded on the south by the
pre-Miocene Alps, on the east by the mountains of the Schwarzwald
and the Odenwald, on the west by the Vosges, and on the north by
the hilly Devonian plateau, of which the mountains of the Taunus
form part, and through which the gorge of the Rhine now passes
between Bingen and Rolandseck (see Map, fig. 2). This deep gorge,
however, had then no existence; and at that time the Rhine had not
begun to flow ; for the drainage of the area under review was partly
in the contrary direction, or from north to south, during the deposi-
tion of the Miocene freshwater strata of the basin of Mainz and the
northern part of Switzerland. The facts as understood by some of
the continental geologists are, that the freshwater Miocene rocks be-
tween Mainz and Basel were altogether deposited, not in a great
lake or lakes which were sometimes invaded by the sea, but rather
in the alternating freshwater and marine beds of a river and river-
mouth, in this respect resembling the estuarine interstratifications
of the Purbeck, Wealden, and fluvio-marine Eocene formations of
the Hampshire basin. In certain strata the pebbly beds of this
Miocene river can be traced all the way from a northern tract formed
partly by the present strata of the Kaiserstuhl, whence brooks
flowed to mingle their waters with tributary streams bearing gravels
coming from those parts of the old Miocene continent now called the
Schwarzwald and the Vosges. These conglomerate beds, with dis-
tinctive pebbles showing whence they came, can be traced as far as
Delsberg, in the Bernese Jura, having always travelled south.

The foregoing summary of events of the Miocene age is necessary, -
otherwise the second part of my argument would not be clearly
intelligible. ina

The upheaval and general disturbance of the Miocene strata over
Jarge Kuropean areas* altered the physical geography of the older

* And others with which we are not concerned in this paper.

OF THE VALLEY OF THE RHINE. 85

continent in a remarkable manner. The Alps received a new eleva-
tion along with the adjacent Miocene rocks, which, previously not
much above the level of the sea, now, in the heart of Switzerland,
vary from about 1200 to 5800 feet in height. In lke manner,
though elevated in less degree, and much less disturbed as
regards curvature of the strata, the Miocene beds that lie between
Basel and Mainz were also upheaved; so that now those strata attain
elevations of at least 1000 feet above the sea near Basel, and
probably of not less than from 700 to 800 feet near Mainz.

The general result was, that in all those parts of Central and
Northern Europe with which I have now to deal the systems of
drainage were wonderfully altered, and in some cases almost entirely
remodelled.

In Switzerland the amount of disturbance and elevation of the
Miocene strata having been greater than that of the Miocene beds
in the valley of the Rhine, the effect of watery erosion in the Alps
must have been proportionate during all the time that the Crag
formations of England and Belgium were being deposited; and during
this period more than the rudiments of many of those valleys were
formed that afterwards were filled by and, over much of the lowlands,
entirely buried beneath the great glaciers of Switzerland, when the
cold of the glacial epoch attained its maximum.

Along the lines of these early valleys, between the Oberland and
the Jura, such as that of the Aar, there can be no doubt that the
glacier sheets lay thickest, and the amount of glacial erosion must
therefore have been greatest; and very important changes in the
outlines of hili and valley were the result. Some of these I have
long ago discussed, and J will not renew the subject; but this seems
certain, that, after the post-Miocene upheaval of the Alps, the
present main drainage of the area began before the glacial episode,
and was in many important respects only established by the influence
of glaciers. The upper valley of the Rhine was at that time filled
by a river of ice which, joined with others, covered the valley more
than halfway from Schaffhausen to Basel, while from its western
edge and end the liquid river followed its present general direction,
hemmed in between the Jura on the south* and the lower hills of the
Schwarzwald that bounded the river on the north. Its level must
then have been higher than now in this part of the channel.

While these were some of the effects of the post-Miocene upheaval
of the Alps and the lowlands of Switzerland, including the Jura, it
is also my opinion that by the same upheaval the contiguous area of
the Rhine between Basel and Mainz was tilted, so that the main
drainage in what is now the basin of Basel and Mainz flowed in a di-
rection opposite to that which it followed during the Miocene epoch ;
that is to say, the new tilting of the country (fig. 1) ended in pro-
ducing a thorough drainage that flowed from south to north instead
of from north to south.

About a hundred miles east of Basel, following the windings of
the Rhine, the Lake of Constance is 1305 feet above the sea, and

* Or, rather, the northern continuation of the rocks of the Jura.

86

A.C. RAMSAY ON THE PHYSICAL HISTORY

502 feet above the average level of the river at the bridge of Basel.
Assuming, as the data seem to warrant, that the basin of Basel and

Fig. 1.—Hypothetical Section along the Valley of the Rhine.

ML

ZL.

TT

= Teo Le ee
TT EL IL ILL IIL ITT TII TIT TILL IPATIT IT ITIL TTT

PARK nem EEE EERE RE RP ID

1. Post-Miocene hills of Switzerland on the south (Jura &c.).

==
——

4, Plain of Miocene strata inclined from south to

2. Pre-Miocene rocks where the gorge now begins near Bingen, on the

3. Mountains of the Schwarzwald &c. on the east side of the valley.
north, over which the Rhine flowed, and bounded on the west by the mountains of the Vosges.

north.

Mainz was filled with Miocene strata to a height
of more than 300 feet at Basel above the present
plain of the Rhine, there was still ample fall for
the river springing from the end of the old Rhine
glacier, or even at a later date from the level of
the Lake of Constance, for the Rhine to have
flowed along the inclined plain of Miocene rocks
that, at levels of at least from 300 to 500 feet
above the present plain, once extended from
Basel to Mainz and the neighbourhood of Bin-
gen. I have already stated that the present.
fall from Basel to Mainz is 531 feet. In the
older epoch adverted to, even allowing the Mio-
cene strata of Mainz to have attained a height
of 500 feet above the level of the present plain,
there may still have been a fall in the river-
channel on the surface of the ancient plain of
more than 300 feet between Basel and Bingen.
This is equal to about 1 foot 9 inches per mile,
an inclination amply sufficient to have carried
forward at a rapid rate the waters of the Rhine.
At this epoch of the history of the river, the
general arrangement of the rocks must have
been somewhat of the kind shown in fig. 1.

In this manner I consider that it must have
happened that the river originally flowed along
the surface of a high inclined plain of Miocene
strata that then filled the valley of the Rhine,
from the confines of Switzerland to what is
now the southern end of the gorge. When this
system of drainage began, the deep river-gorge
that now separates the Taunus and the Hunds-
ruck had no existence (see Map, fig. 2). The
hilly slopes on either side were, so to speak, un-
broken; and a mere minor shallow valley, the
bottom of which was as high as the edge of the
present gorge, received and carried northward
the waters of the Rhine. By degrees the great
river flowing at this high level began to scoop
out a channel, which gradually deepened and
became bounded by cliffs. Just in proportion
as the channel of the gorge deepened, so the
Rhine, wandering through the long inclined
plain of Miocene rocks between Basel and Bin-
gen, by degrees had its surface-level lowered
through the ordinary processes of watery ero-
sion. The wide alluvial flat through which the
river, studded with islands, now flows, shows

87

oO

OF THE VALLEY OF THE RHINE.

that in modern times the Rhine has often changed its channel, winding
hither and thither all across the plain between the Schwarzwald and
the Vosges, and leaving terraces sometimes remote from its present
banks. As it acted then and is doing now, so it must have acted
in the post-Miocene times recorded in this memoir, till at length

Fig. 2.— Map showing the Distribution of Miocene strata north of the
Pre-Miocene Alps, in the area now occupied by the Rhine between

Basel and Mainz.

7 Me < vi
4 4 Hr & =
Ww y 4 “An nt i) aw
“zz Nin AD ——
SE ny =
=; BZ Sql Ss
3, 7 Wwe My — 2 x
Z =BeE f- G
Ayo Gl mnt
Ys aw == SS
Ue S Si aay, z
2g me,
I N& = wy Ly, mane Rg.
I} = ay
f ae W SS Os <
Miz iS fe gt
@ AN ( AS _F iim) YS AY =
RS Y, Saari" 5 AS ——=
i ng AS l= ell ly Cz S>
A
Wis “BX SES
=F = “ SSeS
2 wil z 1) SSS
NS
MnwZ S Zoe S SSS SSS
= ‘CS —= G ——SSSSSSSSSSSSS=——
< WS an DAW

YY;
YY
UP
S
=
SS

Ny

DD

AG i
= ae i il
<<" ii ee Wit ~~ <

are!

s 2«& SS
ZS SALES

SN
B, C, D. Pre-Miocene mountains (Schwarzwald, Vosges,
BE. Pre-Miocene Alps, bordering Mio-

Mn

klilide Wy,
)

iy Wa

AA. Miocene strata.
&c.), bordering Miocene waters.

cene waters.
(The arrow indicates the direction of the drainage in part of what is now the

Rhine valley.)

the greater part of the high plain of Miocene strata was worn away ;
and the tabular uplands near Basel and Mainz now bordering the

modern plain remain to attest the amount of denudation that the

valley has suffered by watery erosion.

This hypothesis readily accords with the occurrence of stratified
banks of gravel at various levels above those of the more modern
river-terraces which are common in the plain. If also much of the
Loess be only river-mud of a comparatively ancient date, and per-
haps partly of glacier origin, this view helps to account for its

A, C. RAMSAY ON THE PHYSICAL HISTORY

88

‘sutyrd of} Sutsepaoq syoor ouooorpy Jo S[piy aepnqey, *vv

“U

abuag fo ymos syooyy auaooy fo spp mjnqny, puv ouysy ayn fo 2.

MM I—

6

cA

OF THE VALLEY OF THE RHINE. 89

being found at many different levels on the slopes that flank the
Rhine valley, far better than that of a hypothetical partial submer-
gence of the country which converted a large part of the valley into
a kind of freshwater estuary. Thus the Jurassic hills between Her-
holz and Ettenheim on the right bank are partly covered by these
old river-deposits to heights of about 200 feet above the river, while
lower down, as at Worms, the Loess rests on river-gravel close to the
great plain ; and at Eltville the Loess covered with vineyards descends
to the level of the Rhine. As the river by degrees lowered the level
of the plain, it left its finer detritus at these and many other levels.

While this plain still retained its original high level, and the
Rhine, as already stated, flowed through the upland slopes formed
of Devonian rocks now lying between Bingen and Konigswinter,
it first began to form the gorge; and as this work of watery
erosion went on, the water, constantly deepening its channel, at
length scooped it to its present depth.

The traces of its temporary levels as the river cut its way down,
may still be seen on the cliffs high above the present surface of the
water. Thus on the hill behind Bingen, called the Rochusberg,
on the spur of Devonian quartz rock on which the Hotel Hartmann
stands, there is a chapel, called St. Roche, standing on the relics of a
plain 341 feet above the river, and which a little further west rises
to a height of nearly 400 feet. This plateau, also in a fragmentary
shape, is continued further down the Rhine. Where the gorge
begins, going down the river, it presents the outline shown in fig. 4.

Fig. 4.—Section across the Rhine Valley, showing the present and old
Levels.

/)

Tinimnnnnn
1. The present level of the Rhine. 2. Part of an old level of the river.
3. Part of the hilly ground beyond, that formed the original river-bank.

Lower down the river, below Niederheimbach, the left bank, looked
at in front, presents the aspect shown in fig. 5.

The plain, 3, slopes gently to the north; and numerous gullies or
minor valleys, which open out just above the present bank of the
Rhine and end in the plain above, have been produced by atmospheric
decomposition and rain and snow during the time that the river
has been cutting its way from 3, its higher, to 1,its present level.

At Wellmich, looking down the river, the general appearance of
the high banks is given in fig. 6, which shows the terraced plain on
the left bank receding to the north in a gradual perspective, the
more hilly ground to the right and left that bounds these terraced
plains not being visible from this point of view.

90 A. C. RAMSAY ON THE PHYSICAL HISTORY

Fig. 5.—View of bank near Niederheimbach.

—=

1. The present level of the Rhine. 2. The cliffy ground of the west side of the
gorge. 3. Anoldriver-plain. 4. The more distant hills that bounded the

earlier river-valley.

Fig. 6.— View near Wellmich.

Fig. 7.—View on the left bank of the Rhine, near Salzig.
a ae — ~

Fig. 7 represents the left bank of the Rhine at Salzig, still further
down. The lower part (No. 1) is cliffy, above which there is the
same kind of plain, 2, gently inclined northward, and which like
those already noticed marks an old level of the river about 400 feet
above its present channel. No. 3 marks the undulating hills
beyond, as in fig. 5. *

The general aspect of much of the country between Coblenz and
Bingen is given in the following rough sketch of part of the scenery,

looking up the river (fig. 8).
Fig. 8.—View, looking up the Rhine, between Coblenz and Bingen.

In this diagram high terraced plains are well marked, and the
same kind of form is occasionally repeated at different levels. The

OF THE VALLEY OF THE RHINE, 91

same general outline is seen near Rheineck (fig. 9), where the terraced
form above the river cliff is well preserved. Finally, just about the

Fig. 9.— View near Rheineck.

mouth of the gorge above the Siebengebirge, looking up the river,
the long hilly slopes on the coast are seen descending towards the
Rhine, as in fig. 10, ending in a terrace (a) similar in height and

Fig. 10.—View near the Siebengebirge, looking south.

general character to those previously mentioned. In fact from end
to end of the gorge there are constant recurrences of these forms,
on approximately corresponding levels above the Rhine, and at other
elevations besides.

As the gorge was being gradually cut out and deepened, and as a
consequence of this the Rhine, wandering through the plain beyond
Bingen, by degrees lowered the surface of that broad part of the
valley ; so just in proportion the Maine, the Neckar, the Murg, the
Kinzig, the Elz, and other tributary rivers also lowered their
channels: in other words, when the Rhine flowed at a higher level,
the valleys of the tributary streams were also proportionately higher ;
and this remark equally applies to the tributaries of the Rhine on
either side of the gorge, such as the Lahn, the Moselle, and many
smaller streams. By this means we arrive at the post-Miocene
history of the deepening of river-valleys over very large areas; and
the reasoning now applied to the Rhine is equally applicable to
the Danube and other European rivers of equal importance.

One other minor point remains with regard to the history of the
Rhine. When the great Rhine-glacier, aided by tributary streams
of ice, spread westward in the valley some distance below the junc-
tion of the Rhine and the Aar, vast quantities of moraine-matter
must have been shed from its western end or edge. The large
bodies of water that then flowed from a glacier so enormous
carried great part of this moraine-matter down the course of the

»

92 A, C. RAMSAY ON THE PHYSICAL HISTORY

river, just as the streams that issue from the smaller glaciers of
to-day attack the terminal moraines and restrict their growth.

A necessary consequence of the powerful flow of such a large
body of glacier-water must have been to carry the waterworn stones
onward to Basel and into the flats beyond; and in time, as the river
changed its channel and wandered hither and thither across the
plain, the gravels got scattered over the whole of its area. The
thick strata of sand, loam, and gravel that form the present plain of
the Rhine are therefore in great part the waterworn débris of old
moraines, just as the gravels of the plains of Piedmont and Lom-
bardy are relics of the moraines of the gigantic glaciers of the
Italian side of the Alps. ‘This view of the origin of the gravels of
the Rhine was pointed out to me thirteen years ago by Professor
Desor, of Neuchatel, though I do not recoilect that he has printed
any thing on the subject.

The substance of the foregoing remarks may be summed up as
follows :—

1. During portions of the Miocene epoch the drainage through
part of the valley that lies between the Schwarzwald and the Vosges
was in great part from north to south, or, in other words, from the
hills north of Mainz into the area now occupied by the Miocene rocks
of Switzerland.

2. After those physical disturbances and elevations that closed
the so-called Miocene epoch in these regions, the direction of the
drainage was reversed, and thus it happened that :—

3. After passing through the hill-country between the lake of
Constance and the place where Basel now stands, the Rhine flowed
along an elevated plain formed of Miocene rocks, the relics of which
still exist between Basel and Mainz.

4. At the same time the Rhine flowed in a minor valley through

the upland country formed of the Devonian rocks that now con-
stitute the Taunus, the Hundsruck, and the contiguous high land’

lying northerly towards Bonn.

5. Then by the ordinary erosive action of the great river the gorge
was gradually formed and deepened to its present level, and—

6. Just in proportion as the gorge deepened, so the gently inclined
Miocene strata of the area between Mainz and Basel were also in
great part worn away, so as to leave the existing plain, which to
the uninstructed eye presents the deceptive appearance of once
having been occupied by a great lake*.

* Since this memoir was sent to the Geological Society I have read a learned
paper sent to me by Professor Fridolin Sandberger “On the Upper Rhine
Valley in Tertiary and Diluvial Times” (Das Ausland, No. 50, Dec. 15, 1873).

This memoir, which is of the highest value, contains a great amount of infor-
mation on the relations, stratigraphical and paleontological, of all the formations
found in the valley of the Rhine from the Upper Bunter Sandstone down to
the times of the Loess and superficial gravels.

Professor Sandberger only here and there incidentally touches upon the physi-
cal questions to which I confine myself, and apparently had no intention of
going into the details by means of which I attempt to prove what I conceive to be
the physical history of the valley, especially in its later stages. Wherever he
does touch on these subjects, however, there is no discrepancy in our views. On

¢

Soc Vol. XXX PL Vil

AP

SENT DISTRIBUTION
DF
STRATA,

pst Muocene-Alps,
N THE
land Mam,

re ss was
pane in

we et

OF THE VALLEY OF THE RHINE. 93

EXPLANATION OF PLATE VIII.

Map showing the present distribution of Miocene strata north of the post-
Miocene Alps, and in the basin of Basel and Mainz.

the contrary there seems to be agreement, a circumstance on which I may be
permitted to congratulate myself. With Professor Sandberger’s permission I
trust soon to publish a translation of his memoir.

From M. Merian, of Basel, and Professor Studer, of Berne, I received much
valuable collateral information, which helped me to work out the theory pro-
pounded in this memoir.

MM. Dufrénoy and Elie de Beaumont attribute the original formation of the
plain of the Rhine, between the Vosges and the Schwarzwald, to the sinking of
the highest part of a long. gently curved arch formed by the ancient junction of
the rocks of these mountains above the space now occupied by the valley. This
space, according to these distinguished authors, was let down by a series of
faults, which produced the scarped sides of the mountains on either side of the
great valley of the Rhine (Explication de la Carte Géologique de la France,
vol. i. p. 437).

Dr. Hibbert, in his celebrated ‘ History of the Extinct Volcanos of the Basin
of Neuwied on the Lower Rhine,’ 1832, states the theory propounded by M. Elie
de Beaumont, and seems partly to consider that, previous to the deposition of
the Tertiary deposits, the solid strata of the Rhine gorge “would be shaken to
their very foundation; while during their very forcible elevation deep fissures
or, according to a late phraseology, deep valleys of disruption (vallées d’écarte-
iment) would ensue, . . . . in which case, owing to the attriting power of torrents
contained in such fissures during an incalculable period of time, it would not
be always easy to distinguish such appearances as have originated amidst the
convulsive effects of uplifting causes, from those which have been caused during
an incalculable lapse of ages by the persistent degradation of meteoric agents.”
He thinks that the Moselle, with all its windings, has ‘‘the close resemblance of
a deep fissure or split,” and the same of the Brihl and the Nette—that the gorge
of the Rhine between Andernach and Rheineck, and thence towards Cologne,
was formed ‘“‘ by the same common convulsion,” and, indeed, “that the valley of

the Rhine, as far even as from Bingen to Bonn, was originally an immense:

fissure, suddenly induced by causes of violence ;” and to this and other fissured
valleys he attributes different dates. All of these were antecedent to the Ter-
tiary period, or probably so (pp. 10-14).

Again, following M. Boué, he considers “the valley from Mayence to Basel
one of the ancient marine basins of Europe during the Tertiary period,” and
that a barrier of high land stretched across “‘the present site of the Straits of
Bingen,” so that “this marine basin had no connexion with the present channel
of the Rhine from Bingen to Cologne.”

“Tts waters flowed in a direction quite opposite to that which they now main-
tain, being from north to south, while its southern extremity [about Basel, I
presume] was connected with the other marine basins of Europe by means of
narrow channels.” ‘The more modern view is that it was directly and broadly
connected with the freshwater and occasional marine interstratifications of the
Swiss Miocene strata, which, quoting from M. Boué, he states to consist of
marine and freshwater débris of shells &e. confusedly intermingled by frequent
débicles(p.17). The basin of Neuwied Dr. Hibbert also considers to have been
a lake which was partly drained by the convulsively disrupted fissure which now
forms the gorge of the Rhine between Andernach and Linz; and this drainage
of the lake was helped by ‘‘the fissured channel of the Brith], which was probably
a remote effect of the volcanic eruptions which in this district distinguished the
commencement of the Tertiary epoch” (p. 19). The different statements seem
to me to be not quite consistent with each other; but they may be partly under-
stood as rather expressing the opinions of M. de Beaumont and M. Boué than
Dr. Hibbert’s own opinion. Further explanation is perhaps wanted in my
memoir respecting the basin of Neuwied and the valley of the Brihl; but I may
return to this subject in a future memoir, ;

Q.J.G.8. No. 118. I

- A Theron

badd

WAP
SHOWING THE PRESENT DISTRIBUTION
’
MIOCENE STRATA

North of the Tost Mocane Alps,
aan *

OPE

n aia

(

SRO

| Miocene Strata

94 A.C. RAMSAY ON THE PHYSICAL HISTORY

Discussion.

Mr. Camppett, in illustration of the paper, and of the effects
which running water is capable of producing, brought forward some
views in Daghestdn and the Caucasus. Very considerable tracts of
this country are drained by cafons, very narrow, but some hundreds
of feet in depth; the streams which pass along them are principally
fed by melting snow. |

Prof. Hucurs doubted the view which referred the present Rhine
to any configuration of the country due to the original level surface
of the Miocene while it ignored the post-Miocene elevation of the
Alps and Jura. He had traced the lava from the Forniche Kopf,
near Andernach, down to within 100 feet of the present Rhine. He
thought, therefore, that the Rhine began to cut back a channel by
rapids and waterfalls, from the Rolandseck end, at the first appear-
ance of the land above the Miocene sea, and before the close of the
earth-movements aud volcanic eruptions which seem to be con-
nected with the later Miocene, but that it took the river so long to
eat its way back as far as Bingen that it had time to wind about
and form a broad valley in the upper part of its course—that the
river continued to run at the higher level at the Bingen end until,
at a comparatively far less remote period, the gorge was eaten back
through this flat valley; and when once the river had got through
the harder rocks near Bingen, it soon disposed of the softer beds in
the Mayence basin.

Mr. Kocu was inclined to think that the Jura chain had existence
in Miocene times, as beds of that age lay horizontally in hollows in
the Jurassic rocks. He quite agreed with Prof. Ramsay as to the
terraces along the sides of the Rhine gorge.

Mr. Tippeman did not see the difficulties suggested by Prof.
Hughes.

The Duke of Areyri remarked that if the physical history of the
Rhine were established, it would throw much light on that of other
great rivers. He considered the paper of great value, but suggested
a doubt whether the history assigned would account for all the phe-
nomena exhibited along the course of the river. There appeared to
have been at the commencement of the existence of the river several
upheavals of the land, but subsequent to the main upheaval of the
Alps. He regarded the channel of the river as entirely due to flu-
yiatile nailer. but he confessed to having some doubt whether the
erosion of the river alone could have effected the enormous amount
of denudation exhibited in the district. He inquired whether there
was not a possibility of some marine denuding power also having
been at work. He also inquired as to the flexure of the rocks and
subterranean movements affecting the conditions of the case, espe-
cially in connexion with the angles at which the Miocene strata are
tilted. A third question alluded to by Prof. Hughes was as to the
volcanic action that had gone on in the valley of the Rhine; and he
wished to know whether that might not also have conduced towards
the present contour of the valley. As to the terraces, he accepted

OF THE VALLRY OF THE RHINE. 95

the author’s view as correct; but the question still remained as to
how far the gorge was due to gradual erosion and how far to some
subterranean action.

Prof. Ramsay, in reply to Mr. Hugh Miller, stated that the de-
tritus brought from the north was found in Switzerland, and that
Prof. Sandberger considered that at the time of its transport the
Jura was not in existence. He had conversed with all the principal
Swiss geologists; and it was from thern he learnt that, along what
is now a part of the valley of the Rhine, a river flowing south-
wards, during episodes in the Miocene period, carried pebbles from
the north along its course. This river must have dated back even
to Eocene times. In reply to Prof. Hughes, he stated that the
attribution of the volcanic outflows to Miocene times was somewhat
problematical, and that there was some probability of their belonging
to a more recent period. Neither did they affect the district mainly
under consideration. There was no great succession of terraces, but
one main terrace of fluviatile origin, of which extensive traces re-
mained. Similar terraces occurred on the Moselle. In reply to the
Duke of Argyll, he stated that he did not think that volcanic action
had any thing to do with the formation of the gorge. The Miocene
strata lay in an approximately horizontal direction, and were not
tilted in such a manner as to suggest that their absence in the basin
was due to any disturbance of the strata. They must, in his view,
have been of necessity scooped away by the action of flowing water.

96 J. CLIFTON WARD ON THE ORIGIN OF SOME

14. The Ortatn of some of the Laxe-Bastns of Cumpernann. By J.
Cuirton Warp, Esq., Assoc. R.S.M., F.G.S., of the Geological
Survey of England and Wales. (Read January 7, 1874.)

[Prares IX. & X.]

Frrst Paper.

Tue subject of the origin of lake-basins is one whieh has received
considerable attention at the hands of several eminent observers,
and notably of Professor Ramsay. At the present time I propose
to discuss the origin of those depressions in which some of our
Cumberland lakes lie, taking as examples Derwentwater and Bas-
senthwaite, and Buttermere, Crummock, and Loweswater. The two
former lakes I have myself carefully sounded; the latter were
sounded many years since by Mr. P. C. Crosthwaite ; and the depths
obtained by him will here be made use of, with some observations of
my own.

The first point is to obtain a true idea of the proportions and
depths of the lakes as compared with the surrounding mountains.
For this purpose I have drawn a number of sections on a true scale
(the same yertical and horizontal), both in the direetion of the
length of the lakes and transverse to that direction. In the case
of the longitudinal sections (figs. 1-3), outlines of the mountains
on the left sides of the respective valleys are inserted, with their
true height, and the lakes are indicated by deep-black lines of a
thickness corresponding to the various depths. To form a strictly
accurate idea of the original size of the lakes, the following points
must also be taken into account. Both Derwentwater and Butter-
mere must formerly have extended nearly a mile higher up their
respective valleys (see Map, Pl. IX.). Derwentwater and Bassen-
thwaite were once continuous, though now parted by some three
miles of alluvial land. Buttermere and Crummock were likewise
originally one, though three quarters of a mile of alluvium now
separates them. The filling up of the head of Derwentwater is due
to the matter brought down by the river Derwent and the Watend-
lath beck ; amd the soundings for a mile and a quarter north of the
mouth of the Derwent show how this proeess is still going on, the
depths for that distance only increasing from 5 to 18 feet, there
being much deeper channels on either side the northern part of this
shallow tongue (fig. 17). The Barrow beck has also done its share
of work in forming an alluvial fan projecting into the lake, and in
shallowing the parts immediately beyond. The tract of alluvium
between Derwentwater and Bassenthwaite has been formed by the
river Greta flowing from the east, and the Newlands beck and its
tributaries flowing from the south and west. The head of Butter-
mere has been filled up by the detritus borne down by the Gates-
garthdale and Warnscale becks, flowing westwards on either side of

OF THE LAKE-BASINS OF CUMBERLAND. 97

Honister Crag ; and the alluvium separating Buttermere from Crum-
mock is mainly the product of Mill beck flowing down from the
north, and Sourmilk Gill from the south, upon the opposite side.

In the absence of borings the thickness of these alluvial deposits
cannot be estimated; but it might be that, formerly, the greatest
depths of the original long lakes were at points somewhere between
Derwentwater and Bassenthwaite in the one case, and Buttermere
and Crummock in the other.

The figures 5-10 and 13-16 inclusive, give the forms of the lakes
in transverse section, while figs. 1, 2, & 3 represent their longitudinal
sections along the lines of greatest depth.

When the true dimensions are thus laid down to scale, the point
that first strikes one is the insignificance of the hollows in which the
lakes le as compared with the elevations of the surrounding ground.
But since these lakes are not sheets of water merely dammed back
by moraine mounds, but lie in hollows scooped out of the solid
rock, it becomes an interesting and legitimate question to inquire
what was the agent which produced the hollows.

Now I think there is no doubt whatever that the principal valleys
of the district in which these larger lakes lie are of very great age.
After more than four years’ intimate acquaintance with mountain
and valley, the fact is very strongly impressed upon my mind that
all the grand mountain-sculpturing to be met with here is due to
the apparently weak agents now in operation, the powers of the
atmosphere, wind, rain, frost, and running water. Nor can I see
any reason to doubt that the elaboration of our present Lake-district
scenery has been going on uninterruptedly, though very likely at
different rates, at least ever since the close of the Carboniferous
period, and perhaps from an earlier date still. It will thus be seen
that I do not consider our Cumberland valleys to be merely gigantic
ice-grooves formed by a mammoth ice-cap, as was recently suggested
to this Society by Mr. Campbell to suit the case of Ireland; but,
rather, I look back through long past ages during which these
valleys were being sketched out, formed, and deepened, under very
varying circumstances and climates, now under almost tropical con-
ditions, and now beneath an arctic mantle of snow and ice—some-
times when the district was far higher above the level of the sea
than it is now, and sometimes perhaps when at a lower elevation.
In all probability glaciers have moved down these old valleys at
more than one period previous to the modern Glacial; and though
they may have effected a large amount of denudation by attrition
and transport, yet did they hinder the work of the denuding agents
for long ages after, just in proportion as they left the country with
a smooth and polished surface against which the atmospheric powers
might beat for long in vain. Such considerations as the foregoing
lead me to conclude that the lake-hollows are of very recent date,
geologically speaking, and represent the removal of but some of the
last rock-shavings by nature’s tools. And what the special tool was
which effected this we have now to consider.

Professor Ramsay has ably shown the difficulties which attend an

98 J. CLIFTON WARD ON THE ORIGIN OF 80ME

explanation of the formation of rock-basins by such agents as the
‘sea, running water, and mere weather-action, or by rock-disturbance,
special depression, or the formation of gaping fissures. Certainly
the lakes in question do not appear to have been formed by any
such actions. They bear no marks of marine action, even supposing
the sea capable of scooping out smooth hollows in closed fiords
running far inland. Such hollows are not now being formed by
running water, but rather filled up through its action. They do not
lie in synclinal troughs of rock; for the general direction of the
two sets of lakes is directly across and at right angles to that of the
strike of the rocks. No one could suppose that the shallow basins
represented in these diagrams were special areas of depression, when
the contour of their bed is seen to conform so closely to that of
the hills on either side, and the lie of the rocks shows no evidence
of so limited a movement. Neither can they possibly be regarded
as resulting from great fissures having taken place; for the rocks
show no signs of such at either end of the long shallow troughs, or
on either side; nor, as far as I can tell, do the lakes lie along lines
of fault, with the exception of Derwentwater, which is bounded on
the east side by a fault throwing together the hard Volcanic Series
of Borrowdale and the soft Skiddaw Slates (fig. 13),

These lakes are, indeed, merely long shallow basins with a
smoothed and well-scratched inner surface, worn out of the Skiddaw
Slate, which is much crumpled, cleaved, and comparatively soft.
The smoothing and the scratching of the rocks, in the direction of
the length of the lakes, may be traced at many points passing under
the water; so that there is no doubt that the hollows have been
at one time filled with glacier-ice ; and it remains to be seen how far
in these cases Professor Ramsay’s theory of the glacial origin of
lake-basins will hold good.

In a former paper, on ‘“ The Glaciation of the Northern Part of the
Lake District” *, I brought forward details to prove that at the
period of greatest glaciation the large glaciers were more or less
confluent, and that the greater part of the district was almost com-
pletely enveloped in ice. In the longitudinal sections (figs. 1 & 3),
however, I have made the thickness of the ice no greater than the
highest ice-scratches pointing down the valleys clearly warrant. In
many cases ice-rounded rocks are common above the point at which
the highest scratches are seen; but for our present purpose we will
ignore this further evidence and take a low estimate of the thickness
of the old glaciers. At the very outset of our examination a diffi-
culty will present itself to some minds. If we take the case of the
old Borrowdale glacier and mark its course along the valley of the
Derwent, over the present sites of Derwentwater and Bassenthwaite,
we are struck by the great flatness of the ground for full thirteen
miles, from Seathwaite to the lower end of Bassenthwaite lake, ten
miles of which is occupied along the line of section either by existing
lake or alluvium representing former lake. If, then, we had to
consider the old glacier a line of ice by itself, arising in the part

* Quart. Journ. Geol. Soc. vol. xxix. p. 422.

OF THE LAKE-BASINS OF CUMBERLAND, 99

of the Derwent valley above Seathwaite, and having no other out-
ward push but that due to gravitation down the two-and-a-half-mile
slope from Allen Crags to Seathwaite, and to molecular gravi-
tation throughout its length, we might be inclined to question
the possibility of its moving over the thirteen miles of flat ground
and scooping out lake-basins in its course. But just as the river
Derwent now receives very many tributaries, some as large as
itself, from the valleys and mountains on either side, and yet does
not occupy a much wider channel, but acquires a faster flow and
greater power of moving material onwards with it, so did the old
Borrowdale glacier, the former representative of the Derwent,
receive additional impulse from the numerous glacier sheets shed
off the mountains on either side and down the tributary valleys.
Let us see what this additional impulse really means. Above the
village of Rosthwaite there met several large ice-streams. The
Derwent glacier proper received reinforcement from ice coming
down from Sty Head, out of Sourmilk Comb between Grey Knotts
and Base Brown, and down the valley in which Comb Gill now
runs due north from Glaramara. Just above Rosthwaite it was
joined by the Stonethwaite glacier, made up of the Longstrath and
Greenup ice-streams (see fig. 11). The union of these two glaciers
(Derwent and Stonethwaite) formed the main Borrowdale glacier,
which had then to be forced through the narrowest part of the
valley, between Rosthwaite and Grange, a great ice-sheet moving
over the fell-tops on the east side more or less coalescing with it
(see Map, Pl. IX., and fig. 12).

The old Rosthwaite Lake, of an oval form, and nearly a mile in
length, occurs, it should be noted, just beyond the junction of the
Derwent and Stonethwaite glaciers, and where the ice, after having
its rate of flow somewhat checked at the junction, would have acquired
an additional impetus (see Map, Pl. IX.). About Grange the ice-
sheet shed off Brund Fell and down the Watendlath valley joined
the main glacier on its escaping from the so-called jaws of Borrow-
dale (just about Castle Crag, fig. 12). Thus increased in mass, the
ice was urged through the valley, over the present site of Derwent-
water, between Cat Bells* and Bleaberry Fell (fig. 13). I have
little doubt that the pressure in a N.N.W. direction was here so
great that the western side of the glacier partly escaped over the
Cat Bell ridge—especially through the Hause Gate, between Maiden
Moor and Cat Bells (fig. 3)—and became confluent with the New-
lands glacier upon the other side. It is just about this part of the
valley (fig. 13) that the greatest depth of Derwentwater is found—
a depth, however, very slight as compared with the thickness of the
ice, the former being 70 feet, and the latter something like 1100.
Tracing the course of the glacier still further down the valley, it
seems that about Keswick it was joined by several other large ice-
sheets. There were the Newlands and Coledale glaciers wedging in
about Braithwaite, and the western part, at any rate, of the great ice-
sheet coming down the Thirlmere valley, pressing along the Naddle

* Skelgill Bank in fig. 13 is the northern end of the Cat Bells range.

100 J. CLIFTON WARD ON THE ORIGIN OF SOME

-yale, and over the low ridge towards Keswick (see Map, Pl. IX.).
Besides these, sheets of ice of greater or less size probably descended
‘the southern slopes of Skiddaw and came down the valley of the
Glenderaterra. The main valley, just below Keswick, is about
three miles in width, between Braithwaite and Latrigg; and it
seems almost certain that the confluent glaciers occupied the whole
width to the thickness at least of a thousand feet. Only two or
three miles further down, however, this great volume of ice had to
pass through a part of the valley not more than a mile wide at its
base, between Barf and Dodd (fig. 14), beyond which it was able to
escape to the west and east of the present Bassenthwaite Lake,
mainly at first, I should fancy, to the west (over into Wythop
vale, fig. 15), on account of the pressure from ice-sheets probably
shed westward and north-westward from the lofty mountain mass
of Skiddaw. It will be seen that the deepest part of Bassenthwaite
is along the western side, just where resistance to the westward
trending of the glacier would be most felt ; and at the same time it
is evident what a small proportion the depth of the long narrow
trough bears to the probable thickness of the ice*.

Turning next to the examination of the case of Buttermere,
Crummock, and Loweswater, we are struck at once with the differ-
ence in form of the valley-bottom, when seen in section (figs. 5-10),
from that of the valley in which Derwentwater and Bassenthwaite lie
(figs. 13-16). In the latter case the contour is like a more or less
wide flat-bottomed pan, in the former (Buttermere, Crummock, and
Loweswater) like a round-bottomed basin.

The head source of the Buttermere glacier was upon Fleetwith,
below Grey Knotts, whence ice-streams flowed down on both north
and south sides of Honister Crag, uniting with one another to
form the main glacier just at the head of the present Buttermere
Lake, where the valley is narrowest, the mountains on either side the
most lofty, and where also the greatest depth of the present lake is
found (see Map, Pl. IX., and figs.5 & 19). The two ice-streams
are represented in fig. 4, and the glacier after their union in fig. 5.
It will also be noticed on reference to fig. 1 that the present lake
commences just at the foot of the steeper part of the glacier-bed, and
that beyond this point to the further end of Crummock there is
scarcely any incline.

Besides the small glaciers shed down the steep hill-side from
Burtnesst and Bleaberry Tarn+ Combs, a considerable ice-stream,
coming down the Mill-Beck valley, must have joined the main
glacier and helped to swell its mass (fig. 6). In my former paper I
showed that the Buttermere glacier on approaching Mellbreak was
probably split into two branches, the main mass continuing down
the valley, but a part of its left side being pressed across the low
watershed by Black Beek, just north of Scale Force (fig. 7), to join

_ * Tt should be borne in mind that a glacier, like a river, has its motion checked
on approaching a narrow part of its valley bed, and flows more swiftly on its
escape. Example, the strait between Barf and Dodd.

+ Between High Stile and High Crag... { Between High Stile and Red Pike.

OF THE LAKE-BASINS OF CUMBERLAND. 101

another ice -stream flowing northwards down Mosedale from Gale Fell
(see Map, Pl. [X.). Any part of the ice-mass thus got rid of from
the main valley at this point, however, was probably more than
made up for by the ice poured down from the flanks of the Grasmoor
and Whiteless Pike range. Between the northern end of Mellbreak
and Grasmoor, the valley is again considerably reduced in width ; and
along this lower half of the lake its depth is greatest. Escaped from
this narrow part, the main mass of the glacier continued down the
vale of Lorton, reinforced by ice shed off from the west sides of the
lofty Grasmoor and Whiteside range; and it is certain that at one
time, at any rate, the left limb of the glacier passed over the present
site of Loweswater, and partly enveloped Low Fell. This must
have been at a time when the outward thrust of the ice was suffi-
ciently powerful to push a portion of the glacier westward up an
incline of 75 feet in a distance of a mile and a half, from Crum-
mock to Loweswater (fig. 2). At the east end of Loweswater other
ice-streams from the south met this branch of the main glacier ; and
the whole mass was squeezed through the narrow valley between
Carling Knott and Low Fell (fig. 10), in which Loweswater now
lies, its deepest part (60 feet) bemg midway between these highest
points (fig. 21). A mile and a quarter beyond the upper end of
Loweswater the ice would pass over the low watershed at Sosgill
(fig. 2) into the flat country beyond.

I think that when the following points are carefully considered
—the fact of the lakes under examination being but long shallow
troughs, the thickness of the glaciers which moved along the valleys
in which the lakes now le, the agreement of the deepest parts of
the lakes with the points at which, from the confluence of several
ice-streams and the narrowing of the valley, the onward pressure of
the ice must have been greatest—one can hardly resist the conclu-
sion that the immediate cause of these lake-basins was the onward
movement of the old glaciers, ploughing up their beds to this slight
depth, in the way Professor Ramsay’s theory suggests. At the
same time it should be noticed that in the case of Buttermere and
Crummock—lying in a valley with rounded bottom, as seen in cross
sections—the action is merely a slight deepening of the basin, or the
formation of a smaller basin of similar general form at the bottom
of the larger; whereas in the case of Derwentwater and Bassen-
thwaite the action has been to produce a long shallow groove, of
yarying width, upon the flat bottom of the wide pan-like valley—
this groove in the case of Bassenthwaite being situated (for pro-
bable reasons before noticed) markedly at one side of the pan, close
- under the rising sides.

It may be urged by some that the fact of the deeper lakes being
situated in the valley bearing the glacier of least thickness is against
the idea of the lake-basins being formed by the ice; but I think, in
this case, the fact is rather due to the original forms of the respec-
tive valleys, the same or a less amount of rock-scooping giving rise
to a deeper lake in an originally round-bottomed valley than in a
flat-bottomed one.

102 J. CLIFTON WARD ON THE ORIGIN OF SOME

At a future time I hope to test the results here obtained for these
lakes by bringing forward like details in the ease of some of the
other lakes, and notably Wastwater, and also in a similar way to
consider the origin of the many mountain-tarns scattered throughout
the district. This latter subject leads me to make the remark that
a great service would be rendered if some one could be induced,
during the summer months, to visit the various tarns with a small
light canoe and sound their depths. The work is one I would
mostly gladly undertake myself; but as my official business will not
allow me the necessary time, I must leave it, hoping some one may
be found to do it out of pure love for scenery and science.

EXPLANATION OF PEATES.

Puate IX.

Map of the northern part of the Lake district, showing the direction of ice-
scratches by arrow-marks, and the tines along which the transverse sections in
Plate X. have been drawn.

Puate X.

Horizontal sections to illustrate the form of the valleys, the depth of the
lakes, the height of the mountains, and the thickness of the ice; together with
plans of the five lakes: all on a true scale of 1 inch to1 mile.

Fig. 1. Longitudinal section of the Buttermere and Crummock valley, from
Grey Knotts to Low Fell.

. Longitudinal section of the Loweswater valley, from the foot of Crum-

mock Water to Sosgill.

. Longitudinal section of the Derwent valley, from Allen Crags to the

foot of Bassenthwaite Lake.

. Transverse section of the head of the Buttermere valley, from Scarf

Gap to Dale Head.
. Transverse section across the deepest part of Buttermere, from High
Stile to Robinson.

. Transverse section across the lower end of Buttermere, from Red Pike

to High Snockrigg, and High Snockrigg to Whiteless Pike.

. Transverse section across the head of Crummock Water, from Black

Beck to Whiteless Pike.

. Transverse section across Crummock Water, from Scale Knott to Ran-

nerdale Knotts.

. Transverse section across Crummock Water, from Mellbreak to Gras-

moor.

10. Transverse section across the deepest part of Loweswater, from Carling
Knotts to Low Fell.

11. Transverse section across the head of Borrowdale, from Great Gable to
Ullscarf.

12. Transverse section across Keskadale, Newlands vale, Borrowdale, and
Watendlath vale, from Knott Rigg on the west to High Tove on the
east.

13. Transverse section across the deepest part of Derwentwater, from Causey
Pike to Bleaberry Fell.

14. Transverse section across the head of Bassenthwaite Lake, from Lord’s.
Seat to Skiddaw.

15. Transverse section across the deepest part of Bassenthwaite Lake, from

. a little south of Wythop Hall to Ullock Pike.

16. qos section aeross Bassenthwaite Lake, from Sale Fell to Little

nott.

bo

~ 09

or

Qe) Gok Ais er)

OL OOG

MAP \
NORTHE o,
OF THE <
LAKE pst
Scale about 4S
P lo 7 mule)

0)

SS

WH NU
dine HS
\

’

SJ

Wy),
VY

aboth a

WE
\ We

Soe Fell

927,

1,Ullscarf 2370. 2,6lar

Steel Fell 18/1, 8, Blencathra, 2847.

Troutbeck

Mell
Fell

¢ Beck

iterate

Common

te

§

Ww
ie
uni!
a
i

W.Robinson, 2417. 12,Hi

| Ligh Crag, 2443, 18 High Stile, 2643. 19,Red Pile, 2479. 20, Mell break, J676.
2,Carling Kinott,/78). 2 56.

BO-BroomEAll, 1670. 3h. d ected; 2084. 3%, Gavel Pike, 2577. 38, Annstone Cragg, 1423. 39 Fairheld, 2862.

9, Skiddaw, 3054. 10, Bart, 1586

27, Dellywaggon Prke, 2810. 28,Ratse, 2889, 29, Lords Seat, 1811.

up

J

Lakes. Mintern Bros. Lith

Quart. Journ. (

MAP OF
NORTHERN PART

OF THE
LAKE DISTRICT,

Scale about '2 inch
tc 7 mile

il |

uli

* SSeatcligx—
Ape,
Ave
{

4 wow ff
‘Seathwaite I \
“\

Rsac™ og

\

2 yy 1.

oO
%, i

1,0 scarf 2370. 2,Glaramara, 2560.

Robinson, 2017... 12,Hindscarth, 2385.
2),Carling Kinott,J78/. 22, Great Borne,
WBroom Fall, 1670. 31,Kirk Fell M76. 32,Grisedale Pike,

3,.Base Brown,2/20, 4, Grey Knotts,2287. 5,Cat Bells, 1482.

13, Dale Head, 2473. /4,Whiteless Pike, 2159. 15, Grassmoor, 279).
2019. 23 Kirk Fell, 2630. 24,Pillar, 2926. 26, Hay cock, 2619.
2593. 33,Crag Hill, 2749. 34, Causey Pike, 2082.

—

il

Sa

6 Helvellyn, 38, 7, Steel Fell, 18/1. 8, Blencathra, 2847.
16, Whiteside, 2317. 17, High Crag, 2443, 18 High Stile, 2643.

WH e%
ry nn

Troutbeck

Matterdale
Common

9, Skiddaw, 3054. +0.Bart: 1536
19, Red Pile, 2479. 20, Mellbreak,/676.

26, Stybarrow Dodd.2756. 27.Dellywaggen Pike, 2810. 28,Raise, 2889, 29, Lords Seat, 18/1.

wititinn = 3OOO 10.

2000 Ki.
The points of the Arrows

woo

10001, unin

Contours

2S Filled up Lakes.

indicate the Tee marks.

35, Bellbarrow, 1363. 36, Starling Dedd, 2084. 37, Gavel Pike, 2577. 38, Annstone Gragg, 423. 39.Fatrfield, 2862
Mintern Bros. Lith

Quart. Journ. Geol. Soc. Vol XXX. Pl. X.

Surface of Water 226Ft.

Surface of Water 238 Ft.

Surface of Water 33/ Ft.

S
Fig. 19. See

0
'

' 126
Ta cao WBE ~~~

1

& Surface of Water 327 Ft.
Fig. 20. Gummock Warter

9.10

_ fark Bees,
“YJ are ,
Surface of Water 396 Fb.

<
RS
Fig. 21. Loweswater

Scale Tinch to 1 mule.

Lakes with their soundings, depths tn feet.

Mintern Bros. Lith.

Quart, Journ Geol Soc. Vol XXX PiX

PLANS AND SECTIONS TO ILLUSTRATE PAPER ON “THE ORIGIN OF SOME OF THE LAKE-BASINS OF CUMBERLAND” BY J. CLIFTON WARD.

5 us Fig |7, Fig.| 8 Foie
“Sa Hay Stacks Scarf 6ap High Gra 2442 ig \7 Fig iy

Méllbreak Tow Fell Fig | 10

Fig 6
tile 2648 Red Pike 2478

aaa Pee a

S SES
Fig. Buttermere Lake Crammeck Water Crummock Loweswater Seagull
Fig. 2
Fig 13 -
Fig th Fig. 12. Crag HUL 2749 Coledale Horse Fig. 4 Pig. 18 Fig | 6.
Allen Crags 257! Aaron Cragae — Base Brown 2120 Little Gatesgarthdale High Scuwdel High Spy Pke Outersite 1863 Sand. Hill Whinletter Puss Beckstone Cull Sale Fell 176
And Barrew OS Grisadale Puke 2593 Seat How 1626 Bart 1535

IMM I TOT
Fig. 3 Seathwaite Old Resthwaite Lake

Surfice of Witer 238 Fe

Grasmoor 27H
Scarf Gap * Honuster Crag Dale Hea A735 High Stile 2643 Has eness Robinson 2417 Red Pike 2476 High Sneckeigg Whateleas Pike AKG Rannerdate linette Whitelens Pee 2169 Crummock Meilbreak 1668 Carling Knotts Loweswoter law Fell
Warnscale Bottom Gateagarthdale Beck B Bleaberry Tarn Buttermere MAL Beck Black Beck Crummeck Scale Kno we Hanrersiale Hinette Oummeck J 77 tA % > a
A A 4 NCO )
“ os ee Se
= =i A \ 5 — J 7 one eet
.: = = ~~ wt
VOD UW QD BS SEES tan ti 80
As il =
Pig. +. Fig. 5. Fig. 8 Z ¥, g
6 Fig 19. ern
Great Gable 2048 Seathowaite Fell 1969 Glaramara 2660 Sergennts Crag 1872, Ulecar’ 2370 Kexkerdale Beck Little Dale Newlands Beck River Derwent Watendlath Tarn Casey Pte 082 Rowling Bnd 22 Skelgill Bank Bleaberry Pell 1932 ;
Styhead Gill River Derwent Leng Strath Greenup Gill Knott Rigg 272 Robinscntrage Hindscarth 2364 High Spy 2142 Castle ag Brand: Fell 1368 High Tove 1686 Newtaneds Beck Derwentwater: 8 ea
Lbs
s Porro Y gg yy
= 7 a= | ws)
3 ; ey Glee
‘ 1) a
Fig.l Fig. 20. Cummocle Water
Lords Seat 1810 Bassenthwaite Dodd W612 Cart Sitle Skid 3054S. Wythop Hall Ullock Pike Sale Fell 1170 Watches Little Knott
Bart 1535 y Bassenthwoite Baasenthwaite
~ KK « Suartace of Water 3.96 Ft.
YS = | Fig. 21. Tweed
\ eee WC J Seale, 1 inch to | mile, both vertical and herixental :
Fig lt. Fig 16
Be Scale Vinch w 1 mile
5 any . . Nove, The Patte bs aude, atari be ih
(UMMM Voteanic Seriew of Borrowdlale SN middaw states SBeacier ice The marks thus ‘Fig\10" show where transverse sections cross the longitudinal sip si Matpalspner ain Dango! Cnet nou

ten formations, bear ne relation te either bedding er clearage Lakes with their sourdings, deptiain feet.

OF THE LAKE-BASINS OF CUMBERLAND. 103

Fig. 17. Plan of Derwentwater,
18.

Venere ae narte dake, with depths given in feet, and dotted

19. ;, Buttermere, 5 ; :
2). i Crummock Water, _| lines along which the sections run.
21. », Loweswater, )

Discussion.

Mr. Campsett said that he had listened with great pleasure to the
able paper of the author. He was not himself acquainted with the
Lake-district ; but he knew many similar districts in which similar
phenomena existed. He agreed with the author’s conclusion, that
these lake-basins were the result of glacial erosion. But if ice
could do so much, it might have done more. In confirmation of the
author’s views, Mr. Campbell said that in the Caucasus there are
very few lakes. He had found no glacial phenomena in the whole
range, except one small moraine near the only lake in Daghestan.

Mr. Evans inquired what effect the varying hardness of the strata,
their trend and dip, might have had on the formation of the basins,
and how the presence of islands was to be explained. He men-
tioned that at the present day the rainfall at Seathwaite was in
some years nearly 200 inches, which, if there were sufficient cold,
would suffice even now for an enormous supply of ice.

Mr. SeeLry inquired whether the position of the lake-basins in
the supposed glacier had any definite relation to the positions of
smaller affluent glaciers, and whether the lake-basins were to be
attributed to that relation.

Mr. Gopwry-Avsten remarked on the acceptance which Prof.
Ramsay’s views had received, and the support which they were
receiving. There was little doubt of the former existence of ice
over a large portion of this part of Europe; but whether it could
have existed in such thickness as was required by some geologists
was another question. He doubted as to the power of glaciers to
drive ice forward to any great extent over land either up a slope or
over a horizontal space. He considered that the paper would add a
great interest to the country to which it related.

Capt. Dovetas Gaxron disputed the power of ice to act in a
manner materially different from that of water. Owing to the
friction of the ice at the bottom of a glacier, he thought its flow
would be so much retarded that its excavating power would be
almost annihilated.

Mr. Warp, in reply, stated that the basins in all the cases he
had cited were excavated in the Skiddaw slate, the hardness of
which was nearly uniform. The dip of the strata was very vari-
able, but he could not point to any spot where the depth of the lake
was connected with the dip.. The islands in Derwentwater might
be the result of an old moraine left by the glaciers in retreating up
the valley. The position of the lakes was in the direct line of the
principal glaciers. The thickness of the ice was proved by the
existence of scratches along the sides of the valleys, such as could
not have been produced in any other way. The probability was

104 J. CLIFTON WARD ON CUMBERLAND LAKE-BASINS.

that the ice had been even thicker than shown by these marks.
Though the laws of motion of ice were the same as those of water,
yet the action of a hard body was of necessity different from that of
a liquid.

Prof. Ramsay was so accustomed to meet with papers such as
this, confirming his original views, that he was almost becoming
weary of the subject. He considered, however, that the sections
given by the author on a true scale were of very great value, as
calculated to give a correct idea of the actual phenomena, and as
showing the value of De la Beche’s maxims with regard to such
diagrams. He inquired whether there could be any difficulty in a
body of ice, some thousands of feet in thickness, cutting out such
inconsiderable hollows as those shown, just in the same manner as
running water sometimes excavates its channel more deeply at one
spot than another, if from local circumstances the nature of its
motion is increased.

NOTES ON A JOURNEY FROM ALGIERS TO THE SAHARA. 105

15, Gronoercat Norrs on a Journny from Axermrs to the SAHARA.
By Grorez Maw, Esq., F.G.S. &ce. (Read February 52, 1874.) |

[Puare XI.]

Durine the spring of 1873 I paid a short visit to the province of
Algiers, and beg to lay before the Society the result of my observa-
tions on the main geological features of the district between Algiers
and the Sahara.

The military read running almost due north and south from the
Mediterranean coast of Algiers to the borders of the desert at
L’Aghouat, the most southern French military outpost, and crossing
nearly at right angles the prevalent line of strike and trend of the
high and low ground, suggested itself to me as affording the most
comprehensive section in a continuous line that could be selected for
my object, which was to compare the geological features of Algeria
with those of Marocco observed during my visits to that country in
the years 1869 and 1871.

Very little has heretofore been written about the geology of Algeria ;
and, indeed, the best published maps fall short of giving a correct
delineation of even the hill- and valley-systems of the country.

My line of route of 285 miles and 3° of latitude (viz. from 36° 47'
N. to 33° 48’ N., or a trifle over 210 miles in a direct line) included
30 miles by rail, from Algiers to Blidah, thence by diligence over the
Lesser Atlas, through the gorge of La Chiffa to Medeah, and thence
over the high land of the Tell to Boukhari, along a well-engineered
road which at present terminates at a point about 110 miles from
Algiers (See Map, Pl. XI.). From Boukhari to L’Aghouat, for the
remaining 200 miles over the Salt Plains and Hauts Plateaux, com-
munication is kept up by a small omnibus, running every fifth day
along a rough track almost devoid of bridges, little better than a
caravan-route, and which would be quite impassable to ordinary
European vehicles. The dotted line on the map indicates my route,
and the black line the section represented in the diagram Pl. XI.,
in which an occasional feature lying a little east or west of the actual
line of section has been introduced. ‘The fine double lines crossing
the main line of section indicate the strike of the beds.

In general terms the portion of Algeria crossed in my route may
be described as a great plateau of an average height of 3000 feet;
bounded on the north by the range of the Lesser Atlas, in precisely
the same way as the great plateau of Spain is backed up on its
north side by the Pyrenees, and shelving off to the north in slopes
and terraces to the sea-level of the Sahara. In detail, the general
mass must be viewed as a series of high plateaux, separated by in-
tervening plains at a relatively lower level ; these plains appear to
correspond with the former relative level of the Sahara submergence,
their present height indicating the subsequent rise of the land, and
represent a great Postpliocene anticlinal, to which the more ancient

106 GEORGE MAW’S NOTES ON A JOURNEY

foldings of the Mesozoic and Tertiary beds are subordinate in their
effect on the contour of the country.

I purpose, in the first place, to follow my route as an itinerary,
and to describe the contour and geological features in passing from
north to south, and then to summarize my observations in the order
of stratigraphical sequence.

The coast in the neighbourhood of Algiers presents, in the absence
of cliffs and escarpments, the same character as the Marocco sea-
board of the Mediterranean and the coast-lines of Corsica and the
Reviera already referred to in previous papers. The hill- and
and valley-system of the land shelves under the adjacent sea, with-
out the intervention of distinct escarpments, indicating, as I believe,
that the existing coast-level is so recent that the sea has not yet had
time to excavate a cliff-boundary. The paleontological evidence on
the connexion of Europe and Africa in Postpliocene times brought
forward by Mr. Boyd Dawkins in his paper ‘“‘ On the Physical Geo-
graphy of the Mediterranean during the Pleistocene age ”’ (Popular
Science Review’ for April 1873) supports the view of the recent
origin of the existing Mediterranean coast-line.

Raised Coast-beaches—I am informed by Mr. Tristram that a
series of raised concrete shell-beaches occurs on the Mediterranean
coast to the west of Oran, ranging in height above the sea-level from
200 to 600 feet. The fragments exhibited were obtained by him at
Mers-el-Kebir, at a height of 400 feet; and among them Mr. Gwyn
Jeffreys has determined the following species :—Pecten opercularis,
Pectunculus glycymeris, Cardium edule, Venus gallina, Turbo rugosus,
and Fusus corneus, all common Mediterranean and Pliocene species.

Neighbourhood of Algiers.—The city of Algiers is built up the
side of a high boss of land standing out as a promontory on the west
side of the bay, and separated from the higher land of the Lesser
Atlas and Tell Plateau by the level plain of the Mitidja, the submer-
gence of which to the extent of 200 or 300 feet would separate the
high boss as an island. It is about eight miles across from north to
south, and the same distance from east to west. It includes the
oldest rocks observed on the line of section, viz. mica-schist and
gneiss, with intercalated quartzite, disposed as an anticlinal with a
north and south strike. Its highest point, about 800 feet above the
sea, is a little to the west of Fort de ’Empereur, from which the
beds dip nearly east and west at an inclination of 30°.

Skirting these ancient micaceous rocks, Tertiary beds are super-
imposed as terraces ranging round the Bay of Algiers to the 8. and
S.E. of the city, and ascend in the vicinity of Moustafa-Supérieur
nearly to the summit of the hill. They have a prevalent dip of from
10° to 20° S.E. and E., and consist of soft, calcareous, cream-coloured
freestone, here and there tufaceous, occasionally including beds of
compact limestone, and closely resemble in mineral character the
Calcaire grossier. They abound in shells, among which Mr. Etheridge
has determined the following—Pecten jacobeus, Arca, Cucullea,
Modiola, sp., casts of Certthium, Natica, sp., and fragments of Pecten
and Ostrea ; and he supposes them to be of Miocene age, and possibly

FROM ALGIERS TO THE SAHARA. 107

in part Pliocene. The Museum at Algiers contains a few corals from
these beds.

Remains of Postpliocene mammals have also been found in the
neighbourhood of Algiers; and the skull of a fossil ox (Bubalus
antiquus), remarkable for the immense length of its horn-cores
(about 20 inches) is preserved in the museum. It was found in
excavating stones for the road to the Penitentiary of Bab-Azzoun,
near Algiers. It resembles in mineral character the Tertiary fossils,
and was probably imbedded in rearranged débris from the Tertiary
freestone beds.

The Plain of the Mitidja.—The railway, for about eight miles,
skirts in a §.E. direction the bay of Algiers close to the sea, then
turns south by La Maison Carrée through some cuttings of shingle
and loam, and by a westerly bend at once enters the plain of the
Mitidja at a height of 100 feet above the sea. The Mitidja is a level
plain, about eight miles across, running out to the sea at the southern
extremity of Algiers Bay, and separates the high boss of land against
which Algiers is built from the range of the Lesser Atlas, forming
its northern boundary. ‘The plain consists of sandy loam inter-
calated with pebble-beds, and rises almost imperceptibly westwards,
from the sea-level at Algiers Bay to the neighbourhood of Blidah,
where a fan-shaped delta of shaly débris stretches from the foot of
the Atlas half across the plain. This appears to have been deposited
during the submergence of the Mitidja by the stream which debouches
from the back of Blidah. Blidah station is situated about halfway
up the regularly sloping delta, at a height of 500 feet; and the town
of Blidah a mile higher up, at a height of 650 feet, and about a mile
from the foot of the Atlasrange. From Blidah station the rail again
descends the western slope of the delta, till it reaches the average
level of the plain at the station of La Chiffa, at a height of about
350 feet. The low ground is here left; and turning due south we
enter the gorge of La Chiffa—a sinuous pass through the Lesser
Atlas, closely resembling many of the lateral valleys of the
Pyrenees.

The Lesser Atlas.—Trending nearly due east and west, on the
south side of the plain of the Mitidja, is the Lesser Atlas, the only
range included in the section which, either in structure or altitude,
can lay claim to the title of a mountain. Compared with the noble
chain south of the city of Marocco, rising 13,000 feet above the sea,
the Algerian chain looks dwarf and diminutive, In altitude it
scarcely averages 4000 feet, and few points exceed 5000 feet. It
presents the character of a single rounded ridge, here and there
rising up as separate blunted points. Very few of the intervening
depressions occur at a lower elevation than 2500 or 3000 feet.
Viewed from the‘south side, the mountainous aspect disappears, as
it forms the northern boundary of the high tableland of the Tell,
almost equalling it in height, excepting that here and there the
summits rise up as shrub-covered hills above the general level of the
Tell plateau. In structure, the Lesser Atlas of Algeria bears no
analogy to the great Atlas of Marocco. It includes no eruptive

108 GEORGE MAW’S NOTES ON A JOURNEY

rocks ; nor did I observe a single eruptive rock on. the line of section
from Algiers to L’Aghouat.

At the foot of the Lesser Atlas, flanking the Mitidja, small
isolated fragments of Tertiary beds are seen clinging to the older
rocks, and, itis probable, were once continuous with the Tertiary beds
on the north side of the plain near Algiers; as these fragments, both
at the back of Blidah and at the entrance of the gorge of La Chiffa,
run up into the Atlas valleys, they must have been deposited since
the Atlas assumed its existing contour.

The central mass of the range, exposed in a number of sections in
the gorge of La Chiffa, consists of a hard, close-grained, dark blue,
argillaceous rock, commencing with a small anticlinal at the entrance
of the gorge, and then dipping south at a high angle, averaging from
50° to 60° and occasionally 70°, or almost vertical, with a strike
nearly east and west, corresponding with the trend of the chain.
The rock is here and there slightly cleaved, the cleavage being nearly
coincident with the stratification.

At first sight the Atlas rocks appear to be a distinct series from
those forming the base of the Tell plateau; but after a careful
working-out Ge this part of the section, I think it probable that they
are merely an altered condition of the lower members of the great
series of Mesozoic beds (probably of Cretaceous age), of which the
Tell and the Hauts Plateaux are composed, the prevailing character
of which resembles that of our Lower Lias. On the north side of
the range, strata of this aspect are seen in the bed of the stream by
the first bridge south of Blidah, and are replaced a mile further south
by the more slaty rock of the Atlas, with a similar strike and dip.
Again, at the south end of the gorge of La Chiffa the slaty rock gra-
duates into stratified alternations of hard bands with intervening
clayey shales; and these, between the south end of the gorge and
Medeah, give place to a great mass of grey marly shales, here and
there somewhat schistose, which are seen to underlie the Tertiary,
series forming the capping of the Tell.

The inven of the Tell consists of an irregular tableland, of an
average height of 3500 feet, about 30 miles across, extending
from the Lesser Atlas on its northern boundary to Boghar on its
southern escarpment. In ascending from the gorge of La Chiffa, the
Mesozoic grey shales and marls are lost, and the capping of the
northern half of the Tell is seen to consist of a Tertiary series ex-
tending about 14 miles from Medeah to within a mile of Berou-
aguia, as an irregular undulating synclinal, presenting a series of
escarpments at its northern and southern boundary respectively,
facing north and south, and including bright red marls resting on
the Mesozoic rocks, overlain by calcareous freestones containing
fossils, interstratified with grey and variegated marls, the whole
having a collective thickness of from 300 to 500 feet. The fossils
are very obscure, most of them occurring only as casts, and few of
the species are determinable, Mr. Etheridge has determined the
following genera, and from their general facies thinks the series
may include beds of both Pliocene and Miocene age, viz. Solen,

FROM ALGIERS TO THE SAHARA, 109

Cyprina, Arca, Pectunculus, and several species of Ostrea and
Pecten.

The calcareous freestone is highly ferruginous, and contains
deposits of hydrous sesquioxide of iron, undistinguishable from the
iron-ores of the Northamptonshire Oolites.

After the Tertiary escarpment is crossed near Berouaguia, at a
height of 4000 feet, the road rapidly descends to a level of 3100
feet, and for a few hundred yards a sharp anticlinal of hard grey
slaty rock similar to that of the Atlas is passed; from this point
to near Boghar the road, at a level of from 3150 to 3600 feet,
passes over an undulating country, the higher parts of which are
crowned with forests of Pinus halepensis. Overlying the anti-
clinal above Berouaguia, beds apparently identical with those in
the southern half of the gorge of La Chiffa again set in and extend
without interruption to the southern escarpment of the Tell, at
Boghar. In physical aspect they resemble our Lower Lias, and
consist of dark bluish marls, often en masse, but occasionally in-
terstratified with well-defined thin courses of hard calcareous and
argillaceous rock. These beds are highly contorted, and present
a succession of dips both north and south at high angles, occa-
sionally vertical, and arranged in shortly repeated series of syn-
clinals and anticlinals, the valleys nearly always occurring on
anticlinals. The strike of these contorted beds is locally variable,
but maintains a general east and west direction. I obtained na
fossils between Berouaguia and Boghar; but at Boukhari (a neigh-
bouring village, situated on an outlying spur of the Tell plateau),
I collected the following species :—Ostrea, allied to O. multicostata,
a fragment of an Ostrea allied to O. eyathula, Ostrea allied to
O. callafera, Pullastra (sp.), Plewrotomaria (sp.), and several examples
of Pecten which are supposed by Mr. Etheridge to be of Miocene
age. In the museum at Algiers, there is a small series of fossils
from the neighbourhood of Boghar, which I have had photographed.
It includes some large Ostrew, which Mr. Etheridge and Professor
Phillips think may be of Tertiary age, but possibly Cretaceous ;
there are also some Lower-Cretaceous fossils from the same locality,
viz. a Turrilites, an Ammonites allied to A. Deshayesir, a Trigonia,
an Inoceramus, and an Astarte; also an Urchin, of the genus
Micraster or Toxaster. This series is important, as affording a key
to the age of the great mass of stratified beds of the Tell plateau,
and tallies in age with some Lower-Cretaceous fossils of other
species obtained by Mr. Tristram in 1861 on the northern escarp-
ment of the Hauts Plateaux. The beds at Boukhari are apparently
on the same horizon as the Boghar escarpment; and I am inclined
to think that the rather imperfect series of fossils I obtained from
the neighbourhood of Boukhari may be of Cretaceous age, especially
as their general facies resembles the fossils I collected from the
Neocomian beds at Saffe, on the coast of Marocco.

Between Boghar and Boukhari the road descends the southern
escarpment of the Tell plateau to the valley of the Cheliff and the
great low-level plain of the northern Sahara, situated at a height of

Q.J.G.8. No. 118. K

110 GEORGE MAW’S NOTES ON A JOURNEY

2150 feet. The Arab village and French fortress of Boghar is
situated near the summit of the escarpment, at a height of between
3000 and 4000 feet, from which a magnificent view is obtained.
Looking north, the nearer parts of the Tell plateau are seen, capped
with forests of Pinus halepensis, the distance being bounded by the
summits of the Lower Atlas. In a southerly direction, the great
level plain of the northern Sahara spreads out as a sea-like expanse,
bounded on the southern horizon by the mammillated chain of the
Djebel Sahari, the curious notched outline of which is faintly seen
in the extreme distance. The level plain runs up amongst the out-
lying spurs of the Tell plateau, which die out like promontories and
islands resting on the flat expanse. From Boghar we look directly
down into the valley of the river Cheliff, which rises in the plain of
the Northern Sahara at a height of a little over 2000 feet. It is
the most important river in Algeria. Its waters, like those of
nearly all the North-African rivers, are very turbid. It first flows
northward through a winding ditch-like channel between high banks
of alluvium, and then westward into the Mediterranean a little to
the east of Oran. In the alluvial banks of the Cheliff, near
Boukhari, Mr. Tristram found some Hippopotamus-bones. The
species was not determined, and the specimens have unfortunately
been lost.

At the carayansary of Bou Guezoul, twelve miles south of
Boukhari, the outlying spurs of the Tell plateau are left behind, the
road terminates, and we enter on the great plain covered with
Artimisia, Sueda, Salicornia, Chenopodium, and other salt-loving
plants, perhaps indicating by their presence a comparatively recent
marine submergence. The uniformity of the plain is only broken
by an occasional low escarpment of gypseous and variegated marls,
which are well exposed near the caravansary of Ain Oussera.

These marls appear to be inferior to the fossiliferous beds of
Boukhari and the Tell; but they are for the most part shrouded over
in the plain by a deposit of grey loam, which, throughout the district
included in the section, seems to mark the limit of the Sahara sub-
mergence. This loam, as well as the older beds that come to the
surface, are in many places coated with a calcareous surface-crust
similar to that observed in the Marocco plain, the formation of which
was, I believe, due to the great heat of the sun, in alternation with
heavy rain, quickly drawing up and evaporating from the substratum
water containing calcareous matter in solution. ©

The Djebel Sahari.—The low-level plain of the Northern Sahara is
crossed, in a direction nearly east and west, by a single range of hills
known as the Djebel Sahari, towards which the plain rises from north
and south to the third caravansary of Guelt-el-stel, at an altitude of
2900 feet. The Djebel Sahari, which are probably 700 or 800 feet
high and 3500 feet above the sea-level, consist of a steep anticlinal
of hard yellow sandstone, rising up from beneath the gypseous marls
here and there isolated as dome-shaped masses, with escarpments
facing both north and south. Denudation has sculptured the range
into a curious mammillated series of hills, separated by deep indenta-

FROM ALGIERS TO THE SAHARA. 111

tions, which are conspicuous from a great distance. Between these
hills and the northern escarpment of the Hauts Plateaux, the low-
level plain continues for about fifteen miles, first descending from
the anticlinal of Guelt-el-stel to a level of 2550 feet, and then gra-
dually ascending to the Rocher de Sel at the foot of the Hauts
Plateaux, at a height of 2800 feet. This second plain appears to be
a repetition, dipping south, of the gypseous and saliferous marls of
Ain Oussera, on the north side of the Guelt-el-Stel anticlinal. Two
large salt-lakes (the Sebka Zahrez) occur at a height of 2550 feet in
the plain, a little to the south of the Djebel Sahari; and a little
further south nitre is obtained by evaporation from square pans ex-
cavated in the grey marls.

The plain south of the Djebel Sahari is covered with a deposit of
grey silty loam, before referred to as the “ Sahara loam ;” and here
we have clear evidence of recent marine submergence in the occur-
rence between the Djebel Sahari and the Hauts Plateaux of an ex-
tensive range of sandhills. I am also informed by Mr. Tristram
that raised beaches of conerete shell-beds, similar to those on the
coast, occur in this neighbourhood. He does not mention their
altitude, which, however, cannot be less than from 2000 to 2500
feet. They form a most important point in the evidence of the
Sahara submergence, to which I shall have to make further
reference.

The Rochers de Sel, or, in Arabic, Hadjar el Mehl, forming the
first and most northern escarpment of the Hauts Plateaux, rise up
from the northern boundary of the plain, which here attains a level
of 2850 feet near the fifth carayansary. The Rochers attain a further
height of 300 or 400 feet, and present a strangely weird aspect.
The mass consists of dark grey marls, mixed with rock-salt and
gypsum, and interstratified with hard grey argillaceous rock-bands,
the whole presenting a broken and tumbled arrangement. In one
place the strike is nearly north and south, and dip nearly vertical ;
but the general dip and strike correspond, as far as I could ascertain,
with the general dip and strike of the northern escarpments of the
Hauts Plateaux, the beds of which the saliferous marls appear to
underlie. The salt-mountain is capped with great blocks of tumbled
rock scattered about in the wildest confusion, the placement of which
is evidently due to the dissolution of the intercalated and underlying
salt-marls ; and I am inclined to think that the irregular dip and
strike of the beds im situ has been materially influenced by the
failure of support from the dissolution of the salt-rock, just as in
some glacial beds the melting of intercalated ice is supposed to have
brought about a confused and contorted arrangement. From these
beds I obtained an Ostrea allied to O. carinata. Mr. Tristram has
given me two specimens of Exogyra sinuata and an Ostrea allied to
O. cyathula he found here in 1860; and Mr. Etheridge thinks they
are of Cretaceous or Neocomian age.

The Hauts Plateaux.—From the south of the Rochers de Sel, the
ground rapidly rises in a series of terraces, forming the northern
escarpments of the Hauts Plateaux. Immediately to the south of

K 2

112 GEORGE MAW’S NOTES ON A JOURNEY

the salt-rock near the fifth carayansary, bright red and green marls
are exposed in the banks of the Oued Melh, dipping south at an
angle of 35°. These bright-coloured marls appear to overlie the
grey saliferous marls of the salt-mountain, and form the base of the
series of the Hauts Plateaux. Their well-marked mineral character
enables their horizon to be identified in several localities further
south. The Hauts Plateaux may be described as a great synclinal,
about 35 miles across, forming an open tableland of an average
height of 3700 feet, bounded north and south by successive series of
terraced escarpments. The beds are probably of Lower Cretaceous .
age, and apparently a repetition of those ef the Tell Plateau, consisting
of grey gypseous marls alternating with well-defined courses of hard
fine-grained argillaceous and ealeareous stone. The town and fortress
of Djelfa is situated near the centre of this tableland, at a height
of 3600 feet; seven miles further south the road attains its highest
point, at a height of a little over 4000 feet, which, however, is not
the actual summit, as higher terraces both right and left surmount.
the main level of the plateau, and probably attain a level of about
4700 feet. The road now descends the southern esearpment, crossing
over the same series of beds dipping north that were passed in
ascending from the Rochers de Sel to Djelfa; and at Ain-el-Ibel,
she sixth earavansary, the bright red and green marls seen near
the salt-mountain are again exposed, underlain by a conglomerate
at a height of 3300 feet. Mr. Tristram, in his ‘Great Sahara,’
records the occurrence of irregular beds of lignite in this neigh-
bourhood.

From the Hauts Plateaux to [’ Agkouat—From Ain-el-Ibel, the
road gradually descends over an undulating plateau of red and yellow
sandstone, probably a recurrence of the sandstone thrown up in the
anticlinal of Guelt-el-Stel. The sandstone south of Ain-el-Ibel dips
21° N.N.E., and further south gradually decreases in inclination till
within a mile or two of the seventh caravansary of Sidi Makhelouf,
where level strata of red sandstone, forming apparently the crown of
the low anticlinal are reached ; these are the lowest beds exposed in
this part of the section. At the caravansary of Sidi Makhelouf,
2725 feet above the sea, the sandstone on the south side of the an-
ticlinal commences to dip 8.8.W. at an angle of from 10° to 15°. At
this point the open plateau merges into a valley shut in by ranges
of the beds overlying the sandstone. These present scarped faces on
either side of the gradually narrowing strath. Between Sidi Makh-
elouf and Metlili, at a height of 2700 feet, several low sandstone
hills (Fig. 1) rise out of the valley, and are remarkable for their
perfectly flat tops at a uniform level independent of their stratifica-
tion, which inclines at an angle of 15°. I can only account for their
contour by marine denudation. Here also the fundamental sand-
stone begins to be shrouded over by grey loam, which is continuous
with the grey loam covering the open Sahara plain further south ;
it is noticeable that this superficial deposit sets in at exactly the
height at which it occurs north of the Hauts Plateaux, near the
Rochers de Sel, and at the height at which the isolated range of

FROM ALGIERS TO THE SAHARA. 11183

Fig. 1.—Flat-topped Sandstone Hills between Sidi Makhelouf and
Metlili. (2700 feet above sea-level.)

“oo ce ie oe SS —e ~. i

Grey loam in bottom of valley.

sand-dunes occurs in the same locality. I shall further on refer to
these evidences of the former extent of the Sahara submergence.

At Metlili the flat valley narrows to a mile and a half in width,
and follows a direct southwest course. Itis bounded by symmetrical
escarpments 250 feet high, composed of gypseous marls interstra-
tified and capped with harder strata, the irregular denudation of
which has resulted in a singular range of isolated bosses and pinnacles
separated by little lateral ravines. Near Metlili, bright red and
green marls similar to those at Ain-el-Ibel and those near the
Rochers de Sel are again cbserved in the bottom of the valley; so
that the escarpments bounding it probably correspond with the beds
of the Hauts Plateaux ; the upper part may be on a higher horizon.
The road here follows for a few miles the line of strike, but on
turning S.E. gradually passes an ascending series of beds, under
which the red and green marls are lost ; and turning through a break
in the eastern escarpment, where the flat valley widens out, drifted
sand and sandhills commence; the open plain is now approached, into
which the cliff-lke escarpments die out as isolated ridges running
N.E, and 8.W., separated by flat-bottomed valleys about a mile wide.
The scene is most singular, and at once suggests to the eye the sub-
mergence of the older land beneath a newer deposit, which continues
as a level plain as far as the eye can reach. Three or four parallel
ridges, separated by intervening straths, terminate the high ground,
each alternate flat occurring between dip slopes and on denuded
anticlinals (Fig. 2); the final escarpment, dipping N.W. at an angle
of 25°, presents a cliff-lke amphitheatre facing the great plain
(Fig. 3). Beyond this, at a distance of about three miles, isolated
rocks with a complementary escarpment facing N.W. rise up out of
the plain, and dip away S.E. at a steep angle under the shroud of
grey lcam and sandhills. On one of these rocks L’Aghouat, the most
southern French fortress in this part of Algeria, rises up from an
oasis amid an immense grove of date-palms, and presents a pictu-
resque association of Arab mud houses, French military buildings,
and a new mosque. The town is situated at a height of 2333 feet
above the sea. The great altitude at which the sandhills commence
in its neighbourhood, was a fact I had not auticipated, as further to
the N.E., south of Biskra, the desert plain is on the level of the sea,

Fig. 2.—Section of Hscarpments on the borders of the Sahara Plain, near L’ Aghouai.

----Open plain and
sandhills 2206
feet above sea-

Ye--------Rocher de Chien
----B’Aghonat.

San —

- Fig. 3.—Last Cliff south of the Harts Platcaua, facing the Sahara Plain, L’ Aghouat. 7

GEORGE MAW’S NOTES OWN A JOURNEY

114

FROM ALGIERS TO THE SAHARA. eles

and the French surveyors have ascer-
tained that some of the lagoons in the
neighbourhood are below the sea-level.
From inquiries I made at L’Aghouat,
L ascertained that the plateau on which
it is built spreads out as a promontory
in a south-easterly direction towards
the level of the main desert, descending
by slopes and terraces to the sea-level.
Mr. Tristram tells me that red sand-
stone, which he supposes to be of
Triassic age, rises up in the M’Zab
country to the 8.H. of L’Aghouat. This
is probably a recurrence of the red
sandstone which comes to the surface
in the Sidi Makhelouf anticlinal.
Although there is good evidence of
a Posttertiary submergence of the dis-
trict surrounding L’Aghouat, the high
level of the sandhills may also be partly
attributed to the agency of wind, the
effect of which in transporting sand to
great distances must be seen to be fully
realized. My approach to L’Aghouat
was through a blinding sand-storm,
which was still in full force on the
following day, when on ascending one
of the rocky eminences over the town I
obtained my first view of the desert
plain. A bitter east wind was blowing,
and the light was quite subdued by the
sand-storm. Looking south over the
plain I could but dimly see the level
horizon through the murky haze; and
nearer the sudden gusts of wind were
picking up from the desert cloud-like
patches of sand, and whirling some of
them away towards the great smoky
mass of moving sand in the distance,
and heaping up the nearer clouds in
talus-like slopes against the boundary
escarpments. To the north, terrace
upon terrace of the cliffs of the Hauts
Plateaux, piled up in receding lines,
formed the boundary (Fig. 4). The
near view was equally remarkable.
East and west I looked along a broken
ridge of rocks, completely isolated in
the plain, crowned by handsome French
buildings and long lines of fortifica-

Fig. 4.—Southern Escarpments of the Hauts Plateaux, facing the Sahara Plain, as seen from LD’ Aghouat.

116 GEORGE MAW’S NOTES ON A JOURNEY

tions. Lower down on either side of the rocky ridge were the Arab
flat-roofed habitations, built of sun-dried mud bricks, embosomed
in a dense forest of date-palms, many of them from 90 to 100 feet
high; and stretching away from this dark green and brown mass,
lines and patches of light green vegetation spread out into the desert,
where its thirsty sand was sucking in the last drainings from the oasis.
During my stay at L’Aghouat, I took sketches and sections of several
of the escarpments and rocky outliers (represented in Figs. 3,
4, and 5), the most important of which is the Rocher de Chien,
W.S.W. of the town, a jagged hill (Fig. 5), rising up 300 feet

Fig. 5.—Rocher de Chien, an isolated Rock, Sahara plain, west of

b)
LP Aghouat.
B.E. p ah N.W.
5 r=] =
3 a
a a3 & a
o (3) 2 ro 3
So mre a
ates s = a
k= oe Ss. 3
S83 Siro PP cyt
a8 : FS ehiahea) S
S) <) rc

ee re es es es ee G

os ba a SS So a Se es a

eS
as ; SSSSSSGNSS

out of the sandy plain, and consisting of a hard calcareo-argilla-
ceous rock, with a N.E. and S8.W. strike, dipping S.E. under the
plain at an angle of 65°. At the foot of its N.W. escarpment:
occurs a calcareous cream-coloured rock, with casts of fossils, inclu-
ding a Chemnitzia or Turritella, underlain by a fine-grained sand-
stone, with casts of bivalve shells, which Mr. Etheridge thinks may
be of Miocene age. The sandstone is succeeded by grey gypseous
marls, which are lost under the Posttertiary deposit of grey loam,
filling up as a plain two miles wide the denuded anticlinal between
the Rocher de Chien and the boundary escarpment to the north.
The lower beds of marl and sandstone en the north side of the
Rocher de Chien have a less steep inclination than the overlying rock,
and dip 8.E. at an angle of about 35°. The face of the main escarp-
ment N.E. of the town (Figs.3 and 4) presents the following succession
of beds in ascending order. At its base white gypseous marls, inter-
stratified with bands of fine-grained white stone, occur as a low
terrace a little in advance of the cliff face, and dip 15° N.N.E. with
a W.S.W. and E.N.E. strike. The main face of the cliff consists of
grey marls interstratified with hard bands and gypseous green marls ;
and its jagged summit is capped with hard calcareous bands, con-
taining curious concretionary masses, which are very fossiliferous,
including a spatangoid Urchin, which Dr. Wright thinks may be a

FROM ALGIERS TO THE SAHARA. 117

Hemiaster or Penaster, resembling Penaster Fournelli, from the Hip-
purite limestone (Brissus?), Avicula, Inoceramus, Lucina, Cardium,
Arca, Byssoarca, Cardita, Cyrena? Cucullea, Clavatula, &c. These
stand out as bosses from the matrix, with a weathered black surface,
and look like blotches of black mud; the colour is only skin-deep,
the mass of the nodules being of a pale grey. They are excessively
hard and difficult to break, and probably consist of ferruginous chert.
The view from the summit of the escarpment is very singular.
To the south the eye ranges over the great plain interrupted only
by the rocky ridge of L’Aghouat; looking north a gentle dip-slope
descends to the flat valley about a mile wide, bounded on the opposite
side by a complementary dip-slope, with a repetition in reverse of the
beds seen in ascending the first escarpment. The second ridge is
soon crossed ; and on descending its escarpment facing north, another
flat valley about a mile wide is reached, bounded on its opposite side
by an escarpment facing S.E. The view from the summits of these
ridges presents a series of ranges running N.E. byS$.W. (Fig. 2), nearly
buried by a Posttertiary deposit of grey loam, forming flat valleys
about a mile wide, each alternate valley occurring between dip-slopes
and in the centre of a denuded anticlinal, suggesting to the eye the
levelling up by a recent deposit of a series of hills and valleys that
had originally been of greater height and depth. The level valley-
bottoms of loam are obviously an inland extension of the plain, and
were deposited when the Sahara sea ran up as fjords between the
promontories of older land that branched out from the isolated land
of Algeria.

Lignite occurs in several localities N. of L’Aghouat, viz. in the
Djebel Amour at Kusra and at Berich, also at El Kheicha on the banks
of the Oued M’Zi, a branch of the L’Aghouat river, where it is a
metre thick, and overlain by black rock ; but I had not an opportu-
nity of seeing it 7m situ. Fossil plants are said to occur a day’s
journey N.W. of L’Aghouat, and are probably associated with the
lignite formation. The only locality where the lignite has been
observed tn situ by 2 geologist is that noticed by Mr. Tristram at
Ain-el-Ibel, where it occurs near the horizon of the band of bright
red and green marls exposed in several localities on my line of section,
and probably a little above the horizon of the salt deposit of the
Rochers de Sel.

Of the occurrence of other minerals, the following localities were
named to me at L’Aghouat :—

Nitrate of potash is found near Touat, manganese at Outhed el
Abiod, lead at Oued Sidi Bilgassem, and copper at Berich; but I
could not ascertain the mineralogical conditions under which the lead
and copper occur.

My return journey from L’Aghouat, following the same line of
route, enabled meto check and correct the altitudes and fill in addi-
tional details in the section.

Stratigraphical Summary.—In conclusion I will give a general
summary in stratigraphical sequence of the formations and of the
successive changes of level and contour of the district comprised in

118 GEORGE MAW’S NOTES ON A JOURNEY

the section. Of eruptive rocks there are no traces in any part of the
district between Algiers and L’Aghouat. ‘The oldest rock is the boss
of micaceous schists and gneiss at the back of Algiers, striking nearly
north and south and dipping away east and west from a low anti-
clinal, the crest of which occurs near the Fort de ’Empereur at the
summit of the hill.

Rocks of the Lower Atlas and Anticlinal north of Berouaguia.—The
pass through the gorge of La Chiffa exposes hard slaty rocks, dipping
south at a ‘high angle and somewhat cleaved, which appear ‘repeated
as an anticlinal on the south side of the higher part of the Tell
Plateau. They contain no fossils; and but for their being confor-
mable to the overlying Neocomian beds, I should, judging from their
physical character, place them on a much lower horizon. There is
no satisfactory evidence as to their age. They strike N.N.E. by
W.S.W., and closely resemble in mineral character the nearly vertical
slaty shales on the northern flanks of the Great Atlas south of the
city of Marocco.

Sandstones of Giuelt-el-Stel and Sidi Makhelouf—tIn the denuded
plain separating the Tell from the Hauts Plateaux, and again on the
south side of the Hauts Plateaux, red and yellow sandstones are
thrown up as anticlinals, between which the overlying beds of the
Hauts Plateaux form an elevated synclinal. There is no paleeontolo-
gical evidence as to their age; but they resemble the Bunter in mineral
character, and are overlain by red marls resembling the Keuper. A
similar succession of beds of Triassic age occurs in the south of
Portugal ; and I am informed by Mr. Tristram that red sandstone,
supposed to be of Triassic age, comes to the surface at Warigla in the
MW Zab country, S.W. of L’Aghouat, probably as a third anticlinal.

Saliferous Marls of the Rochers de Sel.—The northern escarpment
ef the Hauts Plateaux is the only point in the section where the
saliferous marls are clearly exposed. I must take exception to the
view expressed by Mr. Tristram, in his ‘ Great Sahara,’ that they
are “‘an eruption of argillaceous calcareous mud and rock-salt up-
heaved across the Secondary and Tertiary deposits.” Although
they appear isolated, they occupy a well-marked horizon in the
Mesozoic beds at the base of the Hauts Plateaux ; and though absent
in similar positions at other parts of the section, they are clearly
interstratified between the sandstones below and the red marls, grey
marls, and rocky courses above. In mineral character they resemble
the salt deposit at Bex, in Switzerland. Crystals of salt and gypsum
are intimately mixed up with the grey marls. Ten or fifteen miles
north of the Rochers de Sel, the presence of these marls is indicated by
the two great salt lakes, or Zahrez ; but south of the Hauts Plateaux,
although their horizon can be identified, I observed no salt deposits ;
and if they occur in the north Sahara plain, their presence is obscured
by the surface-deposit of grey loam and calcareous crust.

The patchy occurrence, so to speak, of the salt corresponds with
the isolated distribution of salt rock in the Keuper of Worcestershire
and Cheshire.

Red and Green Marls.—Overlying the red sandstones, a thin series

FROM ALGIERS TO THE SAHARA, 119

of very bright red and green marls is exposed as a low escarpment
in the north Sahara plain between the Tell and the Hauts Plateaux,
near the Rocher de Sel at the foot of the Hauts Plateaux, at Ain-
el-Ibel at the foot of the southern escarpment, and again on the
south side of the Sidi Makhelouf sandstone anticlinal from which the
marls have been denuded. Itis probable the red marls are brought
up by rolls of the surface in several parts of the north Sahara plain
both north and south of the Guelt-el-Stel anticlinal, but are obscured
by the calcareous surface-crust, sand, and other superficial deposits.

Grey Marls and rocky bands of the Hauts Plateauw.—Immediately
in upward succession from the band of red and green marls is an
immense series of dark grey marls, interstratified with argillaceo-
calcareous bands, resembling the Lias, the collective thickness of.
which cannot be less than 1500 feet. They compose the great syn-
clinal of the Hauts Plateaux and the contorted mass of the Tell
Plateau, separated by the denuded plain of the northern Sahara.
From these beds I obtained no fossils; but as they immediately overlie
the saliferous marls of the Rochers de Sel containing EHwogyra, they
are probably Cretaceous. These beds also extend south of the Hauts
Plateaux to the borders of the Sahara at L’Aghouat, where they are
overlain by

The Fossiliferous Beds of L’ Aghouat.—These occur in conformable
succession from the marls of the Hauts Plateaux, and consist of
gypseous marls interstratified with bands of fine-grained stone, and
capped with limestone containing curious black concretionary nodules
full of fossils, which Mr. Etheridge considers are of Miocene age.
It is probable that these beds may also occur on the summit of the
Hauts Plateaux, but were not observed along my line of route,
which was below the highest platforms of rock seen to the right
and left.

Tertiary Beds of the Tell and Algiers.—The only other Tertiary
beds included in the section are, first, the soft yellow calcareous
freestone flanking the anticlinal of gneiss and mica-schist of the
Algerian promontory, at a height of from 100 to 900 feet above the
sea ; and, secondly, the red and grey marls and ferruginous freestone
of Medeah, capping the Tell Plateau as an irregular synclinal, and
occurring at a height of from 2500 to 4000 feet, probably of Miocene
age. The great difference in height at which these two masses of
Tertiary beds occur within a short distance, is an important point in
the physical geology of the district presently to be referred to.

Posttertiary Deposits.—Yo the north of the Lesser Atlas, the plain
of the Mitidja consists of grey loam interstratified with shingle-beds,
commencing at the sea-level south of Algiers, and ascending inland by
a gentle slope to a height of from 200 to 300 feet. Again, on the
south side of the Atlas and Tell, the great plain of the northern
Sahara is covered with grey loam and occasional sandhills from a
height of 2000 feet on its northern side to a height of 2700 feet on
its southern boundary ; and after crossing the Hauts Plateaux, the
same deposit sets in on its southern side, commencing in the inter-
vening valleys ata height of 2700 feet, and gradually descending

120 GEORGE MAW’S NOTES ON A JOURNEY

in the open Sahara plain to a height of a little over 2000 feet, where
it is overlain by sandhills.

The raised beaches along the coast, ranging in height from the
sea-level to 600 feet, and similar beaches inland south of the Tell
Plateau, occurring at an elevation of 2000 feet, may be contempo-
raneous in age with the Sahara loam.

In addition to these deposits, which are evidently marine, the allu-
vium of the valley of the Cheliff, near Boukhari, containing bones
of the Hippopotamus, and the superficial calcareous crust, coating
both the Sahara loam and the older beds, may be enumerated as the
most recent formations in Algeria.

Succession of Events, and Changes of Level—(1) The oldest land
on the the line of section is the anticlinal of mica-schist near Algiers.
As the north and south strike is nearly at right angles to the strike
of the neighbouring Atlas and of the Mesozoic beds of the Tell and
Hauts Plateaux, which do not seem to have been affected by the
mica-schist upheaval, these beds must have been more recently de-

osited.
7 (2) The upheaval of the Mesozoic rocks may have been contem-
poraneous with the first upheaval of the chain of the Lesser Atlas,
striking N.E. by E. and 8.W. by W., with which the strike of the
Mesozoic beds corresponds.

(2) A long period of denudation followed, which removed at least
a thousand feet of Mesozoic strata from the area of the north Sahara
plain, and commenced moulding the rocks of the Lesser Atlas into
hill- and valley-contour ; but there is no clear evidence as to the
amount of this denudation, which was Pre- and Posttertiary.

(4) A subsidence of at least 3000 feet during the Tertiary period,

‘in which were deposited the Miocene beds capping the Tell and
similar beds at Algiers, fragments of which are seen skirting the
denuded ravines of the Atlas.

(5) An elevation of at least 4000 feet of the Tell Plateau, and a
lesser elevation of at least a thousand feet of the district north of the
Lesser Atlas chain, including what is now the plain of the Mitidja.
I am inclined to think that the north face of the Lesser Atlas was a
Posttertiary line of fault of at least 3000 feet ; otherwise it is difficult
to account for the great altitude of the Tertiary beds on the Tell
compared with the height of those surrounding Algiers.

(6) A long period of denudation followed, during which the Ter-
tiary beds were removed from the plain of the Mitidja, and a further
denudation took place of the Mesozoic beds from the north Sahara
plain.

(7) This was followed by a Posttertiary depression, which I shall
term the Sahara submergence, during which the formation took place
of the concrete shell-beds along the Mediterranean coast to a height
of 600 feet, and similar beds on the south or inland side of the Tell
Plateau at a height of at least 2000 feet. The deposition of shingle
and loam in the plain of the Mitidja at a height of from 100 to 300 feet,
was probably contemporaneous; and to the same period of submer-
gence must be ascribed the formation of loam-beds and sand-hills on

FROM ALGIERS TO THE SAHARA. ~ 11271

the northern Sahara plain at a height of 2700 feet, and on the open
Sahara south of the Hauts Plateaux at a height of from 2700 to 3000
feet, as well as the coast-denudation of the flat-topped sandstone
hills at Sidi Makhelouf at a height of 2700 feet.

At this time the Algerian promontory was an island, separated
from the Atlas and Tell by a strait six or seven miles wide, now the
plain of the Mitidja. ‘The Lesser Atlas rose directly from the sea,
into which its streams were bringing down subaérial débris, and
depositing the great delta-like mass of drift on which Blidah is built.
Crossing the Atlas and Tell we should soon reach another inland
sea some fifty miles wide, now represented by the north Sahara
plain, out of which the range of the Djebel Sahari rose as a chain
of islands; and its southern limit is now well recorded’ by the long
range of sandhills on the southern margin of the plain north of the -
Hauts Plateaux. The Hauts Plateaux formed another range of
isolated land running nearly east and west, 30 miles wide ; and on its
southern side fjords ran up among the valleys and debouched at
L’Aghouat into the open Sahara sea, some 800 miles across, which
separated by a broad expanse of water the islands of the Tell and
Hauts Plateaux from the highlands of Central Africa.

(8) Reelevation of the Land.—An upward movement followed the
Sahara submergence, the extreme limit of which, whether repre-
sented by the present height of the land, or attaining a higher level,
is open to question. It is, however, certain that an emergence of
the highlands of Algeria took place to the extent of about 3000 feet ;
but the reelevation appears to have been greatest to the south of
the chain of the Lesser Atlas. The contour of the limit of elevation,
as indicated by the present level of the marine deposits of the Mitidja
and Sahara plains, would represent a wide low anticlinal, gradually
rising from the sea-level at Algiers across the plain of the Mitidja
to a height of 2700 feet at the north side of the Hauts Plateaux, and
then descending from 2700 feet on its south side to the sea-level in
the open Sahara. The extent of the elevation following the sub-
' mergence of the Sahara, is a question of considerable interest. In
a former paper read before the Society, and in a paper read before
the British Association at Liverpool, I recorded several facts in proof
that the present coast-line of the Mediterranean, in several far-
removed localities, is gradually subsiding ; and Mr. Boyd Dawkins,
in his paper on the physical geography of the Mediterranean, has
on paleontological and other evidence pointed out the probability
that during the Pleistocene age the land now forming its shore-
line stood at a height of about 3000 fect above its present level.
Again, moraines were observed by Dr. Hooker, Mr. Ball, and myself
in the Great Atlas south of Marocco, at a height of from 6000 to
7000 feet, in a latitude where no perpetual snow now lies at.12,000
feet ; so it is probable that the land stood at a much greater eleva-
tion when these moraines were formed.

On the other hand, Sir C. Lyell has suggested that the former
diminution of temperature and extension to lower levels of alpine
glaciers was probably the result of diminished temperature due to

1 Qy GEORGE MAW’S NOTES ON A JOURNEY

the Sahara submergence; but the Great Atlas chain to the N.W.
would probably have partaken of the general depression, and the
comparatively low level at which moraines there occur would have
been brought still nearer the sea-level, counterbalancing the effect of
increased refrigeration. A small glacier still exists at a height of
between 8000 and 9000 feet in the Sierra Nevada south of Granada,
in latitude 37°; and supposing the Great Atlas had partaken of the
elevation of 3000 feet assumed by Mr. Dawkins as the Postpliocene
height of the present Mediterranean boundary, a corresponding alti-
tude of the Great Atlas of from 15,000 to 16,000 feet would probably
have afforded sufficient height for the bearing of similar small glaciers
at existing temperatures. The high land of Algeria has certainly
been elevated 3000 feet since the Sahara submergence ; and it is
probable that its maximum elevation may have reached 6000 feet
in Postpliocene times, since which a gradual subsidence of 3000
must have taken place and appears to be still going on.

EXPLANATION OF PLATE XI.

Fig. 1. Section across Algeria, from Algiers to L’Aghouat, Sahara. Vertical
scale 1 inch to 9000 feet; horizontal scale 32 inches to 30 miles.
2. Map of route from Algiers to L’Aghouat. The line of route is indi-
cated by a dotted line, the line of section by a thick black line, and
the strike of the beds at different parts by fine double lines.

Discussion.

Mr. W. Bory Dawxrns considered an elevation of from 2000 to 3000
feet necessary to allow of that migration of animals through Spain and
‘Sicily, to and from Europe and Africa, which took place in Pleisto-
cene times; and this view appeared to be confirmed on physical
grounds by Mr. Maw. He believed that a great axis of disturbance
ran east and west along the course of the Mediterranean; and to
this the strike of the beds observed by Mr. Maw was parallel. It
is a remarkable fact that the change in the level of the Sahara
observed by Mr. Maw should be the same as that which has taken
place in the Mediterranean area, but in the opposite direction.
Probably the elevation of the Sahara was coincident with the
depression of the Mediterranean.

Mr. Bavrrman called attention to the excellent drawing of a
desert escarpment exhibited by Mr. Maw. He said that this draw-
ing perfectly represented what is to be seen in every dry desert
country, like the north of Africa or Arabia. In the latter country
the succession of the beds of Neocomian and Tertiary age was
similar to that observed in Algiers. He thought that the disturb-
ances attributed by Mr. Maw to the dissolving out of salt, were in
reality due to the dissolving of gypsum.

Mr. Davipson remarked that thirty years ago M. de Verneuil
found many fossils in the region to which Mr. Maw’s paper related.
These included a great Ostrea, Terebratule, and other forms which
were both Miocene and Pliocene.

Quart. Journ. Geol. Soc . Vol .XXX. Pl. XL.

HAY HILLS, DESERT

ME
= Grey Gypseous Mag Grey Loam overlain by Sandhills

of compact fine
NS Rode / Neacomia

Ss

BNO, Ae TT

G. Ma

Mintern Bros. lith.

(ees
5

—— ayy

Ben fs

e
PLAIN OF LESSER PLATEAU OF
THE METIDJA ATLAS THE eis
g
8

8 S

®

< }
8 § §
{ ee
Bs : g 8 S
&S y “ S x 8 < o ne es S
: iS Coa
: rs \ ee
3 NS S BS SsS SSs = &
eS ae i SZ LN.
‘ ‘ {SHEN SSS

etreec Nes 96h St 5 DN ee
t My
Ze A NIN ct

UW Veiller of the Cheli ff 2150 feet.

' Fig.1. SECTION ACROSS ALGERIA FROM ALGIERS TO LAGHOUAT, SAHARA.
SALT PLAINS DJEBEL SALT
3 3 SSAHARI LAKES
s iS)
3 S 8 Sg |
8 Ss Ss
& & * 5 :
S . e : 3
s : : | ; g 8
s & 8 iS g is CS
S 8 8 § sy N © R S ia)
5 iS 8g 5 8 ~ & a
a cS NOS x not aN
® ~ : Saas Sin aS ges
3 2 é Ses s ge 8 5 gs
: g Ro Ses sy gS

Tertiary Freestenes and Marls Hard Shales _Neocomian ?

Hard Shales £ Gritstones,
reating or Mesozoic: (Neoconian?|Beds. & Gritstones

slightly cleaved).

Metamorphic Gneiss, Micaschist
& Dark Quartaite overlain by
Tertiary Calcareous Freestone.

(NNE ssw ln

Vertical Scale, linch to 9000 feel.
Horizontal. Scale 3% inches to 80 miles.

G.Maw del April /875.

Vertiary (Miocene ?)

Red: £ Grey Gypseous Marls overlain by Post Tertiary loam & Sand, Anticlinal of Compact Sandstone

Neocomian?

YN by w

Fig. 2. MAP OF ROUTE FROM ALGIERS TO L'AGHOUAT.

WD A\\)
RA
x “So

As

SS
Zi

AN N=

HAUTS PLATEAUX®

Highest point of Road F050 fe

\Dyelter 3606 fob

sa TLL

Air eb Lbel Carcransary

3300 feet

3150 feet
Y) 2920 feet

850 feet.

\

WW, Wy

aes sossngmeeet

S by E

Saliferous and Red Marls
Gypseous Marls

{ nw

Grey Gypseous Marls wterstratilied wit bands
of compact tine grained (alcarto-Argdlaceous
Rod: / Neacomian or Cretaceous ? )

ted Marts lignite
& Conglomerate

EF,

~
LF, LY yyy
Ly
YY

2726 feet;

Yaravansary Sidi Makhelouf

fetlili, 2600 feet:

?

i

5)
Me

Wi
WA

Anticlinal of Red & Yellow
Sandstone ( 2 Triassic)

‘idi Makhelouf

ha LLL =
LL ILE.
VM LID

Quart. Journ. Geol. Soc . Vol .XXX. Pl. Xt.

ae | HILLS, DESERT

f L Aghouat 2330 feet.

—
MEY

Red and Gree Marts

/
Uj
y)
H/ Metlili

Gypseous Marls ke

o, i Gre Loam overlain by Sandhills
jocene!

ssw )

ie

(Bz
AGHOUAT

| Mintern Bros. lith .

Sra

fea

FROM ALGIERS TO THE SAHARA. 123

Prof. Ramsay asked for information as to whether there was
evidence of a great sea having extended across the north of Africa
at a comparatively recent period, as he thought that this would
explain certain ethnological phenomena. He was struck by the
difference of the elevation of that part of the Sahara visited by the
author from that described by Prof. Desor.

Mr. Presrwicn considered the occurrence of recent shells at so
great a height something quite new. The former French observers
had referred to their occurrence on the Sahara itself, and below the
level of the sea. Subsidence appeared to be still going on.

The Rey. Mr. Housmaw remarked that the eastern coast of Spain
was still rising.

Mr. Evans observed that the existence of such tracts of high and ~
absorbent soil as those described might, with even a moderate amount
of rainfall, account for the supply of water to the Artesian wells
with which the lower part of the Sahara is dotted. He mentioned
that Cardium edule and Buccinum gibberulum had been found in
the Sahara by Desor, and that the latter had been considered iden-
tical with a shell now found on the N.W. coast of Africa.

Mr. Maw briefly replied, and stated that the sandy plain near
L’Aghouat referred to in his paper was a tongue of elevated land,
east of which, at all events, the desert stretched away at or near
the sea-level ; and this was undoubtedly submerged during the period
of depression. He thought that the sea also probably extended
westward at the same time, perhaps to the Gulf of Guinea.

124 YT, DAVIDSON AND W. KING ON THE TRIMERELLID SE.

16. On the TRIMERELLIDE, a Patmozoic Famity of the Patiio-
-BRaNcHS or BracuiopopA. By Tuomas Davinson, Esq., F.R.S.,
F.G.S., &e., and Professor Wittzam Kine, Sc.D. Honoris Causa,
Queen’s University in Ireland. (Read February 25, 1874.)

[Puates XIJ.—XIX.]

ConTENts.

I. Bibliography of the Trimerellids.
II. Chiefly descriptive of their internal features.
UII. Myology and other characters ef the family.
IV. Affinities of the family.
V. Geological range, Chronogenesis, and Evolution of the family.
VI. Physiography of the Seas tenanted by the Trimerellids, as compared
with that of the Cambrian Seas. 3
VII. Diagnosis of the family.
VIII. The genus Zrtmerella and its species.
IX. The genus Monomerella and its species.
X. The genus Dinobolus and its species.

APPENDIX.
a. Lingulops Whitfieldi. | b. Chelodes Bergmani.

I. BrsrioGRAPHy oF THE TRIMERELLIDS.

YneE important group of shells forming the subject-matter of this
memoir is, comparatively speaking, of recent acquisition in palzon-
tology ; for, though two of its species were made known in 1853, under
- the designations of Obolus Davidsonwi and O. transversus*, it is only
of late years that a beginning was made towards gaining a correet
view of its remarkable internal features. So little understood were
the interiors of these species at first, that they were confidently, but
erroneously, referred to the genus Obolus by one of the present
writers. Billings did not escape the same error when he described
his Obolus canadensis.

The next stage, one decidedly in advance, was gained by Billings,
who, finding some Canadian specimens with a singular interior, was
led to institute a new genus for them under the name 7J'rimerella—
observing, at the same time, that it ‘‘is allied to Obolus, but from
which it differs in the possession of longitudinal septa”+. He erred
again, however, in describing another species of the present family
under the name Obolus galtensis§.

In 1867, Lindstrém published an important paper, describing
some specimens found in the Island of Gotland as far back as 1859,
and which he had no hesitation in referring to the genus T’rimerella
of billings, throwing at the same time further light on the internal
structure of the group. He also referred to some specimens of the

* Davidson, Introduction Br. Foss. Brach. p. 136, 1853.

+ Report on the Geological Survey of Canada, p. 28, 1857.
{ Pal. Foss. Geol. Canada, i. p. 166, June 1862. § Ibid. p. 168.

TI, DAVIDSON AND W. KING ON THE TRIMERELLID. 125

so-termed Qbolus Davidson which had been discovered in the same
island *,

For the first time, in 1868 the public became acquainted with the
fact that an objection had been made by Prof. Hall against the
generic allocation of the so-termed Obolus Davidsoni, though
(bowing to the opinion of other authorities) he was induced to name
an allied species Obolus Conradi+.

In 1870, Dall, from not being well acquainted with the characters
upon which Billings had founded his genus Trimerella, and only
having before him Lindstrém’s description of the Gotland specimens
above referred to, was led to propose for the latter the separate
generic designation, Gotlandiat.

All along our friends were either supplying us with specimens, or
with an account of their discoveries; so that, with close attention to
both, we felt justified in believing ourselves enabled to throw addi-
tional light on the group. In the mean time others were still
working on the same subject. Hall brought out a short notice, in
the early part of the summer of 1871, which contained a description,
with names, of his proposed genera Rhynobolus and Dinobolus§.
Dall published a second paper giving further details (already known
to ourselves) on the internal characters of Trimerella, abandoning
at the same time his genus Gotlandia. In this paper Dall briefly
referred to the different known species of Trimerella, and added two
or three new ones||.

During the same year Meek published a description of Trimerella
ohioensis Y ; and Billings brought out a supplement containing a fuller
description of his 7’. acuminata** ; while, in the same year, Quenstedt
briefly noticed the genus Trimerella, along with Obolus, Schmidiia,
Obolella, and Acritesty.

On the 29th of December, 1871, Billings published a short notice
of a new genus, Monomerdla, which, he intimated, would be soon
fully described, and illustrated by us from both Canadian and
Swedish specimens. He also proposed another genus for his Obolus
canadensis under the name Obolellina tt.

In April, 1872, Hall published a reissue of his paper ‘“ printed in
March, 1871,” to whichis added a plate illustrating the interior of
Trimerella, and the ventral valve of his so-termed genus Rhynobolus.
In the same month, Billings brought out a fuller account, with
figures, of his genus Obolellina§$§.

* Ofv. Vet. Ak. Forh. vol. xxiv. p. 253, 1867; and Geol. Mag. vol, v. p. 441,
1868.
t Twentieth Annual Report of the Regents of the University of the State of
New York, p. 368, 1868.
{ American Journal of Conchology, vol. vi. p. 160, 1870.
§ Notes on some new or imperfectly known forms among the Brachiopoda:
March 1871; and Neues Jahrbuch, p. ¥89, 1871.
|| American Journal of Conchology, vol. vii. p. 79, 1871.
{| American Journal of Science and Art, vol. 1. 8rd ser. p. 305, April 1871.
** Tbid. June 1871; and Annals and Mag. of Nat. Hist. 4th ser. vol. viii.
August 1871.
tt Petrefactenkunde Deutschlands, p. 669, &c. 1871.
{{ Canadian Naturalist, New Series, vol. vi. p. 220, 1871. §§ Ibid. April 1872.
Q.J.G.8. No. 118. L

126 T. DAVIDSON AND W. KING ON THE TRIMERELLID&.

On the 20th of August, 1872, the writers of this memoir com+
municated to the Geological Section, at the Brighton Meeting of the
British Association, a paper entitled “‘ Remarks on the genera T7i-
merella, Dinobolus, and Monomerella,” in which they endeavoured to
point out the characters pertaining to the Trimerellids, as well as
the place this family should occupy among the Palliobranchs*.

Before concluding the present section, it may be mentioned that,
up to this date, about eighteen species of fossils belonging to the
group we are engaged with have been described, or notified, and that
seven different names have been proposed for the genera in which to
include them. Some of these names, it has been admitted, are
synonyms, or belong to one and the same genus; certain others have
been proposed for groups which cannot, we think, take a generic
rank, We are consequently led to restrict the number of genera
to three—namely, Zrimerella, Monomerella, and Dinobolus. These
constitute our family Zrimerelliide. Another genus, Lingulops, has
been proposed by Hall for a curious fossil: we propose giving a de-
scription of it in this paper; but we are yet undecided whether it is
a Trimerellid, or a Lingulid.

-We now subjoin a list of the species :—

Trimerella grandis, Billings. Monomerella orbicularis, Bill.
acuminata, Beil. Lindstromi, Dav. & King.
Lindstromi, Dall. Dinobolus Conradi, Hail.

—— Billingsu, Bill.
— ohioensis, Meek.
—— wisbyensis, Dav. § King.

canadensis, B2d/.
Davidsoni, Salter.
transversus, Salter.

galtensis, Bill. —— Woodwardi, Salter.
Monomerella Walmstedti, Dav.f King. —— Schmidti, Dav. § King.
prisca, Bail. —— magnifica, Bill.

As soon as it became known that we had decided on preparing the
present memoir, specimens from every one who had them were sent
to us. Dr. Lindstrom, Professors Hall, Fr. Schmidt, and Walm-
stedt, Messrs. Fegroeus, Bergman, Dall, Whitfield, Etheridge, and
Meek, and several Dudley friends furnished us with the best things
in their possession, with a liberality and courtesy we cannot forget,
as did also Mr. Billings (Paleontologist of the Canadian Geological
Survey), Sir W. Logan, and Mr. Selwyn. To all we beg to return
our sincere thanks.

TI. CHIEFLY DESCRIPTIVE OF THEIR InTERNAL Parts.

The family Trimerellide includes both transversely and longitudi-
nally elongated species, also thick- as well as thin-valved ones.
The species are generally of considerable size, the largest known
measuring nearly 4 inches in length. Their shell-substance is
chiefly calcareous, though in certain thin species the lime may

* Printed in full in the ‘Brighton Daily News’ newspaper for the 21st of
August, 1872; also in the Geol. Mag. vol. ix. Oct. 1872; and in the Annals and.
Mag. of Nat. Hist. for the same month.

ind

T. DAVIDSON AND W. KING ON THE TRIMERELLID&. 127

possibly have been small in quantity compared with the corneous
element*.

The species often fail in neatness or regularity of form, especially
in their umbonal region. This part is often massive and strongly
projecting, and, it may be, ecurving, incurving, or twisted to one side.
Some species have it obtusely and regularly rounded. All the
species apparently have had their outer surface smooth, or marked
with lines of growth, coarse in some, and fine in others: slight
traces of diverging longitudinal lines are occasionally visible.

Guided by certain evidences indicating the existence of two of the
chief organs belonging to shells of their class, it is readily seen that
the Trimerellids have had the large valve pedunculated, and the
smaller one characterized with oral or (the so-called) brachial ap-
pendages; it is therefore proposed to name the former the pedicle
(Pl. XII. figs. 1,3, 5), and the latter the brachial valvey (Pl. XII.
figs. 2, 4, 6).

The pedicle-valve has the cardinal face of its umbo flattened so as
to form a triangular space or area, which differs remarkably in size
according to genera and species. The longitudinally elongated
species have the largest area; while in those that are transversely
elongated it is small. As in other Palliobranchs, the area is made
up of different parts; two, however, possess exceptional features.

The deltidium (Pl. XII. figs. 1a, 3, 7a), carved, as it were, out
of the central portion of the area, is comparatively level, in relief,
or an excavation: it is crowded with fine imbricated transverse
lamine, in general arching towards either the point of the beak
or the hinge-line. At, and parallel to, its broad end or base,
there is a narrow inclined bard-like space or deltidial slope (b),
distinguished by a few faint longitudinal lines : it is, however, often
so slightly developed as to be with difficulty observed. Next to the
deltidium, or on each side of it, are what may be called the “ delti-

* No microscopic sections of the shell have been made; as in all the speci-
mens possessing remains of the shell-substauce that have been under examina-
tion, these remains, owing to their being a methylosed replacement of calcite, or
dolomite, are not in a condition favourable for the retention of any original
histological characters. On this point, see ‘‘ Monograph of Spirifer cuspidatus,”
Annals and Mag. of Nat. Hist. 4th series, vol. ii. July 1868; and “ Histology of
the Test of the Class Palliobranchiata,” Transactions of the Royal Irish Aca-
demy, vol. xxiv. pp. 453, 455.

+ The pedicle- and brachial valves respectively correspond with those usually
called ventral and dorsal. While describing the various parts which characterize
the Trimerellids, our references will be confined to the figures or diagrams
given in Pl. XIT., unless otherwise stated; and we shall distinguish similarly
situated parts (irrespective of whether they are homologous, or analogous) of the
two valves—those of the pedicle- [or ventral] yalve by roman letters, and those
of the brachial [or dorsal] one by italics. It must not be expected that certain of
these parts will be found unless diligent and prolonged investigations are devoted
to them: they are to be sought for, not only by allowing the light to fall, on
places where they occur, in every favourable direction, but even at different
times of the day, and in different states of the atmosphere. We do not say that
our diagram figures are, strictly speaking, correct representations ; but we offer
them as reliable approximations, subject to immaterial corrections. Such, we
believe, will not vitiate the descriptions herein given of the interior of the shells
under consideration.

12

—

128 T. DAVIDSON AND W. KING ON THE TRIMERELLID&.

dial ridges” (c), prominent in some species, more or less flattened
in others, and marked with arching lamine: they are each defined
on its outer side by a well-marked incised line. The deltidial ridges
terminate near the hinge-margin, each as a cardinal callosity (e),
which is extremely variable in form and size. The areal borders
(d), usually narrow and transversely scored, differ in nothing from
those in other Palliobranchs.

In front of the deltidial slope is situated a tolerably well-defined
cardinal facet (f), the long axis of which corresponds to the trans-
verse direction of the hinge-line: it varies much in size according
to species. This part generally inclines or falls towards the cavity
of the shell: in one species, however, as will be noticed hereafter,
the inclination is in the opposite direction. Although in some
cases the cardinal facet passes without any marked break into the
inner surface of the valve, it is often raised by the hinge underneath
becoming thickened.

The part beneath the cardinal facet thus raised, while in many
species its thickness is in no way diminished vertically, is in some
others materially reduced in lateral dimensions, thereby presenting
itself under the form of a thin vertical hinge-supporting plate or
cardinal buttress (h).

The parts belonging to the hinge of the brachial valve are rude
and extremely variable in form: one of them may be variable in
position. The central portion of the hinge in certain species shows
nothing remarkable, no marks or prominences—due possibly to ero-
sion; but in others it distinctly displays an excavation or cardinal
_scar (PI. XII. fig. 2v), as in 7. ohioensis, also apparently in 7’. Dall

(PL. XV. fig. 2). The excavation is bicupped in 7’. ohioensis (Pl. XVI.
fig. 7); but the same species also presents it transversely grooved
(Pl. XVI. figs. 4,5, 6). There is no prominence in these cases.
In some other species, however, the cardinal scar, though in general
badly seen, appears to be situated on an elevation—rounded, or
squared, and more or less developed, according to species. Nay, in
the same species, as in 7’. Lindstromi, the elevation varies much in
size, certain individuals having it very small, others very large and
standing out like a great tooth—so large that it must have dipped
deeply into the umbonal cavity of the pedicle-valve: fig. 3, Pl. XIV.,
represents the elevation in its median size. In short, the cardinal
scar is so variable that to give a description of it applicable to any
one species would be a somewhat difficult task.

The hinge of the brachial valve is further characterized by another
variable part. Though absent in several species, there exist in others
two depressions or cardinal sockets (¢) on the hinge, one on each side :
when well defined (which, however, is seldom the case) by a promi-
nent border on the inner side, the cardinal sockets appear to have
received the cardinal callosities belonging to the pedicle-valve*.

* Lindstrom, who has represented the lateral margins of: a species of Trzme-
rella “ grooved by a furrow which helps to close the valves more tightly ” (Geol.
Mag. vol. v. p. 14), seems to think that this was the only provision the genus
had for a dentary apparatus. We cannot help thinking that he has taken too

T. DAVIDSON AND W. KING ON THE TRIMERELLIDA, 129

In certain species (Monomerella prisca) the hinge of the brachial
valve or, rather, we believe, its outside, is strongly striated trans-
versely: this peculiarity may be considered the same as the trans-
verse lines on the areal borders. In some cases the point of the
umbo of the brachial valve is replaced by a cavity (Mon. Walm-
stedti, Pl. XVII. fig. 3), apparently due to erosion; in others (7'ri-
merella ohioensis, Pl. XVI. figs. 5 & 6) it is rounded.

As regards the parts belonging to their interior, the Trimerellids
are peculiarly distinguished. In the medio-longitudinal region of the
posterior half of both valves there is what may be called a platform
{j and 7), large or considerably elevated in some species (PI. XIV.

figs. 2, 3, 6, &c.), small and even excavated in others (PI. XVI. fig. 8,

Pl. XVIII. fig. 13, Pl. XIX. fig. 4). In most cases the largest and
most elevated platform occurs in the brachial valve. Generally both
valves have the platform medio-longitudinally depressed on the top,
and rounded at the sides, thereby giving it a biconvex surface: it also
presents characters which, while constant in certain genera and spe-
cies, undergo marked modifications in others. ‘Thus in most Trime-
rellas each platform is distinguished by two tubular cavities—plat-
form vaults (k, k)—short, or nearly obsolete in certain species, and
in others running underneath the platform in its entire denies the
vaults are separated by a median partition (Pl. XVI. figs. 1, 2,3). In
a few species the platform is solid, or a little raised above the inner
surface of the valves, with the slightest indication of vaults (Pl. XVI.
figs. 8, 9). In Monomerella this part, though well developed, is not
vaulted i in either valve; it is the same in Dinobolus.

Returning to Trimerella, the median partition is prolonged aa
the anterior half of the shell under the form of a median plate (1, /).
It characterizes both valves, but is always the largest or longest in
the brachial valve (Pi. XLV. figs. 2, 3, &e.). Some congeneric spe-
cies appear to have had no plate (Pl. XVI. figs. 8, 9).

To complicate the camerated character of the Trimerellids, in the
species which have a cardinal buttress this part divides the interior
of the umbo (of the pedicle-valve) into two hollow sections or
umbonal chambers (1), large and wide-mouthed in the Monomerellas,
long and tubular in 7’rimerella Lindstromi. ‘There are species, how-
ever, belonging to both these genera, also Dinobolus generally, with-
out umbonal chambers; though in some eases indications of them
occur. Occasionally there are appearances of them in the brachial
valve: an unusually developed case occurs in this valve of Trimerella
Lindstromi, as shown by the pair of postlateral prominences (casts)

limited a view, probably arising from an examination of a specimen with long
lamelliform orders: such as occur, one on each side, on the hinge of 7’. ohio-
ensis (P|. XVI. fig. 7). In this species, however, there is a deepish hollow on the
inner or posterior side of each border, and seemingly corresponding callosities
on the hinge (where angulated) of the pedicle-valve: the latter may have be-
longed to the deltidial ridges. So variable are the cardinal sockets and cardinal
callosities, that we are quite prepared to meet with specimens having lamelli-
form borders developed or elongated somewhat as represented by Lindstrom.

130 T. DAVIDSON AND W. KING ON THE TRIMERELLID®.

in fig. 7, Pl. XIV. Nothing is more singular than the aspect of the
interior of the typical Trimerellids—some, as will now be under-
stood, having two specialized cavities (Monomerella), others four
(Pl. XIV. fig. 6, Pl. XV. fig. 44 &.), and others six (7. Lindstrom,
T. Dalli, &e.).

It may be stated, subject to correction, that the platform in each
valve is marked by four sets of scars, which, for the present, we
shall merely distinguish by the names medians (m, m), anteriors
(n, 2), laterals (0, 0), and postmedians (p,p). The postmedians, we
believe, are a set having no relation to the others.

In addition to the foregoing, the entire family 2s characterized by
a submarginal crescent (q, r, 8, g, 7, 8) confined to the posterior half
of each valve. It varies to some extent in each genus, also in dif-
ferent species: even different individuals of the same species do not
unfrequently present variations that seem to depend on differences
of age, thickness of the valves, and other peculiarities. As a conse-
quence, this part, or merely a portion of it, may present itself in relief,
excavated, obscurely, often partially in any of these conditions, or
complicated by successive enlargements of its area; so that we feel a
difficulty in representing it otherwise than diagrammatically, or in
giving a description that is strictly accurate when applied to any
single case. Closing these preliminary remarks, the crescent may
be described as formed of three portions—the crown (q, ¢), sides
(r, 7), and ends (s,s). The crown consists of a line running along

- the hinge, from each side of which it descends into the cavity of the

valves at the origin of their postlateral margins: this portion, in

_ the brachial valve, is pointed in the middle of the hinge (Pl XII.

fig. 6, 7); while, in the pedicle-valve, it passes along and close to
the outer edge of the cardinal facet. The sides, which are pointed
oval in shape in the typical genus, pass forward along and a little
within the postlateral margins of the valves. The ends, for the
most part obscurely defined, occupy the space between the widest
portion of the platform and the adjacent margin of the valve.
Somewhat similar in outline to the crown is a submarginal impres-
sion occurring in the anterior half of both valves: it resembles a small
arch or archlet (u, «),1s simple in form, consists of two halves, has its
two bases attached one to each end of the crescent, and throws itself
forward to within a short distance of the anterior margin of the valves,
apparently to a less extent in the pedicle than in the brachial one.
On each side of both valves one member of a pair of linear trans-
verse scars (t, t) strikes off from near the point of junction of the eres-
vent and archlet to the widest part of the platform, thence in front of
the latter part to the median plate: the other member makes a similar
traverse on the opposite side, and terminates at the same place.
Reverting to the hinge of the pedicle-valve, there is situated on
the cardinal facet an oval-shaped scar or lozenge (g), having its long
axis in the transverse direction of the valye. It is formed poste-
riorly by a line that runs parallel and near to the deltidial slope, and
anteriorly by what may be supposed to be another line running close
to the crown of the crescent. There is much obscurity in the part

T. DAVIDSON AND W. KING ON THE TRIMERELLID&. 131

under consideration, especially its anterior; for generally the latter
portion, and the line forming the crown of the crescent, are so inter-
blended as to prevent either one or the other being individually de-
fined: still we occasionally perceive appearances favouring the idea
that they are distinct.

Besides the foregoing there are some other scars that require
noticing. In the pedicle valve of certain Trimerellas there is an
appearance of an impression (fig. 1, w) situated below the hinge, and
on the cardinal buttress: in the Monomerellas two impressions occur
(fig. 3, w) similarly situated, 2. ¢. at the base of the cardinal buttress
or at the posterior end of the platform (j): and in Dinobolus David-
sont there are two distinct marks (inadvertently omitted in fig. 5, w)
below or on the outer edge (which is rounded off) of the hinge.
Suspecting that all these may form one and the same set, we pro-
pose to call them subcardinal scars.

The brachial valve of Monomerella and Dinobolus equally shows in
a similar position (that is, in front of and below the middle of the
crown of the crescent, or near the centre of the umbonal cavity) a
well-marked impression (figs. 4 & 6, w). There are faint indica-
tions of the same scar in Trimerella ohioensis (Pl. XVI. figs. 5 & 6).

Trimerella and Monomerella have in their pedicle-valve a pair of
umbo-lateral scars (x, figs. 1 & 3), a member being situated close to
and below each cardinal callosity (e). An apparently corresponding
set, but slightly marked, occurs in Dinobolus (x, fig. 5), occupying a
somewhat similar position.

The brachial valve of Monomerella exhibits an analogous pair
(fig. 4, v) within the crescent, a member being situated close to
each of its sides.

III. Myotocy anp oTHER CHARACTERS OF THE FAMILY.

We shall now offer some suggestions as to which organs the
various scars respectively belonged to; but before proceeding further
we find it necessary to anticipate a point which will be treated of in
another section.

Having failed in making out any satisfactory resemblance be-
tween the Trimerellids and Terebratulids in their internal parts,
we were led to extend our investigations to another family. It was
with this object that one of us entered upon an examination of the
various muscles, and scars belonging to them, that characterize Lin-

guia anatina*. The result has been attended with much more
success than in the case of the Terebratulids.

In the species named, the valvular muscles are of two kinds,
direct and slanting. The direct muscles are an anterior pair, situ-
ated in the central region of the shell, and a single one, lodged in
the umbonal cavity. The slanting muscles are four pairs :—one pair,

* Annals and Mag. of Nat. Hist. 4th ser. vol. xii. July 1873. When speaking
of the muscles and other parts of Lingula anatina, our references will be made
to the figures in Plate II. accompanying that paper. Sketches after the two
principal figures are given in the annexed woodcuts in the following page.

132 T. DAVIDSON AND W. KING ON THE TRIMERELLIDZE.

transmedial, crossing in a backward direction from the middle of the
sides of the posterior half of the brachial valve to the reverse sides.
of the pedicle-valve ; the remaining three pairs (laterals) pass from

Fig. 1.—Lingula anatina, to dlustrate the Scars in Trimerella.

>
==N

\ Vico
SS

\ Y
S eZ
= \eZ

ce
= (ES

: fee

3 i |—
Be s

2 \ | =
se Pde Ne - =
B We Nag Oi ie | Se

Z KK We C ' 4

Z a sy) ; ee ea

Ky

Gytliy WS:

ATTN
FITTS
Brachial valve. Pedicle-valve.

g. Umbonal muscle. h. Central muscles. z. Transmedial muscles: 6, Parietal
band. J, &, 2. Lateral muscles (7, anteriors; 4, middles; J, outsiders).
A. Splanchnoceele. B. Pleuroceles. C. Brachioccele.

the sides‘of the same half of both valves to points further forward and
adjacent to the median line of the shell, but without crossing it.
Reverting to the Trimerellids, we shall make one or two sugges-—
tions respecting the subcardinal scars (w, w). First, they may have
belonged to a muscle corresponding with the umbonal one of Lingula.
If, as appears probable, the single umbonal muscle, just referred to,
is the homologue of the two posterior adductors of Discina, there is
no reason against assuming that the presumed corresponding muscle
of the Trimerellids has been simple in certain genera (Pl. XII.
fig. 1, w), and compound or double in others (figs. 3 & 5, w)—or
divided at its attachment in one valve, and undivided in the other.
Second, they may be the same as the part in the brachial valve we
have called the cardinal scar (v, fig. 2)—a view, however, which re-

TY. DAVIDSON AND W. KING ON THE TRIMERELLID®. ieee

quires the pertaining muscle to have been divided at its extremity,
one division attached to the hinge, and the other to the umbonal
cavity. But another idea regarding the cardinal scar (v) has been
adopted by Dall, that it afforded attachment to a muscle homologous
with that belonging to the cardinal process of the Terebratulids.
Certainly from its position, and especially when seated on an eleva-
tion, as in Trimerella Lindstrom, this scar has much the appearance
of being identical with the process characteristic of the hinge in the
family just referred to; but we think a more correct appreciation of
its true relationship is involved in the suggestion that it is homolo-
gous with the small pair of postlateral or converging muscles, of
which, as in Discina, a member passes from near each side of the
pedicle- or flat valve to the space between the posterior adductor -
muscles in the brachial or convex valve. This suggestion, it must
be understood, does not necessitate the abandonment of the idea that
the scar (w, w) belonged to the umbonal muscle: only, instead of
this muscle being divided, with a division implanted on each side of
the pair of converging muscles, as in Discina, it may have been
simple in some cases, and have passed between the latter muscles.
Such an arrangement would be the reverse of what prevails in Dis-
cina ; but it does not seem to be opposed by any valid objections.

With respect to the slanting muscles of Lingula, we strongly sus-
pect that the transmedial pair was absent in the Trimerellids—the
close union of the valves at the hinge, or the possession of a dentary
apparatus, rendering impossible any lateral movement of the valves
in their posterior or umbonal region, such being the function of the
transmedials.

The three remaining pairs of slanting muscles, which have been
generically designated laterals, and severally named “ anteriors,”
«‘middles,” and “ outsiders,” appear to us to have characterized the
fossil family.

The “anterior” pair of Zingula may correspond with, in the Trime-
rellids, the anterior scars, n*, of the brachial valve, and the ends of the
crescent (s), of the pedicle one; the “ middle” pair with the median
scars, m, of the pedicle-valve, and the ends of the crescent, s, of the
brachial one; the “outside” pair with the lateral scars, 0, of the
pedicle-valve, and the ends of the crescent, s, of the brachial one.
In accordance with these suggestions, the letter s (that is, in the
brachial valve) denotes a compound scar; but, instead of being
triple, as in Lingula (which, it must be borne in mind, has, in addi-
tion, one of the terminations of the transmedial muscle implanted in
the corresponding part), it would be double.

Moreover, in Lingula anatina a pair of slanting muscles is im-
planted in the brachial valve, adjacent to, or upon the medio-
longitudinal ridge, a member on each side, where the scars left by
the muscles are generally well seen. We have already assumed these
scars to correspond with those lettered x in figs. 4, 6, that is, on the.

* We have not lettered that portion of the platform in Pl. XIT. fig. 2, where, ap-
parently the scars would be situated, as there are grounds, presently to be noticed,
for believing that they were in this instance transposed to the median plate (/).

134 T. DAVIDSON AND W. KING ON THE TRIMERELLIDA.

projecting or anterior portion of the platform belonging to the
brachial valve of Monomerella and Dinobolus; which genera, it will
be seen, have no median plate, or only a rudimentary one. But
considering that in the brachial valve of Trimerella this part is as
well developed as the ridge in Lingula, or even more so, we are
strongly persuaded that it or its posterior end was marked by the
scars in question.

If the median plate of the brachial valve were not a muscular
fulcrum of the kind, it would have to be considered merely analo-
gous to the medio-longitudinal ridge of Langula, and not its homologue;
or, instead of belonging essentially to the splanchnoccele, it must be
an integral part of the brachiocceele. The latter view, however, has
nothing in its favour; on the contrary, in all known living Pallio-
branchs (and most of them have it) the ridge, when present, lies
within the splanchnoceele, and serves as a muscular fulerum.

The corresponding plate in the pedicle-valve of Trimerella may
have been subordinated in the same way; but we are more inclined
to believe that it merely strengthened the platform, as we think was
also the case with the rudimentary one in the brachial valve of
Dinobolus (fig. 6, 1).

The aberrant Trimerella Billings and T’.. galtensis do not possess
the median plate in the brachial valve: in these cases the muscles
would be attached to the anterior portion of the platform, as in
Dinobolus and Monomerella.

Besides their use as muscular fulera, it may safely be admitted that
the platforms, to some extent, supported the viscera. We are in-
duced to refer the postmedian scar (p, fig. 6) to the ovaries. These
organs have been allocated by Dall and others to the platform-vaults ;
but their usual position in living Palliobranchs would require them
to be lodged in, or connected with, the posterior portion of the
splanchnocele. We are consequently more disposed to believe,
considering their postlateral situation in Rhynchonella*, and in
other shells of its class, that the ovaries were similarly situated in
the Trimerellidst: the large umbonal cavities characteristic of cer-
tain species, seem well adapted for containing them.

The platform-vaults appear, from their contiguity to the usual
locale of the liver, to be the most likely receptacles for the divisions
of this organt. Another view has been taken respecting the use of

* Hancock, Transactions of the Royal Society, vol. cxlviii. pl. lx. fig. 3, 1858.

t Certain genera, especially Camarophoria and Pentamerus, haye hollows in
the umbo of the pedicle-valye, not only corresponding with, but evidently ful-
filling the same office as, the umbonal chambers of the Trimerellids. Many
years ago one of the writers, studying Camarophoria multiplicata, expressed his
belief that the spaces within the “vascular circles” occurring in this species
“were the seats of reproduction.” Monograph Permian Fossils, p. 116. :

{ In our previously published “Remarks on the genera Trimerella, Dino-
bolus, and Monomerella,’ we instanced the excavated scars of Obolus, which haye
an overlapping posterior margin, thereby approaching in appearance to the
platform-vaults of Zrimerella; but these are undoubtedly due to muscular at-
tachment. In Leptena Dutertrii the myophore of one of the valves is doubly
vaulted, possibly for holding portions of the liver and a case homologous with
the platform-vaults of Trimerella.

=

T. DAVIDSON AND W. KING ON THE TRIMERELLIDZ, 135

these vaults. Lindstrém supposes them to have been hollows in which
the muscles were implanted*. We have searched for impressions
likely to have resulted from such implantation, but without success ;
and it appears highly improbable that the vaults, which run the
entire length of the platforms up to the hinge in certain species,
could have served this purpose+.

We may leave for a short while the consideration of the internal
parts of the Trimerellids to take a passing notice of the hinge-cha-
racters. These clearly show that the shells have been furnished
with a massive pedicle, closely agreeing in its attachments with that
of Lingula. As in this shell, the proximal extremity of the pedicle
in the fossil (doubtless in the form of a flatly compressed cylinder—
that is, with two flattened faces) was evidently attached by one of
its faces to the entire surface of the deltidium, the imbricated la-
minee of which are the marks of attachment. The pedicle of Lin-
gua is further attached by the end of the same face, also by two
lateral extremities or lobes, the former to a narrow lineated space in
front of the deltidium, and each of the latter to a point on both sides
of the narrow spacet. It cannot be objected that the deltidial slope
(b) of the Trimerellids marks the attachment of the end of the
adhering face of the pedicle; and we strongly suspect that the del-
tidial ridges (c) and cardinal callosities (e) have resulted from the
attachment of the laterallobes. In certain of these fossils, especially
in Trimerella galtensis, the deltidium displays a deep median groove
(Pl. XVIII. fig. 13), called by Hall the “ central pedicle-groove,”
a peculiarity we would suggest as being due to the strong attach-
ment of the middle part of the adhering face of the pedicle, as
evidence of the kind occasionally occurs in specimens of Lingula
anatina. It requires to be mentioned that in the latter shell the
face, end, and lobes spoken of belong to the corneous sheath of the

* Lindstrom, who speaks of the platform-vaults as “ prolonged muscular
scars,” states that the corresponding parts in Lingula are occupied by two im-
pressions of the adductors. Geol. Mag. vol. v. p. 442, 1868.

+ In a letter received by us from Mr. Whitfield, dated the 4th of Dec., 1871,
he states, ‘In relation to the uses of the cavities beneath the plate [platform],
they certainly could not have contained any part of the adductor muscles; but,
from the large size of these muscles necessary to work such large and, in one
species [7Z\ ohioensis], ponderous shells, the greater part of the surface of this
part (the cardinal half) of the shell would necessarily be taken up by them, and
there would be but little space left for the pallial smuses containing the testes or
ovaries; and the plate [platform] is consequently raised to get these large
muscles out of the way in the centre of the valve, and to let the organs (sexual)
pass beneath them. This I conceive to be the use of them. I have often thought
the matter over when explaining these and allied forms. In one form, Pentamerus
galeatus, many of which I have examined, I know the roughened spaces known
as ovarian spaces are carried inwards, close to the sides of the thin central septum,
leaving no space between them such as we see occupying the centre of Orthis
and allied forms, as well as in most Brachiopoda.”

t We have previously noticed the two incised lines in connexion with the
deltidial ridges: these lines, we believe, are likewise produced by the attach-
ment of the lateral lobes; also the corresponding lines occasionally seen on
the area of certain Spirifers. The transverse strize on the areal borders, as well
as those on the outside of the hinge in the brachial valve, are evidently due to
epidermal or external lines of growth, as in Lingua.

136 TY. DAVIDSON AND W. KING ON THE TRIMERELLID A.

pedicle. Besides these parts, this organ is attached by the end
(compressed) of its inner fleshy or muscular cylinder to an ellipsoidal
space in front of the narrow lineated band, and behind the umbonal
muscle. The part similarly situated in the Trimerellids, and named
lozenge (g, figs. 1 & 7), we attribute to the attachment of the same
cylinder. Referring again to Lingula, the submarginal portions of
the posterior half of its valves are distinguished by a compound scar
consisting of :—a line on the hinge; two eapansions, one near each
of the postlateral margins; and two longitudinally elliptical spaces,
one connected with each of the expansions, and terminating at about
the transverse median line of the valves. The first is formed by
the postparietal of the splanchnoccele ; the second, their inner side
by the post-lateroparietals of this chamber, and their outer side by
the posterior primary vessels belonging to the pleuroceeles; the third
mark the attachment of the slanting muscles, which are limited on
their outer side by the lateral walls of the splanchnoccele.

That the crescent of the Trimerellids is made up of the same parts
(though subject to modifications, as exemplified in Dinobolus), its
crown, sides, and*ends respectively representing the first, second,
and third, is a proposition which, we think, requires nothing further
for its establishment.

In Lingula anatina the splanchnoccele is bounded in part by another
wall, the anteparietal. The transverse scars (t, ¢) in Z’rimerella
occupy a similar position, evidencing that they were produced by the
same wall. A further agreement consists in the anteparietal of the
recent shell running forward, in the brachial valve, with a duckbill-
shaped outline, considerably into the brachioccele ; for doubtless a
' similar peculiarity obtained in the fossil through the attachment of
the same wall to the sides of the median plate, as it does to the cor-
responding ridge of Lingwla, with this difference, however, that the
projection would be longer, and its base narrower. ‘There is no
appearance of the transverse scars starting off from any place in
front of the platform: they run close up to the median plate, thence
apparently forward, and along éach side of it; which would neces-
sarily cause the attached portion of the anteparietal projection, Ons
in the brachial valve, to be comparatively narrow.

The archlet was evidently produced by the attachment of the
anterior pair of primary vessels. In Lingula anatina these vessels
do not, strictly speaking, unite at their anterior extremity; and it
appears to have been the same in 7rimerella, as the archlet seems
often to open at its apex. In the brachial valve of the Trimerellas
(7. grandis, Pl. XIII. fig. 2b) the archlet has also a remarkable
agreement in form with the course pursued by the pair of primary
vessels in the recent species referred to. ‘The opposite valve of the
latter shell shows the same vessels, but more shortened : the archlet
belonging to the corresponding valve of the fossils possesses a similar
peculiarity. We have not yet met with any impressions producing
secondary vessels.

It will be understood that we are now in the brachioccele—the
chamber holding the arms or labial appendages. So far no convoluted

7. DAVIDSON AND W. KING ON THE TRIMERELLID&. 137

impressions referable to these organs have occurred to us: they
have been found in other fossils*. Such impressions, however,
may yet be detected, especially of the apical portion in the pedicle-
valve.

In some specimens a shallow submarginal groove is present, which
is all we have seen that could be attributed to the setal band. Itis
extremely doubtful, from the construction of the hinge, that this
organ would be carried round the posterior margin of the valves, as
in Lingula.

TV. Arrrnitres oF tHE Famity.

Dall places the genus Trimerella in the family Lingulide; of
which he makes two divisions (Linguline and Oboline), remarking
that its nearest affinities are with Lingula and Obolus, but adding
“perhaps it may be eventually placed in the second division of the
family”. Hall, referring to his Dinobolus Conradi, prefers the
same allocation, asserting that it is closely allied to Obolust. The
same may be said of Billings, who states that “both Trimerella and
Monomerella are subgenera [evidently a mistake for subgroups] of
‘Obolus:” he also adds that his O. canadensis and O. galtensis belong
to “a third group” §.

Quenstedt, the only continental paleontologist who has touched
on the subject of its affinities, includes 7’rimerella in his family
Ungulite ; which he makes to consist of Obolus, Schmidtia, Obolella,
and Acrites ||.

One of the points that has led these authors to affine Trimerella
with Obolus is possibly the resemblance there is between the cres-
cent of the former and certain scars which curve forward from the
hinge of the latter. Until one of us had ascertained that the scars
alluded to in Obolus represent the posterior adductor muscles of
Discina4, and not other organs, as in Lingula, we were also inclined
to take the view of our colleagues: now, however, as will be under-
stood from what is stated in the previous sections, we have aban-
doned it**.

Believing that the Trimerellids are closely related to the Lingulids,
but distinct as a family, we shall next enter upon a comparison
between the two groups by tabulating their differences and agree-
ments :—

* Such as Productus, Chonetes, Strophalosia, Strophomena, Koninckina, Ano-
latheca, Davidsonia, &c.
+ American Journal of Conchology, vol. vi. part 2, p. 153, 1870.
¢ Twentieth Annual Report of the Regents of the University of the State of
New York, 1868; and revised edition, p. 376, 1870.
§ Canadian Naturalist, vol. vi. no. 2, p. 222.
|| Petrefactenkunde Deutschlands; Die Brachiopoden, p. 668, 1871.

@ Ann. Nat. Hist. ser. 4, vol. xii. p. 15.

** Tt may be mentioned that an examination of a large number of specimens
belonging to the genera Obolus, Schmidtia, Obolella, Acrites, &e. have led us
to group them all together under the family Obolide, proposed by one of tho
writers in 1850: See Monograph of Permian Fossils, p. 81.

138

Points in which the Trimerellids may be

said to differ from the Lingulids.

. Calcareous shell-substance.
. Closed at the hinge.

. Dentary, though rudely.
. Brachial valve with a cardinal pro-

or

T. DAVIDSON AND W. KING ON THE TRIMERELLIDZ.

Points in which the Trimerellids may be

said to agree with the Lingulids.

Deltidium and its accessories.

Mode of attachment of the pedicle
to the deltidium.

Muscular system generally.

Splanchnoceele in its general out-

line.
. The possession of pleuroceeles.

cess.
. Apophysary system generally ele-
vated platforms, one in each valve;
and the platforms in the type
genus doubly vaulted.
6. Absence of transmedial muscles, | 6.

on PO De
Sk peo).

General arrangement of the vascular

and modifications of the umbonal system.
muscle.

7. Setal band not carried round the | 7. The combination of certain parts
hinge. or organs, producing thereby the

posterior crescent.

The differences between the two groups, we contend, are quite
sufficient to warrant the step we have taken in making of each a
distinct family; while, on the other hand, their points of agreement
are so numerous as to show that they are intimately related. Obolus
is placed by Dall in the family Lingulide; but its close relationship
to Discina opposes the allocation. This relationship alone prevents
our placing the Trimerellids in any family group which embraces
Obolus, though we are quite ready to admit that both belong to one
and the same great division in their class.

At first sight the Trimerellids, considering their general external
resemblance to Uncites, Stringocephalus, and other genera, might be

‘considered to belong to the great subclass, Clistenterata; but, although
certain parts belonging to their hinge seem to favour a different view,
it cannot but be admitted that the consensus of their characters
completely establishes them as Tretenterates*.

V. GeotocicaL Ranex, CHRoNoGENESIS, AND EvoLUTION OF THE
Famiry.

The Trimerellids, as will be seen hereafter, characterize two consecu-
‘live geological systems—the Cambro-Silurian, and Silurian. Neither
the Cambrian nor Devonian rocks have as yet yielded any repre-

* Tretenterata (open-gut) and Clistenterata (closed-gut) form two great sub-
divisions of the Palliobranchs or Brachiopoda. The names Plewropygia and
Apygia, proposed some years ago by Bronn, stand for the same groups. The
latter, bemg a negative term, is totally inadmissible; while the former not. only
prejudges and limits the position of an important organ in a variable group of
shells, but 1 seems likely to turn out to be no more than of sectional value. The
singular aperture, and tube, respectively characterizing the hinge-plate of the
brachial valve of Athyris pectinifera, and A. concentrica, also the foramen
occurring, 77 addition to an open deltidiuwm, in the umbo of the pedicle-valve of
several other palzozoic shells apparently belonging to the Leptenids, Spirife-
rids, Obolids, and Rhynchonellids, may be regarded as indicating, in such cases,
the posterior position of the intestinal outlet.—W. K.

t We include in the Cambrian system the formations below the Arenigs—
namely, the Tremadocs, Ffestiniogs, Menevians, and Longmynds : as such it an-
swers to the Primordial group of Barrande. The Arenig, Llandeilo, and Bala

T. DAVIDSON AND W. KING ON THE TRIMERELLIDZ. 139

sentatives. All the known species of Trimerella and Monomerella,
hitherto only found in Sweden, the United States, Canada, and
Russia, are from beds equivalent in age to the Wenlock and Aymes-
try formations of Great Britain. Dzénobolus, however, has obtained
a more extended vertical and horizontal range: of the seven forms
at present known, D. Schmidti (a Russian species), D. magnificus,
and D. canadensis (both from Canada) are Cambro-Silurian—the two
species last-named having been found in the Trenton or Black-river
Limestone, about the horizon of the English Upper Llandeilo or the base
of the Caradoc. Dinobolus Conradi, confined to the United States, be-
longs to the Niagara Limestone(Upper Silurian). D. Davidsoni appears
to have been found in the Llandovery formation, as well asin that of the
Wenlock in Sweden; but it, with D. Woodwardi and D. transversus,
chiefly characterizes the Wenlock in the counties of Shropshire and
Staffordshire: the latter three are the only British species known
belonging to the family.

It is a remarkable peculiarity in Lingula anatina that the post-
lateral parietals of the splanchnoccele, unlike the membranous coun-
terparts in other Palliobranchs, are highly muscular—so much more
than seems to be needed to protect the viscera as to give rise to
the idea that they also served to some extent to keep the valves
together. It is therefore not improbable that in some forms
related or belonging to the Lingulids the muscularity of the
walls referred to was still more developed, thereby causing them
to become approximately valvulars or adductors. From this con-
dition to complete specialization, such as is presented by the
pair of postadductor muscles that characterize Discina, seems to
be an easy gradation. Obolus may be cited as an additional ex-
ample. Although a considerable number of Cambrian Tretenterates,
usually recognized as Lingulide er Linguloid shells, are known%, it
nevertheless happens that information regarding certain of their in-
ternal characters is too limited to enable paleontologists to select any
of them as constituting the gradational form in question. A better
source seems to offer itself in the family under description.

Neither Trimerella nor Monomereila appears likely to assist us,
inasmuch as the scars (sides of the crescent) due to the post-latero-
parietals do not differ notably from their equivalents in Lingula.
In Dinobolus, however, the same scars are strikingly different; each
one has a composite structure, with two, or more separated indenta-
tions strongly marked—showing that it has been produced bya muscle,
ormuscles, unusually large and powerful. Though evidently belonging
to the postlateral walls of the splanchnoccele, very little modification

or Caradoc formations we group together: and in justice to the labours of
Sedgwick and Murchison, also in accordance with the compromise proposed by
the first cited of these illustrious men, we name them collectively, as others have
already done, the Cambro-Silurian system.

* See paper “On the Earliest Forms of Brachiopoda hitherto discovered in
the British Paleozoic Rocks,” Geol. Mag. vol. v. no. 7, July 1868, by one of
the present writers. Some minute American fossils, described by Hall, Bil-
lings, and others, may be included in the same category.

140 {, DAVIDSON AND W. KING ON THE TRIMERELLID A,

of the muscles represented by the scars referred to would seemingly
convert them into specialized adductors.

The peculiarity exhibited by the s¢des of the crescent, as just
noticed, is of importance in our present inquiry, inasmuch as it cha-
racterizes the earliest genus of the Trimerellids ; and of equal import-
ance is the fact that Dinobolus is the least calcareous of its group.
This genus therefore stands out not only as an initial form, but as
indicating the period of development or chronogenesis of the “family.
We may go further, and assume that, while the Trimerellids strongly
retain many of the leading characters of the more ancient family
(Lingulide), the genus Deas, by its crescent, discloses a relation-
ship to another group strongly differentiated by its myology. As-
suming this to be correct, we shall have to account, on the doctrine
of evolution, for the creation* of the Trimerellids out of some pri-
mordial or early form possessing a stronger tendency to become
characterized with specialized adductor muscles, situated like the
postadductors of Discma, than is displayed in the recent Lingula
anatina, There is no improbability in the idea that a generalized form
will yet be discovered, either in the Cambrian or early Cambro-
Silurian rocks, bringing Discona and Lingula, also Obolus, into close
myotie relationship inter se and with Dinobolus. Such a form would
constitute the root from which the Trimerellids have originated.

VI. PrystogRaAPHY OF THE SEAS TENANTED BY THE 'RIMERELLIDS, AS
COMPARED WITH THAT OF THE CAMBRIAN SEs.

One of the peculiarities which characterize the Trimerellids as a
family, in comparison with the related group the Lingulids, is the
chemical composition of their shell-substance. The latter family
comprises forms whose valves are in great part corneous (consist-
ing of the organic “‘ basement membrane” of Bowerbank), with ad-
ditions of calcium phosphate and carbonate—the phosphate being
the largest in amount ; whereas the former one includes essentially
calcareous species.

These differences seem to be in intimate relation with certain
physiographical phenomena that distinguished the Cambrian from
the Cambro-Silurian and Silurian periods.

In the Cambrian period, and, as far as known, before the Trime-

* The terms creation and evolution, it requires to be understood, are not used
as meaning the production of one species (hypothetically an interreproductive
group) independently of, or owt of another, in the second case effected solely by
extraneous agencies or circumstances. One doctrine is unscientific ; the other is
based on phenomena of no more than a secondary or subsidiary character. Evo-
lution, taking it to be exemplified by metagenesis and metastomotosis, respectively
characterizing the Hydrozoa and Batrachia, is obviously the result of innate
formative action adapting itself to extraneous influences or conditions: the
cases referred to do not differ in their causation from the great progressive
mutations which the life-system of our planet has passed through, as revealed
by paleontology. ‘The variations produced in plants and animals by external
influences constitute a highly important study, giving rise, if considered in con-
nexion with the doctrine of ¢xnate adaptive formativity, to the noblest and most
exalted conceptions that man is capable of forming.—W. K.

T. DAVIDSON AND W. KING ON THE TRIMERELLID®. 141

rellids were in existence, limestones are no more than exceptional
deposits amongst their argillaceous and siliceous associates. On the
contrary, in the Cambro-Silurian and Silurian Periods (that is, during
the role of the Trimerellids), as ever since, depositional phenomena
became remarkably characterized by caicareous elaborations.

It is equally noteworthy that the Cambrian life-system is but
poorly represented by forms in whose skeletons calcium compounds
are essential constituents. Although representing groups most of
which afterwards included members possessing a massive calcareous
framework, the Coelenterates, Mollusks, Echinoderms, and Crusta-
ceans of the Cambrian period had slender skeletal structures, in which
lime was not particularly abundant; and they thus stand out in
striking contrast to their massive successors of subsequent periods.

Doubtless in the Cambro-Silurian and Silurian seas — those
tenanted by the Trimerellids—ordinary marine calcium compounds
prevailed as at present. But, in view of what has already been
stated, are we to suppose that such compounds did not occur to the
same extent in Cambrian seas? Or are we to imagine that the
Cambrian organisms were incapable of abstracting lime from the
medium they lived in? As few, we think, will be found to enter-
tain the latter idea, it may be dismissed to leave room for the con-
sideration of the other one.

It is well known that the Archzan rocks* are of enormous
thickness. In the British Isles they consist of the ordinary meta-
morphics, composed of silacid minerals. In North America, especially
in Canada, where they apparently comprise two (or more) uncon-
formable series—Laurentian and Labradorian—they include both
silacid and silo-carbacid rocks+. The “ zones” (“ bands,” or
“‘masses”) of “ Crystalline limestone,” of the Canadian Geological
Survey, which constitute the silo-carbacid group, attain an aggregate
thickness of some thousands of feet +.

The Cambrian rocks, though varying in certain respects in
different regions, agree pretty generally in their lithological features.

* Dana, rejecting the term “ eozoic,” which has been applied to them “to
indicate the morning of that great creative day in which the lower forms of
animal life were introduced upon our planet” (Dawson), distinguishes the
Laurentian and associated metamorphic rocks, older than the primordials, by
the name Archean. This change very properly does not reject the idea that
undoubted or indisputable organic remains may yet be found in such rocks ;
for it is quite likely that the term eozoie will share the same fate as that of
azoic, as proposed originally for the Cambrians.

+ For the distinctive characters of these rock-divisions, see a paper in tks
Geological Magazine, January 1872, entitled, ‘‘ The Microscopic Characters of
a Silo-carbacid rock from Ceylon; and their Bearings on the Methylotic Origin
of the Laurentian ‘limestones.’ ”’

+ In both the upper and lower series of the Archean rocks, “the united
thickness of which in Canada cannot be less than 30,000 feet, and probably
much exceeded it,” there are several zones of limestone. Of these it has been
ascertained that three at least belong to the lower series.” The “ Zone of
Grenville limestone [the lowest of the series] is in some places 1500 feet
thick.” At Clarendon the Archean rocks have a ‘‘thickness of 4000 feet,
about two thirds of which (2666 feet) consist of Crystalline limestone.”—Logan,
Geology of Canada, p. 31, 1863.

Q.J.G.8. No. 118. M

142 {T. DAVIDSON AND W. KING ON THE TRIMERELLID®.

Now, as regards the silacid members of the Archzans, there is no
difficulty in understanding how they have given rise to the argillites
and sandstones of the Cambrians. But where are the limestones
that we ought to expect would have. heen formed by contemporary
reproduction out of the Silo-carbacid rocks? We cannot be satisfied
with the answer if it refer to the American cases represented by the
primordial limestones of Belle Isle, Troy (N.Y.), and other places, or
the calciferous sand-rock usually included in the Potsdam formation™. .

Turning next to the British Cambrians, we shall be met by a
still less satisfactory answer. The Longmynds, Menevians, Ffes-
tiniogs and Tremadocs, which range from twenty thousand to thirty
thousand feet, or more, in aggregate thickness, consist of all the
rock-forming materials except the one we are searching for. The
lapse of time during which the argillites and sandstones of these
groups were in process of formation must have been immense ; and
not only are they divested of limestone, but calcareous fossils are
exceedingly rare in them.

The cases cited form one of the most difficult problems in prim-
ordial geology ; for considering the “ masses” of ‘‘ Crystalline lime-
stone” characterizing the American Archeans, the totally incom-
mensurate amount of calcareous rocks amongst the Cambrians seems
to be unintelligible: and the difficulty is further increased by the
fact that contemporary organisms rarely had skeletons in which lime
is the predominant constituent.

We shall make no attempt to solve the problem: all we can do
at present is to call attention to the subject.

Certainly, it may be said in conclusion, there are some grounds
for believing that great physiographical differences prevailed between

the Cambrian and the succeeding periods—that great changes in life
and sedimentation, of a persistent nature, began during the early
portion of the Cambro-Silurian, and have been maintained ever
since. ‘The Trimerellids appear to have played no unimportant part
in the beginning of these changes.

VII. Draenosis or THE FAmILy.
Class PaLLioBRANCHIATA, De Blainvillet.
Subclass Trerrnterata, King f.
Family Trimerellide, Davidson & King§.

Shell-substance of the valves calcareous. Species generally massive,
especially in the umbonal region. Umbo of the pedicle-valve often

* The limestones of the Quebec group are excluded, as they are now generally
considered to be on the horizon of the Lower Llandeilo formation ; and it is
even possible that the Calciferous sand-rock is at or near the base of the
Cambro-Silurian system, and equivalent to the Arenig group.

t Brachiopoda, Cuvier 1805; Dumeril, 1806; Lamarck, 1809; and of the
generality of paleontologists. Palliobranchiata, Blainville, 1824.

t Annals and Mag. of Nat. Hist. 4th Ser. vol. xii. July 1873.

§ Cuvier’s name, Brachiopoda, is a misnomer, as it is now generally admitted

T. DAVIDSON AND W. KING ON THE TRIMERELLID&. 143

large ; pointed; projecting variously; solid, or hollow: its hinge-
face with a well-developed area. Area usually of considerable size ;
with a large delttdium, which is solid throughout. Hinge of both
valves rudely or faintly dentary: that of the pedicle-valve thick ;
entire in its whole length ; and more or less elevated ; supported in
the latter case by an upright rib or cardinal buttress of varying thick-
ness ; with a wide median space or cardinal facet enclosing a lozenge-
shaped scar: that of the brachial valve with a more or less elevated
median prominence, or depression. Attached to the interior surface
of the posterior half of both valves is a more or less elevated plat-
form; which is medio-longitudinally situated, and solid, or doubly
vaulted: from the middle of its anterior end a median plate occa-
sionally projects into the anterior half of the valve, especially the
brachial one. Both valves have a compound impression or evescent
running a little within the margins of their posterior half, including
the hinge. A submarginal impression or archlet characterizes the
other or anterior half of the valves.

The detailed description already given of the various parts cha-
racteristic of the family renders it unnecessary to add any thing by
way of general observations. Variations of these parts, to a con-
siderable extent, occur in the whole group. It is principally on the
variations of the platform, and other parts belonging to the cardinal
region, that the genera Trimerella, Monomerella and Dinobolus are
founded. As in other families, the present one is not without
species which cannot be satisfactorily allocated generically.

VIII. Tae Gunvs TRIMERELLA AND ITs SPECIES.

Genus Trimerella, Billings, 1862*. Obolus (galtensis), Billings,
1862. Gotlandia, (Dall), 1870. Rhynobolus (galtensis), Hall,
March 1871. Obolellina (galtensis), Billings, Dec. 1871.

Valves thick ; longitudinally oval. Umbo of pedicle-valve usually
massive; solid; occasionally double-chambered; irregularly pro-
jecting. Area of considerable size; longer than wide. Deltidium
large. Hinge generally thick, and elevated; rudely or slightly
dentary ; and variously modified in different species. Cardinal facet
large. Crescent rather well defined in typical species. Platforms
elevated, and doubly vaulted ; occasionally solid, and slightly raised.

that the brachial appendages are not locomotory organs. It seems probable that
locomotion is effected by the pedicle, at least in shells in the young state: this
view is supported by Morse’s observations on Linguia pyramidata, and by the
evidence of Stoliczka, who, from what he saw of Lingula anatina on the coast of
Arrakan, is inclined to believe that it is capable of changing its place actually
by movements of the pedicle: see Mem. Geol. Survey of India, vol. iv. 1, p. 5.
Blainville’s name, Palliobranchiata, is unobjectionable, as the pallial lobes are
the principal respiratory organs.

* Geological Survey of Canada, Palxontology, p. 166, 1862. The name Z77-
merella, like Monomerella and Dinobolus, is unfortunately not above criticism.

M 2

144 T. DAVIDSON AND W. KING ON THE TRIMERELLID A.

Median plate occurring generally in both valves; longest in the
brachial one.

Type species Trimerella grandis, Billings. ;

The genus varies much in some of its chief characteristics. ‘The
umbo of the pedicle-valve, usually large, subconical, and occasionally
double-chambered, is exceptionally flattened in 7. Billingsii and T.
galtensis ; and the hinge of these species is consequently very slightly
elevated: on the contrary, in 7. Dalli the umbo is prominently
double-chambered, as in the next genus. The hinge has a central
part, projecting in certain species, and excavated in others. In 7.
Billings, and T. galtensis, the platforms are slightly raised or
depressed, a peculiarity which remarkably differentiates them from
the typical species.

The genus Lthynobolus, proposed by Hall for Obolus galtensis,
appears at first sight to possess sufficient distinctive characters ;
but, comparing its type with 7. Billingsit, we cannot see how
it can be separated from the latter generically: both species are
similarly characterized by flat longitudinally-elongated valves, a
flattened umbo in the pedicle-valve, a large area and deltidium, and
solid or slightly elevated platforms. The distinctions made out by
Hall for 7. galtensis appear to us to be insufficient ; as through the
depression of its platforms the anterior portion of these parts, in-
cluding their scars, show nothing more than what is seen in 7’.
Billingsu. Both species may be considered aberrant forms of the
genus; the former departing somewhat further from its type than
the latter.

TRIMERELLA GRANDIS, Bill. Pl. XIII. figs. 2, 3.

Trimerella grandis, Bill. Geol. Survey of Canada, Paleontology,
p- 166, fig. 151, 21 Jan. 1862. Dall, American Journal of
Conch. vol. 1. part 2, p.. 160, 1870, vol. vu. p. 82, 1871.
Quenstedt, Petrefactenkunde, p. 172, pl. lxi. fig. 42, 1871.
Hall, Explanation of figures in “ Notes on some new or
imperfectly known forms among the Brachiopoda,” pl. xiii.
figs. 11-16, 1872. Day. and King, Report of Brighton Meet-
ing of Br. Assoc. 20 August, 1872.

Shell much longer than wide ; broadly rounded anteriorly ; taper-
ing posteriorly to an almost pointed beak; greatest breadth at about
two thirds from the beak; surface smooth. Pedicle-valve—beak
slightly curved: area a little more than a quarter of the length of
the shell: deltadiwm rather raised, but longitudinally depressed along
the middle : deltidial ridges well defined: platform rather wide; oc-
cupying nearly the entire posterior half of the shell; biconvex ;
and most elevated at its anterior margin; its vaults of greater,
or lesser depth: median plate extending to a short distance forward
along the bottom of the shell: crescent of large size: archlet about
the width of the platform; and passing a little way in advance of
the median plate. Brachial valve—hinge with a slightly projecting
cardinal process: crescent with sides and ends of considerable width,
and well outlined: platform comparatively narrow, oecupying half

T. DAVIDSON AND W. KING ON THE TRIMERELLIDA, 145

of the valve; its vaults of greater or lesser depth: median plate
extending to within a short distance of the anterior edge of the
shell, dividing it into two lateral portions: archlet wider than the
platform, passing forward to the end of the median plate.

The discovery of this remarkable species is due to Mr. Billings,
who, in January 1862, published a short description, with figures, of
an internal cast. He also at the same time proposed for it a new
generic appellation, stating that it was allied to Obolus, but differed
from this genus in the possession of three longitudinal septa in each
valye*. Some good figures of the interior of the shell, taken from
gutta-percha squeezes, were subsequently published by Professor
Hall, in 1872.

Billings, when noticing the muscular scars of the present species, ~
states that “‘on each side” of the platform “ there appears to be a
small ovate muscular impression as in Obolus.” The specimen which
displays this impression was kindly placed in our hands by Mr.
Billings. At first we imagined he had correctly interpreted it; but
our subsequent investigations, as stated in another place, have con-
vinced us that the interpretation is not correct.

The impression on each side of the platform belongs to the crescent,
which, as already shown, has its parallel in the posterior half of both
valves of Lingula; therefore, instead of having been produced by
the postadductors, it is more likely to have been formed by the
implantation of the post- and post-lateral parietals of the splanch-
nocele and other organs characterizing the genus referred to.
These organs, as elsewhere made known, it is highly probable, pro-
duced the crown and sides of the crescents; but the ends of the
crescents (terminating in a line intersecting the anterior portion of
the platform) are formed by a scar, or rather scars, which there is
every reason to believe resulted from the posterior attachment of a
group of slanting or lateral muscles +.

The ends of the crescents escaped our notice when fig. 26,
Pl. XIII. of the present species was drawn: indeed it was only
after becoming acquainted with the parts corresponding with them
in Lingula that they disclosed themselves tous. On the right side of
the specimen, as represented by the figure just noticed, there is tole-
rably well seen, as in the annexed woodcut (fig. 2), the scar s, which,
from its relative position, we have no doubt is due, as in Lingula
anatina, to the posterior attachment of the lateral muscles.

The interior of the umbo in the pedicle-valve looks as if it were to
a slight extent doubly chambered; the chambers separated by a
wide cardinal buttress{. The hinge, on the top of the last part,

* We have shown elsewhere, there exists in reality but one septum—that is,
the median plate—the other two forming the sides of the platform: the name
Trimerella is therefore a misnomer.

t+ See paper on “ Lingula anatina,” Annals and Mag. of Nat. Hist. 4th
Series, vol. xii. July 1873; also woodcuts in p. 132.

{ Lwo figures of this species, representing apparently young individuals,
given by Professor Hall (Preliminary Notice, 1872, pl. xiii. figs. 14, 15), show
the umbonal chambers rather well developed, and separated by a narrow
cardinal buttress.

146 T, DAVIDSON AND W. KING ON THE TRIMERELLIDZ.

Fig. 2.—Brachial deine a Trimerella grandis.

7, sides, and s, ends of the crescent.

shows something like a hollow that may have been the seat of
attachment of the umbonal muscle.

Trimerella grandis, chiefly in the form of internal casts, appears
to be a very abundant fossil in the Guelph limestone—a rock imme-
diately overlying the Niagara limestone (Upper Silurian), at
Hespelar, Galt, New Hope, Elora, Guelph, also at Ontario, in
Canada West, or about four hundred miles from Montreal. In the
last locality some specimens have been found three inches and a
half in length by two in breadth and one in depth.

TRIMERELLA AcuMINATA, Billings. ° Pl. XV. figs. 4-7, and Pl. XVI.
figs. 1, 2.

Trimerella acuminata, Bill. Memoirs of the Geol. Survey of
Canada, Paleontology, p. 167, fig. 151, 21 Jan. 1862. (?) Lind-
strom, Ofversigt af Kongl. Vet. Akad. Forhandlingar, p.
253, pl. Boxe figs, 3, 4, 1867. Dall, American Journ. Conch.
vol. vil. part 2, p. 82, 1871. Bill. Am. Journ. Sci. Arts,
ord ser. p. 471, June 1871. Billings, Notes on Trimerella
acuminata, Annals and Mag. of Nat. Hist. 4th ser. vol. viii.
p- 140, August 1871. Dav. and King, Report of Brighton
Meeting of Br. Assoc. 20 August 1872.

Shell large; massive; longitudinally oval; broadly rounded in
front; tapering in its posterior half: swface smooth, or strongly
marked at intervals by concentric lines of growth. Pedicle-valve mode-
rately convex: beak produced ; somewhat rounded at its extremity :
deltxdium triangular ; very wide; deeply concave ; strongly marked
with anticlinal lines: deltidial ridges well pronounced: areal borders
narrow: deltidial slope narrow; marked from side to side by fine

T. DAVIDSON AND W. KING ON THE TRIMERELLIDA. 147

lines: hinge depressed, or concave in the middle: cardinal facet
large ; passing down with a considerable inclination to the cardinal
buttress, from which it is separated by a well-defined edge: plat-
form extending to about half the length of the valve; strongly
depressed along its surface ; with perpendicular lateral walls: vaults
extending to a little way underneath the central part of the hinge:
cardinal buttress low; standing approximately at a right angle to
the plane of the shell; as wide as the platform: wmbonal chambers
extending to some distance under the hinge outside of the deltidium.
Brachial valve deeper and more convex than the opposite one: hinge
raised in the middle; with a strong lamelliform cardinal process :
platform well developed; a little longer than that of the opposite
valve ; its vaults of considerable depth: median plate much elongated.

Mr. Billings, to whom paleontolgists are indebted for the dis-
covery of this species, and who was the first to give a brief descrip-
tion. of it, accompanied by a figure of a fragment of the internal cast
of the ventral valve, informs us that the largest specimen he has
seen is three inches and a half in length, by three inches in width.
He also states “‘ there is sufficient to show that this species is quite
distinct from 7. grandis. Ifa section were made across the beak
of a perfect shell of 7. acuminata, it would show four perforations
arranged in a curve, exactly as in the similar section of the Swedish
species figured by Dr. Lindstrom. But if the beak of 7. grandis
were cut across, 1t would show only two orifices, and they would
be the homologues of the two lateral perforations [our wmbonal
chambers| in the section of 7. acwminata, because in 7’. grandis the
two central tubes [our platform-vaulis| do not extend into the
beak, but terminate before they reach it.”

According to our mode of describing its interior, the species
under description shows six ‘perforations ”—four in the pedicle-,
and two in the brachial valve. Itis extremely difficult to give all
the “ perforations” in one view: our figure 4%, Pl. XV. only shows
those (vaults) belonging to the platforms: there are, however, in the
pedicle-valve of the specimen represented other two (umbonal
chambers), one on each side of the umbo.

The deltidium is usually wide and concave, giving the area a
singular appearance. ‘The hinge is remarkable in having its
middle in the pedicle-valve concave, and the corresponding part in
the opposite valve convex: the convexity of the latter part is in-
creased by the lamelliform cardinal prominence (Pl. XV. fig. 4"). The
cardinal facet exhibits the lozenge faintly, though with close inspec-
tion its bordering lines may be made out. There are two small
““umbo-lateral scars,” one at the entrance, and on the outer side,
of each umbonal chamber (fig. 1, x, Pl. XII.). The scars on
the platform and crescent, though far from satisfactorily defined
(Pl. XV. fig. 4"), appear to be formed and arranged as represented in
PE ene i520

T. acuminata occurs chiefly in the state of casts *, in the Guelph

* We have succeeded in making some very good gutta-percha moulds from
some of these casts.

148 T. DAVIDSON AND W. KING ON THE TRIMERELLID®.

limestone immediately overlying the Niagara limestone (a rock
equal in age to the Aymestry limestone of England, or about that
level, according to Mr. Billings) at Hespelar, Galt, Elora—villages
situated near each other in Ontario (Canada West). Mr. Weston, of
the Geological Survey of Canada, has discovered several new specimens
in the localities, already mentioned in our description, which have
yielded 7. grandis: among these are two exhibiting casts of both
valves in connexion. Mr. Weston discovered also at Hespelar an
incomplete specimen, measuring three and a half inches in length,
which showed that the shell was smooth, or marked only by con-
centric lines of growth. Well-characterized internal casts of 7’.
acuminata were likewise found by Dr. Lindstrom in several parts of
the Island of Gotland, as far back as 1859, in rocks about equivalent
in age to those of Canada.

The species appears to have had a very extended horizontal
range; for it occurs in the same horizon in Canada, the United
States. and in Sweden.

TRIMERELLA Linpstremt, Dall. Pl. XIV. figs. 1-7.

Trimerella? Lindstrom, Ofversigt af Kongl. Vetenskaps-Aka-
demiens Forhandlingar, p. 253, pl. xxi. figs. 1 & 5 to 9, 1867.
Gotlandia Lindstromt, Dall, American Journal of Conch. vol. 11.
part 2, p. 160, 1870. 2? Trimerella, Dall, vol. vii. part 2, p. 84,

.1871. Dav. and King, Report of Brighton Meeting of Br.
Assoc. 20 August, 1872.

Shell very massive; longitudinally oval; anterior half broadly
rounded ; posterior half tapering rather rapidly to the extremity of
the beak, which is pointed ; greatest breadth at about one third of
the length of the shell from the anterior margin: surface smooth
or marked with concentric lines of growth. Pedicle-valve very
thick; rather less deep than the opposite one; sloping from the
point of the umbo to the anterior margin: area equilateral ; trian-
gular; rather longer than one fifth of the length of the shell: areal
borders rather wide: deltidiwm flat, or raised; wide; triangular; with
amore or less marked longitudinal depression along the middle:
deltidial ridges and callosities prominent; crossed by strongly in-
dented lines or ridges: deltidial slope variable ; occasionally its lines
appear to be coarse: cardinal facet large ; variable in its inclination:
umbonal chambers resembling tubular perforations : platform and
vaults nearly as in 7. grandis. Brachial valve massive ; moderately
and uniformly convex, especially about the umbo: hinge generally
with a large, massive, projecting cardinal process in the centre ;
and two depressions, one on either side: platform and vaults agreeing
pretty nearly with those in 7. grandis.

Trimerella Inndstroemi varies remarkably in its cardinal features ;
the variations in some cases being highly puzzling. ‘The cardinal
process is rude and inconstant in shape: the variety given in
Pl. XIV. fig. 3 (the part answering to the cardinal scar v in fig. 2,
Pl. XII.) may be regarded as showing its median state of deve-
lopment—some specimens exhibiting it lower, others even more

T, DAVIDSON AND W. KING ON THE TRIMERELLID®. 149

elevated. In the latter state it appears to have fitted like a tooth
into a deep excavation in the hinge or near the origin of the plat-
form (Pl. XIV. fig. 5), thereby apparently effacing both the cardinal
facet and buttress: this excavation is bounded posteriorly by a
perpendicular wall. Whether the facet was situated on the wall is
a point on which it is difficult to offer any more than a suspicion.
A similar difficulty attaches to the excavation ; was it produced by
the attachment of the umbonal muscle? A variation in the opposite
extreme is represented in fig. 2, Pl. XIV., which shows the facet, with
the lozenge, occupying a place seemingly corresponding to the bottom
of the excavation seen in fig. 5. But certain appearances, presented
to us, favour the suspicion that the facet and lozenge were situated
on the posterior wall, rather than at the bottom of the excavation.
In most of the Trimerellids the facet dips forward from the delti-
dium,—suddenly, we should say, in the present species, if represented
only by such individuals as that in fig.5; but other specimens
exhibit it with a moderate inclination, as in 7’. acuminata (Pl. XV.
fig. 4a); while in the one represented in fig. 2, Pl. XIV., it actually
inclines backwards or in the opposite direction. The umbonal
muscle may or may not have been attached to the bottom of the
excavation in fig. 5 specimen; but it is difficult to point out its
locality of attachment in fig. 2 specimen: possibly it was attached at
the origin of the platform, where occasionally there is something like
a scar.

The dentary system, taking it to be represented by the deltidial
callosities in the pedicle-valve, and the cardinal sockets in the oppo-
site one, is equally variable; though, possibly, some of the appear-
ances of variation may be due to erosion. ‘The callosities are strong
bosses in certain specimens, and thick plates in others ; it is therefore
possible that the appearances which led Lindstrém to believe in the
groove-and-ridge mode of articulation were presented by an extreme
case of lamelliformity *.

The umbonal chambers are of small calibre, being little more than
tubular perforations. Lindstrém has given a figure? showing their
presence in an early stage of growth, also that their tubular form
was persistent. Such a case strongly favours the idea that the
chambers were for some particular purpose; otherwise it is difficult
to understand why they have not been filled up with organic accre-
tions. The umbonal cavity of the brachial valve in the specimen
represented in our fig. 7, Pl. XIV. is doubly chambered; the large size
of the conical prominences (casts), on each side, shows that the
chambers have considerable depth and width.

The platforms do not appear to be so large as in the last species :
they resemble more closely those of 7’. Billings, particularly in
haying short vaults. ‘The scars are represented with approximate
correctness in fig. 2, Pl. XIV. The submarginal impressions, generally
indistinct, are resolvable into the typical crescent.

* The specimen in question, which we have not seen, is preserved in the Mu-
seum of Stockholm.
+ Geol. Mag. vol. vy. pl. xx. fig. 9.

150 T, DAVIDSON AND W. KING ON THE TRIMERELLID®.

Specimens of this remarkable fossil were first discovered by Dr.
Lindstrom in 1859, and soon afterwards forwarded for examination
to one of the writers of the present memoir, accompanied by a MS.
description and figures; but it was only after the publication by
Billings of his genus Z’rimerella in 1862 that certain other characters
of the shell were determined by Dr. Lindstrom. His material being
more complete than that upon which the genus was founded, he
published, in 1867, a description and figures of the species, but without
giving ita name. He was, however, mistaken in believing that the
vaulted platform is smooth, and had no impressions of muscular scars
upon it, as these occurred to us very plainly on one of his specimens,
thus confirming the idea entertained by Billings in 1862.

In 1870 Mr. Dall, from not quite understanding the characters of
Mr. Billings’s genus T'rimerella, and those of Dr. Lindstrém’s species,
proposed for this last a new generic appellation, Gotlandia; but in
1871 he corrected the mistake. In the same year he described the
species under the designation of ? Trimerella Lindstrom.

Some specimens have been found three inches in length by two
and a half in breadth and one in depth.

We are informed by Dr. Lindstrom that this species may be re-
garded as one of the most characteristic fossils of the Silurian rocks
of the Island of Gotland. Almost everywhere may be discerned
fragments of it on the weathered surface of the limestone—often
transverse sections of both valves; so that the hard crystalline
rock seems to be almost entirely made up of them—to such an
extent that it might be termed the “ Trimerella limestone.” ‘There
are, nevertheless, a few localities where tolerably good speci-
- mens may be collected, such as at Klinteberg, Fole, lamsugnen, a
limestone quarry in the parish of Olham, at Furillen on the east
steep coast of Gotland, Widlansudd, situated quite at the extremity
of the island, and in Faro on the north coast of Gotland.

It is, however, only within the range of the “‘ Central Gotland ”
beds of the islands of Gotland and Faro that 7. Lindstreemi (with
T. acuminata) has been found. These beds have been considered
by Dr. Lindstr6m and by Sir R. Murchison equivalent to the
Aymestry rocks of England, a view fully supported by fossil evidence.
We are indebted to Dr. Lindstrom, to Prof. E. Walmstedt of the
University of Upsala, and to Herr Fegrceus for many specimens of
this species.

TRIMERELLA Birziinesi, Dall. Pl. XVI. figs. 8 & 9.

Trimerella Billingsit, Dall, Am. Journal of Conch. vol. vii. part 2,
p. 82, pls. 1, 2, 1871*.

Shell longitudinally oval; compressed: wmbonal region projecting;
twisted to one side. Pedicle-valve—deltidium large ; wide; trans-
versely striated; slightly excavated ; strongly bordered at sides by the

* The fossil referred to 7. Billinsgii in this paper was so labelled by Mr.
Dall; but he must have mistaken some other specimens for this species, as the
longitudinal section given in his fig. 3 appears to have been taken from a speci-
men of Trimerella grandis.

T. DAVIDSON AND W. KING ON THE TRIMERELLID&®. 151

deltidial ridges, and in front by a tolerably well-defined deltcdial slope.
The cardinal facet appears to be small. The platform is solid, long,
and large, but of slight elevation: it has a raised \\-shaped anterior
margin the side divisions answering to the openings of rudimentary
vaults. The median plate scarcely exists, and extends to no more
than a line in advance of the anterior termination of the platform.
Scars occupy the anterior half of the platform: some strongly marked
are situated near its front border. The crescents are faintly indicated.

This remarkably compressed species is one of those aberrant
forms which render it difficult to diagnose a generic group, or to
determine what genera they should be placed in. It has much the
appearance of Tirimerella grandis; and therefore seems naturally to fall
into the same genus. Another species, next to be described, stands out
prominently as a still more aberrant form. 7. Ballingsi does not
possess any thing like a dentary system: its pedicle-valve has no
* median plate ; but the centre of the anterior margin of the platform
is somewhat prolonged. Besides the scars on the anterior half of
the platform, there are linear impressions (answering to the trans-
verse scars, t in fig. 1, and ¢ in fig. 2 of Pl. XII.) passing in
front of the rudimentary vaults. The platform and the openings
of its vaults have much the appearance of a flattened nose, with
the nostrils and their rim prominently exposed.

All we have seen of this species is the internal cast, measuring
two inches three lines in length by one inch five lines in width, of
a single pedicle-valve, found by Mr. Billings in the Guelph Limestone
at New Hope, West Canada.

TRIMERELLA (?) GALTensis, Billings. Pl. XVIII. fig. 13; and
Pl. XIX. figs. 4 & 4°.

Obolus galtensis, Billings, Geol. Survey of Canada, Pal. p. 168,
fig. 151, 21 Jan. 1862. Trimerella minor, Dall, Am. Journ. of
Conch. vol. vil. p. 83, 1871. Rhynobolus galtensis, Hall, Re-
port on the State Cab. of Nat. Hist. preparations of Pal. New
York, p. 5, March 1871. Obolellina, Bill. Can. Nat. vol. vi.
p. 222, 29 Dec. 1871; also April 1872. Dinobolus, Dav. and
King, Report of Brighton Meeting of Br. Assoc. 20 August, 1872.

Shell compressed or moderately convex; tapering rather sharply
behind; broadest, and rounded anteriorly. Pedicle-valve flatly convex:
area large; long ; nearly an equilateral triangle: deltidium raised ;
occupying a considerable portion of the area; crossed by strong
imbricated lines ; with a broad deepish furrow along its entire length :
areal borders transversely striated: platform wide; very slightly
raised ; its anterior very obtusely V-shaped. Brachial valve slightly
convex: platform V-shaped; scarcely raised, or an excavation ;
deeply marked with scars, which give it a tripartite appearance:
crescent well pronounced, especially at the crown, the middle of
which is pointed, the point directed backward.

Under the name of Obolus galtensis this shell was very briefly
described in 1862 by Mr. Billings, who at the same time gave an
incomplete figure of an internal cast of the dorsal valve. In 1871

la? T. DAVIDSON AND W. KING ON THE TRIMERELLIDZ.

Mr. Dall also published a very brief description of the ventral valve
under the designation of Trimerella minor, n. sp.?* Later, in
March 1872, Prof. Hall noticed this shell, calling it Rhynobolus ,
and in December of the same year he gave a fuller account of it.

Only slight traces of the crescent have been observed in the
pedicle-valve. The hinge and beak bear considerable resemblance
to those of 7’. Billings. We suspect that Dall, Hall, and others
view what we call the deltidium in a different light from ourselves—
they taking the median furrow to represent this part; we, on the
contrary, look upon the whole as forming the deltidium. ‘There are
only slight traces of the deltidial ridges: Hall, however, has repre-
sented one of them prominently on the left side of the area. The scars
on the platform of the brachial valve form three sets :—one on the
ante-median portion, which is circular or slightly lozenge-shaped ;
the second or median set immediately behind (partly represented in
fig. 4*, Plate XIX.); the third with its two members separated by ©
the last set, and diverging backwards.

Trimerella galtensis has been placed in different genera. Billings at
first considered it to belong to Obolus, but afterwards included it in his
genus Obolellina: Dall placed it in Trimerella: Hall constructed his
genus Fhynobolus for it: and the present writers, on a former occasion,
allocated itin Dinobolus! It certainly differs in some important fea-
tures from the typical forms of 7rimerella, having partly sunk plat-
forms and no vaults ; but so does 7’. Billingsii, though not to the same
extent. In the last species, as already noticed, the platforms are
considerably reduced (at least the one belonging to the pedicle-valve),
nearly to the exclusion of the vaults.

The absence of vaults in Monomerella, it may be readily seen,
causes the platform in the pedicle-valve to be considerably modified ;
and evidently their absence in the present species has given the
platform of the same valve its peculiar character. We are quite
ready to admit that in its apophysary system 7’. galtensis offers a
close approach to Monomerella; also, looking at the scars on the
platform of the brachial valve, that it very closely agrees in this
respect with Dinobolus: but, with the exception of these scars, we
see very little resemblance between this and any of the known
species of the latter genus. Were it not that its umbonal characters
are characteristic of 7rimerella, we might have been induced to give
the present species a different generic allocation : our views in this
matter are generally influenced by the peculiar characters of
Trimerella Billingsii: besides, excepting the form of the scars on the
platform, we see very little resemblance between 1. galtensis, and
any of the known species of Dinobolus. Another most important
difference prevails: in the shell under description the crescent shows
no appearance of the composite structure (the strong indentations)
which characterizes the inner border of the sides of the same part

* Both Mr. Billings and the authors of this paper, who have seen the speci-
men upon which it was established, are able to assert that Mr. Dall’s Trime-
rella minor was founded on one of the typical forms upon which Mr. Billings
had originally instituted his Obolus galtensis.

T. DAVIDSON AND W. KING ON THE TRIMERELLID®. 153

in the latter genus. The present species appears to us to be an ex-
treme aberrant form linked to Trimerella by meaus of 7’. Billingsi.

Some well-preserved internal casts of both valves have been pro-
cured by Mr. Billings and Prof. Hall from a light-yellow rock of
the Guelph Limestone (Upper Silurian) at Galt, Ontario, in Canada.
The largest examples, we have seen, measured—length one inch eight
lines, breadth one inch one line.

TRIMERELLA OHIOENSIS, Meek. PI. XVI. figs. 3-7; Pl. XIX. figs. 1, 2.

Trimerella ohioensis, Meek, Sill. Am. Journ. Sci. & Art, p. 305,
April 1871. Dall, Am. Journ. of Conch. vol. vii. part 2, p. 83,
pl. i. figs. 3, 4,1871. Dav. & King, Report of Brighton Meet-
ing of Br. Assoc. 20 August, 1872.

Shell large; massive ; globose, or longitudinally oval; anteriorly
broadly rounded; greatest breadth towards the middle: beak
moderately produced ; occasionally much incurved : eaterior smooth.
Pedicle-valve moderately convex, or flattened : deltidium large ; wide ;
triangular ; deeply concave: platform commencing almost imme-
diately under the hinge, and extending to about two thirds of the
length of the valve: vaults very large and deep: median-plate
tolerably long. Brachial valve with an incurved umbo: cardinal
process variable; well developed: platform commencing under the
hinge ; deeply vaulted: median-plate considerably elevated ; extend-
ing to within a very short distance of the margin of the valve.

This, which may be considered a typical form, is more tumid
than any of the preceding species. Mr. Meek states that on com-
paring specimens of 7’. ohwoensis with 7. grandis he found them
quite different and distinct—that proportionally 7. ohioensis is a
much broader shell, and differs in having the beak of the dorsal valve
strongly incurved beyond the plane of the connecting edges of the two
valves—that it is also more convex, its ventral valve more arched, and
the rostral cavity of its dorsal valve deeper. The platforms and
vaults, in their length, resemble those of 7. acumimata. We
have not become acquainted with its platform-scars and crescents ; but
certain obscure indications of the former are observable close to the
almost perpendicular hinge-wall and below the cardinal process :
they seem to have been produced by the attachment of a valvular
muscle, possibly the umbonal. The hinge, much elevated in the
middle of the brachial valve, presents different appearances through
modifications in the mode of the attachment of the muscles belong-
ing to the cardinal scar, v; which in certain individuals appears as
a ridge running along the hinge (figs. 5 & 6, Pl. XVI.); while in others
it is double cup-shaped (fig. 7, Pl. XVI.). Meek mentions that “the
centre of the hinge appears to have been furnished with a massive
gently rounded cardinal process.”

Through the liberality of Mr. Meek, Mr. Joseph Henry, Prof.
Hall, and Mr. Whitfield, we have been able to examine a number of
well-preserved internal casts. These vary considerably ; for while
some examples are almost spherical, or as wide as long, others are a
good deal longer than wide. Mr. Meck assures us that some speci-

154 T, DAVIDSON AND W. KING ON THE TRIMERELLIDA,

mens are nearly four inches in length, two and a half in breadth,
and one inch and a half in depth. The breadth is also greater
in some specimens, as we possess an internal cast in which it
measures three inches.

T’, ohioensis abounds in the Niagara Limestone (Upper Silurian) at
Genoa, Ottawa Co., Ohio; Sinking Springs, Ohio (I. 8. Newberry).
According to Mr. Dall one valve was found in the Guelph Limestone
of Canada.

TrimeRELLA Datut, Dav. & King. Pl. XV. figs. 1-3.

Trimerella Dalli, Dav. & King, Report of Brighton Meeting of
Brit. Assoc. 20th August, 1872.

Shell ovate, or longitudinally oval: surface smooth. Pedvcle-valve
flatly convex, especially along the middle: beak prominent ; nearly
straight ; about one fifth the length of the shell; with two wmbonal
chambers rather deep, conical, separated by a strong cardinal but-
tress: crescent tolerably well pronounced: platform divided at its
posterior end by an extension of the buttress into two lateral por-
tions ; wide and short: vaults not of much depth: median plate of
small extent. Brachial valve uniformly convex: cardinal process
an excavated projection: plaiform long; extending to the middle
half of the valve: vaults rather short: median plate much elon-
gated.

Several internal casts of the present fossil were sent us for exami-
nation by Mr. Billings. After taking gutta-percha squeezes from
them we arrived at the conclusion that they are distinguishable from
all other species of the genus chiefly by the presence of deep um-

‘ ponal chambers separated by a lamelliform cardinal buttress. These
characters approximate the species to the genus Vonomerella.

The largest example that has come under our notice measured
1 inch 4 lines in length, 1 inch 1 line in breadth, and 8 lines in
depth.

T. Dall occurs in the Guelph Limestone at Hespelar, Canada West;
where it was found by Mr. Weston.

TRIMERELLA WISBYENSIS, Day. & King. Pl. XIII. fig. 1.

Trimerella wisbyensis, Dav. & King, Report of Brighton Meeting
of Brit. Assoc. 20th August, 1872.

Brachial valve nearly circular, or a little wider than long: platform
moderately long; depressed along its middle: vaults extending
a the platform to apparently half its length: median plate

ong *.

Of this species all we know is the internal cast of a single dorsal
valve, 13 lines in length by 11 in breadth, found by Dr. Lindstrom
in the Upper Silurian (Wenlock) near Wisby, in Gotland.

* Traces of the crescent are visible on the hinge. Although our figure repre-
sents the platform as if complete, we suspect that it was larger than represented.
Its surface shows median and lateral scars.

T. DAVIDSON AND W. KING ON THE TRIMERELLIDA. 155

IX. Monomerrxya AND ITs SprcrEs.

Genus Monomeretta *, Billings, 1871.

Shell thick ; circular, or transversely oval, in its marginal outline.
Pedicle-valve—umbo usually large ; projecting ; double-chambered :
area of considerable size: deltidium large: hinge generally thick and
elevated ; ledge-shaped ; depressed or concave in the middle portion :
cardinal facet a wall-like space rising out of or behind the ledge or
flat of the hinge: cardinal buttress strong; lamelliform: platform
flat ; slightly elevated ; widest, highest, and very obtusely angulated
in front. Brachial value—umbo rounded: hinge moderately thick :
platform trilobed; usually with a thin margin.

Type species, Monomerella prisca, Billings.

The unvaulted character of its platforms, large umbonal chambers,
and certain other characters differentiate this genus remarkably from
Trimerella. The platform of the pedicle-valve, which widens out
from behind, has its front formed by two edges meeting at a very
obtuse angle. The platform of the brachial valve is widest behind,
and conspicuously trilobed, the latter character deeply incurying its
sides anteriorly: its surface is divided, answering to the attachment
of three sets or pairs of muscles—a lateral pair, with a member be-
longing to each of the lateral lobes, a median pair belonging to the
space between the lateral lobes, and an anterior pair belonging to
the anterior lobe.

The cardinal facet reminds one of the perpendicular wall behind
the deep excavation that characterizes the hinge in certain varieties
of Trimerella Lindstremi (Pl. XIV. fig. 5)—a peculiarity strongly
favouring the idea that in such varieties the cardinal facet was situated
similarly to what it is in the present genus, particularly as traces of
the lozenge have occurred to us on the “ wall-like space” belonging
to the hinge of its typical species. The dentary system of Mono-
merella does not form a prominent feature: we only find evidence

* Mr. Billings, in a letter bearing the date 5th May, 1871, notified to me
the discovery of a species (which he forwarded at the same time for exa-
mination, see Pl. XVII. fig. 5, the type of Monomerelia) that appeared to
him to “look like a cross between Odolus and a Trimerella,’ and added
“ What shall I call it?” I at once perceived that the species was referable
to the same genus as another fossil (herein named M. Walmstedti), occur-
ring in Sweden; a figure of the pedicle-valve of which had likewise been
communicated to me by Dr. Lindstrém, on the 18th of December, 1869,
with the intimation that it belonged to another genus, more akin to Trime-
rella and Obolus than the others (alluding to some other genera spoken
of in the letter). Inadvertently overlooking for the moment the fact last
noticed (for which I feel some apology is due to both Dr. Lindstrom and the
public), and as both the Swedish and the American fossils required a new generic
name, I left it to Mr. Billings to impose one; which he did in the ‘Cana-
dian Naturalist,’ published December 1871. It has been considered necessary to
mention these circumstances, that Dr. Lindstrém’s discovery and opinion might
be adequately recognized.

The figure above alluded to represented the valve we have figured in Pl. XVII.
fig.2. Dr. Lindstrom had also sent me a figure of the brachial valve (the one re-
preserited in Pl. XVIII. fig. 3), ina letter dated so far back as the 18th January,

1860.—T. D.

3 wens

156 T, DAVIDSON AND W. KING ON THE TRIMERELLID A.

of it, apparently sockets, in the brachial valve of M. Walmstedtr.
The cardinal process, slightly developed, is only known to us in this
species. We entertain considerable doubt as to the transverse lines
occurring apparently on the hinge of M. prisca being the repre-
sentatives of this part : their position behind the crown of the crescent
strongly favours our scepticism. ‘The genus includes the following
species.

MonomeERELLA Prisca, Billings. Pl. XVII. figs. 5-8.

Monomerella prisca, Bill. Can. Nat. vol. vi. p. 221, 29 Dec. 1871.
Dav. & King, Remarks on the genera Trimerella &e., Report
of Brighton Meeting of Brit. Assoc. 20th August 1872; Annals
& Mag. Nat. Hist. Oct. 1872, and Geol. Mag. Oct. 1872.

Pedicle-valve obscurely triangular; with rounded angles; longer
than wide; greatest breadth at about two thirds of its length from
the beak; tapering posteriorly ; very much flattened; convex only
at the sides: wmbo almost straight; about one third of the length
of the valve: area broadly triangular: deltidial ridges moderately
wide: areal borders narrow ; raised; strongly marked with trans-
verse lines: deltidial slope and cardinal facet well developed : hinge,
in its central portion, much depressed: wmbonal chambers wide-
mouthed: cardinal buttress strong ; prolonged to some distance along
the bottom of the valve: platform moderately large; obtusely an-
gular in front; with one of its sets of scars striated longitudinally,
and strongly suleated obliquely. Brachial valve almost circular ;
nearly straight in front; evenly convex; and deeper than the oppo-
site one: hinge slightly raised or convex in its middle portion;
transversely striated on what appears to be its outer or exposed por-
tion : crescent in parts strongly marked: platform large; V-shaped ;
round and projecting at its antemedian point; very slightly elevated
above the surface of the valve; its sears tolerably well shown.

This species is strongly marked by characteristic features, espe-
cially its wide deltidium. Its various muscular scars are generally
well displayed. In the brachial valve three pairs (anterior, median,
and lateral) are well seen on the platform: the anterior pair is de-
pressed below the median pair. The subcardinal scar, w, is rather
conspicuous in the umbonal cavity. The hinge appears to be un-
usually slight in depth, showing nothing more than the crown of
the crescent, which forms a deepish furrow. On the outside of the
hinge there is a strongly striated surface; the strie parallel and
running transversely: probably this feature represents the epidermal
lines on the area of the corresponding valve of Lingula*. The an-
terior portion of the crescent, apparently its ends, widens out a little,
and presents a compound appearance. A pair of umbo-lateral scars,
«, lies within the crescent, a member on each side, at its departure
from the hinge.

The pedicle-valve also shows its scars for the most part rather
obviously: two pairs, one behind the other, occur on the platform ;

* Our figure does not represent the striz so fully as could be wished ; they
are in reality more numerous, forming a deeper band, and run out straighter.

fT. DAVIDSON AND W. KING ON THE TRIMERELLIDZX. 157

the hindermost or posterior lying close against the cardinal buttress ;
the anterior a strongly striated pair. There are also faint indica-
tions of a third pair situated at the ante-median point of the platform.
A pair of small umbo-lateral scars, x, occurs on the outer side, a mem-
ber in each, of the umbonal cavities, corresponding to a pair similarly
situated in Trimerelia. The deeply excavated hinge gives a promi-
nent appearance to the cardinal facet: faint but determinable traces
of the berdering lines of the lozenge occur on the last part.

The largest example that has come under our notice measures
2 inches in length by 1 inch and 10 lines in breadth.

Internal casts of this species, discovered by Mr. T. C. Weston, of
the Geological Survey of Canada, do not appear to be rare in the
Guelph Limestone (Upper Silurian) at Hespelar, Ontario, in Canada.
Specimens showing the exterior of the shell have not yet been dis- —
eovered; but it was in all probability smooth, or nearly so, as ap-
pears to be the case with all the species of Trimerellids hitherto
discovered.

Monomeretta Watmstepti, Day. & King. Pl. XVII. figs. 1-4, and
Pl, XIX, fig. 3, 3°?

Monomerella Walmstedti, Dav. & King, Remarks on the Genera
Trimerella, Dinobolus, and Monomerella, Brighton Meeting of
Brit. Assoc., 20th August 1872; Annals & Mag. Nat. Hist.
Ath ser. Oct. 1872; and Geol. Mag. vol. ix. Oct. 1872.

Shell somewhat triangular; with rounded corners; very thick ;
broadest anteriorly; tapering posteriorly; front flatly rounded.
Valves almost equally deep; most convex in the umbonal region:
surface smooth or marked by fine concentric lines of growth.
Pedicle-valve flattened anteriorly: beak large; massive; slightly
incurved: area wide; equilatero-triangular: deléidiwm broad; flat;
strongly laminated: deltidial slope narrow, but well defined: del-
tidial ridges wide; flattened: hinge deeply depressed or excavated
in its middle; projecting to a point in its centre; and backed by a
nearly perpendicular wall: cardinal facet situated on the back-wall
of the hinge: umbonal chambers deep; and wide-mouthed: cardinal
buttress stout; extending to some little distance along the bottom of
the valve: platform obtusely angular in front ; more or less excavated
laterally by three pairs of scars. Brachial valve evenly convex in
its posterior half: wmbo much incurved: Ainge transversely ridged,
and raised in the middle portion ; somewhat excavated at the sides :
platform with a V-shaped anterior outline; sharp, and raised at the
edges ; excavated on its surface by three sets of scars.

The proportions of this species are variable: one example, appa-
rently of ordinary size, measured | inch 8 lines in length by 1 inch
5 lines in breadth and 10 lines in depth.

Monomerella Walmstedti, although agreeing very closely with the
last species, is characterized by some well-marked differences. The
umbo of the pedicle-valve is more elevated ; while that of the opposite
valve is more rounded or incurved. The hinge of the brachial valve
is thicker; has its middle portion more elevated, becoming, it would

Qos, Gas. Nos lis: N

158 T. DAVIDSON AND W. KING ON THE TRIMERELLID&.

appear, a rounded cardinal prominence; and it is excavated on each
side, or where the dentary sockets occur in Trimerella. Instead of
the outside of the hinge being transversely striated, it shows the
centre of the umbo excavated.

The hinge-characters of the pedicle-valve resemble those of MW.
prisca. The cardinal facet is similarly situated, equally large, and
shows traces of the lozenge, as well as the crown of the crescent.
The platform (except in being more elevated) and the umbonal
chambers are also in close agreement with those in the latter spe-
cies. The crescent is not conspicuously marked in either valve.

Very perfect specimens, in separate valves, of this remarkable
species were discovered, as far back as 1867, by Dr. Lindstrom, at
Klinteberg, Central Gotland, in a rock considered to belong to the
Aymestry formation. These specimens are preserved in the Museum
of Wisby ; and other examples may be seen in that of Stockholm.
Some internal casts of the same species were found by Herr Fegroeus
in the same formation in the island of Faro, north of the mainland
of Gotland. We are also under obligations to Prof. F. Schmidt, of
St. Petersburg, for kindly enabling us to add copies of sketches of
casts, probably representing the species under description, which
were found by Count Keyserling at Kerkaw, in Livonia, Russia, in
rocks considered to be the equivalent of the Wenlock Group in
England: the original sketches were found among the papers left
by the late Dr. Pander.

Monomeretia Linpstremti, Dav. & King. Pl. XVII. figs. 9,9 a.

Pedicle-valve transversely elliptical: wmnbo moderately produced :
umbonal cavities shallow, and wide-mouthed : cardinal buttress rather

stout ; attenuated; considerably elongated in advance of the hinge :

platform long, and wide; the elongated portion of the buttress passing
nearly to its front edge; striated longitudinally ; somewhat rugose
transversely ; its front edge obtusely angulated. Brachial valve trans-
versely elliptical in outline; flatly rounded in the middle: platform
with anterior lobe a little more elevated than the lateral lobes.

We are only acquainted with this shell from a single cast showing
both valves in conjunction, and measuring 11 lines in length by 17
in breadth and 5 in depth. We therefore can only describe it im-
perfectly : but the character and shape of the cast leave no doubt of
its belonging to a well-marked species—and different from either of
the preceding in being much wider than long, and more compressed.

Monomerella Lindstremi was procured by its discoverer (after
whom it has been named) from a calcareous bed of the age of the
Wenlock formation in the neighbourhood of Wisby, in Gotland.

MonoMERELLA ORBICULARIS, Billings. Pi xevaie hee, ILO),

Monomerella orbicularis, Bill. The Canadian Naturalist, vol. vi.
p- 221, 29th Dec. 1871; Dav. & King, Report of Brit. Assoe.
Meeting, Brighton, 20th August 1872.

“ Broadly ovate, nearly circular, lenticular, both valves moderately
convex. ‘The casts seem to show that a thin plate extends forwards

T. DAVIDSON AND W. KING ON THE TRIMERELLIDA. 159

a short distance from the cardinal edge, supported by the septum.
The lergth and width appear to be about twelve or fifteen lines”
(Billings).

Only two internal casts ef this species appear to have been found.
The platforms are of the usual form: that in the brachial valve is
excavated ; and at its posterior end is situated the subcardinal scar,
w, which is well pronounced. The crescent in the same valve has
its crown rather deeply impressed on the hinge; and its sides are
expanded, though not shown in the drawing (fig. 10).

We are indebted to the kindness of Mr. Billings for the opportu-
nity of examining the original and typical specimens. It occurs in
the same formation and locality with MW. prisca.

X. DrInoBoLus AND ITs SPECIES.

Dinozotus*, Hall, 1871.

Obolus, Davidson, 1853; Lindstrom, 1867; Hall +, 1868.
Dinobolus, Hall, “March 1871.” £

Ehynobolus (in part), Hall, 1871.

Ungulites, Quenstedt, 1871.

Obolellina, Billings, Dec. 29th, 1871 §, &c.

Shell circular, or generally wider than long. Valves moderately
thick. Pedicle-valve with a slightly prominent evenly formed
umbo: area wider than long: platform more or less sinuated ;
widely V-shaped; and slightly raised in front : crescent prominently
marked in its crown and sides: hinge moderately thick; with a
rounded edge, on which and in front of the cardinal facet is a pair

* The synonymy of the present genus, and the quotations added, have in-
fluenced us in the adoption of the name Dinobolus.

+ ‘Some years since, having these fossils under consideration, I proposed a
new generic name for them; but sending drawings to Mr. Davidson, he gave me
the opinion of himself and Mr. Woodward that they belonged to the genus
Obolus. There are certain points of difference, however, which I have been in-
clined to believe are of generic importance ; and I am not entirely satisfied in
referring them to that genus, as illustrated, though closely allied to it” (Twen-
tieth Report of the State Cabinet, p. 368). (It is only fair I should admit that
Professor Hall is correct in his more recent statement [see ‘American Journal
of Science and Arts’ for August 1872, p. 20] that, in a letter of date 31st October,
1862, he proposed to me the generic name Conradia for this fossil.—T. D.)

‘Notes on some new and imperfectly known forms among the Brachiopoda.
«This pamphlet was published in March 1871, and a number between twenty-five
and thirty copies delivered to meat the time. The type was left standing in order
to print a large number to be accompanied by a plate of figures then in progress,
with descriptions of the same. Of these copies the greater part were distributed
in the United States soon after publication. Copies were sent to the Geological
Society of London, to Mr. Davidson, M. Barrande, Dr. Lindstrom, Dr. Geinitz,
and others. The pamphlet is noticed in the ‘Jahrbuch’ for 1871, p. 989. On
the 7th of April, 1871, the printing establishment of Weed, Parsons, and Co.
was destroyed by fire, together with the Twenty-third Report of the State
Museum (printed to nearly 200 pages), the lithograph-stones, and every thing
else pertaining to the work.”— Op. cit.

§ Canadian Naturalist, vol. vi. p. 222, 1871; and p- 3380, April 1872; also
American Journal of Science, April and May 1872.

nN 2

160 T. DAVIDSON AND W. KING ON THE TRIMERELLIDA.

of subcardinal scars, w. Brachial valve rather tumid at the umbo =
platform somewhat strongly trilobed; its outer margins a little
raised; its antemedian portion rounded, projecting, and terminating
in a slightly developed median plate: crescent a strongly marked
linear scar on the hinge; arching forward in front of the cardinal
facet; inner border of its sides with strongly marked indenta-
tions—one near the hinge, another further forward; outer border
a fine line: a rather strongly marked subcardinal scar, w, in the
umbonal cavity: a large rhomboidal postmedian scar, p, in front of
the latter.

Type species, Obolus Conradi, Hall.

The somewhat general description given in the above diagnosis
points out some of the chief differences between the present genus
and the preceding ones. In the latter the area is longer than wide ;
consequently the umbo is much more projecting. In Dinobolus de-
velopment of the valves is in general bilateral rather than longitu-
dinal; and the areal characters are, as a consequence, not well exhi-
bited: nevertheless the hinge shows the crown of the crescent rather
conspicuously. This last part in both valves differs from its equiva-
lent in Trimerella and Monomerella: the outer border of its sides is
a simple line which runs off from the hinge along and within the post-
lateral margins of the valves, as in these genera; but the inner bor-
der, instead of also being a simple line, is irregular in width, and com-
posite in structure, being broad and strongly indented—at least in two
separate places: the ends of the crescent are not well defined, arising
from interblending growths. The platforms nearly resemble those
of Monomerella; but the one belonging to the pedicle-valve has
its front sinuated (sometimes five-lobed),—not straight, as in the
genus referred to. The umbonal cavity of the pedicle-valve shows
a tendency to become double-chambered, the division being caused
by a slightly developed cardinal buttress (Pl. XVIII. fig. 7, 7a).
This region shows a very small pair of umbo-lateral scars (Pl. XII.
fig. 5x) below the subcardinal scars, w, already noticed. The
deltidium and other parts belonging to the area are imperfectly dis-
played in all the specimens we have examined; though in one spe-
cies (D. canadense, not examined by us), according to Mr. Billings’s
figure of it, copied in Pl. XIX. fig. 7, the deltidium appears to be
well developed. The dentary system is probably rudimentary. The
platforms occasionally show a tendency to become vaulted.

The genus embraces D. Conradi, D. Davidsoni, D. canadensis,
D. transversus, D. magnificus, and some others which may be mere
varieties.

Drvosotus Conrant, Hall. Pl. XVIII. figs. 1-5.

Obolus Conradi, Hall, Twentieth Annual Report of the Regents
of the University of the State of New York, p. 368, 1868; and
revised ed., p. 375, 1870.

Trimerella Conradi, Dall, American Journ. of Conch. vol. vii.
part 2, p. 83, 1871.

Dinobolus Conradi, Hall, Preliminary Notice, Twenty-third Report

Y. DAVIDSON AND W. KING ON THE TRIMERELLID®, 161

-on the State Cabinet of New York, p. 3, March 1872: Dav. &
King, Report of the Brighton Meeting of Brit. Assoc. 20th
August, 1872.

General form orbicular ; depressed; broadly rounded anteriorly ;
‘tapering rapidly in the umbonal region. Pedicle-valve flatly convex:
area small: deltxdium concave: platform undulated on its surface ;
sinuated, and elevated in front: cardinal buttress slightly but
obviously developed: crescent conspicuously displayed. Brachial
valve rather tumid at the umbo: platform sinuated, and raised at
the sides; produced, and depressed in its antemedian portion:
median plate short: crescent conspicuously displayed in part.

Of this important species some specimens have attained to 13 inch
in length by 1 inch 11 lines in width; but the larger number of
specimens that have fallen under our notice presented almost
equal length and breadth. In a specimen measuring 11 lines in
length, the platform of the pedicle-valve is to some extent doubly
vaulted, the vaults attaining two lines in depth (Pl. XVIII. fig. 2, »);
but in none of the other examples we have seen did they exceed a
line or a line and a half.

Dinobolus Conradi occurs in a dolomitic limestone belonging to the
upper part of the Niagara group (= Upper Silurian), at Leclaire,
Iowa; also at Racine, Wisconsin, United States of America,

Divosotus Davinsont, Salter. Pl. XVIII. figs. 6-11.

Obolus Davidsoni, Salter, MS. ; Woodward, Manual of the Mollusca,
p- 240, 1856; Davidson, Mon. Brit. Foss. Brach. vol. i. Intro-
duction, p. 58, pl. 4. figs. 30-39, 1866; Salter, Murchison’s
Siluria, 2nd ed. p. 543, 1859 ; Lindstrom, Sil. Brach. of Got-
land, Ofversigt. af. K. Vet. Akad. Forhandl. p. 375, 1860.

Dinobolus Davidsoni, Dav. & King, Report of Brighton Meeting
of Brit. Assoc. 20th August, 1872.

General form slightly wider than long ; subequivalve ; compressed :
surface marked with concentric lines of growth. Pedicle-valve
moderately convex: umbo very short: area and its component cha-
racters small: crescent well defined in its crown, the central part of
which curves forward ; its sides and ends complex: hinge a T-shaped
ledge, the top branches forming its sides, and the perpendicular
part forming the cardinal buttress : platform with sides straight, and
diverging from its posterior end; a broad sinuated front; and a
strongly undulating surface. Brachial valve moderately convex:
crescent well defined at its crown, the centre of which is pointed, the
point directed backward: platform trilobed in outline; its central
lobe or terminal extremity projecting, and depressed below the lateral
lobes: close behind is situated the postmedian scar, p, which is
rhomboidal in outline: further behind is the subcardinal scar, w.

This species occurs occasionally so well preserved that it has
afforded the principal characters diagnosed for the genus, especially
the sears on and near the hinge, also those belonging to the platform.
The crescent, as will have been observed, is instructively displayed in

aed

162 £, DAVIDSON AND W. KING ON THE TRIMERELLID A.

some of the English specimens: it is also well shown in a cast we
have been favoured with by Dr. Lindstrém, and which remarkably
confirms the accuracy of the drawings (made previously to our seeing

the cast) of this part, as represented by fig. 6 of Pl. XII. The plat

forms have the usual scars—three pairs on each—very well displayed.
The postmedian scar, p, in the brachial valve, has all the appear-
ance of being a visceral impression.

Dinobolus Davidsoni, the first discovered form among the Trime-
rellids, was erroneously referred, in 1853, by one of the writers to
Obolus; with which genus it seemed at the time to bear the closest
affinity, especially as nothing was then known of the shells composing -
the family under description.

From near Dudley, chiefly as internal casts. Mr, C. Ketley informs
us that all the specimens have been found in certain beds of the lower
or “thick” band of limestone of the Wren’s Nest. These beds are
supposed to have their equivalents exposed at the Rushall Canal, near
Walsall; where also similar specimens were obtained during the
cutting of the canal many years since. At Dudley tunnel have been
found a few specimens, doubtfully considered to be D. Davidsona, in
the lower strata. Their absence in the higher strata, with the reverse
erder in another species, D. transversus, may be perhaps considered
worth notice. In the island of Gotland it is very abundant: and
in rocks of a similar age, near Wisby, slabs may be seen covered with
scores of indifferently perserved internal casts and impressions of the
shell. It is stated to oceur at Ferriters Cove, Kerry, Ireland.

Drvozotvs caAnapensts, Biltings, sp. Pl. XIX. fig. 7.

Obolus canadensis, Bill. Report of the Gool. Survey of Canada for
the year 1857, figs. 20-23 (not fig. 19).

Obolellina canadensis, Bill. Can. Naturalist, vol. vi. p. 222, 29th
Dec. 1871; and April 1872, figs. 15°

Dinobolus canadensis, Dav. & King, Report of Brighton Meeting
of Brit. Assoc. 20th August, 1852; Annals, and Geol. Mag. for
same year.

General form longer than wide. Pedicle-valve slightly and gently
convex: 6cak somewhat prominent; very slightly incurved: area
rather wide: deltidium deeply excavated*. Brachial valve convex,
especially in the umbonal region+.

In 1857 Mr. Billings included in the present species both elon-
gated and wide forms: of late, however, he has formed of the latter a
distinct species, the next but two under description. Not having had
the advantage of examining specimens of D. canadensis, we are obliged
to depend entirely upon the description and figures of it published by
Mr. Billings. As now restricted the species is exceptional in form.

Width usually about 2 inches; but some fragments that have been
found undoubtedly belonged to individuals which were 3 inches

* Billings states that the exterior of this valve is not clearly shown in any of
his specimens.

+ Billings has figured the interior of the brachial valve, which par ie! shows
the characteristic platform and scars.

T, DAVIDSON AND W. KING ON THE TRIMERELLID&. 163

wide. As the shells that are wider than long have been removed to
another species, possibly the fragments noticed belong to it.

This species is stated to occur in the Black-river limestone at
Fourth Chutes of Bonne Chére, Pauquette’s rapids ; and in the town-
ships of Stafford and Westmeath, county of Renfrew, Canada. Its
geological position seems to be about the horizon of the Upper
Llandeilo—that is, below the Caradoc formation.

Drnopotvs TRANSVERSUS, Salter. Pl. XVIII. fig. 12.

Obolus Davidsoni, var. transversus, Dav., vol. i., General Introduc-
tion, p. 136, fig. 53, 1853; Sil. Brach. p. 59, pl. v. figs. 1-6,
1866.

Dinobolus Davidsoni, Dav. & King, Report of Brighton Meeting
of Brit. Assoc. 20th August, 1872; Annals, and Geol. Mag. of
the same year.

General form much wider than long; compressed. Valves exter-
nally marked by fine striz parallel to their margins. Brachial valve
thick, and elevated in the umbo-cardinal region; thence thin and
flattening out to the margin: crescent strongly impressed at its sides ;
obscurely impressed at its ends: platform depressed; very slightly
raised at its margins; well defined posteriorly: postmedian scar, p,
subrhomboidal; sharply differentiated from the platform scars: sub-
cardinal scars, w, 1 close contact with the latter.

Very little is known with reference to this species, only a cast of
the interior of the brachial valve having been found: it exceeds 24
inches in breadth by nearly 2 inches in length, indicating: a large
size for the species, apparently larger than any of the others. It
differs from D. Davidson in being wider and more depressed—also
in its internal characters, as denoted in the diagnosis. The sides
(apparently their inner border) of the crescent, which are consider-
ably apart from the lateral margins of the valves, have the pecu-
liarities characteristic of other species, but still more exaggerated.

_ Dinobolus transversus occurs in the calcareous shales at the top of
the Wenlock limestone in the neighbourhood of Dudley; and at the
Rushall canal, near Walsall, Park Hall.

Drxoporus Woopwarnt, Salter.

Obolus Davidsoni, var. Woodwardi, Salter, MS.; Dav. Sil. Brach.
p- 60, pl. v. figs. 7, 8, 1866.

Dinobolus Davidson, Dav. & King, Report of Brighton Meeting
of Brit. Assoc. 20th August, 1872; also in Geol. Mag., and the
Ann, & Mag. Nat. Hist. of the same year.

Certain undetermined differences observable in an imperfect cast,
which is transversely oval and marked with concentric lines of
growth, led Mr. Salter to suggest that it might belong to a species
distinct from D. transversus: possibly, however, when better known,
it may prove to be no more than a variety. It occurs in the Wenlock
limestone of Dormington, Woolhope, and in the associated shales.
near Dudley.

164 T, DAVIDSON AND W. KING ON THE TRIMERELLID&.

Drvnosotvs maentricts, Billings, Pl. XEX. fig. 8.

Obolus canadensis (in part), Billings, Geol. Survey of Canada for
1857, fig. 19 (published in 1858).

Obolellina magnifica, Bill. Can. Nat. for April 1872, p. 17, fig. 7.

Dinobolus magnificus, Dav. & King, Report of Brighton Meeting
of Brit. Assoc., 20th August, 1872.

We have not seen speeimens of this species. Mr. Billmgs has
given the figure of one valve only, showing nothing more than
its exterior. Hestates, however, that the “ dorsal valve” is “ trans-
versely broad ovate; width about one fourth greater than the
length, uniformly and moderately convex; apical angle about 130
degrees ; cardinal edges nearly straight or gently eonvex for about
one third the length of the shell; sides and front rounded, the latter
more broadly than the former. The area seems to be obsolete
altogether or merely linear ; the ventral valve is depressed convex,
with a large beak slightly incurved; area with a wide triangular
peduncular groove, no lateral furrows; surface of both valves econ-
eentrically marked with imbricating lines of growth. In a specimen
which appears to have been about 20 lines in length the height of
the area is nearly 3 lines.”

Dinobolus magnificus, excepting being wider and less long, offers
a close resemblance to D. transversus. It occurs in the Black-river
formation (Lower Silurian) along with D. canadensis.

Drvosoivs Scumipt1, Dav. & Kmg. Pl. XIX. figs. 5 & 6.

This species, only known from two specimens, is almost eircular ;
obtusely tapering at the beak: valves moderately convex: surface
smooth. It elosely resembles D. Conradi; but seemingly differs
therefrom in being of less width. The pedicle-valve, the only one
we are acquainted with, possesses impressions charaeteristic of the
genus.

Professor F. Schmidt, after whom we have named this species,
discovered in 1873 the specimens above noticed at Kirna, Esthonia,
Russia. They oceurred in the Lyckholmer Schicht (2* of Schmidt’s
Classification), which he considers to be nearly equivalent to the
Caradoc of Great Britain, and the Trenton limestone of North
America.

—

APPENDIX.

a. Linevtors Wairrretpi, Hall. Pl, XIX. figs. 9, 9a.

In his recently published “‘ Notes on some new or imperfectly
known forms among the Brachiopoda,” &c., Professor James Hall
states that “in a late examination of some Lower-Silurian species,
usually designated as Lrneuna, one of them was found to possess a
muscular impression differing entirely from any described Linguloid
species. The ventral (?) valve presents a small area with a narrow
pedicel-groove ; a large lobed muscular impression, which in the

T., DAVIDSON AND W. KING ON THE TRIMERELLID.X. 165

cast extends as a narrow groove towards the base of the shell.
The character of the area and foramen differ from the typical
forms of Lingulella; though in that genus the form and character
of the muscular impression has not been determined, as far as I
am aware. For this shell I propose the name Linevtors” *.

In plate 13, fig. 2, of the ‘‘ Generic illustrations ” attached to the
“Notes” the author has given a view “showing the arrangement
of some of the muscular and parietal scars, and also the ramifica-
tions of the vascular lines, which, although originating at nearly
the same points as in recent Lingulide, do not extend so far back
towards the beak.”

Professor Hall has liberally supplied us with gutta-percha squeezes
of this interesting form; and as his description and figure do not .
afford sufficient information respecting its characters, we have been
induced to publish the following observations.

The fossil, which is less than two lines in length by one in width,
is an impression or cast, of the interior and outline of a single valve :
it shows, however, some charactersina _, i ,
remarkably fine state when examined by Fig. 3.—Lingulops Whit-
a good pocket-magnifier. Hall’s figure, fieldi, interior magnified.
taken from the natural impression or the
fossil itself, exhibits these characters
with a faulty appearance ; our gutta-
percha squezees show them in relief, and
depressed, exactly as they would present
themselves in the inner side of the valve.
Even our figure in Pl. XIX. is not so |: ,
satisfactory as could be desired, as cer- "|
tain points have only occurred to. us
since the plate was struck off; for which
reason we haye given an additional figure
in the annexed woodcut (Fig. 3).

The most striking features are :—first,
the posterior semicircular broad zone, eee
with an inner sinused border; second, the arched fillet situated
below the hinge, and on the zone; third, the central space marked
with scars; and, fourth, the linear impressions occurring in the
anterior half of the fossil. The fourth feature, no doubt, represents
the median plate, and the pair of primary vessels belonging to the
brachioceele ; the third obviously forms the apophysary system ; the
second is evidently the equivalent of the crescent characterizing
the Trimerellids; but the first we cannot properly identify with any
part belonging to the Palliobranchs.

Hall is evidently disposed to believe that the fossil represents the
pedicle-valve (‘‘ ventral”): we, however, entertain a strong idea
that it belongs to the opposite or brachial valve, notwithstanding
the statement that it “ presents a small area with a narrow pedicel-

* Preliminary Notice ; twenty-third Report on the State Cabinet of Natural
History for the Palzontology of New York, March 1871, p. 2, and reprint of
the same, March 1872, pl. 13. fig. 2.

166 T, DAVIDSON AND W. KING ON THE TRIMERELLID A.

groove.” The gutta-percha squeezes in our possession show very
faintly what has been taken for the “ groove” (much less so than is
represented by Hall, and which even appears too much like a notch _
in our original figure): there is something like a depression; but,
considering two peculiarities to be noticed presently, we believe it
to be nothing more than such as occasionally occurs on the hinge of
the brachial valve of Lingula anata.

The peculiarities alluded to belong to the crescent and the median
plate. In the pedicle-valve of Trimerella, as will be recollected,
the crown of the crescent has a forward curve in the middle, due, we
believe, asin Zingula, to this portion having been forced out by the
pedicle ; while the crescent in the brachial valve has a pointed
crown, with the point directed backwards: now it is the latter
peculiarity that plainly presents itself in Lingulops. The other pecu-
liarity equally sustains our view; for in the Trimerellids, as also in
Lingula, the pedicle-valve has a short or rudimentary median plate,
whereas in their brachial valve this part is well developed and
elongated as it evidently is in Lingulops.

Believing that the fossil is a brachial valve, we feel tempted to
offer a suggestion in connexion with the semicircular zone. In
Lingula anatina a broadish flattened organ, called the setal band,
is attached, chiefly at both its sides, by a muscular cord to the valves
(differing somewhat in each) a little within their margins: in the
brachial valve the setal band is uninterruptedly (not being interfered
with by the pedicle, as in the pedicle-valve) carried round the hinge
behind the umbonal muscle. _ Our suggestion is that, while the semi-
circular zone itself formed the seat of the mass of the setal band *,
_ the arched fillet was produced by the attachment of the outer mus-
cular cord and associated vessels; also, that the inner scalloped border
is mainly due to the attachment of the posterior and post-lateral
walls of the splanchnoccele f.

We have yet to explain the peculiar sinused character which
gives the semicircular zone such astriking resemblance to a moorish
arch. The inner side of the zone is cut out by five sinuses—one in
the middle, and two on each side—only opposite ones being equal.
The sinuses are each the outer side of a circular scar, its inner
and imperfect side lying near or against the central apophysis
or platform: it is only lately that we have been able to make out
the form of the sinuses. Let us next try to explain their origin.
The middle sinus, we have little doubt, has been produced by the
umbonal muscle pressing against the postparietal or posterior wall of
the splanchnoceele. It might be suggested that the lateral sinuses
have been formed in the same way by other muscles; but as Lingula
is not characterized by any similarly situated (at least the brachial

* The entireness of the zone at its apex forms another evidence in favour of
the valve being the brachial one.

+ See “On some characters of Lingula anatina,” Annals and Mag. of Nat.
Hist., 4th series, vol. i. July 1873. The setal band and its modifications in the
brachial valve are shown in the two woodcuts already given in a preceding page
of the present memoir (p. 132).

‘TY, DAVIDSON AND W. KING ON THE TRIMERELLID®. 167

valve *), we prefer the idea that they represent certain viscera; and
they may be recesses produced by the pressure of ovarian lobes against
the inner side of the postlateral walls of the splanchnoceele.

The vascular impressions in the anterior half of the fossil are, in
their arrangement, unlike any thing we have seen in the Trimerellids,
or in Lingula; for they run at first in front of the platform, and next
along each side of and close to the median plate. In the Trimerellids
and in Lingula they strike directly out from the sides of the platform,
and take a more or less submarginal direction. At their origin, and
in the medio-lateral region of Lingulops, the vascular impressions
are much obscured by apparently others of the kind, seemingly nume-
rous and passing off to the margins. The ends of the semicircular zone
merge into the same regions, and are equally obscured. Neither our —
figure, nor Hall’s gives any thing more than a faint approximation to
the true characters and grouping of these impressions.

We think it probable that lateral muscles similar to those charac-
teristic of Lingula were implanted in the same regions. Much diffi-
culty attaches to gaining a satisfactory clue to the similarly situated
scars in the Trimerellids.

Considerable resemblance prevails between the platform of Tri-
merella and the apophysis of Lingulops in their outline; and the
resemblance appears to be equally close in their scars: there is
apparently a lateral, a median, and an anterior pair.

With reference to the position which Zingulops occupies, the
characters, as now made known, evidently relate it both to the
Trimerellids and Lingulids; but, while thus affined, we feel un-
decided as to which family it should be placed in. So long as the
other valve remains unknown the safest plan will be to waive this
subject, and simply join with Hall in regarding the fossil as a
** Linguloid form.”

Lingulops Whitfieldi is stated to be a “ Lower Silurian species ;”
but its locality has not been mentioned, further than that it is in
the United States.

b. Curnopres+ Breremant, Dayidson & King. Pl. XVIII. fig. 14,
14abe. ;

If doubts prevail in our minds as to the family to which Lingulops
belongs, we haye doubts of a far wider bearing with reference to the
fossil now to be described. Out of respect to our friend Dr. Lindstrém,
who, believing it to be related to them, wished it to be described
along with the Trimerellids he so kindly placed in our hands for
description, we append an account of this fossil. As it possesses
some characters seemingly resembling those peculiar to the pedicle-
valve of a Palliobranch, the following description has accordingly
been drawn up.

* In the pedicle-valve of Lingula anatina the transmedial muscles have their
attachment similarly situated. Were the fossil the same valve we should have
had little hesitation in referring the lateral sinuses to these muscles, notwith-
standing their being single on one side and double on the other in Lingula.

~ + QClaw-shaped.

168 T. DAVIDSON AND W. KING ON THE TRIMERELLID&.

Pedicle-valve thick, claw-shaped, slightly geniculated across the
middle: the geniculation forms the base of the beak, which is a little
less than half the length of the valve, and almost an equilateral tri-
angle in outline. One face of the beak is flatly convex, and slightly
lineated transversely: the opposite face or area is concave, and
crossed by rather strong curving lines. The other half of the fossil
has parallel lateral margins, and a squarish end with a slight sinus
in its middle. One face, the external one, is flatly convex, and
marked with regular and rather distinct lines of growth running
parallel to the end. The opposite face or cavity has an outline
formed by the sides and end of the convex face, also by the curved
basal side (or hinge) of the area: its surface has two slightly raised
oval tracts situated one on each side, separated for the most part by
a medio-longitudinal hour-glass-shaped furrow, and united by an
arcuated band lying near and parallel to the hinge: the oval tracts
are marked by curving subdivided ribs; and the uniting band is
somewhat tuberculated. The hinge is thin, undercut, and marked
in the middle by a pointed tooth-like process; which has been
inadvertently omitted in fig. 14.

There is something in favour of this singular fossil being related
to Trimerella. Its general form does not depart much from that
of T. Billingsti and T. galtensis. The same may be said of the
“beak.” The “area,” it is true, shows no appearance of a del-
tidium ; but it may be surmised that this part has been developed at
the expense and obliteration of the areal borders,—a tendency of the
kind being displayed in several Trimerellas. The “hinge” is thin,
and projecting or ledge-shaped—not an improbable modification ;
_ and it has actually a central V-shaped projection (the “ pointed
tooth-like process”), reminding one of the curving middle that cha-
racterises the crown of the crescent in the pedicle-valve. Even the
characters belonging to the cavity may be considered as placing this
fossil among the Palliobranchs, inasmuch as the lateral tracts have
a striking resemblance to the convoluted impressions representing
the labial appendages, such as are seen in Davidsonia, certain Spiri-
fers, and Productuses. The curved connecting band might be sup-
posed to represent the highly muscular portion forming the base of
the arms. Still, notwithstanding these remarkable points of resem-
blance, it is our opinion that Chelodes Bergman is not a Palliobranch :
on the contrary, we are strongly inclined to the belief that it belongs
to a section of the Ceelenterates, represented by Calceola and Gonio-
phyllum. In contending for this position we are unfortunately at
issue with Dr. Lindstrom, than whom no one has paid so much atten-
tion to the section in question: nevertheless, feeling that our view
requires more than an incidental notice, we hope to be able to publish
a paper on another occasion giving our reasons for maintaining it.

Chelodes Bergmani was discovered by Herr Bergman, to whom we
dedicate it, at Klinteberg (a hill about 120 Swedish feet above the
sea), Isle of Gotland, in the central Gotland formation,—an equi-
valent in age to the Aymestry limestone of England.

T. DAVIDSON AND W. KING ON THE TRIMERELLID®. 169

DESCRIPTION OF PLATES.

Prats XII.

Diagram figures of the cardinal and internal parts of Trimerella (figs. 1, 2,
7, 8), Monomerella (figs. 3, 4), and Dinobolus (figs. 5, 6).

Pedicle- or Ventral Valve. Brachial or Dorsal Valve.

. Deltidium.

. Deltidial slope.

Deltidial ridges.

. Areal borders.

Cardinal callosities.

Cardinal facet.

. Lozenge.

. Cardinal buttress.
Umbonal chambers.
Platform.

. Platform vaults.
Median plate.

m. Median sears.

fs)

Cardinal sockets.

m mono

Platform.
Platform vaults.
Median plate.

m. Median scars.

MHS SAAD ASS

Paso er

>

n. Anterior scars. mn. Anterior scars.
o. Lateral scars. o. Lateral scars.
p- Postmedian scars. p. Postmedian scars.
q. Crown q. Crown
r. Side f cmscent 7. Side | omsset
s. End s. End
t. Transverse scars. é. Transverse scars.
u. Archlet. uw. Archlet.

». Cardinal scars.
w. Subcardinal scars. w. Subcardinal scars.
x. Umbo-lateral scars. az. Umbo-lateral scars.

Fig. 1. Pedicle-valve of the genus Trimerella, principally from specimens of 7.
grandis. k, vaults passing under the platform, j ; i, chambers in the
umbo, one on each side of the cardinal buttress,h. The wmbonal
chambers and their casts are exhibited in fig. 2c, and fig. 2, Pl. XIIT.

2. Brachial valve of the same genus, principally from 7. grandis and
T. ohioensis. The scars, s (also their correspondencies, s, in the pe-
dicle-valve), are usually difficult to define; but the left-hand one, as
shown in a woodcut in the text (p. 146), has been satisfactorily made
out in one of the specimens we have examined.

3. Pedicle-valve of Monomerella, chiefly made up from WM. prisca and M.
Walmstedti. h,a perpendicular plate supporting the hinge (cardinal
buttress) ; it is thinner than usually in Trimerella: x, umbo-laterat
scars that occur one on the outer side of each wmbonal chamber, i.

4. Brachial valve of the same genus. The scars, 2, probably occur in
Trimerella, fig. 2; but they have not been satisfactorily detected (ap-
parently a pair of corresponding scars, x, occur in the pedicle-valve of
Trimerella, fig. 1): they are very obvious, however, in a cast of Mo-
nomerella prisca.

5. Pedicle-valve of Dinobolus, chiefly from specimens of D. Davidsoni and
D. Conradi. The pair of subcardinal scars, w, is on the edge (rounded)
of the hinge. ‘The pair, x, are below the latter, each member in a shal-
low depression corresponding to an internal chamber.

6. Brachial valve of the same genus. p has much the appearance of a
visceral scar. The parts, 7 and s, have been determined from a num-
ber of good examples of D. Davidsoni. Taking in the line that runs
off from the hinge a little distance along and within the postlateral
margins of the valve (the line was detected on the right of one of these

170 T, DAVIDSON AND W. KING ON THE TRIMERELLID A.

margins), the part r (side of the crescent) corresponds to the same part
in Trimerella and Monomerella, except that its inner boundary is much
thicker, and has a compound structure: probably the inner portion of
the deep impression situated at the angle, near the hinge, is the homo-
logue of the scar, x, in Monomerella, fig. 4. The sides of the crescent,
as represented in fig. 5 (pedicle-valve), may not be strictly accurate.

7. Area and hinge of pedicle-valve of Trimerella.

8. Medio-longitudinal section of the area, hinge, platform, and median plate
of pedicle-valve of the same genus. i, wmbonal chamber on left side
of the cardinal buttress, hh; k, platform vault on left side of the median
plate, |.

Prate XIII.

Fig. 1. Trimerella wisbyensis, n. sp. Interior of the brachial valve, from a
gutta-percha squeeze taken from an internal cast. Wisby strata (=
Wenlock), near Wisby, Gotland.

2. Trimerella grandis, Billmgs. 2, internal cast of pedicle-valve (minus
the beak) : 2 a, internal cast of brachial valve : 2 6, interior of brachial
valve: 2, restoration of the pedicle valve: 3, natural cast of pedicle
valve. Guelph limestone (Upper Silurian), Hespelar, Canada.

Prate XIV.

Figs. 1, 1a, 1b. Trimerella Lindstroemi, Dall. Exterior of a large specimen. Cen-
tral Gotland (Aymestry limestone), Klinteberg, Gotland.

Interior of pedicle-valve. Wialmsudd, Gotland.

—— —. Interior of brachial valve. Klinteberg, Gotland.

— —. Restored longitudinal section of both valves.

. Fragment of a pedicle-valve, showing the deep depression

below the hinge.

Transverse section across both valves. e

: . Internal cast of the brachial valve. Island of Faré. Col-
lection of Herr Fegreeus. The projections on each side of the poste-
rior end are casts of umbonal chambers.

SID GR whe

PLATE XV.

Figs. 1,14, b,c,d,e. Trimerella Dalli,n.sp. From an internal cast. Guelph
limestone, Hespelar, Ontario. 1, cast of pedicle-valve; la, of brachial
valve; 16, of both valves in conjunction ; 1c, interior of pedicle-valve,
from a gutta-percha squeeze; 1d, N, vaults, exposed under platform ;
M, umbonal cavities, separated by the cardinal buttress, seen in fig. le;
le, interior of brachial valve,—n, one of the platform vaults.

on Interior of brachial valve, showing cardinal process. Same
formation and locality.
3,3, b. Internal cast, tolerably perfect, in the Museum of the

Geological Survey of Montreal: a, brachial valve ;. 3, pedicle-valve.

4, 4a,b,¢. T. acuminata, Billings. Internal cast of both valves in conjunc-
tion, from the Guelph limestone, Hespelar, Ontario: Museum of Mon-
treal. 4d, mould in gutta percha taken from the umbonal region of
same cast, showing the cardinal process projecting into the interior of
the shell and towards the pedicle- (upper) valve.

Internal cast of pedicle-valve, with one of the projections

(cast of platform-vault) broken.

Internal cast of a pedicle-valve, from original example
figured in 1862. Hespelar; Museum of Montreal. The three lines
are furrows defining two long platform-vaults.

7, 7a. . 7, internal cast; 7a, gutta-percha squeeze, slightly
restored at the margin. Guelph limestone (Upper Silurian), Hespelar,
Canada; Museum of Montreal.

6.

T. DAVIDSON AND W. KING ON THE TRIMERELLID&. AE

Pirate XVI.

Fig.1. Trimerella acuminata. Interior of brachial valve: N, cavity under

platform exposed.

2. Interior of pedicle-valve: N, cavity under platform exposed; O, um-
bonal chamber. These moulds were taken from the internal cast figured
in Pl. XV. fig. 4: the beak is added in fig. 2. Guelph limestone, Hes-
pelar, Ontario.

. T. ohioensis, Meek. Internal cast, showing the pedicle-valve.

The same cast, showing the hinge-characters of both valves.

Niagara Group, Upper Silurian, Otwa County, Ohio.

Internal cast of brachial valve. Same locality.

Gutta-percha mould taken from the hinge portion of the

cast so as to show ridge-like cardinal process, &c. Specimens 3 to 5

were sent us for illustration by Mr. Meek.

. Gutta-percha mould of brachial valve, taken from an internal
cast in the possession of Mr. Whitfield, of Albany. This mould shows
the cardinal process in the form of two circular depressions in the centre
of the hinge, also indications of rude articulation (elongated plates) of
the valves. From the Niagara Group of Yellow Spring, Ohio.

. Trimerella Billingsii, Dall. Internal cast of pedicle-valve.

Gutta-percha mould taken from the same cast so as to show

the interior of the valve. Guelph Limestone, New Hope, West Canada:

Museum of Montreal.

2X Be

an

£© 00

Prate XVII.

Figs. 1, 1a, 6. Monomerella Walmstedti, n.sp. A perfect testiferous specimen.

Fig.

“Central Gotland,” Klinteberg, Gotland: Museum of Wisby College.

2, —— . Interior of pedicle-valve.
3. —— ——. Interior of brachial valve, Same museum and locality.
———— Internal cast of pedicle-valve. Same formation: Island

of Faré. Coll. of Herr Fegreus.

5, 5a, b,c. —— prisca, Billings. Internal cast of both valves in con-
junction. Fig. 5. The pair of scars (the impression on middle)
belong to the brachial valye. Guelph Limestone, Hespelar, Ontario :
Museum of Montreal.

6, 6 a. . 6, internal cast of pedicle-valve. 6a, gutta-percha
mould taken from the same cast, showing the interior of the valve.
Same locality and museum.

Interior of the pedicle-valve, taken in gutta-percha from a

large internal cast. Same locality.

Interior of brachial valve of same species, taken in gutta-
percha from an internal cast. Same locality.

9, 9a, 4, e. Lindstremi,n.sp. 9,9 a, ternal cast. 9b, showing both
valyesin conjunction. 9c, interior of pedicle-valve from a gutta-percha
mould. “ Wisby Strata,” near Wisby, Gotland.

10, 10 a. orbicularis, Bill. 10, external cast of brachial valve; 10a,
of pedicle-valve. Guelph Limestone, Elora, Ontario: Museum of
Montreal.

PruatE XVIII.

1. Dinobolus Conradi, Hall. Interior of pedicle-valve from a gutta-percha
mould.

2. . Another specimen. Same valve, showing deep vault, N, ex-
tending under the platform. From Niagara limestone, Leclaire, Iowa.
Collection of Prof. Hall.

3. Interior of brachial valve.

4, —— ——. Internal cast of pedicle-valve.

5. —— ——. Natural cast of brachial valye. Same museum and locality.

6, 6a. —— Davidsoni, Salter. Exterior. Wenlock shale, near Dudley,

172 T. DAVIDSON AND W.. KING ON THE TRIMERELLID®.

Figs. 7, 7 a. Dinobolus Davidsoni. 7, internal cast of pedicle-valve : 7 a, interior
of same valve taken in gutta-percha from internal cast. Wenlock,
Dudley. Dudley Museum.
8, 8a. 8, internal cast of brachial valve. 8a, interior of same
valve from a gutta-percha mould taken from same specimen: the two
valves are in juxtaposition in the cast.

9: Interior of pedicle-valve. This specimen, which shows no
undercut under the platform, is taken in gutta-percha from an internal
cast. Dudley. In the Museum of Practical Geology, London.

11, 11a. - Internal cast: from the “ Central Gotland” (=Wen-
lock) of the neighbourhood of Wisby, Gotland.

12. transversus, Salter. Internal cast of brachial valve. Top of Wen-

lock, Dudley. Dudley Museum.

13. Trimerella galtensis, Bill., sp. Interior of pedicle-valve. Guelph
limestone, Galt, West Canada. Collection of Prof. Hall.

14, 14.4, 6, c,d. Chelodes Bergmani, nu. g. et n. sp. The tooth-like process
in the centre of what appears to be the hinge is not represented. ‘“‘ Cen-
tral Gotland” (=Aymestry limestone), Klinteberg, Gotland. Collec-
tion of Herr Bergman.

Pratt XIX,

Fig. 1. Trimerella ohioensis, Meek. Exterior of brachial valve; showing also

area and deltidium of pedicle-valve.

Internal cast of pedicle-valve. Upper Silurian, Otwa,
Ohio county. After drawings by Mr. Whitfield, communicated by
Prof. Hall.

3, 3a. ? Monomerella Walmstedti. 3, internal cast of pedicle-valve: the
projections are casts of umbonal chambers : 3 a, internal cast of brachial
valve. Upper Silurian, Kerkaw, Livonia, Russia. After MS. draw-
ings by the late Prof. Pander, communicated by Prof. Schmidt.

A, 4a. Trimerella galtensis, Bill., sp. 4, internal cast of brachial valve ;
4a, interior of same, enlarged. Upper Silurian, Galt, Canada, Mu-
seum of Montreal.

5,6. Dinobolus Schmidti, n. sp.? Pedicle-valve: internal casts. Lower
Silurian; Kiina, Esthonia, Russia. From specimens forwarded by
Prof. F. Schmidt ; and the only two examples of the genus hitherto
found in Russia.

7. Dinobolus canadensis, Bill., sp. After a figure published by Mr. Bil-
lings. ‘Trenton or Black-river limestone. Lower Silurian, Canada.
Museum of Montreal.

8. Dinobolus magnificus, Bill., sp. After a figure published by Mr. Billings.
Same locality and museum.

9, 9a. Lingulops Whitfieldi, J. Hall. 9, natural size: 9a, interior of
brachial (?) valve, much enlarged. Lower Silurian, United States,
America. Collection of Prof. J. Hall.

Discussion.

Mr. Hicxs remarked that the oldest known Lingulid is Zingulella
primeva. The Lingule in the earliest rocks increase in size as they
approach shallow deposits ; and higher up, namely in the Lingula-
flags, a change takes place in their form. In the Menevian beds,
which were deposited in a deep sea, the shells are small. Mr. Hicks
thought, therefore, that we must admit these changes to be due to
changes in the conditions of existence. The Lingule had been
declared to be allied to the Annelids; the Trilobites were also allied
to the Annelids; and these three are the earliest known forms of

Jeu
|

Soc . Vol. XXX.

Quart. Journ. Geol.

Damdson &% King del

Mintern Bros . imp.

TRIMERELLIDE.

Fle phe GaN a eat NeUeg eta ty
lat. JOULILIisE OL, 00C. \

Tho’ Dawidsan del eb hth MXN Hanhart imp

INEM AGI, MILD

a

i

Se eg
ae

anes
raise

7
J)

Vo ~

fal
SVOVG:
}G

n

lan
FEO
TEO

1uagial

tart Jo

0,
We

MXN Hanhart mop

Tho Davidson del et ith

Re Dye

M

8

li

Tho’ Davidson del

BLIDA.,

ICU

T

| Juart Journ.Geol Soc Vol XXX PL XVI.

Ie

Tho’ Davidson d 1 MAN Han)

=~

Quart Journ .Geol.Soc'

Tho’ Davidson del et lith

M&N Hanhart imp

hart imp

M & N Ham

ai

we MIE HD)

(er 2]

Iv

Ae

M&N.Hanhart. imp.

Davidson. del.et hth.

5

Tho

T. DAVIDSON AND W. KING ON THE TRIMERELLIDS. Wis

life. It would be well worth while to follow the inquiry how far
they were related.

Prof. Ramsay was delighted to hear the opinion expressed by
Mr. Davidson, that the distinction of species was merely an abstract
idea. He had always considered that the apparent distinction of
species was due to the absence of the connecting links. He was
thankful to Mr. Davidson for stating that in different geological
formations there were forms which had received distinct specific
names but which he could not distinguish otherwise than by their
derivation. Mr. Hicks had spoken of the three forms of organisms
which occur deep down in the Cambrian, and founded an argument
upon their rarity. But Prof. Ramsay had long maintained the
heterodox opinion that the Lower Cambrian was a freshwater for-
mation, in which marine deposits are here and there intercalated.
He looked forward to the future discovery of marine Lower Cam-
brian beds with a much richer fauna, and considered that it was
only from accidental circumstances that the fauna of the Lower
Cambrian is so poor.

Mr. Eruerine® also called attention to the poverty of the Lower
Cambrian fauna in Great Britain.

Mr. Borp Dawxurs remarked that the most lowly organized forms
seemed to be the most persistent.

Ors, G8: No: lis: 0)

174 D. MACKINTOSH ON GLACIAL TRACES IN

17. On the Traces of a Great Icz-sHEEt in the SoutnmRn Part of
the Laxe District, and in Nort Wares. By D. Macxintosx,
Esq., F.G.S. (Read January 7, 1874.)

(Abridged.)
_ [Puatr XX.]

[| Arrer a few introductory remarks on roches moutonnées, the dis-
tinction between primary and subsequent strie, &c., the author
proceeds to consider the main subject of his paper as follows. |

A tabular statement of facts will, I think, clearly show that the
primary or most persistent glaciation of the south-central part of
the Lake District, must have been produced by an _ice-flow capable
of ignoring the drainage of the country to a much greater extent
than could have resulted from any system of confluent glaciers strictly
so called.

Primary glaciation.
Far Hasdale (near Grasmere), rocks generally smoothed

fromiibetweent:2 ii sacs ee es ae OE each atetineeiog N.N.W. and N.W.
Entrance to Far Easdale (east side), stria from ......... N.W.
Kast of Easdale House, striz from about .............00+- N. 10° W.
Near Sourmilk waterfall, strixe from ................0cee0e. N.W.

Near Blind Tarn moss (crossing the outlet), striz from... N. 30° W.
On summit of the rock and bog tableland between

Kasdule and Great Langdale up to at least 1700 feet

above sea, roaches moutonnées smoothed from about ... N.N.W.

Stris/on ditto fron ee ee N.N.W.
A short distance north of High Close, striz from ...... N. 40° W.
Top of High Close col, striz from ..............s000-seceeee N.
East of Chapel Stile (Great Langdale), strie from ...... N. 30° W.
North of Elterwater Village, rocks smoothed and striated

ALON i ehios Hotisis wolslunis seimasuatee wel eiettattetelsc a ve Gasete elias N. 26° W.

Bottom of Great Langdale west of Elterwater Village
and south-west of Chapel Stile School, rocks smoothed

andistriated: from eons. eces snc aces see orcs seen eee sonts ses N.W.
On roadside (up hill) from Elterwater Village to Dale

1D Aha rs asRoirMs (anc SessadonodaddaioacoeodacdudasasdoRuoKebos N. 40° W.
Higher up on same road, strie from ...............:.-00000 N. 40° W.
North of Little Langdale Tarn, strie from ............... N.W.
Between Little Langdale Tarn and Blea Tarn, striz

PROM eal eece signe assks euaguae name aawsun anaemune saiose meaemeaeod N. 30° W. and N.W.
Near the Parsonage (Grasmere), striae from about ...... N.W.

Jast of Town End (Grasmere), strise from ...........-.-- N.N.W.

Roches moutonnées between Town End and Whitemoss

quarry, smoothed from about .........cecsseeeeeeeereeerees .W.
At Whitemoss quarry, strie from about .................- N.W.
Rydal valley (about 1 m. north of Wordsworth’s house),

strie from ...... aanauelosenana uaciae wane resus ace ameseseesues N. 10° W.
Near Ambleside (on Grasmere road), striz from ......... N. 20° W.
Near Ambleside Church, striz from about ............... N. 15° W.

Rocks on north side of f Loughrigg Fell, smoothed from
SUG CEU ino J saEE JusrBooS anon scee sOd dobhaduocesonodn anos aKebuKdae N.N.W.

THE LAKE-DISTRICT AND NORTH WALES, 175

Primary glaciation.
Loughrigg Tarn and the neighbourhood, strize from
DOU DM eee rece nce spice suicasiveaiios ose aaeaceas seccsbeeeetee N.N.W. and N. by W.
In one place crossed by strize from about N.W.
North side of Brathay valley (between Loughrigg Tarn
and Clappersgate), striz from........+..ssssseessstereeeeee N.W.
Crossed by striz from a little N. of W.
Near Pull (on road from Clappersgate to Hawkstead),

SUIS a MOMy AO OUb pec ceecen xeric cossin santos aaeneseee eee eae N. 27° W.
Quarry near Mr. Atkinson’s house (north-west of Win-
dermere) metrics Sones. Aeon hasiseandsss Meee N. 30° W.
Crossed by later strix from N. 10° W
Windermere churchyard, broad grooves FLOM vs eseseeeee N. 25° W.
Crossed by sharp strize from N. 10° W.
Near Mr. Pritt’s house, Bowness, strie from ............ N. 30° W.
In front of a house called the Ferns, Windermere, strice
fironal between Aiea tte. oa tee ne eee cee eee mae N. 20° and N. 30° W. .
Near Rev. Mr. Stock’s house, Bowness, striee (possibly
Secondary) MmOMts piece uses rirranaatesseecates te see tine N. 10° W.
Near Crossings (Windermere) broad grooves, from
IDELWEENE Nene tenatetteaeceere eater once Steet eae N. 20° and 23° W.

Crossed by strix from about W. 35° N., and a few
grooves from about N. 33° W.
About halfway between Windermere and the Lake,

BUNT ATOM ses oansine secure teeeaeeeadsacie seleanelaemiaclae ses N. 33° W.
Orrest Head (Windermere), smoothed from about ...... N.N.W.
Top of a high ridge called School Knott (near Winder-

mere), crossed by great grooves from...............ss0008 N. 35° W.
Near Ings (east of Windermere) a number of large

roches moutonnées smoothed from about..........-.+--+++ N.N.W.
Broad grooves and striz on ditto, from between ......... N. 20° and 380° W
Rocks at a considerable height on west side of Trout-

beck valley, smoothed from about ...........seeeeeesesees N.N.W.
Watershed between Style End (Kentmere) and Long

Sleddale:striz from: \ i ...sscceneneseaee crete eee eee
West of Stavely Parsonage, primary glaciation from

POON sopdoangdocdocosoondadde DesdoocasndoeanoOnoaud05coaR6G0 eoeeotl NVVis

Crossed by large grooves from nearly.................. W.

Near entrance to Tilberthwaite valley (about High

Yewdale, Coniston), rocks extensively smoothed up hill

AROMA ADOUWG ea cinieeicsia ganas scieaniosionetis umeinmesaneaeeaeeta tate N
North side of Church Beck valley about halfway between

Coniston and Copper-works, rocks extensively grooved

obliquely down hill from about ............06+..2s00e coco Jt)

Crossed by striz running in direction of valley or
from N.N.W

Near Copper-works, strize (covered with stratified sand
Pee anoyorcavel) ih OMerepencmacncertn ester etteneses testcase N.N.W.
‘Between Copper-works and Paddy End, strie from...... N. 30° W.
At mouth of Reddale valley above Copper -works, striz

THE@NE soaoannventesnnensoqodaiaon edodeunscnsancqns°co0anAnAsESK> . N.W.
On steep slope above Copper-works, rocks extensively

striated down hill from about. ...............c:e:essee eee es N. 10° W.
Roches moutonnées on nearly opposite slope, smoothed

vung) Linllll Theasael @]970)1 nepogempeseeonceneecspaeocbosd-ecs sone suns: 3 IN
Between Duddon Bridge and Seathwaite Church, 7 oches

moutonnées with parallel undulations and striz coin-

ciding in direction, from about ...............:s2seeseeee N.N.E.

It will be seen from the above statement of facts that the pri-
mary ice-marks embrace an area extending from near Style End
0 2

176 D. MACKINTOSH ON GLACIAL TRACES IN

and Stavely (Kentmere), on the east, to nearly as far west as
Stickle Tarn and the Coniston Old Man *—and from Far Easdale
on the north to as far south as Bowness and Church-beck valley,
Coniston. In the neighbourhood of Windermere they average
about N. 27° W., and run generally up hill. About Ambleside their
average direction is nearly the same. Around Grasmere they
average about N.40° W. ‘To the north-west and west of Grasmere,
in the upland valleys and on high ridges, they average about
N. 30° W. South of Grasmere and in Great Langdale they average
about N. 35° W. In the Coniston district they average a little
W. of N.

In the neighbourhood of Windermere and Bowness, the ice, be-
sides moving generally up hill, must have ascended and passed over
Orrest Head (700 feet above the sea), crossed a high ridge called
School Knott (760 feet) at right angles, and the upland valley
between Windermere and Stavely at nearly right angles. North-
west and west of Ambleside it must have obliquely crossed Rydal
water, and a high, if not the highest part of Loughrigg Fell (1100
feet). East of Grasmere it would appear to have smoothed the side
of Rydal Fell up nearly to the summit ; but, so far as yet known,
the ice must chiefly have ignored the configuration of the ground in
a district extending for some miles to the north-west, west, and
south-west of Grasmere. From the north slope of Far Hasdale, it
must have obliquely crossed the dale, smoothed the rocks on its
south side up hill, then crossed a high ridge, descended into Kasdale,
smoothed the rocks obliquely across the dale, marched from the
bottom of the dale (900 feet) up hill to the top of a ridge more than
1700 feet above thesea. This ridge consists of hard volcanic breccia
and other felspathic rocks ; the summit is a narrow tableland con-
sisting of alternate rock-basins and bosses; the rock-basins are
partly filled with peat and water ; and many of the bosses are roches
moutonnées. Nowhere in the Lake District have I seen such a
striking series of mammillated rocks: they have been considerably
roughened by the weather ; but the regular curvilinearity of their
forms has been perfectly preserved. From the appearance of the
surface of the great tableland on the north-west (which reaches a
height of between 2300 and 2500 feet), and the rounded rocky
eminences on its southern border, it can scarcely be doubted that
the ice went over it; but I did not examine the ground further west
than Stickle Tarn. From the top of the ridge just described the ice
must have gone down into great Langdale, smoothing the sides of
the projecting rocks which faced it or looked up hill, and leaving
the down sides cliffed or jagged. Numerous examples of this
smoothing of the sides of the rocks which offered the greatest

* The parallel undulations and striz of Duddon valley, though probably
primary, as they coincide in direction with the valley, may at present be left out
of consideration.

t The ice-flow which crossed School Knott must have come from at least as
far as Rydal (a distance of about 7 or 8 miles), and moved the greater part of
the way over longitudinally level ground, and the latter part up hill.

THE LAKE-DISTRICT AND NORTH WALES. 177

resistance to the descending ice, may be seen on the north side of
Great Langdale from High Close westwards; many of them were
pointed out to me by EH. B. Wheatley Balme, Esq., of High
Close. The ice in obliquely crossing Great Langdale (about 300
feet) in one place nearly coincided in direction, without altering its
course, with a bend in the valley. A greater number of distinctly
striated roches moutonnées may be seen in the lower part of Great
Langdale than perhaps anywhere else in the Lake District: they
extend to the base of the ridge called Lingmoor on the south side of
Great Langdale ; and the eastern part of this ridge is striated in
such a manner as to leave no doubt that the ice passed over it. In
some places further west there are indications of the ice having
ascended the north side of the highest part of Lingmoor (about
1500 feet), and glaciated Side Pike, nearly 1200 feet ; but the rocks
are much dilapidated and their bases scree-strewn. On the south
side of Lingmoor the striz and roches moutonnées clearly point to
ice having descended from its summit ; so that it may safely be in-
ferred that the ice-flow crossed over Lingmoor from Great into
Little Langdale. In the Coniston-Old-Man area we meet with
phenomena which cannot be very satisfactorily explained without
supposing that the ice-sheet continued its march not only across
Little Langdale, but (after a westerly deflection of its course) up
the northern slopes of the Coniston Fells, over ground at least 2000
feet above the sea, down into Church-beck valley (600 feet) *, and
’ up the side of the ridge on the south, beyond which I have not
traced the ice-marks. As, however, it is barely possible, though
not probable, that the ice which accomplished such feats among the
Coniston Fells may have belonged to an ice-dome rising to a great
height above tbe northern part of the Fells, and sloping down
towards Little Langdale, we may principally direct our attention to
the Easdale and Langdale ice-sheet until the Coniston mountains
have been more extensively examined, especially to the north and
north-west of the Copper-works.

Source of the Easdale and Langdale Ice-sheet.—Several years ago
I noticed that the northern part of the Lake District must have been
mainly glaciated from the south; and Mr. Ward has brought his 8.
and N. ice-marks to within two miles of where my N. and §. ice-
marks begin, as may be seen from his map (Quart. Journ. Geol. Soc.
vol. xxix. pl. xy.). It is impossible that a strip of ice not two miles
in breadth could have originated two ice-flows in opposite directions,
and both of them many milesin length. The northern ice-flow may
have taken place at a later period than the southern ; but however
this may have been, I think it may be regarded as certain that the
southerly ice-flow could not have performed the feats above specified
without being backed up on the north by an ice-dome rising to a
great height and covering many square miles of country to the
north of Far Easdale. Until the central part of the Lake District

* At intervals along the whole of the northern slope of Church-beck valley
N. and §. primary strix may be seen running obliquely down hill. The instance
noticed in the tabular statement is perliaps the best-defined.

178 D. MACKINTOSH ON GLACIAL TRACES IN

has been carefully searched for ice-marks further west and east (but
especially further east) than the area treated of in this paper, it
might be going too far to invoke an ice-flow assailing the Lake
District from without, and overriding an area extending west and
east from the Duddon valley to Kentmere, if not to the West Riding
of Yorkshire, and as far south at least as Morecambe Bay.

Traces of an Ice-sheet in North Wales.—The state of the basin of
the Irish Sea between the Lake District and North Wales, at the
time of the greatest development of the ice, cannot be well ascer-
tained without a wider induction of facts than we at present possess.
The north-western side of Snowdonia has been principally glaciated
from the 8.E. and S. or in the direction of the lower valleys. In
the high-level valleys, as long ago shown by Professor Ramsay, there
are strize which indicate a thickness of ice sufficient to have enabled
it to cross minor ridges and hollows, and to move along the sides of
hills at great altitudes. During recent visits to North Wales I saw
many glaciated surfaces between the Vale of Conway and Capel
Curig, the striz varying from between 8.W. and W.S.W., and indi-
cating an ice-flow capable of ignoring hills and valleys. In the
great Ogwen Pass, near a farm-house called Wern-go-ischaf, I found
an extensive rock-surface striated at right angles to the pass (or
nearly N. and 8.) which a small glacier coming down from above
did not seem capable of explaining, especially as the lines did not
coincide in direction with the small lateral valleys on the right and
left. To the south of the Snowdonian range of mountains I
happened to alight on a number of phenomena which clearly indi-
cated the southerly movement of a great ice-sheet capable of
ignoring or crossing deep valleys, and which probably had its
source in an ice-dome covering the peak of Snowdon and the sur-
‘ rounding heights, and levelling the area between Snowdon and
Moel-wyn. The group of mountains of which Moel-wyn is the
principal, furnishes evidences of the former existence of such an
ice-sheet. On the southern side of Bwlch-cwm-orthin (which
separates the head of Cwm Croesor from Cwm Orthin) away from
any valley, and at a height of more than 1800 feet above the sea,
J found a number of rock-surfaces, smoothed, mammillated, and
striated from about W. 30° N. There is a possibility of these
surfaces having been glaciated by floating ice ; but when viewed in
connexion with roches moutonnées at a lower level, soon to be noticed,
I think it is more probable that the agent was land-ice of a thick-
ness sufficient to fill up and override the upper part of Cwm
Croesor before it found its way to the irregular plateau on which
the glaciated rocks occur. Between Cwm Croesor and Beddgellert
mammillated rocks may be almost everywhere observed ; but the
greatest display occurs immediately to the south of the bare and
craggy felstone ridge called Yr Arddu; I believe they are unequalled
in any part of Wales or, perhaps, in the Lake District. Their regularly
rounded and approximately dome-shaped forms, not exceptionally,
but as a general rule, furnish an unquestionable evidence of a great
flow of land-ice, as icebergs would have tended to flatten and plane

py

THE LAKE-DISTRICI AND NORTH WALES. 179

down projecting rocks, though, under exceptionally favourable cir-
cumstances, they might have left rocks more or less rounded.
These roches moutonnées are not in a valley, but on an irregular
plateau. They occur at various heights, and must have been
smoothed by ice which moved independently of the drainage of the
country. Indeed the ice by which many of them were smoothed
must have come over Yr Arddu. They are so much smoothed that
it is often difficult to tell the precise direction from which the ice
came ; but taken in connexion with a number of striw, the direction
would appear to have been approximately N. The ice may have
radiated from the Snowdonian dome south-easterly towards Bwlch-
cwm-orthin and southerly towards Yr Arddu. East of Moel Wyn
the mouth of Cwm Orthin is magnificently mammillated, as long
ago mentioned by Professor Ramsay; and a survey of the surface-
configuration would, I think, lead to the conclusion that the mam-
millation was principally caused by a great ice-sheet and not by a
corry glacier.

The above remarks on the primary glaciation of a part of North
Wales are intended as supplementary to those on the Lake District,
and not as exhaustive of the subject.

[The author concludes with observations on the correlation of
the drifts of the Lake District with those of North Wales, the
relation between lake-basins, drifts, and moraines, the commence-
ment of the great submergence while the land was deeply covered
with ice, &¢., &c.}

EXPLANATION OF PLATE XX.

Map intended to show the positions and directions of the primary striz
and rock-smoothing in connexion with the surface-configuration of the prin-
cipal area described in the paper.

Discussion.

Mr. Warp was inclined to regard the scratches in the Lake-dis-
trict described as due to the confluence of several glaciers, so as to
form a large mass of ice, the pressure of which enabled it to travel
over the ridges separating the valleys, especially at their lower ends.
If the phenomena could be explained in this manner, he thought it
needless to invoke the existence of a large general ice-sheet. If
such a thing had existed, it must have brought some of the rocks
from the north and deposited them in the district ; and this was not
the case.

Mr. D. C. Davies thought that the author had left some cireum-
stances out of view, especially the difference of dates of the striz on
the Welsh mountains, which had been cut at different times during
the elevation and depression of the land. He instanced the occur-
rence of fragments of Scotch granite in gravels at an elevation of
from 1500 to 2000 feet above the sea.

180 oN GLACIAL TRACES IN THE LAKE-DISTRICT AND NORTH WALES.

Prof. Ramsay observed that some years ago he had attempted to
show that Anglesea had been glaciated by ice that had come from
the north in the Cumberland district, and attributed this cireum-
stance to the preponderance of this northern ice over that from the
Snowdon range, which was, as it were, set aside by it. He was
inclined to think that the Menai Straits, the direction of which
coincided with the main lines of glaciation in Anglesea, might be
due to the same cause.

W. SHONE ON THE DISCOVERY OF BOULDER-CLAY FORAMINIFERA, 181

18. Discovery of ForaminirEra, &e., in the BouLpER-cLays of
CursHirE. By Witiiam Suonz, Esq. (Read March 11, 1874.)

[Communicated by D. Mackintosh, Esq., F.G.S.]

In the Quarterly Journal of the Geological Society for November
1872, Mr. D. Mackintosh described the Upper and Lower Boulder-
clays of Cheshire, and he also appended lists of the Mollusca found
therein.

I cannot add any thing to Mr. Mackintosh’s description, it being a
most accurate, full, and exact account of the lithological and strati-
graphical characteristics of the drift-clays of Cheshire. For some -
time I had been searching for Foraminifera in these clays in the
vicinity of Chester, but failed to find them. Last September, how-
ever, my friend Mr. J. B. Manning (Constable of Chester Castle)
happened to find in the Upper Boulder-clay of Newton-by-Chester a
boulder bored by Sawicava rugosa, in the cavities of which fragments
of the shells remained. Wishing to possess these fragments, he
proceeded to wash them out; but in doing so he observed that the
holes were not filled with the red clay in which the boulder was
found, but with sand. On examining this sandy débris beneath the
microscope, we found that it contained several shells of Polystomella
crispa. Mr. Manning thereupon remarked, “if we are to succeed in
finding Foraminifera in the Boulder-clay we must look for stones
with holes in them.”

It then occurred to me that the turbinated shells of Turritella
terebra would offer a still more effective shelter to the Foraminifera.
I happened to possess a considerable number of these shells, gathered
from a newly exposed railway cutting, which passed through Upper
Boulder-clay at Newton-by-Chester ; so I put the idea at once into
practice by washing out the substance which filled the inner whorls
of the Turritelle. After pouring off the fine muddy particles there
remained behind a sandy residuum, in which Foraminifera, Ostra-
coda, Sponge-spicules, and the spines of Hchint were abundantly
distributed. The Foraminifera were in a most perfect state of
preservation—so perfect, indeed, that I found it impossible to
distinguish the fossil forms taken out of the Turritelle of the Upper
Boulder-clay from the recent ones of the same species which I had
gathered along the shores of the estuary of the Dee.

On examining a number of T'wrritelle, I observed that some were
entirely filled with a remarkably fine greyish white sand, others only
partly filled with it, though invariably that portion of the shell
nearest the apex; while some T'urritelle were wholly filled with the
red clay in which they were imbedded. I found that the Fora-
minifera &c. occurred in those shells which were wholly or partly
filled with the greyish-white sand, and that the Turritclla filled
with the red clay scarcely contained any Foraminifera &c.

The fact that the Turritelle filled with the greyish-white sand

182 W. SHONE ON. THE DISCOVERY OF FORAMINIFERA ETC.

yielded abundantly Foraminifera and Ostracoda, and that those
wholly filled with the red clay in which they were imbedded con-
tained scarcely any, would appear to suggest that the former
had been transported to their present positions by the same agency
which brought the pebbly gravel and striated erratics which lie
mingled with them in the clay.

On examining the clay in which the Turritelle lay imbedded, I
found that it contained from three to six shells of Foraminifera per
cubic inch. I obtained from it about twenty Foraminifera of five
species identical with those occurring in the Turritelle, but more
worn and fossilized*. I will not, however, at present venture an
opinion as to whether the Foraminifera found free in the clay lived
in situ, or were washed out of Turritelle filled with them previ-
ously to being transported hither from some distant shores.

I am indebted to my friend Mr. J. D. Siddall, Chester, for the
names of the Foraminifera, and to the Rev. H. W. Crosskey, F.G.S.,
and G. 8. Brady, Esq., F.G.S., F.L.S., for those of the Ostracoda.

In the following list U. B. C. signifies Upper Boulder-clay, and
L. B. C. Lower Boulder-clay.

Where found.

———_. ps | fe,

OstRACODA.

Cythore villosa ...........2...00- In Turritelle from U. B.C., Newton, Chester.
antiquata ........0...seeeee 5p ” ”
= {logan ENS Gosdaaocosooode os Ay 0
—— JOMESH ........... ec cee eeees x » »

WEIN) Heh. CrigacoueaHobousooaoNG ” 99 ”
Cytheridea punctillata? ...... ” ” »
Cytheropteron nodosum ...... ” ” 9
Cytherura angulata ............ as: ”
Eucythere? argus............... 9 »
Loxoconcha tamarindus ...... ¥s 0 »
AMPLESSA) | aseeesics ce eee 5 66) 3
Paradoxostoma arcuatum ... % ” ay
HeEXUOSUM .cecesnce seer esse. fi 5 »

ANNELIDA.
Spirorbis nautiloides ......... 3 53 3

EicHINODERMATA.

Spatangide, fragments of
SPINES! ste scas ses csmaencccs 6 i 3
Cidaridz, too much broken
to identify species ............ 3 : _ e]

SPONGIDA.

Triradiate sponge-spicula ... ah 69

_* See note on p. 29 of Quart. Journ. Geol. Soe. vol. xxx. for list of sixteen spe-
cies of Foraminifera found in the Boulder-clay of Liverpool by Mr. Reade, F.G.S.

IN THE BOULDER-CLAYS OF CHESHIRE. 183

Where found.

In Turritelle from

a

FoRAMINIFERA,

Biloculina ringens, very rare ......... U. B. C. Newton- | L. B. C. Dawpool.
by-Chester.
Triloculina oblonga, uncommon ...... ” ” ” ”
trigonula, TARE ceccccceccccvccecrecs ” ” 7 ”
Quinqueloculina seminulum, common ” ” ” ”
secans, frequent.............2..2..2- a ”
Lagena sulcata, rare .........-02s.ceeeee ” ”
USS VAS) PAC) « sirsicisee veissieteite v oiisie nore ” ” ” ”
Biriatas Mares an scemsnecactelsoaeeee cae ” ”
globosa, rare ............sceeeeeeee ” ”
marginata, rare ...........2...6600e- ” ”
BQUAMOSA, LALO ....-.eeeeeeeereeeeee ” ” ” ”
Cristellaria crepidula, rare ............ ” » ” ”
Nodosaria scalaris ?, very rare..... Beas ” ”
Dentalina communis ?, very rare ...... ” ”
Polymorphina lactea, not common ... ” »
Uvigerina angulosa, very rare ......... » ”
Globigerina bulloides, frequent ....-. ” ” ” ”
Textularia variabilis, frequent ......... » ” ” ”
difformis, frequent aidistat=\elalchate’s| e/a ” ” ” ”
globulosa, uncommon ............ ” ”
Bulimina pupoides, very fine large
shells, COMMON .......--e00e eeeeceee eee ” ” ” ”
marginata, uncommon ...........- ” ” ” ”
oyata, not COMMON ...........-46. ” ” ” ”
elegantissima, frequent............ ” ”
Bolivina plicata, frequent ............ 006 ” »
Discorbina rosacea, frequent......... Bn
Planorbulina mediterranensis, fre-
QUCTIE ....ce ee ee ecee eee eet ence eee eeeeeees
Truncatulina lobatula, frequent ......
Rotalia Beccarii, frequent and well
Geyelopedicrace.cccccs aeciersecerevesee=
== nibida, small and rare) yasqacn see

depressula, uncommon............

On comparing the recent Foraminifera from the tidal reaches of
the Dee with the same species from within the Turritelle in the
Upper Boulder-clay, Mr. Siddall and I found it impossible to dis-
tinguish the one group from the other, except that Rotalia Beccartt
and Bulimina pupoides are finer in the Boulder-clay.

The general facies of the Foraminifera of the Upper Boulder-
clay appears to indicate a shallow sea, into which large quantities
of fresh water intermittently flowed.

According to the observations of Mr. Mackintosh, the Upper
Boulder-clay of Cheshire, Lancashire, and Cumberland cannot be
traced much higher than 500 feet above the present sea-level. The
Foraminifera and Mollusca I have found in the Upper Boulder-

184 W. SHONE ON THE DISCOVERY OF FORAMINIFERA ETC.

clay of Chester, so far, appear to confirm Mr. Mackintosh’s conclu-
sions. When the surface of this part of England was submerged
beneath the Upper Boulder-clay sea, and the striated erratics from
the Lake-district, borne on floating ice, were being dropped on the
muddy bottom consisting of this clay, there must have been large
volumes of fresh water poured into it, especially when the accumu-
lated winter’s ice and snow melted before the increasing heat of
summer. Would not such an addition of fresh water to a shallow
sea be likely to have the same effect upon the Foraminifera of that
period as the freshwater of tidal rivers has upon the recent Fora-
miniferal fauna characteristic of estuaries ?

I may here mention that the Foraminifera and Ostracoda from
the Turritelle in the Lower Boulder-clay of Dawpool bear a
more fossilized appearance than those from the Turritelle in the
Upper Boulder-clay of Newton-by-Chester. The Ostracoda from
the Dawpool clay have not yet been named.

I trust that those interested in the fauna of the English drifts
will endeavour to search for Foraminifera &c. in these glacial
deposits. I believe that the distribution of the 7wrritelle containing
Foraminifera and Ostracoda &c. will be found to be coextensive
with the distribution of the Lake-district erratics in the Boulder-
clays.

Note.— For the results of the Rev. H. W. Crosskey’s and
Mr. David Robertson’s investigations of the fauna of the Post-
tertiary fossiliferous beds of Scotland, see the ‘ Transactions of the
Geological Society of Glasgow,’ vols. iii. and iv.

Note on Mollusca from the Boulder-clay at Newton, near Chester.

CoNCHIFERA. GASTEROPODA.

1. Mytilus edulis, Linné. 12. Lacuna divaricata, Fabricius.
2. Cardium echinatum, L. 13. Littorina rudis, Mazon.

3. C. eduie, L. 14. Turritella terebra, Z.

4, Cyprina islandica, L. 15. Aporrhais pes-pelecani, ZL.

5. Astarte suleata, Da Costa. 16. Purpura lapillus, Z.

6. A. borealis, Chemnitz. 17. Buccinum undatum, L.

7. Tellina balthica, L. 18. Murex erinaceus, L.

8. T. calearia, Ch. 19. Trophon truncatus, Strém.
9, Mactra solida, L.; var. elliptica, | 20. Fusus antiquus, L.

Brown. 21. Nassa reticulata, Z.

10. Mya truncata, L. 22. Pleurotoma pyramidalis, Str.
11. Saxicava rugosa, L. :

All the above except Aséarte borealis and Pleurotoma pyramidalis
inhabit the British coasts, and are littoral or sublittoral. _A. borealis
and P. pyramidalis are more northern species [J. Gwyn JErrreys |.

Discussion.

Mr. Evans remarked that there seemed to him to be two principal
points for discussion in the paper :—first, whether the Foraminifera

IN THE BOULDER-CLAYS OF CHESHIRE. 185

eited were peculiar to brackish water ; and, secondly, whether the
Turritelle had been transported.

Mr. Gwyn Jerrreys said that the Foraminifera sent by Mr. Shone
are exactly the same as those found on the coasts of Great Britain.
The Turritelle presented a puzzling question. ‘The Foraminifera in-
habit the Laminarian zone, and their light shells are thrown up at
the edge of high water, so that they would naturally fill any shells
that might be lying on the shore about that line. The Turritelle
might have been transported by ground-ice. The species was T'ur-
ritella terebra, the common European species. Mr. Jeffreys remarked
that we know comparatively little of the Arctic fauna at present,
and that it was highly desirable that an expedition should be sent
to investigate the marine fauna of high northern latitudes.

Prof. T. Ruprrt Jonzs stated that the Rotalie are not identical
all round the coasts, those from different localities presenting dif-
ferent characters, as may be plainly seen in the Rotalia Beccurti of
the Adriatic and of the English coasts. Various circumstances
seem to act in changing the forms, especially whether the animals
have inhabited deep or shallow water, or water more or less fresh.
The Globigerine have thicker shells in deep than in shallow water.
When ill-nourished, Foraminifera alter in the style of their outline.

Prof. Hugurs remarked upon the difference of opinion prevailing
as to the geographical affinities of the shells found in this deposit,
and as to the origin of the deposit itself. He discussed the question
of the origin of the clay, and came to the conclusion that it was not
a true Boulder-clay, but derived.

Mr. Seetey referred to a Boulder-clay, at March, in Cambridge-
shire, containing Foraminifera now common on our present shores.

186 W. TOPLEY ON AREAS OF APPARENT UPHEAVAL.

19. On the CornrEsPONDENCE between some Arras of Apparent Up-
HEAVAL and the TuickENING of Sussacent Beps. By W. Tortezy,
Esq., F.G.S., Geological Survey of England and Wales. (Read
February 4, 1874.)

ConrTENTs.

. Introduction.

The Jurassic and Triassic Rocks of Central England.
The Oolitic Rocks of Yorkshire. —

The Carboniferous Rocks of Yorkshire.

The Carboniferous Rocks of Derbyshire.

. The Weald.

. Observations on Basins.

. Conclusion.

COTS? SH G9 BO

1. Introduction.—The fact that the Secondary strata of England
vary considerably in thickness as we trace them over any wide area
is very well known. Professor Phillips long ago observed the south-
easterly thinning of the Oolites in the Midland Counties *; and
Professor Hull, in 1859, read before this Society a Memoir on this
subject, giving the results obtained by himself and others engaged
on the Geological Surveyy. Recent investigations as to the probable
extension of the coal-measures under the south-eastern counties, have
again drawn attention to this subject ; and much information bearing
upon it is contained in the Reports of the Royal Coal Commission.

In the following paper I shall endeavour to show that, from these
facts, some important conclusions may be drawn concerning the ob-
served dip of the strata, and the disturbances which the strata are as-
“sumed to have undergone, conclusions which I think cannot be denied
if the facts be granted, but which have hitherto been overlooked.

It is known that the Tertiary and Cretaceous beds under London
lie in a “‘ basin,” whilst as we recede from London the beds rise at
varying angles ; so that strata which occur at considerable depths
under London attain great elevations in the surrounding districts.
It has always been assumed that this is due to an upheaval of the
beds in the direction in which they rise. I think, however, that 1f
we allow the full weight to known facts as to the thinning out of
beds (facts which every one accepts), we must conclude that this is
in many cases an erroneous assumption. In some cases the observed
dip and rise of the beds can be wholly accounted for- by the known
or inferred thinning and thickening of the beds.

Not, however, the thinning of beds the dip of which is seen ;
although this has some small effect. But we must take into account
the sum of the thinning of all the underlying beds. If we do this,
we can often account, not for a part only, but for the whole of the
observed dip.

* Manual of Geology, p. 303.
t “On the South-easterly Attenuation of the Lower Secondary Formations of
England,” Quart. Journ. Geol. Soc. vol. xvi. p. 63.

W. TOPLEY ON AREAS OF APPARENT UPHEAVAL, 187

2. The Jurassic and Triassic Rocks of Central England *—As an
illustration of the thinning out of the Oolitic rocks of this area, we
may take a section from Leckhampton to Burford (fig. 1).

Leckhampton Hill is one of the prominent points of the Cotteswold
escarpment ; and here the thickness of the Inferior Oolite has been
carefully measured. If the summit of this hill were capped by Great
Oolite, the base of that formation would be at an elevation of nearly
1000 feet. At Burford, which lies 19 miles a little to the south of
east of Leckhampton Hill, the average elevation of the base of the
Great Oolite is about 400 feet. In ordinary geological language we
should then say, that from Leckhampton Hill to Burford the beds
have a dip of 600 feet in 19 miles.

Along this line the exact amount of the thinning of the Inferior
Oolite and Upper and Middle Lias is well known, and that of the
Lower Lias may be approximately inferred. At Leckhampton the
thickness of strata intervening between the base of the Great Oolite
and the base of the Lias is about 1200 feet; at Burford the total
thickness is probably under 200 feet. There is therefore a differ-
ence in thickness between the two places of about 1000 feet of
Lower Jurassic strata, whilst the dip of the Great Oolite is only
600 feet in the same distance. Therefore the thinning out of the
Jower beds will far more than account for the observed dip of the
higher beds.

To make the matter more plain, let us refer both geological
horizons under consideration to the present sea-level. The base of
the Great Oolite at Leckhampton would be 1000 feet above the sea-
level; at Burford (19 miles distant) it is 400 feet above the sea;
therefore the dip of the Great Oolite is from west to east, 600 feet
in 19 miles. The base of the Lias at Leckhampton is about 200
feet below the sea-level, at Burford it is about 200 feet above it;
therefore the dip of the base of the Lias is from east to west, 400
feet in the same distance of 19 miles.

But we have here only considered the thinning of the Inferior
Oolite and Lias, whereas it is known that the New Red series thins
in like manner in the same direction. The amount of this thinning
cannot be even approximately known between these two places ; but
it will probably be several hundred feet.

The thickness of the Trias in Cheshire is given by Professor Hull
as 5600 feet, and about 600 feet in East Warwickshire ; it is pro-
bably absent under Oxford. From Cheshire to Warwickshire the
Trias thins 5000 feet in about 80 miles, or about 62 feet per mile.
If this district were covered by higher Secondary rocks, the dip of

* The data for the observations are taken from Prof. Hull’s paper already re-
ferred to, and from the same author’s Geological-Survey Memoir, on Sheet 44.
[It was pointed out by Mr. Bauerman and Mr, Etheridge, during the discussion
upon this paper, that borings near Burford have proved the thinning of the Lower
Lias to be less rapid than had formerly been supposed. Therefore the westerly
dip of the base of the Lias would be less than that shown in the figure. The chief
point of my argument, however, still remains, that the easterly dip of the Great
Oolite can be entirely accounted for by the thinning of the Inferior Oolite and
Upper and Middle Lias. ]

W. TOPLEY ON AREAS OF APPARENT UPHEAVAL.

188

"aUTTOO way +f ‘sery aoddq ‘p ‘SBITT JOMOTT “g
‘aYOO aorteyuy “a “OUOISTABT “2 “poy MONT “”

coal “7 0001

00s : » 4008

RELA DP PELE REE

DS SSS SS ae
{
{
1
!
1
1
t
!
!

ong

q20qLooa

GA = “paojang ; “YOROTYIAON, THE, voydueyyoory “M

“900 yang ayy fo dig hpsaysngy quanbasuos ayn pun ajyn0g sowafus
pun sovT ay fo no-buruunpy, oyp moys 02 “paofung o2 paz wordiunyyoary out (synng buagun0) woyoay mouboig— | *8uq

W. TOPLEY ON AREAS OF APPARENT UPHEAVAL. 189

these rocks would be affected by this amount, besides that resulting
from the thinning of the higher Secondary rocks themselves.

We see, then, that in estimating the dip of the beds in any such
district, our results will vary according to the rocks which are exposed
init. Butinasmuch as our knowledge of the great movements which
have affected the earth’s crust is generally deduced from such obser-
vations, this liability to error becomes a very serious matter.

We infer, for instance, because of what we know of the dip
of the Great Oolite in Gloucestershire and Oxfordshire, that the
beds have been upheaved towards the west and north-west, which
has resulted in a dip to the east and south-east, or towards the
London basin. But if, over this area, denudation had gone much
further than it has done, and all the Oolites down to the higher part
of the Lower Lias had been swept away, we should observe that the
beds along this line were approximately flat; and we should infer
that there had been no upheaval here, or that the net results of
such movements as had taken place had been to leave the beds in a
horizontal position. If the whole of the Lias had been denuded, we
should infer a slight westerly dip or from the London basin. If
denudation had gone far into the Trias, sweeping away the whole
of the higher Secondary rocks, we might have inferred a considerable
westerly dip *.

3. The Oolitic Rocks of Yorkshire-—These afford another very
striking example. It is known that the entire Oolitic series, which
is so magnificently developed in the northern part of the county, is
absent under the Wolds, and that the Red Chalk there rests directly
on the Lower Lias. In part, no doubt, this is due to the great un-
conformity between the Cretaceous and Oolitic rocks ; but it is also
partly due to actual thinning of the strata, the Lower Oolites having
disappeared entirely in this manner.

In order to estimate the amount of the dip which is due to this
thinning, we have only to compare the small actual southerly fall
of the bottom of the Inferior Oolite along the base of the western
escarpment with the great southerly dip which the Coral Crag would
have if it now stretched (as once it did) along the top of that escarp-
ment. But this is not all; for the Lias itself is thinning in the same
manner, and we find that the base of the Lias undergoes comparatively
small changes of level along the same north and south line.

4. The Carboniferous locks of Yorkshire-—The Carboniferous
Limestone series of Yorkshire has a general dip, from the high land of

* Since this paper was read, Mr. Whitaker has kindly drawn my attention to
the following remarks by Sir W. V. Guise, in his address to the Cotteswold
Naturalists’ Field Club, April 19, 1869. Referring to the meeting of the Club
held on March 25, 1868, Sir W. Guise says :—‘‘ Mr. Lucy mentioned his having
recently made an excursion in the Cotteswolds round Stow and Burford, and
ealled attention to the gradual thinning out of the beds in a northerly [?]}
direction. On the authority of Mr. Hull’s Memoirs, to illustrate Map 44,
Geological Survey, he stated that, while at Cleeve Cloud the Inferior Oolite
attains an elevation of 1130 fect, the Cornbrash south of Burford is not much
more than onc half that height, This, however, is in some degree due to the
greatly diminished thickness of the underlying strata in the latter locality,”

Oro mone Noel 1S: P

190 W. TOPLEY ON AREAS OF APPARENT UPHEAVAL.

the great Penine escarpment, towards the east and south-east. In this
direction the beds, as a whole, thin out*. In part, then, the westerly
rise of the beds is due to the westerly thickening: A noteworthy
exception to this easterly and south-easterly dip occurs between
Wharfedale and Wensleydale, where the beds have a northerly and
north-easterly dip, from what Professor Phillips has called the
‘“¢ Wharfedale Axis” +. Now it is near Wharfedale that the Lower
or Scar Limestone series attains its maximum thickness. The top
of this series—the upper limit of the ‘‘ main limestone ”—is a well-
defined line; and by it Professor Phillips has estimated the dips.
But the dip of this bed is partly produced by the thinning of the
underlying mass of the Lower Limestone; therefore the ‘‘ Wharfe-
dale Axis” is partly due to unequal thickness of underlying strata.

5. The Carboniferous Rocks of Derbyshire.—The rapid south-east-
erly attenuation of the Millstone Grit and Yoredale Rocks in Derby-
shire is well known. From 3500 feet on the west of Sheffield (between
the Bradfield and the Rivelin valleys) the beds diminish to 1500
feet a few miles north of Belper, a distance of about 20 milesf.

The highest bed of the Millstone Grit has therefore, from this
cause alone, a dip of 2000 feet in 20 miles, whichis rather more than
1°. What further proportion of the dip may be due to the thinning
of still lower beds we do not know. If the Millstone Grit here were
entirely covered up by Coal-measures, we might be in ignorance of
this rapid thinning, and we should probably refer the “ dip ” of these
Coal-measures to upheaval or depression.

6. The Weald.—Perhaps the most striking and most important
example of apparent upheaval being partly due to thickening of
strata, is that afforded by the Weald.

Up to the year 1855, when Mr. Godwin-Austen read before this
Society his memorable paper on the extension of the Coal-Measures §,
no one doubted that the Lower Cretaceous and Oolitic rocks passed
beneath the London basin. The Wealden Beds were supposed to
thin away somewhere, but to be replaced by Oolitic strata, the
whole of the beds under the Chalk being bent into a synelinal
supporting the London Tertiary basin.

Under these circumstances it was natural to infer that the strata
had been uniformly folded into basins and anticlinals, and that the
dip of beds exposed at the surface was a true index to the dip of
those below.

But when it became apparent that the succession was incomplete,
thatin fact the entire Lower Cretaceous and Jurassic series were absent
under London, but gradually came in and thickened as we receded
therefrom, the question was entirely changed ||.

* Prof. Phillips’s Geology of Yorkshire, Part ii, Mountain Limestone Dis-
trict, 1836, pp. 19, 32, 41, 46, 77, 175-7. t Loe. cit. p. 137.

¢ See the Sections by Messrs. Green, Foster, and Dakyns, in the Geology of
.... parts of Derbyshire, p. 139, Mem. Geol. Survey, 1869.

§ Quart. Journ. Geol. Soe. vol. xii. p. 38.

|| Some of the difficulties which are here referred to concerning the upheaval
of the Weald, appear to have presented themselves to Mr. J. H.H. Peyton. See
his lecture on the Boring at Netherfield, St. Leonards-on-Sea, 1873.

W. TOPLEY ON AREAS OF APPARENT UPHEAVAL. 191

The rise of the Cretaceous beds towards the central anticlinal of
the Weald is too well known to need description here. The base of
the Gault, which is 940 feet below sea-level at the Kentish-Town
well, would be rather more than 2000 feet above the sea, if the
Lower Cretaceous beds were restored over Crowborough Beacon,
the highest point along the central anticlinal of the Weald. We
may therefore take 3000 feet as the difference of level of the base
of the Gault between the centre of the Weald and Kentish Town.
For the present we will leave out of consideration the minor anti-
clinals of the Weald, and will speak only of the great and general
anticlinal.

At Kentish Town the Gault rests almost directly upon Paleozoic
rocks* ; at Crowborough there would be at least 2000 feet of strata
between them. How much more than this there may be we shall
shortly know by the sub-Wealden boring. But supposing, in order
to simplify the subject, that the Paleozoic rocks are met with at a
depth of 940 feet below the sea, then is it not plain that the actual
rise of the Gault towards the centre of the Weald is accounted for
by the gradual thickening of the underlying beds? The paleozoic
floor would be a horizontal line (disregarding still the minor folds);
the Oolitic, Wealden, and Neocomian strata would all dip to the
north, the higher beds dipping more rapidly than the lower; the
Gault would have the maximum dip of 3000 feet in 36 miles.

If it be true that great movements which have affected the strata
near the surface have had an equal influence on those below (and upon
this assumption a large portion of our geological theories are based),
then the converse of this should be equally true :—If in any area we
can show that deeply buried strata are undisturbed, should we not
infer that no great disturbance has affected the overlying rocks, and
that the dip of these higher beds must be due to some other cause ?

But since the beds with which we are now dealing are partly of
freshwater origin, we cannot apply the same reasoning to them as
we can to those which are wholly marine. The thinning of the
Jurassic rocks may be due to failure of sediment in a south-easterly
direction, where was probably the open sea; this is rendered likely
by the fact that the limestone beds of the Great Oolite do not, so far
as they can be observed, thin in that direction. If the Wealden rocks
were deposited in a delta, the direction in which the freshwater
sediment failed would also be that in which marine conditions most
prevailed. But, so far as we know, marine or estuarine forms are,
in the English Wealden beds, confined to the highest and lowest
members of that series, and where the beds are seen to thin out
against the old rocks they are still of freshwater origin.

This thinning out, however, may have been against the old shore-
line of the estuary or lake; and in that case the supposed horizontal
position of the paleeozoic floor would probably be due to an upheaval
of the bed of the lake or delta, or to a sinking of the old land. This
necessarily complicates the question, so far as the Weald is concerned ;

* There has been doubt as to the age of the lowest rocks found in this well ;
but Mr. Prestwich now regards them as Old Red Sandstone.
pP2

192 W. LTOPLEY ON AREAS OF APPARENT UPHEAVAL.

but even after making every allowance for this, we are still compelled
to admit that the dip of the higher Cretaceous beds is partly due to
thinning of the underlying Neocomian beds.

The main line of watershed of the Weald runs from near Fair-
light westwards along, or near to, the main anticlinal line, as far as
the western extremity of the Hastings beds, near Horsham; it then
turns northwards over the Weald Clay, and goes over the top of
Leith Hill, on the Lower Greensand range. The top of the Weald
Clay attains its maximum height at Leith Hill, where it is about
750 feet above the sea. If the Weald Clay here had only the thicix-
ness which it is known to have further east, it would be necessary
to assume that the main Wealden anticlinal, or the lime along
which the lower beds attain their greatest elevation, passes under
Leith Hill; in which case the main anticlinal and the main line
of watershed would coincide, as they usually do. From an exa-
mination of railway-cuttings between Dorking and Horsham, it is
now certain that the Weald Clay must be of unusual thickness there,
probably not much under 1000 feet.

The result of this is, that if we consider only the base of the
Weald Clay, we see that the main anticlinal lies near Horsham,
where we should expect to find it: if we consider only the top of
the Weald Clay (or the base of the Lower Greensand), we see that
the main anticlinal lies under Leith Hill.

Mr. Beckles some years back published some notes in the J ournal
of this Society upon the Lowest strata of the cliffs at Hastings*.
He there describes a set of beds, the highest member of which is a
peculiar sandstone, which, when exposed in large masses on the
shore has the appearance of a pavement; hence it has been called
the “'Tessellated Sandstone.” At several places between Hastings
and Cliff End sandstone of this character is seen, which by
Mr. Beckles is believed to be the same bed; but this lies at very
different depths below the base of the Ashdown Sand. Either,
then, Mr. Beckles is mistaken as to the position of these sandstone
beds, or the beds between them and the base of the Ashdown sand
thicken out enormously in a comparatively short distance. This
thickening would occur just where the anticlinal comes ; in fact the
“‘ Fairlight anticlinal,” which is assumed to have affected the Ash-
down Sand and overlying beds, would be largely accounted for by
this thickening. For this reason, and also because the “ tessellated
sandstone ” cannot be continuously traced, I have always thought that
Mr. Beckles was mistaken in the matter ; but the numerous instances
which we have considered of dips and anticlinals being produced by
thickening of underlying strata, certainly very much weakens one of
the arguments against Mr. Beckles’s viewr.

* Vol. xii. p. 288.

t Mr. A. Tylor bas remarked upon the thinning of the lowest Wealden beds
of the Sussex coast. In describing a section of the coast, which is given in the
paper, he says:—‘ The passage of some beds of sand-rock into clay is well shown
on the east of Hastings...... ; and the thinning of the Castle Rock on the same
line is also shown. ‘The bearings of the same strata to the west, through St. Leo-

W. TOPLEY ON AREAS OF APPARENT UPHEAVAL. 193

7. Observations on Basins.—We have seen that there is a very
general tendency for beds to thin towards the dip, and to thicken
towards the rise; but the illustrations given are of strata which
support geological basins. With the strata forming the basins
themselves the reverse appears to be often the case; they thicken
towards the centre of the basin.

This is certainly the case with the Lower Tertiary strata of the
Loudon basin; it seems to be also the case with the Hampshire
basin. Professor Hébert informs me that the same thing occurs in
the Paris basin; whilst the Lower Cretaceous beds which support
that basin thicken as they rise to the west.

It is a point of common knowledge amongst the mining engineers
in the north of England that the strata of the Newcastle coal-
field thicken towards the centre of the basin. The seams of coal .
are ata greater distance apart along the centre of the basin than
towards its western margin, the easterly dip being somewhat greater
in the lower seams than in the higher. The difference of the dip
due to the general easterly thickening is of course usually small ;
but it occasionally happens that the difference is very marked, and
in consulting colliery plans it is necessary always to bear in mind to
which seam the plans refer.

Besides the general easterly thickening of the strata, there are
occasionally local thickenings of certain parts, the result of which is
to cause undulations in the overlying beds which are apparently the
result of movements of the strata. I may just mention one case,
although (as the area in question is not yet published by the Geological
Survey) I cannot give full details, which, indeed, are not necessary.
An important scam of coal on the north-east of Morpeth had been
worked at Ashington to its regular outcrop on the west; borings
earried through the Boulder Clay still further west proved that
this seam rolled in again in a small basin, in which a large colliery
(Longhirst) is now working. A deep bore-hole put down at the
outerop of the Ashington seam proves that some sandstones which
underlie it are of great and unusual thickness there; from this we
may infer that the “ anticlinal,’ which throws out the upper seam
for a short distance, is due, largely if not entirely, to the local
thickening of this sandstone. Yet if these were rocks which were
not deeply explored for mining-purposes, and if we reasoned only
by what we see at the surface, we should certainly refer the anti-
clinal and synclinal in question only to movements of the strata.

8. Conclusion.— Whatever value may be attached to many of the
foregoing remarks, I think this much is certain: enough has been
said to prove that great caution is necessary in reasoning, from the
observed dip of beds, to any conclusions as to great movements of
the earth’s crust. In fact, it would seem that we can never feel

nard’s to Bexhill, are of considerable interest, as they appear to lose much of
their thickness before they pass under the highest part of the Hastings Sand series
and the overlying Weald Clay of Pevensey.” —Quart. Journ. Geol. “oc. vol. xviii,
p. 252, 1862.

194 W. TOPLEY ON AREAS OF APPARENT UPHEAVAL.

absolutely certain that any observed general dip is wholly due to
disturbance. The dip may often be so great that we may be quite
sure that some part of it is due to this cause; but it will gene-
rally happen that the proportion of the dip due to disturbance,
and that due to possible thinning of subjacent beds, cannot be
determined.

Besides the question of the movements which the earth’s crust has
undergone, there is another question involved in this inquiry, of
special interest just now; that is, the influence which such moye-
ments have had in producing the present features of the surface.

Most geologists now believe that this influence has been but small,
that it has had little or no direct effect in producing the present sur-
face-features, and that it has only been exerted in guiding the action
of the denuding agents now quietly at work around us. But others
still believe that the direct influence of disturbance of the strata has
been very great. As regards the Weald it has long been held,
chiefly through the writings of Mr. Martin and Mr. Hopkins, that in
the original upheaval of that area a system of longitudinal and
transverse fissures was formed, which not only marked out, but
immediately form the valley-systems.

It is, perhaps, scarcely necessary to remark that there are other
signs of disturbance met with in the rocks besides the rise and dip
of beds. There are sharp flexures and contortions of strata, as well
as actual fractures and faults, which can only be due to movements
of the rock-masses.

But, after taking all these fully into account, I believe that, as the
result of this inquiry, one supposed cause of the formation of surface-
features is materially weakened ; and yet the fact that great denu-
dation has taken place remains. Not only so, but the amount of
denudation which has actually taken place is greater than is gene-
rally supposed. For if the strata which now remain are thicker at
their outcrops and at their escarpments than they are further to the
dip, we may fairly assume that the strata which once stretched
beyond the present escarpments were thicker than any which are
still preserved.

Discusstron.

The Rev. Mr. Fisner remarked that he had always considered
that the Paleozoic rocks under London had formed an axis against
which the Secondary rocks had abutted, instead of their being carried
over the old rocks, as shown in the diagram. He presumed that
there would be difficulty in any natural cause constantly leading to
the thickening of strata at some particular spot during successive
epochs so as to form a ridge in a certain position, and argued that
the shattered condition of the flints in some tilted rocks showed that
they had been violently upturned.

Mr. Srrrzy thought the paper extremely suggestive, though pos-
sibly its suggestiveness had been carried too far. If the author’s
views were correct, the sea in which the beds had been deposited

W. TOPLEY ON AREAS OF APPARENT UPHEAVAL, 195

must have been of enormous depth; but of this we had no evidence.
He could not believe that the Chalk or any other sedimentary de-
posit could in the process of deposition assume such dome-like
forms as would be necessary under the author’s hypothesis.

Prof. Hueuss considered that there were two kinds of thickening—
(1) by deposits from a shore-line, or (2) by sediment accumulating
in a basin. The instances adduced appeared to be the result of
the thickening of strata in a basin. He accepted the cautions of
the author as to hasty deductions from the dip of strata at the
surface.

Mr. Baverman considered that the Lower Lias was of greater
thickness at Burford than supposed by the author.

Mr. Toptuy, in reply, stated that he did not dispute the fact of
the Paleozoic rocks being much disturbed and crumpled, nor did
he deny that there may have been some disturbance of the upper
beds. What he wished to point out was that the disregarding of
the fact that strata thickened in certain directions might be, and
had been, a fruitful source of error.

196 C. CALLAWAY ON A TREMADOC AREA IN SHROPSHIRE.

20. On the Occurrencr of a Tremapoc Arma near the WREKIN in
Sourh Surorsnire, wth Description of a new Fauna. By
Craries Cattaway, Esq., M.A., B.Sc., &. (Read March 11,
1874.)

(Communicated by Dr. 1. A. Nicholson, F.G.S.)

[ Abstract. |

THe author stated that in an exposure of light-green micaceous
shales dipping south-east at 50° at Shineton, near Cressage, which
are represented as of Caradoc age in the Geological-Survey Map,
he found a series of Trilobites and other fossils which induced him
to regard these Shineton shales as belonging to the Lower Tremadoc
series. He described as new species Asaphus Hos, Conocoryphe
Saltert, C. angulifrons, Platypeltis Croftii, Conophrys salopiensis,
Lichapyge cuspidata, Lingulella Nicholsoni, Metoptoma Sabrine,
and Vheca lineata. The author regarded these shales as the equi-
valents of beds containing Dictyonema found near Malvern and at
Pedwardine.

Discussion.

Mr. Ernrriner differed entirely from the author, and thought
the fossils exhibited by him were of Caradoc age.

Mr. Wicks was inclined to refer the fossils to the Upper Llan-
deilo; but the fragments exhibited were not sufficient to enable the
species to be determined.

H. G. SEELEY ON MURZNOSAURUS LEEDSII. 197

21. On Murznosavrus Leepst, a Piestosavrtan from the OxrorD
Cray. Part I. By Harry G. Szetey, Esq., F.L.S., F.G.S. (Read
May 13, 1874.)

[Puare XXT.]

WHILE on a yisit to Charles E. Leeds, Esq., M.A., of Exeter Col-
lege, Oxford, a gentleman whose specimens have more than once
enriched the writings of Professor Phillips, I was shown a Saurian
in such perfect preservation as previously, so far as I am aware,
had rarely been seen except from the Lias. It was gathered from
the Lower Oxford Clay (a stratum abounding in Plesiosaurians in
the middle of England) in Huntingdonshire, in fragments almost
innumerable, which have been adjusted and reunited with remark-
able skill and zeal by the labours of Mr. Charles Leeds and his
brother Mr. Alfred Leeds, so that now the animal displays :—the
front and hinder parts of the skull; the lower jaw, somewhat over
a foot long; a vertebral column of 79 vertebre, from which, how-
ever, nearly all the tail is missing—the vertebre preserved being
44 cervical, 3 pectoral, 20 dorsal, 4 pelvic, and 8 caudal; numerous
ribs; the coracoids and scapule; the pubes, ischia, and iliac bones,
together with both fore and hind limbs. Whether the tail has be-
come a prey to the casual collector’s unscientific mania for bones I
know not ; but these remains, being the fruit of long zealous collecting
and care, could obviously only have been gathered in a district where
competition was kept subordinate to scientific spirit. Finding that
this noble specimen (from which hardly any important part except
the tail is lost) indicated in my opinion a new genus as well as a
new species, Mr. Charles Leeds volunteered to write from my dic-
tation the account of the animal which I am thus by his kindness
able to submit to the Geological Society. The short time at my
disposal compelled me to leave the hind limbs undescribed, to form
hereafter the subject of another communication, when I hope to
offer to the Society some account also of the other undescribed or
imperfectly indicated Plesiosauria from the Oxford Clay, with which
this species may be properly compared.

The Skull.

The skull is represented by the premaxillaries, portions of the max-
illary bones, the frontal bones, supraorbital, postfrontal, parietal, basi-
sphenoid, exoccipital and basioccipital bones, and the lower jaw.

The premaxillaries carry five teeth on each of the bones, which are
connected by a nearly straight median suture. They have the surface
rough and irregular on the superior and lateral parts as well as on the
palate, the roughnesses being due to a conical bulging of the bone
around the entrance to nutritive bloood-vessels. Each premaxillary
bone is triangular infront; but no sutures can be seen separating the
bones from the nasal bones behind, perhaps because the premaxillary

QJ.G.8. No, 119: a

198 H. G, SEELEY ON MUREHNOSAURUS LEEDSII,

bones extend back to the nares as in Plestosaurus. The anterior al-
veolar border is convex and measures 23 inches round the lateral curve
on each side. The posterior median bird-like or lizard-like extensions
of the premaxillaries (unless they are nasals blended with the premax-
illaries) are narrow and smooth, meet each other in a penthouse
ridge in the median line of the skull, and join the frontal bones at a
distance of 5 inches from the rounded extremity of thesnout*. The
frontal bones are about 22 inches long, and terminate backward in
the foramen parietale, the posterior border of which is formed by
the parietal bones. The frontals are flattened bones, with a groove
on each side of a slightly elevated median ridge; the suture between
them is distinct; they measure 13 inch from side to side where
narrowest, are compressed from above downward; and nearly the
whole of the outer parts of the under-surfaces form the upper bor-
ders of the orbits; the sides of the bones are subparallel; they are
slightly concave from side to side. They may have small distinct
supraorbital ossifications in front, above the orbits, though from the
state of fracture in which the specimen was found this is not quite
certain. From the posterior angle, in which they join the parietal
bone, the narrow postfrontal bones are given off. The postfrontals
are then directed outward and slightly forward, and are traversed by
a ridge which divides their outside surface into anterior and posterior
areas. The foramen parietale is large, and is directed downward
and backward, increasing in size as it descends. The parietal bone
is greatly compressed from side to side, forming a sharp longitudinal
keel above ; as in the nearly mature fowl, it shows no trace of a me-
dian suture, though the frontals are separate; it measures 23 inches
from front to back, and terminates posteriorly in a straight, transverse,
almost smooth and flat, vertical suture: the bone widens from side
to side from before backward, and becomes compressed at its outer
posterior border. The hinder part of the parietal bone is occupied
on the underside by a small part of the cerebral surface. This is
concave from side to side, and straight from front to back, with a
slight median ridge. This surface in front was probably occupied
by the small lizard-like cerebral lobes which, as among Teleosaurs,
were prolonged forward into two long olfactory nerves; posteriorly
the cerebral surface is more excavated, seemingly for the posterior
border of the cerebral lobes, which was raised above the portion of
the brain which succeeded it next behind.

The basisphenoid is united by suture with the basioccipital bone ;
like the basisphenoid of [chthyosaurus, it appears to be perforated by
the carotid, which passes obliquely through it. The basisphenoid
appears to be underlapped by the basitemporal or by a backward
prolongation of the presphenoid bone.

The basioccipital bone is 1} inchlong. It is remarkable for having
a nearly hemispherical surface for the occipital condyle, to which the
exoccipital bones do not contribute as they doin many Plesiosaurians

* This interpretation differs from Prof. Owen’s account of the bones in Plesio-

saurus, but agrees with a memorandum of my own on the bones in Plesiosaurus
Zetlandicus, at York.

A PLESIOSAURIAN FROM THE OXFORD CLAY. 199

and in Teleosaurus. The basioccipital condyle is short from front
to back, and has its chief extension from above downwards, in-
dicating that the head of the animal had more vertical than lateral
movement; the condyle is margined by a slight depression. The
under surface of the bone, as usual, is concave from side to side and
prolonged outward and downward into strong short lateral processes.
The exoccipital bone is preserved only on one side, where it is
slightly fractured. It does not appear to be perforated for the
hypoglossal nerve or for blood vessels. Its external surface is nar-
row, flat, and inclined obliquely towards the median line of the
cranium ; it terminates behind in a sharp border, which is inclined
obliquely inward and forward so as to make the foramen magnum
pyriform.

The anterior nares, which are imperfectly preserved, appear to be ~
small, and situate 33 inches from the anterior termination of the
skull and about 14 inch from each other.

Lower Jaw.

The lower jaw is 13 inches long. The symphysis is 2 inches
long, with the symphysial suture obliterated. In length the rami
are gently convex on the outer side. The jaw is broadest from side
to side at the coronoid process, where it measures 64 inches in
width, and 14 inch in vertical depth. At this point, on the under-
side of the jaw, the dentary bone ceases to be prolonged backward :
it looks obliquely downward, outward, and forward ; its surface is
rough, and towards the symphysis is pitted with the apertures of
blood-vessels and nerves. On the underside of the anterior sym-
physial part of the jaw a concavity runs parallel to its inner
border. Each ramus appears to have sockets for twenty-one teeth.
The portion of the dentary bone interior to the aveolar margin in
the front part of the jaw is an inch wide, becoming narrower be-
hind; it is oblique and convex, but is more vertical at its backward
extension. At 43 inches behind the anterior extremity of the sym-
physis it is overlapped by the splenial bone, which in front is only
half as deep as the inner border of the dentary bone which it over-
laps. The surfaces of both dentary and splenial bones are marked
with the canals for many small blood-vessels, while the canal for
Meckel’s cartilage is open for at least 5 inches in front of the articu-
lation of the lower jaw. Posteriorly to the coronoid process the jaw
is compressed from above downward so as to be only 1 inch deep;
its upper part is formed by the coronoid bone, its lower part by the
surangular bone. The coronoid appears to extend backward s0 as
to form the anterior border of the articulation of the lower jaw.
The articular surface is quadrate, about 1,4 inch from side to
side, and less than 1 inch long; it is concave from front to back,
with a slightly elevated oblique ridge passing from the inner an-
terior corner to the outer posterior side; the posterior margin of.
the articulation is elevated, and consists of an inner and an--outer
part. The articular bone appears to be small, thin from. above

Q2

200 H. G. SEELEY ON MURENOSAURUS LEEDSII,

downward, superimposed upon the angular bone; it widens the
jaw from side to side; and below the articulation the angular bone
is compressed from side to side to form a rounded heel. The heel
behind the articulation is 1? inch long; at first it is triangular in
section from being flattened above, but tapers posteriorly in all
ways and terminates in a rounded vertical concavity about 3 an
inch in diameter. The extreme width of the lower jaw over the
articulations is 54 inches; the heels approximate still more and
appear to have measured 4} inches from side to side; but the left
heel is fractured. ;

The hindermost teeth look directly upward; but as they approach
the front of the jaw they are necessarily directed more and more out-
ward and forward.

Vertebral Column.

The vertebral column, as preserved, comprises 79 vertebree, which
T group as 44 cervical, 3 pectoral, 20 dorsal, 4 pelvic, and 8 caudal.
No vertebrae appear to be missing in the series; and probably these
numbers, except for the missing tail, give the vertebral formula of
the species. Arranged end to end, as Mr. Leeds has placed them,
they measure about 13 feet; so that with the missing part of the
tail the length of the vertebral column may have been something
under 15 feet, or nearly as long as in Plesiosaurus macropterus from
the Lias, though the proportion of length of head to length of body
probably comes nearest to Plesiosaurus dolichodeirus, where it 1s one
to thirteen.

Cervical Vertebre.

The atlas and axis are anchylosed; and all trace of their union
with the wedge-bones is obliterated. From front to back these
vertebra measure rather less than 27 inches. The atlantal cup for the
basioccipital is imperfectly preserved ; it is 13 inch deep, but appears
to have been narrower. The whole mass widens from side to side
behind the posterior articular surface of the centrum, being nearly
circular, 14 inch from above downward and 13 inch from side to
side ; it is flattened, but slightly concave. The neural arches are
not preserved, but appear to have been distinct from each other;
and the arch of the axis was more developed than that of the
atlas. The neural canal is flat below and smooth, and 12 inch long,
so that, as is usual in Plesiosaurians, the inferior wedge-bones pro-
long the articulation for the basioccipital bone forward considerably
on the inferior margin. A hypapophysis is developed in the median
line of the underside, but does not reach within 3 an inch of the
hindmost articular surface; anteriorly it is broad, strong, and
rounded, posteriorly sharp and compressed from side to side, but not
greatly developed. The cervical ribs of these vertebra are broken
away; that of the atlas appears to have been very small, while that
of the axis does not differ in its articular facet from that of an ordinary
cervical vertebra.

The seventh cervical vertebra (Pl. X XI. fig. 1) has the centrum

A PLESIOSAURIAN FROM THE OXFORD CLAY. 201

14 inch long, with the articular surface in front 14 inch wide and
rather less than 1jinch deep; on the posterior surface these dimensions
are somewhat exceeded. The articular surface of the centrum is con-
cave, more so in front than behind ; both surfaces are margined by a
groove, which indicates the union of the epiphysis with the centrum.
The centrum is greatly compressed in its upper half from side to side,
and is therefore concave from front to back laterally ; below the middle
on each side there is a sharp slightly elevated ridge, which dies away
at the articular margins. At the junction of the side of the cen-
trum with its base the cervical ribs are given off: each is attached
by a narrow ovate surface about an inch long, is compressed from
side to side, directed downward and a little outward and backward
in acurve. The anterior margin of the rib is convex in length and
sharp in edge; the posterior margin is concave from above down-
ward, and rounded from side to side; the inner surface of the bone
is flattened ; but the outside is more convex: the free end of the bone
terminates in a concavity indicating a cartilage. The suture between
the neural arch and the centrum is obliterated. The total height
from the base of the centrum to the top of the neural spine is 34
inches. ‘The length from the front of the anterior zygapophyses to
the back of the posterior zygapophyses is 2} inches. The anterior
zygapophyses project about 7 of an inch in front of the centrum ;
the posterior zygapophyses project about 4 an inch behind it. The
anterior zygapophyses are large oval facets looking upward and in-
ward; the part of the neural arch behind them is constricted; and
from their base on each side a ridge arises which is prolonged
backward, upward, and outward to form the upper margin of the
posterior zygapophysis (which measures ? of an inch from side to
side) and form the limits of a small table from which the neural
spine arises ; below this oblique ridge the posterior portion of the
neural arch is compressed from side to side; above it the neural arch
rises 1 inch. The neural spine terminates in a sharp short pos-
terior border; a little concave and nearly vertical, and has a sharp
long anterior border inclined obliquely backward.

In passing down the vertebral column the vertebre get steadily
longer from back to front, and steadily larger; the neural spines
become a little wider, from back to front, and stronger; and at the
eighth vertebra the cervical rib has the usual hatchet-shaped pattern.
Lower down the vertebral column the antero-posterior ridge on the
side of the centrum becomes shorter and less elevated, and finally
disappears about the thirty-second vertebra ; the oblique ridge be-
tween the anterior and posterior zygapophyses gradually becomes
more horizontal and less elevated, so that in the lower part of the
neck there is no trace whatever of a platform from which the
neural spine arises. From about the fifteenth vertebra to about the
thirtieth the neural spine, which has become much widened, is com-
pressed from side to side, and terminates upward in a long flat car-
tilaginous surface, and has its anterior border nearly vertical, and its
posterior border inclined obliquely forward. In the seventeenth
vertebra, which has the centrum Le inch long, the neural spine

202 H. G. SEELEY ON MUR NOSAURUS LEEDSII,

rises more than 2 inches above the zygapophysial ridge, while the
vertebra measures 43 inches in extreme height. In the middle of
the neck the articular surface of the centrum becomes much more
circular, and appears to be, relatively to its size, rather less deeply
cupped. In the twenty-fourth vertebra the centrum is 2,4, inches
from side to side in front, and 14 inch from above downward; at
the side it measures more than 23 inches from back to front.
The circumference of the articular surface has now become rough-
ened for the attachment of small connecting vertebral ligaments.
The upper part of the body of the centrum remains still greatly
compressed from side to side. The extreme height to the top of the
neural arch is 5? inches; the zygapophysial ridge is nearly horizontal;
and the sharp anterior process of the neural spine extends forward
between the anterior zygapophyses of that vertebra, and divides the
posterior zygapophyses of the preceding vertebra from each other;
the neural canal is remarkably small and triangular, higher than
wide, and higher behind than in front.

In the thirty-third vertebra (fig. 2) the centrum is 27 inches long,
in front 27 inches deep, and about ;4, to 3 of an inch wider. The
articular surfaces are relatively flatter, while the margin external to
the epiphyses is now becoming wider. The posterior articular sur-
face is about 23 inches wide and 27 inches deep. The zyga-
pophyses are, in front, on their inner articular surfaces concave from
above downward, so that they almost meet in a median line, forming
the lower half of a cylinder. The extreme height to the top of the
neural arch is 7 inches; the extreme width over the zygapophyses
is 33 inches; the antero-posterior extent of the top of the neural
spine is 1? inch; the anterior border of the neural spine is slightly
concave from above downward. Between the posterior zygapophyses,
convex from above downward, there is a median slit into which the
lower anterior margin of the neural spine of the succeeding vertebra
is wedged, thus constituting a new kind of vertebral joint, compa-
rable in complexity to that of Iguanoid lizards and serpents, though
of somewhat different mechanism and function.

In the thirty-ninth vertebra (fig. 3) the neural canal becomes con-
siderably larger ; the vertebral articular surfaces are becoming flatter ;
the cervical ribs are wider in their attachment, and incline more
markedly to the posterior border of the centrum: this vertebra has an
exostosis on the right posterior articular margin; its height to the
top of the neural spine is rather more than 74 inches. The neural
spine is 13 inch from back to front in the middle. The neural spine
reaches its extreme height in the fortieth vertebra, where it extends
to 74 inches. The posterior zygapophyses are margined above by
a slightly elevated ridge.

In the last cervical, the articulation for the rib, which in the three
preceding vertebre has been rising on the side of the centrum and
becoming circular and elevated, is now higher than wide and in-
clined a little backward. The centrum increases in flatness, and the
neural spine is inclined slightly backward ; the articular edges are
more than usually compressed and expanded ; and the nutritive fora-

4 PLESIOSAURIAN FROM THE OXFORD CLAY. 203

mina are very much smaller than in the early part of the neck, in
which there is only one pair. In length from back to front the
centrum is diminished to rather less than 2 inches; its width from
side to side is increased to 2? inches, while from above downward
the centrum measures 2} inches.

Pectoral Vertebre.

In these vertebre the face of the centrum is somewhat larger
than in the terminal neck-vertebre, and has a small central pit;
the articulation for the rib is long and oblique, formed partly by the
neural arch, partly by the centrum; itis longest in the middle
pectoral, broadest in the third. The under surface of the centrum
is rounded from side to side in the first, flatter in the second,
while in the third it is so flat on the under surface as almost to
make an angle with the sides, a character which is slightly exag-
gerated by compression. The neural arches are imperfectly pre-
served, but appear to be inclined a little backward as in the last
cervical.

Dorsal Vertebre.

The neural arch is not well preserved in the dorsal vertebre till
the seventh ; and none of the neural arches of the dorsal vertebree
retain their transverse processes for the attachment of the ribs. The
centrum is modified in form, owing to the rib being raised on to the
neural arch ; it is flat on the underside from back to front, very
slightly concave there from side to side, and has its sides con-
siderably compressed, and shows below the middle on each side a
large nutritive foramen. What was in the cervical vertebre the
pit in the centre of the articular surface of the centrum now becomes
slightly elevated, with a central puncture, in some respects recalling
a Pliosaurian character. The zygapophyses retain the singular
method of articulation already described in the later cervicals.
The neural spines now, however, widen in antero-posterior exten-
sion, while from below upward they are extremely compressed from
side to side. In all the vertebre both of neck and back the neura-
pophyses join the centrum after the manner characteristic of Plesio-
saurus, and show no approach to the circular pedicles of Pliosaurus.
The transverse processes of the seventh and succeeding vertebre
are directed upward and outward from the part of the neural arch
in front of the posterior zygapophyses ; these transverse processes are
compressed from above downward in a sigmoid fold, being higher
in front than behind, and convex above in front, and concave
behind, with corresponding folds below.

The tenth dorsal vertebra has the centrum much as in the ver-
tebree already described, except that the circumference of the arti-
cular margin of the centrum is now becoming much sharper and
losing the rounded edge seen in the neck ; the base of the transverse
process is getting shorter from back to front; the sides of the
neural spine are subparallel; it is 1Z inch wide, truncated as usual at
the uppermost end, which is 7 inches from the base of the centrum ;

204 H. G. SEELEY ON MURZNOSAURUS LEEDSII,

the neural arch is inclined very slightly backward. Beyond this
vertebra, which is the middle of the back, the vertebre get gradu-
ally smaller, and the neural spines shorten, though they retain their
breadth. The sixteenth dorsal has the centrum 14 inch long, and
measures 63 inches from the base of the centrum to the top of the
neural spine; the position of the transverse processes remains un-
changed, but the base of the pedicle has become shorter from front to
back, and is rhomboidal in section.

Pelvic Vertebre.

The neural arch is not preserved in the pelvic or sacral vertebree,
and the form of the centrum is modified in consequence of the
descending position of the transverse process. The two middle pelvic
vertebre appear to have the centrum shorter from back to front than
the others, measuring only 14 inch. The articular surface of the
centrum appears to be somewhat depressed from above downward,
and to be more concave than in the dorsal region, while the under-
side is less convex from side to side. ‘The transverse process is
strong, and subquadrate where broken off at the base. In the last
pelvic vertebra the transverse process appears to have been much
larger than in the others, and to have come away from the centrum
by a sutural surface 11 inch deep by 14 inch wide, which stands up
from the centrum with an elevated margin.

Caudal Vertebre.

In the next three vertebre the caudal rib has descended to the
side of the centrum, and is marked by a large thick fragment of a
rib, which appears to have been anchylosed to the centrum. These
vertebre are less than 13 inch long, and have the articular centrum-
- margins bevelled as in the neck. Asin the pelvic region, the neural
arch rises from a more anterior part of the centrum than is the
case with the dorsal vertebre, and is shorter from back to front. The
third caudal has the neural arch preserved, and shows that the
zygapophysial facets still retain their quarter-cylinder form: the
extreme height from the base of the centrum to the top of the neural
spine is 5inches. The neural spine is directed backward, and tapers
from below upward, where it is rather more than an inch from back
to front, and is somewhat expanded from side to side at its superior
termination.

The fifth caudal (fig. 5) is the first which gives indications of a
chevron bone, and that only on the left side. The facets for these
bones in the succeeding vertebre are singularly large and elevated,
and somewhat triangular in form. They raise the side of the centrum
near them, and are so placed that the chevron bones attached to them
never impress the articular surface of the succeeding vertebree (fig. 4),
in this differing from Plescosauwrus and recalling Mosasaurus. In
the fifth caudal the centrum is 13 inch from front to back, 13 inch
from side to side, and 134 inch from above downward. As in all the
other vertebree, the posterior measurement is slightly longer than the
anterior measurement. The articular margins do not stand up from

A PLESIOSAURIAN FROM THE OXFORD CLAY. 205

the centrum as in the cervical region ; and the upper part of the
centrum is but slightly compressed laterally, so that the sides of the
bone are not very concave from back to front. The pedicles to
which the chevron bones were attached have left rough granulated
surfaces, which do not enter into the posterior face of the centrum ;
they look obliquely downward and backward ; and often one is some-
what larger than the other ; and occasionally the chevron bones were
anchylosed to them*.

The Coracoids.

The extreme width of the coracoids from side to side immediately
behind the articulations is rather less than 14inches. The bones are
imperfectly preserved, the whole of their interior and posterior parts
being broken away ; moreover they are crushed at the articular sur-
faces, exceptin the median line. As much as is preserved of the me-
dian suture measures 4 inches in length and 2 inches in depth; the
surface is pyriform, tapering most rapidly behind. The two bones ap-
pear to have been placed during life very nearly in the same plane. The
abdominal surface behind the broad thick anterior part and within
the curved lateral parts was a shallow basin; and externally the
bones presented a corresponding bulging, which in the median line
of the body formed a marked keel; externally, at right angles to
this keel, where thickest, the bone is compressed and extends trans-
versely as a rounded surface towards the point where the articular
surfaces for the scapula and humerus meet, but does not reach
them. The whole of the articular surface for the scapula lies in
front of this rounded ridge, which has a concave outline in front on
each side, with the concavities approximating anteriorly so as to
form a strong median process, which extends further forward than
any part of the articular surface for the scapula. In front of this
rounded transverse ridge the inner side of the bone is prolonged
anteriorly, so as to form a compressed anterior extension of the
coracoid, very thin and deeply concave in its antero-posterior ex-
tension of nearly an inch, and very slightly concave from side to
side. Itis broken away towards the median line of the animal; but
towards the outer anterior surface it widens and thickens, blending
with the adjacent bone to support the articulation for the scapula.
The length of the scapula-articulation is rather over 23 inches; it
looks obliquely forward and outward. In its compressed state it is
12 inch deep where it joins the articular surface for the humerus,
which is about equally long and concave from before backward, and
from below upward ; its anterior margin is 6 inches, and its
posterior margin 72 inches from the median line. ‘The least
width across both the coracoids from side to side is 93? inches at a
distance of 7 inches behind the anterior apex of the bones; the
extreme posterior extension is 13} inches behind the most anterior

* T have occasionally seen similar caudal vertebrae from the Kimmeridge
Clay ; and about five years ago Mr. Charlesworth showed me a series of cervical,
dorsal, and caudal vertebre from the Kimmeridge Clay of Ely, with the caudal
similat ; but they belonged to a distinct species.

206 H. G@. SEELEY ON MURHENOSAURUS LEEDSII,

point ; the bones have extended further backward; but at the point
where fractured they would have measured 14 inches from side to
side. The outermost lateral margin is flat, concave from before
backward in a continuous sweep; the bones are convex from side
to side, and considerably compressed towards the distal end. The
coracoids have no anterior articular surfaces in the median line of
the body.

The Scapule.

The scapule are of a form distinct from that seen in any
described Plesiosaurian genus. They are imperfectly preserved, but
were placed inclined inward and forward, like the scapula of Plesio-
saurus. The inferior surface is flat from side to side, and coneave
from before backward. The bone is four-sided, comprising two
posterior portions ; the articulation for the coracoid is 23 inches long,
and the humeral surface rather more than 2 inches long; the
diameter of the bone between these extremes is 34 inches; its
extreme length as preserved, measured from the junction of the two
posterior sides to the junction of the two anterior sides, is about 5
inches. The inner margin, forming the outer part of what would
usually be the scapulo-coracoid foramen, is compressed and sharp,
and concave, so as to continue the curve of the front border of the
coracoid bone; the outer margin is nearly straight and sharp, making
an angle with a part of the bone, which is prolonged from the abdo-
minal to the dorsal part of the body. The lateral part is at about a
right angle with the inferior surface. Posteriorly and inferiorly the
anterior process is thickened and rounded, the ridge being prolonged
downward and outward to the inner and anterior humeral margin.
The surface of the bone between this ridge and the scapulo-coracoid

.foraminal border is oblique, looks inward, and is concave in each
direction. The anterior portion may have extended much further
forward, being exceedingly thin ; but its form is such as to suggest
that no clavicular bones or interclavicle existed, and that the bones
may have met so as to complete the one large foramen which they
indicate.

The Pelvic Bones.

All the pelvic bones are preserved. The pubes and ischia each
unite to form more marked median keels than that formed by
the coracoids, while the pubes and ischia do not meet each other in
the antero-posterior median line, as in Plestosaurus, so as to form
two foramina, but constitute one foramen, 84 inches from side to side,
which is slightly indented by the forward mesial angular margin of
the ischia behind, so that it measures 34 inches from back to front
in the median line, and four inches from back to front where longest,
midway in each lateral moiety.

Neither ischium is perfectly preserved ; and each bone lies entirely
behind its articular junction with the pubis. This line of union is
transverse and looks forward, and is nearly 13 inch long, while
the articular surface for the femur is fully 24 inches long ; but its

A PLESIOSAURIAN FROM THE OXFORD CLAY. 207

shape is probably altered by compression. The external surface of
the bone is slightly convex in each direction, and measures 7 inches
from the median line to the femoral margin; its least width is 24
inches at about 2 inches from the femoral surface. The anterior
margin is sharp and compressed and concave, the concavity looking
forward and slightly inward. The bone appears to have had the
usual triangular form seen among Plesiosaurs ; but its thin posterior
expansion is not entirely preserved; its posterior border is sharp
and compressed, thinner than the anterior border, concave in a curve
which extends inward and backward to within about 24 inches of
the median line, where its antero-posterior extension is 44 inches
measured from border to border. The median symphysis is thick ;
rather less than 3 inches of it is preserved antero-posteriorly ; and,
as compressed, it is more than 11 inch deep. The inner side of the
bone is concave from side to side as well as concave from back to
front. The bones meet at an angle of 120°.

The pubis is thin, and only the posterior portions are preserved ;
the antero-posterior extent of the symphysis appears to be about
64 inches ; but this is inferred from the two incomplete bones.

The bone is thickest at its posterior side and becomes exceedingly
thin anteriorly. The width of the bone from the posterior point of
the symphysis to the outermost margin of the femoral articulation
is 8 inches; the ischiatic union extends 12 inch behind this line; and
its innermost margin is within 43 inches of the symphysial line.
The outer margin of the pubis is compressed from side to side, and
directed outward, so that 34 inches in front of the femoral articula-
tion the concave outer margin is more than 83 inches from the
median line. Like the ischium, the bone is slightly convex on its
external side, slightly concave on its abdominal surface; a slightly
inflated thickening runs parallel to the lateral border.

The iliac bones are 7 inches long. The articular surfaces appear
to have been somewhat expanded so as to extend over the upper
part of the head of the femur; but it is difficult in the present state
of the specimens to determine by what surfaces each could have
joined the pubis and ischium so as to form the acetabulum.

Each bone is compressed from side to side, 24 inches broad at the
proximal end, and expanded at the sacral end in the same direction
to a width of two inches, its least width in the middle being 13 inch.
The anterior border from above downward has a slight sigmoid
flexure from side to side; but its anterior lateral margin is nearly
straight and about 6 inches long. The posterior margin is concave,
and 5 inches long. The compressed sacral end is convex from before
backward, and inclined backward. Half an inch below the upper-
most margin the inner side shows scars, 1 inch long, of two sacral
articulations.

The Fore Lamb.
The fore limb is much smaller than the hindlimb. The humerus

208 H. G@. SEELEY ON MURZNOSAURUS LEEDSII.

is 10 inches long, and but slightly expanded at the proximal end of
the cylindrical shaft; and this expansion appears to be due to the
development of the great trochanter in a transverse direction to the
distal expansion, which is 53 inches from side to side. The bone is
compressed slightly from the proximal to the distal end, being about
2? inch deep proximally and 1? inch distally. The anterior border
is slightly concave; the posterior border is much more concave,
especially towards the distal end. In the middle the shaft is about
2 inches in diameter. There are strong roughnesses towards the
proximal end, indicating the attachment of muscles; and a slightly
elevated longitudinal ridge marks the middle of the posterior border
of the bone. The distal articular surface is divided into two flat-
tened areas, apparently with cartilaginous margins back and front, to
which bones may not have been articulated.

The ulna and radius are short from above downward, as in Plesio-
saurus, and quite distinct in form from those bones in that genus.
The radius is compressed from side to side towards the external border,
and thickened towards the ulnar border; it is more than 22 inches
from side to side, and is nearly 2 inches from above downward
where deepest, near the external border. ‘The humeral surface is
slightly convex ; the distal surface is slightly concave from side to
side; it shows a short second facet for the middle carpal. The
ulnar surface is deeply concave from above downward.

The ulna is shorter from above downward on its radial margin,
where it measures about 1 inch, than on its external border, which
is convex, and in extreme depth from above downward measures
nearly 2 inches. The bone is about 17 inch through from side to
side, and 1? inch along its proximal articular surface. The distal
end ineluaes two apcular surfaces, of which the posterior is

-rather the smaller. The surface facing the radius is concave from
above downward, and appears to have been occupied by muscle
while the external margin, like the external margin of the radius,
was cartilaginous. There are six thick polygonal carpal bones pre-
served. They appear to have formed two rows of three each. The
longest is about 14 to 12 inch long by 1{ inch deep. The phalanges
are strong and thick, like those of Pliosawrus, and not compressed
from side to side, as in typical Plesiosaurs. Towards the distal
end of the limb they become short, and have the proximal and
distal ends of the bone greatly expanded relatively to its length.
There appear to have been five digits ; but the mlaaloee of phalanges
in each is not known.

EXPLANATION OF PLATE XXI.

Fig. 1. Right side of seventh cervical ver tebra of Murenosaurus Leedsii.

2. Front view of thirty-third cervical vertebra.

3. Left side of the thirty-ninth and neural arch of the thirty-eighth cer-
vical vertebre.

4. Front view of centrum of third caudal vertebra.

5

. Under surface of centrum of fifth caudal vertebra.

GH.Ford & C.L..Gmesbach.

; MURANOSA:

Quart. Journ.Geol.Soc Vol . XXX. Pl. XXI.

Mmtern Bros.1mp

fe
© LEEDSII.

:

Quart. Journ.Geol .Soc .Vol XXX. Pl XX].

P i Mintern Bros. imp .
GH Ford & C.1..Griesbach.

MURENOSAURUS LEEDSII.

ON THE UPPER COAL-FORMATION OF NOVA SCOTIA. 209

22. On the Urrmr Coat-rormation of Eastern Nova Scotia and
Privce-Epwarp Isnanp in its Retarrion to the Permian. By J. W.
Dawson, LL.D., F.R.S., F.G.S., McGill College, Montreal. (Read
March 25, 1874.)

Tuts formation was first distinguished as a separate member of the
Carboniterous system in Eastern Nova Scotia by the writer, in a
paper published in the first volume of the ‘ Journal of the Geological
Society,’ in 1845—and was defined to be an upper or overlying
series superimposed on the productive Coal-measures, and distin-
guished by the absence of thick coal-seams, by the prevalence of
red and grey sandstones and red shales, and by-a peculiar group of |
vegetable fossils.

Subsequently, in my paper on the South Joggins* and in my ‘ Aca-
dian Geology,’ this formation was identified with the upper series of
the Joggins section, Divisions 1 & 2 of Sir William Logan’s sectional
list, and with the Upper Barren Measures of the English Coal-fields
and the third or upper zone of Geinitz in the Coal-formation of
Saxony ft.

Still more recently, in a ‘ Report on the Geology of Prince Edward
Island,’ 1871, I have referred to the upper part of the same forma-
tion the lower series of sandstones in Prince-Edward Island, not
previously separated from the overlying Triast.

In Prince-Edward Island, however, where the highest beds of this
series occur, they become nearly horizontal, and are overlain appa-
rently in a conformable manner by the Red Sandstones of the Trias,
which differ very little from them in mineral character. It thus
happens that, but for the occurrence of some of the characteristic

Carboniferous plants in the Lower series and of a few equally cha-

racteristic Triassic forms in the Upper, it would be difficult to affirm
that we have to deal with two formations so different in age.

In connexion with this, the entire absence of the Permian system, __

not only here but throughout Eastern America, raises the question
which I have already suggested in ‘ Acadian Geology,’ whether the
conditions of the Upper Coal-formation may not have continued ~
longer here than in Europe, so that rocks in the former region con-
stituting an upward extension of the Carboniferous may synchronize
with part at least of the Permian. On the one hand, there seems
to be no stratigraphical break to separate these rocks from the
Middle Coal-formation of Nova Scotia; and their fossils are in the
main identical. On the other hand, where the beds are so slightly
inclined that the Trias seems conformable to the Carboniferous, no
very marked break is to be expected; and some of the fossils, as the
conifers of the genus Walchia, and Calamites gigas, have a decided
Permian tendency.
* Quart. Journ. Geol. Soe. vol. x. t Acadian Geology, p. 149.

{ ‘Report on the Geological Structure of Prince-Edward Island,’ by J. W.
Dawson, LL.D. &c., and B. I. Harrington, B.A., Ph.D.

210 J. W. DAWSON ON THE UPPER COAL-FORMATION OF

On the whole, in the ‘ Report’ above referred to, I declined to
separate the red beds of the Lower Series in Prince Edward Island
from the Newer Coal-formation. Prof. Geinitz, however, in noticing
my Report*, and also in a private letter, expresses the opinion that
the fossils have, as an assemblage, so much of a Permian (or Dyadic)
aspect that they may fairly be referred to that formation, more
particularly to its lower part, the Lower Rothliegende. Attaching,
as every one must, great weight to the judgment of Prof. Geinitz on

- such a point, [ have in recent visits to Nova Scotia reexamined the

more instructive sections of the Newer Coal-formation on the eastern
coast of that province, with the view of ascertaining whether any
stratigraphical or paleontological line can be found to divide the
Upper Coal-formation series of my former papers into two members
or to separate it from the Middle Coal-formation. The results of
this reexamination and their bearing on general geological questions
I propose to state shortly as follows :—

The Carboniferous district of Pictou county, extending for about
45 miles along the shores of Northumberland Strait, exposes in that
distance in coast- and river-sections the whole thickness of the Carbo-
niferous system, arranged in three synclinal forms (see Section, fig. 1).
The first or eastern synclinal (No. 1 in the Section), extending from
the older metamorphic rocks on the eastward and southward to a
line running nearly east and west through the town of New Glasgow,
consists entirely of the Lower Carboniferous, Millstone Grit, and
Middle Coal-formation, and contains all the known workable Coal-
measures of the county. Its northern boundary, the New-Glasgow
anticlinal, brings up a bed not recognized in the other Nova-Scotia
Coal-fields—the New-Glasgow Conglomerate, an immense mass,
believed in some parts to be 1600 feet in thickness+, and containing
boulders 3 feet in diameter, with pebbles of all sizes, many of its
largest stones being composed of the hard brown or purplish sand-
stones of the Lower Carboniferous. Its stratigraphical position is
that of the upper part of the Millstone Grit or lower part of the
Middle Coal-formation ; and it is evidently an exceptional bed, re-
presenting an immense bar or beach of gravel and stones stretching
from the eastern end of the metamorphic chain of the Cobequid
Mountains across the Pictou Coal-field, and protecting those deep
swamps in which the Pictou main coal, 36 feet thick, and its black
shale roof, more than 1000 feet thick, were deposited. The theory
of this remarkable deposit, one of the most singular connected with
any Coal-field, is fully discussed in the second edition of my ‘ Aca-
dian Geology.’ I may merely remark that, facing as this bed does
the open sea stretching to the northward in the Coal-formation
period, it is not unreasonable to suppose that it indicates the action
of heavy ice grounding on the shores behind which grew the Srgil-
larva-forests of the Coal-swamps. The arrangement of the beds in

* Neues Jahrbuch, 1872.

t This is Sir W. Logan’s estimate, and is warranted by the breadth which
the bed occupies in the Section; but there are indications that it thins rapidly
toward the dip.

211

EASTERN NOVA SCOTIA AND PRINCE-EDWARD ISLAND.

Fig. 1.—General Section of the Carboniferous rocks of Pictou,

New-Glasgow
conglomerate
and anticlinal.

Nova Scotia.

Carribou Cape-John
anticlinal. anticlinal.

‘
1
1
®
'
'
1
'
1
.
f

a, Silurian. 6b. Lower Carboniferous.

ce. Millstone Grit.
1. Middle Coal-formation. 2 & 3. Upper Coal-formation and Permo-Carboniferous.

QUA J. W. DAWSON ON THE UPPER COAL-FORMATION OF

the first synclinal, which is that of the great Pictou coal-beds, has
recently been worked out in much detail by Sir W. E. Logan and
the late Mr. E. Hartley.

The second or middle synclinal (No. 2 in the Section) extends
from New Glasgow to Carribou Harbour, and centres in the deep
indentation of Bighorn Harbour. On its sontihens side it contains,
north of New Glasgow, the depauperated equivalent of the Middle
Coal-formation ; and the remainder of it is occupied by the Newer
Coal-formation, whose newest beds, however, are not represented in
this trough. The low anticlinal which separates it from the third
trough brings up nothing older than the lower part of the Newer
Coal-formation.

The third synclinal (No. 3 in the Section) extends from Carribou
Harbour to Cape John, and, stretching westward through the Cum-
berland Coal-field, shows in its centre the newest beds of the Upper
Coal-formation.

It is to be observed that in these synclinals the north-west sides
have steeper dips than the south-east sides, and consequently occupy
aless breadth on the map. The south-east sides also show the best
and most continuous sections ; and for this reason I shall select the
section from New Glasgow to Pictou Harbour, and that from Carribou
Harbour towards Cape John, as typical of the lower and upper parts
of the Upper Coal-formation.

1. Section on the East River of Pictou.

1. On the river-section, below New-Glasgow bridge, the conglo-
merate is succeeded in ascending order by a grey concretionary
limestone 20 feet thick, associated with sandstone and shale, and
containing in some layers great numbers of the Spirorbis which I
have described as S. arietinus*, and whose habits of life were pro-
bably not dissimilar to those of S. carbonarius, so abundant in
the Coal-measures. This limestone does not appear in the immediate
river-section, but on the flank of the conglomerate east of New
Glasgow.

2. Above this is a series of black shales and underclays with
grey sandstones and some reddish and purple shales, and thin seams
of bituminous shale and coal. These beds contain Stigmarice, Lepi-
dodendra, Entomostracans and fish-remains, the fossils and the
mineral character of the beds alike corresponding with those seen in
the upper part of the Coal-measures south of the conglomerate.
The thickness of these beds is about 400 feet.

3. This series is succeeded by a thick grey sandstone holding
Calamites, Calamodendron, trunks with aerial roots (Psaronius), &c.,
30 to 50 feet thick. This appears at the mouth of Smelt Brook and
in several quarries to the eastward of that place.

4. Above this is a second series of dark shales and under-clays,

* «Report of Geol. Survey of Canada.’ This limestone may be compared
with the ‘“ Sprorbis-limestone” of the Shrewsbury, Lancashire, and Warwick-
shire Coal- fields i in England. See Hull ‘ Coal-fields of Great Britain’

EASTERN NOVA SCOTIA AND PRINCE-EDWARD ISLAND. Pas

and bituminous shales associated with grey sand-stones and contain-
ing fossils similar to those of the series below. It especially abounds
in fish-secales and Cythere; and several of the fishes are specifically
identical with those of the upper part of the Middle Coal-measures
as seen in the southern trough, south of New Glasgow. These beds
are about 200 feet thick. Mr. H. Poole has described them in the
‘Canadian Naturalist’ for August 1860.

5. The beds up to this point may be considered the equivalents
of the Middle Coal-measures or of the upper part of them, and are
now succeeded in ascending order by thick grey and reddish sand-
stones and reddish and grey shales, including, however, thin coaly bed
and underclays, and clays with nodular limestone. These may be
regarded as belonging to the Upper Coal-formation ; and their aggre-
gate thickness as far as Pictou Harbour may be 2000 feet. They
contain Calamites, trunks of Dadowylon materiarium, Lepidodendron,
Pecopteris arborescens? and Neuropteris.

The dip of the Conglomerate is high; and that this is not alto-
gether due to false stratification is shown by the fact that to the
eastward of New Glasgow the limestone and the Coal-measure beds
rest on the Conglomerate at an angle of 45°; but this rapidly dimi-
nishes to 20°, and in the greater part of the section it is only from
8° to 6°.

The line of demarcation between the Middle and Upper Coal-
formations is not marked here by any great physical break, but merely
by the cessation of the characteristic beds of the Middle Coal-forma-
tion and the change to sandstones associated with red shales.

At first sight 1t might appear that as the beds north of the Con-
glomerate dip uniformly to the north, and mostly at slight angles,
and those south of its outcrop are much more disturbed, there might
be evidence of unconformability. This, however, is due to a line of
fault extending along the outcrop of the Conglomerate, and to the
greater relative disturbance of the beds of the southern synclinal.

2. Section west of Carribou Harbour.

This Section exposes the south side of the third or northern syn-
clinal, and may be supposed to begin not far above the base of the
Upper Coal-formation. It extends in ascending order obliquely across
the synclinal for about ten miles along a coast in which the beds are
on the whole well exposed, with uniform dips of about N. 30° E.
magnetic, or nearly true north, and at an angle of about 10°; and
no break or evidence of unconformability exists throughout the
series, which amounts here in thickness to about 2500 feet.

The lowest beds seen in this section at the mouth of Carribou
River are red and grey shales, and grey, red, and brown sandstones,
including a small bed of coal 5 inches thick, with Stigmaria-rootlets
in the underclay; and at Carribou Island, nearly in the line of strike,
there is a somewhat thicker bed of coal. The overlying series may
be described as consisting of indefinite alternations of shales, mostly
deep red, with sandstones, grey, red, and brown, the latter sometimes

Q.J.G.S8. No. 119. R

214 J. W. DAWSON ON THE UPPER COAL-FORMATION OF

coarse and pebbly, and occasionally in thick massive beds. Several
of the beds of shale contain concretions of limestone, in one case
forming a nearly continuous bed, and with no fossils except a few
casts of a Cythere. In one of the lower beds of sandstone seen on
Carribou River there are concretions of grey copper, and fossil trunks
of trees penetrated by this mineral; and some of the fossil trees
found in the sandstones on the coast are partly mineralized with
sulphate of baryta.

The only material difference in mineral character is that red beds
become more prevalent toward the upper part of the section, where
the general character of the beds is precisely that of the supposed
Upper Coal-formation rocks at Miminigash, Governor’s Island, and
Gallas Point in Prince-Edward Island, and on the coast of New
Brunswick at Cape Jourimain*.

The following statements, reduced from my sectional lists, will
serve to illustrate these points of mineral character.

In the whole section the sandstones, including the argillaceous
sandstones, are to the shales in the proportion of about two to one
in vertical thickness, and the grey and buff sandstones are about
equal to those which are brown and red, while the red and mottled
shales greatly preponderate over those which are grey.

In the lower half of the section, extending to the mouth of Toney
River, the grey sandstone, red sandstone, and shales (mostly redyare
in the proportions of 43, 3,63. In the upper half of the section they
are in the proportions of 43, 53, 3; so that red sandstones become
decidedly more prevalent in the upper part, where there is also a
greater proportion of coarse pebbly sandstones and of light-red shale
with greenish stains.

If we compare this with the upper part of the Joggins section as
given in Sir William Logan’s lists, we find a thickness of 2267 feet ;
and if we regard the Ragged-Reef Sandstones as equivalent to the
heavy sandstones at the base of the Pictou section, it is possible
that the upper part of the latter is not represented at the Joggins.
Taking the proportions of sandstones and shales at the latter place,
we find them to be grey sandstone 12, red and brown sandstone 1,
shale 10; so that here the proportions of sandstones to shales are
not very dissimilar to those in the lower part of the Pictou series,
but the grey sandstones are greatly more prevalent. Like those in
the upper part at Pictou, some of the upper beds at the Joggins are
coarse and pebbly, a character not observed, in either Coal-field, in
the sandstones of the Middle Coal-formation.

If, on the other hand, we turn to Prince-Edward Island, the
geological relations, and especially the fact that the outcrops on
Prince-Edward Island correspond with the extension of two of the
New-Brunswick Carboniferous anticlinals, would lead us to believe
that the upper Coal-formation beds seen, at Gallas Point, and
amounting to about 800 feet in vertical thickness, must belong to
the upper part of the Pictou series, or may even reach some way
above its summit. Accordingly we find the proportions of the
: * Report on Prince-Edward Island.

EASTERN NOVA SCOTIA AND PRINCE-EDWARD ISLAND. 215

several rocks to be grey sandstone 2, red and brown sandstone 4,
shales 2, or a still greater proportion of red sandstone as compared
with Pictou. All this accords with the idea of a gradual increase
of red beds in approaching the summit of the formation, so that the
upper Coal-formation passes in its upper part into beds haying more
the aspect of some parts of the Lower Dyas or Permian. No true
dolomite is present in these beds; but Dr. Harrington’s analyses
show that some of the thin beds of concretionary limestone are
highly magnesian, and the sandstones contain concretions of sul-
phate of copper, while the fossil trees which abound in them are
often mineralized with sulphates of copper and iron, and sulphate
of baryta.

Fossils of the Upper Coal-formation.

Fossils are by no means so abundant in the Upper as in the
Middle Coal-formation, and they are chiefly vegetable. One of the
most characteristic plants is Dadoxylon materiarium, a species with
simple medullary rays, drifted trunks of which abound in a calcified
or silicified condition in the sandstones. The fine specimens of
the Sternbergia pith of this species which I described in 1857 *
and 1871+ are from this formation. In the upper beds leafy
branches of the genus Walchia are common fossils, probably be-
longing to trees of the genus Dadowxylon, the only pines which accom-
pany them. Calamites are also abundant, especially C. Suckovia
and OC. Castiz; and Calamodendron approximatum is not uncommon,
while Calamites gigas occurs rarely in the upper part. Annularia
sphenophylloides is a characteristic plant in the lower part, and Cor-
daites simplex is very abundant in some beds. Lepidodendra are rare,
and represented principally by a species which is identical with, or
very near to L. pictoense. Among ferns the most abundant species
are Pecopteris arborescens and a variety of Alethopteris nervosa.
Stigmarice and Sigillarie are much less frequent even in the lower
part than in the Middle Coal-formation, and have not yet been re-
cognized in the upper part.

The following tabular view may serve as a summary of the flora
of the Upper Coal-formation as at present known. ‘The first two
columns represent the upper and lower parts of the Upper Coal-
formation in Nova Scotia; and the third column represents that of
Prince-Edward Island. Of the species all but about ten, or more
than three fourths, have been found in the Middle Coal-formation
also. It will be observed that the number of species, which in all
is much smaller than that in the Middle Coal-formation, becomes
rapidly reduced in the upper part, and that there is a considerable
similarity between the upper series in Nova Scotia and that in
Prince-Edward Island. This is further noticeable in the great
prevalence of specimens of Dadoaylon materiarium, Walchia, Cor-
daites simplex, and Pecopteris arborescens in this part of the forma-
tion in both districts.

* Proc. Amer. Association, 1857, Canad. Nat. vol. ii.
+ Report on Prince-Edward Island.
R2

216 J. W. DAWSON ON THE UPPER COAL-FORMATION OF

Upper Coal-formation.

: Nova Scotia. :
Species. Prince-

Edward
Lower | Upper Tena

part. part.

. Dadoxylon materiarium, Daws. ........ ... * *

. Walchia (Araucarites) gracilis, Daws.......] ...... *
2 )) THOUS), JOH cbcvccacascnosecee|| oeeane |)" oacaae
. Sigillaria scutellata, Brongn. ............065

1
2
3
4
5. Stigmaria ficoides, Brongn.............1064..
6
7
8
9

kK

*
%
. Calamites Suckovii, Brongn. ............64. * x
HEI, JEVROWGMao0coccc000000009D000R00000% *
9 ERS, JEIROGT0o socooooccnoanq00000040ndoo} coac0s |)“ ‘cosenc
ENEMYS WIS TY OMAP oonaccdoacocoudebabosasdl, | coddda. vl sosnde
10. Calamodendron approximatum, Brongn.
11. Annularia sphenophylloides, Zenker ......
longifolia, Brongn. .........100...0.006
13. Sphenophyllum emarginatum, Brongn....
: longifolium, Geinitz ............0.005-
15. Cyclopteris oblongifolia? Gopp.............
heterophylla, Gopp. 2 )............-
TWIN OMAN), LES saacooodoeosnoenadeds05
18, Neuropteris flexuosa, Brongi................
Cordata, BONG sence sooceoatee se
heterophylla, Brongn...............0.4+
rarinervis, Bunbury .......0.0.00.00--
auriculata, Brongn. ..........-s0se.-+0--
. angustifolia ?, Brongn. ...........005.
24. Odontopteris Schlotheimii, Brongn. ......
25. Sphenopteris latior, Daws. ...............605
PINE he, JEVOAG Os, “ocosogoseedocdo.0ee.oe0 anc
27. Alethopteris nervosa, Brongn..............+.
Serle eBrongmereneece- cease Sieaecdaees
BXSWUIE), JEVAOMG Pb ocano0acc00%00000 Rebate ee
30. Pecopteris arborescens, Brongn. .........+..
abbreviata, Brongi.........0.---sess0+0-
32) —— unite, Brongn, .cicecsaccscecsccsesseeens *
3d. ——-vigida, Daws.” .0.....6........2cceceeses|| oe cee * *
, oreopteroides, Brongn. ...........0... ea * *
30. Bucklandi?, Brongn., or ? Massi-| ©. :
Mons; Ses asap. sesicedto desler Materaceen stalk traraticmtine Mlle hart ae *
| 36. Beinertia Geepperti, Daws. ....,........0.... *
37. Paleopteris acadica, Daws. .................. *
38. Cordaites simplex, Daws. .............ec0e0e *
39. Lepidodendron pictoense, Daws............. *
undulatuml: Ste7;70a ce seensecceseeee *
*
*
*
*

*
aK % OK OK

OK OK OK OK Ok

*

* OK OK KK X

2K OK XK OK CK OK OK OK

41, Lepidophloios parvus, Daws. ...............
42. Lepidophyllum, (various sp.) ...............
415}, « PivaaLWAVENE) 92 aaedbadeseocbsnaddacnobaddoKodoodedun
44, Trigonocarpum Neggerathii, Brongn. ...
8 S1Ok" Chore eRe Taine Oe Eatin encararaoeeaeccel| t uadacky fuel. can aads *
AG. Rind os IMSIPNIS, DAws........0.000-.. *
47. Antholithes squamosus, Daws. ............ *

There is unfortunately no recognized Permian in Eastern America
wherewith to compare the fossils of the upper member of the Newer
Coal-formation ; but inasmuch as the Coal-formation of Nova

EASTERN NOVA SCOTIA AND PRINCE-EDWARD ISLAND. 217

Scotia is, as I have elsewhere shown, more nearly allied in its
fossils to that of Europe than to that of the interior of North
America; and as the Permian flora consists to a great extent of
survivors from the Coal-formation, it will not be unfair to compare
the above list with the species in Geinitz’s and Goppert’s Memoirs
on the European Permian.

The very abundant Dadowylon matercarium is a tree of the same
type with several species found in the European Permian, as for
instance D. saxonicum, Reich., and D. Schrollianum, Gopp. Wal-
chia is also regarded as characteristic of the European Dyas; but
as it is not improbable that it represents merely leafy branches
of Dadoxylon, it belongs to the Carboniferous as well. One of our
species, however, is very near to W. piniformis of the Dyas.
Calamites arenaceus, whether or not an internal axis of Hquisetites,
is Dyadic in Europe; and some of my specimens may well belong to
C. letoderma of the European Permian. C. gigas is a decidedly
and peculiarly Permian species. C. Suckovii and C. Cistiw are
Permian as well as Carboniferous in Europe, as is also Calamoden-
dron approximatum. Annularia longifolia is Permian as well as
Carboniterous. Newropteris rarinervis is peculiarly American and
very widely distributed ; but it is questionable if some of its larger-
leaved varieties are not identical with European forms known by
other names. Veuwropteris flecuosa, N. cordata, and N. auriculata,
as well as Pecopteris (Cyatheites) arborescens, P. oreopteroides, and.
P. abbreviata are both Carboniferous and Permian; and the species
which I have compared doubtfully with P. Bucklandi, and with P.
Massilionis of Lesquereux, has strong points of affinity with P.
densifolius of Goppert. Cordaites simplex is a peculiar American
species, but nearly allied, according to Geinitz, to his C. Resslerianus
from the Lower Dyas. Finally Geinitz thinks the 7rigonocarpum
from Prince-Edward Island to be the same with his Rhabdocarpus
dyadicus.

We thus have an undoubted paleontological resemblance between
the upward extension of the Carboniferous in Nova Scotia and
Prince-Edward Island and the Permian of Europe, though in the
former regions no stratigraphical break enables us to establish on
that ground any well-marked line of division. Taking into con-
sideration the great thickness of the Carboniferous in Nova Scotia
and the large development of this Upper Permo-Carbo+iferous
member, it would not be surprising that in this last we may have
a chronological equivalent of part at least of the European Permian.

We have no evidence as to age derivable from marine shells.
The highest marine limestone known to me, a bed near Wallace
Harbour, which I described many years ago in the Journal of this
Society *, belongs to the base of the Newer Coal-formation, and
contains Productus cora, P. semireticulatus, and Aviculopecten
simplea, all characteristic Lower Carboniferous forms.

In Prince-Edward Island the Upper Carboniferous and the Trias
are appparently conformable, and may almost be said to pass into

* See also ‘Acadian Geology,’ p. 214, 2nd edition.

218 J. W. DAWSON ON THE UPPER COAL-FORMATION OF

each other, though in Nova Scotia the Trias rests unconformably
on the Carboniferous. I believe, however, that this apparent
conformity in Prince-Kdward Island, and the resemblance of the
two series in mineral characters, arises from the almost horizontal
position of the Carboniferous beds, and from the circumstance
that the Trias has been in part formed from their waste. The
Triassic fossils, though few, are of species quite distinct from those of
the Carboniferous. Further details as to the relations of these
formations in Prince-Edward Island will be found in my Report on
that island.

To sum up, it may be said that the beds which overlie the Coal-
field of Pictou and extend into Prince-Edward Island, and which
constitute the upper part of the Upper Coal-formation, have such
strong points of resemblance to the lower part of the European
Permian, both in their mineral character and organic remains, that
they may fairly be named Permo-Carboniferous, a name already
applied to certain marine limestones in the West, in which the
Carboniferous graduates upward into the Permian. They may
also be held to some extent to bridge over the gap which in Kastern
America separates the Carboniferous fromm the Trias.

I may add that in Nova Scotia the Lower Carboniferous beds are
usually more hardened and altered than those of the Middle Coal-
formation, and the latter more than those of the Upper Coal-forma-
tion. Moreover there are instances in Nova Scotia of local uncon-
formability of the Lower Carboniferous beds; and the New-Glasgow
conglomerate affords evidence of extensive denudation of the Lower
Carboniferous before the deposition of the productive Coal-measures.
These facts indicate the long duration of the Carboniferous period
and the extent of the physical changes which it included; and it is
evident that, had unconformability or extensive local denudation
occurred somewhat higher in the system, it might have been
regarded as forming the base of an overlying Permian series.

I have discussed somewhat fully the relations of the flora of the
Lower Carboniferous to those of the Devonian on the one hand, and
of the Upper Members of the Carboniferous on the other, in a ‘ Report
on the Fossil Plants of the Lower Carboniferous and Millstone Grit,’
recently published by the Geological Survey of Canada*. I hope
that I may be able at some future time to describe and illustrate
fully the plants of the Upper Coal-formation in the same manner.

Discussion.

Prof. Ramsay agreed with the author in thinking that these
Upper Carboniferous rocks represented the Permian, and that there
is a gradual passage from the Carboniferous to the Permian. In
North Staffordshire there is some evidence of this passage, but not
in other parts of England. Mr. Binney had argued that the Per-
mian is the uppermost part of the Carboniferous series; but this is
not true in the English area, although it is true if we consider the

* Montreal, 1873.

EASTERN NOVA SCOTIA AND PRINCE-EDWARD ISLAND. 219

globe in general. The Coal-measures are grey, black, and blue; but
in the upper portion they sometimes change to a red tint. During
the Coal-period we have evidence of estuarine conditions ; but subse-
quently the access of the sea was cut off, and the Permian rocks were
formed in vast inland lakes.

Prof. Huenes remarked that the group referred to by Principal
Dawson under the head of Permo-Carboniferous could not be con-
sidered as in any way proving a passage from Carboniferous to Per-
mian, seeing that the Permian was altogether wanting in Eastern
America, unless the fossils approached those of undoubted Permian
in Europe. But he pointed out that many large portions of the so-
called Permian of Europe had been already proved to be only stained
Carboniferous. The fossil lists were founded on a wrong classifi-
cation of the rocks, which had not yet been set right. Believing,
therefore, that the Permian system must be broken up and part
given back to the Lower New Red and Magnesian Limestone series,
previously so well established, and part to the Upper Carboniferous,
he was inclined to refer the Permo-Carboniferous of Principal Daw-
son to the latter, the difference in the plants being only such as
might reasonably be expected between the newer and older portions
of a series representing immense lapse of time and changing condi-
tions. Principal Dawson had shown that the beds in question were
similar in almost all but colour, and conformable to the underlying
undoubted Carboniferous. If, therefore, they were higher than any
Carboniferous beds of England, they must be synchronous with the
lower part of the unrepresented time between the Carboniferous and
so-called Permian ; but being more closely connectod with the lower
rocks, he saw no necessity in the present state of our knowledge for
such a term as Permo-Carboniferous.

Prof. Ramsay could not agree with Prof. Hughes in his opinion
as to the value of the term Permian. ‘The staining of rocks occurs
in two ways—namely, by infiltration from above through overlying
beds, and by direct deposition. Silurian rocks are often stained in
the former manner.

Mr. Evans remarked that this paper had given rise to an inter-
esting discussion. The fact of the two deposits being conformable
in one place and unconformable in another, did not, in his opinion,
necessarily convert them into one system. He thought there were
symptoms that the Permian would eventually be regarded as Upper
Carboniferous. He believed that there was a third mode in which
rocks were stained—namely, by the oxidation of iron already existing
in the beds.

220 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

23. The Sxnconpary Rocks of Scortanp. Second Paper*. On the
Ancient Voicanors of the Hicuianps and the Retations of their
Propucts to the Mesozorc Strata. By Joun W. Jupp, Esq., F.G.8.
(Read January 21, 1874.)

[Puates XXII. & XXITT.]

ConTENTS.
1. Introduction.

1. History of Previous Opinion on the subject.

2. Voleanic Origin of the rocks constituting the great plateaux of the
Hebrides and the North of Ireland.

3. Subaerial Origin of these old Volcanic rocks.

4, Evidences of the Former Existence of great Volcanic mountains
in the district.

II. The Tertiary Volcanoes.

. Classification of the Tertiary Volcanic rocks.

. Nature and origin of the great Volcanic rock-masses :—Lavas, Intru-

sive masses, Volcanic agglomerates and Volcanic breccias.

. Relations of the Volcanic rocks to one another and to the older

deposits in the island of Mull.

. Sections illustrating the structure of the island of Mull;—Beinn

Greig, Beinn Uaig, Craig Craggen, Beinn More.

. Proofs that the central mountain-group of Mull constitutes the relic

of a great volcano.

The Voleano of Ardnamurchan.

. The Volcano of Rum.

. The Voleano of Skye.

The Volcano of St. Kilda.

. Comparison of the great Tertiary Volcanoes.

. Dimensions of the great Tertiary Volcanoes.

. Series of later Volcanic eruptions in the Hebrides, resulting in the
formation of “ Puys.” :

13. Subterranean Phenomena of the Tertiary Volcanoes.

14. Ages of the several Volcanic outbursts already described.

_—
SMI oo FP OF we

=
SS

b

* In the first published paper of this series (vide Quart. Journ. Geol. Sce.
vol. xxix. p. 97) 1 found it possible, in a single communication, not only to
discuss the nature and origin of the remarkable physical relations of the vestiges
of the Secondary strata on the east coast of Scotland, but also to reconstruct
from them the history of the several Mesozoic periods as exemplified in that
district. This, however, was only accomplished by extending the paper to a
somewhat unusual length; and in dealing with the strata of the same age on
the western coast, I have found it impossible, such is the complication of the
questions involved in their study, to deal with both branches of my subject in
a single paper. Consequently I have confined myself, in the present commu-
nication, to a description of the positicns and relations of the fragments of
Secondary strata, and a discussion of the causes to which these are due. In
a third paper, which is already in an advanced stage of preparation, I propose
to illustrate the succession of geological events in the Western Highlands
during the Mesozoic periods ; while in a fourth paper I anticipate being able
to conclude the account of my studies of these rocks by an endeavour to deal
with those problems of ancient physical geography and general paleontology
for the solution of which these remarkable fragments of Secondary strata sup-
ply such valuable materials.

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 221

15. Connexion between the Tertiary Volcanoes of the Hebrides and those
of other districts.

16. General conclusions from the relations of the Volcanic and Plutonic
rocks of the Tertiary period.

III. The Newer-Palzozoic Volcanoes.

1. Lavas of Lorn and the adjacent islands.

2. Characters of the Voleanic rocks of Lorn.

3. Relations of the Volcanic rocks of Lorn.

4. Succession of rocks in Lorn.

5. Conditions under which the Volcanic series of Lorn was deposited.
6. Age of the Volcanic series of Lorn.

7. The Newer-Palxozoic lavas of the Lowlands of Scotland.

8. The Eruptive masses of the Grampian mountains.

9. Relations of the igneous rocks of Beinn Nevis and Glencoe.
10. Physical Features of Northern Scotland during the Newer-Palezozoic

periods.

TV. Conclusion.

1. Comparison of the two great periods of Volcanic activity in Scotland.
2. Influence of Volcanic action in determining the Characters and Rela-
tions of the Secondary rocks of Scotland.
3. The ‘ Geological Record ” in the Highlands.
4. Light thrown upon some problems of Physical Geology by the
Volcanic rocks of the Highlands.

I. Introduction.

For the preservation of the most valuable illustrations of the in-
stitutions, manners, and arts of Ancient Rome, the archxologist is
indebted to the action of a volcano: the relics of Pompeii have
survived in consequence of being buried under the ejections of
Vesuvius. To a similar agency, operating at a distant epoch and
on afar grander scale, the geologist owes the escape from destruction,
in the Western Isles of Scotland, of most wonderful monuments of
physical change and highly interesting records of life-history during
the Secondary periods ; for such, indeed, are those remarkably pre-
served fragments of sedimentary rocks which it is the object of this
memoir to describe.

As he prosecutes an examination and comparison of all the cir-
cumstances under which the scattered relics of the Secondary
formations present themselves in the West of Scotland, the geologist
will be again and again impressed by the extent of the protective
influence which the vast masses of Tertiary lava have evidently
exerted upon the subjacent stratified rocks. And when he has
concluded that survey, he can scarcely have failed to arrive at the
conclusion that, but for this protective influence, every vestige of
the Mesozoic deposits in the district must have been inevitably
swept away by denudation *.

* Those familiar with the geology of Central France will at once recall the
manner in which the sheets of basaltic lava capping the great plateaux have, in
se many cases, secured the preservation of masses of the lacustrine strata on
which they rest, every trace of which must otherwise have been swept away by
denuding forces. (Vide Scrope’s ‘Geology and Extinct Voleanos of Central
France,’ p. 7 &c.) :

222 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

When, therefore, we reflect upon the remarkable combination of
circumstances to which we are indebted for the preservation of these
interesting records of the whole series of Secondary formations (from
the Trias to the Upper Chalk inclusive) we cannot fail to be im-
pressed by the accidental (and often, indeed, exceptional) nature of
the conditions upon which the escape from destruction of fossili-
ferous deposits has in so many cases depended. It would be difficult
to adduce a more striking illustration of the necessarily great imper-
fection of the geological record than that which is suggested by these
strangely preserved fragments, of what were evidently once widely
spread formations representing geological periods of vast duration.

It is impossible rightly to understand the features presented by the
Secondary rocks in the Western Highlands without carefully studying,
in the first place, their relations to the great masses of igneous rocks
among which they lie. These relations are of the most intimate
and often complicated character. Not only have the fragments of
Mesozoic strata which had escaped denudation at the commence-
ment of the Tertiary period, been buried under vast accumulations
of lava sheets to the depth of hundreds and even thousands of
feet, but they are often, as I shall show hereafter, penetrated by
igneous masses connected with three distinct periods of volcanic
activity, from the influence of which they exhibit every conceivable
stage of metamorphism ; and, further, their fragments are found,
often in great abundance, imbedded in the vast masses of scorize
and ashes which have been ejected from the volcanic vents. In
order, therefore, to reconstruct the history of the Mesozoic period,
it is necessary to carefully restore and reunite all these scattered
fragments of evidence; for the same volcanic agency that has so
wonderfully preserved the records, has at the same time unfortu-
nately, in too many instances, sadly mutilated and defaced them.

I shall show, moreover, that although during nearly the whole
of the Secondary periods the voleanic forces were dormant in the
district, yet that era was preceded, as well as followed, by an epoch
of the most intense and prolonged igneous activity. The influences
of this earlier period of volcanic action in determining the characters
of the Secondary deposits, although less marked than those of the
succeeding eruptions of the Tertiary period, are nevertheless clearly
traceable. Further, many of the peculiarities and anomalies pre-
sented by the Secondary rocks in this district appear to find an
adequate explanation in the circumstance that they were deposited
in the interval between these two periods of violent igneous activity,
and that the areas which they occupy may therefore naturally be sup-
posed to have been subject to frequent and excessive disturbance.

But while many points of great importance with respect to the
Secondary rocks are thus dependent for their elucidation on a care-
ful study of the volcanic products with which they are so intimately
associated, much new light is at the same time thrown upon the
nature, age, and history of the latter by an examination of the
relations which subsist between them and the interesting fragments
of fossiliferous, and thereby dated, rocks of the Mesozoic periods.

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 223

In this manner we are led to many very interesting conclusions
with regard to the chronology of the various rocks of the Scottish
Highlands.

Moreover, as I hope to be able to show, the prosecution of this
research concerning the relations between these sedimentary and
eruptive rocks, is calculated to throw new light upon some of the
obscurest problems of physical geology.

Under these circumstances, I have considered it advisable to con-
fine myself in the present communication to this question of the
mutual relations between the Secondary and Volcanic rocks of the
west coast of Scotland—a question of much complication but at the
same time of the highest interest—reserving for a future occasion
the details of the history of the Mesozoic periods in the district, as
deduced from the paleeontological and physical evidence.

1. History of Previous Opinion on the subject—The very inti-
mate manner in which the Secondary and Voleanic rocks of the
Hebrides are associated with one another not unnaturally led the
earlier geological observers to regard them as being of contempora-
neous age. ‘This opinion received its first shock in 1851, through
the discovery by the Duke of Argyll of the leaf-beds of Ardtun, and
the determination by Professor Edward Forbes of the Miocene age
of the fossil plants contained in these deposits*. It then became
evident that a part at least of the Volcanic rocks of the Hebrides
belongs to the Tertiary period. In 1865 Professor A. Geikie had
arrived at the important conclusion that the vast sheets of igneous
rock lying between the Secondary strata in the Hebrides are in
every case intrusive, and therefore not contemporaneous with those
rocks ; and on a review of all the facts of the case, he was led to an-
nounce his conviction that the whole of the volcanic rocks under
consideration belong to the Tertiary period +. This conclusion I have
been able to confirm by showing that the volcanic rocks in question
unconformably overlie even the youngest members of the Chalk.

The unmistakably volcani¢ origin of the so-called “trap rocks ” of
the Western Isles had been noticed by many observers ; and some of
these, especially Dr. Maccullocht¢ and Professor Geikie§, have dwelt
upon the evidently close relations between these and the Plutonic
rocks of the district. The Duke of Argyll has remarked on the
manner in which the two classes of igneous rock, as seen in a section
in Mull, appear to graduate into one another—and also on the indi-

* Quart. Journ. Geol. Soc. vol. vil. pp. 89, 103.

t Proc. Roy. Soc. Edinb. vol. vi. (1866-67), p. 72, and Quart. Journ. Geol.
Soe. (1869). vol. xxvii. p. 283. I gladly take the present opportunity of bearing
witness to the great value of Professor Geikie’s researches among the volcanic
rocks of Scotland. Although the conclusions at which I have arrived are, in
many cases, very different from the opinions which he has expressed on some of
the phenomena of this interesting district, I think that in almost every case it will
be found that the points in which he differs from myself are those in which he
has put forward useful suggestions of a tentative character rather than the
results of direct observation. In almost every case of the latter kind, I am
happy to be able to confirm his great accuracy and acumen.,

{ ‘A Description of the Western Isles of Scotland,’ 1819.
§ Quart. Journ. Geol. Soe. vol. xxvii. p. 282.

224 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

cations of the volcanic character of the mountain of Beinn More, in
the same island, afforded by the highly scoriaceous character of its
materials*.

Although many very valuable geological observations have been
placed on record by Dr. Jameson, Boué, Dr. Macculloch, Principal
J. D. Forbes, and Prof. A. Geikie, but little has been hitherto done
in the systematic examination of the relations of these old voleanic
rocks. The characters of the minerals and rocks themselves, have,
however, been much more successfully investigated by several of
these authors. Dr. Macculloch’s mineralogical knowledge was so
large and accurate that the lapse of more than fifty years has failed
to deprive his descriptions of the rocks of this district of their
interest and value ; and very recently one of the greatest masters of
the methods of petrological research, Prof. Zirkel of Leipsic, has
supplemented these early observations by a series of careful re-
examinations of the same rocks, to the results of which I shall
have occasion to refer more particularly hereafter.

2. Volcanie Origin of the rocks constituting the great plateaus of
the Hebrides and the North of Ireland.—The rocks which constitute
such extensive plateaux, both in Ulster and the Hebrides, have
long excited attention and interest in consequence of the remarkably
picturesque forms which are assumed by them at certain points
where the columnar structure is finely developed; this is especially
the case with those justly celebrated localities the Giant’s Cause-
way and the isle of Staffa. That the rocks which present these
remarkable characters are of voleanic origin, and indeed constitute
the remains of great lava-streams, is a fact which was clearly re-
cognized by some even.of the earliest geological observers +; and
the more minutely and carefully the phenomena presented by the
rocks connected with active or recently extinct volcanoes have been
examined, the more strikingly has the soundness of this conclusion
been made apparent.

At the present day the volcanic origin of these rocks may be
regarded as so far an established portion of geological science as to
render quite superfluous on this occasion any details of the grounds
on which it rests. Whether we regard the chemical composition of
the different varieties of rock, or their mineralogical constitution
(especially as this is revealed to us by the microscope), or the pecu-
liarities of their petrographic structure, such as the remarkable
vesicular and columnar features which they exhibit,—we are alike
struck by the perfect identity of characters between them and the
materials of recent lava-streams. Innumerable minor features serve
to confirm this conclusion—such as, among others, the highly vesi-
cular or scoriaceous character of the upper and under surfaces of
the great masses, the inclusion between them of layers of scoriz,

* Brit. Assoc. Report (1867), Trans. of Sections, p. 55, and Address to Geol.
Soc. 1873, Quart. Journ. Geol. Soe. vol. xxix. p. lxxv.

t Vide Sir Joseph Banks, in Pennant’s ‘Voyage to the Hebrides,’ p. 267;
A. Mills, in Phil. Trans. for 1790, pp. 73-100; Macculloch, Syst. of Geol.
(1831), vol. ii. p.114, &e.

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 225

lapilli and ashes, or the vestiges of ancient soils and vegetation, and
the indications exhibited by the surfaces on which these rock masses
he of having been subjected to the action of heat.

One of the most striking points of similarity between these old
lavas and those seen to be actually connected with existing vol-
canoes has, through a very prevalent misinterpretation of the ap-
pearances presented, been generally overlooked. I refer to that re-
markable peculiarity, connected with the columnar structure, which
is nowhere better exhibited than in the beautiful caves of Staffa,
and which has been so clearly described by Mr. Scrops as giving
rise to such conspicuous features at Pont Gibaud, the Coiron, La
Gravenne de Souillols, Jaujac and other points in Central France*.
In all these cases the same lava stream is found to be composed of
two portions, which at a short distance appear to be very distinctly
separated from one another. The lower of these divisions, which
usually occupies about one third of the thickness of the lava stream,
is composed of very regular, upright, and generally jointed columns,
the articulations of which often exhibit remarkable curved surfaces
and angular processes. The upper part of the stream, however, pre-
sents strikingly different characters, being made up either of nearly
amorphous basalt or of thickly clustered columns of small diameter,
these being usually curved and twisted in the most remarkable
manner. All who have visited Staffa will at once call to mind the
contrast presented by the thick upright pillars which form the
«Colonnade ” and the sides of Fingal’s Cave, and the thin, grace-
fully curved, and intricately interwoven shafts which form the
Buchaille, the Clam-shell Cave, and the roof of Fingal’s Cave. <A
careful examination will convince the geologist that the two varie-
ties of columnar basalt form parts of the same lava stream, and are
not, as is usually stated in guide-books and geological manuals,
the product of two distinct and superposed flows. In the beautiful
basaltic columns of the Giant’s Causeway, in those of Carsaig,
Ulva, and many less known, and more inaccessible localities, I have
been able to verify the correctness of this observation.

Mr. Scrope, in an incidental allusion to the phenomena presented
by Staffa +, shows that the very striking identity of its features with
those so clearly described by him as occurring in Auvergne, had not
escaped his observation. He has suggested that the sharp distinc-
tion in characters between the two portions of the same lava
stream, is due to the different conditions under which they have
parted with their heat—that of the upper portion having escaped
by radiation, and that of the lower portion by conduction through
the subjacent rocks.

3. Subaerial Origin of these old Volcanic rocks.—Accepting, then, as
amply demonstrated, the conclusion that in these rocks of the North
of Ireland and the Hebrides we see the vestiges of extensive lava
streams, the next problem which presents itself to the geologist is

* Vide ‘The Geology and Extinct Volcanos of Central France,’ 2nd edit.
(1858), pp. 57, 163, 191, &c.

t ‘Volcanos,’ 2nd ed. (1872) p. 99.

226 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

the following :—Were these lavas poured out upon a terrestrial sur-
face? or were they the product of a series of submarine eruptions ?

On this question very diverse opinions have been maintained by
different authors, the majority, however, being in favour of the sub-
marine origin of the rocks. This opinion has, in a great degree,
arisen from the long prevalent idea that the lavas in question were
contemporaneous with the Secondary strata with which they are
often so intimately associated. We have already pointed out how
this opinion has been gradually dissipated—tirst by the discovery
of the Ardtun fossils by the Duke of Argyll, and secondly through
the recognition by Professor Geikie of the intrusive character, and
therefore subsequent age, of the sheets of igneous rock which lie in
the midst of the Mesozoic strata. With the supposed evidences of
the Secondary age of these lavas the grounds on which their sub-
marine origin was maintained have also disappeared ; and from the
overwhelming and irresistible mass of evidence which I am now
able to adduce upon the subject, it will, I think, be accepted as
conclusively demonstrated that the lavas were unquestionably of
subaerial or terrestrial origin.

The first point to which it is necessary to allude, is the total
absence of marine sediments and fossils, of contemporary age,
interstratified with the great lava sheets which we are consider-
ing. It is true that this evidence is of a negative character; but
when we reflect on the duration of the periods required for the
gradual accumulation of these enormous masses of laya, and on the
ample proofs which exist of the occurrence of long intervals of time
between the outflow of sheets now directly superimposed the one
upon the other, we cannot but be impressed by the consideration
that, if such an accumulation of lavas took place upon the sea-
pottom, beds of stratified materials containing marine organisms
must, at some points at least, have been deposited in the intervals
between the successive outflows of igneous rock. The actual com-
parison of the volcanic rocks of the Hebrides and Ulster with others
of undoubted submarine character, like those of Central Scotiand
(which we shall have to refer to more particularly in the sequel,
and in which the interbedding of masses of aqueous and igneous
origin respectively is such a constant and characteristic feature),
lends additional weight to the presumption against the submarine
origin of the former.

But this presumption, derived from indirect and negative evidence,
must be regarded as a conclusion satisfactorily established by all
who will examine and weigh the strong, numerous and cumulative
proofs of a direct kind which can be adduced in its support. These
it will be necessary briefly to detail.

The highly vesicular and scoriaceous character of many of the
Tertiary lavas seems to be inconsistent with the supposition that
they have been poured out under a considerable depth of water,
the pressure of which would probably prevent that extensive libera-
tion of volatile materials which has so evidently taken place. Some
of the lava beds must, in consequence of this operation, have origin-

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 228

ally constituted a mass as porous as a sponge, though subsequently
converted, by the chemical action of infiltrating water, into solid
amygdaloidal rocks *,

In the second place I would direct attention to the nature of the
surfaces over which the first emitted layas have flowed. Thus, in
the North of Ireland, where the lavas rest directly upon the chalk,
the old surface of the latter was not only one of great irregularity,
abounding in hills and valleys, but it is almost everywhere covered
with an accumulation of perfectly angular chalk-flints ; these super-
ficial deposits, too, often fill ‘sand-pipes,” which penetrate for con-
siderable distances into the mass of the chalk rock. No one familiar
‘with the characteristic modes of weathering of the chalk can fail to
recognize in these appearances the evidences of subaerial action. At
other points, as indicated by Portlock in the case of Slieve Gallion
and Rathlin Island, there are found, interposed between the old
chalk soils and the superincumbent lavas, beds of lignite containing
wood and amber, which are unquestionably the remains of old
forests belonging to the period immediately preceding that of the
emission of the lavas. In other cases, again, asin Morvern, we find,
in a sunilar position, thick beds of unstratified ash or volcanic dust7.
But in no instance, so far as I am aware, do the surfaces of the
rocks immediately underlying the lavas exhibit the slightest indica-
tion of having been subjected to the action of marine denudation.

In the third place, when we examine the deposits interposed
between the different lava sheets, and which were evidently formed
during the intervals between their eruption, we find numerous indi-
cations of the prevalence of terrestrial, fluviatile, and lacustrine con-
ditions, but never in any instance of those of a marine character.

_ (1) Bands of clay or earth, usually only a few inches in thick-
ness, and of a bright red colour, very frequently occur between
the sheets of Tertiary basalt both in Scctland and Ireland. These
appear to be identical in character with the beds of soil formed by
the weathering of the surface of one lava stream and destroyed by
burning when overwhelmed by a new sheet of lava. ‘This pheno-
menon, as pointed out by Sir Charles Lyell, is frequently exhibited in
existing volcanoes tf.

* By this infiltration various minerals have been formed in the cavities
of the vesicular lava. By very simple chemical reactions the felspars yield
zeolitic minerals, while the pyroxenic constituents are converted into various
hydrous, magnesian, and ferruginous silicates. More complete decompositions
result in the production of chalcedony and calespar. It is a noteworthy cir-
cumstance, capable of frequent verification in the district under consideration,
that in rocks containing numerous amygdaloidal cavities the mass of the rock
is always much decomposed, and often has passed into a ‘‘ wackose” condition,
through the partial removal of its materials by solution.

+ These beds precisely resemble in character those which are found in con-
nexion with volcanoes but recently extinct. Thus, for example, in the island of
Lipari we find a fine-grained ash of chocolate-brown colour, full of small decom-
posed fragments of white colour (probably fragments of crystals of felspar),
which is quite undistinguishable from the rock referred to in the text.

t “On the Lavas of Etna,” &c. Phil. Trans. 1858, pt. ii. p.711. During a
recent examination of many of the most interesting volcanic districts of the

228 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

(2) Masses of vegetable matter, at times containing recognizable
trunks and branches of trees, in some cases converted by the heat of
the overflowing lava into a kind of charcoal, and at others constituting
beds of lignite or coal, which are often of fair quality and considerable
thickness, but seldom of great extent or very constant character,
are by no means of unfrequent occurrence between the sheets of
Tertiary lava. Such relics of the vegetation of the periods during
which the great sheets of basaltic lava were formed have been found
at many points in Ireland—also in the Isle of Mull, near Carsaig,
at Loch Lathaich and Loch Scridain—in Morvern, Ardnamurchan
and the islands of Higg and Canna—and in Skye, at Tahisker, Scori-
breck, Portree, Camiskianevig, Loch Grisornish and other points.”
Attempts to work these masses of coal and lignite have been made
at various times, especially at the Giant’s Causeway in Antrim, at
Carsaig in Mull, and at Portree in Skye. In the case of the last-men-
tioned locality 500 or 600 tons of fuel is said to have been obtained ;
but, as might be anticipated, the deposits have been found to be too
irregular in thickness and inconstant in character to repay the cost
of extensive operations. Owing to the high price of coal, the well-
known lignite bed at the Giant’s Causeway was again opened out by
the country people during the winter of 1872; and I thus had an
opportunity of studying its characters. It is evidently the vestige
of an old forest overwhelmed by a lava stream. At its base is an
‘underclay ” formed from the decomposition of the older lava bed
on which it rests, and intermingled with much vegetable matter ;
the substance of the lignite itself is made up of great masses of
wood, still very clearly exhibiting its tissues; while the upper por-
tions of the mass, which are in contact with the overlying lava, have
been converted by heat into a kind of charcoal.

(3) In the interesting section of Ardtun we have evidence of
mud streams (such as are so commonly formed in the vicinity of
volcanoes in action) having overwhelmed and buried the deposits of
fine sediment with leaves that had accumulated in pond-like
hollows of an old land surface. These mud streams have evidently,
during part of their course, flowed over a disintegrated surface of
the chalk rocks of the district, and have thus caught up in their
mass many unworn chalk flints and angular fragments of the highly
indurated Scottish chalk *. The remains of many similar old mud

Mediterranean area (which, through the kindness of Mr. Scrope, I have been
enabled to make since this paper was read) I have had the opportunity of seeing
how completely in this, as in innumerable other instances, the products of still
active volcanoes agree with those which are connected with vents long since
extinct, like those on the west of Scotland.

* At the time of the discovery and description of the interesting Ardtun
beds the fact of the existence of rocks of chalk and flint 27 sz¢w in the island of
Mull was unknown. The angular fragments in the mud-stream of Ardtun,
which have been called lapilli, are really portions of the peculiar indurated
siliceous chalk of the district, as, indeed, seems to have been strongly suspected
by the Duke of Argyll at the time. The manner in which fragments of rock
of all sizes, many of them perfectly angular, are mingled together in a fine-
grained paste, seems to me to be inexplicable on any other supposition than that

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 229

Streams, but without the interesting accompaniments of the leaf-
beds and entangled chalk detritus which occur at Ardtun, have been
found by me in various parts of the Hebrides. An example of a
very thick mud-stream buried under lava occurs on the shores of
Loch Tuadh in Mull, opposite to the adjoining island of Ulva.

(4) Professor Geikie has called attention to the existence at the
Innimore of Carsaig of a bed, mainly composed of chalk flints, which
is interposed between the sheets of basaltic lava and attains in some
places a thickness of 25 feet. A careful examination of this very
interesting deposit convinces me that we have here preserved por-
tions of an old river-gravel, formed by a stream which flowed over
the surface of the land, in the interval between the outflow of the
two lava streams which enclose it. The gravels are composed of
flint detritus of every degree of fineness, mingled with rounded
fragments of the local hard chalk and of the basaltic lavas of the
district ; the deposit, which thins out rapidly both eastward and
westward, exhibits the alternations of beds of coarse and fine-grained
materials, with occasional seams of sand and loam, the whole being
marked by much false-bedding; indeed, so far as it 1s exposed in
the sections, this interesting mass of gravels, included between the
Tertiary basalts, is perfectly similar in every respect to the
familiar gravels of existing rivers. It is evident that the stream
which formed these gravels must, in the higher part of its course,
have flowed through a valley cut in the chalk strata. I have found
in other cases masses of gravel formed wholly of pebbles of the
different igneous rocks, which have all the appearance of having
been formed by streams cutting valleys in the older lavas and
covered up by subsequent outbursts; but these of course exhibit
much less striking characters than the deposit of the Innimore
of Carsaig.

(5) At very numerous points, both in the North of Ireland and
in the Western Isles of Scotland, we find deposits of pisolitic iron-
ore interbedded with the basalts. That the materials of these
ferruginous masses were derived from the basaltic rocks on which
they rest there can be no doubt. The actual agency which effected
the concentration of the iron appears to have been that indicated
by Mr. David Forbes*, namely the multiplication of organisms of
a similar kind to those which have formed the well-known “ lake-
ores” of Sweden in recent times. This view is confirmed by the
occurrence, in association with some of the beds of pisolitic ore, of
masses of stratified ash, sand, conglomerate, &c., sometimes con-
taining numerous plant-remains, which are evidently of lacustrine
origin. Such are the deposits of Ballypalidy in Antrim, at which

suggested by His Grace, that they were entangled in a mud-stream. The dis-
coloration of the chalk-flints, supposed to be due to the active heat, [ regard as
being the result rather of weathering operations. At Somma and Bagno Secco
in the island of Lipari we find stratified volcanic tufts with leaves and other
plant-remains, which present a very striking resemblance to those of Ardtun,
Ballypalidy, &e.
* Quart. Journ. Geol. Soc. vol. xxvi. (1870), pp. 164, 165.
Q.J.G.8. No. 119. s

230 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. —

place the extensive workings that have now taken place have
exhibited sections which leave no room for doubt as to the true
character of the beds. I have found sections of precisely similar
deposits on the shores of the Sound of Ulva, and at other localities
in the Hebrides. The frequency with which lakes, shallow lagoons,
and pond-like hollows are formed in a district subject to volcanic
eruptions—in consequence of local subsidences, through the dam-
ming up of river-courses by lava-streams, or through the formation
of “ pit-craters ” by explosion, has been dwelt upon both by Mr.
Serope and Mr. Darwin.

(6) Local masses of unstratified volcanic dust, ashes, scorie and
conglomerate are, by no means rarely, found intercalated with the
great streams of lava.

The whole of these facts point to the conclusion that, during the
period of the emission of the great lava floods which form the
enormous plateaux of the Hebrides and Antrim, the surfaces over
which they flowed were above the sea-level, and, further, that
intervals of sufficient duration occurred between the outpourings of
the lava streams to admit of ihe formation of those very interesting

intercalated deposits, which are in every case of a terrestrial, fluvia-
tile, or lacustrine origin.

4, Evidences of the Former Existence of great Volcanic mountains in
the district—In the foregoing paragraphs we have shown that we
are led to the inevitable conclusion, not only that the great pla-
teaux of the Hebrides and the north of Ireland were formed by the
gradual accumulation of a long succession of lava-flows, but that,
like the similar lavas of Sicily, Auvergne, and Iceland, they were
actually poured out on the surface of the land.

Now, in all the cases of recently formed lavas (to which we have
had such frequent cccasion to refer as exhibiting so perfect an
identity in character with the ancient rocks which we are describing),
their extrusion has been accompanied by the piling up of ejected
fragmentary and semifluid matcrials around the vents, so as to
form those more or less conical mountain-piles which we call «“ yol-
canoes ;” and it is from the summits or sides of these that the
great lava streams have usually flowed.

Can we then hesitate to accept the conclusion that rocks which
present such a remarkable identity in characters, even to the
minutest details, must have had a similar origin ?—that the forma-
tion of these ancient subaerial lavas was attended with the produc
tion of phenomena similar to those which, at so many different
points of the ecarth’s surface, we still observe to be the constant
effects of volcanic action ?—that, in short, within our British district
there once existed volcanoes from which those great volumes of
igneous materials we have been describing were ejected ?

Further, when we reflect upon the tact (one patent to the most
tasual observer of the geology of the district) that the masscs of these
ancient lavas which remain for our study, notwithstanding their
vastness, are nevertheless mere isolated relics, which have escaped
destruction by denudation, of plateaux which must originally have

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 231

covered many thousands of square miles*—that, moreover, these
fragments which remain still constitute mountains nearly 2000 feet
in height, entirely made up of almost horizontal lava sheets,—must
we not conclude that the volcanoes from which these wide-spreading
and thickly piled masses of lava were ejected could haye been of no
mean dimensions or insignificant character ?

At what points, then, were these great volcanoes situated ? Where
are the relics of their vast masses, the indications of their violent
action? Can it be that all traces of such great mountain-piles,
belonging as they do to one of the most recent of the geological
periods, have altogether disappeared? Or are we rather to con-
clude that denudation and other causes have so moditied and dis-
guised their characters, that their true nature is not now at first
sight recognizable? These are the questions that everywhere pre-
sent themselves to the geologist in the Western Highlands, and press
for a solution.

In support of the opinion that, even within a comparatively recent
geological period, such vast volcanoes may have entirely disappeared,
while extensive fragments of the lava streams which flowed from them
so conspicuously remain, it may be pointed out, first, that the former
would be composed mainly of loose and fragmentary materials, which
are much more easily acted upon by degrading ferces than the solid
rocks of the latter; secondly, that their greater elevation would faci-
litate their destruction; and, thirdly, that the evidence of denudation,
even on the most stupendous scale, having taken place in the district
since the period of the formation of these volcanoes, geologically re-
cent though they are, is of the most unmistakable character,

But on the other hand it must be remembered that the remains
of the great volumes of fluid materials which ascended through the
volcanic piles, and, sending off ramifications in all directions, served
to bind together the masses of fragmentary matter, must not only
in themselves have constituted rocks of great solidity and perman-
ence, but, by atlording coherence to the volcanoes have retarded
their denudation. Of these solidified igneous springs, which con-
stituted the connexions between the great rivers of lava and the sub-
terranean reservoirs of liquefied rock, it 1s hard, nay, impossible, to
conceive the remains to have been wholly removed while any
portion of the older rocks which they penetrated remained ; though
in many cases we can readily understand the effects of denudation

to be such as to render obscure the connexion between these solidi-

* It is perhaps not possible to adduce any more striking illustration of the
vast changes which have taken place in the relative position of rock-masses,
and the extent to which they Lave been subjected to denuding agencies since a
period so geologically recent as the Miocene, than that which is afforded to us
in the great fauit at Morvern. Here we find a mass of strata compcsed of
Lias uncontormab!y overlain by Upper Cretaceous, and this in turn by Miocene
basalts, let down by a fault, which must have a throw of nearly 20U0 feet, against
the contorted gneissic rocks of the Lower Silurian, Nothing can present a
greater contrast than the green terraced hills of the former, rising in Glashven
to the height of 1516 feet, and the barren and rugged mountains of the latter
culminating in Garbh-shlios (1638 feet). he abrupt junction of these rocks
can be traced in the wild ravine of Innimore.

8 2

932 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

fied reservoirs and rivers, through the removal of all traces of the
ducts which once connected them with one another, and which must
have necessarily existed at higher elevations than either of them.

In the following pages I hope to be able to supply a complete
and satisfactory solution of the interesting question of the position
of the vents from which the great lava streams so conspicuons in
the north-western part of the British archipelago originally flowed ;
and even of the dimensions and features of these old volcanoes I
hope to be able to furnish some data for forming a judgment. At
the same time I shail exhibit, in its outlines at least, the history of
the long series of volcanic eruptions to which the rocks of this
district owe their accumulation. Further, I believe that it will be
possible, from these interesting examples, to deduce some important
conclusions as to the modes of operation of volcanic action, especi-
ally of that by no means insignificant part of it which takes place
far below the surface, and which, in the nature of things, can never
be observed by us in volcanoes which are still active or but recently
extinct ; for in these the mountainous piles of ejected materials
altogether conceal the underlying rocks. Even in the cases of the
deepest ravines on the flanks of existing volcanoes (and I do not here
except even the wonderful natural sections of Somma and the Val del
Bove), interesting as these undoubtedly are, the insight afforded to
us of the internal structure of the mountains must be at the best
very partial indeed ; but in the Hebrides I shall show that we have
supplied to us that great geological desideratum—a number of
voleanoes so dissected by the scalpel of denudation as to constitute,
as it were, a series of anatomical preparations, from which we may
learn directly the internal structure of the piles, and obtain bases
for reasoning on the causes to which that structure owes its origin.
Thus the study of these ancient volcanoes is, as I believe, calculated
to throw new light upon some of the most difficult problems of
physical geology.

As I have already intimated, we have, in the west of Scotland,
the relics of two great and widely separated periods of volcanic
action: one of these was evidently of earlier date than the depo-
sition of the Mesozoic strata which unconformably overlie its
products ; the other was certainly of later date than the formation
of the Secondary rocks, upon which its products are everywhere
superposed, and through which they are intruded. As might be
anticipated, the relations between the various products of the two
volcanic series and the intermediate Secondary strata are often of the
most intricate character.

For the purposes of the present argument it will be convenient to
discuss, in the first instance, the characters presented by the younger
series of volcanic rocks, as these, owing to their comparatively
better state of preservation, often throw much light on the internal
structure and the true relations of volcanic masses. Guided by the
analogies of this younger series of rocks, we shall be the better
prepared to decipher the more obscure, because less perfectly pre-
served, relics of the older series of similar products.

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 233

fl. The Tertiary Volcanoes.

1. Classification of the Tertiary Volcanic Rocks.—In approaching
the consideration of this part of my subject, I hasten to acknowledge
the important assistance which J have received from the very
valuable memoir of Prof. Zirkel, of Leipzig, entitled ‘ Geologische
Skizzen von der Westkiiste Schottlands ” *, which made its appear-
ance in 1871. This author, who is so well known for his im-
portant contributions to petrolog ry, made a tour in the Hebrides in
1868, and in the memoir referred to has given an admirable sketch
of all that had hitherto been done in the elucidation of the geolo-
gical structure of the district, accompanied by the details of his
microscopic studies of the various specimens of rocks collected by him.

Bearing in mind the backward state of petrological investigations in
this country, I have in the present memoir almost uniformly followed
the classification and adopted the nomenclature of Prof. Zirkel. The
few instances in which I have departed from his terminology, such
as, for example, in avoiding the use of the terms “ porphyry ”
and ‘“‘amygdaloid,” and employing their adjective derivatives instead,
will, I hope, commend themselves to all geologists. In the case
of rocks not noticed by Prof. Zirkel in the memoir cited, I have
endeavoured as far as possible to make my use of terms correspond
with the definitions given by the same author in his very admirable
‘Lehrbuch der Petrographie,’ published in 1866.

In studying these rocks and minerals I have also to acknowledge
the valuable assistance of Mr. David Forbes, F.R.S.; of Mr. Thomas
Davies, of the British Museum ; and of Mr. Rudler, of Jermyn Street.

The Tertiary volcanic rocks constitute two well-marked parallel
series. Hach of these series has at one end of it a highly crystalline
rock, and at the other a perfect glass; while the intermediate terms
constitute the most complete and insensible gradation between these
two extremes. The ultimate chemical composition of the rocks of
either series is identical in all the members of it, or rather varies
between the same comparatively narrow limits. But the average
composition of the rocks of the two series presents the most marked
contrast when compared with one another.

These two series of igneous rocks have long been recognized by
petrologists as occurring both in recent volcanoes and in association
with the strata of every geological period. From the marked differ-
ence in the quantity of silica which they contain they have been
designated the acid and basic series: the former is sometimes known
as the felspathic, orthoclastic, or traehytic series; and the latter as
the augitic, plagioclastic, or basaltic.

In the members of the acid series of igneous rocks the percentage
of silica varies from 60 to 80, averaging about 70; while in the
basic its hmits are 45 and 55, and its average 50. In the former
class of rocks the proportion of alkalies is comparatively large, and
that of the alkaline earths and the oxides of iron and manganese
small; but in the latter, lime, magnesia, and oxide of iron con-

* ‘ Zeitschr. d. deutschen geologischen Gesellschaft,’ Jahrg. 1871, pp. 1-124.

234 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.
stitute a very large proportion of the mass, while the quantity of
soda and potash is comparatively insignificant,

The contrast between the ultimate composition of the rocks of the
two séries is illustrated in the following Table :—

Acid series. Basic series.
Max. | Average. | Min. || Max. | Average. | Min.
= — —<———$—_—_ | ———
Srlit@ai ue innene anon: Lssol O 72 61 55 50 45
Adluninaietay ey scincee nese ae 19 15 10 18 14:5 10
INURE OS) Sctbe avon steoodechoe 20 7 0 8 3:5 0:5
Alkaline earths ............ 8 2 0 99 WW rs)
Oxides of iron i) 3 25 0 19 14 8
manganese..........--

2 ee

In the acid series the highly crystalline form is known as granite,
the glass as pitchstone or obsidian. ‘The intermediate terms of the
series have received a great number of names, only a few of which
it will be necessary to notice here,

In the basic series the most highly crystalline form is known as
gabbro, the glass as tachylite ; and the intermediate forms are ale.
as in the fort mer case, very numerous.

In the accompanying Table (p. 235) the composition of the princi-
pal rocks of these two series is illustrated; the averages are for the
most part those calculated by Durocher ; in the instances marked by
an asterisk I have been compelled to employ averages calculated by
myself from published analyses.

The Tertiary granites consist essentially of two species of felspar
with quartz, and either mica or hornbiende, or both of these minerals
together. These granites often, but by no means uniformly, present’
a somewhat cavernous structure, in which case the constituent
minerals are found with perfectly terminated crystals projecting
into the cavities of the rock. The felspars consist of orthoclase
and oligoclase, with much more rarely albite also; their crystals
often present a pale brown or buff colour, owing to the partial
decomposition of ferruginous materials entangled in them. When,
however, the rock is quarried at some depth from the surface the
felspars are usually seen to be of a beautifully white colour ; while
more rarely they present pink and reddish tints. The quartz 18
usually white, but not unfrequently smoky, and in rare cases black ;
it sometimes forms perfect crystals. ‘The felspar and quartz make
up the mass of the rock, which is always a highly silicated one;
and the magnesian and ferruginous silicates are present only in
comparatively small quantities. In the deeper and more central parts
of the greut granitic masses the rock is micaceous; but as we pass

235

Table Illustrating the Average Ultimate Chemical Composition of the Rocks of the Acid and Basic Series.

e Acid Series.

=<

4 ad

8

S Specific gravit Silic | Alumin Potash Soda. Lime Magnesia. Onidesrefaron

i p gravity. Silica. ma. 28 IT ba) a nea. and manganese.

by aS ;. so Jess SF}.

q Max.| Av. | Min.| Max.! Av. |Min.!Max.' Ay. | Min,! Max.) Av. | Min.'Max.) Ay. | Min.) Max, Av. | Min.| Max.| Av. | Min.| Max.} Av. | Min.

§ ee ee al i | Beil | hs a A | 4 | cal

Fe (CAEHOMTUE) ooncosos ,..| 2°73} 2:66 2°60). 78 |728, 66 | 18 | 153) 11 | 9 | G4) 4 | 2°5 14)o |rspOTho f2 |O9%o0 -a5) 17) o

Bb Syenite-granite ...] 2°75] 2°68 2°63) 72 | 69:0 64 | 17 | 150) 1a) 6 Wl eB ge “eg sei2:Bh| ae oh a Dee ea Oa ae ees v2] 2

= Felsite *....... .....| 2°70] 2:65, 2°60] 81 | 743) 68 | 15 | 13) iz | 8 | 44 se tee | DR org gL Teo iose heres cOG On fos ea

p Felstone .......- ...| 2°63] 2°64 2°58] 80 | 754) 68 | 18 |15°0) rr | 6 | Sl) 2 | 6 [Biko || & ORB om anki TIE) oases) 23 ous

9 Pitchstone &c. ...| 2°36] 2°34 2°31] 74 oe G2), a7 190 rr | 6 | BGG 35 | 24) west wes) LS re ee 9) OOh a eae) 2.6) a

Ba | }

a

|

Cy . .

x Basic Series.

A baal Paes

E Gabbro ...... aaeer: 3°10] 2°95) 2°85) 54 |49-0) 45 | 17 [15-0 12 | rt | OS) oO | 4 B5|\ 015 14 190) 6 | 15 | 0 7 fre 11-5] 8

E Augitic-gabbro * .| 3-90 290, 2°75 54 |49°3| 45 | 20 | 165, 14 | 3 L7|o8{6 |3:9)2 | 11/79) 45| 7 | 54) 4 | 16 (13-0) 8

ad Dolerite ..........--| 3°10} 2°95) 2°85) 55 |51-0) 45 | 16 | 140, 12 | 1 | O2)o | 5 BA) 2 | 13 10:0) 7 Ona ge ta at no

bi 18} yf cqeeeneee soe ana 2°96) 2°85) 53 (48 az 118 |1o-8) tor) 3° Fbios is | sO) 2 | 14 102) 7° | 10 65) 3 16 |13 8] 9
Tachylite*........ 2-52) 2°50! 6 | 54:3) so | 18 | 160, 12 Xe}! Te} 40 8 | 72| 6 6 | 2:8] ons} 13 [11-2 \10

J 5 5 4 Sd 3 Sy) ess

2

—

236 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND:

towards the higher and outer parts the mica is gradually replaced
by hornblende, ‘and the rock passes through the varieties of horn-
plendic granite into syenite-granite. The adventitious minerals in
these granites are pyrites, marcasite, chalcopyrite, garnet, apatite,
and epidote. The texture of the rock varies greatly, from coarse
and sometimes porphyritic varieties to others so finely granular that
they might be mistaken at first sight for sandstone.

By the disappearance of the cr vatale of hornblende (which appear
to be replaced in many instances by some easily decomposable
mineral, lining the cavities of the rock) the syenite-granites pass
into felsites. These exhibit every variety, from the most granitic
forms, like eurite, to others which have been called quartz-trachyte,
from which volcanic rock, indeed, they are in no way distinguishable.
These felsites are usually quartziferous, and sometimes very highly
so; at other times they are almost wholly made up of crystals of
felspar, imbedded in an amorphous paste; and not mee
they exhibit the porphyritic structure.

When the felspathic rock becomes compact, as it usually does in
the lava streams, it may be called felstone. The felstones of the
Hebrides vary in colour from black, through various shades of green
and grey, to white; but in almost all cases their surfaces acquire a
white crust as the consequence of weathering action. By this cir-
cumstance, and also by the manner in which they withstand denuding
influences, preserving everywhere the strize and roche-moutonnée
forms impressed upon them during the glacial epoch, the felstone-
lava streams are strikingly distinguished from the basaltic. Both,
however, exhibit the columnar structure (though with characteristic
peculiarities in either case), the globular forms developed by wea-
thering, and the amygdaloidal structure, which in both is developed -
to the greatest extent at the upper and under surfaces of each lava
stream. ‘The felstones appear to be highly siliceous trachytic lavas
which have undergone more or less alteration ; and occasionally, by
weathering action, the obliterated banded and spherulitic structures
so commonly found in, and so characteristic of, quartz-trachytes*,
are restored in their ancient representatives, the felstones.

When the same rock becomes glassy in structure it is pitchstone
or obsidian, the former name being properly applied to the varieties
with a resinous, and the latter to those with a vitreous lustre.

In the various intrusive masses and veins the several forms of
highly felspathic rock are found passing into one another by the
most insensible gradations.

In the basie series of rocks the place of Pranite is occupied by
gabbro, a rock which, from the similarity of its mode of occurrence
and behaviour to that of the former, has been called “‘granitone” and
“‘eranito-di-gabbro.” — It consists essentially of a plagioclase felspar,
one or more pyroxenic minerals, and olivine—the latter mineral being
so uniformly present that, according to Zirkel, it must be regarded as
an essential ingredient of the rock. The felspar of this rock appears

* See Mr. Scrope ‘‘ On the Geology of the Ponza Islands,” Trans. Geol. Soe.
2nd ser. vol. ii. p. 195.

*

J. W. JUDD ON THE SECONDARY ROOKS OF SCOTLAND. BBi7/

to be in almost every case labradorite, of white, green, or purplish tints,
and often belonging to the beautifully glassy varieties. The pyroxenic
ingredient is usually diallage; but this is often replaced on the one
hand by hypersthene, and on the other hand by augite, in the
same manner as Streng has shown to be the case with the gabbros
of the Hartz. In many cases the augite wholly replaces the diallage;
and this is found to be always the case as we pass to the higher and
outer portions of the great intrusive masses; while occasionally,
instead of augite, we find some other variety of pyroxene, such as
the beautiful green coccolite. These brilliant varieties of pyroxene,
with the glassy labradorite and the peculiar forms of olivine, con-
stitute rocks of great beauty. The olivine is usually filled with
innumerable microliths of chromic or titaniferous iron, by which its
ordinary aspect is entirely masked; sometimes, however, it presents
its usual clear pale-green tint, and at others, through partial de-
composition, has assumed a reddish-brown colour. The gabbros
vary in structure from aggregates of crystals, sometimes 2 inches in
length, down to finely granular rocks. Among the adventitious
minerals which they contain, the first place must be assigned to
magnetic iron, which is often present in large quantities; pyrites,
marcasite, chalcopyrite, biotite, garnet, apatite, and epidote also
occur in it ; while serpentine and chlorite are among the results of
the incipient alteration which is often found taking place in its mass.

When the gabbro in which the diallage is wholly replaced by
augite assumes a granular structure, it becomes dolerite ; and when
still finer grained, it passes into anamesite and basalt; finally, when
the basic rock becomes glassy in structure, it is known as tachylite.

Each member of the basic series of rocks can be seen passing into
the others -by the most insensible gradations. Thus the side of a
gabbro vein is often formed by a layer of basalt; that of a basalt
vein by one of tachylite.

The parallelism between the different varieties of the acid and
basic series of igneous rocks, and the similarity in their modes of
occurrence, is very striking. There is one important distinction
between them, however, which it is necessary to bear in mind: the
crystalline acid rocks (granites) are usually among the most stable
and indestructible of the materials of the globe; the crystalline
basic rocks (gabbros) are, on the other hand, particularly liable to
decomposition, the diallage and olivine giving rise to the formation
of various serpentinous minerals.

I have not attempted to describe the interesting details of minute
structure which the microscope has revealed in most of the rocks of
these two series, but must refer the reader for information on this
subject to the excellent memoir of Prof. Zirkel before quoted.

Besides the acid and basic rocks of Tertiary age in the Hebrides,
there occur also a few examples of the intermediate forms—such as
diorites and syenites among the intrusive masses, and ‘“ porphyrites”
and phonolites among the lavas. But these are so small in quantity,
and so local and exceptional in their mode of occurrence, as not to
demand more than a passing notice in this place.

238 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

2. Nature and Origin of the Great Volcanic Rock Masses.—Con-
founded under the common name of “trap,” several very important
distinctions have been, in the case of these igneous rocks, to a great
extent lost sight of; these it will therefore be necessary to refer to
particularly.

a. Luvas.—These, as already noticed, constitute the great mass
of the plateaux, which have been cut up by denudation into those
terraced hills so characteristic of the district. They are usually
composed of compact or crypto-crystalline varieties of rocks, more
rarely of the glassy forms (in the case of the acid and intermediate
varieties), but never of the largely crystalline or granitic forms.
In very many cases they are distinguished by their amygdaloidal
character; but the absence of this feature must not be regarded as
an evidence that any particular rock is not to be considered a lava;
and, indeed, in thick lava streams the amygdaloidal (or altered vesi-
eular) varieties are almost wholly confined to the upper and lower
portions of the mass*. The amygdaloidal lavas often exhibit evidence
of their vesicles having been drawn out by the flowing of the mass,
put in very various degrees in different cases. Both the basic and the
acid lavas at times assume columnar characters; but in those of the
former class these are much more frequently displayed, and in a more
striking form. The basaltic lavas often exhibit the thick regular
columns with equidistant joints, cup-and-socket articulations, and an-
gular processes so well known as occurring at Staffa and the Giant’s
Causeway; in the felspathic lavas, on the other hand, the columns,
when present, are usually of smaller diameter and less regular form,
while they are often of great length, and never exhibit joints at

regular intervals. The masses of felstone lava are seldom found
extending to greater distances than ten miles from their centres of
eruption ; the basaltic lavas, however, have spread in vast sheets to
distances of fifty or stxty miles from those centres.

b. Intrusive Masses.—These vary in dimension, from the thinnest
veins or strings, to vast bosses constituting great mountain-groups
like the Cuchullin Hills of Skye, or the deer-forest of Rum and
Ardnamurchan. In their ultimate chemical composition they coin-
cide perfectly with the rocks composing the lavas ; but in the varieties
of their texture and mineralogical constitution they exhibit a much
wider range. Thus, while we find veins of basalt in which the
rock-structure is identical with that of many of the lavas, we find
also others in which the same rock passes into a glass, tachylite ;
while others, again, are composed of the highly crystalline or eranitic
gabbro rocks. Similarly, felstone veins are related to those of pitch-
stone and obsidian on the one hand, and to masses of felsite, syenite-
granite, and granite on the other. As a general rule it may be
stated that the largest intrusive masses are composed of the most
highly erystalline or granitic rocks. Thus the gabbros on the one
hand, and the granites on the other, constitute, for the most part,

* The manner in which this character is displayed by almost every recent
Java stream is familiar to all who haye visited volcanie districts, and has been
alluded to by many authors.

‘

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND, 239

preat mountain masses; the dolerites and the felsites occur as in-
trusive sheets or bosses ; while the basalts and felstones form narrow
dykes and veins, of which the sides and the smaller offshoots pass
into tachylite and pitchstone respectively. Sometimes different
varieties of rock-texture are exhibited in the same mass; thus the
sides of a mass of granite often pass into felsite, and a vein of
gabbro is bounded by surfaces of dolerite or basalt. That these
masses were actually forced through older rocks, is shown by the
manner in which the latter, whether of aqueous or igneous origin,
are disturbed in the neighbourhood of the larger masses of the kind ;
that they were, at the time of their eruption, in a fluid condition, is
proved by the manner in which they have occupied even the
minutest fissures of the disturbed rocks ; and that this fluidity was
connected with a great development of heat, is indicated by the
changes to which they have given rise in the rocks with which they
have come into contact.
_ The intrusive rocks seldom exhibit the vesicular (resulting in the
amygdaloidal)y structure; but they are, on the other hand, often co-
lumnar. Jn the case of the larger masses the columns are generall
of large proportions, but indistinct character, while in the dykes and
veins they are usually small and often very minute, being in all cases
arranged at right angles to the surfaces of the intersected rock.

c. Volcanic Aqylomerates.—But besides the two series of rocks
which we have described as constituting by tar the greater portion
of the volcanic products of the Hebrides, namely the lavas and
eruptive masses, there are others of a very remarkable character,
which have hitherto almost wholly escaped observation. Like the
two classes of rocks just described, these rocks have also been con-
founded under the general name of “ trap.” They are always found
associated with the great masses of eruptive rock, and are often
remarkable for the fantastic forms to which, in consequence of their
peculiar modes of yielding to denudation, they give rise. On their
fresh surfaces of fracture the rocks of this class often present no
peculiarity which would enable us to infer their mode of origin ; it
is only on the surfaces of masses which have had their remarkable
structure ‘‘ developed ” by weathering operations, that these striking
features are revealed to us. In the same manner that a highly
crystalline and apparently altogether unfossiliferous limestone (like
that of Durness, for example) often exhibits on its weathered faces
preofs that it was originally made up of a congeries of beautiful
organisms, so these apparently structureless igneous rocks are by
the same agency made to reassume to a certain extent their original
form. The weathered surfaces of the blocks show that the rocks in
question are really agglomerates, composed of angular and sub-
angulir masses of rock of all sizes, from blocks several tons in
weight down to lapilli, voleanic sand and dust. Many of these
fragments are shown by the same weathering process to be highly
vesicular and scoriaceous, and among them not a few present the
peculiar characteristics of entire or fragmentary ‘‘ volcanic bombs.”
In many cases crystals of various volcanic minerals, either fractured

240 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND,

or with worn angles or faces, such as commonly occur among the
materials ejected from volcanic vents, are found imbedded in these
agelomerates.

That these peculiar rocks are really portions of old volcanic
piles, and that they are made up of the fragments actually ejected
from volcanic vents, no one who has had an opportunity of studying
the appearances presented by their weathered surfaces can for onemo-
ment doubt. But in the same manner that loose masses of shells and
corals have been converted, through the chemical reactions set up by
the infiltration of water, into crystalline and apparently structureless
limestone, so these old heapsof volcanic cinders have been transformed
by similar agencies into rock-masses of great solidity and hardness.

But lest any doubt should still remain as to the real origin of
these remarkable rocks, I am fortunately able to adduce a kind of
evidence of the same convincing character as that of fossils in the
sedimentary rocks. He who would hesitate to accept as evidence
of aqueous origin in a rock such physical characters as stratifi-
cation, oblique lamination, and ripple-marks, could“scareely con-
tinue to doubt, if shown in its mass, groups of oysters, bored by
sponges and annelids, and overgrown by serpulee and polyzoa.

In the case of the volcanic rocks, though we cannot adduce the
evidence of fossil animals and plants, we can nevertheless point to the
characteristic species and varieties of minerals which they contain.
Every one is aware that the vents and cones of volcanoes constitute
natural chemical laboratories in which many beautifully crystallized
minerals are formed which are seldom or never found except in the ac-
tual vicinity of eruptive centres. Hence such species and varieties are
usually classed by mineralogists as ‘‘ voleanic minerals.”’ These mi-
nerals, as is so well known, are constantly found filling the fissures and
vesicular cavities which occur in the various blocks ejected from the
volcanic vent; and as has been remarked by Daubeny, Sorby, and
other authors, they usually belong to different species and varieties
from those which characterize the lava streams which have solidified
at some distance from the volcanic vent. The agents concerned in
the production of these volcanic minerals appear to be intense heat,
combined with the infiltration of water under immense pressure, and
the penetration of various acid gases and volatile materials.

Many of the peculiar volcanic minerals are of remarkably unstable
character ; and of these it would be as unreasonable to expect the pre-
servation as that jelly-fish, lobworms, and the soft tissues of animals
and plants should be found in sedimentary rocks. But the more
stable species of volcanic minerals actually do occur in a “fossil state”
in the midst of the voleanic agglomerates of the Hebrides ; and the
preservation, under favourable conditions, of their often delicate and
beautiful crystals is the result of a remarkable series of operations.
In the first place, by infiltration, the masses of crystals are surrounded
by, and imbedded in, a mass of zeolitic minerals which is formed by
the decomposition of the felspars of the lava; and thus carefully packed
they escape further injury. Subsequently, on the exposure by denu-
dation of the rock-masses, the “ cotton-wool ” of zeolitic materials is.

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 241

gradually removed : and, under favourable circumstances, the original
crystals are sometimes thus again exposed, exhibiting all their pris-
tine lustre and beauty. Among the minerals thus occurring many
beautiful varieties of the epidote and garnet groups are especially
conspicuous. Some of these minerals appear to have been formed
in the fissures of masses of rock which were ejected from the
voleanic vents; others may have originated from the action of
various gases upon the materials of the lava during the “ solfatara
stage ” of the volcano’s history.

d. Volcanic Breccias—The rocks of the last-mentioned class pass
by insensible gradations into those which it is now our purpose to
describe. Scattered through the masses of the agglomerates of
scoriz and ashes, we find blocks of stone of the most various kinds,
which are evidently of foreign origin; and in some cases these
become so numerous that the rock may be described as a breccia
composed of fragments of foreign rocks imbedded in a matrix of
volcanic ashes, sand, and lapilli.

That these angular blocks, which vary in size from masses several
tons in weight down to the smallest fragments visible to the naked
eye, were actually ejected from volcanic vents, we should be justified
in concluding, not only from the analogy of existing volcanoes, but
from the confused and irregular manner in which they le in the
masses of evidently erupted volcanic materials*. Fortunately, how-
ever, we can point to facts which are altogether conclusive as to the
origin of these blocks. They are found to belong, in every case, to
the particular rocks through which the volcanic vent which they sur-
round has been opened. Thus in the island of Rum (where these
volcanic breccias are admirably displayed, as we shall see hereafter),
which forms the central part of an old volcano that has burst
- through rocks of Cambrian sandstone, the blocks in question consist
exclusively of materials derived from that formation; in Mull a
great volcano has originated in the midst of various Paleozoic and

* The frequency with which blocks torn from the rocks underlying a volcano
are thrown from its vents during eruptions is a fact familiar to all students of
volcanic geology. Alike in the old agglomerates of Somma and among the
modern ejections of Vesuvius, there abound fragments of those stratified rocks
in the midst of which the great fissures to which the Neapolitan volcanoes owe
their origin have been opened... The so-called “lava” ornaments of Naples are
made from the limestones, often more or less dolomitized, of these ejected blocks ;
while from the cracks and cavities of the same are derived the greater portion of
those beautifully crystallized specimens which have made Somma and Vesuvius
so preeminently famous among mineralogists. ‘The blocks thrown out of the
great Neapolitan volcano were, for the most part, evidently derived from the
Subapennine (Tertiary) strata; but some from the Apennine (Secondary) rocks
also occur. My friend Prof. Guiscardi, of the University of Naples, was, in
1856, able to give a list of no less than 112 species of fossils derived from the
blocks thus ejected from the crater of Vesuvius (see his “ Fossile Fauna des
Vesuvs”’ in Roths ‘ Der Vesuv und die Umgebung von Neapel,’ 1857); and this
list he has since increased. Near the great crater now occupied by the Lago
Bracciano, I found great masses of voleanic agglomerate containing such enor-
mous quantities of angular fragments of Subapennine limestone as to constitute
“‘volcanic breccias ’ greatly resembling those of Mull which are referred to in
the text.

249 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

Secondary strata, fragments of the whole of which occur in the vol- |
canic breccias; while at Beinn Shiant, a volcanic vent having been
opened in a tract made up of older lavas, easily recognizable por-
tions of these occur in the great masses of agglomerate which con-
stitute a portion of the relics of the volcanic cone.

In many cases such ejected blocks exhibit evidence of having
been submitted to a certain amount of heat. Being often composed
of materials of much more indestructibie character than the consoli-
dated ashes &c. in which they are imbedded, they are frequently left
lying, like the included blocks of masses of Boulder-clay which have
been destroyed by denudation, upon the surfaces of the mountains ;
and for these, indeed, they may, by a casual observer, be easily mis-
taken in some instances. Such an idea, however, is at once removed
by an attentive study of the whole of the phenomena and conditions
of the case.

3. Relations of the Volcanic Rocks to one another and to the Older
Deposits in the Island of Mull.—We now turn to the consideration
of the very interesting question of the manner in which the several
kinds of volcanic products, which we have described in the preceding
pages, are associated with one another, and of the positions which
they occupy with respect to the strata of the Paleeozoic and Mesozoie
epochs. Owing to a peculiar combination of favourable aircum-
stances, which will be more particularly described in the sequel, the
island of Mull furnishes us with a most complete and beautiful illus-
tration of the relations which the voleanic rocks bear to one another ;
and we therefore propose, in the first instance, to describe the struc-
ture of this district.

The island of Mull, with the adjacent peninsula of Morvern and
various small surrounding islands, form portions of a great plateau,
composed ef sheets of basaltic lava piled upon one another to the
depth of nearly 2000 feet. This plateau, which is now broken up
by denuding agencies and traversed by numerous fiords and
sounds, 1s, on its southern side, brought into abrupt contact with
the Paleozoic rocks through the agency of a great fault, which bas
a throw of at least 1600 feet.

In the centre of this basaltic plateau rises a group of lofty moun-
tains, about twelve miles in diameter, many of the peaks of which rise
to heights of from 2000 to 3000 feet above the sea-level, while one of
them, Beinn More, attains an altitude of no less than 3172 feet.

Nothing can be more striking than the contrast presented by the
mountain-peaks of this central group of mountains and the emi-
nences of the surrounding plateau. While the latter are every-
where characterized by those remarkable tabular and terraced forms
which have suggested the name of “trap” (forms which, by con-
stant repetition in these districts, tend to become monotonous), the
former exhibit the most wonderful diversity of outlines—lofty,
smooth, pyramidal masses, like Beinn ‘’salla, being mingled with
wildly irregular, and fantastically shaped peaks, like Beinn Buy,
Creach Beinn, Beinn Varnach, Dun-da-gu, and Craig Craggen.

A very slight examination of the rocks of Mull is sufficient to

3. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 248

explain the cause of the differences presented by the mountains of
the central and outer portions of the island respectively. While
the latter are made up of nearly horizontally disposed lava streams,
the former consist of a number of intrusive masses composed of
rocks of very various degrees of hardness, and associated with great
deposits of volcanic pe slamerites and breccias—the several rocks
yielding, as might be anticipated, to denuding forces in very un-
equal degrees and in very various manners.

The positions and relations of the rocks of Mull are illustrated in the
accompanying plan and sections (Pls. XXII. & XXIII.). In these an
attempt has been made to exhibit the relations of the different rock=
masses by employing different colours for the two great varieties of
igneous rocks, while the variations in the depth of tint in either case
serve to indicate their crystalline, stony, or glassy character.

Forming the basis of the great central mountain-group of Mull
are masses of highly siliceous intrusive rocks. Where great valleys
have deeply intersected these masses, they are seen to be composed
of granite, usually of the hornblendic variety, but passing, in the
deepest and most central portions of the masses, into ordinary gra-
nite with mica. But in proportion as we trace the rocks of the
mass outwards and upwards, we find the granite passing by insen-
sible gradations into felsite. Lying upon the skirts of these central
bosses of granitic and felsitic rock are thick masses of felstones,
disposed in regular sheets and of amygdaloidal structure, which
alternate with beds of scoriz, lapilli, and ashes, containing many
included blocks of the stratified rocks of the island. While these
latter undoubtedly represent the lavas and fragmentary materials
ejected from a volcanic vent, the central hills of granite and fel-
site are no less certainly great eruptive masses; and that this is the
case is proved by the fact that they give off numerous veins both
into the overlying felstone lavas &c. and into the surrounding older
strata, which veins are often seen to have entangled fragments of
the various rocks which they traverse.

Rising through the midst of these masses of granitic and felsitic
rocks, and constituting the bulk of Beinn Buy with parts of the
surrounding mountains of Creach Beinn, Beinn Varnach, and Beinn
Tsalla, we Gnd a great mass of gabbro or basie rocks eam which
there proceed in every desarcn innumerable veins, sheets, and
intrusive masses of irregular form,.that traverse the whole of the
highly siliceous rocks in every direction. In these great intrusive
masses of gabbro the pyroxenic ingredient in the deeper and more
central parts appears to be dia‘lage and hypersthene, which mine-
rals are replaced in the peripheral portions by various forms of
augite. As we trace the gabbros into the numerous veins which
proceed from the central mass, they are found to graduate into
dolerites, exhibiting every degree of coarsencss of grain, and through
these into ordinary basalt. In the latter condition the veins and
dykes often proceed to great distances from the central mass, inter-
secting alike the lavas of the great plateaux and the various Prima
and Secondary strata which underlie them (see Pl. XXIII. fig. 1),

244 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

Owing to the very different modes of weathering of the igneous
rocks of the acid and basic series respectively, it is comparatively
easy for the geologist to study the relations which they bear to one
another. This study i is conclusive as to their relative age and position.
The rocks of the acid series were first extruded and consolidated ;
and subsequently there was forced through their midst a fluid mass
of basic materials penetrating the amaraera ale fissures of every
size which, by the upheaving force, were rent alike in the earlier-
formed igneous recks and the surrounding strata.

Overlapping the edges of the felstone- lava series we find ie basaltic
lavas of the great surrounding plateaux ; and that these once entirely
buried the former is shown by the small outlying portions of basalt
which at some points still remain capping the felstones.

Finally, we must notice that all round the central masses of
eruptive rocks there occur patches of volcanic agglomerate and
breccias, alternating with lava-sheets and traversed by innumerable
veins and dykes. These patches constitute the last remaining ves-
tiges of the great conical piles of fragmentary materials which so
frequently surmount volcanic vents. As will be hereafter shown,
they belong to two different periods of eruption.

Of the rocks upon which the streams of lava which constitute so
large a portion of Mull were poured out, and through which the great
eruptive masses of the central portion of the island have forced a
passage for themselves, we are not left in doubt. At the south-
western extremity of the island the peninsula of the Ross of Mull,
consisting of Lower Silurian gneiss and mica-schist, with an eruptive
mass of granite (which will be shown hereafter to be of later
Paleozoic age), stretches beyond the lavas of the great plateau, and
is continued in the proximate island of Iona, composed of rocks
supposed to be of Laurentian age. In the south-eastern part of
Mull a series of old volcanic rocks, alternating with breccias, con-
glomerates, and sandstones (all of later Palseozoic age), is exposed
by the agency of great faults ; while, at a great number of different
points around the coast, strata of Middle and Lower Lias age are
exposed beneath the lava streams, and are seen to rest uncon=
formably upon the several paleeozoic rocks ; all these older formations
are covered unconformably at some points by Cretaceous rocks re-
presenting the Upper Greensand and Chalk.

Wherever we approach the great central eruptive masses, whether
of granite or gabbro, the whole of the stratified rocks, both of Pri-
mary and Secondary age, are found to have suffered great disturbance
and intense metamorphism—the limestones passing into saccharoid
marble, the sandstones into quartzites, and the shales into indurated
slaty rocks or the “ Lydian stone”’ of many old authors.

Lastly, as already noticed, fragments of nearly the whole of the
underlying rocks of the island occur imbedded among the ashes, la-
pilli, and scorize forming the volcanic agglomerates already noticed.

4, Sections illustrating the Structure of the Island of Mull.—In
consequence of the facility with which certain of the volcanic rocks
of Mull yield to disintegrating forces, owing to their extremely

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 245

jointed condition, some of the mountains in the island present
splendid natural sections, which afford the geologist the clearest
insight into the relations of the different rocks which constitute
them. By selecting several of these fine natural sections as types,
we shall best be able to illustrate clearly the somewhat compli-
cated relations of the different volcanic rocks of the island to one
another.

a. Beinn Greg (see Section, Pl. XXIII. fig. 2) is a mountain which
rises near the western end of Loch Bah to the height of 1941 feet.
The appearance of this mountain, as viewed from the north, is very
striking, the much-jointed granite and felsite which constitute the
greater part of its mass having crumbled down, leaving an almost
precipitous face. The lower and internal portion of the mountain,
as is well seen on its south side, in the deep ravines which divide it
from Beinn-y-chat and Beinn-a-Gobhar, is composed of a well-
formed typical granite, for the most part of the hornblendic kind.
As we ascend the mountain this granite is found to pass by in-
sensible gradations into a quartziferous felsite, the hornblende being
replaced by easily decomposable minerals, the decay of which greatly
facilitates the disintegration of the rock. Still higher the rock
becomes finely crystalline or granular, the magnesian and ferruginous
materials being apparently no longer separately crystallized, but dif-
fused through the mass as colouring-matters. The porphyritic struc-
ture is locally displayed by all portions of the mass alike, from the
coarsest granite to the finest-grained felsite.

When the northern face of this mountain is viewed from a little
distance, the whole mass presents the appearance of being made up
of a number of concentrically curved beds. This pseudo-stratified
appearance is, as all geologists are aware, very frequently presented
by masses of undoubted igneous origin*; whether in the present
example it is to be regarded as due to the manner in which the
fluid rock was extruded, or to changes which have taken place in the
mass during the process of cooling, I shall not here attempt to
determine.

The granite and felsite of Beinn Greig are traversed by in-
numerable veins. ‘These all appear to be of the ‘‘ contemporaneous ”
class, and to be composed of similar materials to those of the mass
itself, differing for the most part only in the degree of fineness of
grain, colour, &c. Usually the rock of the veins is of much finer
grain than that which it traverses; and in some analogous cases the
acid rock actually passes into the glassy condition, and exists as pitch-
stone veins traversing granite. Ina few instances thin veins of almost
pure quartz and others made up of crystallized felspar are found.

Lying upon the summit and flanks of the eruptive rocks just de-

* T may especially cite the granite of Arran, where similar features to those
described as occurring in Mull were noticed by Macculloch. This concentric
structure appears to be similar to that referred to by Mr. Scrope as occurring in
certain of the domitic puys of Auvergne, such as Clierson and Le Grand Sarcoui
(see ‘The Geology and Extinct Volcanos of Central France,’ 2nd edition, 1858,

p- 68).
Q.J.G.8. No. 119. T

246 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

seribed, are sheets of lava of the highly acid variety (felstones),
often highly vesicular and amygdaloidal in structure, which alternate
with great masses of ash, lapilli, and scoriaceous fragments. The
very rugged forms which these rocks assume in weathering causes
them to present a marked contrast with the rocks on which they
repose. They have evidently been thrust upwards to a certain
extent by the great intrusive masses below them, and are seen
dipping in both directions from them; on the west they are inter-
sected by the shores of Loch-na-Kael, where their characters can
be conveniently studied.

b. Beinn Uaig.—The relations of the intrusive granite and felsite
to the overlying felstone lavas is, owing to the inaccessibility of many
parts of the mountain, not conveniently exposed for study upon Beinn
Greig. But in Beinn Uaig we have a precisely similar section in
which the junction in question is beautifully illustrated. This sec-
tion is represented in the woodcut, fig. 1. The granite and felsite

Fig. 1.—Rocks forming the Summit of Beinn Uaig, Isle of Mull.

- x
\ Vey) me en Dek | LPL RUT.
AHR EAS pe OTE AS

a. Felstone lavas, with agglomerates.
b. Syenite-granite graduating into felsite.

are seen to give off great veins which traverse the masses of felstone
lava and volcanic agglomerate, producing a very sensible degree of
alteration in them along the surfaces of contact; and, further, these
veins are seen to include masses of the traversed rock which have
been caught up in them. In all these features we recognize the
characteristics of intrusive granitic masses, which have been so ad-
mirably illustrated in the writings of Hutton, Playfair, Webb Sey-
mour, and Macculloch.

c. Craig Craggen (Section, Pl. XXIIT. fig. 3).—The valley of the
Forsa or Pennygown river exposes on its western side a fine section
of the mountain called Craig Craggen, which rises to the height of
1885 feet. We here find the granites and felsites presenting the
same relations to the overlying felstone lavas as in the two examples
before cited. As we trace the lavas up the slopes of the mountain
the intercalated masses of agglomerate are found to become gradually
thicker, until they constitute the larger portion of the mass, though
still traversed throughout by lava sheets, and intersected by in-

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 247

numerable dykes. Associated with the agglomerates are numerous
ejected blocks; these comprise examples of nearly all the rocks
found beneath the lavas of Mull, includixg the Lower Silurian gneiss,
quartzite, mica-schist, and slate, the Newer Paleozoic breccias, con-
glomerates, sandstones, and lavas, the Liassic rocks, and the Cre-
taceous quartzuse conglomerates, grits, and sandstones.

At its northern end, as we approach the summit, the whole of
the highly siliceous rocks which compose the mass of the mountain,
granite, felsite, felstone lavas, and felspathic agglomerates, are seen
to be penetrated indiscriminately by numerous intrusive sheets or
dykes on the grandest scale, composed of various forms of gabbro,
passing into dolerite and basait, which have evidently produced a
greater or less amount of alteration at the planes of contact in the
rocks which they traverse. These great intrusive sheets and dykes
of basic igneous rock were, from their positiou, evidently at one time
connected with the great masses of gabbro forming Beinn Varnach
and the eastern portion of Beinn Tsalla, from which they are now
separated by the Forsa Glen. Owing to the manner in which these
gabbro dykes resist denuding influences their course among crumbling
rocks of the acid class can be easily traced by the geological observer.

The last point to be noticed with regard to Craig Craggen is that
the spur known as Nid-a-shoag (the Hawk’s nest) is capped by an
isolated patch of ordinary basalt, which, from its amygdaloidal
character is recognized as a portion of an old lava sheet; while,
overlapping the felstone lavas, towards the northern foot of the
mountain we find portions of the great series of basaltic lavas, which,
though cut off abruptly on the shores of the Sound of Mull, are seen
to be continued on the opposite side in Morvern. The relations of the
several masses of rock of the basic class are such as to leave no room
for doubt that the comparatively soft scoriaceous masses of this moun-
tain owe their preservation to a covering of basaltic lavas, of which
the outlier of Nid-a-shoag constitutes the last vestige ; and it seems
almost equally impossible to resist the inference that the lavas of
the great basaltic sheets were once actually continuous with the
great intrusive gabbro masses, which are composed of identical
materials with them though in a different state of aggregation.

d. Beinn More (Section, Pl. XXIII. fig. 4).—To complete our series
of illustrations of the relations which the several volcanic products
of the island of Mull bear to one another, we cannot do better than
notice this, the highest peak in the island. In the lower parts of
this mountain and the spurs around it the granites, felsites, felstone
lavas, felspathic agglomerates, with the intersecting dykes of gabbro,
dolerite, and basalt may be observed presenting the relations to one
another which have been already described. Resting unconformably
upon these is a mass, many hundreds of feet in thickness and consti-
tuting the whole of the higher portions of the mountain, composed of
basaltic scorie, tuffs, and ashes, alternating with lava sheets and in-
tersected by a plexus of dykes. By the thinning-out of the masses
of agglomerate the basaltic lavas come together, forming the great
peninsula of the Bourg or Gribun, which is made up of a sheets

i

248 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

piled on one another to the depth of more than 1600 feet. It is evi-
dent that the beds forming the summit of Beinn More, which are com-
posed of alternations of lavas and agglomerates, the latter exhibiting
in the fissures of ejected blocks so many beautiful minerals of the
same kind as are found in similar positions in connexion with ex-
isting volcanoes, constitute the last vestige of a volcanic cone
formed during the period at which the basaltic lavas were ejected.
The entire absence of ejected blocks of the stratified rocks in these
later agglomerates becomes a very significant fact when we remem-
ber that, while the earlier eruptions of acid rocks broke through
masses of the older strata, the later basaltic masses forced their way
through the midst of the former.

Although the great mass of the lavas constituting the great
plateaux are of basaltic composition, yet they vary greatly among
themselves in many minor features ; and among them are occasionally
found sheets of clinkstone*, a rock of a more or less acid or inter-
mediate composition. ‘The number of these exceptional layas is
found to greatly increase as we approach their points of origin, as in
passing up Beinn More. This circumstance (one presented in many
existing volcanoes) is easily explained by the well-known fact that
the more acid lavas as a general rule exhibit a much less perfect
fluidity than those of the bane class.

5. Proofs that the Central Mountain-group in Mull comstibutes the
Relic of a great Voleanc.—No one familiar with the phenomena of
active or recently extinct volcanoes, illustrated as these have been
by the researches of Lyell, Darwin, Von Waltershausen, Humboldt,
Von Buch, and many other investigators, but more especially by
Scrope, in his classical work on ‘The Geology and Extinct Volcanos
of Central France,’ can for one moment hesitate as to the interpreta-
tion to be given to the phenomena which I have described as being
presented by the rocks of the island cf Mull.

The group of mountains occupying the central portion of the
island is clearly the greatly denuded core of an immense volcanic
pile, the great accumulations of scorie and lavas which formed the
bulk of this mountain-mass having been to a great extent removed,
and what now remains to our study being little more than the skele-
ton or framework of the vast pile, formed by the consolidation of the
springs of liquefied rock which rose through its mass. “If,” wrote
the late Professor Jukes, “‘ we could follow any stream of lava to its
source in the bowels of the earth, we should probably find it
changing, under varying circumstances of depth and pressure, from
scorize or pumice to granite”. By the study of the relations of the
rocks of Mull the geologist is able to supply the most complete
verification of this conclusion, and indeed to illustrate every stage
of the transformation.

* he term ‘‘porphyrite” has been sometimes applied to these rocks; but
they agree in every essential respect with many of the phonolitic lavas of recent
volcanoes. Rocks precisely resembling those of Mull occur in the central masses
of Mont Dore and the Mezen in Central France.

+ Encyclopedia Britannica, Art. ‘‘ Mineralogy and Geology.”

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 249

Moreover the history of the accumulation of its materials, the
plan of its architecture, and the succession of the changes which
have taken place during its formation may be distinctly read in the
ruins of this old volcanic cone.

At the commencement of the Tertiary Period the area now con-
stituting the island of Mull formed part of an old land-surface com-
posed of many different rocks. It is not difficult to reconstruct the
geological features of the district at that period. Strata of Lauren-
tian age were covered unconformably by highly metamorphosed
Lower Silurian rocks, through the midst of which arose a great
eruptive mass of granite. Lying unconformably upon the Older
Paleozoic rocks were great masses of old lavas &e. belonging to the
Newer Paleozoic periods, while the whole was unconformably over-
lain by many patches of Jurassic strata. Lastly, overlapping the
whole of the rocks already noticed, and lying upon their upturned
and denuded edges, were a number of tracts composed of Cretaceous
rocks only recently upheaved above the sea-level.

Such were the geological features of the land upon which the
grand and long-continued series of volcanic phenomena which we
have to describe were displayed. ‘This volcanic action appears to
have commenced by the opening of vents from which violently
escaping steam hurled aloft great volumes of felspathic ashes, lapilli,
and scorie, mingled with numerous fragments torn from the rocks
through which the explosive discharges took place. These eruptions
of fragmentary matters were accompanied, at intervals, by the out-
flow of streams of acid or trachytic lavas now forming felstones.
And, further, the rising liquid mass from which these streams were
fed, to a certain extent forced upwards the rapidly accumulating
deposits of ejected materials, injecting at the same time with its
substance the numerous fissures which the expansive force could
not fail to create in the superincumbent and surrounding rocks*.
By slow consolidation under pressure the fluid matter gave rise to
the formation of masses and veins of felsite and granite. Thus
there was gradually built up a great volcanic pile, composed princi-
pally of igneous materials of the acid variety.

This period of the ejection of acid materials, which was doubt-
less one of vast duration, appears to have been followed by one
of inactivity ; for there is distinct evidence that before the second
series of eruptions took place the first-formed cone had undergone
great decay: explosions had destroyed its symmetry; subsidences
had taken place in portions of its mass; and denudation had removed
to a large extent its covering of agglomerates and lavas, exposing
the granites and felsites of its central mass.

The second period of eruption was marked by the prevalence of

* Whether some of the granitic and felsitic hills of Mull are the centres
of great masses of acid lavas, extruded in a bulky form at the surface, like the
“domitic puys” of Auvergne, the central trachytic masses of Rocca Monfina, or
the bosses in the midst of the crater of Astroni, it is not possible, in the present
degraded condition of the Mull volcano, absolutely to determine. From the
analogy of the cases we have cited and also many similar ones, however, we
may reasonably infer such to have really been the case.

250 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

materials of the basic class among the products of this volcano.
The liquefied materials of this period were forced through the
older igneous rocks, fissuring them in every direction: those fissures
which did not reach the surface would be filled by the consolidating
rock and form dykes; but, doubtless, many of these great rents
extended to the surface and then gave rise to those eruptions pro-
ducing the “parasitic cones” so common on the flanks of many great
volcanoes. The lavas of this period being possessed of so much
greater fluidity, instead of being confined, like those of older date,
to the neighbourhood of the volcano, flowed to enormous distances ;
and so vast were the volumes of material discharged from the
voleano, that the lava streams accumulated to a thickness of 2000
feet. That the formation of deposits of such vast thickness occu-
pied periods of enormous duration we may reasonably infer; and
the conclusion is confirmed by the proofs of intereruptive denuda-
tion among these old lavas, and by the existence of old soils, forest-
growths, river-gravels, and lacustrine deposits between them.

At the close of this second peried of eruption the volcano of Mull
became extinct, and has gradually fallen into that state of decay and
ruin which it now presents. But, as we shall hereafter show,
volcanic activity in the district did not cease with the extinction of
this and the other similar great volcanic mountains of the period.

Having demonstrated the real nature of the central mountain-
group in the island of Mull (namely, that it constitutes the basal
wreck of a great volcanic pile), I shall proceed to show that, at a
number of other points in the same district, there occur similar
masses of eruptive rocks associated with voleanic agglomerates and
breccias, and that these are the relics of other volcanoes, similar to,
and contemporaneous with, that of Mull. After the details into
which I have entered with regard to this last-mentioned example,
it will not be necessary for me to do more, in describing the other
similar instances, than to poimt out in a general manner the
analogies and variations of their structure as compared with that
which we have chosen as the type.

6. The Volcano of Ardnamurchan.—Immediately to the north-
ward of the island of Mull stretches the peninsula of Ardnamurchan,
which forms the most western portion of the mainland of Scotland.
In this district we have exhibited to us, though in a much more
fragmentary condition, a precisely similar series of rocks, presenting
relations with one another identical with those described in the
island of Mull.

Constituting the peak of Meal-nan-con and a number of adjoining
heights, which stretch in a belt south-westward to near Kalchoan,
we find great masses of intrusive felsite presenting many local
variations, but as a whole exhibiting a striking identity with the
great felsitic rocks of Mull, like which they, in places, assume a.
distinctly granitic character. Lying upon the eastern and southern
flanks of these felsites and penetrated by numerous veins which
proceed from them are numerous sheets of felstone, often of
amygdaloidal structure. These are interstratified with beds of ash

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 251

and scorie, enclosing fragments of the Lower Silurian and Jurassic
rocks which constitute the foundation on which the volcanic rocks
of Ardnamurchan are piled.

The whole of the western extremity of the peninsula of Ardna-
murchan is constituted by masses composed of many varieties of
gabbro, which form a number of wild and barren mountains. At
the western extremity of the peninsula the pyroxenic ingredient of
the gabbro is either diallage or hypersthene; but as we go east-
ward these are found to be replaced in varying degrees by other
members of the pyroxene family, and the gabbro graduates into
dolerite.

Of the intrusive character of this great mass of highly crystalline
basic rocks, and its posteriority alike to the Secondary strata and
the felstone lavas and felsites, we have the most ample proof.
Along the southern side of the peninsula, on either side of the cape
called Stron Beg, the strata of Lias and Inferior Oolite are seen to
be upheaved at very high angles on the flanks of the masses of
gabbro; and as we approach the latter the sedimentary deposits
are found to undergo the most wonderful metamorphism; sand-
stones are seen passing into quartzites, limestones into marble, and
clay into hard shale, slaty rock and ‘“‘ Lydian stone.” Not less striking
are the effects produced by the gabbros on the earlier volcanic rocks,
the felstone lavas which overlie the Secondary strata. In proxi-
mity to the gabbros, these felstone lavas are seen to have been up-
heaved at high angles and to have acquired a peculiar platy structure
and splintery fracture, combined in many cases with the development
of a probably preexisting banded coloration. Even the crystalline
felsites have not escaped tne general metamorphism, but exhibit
peculiarities of texture and fracture near their junction with the
gabbros.

Secondary strata, felstone lavas, and felsite masses are alike
penetrated in all directions by innumerable veins, dykes, and sheets
of gabbro, dolerite, and basalt ; while, to the eastward, the vestiges
of old basaltie lava-flows are seen resting indifferently upon the
Lower Silurian gneiss, the Secondary strata, and the felspathie
lavas of the earlier period of eruption.

Here then, in spite of the more fragmentary character which the
relics have assumed in consequence of the enormous amount of
denudation which they have suffered, their analogy with the rocks
of Mull is unmistakable and points to precisely to the same suc-
cession of events. It is evident that in Ardnamurchan we ‘have
preserved the “sector” of a great circular volcanic pile, which, like
that of Mull, consisted of an older interior mass composed of
igneous rocks of the acid class, intrusive, laval, and fragmentary,
which mass was pierced and enveloped by the basic products of a
second period of eruption.

The manner in which, by means of denudation, the internal
structure of the volcanic masses of Ardnamurchan has been laid
bare to our view in sea-cliffs and deep ravines, renders this frag-
ment of a volcano of special interest to the geologist, Here he is

252 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

enabled to study in detail the results of the potent subterranean
actions which went on side by side with those subaerial phenomena
with which our ideas of volcanic activity are usually associated, but
which do not constitute the only, nor perhaps even the principal
- effects of these grand outbursts. To the consideration of these
subterranean phenomena I shall have to return in the sequel.

7. The Volcano of Rum.—tying to the north-west of the penin-
sula of Ardnamurchan we find the group of the Small Isles, comprising
Rum, Canna, Kigg, and Muck, with several smaller islets. The
whole of these, with the exception of the first mentioned, are
evidently isolated fragments of a plateau composed of basaltic lava-
sheets, which are in some places seen to rest unconformably upon
various Secondary strata. In the island of Rum, however, we find a
great mass of eruptive crystalline rocks which have burst through
the older strata; and the various igneous products, in their nature,
positions, and relations, present such remarkable analogies with
those of Mull and Ardnamurchan as to leave no room for doubt
that in Rum also we have the basal relics of a great volcanic
mountain (Section, Pl. XXIII. fig. 5).

Nothing can be more instructive than the admirable manner in
which, through denudation, the mutual relations of the stratified
and eruptive rocks are displayed in the island of Rum. The former
consist of Cambrian sandstone with some masses of gneissose,
schistose, and quartzose rocks, probably of Lower Silurian age;
these occupy the north-west, north-east, and south-east sides of the
island, while its interior is formed by masses of eruptive rocks,
including both granites and gabbros, which rise into lofty moun-
tains. At a distance from the igneous rocks, the Paleozoic strata
exhibit their usual mineralogical characters and their normal strike
and dip; but wherever they approach the great central intrusive
masses they are found to be violently upheaved and often much
contorted, and are further seen to have undergone a striking meta-
morphism, which often extends to a cousiderable distance from the
igneous rocks; the more argillaceous or felspathic portions of the
Cambrian sandstone pass into a highly micaceous black schist,
and the more siliceous portions into an impure quartzite. The
whole of the stratified rocks surrounding the intrusive masses of
granite and gabbro are intersected in all directions by innumerable
veins and dykes.

Let us now turn our attention to the igneous masses which form
the central group of mountains, and upon the flanks of which the
Primary strata lie on three sides of the island, haying evidently
been removed by denudation on the fourth side. These central
mountains of Rum, as is evident to the most casual observer, are
composed of two very distinct kinds of rocks, the contrast between
which, owing to their different modes of weathering, is extremely
striking.

To the westward, and forming the smooth dome-shaped mountain
mass of Oreval (1872 feet) and its dependencies, is a great develop-
ment of much-jointed highly felspathic rocks, consisting of granite

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. DES

passing by every gradation into felsite. These rocks are identical
in all their characters with the acid rocks of Mull, They occur also
in the south-east of the island of Rum, being there, however, much
broken up by numerous sheets and veins of the crystalline basic
rocks.

The north-eastern mountains of the central group rise into
singularly wild and rugged peaks, among which are Haskeval
(2667 feet), Halival (2367 feet), Sctir-na-gilean (2553 feet), and
Beinn More (2505 feet). They are composed of the varieties of
gabbro in which the pyroxenic constituent consists of several
varieties of the augite group, but in which diallage and hypersthene
only rarely occur. Precisely similar rocks are found, as we have
seen, forming the outer portions of the great basic intrusive masses
and the veins which proceed from them, both in Mull and Ardna-
murchan. The great masses of gabbro in Rum often exhibit that
pseudo-stratification so often observed in igneous rocks, and of
which we have noticed such a remarkuble example in the granites
and felsites of Mull. They are also penetrated by numerous “ con-
temporaneous veins.”

The relations between the eruptive masses of the acid and basic
classes respectively are scarcely less strikingly displayed in Rum
than in Mull; we find numerous offshoots and veins proceeding from
the gabbros and traversing the granites and felsites, as well as the
surrounding older strata and the agglomerates and lavas.

Thus the island of Rum is for the most part made up of great
intrusive masses, and of the more or less disturbed and altered sedi-
mentary deposits through which these have been thrust. Of the
overlying ejected materials (lavas, scoriz, ashes, and breccias) only
comparatively small and isolated vestiges remain; owing, however,
to the clear manner in which the relations of these to the other
rocks are displayed, they are well worthy of attentive study by the
geologist.

On the northern flanks of the mountain of Halival we find a great
mass of felstone lavas intermingled with agglomerates. But asso-
ciated with these are deposits of enormous thickness, consisting of
blocks of Cambrian sandstone of all sizes imbedded in a matrix of
felspathic ash. The included blocks of sandstone in this volcanic
breccia vary greatly in size, from masses several feet in diameter
down to the smallest fragments. As we approach the gabbro masses
which form the bulk of the mountain, the felstone lavas are seen to
undergo analogous changes to those manifested in similar positions
by the same rocks in Ardnamurchan and Mull.

Other patches of felstone lavas with accompanying agglomerates,
which have escaped destruction by denudation, occur at several
points on the south-eastern side of the island. On the western side,
however, there are several small outlers of the ordinary columnar
and basaltic lavas. These exhibit all the usual characters of the
sheets forming the great plateau, and were doubtless once continuous
with the other and larger outliers of the same rocks, which con-
stitute the islands of Canna, Eigg, and Muck. At Scir More, or

954 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

Bloodstone Hill, these rocks present some features of interest to
the mineralogist, owing to the beautiful specimens of heliotrope and
other minerals which are found in their amygdaloidal cavities.

But to the geologist the fact of greatest interest in connexion with
these basaltic lavas is that they rest directly upon the granite and
the Cambrian sandstone, thus affording a most striking proof of the
unconformity which exists between the products of the two great
periods of volcanie eruption, and an indication that a vast interval
must have elapsed between them. This is plain from the fact, of
which we have clear evidence, that rocks of the acid class were
much wasted by denudation before being covered up by the ejection
of basaltic materials.

It is unnecessary for me to remark on the striking manner in
which the peculiar features and relations of the rocks of Mull and
Ardnamurchan are, in every essential detail, repeated in those of
Rum ; nor is it necessary to insist upon the conclusion that, like the
examples before described, these rocks must be regarded as the
relics of a great volcanic pile, and the products of a similar succession
of eruptions.

8. The Volcano of Skye.—It is doubtless a misfortune that the
attempts to unravel the somewhat complicated phenomena presented
by the. igneous rocks of the Hebrides have hitherto, for the most
part, been made in the most frequently visited island of the group,
Skye. For, although the relations of the rocks of this island are suffi-
eiently clear, when viewed in connexion with the analogies presented
by the other islands; yet, considered alone, they are by no means
obvious. It is not surprising therefore that observers, unacquainted
with the clearer sections of Mull, Ardnamurchan, and Rum, should
have been sometimes led to conflicting conclusions upon the subject.

I will in the first imstance point out briefly in how exact a
manner all the essential and distinctive features of the old voleanoes
already described are repeated in that portion of the island of Skye
which is included between Loeh Shgachan and Broadford Bay on
the north, and Lochs Brettle and Eichart on the south.

Nearly the whole of this district is composed of intrusive rocks
which have burst through strata of Paleeozoic (Cambrian and Lower
Silurian) and Mesozoic¢ (Liassic and Oolitic) age: these are seen at
a number of points lying in a greatly disturbed and hightly meta-
morphosed condition upon the flanks of the igneous masses.

These igneous intrusive masses consist of two different kinds of
crystalline rocks, granite and gabbro; and the contrasts between
their modes of weathering are exhibited in their most exaggerated
form in the island of Skye. The granitic masses of the Red
Mountains, culminating in Beinn Glamaig (2670 feet) and Beinn-na-
Cailleach (2385 feet), are as remarkable for their strikingly smooth
pyramidal forms as the gabbro masses of the Cuchullin Hills and
Beinn Blabheinn (both exceeding 3000 feet in height) are un-
rivalled for their wild, jagged, and fantastic outlines*. The granites
of the Red Mountains graduate into many varieties of felsitic rocks,

* Vide Geikie, ‘Scenery of Scotland,’ pls. 2 & 4.

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 255

and are traversed by numerous “ contemporaneous veins,” usually of
similar composition to the mass which they traverse ; the gabbros of
the Cuchullin Hills and Blabheinn pass insensibly into dolerites and
basalts, and are also traversed by many “‘ contemporaneous veins.”
Both of these intrusive rocks, in places, exhibit the pseudo-stratified
appearances so commonly displayed by igneous masses.

That the rocks of basic composition were ejected subsequently to
those of the acid variety, is illustrated by sections at a number of
different points; and this fact was deteeted by the late Principal
J. D. Forbes * in 1845, and by Professor Zirkel+ in 1868.

As in Rum, the ejected rocks, lavas, agglomerates, and breccias
form only disconnected outliers in Skye ; but some of these present
very interesting characters. Patches of volcanic agglomerates and
breccias, including fragments of the several Primary and Secondary
rocks through which this volcano has burst, are found at several
points: the preservation of these is probably due to local subsidences;
and a good example of them may be examined in the hill on the
south of Beinn Dearg, known as Cnoc-na-Fitheach. Outhers of
the felstone lavas, more or less mingled with scorie and ashes, occur
at a number of points round the central mountain-group, and. in,
the island of Scalpa and the peninsula of Strathaird are seen lying
directly upon the various Primary and Secondary strata.

Of the great basaltic plateaux which doubtless once surrounded,
the old volcano of Skye on all sides, we have also only a number of
fragments lett. By far the largest and most important of these is.
the one which, stretching north-westwards, constitutes the great
peninsulas of Trotternish, Vaternish, Durinish, and Minginish. ‘This.
mass of lavas has been upheaved on its eastern side so as to display
in a very interesting manner the Secondary rocks upon which it
lies. Another small outlher of the basaltic plateaux, which has,
escaped from destruction by denudation, oecurs in the adjacent
island of Raasay, where it forms the highest point, the striking hill
of Dun Can. In Applecross, however, denudation has proceeded;
one step further, the lava streams having been altogether removed,
and thus the Seeondary rocks with their included sheets of intrusive
igneous rock exposed at the surface. Further, for great dis-
tances round the central mountain-group of Skye tbe whole of the
stratified rocks are penetrated by innumerable dykes and intrusive
masses both of acid and basic composition.

9. The Volcano of St. Kilda.—I have not yet been able to visit
this remote group of islands; but from the descriptions of Martin
and Macceulloch it is sufficiently obvious that we have in them the
relics of another of the old Tertiary volcanoes. The islands are said
to consist wholly of ‘‘trap-rocks ;” and the central and largest of the
group is described as being formed in its eastern part of granite, and
in its western part of gabbro. Macculloch, who was so competent
a judge of petrological characters, informs us that these rocks were
identical in every respect with those found in the island of Rum.

* Hdinb. New Phil. Journ. New Ser. vol. xl. (1845-46), p. 86.
+ Zeitschr d. deutschen geologischen Gesellschaft, Jahrg. 1871, p. 90.

256 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

10. Comparison of the great Tertiary Volcanoes.—lt has been
shown in the preceding pages that we have evidence of the exist-
ence during the Tertiary period of five great centres of volcanic
action in the northern portion of the Hebrides—and that at each
of these points a great volcanic mountain once existed, the frag-
mentary, but unmistakable, relics of which still remain for our
study. A problem which, however, will naturally suggest itself
to every mind is that of the causes which have led to the very dif-
ferent states of preservation of volcanoes which appear to have been
formed contemporaneously. The portion of this inquiry which is of
more especial interest to the geologist is that concerning the reasons
of that comparatively. perfect state of preservation exhibited by the
volcano of Mull which renders it so admirable a key to the interpre-
tation of the whole series of phenomena presented by the Western
Isles of Scotland.

Fortunately the causes of the different extent of denudation in
the several cases, and of the exceptional state of preservation of
the volcano of Mull, are sufficiently obvious; and as the considera-
tion of these leads us to some highly interesting and important con-
clusions, I shall now proceed to detail them.

If we compare the volcanoes of Mull and Skye, the differences
alluded to, in spite of the identity of the characters and relations of
the various rocks which compose them, are very striking. Thus
while in the former case we observe the most wonderful interlacing
of the two kinds of eruptive rock, and the outliers of the great super-
posed piles of fragmentary materials are still clearly to be traced, in
the latter case the two great eruptive masses rise side by side, with
but comparatively few examples of offshoots from the younger into
the older series, and the lavas, agglomerates, and breccias which
formed the mass of the volcano have almost wholly disappeared.

A careful study of the rocks of Mull shows the cause of this dif-
ference to be that the volcanic pile in this island has suffered far less
from denudation than in Skye, and that in fact the existing surface
in the former island presents a cross section of a voleano taken at a
higher level than in the latter. Further, it can be shown that this
difference is due to a remarkable central subsidence which has taken
place in the case of the Mull volcano.

When we examine the fine sections of the lavas of the basaltic
plateaux which are exposed along the shores of the deep fiords of
Loch Seridain and Loch-na-Kael, we find that the lava sheets, in-
stead of sloping away from the great central masses of eruptive rock,
actually for many miles around dip towards them at angles varying
from 2° or 3° up to 5°, the inclination increasing as we approach the
volcano. Now, as we are certain that this was not the original po-
sition of the lava streams, we are led to the conclusion that a subsi-
dence has taken place in the great central mass of the island, a view
which is confirmed by an examination of the inclination of the whole
of the lavas surrounding the volcano. Still further support to the
inference is afforded by the fact that at certain points, as in the valley
between Kilfinichan and Gribun, where the Cretaceous rocks under-

J. W.JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 257

lying the lavas are suddenly cut off, we have clear evidence of the
existence of faults the downthrow of which is in all cases towards
the great central mass.

By a study of the section of Craig Craggen in connexion with
those exposed at some other points, we are led to infer that a similar
subsidence took place after the period of the eruption of the acid
lavas, and before the outpouring of the basaltic lavas from the same
centres commenced.

Now it is a most interesting fact that precisely similar subsidences
beneath cones of comparatively recent date have been shown to exist
in the case of a volcano in the Cape-Verde Islands by Mr. Darwin*,
and in one in New Zealand by Mr. Heaphyt. The phenomena
described in these two modern examples are perfectly analogous to
those presented in the ancient volcano of Mull; indeed Mr. Scrope
has, in reasoning on the causes of the phenomenon, speculated on
the probability of its very general occurrence to a greater or less
extent in the case of many volcanoes.

According to the observations of Krug von Nidda, Iceland exhi-
bits a precisely similar subsidence of the great volcanic centres to
that which we have been describing in Mull, and on even a grander
scale §.

From the relative positions of the central masses of granite and
gabbro in the great Tertiary volcanoes of the Hebrides we are led
to infer that changes must have taken place in the position of the
axes of eruption of several of them in the same manner as has been
so admirably illustrated in the case of Etna by Sir Charles Lyell
(vide Phil. Trans. for 1858, pp. 738-744, and ‘ Principles of Geology’
1ith ed. vol. 1. pp. 9-14).

With the consideration of the causes of the phenomenon of the
central subsidence of volcanic piles, however, we are not here con-
cerned ; but its effect in the island of Mull, where it is exhibited
in so marked a manner, has been the exceptional preservation of a
series of rocks which afford us the clearest insight into many important
matters connected with the history of these ancient volcanoes.

I have already had occasion to allude incidentally to the wild and
rugged character of the scenery to which the basic crystalline rocks

* Volcanic Islands, p. 9.

+ Quart. Journ. Geol. Soc. vol. xvi. (1860) p. 244.

+ Volcanos, 2nd edit. (1872) p. 225.

§ Karsten’s Archiv, vil. (1834) pp. 247-284. Recent studies of the positions
and relations of the older volcanic rocks of the Lipari Islands have convinced
me that similar central subsidences must have taken place in connexion with
the great volcanic piles of that area. It seems not unreasonable to suppose
that the deflection of the hquefied matter below a volcanic centre to new vents
may very generally give rise to a certain amount of subsidence in the masses
of ejected materials which, often to a very great height, have been piled about
it; the great weight of the latter would assist in producing the same result.
Of course such gradual and comparatively slight subsidences will not be
confounded with, or thought to lend any countenance to, the theory of vio-
lent engulfment of the central parts of volcanic mountains, to which cause
some geologists have felt disposed to refer the formation of their great cra-
teral hollows.

258 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

of the Western Isles give rise, owing to their peculiar modes of
weathering. Acted upon very slowly by the agents of atmospheric
disintegration, the rock-surfaces assume a remarkable roughness
owing to the persistence of the crystals of diallage and augite, ac-
quiring at the same time a deep brown tint from ‘the peroxidation of
the iron; and, moreover, these surfaces, not giving rise to the form-
ation of soil, are altogether destitute of any covering of vegetation.
The lower flanks of the mountains are marked by the dome-shaped
forms resulting from the passage over them of glaciers; and the
grooving and striation are wonderfully preserved upon them*; but
the summits of the same mountains stand up in the form of wild
craggy pinnacles, many of which are altogether inaccessible. More-
over the striking features of these ragged and barren rocks are height-
ened by the contrasts which they present with the verdant, terraced
slopes of the basaltic plateaux on the one hand, and the smooth,
dome-shaped and débris-covered hills of granite on the other.

These features of the gabbro rocks are displayed in a sufficiently
striking manner in Mull, especially in the wild glen between Beinn
Buy and Creach Beinn, and also in Ardnamurchan and Rum; but
in each of these cases the entangled portions of acid rocks give rise
to the formation ef streams and taluses of detritus and the conse-
quent production of soil and vegetation, relieving the monotonous
barrenness of the surfaces of these rocks. But in the island of
Skye we have a considerable area which is formed almost exclu-
sively of the gabbro rocks; and the wonderful features of the
scenery to which these give rise are familiar to ail who have
visited the Cuchullin Hills and Blabheinn, being most remarkably
displayed in the sombre glen which contains the “dark loch” of
Coruiskh.

The deep impression made by this remarkable scene on the mind
of Sir Walter Scott is recorded in his well-known lines in the ‘ Lord
of the Isles ’:—

“ Rarely human eye has known
A scene so stern as that dread lake,
With its dark ledge of barren stone.
Seems that primeval earthquake’s sway
Has rent a strange and shattered way
Through the rude bosom of the hill,
And that each naked precipice,
Sable ravine, and dark abyss,
Tells of the outrage still.”

But for the geologist this justly celebrated, and now tourist-
haunted, locality may claim a more especial interest. For if his
study of the rocks dispels, as it certainly will do, the poet’s beautiful
dream concerning their origin, it will be only to replace it by a far
grander conception ; and as he stands here, in the very centre of
the cooled reservoir of an ancient volcano, his instructed imagination
will revel in the reconstruction from reliable data of the wild and
nes features cf this old “‘ Tierra del Fuego.’

Vide J. D. Forbes, Edinb. New Phil. Journ. New Ser. vol. 4 pp. 90-99
(1s45-46)

J. W. JUDD ON THE SECONDARY BUCKS OF SCOTLAND. 259

11. Dimensions of the great Tertiary Voleanoes.—It might at first

sight appear impossible to arrive at any sound conclusions upon this
question, concerning, as it does, the physical features of a period so
remote from the present. Nevertheless, as I hope to be able to
show, so far is this from being the case, that we have several different
kinds of data, foremost among which must be ranked those derived
from the subsidence which we have shown to have taken place in
the case of the Mull volcano, enabling us to arrive at least at an
approximate estimate.
_ Weare able in the case of each of the great Tertiary volcanoes te
to arrive at some idea concerning the areas covered by their bases.
The base of the voleano of Mull must have had a circumference of
at least forty miles; Etna, which has a greatly truncated form,
nevertheless rises to the height of 10,900 feet from a base only thirty
miles in circumference*; a similar relation between the base and
altitude of the great volcanoes of Sicily and Mull would lead us to
infer that the elevation of the latter was at least 14,500 feet.

By a careful study of the present inclinations of the lava streams
which flowed from the Mull voleano, we may obtain another approx-
imation to the minimum possible elevation of its summit. From an
examination of all the data, especially those furnished by the sec-
tions along the shores of Loch Scridain and Loch-na- Kael, and care-
fully availing myself of every check in making the calculations, I find
that, if the rocks were restored to the positions which they occupied
before the great central subsidence took place, the present summit
of Beinn More, which is 3172 feet above the sea, would be raised to
to an elevation of at least 6000 feet—and, further, that a central
cone reconstructed on the basis thus. obtained would have an eleva-
tion of more than 10,000 feet.

But this must be regarded as only the lowest possible estimate; for
in making it I have totally neglected four different considerations,
all tending to increase our estimate, but by what cxact amounts it is
impossible to state. These are :—

(1) The gradual increase of the inclination of the lava-beds on
the higher slopes of the mountain.

(2) The effect of the great faults before alluded to as contributing
to the central subsidence, the influence of which, though not exactly
determinable, was certainly very considerable.

(3) The originally greater elevation of the surfaces on which the
volcanic rocks lie. This is proved by the fact that many rocks which
were certainly of terrestrial origin are now below the sea-level.

(4) The great masses of agglomerates and lavas which originally
formed the summit and flanks of the mountain, but have now been
removed by denudation.

In speaking, however, of the height of a volcano it must be re-
membered that we are dealing with a quantity subject to constant
and wide variations. During a single paroxysm, hundreds, nay,
even thousands, of feet of materials may be blown from the sum-

* In this estimate I of course only include the mass of rocks constituting the
mountain proper, and not the lava-covered plains around it.

260 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

mit of a volcano, and in a few days, weeks, or months a new
cone of different form and proportions built up. Such changes
doubtless occurred over and over again in the case of the ancient
volcanoes of Scotland as well as in those of more recent date.

From the area which its relics occupy, we may conclude that
the volcano of Skye was not inferior in its dimensions to that of
Mull; and on similar grounds we are led to infer that the voleanoes
of Rum, Ardnamurchan, and St. Kilda, though somewhat smaller
than these, were nevertheless mountains of great extent and
elevation.

Thus we are led to the conclusion that during the Tertiary
epoch there existed in the north-western part of the British
Islands a range of voleanic mountains which were on the grandest
scale—a fact which all who consider the wonderful extent and
thickness of their accumulated lava streams will be fully prepared
to admit. 3

12. Series of later Volcanic Eruptions in the Hebrides resulting
in the formation of ‘Puys.”—I shall now proceed to show that,
subsequently both to the period of the eruption of the acid lavas and
to that of the ejection of basic rocks, and at a time when the great
volcanoes from which these flowed had become extinct and their
products greatly denuded, another series of voleanie outbursts of a
sporadic character took place within the district, and gave rise to the
formation of numerous smaller cones with their accompanying lava
streams. ‘

The researches of Mr. Scrope in Central France, of Mr. Darwin
in the volcanic islands of the Atlantic, of Hamilton and Strickland
in Asia Minor, with others that might be cited, have sufficiently
demonstrated that the extinction of great volcanoes is in many (if
not in all) cases followed by eruptions on a minor scale, which burst
out in the plains at their base and give rise to the formation of nu-
merous small cones, usually arranged along what appear to be lines
of fissure. It would seem that the volcanic forces having to a
great extent expended themselves, and being insufficient to raise
columns of lava to the summits of the mountains which successive
eruptions have continually elevated, have nevertheless been able,
before sinking into absolute quiescence, to open new vents at lower
levels. To the class of smaller volcanoes thus produced it will be
convenient to apply the general term ‘‘puys,”* from the name by

* Tt is of course impossible to draw any absolute line of demarcation between
the class of subsidiary or parasitical cones which are so abundant on the flanks
of many great volcanoes (on Etna there are said to be no less than 800 of them)
and that of the ‘‘ puys,” or cones thrown up in the plains in their vicinity. As
is so well shown by Mr. Scrope, in his work on Central France, the formation
of such puys in a district may be going on during periods of vast duration, and
the several eruptions to which they owe their origin may be separated by very
wide intervals of time. They appear, however, in almost all cases to be of sub-
sequent date to the extinction of the great central volcanic mountains, and some-
times, indeed (as in the case of the Puy de Tartaret in the Mont Dore), to have
been formed in the midst of the ruinous masses of rocks to which such great

volcanoes have been reduced by denudation. The course of events in connexion
with the history of.the decline and decay of a great volcano would appear to be

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 261

which they are known in the district rendered classical by the ad-
mirable researches and descriptions of Mr. Scrope. I shall now
proceed to describe the remarkably interesting evidences which re-
main of the former existence of such “ puys”’ in connexion with the
great volcanoes of the Hebrides.

In the peninsula of Ardnamurchan the highest mountain, Beinn
Shiant, rises to the height of 1759 feet. Its very peculiar outlines (see
woodcuts figs. 2 & 3, p. 262), exhibiting neither the terraced slopes
of the masses composed of superposed lava streams, nor the pecu-
lar contours of either of the two classes of eruptive rocks, are alone
sufficient to arrest the attention of the geologist and to lead him to
its careful examination. And such study is amply rewarded. The
mountain, which exhibits a series of grassy slopes reaching almost
to its summit, is remarkable for the far-stretching and boldly
shaped spurs which on all sides branch out from a common centre.
On a nearer approach we find the cause of these striking features to
be that each of the spurs is capped by masses of lava usually exhi-
biting beautifully columnar forms, while the green slopes below are
composed of softer and more easily weathered rocks. The moun-
tain rises near the line of junction of the older felspathic and
basaltic lavas before described.

The lavas of Beinn Shiant are intermediate in character to the
acid and basic rocks which constitute the masses of the older and great
central volcanoes ; they present distinctive features by which they are
easily recognizable, and are remarkable alike for their great hard-
ness and their power of resisting atmospheric denudation. They
constitute rocks of great beauty, varying from aggregates of crystals
of glassy felspar, through numerous porphyritic and highly crystal-
line varieties to compact felstones, which finally pass into pitch-
stones of more or less porphyritic character, these last being some-
times identical in their mineralogical features and the peculiarities
of columnar structure with the rock of the well-known Scur of
Eigg. These lavas appear to be very similar in character to a
dark-coloured porphyritie trachyte which occurs at Mont Dore les
Bains in Auvergne*.

generally as follows :—By constant ejections from its summit a voleano is con-
tinually increasing its height and adding to the length of the column through
which fluid materials must be raised in order to produce a central eruption.
In the course of this action a point will certainly be arrived at when it becomes
easier for the subterranean forces to rend new fissures in the flanks of the
mountain than to raise a column of lava to its elevated summit. At this stage
of its history Etna now appears to have arrived, the formation of new para-
sitical cones on its flanks alternating with comparatively feeble ejections from
its summit. But, as the bulk and solidity of a volcanic cone are continually
added to by these operations, the subterranean forces will be gradually forced
to seek new vents at lower levels in the plains around the mountain, and
thus give rise to the formation of “puys.” In Ischia, the Lipari Islands, and
many other volcanic districts, we have admirable opportunities afforded to us
for tracing the sequence of this very interesting series of operations.

* In the islands of Lipari and Salina we have many beautiful illustrations of
lava streams, composed of dark-coloured, often nearly black, trachyte, with
disseminated crystals of sanidine. These rocks are precisely similar to those

Q.J.G.8. No. 119. U

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

262

Soa x \

a

“JO2F 6SLT

squint NT

—

ie

“A 92 WoLf waas sv “unYydunUDUpLp we QuoY wnag fo au177InND—’E *BI
M 74, Y ply Ui US Wag INO —~E OUT

Ui}

“9293 6SL1
“MN N 242 wont Uaas SD “UDYIMNUDUPA Ur ‘qunry ¥ UUW fo IUYINGD —G

a |

hil}
a ail(tl Tiaras
\ «ll Pe Sey BZ

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 263

The central and apical portion of the mountain of Beinn Shiant is
formed of rocks of similar petrological character, which, however, are
apparently of intrusive origin. On the northern side of the moun-
tain these rocks form a lofty and precipitous ridge.

The masses of columnar rock evidently constitute portions of a
number of lava streams which appear to have been originally in
connexion with the central mass, and to have diverged in a radial
manner from it. Some of these lava streams can be traced to very
considerable distances.

Where the grassy slopes below the lava cappings of the spurs are
cut through by streams, they are seen to be composed of volcanic ag-
glomerates and breccias. These contain numerous angular frag-
ments of both the felspathic and basaltic lavas of earlier Tertiary
periods with some of the underlying Lower Silurian gneiss, such
being the rocks upon which the mass of Beinn Shiant is built up ;
mingled with these are numerous fragments of felspathic scoriz and
ashes, with many fractured and worn crystals of glassy felspar in-
cluded in the mass*.

Fortunately for the geologist, one of the spurs of Beinn Shiant is
cut across in a sea-cliff forming the headland of Srone More or
Maclean’s Nose; and here we have a magnificent display of the
wonderfully interesting features in the rock-masses composing this
remarkable mountain. The upper part of Srone More, which rises
to a height of 1050 feet, is composed of the columnar lavas already
described ; but the almost precipitous clitfs below are seen to be
made up of nearly angular blocks of all sizes, up to 6 or 8 feet in
diameter, heaped together in the wildest confusion, and presenting
no appearance whatever of stratification or of the sorting of the ma-
terials according to their specific gravity.

Lastly, it is necessary to notice that the great masses of agglome-
rate which constitute so large a portion of Beinn Shiant, together
with the older lavas and gneiss on which they are piled, are tra-
versed in every direction by dykes and veins of every size, composed
of similar rocks to those forming the lava streams—namely, trachyte
and “ pitchstone-porphyry.”

To any one familiar with the characters presented by Arthur’s

of Beinn Shiant, and like them graduate, in consequence of the matrix in which
the felspar crystals are imbedded acquiring a glassy texture, into “ pitchstone-
porphyry.” In speaking of the Hocene acid lavas of the Hebrides, I have
preferred to apply to them the term ‘‘felstone,” although it must be re-
membered that they present no essential points of distinction from recent
quartz-trachytes, either in chemical composition or mineralogical structure.
The later lavas of Beinn Shiant and the Scur of Higg present still fewer points
of difference from those of active volcanoes; and it would perhaps be more
correct to speak of them as “ trachytes” than as ‘“‘ porphyrites.”

* The frequency with which crystals, often more or less worn and fractured,
of various volcanic minerals, such as augite, biotite, the felspars, &., are thrown
out from the vents of volcanoes has been noticed by Mr. Scrope. Such crystals
are found alike in the recent ejections of Stromboli, and imbedded in the tuffs
connected with extinct voleanoes at Vulcano, Albano, Bracciano, Rocca Mon-
fina, and some of the puys of Auvergne.

v2

a

264 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

Seat, the volcanic origin of which has been so admirably illustrated
by Charles Maclaren, Kdward Forbes, and Archibald Geikie, the
similarity of the phenomena presented by Beinn Shiant will be suf-
ficiently striking; nor will he fail to recognize that in either case
we have a much-denuded and ruined voleanic cone submitted to our
study. The voleano of Beinn Shiant, however, was of far larger
dimensions than that of which the ruins constitute Arthur’s Seat.

Although the lava streams of Beinn Shiant are now, by denuda-
tion, reduced to more or less isolated fragments, yet such is their di-
stinctive character that I believe there will be little difficulty, when
the country is accurately mapped, in restoring the main features of
this voleanic cone and of the lava streams which issued from it. Of
the posteriority in date of this volcanic pile to both the felspathic
and basaltic lavas of the great central volcanoes, we have the
clearest evidence; for not only do its products overlie the sheets
forming the great plateaux, but the fragments of the latter are in-
cluded in great numbers in the agglomerates which compose it.
Further, it is clear that these earlier lavas had undergone a vast
amount of denudation before the outflow of those from Beinn Shiant ;
for the latter overlap the older basalts, and in places, where these
have been wholly removed by erosion, rest directly upon the bared
surfaces of Lower Silurian gneiss.

If in the case of Beinn Shiant we are unable to trace the igneous
masses of this volcanic cone penetrating the older lavas and gneiss,
this is solely due, as in the case of recent volcanoes, to the covering
of ejected materials wholly concealing the relations of the rocks
which lie below them. Fortunately, however, many examples of
analogous later eruptions which have taken place within this district
occur, wherein, as a result of the removal by denudation of the
materials of the volcanic pile, the relations between the older lavas
and the newer intrusive masses are very clearly exhibited. One of
the most interesting of these we shall now proceed to describe.

At a distance of three miles to the south-west of the village of
Tobermory in the island of Mull there rises, in the midst of the
great plateau of basaltic lavas, a hill presenting somewhat striking
features, known as Sarsta Beinn (see woodcut, fig. 4). Its height

Fig. 4.—View of Sarsta Beinn from Stot Hill.

ww
pees
ay Yn 7

above the sea-level is apparently about 800 or 900 feet; but the
peculiarity of the mode of weathering of the rock masses which
compose it, as compared with the surrounding tabular basalts,

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 265

makes it a rather conspicuous object. Its summit is occupied by a
small loch, or mountain-tarn, which is said to be of great depth.

When we examine this rocky mass standing up so abruptly in
the midst of the basaltic plateau, we find it to be composed of a
dolerite of the coarsest grain, passing in its lower portions into gabbro.
The rugged, rusty-brown surfaces of these rocks, which resist denu-
dation and the growth of vegetation, form a striking contrast with the
grassy tabular masses of the basaltic lavas in the midst of which
they rise.

The identity in character between this rock mass and others of
eruptive origin which we have before described is sufficiently obvious.
That it was actually upheaved through the older basaltic lava sheets,
we have the clearest and most unmistakable evidence. As shown in
the plan and section (woodcuts, figs. 5 and 6), the contact between

Fig. 5.—Plan of the Hill of Sarsta Beinn.

ANGE
1 ie

a. Ordinary basalts covered by boggy ground.
b. Basalts altered by contact with the intrusive mas:.
c. Great mass of coarse dolerite graduating into gabbro.

Fig. 6.—Section of Sarsta Beinn.

a. Basalts of the great plateaux.
6. Basalts altered by contact with the intrusive mass ©.
c. Intrusive mass, composed of dolerite in places graduating into gabbro.

these highly crystalline rocks and the basalts in the midst of which
they lie is marked by a belt of metamorphosed rock. The basalts
near their junction with the intrusive dolerites and gabbro have
acquired a harder texture, a splintery fracture, and a pecultar platy
mode of weathering, often in concentrically curved planes. In con-

266 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

sequence of their greater hardness, an almost entire belt of the
altered and indurated basalts is found surrounding the crystalline
mass on all sides. From the central mass a number of dykes and
veins can be traced intersecting the surrounding older lavas.

That, subsequently to the eruption of the great sheets of basaltic
lavas, a mass of basic igneous rock was upheaved in their midst at
Sarsta Beinn is evident. And that this eruptive mass was surmounted
by a volcanic cone from which probably lava streams flowed, no one
who has studied the example of Beinn Shiant will, I think, be disposed
to doubt. Indeed fragments of these lava-streams may still remain,
though now forming undistinguishable portions of the great basaltic
plateau. When we remember the observation of Mr. Darwin, that
both in the Cape-Verde Islands and the Galapagos archipelago he
found it impossible to trace the boundaries of recently erupted lava
streams, “‘ except by the size of the bushes growing upon them, or
by the comparative glossiness of their surfaces, characters which a
short lapse of time would be sufficient to obscure,” it will not be sur-
prising that we are not, as arule, able to trace the separate lava
streams of the Hebrides. The fact that in the case of the lavas of
Beinn Shiant we are able so to do, is due solely to the fact that in
them a great capacity for resisting degrading influences is combined
with remarkably distinctive petrological characters.

I have described in some detail these two examples of Beinn
Shiant and Sarsta Beinn on account of their great size and typical
character ; but there is the clearest evidence that at a very great num-
ber of points the basalts of the great plateau were broken through
by similar eruptive masses, sometimes, as in the case of Sarsta
Beinn, composed of basic rocks, at others, as in Beinn Shiant, of more
felspathic materials, and in others, again, of both these varieties. Few
of these later eruptions appear to have been on the same grand scale
as those which we have selected as types ; and many of the intruded
rock masses, which now alone remain to indicate their points of
origin, are of quite insignificant proportions. Thus in the. little
alla of Kige two such masses were detected by Professor Geikie,
and I have found a third of similar character ; in the other islands
also such masses are found scattered in all directions. Until, how-
ever, the whole district has been mapped in the most careful detail,
it will be impossible to give a distinct view of the distribution
of these latest points of eruption in the district, and to show their
exact relations to the great voleanoes. Hven-after such survey,
indeed, many of the smaller intrusive masses may escape the obser-
vation of the geologist through Bema concealed by peat-mosses or the
vegetable covering.

‘Having pointed out that, even in the great mass of Beinn Shiant,
the preservation of the ruins of the volcanic cone are due to very re-
markable and exceptional circumstances, it will be almost unneces-
sary to add that the small cones which doubtless once surmounted
most, if not all, of these eruptive masses, have in almost every instance
been wholly swept away by denudation, and that the lavas which
flowed from them have also disappeared or are now undistinguishable.

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 267

In the case of the celebrated Scur of Eigg, however, I believe that
we have a fragment of a lava stream which proceeded from one of
these ‘‘ puys,” every other trace of which, however, has been re-
moved, ‘The character and relations of this interesting mass of rock
have been very clearly illustrated by Prof. Geikie; and I would
only suggest with regard to it that the characters of the buried con-
glomerate are suggestive of a mountain-ravine subject to the passage
of violent floods rather than of an ordinary river-channel. It seems
to me probable that this ravine was situated on the flanks of the
great volcano of Rum*, then an extinct and rapidly disintegrating
pile; that after the accumulation by the flooded stream of fragments
of all sizes of lava and sandstone (all of which would be naturally
derived from the ruins of that voleano) and the intermingling with
these of uprooted trees (the Pinites eggensis) which grew upon its
slopes, the whole was buried and sealed up in consequence of several
lava streams, issuing from an (at that time) eruptive “ puy,” finding
their way into and filling up the ravine. The course of the subse-
quent changes was similar to those long ago shown by Mr. Scrope to
have taken place in so many instances in Auvergne; and, like these
latter, they have raised up a striking monument to the power and
duration of those forces which, almost unnoticed by us, sculpture
the earth’s surface. The history of the events connected with the
formation of the Scur of Eigg has been very clearly described and
its lessons eloquently enforced by Prof. Geikie.

13. Subterranean Phenomena of the Tertiary Volcanoes.—In the
preceding pages it has been my principal object to show that in the
Hebrides we have evidence, mutilated and fragmentary it may be,
but nevertheless most clear and unmistakable, of the occurrence of
a series of phenomena which, alike in their character and their se-
quence, are identical with those exhibited by existing volcanoes.
But interesting as these conclusions undoubtedly are, they are
not perhaps the most important of the fruits of a geological study
of the beautiful dissected volcanoes of the Highlands. Still more
suggestive, from the circumstance that they enable us to correlate
the familiar subaerial actions of volcanoes with others often regarded
as wholly distinct in character and origin, are the phenomena pre-
sented to us when we study the connexions between the intrusive
masses constituting the centres of these ancient volcanic piles, and
the older stratified rocks through which they rise. To this most
interesting subject I now proceed to direct attention.

The manner in which the surrounding strata are seen to be up-
heaved, contorted, and metamorphosed by the great eruptive erys-
talline masses, whether of acid or basic composition, has been already
pointed out in a number of different localities. But in the case of
the island of Skye, it was first shown by Dr. Macculloch, and after-

* As an example of a “puy” in Auvergne, similarly situated to that which
appears to have given rise to the lavas forming the Sctr of Higg, I may point
to the Puy de Tartaret, which, rising in the Valley of Chambon, one of the
great ravines cut on the flanks of Mont Dore, has poured forth a lava stream
which has occupied all the lower parts of that great watercourse.

968 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

wards more clearly illustrated by Prof. Geikie, that besides the great
central eruptive masses of granite there are other portions of the
same rock, usually of a finer-grained character and passing into
many varieties of felsite, the relations of which to the Primary and
Secondary strata among which they he is of a totally different kind.
Instead of violently breaking through and disturbing them, these
intrusive masses appear to form hills simply overlying the stratified
rocks, but occasionally sending off veins into them and entangling
portions of them in their mass; and in place of effecting a meta-
morphism in the sedimentary roeks extending for a great distance
around, they produce only a comparatively small and local action
upon them. These features, it must be confessed, seem at first sight
sufficiently anomalous; and some authors have even gone so far as
to describe these granites as having been poured out upon the exist-
ing surface “like ordinary trap rocks.”

A careful study and comparison of all the phenomena presented
by masses of this kind, numerous examples of which occur in con-
nexion with the different voleanoes we have been describing, affords
us avery simple explanation of the supposed anomalies. These
masses of fine-grained granite passing into felsite are really portions
proceeding from the great central masses and intruded between stra-
tified rocks. In Skye, and also in some other cases, the overlying
strata have been removed by denudation; and the intrusive rocks,
which have by their hardness resisted denuding influences, now ap-
pear uncovered, and as if actually deposited upon the strata below.
That the explanation now given of this phenomenon is the true one,
I have been able to verify by the comparison of a very great number
of cases, in which every stage of the series of operations here de-
scribed as terminating in the production of the anomalous features
in question, can be clearly traced. Admirable examples showing
that these masses of granite and felsite were actually intruded
among the stratified rocks, are seen in Raasay, Ardnamurchan,
and Mull; I may especially cite the cases of Stron Beg and Craig-
nure, situated in the two last-mentioned of these districts respec-
tively, as illustrating the features and relations of such masses in a
peculiarly interesting manner. A mass of this kind, probably origin-
ally connected with the volcano of Rum, occurs at the northern part
of the island of Eigg; Professor Geikie notices the resemblance of
the rock of which it is composed to the “ quartziferous porphyries”
of Skye and Raasay, and he justly states that the mass “ appears
to have risen approximately along the bedding‘of the Oolitic strata,
and thus to form of itself a large rude bed”’*.

- In the very different modes of their subterranean disposition, the
igneous rocks of acid and basic composition respectively present a
remarkable parallel to the behaviour of the two classes of lavas to
which they give rise. While the acid igneous rocks, when intruded
between strata, tend to form thick lenticular masses, which are
generally confined to within moderate distances from the great centres
of eruption, the basaltic rocks, on the other hand, under like con-

* Quart. Journ. Geol. Soc. vol. xxvii. (1871) p. 294.

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 269

ditions, spread in vast sheets, which insert themselves along a plane
of weakness between two sets of beds, and often proceed to enormous
distances from the centres of eruption.

Very wonderful and striking are the relations between the great
intrusive sheets of dolerite and basalt, connected with the great
centres of volcanic action, and the stratified rocks which they tra-
verse. These relations are exhibited in a very beautiful manner
along the magnificent range of cliffs that forms the eastern boundary
of the peninsula of Trotternish in Skye, and also in Raasay, Apple-
cross, and the south of Mull. So remarkable is the regularity with
which many of these sheets of molten rock have flowed between the
same two strata for great distances, that it is not surprising they
were for a long period regarded as being lava streams contempora-
neous in date with the sediments among which they lie. A close
examination, however, shows that these sheets of igneous material’
alter alike the rocks lying above and below them, and that they are
altogether destitute of the vesicular character leading to the produc-
tion of amygdaloids, as well as other features always presented by true
lavastreams. When, too, their courses are followed over considerable
distances, they are found, in places, either bifurcating into separate
sheets which enclose masses of the stratified rocks, or entangling
fragments of these in their midst, or cutting for a time across the
beds, or sending off processes and veins, or terminating abruptly in
wedge-shaped masses, or breaking suddenly through the superin-
cumbent strata, and so reaching the surface.

The rocks of which the great intrusive felspathic masses are usually
composed, include many different varieties of felsite, usually more
or less quartziferous, and often beautifully porphyritic in struc-
ture; these, in many places, by the appearance of scattered crystals
of hornblende in their mass pass into a fine-grained syenite-granite.
The intrusive sheets of basic rock consist in almost all cases of dole-
rite, often containing much olivine, and passing on the one hand
into fine-grained gabbro, and on the other into many varieties of
basalt.

Besides these larger masses, the igneous intrusions of all ages give
rise to the formation of dykes and veins in prodigious numbers.
Those of felspathic composition appear to be for the most part con-
fined to within comparatively moderate distances from the eruptive
centres ; those of basaltic composition, on the other hand, are found
proceeding to extraordinary distances from them. Prof. Geikie has
even speculated, with much show of probability, on the existence of a
connexion between the great basaltic dykes which traverse the whole
of the rocks in the north of England and the great focus of igneous ac-
tivity in the Hebrides. Basaltic dykes with such a connexion certainly
traverse all the rocks in the west of Scotland in prodigious numbers ;
and the manner in which they sometimes, through greater relative
capacity for resisting denuding forces, stand up like immense walls,
and at others, by their more rapid decay, originate vast chasms, are
facts which must have been observed by all who have travelled in
the western Highlands. In the immediate neighbourhoods of the

270 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

great volcanoes their almost infinite numbers, the complexities of
their intricate interlacings, and the manner in which those of later
date cut across and often laterally displace portions of the earlier
ones, give rise to a series of phenomena of the most instructive and
interesting character. As a general rule, the larger the dyke the
more coarsely crystalline is the mass of rock which composes it;
and in many cases the middle of the dyke is composed of a coarser-
grained rock than the sides.

Frequent allusion has already been made to the metamorphism
produced in the Primary and Secondary strata, and also in pre-
existing rocks, by the passage through them of masses of igneous
rock. Careful examination will serve to detect such alteration at the
surfaces of contact of almost every igneous mass; but in the degree of
this action there is the greatest diversity in different cases. Some-
times it consists of a scarcely perceptible induration extending for a
distance of a few lines only from the surfaces of the igneous mass,
and unattended by any change in chemical characters ; but in other
cases the soft sediments of the Lias, Oolite and Cretaceous forma-
tions have had every trace of their abundant fossil contents wholly
obliterated, and their masses converted into rocks undistinguishable
in appearance and characters from the most highly metamorphic
Primary rocks, with which, indeed, they have in some cases been
confounded.

A careful study of these phenomena of local metamorphism enables
us to enunciate the general law which governs its action to be as
follows :—The degree of metamorphism and the distance to which
its action extends from the intrusive mass is usually proportioned
to the mass of the latter. Thus, the maximum of change is seen
in the vicinity of the great eruptive mountain masses, and the
minimum near the smaller intrusive veins and dykes ; while around
the various intrusive sheets and bosses every intermediate degree of
alteration is exhibited. Even among dykes we can often observe
that the amount of induration in the strata which they traverse, and
the distance to which this extends from their surfaces, are directly
proportioned to their width. There are, it is true, occasional appa-
rent exceptions to this rule; but these are in most cases capable of
easy explanation. ‘Thus masses of rock traversed by a complete plexus
of veins and dykes, such as are seen at some points in Ardnamurchan,
undergo a very great amount of change, their fossils being almost
wholly obliterated and very decided chemical changes induced in
their mass, these changes being evidently the result of the cwmulative
action of the numerous small intrusions. The establishment of this
law of metamorphic action serves to confirm what, indeed, is suffi-
ciently obvious, that the metamorphism in surrounding rocks results
from the passage of heat from the intrusive masses of molten matter
as these gradually pass into the solid condition.

I can scarcely conceive of any series of phenomena more striking in
character, and certainly of none more interesting and suggestive to
a geologist, than those presented by the wonderful complexities pro-
duced by the mutual interferences of different eruptive rock-masses,

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 271

varying in dimensions from great mountain-groups down to the
minutest veins and strings, exhibiting innumerable distinctions of
mineralogical constitution, and belonging to three successive geolo-
gical periods. Nor is it easy to imagine a more striking series of me-
chanical and chemical changes than those which in the surrounding
strata have attended the eruption of these masses, resulting in the
most extraordinary contortions and the extremest metamorphism.

Such features are admirably exhibited alike in the district of
Strath in Skye, on the southern coast of Mull, and at various points
around Rum, but in none in so striking a manner as in the penin-
sula of Ardnamurchan. Here the scalpel of denudation has revealed
the intricate and curious relations of the several varieties of igneous
rocks with one another and with the Primary and Secondary strata.
To represent these, however imperfectly, would require a series of
maps and sections on the very grandest scale; in many cases, indeed,
adequate conceptions of the relations of such very intricate rock
structures could only be conveyed by means of models. It is, how-
ever, to the rocks themselves that the student must be referred ;
and he will find in them illustrations of the characters and action of
igneous intrusions which will amply repay him for the time and
labour expended in their investigation *.

14. Ages of the several Volcanic Outbursts already described.—That
the events of which we have been discussing the evidence, stupen-
dous as they are in scale and complicated as they are in succession,
all took place subsequently to the Mesozoic epoch, we have the most
ample proof :—

First, in the fact of the entire absence of contemporary volcanic
deposits among the Secondary strata. I have already alluded to the
earlier misapprehensions which prevailed upon this subject, and on
the manner in which these were removed by the observations of
Prof. Geikie in 1865.

Secondly, in the circumstance that the volcanic masses are thrust
through and among the representatives of the whole series of Secon-
dary strata, up to and including the Upper Chalk; and that the
volcanic products unconformably overlie and include numerous
fragments of the whole of the Secondary rocks—these fragments
having acquired their present positions either through being
ejected from volcanic vents (as in the case of the breccias of Mull,
Rum, &e.), or by the agency of ordinary denuding agencies opera-

* Below Mingary Castle, near Kilchoan in Ardnamurchan, a section may be
observed which is interesting as throwing light on the probable mode of for-
mation of the celebrated Puy Chopine in Auvergne, the peculiar and appa-
rently, at first sight, anomalous characters of which have been frequently re-
marked upon (see Scrope’s ‘Geology and Extinct Volcanos of Central France,’
pp. 72-76). At Mingary a mass of quartziferous porphyry has been forced
between beds of Lower Lias shale, while a later-formed sheet of basalt has
evidently taken advantage of the plane of weakness, constituted by their junc-
tion, to force itself between them. In the Puy Chopine a mass of the older
rocks (in this case composed of granite) has been forced upwards by an extru-
sion of domite, while a sheet of basalt has similarly inserted itself along their
line of junction.

Die, J. W. JUDD .ON THE SECONDARY ROCKS OF SCOTLAND.

ting in the intervals between the several outflows of lava (as in the
chalk- and flint-detritus of Ardtun and Carsaig)..

That, on the other hand, this remarkable series of events, even
the latest of them, took place at a period very remote from the
present, is proved by the enormous amount of denudation which the
volcanic products have undergone.

I know of no more striking evidence of the vast duration of geo-
logical periods, than that which is afforded to us in the Hebrides,
when we contrast the wonderful freshness of the contours, polished
surfaces, and strize produced during the Glacial period, with the
everywhere abundant proofs of enormous denudation suffered by
rocks of undoubted Tertiary age, many of these, moreover, being
very conspicuous for their intense hardness and great capability of
resisting weathering influences. How almost infinitesimal on such
a comparison appears the time which has elapsed since the Glacial
epoch, to the duration of the great Tertiary periods!

The facts adduced in the present paper show that there were three
well-marked periods of igneous activity in the district, which were
characterized as follows :—

First Period. The outburst in the midst of a terrestrial surface, com-
posed of various Paleozoic and Secondary rocks, of ashes, scorice, and
fragments of the rocks in which the vents were opened, alternating
- with the outflow of streams of highly felspathic lava. With these
subaerial phenomena was connected the ascent of molten masses, in
a manner to a great extent peculiar to rocks of acid composition,
which, injecting the fissures of the surrounding and superposed rocks,
consolidated into felsitic rocks in their outer portions, but in their
deeper and more central portions, under different conditions of slow
cooling and great pressure, assumed well-marked granitic characters.

Second Period. The extrusion from the same volcanic foci of
masses of basic igneous materials, which on reaching the surface
spread, after the usual manner of lavas of this class, into streams
which, following one another at intervals, sometimes of long duration,
gradually built up those enormous plateaux of basalt rock of which
only mere fragments have escaped denudation. The same eruptions
of basic rocks evidently gave rise to the formation of the moun-
tain-masses of gabbro and the intrusion between and among the
surrounding strata of innumerable and widely-spread sheets and
dykes of dolerite and basalt.

Third Period. 'The appearance in a sporadic manner, in the neigh-
bourhood of the grand old extinct volcanoes, of numerous minor out-
bursts of lava (felspathic, basaltic, or intermediate in composition),
which, with the accompanying fragmentary ejections, gave rise to
the formation of a series of volcanic cones, small, indeed, as com-
pared with the vast mountain masses formed during the two pre-
ceding periods of eruption, but of which some at least were of no
mean dimensions. The greater number of these would, to some ex-
tent, compensate for their inferior dimensions.

But from the facts which I have adduced in this paper it is also
evident that periods of enormous duration must have separated these

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 273

three epochs of intense volcanic activity. This is evidenced by the
unconformable relations which the rock-masses produced during the
three periods of eruption bear to one another.

The proofs of extensive denudation having taken place in the in-
terval between the outflow of the great masses of felspathic and
basaltic lavas appear to be of a very decisive character.

(1) The basaltic lavas and their accompanying piles of scoriz, rest
directly in many cases, as is so well seen in both Rum and Mull, not on
the lavas and ashes which must have constituted the outer portions of
the earlier volcanoes, but on the intrusive felsites and granites which
could only have formed their interior and deeply seated portions.

(2) The felspathic lavas, &c., are seen in Mull to have undergone
considerable movements and to have suffered largely from denudation,
before they were buried under the overwhelming products of the
period of basaltic eruptions.

(3) It appears also that the basaltic streams frequently le in
hollows eroded in the preexisting felstones. This fact is not so
easily made out as those before noticed; but I believe there are
distinct proofs of it in Ardnamurchan.

The evidences of erosion on a yery extensive scale having taken
place between the period of the eruptions of basic rocks and that of
the formation of the “ puys”’ are not so easily traced as those of the
former inter-volcanic period, owing to the almost complete removal
of the latest-formed volcanic deposits of the district by denuda-
tion. But in the case of the Sctr of Higg, as so clearly explained
by Prof. Geikie, and in that of Beinn Shiant, which has been de-
scribed in this paper, the proofs of unconformity between the lavas
of the two series are sufficiently distinct and striking.

Lastly, of the enormous changes which have taken place since the
period of the formation of the “ puys,” we have the clearest proofs
in the total removal, with rare exceptions, of the volcanic piles which
doubtless once surmounted those igneous protrusions, which now
alone mark the sites of the latest series of eruptions in the district.

In the almost total absence of paleontological evidence, it cannot,
of course, be absolutely demonstrated that the three series of volcanic
outbursts which we have described were in the case of the different
centres strictly synchronous; but the remarkable uniformity of cha-
racters, in the succession of products in all the different cases, leads
to the very strongest presumptions in favour of a connexion having
existed between these different foci, and of the contemporaneity be-
tween the several analogous phenomena manifested at each of them.

Our means of correlating these three epochs of volanic activity
during the Tertiary period, with the divisions of the same era founded
on the relations which its successive faunas bear to one another and
to the existing creation, must be confessed to be small; and hence
the results attainable by us on this subject are to a certain extent
indecisive ; nevertheless it will be wellin this place to bring together
the whole mass of evidence as yet obtained which bears upon the
question.

The beds intercalated with the earliest series, that of the fel-

fie?

274 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

spathic lavas, have, as yet, furnished no organic remains. They are
evidently, as a mass, post-Cretaceous ; but I have observed some phe-
nomena, which will be described hereafter, apparently pointing to
the conclusion that volcanic activity must have commenced in the
district before the Cretaceous period had altogether closed. May we
not, therefore, bearing in mind-these facts and also the strong proofs
already detailed of unconformable relations between these felspathic
lavas and the Miocene basalts, venture to refer the former, at least
provisionally, to the great Eocene period ?

The proofs of the Miocene age of the basalts constituting the great
plateaux, both in the Hebrides and in Antrim, are as follows :—

At Ardtun in Mull, low down in the series, the following species
of plants have been obtained from the pond-like hollows buried
under streams of mud and lavas as before described.

Sequoia Langsdorffii, 4d. Brongn. Rhamnites lanceolatus, Forbes.
(Taxites Campbellii, Forbes.) Equisetum Campbellii, Forbes.
Rhamnites (?) multinervatus, Filicites (?) hebridicus, Forbes.
Forbes. Alnites (?) MeQuarrii, Forbes.
major, Forbes. Corylus grosse-dentata, Heer.

In the plant-beds intercalated in a series of lacustrine deposits at
Ballypalidy, Co. Antrim, the following have been found :—
Sequoia Du Noyeri, Baily. Eucalyptus oceanica, Unger.

Pinus (?) Plutonis, Baily. Daphnogene Kanu, Heer.
Cupressites MacHenryi, Bazly.

with fragments referred, somewhat doubtfully, to the genera Pla-

tanus?, Fagus, Podocarpus, Andromeda, Quercus, Rhamnus, Hakea,
Celastrus, and Graminites. Also the elytra of two species of beetles.

In “ ash-beds”” between the basalts near Shane’s Castle, Lough
Neagh, Co. Antrim, the following were obtained :—

Platanus aceroides, Gépp. | Sequoia Langsdorfii, 4d. Brongn.

with fragments referred to the genera Juglans, Fagus, Laurus, &c.*

Both Prof. Edward Forbes and Prof. Heer concur in the opinion
that these floras indicate that the beds in which they occur were
deposited during the Miocene period.

Concerning the age of the last of the three periods of eruption,
that of the “ puys,” the evidence from organic remains is much more
scanty. It consists only of the fossil wood (known as Pinites
eggensis, With.) found under the lava stream of the Scur of Higg,
upon which, of course, no conclusion can be founded. Considering,
however, the proofs already adduced, from their unconformable re-
lations, of the separation of the deposits of this period alike from
those of the Miocene and the recent epochs, we can scarcely hesitate
to regard them as belonging, at least approximately, to the Pliocenef.

* Vide Forbes, Quart. Journ. Geol. Soe. vol. vii. (1853), p. 103; W. H. Baily,
ibid. vol. xxv. (1869), p. 357; and Tate and Holden, 7bzd. vol. xxvi. (1870),
pp. 162-63.

t In venturing thus to compare the three periods of volcanic activity in the

north-western parts of the British Islands with those divisions of the Tertiary
epoch which Sir Charles Lyell has founded upon a comparison of the molluscan

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 275

To those familiar only with the English representatives of the
three great Tertiary periods, the Eocene sands, marls, and clays, of
the London and Hampshire basins, the Miocene lignitiferous beds
of Bovey Tracey, and the insignificant Pliocene deposits of the Crags,
the evidences of such enormous physical changes, as we have shown
must have taken place in the north-western portion of these islands
during the same periods, will indeed appear startling. But to re-
move any feelings of difficulty or doubt which may have their source
in such comparisons, it is only necessary to refer to the well-ascer-
tained facts of the geology of the Alps, where events on even a
grander scale than those which we have described, can be shown to
have taken place during the same periods.

15. Connewion between the Tertiary Volcanoes of the Hebrides and
those of other districts—It would be foreign to the objects of the
present memoir to enter upon a discussion of this most interesting
question. I shall therefore only state that, while we have the
clearest proofs of the contemporaneity of the basalts of Antrim with
those of the Inner Hebrides, we have also strong grounds for regard-
ing the granites of Arran and the Mourne Mountains as having
been erupted during the same period with those of Skye, Mull, Rum,
&c.* Thus we are led to the conclusion that along a line stretching
at least 400 miles from north to south, in the north-western part of the
British archipelago, there rose, during a great portion of the Tertiary
period, a chain of volcanoes in a state of violent but intermittent erup-
tion. But continuing this line to the northwards, we find the proofs
of volcanic action during part of the Tertiary period (and in some
cases, at least, this action has not yet become extinct) in the Faroe
Islands, Iceland, Jan Meyen, and Greenland. And southward the
same line of volcanic vents is continued in Central France, the Iberian
peninsula, the Azores, Madeira, the Canaries, Cape-Verde Islands,
Ascension, St. Helena, and Tristan d’Acunha.

When we remember the proofs of the existence of widely spread
terrestrial conditions during the Miocene epoch, and the interesting
facts concerning the distribution of existing terrestrial species, made
known to us by the labours of Edward Forbes, Unger, Heer, and
others, we may well be prepared to regard these isolated masses of
volcanic rocks, as has been suggested by Professor Nordenskiold, as
portions of a great ridge now for the most part submerged beneath
the sea-level and constituting a boundary to the great eastern conti-
nent, similar to that which the Andes, Cordilleras, and Rocky Moun-

faunas of certain deposits with those of recent seas, [ am actuated by a feeling
of the necessity of such a comparison to a due appreciation of the subject,
rather than by a belief in the possibility of establishing any thing like actual
contemporaneity between divisions based respectively on purely paleontological
and physical evidence, both of which series of divisions, moreover, are incapa-
ble of very exact definition and limitation.

* My friend, Mr. Thomas Davies, has called my attention to the fact that the
granite of Lundy Island offers peculiarities of structure strikingly similar to
those presented by the granites of the Mourne Mountains, Arran, and the
Northern Hebrides. It is pcssible, therefore, that in Lundy Island we have
relics of another of the centres of volcanic action during the Tertiary period.

276 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

tains form tothe western. The recent soundings of H.M.S. ‘ Chal-
lenger’ have not only confirmed these conclusions, but have shown
that the elevation of this old Andes of the eastern continent was not
inferior to that of its existing homologue in the western, and that
some of the peaks rose to the height of at least 25000 feet above the
surrounding plains.

16. General Conclusions from the relations between the Volcanic and
Plutonic rocks of the Tertiary period—tThe study of the rocks of
the Inner Hebrides has clearly demonstrated a fact which had been
already strongly suspected by Dr. Macculloch, Prof. Geikie, and other
geologists, that the granitic rocks of that area had a close connexion
and were of contemporaneous age with the old lava streams so ex-
tensively developed in their neighbourhood.

We have shown how, by the more or less complete removal of its
upper portions and the consequent exposure of its deeply svated
rocks through denudation, a voleano may present many different as-
pects—from examples such as that of Mull, in which the relations of
the various volcanic products can still be easily traced, to others, like
that of Skye, in which these relations are by no means so clear at
first sight. It will be easy to conceive a still further stage of ruin
in a voleano in which, all the surrounding lavas and other erupted
materials being removed, we should have left only the central core
of granite or gabbro (the latter perhaps altered into various serpen-
tinous rocks) rising in the midst of a series of stratified rocks; and
in such cases there would be nothing to connect such eruptive masses
with the ordinary subaerial phenomena of volcanic activity.

When we reflect on the striking similarity of the products of this
series of Tertiary volcanoes, we can scarcely doubt of their connexion,
or of their materials having been derived from a common reservoir.
It is, under these circumstances, easy to imagine that, by the total
removal of the superincumbent rocks, the consolidated tontents of
this old reservoir may be exposed. In such cases we might reasonably
expect to find at the surface a tract of granite of wide extent, like
that of Leinster.

Fortunately the same district which we are now studying affords
us, in a series of igneous rocks of older date than those already de-
scribed, a number of beautiful illustrations and convincing proofs of
the truth of these views concerning the relations between the Volcanic
and the Plutonic rocks. These we now proceed to describe.

Ill. The Newer Paleozoic Volcanoes.

That there is evidence in the Highlands of Scotland of a great
period of volcanic activity which preceded the deposition of the Se-
condary strata of the district has been already pointed out. I shall
proceed to show that the eruptions of this earlier period were on a
scale of at least equal magnitude with those of the Tertiary epoch
just described, that like these latter they gave rise to the formation
of vast plateaux composed of lava streams surrounding on all sides
great voleanic mountains, and that under the action of denuding

J. W. JUDD GN THE SECONDARY ROCKS OF SCOTLAND. 2G,

forces these plateaux have been broken up into more or less isolated
fragments, while the central volcanic cones have been so worn away
as to have lost many of their distinctive features. From the facts to
be adduced it will clearly appear that, prior to the Mesozoic epoch,
the greater part of the southern and central districts of Scotland, both
northand south of the Grampians, wascoveredto the depth of thousands
of feet by the products of igneous activity and the strata enclosed
between them. But, as might be anticipated from their far greater
antiquity, the lava plateaux of this earlier period are in a more
fragmentary condition, and its volcanic cones ina far more ruined state
than those of the Tertiary period. Aided, however, by the analogies
of the latter, I hope to be able to demonstrate what were the prin-
cipal characters of this earlier volcanic period, to trace the history
of the succession of events which took place during its continuance ;
but more especially, as bearing upon the objects of the present in-
quiry, my endeavours will be directed to the reconstruction of those
ancient physical features which resulted from these volcanic out-
bursts, and which to so great an extent determined the conditions
under which the Mesozoic strata were deposited.

1. Lavas of Lora and the adjacent islands.—In the same district of
the Western Highlands with the Tertiary lavas already described, and
in close proximity to them, we find another series of volcanic rocks
belonging to a very different geological period. They occupy a large
portion of the district of Lorn, and are also extensively developed in
the south-eastern part of Mull, in Kerrera, Seil, and the smaller is-
lands in their vicinity. By Loch Etive this volcanic tract, which
extends about 22 miles from north to south and 18 miles from east
to west, is divided into two very unequal portions—the smaller and
northern portion extending through Benediraloch towards Loch Cre-
ran, and the larger and southern portion to the neighbourhood of
Loch Awe and Loch Melfort. An examination of the boundaries
and relations of this volcanic tract shows that it must be regarded as
a fragmentary patch of a series of deposits once widely spread, which
has been preserved from denudation, like so many other masses of
rock of various ages in the Highlands of Scotland, through being let
down between great faults or within vast synclinal folds. This con-
clusion is fully borne out, as we shall hereafter see, by the existence
of small and distant isolated patches (outliers) of the same formation.

Until very recently the “ trap-rocks ” of Lorn were confounded
with those of Tertiary age which are developed in their immediate
neighbourhood. Macculloch appears to have considered that the
sandstones and conglomerates which occur at their base were part of
the Secondary series of the Highlands, while Prof. Nicol suggested
that they might represent the Trias. In the map of Scotland pub-
lished in 1861 by Sir R. Murchison and Prof. Geikie the “ traps” of
Lorn are indicated as being of Old Red Sandstone age.

The general features of the district occupied by these rocks, and the
scenery to which they give rise are those which usually characterize
tracts of rocks of volcanic origin. These are well illustrated by the
annexed outline sketch of the mountain which rises at the eastern end

@I.G.8. Na: 119, x

278 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

of Glen Lonnan (woodcut, fig. 7). When compared with the great
plateaux of the Tertiary lavas which have been already described, a
number of points of difference, however, will be remarked. The
wonderful regularity of the terraced features, so strikingly displayed
by the basaltic plateaux of the Tertiary period, are to a certain

Fig. 7.-—_Outhne Sketch of a Mountain near Glen Lonnan, illustrating
the Mode of Weathering of the Porphyrite and Felstone Lavas of
Lorn. (The granitic peaks of Beinn Cruachan are seen in the
background.)

extent wanting in the piles of felspathic lavas which were poured
out before the deposition of the Secondary rocks; the cause of this
difference is of course to be sought in the usually greater fluidity and
consequent even diffusion of the lavas of basic as compared with those
of acid and intermediate composition.

2. Characters of the Volcanic Rocks of Lorn.—As already inti-

--mated, the great mass of these lavas is of felspathic or acid compo-

sition, and includes innumerable varieties of the rocks which were
formerly known by the names of felspar-porphyry, compact felspar,
clinkstone, claystone, &c. For the most part they may be regarded
either as varieties of the rocks included in this country under the
name of ‘‘ felstone,” or of the extensive class called “ porphyrites ”

upon the Continent—the former representing the quartz-trachytes,
. and the latter the ordinary trachytes of modern volcanoes.

The changes which have taken place in many of these rocks sub-

- sequently to their deposition have been such, as in many cases to

mask their real origin as lava streams; in this respect, however,
we find the greatest variations in the same mass of rock. A close
examination of the structure and relations of these rocks will, in
almost every instance, furnish many interesting points of evidence
bearing upon their mode of origin.

The columnar structure is by no means so common among the fel-
spathic lavas as among those of the basaltic class; nevertheless some
interesting examples of it are afforded by the rocks of Lorn. The
eolumnar forms assumed by the acid lavas present distinctive charac-
teristics of their own, as compared with those of the basalts. Inthe
former class of rocks the columns are usually of smaller diameter and

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 279

less regular form, while they often extend to much greater length
than those of the latter class, unlike which again they are never di-
vided into regular blocks by equidistant, curved, joint-planes. One
of the best examples of the columnar structure among the old lavas

Fig. 8.— Chiff-section, S.W. of Oban, exhibiting very fine Columnar
Structure in the Porphyrite Lavas of Lorn. ;

of Lorn is that exhibited in a cliff about two miles south-west of
Oban, where the columns are of great length and beautifully curved
(see woodcut, fig. 8); but more or less perfect examples of the same
structure are exposed on the face of Beinn Lora and at other points.

Some of the volcanic rocks of Lorn present characters which, at
first sight, appear not a little anomalous and puzzling; but this is a
circumstance which, when we remember the metamorphic processes
to which lavas, in common with all other rocks, are subject, need
scarcely occasion surprise. Certain of the lavas appear after their
emission to have weathered into the characteristic spheroidal and
concentric forms, while others have evidently decomposed into a
“‘ wackose” condition ; and in both cases the rocks which result from
the re-induration of such masses, present very peculiar features, and
may, in some instances, be mistaken for consolidated “trap-tuffs.” _

The separate lava-flows were often of enormous thickness. As a
general rule the great body of each of these streams is made up of a
compact, often highly porphyritic rock ; but towards the upper and
under surfaces of the mass, it usually assumes the amygdaloidal
structure. The amygdaloidal cavities, which have evidently served
as chemical laboratories in which very complex operations have been
carried on, are frequently deprived of their contents by recent wea-
thering operations ; and the original structure of the rocks is thus to a
great extent restored. Then is made clearly apparent the originally
highly vesicular character of the upper and under portions of these an-
cient lava streams, the vesicles being often seen to be drawn out in the

x2

280 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

direction of the flow, while the actual surfaces of the stream exhibit
the most strikingly scoriaceous aspect. The same weathering pro-
cess sometimes develops in these old lavas other original structures,
which had become wholly obliterated by infiltration, crystallization,
and other processes taking place in the mass of the rock; and these
structures would have remained altogether unsuspected but for the
action of this cause. Thus some highly crystalline and porphyritic
rocks, when weathered, resume their earthy or compact texture ; and
in certain cases structures like the spheerulites of pearlstone, which
had become wholly obscured in the mass, are again revealed. Simi-
larly many rocks of very solid appearance are seen, when the infil-
trated materials are removed by weathering, to have been originally
aggregates of ashy, pumiceous, and scoriaceous fragments, among
which “ voleanie bombs” or their fragments may not unfrequently
be detected*. ;

3. Relations of the Volcanic Rocks of Lorn.—In considering this
question it is important at the outset to notice two striking facts with
regard to the positions of the lavas and associated beds of Lorn. In
the first place they always rest directly, but unconformably, upon
the Lower Silurian gneissose and schistose rocks, and never exhibit
any of the fossiliferous Secondary strata at their base; in this respect
they present a striking difference in their relations from the Terti-
ary lavas of the adjoining districts. And, secondly, they are, like
the older rocks upon which they repose, penetrated in every direc-
tion by numerous dykes of dolerite and basalt; these are precisely
similar to the intrusive masses associated with the Tertiary volcanic
rocks—of which series, indeed, we can scarcely hesitate to regard
them as forming a part.

These later dykes of dolerite and basalt, which are sometimes of
great width, and often present a prismatic or columnar structure, con-
stitute a most interesting and striking feature in the district. In
consequence of their usually greater relative hardness, they frequently
stand up like gigantic walls amidst the rocks of slate and felstone
which once enclosed them but have now been weathered away from
their sides.

The remarkable series of volcanic rocks of Lorn is probably several
thousand feet in thickness—though, its upper portions having been
everywhere removed by denudation, its original limits in this respect
are quite unknown to us. Its relations to the slate rocks and certain
conglomerates, sandstones, and breccias appear, at first sight, to be

* That the geologist never meets with glassy lavas (obsidians, pitchstones &c.)
among the older rocks will not occasion any surprise when we reflect upon the
facility with which artificial glasses undergo devitrification. That glassy lavas
were formed even in palzozoic times appears to be indicated by the fact recorded.
in the text that weathering sometimes reveals the characteristic spherulitic struc-
ture in some of the lavasof Lorn. Many of the Newer Paleozoic “ porphyrites”
of Scotland are quite undistinguishable in appearance from some trachytes, such
as those of Hungary. The white granular siliceous rocks into which the former
are sometimes found altered, appear to be equally undistinguishable from the
products of the decomposition of the latter (occasioned by the passage of acid
vapours through them), and constituting the so-cailed “ Alaunstein.”

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 281

so anomalous as to have occasioned no little difficulty to the older
geologists who sought to explain them ; indeed the slate rocks were
found to occcur at such various levels among the series of lavas, con-
glomerates, and sandstones, and in such unexpected situations, as
quite to baffle Macculloch and other observers who sought to illustrate
their relations by horizontal sections*.

That in a district so disturbed as this has been shown to be, the
strata in question should be found to be traversed by great faults will
occasion no surprise ; and among rocks of such inconstant characters
as these, it is, moreover, not easy to detect and follow the details of
such dislocations, though their general effect may often be sufficiently
obvious. Thus in the island of Kerrera a N.W. and 8.E. fault, with
a throw of not less than from 500 to 600 feet, is very conspicuous.
It is not, however, to these dislocations, striking as their effects un- |
doubtedly are, that the anomalous position of the rocks in question is
mainly due, but rather to the fact that the volcanic and associated
rocks were accumulated upon an old terrestrial surface of great irre-
gularity, the hills and peaks of which stand up at very various levels
amidst the great mass of overlying rocks. From the examination of
the sections near Oban, in the island of Kerrera, and on the shores
of Lochnell Bay, this relation of the several rocks becomes sufficiently
clear. Sometimes, in favourable sections, peaks of the Lower Silurian
slate rocks are actually seen standing up amid the overlying deposits,
while in other cases hills composed of the former rock stand in such a
manner amid surrounding masses composed of nearly horizontal beds
of conglomerate, sandstone, and lava, that no doubt can remain that
the former constituted, prior to the denudation which has sculptured
the existing surface, the sides of valleys which were subsequently filled

Fig. 9.—Section through the Highest Point of the Island of Kerrera,
illustrating the Relations between the Slate Rocks and the overlying
Volcanic Series of Lorn.

a 7 ee OOS =~

a

a. Lower Silurian slates. 6. Breccia composed of angular fragments of a.
c. Conglomerates and sandstones. d. Porphyrite lava-streams. + Slates
baked and altered by the enveloping lava.

with the deposits of the latter. Interesting examples of such slate
hills again exposed by denudation are found at the “ mythical Bere-
gonium” (where the mass is crowned by a vitrified fort) and near
the town of Oban. The accompanying section (woodcut, fig. 9)

* Vide Macculloch’s ‘Western Isles of Scotland,’ vol. iii. pl. xvi. fig. 3, p. 14.

282 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

will serve to illustrate the peculiar relations of the several rocks in
these cases. eden

The rocks on which the voleanie series of Lorn usually rests are
black clay-slates, abounding in fine cubical crystals of pyrites and
often traversed by numerous veins of quartz. This formation is
perhaps the highest member preserved to us of that great series of
Lower Silurian rocks which, bent into endless folds and greatly me-
tamorphosed, occupies so large a portion of the Scottish Highlands.
At Seil and Easdale it is extensively worked for roofing-slates; as
yet, unfortunately, it has not yielded any fossils; but it exhibits in
the island of Seil interbedded igneous rocks, apparently of contem-
poraneous character.

The only point at which the series of rocks so well displayed in
Lorn is seen in juxtaposition with the Mesozoic sedimentary rocks
and the Tertiary lavas is in the south-eastern part of Mull. Here,
unfortunately, among the wonderfully disturbed and greatly meta-
morphosed rocks, which are exposed only in precipitous and altogether
inaccessible cliffs, I have sought in vain for any simple section illus-
trating the relations of these three series of deposits. Nevertheless,
after carefully tracing the positions of such masses as can be reached
and studied, there appears to be no room for doubt that the various
members of the Mesozoic series rest indifferently upon the denuded
lavas of Lorn and the older rocks, and that they are themselves
covered unconformably by the Tertiary volcanic rocks.

4, Succession of Rocks in Lorn.—The series of rocks which we have
been describing has usually been represented as consisting of beds of
conglomerate and sandstone at the base, overlain by a great mass of
“trap”? rocks. A careful examination of the district, however,
proves that its structure is by no means so simple as this statement
would imply. The outpouring of the great lava streams was, in part
at least, contemporaneous with, as well as subsequent to, the depo-
sition of the conglomerates and sandstones. This is proved by the
alternation of the “ trap” rocks with the conglomerates and sand-
stones, and by the fact that the materials of the latter are to a
great extent derived from the former. Good examples of the alter-
nations of the traps with the sandstones and conglomerates are to be
seen near Dunolly.

At the base of the whole series of the Lorn rocks, and in imme-
diate contact with the subjacent slates, is often found a breccia of
very interesting character. It is wholly composed of perfectly an-
gular fragments, sometimes of considerable size, of the slate and
quartz recks on which it rests; this breccia exhibits no trace of
stratification or of its materials having been sorted or acted upon
by water. It sometimes forms masses of considerable thickness,
which appear to have been accumulated upon old terrestrial surfaces
of the slate rocks by purely subaerial agencies.

The conglomerates of the Lorn series present very remarkable
characters, which are familiar to all who have examined the pictur-
esque cliffs on either side of the beautiful Bay of Oban. They are
made up of blocks of very various sizes, occasionally angular but

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 283

usually subangular or well-rounded, of ‘“ trappean” materials
mingled with others derived from the metamorphic rocks. These
blocks are so firmly cemented together by a sandy matrix, derived
from the same materials, that the great jomts which traverse the
rock in all directions, and give rise to the bold and fantastic forms
which it assumes under denudation, frequently cleanly divide the sepa-
rate pebbles of which the rock is made up, even when they are ccm-
posed of the hardest known materials. These conglomerates, which
attain to thicknesses of hundreds of feet, exhibit great variations in
the size of their materials, sometimes passing into coarse sandstone,
which usually occupies lenticular patches in their midst, and at others
containing blocks of very large dimensions. In the remarkable na-
turally isolated pillar of this conglomerate near Dunolly Castle, which.
is known by the name of Clach-a-choin, there is a single block of
felstone, 8 feet long, 6 feet broad, and 5 feet thick; and many similar
blocks of almost equal dimensions are seen at other points.

The sandstones of the Lorn series vary from very coarse grits,
almost wholly made up of felspathic materials, to fine micaceous
sandstones, exactly resembling those so abundant in, but by no means
peculiar to the Old Red Sandstone. These sandstones are usually of
a grey colour, but sometimes brownish-red. They frequently exhibit
much false-bedding, and sometimes contain fragments of rock of con-
siderable size but of flat form, such as can easily be borne along by
currents of water. In many places they include bands of pebbles of
very various size, and thus graduate into the conglomerates.

With the conglomerates, sandstones, and “ traps” are interstra-
tified a number of subordinate beds of peculiar character ; of these
some are composed of well-stratified materials of excessive fineness,
which appear to occupy pond-like cavities in the midst of the other
beds; and these seem to be made up of the fine volcanic dust so
frequently ejected in great quantities from volcanic vents. In other
cases we find well-stratified beds of small angular fragments of
igneous material (lapilli &c.), constituting “tuffs” precisely similar
to those of modern volcanoes.

As might be anticipated from the characters of these rocks, there is
no appearance of a regular order of succession among them; but on
the contrary the deposits are of the most local description. In order
to illustrate their mode of occurrence I give a section taken along
the south end of the island of Kerrera, where these strata are
especially well exposed (woodcut, fig. 10, p. 284).

From the descriptions already given of the relations of the series
of rocks in Lorn, it is clear that the lava streams must sometimes be
found in contact with the slates, at other times with the conglome-
rates, and at others, again, with the sandstones. The appearance
presented by each of these junctions I propose to describe in illus-
trative examples.

In the island of Kerrera we find a remarkably interesting example
of the contact of one of the lava streams with the slate. Herea pin-
nacle of slate rock protruding above thick masses of sandstone and
conglomerate, in the manner already described has been involved in

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

284

‘souoyspuvs Laas ostzoo Aaa “6

‘ULOrY JO SBALT AY] Jo syooTg pux satqqod Auvur yyIM ‘soyereuto[suOd osavoy f
‘guoyspuvs Aors-Ystueetd JO speq Yor, *3

‘(saztafyqdaiod pte souojs[ay) ULOTT JO svavy “p

‘SODA UBLINIIg aMOTT JO syooT paw sejqqad yo pasoduioo Ayureur ‘seyexomo[suoy “9
‘SyNOd UBLINIIg IaMor] JO syuawsesz BpNduB Jo pesoduioo swroowag “¢

‘suiea zqzenb yytM ‘(URLANTIg IaMorT) syoor ayBTg ~”

wiava ‘ys words) "jae cpg

‘utory “nialtay fo punsy ay) fo pugt yimog ayy buoy wousag—'OL ‘st

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 285

a lava current. The dark-blue highly cleaved slate is changed to a
soft grey rock, divided into angular fragments by numerous joints,
while the brilliant crystals of iron-pyrites with which the slate is
studded are converted into a black amorphous substance still filling
the original cavities (see woodcut, fig. 9, p. 281).

Many examples of the contact of a lava stream with a bed of sand-
stone are exhibited in Lorn and the adjoining islands.. The sandstone,
while still exhibiting more or less clearly, under the lens, its original
granular structure, has for some depth from its surface been con-
verted into a substance of intense hardness, and the mixture of fel-
spathic and siliceous materials of which it is composed has evidently
undergone incipient fusion.

Where the lava has flowed over a bed of conglomerate, the pebbles
of the latter have been frequently caught up and enveloped in the
mass of vesicular and scoriaceous rock forming the base of the lava
stream, in a manner which has often been described as taking place
in the case of the products of recent voleances*.

In a road-cutting in the vicinity of Oban I found a very interesting
illustration of the phenomena presented at the contact of some of the
lava streams with the beds below. The section was as follows :—

(a) Compact dark-coloured felstone lava, which has apparently undergone con-
siderable alteration ; here and there a hollow vesicle occurs in the mass. A
thickness of 15 or 20 feet of this rock is seen in the section.

(6) About nine inches or a foot from its base this mass of lava becomes highly
vesicular or scoriaceous, the cavities being strikingly flattened and drawn
out. Many burnt-looking fragments (lapilli) are caught up in this part of
the lava stream.

(c) For a depth of about 2 feet below we have a confused mass of vesicular lava,
rounded blocks of lava, pebbles of quartz, &c., all imbedded in a matrix of
highly scoriaceous rock. The lower part of this mass passes in places
into a sandy conglomerate; the ordinary coarse sand consisting of lava de-
tritus constituting the matrix.

(d) Conglomerates, composed of pebbles of trap and of the Lower Silurian rocks,
with some unmistakable volcanic bombs, and with a matrix of very coarse
sand, 2 feet thick. In this bed are lenticular patches of a very fine-grained
sedimentary material, probably stratified volcanic dust.

(e) Beds of well-stratified and finely laminated red sandstone, very similar in
character to the typical rocks of the Old Red Sandstone.

The stream of lava of which (a) and (b) form the lower portion
appears to be 40 or 50 feet in thickness, to be highly vesicular in its
upper part, and to be covered directly by another lavastream. The
appearance presented by (6) and (¢) is exactly such as would be
produced by a lava stream with a cindery crust which it rolls over
as it flows along, in the manner so graphically described by Mr.
Scrope, and passing over a mass of shallow-water deposits, the mate-
rials of which it entangles in its course.

The sandstones and conglomerates, which in places attain to a
thickness of several hundreds of feet, appear to be wholly confined to
the lower part of the volcanic series of Lorn. In its upper part

* In the island of Lipari there occur beautiful instances of lava streams of

glassy character having entangled in their mass many fragments of the older
rocks, that have evidently strewed the surfaces over which they flowed.

286 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

that. series consists of stream after stream of lava piled one upon.
the other in almost endless succession; and so far as I have ob-
served, these include between them no stratified deposits. The mode
of weathering peculiar to “ trap” rocks, while it often exhibits large
faces of the different lava currents, causes their junction to be
usually buried beneath a talus of fallen fragments. Nevertheless
we are able to observe at many points, as near Loch Feochan, Loch
Nell, and Glen Lonnan, that between the old lava streams there
occur masses of a peculiar rock, usually of a bright red colour. At
first sight these red rocks present but little resemblance to “ trap-
tuffs,” being often extremely hard and compact in texture, through
the infilling of their pores with crystalline materials. Fresh frac-
tured surfaces, however, show that the rock is made up of angular
fragments of lava, &c. (Japilli), quite unstratified; and here, again,
weathering processes come to the aid of the geologist, and, by re-
moving to some extent the crystallized materials from the interstices
of the rock, reveal its true tufaceous structure. Occasionally too
we find in the beds unmistakable fragments, and more rarely
entire examples, of volcanic bombs; but these it is seldom possible
to extract entire, owing to the jointed structure which they have
acquired since being imbedded in the rock wherein they lie.

5. Conditions under which the Voleane series of Lorn was de-
posited.The highly irregular surface of the old slate rocks, with
its hollows filled by breccias composed of their angular frag-
ments, points to the existence of an old land-surface. Upon
this, as we have seen, the accumulation of lava sheets and great
masses of conglomerate and sandstone appears to have gone on
simultaneously for a very considerable period. The stratified rocks
associated with the lower lava sheets of the Lorn series are of such
an extremely local and irregular character, that they do not by any
means necessarily imply a subsidence of the old land-surface beneath
the sea-level. On the contrary, the accumulation of large blocks
of the local rocks, with alternating seams of sand of various
degrees of fineness, would seem rather to indicate the action of
mountain-streams subject to violent floods. The causes so clearly
described by Mr. Drew* as having given rise to the formation of the
great fan-shaped masses of alluvium in India (the nature and dis-
position of the materials of which would seem to be strikingly simi-
lar to those of the beds we are describing) appear to me to be fully
competent to the production of the peculiar deposits of Lorn.

It therefore appears that the remarkable series of rocks in Lorn
was accumulated along the flanks of a range of lofty volcanoes, the
heaping up of conglomerates by mountain-torrents and the out-
pouring of lava streams, under which these were buried and pre-
served, going on side by sidey. During the formation of the upper

* Quart. Journ. Geol. Soe. vol. xxix. p. 441.

t Mr. Scrope has clearly described the important part which is played by
violent floods (resulting either from the sudden melting of snows on volcanic
cones during their eruptions, or from the condensation of the enormous gquan-
tities of vapour to which they give origin) in distributing and rearranging the
_ materials thrown out from volcanic vents. —

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 287

and larger part of the Lorn series the volcanic forces appear to have
been almost the sole agency concerned in the accumulation of the
rocks; for these consist of subaerial lavas alternating with unstrati-
fied tufts.

6. Age of the Volcanic series of Lorn.—In the entire absence of
paleontological evidence, this question must be admitted to be still
to a certain extent an open one. In forming a judgment concern-
ing it, however, we have the following facts to guide us :—

(1) The very marked unconformity between these rocks and the
strata of Lower Silurian age on which they lie. The latter were
evidently not only deposited, but bent into great contortions and
folds, and brought to their present metamorphosed condition, before
the period of the formation of the former.

(2) The fact that the Mesozoic strata appear to rest upon them,
and, in some instances at least, quite unconformably.

(3) The resemblance which the lava streams, of which they are
so largely composed, present in their petrological characters to
others on the south of the Grampians, which sometimes alternate
with strata containing fishes of Old-Red-Sandstone species.

With these points of evidence before us, we can scarcely hesitate
to regard the volcanic and associated rocks of Lorn as belonging to
some part of the Newer Paleozoic periods; but to hazard any closer
approximation to their age would probably be unsafe.

7. The Newer Paleozoic Lavas of the Lowlands of Scotland.—
South of the Grampian Mountains very large areas of Scotland are
occupied by series of old lavas, which, so far as mineral characters
go, present very striking resemblances to those of Lorn which we
have been describing. These rocks have been more or less fully
described by many authors, and the evidences of their volcanic origin
and of the conditions under which they were accumulated very ably
discussed by Maclaren and Prof. A. Geikie. Moreover the districts
where they are developed are now being mapped by the Geological
Survey, and the relations and characters of these old lavas are being
very admirably illustrated by the maps, sections, and memoirs issued
by that department. Under these circumstances it will only be
necessary for the purpose of the present argument to refer briefly to
the general characters presented by these interesting rocks.

Nearly the whole of the hill-ranges of Central Scotland, such as
the Ochils, Sidlaws, Pentland and Braid Hills, Campsie Fells, Kil-
patrick Hills &c., are composed of these volcanic rocks. The pre-
servation and present positions of the masses of these rocks which
still remain, appear to have been determined by the great N.E. and
S.W. faults which traverse the country. The exact positions and
effects of these lines of fracture are very admirably shown on the
detailed maps of the Survey, so far as it has gone.

The lavas composirg these great ranges of hills, which are in
some cases over 2000 feet in height, consist for the most part of
rocks of a more or less felspathic character, though not usually so
highly siliceous as those of the earliest Tertiary period. Like those
of Lorn they commonly display in avery marked manner the

288 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

porphyritic structure. Occasionally, however, streams of a more
basic character, and approaching to basalts in composition, alternate
with the felstone lavas. With regard to the nomenclature of these
rocks great confusion and uncertainty still unfortunately prevails.
By Jameson and his pupils they were distinguished as porphyries,
clinkstones, compact felspars, claystones and basalts. In the earlier
publications of the Geological Survey they are referred to as “ fel-
stones ;” and in the later works published by that department an
attempt has been made to assimilate their nomenclature to that
adopted on the Continent by applying to them the terms porphyrite,
melaphyre &c.

The lavas of Central Scotland, like those of Lorn, have, as a
rule, undergone a far greater amount of change from chemical action
than those of the Tertiary period. Consequently in many cases
their mode of origin is by no means so obvious as in the case of
these latter; and in some places not only have they undergone
a very great amount of general alteration, but mineral veins have
been formed in their mass.

Strikingly similar as the lavas of Lorn are to those of Central
Scotland in their petrological characters, they nevertheless present
in their relations one very marked point of contrast. While the
lavas of the northern flanks of the Grampian Mountains are, with
the exception of the conglomerates and sandstone near their base,
only separated by thin bands of interstratified ash &c., the similar
lavas of Central Scotland alternate with great masses of sedimentary
rocks, conglomerates, sandstones, limestones, shales, and beds of
stratified ash. While the rocks of the former appear to be the
products of a series of subaerial volcanic eruptions, those of the
latter are not less clearly seen to have been formed by the outburst
from time to time of streams of lava, which flowed over the beds of
seas or lakes in which the accumulation of stratified sediments was
going on.

With regard to the age of the great lava sheets of the Scottish
Lowlands we are fortunately supplied with the most satisfactory
evidence. The associated contemporaneous sedimentary rocks
occasionally contain fossils; and thus we are able to define the age
of different parts of them as Lower, Middle, and Upper Old Red
Sandstone and the lower part of the Carboniferous (Calciferous
Sandstone). Some difficulties still remain, it is true, with regard to
the ages of some of the isolated portions of what once constituted
great plateaux of volcanic rocks covering a very large portion of
Scotland. This is due to the fact that many of the fossils, especially
those of the Old Red Sandstone, are very local in their mode of
occurrence ; but that strata belonging to all the periods enumerated
are found interbedded with these ancient lavas is indisputable.
Towards the latter part of the period indicated (namely, during the
deposition of the Calciferous Sandstone) the lavas show a general
tendency to a more basic character than those of earlier periods, and
thus form a transition to the basaltic eruptions of a sporadic cha-
racter by which the great series of volcanic phenomena extending

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 289

through the whole of the Newer Palewozoic periods was brought to
a close.

8. The Eruptive Masses of the Grampian Mountains.—From Peter-
head, on the north-east, to the Ross of Mull, on the south-west, there
occur, along the whole course of the Grampian Mountains, as is
well known to all geologists, a series of masses of crystalline and
igneous rocks which have evidently been protruded through the con-
torted and metamorphosed strata of the Lower Silurian. At an early
period in the history of geology these rocks excited much attention,
from their bearing upon the controversies then raging between the
supporters of the Neptunian and Plutonic theories respectively. Many
of the phenomena were most carefully studied, especially in Glen
Tilt, where Hutton found the most beautiful illustration of his theory
of the igneous origin of the granites, and where Playfair, Webb Sey- |
mour, and Macculloch by a series of patient observations contributed
so much towards the establishment of the truth of that theory.

The largest and most important of the igneous masses of the
Grampians are the great granitic bosses, which give rise to so many
varieties of physical feature—from the lofty mountains of the Cairn-
gorm, Beinn Nevis, and Beinn Cruachan groups, to the low, undu-
lating, and sometimes almost level tracts of the Ross of Mull, the Moor
of Rannoch, and many districts in Aberdeenshire. Where, through
extensive denudation, the lower and deeper portions of these masses
are exposed, they are found to be composed of an almost uniform mass
of typical granite, such as is so well seen in the Ross of Mull; but
where, on the other hand, they rise into lofty peaks, the granite is
found to become more and more hornblendic, and then to graduate
through euritic varieties into a felsite, usually more or less porphyri-
tic in structure (porphyry of authors). The same changes are often
found to take place in the characters of the rocks composing these
great eruptive masses when we trace them from their central portions
towards their outer margins.

The relations which exist between the great granitic masses and
the stratified rocks among which they lie have been clearly described
by many observers. Whenever we approach such intrusions the
prevalent strike and dip of the beds in their neighbourhood appear
to be more or less interfered with, and the strata are seen to be
frequently affected by sudden disturbance and contortion. Moreover
the uniformly metamorphosed Lower Silurian rocks are found in the
neighbourhood of all these masses to have undergone further and
striking changes, which are of a comparatively local character.

It is evident that these great masses of granitic rock have been
forced in a fluid or semifluid condition through the strata among
which they lie, and which, in their passage, they have disturbed and
altered. This is confirmed by the fact that when we examine the
junction of the granite and stratified rocks; we find the former sending
off numerous veins into the latter, which veins often include angular
fragments of the traversed rock, that have been caught up and
enclosed in their substance. Moreover similar fragments, often of
large size, but more or less altered on their surfaces and frequently

290 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

traversed by granite veins, are found actually imbedded in the gra-
nitic masses. And, further, it appears that considerable tracts of
stratified rocks, preserving all their usual characters, were actually
enveloped by the fluid igneous protrusions; and thus we have often
the most complex entanglements of the granitic and stratified rocks.
These facts are very clearly exemplified in the fine coast-sections of
the Ross of Mull, and more obscurely, owing to the want of sections,
at many points in the Grampian Mountains.

But besides these great granitic masses and the veins in immediate
connexion with them, the whole district for many miles around bears
witness to the igneous activity to which their origin is referred. In
all directions smaller masses of granite, syenite-granite, or felsite
are found, either forced between, or cutting across the strata ; while
dykes of similar materials, passing into felstones, are found traversing
the surrounding rocks, in all directions and often to great distances
from the gréat central igneous masses. Such veins and dykes usually
occur in prodigious numbers near the great granite protrusions, as
is so well seen in the passes of Glencoe and Brander ; but their fre-
quency diminishes in proportion as we recede from the central
intrusive masses.

The smaller intrusive masses and the dykes effect a certain amount
of local metamorphism in the rocks which they traverse ; but this is
far more limited in extent and less striking in character than that
produced by the great central granitic masses. Where, however,
the stratified rocks are traversed by a complete plexus of dykes, as
in Glencoe, the amount of alteration produced by them in rocks
through which they pass is often very considerable.

The period at which these intrusions of igneous rocks (which are
almost uniformly of acid or felspathic composition) took place we can
approximately determine. No one can study the distribution of the
various rocks which make up the Lower Silurian strata of the High-
lands, now that the true succession of these beds has been so clearly
demonstrated by the labours of Sir Roderick Murchison and Profes-
sor Geikie, without perceiving that the igneous protrusions were
posterior, not only to the deposition of the Lower Silurian rocks, but
also to the disturbance and metamorphism by which they have
acquired their present characters. This necessary corollary to the
main conclusions arrived at concerning the succession of the Lower
Silurian strata was clearly recognized by the authors whom I haye
cited *.

On the other hand, that these granitic protrusions took place prior
to the deposition of both the Secondary and Tertiary strata, is suf-
ficiently shown by the fact that no traces of the extensive network
of veins and dykes proceeding from them are in any instance found
traversing these younger rocks.

Thus it appears certain that the igneous rocks of the Grampian
Mountains must be referred to some part of the Newer Paleozoic
periods. Pebbles of these granitic rocks are entirely absent from
the conglomerates of the Old Red Sandstone, which, however, are

* Quart. Journ. Geol. Soc. vol. xvii. p. 228.

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 291

often wholly made up of fragments of Silurian strata. This absence,
is very striking when, as sometimes occurs, masses of granite
are now exposed in the immediate neighbourhood of the conglome-
rates ; and it affords a strong confirmation of the age which we have
assigned to the granites, by showing that at the time when the Old
Red conglomerates were formed, the granites were not exposed at
the surface.

Now, in the whole of the phenomena presented by the igneous
masses of the Grampian Mountains, we have an exact counterpart of
those which we have seen are displayed, as the result of subterranean
action, in the case of the great Tertiary volcanoes of the Hebrides.
In both cases we find great intrusive masses of granite, passing by
insensible gradations into syenite-granite and felsite ; and these, in
rising through the surrounding strata, violently disturb and greatly
metamorphose them. With the great bosses of igneous rock are
evidently connected, in both cases, those smaller masses and dykes
of similar materials, which traverse the surrounding rocks, producing
in them degrees of disturbance and metamorphism proportional to
their bulk.

We have already seen that both north and south of the Grampian
Mountains we have proofs of the former existence of extensive
plateaux composed of lavas, for the most part of a felspathic charac-
ter. Unmistakable evidence, both stratigraphical and palzonto-
logical, and of totally independent character in either case, has led
us to the conclusion that the subaerial felstone and porphyrite lavas
of Lorn, the similar subaqueous lavas of Central Scotland, and the
intrusive masses of identical ultimate chemical composition in the
Grampians, were all formed during the same geological periods, those
constituting the latter part of the Paleozoic epoch. From the rela-
tions shown to exist between the granitic rocks and lavas of Tertiary
age already described, we are thus led to see that there are strong
grounds for the presumption that a connexion between these different
igneous rock masses once existed, and that they all form fragments,
now isolated from one another by extensive denudation, of the same
great series of volcanic outbursts. I shall now proceed to adduce
evidence which is, I think, sufficient to convert this strong presump-
tion into almost absolute certainty.

9. Relations of the Igneous Rocks of Beinn Nevis and Glencoe.—
In spite of the arguments already brought forward in support of the
view that the subaqueous lavas of Central Scotland and the subaerial
lavas of Lorn originally formed parts of one series of wide-spread
plateaux, and that the great igneous masses of the Grampians are
the relics of great centres of eruption during the same period, some
may be inclined to regard it rather as a bold speculation than as an
established scientific conclusion. It is a fortunate circumstance,
therefore, that we are able to point to more than one relic of these
volcanoes of the Newer Paleozoic period, the study of which is suffi-
cient, I believe, to convince the most sceptical upon the subject.

Amid that broad belt of elevated ground constituting the Gram-
pians, the peak which rises superior to all the rest is—that monarch

292 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

among British mountains—Beinn Nevis. The superior elevation of
this peak has resulted in the preservation of rocks elsewhere wholly
removed by denudation; and the mountain group of which it forms
a part is as conspicuous for the value of the evidence which it affords
to the geologist, as itis for its great elevation and its strikingly
picturesque features. Let us therefore ascend this mountain and
study its structure.

Beinn Nevis consists of a well-marked and, to a great extent,
isolated group of mountains, having a circumference of about 20 miles

Fig. 11.—Diagrammatic Section illustrating the Relations of the
Rocks forming Beinn Nevis.

Corrie. Cairn. Lake.
’ '

1
1
1
!
'
1
!
1
1
1
i
!
1
'
\

a. Lower Silurian rocks, greatly contorted and metamorphosed at their junction
with the granite.

b. Coarse-grained porphyritic granite, with many ‘‘contemporaneous veins’
graduating into c.

c. Fine-grained granite graduating into felsite.

d. Felsite sending off veins into ce.

e. Felstone lavas and volcanic agglomerates alternating with one another.

’

and culminating in a peak 4406 feet above the level of the sea
(woodcut, fig. 11).

The outskirts of this mountain group are composed of the gneiss,
schist, quartzite and limestones of the Lower Silurian, the highly
inclined strata of which are seen striking N.E. and 8.W., on either
side of the. great intrusive mass of granite which cuts across them
abruptly. From the great central granitic mass there proceed nume-
rous offshoots, sheets, veins and dykes, composed of granite, syenite-
granite, felsite, &c., which traverse the stratified rocks in all direc-
tions, penetrating between or cutting across their beds.

As we approach the central mass of igneous rocks, the regular
dip and strike of the stratified rocks is found to be greatly interfered
with, considerable disturbance and contortion being manifested among
them. Further, in proximity to the intrusive igneous mass, the
already metamorphosed Lower Silurian rocks are found to have under-
gone still greater alteration; the uniform flaggy masses pass into
highly micaceous, chloritic, taleose, hornblendic and actinolitic schists,
while the associated limestones become more highly crystalline and
intermingled with serpentine,

The great mass of the central mountain-group of Beinn Nevis is

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 293

composed of hornblendic granite, passing by insensible gradations
into ordinary granite on the one hand, and into syenite granite on
the other. The normal character of this rock is that of an aggregate
of white orthoclase and oligoclase felspar with colourless quartz and
hornblende, the latter being usually in part replaced by varying
proportions of lepidomelan ; and the whole mass is rendered beauti-
fully porphyritic by the dissemination through it of fine crystals of
orthoclase felspar of a pink colour. Locally, however, it exhibits
many variations from the typical character. It is traversed, too, in
every direction by veins of various size, composed usually of
granite of finer grain, of euritic felsite, or of crystallized quartz and
felspar. The granitic rocks constitute those great spurs, with sharp
summit-ridges and steeply sloping sides, which divide the deep corries
that form so striking a feature in the Beinn Nevis group.

Let us now ascend the central peak and examine its structure.
Some distance above the well-known lake, which lies in a hollow
upon the shoulders of the mountain, a remarkable change is found
to take place in the character of the granitic mass; it becomes much
finer-grained, and as we still ascend we find the hornblende and
mica gradually disappearing, till in the end the rock becomes a finely
granular felsite of a pale red colour, and often more or less porphy-
ritic in structure.

The highest portion of the mountain, however, is composed of a
mass of rocks of totally different character. Instead of the pale red
granites, eurites, and felsites, we find dark-blue, grey, greeenish,
and purplish felstones; and associated with these are enormous
masses of volcanic agglomerate, composed for the most part of
angular fragments of all sizes, of felspathic materials, heaped
together in the wildest confusion, and compacted into masses of
great solidity and hardness.

The confusion produced by the ever descending fragments which
cover the upper slopes of the mountain and constitute such im-
pressive evidence of the potency of atmospheric agencies in the
work of denudation, is inimical to a complete study of the relations
of the rocks forming the summit of the mountain; but an atten-
tive examination both of these slopes and of the more accessible of
the precipices surrounding the corries shows that the felstones form
great sheets, sometimes exhibiting a rudely columnar structure,
and that between them lie the enormous masses of volcanic agglo-
merate, the whole being traversed by innumerable felstone dykes.

From the same cause, namely the abundance of débris on the
higher slopes of the mountain, the relations of the underlying
felsites and granite to the great cap composed of felstones and
agglomerates is not at first sight very apparent. Careful study,
however, shows that the former rocks send off veins into the latter ;
and the abundance of the fragments of the granitic rocks, above the
level at which the lavas and agglomerates commence, shows that
these veins are by no means few in number.

With regard both to the felstone lavas and the associated agglo-
merates, the changes produced by the chemical action that has takea

Q.J.G.8. No. 119. Y

294 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

place in the mass has been very great. But in this, as in so many
similar instances, weathering action is found to a great extent
undoing the work of metamorphism, and revealing original struc-
tures that had been whoily obscured by it; thus many rocks, now
of highly crystalline or compact character, are shown to have been
originally composed of highly vesicular or scoriaceous materials.

The remarkable characters and relations of the rocks forming
Beinn Nevis were not unnaturally the cause of much doubt and
difficulty during the infancy of geological science. Almost the
only attempt at their explanation hitherto made, consists in the
suggestion that masses of ‘‘ porphyry ” had been forced through the
midst of an earlier-formed mountain of granite.

But to any one who has studied the Tertiary volcanic rocks of
the Hebrides nothing can be clearer than the significance of the
features exhibited by Beinn Nevis. Indeed it is evidently but a
repetition, on a grander scale, of many a Tertiary mountain like
Beinn Uaig in Mull, which we have already described. In both
alike we see the evidence that while felspathic lava streams and
scorie were being erupted at the surface, masses of molten mate-
rials of the same composition were intruded below them, and by
slow consolidation forming those bosses of felsite which pass by
insensible gradations, in their lower and deeper parts, into horn-
blendic and ordinary granite. ‘That both alike constituted portions
of great volcanic piles is, I believe, from the facts adduced in this
memoir, placed beyond dispute.

Scarcely less interesting, in their bearing upon the present inquiry,
are the phenomena displayed in Glencoe. At the northern end of the
celebrated pass we find the Lower Silurian strata undergoing consider-
able disturbance and great metamorphism; and they are also seen
to be traversed by a wonderful plexus of veins, dykes, and intrusive
masses composed sometimes of hornblendic granite, but more usually
of numerous varieties of red felsite (‘‘ porphyry” of authors). The
stratified rocks are overlain by great masses of- felstone, with some
beds of the so-called ‘‘brecciated porphyry”: (consolidated ash,
scorie, &c.), which are, however, by no means’so abundant here as
on Beinn Nevis. These felstones are traversed, like the stratified
rocks upon which they rest, by an almost infinite number of veins
and dykes, usually composed of felsite; and they have in many
cases undergone a certain amount of alteration from heat, similar to
that which we have seen taking place in the Tertiary lavas under
like conditions, and resulting in the development in them of a
peculiar banded structure and splintery fracture.

These felstones occupy the whole of the central parts of the pass,
and weather into those striking forms which characterize the
scenery of this famous spot. At the southern end of the pass
there appear from underneath these altered felstones, the felsites
passing into granites, and enclosing masses of often highly altered
Lower Silurian strata, which constitute the great intrusive mass of
the Black Mount.

Here it is evident that there has escaped destruction by denuda-

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 295

tion a portion of the lavas and associated beds which formed the
outskirts of one of the great Newer Paleozoic volcanoes—the
lavas of Glencoe lying in part on the disturbed and metamorphosed
stratified rocks, and in part on the great eruptive masses which
have produced that disturbance and metamorphism.

Besides Beinn Nevis and Glencoe there are some other points
which exhibit, though in a less striking manner, relics of these
doubtless once grandly developed masses of lava and fragmentary
materials, which during the Newer Paleozoic periods formed those
great volcanic cones of which the granitic masses of the Grampians
constituted the central cores, and which rose above vast pla-
teaux of the felspathic lavas that issued from them; of these
plateaux only small and outlying fragments have, by a combination
of accidents, escaped destruction by denudation. There is one fact
which I have not yet noticed, but which is of a sufficiently striking
character to arrest the attention of every geological observer; I
allude to the very marked similarity in petrological characters
between the vestiges of the lava rocks preserved in Beinn Nevis and
Glencoe on the one hand, and those which make up the larger areas,
as of Lorn and central Scotland on the other.

10. Physical Features of Northern Scotland during the Newer
Paleozoic periods.—That the different deposits of the Old Red
Sandstone were accumulated in lakes, has been suggested by Mr.
Godwin-Austen, and supported by a variety of arguments by Prof.
Ramsay. The peculiar characters of the deposits of this age in the
districts referred to in this paper appear to accord perfectly with
this hypothesis.

The conditions which prevailed during the earlier part of the
Newer Paleozoic periods would appear to have been as follows :—In
what is now the Scottish Highlands an extensive land area existed,
with a number of more or less isolated lakes or inland seas. Along
the line now occupied by the Grampian Mountains rose a range of
great volcanic cones, while numerous intermittent eruptions took
place at the bottom of the great sheets of water; the centres of
eruption of some, at least, of these have been pointed out by Mr.
Maclaren and Prof. Geikie. Thus while great masses of lava, with
agglomerates, tuffs, and ashes, were accumulating round the great
volcanoes, other enormous deposits were simultaneously formed, con-
sisting of alternations of similar lavas with stratified rocks—the latter
being composed in part of the detritus of these lavas, in part of
fragmentary eruptive materials which fell into the lakes, and in part
of the materials derived from the older rocks which formed their
shores. A similar series of operations went on during the next suc-
ceeding period, that of the Calciferous Sandstone, though, in conse-
quence of physical changes, lacustrine were exchanged for estuarine
and terrestrial conditions.

During the whole of these periods much local subsidence must
have taken place, to permit of the accumulation of such enormous
thicknesses of rocks; and the resulting disturbances led to the
production of those local unconformities which have been noticed

y2

296 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

by Prof. Geikie and other authors. These unconformities do not
appear, however, to have the same significance which we naturally
attribute to phenomena of the same kind when exhibited by rocks
which have been formed under less violent conditions.

We have seen that, during the later portions of the Tertiary
epoch, the eruptions from the great centres of volcanic activity were
succeeded by a series of sporadic eruptions, resulting in the formation
of numerous small cones or “ puys;” and in this we find a close
analogy with what has taken place in the case of many modern
voleanoes. The great voleanoes of the Grampians, which were in
activity during the Newer Paleozoic periods, also appear on their
extinction to have been succeeded by the outburst of numerous
“puys.” “The area of the Lothians and Fife,” says Prof. Geikie,
“appears to have been dotted over with innumerable volcanic
vents ;” and these sporadic eruptions seem to have continued
during a great part of the Carboniferous and Permian periods.
These latest eruptions of the Paleozoic epoch consisted of materials
of more or less basic character; and the wide-spreading sheets of
melaphyre and similar rocks connected with them, and intruded
between the older strata, have given rise to those numerous “ crags”
which, often crowned by old castles, form such a striking feature in
Lowland scenery. The most familiar type of these later paleozoic
““puys” is the well-known Arthur’s Seat, near Edinburgh, the
origin of which has been so well illustrated by Maclaren, Edward
Forbes, and Prof. Geikie; but mnumerable other beautiful examples
of the same kind occur, some of which are described in the publica-
tions of the Geological Survey.

IV. Conclusion.

1. Comparison of the two Great Periods of Volcanic Activity in
Scotland.—From the facts described in the present paper it appears
that the British archipelago, in common with the surrounding
districts, has been the theatre at two distinct periods, since the
deposition of the Silurian rocks, of exhibitions of volcanic pheno-
mena upon the very grandest scale. ‘The first of these epochs of
violent igneous activity appears to have lasted from the commence-
ment of the Old-Red-Sandstone period down to the close of the
Paleozoic era ; the second, during nearly the whole of the Tertiary
epoch. The interval between these two grand displays of igneous
forces, namely that during which the Mesozoic strata were de-
posited, appears to have been one of comparative, perhaps of
complete, quiescence of volcanic action. Bearing in mind the
arguments of Mr. Darwin in support of “the identity of the force
which elevates continents with that which occasions volcanic
outbursts,” it is not a little interesting to find that the periods of
maximum volcanic activity, namely the Old-Red-Sandstone and the
Miocene, appear to have coincided with those during which (as

hown from various considerations by Professor Ramsay) a great
extent of continental land prevailed in the same areas.

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 297

A comparison of the products of the volcanoes of the Newer
Palaeozoic and Tertiary periods respectively, leads us to the following
conclusions.

During both of these periods of igneous activity, as in the case
of volcanic outbursts of more recent times, the extrusion of lavas of
highly felspathic composition has, as a general rule, preceded those
of the basaltic varieties.

But, while in the later of these epochs, the Tertiary, we have a
period characterized by the ejection of lavas and the intrusion of
subterranean masses of highly silicic composition, followed after a
long interval by the outburst from the same vents of highly basic
igneous rocks, in the earlier of the two epochs (the Newer Paleo-
zoic) we witness, apparently, a gradual transition during the enor-
mous periods through which the eruptions continued, from rocks of
a moderately acid to others of a moderately basic character.

The last efforts of voleanic activity in each of these epochs, as in
the case of many, if not all, recent exhibitions of voleanic phenomena,
have been of a sporadic character, and have resulted in the for-
mation of numerous, but comparatively small, volcanic cones or
“ puys.”

At the earlier epoch the range of great volcanoes was thrown
up along a line ranging N.E. and 8.W., coincident with the
direction of the subterranean forces which, both long before and
subsequent to their upheaval, appear to have determined the
elevations, foldings, and great fractures of the rocks of the district.
But at the later epoch the volcanoes were thrown up along quite a
different line, one ranging N. and S.; and there is evidence that at
the close of the Mesozoic era new axes of upheaval were originated,
differing in direction from those which had prevailed during the
whole of the periods from the Silurian to the Cretaceous.

But to the consideration of these general conclusions concerning the
subterranean forces, and their influence in determining the charac-
ters and position of the strata, I shall have to return in the con-
cluding division of this memoir.

2. Influence of Volcanic Action in determining the Characters and
ftelations of the Secondary Rocks of Scotland.—The importance of a
clear conception of the history of the two great periods of volcanic
activity, which respectively preceded and followed the deposition of
the Mesozoic strata, to a complete understanding of the nature and
significance of the features presented to us by the latter, is shown
by the following considerations.

(1) The main features of the physical geography of the area in
which the Secondary rocks which we are studying were deposited,
were to a yery great extent originated by that grand exhibition of
volcanic activity which had only just come to a close when the
formation of the latter commenced. Hence this pre-Mesozoic vol-
canic action had much to do in determining the nature of the rocks
which supplied the materials of the Secondary sediments, and also
the conditions under which they were accumulated,

(2) Constituting, as we have seen it did, a period of rest

298 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

between two epochs of the most violent volcanic activity, the
Mesozoic era may well be supposed to have been characterized by
the subdued efforts of those imprisoned forces which were unequal
to the task of opening “ safety-valves” for their violence at the
surface. I shall show in the sequel that many of the anomalous
characters of the Secondary rocks in this district can only be ex-
plained on the supposition that during the period of their deposition
the area was subject to frequent and great oscillations of level.

(3) It is to the overwhelming masses of matter poured out by
the Tertiary volcanoes that the preservation even of such small
vestiges of the Secondary rocks as remain to us is wholly due. Inthe
case of each of the patches of these strata which we are called upon
to study on the west coast of Scotland, the positions and relations
of its beds, and the nature and extent of its metamorphism, are the
result of that complicated series of voleanic phenomena which
followed their deposition.

3. The “ Geological Record” in the Scottish Highlands.—The
purposes which the Alps have served to continental geologists, have
been to a great extent fulfilled in the case of British observers by
the Scottish Highlands ; and in either of these districts the materials
for the solution of some of the most important problems of physical
geology have been not unsuccessfully sought for. It was not unnatural
that, in the case of both these mountain-groups, the greatly altered
and highly crystalline character of the rocks which compose them
should have been looked upon by the older geologists as proofs of their
great antiquity. But the discovery of fossils in some of the less-
altered portions of the Alpine rocks has quite revolutionized the
views of geologists upon the subject of their age; and, similarly, the
results of recent researches in the Scottish Highlands enable us to
’ refer their great crystalline masses to well-known, and in some
cases very recent, geological periods.

Should any one still cling to the old view that highly crystalline
characters in rocks may be regarded as a criterion of antiquity, I
would point to the Cuchullin Hills of Skye, which have been
regarded as of Laurentian age, but which, from the clearest evi-
dence, we have shown to belong to the Miocene period.

The early maps of the Scottish Highlands were of necessity
purely mineralogical. ‘The first attempt at the production of a
geological map, that is of one in which the classification of the rocks
adopted is based upon the periods of their formation, is the Sketch
Map of Murchison and Geikie, published in 1861. The production of
this was rendered possible by Peach’s important discovery of Lower
Silurian fossilsin the limestone of Durness. We are now, however,
in a position to show that the rocks of the Highlands, far from being
of great and unknown antiquity, include representatives of a great
variety of geological periods. The rocks of Lower Silurian age are of
great thickness, and cover a vast extent of country; and beneath them
we find two older series of strata which may, possibly, represent the
Cambrian and the Laurentian. Of the existence of Upper Silurian
rocks in the Highlands we have as yet no certain evidence; but the

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 299

Old Red Sandstone is admirably represented, in all its divisions,
both by sedimentary and igneous products; while the Carboni-
ferous and Permian (?) also find representatives in rocks of the latter
class. As I shall show in the present memoir, we have fragmentary
but almost consecutive examples of all the Mesozoic formations,
from the Trias to the highest rocks of the Cretaceous series. In the
Tertiary epoch we have the felspathic igneous rocks representing
the Kocene period, the basaltic rocks, the Miocene, and the smaller
scattered eruptions of the “‘puys” the Pliocene; while of the
Pleistocene period the interesting glacial deposits of the Highlands
supply us with a sufficient record. Thus we arrive at a basis for
the construction of a true geological map of the Scottish Highlands ;
but the actual performance of this task is one which will demand a
vast expenditure of time and effort.

4. Light thrown on some problems of Physical Geology by the
Volcanic Rocks of the Highlands.—The admirable labours of those
pioneers of Scottish geology, Hutton, Playfair, Webb Seymour,
James Hall, and Macculloch, have clearly demonstrated the relations
which exist between the granitic and the associated strata of the
Highlands. On the other hand the researches of Scrope among the
extinct volcanoes of Central France, supplemented as these have
been by the observations of Darwin, Lyell, and a host of other
geologists, in the most widely scattered districts, have thrown much
light upon the structure of modern volcanoes and upon the nature
and succession of the phenomena which accompany their formation.
In the rocks of the Hebrides we find the means of connecting
these two series of observations, and thus of arriving at a complete
theory of the action of the igneous forces operating under different
conditions.

The characteristics of Stigmaric and Sigillarie respectively were
known to paleontologists long before the discovery of a section in
the Coal-measures enabled them to be identified, in their true
relations, as the roots and trunks of the same plant. Similarly
physical geologists have separately studied the features of Voleanic
and Plutonic rocks, which the wonderful sections of the Hebrides
now warrant us in affirming to have the closest connexion—the great
intrusive masses being, as it were, the roots of a tree of which the
stem, branches, and leaves are represented by the dykes, lava
streams, and cinder piles of great volcanic cones.

From the active volcanoes of Etna, Vesuvius, and Skaptar Jokul,
the step to the ruined piles of the Mont Dore and the Cantal is an
easy one: and with these last we have but little difficulty in
perceiving the parallelism of the phenomena displayed by the rocks
in the central mountain-group of Mull. From the condition of the
voleanic rocks in Mull to that in Skye the transition is obvious, and
from the latter to Beinn Nevis, and thence to the granitic bosses of
Cairngorm, the Moor of Rannoch, and the Ross of Mull, can easily
be made; nor will any difficulty be experienced in passing from
these latter to the wide-spreading tracts of granite, such ag that of
Leinster.

300 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

Circumscribed though the observations of the geologist necessarily
are, within the infinitesimal periods of human chronology, he may
nevertheless learn to triumph over the painful limitations of time
by availing himself of the opportunity which he has of studying
the same series of operations at various stages of its progress, as pre-
sented in different examples. As was so well expressed by Playfair,
“Tt is true we do not see the successive steps of this progress
exemplified in the states of the same individual rock, but we see
them clearly in different individuals; and the conviction thus
produced, when the phenomena are sufficiently multiplied and
varied, is as irresistible as if we saw the changes actually effected
in the moment of observation ’*.

To this mode of reasoning the geologist is continually compelled
to resort. If, as I believe, it is demonstrated that in wide-spread
intrusive masses of granite and gabbro or serpentinous rocks—in
the more or less isolated mountain-groups of granitic and doleritic
rocks—in the great voleanic mountains in varying stages of decay
and destruction by denudation—and in the long lines of parasitical
cones which surround them, we witness but different portions of
the same grand series of phenomena, then the geologist is enabled
to picture to himself the characteristics and results of volcanic
activity in all its developments and under all its phases.

Thus he is led to the conclusion that the distinction between
Plutonic, Trappean, and Volcanic rocks, convenient though it may
be in practice, is a purely artificial one, and that the whole of
these must be regarded as the products of the same igneous forces
when operating under different conditions. While, on the one hand,
we are led to conclude that great tracts of granite, like that of Lein-
ster, may once have been surmounted by vast volcanic piles, we
cannot, on the other hand, doubt that the subaerial volcanic phe-
nomena of Iceland, Sicily, and the Andes are accompanied by
innumerable igneous injections in subjacent strata and the forma-
tion of masses of granite, syenite, diorite, and gabbro at great
depths beneath them.

EXPLANATION OF PLATES XXII. & XXIII.

In these Plates an attempt has been made to illustrate the relations of the
great masses of volcanic rocks in the Hebrides to one another, and to the sedi-
mentary deposits which they have been the means of preserving.

The sketch map on Plate XXII. is an attempt to exhibit the relations of the
outcrops of the great intrusive bosses, sheets, and dykes of igneous rock to the
lavas, agglomerates, and clder strata in the island of Mull; but this, owing to
the small scale employed, could only be accomplished in a somewhat diagram-
matic manner. ‘To display the complicated interlacings of the almost innume-
rable igneous intrusions of the district would require maps of great accuracy on
the largest scale, and even then could only be attempted after a very detailed
and minute survey of the area. Unfortunately the Inner Hebrides have not
yet been surveyed by the Ordnance Department; and, although the coast-lines
are well laid down on the Admiralty Charts, the geologist meets with almost

* Tllustrations of the Huttonian Theory, p. 101.

J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND. 301

insuperable difficulties in attempting to record and correlate his observations
on the interior portions of the larger islands, from the want of even tolerably
accurate maps.

Although I haye endeavoured to make the best possible use of all the mate-
rials which were available as a basis for my study of the position and relations
of the rocks of Mull, yet nothing more could be attempted, in such a sketch
map as the present, than to convey a general idea of the structure of the island.
It must be remembered, in examining this map, that the forms of the outcrops
of the ramifying masses of igneous rock, as there laid down, are in part deter-
mined by the surface-contours in these highly mountainous districts. Thus the
outcrop of a mass of rock filling a rectilinear fissure, when exhibited on a moun-
tain-side, forms acurved line. It is especially necessary to bear this fact in mind
in seeking to realize from the map the actual relations of the great solid rock
masses. Of course in a map on so small a scale, the size of the dykes and sheets
has had to be enormously exaggerated, while only the most imperfect idea is
conveyed of their vast numbers and complicated relations.

By employing strongly contrasted colours for the two great classes of igneous
rock, I have endeavoured to make their relations to one another as clear as.
possible; while by varying the depth of tint in each case I have sought to illus-
trate the gradations from coarsely crystalline to granular, compact, and vitreous
structures, in rocks of the same ultimate chemical composition. In the scale of
colours on the map I have employed the same petrological nomenclature as is
used in the memoir; it must be remembered, however, that I have been com-
pelled to use some terms in a restricted and, to a certain extent, conventional
manner—as, for instance, in applying the name ‘“felsite” to rocks of granular
and ‘‘felstone” to those of compact texture. The index of colours on Plate XXII.
serves also for the sections on Plate XXIII. On the latter, however, an addi-
tional tint (pale red dotted with black) has been employed in fig. 5 for the
Cambrian, which does not occur in the map.

In Plate XXIII. all the sections are constructed, as far as possible, on a true
scale, the proportions, both vertical and horizontal, being the same in each
instance. But this scale is in each section perfectly independent of that of the
others; consequently the sections bear no relation to one another.

In the section across the island of Mull (fig. 1) a dotted line is added to show
the least possible elevation of the volcano before it had undergone central sub-
sidence and extensive denudation. The asterisks show the difference between
the original and the present positions of what is now the summit of Beinn
More, as explained on page 259.

The publication of the general Map of the Western Isles is postponed till the
appearance of the next paper of the series, as it will serve at the same time to
illustrate the relations of the volcanie rocks to one another and the positions of
the scattered fragments of sedimentary deposits, from which the history of the
Mesozoic periods in this district has to be reconstructed.

Discussion.

Mr. Campsett thanked the author for having taught him so much
concerning the country in which he had been born and bred, and
invited him to inspect models which are described in ‘ Frost and
Fire.’

Mr. D. Forses was gratified to find the subject of the igneous
rocks of Britain taken up in so able a manner, and the subject not
left entirely to continental geologists. The author was fortunate in
the fact that he had occupied the same ground as that which had
already been explored by Prof. Zirkel, whose work had been thus
supplemented. He agreed with the author in regarding Volcanic
and Plutonic rocks, from granite down to the most recent lavas, as

302 J. W. JUDD ON THE SECONDARY ROCKS OF SCOTLAND.

of one and the same origin. He suggested a doubt whether the
older granites belonged to so recent a period as that assigned to
them by the author.

Prof. MasksLyneE considered that this paper would go far towards
settling the question between one half of the petrologists of the
present time and the other half. He accepted the author’s view as
to the Tertiary origin of the granite, inconsistent though it was
with the view of the age of granite maintained by the school of
Richthofen. This paper exemplified the necessity of combining the
observations of the geologist in the field with those of the mine-
ralogist in his laboratory.

Mr. SrrLry remarked on the importance of this paper as furnish-
ing lines of departure for investigating the early physical history of
the earth. He had ascertained that the old lines of voleanic activity
corresponded with the agonic lines of magnetism; and the paper
afforded data which would assist in drawing conclusions as to the
changes in the position of the earth’s axis by a comparison of these
lines with those of the magnetic currents.

Mr. W. W. Smyrx congratulated himself on again hearing of
intrusive granitic dykes such as recalled the earlier days of the
Society, and tended to dispel some accepted ideas as to metamor-
phism. He adduced the recent deposits of Etna and Vesuvius in
illustration of the specimens exhibited by the author.

Mr. Branrorp suggested that there might be in this case only one
form of the existence of granite. When the gradual passage from
schists into granite could be traced, there could be little doubt of its
origin ; and even in its eruptive form the granite might be only the
result of a very complete metamorphism. He agreed with the author
as to the probability of the great horizontal outflows of basalt being
subaerial, and not subaqueous, and instanced analogous examples in
India.

Mr. TippEemay, referring to the great length of time alluded to by
the author as represented in the volcanic rocks of Scotland, wished
to call attention to the rocks in the neighbourhood of Seuir-na-
Gillean in Skye, which consisted partly of basaltic lavas of Miocene
age. These appeared to be tilted by the syenite of Marscou &c. ;
and upon both rested unconformably a great thickness of hyper-
sthene. But furthermore, the base of this latest rock appeared to
be thrown down by a fault with a displacement of some thousands
of feet.

Mr. Jupp, though admitting there were two sides to the question
as to the origin of granite, could not regard intrusive masses of
granite passing through rocks of different mineral constitution as
the metamorphosed representatives of each. He briefly replied to
the other points raised in the discussion.

VIE EO ( 2-1 @

TERTIARY
Hoceé we

NEWER
PALAZ0ZO/C

Quart. Journ Geol Soc .Vol XXX Pl. XXII.

Upper 5

¢ #

OF MULL.

Mimtern Bros. Iith .

ICNEOUS ROCKS.

q Volcanic Agglomerates
| of Basic Rocks.

Angitic Gabbro.
Gabbro.

fas Voleanic Agglomerates
a. of Aad Rocks.

5 ah A aie’ le (eae Ws da

Obsidian &Pitchstone.
| Felstone.
Feleite. .

Used

we

NEWER
PALAZOZO/C

Syenite- Granite.

Caliach Pt

Cairn mor
1126

ERT eras

Quart. Journ Geol Soc Vol XXX. Pl]. XXII.

yhed Rocks

, Aulistan PE

Tp ie a
Ww Shest a Beimt aS Se

Ca Drimfirn\ >

Si SoD nee
inde owe Sa

Kilmorf

Gachna cricei
Alasaid Hp is Faenrayls

ofrrean I.
stle x aBoR®

: oh =

SFileam straid eatv

oF

Seale of Miles.

+

oTighgeadh I4

SKETCH MAP ILLUSTRATING THE STRUCTURE OF THE VOLCANO OF MULL.

Mintern Bros. Iith.

DUP OF MULL.

Guachan
Dearg

Mam Chlachaig

Quart. Journ. Geol Soc Vol XXX. P] XOUI .
Fig. 1,
SECTION THROUGH THE CENTRAL MOUNTAIN GROUP OF MULL.

Dun-da-yu Beinn Tsalla Beetin Quachan 3772 % Se
: Beinn Yarnach A

Glashven

Vam-an Tiondre

Glen Mam Chlachaig
Forsa

Fig. 2.
BEINN GREIG.

1941

Fig. 3.
CRAIG CRAGGEN.

Fig. 4.
BEINN MORE.
Fig. 5.
SECTION AGROSS RUM.

SECTIONS ILLUSTRATING THE RELATIONS OF THE VOLCANIC ROCKS IN THE ISLANDS OF MULL AND RUM. acnssaxseotney

ON THE OCCURRENCE OF SAPPHIRES AND RUBIRS IN SITU. 303

24. Nore on the Occurrence of Sarpurres and Rvustrss in sith with
CorunpvuM, at the CULSAGEE ConuNDUM-MINE, Macon County, NortH
Carotina. By Colonel C. W. Jenxs. (Read February 25, 1874.)

(Communicated by David Forbes, Esq., F.R.S., F.G.S.)

Tne oriental ruby and sapphire are, mineralogically considered,
merely coloured crystals of corundum, which mineral species has
been shown by chemical analysis to consist of the earth alumina in
a crystallized state and nearly pure condition. Where these gems
have been met with, they appear almost always, if not invariably,
to have been discovered in the beds of rivers as waterworn pebbles ;
and although the existence of corundum in small quantities in
granular limestone in Asia Minor and the United States has long
been known, any thing like a deposit of this mineral in siti, suffi-
ciently abundant for commercial exploration, appears to have been
altogether unknown, until the author’s attention was directed to the
occurrence of numerous fragments of corundum on the surface and
in the river-beds of Macon County, North Carolina, which encour-
aged him to make a minute examination of this district, and resulted
in the discovery, in the summer of 1871, of the deposits of corundum
now known as the Culsagee Corundum Mine.

The locality of this mine is a hill, situated about nine miles east
of Franklin, the principal town of Macon County, the summit of
which is some 400 feet above the valley and about 2500 feet above
the level of the sea; geologically it is a boss of serpentine pro-
truded through the surrounding granite.

In the bed of the river, which runs past the south side of this
hill, now commonly known as Corundum Hill, numerous waterworn
pebbles of corundum, often of large size, were met with, along with
small fragments of rubies and sapphires; and subsequent explora-
tions revealed the existence of some fine nearly parallel veins con-
taining corundum, cropping out for a length of about a mile, along
the steep side of this hill in a north-east and south-west direction.

These veins all dip to the south-east, at an angle of about 45°,
and, although generally only a few inches across at the surface,
widen out as they descend into the body of the hill. In the deepest
working, now 75 feet, the vein is seen to be 10 feet thick. The
veins themselves consist of a mass of chlorite, jefferisite, and corun-
dum, the latter forming from about a third to one half of the entire
mass, and occurring as more or less well-developed crystals imbedded
in the other minerals. Along with these the following mineral
species were found in minor quantity :—chrysolite, anthophyllite,
margarite, damourite, felspar, talc, sapphire, ruby, spinel, zircon,
hornblende, staurolite, diaspore, black tourmaline, chalcedony, quartz,
chromoferrite, magnetite, and two new silicates to which Professor
Genth has given the names of kerrite and maconite.

304 C. W. JENKS ON THE OCCURRENCE OF

The corundum is invariably crystallized, some of the crystals
weighing more than 300 pounds; usually the large crystals are
found to enclose plates of chlorite or jefferisite, just as if these
minerals had been entangled in the very act of the crystallization of
the corundum itself.

About 200 tons of corundum have already been extracted from
this mine, and, after crushing, used for grinding and polishing
stones, glass, and metals &c., being found far superior to emery for
these purposes. Taking the sapphire as a standard at 100, the
corundum from this mine was found to have an abrasive power of
no less than between 90 and 97, whilst the best Naxos emery only
stood at between 40 and 57.

The colour of the corundum crystals is very variable. Some are
perfectly colourless and transparent, while others have all shades of
yellow, green, blue (sapphire), pik and red (ruby), many of the
crystals showing different colours in different parts of their sub-
stance—some of the large ones, when fractured, revealing portions
coloured as rubies or sapphires, from which many gems have already
been broken out and cut for setting: most of these have as yet,
however, been of small size, owing to the difficulty of finding large
pieces free from flaws or cleavage-planes passing through them.
There can be no doubt as to these stones being the true gems; and
Mr. H.C. Sorby, to whom the specimens have been submitted,
declares, from their microscopic structure, that they must have been
formed at an elevated temperature, and that they contain the same
well-marked fluid-cavities filled with a highly expansive liquid
(without doubt liquefied carbonic acid) as are seen in the sapphires
from Ceylon; he also attributes the colour of the rubies and of the
green sapphire to chromium, but is uncertain as to the cause of
the blue of the sapphire, suggesting that it may be due to protoxide
of iron or possibly to uranium ; and he sees no reason for doubting
these gems being quite identical with those so long known from
the East.

The serpentine rock of the Corundum Hill has been analyzed by
M. Th. M. Chatard with the following results :—

Sil Califrscepanschatebenipeteniss deme es smusaesiins sade seeds cessaenens 41°58
I Mimraberey Soocasnsdoooquar deo. coo cUab oo dnngdauadooadoDsdoKe oBeoDdaBbONe 0-14
ProtoxiGe OLATOM | ovecsena cso oncescecce or wenene emcees ese 749
Protoxide of nickel (trace of cobalt and manganese)......... 0:34
WE ELOGENEY spdoncdcdoscccundsacobodssocn{a00G0 abo GbonsqUDodGoHNGCS cusbhde 49:28
Binion Pe Seat eS oye enerotso tt ol ee eiec ais ac atlacilape ts ete au chincisersoretamrmanta O11
TOES) Gi WeAVUNOI doocgoggsovoqzo0 s0deUosaqaqdocERboOcODABOboAsqa0000000 1-72

100-66

As several of the minerals occurring in the Culsagee Mine have
been analyzed by Professor Genth, M. Chatard, and Dr. Koenig, the
results obtained by these chemists are given below :—

SAPPHIRES AND RUBIES IN SITU WITH CORUNDUM. 305

Sp. gr. 3°766. Sp. gr. 3°797. Sp. gr. 3:695.

SPINEL Mace naemeceeessnesa set Fine-grained. Coarse-grained. Dark green.
Koenig. Koenig Genth.
PAlltummiiian ie eeeacecomeasscceoe cece sacs 5432 56°58 66°63
Sesquioxide of chromium ......... 3°96 2:28 trace.
Sesquioxide of iron ..............-04- 11-51 9°66 1:80
Protos derote irom seeeeteese reece seer 11°16 14-60 11:35
IMIEFETOIEESED, « Shsoodeqone0aagqo600DnBDnOOHOd 19:05 16°88 19-86
Oxideonicoppers.-raseesscescccsssese /aeeties | IV euses O-11
Oxatlevotenickelteerertrcsacccsccoseasey abamaeiy penn Milmniterrd 0:25
100-00 100-00 100-00
GIGIRTANSTS) cone sdecntneseoebooHOse Broadly foliated. Sealy.
Koenig. Chatard. Chatard.
illcapmas Aaa cu ws ses tunamaimercio eco epele 33°93 33°77 34:00
ANhataehn ies cape saadee ened eniconceeseente 17°38 17-56 20°36
Sesquioxide of iron ...............005 5-42 5-61 4:91
Ex otoxderol ION. esas cece 0:50 0:50 0:42
Provoxid exotmickelye-enascetseaseseeee 0:35 ? 0:57
IMAGTIOSIAL vr. cceecinecsmaesanesas cence 23°43, 22°48 21°71
Homitionbloss) seeeeee teen steectsececee TPIT) 20°30 18°50
100-18 100°22 100:47
Cliloniteteescees-c-c Foliated dark green. Fine scaly.
Genth. Genth.
STUN ane oteacaaoenb adequts came oneace 27-56 29°48
FAV MIMNITI A ees Nyaden teenies cciasecenne 22°75 22:22
Sesquioxide of iron .............-- 2°56 0-70
iProvoxide) ol imOneeesesrseeeeeee = 5:43 5:30
Oxide of nickel with eae 0:30 0-11
Oxide of manganese oe 0:17
(Mia pT eSiai estrone eteesiiese sites seoses 28°47 30°99
Ignition loss..... scan oboodpDSEAobe _ 13.80 11-63
100:87 100:60
Margarite, sp. gr. 3087. Staurolite, sp. gr. 3°711.
Chatard. Genth.
HINGE), co desocsocbaeconacHeseesnG0G066 28°80 27°91
Jima athseh pelengacbosasosecocudueusner 49:57 52:92
Sesquioxidetoturonyvs-essessnece eases 6:87
Protoxid eof LOMessesseereseeter ts 0:34 7:80
IWIBYOTVESIE, Gocooanqaoncooonaosonqadce 0-75 3:28
ILPKN® coocessoouecaoacoocsoHaggeANGIeNC 11:33 trace.
Ignition lOss.......2..0s.ssesssessee 6:64 1:59
Soda, potash and lithia ......... undetermined ......
10037
Maconite, sp. gr. 3827. Kerrite, sp. gr. 2°303.
Chatard. Chatard.
[SWIGIG2). anoooopoogaaseconoanedoDnab00ces 3420 38°26
ANEHCRTIR!, Zaaenaccsducons|0ndc0es005c 21-66 11:42
Sesquioxide of iron ............++- 12°54 ie)
Protoxide of ir0n......6........00 0:32 0°32
Nickel with cobalt ............... 0:13 0:22
Mia oniestan ins seeenstenssclsmecceocreas 14-61 26°50
Mnthias eccceteradaeussnecsesscesss tEACEsy | ete hv ay unmberecee
SOGBHarossamesseaectiascessncusesties O250a Vice ry Paes aaa
Potash :satscneceaeueocasneesenee OOo tas Pelee:
Momiblonelosseeasseesria se steerer 11:90 21:28
101-77 99:97

306 ON THE OCCURRENCE OF SAPPHIRES AND RUBIES IN SITU.

Discussion.

Mr. Warineton W. Smyru considered the communication made
by Col. Jenks as a very important and interesting one. He referred
to the nature of these gems, and expressed a hope that Col. Jenks’s
further operations might result in the discovery of large and fine
crystals.

Mr. D. Forsrs remarked that much credit was due to Col. Jenks
for having followed up the fragmentary evidence which he originally
obtained with such good results. The origin of these gems had
long been a disputed point; all those hitherto obtained have been
found in a waterworn state in the beds of streams. Col. Jenks
had discovered the actual home of the so-called oriental ruby and
sapphire.

Prof. Tennant observed that Mr. Sheppard had years ago brought
home sapphires from the same district. They were obtained from
the beds of rivers.

Col. Jenxs gave some further statements with regard to the depth
to which the corundum-veins referred to in his paper have been
worked, and stated that some of the crystals obtained from the
veins could be broken across by a very slight pressure in the fingers
when first taken from the vein, but that they became hard by
exposure to the air.

R. ETHERIDGE ON ECHINOTHURID® AND PERISCHOECHINIDE. 307

25. On the RELATIONSHIP earsting between the EcutnotauRipm, Wyville
Thomson, and the PrriscHoEcHInipm, M‘Coy. By R. Eruerier,
Jun., Esq., F.G.S. (Read March 11, 1874.)

[Puate XXIV.]

A CONSIDERABLE degree of light has lately been thrown on the rela-
tionship existing between palzeozoic and recent Echini, from mate-
rials obtained during the cruise of the ‘ Porcupine’ in 1869.

Prof. Wyville Thomson, in his ‘ Depths of the Sea’*, described
two new and peculiar genera of Echinoidea, Calveria and Phormo-
soma, which, in certain peculiarities of structure, such as the over-
lapping of the constituent plates of the test, appear to unite the
Echini of the present seas with certain of those genera which existed
during palzeozoic times.

Although the structure referred to had previously been noticed, in
the case of Cretaceous Echini, by the late Dr. 8S. P. Woodward, and
in that of a Paleozoic genus by Prof. James Hall, of Albany, yet
it was not until Mr. J. Young, of Glasgow, drew attention to the
nature of the plates in the Carboniferous genus Archeocidaris that
such structure had been observed in any paleozoic genus found in
this country.

Prof. W. Thomson states+ that in the test of Calveria (woodcuts,
figs. 1 & 2) the ambulacral and interambulacral plates, instead of
abutting directly against one another, and thus forming a rigid and
compact test, overlap each other, those of the ambulacra from below
upwards (that is, from the mouth to the apical pole), those of the
interambulacra from above downwards, or from the apical to the
oral pole. The overlapping of both series of plates takes place at
their inner ends, whilst the outer ends of the interambulacral plates
are separated, the one from the other, by intervening membrane, thus
assisting in the general mobility of the framework. At their outer
extremities the interambulacral plates carry each a primary tuber-
cle, and also overlap the outer extremities of the ambulacral plates,
*‘ go that the ambulacral ares are essentially within the interambu-
lacral” §. The pores of the ambulacral plates are arranged in a pe-
culiar manner, in arcs of three, specially placed.

The second genus, Phormosoma (P. placenta, W. T.), resembles the
preceding in the arrangement of the ambulacral and interambula-
cral plates and flexibility of peristome ; but the overlapping of the
plates is not carried to so great an extent as in Calveria, and no
membranous spaces are left ; so that the test is continuous, as in the
normal forms of the Echinoidea. Phormosoma is chiefly remarkable
for the peculiar characters of the under or ventral surface, where the

* ‘The Depths of the Sea, an account of the General Results of the Dredging-
cruise of H.M.SS. ‘ Porcupine” and ‘“ Lightning,” during the Summers of
1868, ’69, 70.2 London. 8vo. 1878.

+ Geol. Mag. 1873, vol. x. p. 301. t Loe. cit. p. 158. § Loe. cit.

308 R. ETHERIDGE ON THE RELATIONSHIP BETWEEN

Fig. 1.—Calveria hystrix, Wyville Thomson. (Two thirds nat. size.) *
(From Thomson’s ‘ Depths of the Sea,’ p. 156, fig. 27.)
\

¥

Fig. 2.—Calveria hystrix, Wyville Thomson*.
(From Thomson’s ‘Depths of the Sea,’ p. 157, fig. 28.)

Inner surface of a portion of the test, showing the structure of the ambulacral
and interambulacral areas.

* [We are indebted to the kindness of Messrs. Macmillan, the publishers of
Prof. Wyville Thomson’s work, for the loan of these engrayings.—Hp. Q. J. G.8.]

THE ECHINOTHURIDA AND THE PERISCHOECHINID®. 309

distinction between the ambulacral and interambulacral ares becomes
entirely lost.

Ihave quoted thus largely from Prof. Wyville Thomson’s work
that the succeeding remarks on the fossil forms may be thoroughly
understood.

The first of these, Echinothuria, S. P. Woodward, together with
the two preceding recent genera, compose Thomson’s family of the
Echinothuride ; and it is to the consideration of the relation of this
family to the Perischoechinide, M‘Coy, that I purpose to devote the
remainder of these remarks.

Echinothuria was obtained some years ago by Mr. Wickham Flower
from the Upper Chalk of Higham, near Rochester*, and in its prin-
cipal characters agrees completely with Calveria and Phormusoma. It
differs from the first named, however, in the wider interambulacral
and ambulacral plates, in the smaller amount of overlapping of the
same, and in the abseuce of membranous intervals; from the latter
by having the structure of the oral and apical surfaces the same.
(Thomson.)

So far we have examined the characters of those genera which
constitute the Hchinothuride ; and if we now turn to the paleozoic
forms of the Echinoidea we shall find an interesting repetition of
many of the foregoing facts.

Paleozoic Echini are represented by a limited number of species,
comprised in the following genera :—Archeocidaris, M‘Coy, Pale-
chinus, Scouler, Perischodomus, M‘Coy, Lepidechinus, Hall, Kocidaris,
Desor, Melonites, D. D. Owen, Lepidocentrus, Miller, and Olagoporus,
Meek & Worthen.

In 1849 Professor M‘Coy + proposed the order Perischoechinide, to
include paleeozoic Echini possessing more than two rows of plates in
each interambulacrum, as distinguished from those of Secondary,
Tertiary, and living genera, where the same arez are composed only
of two rows.

This complexity of the interambulacral system was found by M‘Coy
to be present in the genera Archeocidaris, Palechinus, and Peris-
chodomus. Since the first enunciation of these characters sufficient
information has been obtained to enable us to add the remainder of
the paleozoic genera mentioned (I am not so certain of Lepidocen-
trus, Miller, and therefore do not wish these remarks to be applied
to that genus with so much certainty as to the others) to M‘Coy’s
Perischoechinide.

Mr. Young’st remarks on the interambulacral plates of Archwo-
cidaris clearly indicate that a certain amount of overlapping took
place in that genus. These, as originally pointed out by M‘Coy §, are
of two kinds, a pentagonal and a hexagonal series, arranged in the
test in three or more rows. The larger pentagonal plates have two
of the sides bevelled (Pl. XXIV. a, fig. 1), against which would fit
the corresponding bevelled surface of the next plate, but in this case

* W. Thomson, J. c. p. 162.
+ Annals & Mag. Nat. Hist. 1849, vol. iii. t Loe. cit. p. 302.
§ Synopsis Carb. Foss. Ireland, p. 173.

310 R. ETHERIDGE ON THE RELATIONSHIP BETWEEN

on the lower edge. Mr. Young also observes that one of the other two
sides is provided with a small groove, into which would fit the edge of
a succeeding plate, and so probably prevent too great a laxity of test.

If, for a moment, we now examine the state of matters as regards
Palechinus, we shall find, judging from M‘Coy’s beautiful figures in
his ‘Synopsis of the Carboniferous Fossils of Ireland’ (pl. 24), that
the test in this genus had no overlapping plates, but was rigid, after
the normal character of the Echini*.

Sufficiently good specimens of Archwocidaris have unfortunately
not been obtained to show a series of these overlapping plates in
juxtaposition; but I think a careful examination of well-preserved
interambulacral plates, with their peculiar bevelled edges, cannot
but prove that these edges could have served no other purpose.
Confirmatory evidence of this is to be met with in a closely allied
genus, Lepidechinus, which is described by Hall as having the
“plates of the interambulacral series imbricating from the dorsal
side, and the lower edges of each range overlapping those below ;
while the plates of the ambulacral series are imbricating in the op-
posite direction” 7, a complete repetition of the arrangement seen in
Calveria. {Note.—In the 3rd vol. ‘Illinois Geol. Report, p. 552,
Messrs. Meek & Worthen remark that Prof. Hall has since stated
that his original description of L. cmbricatus, the type of the genus,
was not absolutely correct, but that the imbrication of all the plates
is exactly the reverse of what he had supposed. He suggests that
he may have been mistaken in regard to which was the dorsal and
which the ventral side. |

From a consideration of these facts, and admitting that if the
existence of overlapping plates be accepted in Archwocidaris, whilst
the presence of simply abutting plates is recognized in Palechinus,
we have in M‘Coy’s Perischoechinide two types of structure—one,
Archeocidaris, with overlapping plates, allied to Thomson’s Echino-
thuride, the other, Palechinus, with abutting plates, allied to the
ordinary forms of Echini, but both still possessing their great
distinctive character, viz. the increase in the number of interambu-
lacral plates, which at once presents a clear line of demarcation be-
tween them and all recent types.

Perischodomus, Melonites, Eocidaris, and Oligoporus have not been
shown to possess overlapping plates, although, as will be presently
pointed out, from their resemblance to one of the two preceding,
it is probable that some of them at least will be found to do so.

I shall now proceed to note the more intimate test-structure of the
genera constituting the Perischoechimde, and showhow theyagree with
this family on the one hand, and with the Echinothuride on the other,
simply premising that they all possess the distinctive character of
the former, viz. the great increase in the number of the interambu-
lacral plates.

* Mr. W. H. Baily has also given good figures of Palechinus, which show
the arrangement of the interambulacral plates. See Trans. Roy. Geol. Soc.
Ireland, vol. i.

+ Descr. New Sp. Crinoidea, Prelim. Notice, Albany, 1861, p. 18.

THE ECHINOTHURIDZ AND THE PERISCHOECHINIDZ. Biol

Genus ArcH#mocrpaRIs, M‘Coy, 1844*. Pl. XXIV. fig. 1.

(Echinocrinus, Agassiz, 1841. Palcocidaris, Desor, 1846.)

In this genus the ambulacra are narrow, and composed of two rows
of small plates, of irregular form, each provided with two pores, The
interambulacra are continuous from pole to pole, and are composed
of three or more rows of plates, each provided with a primary tuber-
cle. The rows of plates bordering the ambulacra are pentagonal in
form, the middle row or rows hexagonal; the large oblong penta-
gonal plates have two of the sides bevelled from the upper surface,
the opposite side being bevelled on the lower edge; one of the other
two sides has a small groove into which the edge of the next plate
was received +; the primary tubercles are perforated and hollow;
apical disk unknown ; mouth surrounded apparently by a membrane
bearing numerous minute imbricated plates t.

Taking this genus as the type of the Perischoechinide, we have the
first approach towards the structure of the Echinothuride, as typified
by Calveria. Whether the interambulacral plates of Archceocidaris
were imbricated from above downwards, and the ambulacral from
below upwards, or vice versa, must remain an open question at present,
as we are not in possession of sufficiently good specimens to prove
this point. In Calveria the peristome is covered with small plates §,
and the primary tubercles, as in Archcocidaris, are hollow.

Archeocidaris is confined to rocks of Carboniferous age.

Genus Patacuinus (Scouler), M‘Coy, 1844|/. Pl. XXIV. figs. 2&3.

The plates composing the test of-this genus abut end to end in the
regular way, and, so far as we know, there is no overlapping or in-
termediate membrane. ‘The interambulacral arez are composed of
from five to six or eight rows of plates, those adjoining the ambulacra
pentagonal as in Archeocidaris, the others hexagonal. The ambu-
lacra, narrow, and sometimes slightly furrowed, are composed of two
alternating rows of small plates, wider than long, each perforated by
two rows of pores. The apical disk consists of five genital plates
each perforated by three pores, (ovarian and seminal), the fifth
being somewhat larger (madreporiform plate?), and five ocular
plates, each perforated by two openings 4.

Further, in the apical disk there are eight plates, answering to the
suranal of the Saleniade**.

It becomes apparent that Palechinus belongs to that division of
the Perischoechinide which more nearly approaches the normal
Echini, from which, however, it differs in two most important cha-

* Synop. Carb. Foss. Ireland, 1844, p.173. + J. Young, loc. cit. p. 302.

t Palzontology of Illinois, vol. ii. p. 294. § W.Thomson, J. c. p. 157.

| 2.@ jon Alia | Baily, Geol. Mag. vol. ii. p, 44.

** Baily, two papers read before the Royal Geol. Soc. Ireland, March 9th and
April 12th, 1864, “On the Structure of Palechinus.” [On the other hand Prof.
de Koninck has described and figured (Geol. Mag. vii. p. 259, pl. 7. fig. 1) the
apical disk of Palechinus with genital plates only, four perforated by three
pores, and the fifth by one pore. ] os
Z

x.

o12 -R, ETHERIDGE ON THE RELATIONSHIP BETWEEN

racters, viz. the triple perforation of the genital plates and the double
perforation of the oculars. It has usually been considered that
Palechinus possessed spines only of one kind, small ones, analogous
to the secondary spines of Archwociduris ; but Baily has shown * that
both primary and secondary spines were present, the former with
tubercles perforated for the passage of the ligamentum teres of the
spine.

Pam penne is chiefly confined to Carboniferous rocks. It is said
to occur also in the Silurian.

Genus PrriscHopomus, M‘Coy, 18497.

Nearly related to Archeocidaris, from which it is principally di-
stinguished by the presence of primary tubercles only on the mar-
ginal interambulacral plates bordering the ambulacra. Each ambu-
lacrum is composed of two rows of small plates, each pierced by a
pair of small pores; the interambulacra are wide, and composed of
five rows of plates of irregular form and shape.

Very little is known regarding this genus. It was described
originally from two very imperfect specimens, consisting only of
portions of the test; and I am not aware that any other specimens
have since been described. With the small amount of evidence at
our command it is impossible to form an opinion regarding the con-
struction of the test; all that can be said at present is, that it is
closely allied to Archeocidaris, although the interambulacral plates
appear to be more irregular in ‘form and size.

Perischodomus, so far as is known, is confined to Carboniterane
rocks. ©

Genus Lepipecuinus, Hall, 1861+.

“The form and arrangement of the ambulacral and interambula-
cral series, as in Palechinus, with the plates of the interambulacral
series imbricating from the dorsal side, and the lower edges of each
range overlapping those below, while the plates of the ambulacral
areas are imbricating in the opposite direction, narrow and deeply in-
terlocking at their joining edges, each plate pierced near the opposite
extremity by two pores” §.

Of all the paleozoic genera of Echinozdea the present one shows
in the most complete manner the imbricating character of the Echino-
thuride. According to Prof. Hall, the interambulacral arex are
many times as wide as the ambulacral, and contain a large number
of-plates of somewhat irregular form, judging from the figures ||. The
imbricating nature of the plates of the test closely connects this genus
with Archwocidaris, from which it may be distinguished by the form
of the ambulacral and other plates, and probably, according to Meek
pad Worthen], by the fact of there being no primary tubercles on

* Baily, loc.citt. ft Annals & Mag. Nat. Hist. 1849, 2nd ser. vol. iii. p. 253.

- { Deser.. New Sp. Crinoidea, Prelim. Notice, Albany, 1861, p. 18; also, 20th
Rep. State Cabinet N. Y. 1867, p. 295.

§ See Note, p. 310. || Ld¢d. pl.9.fig.10. § Pal. Illinois, vol. ii. p. 295.

¢

THE ECHINOTHURIDH AND THE PERISCHOECHINIDS. 313:
any of the interambulacral plates on the underside of the test, ir? -¥
on the marginal plates bordering the ambulacra. If we accept this
statement, founded on imperfect specimens, as given by the latter

authors, it must be borne in mind that Perischodomus has pri- sie

mary tubercles only on the marginal interambulacral plates; so,
presuming for one moment that that genus is ultimately proved to
possess imbricating plates, the two genera, Lepidechinus and Peris-
chodomus, may perhaps be considered identical.

Lepidechinus has hitherto been found only in America, in the
Chemung (Devonian) and Burlington-Limestone (sub-Carbouiferous)
groups.

Genus Eocrparis, Desor, 1858*.

Ambulacra composed of a double series of plates, perforated near
their outer extremities by two small pores; interambulacra of five
or more rows of plates, those bordering the ambulacra, as in the other
genera, pentagonal, the remainder hexagonal, some of the rows of
which become obsolete before reaching the poles; each interambu-
lacral plate carries a primary tubercle, but without the usual miliary
ring surrounding it.

The above characters are taken from Prof. Hall’s emended de-
scription ; and from them it appears tolerably certain that this genus
did not possess imbricating plates, but is distinguished from the
other palzeozoic genera by the absence of the ring surrounding the
primary tubercles, and the decrease in the number of the rows of
interambulacral plates towards the poles.

Eocidaris occurs in the Devonian, Carboniferous, and Permian
formations.

Genus Metonirzs, D. D. Owen, 1846}. Pl. XXIV. figs. 5 & 6.

Body ovoid, spherical ; interambulacral plates hexagonal; ambu-
lacral plates of two kinds—one elongated hexagonal series composed
of two rows in the centre of the ambulacral are, elevated into a
prominent ridge, giving rise to a furrow on each side at the junction
of each row with the second kind of plates, which are smaller, poly-
gonal, ranged in four vertical rows on each side the larger central
rows, thus giving to each ambulacrum no less than ten rows of
plates; each ambulacral plate perforated by a pair of pores, which,
in the smaller polygonal plates, are said by Owen to be central,
whereas, in the larger hexagonal plates, they are placed on one side,
that furthest from the central ridge.

In Melonites we find a still further departure from the regular
Echinid type, in the great development of the rows of plates in the
ambulacral ares, five times as many as those seen in Calveria, Pale-
chinus, or Archeocidlaris. The apical disk in this genus presents
some peculiarities worthy of notice. Both ocular and genital plates
are present, as usual; but in some cases the former are without any-

* Synopsis des Hchinides, p. 155.

t 20th Report State Cab. New York, 1867, p. 297.
t American Journal, 1846, 2nd ser. vol. ii. pp. 225 e¢ seq.

‘R, ETHERIDGE ON THE RELATIONSHIP BETWEEN ©

_ trace of pores, at other times a single pore may be present in two of
them. ‘The genital plates vary even more than the oculars ; for in
some instances three of the plates are pierced by four pores, the
other two by five; again, three of the plates may carry four pores,
and the other two plates three pores only*, showing how great may
be the variation in this respect amongst individuals of this genus.

Genus Otteoporus, Meek & Worthen, 1860+. Pl. XXIV. fig. 4.

In the size and arrangement of the plates composing its ambula-
cral and interambulacral aresze Oligoporus agrees very closely with
Melonites, but differs from it in possessing only four rows of plates
and four double rows of pores in each ambulacrum, in the place of
ten. The ambulacra are deeply and doubly furrowed.

_ Oligoporus appears to be intermediate between Melonites and

Palechinus, possessing half the number of ambulacral plates and
pores of the former, and double the number of the latter. It also
agrees with the former in the deeply furrowed nature of the ambu-
lacrat.

Confined to Carboniferous rocks.

In the foregoing remarks some of the more general characters of
the various genera have been touched upon. I have now to notice
a few deductions to be drawn from them.

1. We have seen that the group Perischoechinide was established
by M‘Coy for the reception of those paleeozoic Kchini with more than
two rows of plates in each interambulacrum,—a character common
to all the genera previously noticed, and thereby distinguishing
them from Recent and Secondary forms.

2. Certain members of the Perischoechinide possessed overlapping
plates, both in the ambulacra and interambulacra, in one at least,
Lepidechinus, on a plan similar to that seen in the recent genus
Calveria.

3. We have no conclusive evidence of the presence of mem-
branous interspaces in conjunction with the overlapping plates as
in the Echinothuride ; but the fragmentary condition in which the
remains of Archeociduris are usually found would lead us to the con-
clusion that such did exist.

4. There does not appear to have been any paleozoic genus, so
far as our present knowledge goes, which exhibits the peculiar want
of distinction between ambulacra and interambulacra on the ventral
half of the test seen in the genus Phormosoma.

5. In two closely allied genera, Melonites and Oligoporus, we find
an increase in the number of rows of plates in each of the ambu-
lacra, similar to that which takes place in the interambulacra
throughout the whole group.

6. Lastly, the Perischoechinide differ from recent and later Echini
in an increase in the number of perforations of the ocular and genital

* Pal. Illinois, vol. ii. p. 228.

t Proc. Acad. Nat. Sciences Philad. 1860, p. 474.
{£ Pal. Illinois, vol. ii. p. 249.

THE ECHINOTHURID AND THE PERISCHOECHINID. 315

plates of the apical disk: in Palechinus the ocular plates are doubly
perforated, the genital plates triply so ; in Melonites the perforations
appear to vary in different individuals.

In conclusion, it will be obvious that the diagnosis of the Peris-
choechinide must be so far modified as to include the characters of
the overlapping plates of Lepidechinus and Archeocidaris, and the
increase in the number of the rows of plates in the ambulacra of
Melonites and Oligoporus.

It has also been pointed out that Mr. Baily has discovered primary
tubercles in Palechinus ; how far this will affect M‘Coy’s division
of the Perischoechinide into two families, based on the presence or
absence of primary tubercles, will very much depend on the cha-
racters of these structures in some of the other genera mentioned.

[Since these remarks were written I have seen the description of
another genus of this group, Lepidesthes, Meek & Worthen, in the
3rd vol. of their admirable ‘Illinois Survey Report,’ p. 522. Here
the interambulacra were composed of plates imbricating from below
upwards and outwards from the middle; the ambulacra were com-
posed of ten rows of small pieces imbricating from above downwards,
with two nearly central pores to each piece. The whole surface
appears to have been ornamented with small granules, probably for
the articulation of small spines. Lepidesthes, in the mode of imbri-
cation of its plates, agrees with Lepidechinus, Hall, as now under-
stood, but differs in the relative breadth of the ambulacra and inter-
ambulacra. In the presence of ten rows of plates in the former, it
agrees with Melonites, but differs in their imbricating character.
Messrs. Meek and Worthen remark that this. peculiar imbrieation of
the plates will probably “be found to be of even more than generic
importance.” |

EXPLANATION OF PLATH XXIV.

Fig. 1. Archeocidaris, interambulacral plates, from drawings kindly supplied
by Mr. J. Young, F.G.S.: a, bevelled edges.
2. Palechinus, portion of an ambulacrum, and the bordering interambu-
lacral plates, enlarged. (From Prof. M‘Coy’s figures, ‘Synop. Carb.
Foss. Ireland,’ pl. 24.)
3. Palechinus, apical disk, enlarged: a, genital plates, with triple perfora-
tions; 6, ocular plates, with dual perforations. (From Mr. Baily’s
figure in a paper read before Roy. Geol. Soc. Ireland, March 9, 1864.)
4. Oligoporus, portion of an ambulacrum and bordering interambulacral
guess (After Meek & Worthen’s figure, ‘Illinois Geol. Rep.’ ii. p. 248,
fig. 28.)
5. Bee aites similar figure to last. (L.c. fig. 27.)
, apical disk: a, genital plates, with pores; 8, ocular plates, with-
out pores. (L.c, p. 228, fig. 22.)
7. Oral disk and teeth of a recent Cidaris: a, the five ambulacral segments,
with notched ‘and perforated plates. (From Mr. S. P. Woodward’s
figure, ‘ Geologist,’ 1863, pl, 18.)

Discussion.

Mr. Eruertper described Calveria as resembling an elastic ball
rather than an ordinary Sea-urchin, its calcareous plates being held

316 Rk. ETHERIDGE ON ECHINOTHURIDE AND PERISCHOECHINID A.

in place by a flexible membrane—and as connecting the ordinary
forms with Echinothuria, in which the plates slide over one another
like armour. He remarked that the apical disks vary in each
genus; in the paleeozoic genera the ovarian plates have three or
more and the ocular plates two perforations. The interambulacral
areas in the paleozoic genera have invariably more than two rows
of plates. In Archewocidaris the plates have bevelled edges. The
chief point of the paper was its indicating that a type supposed to
have been long extinct is still represented in our seas.

Mr. Szrtry observed that the buccal membrane in the recent
Echinoidea has overlapping plates, so that if the development of the
plates usually forming the remainder of the test were arrested, forms
would be obtained approaching those described in the paper. He
stated that in his opinion both the Echinoderm-type and the Bra-
chiopod-type have analogies with the Annelids, which are supported
by a comparison of the multiplied poriform zones of Melonites with
the perforated shells of Terebratula; this he thought indicative of
a far-off affinity between these types.

Mr. H. Woopwarp stated that the author of the paper had done
much towards the investigation of the Carboniferous fossils of Scot-
land. He had detected in the Carboniferous rocks undoubted spi-
cules of Synapta and Chirodota, showing that the soft-bodied Echi-
noderms were in existence at the period of their deposition. Mr.
Woodward further remarked that of the Holothuride some forms,
such as Psolus, are protected by calcareous plates having perfect
freedom of motion.

Mr. Gwyn Jerrrerys stated that Calveria hystria was dredged off
the Faroe Islands, and subsequently in the Bay of Biscay. He
remarked that many missing links will probably be found hereafter,
and that nomenclature will be benefited thereby; thus, if Hchino-
thuria and Calveria really belong to the same genus, one of these
names may be discarded. In support of this view he stated that
the paleozoic Huomphalus is identical, as regards the characters of
the shell, with the recent Homalogyra. It is very desirable that
zoologists and paleontologists should employ the same names.

<

imp

Mintern Bros

G.H Ford.

PERISCHOECHINIDE .

T. F. JAMIESON ON THE GLACIAL PERIOD IN NORTH BRITAIN. 317

26. On the Last Stace of the Guactat Prriop in Nortx Briar.
By T. F. Jamizson, Esq., F.G.S. (Read May 27, 1874.)

ContTENTS.

§ 1. Introductory. § 6. Kaims, Eskers, &c.

§ 2. General features of the surface. § 7. Gravel terraces.

§ 3. Disappearance of the marine beds, § 8. Comparative glaciation of east
§ 4. Moraines at low levels. and west coasts.

§ 5. Freshness of the glacial markings. § 9. Conclusion.

§ 1. Introductory.

In Scotland we seem to have at least three well-defined stages in the
history of the Glacial period, viz. :—1st, the great early glaciation by
land ice; 2nd, the period represented by the glacial marine beds,
containing remains of arctic mollusca, when much of the country
was covered by the sea; and, 3rd, what I shall call the time of the
later glaciers. It is to the history of this last part that the follow-
ing pages are chiefly devoted.

During the first stage glacial conditions seem to have attained
their maximum, and the whole of Scotland appears to have been
covered for a long time with snow and thick ice, as North Greenland
is at the present day. An opinion has occasionally been expressed
that it was the polar ice which at this time spread in a continuous
mass from the arctic circle over a considerable part of the northern
hemisphere, covering much of the north of Europe as well as most
of Britain. I cannot say that I have met with evidence of a mass
of glacier-ice from the polar regions haying moved over the district
with which I am acquainted. On the contrary, there are facts
which seem to me inconsistent with such a notion. One of these is
the occurrence of extensive beds of chalk-flints along the crest of a
range of low hills running for six or seven miles inland from Peter-
head.

These beds of loose flint pebbles, lying, as they do, on the very
top of a range of bare exposed hills at altitudes of from 250 to
370 feet, must have inevitably been swept off by a great mass of
glacier-ice moving over them from any quarter. It is true they
have been carried away here and there, and cut off abruptly in cer-
tain directions, as if by the glacier-ice of the adjoining district; but
for a distance of several miles they lie thickly along the very top of
the ridge. Their quantity, and the extent of ground they cover,
forbid us from supposing that they have been drifted from some
foreign region; but assuming that they are native to the locality, it
may still be said that they were not in the state of loose pebbles at
the time of the ice, but may have been imbedded in strata of solid
chalk, which have since been dissolved. This, however, I consider
a very improbable supposition.

This ridge, although it attains the altitude of only a few hundred
feet, is, nevertheless, higher than any ground immediately to the

318 T. F. JAMIESON ON THE LAST STAGE OF

north of it; in fact it may be said to be nearly open sea between it
and the North Pole.

Ever since I visited the district of Lochaber and the Parallel
Roads of Glenroy in 1361-62 I have inclined to the belief that there
must have been a great development of snow and ice in this country
after the submergence, or second stage of the Glacial period ; and an
examination of the glens around Ben Nevis impressed me at that
time with the belief that the glacier, and not the sea, was the last
occupant of the surface. It also seemed to me that the recurrence
of severe glacial conditions on the land after it emerged would ex-
plain many facts otherwise very difficult to account for. In sub-
sequently examining various parts of Scotland I have kept this point
specially in view ; and the result has been to force upon me the con-
viction that the development of snow and ice during this third stage
of the Glacial period must have been far greater than most geologists
suppose.

§ 2. General Features of the Surface.

The general features. of the country, even in the lower districts,
do not appear to me to correspond with the notion that since the sea
retired nothing but the ordinary action of the elements has modified
the surface. In such a case I should expect to find level sheets of
gravel, sand, and silt, containing some remains of marine fossils in
a more or less perfect state; also zones of beach-pebbles mixed with
some littoral shells, and deposits of a similar nature capping emi-
nences that had been in shoal water; and, in particular, I should
look for traces of estuary mud along the curves of the wider valleys,
where the tide and the river had formerly met. Now in Scotland,
so far as I am aware, we have absolutely no trace of any such
estuary beds containing remains of animals peculiar to places of the
sort, except at levels below thirty feet, and which belong, as I have
elsewhere shown, to a more recent period, when glacial conditions
had passed away, the shells indicating a climate rather. warmer
than at present. How could the glacial sea have gradually retired,
or, rather, how could the land have gradually emerged, without some
tidal sediment being left here and there along the valleys where a
pause in the change of level took place? It is true some have
thought they have discovered traces of ancient sea-margins in certain
more or less horizontal banks and terraces, which, however, admit
of a different explanation; but no one, so far as I remember, has
been able to point out any estuary beds, containing estuary fossils,
along the valleys at high levels. Why, also, should we not see some
more distinct lines of old sea-cliff and sea-caves at higher altitudes,
and likewise some heavy masses of blown sand and shells, like
what we find on the coast at present? The beds of glacial marine
clay and sand have been destroyed along the valleys to an extent
inexplicable on the supposition that the sea gradually retired
and nothing but ordinary subaerial action followed. In certain low
districts, where this clay has nearly all disappeared, patches of it
are left on eminences and places just where we might suppose it

THE GLACIAL PERIOD IN NORTH BRITAIN. 319

most likely to have escaped the action of glaciers; and at the mouth
of some valleys (as, for example, that of the Dee at Aberdeen) we
find masses of it which seem to be the denuded remains of beds that
some powerful agency has swept clean out of all the rest of the
valley. And some of these beds appear to have been dislodged from
their original position and thrust out seawards in a confused mass,
as in the banks near the Aberdeen lighthouse and powder-magazine
—a result such as we might suppose would follow from a glacier
moving down the valley over the beds of clay and sand, wasting
them gradually beneath it, and forcing them partly before it.

Now let us consider what would be the effect of a glacier extend-
ing itself over beds of sand and clay such as had been deposited
during the submergence of the country. The advance of the ice
over this old sea-bottom would either destroy the marine beds, by
pushing them before it and wasting them beneath it (the water
flowing from the end of the glacier contributing to the effect), or it
would move over them without entirely destroying them. Much
would depend upon the weight of ice, its duration, the slope of the
country, and also the depth and firmness of the marine strata.

Where the glacier was thick and the valley narrow and steep,
the marine beds would probably be entirely cleared out. The glacier
would push them bodily before it; and where it did slide over them,
the pressure of the ice would work up the clay and sand into a
liquid mud, which would be continually carried off by the water
escaping from beneath the ice; so that the wasting of the mass
would go on rapidly, and result in its entire destruction, leaving
only the stones and washed gravel in front of the glacier. But
where the glacier was not very thick, and the marine strata of con-
siderable depth and firmness, I imagine the ice might slide over such
a deposit without entirely destroying it. The effect would probably
be to compress and wrinkle it to some degree, and to work the upper
portion of it into an unstratified paste or mud mixed with stones.
This would result from the grinding pressure of the ice and the
want of good drainage for the water beneath it; for unless there
was a current of water to carry off the mud, the material would
remain beneath the glacier, although a part would no doubt be con-
tinually carried forward by adhering to its sole and travelling on
along with it. The pressure of the ice pushing into beds of strati-
fied clay and sand would, in some cases, displace them, and wrinkle
them into folds, thus giving rise to beds of contorted drift.

The unstratified, unfossiliferous mass of pebbly clay which occu-
pies so much of the surface, and is occasionally seen overlying marine
beds containing arctic shells, may have been formed in the way
above described. It occurs in many parts of Scotland, and has got
the name of upper drift or upper boulder-clay. Much of what has
been occasionally described as the upper covering of gravel and
boulders I believe to be also a result of the action of the later gla-
ciers. When these upper beds consist of marine deposits ground
into mud by the action of the ice, it is evident that bits of broken
shells may occasionally be found in them.

320 T, F. JAMIESON ON THE LAST STAGE OF

Mr. Trimmer, who devoted much attention to the superficial accu-
mulations of England, and sought to explain them chiefly by the
action of the sea and floating ice, was yet unable to overlook the
remarkable fact that there is a general absence of marine remains,
and of regular beds of these remains, in what he termed the Upper
Erratics, not only in Norfolk, which he had specially studied, but
also in every district of England, Wales, and Ireland he had ex-
amined *.

§ 3. Disappearance of the Marine Beds.

The general disappearance of the glacial marine beds over most
of Scotland is a fact difficult to explain, except upon the sup-
position that there was a subsequent occupation of the ground by
glaciers. The amount of submergence is on this account very diffi-
cult to determine; for the evidence of the sea’s presence has been
destroyed. Several geologists, influenced no doubt by the dis-
coveries in Wales, have supposed, and perhaps rightly, that the sub-
mergence reached to a great amount, and that the former coast-line
must at one time have been 2000 feet above the present, or even
more. But when we ask for proof we get little that is satisfactory,
the presence of far-travelled boulders at great heights being almost
the only fact of any value; and this may admit of explanation by
the agency of land-ice. Supposing, however, that the submergence
was much less, and reached no higher than say 500 feet (which no
one, I imagine, will consider an overestimate), we ought to find
marine beds far more widely spread than we do. For with the ex-
ception of some of the flatter ground along the east coast, away
from the mountains, and here and there over the Scottish coal-field,
beds containing marine arctic fossils are unknown. As a rule, they
are absent from all the Highland valleys, even at low levels, and,
generally speaking, are not to be met with near hills, and from the
country to the west of the Caledonian canal have not been reported
at all.

In the Clyde district many shell-beds are known, but generally
near the shore, and very little above the present reach of the tide.
If we suppose that the glaciers again occupied the surface after the
sea withdrew, it will afford a better explanation of the disappear-
ance of the marine beds than any other I can think of. For much
of Ireland and England the same hypothesis seems to be required.

In Caithness the area occupied by the dark grey drift containing
marine shells probably marks the extent of ground in that quarter
which escaped the action of the later glaciers. This area seems to
be bounded by the hilly ground which borders the plain of Caith-
ness to the west and south. In the latter direction it stops at the
low ridge that divides the water of Dunbeath from Berriedale Glen,
into which the grey shelly drift does not enter. Very likely the
grey drift may have been cleared out of the lower part of Berriedale
by a glacier; for some patches of it occur about the mouth of the

* Quart. Journ. Geol. Soc. vol. vii. p. 24, 1850.

THE GLACIAL PERIOD IN NORTH BRITAIN. 3821

glen, in the cliffs facing the sea. I regret having been prevented
from tracing the outline of this peculiar fossiliferous drift of Caith-
ness along all its western boundary ; for features of interest will no
doubt present themselves where the ice from the hilly ground came
down upon it.

From Berriedale to Inverness, all along the eastern border of
Sutherland and Ross, the later glaciers seem everywhere to have
come down in great force to the present coast-line. At Ardersier,
near Fort George, I came upon a small patch, a few yards in extent,
of grey clay, containing arctic sea-shells. It was buried underneath,
or enveloped in, a brownish unfossiliferous mass of gravel and silt,
and seemed to be a remnant of some bed that had been destroyed
by the action of the later glaciers. Its occurrence, however, in-
terested me much, by showing that glacial marine deposits had once
existed in that neighbourhood. Although no marine fossils were
got in the cuttings for the Caledonian Canal, it is worthy of note
that they have now been discovered near Fort William and Fort
George, at both extremities of the great glen.

Glacial clays containing marine fossils of arctic type are scarcely
known along the borders of the Moray Firth. Dr. Gordon, of Birnie,
tells me the only instances he is aware of are this one at Ardersier
and another at Burghead. In the interior of Nairn and Elgin they
have not been found.

I met with them, however, some distance to the east of Spey-
mouth, between Cullen and Banff, where the sea-cliffs here and
there show deep masses of dark bluish clay, in which remains of
arctic shells may occasionally be detected. Further eastward, at
Gamrie, they again present themselves, but only to a very limited
extent, having apparently been swept out of the little ravines by
small glaciers descending from the neighbouring heights. Without
this hypothesis it is difficult to account for the patchy manner in
which these marine beds have been left there.

In the island of Arran, as has been shown by the Rev. Mr.
Watson and Dr. Bryce, we have evidence of submergence to the
amount of some hundreds of feet above the present sea-level; but
the marine beds have all disappeared from the mountainous part of
the island, and even in the lower southern extremity they have been
wellnigh destroyed and overwhelmed beneath heavy masses of sand,
mud and boulders—the work, as I suppose, of the later glaciers.

Long ago Dr. Scouler pointed out that certain ravines near
Dublin had apparently been formed after the deposition of the shelly
gravel of that district, from the fact that they are completely desti-
tute of any vestige of this marine gravel, although it ascends to
higher levels in the neighbourhood. But he was at a loss to under-
stand on what principle of selection one set of hollows had become
receptacles of this shelly drift while others had escaped (Journ. of
Geol. Soc. of Dublin, vol. i. p. 266). Afterwards Mr. Oldham, in 1848,
in examining this locality and confirming Dr. Scouler’s observations,
was struck with the same remarkable fact, that several of the glens
in Wicklow contain no trace of marine beds, although shelly gravels

322 T. F. JAMIESON ON THE LASL STAGE OF

surround them on all sides and ascend to much higher elevations ;
and he further states that the great boulder deposit of Wicklow is
perfectly distinct from, and in all cases subsequent to, these marine
beds (bid. vol. i1. p. 302).

Mr. Darwin was the first to perceive the true meaning of such |
facts. After studying the similar phenomena of Wales, he pointed
out, in 1842, the evidence they afford of the advance of glaciers
subsequent to the deposition of the marine beds, showing that when
the land had risen to nearly its present height the ice filled some of
the valleys near Snowdon, and cleared out the accumulations left in
them by the sea. In tracing the phenomena downwards, Mr. Darwin
seems to have been unable to say where glacier-action ended and
marine action began. Professor Ramsay, who followed up Darwin’s
observations, and confirmed them by additional details, does not
bring his chief Snowdon glacier far down the valley, supposing the
lower ground near the sea to be covered with marine drift which
the later glaciers have not disturbed ; and he looks upon the high-
lying flats, at elevations of even 2000 and 2300 feet, as terraces of
marine formation, marking pauses in the reelevation of the country
(Old Glaciers of Wales, pp. 95-102, and map).

Judging from what I know of Scotland, I am inclined to think
the later Welsh glaciers must have been far more extensive than
these views would indicate, and that the Snowdon one at least must
have come down to the Menai Straits. Mr. Symonds has pointed out
many facts which imply the action of land-ice and snow, after the
land emerged, not only in various parts of Wales, but also in the
Malvern Hills.

As illustrating the nature of the climate I suppose to have pre-
vailed during the time of the later glaciers, let me give the follow-
ing example.

In the northern extremity of Aberdeenshire there is a hill called
Mormond, about 800 feet high, which lies a few miles south of the
town of Fraserburgh. It rises out of the low surrounding region
like a great mole-heap. Many years ago I had found a good deal of
evidence in the neighbouring district of the presence of the sea up
to heights of nearly 500 feet, and, among other facts, had noted beds
of well-rolled shingle forming the crest of certain low hills, at eleva-
tions of from 200 to 480 feet, which seemed to me to have been
shoals in the sea of that period. In some of these I got remains of
arctic shells. I argued, therefore, that if the sea had covered the
land to the height of 480 or 500 feet, as it evidently did, it should
have encireled this hill of Mormond, and formed a belt of shingle
round it at a corresponding height; or if the submergence was suf-
ficient to completely cover the hill, it ought to show a mass of gravel
on the top. I accordingly spent two long summer days in the month
of June examining the hill, which is a wide-spreading, heath-covered
mass, some eight or nine miles in circumference, bare and brown,
without a bush or tree upon it. J was disappointed and puzzled to
find no rolled shingle anywhere over the whole surface of the hill,
neither on the top nor on the sides of it. But along the base and

THE GLACIAL PERIOD IN NORTH BRITAIN. 323

skirts I came upon mounds of gravel which did not form a zone or
sheet at any regular level, but were disposed in a manner I could
not account for by marine action of any sort. Nowhere did they
reach to the height of 400 or even 300 feet ; and all the top and
sides of the hill seemed completely destitute of marine deposits of
any kind—no beds of gravel, shingle, sand, or silt—nothing but the
angular stony rubbish of the gneiss, quartz-rock, and granite of
which the hill is composed.

Mormond is somewhat of a horseshoe-form, by reason of two
spurs or ridges it throws out towards Fraserburgh, which enclose a
hollow in the northern bosom of the hill. This hollow contains a
bed of peat, below which we find coarse grey mud, lke what occurs
beneath a glacier, full of stones, several of which are ice-scratched.
In front of this hollow, at the distance of a mile or so towards
Fraserburgh, there is a transverse ridge of gravelly débris (some-
what like a kaim), called the Sinclair Hills, the base of which is
not more than 50 feet above the sea. The explanation of the matter
I take to be as follows:—This hill was no doubt encircled by the
sea, or may have been completely under it during the time of the
submergence ; but after the waters withdrew, it appears to have been
covered by snow and ice, which obliterated all trace of the sea’s
presence, and carried down the gravelly débris to its outer edge to
form the mounds we now see along the base of the hill. The hollow
in its northern face was probably occupied by a glacier which
stretched further out, the ridge of the Sinclair Hills marking its
former termination. Some of this gravel may have been originally
of marine formation, and afterwards remodelled by the glacier and
by the water flowing from it. In the north-eastern part of Aber-
deenshire I find traces of heavy snow beds, or small glaciers of the
second order, on the flanks of hills even lower than Mormond; but
in most districts these low hills have been overrun by ice from the
interior of the country.

§ 4. Moraines at Low Levels.

One of the proofs that the glacier and not the sea has been the
last occupant of the surface is the occurrence of well-preserved
moraines at low levels. Now the goodness of the evidence here de-
pends, first, upon the certainty that the masses in question are really
moraines, and, secondly, upon the assurance that they have never
been covered by the sea since they were formed. Differences of
opinion will, no doubt, exist for some time in regard to both of these
points.

When Agassiz visited Dr. Fleming at Aberdeen, in 1840, he was
shown some ridges of gravel near the coast, a little to the north of
that city, and pronounced them to be moraines. Fleming dissented
from this opinion, on the ground that they were of more recent
origin than the beds of fine clay beside them, which contain arctic
shells. I think Agassiz was right as to the mounds being moraines,
and that Fleming was also right, that they were a more recent

324 7, F. JAMIESON ON THE LAST STAGE OF

deposit than the clay containing the shells. They imply, in my
opinion, that the glaciers of the Dee and Don coalesced and reached
the coast after the period of submergence—the left flank of the ice
lying upon Belhelvie, and the right flank on the hills of Nigg. As-
suming that these mounds at Aberdeen are actually moraines of the
later glaciers, we ought to be able, on proceeding up the valley of
the Dee, to point out the various halting-places of the glacier as it
gradually retreated to the mountains. This, I think, can be done.
Ten miles up the valley of the Dee we find indications of a halt, a
crescent-like band of moraine-matter curving out from the base of
the Hill of Fare to the Loch of Drum. At Aboyne (20 miles further
up) there are indications of another halt. But the longest and most
decided pause has been a little below the village of Ballater, about
40 miles from Aberdeen and 600 feet above the sea. Here there is
a great assemblage of mounds on both sides of the river. On the
north side they commence at the base of the hill immediately below
the Pass of Ballater, and extend eastward past Tullich to Culbleen,
where they cover all the ground between the river and the base of
the hill, forming a great mass of hillocks and tumuli of various
forms. To the eastward of Tullich they reach an elevation of about
400 feet above the river, their upper surface forming a nearly hori-
zontal line or platform, which, however, is not quite horizontal, but
slopes slightly to the eastward or down the valley. These moraine-
heaps fill all the curve of the hills between Camus 0’ May and Tul-
lick. The flank of the hill of Culbleen is much bared, and dotted
with perched boulders. Here may be seen split blocks of red gra-
nite, which may have been broken by tumbling off the end or side
of the glacier.

On the south side of the valley, along the base of Pannanich Hill,
the higher part of the moraine does not take the shape of hillocks
so much as it does on the north side, but forms a great bank leaning
against the hill-side, like a rude platform or terrace, strewed with
granite boulders, and corresponding in height with the upper level
of the moraine on the opposite side of the valley. The granite of
the spur of the hill to the east of Pannanich Wells is bare, and
dotted with perched blocks, of which there are some fine examples.
The rock itself is rounded off into pillowy masses, which are more
rugged on their eastern outline; these appearances extend far above
the level of the highest moraines. ‘The granite, however, is too
much weathered to show the finer glacial markings; and the same
remark applies to most of the granite on Deeside. The great sheet
of gravel which covers the moor of Dinnet for some miles to the east
of Camus o’ May has probably been spread out by the waters pour-
ing from the end of the glacier, and issuing from beneath it, aided
by occasional floods when the snow and ice thawed extensively.

Other moraines of later origin than this one below Ballater are to
be found in the upper branches of the Dee—as, for example, in
Glen Lui, near Derry Lodge, at an altitude of about 1600 feet, and
still higher up, near the top of Glen Derry, not far from Loch
Etichan, at the base of Ben Muick Dhui, probably 2000 feet above

THE GLACIAL PERIOD IN NORTH BRITAIN. 325

the sea. These mark the last stages of the glacier. Corresponding
moraines occur in Glen Dee and in the ravines on the north side of
the Cairngorm mountains.

That the glacier of the Dee actually did come far down the valley
after the land emerged is also shown by a remarkably fine series of
terminal moraines in the valley of the Feugh, which is a tributary
of the Dee. These moraines occur at levels of 300 and 400 feet.
They appear to be perfectly undisturbed, and, so far as I can judge,
have never been touched by the sea since they were formed, Oc-
curring as they do at these low levels, they afford good evidence of
the advance made by the later glaciers. The Feugh takes its rise
at the foot of Mount Battock (2555 feet), and, after a course of about
fifteen miles, falls into the south side of the river Dee, at the village
of Banchory, which is eighteen miles from Aberdeen and 150 feet
above the sea. On walking up the valley of the Feugh for about a
couple of miles we meet with a moraine at a place called Gellan, on
the south side of the stream. This moraine must have been formed
by the glacier of the Feugh after it had been joined by those of its
two tributaries, the Avon and the Dye. The moraine here comes
down below the level of 300 feet; and the glacier, at the time of its
formation, was about thirteen miles long. Proceeding two miles
further up the valley, immediately after passing the mouth of
the Dye, we come to a fine moraine at a level of about 300 feet,
which must have been formed by the united glaciers of the Feugh
and the Avon, when the ice-stream was ten or twelve miles in
length. The great size of this moraine marks a very long pause of
the glacier here. The quantity of débris is enormous, especially on
the south side of the valley, where it forms a range of mounds com-
posed of gravel, sand, and rolled boulders. One of these hillocks,
called the Dunimore, is about 120 feet high. In some of these
ridges there is a considerable quantity of washed sand, like river-
sand; others are composed of blocks and boulders of granite mixed
with rough stony débris, while a few consist of a mass of boulders
of all sizes up to 3 or 4 feet in length, with a mixture of disin-
tegrated granite. All the material of the moraine on the south side
of the valley seems to be derived from granite; but on the opposite
side, at Whitestones, much of it consists of gneiss and crystalline
schist. This accords with the distribution of the rocks, which are
all granite on the south side, but mostly of gneiss on the north. It
would seem, therefore, that the débris of the two sides of the valley
has been kept from mingling; and this is just what a glacier would
do, and not what we should expect had these mounds been heaped
together by the sea. In the centre of the valley, which is here very
flat, the moraine has been swept away. The river Avon, which
joins the Feugh a little above this, is a small stream that also rises
at the foot of Mount Battock, and flows along a deep, narrow lonely
glen, or gorge, between Clochnaben (1944 feet) and Peter Hill
(2023 feet). There is a fine little moraine just outside the entrance
to the gorge, at an altitude of about 440 feet above the sea, forming
a characteristic crescent-shaped ridge, dotted over with large granite

Q2d. G. S= Now 1tg: 2A

326 T. F. JAMIESON ON THE LAST STAGE OF

boulders. There it lies, clear as a sunbeam, just as the glacier left
it. I walked up the glen a good long way above this, until I was
near the foot of Mount Battock, but observed no other moraines,
which would seem to show that the glacier must have shrunk rapidly
above this. Here, therefore, as in Lochaber, the glaciers seem to
have retreated by stages, pausing for a long time at certain places
and retiring rapidly at others. ‘There seems to be a corresponding
moraine on the Feugh, near Finzean; and I have no doubt others
will be found in Glen Dye, which, however, I had not time to
explore.

Terminal moraines (at least what I suppose to be so) also occur
on the south side of the Grampians, in localities which imply very
intense glacial conditions. Thus in Kincardineshire I observed one
at Drumlithie, near the railway station there, at an altitude of 280
or 300 feet above the sea, formed apparently by the glacier of the
Bervie, a small stream which takes its rise among hills of from 1500
to 1700 feet in height. When this moraine was formed, the gla-
cier was seven or eight miles long. This moraine, I believe, is
locally known as the Kaims of Candy, and seems to resemble the
kaims of the south of Scotland, many of which have been described
by Mr. Milne-Home, Professor Geikie, and others, and are supposed
by them to be accumulations formed underneath the sea by the
action of conflicting currents or tides. This kaim at Drumlithie,
however, appears to me to be a terminal moraine. It consists of a
long narrow mound, or series of mounds, from 20 to 30 feet high,
with sides sloping at angles of 20° to 30°, curving in a crescent- or
horseshoe-form, with the convexity seawards. So far as I could
make out, it is composed of coarse gravelly débris, irregularly piled
together. These mounds are most sharply defined at their north-
eastern extremity. Outside them (?.¢. seawards) there are traces
of lower and more gently sloping mounds, all under cultivation. A
narrow steep-sided mound like this, over which one can easily pitch
a stone, curving along for more than a mile in a crooked manner, is
not, I think, an accumulation which conflicting tides or currents
would make. The action of the sea would rather level such a
mound than make it. However, I shall touch upon the subject of
the Kaims and Kskers further on.

In the north and west Highlands, where the mountains come near
the coast, moraines of a more decided character are often to be seen
close beside the sea. Those at Brora, for example, on the east coast
of Sutherlandshire, are very striking, and have attracted the atten-
tion of many observers. I examined another very good example
near Muir of Ord, about ten miles west of Inverness, formed appa-
rently by the right flank of a glacier descending Glen Orrin. Along
both sides of the Dornoch Firth I noticed moraine-like masses,
which implied that the glacier had come well down the Firth there.
The great range of gravel ridges and boulders which runs from Cul-
loden Moor to Kildrummy*, near Nairn, consists, in my opinion,

_* Kildrummy is a Gaelic word, and probably means the head or end of the
ridge, which is very descriptive of the place.

THE GLACIAL PERIOD IN NORTH BRITAIN. 327

of moraines, and shows that the later glaciers which filled the valley
of the Caledonian Canal and the neighbouring glens to the west-
ward reached the head of the Moray Firth.

Many of the moraines that attracted the attention of Agassiz in
1840 were at low levels. The most distinct and well-marked of
all he saw in the British Islands were, he tells us, near Florence
Court (the seat of Lord Enniskillen), in Ireland. He also mentions
as good examples those on the banks of Loch Awe and Loch Etive,
especially near Bunaw Ferry ; all these must have been formed by
the later glaciers. Buckland also pointed out a number in Scotland
and England which indicate a very great advance of the ice. Dr.
Hooker, in a communication to the ‘Reader,’ in 1865, mentioned
that in the upper valley of the Tees there are huge moraines, as
well developed and as clearly marked as any in the Alps or Hima-
layas, and that the glacier determined precisely the present course of
the river and its windings. I am therefore inclined to believe that
not only in the north of Scotland, but also in Ireland and much of

England, the glacier, and not the sea, has been the last occupant of —

the surface, and that many of its peculiar features, such as the ©

kaims, eskers, gravel terraces, and unfossiliferous upper drift, will

have to be referred to the action of glaciers and freshwater currents. —

The valley-gravel is well explained by supposing it to have been
formed during the gradual and final retreat of the glaciers; and the
remarkable absence of the bones of elephants and other large quad-
rupeds which occur so frequently in the south of England and
France is no doubt due to the later occupation of the northern
regions by snow and ice.

A great many of the land-locked hollows along the bottom of hills,
now occupied by peat-mosses, probably indicate the position of the
last beds of snow and ice which lingered there before they finally
vanished.

In places where the glaciers did not reach there were probably
heavy beds of snow, which would be partly converted into ice at the
bottom. These would exert a modifying influence on the surface,
and also affect the drainage of the localities they occupied.

Mr. Kinahan reports a well-marked terminal moraine at a height
of only 140 feet above the present sea-level, near Bantry Bay, in the
extreme south-west of Ireland—a district which enjoys the mildest
winter temperature in the British Islands. This moraine, he says,
could never have been under water, or it would have been washed
out of shape (Proc. Geol. Soc. Dublin, March 14,1866). Facts like
these have not been sufficiently weighed ; for if glaciers came down
to within 140 feet of the sea at Bantry Bay, what must have been
their development in the northern parts of Britain ?

§ 5. Freshness of the Glacial Markings.

The freshness of the glacial markings on the bare top of many a
hill, in localities which betoken an immense development of ice,

affords another argument in favour of a late extension of the glaciers ;
2a2

328 ’. F. JAMIESON ON THE LAST STAGE OF

for we can scarcely suppose that these markings would have stood
the action of the weather so long as they must have done had they
been imprinted in the period before the submergence. The length
of time that has elapsed since then must be enormous; and yet on
some hills of 2000 feet, the summits still exhibit the ice-worn sur-
faces left by the glaciers which overflowed them from the central
heights. As no covering of earth or turf seems ever to have pro-
tected them from the weather, it is difficult to conceive how they
could have escaped destruction so long had they existed before the
time of the submergence.

§ 6. Kaims, Eskers, Se.

Many of the narrow, steep-sided, curving ridges of gravel, known
in Scotland by the name of Kaims*, and in Ireland as Kskers, I
believe to be deposits formed along the margin of the later glaciers.
- The fact of these gravel-ridges being composed of water-worn mate-
rials, often stratified, although generally in confused, irregular beds,
has induced most people to think that glaciers could have nothing to
do with them; and British geologists generally refer them to the
agency of marine currents. But in an ice-covered country where
precipices and high rocky escarpments seldom occur, and where the
hills are comparatively low with gentle slopes, the deposits formed
along the margin of glaciers will often be of a gravelly nature,
abounding in beds of sand and rolled pebbles having a sort of stra-
tified arrangement. Such is the case in many of the old moraines
of the Vosges. <A great deal of water escapes from the end of gla-
ciers and pours off their surface during certain seasons; and a
glacier many miles in length, when the snows are melting, will
abound in streams of running water, which will carry along much
gravel and fine sand and produce deposits of irregularly stratified
stuff along the sides and at the end of the glacier.

In order to have moraines of rough rocky débris abounding in
great angular blocks such as occur in alpine districts, there must
be the necessary conditions, viz. alpine heights and rocky cliffs bor-
dering the glacier. Where these occur in Scotland moraines are
found of the usual alpine character ; but in a region of low undula-
ting hills it would be vain to look for such. I have no doubt the
same explanation will apply to many of the Irish eskers. Assuming
masses of glacier ice to have covered the mountains of Ireland and
also to have occupied part of the central plain of that country, what
kind of deposits might be expected to form along their margin, as
they gradually melted away? Something very different no doubt
from the usual character of alpine moraines. At certain seasons
there would be great streams of water running along the margin of

* The word Kazm is in Scotland generally applied to a narrow steep-sided
mound. Some derive the term from the fancied resemblance to a cock’s comb ;
others suppose it to come from the Gaelic word cam, crooked, being often applied
to a crooked or curving hill (see Dr. Jamieson’s ‘Scottish Dictionary’). Mr.
Campbell, of Islay, derives it from Cewm,a path.

THE GLACIAL PERIOD IN NORTH BRITAIN. 329

the ice, and pouring off its surface, when the winter covering of snow
thawed rapidly. And when the glaciers of the Clyde valley were
retreating to the high grounds of Lanarkshire, what sort of deposits
might be expected to form in front of them at the time they were
pausing near Carnwath or Carstairs? Would they not be something
like the Kaims which Mr. Geikie has described as occurring there ?

It is not to be supposed that all the ice which gathered over the
central plain of Ireland and other low regions of a like nature pro-
ceeded from the hills, although glaciers descending from the moun-
tains would, of course, contribute their proportion. For if a great
thickness of snow fell, year after year, over the whole surface of the
country, and lay there without melting, the bottom layers would be
gradually converted into ice, the thickness of which would depend
upon the depth of the superincumbent snow. In course of time, as
the weight of the mass increased, motion would commence by the
ice yielding in the direction of least resistance. The pressure in
certain places being unequal, the ice would move (very slowly, no
doubt) towards the quarter where the pressure was weakest. This,
it is likely, would in most cases be the seaward opening of the chief

valleys; and motion having once begun there, would be gradually — %

propagated into the interior of the country.

I have remarked that these Kaims often lie across valleys in long
sinuous lines, forming curves or segments of a circle, the concavity
of which is presented to the head of the valley, and their convexity
towards the sea or downward end, as in terminal moraines. Such
would seem to be the case with some of those in the south of Scot-
land. For example, in the statistical account of the parish of
Greenlaw, in Berwickshire, the writer makes mention of certain
gravel ridges or Kaims, which he says are “disposed like a horse-
shoe, with the hollow towards the hills”’ Some of the most notable
Irish Eskers lie in lines across the vale of the Shannon, as, for
example, those between Athlone and Parsonstown. Another sweeps
round the northern extremity of the Slieve Bloom mountains; and
Sir Richard Griffiths tells us of a large one called the Horseshoe, from
its peculiar shape. ‘These linear ridges of gravel seem to be most
developed in low districts away from the mountains. In such loca-
lities the ice, being unconfined by hills, would spread out into a wide
cake or sheet. The time of their formation appears to have been at
the close of the glacial period, not during the submergence, but after
it, when the icy masses had begun their final retreat. The material
they consist of and its mode of arrangement point to streams of
water flowing over the surface of the glaciers, and washing the gra-
velly débris into heaps along their margins. A mass of gravel
reposing against the side or end of a glacier would lose its support
when the ice melted away, and, falling down in a slope, would assume
the form of a steep-sided mound. Glaciers are subject to oscillations,
sometimes advancing, sometimes receding, according to the varying
nature of the seasons. Suppose the end of a glacier to push forward
into a sheet of gravel lying in front, the result would be to force it
up into a mound all along the edge of the ice (sce figs. ] & 2).

330 {. F. JAMIESON ON THE LAST STAGE OF

Fig. 1.—Terminal slope of a glacier, with gravelly débris resting
against it.

M. Collomb mentions that the frontal slope of terminal moraines
3 is generally steeper than that of the other side. In the old moraines
; of the Vosges this is constantly the case, the frontal slope sometimes
amounting to an angle of 35°. Sir Richard Griffith tells us that in
Re the case of the Irish horseshoe Eskers one side generally slopes at
4 an angle of 30° and the other at 10° to 15°.
3
2

If the Kaims and Eskers were formed beneath the sea by con-

flicting tides and currents, how comes it they are so destitute of

marine fossils? No satisfactory explanation of this has been

- given. As most of the Irish ridges are composed of limestone gravel,

there would be the less likelihood of the shells being dissolved by the

percolation of water. The only shells I ever found in a Kaim were

in the parish of Slains in Aberdeenshire. These were old-Crag shells

( Voluta Lamberti and others), showing, as I conceive, that the forma-

tion of the Kaim was long posterior to the time when the mollusca

lived which inhabited the shells, and therefore affording no argument

that the ridge owed its shape to the action of the sea. The occur-

rence of these old shells, however, was so far interesting as it showed

that, if the gravel of other Kaims had been originally fossiliferous,

the shells in them ought not to have all disappeared. The marked

absence of marine fossils in the Kaims, Eskers, and upper drift or

upper boulder-clay affords a strong presumption that these deposits
were not accumulated beneath the sea.

I have also a difficulty in believing that narrow crooked (some-
times even zigzag) ridges of gravel, with their sides sloping at angles
of 20° to 30°, would be formed beneath the sea, or at all events
could have emerged from it without being thrown down. I should
think the materials could not lie at such an angle of repose beneath
agitated water, and would have been levelled to some extent as they
came under the influence of the waves and breakers. They could
not have stood the dash of the water in stormy weather, even in

THE GLACIAL PERIOD IN NORTH BRITAIN. 331

comparatively sheltered places. The material of which these ridges
are usually composed is of a loose incoherent nature; and the present
slope of their sides is often as steep as what the sand and pebbles can
lie at. Now the angle of repose of such stuff in strongly agitated
water is much less than it is in air. Long narrow mounds sloping
steeply on both sides could not have preserved this form when lashed
by the waves and breakers ; and I therefore maintain that when the
movement of emergence brought them near the surface of the water
they would have been levelled to the angle of repose which gravel
beaches usually exhibit.

A large iceberg running aground amongst gravel might cause
mounds like the Kaims, but this could take place only in water of
‘considerable depth. Heavy packs of ice driven forcibly ashore might
perhaps also give rise to similar ridges, although scarcely, I should
think, of such great dimensions. But unless the mounds were formed
high up on the beach beyond the extreme limit of the tide they would
be levelled again by the return of the waves. If formed in this way
they should also contain some littoral shells, and their curves ought,
as a rule, to present the convex side to the land, and not to the sea.
The sea cutting into the face of a bank may no doubt cause a steep
slope on one side, but not a narrow crooked ridge, steep on both sides
like a railway embankment.

The Swedish Osar* and the American Horsebacks seem to resemble
the Kaims and Hskers in their linear character, gravelly composition,
steep sides, and want of fossils. Sir Charles Lyell, in his account
of the Swedish Osar, says he met with only one instance of shells
occurring in them, viz. in the top of a ridge near the Castle of
Upsala; but as the species he mentions consist of edible mollusca
(mussels, cockles, and periwinkles), there may be a doubt as to
whether their presence in this very exceptional instance has not been
due to some accidental cause. Agassiz, in his description of the
glacial phenomena of Maine, says the horsebacks “ are unquestionably
of a moraine nature, and yet they are not moraines in the ordinary
sense of the term.” Some of these horsebacks run from north to
south; but occasionally they trend from east to west. ‘ This,” he
says, “is the case where a morainic accumulation of loose materials
may have been pushed forward along the margin, in front of an ex-
tensive sheet of ice moving southward, and then left unchanged by
the subsequent retreat northward of the whole mass. I conceive
that such horsebacks running east and west may be compared to
terminal moraines, which, as is is well known, owe their origin to
oscillations of the front end of a glacier pushing forward a mass of
loose materials, thus throwing it up into a transverse ridge, and then
melting away to some point further back.” (Agassiz in the Atlantic
Monthly, Feb. & March, 1867). He makes no mention of any
marine fossils occurring in these horseback ridges, and does not seem
to think that the sea was concerned in their formation. I may also
mention that some of these American ridges occur in the region of

* Under the name of Osar it,would seem that deposits of more than one kind
have been described, perhaps of different ages and various modes of formation.

332 {. F. JAMIESON ON THE LAST STAGE OF

the freshwater drift of the Western States, where we can scarcely
suppose the sea to have operated.

As the later glaciers in Britain and Ireland in many cases must
have moved over sand and gravel previously accumulated by the sea,
it is evident that their marginal deposits would be influenced by this
circumstance ; for the gravel and pebbles in front of them might
have been originally waterworn by the waves, and some marine
shells might thus be occasionally fourd in them. The occurrence of
such remains in a few Eskers I weuld therefore consider to be no
material objection to my mode of acccunting for them. In reference
to the waterworn and pebbly character which many of these glacial
accumulations present, it may be uselul to bear in mind that deposits
of similar materials are common in &witzerland, and that even the
famous blocks of Monthey lie amongst stratitied sand and gravel.

§ 7. Gravel Terraces.

The terraced banks of sand and gravel that occur along the sides of
our Scottish rivers seem to me to have been formed during the close of
the glacial period. Some geologists have attributed them to marine
action, and have considered them to be ancient sea-margins. I
myself, at one time, endeavoured to account for many of these gravel
accumulations by the off-rushing action of the sea during movements
of upheaval at the time the land was gradually emerging from the
sea of the glacial period, which we know covered much of Great
Britain. But I have become convinced that this explanation is not
the right one, and I now incline to think they are almost all fresh-
water and glacial deposits.

Not a single marine fossil has been found in them. Not a trace
of old estuary mud occurs along the higher reaches of our rivers in
connexion with them. Now it seems to me that if the sea had
occupied a valley after the glaciers finally left it we ought to find some
marine fossils and estuary beds here and there along the valley. The
absence of all organic remains, terrestrial as well as marine, in these
terraces and valley-gravels, agrees better with the supposition that
Scotland was at the time covered with snow and ice. Moreover
these river-terraces are not horizontal, as they ought to be if they
were sea-margins; for as a rule they slope seawards, like the bed
of the present rivers*. They occasionally contain large heavy
boulders, 3 to 4 feet in diameter or more, and are associated with
elongated mounds and confused hillocks of coarse gravel of a more
morainic character. I am of opinion, as I have expressed in former
papers, that these deposits have been for the most part accumulated
during the gradual retreat and melting of the later glacicrs. That
such terraces and mounds can be formed without the presence of the
sea 1s very well shown by the similar deposits that occur in the
valleys of the Pyrenees, the Himalaya, and other mountain-chains in

* For a good account of some Scottish terraces, see a paper on the Old River-

terraces of the Harn and Teith, by the Rev. Thos. Brown. Trans. of Royal
Society of Edinburgh, vol. xxvi., 1870.

THE GLACIAL PERIOD IN NORTH BRITAIN. 333

the interior of continents, where we have reason to believe no sea
has entered since the glacial period.

I have fancied that some of these terraces may have originated
in the following manner. Suppose the channel of one of our
Highland rivers to be occupied .by a retreating glacier when the
climate was gradually growing milder. During warm seasons, when
much snow was melting, gravel, sand, and other débris would be
washed down the hill-sides and lodge against the flank of the glacier.
The rivulets and streams would at such times come down from the
neighbouring heights, carrying with them much matter of this sort.
At present it all finds its way into the bed of the river ; but when the
channel was occupied by a glacier this débris would be arrested by
the mass of ice and lodge in the depression between the glacier and
the hill (fig. 3). This hollow would therefore be gradually filled up
and might be traversed at times by streams of water. When the
glacier melted, the mass of stuff resting against it would lose its sup-
port and fall down in a steep slope, thus giving rise to the terraced
banks we now find (fig. 4).

Fig. 3.—NSection of valley with glacier in retreat.

i

i
Al h

a, 6. Gravel terraces.

It is a mistake, therefore, to refer all these gravel terraces and
platforms of stratified material to the action of the sea, as geologists
have sometimes done, forgetting that the sea is by no means essen-
tial to their formation. The total absence of marine fossils ought
to caution us against such hasty inferences. Extensive beds of stra-
tified gravel, sand, and silt seem to have been formed in all countries
that were formerly occupied by glaciers.

I conceive that in many of the valleys the bed of the stream
might be wholly occupied by ice, the rivers being frozen nearly from
top to bottom—and that, on the approach of summer, a great quantity

bod T. F. JAMIESON ON THE LAST STAGE OF

of snow would be melted before the ice in the central trough of the
valley broke up, so that heavy floods of turbid water would be let loose
over the surface of the frozen rivers and thus rise to great heights
along their banks. The thick solid ice occupying the bed of the
stream would take longer to melt, but would gradually break up,
rising to the surface in great masses, and bringing with it stones and
pebbles from the bottom. These masses of ice would float down the
stream, dropping the boulders here and there as they dissolved.
When the Straits of Dover were dry land and the rivers wandered
away down far below the present coast-line, it is probable that during
the winter season the Seine, the Somme, the Thames, &c. would be
completely frozen along what are now the lower reaches of these
rivers, and at the break up of the snow in summer these frozen
masses would cause the river-floods to rise to heights which now
appear very marvellous.

Mr. Prestwich has discussed this subject in an excellent manner
in the ‘ Philosophical Transactions’ for 1864, p. 286; and I would
only suggest that he has perhaps not sufficiently allowed for the
unmelted ice in the bottom of the valley, forcing the floods to reach
a height they could not otherwise do.

The underground ice of Siberia may have been buried in this way
by the rivers flooding it and flinging down beds of mud on it before
it had time to melt and become detached from the bottom.

§ 8. Comparative Glaciation of East and West Coasts.

IT have often been struck with the remarkable intensity of the
glacial action displayed on the rocks of the west side of Scotland
compared with the east, and in a former paper* I threw out the
suggestion that the precipitation of snow along the west Highlands
had probably much exceeded what fell on the eastern slope of the
island, just as takes place now in regard to the rainfall. Thanks to
the Meteorological Society of Scotland, we have now accurate data
concerning the rain; and Mr. Buchan’s excellent papers on the
subject show us what an immense excess there is in the quantity
that falls on the west side of the country compared with the east.
There are several stations in the west Highlands where the annual
rainfall exceeds 100 inches, whereas along the east coast it ranges
generally from 25 to 30; and it is interesting to note that the gla-
ciation of the rocks corresponds in intensity with the present excess
of the rainfall, showing that the precipitation of snow had been
similarly distributed ; and is it not the case generally that the marks
of ancient glaciers are most decided where the rainfall is now
heaviest ?

The prevailing wind in Scotland is the south-west, which, sweeping
up the moisture from the Atlantic, is cooled as it rises over the
mountain-tops of the west coast and throws down its burden in
copious showers on those hills, so that before it reaches the eastern
side of the island it is a comparatively dry wind. This is well seen

* Quart. Journ, Geol. Soc. xix. p: 258, 1863.

THE GLACIAL PERIOD IN NORTH BRITAIN. 335

along the line of the Caledonian Canal. Fort William (at its south-
west extremity) is one of the wettest places in Scotland, while Cul-
loden (at the north-east end) is one of the driest—86 inches of rain
falling at the one, to 25 at the other; while in the Glen Quoich, near
the summit level of the canal, 102 inches are recorded in the course
of a year.

Can we doubt that in the Glacial period a similar inequality must
have existed in the fall of snow? and must not the glaciers fed by
three times the quantity of snow have been much heavier and more
erosive than those in the eastern valleys? Braemar, at the head of the
river Dee, in Aberdeenshire, although embosomed among the highest
mountains in Scotland, has a rainfall of only 33 inches in the year,
while some of the Argyleshire glens have more than a hundred.

The absence of erratic phenomena along the Ural chain and Siberia
generally may perhaps be traced to the well-known dryness of the
air and lack of snow in that region.

Many of the singular features in the glaciation of the British
Islands seem to be due to excessive falls of snow along the western
hills and valleys, in the localities probably where the heaviest rains
now occur. The phenomena of the Parallel Roads of Glen Roy, for
example, if we are to explain them by the agency of glaciers, require
us to suppose that the ice was much more developed on Ben Nevis
and the region to the west of the Caledonian Canal than in the hilly
district around the sources of the Spey. But the mountain-ridge
that lies between Loch Laggan and Loch Spey is quite as lofty as the
hills which encircle Loch Arkaig, if not more so—four of the
summits exceeding 3000 feet, and one of them being no less than
3700. Yet it is evident the glaciers proceeding from this ridge were
much smaller and waned sooner than those that blocked the
mouths of Glen Gluoy and Glen Spean; otherwise they would have
occupied the watersheds or cols and also the top of Glen Roy at
the time the terraces show that these places were clear of ice. Now
what we know regarding the rainfall of the district harmonizes re-
markably with these results. We have the records of three gauges
in the neighbourhood of Ben Nevis and Loch Arkaig—namely, one at
Fort William, another at Glen Quoich, and a third at Inverie near
Loch Nevis ; and the annual rainfall at each of these places is 86,
102, and 82 inches respectively, being an average of 90 inches in
the year, whereas at Laggan, near the head of the Spey, the annual
amount of rain is only 46 inches, or just about half of this.

Again, there seems reason to believe that the upper reach of the
Moray Firth was at one time occupied by a large glacier composed
of the united ice-streams issuing from all the glens of Ross-shire
and Inverness, which now pour their waters into the Moray and
Dornoch Firths. This great glacier seems not only to have occupied
the top of the Moray Firth, but to have spread eastward past Nairn
and Elgin to near Speymouth, as if the glaciers descending from the
district lying to the east of Inverness (now remarkable for its dry
climate and small amount of rain) were comparatively so small as to
be overborne and pressed aside by the heavier ice issuing from the

336 T. F. JAMIESON ON THE LAST STAGE OF

western glens. The absence of marine fossils in the drift of Elgin
and Nairn favours the supposition that this glacier may have existed
even after the time of the submergence. A belt of gravel and
boulders, forming kaim-like ridges, which may be traced from the
neighbourhood of Inverness eastward along the base of the gneiss
hills of Nairn and Elgin to near Speymouth, may perhaps mark the
southern edge of this great glacier. It is cut through and partly
washed away by the rivers that now cross its path ; but although
some have regarded it as an old sea-terrace, no marine fossils have
have ever been found in it to substantiate that opinion. Now the
climate of the country to the east of Inverness is at present one of
the driest in Scotland, the yearly fall of rain at Culloden and Nairn
being only 24 inches, while in the western glens of Ross and Inver-
ness the rainfall is very heavy. If, therefore, we suppose the fall of
snow to have been similarly proportioned, it would help to explain
how it came to pass that the western glaciers were so large as to
overpower and press aside those of the eastern district. The trans-
port of boulders and denudation of the surface over the low grounds
of Morayshire imply the action of some great force passing from W.
to E., such as would arise from the flank of a glacier moving down
the Moray Firth as I have supposed.

Consider, again, the wonderful facts connected with the glaciation
of Ireland made known lately by Messrs. Kinahan and Close, and by
Mr. Campbell, of Islay; do they not imply an excessive snowfall
along the western heights of that island, and,it may be, a compara-
tively high elevation of all the western side of the country? And
the dispersion of the boulders of Shap granite from Wastdale Crag
in the north of England would admit of a clearer explanation if we
supposed that the region of the lakes, now so noted for its immense
rainfall, was formerly distinguished by a like excess of snow.

I incline to think that the Glacial period, in this country at least,
was distinguished by enormous falls of snow as well as by a low
temperature. If we might suppose that during summer the elephant,
rhinoceros, and other wild animals browsed along the borders of the
ice-covered region, it seems likely that, they would be sometimes
caught in the early storms of approaching winter, when these hap-
pened sooner than usual, and thus get bewildered and smothered in
the heavy snowdrift. Their short-legged heavy carcasses would ill
fit them for wading through deep wreaths of snow; and this would
account for the frequent occurrence of their remains in comparison
with those of the lighter and fleeter animals. When these large
heavy beasts lost themselves amongst the snow and sunk in it, they
would be covered up until the summer thaw came, which would float
off their carcasses and carry them down the rivers, lodging them in
the silt and gravel along their course.

§ 9. Conclusion.

This last phase of the Glacial period was therefore no time of mere
local glaciers lingering among some of the higher mountains, but the

THE GLACIAL PERIOD IN NORTH BRITAIN. 337

return of a great ice-sheet which spread over nearly the whole of
Scotland and Ireland and also the greater part of England. But
the ice seems not to have been so thick and extensive as it was in
the early glaciation—nor so enduring; for it has failed to destroy all
the beds of clay and sand containing arctic shells which the sea left
behind it, whereas in Scotland we find that the ice of the former
period cleaned off every thing down to the hard rock. I would sup-
pose that all the mountain-ranges of Scotland and Wales were coated
with thick ice, which reached the coast in most places, likewise the
hilly ground of the north of England, including the Pennine ridge
along its whole length as far at least as Derby. In the lower dis-
tricts of England further south, there would probably be extensive
snow-beds, more or less converted into ice at the bottom. During
the summer thaws these would send out great streams of muddy
water, occasioning those superficial deposits of brick-earth, warp, and
loess which are so widely spread to the southward. Beds of gravel
would be lodged where the force of the current was stronger, and
when the thaw was unusually rapid. That there must have been
heavy beds of consolidated snow even in the extreme south of
England, I infer from the fact that I find traces of such in my own
neighbourhood, in the low eastern part of Aberdeenshire, on the
flanks of hills only a few hundred feet in height. In the low ground
in front of these places there is generally a land-locked hollow or
shallow basin occupied by peat.

This second ice-sheet gradually shrank into smaller and smaller
compass, separating by degrees into valley-glaciers, which paused for
a time here and there in their retreat, leaving Kaims and Eskers in
the low grounds and more distinct moraines in the mountain-glens.
At times the ice seems to have receded rapidly over a wide area
without leaving any marginal deposits. In these places the surface
is more or less covered with a sheet of coarse earth or mud mixed
with stones, which formed the bed or lair of the ice. During such
times there probably occurred a succession of warmer seasons, causing
a considerable thaw, when vast bodies of water would flow down the
valleys, spreading out much gravel, sand, and silt along their course.
But it is important to observe that no great submergence of the
country again occurred, nothing but that slight depression which is
marked by the estuary beds and raised beaches a little above the
present coast-line.

The last great modification of the surface has been subaerial, and
not submarine. Glaciers, frost, and rain have done the work, not
the waves of the sea.

Discussion.

Mr. Jerrreys considered that the author’s remarks relating to the
beds containing Arctic species of Mollusca were not quite correct.
Pecten islandicus has been tound in the drift of Scotland, but not in
the seas at present surrounding that country. At depths of 30 or
40 fathoms many Arctic shells in a semifossil state have been
dredged, although they do not now live in those waters. Mya trun-

338 1. F, JAMIESON ON THE GLACIAL PERIOD IN NORTH BRITAIN.

cata, a species which lives in very shallow water, has been found in
much deeper water in a semifossil state. At Fort William there is
a bed containing Arctic species of shells 7 or 8 feet above the level
of the sea. Arctic shells of deep-water species have occurred 200
feet above the sea. Different conditions have existed at different
parts of the same seas, altering the character of the Mollusca. The
raising of the sea-beds above the level proper to enable certain Mol-
lusea to flourish, would cause them to become extinct.

Dr. CarPENTER mentioned that cold water may be thrown up into
very small depths under certain circumstances. Near Halifax, N.S.,
the surface-water is tolerably warm, but at no great depth the tem-
perature falls to 35°F. In this case the rotation of the earth causes
the cold water from the north to surge up on its western coast. The
North Sea is a shallow sea, with a shoal in the middle, and having
off the coast of Norway a deep channel which conveys the cold
Arctic undercurrent; hence the east side is 10° F. colder than the
west side. Local peculiarities of disturbance of temperature may
thus occur within short distances.

Prof. Ramsay remarked that the author was not dealing with wide
ocean-deposits, but with ice coming down to the sea from the land.
He had described certain changes—a great Glacial period, a period
of submergence, and a second minor Glacial period.

Mr. Prestwich maintained that temperature was a most im-
portant question in connexion with the subject of Mr. Jamieson’s
paper. The shells of the Glacial deposits, however, were mostly
littoral, and therefore dependent upon the local temperature. The
subject was one of much interest; but the paper was somewhat
speculative, and he should like to see all the evidence upon which
the opinions expressed were founded.

A. W. WATERS ON FOSSILS FROM OBERBURG, STYRIA. 339

27. Remarks on Fosstrs from Osrrsore, Styria. By A. W.
Waters, Hsq., F.G.S. (Read January 21, 1874.)

The Eocene occurs in Styria only in a small district in the south,
in the neighbourhood of Oberburg. These beds are followed by
extensive deposits of later Tertiaries, which attain their fullest
development in the central part of the country. It was these
latter beds (the “ Neogen ” of the Austrians) which were studied by
Sedgwick and Murchison, and described in their classical paper on
the Eastern Alps, the occurrence of the Eocene being then unknown
to them.

But since the researches of these two geologists, who have done
so much for Alpine geology, the occurrence of a Nummulitic deposit
containing a rich store of well-preserved Anthozoa has been made
known to geologists and palzontologists through the publication of a
monograph by the late Professor von Reuss, “ Die fossilen Foramini-
feren, Anthozoen, und Bryozoen,” Denkschriften der Akademie der
Wissenschaften in Wien,1864.

From this monograph it is not a difficult task to determine the
corals which I collected this summer during a short stay in Ober-
burg.

Since the publication of the above, the geological relationships of
the district have been fully discussed by Stur in his “ Geologie der
Steiermark.”

There are two points, about ten miles distant, which have yielded
the richest harvest; these are Gradischneg and Neustift; and the
fossils found in each are designated in Stur’s list. Since I have found
several fossils in both places which had only been found in one of
the localities, I have thought it well to make out a corrected copy
of Stur’s catalogue.

Corrected List of the Fossils from the Nummulitic Formation of
Oberburg (O) and Neustift (N), Styria.

Natica crassatina, Lam. O,N. Melobesia, sp.
obesa, Brongn. O. Lithothamnium nummuliticum,
perusta, Brongn. O. Giimb. O,N.
Delphinula scobina, Brongn. O,N. Verneuilina oberburgensis, Preyer.
Fusus subearinatus, Lam. O. NERO:
Turritella oberburgensis, Fss. Clavulina triquetra, Rss. O, N.
asperula, Brongn. O. Spiroloculina striatella, Rss. N.
Cerithium trochleare, Zam. O. —— Morloti, Rss. N.
Diastoma costellata, Lam. O. — Freyeri, Rss. N.
Panopea, sp. N. Triloculina trigonula, Zam. N.
Venus Aglaure, Brongn. O,N. —— oblonga, Montagu. N.
Crassatella tumida, Lam. O, N. granulata, Rss. N.
Cardium, sp. O, N. Quinqueloculina hiantula, Rss. N.
Pecten Gravesi, D’Arch. O. Vertebralina suleata, Rss. N.
Perna Sandbergeri, Desh. O. Discorbina obtusa, D’ Orb. N.
Ostrea, sp. Truncatulina variabilis, D’Ord. N.

Turritella incisa, Brongn. N. Rotalia formosa, Rss. N, O.

340

Polystomella latidorsata, Rss. N.
Lagena, sp.

Operculina irregularis, Rss. N, O.
Nummulites mamillatus, D’ Arch.

N, O.

striatus, D’Orb. N, O.
Amphistegina, sp. N,O.
Tinoporus vesicularis. O, N.
Trochosmilia subeurvata, Rss. O,N.
Cyathomorpha conglobata, Rss. O.
—— explanata, Rss. O, N.
Calamophyllia fasciculata, Rss. O, N.
Rhizangia Hornesi, Ass. Repenscheg.
Dimorphophyllia oxylopha, ss. O.
lobata, Rss. O.
Mycetophyllia interrupta, Rss. O.
multistellata, Rss. O.
Leptoria eocenica, Rss. O.
Ceeloria ? cerebriformis, Rss. O.
Hydnophora longicollis, Rss. O, N.

N.

Emam) o

Bmivehors annulata, Rss. O, N.

Styloccenia lobato-rotundata, M.-E. §
H, O,N.

taurinensis, M.-H. § H. O,N.

Dictyarea elegans, Leym. O.

Favia dedalea, Rss. O, N.

A. W. WATERS ON FOSSILS FROM OBERBURG, STYRIA.

Heliastreea Boueana, Rss. O, N.
Astrea Morloti, Rss. O, N.
Thamnastreea leptopetala, ss. O, N.
Pseudastrea columnaris, Rss. O.
Podabacia prisca, Rss. O, N.
Dendrophyllia nodosa, Rss, O, N.
Astreopora compressa, fss. O, N.
Dendracis Haidingeri, ss. O, N.
Actinacis Rollei, Rss. N.

Porites nummuliticus, Rss. N, O-
Litharza lobata, Rss. N.
Alveopora rudis, fss. N.
Millepora depauperata, ss. N, O.

cylindrica, [ss.
Membranipora subqualis, Rss. O.
formosa, Rss. O, N.

— Minsteri, Rss. N.
angulosa, fss. O, N.
leptostoma, Rss. O.
Lepralia Reussi, D’Orb. N.

—— rudis, Rss. N.

—— megalota, Rss. N.

—— multiradiata, Rss. O.
Eschara papillosa, Rss. N.
membranacea, fss. N.
Defrancia cumulata, Mich. sp. N.
Heteropora stellulata, Rss. O.

Heliastrza eminens, Rss, O, N.
Fossils or localities in italics indicate that they are additions to
Stur’s list.

Reuss mentioned, besides the Nummuhtes variolarius, two Num-
mulites which he does not determine, but thinks resemble JV. ga-
ransensis and intermedius. Of these three, I find the first (vario-
larius) is not a Nummulite but an Amphistegina; the other two are
N. striatus and mamillatus. I have also found a Lagena, and have
no doubt a more extended examination might add still more to the
list of the Foraminifera. I have further found Tinoporus vesicularis,
P.&J., and Entalophora attenuata, Rss.; and I find there are several
of the Bryozoa from Gradischneg which have not been mentioned,
but of which I have not had time to make a lst. One of the
Pectens is Pecten Gravest. Besides Turritella oberburgensis 1 have
T. innsa.

Stur mentions the occurrence of Lithothamnium nummuliticum,
(Giimbel) ; but a form I cut through much resembling it is a Melo-
besia (Rosanoff).

The coralliferous beds occur only in the immediate neighbourhood
of Oberburg and Neustift; but the Prassberg beds, an arenaceous
limestone, have been found over a much more extended district,
and of considerable thickness. They overlie the Oberburg beds, and
have been joined with them into one series, which passes by the name
of ‘“‘ Oberburg- und Prassberg-Schichten.” This contains fragments
of Pectens, which Stur says are undeterminable. Some I collected
remind me of Pecten Fauwjasvi and quinquecostatus. Above this we
get the Wurzenegg beds, with plant-remains and fish-teeth; then

A. W. WATERS ON FOSSILS FROM OBERBURG, STYRIA. 341

follow strata of the same age as those of Eibiswald and Sotzka, which
occur so well developed in Central Styria.

Reuss found ten species of Anthozoa the same as those from Castel
Gomberto; and as the Mollusca bear out the same conclusion, the
beds must at present be regarded as parallel. It would seem that the
Oberburg beds are not the youngest Nummulitic beds, as they should
probably be correlated with the Lucasana beds of Hantken, as deve-
loped in Hungary, which are overlain by the Tchihatcheffi-Schichten
of western Hungary. These last come to light again near Vienna, at
Stockerau. The Lucasana beds contain a large number of corals
very similar to those of Oberburg; Reuss considered them the equi-
valent of the Castel-Gomberto beds; but the exact correspondence
has been disputed by Hantken.

The Eocene beds were evidently deposited in a large bay extending
from Neustift to St. Florian (near Schénstein) and as far as Prassberg.
Since their deposition these beds have been raised to a height of over
2000 feet.

Discussion.

Prof. Duncan commented on the sad loss which science had lately
sustained in the death of Prof. Reuss, from whose labours he had
received much assistance, and with whom he had long been in the
most kindly relations. He had been disposed to place these beds
among the Upper Eocene or Oligocene beds; but the discovery
of Nummulites in them might cause them to be placed on the lower
horizon.

Q.J.G.8. No. 119. 2m

342 ¥. G. H. PRICE ON THE GAULT OF FOLKESTONE.

28. On the Gavut of Fouxestonn. By F. G. H. Price, Esq., F.G.S.
(Read April 29, 1874.)

[Puate XXV.]

In this paper I have not made any attempt to correlate the Gault
of Folkestone with beds of similar age in other parts of England or
of the Continent, but simply to describe the Gault as seen at Folke-
stone, and to give a list of its fossils.

Not having had time to examine the Gault in other parts of Eng-
land, I am not in a position at present to state whether these divi-
sions will hold good over extended areas.

Upon some future occasion I hope to place before this Society a more
general paper upon the Gault, correlating the various beds.

The Geological Magazine for April, 1868 (vol. v. No. 4), contains
a most valuable paper entitled the *‘ Albian or Gault of Folkestone,”
by Mr. Charles E. De Rance, of the Geological Survey. In this paper
Mr. De Rance divides the Gault into eleven zones, all of which are
well marked and have been recognized more or less by all geologists
and fossil-collectors who have visited the district.

Provisional names have been assigned to these beds, indicative
either of their colour, position, or fossil contents.

Having investigated the whole of the Gault at Copt Point and in
Eastweir Bay, which lies about one mile to the eastward of Folkestone
harbour, where it may be seen resting upon the Folkestone-beds of
the Lower Greensand or Upper Neocomian, and found that, for the
most part, Mr. De Rance’s division into zones or beds holds good, I
consider that we cannot do better than adopt his order of dividing
the Gault into the various beds which I propose describing in this
paper, setting forth one or two points on which I disagree.

It is exceedingly difficult for any one who has not thoroughly in-
vestigated the Gault in sztu to be able to discern the various beds as
they lie on the beach in Kastweir Bay. In consequence of the slipping
of the Chalk above, the Gault has been pressed out and faulted in great
confusion, many of the beds being mixed up together.

The lithological characteristics once familiar to the investigator, he
will plainly see how utterly impossible it would be for any one to
attempt taking accurate measurements of the beds as they are there
placed. They are not merely mixed, but are much expanded in
comparison with what they are at Copt Point, where the only accurate
measurements can be obtained, the Gault at that place being zm situ,
and capped by the Upper Greensand.

The Gault of Hastweir Bay is always moving, and in very wet
weather may almost be seen to move, as the quantity of water per-
colating through the Chalk to the Gault, where it meets with an ob-
stacle, causes the Chalk above to press out these lower beds. I do
not think the day is very remote when we shall have to record a con-

F, G@. H. PRICE ON THE GAULT OF FOLKESTONE, 343

siderable slip of the upper beds over this slippery floor into Eastweir
Bay.

Copt Point is described by Dr. Fitton* as being 130 feet above the
level of the sea, and by Mr. Mr. De Rance as 128 ft. 5 in.

In the autumn of 1873 I endeavoured to take accurate measure-
ments of the Gault at Copt Point and at the Pelter+, a local name
given to a part of Eastweir Bay where the upper beds occur im situ.

1 took my line of measurements from the seam of nodules of sul-
phuret of iron, the base of the Gault, which is found about 4 feet
above a hard seam of sandstone, the zone of Ammonites mammillatus ,
the top of the Upper Neocomian, the intervening four feet consisting
of light sands.

From the sulphuret-of-iron band up to the base of the Upper Green-
sand I found to be 99 feet 4 inches.

The Gault is divided into two groups or divisions, 2. e. the Upper
and Lower Gault, separated by a passage-bed; and these divisions may
be again subdivided into ten well-defined zones, seven being below
and three above the nodule- or passage-bed, as shown in the follow-
ing diagram section (fig. 1, p. 344).

* Trans. Geol. Soc. vol. iv. part 2, 2iad series.
+ A brig of this name was wrecked here, which gives the local name “ Pelter”
to this part of the bay.

QE 2

344 ¥. G. H. PRICE ON THE GAULT OF FOLKESTONE.

Fig. 1.—Diagram Section of the Gault at Copt Point.

---Greensand seam.

Beds.
11. Am. rostratus...

| ~"Nodules (Pecten Raulinianus).

40. Kingena lima...... r=, (Plicatula pectinotdes).

9. Am. varicosus

. Am. cristatus ......

Am. auritus

Am. denarius
Am. lautus .........
Am. Delaruei
. Crustacea .........

go NS SI

2. Am. auritus, var....

1. Am. interruptus...

Iron nodules.

---Upper Neocomian; Folkestone

Am. mammillatus. beds

4 = 1 os.

(Seale 18 feet to 1 inch.)

F. G. H. PRICE ON THE GAULT OF FOLKESTONE. 345

Bed I,

The junction-bed separating the Folkestone-beds of the Lower
Greensand from the Gault consists of a line of nodules of sulphuret
of iron, which is about 1 foot 3 inches in thickness. This is
succeeded by a band of dark greensand, containing two lines of
phosphatic nodules, in which nodules Ammonites interruptus and
pieces of wood bored by Pholadidea sancte-crucis, Pictet et Camp.,
are found; many of the fragments of Ammonites, Inocerami, &c.
which are plentiful in this band are sprinkled over with small
spicules or crystals of selenite. In no other band throughout -the
Gault has this ever been observed. The dark greensand seam goes
off suddenly into Gault, with a line of phosphatic nodules traversing it.
These nodules of phosphate of lime arise, I presume, from the decom-
position of the remains of mollusca, fish, saurians, &c.

The fossils of this Junction-bed are usually in the condition of
rolled casts, and of a fragmentary nature. The following are some
of the most ordinary forms met with :—

Palzeocorystes Stokesii, Mant.
Tnoceramus concentricus, Park.
Nucula pectinata, Sow.

Pecten, sp.

Pholadidea sanctz-crucis, Pict. e¢ Camp., in wood.
Teredo-tubes.

Trigonia Fittoni, Desh.
Dentalium decussatum, Sow.
Pyrula, sp.

Ammonites interruptus, Brug.
Hamites attenuatus, Sow.

H. rotundus, Sow.

Belemnites minimus, List.
Nautilus, sp.

Vertebree of saurians and fish.

Mr. Gardner has obtained a species of Hippurites and Buccinum
gaultinum from this bed.

The above-mentioned fossils are mostly found more or less in-
crusted with phosphate.

Upon many pieces of Ammonites, nearly all of which are Ammonites
interruptus, I have found specimens of Plicatula pectinoides, Pecten
quinquecostatus, Trochosmilia sulcata, and a species of Nucula allied
to obtusa, all with the shell on, apparently indicating that they were
contemporary with the formation of the bed, which appears to have
been derived from some deposit now destroyed.

Whilst examining this bed last October at Copt Point, I met with
a piece of Ammonites Roissyanus, D’Orb., in situ, imbedded in phos-
phate; this is a very rare Ammonite to meet with at all, and
especially in the Lower Gault, as it belongs to the Cristati, or
crested group of Ammonites, which are characteristic of the Upper
Gault.

I also found a fragment of a very large smooth Ammonite, deeply
grooved, which I cannot at present assign to any species.

346 F. @. H. PRICE ON THE GAULT OF FOLKESTONE.

The total thickness of this bed from the top of the Upper Neoco-
mian beds, 7. ¢. the Folkestone-beds of the Lower Greensand, to a
hard seam of nodules and crushed band of fossils is 10 feet 1 inch.

This bed may be considered the zone of Ammonites interruptus,
Brug., as that well-known and varied form occurs in this bed and in
no other, excepting in the band immediately above, which is only 1
inch in thickness.

The following are the synonyma of Ammonites interruptus, Bru-
guicre, 1792 :—Ammonites serratus, Parkinson ; A. dentatus, Sow. ;
A. noricus, De Haan; A. dentatus, Fitton; A. Deluci, D’Orb.; A.
Chabreyanus, Renevier; A. Benettianus, Sow.

Bed Il.

may be considered the bottom bed of the Gault, commencing with
a band of crushed fossils one inch in thickness, which has a line of
phosphatic nodules running through it.

The clay of this bed is of a very dark colour, in faet, very nearly
approaching to black when moist, and is somewhat hard.

It is remarkable for the-deep rich colour of its fossils, and for its
containing several species peculiar to it.

At a distance of three inches upwards from the last-mentioned
band, a line of selenite in large pieces occurs; this mineral is not
met with in any other part of the Gault, and only in the position
here indicated in bed II.

One foot above the selenite a line of phosphatic nodules occurs ;
the space between these two is the horizon of Aporrhais calcarata,
D’Orb., which beautiful little Gasteropod is not met with elsewhere
in the Gault. Yurrilites elegans and Ancyloceras spinigerum like-
wise occur in this horizon and are not again seen.

The following species are peculiar to this bed :—

Fusus Itierianus, D'Oré. Myacites, sp.

Cerithium trimonile, Mich. Pectunculus, sp.

Narica cancellata, Chemn. Nucula De Rancei, Price.
Corbula gaultina, Pict. e¢ Camp. Pollicipes rigidus, Sow.

and a variety of Ammonites auritus having long tubercles, which
may perhaps be a distinct species.

The total thickness of this bed, measured to the base of the “ light
bed” No. III., is 4 feet 2 inches.

The rich colour on the shells of the fossils from this zone is, no
doubt, due to the very large percentage of iron in the form of oxides,
and to the comparatively small percentage of carbonate of lime met
with in the clay of this bed.

My friend Mr. W. H. Hudleston, to whom my best thanks are due,
has given me the following analysis of the clay of bed II.:—

“The Gault of this bed consists of a grey marly clay ; constituent
particles extremely fine and close; nothing granular observed except

a few specks of pyrites and a considerable amount of organic matter.
Dried at 100 C. it contains

F. G. H. PRICE ON THE GAULT OF FOLKESTONE. 347

per cent.
Water andvorsanic matter. os eee een se 8:25
Carbonate or lime -.. . is smc ee Re ee ace 8:61
Silica, silicates, oxides of iron*, pyrites &....... 81:93
Not Chloride of sodium (consid.)

actually |} Carbonate of magnesia (traces) 1-21

ESiiMiasee Suu MUEIC ACI ee ar een ins aims
ted Phosphoric acid (traces) einer
100-00

“This differs principally from the upper bed in containing less
carbonate of lime, and less phosphoric acid, but much more iron and
more organic matter.”

Bed II.

Known as the “light bed,” or “ crab-bed,” from the fact of many
of the Crustaceans found in the Gault being derived from it.

It may be readily recognized from any other by its light fawn-
coloured clay, which exhibits a striking contrast to the very dark co-
lour of the bed last described.

Its thickness is 4 feet 6 inches, being measured from the top of bed
II. to a line of nodules containing many rolled fossils with the shell
preserved on them, carapaces of Palewocorystes Stokesti and P. Bro-
deripii, Hamites and bones of saurians and fishes.

A remarkably fine type of Jnoceramus concentricus occurs in this
line of nodules.

Tabular masses of ironstone occur in various parts of this bed,
being of the same light fawn-colour as the clay. Similar ironstone
also occurs in bed V. It contains many of the fossils common to the
bed, Corbula elegans being met with in it in large numbers.

For the following analysis of this ironstone I am indebted to Mr.
W. H. Hudleston :—

‘«‘ A clay ironstone free from grit. The bulk of the mass consists of
pulverulent carbonate of lime, ferrous carbonate, extremely fine par-
ticles of clay &c. with about 2 per cent of pyrites. Dried at 100 C.: —

Amount of metallic iron (principally as protoxide).... 29-40
Add for iron in combination with sulphur (say) ...... 1:00
Wotalametallic irony aun ee eet ees 30°40”
Bed IV.

This is a narrow band which was apparently overlooked by Mr.
De Rance, or not considered by him of sufficient importance to be
constituted into a separate bed. It is measured from the line of
nodules containing the carapaces of Crustaceans before named, to a
line of phosphatic nodules and rolled casts of fossils. Imbedded in
many of the fragments of phosphate a very small species of Hoplo-
paria is met with, which species is not found in any other part of the
Gault.

* Metallic iron 4:25 per cent.

348 F, G. H. PRICE ON THE GAULT OF FOLKESTONE.

The total thickness of this bed is four inches; in colour it is sightly
darker than the preceding bed No. III. and not mottled like the clay
of bed No. V.

Comparatively few fossils are met with in this bed ; the following
may be regarded as peculiar to it :—

Ammonites Delaruei, D’Orb., a form belonging to the group Cristatz,
which may be considered characteristic of the Upper Gault as the
groups Dentati and Tuberculati are of the Lower. Ammonites cor-
nutus, Pictet, and Ammonites Rorssyanus, D’Orb., are the only other
forms belonging to the group Cristati that I have met with in the
Lower Gault; and as only these three forms have been found in the .
Lower Gault, the Cristati group of Ammonites may still be considered
characteristic of the Upper division.

An Ammonite possessed of very long spines or tubercles, which may
be classed as a variety of Ammonites tuberculatus, Sow., M.C.; asmall
species of Vatica, which will be subsequently described in this paper
as Natica obliqua; and Fusus gaultinus.

The conditions under which this bed was formed appear to have been
especially favourable to the development of Dentaliwm decussatum,
remarkably fine specimens of which occur between these lines of
nodules.

I submitted one of the nodules from the last-deseribed band (in
fact, one containing a specimen of the small Hoploparia) to my friend
Mr. Hudleston, who has reported to me the following results:—

“This is a mixture of carbonate of lime with phosphate of lime
and iron, some clay, organic matter, and pyrites.

“The amount of phosphoric anhydride in the sample dried at
100 C. is

25°53 per cent = 55-60 tricalcic phosphate.”

Bed V.

This bed may consistently be termed the “ coral bed,” on account
of the quantity of corals which it contains.

Its thickness is 1 foot 6 inches. The clay is of a dark colour,
spotted to within a few inches of the top with light fawn-coloured
markings.

It is measured from the line of nodules containing the fossil casts
up to a line where the clay becomes more strongly mottled.

This bed is considered by Mr. De Rance to be the zone of Nautilus
Clementinus ; but as that fossil is by no means peculiar to it, and
makes its first appearance in bed III., and has been met with in every
bed up to and in bed XI., I do no not think we can specialize it as
the zone of Nautilus Clementinus, but rather as the zone of Ammo-
nites lautus, as the typical form of this species makes its first appear-
ance in it.

The following fossils appear to be peculiar to this bed :—

Astarte Dupiniana, D’ Ord. Turritella, two sp. undetermined.

Cerithium Dupinianum, D’ Ord, Solarium albense, D’ Ord.
subspinosum ? Desh. moniliferum, Mich.

F, G. H. PRICE ON THE GAULT OF FOLKESTONE. B49

The following eight forms make their first appearance in this
bed :—

Cardita rotundata, Pictet. Acteeon affinis, D’ Ord.

tenuicosta, Sow., G. T. Bellerophina minuta, Sow., M. C.
Nucula Vibrayeana, )’ Ord. Solarium conoideum, Sow.
Avellana inflata, D’ Ord. Phasianella ervyna, D’ Orb.

Resin is also found in this bed. Mr. Hudleston tells me such
substances, like the Highgate resin, are viewed by Dana as a kind of
copalite. This resin is wholly volatile, burning with a highly lu-
minous flame and fragrant odour. It is not appreciably soluble in
alcohol.

Bed V1.

This bed, commonly styled the ‘ mottled bed,” is of small ver-
tical thickness, being only one foot. It is measured from the top of
bed V. to ahardseam. The clay is darker than that of the preceding
bed, and is mottled throughout with lightish-coloured markings.

It may be considered the zone of Ammonites denarius, Sow., M.C.,
as that fossil is frequently met with in it and is of rare occurrence in
any other bed. The following six forms are peculiar to this zone :—

Homolopsis Edwardsii, Bell. Turrilites Hugardianus, D’ Orb.
Necrocarcinus Bechei, Deslong. Ammonites cornutus, Pictet (a small
Fusus elegans, D’ Ord. variety).

Scalaria gastyna, D’ Ord.
Bed VII.

The top bed of the Lower Gault is measured from the hard seam
constituting the limit of bed VI. up to the first line of nodules forming
the base of the junction-bed. It is 6 feet 2 inches in thickness, and
is usually called the “ dark bed ” from its very dark colour, although
it is not so dark as bed II., neither do the fossils bear so rich a colour
as they do in the latter.

This bed is highly fossiliferous, containing as many as 67 species ;
it is especially rich in Gasteropoda, and possesses six species peculiar
to it, namely :—

Nucula albensis, D’ Orb. Avellana Dupiniana ? D’ Oré.

Aporrhais Parkinsoni, Sow. Cerithium tectum, D’ Ord.
Fusus indecisus, D’ Orb. Phillipsi? Leym.

There are likewise three well-marked forms which make their
first appearance and are abundant in this bed, and pass up into the
junction-bed above, where they only occur sparsely. These are :—

Nucula bivirgata, Witton. Buccinum gaultinum, D’ Ord.
ornatissima, D’ Ord.

Of the last-named fossil I have thought it well to give figures
(Pl. XXYV. figs. 1 & 2), as that given by D’Orbigny in his ‘ Paléon-
tologie Francaise,’ tom. ii. pl. 233, is of so fragmentary a character
as to be difficult to identify with the fine specimens met with in
this bed.

Ammonites auritus and a variety of the same occur in large quan-

350 F, G. H. PRICE ON THE GAULT OF FOLKESTONE.

tities, from which circumstance it may be well to describe this as the
zone of Ammonites auritus.

The genus Cerithium is well represented in the Lower Division,
in which (ranging from bed No. II. up to this bed) as many
as 16 different species have been met with. Of these 10 appear
not to have been described. I have only enumerated 10 species
in my table; of the remaining six I have only found single specimens,
and have therefore deferred noting them until more data shall have
been obtained. This group will form an excellent subject for a mono-
graph at a future time.

Bed VIII.

This constitutes the junction-bed or “ passage ” between the Upper
and Lower Gault, and is measured from the first line of nodules
before cited as forming the top of bed VII. to a second line of
nodules. It has a total thickness of 10 inches, and is well known
by the name of the “ nodule bed.”

This passage-bed, for such it really is, contains many species
belonging to both the Upper and Lower Gault, and marks the total
extinction of many Lower-Gault forms, no less than the introduc-
tion of others which took their place.

Few species of Gasteropoda, which are so plentiful in the top beds
of the lower division, pass into this junction-bed; they are here
generally found in the form of casts or much enveloped in phos-
phate. Several Ammonites belonging to the various groups are
found here. The characteristic species is Ammonites Beudanti, which
is not found in any other part of the Gault. It may be well to
mention that this well-marked form obtains largely in the zone of
Ammonites mammillatus in the Upper Neocomian or Folkestone-beds
of the Lower Greensand.

The nodules of this zone, like the nodules at the base of the
Gault in bed I., contain for the most part a remanié group of fossils.

The Lower Gault appears to have been especially favourable to
the life and development of the Gasteropoda ; for as many as 46
species have been obtained (as may be observed in the accompanying
table) from those beds ; and they reached their maximum of develop-
ment in bed VII. Nine of these species extend into the nodule-
bed No. VIIL., there being but two species peculiar to it.

Out of the 11 species of Gasteropoda found in the Upper Gault,
6 appear to have survived from the Lower, and 5 are peculiar to the
Upper, thus showing that the conditions of the Upper-Gault sea
were unfavourable to the continuance of the more delicate forms of
life, or, if they were continued, that the chemical constituents of the
sediments deposited during the formation of these marls were
unfavourable to the preservation of specimens.

Out of 73 species of Lamellibranchiata that I have met with in
the whole deposit, 39 become extinct in the lower division, or in
bed VIII.; 4 are peculiar to that bed, and 16 to the Upper Gault;
whereas only 14 species are continued from the lower beds.

Taking the whole of the fauna of the Gault, of which I have

¥, @. H. PRICE ON THE GAULT OF FOLKESTONE. 351

collected as many as 240 species, marking their particular horizon
on the table of genera and species at the end of this paper, it will be
seen that 124 forms become extinct in bed No. VIII., and 39 forms
are continued into the Upper Gault. Bed VIII., besides containing
the limit-fauna, possesses as many as 18 species which are peculiar
to it. 59 new forms occur in the Upper Gault, making their first
appearance either in No. VIII. or in a higher zone.

The following is a list of fossils which may be considered peculiar
to bed VIII. :—

Bathycyathus Sowerbyi, Edw. § H. Ammonites splendens, Sow. (avariety).
Vermicularia, sp. — Beudanti, Brongn.
Avicula, sp. —- Brongniartianus ? Pictet.
Cuculleea glabra, Park. —— Itierianus,D’ Ord.
Cyprina quadrata, D’ Ord. Selliguinus, Brongn.
Mytilus Galliennei, D’ Ord. Pachyrhizodus glyphodus, Blake &
Pholas, n. sp. Mackie.

Rhodani, Pict. e¢ Camp. Ptychodus, sp.
Scalaria gaultina, D’ Ord. Fish-jaw and fish-teeth, undescribed.

Murex calear, Sow.

In addition to the above a variety of Ammonites tuberculatus, Sow.
is met with.

This bed may be considered the zone of Ammonites cristatus.

Bed IX.

This bed, forming the base of the Upper Gault, is measured from
the line of nodules at the top of the last-described bed to a line of
nodules containing vast quantities of crushed forms of Jnoceramus
sulcatus; its total thickness is 9 feet 44 inches.

Inoceramus sulcatus is the most characteristic fossil of this bed,
which is everywhere marked with its silvery impressions. This
species has never been met with in the Lower Gault at all, with the
exception of a rolled cast recorded* as being found by Mr. Etheridge
some years ago in the junction-bed I. In the middle junction-
bed, No. VIII., it appears frequently, and occurs in good condi-
tion; but it reaches its maximum of development in the bed
now under consideration, beyond which it does not pass. At a
distance of one foot above the nodule-bed there occurs a seam of
very hard marly Gault, which is filled with specimens of this
fossil.

A hybrid ferm of this Znoceramus is met with in this zone, half
of the shell resembling J. sulcatus, and half I. concentricus. It was
named Jnoceramus subsulcatus by the Rev. Thomas Wiltshirey, and
is figured by Pictet et Roux, Gres Verts, pl. 42. fig. 1.

Ammonites varicosus, Sow., is another fossil met with in great
abundance in this zone, and is decidedly the characteristic Am-
monite of the bed; it makes its first appearance in the line of

* Memoirs of the Geological Survey, ‘The Geology of the country between
Folkestone and Rye.’

+ “On the chalk of Hunstanton,” by the Rey. Thomas Wiltshire, Quart.
Journ. Geol. Soe. vol. xxy. p. 190 (1869).

352 F. G. H. PRICE ON THE GAULT OF FOLKESTONE.

nodules below, and passes up into bed X, where, however, it is not
frequently found. I would suggest that bed IX. be considered the
zone of Ammonites varicosus.

The geologist will at all times when working in the Gault readily
_ be able to fix his position, and know that when he discovers these
two most characteristic fossils he is upon the base of the Upper
Gault.

A variety of Ammonites lautus occurs in this zone, ranging from
the nodule-bed, where it is found in the form of casts, up to a
hard seam near the top, but never having been noted as passing
above that horizon. This Ammonite belonging to the group Dentati,
forms another exception to the rule of the Upper Gault being re-
stricted exclusively to Ammonites of the group Cristati.

Just below the hard seam at the top of this zone, Ammonites ros-
tratus, Sow.= Ammonites inflatus of D’Orbigny, makes its first
appearance. It extends thence up to the top of the Gault.

Pholadomya fabrina, D’Orb., is only met with in situ in this bed;
it certainly is found in the nodule-bed No. VIII., but then only in
the form of rolled casts. No other fossils, so far as I have as yet
been able to ascertain, can be considered to be peculiar to this zone ;
but the two following well-known Upper-Gault forms make their
first appearance in it :—

Pleurotomaria Gibbsii, Sow. M.C. Scaphites eequalis, Sow.

Bed X.

Commences at the crushed band of Jnoceramus sulcatus, and ex-
tends upwards to a line of phosphatic nodules containing specimens of
Pecten Raulinianus, D’Orb., 5 feet 1 inch above the top of bed IX.

The Gault of this zone is homogeneous, hard, and of a pale cold
grey colour, poor in organic remains, and partakes more of the
character of marl than clay, as does the whole of the Gault of the
upper division, which contains as much as 26 per cent. of carbonate
of lime—a fact sufficient of itself to account for the shells of the
fossils being so seldom preserved.

At a distance of 2 feet 8 inches above bed IX. a line of phos-
phatic nodules is met with, in which Plicatula pectinoides occurs in
good condition and in large quantities.

At the base of this bed Rostellaria maxima*, the largest known
Rostellaria from the Gault, was procured.

Between the last-mentioned seam of phosphatic nodules and the
line of nodules containing Pecten Raulinanus, forming the base of
bed XI., the clay, which has a thickness of 2 feet 5 inches, is much
mottled. Pentacrinus Fittont occurs in masses, whereas in all
other parts of the Gault it is only found in fragments. Cidaris
gaultina is only met with between these two lines of nodules.

Many pieces of bones of Chelonians and fishes are obtained from
the same horizon.

Upon two separate occasions Griffiths has discovered in this zone,

* See Geol. Mag. March 1873, vol. x. pl. vi.

F. G@. H. PRICE ON THE GAULT OF FOLKESTONE. 353

between the two lines of nodules, crushed bodies which have been
supposed by some to be the eggs of a species of turtle; they are
more or less round, possessing the texture of egg-shells. Mr. J. 8.
Gardner has two specimens in his collection; and I haye one,
which I showed to Mr. J. W. Hulke, who reported that he did not
think it was a Chelonian egg at all, but that of a crocodile, and
that in the coarse pitting of the shell it most resembles that of
Crocodilus biporcatus.
The following fossils may be considered peculiar to this bed :—

Goniaster, sp. Astarte, sp. (4. omalioides, J.S. Gardn.
Kingena lima, D’ Ord. MS.).
Hinnites, sp.

Many teeth and dorsal spines of Hybodus, and teeth of Otodus
appendiculatus, are found in it.

From the frequency of occurrence of Kingena lima, D’Orb., in
this zone, I adopt the conclusion arrived at by Mr. De Rance, and
propose to name the bed the zone of Kingena lima.

Bed XI.

This bed consists of a marl of a cold pale grey colour, traversed
by a series of partings, the smooth portions between which are
coated over with a smear of oxide of iron. Some difficulty was
experienced in finding the whole of this bed 7m situ, though equal
in thickness to more than half the Gault. We therefore measured
from the seam of phosphatic nodules containing Pecten Raulinianus,
the base of bed XI., up to the base of the dark greensand seam
which traverses this bed, and found that it had a thickness of 35
feet 6 inches. This measurement was taken in that part of East-
weir Bay, eastward of Copt Point, known as the Pelter.

As the upper portion of the greensand seam was wanting at this
point, it having slipped off together with the upper part of the bed,
we went slightly to the east of Copt Point, and southwards of Mar-
tello Tower No. 3, where the top of the Gault is an situ, capped by
the Upper Greensand. We found the base of the greensand seam
by excavating a trench down the cliff-side; we then took measure-
ments of this seam (which, I may mention, Griffiths calls the “‘ Middle
Greensand,” by no means an inappropriate term), and found that it
had a thickness of 3 feet 3 inches. From the top of this seam to
the base of the Upper Greensand was found to measure 17 feet
6 inches, thus giving a thickness of 56 feet 3 inches for bed XI.,
and a total thickness for the whole deposit of 99 feet 4 inches.

In this greensand large quantities of nodules occur in lines, which
may be pyriform sponges; also a small species of Jnoceramus (per-
haps a young form of concentricus), Aporrhais Orbignyana, Trocho-
cyathus conulus, Pecten orbicularis, Belemnites ultimus, Ammmonites
Goodhalli, Avicula gryphcoides, Pollicipes ?, and Phicatula sigil-
lina, fixed to Ammonites, an Ammonite allied to A. lautus, Dre-
phanephorus canaliculatus, Ichthyosaurus campylodon, &c.

The Gault immediately above this greensand seam is of the same

bo4 F. @. H. PRICE ON THE GAULT OF FOLKESTONE.

marly character as that below, only much mixed with grains of sand
for the first few feet.

The bed No. XI. contains but few fossils which are entitled to
be regarded as peculiar to it; they are as follows :—Ostrea frons,
Ptychoceras, Hylospongia of Sollas, only met with just below
the greensand seam; Lnoceramus Crispii, a large smooth species,
and Ammonites varians ? occurring above the greensand seam.

Ammonites rostratus is found of large size in this zone, often
having Plicatula sigillina adhering to it. I have frequently met
with this Ammonite measuring as much as 12 inches in diameter.
Ammonites Goodhalli likewise attains a very large diameter.

Belemnites are rarely found at all above the greensand seam.

For the following analysis of the Gault of bed XI. I am indebted
to Mr. W. H. Hudleston, who has taken considerable time and
trouble in working it out for me.

«Tt consists of a pale grey marl, the particles in a minute state of
subdivision. The bulk of the mass isa mixture of chalk and clay ; a
very few, and extremely small, quartz grains, and, after treatment
with acid, one or two translucent greenish grains were observed ;
also small crystals and fragments of pyrites.

“« Dried at 100° C. this sample contains :—

per cent.
Water andiorcanicumatierweeia «i. Seine Ieee: 6-75
@arbonate\of Inme iach. sso ce Sloe one 26°45

Clay-sand and silica, oxides of iron, pyrites, &c. .. 65°95

Chloride of sodium

Sulphate of lime | found, but not estimated . ‘81
Carbonate of magnesia
Bhosphoriesacid: eicict: Ws! eee eens oe 04

Metallic iron 2°55 per cent.”

There is an enormous field open in the Gault to the palzontolo-
gist ; and it is to be hoped that the day is not far distant when the
large quantity of material obtained and still to be obtained from
the Gault in various parts of England, more especially at Folkestone,
willbe exhausted, and some valuable monographs will be prepared for
the Paleontographical Society.

The data the paleontologist has at present to fall back upon are
principally works, very able in their way, which have been pro-
duced by foreign geologists upon foreign species. Those few scattered
specimens from the Gault that have been figured by Kuglish authors
have been mostly from very imperfect specimens.

Besides the fossils here enumerated, there are many forms which
it is difficult at present to assign to any genus.

I may add that there are many fish-remains (which occur mostly
in the upper division) wholly undescribed. ‘The species of the class
Cephalopoda require entirely reworking out, more particularly the
Hamites, which have been described by Messrs. Sowerby, Pictet,

FE. G. H. PRICE ON THE GAULT OF FOLKESTONE. 355

and others from fragments, and the specimens procured from Folke-
stone are frequently quite perfect; thus any one undertaking the
task of producing a monograph upon this genus would have fewer
difficulties to contend with than those authors who have already
described them from imperfect data.

On some New Species of Gault fossils.

The first species I have to describe is well known to collectors of
Gault fossils, and has been, if I mistake not, erroneously considered
Ringinella lacryma, D’Orb. Pal. Fr. pl. 167. figs. 21-23, which species
is never found in this country. Upon close examination and com-
parison with the figures of &. lacryma, D’Orb., and of Avellana
lacryma, in Pictet et Campiche, Pal. Suisse, Environs de Sainte-
Croix, p. 198, 2nd part, it does not agree, inasmuch as the speci-
mens I have had to refer to in the collection of the Museum of
Practical Geology and my own cabinet are of uniform size and
smaller than those figured by the before-named paleontologists,
besides other differences, to which I shall presently allude.

The following is the description :—

Shell ovate, thick. Spire formed of a regular angle composed of
convex whorls usually five in number ; body-whorl occupying three
fourths of the whole shell; sutures somewhat canaliculated. Two
folds on the columella; aperture narrow. Outer lip thickened and
reflected, emarginate and denticulated on the inside. Shell orna-
mented by fine horizontal lines composed of small ovate punctures,
with spaces between of about three fourths of a millimetre in
breadth ; these spaces are quite smooth, having usually from 15 to

20 lines on the body-whorl. The spaces between the punctate lines
become broader in the middle of the body-whorl; in some instances
they acquire their greatest breadth at the posterior portion of the
body-whorl.

Length of shell 10 millims. ; breadth of shell 6 millims.

This fossil is frequently met with in beds VII., VI., IV., and
III. in the Lower Gault of Folkestone; I propose to name it Avel-
lana pulchella. The genus Avellana, D’Orbigny, synonymous with
Ringinella, D’Orb., is divided into two sections by MM. Pictet et
Campiche, in ‘Pal. Suisse, Environs de Sainte-Croix,’ the first
section being composed of elongated forms with prominent spires,
and the second section of globose forms with short spires.

Of these the following are recorded as having been found in the
Gault of England :—

First section.
1. Avetrana inFuata, D’Orb. pl. 168. figs. 1-4.

= Auricula inflata, Fitton, 1836, Trans. Geol. Soc. vol. iv. pl. 11.
fig. 11.
= Cinulia inflata, 8. P. Woodward, Manual of Moll. p. 314.

2. AveLtana CremEntINA, D’Orb. Pal. Frang. pl. 168.

306 F. @. H. PRICE ON THE GAULT OF FOLKESTONE.

Second section.
3. AvELLANA IncRASSATA, D’Orb. Pal. France. Crét. p. 168.

= Auricula incrassata, Mantell, 1822, Sussex, pl. 19. fig. 33.

= Auricula incrassata, Sow. Min. Conch. pl. 163. figs. 1-3.

= Ringicula incrassata, Geinitz, Grundriss der Versteinerungs-
kunde, p. 337, pl. 16. figs. 3 & 4.

4. AveLiana Duprntana?, D’Orb. Pal. Frang. Crét. pl. 169.
5. AVELLANA PULCHELLA, F. G. H. Price. (Pl. XXV. figs. 4 & 5.)

Besides the above, with the exception of Avellana pulchella, the
following are found in the Gault of France and Switzerland :-—

6. AVELLANA LAcRYMA, D’Orb. Pal. Frang. Crét. pl. 167.

= Tornatella lacryma, Mich. 1834, Mag. de Zool. cl. v. pl. 33.

= T. affinis, Leymerie, 1842, Mém. de la Soc. Géol. t. v. p. 31.

= Auricula acuminata, Deshayes,1842 ; Leymerie, Mém. de la Soe.
Géol. p. 12, pl. 16. fig. 1.

7. AVELLANA VALDENSIS, Pictet et Campiche, Pal. Suisse, Environs
de Sainte-Croix.

8. AVELLANA ALPINA, Pictet et Roux, Moll. des Grés verts, p. 172.

9. AVELLANA suBINcRASSATA, D’Orb. Prodr., & Pal. Frane. Crét.
pl. 168 (not Sowerby).

10. Avettrana Huearprana, D’Orb. Pal. Franc. Cret. pl. 168.
11. Avettana Banpontana, Cotteau, 1854, Moll. Foss. de ’Yonne. ;

12, Avettana ovuzta, D’Orb. Pal. Frang. Cret. pl. 169.

Avellana pulchella approaches very near to Avellana alpina,
Pict. et Roux, but does not possess such a long spire, and dif-
fers from it in ornamentation or sculpture; inasmuch as that of
A. alpina is composed of slightly elevated ridges with little square
punctures, whereas in A, pulchella the ridges are not raised but
simply marked with small ovate punctures.

It will be seen from the above note that this shell differs essenti-
ally from any of the figured species of Avellana, although possessing
certain affinities, inasmuch as the size, ornamentation, the thick
reflected outer lip, the denticulation on the inside of the outer lip,
&c. are not found embodied in any other species.

Narica optiqua, F.G. H. Price*. (Pl. XXV. fig. 3.)

Shell globose, glossy and almost smooth ; the lines of growth very
fine, with indications of longitudinal striz. Spire very short and

* The following is a list of the Natice known from the Gault, with their
synonyms :-—
Natica Cuementina, D'Orb. Pal. Frang. Crét. t. ii. p. 154, pl. 172. fig. 4.

= Littorina pungens, Leymerie.

F, G. H. PRICE ON THE GAULT OF FOLKESTONE. 315//

depressed. Aperture oblique, ovate, the length being nearly double
the width. Inner and outer lips nearly parallel. Outer lip thin ;
the inner lip, which is somewhat reflected over the umbilicus,
is callous and broader at its upper part, where it forms an angle at
the point where it joins the body-whorl.

Umbilicus shallow and open.

millims.
atenethet shell, .ij2. chew te Be eee 12
preaeheot shells... 0. eet. ae eee ee 10
ensth ob aperture’. xi. 22 seen ae 9
iBxeadth of apertures.) 28/2 aces a

This shell has only been met with in bed IV. of the Lower
Gault, Folkestone.

A large Nucula occurs in the Gault of bed I1., which is not met
with elsewhere ; this I have always considered to be a new and un-
described species, since I carefully examined the figure and descrip-
tion given by D’Orbigny of Nucula ovata in the Pal. Frane. Crét.
tom. li. p. 173, purporting to be Mantell’s species. I then found
that he has mistaken the large species, which is peculiar to bed II.
at Folkestone, for what has always been taken as the typical form
of Nucula ovata, Mantell, which has a range throughout the Lower
Gault.

These two separate species are constant in form, and differ as much
the one from the other as it is possible for two Nucule to do. If
this large form, which is figured by D’Orbigny, be taken as the Nucula
ovata of Mantell, then that which has hitherto been considered
Nucula ovata by all English paleontologists, and which is found
throughout the Gault of the south-east of England, and does not
appear to occur at all in France (that is to say, it has not been
described by D’Orbigny), is a separate and distinct species. But
this latter fossil is the typical English Gault form of Nucula
ovata, and was named, described, and badly figured by Dr. Mantell

Narica GAuutina, D’Orb. Pal. Franc. Crét. t. i. p. 156, pl. 173. figs. 3 & 4.
= Ampullaria canaliculata, Mantell, 1822, Geol. of Sussex, pl. 19. fig. 13,

87.
q = Natica canaliculata, Fitton, 1836, Tr. Geol. Soc. ser. 2, vol. iv. pl. 11. fig. 12.

The above two species are common in the Gault of England. Natica Clemen-
tina is met with in beds VI. and VII. of the Lower Gault; whereas NV. gauwl-
tina has a wider range, being first found in the zone of Ammonites interruptus
at the base of the Gault, and passing all through the various beds up to the top.

There are four other species obtained from the Gault of France, which have
not been recorded as occurring in the English Gault. They are :—

Natica Durrxi, Leymerie, 1842, Mém. de la Soc. Géol. t. v. p. 18, pl. 16,
fig. 7.

NATICA EXCAVATA, Michelin, 1386, Mém. de la Soc. Géol. t. iii. pl. 12. fig. 4.

Narica prvyna, D’Orbigny, Pal. Franc. Crét. t. ii. p. 159, pl. 173. fig. 7

Natica Raunintana, D’Orbigny, Pal. Frang. Crét. t. 11. p. 160, Bl 174. fig. 1
@-7.G.s. No. 119: 2¢

358 F. G. H. PRICE ON THE GAULT OF FOLKESTONE.

in his Geology of Sussex, pl. xix. f. 26, 27, in the year 1822, and also
badly figured by Phillips in his ‘ Yorkshire’ in 1829; whereas the
Nucula ovata erroneously described as that of Mantell in the Pal.
Franc. Crét. t. ii. p. 173, was published in the year 1843 by
D’Orbigny, twenty-one years later *.

Therefore, following the rules of priority, the Nucula ovata of
Mantell, figured and described in his ‘ Geology of Sussex,’ must stand
good as a species, and the Nucula ovata of D’Orbigny must receive
a new specific name, as, in my conviction, he made an error in con-
sidering it to be identical with that of Mantell.

Below I give a description of this larger form of Wucula, which I
propose to name Nucula DeRancev.

Noucuta DrRancet, F.G. H. Price. (Pl. XXV. fig. 7.)

Shell ovate, inequilateral, inflated, thick, nacreous, ornamented
with delicate concentric strie ; ventral portion furnished with more
or less concentric lamin. Anterior portion of dorsal margin short
and angular; posterior long, sloping from umbones to posterior
margin. Lunule indistinct; lip of ventral margin very thick ; mus-
cular impressions well-marked and callous.

millims. —
ihenethvofeshelly ten Geo a eee ol
Meishtiofishellevenpre ve wrth sn teoO

This well-marked shell, differing from all other Nuculw in form
and size, is only met with in the lower part of bed II., not even
extending into the horizon of Aporrhais calcarata.

Table of Species of Gault Fossils from Folkestone.

1

(2)
Lower Gault. El.
Genera and Species. ie

I. | II. II1.)IV.) V.| VI] VII.|[VIIr,

PLANTA.

Pinites hexagonus, Carruthers | ... |...|...|..-
AMORPHOZOA.

Hylospongia, Sollas ............ G00. |} 600 ‘000 |} cao 900 | Go0
Pyriform sponges ............... sitio eet |itewtedl ees 200

ZOoPuyta.
Bathycyathus Sowerbyi, Edw.

Yas) neat asncsosonbaboncasodcadal ue" | oad || esa | a0e *
Cyclocyathus Fittoni, M.-Hdw.| ... | * |... |... | e+ | eee * || *
Smilotrochuselongatus,Duncan| ... |...) ...| * | *

insignis, Duncan ......... 6080] doo ft co) ee Il ee
Trochosmiliasuleata, Hdw.§H.| ... |... |... | so *
Trochocyathus conulus, Edw.

P daly concesoasandeanaqcoseaccoss| 3 |] 600'll 000 |! 000
—— Harveyanus (var. of),

AU SIBERDS 5460006060506600000 000 sfatstl| ade lMiaseyllleseses||RersalOseal lane

* There is a very fair figure of Mucula ovata, Mant., in the ‘Tabular view of
characteristic British Fossils.’

F, G. H. PRICE ON THE GAULT OF FOLKESTONE,

Genera and Species.

HCHINODERMATA.
Cidaris gaultina, Forbes ......
Hemiaster asterias, Forbes
minimus, Ag.......... sono
Pseudodiadema, sp. .....-.0.+.-
— Wiltshirii?, Wright Bases
Goniaster, sp............. spnooodas
Pentacrinus Fittoni, Azst.

ARTICULATA.
Serpula articulata, Sow. ......
IEKUS SOC nreatnescee
Wermicularia, sp.°.......<r2s-02-

CrIrRIPEDIA.
Pollicipes levis, Sow. .........
TARECNOS, SoC pacnoanonenatce
—— unguis, Sow. .......6.....--

——,, Sp....... scence cenelasteat a

CrustTAcEA.
AM GHONG Py Sse seen caerct soser aioe
Diaulax Carteriana, Bell ......
Etyus Martini, Mantell.........
Homolopsis Edwardsii, Bell...
aa longimana, Sow....

; ao (minima, MS.) Fatceter
Necrocarcinus Bechii, Deslong.
PaleeocorystesBroderipii, Mant.
—— Stokesii, Mant.......... aes

BRACHIOPODA.
Kingena lima, D’ Ord. .........
Terebratula biplicata, Sow. .

5G] bb ooopacteqeascebecadod nese
LAMELLIBRANCHIATA Mono-
MYARIA.
ANAOWAEY, 0; “Gonaedsodeasdouc secre
9 GJ DS ndancAdesunesnoacosaaOsRDoe
AVIOWIA, SDiis.ceracacsncsncaeses noe
— cenomanensis, D’ Orb.
=e PD statee oiaie se e(ensionivt wisi'otl Hee
— erypheeoides, SOR cece
Exogyra conica, Sow, .........
Gervillia solenoides, Defrance
JSGTAONKEEY, E\9h bascossocoqsece 500086
Tnoceramusconcentricus, Park.
Crispii, Mantell .........
mullethnues JAH Sononooseone
— ima Wiltshire ...

Lima pipes Stari WIG Onccace-
parallela, Sow. M.C. ...
Pecten Beaveri, Sow. M. C....
orbicularis, Sow. M. C....
quinquecostatus, Sow.

eos

Lower Gault.

II. |11I.) IV.) V. | VI.) VII.

Ix.

359

Upper
Gault.

xX.

xI,

*

* *

¥ *

*

*

* *

* |

ooo |

*

*

coe |

*

*

*

*

*

* | %

*

owe |

*

coe |
*
*

*

* | x
¥*
*

*

oon |

*

360 F. @. H. PRICE ON THE GAULT OF FOLKESTONE.

ERS
—
Genera and Species. g
I,

Pecten Raulinianus, D'Oro. ...

» JO cosoadascossacsnacsdono0000 ite
» G3] Db conodoscobasossncsgobqca90C *

rhotomagensis?, D’ Ord. .| ...
Perna Rauliniana?, D’ Orb. ...| *
Pinna tetragona, Sow. M. C....

LAMELLIBRANCHIATA DI-
MYARIA.
Arca nana, D’ Orb, ...........000-
Hugardiana, D’Ord.......
Astarte Dupiniana, D’Ord. ...
, 8p. (omalioides, Gardner,
IVES S) ue eat cect sclsctace eerie
Cardita rotundata, Pict. ......
tenuicosta, Sow. G. T....
Corbula elegans, Sow. M. C....
gaultina, Pict. et Camp. .
Cucullea carinata, Sow. M. C.
glabra, Park,...........0..+
Cyprina quadrata, D’Ord.......
Leda solea (allied to), D’ Ore. .
Lucina tenera, Pict. et Camp. .
sculpta, Phil............--++
Lyonsia ?.......c0cecsereeeeees $0000
Myacites plicatus, Sow. .........
Mytilus Galliennei, D’ Ord. ...
Nucula albensis, D’ Orb.
arduennensis, D’Orb. ...
bivirgata, Fitton, G. T....
— De Rancei, Price .........
impressa, Sow. M. C. ...
———5 Witraks; ID OR0s scboooso50de
ornatissima, D’Or®, ......
— ovata, Mantell ............ *
— obtusa, Sow. G. T. ...... ie
—— pectinata, Sow. M.C. ...| x
—— subrecurva, Phil, .........
—— Vibrayeana, D’Orb. ......
Pectunculus, sp........... atawces
(5) Se GeaseanboasqooebauesanasonG
Pholas rhodani, Pict. et Camp.
sanctsz crucis, Pict. et
(GLENS ooconcondoonnos6aqc0eHN[Gb
FSS WON oocsgaonoconenboosa000
Pholadomya fabrina, D’Or....
Thracia simplex ?, D’ Or.......
Tellina phaseolina, Pict. et
(OLY T 0] )5 600%300000 500500000008 o00000

2

sp. B00
Meredol(tube Of) ete ennsceeeenens *

, allied to ............ * Tra

Ii.

Lower Gault.

Iii.

IV.

|
V. | VI.| VIL.
300 |} $3
x | x
*
*
x
* *
22 \l ooo], 33
*
* *
x
%
d00 *
*
*
+3.) % | 4
* | x
* | x | *
de || doo |] te
Se |leean |G
ma lhe
x *
*
*
*
2 *

KOK OK OK

Upper
Gault.
X. | XI.
%
* *
coe | ¥
Ze ee
x
Le | ag
% | *
*
*
* | *
* | *

F. G. H. PRICE ON THE GAULT OF FOLKESTONE.

Genera and Species.

Thetis Sowerbyi, Romer ......

Trigonia Fittoni, Desh. .........
@ard) etn. ogaoeoos

EENCEPHALA.
G-ASTEROPODA.
Siphonostomata.
Aporrhais calearata, Sow......
Orbignyana, Pict. et

HEV OURO MS Sec dan aie ve ie hsienesecren ee
—— Parkinsoni, Sow. M. C....
Rostellaria carinata, Mantell .
carinella, D’ Orb. ......00.
cingulata, S. P. Wood-
(OURS Tecsecaooe seabaitisehisenasaes
elongata ?, Sow...

, sp. (Griffithi, Gardner,
8)

Pee eee terse ence e eas enenntes

ImMERAITE, JEPUED yaannodoosoe
Pteroceras bicarinatum, D’ Orb,
Fusus elegans, D’ Orb. .........
Itierianus, D’ Orb..........
indecisus, D’ Ord. .........
gaultinus, D'Oro. .........
BELLOWS Daten cemer soseneinc eas
Murex bilineatus ?, Pictet......
@Nlere, SONS coocncadanooose
Pyrula Smithii, Sow. G. T. ...
Buccinum gaultinum, D’ Ord.

Holostomata.
Narica cancellata, Chemn.......
Natica Clementina, D'Ord. ...
gaultina, D’Ord. ..... obod
—— obliqua, Price .........
? an operculum............
Cerithium Dupinianum,D’ Ord.
ervynum, D’ Ord., no loc.
—— Phillipsii?, Leym..........
subspinosum ?, Desh......
tectum, .D’ Orb. ............
—— trimonile, Mich. .........

Ses Clementina, D Orb...

Dupiniana, D’ Orb. ......
gastyna, D’ Orb...... secbacd
gaultina, D’Orb. ..... sche
Turritella granulata, Sow.......
Solarium albense, D’ Orb.......
conoideum, Sow. .........
dentatum, D’Orb..........
—— moniliferum, Mich. ......
ornatum, S020) 2 ..cs- ena

eee

Lower Gault a8 Upper
5 ese Gault.
I. |t1T. IV.) V. | VI.) VIL. || VEIL) 1X.| X. | XI
* * | x
*
* | * | * | * |e] oe | HM] HL
fats eee | ce0 *
% | cee oo | % * * *
eee * #
aac * * *
* *
eee * *
coo | ee #
* *
Sele a *
*
Bele * *
ses *
00 %
500 * ¥
500 *
* *
*
stsfe) * * * *
ur |! doo) o00 |) ur * * * | #
*
*
*
sats *
*
ers oon || #7
*
* *
*
eae *
* eee * *
woe | *
ao6 a6 *
us |e |l ode *
* *
eee * * + *
ace ||) tar *
saath Dist creel ios aaeee nee * | |e | x
* uel hare | |

362 F, G. H. PRICE ON THE GAULT OF FOLKESTONE.

Genera and Species.

Phasianella ervyna, D’ Orb.
Turbo Brunneri?, Pict. e¢ Camp.
Pleurotomaria Gibbsii, Sow...
Acmeza tenuicosta, D'Orb......
Dentalium decussatum, Sow.
= ellipticum?, Sow. .........
Avellana Dupiniana?, D’Orb..
inflata, D’ Orb. ........005-
—— pulchella, Price oad
Actzon affinis, D’ Orb............
Bellerophina minuta, Sow.
Bat CB sie bec Sar aan ean nen OSaeREs

CEPHALOPODA.

Ammonites auritus, Sow. M. C.

4) WED cnangsgoosca0cs
, a var. spined 000000
Beudantii, Brong..........
Brongniartianus ?, Pictet.
— Bouchardianus, D’ Orb...
—— cristatus, Deluc .........
cornutus, Pictet......... “ee
— Delaruei, D’ Ord. .........
denarius, Sow. M.C. .
—— Deshayesii, Leym..........
— Goodhallii, Sow. .........
interruptus, Brag......
, var. Benettianus,
SOU sess cae acsteh camewiaels Se
isles, IDXOR a coecodsoe
— lautus, Sow. HU GUC R aspcee
SVaPMaaneesaees Sacemeue
Mayorianus, D’Ord.......
—- Roissyanus, D' Orb.
rostratus, Sow., =infiltine,

— Selliguinus?, Brongn. 560
splendens, Sow. M. C..

, var., grooved ......
—— tuberculatus, Sow.........

(FOES. soodeosesooson000a 0 doses
varicosus, Sow. ...... sibieteteie
varians ?, SOW. secceseseess

bf Ouidbondenbese0 SppuodeCdnsob.
Aneyloceras spinigerum, Sow..

Baculites, sp. ..........scsseresees
Belemnites attenuatus, Sow..

minimus, Lister .........
, with phragmocone
—— ultimus, DAO O Mee ecnee
, with phragmocone
Crioceras Astierianum, D’ Ord.

Kos

, var.?, very long

eue

3.0

Lower Gault.
III.| LV.| V.| VI.| Vil.
* *
p00 |oce |p ae * *
5 O00 *
* * *
* * oF * %*
* *
* * *
oe Boon Wits *
* x |x | x
900 55 * .
eeon |G
* * * *
* ok *
x
*
* *
%
eee %
*
ood Wr *
* * #
eee | *%
* * ¥
* * *
%

xg
Ee

KOK KK

*

OK KF

mK

Upper
Gault.
IX,| X. | XT

*
* | *
*% |. | ¥
*
€
*
*
% 3
* %
my Wp our | ur
* | *
ece kK
*
*
+

¥F. G. H. PRICE ON THE GAULT OF FOLKESTONE.

Genera and Species.

Hamites armatus, Sow..........
attenuatus, Sow......0+..-
—— elegans, D’Or........ tare
—— maximus, Sow. ............
rotundus, Sow. ............
NautilusBouchardianus,D’ Ord.
Clementinus ...,...........

bon choteuthis quinquecari-

natus ?, Pict. et Camp. .....
Ptychoceras, |b acoponacabtacodées
Scaphites eequalis, Sow. .........
Turrilites elegans, D’Ord.......
Hugardianus, D'Ord......
MSD aes cdesecaecdeesaunumecsee

» [Os cccgdoc05G0 sosocadbonncAed

Piscus.

Drepanephorus enotonatus
LEGA soe sania: sSoltel sen eae ease

Edaphodon Sedgwicki , Ag...
Fish: ear-bones, scales, verte-
bree cern cnaer eater eae
> jaw (undescribed) ......
——: . (undescribed) ......
—— : tooth (undesceribed)......
: 4, (undescribed)......
Hybodus, teeth and dorsal
SIMON seycves sie de cits se ete ciee sae
Ischyodus brevirostris, 4g. ...
Otodus appendiculatus, Ag. ...
——, large..,.....0....0.205
pucherniodud glyphodus,
FEU GIED (69 NAGA soc ocaconencee
Pisodus ?, palate...............06
IE RG OLCIVED EY, " Agacneocanascemone
Se lanciformis,
Harlan...oe Sageeanosodoncaccone

Reprinia &e.
Chelonian bones................+.
Crocodilian ? egg ..2...........»
Ichthyosaurian vertebre ......
Plesiosaurian vertebre .........
Polyptychodon _ interruptus,

CBG coocnonngnenssanonaesacecce

Rhinochelys (lower jaw), Seeley a

Beare seit, ides cauncueseees

dq )
5-3 Lower Gault. Eb
5 5

I. | IL. \IIr.

soe *

*
mK
Ke

i

*

The following are additional species in the collection of Mr. J. S.
Gardner, F.G.S., for which I have no localities :—

“Opis, sp. Mmarginula, sp.

Nezera sabaudiana ?

¥. G. H. PRICE ON THE GAULT OF FOLKESTONE.

304

“UUNAVULLNIIG SD.LBD0.LADT ‘Ssnsr90p
-Usnsnyy Wuosury ung simy ody
‘punassyznUdo "AT “opphannag DynonvT

“sngpysewo “PW “wunpnag
sapuouup ‘vyouponb nurad
-fig ‘vaqnjb vaynongy “awuarn
-p.n sryyhpy ‘svanwo DpaUDvs SDjOYT

“SNIDLISOL
‘y ‘snsoowma sazuouump ‘sn2
-pojnsqns “J ‘sngoajns snup.sao0uy

‘SNIDAISOL SAQUOULULPT
‘saprounzoad ningnoyg ‘Wwozney
snumonuag ‘vypoydeg 0jnj0L9
-alay, ‘vuy nuabury ‘snwiu
-1u, wagsouuary ‘nuygnnb srunpyy

“suDIeDd “VW ‘Sn2DLISVL “VW
“1UIIDY POOL SAIIUOULULP *SNUIU2)
-mg uagoag ‘udsrg snumsa00uy

s

——=|

‘S[ISSOT OTJSTLOJOVIVY

"snp
-1unn sanuomup |G 9

9909
-s2ua sanuomup |e

i=)

*SNS09
“10d sapuoUwupy \EE

lor)

‘puny puabury | 1

Veo)

"Sn QD49
“Sou sapwuowulp |G 9c
“UL 40g
“sped
“*souoz, jo ssoul
“pray

‘soTNpouU JO ouly year

‘so[npou Jo our] puoveg

“paq-4seo
SNgDIINS - SNUDLIDOUT

"snUDwU
-nvy vaqoag surare,
-u0d seynpou jo eur

“puvsucery roddq

ey} 0} dn poansvoyy

AGP eR
jo do} ye mes prey

‘TTA
peg Jo doy omy ye
se_Npou JO oUly sal

‘TITA Peq Ur seTnp
-OU JO SUIT puodeg

"Sn}
“DONS + snmunsa00UT
jo pueq poysnao oy,

“snUprUnNnDay
uajoag sutureyuod
sopupou jo oul] VW

rt

WOIF POANsvo][

“suojsayjog 1 ynny 042 fo spog fo 21907,

*SNO.AOJI[ISSOF
Ayysry ‘Kepo yaeq | “TIA
“STISSOF 92UnU

-a4 pues sojnpou
Jo soul, OM4 SUT
-ureyuoo ‘ea0qe
ueq} Joytep AeTD

*keyo

Ayaeur oars opeg | “XI

‘amoyoo fous oped
B jo [Av paeA |X

“qt ysnosgy
suruuni yory3

‘UL @ 4o0J G pus
-ueeIs jo uUlvos

e@ UjIA ‘yreur
feis = pjoo ‘eT@q | IX
wa | PA
jo oangje Ny oN

‘TITA

365

€
e

F, G. H. PRICE ON THE GAULT OF FOLKESTONE.

ee nnn

‘snpUng
-OL SOZVUDET “MUNDAS H sD.Lad
-010 ‘sngdnisagur sapruowuypy

"IBA ‘SNIIUND SAQIUOUL YY “DID.LDI
-)n0 sipyssody ‘snummany sns
-N ‘apuoulie, wnryzIway “D2DU
-yoad nynonay ‘punu Dou “DULZ
-ynnb vyngsoj “Dsauan DULaNT
‘snpnoungoag ‘snprbi. sadeoyjog

‘snapnuagyy sanuunpy ‘vuobn12a)
DUULT ‘Najos 0} pate vpaT ‘ud
~1apouy ‘ql ‘Usayopy sagshLovoDInT

“pyjayojnd
punjaapy “anunjag sazwwout
-up ‘snurynvb snsng ‘ungos
-snaap wnyojuag ‘nnb2)90 v21QDAT

“sngnD) Sagi
-owup ‘vukasa npyjaunsnyg
“unsafywuouw wniwnjoy ‘su
-bisur “si ‘sngnbuoja snyoouzojug

‘sniwmuap "WF {(7ABA TTRUIS) s2g2nw
-100 sazruomump ‘snunrpimbneT

SaQUywiny, ‘Vwyoag snurawvo0.L0a\T

*smgDpjvUe

~UDU saqmUomupy ****

‘snadnu

~daQuesanuoMmulpy | T.

"IBA ‘$702

20ND sanuowup |

‘SUBIDEISHAY | O

“20NLD)

- sapuowump | F
"snp

-D) sajruomumpy | 9
*SNLUDU

-ap sapvuommy | 0

“see*""* SHAQ OUOISOH[O MT

66 eeecevecscce ssoUyOITy T?40.,

‘oR
OL | S[ssof poysnao Jo purg

F TIT poq 3451] Jo oseg
« : ‘on

Sqvio Jo syoeq SurUILy
F -u0d ‘sejnpou jo oury

"8}S¥0 poy[Or pure sopnp
Q | -ou oyeydsoyd jo oury

iEEIe
-JOUL OLOU seTIODeq
T | Avpo oj} easoya oury

*poq jo doy ‘aves paeyy

re

*spoq WRIULO0O
-oont Jo dog ‘woat Jo
jounydtns Jo soynpoNy

“snadnutap
-Uud sapuoumup fo
ouozZ OY} GAO" STIS

-80F poysnao Jo pueg
"IT poq yoryq Jo dog,

‘029 SQB1I JO SYIVq
Ygta ‘soynpou jo oury

‘oyeydsoyd ut
8880 PoT[OL SUIUTE,
-100 seTNpou Jo oury

"A poq-jetoo jo doy,

sreeeesusesmunsuaad LaMoT

‘soTnpou WodT
pue ‘puvsuseis

yaep ‘Ae yaeq

“sMO[OO YSIY
SuIIeoq ss ISsOT
‘Avo yaep ALI0A

“£eyo po
-ANO[OO-UMVF YYSTT

“pey}jour you
‘eaoqe oy} Uvty
coyysy, = AysITS

*‘punoisd ysTyvep
110) SoUTy ABUL

gst] YA poyjodg
“keyo yaep

uo Suryrem yst

“FUST] TTA POTION

‘Til

366 ¥. G. H, PRICE ON THE GAULT OF FOLKESTONE,

EXPLANATION OF PLATE XXxyV.

. Buccinum gaultinum, D’Orb. (large specimen, decorticated).
(small specimen, with shell preserved).

. Natica obliqua, Price, enlarged.

. Avellana pulchella, Price, front view, enlarged.

, back view, enlarged,

. Nucula ovata, Mantell.

De Raneei, Price.

Fig.

TOD OUR G9 BO

Discussion *.

The Rey. T. WitrsutrE remarked that the various beds were fre-
quently divided by lines of nodules, especially towards the bottom,
and that these nodules had definite forms in the different belts, in
some being all rounded, in others radiate. He thought it was very
desirable to ascertain the reason of this. With regard to Ammonites
rostratus, he stated that its aperture was beaked in the young as well
as in the adult state, and remarked that information as to its mode
of growth, and especially whether the successive beaks were absorbed,
would be of much interest. He also referred to the unexplained fact
that a bed at the base of the Gault contains both shells and casts.

Mr. Carrutuers referred to the discovery in the Gault of Folke-
stone, by Messrs. Gardner and Price, of cones belonging to two spe-
cies of Sequoia, and, associated with these, some species of Pinus,
two of which were to be referred to a group of that genus at present
found associated with the two existing species of Sequoia on the
mountains of western North America. ‘These Sequoie from the
Gault are the oldest known representatives of the genus; and it is
remarkable that they should be thus early associated with species of
the same group of Pines which is now represented only in the same
country where the Sequore also grow. Mr. Carruthers believed this
to be the earliest trace of the geographical distribution of plants
which now exist on the surface of the earth.

Mr. Torrey remarked that the lithological and stratigraphical
break between the Gault and the Neocomian is less marked than is
generally supposed. At a place near Folkestone the lithological
difference is so small that it becomes difficult to say where the one
ends and the other begins. At Folkestone the highest beds of the
Neocomian are false-bedded sands; and, contrary to what might be
expected, these sands are the most constant of the series, always
occurring where the Neocomian is represented below the Gault. As
the Gault is traced westwards we come to places where it is difficult
to discern any difference between it and the Neocomian. He stated
that Prof. Way had found the exterior of the nodules to contain
more phosphate of lime than the interior, which seemed to indicate
that the phosphate came from without.

Mr. CHarteswortu doubted whether the fossil egg, if that of a
Crocodile, could be that of a living species. With regard to the

* This discussion also relates to Mr. Meyer's paper ‘On the Cretaceous
Rocks of Beer Head.”

F. G. H. PRICE ON THE GAULT OF FOLKESTONE. 367

forms of nodules, he remarked that in the Crag the nodules round
the fang-like bases of sharks’ teeth were more or less globular; and
he did not think that the form of nodules has any relation to that
of the nucleus around which they may have been aggregated.

Mr. Hawxrys Jouyson considered the nodules to be due to organic
structures, probably sponges, which had grown upon a hard bottom,
and afterwards been affected by its mineral constituents.

Mr. Tat inquired whether the nodules at the base of the Gault
had been rolled.

Mr. Szrtzy, having examined the supposed Crocodile’s egg, de-
clared that from its form it could not be that of a Crocodile, and he
did not think it was that of a Turtle. It might possibly be the egg
of an Ichthyosaurus or Plesiosaurus. He stated that he had found
nodules of different forms scattered indiscriminately in the Gault,
and that his investigations led him to believe that all these nodules
had been subjected to wear and tear before coming into their present
position. The subdivisions of the Gault recognized at Folkestone
would not, he thought, be represented elsewhere; for their mineral
characters were found to change greatly towards the west, the Gault
itself becoming more sandy and micaceous as it approaches the gra-
nitic rocks. He believed that the Blackdown beds represented both
the Greensand and the Gault.

Prof. Huenss thought that we should take as the base of the
Lower Cretaceous Series the first marine beds which succeed the
freshwater deposits of the 8.E. of England and rest on the Trias and
older rocks in the 8.W. He considered them to be deposits formed
during a considerable period, as successive parts of an irregular
land-surface were being depressed below the sea; so that a shore
deposit might be formed at Blackdown while fine sand or clay
was being thrown down further out to sea over that part of the
south-east of England which had already been submerged to a con-
siderable depth.

Prof. Ramsay observed that the value of such detailed sections, in
a paleontological point of view, was very great; and with respect
to the physical relations of the Gault and Upper Greensand, he stated
that in some parts of England there is lithologically no clear line of
demarcation between the two formations; and, in like manner, in
some other areas there is no very definite boundary line between
the Upper Greensand and the Chalk. He then drew attention to

. the views originally advanced by Mr. Godwin-Austen, who showed
that in this part of the world all these Upper Cretaceous formations
were deposited over a great continental area that was being slowly
submerged, so that while the Upper Greensand began to be deposited
in the sea in one area much of the land still stood above water, and
as it got depressed these Greensand strata were gradually deposited
on the sinking land. For this reason the two ends, so to speak, of a
long section of the Greensand will be of somewhat different age ;
and while the end nearest the land was being deposited as sand,
further out at sea true Chalk was being formed; and thus much of
the Upper Greensand may be considered to have been formed con-

368 F. G. H. PRICE ON THE GAULT OF FOLKESTONE.

temporaneously with much of the Lower Chalk under different local
conditions of proximity of land and depth of water.

Mr. Braxe believed that the animal of Ammonites rostratus might
have lived outside its shell.

Mr. Price stated that he did not wish to imply that the divisions
indicated in his paper would hold good over wide areas. He added
that bones of Turtle were not unfrequent in the Gault.

Mr. Meyer said that the division between the Gault and Green-
sand was not distinct. ‘The Beer stone deposited in a hollow to the
westward exhibited a change of texture. The upper beds of the
Lower Greensand of Folkestone represent a lower horizon of the
Lower Greensand at Guildford.

Quart. Journ Geol . Soc Vol. XXX. Pl. XXV.

G.H.Ford & C.L.Griesbach Mintern Bros imp

GAULT FOSSILS.

C. J. A. MEYER ON THE CRETACEOUS ROCKS OF BEER HEAD. 369

29. On the Cretaczous Rocks of Brer Heap and the adjacent Cx1rr-
sEctions, and on the RetativE Horizons therein of the WaRmIn-
stER and BuackpowNn Fossiiirexovus Deposits. By C.J. A. Meyer,
Esq., F.G.S. (Read April 29, 1874.)

Tne Cretaceous rocks which cap the higher ground over a considerable
portion of the south-east of Devonshire, and from the partial
destruction of which vast beds of gravel have accumulated over the
same area, are exposed on the coast-line between Seaton and the
west of Sidmouth in numerous fine cliff-sections. The strata thus
exposed, while resembling in great measure the various corresponding
deposits of other parts of England, possess at the same time a certain
facies of their own which renders their correlation in part a matter
of uncertainty. I purpose to offer in this paper a description of certain
of these cliff-sections, to point out the various petrological and paleon-
tological subdivisions of the strata therein expused, and to leave to the
consideration of others the chief point of interest which I have been as
yet unable to determine, namely the age of the lowermost deposit.

This paper probably contains little which is not already to be found
in the writings of previous observers. But so far as it is the result
of independent observation, it may not be without its value.

Reference to Previous Descriptions.

The following papers, to some of which I shall have occasion to
refer, are amongst the principal contributions to the literature of
the subject :—

Dr La Becue, Sir H. T. (1822). ‘Remarks on the Geology of
the South Coast of England,” &c. Trans. Geol. Soc. ser. 2,
vol. i. pp. 40 & 95.

Dr La Becur, Sir H. T. (1826). ‘On the Chalk and Sands be-
neath it (usually termed Greensand) in the vicinity of Lyme
Regis,” &c. Trans. Geol. Soc. ser. 2, vol. 11. p. 109.

Frrron, Dr. W. H. (1886). “On the Strata between the Chalk
and the Oxford Oolite in the South-east of England.” ‘Trans.
Geol. Soc. ser. 2, vol. iv. p. 233 &e.

De La Becue, Sir H. T. (1839). Ordn. Surv. ‘ Report on the
Geology of Cornwall, Devon, and West Somerset,’ p. 237.

Gopwin-Austen, R. A. C. (1842). “On the Geology of the South-
east of Devonshire.” ‘Trans. Geol. Soc. ser. 2, vol. vi. p. 433.

Hurcntnson, P. O. (1843). ‘The Geology of Sidmouth and of
South-eastern Devon.’ 8vo. Sidmouth.

Renevier, M. E.(1856). Bull. Soc. Vaudoise Se. Nat. v. pp. 51, 52.

Wuiraxer, W. (1870). “On the Chalk of the Southern Part of
Devon and Dorset.” Quart. Journ. Geol. Soc. vol. xxvii. p. 93.

The Cretaceous rocks in their range westward from the Isle of
Wight repose successively on the abraded surface of older and yet

370 Cc. J. A. MEYER ON THE CRETACEOUS ROCKS OF

older deposits. The Oolitic, the Liassic, the Triassic, and Old-Red-
Sandstone strata support in turn the overlying Cretaceous series.
In passing westward these Cretaceous rocks diminish steadily, yet
at the same time unequally, in thickness. They change slightly,
also, both in mineral character and in their fossil contents. From
Lyme Regis westward to Axmouth, and from Beer Head again west-
wards to their last exposure on the coast at Peak Hill, near Sidmouth,
the base of the series rises gradually higher in the cliff-sections.

In the face of this prevailing rise of the Cretaceous series to the
westward, the Chalk-cliffs of Beer Head, the most westerly chalk
promontory in England, owe their preservation in great measure to
a local synclinal or trough-like arrangement of the strata.

Many writers on the geology of Devonshire have attributed the
present trough-like arrangement of these strata to subsidence since
their original deposition. It is not clear, however, that this view is
correct ; and it may, perhaps, be worth considering presently whether
this seeming subsidence is not due in part to an original inequality of
the ocean-bed. For the moment I must pass on to other matters.

I. Cretaceous Rocks represented in the Cliff-Sections of South
Devonshire.

The Cretaceous rocks of the Beer-Head district include the fol-
lowing principal subdivisions :—

Upper Chalk (in part) ?
Middle Chalk.

Lower Chalk.

Chalk Marl.

Chloritic Marl.

Upper Greensand.
Gault.

(?)

These broader and generally recognized divisions are split up
naturally into many minor beds, either by variation in mineral
composition or by the prevalence of a special fauna.

In describing these minor beds, or groups of strata, it may be most
convenient to take them in their natural or ascending order, and for
facility of reference to number them throughout consecutively.

Description of the minor Subdivisions of the Cretaceous Strata as
seen in the Beer-Head* and adjacent Sections (fig. 1).

Bed 1.—Greenish sandy argillaceous strata. Including often,
near its base, a grit-bed of minute subangular pebbles, or frag-
ments of the underlying rocks.

Thickness variable, usually a few feet only; 3 ft. at White
Cliff, 33 ft. at Dunscombe, 24 feet at Salcombe Hill.

Prevailing fossl—Hxogyra conica (small var.).

* The sections here referred to as adjacent sections are :—those of White Cliff

and Beer ; the Southdown Undercliff, or landslip of 1789 ; the cliffs at Brans-
combe Mouth, Weston Mouth, Dunscombe, Salcombe, and Sidmouth.

BEER HEAD AND THE ADJACENT CLIFF-SECTIONS. 371

Fig. 1.— General Section. Scale 48 ft. to the inch.

Upper (?).-.-..---

Chalk ¢ Middle...--—

IDOE do Ee EY Gaia SE

\ Boor Stone.

Chalk Marl..........J PSSSSSSsS egy
e: :,) }--. Phosphate-bed.

Chloritic Marl..........

---- Warminster beds.

SS
<= : = Chert-beds of Up-
as Beare) (eae per Greensand,
Upper Greensand pecans RXKKKXA KX MKXX MKKXHK KK XW 30

RXR KEN KAR KKK LAN NRE NERY

——<—$———— --- Blackdown beds.

372 Cc. J. A. MEYER ON THE CRETACEOUS ROCKS OF

Fossils :—
Exogyra conica, Sow. (small var.)...... Peak Hill. Salcombe. Weston.
Vermicularia concava, Sow. ............ Salcombe.
Arca carinata, SOwW.............ccceeceeeees White Cliff.
i MareXoyalle}s (3} 95. —-opnoboosbooogensnopacaqeo0Gcee Axmouth.

This, the well-nigh universal base-bed of the Devon Cretaceous
series, is, from its somewhat retentive character, usually the most
difficult to examine in situ. It may be seen in part at White Cliff,
and here and there along the cliffs from Branscombe westward to
Peak Hill. It is present also, in much the same condition, at the
base of the Blackdown strata at Punchey Down, near Collumpton,
and again, further westward, at-Haldon.

Bed 2.—Argillaceous sand, with abundance of green earth.
Varies in colour from dark greyish to pale brownish green, appa-
rently according to its condition of moisture. Fossiliferous semi-
indurated nodules occur near the top of this bed at White Cliff and
in places to the westward, and have much resemblance to the “‘ Cow-
stones” of Lyme Regis.

Thickness variable: 15 ft. at White Cliff, 25 ft. at Dunscombe,
25 ft. at Salcombe Hill, 20 ft. at High Peak.

Fossils, in places, numerous.

May be seen at White Cliff and in the cliff-sections between
Branscombe and High Peak to the west of Sidmouth.

Fossils of bed 2 :—

Miadrepoxal(P)jecceteoseescee ee erence Salcombe.
1B ya KOYA01 (0) SagoaeoeacaspoabasmbudascucoDAd Peak Hill.
Nilicifiediwoodees-ceeeeteet eee eeee eee A
Epiaster Murchisonie, Mant. ......... Axmouth.
Hemipneustes Greenovii, Forbes ...... a Peak Hill. White Cliff.
Fusus quadratus, Sow. ..............0055 Peak Hill.
Rostellaria Parkinsoni, Sow............. “H
AMOMIaLIS Doe tojoeceesaeeeeeeeree eet as
Exogyra conica, Sow. ...............s0s008 Salcombe, &e.
Wk NE (SodR soooogooonsceoscoscocue. Axmouth.
Pecten Millerii, Sow. ....3.......00..0000: Peak Hill.
orbicularis (small var.) .:.......... im White Cliff.
Area carinata, Sow...........cceeceese eens A Axmouth,
Cardium Hillanum, Sow. ............... os) White Cliff.
Cyprina cuneata, Sow. ...............60- 53 Salcombe.
Tnoceramus sulcatus, Sow. ............++: 5 Axmouth.
Lucina orbicularis, Sow................+4- 4 White Cliff.
Mactra angulata, Sow. ...............4.. 5
Nucula impressa, Sow. ............0.000+ 43
lineatas Sows: ccsecnccaccwecsico cost
Pectunculus umbonatus, Sow. ......... Dunscombe. White Cliff.
Tellina striatula, Sow.............0....0065 Peak Hill.
(or Psammobia)..............0c0000 ss

Thetis major, Sow. ...........cceeeeneee ee 3
Trigonia (allied to Coquandiana, D’ Orb.) xp

spectabilis, Sow......... .eccceeee eee White Chiff. Beer Head.
iB) e siniasiacsieeinape ome namanemienaoir tee Dunscombe. Axmouth.
5) 194) cosgaodasese qodbacanacoacsn sod qnecad Axmouth.

Venus immersa, Sow. .......-.0eccceee ens ”

Vermicularia polygonalis, Sow. ......... %
ConcayasSovvie.-neeeccsee esc ceeeeian Salcombe. Peak Hill.

BEER HEAD AND THE ADJACENT CLIFF-SECTIONS. 3/3

At White Cliff these fossils occur only as casts and impressions in
the indurated nodules. At Branscombe some of them are more
perfect. At Peak Hill, where the condition of the bed is slightly
different, many of the species occur precisely as at Black down,the
shell itself having been replaced by chalcedony.

De La Beche* has recorded many other species as having been
obtained from this bed at White Cliff; and he was also one of the
first to point out their resemblance to the Blackdown fauna.

Bed 3.—Buff-coloured and greyish marly sands, with large,
rounded, and spongiform concretions.

Thickness variable: 40 ft. at White Cliff, 30 ft. at Branscombe,
25 ft. at Dunscombe, 20 ft. at Salcombe.

Fossils scarce.

Fossils :—
Exogyra conica, Sow. .........+0-....se0ee White Cliff.
kev tata, SOW. -eeeeanensnescecaee: * Branscombe, &c.
Pecten orbicularis, Sow. (var.) ......... 5 Axmouth.

This bed may be seen at White Cliff and in all the sections from
Branscombe westward to High Peak.

Bed 4.—Light-coloured, yellowish, and slightly marly sands, with
irregular concretions.

Thickness about 20 ft.: 25 ft. at White Cliff (?), 20 ft. at Beer
Head, 25 ft. at Dunscombe.

Prevailing fossilsa—Hxogyra columba, Lam., and Jamra quadri-
costata, Sow.

A shingle-bed or layer of sandstone pebbles, mingled with the
broken valves of Pecten and Exogyra, occurs in the upper part of
these sands at White Cliff. Near Branscombe and at Weston
Mouth the broken shells occur without the pebbles.

The White-Cliff section of this portion of the series gives, in
descending order :—

(@)Arpillaceous sand) (22. eeanenadeectensen e-retertaaeen 1 foot.
(6) Dark, greyish, sandy clay, with pyrites............... 1
(ce) Greenish grey sand, crowded with sandstone-pebbles
ana broken! shells) aerces-heseer cee reaeeeceet eee ee 2 feet fT.
(d) Sands with irregular concretions ............+--...... 20",
Hi oye)! Brppanppnocodbeduétoss05ccouondscdcvoududnteceRe 25 feet.
Fossils :—
Exogyra columba, Lam. Pecten elongatus, Lam.
Janira quadricostata, Sow. Vermicularia umbonata, Sow.

Pecten orbicularis, Sow.

Bed 5.—Light-coloured sand, with chert in nodules or layers,
usually more or less tabular.

Thickness variable: 25 ft. at White Cliff, 20 ft. at Beer Head,
920 ft. at Branscombe, 8 ft. at Weston Mouth, 7 ft. at Dunscombe.

* Trans. Geol. Soc. ser. 2. vol. ii. p. 114.

+ This shingle-bed has been thought by some geologists to mark the upper
limit of the Gault.

Oe nGas. No. 119: 2D

374 C.J. A. MEYER ON THE CRETACEOUS ROCKS OF

Characteristic fossils—Hwogyra columba, Lam., and Janira quadri-
costata, Sow.

The diminution in the thickness of these strata to the westward
is so considerable that I have thought it advisable to give the
sections. They are, in descending order, as follows :—

ft. in.

Sand, with chert in layers) ....6.....s.cccs ste. 00 12 0

White Chiff...... Bed of sandstone-pebbles, with broken shells... 3 6
Sandy with cherti.j.), «scene eee econe qd @

Sand, with chert th layers 22. ote... eiieces eeu ee 6 0

Beer Head ...... Broken-shell sedi eeneiotanaecsteeeeeeee acer ek meer Es 1 0
Sand swith clertiy icant ce daeaueny sas eeenenemenencees 11 0

Sand and chert (shattered) ............cscecesucees 6 0

Mieston\ ity. -ceac: Broken\ shells wy reese Ma ne teen eee eee 0 8
Chert : an almost continuous layer............... 1 0

Dunscombe...... Chert in sand (shattered)...........-.0.cccecneeseee i

At Axmouth the chert and sand about it is crowded with sponge-
spicules.

Bed 6.—Light-coloured sand (in places almost white) passing
into sandstone.

Thickness variable; 1 ft. at White Cliff, 2 ft. 6in. at Beer
Head, 1 ft. at Branscombe, 6 in. at Weston.

Fossi—Lwogyra digitata, Sow.

Bed 7.—Sandstone-pebbles, with fragments of large Ostree.
Thickness O ft. at White Cliff, 2ft. at Beer Head, 1 ft. at Brans-
combe, (?) at Weston.

Bed 8.—Buff-coloured-sand beds, irregular in structure, with
zones of sandstone pebbles.

Thickness variable: 5 ft. at White Cliff, 10 ft. at Beer Head, (?)
at Branscombe, 5 ft. at Weston Mouth, 2 ft. at Dunscombe.

Characteristic fossils—Hwogyra digitata, Sow., and Orbitolina
concava, Lam.

Fossils :-—
Orbitolina concava, Lam. ............65- Axmouth. Weston. Beer Head.
Branscombe.
Exogyra digitata, Sow. ...............055 Axmouth. Weston. Beer Head.
Rhynchonella Schloenbachi, Dav. ...... Weston.
Spans) Cie CHGINS 5 5004 c00009400066000000000 Dunscombe.

Bed 9.—Buff-coloured saccharine-looking sandstone, with fine
green grains. Includes layers of quartzose grit.

Thickness variable: 5 ft. at White Cliff, 20 ft. at Beer Head,
15 ft. at Branscombe, 8 ft. at Weston, 4 ft. at Dunscombe.

Fossils :—
Rhynchonella Schlenbachi, Dav. Spines of Cidaris.
Ditto (fragments). Small fish-teeth.

Bed 10.—Rubbly, yellowish, quartzose sands, in chalky paste.
Thickness variable: 6 ft. at White Cliff, 15 ft. at Beer Head,
10 ft. at Branscombe, 5 ft. at Weston Mouth, 4 ft. at Dunscombe.

BEER HEAD AND THE ADJACENT CLIFF-SECTIONS. 375

Characteristic fossils—Siphonia, sp., Nautilus levigatus, D’Orb.
Fossils :—

(Note.—Abbreviations as to localities :—Ax.=Axmouth, BH.=Beer Head,
Br. =Branscombe, D.= Dunscombe, WC. = White Cliff, W. = Weston.)

PASETEAU CG) coe tncee iron bic s va laste soos hae Oe en e D.
iBryozoa) (Several specios)) «06... sa eniasaneneecete ces D. BH.
PS TSHOM IA SPS 4). Secale sein vas shack eae eabeae eee D. Br. BH. Ax.
Ananchytes subglobosus, Leshe...........0..0cc0 cess BH. Ax.
Cottaldia Benettie, Konig — ......-..eessccceseueeess BH.
Discoidea subuculus, Kleim © .......--.cecceseeeceeseuee D. BH
Echinoconus (Galerites) castaneus, Brorgn. (small
APE )M(S) yiesncuns ceceteceduiecsesisseaceemeaineneeee eee D.
Byrne ovaloiiny Ags A ceken enc mae ksereae meee BH.
Bratti sHOr0es h.dctr chek ea eee eae eee BH.
Salenia petalifera, Desm. \..-2-s2-.ces-seesaseo--e-e ven BH.
Ammonites Mantelli, So2w..........secccessseeeeeeeneens Ax.
Nautilus levigatus, D’ Orb. .2.........0.0c0sceeresseee Ax. D.
LOIERIVEE (3) OS eR ee eRe pee nen ero sacnansenucen csederancndemccccons D.
Pleurotomaria Mailleana, D’ Orb. ..........0..cc cee ees D.
PLELOCRL APN ist see Aen nse Nee eee een NaS eee Ds
SPowtrer st tad tunsacuaasanecsWeanaaeaae marcato anes D.
Phevslens Wiewtligmrck, JEMOyADS. Bocosaconodeccoenodcoade D.
Hrochusyisps Mas uectecs ean eee eae canes D.
Aurrivelln, Bausa eI): Onge nce sastetn eee see sere D.
Caprotin acs pis dances secte sen cose meters see eens D.
Gasirochsenas sp. sac pacers te err ce ne oe eee D. BH.
Tnoceramus latus, ME: Ma eee ee otto. BH. Ax.
LO} Qty S) Cleaner a ern adaeactionatced aaa ideceaocdsadan rc: 1D).
Trigoniasulcataria, Wan .neseceeteasesc neces DE
tenuisulcata, Dujard. Sead hinpaacicosoenademcetee D. W.
nS] Dea pb ar esAc rn Sa oecocdib sho boadoa 600400 syosqaunaoneaadE D.
—— abrupta, Buchs i: eee Haggandes 4 aaoneoors D.
Pernai(oryAvicula)?) cic scenaseter coseee er eee ence D.
iiimaormata, DOr0s ...osceaseanepce enter eenees De
owner JOO, nonacadcodsaoseoneoodo00dNeucd D.
semisuleata, Goldf, ...0-0..cn-.ensncecseseesieices 1D).
Janira quinquecostata, Sow. ..........scececeees eee 1D
Mytilus striato-costatus, D’ Orb. ..........sccsceeee es Ax.
Ostrea (allied to O. Normaniana, D’Ord.) ......... Ax. BH.
Garimata:S Os ss sceosrssoee rere ence eRe eer ee enee D.
Giluyianase/1..... hee eeeenetrreraeceR cee eee err ne DD Ax.
Beetenpas pers PAG cise. due nearei-b men aye ar me ECE: D.
Rhynchonella dimidiata, Sow. .......:..cccsseeeeeenes BH.
ganna J8y Par oobepenoopenends=osccenos +3 700 vodenD ae
OONVEE MSelTthoconoenes603000000600090409 700000003050
Schloenbachi, DOs. wsasendees scweteeheene apne el
Terebratella pectita, Sow. .........::ssccseeneesscensees D, BH.

Bed 11.—Yellowish quartzose sand, with fine green grains, in a
chalky paste, semicrystalline.

Thickness variable: 2 ft. at White Cliff, 3 ft. at Beer Head, 2 ft.
at Branscombe, 6 in. at Weston Mouth, 2 ft. at Dunscombe.

Characteristic fossils—Crustacean (Callianassa 2), Exogyra co-
lumba (small var.), and Terebratella pectita, Sow.

Fossils :—
Crustacean (CallianassaP ..........secsseeescseereerees BH.
Bryozoa (Many SPeCies) ......-6.eeceeeeseenecsnereneven D. BH.

376 C. J. A. MEYER ON THE CRETACEOUS ROCKS OF

Ananchytes subglobosus, Leshe............ceccce cence D.
=. ep yis, Dele) ae aun os Na eeny ny ye can ners St BH.
Discoidea subuculus, Klein ............c ccc e cece ees Bre
Goniophorus lunulatus, B21) [ie abiaa Antone Suan eae D.
Miucrasteryispy eyunscacdateciecmnn sede teehee eee BH.
Pseudodiadema IMEVaelbbaMl, Z1GZ, SoadooceconoaneHasa00e0 BH.
VABBLOERGES JEVROMETOs basadanoacdapsesadouabqngnsdson a: BH.
Mrochus isp ceseeeeeee eee eee a uay eevaa tarenaunee D.
Caprotinayysp cain cscst aasBoracee voces ce semctonncsecenpees D.
Exogyra columba, Lam. (small var.) ............... W.
A Bihar jos enh eagacdagammacdue docaacaanackenosnnanaeibessano. D.
Cyqoe hk W DO ay desodnoasstn nc soscncsavedscoumbasun: D.
Janira quinquecostata, Sow. ............cceeece cence ees BH.
PSL OMRRH Rea cau oBAcorrna a RAamaaNaHe Hie anacucose uonaadat BH.
Pecten rhotomagensis, D’ Orb. ..........2-.0.00000-0 D.
—— Puzosianus, Math........cccecceccceneceneeneennes BH.
aspen; Lamar cnenOasas venue ecanl a uh eee a BH.
Trigonia, sp. (allied to T. sinuata, Park.) ......... D.
Rhynchonella dimidiata, Sow. ..................00005. D.
prasian a, ZOKD sinc 2s a: denon. Aacee eee maeecrt D. BH.
Sid ollosaloryclon, DGB cocoogcoasnaconcosab0b00b0e0ns00s BH.
Merebratellaspectitasy Sow mencrreeoercsceseceec ences D.
Terebratula rugulosa, Morris.............0.0.0.0.0e0 ee D.

Bed 12.—Compact, nodular, rubbly bed, with green grains and
quartz grains, in a chalky paste.

Thickness variable: 1 ft. at White Cliff, 2 ft. at Beer Head, 2 ft.
at Weston, 23 ft. at Dunscombe.

Characteristic fossils—Catopygus carinatus and Ammonites varians.

The top of this bed usually presents an uneven surface at its
junction with the next above, the junction of the two being also
marked in places by a ferruginous line.

May be seen at White Cliff and in all sections from Beer Head
westward to top of Dunscombe cliff.

Fossils :—

Bryozoa (many species) REBRAC HEHE AROEABOR MaSatHEARn GAS D. BH. Ax.
Scyphia (species)...............6.0¢ eo ava ate ROME ae D. BH.
Spongia (several species) ......... pobsesbbannondouc0daE D. BH. Ax.

[These are mostly coated with green Sees and more or less covered with
Serpule and the attached valves of Crania and Plicatula. |

Ananchytes subglobosus, Leske............. matrtlataecteiae BH. &c.
Catopygus carinatus, Gloldf................ceeceecenen es BH. &e.
Discoidea subuculus, Kletm ..... ccc cece cece eee eee BH. D.
Echinoconus (Galerites) castaneus, Brongn. ...... D.
Nucleolites lacunosus, Gloldf. ...1.....c.cce ce eeee eee D.
Pyrina Desmoulinsi, D7 Arch. ':..0.. 6.252. 6..0.- +500. _D.
Pratt; MOOS) saueas saci an tence eae cee nner nees BH.
yuan, ZIG}, Sossoncadoosd006 baa cia cuts on ae ober ei aly
SalemiaycibbayrA gan aerenect creer cece crensnces rer ee BH.
petalifera, Desm. ......0..05.+ a Aik dole koa teas h BH.
Spatangus bufo, Brongn. .......... foaten mace reise Mal sere BH.
Le Vas, Dele hie eeniyh pte pnei ctl cacti ens D.
Ammonites Mantelli, Soww.............ccceeceeceeveeeees Ax.
Copel; Brongn ialscseis. saensscusten tee steee: Ax.
re = VAPIANIS OO Wen aeeurnescebce eciewiniseebee ace eee WC. D.
Nautilus Largilliartianus, D’ Orb.................0004. 23% 1D),
Fittoni, ShArpe .........ccccecc ese ee ees Sra Seselteree Ax.

———— Neevi pats, DOB sien ssjnin voices nseeceemenat st Ax.

BEER HEAD AND THE ADJACENT CLIFF-SECTIONS. Sth

Scaphitesvsequalis) Sozw.|'........c.c+ss+0- 00008 heels WC.

nnrilitesicostatus Lana oes eee naeeee ene ene 1D).

BB CCHeiy SOU) se cise satacrusisawaunic ouane eRe ee iAcxe
A\@lllevaey GHSIES, TOAOTADS seonconoounedeonsecdscesdocococus D.
FETUS IB Nas ete nite fos oialaarols ain aie cavacioneranleelnna anes D.

UN IICA RSP yee nee cane decersindcaiciaemeaad one nmeen notes D.

UNERIGO PSISS SPoeteses che svaves si dacscracceacedse some aeee 1D).

IETS MYOSOMIANTIA 1 BPs saestciho) «21 ccec <ooe cea ee aaa Ax.
[SARI DSE MIE INC, Soonasocccgosocossccocacudecaooor ANsey 10).
CASSISIAN ON 2)" O70 teeter aan see ieee eee Ax.

IESLELOCETANES Puy diene sarees ceceutcameeaens Reena ee BH. D.

Solleramrns ras, Ma osonsenacnososedosenscacosaconca Ax.

AtieGams OtaANE, VHOCCsosnccsc0scosvseno0snesncoucoaebee 1D),

Cardium (allied to C. Hillanum, Sow.) ............ WC.
IWiewll eenayonia, JOVORAs. cosvaancscansooanbeantocctcnsc WC.
(Unicardium) ringmeriense, NOOR oncescerice D.

Ijucina turoniensis; DUOPGs , 4,.cessotacsusiadsoseee 1D}

Trigonia abrupta (?), BUCH, a rtecnemtrectecsnc mene eres D.

=== Gambit, (2) Laie croccconssavaconndons 0s00c008e D.
sulcataria, LE OMS Nera Oates ire ee ets o dsl aoe D.
(allied to T: sinuata, Park.) ......0..0cs00ceeee Ax.

PATTOTIIAS S Dose seeen cascenceeee a seiier seach arate tes BH.

Exogyra columba, Zam. (small var.) ............... Br. W.

Janira quinquecostata, Sow. .............eeceeeeeeeceee 1D).

Lima (allied to L. intermedia, D’Ord.)............... 1D),

Pic] Shang aanbedocouanaresnnacear cloddecone ane sdk someon onua Ax.

ee DEOL OS. Sane EEE ee eee Ax.

BSE RP COLE aceO rR UcdiondornonanaotbodooDepeaeoreadobah Ax. D.

Nonine ligeriensis, LD! OGD sysdaloctaritinas das casein Ax.
Guerangert, D’Orb. 0.2.0.0. wecpece-es-noeeee ee: Dz.

Ostreacarmata, Mam c.ceneeereeree teen eet oreee 1D).
(allied to O. normaniana, D’ Oi: » sacebreneere WC.

IRecten aspen, Wann e cna eeeeher acer ee nce Iplal, lee, 1D),

Crania cenomanensis (?), D’ Orb. ............0..0. ee AXoxe AD):

Megerla lima, Def7n, ... .....ssoese taser ee acre ee BH. D.

MEO Lin) }opang stoadapperosaddorbs yosaduded cog oatoceon D. Br.

iNerebratellapectita, Sas e-a-ncreseeeeeeee een aeecre 1D),

Minizonasemls in certUs, 000... seer ssc eeeee en eeaetcr BH. D.

Merebratularovatas SOcOves-cteeeeeaeeeeeeeeereceee ee cene BH.
TEDW}; TAKS Sepscaeeannctobcoesopabcec-oonogse BH,
ROIMENINO SE), LL spocnapeschoncadeccedonsaseds bode 10),

= ONESH,, SOW: oecied sisrngonss vsihe nents seme ye aamescaae ANG

Rhynchonella dimidiata, Sos tree eee eeece eee eee BH. D.
GoVaKyes: th, VO one deans orssonopscuseceso0u0sresdc5e00 isslals 1D)

-—. Grasiana, DV OPO: s05oss.en soo enor BH.

————Miantellaama,, S020, s.cedsseaeccideseeeeeescenesterr: BH,

Se VVIESUI QUE: occ cacennesiens cnet eeuecoreacare BH.

———— Soaillosmnloeveatl, JOU cooonsaoncscouctoncsnosnseconce. WC.

Bed 13.—Greyish or brownish chalk, with green grains, quartz
grains and phosphatic nodules.

Thickness variable: 4 ft. at White Cliff, 5 ft. at Beer Head,
4 ft. at Weston Mouth, 24 ft. at Dunscombe.

Characteristic fossils—Discoidea cylindrica, Lam.,and Rhynchonella
Mantelliana, Sow.

May be seen at White Cliff and in sections from Beer Head west-
ward to top of cliff east of Salcombe.

Fossils :—

JNRhRGE () pasonibon copebedormbonctndhonoosceacounbonadadaGd. BH.

378 C. J. A. MEYER ON THE CRETACEOUS ROCKS OF

Micrabacia coronula, Goldf. ............ Esai aae anaes Br.
13a (040. Gop penbtdosucccriponocdoaosoospacoctepenaboorvcL bs BH.
Sponge-nodules (phosphatic)................0.0..0..0:- BH. Ax.
Cidaris (spines and plates)...............0.0:eceereeeees Br.
Discoidea cylindrica, Ldin................ce.ee ee ree rece Ax.
—— subuculus, Klein ..........2.0.iceececec eee nce Ax. Br. BH.
Hchinocyphus difficilis, AGE Mees) oak Me BH.
Pseudodiadema variolare, Brongn.......1..c0.cc0ser0e- BH.
Belemmitellas. aed sccicesc nue esas) ou sinelaideronblaceciesteseee BH.
Ammonites Mantelli, Sow......................00-0see0. Ax. D.
TAO ONTAEKAETNSTIES, JOR, sba565555052500003002052053 Ax. BH.
——— SI ALVICUIATI Ss SOW) licen deccatcoteeeece eee nee star WC.
Sy SPOS eee ac ciceit cane cnlclsise sete a ieeier a emer seinen BH.
Scaphites OVOMGPHWWIEE INOQD5 sosophcacuscocodsonboucso.6bde Ax.
Turrilites tuberculatus, Bosc............. isis s ueeyeeae ee a BH.
Becheliseiene as casinesanacertas see sass eae te Ax.
Jisigey Wilenlilcryne, JOY OTHG sonosdocoscdos05900000s4u2ac0000. BH.
Lucina turoniensis, Di Orb... -.:.2..0-..---sesneacese BH.
Janira quinquecostata, S(O Sea eee eae ce RAC ae arn oe Br.
PlicatuJa spinosa, Mant. ................e.eeseeeeeeeeeees D.
Wena @ose), CXOUCHS concoscoscosq9eGoeossouudesoLoduS D.
= lin nmmearseninny JOY OWRD, cosncossstoccsesos00000000- Br.
Rhynchonella dimidiata, Sow. ...............cecee ee WC.
——— (Grasiana, DD? OTD. odsccuscicsmacuececiaeeerere sere D.
Narteltiane: SOW eae rmesaaeth hee wae RUSE E BH. D. WC.
Terebratula ovata, SOW. .......0.c.scceecccecsecucsencees BH. D.
Miegerlia/limay Define cs irascccwscccnceesesscerumecenec aes D.

Note.—Many of the fossils of this bed seem to have been deposited in the
condition of casts; and some at least of these may have been derived from a
partial wearing away of the Chloritic Marl.

Bed 14.—Hard nodular chalk, with fine quartz grains.

Thickness variable: 4 ft. at White Cliff, 5 ft. at Beer Head, 23 ft.
at Dunscombe.

May be seen at White Cliff, Beer, Beer Head, and sections west-
ward to Dunscombe.

Fossils :—

Discoidea subuculus, Leske ..............cesceceeeeues BH. D.
Cardiaster (allied to C. rostratus, Forbes) ......... BH.
Ammomitess Mantel Soweneceereceeeceeteeeceeeeeetre BH.
Rhynchonella Mantelliana, ISO CTEIE Rao spac Saaeecnsceod Beer.
— — limbata, var. robusta, Tate .............e00..00: Ax.

Bed 15.—Creamy white or yellowish sandy chalk, semicrystal-
line. (The Beer Stone of the Beer quarries.)

Thickness variable: 5 ft. at White Cliff, 10 ft.at Beer quarries,
6 ft. at Beer Head, 2 ft. at Dunscombe.

Characteristic fossil—Inoceramus mytiloides, Sow.

May be seen in part from White Cliff westward to Duns-
combe.

Fossils :—
Ananchytes subglobosus, Leshe..............c0.000000 Br. BH.
Discoidea subuculus, Klein ............e0.0cc0ceeeeeee Ax. BH.
Kchinoconus (Galerites) castaneus, Brongn.......... BH.
Tnoceramus mytiloides, Sow. ...............ecceseeee ee BH. Beer.
JUhine rims, OPO. sasgoossoass0asossdosanqne0 obboee Br.

BEER HEAD AND THE ADJACENT CLIFF-SECTIONS. 319

Bed 16.—Hard, nodular, sandy chalk, without flints; semi-
erystalline in structure. (The Bastard Freestone of the Beer
quarries. )

Thickness variable: 8 ft. at White Cliff, 14 ft. at Beer quarries,
14 ft. at Beer Head, 3 ft. at Dunscombe.

Prevailing fossil—TI. noceramus m ytilordes, Sow.

May be seen in part at White Cliff, in the Beer quarries, and in
face of cliffs to west of Beer Head. Passes out in top of cliff to
east of Salcombe Mouth.

Fossils :—
Ananchytes subglobosus, Leshe.........c.csccveeeneees BH.
Cidarisy isp: dara cccss Sess aheahesees deck race Ge coteeee BH.
Discoidea subuculus; Alem ..2.405..0s2-+-c 0s ese BH. BQ.
Echinocyphus difficilis, AGIA ee aciva dae tettsine fetiece ce BH.
Echinoconus (Galerites) castaneus, Brongn.......... BH.
Micrastercor-bovisy Holes ee -seeeee senate cease eee
palentaumibrellasedgeaseetene meee renee ne er tee BH.
Inoceramus Brongniarti, Sow. ...........c.eceeeenee es Br.
miy tiloides') Sows sana. er enassese cence raaeores Br. Ax.
Rhynchonella plicatilis, Sow., var............0.2.0-0+ Beer.
Culvderis, Bron gis senate hmacncnevaeetmcces se: Beer.

Bed 17.—Marly and, in places, nodular chalk, with flints. Chalk
veined with bands of pale grey or yellow. Flints finger-like, dis-
persed. At base a zone of large flints.

Thickness about 10 ft.

Characteristic fossils—Jnoceramus mytiloides, Sow., and Micraster
cor-bovis, Forbes.

This portion of the chalk with flints may be seen at White Cliff,
Beer, Beer Head, Branscombe, and Weston. It passes out near the
top of cliff to east of Dunscombe.

Fossils :-—
Ananchytes subglobosus, Leshe.............cscseeeeeee BH.
Cidaris (allied to C. clavigera, Mant., spines and
POLLIONS Of) oy. cis oasnes soo yaee eee Eee cee e a eee: Br.
Discoidea subuculus, Klein .........0.ccceceeceneenees BH.
Echinoconus (Galerites) subrotundus, Mantell. ... Ax.
Mucraster cor-Dovis, Morbes) <se..c--seeeadeeneeee Ax.
Pseudodiadema Michelini, AG. carsiece Mente eee ees Ax. BH.
Salemiay petalitera, Desi.) ja0,--sece-ee ss eceaee seen a BH.
Bmerimite (portions(oly 02. j-.2.ssecsaecstecmcaeeenettass Br.
Oreaster (Mortions) Of) | esssssece sped eee secant Br.
Tnoceramus Brongniarti, Sow. ............0eceeee eee Axm. Br.
miytiloides, S0W.. ia. pisbaacsnee teapot Reagents Br.
Rhynchonella Cuvicri, Brongn..........1.:12-00re00+- W.

Terebratula semiglobosa (var. subundata), Sow.... BH.

Bed 18.—Chalk with numerous flints, mostly in layers. Chalk more
or less nodular.

Thickness about 40 ft.

Characteristic fossils—Holaster planus, Mant., and Inoceramus

Brongniarti, Sow.
Occupies a considerable part of sections at White Cliff, Beer, Beer

380 C. J. A. MEYER ON THE CRETACEOUS ROCKS OF

Head, and Branscombe. Passes out at top of cliff to east of Brans-
combe.

Fossils :—
Holasterplanuss Vantec renee eceee BH.
Inoceramus Brongniarti, Sow. ............0..ceeeee ees BH.
Ostrea vesicularis, Lam............ SE aera eee BH.
Terebratula semiglobosa, Sow. .............c0eeseeee es WC.
Rhynchonella Cuvieri, Brongn. .............0. secs ees BH.

Bed 19.—Hard chalk without flints.

Thickness about 15 ft.

Characteristic fossil—Jnoceramus Cuvieri, Sow.

A greyish marly-looking band, of about 3 ft. thickness, occurs
above the middle of this bed. Occupies part of cliffs at White
Cliff, Beer, and Beer Head. Runs out near top of cliff to east of
Branscombe Mouth.

Fossils :—

Discoidea (allied to D. minima, 4g.).........008...0.. BH.

Tnoceramus Cuvieri, Sow. ...........+0-ceceseceee sevens ae

Spondy lusts pte ceases ae tssew soca ee kee cree

Terebratulina gracilis, Schloth. . 600 50960 000 00 B BH.
IWiemtiareiaR, JEYOVRA, soocccndoonanceo0s cos 009000a00 BH.

IS EMIAtA.| aba scecteccncn emetic eects shaiene BH.

Rhynchonella plicatilis, Sow.............:0. eee sec eee ees BH

Bed 20.—White chalk with numerous flints.

Thickness about 60 ft. at Beer Head.

Characteristic fossils—G‘alerites albo-galerus, Lam.; Micraster
cor-anguinum, Gmel.; Terebratula carnea, Sow.

Occupies upper part of cliffs about Beer and part of Beer Head.

Fossils :—

Galerites albo-galerus, Lam. .......:-...sseeeeeeeereees WC.
Micraster cor-anguinum, Gimel. .........c ec eeeee cee BH.
Spondylus spinosus, Sow. .........c.cssecevens eee encees WC.
Ostren ice wecaacet or cuca ene nreoearees nattan cena aceite WC.
Crania ienabergensis, fetes...-ccsscc-eseseeseceeeeeescee WC.
Merebratulaycarmeas) Soweensaceeceeecsestceneeeeeeeeeee WC.
Terebratulina gracilis, Schloth. ...........-.0.00c00000- WC. BH.
=== BENIE, UGolnge socdaasoosnseqsn0ss005G0sNb00d000" 000 Wwe.

In the foregoing description many minor details, including purely
local variations in the mineral condition of the deposits, have been
necessarily omitted. The lists of fossils, also, include such species
only as I have observed in situ in the beds described. ‘These lists
are consequently incomplete, and in some cases, as I am well aware,
do not include one half the species obtainable.

II. Classification of the Deposits.

In attempting to classify these various deposits one meets with
considerable difficulty. There is in the first place no well-marked
base-line to the Devon Cretaceous system, the lowest beds, both
lithologically and paleontologically, answering almost equally well

BEER HEAD AND THE ADJACENT CLIFF-SECTIONS. 381

to Upper Greensand, Gault, or Lower Greensand. Then, again, the
upper limit of the series is almost equally ill-defined.

Subject therefore to correction as to the age, or precise geological
horizon, of the highest and lowest deposits, the grouping of the
beds enumerated in the previous pages should probably be as
follows :—

Beds 1 to 3 may represent collectively either Gault, or Gault
and Lower Greensand (in part). They represent also, without
question, the so-called “‘ Greensand ” of Blackdown.

Beds 4 to 9 inclusive are clearly Upper Greensand.

Beds 10 to 12, although attaining in the Beer-Head district a
most unusual thickness, must be referred collectively to the so-called
“‘Chloritic Marl” of the Isle of Wight.

These beds (10 to 12), in their minerai character especially, have
much in common with the higher beds of the Greensand of Warmins-
ter. They include, also, amongst their fossils the greater portion of
those usually quoted as Warminster species. There is little doubt,
therefore, that the fossiliferous portion of the so-called “ Upper
Greensand” of Warminster is, properly speaking, ‘“ Chloritic Marl ”
instead of “ Upper Greensand” as usually stated. And this point
is of some importance.

Bed 13 represents the base-bed of the Chalk Marl, or that part of
the Chalk Marl which, in so many localities, contains lumps and
nodules of phosphatic matter.

In most of the Devon sections, as, indeed, in numerous other
localities, beds 12 and 13 are distinctly separated. That is to say,
the Chalk Marl (bed 13) rests on an uneven surface of the Chloritic
Marl (bed 12). ‘This, again, is a fact of much importance in its
bearing on cretaceous geology, and one to which I propose to refer
again in a separate paper.

Beds 14 to 17 represent apparently the Chalk Marl (in part) and
the Lower Chalk (in part) of our more eastern counties. It is
impossible, however, to say with certainty how much belongs to
either.

Beds 18 and 19, as shown by their fossils, fall in, I think, undoubt-
edly between the Lower and the Upper Chalk, and are so placed,
provisionally, in Table II.

Bed 20 is possibly Upper Chalk (in part). I cannot vouch, hoy, -
ever, for its exact horizon.

III. Relation of the Cretaceous Rocks of the Beer-Head district
to those of certain other localities.

This paper being intended simply as a general description, I have
not thought it necessary to encumber it with special sections. In
the absence of such sections, however, it may be as well to trace
out briefly the relations of the more variable strata of the district
under consideration to those of better-known localities. And here
arises, once again, the question of the age of the lower deposits.

The age of the beds 1 to 3, which form the lower and most

3882 €. J. A. MEYER ON THE CRETACEOUS ROCKS OF

argillaceous portion of the Devon Cretaceous strata, has been at
various times very variously estimated.

Sir Henry de la Beche in 1826 * and again in 1839 T appears to
have classified these strata with the Upper Greensand.

Dr. Fitton in 1836 + makes them Lower Greensand.

Mr. Godwin-Austen in 1842§ described them as possibly “a
sandy condition of the Gault.”

In 1863 ||, and again in 186649], I attempted to show, partly on
paleontological, partly on stratigraphical evidence, that these strata
belonged to the top of the Lower Greensand rather than to the
Gault or Upper Greensand. It is fair to say, however, that the
paleontological evidence then chiefly relied on has proved far from
trustworthy.

Mr. Etheridge, to whose opinion as a paleontologist every one
may well bow submissively, places these beds on the horizon of the
Gault, mainly, I believe, on the evidence of the Black-Ven section.
At the present moment I am almost prepared to agree to his opinion.
The sandy-argillaceous strata near the base of the Blaek-Ven section
contain unquestionably a large number of Gault fossils; and there
is full evidence of the continuance of similar strata to the west-
ward ; but, still, are these strata entirely Gault ?

True Gault, such as one knows so well in the Isle of Wight, is to
be seen near Punfield, in Swanage Bay, where it is underlain by a
considerable thickness of the Lower Greensand. It is traceable as
Gault along the coast-line as far westward as Mewps Bay. At
Lulworth Cove it is to be seen no longer as Gault, which, in its
argillaceous condition, has either thinned out or given place to
dark greenish sandy strata with zones of large concretionary
nodules.

These beds contain Gault fossilsin abundance, and (what is well
worthy of notice) pass upwards insensibly into Upper Greensand,
and downwards into ferruginous sands of the Lower Greensand.

In the altered condition of the Gault as present at Lulworth,
one finds, therefore, an approach already to the condition of the so-
called Gault as seen at Black Ven. And there is no doubt what-
ever that Gault in such condition is present both at Black Ven and
in the sections to the westward. But isthis all? I think it possible
that Lower Greensand, which, as we have seen at Lulworth, actually
accompanies the Gault to its extinction as a clay-bed, may form
a part also of the still doubtful strata to the westward. As to the
correctness of this supposition, it is only fair to say I have no evi-
dence whatever.

The Upper Greerisand proper of the Beer-Head district (beds 4 to
9 in Table II.) exhibits to some extent a purely local development,
The zones of sandstone pebbles or “‘ shingle-beds ” described by Mr.
Godwin-Austen in 1842 **, for instance, and which form so marked

* Trans. Geol. Soc. ser. 2, vol. ii. + Ord. Surv. Report, p. 237.
{ Trans. Geol. Soc. ser. 2, vol. iv. p. 233. § Idzd. vol. vi. p. 449.
|| Geologist, vol. vi. p. 50. | Geol. Mag. vol. iii. p. 13.

** Trans, Geol. Soc. ser. 2, vol. vi. p. 449.

BEER HEAD AND THE ADJACENT CLIFF-SECTIONS. 383

a feature in the Beer-Head and other sections, are in great measure
local. The recurrence of these shingle-beds throughout the several
sections in the district, and their limitation to distinct and definite
horizons, would seem, however, to indicate some special periods of
disturbance. It is at least worth notice that the horizon of one of
them is marked as far eastward as Lulworth by a layer of phos-
phatic nodules. These pebbles, in the Devon sections, are formed
entirely of sandstone, and, as long since also suggested by Mr.
Godwin-Austen *, would seem to have resulted from the destruction
of. Portland strata.

The sand-beds 6 and 8 of the Branscombe and Weston sections
are also in so tar peculiar to the Devon Upper Greensand, that
they are scarcely represented either at Lulworth or in the Isle of
Wight. It would appear, however, that both these beds exist to the
westward, Orbitolina concava, Lam., and Ewogyra digitata, Sow.,
their characteristic and common fossils, being present in the Haldon
Hills 7, in the Irish Greensand +, and, I believe, also in the greensand
of Normandy.

The Chert-beds of the Devon “‘ Upper Greensand,” although
greatly varying in thickness, are traceable continuously from the
Dunscombe Cliffs to the Isle of Wight. From the Blackdown Hills
they have been in great part denuded.

The Chloritic Marl of the Beer-Head section is remarkable (as
already stated) on account of its unusual development, the beds
10 to 12 of Table II., which from their fauna and mineral character
certainly represent the chloritic marl, having in the Beer-Head under-
cliff a thickness collectively of not less than 20 feet. It is true that
this thickness is exceptional, and that the beds diminish rapidly to the
east and west of the Beer basin. Yet their presence in such local force
is worth notice as illustrating the variable character of this division.

Judging from appearances, the Chloritic Marl of many of our coast-
sections would seem to have been a deposit of slow accumulation.
Indeed one may observe in it, in places, an almost entire change
of fossils within the thickness of only a few inches. Such being
the case, an occasional thickening of the strata, as at Beer Head,
and possibly at Warminster, is far from unlikely.

In tracing these beds eastward, the threefold subdivision of the
strata, as seen between Beer Head and Branscombe, becomes less
and less apparent. It may be seen, however, although somewhat
obscurely, in Pinhay Cliffs near Lyme Regis, at White Nore in the
Isle of Purbeck, and more clearly in cliffs just east of Punfield
Cove. At Chard and Chardstock the threefold subdivision of the
strata has been well described by Mr. Davidson (Mon. Pal. Soc.
vol. viii. p. 114), the beds 1V., V., and VI. of the section therein
given answering extremely well to beds 10, 11, 12 of the Beer-Head
strata. Curiously enough, this section of Mr. Davidson’s appears to
have been taken by many continental geologists as a measure of
the whole of the English Upper Greensand.

* Proc. Geol. Soc. vol. iv. p.197. + Trans. Geol. Soc. ser, 2. vol. vi. p. 452.
{ Quart. Journ. Geol. Soe. vol. xxi. p. 35.

384 C. J; A: MEYER ON THE CRETACEOUS ROCKS OF —

The base-bed of the Chalk Marl (bed 13) with its phosphatic
nodules, which is fairly represented in the Beer-Head district, is
also traceable more or less continuously to the eastward. Some-
times, indeed, it almost merges into the Chloritic Marl; yet even then
it is always traceable. Its upper bed, however (bed 14), and the
succeeding representatives of the Lower and Upper Chalk (beds 15
to 20) are, on account of their local peculiarities of structure, far
from easy to correlate with the thicker masses to the eastward.
This difficulty of correlation is also increased considerably by the
present ambiguity of the terms Upper and Lower Chalk. The term
Upper Chalk, in its ordinary, text-book acceptation, appears hitherto
to have signified chalk with flints. And this acceptation of the term
would have been perhaps as good as any other if flints occurred only
above a given horizon. That such is not the case is sufficiently
evident. The chalk of Yorkshire, for example, contains flints on a
lower geological horizon than that of Sussex or of Hampshire.
Now it is evident, as shown by its fossils, that the chalk of the
Dorset and Devon sections contains flints at an exceptionally low
geological horizon. Mr. Whitaker, in a recent communication to
the Geological Society (Quart. Journ. Geol. Soc. vol. xxvu.) has
pointed out this faet very clearly where (pp. 95 & 96) he refers to
the presence of flints in the Chalk Marl itself. It is clear, therefore,
that the beds above the Chalk Marlin the Devon sections are, from
their exceptional character, exceedingly difficult to correlate with
their true equivalents to the eastward.

Taking bed 13 as the true base-bed of the Chalk Marl, bed 14 must
at the least be added to that division.

Beds 15 and 16, which include the Beer Stone, will then fall in the
Lower Chalk. And in this point I fully agree with Mr. Whitaker.
But I believe the lower part of the chalk with flints (beds 17 and
part of 18) of the Beer section to be also Lower Chalk, as compared
with that of Surrey and the Isle of Wight, and that, as above
stated, the flints are here below their ordinary level *. It is bed 17
which caps the cliffs to east and west of Branscombe, and which, by
a partial thinning-out of the intermediate strata, is brought down in
places so nearly on the Upper Greensand.

The succeeding portions of the chalk of the Beer district (beds
18-20) I have had too little opportunity of studying to describe
with confidence. I may say, however, that the zones of fossils are
fairly marked, and that these appear to indicate the presence of a
considerable portion of the higher chalk.

On comparing the Devon cretaceous system with that of the north-
east of Ireland as described by Mr. Ralph Tate (Quart. Journ. Geol.
Soc. vol. xxi. p. 15), one can hardly resist the supposition that the two
series were originally connected. There is at the same time, however,
so marked a difference in the distribution of their respective faunas,
that one must question either their former actual connexion or the
horizons at present assigned to the Irish strata.

_ * Vide a paper on the Surrey Chalk, by Caleb Evans, Esq., F.G-.S. (Proc.
Geologists’ Association for 1870).

BEBR HEAD AND THE ADJACENT CLIFF-SECTIONS. 385

These cretaceous rocks of Ireland have been referred by Mr. Tate
to two formations—‘“ the Hibernian Greensand, and the Upper
Chalk,” the term ‘“ Hibernian Greensand” being applied to the
eretaceous beds underlying the Upper Chalk of Ireland.

The Hibernian series, says Mr. Tate, “forms three lithological
zones, each with its own suite of organic remains.’’ For their full
description, however, I must refer the reader to the original paper.

Now, of these three zones, the lowest, or “‘ Glauconitic Sands ” of
Tate, apparently represents to some extent the beds 1 to 3 of the
Devon sections.

The succeeding “‘ grey marls and sandstones with chert” clearly
represent the true Upper Greensand, in part (beds 4 & 5), of the
Devon sections. ;

The << chloritic sands and sandstones ” forming the upper part of
the Hibernian Greensand would seem by some of their fossils to:
represent the beds immediately beneath the Chloritic Marl (beds 6
to 9) of the Devon sections. They contain, however, apparently a
singular mixture of species, and are consequently not easy to corre-
late. And the same must be said, I think, with respect to the so-
called Upper Chalk.

IV. Conclusione

Taken as a whole, then, and in spite of the as yet doubtful age
of the lowest and highest deposits, the Beer-Head and contiguous
sections of the Devon Cretaceous strata present collectively a highly
important section. One may observe here, as probably nowhere
else in one continuous section, the true relation of the fossiliferous
Blackdown strata to those of Warminster, and, what is at least
almost equally important, the relation of both of these beds to that
of the true Upper Greensand of the Isle of Wight.

The ‘‘ Blackdown beds” in these sections underlie most clearly
the true Upper Greensand, and hold (apparently) the position of the
Gault or of the Gault and Upper Neocomian in part.

The ‘‘ Warminster beds,” on the contrary, are seen to cap the
Upper Greensand, and are therefore in reality Chloritic Marl.

If, then, the above should prove to be a correct reading of these
Devon sections, it cannot but affect to some extent the use of the
term Upper Greensand; and I submit that such term should be
applied exclusively, as far as possible, to beds between the Gault
and the Chloritic Marl, and that the Chloritic Marl should be con-
sidered a distinct division.

c. J. A. MEYER ON THE CRETACEOUS ROCKS OF

386

‘19d. ————
-dQ ‘o[PPUAL

~--—— ——__—_.

ej eeeeeefeeeeceles pe eel tees cerceseenes wT “eorapurXo

x + + fo) ee Cy (od eee leew else ween (enatttlas wewelae Pee ein eee oni bees wary ‘snjmonqns eBaploosiqy
seelee Gog ogni hicieron (icici iii Guy Mequuleucyas eIpleq}OO

(2uo7T “etoS1avyo “CE 04 pare)
sausonl|oo0q0d| po000000 SSOnnC One ncnnnrrericicn) ‘ds Cee
SSO ROOMS nSnnno gna innr iy ‘ds ‘SIIB@pIo,
ss fnjoy ‘snyeutieo snSidoye9
SOOO Oni niin (saquoq
‘snqeajsor “OQ 0} pol[[e) coqserpareg
anja(y ‘IAB
' aysarT ‘snsoqoysqns soqfyqouruy
“VIVMALCONIBOD

“sees 70) “eTNUOIOD vIOVqRAOIT
*VLAHAOOZ,

selec eer clare elereeeeleree COs ie irri) ‘ds ‘eruoydig
wale w ee eel awww eel ener neler err eecenee (sotoods [@t0.A08) erydsog
Bn ged Serine Ieee ici ine ii ii iri iii) sotoods ENOL’ A.

“VaaITaod

wale tense ewww rele een eel eter e eras Cg “eaeouon VUTTOFIQIC,

“VURTININV 40,7

— |—]{—\ — |q—— |] —_—|q— | jj] | qq] \qc— \ —|j—|\—

tH (P| AL

AYO

*TOMO'T

: “spoq
19}SULULIE AA =

PEN

yey | Pe “x0TyO “puesta. doddq

——$-$s&s"__+

‘speq ——
uMOopyor[g= :
eee
HS)

‘aIVI ST JI Jey} O pue ‘QuepuNge ssoy SI 4 yey} — ‘Guepunge st sotoods v yyy sojouep + *az0AT

‘wading Siyg ur panwunua sprssog ay) fo obumy pun uoynguysig oy; bumoyg—y] WIaV I,

387

BEER HEAD AND THE ADJACENT CLIFF-SECTIONS.

Ses ea)

Aopllsooodn lan ert ele we ee elanmenel| terre lasorcelane

we elew wee el ewan elev eneslesererlen wee

eed ees (or RS geod) odoin) agooicia iccicici i) isririciciry

Peewee rene el ewer sale ween alee ereelererenlesereal[stterelsreserlannnee

Oy ees iar? Pee ei in era ard

Beem elem ew elew eee el er ene aleeewenlae eee le tee eeleeeerele reece

ele eeccelecccerisceree|sceseslecenes

mere ale ree eels weet elas ens slr seeeelasereclssencales er eeleeweeelecenee

SO eo) ec ieiord ier) Pied Peed ieee ene ie (i ed I (Oa inary waee
OiCici (aries ei irae GCC cn (ere IC iad Ors eee Ie (Cg iad eee elewenreles ey eid era ary
Seis eid (ary ores

seer rlae eee

Si aC ee oe “mog *‘eoutvo BN yeiqedoy,
seeeeeles se ECOG! *ey1qood eyjoyeaqosa y,
see ewelee BEOGRAD IPOS IL Dat | “eq BI149

evetatevetel| (ere . *YQOTYIS! ‘sI[IoRaa vurNyBaqosa gy,
Wefetetwis}|(siejela(afeteleletulefejerule “far “eunTT el LISP
ietaretar eal Sie cere * -quo7ryas! *(4) IZ}IUTOX) SLORY
Reger OS OROAOOD “290M ‘stsuodoqvust
SOU PEMA SOLON g) (gf ‘(d) sisuouRuoUdD BIUBA)

“VaodOIHOVUg

ee cceeles ere ecccces (4) 0) 4 ‘canqnao wurad gy

cece lew eereccereescecee bE ‘TUTTI TAL —_——
lbuowg ‘oepolea BULOpRIpOpnosg

ntolefevofa||\sterevateversiptereleleiciaye $3910, *stA0q-d00 ee es
ee vcici|'s vec covevesuuvieetpcsen els ‘ds (TOYSVIOLAL

cece e|scecccccecce “QUO TT ‘snueyd Rest (0 |
nna “"saquouy ‘TTAOUOAL) soysnoudrute A
sevecefeccccuccece bp ‘snqepnunyt snzoydormoxy
APROOR EP eae CAG f ‘snaoyes-oqye SO}TMOT BL)
Tecan seal eoceasaiia "JUD ‘Be IUOStyoan JT Joyserdy
wieisleiele (sie) sais br ‘SITLoEytp snyd Aoouryqony
)

mrerateiote | erelatevetoleteretcte 3 (G A) ow ans (—=)
se eeeeleees wesw etre nsec reses ubuong
‘snoueysvo (sozT1o]849) snuo.oUTyORT

‘0G

ZL | OL | St | VT | €1

J. A. MEYER ON THE CRETACEOUS ROCKS OF

SOCOOg Igci oii ici teil icine irs feria

COCR EG (OCI IICIC ICICI ISiCiC iii gy (Ci iii Iii ere naar)

see eee

Bese eee eee lee eee ele renee ler esenlerrnes

DIOGO! IOOCGICIOn CCieicicn! (stir iii arir iy

“wnT “equinjpoo etksoxq,
sqaovnllonucoullodn caclloodagooodeadGaonbo ake sae ‘dg ‘eurjordeQ)
‘ds 6
‘ds ‘eimouy
‘VEaEIHONOD

“psuan® “Set AA ——
“amg “tyoequeoTqog

“nog ‘sttyeor~d ——
“Moy ‘eueljo;ue pA, ——
“agny, “eysnqot “ava “eyequaty ——
SC i ery 40 a “BUBISBL)
(‘wbuowrg “eutyes) «rea ‘
PYoToisleforellofefelefatele “nog “eqeIprurp pone
‘ubuong “t1a1AuG, ——
“mog “exeAuood vyjouoyoudyy
‘aq ‘s4190Ul, snUIEsOUOSIL],
sInjeis\sleieie|eie)6\s[eisieie/e\ela\eje)e\e/e “dg ‘CMIpTO9y I,
“mog “rea “eyepunqns
SOOnOOOOUrOr “Mog! “esoqopstures PES
SODOOOOAOOOHON OT quo yy “esoumenbs
“Loony ‘esoTnsna ——
Meee ee ee tee esses “noy “esoqo ea
“Moy “e{VAo vNyeIqorog,

>
‘od. ——,—— :

“dQ “o[ppHAL = “4oMmorT.
-———

ed

Se nF Be

C.

[IRIN
AWETO

ea
IoJSUIULIE A =

Le ‘t0TyO “pursuasig todd gq

388

*(panurquod) *T LIEV,

“speq ——
uMOpyovlg = :
‘qed

389

BEER HEAD AND THE ADJACENT CLIFF-SECTIONS.

o fre ch Opduljocoanolaoasal|anaced\on ayoiei)| sreteieisin fisfoseisisis!| aiataialaly |(ele!> eieln\||aisielalseleleralsieia “YIO TT ‘snuvisozng ates
ae | eects soo0a0 “(aA TpeUIs)

° al leintotetstsi| nietatetets\| (er sferetels)l(nfa\w}uie/ureis{s\sleisieters “Mog ‘sTaepNoTqIO
fete eeelee Gaon) odsdn scnincrnn eneecee UnT ‘snqesuoja

=e 44 me ttt elec e elev evnnlecereclanscceleccceel(ececes|sosens|eeensslscesseccene eee * -wOTT ‘zadse 90400,
tel atevets||[overalerare'| ejotereraret| (eve cue BOS REO IDLMEOD Cini lg ‘g1aB[NoIse A

5 Hocecddllaaoono|loccndollosoccollsnoocollosns Saolloudovol soonuo|loogenn|locqocoqKacn O00 UU ‘enerAnyIp ——
col osoone|scoadollaccasnllodoscclloodscnloc alevere'|ctererateve| fetefarievetniwrevete(a<reteroreye) “nog “eyeultvo

mae a Ronee © [renee ollantice | ladcodn||\S008o0|ldodooollsoaanollsooedallasoodalléa aYeietallefateyelere (ewermentto N09 pete) BversO
=~ zi |fezorat steel |{ntststa\eye\l{eluiat=s0isl ofetu elsts)|(stsleta{atn|{evstwreter=i| wratets\siei| fers syater™|letsturayers *qlO a ‘gn4e4s00-07¥114S

BEE JOoOdO) OOOO) OO edod lSnining! Incr OOddd ISSSnnnl Shonen Sricinnnl ian asco) souocn |adocodnoicon MO ‘SISUaTIOSI] eS]

WO. Wosteaony snAT |
Wislatevuye/eyeYela/albre "WO a ‘xoydunts

Ce On a rd m0 @ “eyeoTnstmes

ee t[eceeeelececeelerereeleeeeeelenecesleereerlereceelenserelecgeeene POS O GE “eyeurormes ——

a a Be | Frciewtal fae taselesceceles teed icveceleceneslevsese|esecee|scereslaaves ayia) efateretevereis ‘10 ‘eqeu10 ——
Sees oo Plone reel eee wena eee eelee ere rlennnsc(sraneelseeeneleversslsteerelsnnscvacereseseesssee (440.

‘eIpouldiojut ‘Ty 0} portye)

be a (OO nnn nein waleeevenlscereslereevaleesceelecrcceler seeeles DOO! NOOO nCnSoMSoncoomnncnricicricir iit ‘ds “euUT,

se eeeeleceres seer e ela ce rer lee ete eles eave lecenrelseeeselsversalseecetlesuaee (‘wow “eavye 0} peTye)

ae ae = Ey Ui |ictorsta intel taveteratnyal|(ajayorayarni|(ufalelsince||'clele(s' sie doodHo| aA0oOD||oooandlloudocdlacuceodun “nog “gyeysooonbumb

ab ah see emelvcccvelscerselecvccecce “nog “eyeysootapenb BALUB

Re UO IO ira ir “nog ‘snqvops

Ay fae 5 [soon spans al Re erasal eared eeaanelts Sonanl sochdlaaaoo) baduen asqoa aceres orcad deacon ragasa ndianodcodsESecs ‘mog ‘sepropyéu ——

— Jocosdalloaucdallécnacelloo5n6d|jaoqagcclloondodleanodalloooacolibaoacallac biuiete)tietetsteleteteta’atersieretsiie qupyr ‘sngvy

sees OOOO Cn a a i Cn i ge a ee ee ead Rabie "| Wiekenc| IAP RCROCRS RED USOMTON OT Taya ‘T9TAND

seelecereeleveeeslecsseelee tees ‘nog TqARIUsUOL, snuie.o000y

25 |locoosolfacens ye feletsialsrelefoterete(staye “mogy ‘eyepun

aelolarefataiatefcieie iar icieie ‘mog “eyeStacey ——

" Mog “eyeqyistp

= Bes occcoopuncpndon (‘ABA T[eUTs)

Pewee eens “mogy “eoTu0d ———

oes et ee ce Encode eocoéd baone 5 Beaenol sede al bobo |paosoa|(a dosopaopne “+ (awa Tpeuts)

snes tlre www el ewww eel eee ewe tees e slew ewe eleneeeelene reals erttleweneelestreel(estareleerennsleeeseelan

5
v]

2

Q.J.G.8. No. 119.

Cc. J. A. MEYER ON THE CRETACEOUS ROCKS OF

390

OS | 6T | ‘SE | ZT | OL | ST | PL | ST | SL | TT | OL

y= SS eee a eee
‘red ——\-—_— : 3 “spoq
-dq ‘e[ppual "IOMOTT, : TO4SUIULIB AA =
UL —. —,- — —~| “Ley
ALBYO MleYO | [eAL “AOTYO

SC neice (einen

cin DODO) (OOO OOICOOOICGIO CO OOCOTOOOnRIC tii (940 a
‘euerpuenbog ‘J, 0} por) etuostay,
ak eforefafelellistetatelefetsiaiereieletetstetersiexsiale nog ‘rofeut stjoU
es CR) Cay “moy ‘el 4ertys eure,
SE OOO) NSO On Onn itn icin ds “eIqoumuesg
atebelacies oagece ‘mog ‘snjvuoquin snjnounjoog
faisjeiajsis)|(=ielefelafel|taiarsiiele|Kats/scefste;[iatelersisis\| (¢inielelei=\|lelefa/aleininfelecelelefulsleleleteleteleteloletetetstateleiarelsle ds ‘sido
Se OOOO) CCCI tian ay “mog “ey eoUl]
Be OO) ear “moy ‘essord unt eynon NT
ah miovateists)| eivieletslalelslsietsterereys 5 nog “eyepn sue BIVOVIA
See es ei erie ie ee eee (aa ae "QO a ‘SISUSTMO.INY
+ wee teelececcercesece "M0K ‘SLIB[Norg.to eulon'ys
BEOEED| eo0 bab) Ju pac penne 6Qb00) eb old) BCE SU nore ajolafarsiehelshs ‘da “eusBo0198e4)
a. afofeteiele)| laletelelelotlelately/slotet “moy “eyeouno eurrd cg
aapedn|lonoc0dllaocond|Ja00500|lon0000| |sg0Da0| 400000 é QuDy GR osuattomsura sSeae

Sesion) Scgno! (Seicnon (Sccnre an praverel| etevetniare tes 00E CF TUNUROTTIC TAL
Beto CSc ISCnigen Stench Iscteticl Micron! Scricc ic irr rn (04 pee ‘ds)
ae o00G00)|nag00000000nq mog ‘TUNUeL A UINIpAey

ewww ele www e lew ene el reer eele rere elev eres lereuseacenes “QO a “euvoT ley
S" |leba004|lssuo56000a06500000 “Mog ‘eq BULIeo Bory

steep eee e eel ewe weer are el owen ee lerereeleceesese seers eeeeeseeseseesessacs ds (a te=
Piatataiste|| cie\eveterelt atafelereicil lsiovatetevel| (nieletaterel| etetercterel| italetcreretatevsintere mog ‘snsourds snjApuodg
BODDOD| /S00000) |SO0000| [OOO G00! |SO0000! Foou00 bonnonodG QO a ‘aneytmddry
585000] lao0an0llsoanaq|aonv00||ao00G0|lesd50nllan6000400000000 Wis) ‘BsOOTIUdA
BSendn! OOdSo! Sisdden! Secsnn Sond) SSnedol SOoenomnnccn “qUD AT ‘esourds emyeorg
tetoveteleliietatererets) | (ereteretarell eretateterell reretersrel| efetetetersll erevetererann 940 a ‘siIstosRmoyoyr m04090q,

uUMopyory_ = :

“+o!

“puesudes4) sodd qmeg od

*(panuyuod) "J XTaVy,

391

BEER HEAD AND THE ADJACENT CLIFF-SECTIONS.

°

seer welens woleessnelesenssleessse|esnvne

Sn eid a ei Oe I iinet)

RE OOO) OOO OOGOUGtt “moy ‘srauosAjod —————
nn (OOO Oo Coslaancreetererrneraasenvenne moy “eyetoquin
cs). | OSSD SOS "MOG “BABOTOD BLABTNOTUAAO A
BSS) Onn Cds Sciceicr otic! ic deveeccececereceecresesacsodly “eyoyIAn
Souncal eodcor jsdoaculoGo00d lsocced| PodunesGo WO ToS UeIOND oqmy,
wiuje(niove!| a/evure:niet| (eisvurntvis'| e/ats/=(yi«\]/erainierhfyi]/s\s(als's/uini=\n/mve(n\s'8\s(a(uu(a poo 44 ‘snaa1o
sreceefecascalereccefenecealereccelseeseentetteerecetstereernreenses ds ¢
moteteeet | ote ofe ere lBiere eve leaictelltercteiete | elesiate(sie vels{aieis'atelolle)efctsteretetfet= ‘ds ‘snqooay,
SOs ery ery econ) (Cc ir nic "my ‘cInyeuU.to TUMNIIelOG
Be 0g) CCC nono nnng “mos “THOSsuITyIeg
ser eeefeceeeri ee Orel een salen ereeteweeee *Q40 a ‘euBoT eT BIIBT[OISOYy
see neel even colecererleracenleserealerreernaeeeeserrenccnces ‘ds ‘981000109 7
RANG Std! (GGG) GGgong (Ging (Otto iii i ck Ir teseeseredg ‘
eleferersve]] einiereveyerttsvavareleis!| (areferatata| feieratevare||{eirisinis/els\uTu\etalsiate(a “moy’ Lie] ATgoods0d. —-

BRS Otgn) gedigd enn ise ininion! coogi aio CRT@) a “euRo] [BL

fice cr eofesaceetaseccelrcssvelscesselerce q4O a “BUBISISSBO BLIBULOJOINO,_

seeccslncccvafevasmeleeracsfecnscs|seeccscsseteesccsvessene ‘ds ‘stscoq ito Ny
woe ccalececcelrcaccsfecccccfecccsetacesceceererccevsccssevesanere ‘ds ‘gore N
seen celeccceolereceeleeecerlesesesler erases eeersaecrasaescnsssesenes coke, Go
ee Code Sonnoooonnonnnn “mog ‘sngeapenb snsniy
ececcefevcccelesercelesevcelenccee[sccesrececevccse #0 ‘sIssvo BULA
“VGOdOUALSV +)

fe on nog ‘es 1oULUUt snud A
rate tavats]|furetefevelciltetbiavece) wiitexetotevsie)] (sturenmimcelllevalsreveleiw/elntutnys UOT (a) eqenUus.to
sudnoallacoddalloudenol lbooodo loooond| foods0000d “punta ‘eqyvoTsmue4
seme celecncccfeveseelecccvelecscse[eseseseeseeesenece wT ‘Blieyeolns

a ee OOoeCC eco ei “nog ‘stjiqeqoods
BO DOOO otroon tocirocen tocop no loonoon |jpoocooDUboOon GOO OoCDOOM (yng

‘eyenuis “J, 0} pore) “ds ‘

= 5e odnoocadnancaoosduccodconpocenddoy ds ‘

SHucoa soaanollasodon| bacadnlloopscnllondaugooucoodoooaD yong ‘eqadnaqe ee os
E> |||-iolateloielllere'e eteivislolulsln)slelnlelele nsa\sivin(elelsisleiciv'e se dg é

Cc. J. A. MEYER ON THE CRETACEOUS ROCKS OF

392

Mawel ewes eels reer alesse celesscenlesseesleserer|eeeuvs(seursslsoecee

Tete wee el ee wea lesen eles eeelecereel|seeeerleneses|scereslseererierscee

ele e reset ens ceel see cerl ere neal sercari(sseeeri(ersnes

seer le ewe melee eww e lees eee lee sensi eeeslecneesleeeseslttesee

weet ewww eel ieee celer eres ecscesleceees|scescs|erervelsensesl|srsensl(terese

000000) DO0000) DO CO0 0) IGG0000) IGO000C |OU0000) INco000) IS U0 R00] 00000) |DOb000} |G 5umddl lagdu0b

Rall (Ott) GOGO ISIC IC CIOO! ISICICISI in! Sigs! Omigiciig! (SC irigioig Sorin! (einirrrins imiriciry

eset lw wesw eles ewe elec eee else ere eleeeceelecet ee lavensslesesseleeeersl(eerarelevete®

‘0G | 6 | SL | ZT | 9L | ST | PT | "ST | SL | TT | OT | AD ie AS 5 Ia
eee S| | ee — y
‘zo. ——~-—— : : *spoq ‘spoq ~~"
-dq “°[Ppul “TOMOTT : TOISUIULTE AA = uwMopyorlg = :
—_— Oo | OW ; Se
Pike) MPG | Mert “t0TYyO : | ‘puvsusety todd gq “qeg d

meee ete nese ee

“SaOSTG

¢ BsseuRlyeg
‘VEOVLSAUO

jjeeponenenae 080g ‘sngemoseqny <=

s0t5d0G0G00G00060 ‘UDT, ‘S1Y84809
nopogD0008000004 mog Tayooq SOql[Itin, J,
SobaDuADGOOSnOO0eD “mog ‘snnbitqo
sje Mog ‘sipenbes seqiydeog
sondpodcne00006 HO ‘SNYBSIART
setae "QO. ‘snwenwwrpypiscey

sdvoo0uen0dd adunyy “T0441 snpryne NT
conavsuaounnngnDDoOeGua0on eppoyTUMLTO gy

sfofotslsJefefeleleteieverelsheisistetele nog ‘SURLIBA
Pain an “og ‘sIsuaseUOJOYA
siefoiefelele(sleisisteiatele “mog ‘SLIBNOTARUL
SOOOOUDOROOODOOOOG “Mog TTOJUR PL ———
so0000 ‘ubuorg tadnop seytuowuly
“VYacdOTVHda)

*(panunwoo) *T ITAV,

BEER HEAD AND THE ADJACENT CLIFF-SECTIONS. 393

Taste II.—Tllustrating the Succession of the Cretaceous Strata near
Beer Head, Devon.

3 Za
eo a
E EEE
Divisions. | + Minor Divisions. S 3 Zones of Fossils.
é =
=
Upper ft.
(P) 20. | White chalk, with flints ...... 70 | Micraster cor-anguinum.
S Terebratula carnea.
a} 19. | Chalk, without flints............ 15 | Inoceramus Cuvieri.
= Terebratula semiglobosa.
| 18.| Hard chalk, with flints in |40 | Inoceramus Brongniarti.
: layers. Holaster planus.
= 17.| Marly chalk. Flints finger- |10 | Inoceramus mytiloides.
S Pe like, dispersed. Micraster cor-bovis.
8 16. Nodular sandy chalk, without | 14 | Rhynchonella Cuvieri.
flints ; semicrystalline.
15. | Cream-coloured sandy chalk; | 10 | Inoceramus mytiloides.
L semicrystalline.
si (| 14.| Hard nodular chalk, with| 5 | Discoidea subuculus.
S quartz grains.
fs 1 13.) Greyish nodular chalk, with | 5 | Ammonites Mantelli.
3 quartz grains, green grains, Discoidea cylindrica.
GH and Phosphatic Nodules. Rhynchonella Mantelliana.
( 12.| Compact nodular bed, with Ammonites rhotomagensis.
= green grains and quartz Scaphites eequalis.
ra grains in a chalky paste. Holaster subglobosus.
S a Catopygus carinatus.
S 8 11. | Yellowish quartzose grit in a Rhynchonella dimidiata.
Pea 1 chalky paste. Terebratula ete ios
oe Exogyra columba (small var.).
a a 10.| Rubbly, yellowish, quartzose Siphoria.
= | grit in a chalky paste. Nautilus levigatus.
lL Discoidea subuculus.
(| 9.| Buff-coloured sandstone and
grit-beds, with fine green
grains.
: 8. | Buff-coloured sands ............ Orbitolina concava.
S | Exogyra digitata, Sow.
S 7.| Shingle-bed of sandstone-
8 | pebbles. ee
= 4| 6.| Light-coloured sand and/| 23) Exogyra digitata, Sow.
= | sandstone.
a 5. | Chert-beds, in light-coloured Exogyra columba.
2 sand. Janira quadricostata.
= 4,.| Yellowish marly sands, with Exogyra columba.
eee irregular concretions. Janira quadricostata.
s a | Pecten orbicularis.
aS || 3.| Greyish marly sand, with | 30 | Exogyra conica.
S72 nodular concretions. a levigata.
+ g (| 2.| Sands, with green earth (ar- | 25 | Blackdown fauna.
3 Se gillaceous). Tnoceramus sulcatus, Sow.
io) | 1 sue) sands and grit- | 34} Exogyra conica (small var.).
? 4 bed.

BEER HEAD AND THE ADJACENT CLIFF-SECTIONS.

393

Taste I]1.—Zllustrating the Succession of the Cretaceous Strata near

: Sa
Divisions. | +3 Minor Divisions. S =
3 ao
A +
Upper ft.
(?) | 20.| White chalk, with flints ...... 70
E
s 19. | Chalk, without flints............ 15
a
P Cr 18.| Hard chalk, with flints in | 40
4 layers.
3 17.| Marly chalk. Flints finger- | 10
= like, dispersed.
oS 7-_-__—
8 4 16. | Nodular sandy chalk, without | 14:
flints ; semicrystalline.
15. | Cream-coloured sandy chalk ; | 10
U semicrystalline.
Z 14.| Hard nodular chalk, with! 5
S quartz grains.
as 1 13. | Greyish nodular chalk, with | 5
a quartz grains, green grains,
Sie At and Phosphatic Nodules.
i ( 12.| Compact nodular bed, with} 2
a green grains and quartz
= 3 grains in a chalky paste.
a g 11. | Yellowish quartzose grit ina| 3
P=a| 1 chalky paste.
Sa
& S 10. | Rubbly, yellowish, quartzose | 15
= grit in a chalky paste.
ay, (| 9.| Buff-coloured sandstone and | 20
grit-beds, with fine green
grains.
‘ 8. | Buff-coloured sands ............ 10
se}
Es 7.| Shingle-bed of sandstone-| 2
3 | pebbles.
= 4| 6.| Light-coloured sand and| 23
o sandstone.
ol 5. | Chert-beds, in light-coloured | 20
a sand.
= 4. | Yellowish marly sands, with | 20
5 irregular concretions.
sien
a 3 3.| Greyish marly sand, with | 30
=e nodular concretions.
+ & | (| 2.| Sands, with green earth (ar- | 25
2 Sant gillaceous).
oS || 1.| Greenish-grey sands and grit-| 33
? bed.

Beer Head, Devon.

Oru Gas:

No. 120.

Zones of Fossils.

Micraster cor-anguinum.

Terebratula carnea.
Tnoceramus Cuvieri.
Terebratula semiglobosa.
Inoceramus Brongniarti.
Holaster planus.
Tnoceramus mytiloides.
Micraster cor-bovis.

Rhynchonella Cuvieri.

Tnoceramus mytiloides.

Discoidea subuculus.

Ammonites Mantelli.
Discoidea cylindrica.
Rhynchonella Mantelliana.

Ammonites rhotomagensis.
Scaphites sequalis.

Holaster subglobosus.
Catopygus carinatus.
Rhynchonella dimidiata.
Terebratula pectita.

Exogyra columba (small var.).
Siphonia.

Nautilus leevigatus.

Discoidea subuculus.

Orbitolina concava.
Exogyra digitata, Sov.

Exogyra digitata; Sov.

Hxogyra columba.
Janira quadricostata.
Exogyra columba.
Janira quadricostata.
Pecten orbicularis.
Exogyra conica.

» levigata.
Blackdown fauna.
Inoceramus sulcatus, Sow.
Hxogyra conica (small var.).

QF

394 J. G. GOODCHILD ON THE CARBONIFEROUS CONGLOMERATES

30. Nots on the CARBONIFEROUS ConGLOMERATES of the HASTERN Part
of the Basty of the Epun. By J. G. Goopcnip, Esq., H.M. Geo-
logical Survey of England and Wales. (Read March 25, 1874.)

(Communicated by H. W. Bristow, Esq., F.R.S., F.G.S., by permission of the
Director-General of the Geological Surveys of the United Kingdom.)

Prorsssor Puriiies has pointed out, in his ‘Geology of Yorkshire,’
that the Mountain-Limestone between the northern border of the
Lake-district and the river Eden includes a group of red sandstones,
alternating with limestones in the lower part of the series.

They were referred to by him as the “Alternating Limestone and
Red Sandstone Series,” and were considered to be the passage-beds be-
tween the true basement series, or Upper Old Red, and the principal
mass of the Mountain-Limestone which forms the comparatively high
ground between the basins of the rivers Lune and Kden.

This middle group is fairly exhibited about three miles to the
south of Kirkby Stephen, at Ash Fell, where the different members
of the whole series may be examined in the quarries and natural —
sections laid bare in the immediate neighbourhood (fig. 1).

The general sequence in descending order is as follows :—

‘¢q.” Principal mass of the Carboniferous Limestone, including a
few thin beds of stained sandstone and shale between thick masses
of limestone, and having a total thickness of not less than 1000 feet,
between Ash Fell and Kirkby Stephen. :

««b.”” Obliquely laminated, soft, red sandstones, here and there
very calcareous, generally containing many traces of Coal-measure
plants, and frequently assuming a conglomeratic character in conse-
quence of the presence of a few pebbles of milky quartz, and the
occurrence of lenticular fragments of now decomposed shale on some
of the faces of bedding.

These beds alternate with thin shales, which are usually very full
of fossils where calcareous, and with beds of limestone of variable
thickness and different degrees of purity.

Near the top of the series these limestones are usually encrinital —
and crystalline; others are rather earthy and have interlaminated
sandy bands; while those nearer the base, although generally pure
in the mass, have courses of quartz pebbles up to an inch in diameter
on some of the bedding planes.

Here and there amongst the quartz pebbles may be detected pieces
of pinkish felspar or fragments of slaty rock, probably of Silurian
origin, which suggest by their usually subangular character that the
associated quartz was rolled into pebbles ere it began to be drifted
into its present position.

The total thickness of this series at Ash Fell is about 500 feet ;
but the interbedded limestones thin rapidly in a north-westerly
direction, so that near Shap the total thickness cannot be much
more than 300 feet, which is mostly made up of sandstones more or

OF THE EASTERN PART OF THE BASIN OF THE EDEN,

8. River Lune, Ash Fe Roman Fell. N.
o '

Fig. 1.—Diagram Section from the Howgill Fells to Roman Fell.

See=.8| Upper —_— p
peel old Rea, =e nee

5s] Carboniferous

eeSestes) P, i
et ehry ermian.
z-2cI Sandstones. Sea

Fig. 2.— Diagram Section along the Oross-Fell Escarpment.
N.W. Crors Fell. Roman Fell. Brough. §.E.

= !
2 = ————— = i —.— 3 WZ iim Wea r = —
Cra eH Seal" mH tf nr =m = 5 ben i} Hy HI

—————— == fy

Seo

Siete
ST
= SS

396 J. G. GOODCHILD ON THE CARBONIFEROUS CONGLOMERATES

less conglomeratic, and subordinate beds of limestone, which are
similar in lithological character to those in the Ash-Fell sections.

Some sandstones about the middle of this series may be seen in
the railway-cutting in the village of Shap.

“¢,.” The Sandstone series is succeeded by a variable thickness of
limestone, which cannot be less than 500 or 600 feet near Ash Fell.

The beds are not usually so pure as those of the upper series ; but
this lower group is nowhere split up by beds of sandstone or shale.
Near the base it becomes thinner-bedded and passes down by almost
insensible gradations into the next series.

Slee This group in its upper part consists of shales, with thin
impure limestones here and there containing grains of quartz, and
passes down through calcareous beds of a more decidedly conglo-
meratic character into a series of apple-green quartz conglomerates
and chocolate and grey shales.

These, in their turn, are succeeded without any clear line of de-
marcation by the Drift-like series of red conglomerates, sandstones,
and shales which forms the lower part of the Carboniferous base-
ment-beds here, and has always been considered the equivalent of
the Upper Old Red of other parts of the kingdom.

This basement-series is exceedingly variable in thickness and
mineral character; and, so far as it has yet been proved, it seems to
be mainly composed of locally derived materials.

These are the beds which are seen at intervals along the northern
border of the Lake-district between the Silurian rocks and the lime-
stone scars of the Carboniferous series. They are very well exposed
in Birk Beck, between Tebay and Shap Wells.

Between Poola Bridge, at the foot of Ullswater, and the village of
Dacre are thick masses of roughly stratified conglomerates belonging
to this basement-series. Itis these beds which form the striking
conical hills known as Great Mell Fell and Little Mell Fell, as well
as others of the same outline between the foot of Ullswater and the
outcrop of the Carboniferous Limestone to the north.

A thickness of several hundred feet of limestone belonging to
group “‘¢” comes on above the conglomerates in the river-Hamont
sections north of Ullswater; and it is not until we reach Yanwath,
about two miles to the south of Penrith, that we find the Ash-Fell
beds in full force.

Groups “6,” “c,” and “d’’ form much of the valley-ground
skirting the north-eastern border of the Lake-district; and the
generally north-easterly dip of the beds causes these lower groups
to pass beneath the higher members of the Carboniferous series and
the Permian rocks up to the great lines of disturbance at the foot of
the Pennine escarpment, where they are again thrown up to the
surface by faults which have several thousand feet of downthrow
on the south side.

About a mile and a half to the east of Ash Fell a north-easterly
line of disturbance cuts off the Ash-Fell beds on the east, and throws
in limestones belonging to the series “a.”

The strike of the beds on the western side of the faults continues

OF THE EASTERN PART OF THE BASIN OF THE EDEN. 397

nearly parallel to the lines of disturbance from near Ash Fell, north-
ward along the foot of Stainmoor, to the Cross-Fell escarpment (fig. 2,
p- 895); so that limestones belonging to the higher part of “a” are
not lost sight of for any great distance, but may be traced round to the
perfectly clear section in the scars above Brough, where the principal
bed of this series is recognized as the Melmerby-Scar Limestone.

This is the bed which forms, with the Whin Sill, the more striking
and picturesque features of the Carboniferous rocks on the upcast
side of the Pennine faults,

A little to the west of Brough the generally north-easterly dip of
the beds on the north side of the lines of disturbance brings up from
beneath the Melmerby-Scar Limestone a series of sandstones of
various degrees of coarseness interbedded with shales and beds of
impure limestone.

These are clearly seen to overlie a group of conglomerates, gan-
nister-like sandstones with Coal-measure plants, and subordinate
beds of sandy limestone containing many courses of well-rolled
pebbles of quartz.

This conglomeratic series is succeeded by a mass of grey lime-
stone, the thickness of which cannot be made out in the typical
sections.

These beds, from the Melmerby-Scar Limestone downwards, cor-
respond so closely in lithological character and geological position with
the beds seen south of Kirkby Stephen that there is little room left
for doubt as to their identity.

The sandstone and impure limestone series next below the Mel-
merby-Scar Limestone probably represents the lower part of group
“a,” which was described as being split up by thin beds of sand-
stone and shale.

It must be admitted that the interbedded limestones in this sand-
stone group in the escarpment are rarely found to be thicker than
25 or 30 feet, while those on what is supposed to be the same
horizon south of Kirkby Stephen have a thickness of from 100 to
150 feet between the partings of sandstone.

As there is a distance of a little more than eight miles between
the two lines of outcrop, few geologists accustomed to follow indivi-
dual beds over long distances will feel inclined to lay much stress on
the difference in thickness of the limestones in the two sections,
when the thinning can be shown to be at the rate of only 15 feet
ina mile. As an additional argument in favour of the identity, at
least in part, of the escarpment-sandstone and thin limestone series
with the limestone group “a” south of Kirkby Stephen, it may be
mentioned that the thin beds of sandstone observed in the limestone
series “a” north of Ash Fell are found to thicken and the asso-
ciated beds of limestone to diminish in thickness, and become still
further split up by sandstones and shales, as they are followed in a
north-westerly direction towards Penrith.

If it were possible to follow these beds under the Permians still
further to the north-west, we should probably find that the lime-
stones had become quite subordinate to the sandstone series, and

398 J. G. GOODCHILD. ON THE CARBONIFEROCUS CONGLOMERATES

that there would be little, if any, noticeable difference, except that
of colour, between these beds and the rocks below the Melmerby-
Scar limestone at the eastern end of the Cross-Fell escarpment.

The conglomerate series next in order, notwithstanding its super-
ficial resemblance in lithological character to the conglomerates of
the basement-series, as seen at the foot of Ullswater, is distinguished
from them by its including thin beds of limestone; and we should
probably not err in referring this part of the series to beds about the
horizon of the Ash-Fell beds “0b.”

The mass of grey limestone on which the conglomerates lie occu-
pies the relative position of the limestone group “c,” beneath the
Ash-Fell beds, and probably is wholly or in part identical with it.

The base of the series is not clearly seen in the typical section in
the escarpment ; but, as in an adjoining section the conglomerates
are clearly seen to be at no great distance from the floor of Silurian
rocks upon which the Carboniferous beds lie, there cannot be any
great thickness of this lower limestone.

Perhaps the only other distinction of any note between the beds
seen in the escarpment and those along their southern outcrop about
Ash Fell lies in the fact that the very well-marked red tint which
is seen in the Ash-Fell sandstones is nearly or quite absent in the
beds under the Melmerby-Scar limestone.

But as other sandstones higher in the series, which are red on the
downthrown side of the great faults, are clearly seen to be of their
natural colour on the upthrown side, it is inferred that the staining
is due to Permian influence; and the mere absence of the red tint is
therefore not a distinction of any value.

In my opinion this marked absence of staining in the Carboniferous
beds on the escarpment sides of the great lines of disturbance is due
to a considerable upheaval in late geological times, the plain upon
which the Permian rocks once rested being now only in part repre-
sented by the tops of the highest fells on the north side of the
Pennine faults.

Perhaps there is no place along the Cross-Fell escarpment where
this group of sandstones and conglomerates can be so well examined,
and its relation to the beds above and below so clearly made out, as
on Roman Fell and in the sears to the south-east of it.

Along other parts of the escarpment, much of the lower slopes
along which these beds mostly occur is obscured by drift and by
fallen blocks from the scars above.

It will therefore be convenient to speak of the whole series be-
tween the base of the Melmerby-Scar limestone and the top of the
lowest thick limestone as the Roman-Fell beds.

This group will therefore include not only the supposed repre-
sentatives of the Ash-Fell beds “ 6,” but also the lower part of the
Limestone series, indicated by the letter ‘“‘a” in the diagram section
at Ash Fell (fig. 1).

In following the outcrop of the Roman-Fell beds towards Cum-
berland, the conglomeratic beds from the middle of the group down-
wards are found to increase much in thickness, and the included

OF THE EASTERN PART OF THE BASIN OF THE EDEN. 399

stones to become so much larger that there is little difficulty in
finding masses of quartz nine inches or a foot in diameter in some
of the coarser beds.

The upper part of the group, between the conglomerates and the
Melmerby-Scar limestone, does not seem so changeable: except that
the interbedded limestones are thinner and usually of a lesser degree
of purity, and that coals set in to the north, there is but little differ-
ence between these beds under Cross Fell and their eastward exten-
sion as seen near Brough.

As has already been remarked of some of the limestones associated
with the Ash-Fell beds on the south side of the great faults, the
beds above and below the Roman-Fell series become much thinner
in a north-westerly direction, so that under Cross Fell the Melmerby-
Scar limestone is little more than half as thick as the same bed at
Brough ; and there is reason for believing that the lower limestone,
which is seen under the Roman-Fell beds near Brough, dies away
completely before we reach Cumberland.

The exact point where it does so cannot be indicated until a more
detailed examination of the ground has. been made.

So far as the writer has been able to make out in his holiday
rambles, there does not appear to be any undoubted equivalent of
the true Carboniferous basement-series along the escarpment,
although it is possible that those beds may be feebly represented in
places by the lower part of the conglomerates.

The tendency to become coarser in a north-westerly direction is
not so marked along the southern outcrop of the Ash-Fell beds, so
far as they have yet been traced into the basin of the Eden; but at
a short distance west of Penrith some sandstone beds, which must
be very near this horizon, are quite as coarse as any millstone grit,
and in themselves could not be distinguished from it.

Over a considerable area to the east and north of the Silurian
rocks of the Lake-district one of the results of the Geological Survey
has been to prove that, as a rule, the limestones of the Carboniferous
series increase in thickness towards the south-east, and that, on the
other hand, the purely drifted deposits come on in greater force, and
generally tend to increase in coarseness, in proportion as they are
followed towards the north-west.

As this rule can thus be shown to be one of wide application in the
higher parts of the series, one may infer that these lower beds form
no exception, but that the whole of the lower part of the Carbonife-
rous series tends to assume more and more of a littoral character as
it is followed towards the north-west, and that any of the limestones
which can be shown to have died away along the escarpment will
not be found to reappear along the northern outcrops of the local
base of the carboniferous rocks, but that these Roman-Fell beds
over large areas will be found to rest directly upon the older rocks
wherever there are no intervening patches of ‘“‘ Old Red.”

We should perhaps not be far wrong in considering that these Roman
Fell beds are about the horizon of the Calciferous Sandstone series of
the south of Scotland. If one may judge by the published descrip-

400 CARBONIFEROUS CONGLOMERATES OF THE EDEN BASIN.

tions, there is a close agreement between them both in lithological
character and in their position in the Carboniferous series*.

From what has been stated it would therefore appear that certain
beds, which can be shown to occur about the middle of the Carboni-
ferous Limestone in one part of Westmoreland, develop into coarse
conglomerates, which are locally undistinguishable from much of the
true basement series, and have been described, by authors who have
not had opportunities of tracing out the true position of these beds
in the Carboniferous series, as undoubted Old Red Sandstone.

Discussion.

Prof. Hueuxs confirmed the observations of Mr. Goodchild, and
showed their importance as bearing upon inquiries into the changes
of the land which furnished, and of the currents which arranged,
the materials of the beds described.

Prof. Ramsay remarked that the author was entitled to much praise
for the manner in which he had worked out the minor details of the
Carboniferous system. The Cross-Fell conglomerates had been called
Old Red Sandstone ; they are now placed in the Lower Carboniferous,
but no distinct line of demarcation can be determined. Prof. Ramsay
remarked on the distribution of deposits of the Carboniferous series
in Britain, and referred especially to the Carboniferous Limestone.
In the south of Pembroke and Glamorganshire this formation is
2500 or 3000 feet thick; in the Devonshire area 100 feet, and in
the north of Glamorganshire 500-600 feet represent the whole of
the Mountain-Limestone ; in Coalbrook Dale, again, its thickness is
only about 100 feet, in North Wales 800 feet, in Anglesea 500 feet.
In advancing towards Paleozoic districts, where old land existed,
the limestone becomes thin, whilst it becomes thick in the deep-
water areas,—in Derbyshire 3000-4000 feet, falling off to less than
1006 feet in Cumberland, where it is split up by sandstones; and
this is still more strikingly the case in Scotland, where 100 feet of
limestone is a rarity. He thought the existence of coral reefs in
the Mountain-Limestone doubtful. fi

Mr. Trppemawn remarked that the Mountain-Limestone was 3000
or 4000 feet thick in the Lake-district, and in the Pennine Chain,
at a distance of 15 miles, only 600 feet.

* The writer wishes it to be stated that he had independently arrived at this
conclusion before he heard. that Mr. James Geikie held similar views regarding
these Roman-Fell beds.

W. WHITAKER ON THANET BEDS AND CRAG AT SUDBURY. 401

31. On the OccurRrENcE of Tuaner Beps and of Crag at Supsury,
Surrorx. By Witi1am Wuiraxrer, B.A. (Lond.), F.G.S8., of the
Geological Survey of England. (Read June 10, 1874.)

(Communicated by permission of the Director-General of the Geological Surveys
of the United Kingdom.)

1. Tuanet Bens.

Tue newer Tertiary deposits of the Eastern Counties have had such:
a power of fascination over geologists that the older beds have been
left almost unnoticed: indeed since the description of a few sections
in Mr. Prestwich’s essays on the Lower London Tertiaries, which
have now been published 20 years and more, I know of but one
paper (and this probably known to very few geologists)* that gives
us any information about that series in the district in question,
whereas the number of papers that treat of Crag and Drift is
legion. It has therefore been an agreeable surprise to me to find,
in the course of my Geological Survey work, that there are fine
sections of the older Tertiary beds in Suffolk and the neighbouring
part of Essex ; and it was with some astonishment that I saw one
of the finest sets of sections in the London Basin round the town of
Sudbury, on the border of those two counties.

These sections convinced me that the lowest member of the Lower
London Tertiaries, the Thanet Beds, crops out along part of the
northern edge of the Tertiary district, an occurrence of which I had
a very slight suspicion from other sections near Ipswich.

Mr. Prestwich has, indeed, doubtfully referred some of the sands
of the northern outcrop to this division, though without describing
any section; but he does not notice the particular part where it is
now to be seen in many large and clear pits, and the following notes
are all that can be gleaned from his elaborate papers :—

‘In north Essex the zone of outcrop of these sands usually occurs
on the slope of hills, and they therefore form a very narrow belt,
which is further frequently so obscured by drift that they do not
constitute any marked feature in the district. Owing to this cause
and the want of sections, their structure there remains uncertain.
Their thickness may be from 30 to 50 feet” +; and in a foot-note is
added the remark that the Thanet Sands, if present, probably com-
mence (on the west) somewhere near Bishop’s Stortford. Again,
in a later paper, “I am doubtful whether the Thanet Sands range
thus far north. If they do, they must be represented by the bed of
sand which is to be seen reposing on the Chalk in a few pits be-
tween Bishop Stortford and Newport” t.

Since the above was written a great improvement in the matter
of sections has taken place, so that I have had the advantage of

* J. B. Phear, “ On the Geology of some parts of Suffolk, particularly of the
Valley of the Gipping,” Trans. Camb. Phil. Soc. ix. p. 431 (1856).
+ Quart. Journ. Geol. Soc. viii. p. 241 (1852). { Lbid. vol. x. p. 92 (1854),

402 W. WHITAKER ON THE OCCURRENCE OF

further and better evidence. The sands above referred to must in
great part belong to the Reading Beds, as there is no thickness of
even 20 feet of Thanet Sand to be seen in North Essex ; but I think
it can be shown from an examination of the pits at Sudbury that
there is really an outcrop of Thanet Sand—though I believe it does
not reach so far westward as Stortford, unless the bed of clayey
greensand that comes on above the Chalk there should belong to
this division of the Lower London Tertiaries.

Details of the various sections will be given in the proper place
(a Geological-Survey Memoir) ; but it has been thought better to lay
-a statement of the facts (trifling though they be) before the Society
than to keep them hidden away for perhaps some years. It will be

Section of the Lower Tertiary Beds at the Great
Chalk-pit, Balingdon. 1873.

(Scale 8 feet to an inch.)
feet.

a. London Clay, brown and dark grey, with ‘‘race.”
No basement-bed.

(A ——

b. Blackish, rather sandy clay, and
clayeyasand ep rcenccenmr neste ee sec eae

Reading

Bede: | c. Pale greenish-grey, red-mottled, bedded
( sandy clay, the upper part redder... 6

(d. Buff and pale grey, fine, soft, and
somewhat clayey sand, bedded and
firm. The bottom 3 feet or so (d')
pinkish, that tint being stronger at
the lowest half foot (d”)...... about 12

|

: | e. Clayey greensand, very bright-co-

: aa ek 4 loured at top, with 2 few green-

eee eae ; coated flints at the bottom, and here

Oh penser eee and there some higher up. This

: bed is noticed by W. Smith, in

his ‘ Geological Map of Hssex’

(1820), as ‘“ Greensand, called

| Devil’s Dung,” a name still kept in
\ theslocalittypmemsascdeceaiseccs about 2

q" = ee = f. Chalk. A thin bed of tabular flint (of alter-
= == =P nate black and white layers) at top (.).
== Other flints rare (?=the ‘‘ Margate Chalk” of
= Kent). The top flint bed is noticed by W.
TaN {\ \ vn Smith, in his ‘Geological Map of Suffolk’
Wy VAIN i) (1819) thus :—“ Chalk, covered with a Floor
enough here to notice the one great section where all the beds from
the London Clay to the Chalk may be seen. ‘This is in a pit on the
right slope of the valley of the Stour, just south of Balingdon, the
Essex suburb of Sudbury, which is worked down to the level of the

of Stratified Flint.”

THANET BEDS AND CRAG AT SUDBURY, SUFFOLK. 403

river, laying bare the set of beds shown in the figure, drawn on the
same scale as various analogous figures in the Memoir on the
London Basin*.

No fossils have been found in the beds that I have classed as
Thanet Sand, except some pieces of Foraminifera, by Dr. Holden;
the reasons for this classification are purely lithological, and as
follows :—

1. The likeness of the sand, in fineness, compactness, and colour
to that which occurs in the same position in West Kent.

2. The persistence of these characters in all the sections, and the
absence of false-bedding of pebbles and of red-mottling, tending to
show that the sand belongs to the Thanet Beds rather than to the
more varying Reading Series.

3. The fact that the greensand at the bottom is like the “ base-
bed ” of the Thanet Sand; and the occurrence immediately beneath
it of a layer of tabular flint, as is very usual where that bed caps
the Chalk.

Westward from Sudbury I have traced this sand for more than
three miles; but eastward it is soon hidden by Drift, after being
shown in many sections on the outskirts of the town. The green-
sand over the Chalk, near Ipswich, is much like that at Sudbury ;
so perhaps the progress of the Geological Survey may establish the »
occurrence of the Thanet Beds in that direction.

In the northern band here noticed this division seems to be pre-
sent at the expense of the overlying Reading Beds, which are re-
markably thin at Sudbury; and this outcrop most likely joins on
underground to that in South Essex and Kent, many of the deep wells
in the tract between seeming to pass through the Thanet Sand.

2. Crag.

The pits that give evidence of the occurrence of Crag have been
opened only during the last few years, and are all on the left, or
Suffolk, side of the Stour, at the northern and eastern edge of Sud-
bury, the three best sections being close together, on the hill within
a quarter of a mile N.E. of St. Peter’s Church.

The general section of these is as follows :—

A small patch of Drift.

Crag. Ferruginous dark reddish-brown sand, with layers of ironstone, slightly
false-bedded; in parts a light-coloured grit with broken shells;
thin layers of flint-pebbles, phosphatic nodules, and phosphatized
bones in the lower part; and at the bottom a marked bed of the same,
up to a foot or more thick. The whole as much as 10 feet in a hol-
low, elsewhere as much as 7 feet.

Thanet Sand; the lower pinkish bed, and the green base-bed.

Chalk.

In the furthest pit, Mr. Webb’s, the shells are mostly broken up,
and many specimens of Purpura lapillus, var. crispata, with one
valve of a small Anomia ephippium, were all that could be found in
any thing like a perfect state ; but in the nearest pit, Mr. Harding’s,

* Memoirs of the Geological Survey, iv. (1872).

404 W. WHITAKER ON THE OCCURRENCE OF

all the fossils were in the state of casts and impressions in one of
the layers of ironstone, which yielded the following :—

Nassa elongata. Mya arenaria.
Natica occlusa. Mytilus edulis.
Trophon antiquus (lower whorls)- Pecten opercularis.
Cardium Parkinsoni. Tellina.

Mactra. Pollicipes.
Modiola modiolus. Balanophyllia.

From their unsatisfactory state these fossils are of little value in
deciding the age of the bed; indeed Mr. Etheridge assures me that
it is hard in most cases to assign the specific names with any cer-
tainty, and consequently it is open to suggest that the bed might be
a fossiliferous member of the Glacial Drift. The reasons, however,
for referring it to the Crag are as follows :—

1. That no fossiliferous Glacial Drift is known within a very
great distance, much greater than the distance of the Crag tract (on
the east).

2. That no layer of phosphatic nodules is known to oceur at the
base of the Drift, whilst the presence of such a layer at the base of
the Crag is very usual.

3. That the bed is unlike the well-developed Drift of the neigh-
+ bourhood, whilst it is like some of the less fossiliferous beds of the
Red Crag.

As might be expected, this small patch belongs to the upper and
wider-spread division, the Red Crag *, and not to the lower and rarer
White (or Coralline) Crag. Its chief interest consists in the dist-
ance from any yet known masses, it being some miles westward of
all other exposures. Perhaps Crag may yet be found still further
westward ; for my colleague Mr. Penning showed me a section near
Stoke (west of Clare) where, between a mass of Drift sand and the
Chalk, was a bed of angular flints and flint-pebbles, in which I
found one very small piece of phosphatized bone, and on the top of
which we were told.a mass of shells had once been found. We
have since seen some of these, which turn out to be Purpura lapillus,
the common Sudbury shell.

The level of the Crag is, I believe, far higher at Sudbury than in
any other part of the Eastern Counties. I fear that the layer of
phosphatic nodules at the bottom is here too thin, too irregular, and
too much mixed with flint-pebbles to be of economical value.

P.S.—Since the above was written Mr, Penning and myself have
found traces of the bottom bed of the Crag southwards of Thaxted, in
Essex, in the shape of a few small phosphatic nodules and a piece of
phosphatized bone in a very thin layer of gravel at the bottom of
Drift sand and lying on London Clay. [April 1874.]

A few days after this paper was read [June], when with my
colleague Mr. Bennett, I found Thanet Sand a little north of Had-
leigh, and south of that town some nodules of ironstone in sand
(resting on London Clay), which were crowded with casts of shells,

* Mr. S. V. Wood, Jun., thinks it most likely to. belong to the higher part
(Chillesford beds) of this division.

THANET BEDS AND CRAG AT SUDBURY, SUFFOLK. 405

the most common being Cardium angustatum, a species hitherto
found in Crag only, as I am informed by Mr. 8S. V. Wood, who has
kindly examined my Sudbury fossils, which he regards as belonging
to the Crag rather than to the Drift.

DiscussIon.

Mr. Prestwich said that he was quite prepared to accept Mr.
Whitaker’s interpretation of the beds referred to the Thanet Sand.

Rev. Mr. Trurns remarked that the presence of ironstone did not
prove the beds in which it occurs to belong to the Red Crag.

Mr. Gwyn Jzrrreys said that the limits of the Red Crag were
not easy to determine, and that the casts of shells obtained in the
beds near Sudbury were not sufficient to prove the exact position to
which they should be referred, although they represented Crag
Species.

“ite Sretry stated that he had seen fine sections of Thanet Sands
fully 20 feet thick at Hadleigh sixteen years ago. They contained
sharks’ teeth. The sands were capped with pebbles, and above
these with London Clay. The Reading Beds either almost thin out
to the north or are changed to sands at Hadleigh. This would give
the Thanet Sands a more northern extension.

Mr. CHaRLEswortH inquired as to the number of species that
could be identified with true Red-Crag fossils. He thought that
the presence of phosphatic nodules was confirmatory of the beds
belonging to either the Red or the Mammaliferous Crag.

Mr. Gopwin-Austen remarked that the occurrence of Thanet
Sands at Sudbury was a fact of much geological interest.

Prof. Ramsay considered that freshwater conditions generally
succeed marine, and that the submergence of the Chalk areas was
followed by upheaval of land.

Prof. Hueues thought that the base of the Thanet Sands could
not represent a land-surface, but that there had been land in the
neighbourhood, which gave rise to fluviatile conditions.

Mr. Wuiraker, in reply, said that the beds at Hadleigh, referred
to by Mr. Seeley, belonged to the basement-bed of the London Clay,
being above the Reading Beds. He maintained that the balance of
probabilities was in favour of the beds referred by him to the Red
Crag being of that age rather than Drift.

406 N. STORY-MASKELYNE AND W. FLIGHT ON THE CHARACTER

32. On the Cuaracter of the DramantirEeRous Rock of Sourn AFRICA.
By Prof. N. Srory-Masxetyneg, F.R.S., Keeper, and Dr. W. Frieut,
Assistant, of the Mineral Department, British Museum. (Read
June 10, 1874.) ;

Intropuctory Norte.

Tue character of the rock in which the diamonds are found in
South Africa, in the various diggings of Du Toit’s Pan, Bultfontein,
(the Old) De Beer’s, and the Colesberg Kopje or New Rush, is
essentially the same.

In some observations I made to the Society on the occasion of a
notice of Mr. Dunn’s on the subject of the rocks of the Diamond-
fields, I gave a brief account of the minerals of which the particular
rock in question appeared to be composed. I had at that time only
optical, crystallographic, and pyrognostic features to go upon.
Since then these minerals have been carefully selected and sepa-
rated, and have been analyzed in the Laboratory of the Mineral
Department by the careful and accurate hand of my colleague
Dr. Flight; and the results that gentleman has obtained confirm in
almost every particular the forecast I gave to the Society regarding
these minerals.

The specimens of rock examined by us came from various depths—
from near the surface down to 180 feet (in the case of one mass
from the Colesberg Kopje).

Their general character is that of a soft and somewhat pulverulent
ground-mass, composed of a mineral soapy to the touch and of a
light yellowish colour in the upper, and of an olive-green to bluish-
grey colour in the lower regions of the excavations. In this
ground-mass are seen interspersed fragments of shale more or less
altered and a micaceous-looking mineral which sometimes rises into
an important constituent of the rock. It is a mineral of the
vermiculite group. Crystals of a fine bright green colour of a
ferriferous enstatite (bronzite) are not infrequent; and a hornblendic
mineral is occasionally met with, apparently rather as an accidental
than as a constituent ingredient: it closely resembles the mineral
known as smaragdite. Garnet is also frequently but sparsely met
with. Ilmenite also is present, generally in rather greater abun-
dance than the garnet. A more important mineral, however, than
these as a constituent of the rock, since it occurs in much larger and
more generally diffused fragments, is a kind of bronzite of a paler
buff tint, and in this respect, as in the size of its crystals, differing
from the former bronzite, which owes its brilliant green to a small
amount of nickel, and exists uniformly in small prismatic crystals
not larger than a canary seed. A diallage much altered, but
recognizable by its optical as well as by its mineral characters, is
present to some amount in the rock of Du Toit’s Pan.

Opaline silica, occasionally in the form of hyalite and sometimes
resembling hornstone, is disseminated through the greater part of

OF THE DIAMANTIFEROUS ROCK OF SOUTH AFRICA. 407

these rock-masses ; and calcite has penetrated them everywhere, in
some cases even surrounding and, as it were, cementing the diamonds
in their rocky setting.

The analyses of the several minerals composing the rocks are
given below, with the localities from which they were taken. They
will be seen to exhibit this undoubtedly once igneous rock in the
light of a bronzite rock converted (except where the remains of
crystals have still survived the process of metamorphism) into a
hydrated magnesium silicate, which has the chemical character of a
hydrated bronzite. It is possible that the steatite-like magma in
which the other minerals and shale fragments are contained may
have, in fact, originated in an augitic mineral once more plentiful,
but of which part of the calcium and the silica have been separated
as opaline silica and calcite.

It is, however, not very probable that a separation of ingredients
in this way from a decomposed augite would leave a hydrated
bronzite with so definite a character as is the base of this rock.
It is far more probable that the calcite is an infiltered ingredient,
and that the silica has been imported by the agency of water,
probably thermal, and possibly associated with the secondary results
of the volcanic outburst to which the columnar pipes filled with these
rocks, and of which they were first shown by Dr. Cohen to have
been probably the volcanic throats, were due.

It would seem that the diamonds are more plentifully, if not
almost exclusively, found in the neighbourhood of dykes of diorite
that intersect the hydrated rock or occur near it in the horizontal
strata through which the igneous rocks have been projected. This
is the view taken by Mr. William Nevill, whose experience on the
diamond-fields is the more valuable from his having carried with
him to South Africa some mineralogical experience. At Bultfontein,
where the diamantiferous rock is in contact with a narrow dyke of
diorite, small diamonds, numerous and remarkable for the absence of
colour in them, occur in the immediate neighbourhood of the diorite.
At the “ diggings ” of Du Toit’s Pan, on the other hand, where the
diamond-bearing rock intervenes between the diorite and the horizon-
tal strata of shale, large yellow diamonds, accompanied by a few white
ones, were found. Jn the old De Beer, again, diamonds were found
in a rock that abutted on a large boss-like mass of a much more
coarse-grained diorite. The same thing was true of the rich deposits
of diamonds found at the so-called New Rush or Colesberg Kopje.

The distinctive character of the diamonds met with in these
different localities would seem to point to a source in each case not
very remote from the place where they are found, a confirmation of
which is further to be seen in the sharp and unabraded character of
the edges of the crystals, while the fact that so many diamonds
have undergone fracture and in fact are found as fragments, would
not be inconsistent with such a view.

The presence, in intruded masses, of diorite, and the crushed and
broken appearance so often presented by the metamorphosed en-
statite rock containing the diamonds—the fragmentary nature of so

408 N. STORY-MASKELYNE AND W. FLIGHT ON THE CHARACTER

many of these stones, while at the same time they offer no evidence
of attrition—the eroded lines carried on their octahedral faces by so
many of the diamonds, indicating (as shown by Gustav Rose) the
probability of their having beenexposed to an incipient combustion,—
these facts, added to the brecciated character of many of the rocky
masses containing fragments of carbonaceous shale, all seem to
point to a great disturbance of the original enstatite rock. This
rock, probably at some depth below the present surface, and possibly
at the places of its contact with carbonaceous shales, was probably
the original home of the diamond—the alterations that have ensued
from its shattering, at a period subsequent to its becoming solidified,
having aided in effecting the hydration that has so largely changed
it from an enstatite rock into a mixture of enstatite with a
hydrated enstatite, a combination which, both in its composition
and structure, recalls vividly to the mind the similar mixture of the
former mineral with the so-called pseudophite in which it occurs at
Zdar in Moravia.—N. 8. M.

I. Rock or Burtrontein.

The ground-mass of this rock is formed of a drab-coloured mineral,
which appears to be quite broken into fragments that are cemented
by calcite in bar-like forms, much resembling crystals of felspar.
This contains a good deal of the bright green bronzite which will be
hereafter noticed, while garnet, always surrounded by crystals of a
kind of vermiculite, is more abundant than in the varieties of these
rocks from other localities. There is also a paler-coloured bronzite
(see No. 4); and the whole rock appears to have undergone more’
than ordinary change.

This rock has the following composition :—
Oxygen Ratios.

ime carhonaten eae oe 59-625
Magnesia carbonate ........ 4-972
ikronecarbonatemmeon meee: 3°016
Siliciceacidy dees) aes 20°700 11:04
TAMIAMI AM EMER eee 0:553
Tron qpEOtoxId ew ie ere ae 4296 0:95 } 3.97
(Milo risa iii. <f Se eee 5799 2°32 7;
JETT WS Be ee, Sea ace ee 0:5 24
Weitere War Us Oe f eS undetermined

99-485

Treatment with potash removed only a small portion (2-996 per
cent.), which consisted chiefly of silicic acid; some portion of this
constituent was most probably protected by the carbonates from the
action of the alkali. The rock consists, it will be seen, of nearly
68 per cent. of carbonate and a few per cent. of opaline silica, the
remaining constituent being a hydrated ferro-magnesian silicate,
very similar to that which will be described as occurring in the drab-
coloured rock from New Rush.

OF THE DIAMANTIFEROUS ROCK OF SOUTH AFRICA. 409

IJ. Specimens rrom Du Torrt’s Pan.

The drab-coloured mineral, or hydrated bronzite, again forms the
ground-mass of the rock. Through it is disseminated a considerable
amount of a vermiculite mineral, which often takes upon the surface-
layers a fine bluish green-colour like that of clinochlore, the whole
mass in some cases having this colour. Besides these there occur
small imperfect crystals of the bright green bronzite, garnet, and
ilmenite, while fragments of brownish-white zircon and crystals of
a hornblendic mineral, with much the aspect of smaragdite, are
authentically stated to come from this locality, with grossular
garnet and brilliant little black tourmalines. One diamond erystal
in the British Museum is tm setu in a mass of this rock, singularly
gneiss-like in appearance, the mica-like mineral really consisting
of the mineral resembling vermiculite, which we now proceed to
describe.

1. Vaalite.

The mineral resembling vermiculite occurs in hexagonal prisms,
the angles of which are nearly 60° and 120°. Several measurements
gave a mean angle of 59° 50’. The crystals are clinorhombic, the

Fig. 1.—Projection of Vaalite on Plane of Symmetry.
£002

angle (y) of the oblique axes being very near 103° 30’. The faces
measured were those of the forms ¢ (001), a (100), m (110), 6 (010).

100, 001 = 7630

110,110 = 59 50

110, 001 = 83 20, calc. 83° 19’
001, 010 = 90

The plane c offers a very facile cleavage, far less ready cleavages
following the directions of the faces m.
The normal of the face 001 is actually, or very nearly, coincident
with the mean line of the optic axes, which are inclined at so very
Q.5.G.5. No. 120. 26

410 N. STORY-MASKELYNE AND W. FLIGHT ON THE CHARACTER

small an angle as to give a stauroscopic figure almost identical with
that of a uniaxial crystal.

In parallel light, however, the cleavage-plane exhibits principal
sections parallel and perpendicular to the edge c, a; and a slight
opening of the cross perpendicularly to that edge can be detected, so
that the optic axes would lie in the plane of symmetry. The crystal
is negative. It hardly exhibits any dichroism for light traversing
the principal cleavage-plane perpendicularly.

When heated on platinum-foil a little piece expanded to about six
times its original thickness; after it has been reduced to powder it
does not expand at all. By long boiling in strong hydrochloric acid
it gradually expands, and loses constitutional water at abont 110° C.
Heated for many hours at 100° C., it apparently underwent no con-
stitutional change.

The powder dried for an hour and a half at 100° C. lost no weight.
In one portion the loss of weight at higher temperatures was ascer-
tained; in another the total amount of water present was directly
determined. The results are as follow :—

6 6 per cent.

During half an hour at 125-180C. .... 1:309
» half an hour at 125-130 .... 0°205

», one hour at 130 soon Ue

» half an hour at 150 pcg Oeil

» half an hour.at 150 oe PIG)
1:949

Water evolved at a low red heat .... 9-717

All water which passes off at temperatures about 100° C. has been
regarded as constitutional water.
The mineral has the following composition :—
Oxygen Ratios.

Silicichacidis ym. oe 40-833 PALO. sons. IO
PNhTHavIOR, ~ Belo a acess 9-801 4-562 6-616 3
Iron peroxide .... 6°844 2-054 ih heap the
- Chromium oxide .._ trace.
Magnesia ........ 31°338 12:°5385 ....-76
Oda eee ae, 0:670
Wiatere i ais ctrl rat: 9717 OBO, vase | 4!
Carbonic acid .... trace.
99-203

The above ratios correspond with the fatale
R, 0,, 810, + 2 (83Mg0, 2810,, 2H,0).

On comparing the atomic ratios obtained from the analysis of this
mineral with those given in Prof. Cooke’s recently published memoir
on the Vermiculites*, as those of Jefferisite and Hallitet respectively,
ERIM: Cooke, ‘Proc. Amer. Ac. Arts and Sciences,’ December, 1873. ‘Phil.
Mag.,’ April, 1874.

+ The choice of the name of this mineral is unfortunate, as it is already in
use as a synonym for the aluminite (Websterite) of Halle.

OF THE DIAMANTIFEROUS ROCK OF SOUTH AFRICA. : 41h

it will be seen that they do not coincide with those of either of these
minerals.
BOs BkOsrce SiG, gldecduO:

UeHeLIslte? ae aaa ok Ae 12 6 15 18
allipemee iste. es ee 12 2D, 9 12
iteiabele ter eee) ws 12 2 10 8

As we can see no grounds for identifying the mineral under
description with either of the above, we propose to call it Vaalite,
from the Vaal river, and to consider it to be an outlying member of
the vermiculite family.

Analyses are given in the paper of Prof. Cooke of two varieties
of Hallite, a green and a yellow, which vary very slightly in compo-
sition, and are only distinguished by the fact of the green variety
containing somewhat more than 1 per cent. (1°13 per cent.) of iron
protoxide, and the yellow variety 0-32 per cent. This would lead
to the assumption that the development of colour is due to the
presence of a small amount of iron in the lower state of oxidation.
The drab vaalite is accompanied, as has already been stated, by a
few flakes of a fine blue tint; and it was considered important to
determine whether these contained iron protoxide. They were
boiled with strong acid in an atmosphere of carbonic acid, and tested
with gold solution, when a precipitation of metal equivalent te
4-282 per cent. of iron protoxide took place.

2. Bronzite.

Small bright green crystals, with something of an emerald tint,
are among the minerals that occur in the diamantiferous rocks of
Du Toit’s Pan; and they are at the same time a permanent in-
gredient, more particularly in the neighbourhood of the garnet, at
Bultfontein, and occur more or less sparsely in the rock of Coles-
berg Kopje.

These crystals resemble very closely those of the bronzite of the
meteorite of Breitenbach, but are less rich in crystallographic planes ;
in fact they are nearly confined to those of the prism and of the faces
of the form 100. ‘The prism-angle is about 87° 20’, and the prin-
cipal sections are parallel and perpendicular to the zone 001. The
crystal is undoubtedly prismatic.

It was dried at 120°C., and analyzed with the following results :—

Oxygen Ratios.

SHicig acid! he ice 7 eRe 55-908 29-817
PAU UTTITTIae Peeks. hots. eee 2-638
Chromium oxide ...... 0-539
Iron protoxide ........ 4-991 1:109
INackelvoxideyy. sea ae trace 15:073
Maenesian 2.) stasis = ae 34-912 13-964 |
Teme 24 Obi po ee 0:457
Wistentolt 222 36813 not determined

99-445

Qe

412 N. STORY-MASKELYNE AND W. FLIGHT ON THE CHARACTER

With the exception of the enstatite of Aloysthal, in Moravia,
analyzed by Von Hauer, this bronzite is distinguished for containing
less iron oxide than any specimen as yet examined.

Although the mineral was dried at 120° C., a little water was still
retained; but its amount was not determined.

A portion of the powdered mineral was treated with a mixture of
equal volumes of strong hydrochloric acid and water at 100° C. during
twenty hours, and subsequently with potash. Rather more than a
quarter of the material was dissolved, as shown below :—

Oxygen Ratios.
Silicie acide eyes eee 15°445 8-23
IAT UMN AP ecto deren each 0-648
Tron oxide, with a trace of
chromium oxide.... 1:505 0:33 } 3.30
Maomesiay syn. bees coer 8-685 3°47
26:283

The action of the acid, therefore, was that of a solvent only.

3. Smaragdite.

Brilliant greyish-green fragments of crystals are met with among
the minerals in the rock of Du Toit’s Pan. They consist chiefly of the
planes of a prism with the angle, as determined by the goniometer, of —
about 123° 15’, which is that of a mineral of the hornblende type.

Acid had no action upon this mineral beyond removing the
adhering carbonate., After having been dried for several hours at
120° C., it was analyzed with the following result :— .

Oxygen Ratios.

Siliciczacidye: Vere as area 52:967 28°25
Alumina, with a little chromiumoxide 1:939
Ixontprotoxidene cree area ere 4-522 1-005
ALATIVG Wipes y tack ee co on Re eect ee 20-475 5:85 13°85
Monreal ain gaictc Py. cain yu eee hae Fe 17-487 6-995
Soda, with trace of potash ........ ie
Waite ripaine swing mca sgh iste 0-577
99-739

III. Rocx-speciuens rrom New Rusu, ox Cotessere Korse.

The aspect of the rocks from New Rush appears to vary consider-
ably with the depth. In the higher parts and down to 70 or 80
feet the prevailing tint is a drab yellow; at greater depths it as-
sumes a bluer tint, and consists largely of a breccia containing frag-
ments of the shales through which the rock has been intruded, with
occasional fragments of igneous rocks.

In selecting material for the investigation of this much metamor-
phosed rock, those varieties were taken in which the metamorphosis
appears to be the least complete, while they still retained charac-
teristic features of the diamantiferous deposit.

OF THE DIAMANTIFEROUS ROCK OF SOUTH AFRICA. 413

Among these a small specimen offered special advantages for the
investigation. Its first aspect is that of a piece of serpentine of dark
greenish hue pervaded by porphyritic-looking spots, many of which
possess crystalline outline, and, at first sight, somewhat resemble
the crystals of olivine in an augitic lava rich in that mineral. The
minerals that make up this rock are, first, the dark greenish base,
secondly the once crystalline greenish-grey substance scattered
through it, crystals of the mineral we have called vaalite, rarely
garnet, and interspersed calcite, which pervades the whole mass.

A thin section of this rock, seen in the microscope, exhibits, be-
sides the very metamorphosed crystals before alluded to, a ground-
mass containing crystals of a similar kind, only much smaller, an ir-
regular granular mixture of what appears to be the same mineral,
mingled throughout with nearly opaque minute irregularly formed
particles that sometimes seem to present crystalline outline. The
last-mentioned microscopic grains are of a deep olive-yellow to
olive-green ; and of one the crystalline outline gave normal angles
of 57° to 59° 30’ for the measurements of two prism-planes upon a
third plane that appeared to be a pinacoid, which might be taken to
represent the inclination of the plane 101 on the cleavage-plane 100
of olivine. It would be the angle g’m of Descloizeaux, of which the
value of the normal angle would be 59° 37’. The mineral is too
much metamorphosed to present any optical characters, so that it is
impossible to confirm this conclusion by their aid. If, however,
this olive-green ingredient be olivine, it is present only in absolutely
microscopic particles, the bulk of the rock being the mineral which
fills the place once occupied by crystallized bronzite, while throughout
the whole the interstices are filled with infiltrated calcite.

The measurement of the angles afforded by the outlines of what
were once bronzite crystals offers the usual difficulties where the
observations are not controlled by optical characters; a distinct
prism-angle of 87° to 87°30’ in one very symmetrical section would
correspond fairly well with the angle of the Breitenbach bronzite,
viz. 110, 110=88° 16’,

4, Altered Bronzite.

The metamorphosed bronzite-like mineral was carefully picked
and examined. It contained no carbonic acid, and when boiled with
potash gave up avery small amount of silicic acid. The composition
of the mineral is as follows :—

Oxygen Ratios.
Nilicie acids 281.7. 2 53°462 Pera aie Nate +
Alama: sss. 0-948
Tron protoxide .... 8768 1:947
Maenestans yao 257025) © 10-3699% 135-089 27.06. 2
Dior, MMs Se Blea 2:534 = 0728
Water? oh oon ee: 8-363 TAGAY OT ae

100-000

414 N. STORY-MASKELYNE AND W. FLIGHT ON THE CHARACTER

The oxygen ratios correspond very nearly, it will be observed,
with the formula 2(RO SiO,)H,0, or that of a bronzite, two equiva-
lents of which are combined with one equivalent of water.

The darker matrix could not be separated from the above mineral,
which, as has been stated, filled very small cavities in many parts ;
nor could all the vaalite be removed. The specimen employed for
analysis, which was more or less mixed with these substances, had
the following composition :—

SINICIC/ACid Gere wy as Pew an aaron rsay ie 39°732
SA Tum in al a oon: oer eae amen lone caie eaeua Wan eer ee ale 2-309
Tron protoxidess soc ccm muscis elas ouneases 9-690
Magnesia ie ane eras. suka ie nelle cele eee 24-419
SIT eee item noe eee cr cae a eee CHR 10°162
@arbonie ael@yncer sc ce acs ee ee etate 6°556
TEER Res ae PRAT Pah rN 7547

100°415

It may be assumed that the lime and carbonic acid are present as
carbonate of lime, which is distinctly recognizable under the micro-
scope. Asmall amount of alumina invariably accompanies bronzite,
whether as unchanged crystal or in the hydrated form ; so that it is
difficult to base calculations as to the amount of vaalite present on
the amount of this oxide. If, however, we regard the whole of the
alumina as present in the form of that mineral and assume that the
excess thus taken compensates for the iron oxide which must neces-
sarily accompany it, we obtain, after subtracting the oxygen ratios
of the constituents of vaalite, the following residual oxygen ratios
given in column I. :—

alte TI.
Silicreracidie a ee 7 CLO ee as 17°6
Protoxidew pan 9-769 he 9°6
DWiaiGeries teeta nee cM eR SOAGTS al beeen nd 5:2

In column II. of the above Table are given the oxygen ratios
of a mixture of nine equivalents of the hydrated bronzite and two
equivalents of serpentine, of which mixture it is assumed the matrix
of this rock consists.

Among the minerals thus incompletely decomposed in the rocks of
New Rush, beside the hydrated bronzite already described, there
exists, in considerable quantities, a fibrous transparent mineral, ir-
regular in its outline, but exhibiting an augitic cleavage, of a pale
brownish colour, possessing in some lights a violet tinge.

Between the fibres of this mineral there are occasionally seen
minute bars of the brown vaalite ; but it seems tolerably free from
calcite. This variety of vermiculite is occasionally associated with
a yellow wax-like substance, which is probably opal. The mineral
of which we are speaking is one that was described by one of us at
the reading of Mr. Dunn’s paper, from its microscopic and optical
characters, as enstatite. It is in fact a ferriferous enstatite or

OF THE DIAMANTIFEROUS ROCK OF SOUTH AFRICA. 415

bronzite, as the following analysis will show. From the impossibility
of separating the minerals mixed with even the most compaci
specimens that could be selected, the analysis of such specimens was-
followed by treatment with potash.

The percentage composition was as follows :—

Oxygen Ratios.

Silicic acid........ 66°915 :
Atamina ..08F 48 0-970
Chromium oxide .. 0-251
Tron protoxide .... 7-778 1°72
Magnesia ........ 19-785 (OM 9-95
pumer von Ae tors Pree 1-115 0:32
Water ON aa. not determined

96-814

A quantity of the powdered mineral was treated with an excess
of potash during three hours at 100° C., when constituents amounting
to 41-614 per cent. were removed. They were:—

Dulticie acid: 64 scene ators tA eichenaes eed 34°566
Alumni ayia sis Ahlan Sanapieuemiotee a) AAR 1:152
Miageniesiat RAs tits eat. SESE. Ne cites ers ae 1:724
Water (calculated) scmtors-b ses ace) ctens 3186

40°628

The oxygen ratios of the insoluble portion indicate, as will be
seen below, the composition of a bronzite for the other constituent
mineral of this rock.

Oxygen Ratios.
Silicievacids. 6.22). 32349 17-25
ironsprotoxade) = sa 1% @a 1°72 '
: 8-94
Magnesia ........ 18:061 7:22

The matrix of the striated mineral was then examined with the
following results :—

Siliere paid: 1 ck ke eu ee ee Eee 74:776
SAUTE ein cv Gains «Ro ee eee aes 0-411
Chromium toxid ees eee eee ae 0-946
irom yprotoxtdevl. fcusetar ese aera: 6-964
Mig erie Vain dekh tees ueyona eh eis ak pe 7-221
SE rn gs Fe oc che es a en 0-609
Vcatetoeagta.. Side ke eed hie My wl OSs eee oR, Oe alge
Maimeycarwonaten (vs sash ooletcts ovatsnieme ar ks onli

99-235

* The insoluble part was washed out with dilute acetic acid towards the end
of the process, in order to prevent the passage of fine particles through the filter.

416 CHARACTER OF THE DIAMANTIFEROUS ROCK OF SOUTH AFRICA.

When treated with potash, 55°345 per cent. were dissolved. The
residue was somewhat richer in silicic acid than was that of the en-
closed mineral; and this may have been due to insufficient digestion
with alkali. The silicate appears to contain more iron oxide in this
ease, the oxygen ratios of iron protoxide to magnesia being 1:2
here, and 1: 4 in the former instance.

The portion insoluble in potash, amounting to 44-655 per cent.,
was of a dull olive-colour and exceedingly hard. Moistened with a
few drops of cold water, it immediately swelled up to twice or three
times its bulk and crumbled to powder, the action in every way re-
sembling that observed in the slaking of lime. Treated with a mix-
ture of one part of strong hydrochloric acid (spec. grav.=1'16) and
two parts of water in the cold during about two hours, 14:267 per
cent. (or about one third of the material employed) were dissolved.
A comparison of the percentages of iron protoxide and magnesia
dissolved by the acid (1.), with those of the same bases in the residue
from the potash treatment (II.), shows the action of the acid to have
been that of a solvent.

Hf II.
Iron’ protoxide 2... =: SQOO eae oe 6:964
Masnesiag® {aan fete: AeBOO ae 7:225

If we regard the matrix as made up, in addition to calcite, of a
hydrated bronzite of the form already described, and an opal having
the composition

TIME TCRACTOS etrne fe OR Lec Tal aa ae eee 95-766
BWWia Ge Tees ioc aus Si a ee Oe AEE ReaD 4-934.
100-000

and regard the free silicic acid occurring in the striated mineral as
an opal of the same composition, this striated mineral may be re-
garded as made up of

BTOMZULO Hees ens NGO ee Lae 43-850
Hydratedsbronzite: 099 tei ke. cm ee 24-017
Opaline silica wager aa ae ee etna 30:895 —
Adhitmimant sm, ce eee ay. ee Beas eager 0-970
Chromimmioxadev) ae see ee 0-251
99-983

The lithological character of the rock, even down to a depth of
180 feet from the Dempster Claim, is precisely the same as that be-
fore described. The base of the rock consists of the same ingredients
as that described above, the mass of it being the hydrated bronzite.
The rock is further very full of fragments of the shale, which has
been altered but still contains carbon; indeed the character of the
rock is almost that of a breccia, in which these masses of shale are
cemented by hydrated bronzite containing the vaalite and the
bright green bronzite, with ilmenite and the other minerals associ-
ated with it.

LABYRINTHODONTA FROM THE WARWICK KEUPER SANDSTONE. 417

33. On the Remains of Lasyrintnoponta from the Keurnr Sanv-
stonE of Warwick, preserved in the Warwick Museum. By
L. C. Mratt, Esq. (Read May 13, 1874.)

(Communicated by Prof. Huxley, F.R.S., F.G.S8.)
[Puates XXVI.-XXVIII.]

A cortection of Labyrinthodont remains from the Keuper Sandstone
of Warwick has been intrusted to me for description by the Council
of the Warwickshire Natural History and Archeological Society. It
includes all the material acquired by the Warwick Museum since
1842, the date of publication of Professor Owen’s well-known
memoir “‘On Species of Labyrinthodon from Warwickshire” *,
which contained figures of all the fossils accumulated up to that
time. I now propose to describe the additional collection, and after-
wards to review Prof. Owen’s determinations. It may thus be
possible to combine in one paper all the information on the subject
at present accessible. The Warwick Museum is fortunate enough
to possess nearly all the fragments of Triassic Labyrinthodonts
hitherto found in this country.

Nothing contained in this paper will, I hope, imply unconscious-
ness on the part of the writer of the vast superiority of the distin-
guished anatomist and paleontologist, one of whose minor works
covers part of the ground now to betraversed. The interval between
1842 and 1874 1s the period within which all that is most important
for the elucidation of the structure of the Labyrinthodonts has been
accomplished. The writings of Von Meyer, Plieninger, Quenstedt,
Burmeister, Huxley, and Hancock, all subsequent to 1842, now
constitute the chief body of the literature of the subject. In endea-
vouring to interpret the fragmentary remains of the Warwickshire
Labyrinthodonts we shall necessarily rely upon decisions arrived at
by naturalists whose material for comparison was so much more
extensive and perfect than that at Prof. Owen’s command. Reasons
will be given in the following pages for unsettling some of that
author’s determinations. In particular, an impartial reconsideration
of the five species founded by Prof. Owen seems necessary for the
true correlation of the Triassic Labyrinthodonts of England and Ger-
many. ‘The progress of every branch of knowledge calls for such
revision of long-respected conclusions with unwelcome frequency.

The Warwick specimens, whether owing to the hardness of the
matrix, or to the conditions of fossilization, are singularly frag-
mentary. In this respect they present difficulties from which the
fortunate describers of the nearly entire crania found in the shales
and sandstones of Wiirttemberg were exempt. Though fragmentary,
the fossils now under consideration are free from distortion, and
often show details of structure with beautiful clearness. No pains

* Trans. Geol. Soc. 2nd series, vol. vi.

418 L. C. MIALL ON THE REMAINS OF LABYRINTHODONTA

have been spared by the Honorary Curators of the Geological Col-
lection at Warwick to render every feature apparent.

1. MasropoNSAURUS PACHYGNATHUS, Owen—interorbital tract. Coton
End, 1867. Pl. XXVI. fig. 1 A.

This fragment comprises the frontal, with parts of the prefrontal,
postfrontal, and parietal ossifications of each side. The anterior
margin of the right orbit and the internal margins of both orbits are
shown. The surface displays the characteristic sculpture of Masto-
donsaurus, besides the posterior or interorbital portion of the sym-
metrical mucous canals (m.g.) which form the lyra. A narrow, sharp-
cut furrow, occasionally interrupted, occupies the middle line. The
parietal foramen is visible on the broken posterior end (p.f.); the
section, accidentally made by fracture, shows that the foramen is
much more contracted on the subcutaneous surface than below.

The sutures are nearly obliterated; but enough of them may be
distinguished to prove that the frontal bounds a considerable part
of the inner orbital margin. Such an arrangement is uncommon in
the Labyrinthodonta; it occurs, however, in Mastodonsaurus and
Cupitosaurus. It will be seen in a subsequent part of this paper
that something turns upon the character of the prefrontal. This is
clearly seen to have a sharply concave orbital border, bevelled to an
edge on both sides. The lower surface is quite smooth, and exhibits
no foramen or other marked feature. The greatest thickness does
not exceed 3 inch.

The inferior surface of the coronal bones is rarely seen; indeed
I am not aware that it has been exposed in any of the German ex-
amples of Mastodonsaurus. It is therefore interesting to find the
underside of the present fossil clear of matrix, and detached from
the bones of the palatal tract. The inferior surface is nearly even,
and quite smooth over the greater part of itsextent. It has already
been stated that part of the orbital margin is bevelled on the lower
as on the upper surface. The angulation is most apparent upon the
prefrontal bone, and disappears on the underside of the postfrontal.
A pair of longitudinal ridges, { inch apart, and 1 inch high, is seen
upon the underside of the parietals. They subside and diverge for-
wards, not reaching a line drawn to join the centres of the orbits.
These ridges are longitudinally striated, especially in their posterior ex-
tent. Similar ridges exist in other Labyrinthodonts, such as the Car-
boniferous Lovxomma. They have probably served for the attachment
of vertically placed cartilages, which may represent orbito-sphenoids.

This fragment does not perfectly agree with Mastodonsaurus
giganteus, Jager *. There is a row of pits between the internal
orbital margin and the frontal mucous canal. The interorbital
space is much narrower than in the large skull of MW. giganteus from
Gaildorf. Difference of size proves nothing in the case of a single
example; but if I have restored the orbit correctly, its length in
proportion to the width of the interorbital space is considerably less
in the Warwick Mastodonsaurus than in M. giganteus. In other

* = WM. Jegeri, Alberti.

FROM THE KEUPER SANDSTONE OF WARWICK. 419

words, the proportions as well as the dimensions differ. The present
fossil will probably prove to be specifically identical with some of
the remains named Labyrinthodon (Mastodonsaurus) pachygnathus
by Prof. Owen; but that species is as yet inadequately defined.

Measurements :— in.
Width of interorbital space in the narrowest part.... 1£
Length of orbit (restored), about ................ ay
Greatest thi@kmess™ "|. +... sss" scats ote eecnatnn =

Measurements of M. giganteus (for comparison) :—

Width of interorbital space in the narrowest part .... 22
ensth of orbit; about: oy, cuit siame ahccse a oe Senet 6

2. Masroponsavrus PpachyenaTHus, Ow.—part of postero-external
orbital margin. Coton End, 1872. Pl. XXVI. fig. 1 B.

The sculpture, as well as the proportions, of this fragment connect
it with Mastodonsaurus. I am not quite sure of its position in the
skull. The concave and bevelled margin, ab, must have bounded
the orbit; yet it cannot belong to the anterior or prefrontal part of
the orbit, as the underside shows no beyelling. We may pretty
confidently assign it to the outer wall, in which case the suture, ¢,
represents the junction of the postorbital with the jugal, and dd’
the malar mucous canal (Backenfurche of Burmeister). The bone
is stouter than the preceding fragment, and the sculpture is deeper.

Measurements :— in.
Greatest thickness (near the orbital edge) .......... es
Width of masallary erooves a5. see teers) es ee 2

3. MasroponsavuRUs PACHYGNATHUS, Owen—postero-external angle
of skull (left side). Pl. XX VII. figs. 1A, 1B, and woodcut fig. 1.

This fragment comprises part of the roof of the temporal fossa and
the greater part of the articulation for the mandible. These portions
of the skull are respectively represented by a strong process directed
horizontally inwards from the postero-inferior angle, and by an
expanded plate. The plate is gently convex above, and shows the
usual sculpture of a subcutaneous cranial surface; the lower side is
nearly smooth. The greatest thickness is at the postero-inferior
angle. The inferior border is thinned off to an edge ; itis somewhat
undulated, descending gradually from behind so as to overlap the
outer surface of the mandible, and then gradually rising in front.
Behind, where it is rather more depressed than further forwards, the
plate makes, with a horizontal transverse line, an angle of about 60°,
measured as rectilinear. It is mainly formed by the quadrato-jugal
(QJ.); but a portion of the supratemporal (S 7’) remains attached to
the postero-superior angle, and afragment of the jugal is present in
front. Jam not sure whether the fossil includes any part of the
maxilla. The supratemporal and jugal are coarticulated by means of
extended sutures, which overlap considerably the quadrato-jugal. On
the under surface numerous vascular foramina appear, especially in
the angle between the articular process and the quadrato-jugal plate.

420 L. C. MIALL ON THE REMAINS OF LABYRINTHODONTA

The articular process (Art.) is very strong, its broken inner face
being Zinch by 2inch. The lower and posterior surfaces are contri-
buted by the quadrate (Q); and the suture between this and the qua-
drato-jugal can be made out without much difficulty. In its lower
half the process presents roughly the form of a semicylinder placed
transversely, with the convex surface downwards. The antero-poste-
rior measurement of the semicylinder is rather less near the middle
than at either of the two ends; its lower, or articular, surface is there-
fore somewhat saddle-shaped. Above the lower convex part of the
articular process, and on the posterior surface, is a rounded protube-
rance (pr), about 2 inch in transverse width, and not quite so wide in
the vertical direction. This also is furnished by the quadrate. It rises
nearly $ inch from the vertical surface to which it is attached, and
lies directly above the narrowed part of the saddle-shaped articular
surface. Above the prominence a thin vertical plate, lying in the
angle between the articular process and the expanded subcutaneous
part of the quadrato-jugal, connects the two firmly together. It
thins off to an edge internally, and is there deeply arcuated, forming
the external border of an opening in the posterior wall of the tem-
poral fossa, through which an apertor oris, or some analogous muscle,
arising from within the temporal fossa, and inserted into the post-
articular process of the mandible, may have passed. In front of the
semicylindrical articular surface a thin expansion of the quadrato-
jugal partly conceals the quadrate, descending upon it lower at the
sides than in the middle of the articular surface. An articular car-
tilage probably commenced where the descending edge of the
quadrato-jugal terminates.

Measurements :— ine
Greatest lencthiofiragment;) fan ame: +
Greatestiwidthe,: sey ae rs a ea a ee ape as 3
Greatest thickness ofplatOn. Vis eErm eee hae Te
Articular process, transversely, about ............ il

Articular process, longitudinally, about............

olor

Fig. 1.—Section of the mandibular articulation of Mastodonsaurus.
(The bones supposed to be cut through are shaded.)

eet amen

wo

“S SINS
2

line that of the mandible (M). he zigzag line (c) shows the junction
of the quadrate (QU) and quadrato-jugal (Q J).

FROM THE KEUPER SANDSTONE OF WARWICK. 421

4, MasroponsavrRvs PACHYGNATHUS, Owen—postero-external angle
of skull (right side). Warwick, 1864. Pl. XX VII. fig. 2.
This fragment is bounded on all sides by lines of fracture. The
upper surface is much obscured by matrix ; the part visible displays
a boldly sculptured pattern. The lower surface is smooth. Upon
it portions of an extremely jagged suture, which divides the bone
into two nearly equal parts, may be traced. These are probably the
quadrato-jugal and jugal ossifications ; part of a third bone (supra-
temporal ?) may be indistinctly seen in the angle between the other
two. At the postero-external corner a hook-like fragment projects
inwards and downwards. This is evidently part of the articular
process described in the account of the preceding fossil.
The greatest thickness (towards the posterior end) is ;3, inch.
5. Masroponsavrus—portion of occipital border. Warwick, 1864.
Pl. XXVI. figs. 2A, 2B, 2C.

The next fossil to be described comprises the epiotic and portions
of adjacent bones of the left side of the skull. Some of the margins
are rounded, as if by attrition ; but the fossil, though very imperfect,
is for the rest in good condition. I believe it to belong to Masto-
donsaurus, though no other examples of the genus present the same
view of the same parts, so as to permit an accurate comparison. The
large notch, or auditory fossa, upon the posterior border is eminently
characteristic of the Labyrinthodont skull. Its horizontal outline
is, in the present case, nearly semicircular. In none of the German
specimens of Mastodonsaurus is the fossa exhibited without consider-
able distortion ; but the Warwick fossil affords an excellent view of
its boundaries and form. Nevertheless the difficulties of interpre-
tation, arising chiefly from our ignorance of the soft parts, are con-
siderable. The inner wall of the fossa is formed by the epiotic (Hp).
Its original surface has been broken away from the upper half. The
lower half recedes inwards and forwards; it is irregular in form, and
appears to have served for muscular attachment. A broken vertical
section of the epiotic, internal to the fossa, is upwards of half an
inch thick. The bone slightly diminishes in thickness forwards.
Upon the internal wall of the fossa it passes down to a depth of
nearly an inch from the superior surface. Beneath the epiotic plate,
an ascending process of the exoccipital bounds, if I mistake not, the
inner side of the fossa. This ascending process runs out to a nearly
vertical edge behind; it increases in thickness forwards, giving
strong support to the superior wall of the cranium. Below, itis ter-
minated by fracture; it may have been continued downwards as a
ridge leading to the condyle. The large skull of Mastodonsaurus
from Gaildorf gives evidence as to the existence of such a ridge.
On the inner side of this exoccipital buttress, and in the angle be-
tween it and the epiotic, is a deep pit, the analogue, possibly, of the
posterior temporal fossa in the Crocodilian skull*. Behind this pit
the occipital buttress is smooth and gently convex.

* A similar pit exists in Capitosawrus. See Quenstedt, Mastodonsaurier &c.,
tab. ii. fig. 3.

422 L. C. MIALL ON THE REMAINS OF LABYRINTHODONTA

The outer wall of the auditory fossa is constituted by a somewhat
similar vertical buttress, the connexions of which are not quite clear.
The internal face of this buttress, which forms the outer boundary
of the fossa, is furnished by the supratemporal (S 7’) above, and by the
postero-external expansion of the pterygoid below. The fragment
of the pterygoid which remains has been displaced nearly an inch
forwards. The supratemporal extends over the whole of the upper
part of the surface now under description, and joins the epiotic in the
anterior angle of the fossa. The inferior surface of this buttress has
suffered injury. When perfect, it probably consisted of a laterally
compressed vertical mass, which connected the superior part of the
cranial vault with the pterygoid. The outer surface of this
buttress is quite smooth, and doubtless served for muscular
attachment. It is concave from above downwards, and slightly
convex from before backwards. This surface formed part of the
temporal fossa, being overarched in this part of its extent by the
supratemporal and quadrato-jugal. The fractured upper and outer
edge of the fossil shows where the overarching expansion has been
broken off. No.3 nearly represents the complementary portion of
the skull as far as the postero-external angle, but belongs to the
other side.

On the upper surface of the fossil, which is sculptured much as in
Mastodonsaurus, the sutures bounding the epiotic, squamosal, and
supratemporal can be clearly seen. A corner of the parietal is also
shown. ‘The backward extension, or horn, of the epiotic is nearly
smooth ; it has lost the extremity. The auditory fossa passes almost
insensibly into the upper surface on its anterior and external sides.
Internally, the broken edge of the epiotic rises abruptly.

The under surface reveals nothing of consequence, in addition to
the details already mentioned, except part of the upper wall of an
irregular cavity in advance of the auditory fossa. Beyond a general
reference of this cavity to the acoustic chamber it would be unsafe
to go at present.

6. MastoponsauRUS PACHYGNATHUS, Owen—part of right ramus of
mandible. Coton End, 1856. Pl. XXVI. figs. 3A, 3B.

The first of a numerous and highly interesting series of Labyrin-
thodont mandibles, which I shall describe, comprises the posterior
third or so of aright ramus. The outer plate, so far as it extends,
is in excellent preservation, though its inner surface, which forms
part of the wall of the alveolar canal, is obscured by matrix; the
inner plate is almost entirely gone. ‘The articular cavity has been
injured, but not seriously, while the postarticular process, which
yields useful characters in this order, is almost perfect.

The external surface of the angular bone exhibits a remarkably
bold sculpture, which radiates from a point upon the inferior border
of the ramus, about 5 inches distant from its posterior extremity.
Near the centre of radiation it consists of deep and irregular pits,
which become elongated, especially in front, as they recede from the
centre. The upper half of the external surface is furnished by the

FROM THE KEUPER SANDSTONE OF WARWICK. 423

articular element. It is comparatively smooth from the condyle for-
wards, being overlapped to a considerable extent by the lateral expan-
sion of the skull (see woodcut, fig. 1). A shallow mucous canal (h. g.)
bounds the sculptured tract above ; it has a horizontal direction, and
nearly coincides with the suture between the angular and articular
bones. A much deeper and better defined mucous canal (d. g.), about
half an inch wide, and nearly semicircular in section, curves down-
wards and forwards from behind the condyle to near the centre of ra-
diation of the subcutaneous sculpture. At about this point it turns
inwards, as if to cross the inferior border, but loses itself without im-
pressing the smooth internal surface of the ramus. Into this descend-
ing canal the horizontal canal previously described opens. The angu-
lari-articular suture crosses the descending canal nearly at the same
point (that is, in the middle of its course), curving thence upwards,
and terminating in the fractured extremity of the large and strong
postarticular process. Above the suture, in the part of the ramus
behind the descending canal, is a small triangular patch of sculpture ;
beneath it, the narrow space between the canal and the rounded
postero-inferior border of the ramus is quite smooth. A number of
vascular foramina appear upon the external surface of the ramus,
especially in the neighbourhood of the mucous canals. Three of
these, situate upon the lower, or angular, portion of the descending
canal, open upwards and backwards; while those upon the articular
portion of the same canal, and upon the horizontal canal, look for-
wards and downwards.

The ramus terminates behind in a smooth and rounded edge,
which curves gently and continuously upwards from the lower
end of the descending canal to the tip of the postarticular process.

The top of the ramus is occupied by a coronoid ridge, which is
about 4 inches long, and about a quarter of an inch high. It passes
behind into the anterior of two transverse ascending processes which
bound the articular cavity ; towards the broken anterior end of the
fossil it gradually sinks into a groove. ‘This groove was probably
occupied by the dentary bone, which also filled part of the angle now
seen between the coronoid ridge and the smooth outer surface of the
top of the ramus. The surface of the ridge is smooth, except for
longitudinal strie. Other examples of Mastodonsaurus show that the
dentary bone extended backwards nearly as far as the centre of radia-
tion of sculpture upon the angular bone, and terminated in a point.

The condyle and postarticular process are those of Mastodonsaurus,
though much smaller than in WM. giganteus. Though not quite per-
fect, the present fossil exhibits some important features more clearly
than any of the Stuttgard specimens. The articular cavity (Art) is
saddle-shaped, being concave from before backwards, and convex
from side to side. The inner half or so of the original surface has
been broken away. Two transverse ridges define the cavity. Of
these the posterior is the higher, and extends completely across; the
anterior bounds only the inner half of the cavity. The articular
surface rests in part upon a strong internal buttress. The broken
mass which represents this buttress now extends for an inch beyond

424 L. C. MIALL ON THE REMAINS OF LABYRINTHODONTA

the middle line of the ramus ; when perfect, it extended upwards of
an inch further inwards. Below the articular cavity the sides of the
buttress retreat gradually, so that it is broadest above. The postar-
ticular process (Pt. Art.) is very Crocodilian in character, and agrees
with no Labyrinthodont genus except Mastodonsaurus. It is an inch
and three quarters long, and when perfect may have been a quarter
of aninch longer. On the outer side it rises to a sharp and irregular
edge, which falls rapidly backwards. The inner surface is smooth ;
it is concave from before backwards, variable, but on the whole
convex from above downwards. In front it passes gradually into
the under surface of the articular buttress. The junction of the
articular and angular elements may be traced, though not without
difficulty, for a short distance upon this the inner surface of the
fossil. A fracture enables us to see that the plane of the suture
passed nearly vertically through the internal articular buttress, and
that the surface of junction was very extensive. In this part of the
ramus the articular -bone is securely wedged between the V-shaped
plates of the angular. The whole of the articular cavity and the
superior surface of the postarticular process lie above the suture, and
were therefore furnished by the articular bone.

The internal cavity, or alveolar canal, of the mandible was of large
size; but its extent cannot be accurately determined, owing to the
removal by fracture of its inner wall. A small portion of this inner
wall, contributed by the articular bone, remains attached to the
upper border of the fossil in front, forming the inner boundary of
the previously mentioned groove for the dentary bone. Here it
seems to be united by suture to the outer plate of the articular bone.
To complete this part of the ramus, the fragment of the inner. wall
must be produced downwards and backwards into a vertical plate,
the upper half or more being contributed by the articular, the
remainder by the angular bone. The restored vertical plate would
converge behind towards the outer wall of the alveolar canal, leaving
a long, narrow gap along the upper edge of the ramus. I suppose
that this gap was occupied by aslip from the dentary bone ; but there
is no direct evidence that such was the case. The internal wall of
the ramus would be continued to join a broken edge now visible on
the inner and upper surface of the angulari-articular suture beneath
the articular cavity. Were the ramus entire, a large internal man-
dibular foramen would appear in the inner plate, about 6 inches from
the posterior extremity.

This mandible is considerably smaller than that of Mastodon-
saurus giganteus. Its dimensions agree, so far as can be seen, with
those of the four fragments of J. pachygnathus already described in

this paper.

Measurements :— in.
ength of frasment of ramus yee en eee 82
Grestestidenthy cme eke ee eer ae te orem 23
Tieng thvoficoronoid id Genres ee ie eee ie 4

hensth of articwlanicaviby ee aera e ee eee 3
Length of postarticular process (broken) .......... 13

FROM THE KEUPER SANDSTONE OF WARWICK. 425

7. LABYRINTHODON LEPTOGNATHUS, Owen—parts of mandibular rami.
Coton End, 1873.

In this slab are imbedded a fragment from the middle of a left
ramus, and the anterior extremity of a right ramus of a mandible
specifically identical with that figured by Prof. Owen as Labyrin-
thodon leptognathus*. I see no reason to doubt its specific identity
with the portion of the muzzle to which the same name is given } ;
but confirmation by a microscopical comparison of the teeth is
desirable. The present fossil contributes nothing of importance to
our knowledge of the species. A sculptured tract, belonging to the
angular bone, and situate in advance of the centre of radiation, shows
that the ridges, though very distinct, are closer and smaller than in
any of the species of Mastodonsaurus. Over the small patch which
remains they are remarkably continuous, and appear to increase by
intercalation, not by anastomosis. The broken posterior end of the
larger fragment proves that about the middle of the ramus the alve-
olar canal was tin. deep and half as wide. The bases of two tusks
appear in their usual position, near the anterior extremity of the
ramus ; and the row of serial teeth is continued outside them, nearly
or quite up to the symphysis ¢.

8. DIADETOGNATHUS VARVICENSIS, nov. sp.—posterior part of left ra-
mus of mandible. Coton End, 1872. Pl. XXVII. figs.3A,3B.

After prolonged examination and comparison, I find myself com-
pelled to regard this, and the three fragments next to be described,
as belonging to a new genus and species. It is to be regretted that
the characters as yet made out should be so imperfect ; they are,
however, irreconcilable with those of any form previously described.

The present fossil nearly corresponds as to size and extent with
the right ramus of Mastodonsaurus, described above (No. 6, p. 422);
the subcutaneous sculpture and the smooth tract above it, where the
mandible is overlapped by the quadrato-jugal, have the same gene-
ral character. At the same time differences are perceived, some of
which seem to be of generic value. With respect to the external
surface, it is readily seen that the sculpture upon the angular bone is
more rounded than in Mastodonsaurus pachygnathus ; the descend-
ing mucous canal (d.g.) is much less defined, and the surface behind
it, instead of being diversified by a triangular patch of sculpture, is
nearly smooth. On the inner side of the ramus, a considerable part
of the splenial plate, which is furnished by the articular and angu-
lar bones, is preserved. An internal mandibular foramen can be
distinguished, the posterior edge of which is 52 in. distant from the
extremity of the postarticular process, while its lower edge de-
scends within 3 in. of the inferior border of the ramus. The in-

* Trans. Geol. Soc. 2nd series, vol. vi. pl. xliv. figs, 7-9.

t Loe. cit. pl. xlii. figs. 1-3.

t+ The fragment described by Prof. Owen did not exhibit this character.
“ Owing to the circumstance of the anterior part of the jaw having been broken,
it cannot be determined whether any of the smaller or serial teeth were continued
external to the large anterior tusks &e.”—Loce. cit, p, 522.

Q.J.G.8. No. 120. 2a

426 L. C. MIALL ON THE REMAINS OF LABYRINTHODONTA

ternal or splenial plate has suffered injury, and it is impossible to
trace all the details satisfactorily. The course of the angulari-arti-
cular suture, in particular, is uncertain. But one interesting fact is
clear, The strong internal buttress, which in Mastodonsaurus sup-
ports the inwardly produced part of the condyle, is entirely wanting
in the present fossil, which presents in its place a smooth vertical
plate, slightly concave from above downwards, and differing in no
respect from the adjacent splenial tract. The superior, or dentary,
surface of the ramus presents equally striking characteristics. There
is a postarticular process (Pt.Art.) about 2 in. long when perfect. Its
upper surface is triangular, the base being adjacent to the condyle,
and the apex occupying the posterior extremity of the ramus. The
process does not fall away on its inner side, as in Mastodonsaurus,
but is bounded, like the external margin, by a ridge of bone; it
presents, consequently, upon its upper surface a cup enclosed by
three ridges, of which the anterior (the ascending condylar process)
is the highest. The outer ridge is convex upwards; behind, it de-
scends upon the external surface of the ramus, and terminates
without quite reaching the extremity of the postarticular process.
The inner ridge is concave from before backwards, and extends
further back than the other. The cavity enclosed by these ridges
bears much resemblance to a true articular cavity; and I have
hesitated as to whether such might not be its true character. Two
considerations, both influential, oppose this view. First, displace-
ment of the quadrate either backwards, forwards, or inwards would
be easy ; secondly, the homologous part of the mandible in Masto-
donsaurus, Trematosaurus, and several other Labyrinthodont genera
is not articular but postarticular. What I take for the true articular
cavity of Diadetognathus is the space immediately in front of the as-
cending condylar process (Art.). The posterior wall of this space is
formed by the anterior face of the ascending condylar process, which
is about 4 in. high; it is concave forwards at its base, while above it
inclines very slightly forwards. Transversely it is nearly straight,
so that on the whole the cavity has the form of a segment of a
cylinder, or more exactly, of the impression of such a solid figure.
The surface is so placed as to resist backward and nearly horizontal
pressure. In front of the articular surface is seen the commence-
ment of two ridges of bone, both of which are furnished by the arti-
cular element. The outer ridge passes straight forwards, gradually
expanding in width until, at a distance of 4 in. from the ascending
condylar process, it is #in. broad. It is at first convex upwards ;
but ? in. from the ascending condylar process a groove appears upon
its upper surface. The groove increases in width with the ridge,
and ultimately occupies nearly the whole of its transverse extent.
It may have received the posterior and unopposed part of the maxil-
lary series of the teeth. The inner of the two ridges diverges in-
wards, and ends by fracture almost immediately. It was originally
continuous with a similar ridge which now projects inwards from
the upper edge of the articular bone, above the internal mandibular
feramen. It is better seen in the following fossil (No. 9). These
two ridges are the upper edges of the outer and inner wall of the

FROM THE KEUPER SANDSTONE OF WARWICK. 497

alveolar canal. They are separated, for a length of about 34 in.,
by an elongated oval space upwards of + in. wide in its widest part.
Towards the fore part of the fossil the dentary bone overlies the
upper and outer surfaces of the external wall of the ramus. It ter-
minates behind in an elongated pointed slip, which is confined to
the outer surface; the apex was originally about 4 in. distant from
the ascending condylar process. Upon the upper surface of the
dentary bone the bases, or cavities, of about seventeen teeth are
visible. Such as admit of measurement are about } in. from side to
side, and 4 in. in the antero-posterior direction. They are there-
fore much compressed in the direction of the axis of the ramus.
In form and size they seem to resemble the teeth of Capitosaurus ;
but a detailed microscopic examination is necessary to elucidate their
structure and relationship *.

Diadetognathus is sufficiently distinguished from Mastodonsaurus
by the characters of its mandibular articulation as given above. In
Capitosaurus we may note, among other differences, that the ascend-
ing condylar process is concave backwards, instead of forwards as
in Diadetognathus, while there is a well-developed internal articular
process. Zrematosaurus differs essentially in the postarticular pro-
cess, Labyrinthodon in the form and size of the teeth. The data
for a full and accurate comparison with the last-mentioned genus do
not as yet exist.

Measurements :— ; in.
eng th, of fragment on namiusiermvs cys eels cain. 83
Greatest depths (ay aeons et eee meant else ht ot
Kength of postaxticularsprocess’ anaes aes 3 oe 1?
Distance of centre of radiation of sculpture from poste-

TION OXCTEMIGY: 22... 22) eg asee Cee Set etomieh tea 4s
Distance of internal mandibular foramen from posterior

extremuitiyg? first. eipeoeeget ee eee et ecenme cnet vs otol-o ort 54
Height of ascending condylar process.............. i
Width of ascending condylar process.............. 1,4

9. DIAPETOGNATHUS VARVICENSIS—fragment of right ramus of man~
dible.

Both ends are broken off. The external surface shows the radiating
structure upon the angular bone as in the last specimen ; and the ter-
mination of the descending groove is seen upon the inferior border.
Above the angular element, for a distance of nearly 13 in., the
external dentary process of the articular bone (that destined for the
support of the dentary piece) still remains in position. The inner
aspect shows a smooth splenial plate and part of the border of the
internal mandibular foramen. The suture between the articular and
angular pieces can be traced upon the internal surface from the poste-
rior extremity of the internal mandibular foramen backwards to near
the broken end of the ramus. At first it lies nearly halfway between
- the upper and lower borders; about 1 in. behind the foramen it takes
an upward course ; and it descends again somewhat abruptly behind,

* See Addendum, p. 434.
2m 2

428 L. C. MIALL ON THE REMAINS OF LABYRINIHODONTA

The upper surface of the fragment shows the two longitudinal
ridges of the articular bone. Of these, the internal still remains for
33 in.: its upper border is thick and rounded in front, where it
overhangs the internal mandibular foramen ; but behind it is bevelled
off to an edge upon its inner side. The fragment of the external
ridge is only 13 in. in length, corresponding in position to the ante-
rior half of the inner ridge. Its upper surface exhibits irregular
longitudinal ridges and striations. The two ridges converge in
front, nearly touching where they are broken off; but behind they
are separated by a distance of about 4 in. The anterior broken end
of the fragment shows that the section of the alveolar canal, at
about the level of the foramen, is nearly elliptical, the external
plate being strongly curved inwards in its upper half. In both the
foregoing examples of Diadetognathus, as also in the mandible of
Mastodonsaurus pachygnathus, an aperture exists upon the upper
surface of the ramus, between the two parallel and vertical plates
of the articular bone. The upper edges of these plates bear
traces of union with overlying bone or cartilage ; but no element of
the mandible can be proved to have completed the gap. It is pos-
sible that a slip from the dentary, internal to that which is pre-
served in the fossil No. 7, and diverging from it, may have overlain
the articular bone as far back as the condyle; but there is no evi-
dence that such was the case.

This fragment entirely agrees with No. 7, and must be referred
to the same species.

10. DrapEToGNATHUS VARVIcENSIS—fragment of left ramus of man-
dible. Coton End, 1872.

The chief interest of this fossil lies in the row of mandibular
teeth. Eighteen of these are more or less perfectly preserved ; and
there are spaces for eight more. As in the previous fragment, the
teeth are much compressed in an antero-posterior direction. They
are lodged in an alveolus of the dentary bone. As usual in La-
byrinthodonts, the outer wall of the alveolus rises higher than the
inner, the difference in height amounting in the present case to jin.
The teeth are firmly attached both to the floor and the sides of the
alveolus, so that, where they have fallen out, ridges corresponding
to the spaces between the dentinal folds project from the whole
width of the alveolar groove (outer wall, floor, and inner wall).
The base of the tooth is a parallelopiped, whose transverse width is
about twice the longitudinal measurement. From the level of the
inner alveolar parapet, upwards, the tooth tapers and grows more
and more conical. Its sides are ornamented with vertical striz,
easily visible to the naked eye. The largest tooth yet standing is
2 in. high. It may have lost about + of its original height; and as
the bases of some other teeth are much larger than that of the one
selected for measurement, it appears that an unbroken tooth may
easily have attained an inch in height. A detached fragment, im-
bedded in the matrix between two standing teeth, shows the true
crown. This is sharply two-edged, and much longer and more
slender than usual. It is to be hoped that the authorities of the

FROM THE KEUPER SANDSTONE OF WARWICK. 429

Warwick Museum will permit sections to be taken of one or more
of these teeth *.

Nothing in the bony part of the fossil calls for special attention.
An indistinct sculpture, consisting of elongated and separated
grooves, appears upon the angular bone. The fragment seems to be
considerably in advance of the centre of radiation, and to belong to
the fore part of the jaw. A deep and wide groove marks the junc-
tion of the dentary and angular elements upon the outer side of the
ramus.

11. Draperognaruvus vARvicENsis—fragment of jaw with teeth. Co-
ton End, 1872.

About twelve teeth, broken off at their bases, may be made out.
The specimen yields no fresh information ; and its state of preserva-
tion is such that I am not even sure whether it belongs to the
maxilla or the mandible.

Fig. 2.—Thoracic plate of Mastodonsaurus, wrth out-
line from the ‘ Paléontologie Wurttembergs.’

* See Addendum, p. 434.

430 L. C. MIALL ON THE REMAINS OF LABYRINTHODONTA

12. MAsToponsAaURUS GIGANTEUS, Jiger—part of left lateral thoracic
plate. Warwick (Dr. Lloyd). Fig. 2

This agrees so well, both in dimensions and sculpture, with the
Wiirttemberg fossil figured by Von Meyer and Plieninger* that I
have referred it to the same species. The thin, expanded, and non-
sculptured outer margin of the plate is well displayed by our example.
If the specific determination is correct, the dotted outline of the
Palaontologie Wiirttembergs requires to be carried further out.

Measurements :— in.
Greatest width; (@mpertect) Ae. ace eee eee or
Greatest width of sculptured tract ©.....-..2.--5.-- 6
Thickness at the immer side; about 3.2270. -) en eee 3

In the Warwick collection there are also six fragments of bones
belonging to the imbs. Some of these, in all probability, are Laby-
rinthodont; but in the absence of evidence derived either from their
association with determinable remains or from comparison with parts
so authenticated, it is safer to withhold description for the present.

I shall not presume to pass a general opinion upon Prof. Owen’s
two Memoirs, ‘‘ The Labyrinthodons of Wirtemberg and Warwick-
shire” and “‘On Species of Labyrinthodon from Warwickshire”’y.
The new material accumulated since 1842; in England, and especially
in Germany, renders it possible to amend the determinations of parti-
cular bones ; and to this task I shall, as far as possible, confine myself.

Plate 43. figs. 1-3, p.516. Labyrinthodon leptognathus, Owen.

So long ago as 1844 Von Meyer gave excellent reasons for regard-
ing this fragment of the fore part of the skull as distinct from
Mastodonsaurus, or any other Labyrinthodont then describedt. It
is only one fourth of the size of the corresponding part of the large
skull of Mastodonsaurus giganteus, from Gaildorf ; the palatal surface
is more completely closed by bone, the maxillary teeth are hardly
smaller absolutely, and are therefore relatively much larger than in
Mastodonsaurus ; and there is a tusk in the maxillary series. The
mucous canals, visible on the upper surface, are similar to those of
Mastodonsaurus, but not identical. There is a transverse row of
small vomerine teeth, as in Metopias, and a longitudinal row internal
to the posterior nasal aperture, as in Trematosawrus. Both series
occur in the palate of Mastodonsaurus §. This fossil is the type of a
distinct genus and species, for which the name of Labyrinthadon
leptognathus, Owen, may be retained.

Plate 43. fig. 11. p. 530.

This fragment is described as the anterior frontal of Labyrinthodon
pachygnathus. The deep pit, in what is taken for the orbital plate,

* Palaontologie Wirttembergs, t. iv. figs. 1 & 2.

t Geol. Trans. 2nd series, vol. vi. pp. 503-543 (1842).

t Palaontologie Wirttembergs, p- 36.

See Von Meyer’s ‘ Saurier des Muschelkalkes,’ t. 61. fig. 5, and t. 64. fig. 16.

FROM THE KEUPER SANDSTONE OF WARWICK. 43)

is compared to “‘a smaller foramen in the corresponding part of the
smooth orbital plate in the anterior frontal of the crocodile.” But
it is the lachrymal, not the prefrontal, which, in the Crocodilia, is
thus excavated. ‘The prefrontal of Mastodonsaurus, to which genus
the present bone might conjecturally be referred on account of its
proportions and surface-ornament, has been shown, in the earlier
part of this paper, to have a thin, bevelled orbital margin. I have
little doubt that the supposed anterior frontal belongs to the occipital
tract, and that the pit represents a foramen for the passage of a
large vessel, perhaps the carotid artery. In Capitosawrus such fora-
mina upon the occipital surface are clearly seen*. The bone cannot,
as yet, be accurately identified with Mastodonsaurus pachygnathus,
to which species, however, if, may be provisionally referred.

Plate 44. figs. 8,9. Labyrinthodon leptognathus, Owen.

The mandibles, so determined by Professor Owen, may well have
belonged to the cranial fragment which furnishes the type of the
genus and species. The teeth have not, I believe, been microscopic-
ally compared.

Plate 45. figs. 4-11; Plate 44. figs. 1-3. Labyrinthodon pachygna-
thus, Owen.

In the preceding part of this paper reasons have been given for
the belief in a smaller species of Mastodonsaurus, for which the name
of pachygnathus may be retained, The two species differ in size, in
the proportion of the orbit to the interorbital space, in the form of
the auditory fossa, and in the character of the external surface of
the mandibular ramus. As is the case with nearly all the Warwick
Labyrinthodonts, the remains are so fragmentary as to preclude an
entirely satisfactory definition of the species. There seems reason
for including under one name all the fossils here enumerated.

Plate 45.

Of the seven objects ‘represented in this plate there is probably
not one which belongs to a Labyrinthodont. Certainly not one of
them can be identified with any part of an indubitable Labyrintho-
dont specimen. The vertebre (figs. 1-8) are too elongate in the
antero-posterior direction, and they want the neuro-central suture.
The “ episternum” (figs. 9, 10) has no resemblance to any ossification
in the Labyrinthodont skeleton. The ilium (figs. 16, 17) is probably
reptilian. Of the head of the femur (fig. 18) I can say nothing,
except that it is too globose for the articular end of any of the bones
of the extremities of the well-established genera in this order. The
criticisms offeréd by Von Meyery and by Prof. Huxley+ should be
. attentively considered by all who would study, in detail, Prof. Owen’s
Memoir or the Warwick Collection.

* Quenstedt, Mastodonsaurier &c., t. 111. fig. 16.
+ Palaontologie Wiirttembergs, p. 36.
t Quart. Journ. Geol. Soc. vol. xxyi. p. 47.

432 L. C. MIALL ON THE REMAINS OF LABYRINTHODONTA

Plate 46. figs. 1-5.

The remarkable relic represented in fig. 1 has been the basis of
speculation of great ingenuity. Reasons have been given elsewhere*
for reconsidering the question. The collection of limb-bones, upon
which so much stress is laid, does not exhibit a single Labyrintho-
dont feature, while the vertebre and rhomboidal sculptured scale
associated therewith, far from confirming Prof. Owen’s interpretation,
would almost suffice to condemn it. This fossil, like the vertebrae,
ilium, head of femur, and “ episternum” already referred to, would
not, in 1841, have been even provisionally referred to Labyrinthodon,
had the knowledge which we now possess of the German Triassic
Labyrinthodonts been in existence, or had it been suspected that
the Keuper Sandstone of Warwick might contain remains of true
reptiles.

Plate 46. figs. 6,7. “Labyrinthodon pachygnathus” (Diadetogna-
thus ?).

The notch in the left side of the principal mass of bones corre-
sponds with the mandibular articulation. This cavity will be better
understood if the bone is compared with the same part of Mastodon-
saurus pachygnathus (No. 3). Differences will be readily perceived,
such as the absence in the present fossil of the tubercle above the
trochlea of the quadrate, and the greater extension of the quadrato-
jugal behind the glenoid cavity. On the other hand, this fragment
of the skull agrees remarkably well with the mandibular ramus of
Diadetognathus varvicensis, and I believe that it belongs to that

Fig. 3.—Section of the mandibular articulation of Diadetognathus.
(Explanation as before, see fig. 1, p. 420.)

species. The sculpture has the same general character as in Diade-
tognathus; but, in the absence of more distinctive features, this
would not suffice to indicate the genus. The notch near the right-
hand bottom corner of the figure probably marks the position of the
auditory fossa. A suture, dividing the quadrato-jugal from the
supratemporal, intersects the fossil from top to bottom, when placed
in the figured position. The upper (anterior) part of the suture lies

* British Association Report, 1873, p. 243.

FROM THE KEUPER SANDSTONE OF WARWICK. 433

in the mucous canal as far as about the centre of the fragment. It
is continued thence to the internal angle of the auditory notch.

The two adjacent pieces have apparently been detached from each
other, their opposed ends being originally continuous. Thus united,
the fragment would seem to represent the epiotic and some adjoining
bones ; but the means of comparison are inadequate, and I cannot be
sure that this is the true interpretation.

Plate 47.

Casts only of this interesting Jaw are preserved. Microscopical
examination of the teeth is therefore impossible. The casts do not
exhibit the condyle or postarticular process, which would probably
have yielded distinctive characters. The fossil cannot, I believe, be
certainly identified with Mastodonsaurus giganteus, Jiiger, which
differs in its sculpturing.

We are now in a position to sum up the species of Labyrintho-
donts hitherto discovered in the Keuper Sandstone of Warwick.
Prof. Owen’s list* runs thus :—‘“‘ Labyrinthodon Jegeri, L. pachy-
gnathus, L. leptognathus, L. ventricosus, L. scutulatus.”

Of these species L. ventricosus is based upon a slight modification
of tooth-structure. The position in the skull of the tooth examined
is unknown; and no part of the skeleton has been identified. There
is, I believe, no ground for affirming the distinctness of this species.
L. scutulatus has not been proved to be Labyrinthodont; and there
are reasons of weight against such a determination. We have found
it necessary to revert to Jager’s genus Mastodonsaurus, retaining at
the same time Prof. Owen’s quite distinct genus Labyrinthodon.
Adding the new genus and species (Diadetognathus varvicensis), we
get the following list :—

1. Masroponsavrvs, Jiiger.
1. M. giganteus, Jager, = M. Jegeri, Alberti.
2. M. pachygnathus, Owen.

2. LABYRINTHODON, Owen.
1. L. leptognathus, Owen.

3. DrapETogNAtTHUS, Miall.
1. D. varvicensis, Miall.

(Addendum, May 1874.)

5A. Masroponsavrus ?—epiotic and parts of adjacent bones. Blake-
down Hill, Pl. XXVII. figs. 4A, 4B.

This fossil comprises the epiotic of the right side, nearly all the
supraoccipital, and portions of the squamosal and parietal. The
subcutaneous sculpture is very like that of Mastodonsaurus ; but the

* Geol. Trans. 2nd series, vol. vi. p. 543.

434 L. C. MIALL ON THE REMAINS OF LABYRINTHODONTA

fragment cannot be determined with certainty. The epiotic is pro-
duced outwards and backwards into a thin expansion, with a rounded
or spatulate outline. The sutures run in narrow continuous grooves.
On the underside, part of the cerebral surface is exposed. Broken
surfaces indicate the thick vertical walls which bounded the brain-
case behind and on one side. A strong descending ridge passes from
the lateral wall along the epiotic process, subsiding towards the edge.
Part of the passage for the spinal cord is probably indicated by a
smooth surface internal to the inner end of the posterior vertical
wall. Like many other fossils in this collection, the present frag-
ment can only be fully understood when future discoveries have
brought other and more perfect examples to light.

Among a collection of Keuper Labyrinthodonts, forwarded for
examination by the Rev. P. B. Brodie, I find a second example of
these bones, nearly corresponding, as to extent, with the fossil in
the Warwick Museum. The backward projection of the epiotic is
broken off. The bones are somewhat larger and much thicker than
in the fossil from Blakedown Hill. The sculpturing is coarser, but
not quite so well defined. It is impossible as yet to pronounce as
to the species of either fossil. In all probability both are referable
to the same species ; but even this is not quite certain.

Transverse Sections of Labyrinthodont Teeth. Pl. XXVIII.

By permission of the Curators of the Warwick Museum two con-
tiguous teeth, from a mandible of Diadetognathus, have been cut to
show the microscopic structure. Mr. Thomas Atthey kindly made
the preparation; and to his skill we are indebted for a large and
beautiful section.

The tooth of Diadetognathus differs from that of any other Laby-
rinthodont genus. It is much compressed, antero-posteriorly, at
the base; so that its section is a rectangle, with the long sides per-
pendicular to the axis of the jaw. Above, the tooth gradually
becomes conical. The external surface exhibits numerous strie,
but no conspicuous ridges.

The section is made near the base of the tooth. The dentine is
much folded; but there are many intricacies of arrangement, which
no folding, however complicated, can explain. No pulp-cavity is
visible *.

Capitosaurus, to judge from a very rough figure given by Quen-
stedt, is the genus nearest to Diadetognathus in respect of dental
structure ; but the resemblance is not close.

No tusk of Diadetognathus, similar to those which occur at in-
tervals among the serial teeth of nearly every Labyrinthodont genus,
has yet been discovered. When the fore part of the muzzle shall
present itself, we may expect to find palatal, vomerine, or mandi-
bular tusks.

* It must be borne in mind that no opportunity aes yet occurred of ex-
amining the upper part of the tooth.

>

co tr

11238).

me ee

G.H Ford.

TRIASSIC LABYRINTHC

Quart. Journ Geol Soc Vol XXY¥ PL XXV]

ae

a ip ipso
Oi Semen
Se

Mintern Bros imp

NTA, ( WARWICKSHIRE.)

Saye of <8

Hes:

4A.
ming
™ — ——ann__aim j
: lla am ah
{ — ss
OCR aS ie ae
f r / eS oe

GH Ford.

TRIASSIC

,

Paes

ou
be ae ae

Ay

F
LABYRINTHODONTA, ( WARWICKSHIRE. )

Quart. Journ Geol Soc Vol XX¥ Pl XXVI

2.0.

Mmtern Bros imp

GH Ford

TRIASSIC LABYRINTH

Quart. Journ. Gedl. Soc Vol. XXX Pl XYVIE.

Mmtern Bros. imp

Pasties)
dunt soig wssypnyy

TIAXX [d XXX TS 20g fo29 wanop y.tenty

GH Ford.

Quart Journ.Geol.Soc.Vol XXX.Pl XXVIII.

Mantern Bros mp.

DIADETOGNATHUS.

(Tooth - sections )

Fig. 1.

Fig. 1.

Fig. 1.

FROM THE KEUPER SANDSTONE OF WARWICK. 435

EXPLANATION OF THE PLATES.

Prats XXVI.

Mastodonsaurus pachygnathus, Owen. Orbital region. The restored
outline of the orbit is indicated by a dotted line. <A. Interorbital
tract: m.g. Mucous grooves, forming part of the lyra; p.f. Parietal
foramen. B. Postero-external boundary of orbit: ab, Orbital mar-
gin; ¢. Suture; dd'. Malar mucous groove; P¢. O. Postorbital; Jw.
Jugal.

. Mastodonsaurus. Portion of occipital border; left side. A. Superior

surface: Occ. Occipital border; Hp. Hpiotic; S 7. Supratemporal ;
Sq. Squamosal; Pa. Parietal; a. f. Auditory fossa. B. Inferior sur-
face. C. Posterior surface: p. Pit beneath epiotic plate.
pachygnathus, Owen. Posterior part of right ramus of mandible.
A. Side view: .g. Horizontal mucous groove; d.g. Descending
mucous groove. B. Thesame, seenfromabove: Ar¢. Articular cavity ;
Pt. Art. Postarticular cavity.

, Owen. Outline of anterior part of right ramus of mandible.
A. Side view (from ‘Trans. Geol. Soc.’ 2nd ser. vol. yi. pl. 44. fig. 1).
B. The same, seen from above (ibid. fig. 3).

Pratt XXVII.

Mastodonsaurus pachygnathus, Owen. Postero-external angle of skull,
left side. A. Posterior aspect: <Ar¢. Articular surface of quadrate ;
Pr, Protuberance. B. Inferior aspect: @. Quadrate; QJ. Quadrato-
jugal; S Z. Supratemporal.

, Owen. Postero-external angle of skull, right side, inferior

aspect. QJ. Quadrato-jugal; SZ. Supratemporal; Ju. Jugal.

. Diadetognathus varvicensis, Miall. Part of left ramus of mandible.

A. Side view: #.g. Horizontal mucous groove; d.g. Descending
mucous groove. B. The same, seen fromabove: Art. Articular cavity ;
Pt. Art. Postarticular process.

. Mastodonsaurus? Epiotic and adjacent bones. A. Superior surface ;

the dotted line represents the supposed position of the occipital border :
Ep. Hpiotic; SO. Supraoccipital; Sg. Squamosal; Pa. Parietal.
B. The same; inferior surface.

(All the above figures are two thirds the natural size.)

Prats XXVIII.
Teeth of Diadetognathus Varvicensis, magnified 10 diameters.

2. Portion of tooth of the same, magnified 60 diameters.

436 H. G. SEELEY ON SOME GENERIC MODIFICATIONS

34, Nore on some of the Gunertc Mopiricatrons of the Przsto-
SAURIAN Prcrorat Arco. By Harry G. Srztey, Esq., F.L.S.,
F.G.S. (Read May 13, 1874.)

In not having a sternum the Plesiosauria differ from the Croco-
dilia and from all the Lacertian orders of Reptiles. Serpents with
limbs being as yet undiscovered, the only true Reptilia which
admit of comparison with Plesiosaurs in the pectoral bones are the
Chelonians. And even here, at first sight, the resemblance is not
so evident as to command attention ; for the shapes of the plastron-
bones in embryonic Tortoises are more suggestive of the pectoral
and pelvic girdles of Plesiosaurians than are the internal chelonian
bones which support the limbs, since in Plesiosaurs these osteological
elements are expanded shields which cover much of the abdominal
surface. When, however, the embryonic pectoral arch of such a
Chelonian as the Chelone mydas* (fig. 1) is critically looked at, only

Fig. 1.—Pectoral Arch and Bones of Plastron of Chelone mydas.
(After Parker.)

‘sy
Se”
e. Coracoid. pe. Precoracoid. s. Scapula.

unimportant osteological modification is needed to change its cha-
racters to those of a Plesiosaur.

The chelonian coracoid bones (c) are rod-like; but their extension
is entirely posterior to the articulation for the humerus: the bones
approximate somewhat posteriorly, are somewhat concave on their
outer margin, and terminate in cartilages of a shoe-shaped form,
which are so extended inward that their toe-like terminations meet
in the median line. Then, from the humeral articulation the two
precoracoids (yc) extend inward towards the median line; they are
inclined very slightly forward, and join either by their cartilages or
intervening connective tissue.

If, now, a line be drawn to join the median points of meeting of

* Mr. Parker, ‘‘Shoulder-girdle” (Ray Society), pl. xii. fig. 1.

OF THE PLESIOSAURIAN PECTORAL ARCH. 437

the precoracoids and coracoids, and a thin film of ossified connective
tissue extend over the interspace between them and unite these
bones respectively on each side of the animal, a pair of coracoid
bones will be constituted similar in form, position, and relation to
those which characterize the Plesiosauria, the only difference being
that in Plesiosaurs the precoracoid is connate with the coracoid, as
in the Ostrich and many Lizards; while in Chelonians it is connate
with the scapula (s): but much importance cannot be attached to the
condition of the precoracoid, since no one will contend that Plesio-
saurs are either Lizards or Tortoises; while among Amphibians the
precoracoid is a distinct bone, so placed that it might combine indif-
ferently with coracoid or scapula.

It is a noteworthy fact, familiar to all who collect Plesiosaurian
coracoids from soft strata, that the whole of this triangular inter-
space between the precoracoid and coracoid parts, which I covered
in the Turtle with an imaginary thin ossification, is liable to be
broken away in extricating the fossil; and then there remains only
a curious bone forked from the humeral articular surface, which
closely resembles the precoraco-coracoid portion of the chelonian
pectoral arch just described, and in which only the precoracoid parts
similarly meet each other in the median line. A beautiful example
of such a specimen from the Kimmeridge Clay, presented by J. C.
Mansell-Pleydell, Esq., is exhibited in the British Museum (fig. 2)*.

Fig. 2.—Coracords of Mureenosaurus.

H
'
H
Ht
oi

a)

; o
ane” Necaee

Showing connation of coracoid (c) and precoracoid (pc).

Such a specimen goes far to justify an interpretation of the Plesio-
saurian coracoids as the combined coracoid and precoracoid bones.

The Chelonian, however, has the form of its pectoral and pelvic
arches modified by the enormous amount of potential epiphysial
ossification which characterizes the subclass. The epipleural carti-
lages of the Crocodile and the small epipleural bones of Hatteria
and of Birds are, among Chelonians, changed by potential growth
into ossifications, in comparison with which the original pleural ele-
ments are small; while they so grow and unite that an epipleural
skeleton is formed, encased not in muscles as in most other ani-
mals, but in representatives of muscles, now converted, in most
Chelonians, into horny scutes corresponding with muscles in their
extension.

* Tt is at present standing over the remains of Megalosaurus, without osteo-
logical determination, and is placed upside down.

438 H. G. SEELEY ON SOME GENERIC MODIFICATIONS

On the chelonian ventral surface (fig. 3) potential growth has in the
same way repeated, in the connective tissue, the girdle-ossifications,
which were subjected most energetically to the intermittent pressure
of movement, by which all ossifications are originated, extended, and

Fig. 3.—Plastron of Embryonic Testudo. (After Owen.)

LY,
GhSaG took

Ns?

Cl. Clavicle of authors (potential precoracoid). £ Interclavicle.
Hy. Hyosternal. Hp. Hyposternal. Ay. Xyphisternal.

moulded. Thus the hyposternal and xyphisternal bones (Hp, Xy7)
reproduce the pubes and ischia in the region of the posterior abdo-
minal ribs; and in certain Chelonians those pelvic bones become as
firmly united to the elements of the plastron beneath them as the
epiphyses of a mammalian limb-bone to the shaft in old age. In the
same way the coracoids are potentially represented by the hyosternal
bones on which they rest (Hy.). The precoracoids are potentially
represented by the pair of bones which have been variously named
the clavicles or episternal bones (Cl). The chelonian interclayicle
or entosternal bone (7) has no prototype in the chelonian pectoral
girdle, and is probably the interclavicular element pressed into the
external skeleton by the precoracoids, though it might even have
been generated kinetically as a consequence of the alternate pressure
and tension of those ossifications against each other consequent on
locomotion, if there were any reason for supposing the interclavicle
absent.

These remarks upon the chelonian plastron and appendicular girdles
seem the more necessary, because the Plesiosauria show no sign what-
ever of an accumulation of organic energy in their organization which
could not be manifested in extension of the vertebral column; and
we look in those animals for no such elements as characterize the che-
lonian plastron, precisely because we find empirically that the energy
of organization, due to movement of the vertebral column, expended
itself in adding new segments to the body; while the development
of the chelonian plastron is equally in harmony with the specializa-
tion of locomotive energy in the limbs: but since the girdle-bones of
Plesiosaurs occupy similar positions to the pieces of the chelonian plas-
tron, and the similarity of the limbs implies not dissimilar functions
for the abdominal region in the marine groups of both types, I attri-

OF THE PLESIOSAURIAN PECTORAL ARCH. 439

bute the better ossification of the pectoral girdle in Plesiosaurs, as
contrasted with Chelonians, to the larger amount of organic energy
brought to bear upon the bones in consequence of their external
position, and the absence of an epipleural skeleton. The structural
differences of the limbs from those of Chelonians, in view of their
capacity for work, are only a matter of detail.

The other important constituent element of the articular surface’
for the humerus is the scapula. In Chelonians this bone is a slender
rod directed upward to the side of the vertebral column, being thus
unlike that of a Plesiosaur in disposition even more than in form—
since in Plesiosaurs the bones are invariably directed forward in
approximately the same plane with the coracoids, converging ante-
riorly. This consideration led Cuvier to doubt the determination
which named them scapule. But considering the enormous number
of vertebree in the cervical region of Plesiosaurs, as compared with
Chelonians, I would suggest that potential growth, carrying the
vertebree further and further forward, would probably take the free
ends of the scapule onward too, till those bones came to occupy
their singular position anterior to the coracoids. This supposition
perhaps receives some support from scapula and coracoid being
approximately in the same plane (though a different one) in Lizards,
and by the slightly anterior direction that the scapula occasionally
takes.

These two bones are the essential parts of the Plesiosaurian pec-
toral arch; and if clavicles or interclavicle exist, their presence is
exceptional, and only a generic difference.

In 1865 I figured and described, under the name of Plesiosaurus
cliduchus, an animal in which the scapula appeared to carry, on its
outer and anterior margin, a bone holding the same position as the
clavicle in Ichthyosaurus. A clavicle being needed to complete the
resemblance between the two types, I at the time identified the
process as the clavicular bone. Owing to conditions of fossilization,
I believe the Woodwardian specimen was delusive, and that what
appeared to be a separate clavicle was only a thin process of the
scapula, broken and partly displaced so as to show a sharp line of
division from the thicker part of the scapula, against which it abuts
nearly at right angles. Prof. Owen in 1839 (Brit. Assoc. Report,
1840, p. 56), and later, in the ‘Anatomy of the Vertebrates’ (1866,
vol. i. p. 171), stated that the Plesiosaurian “scapula develops an
acromial process representing the clavicle;” but I do not think
Prof. Owen and I intended the same interpretation; for in the
same author’s ‘ Paleontology’ (1861, 2nd edit., p. 247) a figure is
given of the scapula, which does not seem to be in accord with
any figure or published description of the genus. This figure is
reproduced in the ‘Anatomy of the Vertebrates’ (vol. 1. p. 52),
I have seen no skeleton of Plesiosaurus which will justify such a
scapula as Prof. Owen figures (fig. 4, p. 440); and probably the
broad portion of the bone, which is represented as extending dor-
sally and backward over the ribs, bearing upon it the number 51,
should have been directed anteriorly along the interclavicle ; so that

440 H. G. SEELEY ON SOME GENERIC MODIFICATIONS

the bone might be more intelligible were it revolved through half
a circle. ;

Fig. 4.—Pectoral Arch and Humerus of Plesiosaurus, as restored by
Prof. Owen.

KE. Episternumn. S. Scapula. e. Coracoid. H. Humerus.

Prof. Huxley (‘ Anatomy of Vertebrated Animals’) apparently re-
gards the Plesiosaurian pectoral arch as a constant quantity, since
the species are said to differ in little more than the proportions of
the head to the trunk, and the relative length and degree of exca-
vation of the centra of the vertebree. The scapula finds its affinities
in the Lacertilia, the ventral part of the bone in Plesiosaurus being
supposed to correspond to the glenoid part of the scapula of Jguana,
while the mass of the Lizard-scapula is supposed to be represented
by the lateral portion of the scapula of Plesiosawrus. I should not
attach more than epiphysial importance to the glenoid ossifications
in Lizards, seeing how largely epiphyses are developed in the order,
and therefore believe that no part of the Plesiosaurian scapula corre-
sponds to the glenoid ossification adjoining the scapula and coracoid
bones in some Lizards. The two chief objections to Prof. Huxley’s
view are, that in Plesiosaurs there is no evidence of more than one
ossification in the scapula; while the preglenoid part of the bone in
Lizards is commonly in intimate relation with the clavicle, which
on the hypothesis would not be the case in Plesiosaurs. The lateral
part of the scapula is of such varying size, that I prefer to suppose
it moulded from the scapula in the several species by muscular
developmont.

At page 210 of the same work Prof. Huxley gives a diagram of
the Plesiosaurian pectoral girdle (see fig. 5), based, as I learn by
letter, on ‘‘ what he imagines to be sufficient evidence.’ The chief
point in which I dissent from that figure is the treatment of the
interclavicle ; but granting that, as Prof. Huxley remarks, the bone
is not always perfectly ossified, no specimen hitherto described, or,
as far as I know, exhibited in any Museum, shows the characters of
the five-rayed mass named by Prof. Huxley “a, clavicles and inter-
clavicle?” (fig. 5, 7).

OF THE PLESIOSAURIAN PECTORAL ARCH. 44]

Fig. 5.—Pectoral Arch of Plesiosaurus, as restored by Prof. Huwley.

s. Seapula. JZ. Interclavicle and clavicles. g. Glenoid cavity. ¢. Coracoid.

Another interpretation lately given is that of Prof. E. D. Cope,
who, in his splendid memoir “ On the Extinct Reptilia &c. of North
America” (Trans. Am. Phil. Soc., vol. xiv. pt. 1. p. 51, pl. 2), conjec-
tures that the bones which I have here named scapule should be called
“‘clavicles or procoracoids;” and consequently, in restoring the skele-
ton, an imaginary scapula is introduced, which he supposes should ex-
tend dorsally over the ribs after the pattern of Prof. Owen’s diagram.
Prof. Cope’s genus Elasmosawrus, in which this structure is repre-
sented in a restoration (pl. 1), is formed so entirely on the Plesio-
saurian type, that I think the laws of osteological correlation warrant
us in affirming that, since no trace of such structure exists in any
European Plesiosaur, no such bone will ever be found in Hlasmosaurus.
The above hypothesis alone could have induced that distinguished
naturalist to name the scapule clavicles; it may also have induced
him to draw the limits of the glenoid cavities for the humeri entirely
in the coracoid bones, excluding therefrom these elements of the
scapular arch, because it would be contrary to analogy for them to
enter into the glenoid cavity if the scapule are supposed to be
clavicles. I am convinced that clavicles form no part of the ossified
mass figured by Prof. Cope. If it were asked what becomes of the
clavicles? the same question might be repeated with regard to those
bones in Crocodiles ; and if the so-called clavicles of Chelonians are,
as I believe, only potential representatives of the precoracoids, there
would be no need to account for clavicular bones, even to complete
the osteological analogy with the chelonian pectoral arch. If Mr.
Parker’s nomenclature were accepted, we should be led, starting with
a chelonian comparison, to look for the clavicles rather in relation
with the interclavicle than with the scapule.

Professor Phillips, whose loss we have lately had to deplore, ap-
pears, in his ‘ Geology of Oxford,’ to have mistaken the bones of the
pectoral arch in Plesiosawrus for those of the pelvic arch: bones
which have all the characters of coracoids are named pubic bones,
while the scapula is identified with the ischium. These determina-
tions seem to me attributable to the occurrence in the Oxford Clay
of unsuspected generic modifications, and to a Teleosaurian theory of

Q.J.G.8. No. 120. 21

442 Hi. G, SEELEY ON SOME GENERIC MODIFICATIONS

the affinities of Plestosaurus, by which alone it became possible to
regard the pectoral as the pelvic girdle. I allude especially to the
fig. 116, p. 310 (Geology of Oxford), which represents, I think, the
pectoral arch of a new genus with the posterior end turned forwards.
The forms of these bones are exactly paralleled by pectoral bones of
undescribed genera of Plesiosauria from the great Pelolithic Period
of Oxford to Kimmeridge Clay. The fact that no ilium was found
by Professor Phillips is similarly explained. Diagram 180, p. 379,
marked ‘‘ischium,” appears to me to be a left scapula; diagram 179,
p- 378, named “ coracoid of Plestosaurus,” is the right ischium ; dia-
gram 177, called “ischial bones of Plesiosaurus,” appears to me to
represent the scapule of a new genus.

The English Plesiosauria hitherto indicated have been arranged
in the genera Plesiosaurus, Pliosaurus, Polyptychodon, Stereosawrus
(which is a genus instituted for the stiff-backed Plesiosaurs of the
Cambridge Greensand), Placodus, and Tanystropheus*. Very little is
known as yet of most of these, especially of the pectoral arch ; so that
I prefer to limit this note to the genus Plesiosaurus as it is usually
understood.

Plesiosaurs may be divided into those which are furnished with a
separate interclavicle and those in which that bone has no separate
existence. Of the first family, the Plesiosauride, the type should
be Plesiosaurus dolichodewus of Conybeare; but none of the speci-
mens so named in the British Museum gives certain evidence about
its pectoral bones. Of the pectoral bones of other Plesiosaurs, Cony-
beare gave, in 1824, a restoration which has not since been mate-
rially improved upon (fig. 6), chiefly made from specimens said to be

Fig. 6.—Pectoral Arch of Plesiosaurus, restored by Conybeare.

st. Sternum. cl. Clavicle. s. Scapula. e¢. Coracoid. g. Glenoid cavity.

in the Oxford Museum, but which seem to have long been mislaid.
Conybeare correctly identified the coracoids, and accurately placed
the interclavicle (sternum) anterior to them, so that its hinder part is
overlapped and hidden by the coracoids. The scapula correctly formed
part of the humeral articular surface behind; and in front it over-
lapped a lateral wing of the interclavicle. The scapula in this spe-

* The last two genera are from the Rheetic beds,—Placodus in the collection of
Mr. C. Moore, of Bath, Tanystropheus in that of Mr. James Plant, of Leicester.

OF THE PLESIOSAURIAN PECTORAL ARCH. 443

cimen appears to be transversely divided; and Conybeare lettered
the anterior piece (the piece which meets the interelavicle) as the
lateral clavicular bone. Conybeare’s figure and interpretation were
both adopted by Cuvier, except that the bone named clavicle and
scapula is regarded by Cuvier as so singularly placed that its homo-
logy needed investigation.

The author of the article “Plesiosaurus,” in the ‘ Penny Cyclo-
peedia,’ in reproducing Conybeare’s figure, named the whole bone
clavicle, and supposes that the upward and backward process of the
bone (not seen in the figure) is alone to be named scapula.

In 1834 Mr. Hawkins printed his memoirs of Ichthyosauri and
Plesiosauri, and gave therein a restoration or diagram of the pec-
toral bones, professing to be chiefly drawn from the specimen which
he named Plesiosaurus triatarsostinus (a species with three conspi-
cuously large bones in the tarsus), which Prof. Owen, for the inade-
quate reason that there may be one or two more small bones in the
tarsus, proposed to change to Pl. Hawkinsii. Mr. Hawkins’s name
has fair claim to retain its place; for to Mr. Hawkins belonged the
merit of recognizing the species, which Dr. Buckland, who was likely
to have been well advised, in 1836 confounded with Pl. dolichodeirus
in his ‘ Bridgwater Treatise,’ a name which it still retains in the last
edition of that work. But in the pectoral restoration (fig. 7) Mr. Haw-

Fig. 7.—Pectoral Arch of Plesiosaurus, as restored by Hawkins.
st

st. Sternum. cl. Clavicle. s. Scapula. g. Glenoid cavity.

kins was less happy; for, naming the interclavicle sternum as others
had done (and as it theoretically might be), he regarded the scapulee
as clavicles, as Prof. Owen subsequently did, while the coracoids are
named scapule. I am not sure whether Prof. Owen’s description
of the scapula in this species is supported by evidence; for the
second specimen in the British Museum, named Pl. Hawkinsi, in
which an approximation to such a structure is seen, certainly might
be separated as another species. These species being regarded as
the types of the genus Plesiosaurus, the definition of the genus de-
pends upon the accuracy of these osseous determinations, if generic
characters are to be drawn from the pectoral arch; but the girdle-
bones are obscure in Pl. dolichodeirus, and the scapulee are not suffi-
ciently excayated to determine the forms of the bones in Pl. triatar-
212

444 H. G. SEELEY ON SOME GENERIC MODIFICATIONS

sostinus, Hawkins, though analogy would make it not improbable
that they have the form ascribed to them by Prof. Owen. This
doubt existing, the type of the pectoral bones of the genus Plesio-
saurus may be taken from the fine specimen contained in Case 6 of
Room III., North Gallery, British Museum, at present named Pl.
Hawkinswi, and which may appropriately be left so named. The
entire animal is about 6 feet 6 inches long, and has the arrange-
ment of pectoral bones which I here figure reduced (fig. 8).

Fig. 8.—Pectoral Arch of Plesiosaurus, restored from Specimen in
the British Museum.

Z. Interclavicle. cl. Clavicle. s. Scapula. c. Coracoid. g. Glenoid cavity.

The interclavicle, which is very small in Pl. triatarsostinus, only
just emerging in front of the coracoids, in this species is seen to be
of large size and unusual antero-posterior extent (1). This large bone
extends behind the coracoids, much as the interclavicle extends
behind the hyosternal bones of Chelonians, and posteriorly shows
indications of a median cleft. A groove, or perhaps a suture, di-
verging, extends therefrom forward and outward, most clearly seen
on the left side of the animal. If this mark indicates separate ossi-
fications, which were distinct in early life and have become anchy-
losed with the interclavicle in mature growth, those ossifications
would probably represent the hitherto missing clavicular bones.
Similar elements, not well defined, appear to occupy corresponding
positions in the large pectoral arch of the animal in the British
Museum, named by Prof. Owen Pl. laticeps, MS. If these marks
do not indicate the union of clavicle and interclavicle, then, in
Plesiosaurus, clavicles have no existence.

I proceed now to describe and define the pectoral characteristics of
the genus Plestosaurus.

Prestosavrus. (Fig. 8.)

The coracoid bones are longer than broad, chiefly placed behind
the articulations for the humeri, but also extending anterior to them,
contracting in breadth. Part of the curved anterior margin contri-
butes, with the interclavicle and a scapula, to form on each side a
moderately large foramen. The coracoid unites with the scapula,
by suture, to form the articulation for the humerus.

The scapule are very narrow, concave on the inner border, and

OF THE PLESIOSAURIAN PECTORAL ARCH. 445

straight on the outer border, which makes a sharp angle with the
lateral vertical part of the bone, which, arising in front, widens
from before backward, in a long narrow <-shape, to the articuia-
tion, where it is often prolonged as a spur over and above the proxi-
mal end of the humerus. The scapule converge anteriorly, but are
divided from each other by nearly the whole width of the inter-
clavicle, on the extreme lateral wings of which they rest.

The interclavicle is large, often as long as broad, and apparently
may include in its posterior part a pair of subordinate ossifications.
It is of modified V-shape, being concave in front, with concave sides,
which, converging posteriorly, are prolonged behind the coracoids
in a sharp point; the lateral wings are much expanded.

The other British animals usually referred to the genus Plesio-
saurus, which do not conform to this type, are shown, by their pectoral
girdles, to belong to the Hlasmosauride, typified by Prof. Cope’s genus
Hlasmosaurus, in which the interclavicular bone is entirely wanting.
At present I am able to define three new English genera which
seem to agree in this negative character of wanting an interclavicle ;
they are:—Hretmosaurus, founded on the Plesiosaurus rugosus, of
Owen, from the Lias, in the British Museum; Colymbosaurus, to be
indicated by Plesiosaurus megadeirus, of the Kimmeridge Clay, in
the Woodwardian Museum and in that belonging to Marshall Fisher,
Hsq., of Ely; and Murcenosaurus, from the Oxford Clay, indicated
by a new species in the collection of C. E. Leeds, Esq.

I have seen indications of several other genera, which may here-
after be defined; and probably, when other parts of the skeleton are
critically examined, the number. will be increased.

ERETMOSAURTS, g.n. (Fig. 9.)

Prof. Owen has had drawn, with more than usual clearness, in
the ‘ Monograph of Lias Plesiosaurs,’ the chief characters of Pl. ru-

Fig. 9.—-Pectoral Arch of Eretmosaurus, restored from Specimen in
the British Museum.

s. Scapula. c. Coracoid. g. Glenoid cavity.

gosus; but as the pectoral bones are badly preserved in that speci-
men from the Lias of Leicestershire, it is only after some study that
I venture to express ideas of their forms and relations. There may
be a little doubt as to whether a distinct interclavicular ossification
ever existed; but if such a bone were present, which another speci-
men disinclines me to believe, then the bone has entirely lost its
individuality in the mature animal, and is blended with the scapule,
just as the scapular bones in the region between the small precora-

446 H. @. SEELEY ON SOME GENERIC MODIFICATIONS

coid foramen and the glenoid cavity become united with the cora-
coids into one bone. Important light is thrown on this pectoral arch
by an isolated anchylosed scapula and distally imperfect precoraco-
coracoid bone, also from the Lias, clearly belonging to the same
genus, though probably referable to another species (fig. 10). The

Fig. 10.—Left Side of Pectoral Arch of Eretmosaurus.

H .
teen?
.

s. Scapula. e. Coracoid g- Glenoid cavity.

coracoid portion of this mass is extremely thin, but thickens at the
median line of the body of the animal, as is usual. ‘he articular
surface for the humerus shows no indication of being formed by
more than one bone, so perfectly are the scapula and coracoid
blended. On a line with the anterior margin of the articular sur-
face, and nearer the median line than the precoraccid foramen, is a
small vertical perforation, which I suppose to be homologous with
the foramen similarly placed in the coracoid of Lizards, Crocodiles,
and many Dinosaurs, and to be similarly definitive of the limits of
the precoracoid and coracoid elements of the bones, and to indicate
that in some amphibian ancestral race the precoracoids were as di-
' stinct from the coracoids as they are in the Cape frog (Dactylethra),
which has a pectoral region not altogether incomparable with that
of some members of the Plesiosaurian order (see fig. 11). The ante-

Fig. 11.—Peetoral Arch of Dactylethra. (After W. K. Parker.)

pe. Precoracoid. ce. Coracoid.

rior margin of the precoracoid portion of the bone is truncated 1n
the median line, though a suture there extends between it and the

OF THE PLESIOSAURIAN PECTORAL ARCH. 447

scapula. The anterior and inner margin of the scapula similarly
terminates in an edge which gave attachment to cartilage, showing,
I think, that the anterior median portion of the pectoral arch was
occupied not by an interclavicle, but by a common cartilage, at the
expense of which the scapular ossifications extended till they met in
the median line.

The scapula is remarkable chiefly for the length of the acromial
process (which is directed outward), for the straight antero-lateral
margin which the bone thus obtains, for the expansion of its ante-
rior end, and for the small foramen which it combines with the
coracoid to enclose.

These characters are supported by others in all parts of the
skeleton: thus the pubic bones have a prepubic process directed
anteriorly, like that seen in Chelonians; and Prof. Owen describes
and figures distinet olecranon and patella bones. The caudal ver-
tebree give no indication of chevron bones.

CoLymposavrvs, g.n. (Fig. 12.)

The coracoid bones are oblong, widening at the posterior angles,
and not extending anterior to the humeral articulation, except

Fig. 12,—Pectoral Arch of Colymbosaurus, from a Specimen ur the
Museum at Ely.

s. Seapula. ce. Coracoid. g. Glenoid cavity.

slightly in the median line of the body, and not cupped deeply to
form the foramen between the scapula and coracoid.

The scapula is of extraordinary form; the two bones meet in the
median line of the body, meet the coracoids behind by a narrow
union, and enclose the greater part of a large foramen; the ante-
rior margins of the bones are straight, converge at about an angle
of 90°; the posterior end of the straight side is prolonged into a
blunt process, between which and the humeral articulation the side
of the bone is concave.

These Plesiosaurians all have very long slender necks, and closely
approximate to the Elasmosaurus of Cope, with which I hesitate to
identify it till we know more details of the structure of the Ame-

448 H. G. SEELEY ON SOME GENERIC MODIFICATIONS

rican animal. The species are found in the Oxford Clay, Ampthill
Clay, and Kimmeridge Clay.

Moranosavrus. (Fig. 13.)

Both pelvic and pectoral arches are characterized by the bones
having no median antero-posterior osseous union. Thus, instead of

Fig. 13.—Pectoral Arch of Mureenosaurus.

s. Scapula. ec. Coracoid. g. Glenoid cavity.

two obturator foramina, this genus has but one. In the same way
there is but one foramen enclosed between the scapulee and coracoids.
The scapulee converge in front with straight anterior margins, which
make a right angle with enormous lateral processes, which differ
from those of all other Plesiosaurians in being prolonged forward.

With these characters are associated a union of the neural arches
of vertebrae, only comparable to that seen in Serpents and Iguanoid
Lizards, but with semicylindrical zygapophysial facets; while the
chevron bones join tubercles at the base of the centrum instead of
articulating between two centrums; and the ulna and radius, and
tibia and fibula, are distinct in form from those of other genera.
The type of the genus is Murcnosaurus Leedsii, Seeley, Q. J. G. S.
vol. xxx. p. 197. I should place the Plesiosaurus Oxoniensis and
Pl. Manselli in a subgenus of Murcenosaurus. The species are found
in the Oxford Clay, Ampthill Clay, and Kimmeridge Clay.

RHOMALEOSAURDS, g. 0.

One genus, found in the Alum shale of the Upper Lias at Whitby,
is represented by the fine species in the Museum of the Royal Dublin
Society, which has been named, by Mr. Baily and Dr. Carte, Plesio-
saurus Crampton. This type has given no evidence of its sternal
bones; but its other characters differ so far from those usual in
Plesiosaurus, that it may be placed in a genus by itself, to be named
Ethomaleosaurus.

_ The cervical vertebrae are nearly as short, from back to front, as
in Pliosawrus; and the cervical rib is articulated with the centrum
by two facets, as is the case with the early cervical vertebre of
Plosaurus. The ulna and fibula have not the usual reniform shape,
but are short and broad, and resemble the radius and tibia in being

OF THE PLESIOSAURIAN PECTORAL ARCH. 449

slightly constricted. No mention is made of cheyron bones in the
caudal yertebree. There are only six bones, in two rows, in the
carpus and in the tarsus, and only four digits in each limb. The
premaxillary bones appear to extend backward so as to divide the
nasal bones; and the lower jaw is unusually deep at the coronoid
bone. The double articulation for the cervical rib probably indi-
cated relationship to the Pliosauride rather than to either of the
families that I have discussed; but it is seen in isolated vertebra
from the Lias, contained in the British Museum.

I cannot but consider it a matter for regret that, although so large
a number of species have been found in the Cambridge Upper Green-
sand and other Cretaceous strata, we are still ignorant of all those
parts of the body which would warrant us in placing them in genera.
One is thus unable to pronounce any opinion on the evolution of
modifications of the Plesiosaurian girdle in relation to time. And
the only general conclusion at which I have arrived is that Plesio-
sauria, in common with all similar groups, show, in the newer rocks
as compared with the older ones, a greater amount of ossific energy,
probably coincident with higher organization, which manifests itself
in more perfect ossification of the bones, elongation of processes, and
blending of subordinate with the principal ossific centres.

450 J. F, CAMPBELL ON POLAR GLACIATION ETC.

35. About Porar Gractation fc. By J. F. Camppett, Esq., F.G.S.
(Read April 15, 1874.)

InrrRopuction,

In order to be concise and comprehensible, and to assign a
reason for venturing to address this Society upon so large a theme,
I must refer to former work. In 1865 and 1867 I published
and enlarged a book called ‘Frost and Fire. In it I aimed at
showing, by description and by designs, that certain forms are as
letters inscribed on the outside of the globe to record part of its
history.

Example. A voleanic crater and a water-course differ in form,
as do the letters O, 8, A. Both sets of forms convey meaning to
those who have learned to read. A record of volcanic action is
written on the moon in black and white, and can be read by any
one who has seen Naples and its volcanoes, old and new. In
November 1873 no record of aqueous or other “ denudation ” had
been written upon the upper cone of Vesuvius, whose igneous
outside form was remade in 1872, and grew upon the older
mountain-top after 1842. But a deep inscription is scored in water-
courses and deltas within and without the cone and crater of Monte
Nuovo, which grew in 1538, during one scattering discharge of a
fountain of hot vapours, fluids, and solids. That particular gun
has not fired since. A deeper ‘“‘water-mark” is on the older
_ Astroni,” which was a volcano, and now is a wooded preserve for
game, and upon ‘‘Somma.” If we measure the relative depth of
these water-marks upon two volcanic surfaces newly made in 1872
and in 1538, a scale of depth may be constructed by which to
estimate the age of Astroni and Somma, by “ denudation ” of their
igneous surfaces. No such water-marks are seen upon older and
bigger cones and craters in the moon, which cannot be reached,
but which can be seen.

It is recorded on the outside of these two worlds by certain forms,
that it has rained little since 1872 at Naples, that it has rained a
great deal there since 1538, that it has rained a great deal more
there since the unknown date of the last eruption of Astroni and
Somma. There is no record of glacial denudation at Naples. There
is no record of any denudation on the mountains in the moon, which
seems to have no atmosphere and no clouds. I strove to learn this
natural alphabet of form; I copied natural forms, and gave illus-
trations; and I used printers’ signs as symbols to convey my
meaning shortly. In the book quoted, A is like the shape of
mountain peaks ; Y or V the section of a water-course; 8, windings ;
A, fan-shaped deposits of drift, washed out of a water-course and
called a‘ delta.” — expresses the rounded outline of a glaciated
hill country; WW hollows made or widened by glacial erosion.
Water-courses of all sorts and sizes are alike, and record the same

J. F. CAMPBELL ON POLAR GLACIATION ETC. 451

kind of aqueous denudation everywhere. “ Ice-marks” of all kinds
differ from “‘ water-marks;” and both differ from marks of volcanic
action and of subterranean disturbance. Probably few have read
the book; so I give this explanation of my own drift in writing on
the subject.

In May and November 1873, at pp. 198 and 545 of the Journal
of this Society, you did me the honour to print notes on the
Glaciation of Ireland, and of the Hebrides. These and this paper
are sequels to ‘Frost and Fire.’ In July 1873, I left England to
look at a new country; and after travelling 150 days, I returned
in December. At your Meeting on the 7th of January 1874, I
heard able papers read upon the glaciation of the lake-districts, and
upon the confluence of glaciers in parts of England and Wales, and
a discussion in which the existence of an older glacial period was
alluded to. I then offered to give you the results of my last obser-
vations in Europe and Asia in a written form. I now give you
results of work done between 1840 and 1874, referring to my
former writings for details and explanations.

(1) Geography. The observations, of which some account is
given in the writings quoted, were made in Europe and in America.
In Europe they were made in Scandinavia, Finland, and Russia, in
Germany and Switzerland, in Greece, Italy, Sicily, and Spain, in
the British Isles, Fiird, and Iceland. In America they were made
from Hamilton Inlet in Labrador, to St. Louis and Washington.
In 1873 I coasted Norway, rounded the North Cape, and coasted
the Polar basin and the White Sea to Archangel. I crossed Russia
to the Caspian, crossed that sea to Petrovsk, and travelled through
the Caucasus to the Black Sea. I visited the Crimea, Odessa, Con-
stantinople, and Syra, and rounded the south cape of Europe. I
went to Corfu, Brindisi, Naples, Rome, Florence, Pisa, Spezzia, Sestri,
Genoa, Nice, Cannes, and Paris, copying and studying form and
superficial geology everywhere.

During more than thirty three years I have been carefully obser-
ving and describing glacial and other phenomena, within an area
which includes 36 degrees of latitude and 140 degrees of longitude.
I reckon from the North Cape, 71° 11’ 3”, to the Straits of Gibraltar,
from Camara on the Volga, 50° E., to St. Louis on the Mississippi,
90° W. It is a very small bit of the world’s outside after all; and I
have little to add to knowledge previously gathered by my teachers.

(2) Conclusion. From personal observation and reading I had
arrived at certain conclusions, which are stated in my paper on the
Glaciation of Ireland (Q.J.G.8.1873, p.218, § xvii.). I believed that
the majority of the Geological Society had arrived at similar conclu-
sions; but I seem to have gone further than other Members. I was
convinced by my facts and evidence that during the growth and de-
cay, the waxing and waning, of the last of a series of glacial periods,
small and large glaciers existed in the British Isles, that many there
joined to make small and large “local systems,” that many local

systems joined in low grounds whilst the glacial period was wax-
ing till the United Kingdom and the shallows about these islands

452 J. F. CAMPBELL ON POLAR GLACIATION ETC.

were united in bonds of thick ice from the Land’s End to John
o’Groat’s. I had come to believe that the Scandinavian ice-system,
which now has shrunk and broken up, once grew and united till it
equalled and exceeded the dimensions of Greenland ice, and then
crossed the German Ocean and joined the mainland to the islands,
as the Humboldt Glacier now joins small islands to Greenland, wading
far out in shallow seas. In America, from lat. 45° to 39°, I had
seen icebergs afloat and marks on shore, which a great man, now
lost to science, Professor Agassiz, attributed to a general solid polar
glacier. When I started questions for solution at p. 548 of the
Journal, I was inclined to follow that able leader. Then I set off
in search of facts, imitating those who have taught that travel is
needful for geological study.

(3) Hypothesis. If a polar glacier reached Washington at least,
and was part of a “‘ cap” which also overran Scotland and Ireland,
then, as I surmised, that polar system must have flowed, or waded,
or floated over the plains of Europe, down to the latitudes of
Washington and St. Louis,—in Europe say to 40°, 39°, or 38° N. lat.
That being my hypothesis, these are my facts, old and new, which
seem to bear upon polar glaciation and confluent glaciers, and the
marks which are now attributed to a glacial period or to a series of
glacial periods. ;

(4) Norway Sc. In coasting Norway once again I saw much to
confirm my opinion. I saw the coast clearly, and sketched day and
night while I could keep awake, from July 26 to August 7, when I
landed at Vardd. Sea-marks, water-courses, and signs of volcanic
action abound. Lateral pressure has bent and broken the earth’s
crust. A whole country has been first crumpled, and afterwards
* raised bodily from the sea, and bent like a bow. But I hold that
hills and hollows in that country have been shaped chiefly by the
‘‘olacial erosion’ to which Professor Ramsay long ago, and Mr.
Clifton Ward in his late paper, attributes the rock-basins which
hold lakes, and are filled with drift in certain parts of the lake-
district, which he has so well surveyed and mapped (see figs. 1 & 2).
So far as it goes, Mr. Ward’s paper amounts to a demonstration of
the truth of one proposition, which I noted. “The hollows which
contain these lakes were formed by glacial erosion.” I hold the
converse of the second proposition, which I also noted. ‘The
valleys, of which these lake-basins are part, are noé due to glacial
erosion.” I have gone further than this author, and hope that
he will follow. I seek to prove that a great many hollows of
curved section, however they originated, owe their present form
to the glacial erosion, which also produced hollows in them, in
which water lodges; and that many hills and ridges of curved
section between these furrows owe their forms to the same grind-
ing-engines which shaped “Tors” and “ Roches Moutonnées” of
smaller size*.

* Tt still is very often denied, even by geologists, that glaciers wear rocks under

them. _ A glacier near Cape Desolation, in Greenland, about the latitude of
Christiania in Norway, is about 800 feet deep where it enters the sea, in a firth

J. F. CAMPBELL ON POLAR GLACIATION ETC. 453

Fig. 1.—Svari-lis Glaciers, Norway. (The edge of the largest
Ice-tract in Europe.)

454. J. F. CAMPBELL ON POLAR GLACIATION ETC.

(5) The Polar Basin. According to Mr. Skanke, English and
Russian consular agent at Vardo, which port is frequented by Sea-
horsemen and Whalers, ‘“‘ Parry Island” * is the most northern in
Spitzbergen, and is made of gneiss. He gave me a specimen. The
most northern island reached in Novaya Zemlia is in long. 66° E.,
and about 77° N.lat. Consul Skanke gave me a specimen, brought
thence by Captain Mack, which contains a large coral. Bear Island,
near lat. 75° and north of Norway, is rich in fossils which are taken
to indicate a warm climate. Judging by form and information I
concluded that some hills and islands, about the mouth of Kolafjord,
and at the entrance to the White Sea, are patches and outliers of
stratified, unaltered, undisturbed beds, resting unconformably upon
far older, greatly inclined, disturbed, crystalline, altered beds, which
have been “ denuded” and ground into the shape of Russian Lap-
land. I find, on referring to the geological map of Europe, that my
inference was correct, so far as the nature of these rocks is concerned.

At Yeredick the rock is gneiss, the dip nearly vertical, the strike
N.W. 8.E., the direction of glaciation N. and S., magnetic. In
neighbouring hills I saw faults which I take to be postglacial. A
smart earthquake was felt near Kola in February 1873. At other
places on the coast the rocks are gneiss, with veins of pink granite,
like rocks which I have seen in the Hebrides and in Labrador.
Generally the coast of Russian Lapland is made of old, hard, con-
torted, fractured rocks ground into curves. On the east side of the
White-Sea straits the beds are nearly flat, and seem to be little
altered. Limestone is there burned, and it is taken to Archangel’.

(6) Rise of Land. I spent five days on Vardo Island, which is
in the Polar basin. I found on the sea-beach subangular and
rounded stones. I found bits of metamorphic slate, broken from
overhanging beds of rock, in the bay. These dip at an angle of 45°
and strike N.S. Bays and hollows and ridges are on the strike ; but
deep “‘ Gjis” cross at right angles. The general form of the island
is that of glaciation. Up to about 30 feet is the ‘Storm beach” with
larger and more angular stones on it, cast up in a ridge which sweeps
in a crescent round the bay. Up to about 60 feet the raised beach is a
repetition of the actual beach, but grass-grown. On it are large
round blocks of granite. No granite is near im situ. The highest

about three miles wide. Photographs were taken there by Mr. Bradford, which
which were shown at the Meeting. No moraines are on the surface; but under
the ice at the side of this glacier are large boulders, and fine débris in contact
with rounded rocks. All glaciers move at slow rates, varying with circum-
stances. A cubic foot of this ice must weigh about 55 lb. Therefore the
pressure on the bed of this glacier is nearly twenty tons per square foot. If a
train wears wheels and rails, and a carriage its drag and the road—if stones and
sand rolled by rivers and the sea wear rocks, it seems evident that stones and
sand must be crushed and ground, and must crush, break, and wear rocks. under
the great pressure of moving glaciers. Practically, all glacier-rivers are thick
with débris worn from rocks. As glaciers wear rocks, then icebergs driven by
currents must also do the same, under water, when they touch the bottom.

* The Parry Islands of maps are to the north of America.

t Since my return I have read Murchison’s ‘ Russia,’ vol. i. 1845. The rocks
referred to are marked Devonian.

J. F. CAMPBELL ON POLAR GLACIATION ETC. 455

beach in this hollow is about 100 feet above the present sea-level,
and partially covered with peat. Rolled stones and stones packed
by water are close to the highest point in this island at 200 feet.
The summit is 220 feet above the sea, and consists of red altered
sandstone, very hard and ringing to a blow. The east coast is full
of great rifts, in which waves dash and growl.

Peat beds on the raised beaches are brown and dusty, with a
considerable vegetation growing on them. Rock surfaces are un-
mistakably wave-worn; amongst the stones on the old beach I
found granite, gneiss, and red sandstone, with fossils in the shape
of concentric rings, like the rock of Vardé. I also found rolled
brown pumice stones exactly like those which I found in Tiree.
These were abundant: but the nearest volcano is a long way off; so
they have floated far, and mark the equatorial drift from the
Hebrides to Vardo on the way to Novaya Zemlia.

(7) Result. Vard6 Island, at the end of a long promontory and in
the Polar basin, was first ground into shape by ice, as I suppose.
It has been submerged and wave-worn since; and as it rose, drift
upon it was rolled and packed by the sea, which carried light drift,
and may have carried granite on rafts from anywhere. It may
have brought gneiss from Parry Island or from Russian Lapland,
pumice from Spitzbergen or Iceland. Then peat grew; and now
people live there, and fish and roam about in the Polar basin hunting
and sealing and making money. Meantime the waves are breaking
the exposed side of the island into rifts and cliffs. So I read the
forms on the island, without trying to make out the age of the rocks.

Since my return Consul Shanke, of Vardo, has sent me the mete-
orological journal of a crew of Walrus-hunters, who were frozen-in
north of Novaya Zemlia. They went in June 1872; and the journal
ends May 12th, 1873. The men who kept the journal, father and
son, died. The survivors got home to Tromsd in December 1873.
No one else has wintered there, so far as I am informed.

(8) Ocean-currents and their effect. Captain Mack named the
most northern island in Novaya Zemlia “ Castanje 6” or “ Mimosa,”
because he there picked up ‘“ chestnuts” or ‘“ Mimosa” beans,
These brown nuts prove ocean circulation, as I have shown in
‘ Frost and Fire, vol. i. p. 483. They are tropical growths carried
to frozen lands. The island must be about lat. 77° N., long. 66° E.

To that point the equatorial current now reaches; and there the
warm water must plunge under water, which after a certain tem-
perature gets lighter as it gets colder, till it floats ready to freeze.
Solid ice was lighter than sea-water off Labrador as 9 is to 10 by
bulk. In Spitzbergen the “ chestnuts ” get to 20° E., 77° N.; and
the warm water of the equatorial current then goes under the ice
to warm the sea-creatures who live at the bottom. The fortress of
Vardo and the church at Vadso are built in coral sand ; it is there-
fore possible that fossil corals may have lived under similar condi-
tions. Numbers of whales come into the fjord ; thirty-two were
slain by one small steamer in 1873 before August 6th. Fish of
many kinds abound. There is plenty of life in the Arctic basin now

456 J. F. CAMPBELL ON POLAR GLACIATION ETC.

to make fossil-beds with corals in them. At Yeredik, about 70° N
I found a luxuriant vegetation on shore—the wild corn of Iceland,
«* Baldur’s flower,” the cotton-grass, the “ Indian tea” of Labrador,
and many bright wild flowers. Geraniums and cacti live in pots in
the houses; and canaries live through the winter in warm rooms.
The winter’s darkness does not kill; therefore the poles need not be
changed theoretically in order to account for Arctic fossils. The
hills are clothed with firs and birches where there is soil enough
for them to stand in. Peat is common; so materials for coal grow
now. About lat. 69° N., harbours near the mouth of Kola fjord
never freeze. They are sheltered from the eastern drift and remain
open all winter. A few miles further east, at the Seven Isles, the
harbours are blocked every winter. On the 10th June, 1873, they
were still blocked ; on the 15th of August snow patches lay on the
beach, but no ice was anywhere near on sea or land: it had all
drifted away.

(9) Result. The warm equatorial current now affects climate in
the Polar basin to lat. 80° in Spitzbergen and to long. 66° EK. in
Novaya Zemlia. The cold Arctic current affects the coasts of
Greenland, Labrador and North America, and carries floating ice
down to lat. 37° N., long. 47° W.* “That which the sea actually
does for climate now, the sea may have done when the fossils of the
Arctic basin lived, and when land now dry was submerged. Arctic
currents are part of polar glaciation, which is the subject of this
paper.

(10) On the former trip (described, ‘Frost and Fire,’ vol. il.

p- 503) I crossed from the Polar Basin about 70° N. to the Gulf of
Tatnted and searched between 25° and 30° E. long., through
Finland to St. Petersburg and Berlin. I found all known marks
of extensive glaciation continuous as far south as 55° or there-
abouts. In ‘Frost and Fire’ (vol. ii. pp. 3-17) I have described
the marks on which I rely. I found No. 7 “ Hollows” ~ holding
lakes and bogs, No. 9 “ Tors” ~ between the hollows. I found
« Drift ” upon rounded hills and in rounded hollows, on ridges and
in grooves, polished striated rocks under drift, and scratched
polished stones upon the polished rocks. I found “ thirteen wan-
dering blocks” (great masses of a peculiar Finnish rock which is
much used in St. Petersburg, carried southwards and left on sandy
plains) along the shores of the Baltic and in the neighbourhood of
Berlin. I found most of these marks in Devonshire down to lat.
50° 30’ N.+, and in Ireland as far west as long. 10° W. ‘The Alps
and Pyrenees carry European glaciation as far ‘south as 45° N., but
not continuously, so faras I know.

(11) Russian Lapland. In 1873 I found the shores of the
Polar basin between 35° and 40° E. long. universally glaciated at
about 70° N. lat. The hills are all rounded, and their sky-line is
studded with great “perched blocks.” Rounded hollows ~ are
glaciated and partially filled with boulders, with boulder-clay and
drift, with lakes and bogs; all the marks upon which glacialists

* «Frost and Fire,’ vol. ii. p. 247. t Ibid. chap. 42, p. 220.

J. F. CAMPBELL ON POLAR GLACIATION ELC. 457

‘ely abound. Rivers have made little or no impression upon the
rocks, and commonly flow over drift. Thus in Russian Lapland as
n Finland and Scandinavia, from Stavanger to Kola, the action
of very heavy ice is proved by conspicuous marks. The record is
continuous from Kola to Cape Clear, to near the Land’s End and to
Berlin, and to lat. 50° N. in one direction.

(12) fuse of land. <A series of raised beaches follows the whole
northern and western coast thus far, and marks a late unequal rise
of land in this great tract of country. Shells in sitw confirm the
proof by form, which is everywhere conspicuous. Amongst these
marks sea-cayes at considerable elevations are numerous.

(13) Zesult. If ever a solid “ice cap ” radiated from the Pole,
and reached to Washington, and to the south of Ireland and to
Berlin, these glacial marks on the edge of the Polar basin along an
are of 15° about the circle of 70° N. may be work done by the
“ice cap” when it was shrinking northwards to leave isolated
local systems in Greenland, Iceland, and Scandinavia. Something
ground these regions of Northern Europe. An Arctic current like
that which now coasts Greenland and Labrador seems to me in-
sufficient to account for the work done. The new facts tell in
favour of the Polar glaciation for which Agassiz contended. His
theory had not come to the front when Sir Roderick Murchison
travelled in Russia. In his ‘ Geology of Russia’ (vol. i. p. 331) Sir
Roderick mentions the discovery of recent sea-shells on the right
bank of the Vaga. The town of Tromso in Norway stands on peat,
which rests on clay, in which are recent sea-shells in the position
which living shells of the same kinds now hold in the mud. I saw
them in 1873 in drains. In both cases the shells are overlain by
Soul raubtis

(14) The White Sea. The entrance to the White Sea is not
much wider than the Straits of Dover. The shores are not higher
than the French and English coasts; and the eastern shore is a
cliff with a beach on which large stones are strewed. I have
sketches ; they show stratified beds capped by drift on which fir
forests grow.

From Kola to Kandalatz a series of lake-basins extends from north
to south. These forms support the ice-cap theory. The existence of
beds of soft rock on both sides of these harbours does not support
a theory of Polar glaciation by solid ice. I have surmised that
these shallow narrows may possibly represent an old river-course,
widened by sea-currents and by the action of waves. About Arch-
angel the shores of the White Sea are low and flat; and the country
appears to be drift, chiefly fine mud in the delta of the Dwina.
The sea is silting up. I found a great many large polished striated
boulders of very hard rock, many like those which I saw on the
shores of the Polar basin. According to the geological map and my
own observations these must be of northern origin, unless they
came down from the Ural mountains.

(15) Archangel to Astrakhan. Having seen rocks and boulders
about 70° N. from 25° to 40° E., I followed southwards between

Q.J.G.8. No. 120. 2x

458 J. F. CAMPBELL ON POLAR GLACIATION ETC.

40° and 50° E. from the White Sea to the Caspian over Russia, with
this result :—

In the entrance to the White Sea at the mouth of the Dwina, and
along its banks, I found beds of newer formations little disturbed and
covered with Boulder-clay and drift. I found numerous polished and
striated boulders of far older and harder rocks, like those of Russian
Lapland, in the drift as far as I followed the Dwina. Then we
landed and drove 500 versts (about 330 miles). A little north of
lat. 60° N. I came upon a tract like the country of drift-hills which
I saw about Knopio in Finland amongst lakes, and in the archipelago
of the Baltic along the Swedish coast. From long. 15° to 40° E. about
lat. 60° N. is arange of low drift-hills. Where I passed through them
along the Vaga in 1873 the breadth of this range is about 30 miles.
The hills are about 800 feet high at most. They have peculiar
moraine-like forms. Amongst them are detached cones and short
boat-shaped ridges through which the Vaga winds northwards.
Sections made by the river and by its feeders, and surfaces in fields
and roads, show that these mounds are chiefly made of sand, gravel,
and large stones. Most of the stones are rolled; but many are
glaciated ; all seem to have been carried from a region of older
rocks over these plains of newer formations, on which Boulder-clay
and drift are so thickly spread that solid rock is rarely seen. I saw
no rock in driving through these drift-hills*.

A Russian Engineer Officer told me that a range of low hills ex-
tends nearly to the Ural Mountains eastwards. I know that they reach
Sweden westwards about lat. 60° N.; I have passed them thrice.

(16) Result. If the range of drift-hills near Turin, about lat.
45° N., be the moraine of confluent Alpine glaciers, which Professor
Ramsay taught me, and I am willing to believe, then these drift-hills,
forming a large are about the circle of 60° N., may be the moraine of
that polar glacier for which Agassiz contended. I will not attempt
to account for them; but I have seen them in Finland and in Russia
and in the Baltic.

(17) Further south, nearer to Vologda, we passed a conical
mound of drift about 100 feet high in the midst of a plain. Possi-
bly this may be sepulchral. Tombs at Kertch are as large. On it
I found specimens of the gneiss which I had followed from Lapland
over about 10°. Between 60° and 55° N. lat. on the route to
Yaroslay, Moscow, and Nijnii Novgorod, I found numerous boulders
of kinds which I thus followed along 15 degrees of latitude. I was
told by travellers that similar stones abound westwards near the
sources of the Volga. I saw numerous low hills of drift about
Moscow ; but railway travelling by night is not conducive to geologi-
cal study. I suspect that another range of drift will be found about
A5° N. in Russia.

* Murchison’s ‘ Russia,’ vol. i. p.176. “The road which leads from Ust Vaga
to Usting runs for some miles at a short distance from the Dwina, and the great
thickness of the drift-sands which here encumber the surface prevented our
seeing the fundamental rocks.” It is a great satisfaction to find my observa-~
tions, independently made, confirmed by a great authority like Murchison.

J. F. CAMPBELL ON POLAR GLACIATION ETC. 459

(18) The Volga. South of lat. 55° N. and east of Nijnii, I did not
see one of the northern boulders which I was hunting southwards.
The southern limit of northern boulders on the geological map of
Murchison and his comrades, published in 1845, is marked by a
line which passes Nijnii Novgorod, where I saw the last of them
in 1873.

The Volga is more than 2400 miles long, and falls 633 feet
from its source to the Caspian. The right bank is commonly the
highest, and gives a section of the country nearly as long as the
river.

At Yaroslav the right bank is more than 100 feet high, while the
left is low as far as it is visible from the terrace.

At Nijnii Novgorod the right bank is nearly 200 feet high, and
is made of thick beds of stratified gravel, which are noticed in
Murchison’s ‘ Russia.’ The left bank is low as far as visible from
the terrace. At Saratov the right bank is 360 feet high by my
barometer. It is scarped and undercut; and landslips result. The
river is shallow and three miles wide ; and the left or meadow-bank
is a low plain, with a flat horizon distant more than twenty miles.
On a very fine clear day I could see no rising ground eastwards
from the scarped hill above Saratov, which is part of the country
into which the Volga cuts*. I watched this bank while I was
awake during the voyage of six days to Astrakhan, and I did not
see one glacial boulder to be sure of. This part of Central Europe
is covered by thick beds of fine drift, by stratified sands, gravels, and
earth, which rest on stratified rocks which have been little disturbed
since they were deposited. I have sketches to show the work of
rain and rivulets, and of the big river, and the structure of the high
bank.

Below Saratov beds of rock appear to be made of the same kinds
of mud which rivers now carry from the Urals, and from Western
Russia into the Caspian. At some places beds of different colours
alternate. The cliff bank is striped horizontally, and scored verti-
cally by water-courses, and has a strange appearance. I suppose
that the mud varied in colour as the eastern or western branches
were flooded. They have different colours now.

(19) Result. In America I lost the train of northern boulders .
about lat. 39° N., and found the fine drift which colours the Missis-
sippi and extends to the Gulf of Mexico. In Germany I lost the
northern boulders about 55°, 16 degrees further north.

T lost northern boulders in Russia about 56° N. at Nijnii Novgorod,
and followed gravel and mud to the Caucasus and lat.41° N. The
latitudesvary ; the drifts are alike. A Russian officer familiar with
Siberia told me that great blocks of granite are planted amongst
mounds of shells in sandy plains in the steppes of Northern Asia.
He had been railway-surveying. All these facts support the idea
of polar glaciation down to 70° and to 60° at least in Europe, to

* Why that river flowing southwards is working westwards is a question foreign
to the subject of this paper; but the Arctic current may explain the fact.
2x2

460 J. F. CAMPBELL ON POLAR GLACIATION ETC.

some unknown limit in Asia, and to 32° N. in America. Whether ice
flowed, or waded, or floated, it is needed to account for the transport
of large stones from north to south in Europe, Asia, and America.
The bed of the Volga proves that river-ice is not a sufficient
cause.

From 60° to 40° N. in Europe all the drift that I saw might have
been spread by water. But the drift gets finer as it gets further
from the Arctic basin. It may all be of northern origin, glacial at
first, sorted by water afterwards.

(20) The Caspian. As all the world knows, the Caspian has no
outlet and is lower than the ocean, though many large rivers, in-
cluding the Volga, flow into it. The waste is by evaporation; and
much rain falls in Persia about the high land at the southern
end of the Caspian. The water is salt, but not nearly so salt as
the ocean. Far out from the mouth of the Volga for about forty
miles the water is very shallow, and the bottom mud. The delta ends
and the lake deepens suddenly. Lake-steamers wait at the bank
for shallower river-steamers. The water is thick and dirty for a
long way south. Opposite to the end of the Caucasus the basin
deepens greatly ; and thereabouts are naphtha springs and volcanic
phenomena on shore. Manifestly the Caspian, like the White Sea,
is silting up with fine mud, washed off Europe and Asia and their
beds of old drift. Fish of many kinds abound and are caught in
the Volga. J saw many sorts, none quite like any fish previously
seen by me. “ Herrings” for example are more like Bream
for shape. The “Sterlet” has no bones. Since the canal was
cut to join the Volga and Dwina the Sterlet has got to the White
Sea. Sturgeon and Bellugu salted fill barges and are sent over
Russia. The New-Red Caspian formation now forming will have
many fossil fishes in it with ancient tails and bony plates outside.

(21) The Caucasus. Hypothesis. As ice and drift got so far in
the Old and New World, I supposed that glacial phenomena of the
Alps would recur in the Caucasus. These are my observations :—

From the Caspian to the Sea of Azov the range is roughly about
800 miles long. The highest peaks, Kasbeg and Elbrouz, are said
to be over 16,000 and 18,000 feet high.

(22) The Pass of Dariel. In crossing the range from north to
south from Vladikavkas to Tiflis, the road passes the foot of Kasbeg,
which the natives call the “Sword of God.” The road is as easy
as the Simplon. It mounts along the banks of a considerable stream
through a deep gorge which gives a cross section of the range.

On the 3rd of October I saw the Peak of Kasbeg and the range
from the north for more than 150 miles.

Approaching it we drove over undulating plains of clay and
passed a lot of large stones. Elsewhere I should have called them
glacial erratics; but I could find no scratches. In the evening I
walked to the right bank of the river, and found a great ridge of
clay which I took for a moraine; but even here I could find no
scratched stones. I believe it to be part of a delta. I sketched
and inspected brick-pits, and reluctantly gave up my Caucasian

J. F. CAMPBELL ON POLAR GLACIATION ETC. 461

glacial hypothesis, which had been rudely shaken before between
Petrovsk and Vladikavkas.

October 4th. We drove up a beautiful gorge with well-marked
terraces of rolled stones at the mouth of it, and with many very
large stones scattered about; but I saw nothing glacial in the
gorge. The rocks at first were limestone, with springs of clear
water. The limestone, which is much bent, so far as I could make
out from the carriage, overlies unconformably a series of beds.
dipping northwards at a high angle.

I noted in succession “ shales,” “flags,” “ slates,” ‘* mica-schist.”
Through all these beds the river has cut narrow gates of the usual
square pattern, while the side streams have cut the usual V-shaped
furrows. Up to 750 feet the road went over river-pebbles and gravels,
with an occasional fallen stone, as big as a house, near cliffs.

Then we got to igneous rock and to strata dipping north as
before. At Kasbeg Station, 2620 feet above Vladikavkas, we
stopped, and I sketched a high mountain-glen without a single
glacial mark in it that I could recognize. The water in the
river was cold, but certainly not a glacier-stream like those which
I have seen in the Alps, in Iceland, and in Norway.

October 5th. I sketched Kasbeg, which is a tall pyramid of rock
on which a considerable quantity of new snow lay, but so as to
show the rock and its structure. I saw nothing like a glacier from
below. We drove past Sionski Gori, which is all spires and pinnacles
like the rocks above Romsdal in Norway. At one place near the
station of Kobi is a bed of basalt (fig. 3,a) with bent pillars ; it is
nearly horizontal and unconformable to the rest of the strata, which
have the same northern dip. Rock-crystal is found and sold to
travellers. Near the top of the pass water is coloured red and
yellow by mineral springs. Close to the top is a bed of lme-
stone unconformable and nearly horizontal. From this runs a
spring which deposits a yellow stuff which I took for a wreath of
old snow. It is like the Sprudel deposit, and forms on grass and
stones and every thing, in layers like the stuff which forms about
the Geysirs in Iceland. At the summit my aneroid marked 22-800,
7-700 lower than at the Caspian, which gives about 6930 feet.
The height is said to be 8000 feet. From the top we trotted
merrily down into Asia, passing quartz-flags much shattered, and
got into a different climate and country. It was green, with lots
of hay. A plant with red leaves grew in clumps and made the
hill-side like a garden on a lawn. ‘The rivers ran clear.

We had got from the arid rainless country to a watered land.
We trotted rapidly down to 24-700, where we reached a village and
a clear river, and stopped. For some distance down the first pitch
the rocks have the same northerly dip. Near the river is a bed of
basalt (fig. 3, 6) which crops out along the hill-side. The pillars are
very large and crossed by joints or flows of the old lava. Near this
are beds of stone like the rock of Gulbrandsdal in Norway (Silurian ?).
Pebbles in the river are blue, with white crossing veins. The basalt
seems horizontal and unconformable.

462 J. F. CAMPBELL ON POLAR GLACIATION ETC.

Oct. 6th. Lower down we got to the same northerly dip and
followed it along the strike and again across it nearly to Tiflis.
The northerly dip appeared persistent from Balta to Mjeth, for nearly
fifty miles; and the sequence extends southwards from newer to
older rocks. My observations were confirmed by a Russian officer.

A mere traveller bent upon copying general forms could not see
geological details from the jolting springless carts which carried
him rapidly through the Caucasus in four days, allowing scant time
for drawing at stations. From Mjeth the road is on the strike for
a long distance. Near Tiflis the beds are much folded, the folds are
parallel to the main range, and the rocks are like folded beds of the
coal-formation which I have seen in the Alleghany Mountains. I
have drawn a rude section to fit the page (fig. 3), scale twenty miles
to the inch horizontally, exaggerated vertically. The extreme height
of Kasbeg would be expressed by about ,4, of an inch.

Fig. 3.—Section through the Dariel Pass.

(Horizontal scale 20 miles to 1 inch; vertical scale greatly exaggerated.)
a, bo. Basalt.

10----
.. 20----4- 2
30 Si
Honea
B0resenta
60----
70 ----

!

!

nm

a4 : g : : 9

i : : : : :
an :

is 3 g 3 oa § S

3 = 3 3 a) Be B

Gs =a nD

= 3 8 os] ° gs =

e ia) m i iN) nm =

The outlines of the mountains are due to weathering. Except
large stones, I could find no trace of glacial action in the whole
journey of 202 versts (about 1334 miles). Short of Tzalkan, not far
from Tiflis, we saw a lake (“‘ Bazaleti”’); near it is a conical mound
like a moraine. I saw, or thought I saw, some scratched stones
amongst a great many rolled stones, and I believed the lake to be
glacial. After a careful search in the valley lower down, all the
large stones that I could find were smooth water-worn pebbles
taken out of the clay, and out of great beds of rolled stones which
there make large hills. I noticed ploughing with fourteen and
eighteen oxen hereabouts. October 7, got to Tiflis*.

* Since my return I haye read Dr. Lyell’s book, and find that we noticed
many of the same things in the Pass of Dariel, such as the springs, the flowers,
the oxen, &. Klaproth, quoted in a note, mentions the rocks which I noticed.
Keith Johnstone’s Geological Map of Europe colours the Caucasus as Tertiary
and Crystalline.

Montpereux, ‘ Voyage autour du Caucase,’ 6 vols. octavo, and Atlas, 1843.
Tn this work is a section which includes the Pass of Dariel from Vladikavkas to
Tiffis. It confirms what I have said as to the dip and sequence of rocks, and
marks the lake of ‘‘ Bazaleti” on its pebble-beds.

J. F. CAMPBELL ON POLAR GLACIATION ETC. 463

(23) Daghestan (the mountain-country) is on the north side at
the Caspian end of the Caucasus. The rocks are chiefly the folded
limestones which we crossed in the pass of Dariel afterwards, and
apparently some of the dark shales on which they rest unconform-
ably. The limestones are very like beds which appear in the right
bank of the Volga. The long folds are parallel to the main range.
One bed of dark earthy shale contains large globular conere-
tions. We come to it repeatedly in Daghestan in deep water-
courses. Another bed consists of large hard rolled pebbles in a
matrix.

From the nature of the country, the clearness of the air, and the
absence of vegetation, the structure of Daghestan is easily seen.
The edges of the flags and schists and igneous rocks which we after-
wards passed near Kasbeg seem to form the crests of the whole
range, and furnish the boulders and pebbles which the largest rivers
roll down to the plains.

(24) West-end. From Tiflis to Poti on the south side of the
range, and along the shores of the Black Sea, the structure seen
near Tiflis appears to extend into the Crimea along the northern
‘shores of the Black Sea.

The lower hills are made of folded beds of newer stratified rocks,
the higher hills of the edges of harder beds upheaved. I will not
attempt to guess at the age of the rocks; but some appeared like
Mountain-Limestones, the Coal-measures, and superincumbent strata.
Coal has been found in the Caucasus, and is worked north of the Sea
of Azov and south of the Black Sea.

In any case a great thickness of old sedimentary beds have been
tilted up on edge, and a great thickness of newer beds have been
crumpled and folded like dough; and the key to Southern-Russian
geology is in the Caucasus, in the Gate of Asia, in the pass of
Dariel. The general disturbance may be found to coincide with the
geological disturbance which upheaved the Alps and other chains.
Between lat. 30° and 40° N. there is a marvellous coincidence in the
general direction of mountain-ranges in Europe and Asia.

(25) Denudation. Such is a rude outline of the geology which I
thought I saw while rapidly travelling through the Caucasus. The
shapes of the hills and hollows, which I studied and drew, clearly
are due to aqueous denudation (see fig. 4).

If ever glaciers worked in the range, their traces have been
almost entirely obliterated. Very little rain now falls in Daghestan
or anywhere on the northern slopes. The rivers which drain the
range are small. Houses have flat roofs; dresses are not contrived

In the map, Daghestan, which was conquered about eight years ago, is not
coloured.

Koch's Geological Map, 1850, does not colour Daghestan. The Dariel Pass
from north to south is coloured ‘‘Tertiares,” ‘‘Secondares Gestein,” and ‘Thon-
Schiefer.” Kasbeg, and the opposite mountain, are coloured Volcanic, including
therein ‘“ Basalt and Trachyte.”

T have also read Mr. Freshfield’s book, and his paper published by the Geo-
graphical Society, to which I refer below.

464. J. F, CAMPBELL ON POLAR GLACIATION ETC.

for rain; roads made by Russian engineers have no gutters, few
bridges, and no provision against floods. Nevertheless the whole
range for 800 miles is deeply furrowed by water-courses which are
plainly water-work, as furrows are on and in the crater of Monte
Nuovo. They make a pattern like the branch of a fir-tree, or a fern,
with the fine points towards the main range. While driving and
riding westwards for a distance of 512 versts (about 338 miles) from
Petrovsk to Vladikavkas, along the folds of limestone and the strike,
a rise and fall of a couple of thousand feet was a common incident.
We crossed a lofty plateau, scrambled down watercourses of gradually
increasing size, till we got to some rivulet or main river flowing
northwards; but only four bridges are on the road. Having crossed
a river, we climbed to another scarped plateau, and entered by some
queer rift or water-gate, as into a fortress. Then for eight or ten
miles we rode along a different, upland, grassy country till we got to
the edge of it, and plunged down into another arid gorge. I could
find no signs of glaciers even on these remnants of the old surface,
through which the water has dug a couple of thousand feet or more,
as it appeared to me.

Each of these plateaux is like an island inhabited by people who,
like Hebrideans of old, made war upon each other till they were
conquered. ‘Twenty-five languages are spoken in the Caucasus.
Through this country the Russians hunted Schamyl till he was
surrounded upon the Plateau of Gunib. His lower town was fired
into from the opposite plateau over a water-course some miles wide,
and a couple of thousand feet deep. By observation, I made the
bridge 3230 feet above the Caspian, Gunib 4230, the upper town

5130, and the upper edge of the plateau5715. The Russians scaled
‘a cliff to get to Gunib from the west side ; now they have made a
tunnel through it and a practicable road down the mountain-side to
Karadagh. The descent to the next river was equal to the ascent,
2485 feet. Speaking generally, there are no lakes. In all Daghes-
tan only one is known. It is on one of the plateaux, and is about a
couple of miles long. At the foot no stream escapes. The water
evaporates or oozes through a dam of angular gravel, which may be
a terminal moraine. At the head of the lake I found a small median
moraine nearly smothered in the delta of a tiny streamlet. The
lake is deep and clear, and full of trout; I believe that they were
put there by Russians. I could find no glaciated rocks anywhere
about the lake. The moraine is the only mark of glaciation that I
could identify in the whole range while travelling from end to end.
The air was so clear that I often saw a wall of mountains for more
than a hundred miles at a glance. I swept it often with a good
glass. I saw new snow in plenty, and got to new snow several
times, but I could see no blue glacier. The country has been ad-
mirably mapped. The officers at Tiflis only told me of two small
glaciers, which are to the north of the two highest peaks; and these
are marked. On the southern side of the range, I saw one other
small lake, as I have said. From lat. 40° to 45°.N., long. 50° to 35°
E., I could not find one rounded hill or hollow, one scratched rock or

J. F. CAMPBELL ON POLAR GLACIATION ETC. 465

Fig. 4.— Waterwork in the Caucasus.

Karadagh, canon, distant about 12 miles. Gunib.

466 J. F. CAMPBELL ON POLAR GLACIATION ETC.

stone, one perched block, one lake-basin certainly due to glacial
erosion, a glacier or the trace of one. The work of running water
is plain as on a railway embankment after a heavy fall. The
highest hills are jagged sierras, the lower hills pyramidal or scarped.
The valleys of all sizes are shaped like V or like Y (fig.5). The tail of
the Y is often a deep drain like those which I have described in ‘ Frost
and Fire.’ One I guessed at 500 to 600 feet deep. It is scarcely
wide enough for a couple of loaded horses to pass, and it is all water-
work. It is called “ the tunnel of Karadagh,” and is the drain for
a great tract of country, and the road used by its inhabitants.
When I rode through it there was hardly water enough in the
river to fill a London main pipe, and passengers were numerous.
Passages of the same kind abound in Daghestan, and generally are
defended by villages and used as gates. Familiar with glacial
marks, and fresh from Norway and Lapland, where they abound,
the contrast between glacial and aqueous denudations was very
striking.

(26) Rise of land. As I have said, the fall of the Volga from
source to delta is given at 633 feet. Above Petrovsk, on the Caspian,
is a sloping deposit of large rolled stones and sand, which fills up a
hollow between an isolated hill and the first range of the Caucasus.
No river is near to account for this, which I suppose to be a “ raised
sea-margin.”” It seems to bound the northern plains, which fade
into the Caspian. The isthmus at Petrovsk, and the drift-slopes
along the northern range, are more than 500 feet above the Caspian
and the plains of mud which separate it from the Black Sea. If
water stood at the levels marked by raised sea-margins in Scandi-
nayia and in the Caucasus, the Caspian, the Black and White Seas,
and the Baltic would be joined, and most of Europe would be sunk.
Looking to the drift-section on the high right bank of the Volga, I
think it probable that all these European plains were in fact sub-
merged at a late geological period. I found terraces of large rolled
stones elsewhere in the Caucasus at higher levels; but some I attri-
buted to the action of rivers*.

(27) Result. If Kurope was in fact submerged, the circulation
of ocean-currents in the Atlantic is pertinent to the distribution of
northern polar drift in Russia.

(28) Hypothesis. I thought it possible that northern drift might
be found in the low gap which contains the sea of Azov and the
Straits of Zenikali, lat. 45° N. Icebergs do, in fact, reach lat. 37°
N. in the Atlantic now ; and the Crimean winter is severe enough to
make ice-rafts. Northern drift and large boulders reach 39° N. in
America.

(29) Fact. Arrived at Kertch, I saw numerous conical mounds
placed singly or in groups and ranged in rows along the low sky-
line. I was told by different people on board the steamer, and on
shore, that these mounds were “ natural,” that they were “ volcanic,”

a8 IMG page 297 of his great work, Murchison devotes a section to the relies of
a great former eastern Mediterranean Sea which covered these regions and de-
posited the older and newer Caspian rocks, which are marked upon his map.

J. F, CAMPBELL ON POLAR GLACIATION ETC. 467

that they were made of “conglomerate.” I began to think that
they might be “glacial.” They are in fact sepulchral. Hewn-
stone tombs, as big as small churches, are buried under mounds of
earth. Many have been opened, and treasures of art have been
taken from the “ Kourgans” of Kertch. I sought “ drift ” in the
Crimea and found none. I found naphtha-springs, which are volcanic.
They abound, and the naphtha is used as fuel on board steam-boats.

(380) The Black Sea. The northern shores of the Black Sea differ
materially from the sea-coasts further north. Between Poti and
Odessa there are no deep-sea lochs or fjords like those of Scotland
and Norway. There are few bays or indentations, very few har-
bours or islands. No sinking of mountains would change the cha-
racter of the coast in these respects. The coasts appear to have
risen, unless the water-level has fallen. There is no appreciable
tide ; but where land is low, the beach suggests that it is always ebb
and a perpetual spring-tide. Salt lakes abound. These seem to
result from the formation of bars at the mouths of rivers, bays, and
inlets, now ebbed nearly dry. All the plains and islands that I saw
looked like alluvial flats left high and dry. The lofty Crimean coast
seemed to be a sea-cliff, broken out of a small bit of the Caucasus
and made of like materials. I attributed the difference in this
coast-line to the absence of that glacial erosion to which I attri-
bute “ rock-grooves ” as well as “ lake-basins,” ridges left as well as
furrows made, in Scotland and Norway.

(31) Bosphorus. I was assured that miniature arctic and equa-
torial currents flow in and out through the Bosphorus; and I fully
believed what I was told. A passenger assured me that he felt, and
that I ought to feel, the difference in temperature on opposite ‘sides
of the steamer as we entered from the Black Sea. I saw vessels at
anchor heading opposite ways, and vessels under weigh dodging the
stream. I was ready to believe, but I wanted proof.

Oct. 31st. I got a boat and crossed to Asia from Constantinople,
and found the surface-temperature 65° from shore to shore. I found
eddies in plenty, but no opposite surface-currents. With southerly
winds warmer water comes in from the sea of Marmora, according
to the Turkish boatmen. I went to a high hill-top, on the Asian side,
and found nothing glacial anywhere about the Bosphorus. I sought
drift, and found none that could not be explained by the action of
water on the hills. What I saw from the hill might have been a
bit of a great river. It was like Niagara below the falls, the St.
Lawrence near Montreal, the Rhine near Bonn, the Volga near
Saratov, or the Danube near Vienna; a stream flowing through a
wide plain with a few hills in it, all seamed by water-courses. If
the Bosphorus-hollow were now filled to the height of the right
bank of the Volga at Saratov (860 feet), the Black Sea would be
forced back into the Caspian, and the perpetual ebb would be
changed to high water. I am strongly of opinion that the Bosphorus
is, in fact, a drain cut by running water in very late geological
times, and that the Black-Sea level has been lowered by the cutting
of the drain.

468 J. F, CAMPBELL ON POLAR GLACIATION ETC,

(32) Greek Archipelago. Amongst the Greek islands I saw forms
due to volcanic action. I saw volcanic cones in a latitude in which
volcanic rocks and active volcanoes abound from Persia to Sicily ; but
everywhere I saw water-courses on the hill-sides and waye-marks at
the sea-margin. About the South Cape, November 3rd, I saw none of
the glacial marks which I had seen at the North Cape on the 4th of
August, 1873, and in America, near Washington and St. Louis in 1864.

(33) Italy. At Naples I saw voleanoes old and new, and climbed
Vesuvius for the second time after an interval of thirty-two years.
I saw lakes in volcanic craters at Naples and near Rome, and water-
courses everywhere, but I saw no sign of a lake-basin or valley due
to glacial erosion in Italy. I climbed the hills above Genoa and
looked down into large watercourses like those of the Caucasus. I
saw nothing that I could identify as glacial work about Nice,
Cannes, and the hills behind that sunny coast, with its palm-trees
and cacti. The last polar glacial marks that I saw in Europe were
at Nijnu Novgorod, about lat. 56° N.

(34) Result. My observations support the extension of polar
glaciation to lat. 56° N. in the east of Europe, to 55° in Germany, to
nearly 50° in Britain, to 39° in America. The Alpine and Pyre-
rean systems I now suppose to have been separate local systems.
The absence of glaciation in the Caucasus I cannot explain by any-
thing but ocean circulation and atmospheric causes.

(385) Volcanic Phenomena. The enormous disturbance of strati-
fied rocks along ranges of mountains like the Caucasus, and the Bal-
kans, and the Alps, I can only account for by the shrinking of the
earth’s mass and lateral crushing of the crust. The varnish on a

globe is thicker in proportion and is disturbed when the globe shrinks
‘or expands. The undisturbed condition of the northern regions of
Kurope, and the enormous development of igneous rocks south of the
Caucasus can only be explained, as I suppose, by something like a
fracture equal to the length of the range, which is chiefly made of
sedimentary beds lifted like the lid of a box.

(36) Sea-beds. I wish to disclaim the notion, which I never
entertained, that all valleys are “big ice-scratches,” and that all
hollows are due to glacial erosion. Forces which heaved up ‘the
earth’s sedimentary crust in the Himalayas, in the Caucasus, in the
Crimea, Balkan, Alps, and Pyrenees, which also folded the beds of
the Coal-formation and threw them into basins and troughs, synclinal
as well as anticlinal curves—forces which are raising and sinking
whole countries now, are sufficient to account for ocean-beds and for
the Caspian and other basins. But denudation alone accounts for
drift, and for the materials of sedimentary rocks. I seem to have
passed through a sedimentary series nearly forty miles thick in the
Pass of Dariel. The earth’s surface has been ground down to the
extent of the sum of sedimentary beds. Glacial drift, wherever
found, proves glacial erosion, equal to the mass of drift whatever that
may be. The masses of drift which I have seen in Europe and in
America prove glaciation on something approaching the scale for
which Agassiz contended.

J. F. CAMPBELL ON POLAR GLACIATION ETC. 469

(37) Conclusion. I hold with Mr. Clifton Ward and with Profes-
sor Ramsay, who stated the theory long ago, that certain lake-basins
are ‘due to glacial erosion,” and also that many valleys and many
hills between them are chiefly due to the same engines which also
formed glacial drift*. I have endeavoured to explain what I mean by
« denudation ” and “ ice-marks,” and to describe what have I seen in
the writings which I have quoted above. To them I must beg to refer
for details, and for justification of my venturing to address you
on polar glaciation. I cannot yet see my way to a general ice-cap
reaching nearly to the equator.

(38) Travelling. Allow me to say, in conclusion, that through-
out Russia I was treated kindly, courteously, and hospitably. I
was asked for my passport only on arriving at Archangel, and on
starting from Odessa. I had no difficulties or adventures. I en-
dured few hardships, and enjoyed magnificent, bright, calm, hot
weather for five months.

Tho best months for travelling along the route are those which I
selected. The best months for the Caucasus are September, October,
November, and part of December.

The Russians wish travellers to explore; and I believe that En-
glish geologists will be especially welcome visitors. Many officers
especially mentioned the Alpine Club, and hoped to see more of its
members.

(39) Since my return, while writing this paper, I have read Mr.
Freshfield’s paper in vol. xxxix. of the ‘Transactions of the Geogra-
phical Society,’ and his book, ‘ Caucasus and Bashan,’ Longmans,
1869. Our facts may seem to be at variance as to Caucasian glaciers ;
but they may be reconciled. Mr. Freshfield and his comrades climbed ;
I did not. He saw the white side of the shield, I the yellow. He
travelled along the ridge from Kasbeg to Elbrouz, at the rainy end ;
I travelled chiefly at the rainless eastern end of the Caucasus. He
travelled in June and July. I passed Kasbeg in October after sum-
mer, before winter snows. In his book, and in my journal, Kasbeg,
seen from the Station, is depicted as a pyramid of rock, thinly
covered with snow. Possibly the unmelted snows may have led Mr.
Freshfield to an enlarged estimate of the size of the formidable
glaciers on which he climbed during the adventurous journey which
he so well describes. His largest glacier was only 500 yards wide.
In Daghestan valleys are often filled with great masses of drift, the
accumulations of many confluent deltas into which the streams have
cut deeply. Had I seen there unusual waving alluvial deposits
covered with thick snow, while standing upon a glacier myself, I
should have noted them as larger glaciers. I did, in fact, set them
down for moraines when I first saw them. I made a careful sketch
of one such deposit from the plateau above Gunib, believing it to be
glacial. It was only after riding over a great many similar deposits,
and after passing Kasbeg and the ridge, that I was driven from my
Caucasian glacial hypothesis. Between Tiflis and Poti, Petrovsk and
Vladikavkas, | saw no glacier stream. At Soukhum Kali on the

* See footnote on p. 452.

470 J. F. CAMPBELL ON POLAR GLACIATION ETC.

Black Sea, where I landed, I could find no boulders in a stream
which comes from near Elbrouz; and I could see none from the
steamer along the whole coast to Kertch. The glaciers trodden by
Mr. Freshfield I could not possibly see; for they are on the north
side of the range between Kasbeg and Elbrouz. They are marked on
the Russian map. We saw different regions, and the same region
under very different aspects, and we need not fall out like the
knights of old who quarrelled about the gold and silver shield.
What the Russians know of their country is on their Caucasian
map.

Postsoript.—April 27, 1874.

Mr. Belt, whose very interesting book * proves his accuracy in
observing nature, describes (pp. 248-263) marks which he took for
proofs of the action of land-glaciers about lat. 14° N. in Nicaragua.
He says, “The evidences of glacial action between Depilto and
Ocotal were, with one exception, as clear as in any Welsh or High-
land valley.” The exception was scratched rocks. The lowest
marks noted were still 2000 feet above the sea: Believing, as he
does, in “evolution,” and in cold which locked up water enough on
land to lower the whole surface of the ocean 1000 feet at least,
Mr. Belt supposes that tropical hfe survived the cold in wet low
erounds left dry by evaporation. These ancient low lands now are
at the bottom of the salt sea, beneath 1000 feet of melted ice which
turned to fresh water.

Professor Agassiz and his followers go much further. After this
paper on polar glaciation was sent in, a copy of ‘The New York
Tribune’ of December 30, 1873, was sent to me by some person
unknown. On the 15th of April I spoke of the newspaper. It
contains ‘‘A series of six lectures by Professor Agassiz,” which
seem to have been printed from short-hand notes. It is stated that
the first of the series was delivered Feb. 5, 1867, before the Cooper
Institute at New York, after the author’s return from South
America.

It is stated as proved that glaciers covered the whole of temperate
North America, as well as “the whole of Europe.” In Maine the
ice was 12,000 to 13,000 feet thick; and further south glacier
marks were identified. The author concluded at last that the
valley of the Amazon, about the equator, was filled by a vast
glacier which came down from the Andes and went into the
Atlantic, the ice, perhaps, then covering the sea to such an extent
that it is a question whether any open water was left at the
equator then, as it is a question whether there now is open water
at the Pole. But Agassiz found no scratched rocks in the valley of
the Amazon. He is reported to have said, “‘ and if this be so, you
see at once how this intense cold must have modified the surface of
the globe to the extent of excluding life from the surface * * *,
and preparing the surface of the earth for the new creatures which

* The Naturalist in Nicaragua. By Thomas Belt. Murray: 1874.

J. F. CAMPBELL ON POLAR GLACIATION ETC. A471

now exist upon it.” He pointed his argument directly at the
theory of evolution which “is cut at the root by this winter which
put an end to all living beings on the surface of the globe (ap-
plause).”

Any thing said by Professor Agassiz deserves that grave considera-
tion and respect which is due to the memory of a very distinguished
man. These lectures, which were much and deservedly applauded,
contain an epitome of nearly all that has been denied and accepted
by glacialists and geologists within the last 40 years.

Agassiz, those who worked with him, those who followed them,
and those who went on separate paths to their own conclusions
about glaciation, have all begun with the Swiss glaciers, which are so
well described in these lectures. By reasoning from admitted facts
a few have convinced themselves of the recurrence of glacial periods
due to astronomical causes. The complete smothering of the whole
earth in a crust ef ice many thousands of feet thick periodically,
and the consequent periodical destruction of all life on the earth’s
surface, is one of those advanced glacial theories which few accept,
but which cannot be ignored by geologists. The facts on which
theories are founded may be tested; and I have endeavoured to state
the result of my own observations as to Polar glaciation. They go
to prove that the whole of Europe “was not covered by a polar
glacier.”

P.S.—May 1, 1874.—Caucasian Glaciers.

Professor Abich, of Tiflis, an eminent geologist, published in
1870 part of a work on Caucasian glaciers of which I have not got
the sequel. The pamphlet, which was lent to me in the end of
April, contains two views and a section of the Devdoraki glacier.
It is to the N.W. of Kasbeg, in the shadow, and high up. It is
mentioned by Mr. Freshfield. It bursts periodically and sends a
wet avalanche of rubbish into the pass. In 1867 it threatened to
burst again, and damaged the post-road ; therefore it was again sur-
veyed by Professor Abich, who surveyed it in 1861. The conditions
of its growth appear to be these. The top of Kasbeg is more than
16,000 feet high. A very steep cone of more than 6000 feet sheds
avalanches which knead themselves into névé, and produce
glaciers of pure ice with “dirt bands,” which flow and tear and
regelate according to their ascertained laws of motion. One “of
2nd order” named “of Stephen Zminda”’ ends at 9504 feet above
the sea opposite to the Station. The largest ends at 7374 feet above
the sea. In 1867 one branch of this had broken and fallen into
the other, which had grown much since last surveyed. It was
about 350 yards wide, about a couple of miles long, and about
300 feet deep at the end, according to estimate. The end could not
be measured, because of falling stones. The surface was found to be
moving at increased rates, faster than the whole glacier, which is on
a slope of 17°. The views show that the sides of the valley are
deeply furrowed by V-shaped water-courses. The drainage of this
system of “ corries”’ flows into the Tarek at right angles, four kilo-

472 J. F. CAMPBELL ON POLAR GLACIATION ETC.

metres below the Station, which is opposite to the peak of Kasbeg.
From the peak to the junction is hardly five miles on the map.

The glacier-stream there in October 1873 was a rill lost amongst
a great heap of débris, a couple of hundred feet high at least, which
seems to have crossed the valley at some time, so as to block the
main streams and forma lake above the dam. This small glacier is
marked on the Russian map, scale 10 versts to the inch. Thirteen
small glaciers are also marked on it. All these are on the north
side of the ridge, in the shadow, high up, and to the west of Kas-
beg, where the range is highest and nearest to the Black Sea. The
largest glacier, which is mentioned by Mr. Freshfield, and was
described by Klaproth long ago, is mapped as five miles long.
Probably the present dearth of large glaciers in this lofty region is
due to its inland position. South-west winds are dried before they
reach the Caucasus, and nearly drained before they reach the eastern
end, where evaporation is in excess.

Between the Caucasus and Ararat, whose latitude corresponds to
the present limit of Atlantic drift-ice, are numerous large sheets of
water in low grounds. I have not seen them. They have nothing
to do with the drainage of the Caucasian highlands. That gathers
uniform main streams, on which are no lake-basins like those of the
Alps and Scandinavia. Of these streams two flow into the Caspian
and evaporate, two join the Mediterranean.

Professor Abich writes very cautiously about tracks of glaciers in
a glacial period. He founds his opinion upon certain conspicuous
stones, which I also noticed while travelling once along the road,
and upon much local knowledge gained by himself, and by officers,
surveyors, and engineers who conquered and mapped the country and
made the roads. They are making railroads now. The region is sub-
ject to earthquakes and torrents. Professor Abich describes a case
in which stones from 200 to 300 feet in circumference were shaken
down, and moved more than 12 kilometres from a “cirque” in
Ararat by a single débacle of snow, ice, water, and mud during an
earthquake in 1840. A glacier now descends from the top of
Ararat to a level of 9000 feet, in the cirque at the head of the valley,
in which the delta of this sudden mud torrent now has the shape
and semblance of a “ moraine.”

These great Ararat “erratics” are equalled in size by only one Cau-
casian block, named the Stone of Jermoloy, which is 3531 feet above
the sea, near Lars, in the pass of Dariel, five miles below the broken
moraine-dam above mentioned. I took it to be a fallen stone when
I sketched it ; but it has been identified with granite nearer to the
base of Kasbeg, whose top is 10 miles from Lars on the map. This
erratic is equalled in size by a stone on the Unter-Aar glacier, under
which nine of us slept on the 15th of September, 1841. That is slowly
moving on the glacier, and must have gone more than two miles by 1874.
Therefore ice or mud may have carried the stone of Jermolov five
or six miles down hill (10,000 feet) to its present place. These
Caucasian stones are equalled in size by blocks on the Jura, in
Russia, in Lapland, in Scotland and Ireland, and in America, which

J. F. CAMPBELL ON POLAR GLACIATION ETC. A473

have somehow crossed deep valleys and arms of the sea where mud
torrents are out of the question. Professor Abich concludes, with
hesitation, that a Kasbeg glacier carried this erratic so far (five or
six miles) down the course of the Tarek. The pass is a V-shaped
valley of erosion with many large branches. The main stem is
about 40 miles long, three or four thousand feet deep, and at the
bottom it is little wider than the water-course. It is very unlike
any glaciated valley that I have-seen. The drainage of this whole
“basin” escapes through a narrow rock-gorge three or four thou-
sand feet deep; I could see no ice-marks in it ; Professor Abich did.
Every delta must bear some proportion to the mass eroded. In
this case the wedges, which have somehow been carried out of the
solid, have been somehow crumbled and rolled and carried through
the gorge. The débris, or part of it, is spread in a fan-shape out-
side of the narrow gap. The Tarek is cutting into the delta, and
washing the mud into the Caspian Sea. Professor Abich notes a
bed of blue clay with “ Mytilus polymorphus” (a freshwater
bivalve), and sandy clays, low down in the delta-section, and large
stones, granite, schist, trachyte, &c., of which a few only are
scratched, which haye somehow been rolled or carried from the
Kasbeg region, and now rest on the delta or are buried in it, far
outside the gap, down to a level of 1860 feet above the sea. These
stones are exceptional. The country is chiefly mud and shingle.
He concludes, with hesitation, that a Kasbeg glacier came out of
the gap, split on the delta, and carried stones to the right.and left
to places which are about 35 miles from the gap, according to the
Russian survey (of 10 versts to the inch).

The low country suggests arrangement of mud in water.

If, as I have supposed, the plains of Europe were submerged, or
if a lake was at the base of the Caucasus when the shells lived,
then ice-rafts like those which now grow in the sea near Kertch
would suffice for the dispersion of stones carried part of the way by
a glacier and by sudden floods towards the Tarek delta where they
rest. Floods might account for most of them, as I suppose.

But supposing that part of this exceptional delta is a buried
moraine of very unusual form and structure, and that one Cau-
casian glacier did extend 35 miles beyond the gap, and 5000 feet
lower than glaciers do now, to a level of 1800 feet or 1000 feet, that
does not affect my argument about “ polar glaciation.” I saw no
large erratics elsewhere about the base of the Caucasus, none in
other gorges. I saw nothing between Nijnii Novgorod and Vladi-
kavkas to suggest that all Europe had been covered by the polar ice
flood, which was 13,000 feet thick in the latitude of the Caucasus
in North America, which reached the equator, and was as wide as
the world.

I have seen no track of that glacier in crossing meridians between
the Volga and the Mississippi; but I did see the Arctic current at
work in “ drift ” latitudes. I found scratched granite boulders in
Russia, 700 miles from the nearest possible source. But to prove
the “ice-cap” I ought to have found them 900 miles further, in

Q.J.G.8. No. 120. 21

474 J. F. CAMPBELL ON POLAR GLACIATION ETC.

Daghestan, and about the base of Kasbeg, 8000 feet up, where
the “« mountain-ignoring ” cap, 13,000 feet deep, flowed over the Cau-
casian dam.

Somebody ought to find polar drift in Persia, and India, and
Ceylon, in Turkey, Syria and Arabia, in Greece and Egypt, and
about the sources of the Nile. I found Finnish granite in Northern
Germany. To support the ice-cap, samples of northern drift ought
to be found in the Rhine valley and in France, high up on the
shoulders of the Alps and Pyrenees, in Italy and Spain, in Northern
Africa and down to Coomassie. European polar drift ought to
spread at the Equator between the meridians which cross Scandi-
navia and the Ural Mountains.

I found glaciated stones near St. Louis which must have travelled at
least 800 miles southwards in America. To provethe “ice-cap” some-
body ought to find specimens of northern rocks upon the shoulders
of the Rocky and Alleghany Mountains, and all the way to the
Amazon in continuous streams, forming the moraines of the Missis-
sippi-valley glacier, which must have existed when the Amazon
glacier reached the sea. But even if all that is found, if the
severed “‘ boulder” on the top of Adam’s Peak in Ceylon turns out
to be an “erratic,” it has to be proved that it was not carried part
of the way by polar currents and ice rafts like those which exist,
which reach 37° in the Atlantic, and coincide with drift-latitudes on
both shores of the Atlantic basin.

Summary. — Evaporation and condensation, atmospheric and
oceanic circulation are caused by unequal local temperatures.
Rivers and land-glaciers alike result from atmospheric circulation,
and from local condensation, and they flow for the same reasons at
different rates. The existing Arctic current is part of ocean circu-
lation, caused by unequal polar and equatorial temperatures. That
current is a conspicuous fact. It is far larger and more powerful
than any existing river or ice flow on land. It is more than three
thousand miles long, and it is very broad; it moves vast fragments of
the largest existing land-glaciers, and great sheets of thick polar
sea-ice, at great speed. It lowers the perpetual snow-limit where
it flows, and it there modifies climate and all that depends on
climate. Within an area equal to that of India, in Greenland it
increases local condensation, and snow-flakes, and snow-falls, and
ice flows and rivers; and so it increases local denudation on shore
while it works on the shallows. Ice rafts in the ocean-stream carry
drift southwards ; and the water which drags these loaded rafts over
shallows and makes them drop boulders, also rolls and packs mud
and shingle and builds a delta in the sea. This kind of polar
* glaciation” and deposition of “ drift,” caused directly and indirectly
by one of the northern polar currents, cannot be denied. The
agent is a true cause at work; and the work ought to be like that
which I saw in Russia in 1873.

Where a sedimentary rock exists there water has been; and the
drift-plains of Russia are spread upon sedimentary series little dis-
turbed. That part of the earth’s crust has been up and down many

J. F. CAMPBELL ON POLAR GLACIATION ELC. 475

times without breaking much. If glacial periods, due to astronomical
or other causes, have in fact increased evaporation and condensation,
ice-flows, and rivers to any given extent, then polar currents must
have gained power during these periods, because movements in air
and ocean result from unequal local temperatures, on which also
depend the local growth of ice on land.

Problem.— Are the clear traces of ancient glaciation which
abound caused directly and indirectly by polar currents like those
which exist and modify climate, and which must have existed ever
since there was an ocean to circulate upon a revolving round world
warmed by the sun? Or, secondly, are these marks made by polar
ice-flows which do not exist, whose existence has been inferred from
these marks ?

I have contributed my store of facts. Those Members of the
Society who care for this inquiry can form opinions upon all the
facts known to them. If all the knowledge of the Society could be
brought to bear on a point by a debate and a vote, we might advance
geology by expressing a joint opinion upon the nature and extent of
that ancient polar glaciation which is abundantly proved by acknow-
ledged facts.

Postscript dated Salt Lake City, August 12, 1874.

I have now crossed the glacial current of the Atlantic and the
Rocky Mountains, and I have skirted the back of this range south-
wards on the eastern side for 200 miles, seeking for Mississipi mo-
raines. As in Europe, so in America, I have found scratched rocks
and northern drift carried far south, but on certain meridians only;
I have found ancient sea-margins, or something very like them, on
both sides of the Rocky Mountains and at corresponding levels.
I have seen ancient lake-margins corresponding to the wearing
done by the rivers which flow out of the lakes. I have seen lakes
drained and lakes filled up, and all known signs of atmospheric
action in great abundance and on a large scale.

This salt lake is the equivalent of the Caspian, in latitude and in
lack of outlet. As atthe Caspian, so here, the mountains are water-
worn and burrowed by “ cafions,” but a conspicuous old-level
water-line is on the hills at a level which would give the lake an
outlet. Lakes through which waters flood into this salt lake are
sweet like the mountain-streams. The rocks about it are freestone
and coal formation, &c. The people here suppose that this salt lake
is sea~water at 4340 feet above the sea, and a mark of the elevation
of the whole country. I attribute the spread of northern drift on
American and on Russian plains to ancient arctic currents.

There is nothing to indicate any glacial action at all between
Chicago and Salt Lake. In rising from 726 feet above the sea to
8242 feet (the highest point on the railroad), and in descending to
4340 feet, I could see no single mark of ice-action in the region.
For 164° of longitude, about lat. 40° N, from the Volga to this place,
I have found nothing to support the idea that solid ice extended

21L2

476 J. F. CAMPBELL ON POLAR GLACIATION ETC.

from the pole to the equator, and acted as smaller glaciers do on
rocks under and about them. But on “ Pike’s Peak,” which is
14,000 feet above the sea, I found weathered marks of local glaciers,
which carried granite to the plains for some fifteen miles, and 8000
feet down hill. In latitude and altitude and size these nearly
corresponded to the Caucasian glaciers about Elbrouz and Kasbeg
which still exist. The glacial traces which I have found in
America seem to indicate the transfer of oceans with their systems
of circulation from one part of the world to another, by the elevation
or depression of land. I have found nothing to indicate an ice-cap
in these latitudes thus far on my way round the world.

Discussion.

Mr. Bett stated that he had been over much of the same ground
as the author, but had come to quite opposite conclusions. He
agreed with Mr. Campbell that boulders only come down to a
certain point, but he considered that the very plains themselves are
proofs of glacial action. He thought that the absence of boulders
is only a proof that there was no chain behind which could supply
the requisite rock ; and in this case the only signs of the existence
of a polar ice-cap would be due to the blocking up of the drainage
by it, resulting in the formation of the plains of Siberia.

Prof. Ramsay thought that the ideas put forward by the author
were in accordance with those of previous writers ; but he considered
that he exaggerated the power of ice in the shape of icebergs to
effect changes. The question was whether there were ice-caps
moving towards the equator, or whether the configuration of the
mountain-regions might have produced the observed effects. He
expressed himself satisfied that the present configuration would
account, at least to a great extent, for the changes which have taken
place. The boulders found on the great plain of Russia might have
been conveyed either directly by glaciers, or by icebergs broken off
the ice-cap itself. Boulders have been seen 40 miles north of the
Caucasus, proving the existence there of great ancient glaciers. The
absence of boulders on the plains of Siberia was, he thought, to be
accounted for by the absence to the north of Siberia of high land
from which such boulders could be carried.

Mr. Drew thought that sufficient consideration had not been
given to the relative levels of the old and new glaciers. In Cash-
mere he had observed a difference of 5000-6000 feet.

Prof. Huenes thought that the theory of ice-caps spreading in
both hemispheres from the poles to near the equator hardly de-
served discussion, seeing that no facts which could not be otherwise
explained had been adduced in support of it, whilst it involved
great physical difficulties, and was quite inconsistent with the con-
tinuity of the forms of life from pre- to postglacial times. The
question before them seemed to be whether a smaller general drift-
ing from northern circumpolar regions, either of land or floating ice,
would better account for the phenomena observed by the author than

J. F. CAMPBELL ON POLAR GLACIATION ETC. AT7

a dispersion of boulder-bearing ice in all directions from local centres,
especially in the case before them from North Scandinavia. He
always understood that the boulders of N. Germany and W. Russia
could be traced to that mountain-district, and that there was proof
that the ice travelled to the north as well as to the south. Unless,
therefore, the author showed that some of the boulders could have
been derived only from circumpolar regions, he could not see the
necessity of calling in any thing more than changes of level of various
parts of the northern hemisphere along well-known lines of elevation
and depression to explain all the phenomena observed. He ques-
tioned the accuracy of the view that glacial conditions prevailed at
the same time over the whole of even one hemisphere, and, referring
to the observations of Mr. Drew, pointed out that, if the glaciers of
any mountain-region were shown to have once descended from 3000
to 6000 feet lower than the present ice-foot, it was only necessary
to raise that chain from 3000 to 6000 feet to make the glaciers
descend to their ancient level again.

Mr. Wuiraxer thought that the broad points of the paper had
been lost sight of. The rounded configuration of rocks in Norway
he regarded as clearly of glacial origin, but as the effect rather of a
great extension of glaciers than of a true polar ice-cap. In Britain
the glacial drift towards its southern limit is almost wholly marine,
and certainly not due to the action of land-ice ; so that it is di-
stinctly opposed to the notion of the southern extension of the ice-
cap. He could not believe in the existence of such a cap extend-
ing as far as the equator. .

Mr. Torrey thought that Mr. Campbell’s views as to permanent
areas of glaciation and non-glaciation due to cold and warm cur-
rents did not suffice to explain the facts; nor did he think that
a warm current could have passed over the plains of Russia. He
remarked that there is no drift in the south of Europe, and that a
line running nearly in the latitude of Dover would mark the southern
limit of the drift.

Mr. Prestwicu said that he felt much interested in Mr. Camp-
bell’s paper, and recognized that the difficulty in this case lay in
determining the boundary between the action of land- and sea-ice.
He thought that the existence of a Gulf-stream traversing eastern
Europe was not proved. In the eastern, and possibly even in the
southern, counties of Great Britain there is evidence of submergence
in glacial times; and marine drifts also exist in Brittany. Ifnorthern
France were submerged, ingress might be afforded to warmer water.

Mr. Branrorp considered that the difference in the faunas of the
Black Sea and of the Caspian was so great as to furnish a strong
objection to there having been a branch of the Gulf-stream passing
on towards Siberia at so late a geological period, as this must have
brought about a closer resemblance between the faunas. He ob-
jected to blocks found at a considerable distance from mountains
being regarded as true ice-borne boulders unless they were striated
or polished, seeing that blocks three or four feet in diameter might
be seen in Persia, which had been carried five miles or more by the

A783 J. FB. CAMPBELL ON POLAR GLACIATION ETC.

water of insignificant streams. In Persia the country, although
greatly elevated above the sea-level, is covered with drift; but he
had found no signs of striation on the pebbles, nor had he been able
to detect glacial markings on extensive plateaux more than 6000 feet
above the sea, with peaks rising to 12000 feet and even more. In
his opinion the ice-sheets were entirely local.

Mr. Evans remarked that the question of glaciation had always
been a puzzle. He thought that the notion of great ice-caps simul-
taneously covering the northern and southern hemispheres could not
be maintained, since it involved the destruction and new creation
of the entire fauna and flora. If the boulders referred to by Mr.
Campbell were of northern origin, there must have been a flow of
water from north to south, possibly in a great bay bounded by the
Ural Mountains on the east. It seemed to him difficult to conceive
how a current from the Atlantic could have passed over the high
lands of Turkey and the Carpathians. He thought the question as
to the glaciation of the Caucasus of much interest. It was difficult
to conceive how so yast a denudation as that described could have
occurred without either heavy rains or glaciers; but no rains occur
there at present. An alteration of level, however, would of course
affect the temperature and rainfall.

Mr. CampseELtL, in reply, stated that he had seen it given as Prof.
Agassiz’s opinion that the ice-caps had extinguished all life. He
remarked that the boulders have travelled at least 500 miles from
their original bed over level ground.

THE GLACIER-EROSION THEORY OF LAKE-BASINS, 479

36. Nores on the Upper Eneapine and the rattan VaLuEys of
Monre Rosa, and their Revarion to the GuAcIER-EROSION THEORY
of Laxe-Basins. By the Rey. T. G. Bonney, M.A., F.G.S., Fellow
and Tutor of St. John’s College, Cambridge. (Read May 27, 1874.)

THis paper may be regarded as a continuation of two which I have
already had the honour to communicate to the Geological Society * ;
and with a view of preventing misapprehension, I think it will be
well to state clearly what I am seeking to demonstrate, and the
mode in which I am endeavouring to effect my purpose.

As I have already stated, I am unable to accept the theory which
was put forward by Professor Ramsay in a paper communicated to
this Society +, a theory which attributes the existence of the various
Alpine lake-basins mainly to the erosive action of the glaciers by
which they were once occupied. I do not deny the possibility of
some lake-basins being the result of glacial erosion, or of others
being considerably modified by it; the position which I am en-
deayouring to maintain is that none of the greater Alpine lake-
basins has been primarily so formed, or indeed has undergone
any important secondary modification. I will add that all the op-
portunities which I have had of examination of lake-basins in this
and other countries have suggested to me serious doubts as to the
excavating power of glaciers; and, notwithstanding the weighty
arguments which have been advanced in favour of this theory in
more northern regions}, I am still disposed to believe, from what I
have seen, that it is only under very exceptional circumstances that
they can in any proper sense of the word excavate a rock-basin.

The principle which has directed my mode of operation has been
the following :—If a key is to be a master-key to a number of locks,
it must open them all; so if a theory is to be accepted as a general
one, we must apply it to as many instances as possible.

In order then to test this theory of glacier erosion, I have, year
by year, travelled over different portions of the Alps (as well as
other mountain-regions) constantly endeavouring to apply this and
other theories to the physiography of the country, and to ascertain
which appears to satisfy best the difficulties of each problem.

It seemed to me that, viewing the nature of the theory, it could
be better tested in this way than by any minute examination of a
single district. When any advantage appeared likely to accrue I
have done this; but, as the fact of the passage and the general
direction of the flow of the glaciers over certain regions are not dis-
puted, I have not thought it worth while to burden either this com-
munication or my note-book with the results of minor observations,
especially when they led to no important conclusions.

* Quart. Journ. Geol. Soe, vol. xxvii. p. 312, and vol. xxix. p. 382.
t Ibid. vol. xviii. p. 185.
t See especially Mr. J. Geikie’s new work, ‘The Great Ice Age.’

480 T. G. BONNEY ON THE GLACIER-EROSION

Thus I have said but little about direction of striew, &c., because,
as a rule, the bolder configuration of the terrain was sufficient, with-
out the confirmatory evidence of minor marking, to indicate to an
educated eye the path of the ice-streams,—and little about the stra-
tigraphy of the country (although that was often carefully noted),
because, as a rule, owing to its many disturbances, the evidence
furnished by it was neither favourable nor unfavourable to Professor
Ramsay’s theory or to my own.

The observations in this communication concern principally (1)
the district of the Upper Engadine and of the Val Bregaglia with
the Lake of Como, and (2) the Italian valleys of Monte Rosa
including the Lake of Orta. I had already travelled more than
once over the greater part of these regions, together with that of
the other Italian lakes; but, during the summer of 1873, I devoted
myself to a more minute reexamination of these, because, from previ-
ous experience, I knew them to be among the most important.

I assume throughout this communication :—

1. That gorges and V-shaped valleys are mainly cut by “rain and
rivers.”

2. That valleys or basins excavated or materially modified by
glaciers must have a trough-like form, that of a broad U.

3. That, except in those cases where there is a sudden change
from a hard to a soft rock, the contours of a valley, whose form has
been much modified by a glacier, would remain approximately
uniform—yiz. that if a lake-basin has been cut by a glacier, the
valley above must also be more or less U-shaped.

I commence with the Engadine district, where, unfortunately,
. owing to the unusual lateness of the summer, my results in one or
two cases were less complete than I had hoped to make them.

In the Upper Engadine are several small lakes—one group on
the Bernina Pass, the other in that singular trough which forms the
head of the Inn, and has for many centuries been one of the great
highways of Europe across the Alps, that of the Maloya Pass.

The former of these consists of two lakes, the Lago Bianco and the
Lago Nero, with a third unimportant tarn, situated at a height of
about 7300' above the sea, nearly on the summit of the Bernina
Pass (7658'). The Lago Bianco is about 14 mile long and } mile
broad ; the Lago Nero is very much smaller, some 300 yards long.
The former discharges its waters towards the Adda, the latter towards
the Inn.

The configuration of the ground in which these lakes lie is very
remarkable. It is a rather broad upland trough, about two miles
and a quarter long, running NW. and SE., dominated on the
south-western side by the steep precipitous buttresses of the Piz
Cambrena, the summit of which is 11,835’ above the sea, on the
other by rocky hills rising a few hundred feet above the lakes ; the
south-eastern end broadens out into a hummocky plateau, no part of
which is more than about a hundred yards above the water. The
torrent from the Lago Bianco passes over some marshy ground, once
part of the lake, and descends the steep head of the Val di Pila,

THEORY OF LAKE-BASINS. 481

while the carriage-road, hugging the northern side of the above
trough, crosses a slight depression (7658), and zigzags rapidly down
into the cirque-like head of the Val Agone. It needs but a glance
into these valleys to show that their physical features are anterior
to the epoch of the glaciers which have undoubtedly descended
them. The Cambrena glacier, still a fine ice-stream, united to
smaller glaciers which yet cling to the steep sides of the mountain,
once descended on to the trough along which the Bernina road now
runs, and sent forth glaciers on the one side along the Vals Agone
and Pila (covering doubtless all the intervening buttress), on the
other towards Pontresina.

Exactly in the path of this, and where the ice-streams from the
Cambrena, at the foot of its steep slopes, would be arrested by the
opposite ridges (which perhaps had also their glaciers), lie these lakes.

Though discharging their waters to different seas, only a low
barrier of ice-worn rock, nowhere more than a few feet above the
water, separates them*. This is exactly the place where the erosive
action of a glacier would be most intense ; and J should be inclined
to attribute these shallow lakelets mainly to this cause. At the
same time I observed that an important feature in the Lago Bianco,
namely a projecting elbow on the northern side, is formed by a Y-
shaped glen, down which a considerable stream now comes. The
whole of this region, however, is one where a minute examination is
likely to be useful ; this I hope to give on a future occasion, when
the ground is not (as it was on the occasion of my last visit) masked
with snow and the lake with ice; and I should not now have men-
tioned it, had I not considered it one which, so far as it goes, is
favourable to the glacier-erosion theory.

Proceeding now to the second group of lakes, those which form
the principal sources of the Inn, we are again struck by the singular
configuration of the ground. As we ascend the valley a level plain
extends for several miles by the side of the Inn, being very probably
a silted-up lake, until, a short distance above the well-known village
of Samaden, this plain slightly bifurcates, one arm extending a very
short distance along the Flatzbach (the torrent descending from Pont-
resina), the other also soon stopping at the foot of a rocky step in
the floor of the main trough, which runs very uniformly inaS.W. to
N.E. direction. The height of this abrupt step is about 400’; and
above it the valley extends with remarkable uniformity of level? till,
at the Maloya Kulm, the edge of the abrupt descent into Italy is
reached. This trough, about ten miles long, is now occupied by
three lakes, which, however, were once only two in number, the
present Silser See and Silvaplana See haying formed one long lake,
which extended from close to the edge of the Maloya Pass to

* Ball states (Alpine Guide, § 36, k) that in one place it does not exceed
3 feet above the ordinary level of the water; and that the Lago Bianco some-

times overflows into the Lago Nero. The greater part of it appeared to me not
to exceed 15 feet.

t As is shown by the following levels:—Maloya Kulm, 5942’; Silser See,
5892’; Sils, 5896’; Silvaplana See, 5886’; St.-Moritzer See, 5797’. The
height of Samaden above the sea is 5600’.

482 T, G. BONNEY ON THE GLACIER-EROSION

Campfer. Between Campfer and the head of the St.-Moritzer See
is a rocky plain, and the river in places has low cliffs at its
side, so that the upper lake occupies a true rock-basin; the lower
one is also enclosed by a rounded ridge of schistose rock, through
a narrow gorge in which the water escapes in a rapid torrent.

At first sight these lakes seem very favourable to the theory of
glacier-erosion ; but on closer examination several grave difficulties
suggest themselves. Confining our attention for the present to the
lower lake (St. Moritz) we notice :—

1. The gorge corresponds roughly in direction with the axis of
the valley (S.W. to N.E.); but the longer axis of the lake lies
about W.S.W. to E.N.E., the little glen through which the waters
pass to enter the gorge branching off about two thirds of the way
down the lake; the remaining one third of the lake-basin is formed
by the union of three small glens, down each of which a stream
descends, the first a mere streamlet from the barrier ridge, the second
and largest descending from the Statzer See (a small sheet of water
on the low shoulder of the mountain between Pontresina and St.
Moritz), and the third from the side of the main (Surlet) ridge.

2. The northern bank of the lake (on which stands the village of
St. Moritz) is very steep, actually precipitous at places; and this
extends to some distance beyond the gorge *.

Let us then suppose for a moment that a glacier excavated this
lake-basin. This glacier must either have been an offshoot from
the old Pontresina glacier which overlapped the shoulder mentioned
above and descended by the Statzer See, or it must have flowed
down from the flank of the Piz Surlei, or it must have come down
_the main valley from the Maloya Pass.

Doubtless a great glacier did descend over Pontresina ; and it is
possible that it may have overlapped at some period of its existence
in the direction of St. Moritz+; but it seems inconceivable, unless we
attribute far more plasticity to ice than is usually conceded, that it
can have sent off an offshoot at right angles to its main direction of
flow which would have had sufficient power to excavate these hard
rocks. Furthermore a glacier has certainly descended the main (Inn)
valley ; is it likely that it would have been thrust back so far by this
intruder? With regard to the second hypothesis, it is evident that
no important glacier has descended from this part of the Piz Surlei.
If then we adopt the glacier-erosion hypothesis, we are driven to
attribute the lake to the Inn glacier, of whose former passage down
this part of the valley there is sufficient evidence. But in this case
how are we to explain the abrupt termination of the rock-basin ?
The rocks about the gorge mentioned above would be in the “ Stoss-

* Between this cliff and the present margin of the lake is a narrow strip
of level meadow-land ; but it is quite evident that the level of the lake has gradu-
ally been lowered, so that this ridge is one side of the true basin.

t The rocks just in this part have been much affected by subsequent
weathering, which has to a considerable extent obliterated the glacial contours.
Ae evidence on the whole seemed more favourable than opposed to an over-

apping.

THEORY OF LAKE-BASINS. 483

seite,” and so should slope gently upwards from the water. Again,
the whole fall of the valley from the Maloya Kulm to the St.-Moritzer
See, a distance of nearly ten miles, is only 145’, and from the head
of the Silser See, which extends almost to the Kulm, barely 100’, a
fall of about 1 in 500, which could not give much excavating power.
In a word, though the lake is a true rock-basin, the physical fea-
tures of this district in no way lend themselves to a theory of
glacier’ erosion ; they rather seem to be preglacial in their age, and
to have been only very superficially modified by ice-action.

We proceed now to the head of the Silser See, which, as being
obviously the most important part of the district, I examined very
carefully. The present head of the lake is a marshy plain, sur-
rounded by alluvial terraces, which are about 20’ high, and often
rest on rock. The results of my observations were as follows :—

1. The watershed of the Maloya Pass is formed by a low rocky
iceworn barrier which falls down in fine precipices to the Val
Bregaglia.

2. The various glens which come down to the Silser See show
signs of local glacier-action down to very near the water’s edge ; but
all their principal physical features appear to be anterior to the
glacial epoch.

3. The principal affluents to the main Inn Glacier must formerly
have been from the Val Fex and the Val Fedoz, the general direc-
tions of which make angles of about 70° with the axis of the Inn
valley; yet there is no recess of the opposite shore to correspond
with the entrance of these; and they cannot have materially
deepened the lake, because the débris from them has converted it in
the one case into dry land, in the other has made a considerable
delta.

4. In the rocky barrier spoken of above is a gorge which shows
that the lake, or the glacier occupying its place, formerly discharged
its waters into the Val Bregaglia.

5. There are numerous perched blocks on this barrier and several
moraines, some being very large. These extend from the base of
the Pizzo della Margna in a general direction of W. 20° N. to close
to the barrier. One of the largest forms the present actual water-
parting. On the southern side of this are some marshy flats inter-
rupted by moraines, the waters of which are discharged by the
gorge (4). eg

6. Many of the blocks, perched on the barrier and in the moraines,
are a fine porphyritic granite. I found no more erratics of this kind
after reaching the first glen from the Piz Lunghino. They come
from the Val Forno, the floor of which valley is nearly on the level
of the Maloya Pass, and descends very rapidly into the head of the
Val Bregaglia. The rocks near the Maloya-Kulm Hotel are also
glaciated by ice that has moved from the Val Forno.

7. The glaciers about the head of the lake are all connected with
the existing valley-system. That from the Val Forno actually
swept over and, as it were, rested an elbow on the top of the Pass.
The glen under the Piz Lunghino, which joins the lake a short way

484 T. G. BONNEY ON THE GLACIER-EROSION

down its western shore, bears the marks of a glacier that has de-
scended it; and, in a word, looking down the lake, the rocks are
rounded by the glaciers which once descended from the several
mountain-sides; but there is nothing to suggest any scooping action,
no great modification of the outline of the lake corresponding to
the entrance of a glacier valley, while the existence of one or
two promontories is difficult to explain on any theory of glacial
erosion.

I think, therefore, we are justified in concluding :—

That this lake-basin has no immediate connexion with any of the
lateral valleys.

That it once extended to within a few dozen yards of the summit
of the pass, and that some of its drainage was discharged into the
Val Bregaglia.

That the lake-basin (and the above-named gorge) is anterior toa
period when the Forno glacier was of enormous thickness.

That, considering the slight descent of this uppermost trough of
the Inn valley, no glacier formed on the north side of the Maloya Pass
could have excavated the Silser See.

Hence that the basin and all the other physical features of the
head of the Inn valley are antericr to the glacial epoch, and have not
been much altered during it or since.

The first thousand feet of descent from the Maloya Pass into the
Val Bregaglia is very steep, its head almost meriting the title of a
cirque. The descent after that is generally gradual, with the excep-
tion of a sudden fall between Casaccia and Vico Soprano, corre-
sponding with the entrance of the Albigna torrent. The usual signs
. of glacial action, perched blocks, roches moutonnées, and large
moraines, may be detected at intervals all the way down to Chia-
venna. Among the blocks porphyritic granite and coarse gneiss are
conspicuous, evidently brought down by the tributary glaciers from
the left bank ; the rock in the bed of the valley consists of mica- schist
(in one place so fissile as to be quarried for roofing-slates) and
hornblendic schists. The valley is generally narrow, and V-shaped ;
and the ice-marks may be seen sometimes only a short distance above
the torrent. ‘They are very conspicuous after passing Piuro, near
the opening of the valley, just above Chiavenna, where the roches
moutonnées are magnificent, and the characteristic ice-marks may be
traced within some twenty feet above the river. One example, on
the left bank, about a mile from Chiavenna, is very interesting. A
steep knoll of rock, some 40' high, is divided by a narrow gap from
a cliff some 50’ high, which forms part of the range rising on the
left bank of the valley. The lowest part of this gap is perhaps 30
yards above the (Mera) torrent. The rock is a dark (serpentinous ?)
schist ; the whole of the knoll and cliff are magnificently glaciated,
the strize on the latter being very conspicuous. This place, there-
fore, shows (1) that the base of the old Bregaglia glacier was here
at no great height above the present stream ; (2) that this gap was
either excavated by it or anterior to it. As I cannot conceive the
possibility of a glacier cutting out such a gap, I am forced to con-

THEORY OF LAKE-BASINS. 485

clude that the principal features of the scene are preglacial, and
that the glacier has only slightly modified them.

Just below Chiavenna the true basin of the Lake of Como begins,
at the union of the Mera and the Lira (which latter is orographic-
ally the principal valley) ; for a broad level strath extends all the
way down to the waters of the Lago de Mezzola, and Chiavenna
itself is less than 350 feet above the level of the lake. Ice-worn
rocks may be traced on the mountain-sides from at least 2000’ above,
down to the very edge of the meadows.

Is it not then rather perplexing that these glaciers of the Lira and
the Mera, which singly have only slightly modified the valleys in
which they flowed, should, on their union, have been suddenly gifted
with such great excavating power as to scoop out from hard schist
(micaceous and hornblendic) and granite the open valley above the
Lago di Como? And if they have done this, how are we to explain
the comparatively shght effect produced by the junction of the broad
Val Telline, which must have brought down an immense affluent of
ice. About two fifths of the present Lake of Como lies in mica-schists,
the remainder (including its two arms) in sedimentary rocks, fol-
lowing one another in ascending order, and (according to Von Hauer)
members of the Silurian (?), Trias, and Rhetic formations. The
strata, as a rule, dip towards the S., often at a high angle ; but there
are one or two rolls, as near Balbaniello and the Val Intelvi, till
at last (in the Como arm) they roll gradually up and down, and
finally dip sharply away from the fault which brings the Nagelflue
conglomerates against them.

This arm of the Como lake has always seemed to me difficult to
explain on any theory of glacier erosion ; the low promontory of
Bellagio, unless the valley were previously well defined, seems inade-
quate to split a glacier into two such divergent arms; and its form
does not suggest that it has been exposed to any such tremendous
pressure as it would have undergone if this had been the case. This
Como arm, it is well known, has no outlet, and is chiefly cut off
from the plain of Lombardy by a low range of hills composed to a
large extent of Nagelflue, the pebbles being mostly metamorphic
rock, quartzites, schists, and the porphyritic gneissic and granitoid
rocks common in the Val Bregaglia, generally well rolled, not
usually exceeding about a foot in diameter. They lie in a grey
sandy matrix like Molasse, with sandy bands. The occurrence of the
gneissic and granitoid pebbles is important, as showing that the valley
of the Lira was defined at a very early period.

This range of hills is very peculiar in shape. It is a sharp ridge
with steep slopes on both sides, its tops in places being almost an
aréte. The old glacier has indeed passed over it; for on my ascent
to the tower (Castello Baradello) which is perched on the crest, I saw,
in a face of rock near S. Carpofaro, that the pebbles had been planed
level with the matrix, and a few of them yet retained strie. More-
over in several other places the characteristic contours of ice-worn
rocks could yet be recognized, and perched blocks could be seen
dotted here and there; indeed a large one (of the usual granitoid

486 T. @. BONNEY ON THE GLACIER-EROSION

rock) serves as a foundation-stone for the tower. Here, then, in the
very path of the glacier, we have these ridgy hills, which the ice has
but superficially moulded. How, then, could it excavate the lake-
basin? The contours also of the ground all round the head of the
lake seemed to be wholly uninfluenced by any great ice-stream
acting from the lake. I could therefore come to no other con-
clusion than that the glacier in its passage has but slightly altered the
physical features of the district, which all spoke of the usual sub-
aerial agencies.

T must now call attention to a point perhaps of some importance.
From the former head of the lake, near Chiavenna, to its end at Lecco,
is about thirty-four miles. The fall is considerably less than 300
feet; how much, it is difficult to say exactly. Suppose, then, that
all which is at the present time under water is the result of glacial
excayation, and that the slopes of the valley above lake-level are
mainly (as I think all who know the district will admit) the result
of river and subaerial denudation ; it follows that the stream had a
fall of considerably less than 1 in 600. I doubt whether this would
be sufficient for eroding a valley of this character in such hard rocks.

Again, the height of Chiavenna is 1040 feet above the sea; the
deepest soundings in the lake are 1341 feet, or 642 below the sea;
and the height of the hills above Como is about 900 feet. To what
are we to attribute an excavating-power so enormous? A more
unfavourable position for the exercise of any erosive action on the
part of a glacier could hardly be conceived. On the other hand,
supposing this enormous basin to have been hollowed out by ice,
ought not the rocks all around to bear the most marked indications
of the passage of this powerful agent? ought not those ridges of
hills above Como, which lie right in its path, to have had their
sharp summits planed flat? or did the glacier, when it had crawled
up their steep slopes, tumble, like an exhausted caterpillar, helpless
down the other side?

I pass now to what are often called the Italian valleys of Monte
Rosa. Here my first care was to examine the Lake of Orta, which
resembles the Como arm in being separated from the Italian plain
by a low ridge, and in discharging its waters from its northern end.
These hills are a quartzose felsite ; and the eastern bank of the lake
consists chiefly of schists. The most conspicuous of the felsite hills,
that on which the Torre de Baccione stands, has a little valley on
each side, of which the eastern has formed a distinct delta. From
this point the lake is enclosed by a generally steep bank or terrace,
perhaps 100 feet high, above which rather steep slopes rise up to the
plateau on which stands the village of Gozzano. A glacier has evi-
dently passed over this barrier, as moraine heaps of great perched
blocks of granite, gneiss, and schists are strewn about. In the slope
above the lake, stratified drift is seen ; it is considerably below the
moraines; and perched blocks lie on the surface of the bank all
about. If these have not been washed out of some upper layer of

a or rolled from above, the glacier must have passed over this
rift.

THEORY OF LAKE-BASINS. 487

Now, if we suppose the Lago di Orta to have been excavated by a
glacier, we must account, as in the case of the western arm of
Como, for the ridgy outline of the felsite hills, which could not
have resisted so great a grinding force. We must determine which
was the direction of the glacier in order to see how far it would
be likely to exercise a scooping force.

It either descended by the Val Strona, or was an offshoot from the
great Ticino glacier, coming up the valley down which the Strona
torrent now descends, and which appears like a prolongation of the
valley of the Lake of Orta. The difficulties in the former view will
be seen when I speak of the Val Strona. If the latter is correct,
we have to account for this extraordinary offshoot, for its ascent up
some 500 feet, and for its then scooping out the lake-basin.

From the northern extremity of the Lago di Orta, I ascended to
the head of the Val Strona and crossed to Vogogna, in the Val
d’Ossola, and then drove to Ponte Grande, in the Val Anzasca.
Hence I crossed to Fobello and descended the Val Mastalone to its
junction at Varallo with the Val Sesia, ascended that valley to its
head, crossed over to Gressonay, in the Val de Lys, and thence to
the head of the Val d’Ayas, from which I reached Zermatt by the
Schwartz-Thor. The results of this journey, which occupied a week,
may be summed up very shortly ; but they have, I conceive, an
important bearing on the point at issue.

The Val Strona is everywhere a narrow V-shaped valley, in
places almost a gorge. There is abundant evidence that a glacier
has descended it ; but without materially altering the characteristic
outlines of a valley formed by rain and rivers. The Val Anzasca is
also a narrow V-valley, commonly with a gorge at the bottom; yet
a glacier has evidently descended it while its form was much as now,
and the ice-marks may here and there be detected at no great height
above the stream. .

The Val Mastalone is throughout of the above form, everywhere
narrow, in most places a mere glen with little or no level land by
the side of the stream ; yet in places the ice-marks may be detected
only a few feet above the water, as for example about 12 kilometres
above Varallo. The main valley of the Sesia above that town
generally resembles that of the Val Mastalone, except that there is
often, though not always, a small strip of level land by the stream.
It has been filled by a glacier, which has left some of its most
conspicuous traces on an ice-worn mound of rock at Scopello,
perhaps 300 feet high, on one side of which goes the road, on the
other the river.

Among the magnificent scenery at the head of this valley, under
the cliffs of Monte Rosa, I was again and again struck with the
absence of any relation between the principal features of the scenery
and the direction of motion of the glaciers. Surely here, if anywhere,
the projecting buttresses of rock, which at times almost bar the
valley, should have been ground away by the enormous pressure to
which they have been subjected; but though rounded and worn
smooth, they still remain as barriers, suggesting irresistibly the con-

488 T. G. BONNEY OF THE GLACIER-EROSION

clusion that they are older than, and have not been much affected
by, the vast ice-stream which has passed above them.

The portions of the Val de Lys and Val d’Ayas which I examined
did not offer any fresh evidence of importance, except that the
singularly steep head of the latter is very difficult to explain on
any theory of glacier-erosion. _

Assuming, then, that glaciers can excavate such rock-basins as
that of Orta and of Como, which is 1900 feet deep, how are we to
explain their impotent action on the felsite rocks that bar the one,
and on the conglomerate and Molasse that bar the other? how to
explain the absence of all approach to lake-basins in those other
valleys through which the ice-streams must have descended from a
yet longer period, the absence of any but comparatively superficial
effects over all the rest of the district? No alteration in the mineral
character of the strata can account forit. The Lake of Como begins
in crystalline schists, and passes without much marked change in
width and depth through various sedimentary rocks. The Lake of
Orta also appears to lie in rocks no softer than those of Val Strona
and Val Sesia. Did time allow, I might appeal to other evidence on
the Lago Maggiore and the Lago di Lugano, the form of which appears
to me wholly inexplicable on any theory of glacier erosion; but I
think I have done enough to show that, even if this be retained as the
explanation of some of the upland tarns in the Alps, it is wholly
inadequate to explain the great basins of the Italian lakes. These
and their valleys appear to be much older than the ice-age, and to
have been but little modified during the period of maximum exten-
sion of the glaciers.

The theory which I propounded on a former occasion*, of irregular
movements of upheavaland subsidence along lines athwart the valleys,
can be applied to theselakes. There is nothing in the stratigraphy
of the district opposed to it ; so far as the evidence goes, it is rather
favourable to it. The anomalous form of the Lake of Lugano, the
position of the deepest parts of the Lake of Comoy, the variation in
the level of its floor, all agree very well with such a theory ; as does
the slight slope that we must assign to the preglacial valley, compared
with the fall of the other neighbouring valleys. The extraordinary
extension of the old Como lake up the Val Telline also agrees very well
with this, and induces us to suppose that the existence of these lakes
is due to the relative subsidence of a district included roughly between
one line drawn through Chiavenna from east to west, and another
through Lecco, Como, and Arona. This subsidence became less going
westward; but it may have affected an area some distance further to
the east.

Discussion.

Mr. Drew differed from the author, and thought that glaciers
coming from different directions might have produced single results
at certain points.

* Quart. Journ. Geol. Soc. vol. xxix. p. 382.
+ See an important paper by Mr. Ball, Geol. Mag. vol. viii. p. 359.

THEORY OF LAKE-BASINS. 489

Prof. Ramsay was glad to find that the author was gradually
coming round towards his own views, as he now admitted that
glaciers may have excavated some lake-basins. He argued that
the valleys originated before they were covered with a thick coating
of ice; but he maintained that at its origin the ice would have great
weight and excavating power.

Rey. J. F. Braxe thought that if the slope of the bottom of the
glacier changed, its excavating power would be increased. °

Mr. Kocu had sounded many lakes, and found great circular
cavities, which he thought could hardly be due to ice-action. In
winter glaciers move very slightly, and would be able to perform
very little work.

Mr. Bonney, in reply, remarked that he had in no way come
round to Professor Ramsay’s views; for he had never denied that
certain tarns might be the result of ice-action. Ice-denudation is
different from aqueous denudation. A glacier planes and could not
cut cut a notch. He said that he could only admit the smaller
lakelets in special positions as the result of glacier-action. He did
not think there was satisfactory evidence of the existence of an ice-
sheet all over Switzerland. He urged that it is not always safe to
argue from small phenomena to large ones.

Q.9.G, 8. No. l20; 2M

490 T. BELT ON THE STEPPES OF SIBERIA.

37. The Srepres of Stperta. By Tuomas Batt, F.G.S.
(Read June 24, 1874.)

1. Inrropuctton.

Many writers have described the great steppes or “tundras” of
Siberia, and told of their vast extent and monotonous aspect, where
the traveller, day after day, sees the nearly level plains extending
to the horizon in long low undulations, succeeding each other like
the gentle heaving of the surface of an otherwise calm ocean. From
the western base of the Ourals they spread right across the northern
extremity of Asia, nearly to Kamtschatka, unbroken, excepting
where the rivers have cut deep channels, or where some ridges of
rock rise through the level expanse of sand and loam.

Last October I journeyed across the south-west corner of hess
vast plains for more than one thousand miles, from Ekaterinburg,
on the eastern slope of the Ourals, to Byanool, a small town 360
miles south-east of Omsk; and although compared with the whole
area of the great steppes this part forms but a small portion, I saw
enough to impress me with the idea of their enormous extent, and
to excite my curiosity to try to discover the cause of this continental
expanse of sand and loam. I lost no opportunity of examining
natural sections of the ground; and fortunately these were not
scarce, as the rivers had everywhere cut deeply into the sandy
strata. I propose in the present paper to give a description
of the strata composing the steppes, to offer a theory to account
’ for their origin, and to seek to apply it to other parts of the earth’s
surface.

Description of the Strata.—About 60 miles east of Ekaterin-
burg we had left the slope of the Ourals behind us and entered
upon a sandy and nearly level country, the commencement of the
great Siberian steppes. The sandy soil, in some parts covered with
extensive pine forests, continued as far as Shadrinsky, 180 miles
east-south-east of Ekaterinburg; but beyond that town it was of a
more loamy character. Large shallow lakes were numerous; and
much of the land in cultivation was of a peaty character. Around
all the lakes were extensive tracts of peaty ground, proving that
they had at one time been much larger and had been contracted by
the growth of vegetation around their borders.

At Ischim we crossed a tributary of the Irtisch. It runs in an
alluvial plain 4 miles wide, on each side of which the steppe strata
rise to a height of about 80 feet above the river (fig.1). They are com-
posed entirely of loose sand, amongst which I could find no shells
or other organisms. In some parts the sands are false-bedded; and
next the surface they are more loamy than deeper down. They
contained no pebbles.

From Ischim to Omsk, about 160 miles in a direct line, the plains
continue with monotonous uniformity. At Omsk the strata exposed

TI. BELT ON THE STEPPES OF SIBERIA.

in the river-bank are about 60 feet thick, as
shown in fig. 2. The lowest bed exposed is a
finely laminated silt, which, in drying, sets to
an almost stony consistency, but immediately
disintegrates when placed in water ; the bottom
of this bed is not seen. Above it lies a bed of
laminated sand, mostly clean, but in some parts
mixed with silt; in others it is false-bedded.
It is nearly similar to that near Ischim, ex-
cepting. that I found in it some lines of small
pebbles and broken and worn shells of Cyrena
fluminalis. None of the pebbles were larger
than a small cherry.

From Omsk to Pavlodav, a distance of 253
miles, our course lay up the right bank of the
Irtisch, and the sections everywhere showed a
similar arrangement of the beds of sand and
silt as had been exposed at Omsk. At one
point, about 100 miles from Omsk, I noticed
that the lines of pebbles were getting more
numerous and a little larger, some occurring
as big as a walnut. In none of these sections
was the bottom of the steppe strata seen, nor
of course the bed-rock on which they rested,
until I reached the town of Pavloday, in lat.
52° 20’ N., where, on the river-bank, I found
the whole of the strata cut through down to
the bed-rock (fig. 3). This rock (No. 6 in
fig. 3) was magnesian limestone, which at the
top was crushed and broken, and seemed to
graduate from broken rock through broken rock
mixed with fine silt, into fine silt, in which
were numerous blocks of the bed-rock. The
bed of silt was unlaminated, and about 6 feet
thick ; in its upper half it did not contain any
pieces of the bed-rock. It was followed by 15
feet of clean coarse sand, laminated, and in some
parts false-bedded. It contained thin layers of
small gravel; and at the point marked x in
section there was a line of subangular pebbles,
some of which were as large as a man’s fist.
This bed was followed by 8 feet of light-coloured
sandy silt, and the whole capped by 20 feet of
stratified reddish-brown sand containing lines
of small gravel.

At Pavlodav we crossed the river and the
alluvial plain, which is 6 miles wide and was
covered with hay; we then struck across the
country to the south-south-west. After leaving
the river there were no natural sections of the

‘addeqg

‘SnyBy,

addaqg

491

“UnyostT qo hayjo A oY) Ssouon UOYIRG—'T “SIT

(‘poteytoys urerd Teranipy)

492 T, BELT ON THE STEPPES OF SIBERIA.

Fig. 2.—Section of Fig. 3.—Section of Strata at Pavlodav.
Strata at Omsk.

. 1 foot surface-soil.

—

2. 20 feet stratified, red-
dish-brown sand, with
lines of small gravel.

3. 8 feet light - coloured
sandy silt.

4, 15 feet coarse clean
sand, with lines of
small pebbles, and
one line of coarser
pebbles at X.

. 6 feet clayey unlami-
nated silt, with frag-
ments of bed-rock im
lower half.

or

6. Magnesian limestone,
much crushed and
broken at top.

strata exposed; but 20 miles south of Pavloday I noticed that the:
surface soil contained many small subangular pebbles; and having
already determined that the sand and pebbles had come from the
south, towards which I was travelling, I was on the look-out for,
larger stones. The country was now of a more undulating character,
and there were many shallow salt lakes. When 60 miles from
Paylodav, in the middle of the night, the wheel of the tarantass, for the.
first time in a journey of nearly 1000 miles, jolted over a stone; and
on getting out I found there were many angular pieces of quartz
lying cn the surface, from 3 to 4 inches in diameter.

After this the stones increased in number and size; and 20 miles
further south the surface-soil was full of large angular boulders of
quartz, some of which were 8 feet across. Many large tabular
masses of quartz rock looked as if in situ; but several of them had
been quarried to build the enclosures of the government station, and
they were seen to have rested in and on a sandy clay containing
few other stones. ‘There was nothing about these boulders or the
clay in which they rested to suggest that they were moraine accu-
mulations ; they were distributed over a nearly level plain, as if they
had been dropped from floating ice.

Beyond this point, southwards, the bed-rock often comes to the
surface in ridges and low hills of highly metamorphosed crystalline
rocks, separated by level plains composed of a sandy clay, with
numerous small angular stones, invariably of the rocks seen im situ

T, BELT ON THE STEPPES OF SIBERIA. 493

in the immediate neighbourhood ; the
fragments distributed over the plains
were entirely unworn. Where ex-
eavations had been made they were
seen to be much more numerous on
the surface than below. The ridges
and hills were covered with shattered
fragments of the rocks of which they
were composed; nowhere were there
any signs that they had ever suppor-
ted glaciers, though probably they
had been exposed to the action of
intense frost. If wesuppose that the
plains had been covered with water
forming shallow lakes, the distribu-
tion of the angular fragments over
their surface can be easily accounted
for, as these lakes would be frozen
over in winter, and on the ice break-
ing up in the spring it would carry
away fragments of rocks from the
surrounding ridges and drop them in
various parts of the lakes. I have
reason to believe this cause sufficient,
as I have seen it in operation in the

lakes of North America, and described |! Mh

it in a paper read before this Society
in 1864.

The country beyond this to the
southward, although still called the
Kirghese steppes, had lost its level
character, and some of the ridges of
crystalline rocks rose to heights of at
least 2000 feet, none of which showed
any signs of glaciation, excepting that
the surface had been shattered by
intense frost. The true plains may
be said to end on this meridian about
lat. 51° N.; whilst to the north-east
they extend diagonally across Siberia
for 3000 miles. Over this enormous
range of country their internal con-
stitution seems to be the same as that
of the plains I traversed; at Sama-
rova, however, on the same meridian
as Ischim, but 845 milesfurther north,
Erman states that the yellow talcose
clay of which the plains consist rests
on another filled with large fragments
of rock, and says it must be assumed

| me ee ee ee ee

it

v
mika Ht
‘1 hh U
Vy arts
play

‘eddaaig

“IOAN

‘anpojang 3 hayna-wonnr fo worz0ag 1n.19UagH—'F “SLT

“api SoTL g Ve] TRIANTTY

(‘poyeroS8Sevxe YONU o[wOS [vOT}I0 A )

-addayg

494 T. BELT ON THE STEPPES OF SIBERIA.

that west of the meridian of Samarova the blocks lie exposed
on the surface, whilst eastward they are completely covered by the
great deposit of clay; two thousand miles to the eastward of this
point, at Yakutsk, he states that the strata of the steppes con-
sist of loam, fine sand, and magnetic sand. In the deepest parts
twigs and leaves of trees are found; the alluvial plain of the Lena
also consists of similar deposits and the spoils of willow-banks.
“«‘ Everywhere,” he says, “ throughout these immense alluvial deposits
there are now lying the bones of antediluvian quadrupeds along
with vegetable remains.” At Pavloday I obtained teeth of the
Mammoth and Rhinoceros tichorinus from M. Hermann Paulson,
who had made a large collection of the bones of Mammalia from the
banks of the Irtisch. He told me they were exposed in sections of
the steppe-sands after freshets in the spring, when portions of the
banks were carried away ; this information agrees with the position
of the mammalian remains assigned by the traveller Pallas, who
states that they do not occur in the marshes or low lands, but in
the precipices of the river-banks, in what I have called the steppe-
sands. It is interesting to find that they are found in the same
position as the Cyrena fluminalis. Further to the north, around the
coast of the Polar sea, within the Arctic circle, the carcasses of
mammoths have been found along with marine shells, proving, I
think, a small rise of the coast since they were deposited.

2. THEORY OF THE ORIGIN OF THE STEPPES.

It is an interesting question, and one not easily answered, what
. spread out these vast level sheets of sand and loam over the whole
northern base of Asia? For weeks, as I was driven over them and
saw day after day the sun rise on a level horizon and set again on
the far-stretching plains, I puzzled myself with it. I ultimately
arrived at a solution which, I believe, meets all the requirements of
the case; and to obtain its acceptance by geologists, I think it will
be best to describe the steps by which I was led to it.

I soon saw that the plains had no relation to the present river-
system: the rivers simply cut through them; and there are no
defined river-basins, bounded by rocks of greater age,"in which they
might have been deposited. My next idea was that the sands must
be marine; and I find that this is the explanation of their formation
given by the celebrated geologist Prof. Bernhard yon Cotta, who
visited the Altai in 1868. He states that the absence of the entire
series of rocks between the Permian and the Diluvial leads to the
conclusion that during the whole of this long period the region was
dry land, and that during the Diluvial period it was covered with
water up to the foot of the mountains. He says “at this time an
ocean, extending from the Glacial Sea to the Ural, the Altai and the
Caspian and Black Seas, seems to have been a boundary between
Europe and the south and east of Asia. The absence of traces of
glaciers, and, indeed, generally of any vestiges of a glacial period,
such as are so frequently observed in Europe, may be accounted for

T. BELT ON THE STEPPES OF SIBERIA. 495

by the supposition that a current of warm sea-water passing from
the Mediterranean to the Glacial sea took its course along the then
existing Altaic coast. The Mammoths, remains of which have been
found in some of the caves of the Altai, may have lived upon large
flat islands rising out of the Diluvial sea”*. I have given Prof.
Cotta’s theory, but may remark that the supposition of the
Mammoths having lived on islands is not necessary, as, if the
area was once occupied by the ocean, the Mammoths might follow
its retreating coast as the land was gradually elevated. I believe
that the absence of sea-shells is fatal to the marine theory:
I searched diligently for them and could find none; and excepting
in the extreme north, around the present coast, I believe none have
been found by other observers. The mollusca exist all over the
present ocean; they abound around Greenland, even within a short
distance of the foot of the great glaciers. Ido not contend that the
presence of shells of Cyrena fluminalis is a proof that the waters in
which the sands were deposited were not marine, as they might
have been (and, I believe, were) brought down to the area by streams
from the south; but their preservation in the sands proves that
marine shells did not exist there, or their remains would also be
found. I have visited most of the coasts of the world, and dredged
in northern waters, and everywhere found marine mollusks to
abound; and I believe that the absence of sea-shells in any Tertiary
or Post-tertiary strata, excepting In some muddy deposits in which
they could not well live, is a proof that these strata were not depo-
sited in the great ocean. Utterly improbable is it that, according to
von Cotta’s theory, a warm sea-current flowing to the pole passed
over this area. All our experience of the existing ocean proves that
such a warm current would have teemed with marine life.

Rejecting the theory of a marine submergence, I next took into
consideration the possibility of a barrier having existed to the
north that dammed back the drainage of the country and produced
an immense lake in which the sands were spread out. I found it
difficult to conceive of any rise of land to the north that could have
formed such a barrier; and it is also to be remarked that there is
evidence in the marine shells of the northern border that it stood
lower, and not higher, than it now does. If there was now a rise
of the northern boundary of the continent, would the drainage of the
country be pounded back? or would not the rivers cut down their
channels and still keep open their outlets to the sea?

Before reaching Pavloday I had come to the conclusion that it
was probably an overflow of polar ice that had blocked up the
drainage of the country ; and when I reached that town and at last
got a sight of the ancient bed-rock over which the ice must have
moved if it existed, and saw the crushed and shattered surface and
the fragments pushed up into the overlying silt, I no longer
hesitated to adopt the theory; for the appearances presented in the
river-cliff were just such as might be expected to be produced at the
foot of an ice-flow advancing against the slope of the continent.

* Quart. Journ. Geol. Soc. 1869.

AOS J. BELT ON THE STEPPES OF SIBERIA.

Let us see if such a barrier would account for the formation of
the steppes. Siberia is a great continental basin, surrounded on
three sides by the mountain-chains of the Oural, the Altai, and the
Stannovoi; and the only outlets of its drainage are to the north. If
an overflow of ice took place from the north, the waters of all the
rivers would be pounded back. As the ice advanced southwards
it would grind against the bed-rocks, and the water issuing from
below it would wash out sand and silt. Where ranges rose above
it, boulders would be carried southward; and those noticed by
Erman at Samarova were probably brought from the northern end
of the Ourals. At its culmination I do not think it reached
much further south than Pavlodav, as the steppe-sands thin out
rapidly, and there is no sign of glaciation on the surface of the ex-
posed rocks fifty miles to the south. The edge of the iee would
mark the commencement of a great lake into which would run all
the streams from the south and all the water from the melting ice.
At its greatest elevation I believe the overflow of this lake must
have run through the depression between the southern termination
of the Ourals and the western end of the Altai to the Aral and
Caspian Seas.

During the slow retreat of the ice the water would follow its
receding margin ; and every portion of the growing steppe would at
one time or the other form part of the shore of the great lake.

Around this shore, and on the banks of the rivers flowing into the
lake, I believe the great mammalia lived ; and as the water moved
northward they followed its margin; and their remains are thus
found dispersed over the whole area of the steppes. Their frozen
_carcasses found along with sea-shells in the far north prove that
they lived up to the close of the Glacial period, whilst their bones,
in caves far to the south in the Altai, probably mark their range at
the time of its greatest intensity.

I have shown that the ranges to the north of Pavlodav exhibit no
signs of ever having supported glaciers. If we glance at a map of
Asia we shall see that currents of air coming from the south would,
before they reached Siberia, have to cross the high mountains of
India and Central Asia, where they could not fail to part with their
moisture; so that the precipitation in Siberia was probably not
greater, but less, in the Glacial period than it is now. Whence then
came the ice that blocked up the-drainage of Siberia? I believe it
was an overflow from the polar basin, which was filled with ice that
poured into it from the northern extremities of America, Greenland,
and Europe, and which was further heaped up towards the pole by
precipitation from moisture-bearing currents of air that travelled
far north, through Behring’s Straits and past Spitzbergen.

3. APPLICATION OF THE THEORY TO OTHER PARTS OF THE KARTH’S
SURFACE.
Eight years ago, in a paper on the Glacial period in North America,
I suggested that certain phenomena thought to prove that there had

T. BELT ON THE STEPPES OF SIBERIA, 497

been a great submergence of the country in the Glacial period might
be explained by the theory that ice coming down from the north had
dammed up the valley of the St. Lawrence, and changed it into a
great freshwater lake. I also showed that stratified beds of sand
and gravel, without marine shells, might have been, and probably
were, formed by ice flowing down the principal valleys and causing
glacial lakes in the lateral ones, in which the sands and gravels
were deposited. I at the same time urged that the absence of
marine remains in these deposits should make us pause before we
come to the conclusion that they were formed by the sea.

I certainly neither at that time nor until my visit to Siberia had
any idea that the theory was of such wide application ; but it will be
seen that what I have now advanced is only an extension of the
earlier theory, and that I was prepared to accept it more readily than
I might otherwise have done.

The theory of the blocking up by ice of the drainage of countries
in the Glacial period, and the formation of large and small lakes, will,
I believe, explain the existence of many stratified deposits that are
supposed to prove the great submergence of the land below the
waters of the ocean. Thus, if the ice once filled the beds of the
German Ocean and the Irish Sea, all our rivers running to the west
and to the east must have at one time been dammed back by it.
Similarly every lateral valley in the north of England and in Scot-
land was at one time or another occupied by a glacial lake; and
thus, I believe, the stratified sands, often false-bedded, that lie above
the Boulder-clay were formed. In many cases the barrier-ice seems
to have been suddenly removed or broken through, and the escaping
rush of water has cut down through the sands and left them in long
ridges or isolated mounds.

Again, the ice that descended from the mountains of Scandinavia
must haye dammed back all the rivers of Northern Europe; and
I think that a great part of Central Europe must at that time
have been occupied by great freshwater lakes over which floated
icebergs.

It is with some diffidence that I put forward these views, as I
know they are opposed to those held by most geologists; but I am
sure they will credit me with a desire to arrive at a true explanation,
and admit that the arguments I have advanced show, at least, that
the grounds on which it is held that a great part of the northern
hemisphere was depressed below the level of the ocean in the Glacial
period require reconsideration.

Discussion.

Mr. Drew thought that the deposits had a lacustrine aspect.
Their nearest parallel was to be found in the great plain of India,
the origin of which was not quite clear, although it had perhaps
originated as a delta. He did not think that the ground had been
covered by an ice-cap.

498 T. BELT ON THE STEPPES OF SIBERIA.

Mr. Hutxe believed that the author’s view as to the origin of the
deposits was correct. He thought that the plains of Hungary were
of a similar nature, their freshwater deposits having been formed
in consequence of the stoppage of the Danube, changing what had
been a dry plain into a vast lake. The effects described by the
author might have been produced by a rise of land in the north.

Prof. Anstep remarked that the extent of the country under con-
sideration might be estimated roughly at 3,000,000 square miles,
and asked whether this was to be regarded as the area of a fresh-
water lake. Ifso, whence did the water come and where did it go?
He thought such a phenomenon of very unlikely occurrence.

Mr. Betz, in reply, said that he was glad to find that the marine
theory had had no supporters. He maintained as a fact that if a
barrier were placed across the country, the numerous rivers would
soon flood its whole surface.

H, ALLEYNE NICHOLSON ON SPECIES OF CH TETES. 499

38. Descriptions of Sprcres of Comtxres from the LowEr SILvRIAN
Rocxs of Norrn America. By H. Attznyne Nicnoxson, M.D.,D.8c.,
F.R.S.E., F.G.S., Professor of Biology in the College of Physical
Science, Newcastle-on-Tyne. (Read June 24, 1874.)

[Pirates XXIX. & XXX.]

Tae Lower Silurian rocks of North America have yielded a very
large number of small corals, which have been ordinarily referred
by paleontologists to one or other of the three genera Cheetetes,
Monticulipora, and Stenopora. Before considering these, however,
it would seem advisable to very briefly discuss the value of the
above-mentioned generic divisions, the boundaries of which are by
no means yet satisfactorily defined.

The genus Stenopora is defined by Lonsdale as follows :—“< A
ramose, spherical, or amorphous tubular polypidom ; tubes polygonal
or cylindrical, radiated from a centre or an imaginary axis, contracted
at irregular distances, but in planes parallel to the surface of the
specimen; tubular mouths closed at final periods of growth; ridge
bounding the mouths granulated or tuberculated ; additional tubes
interpolated.” (Lonsdale in Strzelecki’s ‘ Physical Description of
New South Wales and Van Diemen’s Land,’ p. 262, 1845.)

The chief features in the above description which are relied upon
as characterizing the genus are the constriction of the corallites at
irregular intervals in planes parallel with the surface, the granula-
tion or tuberculation of the ridges bounding the calices, and the
interpolation of the young tubes between the old by lateral gemma-
tion, so that a fracture exposes the outer walls of the tubes. Mural
pores of course are absent, as in the whole family of the Chetetine.
Lonsdale’s definition, as above given, is accepted in all its essential
details by M‘Coy (Brit. Pal. Foss. p. 24), who describes several
species under this head. Milne-Edwards and Haime, in the Intro-
duction to the ‘ British Fossil Corals’ (p. lxi), accept the genus
Stenopora, defining it simply as “very similar to Chetetes, but
having small styliform processes at the angles of the calices.” They
thus do not notice the characters relied upon by Lonsdale and
MCoy as separating Stenopora and Cheetetes, whilst they introduce
a feature not mentioned by either of these observers. In other
words, they break up Lonsdale’s genus into two portions, one of
which, typified by S. spinigera, Lonsd., they retain under Stenopora ;
whilst the other, comprising all the species enumerated by M‘Coy,
King, Geinitz, and Howse, they place under Chetetes and Favosites.
The type species, S. spinigera, I have never seen; but it appears to
me that the characters relied upon by Lonsdale and M‘Coy as sepa-
rating Stenopora and Chetetes are clearly insufficient for this purpose,
and that Milne-Edwards and Haime are therefore fully justified in
the course they adopted. As far, at any rate, as our American

500 H. ALLEYNE NICHOLSON ON SPECIES OF CHAITETES

examples are concerned, though the occurrence of Stenopora fibrosa
has been often quoted, I have vainly sought for any specimens
which I could refer to this genus. The periodically constricted
tubes, contracted in planes parallel with the surface, do not form a
feature in any of our specimens. The existence of tubercles and
eranules on the bounding-ridges of the calices can be observed in
undoubted species of Cheetetes (Monticulipora), such as OC. Jamesi,
Nich., and C. Orton, Nich. A Devonian species, C. monilifera,
Nich., though in other respects agreeing entirely with Chetetes,
carries a series of vertical spines which project above the surface
from all the angles of the calices; and I have observed the same
feature in an undescribed species from the Trenton Limestone.
Lastly, all the forms which I have examined from the Silurian and
Devonian rocks of America, with the single exception of a large
lobate form from the Trenton Limestone, have the outer walls of
the corallites exposed by a rough fracture, and may thus be sup-
posed to increase by lateral budding. Taking this character alone,
therefore, they would all have to be referred to the genus Stenopora,
though in other respects not agreeing with it, but with Chetetes
or Monticulipora.

The genera Chwitetes, Fischer, and Monticulipora, D’Orbigny, save
in one point to be discussed immediately, are identical with one
another, and may be briefly defined as including ramose, frondescent,
massive, or incrusting corals, composed of closely aggregated, poly-
gonal, circular or subcircular corallites, which are traversed by well-
developed tabule, and which may or may not be separated by very
minute intermediate tubuli. No mural pores. Septa absent or
. quite rudimentary.

The point by which it has been attempted to separate Chetetes
and Monticulipora is the mode of increase of the coral, which is
stated to be by fission in Chetetes and by lateral gemmation in
Monticulipora, the corallum in the two genera being in other respects
precisely similar. It seems to me, however, that it is a pity to
attempt to found a generie distinction upon this point alone, and for
the following reasons :—In the first place, even if other objections
were removed, and the distinction could be shown to be constant, it
might be doubted if this character alone should be regarded as of
more than subgeneric value. Secondly, the character is one under
the best of circumstances difficult of determination, as shown by the
fact that very eminent palzeontologists have come to precisely oppo-
site conclusions as to the mode of growth of the corallum of the
same species. Thus Lonsdale regards Chetetes petropolitanus as
increasing fissiparously, and places it accordingly ; whereas Milne-
Edwards and Haime regard the same species as increasing by gem-
mation, and refer it to Monticulipora. Thirdly, in a great many
specimens of all the species the distinction in question is one which
it is impossible to apply; and in some cases this is true of all the
examples. Thus, in the exceedingly thin incrusting species the
mode of growth cannot be satisfactorily determined ; and one meets
with numerous examples of the ramose, frondescent, or massive

¥ROM THE LOWER SILURIAN OF NORTH AMERICA. 501

forms of which the same is true. It is, however, a pity to separate
two genera by a single character which does not admit of general
observation and application. Lastly, if this distinction be admitted,
it will become a question whether the limits of Lonsdale’s genus
Stenopora should not be extended so as to receive the forms gene-
rally referred to Monticulipora.

The presence of “ monticules” or “mamelons,” from which the
name Monticulipora was derived, cannot be relied upon, since these
elevations are sometimes wanting, and, besides, vary very materially
in their nature even in the cases where they present themselves.

The genus Nebulipora, M‘Coy, would seem to have been rightly
included by Edwards and Haime under Monticulipora, since the
only available distinction is the presence of mural pores, and M‘Coy
is unable to speak with any certainty as to the existence of these
openings in the forms upon which his genus is founded.

For the reasons above given I shall in this communication place
all the corals under consideration in the genus Chetetes, irrespective
of their mode of increase, which is in many instances unknown.
The specimens from which my descriptions are taken have been
partly collected by myself from the Hudson-River group of Canada ;
but most of them have been kindly furnished to me from the
Cincinnati group by Mr. James and Prof. Orton, and from the
Trenton Limestone by Mr. G. J. Hinde. I shall also not only
describe those species which are actually new, but I shall likewise
give a brief account of some of the previously recorded species, the
well-preserved specimens in my possession in many cases enabling
me to adduce new facts of interest and importance.

A completely satisfactory subdivision of the genus Chetetes (in-
eluding Monticulipora) is at present not possible; but the genus
may be conveniently broken up, according to the form of the coral-
lum, into the ramose, frondescent, massive, and incrusting species.

I. Ramost SPECIES.

This group includes those forms in which the corallum is ramose
or dendroid, the corallites disposed in a radiating manner round an
imaginary axis. The corallum is rooted at the base, and the extre-
mities of the branches are rounded. Some of the forms of this
group are slender and regularly divided; but others are more or less
swollen or tumid and irregularly branched, becoming sometimes
almost lobate, and thus passing into one of the sections of the
massive group.

1. Cumreres Dater, Edw. & H. Pl. XXIX. figs. 1, la.

Chetetes Dalei, Edwards & Haime, Pol. Foss. des Terr. Paléoz.
pl. 19. fig. 6.

Corallum branching, the branches cylindrical or elliptical, dividing
dichotomously at short intervals; the diameter of the branches on
an average from 3 to 4 lines, but sometimes only from 1 to 2 lines.
Calices in general from six to eight in the space of | line, polygonal,

502 H, ALLEYNE NICHOLSON ON SPECIES OF CHETETES

with moderately thin walls. A greater or less number of smaller
calices, usually in the form of very minute circular tubuli, situated
at the angles of the larger corallites. Surface covered with conical,
rarely transversely elongated eminences or tubercles, which some-
times attain a height of over half a line, and which are placed at
intervals of from half a line to nearly 1 line apart. These tubercles
are covered with corallites which are not larger than the average;
and their summits may be partially compact or furnished with
corallites of smaller size than usual.

At least two variations appear to be included under this species.
In some forms, which may probably be regarded as the most typical,
the eminences of the surface are very prominent, and the normal or
average corallites are separated by a great number of extraordinarily
minute cylindrical corallites, which vary from three to six or more
placed round each of the larger tubes, which thus acquire a more or
less circular form. In another variety the surface-tubercles are
usually not so pronounced, and the corallites are for the most part
nearly equal in size. Even in these examples, however, there are
occasional excessively minute corallites at the points where the
larger ones meet, though this is never a conspicuous feature as it is
in the preceding variety.

Locality and Formation. Common in the Cincinnati group (Hud-
son-River formation), near Cincinnati, Ohio.

2, Cumretes rucosus, Edw. & H. Pl. XXIX. fig. 2.

Chetetes rugosus, Edwards & Haime, Pol. Foss. des Terr. Pal.
pl. 20. fig. 6.

Branches cylindrical or flattened, from 2 to 3 lines in diameter,
' dividing dichotomously at short intervals. Corallites polygonal or
subcircular, the average ones about six or eight in the space of 1 line,
separated by numerous exceedingly minute circular tubes. Surface
exhibiting numerous elevations, which have a height of from one
quarter of a line to half a line, and are transversely elongated, so as
to constitute so many discontinuous transverse ridges. These ridges
vary in length, but they do not extend round the stems, and they
are usually sharp-edged and placed about half a line apart. The
corallites on the ridges are not larger than those on the other parts
of the coral.

This species differs from C. Dalez in little except in the fact that
the tubercles are elongated in a direction at right angles to the axis
of the branches. It may be only a pronounced variety of C. Dalei.

Locality and Formation. Common in the Cincinnati group, near
Cincinnati, Ohio.

3. CH@TETES APPROXIMATUS, Nich. Pl. XXIX. figs. 3, 3a.

Corallum consisting of cylindrical stems, dividing dichotomously,
and having a diameter of about 2 or 23 lines. Corallites thick-
walled, though not excessively so, subcircular or polygonal, eight or
ten in the space of 1 line, sometimes with excessively minute coral-
lites interspersed amongst the larger ones, though these are never

FROM THE LOWER SILURIAN OF NORTH AMERICA, 503

abundant and may be absent over considerable spaces. Surface
exhibiting a number of small, somewhat transversely elongated
tubercles, placed in diagonal lines at distances of half a line apart
transversely and two thirds of a line measured vertically. The
tubercles are very slightly elevated above the general surface, and
are either solid at their summits, or carry a few excessively small
eylindrical tubuli with or without two or three of the average-sized
corallites.

This species is closely allied to C. Dalei, from which it is distin-
guished by its slightly smaller calices, by the fact that the tubercles
are not conical and prominent and are not covered with corallites of
the ordinary size, whilst the very minute intermediate tubuli are
either absent or are present in very small numbers.

Locality and Formation. Cincinnati group, near Cincinnati, Ohio.
Collected by Mr. U. P. James.

4, Cumretzs atrritus, Nich. Pl. XXIX. figs. 4, 4a.

Corallum composed of subcylindrical branches, having a diameter
of from 5 to 7 lines, and dividing at short intervals. Corallites with
thin walls, polygonal, from eight to ten in the space of 1 line, sub-
equal, some being smaller than the others, but not excessively so.
No very minute intermediate tubuli interspersed among the larger
corallites. Surface with numerous, very close-set, quadrangular or
conical solid eminences or tubercles, each placed at the junction
of from four to six corallites. These tubercles are from four to
five in the space of 1 line, and are apparently quite compact, carry-
ing neither the average corallites nor minute tubuli.

This species is distinguished from all the preceding by its thin-
walled corallites, the absence of excessively minute tubuli placed
round the larger tubes, and the minute, crowded, solid tubercles
which stud the surface. At first sight examples of this species look
as if they had been more or less weathered and worn; but a closer
examination does not bear out this view.

Locality and Formation. Cincinnati group, near Cincinnati, Ohio.
Collected by Mr. U. P. James.

5, Cumretes putcHeLivs, Edw. & H. Pl. XXIX. figs. 5-5 6.

Cheetetes pulchellus, Kdwards & Haime, Pol. Foss. des Terr. Paléoz.

271.
ce Monticulipora pulchella, Edw. & H. Brit. Foss. Corals, pl. 62.
figs. 5-5 b.

Corallum variable in form, usually of subcylindrical branches,
which have a diameter of from 2 to 6 lines, sometimes forming
flattened and expanded subpalmate fronds, sometimes inosculating.
Corallites thin-walled, polygonal, unequal in size, about eight of the
average ones occupying the space of 1 line. Surface exhibiting
rounded or somewhat stellate groups of larger corallites, of which
two or three occupy the space of half a line, and which sometimes
have very minute intermediate tubuli between them. These groups
of larger corallites generally comprise from five to seven individuals,

504 H. ALLEYNE NICHOLSON ON SPECIES OF CHETETES

and they are placed about 1 line apart; they are very slightly or
not at all elevated above the general surface, so that there are no
conspicuous tubercles.

The characters of this species are so well marked as to render its
recognition very easy, even in small fragments.

Locality and Formation. Cincinnati group, near Cincinnati, Ohio.

6. Cumreres Frercueri, Edw. & H. Pl. XXIX. figs. 6, 6a.

Cheetetes Fictchert, Edwards & Haime, Pol. Foss. des Terr. Paléoz.

5 ads
: Cheetetes lycoperdon, Hall (ex parte), Pal. N.Y. vol. ii. pl. 17. figs.
1 g-i (cet. excl.); also Pal. N. Y. vol.i. pl. 24. fig. 1g (cet. excl.) ?

Monticulipora Fletchert, Kdw. & Haime, Brit. Foss. Corals, pl. 62..
figs. 3, 3a.

Corallum ramose, branches from 14 to 3 lines in diameter,
cylindrical or subcylindrical, often irregularly swollen at intervals,
dividing dichotomously usually at considerable intervals. Corallites
with moderately thick walls, the average ones of unequal sizes,
about eight in the space of 1 line, and having very minute tubuli
sparingly interspersed amongst them. Surface smooth, destitute of
tubercles, but occasionally showing groups of corallites which are
very slightly larger than the average.

The examples of this species which I possess from the Clinton
group (Upper Silurian) agree entirely with the description given by
Edwards and Haime, being smooth, and simply having a moderate
number of very minute tubes irregularly intercalated amongst the
ordinary corallites, these latter varying slightly in their dimensions.
. On the other hand, the Lower-Silurian examples which I have
referred to this species, though also smooth, and also having minute
tubuli sparsely interspersed amongst the larger corallites, exhibit
distinct groups of corallites which are of rather larger size than the
average, and which are not set upon distinct elevations. They thus
approach C. pulchellus, from which they are hardly distinguishable
except by the fact that the groups of large-sized corallites are not
nearly so conspicuous, the individual corallites which compose them
being only very slightly above the average size. On the contrary,
in C. pulchellus the groups of large-sized corallites are very con-—
spicuous, as the corallites composing them are nearly or quite twice
as large as the ordinary ones. Perhaps the Lower-Silurian forms
here regarded as referable to C. Fletcheri may constitute a distinct
variety either of this or of C. pulchellus.

Locality and Formation. Common in the Hudson-River group of
the Province of Ontario, and in the Cincinnati group, near Cincinnati,
Ohio; also in the Clinton group (Upper Silurian) at Dundas, Ontario.

7. Cumreres craciiis, James. Pl. XXIX. figs. 7, 7a.

Cheetetes gracilis, James, Catalogue of the Fossils of the Cincinnati
Group, 1871 (named, but not figured or described).

Corallum dendroid, the branches cylindrical or subcylindrical,
from less than 1 line to 2 lines or more in diameter, dividing dicho-

FROM THE LOWER SILURIAN OF NORTH AMERICA, D505

tomously at short intervals. Corallites very small and crowded,
from ten to twelve in the space of 1 line, thick-walled, opening
obliquely on the surface by oval or circular calices, between which
are placed excessively minute circular tubuli. The surface exhibits
no elevations or tubercles, but is entirely smooth ; and there are also
no groups of larger-sized corallites, the ordinary corallites being all
nearly of a size.

This species is allied to C. Fletcheri, Edw. & H., from which it is
distinguished by the thick-walled nearly equal corallites, and the
oblique and very small calices, the dimensions of which are much
more minute than in the latter form. My description is drawn from
type specimens furnished by Mr. U. P. James.

Locality and Formation. Cincinnati group, near Cincinnati, Ohio.

8. Cuaretss peLIcaTuLus, Nich. Pl. XXIX. figs. 8, 8a, 8 6.

Corallum very slender and delicate, ramose, composed of cylindri-
cal stems, terminating in rounded, sometimes thickened extremities,
and apparently springing in some cases from a horizontal footstalk.
Stems sometimes apparently simple, more commonly branched di-
chotomously, the division taking place at acute angles; diameter of
the stems and branches from one quarter of a line to half a line,
rarely reaching two thirds of a line. Corallites very oblique to the
surface, opening by oval apertures, the length of which corresponds
with the axis of the stem and exceeds the breadth. Calices in dia-
gonal rows, about eight in 1 line measured longitudinally, and from
twelve to fourteen in the same space measured diagonally. When
perfect and unworn, the lower lip of the calices is thin and prominent.
The calices are all of equal size; and the surface is destitute of
monticules or tubercles.

This is one of the commonest fossils of the Hudson-River forma-
tion, both in Canada and in the United States. From its very
minute size, I am left somewhat uncertain as to the true position of
this abundant little fossil. It is, I think, certainly the form which
has generally been quoted as a slender variety of Stenopora fibrosa ;
but even if this species were to be retained, our examples could not
be placed under it. It is likewise apparently one of the forms
which has been figured by Hall under the name of Chetetes lyco-
perdon (Pal. N. Y. vol. i. pl. 24. fig. 1h, cet. excl.). It is most
closely allied to Cheetetes gracilis, James; but it is readily separated
by the absence of minute tubuli interspersed amongst the larger
corallites, by the much greater obliquity of the corallites and their
much thinner walls, and by the uniformly slender habit and stunted
growth. From (. Fletcheri, Edw. & H., it is also distinguished by
the uniformity of the size of the corallites and their oblique direction.

Of the fossils of the later rocks, C. delicatulus runs a considerable
chance of being confounded with Helopora fragilis, Hall, which it
much resembles superficially. It is, however, readily distinguished
by its almost always being branched, by the form of the calices, and
by the fact that the calices are not arranged between longitudinal
elevated lines.

Q.J.G.8. No. 120. Qn

506 H, ALLEYNE NICHOLSON ON SPECIES OF CHATETES

Locality and Formation. Common in the Hudson-River group of
Canada almost everywhere; also in the Cincinnati group, near Cin-
cinnati; also in the Trenton Limestone.

9. CozrTETEs? nopuLosus, Nich. Pl. XXIX. figs. 9, 9 a.

Corallum dendroid, minute, of small cylindrical stems, which
branch dichotomously at intervals of 2 lines, and have a diameter of
from two thirds of a line to 1 line. Corallites prismatic or hexa-
gonal, directed somewhat obliquely to the surface, of two sizes.
Larger corallites opening by subcircular or oval apertures, the long
diameter of which coincides with the axis of the stems; from six to
eight in the space of 1 line measured vertically. Smaller corallites
in the form of exceedingly minute circular tubuli interspersed
amongst the larger tubes. Surface exhibiting numerous minute,
sometimes conical, sometimes transversely elongated elevations or
tubercles, which are placed at distances of about half a line apart,
and give to the surface a characteristic nodulose appearance.

This very distinct species is more nearly allied to C. Dale: than
to any other; but it is very readily separated by its much more
slender and graceful proportions, and the much smaller size of the
proportionally remote tubercles. One specimen, indeed, which can
hardly be referred elsewhere, exhibits on transverse section about
twelve very distinct radiating septa meeting in the centre of the
corallites. Though all the other examples possess tabule and have
all the characters of Ohetetes, this raises the suspicion that possibly
the form may require, on more extended investigation, to be removed
from Cheetetes.

_ Locality and Formation. Cincinnati group, near Cincinnati, Ohio.
Collected by Mr. U. P. James.

10. CuHmreres Jamust, Nich. Pl. XXIX. figs. 10-10.

Corallum of cylindrical or subcylindrical hollow branches, the
diameter of which is from 3 to 5 lines; or of somewhat lobate and
subpalmate masses, the extremities of which are rounded. Branches
in the ramose forms dividing dichotomously at irregular intervals,
irregularly thickened and nodulose. Corallites oval or circular in
section, of unequal sizes. The larger corallites about six in the
space of 1 line, with very thick walls, the margins of the round or
subpolygonal calices being obscurely tuberculated or granulated.
The larger corallites are separated by extremely minute cylindrical
tubuli, the number of which varies in different parts of the corallum.
The surface exhibits no eminences or tubercles, or groups of large-
sized corallites ; but typical specimens exhibit at irregular intervals ©
stellate spaces, which are either solid or minutely punctate, and have
a diameter of two thirds of a line.

This species is related to Chetetes tumidus, Phill., of the Car-
boniferous rocks, especially in the rounded, thick-walled corallites,
separated by minute intermediate tubes. I have, however, com-
pared it with specimens of the latter, and find it to be clearly di-
stinguished by the larger size and much thicker walls of the corallites,

FROM THE LOWER SILURIAN OF NORTH AMERICA. 507

the generally greater number of the minute intermediate tubules,
the tuberculated margins of the calices, and the existence of the
curious stellate, solid or porous interspaces. The value of this last
character is diminished by the fact that some specimens, otherwise
the same, do not exhibit these spaces in a conspicuous manner.
The tuberculated margins of the calices, though this feature can
only be detected with the use of a high magnifying-power, remind
one of the species generally considered as belonging to the genus
Stenopora.

Locality and Formation. Not rare in the Cincinnati group, near
Cincinnati, Ohio. Collected by Mr. U. P. James.

11. Cumrerss rHompicus, Nich. Pl. XXIX. figs. 11-116.

Corallum ramose, the branches solid or hollow, subcylindrical,
from 4 to 5 lines in diameter, terminating in acutely pointed, or at
other times swollen and bulbous, extremities. Corallites with very _
thin walls, about eight or ten in the space of one line, variable in
form and arrangement. In parts of the corallum the calices are
regularly hexagonal; but over the greater portion of the surface
they are obliquely rhombic, and are arranged in regular diagonal
rows, the direction of which, however, changes within short dis-
tances, giving to the corallum a most peculiar appearance. In any
given portion of the surface in which this arrangement obtains, the
calices are disposed in a double series of diagonals, the one set
crossing the branch more or less transversely, whilst those of the
other set have a direction more or less corresponding with the axis of
the stem. The corallites are nearly equal in size; and there are no
very minute tubuli interspersed amongst the average tubes. Occa-
sionally a corallite may be slightly larger than the average ; but
the surface exhibits no tubercles, or regular groups of large-sized
corallites.

The present species is related in some degree to Ohcetetes (Monti-
culipora) Bowerbanki, Edw. & H.; but it differs altogether in its
mode of growth, especially when young, whilst the calices are only
about half as large.

Locality and Formation. Cincinnati group, near Cincinnati,
Ohio. Collected by Mr. U. P. James. .

Il. FronpEscENT AND PALMATE SPECIES.

The species of this group of the genus Chetetes have a corallum
which forms a flattened or undulated, often palmate or subpalmate
expansion, of small thickness, but often attaining a considerable
width and height. The corallum is composed primitively of two
layers of corallites, the bases of which are in contact, and which are
directed in opposite directions from a common caleareous membrane.
The corallum is rooted at the base, with which exception the entire
surface is covered by the calices on both sides. In old examples,
additional strata of corallites may be superimposed upon the two
primitive layers.

Aw 2

508 H, ALLEYNE NICHOLSON ON SPECIES OF CHATETES

12. Cuzreres MAMMULATUS, Edw. & H. Pl. XXX. figs. 3, 3a.

Chetetes mammulatus, Edwards & Haime, Pol: Foss. des Terr.
Paléoz. pl. xix. fig. 1.

Corallum forming irregular expansions of very considerable size,
sometimes palmate or lobate, carrying the polypes on both sides, and
from 2 to 4 lines in thickness. Surface covered with well-marked
tuberosities, the prominence of which varies in different specimens,
but which are covered with calices slightly larger than the average,
occasionally with a few intercalated tubes of minute size. The
tubercles are somewhat irregularly arranged in diagonal lines, and
are placed at intervals of from half a line to more than one line
apart. The corailites are subequal, polygonal, with thin walls, from
eight to ten in the space of one line. Very rarely, and quite as an
exceptional thing, one or two very minute tubules may be inter-
calated at the angles of the larger corallites.

This species represents C’. Dalei in the ramose group, but is di-
stinguished by its much less prominent tubercles and its mode of
growth. It is further distinguished by the small, thin-walled, poly-
gonal corallites, the absence of exceedingly minute intermediate
tubuli, and the fact that the surface-tubercles are covered with
calices which in general are rather above the average size. Our
numerous specimens are all broad undulating expansions of com-
paratively small thickness, and thus depart somewhat from the
typical form of the species.

Locality and Formation. Common in the Cincinnati group, near
Cincinnati, Ohio. Collected by Prof. Edward Orton and Mr. U. P.
James.

13. Cuzreres FRonpostvs, D’Orb.(?). Pl. XXX. figs. 2-2 0.

Chetetes frondosus, Edwards & Haime, Pol. Foss. des Terr.
Paléoz. pl. xix. fig. 5.

I am not at all satisfied that the specimens which have been
placed in my hands with this name are really identical with
D’Orbigny’s species, from which they seem to differ in some im-
portant points. I am, however, unable at this moment to refer to
D’Orbigny’s description of the species, and I shall therefore
describe our examples provisionally under the above title. If they
should prove to be new, the name of Chetetes ohioensis might be
applied to them. The characters exhibited by our specimens are as
follows :—Corallum forming erect, flattened, undulating expansions,
polypiferous on both sides, of unknown but considerable height, and
varying from less than 1 line to 3 lines in thickness. Calices
oval or subcircular, from eight to ten in the space of one line,
almost half their own diameter apart, separated by a great number
of exceedingly minute tubuli, which render the spaces between the
average corallites minutely porous. Surface with stellate spaces,
which are sometimes slightly elevated above the surface as very
low, rounded tubercles, but which are more commonly level with
the general surface, and which, instead of bearing average or large-

FROM THE LOWER SILURIAN OF NORTH AMERICA. 509

sized corallites, are entirely occupied by very minute tubuli similar
to those which separate the ordinary calices. -

Some of the thinner examples included under this head have a
resemblance to Monticulipora pavonia, D’Orbigny ; but their charac-
ters appear to be quite different.

Locality and Formation, Common in the Cincinnati group, near
eee Ohio. Collected by Prof. Edward Orton and Mr. U. P.

ames.

14. Cumrerus? cLaTHRatTuLus, James. Pl. XXX. figs. 1-16.

Stictopora clathratula, James, Catalogue of the Fossils of the Cin-
cinnati Group, 1871 (named, but not figured or described).

Corallum forming a thin undulating expansion of considerable
size, and about half a line, less or more, in thickness. The corallum
evidently grew in an erect position, as it consists of two layers of
corallites, which have their bases fixed to a common membrane, and
open on opposite sides of the corallum. The corallites are thus
about one quarter of a line in height, and they are slightly oblique
to the central lamina and also to the surface on which they open.
The calices are somewhat oblique, the lower lip very slightly promi-
nent, the walls moderately thick. About ten of the calices occupy
the space of one line, their size being nearly equal throughout, and
their general form being oval. The calices are arranged in regularly
decussating diagonal lines, which cross the corallum from side to
side in two intersecting directions. Surface with low, rounded,
obscure elevations, arranged in diagonal rows at intervals of from
one line to a line and a half, and occupied by cells which, as a rule,
are neither larger nor smaller than the average. The ordinary
corallites never have minute tubes interspersed amongst them ; but
there are often some small corallites on the elevations.

This beautiful fossil might be taken at first sight for a Ptilodictya,
though it appears to be clearly not of this nature. I have, how-
ever, been unable to make out the presence of tabule, and am
therefore uncertain what its true affinities may be. In the
form of the calices, in its mode of growth, and in the presence of
low tubercles more especially, it resembles a Cheetetes, and I have
provisionally referred it to this genus. The largest specimen in my
possession is an undulated expansion, broken on all sides, about
3 inches long by 24 inches wide, and about 3a line thick. My
description is drawn from type specimens presented to me by Mr.
U. P. James.

Locality and Formation. Cincinnati group, near Cincinnati,
Ohio.

TIT. Massrve anp Discorp SpECTEs.

Corallum fixed or free, massive, discoidal, hemispherical, spheri-
cal, or irregular in shape. The typical forms of this group have the
under surface of the corallum covered by an epitheca ; and these
must have had a free existence, as the under surface of the coral-

510 H. ALLEYNE NICHOLSON ON SPECIES OF CHATETES

lum is more or less deeply concave; others are fixed to some
foreign object; and the mode of existence of other examples is
unknown.

15. CHmTETES PETROPOLITANUS, Pander. Pl. XXX. figs. 5—5e, 6, 7, 8.

Favosites petropolitanus, Pander, Russ. Reiche, pl. 1. figs. 6, 7,
iL Oelalr

Favosites lycopodites, Vanuxem, Geol. New York, 3rd part, p. 46,
fig. 3.

” Cheetotes petropolitanus, Lonsdale, in Murch. Vern. & Keys. Russ.
& Ural, vol. i. pl. A. fig. 10.

Cheetetes Lycoperdon, Hall (ex parte), Pal. N. Y. vol.i. pl. 23. fig. 1,
and pl. 24. figs. 1 a-h, also vol. i. pl. 17. figs. 1 a—f.

Monticulipora petropolitana, Kdw. & Haime, Brit. Foss. Corals,

. 264.
_ Chetetes petropolitanus, Meek & Worthen, Geol. Illinois, vol. ii.
pl. 2. figs. 8 a, 6.

(Only the more important portion of the synonymy of this species
is here given.)

“ Corallum in general free; its basal plate flat or concave, and
completely covered with a concentrically wrinkled epitheca. Upper
surface regularly convex, in general hemispherical, and presenting
obtuse tuberosities about 1 line broad and varying very much in
height. In some specimens these tubercles appear to have been
worn away, and their existence is indicated only by the presence of
small groups of large calices, with thick walls ; the calices are rather
unequal in size, generally polygonal, sometimes almost circular ; the
largest are about one fifth of a line in diameter; the walls are not
perforated ; the tabulz are horizontal, complete, and placed at
about one twelfth of a line from each other. Some vestiges of septa
are often visible. Young specimens are flat and discoidal.” (Hd-
wards & Haime, Brit. Foss. Corals, p. 265.)

The Trenton Limestone and Hudson-River group yield a great
many examples which correspond with the above description in
essential respects—some altogether so, but others with more or less
striking variations. In form this species is protean, being more or
less discoidal when young, but being in its adult state subspherical,
hemispherical, subpyriform, lobate, mushroom-shaped, or shaped
somewhat like a cardinal’s hat. ‘The surface is sometimes mammil-
lated with obtuse tubercles or elevations of different sizes and vari-
able height; but quite commonly it is perfectly smooth, especially
in the smaller examples. Often no definite groups of large-sized
corallites can be recognized; but at other times such are present.
In all the specimens that have come under my notice the calices
are polygonal or subpolygonal, generally about ten in the space of
one line, sometimes fewer, without any very minute tubuli interca-
lated amongst them.

Besides the typical free-living examples of this species, with a con-
cave base and concentrically wrinkled epitheca, the Cincinnati group
yields a number of small, hemispherical, subspherical, smooth or

FROM THE LOWER SILURIAN OF NORTH AMERICA. ale

nodulated masses, which agree in other respects with the free
forms, but which are attached parasitically to foreign bodies, and
which therefore do not possess a concave under surface. I am,
however, unable to decide whether this of itself should be re-
garded as a character of specific value, though it would seem not
to be so.

Locality and Formation. Common, and very variable in form, in
the Trenton Limestone of Belleville, Peterborough, and other locali-
ties in Ontario. Abundant in the Hudson-River group of Ontario.
Less frequent in the Cincinnati group of Ohio.

16, CuzrereEs piscorpEeus, James. Pl. XXX. figs. 4-4d.

Chetetes discordeus, James, Catalogue of Fossils of the Cincinnati
group, 1871 (named, but not figured or described).

Corallum free, discoid, plano-convex, sharp-edged, from 5 to 8
lines in diameter, and about 1 line in greatest thickness. Under
surface concave, covered with a very thin, smooth, and not regu-
larly striated epitheca, which usually exhibits two or three con-
centric wrinkles. In general, the epitheca is so delicate as to reveal
clearly through its substance the bases of the superjacent corallites.
Upper surface gently convex, not exhibiting any tubercles or
elevations of any kind. Corallites subequal; calices with mode-
rately thin walls, polygonal, from eight to ten in the space of one
line. No groups of larger-sized corallites, nor any very minute
intermediate tubuli.

I do not feel altogether certain that this form is distinct from the
young of Chetetes petropolitanus. It is, however, a common form,
and is very constant in its dimensions. Apart from its discoidal
plano-convex form, it is distinguished by its great tenuity (compara-
tively speaking), the sharp thin edges of the disk, the absence of
surface-tuberosities or groups of large-sized corallites, and the
extreme thinness of the epitheca, which is transparent and is not
regularly striated concentrically. But for one character, I should
have been disposed to have placed this species under Cheetetes (Nebu-
lipora) lens, M‘Coy ; and that is the absence in our examples of any
groups of large-sized corallites, whereas their presence is a marked
feature in the latter. The underside of C. discoideus resembles
Lichenalia calycula, James; but it may be distinguished by the
absence of radiating striae, and by other characters as well. The
above description of the species is drawn from type specimens for-
warded to me by Mr. U. P. James.

Locality and Formation. Cincinnati group, near Cincinnati, Ohio.

TV. Incrustine SPEcrEs.

In this group the corallum is parasitic, and forms a thin crust
composed of a single layer of short corallites, which are attached by
their bases to some foreign object, such as the shell of a Brachiopod,
the epitheca of a coral, or the column of a Crinoid.

512 H, ALLEYNE NICHOLSON ON SPECIES OF CHATETES

17. Cuamreres papittatus, M‘Coy. Pl. XXIX. figs. 12-12 d.

Nebulipora papillata, M‘Coy, Brit. Pal. Foss. p. 24, pl. 1 C. fig. 5.

Cheetetes tuberculatus, Edw. & Haime, Pol. Foss. des Terr. Paléoz.
pl. 19. figs. 3, 3a.

Monticulipora papilllata, Kdw. & Haime, Brit. Foss. Corals, pl.
62. figs. 4, 4a.

Corallum forming an exceedingly thin crust, usually not more
than half a line in thickness, growing upon Brachiopods and other
foreign objects. Surface exhibiting rounded tuberosities, sometimes
very slightly elevated, placed about their own diameter apart, and
occupied by corallites of a larger size than the average. Corallites
polygonal, thin-walled, somewhat variable in size, not having very
minute tubuli interspersed amongst them. About eight or ten of
the average corallites in the space of one line. Width of the
tubercles one line or a little more.

Our specimens agree well with the description given by Mr. M‘Coy
of Nebulipora papillata ; but they differ somewhat from that given
by Edwards & Haime. Thus the tubercles in our examples are
rounded, not to any marked extent elongated or compressed; and
the large corallites occupying these eminences are from five to six
in one line (instead of being one-third of a line in width).

The tubercles are not uncommonly perforated by regular and
large circular perforations, which have a diameter of about half a
line, and appear to be the mouths of vertical tubes. I have, how-
ever, noticed similar perforations in several other species of Cheetetes ;
and they are probably extraneous to the coral itself.

Locahty and Formation. Cincinnati group, near Cincinnati,
Ohio.

- 18. Cazrerss corticans, Nich. Pl. XXIX. figs. 13, 13a, 14.

Corallum forming a thin crust, growing parasitically upon foreign
bodies, and having a thickness of less than one quarter of a line, or,
when in superimposed layers, attaining a thickness of half a line.
Surface exhibiting a number of long, narrow, compressed tubercles,
which are all drawn out in one direction, and which, whilst carry-
ing average-sized corallites on their sides, appear to be compact at
their summits. The long diameter of these tubercles varies from
two thirds of a line to two lines, their width not exceeding half a
line, and their height being variable. They are arranged in irre-
gularly diagonal lines, and are placed at intervals apart of from
half a line to two thirds of a line. The corallites are thin-walled,
polygonal, subequal, from eight to ten in the space of one line,
without any intermediate minute tubuli. There are also no groups
of large-sized corallites.

This species is allied to C. papillatus, M‘Coy, but appears to be
sufficiently distinguished by the long, narrow tubercles, which do
not earry any large-sized corallites. The specimens which I have
seen are found coating Orthocerata, and forming thin but extended
crusts. In some examples the tubercles are depressed, and appear

FROM THE LOWER SILURIAN OF NORTH AMERICA. 513

to be solid throughout ; but in other cases they are elevated, and
carry corallites of the ordinary size on their lateral aspects.
Locality and Formation. Cincinnati group, near Cincinnati, Ohio.

19. Cumretzs Ortoni, Nich. Pl. XXIX. figs. 15-15 6.

Corallum forming exceedingly thin crusts, not more than from
one sixth to one eighth of a line in thickness, attached parasitically
to submarine objects. Crusts usually constituting small circular

' expansions, or irregular and indefinite in form. Surface exhibit-
ing numerous, minute, rounded or obtusely conical eminences,
which are placed at intervals apart of half a line, more or less. The
tubercles are somewhat compact at their summits, and carry upon
their sides corallites which are little or not at all larger than the
average. The corallites are somewhat oblique to the surface, mode-
rately thick-walled, subequal, without any minute intercalated
tubuli. Calices small, subpolygonal or oval, from ten to twelve in
the space of one line; their margins thick and surmounted by minute
and crowded miliary granules, which are rounded and not spinous,
and which are placed almost in contact with one another.

In appearance, this species, but for its extreme thinness, and the
small close-set surface-tubercles, might be confounded with C. pa-
pilatus. On examination with a high power, however, it is readily
distinguished from all the known incrusting species of Chetetes by
the fine and close granulation of the margins of the calices, giving
to the surface quite a peculiar appearance. All the specimens I
have seen of this singular species are parasitic upon Strophomena
alternata; and I am informed by Mr. James that it is rarely found
attached to any thing else. In spite of the granulated margins of
the calices, it appears to be a genuine Chetetes, and I see no reason
for removing it from this genus.

I have named the species after my friend Prof. Edward Orton, of
the Geological Survey of Ohio, from whom, as well as from Mr.
U. P. James, I have received the specimens from which the above
description is drawn.

Locality and Formation. Cincinnati group, near Cincinnati, Ohio,

EXPLANATION OF THE PLATES.

Pratt XXIX.

Fig. 1]. Fragment of Chetetes Dalei, Edw. & H., of the natural size. (In this,
as in all the following figures of Chetetes which are not enlarged,
the calices have been omitted for the sake of clearness, and only the
general form of the corallum is shown, together with any superficial
prominences which may be present.)

la. A portion of the preceding, enlarged, showing the very minute inter-
mediate tubuli.

2. Fragment of Chetetes rugosus, Hdw. & H., of the natural size.

3. Fragment of Chetetes approximatus, Nich., of the natural size, showing
the small tubercles.

3a. A portion of the preceding, enlarged, showing a surface-tubercle and
very minute tubuli between the average corallites. Few specimens,
however, exhibit as many of these intermediate tubuli.

514

Fig.

Fig.

H. ALLEYNE NICHOLSON. ON SPECIES OF CH ATETES

4, Fragment of Chetetes attritus, Nich., of the natural size.

4a. Portion of the same enlarged, showing the calices and the small solid
tubercles.

5. A fragment of Chetetes pulchellus, Edw. & H., of the natural size.

5a. Portion of the same, enlarged, to show the groups of larger corallites.

56. or still further enlarged, showing the few minute intermediate
tubuli.

6. Fragment of Chetetes Fletcheri, Edw. & H., of the natural size.

6a. Portion of the same, enlarged.

7. Fragment of Chetetes gracilis, James, of the natural size. :

7a. Fragment of the same, enlarged, showing the small crowded calices
and very minute intermediate tubuli.

8. Fragment of Chetetes delicatulus, Nich., of the natural size.

8a. Fragment of the same, showing a stem which terminates in a rounded
swollen extremity, and apparently springs from a horizontal foot-
stalk, of the natural size.

8b. Portion of fig. 8, enlarged.

9. Fragment of a large individual of Chetetes? nodulosus, Nich., of the
natural size.

9a. Portion of the same, enlarged.

10. Fragment of Chetetes Jamesi, Nich., of the natural size.

10a. Portion of the same, enlarged, showing one of the stellate spaces.

10%. Another portion of the same, still further enlarged, showing the
thick-walled calices and their granulated borders.

11. Fragment of Chetetes rhombicus, Nich., of the natural size.

11a. Portion of the same where the cells are rhombic and arranged in
diagonal rows, enlarged.

11 4. Another portion of the same, where the cells are hexagonal, enlarged.

12. Chetetes papillatus, M‘Coy, growing upon a Strophomena, of the
natural size.

12 a. A few cells of the same, highly magnified.

12%. A portion of the same, not so greatly enlarged, showing one of the
tubercles.

13. Chetetes corticans, Nich., growing on an Orthoceras, of the natural size.

13 a. Portion of the same, enlarged, showing one of the tubercles.

14. Fragment of another specimen of C. corticans, in which the tubercles
are smaller and not so pronounced, growing on a large Orthoceras,
of the natural size.

15. Chetetes Ortoni, Nich., growing on Strophomena alternata, of the
natural size.

15 a. Portion of the same, enlarged, showing the tubercles.

15 }. Portion of the same, enlarged further, showing the granulated margins
of the calices.

Prats XXX.

1. Fragment of Chetetes ? clathratulus, James, enlarged, showing the dia-
gonal arrangement of the cells and the tubercles.

1 a. Section of the same, enlarged, showing the two strata of tubes springing
from a common lamina.

1b. Portion of the same, enlarged still further.

2. Fragment of Chetetes frondosus, D’Orb. (?), of the natural size, showing
the stellate spaces.

2a. Section of the same, enlarged, showing the two strata of corallites
arising from a common membrane.

2b. Portion of the same, enlarged, showing the calices and a stellate space.

3. Fragment of Chetetes mammulatus, Edw. & H., natural size.

3 a. Portion of the same, enlarged.

4. Upper surface of Chetetes discoideus, James, natural size.

4 a. Under surface of the’same, natural size.

4 }. Profile view of the same, natural size.

4c. Portion of the upper surface of the same, enlarged.

4 d. Portion of the lower surface of the same, enlarged.

Quart. Journ .Geol.Soc. Vol. XXX Pl. XXIX.

Per neet id
X64 no hen o/
4.

Mintern Bros.imp.

GH Ford.

AMERICAN CHAV'TETES.

( PLXXX,

Quart. Journ.Geol Soc. Vol.

CUM TT

Ww I

carmran ten
yh

Hi

[EN NSS SENNA SASNSSNARS
\POGATI A OP ot fi
WLLL ee

4

ee

Dk

ton
&

a x

‘as #

Mintern Bros.imp .

GH¥Ford.

AMERICAN CHA'TETES.

FROM THE LOWER SILURIAN OF NORTH AMERICA. 515

Fig. 5. Upper surface of a young specimen of Chetetes petropolitanus, Pander,

of the natural size.

5 a. Under surface of the same, natural size.

5 5. Profile of the same, natural size.

5c. Portion of the surface of the same, enlarged.

6. A specimen of Chetetes petropolitanus (?) attached to a Brachiopod,
natural size.

7. Another specimen of the same (?), forming an apparently unattached
nodulated mass, of the natural size.

8. A fiattened specimen of the same (?), having the form of a cardinal’s
hat, of the natural size.

516 J. W. HULKE ON A REPTILIAN TIBIA AND HUMERUS

39. Nore on a Rupritian Trsta and Humervs ( probably of Hyleeo-
saurus) from the WEALDEN Formation ?n the Istz or Wieut. By
J. W. Hurxe, Esq., F.R.S., F.G.8. (Read June 10, 1874.)

[Puare XXXL]

I am indebted to Dr. Wilkins, of Newport, who has already so fre-
quently afforded me valuable materials for the study of the fossil
Reptilian fauna of the Isle of Wight, for the opportunity of bringing
two very remarkable limb-bones before this Society. They were
obtained several years ago, in Brixton Bay, by a fisherman since dead.
The soft and brittle state of their tissues, the complete substitution
throughout them of the red oxide of iron for the pyrites with which
bones in this locality are usually impregnated, their envelopment
in a concretionary matrix of this oxide and clay, and their incrus-
tation with recent zoophytes and alge concur in making it very
probable that they had been long exposed to the waves and winds
upon the shore between high- and low-water marks before they were
discovered. The horizon of their gisement cannot be fixed more
nearly than somewhere in the mottled purple and grey clays, there-
fore in the beds west of Cowleaze Chine, below the Hypsilophodon-
bed. It is not even known that they were found lying so close
together as to justify the inference from juxtaposition that they
originally formed part of one skeleton. Dr. Wilkins informed me,
when I first saw them in 1870, that he believed they were so asso-
ciated when discovered; and the close agreement of their general
facies, of their texture, of their mineralization, of the matrix about
.them, and of their algal and zoophytic crusts concur in rendering
this extremely probable, and dispose me to regard them as members
of one individual.

Remarkable shortness relatively to their bulk, singularly dwarfed
shafts, and as notably expanded articular extremities characterize
both bones.

Tibia (Pl. XXXI. figs. 1 & 2).—The general resemblance of this
bone to the tibize of such typical Dinosaurs as Jqyuanodon Mantelli
and Megalosaurus Buckland: leaves no doubt of its place in the
skeleton ; but its proportions make it impossible to mistake it for the
tibia of either of these Sauria.

Its total length is 16 inches.

The proximal end is so very massive that it greatly overhangs the
slender shaft. An inner and an outer condyle, the latter separated
behind by a wide, deep, roundish notch from a stout praecnemial
process, are obscurely mapped on the upper or articular surface. Its
antero-internal surface, and the prwecnemial process into which the
former is prolonged outwards, are rough as if for the attachment of
muscles and ligaments; and a rather strong tuberosity marks the
junction of the internal and anterior portions. The long diameter
of this end, passing through the inner condyle and the base of the
preecnemial process, measures about 7 inches; and when the back
of the condyles rests upon a flat surface it includes with this surface

FROM THE WEALDEN FORMATION IN THE ISLE OF WIGHT. 517

an angle of about 35°; and it crosses the long axis of the distal end
at an angle which in this specimen I estimate roughly at about 118°.
Below the knee the tibia contracts quickly to a very short and
slender shaft, in its unexpanded part not more than 23 to 3 inches
long, and with a minimum girth of 7-5 inches, the girth of the prox-
imal articular end measuring 21°5 inches. The cross section of the
shaft is subtrigonal. The cortex of this part is compact, externally
smooth, and finely grained. No large medullary canal is present, as
in Iguanodon Mantel, but a moderately fine cancellated tissue
nearly fills the interior.

Downwards, the shaft expanding and becoming flattened, merges
into the broad triangular ankle end, one angle or malleolus of which
is antero-internal, the other postero-external, while the surfaces are
respectively directed outwards and forwards, and backwards and in-
wards. ‘The antero-external surface is undivided and nearly plane,
while the postero-internal surface is broken by an angulated ridge into
a smaller inner and a wider outer part. The under surface of this end
is pulley-shaped, having a long, narrow, postero-external and a shorter,
wider, antero-internal division. Its broad, shallow, trochlear groove
ascends upon the posterior surface, external to the angulated ridge
lately mentioned ; and opposite to it, on the anterior surface, is a
prominent tubercle. This trochlear part, with its anterior tubercle,
is evidently homologous with the Jywanodon’s separate astragalus.
The condition of the specimen does not allow me to say whether
this representative of the astragalus and the distal end of the
tibia had become confluent during life, or whether they have simply
become undistinguishably blended by post-mortem pressure and
mineralization. The fact that the inner division of this trochlea stops
short of the antero-internal angle or inner malleolus, and is not now
coextensive with the entire under surface of this part of the tibia, is
suggestive of a post-mortem displacement, and, if so, of the absence
of a genuine bony anchylosis.

The proportions of this tibia are, within my knowledge, only re-
peated in the tibie of Hylwosaurus and Polacanthus Foxit. The
materials for comparison are limited to a single example of each; and
the account of the tibia of the latter in the Rev. W. Fox’s narrative
of his discovery of a great part of the skeleton of this remarkable
armour-clad Saurian, communicated by him to the British Associa-
tion in 1866, as also the short memorandum, with a sketch of a few
of the bones, which appeared in the ‘ Illustrated London News’ of
that year, are rather ‘‘ Preliminary Notices,” which scarcely claim
to afford the details so necessary to the anatomist who cannot study
the actual specimen.

Table of Measurements.

inches.
Length of tibia, including distal epiphysis or astragalus 16
Lone diameter of proximalvends.2. 9... .-.--0-. il
Girthjot: Ci GOwe St” eiseiys eee entre sta chaest ea ye cto 21°5
Girthvot shatts So foe ts re pent lee ce tees «ae 75

Nong diameter of distal end)" se. 5... 6 a ae 8°25

518 J. W. HULKE ON A REPTILIAN TIBIA AND HUMERUS

Mr. Fox (loc. cit.) described the tibia and fibula of Polacanthus
as being 13 inches long, with broad ends, like those of Hyleosaurus,
the diameters of the distal end 73 and 33 inches. I am not certain
whether the larger of these latter numbers applies to the breadth of
the tibia alone, or includes the still connected fibula, and also whether
13 inches is the length of the tibia, or of this with its lower epiphysis
or astragalus. The extensive ossification of the tendons made it
probable that the skeleton was that of a fully grown individual.
Mr. Fox has verbally informed me that the bones of his Polacanthus
have a coarse texture, and that the femur is devoid of a medullary
canal. The impression which I retain from a cursory view of Mr.
Fox’s tibia two years ago is of a much less bulky bone than this
specimen.

The tibia of Hyleosawrus obtained by Dr. Mantell at Bolney, now
in the British Museum, probably represents the size of the adult
shin-bone of this Saurian, if it actually formed part of the same
skeleton with the associated scapule ; for these (if their much larger
size, greater massiveness, and stronger features than those in the
well-known Tilgate-forest slab are trustworthy criteria) very likely
belonged to an old fully grown individual. Its length is the same
as that of Dr. Wilkins’s specimen. Its proximal end is similarly
massive, and its distal end is as remarkably expanded; the shaft also
is strongly twisted. The condyles are better marked, and the lower
end, owing to the disassociation of the astragalus, has the usual Di-
nosaurian notch instead of the distal trochlea of Dr. Wilkins’s bone.
With these close resemblances I prefer to refer this tibia to the
same genus, though there are differences of detail which may
possibly denote another species than Mantell’s, possibly also may
merely be the expression of differences of age or sex.

Humerus (Pl. XXXI. figs. 3 & 4)—The bone which I regard asa
left humerus shows in an extreme degree the dwarfing of the shaft,
the expansion of the extremities, and the massiveness relatively to
length exhibited by the tibia. The total length is 16-5 inches, the
breadth of the proximal end 8 inches, and that of the distal end
rather under 7 inches. The shaft is as short as that of the tibia.
The girth of the proximal end is 20 inches, that of the distal end is
18°5 inches, and that of the shaft 9 inches. ‘The proximal border,
very thick and rounded, swells out at about 3 inches distance from
its ulnar end, and assumes a roughly trigonal convex form, the outer
limit of which falls where the proximal and radial borders meet.
The base of this sessile articular caput falls in the ventral lip of the
proximal border; and its rounded dorsal apex marks the stoutest
part of the border, where its thickness reaches 3:6 inches. The
radial border, at first thinner than the proximal, descends straight
for the space of 3 inches, and then expands into a knob-like swelling,
from which it abruptly declines, ending in a ridge which is lost on
the ventral or anterior surface of the short and slender shaft above
the radial tuberosity. The ulnar or posterior border is strongly
concave. ‘The ventral or anterior surface of the expanded proximal
part is very hollow transversely, haying the shape of a wide trough.

FROM THE WEALDEN FORMATION IN THE ISLE OF WIGHT. 519

The dorsal surface of the same part is transversely sinuous, being
laterally hollow, and rendered centrally convex by a stout ridge-
like continuation of the shaft, which, expanding upwards, largely
supports the sessile, trigonal, articular caput. The distal end has
less perfectly preserved its form. A trochlear groove, wide and
shallow distally, prolonged also as a wide shallow trough on the
dorsal surface of the shaft, and deep and notch-like ventrally, very
distinctly mark out an ulnar and a radial condyle. The latter is
ventrally surmounted by a tuberosity, the perfect form of which is
now lost. The short and slender shaft has a roughly trigonal cross
section ; itis most convex dorsally, where one of the rounded angles
falls, and less so laterally and at its ventral surface. The shaft
is smooth, while the rest of the surface of the bone, especially
towards the articular extremities, is very rough.

If, disregarding the hint derivable from the very probable asso-
ciation of this bone with the tibia just described, we look around
amongst the contemporary Sauria of the Wealden formation for the
owner of such a humerus, we find Jguanodon Mantelli and Megalo-
saurus Bucklandi immediately excluded, because their humeri are
well known and very different. There remain for our scrutiny the
Streptospondyli (S. major and S. recentior), Owen, the (Wealden) -
Cetiosauri, Owen, Polacanthus Fox, and Hyleosaurus of Mantell.
Of these, Streptospondylus major has still a doubtful individuality.
Prof. Melville, if I mistake not, long since suggested, with much
probability, that the vertebrae, then and still the only evidence on
which this species was founded, belonged to the cervical and pectoral
series of Iguanodon Mantelli; and although this may not be con-
ceded, the individuality of S. major has not, I think, been so de-
monstrated as to render it impossible that these vertebre may
actually belong to one of the Sauria included in the genus Cetio-
saurus as exemplified in the Wealden fauna—for instance to C. brevis,
Owen, Streptospondylus recentior, Owen, Ornithopsis Hulkii, Seeley.
Eucamerotus, mihi, has, as I have pointed out in a former paper,
strong formal resemblances, as regards its vertebra, to those of
Cetiosaurus oxonensis; but the texture of these bones is very
different. Until recently only vertebrze had been found; but last
autumn portions of limb-bones were recovered by the Rev. W. Fox
and by myself; and as these have the same large-celled texture as
the vertebra, it is presumable that the humerus is similarly consti-
tuted, which this humerus is not. With these Sauria, and with the
Cetiosaurus brevis, Owen, of the Wealden, and also with Pelorosaurus
no comparison can be instituted. Turning to Hyleosaurus and to
Polacanthus Foxii, the fore limb of the latter was missing from
Mr. Fox’s find, and no part of its shoulder-girdle has been recog-
nized; but in the scapula of the former we observe an extraordinary
massiveness of its articular end, and the large characteristic trans-
verse crest—conditions which harmonize well with the great expan-
sion of the proximal end of this humerus. This obvious suitability,
coupled with the strong resemblances of the tibiee, constitute better
grounds for referring the humerus also to Hylwosawrus than to any

520 J. W. HULKE ON A REPTILIAN TIBIA AND HUMERUS.

of its contemporaries. J need hardly say that the bone which
Mantell originally regarded as the humerus of Hylwosaurus was
shown by Prof. Owen to be its tibia. There is also in the National
Collection a bone labelled ‘‘ Humerus of Hyleosaurus ;” but both ex-
tremities are so mutilated that the determination is more than doubt-
' ful; if correct, the bone must have belonged to a very immature in-
dividual.

Supposing that Dr. Wilkins’s specimen should prove to be the
genuine humerus of Hylwosaurus, it is evident, on the assumption that
it and the tibia belonged to one individual, that the proportions of the
fore and hind limbs must have been very different from those which
obtain in Jguanodon and Megalosaurus, in both which the fore limbs
are, relatively to the hind limbs, remarkably reduced.

EXPLANATION OF PLATE XXXtI.
Reptilian Tibia and Humerus.

Fig. 1. Front view of tibia: pr. proximal, d. distal end; 0. m. outer malleolus;
z.m. inner malleolus ; pre. preecnemial crest.
2. Back view of the same tibia.
3. Oblique view of ventral surface of humerus: ec. capitulum; *. radial
condyle; d. deltoid border.
4. Dorsal surface of the same.

G.H Ford .& CL.Griesbach.

TIBIA AND HUMI

Quart. Journ .Geol. Soc. Vol. XXX Pl XXYI .

Mmtern Bros .imp -

Be ene tia

la
ep)
a
5
=
a
e)
Rg
:
a
tn
ip

Quart. Journ .Geal. Soc. Vol. XXX. Pl. XXX1 .

Mintern Bros. imp .

G.H Ford .& CL.Griesbach.

TIBIA AND HUMERUS OF HYLEOSAURUS ?(x3)

J. W. HULKE ON A MODIFIED FORM OF DINOSAURIAN ILIUM, aoe

40. Note on a@ Mopirrep Form of Drnosaurtan Ixium, hitherto
reputed Scaputa. ByJ. W. Hurxs, Esq., F.R.S., F.G.S. (Read

June 10, 1874.)
[Pirate XXXII]

Tux subject of this note is a bone the true place of which in the
skeleton, I suggest, with much deference to the opinions of the emi-
nent authorities who have described it, has been misapprehended.
If the interpretation I suggest should prove correct, it will indicate
an altogether new genus, perhaps even a new order.

The oldest known example originally formed part of Dr. Mantell’s
collection. Subsequently it was purchased by the British Museum,
where, labelled “ Scapula of an unknown Reptile,” it is exhibited
amongst the Megalosaurian remains. Dr. Mantell originally referred
it to his then newly discovered Jguanodon, but not without a caution.
He described it in the following terms :—

«A scapula or omoplate 18 inches long, associated with bones or
teeth of Iguanodon, and probably referable to that animal, presents,
like the coracoid, some important modifications of the usual lacer-
tian type. This bone is of an elongated form, and differs consider-
ably from the omoplate of the Monitors and Iguanas. It somewhat
resembles the scapula of the Scink, and it throws off a long tripar-
tite apophysis, which is imperfect in the only specimen hitherto dis-
covered; this process probably afforded support to a cartilaginous
arch, as in the recent lizards. But although, from circumstances,
I entertain but little doubt that the coracoid and omoplate above
described belong to the Iguanodon, it is so hazardous in paleonto-
logical inquiries to affirm-as certain what is merely probable, and
so many impediments to accurate inductions have been occasioned
by hasty and positive determination of a tooth or bone from imperfect
analogies, that I deem it necessary to repeat that these specimens
were not found in juxtaposition with other parts of the skeleton
of the Iguanodon, but were merely imbedded in the same mass of
rock.” (Phil. Trans. 1841, p. 138.)

This opinion was adopted by Prof. Owen :—

“The scapula,” he wrote (in 1841), “has not hitherto been dis-
covered so associated with other unequivocal portions of the skeleton
of the Jguanodon as to permit the characteristics of this bone to be
confidently recognized. The bone (No. 194, omoplate of Jguanodon,
Mantellian Catalogue) agrees with the undoubted scapula of the
Hylosaur and with that of certain lacertians, especially of the
genus Scincus (Dr. Mantell has pointed out this resemblance in his
Memoir in the ‘ Phil. Trans.’ 1841), in the production of a long
and slender pointed process continued nearly at right angles with
the body of the bone from the anterior part of the articular surface
for the coracoid; but it differs from the scapula of the Hyleosaur
in the presence of two shert processes given off from the lower part
of the base of this long process, and in the absence of the thick and

Q.J.G.8. No. 120. 20

522 J. W. HULKE ON A MODIFIED FORM OF DINOSAURIAN ILIUM.

strong transverse acromial ridge which overhangs the glenoid depres-
sion, and in the deeper concavity of the posterior margin of the as-
cending plate or body of the bone. This part, in its shape, relative
ength and breadth, is intermediate between the crocodilian and lacer-
tian type of the scapula, at least as exemplified in the Monitors and
Iguanas. The Scincs and Chameleons in the more crocodilian pro-
portions of their scapule resemble the Hyleosaur and the great
species of extinct Saurians, most probably the Jgwanodon, to which
the present bone belongs.” (‘‘ Reports on British Fossil Reptiles,” in
Brit. Assoc. Rep. 1841, p. 133.)

Haying subsequently discovered a scapula of a different shape, the
coracoid border of which he saw to be adapted to that of an unequi-
vocal Jguanodon coracoid, and having verified his new acquisition by
identifying it with a pair of bones in the type slab of the Maidstone
Iguanodon, Mantell abandoned his original reference of the first sca-
pula to this herbivorous Saurian, but he continued to regard it as a
scapula. In his ‘ Fossils of the British Museum,’ 1851, at pp. 282,
283, he reproduced from the ‘Phil. Trans.’ 1841 his first description
of it, under the new title “Scapula of an unknown Reptile.”

Mantell’s determination of the genuine Jguanodon scapula was soon
confirmed by Mr. G. B. Holmes’s valuable discovery, near Horsham,
of an Jguanodon scapula in union with its coracoid. ‘These bones
were described and figured by Prof. Owen in his monograph on the
Iguanodon in the ‘ Memoirs of the Palzeont. Soc.,’ issued in 1855.
In the following year, in a monograph on Megalosawrus, in the same
Society’s Memoirs, Prof. Owen, “with a full sense of the inadequacy
of our present evidence for the precise determination of its scapula,”
assigned to this Dinosaur a more perfect example of Mantell’s first
scapula, discovered at Stammerham by G. B. Holmes, Esq.

This second specimen is thus described :—‘ The body of the bone
is an oblong flattened plate proportionally broader and stouter than
in [qguanodon, with the base rounded, not truncate as in Hyleosaurus,
and with the anterior border at first, as it descends, straight, and
then concave, not convex as in Hyleosaurus. The body of the sca-
pula decreases in breadth as it approaches the articular end, near
which there is continued from the anterior border a long and slender
process at least three fourths the length of the entire bone, but the
precise proportions of which cannot be determined in this specimen,
because the extremity of the process is broken off. Near the base
of the process a tuberous projection is developed, which touches the
anterior end of the articular end of the scapula, circumscribing an
elliptical vacuity, probably for the transmission of vessels. The
thickened articular extremity shows indications of a division into two
surfaces, one for the coracoid, the other for the humerus.”

The figure of this specimen (tab. v. fig. 1) is explained to be one
fourth of the natural size. Mr. Holmes writes to me that this is an
error; it is accurately drawn from exact measurements.

To me the ‘Scapula of an unknown Reptile’ was an enigma from
the time I first began to study the Dinosauria; for on the supposi-
tion that it belonged to Megalosaurus, which its place in the National

J. W. HULKE ON A MODIFIED FORM OF DINOSAURIAN ILIUM,. 025

Collection and in Prof. Owen’s monograph on this Saurian scarcely
allowed me to doubt, I could not perceive how such a bone could be
united to that which Cuvier and Owen had taught me to regard as
its coracoid.

The first light reached me from Prof. Huxley’s valuable paper
on the “Classification of the Dinosauria,” read at a Meeting of
our Society, 10th Nov., 1869, in which the previously reputed
coracoid of Megalosaurus was shown to be its ilium, and the forms
of the genuine coracoid and scapula were made known (cf. Prof. J.
Phillips’s letter of July 1, 1869, here given). I now knew certainly
that the “scapula of an unknown reptile,” relegated from [quanodon
to Megalosawrus did not enter into the shoulder-girdle of the latter.
At one time I conjectured it might be an os pubis, in its spatulate
form haying some resemblance to that of the crocodile ; but becoming
better acquainted with Dinosaurian ilia I began to suspect that it
might be a modified form of ilium, and on comparing it with the
ilia of Iguanodon Mantelli and of Hypsilophodon Foawii I was grati-
fied to find its leading features repeated in these. In my following
autumn holiday (1870) I saw in Dr. Wilkins’s collection two pre-
viously unrecognized bones obviously identical with Mantell’s type
* scapula of an unknown reptile.” Wholly stripped of their matrix,
so that they could be viewed from every side, they illustrated some
points in their structure better than the only two previously known
specimens, and substantiated, I thought, my surmise that the reputed
scapula wasanilium. Dr. Wilkins has lately very courteously afforded
me an opportunity of studying these bones at my leisure, and he
kindly permits me to exhibit them this evening.

The smaller bone (Pl. XXXII. fig. 1) formally repeats Mantell’s
and Holmes’s specimens. Its glenoid (or acetabular) end is in some
of its details much more complete than theirs ; but a considerable part
of the body is missing ; it has been broken off at the distance of one inch
above the upper root of the long anterior process, which has been
likened to a connate clavicle. From the level of this process upwards
the front and back margins of the body converge gently, with a corre-
sponding decrease of width of the body, as far as this has been pre-
served. The anterior border is thin and sharp, the posterior border is
thicker. The greatest thickness of the body, where broken across, is
‘7inch. Towards the glenoid (or acetabular) end the body widens out
and it thickens. The terminal aspect of this end is chiefly occupied
by an arched, hollow, smooth, articular surface (fig. 1, a, & fig. 2), the
chord of which forms nearly a right angle with the axis of the ex-
panded blade or body of the bone, and in its now mutilated condition
measures about 3:5 inches. In the direction of its short diameter, or
transversely, this articular surface measures at its anterior extremity
1 inch, which increases to 1°5 inch at its posterior end; and in this
direction the surface is slightly concave. The uninterrupted smooth-
ness of this arched surface throughout its whole extent unmistakably
indicates that the whole formed one part of a joint in which another
bone played freely as a ballin a socket. At each end this arc is termi-
nated by a process, of which the anterior ()) is small (in the speci-

202

524 J. WwW. HULKE ON A MODIFIED FORM OF DINOSAURIAN ILIUM.

men it is mutilated; but in the larger bone it is a thin compressed
lip, which projects strongly downwards and forwards), while the
posterior is stout and blunt. This latter process (¢) forms the poste-
rior inferior angle of the bone, and it is its stoutest part. Its trun-
eated terminal aspect looks downwards and backwards, and has the
characteristic rugosity of a synchondrosis.

The long slender upper process:(Spp) given off from the anterior
margin of the body measures along its upper surface 7inches, and about
1 inch less along its under surface; the free end is missing. From
its base to the free end it makes a gentle double curve, its upper
outline being first convex, then coneave. The cross seetion of this
process is roughly trihedral ; its under surface is marked with coarse
longitudinal grooves. At 4:3 inches from its free end it bears a
small process, now mutilated, separated by a wide subcircular notch
from that which limits anteriorly the large articular arc. This little
process is clearly the anterior of the two small projections at the
under surface of the long slender process which tend to convert the
notch between them into a closed ring in Mantell’s specimen. ‘The
two remaining surfaces of the long slender process (one upper, the
other lateral) are smooth. From their union, near the base of the
process, a ridge (7) passes backwards at the distance of 1 inch above
the articular arc, concentrically with it, which, gradually widening
and becoming less conspicuous, is lost near the posterior inferior
angle of the bone.

Dr. Wilkins’s second specimen is from a much larger individual ;
a much greater portion of the body and smaller piece of the long
anterior process are preserved (Pl. XXXII. figs.3 & 4). The smooth
(enarthrodial) arch (a) at the lower end of the body is flattened and
distorted by pressure ; its chord now measures 7°5 inches. In front
it ends in a sharp compressed lip, which projects strongly downwards
and forwards; and behind it terminates in a low massive hummock,
which, unprotected by matrix, is worn smooth. A splintered, muti-
lated projection, which, as in the smaller specimen, the long slender
process detaches, near its base, from its under surface, inclines down-
wards, and tends to close the notch which separates it from the ar-
ticular are, and to convert it into a foramen (;).

Of the long slender process (Spp) only about 6 inches remain. Its
cross esction is an oblong, the long axis of which is nearly vertical to
that of the same section of the body of the bone. The upper surface is
gently convex transversely, and it is smooth; the under surface is
longitudinally grooved and rough. ‘The greater relative breadth of
the body near the lower articular end, and the obtuse angle which
the long slender process makes with the body, are matters in which
this differs conspicuously from the smaller specimen. Perhaps, if
both bones were complete, these dissimilarities might appear less.

If now we test these four specimens—Mantell’s type in the
British Museum, Mr. G. B. Holmes’s, figured by Prof. Owen in the
Fossil Reptilia of the Wealden Formation, and Dr. Wilkins’s two,
exhibited this evening-—by the scapular hypothesis originally pro-
pounded by Dr. Mantell, and adopted by Prof. Owen and the Rey.

J. W. HULKE ON A MODIFIED FORM OF DINOSAURIAN ILIUM. 025

W. Fox, the large, smooth articular arc (a) at the lower end of the
body must be, in its entirety, the scapular share of the glenoid fossa,
because it alone, as Dr. Wilkins’s smaller specimen demonstrates,
affords evidence of having formed part of a freely movable joint,
and it bears no mark of a subdivision into a glenoid and a coracoid
portion. The ‘“ appearance of a subdivision of the thickened arti-
cular end into two surfaces, one for the humerus, the other for the
coracoid,” which Prof. Owen observed in Mr. Holmes’s specimen*,
if descriptive of this articular arc, was perhaps due to a post-mortem
distortion ; but there is the possibility that Prof. Owen considered
the whole of this are as the coracoid division, and the stout buttress
which posteriorly limits it, as that against which the humerus
rested.

Assuming, then, as proved, that the large, arched articular sur-
face in its entirety represents (on the scapular hypothesis) the sca-
pular part of the glenoid fossa for the humerus, its entirety and
diarthrodial nature being placed beyond doubt by Dr. Wilkins’s spe-
cimens, the coracoid attachment must be sought in front of it. Dr.
Mantell (unless I misunderstand him) supposed that one or both of
the small projections from the under surface of the long slender
process gave attachment to the coracoid bone; but these, even if
complemented by the long slender process, appear very inadequate
for the firm attachment of a bone which contributes about an equal
share with the scapula to the shoulder-joint, and has with it to bear
very considerable pressures ; besides, in Jguanodon Mantelli, Hyp-
silophodon Fox, Megalosaurus Bucklundi, Hylceosaurus Conybeari,
and Scelidosaurus Harrison (the only British Dinosaurs of which
scapulee in union with coracoids have, so far as my knowledge
reaches, as yet been found) the coracoid border of the scapula is
so much longer, relatively to the humeral border, than these two
little projections are to the arch behind them, that it is hardly pos-
sible to conceive that they can be the functional equivalents of the
coracoid border of the scapula as we know it in the Dinosaurs just
mentioned. But conceding that they are corresponding parts, what
explanation can be given of the stout posterior inferior angle of the
body which limits the smooth articular arc behind? Its truncated
terminal surface, directed downwards and backwards, bears, in Dr.
Wilkins’s smaller specimen, the stamp of a firm cartilaginous union
with some other bone. But what bone is there united with the
scapula behind the glenoid fossa ?

There remains only the suggested resemblance of the long, slender,
anterior process detached from the anterior margin of the body
near its lower end, to certain processes found in Hylaosaurus, and
in some existing Lacertians, notably in Scineus.

The part of the Hyleosaurian scapula to which Prof. Owen com-
pares this long slender process is described by him in these terms :—
‘On the outer side of the scapula two broad convex ridges descend
and converge to form the beginning of a thick and strong spine at
fourteen inches distance from the base; this then expands into a

* Brit. Foss. Rept., Megalosaurus, p. 14.

526 J. W. HULKE ON A MODIFIED FORM OF DINOSAURIAN ILIUM.

thick acromial ridge, which extends transversely, and is continued
forwards as a long subprismatic process from the anterior angle of
the head of the scapula. This process, which appears likewise to
be present in the scapula of Jguwanodon, perhaps also in Megalo-
saurus, is broken off in the present specimen about four inches from
the neck of the scapula, with which it forms a right angle”*.

It is not stated from what specimen this description is taken.
Only four are, I believe, yet known; and these are all in the British
Museum. Two of them are the immature pair in the Tilgate-Forest
slab, which has been so often figured ; and in these the crest cross-
ing the outer surface above the humeral and coracoid borders is but
slightly developed, and it ends in front very slightly in advance of
that part of the coracoid border which is still joined to the coracoid
bone. I fail to perceive here any indication of a “ long subprismatic
process ” continued forwards from “the thick (transverse) acromial
ridge ;’’ nor do I find any thing answering to this description in the
two other scapule, which, from their much larger size and massive-
ness, belonged to a much older and probably mature individual. Of
one only the articular end, very mutilated, is preserved; the other
(No. 584, Mantellian Catalogue; No. 2584, Brit. Mus. MS. Catal.)
is a magnificent specimen, almost complete. It wants only the an-
terior inferior angle, formed by the junction of the anterior and the
lower or coracoid border. The fracture which has severed this
missing piece passes vertically, in the same plane, through the cora-
coid border and the anterior termination of the transverse crest,
where it merges into this just in front of the wide, and here shallow,
groove which separates the crest and this articular border. This
scapula also yields no evidence that its crest was produced forwards
in the shape of a slender process beyond the coracoid border. In
the absence of any evidence of the existence of a long slender an-
teriorly projecting process, the resemblance of the “scapula of an
unknown reptile” to that of Hylwosaurus in respect of such a process
seems to me to be wanting in proof, and untenable.

Upon a close comparison of the long slender process with the
processes detached from the anterior border of the scapula in certain
extant lizards, such differences discover themselves as seem to me
to disprove their identity. There is no skeleton of Scincus in the
British Museum, or in that of the Royal College of Surgeons; and
my knowledge of its scapula is derived from the figure in Cuvier’s
‘Ossemens Fossiles.’ In this figure the process, passing off from
the anterior border of the neck of the scapula, is represented as
bending upwards and forwards nearly parallel with this border, a
direction quite different from that of the process in any of the spe-
cimens of the unknown reptile’s scapula, and one which has an
obvious relation to the process detached from the front of the cora-
coid bone, to the end of which it is linked by a cartilaginous loop.
The shape of the scapular process in Scincus is, I believe, also dif-

* Report on British Fossil Reptiles in Brit. Assoc. Report, 1841, p. 117; and

Brit. Foss. Rept. of the Wealden Formations—Monograph Hyleosaurus, where
this extract is reproduced.

J. W. HULKE ON A MODIFIED FORM OF DINOSAURIAN ILIUM. 527

ferent. As in other lizards, it and the coracoid processes are com-
pressed flat blades, the thin edges of which are adapted to the
attachment of the membrane which fills the unossified spaces between
them; whereas the long, slender anterior process of the unknown
reptile’s scapula is subcylindrical, and, where compressed, this is in
the opposite direction to that of the scapular processes of living
lizards. Here, too, I suggest, the comparison fails.

The difficulties of the scapular hypothesis appear to me insur-
mountable. On the other hand, upon a comparison of these reputed
scapule with a typical Dinosaurian ilium, e.g. of Zqguanodon Man-
telli, the essential correspondence of all their several parts is so
evident as to render irresistible the conviction that they are identical
members of the skeleton: @ is the acetabulum, p its pubic pro-
cess, ? its ischial process; Sp the long, upper, anterior, slender
process, is the suprapubic process so largely developed in Jguanodon
Mantelli; and 6 is the expanded part or body of the ilium; the
notch (7) corresponds to that beneath the small anterior process in
existing lizards, beneath which the nerves and vessels wind which
are destined for the front and upper surfaces of the hip-joint and
thigh. The most obvious differences of these ilia from that of Jgua-
nodon are the angle which the axis of their body makes with that
of the long, slender process, and the absence of any distinct stamp
of a sacral attachment.

In the three smaller ilia which have been mentioned in this note,
viz. Dr. Mantell’s, Mr. G. B. Holmes’s, and Dr. Wilkins’s smaller
specimen, the long, slender process makes nearly a right angle with
the axis of the body; in Dr. Wilkins’s larger specimen this angle is
more obtuse, whereas in Jyuwanodon the body and the long, slender
suprapubic process are most extended in nearly the same line. As
the ilium forms the upper part of the acetabulum, the axis of its
body, which is nearly vertical to the chord of the acetabular arc,
is highly inclined or nearly vertical to the vertebral column, and
the iium would therefore necessarily be connected with fewer sacral
vertebree than in typical Dinosaurs—dispositions resembling those
which obtain in existing lizards, where, as is well known, the
number of sacral vertebre is two, and the slant of the ilium is con-
siderable, notably in Hatteria, in which it includes, with the ver-
tebral column, an angle of about 80° (Dr. Giinther), and in Cha-
meleon, in which the ium depends nearly vertically from the sacrum.
These, it is worth notice, exhibit some closer skeletal resemblances
to Dinosauria than do other existing lizards,

As regards the absence of distinct marks of attachment to the
sacrum from both the lateral surfaces of the body, I would speak
cautiously. Only one surface of Dr. Mantell’s and Mr. Holmes’s
specimens is exposed; and Mr. Beccles writes me that one surface
ot his specimen of this bone is backed with plaster of Paris. In
neither of Dr. Wilkins’s bones do I find these marks; but in each
the upper end of the body is missing, and they may have been pre-
sent here, where we find them in existing lizards.

The Rey. W. Fox, to whom I turned for some information on this

528 J. W. HULKE ON A MODIFIED FORM OF DINOSAURIAN ILIUM.

point, (singularly enough, on the day my letter reached him) found a
very perfect example of the bone; and he most obligingly carefully
removed the matrix from both surfaces to look for a sacral impres-
sion, but did not find any. Its absence is, I think, his chief reason
for regarding the bone as a scapula. Should the absence of any in-
dication of a firm cartilaginous union with the coalesced sacral
transverse processes be confirmed, the absence will not disprove the
iliac nature of the bone; for the ilium may have been simply at-
tached by strong fibrous bands passing igamentously between it and
the vertebral column, as in the Enaliosauria. Whichever mode of
attachment obtained, its extent was manifestly much less than in
typical Dinosaurians ; and this is a modification of sufficient impor-
tance to place the possessors of such ilia outside all the known
genera of Dinosauria, and, perhaps, outside this order.

In conclusion I would tender my warm thanks to Mr. G. B.
Holmes, the Rev. W. Fox, and Mr. Beccles for kindly affording me
information respecting the specimens of this bone in their col-
lections, and to Dr. Wilkins especially, for the interest he has shown
and the aid he has afforded me in this inquiry. I feel, also, that
an apology is due to the Society for trespassing upon its time whilst
some difficulties to the acceptance of the interpretation I offer re-
main unsolved; my excuse is my hope that some of our Fellows
may, during the recess, recognize specimens of these ilia which may
be lying unknown in private collections, and seize the opportunity
to confirm or medify my conclusions.

EXPLANATION OF PLATE XXXII.

Modified Dinosaurian Llia in the Collection of Dr. Wilkins, Newport.

Figs. 1 & 2. Smaller specimen. 1. Outer surface, ? nat. size: a. articular are ;
p. pubic process; 7. ischial process; Spp. suprapreepubic process ;
vr. ridge directed backwards from Spp, concentric with a; . body ;
n. notch. 2. Direct view of articular arc, 3 nat. size (fig. 1, a).

3 & 4. Larger specimen, ? nat. size. 3. The outer surface. 4. The inner
surface. Letters a, p, 7, Spp, 6, n, represent the same parts as in
figs. 1 & 2.

Va

GHV¥ord &C.L Griesbach.

x

XX. Pl

Mintern Bros imp.

uart. Journ. Geol. Soe Vol.

Q

ILIUM.

f=

: Quart. Journ. Geol. Soc. Vol. XXX. P] XXXII.

GH Ford. & CL Griesbach. Mintern Bros imp.

DINOSAURIAN ILIUM. :

oS as
ats

ni alate A i.

ae hae p
? By bs Se: es

: rae BO =e wae Et a a - es st zs ee 2 aa eBoy e Ee Bega Be

All

SMITHSONIAN INSTITUTION LIBRARIES

MN