)
i

ree): ate’
ia } wit

4
A

i
4

‘\

wali) lh Bh

shiv’

Ary

Shy Pee
ap Leiter e) awash

ark ei) 34

it oh

ah ey
” ine Ips

prOR AMC Uetr Catgeyoe to
Pith bist mtr ne

ate ka MW Webesthles

ae

* Wer
’ .

reey
Ai yb
a Ase bhey b+ bebe

alps
rh)
Wirt AL
Epil ide!
ith

beloe
, Reels

Tr

4" Feb ac) De hthe

vy

AP ene ee Oe

iis creare rey f ih 4 Ooi ay!

| ded ei) Lage o: | ae Woe

ip he Matt a i, )
an hay? vt) > shat +

ee pata) |

aint

cd 4 bedett
ay Oat ble

Wy
Ke wel bl
if

haa depeiie Ga)
falta win) i
ee A

f Visca
iy my nh!

{
j

1h pee wy
eheene

Aielhee Lota,
” bap Ang as

lot astih 4i)

UP vavee DERE DF

ti ba"
set betd
.) 4

‘Wedel hp ”
yy
aa

ee / "
b Lat Ht tpt

a4 :
Li de Ap reere be

CooGE rapt Ree
’ v Deo veeeone OTe we
hae ele Sere

fants hebet
peor tiytr gel be

« ae
HH, MOT Gong

ss eee

la

ae
-
4 raphotayss

yee
" fess ror ho
Aa > Vr ps Ree a,
hadi hep ree etapa
oon oh 4 Meee
¢

ee tats ek tp
Ath Wished oie yy!
tye

folate:

i? <
alt tet Hite:

heat
nN Hels ty ehh
ott an (rere vag cyt tt led ras §
yn MBG Avibb pei rte) 7

vel be he tod oe,
pire statee

felt eel
ida)

Hoa, yo
,,| RH he
ahs si Ot hen

‘ ait |
ya bi wi
ie birt RAMEETA)

Vidi? WY tad oho
i a

reir
velses Py
hit} tet Hi

; sal
wiv LP) dred . e ake
Peale had os ya"
THs

‘
[ ithe betes a9
+ it Oe
Thebes

2) bo

sibbte

. + ehh!
Piste baad My hil by
Paseo hata begh gt
Orit 4 ie Ma apres
i wily Vong

H bay fee's

ae By

oy)
} Birt

ci rir yaa
ogy! ae ola! ; + wate
Shodiy t {it eee fits i4ya) bey
4 . a i CAPs | rt tat Fiehataees
Thitaes

‘ +4
LA risus

ALG td 4 iy vslaiee

} Lee Lest it reste it
av 7 + ine? ¥ bes vy fs
fr OF MTP rne tare IAD ee eat 7
Fy ' haha Fises none } at tet eee K
u t ( i v — nm
' Pas ay aheg pa ch ctobel ARs

wey pie
qo dedvat beoved eh
Ot pe hag hee ie pee

ray RUarreL

Rey Math
aan

y : railing
kita ed
Ut

iat
if fest AT, teint \

be od

1 ey %)
weber ig!

fivt a? eeF
ne Liat
ie LUe,
Unsere am eA

rie
"
tht a }

Labatt! oY haute
re ee? er Any,
wi He Ce

oti byh ie ein

peal

, : rar a
Vile 1 ee pebh

epee

i! ahi

ba dak 4h mone tee
= 8 yee? 1 hye esas
Cdl dona Lh

i¢
ie ont Die

a

. oro Bel styl
-

ee

rey t ae
peta ibe

CAN
ais a HA9Gnqg
- THE —
} 2,
QUARTERLY JOURNAL
OF THE
GEOLOGICAL SOCIETY OF LONDON.
EDITED BY

THE ASSISTANT-SECRETARY OF THE GEOLOGICAL SOCIETY |

3

et

Quod si cui mortalium cordi et cure sit non tantum inventis herere, atque iis uti, sed ad ulteriora
penetrare ; atque non disputando adversarium, sed opere naturam vincere; denique non belle et probabiliter
opinari, sed certo et ostensive scire; tales, tanquam veri scientiarum filii, nobis (si videbitur) se adjungant.
—Novum Organum, Prefatio.

VOLUME THE THIRTY

1876.

LONDON : <

LONGMANS, GREEN, READER, AND DYER.

PARIS: FRIED. KLINCKSIECK, 11 RUE DE LILLE; F. SAVY, 24 RUE HAUTEFEUILLE.
LEIPZIG: T. 0. WEIGEL.

SOLD ALSO AT THE APARTMENTS OF THE SOCIETY.

MDCCCLXXVI.

Ltst
OF THE

OFFICERS

OF THE

GEOLOGICAL SOCIETY OF LONDON.

PresiVent,
Prof. P. Martin Duncan, M.B., F.R.S.
Wice-Brestents.
Sir P. de M. G. Egerton, Bart., M.P., F.R.S. | J. W. Hulke, Esq., F.R.S.
R. A. C. Godwin-Austen, Esq., F.R.S. Prof. A. C. Ramsay, LL.D., F.R.S.
Secretarics.
David Forbes, Esq., F.R.S. - | Rev. T. Wiltshire, M.A.

Foreiqu Secretary,
Warington W. Smyth, Esq., M.A., F.R.S.

Creasurer,
J. Gwyn Jeffreys, Esq., LL.D., F.R.S .

COUNCIL.
H. Bauerman, Esq. J. W. Hulke, Esq., F.R.S.
Rev. T. G. Bonney, M.A. J. Gwyn Jeffreys, Esq., LL.D., F.R.S.
W. Carruthers, Esq., F.R.S. Prof. T. Rupert Jones, F.R.S.
Frederic Drew, Esq. J. W. Judd, Esq.
Prof. P. Martin Duncan, M.B., F.R.S. Prof. J. Morris.
Sir P. de M. Grey Egerton, Bart., M.P., | Prof. A. C. Ramsay, LL.D., F.R.S.
F.R.S. Samuel Sharp, Esq., F.S.A.
Robert Etheridge, Esq., F.R.S. Warington W. Smyth, Esq., M.A., F.R.S.
John Evans, Esq., F.R.S., F.S.A. Admiral T. A. B. Spratt, C.B., F.R.S.
David Forbes, Esq., F.R.S. W. Whitaker, Esq., B.A.
R. A. C. Godwin-Austen, Esq., F.R.S. Rey. T. Wiltshire, M.A., F.L.S.

Henry Hicks, Esq. Henry Woodward, Esq., F.R.S.

Assistant-Secretary, Librarian, and Curator.
W. S. Dallas, Esq., F.L.S.

Clerk: Library Assistant.
Mr. W. W. Leighton. Mr. W. Rupert Jones,

Museum Assistant,
Mr. B. B. Woodward.

TABLE OF CONTENTS.

Page

AITKEN JoHN, Esq. Observations on the Unequal Distribution of ;
Drift on opposite sides of the Pennine Chain, in the Country
about the source of the river Calder, with suggestions as to the
causes which led to that result, together with some Notices on
the High-level Drift in the upper part of the Valley of the river

ENE Gh tel OSS? SRR del PR ra er MRL Og 184
ALLporT, §., Esq. On the Metamorphic Rocks surrounding the
Land’s-End Mass of Granite. (Plate XXIIL) .............. 407

Brit, Tuomas, Esq. The Drift of Devon and Cornwall, its Origin,
Correlation with that of the South-east of England, and Place in
ART CCUG SONTES ya idWils ss 4 8 a aks heh olla hal as, stethineneiena pels aur 80

Bettany, G. T., Esq. On the Genus Merycocherus (Family Oreo-
dontide), with Descriptions of two new Species. (Plates XVII.

UTE GUT iy Yo sc a's: cca winy s, Sha, « ARE, tie Cae ah SS. hers 259
Bonney, Rey. T. G. OnColumnar, Fissile, and Spheroidal Struc-
PTT MEMOE OR Send ieiann Ca SGN note ai ss Ainge, bias IFT vice 5 GW wees ctw eRe w ae 140

CioveH, C.T., Esq. The Section at the High Force, Teesdale .... 466

Davis, J. W., Esq. On a Bone-bed in the Lower Coal-measures,
with an Enumeration of the Fish-remains of which it is prin-

Gipalhy COMPOSED !..-.2 Fie Ge Sw vc liale ap eee whit Ye ertts Ning hg oye 332
Dawkins, W. Boyp, Esq. On the Mammalia and Traces of Man

found authe fobm-Eood. Cave. ;...cae..: ans. cs he ee 3 Se rhe Oa
Dawson, Dr. J. W. Notes on the Occurrence of Eozoon canadense

aie CECE EM eULCuty (CENeuGeN, )ocety iS aoe aes soe capes feo eeaatal 66
Note onthe Phosphates of the Laurentian and Cambrian

ROCKS Ole OC AMGGAe! en Gee ok eet bs dain whimiee oR ORO wamate 285

Duncan, Prof. P. M. On some Unicellular Alge parasitic within
Silurian and Tertiary Corals, with a notice of their presence in

Calceola sandalina and other Fossils. (Plate XVI.) .......... 205
On some Fossil Reef-building Corals from the Tertiary
Deposits of Tasmania. CR late: PORTS eeememthed, Cet cea aS 341 -

Iv TABLE OF CONTENTS.

Page
ETHERIDGE, RoBERT, Jun., Esq. On the Occurrence of the Genus :
Astrocrinites (Austin) in the Scotch Carboniferous Limestone
Series, with the Description of a New Species (4. Benniev), and
Remarks on the Genus. (Plates XII. and XIII.)............ 103

On an adherent Form of Productus and a small Spiriferima
from the Lower Carboniferous Limestone Group of the East of
Neotland. (Plates XXTY. and XQ) yee eee 454

Gunn, JoHNn, Esq. On the Presence of the Forest-bed Series at
Kessingland and Pakefield, in Suffolk, and its position beneath

the Chillesford Olay eon ake da as be ee ee ee ee 123
Guppy, R. J., Esq. On the Miocene Fossils of Haiti. (Plates

AX VEE and ORS) ieee Siete is eke Bs le Glue owes he 516
Harrison, Wo. J., Esq. On the Occurrence of the Rheetic Beds in

AUGIECSEORSIEO: 5 5's wine diss wea sid a bee eo ese wae se 300 212
Hector, Dr. Note to Mr. H. Woodward’s Paper on a new Fossil

Crab from the Tertiary of New Zealand 23). .0.5 2s 5. 00 ee
Hicks, Henry. Appendix to Mr. J. E. Marr’s Paper on Fossili-

ferous Cambrian Shales near Caernarvon.......-.sssseeeee. 135
Hvuxxe, J. W., Esq. Appendix to “ Note on a Modified form of

Dinosaurian Iium hitherto reputed /Scapula ”) .....< .:.¢% ana 364
Irvine, Rev. A. On some recent Sections near Nottingham...... 5138

Jupp, J. W., Esq. On the Ancient Volcano of the District of Schem-
nitz, Hungar y. (Plate AX) or ciiien wes cs eek ee ae 292

KereEpinc, WALTER, Esq. Notes on the Paleozoic Kchini. (Plate
LU NS eer NAL Ve Ba Ney wR a SR

On the Discovery of Melonites in Britain .............. 395
KENDALL, J. D., Esq. Heomatite in the Silurians ...........3%% 180

Lavis, H. J. Jounston, Esq. On the Triassic Strata which are
exposed in the Cliffsections near Sidmouth, and a Note on the
occurrence of an Ossiferous Zone containing Bones of a Laby-
PUTOMON Oe Ee ol dao ol ddl vP ID oa ee ew Rls BRT 274

Luoyp, T.G. B., Esq. Geological Notes from the State of New York 76

MAckKINTOSH, D., Esq. On the Correlation of-the Deposits in Cefn
and Pont-Newydd Caves with the Drifts of the North-west of
Pnoland el Wy ales IG ape ips os cs ss Rvs ee. ss 91

———. On the Mode of Occurrence and Derivation of Beds of
Drifted Coal near Corwen, North Wales...............5014< 451

Ma.uet, R., Esq. On the Mechanism of Production of Volcanic
Dykes, and on those of Monte Somma................00000- 472

Marr, J. E., Esq. Fossiliferous Cambrian Shales near Caernarvon.
With an Appendix by Hunuy Miers icq! 792.2: oe 134

TABLE OF CONTENTS. v
Page

MELLO, Rey. J. Macrens. On the Bone-caves of Creswell Crags .. 240
Mureuy, J. J., Esq. The Glacial Climate and the Polar Ice-cap .. 400

Newton, E. Tutiry, Esq. On two Chimeroid Jaws from the
Lower Greensand of New Zealand. (Plate XXI.) .......... 326

OwEN, Prof. On a new Modification of Dinosaurian Vertebree.
Meee: Ge Wi) oat oe osu! ois Sa al cnt ae a eto a,

Evidence of a Carnivorous Reptile ( Cynodraco major, Ow.)
about the size of a Lion, with Remarks thereon. (Plate XI.) 95

. Evidence of Theriodonts in Permian Deposits elsewhere than

PEN SOULE ANTEC: WO ot nors avccauslg Bg ay ences he ceen Sao ee Meoeen 352
PrEnninG, W. H., Esq. Notes on the Physical Geology of East An-

glia during the Glacial Period. (Plate XV.) ............400.. 191
Puiuurps, J. A. Esq. On the so-called “ Greenstones”’ of Western

Monin Mann sds oe teg aa w adhe catabudides cad Saaniuaaes 155
Ramsay, Prof. A. C. How Anglesey became an Island. (Plate

RTE tone sagas oa nN era cially ee SPRINT TN te Dark takaoatr 116

Note to Mr. Ramsay’s Paper on the Influence of Various
Substances in Accelerating the Precipitation of Clay suspended
PEMVOUCE He Ceee ese Sage ce oe Lettuce do agreed he Marae nee Seika 132
On the Physical History of the Dee, Wales. (Plate XIV.) 219

Ramsay, Witi1aM, Esq. On the Influence of various Substances in
Accelerating the Precipitation of Clay suspended in Water. With
Ne rea it E ZO ks Pua sayy leno. See ae bana’ a ae alee ele ae Malate ls 129

SEELEY, Prof. H. Govimr. On the Posterior Portion of a Lower
Jaw of Labyrinthodon (L. Lavisi) from the Trias of Sidmouth.
ROE rnicee MOO NC Deena gia sara hirekairaliusia Peers Se a cm OF OS wee mates 278

Notice of the Occurrence of Remains of a British fossil
Zeuglodon (Z. Wanklyni, Seeley) in the Barton Clay of the
PEeeeamen ANT ONO AGE ces. ears its cle’ aay) sista ceca paraye pcam enuanatngs 428

On an associated Series of Cervical and Dorsal Vertebree of
Polyptychodon, from the Cambridge Upper Greensand, in the
Woodwardian Museum of the University of Cambridge ...... 433

. On Crocodilus icenicus (Seeley), a second and larger Species
of Crocodile from tha Cambridge Upper Greensand, contained in
the Woodwardian Museum of the University of Cambridge ., 437

. On Macrurosaurus semnus (Seeley), a Long-tailed Animal
with Proccelous Vertebree from the Cambridge Upper Greensand,
preserved in the Woodwardian Museum of the University of |
SAMIR Seer yee Wider Shia. =! 00 aca a:b barnk aio apache RNR 440

. On Remains of Emys hordwellensis (Seeley) from the Lower
Hordwell Beds in the Hordwell Cliff, contained in the Wood-
wardian Museum of the University of Cambridge ............ 445

. On the British Fossil Cretaceous Birds. (Plates XX VI. &
PS OTE RR) Se Go a 496

vi TABLE OF CONTENTS.

Page
Ussumr, W.A.E., Esq. On the Triassic Rocks of Somerset and ;
| UE 0) 0 eg ee MP IER coe Rs open) CME Bg DE th 367

Warp, J. Cxiirron, Esq. On the Granitic, Granitoid and Associated
Metamorphic Rocks of the Lake-district. (Plates I.& I.).... 1

Wusovy, E., Esq. On the Permians of the North-cast of England (at
their Southern margin), and their Relations to the Under- and
Over-lying Formations... ...:... <.oca ee eRee OR bee 535

Woopwarb, Henry, Esq. On some New Macrurous Crustacea from
the Kimmeridge Clay of the Sub-Wealden Boring, Pea and
from. Boulogne-sur-Mer. (Plate VI.) 10. ons. 6 oc acu cee 47

. Ona new fossil Crab from the Tertiary of New Zealand,
collected by Dr. Hector, F.R.S., F.G.S., Director of the Geological
Survey of New Zealand. With a Note by Dr. Hector. (Plate

IVT eit nase ite rataleseeliate eas SEP cba bene cede Oa ay Gila Cae, Ree sisi. ogi
On the Discovery of a Fossil Scorpion in the British Coal-
measures. (Plate, VU.) voc ievsiast eer n eens ve 0s oe 57
——.. Ona remarkable fossil Orthopterous Insect from the Coal-
measures of Scotland. (Plate TX.) oo.) oes wae ses, t ec ee 60

Woopwarb, Horace B., Esq. Notes on the Gravels, Sands, and
other Superficial Deposits i in the Neighbourhood of Newton Ab-

bot, Devonshire -). i acces shes saves codes Coen oe eee 230
Wort, R.N., Esq. On certain Alluvial Deposits associated with

the Plymouth Limestone! ie20! saat. 20h). 555 2b ee eee 236
PROVARIEL ROOM bis aig ied @ ew bb aes Melee eo eames ua he ee ee 15
Enstial Poraiem Niembers iejes 060s ls he oc ele onda sy 35
List of Foreien Correspondents : 32... 6.02 20005 SS, Ge 36
Tast.o: Wollaston Medallists 5.20.00. 00 220. 22 PG 37
Mist of Murchison Medallists «iii. vais crawan dow aie see 39
hash or tayell WMiedalliste’s\.0/04.0 Ps. Caines oben went Oe Renae 39
Pinay erat y wR Mresy to 4:siq70y jie Mo iwihye Wm ey dpa ee dg Hk on) oe Bg

Donations to the Library (with Bibliography) ............0-000, 133

LIST OF THE FOSSILS FIGURED AND DESCRIBED
IN THIS VOLUME. }

[In this list, those fossils the names of which are printed in Roman type have
been previously described. ]

Name of Species. Formation. Locality. Page.
PLANTS.
Paleachiya perforans. Pl. xv....... Feat pe ale le eAmetack ee | 210
AMORPHOZOA.

Archeospherine. Pl. x. f. 1-3...) | oes | 71
Hee oe dences: Pl. x: £ 8-10. Laurentian chishs 0 ee di RBaa | 66
C@LENTERATA.

Heliastrea tasmaniensis. P\. xxii. f. \ | (
MMR end Serie esata views wan og unclean | : : 342
Thamnastrea sera. PI. xxii. f. 4-6. (SESE oo a 343
SB arcana ga nateed vada cai ace {| 344
MoLuivsca.
(Brachiopoda.)
Productus complectens. Pl. xxiv.,
MIME nca hua ioe eee sidivice worries |\Carboniferous ...|Scotland 454
( Gasteropoda.)
Cancellaria epistomifera. P.xxviii. \ | f
Fe os se os oR ee ee | 520
Cerithium uniseriale. Pl. xxix. f. 4. 519
obesum. Pi xxix, f. 9.2 .25.:. 519
Clavatula labiata. Pl. xxviii.f. 3... } Miocene ......... EVGA, issn so oecass 4 | 527
Conus marginatus. PI. xxix. f. 5... 528

528
524

Cyprea Gabbiana. P1.xxix. f.10...

Glea trancata. Pl. xxix. f. 6......:.- | | 524
Fusus Henekeni. Pl. xxviii. f. 2, 6..) | (

Vili FOSSILS FIGURED AND DESCRIBED,

Motuuvusca (continued).

Name of Species. | Formation. Locality.

Marginella Sowerbyi. PI. xxviii. f. 1 \ |
Murex cornurectus. Pl. xxviii. f. 4
textiis: -Pl xxix. fit 22
Orthaulax inornata. PI. xxviii. f. 8
Phorus delectus. Pl. xxviii. f.10... |
Pleurotoma consors. Pl. xxxviil.

|

}|Miocene ......... Elaita s2.csene

squamosa. Pl. xxix. £7 ....<-
Sigaretus excentricus. Pl. xxix. f. 11
Terebra sulcifera. Pl. xxix. f. 8 ...
Triton domingensis. Pl. xxix. f. 2
Turbinellus edificatus. Pl. xxviii. | |

Behe ee oeeeme tatese nee Sethis ea ton side

—— haitensis. Pl. xxix. f.3 ......

EcHINODERMATA.

Archeocidaris Urii. Pl. iii. f. 14-18...|Carboniferous ....North Wales
Astrocrinites ? Benniei. PI. xii., xili....|Carboniferous .../Scotland ......

Melonites Etheridgti ........0...sseeeeee- Carboniferous ...|Derbyshire .........
Palechinus (?) intermedius. Pl. iii.
RUN Coy nog i os Fn es nike ve mane Carboniferous ...| Wexford ............
Pies, EL il, fs. U2, Ld) ..<-venanena Carboniferous ...|Ireland............00.
Perischodomus biserialis. Pl. iii. f.
ee Fn ce 2 a Ventanas Carboniferous ...\Ireland....... oateeee
Rhoechinus irregularis. Pl. iii. f.
TE Se SO te ame he Y Carboniferous ...|Wexford ............
ANNULOSA,
( Crustacea.)
Callianassa isochela, Pl. vi. f. 1,2. Kimmeridge ...|/Sussex ............

Pd

seeeee

Harpactocarcinus tumidus. Pl. vii. |Tertiary .........,New Zealand ...

‘Sussex and

: ; ; ie Sie)
Mecochirus Peytoni, Pl. vi. f. 3, 4...... Kimmeridge | |Boulogne......
( Arachnida.)
Euscorpius anglicus. Pl. viii. ............ |Carboniferous ...|Scotland ......
(Insecta.)
Gryllacris Erongniarti. Pl. ix. f. 2...... Carboniferous .../Scotland ......
Libhomantis carbonarits, Pl. ix f. 1....\Carboniferous ".|Coalbrookdale

nee

eeostee

oeseen

ewates

——

48
48

52

58

60
62

FOSSILS FIGURED AND DESCRIBED.

1x

Name of Species. Formation. Locality. Page
VERTEBRATA.
(Pisces.)
Callorhynchus Hectori. Pl. xxi. f. 6-9\|Lower Greensand|New Zealand ...... | 329
BLEMACARGNUS, SP. .cecccsssceewereseeeens Carboniferous ... |Yorkshire ea bee)
eR Co ois Sa Un duke adisisisns cgarawacaee Carboniferous... |Yorkshire ......... 339
ORO a ee Carboniferous .../Yorkshire ......... | 339
Ischyodus brevirostris. PI. xxi. f. 1-5|Lower Greensand|New Zealand ...... | 327
(Batrachia.)
Labyrinthodon Lavisi. P|, xix............|Triassic .........|Devonshire ......... 278
( Reptilia.)
Cynodraco major. PI. Xi. ..........5. \ 95
Pareiasaurus bombidens. Pl. v. f. ! :
BR SI ow ankle tua st Uh cng ds vaca ee anys . 44
[3 ea eer al f Priassie F028, South Africa ...< 45
Tapinocephalus Atherstonii. Pl. iv. |
| ag ant Bele atin wstoeneeeekkiats abt \ 43
( Aves.)

Enaliornis Barretti. Pl. xxvi. f. 1-)
11, 14-27, and Pl. xxvii. f. 1-5,

|
#1 ae 19-25 eee tere ecreeerecereeereseeseas \ Upper Greensand Cambridge eae

Sedgwicki. Pl. xxvi. f. 12, 13, |
and PI. xxvii. f. 6, 7,9-11, and 13 |

SRR rie Mins aaah Sue OeMaah Oe J}
( Mammalia.)
Merycocherus Leidyi. Pl. xviii. ... ) |y,-
Ee emarae Pi. SEV ca coeds \ Miocene UU ABR [Oregon BRA. of

A®

La
|

506

| 270
269

EXPLANATION OF THE PLATES.

PLATE PacE
Sections or Granitorp Rocks of the Lake District, to illus-
trate Mr. J. Clifton Ward’s paper on these rocks ............ 34

I
if
EHVOBS AROSSIES cre... Sepcinavisawepabemr esha acousidwen Gece tee 42

VERTEBR& OF TAPINOCEPHALUS AND OF PAREIASAURUS BOMBI-
pENs, to illustrate Prof. Owen’s paper on some Dino-

ll ae Ecuarnt, to illustrate Mr. W. Keeping’s paper on
; | BAUMAN. VertOOrss ...csionciy sden «gas qocewss ewan bans aoe 46

VI New Kiwmeripee-Ciay Crustacea, to illustrate Mr. H.
Woodward's paper on those fossils..............:esssecseusecaue 50

VII { HaARPACTOCARCINUS TuMIDUS, to illustrate Mr. Woodward’s
"| ‘paper on that species“ ....cc,.:c005.0.<0essasanana se aareee eee 56

VII Britisu Fosstzt Scorpions, to illustrate Mr. H. Woodward's
; paper on Those LOssS “veer. sv <eesenne suns ongvnite coud) cee 58

1x LITHOMANTIS CARBONARIUS, to illustrate Mr. H. Woodward's
paper onithat Tnspet A job. cocci icagends ones caeieaiss cette eae 64

- { Hozoon AND ARCHAOSPHARINA, to illustrate Mr. J. W. Daw-
son’s paper on the occurrence of Hozoon at Cote St. Pierre. 74

XI Cynopraco magor, to illustrate Prof. Owen’s paper on that

REDE ssapenienitgenae cores gated nasnneants.dietnecuine sadhana 102
XII. ) Astrocrinites? Benniet, to illustrate Mr. R. Etheridge Jun.’s
XIII paper on that Fossil ...co ane ied secu auuees dceess rset ses Oe Enea mmm

GroLoaicAL Map anv Section of Anglesey and the adjacent
SY. part of Wales, to illustrate Prof. Ramsay’s papers on the
Island of Anglesey and the Valley of the Dee ............... 122

W. H. Penning’s paper on the Physical Geology of East

Map anp Secrion of a part of East Anglia, to illustrate Mr.
XV
Anglia durine the Glacial Period ..........0s8-+.1.>sesnasasmere 204

XVI PALHACHYLA PERFORANS, to illustrate Prof. Duncan’s paper
Br aa OF thiah PAG ee aiaire ada st aneaas oc -seevnasanasesanans seas. Aoanee 210

XVII MERYCOCHG@RUS TEMPORALIS, to illustrate Mr. Bettany’s paper
: on. that fossil meccdecc vcs dnncnevnsenccedonierosan scence teen een 272

: Merycocuarus Lerpy1, to illustrate Mr. Bettany’s paper on
BT that fossil <scsessse ssiiteanneeiann ines ane ae eee 272

EXPLANATION OF PLATES. xi

PLATE Pace
xx, | LAsyemruovon Lavisi, to illustrate Prof. Seeley’s paper on

: 1 that fossil ........ Sahin uc nate eaigerie Me EOE ot aa st calves osctucmabn Ge 284

Map of the District of Schemnitz in Hungary, with Section,
XX. to illustrate Mr. Judd’s paper on the Ancient Volcano of
that District ......... sh os Jor ARE ae tee ern anin SS ktcisecesaues 324

Cretaczous Cuimmrorps, to illustrate Mr. Newton’s paper
XXiI. 2
OH, CHOKS FOSSTIS: ciescaias aos yada ee eteeene Mice aoe atboae once s 330

TAsMANIAN Corazs, to illustrate Prof. Duncan’s paper on
XU. { EGS TORSUAL ”L),.5.0sshe is tadeus ies GEER ERO Meee ee poe ee 350

xxi, { Rock Sections, to illustrate Mr. Allport’s paper on the Meta-
‘ morphic rocks surrounding the Land’s-End mass of Granite 426

XXIV.) ApHERENT Propucti AND SPIRIFERINA, to illustrate Mr. R.
XXV. Etheridge Jun.’s paper on those fossils ................000000 08 464

XXVI. ) Remains oF Britisu Fossiz Cretaceous Birps, to illustrate
XXVII. Prof. Seeley’s paper on those fossils ..............ceee.s-cesees 512

XXVIII. | Hairran Fossin SHExzs, to illustrate Mr. R. J. L. Guppy’s
XXIX. paper on the Miocene fossils of Haiti............. “runeone eer 532

ERRATA ET CORRIGENDA.

At p. 30, line 22, for substnces read substances.
45, line 23, for post-liassic read prze-Liassic ?
line 30, for Guibon read Guybon.
46, last line, for Ketospondylian read Tretospondylian.
59, line 2, for carbonarius read anglicus.
124, figure, the bracket of the ‘‘ Laminated series” should include only
beds 2-4.
205, line 5, and 210, line 8 from bottom, for Plate XV. read Plate XVI.

GENERAL INDEX

TO

THE QUARTERLY JOURNAL

AND

PROCEEDINGS OF THE GEOLOGICAL SOCIETY.

a

[The fossils referred to are described ; and those of which the names are printed
in italics are also figured. |

Acanthodes Wardi, 335.

Adamello group, analysis of tonalite
of the, 309.

AXglina Hughesii, 138.

Afon Trystion, section along the val-
ley of the Dee through, 228.

Agglomerate, spheroidal structure in,
at Burntisland, 150.

Agglomerates of the Schemnitz dis-
trict, 294.

Aitken, John, Esq., observations on
the unequal distribution of drift on
opposite sides of the Pennine chain
in the country about the source of
the Calder, with suggestions as to
the causes which led to that result,
together with some notices on the
high-level drift in the upper part
of the valley of the river Irwell,
184.

Algee, Prof. Duncan on some unicel-
lular, parasitic within Silurian and
Tertiary corals, with a notice of
their presence in Calceola sanda-
lina, 205.

Allport, S., Esq., on the metamorphic
rocks surrounding the Land’s-End
mass of granite, 407.

Alluvial deposits, Mr. Worth on cer-
tain, associated with the Plymouth
limestone, 236.

Alpes Suisses, Prof. A. Favre sur les
anciens glaciers du revers septen-
trional des, Proc. 125.

Alps, section from the, to Brighton,
New Zealand, 54.

Amphiccelian vertebree, 44.

Analyses of slates of the Lake-district,
VOL, XXXII.

5; of altered Skiddaw slates, quartz
felsite, syenitic granite, and altered
ash, 22; of altered crystalline rocks
from the Penzance district, 162; of
erystalline slaty rocks from Western
Cornwall, 165; of volcanic rocks of
Schemnitz, 303; of the syenite of
Hodritsch, ‘‘ greenstone-trachyte”
of Gelnerowsky Wrch, and horn-
blende-andesite of Rybnik, 308 ; of
the dacite of Kis Sebes, banatite of
Szaska, tonalite of the Adamello
group, and quartziferous horn-
blende-andesites of the Euganean
hills, 309.

Analysis of rock from Chapel rock,
164; of crystalline rock from New-
lyn, 167; of hornblende from Bo-
tallack, 167; of crystalline rock
from Pendower Cove, 171; of green-
stone from near Camborne, 174; of
greenstone from Newlyn Kast, 175;
of elvan from near St. Stephens, 176.

Andesite, hornblende-, of Rybnik,
analysis of, 308.

Andesites, quartziferous hornblende-,
of the Euganean Hills, analysis of,
309.

Andesitic rocks of the Schemnitz dis-
trict, 295.

Anglesey, Prof. Ramsay on, 116.

Anniversary Address of the President,
Proc. 53-121. See also, Evans,
John, Esq.

Annual Report for 1875, Proc. 15.

Antler, implements of, from the
gas Cave, Creswell Crags,

GENERAL INDEX.

Aplite forming a vein in gneiss at
Schuttrisberg, 320.

Arachnida, fossil, of the Paleozoic
rocks, 63.

Archeocidaris Urii, 39.

Archeospherine, 73.

Arta, submarine crater within the har-
bour of Karavossera, in the Gulf of,
Proc. 12.3.

Arvicola amphibia, occurrence of, in
the Robin-Hood Cave, Creswell
Crags, 247, 249.

Ash, altered, analysis of, 22.

Astrocrinites, Mr. R. Etheridge, jun.,
on the oecurrence of the genus, in
the Scotch Carboniferous Limestone,
103.

Astrocrinites Benniei, 105.

Augite-andesites of Schemnitz, analy-
sis of, 803, 304.

Australian seas, Cainozoic physical
geography of the, 345.

Bala lake, section of the valley of the
Dee through the, 228.

Balliway Rigg, hypersthenite of, 21.

Banatite of Szaska, analysis of, 309.

Barton clay, Prof. H. G. Seeley on
the occurrence of a British fossil
Zeuglodon in the, of the Hamp-
shire coast, 429.

Basalt, spheroidal structure in, 150,
151; thinning out of the, at the
High Force, 470.

Basalts of the Schemnitz district,
301.

Bathygnathus borealis, 360.

Bear, brown, occurrence of, in the
Robin-Hood Cave, Creswell Crags,
247, 248.

, grizzly, occurrence of, in the
Robin-Hood Cave, 247, 248.

Beer Head, section from Sidmouth to,
276.

Belt, Thomas, Esq., the drift of Devon
and Cornwall, its origin, correla-
tion with that of the south-east of
England, and place in the Glacial
series, 80.

Bettany, G. T., Esq., on the genus
Merycocherus (family Oreodonti-
dz), with descriptions of two new
species, 259.

Birds, British fossil cretaceous, Prof.
H. G. Seeley on the, 496.

Bison priscus in the Robin-Hood Cave,
Creswell Crags, 247, 249.

Black river, New York, grooves and
channellings, 76.

Bone-bed, Mr. Davis on a, of the lower
Coal-measures, and an enumeration

of the fish-remains of which it is
principally composed, 332.

Bone-bed, section to show the position
of, near Bradford and Clifton, York-
shire, 333. ;

Bone-caves, Rev. J. M. Mello on the,
of Creswell Crags, 240.

Bonnevie, cross joints at, 148.

Bonney, Rev. T. G., on columnar,
fissile, and spheroidal structure,
140. 3

Bos longifrons in the surface-soil of
the Robin-Hood Cave, 247.

Botallack, analysis of hornblende
from, 167.

‘Botter rock, altered dolerite from, 422.

Boulder-clay in East Anglia, 197.

Boulogne-sur-Mer, Mr. H. Wood-
ward on some new Macrurous Crus-
tacea from the Kimmeridge clay of,
47.

Bradford, section to show the position
of bone-bed near, Yorkshire, 333.
Branscomb Mouth, section from Sid-

mouth to Beer Head through, 276.

Brentor, altered dolerites from, 421.

Bridgewater area, trias of the, 373.

Brighton, section from the Alps to,
New Zealand, 54.

Brithopus priscus, 358.

British Coal-measures, Mr.. H. Wood-
ward on the discovery of a fossil
scorpion in the, 57.

fossil cretaceous birds, Prof.

H. G. Seeley on the, 496.

Zeuglodon, Prof. H. G.
Seeley on the occurrence of the re-
mains of a, (Z. Wanklyni, Seeley)
in the Barton clay of the Hamp-
shire coast, 429.

Budleigh Salterton, pebble-beds of,
382.

Burlescombe, pebble-beds of, 382.

Burntisland, spheroidal structure in
agglomerate, 150.

Buttermere, syenitic granite of, 14, 18.

and Ennerdale syenitic gra-

nite, map showing the distribution

of, 15.

Caernarvon, Mr. Marr on _fossili-
ferous Cambrian shales near, 134.
Cainozoie physical geography of the

Australian seas, 345.

Calceola sandalina, Prof. Duncan on
the presence of unicellular algze in,
205.

Calder, Mr. Aitken on the unequal
distribution of drift about the source
of the river, 184.

Caldew, granite of, 3.

GENERAL INDEX.

Callianassa isochela, 48.

Callorhynchus Hectori, 329.

Camborne, analysis of greenstone from
near, 174.

and St. Erth, rocks of the dis-
trict between, 172.

Cambrian rocks, Dr. J. W. Dawson on
the phosphates of the, of Canada,
285.

Cambridge Upper Greensand, Prof. H.
G. Seeiey on an associated series of
cervical and dorsal vertebre of
Polyptychodon from the, 433; Prof.
H. G. Seeley on Crocodilus icenicus
(Seeley), a second and a larger
species of crocodile from the, 457;
Prof. H. G. Seeley on Macruro-
saurus semnus (Seeley), a long-tailed
animal with procclous vertebre
from the, 440.

valley, drift-deposits in the, 196.

Canada, Dr. J. W. Dawson on the
phosphates of the Laurentian and
Cambrian rocks of, 285.

Cancellaria epistomifera, 520.

Canis familiaris in the surface-soil of
Robin-Hood Cave, 247.

—— lupus in the Robin-Hood Cave,
247, 248.

—— vulpes in the Robin-Hood Cave,
247, 248.

Cape Cornwall, altered slates from,
410

Capra hircus in the surface-soil of the
Robin-Hood Cave, 247.

Carboniferous limestone, Mr. R. Ethe-
ridge, jun., on the occurrence of the
genus Astrocrinites in the Scotch,

103.
, section of hzmatite vein
in, Parkside, 183.
group, lower, Mr. R. Ethe-
ridge, jun., on an adherent form of
Productus and a small Spiriferina
from the, of the east of Scotland,
454.
Carnivorous reptile, Prof. Owen on a,

95.

Carrock Fell, rocks of, 16, 20; hyper-
sthenite of, 21; map showing the
distribution of, 16.

Caryocaris Marrii, 138.

Castle Horneck, crystalline slaty rock
from Rosehill, 166.

Cefn caves, Mr. D. Mackintosh on the
correlation of the deposits of the,
with the drifts of the north-west of
England, 91.

Cerithium uniseriale, 519.

obesum, 519.

Cervus megaceros and tarandus in

the Robin-Hood Cave, Creswell
Crags, 247, 249.

Chapel rock, rock from, 163; analysis
of, 164.

Chesil beach, Proc. 6.

Chiastolite slate of the Lake-district,
Dy.d5 De

Chillesford clay, Mr. J. Gunn on the
position of the forest-bed series be-
neath the, 123.

Chimezroid jaws, Mr. Newton on two,
from the Lower Greensand of New
Zealand, 326.

Chit Rocks, section from Ladram Bay
to Sidmouth through, 276.

Chyandour, Penzance, crystalline slaty
rock from, 166.

Cinderhill cutting, 533.

Clavatula labiata, 527.

Clay, Mr. W. Ramsay on the influ-
ence of various substances in accele-
rating the precipitation of, 129.

Clay-slate, metamorphism of, around
granite, 9.

Clea truncata, 524. :

Clicker Tor, altered dolerite from,
422,

Cliff sections, Mr. Johnston Lavis on
the Triassic strata exposed in the;
near Sidmouth, 274.

Clifton, Yorkshire, section to show the
position of bone-bed near, 333.

Clough, C. T., Esq., on the section at
the High Force, Teesdale, 466.

Coal, Mr. D. Mackintosh on the mode
of occurrence and derivation of beds

of drifted, near Corwen, North
Wales, 451.

Coalbrook Dale, remanié gravels of
Proc. 5

Coal-measures, Mr. H. Woodward on
the discovery of a fossil scorpion in
the British, 57.

— of Scotland, Mr. H. Woodward
on a remarkable Orthopterous in-
sect from the, 60.

Celacanthus lepturus, 339.

Columnar structure, Rey. T’.. G. Bon-
ney on, 140.

and fissile structures in the
Roche Sanadoire, 143.

Columns in lava at Royat, 148.

-—— and cross joints, Roche Tuilliére,
144.

—— and joints, Gross Weilberg,
145.

Conglomerates of Milverton, 382.

Conus marginatus, 528.

Corals, Prof. Duncan on the presence
of unicellular algz parasitic within
Silurian and Tertiary, 205,

GENERAL INDEX.

Corals, Prof. Duncan on some fossil
reef-building, from the Tertiary de-
posits of Tasmania, 341.

Cornwall, Mr. Belt on the drift of,
80.

, Mr. J. A. Phillips on the so-
called “‘Greenstones” of Western,
155.

Corwen, Mr. 1D. Mackintosh on the
mode of occurrence and derivation
of beds of drifted coal near, North
Wales, 451.

Cote St. Pierre, Dr. J. W. Dawson on
the occurrence of Eozoon canadense
at, 66.

Crab, Mr. Henry Woodward ona new
fossil, from the Tertiary of New
Zealand, 51.

Crag Bridge, quartz-felsite of, 14, 18.

Crater, submarine, within the har-
bour of Karavossera, in the Gulf of
Arta, Proc. 123.

Creswell Crags, Rev. J. M. Mello on
the bone-caves of, 240.

Cretaceous birds, Prof. H. G. Seeley
on the British, 496.

Crocodilus icenicus (Seeley), Prof. H.
G. Seeley on, a second and larger
species of crocodile from the Cam-
bridge Upper Greensand, 487.

Croll, James, Esq., award of the Mur-
chison Geological Fund to, Proc. 50.

Cross joints and columns, Roche Tuil-
liére, 144.

at Bonnevie, 148.

Crown Hili, section across the valley
of the Willow Brook through, 217.

Crustacea, macrurous, from the Kim-
meridge Clay of the Sub-Wealden
boring, Sussex, and from Boulogne-
sur-Mer, Mr. H.Woodward on some
new, 47.

Crystalline rocks, altered, from the
Penzance district, analysis of, 162.

slaty rocks from Western Corn-
wall, analysis of, 165.

Ctenacanthus hybodoides, 335.

, 8p., 335

Cumberland, section across hematite
deposits, Millom, 181.

Cup-and-ball structure, 147.

Curvitabular structure, 146.

Cynodraco major, 95.

serridens, 357.

Cyprea Gabbiana, 528.

Dacite of Kis Sebes, analysis of, 309.

of Schemnitz, analysis of, 303.

Davis, J. W., Esq., on a bone-bed in
the Lower Coal-measures, with an
enumeration of the fish-remains of

which it is principally composed,
332.

Dawkins, W. Boyd, Esq., on the Mamn-
malia and traces of man found in
the Robin-Hood Cave, 245.

Dawson, Dr. J. W., onthe occurrence
of Kozoon canadense at Cote St.
Pierre, 66.

——, on the phosphates of the Lau-
rentian and Cambrian rocks of
Canada, 285.

Dee, Prof. Ramsay on the physical
history of the, 219; section along
the course of the valley of the, 228.

Denuding agencies, Proc. 4.

Deposition, geological, Proc. 4.

Deposits, Mr. D. Mackintosh on the,
of the Cefn and Pont-Newydd caves,
os

Deuterosaurus biarmicus, 357.

Devon, Mr. Belt on the drift of, 80.

, Mr. W. A. E. Ussher on the
Triassic rocks of Somerset and, 367.

Devonshire, Mr. H. B. Woodward on
gravels and sands of Newton Abbot,
230.

, Trias of, 377.

Dinosaurian ilium, Mr. Hulke on a
modified form of, hitherto reputed
scapula, 364.

Diorite of Miton Hill and Round
Knoit, 16, 21.

Diplodus gibbosus, 336.

Diszowa Wrch, section from the
Hisenbach valley to the Hodritsch
valley through the, 319.

Dog, absence of, in the breccia of the
Robin-Hood Care, Creswell Crags,
246; occurrence of, in the surface-
soil of Robin-Hood Cave, 247.

Dolerites, altered, 421.

Dorsetshire, drift-beds of, 86.

Drift, Mr. Belt on the, of Devon and
Cornwall, 80.

, Mr. J. Aitken on the high-level,

on the upper part of the valley of

the river Irwell, 184.

, Mr. Aitken on the unequal dis-

tribution of, on opposite sides of

the Pennine chain, 184.

: aaa in the Cambridge valley,

south and east of the
pak escarpment in Hast Anglia,
192.

Drifts, Mr. D. Mackintosh on the, of
the north-west of England and
Wales, 91.

Duncan, Prof. P. M., on some uni-
cellular alge parasitic within
Silurian and Tertiary corals, with

GENERAL INDEX.

a notice of their presence in Calce-
ola sandalina and other fossils, 205.
Duncan, Prof. P. M., on some fossil
reef-building corals from the Ter-
tiary deposits of Tasmania, 341.
Dunscombe Hill, section from Sid-
mouth to Beer Head through, 276.
Dykes, Mr. R. Mallet on the mecha-
nism of production of volcanic, 472.

Earthquakes of the Schemnitz district,
305

East Anglia during the Glacial period,
Mr. Penning on the physical geo-
logy of, 191.

Echini, Paleozoic, Mr. W. Keeping
on the, 35.

Hehinoidea, classification of, 40.

Ectracheliza, 524.

Hisenbach valley, section from the, to
the Hodritsch valley, 319; section
through the Schwatzer Berg passing
through, 319.

Elephas primigenius, occurrence of,
in the Robin-Hood Cave, 247, 249.

Elk, Ivish, occurrence of, in the
Robin-Hood Cave, 247, 249.

Elvan near St. Stephens, 175; ana-
lysis of, 176.

Emys hordwellensis (Seeley), Prof. H.
G. Seeley on remains of, from the
Lower Hordwell beds in the Hord-
well Cliff, 445.

Enaliornis Barretti, 499.

Sedgwicki, 506.

England, Mr. Belt on the correlation
of the drift of Devon and Cornwall
with that of the south-east of, 80.

, Mr. E. Wilson on the Permians

of the north-east of, 533.

and Wales, Mr. D. Mackin-
tosh on the correlation of the de-
posits in Cefn and Pont-Newydd
caves with the drifts of the north-
west of, 91.

English Channel, shape of the, Proc. 6.

Ennerdale and Buttermere syenitic
granite, map showing the distribu-
tion of, 15.

and Buttermere syenitic granite,
microscopical examination of, 18.

Eozoon canadense, Dr. J. W. Dawson
on the occurrence of, at Cote St.
Pierre, 66.

Equus caballus in the Robin-Hood
Cave, Creswell Crags, 247, 249.

Etheridge, Robert, Jun., Esq., on the
occurrence of the genus Astrocri-
nites (Austin) in the Scotch Car-

_ boniferous-Limestone series, with
the description of a new species (A.

Benniet), and remarks on the genus,

103. :

Etheridge, Robert, Jun., on an ad-
herent form of Productus and a
small Spirtferina from the Lower
Carboniferous Limestone group of
the east of Scotland, 454.

Etna, 472.

Huganean Hills, analysis of quartzife-
rous hornblende-andesite of the,
309.

Eurosaurus uralensis, 356.

Euscorpius anglicus, 58.

Evans, John, Esq. (President), Ad-
dress on presenting the Wollaston
Gold Medal to Prof. Huxley, Proc.
47; Address on presenting the ba-
lance of the Wollaston Donation
Fund to J. Gwyn Jeffreys, Esq.,
for transmission to Prof. Giuseppe
Seguenza of Messina, 48; Address
on handing the Murchison Medal
to Prof. Ramsay, for transmission
to A. R. C. Selwyn, Esq., 49; Ad-
dress on presenting the balance of
the Murchison Geological Fund to
Prof. Ramsay for transmission to
James Croll, Esq., 50; Address on
presenting the Lyell Medal and
Fund to Prof. Morris, 51. Anni-
versary Address, February 18th,
1876: Obituary Notices of Deceased
Fellows :—Sir Charles Lyell, 53;
Mr. George Poulett Scrope, 69 ; Sir
William Edmond Logan, 76; M.
G. P. Deshayes, For. Memb., 80;
Mr. W. J. Henwood, 82; Mr. Wil-
liam Sanders, 85; the Ven. Arch-
deacon W. EK. Hony, 86; the Right
Hon. Sir Edward Ryan, 86; Col.
R. Baird Smith, C.B., 87; Mr. T.
G. B. Lloyd, 87; Mr. Nathaniel
Plant, 88; Mr. M. W. T. Scott, 88 ;
Mr. James Middleton, 89; Mr.
F. E. Edwards, 89; and Mr. N.-
T. Wetherell, 90; Address on the
recent progress and probable future
of geological research, with parti-
cular reference to deep-sea deposits
and the causes of the changes of
climate, and observations on the
sources of the water-supply of Lon-
don, 91-121.

Favre, Prof. A., sur les anciens glaciers
du revers septentrional des Alpes
Suisses, Proc. 125.

Felis spelaea in the Robin-Hood Cave,
Creswell Crags, 247, 248.

Felsite, quartz-, map showing the dis-
tribution of, of St. John’s vale, 138.

GENERAL INDEX.

Felsite, spherulitic, of Carrock Fell and
Great Lingy Hill, 16, 20.

Fish-remains, Mr. Davis on the, com-
posing a bone-bed of the Lower
Coal-measures, 3382.

Fissile and columnar structures in
the Roche Sanadoire, 148.

structure, Rev. T. G. Bonney
on, 140.

Flint implements from the Robin-
Hood Cave, Creswell Crags, 252.

Fluid-cavities, 415.

Forest-bed series, Mr. J. Gunn on the
presence of the, at Kessingland and
Pakefield, in Suffolk, 123.

Fossil Zeuglodon, Prof. H. G. Seeley
on the occurrence of remains of a
British, (Z. Wanklyn7, Seeley) in
the Barton Clay of the Hampshire
coast, 429.

Fox, occurrence of, in the Robin-Hood
Cave, Creswell Crags, 247, 248.

Fusus Henekeni, 524.

Gelnerowsky Wrch, analysis of green-
stone-trachyte of, 308.

Geological structure of the Schemnitz
district, 294.

Geology of New York, Mr. Lloyd on
the, 76.

Giant’s Kettle, section of, in Lauren-
tian granite, in Pulpit Rock, near
Oxbow, 78.

Glacial climate, Mr. J. J. Murphy on
the, and the polar ice-cap, 400.

deposits south and east of the

chalk escarpment in East Anglia,

192.

period, Mr. Penning on the phy-

sical geology of East Anglia during

the, 191.

phenomena,

classification of,

series, Mr. Belt on the place of
the drift of Cornwall and Devon in
the, 80.

Glaciers, anciens, du revers septen-
trional des Alpes Suisses, Prof. A.
Favre sur les, Proc. 125.

Glashiitte valley, section through the
Kohl-Berg passing through the,
318.

Goat, occurrence of, in the surface-
soil of the Robin- Hood Cave,
247.

Granite, formation of, at various

. depths, 28; as a product of meta-
morphism, 30.

, map showing the distribution of

the Buttermere and Ennerdale

syenitic, 15,

Granite, metamorphism of clay-slate
around, 9.

, Mr. 8. Allport on the meta-

morphic rocks surrounding the

Land’s-End mags of, 407.

of the Lake-district, 3, 5.

——., Skiddaw, 5.

, Syenitic, of Buttermere and En-
nerdale, 14, 18.

Granitic, granitoid, and associated
metamorphic rocks of the Lake-
district, Mr. J. C. Ward on the, 1.

Granulite, 320.

Gravels, Mr. H. B. Woodward on the,
of Newton Abbot, Devonshire, 230.

Great Lingy Hill, spherulitic felsite
of, 16, 20.

Greenland, Proc. 5.

Green Point, section from Ladram
Bay to Sidmouth through, 276.

Greensand, Lower, Mr. Newton on
two chimeroid jaws from the, of
New Zealand, 326.

, Upper, Prof. H. G. Seeley on

an associated series of cervical and

dorsal vertebra of Polyptychodon

from the Cambridge, 433.

: on Crocodilus icenicus

(Seeley), a second and larger species

a crocodile from the Cambridge,
‘€

, —- on Macrurosaurus
semnus (Seeley), a long-tailed ani-
mal with procclous vertebre from
the Cambridge, 440.

Greenstone-trachytes of Gelnerowsky
Wreh, analysis of, 308.

of Hungary, 298.

Greenstones of Western Cornwall,
Mr. J. A. Phillips on the so-called,
155.

Grooves and channellings,
River, New York, 76.

Gross Weilberg, columns and joints in,
145.

Gulval, crystalline slaty rock from
Rosemorran, 166.

Gunn, John, Esq., on the presence of
the Forest-bed series at Kessingland
and Pakefield, in Suffolk, and its
position beneath the Chillesford
Clay, 128.

Guppy, R. J. L., on the Miocene fos-
sils of Haiti, 516.

Gurnard’s-Head district, rocks of, 170.

Gyracanthus formosus, 334.

-—— tuberculatus, 334.

Gyrolepis Rankinii, 338.

Black

Hematite deposits, section across,
Millom, Cumberland, 181.

GENERAL INDEX,

Hematite in the Silurians, Mr. Ken-
dall on, 180.

vein, section of, in carboni-

ferous limestone, Parkside, 183.

, vertical section of, 182.

Haiti, Mr. R. J. L. Guppy on the
Miocene fossils of, 516.

Hampshire, drift-beds of, 86.

coast, Prof. H. G. Seeley on the
oceurrence of a British fossil Zeu-
glcdon in the Barton Clay of the,
429,

Hare, occurrence of, in the Robin-
Hood Cave, 246, 247, 249.

Harpactocarcinus tumidus, 52.

Harrison, W. J., Esq., on the occur-
rence of Rheetic beds in Leicester-
shire, 212.

Hector, Dr., on the geology of part of
New Zealand, 53.

Heliastrea tasmaniensis, 342.

Helston, altered dolerite from, 422.

Hempshill, section at, 533.

Hennock, altered dolerites
421.

Hicks, H., Esq., on the physical con-
ditions under which the Upper Si-
lurian and succeeding Palzozoic
rocks were probably deposited over
the northern hemisphere, Proc. to.

, appendix to Mr. Marvr’s paper
on fossiliferous Cambrian shales
near Caernarvon, 154.

High Force, Mr. C. T. Clough on the
section at the, Teesdale, 466.

, section at the, 467; thin-
ning out of the basalt at the, 470.
Hodritsch, analysis of syenite of,

308.

, section through the Schwatzer

Berg passing through, 319.

valley, section from the Hisen-
bach valley to the, 319.

Hog, occurrence of, in the surface-soil
of Robin-Hood Cave, 247.

Hoplonchus, 336.

Hordwell Cliff, Prof. H. G. Seeley on
remains of mys hordwellensis
(Seeley) from the Lower Hordwell
beds in the, 445.

Hornblende from Botallack, analysis
of, 167.

Hornblende-andesite of Rybnik, ana-
lysis of, 308.

Hornblende- and mica-andesites of
Schemnitz, analysis of, 303.

Horn Rock, section from Ladram Bay
to Sidmouth, through, 276.

Horse, occurrence of, in the Robin-
Hood Cave, Creswell Crags, 247,
249.

from,

He springs of the Schemnitz district,
04.

Hulke, J. W., Esq., appendix to
** Note on a modified form of Dino-
saurian ilium, hitherto reputed sca-
pula,” 364.

Hungary, Mr. Judd on the ancient
younne of the district of Schemnitz,

Hunters, palzolithic, north-western
range of, 255.

Huxley, Prof., award of the Wollaston
Medal to, Proc. 47.

Hyznas in the Robin-Hood Cave,
Creswell Crags, 245, 246, 247, 248.

Hypersthenite of Mosedale Crags and
Round Dale, 17, 21; of White
Crags, analysis of, 23.

Ice, action of, Proc. 4.
Igneous origin, metamorphic rocks of,
418

Iguanodon, pelvis of, 364.

lium, Mr. Hulke on a form of Dino-
saurian, hitherto reputed scapula,
364.

Implements from the Robin-Hood
Cave, Creswell Crags, 249.

Insect, Mr. H. Woodward on a re-
markable Orthopterous, from the
Coal-measures of Scotland, 60.

Insects, fossil, of the Paleozoic rocks,
63

Trish Sea, Proc. 6.

Tronstone, implements of, from the
Robin-Hood Cave, Creswell Crags,
250.

Irving, Rev. A., on some recent sec-
tions near Nottingham, 515.

Irwell, Mr. J. Aitken on the high-level
drift in the upper part of the valley
of the river, 184.

Ischyodus brevirostris, 327.

Jefferson County, New York, grooves
and channellings on the Black River
at Watertown, 76; Giant’s Kettle
near Oxbow, 78.

Judd, J. W., Esq., on the ancient vol-
cano of the district of Schemnitz,
Hungary, 292.

Kanzian Wrch, section through the
Schwatzer Berg passing through,
319.

Karavossera, submarine crater within
the harbour of, Proc. 123.

Keeping, Walter, Esq., on the Palmo-
zoie Kehini, 35.

, on the discovery of Melonites in

Britain, 395.

GENERAL INDEX.

Kendall, J. D., Esq., hematite in the
Silurians, 180.

Kent, drift-beds of, 87.

Kessingland, Mr. J. Gunn on the pre-
sence of the Forest-bed series at,
123; section of cliff at, 124.

Kimmeridge Clay of the Sub-Wealden
boring, Sussex, and of Boulogne-
sur-Mer, Mr. H. Woodward on
some new macrurous crustacea from
the, 47.

Kis Sebes, analysis of dacite of, 309.

Klokee Wrch, section from the Eisen-
bach valley to the Hodritsch valley
through the, 319.

Kohlberg, section through the, 318.

Kompberg, section from the Eisen-
bach valley to the Hodritsch valley
through, 319.

Labyrinthodon, Prof. Seeley on the
posterior portion of the lower jaw
of a (L. Lavisi), 278.

Labyrinthodon Lavisi, 278.

Labyrinthodont, Mr. Johnston Lavis
on the occurrence of an ossiferous
zone containing bones of a, 274.

Ladram Bay, section from, to Sid-
mouth, 276.

Lake-district, Mr. J. C. Ward on the
granitic, granitoid, and associated
metamorphic rocks of the, 1.

Land’s-End mass of granite, Mr. S.
Allport on the metamorphic rocks
surrounding the, 407.

Lang Dale, hypersthenite of, 21.

La Plata, estuary of the, Proc. 6.

Laurentian granite, section of Giant's
Kettle in, in Pulpit Rock near
Oxbow, 78.

rocks, Dr. Dawson on the phos-
phates of the, of Canada, 285.

Lava, columns in, at Royat, 148.

Lavas, ancient, of the Schemnitz di-
strict, 294.

Lavis, H. J. Johnston, Esq., on the
Triassic strata exposed in the cliff-
sections near Sidmouth, and a note
on the occurrence of an ossiferous
zone containing bones of a Laby-
rinthodont, 274.

Leicester, section across the valley of
the Willow Brook at, 217; section
of rhxtic beds, Spinney Hills, near,
213.

Leicestershire, Mr. Harrison on the
occurrence of rheetie beds in, 212.

Lepidermata, Echinoidea, 40.

Lepracanthus Colei, 335.

Lepus timidus in the Robin-Hood
Cave, Creswell Crags, 247, 249.

Le Puy, spheroids in an unjointed
column near, 151.

Limestone, Plymouth, Mr. R. N.
Worth on certain alluvial deposits
associated with the, 236.

Lion, occurrence of the, in the Robin-
Hood Cave, Creswell Crags, 247,
248.

Lithia mica, 417.

Lithomantis carbonarius, 63.

Littlecomb Hill, section from Sid-
mouth to Beer Head through, 276.

Lloyd, T. G. B., Esq., geological notes
from the State of New York, 76.

Lodes of Schemnitz, 322.

Lower Carboniferous Limestone group,
Mr. R. Etheridge, jun., on an ad-
herent form of Productus and a
small Spiriferina from the, of the
East of Scotland, 454.

Coal-measures, Mr. Davis on

a bone-bed of the, 332.

Glacial deposits south and east

of the chalk escarpment in East

Anglia, 192.

Greensand, Mr. Newton on

two chimeroid jaws from the, of

New Zealand, 326.

Hordwell beds, Prof. H. G.
Seeley on remains of Hmys hord-
wellensis (Seeley) from the, in the
Hordwell Cliff, 445.

Lyell Medal and Fund, award of the,
to Prof. Morris, Proc. 51.

Lyell, Sir Charles, extract from the
Will of, Proc. 17.

Lycosaurus curvimola, 357, 360.

Mackintosh, D., Esq., on the correla-
tion of the deposits in Cefn and
Pont-Newydd caves with the drifts
of the north-west of England and
Wales, 91.

, on the mode of occurrence and
derivation of beds of drifted coal
near Corwen, North Wales, 451.

Macrurous Crustacea from the Kim-
meridge Clay of the Sub-Wealden
boring, Sussex, and from Boulogne-
sur-Mer, Mr. H. Weodward on
some new, 47.

Macrurosaurus semnus (Seeley), Prof.
H. G. Seeley on a long-tailed
animal with procclous vertebre
from the Cambridge Upper Green-
sand, 440.

Mallet, R., Esq., on the mechanism of
production of volcanic dykes, 472.
Mammalia, Mr. W. Boyd Dawkins on
the, found in the Robin-Hood Cave,

245.

GENERAL INDEX.

Mammoth, occurrence of, in the
Robin-Hood Cave, Creswell Crags,
247, 249.

tooth, implements of, from the
aoe Cave, Creswell Crags,
249.

Man, Mr. W. Boyd Dawkins on the
traces of, found in Robin-Hood
Cave, 245.

Map showing the distribution of the
quartz felsite of St. John’s vale,
13 ; showing a part of the boundary
of the Buttermere and Ennerdale
syenitic granite, 15; showing the
distribution of the Carrock rocks, 16.

Marginella Sowerbyi, 528.

Marine deposits, Proc. 4.

Marr, J. E., Esq., fossiliferous Cam-
brian shales near Caernarvon, with
an Appendix by H. Hicks, Hsq., 134.

Mecochirus Peytoni, 48.

Megalichthys Hibberti, 338.

Mello, Rev. J. Magens, on the bone-
caves of Creswell Crags, 240.

Melonites, Mr. W. Keeping on the
discovery of, in Britain, 395.

Etheridgit, 397.

Melosaurus uralensis, 355.

Mendip area, Trias of the, 367.

Merychyus, 262.

Merycocherus, Mr. Bettany on the
genus, 259.

— Leidyi, 270.

temporalis, 269.

Metamorphic action, minerals pro-
duced by, 411.

, granitic, and granitoid rocks of

the Lake-district, Mr. J. C. Ward

on the, 1.

rocks, Mr. 8S. Allport on the,
surrounding the Land’s-End mass
of granite, 407.

Metamorphism, distribution of, 30;
of clay-slate around granite, 9.

Miaulis, Capt., on the occurrence of a
submarine crater within the harbour

. of Karavossera, in the gulf of Arta,
PROC. 12%.

Mica and hornblende andesites of
Schemnitz, analysis of, 303.

Mica, lithia, 417.

Mica-schist, 409 ; of the Lake-district,
3, 4, 5.

Microscopie characters of St.-John’s
quartz felsite and altered Skiddaw
slate, 18; of Crag-Bridge quartz
felsite, 18; of Buttermere and
Ennerdale granite and altered
Skiddaw slate, 18; of Carrock-Fell
rocks, 19; of the spherulitic felsite
of Carrock Fell, Rae Crags, and

Great Lingy Hill, 20; of the dio-
rite of Round Knott and Miton
Hill, 21; of the hypersthenite of
Carrock Fell, Mosedale, Langdale,
and Balliway Rigg, 21.

Middle Glacial deposits south and east
of the chalk escarpment in East
Anglia, 192.

Milber Down, section from near New-
ton Abbot to, 230.

Millom, Cumberland, section across
hematite deposits, 181.

Milverton, conglomerates of, 382.

Minerals produced by metamorphic
action, 412.

Miocene fossils of Haiti, Mr. R. J. L.
Guppy on the, 516.

volcanic activity in Hungary, 312.

Mississippi, Proc. 4.

Miton Hill, diorite of, 16, 21.

Moel-Tryfaen, Proc. 5.

Monte Somma, voleanic dykes of, 472.

Morris, Prof., award of the Lyell
Medal and Fund to, Proc. 51.

Mosedale Crags, hypersthenite of, 17.

Murat, spheroid in basalt near, 150.

Murchison Medal, award of the to
Mr. A. R. C. Selwyn, Proc. 49;
Geological Fund, award of the, to
Mr. James Croll, Proc. 50.

Murex cornurectus, 521.

textilis, 522.

Murphy, Joseph John, Esq., the
Glacial climate and the Polar ice-
cap, 400.

Newlyn, analysis of crystalline rock
from, 167.

, Hast, analysis of greenstone
from, 175.

Newton, H. T., Esq., on two chime-
roid jaws from the Lower Green-
sand of New Zealand, 326.

Newton Abbot, Mr. H. B. Woodward
on the gravels, sands, &c. of, 230.

, section from near, to Milber
Down, 280.

New York, Mr. Lloyd on the geology
of the State of, 76.

New Zealand, Dr. Hector
geology of part of, 53.

——, Mr. Newton on chimeroid

jaws from the Lower Greensand of,

326.

on the

, Mr. H.Woodward on a new

fossil crab from the tertiary of, 51.

-——, section from the Alps to
Brighton, 54.

Northern hemisphere, Mr. H. Hicks
on the physical conditions under
which the Upper Silurian and suc-

GENERAL INDEX.

ceeding Paleozoic rocks were pru-
bably deposited over the, Proc. 10.

North Huel Rose, greenstone from,
175; analysis of, 175.

Wales, Mr. D. Mackintosh on
the mode of occurrence and deri-
vation of beds of drifted coal near
Corwen, 451.

Nottingham, section of Permian and
Triassic rocks on the Great Northern
Railway near, 536.

— —, Mr. A. Irving on some recent
sections near, 513.

Oreodon, 262.

Orthacanthus cylindricus, 336.

Orthopterous insect, Mr. H. Wood-
ward on a remarkable, from the
Coal-measures of Scotland, 60.

Orthopus primevus, 353.

Ossiferous zone, Mr. Johnston Lavis
on an, containing bones of a Laby-
rinthodont, 274.

Owen, Prof., evidence of Theriodonts
in Permian deposits elsewhere than
in South Africa, 352.

, evidence of a carnivorous reptile
(Cynodraco major) about the size
of a lion, with remarks thereon, 95.

Ox, short-horned, occurrence of, in
the Robin-Hood Cave, 247.

Oxbow, Jefferson County, New York,
Giant’s Kettle near, 78.

Pakefield, Mr. J. Gunn on the pre-
sence of the forest-bed series at, 123 ;
section of cliff at, 124.

Paleachlya perforans, 210.

Palechinus intermedius, 37.

gigas, 38.

sphericus, 38.

Paleolithic hunters,
range of, 255.

Palzozoie Echini, Mr. W. Keeping on
the, 35.

insects and arachnida, 63.

rocks, Mr. H. Hicks on the
physical conditions under which the
Upper Silurian and succeeding, were
probably deposited overthenorthern
hemisphere, Proc. 10.

Pareiasawrus bombidens, 44.

serridens, 45.

Parkside, section of hzmatite vein in
Carboniferous Limestone at, 183.
Paul Hill, crystalline slaty rock from,

165.

Peake Hill, section from Ladram Bay
to Sidmouth through, 276.

Pebble-beds of Budleigh Salterton and
Burlescombe, 382.

north-western

Pendower cove, analysis of crystalline
rock from, 171.

Pennine chain, Mr. Aitken on the un-
equal distribution of drift on op-
posite sides of the, 184.

Penning, W. H., Esq., notes on the
physical geology of East Anglia
during the glacial period, 191.

Penzance, crystalline slaty rock from
Chyandour, 166.

district, analyses of altered crys-
talline rocks from the, 162.

Perischodomus, 35.

Permian deposits, Prof. Owen on evi-
dence of Theriodonts in, elsewhere
than in South Africa, 352.

and Triassic strata near Not-
tingham, 513, 533.

Permians of the north-east of Eng-
land, Mr. E. Wilson on the, 533.
Pic de Sancy, columns of trachyte

curving to a joint in the, 142.

Picked rock, section from Jadram
Bay to Sidmouth through the, 276.

Phillips, J. A., Esq., on the so-called
‘‘oreenstones” of Western Corn-
wall, 155.

Phorus delectus, 529.

Phosphates, Dr. Dawson on the, of
the Laurentian and Cambrian rocks
of Canada, 285.

Physical geography of the Australian
seas in Cainozoic periods, 345.

geology of East Anglia during
the Glacial period, 191.

Plan of arrangement of hematite de-
posits, Millom, Cumberland, 181.

Pleuracanthus levissimus, 336.

Pleurodus affinis, 337.

Rankini, 337.

Pleurotoma consors, 527.

squamosa, 527.

Pliocene deposits south and east of
the chalk escarpment in East
Anglia, 192.

Plymouth limestone, Mr. R. N. Worth
on certain alluvial deposits associ-
ated with the, 236.

Polar ice-cap, Mr. J. J. Murphy on
the Glacial climate and the, 400.
Polden hills, trias north of the, 367;

trias of the area of the, 373.

Polyptychodon, Prof. H. G. Seeley on
an associated series of cervical and
dorsal vertebre of, from the Upper
Greensand, 433.

Pont-Newydd caves, Mr. D. Mackin-
tosh on the deposits of the, 91.

Postglacial deposits south and east of
the chalk escarpment in Hast Anglia,
195.

GENERAL INDEX.

Precipitation of clay suspended in
water, Mr. W. Ramsay on the
influence of various substances in
accelerating the, 129.

Productus, Mr. R. Etheridge, jun., on
an anherent form of, and a small
Spiriferina from the Lower Carbo-
niferous Limestone group of the
east of Scotland, 454.

Productus complectens, 462.

Prudelle, Plateau de la, curvitabular
structure in the, 146.

Pulpit Rock, section of Giant’s Kettle
in Laurentian granite in the, 78.

Quartz, 413.
Quartz-andesites of Schemnitz, analy-
sis of, 303.
Quartz-felsite of St. John’s vale, 12,
18, 22; of Crag Bridge, 14, 18.
Quartz-trachytes of Schemnitz, analy-
sis of, 303.

Quartzite, implements of, from the
Robin-Hood Cave, Creswell Crags,
250.

Rae Crags, spherulitic felsite of, 20.

Ramsay, Prof. A. C., how Anglesey
became an island, 116.

, note to Mr. W. Ramsay’s

paper on the influence of various

substances in accelerating the pre-
cipitation of clay suspended in

water, 132.

, on the physical history of the
Dee, Wales, 219.

Ramsay, William, Esq., on the in-
fluence of various substances in
accelerating the precipitation of
clay suspended in water, with a

_ note by Prof. Ramsay, 129.

Reindeer, occurrence of, in the Robin-
Hood Cave, Creswell Crags, 247,
249.

Reptile, Prof. Owen on a carnivorous,
95

Rheetic beds, Mr. Harrison on the oc-
currence of, in Leicestershire, 212.

, Section of, Spinney Hills,
near Leicester, 213.

Rhinoceros tichorhinus in the Robin-
Hood Cave, Creswell Crags, 247,
249.

Rhoechinus irregularis, 37.

Rhopalodon Wangenheimii, 355.

Rhyolites, Hungarian, 299.

of Schemnitz, analysis of, 303.

River action, Proc. 4.

Lemon, section from near

Newton Abbot to Milber Down

through, 250.

Robin-Hood Cave, Mr. Boyd Daw-
kins on the Mammalia and traces
of man found in the, 245.

, sections in the, 242, 248.

Roche Sanadoire, columnar and fissile
structures in the, 143.

Tuilliére, columns and cross
joints in, 144.

Rosehill, Castle Horneck, crystalline
slaty rock from, 166.

Rosemorran, Gulval, crystalline slaty
rock from, 166.

Rougeae, section at, 147.

Round Dale, hypersthenite of, 17.

Knott, diorite of, 16, 21.

Rowley - Regis basalt, complicated
spheroidal structure in, 151.

Royat, columns in lava at, 148.

Rudna valley, section from the Hisen-
bach valley to the Hodritsch valley
through the, 319.

Rybnik, hornblende-andesite of, analy-
sis of, 308.

St. Austell, altered dolerite from, 422.

St. Erth and Camborne, rocks of the
district between, 172.

St. Ives district, rocks of, 172.

St. John’s vale, quartz-felsite of, 12,
18.

St. Michael’s Mount, altered slates
from, 411.

St. Stephens, elvan near, 175; analy-
sis of, 176.

Saleombe Hill, section from Sidmouth
to Beer Head through, 276.

Sanadoire, Roche, columnar and fis-
sile structures in the, 143.

Sandstone, flowerpot-shaped blocks of,
at Theresa, Jefferson County, New
York, 78.

Schemnitz district, geological struc-
ture of the, 294.

, Mr. Judd on the ancient vol-
cano of the district of, 292.

Schist, mica-, 409.

, spotted, of the Lake-district, 2,
4, 5.

—, tourmaline-, 408.

Schuttrisberg, aplite forming a vein
in gneiss at, 320. ;

, section through the Kohl Berg,
passing through, 318.

Schwatzer Berg, section through the,
319.

Scorpion, Mr. H. Woodward on the
discovery of a fossil, in the British
Coal-measures, 57.

Scotland, Mr. H. Woodward on a re-
markable fossil Orthopterous insect
from the Coal-measures of, 60.

GENERAL INDEX.

Scotland, east of, Mr. R. Etheridge,
jun., on an adherent form of Pro-
ductus and a small Spiriferina from
the Lower Carboniferous Limestone
group of the, 454.

Sections from tle Alps to Brighton,
New Zealand, 54; of Giant's Kettle
in Laurentian granite in Pulpit
Rock, near Oxbow, 78; of excava-
tion made in search of stream-tin,
81; of cliff at Kessingland and
Pakefield, 124; at Rougeac, 147;
across hematite deposits, Millom,
Cumberland, 181; of hematite
vein, 182: of hematite vein in Car-
boniferous Limestone at Parkside,
183 ; of Rheetic beds, Spinney Hills,
near Leicester, 213; across the
valley of the Willow Brook, at
Leicester, 217; along the course of
the valley of the Dee, 228; from
near Newton Abbot to Milber
Down, 230; in the Robin-Hood
Cave, 242, 243; from Ladram Bay
to Sidmouth, 276; from Sidmouth
to Beer Head, 276; through the
Kohl Berg, 318 ; through the Schwat-
zer Berg, 319; from the Eisenbach
valley to the Hodritsch valley, 319 ;
of strata to show position of hone-
bed near Bradford and Clifton,
Yorkshire, 333 ; at the High Force,
467 ; of Permian and Triassic rocks
near Nottingham, 536.

Seeley, Prof. H. G., on the posterior
portion of a lower jaw of a Laby-
rinthodon (ZL. Lavisi) from the
Trias of Sidmouth, 278.

, on an associated series of cervi-

cal and dorsal vertebre of Poly-

ptychodon, from the Cambridge

Upper Greensand, 433.

, notice of the occurrence of re-

mains of a British fossil Zeuglodon

(Z. Wanklyni, Seeley) in the Barton

clay of the Hampshire coast, 429.

, on Crocodilus icenicus (Seeley),

a second and larger species of cro-

codile from the Cambridge Upper

Greensand, 437.

, on Macrurosaurus semnus (See-

ley), a long-tailed animal with pro-

ccelous vertebre from the Cambridge

Upper Greensand, 440.

, on remains of Hmys hordwellensis

(Seeley) from the Lower Hordwell

beds in the Hordwell Cliff, 445.

, on the British fossil cretaceous
birds, 496.

Seguenza, Prof., award of the Wollas-
ton Donation Fund to, Proc. 49.

Selwyn, A. R. C., Esq., award of the
Murchison Medal to, Proc. 49.

Sidmouth, Mr. Lavis, on the Triassic
strata in the cliff-sections near, 274.

, Prof. Seeley on a portion of the

lower jaw of a Labyrinthodon (L.

Lavisi) from the Trias of, 278.

, Section from Ladram Bay to, 276.

, section from, to Beer Head, 276.

Sigaretus excentricus, 519.

Silurian corals, Prof. Duncan on the
presence of unicellular algze within,
205.

, Upper, and succeeding paleo-
zoic rocks, Mr. H. Hicks on the
physical conditions under which
the, were probably deposited over
the northern hemisphere, Proe. 1o.

Silurians, Mr. Kendall on hematite in
the, 180.

Sinen Gill, granite of, 3.

Skiddaw granite, 5.

slate, altered, microscopic cha-
racters of, 18; analysis of, 22.

Slates altered by contact with granite,

08.
of the Lake-district, 2; analysis
of, 5.

Slaty rocks, crystalline, from Western
Cornwall, analyses of, 165.

Somerset, Mr. W. A. EH. Ussher on the
Triassic rocks of, and Devon, 367.

, West, Trias of, 377.

Somma, volcanic dykes of, 472.

Spheroidal structure, Rey. T. G. Bon-
ney on, 140.

in agglomerate, Burntis-
land, 150; in basalt near Murat,
150 ; in Rowley-Regis basalt, 151 ;
in an unjointed column near Le
Puy, 151.

Spherulitic felsite of Carrock Fell,
anal of, 23; and Lingy Hill, 16,
2

Spinney Hills, section of Rhetic beds
near, Leicester, 213; section across
the valley of the Willow Brook
through the, 217.

Spiriferina, Mr. R. Etheridge, Jun.,
on.an adherent form of Productus
and a small, from the Lower Carbo-
niferous Limestone group of the
east of Scotland, 454.

Spotted schist of the Lake-district, 2,

Stereodermata, Hchinoidea, 40.

Stream-tin, diagram section of excaya-
tion made in search of, 81. .

Strepsodus sanroides, 338.

Sub-Wealden boring, Sussex, Mr. H.
Woodward on some new Macrurous

GENERAL INDEX.

Crustacea from the Kimmeridge
Clay of the, 47.

Suffolk, Mr. J. Gunn on the presence
of the Forest-bed series at Kessing-
land and Pakefield in, 123.

Superficial deposits, Mr. H. B. Wood-
ward on the, of Newton Abbot,
Devonshire, 230.

Surrey, drift-beds of, 87.

Sus scrofa domesticus in the surface-
soil of the Robin-Hood Cave, Cres-
well Crags, 247.

ferus in the Robin-Hood
Cave, Creswell Crags, 247, 249.

Sussex, drift-beds of, 87.

Switzerland, rainfall of, Proc. 4.

Syenite of Hodritsch, analysis of, 308.

Syenitic granite, map showing the
distribution of Buttermere and En-
nerdale, 15.

of Buttermere, analysis of,

22; of Buttermere and Ennerdale,

14, 18.

rock of the Schemnitz area, 298.

Szaska, analysis of banatite of, 309.

Tabular structure, 145.

Tapinocephalus Atherstoni, 43.

Tasmania, Prof. Duncan on some
fossil reef-building corals from the
Tertiary deposits of, 341.

Taunton-valearea, Trias of the,373,375.

Tavistock, altered dolerites from, 421.

Teesdale, Mr. C. T. Clough on the
section at the High Force, 466.

Terebra sulcifera, 525.

Tertiary corals, Prof. Duncan on the
presence of unicellular algee in, 205.

deposits, Prof. Duncan on some
fossil reef-building corals from the,
of Tasmania, 341.

— of New Zealand, Mr. Henry
Woodward on a new fossil crab
from the, 51.

Thamnastrea sera, 348.

sp., 344.
Theresa, New York, geological phe-
nomena of the district round, 76.
Theriodonts, Prof. Owen on the evi-
dence of, in Permian deposits else-
where than in South Africa, 352.

Tides, Proc. 6.

Tolcarn, rocks from, 162, 163.

Tonalite of the Adamello group, ana-
lysis of, 309.

Tourmaline, 415.

schist, 408.

Trachyte, columns of, curving to a
joint in the Pic de Sancy, 142.

, greenstone-, of Gelnerowsky

Wrch, analysis of, 308.

Tretospondylia, 45.

Trias, Prof. Seeley ona portion of the
lower jaw of a Labyrinthodon (L.
Lavisi) from the, of Sidmouth,
278.

Triassic rocks, Mr. W. A. E. Ussher
on the, of Somerset and Devon,
367. :

strata, Mr. Johnston Lavis on

the, of the cliff-sections near Sid-

mouth, 274.

and Permian strata near Not-
tingham, 513, 533.

Triton domingensis, 522.

Tufts of the Schemnitz district, 294.

Tuilliére, Roche, columns and cross
joints in, 144.

Turbinellus edificatus, 523.

haitensis, 523.

Tylor, A., Esq., denuding agencies

and geological deposition under the

flow of ice and water, with the laws
which regulate these actions, and
the special bearing on river-action
of observations on the Mississippi
and other great rivers, and their
present and past meteorological
conditions, and similar remarks on
marine deposits, illustrated by the
Irish Sea and the Chesil Beach,
Proc. 4.

Upper Glacial deposits south and east
of the chalk escarpment in Hast
Anglia, 195.

Greensand, Prof. H. G. Seeley

on an associated series of cervical

and dorsal vertebra of Polyptycho-

don from the Cambridge, 433.

, Prof. H. G. Seeley on

Crocodilus icenicus (Seeley), a

second and larger species of Cro-

codile from the Cambridge, 437.

, Prof. H. G. Seeley on Ma-

crurosaurus semnus (Seeley), a long-

tailed animal with procclous ver-

tebre from the Cambridge, 440.

Silurian and succeeding Pale-
ozoic rocks, Mr. H. Hicks on the
physical conditions under which
the, were probably deposited over
the northern hemisphere, Proc. 10.

Ursus arctos and ferox in the Robin-
Hood Cave, Creswell Crags, 247,
248.

Ussher, W. A. E., Esq., on the trias-
sic rocks of Somerset and Devon,

367.

Val del Bove, 472.
Vesuvius, 473.

GENERAL INDEX.

Volcanic dykes, Mr. R. Mallet on the
mechanism of production of, and
on those of Monte Somma, 472.

Voleano, Mr. Judd on the ancient, of
the district of Schemnitz, Hungary,
292.

Wales, North, Mr. D. Mackintosh on
the mode of occurrence and deriva-
tion of beds of drifted coal near
Corwen, 451.

, Prof. Ramsay on the physical
history of the Dee, 219.

Ward, J. Clifton, Esq., on the grani-
tic, granitoid, and associated meta-
morphic rocks of the Lake-district,
1

Water, action of, Proc. 4.

Watertown, Jefferson County, grooves
and channellings on the Black river
at, 76.

Westmoreland, rainfall of, Proc. 4.

Weston Mouth, section from Sidmouth
to Beer Head through, 276.

White Crags, analysis of hypersthe-
nite of, 24.

Wild boar, occurrence of, in the
Robin-Hood Cave, Creswell Crags,
247, 249.

Willow Brook, at Leicester, section
across the valley of the, 217.

Wilson, E., Esq., on the Permians of
the north-east of England (at their
southern margin), and their rela-
tions to the under- and overlying
formations, 533.

Wiltshire, drift-beds of, 86.

Windy Gap, section from Ladram
Bay to Sidmouth through, 276.

Wolf, occurrence of, in the Robin-
Hood Cave, Creswell Crags, 246,
247, 248.

Wollaston Donation Fund, award of
the, to Prof. Seguenza, Proc. 49.
Medal, award of the, to Prof.

Huxley, F.R.S., Proc. 47.

Woodward, Henry, Hsq., on some
new Macrurous Crustacea from the
Kimmeridge Clay of the Sub-
Wealden boring, Sussex, and from
Boulogne-sur-Mer, 47.

, on a new fossil crab from the

Tertiary of New Zealand, collected

by Dr. Hector, Director of the

Geological Survey of New Zealand ;

with a note by Dr. Hector, 51.

, on the discovery of a fossil

scorpion in the British Coal-mea-

sures, 57.

, on a remarkable fossil Ortho-

pterous insect from the Coal-mea-

sures of Scotland, 60.

, H.B., Esq., on the gravels, sands,
and other superficial deposits in the
neighbourhood of Newton Abbot,
Devonshire, 230.

Woolborough Hill, section from near
Newton Abbot to Mglber Down
through, 230.

Worth, R. N., Esq., on certain allu-
vial deposits associated with the
Plymouth limestone, 236.

Yorkshire, section of strata to show
position of bone-bed near Bradford
and Clifton, 333.

Zeuglodon, Prof. H. G. Seeley on the
occurrence of remains of a British
fossil, (Z. Wanklyni, Seeley) in the
Barton Clay of the Hampshire
coast, 428.

Zeuglodon Wanklymi, 428.

Zygosaurus lucius, 355.

Printed by TayLoR and FRANCIS, Red Lion Court, Fleet Street.

THE

QUARTERLY JOURNAL

OF

THE GEOLOGICAL SOCIETY OF LONDON.

VoL. XXXIT.

1. On the Grantric, GRANITOID, and AssocrsTED MreramMorpuic Rocks
of the Laxe-pistrict. PartsIII—Y. By J.Cirrron Warp, Esq.,
Assoc. R.S.M., F.G.S., of the Geological Survey of England and
Wales. (Read December 1, 1875.)

[Pratss I. & II.]

Part III.—On the Skiddaw Granite and its associated Metamorphic
Rocks.

[Puiate IJ

ConTENTS,
Introduction.

I. Examination in the Field.

- Unaltered Slate.

. First Stage of Alteration : Chiastolite Slate.

. Second Stage of Alteration: Spotted Schist (Knotenschiefer).
. Third Stage of Alteration : Mica-schist.

. Granite and its Junctions with the Altered Rocks,

Ouryp cob

II. Microscopical Examination.

1. Chiastolite Slate.
2. Spotted Schist (Knotenschiefer).
3. Mica-schist.

III. Chemical Examination.
Summary.

Appendix: Notices of papers on similar rocks.

Q.5.G.S. No. 125.

2 J. CLIFTON WARD ON THE GRANITIC, GRANITOID, AND

Introduction.

The Skiddaw Granite occurs over several detached areas of small
extent in the mountain tract on the north side of Skiddaw and Blen-
cathra, much of which is known as Skiddaw Forest, the birth-place
of the river Caldew. The alteration in the surrounding Skiddaw
Slate extends in every direction for two or three miles from the cen-
tral mass, and is necessarily somewhat varied in character by the
changeable nature of the slate, which not unfrequently contains beds
of a sandy or even gritty texture. This last fact must be carefully
borne in mind when we consider the metamorphism which has been
effected.

As in the last part of this memoir, the subject shall be dealt with
under the three heads of :—I. Examination in the field. II. Micro-
scopical Examination. III. Chemical Examination.

In an Appendix short notices are given of similar researches on
these and kindred rocks.

I. Examination in the Field.

1. Unaltered Slate-—The general character of the unaltered clay-
slate is that of a bluish grey cleaved rock, seldom forming slates of
any value, but generally weathering into small flakes and chips or
into pencil-like fragments. The sandy or gritty beds are usually
uncleayed—but are very inconstant in character, and over the area in
question do not form beds of any thickness.

2. First Stage of Alteration: Chiastolite Slate.—On approaching
the Skiddaw-granite area from the south, as on walking up the valley
between Skiddaw and Blencathra, the first trace of alteration met
with is the appearance of small faint spots on the cleaved surfaces of
the slate. These increase in number and size; and many of them,
becoming more definite, assume the form of crystals of chiastolite,
which are soon found to traverse the slate in every direction, so that
very various sections of their rhombic form are displayed, frequently
showing a dark centre. The band of the chiastolite slate is seldom
more than a quarter of a mile in width.

3. Second Stage of Alteration: Spotted Schist (Knotenschiefer).—
The slate just described passes gradually into a rock of a general
schistose character. It becomes more massive and less cleaved, the
planes of cleavage being replaced by those of an imperfect foliation,
which retain, however, the same general strike and dip, though they
are frequently contorted.

The foliation seems to be due to the development of innumerable
little spots, mostly of a rectangular or oblong form, lying with their
longer axes in the planes of foliation ; and they give the rock a gene-
ral dark and spotted appearance. Large crystals of chiastolite and
andalusite * frequently traverse the mass; and in many parts minute
spangles of mica appear.

* Tt will be borne in mind that chiastolite is but a variety of andalusite, and
the latter term may sometimes be used to include both.

ASSOCIATED METAMORPHIC ROCKS OF THE LAKE-DISTRICT. a

This rock, misnamed Hornblende Slate, is frequently well jointed,

and has furnished a great many boulders which have been transported
southwards out of the tract of Skiddaw Forest. Its texture has also
favoured its use for making sets of musical stones, several of which
may be seen in Keswick.
4. Third Stage of Alteration: Mica-schist—The spotted schist,
becoming increasingly micaceous, passes over into mica-schist,
which, however, is seldom well and typically developed, except close
round the margin of the granite. It generally retains more or less
of the spotty character, the spots becoming more and more faint,
while sometimes crystals of chiastolite also occur.

The general colour of this schist is a light grey, passing into bluish
grey. ‘The foliations are wonderfully contorted, as may be especially
well seen near the granite, at the junction of Grainsgill Beck with
the river Caldew. This rock also weathers in places into large blocks,
which often lie thickly strewn upon the mountain-slopes; but in other
parts it readily crumbles away, forming a sandy wash.

5. Granite and its Junctions with the Altered Rocks.—The granite,
around which these altered rocks lie, occurs in three detached masses,
a small patch exposed in Sinen Gill (101 8.E.), a larger in the upper
course of the Caldew, and a third a little lower down the same river,
and extending northwards to the foot of Brandy Gill (101 N.E.).
There is a fourth small area of granite at the head of Brandy Gill;
but it is only poorly developed, and is much decomposed by weather-
ing. It is surrounded by other rocks than Skiddaw slates, and will
therefore be dealt with elsewhere.

The Sinen-Gill and upper Caldew granites are precisely similar in
character, consisting of white felspar and dark mica ; and the Brandy-
Gill rock differs but slightly, and that principally on the north, where
it borders on rocks which do not belong to the Skiddaw Series.

In almost every case observed the junction between the schist and
the granite is well marked. Occasionally the schist at the point of
junction is the dark and spotted rock, in which case the line of di-
vision between the two is very clear ; and in no instance is there any
difficulty in drawing a geological boundary to these three granitic
masses. Sometimes close to the granite the rock becomes a little
gneissic ; but there is no general passage from mica-schist into gneiss
and from gneiss into granite.

II. Meeroscopical Examination.

The microscopic structure of the chiastolite slate and spotted schists
has been figured in the Official Survey Memoir on 101 S.E. ; but the
accompanying illustrations in Plate I. will, I think, throw additional
light on the minute structure of these rocks. As in the preceding
section of the memoir, one figure shows the actual junction of the
granite and metamorphic rock.

1. Chiastolite Slate. (Plate I. fig. 1.)—The example figured is a
fairly typical specimen of chiastolite slate. In many parts the non-

erystalline character of the base is clearly seen, the various particles
B2

4 J. CLIFTON WARD ON THE GRANITIC, GRANITOID, AND

of which the rock is composed standing out with sufficient distinctness;
but in other parts the granular structure is more or less obscured by a
wavy and indistinct meshwork of greenish and brown matter, among
which small mica particles are pretty generally diffused.

The crystals of chiastolite are variously developed and are cut in
every direction, thus yielding many different forms (see figure),
the sections being longitudinal, transverse, or oblique to the principal
axis. They invariably contain small particles of carbon, grouped
mostly, and often exclusively, at the centre; and associated with
these and also occurring in the base, are short black bars, probably
of pyrites.

The crystal-sections have a narrow edging with a netted appear-
ance, which appears as a minute coloured mosaic in polarized light,
while under the same conditions the interior is either coloured or
dark in spaces separated from one another by minute canals. Some-
tiiues, as in the case of the crystal the pointed end of which is seen
at the bottom of the figure, when the interior appears dark under
crossed prisms, the margins and canals stand out brightly illuminated.
In ordinary light the interior has a general light-brown tint between
the canals.

In the figure it would almost seem as if the long crystal had been
broken and its parts severed by the formation of the central lozenge-
shaped one; and it is curious to observe how markedly wavy the base
is at the upper end of one of the long fragments. There also occur
a few very minute scattered prisms of a mineral which may be anda-
lusite, showing mostly red and green colours in polarized light.

2. Spotted Schist (Knotenschiefer). (Plate I. figs. 2, 3, & 4.)—
The chief points to be noted in the microscopic examination of this
rock are the great number of the rectangular spots and the crystalline
nature of the base. From fig. 2 an idea may be gathered of the
distribution of the spots, frequently much along the planes of cleavage,
which now become more or less distinct planes of foliation. Some-
times the spots are long and narrow ; but the more general form is
rectangular or oblong. When viewed with polarized light, most of
them distinctly exhibit shades of colour arranged in the form of a
cross, as shown in fig. 3; and there seems to be little doubt that the
spots are undeveloped chiastolite crystals. Particles of carbon (?)
frequently occur in connexion with them as well as in the base, as-
sociated with bars of pyrites. In some specimens the numerous
decomposing pyrite particles are very evident upon the weathered
outside as well as under the microscope. The base around these spots
is made up of very minute prisms, showing red and green colours
under polarized light, and frequently taking a flowing arrangement
(fig. 4), whilst mingled with them are numerous flakes of mica, black
bars and dark patches, and some quartz. ‘The prisms, which appear
to be rhombic and show no dichroism, are possibly andalusite. The
light-yellowish parts near the upper edges of figs. 3 and 4 are quartz,
and contain some few liquid-cavities with vacuities. Fine brown
cracks frequently traverse the mass.

3, Mica-schist. (Plate I. fig. 5, right-hand side.)—The passage from

ASSOCIATED METAMORPHIC ROCKS OF THE LAKE-DISTRICT. +)

the spotted to the mica-schist takes place by the increase of mica and
quartz and the disappearance in great part of the andalusite and
chiastolite. In the most fully developed mica-schist the whole rock
consists of quartz and two micas, brown and white; but in most
cases there are either traces left of the rectangular spots, often very
faint, or sometimes even of well-developed crystals of chiastolite or
andalusite. Occasionally, also a little hornblende appears to be pre-
sent; but this mineral is by no means a constant or characteristic
one in any of the rocks of this metamorphic series.

In fig. 5 is shown the junction of the granite and schist, the former
appearing white in the specimen, and the latter being a dark mica-
ceous schist. The junction is well defined, microscopically as well
as lithologically, the crystalline particles of mica and quartz in
the schist being small as compared with those of the same minerals
in the granite. Crystals of felspar in the schist seem also com-
paratively rare. In the figure the variously clouded patches are
mostly quartz, and some crystals of triclinic felspar occur on the
left-hand side: small plates of mica are also present.

From the above descriptions it will be seen that, the mineral horn-
blende being scarce, the old name of hornblende slate is false when
applied to the schistose rock ; and I think there can be no question
that it is an andalusite schist, probably the equivalent of the German
Knotenschiefer.

Ill. Chemical Examination.

Having noticed the appearance and bearing of these rocks in the
field, and examined their microscopic structure, we turn to their
chemical composition. The following analyses have been made for
me by Mr. John Hughes, F.C.S.; and I prefix an analysis of Welsh
slate, after Haughton, to compare with A.

|

A B. C. D.
aecili of ie yh ele 2Se| 25
= Shoo Ke BS nlo = a=
SEs Pte EREEe
Fe? slau oa ene lSch el Bea
“nn? One as 60:50 ]| 65-725 | 54-448 | 53-174 | 75-228
LSS Sh 19°70 | 14182 | 23-930 | 24-460 | 11-140
MEG cians osscdeeparessenaitensss 1-12 ENiat\. U'Ze3 1-512 1624
WER EMA 365). - cee dneeeseu se 2°20 2°342°| .2°379 | 1:946') 1-081
POMBE iia. ces atyieedeti/cizs- 3°18 3°261 3010 | 5037 | 4516
Le ee ee: eee 2-20 1-981 2086 | 2:930 | 3-996
Herrous OXIdE oc, sores. .000- 783 7306 8883 8634 Sarg
Berrie OKIE 2.2 ests cscwnees seh trace trace ‘246 | trace
Bisulphide of iron ......... we trace 1-801
‘| Phosphoric acid .........++ i trace ‘085 "128 149
Sulphuric Acid) wos .anicex 2d.) aed ‘294 | trace trace trace
CALEODIC ACIA, s0cusaned «650s ae trace trace trace trace
Loss on ignition, and water| 3°30 3°733 | 2:090 1-933 ‘500
100-060 | 100:000 | 100-000 | 100-000 |

6 J. CLIFf0ON WARD ON THE GRANITIC, GRANITOID, AND

In selecting the specimens for analysis, I have endeavoured as
far as possible to take samples which should be fairly representative
of the rocks.

The correspondence, in chemical composition, between the spotted
and mica-schist is exceedingly close, although the specimens them-
selves were very different in appearance ; while the analyses of these
rocks differ considerably from those of the chiastolite slate and the
granite. Indeed, chemically the chiastolite slate more nearly ap-
proaches the granite than do the intermediate rocks, B and ©. The
large proportion of alumina in B and C accounts for the great
development, especially in B, of the little concretionary spots, in
which the excess of alumina would appear to be secreted. Carius
gives a set of analyses of spotted schist which show a higher per-
centage of silica than do either B or C (see Appendix, p. 9).

Now there can be no doubt whatever about the fact of the chias-
tolite slate passing into the spotted schist; and it is somewhat
surprising, at first sight, to see such a manifest decrease in the per-
centage of silica and increase in that of alumina. But the analyses
of clay slate are very variable, and it is quite possible that another
sample of a similar rock from the same neighbourhood would present
a closer correspondence in chemical composition than does A to B,
more nearly resembling the Welsh slate.

The chief difference, chemically, between the metamorphic rocks
A, B, C, and the granite D, is that the latter shows a very marked
increase in the percentage of siliea and a decrease in that of alumina.
The question therefore arises, Is it possible to consider the granite
in this case the extreme term of metamorphism of these rocks?
Prof. Fuchs, in describing a similar series of metamorphic rocks
occurring in the French Pyrenees*, shows that, in the extreme
terms of the typical mica-schist, gneiss, and granite, the silica and
alkalies are increased and the alkaline earths and iron are decreased.
But at the same time he states that the increase in silica seems to
have been independent of the metamorphism, and mentions unaltered
clay slates traversed by white quartz veins. It seems, however,
very hard to make this distinction ; for the formation of quartz veins
implies metamorphic action produced by heated waters carrying
silica in solution, and this action probably effected under great
pressure. Indeed the whole series of changes, from the unaltered
clay slate to the mica-schist, is one pointing strongly to aqueous
action under great pressure—in other words, to a very moist heat.
It is quite conceivable that, under such circumstances, those rocks
most deeply buried would become most highly silicated ; and there-
fore it will not do to say, in all cases, that a rock A could not have
been converted into another, D, because of the much lower per-
centage of silica in A than D; but, at the same time, when two
rocks such as our C and D are found close to one another, it seems
highly improbable that D has been produced by metamorphism out
of the surrounding rock C.

When, therefore, all the evidence in this case is taken into

* Leonhard’s Jahrbuch, 1870, p. 717.

ASSOCIATED METAMORPHIC ROCKS OF THE LAKE-DISTRICT. ‘4

account, it seems to me most reasonable to infer that the granite of
Skiddaw, which we see exposed at the surface, was not formed by
the extreme metamorphism of the rocks immediately surrounding it.
Had this been the case, there would probably have been a more
gradual passage, a transition from mica-schist into gneiss, and from
gneiss into granite, and the chemical composition of the junction
rocks would not have differed by so much as 22 per cent. of silica.
At the same time, as just now hinted, it is by no means improbable
that the granite was formed out of rocks of a similar class, at a
somewhat greater depth; and the granite thus formed, and eating
its way upwards, would probably in great measure absorb into
itself the rocks immediately around. In connexion with this pro-
bable partially intrusive nature of the granite, the much-contorted
character of the mica-schist close around it may be noted.

A further examination of the analyses shows that the last item
mentioned in each case represents both water and carbon, thus :—

A B C D
Waele. Stee si. S45 1:698 1:298 1-216 0:397
Carbonaceous matter.. 2°035 0-792 0-717 0:103

3°733 2-090 1:933 0-500

We have already seen that microscopical examination shows the
presence of carbon in considerable quantity in connexion with the
erystals of chiastolite; and it decreases in amount through the
various terms of the series; the same seems to be the case with the
proportion of water.

It is interesting to note that the proportion of phosphoric acid
increases from A to D, just as, in Part I1., we saw it increase among
the more highly altered members of the volcanic series, accompanied
by a decrease of carbonic acid ; but the percentage of phosphoric acid
in the specimens of slate analyzed (see also Part IV.) scarcely
seems to warrant the idea that what occurs among the volcanic
rocks was solely derived from these underlying slates.

Summary.

1. The spotted rock hitherto called hornblende slate, and well
developed in the tract known as Skiddaw Forest, may be con-
veniently termed spotted (or andalusite) schist (the indefinite
Knotenschiefer of the Germans).

2. There is a complete passage in the field from unaltered clay-
slate, through chiastolite slate and spotted schist, to mica-schist.

3. The junction between the mica-schist and the granite is well
defined, and there is no general transition from mica-schist into
gneiss (proper) and from gneiss into granite.

4, Alteration of the clay slate begins by the formation of small
spots, which become developed into chiastolite and andalusite.

5. The chiastolite slate passes into spotted schist by the great
increase of the small oblong or rectangular spots arranged along

8 J. CLIFTON WARD ON THE GRANITIC, GRANITOID, AND

planes of foliation ; and mica appears, often in considerable quantity,
probably formed from some of the andalusite or rectangular spots.

6. The spotted schist passes into mica-schist, consisting of quartz
and brown and white mica, but frequently retains, to the last, faint
spots, and occasionally chiastolite crystals. The planes of foliation
in the mica-schist are very much contorted all round the granite
margin.

7. When viewed microscopically, the chiastolite slate is found to
contain in many parts small crystals, besides the large ones of
chiastolite, and the spots, which last are often plainly seen to be
but undeveloped chiastolite.

8. In the spotted schist the chief difference, microscopically, is
the increase in number of the spots and of the minute crystals in
the base, while mica begins to appear pretty plentifully (taking
the place of some of the andalusite), and there is some quartz.

9. Lastly, in the mica-schist the andalusite and the spots give
way more and more to mica and quartz.

10, Chemically, the altered rocks differ chiefly from the granite
in containing far less silica and much more alumina and ferrous
oxide. The analyses of the spotted and mica-schists are very
nearly identical; that of the chiastolite slate shows more silica and
less alumina, but is in other respects similar to the analyses of the
schists, though it contains more carbon.

11. On the whole, the evidence seems to be against regarding
the granite, now exposed, as the result of the extreme metamor-
phism of the Skiddaw slates immediately around it; but whether it
may not have resulted from the metamorphism of underlying parts
of the same series is an open question. The great contortion of the
mica-schist may be in part due to the, at any rate partially, intru-
sive character of the granite.

Appendix.
The following papers by foreign authors have been written on rocks
similar to those just described :-—

Naumann, “ Knotenschiefer,” Erlauter. z. geogn. Karte v. Sachsen, 1838, No. ii.
p- 264, and 1845, No. v. p. 50.
Kersten, ‘ Knotenschiefer,” Journ. f. prakt. Chemie, vol. xxxi. p. 108.
DvrocueEr, ‘Etudes sur le métamorphisme des roches,” Bullet. de la Soe. géol.
de France, 1846, vol. iii. p. 546.
Canrivs, ‘‘ Thonschiefer-Metamorphose bei Eichgriin,” Leonh. u. Br. Jahrb. 1856,
. 595.

Meccan “ Knotenschiefer,” Berg- u. hiittenm. Zeitung, 1858, p. 107.
JERMEEJEFF, Prm., “Ueber russische Andalusite,” Verhdl. d. k. Ges. fir die ges.
Mineral. St. Petersburg, 1864, pp. 135-147. Leonh. Jahrb. 1866, p.'724.

Kennoort, Elemente der Petrographie (1868), p. 207.
Fucus, ‘‘ Die alten Sediment-Formationen und ihre Metamorphose in den fran-
zosischen Pyrenaen,” Leonh. Jahrb. 1870, p. 717.

For microscopic descriptions of the minerals andalusite and chias-

tolite, see also :—

ZirKEL, Mikroskopische Beschaffenhcit.
Rosensuscn, Mikroskopische Physiographie.

ASSOCIATED METAMORPHIC ROCKS OF THE LAKE-DISTRICT. 9

The following are the results at which Prof. Fuchs arrives in the
memoir quoted, which deals with a set of rocks apparently very
similar to those around Skiddaw :—

While the metamorphism of the clay slate generally increases as
the granite is approached, yet do alternations occur of beds in all
stages of metamorphism.

The metamorphism begins with the appearance of little specks
(concretionary), which pass into andalusite and chiastolite; and the
rest of the rock-mass passes by degrees into an indistinct mixture of
mica and quartz with some felspar (mica-schist and gneiss).

The andalusite and concretions are still met with in the mica-schist
and gneiss, but are at last changed by pseudomorphism into mica.

The cause of these mineral changesis, first of all, molecular arrange-
ment; and afterwards an interchange of chemical constituents comes
in to help. When the clay slate is changed into mica-schist and
gneiss, the superfluous alumina goes to form the concretions and the
andalusite. The alkaline earths and iron are decreased in quantity,
but the alkalies and silica are increased.

There is a perfect petrographical passage in many places from the
gneiss into the granite. The organic matter of the clay slate gradu-
ally disappears during the process of metamorphism ; but traces of it
can be found in all stages of the alteration and even in the granite
itself.

The elaborate memoir by Durocher upon metamorphism, quoted
above, tends as follows :—

For the idea of semifusion he substitutes that of a flow of secular
heat emanating from below and from centres of plutonic action which
perhaps did not even attain to dull red heat. Two kinds of meta-
morphism are distinguished—that resulting from a movement of the
particles composing stratified beds, and that produced by the intro-
duction of a foreign body playing the part of a cement and penetrating
to the interior of the rocks, as carbon does to iron. Generally the
igneous rocks develop in the adjacent beds just those minerals of
which they are themselves composed.

Carius describes the metamorphism of clay slate around granite
masses, near Hichgrin. At a certain distance from the granite the
rock is full of small concretions (spotted schist), and becomes more
and more micaceous the nearer the granitic mass is approached. He
gives the following analyses of six specimens of these rocks :—

i. 2. 3. 4. 5. 6.
eee ee 59°385 60°028 60605 63:174 60:005 61-387
lori noon. oscar 22069 197113 24055 19°288 24104 20-803
Tron oxide” &.0-..22.<- 6816 7373 5687 4-935 6°436 6606
Manganic oxide ...... 0°273 0-141 0-280 0537 0°137 0:246
BRIG oi5 Hn 2es pseu nas 0236 1-165 0-412 0°388 0°173 0-903
Ua ee 3°608 2°186 1-781 1-599 1-872 2-105
LoL Sie see aaa 2-109 3°198 0-776 1:829 2-087 3°262
PE avian nin eine wns 3°849 3789 3648 4-193 2797 .2:966
NIE oni Pas anid nce 3471 3°993 3°305 3962 2°752 1-476

101°816 100982 100:349 99-905 100-363 99°754

10 J. CLIFTON WARD ON THE GRANITIC, GRANITOID, AND

Of these analyses, No. 1 is of unmetamorphosed slate, Nos. 2-5
contain the small concretions and mica, gradually increasing in the
number and size of its flakes, and No. 6 shows the composition of
Cornubianite, an extremely hard and crystalline metamorphic rock.
The author observes that these analyses prove the metamorphism not
to consist in an increase or loss of chemical elements, but rather ina
chemical or mechanical change among those previously existing in the
rock, and regards the cause of this change, whether the direct action
of the granite or watery influence alone, as still an open question.

Kenngott alludes to Knotenschiefer, in his ‘Elements of Petrography,’
as a rock of an uncertain character to which various names have been
given, such as Fleck-, Frucht-, or Garbenschiefer, according to the
appearance of the small spots, and which should be definitely named
when recognized minerals such as chiastolite &c. are present.

The metamorphosed Skiddaw slates have been thus described by
various observers :—

In 1820 Jonathan Otley, whose name will always be held in re-
verence in the Lake-district, first separated the Skiddaw slates as a
group from the other neighbouring formations*, and noticed the
‘mica slate,” the ‘ whintin” (=spotted schist), and the “ chiastolite
slate” developed around the granite. (See Otley’s ‘Guide to the
Lakes.’)

In 1824 and following years Prof. Sedgwick carried on his un-
tiring and thorough investigation of the district. In 1832 he thus
writes of the Skiddaw group where metamorphosed f :—*“ (1) Skiddaw
slate with interspersed crystals of chiastolite, alternating with and
passing into the preceding group (unaltered Skiddaw slates). (2) A
similar slate with numerous crystals of chiastolite, passing in the
descending order into a crystalline slate sometimes almost composed
of matted crystals of chiastolite. (3) Mica-slate spotted with chias-
tolite. (4) Quartzose and micaceous slates sometimes passing into
the character of gneiss.”

The same author, in his ‘ Letters,’ dated 1842, says “I believe
that this beautiful mineral group is nothing more than the Skiddaw
slate altered and mineralized by the long-continued action of subter-
ranean heat. The granite, though a fused rock, may not have pro-
duced the whole of this change ; butit is at least an indication of the
kind of power by which the ‘‘ metamorphic structure was brought
about.”

Otley, in the article already quoted, first drew attention to the
distinction between cleavage and stratification in connexion with
these rocks ; and Sedgwick, in 1835, gave a definition of cleavage.

Prof. Harkness, in 1858, alludes thus to the rocks in question :—
“This Skiddaw slate (on the north side of the granite), as it approx-
imates the granite of Skiddaw Forest, passes into chiastolite slate,
chiastolite rock, and a pseudo-gneiss. On the south side of the gra-

* « Remarks on the Succession of Rocks in the District of the Lakes,” Phil.
Mag. vol. lvi. p. 257 (from the Lonsdale Magazine, vol. i. p. 433).

t “On the Geological Relations of the Stratified and Unstratified groups of
rocks composing the Cumbrian Mountains,” Proc. Geol. Soc, vol. i. p. 399.

ASSOCIATED METAMORPHIC ROCKS OF THE LAKE-DISTRICT. th

nite the same phenomena occur ; but on this side hornblende rock and
actinolite rock also appear.”

In 1858 Mr. J. G. Marshall* stated his conviction, as mentioned
in Appendix to Part II., that all the Lake-district granites were
“truly metamorphic rocks,” the clay-slate and green-slate series
having ‘‘been generally subjected to the metamorphic action of heat,
pressure, and moisture.”

Prof. Harkness again writes on the Skiddaw-slate series in 18637 ;
and Prof. Phillips, in his article on the geology of the district for
Black’s Guide, first published in 1846 and revised in 1865, speaks of
the gneiss and mica-schist surrounding the granite, and adds:—
«« Above them appears a thick series of dark slaty rocks, very regu-
larly laminated and full of black spots of hornblende. This rock is
locally called ‘ whintin,’ and by geologists has been named ‘ horn-
blende slate.’”

The ‘‘ hornblende rock” and “ hornblende slate” thus alluded to
by Professors Harkness and Phillips are what, for reasons given in the
preceding pages, I have called spotted schist (German Knotenschiefer).

Prof. Nicholson also mentions “ chiastolite rock” resting ‘‘ upon
hornblendic gneiss and, in some cases, mica-schist ” +.

EXPLANATION OF PLATE I.

Fig. 1. Chiastolite slate, How Gill, Skiddaw. x 6.
2. Spotted schist (Knotenschiefer), Skiddaw Forest. x 6.
3. The same as fig. 2, viewed with polarized light. x 20.
4, The same as fig. 2, viewed with polarized light; a portion of the base
between the rectangular spots. x 50.
5. Junction of Skiddaw granite and spotted mica-schist. The granite is on
the left, and the schist on the right. 10, polarized light.

These figures have been copied from drawings in water-colour made by the
author direct from the microscope.

Part LV.—On the Quartz-Felsite, Syenitic, and associated Meta-
morphic Rocks of the Lake-district.

[Puarte IT.]

CONTENTS.
Introduction.

I. Examination in the Field.
1. St.-John’s Quartz Felsite and surrounding rocks.
2. Crag-Bridge Quartz Felsite.
8. Buttermere and Ennerdale Syenitic Granite, and surrounding rocks.
4. Carrock-Fell Rocks.
a. Spherulitic Felsite of Carrock Fell and Great Lingy Hill.
b. Diorite (?) of Miton Hill and Round Knott.
c. Hypersthenite of Mosedale Crags and Long Dale.

* « On the Geology of the Lake-district, in reference especially to the Meta-
morphic and Igneous Rocks,” Brit. Assoc. Rep. for 1858, Trans. Sect. p. 84,

+ Quart. Journ. Geol. Soe. vol. xix. p. 113,

t ‘Essay on the Geology of Cumberland and Westmoreland,’ 1868.

se J. CLIFION WARD ON THE GRANITIC, GRANITOID, AND

II. Microscopical Examination.
1. St.-John’s Quartz Felsite and altered Skiddaw Slate.
2. Crag-Bridge Quartz Felsite.
3. Buttermere and Ennerdale Syenitic Granite, and altered Skiddaw
Slate.
4. Carrock-Fell Rocks.
a. Spherulitic Felsite.
6. Diorite ? (altered dolerite).
ce. Hypersthenite (altered dolerite ?).
III. Chemical Examination.
1. Rocks of St. John’s Vale and of Buttermere.
2. Rocks of Carrock Fell.
Summary.

Appendix: Notices of papers on these rocks by other authors.

Introduction.

I now propose to describe the mode of occurrence of several masses
of quartz felsite, syenitic granite, and other rocks, together with
their metamorphic associates. Following the plan adopted in the
two previous Parts, the appearance of these rocks in the open country
will first be noticed, and be followed by the microscopical and
chemical examinations. Although the several masses are detached
from one another, it will, I think, be better to describe the rocks
together, in each section, as noted above, rather than to treat of each
mass separately in all its bearings. The same order of description
being preserved, however, in each section, it will be possible, when
desired, to take each rock separately under the various heads of
examination, and read all that relates to it independently of the
others.

I. Examination in the Feld.

1. St.-John'’s Quartz Felsite and surrounding rocks—This rock
occurs on either side of the lower end of the Vale of St. John,
forming two masses, each about one mile long from north to
south and a half to three quarters of a mile broad (Q.F. fig. 1).
It is usually of a pale colour, though sometimes assuming a red tint,
and consists mainly of quartz and felspar in a felsitic base. The
quartz is crystallized. The felspar is orthoclase and oligoclase, the
latter being much altered into a yellowish-green steatitic mineral.
Mica is sparingly developed; and a small quantity of hornblende
may perhaps occasionally be present. The rock not unfrequently
contains large fragments, much altered ; it is generally well jointed,
as at Skundraw, and in some places shows lines of what appear to
be former bedding. .

Its distribution and appearance in the mass is very striking, ex-
cept on the south. Each tract appears to be surrounded by Skiddaw
Slate, although, the ground being much dritt-covered, the slate is
only seen at a few points, and at some of these shows little sign
of alteration. The western tract is, I believe, faulted on the
south against the volcanic rocks of High Rigg; and when looked at

ASSOCIATED METAMORPHIC ROCKS OF THE LAKE-DISIRICT. 13

from Wanthwaite Bank (a part of the eastern quartz-felsite tract)
there may be noticed several well-marked features, apparently es-
carpments and dip-slopes, quite similar in strike and angle of dip to
the well-scarped beds of High Rigg. When the ground is closely
examined, Skiddaw Slate is found torun in among the quartz felsite
in two places. One of these, just west of Bridge House, though only
a short inlet (a, fig. 1), is quite in the line of the most marked

Fig. 1.—Sketch Map showing the Distribution of the Quartz Felsite of
St. John’s Vale and its Position with regard to the surrounding
fiocks. (Scale 1 inch to a mile.)

Fi1t11T Skiddaw Slate.
a Faults and veins.

transverse depression as seen from the other side of the valley. The
other is a band of slate (b) which, in a much altered condition, may
be traced from the fault near St. John’s church, through and across
the south-western corner of the felsite to Sikes, where both are
hidden under alluvium. The appearance in this case is quite that
of a bed conformable to the felsite above and below. Along the
western side of this felsite tract there is a change in what appears
like bedding-dip, to the N.N.E.

Again, when the eastern mass is looked at from the opposite or
western side, indications of bedding-planes are also seen ; and on the
slope above Wanthwaite farm a bed of ash (c) appears with felsite
above and below and apparently striking northwards, generally par-
allel to the margin of the felsitic mass. The southern margin of
this eastern tract is also partly bounded by vein-faults ; but between
it and the crags of Wanthwaite and Clough Head, formed of vol-
eanic rocks, occurs a small tract of, for the most part, highly altered
slate, containing two or three bands of ash, bounded on the east by
a probable north-and-south fault. It is very hard to distinguish

14 J. CLIFTON WARD ON THE GRANITIC, GRANITOID, AND

between some of the altered slate and some of these ash-intercala-
tions, or between these and the partially developed felsite ; some
bands of black shaly slate, however, seem to have undergone but little
alteration. Between Clough Head and White Pike, and forming the
latter, are several small masses of partially developed quartz felsite
(d, e, f,g), the slate generally passing into them all round ; and
much of the black slate is here converted into a very hard and flinty
rock of the nature of hornstone.

Altogether the disposition and relation of these quartz-felsite
masses, especially to the neighbouring volcanic rocks, rather suggest
that they occupy the place of basement beds of that series, such beds
probably presenting alternations of Skiddaw Slate with volcanic
deposits. Itwill be seen in the sequel how far such a surmise,
derived from field-evidence, is supported by the results of microscopic
and chemical research.

A third small area of quartz felsite is exposed beneath the vol-
canic rocks of Wanthwaite Crags; and its upper boundary conforms
so precisely to the strike of the overlying strata as to suggest that
it also represents the lower part of the series altered zn situ, the
character of the beds above resembling those generally occurring
near the base.

2. Crag-Bridge Quartz Felsite—A similar quartz felsite, though
occupying a much smaller tract than those of the Vale of St. John,
occurs at Crag Bridge, Rosgill, near Shap; and here again the
Skiddaw Slate is much altered immediately around the junction, and
in some places may be seen to dovetail into the felsite in the most
decided manner.

3. Buttermere and Ennerdale Syenitic Granite, and surrounding
rocks.—This large mass of syenite and syenitic granite extends
almost uninterruptedly for a distance of nine miles, from Buttermere
on the north, to Wastwater on the south, forming for more than half
the distance the boundary between the volcanic series on the south-
east, and the Skiddaw Slate on the north-west. It is generally of
a reddish tinge ; and the hornblende and mica are often sparingly dis-
tributed, though occasionally one or the other is found in greater
quantity. In the south, where detached masses occur surrounded
by the volcanic series, and in some other parts, the rock should
rather be called a felstone or sometimes a granulite, as, for example,
along the northern shores of the western half of Wastwater.

The eastern boundary, from Red Pike, Buttermere, across the
Ennerdale valley, to below Haycock, runs remarkably parallel with
the outcrop of the various beds of the voleanic series, and appears
in places to be faulted with them ; this will be seen by a reference
to the accompanying figure (fig. 2). In addition to this it should be —
noted that in several spots beneath Red Pike the syenite contains
bands showing distinct lines of bedding, or, in one case, a markedly
brecciated structure very like some of the beds of volcanic ash.

The Skiddaw Slate has experienced the same kind of alteration
in this case as around the St.-John’s quartz felsite, only that it
frequently extends for much greater distances; but in a few cases,

ASSOCIATED METAMORPHIC ROCKS OF THE LAKE-DISTRICT. 15

Fig. 2.—Sketch Map showing a part of the Boundary of the Butter-
mere and Ennerdale Syenitic Granite, and its Relation to the
neighbouring Rocks. (Scale 1 inch to a mile.)

Haycock.

S. Syenitic granite. The rest as in fig. 1.

even here, unaltered black slate occurs close to the junction. Near
Scale Force the altered slate has quite a streaky appearance outside ;
and on the summit of Red Pike, where only a small exposure occurs,
it is very hard and massive, approximating to the syeniticrock. On
either side of Ennerdale the slate is equally hard and flint-like along
the junction and for some distance from it.

The volcanic rocks in its neighbourhood are generally much
altered, being rendered very compact and flint-like; but it is often
difficult, among these rocks in connexion with the syenite, to sepa-
rate metamorphic phases from those which are probably not strictly
dependent upon it ; but in no case have I seen any thing like the series
of changes described in Part II. as occurring around the granites of
Eskdale and Shap; there are none of the purplish porphyritic rocks
so generally found.around these granites.

16 J. CLIFTON WARD ON THE GRANITIC, GRANITOID, AND

4. Carrock-Fell Rocks.—It wiil be seen by the accompanying
figure (fig. 3) that the various igneous masses of Carrock Fell and its

Fiz. 3.-—Sketch Map showing the Distribution of the Carroch Tocks.
(Scale 1 inch to a mile.)

+
A
“ VY @ ¢ exw 2€
: = a
Oy
q
Skiddar;,
ge Slatez *°"
Pad re LNG F
ro r # VotcanSer.
a ee Rs
Rg 4 cep > nes} Ca : F
Z 4 ” +. = 5 Y
fv | AQYY Great Linge setae Carroc _—
<o La Sau 2” & A Fett
as ok 70 fi
eae > Sh Hty 50
pe ae a Hy.

SU. <<
a PS OR :
ie rap\
Eo Se
ees <P. ST.
Ps T 2

Mosefale_

Cal den,

~ Alluvium. J. Spherulitic felsite. G2. Diorite. Hy. Hypersthenite.
G. Granite. &G. Bastard granite. m. Mica-schist. S/. Skiddaw Slate.

neighbourhood dovetail more or less into each other along an east-
and-west strike.

4a. Spherulitic Felsite of Carrock Fell and Great Lingy Hill.—
The colour of this rock varies from a pale reddish tint to a brownish-
grey and grey, its colour generally serving to distinguish it from
the other rocks having the same general range. There is nothing,
lithologically, to indicate the spherulitic nature of much of this
rock ; but this character, as hereafter shown, is clearly revealed by
microscopic examination. Its appearance is that of a rather coarse-
grained felsite, with scattered greenish spots, and minute, porphy-
ritically imbedded felspar crystals.

The best exposure of this rock is on the summit of Carrock Fell
and in the crags just east of the summit. In some parts it has
quite the appearance of passing into a trap of much the same nature
as many of the contemporaneous beds of the district. A good
example of this occurs on the Pike, just east of the Stone Circle.

The felsite graduates on the west and south of Carrock-Fell
summit into diorite (?) on the one hand and hypersthenite on the
other.

46. Diorite(?) of Miton Hill and Round Knott.—This rock has a
somewhat limited extension, though both felsite and hypersthenite
seem to pass over into it, occasionally, in other parts than the area
given as its particular range.

‘Lithologically it is a dark green, highly crystalline compound of
felspar and hornblende (?), becoming coarse-grained in the south,

ASSOCIATED METAMORPHIC ROCKS OF THE LAKE-DISTRICT. iy;

and passing into hypersthenite—and more compact and reddish in
the north, passing into the felsite of Rae Crags and Carrock Fell.

At Round Knott this rock contains most distinct lines of bedding,
so that several dips could be taken, the general inclination corre-
sponding to that prevailing among alternations of volcanic rocks and
Skiddaw Slate in Dry Gill to the north-west.

4c. Hypersthenite of Mosedale Crags and Lang Dale.—Lithologi-
cally, this rock varies a good deal in character. When most fully
developed it is a very coarsely crystalline mixture of hypersthene
and darkish felspar, the hypersthene crystals being sometimes an
inch or more in length. The felspar, in such a case, weathers most
easily, and the hypersthene stands out prominently, producing a
very rough-looking exterior. Occasionally, as at White Crags, the
felsparis of a pale or white colour, and the hypersthene occurs gene-
rally in smaller crystals than in the rock just described ; so that the
aspect varies from a white or grey coarse-grained crystalline rock to
a dark and more finely grained one.

At several parts along the strike of this rock, and especially a
little south of Carrock-Fell summit, the coarsely crystalline hyper-
sthenite presents a most marked appearance of bedding. Alternating
with very coarsely crystalline layers are others of very fine grain,
the aspect of the rock from a little distance quite resembling the
bedding among the volcanic series. Moreover the dip of this bed-
ding is tolerably constant along a line of strike answering to the
strike of the volcanic beds on the north, the dip being in this case
northwards at angles of from 50° to 70°. Again, at the crags just
north of Mosedale, there are most undoubted cases of beds of con-
temporaneous trap alternating for short distances with hypersthe-
nite and passing into it along the east-and-west line of strike.

It is somewhat doubtful whether or not the southern boundary
of this rock with the Skiddaw Slate is a natural or a faulted one;
most probably it is a faulted boundary, being very straight, for the
most part very definite, and corresponding with a probable fault
further to the east, on the other side of the valley. The slate,
however, is much altered along the line of junction.

All three of these rocks become somewhat changed in character
on approaching Brandy Gill, in and parallel with which there run
several strong mineral veins. At the head of the Gill is a small
area of bastard or half-developed granite. Below this it would
seem that the dioritic and hypersthenic bands approach or cross the
Gill, becoming at the same time markedly micaceous. At the foot
of the Gill is mica-schist (altered Skiddaw Slate) bordering a very
quartzo-micaceous granite.

II. Mieroscopical Examination.

The microscopic structure of the St.John’s quartz felsite and of
the Ennerdale syenitic granite has been figured in the Survey
Memoir on the Keswick district, pl. i. figs. 5 & 6. In the Plate
accompanying the present paper I haye given their structure as

Qd.G 5. No, 125. C

18 J. CLIFTON WARD ON THE GRANITIC, @RANITOID, AND

more highly magnified, in order that comparisons may be made with
that of the altered slate in their neighbourhood.

1. St.-John’s Quartz Felsite and altered Skiddaw Slate. a my
Memoir on the Keswick district I have described the microscopic
structure of the varicus stages of altered slate around the quartz
felsite, but no figures are given. The chief characteristics are the
following : —

Specimens of very decided slate, but with evident signs of altera-
tion, show the following structure under a high power:—an irre-
gular and patchy network of a yellowish-green mineral, sometimes
crystallized in fan-shaped masses, faintly dichroic and more or less
fibrous in appearance; mingled with this apparently chloritic or
semi-micaceous mineral are an infinite number of minute quartz
grains, while dark particles both of magnetite and carbonaceous
matter are diffused throughout.

In the next stage of alteration, represented in fig. 1, Pl. II., the
base has a structure which is between granular and felsitic, though
more closely allied to the latter. In this base are many crystalline
grains of quartz, much larger than any occurring in the previous
stage; and imperfectly formed felspar crystals are here and there
apparent. The yellowish-green mineral frequently occurs in larger
portions, and puts on in many cases more the appearance of mica
than in the last example. The quartz contains exceedingly minute
liquid-cavities, occasional grains of magnetite, and small portions of
the greenish mineral.

Lastly, the rock assumes its fully crystalline structure (fig. 2):
the felsitic base contains numerous crystals of felspar, while those
of quartz are developed in greater abundance and are generally of
larger size than in the previous stage, the liquid-cavities being also
larger. Mica is distinctly apparent, generally of a greenish colour.

2. Crag-Bridge Quartz Felsite—In the specimens examined the
base of this rock appears somewhat intermediate between that of
altered Skiddaw Slate and one truly felsitic, being composed largely
of quartz in small grains, and a great quantity of the yellowish-
green chloritic mineral frequently crystallized in radiate and fan-
shaped groups. Crystallized felspar seems rare, while mica is
present in long greenish flakes, and crystals of quartz are here and
there porphyritically imbedded. Besides these minerals a little
hornblende may also be present. This microscopic character of the
rock agrees with what might be inferred from its lithological appear-
ance, being a less decided quartz felsite than that of St. John’s Vale,
and more resembling some of the minor patches.

3. Buttermere and Ennerdale Syenitic Granite, and altered Skid-
daw Slate.—Figs. 3 and 4 (Pl. IL.) represent the highly magnified
microscopic structure of specimens of altered Skiddaw Slate from Red
Pike and from near Scale Force, respectively. As before, we have
the yellowish-green mineral in profusion, grains of quartz, some-
times assuming a more or less crystalline form, and particles of
magnetite, together, probably, with some carbonaceous matter. In
the example from near Scale Force the green mineral is frequently

ASSOCIATED METAMORPHIC ROCKS OF THE LAKE-DISTRICT. 19

arranged along bands, and the quartz grains are of various degrees
of fineness between them. The quartz contains great numbers of
liquid-cavities ; one of them in fig. 4 is enlarged three times the
rest of the figure; and the small contained vacuity in this, as in
many other cases, keeps up a continual active movement to all parts
of the cavity, never going out of focus in any position.

Fig. 5. represents the structure of the syenitic granite near its
border, at Scale Force. The quartz is generally not crystallized,
but interstitial and full of minute liquid-cavities; the base consists
almost wholly of felspar crystals, both plagioclase and orthoclase,
having a finely dotted appearance caused by innumerable very
minute cavities. Mica and a chloritic mineral occur, the former
scantily ; the latter, though very soft, is frequently taken for horn-
blende. Indeed, although hornblende does occur in some parts of
this rock, its presence generally is more than doubtful, especially in
an unaltered form *; and therefore the term syenite, usually applied
to this mass, or even that of syenitic granite, is scarcely an exact
one. Although varying much in its character, yet, asit is distinct
from true granite on the one hand and from true quartz felsite on
the other, some distinctive name is needed.

The somewhat doubtful green mineral is frequently accompanied
by patches of a black substance or iron-staining ; and it seems pro-
bable that this is in most cases due to a decomposition of the
mineral and peroxidation of the contained iron, the mineral thus
resembling chloropheite. At any rate it seems certain that the
green mineral occurring so abundantly in the altered slate and
almost universally in the syenitic granite, is one and the same, or
different phases of development of the same group, that of the
chlorites or hydrous micas. Generally speaking, in the crystalline
rock, this greenish mineral occurs in larger flakes than in the
altered rocks. Thus, in the latter we have quartz and a hydrous
mica in some form, and in the former, quartz, both hydrous and
anhydrous mica, and felspar. The bearing of this will be consi-
dered hereafter.

The microscopic structure of the Buttermere and Ennerdale rock
necessarily varies somewhat over so wide a spread, though the
characters given above and the figure (fig. 5, Pl. II.) contained in this
paper and that in the Survey Memoir (pl. i. fig. 6) represent the
prevailing structure. Sometimes, where the syenitic granite adjoims
the beds of the volcanic series, the transition rock is a hornblendic
felsite, collections of small hornblende crystals occurring in a fel-
sitic base.

In some parts the quartz is much diminished in quantity, and the
soft green mineral is very abundant ; this gives the rock a dark tinge.

4. Carrock-Fell Rocks.—We have already seen that the field-
examination led to the division of the Carrock-Fell rocks into three

* Prof. Phillips remarks of this rock, ‘‘ Rarely distinct hornblende is obser-
vable” (Black’s ‘Guide to the Lakes’). a
+ The ordinary micas, it will be remembered, contain small quantities of

water, but far less than the chlorite group. bs
C2

20 _ J. CLIFTON WARD ON THE GRANITIC, GRANITOID, AND

groups, separated from one another by uncertain lines. We shall
now see that the microscopic examination, made after the field-
work was completed, fully confirms this.

4a, Spherulitic Felsite (Carrock Fell, Rae Crags, Great Lingy
Hill).—Specimens taken from various parts, along the line of strike
of this rock, all show the same general microscopic structure, which
is as follows *.

In piain light the base has a felsitic appearance but with indica-
tions of a radiate structure from numerous points, while scattered
abundantly over this generally pale and hazy base are long greenish
and dichroic fibres, and irregular patches and streaks of a green and
greenish-brown mineral, which is apparently converted in many
parts into a dark brown or black product. Occasionally a few
small grains of quartz may be distinguished apart from the felsitic
base, and crystals of felspar occur porphyritically imbedded.

In polarized light, between crossed Nicols, the radiate spherulitic
structure of the base comes out admirably, a dark cross rotating
with the turning of either prism. The spherulites have invariably
a radiated and no concentric structure; and their margins are very
seldom definite, but interfere with cach other or become blended
with the general felsitic base (fig. 6, Pl. 1I.). Their average size is
about ;1, of an inch in diameter. Under crossed Nicols a good
many particles of free quartz may be detected in the base, which
exhibits the usual felsitic reaction on rotating either prism. The
felspar crystals seem to be mostly orthoclase in some specimens,
while plagioclase abounds in others. The green fibres appear
to be hornblende; but very little of the greenish mineral seems to be
in an unaltered state.

The rock, or parts of the rock, on Great Lingy Hill, are less
spherulitic and more crystalline than that of Carrock summit; pla-
gioclase crystals are numerous, and free quartz abounds, while a
highly reticulated felsitic structure appears between the crystalline
portions, apparently the remains of the spherulites.

A specimen from Rae Crags is intermediate between this last and
the Carrock-summit rock. It is more crystalline than the latter,
and more distinctly spherulitic than the former.

The example of contemporaneous trap-like rock mentioned as
occurring on Pike, Carrock Fell, has the following microscopic
structure. In ordinary light, numerous green dichroic fibres and
particles appear scattered on a hazy felsitic ground; but with a high
power (+ inch) some ghosts of small crystailine needles appear.
Between crossed Nicols the base at once assumes a semicrystalline
appearance, many small white prisms upon a felsitic ground, with
scattered particles of the altered green mineral. I could, however,
detect no trace of the development of a spherulitic structure.
Altogether I am strongly inclined to think that the microscopic
examination tends to confirm the suggestion of its contemporaneous
trap-like origin derived from its outward aspect and I therefore

* Only the general structure of each group will be here described, all detgils
being reserved for Survey publication.

ASSOCIATED METAMORPHIC ROCKS OF THE LAKE-DISTRICT. eA]

regard the rock as part of an original lava-flow, with the minutely
crystalline structure so common in that class of rock in this country.
Whether the spherulitic felsite may be a further step in the altera-
tion of somewhat similar beds, shall be considered when the other
two divisions of the Carrock rocks have been described.

4b. Diorite (?), altered dolerite (Round Knott and Miton Hill).—
This generally fine-grained, dark, crystalline rock, presents a beau-
tiful appearance under the microscope. Even specimens taken from
that part of the rock at Round Knott which shows traces of bedding,
have a fine crystalline microscopic structure, consisting of felspar,
mostly plagioclase, a hornblendic mineral, and some nearly unaltered
augite, and more or less quartz, probably of secondary formation.
But, from the manner of occurrence of the green hornblendic
mineral, and the indications of partially altered augite, one would
be inclined to suspect that the rock may be an extremely altered
one, belonging, in parts at any rate, to the doleritic class *.

Looking at its microscopic structure alone, one would but little
expect to find such a rock having well-marked lines of bedding.

4c. Hypersthenite (altered dolerite?), south of Carrock-Fell
summit, crags above and north of Mosedale Village, Lang Dale, and
Balliiway Rigg.—This rock presents the following gencral micro-
scopic characters. A coarsely crystalline compound of plagioclase
felspar and hypersthene, with scattered grains of titaniferous iron
oxide, and more or less quartz. The dark-green, coarsely crystal-
line varieties of the rock show little or no quartz; but the white or
grey varieties (lighter-coloured felspar) contain a good deal, occu-
pying spaces between the other minerals. The hypersthene may be
replaced by a hornblendic mineral or may occur in conjunction with
it. Fig. 7 (Pl. II.) represents the microscopic structure of the white
variety from White Crags (an analysis of which is given in this paper)
as viewed by polarized light. The quartz is seen in triangular and
other spaces between the banded plagioclase crystals; and part of
an altered crystal of hypersthene is shown with patches of the
black iron oxide.

Mention has been made of the apparent stratification in this rock,
immediately south of Carrock Fell and Round Knott. The coarse
bands are, microscopically, a crystalline mixture of plagioclase and
hypersthene; but the fine-grained bands, which alternate with the
former, show a very different microscopic structure. This I have
endeavoured to reproduce in fig. 8, as seen in polarized light,
between crossed Nicols. In ordinary light a great number of well-
defined grains, clear and transparent, are seen to be closely touching
one another or parted by a green dichroic mineral, and minute
particles of magnetite to be scattered plentifully throughout. Be-
tween crossed Nicols, the clear grains are seen to consist of partly
crystallized felspar and quartz, besides which there occur some
crystals of plagioclase of a larger size; and the green mineral seems
to be in part hypersthene, and in part, probably, hornblende and

* This was Mr. Allport’s opinion when he kindly glanced through these
slides.

22 J. CLIFTON WARD ON THE GRANITIC, GRANITOID, AND

altered augite; but both are a great deal changed. The divisions
which appear so frequently to traverse the small clear spaces appear
only in polarized light. This kind of erypto-crystalline structure is
akin to that described in Part II. of this Memoir as occurring in a
highly altered volcanic rock close to the granite.

These fine-grained bedded portions are quite distinct from irre-
gular segregations, some of which occur in the coarse-grained
hypersthenite lower down the same hill-side. These last are of
irregular form, often more or less circular, of a whitish tint, and
fine- ~erained. Microscopic examination shows them to consist of a
compound of well-developed felspar crystals, a green hornblendic-
looking mineral, and quartz.

Til. Chemical Examination.

1. Rocks of St. John’s Vale and of Buttermere.—The following
are analyses of carefully selected specimens of (A) altered Skiddaw
Slate from the summit of Red Pike, (B) St.-John’s quartz felsite,
(C) Buttermere syenitic granite. (See figs. 3, 2, and 5, Plate IT.)

A. B. C. - D>. ‘ HR.

Skid. Slate uartz Syenitic meek ar
of Bed | fetatte af evn of wo ae
Pike. St. John’s. |Buttermere. analyses (altered).
GR yd pe tes eis 54-480 | 67:180 | 71:442 || 60:102 | 68-421
Alumina ...... 20-720 | 16-650 | 15:340 | 17-451 7 Teeeaa
RN ae end 1:624 9-352 1-064 1:400 2-:016
Magnesia...... 1-946 1:549 720 2:144 “792
Poa oo oe es 3:203 2:914 4-439 o'252 3°338
UE ti x oe Ae 6:217 4:032 3°951 4-099 5627
Ferrous oxide..| 8188 2-151 1:107 7747 2°855
Ferric oxide .. -988 “559 1-230 494 172
Phosphoric acid. “569 "179 118 "284 204

Sulphuric acid..| trace trace trace 147

Carbonic acid ..| trace *885 trace trace trace »
Loss on ignition} 2:065 1:549 “589 2°899 °720

100-000 | 100-000 | 100-000 || 99-999 | 100-000

In the fourth column is the mean of two extreme analyses of
Skiddaw Slate, that of A and that of the How-Gill slate brought
forward in Part Ill. The result is a very near approximation to
Haughton’s analysis of Welsh roofing-slate (see page 5, Part II).
In the fifth column I have reinserted an analysis of coarse altered
ash, brought forward in Part II. of this memoir *.

The question now arises, Have we any evidence, in chemical com-
position, to enable us to decide whether the quartz felsite of St.
John’s or the syenitic granite of Buttermere has been formed deep

* Quart, Journ, Geol, Soc, vol, xxxi. p. 597.

ASSOCIATED METAMORPHIC ROCKS OF THE LAKE-DISTRICT, 3

beneath the rocks in connexion with which it now occurs, and in-
truded into its present position, or by metamorphism in situ?
If A, or even D, be contrasted with C, the difference in composition
is at once seen to be striking, especially with regard to the relative
proportions of silica and of ferrous oxide. On the other hand,
when C is contrasted with E, the general correspondence is well
marked. The same applies also to B—a considerable divergence
between it and A or D, and a marked correspondence between
it and E. Hence we may infer that, in so far as these few
analyses represent the average composition of the rocks, it is
quite possible that both the quartz felsite and the syenitic granite
may be due to the extreme metamorphism 7n situ of beds of the
volcanic series. And, further, the presence of bands of Skiddaw
slate, interstratified with volcanic ash or trap, might tend to modify
the result, some bands of slate (probably the black shaly bands) not
yielding themselves readily to the change, and therefore remaining
as such, and others, more highly silicated, such as the example from
How Gill, becoming likewise transformed into the crystalline rock,
though, perhaps, less decidedly so. We have already seen what
evidence there is in the field to support the view that these felsitic
and syenitic rocks represent or take the place of beds of the volcanic
and Skiddaw series ; and I am inclined to think that the evidence of
chemical composition goes some way towards strengthening that view.

That these masses of St. John’s and of Buttermere were probably
consolidated at a very considerable depth has already been shown
in the first part of this memoir, when dealing with the evidence of
the liquid-cayities in the quartz; and it becomes a question how far
metamorphism, carried on thus deeply, may be accompanied by an
introduction of an additional amount of silica from below.

Finally, I would by no means wish to assert that no part of the
masses in question is intrusive. With regard to that of St. John’s,
occupying only a small area, it may perhaps be so, any intrusive
appearance which’ the rock may now have being probably the result
of the subsequent faulting (see map, fig.1,p.13). But in the case of
so large an area as that of the Buttermere and Ennerdale rocks, it is
most probable that, even if we suppose the whole mass to be due to
the metamorphism of beds of the volcanic and Skiddaw series, some
measure of intrusion among and absorption of overlying beds may
have taken place over certain tracts.

The comparatively large percentage due to “loss on ignition” in
A and B represents carbon and water in the following proportions :—

AG B.

Ce a ee en oe a 1-704 ‘75
Carbonaceous matter.... ‘361 “797
2-065 1-549

2: Rocks of Carrock Fell.—I have had analyses made of typical
samples of (F) spherulitic felsite of Carrock Fell, and (G) hyper-

24 J. CLIFTON WARD ON THE GRANITIC, GRANITOID, AND

sthenite of White Crags. The figures 6 and 7 in Plate II. are from
slices of these samples.

: aH. ‘ Ga. HH,
arrock- 4

fa |e eee

summit. g

SiliGa, us eee osu ee 69:044 | 56:656 || 59-511
ATi cue chee ote ee ee 11-660: | 16°129' 7 17-460
ime; «cee ha aes 1:456 7-068 5:376
MAEM A No avast ae x. os -540 2-306 1:801
Bonds ts ac are Sway dl) 1:293 3°705
BYORI ace ce eas, see 9-673 8:296 3093
Ferrous oxide........ 1°449 6-590 4:926
Femie uxrde’ =... oc. < 2-030 *549 L271
Fisulpinde of won S..-) . . os sue “604
Phosphoric acid ...... trace 153 "115
Carbomute acid . oi... trace trace 1:569
ROUND 7c, minaay eve es 137 356 086

Loss on ienifion ...... “A4] “604 483

100-000 | 160-900 || 100-000

In the third column I have placed an analysis, H, of one of the
contemporaneous traps of the district, to contrast with G.

When I first entered upon the examination of the Carrock district
I fully expected to find a group of unmistakably intrusive rocks ;
but, as already detailed, I soon found evidence, in the strike of the
various rocks above described, in the appearance of apparent dips,
and in the existence of included portions of contemporaneous trap,
to lead me to suspect that here also might be a case of the extreme
metamorphism of rocks in situ. When we come to consider what
the rocks may be which have possibly been thus altered, we find
that the bands of felsite, diorite(?),-and hypersthenite occupy a
tract of country at the junction of the Skiddaw-slate series with the
volcanic series. The volcanic rocks immediately eastwards, upon
the other side of the valley, consist of a thick series of lavas, very.
similar to those of Eycott Hill to the south-east. The country to
the north is formed of both ash and lava, the former much altered,
and exceedingly weathered, and containing some bands of Skiddaw
Slate about Dry Gill, dipping southward, as if under the rocks of
Carrock. Hence these rocks, if they really represent any beds of
the volcanic series, may occupy the place of the lowest series of
traps and ashes, originally lying in a synclinal, and being the equi-
valents of the series immediately north and north-west, and east
and south-east. Afterwards, on looking over the literature of the
area in question, I found that Prof. Sedgwick had written thus of
the hypersthenic rock :—‘“‘ When on the spot, I considered;it only
as an instance of one of the porphyries near the base of the middle

ASSOCIATED METAMORPHIC ROCKS OF THE LAKE-DISIRICT, 25

division (green slate and porphyry) in an unusual state of crystal-
lization” (see Appendix, p. 26).

If the analysis just given be now considered, we shall see how
far this old view of Sedgwick’s is rendered probable. Unfortunately
I have not at present analyses of the lavas of Eycott Hill, which
seem to be of a generally more basic character than the one chosen
as representative of the more southern contemporaneous traps.
However, if G and H be contrasted, the likeness in composition is
sufficiently evident; and I doubt not that an analysis of the por-
phyritic dolerite (contemporaneous) of Eycott Hill and the country
to the north would correspond still more closely with G. On the
other hand, if F be compared with E (p. 22), representing an ashy
and more highly silicated member of the volcanic series, consider-
able correspondence is again found to prevail. When these facts
are taken into account, together with those previously brought
forward, it seems, I think, in the highest degree probable that these
various masses represent neither truly intrusive rocks nor old centres
or foci of volcanic eruption, but rather a series of interbedded vol-
canic rocks of somewhat varying composition which have been me-
tamorphosed im situ, the result being masses of felsite, diorite, and
hypersthenite—the first the product of the ‘most highly silicated
beds, and the latter two of the less silicated and augitic lavas. If
this be the case, the names given above to these masses must be re-
garded as correct only in a limited sense, as suggesting that these
old volcanic beds have been so metamorphosed as to take on felsitic,
dioritic, or hypersthenitic characters, these characters being by no
means persistent and universal over large areas. Owing to the
depth at which the metamorphism took place, the rocks have pro-
bably been partly silicated from below; and certainly much of the
quartz is of secondary formation.

Summary.

1. The quartz felsite of St. John’s Vale and the syenitic granite
of Buttermere and Ennerdale lie, for the most part, at the junction
of the volcanic and Skiddaw series, and seem, by their line of strike
and by the occasional presence of bands of slate or voleanic rock
enclosed within or running through them, to represent the transition
beds between the two series metamorphosed in great measure in situ.

2. Microscopic examination seems to proye the occasional passage
of certain parts of the Skiddaw Slate into quartz felsite (Pl. Il.
figs. 1 & 2).

3. The chemical examination, so far as it goes, shows the pos-
sibility of these rocks being formed by metamorphism out of the
volcanic series and included beds of Skiddaw Slate.

4. Evidence gathered in the field, and microscopic and chemical
examination, all seem to suggest that the rocks of Carrock Fell &e.
represent the base of the volcanic series, consisting largely of con-
temporaneous traps thrown into a synclinal the axis of which ranges
generally east and west, and metamorphosed into rocks of greatly
varying character. .

26 J. CLIFTON WARD ON THE GRANITIC, GRANITOID, AND

5. All the instances of metamorphism brought forward in this
part were produced at‘great depths; and the rocks may have been
considerably silicated from below.

6. Although all these various masses treated of were probably
formed in the main by the metamorphism of beds in situ, it is pro-
bable that some parts of the resulting magma became occasionally
intrusive among and absorptive of higher beds.

APPENDIX.
Notices of papers on these Rocks by other Authors.

But little more than a bare mention is made by Otley, in Bie
‘Guide,’ of the various rocks treated of in this part.

Prof. Sedgwick described them in his ‘ Letters’ (1842). He notes
the occurrence of the St.-John’s rock in two principal masses, and
remarks :—‘* When the largest mass was protruded, it bore upon its
surface an enormous fragment of Skiddaw slate, which was thus
elevated far above its natural level, mineralized by heat, and jammed
against the base of Wanthwaite crag.” (This supposed enormous
fragment represents the area occupied by the alternations of slate
and volcanic rocks at Clough Head, on the south of the eastern
quartz-felsite mass. )

After describing the distribution of the syenite of Ennerdale and
Buttermere, Prof. Sedgwick alludes to the alteration immediately
around it, remarking that in some places “the black slates are so
changed that they can hardly be distinguished from the porphyries
of the middle division” (volcanic series). Again—‘ In some places
the formations are in almost inextricable confusion, the slate rocks
in one place abutting on the syenite, in another supporting it, and
in a third resting upon it. A great mass of the Skiddaw Slate has
been caught up by the syenite, carried to the top of Red Pike, and
wedged against the green porphyries of High Stile.” Finally, he
adds :—<* In no one case, however, has this syenite in mass penetrated
the green slates or passed over them.”

The same author notices the “‘ almost endless varieties of struc-
ture’’ exhibited by the syenite of Carrock Fell, its crystals of hy-
persthene and great quantities of titaniferous oxide of iron dis-
seminated through the mass, remarking :—‘‘ when on the spot I
considered it only as an instance of one of the porphyries near the
base of the middle division (green slate and porphyry) in an unusual
state of crystallization.”

Mr. J. G. Marshall, in his paper “ On the Geology of the Lake-
district ” &c.*, includes all the syenites with the granites as meta-
morphic, being, he thinks, “‘ but altered beds of slate rock i situ.”

Prof. Harkness, in 1863, alluded to the Carrock-Fell syenitet.

Dr. Nicholson, in 1869, contributed a paper { entitled “ Notes on
certain of the Intrusive Igneous Rocks of the Lake District.” He

* Brit. Assoc. Rep. for 1858, Trans. Sect. p. 84.

+ “On the Skiddaw Slate Series,” Quart. Journ, Geol. Soc. vol. xix. p. 115.
+ Quart. Journ. Geol. Soc. vol. xxv. p. 435.

ASSOCIATED METAMORPHIC ROCKS OF THE LAKE-DISTRICT, Pa

notes the regular jointage of the quartz-felsite (called syenite) of
St. John’s, and its resemblance to bedding in the quarry nearest
Threlkeld, and describes the rock in general. The metamorphism
of the Skiddaw Slate in the Ennerdale district, around the syenite,
is contrasted with the less amount of change occurring along the
junction near Buttermere. The strike of the Carrock-Hill porphyry
is noticed as similar to that of the bedded trap. In conclusion, the
author is disposed “‘ to look upon the igneous masses in question as
the roots of the ancient vents from which were derived the alter-
nating ashes and traps which together compose almost the whole of
the green-slate series. Whether they are to be regarded as being
produced by an alteration and fusion of the Skiddaw slates in situ
is another question, and one which is at present incapable of solu-
tion. Phenomena, however, are not wanting which would appear
to favour this view.”

Prof. Phillips, in his article in Black’s ‘Guide,’ speaks of the
‘“‘ principal mass” of “the syenitic and porphyritic rocks of the
Vale of St. John” having “ been forced up amongst the brecciated
slates.” The Ennerdale and Buttermere syenitic masses, he states,
are “evidently irruptive,’ and “have been forced through, and
perhaps have spread over the middle slates” (volcanic series). The
same conclusion is also drawn in the case of the “syenites and por-
phyries round Carrock Fell and High Pike.”

EXPLANATION OF PLATE II.

Fig. 1. Altered Skiddaw Slate passing into quartz felsite, Clough Head, St.
John’s Vale. Magnified 30 times.

. Quartz felsite of St. John’s Vale. Magnified 30 times.

. Altered Skiddaw Slate, summit of Red Pike. Magnified 70 times.

. Altered Skiddaw Slate, near Scale Force, Buttermere. Magnified 70
times. The somewhat oval liquid-cavity in the large quartz space is
magnified 210 times.

. Syenitic granite, Scale Force, Buttermere. Magnified 30 times.

. Spherulitic felsite, Carrock-Fell summit. Magnified 10 times, pola-
rized light.

. Hypersthenite, White Crags, Carrock. Magnified 10 times, polarized
light.

. Fine-grained band in coarsely crystalline hypersthenite; these bands
dip regularly into the mountain-side, like ordinary bedded deposits.
Magnified 10 times, polarized light.

The above figures were drawn in water-colours, direct from the microscope,
by the author.

H> CO bo

co -F Go

Part V.—General Summary.

ConTENTS.

Leading Results of the several parts—Considerations relating to
Metamorphism.

1. Granite at various depths.

2. How far Granite may be an ultimate universal product of
Metamorphism.

3. Distribution of Metamorphism,

4, Classes of Metamorphism.

28 J. CLIFION WARD ON THE GRANITIC, GRANITOID, AND

Leading Results of the several Parts——In the four preceding parts
of this memoir I have endeavoured to bring forward all the prin-
cipal facts bearing upon the origin of the Plutonic rocks of the
district, and the degree of alteration to which the surrounding rocks
have been subjected.

In Part I. the evidence of the liquid-cavities in the quartz of the
granitic and granitoid rocks was considered, the general conelusion
being that the granites, syenitic granites, and quartz felsites were
all consolidated at very considerable depths, under great pressure,
this pressure being much greater than could be due to the thickness
of overlying rocks, and therefore exerted mainly from below and
laterally, and resulting in the work of upheaval and contortion of
the overlying strata. The period at which the principal formation
of these granitic and granitoid rocks took place was considered to
be that of the Old Red; and the work of elevation, consequent on
the great surplus of upward and lateral. pressure, was accompanied
by an enormous denudation of rocks at the surface during the greater
part of Old Red times.

In the subsequent divisions of this memoir, the mode of origin of
these various masses was discussed. In Part II. the granites of
Kskdale and Shap were dealt with, and it was shown to be at least
probable, from evidence gathered in the field, and by microscopic
and chemical examination, that these granites had been formed by
the extreme metamorphism of rocks of the volcanic series, while at
the same time the partially intrusive character of the Shap granite
was suggested by various considerations.

Part ILI. discussed the origin of the Skiddaw granite from points
of view furnished by field, microscopic, and chemical investigation.
The gradual transition from unaltered Skiddaw Slate to mica-schist
was proved under these three heads, while at the same time the
abrupt passage from the mica-schist to the granite appeared to
negative the idea of the next step, into granite, necessarily follow-
ing in this case.

In Part IV. the quartz felsite of St. John’s and the syenitic
eranite of Buttermere and Ennerdale were examined as to their
origin; and there was found to be much evidence in favour of their
representing transition beds between the volcanic series and Skiddaw
Slates, metamorphosed in situ.

The interesting rocks of Carrock Fell were then considered, and
field, microscopic, and chemical evidence were all thought to lead to
the inference that these masses of felsitic, dioritic(?), and hyper-
sthenitic rocks were due to the metamorphism in situ of the beds
forming the lower part of the volcanic series.

Such, then, are the leading results of this inquiry; and it now
only remains to point out several considerations relating to meta-
morphism to which the geological facts of this district seem to lead.

1. Granite at various depths.—Granite, it appears, may be formed
and undergo consolidation at very different depths, and uuder vary-
ing circumstances. If the agents (pressure and moist heat) acting
in the formation of granite work tolerably near the surface or

ASSOCIATED METAMORPHIC ROCKS OF THE LAKE-DISTRICT. 29

beneath some line of weakness in the overlying rocks, then the great
pressure (one of these agents) under which granitic rocks seem in-
variably to be formed, is likely to relieve itself in volcanic outbursts,
and the granite may be said to be directly the root of the volcano.
But if, on the other hand, these agents are at work at very great
depths beneath the surface, the pressure under which the granite
may be formed, although very intense in its degree, and far exceed-
ing that due to the superincumbent rocks alone, being unable to
open direct connexion with the surface, is spent in the work of
elevation and contortion of the overlying rocks. Then, if this work
of contortion and fracturing of the rocks above open out great lines
of weakness, or produce indirectly, by concomitant denudation taking
place at the surface on a large scale, a decrease in the thickness of
deposits overlying the granite, this last may then become eruptive,
and volcanic outbursts take place at the surface. Hence it would
follow, as is found to be the case, that volcanic phenomena succeed
or accompany the elevation of large tracts, and occur in connexion
with areas of depression.

It would therefore be in the highest degree unwise to suppose that
every granitic mass has been the root and origin of some past series
of voleanic phenomena; for it may represent (1) a mass so deeply
formed that, notwithstanding all the elevation and contortion pro-
duced by the intense pressure, no point of sufficient weakness was
found whereby relief might be obtained and a volcanic centre be
established, (2) a mass which, though very deeply formed, was yet
able to penetrate upwards for a certain distance along some area or
line of weakness, though its final point of consolidation was still far
below the surface, with which it for ever remained unconnected by
any voleanic neck; or (3) it may actually represent the very foun-
dation of a true volcanic neck.

The granite of most deeply-seated formation will thus be more
usually surrounded by widely metamorphosed and contorted rocks,
of which, indeed, the granite itself may generally be considered but
a part, whereas the granite directly connected, at a moderate depth,
with volcanic phenomena will bear more clearly the marks of an
intrusive character. Although all volcanic phenomena may have
so deep-seated an origin that they are invariably connected with a
granitic magma, it of course does not follow that all consolidated
granitic masses must necessarily have been at one time connected
with volcanoes. If this view be true, then, in order for the pro-
duction of the several classes of volcanic rocks, either (A) that so-
called granitic magma must vary considerably in its composition
over different areas, and at different periods over the same area, o7
(B) this magma must become variously changed on passing upwards
through and among rocks of different composition and character.
In the case, however, of a single vent erupting both basic and acidic
material at different periods, we must suppose, in the latter alter-
native (B), that the eruptions had their chief source at somewhat
different depths along the same line, where the surrounding rocks
by which the character of the magma might be influenced were

30 J. CLIFION WARD ON THE GRANITIC, GRANITOID, AND

likewise different. It is, indeed, hard to suppose that the matter
forming the root of a volcano could exist in a state of igneous fusion
without at the same time absorbing into itself portions of the sur-
rounding rocks in various degrees, thus haying its character more or
less modified.

In this connexion it is well to remember that a granitic mass
formed by metamorphism of rocks in sitw is almost sure to become
an intrusive mass at some part, even among rocks of a similar kind
to that out of which it has been formed.

2. How far Granite may be an ultimate biecivemeal product of Meta-
morphism.—It has been seen that in the Lake-district there is one
group of rocks (the volcanic) which in many of. its members (espe-
cially the ashes) approaches the granites very nearly in chemical
composition, while there is another (the Skiddaw Slate) which
generally differs more widely, though having some parts much more
highly silicated than others. Now what is true of the rocks of this
limited district is true also of rocks in general; and we might be in-
clined to say decidedly that the latter group could not be metamor-
phosed into granite, while the former might be. But metamorphic
action is varied; and although a simple melting down of clay-slate
might never produce granite, yet a moist fusion, accompanied by
elementary substnces, brought upwards from still greater depths,
might effect a great transformation. We know, indeed, that our
slate-rocks are often abundantly seamed with veins of quartz, which
mineral must have arisen from below in a state of solution, as is
evidenced by its now containing myriads of liquid cavities. Now,
just as silica has permeated these slightly altered rocks, so we may
suppose silica to permeate, probably in still greater quantities, those
rocks more highly altered at a greater depth, and the masses in a
state of aqueo-igneous fusion at a greater depth still. Hence a rock
of the composition of slate may be converted into granite by under-
going a change in the arrangement of the elements preexisting in it,
together with the addition from below of some of those elements
which were lacking in the original slate. In this sense, for ex-
ample, the Skiddaw granite might be metamorphic in its origin,
though possibly formed out of lower beds than those among which
it now lies, and with the addition from below of an additional quan-
tity of silica.

All writers on metamorphism have agreed that there are but three
ways in whieh it can be effected—(1) change in the mechanical
arrangement of the particles, (2) change in the arrangement of the
chemical elements, (3) withdrawal or addition of some of these
elements.

3. Distribution of Metamorphism.—There is, perhaps, no more
striking geological fact than the manner in which, over highly me-
tamorphosed areas, beds or tracts of rock showing little alteration
alternate with others most intensely metamorphosed. I believe
that the study of alterations effected in such a series of rocks as the
‘‘ voleanic ” in the Lake-district throws light upon this point. In
this case there are frequent alternations of beds of very different

ASSOCIATED METAMORPHIC ROCKS OF THE LAKE-DISTRICT. 31

degrees of fineness or coarseness of material, and many alternations
of various chemical composition. ‘The changes produced have been
sometimes purely mechanical—such as the compression of ashy
particles into a very solid rock, or the elongation and fresh arrange-
ment of such particles produced in the process of cleavage by intense
lateral pressure. But, besides these, other and far greater changes
have taken place, variously in different beds, by that most powerful
of all agents, water, acting in a more or less heated condition and
under pressure. The amount of alteration which has taken place
in the beds by atomic or molecular rearrangement, and the addition
or withdrawal of chemical elements, is most surprising ; and the mi-
eroscopic study of the rocks brings this fact out in a most striking
manner. No trace of the original structure is left in many a rock
which outwardly presents no special signs of intense metamorphism.
Slice after slice may be cut and examined, and yet, perhaps, not a
single fragment of the most prominent mineral be found in a wholly
unaltered condition. And if this is the case among the less out-
wardly metamorphosed beds—if water at a comparatively low tem-
perature, though acting perhaps through vast periods of time, be
competent to produce such striking internal changes, what may not
result from that far more intensely acting metamorphism, of the
same class, which has completely obliterated all signs of former ex-
ternal structure? For it is a fact well worthy of careful notice,
that this original external structure (whether of bedding or frag-
mental) very long survives the destruction of internal structure ; and,
from the mere outside appearance of many a rock, none would have
the faintest idea of the enormous amount of change in its internal
and microscopic structure.

Now, if we consider for a moment the action of metamorphism
upon a series of rocks formed of alternating beds of (1) moderately
fine-grained ash, (2) compact basaltic lava, (3) coarse brecciated
ash, (4) fine stratified volcanic dust of a general felspathic cha-
racter, (5) thin vesicular lava, (6) very fine-grained unstratified
volcanic dust, (7) volcanic breccia, we may form a fair idea of what
might be called selective metamorphism. The above-named deposits
might be thus changed into:—1, a hard flinty-looking felstone
roughly cleaved; 2, a rock with scarcely a trace of unaltered augite,
but full of chlorite, with quartz filling up spaces, and original crys-
talline structure almost or wholly obliterated; 3, a ringing, white-
weathering felstone, perfectly structureless inside, but revealing the
outline of fragments on the exterior; 4, a felstone-like banded
rock, exceedingly trap-like within, and containing imperfectly
formed crystals of felspar; 5, an amygdaloidal rock, charged through-
out with added material; 6, apparently a fine-grained trap; 7, a
massive rock, seeming to be composed of all sorts of metamorphic
rocks welded together. It is more than probable that in such a
series No. 6 would be mistaken for a bed of contemporaneous trap
(lava), of which 5 was the vesicular base, while 4 might be taken
to be a contemporaneous felstone trap. If these various beds were
of some thickness, and had an extended range, the errors supposed

4 J. CLIFTON WARD ON THE GRANITIC, GRANITOID, AND

above would be all the more likely to occur, the regularity of meta-
morphic selection along bands masking the very existence of great
metamorphism. This selective metamorphism, in whatever kinds of
rock it shows itself, will be much facilitated by the fact that very
generally the strike of the rocks corresponds with the main axis of
metamorphism, both being the result of lateral pressure acting at
right angles. It is important to remember that in the case of such
a series of beds as that just now supposed, while all may have been
subjected to the same general amount of metamorphic action, some
will show that metamorphism outwardly far more than others,
while at the same time the various beds may have been changed
in different ways according to their texture and composition.

4. Classes of Metamorphism.—In conclusion, it may be as well
just to summarize the principal conditions under which rocks occur
with regard to metamorphism.

(1) That state in which zgneous fusion is the most important or
conspicuous element. ‘This is seen in the case of lava flowing from
a volcanic vent. The presence of water, however, is distinctly
recognized ; and its sudden expansion into vapour with explosive
violence is acknowledged as the chief agent in those paroxysmal
eruptions which have sometimes resulted in the formation of crater-
lakes of great size.

(2) A state of aqueo-igneous fusion, occurring at a much greater
depth than the last, and reaching only a dull red heat as a maximum.
In this state we have every reason to suppose the granitic magma to be
which may now be solidifying at various remote spots beneath our
feet. ‘The metamorphism due to moist heat under pressure, at a
temperature not exceeding 400° C., may show itself in the complete
changing of the mineralogical and physical aspects of a rock, and in
the production of such anhydrous silicates as felspar, augite, &c.,
which when produced in the dry way require a very much higher
temperature for their formation (Daubrée*). Thus, while granite
may be solidifying at a certain spot, the rocks all around may
become granitoid and porphyritic.

(3) A state in which the rocks are permeated by water at a con-
siderably lower temperature than 400°C. The metamorphism in
this case, though not so generally conspicuous, is still highly im-
portant. Thus, the water with diminished temperature can now
decompose those very felspar crystals which at a higher temperature
it could give rise to; and minor changes innumerable may take
place throughout the rock subject to it.

Discussion.

Mr. Rurrey partly agreed with the author in his views as to
metamorphism, which seemed to him to be to a great extent a
bathymetrical question. Granite had no doubt in many instances
been formed from other recks subjected to great heat accompanied

* This author has shown that the action of water-vapour is similar to that
of liquid water.

ASSOCIATED METAMORPHIC ROCKS OF THE LAKE-DISTRICT. 33

by pressure. The nature of felsitic matter was not, in his opinion,
well understood. In many instances it is exceedingly difficult to
distinguish between mica and hornblende, hornblende and mica
plates of microscopic size being much alike under certain circum-
stances. He had never seen lines of bedding in hypersthenite as
described by the author. Quartz-porphyry and rocks of that de-
scription were merely varieties of granite. The felsitic dykes
occurring in the Upper Silurians of the Lake-district nearly all
contain magnesian mica. He thought it a curious fact that, although
the Carboniferous Limestone must at one time have overlain a great
part of the Silurian rocks of the Lake-district, there are no traces of
the typical doleritic dykes of the Carboniferous Limestone in the
latter rocks, through which, however, the dykes must in the first
instance have passed; and it would be interesting to ascertain what
was the mineral constitution of basalt dykes of Carboniferous age
where they passed through underlying sub-Carboniferous rocks.
The spots in the specimens described by the author as “ Spotted
Slate” appear to be crystals.

Prof. Hugues drew attention to a paper by Mr. Marshall on the
metamorphism of the Lake-district, which he considered correct in
principle, though in the application the author might be often wrong.
He thought that we should be cautious not to attach too great
importance to the apparent dip of metamorphic rocks to or from a
granitic boss or ridge, as the phenomena might be produced during
movements at any period subsequent to the consolidation of the
granite by the greater compressibility of the surrounding rocks as
compared with the granite, the extreme of this action producing
the protrusion of solid granite, also that the withdrawal of a large
amount of matter from below to furnish the materials of the volcanic
ejectamenta must cause extensive sinkings of the area from below
which this matter was withdrawn. He gave examples, on a small
scale, of what Mr. Ward had termed selective metamorphism, and
showed that, in the case of the Whin Sill, it seemed to affect most
the more siliceous rocks. He also referred to examples of certain
mica-trap dykes on the borders of the Lake-district, the mode of
occurrence of which suggested that they were due to the alteration
of the Silurian rock in place along lines of joints.

Mr. Jupp maintained that the relations of the igneous to the
stratified rocks of the Lake-district, which had been rendered so
clear by the author’s surveys, were perfectly consistent with a
totally different conclusion from that at which he had arrived; and
he doubted whether the chemical analyses cited in the paper could
really be regarded as bearing out the author’s views. With respect
to the application to the district of Mr. Sorby’s deductions from the
study of liquid-cavities in the crystals of igneous rocks, he argued
that (as many of the necessary data employed in the solution of the
problem were confessedly only the rudest approximations instead of
actual values) but little importance could be attached to the results
arrived at.

Mr. Hicks remarked that, in the Cambrian series, rocks which had

Q.J.G.8. No. 125. D

34 METAMORPHIC ROCKS OF THE LAKE-DISTRICT.

‘been subjected to the conditions of heat and pressure existing at a
depth of 50,000 feet are found not to be sensibly altered, whilst in
the same rocks the passage of a trap dyke caused a considerable
change in the mineral ingredients in its neighbourhood. The vol-
canic rocks described by the author were produced, Mr. Hicks
thought, in Silurian times. |

Mr. Warp replied as follows :

To Mr. Ruttley.—That the passage of the hypersthenite into the
felsite, though most decidedly gradual in some parts, was frequently
tolerably abrupt.

To Prof. Hughes.—That Mr. Marshall’s views as to metamorphism
were carried much too far, inasmuch as he considered the whole
volcanic series to represent beds of an ordinary sedimentary origin
highly metamorphosed.

To Mr. Judd.—That the author had always endeavoured to be
especially on his guard against mistaking lines of viscosity in vol-
canic lavas and other rocks for indications of former bedding. He
defended the use of chemical analyses, when care was taken to
insure the analysis of truly typical and well-selected specimens,
adding that, although the results might not go for much when taken
alone, yet they were of some value if united with other lines of re-
search on the same rocks, such as microscopic and field examination.
With reference to deductions drawn from the examination of liquid-
cavities in the quartz of the granites, the author mentioned that
Mr. Sorby had expressed considerable surprise and satisfaction at
the general agreement of the author’s results with those he had
published many years since on various granites other than those of
the Lake-country, and that a German physicist had arrived at
parallel results by an entirely different method of examination.
With regard to Mr. Judd’s suggestion, that all or much of the
Lower Old Red may once have been present in the Lake-district,
thus adding to the thickness of rocks above the granite of Skiddaw
&e., Mr. Ward remarked that he would leave the Society to judge
whether we should consider that at the close of the Upper Silurian
some 20,000 feet of strata of Old Red age were deposited and then
denuded, ov, considering there is no trace of these Lower Old Red strata
in the district, whether at the close of the Upper Silurian, when the
Skiddaw Slates were covered by some 20,000 feet of strata, that
vast denudation alone took place, which bared the rocks round
Skiddaw, and was succeeded by the deposition of only the upper-
most beds of the Old Red.

To Mr. Hicks.—That he was of opinion that it was rather too
strong a statement to make, that because certain beds buried—as
Mr. Hicks affirmed—under 50,000 feet of strata were not much
metamorphosed, therefore no granite could be formed at a less depth
than this; he believed that the Lake-district work completely con-
futed such a statement.

Quart. Journ. Geol .Soc Vol. XXXII. PL. I.

GH Ford lith. Mintern Bros. imp.
SECTIONS OF GRANITOID ROCKS OF THE LAKE - DISTRICT.

“ vy ie
* ‘ : (20 i: f el .
re ; M E oe “ "pee gta Semaine yes bihmne Few dt hla

GH Ford lith.
SECTIONS OF GRANITOID

Quart. Journ .Geol. Soc .Vol. XXXII PL I.

Mmtern Bros.imp.

ROCKS OF THE LAKE- DISTRICT.

W, KEEPING ON THE PALZXOZOIC ECHINI. 39

2. Norss on the Patmozorc Ecuint. By Water Kneprne, Esq., of
the Woodwardian Museum, Scholar of Christ’s College, Cam-

bridge. (Read June 9, 1875.)
(Communicated by Professor T. McK. Hughes.)
[Puate IIT.]

THE interest excited by the discovery, in 1869, of that remarkable
flexible Echinoderm, Asthenosoma or Calveria, in the deep Atlantic,
is increased by the fact noticed by Mr. J. Young* and other paleon-
tologists, that certain other forms, whose day ended far back in geolo-
gical time, also possessed the same overlapping plates and consequent
flexibility. This structure was first described by Prof. J. Hall in
Lepidechinus, and has since been noticed by Meek and Worthen in
another American genus, Lepidesthes, and by Mr. Young in the
scattered plates of Archewocidaris from our own Carboniferous rocks.
This character I have also detected in other genera of the same group
of Perischoechinide from Co. Wexford, Ireland.

It is worthy of note that whereas in the more modern (Creta-
ceous and Recent) group of flexible Echini (Echinothuride) the inter-
ambulacral plates imbricate from above downwards, and the ambu-
lacral from below upwards, in these Paleeozoic forms (Perischoechi-

nidz) the opposite arrangement obtains.

1. PeriscHopomus. Pl. III. figs. 1-5.

Since the first notice of the genus Perischodomus by Prof. McCoy, in
the ‘ Annals of Natural History,’ vol. i1.(1849) p. 251, this specimen,
which is one of McCoy’s two types, seems to have been lost sight of
by paleontologists ; for we find Mr. Etheridge, Jun., observing + that
‘very little is known regarding this genus,” and Meek and Worthen
sayt “only a single very imperfect specimen of it has, we be-
lieve, yet been found.”

McCoy’s diagnosis of this genus is :—‘‘ Spheroidal, depressed, sub-
pentagonal; ambulacra narrow, of two rows of small plates, most
usually of a transversely elongate pentagonal figure, and each
pierced by one pair of simple pores; interambulacra wide, of five
rows of plates very irregular in size and shape, all the plates
covered with small equal granules or secondary tubercles, while the
rows on each side adjoining the ambulacra alone bear the small
smooth primary spines, one on each, the supporting tubercle being
small, mammillated, perforated but not crenulated, surrounded by a
double ring, and situated not in the centre but near the ambulacral
edge, a little above the middle; ovarian plates pierced each with 6
foramina; mouth and anus small, both central.”

To this must now be added :—interambulacral plates overlapping

* Geol. Mag. vol. x, p. 302. t Quart. Journ. Geol. Soc. vol. xxx. p. 312.
t Pal. llinois, vol. ii. p. 295.
D2

36 W. KEEPING ON THE PALOZOIC ECHINI.

from below upwards, and from the centres of the areas outwards;
ambulacral from the dorsal side downwards.

The specimen which has disclosed this structure is a flattened,
nearly perfect, dorsal half of P. biserialis (McCoy) *.

The interambulacral plates are convex and irregular in form, as
shown in the figure given by McCoy?+; but those of the median row
are much more regular, being usually trapezoidal. The inferior over-
lapped surface of articulation in a well-exposed interambulacral plate
is seen to be about half the breadth of the exposed part of the plate
itself; it is not simply bevelled off flat, but is slightly hollowed
to receive the corresponding convex surface of the plate below
(fig. 3); on the contrary, in the lateral imbrication, the underlying
overlapped surface is slightly convex (fig. 2).

The plates of the median range overlap those on each side of them
as well as those above, like the summit tiles of a roof. Of the five
ranges of interambulacral plates, two end suddenly at about the

hird plate from the apical disk, the last one being scarcely smaller
than the rest in the range; the plates of the lateral ranges next
above are larger than usual by way of compensation (fig. 4).

The ambulacral areas are slightly sunk, their plates overlapped on
both sides by the interambulacral, and deeply interlocking with each
other. Like the interambulacral plates they are irregular, in places
club-shaped, the thick and thin ends alternating, in other places
more regular (as in McCoy’s figure); their surface is minutely
granulated as in the interambulacral plates.

Of the apical disk (fig. 4), the five genital plates are preserved in this
specimen. These are pentagonal in form, and are overlapped by the
interambulacral plates. They are perforated by numerous pores—
6 according to Prof. McCoy; but one plate has 8 distinct perfora-
tions, while another almost as obviously could not have possessed
more than 6; so that in this genus we have the same remarkable
irregularity as is observed by Messrs. Meek and Worthen in the
genus Melonites ft.

Other small triangular plates are seen, which may possibly be the
ocular plates; but I am unable to detect any perforations in them ;
moreover they are somewhat displaced, so as to be opposite the
interambulacral plates; this displacement is shared by the other
apical plates. If these are the ocular plates, it is singular that they
should overlap the genital plates. Within the circle are seen other
plates, two of which appear to be in situ; these may represent the
suranal series of the Saleniadz and Cidaride.

Two sets of spines were present in this species (P. biserialis).
Prof. McCoy described the primary one as cylindrical and smooth
‘‘as far asseen.” Only the secondary ones (fig. 5) are preserved on this
specimen; and these, under the microscope, are seen to be delicate,
needle-shaped, and coarsely ribbed. Length 3-5 mm.

Tn its general aspect, and in the complex nature of its imbrication,
Prof. Hall’s genus Lepidechinus approaches near to Perischodomus—
so much so that we find Mr. Etheridge saying, “‘ presuming for one

* Ann. & Mag. Nat. Hist. vol. iii. 1849, p. 254. T Loe. cit. p. 253.
t Pal. Illinois, vol. ii. p. 228.

W. KEEPING ON THE PALZOZOIC ECHINT. 37

moment that that genus is ultimately proved to possess imbricating
plates, the two genera Lepzdechinus and Perischodomus may perhaps
be considered identical”*. Still I think there is abundant differ-
ence between the two genera. The interambulacral plates in Prof.
Hall’s genus are far more numerous (as many as eleven ranges of
plates); and these dwindle in size towards the summit; also, on the
upper surface at least, well-developed tubercles are present on all
the ranges of interambulacral plates} instead of being confined to
the two lateral ranges.

The “imperfect specimens” mentioned by Messrs. Meek and
Worthen as “doubtless belonging to Lepidechinus”+t, would pro-
bably have been referred by them to Perischodomus, had they known
of the imbricating character of the plates in this genus. Lepides-
thes§ is distinguished from this by its numerous (10) ranges of
ambulacral plates; Archeocidaris by its large plates, cach with a
primary tubercle.

2. RHOECHINUS ||, Noy. gen.

Test rather small, spheroidal (?), interambulacral areas consisting of
five ranges of irregular plates, two of which disappear before
reaching the apical pole; plates covered with small granules; no
large tubercles. Ambulacral areas narrow, composed of two ranges
of interlocking plates, each perforated by a pair of pores. The
interambulacral plates imbricate from below upwards, and from the
central range outwards; the ambulacral from above downwards.

This genus is nearly allied to Perischodomus and Lepidechinus,
from both of which it is distinguished by the absence of primary
tubercles.

R. mREGULARIS (Keeping). Pl. III. figs. 6-8.

A small Urchin (diam. about 25 mm.), depressed at the poles.
Interambulacral plates echinated, with little sharp granules (fig. 8).
The median range is the first to disappear towards the poles, being only
seen near the equator of the test ; another range disappears at about
two plates from the summit. Ambulacral areas sunk, narrow (2-3
mm.); plates transversely elongated, pentagonal, the pores situated
about the centre of the plate. There are 4 or 5 ambulacral plates
to one interambulacral.

Loe. Hook Head, Co. Wexford, Carboniferous Limestone.

3. PALZCHINUS.

PaLecuHInus? INTERMEDIUs (Keeping). Pl. ITI. figs. 9-11.

Test spheroidal; interambulacral areas with five (?) ranges of
regular plates, outer ranges pentagonal, inner ranges hexagonal.
Surface of the plates uniformly granulated with small tubercles

* Quart. Journ. Geol. Soc. vol. xxx. p. 313.
+t Prof, Hall, Twentieth Report, State Cabinet, New York, t. ix. fig. 10.
¢ Pal. Ill. ii. p. 295. § Loc, cit. iii. p. 522. || pod, a pomegranate.

38 W. KEEPING ON THE PALAZOZOIC ECHINI.

surrounded by a smooth ring. Ambulacral areas as broad as one
of the adjacent interambulacral plates (6 mm.); composed of two
rows of plates, which are wedge-shaped, and granulated like the inter-
ambulacrals; pores near the outer edges of the plates, bigeminal.

The interambulacral plates overlap from below upwards (?)*; and
the antero-lateral overlaps the postero-lateral edge+; the ambula-
cral plates overlap in the opposite direction. As a consequence of
the alternating wedge-shaped ambulacral plates, only every other
one makes an impression on the interambulacral plates, which are
thus marked with 4—6 indentations.

Near to P. gigas (McCoy), from which it differs in the greater
relative number of ambulacral plates, each of which has only two
instead of four pores, and in the overlapping of the plates. This
latter structure is not nearly so pronounced as in Perischodomus,
and is, indeed, of a different type. In that genus the plates are
thin and overlap like scales; here they are thick and bevelled off as
a carpenter planes off his plank in joining (fig. 11); so that the test
here could only have possessed a slight degree of mobility, and would
scarcely be capable of “panting” ¢ like its modern representative
Calveria. The specific name refers to the overlapping of the plates,
which is interesting as being of a character intermediate between the
simply opposed plates of ordinary Hchini and those of Perischodomus.

Loc. Hook Head, Fethard.

Patmenines Gieas (McCoy). Pl. III. figs. 12, 13.

In this species the ambulacral plates seem to be welded in pairs.
They were described by Prof. McCoy$ as each possessing two
pairs of pores (fig. 12)—a character which M. Desor || suggests might
separate it as a genus from Palechinus; but its true nature is
clearly shown by a specimen in a part of our collection arranged by
Prof. McCoy himself, where the double character is indicated by a
pair of median interlocking articular angles to each plate (fig. 13).
This is the only example recorded of the fusion of the ambulacral

plates in the Paleozoic Echini, a feature so common among recent
Urchins.

PALZHcCHINUs sPHxXRICUS (McCoy).

A specimen of this species shows a loose ocular plate with two
pores, agreeing with those described by Mr. Baily J in P. ellipticus.

* Since both poles are absent, I am unable to determine the direction of the
imbrication ; but we may presume that it was the same as in the other Perischo-
echinide.

t This character may be taken as generic; and Messrs. Meek and Worthen
suggest that it is of “even more than generic importance” (Pall. Ill. iii. p. 522).
But since in every characteristic, except the bevelling of the plate, it agrees with
Palechinus, it is perhaps best to retain it in that genus till we obtain further
evidence.

{ The Depths of the Sea, by Prof. Wyville Thomson, p. 156.

§ Carb. Foss. Ireland, pl. xxiv. f. 4, p. 172.

|| Desor, Synopsis des Kchinides, p. 158.

§| Royal Geol. Soc. Ireland, 1864.

W. KEEPING ON THE PALZXOZOIC ECHINI. 39

4, ARCHZOCIDARIS.

ArcHzocrpaRIs Unit (Fleming). Pl. III. figs. 14-18.

A specimen found by Professor Hughes at Hafod-y-calch, Corwen,
possessing sixteen large plates but slightly displaced from their
natural position, indicates that this Urchin possessed as many as five
ranges of interambulacral plates * in each area. ‘These plates are
very thin, and imbricate from the central ranges outwards (fig. 17) as
in Perischodomus. The dorsal and ventral edges were more firmly
united by a sort of hinge (fig. 16), the edge of one fitting into a groove
in the other, as described by Mr. J. Young +, from which description
it also follows that the ambulacral plates were overlapped by the
contiguous interambulacrals.

Part of an ambulacral range is preserved on one side, showing
the plates to be very irregular and perforated by two large pores.
A very good idea of the aspect of these plates may be got by ima-
gining a couple of corallites cut from the Dudley chain coral,
Halysites catenulatus. The only interambulacral plate which is
contiguous to this series of plates is not pentagonal, but hepta-
gonal, with rounded angles (fig. 15); this was explained by the dis-
covery of a broadly wedge-shaped scale-like ambulacral plate (fig. 18),
the head of the wedge being produced and rounded to fit into the
space made by cntting off the angles of the interambulacral plates
as above described. Such an arrangement would help to keep the
plates in position during flexures of the test.

Associated with the plates and large thorny spines of this species
the small secondary spines may nearly always be found. These re-
semble those of Perischodomus in appearance; they are tubular and
coarsely ribbed, almost as coarsely as the larger spines, but possess
no thorns upon them.

Opinions have differed widely as to the value of the group
Perischoechinida in classification. By Prof. McCoy it was placed as
a distinct order of the Echinodermata, while Pictet includes its mem-
bers in the family Cidaride. But though no structure is known in
them which is not represented in the Echinoidea, they are well
distinguished from all the rest of the order by the greater number
of ranges of interambulacra] plates, and, so far as is yet known, also
by the fact that when the plates of the test overlapped, this over-
lapping was in the opposite direction to that of the flexible Echini
of the more modern group, and, again, by the greater number of
pores in the apical series of plates, a character indicating other
physiological variationst. They are equally well marked offin time—
all the Perischoechinide having disappeared at the end of the Paleozoic
period, before any of the modern types of Hchine had appeared.

Prof. Wyville Thomson’s remark in connexion with the Echino-

* A, Wortheni (Hall, Geol. Iowa, vol. ii. p. 700) has four rows of plates in
each interambulacral area.

t Geol. Mag. vol. x. p. 302.

t Double genital pores occur in the genus Cidaris (Agassiz, Rey. Echin.
p. O47):

40 W. KEEPING ON THE PALZOZOIC ECHINI.

thuride, that ‘“‘they differ to the widest extent compatible with
belonging to the same suborder,” might, it seems to me, apply with
greater force to the Perischoechinide. Such was Romer’s opinion,
who ranked them as a separate suborder. ;

I append the following classification of the Echinoidea.

Order EcHINOIDEA*,

2 rows of plates in each area. More than two rows
Suborder Ecurnipa. of plates in the
interambulacral areas.
Suborder PEriscro-
ECHINIDA (McCoy).

Plates not Plates imbrica-
imbricating ; ting; test flexible.
test rigid. Section Echino-
Section Stereo- thuride Plates not Plates
dermatat. (W. Thomson). —° imbricating. imbricating.
Section Section
Tessellata. Lepidermata.
Vent sur- Vent not
rounded by surrounded by
the apical the apical
plates. plates.
Subsection Subsection
Endocyclica Exocyclica
(Wright). (Wright).
APPENDIX.

Since writing the above I have had an opportunity of examining
cursorily some specimens in the Jermyn-Street Museum. Amongst
them is a giant Hchinus, from the Carboniferous Limestone of Derby-
shire, crushed and much disarranged. It covers an area of about 7}
by 63 inches. ‘Traces of the five ambulacra are seen converging to
what is, I believe, the mouth, of which aperture no trace is now left.

The interambulacral plates are extremely numerous, mostly hexa-
gonal, but some (the marginal rows) pentagonal; the plates have
suffered much from weathering, so that only a few of them show
that they were uniformly covered with small imperforate granules
surrounded by a faint ring. There must have been from 10 to 15
ranges of these plates in each interambulacral area.

The ambulacral areas are very broad, and are composed of nume-
rous very irregular plates, each perforated by a pair of pores. There
are at least five rows of these plates ; and certain appearances suggest
that there may have been as many as 8 or 10 ranges, as in the

Pe definition see Huxley ‘Introduction to Classification of Animals,’
p. 128.
t orepeds, stiff, fixed,

W. KEEPING ON THE PALZOZOIC ECHINI. Al

genus Lepidesthes. These ambulacral areas exactly resemble the
figures of the American Melonites and Oligoporus; the plates are
covered with granules similar to those of the interambulacral areas.
Some of the interambulacral plates are distinctly overlapping; but
in this character different plates vary very much, some of them
showing no trace of it.

The thickness of the plates is great, being about equal to half the
exposed surface in the case of the interambulacral plates.

Minute spines are profusely scattered over the whole surface of
the specimen.

This fossil is, I believe, the first example recorded in Britain of
that group of Echini, with numerous ranges of ambulacral plates,
represented in America by the genera Melonites, Oligoporus, and
Lepicdesthes.

I hope soon to find time to give this specimen a more careful
examination, and to determine its affinities.

EXPLANATION OF PLATE III.

Fig. 1. Perischodomus biserialis (M‘Coy); nat. size. The overlapped surface of
an interambulacral plate is exposed at x.

Diagram to show the method of overlapping of the inter-

ambulacral plates from the central range outwards.

Diagram to show the method of overlapping of the inter-

ambulacral plates from pole to pole.

Apical disk and portion of corona, enlarged 2 diameters.

Secondary spine, enlarged 4 diameters.

Rhoechinus trregularis (Keeping), nat. size.

Portion of fig. 6, enlarged 2 diameters.

. An interambulacral plate, enlarged.

. Palechinus(?) intermedius (Keeping), nat. size.

Interambulacral ‘plate, showing its bevelled edges, twice

SOMIM UR oo bw

=

nat. size. 2

Diagram to show the kind of overlapping of the inter-
ambulacral plates.

12. Palechinus gigas (M‘Coy). Professor M‘Coy’s figure.

Corrected figure.

14. Archeocidaris Urii (Fleming), nat. size.

15, — A heptagonal interambulacral plate of the same.

Diagram to show the method of articulation of the inter-
ambulacral plates from pole to pole.

17. —-—- ——. Diagram to show the method of articulation of the inter-
ambulacral plates from the central range outwards,

18. A peculiar ambulacral plate from fig. 14, enlarged.

Discussion.

Mr. Erurriper remarked that the genus established by the author
in this paper was of particular interest as representing an American
type, and it was of great importance in keeping up the correlation
of the geology of Europe and America. He then commented at
some length on the differences between the Paleozoic and more
modern Kchini. The type with flexible tests seemed never to have
become extinct. In conclusion he expressed his approval of Mr.
Keeping’s proposed classification.

42 W. KEEPING ON THE PALM0ZOIC ECHINI.

Prof. Duncan said that the analogies of the modern Echini with
the Paleeozoic are closer than is commonly imagined. ‘The regular
Kchini of modern times do not always preserve the same number of
plates in the ambulacral areas throughout. In modern forms the
ambulacral pores run in straight series; in the older ones they are
grouped obliquely. He remarked that some regular Echini have the
plates jammed up and broken near the buccal orifice, so that if found
separately their origin could not be determined.

Mr. Kererrrine, in replying, stated that he had received great assist-
ance from the paper by Mr. R. Etheridge, jun., in the Quarterly
Journal. The fossil referred to as in the Jermyn-Street Museum
had the ambulacral and interambulacral plates very numerous; no
such form had hitherto been recorded in Britain. He did not think
the modern flexible Echini were like the Palsozoic. As to the
presence of crowded ambulacral plates in recent Echini, this was a
character which he thought did not affect the classification now
proposed.

Quart. Journ. Ged Soc .Vol. XXXII. PLL.

ren
nay

Mintern Bros. amp.

G.H.Ford lith.

PALZOZOIC ECHINI.

PROF. OWEN ON SOME DINOSAURIAN VERTEBR&. : 43

3. On a new Montrication of DinosAuRIAN VERTEBR&. By Prof.
Owen, C.B., F.R.S., F.G.S., &c. (Read November 17, 1875.)

[Puates IV. & V.]

In Dinosauria, after the character of ‘‘ number of coalesced vertebrze
with alternate disposition of their centrums and neural arches in the
sacrum,” that of “‘ the complex development of the neural arch with
par- and diapophyses in antecedent trunk-vertebre” prevails; the
modifications of the articular surfaces of the centrums are less
constant.

Most commonly those surfaces, through the major part of the
spinal column, are more or less flattened, the deviation from flat-
ness being usually toward concavity on the hinder surface.

In this modification it is usual to find a few vertebre at the fore
part of the column with such concavity deepened to a cup and fitted
to a ball on the fore end of the next centrum: such modification is
termed “ opisthoccelian ;” but I have not seen any example of a Dino-
saur in which it prevailed throughout the dorsal region. In certain
caudal vertebre the deviation from flatness is toward concavity on
both articular surfaces of the centrum; but I have not met with
any instance in which the depth was such (as in the Ichthyosaurus
e.g.) as to merit the term “‘ amphiccelous ;” and I know of no Dinosaur
from the Lias upwards in which the dorsal vertebree are so modified.
The nearest approach is made by the vertebre of the phytophagous
Dinosaurian genera Tapinocephalus and Parciasaurus; but it is so
far from the true, or commonly understood, amphiccelian character,
that I venture to hope the Society will hold me justified in proposing
another term for such vertebral modification.

Of the dorsal series of vertebra in T'apinocephalus Atherstonii, Ow.,
the subject of Pl. IV. is a type. The centrum, 23 inches in length,
is 5 inches in breadth, and 43 inches in height. Both articular sur-
faces deviate from flatness by a feeble concavity, a little more marked
on the hinder one.

As at first wrought out of the matrix, which the petrified bone
closely resembles in colour, no deviation from this common Dinosau-
rian character presented itself; and when, persisting in chiseling
away whatever layer was doubtful, the defined central depression (fig.
1, c’) came into view, a doubt whether it was natural at first suggested
itself. Similar thoroughness in exposing the veritable bony surface
in other vertebrx, however, was attended with the same result. The
small, central, subcircular pit, from 4 to 5 (rarely 6) lines in diameter,
was repeated in each case. On the free surface of the anterior
sacral, as on that of the posterior of four anchylosed sacrals of the
Dinosaurian type, in Tapinocephalus, the foramen was relatively
larger than in the dorso-lumbar series. The neural arch of the
above fourth sacrals had been severed by fracture from an anchy-
losed fifth sacral; but the intervertebral space of the centrums had
not been obliterated.

44 PROF. OWEN ON SOME DINOSAURIAN VERTEBR2.

A vertical longitudinal section was then made of a dorsal centrum.
A fine, close, cancellous texture prevailed, more compact at the peri-
phery, with a slight enlargement of the cancelli toward the centre.
The central pit or foramen led to a cylindrical canal 6 lines long,
slightly expanding, and then rapidly contracting to a point (Pl. IY.
fig. 1 a,c’), which met the apex of the opposite cone at the middle of
the centrum. Each canal had admitted the original mud of the matrix
to near the base of the cone; but that cavity contained crystals of
spar, with matrix of a lighter colour than the rest. Some substance
in the recent carcass had resisted further entry of the mud until the
whole had hardened, and in the remaining cavities spar had slowly
crystallized.

In developing from a similar almost intractable matrix the parts
of the skeleton of another genus (Pareiasaurus, Ow.) of phytopha-
gous Dinosaur, the dorso-lumbar vertebre of a species (Par. bom-
bidens) of somewhat smaller size than Tapinocephalus Atherstonii
presented centrums proportionally longer to their breadth and height
than in the representative of the latter genus. A centrum 3
inches 9 lines in breadth, and 3 inches 4 lines in height, has a
length of 2 inches 8 lines along its lower part.

The articular surface of the centrum is subundulate, convex along a
fourth of the periphery, concave at the centre, feebly so at both parts,
and with the slight convexity continuing from the upper border to
near the centre. The appearance of any pit here was so inappreci-
able that the tentative section of the first centrum so divided was
made with little expectation of a repetition of the canaliculo-coni-
cal excavation. There it was, however (Pl. V. fig. 3, ¢’, c’); its
aperture or outlet a little wider relatively, the entering canal rather
more constricted, the terminal cone with a somewhat dilated base,
and an interval of the osseous texture, of two to four lines extent,
separating the apices. Here also, the canal was filled by matrix,
the cone by mixed discoloured matrix and spar.

Save as regards this central depression, the articular surfaces of
the Pareiasaurian centrums, like those of Tapinocephalus, would be
most correctly characterized as “‘amphiplatyan.”’ But a vertebra of
the same column of Pareiasaurus bombidens, of somewhat smaller
size (Pl. V. figs. 4-7) deepened so regularly beyond the peripheral
convexity of the articular surface of the centrum as to entitle it to
be characterized as ‘‘amphicelian.” From this character, which,
in some cases is approached by the caudal vertebrae of amphiplatyan
columns, I was at first disposed to view this vertebra as caudal,
more especially as the lower border of both terminal surfaces was
obliquely bevelled off, the anterior one (figs. 4 & 5, hy) in a great
degree, and roughened as if for ligamentous or synchondrosal
attachment of a hem- or hypapophysis.

But with the foregoing characters were associated a wide neural
canal (7). fig. 6,2), a parapophysis near the lower part of the side
of the centrum (7d. fig. 5, ), and the base of a diapophysis (d) out-
standing from the basal part of the neurapophysis (np). These cha-
racters determine the vertebra to be cervical, and perhaps the an-

PROF. OWEN ON SOME DINOSAURIAN VERTEBRZE. 45

terior one, receiving the single .occipital condyle by the somewhat
deeper anterior concavity, and affording the subtriangular rough
surface (figs. 4 & 5, hy) for the attachment of an atlanto-occipital
“‘wedge-bone:” the smaller hinder surface (fig. 5, Ay’) indicates
a second smaller “‘wedge-bone” in the neck of Pareiasaurus. After
securing five views of this vertebra, I made a longitudinal trans-
verse section of the centrum, and, with the finely cancellous texture,
exposed the pair of conical pits (fig 7, c',c'). The initial canal is
feebly marked; the excavation is chiefly conical, the cones wider
and shorter than in the succeeding vertebre, and their apices se-
parated by osseous tissue of a line in breadth.

In the above described and figured unossified tracts of the middle
of the centrum in Tapinocephalus and Parevasaurus, we have indica-
tions of a persistent trace of the primitive “ chorda dorsalis.” In
the larger Dinosaur a beaded remnant would seem to traverse some
proportion of the vertebral column, as in certain fishes*. It is a
trace of a lower grade of vertebral structure, or stage of develop-
ment of the centrums, and is interesting as being associated with a
Dinosaurian type of probably Triassic age.

It recalls the more widely perforated, still less ossified, centrums
of the vertebre of Ganocephalous reptiles of the Carboniferous
series, represented by Parabatrachus, Hylonomus, Dendrerpeton, &c.,
and thus strengthens the evidence of the post-Liassic age of the
Karroo series given in the Quart. Journ. Geol. Soc. vol. xxiii.
p: 140.

For the family of Dinosauria, represented by the genera Tapino-
cephalus and Pareiasaurus, I haye proposed the name Tretospon-
dylia.

In conclusion, I have to express my deep obligations to the late
Andrew Geddes Bain, F.G.S., and to my friend Dr. Guibon Ather-
stone, F.G.S., for the fossil materials from Blinkwater and Gats
Plaatz, South Africa, from which have been derived the above-
described structures.

EXPLANATION OF THE PLATES.

Puate LY.

Fig. 1. Anterier view of centrum, and anchylosed base of neural arch of a dorsal
vertebra, of Tapinocephalus Atherstonii, Ow.
la. Outline of section and plane of the articular surface of the centrum.
2. Side view of the same vertebra.

PLaTE V.

Fig. 3. Longitudinal and vertical section of a dorsal centrum of Pareiasaurus
serridens.
4, Anterior view of centrum, and anchylosed base of neural arch, of a cer-
vical vertebra of Parezasaurus bombidens, Ow.

. Side view of the same vertebra.

. Upper view of the same vertebra.

. Longitudinal and transverse section of the centrum of the same ver-
tebra. (The dotted lines show the lay of the surface as wrought out
prior to the section.)

(All the figures are of the natural size.)

“I Or

* Scymnus, Mugil, e. g., ‘Anat. of Vertebrates,’ 8vo, vol. i. pp. 32, 37, fig, 31.

46 PROF. OWEN ON SOME DINOSAURIAN VERTEBRA.

Discussion.

Prof. Srztey wished to ask the author whether other specimens
gave a better idea of these Dinosaurs. He thought that the Dino-
sauria and the Dicynodonts, in the structure of the pelvis and of the
fore and hind limbs, resembled each other much more closely than
any other two orders. He inquired whether this evidence did not
break down the classificational value of the Dinosauria. The aper-
tures seemed to him to be equivalent to the hollows in the vertebree
of birds, which are connected with the extension of the respiratory
system throughout the body; and thus, as they occur in Dinosaurs
and in Pterodactyles, they seem to aid in breaking down the de-
markation between these groups. He also thought that he had
observed the same structure in Dicynodonts. The specimens de-
scribed by Prof. Owen appeared to him to open up some very broad
philosophical questions.

Prof. Owen said :—‘‘The ‘Philosophical Transactions’ (1862), it
is true, show that a Dicynodont had a sacrum of several vertebra,
as in the Dinosaurs. The structure of the hind limbs of Dicynodonts
is not known ; but that of the fore limbs shows a humerus perforated
above the inner condyle, and three-jointed digits, save the pollex,
which has two joints, in their pentadactyle fore paw. Besides this
difference from Dinosaurs, the premaxillaries are confluent as one
bone, with an edentulous trenchant border, evident.y, like the cor-
responding border of the edentulous lower jaw, sheathed with a
beak-like substance. The Dicynodonts in their pair of long upper
tusks, and the Oudenodonts in their toothless jaws, further differ
from Dinosaurs; and both agree in having vertebre cupped as in
Ichthyosaurus, a character which no Dinosaur of Liassic and later age
presents. The ordinal terms Dinosauria, Anomodontia, Pterosauria,
Crocodilia, so long as they are expressive of distinctive characters of
the groups of Reptilia so associated, are useful instruments in gaining
farther knowledge of the class. To extend any one of those terms
to another group which may have a single character, added to those
of class, in common with another, is to cast away the advantage of
an ordinal term significant of actual knowledge of a sum of charac-
ters.” He added that no Dicynodont, Oudenodont, or Endothiodont
species of the order Anomodontia has hitherto been found in strata
unquestionably so recent as those that have yielded species of Dino-
sauria, other than Ketospondylian.

SS gee See el

|
.

Quart. Journ. Goel. Soc. Vol. XXXII. PL IV.

Mintern Bros imp.
7 7 S Ss 5
R.Owen del. GH Ford hth. fon VERTEBRA OF TAPINO CEPHALUS ATHERSTON

cet ee ; : j SK oe

Quart. Journ Geol. Soe Vol. XXXII Ave

r
:

&
VERTEBRE OF PARBIASAURUS BOMBIDENS. Mintern Broa, imp.

ROwen del GH Ford ith.

H. WOODWARD ON NEW KIMMERIDGE-CLAY MACRURA. 47

4. On some New Macrvrovs Crustacea from the KimmeEripes
Cray of the Sus-WEALDEN Borne, Sussex, and from BovuLoene-
suR-MER. By Henry Woopwarp, Ksq., F.R.S., F.G.S., of the
British Museum. (Read November 3, 1875.)

[Puate VI.]

Ir has always appeared to me to be a point of special interest to
geologists to record those forms found in a fossil state which have a
considerable vertical range, and yet belong to genera existing at the
present day. Among higher groups now living, we find the vertical
range exceedingly small; but when we examine the Invertebrata,
we meet with such genera as Lingula, Pentacrinus, and Limulus
having an extremely high antiquity; but the higher forms of these
types follow precisely the same general law, having a much more
restricted range in time than the lower and humbler genera.

One of the Crustacea about to be described by me belongs to a
very interesting group, the family of the Thalassinide.

Most if not all of the members of this family are fossorial in their
habits, burrowing in the sand or mud with great rapidity and remain-
ing concealed in their burrows, save their large claws, which have in
many instances been caught and devoured by fishes, or have been cut
off by the dredge, whilst the animal itself in its burrow escaped.

Of living British representatives we have four genera :—

1. Callianassa, Leach (1 species): C. subterranea, Leach.

2. Gebia, Leach (2 species): G. stellata, Leach; G'. deltura, Leach.
3. Awius, Leach (1 species): A. stirhynchus, Leach.

A, Calocaris, Bell (1 species): C. Macandree, Bell.

Formerly only one fossil species of this remarkable group of bur-
rowing Crustaceans was known, the Callianassa (Pagurus) Faujasii,
from the uppermost Chalk of Maestricht, first noticed in an admi-
rable work by Faujas-Saint-Fond entitled ‘ Histoire Naturelle de
“la Montagne de St.-Pierre, Maestricht, 1799 (p. 179, pl. 32. figs.
5 & 6), and subsequently described by Desmarest in 1822*, and by
Konig in 18237.

Another species was added in 1845, by Ottot from the Green-
sand§ of Triebitz in Bohemia and Kieslingswalde Glatz, Silesia, and
named by him Callianassa antiqua.

In 1867 Dr Fritsch of Prague added five new species from the
Chalk of Bohemia||, namely C. Turtia, C. bohemica, C. brevis, C.
elongata, and C. gracilis,

* Brongniart et Desmarest, Hist. Nat. des Crustacés, Paris, 1822, 4to, p. 127,
PL ae Fe

d t Konig, Teones Foss. Sect. (Lond. 1823) Centuria 1. p. 2, pl. 2. fig. 20.

t Otto in H. B. Geinitz’s Grundriss der Verstein. p. 21], pl. 8. figs. 12 & 18
(Dresden, 1845).

§ ‘* Greensand” = oberer Quadersandstein = U. Cretaceous.

|| Ueber die Callianassen der bohmischen Kreideformation (Prag, 1867),
tabs. 1 & 2.

48 H. WOODWARD ON NEW KIMMERIDGE-CLAY MACRURA,.

- In 1868 I called attention to two new species of Callianassa at
the Norwich Meeting of the British Association *—one, the Callia-
nassa neocomiensis, from the Greensand of Colin Glen near Belfast,
the other, C. Batei, from the Upper Marine series (Eocene), Hemp-
stead, Isle of Wight.

I have now, by the kindness of Mr. Henry Willett, F.G.S., and
the Committee of the Sub- Wealden Exploration, been favoured with
portions of cores from the boring obtained at a depth of 1057 feet,
containing, besides the fore limbs of a small macrurous Crus-
tacean, to be presently described, three or four more or less perfect
examples of a new species of Callianassa, and the oldest hitherto
known of this remarkable type of fossorial Crustaceans (Pl. VI.
figs. 1 & 2).

The fossil, which measures about 42 mm. in length, is seen in
profile on several sections of cores, and indicates the horizontality
of the beds penetrated. The parts of the fossil are all em situ, save
the carapace, which has been displaced in each case.

The first pair of feet, which are longer than the entire body, have
the penultimate joint, or manus, enormously developed (measuring
7 lines in length); the wrist is quadrate, broader than it is long;
it decreases greatly in size from its broad articulation with the
manus to its narrow proximal end, where it joins the arm (or
“meros”), which is short and strongly curved, and is united by two
very small and extremely slender joints (the “ ischium” and “ basos ””)
to the cephalothorax.

The hands of the fore limbs are more equal in size than in the
living species of Callianassa ; but in other genera of Thalassinide,
as for instance in Awius, the inequality is less marked. In the
Eocene and Cretaceous species (already referred to) the disparity in
the relative size of the hands is also somewhat diminished as com-
pared with existing species. Iam therefore not disposed to lay much
stress upon this point.

The feet of the second pair are also didactyle; and the extremities of
all the four pairs of smaller limbs are somewhat flattened and ex-
panded, to assist in digging.

The carapace and the segments of the abdomen are smooth; the
latter are somewhat quadrate in profile, not pointed ; and the shelly
covering of both is extremely thin, as in all the forms which habi-
tually conceal themselves in foreign bodies—a peculiarity developed
in the highest degree in the Paguride. The segments of the abdo-
men contract somewhat at each extremity as in other Callianasse ;
and the caudal plates are oval.

I propose to designate this new species Callianassa isochela.

2. Mercocurrvs Prytont, sp.nov. Pl. VI. fig. 3

Kimmeridge Clay, Boulogne-sur-mer, and Sub-Wealden boring.

I should have hesitated to describe the fossil macruran repre-
sented by the two small fore limbs preserved in one of the cores for-

* Brit. Assoc. Report, Norwich, 1868, p. 72, pl. ii. figs. 4 & 5.

H. WOODWARD ON NEW KIMMERIDGE-CLAY MACRURA. 49

warded to me for examination by Mr. Henry Willett, F.G.S., had
not my friend, Mr. John E. H. Peyton, F.G.S8., of St. Leonards (who
has devoted much attention to the question of the true geological
horizon of the Sub- Wealden boring) gone over to Boulogne-sur-mer
in September 1873, to examine the Kimmeridge Clay of that locality,
where he not only obtained several specimens of Lingula ovalis, Sow.
(new to that locality), but also a small but nearly perfect example of
a Crustacean referable to the genus Mecochirus.

This genus (like Caldianassa) early attracted the notice of geo-
logical writers. It was first described as a “‘ Locust” by Knorr in
1755, then as an Astacus by Bajer in 1757, as a Macrourites by
Schlotheim in 1820, and as a Palemon by Kriiger in 1823 & 1825.
It was not, however, correctly defined and separated from other genera
until 1827 (by Germar, in Keferstein’s Deutschl. Bd. iv. p. 102), since
when it has been repeatedly noticed and the species figured by Holl,
Bronn, Minster, Mayer, Pearce, Quenstedt, M‘Coy, Pictet, Oppel,
and others.

The oldest species is the Mecochirus olifex, Quenstedt, from the
Lower Lias of Dusslingen, near Tubingen, Wurttemberg*.

Next in ascending order is the Mecochirus (Cancrinium) socials,
of Meyer (1841) from the Kelloway group, Wurttemberg and Nor-
mandy.

To ails species (Mecochirus socialis, Meyer, sp.) Dr. Albert Oppel
has referred the Mecochirus (Ammonicolax) Pearcei, from the Oxford
Clay of Christian Malford, near Chippenham, Wilts, which he con-
siders a part of the Kelloway group.

I have not had the good fortune to see a specimen of Meyer’s .
socialis ; but, comparing his figures of that species with specimens of
M. Pearcet in the British Museum (from the collection of Mr.
William Cunnington, F.G.S.), and with other specimens obtained di-
rect from Mr. William Buy (late of Sutton near Chippenham), I
am satisfied the specimens from Wiltshire should still be retained as
belonging to a distinct species as established by M’Coy in 1849
(Ann. & Mag. Nat. Hist. series 2, vol. iv. p. 172).

The remaining four species of Mecochirus, namely M. longimanus,
Sehloth., WZ. Bajert, Germ., M. brevimanus, Minst., and M. dubius,
Miinst., are from the lithographic stone of Solenhofen, Bavaria.

Like other species of this remarkable genus, the fore limbs of
Mecochirus Peytoni are equal to the length of the entire body. They
measure 75 millims. in length, of which the terminal joint (dactylus)
measures 18, the propodos 30, the carpus 5, the meros 18, and the
ischium and basos 4 millims. The surface of the fore legs is very
finely punctate.

The carapace, which is finely granulated, measures 30 millims
from the rostrum to the posterior border, and 14 millims. in depth
from the mesial dorsal line to the lower margin of the branchial
region. The rostrum is somewhat produced.

The antenne are not well preserved, but are long and slender.

* Wirttemb. naturw. Jahresh. 1850, Bd. vi. p. 186; Quenstedt, der Jura,
1856, p. 89, tab. 2. fig. 17.

Q.5.G.8. No. 125. i)

50 H. WOODWARD ON NEW KIMMERIDGE-CLAY MACRURA.

The abdominal segments are 45 millims. in length and about 10 mil-
lims in depth (their epimeral borders are faleate); they are about
6 millims. in length each, and the caudal segment about 9 millims.

The smaller feet are not very clearly seen; but they measure
about 25 millims. in length.

IT believe the two fore limbs (Pl. VI. fig. 4) from the Sub- Wealden
boring, though imperfect, may be referred to the same species, .
Peytoni; and in examining one of the specimens I was delighted to
observe that it was associated (like that from Boulogne) with Lingula
ovalis, Sow.

In size this species is intermediate between Mecochirus socialis,
Meyer, sp., and M. Pearcet, M‘Coy, being about 130 millims., the
former being 60 millims. in total length (or less than half the size
of VM. Peyton), the latter being 175 millims. long, and robust in
proportion.

EXPLANATION OF PLATE VI.

. A nearly entire specimen of Callianassa isochela, sp. nov.

. Callianassa isochela, sp. nov. Another specimen, showing the caudal
series very well preserved. (Both figures twice nat. size.) From the
Kimmeridge Clay of the Sub-Wealden boring, near Battle, Sussex.

3. Mecochirus Peytoni, sp. nov. From the Kimmeridge Clay, Boulogne-

sur-mer, associated with Lingula ovalis in the cliff-section ; nat. size.

4, Hands of M. Peytoni. From the Kimmeridge Clay, Sub-Wealden boring,

near Battle, Sussex; nat. size.

Fig.

bd =

C:L.Gnesbach.

Quart.

NEW KIMMERIDGE CLAY CRUSTACEA.

Journ.Geol. Soc Vol. XXXII. P1. VI.

Mintern Bros. imp.

H. WOODWARD ON A CRAB FROM THE NEW-ZEALAND TERTIARY. 51

5. On a new Fosstt Cras from the Tertiary of New Zuatanp, collected
by Dr. Huctor, F.R.S., F.G.S., Director of the Geological Survey
of New Zealand. By Henry Woopwarp, Esq., F.R.S., F.G.S.,
of the British Museum. With a Note by Dr. Hucror. (Read
November 3, 1875.)
[Puare VII.]

Tue fine fossil Crab which I have now the pleasure to describe was
obtained by my friend Dr. Hector, F.R.S., F.G.S., from the “ passage-
beds,” Ototara series, Woodpecker Bay, Brighton, north-west coast
of the South Island, New Zealand, considered by him to be pro-
bably equivalent in position to the very lowest Eocene, or the
uppermost Cretaceous of Europe.

I have compared it with the various genera of fossil Crabs with
which I am acquainted, and am of opinion that its nearest alliance is
with the genus Harpactocarcinus of Alphonse Milne-Edwards*, from
the Nummulitic series of France, Spain, and Italy. Of this genus
Alphonse Milne-Edwards has described the following six species,
namely :—

Harpactocarcinus, Alph. Milne-Edw.

1. punctulatus, Desm. M. Eocene, Barcelona and N. Italy.

2. macrodactylus, A. M.-Edw. ,, ha “

3. rotundatus, A. M.-Edw. A - Fe

4. ovalis, A. M.-Edw. ie pe

5. Souverbier, A. M.-Edw. 3 Dax, S.W. of France.

6. quadrilobatus, Desm. = Dax; Perpignan; and Vicentin,
N. Italy.

Of the above species the specimen from New Zealand most nearly
approaches to H. quadrilobatus, Desm., in general proportions; but
the carapace of the New-Zealand crab is much more tumid, and the
relative proportions between the depth of the anterior and posterior
halves of the carapace differ considerably. Thus, if an imaginary
line be drawn transversely across each carapace from the epibranchial
spine on the lateral border of each, we shall find the proportions to
be as under :—

Depth of Depth of
anterior half. posterior half,
Harpactocarcinus quadrilobatus........ 29 46
£, HOW, SPECleS: ... ea ce 35 40

The carapace of the New-Zealand crab is 83 millims. broad and 73
millims. in depth from the rostrum to the posterior border. If
measured along the curve of the carapace, the depth is 83 millims. ;
but, owing to the curvature of the carapace this is less apparent.
The carapace is remarkably tumid, especially on the branchial

* Alph. Milne-Edwards, Hist. Crust. Podophth. fossiles, 1861-65. Family
of Canceride, pp. 196-207.
; E2

52 HH. WOUDWARD ON A CRAB FROM THE NEW-ZEALAND TERTIARY.

and gastric regions; the surface of the posterior half of the carapace
is very finely granulated ; but the anterior half is nearly smooth. The
rostrum is short and bicuspid, as in other Harpactocarcini ; the orbits
are shallow and rounded; the space between the orbits measures
15 millims. |

The hepatic margin, at first partially obscured by matrix, now
presents.a nearly smoothly rounded border, a blunt and rounded
denticle marking the lateral angles of the epibranchial border, as in
HI. quadrilobatus.

The divisions between the branchial, cardiac, and gastric regions
are faintly indicated by a shallow undulation of the surface of the
carapace, and a slightly roughened and incised line of short oblique
markings on the sides of the gastric intumescence.

The external jaw-feet, or maxillipeds, are well preserved*. The
endopodite is broad and straight-sided, and divided by a suture near
its anterior third ; the surface is marked by a longitudinal furrow ;
the exopodite is straight and narrow: both take their rise from a
common basal joint.

The chele, or first pair of true limbs, are robust and, as is com-
monly the case among the Canceride, very unequal in size, the right
being considerably larger than the left hand.

The ischium measures 8 millims. in length, the arm 31 millims. ;
these, with the short basal joint, are concealed beneath the carapace ;
the carpus is a short strong joint 26 millims. long, and having its distal
end broadest (measuring 30 millims.) and armed at each angle with
a stout short spine. Its proximal end contracts greatly at its articu-
lation with the arm, giving it a triangular form. The length of the
larger hand is 65 millims., breadth 30; length of the smaller hand
53 millims., breadth 21. The chelate ends of the limbs are consider-
ably recurved, as in other Canceride.

The abdomen has nearly all flaked off; but a sufficient portion
remains attached to the plastrosternum to show that the fossil crab
was a female. In general form the plates of the plastrosternum
agree with those of Harpactocarcinus. The four.pairs of simple mono-
dactylous fect are broken off short, but they present no characters
dissimilar to those of other Cancerines.

I propose to name this interesting fossil Harpactocarcinus tumidus.

The presence of this Crab alone would sufficiently attest the near-
shore character of this deposit at Woodpecker Bay; but, besides
another specimen of Harpactocarcinus tumidus obtained from this
deposit, a number of portions of chelz of other Cancerine claws, with
remains of an extinct gigantic Penguin, attest the correctness of this
opinion.

For the present, however, I will refrain from generalizing on the
exact horizon of this fossil from New Zealand as compared with the
beds containing Harpactocarcint in Europe; but Dr. Hector has
added a few remarks to my paper ; and as he has visited the locality

* The entire underside has been carefully worked out by Mr. Barlow, the

intelligent and able mason attached to the geological department of the British
Museum.

DR. HECTOR ON THE GEOLOGY OF NEW ZEALAND. 53

and also obtained this and many other fossils therefrom, and worked
out the geology of the district, I leave him to notice the geological
facts of the case.

Note by Dr. Hxcror.

The locality where both the specimens of this fossil were obtained
is a small bay on the exposed west coast of the South Island, where
the shelter of the Seal-rocks affords a precarious landing-place for
supplying stores to the gold-miners around Brighton. Northwards
the coast is formed of granite for many miles; but the crystalline
rocks at this place disappear under Upper Mesozoic and Lower
Tertiary formations, which form the coast range, and;expand south-
wards to a width of 15 miles, covering the important area of the
Grey-River Coal-field.

The Seal-rocks belong to the upper part of the formations re-
ferred to, representing on the west coast the Ototara group of the
north-east of Otago Province, a formation which is widely distributed
throughout New Zealand, and nowhere, so far as I have seen, passes
into the Tertiary strata of later date.

The Ototara group was referred to by Profs. E. Forbes and Ru-
pert Jones*, who from fossils collected by Mr. Walter Mantell,
considered it to possess mixed characters of the Upper Cretaceous
and Eocene periods of Europe.

In the same formation Prof. Huxley, from a single bone brought
also by Mr. Mantell, determined the former existence of a gigantic
Penguin (Paleeudyptes antarcticust); and it is worthy of remark
that not only have further remains of the skeleton of that bird been
obtained in the Calcareous Sandstone at Ototara, but that a large
part of a skeleton was extracted from the same stratum in the Seal-
rocks with the first-found specimen of the Crustacean under notice,
thus satisfactorily correlating the formations on the opposite sides of
the islandt. Concerning the paleontological value of these Penguin
remains Prof. Huxley makes the following important observations :—

“Whatever be the precise age of the fossil, it is not a little re-
markable to find in strata of such antiquity the remains of a bird
the whole of whose congeners are at present absolutely confined to
the southern hemisphere, and therefore, in a broad sense, to the
same great distributional area. If the strata be of Pliocene age,
the fact is in accordance with the relations which have been observed
between the recent and Pliocene faune of the Northern Hemisphere.
On the other hand, the little that is at present known respecting the
distribution of Birds in time is not inconsistent with the ascription
of a far higher antiquity to a genus as closely allied as Paleweudyptes
to those which now exist” (J. ¢. p. 675).

The Ototara formation has, since the above was written, been ex-
amined in most distant parts of New Zealand; and though it varies
in mineral aspect, it always maintains, at least in its upper part, the

* Quart. Journ. Geol. Soc. vol. vi. p.329, 1850. — t+ Lbid. vol. xv. p. 672, 1859.
{ Hector, Trans. New-Zeal. Inst. vol. iv. p. 341.

DR. HECTOR ON THE GEOLOGY OF NEW ZEALAND.

54

‘UVIMOA, puL snodoeyruoqaey ATquqoad ore “ATX PUe “TITX x

‘SUIBITINOUT poyepost ATTe19UIED ynq PAISSVUT UT PUBL _) “A ‘yeoo youd UMOIT 1
‘sty ot} JO eseq 4soM OY} SUOTL SYOOT ESOySTIPS “7 ‘ow ‘soyLUOUIULY
“ory ‘souog URLAINeY ‘spURsMOdIS PUB OUOJspULS OFLUOMMIT “Y pe “TITA
SOPIGOTIAT, YA Soqiz}renb pure souo0zsouryy s10deapepy “2 Tee » ATX ‘sXvyo Ysno} yarrq. “2
- ‘sd py oy} Jo souoyspurs O10z0% ed teddy ‘s ***yTITX TIX "sqlId puesucedly “Y
‘gayv[S PUB SeTIOJS “sopoauoinayd wagog—spreut vpory “O°
~pues snosovjny Weeds ‘Uoysjooyy JO SYOOI SNOTEFTAMy “YT Try ‘QUOJSOUN] [VploonyT Aopun spreul CUO Sf | .vceeeeee
“erpuy ‘Teyeuuuleyy 7B SB SspIssoF ouLLS OY} PUB “quoyspuRs OTJTUOONVT *a ITA
seagdowuay, UJI OUO}s[AVUL PUB OUOJSpUBS WOT{) 7 a ee X ‘dnoiZ vivjojQ—ouojspus snoodeopuy “P
“QYVAIULOTGUOD JOMOTT *O ‘OIJINUUMUNU PU SHOeTvOTBO *2
*(soavor snouopet{joory) [woo snourumngyre, “sp -***** cee 2 ‘spre Avo onyq Ysnoy, *@ } Bee Gi reece
‘sayedotmopsuoo aoddy ‘ww “QUOJSPULS JOAVLY) “DV
. ‘SUOTJBVULLOT

"SUOLJVUILO

» woyoag fo spay

‘punway many “uopybiug 07 sdyp ayp wort worpoagy

DR. HECTOR ON THE GEOLOGY OF NEW ZEALAND. 55

character of a littoral or shallow-water calcareo-arenaceous deposit,
rich in distinctive fossils associated with other forms that are com-
mon in the underlying formations. The stratigraphical position of
the Ototara group, as the upper part of what for convenience of
mapping, I have termed the Cretaceo-Tertiary formation, is shown
in the following general schedule, in which the local names and sub-
divisions used in surveying have been suppressed, and a comparison
attempted with European formations :—

Formation. Physical Character.
( I. Recent and Postpliocene. a. Fluviatile and littoral. M
a IT. Upper Pliocene. Fluviatile, extended aie one
a Fluviatile, oldest gold-drift. } P A
ae ITI. Older Pliocene. b. Marine, craglike and conglomerates.
eS IV. Upper Miocene. a. Marine, argillaceous, concretionary.
V. Lower Miocene. Marine, arenaceous, concretionary.
VI. Upper Eocene. Marine calcareous.

6. Estuarine lignitiferous.
a. OroTaRA group—cealcareous sandstone.
b. Chalk and marl.
a. Marine, ferruginous clays and green-
| sands.
| b. Estuarine, gravels and hydrous brown
4

VII. Cretaceo-Tertiary.
VIII. Middle Cretaceous.

coal ; chief SAURIAN BEDS.

m
( IX. Lower Cretaceous, a a. Fluviatile grits and conglomerates;
=| anhydrous brown coal.
2 wa | 2. Marine, grits, green and grey sand-
8 4 [ | stone.
R X. Jurassic. Green sandstone and shales, with thin
= | coal, Pecopteris, Teniopteris, &e.
XI. Liassic. Red and blue marlstones.
( XII. Trias. eens and shales, plant-beds.

About 320 specimens of fossil Mollusca and Echinodermata have
been described from the first seven of the foregoing divisions, the
collections having been made from 90 different localities* ; and I
have prepared the following Table to show the actual number of
recent and extinct species, comparing the recent fauna with the
Upper, Middle, and Lower Tertiary and Cretaceo-Tertiary forma-
tions, omitting those which are of doubtful specific identity or
stratigraphical position.

Univalves. | Bivalves. te ae Totals.

Formations. Ree ibe ee tbe) Rh. | Mx. | OR | xe Re Ex.
Hee ee beeen 1k k 134 9 26 454
700d 9 Pe ae 67 14 45 9 6 (@) 3 0 | 120 23
Ve Wrest carn 30 60 22 33 2 e di a 5d 109
UGE) et Ree an wf 28 4 46 2 8 1 i 14 89
10 LL a, ae a a 0 | 28 9; 0) 20 70
Extinct species,

eS 1 bee 87 Es Th ah Oe eee ee 194

* Catalogue of Tertiary Fossils in the Col. Museum, by F. W. Hutton, Assist.
Geologist. New Zealand, 1873.

56 DR. HECTOR ON THE GEOLOGY OF NEW ZEALAND.

From the comparison which this Table affords with the recent
fauna of the same area, the Otatara formation would seem to have
no claim to a place among Eocene formations. This is confirmed by
the occurrence of a few fossils of decidedly Cretaceous type, such as
Saurian forms and fragments of the shell of Znoceramus, and the pre-
sence of many forms that are associated with decided Mesozoic fossils
in the underlying strata. On the other hand, the occurrence of
decidedly Tertiary fossils (among which is Nautilus ziczac, or a closely
allied form), the gigantic Penguin bones, and the recent discovery
of the bones of a Turtle, also from the Seal-rocks, indicate a fauna
not dissimilar to that at present existing in adjoining areas to the
north and south; so that any additional evidence bearing on this
matter, such as is afforded by Mr. Woodward’s determination of the
affinities of this Crustacean, is an important contribution to New-
Zealand paleontology.

The accompanying section (p. 54) explains the general sequence of
the formations referred to in the foregoing paper on the west coast
of the South Island, from Brighton to the Alps.

EXPLANATION OF PLATE VII.

Harpactocarcinus tumidus, H. Woodw. Three fourths nat. size.

Fig. 1. Upperside of carapace.
2. Underside of carapace.

The specimen is preserved in the British Museum.

Quart. Journ Geol. Soc Vol. XXXIL.PL.VIL

ra yy

GHFord & CO L..Gnesbach. Mintern Bros. imp .

HARPACTOCARCINUS TUMIDUS.

“7

eae Av pr At ee Bie © ro
re Py ‘ : 2% ? MM A i : :
re Rr te lh t
> i
, i)
: :
’
e
‘
:
‘
.
4
.
%
F
a
“f
"
‘
%
er,
¢
r
‘
‘
s .
j
* ™ ;

PS Te

H, WOODWARD ON A BRITISH COAL-MEASURE SCORPION, 57

6. On the Discovery of a Fossiz Scorpion in the British Coar-
Measures. By Henry Woopwarp, Esea., F.R.S., F.G.S., of the
British Museum. (Read November 3, 1875.)

[Puate VIII. ]

THE paucity of the remains of any class of terrestrial air-breathing
animals preserved in a fossil state, has given to such organisms a
special interest in the eyes of geologists, not only as marking the
probable position of old land-surfaces, but also as giving us some
slight indication of the climatal and zoological conditions of this,
to us, otherwise terra incognita.

But this interest is greatly enhanced when such remains occur
in strata of paleozoic age, in which evidences of land-dwelling
animals are extremely rare. Indeed, but for our paleozoic coal-
strata, we should have little left, save evidences of marine life, older
than the Trias.

I have now the pleasure of announcing the discovery, in two dis-
tinct localities in England, of the remains of a fossil Scorpion in
strata of Coal-measure age, and possibly in a third locality in
Scotland.

So long ago as 1835, Count Sternberg published the discovery of
a remarkable fossil Scorpion in the Coal formation at the village
of Chomle, near Radnitz, Bohemia, afterwards named and described
by Corda as Cyclophthalmus senior (Corda in ‘ Bohmischen Verhand-
lungen,’ 1836, and ‘ Wiegmann’s Archiv,’ 1836, i. p. 360).

This is the first authentic record of an Arachnide in the Coal-
measures. It has been figured in the ‘Transactions of the Royal
Bohemian Museum ;’ and these figures were reproduccd by the late
Dr. Buckland in his ‘ Geology and Mineralogy’ (1836, plate 46’
and 46”. fig. 13).

*‘This fossil Scorpion,” writes Buckland, “ differs from existing
species less in general structure than in the position of the eyes.
In the latter respect it approaches nearest to the genus Androctonus,
which, like it, has twelve eyes, but differently disposed from those
of the fossil species. From the nearly circular arrangement of these
organs in the latter animal, it has been ranged under a new genus,
Cyclophthalmus”’ (Buckland, op. cit. vol. i. p. 407).

This remarkable fossil remained unique until 1839, when Corda
added a new genus of Pseudo-scorpions, also from the coal of
Bohemia. which he named Microlabis.

No other Scorpion was noticed from the Coal-measures until
1868, when Messrs. Meek and Worthen, in their ‘ Paleontology of
Illinois,’ described a new form of this ancient family (from the lower
part of the true Coal-measures of Mazon Creek, Morris, Grundy
County, Illinois) * under the name of Hoscorpius carbonarius. This

* Geological Survey of Illinois (A. H. Worthen, Director), vol. iii. 1868
(Paleontology, by F. B. Meek and A. H. Worthen, p. 560, with woodcut).

58 H. WOODWARD ON A BRITISH COAL-MEASURE SCORPION.

form is considered by its describers to be near to the recent Ameri-
can species Buthus hirsutus from California, with which it accords
well in its general proportions; but as neither the palpi nor the
last three segments of the body are preserved, we fully agree in the
conclusions of the learned authors as to the desirability of keeping
it distinct generically.

The same locality in Illinois has yielded, in addition to many new
and interesting forms of Crustacea, a new genus of Myriopods, named
by Messrs. Meek and Worthen Huphoberia, with two species H. ar-
migera and EL. major, M. & W., two new Arachnides, named Mazonia
Woodiana, M. & W., and Architarbus rotundatus, Scudder, and eight
insects, determined by Mr. 8. H. Scudder.

In April 1874 I visited an experimental sinking for coal on the
estate of the Earl of Dartmouth at Sandwell Park near Birmingham,
for the purpose of examining (on behalf of the British Museum) a
large series of fossil plants of great interest, from the “‘ Red Rocks ”
overlying the “ Thick Coal.”

My attention was specially called to an obscure specimen (dis-
covered by Mr. Henry Johnson, C.E., the Secretary and Manager of
the Sandwell-Park Colliery Co.) in black shale, which, to my surprise
and pleasure, proved on examination to be the remains of a fossil
Scorpion (Pl. VIII. fig. 1).

In November 1874 I received from Dr. D. R. Rankin, of Carluke,
a single segment (preserved in the round) of an Articulate from the
shale-beds of the upper limestone series of the Coal-measures near
Carluke, which I am unable to assign to any other group of Articu-
lata; and I believe it to be an abdominal segment of a Scorpion
(figs. 5, 5a, 5b).

In August last, Mr. Edward Wilson of Nottingham (a most

zealous and careful collector of geological specimens, and thoroughly .

acquainted with the geology of the country around Nottingham)
forwarded me several specimens from the clay-ironstone nodules of
the Coal-measures, Skegby New Colliery, near Mansfield. Two of
these clay-ironstone nodules exhibit, now that they have been split
open—one, the body-segments of a fossil Scorpion (fig. 2); and the
other, one of its palpi (fig. 3). So interesting an addition to the
already numerous forms of Arthropoda from the English Coal-
measures deserves to be made known as widely as possible, that
those geologists who live near workable coal-beds where the clay-
ironstone nodules occur may be encouraged by these “finds” to
make use of their hammers and their eyes to furnish us with more
perfect remains of this ancient air-breather of the coal-period.

As the specimens of Scorpion from the coal are at present very
imperfect, I will only point out that our Scorpion agrees best with
an Indian form which I cannot very well distinguish from Scorpio
afer; but I prefer to adopt Messrs. Meek and Worthen’s genus
Eoscorpius with the specific designation of_anglicus for the fossil
form, and to await the discovery of more perfect remains before
venturing to define its characters more minutely.

ee ee eee

Quart. Journ. Geol. Soc. Vol. XXXII. Pl. VII.

‘4
.
4
)
ees
Hi}
14— |
Z
GHFord & C.L.Griesbach. Mintern Bros. imp.

BRITISH FOSSIL SCORPIONS.

H. WOODWARD ON A BRITISH COAL-MEASURE SCORPION. 59

EXPLANATION OF PLATE VIII.

Fig. 1. Eoscorpius carbonarius, sp. nov., from the Sandwell-Park Colliery.

2. The abdominal segments, from near Mansfield.

3, One of the chelate palpi, from near Mansfield, enlarged. 3a. The same,
nat. size. ;

4. Scorpio afer (recent), nat. size.

5. Single articulation preserved in the round from the Coal-measures,
Carluke, Scotland; seen from above. 5a. The same, end view. 50.
The same, seen from below.

60 H. WOODWARD ON AN ORTHOPTEROUS INSECT

7. On a remarkable fossil Ornrnoprerous Insect from the Coat-
MEASURES of ScortanpD. By Henry Woopwarp, Ksa., F.R.S.,
F.G.S., &c., British Museum. (Read November 3, 1875.)

[Puate IX.]

Tue clay-ironstone of Coalbrook Dale, Shropshire, has already yielded
the wing of one most interesting form of fossil locust, the Gryllacris
(Corydalis) Brongniarti (Pl. LX. fig. 2), figured and described in
Mantell’s ‘ Medals of Creation,’ in Murchison’s ‘ Siluria,’ 4th edition,
1867, p. 300, woodcut 80, and, lastly, in the Geol. Mag. 1874, new
series, Dec. 11. vol. 1. pl. xiv. fig. 3; whilst the Coal-measures of
Saarbruck and of Westphalia have yielded fossil remains of a Giryl-
lacris and seven species of Blattinw, or cockroaches, 4 Termites
(white ants), 1 Acridites, 3 species of the Neuropterous genus Dic-
tyoneura; and from those of Grundy co., Illinois, U.S. A., 3 species
of the genus Miamia, 1 of Mylacris, 1 of Megathentomum, 8 species
of Huphemerites, and 1 doubtful species of Mantis have been described
by Mr. 8. H. Scudder, and from the Coal-measures of Cape Breton
the genus Haplophlebiwm by the same author*.

Mr. Scudder has likewise described seven genera from more or less
fragmentary remains in the Devonian rocks of New Brunswick?.

Dr. Anton Dohrn (Director of the Zoological Station at Naples)
has described a new and most remarkable Neuropterous insect
from the Permian (Todtliegende) of Birkenteld, which he has named
Eugereon Bockingii, and regards as possibly belonging to an extinct
order (Dicryoprera) t.

It has been my good fortune during the past year to have placed
in my hands, through the kindness of Edward Charlesworth, Esq.,
F.G.8., a most interesting fossil insect from the Coal-measures of
Scotland.

The insect in question has been preserved on the half of a clay-
ironstone nodule, which had been split open for the purpose of
examination ; the fate of the corresponding half is unknown.

These clay-ironstone nodules have been known from the earliest
times to contain fossil remains, and have in consequence been dili-
gently examined; but as they occur in infinite numbers, and the
geologists are few who devote themselves to their study, we know
but little, after all, of the buried treasures they contain, and they too
often pass into the smelting-furnaces unnoticed and unseen. In one
clay-ironstone nodule larger than usual, Mr. George Maw, F.G.S.,
of Benthall Hall, near Broseley, obtained the entire and uncrushed
skull of a Labyrinthodont, one of the most completely preserved
of such remains ever found in the Coal. I had the good fortune
to procure, through Mr. Hollier, of Dudley, in a small nodule,

* Geol. Mag. 1867, vol. iv. p. 386, pl. xvii. fig. 1.

t Ibid. 1868, vol. v. pp. 172, 216.

+ Dr. Dohrn in ‘ Paleontographica,’ 1866, Bd. xiii. p. 333, taf. xli., and
op. cit. 1869, Bd. xvi. p. 129, taf. viii.

a al

FROM THE COAL-MEASURES OF SCOTLAND. 61

the nearly perfect remains of an Arachnide, Hophrynus Prestvicit,
having both sides seen in the two halves of the same stone*.
I allude to this in order to stimulate geologists in the neigh-
bourhood of coal-fields to devote some time to splitting these
nodules of the ‘‘pennystone” ironstone, feeling assured they will
be well rewarded by finding king-crabs, Arachnides, Myriopods,
insects, ferns, fruits, and other fossil remains, which may serve to
enrich their own and the public museums, and increase our know-
ledge of the fauna and flora of the Carboniferous epoch.

Owing to the large expanse of surface covered by the fossil insect
under consideration, only two of the four wings are well preserved ;
and of these two we have not the extremities, or the posterior
border of the hinder wing.

The two best-preserved wings measure 24 inches in length and
1 inch and 13 inch respectively in breadth, the hind wing being
the broadest, as is the case among the OrrHoprsra generally.

The anterior margin of the fore wing is not well preserved; but
a careful examination brings to light the characteristic strong lon-
gitudinal vein running parallel with the fore margin, so necessary
for giving support to the wing in flight.

This is followed by a second equally straight vein.

The third vein is likewise parallel, but bifurcates near the ex-
tremity of the wing.

The fourth vein is much curved, and gives rise to six branches,
three of which again bifurcate, and all bend outwards and back-
wards towards the hinder border of the wing. About 2 lines from
the base of this strong fourth vein it unites with a fifth short vein
at an acute angle, enclosing with the proximal border of the wing
a triangular space. This triangular area is observable in all four
wings, and appears to correspond exactly with a similar vein-area
in the wing of Gryllacris (Corydalis) Brongniarti described by Mr.
A. H. Swinton in the ‘ Geological Magazine,’ 1874, Decade ii. vol. i.
p- 337, pl. xiv. fig. 3, as ‘‘a vocal organ, and which in the Locustinz
appears to be greatly developed, and to be furnished with a series of
minute arches, which the insect causes to vibrate by moving one
wing across the other, as the fiddle-bow is moved across the strings
of the violin.”

This shrill-vein is by no means always developed as an organ of
sound in all the members of the Orthoptera; but few are without
if In some measure.

The other veins of the wing take their rise from the centre of
this triangular vein, and, branching into three and into two, ter-
minate in the hinder border.

The venation of the hinder wing differs but little from that of
the fore wing, save in its greater depth, and consequently in the
prolongation backward of the hindermost veins. (See Pl. IX. fig. 1.)

The transverse veinlets which brace the principal veins together,
have a general rectangular arrangement, dividing up the wing into
a great number of chiefly square and oblong meshes.

* Geol. Mag. 1871, vol. viii. p. 385, pl. xl.

62 H. WOODWARD ON AN ORTHOPTEROUS INSECT

Of the body we see but little; a transverse band marks the pro-
pbable division between the meso- and metathorax; and from the
undisturbed position of the bases of the four wings we may conclude
that the insect was not less than five lines in breadth across the
thorax.

Immediately in front of the wings, and evidently in situ, is the
prothorax, consisting of two dilated and rounded lobes, veined like
the wings themselves, but rather more irregularly, and measuring
14 lines across and 6 lines in length. In the centre of these two
rounded lobes, and extending 8 lines in advance of them, is what
appears to be a slender frontal process, 3 lines long, behind which is
the small head with its eyes, measuring 3 lines in breadth. In
front and from one side of the prothorax, one leg is seen remaining
um situ.

Compared with the ordinary Neuroptera, and even with Dr.
Dohrn’s Eugereon Backingu, we see that in the neuration of the
wings there is a marked distinction ; for whereas the cross veins in
our fossil run nearly always at right angles to the longitudinal veins,
leaving more or less quadrangular meshes, the wings in Hugereon
and many other Nevroprera have the interspaces between the veins
filled by polygonal meshes, especially in the hinder part of the wings.
Nor does the arrangement of the main veins correspond.

Compared with modern Orrnoprmra, we find that in the fossil
there is a tendency to greater branching of the large longitudinal
veins, and a more frequent branching of the transverse veinlets that
brace them together; and that there is a less parallel disposition of
the former.

Between this form and Corydalis (Gryllacris) Brongniarti (Pl. IX.
fig. 2), from the Coal-measures, Coalbrook Dale, there is a marked
sinilarity.

Having corresponded with Prof. J. O. Westwood, of Oxford, and
also communicated with my esteemed colleague Charles O. Water-
house, Esq., of the Zoological Department, I am confirmed by both
these entomologists and also by R. McLachlan, Esq., F.L.S., in re-
ferring this remarkable fossil to the neighbourhood of the Manrrp am.
In no other group do we meet with such a dilatation of the pro-
thorax; and although the frontal process seems at first sight ab-
normal, nevertheless I find it, combined with the prothoracic
dilatations, in the genus Blepharis, and notably in B. domina, from
the White Nile. ‘There is also a general agreement in the neura-
tion of the wings, which leads me to rest satisfied in this deter-
mination. Nevertheless there are points of affinity with the Nxrv-
ROPTERA Which mark this insect, and which led Audouin to refer the
Coalbrook-dale example to Corydalis, although Mr. A. H. Swinton
has subsequently referred it to Gryllacris*.

One great distinction between Blepharis domina and the fossil
under consideration is, that in the latter the dilatations of the pro-
thorax are tumid and rounded, whilst in the former they are per-
fectly flat and subtriangular.

* Geol. Mag. loc. cit.

FROM THE COAL-MEASURES OF SCOTLAND. 63

Without following the example of Dr. Dohrn and proposing a

new order for the fossil before us, I have no hesitation in placing it

in

a new genus, for which I would propose the name of Lithomantis,

calling the species L. carbonarius.

The following is a list of the hitherto described fossil Insects and

Arachnida found in the Palaeozoic rocks :—

Al

meen
: ee
Our G bor

LS
—
—

| 20.

PERMIAN.
EKugereon Beckingii, Dohrn. Birkenfield.

COAL-MEASURES.
ARACHNIDA.

I. Scorpionide.

. Microlabis Sternbergi, Corda. Coal-measures, Bohemia.

. Cyclophthalmus senior, Corda. Ditto, ditto.

. Mazonia Woodiana, M. & W. Ditto, Illinois.

. Hoscorpius carbonarius, JZ. d W. Ditto, ditto.

. Hoscorpius anglicus, H. Woodw. Ditto, Sandwell-Park and Skegby Col-

lieries.

II. False Scorpions.

. Architarbus rotundatus, Sewdder. Coal-measures, I1linois.

subovalis, H. Woodw. Ditto, near Padiham, Lancashire.

. Hophrynus Prestvicii, H. Woodw. Ditto, Coalbrook Dale and Dudley.

Ill. Araneide.

Aranea, sp. Coal-measures, Bohemia.
Protolycosa anthracophila, Rémer. Ditto, Silesia.

. Arthrolycosa antiqua, Harger. Ditto, Morris, Illinois.

MyrioropaA.

. Xylobius sigillariz, Dawson. Coal-measures, Glasgow, Huddersfield, and

Nova Scotia.

. Euphoberia armigera, M. & W. Ditto, Illinois, U.S., and Nova Scotia.

Brownii, H. W. Ditto, Glasgow.

—— major, M.§ W. Ditto, Illinois, U.S.
anthrax, Salt., sp.(?). Ditto, Coalbrook Dale.
ferox, Salt., sp. Ditto, ditto.

CoLmOPTERA.

. Curculioides Ansticii, Buckl. Coal-measures, Coalbrook Dale.

. Scarabeeus, sp. Ditto, Saarbruck (a fossil fruit! Rémer).
Troxites Germari, /. Gold. Ditto, Saarbruck.
ORTHOPTERA,
(Blattide.)

. Blattina primeva, F. Gold. Coal-measures, Saarbruck. ©“ te thine (f°

Lebachensis, #. Gold. Ditto, ditto.
gracilis, f. Gold. Ditto, ditto.
anaglyptica, Germar. Ditto, Westphalia.
anthracophila, Germar. Ditto, ditto.
didyma, Germar. Ditto, Westphalia.
flabellata, Germar. Ditto, ditto.

, —— yenusta, Lesgrx. Ditto, Frog Bayou, Arkansas.

64 H. WOODWARD ON AN ORTHOPTEROUS INSECT

(Locustide.)

29. Gryllacris lithanthraca, /. Gold. Coal-measures, Saarbruck.
(Corydalis) Brongniarti, Mant. Ditto, Coalbrook-dale.
31. Acridites carbonatus, Germar. Ditto, Westphalia.

(Mantide ?)

32. Lithomantis carbonarius, H. Woodw. Coal-measures, Scotland.
33. Mantis ?, sp., Sewdder. Ditto, Grundy Co., Ilinois.

NEUROPTERA.

34. Termes Heeri, /. Gold. Coal-measures, Saarbruck.
: formosus, Ff’. Gold. Ditto, ditto.
36. —— Decheni, F/. Gold. Ditto, ditto.
affinis, F. Gold. Ditto, ditto,
38. Dictyoneura anthracophila, F. Gold. Coal-measures, Saarbruck.

39 Humboldtiana, #. Gold. Ditto, ditto.

40. libelluloides, F. Gold. Ditto, ditto.
J 41. Miamia Bronsoni, Dana. Ditto, Grundy Co., Illinois.
J 42. Dane, Scudder. Ditto, ditto.
Vv 43. —— (Chrestotes) lapidea, Scudder. Ditto, ditto.

44. Hemeristia occidentalis, Dana. Ditto, ditto.

45. Mylaeris anthracophila, Scudder (in Dana).

46. Megathentomum pustulatum, Sewdder. Ditto, ditto.
47. Eupbemerites simplex, Scudder. Ditto, ditto.

gigas, Scudder. Ditto, ditto.

affinis, Scudder. Ditto, ditto.

50. Haplophlebium Barnesii, Scudder. Ditto, Cape Breton.

LEPIDOPTERA?
51. Tinea, sp., Fabricius. Coal-measures.

DEVONIAN.

. Platephemera antiqua, Sexdder. Devonian, New Brunswick.
. Homothetus fossilis, Sewdder. Ditto, ditto.

Lithentomum Harttii, Sewdder. Ditto, ditto.

. Xenoneura antiquorum, Scudder. Ditto, ditto.
Gerephemera simplex, Scudder. Ditto, ditto.

. Dyscritus vetustus, Scudder. Ditto, ditto.

. Archimylacris acadicus, Scudder, Ditto, ditto.

STS OUR 09 bo

EXPLANATION OF PLATE IX.

Fig. 1. Lithomantis carbonarius, sp. noy.; nat. size. Coal-measures, Scotland.
2. Gryllacris (Corydalis) Brongniarti; nat. size. Ditto, Coalbrook Dale.
3. Blepharis domina (recent); nat. size. White Nile, Abyssinia.

Discussion.

Mr. CuarteswortH inquired whether the presence of the few
Cretaceous fossils found in the deposit which had furnished the
New-Zealand Crab described might not be the result of the degra-
dation of preexisting rocks.

Dr. Hector replied that, on stratigraphical grounds, this could not
be the case.

Mr. CuarteswortH stated that he had been unable to ascertain
the precise locality of the fossil Orthopterous insect described, but
that he was informed by the gentleman from whom he received it —

Gee doen eel Soc’ Vol XM FLIX.

CS

Vi

i

i gainme:s ry ahY) Ren nlarae Acree

Sr tN ee RTI pal AS

MEERA Yo AI OL OM EO
Miigg Sowa SAN wicage hia manele KIB biel >>

rod) 5 i)
{2 PLAC ERD
SOAS Lf? Ey Reg
ISA bps: ac

PPL | ALL EEN
<¢ KR iy, ATTY Lem ere Vyas: CORE, Lipp uN
apg iag Minin Ma ea ua ene one
| SOE A ca
OG iy
NSIT
USK SISO

sith
Lda

ely

LIAISE

Mintern Bros. imp.

GH¥Ford & C.L.Gnmesbach.
LITHOMANTIS CARBON ARIUS.

FROM THE COAL-MEASURES OF SCOTLAND. 65

that the nodule containing the specimen was picked up by a lady in
Scotland.

Prof. Morris remarked that the New Zealand Crab was of espe-
cial interest. All the previously described species of Harpactocar-
cus had been obtained from Nummulitic deposits in the south of
Kurope; and the same concurrence was observed in New Zealand.
Similar phenomena occurred in Australia, where many species re-
sembling European forms had been discovered by M‘Coy.

Mr. ErnerincGeE said that one of Mr. Woodward’s papers demon-
strated the value of the Sub-Wealden boring. He had examined
the cores, and had come to the conclusion that the Oxford Clay was
reached at 500 feet; but in this he was mistaken, owing to his
having wrongly identified the Ammonite discovered at that depth
with Ammonites jason. The occurrence of the same species of
Crustacean at Boulogne and in Sussex was of great interest, as
marking the identity of the deposit in the two localities. Lingula
ovalis occurred with other fossils throughout the Kimmeridge Clay
of the boring.

Mr. Woopwarp thanked Mr. Charlesworth for his endeavours to
ascertain the locality from which his Lithomantis was obtained, but
remarked that, whatever this might be, there could be no doubt as
to its geological horizon.

Q. J.G.S.- No. 125. F

66 J. W. DAWSON ON THE OCCURRENCE OF

8. Nores on the Occurrence of Eozoon canapense at Core St.
Pirrre. By J. W. Dawson, LL.D., F.R.S., F.G.S. (Read

May 12, 1875.)
[Piate X.]

Core Sr. Prerre, in the Seigniory of Petite Nation, on the river
Ottawa, is the locality whence some of the most instructive speci-
mens of Hozoon were obtained by the late Mr. Lowe, whose collec-
tions are referred to in papers presented to this Society by Sir W.
E. Logan and the writer. Believing that a reexamination of this
place would afford a good opportunity for collecting additional speci-
mens, and for the study of the fossil in situ, as well as for testing the
validity of objections recently raised to the animal nature of Hozoon,
I made arrangements for visiting it in September last; and, through
the kindness of Mr. Selwyn, Mr. T. C. Weston, of the Geological
Survey, a skilful collector, and who has had much experience in
preparing and examining specimens of Hozoon, was permitted to
accompany me, and subsequently prepared slices and photographs of
some of the specimens obtained.

The Lower Laurentian rocks of this region have been carefully
mapped and described in the Reports of the Geological Survey, to
which I may refer for their general description. The limestone,
which has afforded Hozoon at Cote St. Pierre, is a thick bed belong-
ing to the Grenville band of Sir W. E. Logan, and included between
the two great belts of orthoclase gneiss (the third and fourth gneiss)
which in this region constitute the upper beds of the Lower Lauren-
tian. Its average thickness, according to the measurements of Sir
William Logan, is 750 feet; but it varies from 1500 feet to 60 feet.
Its outcrop has been traced in the country north of the Ottawa for
at least 100 miles, along several anticlinal and synclinal folds*.

At Cote St. Pierre this limestone occurs on the flank of a hill of
gneiss and stratified diorite, with a dip to the south-east at angles
of 70° to 80°. The dip, however, is very inconstant, owing to the
contortions of the beds.

The limestone is white and crystalline, and may be described as
thin-bedded, since it presents a great number of layers of no great
individual thickness, and differing in the coarseness of the crystal-
lization and in the presence of dolomite, serpentine, and layers of
gneissose matter in some of them. The specimens of Hozoon were
found to be abundant in only one bed, not more than four feet in
thickness, though occasional specimens and layers of fragments
occur in other parts of the band. The exposures are in part natural
weathered surfaces seen on a wooded bank, in part an opening
made by Mr. Lowe to extract specimens of Hozoon, and a larger
opening made, as we were informed, by parties in search of fibrous
serpentine, or “ rock-cotton,” for economic purposes.

* See map in ‘Geology of Canada,’ 1863.

EOZOON CANADENSE AT COTE ST. PIERRE. 67

The sections seen in the artificial exposures may be tabulated as
follows, though from the highly inclined position of the beds and
the irregularity of the excavations, perfect accuracy was not at-
tainable ;—

Mr. Lowe's Excavation (order descending).

1. Limestone with serpentine and entire specimens of Hozoon—3 feet.
2. ee crystalline limestone, with layers containing fragments of Hozoon—
eet.
3. Limestone with concretions and layers of serpentine, and a few specimens
of Eozoon—several feet, to the bottom of the excavation.

Fisher’s Excavation (order descending).

. Laminated limestone with bands of serpentine—6 feet.

. White laminated limestone traversed by small veins of chrysotile—8 feet.

. Limestone with concretions and interrupted bands of serpentine and
pyroxene, and fragments of Hozoon—10 feet. (Crystals and layers of
dolomite occur in this and the preceding beds.)

Go NS

_ 4. Limestone with large concretions of serpentine, and in one layer fine-grained

variety of Hozoon (var. minor)—20 feet.

5. Limestone with serpentine and perfect specimens of Hozoon. (This pro-
bably corresponds to Lowe’s excavation)—12 feet.

6. Coarse-grained limestone and dolomite—several feet.

(After a break of several yards)
7. Limestone with masses of pyroxene and veins of chrysotile and some imper-
fect Hozoon.
(After a break of several yards)
8. Coarse-grained diorite, resting on a thick band of gneiss.

In front of Lowe’s excavation, and apparently overlying the lime-
stone exposed in it, is a narrow ledge of fine-grained gneiss; and
beyond this other and probably overlying limestone appears, holding
pyroxene and mica. The whole vertical thickness of the limestone
exposed can scarcely exceed 150 feet; but this is probably only a
small part of the development of the band at this place.

In the strike of the limestone to the south it appears to bend
abruptly, or to be thrown by a fault, to the south-east, the gneiss
and diorite coming forward into a line with it, and the limestone
appearing in a liitle bare knoll in front of these. On the surface of
this limestone were found some fine specimens of weathered Hozoon.

I examined carefully the relation of the bedded serpentine and
the veins of chrysotile or fibrous serpentine to the limestone. The
compact serpentine is evidently an original part of the deposit,
occurring in layers and lenticular concretions. In some beds it
shows no indication of the structure of Hozoon; but in others it fills
the cavities of the fossils, and there are many regular layers of
fragmental Hozoon of considerable thickness in which it fills the
cells, while in other layers interstratified with these the fossil is
associated with dolomite. I satisfied myself on this point not only
on the ground, but also by taking away large specimens represent-
ing several thin layers, and treating them with dilute acid so as to
being out the structure. The following is a section of such a speci-
men, 54 inches in vertical thickness, treated with acid and examined

with a lens :-—
F2

68 J. W. DAWSON ON THE OCCURRENCE OF

. Limestone with crystals of dolomite and a few fragments of Eozoon.

. Fime-grained limestone with granules of serpentine—the latter filling the
chamberlets of fragments of Hozoon and small globigerine Foraminifera.

. Limestone with dolomite and including a thin layer of serpentine as above.

. Limestone and dolomite with grains of serpentine and fragments of supple-
mental skeleton of Hozoon.

. Crystallized dolomite, holding a few fragments of Hozoon in the state of
calcite.

. Limestone with disseminated serpentine as above, chamberlets of Hozoon and
fragments of its supplemental skeleton, also small groups of chamberlets,
perhaps globigerine Foraminifera.

Om om POO Ne

In other specimens a like thickness of rock presented a mass of
fragments of supplemental skeleton with the canals injected with
serpentine, and granules of the same filling the chambers.

The chrysotile veins, which are sometimes an inch or more in
thickness, but branch off into the most minute fibres, are evidently
altogether subsequent in origin to the bedded limestone and ser-
pentine. They are undoubtedly of aqueous origin, and in their
mode of occurrence strongly resemble the veins of fibrous gypsum
penetrating the Lower Carboniferous marls of Nova Scotia. They
cross the bedding in all directions, and pass through the structure
of Hozoon, though sometimes running parallel to its laminee for short
distances. They must have been introduced after the Hozoon was
mineralized, and have evidently no connexion with its structure.

In the diagram (Pl. X. fig. 2) I have attempted to represent this
relation ; and I have no hesitation in stating that the assertion that
these chrysotile veins are identical with or similar to the proper wall
of Hozoon either in structure or distribution, is wholly without foun-
dation, other than that which may arise from confounding dissimilar
structures accidentally associated with each other.

Some slickensided joints lined with a lamellar and fibrous ser-
pentine traverse the beds, and, as the chrysotile veins sometimes
terminate in them, may be older than the latter. These also were
observed to cross the masses of Hozoon.

Few disseminated minerals, other than those already mentioned,
were observed in the Hozoon limestone. A few detached crystals of
mica, pyroxene, and pyrite were found in the fragmental layers, and
also a few rounded particles of quartz, probably grains of sand.

The perfect examples of Hozoon, at least those rendered evident
by mineralization with serpentine, are confined to certain bands of
limestone, and notably to one band—that originally opened by Mr.
Lowe. In this bed the fossil occurs in patches of various sizes, some
of them two feet or more in diameter, and bent or folded by the
contortions of the strata; others are much smaller, down to a few
inches. On the weathered surfaces the specimens mineralized with
serpentine project, and exhibit their lamination in great perfection,
resembling very closely the silicified Stromatopore of the Corniferous
Limestone.

None of the specimens of Hozoon is of any great vertical thick-
ness. The lower lamine are generally the best developed and with
the thickest supplemental or intermediate skeleton. The upper

EOZOON CANADENSE AT COTE ST. PIERRE. 69

laminz become thin-walled, though often very regular; and after
about 100 laminz at most, the superficial portions become acer-
vuline for an inch or so and then terminate. In some speci-
mens only a few regular lamin are formed, succeeded by acervu-
line structures. A very fine and regular specimen in my collection
has 100 lamine in a thickness of 34 inches, giving a little more
than a thirtieth of an inch for the average height of each lamina of
sarcode and shell-wall.

In order to test the state of preservation of the canal system and
nummuline layer, I treated a great number of specimens from diffe-
rent parts of the bed with a dilute acid. The result was, that in all
the canal system could be detected in greater or less perfection in
the thicker lamine near the base of the forms, and in some through
a great number of the lamine. The structure of the nummuline
layer was not so constantly preserved, its tubuli not being infiltrated
in some parts, so that it appears as a structureless band, or fails
altogether to be visible. In no instance could it be seen to pass
into chrysotile, as recently affirmed by Messrs. Rowney and King’,
although chrysotile veins often run very near to or across the walls.
The nummuline layer is almost always distinctly limited by parallel
surfaces, with its tubes at right angles, or nearly so, to these. The
sort of chevron arrangement figured by Rowney and King in fig. 7
of their plate, in the number of the ‘ Annals of Natural History’ for
October 1874, I have never observed; and Mr. Weston, who has pre-
pared and examined microscopically hundreds of specimens of Hozoon,
was struck with the inaccuracy of the representations in this plate,
and remarked on it the first time that I met him after he had seen
the paper referred to. Figs. 2 and 3, Pl. X., relate to these points, and
show the actual nature of the nummuline wall and its relation to
the intermediate skeleton and chrysotile veins, as do also the figures
recently published by Dr. Carpenter in his paper in the ‘ Annals’f.

By careful scrutiny of the beds we were enabled to detect two
new forms of Hozoon, which may eventually prove to be distinct
species, but which for the present may be regarded as varietal
forms.

One of these, found im situ by Mr. Weston, is flat in form and
very finely laminated, with thin walls except near the lower part,
where there is some supplemental skeleton with finer and more
curved canals than usual. The thin walls or lamine of the ordi-
nary skeleton are connected by very frequent vertical pillars or par-
titions, and are as numerous as thirty in half an inch, while the
whole thickness of the specimen does not exceed an inch. It thus
very closely resembles some of the Devonian and Silurian Stromato-
pore, especially when seen on the weathered surface. It may be
named in the mean time variety minor.

The second occurs in more or less oval patches a few inches in
diameter, limited by a sort of frame or border of compact serpen-
tine, and presenting under the microscope an aggregation of small
acervuline chamberlets, having the proper wall filled with unusually

* Ann. & Mag. of Nat. Hist. May 1874. t Lbid, June 1874.

70 J. W. DAWSON ON THE OCCURRENCE OF

long parallel tubes, and with little development of supplemental
or intermediate skeleton. The appearance of parallel tubulation
running through and past several successive chamberlets was more
conspicuous in these specimens than in the ordinary acervuline
Eozoon, and the chamberlets themselves more cylindrical and tortu-
ous. These specimens may either be portions of the acervuline
superficial part of Hozoon broken off and separately preserved, or
they may constitute a distinct varietal form. As the latter seems on
the whole most probable, I would name this form variety acervulina.

These varieties are of much more rare occurrence than the ordi-
nary type of Hozoon.

The ordinary specimens of Zozoon found at St. Pierre are mineral-
ized with serpentine; but fragments imbedded in the dolomitic
limestones have their canals filled with a transparent mineral which,
from its optical character, is evidently dolomite, though the quan-
tity obtained was not sufficient for any definite chemical test.
Parts of the canals in these specimens were filled with calcite, as
shown by its dissolving entirely away in a dilute acid. In one of
the serpentinous specimens also I have observed that, while portions
of the groups of canals, especially the basal portions, are filled with
serpentine, the extremities of the canals and their finer branches
present, under polarized light, the aspect of calcite; and that they
. are filled with this mineral is proved by these portions of the canal-
filling being entirely removed when treated with dilute acid. It
would thus appear that in these specimens, while the terminal parts
of the canals have been filled with calcite, the basal portions have
been occupied by serpentine. ‘This is not, however, a new fact, as
similar appearances have been already described both by Dr. Car-
penter and the writer.

In one specimen I observed a portionof the fossil entirely replaced
by serpentine, the walls of the skeleton being represented by a
lighter-coloured serpentine than that filling the chambers, and still
retaining traces of the canals. The walls thus replaced by serpen-
tine could be clearly traced into connexion with the portions of
those still existing as calcite. This shows that the serpentine, like
the quartz in silicified shells and corals, has had the power of re-
placing the calcite of the fossils; and I believe that its partial
action in this way accounts four some irregularities observed in the
less perfectly preserved specimens. Nor is it improbable, as Dr.
Hunt has already suggested, that some of the masses of serpentine
and pyroxene on which specimens of Hozoon are based, may represent
older and more perfectly mineralized masses of the fossil.

In some of the specimens of Hozoon, the superficial lamine are
apparently broken and displaced in such a mariner as to suggest
the idea that partial disintegration by the waves had taken place
before they were finally buried. It is also observable that in some
of the masses the compression to which they have been subjected
has produced a microscopic faulting, which slightly displaces the
laminee.

One of the most interesting features of the St.-Pierre limestone,

EOZOON CANADENSE AT COTE ST. PIERRE. 71

not noticed by previous observers, is the occurrence of layers filled
with little globose casts of chamberlets, single or attached in groups,
and often exactly resembling the casts of Globigerine in greensand
(Pl. X. figs.4-10). On weathered surfaces they were often especially
striking when examined with the lens. In some cases the chamberlets
seem to have been merely lined with serpentine, so that they weather
into hollow shells. The walls of these chamberlets have had the same
tubulated structure as Hozoon (fig. 4) ; so that they are in their essen-
tial characters minute acervuline specimens of that species, and similar
to those which I described in my paper of 1867* as occurring in the
limestones of Long Lake and Wentworth, and also in the loganite
filling the chambers of specimens of Hozoon from Burgess. Some of
them are connected with each other by necks or processes, in the
manner of the groups of chamberlets described by Gumbel as occur-
ring in a limestone from Finland, examined by him. That they are
organic [ cannot doubt, and also that they have been distributed by
currents over the surface of the layers along with fragments of
Eozoon. Whether they are connected with that fossil or are speci-
fically distinct may admit of more doubt. They may be merely
minute portions detached from the acervuline surface of Hozoon,
and possibly of the nature of reproductive buds. On the other
hand, they may be distinct organisms growing in the manner of
Globigerina. As this is at present uncertain, and as it is convenient
to have some name for them, I propose to term them Archeosphe-
rine, understanding by that name minute Foraminiferal organisms,
having the form and mode of aggregation of Globigerina, but with
the proper wall of Hozoon.

In slicing one of my specimens from Cote St. Pierre I have
recently observed a very interesting peculiarity of structure, which
deserves mention. It is an abnormal thickening of the calcareous
wall in patches extending across the thickness of four or five
lamelle, the latter becoming slightly bent in approaching the
thickened portion. This thickened portion is traversed by regu-
larly placed parallel canals of large size, filled with dolomite, while
the intervening calcite presents a very fine dendritic tubulation.
The longitudinal axes of the canals lie nearly in the plane of
the adjacent lamin. This structure reminds an observer of the
Cenostroma type of Stromatopora, and may be either an abnormal
growth of Hozoon, consequent on some injury, or a parasitic mass of
some Stromatoporoid organism finally overgrown by the Lozoon.
The structure of the dolomite shows that it first inerusted the
interior of the canals, and subsequently filled them—an appearance
which I have also observed in some of the larger canals filled with
serpentine, and which is very instructive as to their true nature.

From the above facts the true nature of Hozoon may, I think, be
rendered evident to any geologist, however little he may have made
the fossil Foraminifera a subject of study. The theories as to its
origin may be sumined up thus :—

1. The complicated theory of pseudomorphism and replacement,

* Quart. Journ. Geol. Soe. vol. xxii. p. 260.

72 J. W. DAWSON ON THE OCCURRENCE OF

advocated by Messrs. Rowney and King, may be dismissed at once.
Independently of the insuperable chemical difficulties which have
been pointed out by Dr. Hunt*, and which he proposes to discuss
more fully in his papers on Chemical Geology, now in the press, we
have the further facts that no replacement of serpentine by calcite
is indicated by the relations of these minerals to each other, while
such replacement as does occur is in the other direction, or the change
of calcite into serpentine, as evidenced by the state of preservation
of some specimens of Hozoon, above referred to. Further, this
theory fails to give any explanation of the specimens mineralized
by pyroxene, dolomite, and calcite, or to account for the nummuline
wall, except by attributing it to the alteration of chrysotile, which
is inadmissible, as the veins of this mineral are newer than the
walls supposed to have been derived from them.

2. Inasmuch as many apparently concretionary grains and lenti-
cular masses of serpentine exist in the Laurentian limestones, it
may be supposed possible that Hozoon is merely a modification of
these concretionary forms. In this case, the filling of each lamina
and chamberlet of Hozoon must be regarded as a separate concretion ;
and even if we could suppose some special cause to give regularity
and uniformity to such concretions in some places and not in others,
we still have unaccounted for the canals and tubuli, or the delicate
threads of serpentine representing them. Further, we have to sup-
pose that a tendency to this regular and complicated arrangement
has affected in the same way minerals so diverse as serpentine,
loganite, pyroxene, and dolomite.

3. The only remaining theory is that of infiltration by serpentine
of cavities previously existing in the calcite. There is no chemical
objection to this, inasmuck as we know of the infiltration of fossils
in other formations by minerals akin to serpentine; and in these
limestones the veins of fibrous serpentine have evidently been intro-
duced by aqueous action subsequently to the production or fossiliza-
tion of the Hozoon. Further, the white pyroxene of the Laurentian
limestones, and the loganite and dolomite, are all known to have
been produced by aqueous deposition. The only question remaining
is, Whence came the original calcite skeleton with lamine, chambers,
canals, and tubuli to be so infiltrated? The answer is given in the
comparison with the tests of Foraminifera, originally proposed by
the writer, and illustrated in so conclusive a manner by the re-
searches of Dr. Carpenter.

I may add, in conclusion, that had geologists generally the oppor-
tunity of studying Hozoon in situ, in good exposures, like that at
St. Pierre, they would much more fully understand and appreciate
the arguments for its organic nature, than when they have had
opportunities of examining only polished specimens and slicest. Its

* Trans. Royal Irish Academy, 1871.

t I have been sorry to find, from specimens in the cabinets of my friends, that
some London dealers are in the habit of circulating specimens labelled “ Hozoon
canadense” which have no trace of the structures of the fossil, but are either badly

preserved acervuline portions or merely ordinary serpentine marble. Such speci-
mens can, of course, only mislead, and may produce much unnecessary scepticism.

EOZOON CANADENSE AT COTE ST. PIERRE. 73

Stromatoporoid masses, projecting from the weathered beds of lime-
stone, would at once attract the attention of any collector ; and the
whole conditions of its occurrence, whether entire or in fragments,
are precisely those of fossil corals in the Silurian limestones. Fur-
ther, the symmetry and uniformity of its habit of growth are much
more apparent when they can be studied in large specimens prepared
by natural weathering or by treatment with an acid.

[ Wote (Oct. 30).—Messrs. Richardson and Weston, of the Geolo-
gical Survey, have recently revisited the locality of Hozoon at Cote
St. Pierre mentioned in the above paper, and have collected some
additional specimens. One of these deserves notice, as illustrating
the nature of Archwospherine. It is a small or young small speci-

Figs. 1-4.—Small weathered Specimen of Eozoon from Petite
Nation.

Fig. 1. Natural size; showing general form, and acervuline portion above and
laminated portion below.
2. Enlarged casts of cells from upper part.
3. Enlarged casts of cells from the lower part of the acervuline portion.
4, Casts of sarcode layers from the laminated part; enlarged.

men, of a flattened oval form, 24 inches in its greatest diameter, and
of no great thickness (fig. 1). It is a perfect cast in serpentine, and
completely weathered out of the matrix, except a small portion of
the upper surface, which was covered with limestone which I have
carefully removed with a dilute acid. The serpentinous casts of the

74 J. W. DAWSON ON THE OCCURRENCE OF

chambers are in the lower part regularly laminated; but they are
remarkable for their finely mammillated appearance, arising from
their division into innumerable connected chamberlets resembling
those of Archeospherina (fig. 4). In the upper part the structure be-
comes acervuline, and the chamberlets rise into irregular prominences,
which in the recent state must have been extremely friable, and, if
broken up and scattered over the surfaces of beds, would not be
distinguishable from the ordinary Archwospherine, This specimen
thus gives further probability to the view that the Archwospherine
may be for the most part detached chamberlets of Hozoon, perhaps
dispersed in a living state and capable of acting as germs. ]

EXPLANATION OF PLATE X.

Fig. 1. Fragments of skeleton of Hozoon, imbedded in dolomite limestone.
(a) Fragment with canals. ()) Fragments not showing canals. (¢)
Dolomite. (Magnified 10 diameters.)

2, Laminated Hozoon, with vein of chrysotile. (@) Calcareous wall, slightly
eroded with acid. (4) Serpentine filling chambers. (¢) Chrysotile
vein crossing the structures. (10 diam.)

3. Portion of a specimens imilar to that in fig. 2; a very thin slice more
highly magnified. (a) Intermediate skeleton with portions of two
large canals. (a’) Proper wall with fine tubulation. (0) Serpentine
filling chambers. (¢) Chrysotile vein traversing serpentine. (Magnified
90 diam.)

4. Small Archeospherina, showing tubulated wall. (200 diam.)

5, 6, 7, 8. Archeospherine, casts, as opaque objects, of some of the varieties,
(75 diam.)

9 and 10. Similar specimens seen in section. (75 diam.)

The specimen represented in fig. 4 is from Long Lake ; all the others are
from Petite Nation.

DIscussIon.

Prof. Duncan said that he thought the author had run the mine-
ralogists rather hard. For his own part, when he first examined
specimens of Hozoon he had come to the conclusion that they were
ancient Foraminifera with Nummuline peculiarities; and since he
had acquired a more intimate acquaintance with fibrous minerals
and serpentines, he found himself more than ever confirmed in this
view. The discovery of isolated masses was very interesting, seeing
that, whether they were separated fragments or distinct organisms,
they still showed the Nummuline structure. Prof. Duncan com-
pared the habit of growth of Hozoon to that of the Nullipores, and
suggested that it would be more philosophical to refer both the
latter and the Foraminifera to Hiickel’s group “ Protista.”

Mr. Eruertmer remarked upon the singular fact that whilst, as
a general rule, we were disappointed in obtaining instructive sec-
tions of Hozoon, we had only to go to Dr. Carpenter to see sections
which seemed to be convincing. He thought the difference of
opinion that prevailed as to the nature of Hozoon was due mainly to
the difficulty that certainly existed of procuring specimens to show
the so-called tubuli and stolons. He stated that he had received
from Jersey specimens which at the first glance he said were like

Arde

Hs

Quart. Journ .Geal. Soc .Vol.

Z

ee pease:

badare i) Verna ag Oa Yeon I (Smt RAN

eer RnM tee Rhee

me
reewnnne

Sas og a

ny PLN Me aD

at
has AEN ae

y=.
ever mana TT

8

s
ne

Mintern Bros .imp.

GH Ford hth.

as ho
af
°
i
i
ai
Ye
‘
‘
*
’
?

: i
pit

a a eh a ep teeas seats |

Pray nad
ys Hse pares ng ery

EOZOON CANADENSE AT COTE ST. PIERRE. 75

Eozoon; and on getting sections of these made, organic structure
at any rate seemed to be present. He had specimens of Stromato-
pora so like Hozoon in external appearance, that Sir W. Logan on
seeing them immediately mistook them for the latter.

Prof. Ramsay said that he was pleased to find that there might be
some relation between Zozoon and Stromatopora. For his own part
he did not see how the structure described as Hozoon could have
been formed, unless it was of organic origin.

Mr. Evans remarked that, it seemed to him, one of the most
interesting points of the paper consisted in the indication (he
believed, for the first time) of the occurrence of possibly separate
organisms associated with Hozoon. They apparently bore a close
resemblance to Globigerina; and, considering the conditions of
fossilization, their presence seemed to confirm the notion of the
organic and indeed Foraminiferal origin of Eozoon.

76 T. G. B. LLOYD——NOTES FROM THE STATE OF NEW YORK.

9. Grotocicat Notzs from the State of New York.
By T. G. B. Luoyp, Esq., C.E., F.G.S. (Read June 23, 1875.)

[ Abstract. ]

Tue following notes of various geological facts occurring in the
northern part of the State of New York were collected cui a
residence there some five or six years ago.

Grooves and Channellings, Black River, Watertown, Jefferson
County.

The grooves and channellings which are seen cutting into the
ledge of birdseye limestone at the base of the bank of sand, gravel,
and boulders, pass obliquely across the bed of the river.

Fig. 1.—View of the Pulpit Rock, near Oxbow, Jefferson Co., New
¥ ork, showing remains of Giant's Kettles in Laurentian Granite

_ ———<——<—
0510 20 30 40 50
We tt tt ET

Further down stream, beyond the railroad bridge, the channel-
lings were discontinued; but from the south side of the river they
continued in a south-west direction to Lake Ontario, a distance of
about 10 miles, according to Dr. Emmons.

Glacial Phenomena of the District around the village of Theresa,
Jefferson Co.

The surfaces of the Potsdam Sandstone and of the Laurentian
rocks, between Philadelphia on the south-east and the village of
Redwood on the north-east of Theresa, are grooved and striated in
a direction north-east and south-west, corresponding with the longer

T. G. B. LLOYD—NOTES FROM THE STATE OF NEW YORK. Th

axes of a number of small lakes which lie between Theresa and
Redwood. Long parallel ridges, from 100 to 200 feet in width, of
Laurentian gneiss are seen on the high ground between Philadelphia
and Theresa, running in a direction about N.N.E. and 8.8.W., with
the strike of the beds. They rise some 40 or 50 feet above the sur-
rounding plain, and are distant from each other about 7 mile. The
beds composing the ridges vary in dip from about 75° to 90°. The
surfaces exhibit well-marked grooves and strie. Hog’s-backs of
gneiss, with their steeper ends facing the south-west, rise here and
there above the surface of the ground. A stiff, tenacious sandy
clay (“ Till;” D in table below), containing rounded boulders, chiefly
of local origin, forms a capping to the ridges.

In the valley of Indian River, below the village of Theresa, the
base of the cliff is grooved in a manner somewhat similar to that
observed on Black River.

Tabular List of Beds.

A. Light red dry sandstone (laminated) ; containing thin seams of carbonaceous
matter, and rounded lumps of clay in layers.

B. Dark brown and bluish-coloured sandy clays (laminated), containing occa-
sional seams of light red sand, without boulders.

B!. Dark red unstratified clay without boulders.

C. Beds of mainly unstratified gravel, clay, and sand, containing boulders.

D. A tough sandy clay with rounded and some angular boulders (see Croskey’s
definition of boulder-clay). The bed-rock of sandstone or gneiss is
rounded, smoothed, and grooved, and marked with fine parallel striz.

The beds (A) occur at the head of the valley near Theresa,
about 4 mile south of the entrance of the gorge of Indian River.
They were about 20 feet in thickness, and rested partly on the bed-
rock of gneiss and on the surface of the bed B. ‘The sand was
laminated, and showed in places oblique-current action ; besides the
thin seams of carbonaceous matter, it contained seams of rounded
nodules of clay, not concretionary. They varied in size from a
walnut to a duck’s egg; the clay composing them was like that of
bed B.

It is probable that B and B' are only varieties of the same deposit.

C and B only differ in the relative proportions of their constituents.

Summary of preceding Observations.

1. The persistent direction of the groovings and striations of the
surface rocks over a considerable space.

2. The coincidence of the directions of the groovings &c. with
the longer axes of most of the lakes of the district, and with the
general course of Indian River.

3. The local character of the boulders.

4, The directions and forms of the parallel ridges and hog’s-backs
or roches moutonnées.

The ridges and troughs are quite distinct from the undulations in
the Laurentian group so common in other parts of North America.

78 T. G. B. LLOYD—-NOTES FROM THE STATE OF NEW YORK.

Giant's Kettle near Oxbow, Jefferson Co.

The cliff on the face of which the section of the kettle is exposed,
has been named the Pulpit Rock, from the circumstance that the
preachers of old used to hold forth from the interior of the kettle
during their open-air meetings.

The drawing (fig. 1), for which I am indebted to the kindness of
Mr. Ruddle, F.G.S., was made from a rough sketch taken on the spot.

The rock is of Laurentian granite, and is situated on the western
side of a dry valley 450 feet wide. Its upper surface is about 70 feet
above the road, and about 100 feet above the river Oswegatchie.

Besides the large kettle there were faint outlines of two others,
situated at a higher level than the principal one.

Fig. 2 is a section through the centre of the kettle in a direction
corresponding with the face of the rock. On the line C D it formed
about three quarters of a circle; on A B it was nearly semicir-
cular. A spiral groove ascended from right to left, as high as the

Fig. 2.—Section of Giant’s Kettle in Laurentian Granite in the
Pulpit Rock near Oxbow. (Scale +; inch to 1 foot.)

Scale of

LEA ef ae feet.
Z By
Zi -3
ZW -4
w LA, ~
Z a:
@ -9
® -10
Z mT
Z -12
LZ =i3
Z -l4
ZZ -15
eZ :
Zz 80 4
3 ae ~18
Z -19
ZW 22
ZA =38
ZQ 24
GZ r =—25
Zijxe oo TO O---.--- 26
QZ LZ GZ =
Cz g -29

point F. The talus outside mainly consisted of the large mass of
stones which formerly occupied the bottom of the kettle. Its
resemblance to the kettles figured in Quart. Journ. Geol. Soe. vol.
xxx. pp. 757, 759, figs. 4 & 5, will be apparent.

Flowerpot-shaped Blocks of Sandstone, Theresa, Jefferson Co.

The quarry from which the blocks were extracted was situated on
a high cliff of Potsdam sandstone, overlooking the gorge of Indian

T. G. B. LLOYD—NOTES FROM THE STATE OF NEW YORK. 79

River. The beds of Potsdam sandstone passed from a fine-grained
grit into a quartzite, the grains of which could not be distinguished
without a lens. The upper surfaces of the beds were ripple-marked.
At a depth of about 6 or 8 feet below the surface the peculiarly
shaped blocks, or columns, were discovered. I observed a few lying
about, which had been extracted by the quarrymen; one of the
largest was 2 feet high, 2 feet in diameter at upper surface, and 14
foot on the lower surface. They were broken off at the lines
separating the beds of sandstone. They corresponded exactly in
texture and colour with the surrounding rock. On the edges of the
blocks were thin layers of oxide of iron, easily detached by a blow.
The surrounding rock was here and there streaked with the same
substance. A short distance away from the quarry was a large
mass of rock of a similar shape, and about 5 feet in diameter, which
had apparently been dislodged from a hole close by it. There were
no markings on the surface of the blocks. Prof. James Hall, of Al-
bany, informed me that the true nature of the blocks was doubtful.

80 T. BELT ON THE DRIFT OF DEVON AND CORNWALL.

10. The Drirt of Devon and Cornwatt, its OrteIn, CoRRELATION
with that of the Sourn-East of EneLanp, and Pace in the
GractaL Serres. By Tuomas Ber, Esq., F.G.S. (Read No-

vember 3, 1875.)
[ Abridged. |

1. Intropuction.—If we separate the southern counties from the
rest of England by a line following up the Thames to near Ciren-
cester, then along the southern boundary of Gloucestershire to the
Severn at the mouth of the Avon, we get an irregular strip of
country, shaped somewhat like a fish, of which Kent forms the
head, and Cornwall the tail. Excepting, perhaps, some of the
hills of Somersetshire, this part of England appears to have been
entirely free from the action of land-ice in the Glacial period;
and there are no glaciated rock-surfaces, no true till, and no mo-
raines. Along with, and, as I believe, in consequence of, this
freedom from ice-action, there is also an utter absence of those frag-
mentary marine shells which, further north, are found in drift on
both the eastern and western coasts, where the great streams of ice
from the north, after crossing arms or channels of the sea, advanced
upon the land—and the presence of which, as they are never found
above heights to which the circumpolar ice reached, and for other
reasons I have urged, is due, not to the submergence of the land
beneath the ocean, but to portions of the ocean-bed having been
carried up by the rising ice*.

The purpose of the present paper is to describe the surface-geology
of the counties of Devon and Cornwall, to examine the theories that
have been proposed to account for the origin of the drift beds, to give
my own views on the same question, and to support them by apply-
ing the theory to the drift beds of the whole of the south of
England.

2. Drirr Beeps or Devon anp Cornwatt.—The superficial deposits
of Devon and Cornwall consist of two divisions :—a. Upland de-
posits (on the tops and sides of hills, mostly more than 300 feet above
the present sea-level, and reaching up to about 1200. feet; these
consist of gravels, clays, and transported boulders left in frag-
mentary patches, and evidently have had at one time a much
greater extension) ; b. Lowland deposits (gravels, clays, and boulders
spread out in more persistent and regular beds at low levels; they
are most strongly developed in the valleys, but are not confined to
them).

a. Fpland deposits—The whole of the surface of Dartmoor is
sprinkled over with large blocks of granite, some of which are of
enormous size. On the road from Tavistock to Princetown, about a
mile west of Merriville bridge, one of the numerous excavations
made in search of stream-tin exhibits the following section :—

* Nature, May 14 & 28, 1874.

ss

T, BELT ON THE DRIFT OF DEVON AND CORNWALL. 81

Fig. 1.—Diagram Section of Excavation made in search of Stream-tin.

Excavation.

a. Soft decomposed granite. 6. Stratified granite sand. c. Gravels and sand
with boulders. ad. Large blocks of granite lying on the surface.

Many other sections in the old stream-tin workings and in the
banks of rivulets show a similar succession, If the blocks of granite
had been left by the decomposition of softer portions of the bed-rock,
they would have rested upon it. Neither can they have slipped
down from the “tors” that crown many of the hills; for they are
not confined to the slopes, but are spread out over the level ground
as well, neither are they concentrated at the bottom of slopes,
as they would have been by such an action. Here and there two
or three, or even a dozen may be crowded together ; but generally
they are separated from each other.

Near to Okehampton, on the northern’part of the moors, the bed-
rock is composed of hardened shales and sandstones. The sides of
the hills are covered with a stony clay, sometimes as much as 30
feet in thickness, containing many large angular blocks of stone,
especially near to, and on, the surface. Amongst these, blocks of
granite, that must have been brought at least two miles, are not
unfrequent.

On the eastern side of Dartmoor, Mr. G. W. Ormerod has noticed
beds of gravel reaching up to about 900 feet above the sea, and
has given instances of many transported blocks of granite and Car-
boniferous rocks. Thus, to the east of Cranbrook Castle (1110 feet
above the sea) and near Wooston Castle, large transported granitic
blocks overlie the Carboniferous beds. On the Newton and Moreton-
Hampstead Railway, fragments of granite are spread over the Car-
boniferous strata; and gravel formed of elvanic and Carboniferous
rocks occurs at Riddy Hill, its nearest point of derivation being on
the opposite side of the valley*. Mr. George Maw has described
the occurrence at Petroclistow, near the centre of the county, of an
isolated bed of gravel, composed almost entirely of the detritus of
Dartmoor granite, twelve miles distant from the nearest granitic
mass, and considers that it can only be accounted for by an amount
of submergence covering the whole of the ridges‘.

In one of the many able communications Mr. D. Mackintosh has
made on the Glacial beds, he has advanced arguments to prove that

* Quart. Journ. Geol. Soc. vol. xxiii. p. 418, t Ibid, vol. xx. p. 451.
Q.J.G.8. No. 125, @

82 T, BELT ON THE DRIFT OF DEVON AND CORNWALL.

the blocks of granite spread over Dartmoor must have been carried by
currents of water and floating ice*. The experienced glacialist Mr. J.
F. Campbell also believed that the denudation of the tors was due to
floating icey. Many miles to the east, on the Haldon Hills, gravels
are found containing pebbles of granite and other crystalline rocks,
mingled with chalk flints and chert. At Little Haldon Mr. Mackin-
tosh found the gravels 10 feet thick, extending down the sides of
the hill, proving that it had its present contour before they were
deposited. The top of Little Haldon is about 800 feet above the
sea ; and the gravels extend over its summit.

Gravels with foreign pebbles are found on Straightway Hill, and on
the summits of the hills surrounding the vale of Charmouth?. Mr.
H. B. Woodward has kindly furnished me with much information
respecting the drift-gravels that cover the Black-Down Hills, first
described, I believe, by Dr. Buckland.

The bed-rocks belong to the Greensand formation ; and on the tops
of the hills, at heights of from 600 to 700 feet above the sea, occur
patches of gravel with large, well-worn boulders of chert, large
pebbles of quartz, rolled flints, and a few boulders of quartzite§.
Mr. Woodward, in a letter to me, says:—‘‘ The occurrence of old
pebbles in a clayey drift on the tops of the Greensand heights is the
more important because, on the palozoic rocks of Devonshire, it is
often very difficult to decide between a drift gravel and the disin-
tegrated conglomerates of the New Red series.”

In Cornwall, upland gravels occur on Crousa Downs, near St.
Keverne, and blocks of syenite are spread over the surface of them ||.
On the isolated hill of St. Agnes, on the northern coast, Mr. Bene-
dict Kitto has described gravels containing angular and waterworn
stones covered by clay, and that again by rubble and rounded
pebbles §].

When the nature of the evidence is fully understood, there is not
likely to be any difference of opinion respecting the agent that trans-
ported the travelled blocks of Devon and Cornwall. The probability
that they have been carried by floating ice is great, and is enhanced by
the occurrence of rounded pebbles. ‘These require for their forma-
tion the presence of water; and granting the existence of this up to
the heights at which they are found, we may easily imagine that, in
the Glacial Period, coast ice would form round the shores in winter,
and float off pebbles and boulders in the spring.

b. Lowland deposits—The Upland deposits are separated from
those of lower levels by tracts over which no gravels are now found.
They are distinguished from the Lowland deposits by evidence of
tranquil deposition, whilst those in the lower grounds occur spread
out in wide sheets, exhibiting many signs of sudden and tumultuous

* Geological Magazine, vol. iv. p. 399. t Frost and Fire, vol. ii. p. 220.

{ Sir H. T. De la Beche, Report on the Geology of Cornwall, p. 395.

§ Mr. H. B. Woodward, Geological Magazine, 1874, p. 335.

|| Rev. E. Budge, Trans. Roy. Soc. Cornwall, vol. vi. p. 97; and Sir H. T.
De la Beche, Report on the Geology of Cornwall, p. 396.

€| Miners’ Assoc. Cornwall and Devon, August, 1874.

f, BELT ON THE DRIFT OF DEVON AND CORNWALL. 83

formation. On the hills large stones are found scattered over the
surface; in the Lowland deposits they occur at the bottom of the
gravels. In Cornwall, the stream-tin is found at the bottom of the
gravel, along with rocks of quartz of great size. In 1830, Mr.
Joseph Carne urged that the occurrence of the stream-tin invariably
at the bottom of the gravels proves that the whole of the materials
had been in motion at once, and that these deposits furnish the
strongest evidence of a sweeping inundation *.

In this view Mr. Carne followed Prof. Sedgwick, who, fifty years
ago, had satisfied himself that the gravels had been spread out by a
great débacley. Mr. Carne supported this theory with great ability,
and referred to the fact that all the productive stream-tin works
were situated in valleys opening to the south, whilst the richest tin
veins are situated near the northern coast, and concluded that a deluge
of water from the north had swept the tin into the southern valleys.
Sir H. De la Beche fully concurred in this conclusion, and furnished
additional arguments in its support. He says that any detritus that
existed previously to the deposition of the stanniferous gravels ap-
pears to have been “fairly washed out by a mass of water rushing
over the land, rolling and driving various loose materials before it,
and allowing the tinstone, from its greater specific gravity, to be
strewed along the bottom” +.

In Devonshire, in every valley there is a repetition of the same
facts—a scouring out of their upper reaches of all loose materials,
which are spread out in great sheets towards their mouths. In the
valleys of the Teign and its tributaries, there are some of the largest
accumulations of gravel in Devonshire; and we have distinct evi-
dence that it was not deposited during the excavation of the
valleys, but that the latter were formed long before Miocene or pre-
Miocene times; for at, and near, the junction of the Bovey river
with the Teign are found the Bovey-Tracey lignites of Miocene age,
lying in the valleys and covered by the lowland gravels. The Bovey-
Tracey lignites have been bored through to a depth of 200 feet with-
out the bottom of the old pra-Miocene valley being reached. The
drift-beds extend up the sides of the valleys, and to the tops of the
ranges bounding them.

3. THEORIES Or oRIGIN.—The facts to be explained, not only in
Devon and Cornwall, but, as we shall see, over the whole of the
south of England, are:—(1) gravels containing travelled boulders,
on the sides and tops of hills up to about 1200 feet above the sea; (2)
denudation of the surface-gravels from an intervening tract between
the Upland and Lowland deposits; (3) in the lowlands, especially
in the valleys within 100 feet above the level of the sea, a wide
spreading-out of gravels that show signs of sudden and tumultuous
deposition.

Subaérial denudation could not produce the hill-gravels containing
rocks brought from a distance. Nor could the deposits lying on the

* Trans. Roy. Soc. Cornwall, vol. iv. p. 47.
t Annals of Philosophy, vol. ix. p. 241.
t Geology of Cornwall, pp. 396, 398, & 410.
G2

84 T. BELT ON THE DRIFT OF DEVON AND CORNWALL.

sides of the valleys have been left there during their excavation ; for
as we have seen, the configuration of the country dates back at least
to early Tertiary times.

Water must have been present to form the gravels of Dartmoor
up to about 1200 feet above the sea, and also to allow ice to trans-
port the erratic blocks. Was the water that of the ocean, or of a
great freshwater lake? With respect to the area in question, there
is this insuperable objection to the theory of marine submergence,
that over the whole district no marine beaches, and not a single
marine organism has been found, excepting within a few feet of the
present sea-level. Nor can sea-shells have once existed in the de-
posits, and been since destroyed; for mammalian remains and land
and freshwater shells are preserved ; and any agency that would have
obliterated the traces of the one, would not have spared the others.

After much study of this question, not only with regard to the
part of the country now under consideration, but with reference to the
erratic blocks and the stratified and terraced gravels of nearly the
whole of Northern Europe, the only theory I can find that meets
all the facts of the case and deals with all its difficulties, is :—
(1) that at the height of the Glacial Period the bed of the Atlantic
was occupied by ice flowing from the north-west, from the direction
of Greenland, reaching the coast of Europe either at Brest or
some other point further south, and damming back the whole of the
drainage of northern Europe into an immense lake of fresh water,
which, at its greatest extension, reached to a height of at least 1200
feet above the present level of the sea; (2) that this lake was gra-
dually lowered for some hundreds of feet, probably by the deepening
of a channel of outlet, and then suddenly and completely drained
by the breaking away of the ice-dam causing an enormous flood or
débicle ; (3) that the complete denudation of the detritus from
the low grounds was prevented, and its outspread in great sheets
caused, by the outflow of the waters being checked by the immense
area drained through the single outlet of the British Channel.

In some Alpine lakes caused by glaciers, the waters periodically
break away, and are again pent up by the readvance of the ice; and
so it is possible that a series of floods might be caused by the great
Atlantic glacier. But I can find no evidence of more than one
debacle in the southern counties, and I am inclined to agree fully
with the conclusion that Professor Sedgwick arrived at fifty years
ago—that “ diluvial torrents have swept over the land, driving be-
fore them enormous masses of gravel,” and that “all the diluvial
detritus originated in the same system of causes, which, having pro-
duced their effects, were never repeated” *. If, however, I were
discussing the origin of the glacial beds north of the Thames, I
should have to show that although there is evidence of only one
sudden and tumultuous discharge of the waters, yet, after the
débacle so caused had occurred, they rose again, but were this time
more quietly lowered.

Completely as this theory seems to explain the production and

* Annals of Philosophy, vol. ix. p. 248. |

T. BELT ON THE DRIFT OF DEVON AND CORNWALL. 85

distribution of the surface deposits, I was not led to adopt it on
that account alone, but because, in addition, I found abundant evi-
dence that the ice from the north had moved down the bed of the
Atlantic, and from the direction of Greenland as its chief gathering-
point. Thus Newfoundland is glaciated from the north-east, and -
the ice advanced up the valley of the St. Lawrence, and forced the
drainage from the great lakes to run into the basin of the Missis-
sippi. Even so far westward as the Lake of the Woods, we learn
from Mr. G. M. Dawson’s elaborate and instructive report, that the
ice moved persistently from the north-east*. The whole of Cape
Cod is a great moraine, proving that the bed of the Atlantic was
occupied by ice several degrees of latitude further south than is re-
quired on the European side of the basin to block up the English
Channel.

In the Old World I found that the ice had advanced upon Siberia
from the north, had overridden Iceland, the Hebrides, and the north
of Scotland from the north-west, and at the extreme south-west of
Ireland was 2000 feet thick. Mr.T. F. Jamieson has clearly shown
that the ice that glaciated Caithness came from the north-west, and
not from the south-east, as had been supposed by Mr. Croll. The
Shetland Isles are too small to have borne local glaciers; yet we find
from the interesting observations of Mr. C. W. Peach that they are
hugely glaciated. On Lerwick he found the grooves and striz
pointing north and south, and that the drift had evidently come from
the north. In North Uist he found great ruts and scorings pointing
W.N.W. On reaching the top of Muckle Heog, 500 feet above the
sea, he found the W.N.W. end vertical and polished down to at least
150 feet from the top. To what agent can we ascribe the glaciation
of these islets, excepting an immense body of ice moving from the
north-west t ?

Greenland is now covered with ice ; and this theory assumes that in
the Glacial Period ice advanced from that direction, and flowed furthest
down the deepest channel, which was the bed of the Atlantic. That

* Report on the Geology of the region in the vicinity of the 49th parallel,
1875. Map showing Glacial markings. I am indebted to Prof. Hall for infor-
mation about Cape Cod.

t For Caithness see Quart. Journ. Geol. Soe. vol. xxii. p. 261; and for the
Shetland Isles, Geological Magazine, 1865. So long ago as 1852, Dr. Robert
Chambers, in a remarkably able paper read before the Royal Society of Edin-
burgh, and published in full in the ‘ Edinburgh Philosophical Journal’ for 1853,
urged that the whole of the glaciation of Scotland could not have been produced
by ice gathering on the mountain-chains, as they themselves had been overrid-
den, and that the ice that had left the highest markings had flowed from the
north-west. He argued that both the hills and valleys, including the lake-
basins of Sutherland and Ross-shire were due to the erosion of the rocks by an
immense body of ice that had moved down upon Scotland from the north-west.
Mr. T. F. Jamieson, in his valuable papers on the Glacial Period, published in
the Journal of this Society, has also mentioned many facts pointing to a flow
of ice from this direction ; but Mr. James Geikie, in the otherwise very com-
plete account of the glacial phenomena of Scotland in his ‘Great Ice Age,’ has
not discussed this important question, though many of the facts mentioned by
him, such as the blocking-up of the Clyde valley, would receive an easy solution
by this theory.

86 ' T. BELT ON THE DRIFT OF DEVON AND CORNWALL.

ice from the north could have scored the eastern sides of the Ameri-
can hills, nearly at right angles to the slope of the continent, or
have been forced up the valley of the St. Lawrence, before the ocean-
depression was filled, seems as improbable as that it should flow
‘ along the ridges of the Alps, and not down the valleys. Doubtless
the culmination of the Glacial Period was preceded and followed by
both local and continental ice systems; but, in addition to these
agencies, there is required, to explain the grander phenomena, the
action of circumpolar ice filling the northern part of the Atlantic
basin down to about latitude 39° on the American side, and about
latitude 40° on the European side.

The main feature of the Glacial theory that I advocate is, that
the glaciation of North-eastern America and North-western Europe
was principally due, not to local and continental ice, but to that
filling the northern part of the Atlantic basin, and blocking up the
drainage of the continents as far as it extended. The theory re-
quires no interglacial periods, no oscillations of the earth’s surface ;
yet I know not of any fact in the composition and distribution of
the glacial beds on either side of the Atlantic that does not find in it
a natural and easy interpretation.

4, CoRRESPONDENCE WITH THE DRIFT OF THE SOUTH-EAST OF
Enetanp.—-It is clear that, if this theory is correct, evidence of the
existence of the great lake and of its sudden and torrential discharge
ought to be found over the whole of the vast area its waters once
covered. I have found this evidence over much of Northern Europe,
and in our own country, up the valley of the Severn, and over the
whole of the eastern coast: but in this paper I confine myself to the
south of England, as I thus avoid questions connected with the
former presence of land ice, and with what many geologists consider
evidence of marine submergence. I shall pass rapidly in review
the surface geology of the rest of the southern counties.

Dorsetshire.—Upland deposits have been deseribed on the Black
Downs,.to the 8.W. of Dorchester, at Preston and Osmington, and
in the Admiralty quarries on the top of the Isle of Portland*. The
latter are remarkable, as they overlie remains of Hlephas prim-
genius and FE. antiquus. The raised beach with marine shells at
Portland Bill is covered with the Lowland deposits, which appear to
have been suddenly and tumultuously produced. Prof. Prestwich
believes they were formed by the transient passage of a body of
water sweeping the land detritus down to the shore-line, and to a
certain distance beyond it—a conclusion identical with that arrived
at by Sedgwick and De la Beche with regard to the Lowland deposits
of Cornwall. |

Hants and Wilts.—The superficial deposits of these counties have
been admirably described by Mr. Codringtony. The gravels are
not confined to the valleys, but cover the plateaux between them up toa
height of 420 feet above the sea, and extend in width for a distance
of 20 miles. At Milford Hill, near Salisbury, Upland gravels from

* Prof. Prestwich, Quart. Journ. Geol. Soe: vol. xxxi. p. 29.
Tt Quart. Journ. Geol. Soc. vol. xxvi. p. 528.

T. BELT ON THE DRIFT OF DEVON AND CORNWALL. 87

10 to 12 feet in depth contain paleolithic flint implements, which
occur principally at the base of the deposit.

Sussex, Kent, and Surrey.—Sir Roderick Murchison very fully
described the surface geology of these counties, and followed Sedg-
wick and De la Beche in ascribing the formation of the Lowland
deposits to cataclysmal action. Of late years, however, the evidence
that abounds of violent and tumultuous deposition has been almost
ignored under the influence of the teachers of the theory of subaerial
denudation.

In the distribution of the gravels between the escarpments of the
North and South Downs there are two points to be explained—the
complete denudation of many large tracts lying above 200 feet above
the sea, and the outspread of the gravel in great sheets below that
level and especially below 100 feet above the sea. To the former
of these distinct yet complementary phenomena my attention was
first drawn by Mr. J. H. Farrer, who pointed out to me from the
top of Hawkhurst Down the remarkable absence of flints from
between the chalk escarpment and the top of Leith Hill, and urged
that if the Chalk that once covered the greensand slope had been
removed by subaerial denudation, the flints would still have been left
behind, and asked what force had cleared them away. In every
county of the area to which this paper is confined there are Upland
gravels, with transported stones, showing signs of gradual and
tranquil deposition, and Lowland gravels affording evidence of sud-
den and tumultuous formation. Nowhere are there any marine
remains or old sea-beaches, excepting near the present shore-line.
In Cornwall and in Devon, as well as in Sussex and Surrey, inde-
pendent observers, including the most illustrious of our geologists,
have published their convictions that great currents of water poured
across the country. Ido not know what other geological conclusion
has more evidence to support it. Yet because no acceptable theory
has been advanced to account for the origin of such a tumultuous
and overwhelming flood, the evidences of its occurrence have of late
been neglected. Can we have a stronger argument of the necessity
of a good theory of origin to enable us to appreciate the importance
of the facts with which we have to deal?

5. CLASSIFICATION OF THE GLACIAL PHENOMENA.—I have in former
papers* urged that Mr. Alfred Tylor’s theory that the sea-level
was greatly lowered in the Glacial period through the vast quantity
of water that, in the form of ice, was piled up around the poles is
borne out by much geological, physical, and biological evidence.
Possibly a point of departure for the commencement of the Glacial
period may be found in the first evidence of the lowering of the sea-
level. Thus, in the Weybourne sands, the Portland-Bill marine
beds, and the Bridlington Crag we have evidence of the sea rising
to about the same level as it now does; and these deposits are
clearly of older date than the overlying glacial beds.

The first stage of the Glacial Period, as thus defined, is marked by

* Naturalist in Nicaragua, p. 266; and Quarterly Journal of Science, Oct.

88 T. BELT ON THE DRIFT OF DEVON AND CORNWALL.

the older forest-beds and the immigration of the great mammalia,
indicating that the sea had retired from the British Channel and the
German Ocean. A great river appears to have run to the south-
west through this now submerged area, on the banks of which and
of its tributaries paleolithic man and the great mammalia lived, the
latter frequenting low marshy tracks near to the streams, the former
higher grounds more distant from them. That this river flowed
southward, and not to the north, as supposed by Mr. Godwin-Austen,
is indicated by the fact that the fauna is the same on the Somme and
the Thames, which would scarcely have been the case if the Straits
of Dover then formed the division between watersheds draining to
the north and the south. This apples withparticular force to the dis-
tribution of the hippopotamus. Its remains are found not only in all
the southern parts of the area, but as far northas Yorkshire. There
is evidence that the winters were more severe then than now; and it
would have had no refuge from them to the north of Dover if the
rivers there ran northwards; but if they drained into a great stream
running southward as far the Bay of Biscay, it might easily have
retreated down this in the autumn and reascended in the spring.

The presence of southern shells, such as Cyrena fluminalis, Unio
littoralis, and Paludina marginata, in the old mammalian gravels
both to the north and to the south of the Straits of Dover, is evidence
nearly as strong that the streams in which they lived all belonged
to the same watershed, and that the course of the main river was
southward. For this great stream I propose the name of the Ger-
manic River, as its principal drainage-area was the bed of the
present German Ocean and the waters discharging into it.

The second stage of the Glacial Period is marked by the continued
advance of the ice from the north, and, possibly, by the retreat of the
southern fauna and the arrival of the Arctic mammals, whilst the
woolly rhinoceros and the mammoth still roamed in summer over the
area that had formerly been their winter retreat.

In the third stage the Atlantic Glacier reached the coast of Europe,
and blocked up the English Channel and with it the drainage of the
whole of Northern Europe, forming a great continental lake, over
which floated icebergs carrying boulders from the Scandinavian
of mountains.

In the fourth stage the barrier of ice blocking up the English
Channel was suddenly broken away, causing the tumultuous discharge
of the waters of the great lake and the consequent outspread of the
Lowland deposits, including the Middle Glacial sands and gravels
Norfolk and Suffclk.

In the fifth stage the Atlantic glacier again advanced, and the
great lake was reformed; but its discharge this time was more gra-
dual. ‘To this period belongs the formation of the Upper Boulder-
clay of Norfolk and Suffolk, which I have not been able to satisfy
myself is represented south of the Thames except by the “Trail” of
Mr. Fisher; but some of the Upland deposits may belong to this and
not to the third stage.

In the sixth and last stage, I consider that the Atlantic ice had

T. BELT ON THE DRIFT OF DEVON AND CORNWALL. 89

rapidly retired as far,back as the extreme north of Scotland ; but the
sea had not yet returned to its former level, and the British Isles
were still joined to the continent. To it I assign the last great
forest-period and the arrival of Neolithic man with the associated
fauna from the continent.

Postscript.

January 17, 1876.—Since this paper was read before the Society
I have again crossed Europe and examined the southern limits of
the northern drift as far as the cliffs of glacial clay around the
shores of the Sea of Azoff. I now attach much more importance to
the second advance of the Atlantic glacier, and believe it marks the
culmination of the Glacial Period. The restriction of the middle
glacial sands and gravels to heights not exceeding, if reaching, 500
feet above the sea, leads me to think that the first great Huropean
lake drained around the hills of Brittany into the valley of the Loire,
and that the principal distribution of the Scandinavian drift and the
formation of most of the upland deposits belong to the period of the
second great lake, when the col to the east of Brittany was also
blocked up and the waters drained through the Black Sea to the
Mediterranean, cutting out gradually a passage through the Darda-
nelles that had not before existed.

DISCUSSION.

Mr. Hicks stated that he had noticed that the glaciation at St.
David’s is from the north-west. He had already stated before the
Society his opinion that there had been depressions proceeding from
a point in the Atlantic, probably not far from the coast of South
America, to the north-west and north-east; and this might perhaps
have something to do with causing a flow of ice from Greenland to
the south-west in North America and to the south-east in Europe.

Rev. O. Fisher wished to know what would be the area of the
great freshwater lake supposed to be produced by the damming
action of the great Atlantic glacier.

Mr. Brrr stated that the area blocked up by the ice would be
about 400,000 square miles (2000 x 200), and would include all the
region drained by the present northern rivers.

Prof. Hueues wished to know why the waters could not drain off
by way of the Black Sea, and why the advancing Atlantic glacier
should be supposed to stop just at the western poiut of Cornwall.
He could discover no evidence of there ever having been a lake
such as the author described. The drift gravels were the result of
all the agencies of denudation which had ever been at work; and the
boulders at the bottom of the low-ground drifts were probably due
to the fall of débris from the summits. The boulders of the drift,
if the author’s theory were true, ought to consist of Greenland rocks,
whereas they were really of local origin. With regard to the direc-
tion of glaciation in Britain and Northern Europe being from the
north-west, he could not agree with the author. In many places

90 T. BELT ON THE DRIFT OF DEVON AND CORNWALL.

glaciation might be observed running in every direction; and it was
not fair to note only certain striz, and neglect those which were
not in favour of a foregone conclusion. He thought that Mr. Camp-
bell had shown clearly that the glaciation of Ireland took place from
the north-east.

Mr. Moeeriver remarked that a very flat country extended across
the continent of Europe from England to the Black Sea, and thought
that in that direction there was no land sufficiently high to form the
boundary of such a lake as that required by the author’s theory.

Rev. T. G. Bonney said that, in addition to the difficulties which
Prof. Hughes had mentioned, four others at least occurred to him :—
that the barrier to Mr. Belt’s lake was defective between the high-
lands of Brittany and the Auvergne; that the ice in its course from
Greenland would have to cross a part of the Atlantic where the
depth approached 2000 fathoms, which seemed to demand an incon-
ceivable accumulation in that country; that under such circum-
stances Wales, Scotland, and Scandinavia must have had their own
ice-systems ; and that to reach Scandinavia (which certainly had its
own ice-system), this great sheet must have crossed the Lofoten
Islands, yet all the higher hills in these were remarkably sharp and
broken. Further, in regard to what Mr. Belt had said about the
lowering of the general level of the sea, it must be remembered that
such an ice-cap would raise the level in the hemisphere where it
occurred.

Mr. Brrr, in reply, said that he did not want the ice to stop at
Cornwall, but that his statement as to its limits was founded on
observed marks of glaciation. He thought the absence of marine
remains throughout the drifts of the northern plains of Europe was
a highly important and suggestive fact. With regard to the glacia-
tion of Ireland, he remarked that the ice flowing south-east from
Greenland would strike against the high lands of Scotland and
England, and be turned back over Ireland. The lowering of the
sea was not absolutely required by the necessities of the paper ;
but if the accumulation of ice took place simultaneously at both
poles, the sea must necessarily be greatly lowered.

D. MACKINTOSH ON THE CEFN AND PONT-NEWYDD CAVE-DEPOsITs. 91

11. On the Corrzetation of the Deposits in Cern and Pont-NEwypp
Caves with the Drirts of the Norru-west of EneLanp and Watzs.
By D. Macxinrosn, Esq., F.G.8. (Read June 23, 1875.)

In the present state of Posttertiary geology it is of very great im-
portance (as may be inferred from the Presidential address just pub-
lished, May 1875) that some one should attempt to correlate the
deposits in caves with the glacial drifts of the neighbourhood. I
therefore venture to bring before the Society a brief statement of
the results of observations lately made in and around the Cefn and
Pont-newydd Caves, Denbighshire. These caves are situated near
to each other in the face of a limestone escarpment, on the north
bank of the river Elwy. The Pont-newydd cave has been described
by Professor M‘Kenny Hughes and the Rev. D. R. Thomas, in the
Journal of the Anthropological Institute (vol. iii. p. 387), and by
Mr. W. Boyd Dawkins, F.R.S8., in his work on Cave Hunting. By
these writers the cave-deposits are regarded as Postglacial. The
best account of Cefn Cave, as it existed before the deposits were
nearly all cleared out, is perhaps to be found in Mr. Joshua
Trimmer’s ‘ Practical Geology,’ published in 1841%*, the following
being the order of succession therein stated or implied :—

1. Sand, silt, and marl, with sea-shells in one or more places (uppermost).

2. Loam, with*angular fragments of limestone and bones, filling the cavern
nearly to the roof (dzlwviwm of old authors).

3. Crust of stalagmite.

4. Loam, with smooth pebbles, bones, teeth, and fragments of wood (lowest).

Mr. Trimmer believed that the lowest of these deposits was intro-
duced, before the glacial submergence, by the adjacent river while
flowing at a considerably higher level than now. But that the
river-channel must then have been excavated to a level as low,
if not lower, than at present, is evident from the fact that the
Upper Eoulder-clay extends down to, and in some places runs
under, the river-bed, in a manner showing that here (as elsewhere
in the north-west of England and Wales) the river, since the
glacial submergence, has been principally occupied in reexcavating
its choked-up channel. Trimmer likewise believed that the deposit
with sea-shells was introduced by the sea through a fissure in the
roof of the cave.

From the nature and sequence of the deposits in Pont-newydd
cave (a considerable portion of which has not yet been cleared out),
compared with what I have seen of the remnants still visible in
the Cefn cave, and from a further comparison of the facts thus
obtained with accounts given by Mr. Trimmer, Mr. J. Price, M.A. (of
Chester), Professor Hughes, Mr. Boyd Dawkins, &c., I have been
led to regard the following as the sequence of the beds (order
descending) :—

* See also paper by Rey. E. Stanley, Proc. Geol. Soc. vol. i. p. 402.

92 p. MACKINTOSH ON THE CEFN AND PONT-NEWYDD CAVE-DEPOSITS.

1. Coarse sand charged with minute fragments of sea-shells, still found adher-
ing to one side of a rising branch (ascended by steps) of the Cefn cave*.
In the Pont-newydd cave a bed of very fine stoneless clay.

2. Clay, with angular and subangular fragments of limestone, a few polished
fragments and pebbles of limestone, likewise a few pebbles of Denbighshire
sandstone and grit, felstone, &. This deposit (which contains bones of a
number of the usual cave Mammalia) is horizontally continuous with the
Upper Boulder-clay of the district (see sequel).

3. Stalagmitic crust, from Jess than an inch to 2 feet in thickness. Very little
now left in the Cefn cave; apparently absent in the Pont-newydd cave.

4. Loam, with rounded and smoothed pebbles, bones, teeth, and fragments of
bone and wood, in the Cefn cave. In the Pont-newydd cave a bed of extra-
rounded pebbles, more or less cemented by stalagmitic matter.

The coarse sand in the uppermost bed (1) in no respect resembles
the Upper Boulder-clay on the summit of the plateau above the cave;
it is not what would result from a subaerial rearrangement of the
clay ; and the proportional number of fragments of shells is very
much greater than that found in the clay t. It is therefore probable
that the coarse sand was introduced by the sea through a fissure or
fissures in the roof, as Trimmer supposed.

The clay (2) can be traced along the plain of Lancashire and
Cheshire, the coast of Flintshire, and up the vale of Clwyd. It
spreads over the gently rising ground between St. Asaph and the
Cefn and Pont-newyd caves ; and it may be seen all around the
caves, in some places filling up hollows, in others covering plateaux,
and in not a few instances clinging to the face of steep slopes, or
even adhering to narrow rocky terraces or ledges. I have been
familiar with this clay in Cheshire and Flintshire for four years,
and have therefore little hesitation in asserting that traces of it, in
an unmodified state, may be found at the entrances of both the
Pont-newydd and Cefn cayes—that in the interior of the Cefn cave,
for a considerable distance from the entrance, there are indications
of this clay haying once filled the cave nearly, if not quite, to the
roof—that in the interior of the Pont-newydd cave it maintains its
unmodified character for a considerable distance from the entrance—
and that in no part of these two caves has this clay been modified
further than what may have resulted from the dropping of calcareous
matter, from the temporary ponding back of water in the recesses
or hollows, or from accumulation within the caves under condi-
tions which may have differed from those without. The angular
limestone fragments may have fallen from the roof or sides of the
caves during the period of accumulation ; or previously fallen fragments,
along with the bones of animals, may have been washed up into the
clay by the waves of the Upper-Boulder-clay sea. It ought not to
be forgotten that in caves sea-waves are often possessed of very
ereat force, and that they are capable of insinuating themselves into
the remotest recesses.

* T had no difficulty in raking off large quantities witha hammer, It would
appear to be present in at least one other branch of the cave.

t I could see none at all in the clay of the immediate neighbourhood of the
cave, though they probably might be found, as I believe Prof. Hughes has col-
lected them from the same clay in the Vale of Clwyd.

D. MACKINTOSH ON THE CEFN AND PONT-NEWYDD CAVE-DEPOsITS. 93

The stalagmitic crust (3) must have been accumulated during a
period when conditions were favourable; and these conditions must
have varied in different parts of the Cefn cave.

The bed under the stalagmite in the Cefn cave, with its repre-
sentative in the Pont-newydd cave, is the most difficult to explain.
I am unable to say whether it most resembles an accumulation of
fine river-shingle or a raised beach. In the caves it does not occur
at the height above the present sea-level of the more typical raised
beaches. It is not confined to the caves, but may be seen at inter-
vals along the sloping north bank of the Elwy. It may likewise be
found in patches along the Elwy valley as far as the Vale of Clwyd,
where it would appear to graduate into the middle gravel and sand
which underlies the upper clay. However it may have originated,
I cannot believe with Professor Hughes that, in the Pont-newydd
cave, it was washed in through a swallow-hole, from the Boulder-
clay of the neighbourhood, by a freshwater stream. The relative
proportion of stones of different kinds is not nearly the same in the
pebble-bed and in the Boulder-clay covering the surface of the
ground at a higher level. In the pebble-bed nearly all the stones,
so far as I could see, are Denbighshire sandstone or grit. In the
Boulder-clay a large proportion of the pebbles are limestone. I
could only see one felstone specimen in the pebble-bed (though
others probably might be found). In the Boulder-clay felstone
pebbles and boulders are far from being rare*. The pebble-bed is
not confined to the cavern or the ground straight in front of it, as it
would have been if deposited by a stream flowing out of the cavern;
but it may be seen in a recess a short distance east of the mouth of
the cavern, and, as I have already remarked, at intervals further
east. Several examples of it may be found near the Cefn cave,
clinging to the rocky slope. This bed could not have been accumu-
lated before the glacial submergence, as it contains a few erratic
pebbles +, which must have been transported by ice from regions far
beyond the basin of the river Elwy.

Before endeavouring to offer a general explanation of the mode
in which the various deposits were introduced into the Cefn and
Pont-newydd caves, it may be necessary to state that the drift-
deposits of the north-west of England and the borders of North
Wales are separable into :-—(1) a lower stony Boulder-clay (with
glaciated stones), which to the south of the Mersey and in the
eastern part of Wales can only be detected at intervals; (2) a mid-
dle sand and gravel (without any large boulders or glaciated stones,
excepting a few among or near to the mountains), which extends
almost continuously over very wide areas, and often attains a thick-
ness of nearly 200 feet ; (3) an upper Boulder- or brick-clay (with

* Tt likewise contains Eskdale granite.

t+ Professor Hughes mentions the occurrence of felstone in this bed in Pont-
newydd cave. I sawa large boulder of very typical Arenig felstone, about half-
way between Denbigh and Cefn ; and I believe that a boulder which may be seen
a short distance N.W. ofthe Cefn cave must have come from the same moun-
tain.

94 D. MACKINTOSH ON THE CEFN AND PONT-NEWYDD CAVE-DEPOSITS,

glaciated stones and a few boulders), which is quite as persistent as
the sand and gravel, but which has not yet been clearly traced to a
height of more than a few hundred feet above the present sea-level.
The first indicates a cold period, the second a comparatively mild
period, and the third, I believe, a second cold period, though this
has been disputed by Professor Hughes and Mr. Kinahan. Believing
with Professor Ramsay that the great glacial submergence com-
menced before the ice-sheet or ice-sheets disappeared from the
country, that the lower Boulder-clay (so far as it is of marine
origin) was accumulated while the land was sinking, and that the
sand and gravel formation was deposited while the land was rising—
and likewise believing in an interglacial period, during the first part
of which the land was still submerged, dry land prevailing during
the second part, it follows that there must have been a second sub-
mergence, during which the upper Boulder-clay was deposited. It
is possible that the pebble-bed in the Pont-newydd cave may have
been deposited as the land was rising out of the interglacial or
Middle-Sand-and-gravel sea, and that pebbles and loam may then
have been introduced into the Cefn cave; but as the neighbourhood
could not have been inhabited by land animals until after the
emergence of the land, the bones, teeth, and fragments of wood
which were found associated with the lowest deposit in the Cefn
cave may have been washed in by rain through fissures in the roof.
After the accumulation of the stalagmite, more bones must have
been introduced, and the cave may have been temporarily inhabited
by the hyena. This state of things may have been brought to a
close by the submergence of the cave beneath the waters of the
Upper-Boulder-clay sea, which filled the cave nearly (if not, in many
places, quite) to the roof. The sand with sea-shells may have been
introduced through fissures in the roof while the plateau above the
cave was gradually rising above the sea-level.

The above imperfect explanation is the best I can offer concerning
the mode in which the different and very dissimilar deposits in the
Cefn and Pont-newydd caves were accumulated.

P.S.—As may be learned from his paper read before the Anthro-
pological Institute, Professor Hughes found Paleolithic flint imple-
ments and a human tooth, which he believed came from the bed I
have called Upper Boulder-clay, in the Pont-newydd cayve*.

* Since this paper was written, Professor Hughes has found a large frag-
ment of felstone in the lowest bed of the Pont-newydd cave.

PROF, OWEN ON A CARNIVOROUS REPTILE. 95

12. Evidence of a carnivorous Reprite (CrnoprRaco Masor, Ow.)
about the size of a Lion, with Remarks thereon. By Prof. Owen,
C.B., F.R.S., F.G.8., &c. (Read February 2, 1876.)

[Prats XI.]

SEearcuine over the residuary, more or less shapeless blocks of

matrix from the Karoo lacustrine deposits of South Africa, trans-

mitted by their discoverer the late Andrew Geddes Bain, Hsq., F.G.S.,

I came upon a lump about the size of one’s fist, in which the sole indi-

cation of organic remains was a pair of mutilated canine-shaped teeth.

Part of the outer enamelled surface was exposed to view, and
suggested that the block might contain a fragment of the upper.jaw
ot a Dicynodon. On clearing away the matrix from the teeth, how-
ever, they presented a difference of shape from the tusks in that
genus: the crown was narrow instead of round; the transverse
section was a long oval (Pl. XI. fig. 3) with the small end pointed :
when the whole of the outer side of the crown became exposed,
the shape of the best-preserved canine (2d. fig. 2) resembled that in
Machairodus, and most so that of Machairodus latidens*. On care-
fully relieving the hinder trenchant margin of the tooth from the
matrix, I was much interested in fmding that it carried the resem-
blance to the canine of the extinct feline mammal to correspondence
in the minute serration of that margin (2d. 20. ¢’).

A small part only of the non-enamelled base was preserved, the
canines having been broken off a little above their exit from the
socket ; but the breadth of the broken base and the reduction of
the pulp-cavity to a linear trace were indicative of a long and
deeply implanted fang. Nevertheless it was evident that, in pro-
portion to the antero-posterior diameter of the base of the enamelled
crown, that part of the tooth was relatively longer than in Machair-
odus latudens: it differed also in the absence of the serrate border
at the fore part of the crown (7b. 7b. ¢). The only part of the skull pre-
served posterior to that supporting the upper canines was a portion
of the lower jaw, of which the alveolar border of the right ramus ex-
tended one inch behind the right better-preserved upper canine. No
trace of tooth could be detected in this border.

Thesymphysial part of the lower Jaw (fig. 5) extended forward about
2 inches in advance of the upper canines (2b. c.c). The animal had
become fossilized with its mouth shut; and the upper canines
descended along a laterally compressed part of the lower jaw, also
as in Machatrodus, one on each side, with their points projecting
beyond the lower border of the jaw (fig. 1, ¢,¢). The next thought
was as to possible evidences of the teeth at the fore end of the
lower jaw; but not until the extremely hard matrix had been
ground down to the alveolar border were any such traces visible.
The crowns of these teeth seemed to have been broken away prior
to interment; but the bases of eight incisors and of two lower
canines were exposed, in transverse scctions at the level stated.

* British Fossil Mammals, 8vo, 1846, p. 180, fig. 69.

Q.J.G.8. No. 126. H

96 PROF. OWEN ON A CARNIVOROUS REPTILE.

The lower canines (2. fig. 5,¢', ¢) had risen, as in Machairodus,
immediately in front of the upper ones, and presented the same
inferiority of size; but they are divided, in Cynodraco, by a tooth-
less interval, or “‘ diastema,” from the lower incisors (70. 7, 1, 2, 3,
4). In this character, as in the number of incisors, the South-
African Karoo fossil resembles the Marsupial genus Didelphis.
The lower incisors are subequal, subcompressed, and elliptical in
transverse section, at least at the base of the crown, with the long
axis of the section directed from the fore to the hind part of the
alveolar border; and they are close-set, as in carnivorous mammals.
As in these, also, the dentine, in both canines and incisors, is of the
hard unvascular kind, and the enamel as distinct in tissue and as thick.

The decrease in size is from the 1st to the 4th; but the degree
shown in the fossil and fig. 5 may be due to section at different
heights from the base.

Associated with this fossil, or from near the same locality, was a
larger oblong block of the same matrix, with the ends of a long
bone partially visible. Out of this block an entire humerus was
developed (Pl. XI. figs. 6-9, half nat. size). It is of a left fore limb, in
length 10 inches 6 lines, with some loss by abrasion of both articular
extremities, the shaft showing well-marked developments for mus-
cular attachments and other characters unusual or unknown in the
Reptilian class.

The breadth of the distal end—the extension of strong ridges from
both the outer (e, e’) and inner (f, f’) sides, just above the elbow-
joint, indicative of strong supinators, flexors, and extensors of the
forearm and paw—the modification of the articular surfaces of that
end, better preserved than those above, for the combination of due
attachment of two bones of the forearm with freedom of motion,
not only in the bending and extending, but in rotating on each
other, so that the paw could be turned “prone” and ‘‘ supine,”
whereby its application as an instrument for seizing and lacerating
is adyantaged,—add to this the structure, hitherto known only in
the Mammalian class and preeminently in the feline family, of a
defence of the main artery and nerve of the forearm from com-
pression during the action of the above-named muscles by a strong
bridge of bone (h) spanning across them, furthermore the exten-
sive and powerful ridge (6, 6’) at the proximal half of the humerus
for the attachment of arm-muscles, especially the deltoid,—the
combination of these characteristics, which Cuvier dwells upon in
contrasting the humerus of the feline and bovine mammals*, are
here exemplified in a fossil homologue, from a formation of the
Triassic or Permian division of geological time.

Extending, however, the comparison of the present humerus
beyond the salient features above defined, the head of the bone (fig. 8)
differs from that of the feline humerus in being broader transversely,
instead of from before backward; the articular part is oblong and
narrow, not hemispheroid nor nearly so convex ; there is no eleya-
tion of an outer transverse tuberosity. The representative of the

* Ossemens Fossiles, vol. i. (4to, 1821) p. xlvii.

PROF. OWEN ON A CARNIVOROUS REPTILE, 97

deltoid ridge (ib. 6, 6’), of similar basal extent to that in Felis, is more
produced and is much thinner, extending as a broad plate of bone
outward (radiad) and forward (thenad); it seems to be rather a
development in those directions of the entire shaft than to be a
superadded process of the shaft. The more rounded, thickened char-
acter which prevails through nearly the whole extent of the humerus
of the feline as of most mammals, is here confined to a constriction
barely an inch in extent between the subsidence of the deltoidal
or delto-pectoral crest and the beginning of the neuro-arterial bridge
(A) on the inner side, and of the supinator crest (¢) on the outer side
of the lower two fifths ofthe bone. The perforation itself (/) is more
highly placed, is further from the articular surface for the ulna,
thanin Felis. The ulnar articulation or “ trochlea ” is less defined ;
the inner boundary ridge, so prominent in Felis, is here undeveloped.
The convexity (fig. 6, 7) for the radius, on the other hand, is relatively
larger, and its ball advances further upon the fore part of the shaft,
in the present fossil. In this, likewise, the anconal pit (fig. 7, d), so
deep and well-defined by outer and inner ridges in /élis, is a mere
wide and shallow triangular depression.

The delto-pectoral crest (0, 6’) which seems to be a forward and
inward production of the outer border of the proximal part of the
humerus, is essentially a reptilian character of the bone. Its
ordinary proportions in existing Saurians are shown in Uroias-
tiv spinipes (fig. 10, 6, b') and in Monitor niloticus * ; in Croco-
diles the shortening of its base gives it more the character of a
distinct process; in Pterodactyles its development is chiefly in
transverse extent+; in Cynodraco the longitudinal extent prevails;
in Omosaurus the development in both directions of the delto-
pectoral crest is such as to have suggested the generic name of this
Dinosaur. Sut the perforation or canal (A, &, fig. 6) is not present
in any existing Reptile. There is, indeed, what may be loosely
termed a ‘‘supracondyloid foramen” in several Reptilia. It is
noticed in the osteology of the genus Zrionya; but its position in
the humerus is defined :—‘‘The bone is perforated from before
backwards at the outer angle of the distal extremity”t. The
homologous supracondyloid foramen is shown in the humerus of the
Monitor niloticus, in the under-cited monograph (Note *), in pl.
xvii.- fig. 6, at e', where it perforates a low supinator crest. Its
position in Uromastix spinipes is shown at U' and m in figs. 10 and 11
(Plate XI.). Dr. Giinther notes the homologous foramen in the
humerus of Zestudo elephantopus and its allies (L'estudo ephippiwm
e.g.), and describes it as ‘‘ the canal on the radial edge of the bone,
close to the elbow-joint, perforating the substance of the bone from
the front to the hinder side ’’§.

* Monogr. on Omosaurus, pl. xvii. fig. 6, 6, and fig. 6’.

t Monogr. on Cretaceous Pterodactyles, 4to, 1860, pl. iii. fig. 9, 4.

t ‘Descriptive Catalogue of the Osteological Series contained in the Museum
of the Royal College of Surgeons of England,’ 4to, 18538, p. 184, Nos.
942, 943.

§ Philos. Trans. 1875, p. 266, plate xlii.; ‘'d, radial canal for blood-yersels,”
ib. p, 285.

BR 2

98 PROF, OWEN ON A CARNIVOROUS REPTILE.

Cuvier thus defines the supracondyloid canal in the Lion :—‘ Au-
dessus du condyle interne, la ligne apre est aussi percée d’un trou pour
le passage de l’artére cubitale”*. But this is not the homologue of
the commonly called “ supracondyloid canal” in the humerus of
certain Reptilia. The “internal condyle” of Cuvier and anthropo-
tomy is the “ ulnar condyle ;” the “ external condyle” is that which
is termed ‘‘ radial” in vertebrate anatomy. Thus the zootomist has
to take into account, in the application of his science to Palzonto-
logy, of an “ ulnar supracondyloid foramen,” and a ‘radial supra-
condyloid foramen.” There is, also, a third perforation of the distal
end of the humerus distinct from both, which may be termed an
‘“‘inter-condyloid foramen.” It is that which is present in the
humerus of the wolf f and some other mammals, but of which I
have not found any example in recent or extinct Reptilia. While
upon this comparison I am tempted further to remark that, in com-
parative anatomy, zootomy, or anatomy properly so called, the term
‘*‘condyle” is restricted to portions of bone more or less convex,
modified for the articulation of two bones, and usually covered with
synovial cartilage. Anthropotomy, however, in reference to such
portions of the distal articular surface of the human humerus, calls
the radial condyle the “radial head,” and the ulnar condyle the
‘‘ ulnar trochlea,” and restricts the term ‘‘ condyle” to the tuberosi-
ties which project beyond and somewhat above the articular
prominences, which, in the humerus, answer to the corresponding
prominences rightly termed “condyles” in the femurt. I have
found it useful, in comparisons of the humerus akin to those in the
present paper, to call the anthropotomical condyles “ epicondyles,”
as being parts projecting somewhat above, or proximad of, the true
condyles, and to distinguish them as “ entepicondyle” and “ ectepi-
condyle” respectively. Thus the canal which gives passage to an
artery and commonly a nerve at the distal end of the humerus in
Felines, and most Marsupials, is an entepicondylar canal or foramen ;
and this it is which characterizes the humerus of Cynodraco. The
canal which gives passage to a blood-vessel in the humerus of certain
Chelonia and Lacertilia is an ectepicondylar canal; and its presence
in no way affects the resemblance (I will not say affinity) to the
feline Mammalia which the extinct Cynodraco presents in its
humerus as in its dentition.

I am of opinion, though it is difficult to judge from the wood-
cuts of small and fragmentary humeri ascribed to Dicynodon by
Prof. Huxley in the ‘ Memoirs of the Geological Survey of India,’
that the “supracondyloid foramen” there indicated at a, figs A
and B, p. 10,is not the homologue of the ‘‘supracondyloid foramen ”
which “occurs not unfrequently among Lacertian Reptiles,” but
that it differs not only in “form and proportions,” but likewise in
relative position, and that in Dicynodon it concurs with the pair of

* *Lecons d’Anatomie comparée,’ ed. 1835, vol. i. p. 884.

t ‘British Fossil Mammals,’ 8yo, 1846, p. 129, fig. 47, a.

{ See for example, the instructive plate “77, left human humerus,” p. 92, of
the ‘Anatomy Descriptive and Surgical,’ Syo, 1858, by Henry Gray, F.R.S.

PROF. OWEN ON A CARNIVOROUS REPTILE. 99

long upper tusks in exemplifying Mammalian characters, not Lacer-
tian ones. The same concurrence, not only with well-defined
canines, but likewise with equally well-defined incisors and molars,
has been observed by me in several species of extinct Reptilia from
the Karoo deposits of South Africa.
_ The earliest examples (Galesawrus, Cynochampsa) of such Reptilia
were made known through this Society by its ‘Quarterly Journal,’
vol. xvi. (for the year 1859-60). Since that date I have had evi-
dences of the genera Lycosaurus, Tigrisuchus, Cynosuchus, Nytho-
saurus, Scaloposaurus, Procolophon, Gorgonops, as well as of the genus
Cynodraco, of one species of which the dental and humeral charac-
ters have supplied the chief subject of the present communication.
Most of the genera above mentioned are represented by more than
one species *: and both genera and species are classifiable in groups
characterized by modifications of the structure of the external bony
nostrils—as, for example, into the “tectinarial,” ‘‘ binarial,” and
‘¢ mononarial” families, the skull in the latter presenting, in the
aspect and position of the single terminal nostril, a strikingly mam-
malian facies. For the name of these extinct carnivorous Saurians
I find it convenient, and believe it will be generally acceptable, to
form a distinct order of Reptilia under the denomination of Therio-
dontia, with the following characters :—‘‘ Dentition of the carnivo-
rous type; incisors defined by position and divided from the molars
by a large laniariform canine on each side of both upper and lower
jaws, the lower canine crossing in front of the upper; no ectoptery-
goids; humerus with an entepicondylar foramen; digital formula of
fore foot 2, 3, 3, 3, 3 phalanges.”

When we contrast the grand development and notable modifica-
tions of the extinct species of Reptilia with the poor scanty evidences
of the class still lingering on in life, we seem to be witnessing a
course of degencration, of retrogradation, rather than of organic
progression in the course of time.

Although there be none now, there have been Saurians in which
the articular surfaces of certain vertebree in the same column were
modified—as, e. g., of the anterior ones in great herbivorous Dino-
saurs, for freer movements of the head, by means of an anterior ball
playing in a posterior cup, such as we now find in the Rhinoceros
and some other large mammalian herbivores. In, another part of
the same vertebral column of these extinct reptiles we find a sacrum
not limited to two, but composed of five or six ankylosed vertebre,
also as in these and many other mammals.

Again, some of these vegetarian Dinosaurs had masticatory teeth
of a complex structure + unknown in existing phytophagous Sauria,
but resembling those teeth of certain phytophagous mammals, e. g.
Megatheriumt, Mylodon §. With the modification of a portion of the

* These will be characterized and the characters illustrated in the Plates cf
the Catalogue of South-African Reptilia, on which I am now engaged by
direction of the Trustees of the British Museum.

t ‘Odontography,’ p. 250, pl. 71: Ato, 1840.

{ Ibid. pl. 34. § Ibid. pl. 79.

100 PROF. OWEN ON A CARNIVOROUS REPTILE.

dental tissue of each tooth, by retention in the dentine of sub-
parallel vascular canals, is associated a downward development of the
zygomatic arch, for extending the origin and augmenting the size
and force of the masseteric masticatory muscles.

This structure, unknown in any existing Reptiles, but exemplified
in Iguanodon, Scelidosaurus, Parevasaurus, reappears, so to speak, in
certain Mammalia, and here in members of the class which by their
low position according to cerebral characters, with genital and con-
comitant modifications, show a near approach to the cold-blooded
Ovipara: the Sloth and Kangaroo are examples.

But cranial characters of lack, as of gain, are not wanting in this
comparative glance. The lingering evidence, in Reptiles, of vegeta-
tive repetition, as manifested by multiplied centres of ossification in
the facial blastema, has disappeared. The ectopterygoid ceases to
exist in both Theri- and Dicynodonts ; that bone never reappears
in the mammalian series.

In no existing Reptilia is the principle of differentiation of struc-
ture adapting particular teeth for special functions exemplified as in
the extinct Theriodontia. Not until the discovery of members of
this order of Reptilia could the anatomist specify “ incisors,”
‘‘molars,” ‘ canines,” in the dental series, by characters of size,
shape, and relative position, with the same certainty, or on as satis-
factory grounds, as in the warm-blooded quadrupeds. With the
carnivorous type of incisors, canines, and molars we now have eyi-
dence of an associated humeral structure unknown in any lacertian
or other existing reptilian fore limb, but a structure reappearing in
certain mammals and notably in the implacental group of Mar-
supialia and in the Ovo-viviparous Monotremes.

In one of the species of Saurians with what is now the mam-
malian type of humerus and of dentition, evidence, for which I am
indebted to the Governor of the Cape of Good Hope, Sir Henry
Barkly, K.C.B., has reached me of the bony structure of the fore
paw, which again shows an advance towards the mammalian type.
The pollex has two phalanges ; the four other digits have each three
phalanges. The slight difference in length in these fingers is due,
as in a dog’s paw, to difference of length in such phalanges, not to
difference of number of these, not to excess in the third and fourth
digits beyond the number three, which number rules in the fingers
of all terrestrial mammals.

Reverting to the chief character of the Theriodont reptiles, a fact
of some significance may be noted, viz. that the incisive formula of
some of the species is repeated in the low marsupial order of mam-

mals. Didelphis, e. g., has 2. = as in Cynodraco; Thylacinus

and Sarcophilus have 1. — as in Cynochampsa. In no placental

3-3
carnivore do the incisors exceed =—

In the class of cald dave: ai) le tito naked Ovipara, as
known to zoology by existing species, the characters above specified,
discussed, and compared are wanting. They would have been un-

PROF. OWEN ON A CARNIVOROUS REPTILE. 101

known and unsuspected as reptilian ones by zootomists, save through
the researches of the paleontologist.

7° the gap in the animal series between the Mesozoic and
himanous air-breathers had not been filled up otherwise than by
reptiles, the remnant of that class which has survived and reached
our times would have testified to total loss of such gains of organiza-
tion as had enriched the ancestors or predecessors of modern
tortoises, lizards, and crocodiles.

We now know that not one of these gains has been lost, but has
been handed on and advanced through a higher type of Vertebrata,
of which type we trace the dawn back to the period when Reptiles
were at their best, grandest in bulk, most numerous in individuals,
most varied in species, best endowed with kinds and powers of loco-
motion and with the instruments for obtaining and dealing with
food.

Has the transference of structures from the reptilian to the
mammalian type been a seeming one, delusive, due to accidental
coincidence in animal species independently (thaumatogenously)
created? or was the transference real, consequent on nomogeny or
the incoming of species by secondary law, the mode of operation of
which we haye still to learn? Certain it is that the lost reptilian
structures dealt with in the present paper are now manifested by
quadrupeds with a higher condition of cerebral, circulatory, respira-
tory and tegumentary systems, the acquisition of which is not
intelligible to the writer on either the Lamarckian or the Darwinian
hypothesis,

EXPLANATION OF PLATE XI. '
Cynodraco major.

Fig. 1. Nodule of Karoo clay-stone with fore part of skull, the upper canines

exposed (ce, ¢.).

. Side view of crown of right upper canine.

. Transverse section of ditto at the dotted line ¢, fig. 2.

. Portion of canine, magn. 2 diam., showing serrations of the trenchant

hind border, ec.

Upper view of fore end of mandible, showing transverse sections of the
crown-base of the incisors (2 1, 2, 3, 4) and canines (c’), with the
crowns of the upper canines (¢, ¢).

. Front (thenal) view of left humerus, half natural size.

. Back (anconal) view of left humerus, half nat. size.

. Proximal end, abraded, of left humerus, half nat. size.

. Distal end, less abraded, of left humerus, half nat. size.

10. Front (thenal) view of left humerus of Uromastix spinipes, nat. size.
11. Back (anconal) view of left humerus of Uromastix spinipes, nat. size.
In the figures of the humerus, a, “head;” 6, ectotuberosity, here the be-

ginning of the delto-pectoral crest () 6'); ¢c, entotuberosity, not well defined; d,
olecranal depression; ¢, beginning of ‘‘supinator ridge,” ending at e’, the
ectepicondyle (the ridge is less developed in Uromastix); f, entepicondylar
ridge; jf’, entepicondyle ; /, ulnar condyle; g, radial condyle; h, bridge de-
fining #4, the entepicondylar canal (fig. 6); m, bridge defining the ectepi-
condylar canal (figs. 10and 11). All the figures of the natural size, save where
otherwise expressed.

ore GOS

Noe oh? Ber)

102 PROF. OWEN ON A CARNIVOROUS REPTILE.

Discussion.

The Presrpent (Mr. Evans) remarked that Prof. Owen’s paper
was a most important and suggestive one, especially as regarded
the views advanced respecting the connexion between these old
Carnivorous reptiles and the Mammalian Carnivores.

Prof. Sertey remarked upon the extraordinary characters pre-
sented by the creature described by the author, and expressed his
regret at the want of additional materials, which might have thrown
a further light upon the difficult and important questions raised in
the paper. He thought that if all the forms referred to the Reptilia
were to be regarded as belonging to that class, the latter would be
rather difficult to define. The present representatives of the Reptilia
are the Chelonia, Crocodilia, Lacertilia, and Ophidia; and any forms
departing from these are not strictly Reptiles in the ordinary sense
of the term. He thought the present fossil presented some Chelonian
characters, but that in many Lizards we may find indications of a
dentition similar to that of the fossil. He considered that there
could be no doubt as to the connexion between Reptiles and Mammals,
and that Prof. Huxley was wrong in his views as to the relation-
ship between Birds and Reptiles. Every mammalian type has a
reptilian brain in its earliest stages. The suggestion of the forma-
tion of a new order seemed to him to be founded upon certain
points which could not be regarded as absolutely proved.

Prof. T. Ruprrr Jones congratulated the Society on having been
the medium of publication of the magnificent series of Fossil Reptiles
characteristic of South Africa. He was sure that to Prof. Owen it
must be a heartfelt pleasure to have been the immediate elucidator
of these wonderful creatures of manifold and rare structures, brought
out by his many years of continued labour on the collections made
by Bain, Orpen, Atherston, and others. Together with the illus-
trated descriptions of Professor Huxley, his lucid and powerful
expositions have made the history of these creatures known to the
world ; and they will prove a lasting monument of his persevering and
elucidative work. Prof. Jones added a few words on the geological
occurrence and distribution of the Dicynodont and associated Reptiles
in the Karoo formation of South Africa, its lacustrine or estuarine
origin, its enormous thickness, wide extent, and probable age as early
Mesozoic.

Prof. Duncan maintained the necessity of accepting the Reptilian
type as here understood by the author, and remarked that the em-
bryonic forms of mammals are reptilian. The question seemed to
him to be one of probabilities. The old beds contain the foreshadow-
ings of higher forms of animals.

Prof. OwsEy, in reply, stated that after thirty years of work on fossils
he had arrived at the conclusion that the artificial line between the
Paleozoic and Mesozoic series seems to need to be raised so as to in-
clude the Trias. The only fossil fish from the beds yielding the
fossil described appeared to be Paleozoic. He justified the reference
of the fossil to the Reptilia, and remarked that in no Reptile does
the ramus of the lower jaw consist of one piece.

‘6
.

A eras
SS ee

is 5 re
Age eet eee

a
a

Quart .Journ.Geal. Soc Val XXXII.PI. XI.

Mintern Bros. imp

GH Ford & ©.L.Gnmesbach.

R. ETHERIDGE, JUN., ON ASTROCRINITES. 1038

13. On the Occurrence of the Genus AsrrocrinirEes (Austin) in the
Scotch Carsonirerovs Limestony Series; with the Description of
a new Species (A.? Bunnter), and Remarks on the Genus. By R.
ErHEriner, Jun., Esq., F.G.S. (Read February 2, 1876.)

[Puates XII. & XIII.)

Inrropuction.—In a paper entitled “‘ Proposed Arrangement of the
Echinodermata” &c.*, Messrs. Austin referred to a genus which
they proposed to establish under the name of Astracrinites, in the
following words :—“ Another genus (Astracrinites of our MS.) offers
so many affinities to the recent and fossil Echinodermata, that we
consider it the most remarkable of all the known genera. By its
being lobed it approaches to the Lobistellay ; its ambulacra, spines,
and arms mark it as allied to the Echinites, while the arrangement
of its calcareous plates connects it with the Liries of the ocean. In
short, it possesses the lobes of a Starfish, the ambulacra and spines
of a Sea-urchin, and the plates of a Crinoid. It is further remark-
able by deviating from the quinary type so prevalent in the Echi-
nodermata: the lobes and ambulacra of this new genus are each
four in number.”

In the following year Messrs. Austin described this curious and
aberrant form under the name of Astracrinites tetragonus. Their
description is as followst :—

“Family ASTRACRINIDA,
* Consisting of the genera Astracrinites and Aporocrinites ;

‘*Genus Asrrocrinites, Austin.

‘« Def.—Dorso-central plate quadrangular, to which four pairs of
elongated plates are attached, imparting a lobed shape to the fossil.
In the retiring angles at the base of the four lobes are a like number
of ambulacra. Mouth central. Anus lateral.

“A. tetragonus, Austin.

“‘ Def.—The plates of this species agree with the generic defini-
tion. Hach of the elongated plates has two or three rows of minute
tubercles around its outer margin, apparently for the attachment of
spines. The ambulacra have each a double row of pores placed
centrally, with marginal tubercles. Near the centre of the dorso-
central plate is an oval eminence, apparently analogous to the ma-
dreporiform tubercle on the dorsal surface of the true Starfishes.”

In neither of these communications did the authors assign any
locality to their fossil; but afterwards one of them, Fort-Major
Austin, in a paper read before this Society, “‘ Observations on the

* Ann. & Mag. Nat. Hist. 1842, x. p. 112.

+ The Lodistella of Austin correspond to the Cirrhigrada of Forbes.
{ Op. cit. 1848, xi. pp. 205, 206.

104 R. ETHERIDGE, JUN., ON ASTROCRINITES.

Cystidea of M. von Buch, and the Crinoidea generally,” gave the
horizon and locality as Carboniferous Limestone, Yorkshire*. The
name Astracrinites has been rejected by Dr. H. G. Bronn, on
account of its close resemblance to Asterocrinites, Miinster; and he
has proposed in its place that of Zygocrinust. Dr. F. Romer
observes that the four-rayed structure of this fossil more properly
allies it with the Cystoidea than with the Blastoidea, which i other-
wise considerably resembles +t.

Prof. de Koninck and M. le Hon § refer Zygocrinus to the Blas-
toidea, and state that the genus differs from Pentremites by the
absence of ovarian apertures and the possession of separate openings
for the mouth and anus.

Prof. Morris has altered Austin’s name Astracrinites into Astro-
cerinus ||, and does not recognize Bronn’s name Zygocrinus in con-
nexion with it. He gives Settle as the locality.

The late Prof. Pictet provisionally placed Zygocrinus in the Blas-
toidea, stating at the same time that its structure indicated a com-
plete analogy with that of Codaster [ Codonaster], M‘Coy, except that
it possesses four pseudambulacra, instead of five as in the latter ¥.

The substitution of generic names by various authors for trivial
causes has given rise to endless confusion; and as I cannot perceive
the desirability of adopting Dr..Bronn’s substitution of Zygocrinus
for Austin’s Astrocrinites, I have much pleasure in restoring the
latter, notwithstanding its resemblance to <Asterocrinites, Munster,
the difference being sufficiently well marked to prevent any confu-
sion of the two names. <A very eminent authority, Prof. A. Agassiz,
thus writes on this subject **:—‘ We ought not to reject names dif-
fering so little as Moulinia, Moulinsia, Moulinsium, Cassidula, Cas-
sidulus, simply because they are likely to be mistaken for one
another ; in our present condition, with an infinite number of
genera, a difference, no matter how slight, should be sufficient
reason for retaining the name instead of coining a new one, which
is just as likely to fall into the same category, and resemble another
name in a different department to as great an extent.”

These quotations comprise all the published information I have
been able to gather concerning the history of Astrocrinites tetra-
gonus, Austin.

I have now the pleasure of bringing under the notice of the
Society some small and highly interesting fossils lately discovered
by Mr. James Bennie, during his duties as one of the fossil-collec-
tors of the Scotch branch of H.M. Geological Survey. In doing
so I have to thank Prof. Ramsay, F.R.S., Director General of the
Survey, for permission, obtained through Prof. Geikie, F.R.S., to

* Quart. Journ. Geol. Soc. 1848, iv. p. 293.

t Index Pal. Nomenclator., 1848, p. 1381.

t Wiegmann’s Archiv f. Naturg. 1851, p. 390.

§ Rech. sur les Crinoides du Terr. Carb. Belg., 1854, pp. 189 & 193.
|| Cat. Brit. Foss. 2nd edit. (1854) p. 72.

{| Traité de Pal., 1857, iv. p. 295.

** Revision of the Hchini, pts. 1 & 2 (1872), pp. 23, 24.

R. ETHERIDGE, JUN., ON. ASTROCRINITES. 105

publish these notes otherwise than through the regular channel. The
fossils in question, found by Mr. Bennie both in the Carboniferous
series of the Lothians and Fife, are very closely allied, if not iden-
tical, with the genus Astrocrinites; but as they appear to differ in
certain minor points from the typical species, I have considered the
form to be an undescribed species, and have named it A.? Bennier,
after Mr. Bennie, to whose acute observation we are indebted for
the discovery of the specimens. Unfortunately the majority are
much crushed, although a few are intact; and zn all, the individual
component parts retained in each specimen are in a fine state of
preservation.

During my investigations into the structure of A.? Benniei I have
been favoured by Fort-Major Austin with a drawing and a specimen
of his A. fetragonus, and by Prof. T. McK. Hughes with the loan
of specimens from the Cambridge Museum, which I recognized
in the collection when visiting that Museum some time since. I
take this opportunity of thanking both these gentlemen for their
attention and kindness.

1. Description of the Fossils.

(a) General Form.—The body or calyx is quadriradiate ,with con-
vex prominent lobes, three of which are alike and project more than
the fourth, which varies considerably in its structure from the other
three. The deep reentering angles between the lobes are occupied
by the pseudambulacra. The dorsal surface, or that answering to
the attached surface of a Crinoid, is convex, and covered by closely
set tubercles, but without any sign of a point of attachment for a stem.
The ventral surface is flattened, with a large central aperture, from
which radiate the four pseudambulacra. On the non-symmetrical
lobe, and excentric, as compared with the ambulacral system, is a
second and pyriform aperture of complex structure.

(b) Component Parts——It will perhaps be more intelligible if I
describe the plates constituting A.? Bennier as those forming the
three symmetrical lobes and those composing the unsymmetrical
lobe. ‘The minute size and crushed nature of the majority of the
specimens, added to the very peculiar arrangement of their parts,
make this anything but an easy task, which has, however, been
lightened by the skill with which Mr. Bennie has sifted and washed
samples of the disintegrated shale in which A.? Benniet occurs, and
so obtained some, at least, of the plates in a free and separate con-
dition. Immediately surrounding the central aperture, on the ven-
tral surface, and in the angles between the ambulacra, are what I
have, for the want of a better term, called the three ‘spearhead-
shaped” plates (a, figs. 5 & 8), from their general resemblance to
the head of a spear, having a small truncated base, constricted shaft,
and expanded head. We have never succeeded in obtaining these
little plates separate, but only in conjunction (a, fig. 8) with those
next about to be described. These “spearhead” plates are suc-
ceeded in radial order by three larger pieces (4, figs. 1, 4, 5, & 8),
thick, strongly arched, and convex, ornamented along the median

106 R. ETHERIDGE, JUN., ON ASTROCRINITES.

line or crest with several rows of peculiar tubercles (figs. 21 &
22). A few examples of these plates have been obtained by Mr.
Bennie, in the manner previously explained, separate or united
to their “spearhead” plates (6, fig. 8). So far these are the
plates of the three symmetrical lobes visible on the ventral sur-
face of the fossils ; whilst the points of junction of the crested and
tuberculated plates just mentioned with the succeeding pieces
forming the dorsal surface are the most acute or prominently ex-
tended portions in the general periphery (c, fig. 1). These three
crested plates, to the convexity of which the fossils owe much of
their lobate character, are united with the truncated apices of four
other plates, which enter into and chiefly form the dorsal surface,
and with which they alternate. Two of these are arrowhead-
shaped or forked plates, and alike (e, fig. 3; 6, fig. 6; figs. 7 & 7*);
whilst the other two appear to be different; but of their exact form
I am not at present quite satisfied. A number of the forked plates
have been picked out by Mr. Bennie (figs. 7 & 7*); they are
slightly concave internally, with, on each side of the fork, two
notched or angular shoulders (0, fig. 7*) and a central concave de-
pression (a, fig. 7*), into which the apex of the pseudambulacrum fits.
Each half or fork of the forked plates is convex (c, fig. 7); and the
lateral union of every two contiguous halves gives rise to the lobes,
which are continuations of those formed by the crested and tuber-
culated plates on the ventral surface (6, fig. 8). The closely packed
tubercles on the dorsal surface usually hide the sutures between these
forked plates; but they are visible in some specimens (J, fig. 3,
a, fig. 6). As previously stated, I can with certainty only assert
that two of these plates forming the dorsal surface have the form
represented by fig. 7—the other two (a, fig. 10), bounding the pseud-
ambulacra bordering the unsymmetrical lobe, being apparently
different in form. In A. tetragonus the Messrs. Austin describe a
dorsocentral plate; but I am again uncertain whether the present
little fossils possess one, although in some specimens there are
certain small grooves amongst the tubercles which appear to mark
the outline of such a plate. If a dorsocentral plate is present, then
it is to be seen in fig. 12, where lateral pressure has thrust it up
out of its place (a, fig. 12).

Passing now to the fourth or unsymmetrical lobe, we meet with
a much more complex structure, exceedingly well shown in fig. 5.
The surface is much more flattened than that of either of the other
three lobes (8, fig. 5), and occupied by an elongated pyriform aper-
ture (fig. 5; a, fig. 20). On one side of the latter is an elongated
plate (c, fig. 5; 6, fig. 20), the ambulacral extremity of which has
much the appearance of being haif one of the “spearhead” plates
(a, fig. 5). If this view is correct, the corresponding half on the
other side (¢, fig. 20) would occupy, during life, what is now a
vacant space (d, fig. 5) merely filled with matrix. These two halves
partly bound the excentric aperture, the remainder of the latter
being excavated out of another plate (e, fig. 5; figs. 13 & 14; d, fig. 20),
triangular or somewhat deltoid in form, and broader than long,

R. ETHERIDGE, JUN., ON ASTROCRINITES. 107

crossed by vertical scale-like ridges (fig. 13) instead of being orna-
mented with tubercles. It has a projecting lip (f, fig. 5; e, fig. 20),
with a minute inward ridge (g, fig.5). Following this plate round
towards the dorsal surface, we find that its pointed extremity
(a, figs. 138 & 14; f, fig. 20) is received into the reentering angle of
yet another plate, which may be seen in the end view of a specimen
represented in fig. 15 (a), passing on to the dorsal surface, with
which it unites. Whether this (a, fig. 15) is a single plate or com-
posed of two portions, | am in doubt. In some individuals it ap-
pears to be entire (0, fig. 12; d, fig. 11), whilst in others there is
certainly a division down the centre from the reentering angle
separating it into two halves (a, a’, figs. 15 & 16). In fig. 16 we
have represented an internal view of the crushed hollow dorsal surface
(with, at 5, such a separation of the plate in question into two halves),
the only remaining portion of the unsymmetrical lobe. The aper-
ture-plate (e, fig. 5; a, fig. 15; d, fig. 20) presents, when viewed
internally, some points of interest. The inward projecting ridge
described as at g (fig. 5) is seen to be a portion of a minute circular
cup-like depression (0, fig. 14), which leads inwards and downwards
by a small channel or groove (c, fig. 14) towards the general cavity
of the body. At the edge of the interior and vertical portion of the
plate this channel appears to break up into two or three folds
(d, fig. 14).

1 have endeavoured in the foregoing description, with the aid of
Mr. Sharman’s truthful figures, to convey some idea of the plates
composing this little organism, although it is difficult to convey to
others a clear and comprehensive view of its structure, especially
when we take into consideration its small size and usually crushed
condition.

(c) Pseudambulacra.—The pseudambulacra are four in number
(a, fig. 1), long, narrow, very slightly petaloid, and radiate outwards
from the central aperture, one to each of the four concavities or inter-
lobular spaces between the arched and convex lobes of the calyx. Each
pseudambulacrum is more or less hexagonal in section, as indicated in
fig. 18, which is a diagrammatic cross section of one Mr. Bennie has
succeeded in obtaining separated from the remainder of the organism
(fig. 19). The component parts of the pseudambulacra are usually so
closely fitted together that the plates can only be occasionally dis-
tinguished. The more or less nearly horizontal upper surface
is traversed by a median (and for the size of the organism),
deep longitudinal groove. The sides of each pseudambulacrum are
bent down at a considerable angle, and in weathered specimens
are seen to be formed by two series of small plates on each side. The
pseudambulacral groove (a, fig. 17) gives off other, short, lateral, alter-
nate grooves (0, fig. 17), which terminate in small and nearly round
depressions (c, fig. 17) or sockets. Contiguous to each of these, and
separated only by a small interspace, is another depression (d, fig. 17)
somewhat more than halfwaybetween the pseudambulacral groove and
the margin, also connected by another groove (f, fig. 17) with a slit-
like opening (or pore?) (e, fig. 17) between the side plates g and h.

108 R. ETHERIDGE, JUN., ON ASTROCRINITES.

The small spaces separating the inner contiguous sockets (c and d,
fig. 17) are prominent, and give a longitudinal ridge-like appearance
to both sides of the pseudambulacrum, better seen in the cross sec-
tion (c, fig. 18). The slit-like opening (¢, fig. 17) may represent the
pore of the pore-plate in a Pentremite, when the smaller plate (A,
fig. 17) would perhaps be identical with the supplementary pore-
plate in the same genus; it appears, however, in the present organ-
ism to be itself sometimes pierced by a pore. On the underside
of the pseudambulacrum, at the base or the end near the ventral
aperture, is a peculiar projecting process (e, fig. 19). '

(d) Apertures.—The ventral central aperture, when the spear-
head-shaped plates (a, fig. 5) are in position, is in the form of a
St. George’s cross; but when these are removed it is seen as in
fig. 4, much larger and quadrangular. The crushed condition and
delicate nature of this part of the calyx in all these little fossils
render it extremely difficult to form an absolutely correct idea as
to what was the perfect state of the central aperture. I have not
seen any evidence of supplementary pores surrounding the latter
with which the pseudambulacral grooves communicate, similar to
those discovered by Mr. Billings in the apex of Pentremites conoideus,
Hall *.

The excentric aperture (fig. 5, & a, fig. 20) is pyriform and filled
in, or perhaps covered by a plate very valve- or lid-like in cha-
racter. It is a narrow elongate plate (A, fig. 5; g, fig. 20), slightly
broader at the middle portion than at each extremity, the outer of
which is split up into a series of small prolongations or fingers,
resting on the margins of the aperture. In the figured specimen
(fig. 5) the two terminal ones are the longest, whilst on the best-
preserved side of the plate there are three lateral digitations; the
former, or two terminal ones, rest one on each side of the previously
mentioned ridge, running inwards towards a small cup-like depression
(9, fig. 5; 6, fig. 14).

(e) Ornamentation.—The dorsal surface and interradial lobes are
ornamented with tubercles—those of the dorsal surface sharp and
tooth-like (a, fig. 24), those of the plates short, conical, bluntly
pointed, and impunctate, fluted down the sides, all set independently
of one another, and with a plain smooth apex (figs. 21 & 22). The
convex-crested plates on the ventral surface have usually each three
or more rows of these tubercles well developed, the central row
being the largest.

(£) Spines.—Adhering to one of the specimens by some particles
of matrix, but not attached in place, is a microscopic spine (fig. 23),

(g) Madreporiform Tubercle.—I have quite failed to detect any
body or projection of or on the surface which could for one moment
be construed as representing this organ. It is true that on one of
the specimens (c, fig. 3; 0b, fig. 24) obtained by Mr. Bennie at
Carlops, there is a peculiar body, which I at first thought might
be the “eminence on the dorsal surface” mentioned by Messrs.
Austin in A. tetragonus, but was puzzled how to account for its pre-

* Pal. Fossils (Geol, Sury. Canada), ii. pt. 1, p. 103.

R. ETHERIDGE, JUN., ON ASTROCRINITES. 109

sence on only one individual out of the series of specimens in the
Survey collection. When making the excellent drawings which
accompany this paper, my friend and colleague, Mr. G. Sharman,
noticed that the little body represented in the figure, although a
good deal worn, possessed a spiral form—a fact which I had over-
looked. It then struck me that this might be an accidentally adherent
specimen of a somewhat common Carboniferous Foraminifer, Valvu-
lina. For the sake of comparison, an enlarged figure of a species of
Valvulina is also given (fig. 25); and it will be observed that the re-
semblance is very. close. In some specimens the tubercles on the
dorsocentral plate certainly appear to be somewhat worn off, or per-
haps less developed than on other parts of the surface; but still in
no way is there to be distinguished any resemblance to a Madrepori-
form tubercle, such as is described by Major Austin in his species.

2. Affinities of A.? Benniei with A. tetragonus, Austin.

The specimen kindly forwarded to me by Fort-Major Austin, like
those in the Cambridge Museum, is from limestone, and in a bad
state of preservation; nevertheless in the former I can distinguish
the two pieces which I have called the forked plates (}, fig. 6,
fig. 7), and the three arched and crested plates bearing tubercles
(c, fig. 1; 6, fig. 5), although I cannot see the succeeding spear-
head-shaped pieces (a, fig. 5). The dorsocentral plate has been
accidentally removed in the specimen in question, leaving only
its cavity (a, fig. 26). From a comparison:of these with the spe-
cimen in our collection, showing the apparently displaced plate on
the dorsal surface (a, fig. 12), I think it more than probable that
this dorsocentral plate did exist in A.? Benne. The fourth or un-
symmetrical lobe in this specimen of .A. tetragonus is not well pre-
served, but appears to be generally constituted as in our species.
Under these circumstances I think there can be very little doubt as
to the close congeneric nature of these two fossils; for they coin-
cide in the following points—the general tetraradiate and lobate
form, want of symmetry between all the lobes, presence of both a
central and interradial aperture, possession of a tubercular orna-
mentation, absence of a stalk or column, and the arrangement of
the pseudambulacra. On the other hand, as previously stated, I have
quite failed to detect in A.? Bennier any body resembling a madre-
poriform tubercle permanently attached; it is also smaller, less
lobate, and apparently of more delicate construction than <A. tetra-

onus.
: 3. Systematic position of Astrocrinites.

In their “ Proposed Arrangement of the Echinodermata” &c.,
Messrs. Austin proposed a new family for the reception of their
genus Astracrinites* (then only known as a MS. ee the Astra-
crinoidea, afterwards changed to Astracrindey.

Their first class, the’ Pinnastella (= Crinoidea auct.), is divided

* Ann. & Mag. Nat. Hist, 1842, x. pp. 106-118,
t L.c. 1843, xi. p. 205.

110 R. ETHERIDGE, JUN., ON ASTROCRINITES.

into several orders, the first two of which are the Ctonacinete
(Austin), or those forms with a jointed flexible stalk, and the
Liberide (Austin), the latter containing the family in question.

The following remarks are devoted to a consideration of the sys-
tematic position of Astrocrinites? Bennier with regard to the various
orders of the Echinodermata as at present recognized. I believe I
am correct in stating that throughout the Echinoidea, Asteroidea,
and Blastoidea the quinqueradial arrangement is steadily maintained,
and that in the Crinoidea, although the arms become much mul-
tiplied, they are primarily from five to ten in number. On the
other hand, in the Cystoidea (e.g. Apiocystites, &c.) the radii are
sometimes reduced to four; but although such is the case, they can
hardly be compared to the radii of A.? Benniei; for in Aprocystites
they are only recumbent arms, whereas in the latter they are pseud-
ambulacra imbedded in the plates of the test. However, 4.? Benniet
so far corresponds with some Cystideans; but, on the other hand,
there are not any organs similar to the pectinated rhombs of that
Order.

The absence of a stem or column for attachment tends to separate
this species from the Crinoidea, Blastoidea, and Cystoidea, although
we know that in the adult condition some Crinoidea are free, and
that certain Cystideans (Agelacrinites &c.) are asserted to be without
any column; whilst in another, Lepadocrinus, there is evidence to
show that it was free, or that its means of attachment were not of
a permanent description.

A, tetragonus, Austin, has been described as possessing a madre-
poriform body, which, if it exists as described, would indicate
Asteroid or Echinoid affinities; there is, however, no trace of such
an organ in A.? Benniei. Furthermore, in the Asteroidea the ske-
leton is composed of a large number of small plates united by mem-
brane; and when a second aperture in addition to the mouth is
present, it is dorsal and nearly central. In the form under con-
sideration the arrangement of the plates of the calyx is quite different
from the foregoing ; and both apertures, which are constant, are on
the ventral surface.

It appears to me that the forked plates (figs. 7 & 7*) into which
are received the apices of two of the pseudambulacra, correspond with
the forked “ radial” plates of the Blastoidea, although there is this
curious divergence from the latter type :—If we take a Pentremite
for example, any lobe-like projections existing in the general peri-
phery are caused by the projection of that part of the radial plates
into which are received the apices of the pseudambulacra (a, fig. 27),
whereby the suture or union of each prong of the forked ‘plates
with its neighbour lies more or less in an angular concavity or re<
entering angle (6, fig. 27); in Astrocrinites this arrangement is
reversed, the pseudambulacra all lie in deep concavities between the
convex and projecting lobes, so that in this case the line of union
of the contiguous prongs of the forked plates lies on a convewity Ys
instead of in a concavity as in the Pentremite.

Mr. Billings has pointed out that in the Pentremite the forked

R. ETHERIDGE, JUN., ON ASTROCRINITES. Lis

plates, although called “radials,” do not support the bases of the pseud-

-ambulacra as do those plates in the Crinoidea, but, on the contrary,
these are situated at the apex of the Pentremite, the smaller extre-
mity of each corresponding with the “‘ apex of the ambulacrum of a
Sea-urchin or of a Starfish. It also represents the tip of the arm of
a Crinoid.” Mr. Billing’s deduction from this is, that the ‘“ forked
plates do not belong to the radial but to the perisomatic system.”
In asimilar manner the bases of the pseadambulacra in Astrocrinites
are situated at the apex of the fossil, the tips being received in the
“radial” or forked plates, occupying the position of the bases of the
arms of a Crinoid*. The three convex arched plates (6, fig. 5) are
probably analogous to the interradials in the Blastoidea (a, fig. 27),
whilst for the truncated plate (figs. 138 & 14, a, & fig. 20) partly sur-
rounding the excentric aperture it is more difficult to find a homo-
logue. I would ask, Can it be compared to the fifth interradial
plate of Nucleocrinus (which, according to Mr. Billings, “is trun-
cated at its apex for the reception of the oro-anal orifice” +)? This
oro-anal area of Nucleocrinus, Messrs. Meek and Worthen tell us, is
wider and “often more prominent above . . ._ , and occupied
by three large, elongated pieces, the middle one of which, the anal
piece, is lanceolate in form, and, with the two interradials, fills all
the large anal area down to the base’ t.

I have not succeeded in finding any specimen with the central
opening closed by small plates as in Nucleocrinus§$, Granatocrinus ||,
and some Pentremite 4[; nor are there any apertures to be seen
analogous to those occurring at the apices of the deltoid or inter-
radial plates of these genera, the ovarian orifices of some authors,
the ‘respiratory spiracles” of Mr. Billings**. Finally, I believe
that the peculiar lid or valve (A, fig. 5, g, fig.20) covering the inter-
radial aperture in Astrocrinites is analogous to that figuredt} by
Messrs. Meek and Worthen as covering the so-called anal aperture
(the oro-anal of Billings) in Granatocrinus Norwoodi (Owen and
Shumard), although in position it corresponds with the anal opening
of Codonaster, McCoy, and Codonites, Meek and Worthen. Both these
genera have been transferred by Mr. Billings to the Cystoidea tt.

The structure of the ambulacra appears to be to a great extent
on the plan of the Blastoidea. Mr. Billings has described those of
Pentremites pyriformis at some length. He says, ‘‘The median
groove . . . sends off branches, right and left alternately, towards
the sides of the ambulacrum. These branches do not run directly
to the ambulacrum-pores. Each of them terminates at a point
between the inner extremities of two of the pores. There is at this

* Geol. Surv. Canada, Pal. Foss. ii. pt. i. p. 112.

t Op. cit. p. 115. { Illinois Geol. Rept. ii. p. 275.

§ Billings, op. cit. p. 117 and p. 114, f. 73.

|| Meek & Worthen, cakes Of. 2a & 6.

Gmee ah biir arth yeh, Lieil9, 2.0

tt See his views, American Journ. Sc. 1869, xlviii., 1870, xlix.; Ann. & Mag.
Nat. Hist. 1870, v., 1871, vii.; Mem. Geol. Surv. Canada, Pal. Foss, ii. pt. i.
p. 101.

O25/G,8. No. 126: I

112 R. ETHERIDGE, JUN., ON ASTROCRINITES.

point a small pit, which appears to be the socket of an appendage
quite distinct from the pinnule. The groove does not reach the
socket of the pinnule, which is situated further out, between two
of the pores. On the other hand, a small groove runs from each pore,
inward, and terminates at ee socket, about halfway between
the pore and the main median groove of the ambulacrum”*. If the
description of the pseudambulacrum of A? Benniei be referred to
(ante, p. 107) and fig. 17 examined, a close affinity to the foregoing
will be recognized. The sockets at the termination of the short
lateral grooves (c, fig. 17), I think, represent the small pits de-
scribed by Mr. Billings in P. pyriformis as terminating the lateral
grooves ‘‘at a peint between the inner extremities of two of the
pores.” The socket (d, fig. 17), on the other hand, at the end of
the groove (f, fig. 17), running inwards from each pore (e, fig. 17),
is similar to those in P. pyr iformis, “about halfway between the
pore and the main median groove of the ambulacrum.”

As previously stated, the Blastoid affinities of Astrocrinites (= Zy-
gocrinus) have been pointed out by Dr. F. Rémer, Prof. de Koninck,
M. le Hon, and the late Prof. Pictet. It is to this order that I
would provisionally refer Astrocrinites, at any rate until farther evi-
dence, obtained from the examination of more perfect specimens
than ours, can be adduced for or against such a reference.

4. Affinities and Differences.

With the Blastoidea Astrocrinites agrees (1) in the nature and func-
tion of certain of the plates of the calyx, (2) in the nature, arrange-
ment, and mode of growth of the pseudambulacra, (3) in the
possession of a central ventral aperture, around which the latter
radiate, (4) in the presence of a second ventral aperture, excentric
and interradial, closed by a valve or lid. On the other hand, this genus
differs from the Blastoidea (1) in the quadriradiate arrangement of
its parts, (2) its free and unattached habit, (3) the presence of the
tubercular ornamentation ; added to which the typical species has been
described as possessing a madreporiform body, which ours does not.

5. Localities and Geological Horizon.

Astrocrinites? Bennie was first found by Mr. Bennie at Carlops
Quarry, near Carlops, Peeblesshire, in shale overlying the No. 2
limestone of the Lower Carboniferous Limestone group of the Mid-
lothian Carboniferous Series, where it occurs in considerable numbers.
It was subsequently found plentifully in a bed of shale occupying
the same geological horizon at Kidlaw, near Gifford, Haddingtonshire,
and more sparingly at East Salton and Salton lime-works, chiefly
as fragments ; again at Skateraw and Kast Barnes, near Dunbar,
in the same horizon, but usually in a fragmentary condition. A few
individuals were found in the shale overlying the No. 2 limestone
at Lampland, near Path-head, Haddingtonshire, and two fragments
at Roscobie, Fifeshire, in shale connected with one of the lower

* Dic. p. 14.

R. ETHERIDGE, JUN., ON ASTROCRINITES. 1138

limestones, probably either No. 1 or No. 2. The No. 2 limestone
of the Midlothian Carboniferous Limestone series is probably the
equivalent of the Hosie limestone of the west of Sectland.

P.S. Since this paper was written .? Bennie: has also been met
with at Cowdens Quarry, near Dumfermline, Fife, in shale above the
Linn limestone, probably the equivalent of the No. 1 limestone of
the Midlothian Lower Carboniferous Limestone groap. (Hdinb.
March 21, 1876.)

EXPLANATION OF PLATES XII. & XIII.

Astrocrinites? Benniet (R. Eth. Jun.).

Puate XII.

Fig. 1. View of ventral surface, crushed: @, pseudambulacra; 0, arched, con-
vex, and crested plates ornamented with fluted tubercles (figs. 21 & 22);
ec, suture between 6 and half the succeeding forked plate (figs. 7 &
7*); d, central ventral aperture; ¢. position of the unsymmetrical
lobe, broken away.

2. View of dorsal surface, crushed, covered with spike-like tubercles
(a, fig. 24): a, apices of two of the pseudambulacra.

3. View of dorsal surface, somewhat distorted: a, pseudambulacrum; 8,
sutures between the forked plates ¢, e'; ¢, adherent Foraminifer ?
(see fig. 24, 6): d, small encrinite ossicle adhering by matrix; f, base
of the unsymmetrical lobe.

4. Ventral surface, much weathered: a, central aperture; ), two convex
crested plates; 4', position of the third convex crested plate, which
has been removed; 0”, position of the unsymmetrical lobe, which has
also been removed; ¢, pseudambulacra.

5. Ventral surface of a nearly perfect specimen: a. three spearhead plates
projecting over the pseudambulacra (the sutures between them ard the
arched plates are well shown); 2, convex crested plates, ornamented
with the fluted tubercles (figs. 21 & 22): ¢, one of the piates bounding
the excentric aperture (see 4, fig. 20), probably representing half one
of the spearhead plates ~@; d, space which would be occupied by the
other half of this plate, with a small portion of the plate remaining
(see c, fig. 20’; e, the excentric-aperture plate (see figs. 13 & 14, and
d, fig. 20); f, the projecting lip of this plate; g, inward process of the
same; h, lid or valve covering the excentric aperture with its lateral
digitations; #, three pseudambulacra; /, space which would he occu-
pied by the fourth pseudambulacrum.

6. Dorsal surface: a, sutures between the forked plates; 0, 0, the two
forked plates; ¢, base of the unsymmetrical lobe.

7. One of the forked plates, exterior view (see 4, fig. 6): @, concavity into
which the apex of the pseudambulacrum is received: 8, crenulated
ridge; ¢, convex halves of the forked plate.

7*, Interior view of the same: a, same as in fig. 7; 4, angular shoulders.

8. Conjoined spearhead and convex crested plate, ornamented with fiuted
tubercles. a little laterally distorted: a, spearhead plate; 0, convex
crested plate.

9. Side view of the same: the dotted line in both these figures shows posi-
tion of suture between a and 6.

10. Side view of fig. 5: a, elongated plate, corresponding to half one of
the forked plates. 5, pseudambulacrum ; ¢, excentric-aperture plate.
side view; d, the lip-like projection of the game.

12

14

Fig. 11.

Fig. 12.

16.

17.

18.

19.
20.

23.
24.

25.
26.

ate

R. ETHERIDGE, JUN., ON ASTROCRINITES

Dorsal view of fig. 5: a, one of the pseudambulacra; 6, position of
the second pseudambulacrum ; ¢, apex of the third pseudambulacrum ;
d, base of the unsymmetrical lobe.

Puate XIII.

Dorsal view of a rather elongately lobed specimen: a, dorso-central

plate pressed upwards; 0, the base of the unsymmetrical lobe; ¢, ¢, ¢,
the three symmetrical lobes.

. Exterior view of the excentric-aperture plate: a, the pointed extre-

mity received into the re-entering angle of the plate or plates.

. Interior view of the same: a, the pointed extremity ; 4, circular cup-

like depression on which the apex of the lid appears to rest, placed at
the end of the inward ridge seen at g, fig. 5; c, channel leading in-
wards from it; d, folds into which the channel appears to break up.

. End view of the unsymmetrical lobe: a, the excentric-aperture plate

(figs. 13 & 14 &e.); 4 3’, the plate into which fits the excentric-aper-
ture plate, with the suture apparently dividing it into two; ¢, lip of
the excentric-aperture plate.
Interior view of the dorsal surface of a crushed-down specimen: a a’,
the plate or plates similar to 0 0' in fig. 15, crushed inwards; 3, 8, 8,
the three symmetrical {lobes ; ¢, c, hollows of the interlobular plates
into which the apices of two of the pseudambulacra fit, similar to a,
figs. 7 &7*; d, d, position of the other two interlobular spaces in the
perfect organism; ¢, interior of the dorsal surface, with a thin cover-
ing of matrix.
Enlarged view of a portion of one of the pseudambulacra: a, ambu-
lacral groove; 4, alternate, lateral short groove ; c, inner depression or
socket at end of 6; d, outer depression or socket at end of groove f
proceeding inwards from the slit or opening e; g, lateral plate, pro-
bably equivalent to the pore-plate in Pentremites; h, supplementary
plate, identical with the supplementary pore-plate in the same genus?
Diagrainmatic view of the cross section of a pseudambulacrum: @, am-
bulacral groove ; 6, more or less horizontal portion of the pseudambula-
crum ; ¢c, ridge forming an interspace between the inner pair of sockets ;
d, exposed sides of the pseudambulacrum bent down at a considerable
angle; e, concealed portion below the general surface of the organism.
Side view of a pseudambulacrum: a, base; 4, apex; c, exposed lateral
portion ; d, concealed lateral portion ; e, inward process.
The excentric aperture and plates surrounding it: a, the more or less
pyriform aperture; 6 & c, the side plates which complete the aperture,
and which probably correspond to one of the spearhead plates of the
other lobes (a, fig. 3); the plate ¢ is absent in the specimen, and its
position is therefore only indicated ; d, the excentric-aperture plate ;
é, its lip; f,pointed extremity of the plate which is received into the re-
entering angle of b 0’, fig. 15; g, lid or valve filling-in the aperture.

21 & 22. Tubercles of the arched and crested plates (4, fig. 5, and figs 8 & 9)

Microscopic spine attached by matrix to one of the specimens.

Small body (4) adhering to dorsal surface of fig. 3, supposed to be a
Foraminifer (Valvulina, sp.); a, tubercles of dorsal surface.

Greatly enlarged view of a specimen of Valvulina, sp., for comparison
with 4, fig. 24.

Dorsal surface of A. tetragonus, Austin, from a weathered specimen
kindly presented to me by Major Austin: a, cavity left by the dorso-
central plate, which has been accidentally removed; 6, b, the two
forked plates; ¢, the unsymmetrical lobe.

Pentremites cervinus, Hall (Geol. Rept. State Iowa, 1858, i. pt. 2. t. 25.
f. ila): a, a, two of the radial plates; b, suture between them.

N.B. -All the parts figured are magnified, their true size varying from ,}, inch
to ¢inch, Nos. 1 and 5, which are entire forms, are } inch in diameter. ‘The
remainder are individual parts enlarged. :

Quart Journ.Geol.Soc Vol XXXII PI XII

GSharman del.et lth. REtheridge Jun® direxit. M&NHanhart imp.

ASTROCRINITES ? BENNIE.

T
ay

Piexh.

=
i

Geol.Soc Vol. XXXI

t. Journ

Quar

« N Hanhart imp.

9
ev

M

R Etheridge Junt direst

G.Sharman del.et lith.

ale

ASTROCRINITES? BENNIE

i es
te 4
ror
Ja
.
‘ee
i " d
- r
¢
~
’
s
. E 7
.
‘
.
*
:
.
‘
.
i
4
'
\
*
‘
’
’
4

R. ETHERIDGE, JUN., ON ASTROCRINITES. 115

Discussion.

Prof. Morris complimented the author upon the care and labour
he had bestowed in working out this genus from the fragmentary
and minute specimens at his disposal. In the genera of Blastoidea,
to which this form belongs, the plates of the calyx vary in number
and form, as do also the orifices of the upper surface, in addition to
which the normal number of the pseudambulacra appears to be
only four, whilst in the other Blastoidea there are usually five. Oc-
casionally, however, some Echinodermata present the abnormal con-
dition of having only four rays. Besides the character of the rays,
the genus is further distinguished by being stemless, in which cha-
racter Astrocrinites, if rightly interpreted, bears the same relation to
the stalked Blastoidea that Marsupites does to the Crinoidea.

116 A. C. RAMSAY ON THE ISLAND OF ANGLESEY.

14. How Anettsey became an Istanp. By Professor A. C. Ramsay,
LL.D., F.R.S., V.P.G.S. (Read January 19, 1876.)

[Prate XIV.]

Tue structure of the low ground of Caernarvonshire within three or
four miles of the Menai Straits in almost all respects resembles that.
of Anglesey, both in its geology and physical geography. The
Menai Straits divide the two regions ; but carboniferous rocks form
the larger part of either shore, and the Straits may be considered
simply as a long shallow valley, the bottom of which happens to lie
beneath the level of the sea. The question thus arises, At what
epoch and by what means was Auglesey separated from the main-
land ?

The whole of Anglesey is low ; and only one steep escarpment, a
minor one, occurs in the island—that of the Old Red Sandstone
overlooking Traeth Dulas, which rises abruptly above the tidal flat
of the Traeth to the height of about 250 feet. (See Map and
Section Pl. XIV.).

The entire island may, indeed, be looked cn as a gently undulating
plain, the higher parts of which attain an average elevation of from
200 to 300 feet above the level of the sea; while most of its principal
brooks and small rivers run north-east and south-west, in depres-
sions with gently sloping sides; and onlv one inland valley, with the
same trend, is of any marked importance, namely that of Malldraeth
Marsh, in which the small coalfield lies. There are, however, a few
exceptions to the averagelevelsmentioned above—the summitof Holy-
head mountain being 709 feet, and Garn, near Llanfairynghornwy,
598 feet above the sea, while the greatest elevation crossed by the
sections of the Geological Survey (sheet No. 40) is only about
400 feet high.

On the opposite side of the Straits the same kind of low undu-
lating scenery prevails for several miles inland, with the same kind
of minor north-east valleys, one marked instance of which occurs in
a long shallow and narrow valley in or alongside cf which the Caer-
narvon and Bangor road runs for several miles.

The surface of the ground on both sides of the Straits is to a con-
siderable extent composed of glacial detritus, with erratic boulders,
large and small (from the north), gravel, sometimes sand, and clay
from which any number of ice-scratched stones may be gathered
from well-exposed sections, as, for example, in the boulder-clay coast
cliff of the Mount at Beaumaris (which contains minute fragments
of sea-shells), or anywhere else in similar cliffs round the shores of
Anglesey, or inland in occasional pits and fresh cuttings on both
sides of the Straits. ‘Uhrough these glacial accumulations the rocks
of the country frequently appear, sometimes in barren tracts of
considerable extent, sometimes in small isolated bosses of gneiss or
grit, often covered with heath or furze, while the more fertile grounds

A. C. RAMSAY ON THE ISLAND OF ANGLESEY. 117

of the whole of Anglesey consist chiefly of glacial detritus, with here
and there small alluvial meadows by the sides of the streams.

When freshly stripped of glacial débris, or even of a mere thin
turfy soil, the underlying rocks are often found to be ice-smoothed
and marked with glacial strive, running generally from about 3U° to
40° west of south. The larger valleys of Malldraeth Marsh and
the Menai Straits (with others of minor note) run in hollows in
the same general direction.

This memoir is not the place in which to discuss the theory of
the Glacial epoch ; but for my general argument it is necessary to
touch on one or two points connected therewith.

It is well known that in all mountain regions where glaciers
exist, or have in past times existed, the disturbances of the earth’s
crust that produced the elevation of the mountains go back to
periods long antecedent to the last great Glacial epoch. Thus the
first great upheaval of the Alps is of pra-Miocene age, and the
mountains of Scotland and Cumberland were mountains before Old-
lted-Sandstone times, while the last great movement of the rocks of
Wales is certainly older than the Permian epoch, and, probably, like
the mountains of Cumberland, very much older,

There was therefore plenty of time, in what is now Wales, long
before the beginning of the great Glacial episode, for the more ordi-
nary agents of denudation to have formed deep valleys, down which,
when that episode began, the growing glaciers might gravitate,
deepening their channels as they pressed forward, and mamillating
and striating the rocks over which they slid; for the great original
valleys of the mountains were by no means entirely scooped out, but
merely modified by the glaciers.

Thus, for example, it happens in Wales that all the striations in
the valley of Dolgelly and the estuary of the Mawddach, in Merio-
nethshire, follow the south-westerly trend of the valley, the glacier
that filled it when at its greatest being fed by the snows of the
slopes of Cader Idris and Aran Mowddwy and those of the tributary
valleys of Afon Eden and the Mawddach, that joimed it from the
north, while from a central low watershed near the sources of the
Wnhion, another branch pressed north-easterly into and beyond the
region now occupied by Bala lake.

The striated rocks exposed at low tide in the estuary of the
Mawddach, and the islet-like heathy bosses of rock that stand out
amid the marshy moss opposite Barmouth, are merely roches mou-
tonnées, once buried deep beneath the glacier that pressed forward
into Cardigan Bay.

In like manner all the western valleys of the Cambrian mountains
of Merionethshire, from Barmouth to Diphwys west of Trawsfynydd,
such as those of Afon Atro and Ardudwy, are marked by deep
grooves and striations pointing more or less westward, according to
the trend of the valleys. The broad flats and roughly hilly, but not
mountainous, country of Cors-goch, Afon Eden, Trawsfynydd, and the
ground bounded by the amphitheatre of scarped mountains formed
by the Arenigs, the Manods, the Moelwyns, and the Cambrian steeps

Du

ius A. G. RAMSAY ON THE ISLAND OF ANGLESEY.

of Graig-ddrwg were at the same time filled with deep accumulations
of snow and ice, from which were discharged a radiating series of
glaciers, one pressing southward to swell the ice-stream that filled the
valley of what is now the estuary of the Mawddach, another through
the Pass of Afon Treweryn between Arenig Mawr and Arenig Bach
eastward towards Bala and the valley of the Dee, whilst a third, swelled
by all the snows of the Manods and the Moelwyns, flowed south-west,
into what is now the broad flat of Traeth Bach, there to be joined
by another great ice-stream which, partly descending from the high
eastern slopes of Snowdon, filled Nant Gwynant, and debouched into
the area now occupied by the marshy flat of Traeth Mawr. In all
of these the directions of the striations necessarily conform to the
trend of the valleys—easterly, southerly, or south-west, as the case
may be.

I mention these specialities of striation partly to show that the
Welsh group of mountains, as a whole, was not overridden by a great
universal ice-sheet proceeding from the far north, as some have
ventured to suppose. In the main, every large valley was filled by its
own special glacier—a circumstance long ago stated in a broad way
by Buckland, and definitely worked out by me for Caernarvonshire
in my essay in the first volume of ‘ Peaks, Passes, and Glaciers,’ after-
wards republished in a little book. In North Wales, when the
country was absolutely full of snow and ice, each great valley yet
maintained its separate ice-flow. In such a case, however, there
were many deviations consequent on under and upper ice-currents,
the upper parts of glaciers diverging from the direction of the under-
flow, and passing across what are now low watersheds, like that
of Llyn Cawlyd, into neighbouring vales, such as that of the Conwy—
a circumstance to which special attention has been called by the
Reverend W. T. Kingsley, while engaged within the last few years
in sounding the lakes of North Wales, with the view of proving
them to be true rock-bound basins, in which he succeeded.

Such statements as these are preliminary to, and necessary for
the understanding of, the main point of this paper, which I will now
discuss.

On the north-west slopes of the Snowdonian range* (see Map, PI.
XIYV.), great glaciers poured their ice-streams down the valleys of
Llyniau Nant-y-llef to the west, and of Llanberis and Nant-ffrancon
to the north-west and north; and these never quite reached the region
now occupied by the Menai Straits, but, escaping from the higher
bounding-walls of their valleys, spread out in the shape of broad fans
on the north-western slopes of the minor hills that now overlook the
Straits. This is proved by the northerly curve of the glacial striations
at the mouth of the Pass of Llanberis, on the flatter area above the
steep slopes of the slate-quarries by Llyn Peris and Llyn Padarn, and
strongly hinted at by the long, smooth, grass-covered terraces of rock
that pass from the opening of the valley of the Ogwen north-eastward
along the seaward slope of Moel Wnion, and are still further continued

* Tuse the word range as a convenient term. There is no range of moun-
tains in North Wales. Taken collectively they form a great group.

A. GC. RAMSAY ON THE ISLAND OF ANGLESEY. 119

across the Aber valley towards Penmaen-mawr. These, when stand-
ing on or near them, are barely appreciable—but seen from Beau-
maris across the Lavan sands, form unmistakable features in the
landscape. I have often thought (though this is yet unproved) that
the broad mound of glacial débris on which Penrhyn Castle stands
may only be the relics of the terminal moraine, at a comparatively late
period of the history of the large glacier that flowed down the valley
of Nant-ffrancon and the Ogwen, fed by the snows of the Glyders,
Foel-goch, Mynydd Perfedd, and Elidyr-fawr on one side, and by
those of Carnedd Dafydd, Carnedd Llewelyn, and Foel-fras on the
other.

It thus appears that at this period the glaciers of Snowdonia did
not cross the Menai Straits into what is now the island of Anglesey ;
and it is therefore clear that the north-east and south-west striations
' which mark the whole of that broad region must have been pro-
duced by some other power.

These striations point directly to the mountains of Cumberland, a
country which, lying further north, was at one time buried so deeply
under snow and ice, that almost all its mountains look simply like
gigantic roches moutonnées ; for Cumberland was far more intensely
glaciated than the more southern region of Wales. From Cum-
berland a vast mass of ice flowed southward ; and, reinforced by
the ice-streams that came from the mountains of Carrick, in the
south of Scotland, and perhaps even from the basin of the Clyde,
it overspread the region now occupied by the shallow sea of
Morecambe, Lancaster, and Liverpool bays, that lie between Cum-
berland and Anglesey (nowhere more than 30 fathoms deep), and
pressing still further to the south-west, covered the whole of the
low ground of Anglesey, and went to some unknown distance beyond.
Furthermore, in my opinion, so great was the size and power of this
ice-flow, that it hindered the glaciers of Llanberis and Nant-ffrancon
from encroaching on the territory of Anglesey, and these simply
joined the larger on-flowing glacier as minor tributary ice-streams.
For this reason it happens that the glacial striations of Anglesey, as
we might at first expect, do not point towards the old glacier-valleys
of Snowdoma that open on the Strarts, but run at right angles to the
courses of these comparatively minor glaciers.

If we now turn to the rocks that form the banks of the Menai
Straits, we find that they chiefly consist of nearly flat-lying Carbo-
niferous strata (see Map and Section, Pl. XIV); and looking at the
disposition of these strata from Traeth Melyn, opposite Caernarvon,
to Llanfair-pwll-gwyngyll, in Anglesey, and on the opposite shore
from Caernarvon to Bangor, there is reason to believe that from end
to end they once filled the whole of the region now occupied by the
Straits. The larger part of this region, as it now exists, is of Carbo-
niferous Limestone age; but it by no means consists entirely of solid
limestone. On the contrary, numerous bands of shale and friable
sandstones and conglomerates are intermingled with the limestones,
together with beds of soft red marl. On the coast opposite Caer-
narvon, the low cliffs are entirely formed of red marl overlying the

120 A. C. RAMSAY ON THE ISLAND OF ANGLESEY.

limestone ; and on the Caernarvonshire coast for three miles north
of the town, also overlying the limestone, there are soft shales of
the Coal-measures, sometimes red and marly, and containing thin
seams of coal.

In Anglesey, from three to four miles north-west of the Straits,
lies the valley of Malldraeth Marsh, the rocks of which also consist
ot Carboniferous Limestone, Millstone grit, soft Coal-measure shales,
with a little sandstone, beds of coal, and Permian strata; and this
valley, nine miles in length, runs almost exactly parallel to the valley
of the Menai Straits. Many years ago, at its north-eastern end,
I saw deep glacial striations on the Millstone Grit, running straight
down the shallow valley towards Caernarvon Bay.

Considering that the south-westerly trend of each of these valleys
(and of others of minor note) corresponds with the general direction
of the glacial striations of Anglesey, and therefore with the onward
course of the great glacier that produced them, I have been led to the
conclusion that both of the shallow valleys were scooped out in com-
paratively soft rocks by the grinding power of the vast glacier coming
from the north-east, and that when in the course of time the
climate ameliorated and the glacier disappeared the sea flowed in
where part of the glacier had been, and Anglesey first became an
island, the islets in the narrower and shallower part of the Straits
at the Menai and tubular bridges being merely weathered roches
moutonnées, once overridden by the moving glacier.

Malldraeth Marsh, as its name (the sodden sands) implies, is for
the most part deeply buried under watery moss and alluvium; the
river that traverses it is artificially banked and tidal for miles from
its mouth; and if the alluvium that covers its surface, and the blown
sands at its mouth were removed, it would again become, as it was
in old times, an arm of the sea, a kind of low-banked fjord (like the
Menai Straits), about nine miles in length, but, unlike the Straits,
closed at its north-east end. Nowhere in the Straits is the water
more than from 6 to 10 fathoms deep at high water ; and around the
islets near the bridges it varies from 3 to 7 fathoms. An elevation
of the country of only 42 feet, would in time, by silting up and sub-
sequent accumulation of alluvium, turn the north-eastern and south-
western parts of the Straits on either side of the islets into valleys
like that of Malldraeth Marsh.

There are some details connected with the partial submergence of
North Wales during part of the Glacial epoch which it is needless
to enter on with respect to this special subject. Anglesey was, after
emergence, apparently joined to Caernarvonshire by an undulating
plain of Boulder-clay ; but by and by, through marine waste and
subaerial drainage, this material was so much worn away that only
the relics of it now remain in occasional cliffs and low banks on
either side of the Straits. Then it was that at length Anglesey
fairly became an island such as it now appears, though the long
channel through which the waters of the Straits flow was ground
out at a comparatively early part of the Glacial epoch.

Such I believe to have been the physical history of Anglesey

A. C. RAMSAY ON THE ISLAND OF ANGLESEY. 137

before, during, and after the Glacial epoch; and thus it was that
Anglesey got separated from the mainland of North Wales by what
is, after all, merely a long and broad glacial groove.

This with me is no hasty conclusion. Having at intervals been
much in Anglesey, and occasionally living on the banks of the Straits
for years, ! wondered what might be their meaning. It is difficult or,
perhaps, impossible to look at this long narrow channel without the
idea entering the mind that at some unknown but comparatively recent
time it may have formed part of the course of a large river flowing
from some unknown land now: submerged or destroyed by denuda-
tion ; and this hypothesis possessed me so strongly for years, that,
though always in doubt and in search of a better, it helped to blind
me to any other. Last summer, however, while looking at the
Straits, the whole truth of the case seemed suddenly to flash into my
mind. ‘The result I have given in this paper.

EXPLANATION OF PLATE XIV.

Geological Map of Anglesey and the adjacent parts of North Wales. with a
contoured section of the form of the ground along the line A B.

Seale of Map, 12 miles to an inch.
Seale of Section, 4 miles to an inch.

Discussion.

Mr. Hicxs said he did not agree with Prof. Ramsay that the
Menai Straits were formed by a glacier passing over the land. The
hollow was made by the great changes which took place towards or
at the close of the Paleozoic epoch. The great N.E. to S.W. faults
on both sides of the Straits dropped the Carboniferous beds against
the older and harder Paleozoic rocks, and a depression containing
broken-up and soft beds was thereby formed at the close of the Paleo-
zoic epoch. During the Glacial epoch the land here stood very much
higher than at present, and the present hollow formed then a valley
above sea-level. ‘The separation cf Anglesey from the main land
occurred at a comparatively recent period, and is entirely due to the
gradual encroachment of the sea over a gradually subsiding area. The
subsidence which has taken place along the western coast, even in
historic times, has been considerable in extent; and the sea is still
continually encroaching on the land and along the lines of valleys.

Mr. J. F. Campsett confirmed the author’s views. He had found
strie high up on the Snowdonian range, in Ireland, and all over
the British Islands, pointing in a direction towards Scandinavia, and
thence towards the North Pole.

Mr. Cartes Moorz remarked that the Mendips were upheaved
before the formation of the New Red Sandstone, which lies uncon-
formably upon the up-turned Coal-measures, and that the Carboni-

L232 A. GC. RAMSAY ON THE ISLAND OF ANGLESEY.

ferous strata of that region had been pounded and crushed, so as
greatly to facilitate their removal by denuding agencies. He in-
quired whether Prof. Ramsay could define more closely the period
of the upheaval near the Menai Straits.

Mr. Henry Woopwarp remarked that since the Glacial period a
considerable subsidence of Anglesey had taken place, and instanced,
in support of this view, the finding of a jaw of Elephas primigenius
associated with old forests at a great depth during the excavations
for the enlargement of Holyhead Harbour. He thought that ice
from Cumberland would suffice for the glaciation of the surface of
Anglesey without going so far as the North Pole.

Mr. Forpnam inquired why we should go to Cumberland for ice
when Snowdon was so near. If thére were ever a polar ice-sheet,
the mountains of Cumberland and Wales, he thought, were not
worth mentioning. He inquired whether there were any erratic
blocks in Anglesey which might indicate the source of the ice by
which it was glaciated, and also whether the Great Orme’s Head
was striated. He thought there was some reference to the Straits
having been much shallower than at present within historic times.

Mr. Drew confirmed Prof Ramsay’s views. The Straits are now
margined by cliffs, owing probably to erosion subsequent to their
formation.

Mr. Jorpan remarked that faults, as a rule, do not affect the
surface.

The Presrpent (Mr. Evans) remarked that he understood Prof.
Ramsay to maintain that the general features of the country were
preexistent to the Glacial period, but that where there were softer
rocks the ice cut through them, the hollows thus formed being after-
wards often filled up and reexcavated by subaerial denudation.

The AvrHor, in reply, stated that at the close of the Carboniferous
period Anglesey was covered with Carboniferous strata, and that,
although faults did occur in places, letting down portions of these
rocks, these had nothing to do with the production of the Menai
valley. The Millstone Grit and Coal-measures, not less than 1500 feet
thick, originally lay over the Carboniferous Limestone of Menai Straits
when the faults were produced. The ice came from the north-east
and not from Wales; and the striations pointed directly towards
Cumberland. The Great Orme’s Head could hardly show any strie,
as its surface had been greatly affected by weathering. Many erratics
occur near Beaumaris which are similar to the rocks of Cumberland or
Galloway, and which may perhaps be identified with them.

*

Quart. Journ. Geol. Soc Vol. XXXII. Pl. XIV.

LIVERPOOL
BAY

YB)
Snowdon

“/
WT

NY
\

Z ms \

Z
A
d)

°

})
\
=dllinN\

) Pn~kpemees

CR
S Ds ates wht

A
Ss See

NW

Dinas

Mintern Bros. hth.

Quart. Journ. Geol. Soc Vel. XXXI.Pl XIV.

LIVERPOOL
BAY

New Red. Sandstone.

iH Coal) Measures
IMM) De2lstorne Grit,
555 Carboniferous Lunestone.

*CHES

SSS et Bed. Sandstone
BW wala Shale Grit &e.

Lower Silurian, Cambrian
& Igneous Rocks

4)

Ce

GEOLOGIGAL MAP

ANGLESEY

and the adjacent part

ill is

RY

ay Act ss
S iE
a =: s g g Z E Snowdon

Glocho

Wee Dinas
Straits
ei eee

Mintern Bros hth

ON THE FOREST-BED SERIES AT KESSINGLAND AND PAKEFIELD. 123

15. On the presence of the Forust-nep Serres at Kusstnetanp and
PakErtELD, in SuF¥YOLK, and its position beneath the CuILLEsroRD
Cray. By Joun Gunny, Esq., M.A., F.G.S. (Read November
17, 1876.)

Tnx following remarks, together with the accompanying section and
the Elephantine and Cervine remains exhibited from the soil of the
forest-bed at Kessingland and Pakefield, are intended to be supplemen-
tary to a paper “ On the Relative Position of the Forest-bed and the
Chillesford Clay in Norfolk and Suffolk, and on the Real Position of the
Forest-bed”*. Ihave but little to add to that paper and nothing
in it to withdraw or alter. My object is to describe the rootlet-bed,
part of the freshwater formation, which succeeded the Forest-bed,
and to exhibit the Elephantine and Cervine remains as proofs that
the soil of the Forest-bed lies beneath the Chillesford Clay.

As far back as 1868 Professor Huxley, then President, in dis-
cussing a paper by Mr. Prestwich on the Red Crag of Suffolk, ob-
served that Mr. Gunn had stated as “a fact, and a, very important
fact,” that he had seen the Forest-bed exposed on the beach at
Easton Bavent, Kessingland, and Pakefield underlying the Chilles-
ford Clays and Sands. This was opposed by Mr. Prestwich, who
said he had seen the roots of trees in the Chillesford Clay, which he
considered to be a proof that the Forest grew above it and upon it.
It has also since been stated by others who have visited the spot,
and have not been able to see the Forest-bed in consequence of its
being covered with beach-sand and shingle, that Mr. Gunn’s ipse
dixit could not be accepted. I determined therefore to collect all
the specimens I could, and carefully to note the beds from which
they were derived. It would be foreign to my purpose to describe
and particularize the specimens, which would fill a long paper; but
it will suffice to say that they correspond with those from similar
beds at Bacton and Cromer. If the reader will turn to the section
(fig. 1), which is a general one of the cliffs, about three fourths of
a mile in extent, I will indicate the beds from which they have
been obtained.

No. 7, at the foot of the cliff, represents the estuarine soil of the
Forest-bed, on which the forest grew after this bed was raised above
the level of the water. It consists of two parts, blue clay and gravel,
the latter called the “ elephant-bed,” as described in my former paper;
and from these two deposits every specimen exhibited has been ob-
tained ; the colours of the specimens mark their respective beds ; or,
if any be beach specimens, the matrix will do so equally. This bed
ranges to the full extent of the Forest-bed ; and its thickness has not
been ascertained. It lies between high- and low-water mark ; and
the water rises so as to prevent one’s searching with the spade.

No. 6 represents the Forest-bed, upon the soil last mentioned.

* Quart. Journ. Geol. Soc. vol. xxvi. p. 551 et seqq.

124 J. GUNN ON THE FOREST-BED SERIES AT

This I have seen here on one spot (a mere patch of a few yards
length) which has escaped denudation. It is well developed, with
blackened earth and large stools and stems of trees; but I did
not find any mammalian remains in it; and Mr. Thomson, of Pake-

Fig. 1.—Section of Cliff at Kessingland and Pakefield.

Sueface sols ois dhe apace

-
|

—

. Upper Boulder-clay ...
oolitic.

on rocks, chiefly
)

Seg aeta aes

~)

iitiddle Deft y. 122k.

t

A ee ee ee
=~

TJAMINATED SERIES.

Cetacean bones, marine

NC Fin baie chs mabananyil
shells.

|

| 4, Norwich Crag (fluvio-
\ IATANE), siangsinsabith,iss
(
|

|
5. Rootlet-bed and Unio- Ee 2
bed (freshwater)......

‘| Freshwater & marine shells.
Mammalian remains (rare).
Freshwater shells abundant
in black Unio-bed.
Present beach level.
Elephas antiquus, Rhino-
ceros etruscus, Trogon-
therium Cuviert.

6. Forest-bed with stools { =
On taSee (52. ssh. Fale. | ry

Elephas meridionalis, Rhi-
noceros megarhinus
(rare). Cetacean re-
mains (common).

7. Soil of the forest-bed
(Elephant-bed and
Blue Clay; estua-
rine).

ForrEStT-BED SERIES.

field, whom I employed to collect specimens, says that he never
found any except in No. 7, in which they abound for about three
quarters of a mile. I mention this to show from what bed the spe-
cimens exhibited are taken, and not to affirm that they are confined
to No. 7, because a fine molar of Elephas antiquus has been ob-

KESSINGLAND AND PAKEFIELD, IN SUFFOLK. 125

tained from the Forest-bed here; and the researches of Mr. Blake,
F.G.8., of the Geological Survey, have proved that mammalian re-
mains are to be found in No. 5.

No. 5 represents the freshwater beds, about ¢ feet in thickness,
formed on the subsidence of the Forest-bed. They consist in part
of a black soil, only a few yards long, with freshwater shells, cor-
responding with those at Mundesley and Kunton, commonly called
Unio-beds. They consist in part also of the rootlet-bed, about 500
yards long, of green oozy clay, similar to that at Runton, pierced
vertically with small roots, about the size of the little finger. This,
Mr. Prestwich has pointed out as an indication of the presence of the
Forest-bed (Quart. Journ. Geol. Soc. vol. xxvii. pp. 463, 464). With
all submission to so high an authority, it appears to me to represent
brushwood, which succeeded the true forest very extensively. The
rootlets are very different from the grand roots which extend laterally
from the stools of the forest-trees ; and, besides, even these could not
be expected to grow from such decidedly marine formations as
the Chillesford Clays or Sands, which were continuously being sub-
merged in deeper and deeper water, till the Westleton (or pebbly) beds
and the Lower Boulder-clay were deposited without any break.

On the forest going down, this freshwater bed seems to have been
formed; and the rootlets remain in it. At Happisburgh there appears
to have been a hill, like that at Kessingland, which continued above
water after the submersion of the lower land. On this hill hazel-
nuts and the bones of sheep or goats, now in the Norwich Museum,
have been found; and a part of the sludge, as Dr. Falconer called it,
was carted away at his suggestion; and a portion of it, full of the
leaves of the willow, is now exhibited. This is evidently more recent
than the Forest-bed, and probably of the same age as the rootlet-
bed at Pakefield, Kessingland, and Runton.

No. 4 represents the Fluvio-marine deposits, about 2 feet thick,
which succeeded the fresh water on the influx of the sea. Crag-
shells have been seen here, but rarely. The Norwich fluvio-marine
Crag most probably belongs to this division, which, at Bramerton,
underlies the next-mentioned deposit. Here, as at other places, an
interval of several miles may be interposed between beds of crag-
shells, and yet the deposits are continuous and are stratigraphically
the same. This remark applies equally to No. 3, next mentioned.
The Chillesford Clay sometimes yields shells, as at Easton Bavent,
and, slightly, at Aldeby; but at Bramerton, Brundall, Wroxham,
Horstead, Coltishall, Hoveton, Barton Turf, Ludham, Burgh, near
Aylsham, and other places it is destitute of them; and yet, from its
mineral character, it can scarcely be mistaken. The fluvio-marine
beds, both here and at Happisburgh, are cut short by the rise of
the freshwater deposits, and are but partially and faintly represented.
Owing to this cause, at Happisburgh the Lower Boulder-clay rests
immediately upon the Forest-bed, without the intervention of any
of the intermediate beds.

No. 3 represents the marine formations, and contains the Chilles-
ford Sands and Clays, which are finely developed here, notwithstand-

126 J. GUNN ON THE FOREST-BED SERIES AT

ing that Messrs. Wood and Harmer, in their Paleontographical
Monograph almost ignore them, except at the Pakefield-Lighthouse
Gorge, near which they represent them as dipping down or trun-
cated and not reappearing throughout the entire length of the coast.
They average about 12 feet in thickness,

No. 2 has generally been regarded as the Middle Drift; but, from
the admixture of occasional plots of pebbles, I am inclined to call it
the Pebbly Bed, or Westleton Sands. The upper part is, perhaps,
Middle Drift. It is about 15 feet thick. _

No. 1 represents a grand display of the Upper Boulder-clay, ave-
raging 15 feet; above is the warp and humus, 4 feet.

If I have erred in any respect, I trust I shall be corrected by the
geological surveyors now at work in the ncighbourhood.

Discussion.

Mr. A. Tytor inquired whether the lowest beds in Mr. Gunn’s
section represent the Antwerp Crag, as in this case the succeeding
deposits would represent the other three Crags. Great quantities of
fossil bones and teeth had been collected at Antwerp during the con-
struction of the fortifications of that city.

M. Cuar.eswors could not understand how the Norwich Crag
can be above the Forest-bed. Remains of Mastodon have not been
found in the Forest-bed ; but they occur in the Norwich Crag.

Mr. Guyy said that it was now agreed at Norwich that the so-
called Mammaliferous Crag is not really Mammaliferous, but that
the mammalian remains found in it are derived from the Stony Bed
resting immediately upon the Chalk. Dr. Falconer entertained the
same opinion.

Mr. Cuarresworts stated that what he wanted was the evidence
upon which the bed indicated in the section brought before the
Society by Mr. Gunn was identified with the Norwich Crag.

Mr. Gwyn Jerrreys remarked that certain Arctic shells oecur
in the Chillesford Clay, and instanced particularly Leda hyperborea,
a species which was obtained from a depth of 450 fathoms in the:
Norwegian Sea, and which he had found in 175 fathoms at the
entrance to Baffin’s Bay. The Lede are deep-water species ; and
the occurrence of L. hyperborea in the Chillesford Clay would seem to
indicate that that deposit had been formed at a considerable depth,
and afterwards upheaved.

Mr. Wuriraker stated that he had been with Mr. Blake along the
coast to examine the lower beds referred to by Mr. Gunn. Mr.
Blake had clearly established that in the case of these beds, which
were only about 20 feet thick, different observers had given different
names to the same things, and in fact had given names to con-
ditions and not to beds. He was glad that Mr. Gunn had called
the ‘* Rootlet-bed ” by that name, as it was certainly not a ‘ Forest
bed” though it had been so called. From the varying nature of the
coast, and the fact that the sections were interrupted, he could easily
understand the difficulty that had been experienced in establishing

Land

KESSINGLAND AND PAKEFIELD, IN SUFFOLK. 127

the sequence of the deposits and their identity in different sections.
He thought that the Forest-bed, the Chillesford beds, the “laminated
beds” and the gravelly sands belonged to the same series. It is a
question whether the so-called Norwich Crag of the coast is the same
as that inland; Messrs. Wood and Harmer think that itisnot. He
would caution geologists against forming subdivisions among these
beds, as they vary constantly. He thought it probable that the
Norwich Crag is newer than the Red Crag, and that there is no
break between the Crag and the Drift.

Prof. Szerry stated that in 1863 he had been along the whole of
the coast, in company with the late Prof. Sedgwick. The coast was
then quite clear, and the beds could be traced all along. The Forest-
bed was full of trees, with bones &c. lying about among their roots.
Kast of Cromer a series of beds was shown different from those to
the west. To the west of Cromer the Norwich Crag changed its
character, being reduced to a thin calcareous band half an inch thick.
in which shells such as usually occur in the Norwich Crag are to be
met with. This thin bed rested immediately upon the Forest-bed,
which was therefore below, but possibly in part contemporaneous
with the Norwich Crag.

Mr. Woopwarp considered that Mr. Gunn had done great service
to paleontology by the indefatigable zeal with which he had collected
the mammalian remains from the soil of the Forest-bed. He con-
sidered that Mr. Gunn had conclusively proved the existence of the
Forest-bed all along the coast; but in former times it must have
extended far out into the German Ocean, as specimens of Mammoths’
teeth and tusks have been dredged up in abundance by the trawlers
off the Dogger bank, together with a nearly perfect skull of Rhino-
ceros leptorhinus. The Antwerp collection mentioned by Mr. Tylor
belonged to the same series as the so-called Coprolite diggings ;
many of the shells were distinct from those of any of the Crags. He
quite agreed with Mr. Gunn that all the mammalian remains found
in the Norwich Crag belonged to the Stony bed at the base, resting
directly on the denuded surface of the Chali.

Mr. CuarteswortH reiterated his question as to the evidence by
which the bed marked Norwich Crag in the section exhibited by
Mr. Gunn had been identified. He maintained that the newer beds
of the Norwich Crag could only be identified by means of their cha-
racteristic shells; and he asked Mr. Gunn what evidence of this
nature he had to adduce in support of his opinion that the deposit
in question was Norwich Crag. He denied positively that the mam-
malian remains were only found in the Stony bed.

Prof. Ramsay congratulated the author on having proved the
existence of the Forest-bed, and expressed his gratification at seeing
such local points of great geological interest so carefully worked out
by observers residing on the spot.

The AvrHor thanked Mr. Charlesworth and the President for dis-
cussing the question as to the identity of the Norwich Crag. He
maintained that there had been a regular sequence of deepening
conditions. At Haston Bavent the Chillesford Clay rested on Nor-

Q.J.G.S. No. 126. K

128 ON THE FOREST-BED SERIES AT KESSINGLAND AND PAKEFIELD.

wich Crag; but he did not think that it was necessary that shells
should be found in order to prove the nature of the latter bed: in
fact he considered that far too much weight was laid upon concho-
logical evidence in such matters, and far too little upon stratigraphi-
cal considerations. He differed from Mr. Prestwich with regard to
the position of the Forest-bed, which no doubt extended far under
the North Sea, perhaps even as far as Belgium, the beds in which
country probably correspond with those of Hast Anglia. He stated
that the Stony bed is not to be seen at Pakefield, because the
Chalk upon which it immediately lies dips down at a high angle
and passes at some depth below the Forest-bed.

ee)

ON THE PRECIPITATION OF CLAY SUSPENDED IN WATER. i2

16. On the Inrnvunce of various Substances in ACCELERATING the
Precipitation of Ciay suspended in Water. By Witt1aM
Ramsay, Esq., Tutorial Assistant in Glasgow University Labo-
ratory. With a Note by Pror. Ramsay. (Read March 8, 1876.)

(Communicated by Prof. Ramsay, F.R.S., V.P.G.S.)

Ir has been noticed by several observers that clay suspended in
water settles more quickly if the water is salt than if it is fresh.
This fact is contrary to what would naturally be supposed—namely,
that suspended matter should settle more easily in a liquid of low
than in one of high specific gravity.

The earliest notice which I have been able to find relating to
this subject, is a note by Skey in the ‘Chemical News’ for 1868,
p- 160, in which he advances the following hypothesis to explain
the phenomenon.

After taking notice of the fact that those salts which accelerate
the precipitation of clay are too stable to be decomposed by mere
contact with it, he suggests that such salts have a strong affinity
for water compared with that of clay for water. They therefore
tend to abstract water from the clay, and cause it to coagulate and
settle. In corroboration of this he instances the fact that iron
ferrocyanide is thrown down by numerous salts; that silica is
precipitated from its solution in ammonia by chloride of ammonium;
and that nitrate of baryta is precipitated by nitric acid.

In the ‘Chemical News’ for August, 1874, Durham advances an
electric hypothesis ; he supposes that clay, in falling through water,
** generates electricity by friction, and, as water is a bad conductor,
the difference in potential between the clay and the water remains
for some time, hence they are mutually attracted; but when an
acid or salt are added, the liquid becomes a good conductor, the
potentials are equalized, and the clay falls.”

Durham also states that in solutions of the same salt of different
strengths, the rate of settling is in the order of the specific gravity
of the solutions, and that clay remains longest suspended in the
liquid of highest specific gravity.

Before reading either of these papers my attention was drawn to
this subject by a communication from Mr. Peter Robertson to the
Geological Society of Glasgow. It struck me that a probable solu-
tion of the question was to be found in the relative amounts of heat
absorbed by various salts in passing into solution.

Does the rapidity of precipitation of clay in-solutions of salts
bear any relation to the absorption of heat by the salts in passing
into the liquid state? In order to answer this question, some
experiments were made which I shall now proceed to describe.

Some very fine clay was procured, which on being put into water
broke up of its own accord into very minute particles. A number
of solutions of different salts were prepared, and the specific gravi-

K 2

130 W. RAMSAY ON THE PRECIPITATION

ties determined. A measured quantity of the muddy water was
added to each solution, and the rates of settling compared with one
another. It was found impossible to obtain any definite measure
of the rates of settling; for neither by weighing the clay which had
settled in a given time, nor by comparing one stratum of mud
settling in one solution of salt with another stratum in another salt
solution, could a reliable measure be obtained. It was always easy,
however, to pronounce which was settling most quickly.

The results will be seen in the following Table. In each case
one part of salt by weight was dissolved in four parts of water.
The salts are arranged in the order of settling, those in which the
clay subsided most quickly being placed first, with regular progres-
sion to the one which kept the mud longest in suspension.

Reduction of temper- Specific gravity of

Salt. ature on dissolving. solution at 10° C.
Ammonium chloride ........ 15:19 C. 1:0604
Ammonium nitrate ........ 14°10 1:0863
Potassium chloride.......... 11°81 1:1361
poditim aiitratel,.«. ac. awlads x 9°46 1:1355
Barium -chloride:'. ..!ea.cs ss 4-50 11733
Magnesium sulphate ........ 4:50 1:1035
Asie eulphate: sis pa Sytecs Ault 3°10 1:1258
Sodium ehloride.....i.5. «6 .% 2°10 Lelibis
head mitratedlic. acceansiiehk te 1:90 1:2099

It will be at once seen that those salts which absorb most heat on
going into solution allow clay to deposit most easily.

In order to ascertain if the fluidity of the solution had any
influence on the rate of settling, the rate of flow through a capil-
lary tube, of the following six salts, was determined. The numbers
are seconds, and are simply comparative. ‘The temperature for all
was 15° C.; and there was a constant pressure of one metre of
water.

Time of flow through

Salt. a capillary tube.
seconds.
Ammonium chiecride .....«ssaesoeeun Sree
Ammonia) MUPBLE iwi ingavact ou walls 366-0
Potanstim enorme, ious uaidaadsacsetion 381:0
OCU Hires, dais yko. th. teas hk we 456-0
Barigm wiieride. ts sd. |ic eck tetl cih ase 455-0
HOdiid ChIGKHIO.. 55). alee ath wel Leak 554:0
VAP, ged soraihes «lays .jorwibanions beeareke tease 380:0

The order holds good for several of the salts, but it is not constant
enough to allow of the deduction of a general rule. A connexion
appears, however, to subsist between the fluidity of the solutions
and their absorption of heat on being dissolved. Water flows more
rapidly through a capillary tube than any of the solutions, except

OF CLAY SUSPENDED IN WATER. 131

those of ammonium chloride and nitrate ; and yet mud settles much
more slowly in water than in any of the solutions.

Clay does not deposit so quickly in acid solutions as in solutions
of salts. The order of settling for the four acids hydrochloric,
nitric, sulphuric, and acetic acids of 16 per cent. was

1. Nitric acid.

2. Hydrochloric acid.
3. Acetic acid.

4, Sulphuric acid.

The difference between the rates in hydrochloric and acetic acids
was slight, but still perceptible. Hydrochloric acid has a solvent
action on the iron contained in the mud.

Mud deposits more rapidly in a solution of caustic soda than in a
solution of potash; and as potassium hydrate evolves more heat on
solution than sodium hydrate, the rule appears to hold in their
case.

It appeared desirable to ascertain if the rate of deposition varied
with the density of the solution. For this purpose four solutions of
common salt were prepared, the strongest containing 100 grains
dissolved in 500 cub. centims. of water; the second was half-
strength, the third one quarter, and the fourth one eighth.

The specific gravities of the solutions were :—

Common salt. sp. grav.
100 grams dissolved in 500 cub. centims. water. . 11222
50 a Ss - 1:0631

25 Ps “2 10220
12:5 M a a 1:0122

The mud deposited in the order of their specific gravities;
it precipitated most quickly from the strongest solution. It thus
appears, that, although probably the specific gravity of the solution
has some influence on the rate of settling, the real cause is most
probably the absorption of heat; for the more salt passes into solu-
tion, the more heat is absorbed. This explanation is confirmed by
the deportment of fine mud in water of different temperatures. It
is well known that precipitates will settle more rapidly if the liquid in
which they are suspended be boiled. This phenomenon is, I believe,
usually attributed to the coagulation of the precipitate; and in some
cases, for instance gelatinous, precipitates, this undoubtedly happens ;
but in the case of finely divided precipitates I am disposed to attri-
bute the settling to the absorption of heat by the water.

Experiments were made with a view to decide this point. The
water, however, after being heated up to the required temperature,
must be kept at that temperature by means of a heat equal on all
sides, while the mud is depositing. This was done by means of
water-baths, in which the beakers containing the water and mud
were placed. In every case it was observed that the hotter the
water the more rapidly the mud settled.

Perhaps I may be allowed to attempt a possible explanation of

132 Ww. BAMSAY ON THE PRECIPITATION

these phenomena. It is generally agreed that the molecules of a
fluid acquire, when heat is applied, a greater amplitude of vibra-
tion, but that the duration of each vibration remains constant.
To take the simplest case, suppose a particle a vibrating in
a horizontal plane, and that it performs each vibration in equal
times, whether the amplitude of the vibration remain constant or
not. Another particle } is descending at right angles to the plane
of vibration of a. If the amplitude of the vibration of a be doubled,
the particle 6 will have twice as great a chance of avoiding collision
with aas before; for the particle a will traverse a given space in
half the time it formerly took, leaving it vacant the other half.
The particle 6 will therefore be at liberty to fall through this space
for twice as long a time as was at its disposal before the amplitude
of vibration was doubled.

Of course the problem is infinitely more complicated than the
very rough and ready instance I have given; but I think it is
simply an extension of the phenomenon, and can therefore be ac-
counted for on the same principle.

Note by Prof. A. C. Ramsay, F.R.S., F.G.S.

Though the author does not draw any geological inferences from
the foregoing data, it is yet obvious that they have geological
bearings. Thus, given equal amounts of mud in suspension in a
salt lake or in the sea, and in a freshwater lake, the water of each
being still, a greater quantity of sediment in a given time will be
precipitated to the bottom in the salt water than in the fresh water,
and the amount of salt in solution will relatively retard or hasten
precipitation. But supposing mud in suspension to be carried into
the sea by a river, the mixture of fresh and salt water, the current
at its mouth, and the movements of the sea beyond that current,
due to winds, tides, and ocean currents, will also affect the question
to such an extent that under these circumstances no absolute rate
of precipitation can be estimated in a given time, over a given area,
for a given quantity of mud. Still salt water, as compared with
fresh water, must under like circumstances have an effect—as, for
example, in the case of the mud carried by rivers into the great
lakes of America and into Hudson’s Bay, or the mud carried into
the salt lakes of Central Asia, and that carried into the large lakes
of Switzerland and the north of Italy. Examples of fresh and salt
lakes occurred in old geological times; take the case of the Swiss
Miocene deposits for the first, and of the British Permian and
Triassic (and some other areas) for the second; and the rates of
deposition of mud in such instances may possibly have been modi-
fied by the freshness or saltness of the water.

Discusston.

Prof. Hveues inquired whether the observations had been suf-
ficiently carefully made to determine the amount ef precipitation in

OF CLAY SUSPENDED IN WATER. 133

a given time. If particles put into heated water were not of the
same temperature as the water, the gases would be driven off from
them, and their place would be taken, on cooling, by water, which
would hasten the deposition. He asked whether, in the case of the
solutions, the difference of the specific gravity of the fluids might not
somewhat similarly affect the air entangled in the particles.

Prof. Masxrtynz said that he thought this paper was hardly suited
to the Geological Society, the question discussed in it being a purely
physical one. Absolutely pure water is viscous; and judging by
the influence of salts in solution in aiding precipitation and filtration
in the laboratory, it would seem that this viscosity is impaired by
salts dissolved in it. The results obtained might therefore be due to
the different viscosities of the solutions employed.

Prof Ramsay remarked that at any rate the fact remained that
the mud went down faster in salt water than in fresh; and that
fact seemed to him to have important geological bearings.

Mr. Evans expressed a hope that these experiments might be
carried still further, as, although they were undoubtedly of a phy-
sical nature, their results appeared to be of considerable importance
from a geological point of view. He suggested that experiments
might be made with solutions which, having been already saturated
with one salt, would yet dissolve another without undergoing any
change of volume.

Mr. D. Forsus stated that the author’s results seemed to him to
be deserving of great confidence; his experiments had evidently
been conducted with all due care.

Prof. Morris remarked that the subject was not a new one.
Mr. D. Robertson studied it in 1873, and found that mud is less
rapidly precipitated in weak than in strong saline solutions. The spe-
cific gravity of the matter in suspension must also be taken into
consideration in such experiments.

The Prestipent (Prof. Duncan) did not feel quite sure that the results
embodied in this paper would be found to have much practical bearing
on geological questions. Very fine sediments take along time to fall,
and would probably be carried far out to sea. From his own
observations he did not think that the greatest quantity of mud was
precipitated at the mouths of rivers.

(ot J, E. MARB ON THE FOSSILIFERCUS

17. Fossiliferous CAMBRIAN SHALES near Cuernarvon. By J. EH. Marr,
Esq., St. John’s College, Cambridge. (Read March 8, 1876.)

(Communicated by Prof. T. Mc K. Hughes, F.G.S8.)

Tue shales under consideration extend from about three miles 8. W.
of Caernarvon to Bangor, being roughly parallel to the Menai Straits.
They are faulted against Lower Cambrian to the east, and disappear
against a dyke on the west. They are, as a rule, greyish black to
bluish black in colour, most commonly tolerably sandy and mi-
caceous ; but in places, as where the fossils are found, they are chiefly
clayey with hardly any admixture of sand.

The fossils were obtained from three places, all within a mile of
Caernarvon and on the banks of the Seiont.

The first place is close to Pont Seiont, on the E. side of the river,
where an arrow marks the dip inthe Ordnance map. The shales
here are bluish black, dipping at about 45° S.E., containing a large
amount of iron oxide, from the decomposition of iron pyrites, which
is itself found in nests and veins scattered through the mass of shales,
but especially towards the top of the section; and the outer surfaces
of the rock weather to a rusty or olive-brown hue.

The rock is of a splintery nature; and it is rather difficult to get
large pieces. It has a concretionary structure all through, breaking
up into lenticular pieces ; and itis in the interior of these pieces that
most of the fossils occur.

Most of the fossils came from the bottom of the section, and were
scattered somewhat capriciously, so that one might work for two or
three hours without finding a single specimen, and then suddenly
come upon two or three. Graptolites are the most abundant fossils
in this place. About 35 yards 8.W. of the place from which the
fossils were obtained runs a greenstone dyke, parallel to the bedding
of the rock, and altering the shales to the distance of about four
yards from the edge of the dyke. The shale, where altered, has
a subconchoidal fracture, and a glazed porcellaneous appearance
with slickensided and iridescent joint-surfaces. One Graptolite oc-
curred here at the outer edge of the altered rock ; but unfortunately
it fell to pieces. On the other side of the dyke no rock is seen ex-
posed ; but about 50 yards further on another dyke is just visible, in
weathered bosses, at the side of the road.

The second place where the fossils were found is on the 8.W. side
of the river, and a few hundred yards nearer to the mouth of the
river than the preceding place. Here aseries of sections is exhibited
along the old tramway from Caernarvon to Wantlle. The section
where the fossils were found is the third from the tramway bridge
over the river, and about opposite the letter “a” in the “road”
of “ Wantlle railroad.” ‘The shale at this place is very similar to
that just described, but breaks up into larger masses. As the con-
cretionary structure is not so strongly developed, Graptolites were

CAMBRIAN SHALES NEAR CABRNARVON. 135

not so numerous here as in the first place ; but Phyllopod crustaceans
appeared to predominate.

The last place is on the same side of the river as the first, but
about three quarters of a mile higher up, close to the mill below
Peblig Bridge, but on the opposite side of the river; and the only
two specimens found were got out of the material taken from a pit
which was said to be sunk for the purpose of getting coal!

The rock here has quite a different texture from that of the formerly
described rocks, being finely laminated and without the jointed and
concretionary structure of the other rocks. It has a fine texture
and greasy tecl. In it occur veins of quartz &c., which show most
beautiful instances of slickensides. ‘The pieces in which the fossils
occur were got from a depth of about 60 feet from the surface.
Fossils seem to be scarcer here than at the other places. I only found
two pieces, one having on it a tube, and the other a tube similar to
the former on one side, and a Graptolite on the other.

On the whole the first place has afforded a much greater variety
than the other two; but as it was worked for a very much longer
time, it is doubtful whether, with thorough search, the second would
not yield a much more abundant fauna than it has hitherto done.
The third place seems to be much less fossiliferous than the other
two, the more so as, although I tried a fourth place about a quarter
of a mile higher up the river than the third, and on the same bank,
and although the shale was of exactly the same character as that of
the third place, it did not yield a sign of any organism whatever.

The whole deposit is the more interesting as the only remains
hitherto found are a single specimen of Bellerophon perturbatus
below Penrhyn Park, near Bangor, some Graptolites and an obscure
fragment of a Crustacean near Caernarvon, at the place first described
in this paper. Cf. Prof. Ramsay’s Memoir on the Geology of North
Wales (Mem. Geol. Survey, vol. iii. p. 161).

The fossils seem to indicate that the deposit belongs to the upper
part of the Arenig group.

Aprprnpix. By Henry Hicks, Esq., F.G.S.

In the interesting serics of fossils collected by Mr. Marr, and which
have been placed in my hands for identification by Professor Hughes,
I have been able to recognize the following forms :—a new Caryocaris,
anew Aiglina; Trinucleus, sp.; Barrandea, sp.; Lingula, sp.; Dis-
cina, sp.; Obolella, sp.'; Orthoceras caereesiense (Hicks), Didymograptus
bifidus (Hall), D. indentus (Hall), D. Murchisoni (Beck) and var.
furcillatus (Lapw.). The specimens hitherto discovered are for the
most part fragmentary; and hence identification of some of the species
is difficult. I have therefore preferred in such cases to note the
occurrence of the genus only. Others are undoubtedly new species
or varieties, to which I propose to give names, though I do not think
it would be advisable to figure or describe them fully until better spe-
cimensarefound. Fortunately the species recognized enableus tomake

136 _ HH. HICKS ON THE FOSSILIFEROUS

out without any doubt the position of the beds in the series, and to
correlate them with those in other districts.

As described in Mr. Marr’s paper, the rocks have been faulted
down against very much older beds, and the stratigraphical evidence
of their position is wanting; we have therefore to rely almost
entirely upon the fossils and some slight assistance which we
are able to derive from lithological characters. The whole fauna,
speaking generally, has an Arenig look about it; and even were no
species capable of being clearly made out, I should have no hesitation
in placing the beds in that group. It is, however, far more satisfac-
tory to be able to point out, from evidence derived from other districts
where the succession is perfectly clear, what the real position of the
beds must be as proved by the contained fossils. At St. Davids the
Graptolites D. bifidus and D. indentus are found only in the upper
Arenig rocks, and D. furcillatus and D. Murchisoni in the Lower
Llandeilo. According to Messrs. Hopkinson & Lapworth, this is
also the position of these fossils in the neighbourhood of Shelve in
Shropshire, and in Cumberland.

Another of the fossils, Orthoceras caereesiense, is also found in the
Upper Arenig rocks at St. Davids; and the genus Caryocaris is only
known to occur in the Skiddaw slates of Cumberland, associated
with the same Graptolites as are found here. I think we are justi-
fied therefore in correlating these beds with the Upper Arenig
rocks elsewhere.

In some respects the fauna would indicate a position intermediate
between the Upper Arenig and Lower Llandeilo, a passage from the
one into the other. However, on grounds mentioned in a former
paper to the Society, viz. that no beds so low are anywhere exposed
in the neighbourhood of Llandeilo, and that these were distinctly
recognized in parts of North Wales by Prof. Sedgwick many years
ago, and called the Arenig group by him, I feel it is but right that
they should be included in that group.

The lithological evidence also tends to bear out in a marked
degree the foregoing conclusions. From the fact that it was about
this time that volcanic action commenced in Wales and Cumberland,
the rocks in each of these areas assumed much the same cha-
racter; specimens from St. Davids, Shelve, Cumberland, and near
Caernarvon are all bluish-black micaceous shales, with little or no
cleavage. In no other part perhaps in these early series have
we this character so universally shown; and I attribute the cause
to the admixture of a considerable amount. of fine volcanic material
with the ordinary sediment. Why these were not formed into
slates, as the previous and succeeding beds, may have been from the
presence of too little argillaceous in proportion to the amount of
voleanic material; for they were subjected to the same amount of
pressure as those in which the cleavage is now most marked. These
beds are succeeded in the places mentioned by contemporaneous
tuff beds, the first of which we have evidence in the British area.
There can be no doubt in my opinion that a moderately deep sea
prevailed over these areas at the time that these beds were

CAMBRIAN SHALES NEAR CAERNARVON. 137

deposited, and that the volcanoes were submarine. The presence
of fine sediment over so many parts in Britain, and in other
regions in Western Europe proves also that there was no land of
any extent near.

The immediately preceding and the succeeding sediments also are
for the most part such only as would be formed in a moderately deep
sea and away from shore. The fine argillaceous material, however,
was undoubtedly the result of denudation, though in this case of dis-
tant land. Up to this time the sediments had been gradually
changing from the rougher to the finer materials; and in the next
group, wherever volcanic action was not taking place over these
areas, the rocks heaped up were formed almost entirely by marine
life, showing the presence then of a-clear sea. Exception might
perhaps be taken to this view on the evidence of the great grit bed
in parts of Caernarvonshire, and also of the Stiper Stones in Shrop-
shire. It must not be forgotten, however, that in each of these
places very fine sediments directly overlie and underlie these quart-
zose sandstones, and quite conformably. My own opinion, already
expressed in a former paper*, is that the grit bed and the Stiper
Stones are parts of what was a shifting sandbank at a considerable
distance from land, and in a sea of moderate depth.

The discovery of these fossils by Mr. Marr must be looked upon
as of considerable importance, since it enables us to read the succes-
sion of these early Paleozoic rocks in Caernarvonshire in a much
clearer manner than we have hitherto done, and it also carries the
succession there a stage higher.

The following attempt at a comparison of the series which make
up the Arenig group in Caernarvonshire and in Pembrokeshire
may now be useful.

Pembrokeshire. Caernarvonshire.
( 4, Black shale followed a, Shales of Bangor and
by interbedded tuff. Caernarvon, and shales

under and mixed up
with the lower ash-beds
in the Arenig moun-
Upper Arenig. 4 tains, and to the N.E.
and N.W. of Portma-
doe.
b, Fine slates ........00 . 0, Slates at Ty Obrey, and
under the above near
\ Arenig.
Middle Arenig. Slates and flags ......... Grit bed at Ty Obrey &e.
Lower Arenig. Slates and flags ......... Upper Tremadoc rocks
near Portmadoc, and
black slates under the
grit bednear Arenig &e. T

* Quart. Journ. Geol. Soc. vol. xxxi. page 175.

t In addition to the evidence given in my former paper in regard to these
beds, I have lately received from Mr. Homfray a good specimen of Callograptus,
which was discovered by Mr. Ash in the neighbourhood of Portmadoe, in the
lower part of the Upper Tremadoc rocks, and which appears to- be identical
with the species found at St. Davids.

138 H. HICKS ON THE FCSSILIFEROUS

A very short diagnosis of the two new species may be added,
leaving the fuller descriptions until better specimens have been
found. .

CaryocaRis Marri, 0. sp.

Carapace about 2 inch long, more pointed than rounded ante-
riorly, subtruncate posteriorly ; width at the broadest part about +
inch. Shell thick and horny-looking, probably similar in composition
to the shell of a Trilobite. The surface is convex over about three
- quarters of its length, and is marked with at least three moderately
deep furrows, bent considerably forwards, and which seem to divide
it into lobes. The carapace has the appearance of being much
narrower than in C. Wrighti.

AXciina Hueresit, n. sp.

Apparently not more than 4 inch in length, sometimes much
less. Head smooth, inflated, and convex, longer and wider than
the tail. Glabella-furrows indistinct. Eyes small. Body-rings
six: axis arched, very wide anteriorly, but tapering regularly
backward. ‘Tail triangular and strongly margined, axis raised and
reaching nearly to the hinder margin. This is a very good species,
and cannot well be confounded with any other British species. Its
small size, very strongly marked tapering axis, and large inflated
head, are good characteristics.

DISCUSSION.

Prof. Ramsay was glad to find that the Upper Arenigs mark a
passage, as, indeed, he had previously thought. He had never seen
voleanic materials in that part of Caernarvonshire ; but intrusive
porphyries occur there. It seemed to him that the series was in-
complete ; there is no trace of the Tremadoc rocks and Lingula-
flags near the Menai Straits, the Arenig beds lying unfaulted upon
Cambrian in Caernarvonshire and quite unconformably in Angle-
sey. Certain ferruginous beds of the Arenig series which occur
near the Menai Straits are found further south lying on Cambrian
with no Tremadoc beds or Lingula-flags, and so also in Anglesey.

Mr. Jupp suggested that, in appealing to the case of the Scottish
rocks as supporting his views concerning the unconformity between
the Cambrian and Silurian formations, Professor Ramsay was
reasoning in a circle, seeing that the so-called Cambrian rocks of
Scotland had only been identified as such on the ground of their
unconformable infraposition to the Silurian.

Prof. Huenes remarked that the question seemed to be what is to
be regarded as the base of the Arenig. The North-Wales district
was evidently highly faulted, which gave rise to most of the dif-
ficulties. He objected to certain beds being classed as Cambrian
‘without fossil evidence, and maintained that we have as yet no
right to apply the term Laurentian to any rock on this’ side of the
Atlantic.

CAMBRIAN SHALES NEAR CAERNARVON. 139

Mr. Hicks, in reply, said he could not agree with Prof. Ramsay
that there was a break in the series at this point, either in the
south of Scotland, in Cumberland, in Shropshire, or in Wales. The
faunas and the sediments in each of these places show that very
similar conditions prevailed when the rocks were deposited. In
nearly all, the conformity between the beds also is most clearly shown.
These shales near Caernarvon, he thought, were undoubtedly dropped
down against the older beds by two or more faults, and were not
unconformable to the older rocks, as supposed by Prof. Ramsay.
There is no evidence to show that any changes took place over this
limited area at or before the time these beds were deposited which
could have produced such an effect. The sections are acknowledged
to be for the most part hidden by drift; there is no actual uncon-
formity seen, and the whole is assumed on the strength of very
imperfect data. There are no alterations in the sediments such as
would indicate beach-conditions ; and many of the beds are greatly
metamorphosed, and their age in most cases doubtful. It seems,
therefore, far more reasonable to suppose that the ground here is
greatly faulted, and that, as in very many other areas, the absence
of a series at any one or more places is due to some accidental cause,
and not to an unconformity.

140 T. G. BONNEY ON COLUMNAR, FISSILE,

18. On Cotumnar, Fisstze, and SuHEROPDAT, Srructurr. By the
Rey. T. G. Bonney, M.A., F.G.S., Fellow and Tutor of St.
John’s College, Cambridge. (Read February 23, 1876.)

Tue subject of the columnar jointing and spheroidal structure of
rocks has of late received considerable attention, the most im-
portant papers on the subject being those by Professor James
Thomson * and by Mr. R. Mallet. The former explains the colum-
nar structure by a contraction of the mass in cooling, and advances
a novel theory to account for the cup-and-ball structure observed in
cross-jointing , viz.—that the fracture commenced at the centre,
owing to a longitudinal tensile stress, starting often from a ‘small
[included } mass of stone different in texture and in hardness from
the rest” of the rock, and proceeded outwards towards the peri-
phery. The cause of this tensile stress he considers to be, probably,
chemical action set up by infiltration of water, which has produced
an expansion of the outside of the column, so that the outer part,
growing longer, has strained and finally snapped the interior.
The spheroidal structure, often manifested in decaying basalts, he
considered not to be “an original concretionary structure, but due
to decomposition penetrating from without inwards in blocks or
fragments, into which the rock has been fissured.”

Mr. R. Mallet commences his paper by an ingenious mathematical
demonstration of the cause of hexagonal fracture in the contracting
body. He then passes on to consider the cup-and-ball structure.
This he regards as a further product of contraction from loss of heat
in a prism which is now cooling from the sides as well as from an
end, so that the curved surface of the joint is always concave to the
end which is losing heat; and he regards the spheroidal structure
as the result of the residual forces of contraction which yet remain
in the imperfectly cooled prismatic blocks into which the column is
divided.

That columnar structure was due to contraction was clearly
pointed out many years since by Mr. Scrope in his admirable work
on the Auvergnet; so that the confused statements in subsequent
text-books of geology well deserve the severe comments of the above-
named authors. With their explanation I fully concur; and the
demonstration of Mr. R. Mallet seems to me unanswerable. I shall
therefore pass rapidly over this part of the subject, merely calling
attention to oue or two points of interest in connexion with
columnar structure. As, however, I entirely dissent from Professor
J. Thomson’s conclusions in the rest of his paper, and differ to some
extent from those of Mr. Mallet, I venture to think there is yet
room for a few remarks on the subject—one which I have for
some time past lost no opportunity of studying.

* Report of Belfast Naturalists’ Field Club, 1869-70.

t Phil. Mag. ser. 4, vol. 1. pp. 122, 201.
t Volcanoes of Central France, p. 92.

AND SPHEROIDAL STRUCTURE. 141

I shall begin by describing the various structures which may be
observed in rocks, more especially volcanic, and then consider how
far they can be explained by the theory of contraction, as stated by
the above-named authors. The advantage of this method will be
that we shall see more clearly what phenomena have to be explained.
I shall consider the spheroidal structure apart from the others.

It is almost needless to remark that though columnar structure*

appears to be most frequent in basalt, it is not confined to that rock; I
haye myself seen it in trachyte (district of Mont Dor, Auvergne),
pitchstone (Arran), felstone (Cader Idris, &c.), phonolite (Roche Sa-
nadoire &c., Auvergne). Nor is it confined to igneous rock. I have
observed it in volcanic mud (beneath a bed of basalt in Tideswell Dale,
Derbyshire), in coal at contact with basalt from Ayrshire (Geological
Museum of Edinburgh) and from Yorkshire (Woodwardian Museum,
Cambridge), in hematite iron-ore (tb. ; the columns are about $ inch
in diameter), and, though rather imperfect, in palagonite tuff
(Iceland), and in a large quartz vein (Svolvaer, Lofoten Islands).
Ice also, when kept for some time very near a temperature of 32°
Fahrenheit, as it is during a slow thaw or in those singular caverns
termed glaciéres, in the Alps, also exhibits a beautifully regular
columnar structure, which I can only attribute to a contraction of
the mass, probably as it passes from the point of minimum den-
sity to the melting-point f.
. Occasionally one set of parallel divisional planes is more strongly
marked than the others, so that, while the majority of the columns
retain the hexagonal type, an oblong form dominates, and a some-
what platy or bedded aspect is given to the rock-mass. I remember
observing this especially in a trachyte in the ravine of the cascade
of the Dore, on the Pic de Sancy; and it is very conspicuous in the
great mass of felsite which rises above Llyn-y-Gader (Cader Idris).

Columnar structure is rarely set up quite close to the exterior of
the mass; there is usually an interval, varying from a few inches to
several yards ; this is either affected by some other form of jointing
or is quite irregularly cracked..

Where the space intervening between the columnar part and the
exterior is but small, as, for example, in “the Spindle,’ near St.
Andrew’s, Fife, and at Montiquey quarry, near Balmuto in the same
county, it may be concluded that the mass soon began to lose heat
very uniformly. Henee this structure is more likely to occur in
intrusive masses (especially when the lava cooled at some depth like
the above) than in streams of subaerial lava. In the latter also
the columns often approach nearer to the under than to the upper
surface. Thus the lava-stream of Royat is rudely columnar almost
close to its base at the grotto of the Tirtaine, as is the stream from
the eastern side of the Puy de Gravenoir, where it is exposed in a
quarry on the east of the road from Clermont-Ferrand to Beaumont,
while the upper part, still some yards thick, exhibits a far less

* As it may be considered to be now agreed that the sides of the prisms are

normal to the surface of cooling, I shall not multiply instances to prove this.
t See ‘Nature,’ vol. i. p. 481, vol. ii. p. 288.

142 T. G. BONNEY ON COLUMNAR, FISSILE,

regular structure. The same may be observed in the prismatic
lava at Torre del Greco (Naples), and at Fingal’s Cave (Staffa).

The curved forms of the columns have been so fully discussed by
Mr. Scrope and Mr. R. Mallet that I pass them by, merely stating
that while some of the cases of slight flexure may be produced by a
small motion of the lava-stream, after the formation of the columns,
but before it has become perfectly solid, the majority must be duc
to changes in the form of the surface of what we may call breaking-
tension in the cooling mass. It is extremely difficult to explain
such complicated cases as the Clam-shell Cave (Staffa), Pouk Hill
(Walsall), the Roche Sanadoire (Auvergne), and Jaujac (Ardéche) ;
but we must remember that the surface of cooling must be closely
related to the surface of the mass, and that partial denudation has
often deprived us of the data for that part of the problem. Varia-
tion in the conductivity of the superincumbent materials might also
produce some effect. Further complication has been sometimes
introduced by large fissures, which appear to have opened out in the
the mass somewhat anterior to its breaking into columns, and have
allowed of a more rapid escape of heat from their surfaces. Thus
columns will often be found to curve to a large, more or less hori-
zontal joint (see fig. 1).

Fig. 1.—Columns of Trachyte curving to a joint (Pic de Sancy,
near Cascade of the Dore).

Ay

. C
A, B. Great joint. C. Minute cross-joints.

>
AA
LZ

The structures more or less parallel to the surfaces of the mass,
and consequently generally at right angles to the columnar, have
next to be noticed. These may be roughly classified as (a) Fissile
and Platy, (0) Tabular, (c) Curvitabular, (d) Cup-and-ball struc-
ture.

(a) Fissile—A distinct tendency to a cleavage, at first sight
closely resembling that of ordinary slates, is often visible in igneous
rocks. A pitchstone on the north side of the hill of Dunfion (Arran)
splits with tolerable ease in one direction. .The great intrusive
sheet of pitchstone on the Corriegills shore of the same island also
has a distinctly fissile character, especially near the upper and lower
surfaces, to which the structure is parallel. The same structure
may be often observed near the surface of basalt dykes, as on this
same shore; usually, however, it extends only for a few inches into

AND SPHEROIDAL STRUCTURE. 143

the mass. The outer part of the mass of basalt at the Spindle and
Montiquey (mentioned above) exfoliates in rather thin plates; and
the columns are, so to speak, enclosed in a fissile spheroidal shell a
few inches thick.

The phenomenon is very conspicuous in some of the trachytic
rocks of the Auvergne; those in the Cantal and Mont Dore are so
fissile as to be used for roofing-slates. The well-known Roche
Sanadoire and Roche Tuillicre*, the latter of which is extensively
quarried for the above purpose, are very remarkable instances of this
structure, and throw much light upon its origin. I shall therefore
describe them in some little detail. Near the highest part of the
great Auvergne plateau, from which rise the groups of the Puy-de-
Déme volcanoes and the chain of the Pic de Sancy, and at the
northern extremity of the latter, two rocky bluffs rise on either side
of the deep cirque-like head of a wooded glen. That on the right
bank of the valley is the Sanadoire, on the left La Tuilliere. Both
are phonolite; but the character of the rocky, and still more its
structure, are different.

The Roche Sanadoire forms the extremity of a rather short spur
from the Puy de Angle. On the southern side it is not very con-
spicuously columnar; but the rock, which is exposed in several
abrupt crags, is divided into a vast number of small rather irregular
prisms, with numerous cross joints. When, however, a view is
gained of the precipitous western face, the columnar structure
becomes more conspicuous, together with another set of divisional
planes, which give a more fissile character to the rock. A reference

Fig. 2.—Columnar and Fissile structures. Roche Sanadoire.

A, C. Jointed columns. 3B. Fissile bands.

* Mr. R. Mallet (Phil. Mag. ser. 4, vol. 1. p. 218) calls the rock of La
Tuilliére basalt, and the fissile structure a “slaty cleavage” due to subsequent
pressure. The rock is a phonolite, and has not the slightest resemblance to
basalt; and the structure, I trust to show, is not a case of ordinary cleavage.

+ For full account see ‘ Neues Jahrbuch fiir Mineralogie,’ 1872, p. 351.

Q. J. G. S. No. 126. I

144 T. G. BONNEY ON COLUMNAR, FISSILE,

to the annexed diagram (fig. 2), together with the following extract
from my note-book may make this ‘plain: —

‘““The columns are frequently rather rude and quadrangular
in shape; in part, however, they are tolerably regular and hex-
agonal. They are bent in a most singular manner, so that those
toward the outside curve round and roughly overarch an interior nearly
vertical sheaf or boss. These columns are cut by numerous cross joints
which run parallel to great curving divisional planes, and traverse
their axes at a high angle. Thus the columns appear to be built
up of slabs. The cross joints are commonly not at right angles to
the axis, the inclination sometimes not exceeding 50°. In parts
_ this platy structure becomes so conspicuous as almost to obliterate
the prismatic ; and a very singular effect is produced by this alterna-
tion of bands of ‘frilled’ and ‘crimped’ rock. The whole mass,
whether distinctly columnar or not, has more or less of a fissile
structure, which is of course most distinct where the columns are
least so, and is (so far as I could see, and as one would expect)
parallel to the great curving divisional planes. This description
applies especially to the south-western part of the hill; when a
view is obtained of the whole of the western face, as from the slopes
on the opposite sides of the valley, the same complicated double
structure will be seen continued throughout.”

Turning now to the Roche Tuilliére, we find that its eastern face
is divided into a number of large columns ; these are very distinctly
marked, and are cut by a great number of cross joints, which are
not at right angles to the axis of the columns, but dip outwards.
These joints also, when viewed from a distance, are scen not to
traverse the face of the cliff horizontally, but to arch upward from
its two extremities so as to become horizontal in the centre,

Fig. 3.—Diagram of Columns and Cross Fig. 4+.—Fissile structure.
joints (Ioche Tuilhére). (Roche Tuilliére).

AB. Vertical joints. CC. Direction of cross
joints. D. Direction of southern end of cliff.

following very uearly the curve of its sky-line, which has no doubt
been determined by the direction of these divisional planes. On
climbing the steep wooded talus to the base of the columns their
structure becomes more evident. In shape they are rather less

-

AND SPHEROIDAL STRUCTURE, 145

regular than their distant appearance would lead one to infer, as
they vary from irregular quadrangular to hexagonal, the majority
appearing to be pentagonal. They are often as much as 6 feet in
diameter. ‘The great cross joints cut the columns at distances from
2 to 5 feet apart; and the whole mass is traversed by a fissile struc-
ture curiously resembling that of a roughly cleaved slate, parallel
to these joints. As the outward dip of these planes is some 20°, it
produces a most unpleasant idea of instability upon the mind; for
the columns look as if the least disturbance would cause them to
slip forward. At the southern end of the cliff the columns are less
distinct ; and here the curve of the divisional planes makes a much
higher angle with the horizon than elsewhere, becoming nearly
vertical (see figs. 3 & 4).

From the above descriptions it will be evident that the fissile
structure in these rocks cannot be a true cleavage ; its explanation
will be attempted presently.

(b) Tabular Structure.—In many igneous rocks the horizontal
joints are so regularly and strongly defined as to produce a tabular
or bedded structure. Instances of this may be seen in many
granites, as, for example, at Pardenick Point, near the Land’s
End; in dolerite, in an intrusive mass under “ the Castle,” Burntis-
land (Fife); in the gabbro of the Gimlet rock, Pwlhelli, and of the
Cuchullin Hills; and, though less perfectly, in some of the “ traps ”
in the Lake-district. A case which I have more than once observed
seems to suggest some connexion between this and the platy struc-
ture described above. An excellent example of it may be found in a
quarry on the Gross Weilberg (Siebengebirge) (fig. 5). Here a mass of

Fig. 5.—Columns &c. (Gross Weilberg).

A, Trachytic tuff. 3B. Closely jointed rude columns. C. Bent columns.
D. Upper stage of quarry with good columns. FE. Lower stage.
F, Platy structure. G. The same in rude columns.

basalt is intruded into a trachytic tuff. The exterior of the mass at
the top is for the space of 3 or 4 feet divided into rude quadran-
LZ

146 T. G. BONNEY ON COLUMNAR, FISSILE,

gular prisms, which are traversed by so many cross joints as to
make the rock almost platy in structure; then, after a distance of
about a yard, where the normal joints become rather more frequent
and somewhat confused, a series of magnificent columns commences,
which run with great regularity, and are singularly free from cross
joints, the shafts often being unbroken for a distance of several
yards. On approaching a contact with the tuff on the lower side of
the mass a platy structure again sets in.

This tabular structure, it will be observed, is roughly parallel to
the bounding surface of the mass, as is the bedding in the case
noticed under Burntisland Castle. Intrusive dykes also sometimes
show a tabular or roughly platy structure running parallel to their
bounding surfaces.

(c) Curvitabular Structure.—I proceed next to describe a struc-
ture which I have not often noticed myself, and do not remember to
have seen described by others—which, for the sake of a name, I may
denominate curvitabular jointing. As an example of it I will de-
scribe the basalt of the Plateau de Prudelle, above the valley of
Villar (Auvergne), a sketch of which is engraved by Mr. Scrope, in
plate vi. of his work on the Auvergne. This basalt crowns a rib or
promontory of granite, resting on this or on a thin intervening bed of
sandy (and ashy ?) marl containing rolled felspar crystals and a few
bits of scoriaceous basalt. Above the marl comes a rather irregular,
hard, “ropy”-looking, brown band, probably only the baked crust
of the marl, which is united irregularly to the base of the basalt
flow. This is very rough, slaggy and scoriaceous, in places re-
sembling an agglomerate of cinders, sometimes partly fused together,
sometimes loose. This crust is about a foot thick, and passes rather
suddenly into a somewhat slaggy-looking basalt, with a few
elongated vesicles and occasional irregular vertical joints. This
continues for about 2 feet; and then the curvitabular structure sets
in rather suddenly. The mass is traversed by a series of more or less

Fig. 6.—Diagram of Curvitabular structure (Plateau de la Prudelle).

AND SPHEROIDAL STRUCTURE. 147

horizontal curving joints with their convexities upward; so that the
whole is divided into a series of plano-convex, meniscoid, or
concavo-convex blocks, in length perhaps about 2 to 4 feet, and in
thickness 4 to 12 inches (fig. 6). ‘This structure continues with but
slight indications of any tendency to vertical master joints for some
35 feet ; and then the basalt for the remaining space (perhaps about
4 yards) becomes rather suddenly rudely columnar*. How much
has been removed by denudation from the top of the flow it
is impossible to say. At the end of the spur there is evidence
that its surface was uneven and a ridge already defined when
the basalt flowed; for the tabular structure curves in both the
longitudinal and transverse sections, so as to remain symmetrical with
the surface of the ground. It is therefore evident that it is con-
nected with the form of the surface of the rock. Close to the
station of Rougeac (railway from Le Puy to Arvant) I saw a similar
structure in a mass of basalt. A very rough sketch of this in-
teresting section (jotted down from the railway-carriage window)
is given below (fig. 7).

Fig. 7.—Section at Rougeae.

A. Basalt. B. Stratified volcanic ash. C. Agglomeratic volcanic ash.
D. Basalt, with curvitabular structure.

Here also the structure is affected by the form of the bounding
surface; but in this case the convexities of the curves are turned
downward. |

(d) Cup-and-ball Structure.—The next structure to be noticed
is that form of cross-jointing of columnar rock which goes by the
name of cup-and-ball structure. Usually the columns are divided
across, at variable distances, by plane cross joints at right angles to
their axes. These sometimes, as in the Gross Oelberg and Weilberg
(Siebengebirge), and at Murat (Département de Cantal, Auvergne),
are as much as 15 or 20 feet apart ; but commonly they are not more
than a yard, and sometimes less. Occasionally, however, the joint-
surface is not a plane, but more or less curved, so that the convex
extremity of one segment fits into the concave extremity of the next.
The form of this curve varies: sometimes its curvature is but slight,
and it is continued down to the sides of the prism; at others the
curvature is more marked, the outline of the ball being more nearly
circular, and is not continued to the sides of the prism, a flat space

* Mr. Scrope (p. 106) describes this basalt as separated at some points into
very regular prisms of five or six sides, which extoliate by decomposition in
slaty lamine at right angles to their axes. I do not remember to have seen
this structure conspicuous in the part which I examined.

148 T. G. BONNEY ON COLUMNAR, FISSILE,

intervening between these and the curved boundary of the ball,
which, as arule, nowhere touches them.

These curves generally point in one direction, sometimes upward,
sometimes downward; occasionally, however, cases may be noticed
where they are differently directed in adjoining columns*. So far —
as I can remember, they are most frequent where the rock exhibits
the spheroidal structure which I am about to describe; but before
doing this I should wish to call attention to a curious form of cross-
jointing which I observed in some of the basalt columns at the upper
part of the hill of Bonnevie, Murat (fig. 8).

Here, in the upper part of the excavation opened for quarrying,
a curious interruption to the regularity of the columns may be
observed in the form of a sort of projection, like a part of a capital
from which springs the base (curving till the old direction is

Fig. 8.— Cross joints at Fig. 9.—Structure of Columns
Bonnevie. in the Lava at Jioyat.

A. Plane of parallel joints.

resumed) of another column. A closer examination shows that this
structure is connected with a series of cross joints, which cut the
columns (here at angles of about 60°), and to whose plane the axes
of the imperfect prisms forming the brackets are parallel. These
joints are from 4-6’ apart; and the ordinary cross-jointing in the
other parts of the hill seems rather regular,

A singular case of columnar and joint-structure occurs in the
valley of the Tirtaine, near the grotto of the springs between Old
and New Royat (Auvergne). Here the lava-stream, which is
generally irregularly jointed, and sometimes (according to Mr.
Scrope) spheroidal, is rudely columnar at its base (fig. 9). he
columns are irregularly hexagonal, and about 18" in diameter. At
first sight they seem cross-jointed at intervals of about 12” or 15";

_ * As is the case (pointed out: by Mr. Serope, Geol. Mag. Dec. 11, vol. ii. p. 412)
in the columns in possession of our Society. .

AND SPHEROIDAL STRUCTURE, 149

but on closer examination it becomes evident that this apparent
division is caused by a change in the direction of the faces of the
column, so that for a very short space, at this point, they are no
longer vertical, but bend very slightly in or out. Nearly all these
apparent bands curve a little upwards. Sometimes they correspond
with actual cross joints; at others there may be very fine con-
cealed cross joints; but in many there is not the slightest sign
of any division at all. Still these bands indicate a tendency to
division ; for at a short distance the columns become less regular, and
the real cross joints (still curving upwards) more numerous, so that
the lava is first divided into irregular rectangular blocks, with
curved upper and under surfaces, and then broken by irregular
curving joints, which in a few cases approach the curvitabular
structure already described ; so that in one part the columnar, in
another the curvitabular structure is dominant.

To proceed to the spheroidal structure. Professor J. Thomson
regards it as the result of a process of exfoliation due to the action
of the weather on a tolerably regular-shaped homogeneous mass.
Though undoubtedly cuboidal blocks of rock have, for obvious
reasons, a tendency to weather into rough spheroids, and though
further action of the weather might occasionally produce concentric
exfoliation in such spheroids, yet I trust to show that this explana-
tion is wholly inadequate in the present case, although doubtless
the weather has great effect in developing the structure. The ex-
amples which I am about to mention will, I think, establish these
two propositions :—

(1) Spheroidal structure is to be seen in rocks which are not
homogeneous and are not at all cuboidal in form.

(2) Spheroids may be found in columnar rock which have evi-
dently formed inside a prism, the exterior of which was not broken
by joints.

Spheroidal structure has been observed in plaster on a wall*.
A very fine example of it in bedded shale is figured by Mr. Jukes in
his ‘ Manual of Geology’+. I have seen it well developed (of an
ellipsoidal form) in a lenticular fragment of shale caught up in
basalt on the Fifeshire coast, near Elie. A very fine instance of it
may be seen in volcanic ash near the village of Santa Lucia (Valle de
Cordevole, Italian Tyrol). Here it is so conspicuous that the rock
at a short distance might be readily mistaken for a decomposing
basalt. Instances of it can also be found in the agglomeratic ash
of the Binns, Burntisland (Fife). The annexed diagram (fig. 10)
will show that here the structure is wholly independent of the form
of the rock. Sometimes also it occurs where a distinct stratification
may be observed, through the planes of which it cuts.

Again, perlitict obsidian and pitchstone is a true case of sphe-

* Geol. Mag. vol. viii. p. 333. + Manual of Geology, p. 311 (ed. Geikie).
t I distinguish, of course, perlitic from sphzrulitic structure. The best spe-
cimens which have come under my own notice have been from Hungary: for
the opportunity of examining a very interesting example I am indebted to Mr.

J udd’s kindness.

150 T. G. BONNEY ON COLUMNAR, FISSILF,

roidal structure, though on a minute scale; and here examination
of the rocks, both macroscopic and microscopic, proves that the
structure is in no way connected with previously formed cubes, or
capable of being explained by simple exfoliation during decompo-
sition. But the following instances of its occurrence will, I think,
show that no theory of mere decomposition is adequate to explain
the facts.

The annexed diagram (fig. 11) is taken from a large spheroid ex-

Fig. 10.—Spheroidal structure in Fig. 11.—Spheroid in Basalt
Agglomerate (Burntisland). near Murat.

hibited in a mass of basalt by the roadside on the way from Murat
to the bridge over the Alagnon (Auvergne). A glance at it will,
I think, show that the curious prolongation of the outer layer is
impossible to explain on any principle of weathering. I may add
that this occurrence of an isolated spheroid or small group of sphe-
roids in a mass which generally does not show the structure is by
no means rare.

But I can produce yet stronger cases. A few kilometres from Le
Puy on the Route de Brioude, close to where a road turns off to
Polignac, is a mass of columnar basalt rather decomposed, part of
which exhibits very well the spheroidal structure. Here spheroids
may be seen, one above another, enclosed three or four at a time in
a columnar shell without any dividing cross joints, so that they are
just like Dutch cheeses packed in hexagonal cases (the interstices
being filled up). The lid of the box has more or less fallen away,
and exposed the contained spheroids (See fig. 12).

Again, a mass of trachyte, intrusive in trachytic ash, by the road-
side on the flank of the Puy Gros (near the Roche Sanadoire men-
tioned above) showed structures bearing on this as well as other
points of interest. The surface in contact with underlying tuff
exhibited a_platy structure for about three inches; then came a
layer rudely spheroidal; above that were some irregularly vertical
joints ; two of these curved in opposite directions so as to enclose a

AND SPHEROIDAL STRUCTURE. poh

lenticular space. In this, two internal spheroids were exposed by
the exfoliation of the face, which were not connected with any cross
joints. In the adjoining mass four large, rude and very irregular
spheroids were shown (the bounding surfaces being incomplete or
with more than one centre). Similar instances of spheroids discon-
nected from the main joints may be seen in a quarry called Turner’s
Pit in the Rowley-Regis basalt, where also, in some cases, the nuclei
of the spheroids are further subdivided, so that imperfect spheroids
are enclosed by a spheroidal shell, like the twin kernels of a nut (fig.
13). It is therefore clear that, though the spheroids often do corre-

Fig. 12.—Spheroids in an un- Fig. 18.— Complicated spheroidal
jointed column near Le Puy. structure (Rowley-Regis basalt).

spond with the spaces between cross joints in the column, there is no
necessary connexion ; both may be due to a somewhat similar cause;
but the spheroid can exist without the cross joint, and vice versd.
Is it, then, possible to find any one cause which will explain these
various divisional structures from the fissile to the spheroidal? I
think that which has already been suggested (I may say proved)
for some of them, viz. the contraction of a cooling mass, is capable
of explaining all. Mr. Mallet has shown that if a mass of molten
rock be cooling uniformly from a surface, it will, in consequence of
the mathematical principle of least action, break into hexagonal
prisms at right angles to the surface of cooling. By the same
principle, if a cube were contracting in consequence of a uniform
loss of heat from each of its sides, it would be more likely to rupture
internally (supposing that the solidification of the exterior prevented
diminution of volume) in spheroidal shells; for the more rapid loss
of heat from the angles would tend to bring the isothermal surfaces
within into a rudely spherical form; and then, when the strain

152 T. G@. BONNEY ON COLUMNAR, FISSILE,

caused shells to be formed, they would be spherical, because the
sphere is, for an equal volume, the figure of least area, and there-
fore of least resistance ; and as its surface is at every point at right
angles to the radius, there is no tangential component to the central
force, and thus the whole of it is effective in rupturing. Thus a
hexagon is the figure which will result from uniform contraction in
two dimensions, a sphere from contraction in three dimensions.

But now, supposing that the contraction is mainly in one dimen-
sion, or, to put it otherwise, suppose that all the points lying in one
surface in a body are in a state of uniform strain in one direction,
naturally there would be a tendency to rupture along a surface at
right angles to the strain. Suppose, for example, a number of tiles
placed on a floor were subjected to strains perpendicular to the floor,
they would naturally split parallel to it. . Something analogous to
this happens in the cooling of an igneous mass, where heat is lost
from a surface (suppose the upper). The strains in a horizontal
direction, due to contraction, are at once eased by the formation of
joints, more or less regular ; but if the loss of heat from the surface
be rather rapid, there will be a strong normal strain, which will not
be relieved thus, and so slabs or tabule will be broken off by a kind
of exfoliation ; and the more the adjacent particles in a straight line
normal to the cooling surface differ in temperature, as will be the
case in rapid cooling, the more frequent will these cross joints be.
Thus the mass near the surface is generally platy or tabular.

The matter may be expressed, perhaps, rather more simply in
another way. Suppose a body contracting uniformly towards a
point within it, and its particles incapable of differential motion ;
then if rupture takes place, a series of spherical shells, concentric
with this point, will be formed. Suppose now (the law remaining
the same) this point be in the surface of the body; then it will
break in concentric hemispherical shells. Suppose the point towards
which contraction takes place be outside, still the body will break
into segments of large concentric spheres whose curvature will
become less and less as the point becomes more remote, the limit
being, of course, a plane when the distance of the point is infinite.
Read, for a force causing contraction to a point, loss of heat from a
surface causing contraction, and the case remains the same. When
heat was lost with tolerable uniformity throughout any part of the
mass, spheroids both large and small would be formed ; when it was
lost from a more or less plane surface, but from certain points on it
more than others (which is equivalent to what would happen when
cooling had advanced a little distance within a lava-stream, owing
to either superficial irregularities or surface-fissures), curved cross
joints and curvitabular joints would be formed; and when cooling
was taking place uniformly from all the points of a plane surface,
then platy or tabular forms would result. It must also be remem-
bered that the form of the exterior surface would greatly modify
these results; for, speaking generally, all points of equal tension
would lie in surfaces parallel to the exterior one, whatever it might
be. Still, the principle of least action would cause a certain sym-

AND SPHEROIDAL STRUCTURE. 153

metry in the surface of rupture; and thus surfaces of revolution
(such as spheroids, hyperboloids, &c.) would be generally produced.

There is thus a certain relation between the spheroids, the curved
horizontal joints, and the great curving cross joints, such as are seen
at the Roche Tuilliére. Hence the fissile structure so well exhibited
both here and at the Roche Sanadoire is not a cleavage structure in
the ordinary * sense of the word; for it is the result of force acting
in exactly the opposite direction, being the result of tension ; whilst
cleavage is due to compression—somewhat similar results being
thus produced by forces with changed algebraic signs.

I consider the cup-and-ball joint structure a special case of the
spheroidal; and the only point on which I feel some doubt is whether
one should regard the division as beginning at the centre or at the
exterior, or as to some extent independent,—namely, whether as a
column cools tolerably uniformly, it tends by symmetry to divide into
approximately equal lengths, and so spheroids are formed in the more
plastic though warmer parts (the exterior shell being a little more
rapidly chilled), and whether the cracks thus formed between two
adjacent spheroids are continued toward the exterior, which then,
being in a state of strain, is cracked horizontally at these points of
weakness ; or whether the exterior, cracking first, caused a line of
weakness, which determined the commencement of a spheroid in the
inner parts; or, as a third possibility, whether the two divisional
surfaces are to some extent independent—the spheroids forming
within in the more plastic part of the column, the cracks opening
from without to the more solidified part, and the two surfaces of
division running together at last so as to complete the separation.

Some independence in these surfaces of division seems to be indi-
cated by the occurrence of distinctly formed spheroids in an unbroken
column, and by the fact that the spheroids are occasionally again
subdivided into segments which to some extent continue the same
structure, two or even more of these being enclosed in a more
regular spheroidal shell (see fig. 13). This independence, too,
would explain the fact that the ends of columns pointing in the same
direction show sometimes cups and sometimes balls. It is quite true
that the divisional curves, if due to strain from contraction, should
be concave to the surface of cooling, as Mr. Mallet has proved (and
this, I have little doubt, is the case in the curvitabular structure) ;
but, as Mr. Scrope objects, and as my own experience has shown
me, in the case of, cup-and-ball structure in columns there is great
uncertainty, adjacent columns showing at top, one a cup, the othera
ball. If, however, the two fissures were formed to some extent inde-
pendently (the curved one beginning at the interior, the plane one
at the exterior), this would be likely to happen, though still the one
or the other structure might predominate : the outside cracks would

* That it cannot bea true cleavage is shown (1) by the extreme improbability
of this being produced in rocks which are not likely to have been subjected to
great earth-movements and must have always been of rather a superficial cha-
racter, and (2), in the cases quoted, by the impossibility of explaining the ar-
rangement of the divisional planes on any theory of cleavage.

154 ON COLUMNAR, FISSILE, AND SPHEROIDAL STRUCTURE.

generally form at the points where the spheroid’s surface was nearest
to the outside of the column, because the shell would be rather
weaker there.

The independence of the structures is to some extent confirmed by
the fact that a thin plane lamina of rock sometimes exists between
two spheroids in a column.

All the divisional structures, therefore, which can be observed in
igneous rocks are, if the above considerations be admitted, to be re-
ferred to one and the same cause, viz. contraction of the mass while
cooling. It is not, of course, denied that in some cases effects may
have been produced by subsequent chemical or mechanical action (as
the true cleavage in certain of the Welsh felstones) ; but it is con-
tended that those above described are all the result of one and the
same cause.

Discussion.

Prof. Ramsay inquired whether Mr. Bonney had observed the cup-
and-ball structure in narrow dykes at right angles to the planes of
cooling.

Mr. Kocn stated that he had tried many experiments bearing on
the subject of prismatic and spheroidal structure. Slag at 1500°C.,
when suffered to run into water, has a perfectly smooth surface, and
exhibits distinct prismatic structure. Fragments of ironstone placed
on the sole of a furnace cracked off in shells, leaving spheres; and
quartzite under similar circumstances flaked off in the same way.

Mr. Jupp noticed the interesting bearing of the numerous facts
described in this paper on the theories concerning the production of
columnar structures propounded by Prof. James Thomson and Mr.
Robert Mallet. He, however, deprecated the introduction by the
author of the question of the fissile structure of phonolites (which
microscopic examination proved to be the result of peculiarities in
the arrangement of the crystalline constituents of the rock) as
haying any relation to the structures resulting from contraction.

Mr. Rurtey stated that one of the felstone dykes in the Lake-
district showed spurious cleavage in the direction of the dyke.

Mr. Mureny remarked that as the cooling commenced at the
surface, the percolation of water between the first formed columns
would aid in the production of the columnar structure lower down.

The AurHor, in reply to Mr. Judd, said that he did not regret
having introduced the Auvergne phonolites into his paper, for he was
convinced that their structure could not be explained by pressure ;
he considered that} the intricacy of the arrangement shown by his
diagrams negatived the supposition. Separation of the minerals, as
described by Mr. Judd, would not necessarily be fatal to his theory ;
but in the case of the Roche Sanadoire, he had examined a section
to see if it existed, and had not found it.

‘* GREENSTONES ”’ OF WESTERN CORNWALL. 155

19. On the so-called ‘‘ GREeENstonEs” of WESTERN CoRNWALL.
By J. Anraur Puiiiies, Ese., F.G.S. &e. (Read February 23, 1876.)

In a paper which I had the honour of reading before this Society in
March 1874, 1 described and gave analyses of two varieties of
Cornish Greenstone, both of which were characterized as being pro-
bably altered hornblendic slates. In the present communication it
is my intention to give the results of a chemical and microscopical
€xamination of some of the more important of the so-called Green-
stones of West Cornwall, reserving for a future occasion an account
of those found east of the parish of St. Stephens.

Penzance District.—The most extensive group of these rocks is
probably that situated on the margin of Mount’s Bay, in the vicinity
of Penzance, and extending, with comparatively little interrup-
tion, from Penlee Point, about two and a half miles south-west of
that town, to a mile east of the village of Gulval. From the latter
locality to Marazion there does not appear to be any extensive ex-
posure of rocks of this class, but they again make their appearance
in considerable masses in the sea-cliff extending eastward from
Marazion to Cuddan Point.

This district may be regarded as being to a certain extent classic
ground, since the rocks which now occupy our attention have fre-
quently been the subject of the investigations and speculations of
the earlier geologists.

In a paper read by Prof. Sedgwick before the Cambridge Philoso-
phical Society, March 20th, 1820, he says :—*“ In the cliffs between
Penzance and Mousehole, we found the ordinary varieties of clay-
slate, much intersected by contemporaneous veins of quartz. These
rocks were succeeded by, and alternated with, beds of compact fel-
spar, greenstone, hornblende rock, and the purple schist before
mentioned, all resting conformably, and, on the great scale, exhibit-
ing a slaty texture, although many of the hand specimens derived
from them broke into irregular fragments ’’*.

Dr. J. Forbes, who wrote about the year 1822, remarks, with
regard to what he calls the “slate formation” of Cornwall :—
‘‘This formation is much more complicated than the last (the
granite), and affords much greater scope for geological research.
It comprehends, as far as I have been able to ascertain, five distinct
rocks. These are clay-slate, hornblende rock, greenstone, compact
felspar, and slaty felspar. The three first of these are so well
known, have such distinctive characters, and are of such general oc-
currence, that we are not very likely to be mistaken in discriminating
and giving them their proper appellation. Ofthe correctness of appli-
cation of the two last terms I am less certain. From the peculiarity

* “On the Physical Structure of those Formations which are immediately
associated with the Primitive Ridge of Devonshire and Cornwall,” by the Rey.
Adam Sedgwick, M.A., M.G.S. (Trans. Cambridge Phil. Soc. vol. i. p. 134).

156 J. A. PHILLIPS ON THE SO-CALLED

of their characters, and their distinct nature, however, I am sure
that, whether rightly named or not, they are as well entitled to
appropriate appellations as any rocks can be.

‘‘ By felspar rock I mean a rock of small granular structure, con-
sisting, apparently, principally or almost wholly of felspar. By
slaty felspar I mean a rock apparently of the same composition, or
only with the addition of a very small portion of mica, with a dis-
tinct slaty fracture.

“These five rocks, constituting the assemblage to which I have
given the name of the slate formation, occur in beds of various
magnitude, alternating with each other; but, with one very small
exception, I have uniformly found the slaty felspar rock in imme-
diate contact with the granite ; and I think it not improbable that,
in proportion as we recede from this central rock, we shall find the
slaty felspar become less frequent, and be finally superseded by
some of the varieties of clay-slate”*. |

Dr. Forbes subsequently goes on to say that, in tracing the shore-
line from Mousehole to Newlyn, nineteen different beds may be
distinguished ; and he remarks that, in advancing from Newlyn to
Penzance, the rocks consist of a series of hornblendic beds of a more
or less slaty character, together with slaty rocks intermediate be-
tween felsite and hornblende slate. He describes the rocks imme-
diately south of Penzance pier as being precisely similar to the
others, excepting that the hornblendic varieties have more decidedly
the character of greenstone, and that the slaty rocks approximate
more closely to clay-slates. Of this locality he further remarks, that
several of the beds are decided and well-characterized clay-slates.

The rocks extending from the pier to Chyandour are stated to be
similar to those between Newlyn and Penzance; but their dip,
direction, and succession are described as being somewhat less dis-
tinct. The same author calls attention to a greater irregularity in
the direction and inclination of the strata between Marazion and
Cuddan Point.

With the view of ascertaining the correct names of some of the prin-
cipal Cornish rocks, Mr. John Hawkins forwarded, in 1793, a care-
fully selected collection of specimens to Prof. Werner, and during the
autumn of the same year received the decision of that then eminent
petrologist. Mr. Hawkins tells us, “‘no doubt was expressed of the
nature of our metalliferous rock, the killas.” It was pronounced to
be a genuine T'honschiefer, or argillaceous slate, in no respect differ-
ing from that which oceurs in Saxony..... . Those taken from the
neighbourhood of Penzance were, after a rigid scrutiny, pronounced
‘to be hornblende-rock, intimately mixed with the constituent mass
of argillaceous slate, and in part genuine hornblende-slate” f.

Writing in 1827, Messrs. Von Oeynhausen and Von Dechen
remark, “the Killas (of Cornwall) is, at its contact with the

* “On the Geology of the Land’s End District,’ by John Forbes, M.D.
(Trans. Royal Geol. Society of Cornwall, vol. ii. (1822) p. 251).

+ ‘On the Nomenclature of the Cornish Rocks,’ by John Hawkins, Esq.,
F.R.S. (Trans. Royal Geol. Society of Cornwall, vol. i. (1822) p. 155).

‘ GREENSTONES ”’ OF WESTERN CORNWALL. 137

granite, rather hornblende-slate and greenstone than clay-slate ; in
the Lizard district we have seen the greenstone intimately mixed
with granite, both occurring in the same vein. ‘The transition from
clay-slate into hornblende-slate and greenstone is commonly so
gradual that we have not been able to trace anywhere a line of
junction between both rocks ’’*.

Dr. Boase, in his ‘Geology of Cornwall,’ observes, that “the
compact and schistose greenstones repeatedly pass into each other, so
that in traversing them from the granite they appear to alternate ;
but this does not hold good to any extent in the length of their beds,
as may be seen on the shores of Mount’s Bay, where the massive
rocks are insulated in the body of the slate” 7.

The same author, in his ‘ Primary Geology,’ makes the following
observations :—‘‘ The most abundant species of actynolite rock is
hard and compact, known provincially by the names of blue-stone or
blue elvan. ‘This variety is accompanied by, and gradually passes into
blue slate, which is fissile in various degrees, and may also be easily
separated into small rhomboidal pieces. The massive kind occurs
in elevated ridges on the coast, and is very durable, whilst the slate
is decayed to a considerable depth, still retaining its form, but pro-
gressively diminishing in tenacity from the perfect rock to the sur-
face” +. He further states that each of the schistose rocks of this
district, but particularly the ‘“‘ greenstone,” passes into thick lamellar
slates, which exhibit various shades of blue and green.

These rocks, as well as all the others forming the subject of this
investigation, are laid down on the Map of the Geological Survey as
‘“‘sreenstones;” and this classification appears to have been gene-
rally accepted. In 1867, however, Mr. David Forbes, F.R.S.,
writing in the ‘ Geological Magazine,’ makes the following observa-
tions :—‘“ The writer of these remarks, finding, from an examination
of the sheets of the Geological Survey, that large masses of green-
stone were represented as occurring in Cornwall, near Penzance,
and at the Botallack Mines, immediately imagined that he would
there find the same relations of this greenstone to the metallic
lodes occurring as he had found to be the case in South and North
America, Spain, Norway, Sweden, &c., and made a journey expressly
for this examination; on arrival he at once found that the rocks
had evidently been metamorphosed in situ; and they no doubt ori-
ginally had only been the ordinary sedimentary clay-slates. Had
he now been content with the decision of the Geological Survey, that
the rocks in question really were greenstones, then he must at once
have come to the conclusion that greenstones could be formed by the
alteration of clay-slates im situ. It did not, however, require a long
examination to prove that the rocks were neither petrologically,

* Messrs. Von Oeynhausen and Von Dechen “ On the Junction of the Granite
and the Killas Rocks in Cornwall” (Phil. Mag. vol. v. p. 246. London, 1829).
+ ‘Contributions towards a Knowledge of the Geology of Cornwall,’ by
Henry 8S. Boase, M.D., p. 235. Penzance, 1832.
4 t ae Treatise on Primary Geology,’ by Henry S. Boase, M.D., p. 44. Lon-
on, 18384.

158 J. A. PHILLIPS ON THE SO-CALLED

mineralogically, or chemically greenstones, or even any allied rock, ,

being nothing more than clay-slates altered in situ, and possessing
none of the properties of greenstones beyond the greenish tinge
which coloured them’’*.

Still more recently (1872) the Rev. W. 8. Symonds, F.G.S., says
that at Penzance the slates are traversed by greenstones and fel-
stones, which occur again at Newlyn, but that what these rocks
are it is difficult to determine ; he is, however, decided in his state-
ment that there are numerous interbedded traps or elvans every-
where associated with these rocks +.

Having in view the differences of opinion which have at various
times been entertained with regard to these rocks, I have endea-
voured, by the aid of chemical analysis, and by an examination of
thin sections, to obtain such data as may be necessary for their
more correct classification, as well as for the approximate determi-
nation of the nature of the different changes they may have severally
undergone, The various specimens submitted to examination were
selected during repeated visits to the locality ; and on one of these
occasions I had the advantage of being accompanied by Mr. S.
Allport, who, I am pleased to learn, purposes laying before this
Society the results of a careful microscopical examination of the
rocks of the Penzance district.

Their position around the shores of Mount’s Bay will be under-
stood by reference to Sheet xxxiil. of the Government Geological
Map. In the neighbourhood of Penzance these rocks, at first sight,
appear to consist of a series of fissile greenish slates dipping from
the granite, and consequently in directions varying from south-east
to nearly south. A closer examination, however, reveals the fact,
not only that some have a much less slaty structure than others, but
also that many of them form compact crystalline beds +, in which
no distinct trace of lamination can be detected; from these hard
rocks, which sometimes, as at Tolearn, project boldly above the sur-
face, the beds appear to graduate through different varieties of horn-
blendic slate until they become ordinary killas. In the rocks ex-
posed along the sea-shore these alternations have sometimes the
appearance of being very gradual; but where they have been opened
upon by quarrying, as on the left of the road at Tolcarn, nearly
opposite St. Peter’s church, the divisions are more distinctly

marked. In this locality an ordinary grey killas overlies a band,

some 4 feet in thickness, of a fine-grained slightly lamellar rock,
somewhat resembling certain varieties of basalt, which is again
succeeded by a tougher and more distinctly crystalline mass. As
there is no exposure across this latter band, its width cannot be
accurately determined; but it probably does not exceed 20 yards.
On the opposite side of the valley it projects above the surface of

* «On the Alleged Hydrothermal Origin of certain Granites and Metamor-
phie Rocks,” by David Forbes, F.R.S. (Geological Mag. vol. iv. p. 57).
t ‘Records of the Rocks,’ by the Rev. W. S. Symonds, F.G.S., Rector of
Pendock, p. 297. London, 1872.
Some of these ‘‘ beds ” are probably intrusive.

a

‘¢ GREENSTONES ” OF WESTERN CORNWALL, 159

the ground to a height of several feet, and is traversed in all direc-
tions by narrow veins having a similar composition; they are, how-
ever, harder, and consequently the weathering of the matrix has
caused them to stand out in considerable relief. Below this outcrop,
and a few feet north-west of the foundations of the new vicarage,
some fresh specimens of a highly crystalline rock were obtained.

This, which is doubtless the “‘ greenstone” of Dr. Forbes, has the
usual greenish colour, is crystalline, and appears, in ordinary hand
Specimens, to consist of an intimate mixture of a dark-coloured
mineral with a considerably less amount of crystalline felspar; in
other localities crystals of felspar predominate.

Although there is an almost continuous exposure of slates for a
considerable distance in a lane immediately east of the Tolcarn out-
crop, no distinctly crystalline rocks are visible; and it consequently
becomes probable that they do not here extend many yards in that
direction, since the road runs nearly at right angles to their usual
line of strike.

When thin sections of these rocks are examined under the micro-
scope, they are generally found to have experienced much alteration ;
but the extent of the change which they have undergone is far more
considerable in some cases than in others; and the nature of this
alteration varies as well as its degree.

Those which have undergone the least amount of metamorphism
are composed of crystalline felspar, augite, or diallage, magnetite,
titanic iron *, and occasional specks of pyrites, with a little apatite ;
they also occasionally contain flakes of brown mica, and, as products
of alteration, viridite and hornblende. The more altered rocks,
on the other hand, consist of a colourless semitransparent base,
through which hornblende and viridite are thickly disseminated.
In this are enclosed green pseudomorphs after augite, shadowy out-
lines of felspar crystals rendered to a considerable extent opaque
by flocculent secondary formations, crystals of magnetite, and pseudo-
morphs after that mineral. Some specimens also enclose crystalline
grains or granular patches of quartz. It will thus be observed
that the ultimate result of metamorphism has been the removal of
almost all the original minerals, and their replacement by others,
which nevertheless frequently admit of the forms of the primary
constituents being distinguished.

It has been before stated that the nature of the successive changes
which have taken place varies in different localities. Thus in the

* A large proportion of the enclosed black opaque mineral appears to have
the form of magnetite; but it also occurs imperfectly crystallized, and as irre-
gular disseminated grains, &c. In order, therefore, to ascertain whether titano-
ferrite is likewise present, a small quantity of black sand was separated by a
careful washing of a considerable amount of pulverized rock, and the ferru-
ginous residue subsequently examined for titanic oxide. In this way the pre-
sence of that substance was clearly established ; but as the quantity is small, and
the exact proportion in which it occurs is of little importance to this investiga-
tion, its presence has been neglected in the analyses; it is, however, probable
that the magnetite in these rocks is generally titaniferous.

& 2. G. Sz No, 126. M

160 J, A. PHILLIPS ON THE SO-CALLED

rock projecting above the surface on the eastern side of the valley
of Tolcarn the majority of the large crystals of felspar are tolerably
perfect, and, when seen in polarized light, exhibit. very distinctly
the coloured strie usually considered to be characteristic of plagio-
clase; on the other hand, the augite has been entirely replaced
either by hornblende or viridite, while magnetite and apatite remain
unchanged. The material constituting the numerous veins by
which this rock is traversed is exactly similar to that forming the
matrix, excepting that it contains a little more green hornblende.

The augite, in specimens broken from the rock near the north-
western foundations of the new parsonage, is, on the contrary, but
slightly altered, while the felspar is much decomposed, the crystals
of this mineral having frequently become almost merged in a
colourless slightly opaque base, in which the form of the original
felspar crystals may still be traced. The rock from this locality
contains, disseminated through it, a considerable amount of brown
hornblende, viridite, magnetite, a little mica, and apatite, with
felspar, and pseudomorphs after felspar interpenetrated by green
belonites.

The compact greenstone exposed in a quarry on the opposite side
of the road has been subjected to a much larger amount of alteration
than either of the foregoing; there are, however, to be observed
outlines of felspar crystals, some portions of which exhibit stris
when seen in polarized light; and pseudomorphs after augite are
tolerably distinct. The apatite is unchanged; but many crystals of
magnetite have been, to a great extent, replaced by a grey siliceous
mineral *,

The somewhat slaty rock which comes in immediate contact with
the above consists of a translucent cloudy base, through which
hornblende and magnetite are thickly disseminated, together with a
little, apatite. Although its original structure is, to a very great
extent, obliterated, faintly delineated outlines both of augite and
felspar can be detected.

At Battery Point there is a rock mainly consisting of tolerably

* In the first specimen in which these pseudomorphs were observed the whole
of the magnetite had been replaced. Iwas therefore at a loss to determine
the nature of the original crystals, and submitted a section for examination to Mr.
Allport, who returned it without having been able to arrive at a decision. A
few days after the return of this section, I received a note from Mr. Allport,
in which he stated that he had then prepared a specimen containing pseudo-
morphic bodies, enclosing unaltered magnetite, and that he was satisfied they
were pseudomorphs after that mineral, adding, “I have no doubt they will be
found to be siderite.” Subsequent examination of a large number of such _
patches in various stages of alteration leaves no doubt of their being pseudo-
morphs after magnetite; but the substance by which that mineral has been
replaced is not generally siderite. On heating a thin section of one of these
rocks with hydrochloric acid in a hollow cell before the microscope, no evolution
of carbonic anhydride took place ; and no action upon the pseudomorphs was
apparent after boiling for twenty minutes in dilute hydrochloric acid; the
whole of the apatite, however, was removed by this treatment, leaving numerous
hexagonal holes through the section, ;

“¢ GREENSTONES ” OF WESTERN CORNWALL. 161

fresh triclinic felspar, together with viridite, hornblende, apatite,
and magnetite. There are also indications of pseudomorphs after
augite ; but this mineral has been entirely replaced by various horn-
blendie and chloritic substances. This is probably the ‘“ compact
felspar” of Dr. Forbes.

The upper portion of the Chapel Rock, immediately west of the
causeway leading to St. Michael’s Mount, consists of a mixture of
felspar, hornblende, viridite, magnetite, and apatite. The felspar,
which is chiefly plagioclase, is little altered ; but pseudomorphs after
augite cannot be very distinctly traced, being represented by patches
of viridite and fibrous hornblende; the magnetite is unaltered. This
rock contains a few crystalline granules of quartz, and very closely
resembles that at Battery Point.

On the sea-shore of the mainland, nearly opposite St. Michael’s
Mount, the felspar is less distinctly defined, but augite is more
readily detected. This rock contains but little apatite, and much
of the magnetite has been replaced by a greenish-grey mineral.

Specimens broken from the nearest point of rock east of the last-
mentioned locality resemble those obtained nearer Marazion, but the
felspar has become still more decomposed; pseudomorphic forms
after augite may nevertheless be traced; apatite occurs in small
quantities; and magnetite is, to a certain extent, replaced.

The rock immediately west of the elvan-course which enters the
sea a little to the east of this place consists of crystalline but much
altered felspar, with crystals and patches of augite in a compara-
tively fresh state ; in addition to which it contains green hornblende,
perhaps tremolite, a little viridite, and partially replaced crystals of
magnetite; no apatite could be detected by the microscope. , Sec-
tions prepared from a rock on the eastern side of the elvan-dyke
do not differ from the foregoing, excepting that the augite is
generally in a still less altered condition.

Altogether twenty-three thin sections of crystalline rocks from
the shores of Mount’s Bay were prepared and examined; but as
none of them exhibits any peculiarity not observed in one or other
of those already noticed, it would be useless to enter into a length-
ened description of them. Im order, however, to compare the che-
mical composition of rocks of this class, differing somewhat in cha-
racter and in various stages of alteration, analyses were made of
four carefully selected specimens, each of which may be regarded as
being to some extent typical. Although made in duplicate, the
mean results only are, for the sake of convenient comparison, given
in the following Table. The mineralogical characteristics of the
rock, as deduced from a microscopical examination of thin sections,
have also in each case been added ; for, although these have already
been generally described, it is desirable that the records of the two
series of investigations should, as far as practicable, be placed side
by side. y

mM 2

162 J. A. PHILLIPS ON THE SO-CALLED

Table showing the Composition of Four Varieties of Altered Crys-
talline Rock from the Penzance District*.

iT. ih TT: IV.

hygrometric ...... 24 “21 22 29

Rite nat 76 | —-80 117 1-92
SULT ena are Bn nnan eae saa Ran 46°32 | 47:22 43:48 47°26
Phosphoric anhydride ...... 52 ‘95 | distinct trace 33
Advising Cleats yen 18:18 19°85 18-60 21°64
Peptic nie vs iv, eds. taxsseeee 82 101 3°68 3°97
os 4, (porsnlphide j, .c..ss.s: "32 wi, traces traces
Rerrots Ox106:.. ...seec-cn~ ce 10°92 9°87 11:38 8:92
Manganous oxide ............ trace trace trace trace
Ni DESEO, eRe cet ay Bie 9:32 | 10-18 12°31 6°01
IMsomeste vee. dae. sei ek 7°46 6:30 6:01 4:02
Las cit a ae ee eRe a Se | 2:07 1:12 1:12 1:91
201 ie a OR “So ERR IR Ri 2-95 2:13 1:69 377
99°88 | 99°81 99°66 100-04
Specific gravity... | 302 | 2-97 3-01 2-98

I. Rock immediately North of St. Peter’s Vicarage, Tolearn.—
This is greenish and crystalline, presenting, in hand specimens, the
appearance of a mixture of felspar and augite, in which the latter
greatly predominates: some of the crystals of this mineral are one
fifth of an inch in length; but those of felspar are usually less dis-
tinct. Thin sections examined under the microscope are found to
contain felspar, augite, viridite, brown and green hornblende, tre-
molite (?), magnetite, occasional specks of pyrites, apatite, a little
granular quartz, and a few flakes of mica. The felspar is much
altered, some of the crystals haying become merged in a colour-
less slightly opaque base; a few, when examined in polarized
light, still exhibit the characteristic striz of plagioclase, while the
remainder are pseudomorphs, either enclosing viridite or traversed
by acicular crystals of hornblende. The augite, which occurs in the
form of large crystals and in crystalline patches, has undergone
comparatively little change, excepting that in some places, and par-
ticularly along certain lines of fissure, it has become partly con-
verted into either viridite or hornblende. Magnetite is sometimes
partially replaced, while apatite remains unaltered.

II. Compact Crystalline Rock from Tolcarn Quarry.—Specimens
of this rock do not materially differ in appearance from those of the
foregoing, excepting that it is less distinctly crystalline, is some-

* It must be borne in mind that the first three of these analyses are of rocks
from the same locality, in different stages of alteration, while the fourth is of a
greenstone from a distance of some three miles. In addition to their other
constituents the rocks described in this paper not unfrequently afford, as pro-
ducts of alteration, traces both of carbonic and sulphuric acids.

*¢ GREENSTONES ” OF WESTERN CORNWALL. 163

what lighter in colour, and has a browner tint. Examined under
the microscope, it is seen to have been subjected to a greater amount
of alteration; but it nevertheless retains cloudy outlines of felspar
erystals, portions of which afford, in polarized light, the usual striae of
plagioclase. The augite is represented by pseudomorphs after that
mineral, coloured by viridite, while magnetite has been partially
replaced by a greenish-grey material, and apatite appears to be
unchanged.

III. Fine-qrained Crystalline Rock from Tolearn Quarry.—This
rock is of a dark sage-green colour, is exceedingly fine-grained, and
breaks with a somewhat conchoidal fracture. A microscopical ex-
amination of thin sections shows it to consist of a granular base,
containing quartz, through which hornblende and magnetite are
plentifully disseminated.

In spite of the large amount of alteration to which this rock
has evidently been subjected, its original constitution may still,
to some extent, be traced, augite having become replaced by patches
of viridite and hornblende, while felspar is either represented by
indistinct pseudomorphic forms, or completely merged in the granular
translucent base; magnetite, on the contrary, remains comparatively
unchanged, and apatite is apparently absent.

IV. Chapel Rock, west of causeway leading to St. Michael's
Mount.—This rock, which rests upon slate, is crystalline, of a
greenish-grey colour, and is traversed by numerous fissures, some
of which have become filled with hornblende; in its general ap-
pearance it more nearly resembles a felsite than either of the fore-
going. Under the microscope it is seen to be composed of a fel-
spathic base, porphyritically enclosing crystals of somewhat altered
plagioclase, together with a little granular quartz ; throughout this
are disseminated viridite, apatite, crystals of magnetite, and angular
patches of hornblende, the latter representing replaced crystals of
augite.

With regard to the petrological character of the rocks above
described, I would remark that No. I. is probably a rock resulting
from the alteration either of gabbro or dolerite, while No. II. and
No. III., which do not differ very materially from it in chemical
composition, may be regarded as the results of a still more advanced
stage of metamorphism *. In No. IV. the proportion of silica is the
same as in Nos. I. and IIL., but the amount of lime and magnesia is
somewhat less, while that of alumina is greater.

Associated with the before-described rocks are others, possessing a
distinctly lamellar structure, but which appear to graduate from
ordinary killas, on the one hand, to imperfectly cleavable slates, on
the other.

A slate belonging to the killas class, which comes in contact with
highly crystalline rocks in the Tolearn Quarry, analyzed in dupli-
cate, afforded the following results (sp. gr. = 2°76) :—

* Mr. Allport, when on the ground, paczinge’ that these rocks might be
altered gabbros.

164 J. A. PHILLIPS ON THE SO-CALLED

Iu, ‘eB:

hyarometHc?,.ss:¢ ws dein 68 "58
pies { combined ;,. 51, gin taste 3:20 3°14
DLIEA of: ¢ bee 'b o:x'eXe ghlgieehe eles 63:45 63°37
Phosphoric anhydride ....,. ua) abrace trace
Almmung ) liga eain chtatae ee 18°33 18°31
Verrie oxide |. 5/2. ¢% xis aa Gs 2°68 2°63

ws . persuiphide.*). cuss: haere traces traces
Ferrqus oxide... ah yee she 3°66 3°60
Manganous Ode . 245 ws ine ae trace __ trace
TE ee or "43 "59
Maeneetey |. gon k ah db aalasty hi 3°00 2°92
0) CE eee oe 5 AER hata ale 3°73) 4:00
N11 (a ae a RR ig AER, ‘78 65

99°94 99-79

This is a grey killas, of more than usual hardness, and of which
some of the transverse headings are stained of a reddish-brown
colour by hydrated ferric oxide, while others are coated with
chlorite.

Under the microscope it is found to consist of a nearly colour-
less base, enclosing a little viridite with mica, and numerous dis-
tinct fragments of quartz, together with a few aggregations of a
granular mineral, having the appearance of minute garnets. These
groups vary from >,),, to =}, imch in diameter, and are sometimes
darkened and rendered opaque by.the liberation of hydrated ferric
oxide ; when viewed between crossed nicols the base presents the
usual coloured mosaic characteristic of such rocks.

Upon a hill a little west of the tin-smelting works in the valley
above Tolcarn, a quarry has been opened for the purpose of ex-
tracting the hard pyroxenic rock to be used as road material. The
killas is here seen between two beds of crystalline rock, and differs
from that in the Tolcarn quarry in containing thin foliations of
hornblende, with perhaps a little mica, in planes parallel with its
cleavage. This rock appears to be intermediate in character be-
tween the ordinary clay-slate of the district and the crystalline fissile
rocks, which now remain to be described.

These rocks, which are generally fine-grained, vary in colour from
bluish green to greenish grey, and differ considerably in the facility
with which they may be divided into thin plates. In some cases
slaty cleavage is distinctly recognizable, while in others nearly all
trace of such a. structure has become obliterated, under which cir-
cumstances the rock breaks with a subconchoidal fracture ; between
these extreme limits various degrees of fissility may be observed *.
When thin sections are examined under the microscope they are
found to consist to a large extent of hornblende, which occurs both
as crystalline patches and as felted acicular crystals. Sometimes

* It is often impossible to determine to what extent the fissility of these

rocks is due to true: slaty cleavage, and how far it may be the result of
bedding.

6 GREENSTONES” OF WESTERN CORNWALL. 165

this rock is almost entirely composed of closely felted crystals,
among which is disseminated a little flocculent viridite, or more
rarely magnetite; not unfrequently the hornblendic mineral is
thickly distributed through a colourless slightly opaque base,
which forms but a small proportion of the whole. Patches of
quartz may also occasionally be observed; and into these minute
green crystals project from the surrounding mass like the foliage of
a tree into the open spaces between its branches. In addition to
green hornblende, and perhaps a little chlorite, these rocks fre-
quently contain brown hornblende, together with colourless crystals
of a mineral which is probably tremolite. Magnetite is likewise
sometimes present; but instead of being in a crystalline form, as in
the case of the rocks before described, it is generally disseminated
either as a sooty powder or in rounded aggregations.

Twenty-six different sections of these rocks were cut in various
directions, and examined under the microscope with the foregoing
general results, while four specimens were subjected to chemical
analysis ; in each case the mean of two separate estimations is given
in the following table; and a description of the appearance of thin
sections, seen under the microscope, is appended.

Table showing the Composition of Four Varieties of Crystalline
Slaty Rock.

I. II.* III. IV.
hygrometric ......... “54 1:37 “53 68
Water { Waisted 77 1:56 87 | 1-06
oh aS AAI ESAT UE 37°44 35°58 39:20 | 36°57
Phosphoric anhydride ...... ‘27 | distinct trace “14 68
PTO... dr45 donb deessiva 1632 21°14 16:22 | 15°34
WERE OIE... 52. ced ce pnnene 7°33 14-74 4:82 T57
3, persulphide............ 23 traces as aes
PSEFOUS OMIA oe ccinnwvenene 14°75 10°32 15°88 14°84
Manganous oxide ........0... trace trace aes trace
ARB Beier «Sich a SUE RR widows 12°83 9-20 10°70 | 12:58
MMH Ss dence snncceneskannse' 6°06 2°78 6°56 5°98
DGTABSN © sas ib sak takenceiencs 1:50 ‘95 2°69 1-44
PORE iek Ainicous Mis <cekpene 1°85 2°11 1°45 2°99
99°89 99°75 99°66 | 99°73
Specific gravity o........00...| 3°29 31d 3°26 3°28

I. Paul Hill, Paul—This rock is hard, dark in colour, very fine
in grain, and possesses a distinctly slaty structure. Thin sections,
examined under the microscope, are found to be composed chiefly of
minute thickly matted green crystals, through which are disse-
minated a little flocculent viridite, and occasionally some dust-like
magnetite. It encloses small irregular patches of milky or trans-
parent quartz, traversed by minute hornblendic crystals.

* This analysis was published in the Quart. Journ. Geol. Soc. vol. xxvi-
p. 329, but is introduced here for the sake of comparison.

166 J. A. PHILLIPS ON THE SO-CALLED

II. Rosehill, Castle Horneck.—This rock is brown in colour, and
is more fissile than the foregoing, consisting largely of crystalline
hornblende, through which hydrated ferric oxide and a chloritic
mineral are disseminated ; it includes patches of some nearly colour-
less translucent material, into which minute green crystals project
from the circumference *.

III. Chyandour, Penzance.—This is a very hard rock, dark in
colour, frequently breaking with a subconchoidal fracture, and in
which nearly all trace of cleavage or lamination has disappeared ;
when freshly broken its surface is sometimes spotted with patches
of the size of small shot of brown crystalline hornblende. Thin
sections are found to be composed of small interlaced acicular
crystals, through which a flocculent mineral of a darker green
colour is disseminated.

IV. Rosemorran, Gulval._—A distinctly lamellar rock of the usual
dark green colour, and to a considerable extent fissile. Thin sec-
tions are seen to consist of a felted aggregation of a fibrous horn-
blendic mineral, in which are imbedded a few larger crystals of
hornblende. Irregular minute patches of granular quartz are scat-
tered throughout the mass; these, which contain liquid-cavities, are
sometimes milky, and at others transparent, being traversed in all
directions by belonitic crystals, which penetrate from the exterior.
A few nebulous stains of viridite and hydrated oxide of iron, together
with occasional needles of apatite, are disseminated throughout.

It will be observed that the foregoing specimens of slaty rock,
obtained from localities at considerable distances apart, closely
resemble one another in general chemical composition, and are
almost identical in their petrological characteristics. They contain,
on an average, ten per cent. less silica than the more highly
crystalline rocks of which analyses are given p. 162; and a small
proportion of phosphoric anhydride is present in all of them.

The microscopic closely felted crystals, of which these rocks are
often to a considerable extent composed, seldom exhibit distinct
outlines, and not unfrequently assume a more or less flocculent
appearance.

This mineral, which is dichroic, is believed to be hornblende,
although the low percentage of silica found by analysis might appear
to indicate the presence of chlorite. When we take into considera-
tion, however, the small proportion of combined water in these
rocks, it becomes apparent that if chlorite be present the amount
must be very small 7.

* All the rocks belonging to this class are to a large extent decomposed by
boiling hydrochloric acid; but on placing an unmounted section from this
locality on a glass slip under the microscope, and covering it with hot acid,
some of the lighter patches were observed to be more readily attacked than the
remainder of the mass. In one or two casesa few minute bubbles of carbonic
anhydride were evolved, but the quantity was too small for estimation by
analysis.

if ‘The proportion of silica in hornblende usually varies between 40 and 50
per cent.; but according to Rammelsberg (‘ Handbuch der Mineralchemie,’
pp. 492, 493) a black crystal of this mineral from Filipstad, Warmland, con-

“ GREENSTONES ” OF WESTERN CORNWALL. 167

In addition to killas and these green slates there is a variety of
slaty rock which is somewhat greyer in colour than the fissile
hornblendic rocks. This contains a larger amount of magnetite ;
and the proportion of colourless base is also more consider-
able. Sections prepared from a rock of this description, which has
been quarried near Penlee Point, on the right of the road from
Newlyn to Mousehole, were found to consist of a mixture of horn-
blende, viridite, and magnetite enclosed in a transparent base.
The granular colourless base is traversed by parallel undulating
irregular bands of feathery green crystals; spongy magnetite is also
present, and is either disseminated throughout the mass, or forms a
border to bands or patches of hornblende. The base is thickly
interlaced with belonites; and its granular character becomes appa-
rent on being examined by the aid of polarized light.

A freshly broken specimen of a greenish-grey rock of this class,
from a quarry in the village of Newlyn, was analyzed in duplicate,
and afforded the following results (sp. gr. 2°89) :—

- TI.
hygrometrie: v.00. “19 “20
Hale { etait ee! h 9: ‘55 63
Se eg es US U.S aioe 50°57 50°51
Phosphoric anhydride ........ 12 13
nneeinNe? Vs Ser aldiw a pales 19°65 19°69
Merrermrmrrel Soe SE ee. 1:46 1:28
pr permulpiiders es. ee traces traces
ea i ee 7°24 7-40
Manganous oxide 2.0.0. 250... trace trace
rere, SPOS Seti) hes TAT 7°38
eee Le 2 2a Skalelsais's 3 7:93 8°04
eee SES SU. SWS St 1:08 1:00
Pere teehee aed tod: 3°46 3°41
99°72 99°67

tained only 37:84 per cent., while the percentage found in a specimen from
Fredriksvirn was still less, or 37-34 per cent.

A specimen of black crystalline hornblende from a vein in the cliff near
Botallack, analyzed by A. G. Phillips in the author’s laboratory, afforded the
following results (sp. gr. 3°24) :-—

WY BNGE Sac <<nadeesss See cieia 94
LU meel) St Repeal aaah hell 40:12
PUI HRNINE Yo dons idenpaxcadaecnscuss 16°75
ERI GRIGG ok. vdccececsaccduaries 4:32
WMGPPOUS OXIDR-. codnisnsuedeacnans 12-70
Manganous oxide .....+......... 3°42
MA tay hk aac aidan sddone cuts 12°37
DUCE, sive sen <-nscdasn can -case 6°82
IPRA. Deh tao y cada ackenves sondsas . trace
PP Gade ates «a thckaenes sbbddidns 2°65

10009

Another dark-coloured crystal from a neighbouring locality contained
39°73 per cent. of silica; it is therefore evident that the proportion of this sub-
stance in Cornish hornblendes is sometimes unusually low.

168 J. A. PHILLIPS ON THE SO-CALLED

This is a fissile and distinctly lamellar rock, which is employed
for road-making. Although its prevailing tint is grey, itis spotted.
with patches of light brown hornblende mixed with a little greenish
mica. An examination of thin sections shows it to consist largely
of greyish microlites, so thickly matted together in a transparent
base as to render the general mass to some extent opaque. In this
mixture are patches of brown hornblende, a litttle mica, tremolite,
magnetite, and a few spots of iron pyrites: the magnetite is not
generally crystalline; and when seen in polarized light the base
exhibits no indication of crystalline structure. This rock contains
a few distinct markings having a resemblance to cloudy crystals of
plagioclase seen in certain varieties of dolerite.

The foregoing examination of rocks forming the shores of Mount’s
Bay indicates that they principally belong to three distinct series,
differing alike in their chemical composition and mineralogical con-
stitution :—I1st, augitic rocks of a class generally believed to be of
igneous origin; 2ndly, ordinary clay-slates; 3rdly, hornblendic
slaty rocks.

The rocks included in the first of these divisions appear to have
been originally composed of felspar, augite, magnetite, and apatite ;
all have, however, in a greater or less degree, been subjected to
influences through the agency of which the crystalline character of
the felspar has to some extent been destroyed, while viridite and
hornblende have been produced by the alteration of augite. In
some instances these secondary changes have reached such an
advanced stage that the felspar is either represented by pseudo-
morphic forms, or has become transformed into an amorphous
or granular base ; in other cases the augite has been completely
removed, giving rise by its decomposition to pseudomorphs of which
hornblende or viridite are the chief constituents. In such specimens
apatite sometimes appears to be the only original mineral which has
not suffered from alteration. That these changes have, to a great
extent, been produced by a rearrangement of the ultimate con-
stituents of the rock, rather than by their removal and substitution,
is probable from the results of analyses of variously altered
specimens.

The rocks of the second class are the ordinary clay-slates or
killas of the Cornish mines.

The green lamellar rocks constituting the third class appear to
present evidences of sedimentary origin, and may have been pro-
duced either by the metamorphism of other rocks subsequent to
consolidation ; or, when in the state of paste, magma, or mud, they
may have become consolidated under such chemical and physical
conditions as to lead to the generation and crystallization of the
various minerals which they enclose. Reference to the table,
p- 165, shows that they not only possess a great resemblance to
one another in chemical composition, but also that in this respect
they more nearly approach the crystallized igneous rocks than
ordinary clay-slates; the proportion of silica present, however, is
considerably less than is usually contained in hornblendic slates.

Carr-Cornwatt, Disrrict.—The so-called greenstones of Cape

‘¢ GREENSTONES ”? OF WESTERN CORNWALL. 169

Cornwall extend from the northern shore of Porthleden Cove to a
quarter of a mile beyond the Levant Mines; their total length is
about two miles, and their greatest width not above three hundred
yards. In many places, however, they form a mere border along
the coast line.

These rocks, which are stratified, and have a general dip from
the granite, vary considerably in colour, since they graduate from a
bluish-grey through sundry shades of green to dark brown or black.
Slaty cleavage is often well defined, but in all cases in which it has
become partially obliterated it is rendered apparent by weathering,
and there is always a tendency to break into thin plates.

They are generally lighter in colour than the slates bordering
Mount’s Bay, and hornblende is not unfrequently more or less con-
centrated in laminz parallel with the cleavage. In some places,
and particularly in the cliff near Botallack, the rocks are traversed
by veins of hornblende, axinite, garnet, or magnetite; garnets are
also sometimes plentifully imbedded throughout the rock.

In some localities along this part of the coast the slaty beds
enclose a considerable amount of quartz in the form of layers fol-
lowing the lamination of the rock; more rarely it forms distinct
yeins crossing the lines of bedding. Wherever such an admixture
of quartz with slaty rocks is observed, the parallelism of the lamine
is interfered with, and the rock becomes more or less bent, cor-
rugated, or twisted; this contortion may perhaps to some extent
have been the result of crystallization*.

Thin sections of these rocks are found to be composed of a colour-
less granular base, throughout which reticulated feathery horn-
blendic crystals are thickly disseminated, together with grains of
quartz, magnetite, sometimes minute garnets, and a few flakes of
brown mica. This mixture is occasionally traversed by acicular
hornblendic crystals, which are frequently arranged in stellate
groups of about +6, inch in diameter. Sections cut perpendicularly
to the foliation show that the magnetite and hornblende are some-
times arranged in corrugated bands, and that the proportion of
transparent base is much larger than in the somewhat similar rocks
of the vicinity of Penzance.

* T stated in a former paper (Quart. Journ. Geol. Soc. vol. xxxi. p. 341)
that it is not improbable that the repeated widenings which have evidentiy
been experienced by fissures enclosing mineral veins, may sometimes have been
produced by mechanical forces resulting from crystallization. As somewhat
supporting this hypothesis, it may be stated that, in the manufacture of “soda
crystals,” at the chemical works of Messrs. Gaskell, Deacon & Co. at Widnes,
Lancashire, vessels made of thin boiler-plates have been substituted for the
ordinary erystallizers of cast iron. The result has been that a film of the
solution of carbonate of sodium found its way between the laps where the
plates are rivetted, and in crystallizing foreed the sheets of metal apart.
To such an extent were the vessels damaged by this action, that they very shortly”
required to have their joints reclosed by careful caulking. Since the pub-
lication of the paper above referred to I find that Volger attributes the forma-
tion of the fissures enclosing mineral veins to the force of crystallization, and
that the notion of vein-fissures being opened as crystallization advances is
defended by Griimer. See ‘Chemical and Geological Essays, by T. S. Hunt,
E.RB.S. &e. p. 202.

170 J. A, PHILLIPS ON THE SO-CALLED

In other specimens the crystals are exceedingly minute, and
closely matted together, without the stellate or banded arrangement
above described. All the sections examined contain magnetite,
which is often finely granular, and is seen, with a combination mag-
nifying 250 linear, to be in the form of imperfect crystals, with
rounded edges. Less frequently the magnetite is more distinctly
crystalline; and when this is the case, it is generally, in part, re-
placed by some siliceous mineral.

Occasionally the transparent granular base of which these rocks are,
to a large extent, composed, instead of containing radial or flocculent
crystals, encloses numerous irregular flakes of brown hornblende or
mica; magnetite and viridite are also present in this variety.

No altered gabbros or dolerites of the class found in the neigh-
bourhood of Penzance were observed in the Cape-Cornwall district ;
and the majority of the sections examined, nineteen in number,
afforded evidence of the slaty or schistose nature of these rocks *.

Gurnarp’s-Heap Districtr.—The greenstones in the vicinity of the
Gurnard’s Head, as laid down upon the Map of the Geological
Survey, extend, with certain interruptions, from that promontory
to Porthzennor Cove. An examination of these rocks in situ soon
renders it apparent that they very closely resemble those found in
the neighbourhood of Newlyn and Penzance. The slaty rocks,
which here, as elsewhere, dip from the granite, are perhaps more
crystalline and somewhat darker in colour than those of the former
locality ; the gabbros or dolerites, on the other hand, have under-
gone a still larger amount of metamorphism.

The upper portions of the Gurnard’s Head are composed of a
hard, dark, bluish-green slate, consisting of a transparent base,
through which brown and green microlites, with granular mag-
netite, are plentifully and regularly disseminated. Beneath this is
a dark crystalline rock, which, although apparently stratified, ex-
hibits no evidence of cleavage. ‘Thin sections are seen to consist of
the usual transparent base, enclosing hornblende and viridite, with
pseudomorphs after magnetite, and small irregular crystalline
patches of augite or diallage. It also contains a few small and
much-altered crystals of felspar, together with various shadowy
patches and outlines suggestive of extensive metamorphism.

A wide band of rock, coloured upon the Geological Map as green-
stone, immediately south of the headland, is shown by the micro-
scope to consist of a translucent granular base, containing grains of
quartz, throughout which the ordinary minerals are disseminated in
the usual way; it also encloses a few minute garnets.

A narrower band, a little south of the above, has sometimes a slaty
structure; and sections of it are found to have a similar composition.

The two next headlands north-east of the Gurnard’s Head, form-
ing the extreme limits of Porthglaze Cove, are, to a great extent,

* A greenstone from Botallack, analyzed by the author (Quart. Journ. Geol.
Soc. vol. xxxi. p. 329), was found to contain 48 per cent. of silica; in all
other respects its ultimate composition closely agrees with that of the pyroxenic
rocks of Penzance, and the slaty blue elvans occurring between St. Erth and St.

Stephen’s. These latter rocks do not materially differ in chemical composition
from the ash beds of St. Kew.

“‘ GREENSTONES” OF WESTERN CORNWALL. LTA

composed of a crystalline rock of the same class as that near St.
Peter’s vicarage, Penzance. This, however, has experienced a
larger amount of alteration than the majority of those on the shores
of Mount’s Bay, since none of the sections examined retain any
trace of unaltercd augite or diallage, the magnetite has been par-
tially replaced, and apatite sometimes appears to be the only un-
altered mineral present. Diallage has been found in sinking a shaft
at Carnellow Consols, near the point of junction of the crystalline
rock with slates. The nature and distribution of the various minerals
seen in thin sections of the more highly metamorphosed portions of
this rock tend to the supposition that the darker beds forming the
base of Gurnard’s Head have had a similar origin ; but if this be the
case, they have been subjected to still more extensive alteration.

At Porthglaze Cove there is a hardened killas, sections of which
show it to be composed of a transparent quartzose base, coloured by
flocculent green and brown microlites, which are plentifully distri-
buted throughout.

A little north of this point the cliff becomes darker, more highly
crystalline, and less distinctly slaty: hand specimens resemble those
of the rock forming the eastern base of Gurnard’s Head; but when
thin sections are examined a marked difference is observed. This
rock is chiefly composed of closely matted hornblende, among which
are distributed a flocculent mineral of a greyish-green colour,
numerous grains of magnetite, a few patches of a granular base,
and some grains of quartz.

The northern side of the promontory forming the southern
boundary of Pendower Cove is composed of a dark-green crystalline
rock approaching to black. In hand specimens it resembles the
bottom rock of Gurnard’s Head; an examination of thin sections,
however, shows that it differs materially from it, being principally
composed of matted green hornblende of various shades, a small
quantity of transparent base, and occasional grains of quartz, but no
magnetite.

An analysis of this rock, made in dapiivates afforded the following
results (sp. gr.=3°35) :—

E ite

Hiyprometri¢ }.’.. 0)... 34 “40

Male encuined Cece 1:05 ~—«-:107
OR oe Se Pear a ere 37°72 37°63
Phosphoric anhydride.......... “41 “46
Peliimia see ee. Sue ee SS E719 17°15
Perrig oxiden ery}. Dec 2. 7°75 7:53
ae Berauepniae Hoe Se oe yee! traces traces
Perrousvemae so. oie Pe. 15°81 16°03
Manganous oxide ............ “43 “40
PIG: AE PP ey AP I 14-00 13°95
Monesia eye loL ae APM 2 3°41 3°52
1 gan edna Bee a) ptiee ro SA hs 1:02 94
LT RRR ARES sie on MR A Ad 1:06 95

100719 = 100-08

172 J. A. PHILLIPS ON THE SO-CALLED

It will be observed that its composition very nearly agrees with
that of the crystalline hornblendic slates in the neighbourhood of
Penzance.

A similar rock occurs on the headland forming the northern
boundary of the Cove, and is associated with a brecciated quartzite,
the angular fragments of which have been cemented together by
crystalline hornblende.

From the foregoing description of the rocks of this district it will
be seen that they are in all respects almost identical with those in
the neighbourhood of Penzance. Nineteen sections of rocks from
this part of the coast were cut and examined.

Sr.-Ives Distrror.—The next considerable exposure of rocks be-
longing to this class is that laid down on the Government Geological
Map as extending, with but one inconsiderable break, from a short
distance east of Trevail Point to a little beyond Porthminster Point,
in St.-Ives Bay; its total length is three miles, and its greatest
width about a mile. The only interval in the continuity of
these rocks occurs along the valley through which passes the turn-
pike-road leading from St. Ives to Halsetown and Towednack; this
has been formed by the erosion of clay-slates which here come
to the surface. The rocks in this neighbourhood are very similar
character to those forming the shores of Mount’s Bay, and consist of
hornblendic slates, interbedded with augite rocks and occasional
bands of clay-slate.

Although numerous thin sections were prepared from the rocks
of this district, they are all so nearly identical in their petrological
characteristics with those from the neighbourhood of Penzance, that
it has not been thought desirable to subject any of them to analysis,
and a detailed description of their appearance under the microscope
becomes unnecessary. It may, however, be stated, generally, that
a large outcrop of crystalline rock, very similar to that of Tolearn,
occurs in the immediate vicinity of St.-Ives pier. At Clodgy Point
the rock contains well-defined and unaltered crystals of plagioclase,
slightly altered augite or diallage, with pseudomorphs after one or:
other of these minerals, and pseudomorphic crystals after magnetite ;
it also contains viridite, a little granular quartz, and a few crystals
of apatite.

Near Carrack Olu there. are large naa masses of very hard and
tough greenish slates, consisting of reticulated crystals of hornblende,
with a little viridite: and quartz, disseminated throughout a subor-
dinate amount of apparently felspathic -base. . .Near Porthminster
Point, at the eastern extremity of the formation, ordinary clay-slate
alternates with more or less crystalline rocks. The slate, which is
here much contorted, is hardened by metamorphic action, and is
traversed by numerous veins of quartz; quartz also frequently
occurs in thin layers following the lines of cleavage ; thin sections
of these slates present the usual characteristics of ordinary killas.
Twelve sections were prepared from these rocks.

DisTRict BETWEEN Sr. Err anp CamBorne.—On referring to the
Geological Map, numerous bands of greenstone will be observed ex-

“‘ @REENSTONES ” OF WESTERN CORNWALL. 173

tending in a north-easterly direction from the former locality to the
latter, or for a distance of about six and a half miles: their greatest
ageregate thickness approaches half a mile; and all are compre-
hended within a belt of less than a mile in width. On either side
of these rocks, and in their close proximity, at least two large elvan-
courses run parallel with them throughout their extent; and small
elvans, or branches from the larger ones, sometimes take their
course between the different bands.

On the Map of the Geological Survey, Camborne is indicated as
being the most eastern limit of these rocks; but it is evident that,
although considerably reduced in width, they must be continued at
least two miles further east, as they again make their appearance in
the workings of South-Roskear, East-Pool, and various other mines.
They are generally known as “ blue elvans,” or ‘irestones” (iron-
. stones), the latter name being applied to them by miners on account
of their extreme hardness.

In some places these rocks stand out boldly above the surface of
the soil, as a little north of the viaduct situated two miles west of
Camborne; in others they have been laid open either by railway-
cuttings, mining-operations, or quarrying. On account of their
hardness and exceptional toughness, they are much used throughout
the district as a material for metalling roads, and are occasionally
sent to considerable distances for that purpose. Their structure is
ordinarily more or less slaty; but they sometimes break with a
hackly or even subconchoidal fracture. In colour they vary from
grey to greyish or bluish green. Rocks of this class often enclose
specks of pyrites, and are occasionally mottled with bands or veins
of a brown or greenish-brown material, which is sometimes imper-
fectly crystallized. These veins are probably composed of a mixture
of garnet and axinite ; but the crystals are so imperfectly developed
in the specimens examined as to render their identification somewhat
difficult.

At No-Man’s-Land, two miles east of the village of St. Erth, a
quarry has been opened, for road material, upon a band of exceedingly
hard bluish-grey rock, which is frequently spotted with pyrites. It
is contorted, and somewhat coarse in texture, exhibiting a disposition
to break into plates, in accordance with certain parallel planes.
This is mixed with a considerable amount of clay-slate, the cleavage
of which approximately coincides with the planes of fracture of the
enclosing rock ; this slate has evidently been subjected to influences
by which its iron has, to a large extent, been converted into hydrated
oxide.

Under the microscope thin sections are found to be composed of
filamentary reticulated crystals of various shades of green, very
thickly pervading a transparent base, which encloses patches of
quartz, magnetite, and pyrites. In addition to the greenish crystal-
line mineral before referred to, there is also present a greyish-green
nebulous substance, which usually occurs in patches,

Two miles west of Camborne, after passing through an archway
in the embankment supporting the railway viaduct near Carn-Cam-

174 J. A. PHILLIPS ON THE §0-CALLED

borne Dressing Floors, a wide band of bluish-grey greenstone, which
has a distinctly slaty structure, crops out some thirty or forty feet
above the surface.

An analysis, in duplicate, of this rock, afforded the following
results (sp. gr.=3:03) :—

{. ET,

hygrometric ..cws).iWaed “14 12

Water combined . ,.eaektan 60° +65
UCR. Jb i Gb ght. Ritik, Lee 48°30 43°41
Phosphoric anhydride ........ trace trace
AMIN GIG Midi ihie b chet bee 17:04 17:02
Bermcioxides 14h alekiok ARE RARE 2°73 2°68
jp suqpersulphideyc 2ct- cu owes traces _— traces
Nermpus,oxide. ti .4).eule veces 9-50 9-41
Manganous oxide j.w iw 5 .e% 28s trace trace
PANE noise ys eee ee 13°30 13:15
Marmesis.n igiiids Usha k hee 6:18 6°20
POA SR td se Wes <2 ERE ok 30 30
RIO Feats Ate rile, le Mate eae 2:01 2:13

100-10 10007

Sections examined under a low power are seen to consist of
flaky hornblende and green or brown microlites, thickly dissemi-
nated through a transparent base, in which sooty magnetite and
cloudy viridite are somewhat abundant; under a moderately high
power the latter is observed to closely resemble certain varieties of
chlorite. Some of the sections prepared from this rock expose irre-
gular patches of transparent base, into which imperfect hornblendic
crystals project from the surrounding mass, as observed in the case
of hornblendic slates from the neighbourhood of Penzance. This
rock sometimes also encloses small imperfect crystals, which are
probably garnets; and a general tendency to.a banded arrangement
of its constituent minerals is observable.

A little west of the Camborne station the same mass of green-
stone has been laid open by the railway-cutting; here the rock is
similar to that last described, but a little darker in colour, and
perhaps somewhat closer in texture. It is, however, traversed both
by bands and veins of a granular brown substance, which has occa-
sionally a pinkish or greenish tint. As in the case of specimens
obtained from the same mass of rock two miles further west, the
minerals distributed through the base exhibit a tendency to arrange
themselves in nearly parallel bands.

A microscopical examination of numerous sections from this loca-
lity shows no considerable difference between this rock and that from
Carn Camborne: the granular brown substance before referred to
as occurring in specimens from this cutting consists principally of
small imperfect garnets, with which is probably associated a little
axinite. One of the bands of greenstone here laid open, which in
hand specimens does not appear to differ materially from the others,

““ GREENSTONES ”? OF WESTERN CORNWALL. 175

is seen, in thin sections, to consist principally of an aggregation of
rounded grains of quartz, having an average diameter of ;4, inch.
_ These are closely cemented together by a mineral of a greenish
colour; and among them are small fragments of hornblende.

At South Roskear the greenstone is coarser and more slaty in
texture than at Camborne; magnetite occurs more plentifully, but
has been to some extent replaced by a siliceous material ; no garnets
were observed in this rock.

The specimens obtained from East Pool are rather fine-grained,
contain a less amount of hornblende, and are much mottled by the
presence of the brown crystalline mineral before described. Eighteen
sections of rocks from this district were cut and examined.

Newtyn East.—At North Huel Rose, in this parish, there is a
band of greyish greenstone, which is not laid down upon the Map
of the Geological Survey. This rock has a very compact structure,
almost resembling that of chert, and although exceedingly hard and
tough, exhibits distinct evidence of a disposition to break into plates.
An analysis, in duplicate, afforded the following results (sp. gr.
3°02) :—

i. Il.

hygrometric..........- 24 20

ay { somibined <0... 000: 64 52
Se co ee oe 46°57 46-70
Phosphoric anhydride ........ 63 D9
eo 19-50 19°74
MEI ORGOG how iie les 8 Se ws § 1°84 1°81
ee traces traces
Ferrous oxide........ he iv ial 12°80 12°81
Manganous oxide ............ trace trace
BROS SSM egrdhs 2 ais Beet ves » 9-91 9°89
A Bo aces Mas Woes Riven At 4 3°93 3°84
Be is fle Users, Hyd ain cack dteie ee ‘51 53
aE Eien. Cite. hung softer 3°50 3°40
RT ech, aac tet aah Mane vais trace trace

100-07 100-03

Under the microscope it is seen to consist of a transparent base,
enclosing matted crystals of hornblende, together with magnetite
and viridite. Between crossed prisms the base presents the ap-
pearance of a coloured mosaic, indicating its granular constitution.
The hornblende is disseminated, principally in the form of undulating
filamentary and approximately parallel bands, between which the
base is more or less charged with a flocculent green mineral; a
nebulous greyish substance is also usually present, and angular
patches of magnetite are somewhat numerous. Four sections of
this rock were examined.

Sr. SrepHens.—At Terrace, about a mile and a half west of St.-
Stephens church, there is a large blue elvan, which has for many
years been quarried as a material for repairing the parish roads.
This greenstone is not shown on the Geological Map, but is several.

Q.J.G.8. No. 126. N

176 J. A. PHILLIPS ON THE SO-CALLED

fathoms in width, and has a direction north of east and south of
west; its dip and apparent cleavage very nearly coincide with those
of the enclosing slates. Although large quantities of stone have
at various times been removed from this locality, the quarry has not
been worked in such a way as to expose a complete transverse section
across the band. It is, however, evidently traversed for a con-
siderable distance by a nearly perpendicular fault; and at a point
about two hundred fathoms south-west of the quarry it is inter-
sected by the Great Terrace Elvan. This elvan is remarkable from
the circumstance that many of its felspar crystals have become re-
placed by a mixture of schorl and cassiterite, which at different
periods has caused it to be worked as a source of tin-ore.

A fresh specimen of this rock afforded by analysis the following
results (sp. gr. 3:03) :—

KE. it:

hypromebaret o.5 ai), doesiber -20 "22

—— { COMPA, hoc 5: bee erapeete vi? ‘67
DIG. dichammctet makes haeeete ce 46:89 46:96
Phosphoric anhydride ........ 31 B82
PRRRNER, AB co icree aig Sake wii cans eae 20:46 20°53
POTIe ORE oan Vike oe he 2°19 2-09
0 DEEDS i whee h. | traces traces
Pesirows Ome) oo. 05 ee Sars WS Ligls 1isks
Manganotis oxide, +... ..... «. SIMs trace trace
fiends: BREE es he. BR 8°37 8:42
Maonpsis lye? s,s a Hee eR. 5°83 5°91
DOERSSS RIES 3 ji aloG Vig ae trace trace
Pee: ©. A Be. Fae) Bis Se 3°87 3°80

100:04 100-10

Thin sections, examined microscopically, are not unlike those of
the rock from North Huel Rose, excepting that the hornblende is
disposed in more regular bands, and the magnetite is also more dis-
tinctly banded. The edges of the green hornblendic stripes are
fringed by feathery crystals of that mineral; and the patches and
crystals of magnetite are often cracked, or have become replaced by
a siliceous mineral.

_ Between the nearly parallel hornblendic bands before referred to,
the transparent colourless base is traversed by acicular hornblendic
crystals, the longer axes of which lie (mostly in one direction) in
planes coinciding with the cleavage. Viridite is always present in
small quantity; and one of the sections prepared from this rock en-
closes small patches of a blue mineral, which is perhaps indicolite.
Among the fissile hornblendic rocks, several have been observed to
contain occasional small crystals of apparently a schorlaceous mineral.
Five sections of the Terrace blue elvan were cut and examined.

- It will be observed not only that the blue elvans, or greenstones,
occurring between St. Erth on the west and St. Stephens on the
east, although varying somewhat in the relative proportions of their
several constituents, resemble one another in their general chemical

‘€ GREENSTONES ” OF WESTERN CORNWALL. L77

composition, but also that in this respect they do not materially
differ from the highly crystalline augitic rocks of the neighbourhood
of Penzance. It is also remarkable that in all these rocks the pro-
portion of silica is very nearly constant. The greenstones, how-
ever, which occur between these limits afford distinct indications of
slaty cleavage; and thin sections do not exhibit the distinctive
angular arrangement of crystals commonly seen in rocks of known
igneous origin.

In this latter respect they essentially differ from a greenstone
quarried in a field known as the Sanctuaries, a little west of the
town of St. Austell, where the central portions of the mass enclose
felspathic and other crystals, arranged in such a way as to indicate
the probable igneous origin of the rock. Although the central
portions of this dyke are crystalline, it gradually becomes slaty on
either side; and it is consequently difficult to trace a precise line of
demarcation between it and the enclosing killas*.

With regard to the blue elvans found between St. Erth and
St. Stephens, it may be remarked that their texture is frequently
coarser in the centre than towards the sides of the bands, and
that they eventually merge into the enclosing rocks, of which they
gradually assume the structure and cleavage. Their ultimate che-
mical composition agrees closely with that of the crystalline au-
gitic rocks before described ; but their structure is invariably slaty,
and thin sections do not exhibit any of the characteristics peculiar
to igneous rocks. It follows that if these were originally igneous
rocks other than ash-beds, they must have been subjected to the
action of influences by which the arrangement of their constituent
minerals has been completely changed and slaty cleavage produced.
They are met with only in the neighbourhood of the granite, or in
the vicinity of elvan courses; and when describing those in the dis-
trict around Camborne, Messrs. Von Oeynhausen and Von Dechen
state that they have not the appearance of regular dykes or veins
enclosed in clay-slatest. The greenstones occurring in the killas
surrounding the granite of Dartmoor are described by the same geolo-
gists as being of a highly schistose character, suggestive of the pro-
bability of their being highly metamorphosed slates.

Swamary.

The foregoing facts may be briefly summarized as follows.
The rocks of the so-called greenstone district of Penzance belong
principally to three distinct classes :—

* The writer some years ago examined this greenstone, and described it
(Phil. Mag. vol. xli. p. 97, 1871) as being probably a diorite which had urder-
gone extensive alteration. Since that time the quarry has been extensively
worked ; and sections recently prepared have been found to enclose small patches
of some augitic mineral. They alsu contain felspar (of which a portion is
triclinic), magnetite, green and brown microlites, a nebulous greenish-grey
substance, and apatite. It is therefore more probable that this is a highly
altered rock belonging to the class found near St. Michael’s Mount and in
various other localities in the neighbourhood of Penzance, St. Ives, and the
Gurnard’s Head.

+ Phil. Mag. vol. y. p.247. London, 1829.

N 2

178 J. A. PHILLIPS ON THE SO-CALLED

a. Gabbros or dolerites, in which the originally constituent
minerals are occasionally, to a great extent, unchanged, but are
sometimesalmost entirely represented by pseudomorphic forms.

b. ‘ Killas,” or ordinary clay-slates.

c. Highly basic hornblendic rocks, which exhibit a tendency to
break into thin plates; these, under the microscope, present
the characteristic appearance of metamorphic slates.

In addition to the foregoing, there are various slaty rocks of a
character intermediate between 6 and ¢, as in the village of Newlyn,
and on the western side of the valley north of Tolcarn.

It follows that these rocks have been more correctly described by
some of the earlier geologists than by the more recent observers, who
have regarded them either as being exclusively “‘ greenstones,” or as
consisting of metamorphosed slates only.

The rocks of the Cape-Cornwall district, coloured as greenstones
on the Geological Map, are principally hornblendic slates, which some-
times enclose veins or bands of garnet, magnetite, or axinite. Minute
garnets are not unfrequently disseminated throughout these slates.

The rocks in the neighbourhood of the Gurnard’s Head are almost
identical in character with those on the shores of Mount’s Bay; but
the slates are generally darker in colour, and are perhaps somewhat
more crystalline in structure, while the gabbros or dolerites have
sometimes been subjected to a still higher degree of metamorphism.

The crystalline pyroxenic rocks and metamorphic slates of the
St.-Ives district exactly resemble those in the neighbourhood of
Penzance.

The greenstones exposed between St. Erth and St. Stephens, un-
like the hornblendic and augitic rocks of the other districts, do not
occur in the immediate vicinity of granite; but elvan courses are
invariably found at no considerable distance from them. The struc-
ture of these rocks would indicate that they are probably altered
ash-beds or hardened hornblendic slates, frequently containing
pyrites and minute garnets, with perhaps axinite and some other
minerals. Neither augite nor diallage was detected in any of the
sections examined.

It is worthy of remark, that the chemical composition of these
slaty greenstones, or blue elvans, does not materially differ from
that of the gabbros or dolerites in the neighbourhood of Penzance,
the percentage of silica in the two series being as (will be observed)
nearly constant.

The hornblende slates contain about ten per cent. less silica than
the crystalline pyroxenic rocks; and there is an excess of iron
oxides present to nearly the same extent; in other respects their
composition is very similar.

_ Apatite is to be detected by the microscope in thin sections of the
augitic rocks only; but phosphoric anhydride is to a small extent
present in rocks belonging to each of the several classes.

The killas is a highly siliceous rock essentially differing from the
foregoing in chemical composition.

‘‘ GREENSTONES ” OF WESTERN CORNWALL. 179

Discussion.

Mr. Sorsy stated that he had come to nearly the same conclusions
with Mr. Phillips with respect to these rocks.

Dr. C. Le Neve Fosrer agreed with the author in considering
some of these so-called greenstones altered rocks, and not intrusive,
as had been supposed. He congratulated geologists on having such a
combination of chemical and microscopical analysis as that furnished
by Mr. Phillips.

Mr. Rurtey thought that many of the bands might be identical
with those occurring in Devonshire, especially near Tavistock. The
dipping away from the granite described by the author might be
due to the granite having been subsequently intruded. He thought
it probable that many of the beds referred to may be ash-beds, and
that it was quite possible the schistose character might be super-
induced in beds of ashes by the pressure of superincumbent rocks.
The mapping of Sir Henry De la Beche is remarkable for the accuracy
with which these boundary-lines are drawn; furthermore his deter-
minations of these rocks, as shown in his report, are either absolutely
correct or close approximations to the truth, although the small scale
of the map does not permit this to be shown.

Mr. Koc remarked that, in considering the nature and origin of
such rocks as these, we ought to notice what rocks overlie them,
and to take into account the probable effects of water percolating
through such overlying beds and containing much carbonic acid,
by deposition of silica and removal of carbonate of lime. The
analysis of the killas given by the author was almost that of altered
trachyte. He thought that in such investigations the microscope
must be our principal guide, but aided by the consideration of the
relative positions of the rocks.

Prof. Ramsay inquired what was the definition of a volcanic rock.
He thought that volcanic rocks were either lavas or ashes, and that
it was dangerous to affirm that the blue elvans were ashes in so
metamorphic a district.

Prof. Masxrtyne, after thanking the author for his paper,
inquired as to the meaning in which he used the terms “ gabbro”
and “igneous rock.” He also remarked that it was strange to find
apatite the only mineral that was unaltered.

Mr. Drew inquired as to the possible future colouring of the
Survey Map of this district. It appeared from the author’s paper
that the parts coloured as “greenstone” in the present maps
included various rocks which ought to be subdivided.

The AvrHor, in his reply, stated that igneous rocks have a certain
amount of cleavage, which is often rendered very distinct by
weathering. The crystals of hornblende are very distinct in the
rocks described. Gabbro is a felspathic rock with diallage, sma-
ragdite, or hypersthene instead of augite: but with regard to some
of them it may be doubtful whether they are gabbros or dolerites.

180 J. D. KENDALL ON HEMATITE IN THE SILURIANS.

20. Hamatize in the Sttuntans. By J. D. Kenpatt, Esq., F.G.S.
(Read March 8, 1876.)

In a paper read last session before the Society, I pointed out some
of the most important features of the Hzmatite deposits that occur
in the Carboniferous Limestone of Cumberland and Lancashire. |
Amongst other things I showed that the direction of the deposits
was invariably parallel to that of the meridional divisional planes’
by which the rocks are intersected ; and as an explanation of that
fact 1 suggested that the hematite had probably been deposited in
caverns which had been hollowed out by chemical action along the
meridional planes, or that the hematite had replaced the Limestone
along those lines.

The paper also briefly referred to the deposits of hzmatite in the
Silurians of the same districts, and went to show that these deposits
(which are in the form of veins), like those in the Carboniferous
Limestone, were in direction paraliel to the divisional planes. There
was, however, this difference between these Silurian deposits and
those in the Carboniferous Limestone, that whereas all the deposits
in the latter were parallel to one another, and had a direction nearly
north and south (that is the same as the meridional divisional
planes), those in the Silurian had two directions, some being parallel
to one set of divisional planes and some to the other.

Since the delivery of that paper I have had an opportunity of
examining a Silurian deposit which appears to me to be altogether
unlike (as regards the relation between the direction of the deposit
and the divisional planes) either those deposits to which I before
alluded in the same series of rocks, or those in the Carboniferous
Limestone. I therefore deem it may be interesting to the Society
if I give an account of my observations on that deposit.

The deposit to which I refer is in the parish of Millom and
county of Cumberland, and is now being worked by the Water-
Blean Mining Co., their works being known as the Water-Blean
Mines.

The geological structure of the district in which the deposit
occurs is somewhat as shown in the following plan and section
(figs. 1 and 2).

The Silurians are all conformable, and strike about 65° N.E. and
S.W., with a dip of about 80° to the N.W. ‘Their order is thus.
inverted, the Limestone appearing to dip in below the Ash-beds,
and the Flags below the Limestone, whereas in reality the Ash-
beds underlie the Limestone, which in its turn forms the base of the
Flags.

The Limestone is part of that narrow band which stretches through
the country in a nearly straight line from Millom to Shap, a distance
of about thirty miles. It is in this rock that the haematite occurs
to which I wish to call attention. The form of the deposit (or,

J. D, KENDALL ON HEMATITE IN THE SILURIANS. 181

Fig. 1. Plan of arrangement of Hematite deposits, Millom,
Cumberland.

)

3
o

| py

Kiihii. Vy \|_ Ns =
eae emery et ee ea

cer: oo
2 2 ZZ

NR | mM
|

|

A. Carboniferous Limestone.

B. Coniston Flags ; ; P .

C. Coniston Limestone : : : : |storian.
D. Ash-beds (Green Slates and Porphyries)

— Hematite Veins.

a, >. Line of Section, fig. 2.

Fig. 2. Vertical Section across Hematite deposits, Millom, Cumber-

Bae is

For explanation se e fig. 1.

rather, deposits ; for there are several of them) is that of short veins,
which vary in width from a few inches to 9 feet. The ore is very
hard, and has a high red stain, and where it joins the Limestone
appears in most cases to be grown to that rock, as the ore in the
Carboniferous Limestone is with it. The direction of the veins is
about 65° N.E. and 8.W., and they dip at an angle of about 80° to
the N.W.~ The following vertical cross section (fig. 3) will partly
show their mode of occurrence. ©

182 J.D. KENDALL ON HEMATITE IN THE SILURIANS.

Fig. 3. Vertical Section of Hematite vein.

——— =
=

Sic

\ Ny

A. Glacial drift. |B. Limestone. C. Hematite. d,d,d. Bed-joints.

The direction and dip of the veins, it will be observed, are exactly
the same as the direction and dip of the bed-joints of the Lime-
stone, the strike of both being about 65° N.E. and §.W., and their
dip about 80° tothe N.W. This is the point which I wish chiefly
to place before the Society. In the paper before referred to, as
already stated, I pointed out that the hematite deposits in the Car-
boniferous Limestone of Whitehaven and Furness invariably coin-
cide in direction with the meridional divisional planes intersecting
that rock, and that the deposits in the Silurian slates at Knock-
murton, near Whitehaven, have two directions, which coincide with
either one or the other of the divisional planes of those rocks. The
Water-Blean deposit, however, appears to me to be different from
either of these, the deposition of the ore having taken place along the
bed-joints of the rock. In some of the Whitehaven deposits, as at
Parkside and Bigrigg Moor, a phenomenon ‘somewhat similar to
this is to be seen, as shown in fig. 4, which is a vertical section of
the Parkside deposit; but there the strata lie at such a low angle
that the ore-deposit is more like a bed than a vein. Their likeness
to the Water-Blean deposits, however, is only partial; for, notwith-
standing their bed-like appearance, and the fact of their being
parallel to the bed-joints of the rocks, the deposits nevertheless
have their longest axis parallel to the meridional divisional planes; .
and the strike of the rocks does not in any way determine the direction
of the deposits, as it does at Water Blean.

The direction of the Water-Blean deposit is exactly the same as
the strike of the rocks in which it ocew's, and altogether different from
the directions of either of the divisional planes. f

J. D. KENDALL ON H-ZMATITE IN THE SILURIANS. 183

Fig, 4. Vertical section of Hematite vein in Carboniferous Limestone
at Purkside.

A. Drift. C. Hematite.
B. Millstone Grit. D. Carboniferous Limestone.

A question which here naturally suggests itself is, Why was not
the ore deposited along the meridional divisional planes of the Lime-
stone at Water Blean, in the same way that it is in the Carboniferous
Limestone of Furness and Whitehaven? To answer that question
it appears to me to be necessary to state that the bed-joints in the
Limestone at Water Blean are much more persistent than the divi-
sional planes are, the latter being very irregular in direction, and
not by any means so strong and open as the bed-joints. We should
_ therefore naturally expect the carbonated water to act most power-
fully on the bed-joints, as it, in my opinion, for the same reason, has
done on the meridional, in preference: to the transverse divisional
planes in the Carboniferous Limestone of the district of Whitehaven
and Furness.

Discussion.

Mr. Forsus considered that this paper was equivalent to a with-
drawal on the part of the author of the opinions expressed in his
former paper. Mr. Forbes maintained that the direction of the
vein-like deposits is due to their being formed in preexisting fissures,
into which the hematite has been injected. When hematite is
found in caverns it has been washed in by water.

Prof. Hueurs thought that the only rule as to the mode of oc-
currence of these deposits was dependent on the direction of the
joints and bedding, pointing out that in the Mountain Limestone it
was determined chiefly by the joints, while in the highly inclined
Coniston Limestone the ore was found along the Limestone bands,
which were separated by impervious shales.

Mr. Warrneton W. Suytu remarked that this is a problem in the
investigation of which great caution is necessary. In some localities
undoubtedly the veins of hematite have a general direction ; and in
many cases there is a considerable coincidence of the veins of
hematite with joints.

184 J. AITKEN ON THE DISTRIBUTION OF DRIFT

21. OssERVATIONS on the UnEauaL Distrisution of Drirr on op-
POSITE SIDES of the PENNINE CHAIN, in the country about the souRCE
of the RIVER CALDER, with Suggestions as to the causns which led
to that RESULT, together with some Notices on the HigH - LEVEL
Drirt in the UPPER PART of the VautEY of the RIVER IRWELL.
By Joun ArrxeEn, Esq., F.G.S. (Read June 23, 1875.)

[ Abridged. ]

For many years past the attention of geologists has been more or
less directed to the fact that a marked difference exists in the distri-
bution of drift on opposite sides of the Pennine chain, a difference
amounting in some instances to an entire absence of that material on
the easterly slopes for many miles from the watershed of the country,
and over a considerable portion of its length, notwithstanding that
thése deposits overspread the great plains of Lancashire and Cheshire
in great force, and are found mounting up upon the flanks of those
hills on their western sides to very considerable elevations, ap-
proaching closely in some cases to the culminating ridge, and in
others, where the chain is crossed by intersecting valleys, to some
hundreds of feet in excess of the summit-level of these gorges. It
would further appear from facts hereafter to be adduced, that
although some of these cross valleys attain to only very moderate
altitudes, no communication existed during the glacial period of
such a character as to permit of the passage of a body of land ice
from one side of the chain to the other. Whilst, however, these
phenomena have not wholly escaped the notice of those observers
who have, more particularly of late years, directed their attention to
the surface accumulations of the northern counties of England,
allusion having been made to the subject by Messrs. Binney, 'Tidde-
man, Green, Foster, Dakyns, Goodchild, and others, yet no serious
attempt has, I believe, so far been made to grapple fully with the
subject by any of those who have hitherto given attention to it.

The subject has long perplexed me, amongst other observers; and
it is only after a lengthened consideration that I have ventured to
suggest a theory which, whilst offering an explanation of the pheno-
mena, does no violence to any of the well-established principles of
physical or geological science. My object, then, in presenting the
present communication is to show that the whole of these pheno-
mena are explicable on the supposition that, during the flow of the
great ice-sheet over this region, these intersecting channels were
sealed and blocked up so effectually as to completely cut off all com-
munication between the eastern and western sides of the chain, and
that the only agency fully meeting the requirements of this supposi-
tion is that of ice or snow so consolidated and fixed in the sinuous
channels as to remain stationary and inert while the great mass
of glacial ice, in two sheets separated by the more elevated portions

ON OPPOSITE SIDES OF THE PENNINE CHAIN. 185

of the great anticlinal range, traversed its eastern and western sides
on its march from north to south.

That vast masses of drift deposits overspread the wide plains of
Western Lancashire and adjoining counties, extending from the
Irish Sea in one unbroken sheet to the slopes of the Pennine Hills,
and thence mounting up to higher levels, and penetrating almost
every upland valley and mountain-gorge to considerable elevations,
is too well known and too generally acknowledged to require any
amplification from me on the present occasion.

A great number of facts, and the statements of various autho-
rities as to the altitude of high-level drift on the western slopes
of the Pennine chain, indicate a general elevation of from 1100
to 1200 feet, whilst scattered boulders and pebbles of travelled rocks
may be detected at various places in the locality some 200 feet in
excess of that; and I contend that these more elevated stray boulders
and pebbles owed their origin to icebergs during the interglacial
period, and not to the period of land ice.

I wish particularly to direct attention to the circumstance of drift
having been found at an elevation of 930 feet above sea-level, on
the shoulders of the hills on both sides of the southerly termination
of the Walsden defile at the summit near Littleborough, whilst
all traces of them vanish in a line almost coincident with the
watershed of the valley, which has an elevation of 627 fect, beyond
which no accumulations of this character are met with in travelling
east for a distance of about 15 miles. The first indication of their
presence as a regular bed of drift occurs in the valley of the Calder at
North Dean, near Halifax, where, in sinking for the foundations of
a railway viaduct, the following section was observed by Mr. James
Spencer, of Halifax, to whom I am indebted for the accompanying
particulars :—

ft.
HE ee ee faecal bearer eo resets Undue rou Mtectt: 1
Peg UL Ue Unc be Mee EP Cae ee Ce RS ey ee a 8
Gravel derived from local rocks........0....--ceccecsscssceeeses 2

It is, however, stated that an occasional pebble foreign to the
district has been noticed in the bed of the river Calder, a little
further to the west than the spot just named*.

Close to the railway bridge at Elland, in the bank on the south
side of the river Calder, under about 5 feet of sandy loam, a bed of
gravel exists, about 4 feet of which is exposed, and which I am
informed by Mr. Davis, F.G.S., of Greetland, has been proved by a
well-sinking to be from 15 to 20 feet in thickness, in which red

* Dr. Alexander, in describing the geology of the parish of Halifax, in the
‘Tracts’ of the Geological and Polytechnic Society of the West Riding of
Yorkshire, 3rd June, 1841. p. 201, mentions that “in prosecuting the works on
the line of railway some blocks of granite had been found near Hebden Bridge ;”
and Mr. Binney also informs me that he has seen foreign pebbles at Hebden
Bridge.

186 J. AITKEN ON THE DISTRIBUTION OF DRIFT

granite, silurian porphyritic greenstone, and other foreign pebbles
occur ; and at and below this point examples of these rocks may be
detected in the bed of the river wherever gravel is exposed. ‘This is
situated at an elevation of about 200 feet above the sea level.

At the Elland railway-station, the following section has been
exposed by some recent operations of the Company :—

ft. in
ROHS Sasnce apis octane Wallace ok ekia ie hae ac 1 4
Fine subangular gravelly surface wash ..........0...... 2 6
TOnmy Sand? 277 ee cei . isaind= ob hlewtsleateas 6 inches to 1 O
Bimo, sandy; paved, . a... cc 5-0 diseneadtaas ses chines sey eeneerat 2 6

Gravel coarser than that above; stones all waterworn
and much abraded, of local origin, resting upon a
bed ‘of Carboniterbus shale ys 2.2... oueksesancccesenbetee 50

At the cemetery, a few hundred yards to the north-west of the
last-named place, and at an elevation of about 300 feet, a bed of
gravel, sand, and loam, varying in thickness from 5 to 10 feet, ana-
logous in character to the section last described, reposes upon the
Carboniferous shale. And in the Shibden valley, deposits precisely
similar to these are found at about a corresponding elevation.

In the three last-named instances the stones composing the gravel
consist of well-rounded waterworn rocks, derived exclusively from
the basin of the Calder and its tributaries, not a foreign pebble being
found incorporated with them. This gravel may therefore be re-
garded as having a purely local origin, and probably dates from the
time when the river ran at this level, and is consequently in no
way connected with the glacial drift which occurs in the valley
below.

Precisely analogous conditions to those already described as oc-
- curring in the Walsden defile exist in the Cliviger valley, the drift
running up on the sides of the hills near the entrance of the gorge
to a height of 1075 feet, its southerly termination taking place both
on the sides of the hills and in the valley in a line almost cor-
responding with the summit-level of the latter at Calder Head, which
has an elevation of 768 feet.

It appears from a paper* read some time ago by Mr. R. H. Tid-
deman, of the Geological Survey, that a very similar state of things
occurs with regard to the distribution of drift in the great water-
shed of the north, opening between the basins of the Ribble and
the Aire, which has a summit-level of 700 feet. No ice-scratches
are shown on the excellent map which accompanies that paper as
existing east of the watershed of the Pennine range in that part of
the country ; and those which do occur near to that line run, with
one exception, in’a direction closely approaching to north and
south.

* “On the Evidence for the Ice-sheet in North Lancashire and adjacent

parts of Yorkshire and Westmoreland,” Quart. Journ. Geol. Soc. for Noy. 1872,
p. 478, a contribution the value of which it is difficult to overestimate.

ON OPPOSITE SIDES OF THE PENNINE CHAIN. 187

In the southerly portion of the chain the behaviour of the drift in
the neighbourhood of Castleton (a town situated just beyond the
central axis, on the easterly slope of the range) entirely corresponds
with the instances already adduced; and this is confirmed by the
following quotation from a paper by Prof. W. B. Dawkins * :—
** From what I can see, there is nothing whatever in this deposit
to separate it from the superficial deposits in the valley near Cas-
tleton, which you will find pretty well everywhere. When I ex-
amined these the other day in the company of Mr. Prestwich and
Mr. Tiddeman, we came to the conclusion that they are not glacial,
but the result of subaerial wear and tear of the rocks in the neigh-
bourhood.” ‘These examples will, I trust, suffice to show that over
this large area, extending from Skipton in the north, by Tod-
morden, to Castleton in the south, a general absence of drift de-
posits and of all other indications of glacial action characterizes the
eastern slope of the great axial chain in this division of the country,
thus establishing the first part of the proposition with which I set
out.

We may now inquire if there is any legitimate reason for sup-
posing that drift was at one time deposited over the now barren
area, and that its absence at the present time is to be accounted for
by subsequent denudation, as suggested by Mr. Tiddeman in the
paper previously referred to. It appears to me, from a full con-
sideration of all the facts, exceedingly problematical that this can
have been the case ; for it is difficult, if not impossible, to conceive of
any known force which could act in so capricious and anomalous a
manner as to have swept and cleared off the surface of the driftless
area to the east so completely that not a trace of its former existence
should remain, even in the deep sheltered valleys which intersect the
district in all directions and thus present conditions of the most
favourable character for its preservation, and yet at the same time
leave intact the vast bodies of drift which so completely envelop
the country to the west.

The drift found in the lower reaches of the eastern valleys was
probably derived from the section of the great ice sheet which
traversed that side of the country; and its presence there is no
indication of its having been carried through the cross valleys from
the west.

Having so far endeavoured to prove, and, I hope, successfully, that
a marked difference exists in the distribution of drift on the two
sides of the Pennine chain, it now remains to show that this pheno-
menon has in all probability been induced by the valleys in question —
having been blocked up during the glacial period by accumulations
of ice or snow held firmly in their places by the physical and other
characters of the gorges themselves, acting in combination with the
force exerted by glaciers pressing against the ends of these blocked-
up channels, thus effectually holding stationary Le imbedded masses
in the places of their deposition.

* Trans.’Manchester Geological Society, vol. xiii. p. 124.

188 J. AITKEN ON THE DISTRIBUTION OF DRIFT

For a full and clear comprehension of this subject, it will he
necessary to glance at the physical and geographical features of the
district and to trace their relation and bearing to the surrounding
country. In treating of this branch of the subject, our attention
will be principally directed to that part lying between Burnley and
Littleborough on one side, and Todmorden on the other—a district
embracing the deep valleys of Cliviger and Walsden, in which are
contained the two lowest watersheds in the whole of this central
range of hills.

If, then, we take Todmorden as the central point of the district,
from which to commence our observations, we shall find that that town
is situated almost on a line with the highest summits of the Pennine
Hills, in a deep gorge at the junction of the two streams which flow
through the two defiles already mentioned, and on the main axis of
that series of dislocations known as the great Pennine fault, which run
in a line roughly coincident with the Walsden and Cliviger valleys,
and the effect of which has been, not only to fracture and break up
the rocks to a marvellous extent, but also to cause a reversal in the
dip of the strata, those on the east dipping at an angle of about
5° to the S.E., and those to the west declining at an angle of from
20° to 45° to the W.S.W.

Starting from this centre, three valleys radiate in different direc-
tions, in lines almost equidistant from each other :—one stretching
off through Walsden to Littleborough for a space of about 5 miles in
a direction a little W. of S.; another taking a N.W. course through
the sinuous gorge of Cliviger to Burnley, a distance of about 8 miles ;
whilst the third, the valley of the Calder, takes an H.N.E. direction.

The appearance thus presented, when traced on a geological map,
reminds one forcibly of the device which characterizes the Manx
copper coinage. By an examination of these arms, we shall find
that they are three deep, narrow, rocky, tortuous channels, hemmed
in on every side by massive walls of rock, cut into at intervals
along their course by the entrance of tributary streams.

The watersheds of the Walsden and Cliviger valleys are at a com-
paratively low level, that of the first having an altitude of 627 feet
above sea-level, and being eaten into to a depth of 330 feet below
the shoulder, or first platform bounding the valley*; whilst that of
the second is crossed at an elevation of 768 feet, and is excavated to
a depth below the immediate country of 475 feet, a depression which
would be greatly increased if the summit-level of the surrounding
hills were taken as the datum-line for measurement.

The valley of the Calder, which forms the line of drainage through
which the waters of the aforesaid tributary valleys, from their
junction at Todmorden, pass off to the German Ocean, is cut in to
a depth considerably in excess of those already described.

If, then, we view these valleys in relation to the initial period of
their formation, structure, and physical characters, the total absence
of drift or other indication of ice-action within their boundaries, or
the marked contrast presented in the configuration of these and other

* Prof. Hull, Geol. Mag. vol. iii. p. 475.

ON OPPOSITE SIDES OF THE PENNINE CHAIN. 189

valleys in the neighbourhood, and the general surface-contour of the
surrounding district as compared with the country on the western
slope of the chain immediately beyond the watershed, where ice is
known to have been, and where the hills and dales lose their stern
angular character, and become characterized by a soft, rounded, and
flowing appearance, indicating unmistakably the operation of diver-
gent forces, and the existence of wholly different conditions, we shall
have no difficulty in concluding that glacial or land ice has had no
share in their formation or subsequent modification, but that they
owe their origin in the first instance to the fracturing of the rocks,
and subsequently to the operation of subaerial forces.

Having now pointed out the physical features of the district, I
shall proceed to offer some reasons for concluding that during the
glacial period these lines of communication were blocked up by snow
or ice so as to cut off all connexion between the two sides of this
chain of hills. On the approach of the long winter which preceded
the period when ice overspread the county of Lancaster and the
whole of the northern part of our isiand, the severity of the climate
would gradually increase ; and during this time large accumulations
of snow would take place, and ice be formed in deep and sheltered
situations lke those presented by the Walsden, Cliviger, and Tod-
morden valleys. The snow, thus accumulated to a depth of from
800 to 1000 feet, would, by being partially melted and recongealed,
become in time a consolidated compact mass, little if at all inferior
to that of ice itself in density and consistency.

Accepting, then, this proposition, we shall have no difficulty in con-
cluding that these bodies, held firmly in their places in the tortuous
serrated valleys which are so constructed, and holding such relations
to each other that any attempt to move on the part of the accumu-
lations in one of the arms would be resisted and counterbalanced
by the opposing force of the other two branches, each of which
would have a tendency to move in an opposite direction, would thus
be able to offer an effectual resistance to the force of the great
northern glacier, on its approach so far south, to dislodge them from
their strongly intrenched positions,—the resistance to all motion on
their part being still further increased by the presence of the gla-
ciers acting upon the terminal portions of the two arms at the
mouths of the valleys by which those bodies, if in motion, would
have to emerge, with a force equal to that exerted at the northerly
end of the Cliviger gorge, where the glacier would seek to effect an
entrance therein, thus completely neutralizing each other, and pre-
venting any motion on their part taking place. The progress of the
glacier in its journey south being thus arrested by an impassable
barrier, at or near Holmes Chapel, in the Cliviger gorge, where the
valley suddenly contracts and becomes hemmed in on both hands
by massive beds of grit rocks, in some places almost vertical, it
is evident that the ice forcing its way so far would be compelled
either to move on over the imbedded and stationary mass, or be de-
flected from its course, and compelled to take a more westerly direc-
tion, part of it finding its way over the pass separating the Easden

190 J. AITKEN ON THE DRIFT OF THE PENNINE CHAIN.

Clough from the Whitewell valley at an elevation of 1170 feet, and
passing on east of Deerplay Hill and the high land adjoining into
the valley of the Irwell, thence skirting the hilly district east of
Bacup and crossing Brandwood Moor, the. dividing line between
the basin of the Roach and Irwell, and so continuing its course
on to Manchester and the great Cheshire plain; whilst another
stream of ice would doubtless cross from Burnley into the Whitewell
valley by the Wholaw-Nook pass, becoming there confluent with the
arm already described, the main body moving off in the direction
of Burnley and Accrington, on the westerly side of Hambledon Hill,
thence pursuing its course south and uniting with the other streams,
finally debouching onto the plains of South Lancashire.

In support of the theory that a large body of moving ice meeting
with resistance sufficiently powerful can be either entirely arrested,
or its various portions be endowed with varying and diverse motions,
we have the authority of the Duke of Argyll, Prof. Ramsay, and
Mr. Goodchild, some of whose statements fully justify the conclusion
that such was the behaviour of the ice in this instance.

In all probability ice did not overflow the higher hills in this part
of the axial range, as is manifested by the absence of all trace of
its action upon the surface-contour of the country, and by the fact
that there are no remains of morainic or other matter to indicate
that it had ever been in those situations.

It may be assumed that the great ice-sheet in this part of Lan-
cashire did not rise to so great an elevation by some hundreds of
feet as it is proved to have done further to the north by the obser-
vations of Mr. Tiddeman, Mr. J. Clifton Ward, and Mr. Goodchild ;
and I contend that satisfactory reasons may be given for this vari-
ation by taking into consideration the gradual slope of the ground,
the waste by melting to which the ice would be subjected in tra-
velling south, and its tendency to spread out and become diffused
on reaching the flatter and more level plains of Lancashire and
Cheshire.

Statements contained in the memoir by A. H. Green, C. Le Neve
Foster, and J. R. Dakyns on the Geology of the Carboniferous Lime-
stone, Yoredale Rocks, and Millstone Grit of North Derbyshire and
the adjoining parts of Yorkshire, confirm many of these conclusions
arrived at by me on independent grounds.

I have thus endeavoured to lay before the reader such facts as
have come within my reach tending to illustrate the question forming
the subject of this essay, with what success he will be able to judge.
If, however, the facts and arguments should fail to carry conviction,
I trust that the effort I have made to solve this somewhat difficult
and perplexing question may not prove altogether fruitless; for
should no other result follow than that of attracting the attention
of other observers to the subject, so that eventually a satisfactory
solution may be arrived at, my object will, at all events to some ex-
tent, have been attained.

ON THE GLACIAL GEOLOGY OF EAST ANGLIA. 191

22. Norms on the Puystcan Guotoey of East Anenia during the
GuactaL Prrtop. By W. H. Pennine, Esq., F.G.S. (Read

December 15, 1875.)
[Puate XV. ].

INTRODUCTION.

In submitting to the consideration of the Society the following ob-
servations, I would remark that they are intended to form, not a
full description, but a sketch of the physical geology of East Anglia
during the glacial period. The evidence upon which they are
founded is derived from deposits that either have been, or will be,
mapped by the officers of the Geological Survey. Full descriptions
will appear in their published maps and memoirs; therefore the
subject will here be treated generally, and all details omitted.

The object I have more especially in view, is to offer an explana-
tion of the origin of a somewhat puzzling series of gravels and
sands, classed by Mr. 8. V. Wood, Jun., as ‘* Middle Glacial,” to give
a reason for their occurrence in certain areas and non-occurrence in
others, and to account, on mechanical grounds alone, for the almost
total absence therefrom of any (except derived) fossil remains. I
wish also to remark briefly on the probable means of formation of
the so-called ‘** Denudation gravels.”

The subject is divided, for reasons that will appear, into two
parts; the first relates to the district south and east, and the
second to that north and west, of the great Chalk escarpment.

By the term “ East Anglia” is meant a tract of country, north of
the Thames, which may he considered to be bounded on the west by
a line passing from London through Hertford, Royston, and St. Ives
to the estuary of the ‘“‘ Wash,” and which includes all Norfolk,
Suffolk, and Essex, with part of Cambridgeshire, Hertfordshire,
aud Middlesex. The general contours of the southern part of the
district at every hundred feet above the sea are shown by the dot-
ted lines on the map (fig. 1, Pl. XV.), the smaller and later-formed
features being omitted as immaterial to the argument.

The lowest tract in the district may conveniently be termed the
‘*‘ Cambridge Valley :” a large portion of it is fen-land bounded on
the south and east by Gault, which is, with trifling exceptions, the
lowest geological formation that comes to the surface in the area in
question. On the right side of the Cambridge valley runs the chalk
escarpment, divided as usual into two main lines, one of the Lower
and one of the Upper Chalk. The latter forms some of the highest
ground in the district *, and may be traced as an uneven ridge,
with rounded hills and hollows, from Buntingford, by Saffron
Walden, Haverhill, Thetford and Swaffham, to the sea. The ground
slopes gradually away from this ridge towards the London Basin,
where the Tertiaries set in, one of the series (the London Clay)

* At Tharfield (S. of Royston) 550 feet, and in West Norfolk 450 to 650 feet
above the sea.

Q: J..Go8;, No. 126. 0

192 W. H. PENNING ON THE PHYSICAL GEOLOGY OF

forming an escarpment roughly parallel to that of the Chalk. The
northern and eastern portions of the district are in part occupied
by deposits of later Tertiary age—the Crags, the Chillesford beds,
and the Forest-beds.

The general altitude of the line of escarpment of the Upper Chalk
may be taken as from 300 to 500 feet, that of the London Clay
from 200 to 300 feet above the sea.

Part 1.—The Drift-deposits to the South and East of the Chalk

escarpment.

Plocene.—Along the coast of Norfolk, Suffolk, and part of Essex,
as well as for some distance inland, occur the ‘‘ Crag” deposits. Of
these sands, the older or *‘ Coralline Crag” contains the remains of
a deep-sea fauna; in the “ Red Crag,” immediately succeeding,
littoral forms prevail; while the newer or “ Norwich Crag” also
contains littoral forms with land and freshwater shells in addition,
A gradual rising of the coast during the Crag period is thus indi-
cated ; and this continued until the land stood slightly higher than
at present, admitting the growth of the Cromer ‘‘ Forest-bed.”

These Crag-beds perhaps scarcely belong to the drift-deposits that
present the evidence on which the following arguments are founded,
although I believe that there is a gradual passage up from them to
those of recent date. This gradual passage will probably be found
to exist stratigraphically, as well as in regard to their enclosed fos-
sil remains and to the climate* that prevailed during the different
periods of their deposition.

Lower Glacial.—After the formation of the Cromer Forest-bed
the land again sank beneath the water, gradually and to a depth
of certainly not less than 400 or’500 feet +: during the progress of
the submergence the advancing shore-line gave rise to the *‘ pebbly ©
sands” which, according to Messrs. Wood and Harmer, ‘‘form the
base of the whole glacial series, indicate shore-conditions and the first
setting-in of the great glacial subsidence” +. Arctic conditions of
climate began to prevail; and here and there patches of clay were
dropped by icebergs, heralds of those to follow in such vast numbers
during the Upper Glacial epoch. The contorted drift, which occupies
a large area in the north-east portion of the district, was deposited
during the early part of this submergence of the land ; its mode of
formation I consider to have been much the same as that of the
Upper Boulder-clay presently to be considered. The contortions
sometimes exhibited by this deposit are probably due to agencies
acting on the clay at a subsequent period.

Middle Glacial—The Lower Glacial beds are overlain and con-
siderably overlapped by a series of gravels and sands called the
‘“‘ Middle Glacial.” These beds occupy, or have occupied, almost all
the area covered by the Lower Glacial, and extend far beyond it in a
southerly direction. It will be seen on reference to the map (fig. 1,

* Lyell, ‘Student’s Elements,’ p. 177.

t Lyell, ‘Student’s Elements,’ p. 167.
t ‘Outline of the Geology of the Upper Tertiaries of East Anglia,’ p. 16. ©

EAST ANGLIA DURING THE GLACIAL PERIOD. 193

Pl. XV.), that many of the valleys running 8. and E. from the chalk
escarpment to the sea are, almost to their source, cut through the
Upper Boulder-clay, and expose these gravels beneath it. They
are thus proved to be very persistent over a definite area, within
which it is a rare occurrence for the clay to be seen resting on the
older geological formations. It is probable that the gravels extend
beneath the clay over nearly all the intervening area, in greater
force along the lines of the larger valleys (which I believe to have
been carved into nearly their present form in preglacial times),
and in a more attenuated, or even patchy condition under the higher
grounds.

But the gravels can in no instance of which I am aware be
traced up to the escarpment of the Chalk, or, in other words, beyond
a certain definite level. It is not that they disappear beneath a
great thickness of Boulder-clay to reappear at its opposite boundary;
on the contrary, it is evident that they gradually thin out; and a
few miles before the escarpment is reached we find the Boulder-clay
overlapping them and resting directly on the Chalk (fig. 2, Pl. XV.).
This is well seen in many sections, also along the N.W. face of the
scarp, where the junction of the Chalk and the clay forms for many
miles a well-marked line, the absence of any intervening sand or
gravel being constant and remarkable. The fact of the gravels not
running up to the escarpment was noticed, but no inference drawn
therefrom, in 1835, by Mr. Caleb Burrell Rose, who writes * :—
** The general surface of the chalk must have suffered prodigious
abrasions from the violence of the elements, as evidenced by the
immense quantity of gravel formed and collected in various
situations, as well as by the different altitudes at which the chalk
is found, it appearing immediately beneath the vegetable soil, even
on the highest ground; and at a level of not less than 50 feet lower
it may be found covered by more than 150 feet of sand and clay
containing boulders.”

In the thinning-out of the Middle Glacial beds against the Chalk,
and their not rising beyond a certain height, we have a clue to
the conditions under which they were formed. The great chalk
escarpment, as a long and narrow ridge, standing at this time well
above the sea, was an important feature in the physical geography
of the period; and the Middle Glacial sands and gravels, as such,
are entirely owing to its existence. For it formed a barrier op-
posing itself to the strong current which must at that time have been
sweeping round from the North Sea to the Atlantic Ocean. By
this current were brought down the materials of which the gravels
are composed, and which consist of pebbles derived from the rocks
of the northern and eastern coast along which it travelled, mixed
with a large percentage of flints from the chalk barrier itself.

In the gravels are intercalated occasional masses of Boulder-ciay
which were brought down by icebergs, and which, heavily descend-
ing from them, have distorted the gravels wherein they now lie
without any approach to order or arrangement. Further evidence

* “ Geology of West Norfolk,” Phil. Mag. vols. vii. and ~
(0) e

194 W. H. PENNING ON THE PHYSICAL GEOLOGY OF

of icebergs during this period is found in the associated series, or
parts of series, of fossil remains taken from the gravels—as, for
instance, at Bishop Stortford, where a good number of vertebre
and other bones of Pliosaurus, parts of the same animal, were
discovered. These of course were derived from some of the Second-
ary rocks to the northward, and must have been transported in a
large mass of the matrix in which they had been previously
fossilized.

The submergence, which began at the commencement of the Lower
Glacial series went on until the waters af the North Sea were again
united to those of the Atlantic *, when a strong current was set up
in that direction. It steadily continued during the accumulation
of the Middle Glacial beds, bringing them up to the level at which
they are now found. But while yet the higher ranges of the Chalk
were above the water, the lower parts of the escarpment had gone
down sufficiently to admit of the passage over them of the sea; -and
this of course began to give rise to a set of very different physical
conditions. In place of the strong southerly current confined to
the eastern side of the chalk ridge, the waters having admission to
a much larger area, their power (within that area) of transporting
gravels was materially lessened, and, as submergence went on, abso-
lutely lost. For this reason the gravels are found running not
quite although nearly up to the lower levels of the chalk escarp-
ment.

[The ground to the N. and W. of the escarpment (fig. 2, Pl. XV.)
being at even a lower level than where the gravels occur, was of
course also under water; it was, however, land-locked on every side
but one, and formed a bay in which no current of any extent was
possible ; consequently we find in it but few (if any) deposits that
undoubtedly belong to the Middle Glacial series. The waters from
the bay flowed outward through the ‘‘ Wash” to join the southerly
current, and contributed their share of pebbles to the gravel that it
deposited. |

It is almost impossible to locally classify in detail these drifts;
but, speaking generally, the coarser gravels are found, as might
naturally be expected, in positions nearest to the chalk ridge, as in
the neighbourhood of Clare, the finer gravels in the intermediate
country between it and the most distant deposits at Hertford, where
they take the form of brick-earth. Throughout are interspersed
masses of current-bedded sand and occasional patches of iceborne
clay.

The general absence from these beds of fossil remains may be ac-
counted for by the conditions of deposit ; they were formed in a
strong current, which would be (except, perhaps, at the commence-
ment and end of the period) unfavourable to the existence of animal
life, while the remains of any plants or animals that may have
come within its sweep would speedily be reduced by attrition,
except of course those which were previously fossilized, and all of

* “England was again joined to the continent during the time that the
vegetation of the ‘ Forest-bed’ flourished ” (Ramsay, ‘ Physical Geology,’ p. 178).

EAST ANGLIA DURING THE GLACIAL PERIOD. 195

which are much rolled and waterworn. In the exceptional cases
where shells do occur, they are found to be of such species as would
perhaps indicate a climate warmer than that which prevailed before
and after the Middle Glacial period. ‘This fact is probably due not
so much to a real difference of climate as to the temperature of the
current, which would be governed by its direction and the condi-
tions affecting it before reaching these islands.

If these views be correct, there is between the Lower and Middle
Glacial no definite line of demarcation.

Upper Glacial_—The Great Chalky Boulder-clay in East Anglia
extends in mass, or in patches that once formed part of the main
mass, over the northern half of the London Basin, the Pliocene area,
and the dip slope of the Chalk, rises over the escarpment of the same
formation, and plunges down into the Cambridge valley. It now
caps the highest hills; and it occupies the deepest valleys, except
where it has been removed by recent denudation.

The submergence which began with the Lower Glacial and con-
tinued during the Middle Glacial periods still proceeded; conse-
quently the strong northern current was, as we have seen, gradually
replaced by a more open sea. During the succeeding era the bot-
tom of this sea became covered with a thick deposit of ice-trans-
ported clay. This clay, made up of chalk and the débris of other
rocks also found to the northward of the area, having been brought
down and dropped in masses, presents within itself no traces of
stratification. Still it is seen in most instances resting evenly on
the Middle Glacial beds; it is, indeed, as a bed, stratified with them,
although its own structure is not the result of stratification. There
are no signs of grinding or thrusting of their surface, such as
must have been apparent had the clay been formed beneath a cover-
ing of ice sliding over the land. This latter mode of formation is
accepted by many as a true explanation of the phenomena presented
by the Boulder-clay ; but for this reason especially I differ from that
conclusion. Moreover, if the clay be a direct result of ice moving
over the land, an emergence after the deposition of the Middle
Glacial gravels must have intervened ; but of this there seems to be
no evidence beyond the disputed point of the method of the: clay’s
formation.

In this district we have no indications of the greatest depth of
the glacial submergence; it was probably many hundreds of feet.
That it continued for a lengthened period is certain, judging from
the great thickness in many places of the Boulder-clay. Whatever
oscillations of level may have occurred elsewhere during the Glacial
period, there appear to be in East Anglia no marks of any but one,
and that a gradual and long-continued movement of depression
succeeded by another, equally gradual, of reelevation.

Postglacial—When at length the land again assumed an up-
ward movement, and as it rose from beneath the sea, every part in
turn, as a receding shore-line, would be subject to the action of the
waves, and the surface of the Boulder-clay thereby eroded and to
some extent reassorted. A clayey gravel would naturally result;

196 W. H. PENNING ON THE PHYSICAL GEOLOGY OF

and the whole of the uneven surface of the clay would be more or
less covered by such material, although on the flats, in the hollows,
and in the channels the larger portions of it would be accumulated.

As it invariably happens that subaerial erosion acts most power-
fully along an anticlinal line, the gravels in the hollows and
channels would be the longest-preserved from destruction. And
we find here and there on the surface of the Boulder-clay patches
of this clayey gravel or loam generally capping the higher grounds,
frequently many of them in a curved or straight line, showing
the direction of the old channel in which they were originally
formed. These so-called ‘“ Denudation gravels” are never of any
great extent or thickness, nearly all having been removed by
erosion. Where they exist it may be inferred that there the Boulder-
clay still retains its original thickness ; and a line drawn from point
to point where they occur would give a rough measure of the
muumum amount of Postglacial denudation.

Part 2.—The Drift-deposits in the Cambridge Valley.

The title of “* Cambridge Valley ” I consider to be most applicable
to that main branch of the valley which is occupied by the river
Cam, or Rhee, a stream that rises on the Chalk a few miles from
the town of Royston (fig. 1, Pl. XV.) This stream runs in a N.E.
direction along a line roughly parallel to the Chalk escarpment, until,
south of Cambridge, it is jomed by the Cam, or Granta. Thence
the united streams, under the name of the river Cam, run nearly
due north, still parallel to the escarpment. By Ely their waters flow
into the Ouse, a river which continues the course hitherto taken by
the Cam, and falls into the “ Wash” at King’s Lynn. The valley
occupied by the Cam and its extension the Ouse thus lies parallel
to and near the base of the Chalk; it has been in fact cut back, and
is still being cut back into the escarpment. In this operation the
river is aided by several streams having their source well up in the
Chalk district and running across the strike into the Cambridge
valley. ‘These are :—the Cam, or Granta, which rises near Saffron
Walden, and has a branch from Bartlow; the Lark, from Bury St.
Edmunds and Mildenhall; the Little Ouse, from Brandon; the
Wissey, from Watton ; and the Nar, from the west of Swaffham.

The valley has been, along the greater part of its course, cut down
to the horizon of the Gault (fig. 2, Pl. XV.)—its upper portion
being enclosed on either side by gentle slopes of Chalk-marl, suc-
ceeded by hills of the Lower Chalk. North of Cambridge it opens
out into a broad expanse of fen-land, overlooked from the east side
only by the high chalk range. Rarely do any beds lower than the
Neocomian crop out ; but by these and the Gault the Fen is skirted
all the way to the sea.

The Chalk escarpment receded to its present position or there-
abouts in the Pliocene period: during the progress of its submergence
for, and reelevation after, the Eocene deposits it had been subjected
to some disturbance, many small faults, flexures, and contortions

EAST ANGLIA DURING THE GLACIAL PERIOD. 197

being the result. To a considerable flexure of this period is owing
the present position of an upper extremity of the valley.

That the Cambridge valley, as such, is mainly preglacial is
evident from the consideration of several reasons, amongst others
the position which the Boulder-clay now occupies therein*. It
rests on the escarpment at an elevation of more than 500 feet; it is
found here and there on the flank at lower and lower levels; and it
occurs in the bottom of the valley at but a small height above the
sea. It is true that in those instances where Boulder-clay is found
on the flank it is in small outlying patches only (fig. 2, Pl. XY.);
but this is inevitable from the progress of recent erosion; and they
are quite sufficient to prove the former extension of the mass,
These patches now cap small elevations resulting from denudation ;
and a line drawn from the escarpment to the low ground would
intersect them all.

Although the escarpment had been cut back to nearly its present
position during the Pliocene period, and the land was somewhat
submerged during the deposition on the east coast of the Pliocene
beds, we find in this valley no signs of their present or former
existence. Tor the valley was not excavated to its present depth,
by perhaps 50 or 100 feet; consequently any deposits that may
have been left in it during the Pliocene era have long been swept
away. For the same reason, perhaps, we have no beds of Lower
Glacial age, which are, so far as I have seen, confined to areas of
comparatively slight elevation. It is not difficult to suppose that ~
these Lower Glacial deposits would be excluded from this valley by
its then height relatively to that at which they are now found, or
that any which may possibly have once occupied the area have been
subsequently removed.

But it is far more difficult to account for the apparently total
absence of the Middle Glacial deposits, which, just over the Chalk
escarpment, run up to a height of 300 feet or more above the sea,
and to at least 200 feet above what must have been the bottom of
this valley at the time of their deposition. Thesubsequent physical
conditions were not favourable to, nor was the time sufficient for,
the removal from the valley of beds of any extent (assuming them
to have been therein deposited) before the deposition of the Great
Chalky Upper Boulder-clay. We should find them still, however
ereat the subsequent denudation, between 'the older rocks and the
Boulder-clay in some at least of the many instances in which it
occurs. It cannot be assumed, as in the case of the Pliocene and
the Lower Glacial, that the Middle Glacial beds at one time occupied
the valley, and were afterwards, and before the Upper Boulder-clay
period, removed; yet in no instance do we find the Boulder-clay
resting on any other than preglacial formations. As suggested in
the first part of this papery, the currents from the north that
formed the Middle Glacial gravels and sands were confined to the
seaward side of the Chalk range, which was not wholly submerged,
and were entirely excluded from the Cambridge valley; for, as we

* Ramsay, ‘ Physical Geology,’ p. 211. “i Ante, p. 193.

198 W. H. PENNING ON THE PHYSICAL GEOLOGY OF

have seen, if they had had access to the valley, and had not been ex-
cluded therefrom as suggested, some undoubted traces of the series of
gravels and sands would remain and testify to the fact. There are
certain gravels which, | admit, bear a striking resemblance to those
of undoubted Middle Glacial origin; but my reasons will presently
be given for believing them to be of much more recent date. In some
cases the Boulder-clay has a gravelly base; but it is that and nothing
more, representing the gradual change of physical conditions which
occurred when the Chalk-ridge was nearly submerged. At this
time, although the southerly set of the water still continued*, its
currents possessed much less power of transporting and rearranging
gravels, and their deposition ceased. But the change was gradual ;
and we may readily conceive that during the transition-period the
surface of the Chalk, being slowly encroached upon, would be partly
covered over with a wash from the rock itself, and perhaps also
from the gravels. This wash is not a clean gravel; and it
gradually passes up into the Boulder-clay, which then began to be
deposited.

We have seen that the Cambridge Valley was excavated in pre-
glacial times, and that in all probability no Lower or Middle glacial
beds were formed within its area. But when the Chalk was wholly
submerged t+, the sweeping currents, hitherto confined to the east
side of the range, were replaced by an open sea extending over every
part of East Anglia that does not now attain to an elevation of
500 feet. In this sea the icebergs laden with rocky débris from the
north were slowly melted; and their freights, descending 7 mass,
formed on the bottom the unstratified Boulder-clay. There had
been icebergs borne along in the -Middle-glacial currents also; but
they were swept off more quickly to the southward, and probably
a great majority of them may have melted over an area still beneath
the waters of the Atlantic.

The Upper Boulder-clay is found all along the top of the Chalk
range ; it caps the minor elevations on the flank, and it occupies
some of the lower ground in the valley. At one time it doubtless
spread as a sheet over the whole area, from the highest point of the
’scarp down to, if not below, the present sea-level. It has since been
so much denuded that its main mass on the hills is disconnected
from the remainder, which now exists merely as outliers on the
smaller hills and ridges. '

Noting that the Boulder-clay is found on the back and top of the
escarpment, as well as on the low ground beyond the foot of it, we
might assume that there is no “‘ horizontality ” in its mode of occur-
rence, and that its present boundary-lines would ignore all the con-
tours, and features even, of the country. To some extent, and in
small areas, this may be so; but looking at the Boulder-clay as a
whole, there is a striking regularity in its occurrence. When viewed
on a true scale, in the exceptional case of the Chalk escarpment it _
is seen to plunge down about 500 feet ; but this being in a distance
of not less than 10 miles, represents a fall of 1 in 100 only, or an

* Ante, p. 194. + Ante, p. 195.

EAST ANGLIA DURING THE GLACIAL PERIOD. 199

angle of about half a degree. It touches, as any other formation
. would be expected to do, the outliers and hills within the larger
valley which are sufficiently high to reach its horizon; and its
boundary-line conforms generally to the present features of the
district.

As the surface of the Chalk, when gradually encroached on by
the glacial sea, became partly covered by a gravelly wash, so also
did the Boulder-clay on its emergence*. Of this gravel or loam but
little remains, and that invariably on the higher grounds, in
patches which once formed portions of extensive plateaux. I think
it can scarcely be assumed that so large a mass of material as the
Boulder-clay which once filled the valley could have been removed
by denudation, marine or subaerial, or by both combined, without
its component particles having been more or less reassorted and re-
deposited on the denuded surface. This may have been the origin
of certain deposits of doubtful age which occur, here in the form of
loam, there as an impure gravel.

Besides the high-level deposits, there are in the valley broad
sheets of much newer valley-gravel, enclosing the remains of recent
shells and of extinct mammalia. They occur in three or more
terraces at different levels, marking as many points in the progress
of the valley’s formation, the higher and older terraces haying been
already reduced to mere patches, indicating a former extension, and
testifying to the rapidity of denudation. These gravels have been
made up mainly from the waste of the Chalk and superincumbent
Boulder-clay, and may be seen in many sections in the neighbour-
hood of Cambridge. Much has been already written about them ;
they have now been mapped by the Geological-Survey Officers, and
will be fully described by them also in their publications. The
same remark applies to the Fen-lands, which are of more recent
date than the Drifts proper that form the subject of this communi-
cation.

But there are within the area certain other gravels upon which a
few remarks are necessary, owing to their resemblance to those of
Middle-glacial age. These deposits occur at an elevation of 20 to
60 feet, or thereabouts, above the level of the river, and consist of
gravels and sands with intercalated masses of loam and clay, the
latter having somewhat the appearance of Boulder-clay, or, rather, of
a wash from Boulder-clay in its immediate vicinity. The lines of
stratification are irregular, sometimes horizontal, more frequently
inclined, and, in the two most noteworthy sections, several miles
apart, dip north at an angle of 15°. In one of these sections are two
erayels, or, rather, a gravel and a loam, the latter being banked up
against a scooped-out edge of the former, thus presenting the ap-
pearance of a fault.

There is a Middle-glacial appearance about the gravels; but still
I think that all the phenomena might occur as well either in an
esker or in deposits left by a river of tolerable magnitude. In-
clined bedding is not unusual in river-gravels; and the included

* Ante, p. 198 and Part 1, p. 196.

200 W. H. PENNING ON THE PHYSICAL GEOLOGY OF

masses of wash from Boulder-clay are easily accounted for, if we
assume that the gravels were formed during the removal by denuda-
tion of the Boulder-clay which once filled the valley. The uniformity
of level at which they occur would also seem to indicate a fluviatile
rather than a glacial origin. But the strongest argument in support
of this view is the fact that the gravels in places contain recent land
and freshwater shells (Heliw, Pupa, Succinea) ; this would be con-
clusive evidence if the shells could be more generally found. But
it is just possible, although far from probable, that in the few in-
stances only in which (so far as I know) they have been discovered,
a newer gravel may have been deposited in channels cut by streams
through that which is of older date. In my opinion the parts of
the gravel where the shells do occur so exactly resemble the other
portions where they do not, that such a proposition cannot be enter-
tained. Therefore without hesitation I refer these deposits to the
existence of an ancient river running, as does the present one, along
the foot of the escarpment, although, of course, not on the ground
occupied by the Cam of our own time. The Chalk ’scarp was then
somewhat different from what it is now; it had been cut down to
such a level only that its foot was higher by as much as these gravels
are now above the river. The line of this old river is now indicated
by an elongated series of patches of the gravel in question which
formerly occupied its channel. The remaining portions of these
eravels now form long, low, and rounded ridges, occupying, pro-
bably, slight depressions in the Chalk, which depressions (formerly
a part of the old river-course) are now somewhat elevated above the
surrounding area, the gravels having partly preserved, as usual, the
lines of hollow from denudation. This series commences near
Royston, in a hollow formed by the flexure in the Chalk previously
mentioned*, runs in an easterly direction about parallel to the
present course of the river, crosses the minor valley of the Cam or
Granta, is joined by another line of similar patches from the east,
and then sweeps round to the north of Cambridge, being lost in
the fens beyond (fig. 1, Pl. XV.).

I am well aware that this solution may not explain the origin of
all the patches of gravel that are found in the valley and cannot
with certainty be referred either to the Glacial, Postglacial, or Recent
period. But I do not think that any of them have been actually
traced under the Boulder-clay, as they would have been if of Lower-
or Middle-glacial origin. It is possible that during the period in
which the drifts were being formed by the Middle-glacial currents
traversing the east coast, a stray iceberg may have occasionally
found its way into the land-locked inlet of the Wash and its exten-
sion the Ouse and Cam valleys, at that time so much broader
through greater submergence. But this would be, I apprehend, a
very exceptional occurrence; and any deposits dropped from such
bergs are not to be classed aoe the current-formed gravels that occur
in such, force on the other side of the escarpment.

Taking all the facts into consideration, I feel justified i in drawing

* Ante, p. 197.

EAST ANGLIA DURING THE GLACIAL PERIOD. 201

the conclusion that, strictly speaking, there are no Lower or Middle
Glacial deposits within the area of the Cambridge valley *.

SUPPLEMENTARY Norns.

Since the above notes and the conclusions therefrom were brought
before the Society, the completion of the mapping of some of the
gravels, mentioned as of doubtful age, seems to have thrown light
upon the date and mode of their formation. It has afforded, also,
additional evidence in support of the theory advanced respecting
those gravels which were thereby asserted to be relics of an ancient
river Cam.

There is a hill called “‘ The Rivey,” situated about ten miles 8.E.
of Cambridge ; its height above the Cam is about 325 feet ; it is, as it
were, an outlier from the escarpment, and it is similarly covered by
Boulder-clay. This Boulder-clay rests directly on the Chalk, and is
capped by a few feet of gravel; it is supposed that this is a patch
of ** Denudation gravel,” and that the Boulder-clay beneath it re-
tains its full thickness ; or it may be a higher example of those to
be described. In a direction running away from the escarpment
are patches of gravel at a level several feet lower ; then, further on,
other patches at a still lower level, and others, still further away,
with less elevation, the Boulder-clay being continuous beneath them
all. At Hildersham, in the valley, 13 mile from the Rivey and
200 feet below the summit of the hill, the gravel and Boulder-clay
occur in the same relative positions; and a little further on, the
gravel overlaps the clay and rests on the Chalk.

This interesting and well-marked geological feature has been pre-
served, it may be, by its position in the bend to the north of the escarp-
ment; and it affords a key to many other patches of gravel in similar
positions, and heretofore considered as of doubtful age (p. 200).
All the patches (many of which are small), and the indications of
several other patches removed by recent denudation, seem to form
parts of lines that once were continuous, and which are more or
less at right angles to tho long series ascribed to an old course of
the Cam. They continue in a gradually descending order from .
almost the top of the ’scarp down to the level of that series—that
is, from 325 feet down to 20 feet above the present river.

The explanation proposed is, that the slope of the Preglacial
‘scarp was entirely covered by Boulder-clay, that the valley was, in
fact, almost filled by that deposit, the patches of Boulder-clay which
here and there remain marking out the line of that Preglacial
escarpment (see ante, p. 198). This Boulder-clay has been removed

* Since the above paper was read the authoyr’s attention has been called to
the fact that in the ‘Geological Magazine’ for February 1870, Mr. 8. V.
Wood, Jun., published a statement to the effect that the Middle Glacial beds in
East Anglia do not occur at a greater elevation than 250 feet ; also, that in a
section illustrating a paper by him read before the Society, June 19, 1867, the
overlap of those beds by the Boulder-clay, and the occurrence of the latter in
the Camb:idge valley without any Middle Glacial are represented. Although

Mr. Wood did not draw the same inferences from these facts, he is entitled to
priority in their observation.—W. H. P.

202 W. H. PENNING ON THE PHYSICAL GEOLOGY OF

entirely by postglacial erosion ; and the Chalk has been cut back
beyond the old line, except where such patches still remain. In
the early Postglacial times, when the valley began to be reexcavated,
a Inain stream must have run approximately in the same direction
as the present river, but then, of course, at a height considerably
above it—as much higher really as the thickness of the Boulder-
clay that has since been removed. The escarpment covered by
Boulder-clay was cut back then in the same way as it is being cut
back now, by minor streams running across the strike (p. 196) into
the main river parallel thereto. This operation began immediately
upon the reelevation of the land after the great submergence, and
has continued up to the present time.

There must have been very considerable accumulation of detritus
from the removal of such a mass of Boulder-clay ; and that removal
would be the more rapid in proportion to the greater steepness
of the area subject to denudation. These accumulations would,
in turn, be removed ; but it is suggested that in the mass of Boulder-
clay and overlying patches of gravel leading down from the Rivey
to the valley below we have evidence of the later stages of that
clay’s denudation, and small remnants of the resulting accumulation.

The gravels are made up of exactly such material as would be
derived from the waste of Boulder-clay—rolled lumps of hard
and soft chalk, flint pebbles, with some fragments of quartz, liime-
stone, and other derived rocks, enclosing broken pieces of derived
fossils. They constitute also a gradually descending series, with-
out any very great break in their continuity, either in regard to
level or position. All the patches appear to occupy hollows in the
surface of the clay, by which circumstance they have, indeed, been
preserved—that is, down to the level of the Preglacial valley-
system ; below it the gravels rest on the excavated chalk. In many
instances they occur in hollows along the top of a ridge—the old
stream-channels in which they were left now forming synclinals,
and thus preserving the gravels and the clay or chalk immediately
beneath, while that on either side has been worn away.

The marine Middle-glacial currents could not have formed these
cravels, even if they had access to the valley, as some of the ridges
run up into coombe-like valleys where such currents would have
been impossible. The streams that deposited them were doubtless
rapid, and occupied gorges in the clay; this would account for the
false-bedding sometimes exhibited.

The series described as indicating the ancient course of the Cam
represents the latest stage of all, previous to the present system of
drainage ; in this series and in patches at or slightly above the same
level occur the recent land and freshwater shells. At one place in
this series shells of Cardium edule have been found ; and their occur-
rence would seem to bear out the suggestion made, that during the
Glacial period the Cam valley formed a land-locked bay, on the sandy
shores of which this mollusk flourished (p. 194). It occurs, but not
plentifully, at the base of the deposit, and probably indicates the
position of the old shore-line.

EAST ANGLIA DURING THE GLACIAL PERIOD. 203

If the explanation offered be the true one, we have here a series
of river-gravel terraces occurring at all heights, from nearly the
bottom of the present valley up almost to the top of the Chalk
escarpment.

Approawimate Elevations (above the Cam) of two lines of
Gravel Patches.

feet. feet.

EMMERIVGY |e ceceee ers casio eons 325 Bale hadi: ic. siacs bbcaysuan givens Sik
; 270 Small patch W. of Balsham 25
sp alld ar a { 200 Girne 205
BPRIIOPSUHEL +... 0.002. ccene ence 125 (Possible continuation of the above.)
(Indications of other patches between Grayel-pits N.W. of Dun- | ,--
175
the above.) PRB = sid daan ad alapanlgan abate
Wratting valley ............ 110

Ancient course of the Cam, 20 to 60 feet.

EXPLANATION OF PLATE XV.
Fig. 1. Map of a part of East Anglia, showing the Drift-deposits. Scale 8
miles to 1 inch,
_ 2. Section from London-Clay escarpment to the Cambridge valley along

the line indicated on the map. Horizontal scale 4 miles to 1 inch;
vertical scale 800 feet to 1 inch.

Discussion.

Prof. Duncan said that his impressions, formed twenty years ago,
agreed with those of the author. It was a very simple case of
geology complicated by geologists. He inquired whether the
author thought that such valleys as extend up to the Chalk from the
German Ocean were excavated before the Glacial period, or whether
they were then filled up and afterwards excavated. He also asked
whether the author was quite satisfied that all the gravels are water-
worn by tidal action. It seemed to him exceedingly interesting to
find that Mr. Penning pushed the ice-cap a little further back.

Prof. Morris remarked that the valley of the Cam presents two
or three different conditions; near Cambridge the river flows over
the Gault, cutting through the old gravel between Barnwell and
Chesterton. The valley was probably excavated in the Boulder-
clay, which occurs at higher levels, and once no doubt covered the
whole district.

Mr. Juxes-Browne thought that the Boulder-clay did fill the
valleys, and that the boulders referred to by the author had been
derived from the Boulder-clay. One point of special interest in the
paper seemed to him to be that the author regarded the whole
Glacial series as continuous, in opposition to those who would place
the excavation of all the valleys in the Glacial period, while the
author considers them preglacial. At Cromer the so-called Middle
Glacial is nothing but blown sand.

Mr. Grorez Maw inquired as to the order of sequence of certain
beds known as Lower, Middle, and Upper Glacial, what evidence

204 ‘ON THE GLACIAL GEOLOGY OF EAST ANGLIA.

there is with regard to those extensive coast-beds which occur all
along the coast from Yarmouth to Cromer, and especially what evi-
dence we have as to the so-called Upper Boulder-clay on the coast
being really newer than the great mass of the drift of high Suffolk.

Prof. Ramsay thought that the question raised by Mr. Maw had
been thoroughly discussed at the last meeting of the Society. He
had had much pleasure in listening to Mr. Penning’s paper, because
that gentleman seemed to have aimed at getting rid of the infinite
refinements in the classification of these beds as Upper, Lower, and
Middle Glacial deposits, these things being generally treated in such
a manner as to produce, at least in the popular mind, a notion that
these deposits are of as great importance as Upper and Lower Silu-
rian, for example, with beds thousands of feet thick. He thought
that the effects of Glacial action had been immensely exaggerated,
and believed that all the great features of the country existed before
the Glacial period.

The Prestpent (Mr. Evans) considered Mr. Penning’s paper of great
importance from two points of view—first, with respect to the sequence
of the Glacial beds, and, secondly, with regard to the Cambridge valley.
He thought that the author’s views as to the southern part of the
Cambridge valley would be borne out when he came to examine the
portion north of the Lark, and further north in the neighbourhood
of Brandon, where he would be able to trace the connexion between
the Glacial and Postglacial gravels. With regard to the ice-cap,
there were certain points that deserve great attention, especially the
question whence are derived the materials of the Boulder-clay.
These seem to be in great measure derived from the Upper Cre-
taceous and Tertiary beds in the neighbourhood. And in some cases
unworn flints show glacial scratches; hence he suggested that
inquiry should be made whether they were due to local glaciers.

The AvurHor, in reply, stated that the Cambridge valley was ex-
cavated in Preglacial times. There is no evidence of extended de-
nudation in Glacial times, but only of local unconformity.

We publ) re or) 43) ProppIay'2z plpTpces wre (L

perouwy soar b Ng sapog 4 "pownpyy~pyy'® Koy) uopuoT’ p — spog Burpvaya yoy -q-ammg-y

yoxnuy i SS eee - 2
Kanyon be —— - a
PIO a hep Saree 2s UaALY pramdimmaey
WHAIY PTD gd el! med dO ; Apy) Uopuo'y
aon? quads yy “¥POU) Me

‘$

‘YIU [ OF JAA OOQ FPRIG PVIIpLa4 “YOU [ Of SAPTUL ¢ ‘OYDIG YOYUOrAOTy
soy ebpruquny) ey) 07 quewdsposy Nop) Uuopua'y moss WOIY®Y *Z OI 4
(op YOU? /syooy LPO IAC

IP YONDAD FOVIVID ayp pr et ae
“DAS IY} AAOGY YAS OOk PUO QO O0e ‘QOL £ ] PIN EE SH z

pny mang = Mp] «opynog ae
| [AVIA TOIT D FSOT =.

“YIU | OP SIYNM Y ‘aYOIG ummMamy {- care
f g P i eae

a ==

“sprsodagy YA] ay} hbinmayy

SCELONY LS i \ Ie
Jo y.vd Y Jo ig | | me
dVW ‘1 9I4 * tee =

pase tele ie ea titth, = © S!4g¢--——

AY 1) META re Pod Tee TTT

trays Weng 4g ESOypAU'ae vraTESs UCL

“Ang sappnog J AE CADIS 2 PAG ae

Ca) 7 n prada
ene App) uopuoTy
“YIU [OF PIAS YYZ PPPS PPM, “YPM | O} SATIUL § ‘ADI TopUoL LOT ‘Ss

Aan ebprquay oy) OF qunudinasy NVy) Uopuo'y wo0.4) WOIDEY ‘*Z OI 4

(op yrya/sy20y OPO

Ip JINDAL POTD IP PIT
Noy «opnog §

poxDag PMD I

b you fh Oy Saye g ‘apna. UO

DAB IY} BOTY YAY OD PUB" VOC: ‘Ve: OOF

MINT NOTED) ee

so 7
fe ee

eT oo an oe

[ psojbumunig 7

ea aaa NN
Rae aye ee ee

es B ecemapee i ofa casa
Ses ae. ody, ‘I iil

Seen Gaaeot

ieeieeraeee oe = a ae ee ae

O.—+— 9 — eo

Pagar

“spisoday Luigy ay} bunmays

evel NOV ESV el

Jo yd D Jo

dVW ‘1914

nan a 1p Tm

~
.
‘
.
‘
e ‘
’
.
*
ee)
ud

4
F
“*
:
‘
‘
°
‘
af
'
i
i.
W J
A,
i
he

oan

oc athagae De pyc

4

vet

‘
f
t
3
:
.
L
‘
*
ae"
ta
.
‘ -
%
’
,
*
.
‘ “N
Ve ,
, yas Boat

P. M. DUNCAN ON PARASITIC UNICELLULAR ALG. 205

23. On some UNICELLULAR ALG parasitic within SILURIAN and
Tertiary Corats, with a Notice of their Presence in CaLcroLa
SANDALINA and other Fossils. By Professor P. Martin Duncan,
F RB.S., V.P.Geol. Soc., &. (Read January 19, 1876.)

[Puate XV.]

Jonn QueExerr described, in his lectures at the Royal College of
Surgeons in 1851-1852, certain tubes or canals as being very fre-
quently met with in the skeletons of corals; and in his Lectures on
Histology, published in 1854, and which contained the subject
matter of his previous discourses, he wrote as follows: ‘‘ Confervoid
growths also are very frequently met with in the skeletons of corals,
as all these bodies possess animal matter which, decomposing after
death, becomes a nidus for the development of conferve ; and hardly a
section can be examined without exhibiting such an appearance as
shown in fig. 78”*. This figure shows long, short, and almost
straight canals cutting across the normal coral structure at different
angles.

In 1859 Kolliker gave the results of his examination of some
corals to the Royal Society, in a communication “ On the frequent
occurrence of vegetable parasites in the hard structures of animals”.
He said :—‘ All the genera of corals which I investigated contained
parasitical fungi, viz. Astrwa diffusa, Porites clavaria, Tubipora
musica, Corallium rubrum, Oculina diffusa, Oculina, sp., Alloporina
mirabilis, Madrepora cornuta, Lophohela prolifera and Nullipora
alecornis. The fungi were most frequent in the genera T'ubipora,
Astrea, Porites and Oculina, the last three of which contained also
many sporangia, which in the red coral were very scarce or
wanting.”

Before Quekett lectured, and contemporaneously with his and
Kolliker’s researches, Dr. Carpentert and C. Wedl§ investigated
the corresponding tubes or canals in shells; and the last-named
naturalist communicated a most admirable paper on the sub-
ject just before Kolker came forward. Wedl described and de-
lineated the tubes in perfect and in decalcified specimens of shells,
and obtained a view of the parasite itself when removed from its
nidus in Melania Hollandri under the effects of a dilute acid. He
agreed with Quekett in ascribing the parasite to the Confervee, and

* John Quekett, Lectures on Histology, vol. ii. p. 153. Parasitic borings
corresponding to those noticed by Quekett were described by C. B. Rose, F.G.S.,
in fossil-fish scales from the Chalk and Kimmeridge Clay, in the Transactions
of the Microscopical Society, 1855, p. 7. He figured them ; and his faithful
delineations show ramifying tubes and occasional globular enlargements.
He attributed them to the operation of infusorial parasites.

t+ A. W. Kolliker, of Wurzburg, Proc. Royal Soc. June 9, 1859, vol. x.

+ Carpenter, Cyclop. Anat. Physiol. art. “ Shell.”

§ Wedl, Sitzungsber. d. kais. Akad. d. Wiss. in Wien, B. xxxiii. no. 28. Dec.
1858.

206 P. M. DUNCAN ON SOME UNICELLULAR ALG

believed it to be the Saprolegna ferax of Kitzing. He did not
examine the corals; but after satisfying himself about the occurrence
of the parasite in recent species, he examined some fossil shells, and
detected it (amongst others) in Leptena levis from the Devonian.

Thus these investigations showed that there were parasitic vege-
table growths in modern corals and in many shells even as old as the
Devonian, and that they either were conferve or fungi.

During the course of some investigations into the minute structures
of Tertiary corals, | was greatly puzzled by the omnipresence of a
system of branching canals ending in culs-de-sac and having dark
borders and a refractive central area. In one instance (in a
Thanmastrea * from Tasmania) the tubes frequently merged into an
irregular dark mass, and resembled the tubuli of bone (Haversian
canals) passing into lacune; and as they surrounded in a circular
series a radiating mass of closely approximated normal spicula, the
resemblance to a low class of osseous tissue was extraordinary. These
canals had their length, direction, and frequency evidently in relation
with the situation and regularity of disposition of the denser and
normal coral structures: where the spicula were closely united
laterally, their extremities radiating from a common centre, the
tubes did not pass into, but surrounded the mass; and when
these nodules of normal tissues were in long series the tubes ran
down by their sides. Hence sections cut across the calices and
septa did not exhibit many tubes, but numerous dot-like markings,
which were their cut ends; on the other hand, sections longitudinal
to the septa and costz presented long lines of tubes with ramifica-
tions and swellings. The resemblance in shape and size of these
tubes to Quekett and Wedl’s figures led to the belief that even this
hard coral had not been without its parasite, and incited me to
follow up the subject in living, recently dead, and other fossil species.

The results of my work on the recent forms, as they enter especially
into questions foreign to those considered by this Society, are about
to be presented to the Royal Society ; but I thought that a notice of
the occurrence of these interesting parasites in such old forms as
Goniophyllum and Calceola would be of interest.

I chose this species of coral and Calceola because their hard parts
admit of thin and wide sections, and also because I was working
upon them in conjunction with Mr. H. Woodward, F.R.S., im in-
vestigating the question of the Rugosa operculata of Lindstrom. But
I have examined others also, with different degrees of success. In
the spevimens which Mr. H. Woodward had had cut, the fossilization
had been very perfect, and much of the calcite had been replaced by
crystalline carbonate of lime. The calicular fossa and the cavities,

* On decalcifying a part of a thin transverse section of this Thamnastrea, and
after careful washing, a ;!,-in. immersion-lens was used. It showed a basement
tissue almost homogeneous, on which were numerous straight and a few rami-
fying tubes, resembling in shape the refractive tubes of the perfect specimen.
The tubes were such, and not excavations in the homogeneous tissue ; and they
had each a proper and hyaline-looking wall; the contents were more refractive
than it and than the surrounding structures. Hence the cellulose wall was
preserved during the process of fossilization.

PARASITIC WITHIN SILURIAN AND TERTIARY CORALS. 207

generally speaking, were filled either with sand and minute fossils of
the same period as the mass, or with crystalline carbonate of lime.
The specimens were very perfect and had not been rolled.

In investigating these parasitic growths and perforations in recent
and fossil specimens, it is necessary to use thin sections for transmitted
light ; but a thick section often exhibits the refractive tubes when
reflected light is employed. The sections of the fossils should be
carefully made, and scratches on their surfaces noted; and in order
to prevent erroneous interpretations, it is as well not to record any
tubes as parasitical in their formation unless they contain matters
resembling more or less those of the recent forms, and unless they
can be traced in and amongst the normal tissue, and not only on the
surface.

The best plan is to examine, first of all, that part of the section of
the coral or shell which was formerly exposed to the sea; there it
is exceptional not to find one or more straight dark lines passing
from close to the external margin or old surface inwards at different
angles (fig. 6). (A magnifying power of 350 linear is necessary, and
careful and good illumination and definition.) When they are satis-
factorily seen it is necessary to examine them throughout their length,
and to establish, if possible, their relation with others and with the
outside, and to notice their contents.

These perforations or tubes, and the concavities and little loculi
with which they are often connected at the surface, cannot be mis-
taken for Cliona-borings; for these last are larger, contain spicules,
and do not present the long and often tubular branchings of the
vegetable parasites, which, moreover, never contain spicula. But
the loculi, when some of the Algx get into the corals, do often
resemble the results of the early efforts of Cliona to perforate ; and
it is quite possible that the Algae may have subsequently occupied
the space where a Cliona had been at work ineffectually.

Their length, minute size, and general characters separate them
from some very ill-defined organic perforations seen in Belemnites and
modern shells, and which Fischer* and Quenstedt have termed
Dendrine. It is hardly necessary to suggest that the edges of the
planes of crystallization in no way resemble the tubes.

Appearance of the tubes and other parasitic productions in Gonio-
phylum pyramidale (Plate XVI.).—The microscopical elements to
be observed, are :—

1. Tubes which have no proper wall, and which are excavations
out of the coral-structures. They are found (a) just beneath
the surface, running parallel to it; but these are rare (fig. 2); (3)
running more or less inwards at different angles to the surface,
many being found near to the edge of the coral-wall, and a few far
away towards the interior (figs. 6 & 9). These last-mentioned
tubes do not vary much in size, and average in diameter about
0-008 in. Their calibre does not alter in different parts of their
‘course, which is rarely curved, usually straight, and occasionally
branching, the branches being often as large as the parent tube.

* Fischer, ‘Comptes Rendus.’ Dec. 6, 1875.
Q.J.G.S. No. 126. P

208 P, M. DUNCAN ON SOME UNICELLULAR ALG

Both the kinds (a and {) are usually filled, except in the axis, bya
dark granular matter; and the tubes therefore present the appearance
of dark edges with a longitudinal central clear line. The length, in
some instances, extended over more than one field of the microscope..
Each ends in a cul-de-sac, which in this instance is not swollen out;
and often where there are no cell-contents for a little space the
absence of special walls can be readily determined. Usually no
origin or ending of the tubes can be seen; but in a few instances
their commencement in dark spots at the margin of the coral-wall
can be readily seen (fig. 6). These spots are either not much
larger than the tube, or are very much bigger, and are filled with a
mass of globules with whose exterior the tubes seem to be continuous.
Occasionally small dark pigment-masses with a definite globular
shape are in contact with one end of the tubes (fig. 14). The large
masses are probably the remains of resting spores or oospores ; and
the others are conidia (fig. 3). With regard to the numbers of the
tubes (3 there appears to be no regularity; in some places they
are very widely apart, and in others crowded ; but they never appear
to inosculate with others, but simply branch.

In one or two tubes in the specimens examined there are dark spots ;
and in one the calibre is swollen out at one spot.

y. There are here and there very minute tubes which ramify
frequently and in a short space, so as to be very dendritic in
appearance; they are densely black and opaque, and their diameter
is about one half that of the other tubes.

2. More or less globular conidium-like masses are either separate
_or crowded, and in this last instance often are in linear series.
(a) They constitute moniliform bodies (fig. 3), sometimes with
tubular projections. (3) They are in evident linear series, but are
disconnected; nevertheless traces of excavations, which probably
are relics of old tubes which once contained them, are occasionally
visible.

3. Tubes having a calibre twice as large as the others, or even more,
and whose contents are discontinuous, dark and often in the form of
the conidium-globule (fig. 8).

There is a piece of a Brachiopod shell in the matrix within the cali-
cular fossa of the coral; and it shows tubes /2 to perfection; and
they look like so many straight and curved wires (fig. 4).

The tubes mentioned under section y, and the more or less
irregular black spots with which they are continuous, readily receive
explanation after the study of the Alge parasitic in the Thamnastrea
from Tasmania (fig. 1), and of Calceola sandalina (fig. 11).

In the Thamnastrea the enormous multitude of tubes simulate
radiating spicula in appearance, and here and there one or two can
readily be traced running into a black mass. This irregular shape
produced by the growth of the Alga depends on the special molecular
structure of the coral. Ifthe tubes were obliterated by fossilization,
and the black spaces, not unlike the lacune of bone, remained,
the appearance would greatly resemble that of some parts of Gonto-
phyllum.

PARASITIC WITHIN SILURIAN AND TERTIARY CORALS. 209

Examination of Calceola sandalina.—In the Calceola examined
the tubes of all kinds were seen and conidia of globular shape
included in the larger ones. The largest tubes (fig. 12) are four
or five times as broad as the medium-sized straight tubes (fig. 11);
and their exit in a loculus opening outwards at the surface can be
seen. The loculus (figs. 12 & 13) sometimes contains a crowd of spores ;
and little wavy canals pass out from all sides. The contents of
the tubes have undergone alteration in some, and a reddish tint has
replaced the ordinary greenish-black colour.

The sections of such fossils necessarily contain tubes at different
angles ; and some which lie parallel to the line of incision are injured ;
hence the continuity of the conidium-bearing tubes is interfered with,
and these are often left without a trace of the former tube. The
same occurs with regard to the small branching tubes, which become
broken up and isolated by the section. This is seen in a Lower
Silurian Foraminifer (fig. 5).

REMARKS.

A comparison of the parasitic excavations of recent corals with
those of the Secondary and Paleozoic ages presents most remark-
able resemblances. Tube may be compared with tube in all its
parts ; but fossilization has produced appearances in the spores
and conidia which suggest distinction between the recent and the
fossil kinds of Alge. Nevertheless the general character of the re-
productive resting spores and the conidia arising from the vegetative
part of the organism remain much the same. The large tubes in
the paleozoic coral and Brachiopod, or whatever else Calceola may be,
would at first sight indicate a different species of parasite from those
which formed the smaller penetrations ; but both large and medium-
sized tubes often exist in the same recent corallum, and these last
now and then give off others so small and so finely linear that their
diameter cannot be measured. Whilst recognizing two or three
forms of parasitic Algze within these sclerenchymatous structures of
recent and ancient date, it does not follow that they are to be made
into different species. They may all be parts of the same mycelium-
like growth of the parasite, and may depend upon the nature of the
nidus in which growth has taken place.

Tubes of analogous sizes and shapes are found together in recent
corals, and they are often continuous.

Wedl suggested that the Conferva which grew into the shells was
Saprolegnia ferax. Kiitzing and Kolliker, from the want of cell-
like partitions in the tubes, objected to the confervoid nature of the
parasite, and urged that it is one of the Fungi—one of a group
which grows at the expense of animal tissues, and secretes carbonic
anhydride.

The distinction between Saprolegnia and the Fungi, however, is
but doubtful. Its spores vegetate; and the tube growing from
them, in some species, speedily perforates Conferve and dives into
their cells, growing and developing at their cost. It is really in-

210 P. M. DUNCAN ON SOME UNICELLULAR ALG

separable from the group of the <Achlye, or, rather, from a group
which embraces Achlya, Empusina, Sporendonema and possibly
Botrytis*, all being members of the Protista group, whose natural
distinctions, evident enough sometimes, are not so in some parts of
their life-cycle. That one becomes the other, on a change of the
surrounding medium occurring, is one of the most interesting and
suggestive of facts. Evidently the parasite got into the old corals,
as it does into those now living, from the outside, and by contact,
growth, pressure, protoplasmic movement and the dissolving effect
of the evolved gas, slowly and surely penetrated. The course, size,
shape, and length of the tube being determined by the presence of
the organic matter within the sclerenchyma and the arrangement of
the coral-spicula, it is best to term the eritophyte an Alga, and
to classify it amongst the unicellular types in the neighbourhood
of Achlya, calling it Paleachlya perforans. It is of course im-
portant to decide when the perforations were made. Were they
forming contemporaneously with the growth of the coral and shell,
or were they of subsequent date? In the first case the Silurian and
Devonian age of the Achlyan becomes apparent ; but if the second
supposition be at all consistent with facts, the whole interest of the
subject vanishes.

In favour of the theory of the simultaneous life of the host and
the parasite, the theory of the growth of the Algez in recent
times must be advanced and the presence of sea-water and of
animal tissue of a low vitality assumed. In addition there is the
fact that a portion of a Brachiopod included in a sandy matrix
within the coral, and not continuous with coral-structures, con-
tains the tubes. Moreover the crystalline mass of the inter-
stices of the coral, although it holds mechanically abundance of
spores, does not present tubes or any evidence of growth. It is
therefore in accordance with our knowledge to assert that the
parasites lived at the same time as the organisms which they
penetrated, and that this minute Alga presents one of the most
singular proofs of the persistence of form and life-cycle from the
paleeozoic age to the present. In conclusion, I have.to thank Mr.
W.S. Dallas, Prof. Morris, and Mr. H. Woodward for references,
suggestions, and sections.

The characteristic tubes of this Alga have been found by me in
species of Cyathophyllum from the Upper Silurian, and in a
Foraminifer from the Lower Silurian of Canada.

EXPLANATION OF PLATE XV.

Fig. 1. Tubes in Thamnastrea from the Miocene of Tasmania, x 350.
2. Tubes in Goniophyllum pyramidale from the Upper Silurian, close to
the wall, x 350.
3. Spores in Goniophyllum pyramidale, Xx 450.
4. Tubes in the.shell of a Brachiopod imbedded in Goniophylinm pyra-
midale, x 350.

* See Micrographic Dictionary, 1875, 3rd edit. for these genera.

h Pai ’ i
Ail

i
an
5

OM we sys Was F oi
. a rte - Nr Ge Aire

Quart. Journ .Geol. Soc Vol. XXXII. Pl. XVI.

Mintern Bros imp.

D.Blair FILS. del et lith.

PALAACHLYA PERFORANS.

PARASITIC WITHIN SILURIAN AND TERTIARY CORALS. ait

Fig. 5. Tube with conidia, from an American Lower-Silurian Foraminifer, x

400.

6. Tube near the edge of Goniophyllum pyramidale, showing the entrance
of the tube from the outside, x 350.

7. Tubes in Goniophyllum py pee alternately dark and light.

8. Tubes with conidia in Goniophyllum pyramidale, xX 400.

9. Tube of moniliform appearance from Goniophyllum, x 350.

10. Tubes cut across, showing their lumen, from Goniophyllum, 350.

11. Tubes in Calceola sandalina, from the Devonian, X 350.

12. Large tube, opening at the surface of Calceola sandalina, reaching in
but a short distance, and containing oospores, some of which have
germinated and formed canals, which radiate from the end of the
tube, x 400.

13. A similar tube from Caleeola sandalina, containing oospores, x 400.

14. Tube with spherical spore-sac, from Goniophyllum pyramidale, X 350.

Discussion.

Prof. Morris expressed his agreement with the author’s con-
cluding remarks, and thought that the discovery of these low para-
sitic organisms so far back in geological time was especially interest-
ing, as showing the identity of conditions in all periods. He
remarked that another tubular structure different from those de-
scribed by Dr. Duncan had been observed by M‘Coy in Pterinea de-
missa, from the Wenlock Limestone, and referred to Cliona. This
latter structure had also been observed in some Jurassic and Creta-
ceous Mollusca.

Mr. Erneriper agreed with Prof. Duncan in regarding these
parasites as Algze, and remarked especially the similarity of the
phenomena to those presented by many freshwater Alge, referring
especially to Vaucheria. The continuity of conditions of life thus
manifested was, he thought, very remarkable. He stated that
numerous sections of Scandinavian paleozoic Corals given to him by
Sir Roderick Murchison were full of tubular forms like those de-
scribed in the paper.

Prof. Ramsay asked Prof. Duncan to state how much of the con-
tents of his paper was to be regarded as new.

The AvrHor, in reply, suggested that a similar question might
just as well be put to the author of every paper. His new points
were the establishment of the extension of these parasitic organisms
to a much earlier period than any in which they had previously been
recognized, especially their occurrence in the very ancient Corals to
which he had referred, and certain particulars as to their characters
and mode of occurrence. The parasitic Alge could not be con-
founded with such sponges as that found by M‘Coy or with Clona,
as their tubules were very much smaller than those formed by the
sponge; in fact the tubules of the parasitic plants had about the
same diameter as the spicules of Cliona.

212 W. J. HARRISON ON THE OCCURRENCE

24. On the OccuRRENCE of the Ruaric Beps in LeicesTERSHIRE.
By Wm. J. Harrison, Esq., F.G.S., Curator of the Town
Museum, Leicester. (Read March 8, 1876.)

Goop inland sections, exhibiting the junction of the Triassic and
Liassic beds, are rather scarce in this country. I wish, therefore,
to describe an exposure of the Rhetic beds near this town, which
is of interest as proving the continuity of that formation and the
remarkable persistence of lithological conditions in its strata; whilst
the occurrence of some new species of fossils shows that our know-
ledge of the life of that period is, as yet, very incomplete.

These Rhetic beds are to be seen in three brick-pits situated at
the northern extremity of the Spinney Hills, a low range forming
the eastern boundary of the town of Leicester and the Soar valley
(fig. 2).

The Rheetics form a capping to the hills at this northern end; but
southwards, as the ridge rises, they are overlain by a thick covering
of drift to the depth of at least 20 or 30 feet.

Eastwards the upper members have been denuded by a little
stream, the Willow Brook, although the lowest bed (the Grey Marl)
is nearly, but not quite, continuous right across to Crown Hill,
where Lower Lias beds (yellow fucoidal limestones) first put in an
appearance (see fig. 2).

The floor of the brick-pits just mentioned is about 10 feet deep in
red Upper Keuper Marls. Descending sections in neighbouring
pits and in the railway-cutting near the station show an alterna-
tion of Red, Grey, and Blue Marls to a depth of from 80 to 100 feet.
It is noticeable that the relative thickness of the red bands becomes
less as we approach the top. Selenitic crystals and salt pseudo-
morphs occur in these beds; but they have as yet exhibited no traces
of life. A thick nodular band of gypsum occurs about 60 feet down.

Close to the eastern foot of the Spinney Hills a boring for coal has
reached the depth of 741 feet (fig. 2). This commences just below the
Rheetic Grey Marl, and passes through 690 feet of Keuper Red Marls
containing much fibrous gypsum, and in the lower part thick red
clays, and then enters a bed of sandstone, through which it is now
passing. It would thus seem probable that these Triassic beds
thicken in this direction; and as we are receding from Charnwood
Forest (a Triassic island), this would be a likely consequence. - Of
the 51 feet of Lower Keuper Sandstone, the first 20 feet is described
by Mr. J. A. Bosworth, F.G.S., the engineer, as a perfect quicksand,
no solid cores being obtained. »At Hinckley, twelve miles to the
S.W., the Red Marls are of the same thickness, as proved by a
borehole for water, of which a good supply was obtained, but too
much impregnated with mineral matters for use; the same thickness
of marl was also proved at Rugby, where the water was equally bad.

In the brick-pits the Rhetics are seen to rise nearly vertically for
about 30 feet above the Red Marl, to which their stratification is

OF RHZTIC BEDS IN LEICESTERSHIRE. 213

Fig. 1.— Vertical Section of Rhetic Beds, Spinney Hills, near

Leicester.
Lithological character. Thickness. Fossils &c.
ft. in N= . . eons
Z ZS Fates. oan Flint implements (Neolithic),
10. Soil and drift ......-...+0-+. 10 |esfe® FESS 2 ie ee ey
9. madgler band of nee 0 6 OPO? CO*N .sEstheria (casts), Avicula con-
MUG? Seiceufcinss ecccst=s ————————————

SS eS SSS torta.

8, Light-coloured apa
with sandy partings .

cs
Oo

Cardium rheticum.

TH

ii

\
aS — ——— | Modiola minima,
es I? Avicula contorta,

PO at aad Damesii,

eeere seer with sat kc } Pholidophorus Mottiana.

PATTINGS ....--.ereeeeee es
6. Finely-laminated black t 20 Meitube contertes
NEBILDH cio hacss>-snaecnenes Ticiiaes #ickcaien
Be Sandstone .........0.0.-.000-s 0 5 il, Aeie Shh aaa:
4. Black shales, orely & 2 6 ( Gyrolepis, Saurichthys api-
Taminated ........:..0-+ calis, Hybodus minor,
0 3 Ceratodus, Nemacanthus
3. Bone-bed ...........:ecseeees

Plesiosaurus, Coprolites,
Sargodon tomicus, Acro-
| dus, Axinus elongatua, A,
depressus.

monilifer, Ichthyosaurus,

2. Light buff sandy mars, :
Fish-scales, Selenite, Salt

with hard courses ;
bluish nodules in sills pseudomorphs.
lower part..............-

Selenite,

1. Red and blue pet 10 0

Keuper Marls ......... Salt pseudomorphs.

| i!
. ! | | | i | |
ee eS. —e———————————E——————— eee

214 W. J. HARRISON ON THE OCCURRENCE

parallel. ‘The upper surface of the Red Marl, however, is somewhat
uneven, being hollowed out here and therein long curves. There isa
slight dip to the south-east. The bed which I consider to be the
lowest member of the Rheetics is No. 2 in the accompanying section
(fig. 1). It isa light buff-coloured hard sandy marl, some 16 feet in
thickness, often of a greenish tint, with blue nodules here and there
in the lower part. Itis very much fissured, and has a conchoidal
fracture. It is traversed by three or four courses of harder, whiter
stone, which weather out from the rest, and are more laminated.
Crystals of selenite are plentiful in this bed; and I have also speci-
mens from it of pseudomorphic salt-crystals, and a slab showing
ripple-marks. Small fish-scales are numerous, and dark brown
markings as of vegetable matter. I have obtained a single insect-
wing from this bed. Pittings, as of rain-drops, occur in all parts,
and the upper surface is very uneven. This lowest Rheetic bed ex-
tends for some distance to the north and east of the Spinney Hills.

On the uneven upper surface just mentioned rests the bone-bed,
a stratum not more than 2 or 3 inches in thickness. This bed is
not easy to note on the “face” of the working, which is usually in
avery “mashy” soft condition; and I had much difficulty in finding
it. On digging back some distance it becomes comparatively hard ;
but its contents are then so brittle as to be very difficult of extrac-
tion. For the detection of the numerous small teeth, scales, &c.
which it contains I have found the best plan to be to take home
baskets full, which can then be sorted and examined at leisure with
a lens. From this bone-bed I have obtained large vertebre of
Ichthyosaurus, numerous fragments of rib-bones, one probably of
Plesiosaurus, about 18 inches long, together with numerous unde-
termined bones, some of which have a Labyrinthodont character.

Spines of Nemacanthus monilifer and Hybodus minor, teeth, scales,
&e. of Hybodus, Acrodus, Saurichthys apicalis, Sargodon tomicus,
Gyrolepis, Ceratodus, large Saurian teeth, worn and rolled bones,
phosphatic nodules, coprolites containing fish-scales, &c. are all of
more or less frequent occurrence in this bed. I have found in some
coprolites small quartz pebbles, and fragments of sandstone pre-
cisely resembling the Upper Keuper Sandstone of the Dane Hills,
on the opposite side of the Soar valley.

Pebbles of all sizes are numerous in this bone-bed. They are
mostly quartzose or slaty, and rounded; many are 3 or 4 inches
in length; and I should refer them all to our Charnwood-Forest
rocks. All the bones are highly mineralized, being heavy, and of
a brownish colour. Small concavo-convex bodies are common, much
like Discina in appearance; these are probably the ends of bi-
concave vertebre, of which the central part has decayed away.

Two small bivalve shells (Awinus elongatus and A. depressus) occur
loose in the bone-bed; they also occur in the “ flinty bed” at Beer
Crowcombe*.

Next above the bone-bed come about 2 feet 6 inches of coarse
black shales (No. 4 in section), very pyritous, the fossils being

* Quart. Journ. Geol. Soe. vol. xvii. p. 503.

OF RH-ELTIC BEDS IN LEICESTERSHIRE. 215

decomposed ; and these are overlain by a thin irregular band of hard
reddish sandstone from 3 an inch to 1 inch thick, whose surface is
covered with casts of Awinus (formerly called Pullastra).

Then come about 2 feet of finely laminated black shales (No. 6).
It was this bed which, in February 1874, yielded me the first fossil
evidence (Cardium rheticum and Avicula contorta) by which I was
enabled to prove these beds to be of Rhetic age.

From the next stratum (No. 7) in March 1874, I got a Starfish, the
first found in British Rhetics. Curiously enough, the same species
occurred, about that time, to Professor F. Romer in the Rhetic beds
of Hildesheim *, whilst it has since been found by Mr. G. Embrey at
Westbury-on-Severn. It has been determined by Dr. Wright to be
his Ophiolepis Damesii. I have recently found a thin band in the
shales almost made up of the remains of these beautiful Starfishes,
their joints occurring by thousands. There are apparently at least
two distinct species.

This bed (No. 7) consists of rather dark shales with sandy partings.
Here I also found a new species of Pholidophorus, which I propose,
provisionally, to name P. Mottiana, after my friend Mr. F. T.
Mott, President of the Leicester Literary and Philosophical Society.
Avicula contorta and Cardium rheticum also occur. There are, too,
some curious oval markings, with fine striz radiating from the centre ;
but these may be inorganic. Worm-tracks are numerous. A white
amorphous mineral, Kaolinite, with a little bitumen, fills many fissures .
in these beds of shale ; and the cavities left by radiating selenite crystals
cover the surfaces in great abundance. Flakes of mica spangle many
of the sandy partings which occur in the upper part.

The uppermost shales seen in the section (No. 8) are very light
in colour, and about 4 feet thick. All the shells already men-
tioned occur in them, together with Modiola minima.

Indications of a bed of hard rubbly limestone are to be seen
capping the brick-pit sections. Drainage-operations higher up the
crest of the Spinney Hills, in connexion with new streets to be
built there, have, however, lately offered a good opportunity. of
examining beds somewhat higher in the series than those already
noticed. |

The bed of limestone (No. 9) is nodular, the nodules occurring at
intervals of a foot or more. They are intensely hard, but soon
break up into cubical masses on exposure to the air, being traversed
in all directions by cracks filled with calcite. Ihave recognized this
limestone as entering into the composition of Roman pavements
found in Leicester; it would, in fact, present ready-made tesserz to
the hand of the artisan. It has a conchoidal fracture, and is of a
bluish tint, but grey on the outside. Fossils are very rare in it; but
I have found casts of Estheria minuta and Avicula contorta on the
outer surfaces.

A second nodular bed of limestone exists, I believe, about 2 feet
above the one just mentioned, and then beds of light-coloured clay
and sand; but here the drift obscures the section, and, as it thickens

* Zeitschr. d. deutsch. geol. Ges. 1874.
Q

216 W. J. HARRISON ON THE OCCURRENCE

rapidly along the crest of the hill, prevents further investigation.
The existence, however, of beds higher in the series is shown by the
occurrence of blocks of limestone in the drift, with Monotis decussata
and Anoplophora musculoides.

I have found further traces of the Rheetics at a point about a mile
north of the Spinney Hills, near the site of the Borough Asylum ;
and I believe Mr. R. Etheridge, F.R.S., has noted faeinl occurrence
at a spot about 9 miles further north between Barrow and Sileby.
I noted them here during the widening of the Midland main line in
1873, and the section is mentioned by Mr. H. B. Woodward (Geol.
Mag. 1874, p. 480).

Still further to the north-east the same beds have been cut through
near Stanton by the railway now in course of construction from
Nottingham to Melton ; but here, again, the higher Rheetic beds seem
to have been much denuded. The nearest southern exposure seems
to be at Copt Heath, near Knowle, in Warwickshire, as noted by the
Rev. P. B. Brodie (Quart. Journ. Geol. Soc. vol. xxx. p. 746).

In the deep boring for water at Rugby the Rheetics were reached
at a depth of about 450 feet. The black shales were there 8 feet
thick, and the hard sandy marls beneath about 10 feet.

Although good sections of the Rhetic beds are rare in Leicester-
shire, yet the line of outcrop of the strata can be clearly traced from
Leicester northwards, forming the eastern boundary of the Soar
valley, and varying from a quarter to about half a mile in width.
Southwards the great thickness of drift makes an exact tracing
difficult.

At Crown Hill (fig. 2) I have lately detected, in the preliminary
works for the Great Northern line from Melton to Leicester, certain
beds of limestone, called in Warwickshire the ‘ Firestones” and
“Guineas,” which mark the passage of the Rheetics into the Lower
Lias, together with a thick bed which is probably true White Lias.
The cutting here will, I believe, exhibit a very interesting series of
beds, and enable us to trace the succession for a long way up.

Wherever the true junction of the Lias and Trias is exposed in
this country, the Rheetic beds appear to be invariably present. In
1874 I examined a fine section exposed in widening the Midland
main line between Leicester and Wigston, which seemed to contra-
dict this; Lower Lias Shales and Limestones, with Ammonites
planorbis, appeared to be resting on Upper Keuper Marls. A close
examination of the latter, however, showed them to be recomposed
stuff; the Rhetics had been eroded, probably by glacial action, and
their place filled with transported material.

I am much indebted to the Rev. P. B. Brodie, Messrs. W. H.
Hudleston, W. Davies, L. C. Miall, C. Moore, and others, for their
kind assistance in the determination of specimens.

The students in my science classes, especially Messrs. J. E. Elgood,
J. R. Plant, L. H. Llewellyn, and W. J. Harrison, jun., have ren-
dered valuable aii in searching the beds.

~ 8
a -_
¢ os
“25
Be"
oe
|
oO
ro
i=]
es)
ee
& !
a
ou
=
c
®

“peq-ou0g ‘9

“‘so/ eqs DJL0ZU0I-VINIAP “Pp

(oxy) SMOFSOUTL] AVINPON “2

ta Ss
a
ne &
Boos
PES
é
i=

4
S

=}

£

b<-]

=)
‘fu

i]
}

«fe

{! HABER |

il
y i
|].

|

ale

efefed ede pede

5 = * :_— =

|
|
I
i

|:

SS

=
—-
=~
—

—— ry
—

.

i

ti

it.

!

—T
—
= =

Fl

I:

‘:} py
HABA RRA

|

= =>

= Ta

— — 5
—

Hh eL Es
fh} ; | j | : Spinney Hills.

OF RHZTIC BEDS IN LEICESTERSHIRE.

[i
, R Hah oe Sr ae Leicester.

i iH

he co

LY Eby ij
leit i
IRL HE il)

~--- §8t. Saviour’s Church.

(6

ih

= = Se ae Boring for Coal.
HELE -—---—~—-—-~—----- Willow Brook.
HH eS
|
Whe
Wil S
Ht 3
hi 9
a Ee

“aaysanary yo ‘yoorg mopjrsy 2Y3 fo hayn, ay) ss019v uorjmg woibng—Z “Siz

218 ON THE OCCURRENCE OF RHAZTIC BEDS IN LEICESTERSHIRE.

Discussion.

Prof. T. Ruerrt Jonzs referred to Mr. Harrison’s exactness and
praiseworthy energy in working out the geology of Leicestershire,
and also alluded to the enlightened and liberal support of science on
the part of the Corporation of Leicester. He remarked that some
of the bones found in the Bone-bed of the Spinney Hills were of
large size and apparently perfect.

Mr. Jupp pointed out the great interest and value of this new
exposure of Rheetic strata which had been so admirably described in
Mr. Harrison’s paper. | Hitherto we had no knowledge of any
sections of the beds of this age between those of the Warwickshire
outliers described by the Rev. P. B. Brodie and that seen at
Newark.

Prof. SrELEy wished to correct an error into which he had fallen
in describing a species of Tanystropheus from Leicester. He had
ascribed it to the Rhetic beds; but it was really from the Keuper.

ON THE PHYSICAL HISTORY OF THE DEE, WALES. 219

25. On the Puysicat History of the Dez, Wars. By Professor A.
C. Ramsay, LL.D., F.R.S., V.P.G.S. (Read April 26, 1876.)

[Prats XIV.]

Tue surface of Bala Lake is 600 feet above the level of the sea, and
is bounded on the north-west and south-east by hills that near the
lake have elevations of about 1200 and 1300 feet, and therefore rise
to heights of 600 and 700 feet above its level, while at greater dis-
tances, but still not far off, they rise to heights of 1800 and 1900
feet. On the south-west there is a long gentle slope, which attains
a height of about 800 feet above the sea at the watershed close by
the road to Dolgelli, or 200 feet above the lake, beyond which the
river Wnion drains the country in a south-westerly direction into
the estuary of the Mawddach.

The lake is about 130 feet in depth; and at its north-eastern end
the waters of the Dee, escaping from this basin, first traverse an
alluvial tract nearly three miles in length, the detritus of which has
been carried from the adjoining hills by rivers, the largest of which
enters the Dee near Pont Treweryn, about half a mile below Bala.
Other good-sized streams form tributaries of the Dee below this
point, and have helped to distribute the materials of the alluvial
plain, beyond which, for a space of about a mile, the hills close in
upon the banks of the river between Glan-dwynant and Pont Llan-
derfel. Between this point and the neighbourhood of Corwen, the
river, for nfore than six miles, again wanders through broad allu-
vial flats, after which it flows in many curvatures, sometimes making
great loops, between the steep slopes of the vale of Llangollen, which
form the high and often flat-topped hills of Wenlock shale and slate,
that sometimes, not far from the river, slope up to 800 and 1200
feet above its banks.

Leaving this part of the valley, the Dee, with an exceedingly tor-
tuous course, cuts through the marked escarpment of the Carboni-
ferous rocks of the Denbighshire coal-fields east of Llangollen, and,
passing through a less elevated tract of Permian strata, emerges on
the low undulating country of New Red sandstone south of Chester,
and finally, by a winding channel, enters the large tidal estuary
known as that of the Dee.

Emerging from the Permian region at Erbistock, the country
through which the Dee flows to Chester may still be called a valley,
in so far that, miles to the west, there rises the long slope of Per-
mian and Carboniferous strata, while the eastern boundary is formed
of the escarpments of Keuper Sandstone of the Peckforton Hills
and Delamere Forest. Eliminating the minor modern effect on the
scenery produced by a shroud of glacial drift, with other geologists
I do not doubt that before the Glacial epoch the contours of this
wide valley were much the same as now, or, in other words, that,
like the valley of the Mersey and others in the west of England,

a. 0.G. 5. No. 127. R

220 A. C. RAMSAY ON THE PHYSICAL

it is preglacial. The broad features of the history of this part of
the valley are sufficiently obvious; but from Erbistock to the source
of the river in Bala Lake, the case is different; and the physical
history of that long and often steep-sided river-valley has, as far
as I know, not been analyzed.

At Holt, seven miles south of Chester, the river is about 100 feet
above the level of the sea; near Erbistock, eight miles further
south, where the valley enters the Permian strata, about 225 feet ;
in the vale of Llangollen, four miles west of the town, about 475
feet; near Corwen, about 500 feet; while the level of the lake where
the Dee rises is only 600 feet above the level of the sea. The fall,
therefore, from Erbistock to Chester, where the river is tidal, is
about 225 feet in a north and south distance of 15 miles as the crow
flies, and from Bala Lake to Erbistock about 375 feet on a direct
line nearly east and west; so that the average slope is nearly the
same the whole way, or about 15 feet per mile, taking no account
of the numerous minor curvatures of the river.

I have said in previous memoirs that the main great contours of
Wales were approximately the same as now, before the beginning of
the Glacial epoch, and that all the glacial striations of the mountain
regious of North Wales correspond to the trend of the greater
valleys. The chief effect of the glaciers was therefore to grind and
smooth the rocks into moutonnée forms in the bottoms of the valleys
and on the sides of the mountains, and in doing this more or less to
deepen and widen these valleys, and here and there to scoop out
those rock-bound lake-basins so common in all old intensely gla-
ciated regions, of which Bala Lake is one. The proof of its being a
true rock-basin is simple.

At present the outflow of the lake is through an alluvial flat, the
western end of which helps to dam up the water. But this allu-
vium, as already shown, is the gift of the tributary rivers that here
enter the Dee. In older times the lake was more than double its
present Jength ; for while a mile and a half of delta at its south-
western end, where the river Wnion enters, marks its ancient extent
in that direction, the sheet of water also once extended 22 miles
east of Bala over the area now occupied by alluvium, giving the
ancient lake a curved form, before torrents and rivers from the north
and south filled that space with detritus borne down valleys that
penetrate far in among the hills. The effect of this is, that the
lake now is little more that 34 miles long, while formerly it was
8 miles in length.

Bala Lake being 600 feet above the sea, and about 130 feet deep,
its bottom is 470 feet above the sea-level: and probably its rock-
bottom may be deeper still; for two large and three small streams
enter the lake on either side as it now stands, in addition to the
main rivers of Afon Llyn, and Dwfrdwy, which have deposited the
delta at its south-western end; and doubtless the bottom of the
lake, by means of these rivers, may be covered with detrital matter.

But even assuming that the present depth of the lake represents
its rocky bottom, the fall of the river is only a few feet in the whole

HISTORY OF THE DEE, WALES. 224

of the alluvial tract at its eastern end. Beyond this the Dee runs
through a narrow valley, with solid rocks down, or almost down,
to its banks, where in older times the original outflow of the
Lake was; and it may be considered certain that the old point
of outflow is a true rock barrier. If, however, there should still
be any doubt about this gorge having formed the original rocky
dam of Bala Lake, there remains other evidence on the question of
its being a true rock-basin; for, about two miles further on towards
Llandrillo, the rocky bottom of the river is from 500 to 510 feet
above the sea, while the bottom of the lake is only 470 feet above
that level; and even beyond this, at the ferry halfway between
Corwen and Llangollen, 14 miles in a straight line E.N.E. of the
lake, the rocky bottom of the river is about 475 feet above the sea,
or 5 feet above the bottom of Bala Lake. As there is not, and never
could have been, any other possible outlet for the water of the lake
than the channel of the Dee, it is quite certain that it lies in a rock-
bound basin; and indeed it seems to me more than likely that the
large alluvial flat more than six miles in length between Llandrillo
and Corwen may conceal one or more lake-basins, now filled with
detritus brought down to the hollow by the numerous streams, large
and small, that, over this space, now form tributaries of the Dee*.
Similar alluvial flats, once lake-basins, are common in many old
glacier-valleys in Switzerland, Cumberland, Scotland, and elsewhere,

The glaciers that did this work of erosion came chiefly from both
sides of Aran Mowddwy, down Cwm Croes and the valley of
Dwfrdwy; and these converging near where Llanuwchllyn now
stands, and aided by other tributary ice streams, scooped out the
basin of Bala Lake, and went far beyond its ancient boundary, in a
valley older than the beginning of the Glacial epoch, like all the other
great valleys of Wales.

Such being certainly the case with this upper part of the valley of
the Dee, it is natural, and indeed inevitable, to suppose that the
further continuation of the valley towards and beyond Llangollen is
also of preglacial origin; and this, in my opinion, seems to be
strengthened by the erratic blocks and glacial striations mentioned
by Mr. Mackintosh in a memoir on boulders and erratics in Denbigh-
shire, even though the author seems doubtfully to attribute some of
these striations to the action of floating icer. If, for the sake of
argument, we assume that this part of the valley (unlike the upper
part) is not preglacial, then the mass of country through which it
runs must have formed a dam to the drainage of the waters of the
upper Dee, which would have had the effect, even if these waters
were only raised 200 feet, of filling with a lake ramifying in all di-
rections among the hills, vast tracts of Merionethshire. Of this
supposititious lake there is no sign whatever.

* All the heights mentioned above were determined by anguiar measurements
with the theodolite by the late Professor Jukes, Mr. Aveline, and others. They
are therefore not to be depended on within a few feet; but any possible errors
of this kind are too small to affect the general argument.

+ Quart. Journ. Geol. Soc. vol. xxx. p. 711 (1874).

R 2

222 A. C. RAMSAY ON THE PHYSICAL

Beyond Erbistock, where the Dee leaves the hills, and all down
the estuary, the valley is undoubtedly preglacial, in so far that the
country is largely shrouded by glacial drifts.

The question thus arises, by what means was the valley of the Dee
excavated, and if it be possible to discover during what geological
epoch that excavation began.

The valley, in some aspects, may be compared to that of the Mo-
selle, the physical character of which I have elsewhere indicated*;
for, especially east of Corwen, between banks of Wenlock shale or
slate, that form a high and much-croded tableland, the Dee winds
in many a curve, two of which, about two miles above Llangollen,
form great loops, closely resembling the well-known loops of the
Moselle above Alf. In both rivers a steep rocky high bank is apt
to be opposed to the concave curve of the stream, while, on the
opposite, or convex side, the ground generally rises from the river
with a long and gentle slope. This form of ground is characteristic
of many rivers that flow through deep valleys, the elevated ground of
which on either side is more or less flat-topped, and invariably in-
dicates a long-continued course of fluviatile erosion, begun at first by
a wandering river on a high and slightly undulating tableland or
plain.

The Silurian rocks of the region now called North Wales un-
derwent great disturbance and erosion, both atmospheric and marine,
both before and during the deposition of the Carboniferous Lime-
stone—one result of this being that as the country was in great
part slowly submerged the limestone series was laid in a very
unconformable manner on the denuded edges of both the Lower and
Upper Silurian strata.

The high escarpment of Carboniferous Limestone in Denbighshire
facing west clearly shows that this limestone once extended a great
deal further in that direction across much of the Lower Silurian
rocks of the Berwyn Hills and the Wenlock slates and shales that
form the high land on either side of the valley of the Dee. This
fact, if it needed confirmation, is further proved by the well-
known outlier of Carboniferous Limestone that lies, far in among
the hills in the valley of the Dee, two miles west of Corwen, ten
miles distant from the nearest point of the escarpment, and six
miles from the limestone at the southern end of the Vale of Clwyd.
This little patch, which occupies an area of something less than
three quarters of a square mile, lies quite unconformably on the
Wenlock shale on its south-western side, while on the south-east
and north-east it is, as it were, dovetailed into the Silurian strata
by two faults; and as its thickness, according to the late Professor
Jukes, may be about 2000 feet, it is clear that the Carboniferous
Limestone, in force, once spread well across this and the neighbour-
ing region.

Furthermore, everywhere to the east in Denbighshire and Flint-
shire, the limestone is overlain by the Millstone Grit and Coal-

* Physical Geology and Geography of Great Britain, chap. 15.

HISTORY OF THE DEE, WALES. 223

measures ; and I have no doubt that these upper Carboniferous strata
at one time also spread far to the westward above the limestone
across those parts of the Silurian strata now under review; but how
far,it may perhaps be impossible to determine. I feel certain, how-
ever, that what is now the high mountain region of Snowdonia,
at that time very different from its present form, was never,
except on its low western outskirts, covered by Carboniferous strata.

The statement that so much of Wales was once buried beneath
Carboniferous rocks is, at first sight, somewhat startling, involving,
as it does, such large revolutions in the physical geography of the
district in early times.

First. An immense amount of denudation undergone by the Silu-
rian rock before the deposition of the Upper Old Red Sandstone
(which sparingly exists in North Wales) and the Carboniferous
Limestone*.

Secondly. It appears to me that this waste was accompanied by a
gradual depression of the old land, so that, over part of Wales, what,
many years ago, I termed ‘‘a plain of marine denudation” was pro-
duced, similar to that well-known high plain of Carboniferous Lime-
stone and Old Red Sandstone of the Mendip Hills, on which the
Liassic and Oolitic strata lie unconformably. In like manner the
Carboniferous Limestone overlain by Millstone Grit and Coal-mea-
sures, covered in flat-lying strata the much older Silurian plain
of marine denudation.

Thirdly. By succeeding changes of physical geography after the
deposition of the Millstone Grit, the region was converted into part
of one of those vast deltas, or at all events broad swamps, under-
going slow intermittent depressions, during which the Coal-measures,
with all their successive terrestrial growths, were formed.

Fourthly. The region was again disturbed and elevated before
and during the deposition of the Permian strata, which, taken on a
large scale, often lie unconformably on the Coal-measures, and were,
in my opinion, true continental deposits. Then by slow degrees,
under subaerial agencies, a vast amount of the Carboniferous rocks
were removed from above the Silurian formations, and the plain
of marine denudation was again to a great extent exposed, not
less high, but probably even more elevated than it is now, lying as
it then did in the midst of a broad continental area.

This plain (or, rather, its relics) is of much wider extent than that
part of it with which I am now concerned, lying on either side of
the valley of the Dee; for beyond that area it extends far and wide
into the northern part of Denbighshire, and also into Montgomery-
shire and South Wales. Standing on the summit of Cader Idris or
of Aran Mowddwy, 2960 feet high, and looking east and south, the
eye, as far as it can reach, ranges across a vast extent of old table-

* This amount could be approximately estimated by any one who would take
the pains to restore the pre-Carboniferous anticlinal curves of the Silurian
strata; but the doing of this in detail is not essential to my present argument.
For the method see memoir ‘‘On the denudation of South Wales,” ete., Memoirs
of the Geological Survey, vol. i. 1846.

224: A. C. RAMSAY ON THE PHYSICAL

land, the plane surface of which near the Arans is about 1900 feet
above the level of the sea, or more than 1000 feet below the summits
of the neighbouring mountains. All intersected by unnumbered
valleys, to the ordinary observer it is merely a hilly country, while
an eye versed in physical geology at once recognizes that all the
diversities of feature are due to fluviatile erosions that have scooped
out the valleys.

This brings me to the main object of this memoir, the origin and
geological date of the valley of the Dee.

The country having been upheaved after the deposition of the
Carboniferous strata, and a large part of these strata having been
removed by denudation, the valley of the Dee did not then exist, or
existed only in a very rudimentary stage on the surface of a wide
slightly undulating tableland having a gentle slope towards the east.
It was then that first the Permian rocks were deposited, and after-
wards the New Red Sandstone and Marl; and it may be recollected
that, at all events in my opinion, both sets of these formations were
deposited in inland salt lakes or seas. During these epochs, there-
fore, I hold that the old tableland lay high above the level of the
sea, probably even higher than it does now, and that it formed part
of a wide continental area.

After the Trias came the Jurassic formations, deposited around
groups of islands in shallow seas, the effect of partial submergence
of an older land, and themselves, in the middle and north of England,
partly of fluviatile origin, the result of minor oscillations of level.
Then followed the long fluviatile conditions under which the Purbeck
and Wealden beds were deposited, during which epoch the greater
part of what is now Britain must have remained as part. of a much
larger land. After this came the Cretaceous formations, the lower
half of which in the British area were certainly deposited near
shore ; while in later times, during the deposition of the Chalk, there
is no proof that the higher parts of Wales were submerged, but
rather the opposite. Later still come the Eocene deposits, some of
them of freshwater origin, and which in a large sense may be re-
garded as estuarine, even including the London Clay, which seems
to have been formed at the mouth of a great river such as the
Amazons or the Ganges. Our Miocene rocks are unmistakably of
terrestrial and freshwater origin; and the different members of the
Crag were mere eastern shallow-sea or shore deposits.

It thus appears that, since the beginning of the Permian epoch,
the higher ground of Wales has formed land well raised above the
level of the sea; and even if there be some doubt about this with
respect to the time of the deposition of part of the Chalk (to which
doubt I do not attach much importance), that must have been but a
short episode when compared with the whole of the terrestrial
period.

During that long lapse of geological ages, there was therefore
ample time for the action of all the ordinary processes of subaerial
denudation, the most powerful of which is the action of rain, rivers,
and glaciers.

HISTORY OF THE DEF, WALES. 225

In the most elevated parts of North Wales, the height of the
highest mountains as they now stand between Cader Idris, Aran
Mowddwy, and Snowdonia varies from about 2900 to 3571 feet.
During the time of submergence, when the ground further east was
buried under Carboniferous strata, the western mountain region
stood well above the level of the sea, and higher relatively to the
plain of denudation than it does now; for long before the Carboni-
ferous epoch, and ever since that time, it has suffered from un-
ceasing subaerial degradation, and long-lost hills and valleys must
have prevailed in that old land, the shape of which it is now vain to
speculate about.

Outside of this region, after the partial removal of the Carboni-
ferous rocks, there lay, as already said, a wide tableland, extending
far to the south, and also to the east and north-east. In the latter
district this tableland sloped gently north-easterly and easterly ;
judging by those elevated parts of it that still remain more or less
entire, in a distance of about 30 miles between Aran Mowddwy and
the highest part of the Carboniferous escarpment, not far north of
the Dee, the slope probably fell in height 700 or 800 feet, at an
angle of about 14°, equal to 23 feet in a mile.

When by the drainage of this old land, the Dee began to flow in
its earliest channel, induced by minor undulations of the ground, it
is clear that its present source, Bala Lake, had no existence ; for
whereas the river at that time must have flowed on a surface of
land not less high than that on either side of the present valley near
Corwen and Llangollen (now, in places, from 1600 to 1800 feet high),
the surface of Bala Lake is only 600 feet above the level of the sea,
while the neighbouring watershed between the lake and Dolgelli is
only 200 feet higher. As the river could not flow up hill, it is clear
that in that early stage of its history the valley of the Dee about
Bala must have been at least from 1300 to 1400 feet higher than it
is now, and consisted of a mass of Silurian rocks, great part of
which has since been removed by denudation.

As that country, according to the view already stated, has in the
main been land ever since the Permian epoch, it seems to me that
this waste of rocks was essentially produced by watery erosion, aided
in a much less degree at a late epoch by glacier ice; and thus
it happened that the Dee, a river of very ancient date, wandering
hither and thither, by degrees deepened its channel in the same
manner that the Rhine and the tortuous Moselle have cut out theirs,
as described in my memoir “ On the Physical History of the Valley
of the Rhine”*.

For this reason it also happens that the Dee now cuts right
across the Carboniferous escarpment west of Erbistock and the
lower area of the Permian strata; for when the Dee began to
run, the Carboniferous escarpment had no existence, and the strata
of these formations stretched further to the west, ending along some
line now unknown in a sort of feather edge, and forming part of the

* Quart. Journ. Geol. Soc. vol. xxx. (1874), p. 81.

226 A. C, RAMSAY ON THE PHYSICAL

great inclined plane over which the Dee ran at a level hundreds of
feet above the bottom of its present valley. By and by, as the river-
channel deepened, the escarpment began to be formed, its face
sloping in a direction at right angles to the general dip of the
strata (after the habit of all such escarpments), by a process of
recession that has often been-explained. The whole was strictly
analogous to the manner in which the rivers of the Weald acted at
a later date, as described in ‘The Physical Geology and Geo-
graphy of Great Britain ’—and also for the same reason that the
Thames now cuts across the escarpment of the Chalk, as described
in my memoir ‘On the River-courses of England and Wales’*.
Escaped into the low country of the New Red series, the history of
the Dee becomes simple, and requires no special illustration.

But this process of ordinary fluviatile erosion is not the only
agent that has been at work in Wales; for in late geological times
the Glacial epoch supervened, and the moving ice of thick glaciers
exercised a strong abrading power. ‘Then it was that in the moun-
tain-region of the west so many lake-basins were scooped out, and
among others the rock-bound basin of Bala Lake; and though the
face of the country is always being slowly changed, the time that
has elapsed since the close of the Glacial epoch is compara-
tively so short that the large essential rocky features of the region
traversed by the river have since that time undergone no important
alteration.

I shall conclude with a few words about the length of geological
time occupied in the excavation of the valley.

It may be remembered that in the paper on the excavation of the
valley and gorge of the Rhine, I explained that the work of scooping
it down to the present levels of the river has been accomplished
since the close of the Miocene epoch, and that the direction of the
flow of its water then and now is connected with changes of physical
geography brought about by a great post-Miocene disturbance of
the Alps.

The height of the old high-level terrace or old bed of the Rhine
is about 400 feet above the present river; and the same is the case
with its tributary, the Moselle. The height of the relics of the old
plain of marine denudation above the Dee where it passes through
the Wenlock strata is, to take a low estimate, not less than from
1100 to 1200 feet; and the time occupied in cutting that depth
extends from the Permian epoch down to the present day. The
depth excavated in that period of time is not less than three times
as great as that performed by the Rhine and Moselle; and I think
no geologist will doubt that the time that elapsed between the
Permian and Miocene epochs was much more than three times-
the length of that which has elapsed since the close of Miocene
times down to the present day. The Dee is but a petty stream
when compared with even the smaller of the two German rivers ;
and it is obvious that the greater the body of water in a river, taken

* Quart. Journ. Geol. Soc. vol. xxviii. (1872), p. 148.

HISTORY OF THE DEE, WALES. 227

in connexion with the angle of its fall, the more powerful is its
erosive force; and this may account for the proportionately much
greater amount of time taken by our little river in the work of
excavating its valley as compared with the work done by the
Rhine.

This great lapse of time also accounts for what I may call the
ruinous state of the old tableland that lies on either side of the
valley through which the Dee flows; for in the long geological ages
that the region has remained as land there has been ample time, as
the river gradually excavated its valley to lower levels, for the
drainage from the north and south to cut out tributary valleys,
which by watery erosion got deeper and deeper in proportion to the
gradual excavation of the main artery of the valley of the Dee; and
thus it happens that what was once a long slightly undulating
inclined plane has by degrees been converted into a very hilly country,
simply by the power of rain and running water, and, in a minor
degree, for a time by glacier ice.

According to these views the sum of the whole matter is :—

1. That after the last important disturbance of the pre-Permian
rocks, North Wales was carved by slow degrees and by subaerial
agencies into its present mountainous form, chiefly between Permian
times and the present day.

2. That by far the greater part of that valley-excavating work
was performed between Permian and pre-Glacial times.

3. That the work of the glaciers of the latter period some-
what deepened, widened, smoothened, and striated the outlines of
the mountains and valleys, and excavated many rock-bound lake-
basins, but on a grand scale did not effect any great changes on the
preexisting larger contours of the country.

4, That a minor submergence of part of Britain during part of
the Glacial epoch produced no important effects on the large out-
lines of the rocky scenery.

5. That the effects of subacrial waste subsequent to the Glacial
epoch have been comparatively small, simply for lack of time.

I may also remark that the method of analysis employed in this
paper is, I am convinced, applicable, with variations, to the reading
of the history of all the more important rivers of Wales, to Cum-
berland, and to much of Scotland.

Facts of a broad kind, setting forth the general principles of
watery erosion, have been well knownever since the days of Hutton—
though for long, with a large class of geologists, his views fell into
undeserved neglect. The subject, indeed, with some definiteness, is
far older than Hutton ; for it was insisted upon by Ray, who, while
the fourth edition of his ‘ Physico-Theological Discourses’* was
passing through the press, was surprised and delighted to find that
his opinion had been anticipated by Josephus Blancanus, in his book
‘De Mundi Fabrica,’ in which he stated, among other physical
matters bearing on modern geology, that ‘‘ our Appenine was at first

* Edit. 4, p. 359.

998 A. C. RAMSAY ON THE PHYSICAL

one continued, even, eminent ridge of land, not divided into parti-
cular mountains and hills by intervening valleys, as now it is, but
that after the rivers began to flow down from the top of it, by little
and little fretting and corroding the ground, they made valleys, and
daily more and more ; and by this means the whole Appenine came
to be divided into many hills and mountains.”

It is not, however, to insist on these simple principles, enlarged
by modern research, versus unnumbered unproved open faults, frac-
tures, and tumultuous splittings of mountains, that this paper has
been prepared ; but having already written on the physical history
of rivers both foreign and British, the details of which, comparatively
simple, only grew on my mind by slow degrees, I have at length
fallen on the Dee as an example of one of many rivers, the history
of which is much more complicated than, for example, that of the
Po, the Rhine, the Severn, or the Thames; and I hope I have
succeeded in unravelling that history. If so, the Dee, as a living
river, has a history compared with which those of the Thames, the
Rhine, the Ganges, and many other great rivers of the modern
world are matters of yesterday.

Discussion.

Mr. Tuorre stated that Arenig Lake, and not the Bala Lake,
was the source of the Dee, and wished to know what origin Prof.
Ramsay attributed to the former.

Mr. Hicxs considered that the author’s explanation of the origin
of the lake and the gorge through which the Dee afterwards flows
was not a correct one, and that these hills were only elevated at the
close of the Paleozoic period, and not, as maintained by the author,
in Silurian times. In a previous paper Prof. Ramsay claimed that
the Menai Straits had been excavated by a huge glacier moving
from N.E. to 8.W. ; whilst in the present communication he attri-
buted Bala Lake to a glacier moving in exactly the opposite direc-
tion. Now, as the valley of the Dee and the basin of Bala Lake
are roughly parallel to the Menai Straits, they should have been
formed by the same glacier. Neither, however, is due to glacial
action ; but both are produced by two parallel faults that were
formed at the close of the Paleeozoic period.

Prof. Srrtry inquired what grounds there were for attributing
such an ancient date to the Dee. There was no proof that the Lias
had not formerly extended over this district. With regard to this
old tableland, which the author considered to have been formed in
pre-Carboniferous times, and subsequently covered by Carboniferous
strata, he thought it was impossible that these Carboniferous strata
could have been denuded and the old tableland left intact. The
water running off the land as it rose would form the first valley ; and
both would suffer together.

Prof. Ramsay, in reply, stated that he considered Bala Lake to be
the commonly acknowledged source of the Dee. The Arenig lakes
were also due to glacial action. He had treated the disturbances of

[Zo face page 228.

E,S.E.

Coal-measures.

lillstone
| grit.

|
|

Permian.

me

.

[Vo face page 228.

Section, to scale, along the Course of the Valley of the Dee, on line C—D of Map, Plate XIV.

Rea : BSE.
W.S.W Loy ground, Near Llantisilio, =
AlUGii ORERERD Be! west of Llangollen.
Source of the Dee, : . D D D
y-Wenallt. Bala Lake. Dee, near Llandderfel. Cilan. Moel-yr-henfae, Afon Trystion. Moel-gwynt. BAR SOMES C2 Cefn-y-Pedw.

General slope of old inclined tableland_

Millstone Coal-measures.
grit.

|
} H
|

Felspathic porphyries and ashes. Bala Beds, Wenlock or Bala Beds, south of Corwen. Wenlock Shales, Slate
Denbighshire Grits. F

3, ke.

Old Red
Sandstone. -|

Permian,

Carboniferous
Limestone.

i in Ma
\ my
= ’
fi
k
i
‘ue
“ J
>
*
.
P|
3: ans OF
th ie
. .
-
+ i ° s
uv Poe ; i
i \ L 7
ad - ~“
* ate ie Bd cee a dh
’
.
“
1
My

~~

HISTORY OF THE DEE, WALES. 229

the Lower and Upper Silurian rocks as occurring at one time for the
sake of convenience, there being no unconformity locally visible in this
area. ‘There was no fault down the Menai Straits, though faults
occurred to the right and left of it. The point which Prof.
Seeley had raised was an important one, and one which it was
difficult to give a direct reply to. The Carboniferous strata, by
covering the tableland, would in a great measure tend to preserve it.
They probably thinned out towards Snowdon.

230 H. B. WOODWARD ON THE SUPERFICIAL DEPOSITS

26. Norns on the GRAVELS, SANDS, and other SurerFictaL Deposits in
the NeteHBouRHoop 0f Newron Anzot, DevonsHireE*. By Horace
B. Woopwazp, Hsq., F.G.S., of the Geological Survey of England
and Wales. (Read April 5, 1876.)

THE visitor to Newton Abbot cannot fail to be struck with the
number of sand- and gravel-pits, some of great size, which are
scattered about the neighbourhood, not only in the valleys, but
high up the slopes of the hills, and in several instances on their
summits.

On giving some attention to the character and distribution of
these deposits of sand and gravel, it will be found :—/irst, that two
main divisions may be made in them, namely those of an evi-
dently alluvial character, and those which occupy positions quite
unconnected with the present systems of drainage ; and, secondly,
that these latter deposits are nevertheless confined to the Bovey
Basin and bordering hills, not descending into any of the other
lateral or outlying valleys.

A reference to the geological maps of the country will not serve
to throw much light on the correlation or identification of the beds
when they are examined in detail. But since the days when
De la Beche and Godwin-Austen recorded their observations, many
new sections have been opened up. The rapid growth of Torquay
and Newton Abbot has been attended by a call for sand, which
is largely used in the preparation of “ cobb,” a kind of plaster,
unfortunately necessary to envelop and keep the damp out of
houses for the most part built of marble.

Diagram Section from near Newton Abbot to Milber Down.
(Distance about three miles.)

8.E.

N.W. Nees
Wool- ou
River Lemon,, borough eee Milber Down.
near Hill. allway. e
Newton Abbot. Hill south

of Keybury
Mill.

a. Gravel and sand. c. Triasste Breccia. e. Traces of Greensand.
b. Bovey clays. d. Devonian slates.

The examination of these sections, together with those dug
for the foundations of new villas on Woolborough Hill, has led me
to differ from the opinions and conclusions which have been hitherto
accepted.

Since Mr. Godwin-Austen’s paper was published f, no particular

* This paper is communicated by permission of the Director-General of the

Geological Survey of the United Kingdom.
+ Trans. Geol. Soc. 2nd series, vol. viii. p. 433.

IN THE NEIGHBOURHOOD OF NEWTON ABBOT. 231

account of these deposits has been given; but in 1867, in an
address delivered to the Devonshire Association at Barnstaple,
Mr. Pengelly observed that it may be doubted whether all the
localities so represented in the maps of the Geological Survey are
really true Greensand localities.

While engaged in carrying on the Geological Survey in the
neighbourhood of Newton Abbot (1874-75) for the new edition of
Sheet 22, I was for some time sorely perplexed with the various
sands and gravels met with. ‘To give one instance :—certain sandy
beds opened up in front of my own house were at the first glance
thought to be Triassic; a short examination led me to class them
doubtfully as Greensand; further observation caused me to regard
them as possibly Bovey beds; and before I left the country I was
persuaded that they were Drifts, using this last term in its con-
veniently indefinite sense.

The fact is, the sections vary very much in detail. There is
sand, gravel, clay, and fine sandy clay, sometimes dovetailing one
into the other, but sometimes forming individually the entire mass
of the open sectious ; so that the geologist familiar with the Green-
sand of Haldon and the Blackdown Hills may recognize some beds
similar in character to that deposit; and if he be acquainted with
the sandy strata of the Bovey formation, he may, in isolated sec-
tions, be puzzled to discriminate between the two.

The examination and comparison of the numerous fine sections,
and, above all, the tracing of the beds on the ground, prove the
convexion of deposits that might otherwise be considered distinct,
and have led me to class all the sands and gravels in the neighbour-
hood of Newton Abbot, with one or, at most, two exceptions, as cer-
tainly of an age posterior to the Bovey beds.

Beds regarded as Upper Greensand have been mapped on Wool-
borough Hill and Milber Down, at White Hill, Staple Hill, and near
Kingsteignton. At each of these localities there are beds of sand that,
on lithological grounds, might be identified with the lower beds of
the Greensand series; but, except at the very summit of Milber
Down, and perhaps also near Coombe Farm, north of Sandy Gate,
there is evidence for connecting them with the coarse gravels con-
taining fragments of Greensand chert and Chalk flint, thus indi-
cating their post-Cretaceous age; while, again, the fact that in
places the gravels and sands overlie the clays of the Bovey formation
at once indicates their age to be still more recent.

The sands are generally composed of quartz, sometimes rounded,
sometimes angular; and they graduate into gravel or shingle, com-
posed of pebbles of quartz, grit, &c.; and these pass into or become
interstratified with very coarse gravel-beds, composed of large
blocks and pebbles of grit and chert, flint and quartz. All the
fragments are apparently of local derivation.

Some sections show only sand, as at Whitehill and in the Torquay
road at Newton Abbot; others display only gravel, coarse, rolled and
angular, as shown in several pits on the summit of Milber Down.
Again, some of the largest pits exhibit all the different kinds of de-

232 H. B. WOODWARD ON THE SUPERFICIAL DEPOSITS

posit in ever varying order, and very irregularly accumulated ; such
are the pits on Woolborough Hill, near the base of the Milber Down,
and Kingsteignton. Here white clay, coarse gravel, fine and coarse
sand occur anywhere and at all horizons in the series.

The coarse quartzose sands, largely dug on Woolborough Hill,
may be traced over the Bovey clays at the Newton station ; and at
Courtenay Park I have been informed by Mr. C. D. Blake that the
clays were similarly reached beneath them.

A very variable set of beds occupies the hill-sides between Wool-
borough and Longford. One pit, dug in very fine and coarse sand,
situated on the ridge just north of Longford, might be thought to be
opened in Greensand ; but an adjoining pit, at a little lower level,
shows about 25 feet of sand, gravel, and whitish clay, with a
wedge of the very fine sand in the middle of it, which wedge is
evidently the continuation and termination of the sand worked in
the other pit. The gravel is composed mainly of quartz, grit, flint,
and chert. .

These old deposits of gravel and sand extend along the borders of
the valley from Staple Hill to Kingskerswell, and from Milber Down
to Sandy Gate and Ugbrooke Park. They occupy the old Bovey
Basin, which did not, when the beds were deposited, extend very
far eastwards of Kingsteignton, there being then no opening at the
present mouth of the Teign.

They extend in places from the tops of the hills bordering the
basin to the bottom of it, attaining between Haccombe and Coffins-
well a height of 540 feet.

The beds moreover exhibit, in places, at Whitehill, Woolborough,
and near Kingskerswell, a very remarkable dip into the valley,—a
feature observable in both sands and gravels, and one which tends
to connect the several deposits.

The position occupied by these deposits is thus very dissimilar to
that occupied by the Greensand of Haldon and the Blackdown Hills.

At Staple Hill the beds consist for the most part of sands banked
up against the Devonian and Carbonaceous slates, at such a high
angle (noticed by Mr. Godwin-Austen) that I was at one time
disposed to consider a fault, which affects the older rocks in the
neighbourhood, to have been instrumental in producing the very
remarkable inclination of the sands*. But this inclination seems
evidently one produced during the deposition of the beds, because it
is marked at several other points corresponding with the course of
the valley, as before mentioned, and it does not always affect the
whole of the beds seen in one section, being very irregular, although
inclining into the valley rf. ;

* De la Beche, referring to the deposits of Greensaud in the neighbourhood
of Newton, and which had been brought to light by Mr. Godwin-Austen, con-
cludes that probably a combination of faults and of a depression, produced
either previously or subsequently to the deposit of the Greensand, may best
explain the phenomena observed (Geol. Report, p. 236).

+ Mr. J. H. Key, in describing the Bovey Clays and Lignites, has pointed out
how the dip of these beds increases from the sides towards the centre of the
basin (Quart. Journ. Geol. Soc. vol. xviii. p. 14).

IN THE NEIGHBOURHOOD OF NEWTON ABBOT. aaa

At Staple Hill, moreover, there are coarse gravelly seams con-
taining a few flints, as well as many other fragments of local rocks
interstratified in the sands, in one of the large sand-pits now worked
at that locality.

On the western margin of Milber Down, and just inside the lodge-
gates on the carriage-road leading to Haccombe, is a small pit
exposing fine and rather loamy sands with a few irregular chert
blocks, which may very likely represent an outlier of Upper Green-
sand. In any case it would only extend along the top of the ridge,
and be almost entirely concealed by the coarse gravelly deposits
which descend towards the bottom of the valley, and there become
intermingled with finer sediments. Another small outler may be
present beneath the gravels worked east of Combe Farm on the old
and new high roads between Newton and Exeter. At the base of
the gravels there is exposed sand much like the Greensand ; but I
could not determine whether it was in any way intercalated (as
exactly similar beds are in other places) with the coarse gravel con-
taining flint and chert.

These two patches, however, occur in positions autho might
connect them with the Greensand of Haldon.

At Haccombe, indeed, Mr. Godwin-Austen recorded beds of true
Greensand, with included layers of chert and whetstone, and
covered with characteristic fossils; but I cannot agree with him in
considering that the Greensand ranges over the summits of the hills
above Coffinswell, that it passes into the valley a little below Kings-
kerswell, and then rises into the opposite hills. Near Combe Farm
Mr. Godwin-Austen obtained Evogyra and Pecten quinquecostatus.

He has, however, recorded some fossils from localities where I
have been unable to identify any Greensand im situ. Thus at
Staple Hill he has found 7rigonia and Jnoceramus concentricus, at
Lyndridge Hill Orbicule and Orbitolites.

Although I searched diligently, and had the assistance of one of
the fossil-collectors of the Geological Survey, we obtained not the
fragment of a fossil in any of the sandy beds. This of course is in
itself negative evidence, and of no great value ; for in many sections
of undoubted Upper Greensand one may search long and find no
reward in the shape of a fossil.

Looking to the facts that at Staple Hill one of the large pits con-
tains one or two bands of gravel yielding flints sparingly, that the
beds have that remarkable inclination repeated at Whitehill and
between Newton and Kingskerswell, where the same sands become
interbedded with coarse gravel containing flint and chert, and
overlie the Bovey beds, I feel no hesitation in considering the few
fossils that have been found at these localities to have been rede-
posited along with much material derived from the denudation of
the Greensand.

Although these older deposits are roughly confined to the main
valley of the Teign and its tributaries west of Kingsteignton, yet, as
I have noticed, they extend to a height of 540 feet between Hac-
combe and Coffinswell, and they do not descend into valleys beyond

234 H. B. WOODWARD ON THE SUPERFICIAL DEPOSITS

the Bovey basin. This tends to show that most of the minor
features of the country have been produced subsequently to the
deposition of the gravel.

Further, these deposits can be traced from Sandy Gate upwards to
Gappath and the borders of Ugbrooke Park; and were the slope
continued without break, we might reach the Greensand of Haldon,
which is capped by analogous deposits of drift sand and gravel.
There seems to be some connexion between these plateau-gravels
which cap all the Greensand heights of Devonshire, and those which
were accumulated with this high marginal dip in the Bovey basin.

The question of their formation seems to me to be a large one,
and to require for its solution a more extended study of the super-
ficial deposits of Devonshire than I have been enabled to make.
Mr. Belt, however, has taken up the subject boldly; and my col-
league, Mr. Ussher, is gathering facts which will, I hope, soon lead
to very interesting results.

It is not always easy and is sometimes impossible to tell where
these older deposits of gravel end in the valley, and where the modern
alluvial gravels come on. This, however, is only natural, because
the alluvial deposits are evidently very largely made up of the older
materials.

The Bovey Clays are worked beneath the ‘‘ Head” of gravel,
sometimes called ‘‘ Pengelly’s Head,” which in most instances is a
comparatively modern fluviatile and estuarine deposit. It often
occupies a position above the alluvium proper, but is sometimes not
to be distinguished from it.

Mr.J.H. Key, who has given an excellent description of this deposit,
has found in it the rib and jaw of a deer, and shells of the oyster
and cockle *. In the collection of Messrs. Watts, Blake, Bearne
and Co., of Newton Abbot, are bones of the Ox, Deer, and Man,
obtained from the same deposit, and what is most remarkable, a
wooden doll, or idol, about a foot in length, which was obtained at a
depth of 20 feet in the coarse gravel, on the left bank of the river
Teign, in a pit called the Zitherixon pit, near the toll-gate between
Newton Abbot and Kingsteignton. A bronze spear-head was also
found in the same pit at a depth of 15 feet.

There is no doubt that this deposit was on the whole the latest
formed in the valley, partly by river and partly by estuarine
action.

The material: was evidently for the most part obtained “ ready-
made,” as it were, from the older gravels ; and therefore the de-
ductions are indicative of the rapid accumulation of the gravels
rather than of the antiquity of the remains found in them.

Mr. John Evans considers that the spear-head ‘belonged to
the latter part of the British Bronze Period—say some few cen-
turies B.c., or even only one or two. The wooden figure,” he adds,
‘“‘ig much like some of those from the peat of the Somme valley, de-
scribed by Boucher de Perthes. It is hard to say what is their age
or object.” Mr. Carruthers, who examined the wooden figure, con-

* See Quart. Journ. Geol. Soe. vol. xviii. (1862), p. 9.

IN THE NEIGHBOURHOOD OF NEWTON ABBOT. 235

sidered that it was formed of bog-oak, and was probably shaped from
a piece of fossil and not recent wood*.

I mention these facts as affording interesting evidence of the
accumulation of alluvial deposits, and as furnishing one indication
out of many that might be given that our modern river-gravels are
to a great extent made up of older gravels.

The deposits now being formed in the bed and along the margin
of the estuary of the Teign between Newton Abbot and Teignmouth
are many of them identical in character with the Triassic breccia.
Low cliffs of Devonian slate, with here and there igneous rocks,
border the water in places. These slaty rocks by constant satura-
tion become soft, and break up into small angular fragments, form-
ing a beach, which sometimes lies at an angle of 8° or 9°. This beach
contains a few pebbles of quartz, grit, &c., which are derived from
bordering patches of gravel. The slates break up along lines of
jointage and cleavage; and large masses frequently slip into the
water and gradually fall to pieces. Large blocks of hard rock fall
by atmospheric wear and tear, and become imbedded in time amidst
the slaty fragments without being worn or rolled about, a fact
which might account for some, at least, of the large blocks of igneous
rock met with in the Triassic breccia.

[For the Driscusston on this paper see p. 238. ]

* The wooden image has been figured by Mr. Pengelly in the Trans. Devon
Assoc. for 1875, vol. vii. p. 200. See also Trans. Anthropol. Inst. vol. v. p. 299.

Q. J. G.S. No. 127.

h

236 R. N. WORTH ON CERTAIN ALLUVIAL DEPOSITS

27. On certain AttuviaL Depostts associated with the PLyMouTH
Limzstone. By R. N. Wortu, Esq., F.G.S. (Read April 5,
1876.)

In the ingenious paper on the Drift of Devon and Cornwall,
read by Mr. Belt before this Society in November last, that gentle-
man speaks of the lowland gravels of Devon and Cornwall as show-
ing “signs of sudden and tumultuous deposition.” Whether he is
right or wrong in this assumption can best be settled by a reference
to detail in various localities ; and it is with the view of aiding in
this work that I would direct attention to certain alluvial deposits
associated with the Plymouth limestone, which seem, so far as that
locality is concerned, to negative Mr. Belt’s hypothesis.

The limestone of Plymouth, and that of Yealmpton (which com-
mences about two miles distant from the eastward termination of the
first), abound in fissures and caverns. Some of the latter have become
celebrated by their ossiferous contents. Some of the former contain
the alluvial deposits to which I refer, and which, denuded off else-
where, are therein preserved.

The best examples of these deposits have occurred in connexion
with Plymouth Hoe, where the so-called drift-gravel fills sundry
fissures, and has been found to entomb remains corresponding very
closely to the fauna of the Oreston caverns, which are less than a mile
distant, on the other side of the Cattewater.

The chief deposit occurs about midway on the Hoe, and fills a
huge ‘‘ pocket” in the limestone. Beneath the turf there is a bed
of clayey soil containing pebbles, which range from a very small size
up to boulders a dozen pounds or more in weight. They are chiefly
quartzose; some composed of quartz and schorl, others granitoid,
and a few of slate. But for the absence of granite, this part of the
deposit in no respect differs from the ordinary alluvium of our Dart-
moor rivers,

Beneath this pebble-bed are patches of white and red clay, con-
taining few pebbles; and beneath this, again, a large quantity of
siliceous sand, the depth of which has not yet been ascertained, as
the excavations have not reached the bottom of: the fissure. The
sand is chiefly of a cream-colour.

Similar deposits to this series are by no means of unfrequent oc-
currence in association with the limestone, though sometimes the
sand, sometimes the clay, and sometimes the pebbles appear to pre-
dominate. They are found not only on the Hoe, but at Cattedown
(where, in Deadman’s Bay, is a “ pocket” at least 40 feet in
depth, containing white clay similar to that described), at Billa-
combe (near the termination eastward of the Plymouth limestone),
and at Yealm Bridge (where a fissure contains pebbles identical
in character with those in the bed of the Yealm, 50 feet below).

ASSOCIATED WITH THE PLYMOUTH LIMESTONE. SST

We have here, I think, clearly the remains of considerable de-
posits which once occupied large areas in the valleys of the south
of Devon, and which, in part at least, are fairly entitled to be con-
sidered the lowland gravels of which Mr. Belt speaks. If they are
not, then in this immediate district these gravels are not repre-
sented. As to their date, there is substantial evidence that the same
stream, or streams, to which they are due, also carried into the
Oreston caverns the extinct mammals whose remains have been
found there, and that the cavern- and fissure- or ‘‘ pocket ”-deposits
are contemporaneous.

Mr. Belt, following authorities of such eminence as Professor
Sedgwick and Sir H. De la Beche, and concurring with Mr. Joseph
Carne, holds that the lowland gravels were spread out by a great
débicle. If this view be correct, the evidence should be continuous
throughout the district presumed to be exposed to its action. But,
in the deposits here briefly described, there is not only no evidence
whatever of a cataclysmal character, but every indication of orderly
deposition :—inland, nearer to the source of the débris, the bulk of
the pebbles and gravels proper; further from their origin the sands
and clays, in fair orderly succession.

The idea that the stanniferous gravels of Cornwall and Devon
were formed by some great catastrophe is at least three centuries
old; for Carew, in his ‘Survey,’ says that the miners attributed
them to the deluge. Nearly fifty years since, Mr. Carne contended
that these gravels were deposited by a deluge of water from the north,
because the productive stream-works are situated in valleys opening
to the south. This fact of position, however, is easily explained by
reference to the peculiarity of the watershed of Cornwall. There
are only two rivers in the north of that county; and one of these
does not pass through a stanniferous district. On the south coast
rivers abound. From its superior weight, the tin ore would
naturally form the lower layer of the deposits with which it was
associated—and this whether the whole of the materials were
in motion at once or not, since the effect of the continuous action
of a stream on gravel in its channel is to carry forward the lighter
and leave the heavier behind. Moreover the natural separation of
the tin-gravel would be materially aided by the occurrence of times
of flood.

Sir Henry De la Beche clearly shows that all that is required to
account for the production of the stanniferous deposits is the exist-
ence of a mass of decomposed granite containing tin-ore, and its
subjection to the action of water—though it is not necessary, as I
hold, to assume with him that this water should be “ driven vio-
lently against or over it.” In such a body of decomposed granite I
conceive, not merely the stanniferous gravels of the Dartmoor rivers,
but the deposits of clay and sand here noticed, have their origin—
the clay representing the felspar, and the sand the quartz. This
accounts also for the absence of granitic pebbles in a deposit derived

from the Dartmoor Granite.
s2

238 R. N. WORTH ON CERTAIN ALLUVIAL DEPOSITS

e

Discussion.

Mr. Ussuer stated that he had been inclined to consider that
the sands flanking the Bovey valley were of peculiarly local occur-
rence ; but the observations of Mr. Woodward as to their similarity
to certain arenaceous deposits on Great Haldon and the Blackdown
Hills, recalled to his mind an instance where, in a pit on the
former, sand of granitic origin, much resembling portions of the
sands in the neighbourhood of Newton, is shown, underlying an
accumulation of broken unworn chalk flints in a clayey matrix,
ascribed by Sir H. de la Beche and French geologists to the Plastic-
Clay formation. Owing to the removal of Post-tertiary deposits, he
was unfortunately unable to trace any connexion between them.

The sands of which Mr. Woodward’s paper treats skirt the Bovey
valley at heights of from less than 100 to more than 500 feet above
the sea-level, dipping down the slopes of the hills at angles roughly
coincident with them, as gravel seams in the deposits show. Hence
it seems evident that they cannot have been originally deposited
upon their present uneven surface, but owe their position to those
forces of elevation and depression the prevalence of which in Devon
during the Post-tertiary period is elsewhere distinctly evidenced.
With regard to one statement, he desired further information from
Mr. Woodward—namely, as to the direct superposition of these sands
upon the clays of the Bovey valley; for he agreed with Mr. Wood-
ward that the superficial deposits of the valley, including the gravels
of the Zitherixon pit (in which the wooden doll or idol was found),
were of much later date, and due to the redistributing action of the
stream now occupying the valley at a period when its volume was
much greater than at present.

With Mr. Worth he entirely agreed in considering that Mr. Belt’s
explanation of the gravels of Devon and Cornwall was unsatis-
factory; for not only were there no signs in their mode of occur-
rence of a general dispersion involving more or less of cataclysmal
action, but indications of a clear order of sequence, manifested, for
mstance, by the gravels on the sides of such valleys as those of the
Exe, Otter, Taw, and Torridge, indicating successive deposition by
the former representatives of the present streams in their erosion of
the valleys, the most recent deposits occurring as alluvia and gravels
occupying flat tracts of land through which the rivers now flow.

He considered that if, in addition to the evidence furnished by
these gravels, the raised beaches on the Devon and Cornish coasts,
the stony loam or head of angular débris overlying them, the
cavern deposits of Devon and other phenomena were taken into
account, an order of sequence proving physical changes might be
established, by which some conclusion respecting the vicissitudes
experienced by the counties of Somerset, Devon, and Cornwall
during the Pleistocene epoch might be arrived at.

Mr. Tuorpz was glad that steps had been taken for properly
working out this interesting district, as there were -few reliable

ASSOCIATED WITH THE PLYMOUTH LIMESTONE. 239

papers on the subject, Mr. Holles’ elaborate paper being in many
points untrustworthy.

The Prusrpenr (Prof. Duncan) read a letter from Mr. H. B.
Woodward, in which the latter stated that his reason for bringing
this paper forward was that many sections were now to be seen in
the course of excavations for building around Newton-Abbot, which
would soon be obliterated. The President also said that he had
been over the ground mentioned by Mr. Worth in his paper, and
entirely agreed with the views expressed by that gentleman.

240 J. MAGENS MELLO ON THE BONE-CAVES OF CRESWELL CRAGS.

28. The Bonz-caves of CrEsweLt Craes.—2nd Paper. By the
Rey. J. Magzns Mztto, M.A., F.G.8. (Read April 5, 1876.)

I wap the honour last June of reporting to the Society my dis-
covery of some interesting bone-caves and fissures in Creswell Crags,
in the Lower Magnesian Limestone of N.E. Derbyshire, which con-
tained a large number of species of Pleistocene mammalia, together
with some traces of the presence of man*. I was able on that
occasion to exhibit portions of some 15 or 16 species belonging to
no fewer than 12 genera obtained from one cave alone, locally
known as the Pin-hole, so called from a curious custom which
prevails amongst some of its visitors of dropping a pin into a small
water-filled hollow, removing at the same time a pin deposited by
some previous visitor. Amongst the animals which had left their
remains in this cave the most important were the Irish Elk, the
Glutton, and the Arctic Fox; together with these were a large
number of Hyzna remains, and also bones and teeth of the Mam-
moth, the Woolly Rhinoceros, the Brown Bear, the Reindeer, the
Urus, and of some other animals. During the past summer I have
been able to carry on the work of exploration, assisted by Mr.
Thomas Heath, F.R.H.8., Curator of the Derby Museum, who has
been able to devote a good deal of time to it, and whose skilful help
has been of great value in carrying on the researches ; and I must
also acknowledge the energetic assistance given us by Mr. F. Tebbet,
superintendent at the Creswell quarries, who from the beginning has
taken much interest in these discoveries.

We commenced the renewed search by continuing the excavation
of the floor of the Pin-hole, trusting that amongst other remains we
might obtain some more evidence of the presence of the Arctic Fox ;
in this, however, we were not successful. All the front of the
eavern was thoroughly searched, the chief bones found being
two perfect pelves of Rhinoceros tichorhinus, and also two atlases
of the same animal, together with a few Reindeer- and other bones
of no particular interest. As we worked our way further into the
fissure the number of bones found was very small indeed; the bed
of red sand, which at the entrance of the cave had proved so rich in
its contents, became filled with limestone fragments, and was nearly
destitute of bones; and we determined, under these circumstances,
to desert that cave for the time and begin the exploration of a
neighbouring one a little lower down the ravine, and in the same
side of it. This, which is called the Robin-Hood Cave, is of moderate
size, containing several chambers communicating with each other,
the separation merely consisting of narrow walls of the limestone
rock (fig. 1).

We began work here by making a section of the floor at the
entrance, cutting down to a depth of about 8 feet, where blocks of
limestone were met with, which probably form part of the original

* Quart. Journ. Geol. Soe. vol. xxx1. p. 679.

J. MAGENS MELLO ON THE BONE-CAVES OF CRESWELL CRAGS. 241 |

Wit;

The curved broken lines indicate the portion worked out.

floor of the cave. We gradually worked forward into the cave,
carefully examining each stratum as it was removed. As in the
Pin-hole, so here there was a certain amount of dark surface-soil of
inconsiderable thickness, seldom, if ever, exceeding 5 or 6 inches,
and near the entrance not above 2 inches; in this in different
parts of the cavern we found some broken fragments of Roman and
Medizeval pottery, a human incisor, and some bones of recent animals
(sheep etc.). On the left-hand side of the cave, and extending a con-
siderable way across its mouth, there was a very hard limestone
breccia, varying in thickness from a few inches up to about 3 feet;
beneath the breccia was a deposit of light-coloured cave-earth, more
or less sandy and very calcareous; where the breccia attained its
greatest thickness the cave-earth was almost wanting, being only a
few inches thick at the side of the cave; but further in, under the
thinner portions of the breccia, the cave-earth was fully 3 feet
thick ; the succeeding layer was of dark-red sand, very similar in

242 J. MAGENS MELLO ON THE BONE-CAVES OF CRESWELL CRAGS.

character to the bone-bearing bed of the Pin-hole, but differing
from it in containing, near its base, patches of highly laminated red
clay. Small nodular masses of black oxide of manganese occurred
here and there in the sand, and also some quartzite and other
pebbles. This middle red sand bed was about 3 feet thick, and
rested upon a bed of lighter-coloured sand, containing many rough
blocks of limestone already mentioned as apparently forming part
of the original floor of the cavern. Fig. 2 gives a section across the

Fig. 2.—Section in the Robin-Hoed Cave, in line 1, Fig. 1.

+ Stalactite uniting breccia with roof.

a. Stalagmitic breccia, with bones and implements, 18 in. to 3 ft.

6. Cave-earth, with bones and implements, of variable thickness.

c. Middle red sand with laminated red clay at base, containing bones, 3 ft.
d, Lighter-coloured sand with limestone fragments, 2 ft, ?

side of the cavern where the thickness of the breccia was greatest.
Fig. 3 is another section, taken across the cavern, facing the openings
of the chambers, which are shown above the cutting.

The Breccia.—The Breccia (a) was firmly cemented together by
stalagmite, many thick masses of which were interspersed with it ;
and the whole deposit was so hard that it had frequently to be
blasted in order to remove it and examine its contents. It was found
to contain a large number of bones, mostly of small animals, including
the-Water-vole ; but together with these were some of the Reindeer,
and also teeth of the Hyena, Rhinoceros tichorhinus, and Horse.

J. MAGENS MELLO ON THE BONE-CAVES OF CRESWELL crAGs. 243

Fig. 3,—Section in the Robin-Hood Cave, in line 2, Fig. 1.

a. Surface-soil and thin breccia, 2-3 in.

6, Cave-earth, with fiint and quartzite implements, teeth, bones, angular lime-
stone fragments, and charcoal, 3 ft.

c. Red Sand with laminated clay, few bones, 3 ft.

1. Fox-hole?

Besides the bones, the breccia contained numerous flint flakes and
chips, as well as one or two flint cores. Although the majority of
the flints were chips, or flakes of the simplest character, some few
of them were of superior workmanship, being well-shaped spear-
heads, or large arrow-heads chipped on both surfaces; there were
also one or two sharply pointed fiints of the awl type. A few
quartzite implements, similar to those to be described in the suc-
ceeding bed, were found in the breccia.

The Cave-earth (b).—The cave-earth contained some flint imple-
-ments; but besides these were a large number of split or chipped
quartzite pebbles, evidently fashioned by man, most of them having
a very definite bulb of percussion, and a general uniformity of
design prevailing amongst them. Amongst these quartzite imple-
ments are some fashioned on both faces, and which present in their
oval form a decidedly paleolithic aspect ; a similarly shaped imple-
ment of clay-ironstone was found with these. Some of the pebble
implements were designed for hammers, one large one having a few

244 J. MAGENS MELLO ON THE BONE-CAVES OF CRESWELL CRAGS.

chips struck off one end, and the opposite face being much bruised.
Besides these implements there were a considerable number _of
unfashioned pebbles of quartzite, together with one or two of black
chert and of quartz, both in the cave-earth and in the red sand.
The cave-earth, as well as the breccia, contained a good many small
fragments of charcoal. We found numerous animal remains in this
bed. Horse-teeth were specially numerous, and also those of the
Rhinoceros tichorhinus, and of the Hyzena ; a good many very perfect
fragments of both the upper and lower jawbones of the latter
animal were found, one of the upper jaws still retaining, in addition
to the canine and large molars, the small fifth molar that is so
generally lost. A left ramus of a lower jaw has also some of the
incisors 7n situ. The numerous Rhinoceros-teeth are both upper and
lower molars and premolars, together with some milk-teeth, most of
them being very perfect. In this bed were found bones of the
Reindeer, as well as teeth, and also some fine teeth of the Cayve-
Lion and of Bears. At the base of the cave-earth, at one place on
the left-hand side of the cavern there was a thin bed of small
pebbles, apparently deposited by water; this and the laminated red
clay were the only traces of any thing like regular bedding in the
floor of the cave, apart from the chief divisions already mentioned.
At one point the cave-earth had been considerably disturbed by a
fox-hole ; this was the only instance in which such disturbance was
seen (see 1, fig. 2).

The Red Sand (c).—The Red Sand underlying the cave-earth
contained comparatively few bones, except in one place near the
entrance of the cavern, where a considerable number of large bones
were found at its base, almost resting on the underlying light sand.
The bones consisted of portions of antlers and other remains of the
Reindeer, some very perfect metacarpals and metatarsals, and ver-
tebrz of the Bison, and some bones of the Hyzna etc. At another
place was found an extremely perfect small molar of Hlephas primi-
genius with a portion of the jawbone still attached to it. A further
and detailed account of the implements and bones of this cavern has
been kindly prepared for me by my friend Prof. Boyd Dawkins, and
will be found in the following paper. One thing to which I should call
attention is the somewhat strange fact that although a very large
majority of the bones discovered have been gnawed by hyzenas, to
whose agency we must attribute the presence of most of the animal
remains found in these caves, yet up to the present time we have
been able to find no trace of the coprolites of these animals, which
are usually so numerous in hyzna-dens; I hardly know how to
account for their absence. I hope that we shall be able to continue,
the work of exploration during the ensuing summer ; and the results
of such further explorations I trust I shall have the honour of
bringing before the Society at a later period.

[For the Discusston on this paper see p. 256. |

ON MAMMALIA ETC. FOUND IN THE ROBIN-HOOD CAVE. 245

29. On the Mammata and Traces of Man found in the Rosin-Hoop
Cave. By W. Borp Dawxins, Esgq., M.A., F.BS., F.GS.,
F.S.A., Professor of Geology and Paleontology in the Owens
College. (Read April 5, 1876.)

ConTENTS.

. Caye oceupied by Hyzenas during deposition of Lower and Middle strata.

. Occupation of cave by Hyzenas interrupted by floods.

. Occupation of cave by Hyzenas during deposition of cave-earth interrupted
by Man,

Occupation by Man during the period of the Breccia,

. Dog not present.

. Fauna of the Breccia different from that of underlying strata,

. Remains from surface-soil.

. Distribution of Species in the cave.

. Notes on Species.

10. Traces of Man in the cave.

11. Implements of antler and mammoth-tooth.

12. Implements of quartzite and ironstone,

13. Implements of flint.

14, Distribution of implements in cave,

15, The ruder implements the older.

16. North-western range of Palzolithic hunters.

COIS STP whe

Tur discovery of the fossil remains in the Robin-Hood Cave,
brought before the Geological Society by the Rey. J. Magens Mello
in the preceding paper, is of no common interest. It proves
not only that the caves of Derbyshire were the lairs of Hyznas in
ancient times, but that they were inhabited by the same kind of
Paleolithic men as those of the caves of the south of England,
of France, Belgium, and Switzerland. The remains have been
handed over to me for description by Mr. Mello and his coadjutor
Mr. Heath, and are the subject of the following remarks,

1. Cave occupied by Hycenas during deposition of Lower
and Middle strata.

A comparison of the bones and teeth from the lower red sand
and clay, and from the cave-earth, with those from Wookey Hole,
Kent’s Hole, Kirkdale, and other hyzna-dens, renders it impossible
to doubt that the great majority of the animals in the cave were
killed and eaten by the Hyznas. With few exceptions the solid
bones are alone perfect, the long bones containing marrow and the
vertebre being represented merely by gnawed fragments. All the
lower jaws have lost their angles and coronoid process; and the
number of teeth stands in a greater ratio to the quantity of bones,
than would have been the case had not their possessors fallen a prey
to a bone-destroying animal. The only long bones and vertebree
which were found without marks of the teeth of Hyzna were met
with in the lowest ossiferous stratum, and belong to the Bison and
Reindeer. Their presence may be explained by the supposition that
they were introduced by the stream flowing past the entrance.

246 W. BOYD DAWKINS ON THE MAMMALIA AND

2. Occupation of cave by Hycenas interrupted by Floods.

The red sand and clay of the lower ossiferous stratum are the
results of an occasional flooding; and the smoothed and rounded
surfaces of many of the bones and teeth are due to the friction of
the sand set in motion by the currents of water in the cave.

It seems, therefore, tolerably clear that the occupation of the cave
by the Hyena, during the time of the deposition of the lower ossiferous
stratum, was occasionally interrupted by floods.

3. Occupation of cave by Hycenas during deposition of cave-earth

umterrupted by Man.

The middle deposit of red loam, of the kind so abundant in
caves in the south of England, has probably been introduced during
heavy rains, and is to some extent the result of the decomposition
of the limestone. During its accumulation Hyznas inhabited
the cave; but their occupation was disturbed by the visits of Paleo-
lithic hunters, who left behind them the implements to be presently
described.

4, Occupation by Man during the period of the Breccia.

The remains from the breccia above the caye-earth seem to me
to indicate that the cave at that time was inhabited by Man, and
that it was not so frequently visited by the Hyzenas as before.
The pieces of breccia, which I have carefully broken up in search
after bone needles, contain for the most part the split and broken
bones and a few vertebre of the Hare. Had the Hyeenas then
frequented the cave, the vertebrae would have been eaten, and some
of the bones would have been gnawed.

5d. Dog not present.

The presence of the vertebrae also implies that the Dog was not
used by the hunters who then lived in the cave. It will be remem-
bered that a similar conclusion was drawn by MM. Lartet and
Christy from the vertebrae of Reindeer in the caves of Perigord with
regard to the Paleolithic hunters in the south of France. It seems,
to me, indeed, after a careful examination of all the evidence, that
the Dog was not the servant of Man in the Paleolithic age in
Europe, and that the reputed occurrence of its remains in deposits
of Pleistocene age is the result of mistaken identity, or of mistaken
gusement.

6. Fauna of the Breccia different from that of underlying strata.

The group of remains from the breccia differs, as may be seen
from the following Table, considerably from those of the underlying
strata. The carnivores of the latter, with the exception of the
Hyzenas and Wolf, are conspicuous by their absence; while all the
herbivores are represented, with the exception of the Bison. The
Hare, on the other hand, so rare in the latter, is abundant in the
breccia. This difference in the mammal faunas is accompanied, as
will be seen, by corresponding differences in the implements. The

TRACES OF MAN FOUND IN THE ROBIN-HOOD CAVE. 247

numerous flint flakes and fragments of charcoal in the breccia
prove that then Man was the normal inhabitant of the cave, while
the broken bones prove that he fed for the most part on hares.

7. Remains from Surface-soil.

The remains from the surface-soil belong to domestic animals,
such as the Short-horned Ox (Bos longifrons), Goat, Hog, and Dog,
or to Hares and Rabbits. A fragment of Samian ware and pieces of
rude black pottery prove that the cave was visited after the Roman
conquest, and sherds of a glazed jar that it was also visited in the
middle ages.

8. Distribution of Species in the Cave.

The accompanying list of species shows the vertical distribution of
the animals in the cave, and enables us to form a rough idea of the
mammalian fauna of the district. It will be seen that during the
accumulation of the lower and middle strata the Horse was more
often the prey of the Hyena than any of the other animals, and
next to it the woolly Rhinoceros, just as in the Hyzna-den of
Wookey Hole, while the Bison, so enormously abundant in the
deposit at Windy Knoll (Derbyshire), is comparatively rare. It
was, however, more abundant in the lower stratum in the cave than
in the cave-earth, the numbers of specimens in each being as thirty
to six.

Lower red | Middle cave- == Upper
sand and clay. | earth. | Breccia.
ee ae Gat on eh tae | ee | te
ga| 8) 4 gs] 2) 4 les) 2/4
SsalaSslalsa Ss ala
URES GS 7) ere Senet ead tac some ake Te ee * | x
2. Lion (var. Felis spelea)......| 1 | |... | 3 | |
3. Spotted Hyzna (var. #.
SPUR innvavieavensnce satan | 22) 3] ../69) 2)... ) 6 |
4. Fox (Canis vulpes)............ eee ae ae ee |
5. Wolf (Canis lupus) ...... gt Ue EE as So ers reat ie
6. Grizzly Bear? (Ursus feror)| 11 | 1]...| 2] 1
7. Brown Bear (Ursus arctos).| 1
8. Irish Elk (Cervus megaceros)| ... |... | oo. | 2] |... | 1} 1
9. Reindeer (Cervus tarandus).| 6 | 17 |... | 6 | 22 Pores |
10. Bison (var. Bison priscus)...| 4 | 26]... |... | 6
11. Horse (Hguus caballus) ...... SE ie | wee eke hD 3
12. Woolly Rhieoceros (2. ticho-
PURTUS) ies. Glade eosecns meni ee Mew. Adel Abe Lesthedet
13. Mammoth (Elephas primi-
NDEI Meck abcd cite eb wane te TO Bh SR i |
14. Wile) Bont (SUsscroy@jerws)| 1. |... | ose | s->, | oe.{ wee | 2
15. Hare (Lepus timidus) ......| «+ er PA eet ER ee
16. Arvicola amphibia ............ Peiheve [soi] ee
eet a ae LOR LL o> a | ee eee ee es —— fe
17. Dog (Canis familiaris) ......| ... Ee ee OCC eer ees eames. poem
18. Goat (Capra hircus) .........| ... Sei iath- | wns. ff swe.|, iran hy aaa ee ee
Td. Celtie Shorthomy Bes fomgi-) s.5 | ane | oes | cae | oon | vee | cee | ace | on he oe
PROMS). <.jst tee

A). Hop (Susiserafa domestiems).| sc. | 2. | su. | «+0 | 00 | soe | oon | ave |) ee |

Surface-soil,

248 W. BOYD DAWKINS ON THE MAMMALIA AND

The faunas of the three lowest stages in the cave present no
important differences worthy of note, with the exception of that
between the breccia and the cave-earth already pointed out—a dif-
ference, be it remarked, that may be the result of one animal falling
more easily into the hands of man than another, and not of a change
in the fauna of the district.

9. Notes on Species.
Order CARNIVORA.

Iion.—The Lion is represented by three canines, two worn down
to the stump and one belonging to a young adult, and by ml.
These teeth belong to four individuals, and are rather smaller than
the average of those from the British caves described in the Mono-
graph on Felis spelea by Mr. Ayshford Sanford and myself (Pal.
Soc. 1866 et seq.)

Spotted Hycena.—The cave variety of the Spotted Hyena is
proved to have inhabited the cave through many generations by the
numerous jaws and teeth of all ages, ranging from whelphood to
the extreme of old age. All the jaws are gnawed to the patterns
figured in Buckland’s ‘Reliquie Diluviane,’ pls. 3, 4,5, and in
‘Cave-hunting,’ fig. 92.

Fox (Canis vulpes)—A femur, two lower jaws, and a few
separate teeth fall within the limits of size offered by the corre-
sponding parts of the common Fox; and I therefore regarded them
as belonging to that rather than to the arctic species, which has
been determined by Prof. Busk, F.R.S., from a neighbouring cave
(Quart. Journ. Geol. Soc. 1875, p. 686).

Wolf.—The jaws, teeth, and phalanges of Wolf cannot be distin-
guished from those of the Canis lwpus of Europe and Asia. They
belonged to adult animals.

Bear.—The teeth of Bear (consisting of canines, premolars, and
molars) indicate the existence of two closely allied forms. One of
the canines is identical with that of U. arctos, while the rest, and
the molars and premolars, are undistinguishable from those of the
U. ferow or Grisly Bear as defined by Prof. Busk. None of them
belong to cubs.

With regard to the further question as to whether the two Bears
are closely related species, or well-marked varieties of one species,
speech may be silvern, but silence is golden. The examination of a
vast number of the remains of fossil Bears in this country and on
the continent has forced on my mind the extreme difficulty of
defining the one from the other from the study of the hard parts.
Both forms, however, are quite distinct from the largest of the cave-
haunting Bears (Ursus speleus).

TRACES OF MAN FOUND IN THE ROBIN-HOOD CAVE. 249

Order ARTIODACTYLA.

Suborder RumINantTIA.

Trish Elk and Rendeer.—-The alveolar portion of the lower jaw,
a fragment of a metacarpal, and two molars belong to the Irish Elk,
while numerous antlers, some shed, and others torn from the head,
teeth, and gnawed bones may be referred to the Reindeer. It is
worthy of remark that all these remains are those of adults.

Bison.—Of the Bison it need only be remarked that the remains
fall within the limits of measurement of those from Windy Knoll
(Q. J. G. 8. vol. xxxi. 1875, p. 247). Some of the long bones
belong to young animals.

Suborder Non-RumInanTIA.

Wild Boar.—A small fragment of jaw containing the first three
premolars proves that the Wild Boar also lived in the district during
the time of the accumulation of the breccia.

Order PERISSODACTYLA.

The Horse and Woolly Rhinoceros.—The teeth and the bones of
Horses prove that the adults were here with their foals: and this
may also be observed of the Woolly Rhinoceroses, the milk-teeth of
which amounted to 19, or very nearly one third of the total number
of teeth of that animal in the cave. All the long bones are gnawed
to the usual patterns (see ‘ Cave-hunting,’ p. 314 et seq.). The humeri
have been docked of their proximal, the ulne of their distal ends,
while the radii, femora, and tibiz are represented merely by the
stout middle portion of the shaft; and in all, the nutriment stored in
the cancellous portion of the interior of the bones has been scooped
out by the jaws and tongues of the Hyzenas.

Order PROBOSCIDEA.

Mammoth.—Five out of the six molars of Mammoth belong to
the milk-series, and three of them are dm 2. The true molar is a
stump worn down to the fangs. There are also fragments of tusks
and of bones.

Order RODENTIA.

The Hare—tThe split and broken bones of the Hare were very
abundant in the breccia; and that animal had formed the principal
food of the inhabitants during the time of its accumulation. All
the long bones had been split for the sake of their contents. There
are also numerous jaws and teeth of Arvicola.

10. Traces of Man in the Cave.

The traces of Man found in the cave in association with the
extinct mammalia, consist of fragments of charcoal, and implements
of antler and mammoth-tooth, of quartzite, ironstone, greenstone,
and of flint. ,

11. Implements of Antler and Mammoth-tooth.
An awl, pin, or possibly a lance-head, has been fashioned out of

250 WwW. BOYD DAWKINS ON THE MAMMALIA AND

a tyne, which has been ground down to a sharp point (fig. 1), simi-
lar to many of those found in the caves of the Dordogne,
Fig. 1. and figured in the ‘ Reliquiz Aquitanice.’ It may have
. formed part of a pin such as that figured by Mr. John
Evans from Kent’s Hole (Ancient Stone Implements,
fig. 406). Another pointed tyne may also have been
used for the same purpose as fig. 1; but the marks of
Man’s handiwork are not so decided. A fragment of
the base of a Reindeer-antler from the caye-earth may
perhaps have been cut and perforated by the hand of
Man. Another is a triangular sharp-pointed arrow-head
or piercer (?), formed of one of the plates of an Elephant’s
molar, probably of a milk-molar. Its surface is highly —
Point made Polished, and it has been formed by the loss of the enamel
ofantler. and the grinding of the surface of the dentine until it as-
hull size sumed its present form*. A loose plate of the milk-molar
of a Mammoth was also found in the cave.

12. Implements of Quartzite and Ironstone.

The implements of quartzite and ironstone, eighty-six in number
(irrespective of splinters), belong to well-known types in other regions,
which are generally fashioned out of flint. They have all been made

ik : i | iy

( is,

oe ie
ie | my

Quartzite flake, x3. Cavye-earth.
a. Section.

SYS

Quartzite hache, x3. Cave-earth. «a. Section.

* This singular specimen is considered by Mr. Evans to be non-artificial.
After a further examination of it in the British Museum, Mr. Davies and myself
eannot look upon it otherwise than as expressed in the text.

TRACES OF MAN FOUND IN THE ROBIN-HOOD CAVE. Zak

out of pebbles, in which advantage has been taken of the smooth sur-
face to form one side of the implement.

A triangular implement (fig. 2), tapering from the unworked base
to an obtuse point, and carefully worked at the sides and point, was
obviously intended to be used in the hand like those of the Pleisto-
cene river-gravels. The side which is not figured is flat, and has
been produced by one blow which has split the pebble; it measures
4°4 inches.

A second rudely chipped stone has also probably been intended
for use in the hand; its point, however, has been broken away, so
that its exact original form cannot be ascertained.

A third, with curved hatchet-edge, is notched by use; and its
rounded shape bears a resemblance to some of the ‘ choppers” of
Le Moustier, La Madelaine, and of the British river-gravels figured
by Mr. Evans (fig. 443). It consists of a pebble broken so that
the blunt end may be easily grasped in the fingers.

Figure 3 represents a rude flake which has the bulb of percussion
and the surface of the chipping-block as well marked as in most
flint implements of the same kind. One of its surfaces is composed
of the water-worn surface of the pebble; and both its edges have
been chipped by wear.

Another form may be described as a flake of the rude kind found
in the valley-gravels and in the cave of Moustier. Its edges are
considerably worn by use. It is the only quartzite implement found
in the breccia, the two others being splinters.

The implements, fig. 4, of quartzite, and, fig. 5, of ironstone,

Fig. 4.

Oval quartzite implement, x 4. Oval implenient of ironstone, x }.
Cave-earth. «. Section. Cave-earth. a. Section.

have been chipped all round, and are of the oval shape common at

St. Acheul and at Moustier. The first corresponds in form with

that which I have figured from Wookey Hole in the Journal of the

Society, and the second with one which I have in my possession
Qo. Gr, Bs. No. 127. T

252 W. BOYD DAWKINS ON THE MAMMALIA AND

from the same cavern, although the former is thicker than the latter.
They measure respectively 3 x 2:2 and 2°4 x 1°6 inches.

Five quartzite hammer-stones bruised and broken by use were
also found, and many quartzite pebbles.

It may be remarked that paleeolithic implements of quartzite have
been met with in India in the Lateritic strata and in a river-deposit
in the valley of the Narbada (‘Cave Hunting,’ p. 426). In the
Museum at Toulouse they are also to be seen from the valley-gravels
of that neighbourhood.

Among the specimens sent to me by the Rev. J. M. Mello is a
ground and polished fragment of Silurian or Cambrian flagstone
(measuring 4x 3x 0-5 inches). It has formed part of a slab which
has been broken up.

13. The Flint Implements.

The flints which bear the trace of man’s handiwork in the cave
amount to upwards of 267. Among them the most striking are the
flakes which have been wrought into finely pointed lanceolate
forms. One of these, fig. 6 (measuring 3°35 x 0-95 inches), is care-
fully worked on the flat side (a), while the opposite surfaces are
worked at the extremities. It is carefully chipped at both ends.

A second and smaller specimen presents the same general form.
This type is identical with that figured by Mr. Evans from Kent’s
Hole, fig. 300, except that in the specimens from Robin-Hood Cave
the greater part of the working is to be seen on the flat side.

Fig. 6. :

Lanceolate flint flake, x}. Breccia. Lance-head of flint,x4. Breccia.

The double-pointed lanceolate flake (fig. 7), which tapers from
the wide central portion to the ends, from which apparently the
points have been broken, is also carefully worked on the upper half
of the flat surface ; and the chipping has been continued to the end
of the left lower side of the implement (a). On the other side (6)
the opposite edge is worked on the opposite surface, with the prac-
tical result of producing a twist in the edges analogous to that

TRACES OF MAN FOUND IN THE ROBIN-HOOD CAVE. 25a

which has been observed in neolithic arrow-heads, intended to make
the arrow revolve in its flight. It measures 3°8 by 1:42 inches. A
fragment of a second specimen was also found corresponding with
the upper part of fig. 7, and as nearly as possible of the same size and
form as a specimen in the Oxford Museum from Wookey Hole. In all
these implements the salient midrib of the flake has been left intact.

If this latter form (fig. 7) be compared with those figured in the
‘ Reliquize Aquitanice’ from Périgord, it will be seen that it bears
a strong family resemblance to some of those from Laugerie Haute
(A, pl. iv. figs. 7, 8,9). They are also of the same type as those
from the Pleistocene portion of the deposits at Solutré figured by
MM. Ducrost and Lortet (Archives du Muséum d’Histoire Naturelle
de Lyon, ii. pl. v.), and considered by M. de Mortillet to be cha-
racteristic of a stage in the Paleolithic culture.

Two other fragments of trimmed flakes were found, both with the
salient ribs worked carefully off and the flat inferior surface intact.

In fig. 8 an implement is represented formed of a flake with
the cutting-edge carefully and minutely chipped. The inferior sur-
face is, for the most part, flat and unworked, while the superior is
occupied by the natural surface of the flint-pebble. It is probably
a scraper, analogous to that from Kent’s Hole, fig. 392 of * Ancient
Stone Implements.’

Eleven flakes, with one of their extremities trimmed to a rounded
edge, were met with. They are of the usual type so common in the
caves of Périgord. One fragment is of the same form as the upper
half of fig. 396 of ‘ Ancient Stone Implements,’ and is composed of a
flake struck from the outer side of a flint-pebble.

Fig. 8. Fig. 9.

it i

i

i)

Mil

Flint implement. Flint implement.
Cave-earth. a. Section. Full size. Breccia. Full size. Breccia.

Figs. 9 and 10 represent two very singularly worn flakes like those
which have been figured by Mr. Evans (figs. 398,399,400) from Kent’s
Hole. The first consists of a flake with “an oblique straight scra-
ping edge, forming an obtuse angle with one side of the flake, and an

tT 2

254 W. BOYD DAWKINS ON THE MAMMALIA AND

acute angle with the other.” The angle made by this edge with the
major axis of the flake is 38° in two specimens, and 31° in a third.
In the second (fig. 10) both ends have been slanted off, and the

curved edgeof the flake has been worn, while the straight

one is sharp. The unworn edge has probably been in-

serted into a handle of some sort or another.

the first kind were met with, and one of the second.

A drill, or piercer, is represented in fig. 11,
formed out of a small flint flake.
shape to those figured from the caves of Périgord
in the ‘ Reliquiz Aquitanice.’
of implements is proved to have gone on in the
cave by the presence of large numbers of flakes and

a few chipping-blocks.

The general facies of the whole collection of
implements, and their association with the extinct
mammalia in the cave-earth and breccia, prove them

to be of Paleolithic age.

Table of distribution of traces of Man

It is of similar

The manufacture

Three of

Flint borer. Full
i Breccia.

g1ze.

in the Robin-Hood Cave.

|

a |
ea |
| She ieee |
| Sel | o, \oolee
: - ms |. Slee ul ones
| Ex) e |e | oe
o*| 3 | 2 | samme
| 4 CO lAl|nalea
} arrays nat He OT ej 20 onkan wane nc acer epemeemmbeseade| mute an * i
Potnited ‘antler (GW) .0c ails. senen tenaataanonaeee *
per cpanel Piri. 2a. chasse MRM sian seee cikiee aed eeetee *
Worked quartzite pebbles ........... ....../eeeee pS) MBC ror 2: 94
Quartzite hache (fig. 2) .....1.0....cscseecee[rerees * |
Quartzite Chopper ........ccsseerseeceecseeen|eeeees i eseeee[eeeeee 2 |
Quartzite flake (fig. 3)............ceseseeeeeeefereees %
” Te ee er et ee ee ee *
Quartzite implement (fig. 4) ...............Jeceee Bey |
Ironstone implement (fig. 5) ...............Je0e08 * |
Quartzite hammer-stones .............ceeeeleeere % |eereee| ooease 5
| Worked Bits co. i.ls.) Saag: « s deten|we eee 8 | 259 |...00 |267
| Lanceolate flint flake (fig. 6) ...........-...[eereee|eee eee ® [econ] 2
| Double-pointed lanceolate flake (fig. 7)...|-+++++|-++++ H eaeeee 2
Flint scraper (fig. S) ccseeeeouseereep esse. [onsen |
IONE ORETAPEEA” 2... 1. cnn Pemdae tender «nes as0)eeese: |eeeees Ho feeeees 11
Wrimmred flakes covisveansaon sce see asciees oaeive [errr seers H ewwees 2
Worn 5 bye OF BESO. ..0sdaiteaees jeeeees oes % [eeeeee 2
Worn ie CS ap a ee ee licdavealealeren eee 1
Piint-boter (NG, 1S) ooo cravs. tao psceeenaveue eeeeasjereeee H eveeee 1
OHippinie-WIGCks Joc seweseserssss ne ceasensseene pos ae % |eseee 2
Black Roman, POthery <3): vse aes chewsmas seine a giese tal eteee #
Sammie, WANG: .wsvsi fod tdest autedesneessnsatsuied are ate ete: #
Moditeval gherttss. Jo. :..tasassescel abs oleae | - oda caseels aia * |
| ' i

TRACES OF MAN FOUND IN THE ROBIN-HOUD CAVE. 255

14. Distribution of Implements in the Cave.

The distribution of implements in the cave represents, as may be
seen from the preceding Table, three distinct stages. During the
time of the deposit of the lower stratum Man is not represented
among the fauna of the district. While the cave-earth was being
accumulated, his presence is marked principally by the quartzite
implements formed out of an intractable material, and far ruder
than those which are generally formed out of the more easily
fashioned flint.

Of ninety-four worked quartzite pebbles, only three were found
in the breccia, while eight only of the 267 worked flints were met
with in the cave-earth (including fig. 8). The hunter, therefore, of
the cave-earth period used quartzite for most of his implements,
while that of the age of the breccia used flint, the overlapping of
the two materials in this cave being comparatively slight.

15. The Ruder Implements the Older.

The workmanship of the later of these two periods of human
occupation is of a higher order than the former. If, for example,
we compare figs. 6, 9, 10, 11 with figs. 2, 3, 4, 5, it is impossible
to resist the conclusion that the hunter of the breccia-age was better
equipped than his predecessor of the era of the cave-earth.

lf these groups of implements be compared with those found in
other paleolithic deposits, it will be seen that the older quartzite
division corresponds in its general form with that series which is
assigned by M. de Mortillet (‘ Matériaux,’ Mars 1869, “ Essai d’une
Classification ”) to ** the age of Moustier and St. Acheul,” and which
is represented in this country by the rude implements of the lower
breccia in Kent’s Hole. The newer or flint division, on the other
hand, contains among its forms more highly finished implements,
such as figs. 6 and 7, which correspond with those which are con-
sidered by M. de Mortillet to belong to “the age of Solutré,” and
which are found in this country in the cave-earth of Kent’s Hole
and Wookey Hole. In this cave, therefore, we have a direct relation,
in point of time, established between the rude types of implements
below and the more finished ones above, which is a fact of no small
importance in the classification of Paleolithic implements. In all
future cave-explorations it will be necessary to keep a keen look-out
for broken pebbles and roughly-edged stones, with scarcely any
marks of design, which may have served the ends of savages of a far
lower culture than those whose history has been revealed to us in
the caves of Périgord and of Belgium.

16. North-western Range of Paleolithic Hunters.

This discovery of implements in Derbyshire extends considerably
the known range of the Paleolithic hunter to the north and to the
west. Hitherto the Vale of Clwyd has been the district furthest to
the north in this country in which his implements have been dis-
covered. The fragment of human fibula in the Victoria cave has

256 W. BOYD DAWKINS ON THE MAMMALIA AND

proved his presence in the Pleistocene age in Yorkshire ; the caves of
the south of England prove that he wandered over the plains now
submerged beneath the waters of the Channel; those of Somerset,
Pembroke, and Herefordshire, that he inhabited the valleys of the
British Channel, the Severn, and the Wye. He is proved by M.
Dupont to have hunted the reindeer and mammoth in Belgium, on
the eastern side of the great valley which, during the latest stage of
the Pleistocene, extended across where now exists the German ocean,
and joined the eastern counties to Belgium. That savage tribes living
on the chase should be found on one side only of a great valley, while
the animals which they hunted were equally abundant on both,
was in the highest degree improbable. We now have proof that
their hunting-grounds extended as far to the west as the hills of
Derbyshire-—hills which in those times abounded with Bisons, Rein-
deer, Horses and Woolly Rhinoceroses, and in which (as I have shown
in my “ Essay on the Animals found at Windy Knoll”) there was a
continual swinging to and fro of migratory animals as in North
America. And further, we now have proof of the presence of the
Paleolithic hunters close to the glaciated region to the north-west,
which was probably covered with glaciers in the late Pleistocene
age. We have, however, no evidence as to the relation of the con-
tents of the Creswell Caves to the Boulder Clays.

Discussion.

Mr. Evans considered the cave to be of a most interesting cha-
racter. Implements of worked flints were here found in one bed
lying on another in which implements of quartzite occurred, the
latter being the ruder. That the implements from the lower bed
were less finished than those of the upper was doubtless to a great
extent due to the nature of the material of which they were formed.
Some of them resemble in character forms from the upper valley-
gravels of the Somme. Almost identical specimens had been ob-
tained, from what were apparently valley-gravels, near Toulouse.
On the other hand, the implements in the upper layer resembled
those of Solutré, Aurignac, and Kent’s Cavern, and possibly repre-
sented a period earlier than that of La Madelaine. Amongst the
specimens were some of great interest. One flake was worked off at
the end in a diagonal direction, like specimens from Kent’s Cavern.
There was also a borer like those from La Madelaine. Some of the
scrapers, however, might be taken to represent any period in the
French caves subsequent to that of Le Moustier, while one “ side-
scraper,” or “ chopper,’ was much like those from Le Moustier.
The paper was one of the greatest interest; and he hoped that it
would be supplemented by further reports as the work progressed.

Prof. Prestwicu remarked that the superposition of the bed con-
taining the more perfectly formed implements over that in which those
less highly finished were found, was better marked in this cave than
in any other in this country. The absence of the coprolites of the
Hyena in this as in the Brixham Cave was probably due to their

TRACES OF MAN FOUND IN THE ROBIN-HOOD CAVE. Dak

being acted on by water, during floods, and washed away. It was
not to be assumed, from the absence of any traces of man in the
lowermost deposit, that he did not exist in that spot at this period,
as he would hardly be likely to occupy dens along with the Hyzna.
Some of the flint implements found in the valley-gravels were of
extreme rudeness as compared with those of the high-level gravels,
though the latter were the older. Fineness of finish does not neces-
sarily prove more modern date, as is exemplified by the Shrub-Hill
implements ;.but much would depend on the material, and he had
not yet seen Mr. Mello’s specimens,

Mr. Ernerie# inquired whether the absence of coprolites from
these dens might not be accounted for by the cleanly habits of the
animals. The Carnivora, as a rule, were exceedingly cleanly in this
respect.

Prof. Hugues suggested that the difference of material might be
explained on the supposition that the people who left the quartzite
implements had lived in that or some other district where quartzite
was the only or most common material; while the flint was brought
by a tribe who came from a district where flint was abundant. The
rougher material did not of itself prove greater antiquity. Quartz
and quartzite had been used at all periods from that of the laterite
of India to that of the neolithic graves of Britain, and even later
elsewhere. So also he had found grey felstone implements in Wales
of neolithic as well as of palzolithic type; while polished weapons
of the same material were occasionally found in the fenlands near
Cambridge.

Mr. Howarrn stated that the African Hyzna does not resort to
caves, and he could not understand why the extinct Hyzna should
have done so. The remains of Hyznas had been found in caves on
the banks of the Lena and Obi in Siberia; and this led him to think
that possibly the Hyzena of that period, owing to the intensity of
cold, was in the habit of hibernating, as the bears of cold climates
do at the present day. The African Hyzenas only prey on sickly
members of the antelope tribe; and it therefore seemed to him doubt-
ful that those found in caves would have preyed on such large
animals as the Rhinoceros.

The PREsIpENT inquired at what distance from the cave flint was
obtainable, and also where quartzite pebbles could be found.

Dr. Merron considered that the development of man was more
clearly proved by the progressive improvements in the manufacture
of his implements than by the physical formation of his skull.

Mr. Bryyey stated that quartzite pebbles could be obtained within
seven or eight miles of the spot, whilst flints were abundant at not
more than about forty miles distant.

Mr. MEt1o, in reply, said that quartzite pebbles occurred abun-
dantly in some sands near the caverns. They were probably
derived originally from the Bunter. The flints might have been
obtained at no great distance, possibly from the valley of the
Trent; for some specimens were weathered, and evidently derived
from gravel; others, however, were probably obtained directly from

258 oN THE MAMMALIA ETC. FOUND IN THE ROBIN-HOOD CAVE.

the Chalk. He could not account, however, for the implement
fashioned out of ironstone, which is not found in the vicinity of
these caves. He had been under the impression that coprolites
were of frequent occurrence in the Hyzna-dens. The filling-in of
the cavern had doubtless been accelerated by the water overflowing
into it from the stream now flowing at the base of the Crags, but
which at that period ran at a higher level. There were slight
traces of a rearrangement of the bones such as would be effected
by water.

Prof. Boyp Dawxins, in reply, stated that he had no intention
of generalizing from this single example; but it was worthy of
note that here, as in M. Mortillet’s classification, the ruder imple-
ments were older than the more highly wrought flint flakes. The
Hyzena, like other animals, takes refuge in whatever place will best
suit him, and round the mouth of his den bones and coprolites ac-
cumulate. The absence of coprolites in this cave was probably due
to moisture dripping from the roof, which would dissolve and wash
them away; the caves in which coprolites are found were probably
dry. . The evidence of the cave-Hyzenas having preyed on large
game, such as the Rhinoceros, rests on the fact that the bones of
those animals found in the caves show marks indicating that they
have been gnawed. A sick or wounded animal, even though of
large size, would easily fall a prey to their numbers. The occur-
rence of the remains of Hyznas as far north in Siberia as the banks
of the Lena and Obi was an important fact and new to him. It in-
dicated a difference of climatal conditions in that country of a
marked character.

G. T. BETTANY ON THE GENUS MERYCOCH@RUS. 259

30. On the Genus Merycocuervus (Family OrxopontTIp&), with
Descriptions of two New Species. By G. T. Burrany, Esq.,
B.A., B.Sc., Shuttleworth Scholar, Caius College, Cambridge.
(Read February 2, 1876.)

(Communicated by Prof. Hughes, F.G.S8.)
[Prares XVII. & XVIITI.}

A VALUABLE series of mammalian remains, chiefly consisting of
skulls and portions of skulls, collected by Lord Walsingham in the
valley of John Day’s river, a tributary of the Columbia, in Upper
Oregon, in the winter of 1871-72, was presented by him to the
Woodwardian Museum, Cambridge. Most of the specimens were
obtained from near the head of a small stream called Bridge Creek ;
but a few were from the Great Cafion, higher up on John Day’s
River, nearly opposite Old Camp Watson. Lord Walsingham se-
cured indeed a rich prize in the portions of three days which he
devoted to a search for these remains.

The specimens were, fortunately, for the most part brought to
England in masses of the original rock, and had therefore the great
advantage of Mr. H. Keeping’s care and skill in developing them
from the matrix. His labours have given us a valuable and, I
believe, unique series of specimens, which serve to extend very
largely our acquaintance with the two principal genera of the family
Oreodontidee.

A summary of these fossils may here be given :—

'l. A large nearly complete skull, with lower jaw attached, the
zygomatic arches being, however, almost destroyed (Merycocherus
Leidy?, 0. sp.).

2. The greater part of a large skull preserving very completely
one zygomatic arch with posterior crest (Merycocherus temporalis,
nh. 8p.).

3. Another skull of Merycochwrus temporals, showing the part
anterior to the bifurcation of the sagittal crest.

4. Another skull of Merycochwrus, probably M. Leidyi, wanting
the greater part of the face.

5. A nearly complete skull of Oreodon major.

6. The greater portions of two skulls of Oreodon Culbertsoni (the
first-described species of the family).

7. Half of the frontal region of a Meryococherus larger than any
of the others.

8. Casts of the brain of a large and of a small species, with deter-
minable parts of bone attached.

9. Many portions of skulls, chiefly parts of upper and lower jaws
with teeth, including a number which show the canine and incisor
teeth.

10. Portions of limb-bones and a number of vertebre..

260 G. T. BETTANY ON THE GENUS MERYCOCHERUS.

Professor Hughes having kindly permitted me to examine these
remains, I now present an account of Merycocherus, so far as I am
at present able to give one, together with a description of two new
species, which are remarkable additions to the Tertiary vertebrate
fauna—hoping at some future time to furnish a more complete
account of the family and its relations to extinct and existing
Ungulates.

The progress of our knowledge of the Oreodontide has been com-
paratively slow, extending now over nearly 30 years. Fossils
belonging to this family were first obtained from the Mauvaises
Terres, Nebraska; and a jaw of a large species, supposed to be a
Palcotherium, was described in 1846 and 1847 by Dr. Prout, in the
‘American Journal of Science and Art.’ Gradually other specimens
came to light, and were described by Prof. Leidy, who gives the
following account of their mode of occurrence. ‘ The deposits of
the Mauvaises Terres are remarkable for the great quantity of fossil
remains of mammals and turtles they have yielded without further
exploration than picking them up from the surface of the country.
Detached from the neighbouring soft and readily disintegrating
rocks, the fossils lie strewn about, and have often attracted the
attention of the least curious of those who have traversed the
district. Many of the loose fossils have gradually been collected by
travellers and others. Of those coilected, by far the greater part
have been submitted to my investigation; and these have amounted
to the enormous quantity of between three and four tons in weight.”
Prof. Leidy collected and completed his earlier descriptions in 1852,
when he published, in the Smithsonian Contributions, “The Ancient
Fauna of Nebraska,” consisting of 126 pages, with 24 splendid
plates. In succeeding years the Mauvaises Terres were further
explored by Drs. David Dale Owen, John Evans, and F. V. Hayden,
who brought to Philadelphia large collections of fossils. Altogether
Prof. Leidy supposed, some years ago, that he had seen entire skulls
or portions of skulls of about 500 individual Oreodonts, a very
large proportion of them belonging to one species, Oreodon Culbert-
sont. In 1869 the results of his twenty years’ labour were pub-
lished, forming the seventh volume of the second series of the
‘Journal of the Academy of Natural Sciences of Philadelphia,’ and
entitled ‘“‘The Extinct Mammalian Fauna of Dakota and Nebraska.”
This work contains 472 pages and 29 plates, and gives a synopsis of
the entire mammalian remains of North America, with the most
complete references and the author’s valuable critical opinions. A
considerable portion of this work is devoted to the Oreodontide.
Still more recently, Prof. Leidy’s ‘ Contributions to the Extinct
Vertebrate Fauna of the Western Territories,’ published in 1873,
records further progress in reference to the Oreodontide, especially
Merycocherus, which, however, is still described only from incomplete
upper and lower jaws. In a letter to myself, dated Oct. 26, 1875,
Prof. Leidy says, “‘ In answer to your question about Merycocherus,
IT have seen no other remains than those described in my two
works” (those already referred to).

G. T. BETTANY ON THE GENUS MERYCOCHERUS. 261

The following is given as a brief definition of the family Oreo-
dontidee :—

Hornless Ungulates, in which the posterior region of the skull
presents a strong sagittal crest flanked by large temporal fosse.
Orbits closed. Zygomatic arches strong. No unossified spaces on
the side of the face. Teeth 44, a complete series. Canines large,
the lower being transformed premolars. Molars of typical rumi-
nant form.

The following more extended definition is copied, for convenience
of reference, from the “‘ Extinct Mammalian Fauna,” p. 71.

**QrzopontIpD#. ‘The skull has somewhat the form of that of the
peccaries ; the cranial portion especially resembles that of the camel.
It is hornless. The temporal fossz are large, and separated by a
median sagittal crest as in the camel. The zygomatic arches are
strong. ‘The orbits are closed behind by an arch. Large and com-
paratively deep fossee impress the lachrymal bones in advance of the
orbits. No unossified spaces occupy any part of the face. The
auditory capsules are variable in degree of development. The para-
mastoids are long and strong. The lower jaw is broad and deep
posteriorly, and impressed with a comparatively deep fossa below
the lunar notch. The teeth in both is form nearly unbroken
| 14 3-3
=44; well-developed incisors in both jaws, the fourth of the lower
jaw being a transformed canine, as in ordinary ruminants. Canines
well developed and strong in both jaws, suilline in their resem-
blance, those of the lower jaw being transformed premolars. The
anterior three premolars haying the crown in the form of a demi-
cone, with more or less rudimental elements at the base internally.
The fourth upper premolars and the true molars of both jaws con-
structed after the ordinary ruminant type, aud most nearly resem-
bling in form those of the Deer family.”

This description may be supplemented and modified by observa-
tion of our specimens as follows:—In the first place, the general
resemblance to the skull of the Peccary is very much less strong in
the larger species than in the smaller. The temporal fossz are in
some very large, and the median sagittal crest may be very strong
and prominent. The only breaks in the series of teeth are those
caused by the protrusion of the large canine teeth in each jaw into
corresponding intervals in the series of teeth in the other jaw. It
is better to amend the dental formula according to the definition
which defines the canine of the lower Jaw as the tooth which bites
in front of the first tooth in the superior maxillary bone; the first
premolar in the lower jaw of Oreodontidee will then be called a
caniniform premolar. Thus we get the normal dental formula, so
frequently met with in early Mammalia,

3—
arches. The formula of dentition is 2. 3 —, C. 79 PM. ss

- 3—3 1—1 4—4 3—3 44
U. wet) C. oeslte) pm. 4—4? Mm. 3 a ae .

The suilline resemblance attributed to the canines should be omitted,

262 &. T. BETTANY ON *HE GENUS MERYCOCHERUS.

as it gives rise to an erroneous conception of the nature and appear-
ance of these teeth.

The following are the distinctive characters assigned by Prof.
Leidy to the genera of Oreodontide established by himself (‘ Con-
tributions,” pp. 201, 202).

“Oreodon: Molar teeth with short crowns, as in the deer, and, as
in this, at maturity inserted by fangs. Anterior premolars straight,
with the diameters nearly equal, and with their points median or
nearly so. Face gradually convergent, conical. Infraorbital arch
narrow or of moderate depth, gradually declining upon the side of
the face. Infraorbital foramen small and situated above the third
premolar. Nasal orifice nearly as wide as high, and situated imme-
diately above the incisive alveolar border, as usual in most animals.
Premaxillaries and maxillaries remaining distinct from one another.
Incisive foramina of moderate size.

“Merycocherus: Crowns of the molars proportionately longer than
in Oreodon, and protruding gradually as they were worn away, the
anterior having their sculptured triturating surface obliterated before
the posterior are fully protruded. Anterior premolars with the
length and breadth exceeding the width, and the upper ones incli-
ning posteriorly, and with their points in advance of the middle.
Facial cone abruptly narrowed in advance of the orbits. Infra-
orbital arches deep and rapidly declining on the face. Orbits
smaller and more externally situated than in Oreodon. Infraorbital
foramen above the interval of the first and second molars. Nasal
orifice situated far above the alveolar border, as in the tapir, and
commencing below as an angular notch of the premaxillaries, which
are firmly coossified together and with the maxillaries. Incisive
foramen large.

“Merychyus: Teeth asin Merycocherus. Facial cone intermediate
in character to the latter and Oreodon. Infraorbital foramen
situated above the last premolar, or in a position intermediate to
that of Oreodon and Merycocherus.”

The following corrections must be made in the generic characters
of Merycocherus, in consequence of the greater perfection of our
specimens :—

1. The facial narrowing in front of the orbits is very little greater
or more sudden in Merycocherus than in Oreodon, although it is
considerable.

2. The orbits are not relatively smaller than in Oreodon.

3. The position of the infraorbital foramen relatively to the
teeth seems to be variable. In our specimens it is above the 3rd
premolar, or the interval between the 3rd and 4th.

4, The nasal orifice extends far above the alveolar border, and ap-
pears to be uninterruptedly open from the apex of the notch between
the premaxillaries, only half an inch above their continuous alveolar
border, to the horizontal projecting spine of the connate nasals.

Since the position of the infraorbital foramen varies in different
species, the genus Merychyus may very well be dropped, as it is
founded upon this single character. The teeth and portions of jaws

&, T. BEI'TANY ON THE GENUS MERYCOCHERUS. 263

assigned to Merychyus elegans and medius (Extinct Mammalian
Fauna, pp. 118-121) appear to belong to Oreodon, while the teeth
figured under the name of Merychyus major are referable on equally
good grounds to Merycocherus.

In the third of his great works above referred to (‘Contributions,’
p. 201), Prof. Leidy says ‘‘ Merychyus would appear to be the same
as Merycocherus, and the fossils which had been referred to it
belong to the same geological horizon.”

The following are to be added to the characters of Merycocherus:—

1. The side of the face is very nearly oblong (that of Oreodon
being a triangle truncated in front); and the whole upper surface
of the skull slants downward much less from behind forward than
that of Oreodon.

2. The temporal fossa is much wider and deeper than in Oreodon.

3. The outward curvature of the temporal ridges from the sagit-
tal crest is much more gentle and prolonged than in Oreodon.

4. The whole squamous portion of the temporal bone is very
large. The length, breadth, height, and outward curvature of the
zygoma are much greater than in Oreodon. The form of the zygo-
matic crest differs very considerably, being greatly extended trans-
versely, and looking principally forwards and backwards, while its
aspect in Oreodon is principally inwards and outwards.

5. The postglenoid processes are large and transversely extended.

6. The suborbital arch is unusually deep and strong.

7. The occipital foramen approaches nearly to a circular shape (it
is transversely oval in Oreodon).

8. The posterior part of the basicranial axis is set at a much
greater angle with the longitudinal axis of the palate than in
Oreodon; and its shape is that of a continuous curve of compara-
tively small radius. In J. temporalis the posterior extremity of the
basioccipital axis descends even below the plane of the palatal
surface.

9. The palate extends considerably behind the last molar tooth,
viz. to one third of the distance between the latter and the auditory
bullz. In Oreodon the palate only extends to about the level of
the last molar tooth.

I now proceed to a detailed description of this remarkable skull.

The skull in Merycocherus is elongated, high, and massive. Its
median longitudinal section would be an oblong with the length
between three and four times as great as the height. Viewed from
above, a high and strong sagittal crest is seen proceeding forwards
from a small transverse occipital crest. The sagittal crest is flanked
on each side by the convex wall of the cranial cavity, by a wide
temporal fossa, and a strongly arched zygoma, reaching out much
beyond the glenoid cavity. The sagittal crest bifurcates into two
gently and very symmetrically curved temporal ridges, which extend
to the posterior border of the orbit and then become lost on the
malar bone. The frontal surface of the skull is of considerable size,
and very nearly lozenge-shaped, the posterior angle being situated
between the two temporal ridges, the lateral angles corresponding

264 G. T. BETTANY ON THE GENUS MERYCOCHERUS.

with the posterior border of the orbit, and the anterior angle being
undefined, as the surface is continued by the nasal bones. The
frontal surface is generally somewhat convex; but it is slightly
concave behind, in the angle between the two temporal ridges. and
also in front of the supraorbital foramina, which are about three
quarters of an inch distant from one another, on a level rather
behind the middle of the orbit, and about half an inch in front of
the median point to which the nasal bones converge. From each
foramen a groove, more or less marked in different species, runs
forward and is gradually lost on the nasal bones.

The upper surface of the face is elongated and comparatively nar-
row. It is constituted entirely by the nasal bones, which do not
extend downwards on the side of the face. Hach nasal bone is
slightly convex transversely; and the two together form a very
regular convexity, which is greater or less in different species.
Posteriorly the outer margins of the two nasals converge gently to
a point in the middle line. Anteriorly they together form a pro-
jecting triangle, which is thickened below at the apex; this latter
overhangs the nasal aperture, but does not extend quite so far
forward as to overhang the extremity of the preemaxille.

The upper part of the nasal aperture is almost square; but infe-
riorly it is produced to an angle of 30°, ending in a rounded form as
a notch between the diverging premaxillary bones, which are
anchylosed at the alveolar border and for about half an inch above
it. The inner margin of each premaxilla bounds the nasal aperture
entirely in rather more than its lower half; the maxillary bone
bounds it at its widest upper part, while the nasals complete the
boundary above.

The lateral aspect of the face is nearly oblong, and almost vertical ;
and its height is, within a little, equal to its length. The posterior
part of the face is not more than three quarters of an inch higher
than the anterior. The facial surface presents, in front of the
orbits, the following inequalities. First, the lachrymal fosse, just
in front of the middle of the inner boundary of the orbit. This
fossa may vary from comparative shallowness to considerable depth,
and from a rounded to a nearly conical shape. Secondly, the gently .
convex continuation of the lower part of the zygoma on to the face,
graduating above into the lachrymal fossa, in front into the vertical
and flat anterior part of the face, and below into the depression
above the premolar teeth. Thirdly, the latter depression, more
marked anteriorly than posteriorly, and containing just below the
anterior termination of the zygomatic ridge the infraorbital fora-
men, a vertically-placed oval opening directed forwards ; its position
varies, as also does the depth of the fossa. Fourthly, a rounded
convexity corresponding to the fang of the upper canine tooth,
forming the extreme external part of the front of the face. Be-
tween this prominence and the upper vertical part of the side face,
an oblique groove, which probably served to convey a nerve to the
upper lip, passes downwards to the front of the premaxilla.

Unfortunately I have to describe the zygoma and its posterior

G. T. BETTANY ON THE GENUS MERYCOCHERUS. 265

crest from a single specimen of M. temporalis, where it is almost
perfectly preserved on the right side. I have thought it best to
describe the temporal fossa and its boundaries fully here, instead of
leaving it for the subsequent description of the species.

The long axis of the temporal fossa is directed forwards and out-
wards at an angle of about 40° with the middle line. Its extreme
length could not have been less than eight inches in MW. temporalis ;
but this cannot be exactly determined, because the upper and pos-
terior boundary is almost entirely broken off. But it may be stated,
to give some idea of its size, that, notwithstanding the high develop-
ment of the face and nasal cavities, the temporal fossa is larger than
the face. It is bounded behind by the transverse occipital crest,
thus extending almost to the extreme posterior limit of the skull.
This boundary was continued by the anterior of the forks into
which that crest divides, running outwards and somewhat forwards.
The height of this crest cannot be determined; but it appears pro-
bable that it was not less than an inch high. The external angle
of the temporal fossa, situated midway on its lateral aspect, is
occupied by the great posterior transversely-placed zygomatic crest.
The crest slopes away in front so as to leave no external boundary
- to the fossa. Internally the fossa was bounded by the high sagittal
crest, which appears to have continued at the same level as the
frontal surface, or very little lower than it. The bifurcation of the
sagittal crest forms the antero-internal wall of the fossa, which was
in front limited by the posterior orbital arch.

The whole internal surface of the temporal fossa is very smooth,
and gently and beautifully curved. The lateral wall of the cranium,
which may be said to constitute the internal floor of the fossa, is
convex, and is formed almost equally by the squamous and the
parietal bones. Their suture is along the line of the most marked
prominence of the convexity, and lies midway between the middle
line and the outer boundary of the fossa. This prominence, together
with a slighter ridge above it and parallel with it, constitutes a
rather broad shallow groove. I would suggest that this ridge marks
the limit separating the upper and anterior from the lower and
posterior portion of the origin of the temporal muscle. Near the
posterior termination of this ridge, which is on the parietal, and
between it and the squamous suture, is a large foramen which Dr.
Leidy identifies as venous. There is in our specimens a smaller
foramen just behind and below this.

The internal lateral boundary and floor of the fossa in its middle
portion presents, from above downwards :—first, a considerable con-
cavity formed by the nearly vertical rise of the sagittal ridge ;
secondly, the convexity on the side wall of the cranium ; and, lastly,
another deep but gently curved concavity directed downwards, for-
wards, and outwards along the inside of the posterior and external
ridge, which extends forwards from the occipital crest.

The extreme depth of the fossa depends on the height of the
sagittal ridge; but it was at any rate equal to two thirds of the
total length of the fossa.

266 G. T. BETTANY ON THE GENUS MERYCOCH@RUS.

Unfortunately the surface of the postero-inferior part of the orbit
and the suborbital arch are broken away on the right side in our
specimens, so that it is impossible to determine how the very deep
suborbital arch graduated into the much less deep middle portion of
the zygoma. Certainly the zygoma projected forward and was
intercalated between the prongs of a fork formed by the processes
of the malar. The two prongs of the fork seem gradually to have
diminished to nothing, while the zygoma descended to the same
level as the lower margin of the malar. The broken-off anterior
extremity of the zygoma, as we possess it, is about half the height of
the suborbital arch. Its external surface is smooth and gently
rounded. Its upper surface is as broad as its external is high; in
front it is obliquely directed inwards; but posteriorly its direction
changes until it is quite parallel with the middle line of the skull.
At the same time the internal height of the zygoma becomes very
much less than the external, and ultimately thins away to a sharp
edge.

The upper surface of the zygoma in passing backwards now
becomes very much wider. Its inner edge curves round until it
passes transversely inwards to join with the lateral wall of the
cranium at about its anterior third; while its external border rises
in a concave curve, passing inwards at the same time, till it has
reached a height of 23 inches above its origin, and of 3} inches
above the lowest part of the zygoma.

This crest and arch differ from those of Oreodon in their much
greater proportional size and in their form. In Oreodon the
zygoma continues the direction of the infraorbital arch, or even
curves inwards to the cranial wall. In Merycocherus it diverges
considerably further from the middle line than the infraorbital arch.
In Oreodon the crest on the hinder part of the zygoma is very little
higher than the rest of the arch; it is almost constituted by the
arch thinning in a vertical direction behind; and its concave surface,
of no great extent, looks mainly inwards, only slightly forwards. In
Merycocherus the crest is much higher than the rest of the zygoma,
and, far from being constituted by the compression of the latter, it
is set at right angles with its anterior part; and its large concave
surface looks directly forwards. Its concave area is about six
square inches; in a good-sized Oreodon it is only about one square
inch.

The posterior aspect of the skull presents, in the middle line
below, the nearly vertical and circular occipital foramen, flanked by
the antero-posteriorly compressed occipital condyles. The upper
and posterior of the two surfaces produced by the compression of
the condyle, is separated from the lower and anterior by a blunt
ridge. The former is convex and oval, with an acute termination
below. The condyles are considerably separated from one another;
and there appears to have been an emargination of the basioccipital
in the middle line.

The lateral margins of the foramen are continued above into
blunt ridges, which soon join and then bifurcate to form strong

G. T. BETTANY ON THE GENUS MERYCOCH@RUS. 267

sharp projecting ridges which extend upwards to the transverse
occipital crest, enclosing between them a fossa of moderate size and
varying depth. Outside these ridges, on each side, is a much larger
and deeper fossa, bounded above by the lateral occipital crest pro-
ceeding obliquely downwards and forwards to the zygoma, below by
the occipital condyle. of either side, and externally by the large
blunt convex postauditory process. Below this is the large par-
occipital process, which is triangular at the base, and extends down-
wards and somewhat inwards for an indeterminate distance closely
behind and externally to the auditory bulla. The smallest length
that can be allowed for the paroccipital process, consistently with the
size of its basal part and its length in Oreodon, is aninch and a half.
The external auditory meatus looks upwards and outwards in the
angle between the posterior zygomatic root and the postauditory
process.

The hinder parts of the zygomatic arch, the postglenoid processes,
the auditory bulle, and paroccipital processes are important features
in the widely extended posterior part of the under surface of the
skull. The under surface of the basioccipital is convex, with a
strong median longitudinal ridge expanding into a blunt eminence
of considerable size, which is gradually lost anteriorly on the basi-
sphenoid. The latter bone, as well as the presphenoid, is smooth,
nearly flat, and narrow. The base of the cranium forms a very
regular curve, extending very much upwards and forwards.

The auditory bulle, in the two specimens in which they have
been perfectly preserved, arise just in front of the foramina lacera
posteriora internally, and of the paroccipital processes externally.
Their hinder limit extends slightly behind a line joining the poste~
rior surfaces of the postglenoid processes. The bulle are large and
somewhat nipple-shaped and compressed, their long axis making an
angle of about 10° with the median line of the skull. The external
surface is more irregular than the internal, which may be perfectly
smooth. The large foramina lacera media are situated at the inner
anterior side of the base of the bulla, and are elongated parallel
with the bull, so as to terminate acutely behind. Hach foramen
is divided by a process projecting from the anterior boundary into a
smaller anterior and a larger posterior region.

The glenoid articular surface, with its postglenoid process, is
situated externally to the auditory bulla, separated from it and the
paroccipital process by a deep sigmoid fossa. The fore part of the
articular surface is nearly horizontally situated. It is slightly
convex antero-posteriorly, shelving away in front and laterally,
where it is continuous with the under surface of the zygoma. <A
posterior vertically descending portion is contributed to the glenoid
surface by the anterior fiat smooth surface of the very large post-
glenoid process. ‘This latter is transversely extended, somewhat
compressed antero-posteriorly, and strongly convex behind. It pro-
jects less deeply than the auditory bulla by a quarter or three
eighths of an inch.

The under surface of the zygoma extends far outward beyond the

Ged, GS. No, 127, U

268 G. T. BETTANY ON THE GENUS MERYCOCH@RUS.

glenoid surface. Its breadth is about equal to the distance between
the middle line and the outer edge of the glenoid surface. Between
the postglenoid tubercle and the outer margin of the zygoma the
latter forms a wide concavity ; for the lower edge of the zygoma at
its most prominent part descends to as low a level as the apex of the
auditory bulla. The surface is smooth. The lower fork of the
malar extends backwards, beneath and internally to the zygoma,
lying ou it, but not anchylosed with it, as far as the region of the
great transverse crest. It ends in a thin flat rounded lamina. The
posterior surface of the crest is smooth and slightly concave below.

The palate is broad and arched, and only widens slightly between
the molar teeth. There is an anterior major concavity between the
premolars and the first two molars of each side, and a posterior
smaller concavity behind the level of the last molars. The palate
is continued behind the alveolar processes for about one third of the
distance between them and the auditory bulle; its width di-
minishes to about half that of the anterior portion; and-there is a
deep notch on each side between this posterior portion of the palate
and the alveolar processes. The palatal surface ends by a posterior
edge which possesses a median prominence and two lateral shallow
concayities, which pass on each side into the vertical pterygoid plates.
The latter extend, at about the same level as the palate, nearly back
to the auditory bulle, and consequently enclose a large and deep
median fossa behind the palate. I am not able to describe the ptery-
goid fossee; they were certainly comparatively small.

The teeth do not need much remark, since Leidy has described
the teeth of Oreodon so elaborately in the ‘ Extinct Mammalian
Fauna of Dakota and Nebraska,’ and those of Merycocherus do
not differ from them in many points except those mentioned in the
generic definition given above. One of the most conspicuous cha-
racters appears to be the more rapid increase of the molars in size
from before backwards, and the increased strength of the external
columns of the upper molars.

The lower jaw exhibits a considerably: greater size and prominence ~
of the lower posterior rounded angle than in Oreodon. The con-
dyles are very narrow, and are transversely elongated. The as-
cending rami are vertical, and consequently parallel to one another.
Their external surfaces are concave posteriorly and convex ante-
riorly, the convexity being lost on the horizontal rami. The as-
cending rami are most remarkable for their breadth, though their
height is considerable. The horizontal rami are not very deep;
they are vertical, but have a-strong convexity externally, proceeding
from behind and above obliquely downwards and forwards to the
lower extremity of the ramus. The mental foramen is large, oval,
and directed obliquely forwards; it is placed just below the third
premolar (the caniniform tooth being reckoned as the first). The
anterior symphysial surface of the jaw is broad, in correspondence
with the breadth of the row of incisor teeth with the caniniform
premolar. It is twice as broad above as below, and is regularly
convex from side to side. From its lower extremity it ascends at

G. T. BETTANY ON THE GENUS MERYCOCHERUS. 269

an angle of 40° with the horizontal. The two halves of the man-
dible do not appear to have become anchylosed at any period. They
are separated at the symphysis by an eighth to a quarter of an inch
in our specimens ; and the opposed surfaces are very simply denticu-
lated. The posterior and inferior margins of the mandible are
rounded and thickened, but are less thick at the angle. The lower
margins of the horizontal rami very gradually approach each other,
and meet anteriorly by forming a small semicircular curve at the
symphysis. Nothing can be said about the coronoid process or the
inner surfaces of the rami, because the space between them in our
M. Leidy? is filled up by peculiarly hard matrix.

Since the species of Merycocherus founded by Prof. Leidy are
chiefly based on their teeth with portions of jaws, and I am not able
to determine their identity or non-identity with those in the Wood-
wardian Museum, I think it better to describe the two principal
forms which I have determined as new species, especially as they
are distinguished on grounds independent of the teeth. -

MERYCOCHGRUS TEMPORALIS, n. sp. Pl. XVII.

This species is founded mainly upon two specimens, numbered
2 and 3 in the list given above. The first of these is of unique
interest. Anteriorly it is vertically truncated in the region of the
third premclar. The frontal surface and the lateral walls of the
skull are entire. On the right side the margins of the orbit are
much injured, and the infraorbital arch is almost entirely lost ; but
the great zygomatic crest behind this is complete. Unfortunately
the ridge leading from the posterior part of the zygoma to the trans-
verse occipital crest is broken off, as is almost all the sagittal crest.
On the left side the orbit with its infraorbital arch is almost perfect,
but the zygomatic crest is deficient. Posteriorly the lateral occipital
fosse are plain, but they lack the bounding projecting ridges; the
condyles are broken off. On the under surface of the skull the sur-
faces of the basioccipital and basisphenoid are almost perfect ; the
auditory bulla, the glenoid surfaces, and the postglenoid processes
are in excellent preservation ; and the palate, so far as the anterior
truncation, is complete. The teeth, however, are broken off, not
simply worn down.

The other specimen is more complete anteriorly, and includes
considerable portions of the canines and the nasals up to their ante-
rior terminations. The sides of the face are more or less damaged ;
but the orbits can be pretty well defined, and the suborbital arches
are entire, with a little piece of the zygoma on the right side. The
greater part of the frontal surface is present; but nearly the whole
skull behind the posterior boundary of the orbits is broken off.
Below, the palate is in a very imperfect condition; the outlines
of a number of teeth are seen broken off down to the level of
the palate.

The chief characteristics of MW. temporalis are the following :—

1. The lachrymal pits are wide and shallow.

2. The infraorbital arch is directed somewhat 1 outwards

U

270 G. T. BETTANY ON THE GENUS MERYCOCH@RUS.

and downwards; its surface is remarkably even, unmarked by
furrows or prominences.

3. A plane lying on the orbit is directed more obliquely down-
ward and forward than the infraorbital arch. Thus the eye would
look considerably forwards and upwards as well as outwards. The
orbit is larger and more oval than in M. Leidyt.

4, The very great outward extension of the zygomatic arch.

5. The great size of the posterior zygomatic crest and of the tem-
poral fossa, whose long axis makes between 10° and 15° greater
angle with the middle line than in the following species.

6. The greater breadth and size of the postglenoid process.

Merycocnarvs Lempyi,n.sp. Pl. XVIII.

I desire to connect this species with the name of Prof. Leidy,
whose magnificent works have done so much for our knowledge of
the palzeontology of North America.

It is founded on a large skull, which is nearly complete, lacking
unfortunately the posterior part of the zygoma with its crest on
both sides, and also the orbit and the suborbital arch on the left
side. ‘The face, however, is almost perfect, the fore part of the
nasals only being injured. The mandible is firmly united by the
matrix to the skull almost i situ, but a little laterally displaced
behind. The front teeth are a good deal broken; the lateral view
of the teeth of the upper jaw is very good, especially on the left
side; the lower premolars are also partially visible. The occipital
crest is injured behind. The comparatively small portion of the
base of the skull that exists behind the glenoid articulation is well
exposed; but the auditory bulls and paroccipital processes are
almost entirely destroyed. The occipital condyles are very per-
fectly preserved.

The characteristics of MW. Leidyi are as follows, in contrast to
those of M. temporalis given above :—

1. The lachrymal pits are conical at the bottom and of consider-
able depth ; there is a moderate-sized lachrymal tubercle on the
margin of the orbit.

2. The infraorbital arch, of somewhat greater depth than that of
M. temporalis, is almost vertically situated. Instead of being flat,
it presents from above downwards :—first, the raised rim of the orbit ;
secondly, a broad groove passing round from the fore part of the
lachrymal depression and backwards on the lower fork of the malar,
being overhung by the strong prominence of the anterior part of
the zygomatic process itself where it fits into the malar; and,
thirdly, the convexity of the malar process of the maxillary bone,
which gradually dies away forwards on the side of the face, and
becomes quickly lost behind.

3. The orbit is considerably more vertically placed than in
M. temporalis, and looks much more directly outward; it is nearly
circular.

4. It is impossible to determine how far the zygomatic arches
extended outwards; but from the direction of the parts existing,

G. T. BETTANY ON THE GENUS MERYCOCHERUS. 271

especially of the inferior fork of the malar, it would appear that
they could not have extended outwards so far as in WM. temporalis.

5. The smaller size of the temporal fosse, and the more acute
angle (by about 10°) which its long axis makes with the middle
line of the skull. The cranium itself is considerably smaller in pro-
portion to the size of the skull than in M. temporalis.

6. The postglenoid = are much less broad and propor-
tionally deeper.

The relative position of the auditory bulle, postglenoid processes,
and occipital condyles differs in the two species. A line drawn
through the postglenoid and paroccipital processes to the middle
posterior point of the basioccipital in WM. temporalis makes an angle
of about 60° with the median line; the same angle is only about
45° in M. Leidyi. The absolute distance is greater in the latter ;
and the postglenoid processes, instead of being opposite the middle
of the auditory bulla, are situated opposite its anterior extremity.

The following are the dimensions of I. temporalis :—

Probable extreme length of skull................ 14
Length of the larger portion in the Woodwardian
a head wae Se ne Dey ae, eee 92
Reneth.of the pmall portion ..- 0... sane ee een 72
Half width across frontal surface ................ 22
Distance from median line to external limit of zygo-
MEAS MOR ot Lk EE tht aS ys os Bide a'
Greatest width of brain-case.. 6... e eee ee O.
Width of palate within teeth. ..... 2... 22s eennes
Breadth between outer surfaces of postglenoid pro-
Sen OE ee ee ere ere eet eee
Depth of auditory bulla GMMR 2. ate 8 een 36
Depth of auditory bulla in front ................
Height of zy gomatic See OE ih ite OO oka fate
Depth of pkall ip malidie mos Ook. woes ceigels bs
Probable depth at posterior extremity ............
Pigment CiaMener GF. OFT a nia 5 mn he Fees
Depth CU y COLE SS SR Seen eee hy ee
Width of face just in front of EE os isan 5 oe
Heiett at face mm fromt). 2... 2 cl es
Length eee MORTAR SN te oie Sa sk Sh 'enai a ede

PWR OWE oO won
S| OCOD] Blo

{OO HS | >| 00/—1

Dimensions of MW. Leidyt :—
Total length of skull (possibly the actual length was

1 inch greater when uninjured) .............. 123
Half width across frontal surface .............-.. a=
Greatest width from middle line to exterior of zygoma,

pe dae MSC ECR Se So teclate Ses 6 = acs ee OF
Greatest width GF Mraimi—case -~ +). sok eh ce ee 2i
Width between external surfaces of postglenoid pro-

SERS ne weet tet oir sey os. ws a's heel ee era 4;

2f2 G. T. BETTANY ON THE GENUS MERYCOCH@RUS.

inches
fone diameter Of OTDIG. 6 cas en See er as 1t
Wepth of intractbicel arene rye ee ee ere eens 1,
Width of face 23 inches in front of orbit .......... 255
Heieht of face anteriorly oo. en en eine 32
fengeth ‘of last ‘molar tooth” 20a eee ee ee 1i
Extreme length ef lower jaw... - >...) a. ee 10

Appendix on the Geological Position of the Oreodontide.

It has been thought desirable to add to the foregoing paper a
view of the species of Oreodontide already established, the loca-
lities from which they have been obtained, and thé age of the
deposits, so far as they have been determined. There can be little
doubt that the number of species is less than is here indicated; but
it would be futile at present to attempt such a revision as is required.
For similar reasons the name Merychyus is retained for the species
characterized by Prof. Leidy.

Oreodon affinis?, Leidy. Bed A, Mauvaises Terres, White R., Dakota. Miocene.

hybridus ?, Leidy.

— bullatus? , Leidy.

Culber tsoni, Leidy. Beds B, OC, and iD Mauvaises Terres, White R.,
Dakota. Miocene.

— Culbertsoni, Cope. North Colorado, equivalent to White-river beds.

Miocene.

— Culbertsoni, xobis. Bridge Creek, John Day’s R., Oregon. Miocene.

gracilis, Leidy. Beds B, C, and D, White R., Dakota. Miocene.

gracilis, Cope. North Colorado, Miocene.

major, Leidy. Bed D, White R., Dakota. Miocene.

major, zobis. John Day’s R., Oregon. Miocene.

superbus, Leidy.

occidentalis ?, Marsh.

Merycocherus proprius, Leid; yy Bed D, "Niobrara. R., Nebraska. Miocene.

temporalis, nobis. John Day’s R., Oregon. Miocene.

Leidyi, nobis.

rusticus, Leidy. Sweetwater R. , Wyoming. Pliocene.

Merychyus elegans, Cope. North Colorado. Miocene.

elegans, Leidy. Niobrara R., Nebraska. Pliocene.

medius, Ledy. = i

major, Cope. North Colorado. Miocene.

—— major, Leidy. Niobrara R., Nebraska. Pliocene.

2? ” ”

” 9 ”

>

The John-Day’s-River deposits are indicated as Miocene in accord-
ance with Prof. Leidy’s conclusion that the zoological character and
state of preservation of the fossils renders it probable that the for-
mation to which they belong is of contemporaneous age with those
of the White River, Nebraska. With regard to the first-named
species in the list, Prof. Leidy says that certain facts “indicate
Oreodon affinis, hybridus, and bullatus to be species which preceded
the others in time, and were perhaps their ancestors, from one or
another of which they may have been derived.” A further state-
ment of interest, borne out by the above Table, is that “Meryco-
cherus proprius appears not to have inhabited the same locality with
the Oreodons, at least contemporaneously.”’

pee

a Si Ss

ee is

a

~

ie Mite hi ie * iy a8
Pe Marin th

: ra) Bair eck

Quart. Journ Geol. Soc Vol XXXTPL XVI.

Mintsrn Bros imp
i 1
Pal

C.L.Gnesbach. bo /

MERYCO CHGRUS TEMPORALIS.

CL.Griesbach

MERYCOCHERUS LEIDYI.

Quart. Journ Geol Soe Vol. XXXL. Pl. XVII

Mintern Bros imp.

G. T. BETTANY ON THE GENUS MERYCOCHE@RUS. a8

EXPLANATION OF THE PLATES.

Puate XVII.
Merycocherus temporalis, half natural size. All the views are from the
same specimen.

Fig. 1. Upper view of skull.
2. Skull seen from behind.

3. View of right side; orbital region partly restored from the
left side.

4, Posterior basicranial region.

(au.b, auditory bulla; yo, paroccipital process ; p.gl, postglenoid process. )

PuatE XVITI.
Merycocherus Leidyi, two thirds natural size.

Fig. 1. Right side of skull, somewhat completed from the left side.
2. Front view of face of same specimen.

(0c.c, occipital condyle; po, paroccipital process; pg/, postglenoid process.)

Discussion.

Prof. Duncan remarked that these late American forms seem to
be very variable. Notwithstanding its name, the affinities of this
genus to the Pigs was but small. The teeth appeared to be very
variable; and this tendency to variation would seem to indicate a
near passing-away of this type.

274. H. J. JOHNSTON-LAVIS ON THE

31. On the Triassic StRATA which are EXPOSED in the CLIFF-SECTIONS
near StpmoutH, and a Nove on the occURRENCE of an OssIFEROUS
Zone containing Bonss of a Lasyrintuopon. By H. J. Jounston-
Lavis, Esq., F.G.8. (Read March 22, 1876.)

In describing the locality and geological position of the vertebrate
fossils which were obtained from the Triassic rocks near Sidmouth,
perhaps it would be as well to commence with a description of the
coast-sections for a short distance east and west of that town.

Starting from the east, we find that the Marl* (which is the upper-
most subdivision of the Trias of South Devon) makes its appearance at
Branscomb Mouth, exposed beneath the Greensand and Chalk in the
cliff-sections, and in the ramifying valley cut through by the small
stream which runs out to sea at this point. It nowforms the lower
portion of the cliff under Littlecomb Hill and Branscomb Hill, being
overlain by the Greensand and Chalk until it is exposed inland at
Weston Mouth by the action of the little stream called Weston
Water, which runs out here.

It again forms the base of the cliff under Dunscomb Hill, being
still overlain by the Greensand and a small patch of Chalk; again it
is exposed and partly excavated by the little brook at Salcomb
Mouth ; thence it forms the base of the cliff under Salcomb Hill,
being now capped by the Greensand alone; itis then largely exposed
on the surface in the valley of the river Sid.

A few yards east of where the Sid runs into the sea, the Upper Sand-
stone (of Mr. Ussher) crops out, forming a cliff overhanging the Sid, and
constituting the bed of the stream for half a mile from its mouth, and
also the whole valley except where covered by gravel, which Mr.
Ussher tells me is in some places 15 feet thick, containing a bed of
peat about a foot in thickness. The gravel is chiefly composed of
chert, and contains teeth of Elephant, numerous specimens of which
have. been found by Mr. P. O. Hutchinson and others.

To the west of Sidmouth, at the end of the Parade, we meet with
a low projecting cliff, called Chit Rock. Mr. Ussher, who has sur-
veyed this district, tells me he has met with no evidence of a fault
haying existed in the valley; and therefore we may conclude that it
is the continuation of the small exposure of sandstone which is seen
to exist east of the river, as in section, fig. 1(p. 276). At the western
end of the Chit Rock we find a fault which has given the Chit Rock an
upthrow of at least 40 feet ; but it is very possible it may be as much
as 80 feet, since it has no marl capping it, and in its lithological
character resembles the middle of the Upper Sandstone. We see, on
the western side of the fault, the Marl brought down within a short
distance of the beach, there being a small mass of Sandstone exposed
beneath it. All the Triassic beds from Branscomb up to the present
place dip gently to the east; but now we find them dipping to the
west ; this only takes place for the distance of about half a mile;

* See Ussher, ‘ Geological Magazine,’ Decade IT. vol. ii. No. 4, April 1875.

TRIASSIC STRATA NEAR SIDMOUTH. 275

for the Sandstone which had disappeared, soon makes its appearance
again, having formed a synclinal curve. During the whole of this
distance it has been covered by the marl, which, as it advances
westward becomes thicker, being less denuded; it has cappings of
Greensand and chalk-gravel at Peake Hill and High Peake. The
Sandstone continues to rise gradually to the westward ; but the Marl
and overlying Greensand have been cut down by atmospheric denu-
dation, forming Windy Gap, which separates High Peake and Peake
Hill—High Peake being the higher of the two, but resembling Peake
in every other respect. The Sandstone gradually rises until, at a
short distance to the west of High Peake, the Marl has been
entirely denuded (save in a few places where through faulting it
has been brought to a lower level) and it appears on the surface.

The upper marls are variegated, and especially in the higher part,
east of Sidmouth, contain very thin layers of a greenish-grey sand
mixed with a large quantity of mica, intercalated with layers of marl,
varying in thickness up to two inches, but of the same light colour ;
they show ripple-marks, and occasionally contain pseudomorphs of
rock-salt.

The marls between Weston and Branscomb Mouths contain a large
quantity of gypsum, which at one time was worked at Branscomb
Mouth. A few small veins are to be seen between Salcomb and
Weston Mouths. The marls also contain bands of potato-stones
enclosing a cavity lined with calcite. Mr. H. B. Woodward objects
to my giving them the name of potato-stones, as they do not con-
tain quartz crystals.

The Sandstone, especially at its upper part, where it resembles
very much in lithological characters the upper beds of marl, contains
a large number of pseudomorphs of rock salt, ripple-marks, and
sun-cracks; but in no case have I met with rain-marks, which we
might expect; neither have I met with any foot-prints. Some of the
upper sandstones effervesce with hydrochloric acid.. Between High
Peake and Otterton Point the sandstones contain spherical masses
sometimes almost like a cannon-ball, composed of iron pyrites;
these are washed out of their matrix and lodge at the bottom of
the little rocky pools.

The sandstones also contain curious irregular branching-shaped
masses of a harder texture, which withstand the weathering and
give the cliff a rugged aspect. It is worthy of notice that, at the
points where the Marl reaches down to the beach, there are no
reefs on the foreshore opposite, but a beautiful fine-grained red sand,
except where large blocks of chert have fallen from the Greensand
capping the cliffs; but wherever the Sandstone appears above the
beach one sees large reefs running out to sea for nearly half a mile at
low tide ; it does not seem to be the sandstone itself which withstands
the weathering, but these curious hard masses contained in it, since
they occur in all the projecting points of the cliff formed by the
sandstone strata.

Last autumn, while on a visit to Sidmouth for the second time, I
had the good fortune to find the bones of a Labyrinthodont.

H. J. JOHNSTON-LAVIS ON THE

(

276

hy

a

Ossiferous -
Zone

Seems +e

\
woot ee ee eee

River Sid.

ut
oe = — — —_. — oe Hy a

Sidmouth.

"[ Sy 908 suaq4o] Jo Moreuedxe 10,7

Salcombe
petted: ||| |
Ossiferous ________....
Zone

’
:
;
oO
2-8
g5
oo
|
S
a)

ro io) -Q a ® 5

2 ed S FE ee : >

i ao ea ga He =

S a 5 F oe

"pvayy laagq 02 ynoupry mowf wous09—'g “BIT
"TOACAS-HTeYQ) *a "¥TVyO *p *puvstoaly °a "ICT 9 ‘auoyspuLg *”
SOW WMO “OL "Yooy, pop-jo-uryy “g WOOY JY “9 ‘OORT SOUIS poyLjep yooy “F ‘Wory

"YOY [[OYSostoj10T, *G ‘qULOg Weal *J, "‘(oOWY Joyog op}WT “c ‘yO0yy UAOFT “E ‘Avg, weapey *T

vy SE
!
= ' f
= =
1 = SS 1 !
~ sent 1 |
' ‘ \ i
| : : “AA
a a 2 2 rf
5 = & os 5 3
o Pa) Ss
aa a ‘. Ais “8 Fe
3 : x gS
o i) Fons] [=]
mo e as

‘yynowpry 02 ing wospoT mort uoyoas-fyp ayy fo marg oha-spug—T ‘3tq

TRIASSIC STRATA NEAR SIDMOUTH. 277

These were brought to light at different periods during my month’s
stay. It may be mentioned that those described by Prof. Seeley are
merely the bones which I considered capable of identification ; for I
met with a great number of small fragments dispersed throughout
a particular series of beds situated about 10 feet from the top of
the Sandstone. These bones were mostly found in fallen blocks
which were derived from these beds in a little cove known as Picket-
Rock Cove. It may be well termed an ossiferous zone, as it does not
consist of one single bed, but of from one to four; not that I mean
to imply that bones are only found in this zone, since Mr. Whitaker’s
Hyperodapedon was found at the very bottom of the Sandstone. This
zone is characterized by lithological differences, inasmuch as the
matrix is composed of much coarser sandstone, containing here and
there masses of marl varying in size from that of a pea to that of
a hen’s egg. It is nearly hard enough in some places for building-
purposes. In these beds ripple-marks are very plentiful. The
fragments of bone which are found in this zone seem to be very
slightly waterworn.

I cannot conclude without expressing my thanks to Messrs. H. B.
Woodward, W.A. E. Ussher, and W. Whitaker for the kind assist-
ance they have afforded me, and to Mr. P. O. Hutchinson, of Sid-
mouth, for the artistic diagrams with which he has furnished me.

P.S.—Since writing the above I have received from the Rey. S. H,
Cook some fragments of bone obtained by him about twenty years
ago from the same zone west of Sidmouth, as well as one which he
recently obtained from the small outcrop of sandstone with the os-
siferous zone on the western side of the Chit-rock fault.

Below the ordinary red sand on the beach one finds a stratum of
black sand which has been derived from the sandstone, but is of
ereater specific gravity than the red sand, which accounts for its
position on the beach. Mr. R. W. Cheadle has kindly made a quali-
tative analysis for me, with the following rough results—Silica,
magnetic iron oxide, manganese, titanium, and alumina.

[For the Discusston on this paper see p. 283, |

278 H. G. SEELEY ON A PORTION OF

32. On the PosrERIoR Portion of a Lowrmr Jaw of Labyrinthodon
(L. Lavist), from the Trras of SipmoutH. By Harry Govier
SrEtey, Esq., F.L.S., F.Z.8., F.G.8., Professor of Physical Geo-
graphy in Bedford College, London. (Read March 22, 1876.)

[Puate XIX.]

Norwitustanpine all that has been done of late years by Prof. Hux-
ley and those who have continued his labours, for the elucidation of
the Labyrinthodontia, less is probably known of the structure and
affinities of that group than of any other long-established Order of
vertebrates. It may perhaps be an open question whether Laby-
rinthodonts can rightly be referred to an existing class of animals;
for the nature of the affinities of these fossils with Amphibia, Palzeo-
sauria, and Reptilia is a subject rather for inquiry than for dogma-
tism; and unfortunately the new materials now to be described are
too few. to do more than slightly modify accepted surmises.

No apology seems to be necessary for the restoration of Prof.
Owen’s name Labyrinthodon, because neither Prof. Huxley nor Mr.
Miall have attempted to justify either its suppression or reconstruc-
tion, while Iam inclined to urge that demonstration is wanting that
any foreign Labyrinthodont genus, such as Mastodonsaurus, has ever
occurred in the British Isles. The single dermal plate figured by
Mr. Miall, so similar to one hereinafter described from Sidmouth,
seems to me no evidence; for Labyrinthodonts like Crocodilia and
Teleosauria may well have resembled each other in the dermal ar-
mour of allied genera; and in describing an English fossil closely
allied to one of those originally described by Prof. Owen from War-
wickshire, it seems better to refer it to the genus Labyrinthodon as
constituted by Prof. Owen, there being no evidence adduced or ayail-
able for the subdivision of that genus in the manner proposed by
Mr. Miall.

To H. J. J. Lavis, Esq., F.G.S., we are indebted not only for the
discovery of the present specimen, unexampled among Labyrin-
thodont remains for its beautiful preservation, but also for untiring
labour and skill in extricating the fossil from the matrix, and re-
uniting its multitudinous fragments into the handsome and instruc-
tive remains of a right lower jaw exhibited.

The fossil is 13 inches long, and absolutely free from matrix, so
that every aspect and its component structures can be studied. It
demonstrates that the lower jaw does not consist of articular, angu-
lar, and dentary elements only, as generally described, from which
it is evident that the jaw is not constructed according to the Batra-
chian type. In addition to those three bones, there are certainly
separate splenial and surangular elements, and not improbably a
separate coronoid bone also. ‘These five or six bones in each ramus
are arranged so as to enclose a deep narrow cavity at least 9 inches
long, in the hinder part of the jaw, extending forward from beneath

—

A LOWER JAW OF LABYRINTHODON, 279

the articulation (Pl. XTX. fig. 2). This cavity has two perforations.
There is one long narrow superior perforation, which looks upward and
extends for more than 4 inches in front of the articulation ; it is mar-
gined externally by the surangular bone, internally, presumably, by
the splenial. The internal lateral perforation is much smaller; its
place is from 7 to 8 inches in front of the hindmost extremity of the
jaw, and half an inch from its inferior border. Its lower border is
_made by the splenial bone (fig. 2 a), and its upper border probably
by the surangular (fig. 2d), unless a small coronoid were present.
The specimen does not show whether the perforations communicated
or were separate. The internal cavity, like the Jaw itself, is some-
what compressed from side to side behind on its inferior border, and
widens a little anteriorly ; it widens more from below upward. It
is straight and in the same plane.

All the bones unite by sagittate sutures, and scarcely show any
indications of squamous overlap. They are all more or less seen on
the external aspect of the jaw (fig. 1), except the splenial bone ; while
every bone is seen on the internal aspect (fig. 2), where all the bones
are free from sculpture. The internal aspect is largely occupied by
the splenial element (fig. 2 a), which is fractured at its anterior ex-
tremity, and along all its upper border behind the lateral perforation.
As preserved, it is 114 inches long, and was probably some inches
longer. It terminates behind by a suture inclined slightly backward,
but nearly vertical where it joins the articular bone (fig. 2 e) ; and it
rests upon the angular bone (fig. 2 ¢) below, which extends along this
aspect of the jaw for 7} inches. This suture with the angular bone
has the outline of a low hill, of which the summit is at 3? inches
from the hinder extremity, and the height about 1} inch. Hence
the splenial bone descends in position as it passes forward, and then
for 12 inch it crosses obliquely on the underside of the jaw, and so
appears on its exterior aspect for 43 inches, extending forward as a
very narrow strip, not a quarter of an inch deep (fig. 1 a), which
has the angular bone above it for 25 inches behind, and the dentary
bone above it in front. It is strongly marked on the flattened
underside of the jaw with a radiate ornament reaching through
half a circle. The ornament extends from a centre on the inner
margin of the bone, half an inch below an oblong vascular perfor-
ation on the inner aspect; around the centre are close-set pits; but
further away these are elongated into grooves. Every pit and groove
has a perforation or channel for one large blood-vessel.

Of the dentary bone only a small posterior part is preserved (fig. 1b).
Tt extends backward on the upper surface of the jaw to within 2
inches of the articulation, and 4 inches of the extremity of the heel,
terminating ina point. It rapidly widens and deepens. The surangu-
lar bone, over which it extends by squamous overlap, is internal to it
on the superior aspect. Externally it descends down the side of the
jaw by a long oblique suture, resting at first on the surangular bone
for 12 inch, then on the angular bone for 53 inches, and finally upon
the splenial bone. The dentary differs from the other external bones
in being smooth and free from ornament, except faint longitudinal

280 H. G. SEELEY ON A PORTION OF

striations. It is separated from the angular by a deep depression,
looking as though due to its extension behind the angular, but pro-
bably resulting from the elevation of the ornament on the angular
bone. The dentary bone makes an angular bend, so as to form a
superior surface for the teeth, which appear to look upward and out-
ward. Where the bone ends behind, the jaw extends for 27 inches
below it, and 7 inch above it, being 25 inches deep.

The angular bone (fig. 1 c)is the most striking part of the specimen. _
Tt forms the lower portion of the jaw on both aspects, and is an elon-
gated bone tapering towards both ends on both sides. Externally
it is 103 inches long, is received between the splenial and dentary
bones in front, and posteriorly becomes depressed and reduced in
depth from 13 inch at 5 inches from the end of the bone, to less than
half an inch under the articulation. Its basal border is straight to
within the posterior two inches, when it curves gently upwards, ter-
minating half an inch short of the surangular bone, which extends
along its superior 53 inches behind the point where the dentary bone
terminates. This bone is marked by a semicircle of radiated sculp-
ture, the centre of which is rather behind the middle of the bone on
its basal border. As on the splenial bone, the ornament consists of
an irregular honeycomb of pits near the centre, these becoming
elongated laterally into grooves which run towards the extremities.
of the bone, harmonizing with the sculpture on the splenial and sur-
angular elements. The grooves as they proceed become subdivided
by ridges, of which five may be counted above the suture with the
splenial bone at 8 inches from the end of the bone. The pits and
grooves are full of perforations for blood-vessels, which take the same
directions as the grooves. A broader groove, also vascular, shaped
like the tail of a cursive y reversed, ascends to join the y -shaped
groove on the surangular bone, which extends below the articulation
of the jaw. ‘These grooves only differ from the others in being
broader. Altogether there are about 60 pits or grooves on the an-
gular bone, mostly averaging 7% inch in diameter, though some are
punctures ;; inch in size; while the broad posterior groove is
half an inch in diameter, and the long anterior groove is 6 inches in
length.

On the interior aspect of the jaw the angular bone occupies a
smaller area. (fig. 2,c). It reaches as far back, but not so far for-
ward, as in front, being only about 77 inches long. Its greatest depth
is 13 inch at 3 inches from its posterior end. The hinder part of the
bone is concave from above downward, and has two conspicuous de-
scending foramina under the part where it is deepest.

The surangular element (figs. 1, 2, 3, d) has hitherto been named
the articular ; but it extends in front of the articular bone so as to hide
the whole of it except the surface seen in the articulation. The
branch of the surangular which extends between the angular and
dentary bones is about 52 inches long ; but the part which, extending
interior to the alveolar margin of the dentary bone, is imperfect an-
- teriorly, is 8} inches long from the heel of the bone. Superiorly the
bone bends over towards the inner side, so as to fort a sort of back-

A LOWER JAW OF LABYRINTHODON. 281

ward prolongation of the alveolar table, nearly aninch wide. From
this a sharp slightly elevated coronoid ridge [imperfectly preserved |
arises, which extends forward in the median line, originating at the
outer and anterior border of the articulation. Another ridge at right
angles to this, and somewhat thicker, is directed outward, so that a
concavity extends along the bone between them. This ridge extends
backward for 3 inches, inclined very slightly downward, and then
curves in a semicircle upward, to terminate just behind the articu-
lation. Below this ridge, under the articulation, is a shallow groove
less than 7 inch wide, and below this an elevated W-shaped
rudely sculptured mass of ornament, 1} inch long and 4 inch deep.
Below this is the broad v-shaped groove, with a few tear-like
ornaments. The hinder arm of the y extends upward behind the
articulation ; the front arm extends upward and forward; both arms
are connected by the narrow curved groove below the articulation.

Behind the v-shaped groove is a triangular piece of vascular
honeycomb-like ornament, 1? inch deep and 12 inch long at its
base, where it is underlapped by the angular bone. It extends the
jaw externally behind the articulation. Nothing is behind the
straight inclined posterior line of this sculpture except the heel of
the bone—a subquadrate process half an inch deep, developed back-
ward and inward, so as with the angular bone below to nearly
enclose the hindmost termination of the articular bone, which is
greatly attenuated. The surangular bone is reflected behind the
articular element, so that superiorly it almost meets the splenial
element. Posteriorly, below the articulation, the bone is deeply
impressed on the irregular smooth hinder surface.

The articular bone (figs. 1, 2, 3, ¢) forms a narrow strip on the inner
side of the jaw, oblique and curved backward, extending from the
articulation to below the heel of the surangular bone. The articular
surface is subrhomboid, extending further backward and outward on
the external margin, and further forward on the inner margin, which
slightly overhangs the bones beneath. It measures ? inch from
front to back, and 13 from within outward. Its greatest oblique
measurement is less than 2 inches. The hinder margin (formed by
the surangular bone) rises above the anterior margin. From back
to front the area is semicylindrically concave, and marked with two
broad slightly elevated ridges with a longitudinal groove between
them, while a second incomplete sulcus forms the internal margin to
the articulation.

Though the bones now described have a reptilian aspect, and
somewhat of a reptilian arrangement, they do not go far towards
suggesting the affinities of Lubyrinthodon. There seems no reason
to doubt that the central hollow space in the jaw received the
primitive cartilage round which the bones were ossified ; and the
persistence of this character is rather a link with the lower than
with the higher Vertebrata. The jaw differs from the Batrachian
mandible in possessing well-developed angular and surangular ele-
ments, by which this cartilage and its articular ossification are en-
closed. But some reptiles, such as Crocodiles and the marine

282 H. G. SEELEY ON A PORTION OF

Chelonia, present many analogies in the perforations, in the struc-
ture of the jaw, and in the sculpture of the bones. But in the
absence of more important parts of the skeleton, no conclusions can
be drawn from these resemblances.

The species indicated by this specimen is almost identical in size
with the fragment attributed by Mr. Miall to Labyrinthodon pachy-
gnathus, and, so far as I can judge from figures, resembles that fossil
closely both in form and ornamentation. But the depth and outline
of the postarticular part of the jaw, and the differences of sculpture
in the lateral subarticular ornament, supply characters for distine-
tion which, if the Warwickshire specimens are accurately figured,
must separate the Dorsetshire fossil as a distinct species: this I
propose to name Labyrinthodon Lavisi, in honour of its discoverer.

Associated Fragments.—Within a yard or two of the spot where
the jaw was found, Mr. Lavis picked up several other Labyrintho-
dont fragments, which he supposes, with good reason, to have been
portions of the same skeleton. These comprise two small pieces of
bones from the upper part of the skull, which are too imperfect to
be determined with certainty at present, though one forms a portion
probably of the orbit of the eye, and the other shows a portion of a
narrow channel of the kind which has been called mucous groove:
its radiated sculpture is finer than on the other specimens; and it
may belong to a young animal or a small species. Another com-
pressed curved bone about 24 inches long, 2 inch deep at one end,
narrowing along the curve, and less than 7 inch thick, I am disposed
to regard (if Labyrinthodont at all) as possibly a vomer. The concave
border of the specimen appears to have been palatal. Along it, and
on part of the adjacent side of the bone, extend three or four ill-
preserved rows of small conical teeth, which vary in size. These
present a suggestive resemblance to the rows of teeth on the spe-
cimen figured by Prof. Huxley, and named Hyperodapedon, espe-
cially as some of them are similarly worn down; and it is possible
that this fragment may pertain to that genus.

Two other specimens are portions of thoracic plates. A narrow
fragment, 24 inches long, probably belongs to the median thoracic
plate; while the principal fragment, 3 inches long and 12 inch
wide, may be the right thoracic plate. It is thin, and marked by
the usual radiated sculpture, but has the broad external border
compressed and smooth.

Other smaller but imperfect plates were found, which are marked
with shallow pits which do not become elongated into grooves.

Another portion of bone may be a segment from the upper part of
a strong rib.

A chevron bone (fig. 4) shows the articular surfaces on one side,
which prove it to have been articulated between two vertebree, as in
most of the Reptilia and Paleeosauria.

Portions of dense cylindrical or more compressed bones, having
the appearance of weathered teeth, were found; but the transverse
sections which Mr, Cuttell has endeavoured to prepare are too thick
and opaque to demonstrate their real nature.

A LOWER JAW OF LABYRINTHODON. 283

EXPLANATION OF PLATE XIX.

Fig. 1. External view of posterior part of right ramus of lower jaw.
2, Internal aspect of the same specimen.
3. Articular region, seen from above.
4, Chevron bone.

a, splenial bone; 5, dentary bone; ¢, angular bone; d, surangular bone ;
é, articular bone.

DiscussIon.

Mr. Ussuer remarked that the Trias of South Devon was too
variable in character to furnish reliable evidence as to the persistence
of individual beds as lithological or paleontological zones. The con-
glomerate bed of Otterton Point, which crops out below High-Peake
Hill, is cut out at Ladram Bay by a fault, but reappears towards
Otterton Point for a distance of a mile and a half. It varies very
much in character, and exhibits false bedding in parts. Organie
remains have been found in the Trias at Ruishton, in a marly gritty
conglomerate, which contains Labyrinthodon, Estheria, &e.—also near
Stoke St. Mary’s, where Lstheria minuta occurs,—in the Vale of
Taunton, in sandstone not contemporaneous with that on the coast,
and at North Curry, where Estherie and plant-remains are found in
arenaceous beds under marl.

Mr. Wurraxker stated that while Mr. Lavis’s Labyrinthodon was
obtained from near the top of the sandstone, his Wyperodapedon
came from near its base.

Prof. Ramsay remarked that all these bones were from the
Keuper, and not from the Bunter.

The Presipent (Prof. Duncan) said that Professor Seeley had very
properly treated this bone from his own point of view; but he would no
doubt admit that there were others from which it might be looked at.
It seemed to him that if this were the lower jaw of Labyrinthodon, it
was very distinct from any other specimen. He noticed that from Mr.
Miall’s description it would seem that he had got upon the trace of the
surangular bone in Diadetognathus, as in one of his specimens he
describes a ridge which may be the indication of a bone wanting in .
the Warwickshire specimen but present in this one. He thought
that some fish exhibited a somewhat similar construction, but that
the lower jaw alone could never be sufficient for the purpose of
classification ; and here the other and more essential parts of the
skull were wanting. The Labyrinthodonts seemed to him to be
unquestionably an order of Amphibia.

Mr. Lavis, in reply, said that his estimate of the upthrow was
made from the lithological characters of the beds. The zone indi-
cated by him was not conspicuous by its lithological characters. It
contained small grains of quartzite and pockets of marl.

Prof. Sretzy, after thanking the President for his remarks, stated

Q.J.G.8. No. 127. x

(284 H. G. SEELEY ON A LOWER JAW OF LABYRINTEODON.

that the fossil described by him presented analogies to the Teleosauria
and marine Chelonia, a fact which supports the evidence derived
from. the skeleton of the Labyrinthodontia. The skulls in those
animals present characters found nowhere except among the true
Reptilia. With regard to the identification of the specimen, he said
that if those from Warwickshire described by Mr. Miall were Laby-
rinthodonts, then this was one also.

.

Quart. Journ. Geol. Soc Vol. XXXII. Pl. XIX.

‘S)

rf

i

Mimtern Bros. imp 3

Quart. Journ. Geol. Soe Vol. XXXI. Pl. XTX,

C L Griesbach

LABYRINTHODON LAVISI.

Peas |
A,

ae

aa a W

®
a) a
ies, , ;
i }

Ps om
tn ht a
ae BaD,
4
wt
5
j
é
'
5
.
v9
Peo: Vy
>: '
']
*
ad
i
yi
ba a
s iy
"7
‘ rere
i et an
¢ “Ff
Ne

tah! pe»

J. W. DAWSON ON THE PHOSPHATES OF CANADA, 285

33. Norm on the Puospnares of the Laurentian and CaMBRIAN
Rooxs of Canapa. By J. W. Dawson, LL.D., F.R.S., F.GS,
(Read March 22, 1876.)

Tnx extent and distribution of the deposits of apatite contained in
the Laurentian of Canada and in the succeeding Paleozoic forma-
tions, have not escaped the notice of our Geological Survey, and
have been referred to in some detail in Reports of Mr. Vennor,
Mr. Richardson, and others, as well as in the General Report
prepared by Sir W. E. Logan in 1863. Some attention has also
been given, more especially by Dr. Sterry Hunt, to the question of
the probable origin of these deposits*. My own attention has been
directed to the subject by its close connexion with the discussions
concerning Hozoon ; and I have therefore embraced such opportunities
as offered to visit the localities in which phosphates occur, and to
examine their relations and structure. I would now present some
facts and conclusions respecting these minerals, more especially in
their relation to the life of the Laurentian period, but which may
also be of interest to British geologists in connexion with the facts
recently published in the ‘Journal’ of this Society in relation to
the similar deposits found in the Cambrian and Silurian of Wales f.

In the Lower Silurian and Cambrian rocks of Canada phosphatic
deposits oceur in many localities, though apparently not of sufficient
extent to compete successfully for commercial purposes with the
rich Laurentian beds and veins of crystalline apatite.

In the Chazy formation, at Alumette Island, and also at Gren-
ville, Hawkesbury, and Lochiel, dark-coloured phosphatic nodules
abound. ‘They hold fragments of Lingule, which also occur in the
containing beds. They also contain grains of sand, and, when
heated, emit an ammoniacal odour. They are regarded by Sir W.
Logan and Dr. Hunt as coprolitic, and are said to consist of ‘‘ a paste
of comminuted fragments of Lingule, evidently the food of the
animals from which the coprolites were derived” t. It has also
been suggested that these animals may have been some of the larger
species of Trilobites. In the same formation, at some of the above
places, phosphatic matter is seen to fill the moulds of shells of
Pleurotomaria and Holopea.

In the Graptolite shales of the Quebec group, at Point Levis,
similar nodules occur; and they are found at Riviere Ouelle,
Kamouraska, and elsewhere on the Lower St. Lawrence, in lime-
stones and limestone conglomerates of the Lower Potsdam group,
which is probably only a little above the horizon of the Menevian
or Acadian series. In these beds there are also small phosphatic
tubes with thick walls, which have been compared to the supposed
worm-tubes of the genus Serpulites§.

* Geology of Canada, 1868; Chemical and Geological Essays, 1875,
t Davies & Hicks in Quart. Journ. Geol. Soc. August 1875.
{ Geology of Canada, p. 125.
§ Geology of Canada, p. 259; Richardson’s Report, 1869.
x2

286 J. W. DAWSON ON THE PHOSPHATES OF CANADA,

The Acadian or Menevian group, as developed near St. John,
New Brunswick, contains layers of calcareous sandstone blackened
with phosphatic matter, which can be seen, under the lens, to consist
entirely of shells of Lingule, often entire, and lying close together
in the plane of the deposit, of which in some thin layers they appear
to constitute the principal part*. Mr. Matthew informs me that
these layers belong to the upper part of the formation, and that the
layers crowded with Zingule are thin, none of them exceeding two
inches in thickness; but he thinks that the dark colour of some of
the associated sandstones and shales is due to comminuted Lingule.

At Kamouraska, where I have studied these deposits, the ordinary
phosphatic nodules are of a black colour, appearing brown with blue
spots when examined in thin slices with transmitted light. They
are of rounded forms, having a glazed but somewhat pitted surface—
and are very hard and compact, breaking with glistening surfaces.
They occur in thin bands of compact or brecciated limestone, which
are very sparingly fossiliferous, holding only a few shells of Hyolithes
and certain Scoltthus-like cylindrical markings. In some of these
beds siliceous pebbles occur with the nodules, rendering it possible
that the latter may have been derived from the disintegration of
older beds; but their forms show that they are not themselves
pebbles. Phosphatic nodules also occur sparingly in the thick beds
of limestone conglomerate which are characteristic of this formation ;
they are found both in the included fragments of limestone and in the
paste. The conglomerates contain large slabs and boulders of lime-
stone rich in Trilobites and Hyolithes; but in these I have not
observed phosphatic nodules.

In some of the limestones the phosphatic bodies present a very
different appearance, first noticed by Richardson at Riviere Ouelle,
and of which I have found numerous examples at Kamouraska. A
specimen now before me is a portion of a band of grey limestone,
about four inches in thickness, and imbedded in dark red or purple
shale. It is filled with irregular, black, thick-walled, cylindrical
tubes, and fragments of such tubes, along with phosphatic nodules
—the whole crushed together confusedly, and constituting half of
the mass of the rock. The tubes are of various diameters, from a
quarter of an inch downward; and the colour and texture of their
walls are similar to those of the ordinary phosphatic nodules.

Under the microscope the nodules and the walls of the tubes show
no organic structure or lamination, but appear to consist ‘of a finely
granular paste holding a few grains of sand, a few small fragments
of shells without apparent structure, and some small spicular bodies
or minute sete. The general colour by transmitted light 1s brown ;
but irregular spots show a bright blue colour, due probably to the
presence of phosphate of iron (vivianite). The enclosing limestone
and the filling of the tubes present a coarser texture, and appear
made up of fragments of limestone and broken shells, with some
dark-coloured fibres, probably portions of Zoophytes. Scattered

* Bailey and Matthew, “ Geology of New Brunswick,” Geol. Survey Reports.

J. W. DAWSON ON THE PHOSPHATES OF CANADA. 287

through the matrix there are also small fragments, invisible to the
naked eye, of brown and blue phosphatic matter.

One of the nodules from Alumette gave to Dr. Hunt 36°38 of
calcic phosphate; one from Hawkesbury 44:70; another from
Riviere Ouelle 40°34; and a tube from the same place 67:°53*. A
specimen from Kamouraska, analyzed by Dr. Harrington, gave
55°68 per cent. One of the richest pieces of the linguliferous sand-
stone from St. John yielded to the same chemist 30°82 of calcic phos-
phate and 32-44 of insoluble siliceous sand, the remainder being
chiefly carbonate of lime.

Various opinions may be entertained as to the origin of these phos-
phatic bodies ; but the weight of evidence inclines to the view origin-
ally put forward by Dr. Hunt7, that the nodules are coprolitic ; and
I would extend this conclusion with some little modification to the
tubes as well. The forms, both of the tubes and nodules, and the
nature of the matrix, seem to exclude the idea that they are simply
concretionary, though they may in some cases have been modified
by concretionary action. There are in the same beds little piles of
worm-castings of much smaller diameter than the tubes, and less
phosphatic; and there are also Scolithus-like burrows penetrating
some of the limestones, and lined with thin coatings of phosphatic
matter similar to that of the tubes. Further, the association of
similar nodules in the Chazy limestone with comminuted Lingule,
as already stated, is a strongly confirmatory fact.

The tubes are of unusual form when regarded as coprolitic; but
they may have been moulded on the sides of the burrows of marine
worms ; or these creatures may have constructed their tubes of this
material, either consisting of their own excreta or of that of other
animals lying on the sea-bottom. In any case, the food of the
animals producing such excreta must have been very rich in solid
phosphates, and these animals must have abounded on the sea-
bottoms on which the remains have accumulated. It is also
evident that such phosphatic dejections might either retain their
original forms, or be aggregated into nodular masses, or shaped
into tubes or burrows of Annelids, or, if accumulated in mass,
might form more or less continuous beds.

The food of the animals producing such coprolites can scarcely
have been vegetable; for though marine plants collect and contain
phosphates, the quantity in these is very minute, and usually not
more than that required by the animals feeding on them.

We must therefore look to the animal kingdom for such highly
phosphatic food. Here we find that a large proportion of the
animals inhabiting the primordial seas employed calcic phosphate in
the construction of their hard parts. Dr. Hunt has shown that the
shells of Lingula and some of its allies are composed of calcic
phosphate; and he has found the same to be the case with certain
Pteropods, as Conularia, and with the supposed worm-tubes called
Serpulites, which, however, are very different in structure from the
tubes above referred to.

* Geology of Canada, p. 461. Tt Ibid.

288 J. W. DAWSON ON THE PHOSPHATES OF CANADA,

It has long been known that the crusts. of modern Crustaceans:
contain a notable percentage of calcic phosphate; and Hicks and
Hudleston have shown that this is the case also with the Cambrian
Trilobites. Dr. Harrington has kindly verified this for me by.
analyzing a specimen of highly trilobitic limestone from the Lower
Potsdam formation at St. Simon, in which the crusts of these
animals are so well preserved that they show their minutely tubu-
lated structure in great perfection under the microscope. He finds
the percentage of calcic phosphate due to these crusts to be 1°49
per cent. of the whole mass. It is to be observed, however, that
the crusts of Trilobites must have consisted very largely of chitinous
matter, which, in some cases, still exists in them in a carbonized
state. A crust of the modern Zimulus, or King Crab, which I had
supposed might resemble in this respect that of the Trilobites, was
analyzed also by Dr. Harrington. It belonged to a half-grown
individual, measuring 5-25 inches across, and was found to contain
only 1845 per cent. of ashes, and of this only 1:51 per cent. of
calcic phosphate. The crusts of some Trilobites may have contained
as large a proportion of organic matter; but they would seem to
have been richer in phosphates. Next to Lingule and Trilo-
bites, the most abundant fossils in the formations containing the
phosphatic nodules are the shells of the genus Myolithes, of which
several species have been described by Mr. Billings*. Dr. Har-
rington has ascertained that these shells also contain calcic phosphate
in considerable proportion. The proportion of this substance in a
shell not quite freed from matrix was 2:09 per cent. These shells
have usually been regarded as Pteropods; but I find that the
Canadian primordial species show a structure very different from
that of this group. They are much thicker than the shells of proper
Pteropods; and the outer layer of shell is perforated with round
pores, which in one species are arranged in vertical rows. The
inner layer, which is usually very thin, is imperforate. In one
species (I believe, the H. americanus of Billings) the perforations
resemble in size and appearance those in the shells of Terebratule.
In another species (H. micans probably) they are very fine and close
together, as in some shells of tubicolous worms. I am therefore
disposed to regard the claim of these shells to the rank of Pteropods
as very doubtful. ‘They may be tubicolous worms, or even some
peculiar and abnormal type of Brachiopod. In connexion with
this last view, it may be remarked that the operculum of some of
the species much resembles a valye of a Brachiopod, and that the
conical tube is in some of them not a much greater exaggera-
tion of the ventral valve of one of these shells than the peculiar
Calceola of the Upper Silurian and Devonian, which has been
regarded by some paleeontologists as a true Brachiopod. I have not,
however, had any opportunity of comparing the intimate structure
of Calceola with that of these shells. Shells of Hyolithes occur in
the Lower Potsdam in the same beds with the phosphatic nodules ;
and in one of these Mr. Weston has found a series of conical shells

* Canadian Naturalist, Dec. 1871.

J. W. DAWSON ON THE PHOSPHATES OF CANADA. 289

of Hyolithes pressed one within another, as if they had passed in an
entire state through the intestine of the animal which produced the
coprolite.

Inasmuch, then, as some of the most common invertebrates of
the Cambrian seas secreted phosphatic shells, it is not more incredible
that carnivorous animals feeding on them should produce phosphatic
coprolites than that this should occur in the case of more modern
animals feeding on fishes and other vertebrates.

We may now turn to the question as to the source of the abun-
dant apatite of the Laurentian rocks. Were this diffused uniformly
through the beds of this great system, or collected merely in fissure
or segregation veins, it might be regarded as having no connexion
with other than merely mineral causes of deposit. It appears,
however, from the careful stratigraphical explorations of the Cana-
dian Survey, in the districts of Burgess and Elmsley, which are
especially rich in apatite, that the mineral occurs largely in beds
interstratified with the other members of the series, though deposits
of the nature of veins likewise occur. It also appears that the
principal beds are confined to certain horizons in the upper part of
the Lower Laurentian, above the limestones containing Lozoon,
though some less important deposits occur in lower positions*.

The principal apatite-bearing band of the Laurentian consists of
beds of gneiss, limestone, and pyroxene-rock, and has a thickness of
from 2600 to 3600 feet. It has been traced over a great extent of
country west of the Ottawa river, and has also been recognized on
the east side of that river as well. The mineral often forms compact
beds with little foreign intermixture ; and these sometimes attain a
thickness of several feet, though it has been observed that their
thickness is variable in tracing them along their outcrops. Several
beds often lie near to each other in the same member of the series.
Thin layers of apatite also occur in the lines of bedding of the
pyroxene-rock. In other cases disseminated crystals are found
throughout thick beds of limestone, sometimes, according to Dr.
Hunt, amounting to two or three per cent. of the whole mass.
Disseminated crystals also occur in some of the beds of magnetite, a
mode of occurrence which, according to Dr. Hunt, has also been
observed in Sweden and in New York in the Laurentian magnetites
of those regions.

The veins of apatite fill narrow and usually irregular fissures ;
and the mineral is associated in these veins with calcite and with
large crystals of mica. In one instance, at Ticonderoga, in New
York, the apatite, instead of its usual crystalline condition, assumes
the form of radiating and botryoidal masses, constituting the Kupyr-
chroite of Emmons. Since these veins are found principally in the
same members of the series in which the beds occur, it is a fair
inference that the former are a secondary formation, dependent on
the original deposition of apatite in the latter, which must belong
to the time when the gneisses and limestones were laid down as
sediments and ongane accumulations.

* Vennor’s Reports, 1872-73 & 1875-74.

290 J. W. DAWSON ON THE PHOSPHATES OF CANADA.

In all the localities in which I have been able to examine the
Laurentian apatite, it presents a perfectly crystalline texture, while
the containing strata are highly metamorphosed; and this appears
to be its general condition wherever it has been examined. Nume-
rous slices of the more compact apatite of the beds have been
prepared by Mr. Weston, of the Geological Survey; but, as might
be expected, they show no trace of organic structure. All direct
evidence for the organic origin of this substance is therefore still
wanting. There are, however, certain considerations, based on its
mode of occurrence, which may be considered to afford some indirect
testimony.

If, with Hunt, we regard the iron ores of the Laurentian as
organic in origin, the apatite which occurs in them may reasonably
be supposed to be of the same character with the phosphatic matter
which contaminates the fossiliferous iron ores of the Silurian and
Devonian, and which is manifestly derived from the included
organic remains.

If we consider the evidence of Hozoon sufficient to establish the
organic origin, in part at least, of the Laurentian limestones, we
may suppose the disseminated crystals of apatite to represent copro-
litic masses or the débris of phosphatic shells and crusts, the structure
of which may have been obliterated by concretionary action and
metamorphism.

Such Silurian beds of compact and concretionary apatite (without
structure, yet manifestly of organic origin) as that described by
Mr. Davies in the ‘Journal’ of this Society, may be taken as fair
representatives of the bedded apatite of the Laurentian. Further,
the presence of graphite in association with the apatite in both cases
may not be an accidental circumstance, but may depend in both on
the association of carbonaceous organisms, whether vegetable or
animal,

Again, the linguliferous sandstone of the Acadian group is a
material which, by metamorphism, might readily afford a pyroxenite
with layers of apatite like those which occur in the Laurentian.

The probability of the animal origin of the Laurentian apatite is
perhaps further strengthened by the prevalence of animals with
phosphatic crusts and skeletons in the Primordial age, giving a
presumption that in the still earlier Laurentian a similar preference
for phosphatic matter may have existed, and, perhaps, may have
extended to still lower forms of life, just as the appropriation in
more modern times of phosphate of lime by the higher animals for
their bones seems to have been accompanied by a diminution of its
use In animals of lower grade.

The Laurentian apatite pretty constantly contains a small per-
centage of calcium fluoride ; and this salt also occurs in bones, more
especially in certain fossil bones. This may in both cases be a
chemical accident ; but it supplies an additional coincidence.

In the lowest portions of the Lower Laurentian no organic remains
have yet been detected ; and these beds are also poor in phosphates.
The horizon of special prevalence of Hozcon is the Grenville band of

J. W. DAWSON ON THE PHOSPHATES OF CANADA. 291

limestone, which, according to Sir William Logan’s sections, is about
11,500 feet above the fundamental gneiss. It appears, from recent
observations of Mr. Vennor and Mr. W. T. Morris, that the bed
holding the Burgess Zozoon is on the same horizon with the lime-
stone of Grenville. The phosphates are most abundant in the beds
overlying this band. This gives a further presumption that the
collection and separation of the apatite is due to some organic
agency, and may indicate that animals having phosphatic skeletons
first became abundant after the sea-bottom had been largely occupied
by Hozoon.

I would not attach too great value to the above considerations ;
but, taken together, and in connexion with the occurrence of apatite
in the Cambrian and Silurian, they seem to afford at least a
probability that the separation of the Laurentian phosphate from
the sea-water, and its accumulation in particular beds, may have
been due to the agency of marine life. Positive proof of this can
be obtained only by the recognition of organic form and structure ;
and for this we can scarcely hope, unless we should be so fortunate
as to find some portion of the Lower Laurentian series in a less
altered condition than thatin which it occurs in the apatite districts
of Canada. Should such structures be found, however, it is not
improbable that they may belong to forms of life almost as much
lower than the Lingule and Trilobites of the Cambrian as these are
inferior to the fishes and reptiles of the Mesozoic.

Discussion.

Mr. Hicks said that the author had traced the existence of these
phosphatic nodules considerably lower in America than had been
done in England. He was inclined to accept the view of their
organic origin, seeing that in England certainly the abundance of
phosphates depends on that of organic life. The phosphate was due
chiefly to the shell and to the decomposition of the substance of the
body of the animals.

Mr. Krrpine remarked that the presence of graphite in these rocks
had been ascribed to plants; if so, the deposit was formed not far
from shore, and consequently we should not get that freedom from
sediment which is necessary for the production of an‘extensive de-
posit of animal origin. It seemed to him that attributing to these
deposits an animal origin was like going round to the back door when
the front door was open; for there was plenty of apatite in the
igneous rocks. He believed that even in later ages, when life was
more abundant, no workable deposits of coprolite nodules had been
formed where now found, but they had been sifted out by the action
of water from older deposits; for example, those of Cambridge from
the Gault, and those of the Red Crag from the London Clay.

292 _ ° J. W. JUDD ON THE ANCIENT VOLCANO OF -

34. On the Ancient Votcano of the Disrricz of ScHEMNITZ,
Huneary. By Joun W. Jupp, Esq., F.G.S. (Read April 26, 1876.)

[Puate XX.]
INTRODUCTORY.

Wuen, two years ago, the conclusions at which I had arrived con-
cerning the relations of the so-calied plutonic rocks to those of
undoubted volcanic origin in the Scottish Highlands were laid before
this Society, it was suggested by the late Sir Charles Lyell that pro-
bably no series of igneous rocks would be found to offer better
illustrations of the phenomena which I then described, or would
furnish more valuable opportunities for testing the truth of the
general conclusions to which I had been led, than the old volcanic
masses encircled by the mountain-ranges of the Carpathians.
From Mr. Poulett Scrope, ever ready with generous aid to those
who devoted themselves to the studies which he had himself so
successfully prosecuted in earlier days, a suggestion from his life-
long friend was always sure to meet with a ready response ; and
thus I found myself afforded an opportunity of continuing the ex-
amination of these interesting questions, by a careful study of the
voleanic rocks of Hungary and Transylvania. From both the
friends I was privileged to receive much valuable advice in pre-
paring for the work I proposed to undertake; and when the one
was taken from our midst, the other seemed to redouble his efforts
to counsel and aid me in its execution. In laying before this So-
ciety the first portion of the results of an investigation, in the con-
ception and conduct of which these fathers of our science so warmly
interested themselves, it is impossible to avoid giving expression to
those feelings of sadness which the reflection that neither of them
has survived to witness the completion of my task so powerfully
awakens within me.

There is probably no country in the world where the geologist
finds more valuable aid in carrying on his researches than the
Austrian Empire. The Imperial Geological Institute of Vienna has,
under the successive direction of Haidinger and Von Hauer, issued
geologically coloured copies of the whole of the 165 sheets of the
military map of the empire; and these have been accompanied by
most valuable memoirs on the different districts, published in the
well-known ‘Jahrbuch’ of the institute. Franz von Hauer has,
further, completed a reduction of these large-scale maps to a gene-
ral map consisting of twelve sheets, with a memoir descriptive of
eich, and has, finally, in his most valuable and useful work ‘ Die
Geologie und ihre Anwendung auf die Kenntniss der Bodenbeschaf-
fenheit der 6sterr.-ungar. Monarchie,’ which is accompanied by a
single-sheet map of the whole country, summarized in a most able
manner the entire mass of information hitherto obtained con-
cerning the geology of the empire. Of the great importance of

THE DISTRICT OF SCHEMNITZ, HUNGARY. 293

the paleontological, petrographical, and chemical researches which
have been carried on in the Vienna Institute, in illustration of the
work done in the field, it is quite unnecessary to speak in this
place ; they are far too well known to all geologists to need reca-
pitulation.

Nor do the results already enumerated by any means exhaust the
mass of materials which the geologist finds ready to aid him in his
researches withinthe Austrian dominions. The Geological Institutes
of Vienna and Buda-Pest still continue to carry their valuable
explorations of the geology of the country into still greater detail ;
and numerous private associations of scientific workers in every
state of the Empire, or connected with its several universities, con-
stantly make the most valuable new contributions to our science.

I have to thankfully acknowledge the very great liberality and
courtesy with which the various sources of information were thrown
open to aid me in the task I had undertaken, In Vienna the
director of the Reichsanstalt and his colleagues afforded me every
facility in my inquiries, while, in the University, Prof. Suess and Dr,
Neumayr, and, in the Hof-Mineralien-Cabinet, Prof. Tschermak and
Dr. Brezina rendered me equally kind assistance. In Buda-Pest,
Prof. Szabé, at the University, and Dr. von Hantken, as director of
of the Hungarian Survey, spared no pains in affording me all
needful help; while in Schemnitz I had the most ready aid of the
veteran geologist, Von Pettko, of Directors Poeschl, and Pech, and
of Prof. Winkler of the Berg-Akademie.

In spite, however, of all the important assistance afforded by the
labours of preceding investigators, the student of the Hungarian
voleanic rocks finds himself placed in one respect at a great disad-
vantage, in comparison with the opportunities which he enjoys in
examining the similar rocks in Scotland. ‘The absence of the mag-
nificent sea-cliffs with their extensive and wonderfully clear sections
(which, as I have shown, afford us such invaluable aid in determining
the true relations of the various rock-masses in the latter country)
constitutes a great drawback to the successful investigation of the
structure of the Hungarian voleanos; and a further difficulty is
created by the circumstance that very large portions of the country
are covered by almost impenetrable forests. The analogies which
are presented by a district of which the internal structure is so
admirably exposed by denuding agencies as is that of the Western
Isles of Scotland, may therefore be not unreasonably supposed capa-
ble of throwing new light upon the more intricate and difficult,
because less clearly exhibited, relations of the volcanic, plutonic,
metamorphic and sedimentary rock-masses of Hungary and Tran-
sylvania.

There is no portion of the Transleithan provinces of the Austrian
Empire in which the characters and relative positions of the different
volcanic rocks can be so conveniently studied as in the district
which surrounds the famous old mining-towns of Schemnitz,
Kremnitz, and Konigsberg. And perhaps few areas could be named
which, within so small a space, present rock-masses in more puzzling

294 J. W. JUDD ON THE ANCIENT VOLCANO OF

and apparently inexplicable relations, or concerning the true inter-
pretration of the structure of which greater diversities of opinion
have existed among geologists. If, therefore, I venture on this
occasion to offer a new explanation of the phenomena of a district
which has already been so patiently and accurately studied and so
often described by able geologists, it 1s not in any spirit of presump-
tion, but rather from the conviction that what is necessary to the
more perfect elucidation of the singularly involved questions con-
cerning the age and relations of the different geological formations
of the district is the means of comparison with another area in which
the rock-masses are more perfectly exhibited. And such a key to
the right understanding of the often perplexing and sometimes
seemingly contradictory phenomena presented by the Hungarian
rocks, I hope to show in the present memoir, is afforded by those
districts of Scotland which furnish us with such singularly favour-
able opportunities for geological research. And if, as I hope to
make manifest in the sequel, harmony and simplicity are by this
explanation introduced where all before was complication and con-
fusion, the circumstance may be fairly claimed as an additional
confirmation of the truth of the solution offered.

In treating this question it will be most convenient to present,
in the first place, a general sketch of the universally accepted facts
concerning the geological structure of the district in question, and,
in the second, to discuss the various interpretations which have been
offered with a view to explaining the origin of that structure.

Part I.—Sketch of the Geological Structure of the Schemnitz
District.

The volcanic rocks of the Schemnitz district cover an area
about 50 miles in diameter. It will not be necessary for our pre-
sent purpose to notice in detail the various sedimentary formations
of more ancient date, through which these Tertiary volcanic masses
have been erupted, and upon which they are seen at many points to
be superimposed ; suffice it to remark that the rocks subjacent to
those igneous formations with which we are more particularly en-
gaged in the present memoir include many old granitic and meta-
morphic masses, together with representatives of nearly the whole
series of fossiliferous strata, from the Trias to the Eocene Nummu-
litic inclusive—and that the whole succession of Secondary and older
Tertiary rocks in this district appears to have been continuously
deposited, and without any of those great breaks which are indi-
cated either by a sudden change in fauna or by wide-spread physi-
cal unconformity between them.

The outer portion of the great tract of voleanic rocks which we are
here considering, constitutes a broad girdle consisting of lavas with
their associated agglomerates and tuffs ; these, reduced by denuda-
tion to a number of more or less isolated mountain-masses, from
3000 to 4000 feet above the sea, and almost everywhere covered by
dense forests, overlook the central depressed area in which lie the

THE DISTRICT OF SCHEMNITZ, HUNGARY, 295

well-known mining-towns of Schemnitz, Kremnitz, and Konigsberg,
with numerous populous villages, and the health-resorts of Eisen-
bach and Glashtitte, which enjoy a considerable local celebrity.
Traversing this area in a diagonal direction from north-east to south-
west is the deep valley of the Gran, one of the most considerable
feeders of the Danube.

The lavas which everywhere compose the outer mountain girdle
of the Schemnitz district are all of the type intermediate between
the acid quartz-trachytes and the basic basalts. In all of them the
predominating felspar is plagioclase; and they therefore belong to
the class of igneous rocks now very generally called “ andesites,”
and not to the true or sanidine-trachytes, in which latter the most
abundant felspar is always orthoclase. With regard to the exact
species of felspar which constitutes the principal portion of the
widely spread Hungarian lavas, the researches made by Karl von
Hauer, in the laboratory of the Reichsanstalt of Vienna, and those
by Prof. Szabo, of the University of Buda-Pest, lead to closely con-
cordant results—the former being made by the ordinary analytical
methods, and the latter by the application of Bunsen’s-flame reactions,
in a manner devised by the Hungarian Professor himself for dis-
tinguishing the chemical composition of minute quantities of mine-
rals contained in rocks*. From the researches of both these in-
vestigators it appears that the principal ingredient of the Schem-
nitz lavas is a lime-soda felspar, sometimes approaching to the
labradorite and at others to the andesine type. These so-called
andesite rocks have been further classed by Tschermak and Doelter,
according to the additional constituents which distinguish them, into
augite-andesites, hornblende-andesites, and mica-andesites. The
resemblance of these modern andesite lavas to the old “ porphy-
rites’ of Devonian and Carboniferous age, in Scotland, is very
striking; in the volcano of Santorin, rocks of precisely similar
chemical and mineralogical character are being erupted at the
present day.

The most interesting fact with regard to the constitution of these
Hungarian lavas, which in the central parts of their masses are
often found to assume a very coarsely crystalline and almost granitic
character, while their outer portions present a strikingly scoriaceous
or slaggy appearance, remains to be noticed. It is that though the
predominant felspar in them is always of the basic type, yet they
not unfrequently contain free quartz, sometimes in very large pro-
portion. This free quartz is in some cases found to constitute large
irregular crystalline grains in the mass, just like those of the ordinary
orthoclase quartz-trachytes ; but at other times its presence can only
be detected by the microscope in thin sections. These quartziferous
andesites were by Stache, who first clearly pointed out their true
character, styled “ Dacites,” from the circumstance of their preva-
lence in Transylvania (the ancient ‘‘ Dacia”). They find their
exact analogues, among the more ancient volcanic rocks, in the

* Ueber cine neue Methode die Felspathe auch in Gesteinen zu bestimmen,
von Dr. Josef Szabé. (Hungarian edition, 1873; German edition in 1876.)

296 J. W. JUDD ON THE ANCIENT VOLCANO OF

‘* quartz-porphyrites.” The ‘dacites,” or quartz-andesites, are
not of frequent occurrence in the Schemnitz district, but are seen
at a few points, as the Hodritsch Thal near Schemnitz, the Kopa-
nitzer Thal near Dilln, and the Berg Szitnya near Sz. Antal; but
in some other parts of Hungary and throughout Transylvania they
have a very wide distribution. )

The agglomerates, tuffs, and ash-beds which are associated and
alternate with the sheets of lava are almost wholly made up of
fragments of volcanic rock, sometimes mingled with portions of the
subjacent sedimentary deposits. In some places the agglomerates
consist of irregular masses of all dimensions, including blocks of
prodigious size and weight, confusedly heaped together ; at others
they pass into the most perfectly stratified tuffs and ashes. Such
stratification in the ejected volcanic lapilli and sand must not, how-
ever, in the absence of marine or freshwater organic remains, be ac-
cepted as any proof of their having been accumulated beneath water ;
for, as we know from the example of many recent volcanos, ejected
materials in falling through the air, or while being swept down the
mountain-slopes in those torrents of mud so characteristic of great
volcanic outbursts, may assume the most perfectly stratified and even
jinely laminated condition. The masses of volcanic agglomerate, as
well as the rocks on which they rest, are frequently seen to be
traversed by numerous dykes of lava.

The more finely divided tuffs and ashes are not unfrequently
found to contain numerous remains of plants. These have been
most carefully studied, and their relations to the existing flora and
to those of earlier geological periods very clearly determined by
Unger, Von Ettingshausen, Kovats, Andrae, Heer, and Stur. From
the examination of these plant-remains we are enabled to infer the
conditions of climate, soil, and elevation which distinguished the old
land on which these volcanic eruptions took place. Besides the
plants, some remains of terrestrial mollusca, reptiles, and mammals
have also been found imbedded in these volcanic tuffs.

That these agglomerates, tuffs, and ashes were formed by
the ordinary explosive volcanic action, there is, I think, not the
smallest room to doubt ; and that the voleano which they helped
to build up was a subaerial one, there is also abundant, proof.
That, as is so common among volcanic cones, numerous ponds,
marshes, and even lakes of considerable size, were successively
formed and filled up by the ejected materials and sediments derived
from them, is also, I believe, rendered sufficiently clear from the
lacustrine deposits with remains of freshwater mollusca and fishes
which are so frequently found alternating with the agglomerates and
lavas. But it is also evident that the ancient volcanos of Hungary,
like so many existing ones, rose directly from the margin of the
ocean, or rather of a more or less isolated inland sea, with semi-
marine or brackish-water conditions. ‘That into some of the de-
pressions among the great volcanic masses of Schemnitz the sea
must, from time to time, have had access, is proved by the marine
or brackish-water shells found in the basins of Handlova and Rybnik,

THE DISTRICT OF SCHEMNITZ, HUNGARY. 297

both of which, however, are on the outer margin of the volcanic
area in question.

It would lead me too far from the main objects of the present
memoir were I to attempt any thing like an analysis of the splendid
flora associated with these tuffs, numbering as it at present does
many hundreds of species. Suffice it at present to notice that
in the opinion of Th. Fuchs, the greatest living authority on the
subject of the Tertiaries of Kastern Kurope, they indicate the exist-
ence at the period of these volcanic outbursts of a warm temperate
climate—and that the nearest living analogues of the plants of the
period are now found in Asia Minor, the Caucasus, the Himalaya,
Central and Northern Asia, and Japan.

The exact period of the grand outbursts of andesitic lavas and
agglomerates in Hungary, which gave rise to the formation of a
series of volcanos exceeding Htna in dimensions, the relics of one
of which we are now studying in detail, is fixed, by means of the
paleontological evidence we have above alluded to, as that of the
“Sarmatische Stufe,” a geological horizon which may be paralleled
with that of the beds which in Western Europe are classified as
the Miocene proper of Beyrich, or the Upper Miocene of Lyell.

The structure, then, of the girdle of volcanic rocks which sur-
rounds the Schemnitz mining-district is, as will have appeared from
the foregoing description, sufficiently simple and intelligible. We
see in it the ruins, greatly wasted by denuding causes, of an
enormous volcanic cone, built up both by eruptions from its centre
and also by innumerable outbursts upon its flanks.

But when we proceed to examine the central area which the
great girdle. of lavas and tuffs encircles, we discover aseries of rocks
presenting the most complicated and puzzling relations with one
another.

We may notice, in the first place (see Plate XX.), the masses of strati-
fied rocks, including dolomitic limestones, with sandstones and shales,
containing fossils which clearly show that they are of the age of the
Werfener Schichten (Lower Trias), and upon which is seen resting at
ene point, namely near Hisenbach, a small patch of Eocene lime-
stone crowded with Nwmmutlites. Very intimately associated with
these sedimentary rocks, of which the exact geological age is thus
rendered perfectly clear by their contained fossils, we find a series
vI untossiliferous and highly metamorphic rocks, consisting of erys-
talline limestone, quartzite and quartz-schiefer, various schistose
‘rocks, gneiss, and aplite or granulite. In equally close connexion
with these metamorphic rocks we find a number of clearly intru-
‘Sive igneous masses, consisting of greenstone or diorite, and of rocks
which have been almost universally classed, by the geologists who
have studied the district, as ‘‘ syenite and granite.”

By all the earlier writers on the geology of the Schemnitz area,
this central mass of metamorphic and igneous rocks was regarded
as beg of Primary age; but it has been gradually made clear by
later observations that, so far as the so-called “ greenstones” are
concerned, they are undoubtedly connected with the Tertiary ande-

298 J. W. JUDD ON THE ANCIENT VOLCANO OF

sites of the surrounding volcanic girdle, and indeed pass into the
latter by the most insensible gradations ; hence by the more modern
writers on the district they have been called “ greenstone-trachytes,”
while Richthofen suggested for them the name of “ propylites,” a
term which has not, however, met with any thing like general
acceptance among petrologists.

The “ greenstone-trachytes”’ of Hungary are rocks usually con-
sisting of a more or less felspathic base, in which large crystals of
plagioclase felspar with others of hornblende and mica are im-
bedded, so as to give them a more or less strikingly porphyritic
character. In places, however, the rock becomes granular, and
sometimes almost compact, in which cases it is dark-coloured and is
known locally as “ aphanitic greenstone.” Occasionally this rock
is found to contain spherical concretionary masses, and is then
known as “ Kugelgriinstein ;” sometimes it contains irregular
cavities, which are lined with various crystallized minerals. Occa-
sionally the greenstone-trachytes contain grains of free quartz,
and so pass into “dacites” or quartziferous-andesites of porphy-
roid or granitoid structure. Such rocks present the closest analo-
gies with the quartz-diorites of the older geological periods.

The greenstone-trachytes are clearly intrusive rocks, and are
found traversing the Triassic strata in dykes, and also enclosing in
their midst enormous masses of these latter deposits, which have
been subjected to intense metamorphism. Many portions of these
greenstone-trachytes have undergone great alteration, evidently
through having been traversed by acid vapours; and they now con-
tain, disseminated through their mass, large quantities of pyrites
and other metallic minerals, including gold and silver. The rock
occasionally assumes a brecciated appearance and would appear at
times to constitute a true ‘“ friction-breccia;” some of the masses
which have been most profoundly altered by the percolation through
them of acid gases have assumed a white and earthy appearance,
and have been called, though erroneously, ‘ greenstone-tuffs.”
The greenstone-trachytes are traversed by veins of quartz, often of
large size, which constitute the principal receptacles of those valu-
able metallic ores for which the district is so famous. These
metallic lodes traverse the so-called ‘“‘syenite and granite,” the
‘‘ oreenstone-trachyte,” and the various stratified rocks through
which these igneous masses have been extruded. The most pro-
ductive deposits of ore, however, appear to be found near the
junctions of the intrusive masses with the metamorphosed sedimen-
tary rocks.

The beautiful so-called syenitic rock of the Schemnitz area forms
four distinct bosses, which are surrounded by highly metamor-
phosed rocks, and are at the same time very intimately associated
with the greenstone-trachytes. The rock in question is a per-
fectly crystalline aggregate, either coarse or fine-grained, and occa-
sionally porphyritic in structure. So strikingly granitic is its aspect,
that all geologists who have studied the district have named the
rock either syenite or granite, according to the absence or presence

THE DISTRICT OF SCHEMNITZ, HUNGARY, 299

in it of free quartz. Microscopic study of this rock, however,
shows that its predominating constituent is always a plagioclase
felspar, to which hornblende and mica, and a variable quantity of
orthoclase is added. This rock ought therefore, it is clear, to be
classed with the diorites rather than with the syenites; but its
felspar is probably not oligoclase but labradorite. When, as is
frequently the case, a considerable quantity of free quartz is present
in it, the rock, though called a “granite,” should be rather referred
to the quartziferous diorites. The rock with which these granitic
masses of Hungary appear to have the closest points of resemblance
is the so-called granite of the Adamello group in the Alps, which
Vom Rath has distinguished under the name of Tonalite. Of the
exact relations of these bosses of granitic rock in the Schemnitz
district to the stratified, metamorphic, and volcanic rocks which
surround them we shall have to treat more fully in the sequel.

Besides those voleanic rocks of the Schemnitz area already de-
scribed, and which belong, as we have seen, to the class intermediate
in composition between the acid and basic types, there are evidences
of a number of eruptions of rocks of totally different composition,
almost all of which are of later date than the widely spread ande-
sitic lavas of Hungary. They belong either to the extremely acid
class (of liparites, perlites, obsidians, &c.) or to the extremely basic
type (basalts and augite-andesites).

Of certainly later date than the great masses of amphibole-ande-
sites, with their associated ‘ greenstone-trachytes” and dacites
(propylites), are those interesting rhyolitic rocks which rise in the
very centre of the district we are describing. They are espe-
cially well exposed at Konigsberg, near Vichnye (HKisenbach), in
the Hliniker Thal, and in the country north of the Gran, extending
from Heiligen Kreuz, nearly as far as Kremnitz (see the map and
section, Plate XX.). These rocks present, in their chemical com-
position, in their mineralogical constitution, and in their general
features, the most marked contrasts with the andesitic rocks which
cover so large a portion of the surrounding country. The Hungarian
rhyolites contain from 70 to 81 per cent. of silica; the proportion
of iron and of the alkaline earths is very small in them, while that
of the alkalies is very high, the quantity of the potash usually ex-
ceeding that of the soda. These rhyolites always contain free
quartz, sometimes in considerable quantity; the predominating
felspar in them is always orthoclase; and their other principal
ingredients are hornblende and biotite, which occur sometimes
separately, and at others in association with one another.

The features of greatest interest presented by these rhyolitic
rocks, however, are those connected with their structure. They ex-
hibit, as is so constantly the case with rocks of the most acid class,
a great tendency to assume the vitreous character; and hence arise
all those interesting modifications of rock-structure which were
first clearly described by Beudant in 1822, and which have made
the loeality of Hlinik so famous among petrologists.

In the Schemnitz district, however, the rhyolitic rocks are seldom

med. Goss NO. 127, x

300 J. W. JUDD ON THE ANCIENT VOLCANO OF

found assuming the perfectly glassy character and becoming true
obsidians, like those of Lipari, Iceland, Mexico, Ascension, &c.;
their lustre scarcely ever passes beyond the semivitreous or resi-
nous; and hence they may be rather classed with the pitchstones,
or, being frequently porphyritic in structure, with the “ pitchstone-
porphyries.” Throughout Hungary, indeed, true obsidians are very
rare; and although most interesting varieties of this rock of black,
red, and green colours are found in the Tokay-Hegyalja, some of
which have been well described by Vogelsang, yet, so far as my own
observations go, these never form large rock-masses as in Lipari, but
occur only as “ Saalbands” on the sides of intrusive dykes of rhyo-
litie or quartz-trachyte rock. |

Of the different varieties of the curious perltic variety of structure,
however, the Hungarian rhyolites, and especially those of the vicinity
of Schemnitz, afford the most interesting and instructive examples.
Indeed almost all the vitreous rocks of Hungary exhibit, under the
microscope, the presence of those peculiar and little-understood
spiral and concentric structures so characteristic of perlite; and in
addition to these we frequently find various modifications of true
spheerulitic concretions which are always quite independent of, and
evidently owe their formation to quite different causes from the
former.

Although the stony and slightly vitreous varieties of the Schem-
nitz rhyolites frequently assume a scoriaceous character, and thus
give rise to the formation of those beautiful ‘‘ millstone- porphyries ”
which are so extensively worked at Do Sminje, near Hlinik, and at
Himmelreich, near Konigsberg, yet pumice is rare. The so-called
pumice of the Hliniker Thal presents, indeed, a slight approximation
to the typical characters of this rock, and, under the microscope, is
seen to be full of gas-pores; yet the sponge-like character which in
true pumice is so manifest to the naked eye, is almost always want-
ing in the Hungarian rock, and I have never found specimens of the
latter that would float in water. This may in part be accounted
for, no doubt, by the circumstance that the rhyolitic rocks have un-
dergone a very considerable amount of denudation, by whieh the
lighter and more fragile materials would be easily swept away; but
I believe there is also evidence that a considerable amount of subsi-
dence took place before the extrusion of these rhyolitic lavas, and
that they were erupted under a pressure of water sufficient to pre-
vent the full distention of the liquefied mass by the escaping gases
and vapours.

Through the so-called hornstone modification, the semivitreous
and perlite rocks of Hungary graduate into the stony varieties.
These exhibit great diversities of character, being sometimes exceed-
ingly compact and fine-grained, and at others highly porphyritic, in
the latter case frequently exhibiting approximations towards the
granitic structure. The beautiful red rhyolite of the “ Stein- Meer,”
near Hisenbach, containing large but imperfect double pyramids and
rounded grains of quartz, is almost identical in appearance and
characters with many quartz-porphyries of the older geological .
periods.

THE DISTRICT OF SCHEMNITZ, HUNGARY. 301

The rhyolites of the Schemnitz district frequently assume those
remarkably banded and ribboned structures which are so admirably
exhibited by the rocks of similar composition in the Ponza Islands,
Lipari, Ascension, &c.; but, as I propose to discuss the nature and
origin of this remarkable structure in all these instances in a special
communication to this Society, I will not dwell longer on the subject
on the present occasion.

With regard to the relations of the rhyolitic to the andesitic rocks,
there is fortunately no room for the smallest doubt, inasmuch as the
beautiful natural sections of the Hliniker Thal, and the fine artificial
exposures of the rocks along the line of railway which has recently
been carried across the Carpathians by the pass of Berg, furnish the
geologist with abundant evidence upon the subject. In studying
these very instructive sections, I am much indebted to the kind aid
of Dr. Gustav Zechenter, of Kremnitz, who has constructed a detailed
geological map of the country through which the railway passes.
These clear sections seen in the railway-cuttings show that the rhy-
olites constitute great vertical intrusive masses, forced through the
midst of the older andesite lavas, and that the tuffs and agglomerates
belonging to the former class of rocks always overlie those of the
latter.

The tuffs and agglomerates, which are not unfrequently associated
with the rocks of the rhyolitic class, are in many localities found to
yield numerous plant-remains. The rhyolitic tuffs of Heiligen Kreuz
and of Jastraba, south of Kremnitz, are especially famous for the
fine floras they have afforded. These, and some animal remains
found with them, enable us to fix the period of the rhyolitic out-
bursts as coinciding with the latter part of the Miocene epoch; and
they also serve to show that the climatal conditions of the period
were considerably more temperate than during the great andesitic
eruptions, the flora being more nearly related to that existing in the
same area at the present day.

We have spoken of the rhyolitic as younger than the andesitic
eruptions of the Schemnitz area. There is proof, indeed, that the
great volcano built up by the earliest and grandest ejections, which
supplied such prodigions volumes of lavas and agglomerates, had -
undergone great denudation, and was, in fact, reduced almost to a
state of ruin before the occurrence of the second series of eruptions
in their midst. The former, indeed, sustain to the latter precisely
the same relations which in Auvergne are seen to subsist between
the skeleton volcanos of Mont Dore, the Cantal, and the Mezen, and
the more recent volcanic outbursts (forming the “ puys”) which
have taken place in the midst of their ruins.

Of still more recent date than the rhyolites of the Schemnitz dis-
trict are the sporadic outbursts of basaltic rocks, of which, for the
most part, there remain only the central lava masses, which filled”
the throats of these small volcanos (“‘puys”). Such centres of vol-
canic action, now marked only by intrusive plugs of basalt, are seen
at many points within and around the grand old volcano of the
Schemnitz district. At one locality, however, namely Ostra Hora,

y2

302 J. W. JUDD ON THE ANCIENT VOLCANO OF

near Heiligen Kreuz, the remains of a central cone and of a lava-
stream proceeding from it, the latter broken up by denudation into
miniature plateaux like those of the Auvergne, can be very distinctly
traced, as was long ago shown by Von Pettko*.

The basalts of the Schemnitz district exhibit some very interest-
ing diversities in character. In some instances, like that of the
well-known Calvarien Berg of Schemnitz, they contain very large
quantities of olivine, while in others, as, for example, the intrusive
mass of Gieshibl, that mineral is almost wholly wanting. The
latter variety constitutes a link between the true basalts and a rock
which is very extensively developed in Hungary, consisting of a
triclinic felspar, augite, and magnetite, with occasional crystals of
hornblende. These are the rocks now usually called augite-ande-
sites; but, according to Prof. Szabo, the felspathic ingredient in
them may be any variety of the lime-soda felspars, from andesine to
anorthite. In the volcanic district of the Matra, which les to the
south-east of that of the Schemnitz area, the augite-anorthite rocks
are very extensively developed ; and hence Prof. Szabé has proposed
for this particular variety of the augite-andesite group the name of
‘‘ matraite.”

With regard to the exact age of the basaltic rocks of the neigh-
bourhood of Schemnitz we have not such clear and unmistakable
evidence as in the cases of the andesitic and rhyolitic varieties.
But it is manifest that while these basaltic rocks are certainly
younger than the rhyolites, they must nevertheless be of consider-
able antiquity, seeing how greatly they have suffered from denuda-
tion. As we have in other parts of Hungary, as well as Transylvania
and Styria, good paleontological evidence that numerous outbursts
of rock of precisely similar character to the basalts of Schemnitz
took place during the formation of the Congerian and Inzendorf .
beds, we can scarcely be wrong in assigning the latter volcanic rocks
to the same age—that is, to the older portion of the Pliocene
period.

We have spoken of the outburst of the andesitic, rhyolitic, and
basaltic lavas of the Schemnitz area as taking place successively; it
must not, however, be assumed that the production of the different
varieties of volcanic materials at this igneous centre can be corre-
lated with sharply defined and distinctly separated geological periods.
On the contrary, it is clear that although the great mass of the an-
desitic lavas was erupted before, and most of the basalts subse-
quently to the rhyolites, yet the three periods of eruption of dif-
ferent classes of rock to some extent overlapped one another. In-
deed at Zapolenka, near Hodritsch, we find evidence, as Yon Pettko
has so well shown?y, that a cone and lava-stream of comparatively
modern date exist; for its features can be distinctly traced. The
lava-stream, it is clear, must have dammed up the river-course, and

* “Geognostische Skizze der Gegend von Kremnitz, von Johann von Pettko,”
Haidinger’s ‘ Naturwissenschaftliche Abhandlungen,’ Bd. i. (Wien, 1847) p. 302.

t “Ueber den erloschenen Vulcan Zapolenka in der Nahe von Schemnitz,
von Herrn Johann von Pettko,” Haidinger, Bericht vi. (1850) p. 168.

THE DISTRICT OF SCHEMNITZ, HUNGARY. 303

thus given rise to the formation of a rather extensive lake, which
has been partially filled up with sediments and then drained by the
cutting-through by the stream of the barrier of lava which retained
it. This comparatively modern stream of lava, however, is of ande-
sitic composition; and the cone is formed by scorie of the same
rock.

The following table exhibits the average composition of the most
characteristic varieties of voleanic rock which have been erupted in
Hungary. These averages have been calculated from a considerable
number of published analyses.

III. Horn-
blende and _ IV, Augite-
Mica-Ande- Andesites.

I, Rhyolites II. Dacites
or Quartz- or Quartz-
Trachytes. Andesites,

sites,
Specific gravity .. 2°375 2°629 2-643 2-715
EE ee 73-49 65°03 58:49 56°21
PAID) Fs vc. a’ n« 13-63 14-80 20°45 17:26
peeps won and } 2:27 6-20 765 11-56
anganese '
a ee 1°24 3°61 5°58 6°79
Magnesia ..:..... 0-32 1:01 i-72 1-49
eM os in vied ia» 7°23 5°58 6:19 4°43

The quantities of soda and potash are not given separately in these
averages; for in the analyses of Von Sommaruga, on which they are
based, the separation of the alkalies was effected by a method which,
as has been since shown by Tschermak and Doelter, does not yield
reliable results.

The voleanic formations of Hungary offer some peculiarities of
great interest, especially in the remarkable abundance of rocks
mainly composed of plagioclase felspar, with or without free quartz.
To the British geologist they are of especial interest, on account of
their numerous and striking points of analogy with the newer Pa-
lzozoic volcanic rocks of Scotland. Indeed we may safely assert
that the Paleeozoic hornblende-, mica-, and augite-porphyrites of our
own island present the exact counterparts of the Tertiary hornblende-,
mica-, and augite-andesites of Hungary; and, excepting in the condi-
tion of their preservation, I know of no grounds whatever for placing
them in different categories, and assigning to them distinct names.
Rocks of precisely similar composition to those of Hungary are
abundantly developed in various parts of the Turkish Empire; and
in the island of Santorin, as we have already remarked, they are
being erupted at the present day.

No contrast, however, can possibly be greater than that afforded
by the volcanic rocks of Hungary on the one hand, and those contem-
poraneously formed in Bohemia and Central Germany on the other.
While, in the former, plagioclase felspars almost always predominate,
even when the rocks are highly quartziferous, in the latter ortho-
clase is everywhere the most abundant constituent, occurring in
rocks eyen of the most basic type. On the north-west side of the

304 J. W. JUDD ON THE ANCIENT VOLCANO OF

Carpathian chain, leucite, sodalite, nepheline, nosean, and hatiyne are
most abundant rock-forming constituents; while on the south-east
of that mountain-range these minerals appear to be wholly unknown.
A glance at the ultimate analyses of series of the Tertiary volcanic
rocks of Hungary and Bohemia respectively will show that while
potash is an abundant constituent in the latter, it is present only in
very small proportion in the former.

In no way, perhaps, will it be possible to make this contrast in
composition between the volcanic ejections of the two areas more
striking than by placing side by side the analyses of rocks from either
of them which contain nearly the same percentage of silica. LI. is
the average of two analyses of Bohemian phonolites by Rammelsberg ;
II. the average of three analyses of Hungarian augite-andesites by
Doelter.

I. .
eimai <:. 0c Oheddw. heey ee 57:05 58°75
Piging iba MP aos 21:57 17:02
Oxides of Iron and Manganese 2°91 8:05

" Lime and Magnesia.......... 2-16 9°43
pl We te. a ah Os eer 5°89 1:26
LS Rp eee. hey Gedicenrgee 4 ta teas 8°69 3°50

We shall not here stay to discuss the interesting problem which
is suggested by these curious differences. Suffice it for the present
to notice the fact that while on the one side of the Carpathian chain
nepheline-, leucite-, and hatiyne-basalts with various phonolites were
building up the vast volcanos of Bohemia and Central Germany,
on the other side of that mountain-range andesitic lavas with some
rhyolites and felspar-basalts were being erupted on an equally
gigantic scale. In the two districts the volcanic activity began,
attained its climax of intensity, and fell into decline in correspon-
ding geological periods ; yet the products of the volcanic action in
the two districts are as different from one another as it is possible
to conceive.

Since the extinction of the volcano of Schemnitz the igneous
forces have given rise to a number of minorsymptoms. Hot springs
(probably forming geysers like those of Iceland and the Rocky-
Mountains) have deposited enormous quantities of siliceous sinter
(freshwater quartz), which is found to yield at many points the
beautifully preserved remains of a flora not very different from that
existing at the present day. The localities of Ilia, Hlinik, and
Lutilla are particularly famous for presenting us with the relics of
this flora. Other springs, like those so abundant in Auvergne and
Central Italy, were highly carbonated, and have deposited enormous
quantities of calcareous tufa.

Several hot and mineral springs still rise in the midst of the old
voleanic district of Schemnitz. Two of these, namely those of
Vichnye (Hisenbach), which has a temperature of 101° Fahr., and
Skleno (Glashiitte), with a temperature of 120° Fahr., have given
origin to locally famous bathing-places. At both points a consi-

THE DISTRICT OF SCHEMNITZ, HUNGARY. 305

derable quantity of tufa is deposited by the hot springs. We shall
show in the sequel how large a part must probably be assigned to
the action of mineral springs and fumaroles in filling those fissures
in the old volcanic rocks of the Schemnitz district which now con-
stitute its far-famed mineral veins. Some of these are, as Von
Pettko has shown, of later date than the eruption of the basalts.

That since the country has assumed its present form it has been
visited with severe earthquake-shocks, we have sufficient proof.
Near Eisenbach there occurs, on the slope of a mountain composed
of quartz-trachyte, a most remarkable accumulation of gigantic
blocks, which is known as the “Stein-Meer.” Although no re-
cord or tradition of the formation of this singular mass of débris
appears to be in existence, I think that no geologist will be in-
clined to doubt that its origin must be assigned to the secondary
effects of some great earthquake-shock, which has affected the dis-
trict and hurled down from the mountain-side the vast masses
already conveniently prepared for the catastrophe by a peculiar
system of jointing.

The district is still subject to earthquakes. In 1854 and 1855
three shocks were felt at Schemnitz within the space of nine months,
the attendant phenomena and effects of which were carefully noted
by M. Russegger*. Each shock is said to have been attended with
a loud noise, ike the report of a cannon; and the whole town was
violently shaken. The vibrations appear to have followed the line
of the principal mineral vein of the district, but not to have ex-
tended to any very great distance from it. The secondary effects of
the shocks seem to have gradually increased to greater depths, as
was proved by a subsequent careful examination of the workings of
the mines ; and the maximum action appears to have been experi-
enced at a depth of between 600 and 700 feet from the surface.

Part II.—On the Interpretation of the Geological Structure of the
Schemnitz District.

Haying now described the main features of this very interesting
district, we will proceed to discuss the question of their true inter-
pretation, and endeavour to define the exact relations and geological
age of the several rock-masses of the central area; for on these
points there have arisen, as we have before remarked, the widest
diversities of opinion among geologists.

Without attempting to notice the works of all the earlier writers
on Hungarian geology, it will be sufficient if in this place we make
our retrospect of geological opinion on the questions at issue com-
mence with the justly celebrated work of Beudant}, published in
1822. The distinguished French geologist, as is well known, de-

* «Pas Erdbeben in Schemnitz am 31. Januar. 1855, von J. Russeger,” Sitz-
ungsber. Wien. Akad. xv. p. 368. “Bericht ttber das am 30. September, 1855,
stattgefundene Erdbeben, von J. Russeger,” ibid. xvii. p. 479.

+ Voyage minéralogique et géologique en Hongrie pendant l’année 1818.
3 vols. Paris, 1822.

806 J. W. JUDD ON THE ANCIENT VOLCANO OF

voted especial attention to the Schemnitz area; and the conclusions
at which he arrived may be briefly stated as follows :—

Beudant seems to have regarded the -outer girdle of andesitic
rocks as being formed synchronously with the whole “ Secondary
series,” while the syenite and greenstone of the central area he
referred to the earlier ‘‘ Transition period,” and the metamorphic
rocks of the same tract to the Primary. The rhyolitic and basaltic
rocks he justly classed, in part at least, as Tertiary, although,
through his unfortunate but not unnatural mistake of confounding
the Pliocene Congeria-beds with the Eocene Calcaire grossier, he
threw the whole too far back in time. The one point which all the
earlier geological interpretations of the district appear to have had
in common was the separation of the gneiss, granite, syenite, and
diorite, which were classed as Primary and Transition rocks, from
the trachytic and basaltic, which were grouped with the Secondary
and Tertiary.

The more accurate study of the Tertiary faunas and floras, which
was initiated by the labours of Lyell and Deshayes in Western
Europe, and has been so admirably extended to the east of Hurope
by Hornes, Suess, Von Hauer, Fuchs, and others, soon, however,
led to the recognition of the fact that all the volcanic rocks of the
Schemnitz area belong to the Tertiary period, and even to the latter
half of it. Thus there was brought about a wide separation of the
supposed ancient crystalline “granites, syenites, and diorites”
from the evidently modern “ trachytes.”’

But, at the same time, petrologists began to recognize the circum-
stance that the so-called old greenstones or diorites of the district
not only presented the most remarkable points of resemblance to
the Tertiary trachytes or andesites by which they were surrounded
on all sides, but that in many instances the most perfect transition
of the one rock into the other could be clearly traced. As early as
1848 we find Von Pettko* (to whose labours, next to those of Beu-
dant, we are most indebted for the elucidation of the structure of
the complicated Schemnitz area) making the declaration that ‘“ there
is nothing to show that the greenstone of this district is older than
the associated trachyte, while the frequent transition of the two
rocks into each other, and their common local disposition, forming
together the great ring-shaped mountain-ridge, decidedly point to
their synchronous formation.”

As this conviction of the intimate connexion existing between the
Hungarian trachytes and the so-called diorites became more firmly
established in the minds of geologists, the latter rocks acquired the
name of “ greenstone-trachytes.” This view of the relations of the
two classes of rocks was in part adopted by Richthofen, when, in
18607, he made his general survey of the Hungarian volcanic rock-
masses ; and at a later date this author suggested, as we have already

* Berichte tber die Mittheilungen von Freunden der Naturwissenschaften in
Wien, Bad. iii.

t “Studien aus den ungarisch-siebenbiirgisthen Trachytgebirgen,” Jahrbuch
der k.-k. Reichsanstalt Wien, Bd. xi. S. 153.

THE DISTRICT OF SCHEMNITZ, HUNGARY. 307

seen, the name of ‘“ propylites” for the ‘ greenstone-trachytes.”’
Richthofen, however, in his later works, appears to have regarded
his ‘“propylites” as of slightly older date than the trachytes,
- though forming part of the same series of igneous outbursts during
the Tertiary period.

At the present day, so convinced are some of the most eminent
petrologists of the essential identity of these two classes of rocks,
in Hungary, the Tertiary greenstones and the trachytes, that they
make no distinction whatever in the terms which they apply to
them—Tschermak and Doelter, for example, calling them both
* hornblende-andesites,” and Szabo designating them equally as
“ trachytes.”

With regard to the stratified and metamorphic rocks, the discovery
of fossils in some of the less crystalline schists and limestones has
fully established their Triassic age. An isolated mass of limestone-
conglomerate, which occurs near Kisenbach, is crowded with Num-
mulites, and therefore clearly belongs to the Eocene. With respect
to the more altered and granitic rocks, namely the masses of gneiss,
mica-schist, quartzite, crystalline limestone, &c., with the intrusive
syenite and granite, it was suggested by Von Adrian in 1866* that
they might belong to the Devonian, an opinion that was adopted
with some hesitation by Von Lipold in 18677; in the more recent
publications of the Vienna Reichsanstaltt, the probability of their
being of Dyas or Permian age is entertained by Franz von Hauer.
But all these identifications rest upon the insecure foundation of
mere mineralogical resemblance, palzontological evidence being
entirely wanting.

Dr. Daubeny, who paid a somewhat hurried visit to Hungary in
1823, appears to have been greatly impressed by the complicated re-
lations of the granitic, dioritic, and trachytic masses of the Schemnitz
area ; and in 1848§ he made the remark that if the granitic rocks of
this district could be regarded as of equally modern date with the
trachytes, many difficulties in the interpretation of the geological
structure of the area would thereby be removed. Unfortunately,
however, he had no opportunity of revisiting the country, and of
testing how far his suggestion might be tenable.

It isin this important suggestion of Dr. Daubeny, as I propose to
show, that the true key to the correct understanding of the struc-
ture of the Schemnitz district is to be found.

Beudant is undoubtedly right in regarding the “granite and
syenite” of the central tract at Schemnitz as being most intimately
associated with the “greenstone,” and, indeed, as forming part of
the same mass. So strikingly is this the case, that in none of the
geological maps of the district that have been published is there

* “Das siidwestliche Ende des Schemnitzer-Kremnitzer Trachy}stockes,”
Jahrbuch der k.-k, Reichsanstalt Wien, Bd. xvi. p. 355-477.

t “Der Bergbau von Schemnitz in Ungarn,” ibid. Bd. xv. p. 317-458. ~

t ‘‘Geologische Uebersichtskarte der osterreichisch-ungarischen Monarchie,

Blatt iii. Westkarpathen,” ibid. Bd. xix. (1869) 8. 512.
§ A Description of Active and Extinct Volcanos, 2nd ed. (1848) pp. 129,
130.

308 J. W. JUDD ON THE ANCIENT VOLCANO OF

any thing like agreement as to the lines of separation between the
‘“‘syenite” and the “ greenstone”—rocks by one author assigned
to the former being by others referred to the latter. And it is only
necessary for the geologist to study the interesting series of sections
exposed in the Hisenbach and Hodritsch valleys, to be convinced
that the two classes of rocks pass into one another by the most in-
sensible gradations.

But, on the other hand, it is equally certain, as is, indeed, I
believe now universally admitted alike by petrologists and physical
geologists, that the so-called greenstones, greenstone-trachytes, and
trachytes are all merely varieties of the same rock, and constitute
indeed portions of the same great eruptive masses.

Now, that the so-called trachytes (or andesites) are true vol-
canic rocks, erupted during the Miocene period, we have, as already
shown, the most convincing proofs. Hence we are irresistibly led
to the conclusion that all these igneous rocks of the Schemnitz area,
‘‘ oranites and syenites,” ‘ greenstone-trachytes,” andesitic lavas
and tuffs, are parts of the same great eruptive masses, and are of
contemporaneous date.

The truth of this conclusion is borne out in a very striking manner
by a comparison both of the chemical and microscopical characters
of the three different classes of rocks in question. In the following
table I place side by side analyses of typical examples of the so-
called syenite, of the ‘‘ greenstone-trachyte” and of the andesitic
lava of the Schemnitz district.

iis ri. ITI.
SLE: ae ae Re ee ae 61:73 60°26 61:62
A LOMA, ve, (nse, Fo 17:45 18-25 20°66
Oxides of iron and
manganese .... 5:94 6°83 ~ 6°64
sure ae Sie os 4°52 3°08 4:27
Magnesia ...... 2-29 Td. 1°35
Ue T=) ae ay oe ae 7°00 5°61 4°55
L088 + ion calee oetees 1:16 3°40 2°40

I. is the so-called syenite of Hodritsch, analyzed by Karl von
Hauer.

II. the “greenstone-trachyte” of Gelnerowsky Wrch, analyzed
by Von Sommaruga.

III. the hornblende-andesite of Rybnik, also analyzed by Von
Sommaruga.

Allowing for the different degrees of decomposition, as shown by
the amount of volatile material in each case, the rocks are seen to
be almost identical in composition.

When we turn to the consideration of the microscopical characters
of the so-called “syenite and granite” of Hodritsch, we find that,
while offering the most marked points of contrast with all the un-
doubtedly ancient Hungarian granites, it presents some very remark-
able features of agreement with the “ greenstone-trachytes” and

THE DISTRICT OF SCHEMNITZ, HUNGARY. 309

“ hornblende-andesites ” of the same area, with which they are so
intimately associated.

In all these rocks it is interesting to find that the prevailing fel-
spar is always plagioclase ; in all of them orthoclase occurs only as
a subordinate ingredient, and hornblende and mica, or both of these
minerals together, form the next most important constituents ; and
in all of them grains of free quartz occasionally appear in the mass,
converting the “trachytes” into “‘dacites” and the ‘syenites ”
(diorites) into “ granites ” (quartz-diorites).

Rocks of precisely similar constitution to the so-called ‘‘ granite
and syenite ” of Hodritsch are very widely diffused in Hungary and
Transylvania, and in the countries lying southwards. In the
Vlegyasza, Bihar, and Rodnaer ranges in Transylvania, Dr. Doelter
has described the andesites and quartz-andesites as becoming
highly porphyritic, and even granitiform in structure*. The rock
of Ilova Thal, which Richthofen compared with the ‘“ Nevadites”
of the Rocky Mountains, appears from the examples of it which I
was shown by Dr. Tschermak to be identical in character. In the
Banat there occurs a precisely similar rock, to which Cotta applied
the name of ‘‘ Banatite ;” and this forms intrusive masses which, as I
am informed by Prof. Szabd, are of clearly younger date than all the
members of the Cretaceous series. In the Euganean Hills we have
quartziferous hornblende-andesites which are certainly as recent
as the Eocene, if not of still later date. In the Alps we have the
granitic mass of the Adamello, for the rock of which Vom Rath has
proposed the name of “ Tonalite;” and a rock of very similar cha-
racter from Queensland has been described by Daintree. In order
to illustrate the composition of these different granitic rocks, all
distinguished by the presence of free quartz with a triclinic felspar
as the predominating ingredient, I place side by side the following
analyses :—

III. Tonaliteot VY. eee
I. Dacite (Quartz- II. Banatite the Adamello ee oa
Andesite) of Kis of Szaska, | Group in the blende-ande-

Sebes, Transyl- Hungary, Alps (Vom sites, Huga-
vania Gigelian. (lesen. Bath) (ies a
SI ee rere 66°32 65:84 66°91 67°98
iin: 14°33 15°23 15:20 13°05
Oxides of iron and
manganese .. 5°78 3°50 6°45 5°69
MITE. 3 Stale hee ss 4-64 4-74 3°73 1:63
Macnesia: .:%.. - 2°45 2-31 2°35 0-14
Potash 1°61 3°06 0:86 323
aia: . tee 3°90 2:96 3°33 7°96

OSS. 25 teen se PhS 0:98 (EA iy 0-32

* “Zur Kenntniss der quartzfiihrenden Andesite in Siebenbirgen und Un-
garn, von Dr, C, Doelter,” Tschermak’s Mineralogischen Mittheilungen, 1873,
p. 51.

310 J. W. JUDD ON THE ANCIENT VOLCANO OF

According to the analyses of Karl von Hauer, the felspar in the
different varieties of the Hungarian rocks would appear to be usually
either labradorite or some variety intermediate between that species
and andesite. The results obtained by Szabé by a totally different
mode of investigation are not very divergent from those arrived at by
Von Hauer.

The microscopic examination, which proves that the mineralo-
gical constituents of all these rock-varieties—the andesitic lavas, the
“* sreenstone-trachytes,” and the “‘ syenite and granite” of Hodritsch
—are quite identical, consisting in each case principally of a plagio-
clase felspar with orthoclase, hornblende, mica, and sometimes quartz
superadded, also demonstrates what are the real differences, namely
those of structure, which serve to distinguish, though not by any
hard lines of demarcation, the three several groups of rocks. The
materials of the so-called granite and syenite are all distinctly
crystallized ; the ‘‘ greenstone-trachytes ” consist of a micro-crys-
talline or felsitic base, in which the crystals, often injured and
broken, are imbedded ; while in the andesite lavas the base is usually
of a more compact texture, passing into a glassy condition. I hope
to be able to show in the sequel that these variations of structure
are fully accounted for by the differences of the conditions under
which the several kinds of rocks can be shown to have become con-
solidated.

Having pointed out what appear to me to be the true relations of
the several varieties of igneous rocks in this complicated district, I
will now proceed to notice the different explanations which have
been offered of its general physical structure. Beudant correctly
described the voleanic rocks of Hungary as constituting five well-
marked and more or less completely isolated mountain-groups—those,
namely, of (1) the Schemnitz Mountains, (2) the Visegrad Moun-
tains, (3) the Matra, (4) the Tokay-Eperies group, (5) the Vihorlat
Mountains ; and to these we must add, for Transylvania, two other
groups (6) the Hargitta Mountains and (7) the Transylvanian
Erzgebirge. Other similar masses of volcanic rocks of con-
temporaneous date and analogous composition stretch through
the provinces of European Turkey, forming a link between the
Miocene volcanic rocks of our continent and those of Asia. So
enormously have the volcanic rock-masses of Hungary suffered
from denuding causes, and so much more difficult is it to trace the
relations between the isolated fragments of their short and bulky
trachytic lava-streams, as compared with the wide-spreading plateaux
constituted by the basaltic lavas of Auvergne, that it is not surpri-
sing to find Beudant altogether failing to perceive any connexion
between the different isolated masses, and declaring that almost
every separate mountain appears to be the result of a distinct
eruption.

It is to Von Pettko, whose studies and writings have thrown so
much new light upon the structure of the Schemnitz district, that
we are indebted for the first clear exposition of the true relations of
the different masses of volcanic rocks, and for some most valuable

THE DISTRICT OF SCHEMNITZ, HUNGARY. 311

suggestions as to the form and structure of the great volcano of
which they constitute the ruins. He showed* that the mining-
district of Schemnitz, a comparatively low-lying tract of oval
form, about 20 miles long by 15 broad, was surrounded and over-
looked on all sides by steeply scarped mountain-masses, rising to
heights of from 3000 to 4000 feet above the sea. This encircling
wall, as he pointed out, consists of great masses of andesitic lava, all
exhibiting a slight dip away from the central depressed area, and
alternating with enormous accumulations of tuffs and agglomerates,
the whole being not unfrequently traversed by mineral veins ; while
in the central area rise numerous masses of rhyolitic and basaltic
lavas and tuffs. Here, as the able Hungarian geologist clearly per-
ceived, we have presented to us the characteristic features of a great
voleanic cone, with a central crater of enormous dimensions ; and
he not inappropriately compared the Schemnitz crater with its
central masses of acid rocks and its walls of more basic materials,
to the well-known crater of Rocca Monfina with its enclosed bosses
of rocks, forming the mountains of Santa Croce.

At the time when Von Pettko wrote (1848), the ‘“ elevation-
crater” theory of Von Buch was almost universally accepted by
foreign geologists, and the example of Rocca Monfina was regarded,
on the authority of Abich, as affording a most signal illustration of
its truth. It is not surprising, therefore, that Von Pettko should
have adopted this view of the origin of the great crater of Schemnitz,
and declared it to be a magnificent example of an “ elevation-
erater.” The facts and arguments adduced in support of this once
famous theory by Von Buch, Humboldt, and Elie de Beaumont have
been so utterly demolished by the observations and reasonings of
Lyell, Scrope, and Constant Prévost, that it would be mere waste of
time at the present date to treat the theory as one still possessed of
any vitality. But as regards all the other conclusions of Von Pettko
concerning the Schemnitz district, I believe that they have been
fully substantiated by further observation.

Richthofen, who, at a somewhat later date, made a general study
of the Hungarian volcanic rocks, arrived at some interesting conclu-
sions as to their relations. He correctly defined their true sequence
of eruption to have been as follows—first the trachytes, secondly
the rhyolites, and thirdly the basalts ; but, probably from failing to
perceive and take account of the enormous amount of denudation
which these volcanic masses had been subjected to, he was led to
conclude that only the latter two classes of rocks could be regarded
as having had a true volcanic origin. The vast masses of trachytic
lavas with their tuffs he referred to a series of ‘‘ massive-eruptions ;”
and not recognizing the essential identity in the chemical and mine-
ralogical constitution between the ‘‘ greenstone trachytes” and the
ordinary trachytes of the district, he applied to the former the name
of “‘ propylites,” and assigned their origin to an older series of

* “Hrhebungskrater in der Umgebung von Schemnitz und Kremnitz,” Be-

richte iiber die Mittheilungen von Freunden der Naturwissenschaften in Wien,
Bd. iii. p. 199.

3) Be J. W. JUDD ON THE ANCIENT VOLCANO OF

“‘ massive-eruptions”” than that which he supposed to have given
rise to the latter.

I will now proceeed to describe in detail the conclusions to which
I have been conducted by a careful study of the volcanic rocks of the
Schemnitz district, with the aid of the light thrown upon them by
the labours of former investigators.

The epoch of maximum volcanic activity in Hungary was that
true “‘ age of fire” in our hemisphere, the Miocene period. So grand
was the scale of the outburst of igneous forces at that time, so
great the destructive effects on older deposits, and so enormous the
accumulations of ejected materials piled above them, that the task
of tracing the nature of the first symptoms and movements which
heralded the appearance, and constituted the earlier stages of this
tremendous manifestation of subterranean activity is a very difficult
one.

The volcanic outbursts of the Miocene period were certainly pre-
ceded by a period of long and continuous, though irregular, upheaval
throughout the areas in which they took place. The marine
deposits of the Cretaceous and Nummulitic periods were succeeded
by the wide-spreading estuarine and lacustrine sediments of the
Oligocene, the latter formations consisting in Hungary and Transyl-
vania of fresh- and brackish-water strata, having a thickness of
several thousands of feet and containing beds of true coal, in one
case of a thickness of 90 feet. Now various deposits of rocks of
chemical origin testify to the fact that even before the close of the
Kocene period, the subterranean forces were beginning to exhibit at
the surface their most feeble manifestations, namely gaseous ema-
nations and hot and mineral springs. But that even at this earlier
stage some true volcanic fissures were formed, along which were
thrown up a succession of cones, the isolated ‘‘ necks ” of teschenite
and pikrite in Silesia, which have been so ably investigated by Ho-
henegger and Tschermak, abundantly testify. Nor is similar evidence
wanting in Hungary. Dr. von Hantken, the Director of the Hun-
garian Geological Survey, pointed out to me that in the midst of the
Visegrad group of mountains small deposits of tuff occur interstra-
tified with the coal-series of the Oligocene; and this fact is recorded
in the memoir which is published on this district by the Geological
Institute of Hungary.

Dr. Szabé also suggested to me that some isolated masses of
quartz-trachyte rocks near Buda-Pesth, which are totally different in
composition and character from the widely spread andesitic lavas of
the Miocene, might not improbably belong to this-earliest stage of
the great volcanic outbursts. It is not surprising, however, to find,
both in Hungary and the adjoining countries, that the subsequent
eruptions of the Miocene period on so grand a scale, and upon the
same sites, have almost entirely destroyed the evidence concerning
‘the first feeble manifestations of those forces which culminated in
such tremendous displays of violence.

That the grandest of the Tertiary volcanic outbursts into Hungary
and Transylvania—those namely which gave rise to the formation

THF DISTRICT OF SCHEMNITZ, HUNGARY. 313

of the vast masses of andesitic rocks—commenced at the beginning
and continued during all the earlier portions of the Neogene period
(that is to say, that they belong to the age of the Miocene proper,
the Upper Miocene of Lyell), there is not the smallest room for
doubt. At many points, as in the Piliser Gebirge, the tuffs and
lavas of this voleanic series are found overlying even the highest
members of the Oligocene formation, while the fine flora and fauna
yielded by the tuffs themselves, and certain fossiliferous sedimentary
deposits, which at some points are found alternating with the volcanic
rocks, are quite sufficient to establish this conclusion as to the age of
the latter.

An examination of the vast masses of lava, agglomerates, tuffs,
and ashes themselves, as well as of the fossils associated with them,
leads us to infer that the eruptions which produced them were of a
subaerial character; and upon a review of the whole of the pheno-
mena, we cannot escape the conclusion that during that great ex-
tension of the land-area which characterized the Miocene epoch in
this part of the globe a chain of grand volcanos rose parallel to the
great curved mountain-axis of the Carpathians. Of the still active
volcanos of Europe, the only one which will bear comparison in
point of dimensions with the great eruptive centres of Hungary is
Etna. But during the period at which the Hungarian volcanos were
formed, a complete girdle of volcanic mountains of almost equal di-
mensions, though pouring out various materials at different points,
surrounded the whole Alpine system and stretched far eastward into
Asia. At the same period another chain of equally grand volcanos
stretched from the Arctic circle southward, through Iceland, the
Faroes, the western side of the British Islands, and thence through
the district which is now occupied by the Atlantic, to a latitude
south of that of the Cape of Good Hope.

‘No mistake could, however, possibly be greater than to suppose
that during this period of the tremendous paroxysmal displays of
volcanic energy, there was any thing like an interruption of geologi-
cal continuity in the area, or even to imagine that our portion of
the globe exhibited the effects of more powerful subterranean action
than can, at the present time, be witnessed upon other parts of its
surface. On the contrary, we find the clearest proofs that these
ejections of yoleanic material which have produced such a vast ag-
gregate result, were separated by very wide intervals of time, and
that the whole continent, and even the slopes of the voleanos them-
selyes, must have been clothed with that magnificent flora of which
such interesting relics have been preserved for our study. We have
also abundant evidence that innumerable lakes were formed among
these old voleanos, and quietly filled with sediments, and that these
were in turn buried under subsequent ejections from the igneous vents.
Nowhere, indeed, can the vastness of the periods of time represented
by the great Tertiary epochs be better appreciated than in Eastern
Europe, where we find each of them represented by deposits several
thousands of feet in thickness.

Confining our attention again more particularly to the Schemnitz

314 J. W. JUDD ON THE ANCIENT VOLCANO OF

district, we find the andesitic lavas and tuffs ejected during the Mio-
cene period gradually building up a cone of vast dimensions which
eventually covered an area nearly fifty miles in diameter. That,
as in the case of so many existing volcanos, eruptions took place
sometimes from the summit, and at others on the flanks of the
mountain, we have no room to doubt. The great volcano of
Schemnitz then doubtless presented an appearance very similar to
that of Etna at the present day; that is to say, it consisted of an
enormous central mountain, upon the sides of which innumerable
parasitical cones had been thrown up, each pouring forth its lava
streams. Like Etna, too, the slopes of this old Hungarian voleano
were covered by the magnificent forests characteristic of a warm
temperate climate—gigantic Sequoias and pines, with oaks, beeches,
chestnuts, planes, poplars, and willows, and an undergrowth of vines
and herbaceous plants of subtropical character. And through these
forests roamed the Mastodon, the Dinotherium, the Anchitherium,
and the Hyotherium, with species of the Rhinoceros, Tapir, and
many other forms, all of which are now extinct.

This period of most violent eruption in the great Schemnitz volcano
appears to have been terminated by a paroxysmal outburst on the
very grandest scale, which resulted in the formation of a prodigious
crater of oval form. Vast as must have been the dimensions of this
crater, they did not exceed those of some of the still active volcanos
of the globe; and cavities of equal size have perhaps been produced
by eruptions that have taken place during the historical epoch. The
result of this tremendous paroxysm was to reduce the Schemnitz cone
to a crater-ring of vast dimensions, one of those ‘*‘ basal wrecks” of a
volcano, as Mr. Darwin so appropriately named the similar magnifi-
cent examples which are found among the islands of the Atlantic
Ocean.

The great paroxysmal outbursts which produced this vast crater
must have been succeeded, as is so frequently the case under similar
circumstances, by a considerable subsidence of the whole mountain
mass; and in consequence of this, not only were all except the highest
portions of the crater-ring submerged beneath the sea, but the latter
found its way into the interior of the crater. The wider-spread
movements which had been taking taking place in Eastern Europe
had resulted, as Von Hauer has so well shown, in the formation of a
number of inland seas with which the waters of the ocean had
not a perfectly free communication, and which are accordingly cha-
racterized by a stunted marine fauna. By a continuation of the
same earth-movements the communication of these seas with the
ocean was completely cut off, and a series of Caspians formed, cha-
racterized by a very remarkable brackish-water fauna. During these
periods (those, namely, of the “‘ Sarmatische Stufe,” and of the ‘« Con-
geria- and Inzendorf beds”’) the Hungarian volcanos rose as islands,
still clothed with a flora very similar in character to that of the pre-
ceding period, but in which the mammalian fauna had undergone
considerable change, through the disappearance of many of the old
forms and the addition of such new and more modern types as Hip-

THE DISTRICT OF SCHEMNITZ, HUNGARY. 315

potherium, Machairodus, and Antilope, with species of Sus and Hy-
ena. The volcanic islands of Hungary must at this stage of their
history have been very similar in appearance to those of the Grecian
archipelago.

That during this period of subsidence the basal wreck of the
Schemnitz cone underwent a vast amount of denudation we have the
clearest proofs ; the crater-ring was evidently breached at a number
of points, and “ barrancos” formed leading into it, the sea being
thus permitted to enter it and to eat back the cliffs around it, with
the same kind of action that Lyell and Darwin have shown to be
taking place in the submerged volcanos of the Atlantic; and in this
manner the extent of the vast “caldera” was probably greatly
increased. Around the flanks and in the centre of the ruined vol-
cano, deposits enclosing relics of the interesting fauna and flora of
the period were accumulated—beds with marine, brackish-water,
and freshwater shells alternating with lignites, formed by the
numerous vegetable remains carried down into the lagoons, and
with masses of tuffs ejected from the volcanic vents.

It was during this condition of the volcano that the last eruptions
of the Schemnitz area took place ; and these were almost entirely
confined to the central lagoon occupying the crateral hollow of the
old voleano. The materials ejected during these later outbursts
consisted in the first instance of quartz-trachytes (liparites or rhy-
olites) with the beautiful varieties of perlitic rocks which we have
already referred to as having probably been formed for the most part
under water and not constituting true pumice. The eruption of these
highly acid rocks was followed by that of those of extreme basic
character—the basalts. At this period of its history the appearance
of the voleano of Schemnitz must have presented a most striking re-
semblance to that now exhibited by Santorin. The latteris a nearly
submerged and greatly ruined crateral ring, the interior of which is
occupied by the sea, in the midst of which a number of new and
smaller cones are being gradually built up by the eruptions which
are taking place at the present day. In the Schemnitz caldera the
ejection of the rhyolitic lavas and tuffs was followed by that of the
basalts, and a number of cones were formed which gradually rose
above the surface of the central lagoon ; and thus the eruptions that
were at first submarine gradually became subaérial ones. By a
continuance of these outbursts, the depression of the central area
appears to have become eventually entirely filled up, and the sea ex-
cluded from it, the process being probably aided by a process of slow
upheaval which the whole district was gradually undergoing.

It is impossible to pass by in silence the points of analogy in the
succession of the different lavas of the Tertiary Hungarian volcanos
and that of the more recent ones of the Lipari Islands*. In both
cases the ejected materials which have formed the great mass of the
volcanos, consist of rocks of composition intermediate between the
acid and the basic types; and the eruption of these was followed,

* See Geological Magazine, Decade ii. vol. ii. p. 60.
OJ. Gs. Sano, 127. Z

316 J. W. JUDD ON THE ANCIENT VOLCANO OF

after an interval of quiescence, by the final outbursts, first of a highly
acid, and then of an extremely basic character. Have we here an
indication that during a period of temporary rest in a volcano, the un-
ejected materials below it may undergo a separation into two
“magmas” of different composition and specific gravity, similar to
those which, according to the well-known theory of Durocher, have
been universally formed below the earth’s crust?

The formation of the scattered basaltic cones within the great
erater-ring of the Schemnitz volcano, and of a few similar ones
(‘‘ puys ”) around its flanks, were the last great efforts of the igneous
forces of the district. Subsequently their declining energies appear to
have been equal only to the origination of fumaroles and hot springs or
to the production of those earthquakes to which the region is still occa-
sionally lable. The extent of the deposits, both siliceous (“sinter”)
and calcareous (‘‘ travertine”), formed by these springs shows clearly
how long, even after the final extinction of the volcano, the subter-
ranean forces retained a portion of their vitality, and how very slowly
and gradually the igneous activity of this region sank into complete
quiescence.

We have endeavoured in the preceding pages to investigate the
several varieties of rocks of which the old Schemnitz volcano was
built up, and, by the aid of the analogies of more perfect and still
active volcanos, to reconstruct the history of the events to which its
original features and its present condition are to be ascribed. Let
us now proceed to inquire what are the appearances exhibited in the
interior of this old volcanic pile as now exposed to us, which serve
to illustrate the nature of the operations going on deep below the
surface, synchronously with those phenomena which we ordinarily
witness at the surface as the accompaniments of volcanic activity.

And although we have not here the remarkably clear sections af-
forded by sea-cliffs, as in the ancient Hebridean volcanos, yet in the
great volcano of Schemnitz—the whole heart of which has been blown
out by tremendous paroxysms, while the vast crater thus originated
has been further enlarged by the action of the sea which entered it
after the partial submergence of the mass—many valuable opportu-
nities are afforded to the geologist for studying the relations of the
originally deeply seated masses of the volcanic pile. Since the final
upheaval of the district, too, the great rock-masses have been greatly
reduced by subaérial denudation, and important sections of them ex-
posed in the valleys; and, finally, the innumerable mining-works
opened all over the central area, and a series of deep railway-cut-
tings have all served to throw light upon the relations which the
different rocks of the district bear to one another. Aided by all
these different kinds of exposures of the rock-masses of the Schem-
nitz area, we propose to describe the structure of the interior of this
old volcano and to investigate the connexion between its internal
mechanism and the operations which we have shown to have taken
place at the surface.

We have already exposed the grounds on which we have been led
to the conclusion that the so-called “granite and syenite,” the

THE DISTRICT OF SCHEMNITZ, HUNGARY. OL:

“ sreenstone-trachyte’”’ and “ dacite,” and the different varieties of
andesitic rocks in the Schemnitz district are chemically and minera-
logically identical, and that their variations in structure are simply
the result of differences in the conditions under which they have
been consolidated. That the hornblende- and mica-andesites were
actual lavas poured out at the surface, we have the clearest proofs ;
and that the “ greenstone-trachytes,” the dacites, and the “ syenite
and granite”’ constitute intrusive masses which have been forced,
while in a plastic and highly heated condition, through the midst of
the sedimentary rocks, is clearly shown by the phenomena presented
at the planes of contact of the two classes of rocks, and by the
dykes and veins which are seen proceeding from the one into the
other.

The rocks of sedimentary origin, through which the igneous masses
forming the centre of the old Schemnitz voleano have forced their
way, have been supposed to include certain members of the Paleo-
zoic series; and these rocks, which, as we have already pointed out,
were at first generally referred to the Primary group, have been sub-
sequently identified, though on very doubtful grounds, seeing that
they contain no fossils, first as Devonian, and secondly as Dyas.
They are supposed to be overlain by the Triassic and Nummulitic
rocks of the district, which are unmistakably identified as such by
the fossils which they contain.

But after a very minute and careful examination of the central
district of the Schemnitz voleano, I have been led to the conclusion
that there exists within it no series of rocks of older date than the
Trias, and that the supposed Paleozoic rocks—the quartzite and
quartzschiefer, the crystalline and schistose limestone, and the
schists, gneisses, and aplite—are nothing but the arenaceous, cal-
careous, and argillaceous beds of the Trias, metamorphosed by their
contact with the Miocene intrusive masses.

I have already enumerated the chemical and petrological grounds
on which I have been led to infer that the rocks of granitic structure
in the midst of the Schemnitz volcano are identical in composition and
age with the surrounding andesitic lavas. ‘here are also several
circumstances in connexion with the physical structure of the district
which lend the strongest confirmation to this view. Were the igne-
ous masses of Hodritsch, as is asserted, an old granitic boss upon
which the so-called Permian and the Triassic strata have been de-
posited, the conglomerates associated with these latter could scarcely
fail to contain pebbles of the granite and syenite. Such, however,
is never the case, as I can testify after a most careful examination
of the whole district, with a view to the determination of this point.
Similarly, the absence of all pebbles of the metamorphic rocks (the
quartzites, schists, and gneisses) of the district in the conglomerates
of the Trias is quite inconsistent with the view that the former were
in existence before the latter and constituted the basis on which
they were deposited. Another proof that the rock of Hodritsch is
not an old granite boss upon which the sedimentary rocks were un-
conformably deposited, is found in the circumstance that the Triassic

Z 2

318 J. W. JUDD ON THE ANCIENT VOLCANO OF

rocks are never seen resting directly upon a denuded surface of the
granite*, but are always separated from it by a zone of highly meta-
morphosed rocks.

The most satisfactory method of determining the true relations of
the igneous, metamorphic, and sedimentary rocks of this district is
by studying their exposures along the Hodritsch, Eisenbach, and
Glashiitte valleys and in the mountain-masses of the Kohl-Berg
(see fig. 1) and the Schwatzer Berg (see fig. 2) which separate

Fig. 1.—Section through the Kohl-Berg.

Schuttris- Glashiitte
Ss. berg. valley. iN),

Schistose rocks. | 3
. pa Benes ‘ B x. “Syenite” (dio-
a. Werfener 3 q. cua oe ‘ | g : aa (
Schichten. = es eri cake BS ge. ‘* Greenstone - tra-
b. Limestone and f 5 & Gane red chyte” (Horn-
dolomite. E t. Crystalline | = oe - ande-
al limestone. Js ie

these valleys. Here we find, in innumerable instances, the lime-
stones, sandstones, and shales of the Trias, wherever we approach
a mass of ‘“ syenite ” or ‘ greenstone-trachyte,” undergo a marked
change and acquire by the most imperceptible gradations a more
erystalline character, till finally they assume the condition of a
crystalline or schistose limestone, of a quartzite or quartz-schist, or
even of a gneiss or aplite (granulite). I believe that it is impos-
sible for a geologist to examine all the interesting lines of junction
between the igneous and stratified rocks without being fully con-
vinced that the metamorphic masses which in every instance occur
between them have been produced by the action of the Miocene
intrusive rocks upon the Triassic strata which they have penetrated
(see fig. 3).

I do not propose in this place to enter upon a discussion of the
details of the interesting problems which are so admirably illustrated
by these beautiful examples of local or contact metamorphism, and of
their bearing on the wider question of the nature and causes of the
phenomena of general metamorphism. On a future occasion I hope
to lay before this Society the results at which I have arrived from
the study and comparison, both in the field and in the cabinet, of
the rocks which exhibit the most distinct and rapid transition from

* In this place, as in many other portions of this memoir, I have used the
names by which the rocks of the district have been so long indicated, rather
than those which I have endeavoured to show in the present paper would be

more appropriately assigned to them, in order to make my line of reasoning
more clearly manifest.

= Fig. 2.—Section through the Schwatzer Berg.
aa)
Schwatzer Kanzian Hisenbach
Hodritsch. Berg. Wrch. valley.

x Zz
p. Schistose rocks. A es Sees os
a. Werfener Schichten g. Quartzite. Metamor a G! Syenite ” (diorite). i‘
(Lower Trias) hc Anite phosed z. “Greenstone-trachyte
é ss Giclee | Trias. (Hornblende-andesite).

Fig. 3.—Seetion from the Hisenbach valley to the Hodritsch valley (based on that of Von Adrian).

Hodritsch Wartacik Klokec Rudna Diszowa Hisenbach
valley. Kompberg. Wrch. Wrch. valley. Wrch. valley.
! i i | {

THE DISTRICT OF SCHEMNITZ, HUNGARY.

Ww ; p. Schistose a4 M ‘i
a. Werfener Schichten. ; Lower Trias. g. Quartzite. peumer x. “Syenite” (diorite).
h ; old
6. Limestone and dolomite. : phosed ti 9 ;
ce. Nummulitic limestone Hocene e ais Trias. y. “ Dacite” (quartz-diorite).
. : s. Gneiss,

320 J. W. JUDD ON THE ANCIENT YOLCANO OF

the unaltered to the most highly metamorphosed condition in Scotland
and Hungary, and of the light which they throw upon the important
question of the origin of foliation. At the present time I propose only
to notice the interesting relations of the rock called aplite, which ap-
pears to be the final product of metamorphism of some of the Triassic
rocks. Upon the old theory that the granitic mass of Hodritsch was of
ancient date, and that the Dyas and Trias strata were deposited upon
it, the relations of the aplite rock appeared to be altogether inexpli-
cable; while the views of the relations of the igneous, metamorphic,
aid sedimentary rocks of the Schemnitz area which I have endea-
voured to enforce in the present paper suggest, as I hope to show,
a remarkably simple explanation of the phenomena, which I think
will commend itself to the minds of all geologists.

The aplite (a variety of granulite) is a crystalline rock composed
of orthoclase and quartz, to which hornblende, tourmaline, and other
minerals are sometimes added. It always occurs at the junction of
the “syenite” and gneiss, and while it sometimes forms masses
having a general parallelism with, and graduating insensibly into,
the latter, it is not unfrequently found sending off veins into the
gneissic rocks, and thus behaves like an intrusive rock, as von Lipold
has well pointed out (see fig. 4).

Fig. 4.— Aplite forming a vein in Gneiss at Schuttrisberg.

;
a ~

iets aS
fel St Set
a
ees Ss
Sie exes
<

r. Aplite. s. Gneiss.

At first sight the occurrence of this granulite, or ‘“‘ semi-granite ”
(as it is well called), at the junction of the granitic and gneissic
masses might be thought to lend some support to the views enter-
tained by several geologists of the metamorphic origin of the granitic
rocks themselves. But a more careful study of the conditions here
presented will be sufficient to convince us that, here at least, any
such doctrine is altogether untenable. Microscopic and chemical
examination alike demonstrate conclusively that the aplite cannot
possibly be regarded as an intermediate stage between the gneiss and

THE DISTRICT OF SCHEMNITZ, HUNGARY. 321

the “syenite.” Of the former it certainly represents the last stage
of alteration ; but with the latter it has absolutely nothing in com-
mon. Itis a rock of far more acid composition, and it cortains
much free quartz and orthoclase felspar ; while the igneous masses in
contact with it have little or no free quartz, and the felspar is prin-
cipally plagioclase.

The phenomena presented by these masses of aplite can only be
explained, as it appears to me, by regarding the rock as the last
stage of metamorphism of the Triassic sediments. That their par-
ticles must have attained the condition of almost absolute internal
mobility is shown by the nearly perfectly crystalline character of
the rock ; and that, as a consequence of this, the mass was in a
truly plastic state is shown by the fact that it was capable of being
injected in the form of veins into the fissures of the surrounding
masses. ‘Tio the same cause (its perfect internal mobility), as I
shall hereafter show, we must assign that disappearance of the folia-
tion which distinguishes the surrounding gneissic rocks, into which
it frequently passes, however, by the most insensible gradations.

We thus see that the examination of the ruins of the old Hun-
garian volcano leads us to results identical with those which we
obtain by the study of the similar ones of the same age in the West-
ern Isles of Scotland—namely, that portions of the same liquefied
masses of rocks, which at the surface flowed in lava-streams or
were scattered by explosions as scoriz or ashes, consolidated at a
great depth and under considerable pressure in the interior of the
voleano, in a highly crystalline and even perfectly granitic condi-
tion. We have also shown that around these vast and slowly
consolidating masses of intensely heated rock, a series of chemical
changes were set up in the sediments which they penetrated,
resulting in the transformation of the latter into schistose and
highly crystalline rocks.

As in the case of the rock-masses of intermediate composition
(the andesites and diorites) we find the more deep-seated portions
assuming a highly crystalline and even granitic character, so we
also have grounds for the conclusion that, under the same condi-
tions, the rhyolitic and vitreous rocks of acid composition pass
into highly erystalline quartz-trachytes, presenting precisely those
characters which are regarded by some petrologists as belonging
only to rocks of very ancient date. This highly porphyritic or semi-
granitic form of the rhyolitic rocks is well exhibited by the great
intrusive masses near Eisenbach and Konigsberg.

Of the interesting minerals which result from the long-continued
reactions taking place between masses of highly heated volcanic
materials and the surrounding sedimentary rocks—and with which
we are so familiar at Monzoni, an old volcanic vent in which they
are seen lying in their original situation, and at Vesuvius, where
they are continually being ejected from the crater—we find several
very interesting examples in the case of the great Schemnitz vol-
eano. Thus, near Hodritsch, a mass of Trias dolomite entangled
in the dioritic rock exhibits, near the planes of its junction with the

322 J. W. JUDD ON THE ANCIENT VOLCANO OF

igneous mass, beautiful examples of ophicalcite; and in a similar
situation at no great distance, masses of finely crystallized
fassaite and pleonaste also make their appearance. LEpidote, mica,
garnet, and other minerals* are also found under similar conditions
at several points in the district.

Although there are, as we have seen, so many points of resemblance
between the volcanoes of Hungary and those of Scotland, yet there is
one respect in which they afford a very remarkable contrast. In the
case of the Scottish volcanoes we observe but little evidence of those hot
and mineral springs and of those discharges of gas and vapour which
usually constitute the final stage of volcanic activity; while of the
existence and action of these, as we have seen, there are in the case
of the Hungarian volcanoes most abundant proofs. Now when we
come to examine the deep-seated rocks of each of these districts,
we find also a corresponding difference. While the rocks which once
formed the central masses of the old Scottish volcanoes have undergone
comparatively little chemical change since their eruption, those in
the midst of the Hungarian volcanoes have clearly been affected
in the most powerful manner through the penetration of their mass
by acid gases and vapours. Oiten the whole mass of rock over a
considerable area is found to be completely decomposed and reduced
to a white or variegated friable product, to which the name of
‘* oreenstone-tuff” has been sometimes, though erroneously, applied.
At other times the change has been a more partial one; and over
very considerable tracts the ‘ greenstone-trachytes” or “ propy-
lites” are found to be charged with the sulphides of iron, copper,
lead, and antimony, often containing some silver and gold. Where,
however, fissures have existed in these rocks, these various metallic
substances have been accumulated in greater abundance, and have
given rise to the formation of those valuable lodes for which the
district is so famous.

It does not of course fall within the scope of the present memoir
to attempt to give any account of the great mineral deposits of the
Schemnitz district, further than is necessary to show the connexion
between the causes by which they are formed and the volcanic
activity of the area in which they occur.

The fissures occupied by the Schemnitz lodes have a general
direction from N.N.E. toS$.8.W. They traverse the so-called granite
and syenite, the ‘ greenstone-trachyte,” and the metamorphosed
Triassic strata, und also extend into the midst of the trachytes and
other rocks which surround the centre of the old volcanic mountain,
the most productive portions of the lodes appearing to be always
situated in the central masses of “ greenstone-trachytes.” According
to Von Pettko there is evidence that in some cases the infilling of
the mineral veins 1s even of later date than the eruption of the basalts.

The phenomena which we have described as being exhibited by
the old volcano of Schemnitz find a very close parallel in the other
great centres of igneous activity in Hungary and Transylvania. In
each case we find a great circular or, sometimes, linear group of

* Jahrbuch der k.-k. Reichsanstalt, Bd. 1v. (1855) p. 183.

THE DISTRICT OF SCHEMNITZ, HUNGARY. 323

mountains (the centre of volcanic activity sometimes showing clear
evidence of having shifted along a line of fissure), which has been
formed by enormous and frequently repeated eruptions of andesitic
lavas and tufts. In every instance we find proofs that the more
deeply seated masses of andesitic lava have, in consolidating, assumed
a highly porphyritic or granitiform structure, and that the action
upon these of acid gases and vapours has resulted in the decomposi-
tion of the mass, with the diffusion of valuable metallic ores
throughout the substance of the rock and their accumulation in
‘considerable quantities wherever a suitable fissure occurred in it.
In each of these volcanos, too, the eruption of the andesitic lavas
appears to have been followed, after an interval of subsidence, by the
extrusion of highly siliceous rocks (rhyolites) and, finally, of extremely
basic ones (basalt and augite-andesite) on a much more restricted
scale than in the case of the older ejections. The fossil remains
associated with the masses of tuffs at the several volcanic centres
show that the ejection of the different classes of lavas took place
at periods which were approximately contemporaneous in each case ;
and at every point the decline of volcanic activity was marked by
the appearance of hot and mineral springs and fumaroles.

Conclusion.

The more important results to which we have been led by the
facts and arguments of the foregoing memoir may be briefly
summed up as follows :—

(1) The igneous masses which in Hungary and Transylvania have
been poured forth at the surface, in the form of andesite and quartz-
andesite lavas, have, where undergoing consolidation at some depth
from the surface, assumed a most perfectly granitic character
(‘‘ syenite ” and “granite” of most authors, or, more properly
perhaps, diorite and quartz-diorite). There is the most complete and
insensible transition from these granitic rocks to the true lavas; and
the whole of them are of Miocene date.

(2) Around each of the intrusive masses which constitute the
links between the voleanic rocks poured forth at the surface and the
subterranean reservoirs from which they have been derived, various
chemical actions have been set up in the masses of Triassic sediments
through which these have been extruded; and these actions have
resulted in the production of highly crystalline and foliated rocks
which are identical in petrological character with materials con-
stituting widely spread metamorphic areas.

(3) But, while there appears to be at first sight a gradual passage
from the stratified rocks through these metamorphic masses into the
truly granitic rocks, yet the phenomena displayed by this district
really afford no support whatever to the theory that the granitic
masses are not truly intrusive at all, but represent sediments which
have undergone their final and most extreme metamorphism Zn sttu.
On the contrary, not only is the chemical and mineralogical consti-
tution of the granitic masses far too uniform throughout and too

324 J. W. JUDD ON THE ANCIENT VOLCANO OF

different from that of the surrounding stratified rocks to admit of
any such interpretation, but the final product of metamorphism of
the Triassic rocks is actually seen in the aplite—which must have
been reduced to a liquefied condition, and graduates most insensibly
through the gneissose and schistose rocks into the sedimentary
masses ; and yet the distinction in character and composition between
the aplite and the dioritic rocks is most clear and unmistakable.

(4) The mineral veins of Hungary and Transylvania, with their
rich deposits of gold and silver, cannot be of older date than the
Miocene, while some of them are certainly more recent than the~
Pliocene. Hence these deposits of ore must all have been formed at
a later period than the clays and sands on which London stands ;
while in some cases they appear to be of even younger date than the
gravelly beds of our Crags !

EXPLANATION OF PLATE XX.

In this Plate the same plan of colouring has been adopted as in those illus-
trating the paper “On the Ancient Volcanoes of the Highlands,” published in
vol. xxx. of this Journal. The distinctive chemical character of the several
voleanie rocks is indicated by different colours—blue representing the basic
lavas and red the acid, while the varieties of intermediate composition are di-
stinguished by shades of purple. The depth of tint in each case indicates the
more or less highly crystalline character of the several rock-masses ;* thus dark
purple represents the granitic andesite, or diorite, and fainter tints the ordi-
nary andesitic lavas. ‘The same plan of colouring is adopted both in the map
and section. In the former the broken line shows the approximate boundary
of the great “Caldera” to which the crater of Schemnitz was reduced by pro-
longed denudation ; the continuous line is the direction of the section below,
which is drawn on a true vertical and horizontal scale, twice as great as that of
the map. ‘The scale of the map is about one inch to six English miles; that
of the section one inch to three miles.

Discussion.

The Prestpunt (Prof. Duncan) remarked that it was curious to
note that the same volcanic rocks had been ejected in past
times as were being thrown out at the present day; whilst, on the
other hand, in two neighbouring districts volcanic rocks of com-
pletely distinct natures were being ejected at the same time. He
dia not consider plants to be a safe guide to the age of beds, as
they are known to extend through long periods.

The Dux oF ARGYLL expressed his personal thanks to Mr. Judd
for the light he had thrown on volcanic rocks, more especially those
of Scotland. The most important point was the identity of rocks
hitherto considered distinct. In the Isle of Mull granite had been
found passing insensibly into porphyritic and volcanic rocks.

Mr. W. W.Smyru said that formerly the rocks of this district were
divided into metalliferous and non-metalliferous. In going over this
spot with Von Hauer and others, he had pointed out Nummulites at
Kisenbach, and had then maintained that the crystalline rocks there
were metamorphosed Nummulitic Limestone. A map of part of
this regiun, which he had drawn at that time, agreed in most re-

Quart. Journ. Geol. Soc .Vol. XXXII. Pl. XX.

Aqueous Rocks.

COGEIE
Sa ata

Eocene
(Nammaltic )
Rocks.

Triassic Rocks
metamorphosed

wv places and
passing into Quartzte
Schist, Gneiss & Aplite

Igneous Rocks Age.
Basaltic s
tatis a)
Basalt and aS
Augite andesite Ny
lavas. NI >
aa S
Rh yolitic and s
Pamice tufts S
Quarts porphyry (' S
and Weyolités S

lavas

Cr

Andesttic
tufts
NI
S
‘Greenstone - S
Trachyte by
ud Bogle F S
~S
Syertate =
and
Granite”
2 :
BS Py
%
i
we TE

Mintern Bros .imp.

Quart. Journ, Geol. Soc Vol. XXXII. Pl. XX.
DISTRICT OF SCHEMNITZ IN HUNGARY.

SKETCH MAP OF THE

Aqueous Rocks.

Yeogente
Strata

| Eocene
(Nemmuatac)
Rocks.

Igneous Rocks

Basaltic

tall

Basalt and

Augite andesite
BY lcaas

Rhyolitic and
Pumice uitts
Quartx porphyry (

and Rhyolitic

Y lavas

Triassic Rocks
metamorphosed

tv places and
(passing into Guartate
Schast, Gneiss & Aplite

—— =
Strata

Crit Strata Congerva Si

Andesite
tults

Hornblende
Andesite
and Dacite

‘Greenstone - i
Trachyte
and Fropylite”

Older Neogene

Syenite
and

al)
Granite

Kiremitir Stoss

Blaufiss Stoss

Ostra Hora
'BartosL chotha

| gi

SECTION ACROSS THE GREAT CRATER OF SCHEMNITZ.
Mustern Bros imp

THE DISTRICT OF SCHEMNITZ, HUNGARY. 325

spects with Mr. Judd’s; but Mr. Judd had greatly simplified our
notions of what had hitherto appeared to be an excessively complex
district.

Mr. Jupp, in reply, stated that marine shells had been found
along with the plant-remains in the beds associated with andesitic
lavas, and these had borne out the views held by botanists with
regard to the age of these lavas. The Duke of Argyll’s discovery of
the leaf-bed in the Isle of Mull constituted the starting-point for all
our knowledge of the age of the voleanic rocks of the Western Isles.
The transition of granite into the voleanic rocks was very marked
in that island. The difference between the rocks with glassy plagi-
oclase and the earthy rocks in the Schemnitz district was possibly
connected with the presence or absence of metallic deposits. In
Scotland the rocks were destitute of ores ; but in Hungary they had
evidently been subjected to the action of acid gases, and valuable
metalliferous minerals are found abundantly in them.

326 E. T. NEWTON ON TWO CHIMZROID JAWS FROM

35. On two Coimarorp Jaws from the Lower Greznsanp of New
ZEALAND. By EK. Turrtzy Newron, Esq., F.G.S., of H.M.
Geological Survey. (Read June 7, 1876.)

(By permission of the Director-General of H.M. Geological Survey.)
[Puate XXI.]

Havine been working for some time past at the Cretaceous Fishes,
my interest has been especially centred in the group of the Edapho-
dontide ; for the specimens belonging to this group, which have
been brought to light within the last few years, have much enlarged
our knowledge of these peculiar forms of fossil fishes. The general
results of my examination of this group, will shortly be pub-
lished; and in the present communication it is only proposed to
consider two specimens from the Lower Greensand of New Zealand,
which have been deposited in the British Museum by Dr. Hector.
My attention was first directed to these specimens by Mr. W. Davies,
of the British Museum, when I was examining the fine series of
Chimeeroid jaws now in the national collection.

One of these New-Zealand specimens is the right mandible of a
species of Ischyodus already known to us from the Gault of Folke-
stone; and the second is a small right maxilla, which is altogether
new and appears to be generically distinct from any of the fossil
forms hitherto described.

Before considering the peculiarities of the mandible, it will be ne-
cessary to enter into some explanation and description of the species
to which it is referred.

It appears that the late Prof. Agassiz, some years ago, saw in the
Earl of Enniskillen’s collection of fossil fishes, a Chimeeroid mandi-
ble from the Gault of Folkestone, which he named, in manuscript,
Chimera brevirostris. Subsequently, in his ‘ Poissons Fossiles’
(1843)*, he alluded to this specimen under the same specific name,
placing it in his subgenus Jschyodon t; but, most unfortunately, he
gave no description or figures. Since the publication of that
standard work of fossil ichthyology, several examples of Ischyodus-
mandibles have been obtained from the Gault, nearly all of which
appear to belong to one species. These mandibles have generally
been referred to Agassiz’s J. brevirostris, simply because they were
the only forms of Jschyodus known from the Gault ; but the want of
a figure or description of the original specimen left the matter in
great uncertainty.

At the request of my colleague, Mr. Etheridge, the Earl of Ennis-
killen has very kindly sent the original type specimen of Agassiz’s
I. brevirostris to the Museum of Practical Geology for examination
and comparison; and I am now therefore enabled to give for the first

* Recherches sur les Poissons Fossiles, ili. p. 344.
+ Agassiz provisionally adopted Sir Philip Egerton’s genus Jschyodus, with
the above modification.

THE LOWER GREENSAND OF NEW ZEALAND. 327

time a figure of the type of this species(Pl. XXI. fig. 4). Although
the specimen is much broken, there can be no doubt of its specific
identity with those mandibles which have been more recently
obtained from the same horizon, and in a more perfect condition.

Among the fish-remains from the Cambridge phosphatic deposits,
examples of J. brevirostris (which can now be definitely determined
to be this species) are not uncommon; and one of the varieties there
found agrees very closely, not only with the type specimen from
the Gault of Folkestone, but also with the New-Zealand Lower-
Greensand example, which forms, in part, the subject of the present
communication.

As I. brevirostris has never been described, it is proposed here to
give the chief peculiarities of the mandible; and considering the
fragmentary character of the type specimen, it is thought better to
describe a more perfect example, and afterwards to compare with
this, first the original type specimen, and then the one from New
Zealand.

Good evidence of the form of the maxilla and premaxilla of J.
brevirostris has now been obtained; but as these parts are not re-
quired here for comparison, I do not purpose describing them.

IscHYODUS BREVIROSTRIS, Agassiz. (Plate XXI. figs. 1-5.)

The specimen of a right mandible, represented by fig. 1, has been
chosen for description on account of its being extremely like Agas-
siz’s type specimen, and also because it is intermediate in form be-
tween two or three extreme varieties, this species being, like others
of the family, very variable in form. The general appearance of the
jaw will be best understood by a reference tothe figure. It is some-
what more triangular in outline than is usually the case in this
genus, or even in other specimens of the present species. The oral
margin (a to d) is seen to be deeply indented anteriorly and pos-
teriorly. The front or symphysial margin is slightly convex. The
posterior margin is irregular in outline, being always more or less
broken. It is evident from the form of these jaws, that they were
continually growing at the hinder part, and being thrust forward as
the anterior portions were worn away ; from this arrangement it is
obvious that at the hinder part the bone would be only imperfectly
ossified, and consequently easily broken ; and as a matter of fact
the hinder part of these fossils is always more or less broken.

Viewed from the inner side, as in fig. 1, the oral surface (0s) of
this mandible is seen to be armed with five of those peculiar denti-
nal surfaces which have been sometimes termed tritors, but which
will be alluded to in this paper as teeth (fig. 1,abcde). The
first of these is placed at the end of the beak (a). The dentinal
substance of which this is composed is arranged in a series of plates
or lamella, set somewhat obliquely to the anterior border. These
lamelle are not seen in fig. 1; but in specimens with the outer
surface denuded they are seen to extend along near the front mar-
gin asinfig.2. The dentinal substance of which the other four teeth
are composed is not arranged in lamelle, but each tooth consists of a

328 E. IT. NEWION ON TWO CHIMAROID JAWS FROM

number of tubes (filled with dark substance), generally set perpendicu-
larly to the surface, around which the dentinal substance is deposited.
A surface-view of a portion of one of these tubular teeth is shown
in tig. 8. The tooth 6 (fig. 1) is small, and placed at the anterior
end of the upper margin of the symphysis. The tooth near the let-
ter c causes the projection upon the oral margin, and extends inwards
almost to the large central one, e. The posterior and outer tooth,
d, extends for some little distance near to the margin of the bone.
The median tooth, e, is much larger than either of the others;
in this species there is always a space between it and the symphy-
sial tooth, 6. Below and behind the teeth the bone is covered by a
delicate layer of enamel-like substance, which is smooth and shining
and shows definite lines of growth. The symphysial surface (sy)
is slightly convex ; but towards its oral margin it is marked by a
deep groove with a rounded and projecting ridge immediately above it.

If Agassiz’s type specimen (fig. 4) be compared with the more
perfect one described above (fig. 1), the close resemblance in every
particular will, I think, be obvious, the greatest difference being
that the middle tooth (e) is larger in the type (fig. 4); this differ-
ence, however, must be regarded as of little importance, as it is al-
most wholly due to the fact that the bony covering is more broken
away in the latter. .

It might be thought that the differences in general outline and in
the form of the teeth between the two specimens represented by figs.
1 and 4 would justify their being referred to different species ; but the
examination of a large series from the Gault and Cambridge deposits
has convinced me that J. brevirostris varies to a far greater extent
than is shown in these two specimens. In some instances the tooth
e is narrower than it is in fig. 1, and extends almost to the oral
margin atc; in other cases the same tooth will be proportionally
much wider, and placed further back. The extent to which the
oral margin is indented also varies much, being sometimes more,
and sometimes less than in fig. 1. The proportional length of the
oral margin from a to d likewise varies.

The New-Zealand specimen (fig. 5) has the outer surface imbedded
in the matrix, and a portion of the beak broken off; but still the
fossil is sufficiently perfect to allow of a very close comparison.
The description of the Cambridge mandible given above would
answer equally well for this one from the Antipodes; -but a
comparison of the two (figs. 1 & 5) will show that there are some
differences. When, however, the variations which are known to
occur among the Gault specimens belonging to this species are duly
considered, these differences will be regarded as of minor impor-
tance. The greatest dissimilarities are these: in the New-Zealand
specimen the oral margin was probably shorter and less indented,
the margin from d to # is proportionally longer, and the middle
tooth ¢ is larger, and extends further forward and nearer to the
symphysial tooth, 6. If now the figs. 1, 4, & 5 be compared, it
will, I think, be acknowledged that there is as great a difference
between the Cambridge and the Folkestone specimens (figs. 1 & 4)

THE LOWER GREENSAND OF NEW ZEALAND. 329

as there is between the New-Zealand form (fig. 5) and the original
type specimen of J. brevirostris (fig. 4).

-‘Cattoruyncuvus* Hecrori, nov. sp. (Pl. XXI. figs. 6-9.)

The second Chimeeroid jaw, alluded to above as having been
brought from New Zealand by Dr. Hector, is a small right maxilla
which presents certain peculiar and interesting characters.

This maxilla measures little more than an inch in length, and not
more than three fourths of an inch in width at its widest part. The
oral surface of the specimen is exposed, the upper portion being
imbedded in the matrix. The entire bone is much depressed, as
shown in figs. 8 & 9; and this is evidently its natural form, for
there is not the slightest evidence of its having been distorted in
the process of fossilization. Itis narrowest in front, and widens out
at the hinder part, the outer angle being produced backwards,
downwards, and outwards, so as to form a conspicuous process, The
oral surface is provided with but one tooth, composed of the tubular
kind of dentinal substance; this one, however, has a peculiar
form. Posteriorly the tooth appears to pass into the substance of
the bone, as in the other forms of the Edaphodontide ; and at this .
part it is proportionally wide, occupying the greater part of the
width of the bone ; passing forwards it is seen, at a distance of about
z of an inch from its hinder border, to separate into two processes:
one of these passes along the inner side near the symphysis, almost
to the anterior end of the maxilla ; the other extends along the outer
margin for some little distance, but does not reach so far forward as
theinner one. Between these two processes there is a depressed
space, the inner process standing up in relief as a rounded ridge.
Both sides of each of these processes appear to have been worn by
attrition against the teeth of the mandible. Judging from the man-
ner in which this maxillary tooth is worn, it seems at first sight that
there must have been at least three teeth in the mandible—one to
cut out the central groove in the maxillary tooth, and a second and
a third to wear away the inner and the outer sides. But a study of
the recent Callorhynchus shows that the mandible was probably pro-
vided with but one tooth.

It will be remembered that the maxilla of Hdaphodon is charac-
terized by three teeth, that of Zschyodus by four, and that of Hlas-
modus by possessing two elongated teeth and a series of dental
lamin upon the outer margin. In the young condition it appears
that the two long teeth are joined posteriorly and rolled in a pecu-
liar scroll-like manner. In general appearance this tooth resembles
that of the New-Zealand maxilla (compare Pl. XXI. fig. 6 with

* When the paper was read before the Society, this specimen was described
as the type of a new genus under the name of Upsilodus; but some remarks
made by the referee induced me to reexamine the matter. Through the
kindness of Dr. Giinther, I have had the advantage of studying the recent
Callorhynchus antarcticus, and have convinced myself that there are no such
important differences between its maxilla and this fossil as, from descriptions
alone, I had supposed to exist. Ihave therefore felt compelled to refer the
fossil maxilla to Cal’orhynchus.

330 E. T. NEWTON ON TWO CHIMROID JAWS FROM

Geol. Surv. Dec. 6, pl. i. fig. 8); but this resemblance is more appa-
rent than real.

Concerning the maxilla of Ganodus, one would like to be better
informed. Agassiz says (‘ Poissons Fossiles,’ vol. iii. p. 339) “that
the upper jaw of this genus resembles that of Callorhynchus ; but its
external face, in place of being entirely concave, presents only a
deep groove along the external part, and the remainder is convex.”
The New-Zealand specimen does not agree with this description, nor
with the figures which are given. Concerning the mandibular teeth
Agassiz goes on to say that they are near together and united into
a protuberance covered by a bony layer, and that they are placed
so obliquely as to be parallel with the dentary margin.

From these considerations, I think it will be obvious that the
New-Zealand specimen cannot be referred to either [schyodus, Eda-
phodon, or Elasmodus, and also that we should not be justified in
placing it in the genus Ganodus; for it seems extremely improbable
that teeth placed obliquely in the mandible should wear a single
maxillary tooth into longitudinal grooves.

A comparison of the New-Zealand specimen with the maxilla of
Callorhynchus antarcticus shows that, although there are minor dif-
ferences, yet in all their chief points they are remarkably alike.
Their general form and size closely agree. In both there is but a
single tooth, forked anteriorly and covered by the bone posteriorly.
The greatest differences between the two are these: the tooth of
the New-Zealand maxilla is flatter, larger, and has longer anterior
processes than that of C. antarcticus, the tooth of the latter being
more rounded and not approaching so nearly to the borders of the
bone, while the anterior portion is much less deeply indented.

Each mandible of C. antarcticus possesses but one tooth; and it is
this which cuts out the indentation in the front of the maxillary
tooth, thus forming the two processes. The outer margin of the
mandible, although not armed with teeth, is trenchant and wears
away the outer surfaces of the maxillary bone and tooth. The inner
or symphysial edge of the maxillary tooth stands above the surface
of the bone, so that when the two maxille are together there is a
groove between the teeth. This conformation is markedly shown
in the New-Zealand maxilla.

The differences above-mentioned between the maxilla of C. antare-
ticus and the New-Zealand fossil are of such a character that they
can scarcely be regarded as of more than specific value; and it is
suggested that the name of the Director-General of the New-Zealand
Survey, who brought the specimen to this country, should be as-
sociated with it as a specific appellation, and that this Chimeeroid
fish should be known as Callorhynchus Hectori.

With regard to the geological horizon from which the two speci-
mens constituting the subject of the present communication were
derived, the only information I have been able to obtain is, that
they were found, together with other fish remains, imbedded in a
fine conglomerate, believed to be of Lower Greensand age, which
occurs at Amuri Bluff, New Zealand.

i. Quart. Journ.Geol.Soc Vol. XXXII.Pl. XX.

Sharman ad nat del et Eth. M&N.Hanhart inp

CRETACEOUS. CHIMAROIDS.

TIE LOWER GREENSAND OF NEW ZEALAND. 3

DESCRIPTION OF PLATE XXI.

* Iscuyopus BrEvirostris, Ag. Figs. 1-5.

Fig. 1. A right mandible, from the Cambridge phosphatic deposits, seen from
the inner side: a, laminated tooth at extremity of beak ; 6, symphysial
tooth of tubular dentine ; ; ¢, anterior and outer tooth of tubular den-
tine; d, posterior and outer tooth of tubular dentine; e, median tooth
of tubular dentine ; 0s, oral surface; sy, symphysial surface ; ; @, pos-
terior angle. In the Museum of Practical Geology

Fig. 2. A portion of the beak of a left mandible from Gee beideas having the
outer surface ground away to show the laminated structure of the
tooth. In the Museum of Practical Geology.

Fig. 3. A portion of the surface of one of the tubular teeth, enlarged six dia-
meters, showing the central tubes and the surrounding dentine. In
the Museum of Practical Geology.

Fig. 4. A right mandible from the Gault of Folkestone. The original type of
the species alluded to in Agassiz’s ‘Poissons Fossiles,’ iii. p. 344. In
the collection of the Earl of Enniskillen. Letters as in fig. 1.

Fig. 5. A right mandible from the Lower Greensand of New Zealand. In the
British Museum. Letters as in fig. 1.

CaLLoruyncnus Hecrort, sp. nov. Figs. 6-9.

Fig. 6. A right maxilla from the Lower Greensand of New Zealand, showing the
lower or oral surface: a, the inner spur of the tooth; 0, the outer
spur of the tooth. In the British Museum.

Fig. 7. Same specimen, enlarged.

Fig. 8. Same specimen, enlarged, front view.

Fig. 9. & outer side.

3?

Discussion.

Mr. Erneripes stated that he had been called upon to examine
carefully the large series of fossils brought from New Zealand by Dr.
Hector, and that he had been much struck with those associated
with the specimen of Jschyodus noticed by Mr. Newton. These
were undoubtedly of Cretaceous type; and the deposit, in its general
conditions, appeared closely to resemble the remarkable beds of
Upware and Farringdon. Mr. Newton’s paper was of great import-
ance, as throwing much light upon the distribution in time and
space of so interesting and remarkable a group as the Chimeroid
fishes.

Sir Puitre Ecrrton was gratified to find that the divisions esta-
blished by him in this group of fishes some twenty years ago still
held good. Mr. Newton’s determination of the Jschyodus was cer-
tainly of great interest, as demonstrating the occurrence in New
Zealand of a species belonging to our own Greensand.

The AvrHor stated that “the nearest ally of his new form wena
to be the living Callorhynchus. The maxilla was remarkably well
preserved ; and its single tooth seemed to furnish evidence of the
existence of three teeth in the lower jaw. .

Q.J.G.8. No. 127. 2a

38382 J. W. DAVIS ON A BONE-BED IN THE LOWER COAL-MEASURES.

36. On a Bonz-BED in the LowreR CoAL-MEASURES, with an ENUME-
RATION of the FIsH-REMAINS of which it is principally composed.
By James W. Davis, Esq., F.G.S., F.L.8. (Read June 7, 1876.)

Near Bradford and Clifton, in Yorkshire, a peculiar stratum of shale,
which contains so great a number of remains of fossil fishes that I
venture to call it a bone-bed, occurs immediately above the Better-
bed coal, and is known to extend over a surface four or five miles
in length by about two in average breadth.

The Better-bed coal is a member of the Lower Coal-measures, or
Gannister series. It occurs about 700 feet above the Rough rock,
the uppermost bed of the Millstone grits, and is separated from the
Black-bed coal by overlying strata of an average thickness of 120
feet. It is extensively worked by the Low-moor Iron Company,
and used by them in smelting the clay-ironstone of the district. It
is peculiarly valuable for this purpose on account of its freedom from
sulphur, the excellence of the iron manufactured by this firm being
in a great measure ascribed to the use of the Better-bed coal.

The following section (p. 333) will explain the position of the
bone-bed. The section extends from the thick series of sandstones
known as the Elland Flag rock, below the Better-bed coal, to the
Black-bed coal above.

The beds vary very much in thickness within even a small
area, the sandstones and shales often thinning out from a thickness
of from twenty to forty feet in less than a mile, and altogether
disappearing. The most persistent beds are the coals and the fire-
clays, or seat-earths on which they rest. The section may be taken
as a fair average in this locality. |

The bone-bed rests immediately on the surface of the coal, and
varies from a quarter to five eighths of an inch in thickness. Above
is a thick bed of blue argillaceous shale containing layers of iron-
stone nodules, the only organic remains found in it being those of
plants. The bone-bed is composed in a great measure of commi-
-nuted bones, principally of fishes, though remains of Labyrinthodonts
are sometimes found; mixed with these are minute fragments of
coal, often in thin layers of small extent. The whole presents
the appearance of a brownish-black, argillaceous shale, and is easily
distinguished from the light-bluish shale above. It is continuous
over a large area, being invariably found (where the coal has been
worked) from the north-west of Wyke, near Bradford, to Clifton.
Nearly all my specimens are from the latter district.

I would briefly draw attention to the section, in order to point
to the. probable circumstances attending the aggregation and depo-
sition of the strata composing it. Beginning with the Elland Flag-
rock, we find it composed of a great thickness of sandstones, with
intercalations of shale, which were probably of littoral origin, or
may have been the estuary of a large river. A gradual elevation of

J. W. DAVIS ON A BONE-BED IN THE LOWER COAL-MEASURES. 333

Diagram Section of Strata to show Position of Bone-bed near
Bradford and Clifton, Yorkshire.

ft. * in.
Pemsineke bod Coal) 1.ici. bo ib. 6 8 eee octane os 2 10
eMC AMIN cass cdlclenc nay sic s beiusies «sa 0ealn deen dey pee. o@
3. Argillaceous shale ...............seeesereeees 20 O
\
Mi Simones stone ....7....6 00000 ae. { = i566 8
\
IAT 5.5 < asic seep nisepnaniabager ak anes ane tas ge 5. ao
(F
t E
G. SHON sbOME .......cesradearsrersenesens ene — 7
(See
——
Peer enie ye ROU Ste .. { SS 10 0
DO rath 4 idan tnid hy me ala = nel
EP ibiiie shila swith iim, 09), onr-ncon-csie ee 2 0
Cu, | RC ara Se OR OC Caer Pes Wa Seem) = 0 6
ee dirneab uN AIG oR Tia) dea == 15 0
12. Blue shale with ironstone ...........-+.++- 20 «OCO
13. Bone-bed .2.......ccrccoveeseseareconvenerencess i to gin.
14. Better-bed coal ......sse.e..eseeeeeereeeee es 1°3 to 2°6
15. Seat-earth or Galliard ...........-ss000- 208 Sore Oo
= = \
16. Shales and Raggy stone. .......00:..-s8eee 80, 0
17. Elland Flag-rock with partings of shale. 4 <i ea

334 J. W. DAVIS ON A BONE-BED IN THE LOWER COAL-MEASURES.

the land took place and was in course of time covered by a dense
vegetation, which ultimately formed the Better-bed coal. The roots
of the plants are found in the seat-earth below the coal, from which
the inference may be drawn that they grew and decayed where they
are now found, and were not washed from a distanee. The land
became again submerged, probably forming an estuary, whilst the
materials forming the bone-bed were deposited on the coal. That it
was an estuary-deposit receives confirmation from several sources.
The fish-remains found imbedded in it belong to two distinct groups :
the Elasmobranchii, of which the sharks and rays are existing repre-
sentatives, are chiefly confined to salt waters ; whilst a second group,
the Ganoidei, have several living representatives which are only found
in fresh water, as the Lepidosteus, Amia, and Calamoichthys of the
rivers of America and Africa, whilst others, as the sturgeon, can live
either in salt or brackish water. It is by no means certain that the
habits of the fish during the Carboniferous period were exactly sim1-
lar to those of their descendants; but supposing that they were so,
it is not difficult to conceive, when the land was lowered sufficiently
for the formation of an estuarine deposit, that the detached teeth and
spines of the Elasmobranchs might be washed towards the shore by
the tides, and there mixed with the remains of Ganoids from the
river. This also appears probable when the broken and frag-
mentary condition of the fossils is considered; for being constantly
subject to attrition by each succeeding wave, the state in which they
are found would be the natural result.

After some time, the land was lowered still more, and the mud
brought down by the stream, being deposited in deeper water,
formed the bed of blue shale resting upon the bone-bed. ‘The over-
lying strata of alternating beds of coal, shale, and sandstone lead to
the inference that the elevation and subsidence of the land occurred
repeatedly.

Before proceeding to the enumeration of the fish-remains I have
been able to identify from the bone-bed, I wish to acknowledge my
indebtedness to the Earl of Enniskillen, Sir Philip de M. Grey Eger-
ton, and W. P. Sladen, Esq., who have kindly placed their collec-
tions at my disposal for comparison with my specimens, and also to
L. C. Miall, Esq., who identified the remains of Labyrinthodonts.

ELASMOBRANCHS.
1. Gyracanruvs Formosvs, Agass,

2. G. TUBERCULATUS, Ag.

Pectoral and dorsal spines of this genus are frequently met with.
The bones of the pectoral arch are also found, and occasionally
patches of shagreen covered with small tubercles.

The gyrating lines of G. formosus are sometimes found divided into
tubercles, which are characteristic of the second species, G. tubercu-
latus. hut as the various intermediate stages can be traced in a

J. W. DAVIS ON A BONE-BED IN THE LOWER COAL-MEASURES. 335

series of the spines, I infer that there can be no sufficient grounds
a considering the latter more than a variation from the original
ype.

3. CrenacANTHUS HYBoDOIDES, Eg.

Tolerably abundant. In good state of preservation. Spines 12
to 15 inches long.

4, CrenacanTHvs, sp. ?

Straight. Ridges much broader than in C. hybodoides, and much
fewer in number, being only 5 or 6, whilst C. hybodoides has 16 or 18.
The spine presents a fibrous appearance, as though the ganoine had
been dissolved away, exhibiting the bony structure underneath. Its
length is from 4 to 6 inches.

5. CrENACANTHUS, sp. ?

This spine differs from the preceding in being more curved, with
the ridges symmetrical, and not branching into each other, and
closer and finer in appearance. It is 53 inches long, 3 inch broad
at the base, and tapers to a fine point.

6. Lepracantuvs Corer, Egerton.

Was described by Prof. Owen in the ‘Geological Magazine’ for
1869, p. 481, from a single specimen found by the Earl of Ennis-
killen in the coal-shale at Ruabon. ‘The species occurs in the bone-
bed, about a score of specimens having been found. The largest mea-
sures 33 inches in length, and ;3, of an inch in width, being an inch
longer than the one described by Prof. Owen. He says, ‘“‘The spine
is gently curved, moderately compressed, with the back or convex
border rounded ; the thinner concave border is armed by relatively
large recurved pointed denticles, subcompressed and strengthened by
an almost ridge-like swelling along the middle of each side. These
denticles are few in number compared with most similarly barbed
fossil fish-spines: four project from about one third of the length
of the body of the spine; and not more than seven are traceable in
the present specimen.” In one or two of the specimens from the
bone-bed, which have left the matrix without being fractured, the
spine is seen to have been hollow, with a row of denticles on each
side of the concave border; these are placed alternately, and not op-
posite to each other, so that there are 13 or 14 denticles, in place of
seven as seen by Prof. Owen, one half of the type specimen being
hid in the matrix.

7. AcanrHopEs. Warp1, Egerton.

Spines of this species are common. They range in size from an
inch and a half long, and one eighth of an inch broad, to 7 or 8
inches in length, and fully half an inch in breadth. All are
characterized by the deep furrow running parallel to the convex
portion of the spine.- - oe Fae ests ait aay

336 J. W. DAVIS ON A BONE-BED IN THE LOWER COAL-MEASURES.

8. PLEURACANTHUS LEVISsImts, Agass.

Very rare. One specimen very perfect, nearly 8 inches in length,
and five eighths broad at the base, gradually tapering to a blunt
point. A double row of denticles extends, on each side from the
point, five inches along the spine.

9. ORTHACANTHUS CYLINDRIcUS, Agass.

The specimens of this species are generally small, rarely exceeding
three or four inches in length. They are found rarely in the bone-
bed.

10. Dirtopts erBzBosus, Agass.

Teeth rare. They are probably the teeth of Pleuracanthus or
Orthacanthus. Sir P. Grey Egerton has established their connexion
with the former; and Dr. Newberry has found the spines of the latter
associated with the teeth of Diplodus in such a manner as to leave no
doubt that they belonged to the same fish. Both spines were
cephalic and nearly related. It is probable that different species of
Diplodus belong to each of these genera.

11. Gen. nov.

A spine hitherto undescribed. It is 43 inches in length, with a
diameter of 1 of an inch. The only specimen I have is crushed from
the base for 2 inches towards the apex; the remaining portion is round,
without any hollow for the insertion of fin-rays, and gradually tapers
to a point. It is slightly curved, principally on the convex or pos-
terior portion of the spine. The surface presents a fibrous appear-
ance; otherwise it is quite plain and smooth. On the concave surface
of the spine are eight denticles, separated from each other by about 3
of an inch, the intervals between those nearer the base being wider and
gradually becoming smaller nearer the point. The denticles are com-
paratively large, flattened, and blunt, and extend from the spine
nearly 3 of an inch. A second specimen, in the cabinet of W. P.
Sladen, Esq., is completely separated from the matrix; it is nearly
straight, and shows no evidence of a second row of denticles. A
transverse section, cut from the spine about one third its length
from the base and magnified twenty diameters, shows the internal
part to be rather more than half the diameter of the spine, and ap-
parently hollow. The bony structure surrounding the central part
is composed of numerous ridges without definite arrangement, be-
tween and amongst which are pits or vacuoles, the whole presenting
a labyrinthoid appearance. The bony part of the section has been
cracked both from the external and internal margins. The cracks
are filled up with iron pyrites.

It is probably a cephalic spine allied to Pleuracanthus.

12. Hortoncuts, gen. nov.

Three or four pines found in the bone-bed present characters
differing from any previously described. They are 13 inch long
and 3, broad at the base. They are straight except a slight curve

J. W. DAVIS ON A BONE-BED IN THE LOWER COAL-MEASURES. 307

on the anterior margin towards the point. From the apex to 7
of an inch from the base the side of the spine is ornamented by lon-
gitudinal enamelled ridges, six in number; the remaining portion
has been imbedded in the integument of the fish. The line dividing
the two parts lies obliquely across the spine, the posterior border
being the shorter one. ‘The ridges in one or two instances, by run-
ning into each other, present a bifurcated appearance. The whole
of the spine is slightly compressed. Along the posterior margin is
a series of minute denticles scarcely distinguishable without the aid
of a magnifying-glass.

The two genera to which these spines are most nearly allied are
Onchus and Homacanthus of Agassiz. They differ from the former
in having denticles, the genus Onchus being described by Agassiz in
the ‘Monographie des Poissons du Vieux Grés Rouge’ as “ compre-
hending only the Icthyodorulites which are straight or feebly arched,
with longitudinal, smooth, and uniform furrows, and having the base
bevelled ;” whilst in the same monograph Homacanthus is described as
including very small Ichthyodorulites “armed with crenulations on
their posterior border, and their sides ornamented by homogeneous
longitudinal furrows.” The only specimens known are named H.
arenatus “‘ from the arched form of the spines, which are curved like a
sickle.” The spines from the coal-measures differ from Homacanthus
in being straight and having the ridges bifurcating.

Sir Philip Egerton, who has a specimen of the spine, suggests the
generic name ‘‘ Hoplonchus” as signifying the character distinguish-
ing them from Onchus, which appears to be their nearest relative.

13. Pievropus Ranxini, Agass.

14. PLevRODUS AFFINIS, Agass.

I have nearly two hundred teeth of these species from the bone-
bed. They differ very much in form, size, and character. Some
have broad wing-like processes spreading from the main ridge of the
tooth; others are narrow and elongated. Many are ornamented
with a series of well-defined striz running across them transversely,
whilst the majority are smooth and plain or slightly punctured.
The size ranges from two tenths of an inch to seven tenths. Prof.
Agassiz named the two species, but did not describe them. Judging
from the manner in which the species run into each other in speci-
mens in my cabinet, it will probably be found, when the fish is well
known, that the teeth are from different parts of the mouth of one
species rather than belonging to separate ones.

15. Hetopus stmpiex, Agass.
Rare. Two or three teeth.

16. Hetopvs, sp. ?
Rare. Two teeth.

17. CLaDoDUS MIRABILIS, Agass.
Not common. Teeth.

338 J. WwW. DAVIS ON A BONE-BED IN THE LOWER COAL-MEASURES,

18. Pacrtopvs, sp. ?
Teeth.

19. Peratopus Hastinesra, Owen.
Not uncommon. Teeth.

20. Harpaconvs, sp. ?
Teeth.

21. CrenoprycHivs ApPIcatis, Agass.
Teeth.

GANOIDS.

22. Mrearicutays Hippertt, Agass.

The remains of this fish are abundant. The scales are usually
found detached ; occasionally masses are discovered connected ; teeth,
vertebree, and head-bones are common. The fish varied much in
size. One large specimen measures 2 feet 6 inches in length with-
out the head and tip of the tail, which are unfortunately wanting.

23. Hotoprycurus savroipes, Agassiz.

24, SrrePsopus savRorpEs, Huxley.

Scales of the former and teeth of the latter are found in tolerable
abundance, nearly always detached. One or two large teeth, pro-
bably belonging to Strepsodus, are anchylosed to the bone of the jaw.
25. ACROLEPIS, sp. ?

Rare. Very beautifully preserved scales and broken head-plates.

26. Piatysomvs, sp. ?
Rare. A mass of scales is the only example I have seen.

27. AcantHoporsis Ecrrtont, Hane. & Atthey.
Part of a jaw with teeth.

28. AMPHICENTRUM, sp. ?
A tooth and a spine attached to dorsal fin ?

29. Rutzopoprsis, sp. ?
Very rare.

30. CycLoprycHIvs, sp. ?
Rare. Part of a jaw.
31. Gyrotepis Ranxrni, Agass.

One of the few examples from this bed of a fish being found
with tae fins and scales in situ. It is 5 inches long, and exhibits
part of the head, two dorsal fins, and a mass of small scales.

J. W. DAVIS ON A BONE-BED IN THE LOWER COAL-MEASURES. 359

32. PAaLmonIscus.

I have found no good specimen of this usually abundant genus ;
occasionally detached scales and teeth are found which exhibit its
peculiarities.

33. C@HLACANTHUS LEPTURUS, Agass.

The remains of this species are fairly numerous, and indicate a
fish from 12 to 15 inches in length ; and as the specimens figured by
Prof. Huxley in dec. xii. of the ‘ Memoirs of the Geological Survey’
are only 4 to 5 inches Jong, the present examples may be regarded
as large. All my specimens are disarticulated ; as single bones they
are very perfect; and many of the external ones exhibit most beau-
tifully the sculpturing peculiar to the genus. Of the bones forming
the head are found the hyomandibular and palato-quadrate jugular
plates, ramus of lower jaw with teeth, opercula and frontal bones,
bones forming the pectoral arch, and large interspinous bones con-
necting the unossified vertebree with the dorsal fin-rays. A number
of bones presenting very much the appearance of vertebra, but of
a more cruciform shape, were doubtless the ossified centres from
which were suspended the branchiostegal rays ; some examples show
the rays in situ. Fragments of the rays exhibit ossified processes
developed along the two flattened edges of the ray, which possibly
served to support the gills. The ossified walls of the air-bladder are
occasionally found; one specimen measures 6 inches in length by 3
in breadth.

34. Crenopus Eetrrpricus, Hane. & Atthey.
Rare. Two or three teeth, a few head-bones, and ribs.

35. Crenopus TUBERCULATUS, Hance. & Atthey.
Teeth.

LABYRINTHODONTS.

36. Vertebre and other bones of Labyrinthodonts, recognized by L.
C. Miall, Esq., as those of Lowomma, have been found. They are
rare.

Discussion.

Sir Partie Grey Eorrton remarked upon the advantage accruing
to geological progress by the activity of local observers, as furnished
by the detection and thorough working-out of this bone-bed, which,
being only half an inch thick, would in all probability have entirely
evaded the notice of geologists not resident in the immediate dis-
trict. Local observers (he did not mean mere collectors, but men
who knew how to follow out a course of investigation indicated by
the local phenomena noticed by them) ought in every way to be
encouraged. He said that Onchus, as now restricted, is essentially
an Old-Red type, but extends up into the Carboniferous Limestone,

340 J. W. DAVIS ON A BONE-BED IN THE LOWER COAL-MEASURES.

and even into the Coal-measures, from which Agassiz described one
species under the name of O. subulatus. This will in all probability
prove to belong to Mr. Davis’s new genus Hoplonchus. He approved
of the author’s proposed union of the two supposed species of
Gyracanthus, and agreed with him in his determination of the
Ctenacanths. To show the generality and wide distribution of the
characteristic forms of this formation, he mentioned that M. De-
walque had recently furnished him with examples of several of the
species from Central Russia.

Mr. Eruerrper remarked that the relation of the Bone-bed to the
Better-bed coal gave it a perfectly definite geological horizon.

P. M. DUNCAN ON SOME FOSSIL REEF-BUILDING CORALS, 341

37. On some Fosstn Renr-surrpine Corats from the TERTIARY
Deposrts of Tasmanta. By Professor P. Martin Duncan, M.B.
Lond., F.R.S., &c., President. (Read May 10, 1876.)

[Puate XXII.]

A Description of a very remarkable species of Dendrophyllia from
the Tertiary deposits of Table Cape, in North Tasmania, was read
before this Society on June 9, 1875*; and shortly afterwards I
received a parcel of other kinds of corals from the same locality,
accompanied by a request from the Royal Society of Tasmania that
I would undertake their examination. I was then made aware,
from an abstract of a paper read before the Royal Society of
Tasmania, that all these corals had been under the careful hands of
the Rev. Julian Woods, to whom the paleontology of the Australian
province is so much indebted.

The Rev. Mr. Woods gave his reasons for believing the Table-
Cape deposits to be of the same Lower Cainozoic age as that which
I had given them, and supported his opinions by references to the
similarity and identity of the species of Echinodermata, Mollusca,
and Corals found in them and in the Lower Cainozoic deposits of the
mainland. As my inferences were derived second-hand from Mr.
Woods, he clearly has the priority of haying decided the geological
position of the Table-Cape beds. He stated that, after a comparison
of the Tasmanian and Australian specimens, he found in the deposits
of both countries such well-known forms as Hemipatagus Forbesi,
Woods & Dunce., Cellepora gambierensis, Pectunculus laticostatus, Cu-
cullea concamerata, Dentalium Hicksit, Trigonia semiundulata, Cor-
bula sulcata, Cyprea eximia, Voluta Hannafordir, Voluta antisca-
laris, Conotrochus M’ Coy, and a large Placotrochus deltoideus. This
is a fauna which is characteristic of the Muddy-creek series in
Hamilton, Victoria, and partly of the Mount-Gambier limestone,
deposits which are low down in the Australian Cainozoic series (Mio-
cene of some geologists).

In noticing one of the corals the description of which forms part
of this communication, Mr. Woods considered it to belong to the
genus Jsastrea, explaining that it was new to science, and that its
presence indicated that there was evidence of a deeper sea and
warmer climate than now existed on the area.

The examination of the coral in question and of some other spe-
cimens from the same locality, whilst it necessitates the rejection of
the generic nature of Mr. Woods’s species and of its bathymetry,
quite confirms his opinion regarding the former climate of Tasmania.

The specimens about to be described are included in the group of
compound Astreida, called the Astreeacese by Milne-~Edwards and
Jules Haime, and belong to the genera Heliastrea and Thamnastrea.

Heliastrea, a large genus, culminated during the Miocene, and

* Quart. Journ. Geol. Soc. vol. xxxi. p. 673.

342 P., M. DUNCAN ON SOME FOSSIL REEF-BUILDING CORALS

was, and is at the present day, a reef-building and not a deep-sea
group. Thamnastrea, so common in the Jurassic ages, was then a
reef-builder and a littoral form, and after a great number of species
had been evolved, it became rare in the Nummulitic period, and
died out in the subsequent geological age in the Australian region,
having been probably destroyed in the European areas by the
changes which ensued upon the destruction of the Eocene reefs.
List of species from Table Cape, Tasmania :—

1. Dendrophyllia epithecata, nobis *.
2. Heliastrzea tasmaniensis, sp. nov.
3. Thamnastrea sera, sp. nov.

Genus HELIASTRZNA,

HELIASTR#A TASMANIENSIS, sp. nov. Plate XXII. figs. 1-3.

The corallum is incrusting, and the corallites are subcylindrical
and distant. The calices project but slightly, and differ much in
size; the fossa varies in depth, being in some instances slight and
in others more than equal to the diameter of the calyx; and the
margin is thin. The septa are alternately long and short, are
straight and thin, and are marked with separate granules or with
linear groups of them. They are not exsert, and are only slightly
thicker at the wall than elsewhere: their arrangement is irregular ;
for in the largest corallites there are four cycles in six systems, and
a few members of the fifth cycle in one or two also, whilst in the
majority the fourth cycle is incomplete, there being three cycles in
six systems, and in one or two members of the fourth. In some
ealices the quaternary arrangement of the septal numbers exists,
there being only four primaries and four cycles in four systems.
The distinction between the primary and secondary septa is slight ;
but the existence of a very small septum between two much larger
ones is very decided. The large septa reach far inwards and are entire.
The columella is very small, and is formed of trabeculee between the
septal ends. The wall is thin, and in some instances more so than
are the larger septa. The endotheca is largely developed, and the
dissepiments are thin, long, and curved downwards.

The coste of the large septa are well developed; and they are
either long and wavy over the coenenchyma, or are short and re-
stricted to the corallite; those of the small septa are rudimentary,
and exist either as faint projecting lines or as spines; and their pre-
sence between the more prominent ones is very marked; they are
all thin and delicate. The exotheca is greatly developed and is
largely cellular, the direction of the upper parts of the cells being
nearly horizontal; but all parts of it, including the vertical partitions,
are thin.

The diameter of the calices is under + inch, and their distance
apart is rather more or slightly less.

The species is remarkable for the tenuity of the whole of the struc-

* Quart. Journ. Geol. Soc. vol. xxxi. p. 677.

FROM THE TERTIARY DEPOSITS OF TASMANIA. 343

tures of the corallum, the great development of the exotheca, the
long filiform continuations of the greater coste which connect the
corallites, and their occasional absence, and also for the small
columella and indefinite septal number. The presence of corallites
with the quaternary arrangement is very suggestive.

Alliances. The species is eminently Solenastreean in its aspect;
and were it not for the costal development, it would be associated
with that genus. In some parts of the corallum the distinction
between the structures and peculiarities of the genera Heliastraa
and Solenastrea are by no means decided; but in others the long
wavy coste mix In and amongst the exotheca, and unite with those
of the neighbouring calices. The alliances of this Heliastraan
must be sought amongst those with long costa, much exotheca, a fair
amount of endotheca, and a small columella. Species thus distin-
guished are not found amongst the recent reef-building coral faunas.
The nearest allies of the new species have been found in the same
strata as the congeners of the Australian Placotrochi, namely in the
Miocene of the West Indies; but in estimating the value of this
remote connexion, it must be remembered that there is hardly any
paleontological evidence relating to the Tertiary reef-corals of the
Pacitic area.

The species differs much from the Heliastrwa I described from
the Tertiaries of Java*.

Heliastrea immersa, Reuss. tf, from the older Tertiaries of Monte
Grumi, near Castelgomberto, is of the same general type as the new
form, as it has distant corallites, a peculiar septal number, long
wavy coste, and apparently a thin wall and a small columella.

The new species, when compared with the Heliastreans of the
Ootatoor group of the Cretaceous formation of Southern India, pre-
sents remarkable affinities with Helvastrea cortica, Stol.t; for this
interesting Indian reef-coral has all the characters of the Tas-
manian form, but differs in possessing thick coste, those of the
small intermediate septa being large. Even the other species from
these Indian rocks (Heliastrea rotunda, Stol.) is not without its
resemblance to the Tasmanian species.

THAMNASTRHA SERA, sp. nov. Plate XXII. figs. 4—6.

The corallum is solid and short, being about 5, inch in height ;
and the base is incrusting; the upper surface is subplane ; and the
corallites are distant and widely apart. The calices are slightly
depressed below the level of the long and numerous cost; their
fossula is small and shallow ; and the columella is papillary, being
formed by oblique and rounded processes from the free ends of the
septa. The septa are continuous with the coste, and are at the mar-
gin of the fossula about twenty-six in number; they are subequal,

* Quart. Journ. Geol. Soc. vol. xx. 1864, p. 72.
t+ Reuss. Pal. Stud. tber die alt. Tertiar. der Alpen, 1868, 1. Abtheil,
. 30. :
P t Stoliczka, Paleeontol. Indica, vol, 14, 4 ser. pl. viii. f. 4-8; The Corals or An-
thozoa, Mem. Geol. Survey of India, 1873.

344 P. M. DUNCAN ON SOME FOSSIL REEF-BUILDING CORALS

and their upper edge, finely dentate, dips down very slightly towards
the columella. Some septa are continued on to long and wavy
coste, which become thickened halfway between the calices; and
others soon unite to two or even more cost, or simply bifurcate in
their path towards the nearest corallites. Occasionally there are
three coste united to one septum. The costx are long, slightly
wavy, slightly exsert, narrow, except midway, and are dentate at
their free edge. The small spiny dentations, when removed, leave open
pits on the surface of the coste, which give them a very characteristic
appearance. The cost are also finely and distinctly granular late-
rally ; but the tips of the grains do not meet. There are eighty costa
in connexion with the calicular septa in the largest corallite.

The sides of the septa are marked with closely packed, large, but
flat granules, as are those of the cost (as seen in sections) ; and the
endotheca is largely developed, being close, and curved both upwards
and downwards. It reaches to the columellary space, across it, and
high up in it.

The wall appears to be rudimentary; and the cost are united by
exotheca and by a growth from their sides. Synapticule barely
exist ; for it is very rare to find a few of the lateral granules of the
costz and septa attached by their ends.

Diameter of largest calice =5; inch.

Alliances. This is a well-marked species, and has large calices,
long dentate costa, and largely developed endotheca and exotheca,
the synapticular element being very small. It is the only form of
the genus which has hitherto been discovered in deposits later in
age than the Nummulitic; and it does not closely resemble those
from that series*. Very Jurassic in its appearance, the coral would
almost pass for Thamnastrea Walcott, nobis}, from the Inferior
Oolite of England; and it has no alliance, except that of a generic
nature, with the species from the Indian Cretaceous rocks.

THAMNASTRAA, species.

A much-worn specimen of a Thamnastrean, greatly resenbling
and probably identical with the last-described species, was cut, and
microscopical sections were made. One section passed through
several corallites at a slight distance from the surface ; and another
was taken in a longitudinal direction and parallel with the septa.
A wall separating the corallites is not to be found, and the coste
are continuous with the septa of different calices, the lamin being
bound together by their sides. The septa and dissepiments, es-
pecially the former, are marked by opaque and either rounded or
more or less elongate spots, which apparently coincide with the
large granules which constitute the beauty of the ornamentation of
the septa. These granules are very numerous and evidently are
_of more importance than simple ornaments; for the microscope
shows them to be centres of sclerenchymatous spicular growth.
They are probably ill-developed synapticule.

* Consult D’Achiardi, Coralli Eocenici del Friuli, 1875.
+ Pal. Soc. Supplement to Brit. Foss. Corals, Oolitic part.

FROM THE TERTIARY DEPOSITS OF TASMANIA. 345

The sclerenchyma, as a whole, may be said to consist of spicules
radiating from centres in more or less linear series; and each spicule
is joined laterally and before and behind to its fellows. The whole
coral is infested with the tube-like penetrations and ramifications of
a parasitic unicellular Alga; and this has been described in a former
communication (see Quart. Journ. Geol. Soc. vol. xxxii. p. 205),

Remarks on the Species.

The Heliastrzean just described was evidently a rapid grower, and
a true reef-building form, having its bathymetrical distribution
restricted to 20 fathoms; and the Thamnastrzean (so solid, yet so
abundantly supplied with endothecal structures) appears to have
had a corresponding habitat. They required the external conditions
peculiar to coral-reefs.

In considering how these physical conditions could be found on
the area from which their absence is now so conspicuous, the general
physical geography of the Australian seas during the Cainozoic
periods must be considered. This was attempted to be ex-
plained in 1870 (Quart. Journ. Geol. Soc. vol. xxvi. p. 284), by the
author of this communication; and the relation of the past
and existing coral-faunas of Southern Australia, and the former
distribution of land and sea, were noticed in the concluding parts of
his essay on the fossil corals of the Australian Tertiary deposits.
As there is no reason for altering the opinions therein expressed,
reference must be made to that communication ; and the bearings of
the suggestions therein contained will be found to be explanatory
of the existence of reefsin Tasmania. The open sea of the vast area
to the west of Cape Howe, running up into the tropics, would place
the old hills of Tasmania and those of Eastern Australia in the midst
of an ocean, even if the Australian land were prolonged away to the
north-east, as it probably was in those days. But even admitting
that probable insular distribution of land in the South Pacific during
the Miocene epoch which has been so frequently suggested by
zoologists, botanists, and geologists, the fact of the sea-temperature
being sufficient for the development of reefs in Tasmania is insuffi-
ciently explained.

There is a faint relic of the old reef-fauna still lingering on the
Tasmanian shores in the form of Echinopora rosularia, Linn. It
forms thin incrusting layers, and not masses of limestone; but it is
clearly a relic of that reef-building coral-fauna which has long since
died off from the area. Even the hardiest of the Pacific corals, the
stronger forms of Porites, are absent. In fact, the only surface-
water coral lingers on in a temperature in which Porites could
neither exist nor propagate.

Evidently the reefs around Tasmania, now long extinct, existed
amidst all the physical conditions peculiar to coral-growth on 2
large scale. Pure sea-water in rapid movement and having a
temperature of not less than 74°Fahrenheit, was as necessary to
them as it is to those far away to the north and north-east at
the present day.

346 P. M. DUNCAN ON SOME FOSSIL REEF-BUILDING CORALS

The coral-isotherm would have to be 15° of latitude south of its
present position in order that reefs should flourish south of Cape
Howe ; and this could only be produced by a different distribution of
land and sea, and by a different position of the polar axis to that
which now prevails.

A south-polar continent reaching, in the Miocene age, northwards
to 50° S. lat., would meet the requirement of a land reflecting the
equatorial currents and tides and adding in no way any great
amount of cold water ; and large islands off its coast extending to the
American and African coast would satisfy most of the biological
requirements of the period. To the north and north-east there
existed then the land whose memorials are the atolls and fringing
reefs of the great Pacifie. On the other hand there was the open
sea, already noticed, of Central Australia, and to the north the
great volcanic islands were sea-floors on which Miocene detritus was
accumulating. ‘ Lemuria’ may be assumed to have existed.

Vast surfaces of South and Central America, and some of North
America, were sea-floors; and the ocean and coral-tracts prevailed
over a great space in Europe and Western Asia. Much of North
Africa was still submerged. On the other hand there was probably an
Atlantis, and a huge continent existed in the north of Central Asia,
N. America, and N. and Central Europe. In those days and before,
even during the Nummulitic period, the coral-isotherm of 74° F.
reached fully 25° N. of its present possible position in the portion of
the globe antipodean to Tasmania; and the winter’s cold could not
have been sufficient to chill the surface-water in the latitude of
Vienna and N. Italy. The question arises,—Could the temperature
of this broad belt of warm water be maintained by geographical
causes alone ? and a second question requires a satisfactory solution,
bearing as it does on the question of alternations of season and of
light and darkness,—Could such geographical causes as the distri-
bution of the land in polar masses and in central islands overcome,
in high latitudes, the effects of the position of the globe in relation
to the sun in perihelion and aphelion? It appears to me that they
might to a certain extent modify the severity of climate, but not
sufficiently to permit of important reefs existing in Western, Central,
and Southern Europe, and in Tasmania synchronously. That the
reefs were contemporaneous, there is every reason to believe.

An examination of the flora which underlies the marine Cainozoic
deposits of the mainland of Victoria has shown that the plants found
there resemble those of Tropical rather than of Extra-tropical Aus-
tralia; and as will be noticed in a communication about to be pre-
sented to the Society, the Echinodermata of the succeeding strata
afford the same kind of evidence.

The proofs of the existence of higher temperatures, and of dis
absence of the long and dark months in the Arctic regions, are

abundant from the age of the Carboniferous formation to the Mio-
cene. During all the ages of the globe down to that period when
such vast alterations in the relative level of land and sea occurred,
there is found evidence in fayour of the theory that perpetual frost

FROM THE TERTIARY DEPOSITS OF TASMANTA. 347

and snow and months of darkness did not exist in the Arctic regions.
Contemporaneously a corresponding genial climate existed in the
southern hemisphere. It is perfectly certain that, under the existing
astronomical arrangements, there must be prolonged day and pro-
longed night at the poles and for a certain number of degrees in the
equatorial direction ; and the inclination of the polar axis to the
plane of the ecliptic necessitates the perpetuation of the present arctic
and antarctic climates. The astronomical conditions under which
a sufficient amount of light could be given to the plants within twenty
degrees of the pole are not those which now prevail; but were the polar
axis at right angles to the plane of the ecliptic, and were there no
greater node than at present, there would be equal day and equal
night. The biologist claims this as the earliest position of the globe.
The arguments against such an inference relate, of course, to the
nature of the forces which could bring down the north pole, or rather
which would incline the polar axis, and to the amount of solar heat
which would reach the polar regions. Probably no force acting on
the globe as a mass, such as gravitational energy, could alter the
position of the poles as they are now, with reference to the ecliptic ;
for after a very slight inclination had been produced, the force
would produce no more obliquity, but only a more or less rapid pre-
cessional movement.

Nor is it possible to understand how any external force could cause
the approach of one pole to the sun and the recession of the other,
the globe being comparatively homogeneous, for the sake of argu-
ment, and the polar axis being supposed to be at right angles to the
plane of the ecliptic.

But it would be possible if a vast alteration in the relative distri-
bution of land and sea occurred, in one hemisphere especially. It
is perfectly reasonable to infer that the great subsidences of the
Miocene lands, and the formation of the Southern Ocean, whose
area is greater than that of all the land to the north, and the vast
upheaval of the Central-Asian, Caucasian, and Alpine and other
areas, producing great alterations in the homogeneous condition,
brought the land-surfaces of the north with their higher specific gra-
vity and great mass within the influence of the gravitational energy
of the sun. From that time dated the long winter’s night and the
presence of perpetual ice and frost in the highest latitudes; and
those changes in the climatal conditions of the northern and southern
areas, where reef-corals had built and the light-requiring floras had
flourished. The objection regarding the small amount of heat
which would be granted to the high latitudes under the conditions of
a vertical polar axis are of the nature of those which sadly troubled
our science with respect to the impossibility of animals living at
very great depths in the sea in consequence of the pressure to
which they would have to submit, and to the warm temperatures
which must prevail in the oceanic abyss.

The presence of land in the extreme north or south on which no
perpetual ice rested except on the high hills, would introduce an ele-
ment into the argument which would suffice to demonstrate a tem-

=e. 3.6.8. No. 127. 2B

348 P. M. DUNCAN ON SOME FOSSIL REEF-BUILDING CORALS

perate zone. Oblique as would be the path of many solar rays, still
the corresponding loss of temperature would be compensated for by
the warming of the atmosphere by the radiation from. the masses
of the land-surfaces to the north and south. It is, moreover, reason-
able, according to the principles of thermodynamics, to assert that
the sun was then producing more heat, and that the internal tem-
perature of the globe was greater than now; and this may have had
some slight influence.

In bringing these theories before you, which have been in the
minds of so many geologists, and which have been brought forward
by succeeding generations of us, Belt and Woodward being their
last supporters, my excuse must be that possibly they may be im-
proved upon by those physicists who will admit the necessity of
comprehending biology in their dynamical and kinetic arguments,
—or that in the true interests of science they may be disproved, so
that we may seek explanations of the facts brought forward in this
and other cognate essays in other directions.

Note.—This communication was forwarded to the Society before
my predecessor in the Presidential office read his admirable Address.
I now refer the reader to his remarks on those questions of astro-
nomical interest which I have introduced in this paper. (See
Anniversary Address of the President, John Evans, Esq., F.R.S.,
Quart. Journ. Geol. Soc. vol. xxxil. Proc. p- 101 et seq.)

EXPLANATION OF PLATE XXII.

Fig. 1. Corallum of Heliastrea tasmaniensis, sp. nov.

1
2. Calice, magnified 3 diameters.

3. Costz and exotheca, x 3.

4, Calices of Zhamnastrea sera, sp. nov.
5. Longitudinal section of cora lum.

6. Details of septa, x 3.

Discussion *.

Mr. Evans was glad to find that this subject, concerning which he
had’ lately expressed his own views, had been taken up by the
author; but he thought it possible that Dr. Duncan would, on
further consideration, be inclined to modify somewhat the theory
promulgated in this paper in favour of some other view. In order
to account for the occurrence of reef-building corals of Miocene
age in latitudes now too cold for them, the author had reverted to
the oldjidea of the vertical position of the poles. If the interior of
the earth is fluid, a sliding crust such as the speaker had formerly
suggested is possible, though it would be difficult to prove the ex-
istence of a fluid interior, and still more difficult, did that exist, to
prove that the crust would slide on it. But even supposing the
earth to be a nearly solid body, elevations and depressions enough

* This discussion relates also to Prof. Duncan’ s paper “ On the Echinoder-
mata of the Australian Cainozoic Deposits.”

FROM THE TERTIARY DEPOSITS OF TASMANIA. 349

must take place on the surface to alter the relative positions of the
poles with regard to the surface of the earth. Because there were
proofs of warmer climate having existed in Miocene times in Green-
land, near the one pole, and in New Zealand near the other, there
was no need to suppose that belts of warmer temperature had ex-
tended nearer the poles than at present, for the same sliding of the
crust that brought Greenland nearer the equator would also bring
New Zealand nearer the tropics, both being on nearly the same
meridian, but on opposite sides of the globe. The subject was one
that deserved the attention of geologists, as it lay at the root of
many important questions affecting the past history of the earth.

Professor Hvuenxs believed implicitly what the astronomers told
him must be; and if observations on the distribution of life neces-
sitated any thing more than such alterations of climate as could be
accounted for by geographical changes and modification and adapta-
bility in the forms of lite, he would prefer to leave it as one of the
many things he could not explain, than accept an explanation in-
consistent with accepted astronomical theories. If, as explained by
Sir John Herschel, the transference of large masses from one part of
the earth’s surface to another would disturb the equilibrium, we
must remember that this action would be mostly compensative ; and
if the cumulative effect of many such disturbances might be a partial
readjustment of the mass, we must regard such movements only as
a tendency to keep the whole mass and its axis of rotation as it was
in spite of the transference of portions from one place to another
by denudation. Moreover he disputed the data on which the views
advocated by both the present and the late President were founded.
He asked whether we should say that the climate of the period of
our older river-gravels was that of Egypt or of Northern Siberia,
seeing that the Corbicula fluminalis and Unio littoralis were now
found only much further south ; while the hairy elephant and rein-
deer, which had once lived with them, were now held to prove an
arctic climate. When we know that flowering plants and ever-
greens now live in Alpine regions, where they are buried in total
darkness under snow for four months, shall we say that the absence
of light would render it impossible for evergreens and flowers to
have flourished where the arctic winter-night is four months long,
even though we could account for a milder climate by geographical
changes.

Mr. Woopwarp remarked that as it was not merely a question of
one fauna and flora, Mr. Hughes’s statements must be received with
caution. There were evidences in northern latitudes, not only of a
Miocene, but also of a Carboniferous, a Jurassic, and a Cretaceous flora.
Nor was ita question merely of lowly organized plants which would
be more likely to withstand the climate; for Prof. Nordenskidld had
found tree-trunks standing erect in the soil in which they grew, and
it was impossible for them to have grown in a climate so rigorous as
now exists at that latitude. If the geologists are wrong in the con-
clusions they have drawn from these facts, let the astronomers show

2Bn2

350 P, M. DUNCAN ON SOME FOSSIL REEF-BUILDING CORALS

them how to account for the occurrence of fossils indicative of such
a warm climate in such high northern latitudes ; for the absence of
cold must be accounted for to explain the growth of these trees in
that spot.

Prof. ANstED maintained that the geologists had certain natural-
history facts on their side with regard to the occurrence of fossils
near the poles; it remained, therefore, for the astronomers and
physicists to find a new theory to account for these facts.

Prof. Grrrn thought that the astronomers should be asked if the
change of axis was a possible explanation, and to calculate what
would be the result of a change in the distribution of land and sea,
and how wonld such a change affect the position of the poles. The
question was one of mechanics.

Mr. Sorsy considered that the amount of heat and light received
from the sun should also be taken into account, and the fact that
this may have varied at different periods.

Sir Antonio Brapy stated that there were many facts which
tended to prove that the sun had varied in heat &c.; but the sun
had probably little to do with the warmer climate of the poles in
past ages. The heat of the earth in its various stages of cooling
would be sufficient to account for these changes of climate on the
surface of the globe.

Prof. Ramsay could not agree with the last speaker in thinking
that radiation in cooling would produce any palpable effect on the
surface of the globe. So far from there being any proof that the
climate had been gradually growing colder from the earliest times
down to the present date, there was every evidence to show that
glacial periods had recurred at different periods in past time. Dr.
Duncan and Mr. Evans had merely given suggestions, but had not
attempted to solve the problem. The poles probably occupied the
same position in Miocene times that they do to-day. Darwin and
Dana were both agreed in thinking the present continents to be of
extreme antiquity. Great elevations of land had taken place prior |
to the Miocene epoch. The Alps and the Himalayas were both
pre-Miocene, and were probably higher in pre-Miocene times than
at present, haying been subjected to great denudation.

Mr. Gwyn Jerrreys pointed out that certain species of shallow-
water mollusca now found in the Arctic Ocean had formerly in
post-Tertiary times lived as far south as Sicily.

Dr. Wrieut remarked that there was a wonderful similarity be-
tween the Miocene echinoderms from Australia and those found at
Malta.

Prof. Morris considered the abundance of echinoderms belonging
to the Spatangoid group in these Australian beds to be very inter-
esting. The feature presents itself in the New-Zealand Tertiaries,
where forms allied to Arachnoides occur. The distribution of these
echinoderms in New Zealand was excessively complex and difficult
to understand. There was a remarkable similarity between the
Miocene floras of Greenland and Central Europe; and the question
to be asked was, Did they spread over a continent formerly existing

Quart.Journ.Geol.Soc.Vol. XXXII. Pl]. XXII

Foord ith. M&N.Hanhart imp.
TASMANIAN CORALS.

FROM THE TERTIARY DEPOSITS OF TASMANIA. 351

between these points, or had they emigrated from some one central
spot ?

a. Duncan, in reply, stated that it was only by the united inves-
tigations of all students of geology that the question could be in
any way settled. ‘The belief in the recurrence of glacial epochs was
founded on some erroneous conclusions drawn from beds in England,
South Africa, and in India, which were related to local glacier action
or to volcanic agglomerates. The Miocene plants could not have
existed without sufficient light; and severe frost would have destroyed
them ; and therefore they could not have extended so far north under
conditions similar to the present. The Echinoderms did certainly
present a striking resemblance to those found in the Miocene beds
of Malta; but there were still sufficient specific and generic differ-
ences to justify him in describing them as distinct.

352 PROF, OWEN ON EVIDENCES OF THERIODONTS IN PERMIAN

38. Evipences of THERIODONTS in Permian Dupostts elsewhere than
Sovrn Arrica. By Prof. Owzn, C.B., F.R.S., F.G.8. (Read
May 24, 1876.)

A Few days ago Mr. W. Davies brought to my notice the cast of a
fossil which had been purchased by the British Museum in 1865 of
Krantz, the dealer, with the label “ Hurosaurus uralensis, H. v.
Meyer, Brithopus priscus, Kutorga. Permischer Sandstein von
Perm.”

The subject of this cast is that of Kutorga’s plate i., and is, as
he truly describes it, the lower end of a humerus, with the perfora-
tion or canal above the inner condyle, on which character he mainly
rests his determination of the fossil as evidence of an extinct mam-
mal and as the basis of his genus and species Brithopus priscus *.

The discovery of that character in the humerus of Cynodraco +
and other genera of South African reptilia, led to the rescue of
the cast in question from the obscurity in which it had remained
since its acquisition, and to the present retrospect of the circum-
stances under which Kutorga’s contribution to Permian paleontology
has been thrown into the background, and his supposed mamma-
lian genera systematically ignored.

Kutorga, accepting the evidences summed up, in 1831 +, of the
mammalian “nature of Thylacothertum and Phascolotheriwm, and
the proofs of the Oolitic age of their matrix, saw no geological
objection to the remains of that class being discovered in the Per-
mian deposits.

With respect to the most conspicuous test-character of the fossil
humerus §, he truly states that such character was known only (in
1838) in that bone of certain unguiculate mammals; and he quotes
the instances given by Cuvier ||, out of which instances, by reason
of the breadth of the condylar part of his fossil, and the high posi-
tion of the canal, he adopts the Edentate genera as those to which
his Brithopus was most nearly allied, and proposes to place his
extinct mammalian genus between Bradypus and Dasypus 4].

* Beitrag zur Kenntniss der organischen Ueberreste des Kupfersandsteins
am westlichen Abhange des Urals, von Dr. Stephan Kutorga, &c., mit vii.
Steindrucktafeln (8vo, St. Petersburg, 1838), p. 9.

Tt Quart. Journ. Geol. Soc. vol. xxxii. p. 95, pl. xi. figs. 6-9.

t Buckland, Bridgewater Treatise, vol. i.: supplemented by papers in Pro-
ceedings of the Geological Society of London, December, 1838, vol. iii. p. 17:
and Transactions of the Geological Society, 2nd Series, vol. vi. p. 58.

§ “Das am meisten im Auge fallende Kennzeichen dieses Knochens, ist das
grosse ovale, itber dem ‘ Condylus internus’ stehende Loch, welches zum Durch-
gange der Arteria ulnaris diente.”— Op. cit. p. 10.

|| Legons d’Anat. Comp. ed. 1835, vol. i. p. 384.

“| “Aus allem oben Gesagten ergiebt sich, dass das Thier, dessen Oberarm-
knochentheil wir vor uns haben, zur Ordnung der Edentaten gehorte, und zwar
zu einer neuen, zwischen Bradypus und Dasypus zu stellenden, jedoch mehr mit
der ersten verwandten Gattung.”—J2. p. 13.

DEPOSITS ELSEWHERE THAN IN SOUTH AFRICA. 353

For the generic distinctness of Brithopus he specially urges the
presence of a second perforation or canal, near the upper part or
beginning of the supinator crest, which canal he calls the foramen
condyloideum externum, and conceives that it may have served for the
passage of the arteria radialis *. This canal, Kutorga states, had
not been previously observed in the humerus of any vertebrate ani-
mal; and at that date (1838) the statement was true.

I may not be mistaken in supposing that the reduced copy of
Kutorga’s figure of the above-defined Permian fossil will be accepta-
ble, to which I have added (Cut, fig. 1) letters indicative of the

Fig. 1. Brithopus priscus, Kut. Fig. 2. Orthopus primevus, Kut.

same parts as in figs. 6-11, pl. xi., Quart. Journ. Geol. Soc. vol.
xxxli. The significance of the foramen marked m will thus be
appreciated.

Kutorga finds evidence of a second mammalian genus in another
fossil from the same locality and formation, which he describes and,
fortunately, figures in the same treatise ft. Conceiving this fossil to
be the lower portion of a humerus, with also the characters of the
foramen condyloideum internum, heindicates differences which support
his conclusion that it formed part of a distinct mammalian genus
and species, for which he proposes the name Orthopus primevus.

* «Die Basis dieses Kammes ist schrage von oben nach unten mit einem
cylindrischen Kanale durchbohrt, der in seinen beiden Enden in flachorale
Gruben ibergeht, und eine Arterie, wahrscheinlich die arteria radialis, im
oberen Theile ihres Verlaufes beherbergte.”—J0. p. 11.

t Op. cit. p. 15, Taf. ii.

354 PROF: OWEN ON EVIDENCES OF THERIODONTS IN PERMIAN

This fossil, however, is the upper (proximal) portion of a reptilian
humerus, and most probably of the same species, if not bone, as
that on which the genus Brithopus is founded. Figured upside
down, as in Kutorga’s plate i1., the articular head of the humerus,
(a, 6 in Out, fig. 2) is made the “trochlea;” the entotuberosity
(ib. c) becomes the ‘‘ condylus internus ;” the ectotuberosity is the
“condylus externus;” the delto-pectoral plate (ib. b*, 6’) is the
‘‘supinator ridge” (“der tiber diesem Condylus stehende kamm”’);
while the groove at the inner part of the fractured shaft, rightly
referred to as evidence of a “foramen condyloideum internum,” is
the upper (proximal) commencement of that canal, which is shown
in its entirety and with its lower (distal) outlet in what I hold to
be the distal portion (Kutorga’s pl. i.) of the humerus of the same
species or individual (Cut, fig. 1).

The humerus of Orthopus being reversed in position, as in Cut,
fig. 2, and being juxtaposed to fig. 1 (Brithopus), the characters
of the reptilian humerus, under theriodontal modifications, come
out unequivocally ; and the proportions are nearer those of the per-
forated humerus of Galesawrus and Dicynodon than of Cynodraco.

The proximal articular convexity, or head, of the bone (Cut,
fig, 2,a) is a narrow semi-ellipse, its long axis transverse to that of
the shaft, its curved border being due to the usual production of
this articular surface backward, or anconad, in Reptiles. The ento-
- tuberosity, c, is more definitely marked than in Cynodraco. The
ectotuberosity, 6, 1s represented by the angle formed by the out-
standing of the delto-pectoral crest as it descends from the axial line
of the head of the humerus to 6’. The outer border of this large
and characteristic crest forms a low angle about halfway down
at 6+; then forms a second (almost right) angle at 6’, where it
returns to subside on the shaft of the humerus. The forward
(thenal) bend of the crest is not so marked as in Dicynodon; it
does not exceed that shown in the humerus of Cynochampsa and
Cynodraco. ‘The shaft of the humerus between the delto-pectoral
and supinator crests is as short and as contracted, compared with the
proximal and distal expansions of the bone, as in Cynodraco (loc.
cit. pl. xi. figs. 6, 7). The reptilian character of the distal portion
of the Brithopus bone (Cut, fig. 1) is shown by the shallow ill-
defined olecranal depression, by the absence of the definite trochlear
character of the distal articulation for the ulna, and by the presence
of the ectocondylar perforation, m.

There is thus satisfactory evidence of an extinct reptile in the
Permian deposits of the Ural, with a humerus showing the
same mammalian entepicondylar perforation as in certain extinct
reptiles of the Karoo series of South Africa, and with other modifi-
cations in which it more nearly resembles the Theriodont species. of
those deposits.

Brithopus priscus, Kutorga, as represented by the plaster cast,
is superseded by Hurosaurus uralensis, H. v. Meyer; and I next
proceed to give the results of investigation of the ground of this
supercession and synonymy in the printed label of the dealer Krantz.

:

DEPOSITS ELSEWHERE THAN IN SOUTH AFRICA. 355

The earliest ascription of the generic term Hurosaurus to Permian
fossils I find in a paper by Fischer von Waldheim, in tome xy.
1842, of the ‘ Bulletin de la Société Impériale des Naturalistes de
Moscou’*. Fischer there states that he has several portions of
bone (not named) belonging to the same Saurian. A striking cha-
racter, he remarks, is the great breadth of the articular head com-
pared with the insignificant length of the bone, ‘as if,’ he says,
“that articular end (‘ Gelenkkopf’) of the bone had been provided
with an aliform process ;” and he proposes for the new Saurian, so
indicated, the name of Hurosaurus. No other characters of the
genus nor any figure of the fossils are given. We recognize, how-
ever, in this curt notice, a characteristic of the proximal portion of
the humerus of Kutorga’s Orthopus.

The author next cites a ‘“‘remarkable humerus distinguished by
its broad processes and the perforated articular surface for the
reception of the ulna, from the same locality as that in which the
mandible of the reptile called Rhopalodon was found ;” and this
humerus, the perforated character of which is perhaps more accu-
rately described by Kutorga in his Brithopus, Fischer thinks may
belong to his genus Rhopalodon +.

Itis significant, and probably by reason of the insufficient grounds
assigned by Fischer for his genus Lurosaurus of 1842, that Agassiz
| has omitted it in the list of Reptilian genera in his ‘ Nomenclator
| Zoologicus,’ 4to, 1842, and in the ‘ Addenda,’ 1846. He gives
| amongst the mammalian genera Kutorga’s Brithopus and Orthopus,
: but adds ‘“ Reptilia?” after each entry.
| Fischer’s Rhopalodon Wangenheimii was founded on a portion of a
| mandibular ramus, 2+ inches in length, containing “‘ nine molars
with short, thick, pointed, subcompressed crowns having two tren-
chant and serrate borders ;” and, further, “ at the anterior fractured
part of the fossil was the implanted base of a large canine.” A
figure of the fossil is given by Eichwald in the ‘Lethea Rossica,’
1860, pl. lvii. fig. 9. So far as the characters are noted they
are “ theriodont.”

In 1858 Hermann von Meyer described a fossil skull, presented
by Major von Qualen to the ‘Mineralien Kabinet’ at Berlin, as
from the Permian system of West Ural, probably from the same

locality as the fossils described by Fischer. A skull from this
locality had previously been described by Eichwald under the name
of Zygosaurus lucius; but v. Meyer’s subject is referred by him,
under the name of Melosaurus uralensis, to the section of the Laby-
rinthodonts with embryonal vertebral column. The specimen

j * “Zweiter Nachtrag zu den von Hrn. Major von Qualen am westlichen
Abhange des Urals gesammelten Versteinerungen,” p. 463.
+t ‘Kin merkwirdiger Humerus, ausgezeichnet durch seine breiten Fortsatze.
und die durchbohrte Gelenkflache zur Aufnahme der Ulna, ist-in derselben
Gegend gefunden worden, und scheint dem Rhopalodon anzugehoren.”—
. 464.
: t ‘* Bei diesen Abweichungen habe ich den zweiten Schadel aus den Ural unter
der Benennung Melosaurus uralensis za den Labyrinthodonten mit embryonaler
Wirbelsaule gestellt.”—Newes Jahrbuch fiir Mineralogie, &c., von K. C. von
Leonhard und H, G. Bronn, Jahrgang 1858, p. 301. ;

356 PROF. OWEN ON EVIDENCES OF THERIODONTS IN PERMIAN

is figured, of the natural size, by Eichwald in pl. lvii. fig. 25, op.
cit. In the text of this work the genus Melosaurus of von Meyer,
together with the genera Brithopus and Orthopus of Kutorga, are
made synonyms of the Hurosawrus of Fischer*: no characters of
the Permian humeri are pointed out to show the reptilian nature of
Kutorga’s fossils ; but Kutorga’s determination of their homology is
called in question; they are affirmed to be portions of the coracoid bone
of Hurosaurus; and the perforations in the bone assigned to Britho-
pus are homologized by Eichwald with those in the coracoid of
Ichthyosaurus and Hyleosaurus +.

The three folio plates of Eichwald’s ‘Lethzea Rossica’ are devoted
to the reptilian fossils of the Permian deposits in the Ural. Long
bones and portions of long bones are figured, as of Hurosaurus
uralensis, but give not any indication of correspondence with such
bones from the Karoo beds.

The characters described and figured of the skull of the Melo-
saurus, Vv. M., are Labyrinthodont. The strong mandible and its
canine-like anterior teeth suggest the need of as firm a junction of
the head to the trunk as in the Salamandroides of Jaeger $. Whe-
ther the occiput in the subject of v. Meyer’s description be in such
a state of preservation as to support the conclusion that the basi-
occipital had not advanced beyond its embryonic notochordal state,
may be worthy of fresh scrutiny of the specimen in the “‘ Mineralien-
Kabinet.”” With whatever result, it would give no support to Eich-
wald’s reference of Kutorga’s Brithopus and Orthopus to such
Labyrinthodont genus, or to the synonymy of Melosaurus with
Fischer’s Hurosaurus.

The homology and reptilian characters of the portions of Permian
fossils described and figured by Kutorga being now beyond question,
it is to the Theriodont, not the Labyrinthodont order that such
humerus must be referred. I say “'Theriodont” rather than
“ Anomodont,” because of the indicatien of the more powerful
muscles working the forearm and paw-which is afforded by the
degree of development of the ‘“supinator” or ‘“ ectocondylar” crest
in Brithopus, as in Cynodraco. A minor development differentiates
the humerus in the genera Dicynodon and Oudenodon.

I therefore proceed to seek for other evidences of Theriodontia in
Permian fossils of the Ural.

* Lethxa Rossica, 8vo, 1860, p. 1621.

t “Tos coracoide” de l’Eurosaurus “a été pris! antérieurement pour la
partie inférieure de humerus d’un Mammifére ; il semble plutot se rapporter
a los coracoidien de l’omoplate, qui offre une forme trés-large dans 1’Jchthyo-
saurus et surtout dans le Hyl@osaurus Mantelli.”...“L’os coracoide de l Hyl@o-
saurus est perforé d’un trou artériel et nerveux; le coracoide de l’Hurosaurus
en offre deux, un trou large, percé de chaque cdté de sa partie retrécie par
laquelle passaient des artéres et des nerfs de l’extrémité antérieure.”

“' Kutorga die organ. Ueberreste des Kupfersandsteins, &c. pl. i. fig. 1-3”
(Hichwald, op. cit. p. 1626). ‘“M. Kutorga a décrit un autre os fragmentaire
comme ‘ pars condyloidea humeri’ d’un Mammifére inconnu; je préférerais le
nommer aussi lomoplate, en l’ajoutant a l’os coracoide” (7b. p. 1626).

t Fossile Reptilien im Wiirttemberg, 4to, 1828, p. 38, Taf.’v. figs. 1 & 2.

DEPOSITS ELSEWHERE THAN IN SOUTH AFRICA. 357

So long as such evidence afforded by the arm-bone remained un-
recognized, the third fossil which Kutorga figures (in his Taf. iii. fig.
3, A, B, C, D, E), and describes (p. 19) as the representative of his
Pachydermal genus Syodon, had little chance of being extricated
from the neglect and oblivion into which it had fallen. This fossil
is a recurved subcompressed caniniform tooth, 3 inches in length,
(following the convex curve), 7 lines in antero-posterior basal breadth
4 lines in transverse basal breadth, which diminishes more rapidly
than the antero-posterior one to the point; both fore and hind
margins are trenchant, and need only the fine serrations, which may
have been overlooked, to make the correspondence of this Permian
tooth with the upper canine of Cynodraco instructively close.
Kutorga’s fossil is certainly more like that reptilian tooth than the
tusk of a wild boar or other porcine species.

A more decisive dentary evidence of a Permian Theriodont is
afforded by the Deuterosaurus biarmicus of Eichwald*. This species
is represented by the fore part of both upper and lower jaws, which
are remarkable for the predominance of the vertical over the trans-
verse diameter. The disproportionate development of the canines,
especially of the upper pair, is noted in the generic character :—“ les
canines etaient trés-longues, surtout celles dela machoire supérieure ”
(J. c. p. 1608). I submit a reduced copy of Eichwald’s figures of the
fossils (Cuts, figs. 3 & 4, p. 358), which indicate the following incisive-
canine formula :—i. =”, c. =. The lower canine crosses in front
of the upper one.

One small molar (fig. 4, m) is shown at a short distance behind
the upper canine, and a still smaller one (ib. m’) behind the lower
canine in the postcanine portion of the fragment of skull. All the
teeth have conical acuminate crowns with denticulate or crenulate
trenchant borders. In the incisors the fore-and-aft diameter pre-
vails at the base of the crown chiefly through a posterior protuberance
of that base: in the canines the antero-posterior diameter is greatest
at the base of the crown.

In all the more essential characters there is a close agreement be-
tween Deuterosaurus and Cynodraco; the incisive-formula is the
same (Cut, fig. 5, Cynodraco serridens, from Karoo beds, Fort
Beaufort). In the more compressed form of both upper and lower
jaws, the Lycosaurus of the Karoo beds of the Sneewberg range
(Cut, figs. 6, 7) more resembles the Deuterosaurus of the Ural Per-

mian. But the incisive-formula of Lycosaurus is — The canines
of Deuterosaurus, though large in proportion to the contiguous in-
cisors and molars, appear relatively less than in Lycosaurus or Cy-
nodraco. The postbasal swelling, or partial “cingulum,” in the
incisors of Deuterosaurus, and the greater relative depth of the man-
dibular ramus behind the canines compared with that of the sym-
physial part supporting the incisors, preclude a reference of the Cape
species of Theriodontia to the genus of that reptilian order from the

oe Beal

* Lethza Rossica, p. 1607, tab. lviii. figs. 1, 2, 3.

358 PROF. OWEN ON EVIDENCES OF THERIODONTS IN PERMIAN

Ural. But their ordinal affinity is as satisfactorily manifested by
the maxillo-mandibular and dental characters as by the humeral
ones.

Fig. 3. Fig. 4.

Lycosaurus, Ow. Cynodraco serridens, Ow. Lycosaurus, Ow.

I submit, therefore, as a result of the foregoing analysis of the
accessible records and representatives of the Permian fossils of the

DEPOSITS ELSEWHERE THAN IN SOUTH AFRICA. 359

Ural, that we have an additional and weighty evidence of the palao-
zoic date of the ‘* Karoo series” of South Africa.

I next propose to offer evidence of a Theriodont reptile affined to
Lycosaurus, from a probably Permian deposit in North America.

In a “red-sandstone formation” in Prince-Edward Island, a
portion of the skull of a reptile, including the left maxillary (21),
premaxillary (22), and nasal (15) bones (Cut, fig. 9), was discovered

Fig. 9.

Fig. 8. Lycosaurus, Ow. Fig. 9. Bathygnathus, Leidy.

and has been described and figured by the eminent paleontologist,
Joseph Leidy, M.D., in the ‘Journal of the Academy of Natural
Sciences of Philadelphia ’*.

The teeth, of which seven are preserved, are implanted in distinct
sockets, and have subcompressed, recurved, conical, acuminate crowns,
with anterior and posterior trenchant borders, of which the latter is
minutely crenulatedt. Of these teeth the foremost in the maxillary
bone, protruding close to the maxillo-premaxillary suture, claims by

* Vol. ii. p. 327, plate xxxii. T Tor ib:

360 PROF. OWEN ON EVIDENCES OF THERIODONTS IN PERMIAN

its superiority of size and length of crown to be regarded as a
‘canine ” (Cut, fig. 9, c). It is followed by five much smaller but
similarly shaped teeth, to be classed by position as “molars” (ib. ib.m).
In the premaxillary bone but one incisor has been preserved (ib. 2):
‘‘its anterior border is obtuse and not crenulated, while the posterior
border is acute, but is too much broken to judge whether it is cre-
nulated ; the enamelled crown has been about 9 lines long by 4 lines
in breadth at its base. The fang of the tooth, like that of the other
ones, is oval in section” *. ‘This incisor is separated by a diastema
an inch in extent from the canine. The length of the crown of
the tooth is 1 inch 9 lines, with a breadth of base of 7 lines. “ Its
fang can be seen in the wide fissure of the jaw extending two inches
from the alveolar border ” f.

The foremost molar, with a visible crown 6 lines in length, pro-
trudes at a distance equal to its own antero-posterior breadth from
the canine. The crown in the succeeding molars, which are com-
pletely protruded, is 9 lines inlength. Dr. Leidy describes a minute
crenation of the anterior as well as posterior trenchant borders of
the crown +.

On the hypothesis that a portion of the upper jaw is here de-
scribed, I see in a semicircular notch more than an inch across the
base, at the upper anterior angle of the fossil, evidence of an external
nostril (Cut, fig. 9,n). Dr. Leidy, however, has described and figured
this fossil as “‘ the right dental bone ”’ of the lower jaw. The grounds
of the contrary homology, which I here submit, are afforded by an
almost complete skull of a Lycosaurus, from the Karoo series of 8.
Africa (Cut, fig. 8).

In this, as in other Theriodonts, a well-marked canine (ib. ib. ¢)
divides a series of small incisors (ib. ib. 7) from a similar series of
small molars (ib. ib. m). The diastema between the canine and in-
cisors is rather longer than between the canine and the molars. The
crowns of all the teeth have the same general shape, even to the
‘“minutely crenulated ” hinder border, as in Bathygnathus, Leidy ;
the incisors also show the distinction of the “more obtuse anterior
border.”

Both fossils are remarkable for the depth or vertical extent of the
upper jaw, 21; and, in both, the alveolar border of the maxillary de-
scribes a moderate convex curve. This seems to me to be fatal to
the mandibular hypothesis. I exhibit an impression of the plate
(Ixviii.) of my ‘ Catalogue of S. African Reptilia,’ giving a side view
of the skull of Zycosawrus curvimola, also an impression of Leidy’s
plate xxxiii. of his Bathygnathus borealis (reduced in Cuts 8 & 9),
from which may be better appreciated the comparisons leading
me to refer Bathygnathus to an allied genus of the Theriodont order
of Reptilia.

If the proposed homology and affinity be accepted, then the students
of the South-African fossils have a new interest in the determination of

* Op. cit. p. 328. pellcha! os

{ This character is repeated in certain teeth of Cynodraco_serridens, ‘ Catal.
of 8.-African Reptilia,’ 4to, 1876, pl. xvi. fig. 7.

DEPOSITS ELSEWHERE THAN IN SOUTH AFRICA. 361

the horizon of the Nova-Scotian deposits in which the Bathygnathus
borealis was discovered. On this subject the experienced geologist
Principal Dawson, F.R.S., F.G.S., adds a note to Dr. Leidy’s paper,
a quotation from which may not be unacceptable. ‘‘ The fossil was
found at New London, on the northern side of the island, imbedded
to the depth of 9 feet in red sandstone with calcareous cement. - The
total depth from the surface was 21 feet 9 inches; the discovery
was made when digging a well. The sandstone in question belongs
to a formation which occupies nearly the whole of Prince-Edward
Island, generally dipping at a small angle to the northward. It
includes thin beds of coarse concretionary limestone; and. at the
southern side of the island, where the oldest beds of the formation
appear, there are beds of grey clay or soft shale, and brown and grey
sandstone, containing silicified trunks of coniferous trees, with in-
distinct vegetable impressions, perhaps Calamites. These beds may
either belong to the top of the Carboniferous system or to an over-
lying deposit of the Permian or Triassic age ; and in either case the
red sandstones which conformably overlie them will be equivalent to
the New Red of western Nova Scotia* and Connecticut, and are
probably Triassic or Permian.” ‘To this note there is the following
rider :—‘‘ Mr. Lea states he agrees with Mr. Dawson in supposing
these Red Sandstones to be equivalent to those of Connecticut,” &e. ;
and he further observes, ‘‘ there are many reasons in fayour of re-
ferring them to the superior strata of the Permian series.” How-
soever this question, viz. as between Permian and Trias, may be
decided, no fact is elicited in the foregoig survey indicative of
a Liassic age, and the majority of instances weighs in favour
of a Permian period. If the affinities of the Ural and Prince-
Edward-Island fossils have been correctly determined, we have eyi-
dence already of the wide geographical range (North America,
Europe, South Africa) over which the Reptilia distinguished by
certain mammalian structures extended.

The Saurians which have been referred to the family of Thecodonts,
the remains of which have been obtained from the dolomitic conglo-
merate of Bristol, have not only the teeth implanted in distinct
sockets, but have crowns, conical, compressed, acuminate, with an
anterior and posterior finely serrated edgey. A similar type of
tooth (Cladiodon) has been discovered in the red sandstone of War-
wickshire. These ‘“ thecodonts ” may form a family in the Therio-
dont order, if less fragmentary evidences than we now possess do
not show even a closer similarity in their dentition to the type
genera of the Karoo beds.

The steps leading to the above results may be summed up as
follows :—

To Kutorga belongs the merit of having first pointed out and
figured the humeral characters which he terms “foramen condy-
loideum internum” and “foramen condyloideum externum,” but
under the belief that both were mammalian characteristics. In the

* Op. cit..p, 329. + ‘Odontography,’ 4to, 1840, p. 266.

~ 362 PROF. OWEN ON EVIDENCES OF THERIODONTS IN PERMIAN

‘ Descriptive Catalogue of the Osteological Series contained in the
Museum of the Royal College of Surgeons,’ it was first shown that
the “ foramen condyloideum externum ” was a character of a rep-
tilian humerus*.

Professor Huxley was the first to figure the canal homologous with
Kutorga’s “ foramen condyloideum internum,” under the term “ supra-
condyloid foramen or canal,” in a humerus of a Dicynodon from the
Panchet rocks of India, and in the humerus of a Dicynodon from
the Karoo series of the Cape, in elucidation of which character he
remarks :—‘“ A supracondyloid foramen occurs not unfrequently
among Lacertian Reptiles, but not with the precise form and pro-
portions which it presents in these two humeri” f.

The determination of the homology of the supracondyloid foramen
in the humerus of the Dicynodonts and Theriodonts is given in the
‘ Catalogue of S.-African Reptilia’ f, and in the paper on Cynodraco §.
It illustrates the mutual affinity of the two extinct groups, and their
common possession of a mammalian structure. A like determination
of the homology of Kutorga’s ‘‘ foramen condyloideum externum ”
converts into certainty the suspicion expressed by Agassiz of the
reptilian nature of Brithopus and Orthopus. An ectepicondylar
foramen or canal has hitherto been seen only in the humeri of cold-
blooded vertebrates ; an ‘“ entepicondylar foramen or canal,” charac-
teristic of many genera of Mammalia, is also characteristic of the
humerus in certain Reptilia, but has hitherto been found only
in extinct genera and species of Palseozoic or Permian age.

Thus is exemplified the value of close study and comparison of
small and seemingly insignificant foramina and other minute cha-
racters of fossil bones, the advantage of defining such characters,
their homologues being determined under distinct names, and the
indispensability of such labour in all who would contribute to “ the
progress of sound paleontology ” ||.

Discussion.

Professor Srrrzey could not realize Prof. Owen’s new order. It
seemed to him to rest upon the collocation of specimens, for which
there seemed to be no sufficient evidence to justify the foundation
of a new order upon them. It seemed to be stretching a point a
very long way to refer all these isolated specimens to a new order
of which we know so little. The Bristol specimens of Palceosaurus

* Ato, vol. i. p. 184, no. 943 (1853).

+ “On Vertebrate Fossils from the Panchet Rocks,” &c. in ‘Memoirs of the
Geological Survey in India,’ 4to, 1865, p. 10, figs. 3.4, B. See, however, the
remarks in the Quart. Journ. Geol. Soc. vol. xxxii. p. 98.

The figure and description of the left humerus of the Dicynodon Murrayi,
Huxley, in ‘Quart. Journ. Geol. Soc.’ vol. xv. 1859, exemplify the difficulty
attending the investigation of fossils from the dark hard matrix of the Karoo
beds.

- Quarto, 1876, pp. 35, 43, 53; Nos. 8. A. 59, 8. A. 72, S. A. 88; pls. xxvii.
xli., xii.
§ Quart. Journ. Geol. Soc. vol. xxxii. (1876) p. 96, pl. xi. figs. 6-11.
|| Quart. Journ. Geol. Soc. vol. xxvi. (1869) p. 32.

DEPOSITS ELSEWHERE THAN IN SOUTH AFRICA. 3863

did not seem to him to warrant the removal of that genus from the
Dinosauria.

The Prestpenr (Prof. Duncan) remarked on the interesting mode
in which these’ Reptilian jaws seem to foreshadow those of the
Mammalia. During the existence of the great land surface which
persisted through the Carboniferous, the Permian, and the Trias,
many land reptiles lived upon it, and doubtless developed those cha-
racters which seem to be Mammalian, just as in more recent times
we have similar characters developed under similar conditions in the
Marsupial Mammalia.

Mr. Erurrimer stated that the deposits from which the Bristol
Saurians were obtained certainly belonged to the middle of the
British Triassic series, forming the base of the Keuper.

The Avrnor stated that the teeth of Megalosawrus among the
Dinosaurs approached nearest to those of his Theriodontia, but they
are not Thecodont. Megalosawrus had amphiplatyan vertebre,
whilst those of the Theriodonts were biconcave. Considering the
fragmentary evidence upon which the order Dinosauria was first
established, and the fact of its being now generally accepted, he
did not think that the argument from imperfect evidence against
the order Theriodontia was of much value.

Q.5.G:8. No. 127. 2c

364 J. W. HULKE ON A MODIFIED

39. Appendix to “ NotE on a Mopirrep Form of Dinosavrian Itium,
hitherto reputed Scaputa.” By J. W. Hurxs, Esq., F.R.S.,
F.G.S. (Read June 9, 1875.)

A RECENT discovery of Jguanodon-remains in the Isle of Wight by
the Rey. W. Fox has solved the riddle of the bone which, in the
above note, I brought under the Society’s notice towards the close
of last Session*, and it has also completed our knowledge of the
pelvic elements of Mantell’s Iguanodon. It will be remembered that
Mantell and, later, Owen had regarded the bone as a scapula, first
of Jquanodon, and afterwards of some unknown reptile, that this
decision seemed to be untenable, and that the balance of all the
evidence I could up to that time collect respecting it favoured its
being an ilium.

This fresh material, which Mr. Fox has courteously permitted me
to examine, consists of a series of vertebra, of fragments of large flat
bones which cannot be any thing else than pieces of the ilia, of a pair
of ischia, the former clavicles, and a pair of the bones in question.

The vertebre comprise the sacrum and a few of the adjoining lumbar
and caudal vertebre. Their form, and also that of the ischia, agree
so closely with those of the types of Mantell’s Zguanodon in the
British Museum as to leave no doubt of their generic identity. The
third pair of bones in the pelvic girdle, which are neither ilia nor
ischia, must be pubes (fig. 1, P). Their shape closely repeats that
of the smaller of the two specimens illustrating my note (pl. xxxii.
fig. 1, vol. xxx.). Inverting the bone there figured, its long slender
process (Spp') is easily identified with the thin elongated rod which,
in the Mantell-Bowerbank specimen of the nearly allied Hypsilo-
phodon, Prof. Huxley recognized as the pubis (Pd, pl. 1. Quart.
Journ. Geol. Soc. vol. xxvi.). This identification roughly gives the
direction of the whole bone. Its long slender branch slanted down-
wards and backwards parallel to the ischium ; and the little process
detached from its posterior surface meeting a corresponding process
observable in all well-preserved ischia, converted the upper end of
a long narrow obturator-space into a foramen having the position
of that which in birds (Ostrich) transmits the tendon of the obtu-
rator internus muscle (figs. 1 & 2,0bt). The process marked 7 in
the figure (pl. xxxii. vol. xxx.) jomed the ilium, and that marked
p the ischium; the smooth intermediate arc formed the lower and
front part of the acetabulum, whilst the broad blade-like part sloped
ventrally inwards and forwards (figs. 1-3, p). Iam inclined to think
that the free end of this part met the corresponding extremity of the
os pubis of the other side in a median symphysis; but I have not
yet any positive evidence of this in Jguanodon.

In respect of its large contribution to the acetabulum, the pubis
of Mantell’s Iguanodon resembles that of existing Lacertilia, as it

* Quart. Journ. Geol. Soc. vol. xxx. p. 521, pl. xxxii.

FORM OF DINOSAURIAN ILIUM. 365.

Hig. 1. Restoration of Iguanodon’s pelvis, about one fourteenth natural size.
», 2. A corresponding view of pelvis of Ostrich. } ?
,», 3. A corresponding view of pelvis of an existing species of South-American Lizard,
about twice natural size.
,», 4. A corresponding view of pelvis of Crocodilus niloticus, about one fifth natural size.

The homologous parts in each figure are indicated by the same letters, viz. A, aceta-
bulum ; J/, ilium; pra, its preeacetabular process ; psa, its postacetabular part ; Js, ischium ;
P, os pubis ; p, the part directed beneath the belly where in the Lizard it meets its fellow
in a mesial symphysis ; p', its postacetabular part directed backwards parallel with the
ischium, wanting in Lizards and Crocodiles; ob.f, obturator foramen ; 0b, foramen for the
transmission of the tendon of the musculus obturator internus.

366 J. W. HULKE ON A MODIFIED DINOSAURIAN ILIUM.

does also in respect of its broad ventral extension and probable
median symphysis (compare figs. 1 & 3); whilst its long slender
branch repeats the avian resemblance noticed by Prof. Huxley in
Hypsilophodon (figs. 1 & 2).

My warmest thanks are due to the Rev. W. Fox for allowing me
to study these very instructive fossils.

Discussion.

Prof. SreLry said he thought this paper was one of the most
important contributions to the osteology of the Dinosauria that had
been laid before the Geological Society within his memory. He
entirely agreed with the interpretation which Mr. Hulke had given
of the bones. He said that in the Apteryx the preepubie process of
the pubis is more developed than in any other known bird, while
the whole pelvis in other respects approximates closely to that of
Iguanodon. It was of great interest to find the fossil showing that
this process attained at least as great a development as in the Chelonia
and Lacertilia, because it showed that the pelvis of Dinosaurs was
not entirely Avian, but, like the remainder of the skeleton, showed
a blending of Avian and Reptilian peculiarities. ,

ON THE TRIASSIC ROCKS OF SOMERSET AND DEVON. 367

40. On the Trrasstc Rocks of Somzrser and Devon. By W. A. E.
Ussner, Esq.,-F'.G.8., of the Geological Survey of England and
Wales. (Read June 7, 1876.)

(Communicated by permission of the Director of the Geological Survey.)

ConTENTs.

I. The Mendip Area.
II. Polden Hills, Bridgewater and Taunton Vale.
III. West-Somerset and Devon Area.

INTRODUCTION.

Tux Triassic areas of Devon and Somerset may be divided into three
lithological districts :—the first, on the north and north-east of
the Polden Hills; the second, on the north-east, east, and south-
east of the Quantocks; the third, comprising the rest of the
Triassic area, lies between Watchet and the south coast of Devon,
being bounded on the west by the Devonian and Culmiferous high-
lands, and on the east by the Quantock and Blackdown ranges.

The first of these districts is too well known to call for more than
a general description in this place. In the Geological-Survey Memoir
on the Bristol Coalfield it has been already treated of, the labours of
Conybeare and Phillips, De la Beche, and the late Mr. Wm. Sanders,
with those of Messrs. Moore and Etheridge having left little to be
desired as far as the description of the Triassic strata is concerned.
With reference to the last-mentioned districts, of which we purpose
to give a more detailed though far from exhaustive description, it is
different, the relations of the Trias not having been before estab-
lished, except so far as might be inferred from a study of the
splendid section along the coast between Seaton and Torquay.

Sir Henry De la Beche, in his report upon the geology of Devon
ahd Cornwall, gives a very general description of the Trias, without
attempting a classification of its component marls, sands, and con-
glomerates, which the greater time occupied in the resurvey of the
district has enabled me to make, and, by the kind permission of the
Director of the Geological Survey, to lay before the Society. <A
general description, adducing only such evidence as becomes neces-
sary to prove the sequence which we have adopted, and to the
exclusion of such details as must be reserved for Geological-Survey
publication, is here attempted.

I. The Mendip Area. Trias North of the Polden Hills.

Throughout the districts occupied by Triassic strata between
Bristol in the north and Glastonbury and Shepton Mallet in the
south, and between the mouth of the Severn in the west and
Frome in the east, their lithological characters seldom vary, the
‘ major part consisting of red and greyish green marls, in the
neighbourhood of Paleozoic hills resting upon or dovetailing into
conglomerates with a dolomitic matrix, which from their marginal
nature are but locally observable and superficially occupy a very
limited area. ;

Q.J.G.8. No. 128. 2D

368 Ww. A. E. USSHER ON THE TRIASSIC

The variegated marls outcropping. from beneath the Liassic
strata in the east occupy the lowest tracts of land covered by
the secondary rocks, capped here and there by Liassic beds, where
hills of sufficient height to allow of their presence occur, or
where exceptional cases of fault have rendered both equally sus-
ceptible to the forces of denudation. The scanty herbage some-
times permits the junction between the marls and overlying
Rheetic beds to be seen, even at a considerable distance, on the
slopes of these outliers, the grey and cream-coloured marls of the
Rheetic offering a marked contrast to the green-, grey-, and red-
banded marls of the Trias.

As a general rule, greenish and bluish grey tints are most pre-
valent in the upper portions of the marls. Bands of greenish grey
calcareous stone are frequently met with, sometimes occurring in
lamin, upon the surfaces of which pseudomorphs of crystals of
rock salt have been occasionally discovered. Mr. Bristow found a
specimen in the. railway-cutting between Shepton Mallet and Wells
during my noviciate in field-work. Bluish mottling and banding
is not uncommon in the marl.

Sandstones appear to be intercalated with it, but in very excep-
tional cases, where we were unable to ascertain their thickness or
relations to the underlying beds.

At Claverham, near Yatton, grey sandstones occur which are
much used in the neighbourhood for building-purposes, owing
to their durability and hardening on exposure. The local epithet
for these beds is Clar’ham stone ; but whether they occur as a small
lenticular patch in the marl, or as a boss of a subjacent extensive
bed, there only brought to the surface, we are unable positively to
say, though from the. proximity of Carboniferous sandstones capable
of furnishing them, and the fact that similar beds have not been
traced, we are inclined to the former hypothesis.

At Ridge Hill, on the west of Chew Stoke, beds of soft sandstone
are shown under the marls in a road-cutting. Here again the
proximity of a patch of Millstone Grit at Leigh “Down, about half a
mile to the northward, is sufficient to explain their presence ; but
their vicinity to the conglomerates on the north, and the occasional
presence of arenaceous beds at the base of the marls, render it
difficult to determine whether they do or do not rest on the con-
glomerates directly, their junction with them being concealed by an
overlap of marl.

Arenaceous beds outcrop at the base of the marls, near Backwell,

between Brockley Elm and Yatton, between Yatton and Wrington,
and also at Churchill; they are best developed near Roddy to the
north of Congresbury, and at Churchill.
' The irregular junction sometimes noticeable between the marls
and conglomerates, and the very unequal breadth of outcrop
exhibited within short distances by the latter, render it probable
that these sandstones may represent portions of arenaceous deposits
thrown down further from the old Triassic shores, and subsequently
overlapped by muddy sediments, as a progressive subsidence nar- —
rowed the limits of the existing land.

ROCKS OF SOMERSET AND DEVON. 369

We do. not know of the occurrence of similar sandy beds on the
flanks of the Mendips; superficial evidence confines them to Broad-
field Down.

With very few exceptions the old Triassic coast line in this
area is fringed with conglomerates, sometimes occurring super-
ficially as very thin fringes resting on the Palwozoic rocks, as
on Banwell Hill, Sandford Hill, &c., sometimes occupying con-
siderable local superficies, as near Wrington, Blagdon, Shipham,
Chileompton, Mells, Croscombe, &c. (in which places they occupy
high grounds sloping towards the principal valleys, with minor hills
and undulations of their own), sometimes rising in bold hills wrap-
ping a nucleus of Mountain Limestone which is here and there un-
covered, as at Churchill and Decoy Nyland, near Cheddar, some-
times as inliers in valley bottoms as at Midsomer Norton, some-
times as outliers on the older rocks, as at Slab-Ho (north of Dinder)
and near Binegar and Nettlebridge. In all cases where it rests upon
the older rocks the conglomerate fills the inequalities in their out-
line, presenting a very irregular junction with them in places where
the old coast was much indented.

De la Beche thus describes the composition of these conglomerates
(Mem. Geol. Surv. of Gt. Brit. vol. i. p. 240) :—* The lowest part of
the poicilitic, or new red sandstone series in our district has gene-
rally been considered as composed of a conglomerate formed of
rounded portions or fragments of the subjacent rocks cemented by
carbonate of lime, mixed occasionally with so much carbonate of
magnesia that the name magnesian conglomerate was applied to it,
and magnesian limestone to the limestone into which it sometimes
graduates by the disappearance of the pebbles of the subjacent
rocks, and the presence of little else than the matter of the cement.
Instead of these terms the names of dolomite conglomerate and
limestone were suggested by Dr. Buckland and Mr. Conybeare,
they at the same time observing that the cement or limestone, in
the absence of pebbles and fragments, was sometimes dolomite
(carbonate of lime and magnesia) mixed with carbonate of lime, at
others merely carbonate of lime. ‘They also pointed out that the
pebbles and fragments in the conglomerate or breccia were those of
the nearest hard rock, fragments of coal sandstones being scarce,
probably from their texture being commonly too friable to resist the
friction to which the other pebbles had been exposed.”

De la Beche (ibid. p. 241), after mentioning the irregular pro-
jection of the conglomerates from the margin at intervals into the
marl (of which we recall to mind an instance at Dunyate, near
Axbridge, on the south side of the Mendip hills), proceeds to say :—.
“It will be evident that whenever conditions may arise for the ~
deposit of the carbonate of lime, or of carbonate of lime and
magnesia, as the case may be, without including the pebbles, the
travertino, for such it would be, would cover up the conglomerate
beneath. In supposing, therefore, that these dolomitic conglomerates
may often have been beaches skirting the land of the time, rising
higher and higher up ‘its flanks as such land became depressed

2D 2

So

370 Ww. A. E. USSHER ON THE TRIASSIC

relatively to the sea, we do not suggest a mode of explanation at
variance with the manner in which such things may happen at the
present time, but on the contrary, one in accordance with it.”

He says further (ibid. p. 246), «‘ The conglomerate of the Vale of
Westbury is pointed out as containing an abundance of sulphate of
strontia, forming part of the matter cementing the pebbles and frag-
ments together. The frequent occurrence also of nests of sand in
many places, the irregular cavities filled with rock crystal, chal-
cedony, carbonate of lime and other minerals, and which, when
detached from the matrix surrounding them, are locally known as
Potato-stones, is mentioned (Geol. Trans. vol. i. pp. 212, 291-294,
&c. ; ibid. Ist ser. vol. iv.). These cavities so filled, and shown to be
particularly abundant near Mells (Mendip Hill district), are of the
same kind as those noticed at Kenfig Point, Glamorganshire,” where
the most westerly patch of conglomerate occurs.

On p. 249, De la Beche notices intercalations of flesh-coloured
dolomitic limestones with the marls, on the south of the Mendips,
near Croscombe. They seem to be passage-beds into the underlying
dolomitic conglomerate. The presence of arenaceous matter in the
lower part of, the Trias, in neighbourhoods where Paleozoic sand-
stones occur, has been already noticed; it might be of interest,
however, to quote an instance at Portishead, given by De la Beche
in the work before referred to (pp. 248, 249 *), the beds being
given in descending order :—

Conglomerate.

. Layers of red and buff-coloured sandstone.

. Conglomerate.

. White crystalline dolomite,

. Buff-coloured sandstone.

, Conglomerate reposing on the edges of the coal-measure sandstone
(Pennant grit).

D OUR oo bo

In fact, throughout this district, the general absence of arenaceous
matter of Triassic age may be accounted for by the great quantity
of calcareous sediments afforded to the denuding agents of that
period, through the prevalence over the area of Carboniferous lime-
stones. yen in cases where the shores were wholly composed of
Carboniferous sandstones the transport of calcareous matter from
adjoining limestone coast-lines prevented the representation of the
marginal deposits by sandy beds alone, but produced an intercalation
of sandy and calcareous matter, according as the degradation of the
sandstone cliffs was counteracted by the interception of beach
materials, or counterbalanced by the influx of calcareous sediment
from the vicinity.

The reverse is the case in the Devon area, where the shales and
grits of the Devonian and Culm-measure strata afforded a mass of
arenaceous sediment, in which the occasional limestones occurring
in them were in their degradation unable to produce any marked

* Vide Moore, “ Abnormal Secondary Deposits,” Quart. Journ. Geol. Soe. xxii.
p. 456: a gallery driven in search of manganese, through conglomerates with
intercalated masses of bluish and variegated marl, at Croscombe.

ROCKS OF SOMERSET ANI) DEVON. on

change, or to atford intercalations of calcareous matter other than
is locally to be seen in certain of the lowermost breccio-conglomerates
in the neighbourhood of Sampford Peverell (vide Conybeare and
Phillips, ‘ Outlines of Geology of England and Wales,’ p. 308), and
in the vicinity of Paignton and Teignmouth, in the middle series of
conglomerates near Wiveliscombe, Milverton, and Vellow (near
Stogumber), and the marlstones between Bishops Lydeard and
Crowcombe Heathfield, and the calcareous nodules and beds in the
same sandstones in other places. The fragments contained in the
conglomerates are principally composed of Carboniferous limestone,
exhibiting various degrees of. attrition, from well-rounded boulders
and pebbles to nearly angular fragments.

At Wrington (near Yatton) and Draycot (near Cheddar) beds
rather approaching to the character of a breccia than a conglomerate
occur, and in the latter locality form a very serviceable building-stone.

Organic Remains.—Sir H. De la Beche, in the work before cited
(p. 268), says :—‘‘ In the dolomitic or magnesian conglomerate of
Redland, near Bristol, Dr. Riley and Mr. Stutchbury have found the
remains of two Saurians, to which they gave the names of Z’hecodon-
tosaurus and Paleosaurus.” (Vide Moore, ‘* Abnormal Secondary
Deposits,” J. ce. p. 456.)

We may therefore infer that, from its first known appearance
in the Coal-measures, Sauroid life had continuously progressed
through the Triassic period (of which Mr. Lavis’s recent interesting
discovery at Sidmouth affords further proof) wherever the waters of
that period were favourable to their existence, till in the Liassic
age more suitable conditions for their preservation, if not for their
increase, prevailed.

Thickness.—According to Conybeare and Phillips (p. 299), “ At
Pucklechurch, in Gloucestershire, shafts have been sunk from the
Lias to the coal-measures ;” the Trias was found to be 153 feet in
thickness.

In a section of a coal-pit, 2 mile south-east of the village of
Paulton (ibid. p. 429), the Trias underlying Liassic and Oolitic beds
was found to consist or—

ft. in
BSCR THA ANCL BANG, | 5. 4. swans cenwe ne nds esis sles eav'nses snes). omen cee 132 0
Calcareo-magnesian conglomerate, called millstone ............ 6 O

Total thickness of Trias... 138 0O

In a pit-section at Chilecompton, given by Messrs. Buckland and
Conybeare (in observations on the south-western coal-district), the
dolomitic conglomerate continued for 138 feet from the surface.
(It should be mentioned that the major part of the formation in this
locality is represented by conglomerates, and that, in places, the
Rheetic beds appear to rest directly upon them, so that the formation
of conglomerate appears to have been locally continuous and con-
temporaneous with that of the upper beds of marl.)

The same observers give a section of a pit at Welton, to the
north of Midsomer Norton, which was sunk through 114 feet of
red marl, resting on 6 feet of dolomitic conglomerate,

.3e W. A. E, USSHER ON THE TRIASSIC

“There is a large development of this conglomerate at Stratton-
on-the-Fosse, where it comes in contact with the exposed edges of
the Coal-measures, and again at Mells, whence a narrow belt passes
beyond the village of Elm, skirting the limestones to Vallis. Con-
glomerates are also found continuously along the limestone escarp-
ments, filling up the inequalities of the surface on both the north
and south flanks of the southern portion of the Mendips. ... Few if
any organic remains of a locality so disturbed as were the probably
local coast lines on which these pebbly deposits were laid down can
be expected to have been preserved.” (Moore, ‘Abnormal Secondary
Deposits,” J. ¢. p. 455.)

Mr. Moore says the dolomitic conglomerate is found to vary
from 3 to 60 feet in thickness.

Mr. Moore (“ Abnormal Secondary Deposits,” /. c. p. 498) gives the
following representation of the Trias in the Mangotsfield railway-
cutting :—

Under Rhetic Beds.

Keuper, consisting of variegated red and blue marls with sul- ft. in.
phate of strontia, passing into regularly bedded New Red
Sandstones without orgamismS ...........sseeeeseseeeeees about 25 0

Resting on Pennant Rock.

In the same paper (ibid. p. 458) he says :—“ It will be seen that
within this part of the Somersetshire coal-basin the Triassic rocks
are reduced to a little over 50 feet, whilst south of the Mendips,
and where only partially proved, they attain a thickness of 609 feet.”

He says further (p. 476), “‘ Comparing the greatest thickness

‘of the Trias’ without the coal-basin with the greatest reduction
above the coal-measures within it, we arrive “at the following
remarkable result.”

Without the Coal-basin. Within the Coal-basin,
ie, en, Tt. 1 ate

Triassic beds ” 2000 O 50. 0

Mr. Moore gives the thickness of the Trias at the Tyning Pit
(near Radstock) as 186 feet, of which 19 feet at the base consist
of dolomitic conglomerate.

At Norton- Hill Pit the marls are given as 118, and the. con-
glomerates 54 feet in thickness.

At Batheaston a pit, in 1812, showed 54 feet of Trias under
Rheetic beds.

The red marls of the northern area were deposited continuously
with the upper portions of the marls, upon which the Blackdown
greensands to a great extent rest directly.

The same general description is applicable to the Jatest stage of
deposition in the three districts under consideration ; but the great
attenuation exhibited by the Trias of the northern district precludes
any attempt at correlation as regards time with the much greater
development of marls which in South Devon form only the upper part
of the Trias.

As we have no reason for inferring that deposition was more

ROCKS OF SOMERSET AND DEYON. 373

rapid to the south than to the north of the Mendip hills, and certainly
could never have been so unequal as to cause the sediments formed
in the northern-area to bear so small a proportion to those thrown
down in the southern districts, on the assumption that deposition
took place contemporaneously in both, we have every ground for
inferring with Sir H. De la Beche that “ deposits were not effected
over the northern area to any great extent until the period when
the red marls prevailed, when the waters of the time evidently
overspread a considerable portion, there being a free passage of sea
between Wales and Devon, with outstanding islands.”

II. Area of the Polden Hills, Bridgewater and Taunton Vale.
(a) Bridgewater District.

We have placed this area second in order, not only because it
separates the northern from the southern area, but because it seems
to present a transition stage between the simple uniformity of the
Triassic deposits in the one, and their more complex and variable
nature in the other.

We include in this district the country on the south of the
Quantocks as far west as Taunton. It is therefore bounded on the
south by the Liassic tableland (on the north-eastern side of the
Blackdown Range), on the east by the Polden hills, and on the
north by the Otterhampton Liassic strata, the beds resting upon the
slopes of the Quantocks upon their north-east, east, and south flanks.

The Trias in this area, east of a line drawn north and south
through Bridgewater, and south of a line drawn in an east and
west direction through Monkton Heathfield, appears to consist
entirely of red variegated marls of the same character as those on
the north; that is, everywhere marls are at the surface; but whether
they rest directly on the older rocks or conceal lower beds of dif-
ferent characters, we are, through the want of sufficiently deep
excavations, unable to say—though, from the presence of arenaceous
beds in places on the south slopes of the Quantocks, we are inclined
to believe that an intercalation at least of arenaceous matter with the
marl occurs near the base of the Vale-of-Taunton Trias, marking the
marginal accumulations of a coast receding with gradual subsidence.

This view is in a measure borne out by the observations of Mr.
Moore, to whom we are again largely indebted. He gives a section at
Ruishton, north of Taunton*, 21 feet 5 inches in depth, consisting of—

fG,° in.
Red, bluish sreon, and grey marls........1.....0..ss00sseeerseseeeese 9 7
Gritty conglomerate with occasional sandy bands and inter-

mediate layers of marl, with Fish, Reptile, and Batrachian

remains (including teeth of Labyrinthodon,? Belodon,

Acrodus keuperinus, Estheria minuta, &C.) eeccccccececcscavenes te
RUM NRE NOE SPIO Go aan eles Sse e a wale a nicnk oneonees onsen anes aeeaeeaee | Se
Braap eat ee Maal are hand Sais nae sis SE 5 oie ods dl a e'v' waa gg RAR 0.6
Ben EE FS wan scangasndnaecpa sae sPasnmadne Male 10 0

At North Curry, under an undetermined thickness of variegated
marls, Mr. Moore notes the occurrence of various beds of dull grey

* Quart. Journ. Geol. Soe. vol. xxiii. p. 468.

374 W. A. E. USSHER ON THE TRIASSIC

and brown sandstone, enclosing nodules of marl, and containing
Estheria, plants, traces of fish-scales, and a Reptilian bone—in all
3 feet 6 inches thick, and resting upon marls exposed to a depth of
more than 50 feet.

At Knap, he gives a section of beds to which he assigns a lower
position than those exposed in the North-Curry section :-—

ft. in.
Variegated marls witha hard grey band of stone 2 inchesthick 10 9
Grey sandstone in thin COUSES; .... 600 c0-500+ sen ye jena acne eee 3 0
IAMESUOMC a cocis svs dese baenccisneriaswkcayaiveleseutinca aus cvinee «ques eee 1 10
Red gnarl 3... seeds ad aceiee | Satie ens Veneta ae kisls te te teeta 1 10
TCV SOMGSCODCAY. 05 aca spade arearpeniiges deal cisgslots se mepaualne eee 10 11

28 4

Near Stoke St. Mary,. Mr. Moore noticed thin beds in the
Keuper containing Posidonomya minuta.

Upon the south-eastern flanks of the Quantocks, between North
Petherton and Durston, beds of rock-sand rest upon the Devonian
strata, and attain their greatest superficial breadth (of one mile
and a half) between St. Michael’s church and Thurloxton. From
North Petherton they gradually diminish in superficial breadth upon
the north-eastern flanks of the Quantocks; and at Enmore, where
they are in parts slightly brecciated, they disappear beneath the
overlying marls, which rest directly upon the older rocks for a dis-
tance of three miles. At this point a deep valley has been cut
through the marls, exposing sandstones resting upon the older rocks
and continuing for a mile down the valley to a point beyond Spaxton,
where they are apparently cut off by a fault bringing down mars.

The area between Bridgewater and the Quantocks is character-
ized by the presence of numerous inlying hills of Devonian grit, and
by the prevalence of faults in an easterly and ,westerly direction ;
three of these to the north of Bridgewater are especially noticeable,
as they intersect the Triassic area, affecting its component marls and
sandstones between the alluvial flats of the river Parret and the
slopes of the Quantocks.

The Devonian inliers are frequently flanked by beds of sandstone,
as at Cannington and Radlet farm; but the overlying marls seldom
allow more than a partial exposure (of sandstone) upon their flanks.

Between Charlinch and Cannington, beds of sandstone are inter-
calated with the marls, and are either passage-beds between them
and the marginal sandstones upon the Quantock slopes, or a local
modification caused by the interchange of calcareous matter as fine
mud with arenaceous sediment.

Upon Wembdon Hill another modification of the Bridgewater
Trias is shown in the occurrence of hard bedded breccia, consisting
of subangular and occasionally rounded fragments in a matrix of
calcareous sandstone, and containing beds of sandstone unbrec-
ciated. ‘These breccias are of very local occurrence; as far as we
can judge by superficial evidence, they appear to occupy a position
at the base of the sandstones, as they are always found to pass up-
wards into them; but the relations of the Wembdon sandstones to
the marls on the north and south, against which they are faulted,
cannot be definitely fixed, nor can we at present say whether the

ROCKS OF SOMERSET AND DEVON. SYD

breccia rests directly upon the older rocks, though we are inclined
to consider its position as indicative of a nucleus of older rock,
brought up by faults, but not exposed, through the greater durability
of the 'I'riassic beds thrown down upon it.

To continue the thread of our notice of the marginal sandstones
on the Quantock slopes, interrupted by this necessary digression,—
between the Spaxton valley and Over Stowey we find sandstones
fringing the older rocks, but superficially forming a band of no
ereat breadth. At Nether Stowey, however, we find the sandstones
flanking the older rocks, and at Doddington overlain by marls which
continue to the Liassic districts on the north, extending for some
distance from their margin, and connecting sandstone beds, in places
flanking the Devonian inlier of Radlet Farm, with the marginal sand-
stones on the Quantock slopes. The southern boundary of this
exposure of sandstone is formed by one of the three great faults
before mentioned, bringing up marls on the south.

Upon the north side of the most northerly of these three great
faults, the sandstones of Doddington appear to the south of Fid-
dington, in a small valley between that point and Cannington Park,
and to the south of Cannington Park, proving their continuity in
that district under the marls.

By similar reasoning, into which we cannot now enter, we are
led to infer that the sandstones flanking the Quantocks are of the
same age with those brought up here and there by faults and
exposed on the sides of Devonian inliers. Whether, when the area
of Bridgewater first came within the influence of Triassic waters,
breccias were thrown down over a much larger area than the
vicinity of Wembdon Hill and subsequently overlapped by sand-
stones, or formed merely a local variation of arenaceous beds at
first deposited, we are unable te say.

(B) Vale-of-Taunton District.

In the vale of Taunton we find, superficially at least, an order of
deposition and mode of occurrence in the Triassic strata as simple
as that exhibited in the Mendip country; but on either side of it
the relations of the beds are more obscure, owing to the greater
lithological differences observable and the occurrence of numerous
faults ; ; yet when we trace the marginal sandstones on the Quantocks
from Durston westwards, as we have already done in the Bridge-
water district, we shall find that they form a connecting link be-
tween the deposits of the Bridgewater area, and the more complicated
divisions of west Somerset and Deyon,—in other words, that a con-
temporaneous deposition of arenaceous material took place in the
Watchet-valley, Vale-of-Taunton, and Bridgewater district. From
Durston these sandstones gradually attenuate at the surface, being
concealed by marl resting directly on the older rocks to the south-
east of Hestercombe; at Cheddon Fitzpaine they reappear, being
brought up by fault, ‘and, with an average breadth of half a mile,
continue to within a mile of Kingston (north of Taunton). Here
marls rest upon the older rocks. To the west of Kingston, how-
ever, sandstones reappear and exhibit some superficial development :

376 WwW. A. E. USSHER ON THE TRIASSIC

at Yawford they are concealed by marl, but at once emerge, and
thence continue along the southern ania of the Quantocks at
Bishops Lydeard and in the Watchet valley, finally becoming por-
tions of the Upper Sandstone division of the Devon and West-
Somerset Trias, hereafter to be treated of.

Thickness—Considerable difficulties exist in estimating the thick-
ness of the Trias in this area. Mr. Moore (“‘ Abnormal Secondary
Deposits,” l.¢. p. 458) mentions a boring made at Compton Dundon,
south of Street, through a small thickness of Rheetic beds, in which
the base of the Keuper was not reached at a depth of 609 feet.
If we allow a persistent dip of 3° in the Bridgewater Trias, measur-
ing its outcropping breadth from the Rhetic beds on the Polden
Hills, near Puriton mill, to the Quantocks near north Petherton,
we get a thickness of about 1200 feet; but the conditions for
forming such an estimate are too unfavourable to allow of its being
put forward except as a theoretical possibility. In the Vale of
Taunton, De la Beche allowed a thickness of nearly 300 feet for
the Trias ; but it is plain that the materials for forming an estimate
are equally wanting there; for the vicinity of the district to the
Watchet and Wellington areas, where a very much greater thick-:
ness and more variable composition are exhibited, and the uncertainty
as to the depth of the old rock floor under Taunton render it pos-
sible that the marls, superficially constituting the Trias, may con-
ceal beds of a different character, of the same age as the conglomer-
ates or lower marls of the Watchet and South-Devon area.

We can only partially account for so small an estimate of the
thickness of the Vale-of-Taunton Trias by assuming that the rocky
floor on which it rests formed, as a ridge extending southwards
from the Quantocks, a barrier to the incursion of Triassic waters
during the deposition of the breccias, lower marls, and conglome-
rates in the area between Watchet and the south coast of Devon,
and during the earlier deposition of sandstones in the Bridgewater
district, not finally coming within their influence till the later
stages in the deposition of the Upper Sandstones of West Somerset
and Devon. Even on this assumption, allowing a persistent dip of
3°, and estimating breadth of outcrops from the base of the Rhetic
beds north of Stoke St. Mary to the Quantocks at West Monkton,
the resultant 700 feet can scarcely be considered excessive.

Concluding remarks.

We have hitherto been considering districts in which the Trias of
Somerset presents its simplest form, being mainly divisible into two
portions, the lowermost of which appears to be far the least developed.

In the Mendip area we find :—(1) Variegated marls occasionally
arenaceous or passing into sandstones at and near their base, over-
lying or doyetailing into (2) Dolomitic conglomerate, which though
apparently the first deposit of Triassic times in the area, has, in
places, as a marginal deposit, continued during the entire deposition
of the marls.

In the Bridgewater district, and in the area between Taunton and the
Polden Hills (as far as we can judge by superficial evidence) we find :—

ROCKS OF SOMERSET AND DEVON, O77

(1). Variegated marls occasionally containing arenaceous beds.

(2). Sandstones, or rock sands, apparently forming the marginal
deposit, marking, with few exceptions, the first stages of deposition
in these areas, passing upwards, in places, by intercalation into the
overlying marls, at Wembdon passing downwards into hard bedded
breccio-conglomerate.

If the Wembdon breccias should prove persistent beneath the sand-
stones in the Bridgewater district, a threefold division of (1) marl,
(2) sandstone, (3) breccia, would be necessary, presenting some
slight affinity to the (1) marls, (2) sandstone, (3) dolomitie con-
glomerate flanking Broadfield Down (to the north of the Mendips).

Til. West-Somerset and Devon Area.

The area now to be treated of is bounded on the north by the
Bristol Channel, on the south by the English Channel, whilst the
Quantocks, Taunton Vale, and the Blackdown range form a provi-
sional boundary on the east; the Devonian and Culmiferous highlands,
stretching south from the Brendon Hills, define its western limits.

Much has already been written about the Trias of Devon as
shown in its greatest exposure on the south coast. The whole
section has been described by Mr. Pengelly, and attention has been
specially directed to individual divisions, to the marls and sandstones
between Exmouth and Straight Point by Mr. Ormerod, to the Pebble
beds of Budleigh Salterton by the late Mr. Salter and my friend
Mr. Vicary of Exeter, to the Upper Sandstones by my friends
Messrs. Lavis and Whitaker; so that in the following brief notice
my object is rather to prove the maintenance over the area under
consideration of the sequence observable in the south-coast section,
and to observe the relations of the beds in that section, than to
attempt any detailed account of the lithological characteristics of
separate’ divisions. Inasmuch as one section of any set of beds, be
it ever so clear, cannot be taken without further research as the
absolute and unvarying type of the whole, I propose, by the cor-
relation of the beds in four distinct parts of the area and by a
glance at the structure of each, to endeavour to prove the persistence
of the south-ceast divisions throughout the area, and that any hiatus
is rather due to concealment by faults or overlap* than to impersist-
ence, and, whilst showing the difficulties in the way of forming esti-
mates of their thickness through the prevalence of faults repeating
and (in some cases) cutting out the beds, to attempt an approxi-
mation to the thickness of the several divisions.

I made my first acquaintance with the Trias of this area at
Wellington, in the year 1871, and was then led to make, with one
exception, the same divisions as are displayed in the South-Devon
coast. In the year 1872, the division underlying the conglomerates
was indecisively indicated; but the discovery of its true character in
the following year necessitated a reinvestigation of its boundaries, in
doing which considerable light was thrown upon the relations of

* By the term overlap no unconformity is here meant, but the conceal-

ment of lower beds by the extension, through a progressive subsidence, of those
next in time upon the old rocks.

Os

or

378

W. A. E. USSHER ON THE TRIASSIC

the other divisions, and the distinction between natural and faulted
boundary-lines became apparent.

In the correlation of the divisions displayed in the south-coast
section with those inland, we should hardly expect to find the same
identical lithological characters or identification tests throughout. Nor
is this the case ; for, with the exception of two divisions, both marls,
we find considerable local variations, distinct derivation, and in the
members of the lowest division, much horizontal change or replace-

ment.

The subjoined is a rough table of the Triassic beds in four distinct
and typical localities, between Watchet and the south coast of Devon*.

4,
Ay

Coast SECTION.

marls, veins of gypsum,
calcareous in upper and
middle beds, loamy and
sometimes passing into
rock-sand in bottom beds.

|
(Upper) red variegated

BURLESCOMBE.

Ditto, but no gypsum

observable.

(Upper) red sandstones
aa rock-sand, in places
with calcareous nodules,
mottled greyish, and con-
taining bands and poc-
kets of red clay; slightly
conglomeratic at about 50
feet from bottom.

Pebble-beds of Budleigh
Salterton, large ellipsoidal
pebbles, mostly quartzite
(foreign derivation) in red-
sand matrix, with imper-
sistent beds and bands of
rock-sand.

Red variegated marls
slightly calcareous above,
loamy below, and contain-
ing beds of sandstone often
impersistent. Brickpits in
soil at Exmouth.

Faulted against 4. Brec-
cias of angular stones in red
rock-sand, as at Exmouth
and Dawlish ; with inter-
calated rock-sands near
the latter, and underlying
rock-sands at Dawlish.
Hard thick-bedded (com-
pare Heavitree) breccias
of Teignmouth, breccias,
breccio-conglomerates, and
variegated sands of Tor-
quay and Paignton. (At
Exeter brecciated loamy
clay with beds of sand-
stone.)

Red rock-sand, some-
times buff and grey,

and occasionally con-
taining seams of
clay.

Small quartz pebbles
in red-sand matrix,
occasional impersis-
tent strips of sand,
sometimes compac-
ted,with grit pebbles
locally predominant,

Ditto; structure not
observable, owing to
a thick loamy clay
soil, in which are
brickpits.

Hither faulted out or
concealed by 4. In
places a little breccia
of shale fragments
in sand and dark-
coloured sandstones.
At Tiverton breccia
and breccio-conglo-
merate, hard-rock-
sand, shale breccia
in sand. At Brad-
ninch gravelly brec-
cia of gritstones with
intercalated beds of
red rock-sand. Gra-

velsnear Tiverton,&c.

TILVERTON.

Ditto. Brickpits in
marls at Taunton,
and in marl bottom
beds at Wellington.

Red rock-sand and
sandstone with cal-
eareous nodules in
places, and grey
mottling in streaks
and spots.

into pebble-beds,

Ditto, much faulted;
whole thickness not
shown by the out-
crop.

Redand blackishrock-
sand,in places slight-
ly brecciated in lines
of false - bedding.
Near Wivelacoice
hard breccia, some-
times slightly con-
glomeratic, and red
sandstones, red and
blackish{rock-sands.

|

Ditto.

WILLITON.

Brickpit at
Blue Anchor; gyp-
sum veins on coast.

Red, buff, and grey

rock-sand and sand-
stones; between Bi-
shop’s "Lydeard and
Willton very cal-
careous, almost a
marl-stone in places.

Conglomerate of limestone grit and quartz
pebbles, in hard sandstone matrix, often
thick-bedded. Near Whiteball-Hill tunnel,
north of Burlescombe, seems to be passing

Ditto.
Breccia of angular
fragments in red

sand; hard breccias;
red and _ blackish
rock-sand and sand-
stone, occasionally
slightly calcareous.
Breccias of shale
fragments in sand
and loam.

* The relations of the divisions in the coast section, Burlescombe district,
and the Williton neighbourhood are shown in figs. 1, 2, and 3 respectively, the

sections accompanying this paper.

The similarity of the Milverton and Wil-

liton districts renders a section illustrative of the former unnecessary.

[To face page 378.

Ex- Lime Highland Straight
mouth. Kiln. of Orcomb. Point.
' i

nm OF. tan
19 * gand Fault. g

Numerous faults affecting inter-
calated marl and sandstone beds.

| =

o

oa

’ 4 . v

| E 5 3

mA 5 b 5

nM ~~ ovo -

a S fr}

jaa) nN om iS)

; a : P

Dei wa soe

=f
- -—— - .
ree ------ ew 5 =

> - ~~
gp ne ae
~

a"
te

nches to a mile.)

W. 15° N, E. 15° S.
Stream,

Blackdown Hills.

I a
|
|
!
|
{
1
!

Gravelly 1
]
arly 2 inches to a mile.)

W. 12°S. Be Zo EN.

Stream. Bagborough.

a, ee Limestone. o. Greenstone.
Shales and Grits. * Red clay, probably belonging to
b. oe and Shales. the lowest Triassic series.

¢. Upmestones. + Rheetic beds.

[To face page 378.

Fig. 1.—South-Devon Coast Section. (Horizontal scale about 55; inch-to a mile.)
Th River _‘Teign- ; Langstone Month of Ex- Lime Highland Straig
Petit Tor. Wess Peinn. mouth, Clerk Rock. Dawlish. eT, River Ixe. mouth. Kiln. of cat, pare Stream
1 LS 3 i i i ;

Fault.

reccia and breccio-conglomerate, containing boulders, Faults, Numerous faults affecting inter-
a z : = ealated marl and sandstone beds.
: a
i E ee : 3 2
2 q =| = S =| cl
= . a A} =| s
: a 2 ; 6 :
a g a dd | ; A Hi eA | 3 3 3 a
es a : cl 3 32 2 2 8 3 i 3
Ee) iF} AY S Ba | | a 8 3 P 4 3
‘os D g I So 3 3 £
Gj b £ “sl “4 5 2 8 a 5 & s 5 =
’ ite g <7 eB 2 5.8 3 4 3 Ei 3 2 8 is
A 4 i a 6 aes a a = A A a Steer aes &
oe este ae eat ; Ee eal ie ieclee
1 ' ' t 1 x 1 ~ ia
1 ' = 1 1 1 == 1 IheooS
ee hoa!
nem
i thes

c e" €
Small Small Probable
faults. faults. fault.

Upper sandstone, with calcareous nodules and occasional Upper marls, loamy in the lower parts, gypsiferous in places, Fault.
bands and pockets of clay.

Fig. 2.—Section across the Triassic Valley near Burlescombe. (Horizontal scale nearly 2 inches to a mile.)

W. 15° N. TE, 15° 8.

West-Leigh
Hill.

Stream, Redball Inn, Stream,

Blackdown Hills.

Dower Hill.

Gravelly Triassic breccia, on Culm-Measure grits and shales.

a Fig. 3.—Section across the Valley between the Quantock and Brendon Hills. (Horizontal scale nearly 2 inches to a mile.)
1208, E. 12°.N.

Tolland. Stream, Stream. High Road. Stream. Bagborough. Quantock Hills

H
1
I
1
!
D

g f
a. Clay with flint or chert ‘ d. Upper Marls. \s g. Lower Marls. 8 k, Culm-Measure Limestone. o. Greenstone. :
stones. ¢, Upper Sandstones: ¢’, caleare- | = h. Gower Sandstones. | 3: ers », Shales and Grits. * Red clay, probably belonging to
b. Chalk. ous; e”, conglomeratic. | ioe i. Breccia. e m. Devonian Grits and Shales. the lowest Triassic series.
c. Upper Greensand. f. Pebble-beds and Conglomerate. J S 8 n. Devonian Limestones. + Rheetic beds.

yeas
a

Cah MAP

\ ; [ional ean vy dloiny eps ne ebexctee t
; f mh op hie.
Sue yee

m pit: ye

ae

ROCKS OF SOMERSET AND DEVON. 379

Here, then, we have five distinct and mappable divisions of the Trias,
occupying an area of about 500 square miles—that is, between
Watchet and Minehead on the north and Paignton and Axmouth
on the south, and between Durston on the cast and Wiveliscombe
on the west. Of this area, however, the only member of the
Trias of Devon and Somerset that we can speak confidently about,
regarding persistence, viz. the Upper Marls, occupies about a half.
Whether borings made at Taunton or Axmouth would cut through
all the other divisions in vertical succession, beneath the Upper
Marls, or subsidence over the whole area proceeded unequally, so
that the other divisions were local, and conformably overlain in some
places, and overlapped in others, by the marls resting on the older
rocks, as we proceed eastwards, is entirely conjectural.

We will now treat briefly of each division in descending order,
beginning with

The Upper Marls.

These, from their outcrop beneath the Penarth, or Rhetic beds,
cover a larger area, superficially, than any other member of the
series. Between Durston and Langport, East Ninehead (north
of Wellington), and the Lias plateau of Hatch Beauchamp, Sidmouth
and Axmouth, they extend, concealed by the everlapping Upper
Greensands of the Black Downs, over many places between Taunton
and the coast.

They consist of red marls, variegated greenish grey, sometimes
with slightly bluish bands in their upper portions—locally, as at
Dumpdon, north of Honiton, greenish grey, mottled red. In the
coast sections of Watchet and South Devon they are intersected,
in places, by numerous gypseous veins; occasionally, as on the
‘coast* west of Sidmonth, potato-stones are to be found in them.
In the coast section, in their lower beds, we find them retaining
their marly structure, but losing their calcareous nature, and in the
beds immediately overlying the sandstones, in places, almost passing
into rock-sand; in the cuttings north of Sidmouth, on the branch
line, a thick bed or two of sandstone occurs in them near their
junction with the underlying sandstones. Many disused marl-pits
are to be met with over the area they cover.

Thickness.—In estimating the thickness of these beds we cannot
take a persistent dip across the wide area they cover, ignoring
probable undulations and faults; this would give an enormous
thickness, in excess even of that which we should infer from their
exposure on the coast, where we have a distance of about 8 miles
between their outcrop south of Axmouth and the mouth of the
river Sid, where they overlie the sandstones. In this distance their
continuity is broken by a large fault at Whitecliff, near Seaton. A.
considerable allowance must be made for the synclinal trough which

* Pseudomorphs of rock-salt crystals have been obtained between Peak and
High-Peak Hills and near Salcombe mouth in the coast-section marls by Mr.
Ormerod, the author, and others.

880 w. A. E. USSHER ON THE TRIASSIC

lets down the Chalk and Upper Greensand, between Whitecliff and
Branscombe Mouth. In assigning a dip of 3°, persistent over
an area of 5 miles, allowing a deduction of 3 miles for the synclinal
at Beer, &c., we are scarcely making an underestimate. This would
give a thickness of about 1350 feet. I do not think ory exceed
1000 feet further inland.

The lower beds, sandy marls, lead us by an easy conformable
passage to the consideration of the next member of the group.

The Upper Sandstones.

These consist of sandstones and rock-sands, red in the coast
section ; generally red but often yellowish, buff and grey, resembling
the Upper Greensand in colour in the interior ; locally mottled grey,
in spots and streaks as at Ninehead, near Wellington ; containing
pockets of red clay in parts of the coast section, and in the railway-
cuttings north of Sidmouth lenticular bands of marl in the top beds,
and near Milverton and Wellington beds of clay in their lower
portions (= WATERSTONES).

At Otterton Point they contain two or three conglomeratic beds,
and a new. pebbles in false-bedding lines, also observable near
Harpford, 5 miles north of the Point. |

Calcareous nodules, and veins and nodules of ironstone occur in
them on the coast, and in many localities inland. Between Crow-
combe-Heathfield station (south of Watchet) and Bishops Lydeard
these beds become exceedingly calcareous; about the middle of
the division they present a nubbly appearance and bluish grey colour
with occasional red spots, and might locally be almost considered
a limestone. From a minute investigation of the Watchet district,
my colleague, Mr. Blake, suggested the possibility of these marlstone
beds representing a portion of the hitherto unfound Muschelkalk ;
but from an acquaintance with the whole area where the Upper
Sandstones are present, I can only regard the local variation the
beds here assume as an intensified form of the calcareous bands
and nodules to be met with in many places between Bishops Lydeard
and the south coast.

The discovery of organic remains in the South-Devon and West-
Somerset Trias seems to be confined to this division, the respective
finds of my friends Messrs. Layis and Whitaker being in its upper
and lower beds.

A conglomeratic bed occurs in the sandstone at Ninehead (near
Wellington), but apparently much higher up than the beds of
Otterton Point. The Otterton beds were observed and commented
on by Mr. Pengelly and many other able observers long before I
knew of their existence, and have become of peculiar interest as the
site of the Hyperodapedon found by Mr. Whitaker.

On Woodbury Common, north of Exmouth, the rock-sands very
much resemble Greensand in colour, and in the older Survey map
have been mistaken for it.

These sandstones frequently exhibit false-bedding.

ROCKS OF SOMERSET AND DEVON. 301

Extent.—With one exception south of Williton, where these beds
are faulted out, they run with an average breadth of outcrop of 14
mile, between Williton and Otterton Point, a distance of over 30
miles: in sheet 22 of the Ordnance map, north of Otterton Point
their breath of outcrop appears to be from 3 to 4 miles; but this is
probably due to repetition by faults.

Economic Uses.—At Watt's Bridge, near Bishops Lydeard, some
hard thick-bedded sandstones, used for building-purposes, occur in
this division. Between Bishops Lydeard and Crowcombe the
calcareous sandstones are largely quarried to burn for lime.

‘Thickness.—In estimating the thickness of these beds we cannot
allow more than a mile and a half of outcrop, north of Ottery St.
Mary, which, with a dip of 3° persistent,-would give a thickness
of about 400 feet. If we take the coast section, it is plain that an
estimate made at right angles to the strike, measuring outcrop from
their junction with the last capping of marl (supposing the absence
of disturbances) to the outcrop of the. pebble-beds of Budleigh
Salterton, would give the true thickness.

Estimates have been previously made ignoring any disturbances
in these rocks as displayed on the South-Deyon coast, on the ground
that the beds cut out by faults, would balance those repeated: a
glance at the section of these sandstones made from close observa-
tion of the coast will, at least as far as this division is concerned,
show the fallacy of this reasoning. The fault at Chit Rock brings
down about 200 or 300 feet of Upper Marls—the same sandstones
_ there cut off appearing again at the base of the cliff south of Peak
House, and again outcropping on the surface south of High Peak,
but not finally, as a fault with a downthrow of about 50 feet again
brings down Upper Marl a little to the north of the Preventive
Station at Ladram Bay. About 80-yards further south the first:
cutting-out fault occurs, which disposes of about 5 feet of marl ;
but this is more than counteracted by another fault not 80 yards
further south, which throws down about 14 feet of marl, the sand-
stone finally emerging from the last capping half a mile south of the
road to the beach, at Ladram. Even from this point we cannot
take the coast section, as it frequently represents the strike of the
beds, and in many places they appear perfectly horizontal, whilst
in one place, as may be seen in the section, either a fault or a
crack (I have not yet succeeded in ascertaining which) by inland
evidence occurs. ‘So, allowing the coast-section sandstones an out-
crop of 22 miles, and a persistent dip of 2°, we have a thickness of
about 530 feet—if any thing, an excessive estimate.

Both from a better acquaintance with these beds as shown on
the coast, and to point out a few of the difficulties in the way of
estimating the thickness of the divisions, even where the relations
between the under and overlying members are observable, I have
gone into their occurrence in the coast sections at greater length.

The traces of conglomerate at Otterton Point may be considered
indicative of a passage into the next series—

382 W. A. E. USSHER ON THE TRIASSIC

The Pebble-beds of Budleigh Salterton and of Burlescombe, and the
Conglomerates of Milverton.

This division affords 3 distinct lithological types, passing from one
to the other, I believe horizontally ; but, owing to this being the
thinnest member of the Trias of Devon and Somerset, it is the most
affected by faults, so that the changes in composition cannot be
always traced.

We will commence with the coast, as the conglomerate in the
Williton district is so much faulted where present and so often ab-
sent, owing to the same cause, as well as so similar to that of
Milverton and Wiveliscombe, that it would form a bad starting-
point.

The following are general characteristics of all three types—well-
worn nature of contained fragments, impersistent strips and beds of
rock-sand and sandstone intercalated, and general thickness of
beds.

At Budleigh Salterton the pebble-beds form, for nearly 20 chains,
the substratum of the beach, dipping conformably under the Upper
Sandstones and overlying Marls which dip towards the north-east at
an angle of 5°. They consist of about 100 feet of large ellipsoidal
quartzite pebbles of foreign derivation (as far as at present known)*
in red sand, and containing impersistent strips and beds of rock-
sand. As far as I have yet seen, the quartzite pebbles seem to
give way gradually to grit and quartz, and to become. smaller
north of Aylesbere Hill, between Whimple and Ottery St. Mary.
. Between Tallaton and Uffculm, though the feature is generally
persistent, the pebble-beds are present only at intervals along the
line of feature, but sufficiently close to justify me in considering
their absence due to faults. Thus they are shown at Blue Anchor,
Higher Tale, Clist, William’s Cross, Kentisbere, Pirzwell, Higher
South Hill, and Bull Moor, near Uffculm. In most of these patches
the pebbles are smaller than those south of Tallaton. At Uffculm
and Uffculm Down they consist of pebbles of grit and quartz (in
some places compacted by ironstone into comparatively hard beds),
in a red-sand matrix, and containing strips of sand. At Burlescombe
they exhibit the appearance of a loose gravel of small quartz
pebbles in sand. Between Uffculm and Whiteball Hill their ab-
sence in one or two places is manifestly due to faults. Near Thorn
St. Margaret, they appear to be passing into a hard rock, but the
immediate change into the conglomerates, faulted against them,
cannot be observed. The conglomerates of Thorn St. Margaret,
Milverton, Castle Hill, near Wiveliscombe, Combe Florey, &c. con-
tain pebbles (generally large) of limestone and Old Red Sandstone,
and smaller ones of quartz (the limestone pebbles being in excess .
of or less than the others in different places) in a matrix of
hard red sandstone, generally slightly calcareous. The beds are
generally massive; but those at the top of sections often appear
much thinner than the rest. Subangular fragments of grit and

* Mr. Peach advocates their derivation from certain quartzites in Cornwall,

ROCKS OF SOMERSET AND DEVON. 383

quartz sometimes occur in the matrix, and give the beds the slightly
hybrid character of a breccio-conglomerate, as at Vellow near
Stogumber.

It is possible that the thick-bedded breccias of Wembdon, overlain
as they are by sandstones, and containing intercalated beds of
sandstone, may prove to be the representative of the conglomerate
division in the Bridgewater area.

Thickness.—As the conglomerates and pebble-beds seldom exceed
100 feet in thickness, and both contain intercalated beds of sand-
stone, so that in some places a section might display more sand-
stone than conglomerate, it is probable that the local absence of
this division may be due to replacement, and the sandstone beds
traced from Otterton Point (between the conglomeratic bed there
and the pebble-beds) pass into conglomerate in other parts of the
district. ;

Mr. Pengelly noticed the feature made by the pebble-beds of
Budleigh Salterton, being the line of hills running thence to
Tallaton. In tracing the conglomerates from Wellington south-
wards, I found that feature maintained, even where the beds them-
selves were absent, indicating a fault junction between Upper
Sandstone and Lower Marls.

Economic Uses.—The pebble-beds of Uffculm and Burlescombe
are quarried for gravel*. The conglomerates of Milverton, between
Thorn St. Margaret and Alcombe, near Minehead, are quarried out
for building-purposes, but more often for the sake of the contained
limestone pebbles to burn for lime.

I have not found seams of clay in either the conglomeratesf or
the pebble-beds. In one instance (east of Whimple) at Straightwayt
Head, I was informed that they had worked out a bed of clay in
excavating the gravel.

In mapping the conglomerates west of Milverton I noticed clay,
but concluded it was merely a soil or beds in the conglomerate.
Near Tiverton Junction, however, the extent of ground it occupied
led me to a better understanding ; but it was not till the summer of
1873, in making a rapid survey of the coast section, that I recog-
nized the true nature of the loamy clays in

The Second Series of Marls.

Underlying the pebble-beds conformably, we find red marls, varie-
gated greenish grey in bands, and slightly calcareous in the upper
beds. They seem to continue only for ? mile from their outcrop,
under the pebble-beds on the top of the cliffs, in unbroken succession,

* P.O. Hutchinson, Esq., of Sidmouth, informs me that manganese was
worked in the pebble-beds at Yattington (or Yettington), a hamlet to the
north of Budleigh. In sheet 22 of the Ordnance map,

+ Except where the base of the overlying sandstones consists of intercalated
beds of clay and sandstone, when the top beds of conglomerate may occur in
or with clay.

+ Straightway Farm is called by the inhabitants Straightgate Farm.

@.J.G.8. No. 128. 2 £

384 W. A. E. USSHER ON THE TRIASSIC

and for 17 mile along the beach, but long enough to expose all the
homogeneous portion of the deposit in the coast section. South of
Westdown Farm some grey sandy beds begin to emerge on the
beach; but a fault throws them against coarse red and greyish
sandstones, of which, with a capping of marl, Straight Point consists.
Beds of clay occur in these sandstones, which may either be the
beds under the grey bed on the beach before alluded to, faulted up
(in which case they would probably be beds of sandstone in the marl),
or lower sandstones. The Lower Marls contain many beds of sand-
stone frequently impersistent in their lower portion. Theseare only |
mappable as far as Woodbury, owing to a thick loamy clay soil which
effectually conceals the marly structure as well as any changes the
deposit may undergo. From Straight Point to Exmouth numerous
faults obscure the relations of the beds affected by them*; I have
noticed nine of these. The relations of* the marls are still further
obscured by the last fault at Exmouth, which brings up the lower
divisions and not even the upper beds of that.

Between Straight Point and Exmouth the principal difficulty con-
sists in the position to which we assign the faulted sandstones.
Some of them are undoubtedly beds in the marls, and as such
indicate a passage into the lower series; but the sandstones of
Straight Point, near Littleham, Withecombe, and Lympstone might
well be taken as belonging to the lower division of sandstones,
underlying the Lower Marls, and overlying the breccias of Heavitree,
east of Exeter, and the breccia at Dawlish.

The Lower Marls occupy a considerable area, with an average
breadth of 2 miles between Exmouth and Burlescombe. They lie
on the slopes and low ground below the pebble-bed feature; but,
owing to the thick loamy clay soil concealing them, it is difficult
either to obtain dips or to trace faults, except by inference, in them.
This loamy soil may be in a measure due to washes of sand from
the feature above, whilst the superficial portions of the marl were
redeposited by the former streams that assisted in the excavation of
the valley of the Culm.

Where the valleys are broad, as in the Vale of Taunton, and
between Topsham and Exmouth, I have not found much difficulty in
observing the structure of the Upper or Lower Marls. Between
Burlescombe and Castle Hill, near Wiveliscombe, the Lower Marls are
much cut up by faults throwing them against conglomerate and
Lower Sandstone, and near Westford, west of Wellington against
Upper Sandstones. Northey Farm, near Fitzhead, north of Welling-
ton, is the only instance in which they are altogether absent between
Stogumber and Exmouth. This had been mapped by my colleague
Mr. H. B. Woodward before we recognized this division at all; and
the faults drawn by him in that instance, and in a few square miles
he had mapped in the vicinity, amply account for the absence of the
marls at Northey Farm, and are not susceptible of correction.
Between this farm and Bicknoller (near Williton) the Lower Marls

* Marls with beds of sandstone, and sandstones with beds of marl.

ROCKS OF SOMERSET AND DEVON. 385

are faulted against breccia (lowest series), Conglomerates, Upper
Sandstones, and Upper Marls.

Economic Uses.—The loamy clay soil affords material for brick-
making. The continuity of this division might almost be proved
by the numerous brick-pits in its soil between Burlescombe and
the coast, as in Leonard Moor, north of Tiverton Junction, east of
Collumpton, Clis| Hydon, near Exmouth, north of Burlescombe, at
Cutcomb Farm east of Wiveliscombe. South of Williton, a thick
drift-gravel often conceals the second marls.

Thickness—We have now to consider the slender evidence pro-
curable to assist us In arriving at an approximate estimate of the
thickness of this division. If we could take the bed of grey sand-
stone on the north side of the fault, south-east of Westdown Farm,
as the top bed of the Lower Sandstones, the breadth of outcrop*,
allowing a persistent dip of 5°, would give a thickness of 530 feet.
600 feet is the utmost I should be inclined to allow for the whole
division inland.

If the sandstones of Straight Point are beds in the Marl, indi-
cating a local passage into the Lower Sandstones, and do not
represent the upper portion of the Lower Sandstones themselves,
they can scarcely add more than 300 feet to the thickness of the
Lower Marls in this part of the district, giving a total thickness of
830 feet.

Unfortunately, inland we have no better data to go upon, as, in
the cases where the relation between the marls and their over- and
underlying beds are observable, their own structure is concealed.
Allowing the average breadth of areat they cover as the breadth of
outcrop, with a dip of 2° persistent, their thickness would be about
360 to 400 feet.

The intercalations of sandstones with the marls on the east of
of Exmouth seem to indicate a passage into the upper member of
the lowest division in South Devon.

Lowest Division.

In considering this division (so far as we know, the lowest, and
certainly the most varied member) of the Devonshire Trias we must
treat the constituent parts, locally and lithologically, according as
they are represented by (lower) sandstones, hard breccias, breccio-
conglomerates, brecciated sands, brecciated loamy clays, or clays.
In so doing it must not be implied that beds in different localities,
of the same lithological character and composition, are either con-
temporaneous or occupy the same position in the division.

Where present, in the area between Minehead and the south
coast, these beds present many different phases, occurring as sand-
stones and sands, brecciated sands and sandstones, gravelly breccias
(conglomeratic) clays, often brecciated with shale-fragments, sand-
stones, and sands which occur intercalated at any horizon. Gene-

* 12 mile.
+ About 2 miles, z. e, varying from 1 to 3.

2E 2

386 Ww. A. E. USSHER ON THE TRIASSIC

rally in the southern part of the area sandstone seems to be the
upper variety, faulted out at Exmouth*, but appearing immediately
north of it, and probably occupying the bed of the Exe as far as
Topsham, where it is developed. Between Topsham and Silverton
it covers a considerable area with an average breadth of about two
miles. It probably conceals other varieties of the division at Polti-
more. Between Heavitree and Honitons Clist (east of Exeter) these
sandstones appear to dip with the ground, and to contain traces of
clay. Owing to the sudden changes the members of this division
present, it is almost impossible to say whether the absence of Lower
Sandstones about Bradninch is due to replacement by breccia, or
whether their development north of Collumpton is due to the con-
cealment of lower beds of different character, or to change from
breccias. .

Due east of Burlescombe the whole division is either cencealed by
Lower Marls, faulted out, or represented by clay-beds. Traces of
the sandstones (Lower) occur at Ford Place and near Whipcoats.
West of Thorn St. Margaret the Lower Sandstones either conceal
or represent all the lower varieties of the lower division; for their
junction with the Lower Marls is a faulted one, as also north and
east of Thorn St. Margaret, where they are let in by faults. Be-
tween Thorn St. Margaret and Wiveliscombe the visible part of the
lower division is chiefly represented by sandstones, faulted against
the Devonian at Horridge Down, south of Wiveliscombe (fault shown
in the railway-cutting).

Itis very likely that north of Burlescombe these sandstones repre-
sent the middle or lower beds of the division, and not the upper part,
as at Honiton’s Clist and Broadclist. A large east-and-west fault from
Grant’s Farm, north of Wiveliscombe, shifts the position of all the
divisions on the north side of it eastwards; so that we find sand-
stones and breccias of the Lower Series faulted against Upper Sand-
stones, and Conglomerates and Upper Sandstones against Upper Marls.
From this fault to Washford (near Williton) the lower division is
well developed. It is much faulted north of Stogumber. In this
district sandstones generally occur intercalated with breccias; but,
where separable, are found to underlie them.

From a point between Bilbrook and Roadwater, to Dunster, marls
generally conceal the sandstone, which is shown at Withycombe and
Dunster Park. I am inclined to consider these last belonging to
the Upper divisions, and the absence of the rest of the series due
both to numerous faults and overlap; but this is only a supposition,
based on lithological affinities which are not always trustworthy.

From Dunster to Minehead sandstones occur in the Lower division,
where it is thrown up by faults. Between Bratton and Selworthy,
west of Minehead, sandstones with intercalated beds of breccia,
which Mr. H. B. Woodward concurs with me in referring to the
Lower series, occur. South-west of Grabbist Hill a thin fringe of
sandstones and intercalated breccias occurs, probably overlapped

* At Dawlish red sandstones overlie the breccia, and are apparently brought
down by successive step-like faults between Langstone Point and Dawlish.

ROCKS OF SOMERSET AND DEVON. 387

and to a great extent concealed by Lower Marls, as, between the
western extremity of Grabbist Hill and Horner (south-east of Por-
lock) the Lower series is developed principally in the form of breccia.
South and west of Luckham sandstones occur either resting on or
faulted against the older rocks.

Character of Lower Sandstones.—These beds consist of red-rock
sands and hard sandstones, and differ from the Upper division in
being almost uniformly red. They are stained blackish at Lower
Welshford, near Thorn St. Margaret, near Gaulden Farm, south of
Tolland, and in the neighbourhood of Luckham ; though sometimes
slightly calcareous, the network of concretionary veins and the
calcareous nodules, often locally distinctive of the Upper Sandstones,
are generally wanting*. They are also found passing into, inter-
calated with, and containing seams and beds of breccia, and at
Bradninch associated inseparably with a gravelly breccio-conglo-
merate. In distinguishing them from the Upper Sandstones by
colour I should have made a notable local exception in the light-
coloured and banded sands and sandstones of Torbay. The Lower
Sandstones, where the Lower division is developed, are found over
the area treated of to occur at almost any horizon with reference to
the other varieties, and also inseparably associated with them.

Breccias.

In describing the different purely lithological varieties of breccia
we will first consider the apparently uppermost form exhibited
in the coast-section. In the Shrubbery at Exmouth we find
red sandstones brecciated with numerous angular grit-fragments,
generally of a uniform size; the same general character is dis-
played by the breccias of Langstone Point and Dawlish, and, in a
less degree, by breccia at the Docks in Exetert. North of Exeter
in many places brecciated rock-sands and sandstones occur, differing
from the Dawlish breccias either by the fragments contained being
closer and the matrix more impure, as shown in some of the rail-
way-cuttings north of Stogumber, by a difference in the material,
as displaved in the breccias containing shale and grit-fragments
resting on Culm-measures on the north of the Tiverton Valley, or by
resemblance to a gravel in containing rounded as well as angular and
subangular fragments close together and _ of various sizes, as in the
neighbourhood of Bradninch and Roundham Head, near Paignton,

* T have hitherto failed to detect the presence of gypsum in the Lower
Marls; though it is possible that marls containing veins of gypsum near Withy-
eombe (Williton district) may belong to this division, I am inclined upon other
grounds to ascribe them to the Upper-Marl division.

+ Fragments of igneous rocks are numerous in the Teignmouth and Dawlish
breccias ; and in parts of the Crediton valley the breccia is almost wholly made
up of them.

Murchisonite is most plentiful in the breccia south of Exeter and in the
Crediton valley. The crystals when found singly, as is often the case in the
last-named locality, exhibit signs of attrition on their edges.

So far as I am aware, few, if any, fragments of igneous origin are found in
the conglomerate and pebble-bed division

388 W. A. E. USSHER ON THE TRIASSIC

on the coast. The gravelly breccias in the vicinity of Tiverton, Col-
lumpton, and Cadbury, near Crediton exhibit a matrix of brown
earthy sand in which the fragments are often loose.

Hard Breccias.—Between Clerk Rock, north of Teignmouth, and
St. Mary’s church the coast-section exhibits hard breccias, often very
thickly bedded, readily distinguishable from the Dawlish variety by
their presenting all the characteristics of a hard rock in which matrix
and contents are weathered alike, and not like the Dawlish breccias,
from the weathered surfaces of which the fragments stand out, being
much more indestructible than the matrix. The stones contained
in the Teignmouth breccias are of all sizes, from boulders as large as
a man’s head to pebbles the size of a pea, and exhibit a much greater
variety of derivation than the rocks of Dawlish; they also show a
various amount of attrition, from perfectly unworn fragments to
rounded boulders. The contained fragments consist of Devonian
grit, granite, trap, quartz, and limestone rocks, in varying propor-
tions, Devonian-grit stones being in considerable majority, limestone
fragments occurring locally. The matrix consists of hard reddish
sandstone, often coarse-grained. Inland the breccias of Heavitree,
east of Exeter, and of Sampford Peverell exhibit some resemblance
to those of Teignmouth: the contained fragments are angular and
subangular, smaller than the average of those on the coast; and
limestone is absent in the former. At Halberton and Sampford
Peverell, at the mouth of the Tiverton valley, hard-bedded breccias
occur, containing fragments of culm grit, limestone, and quartz;
north of Sampford Peverell these beds present the appearance of a
conglomerate.

At Canon Leigh, near West Leigh, a few beds of hard breccia are
visible, faulted against culm limestones ; east of Bathealton, south of
Wiveliscombe, they dip over Lower Sandstones, and are cut out by a
fault bringing up Lower Marls. North and north-east of Wivelis-
combe breccias occupy high ground, dipping over Lower Sandstones,
and at Castle Hill faulted against Lower Marls; here they contain
rounded fragments and are not thick-bedded.

Between the great fault north of Fitzhead (near Wiveliscombe) and
Lydeard St. Lawrence the beds present more the appearance of a
conglomerate than a breccia, containing pebbles of grit, quartz, and
limestone. North of Lydeard St. Lawrence beds of breccio-conglo-
merate and hard breccias occur, intercalated in brecciated sands and
Lower Sandstones; but the general character of the breccia in the
neighbourhood of Luckham*, Williton, and Stogumber is more in
accordance with the breccias of Dawlish (with many exceptional
phases, all of which have their congeners between the Watchet
district and the south coast of Devon) than those of Teignmouth.

We cannot here go into all these minor modifications which the
breccias present. In numerous instances, on the older rocks,
bordering the Tiverton and Crediton valleys, and on Spray Down,

* A band of gravelly breccio-conglomerate, first noticed and mapped by my
colleague Mr. J. H. Blake, occurs in the breccia near Luckham, and somewhat
resembles the gravelly breccia of Bradninch in character,

ROCKS OF SOMERSET AND DEVON. 389

near Broadclist, the Trias assumes the character of a brown gravelly
mass containing numerous large grit boulders, with an earthy
matrix and very feeble evidence of stratification. In all the varieties
of breccia previously alluded to, local beds and impersistent bands
of sand and sandstone may occur at any horizon.

The next (and, so far as I am aware, the final) variation this division
assumes is not present on the coast, and only very locally observable
inland, viz. brecciated loam and clay and intercalated clays and
coarse sand finely brecciated. Exeter and the Crediton valley, and
between Silverton and Bradninch, are the only localities in which I
can speak with certainty of its occurrence.

The railway-cutting near Queen’s-street Station, Exeter, shows
red loamy clay, brecciated with shale-fragments, and containing a
few even red-sandstone beds intercalated. Culm-shales are ob-
servable cropping out in one place at the base of the cutting; so that
either there is a fault between Queen’s-street and the Docks, where
breccias somewhat resembling those of Dawlish occur at a much
lower level, or the floor of the older rocks on which the Triassic beds
were deposited is exceedingly uneven. However, as the mapping of
Exeter is not quite completed, the question is open to settlement.

In St. Sidwell’s parish, Exeter, brecciated loamy clay (fragments
of culm-shales), with beds of red clay and loamy clay, occasionally
brecciated, are shown. In places north of Newton St. Cyres, near
Whipton, and near Polestow, north-east of Exeter, dark-red clay,
mottled grey in the latter place, and in parts thickly brecciated with
shale-fragments, occurs.

South of Thorverton and east of Crediton the beds are very vari-
able, showing dark-red loamy clay, bands of coarse reddish-brown
brecciated sand with lines of grey sandy loam, coarse dark-red sand
and sandy loam, reddish-brown and red laminated sand, breccia of
shale, igneous and grit-fragments in sand and loam, and breccia of
small grit, igneous and quartz-fragments.

No definite position can be assigned to these clay-beds in the
lowest division, as they probably locally replace the other lithological
constituents.

We have now glanced at the lowest division of the Devon and
West Somerset Trias, in itself exhibiting affinities to each and every
other member of the group,—the clay-beds to the Upper and Lower
Marls, the sandstones to Upper Sandstones, and the breccio-conglo-
merates to the Conglomerates, whilst the gravelly breccias exhibit
some points of similarity to the Pebble-beds. We find the greatest
and most complete development of this division exhibited in the
coast of Devon, from Exmouth to Torquay and Paignton, in the
valleys of Crediton and Tiverton, and north of the great fault north
of Milverton. In the intervening districts we have pointed out the
probability of some of the unrepresented varieties being concealed
by the one superficially developed, and, by mentioning the occurrence
of sandstones at the top of the division about Exeter and Broadclist
and at apparently the bottom, from Wiveliscombe and Bathealton
northwards, shown the possibility of the absent breccias being

390 W. A. E, USSHER ON THE TRIASSIC

represented by replacing sandstones. In close proximity to those
localities where the Lower Division seems entirely absent we have
pointed out at least two* considerable faults affecting the junction
of the Trias with the older rocks, and shown the numerous disturb-
ances displacing the members of the Trias; so that, though there
is a possibility of the Lower Division being absent altogether in one
or two places, facts seem to point to its elimination by faults or
concealment by the Lower Marls.

Economic Uses.—At Woodcock’s Well, west of Collumpton, and
Cowley Moor, north of Tiverton, bands or lenticular masses of clay
occur in the Lower Sandstones, and afford material for brick-making
in both those places.

At Langford (Langord on Ordnance Map), near Upton Pyne,
brecciated sandstones have been worked for manganese. Breccias
and sands at Luckham, stained blackish, have been worked for iron-
ore. In two brick-pits near Montle Grand, in Exeter, clay and
brecciated clay are used for brick-making, the shale-fragments being
burnt in the brick, which is hard and rough.

The breccias of Heavitree, Luckham, and Sampford Peverell are
quarried for building-purposes, and, in the last-mentioned place also
for burning lime, owing to the contained limestone.

Thickness.—Here, again, we are at a loss in forming an estimate
of the thickness of the division as exposed in the south coast. In
the first place, there is a probability of at least a hundred feet of
sandstone being cut out at Exmouth; in the next, though many
high dips are to be seen in the beds between Exmouth and Torquay,
my friend Mr. H. B. Woodward assures me that they are in many
cases due to faults, and that the beds are in numerous instances
comparatively horizontal; it is also evident that the coast-line
cannot represent their thickness, as it crosses the line of dip
diagonally, so that, instead of 10 miles from Exmouth to Torquay,
we must take 7 miles between Exmouth and Chudleigh. Starting
with a fault, and the probability of several more occurring, but not
to be traced, owing to homogeneity of beds traversed or variability
of the division, any attempt at estimating their thickness would be
very untrustworthy.

South of Exeter the area occupied by this division has a breadth
of from 6 to 8 miles; north-east of Exeter about 3 miles, two of
which may be taken as occupied by tolerably horizontal Lower
Sandstones, and the lower beds as concealed. The broadest part of
the Crediton valley occupied by these beds is 33 miles ; but, both on
the north and south side of the valley, they appear to dip away from
the older rocks, forming a synclinal trough, the sandstones at Upton
Pyne, on the south side of the valley, apparently concealing breccias.
The axis of depression is a little to the north of Bramford Speke+.

* At Horridge Down, south of Wiveliscombe, and at Canon Leigh, near
Westleigh.

t This synclinal axis seems to pass into the Culm-measure area on the south-
west of Newton St. Cyres. West of St. Cyres the Trias in the Crediton valley

is almost exclusively composed of breccia containing numerous fragments of
igneous rocks,

ROCKS OF SOMERSET AND DEVON. 391

Assuming the rocks north of this axis to be unfaulted, the few dips I
have taken in them, averaging 10° due south, must not be taken into
account, as they are only local; and any estimate before the whole
valley is resurveyed and a more minute examination made of the
part already gone over, would be premature.

The Tiverton valley, with a breadth of 4 miles north and south
through Halberton, seems also to present a synclinal in these beds.

If we allow a tolerably persistent thickness for the Lower Division
over the whole area, which is very unlikely, the Watchet district
might afford a clue.~ Assuming that their average width, about a
mile (south of Stogumber), is a persistent outcrop where the whole
thickness of the division is represented, a persistent dip of 8° would
give a thickness of about 720 feet. Allowing the coast-section beds
to represent a maximum development, Mr. Woodward agrees with
me in considering 1000 feet as an outside estimate for them.

General Observations.

In the foregoing all the estimates made of the thickness of the
divisions must be considered merely approximations based upon
such material as the survey of most of the district has supplied me
with. Tracts yet remain to be investigated which will probably
throw more light upon the relations of two of the divisions—the
Pebble-beds and the Lowest Division.

Mr. H. B. Woodward has subdivided the Triassic rocks in the
immediate neighbourhood of Williton, and in that district fully
corroborated the sequence of which it is the object of this paper to
attempt a description. As previously stated, I do not insist that
this sequence is present over the whole area covered by Triassic
rocks in West Somerset and Devon, or even over a much greater
space than where it is exposed.

In a slowly and unequally subsiding area beds of breccia may
have been forming before the land under the area now covered by
the Upper Marls -had come within the influence of the destructive
and accumulating agencies at work in other portions of the district ;
and, when it did, the sediment deposited may have been of a
totally different character. I only desire to show that, from the
outcrop of the Upper Sandstones to the older rock-margin, this
sequence, considering the impersistent nature of the divisions, is,
we might almost say, abnormally persistent.

At Exeter, about Silverton and south of it, in the Crediton valley,
and at the western extremity of the Tiverton valley, patches of
basalt occur in connexion with the Lowest Division. Their mode
of occurrence and composition will be treated of in the Geological-
Survey memoir. |

The frequent high dips exhibited by the lower beds of the Trias
down the slopes of the older rocks, upon which they rest, decreasing
in amount of inclination, though generally persistent in direction as
we recede from their (the older rocks’) superficial limits, the presence
of patches of the lower beds of the Trias on the Paleozoic Highlands,.

392 Ww. A. E. USSHER ON THE TRIASSIC

as at Stoodleigh Beacon and Spray Down, and the rough parallelism to
the old rock-margin exhibited by the majority of the faults affecting
the Trias seem to lead to the conclusion that a partial upheaval, at
least, of the rocks on which the Lower Division of the Trias rests, in
Post-Triassic times gave their easterly tilt to the Secondary Rocks,
and probably aided in producing the folds observable in the Culm-
limestones of West Leigh and its vicinity.

To what extent the Triassic beds may have suffered from denu-
ding forces since that elevation, and how far the upheaval of the
older rocks has been Post-Triassic, is a question upon which future
investigations and careful survey will probably throw some light.
The existence of gravel patches upon the older rocks, near the
margin of the Triassic area, in some districts, the fact that material
for the formation of such gravels would be readily furnished by the
disintegration of certain varieties of breccia of the Lowest Triassic
division *, the fact that this gravel actually overlies and is hardly
separable from an outlying patch of brecciated sand near Stoodleigh
Beacon, 6 miles north-west of Tiverton, and isolated by nearly 2
miles of Culm-measure shales and grits from the main mass of the
Trias, and the careful indication of three minute patches of sandstone
(probably of the lower series) on the top of Grabbist Hill near
Minehead, by my friend and colleague Mr. Blake, all lead to the
inference of a very much greater extension of the Devonshire and
West Somerset Triassic rocks before the upheaval which (probably
successively ?) exposed them to denuding forces, and, leaving but
very few traces as beacons of their former site, afforded by their
destruction plentiful material for the formation of subaerial gravels
and thick soils.

Thickness *.

Summing up the outside and very problematical estimates want
of better compel me to offer, we have for—

Upper Wars se nas ees snes aie aceite 1350
Duper Sandstone io... sues i ae
(TOR PIOMOTAGES, fates os how oe hors 100
gwen Wag cts os sap ahewoaspeme hae 600

3580 feet in all,
though 2500 feet appears a more likely estimate.

* Some of these outlying patches of High-Level Gravel have been visited
since the above had gone to press; and I do not now entertain a doubt as to
their Triassic age.

t The different amount of the dips allowed in forming estimates of the
thickness of each division would at first sight appear anomalous in a conform-
able series of beds; but when the different localities from which the estimates
have been made, and the local variations so commonly experienced in the
occurrence of the divisions, are taken into account, the difficulty of obtaining
results from one locality alone will be appreciated.

ROCKS OF SOMERSET AND DEVON. 393

General Deductions.

We find in the area last treated of a tolerably persistent series of
conformable strata of much greater thickness and more variable
composition than that of the Bridgewater district and the other
tracts covered by Triassic rocks south of the Mendips, whilst in the
area north of the Mendips they seldom exceed 200 feet.

Hence we are justified in considering that the area east of the
Quantocks and of Taunton, and south of the Mendips, was dry land
during the deposition of the Breccias, Lower Marls, and Con-
glomerates in the West-Somerset and South-Deyon areas, and did
not come within the influence of the Triassic waters till after the
commencement of the deposition of Upper Sandstones in the latter.
The question as to the correlation of the Devon and Somerset Trias
with that of Lancashire, Cheshire, Staffordshire, Leicestershire, and
Warwickshire yet remains to besolved. Whilst the Upper Marls and
Sandstones of the South-coast and Watchet areas, and their easterly
extension to the Bridgewater area, and that of the former, at any
rate, to the southern flanks of the Mendips, leave no room for doubt
as to their Keuper age, and the dolomitic conglomerate and marls
north of the Mendips are evidently the later deposits of the Keuper
period, we are at a loss to account for the remaining members of
the South-Devon Trias.

Taking an average of 2187 feet for the Keuper in the midland
counties, and 924 feet as the average thickness of the Bunter, from
Prof. Hull’s table of thicknesses in his memoir on the Permian and
Trias (p. 108), we have a total of 3111, or about 460 feet under
the outside estimate allowed for the South-Devon Trias. Of that
estimate of 3580 feet, the Upper Marls and Sandstones, which are
evidently of Keuper age, constitute 1880 feet.

If the remaining 1700 feet be taken as Keuper, we have an ab-
normal development; but if, as appears most likely, the lower
divisions are of Bunter age, the absence of unconformity throughout
the series makes it evident that no break here took place during the
Muschelkalk period, but that that hitherto unfound division must
have its British representative in South Devon. Whether we con-
sider the Pebble-bed division or the Lower series of marls as repre-
senting the Muschelkalk, it is evident that if we consider the Lower
beds as Bunter, it must also be represented.

It would be puerile were I to infer that, because no lithological
equivalent occurs, contemporaneous deposition did not take place,
and would be in the face of the evidence I have endeavoured to bring
forward as to the interchangeable nature of the constituents of single
divisions within a limited area, as, for instance, the marlstones almost
replacing the Upper Sandstones near Bishop’s Lydeard and at
Cothelstone, and the different positions occupied by the sandstones
in the Breccia division in the north and south parts of the area.

As every one knows, the nature of accumulating sediments depends
largely upon the materials whence they were derived, which
is well instanced in the dolomitic conglomerates of the Mendip

394 ON THE TRIASSIC ROCKS OF SOMERSET AND DEVON.

country (where the old rocks were chiefly composed of Carboniferous
Limestone), the prevalence of arenaceous deposits at the base of the
Trias fringing the north-east and south slopes of the Devonian grits
and shales of the Quantocks, the basement breccias and sands of the
West Somerset and South Devon Trias (where the older rocks chiefly
consist of Culmiferous and Devonian shales and grits), and the ad-
mixture of calcareous material in the form of derived fragments or
in the matrix in places where neighbouring exposures of limestone
were favourable for their production; so that we are not only
unable to see any reason why the Muschelkalk should present similar
characteristics in different areas, but fail to see why its repre-
sentative should form a perfectly distinct division.

. Whether the Budleigh pebbles travelled from Normandy or Corn-
wall, their advent seems to mark an epoch in the Triassic history of
Devon, when either by the breaking of a barrier, allowing the in-
cursion of foreign sediments into a salt lake or inland sea, or by a
natural extension of the local sources of derivation, a supply of
foreign material was swept into the area, furnishing the depositing
agents with a different supply from that which had before been and
was afterwards afforded by the local rocks.

In the Geological-Survey memoir on this area the bibliography of
the subject will be gone into, and these few precursory notes will be
modified and strengthened, when additional observations, greater
latitude, and time allow us to enter into details and amplify many in
our possession not alluded to here*.

DIscussIon.

Mr. Ernerinecr remarked that the Rhetics form a most im-
portant feature in Somersetshire. He thought that the ignorance
of the Triassic rocks so common among geologists was to be ascribed
to a great extent to the want of interest attaching to these non-
fossiliferous red beds; and hence our thanks were due to those who,
like Mr. Ussher, would work upon them.

Prof. Ramsay said that Sir Henry De la Beche grouped the whole
of the Triassic rocks together, but nevertheless there was no doubt
that he fully understood their nature. In passing over part of the
country, he had once fancied that some of the lower breccias might
be Permian; but he had never been able to investigate the matter.

The AurHor remarked that if the Lower Breccias were Permian,
the Bunter and Muschelkalk must be also represented, in the absence
of unconformity.

* Should any of my observations tally with those of my predecessors in the
field, I have only to state that the latter part of this epitome has in a still more
abstracted form appeared in the ‘Geological Magazine,’ dec. 2, vol. ii. No. 4,
April 1875, and has been strung together and condensed entirely from my own
notes and observations.

W. KEEPING ON THE DISCOVERY OF MELONITES IN BRITAIN. 395

41. On the Discovery of Mrnonrres in Britain. By Watter KEeEprne,
Esq., B.A., of the Woodwardian Museum, Scholar of Christ’s
College, Cambridge. (Read March 22, 1876.)

(Communicated by Professor Hughes.)

In the appendix to a paper read before the Geological Society, June 9,
1875, I noticed the discovery of a large Urchin, nearly allied to
the American genus Melonites, from the Carboniferous Limestone of
Derbyshire, and preserved in the museum of the Geological Survey.
I have since been permitted to examine the specimen more carefully,
so as to determine it to be a new species of Melonites, a type of Echi-
noids characterized by its numerous ranges of ambulacral plates, which
until now was unknown beyond the boundaries of the New World.

Another fragment, smaller, but well preserved and presenting all
the characteristics of the species, may be seen in the British Museum.

Description of the specimens.—The larger specimen is a confused
mass of stout plates covering an area of 7 inches by 7 inches.
Among these may be seen some broad bands and broken masses of
another set of plates, which are each perforated by a pair of pores.
From the arrangement of this latter system of plates, converging, as
they do, roughly to the centre of the specimen, we see that this is
the remains of one large Echinoid broken up and much disarranged,
the larger scattered plates having formed its interambulacral areas,
the masses of smaller plates its ambulacral zones.

The interambulacral plates (figs. 3,4) are very numerous, and are all
about the same size, which shows that new series of plates were
intercalated to form the increasing circumference towards the equator
of the test, where there must have been as many as eight or nine*
ranges of plates. The plates themselves are very thick, and are
rather irregular in form, most of them being unequally six-sided ;
others are pentagonal, with one face rounded. These latter formed
- the marginal ranges of plates, and they are marked with indenta-
tions for the ambulacral plates, with which they articulated. The
surfaces of all are ornamented with closely set minute tubercles.

The plates were articulated with each other in a very irregular
manner, some of their edges being perpendicular to the surface, while
others are inclined as much as 30°, every intermediate stage being
seen. The marginal ranges share the irregularity so characteristic
of this Urchin; for some of these plates are bevelled off from their
upper edges, others from the lower, so that in some parts of the
test the ambulacral areas overlapped the interambulacrals, in others
vice versa.

The ambulacral areas are in better condition. When the test was
crushed their plates still cohered, and the areas were broken up into

* Supposing, as the specimen seems to indicate, that the circumference of the
test was that of a circle with 33 inches radius, then, the ambulacral areas being
36 millims. broad (see infra), the interambulacral areas were 72 millims. broad ;

and allowing each interambulacral plate 8 millims., there were nine ranges of
such plates at the ambitus.

396 W. KEEPING ON THE DISCOVERY OF MELONITES IN BRITAIN.

masses of various shapes, the larger ones remaining as broad bands
converging to the centre. Their upper surface is distinctly convex
both longitudinally and laterally ; on one border they are festooned
to receive the convex articulations of the adjoining interambulacral
plates ; but the opposite edge is a regular simply denticulated border
(w, fig. 1). I therefore believe that these masses are broken halves
only of the ambulacral areas—the festooned edge being the outer
border of the area and the denticulated edge its middle line. An
examination of the constituent plates confirms this view; for the
pores which they bear are all situated towards that edge of the plate
which is nearest the festooned border.

We have seen that each half of an ambulacral area is convex
laterally ; they are in fact much more convex than the general sur-
face of a test 7 inches in diameter could have been; and they must
therefore have stood out as ten broad ribs on the test, set in pairs
separated by a gentle valley in the median part of the ambulacral
areas, and by the broader planes of the interambulacral areas.

The half ambulacral areas, as here preserved (see fig. 2), are com-
posed of six or seven ranges of small irregular plates, which are
broader than high, thicker than broad, and ornamented with minute
tubercles like the interambulacral plates. The middle ranges are
more regular. Each plate is perforated by a pair of pores, subtri-
angular where best preserved, and situated close to that edge of the
plate which is nearest the festooned border. Although the plates
are so irregular, yet the pores are most of them arranged in lines
and set in shallow grooves running longitudinally upon the areas.
On the internal surface of the test these grooves (or ‘ poriferous
furrows,” as they may be called) are represented by a correspond-
ing set of ridges.

The tubercles form a very uniform granulation over the whole
test, but are best marked on the ambulacral areas. They are im- —
perforate, without boss, and are of two orders: the larger ones, which ~
are most numerous, are surrounded by a smooth areola, bounded by
an elevated ring (fig. 5); contiguous rings are almost in contact.

On the surface of one of the ambulacral masses and elsewhere on
the block some minute acicular spines are abundantly scattered; the
microscope shows them to be sulcated, with a prominent collar at
the base (fig. 6).

The dimensions of the specimen are as follows, viz. :—

Greatest length of specimen ........ =
Greatest breadth

Average length of interambulacral plates .. 8
Average breadth ss oa 6
Average thickness Pd a 34
Length of an ambulacral plate ............ 4
Breadth . oe Py
Thickness ,, 32
Breadth of half ambulacral hers ee eee 19
Length of spine 2.00.2. 242. ¢.. + ae eee 3-6

W. KEEPING ON THE DISCOVERY OF MELONITES IN BRITAIN. 397

Figs. 1-6.—Melonites Etheridgii, sp. n., from the Carboniferous
Limestone of Derbyshire.

Fig. 1. Crushed specimen of Melonites Etheridgii, n. sp., half nat. size (Museum
of Practical Geology).

Fig. 2. The ambulacral mass marked y in fig. 1. The outer festooned border
and the spines are shown.

Fig. 8. An interambulacral plate, showing the bevelled edge and the tubercles.

Fig. 4. One of the marginal interambulacral plates, showing the grooves for
the articulation of the ambulacral plates.

Fig. 5. Tubercles, magnified. N.B. The rings around the tubercles are here
made to look too angular and too prominent.

Fig. 6. Spine, much magnified.

398 Ww. KEEPING ON THE DISCOVERY OF MELONITES IN BRITAIN.

The British-Museum specimen is smaller (4 inches by 3 inches) ;
but it is a fragment of probably quite as large an individual, as we
see from the size of its plates. |

On one side the test has been much disturbed, the whole surface
being covered with a confused mass of interambulacral plates and
a few ambulacrals ; but on the other surface we have a well-preserved
interambulacral area composed of five or six ranges of plates, an
extra range being intercalated between the others as we approach
the equator. The outer edges of the bordering ranges are bevelled
off, showing that they were overlapped by the ambulacral plates ;
these latter are seen confusedly heaped together in two masses bor-
dering the interambulacral area on either side.

The tubercles and spines are well preserved.

It is remarkable that whereas in the larger specimen the ambula-
cral areas are preserved with their plates in position amid the
scattered plates of the interambulacral system, here it is the ambu-
lacral ranges that have given way, while the interambulacrals still
preserve their natural relations.

Metonires ETHERIDGII, 0. sp.

Specific Characters.—Test large (diameter of crushed specimen 74
inches), spheroidal (?), composed of very thick plates arranged in
five ambulacral and five interambulacral areas. All the plates
ornamented with minute tubercles for the support of the spines.

Interambulacral areas broader than (twice as broad as?) the
ambulacrals, composed of numerous (nine?) ranges of plates,
marginal ranges pentagonal, the rest hexagonal, articulating with
each other by faces which vary from a right angle to one of thirty
degrees with the exposed surface. -Ambulacral areas large (14 inch
wide), each consisting of two broad ribs separated by a slight de-
pression along its median line running from mouth to anus; com-
posed of numerous (twelve to fourteen) ranges of irregular plates,
each perforated by a pair of simple pores on its outer margin.

Tubercles minute, imperforate, without boss, of two orders, the
larger kind surrounded by a smooth areola, bounded by an elevated
ring.

Spines small (length 3-6 millims.), tapering, coarsely sulcate,
with a prominent collar round the articular end.

Affinities and Differences.—The only forms which require a close *
comparison with our fossil belong to the genera Melonites and Oligo-
porus of the group Perischoechinide ; with these it agrees in the
thickness of its test, its numerous ambulacral and interambulacral
plates, and in the absence of large tubercles; but Oligoporus has
not more than four ranges of plates in its ambulacral areas.

As a species, this is well distinguished from Melonites multiporus,
N. & O., by the characters of its ambulacral areas: these are com-
posed of twelve or fourteen ranges of plates separated into two zones
by a median depression, and the two middle ranges are not larger than
the others; in M. multiporus, N. & O., they are composed of eight

W. KEEPING ON THE DISCOVERY OF MELONITES IN BRITAIN. 399

ranges of plates, with a prominent median ridge formed by arched
plates which are twice as large as the others.

M. EHtheridgwt is perhaps further distinguished by the size of its
tubercles and spines, no trace of either of these being seen in the
beautiful specimens of MW. multeporus preserved in the British Museum*.
Although some of the plates in this Urchin articulate with each other
by inclined faces, yet I do not believe that its test was flexible, but
rigid, as is proved by the thickness of the plates and the mode of
preservation of the ambulacral areas; therefore it must be classed
in the group of Perischoéchinida tessellatat.

I have named this species after Mr. R. Etheridge, F.R.S., and
take this opportunity of thanking him for that assistance which he
is ever ready to give.

Locality and Formation.—Both the specimens are from the Car-
boniferous Limestone of Derbyshire.

Discussion.

Prof. Morris remarked that certainly Melonites was hitherto un-
known in England. The American species has 8 ambulacral and 7
interambulacral ranges of plates. He thought the specimen was in
too imperfect a state of preservation to show whether the valleys
described by the author as dividing the ambulacral areas really
existed.

The PrestpEnt remarked that the presence of the spines in these
specimens was very interesting, as the spines of Melonites were pre-
viously unknown. He inquired as to the true relationships of this
genus, and whether it approached the Crinoids, and stated that
similar plates were to be met with in abundance in a separate state
in the Lower Carboniferous shales near Glasgow.

Mr. Keeprrne stated that Mr. J. Young had mentioned to him the
existence of traces of Melonites in the Glasgow shales. He explained
further the grounds on which he assumed that the ambulacral areas
showed a double convexity ; and, in reply to the President, stated
that he regarded these Urchins as allied to the Cystoids, and men-
tioned that a Russian form of Caradoc age, described by Schmidt,
is intermediate between the Echinoids and Cystoids, the part de-
scribed by that author as the mouth being really the mark of a
stalk.

* Prof. J. Hall kindly informs me that Melonites multiporus has no tubercles
or spines on the test, the plates being simply pustulose over the entire surface.
+ Perischoéchinida tessellata, as defined by me, Q. J. G. 8. Feb. 1876.

LNs)
Yon

Q.J. GS. No. 128.

400 JOSEPH JOHN MURPHY ON THE GLACIAL

42. The Guactat Cuimate and the Powar Icu-car. By Joseru Joux
Mourrry, Esq., F.G.8. (Read June 21, 1876.)

In a paper “ On the Nature and Cause of the Glacial Climate,’’ in
the Journal of the Society for 1869 (p. 350), E gave my reasons for
thinking that the glacial climate was not one of intense cold, but of
snowy winters and cold summers, with a small range of temperature —
—in four words, not Siberian but Fuegian. I agree with Mr. Croll
that a glacial epoch is one of maximum eccentricity of the earth’s
orbit, and that the northern and southern hemispheres, during such
an epoch, are glaciated alternately. Where I differ from him
is, that while he thinks the glaciated hemisphere has its winter in
aphelion, I maintain, on the contrary, that the glaciated hemisphere
is that which has its swmmer in aphelion. Mr. Croll, im Ins work
on ‘Climate and Time,’ has replied to me; and I propose in this
paper to supplement my former one and give a fuller exposition of
my views, showing also where I think he has fallen into error.

I quote the following from ‘ Climate and Time’ (p. 54}:—

‘* According to the calculations of Leverrier, the superior limit of
the earth’s eccentricity is 0°07075. Lagrange’s determination makes
the superior limit 0°07641. Recently the laborious task of reinves-
tigating the whole subject has been undertaken by Mr. Stockwell, of
the United States. He has taken into aecount the disturbing in-
fluence of the planet Neptune, the existence of which was not known
when Leverrier’s computations were made; and he finds that the
eccentricity of the earth’s orbit will always be imecluded within
the limits of 0 and 0-0653888.”

In order not to take too high a value, and for facility of calcula-
tion where minutely accurate results as to the effect on elimate
are unattainable, I take the maximum eccentricity at 0°069. The
following tabular statement shows the sun’s mean, perihelion, and
aphelion distances, at present, and at maximum eccentricity, the
ratios of the same, and the ratios of heat received by the earth
at each distance :—

At present. At maximum eccentricity.

Sun’s distance in miles.

Means «air Fos 91,400,000 91,400,000

Perihelion .... 89,864,000 85,093,400

Aphelion .... 92,936,000 97,706,600
Ratios of distance.

Hy cies RnR le 1-000 1:000

Perhelton.. ..5.: . 0-983 0-931

Aphelioni ls pie 34. 1:017 1-069
Ratios of heat recerved.

Means 328) sou tneleOO0 1-000

Perthehon i 2. 1:035 1-154

Aphelion 7.234). -0:967 0-875

CLIMATE AND THE POLAR ICE-CAP. 401

The sun’s aphelion distance occurs at present near the mid-
summer of the northern hemisphere; so that if, as I maintain,
glaciation is the result of a cold summer due to the remoteness
of the sun, the glaciation has to be accounted for by the heat
received by the earth at the northern midsummer being less than
at present in the ratio of 0°875 to 0-967, equal to 0-905 to 1-000,
or nearly a tenth part less.

The following attempt to estimate the effect of this difference on
terrestrial temperatures is by the method adopted by Mr. Croll from
Sir John Herschel (¢ Climate and Time,’ p. 37).

The temperature of space is estimated by Pouillet and Sir John
Herschel at — 239° Fahr. The mean temperature of the entire
northern hemisphere for July is estimated by Dove (‘ British-Asso-
ciation Report, 1848) at 71°. The mean July temperature of the
northern hemisphere is ‘therefore 310° above that of space—in
other words, 310° degrees warmer than it would be in the absence
of the sun. If, then, the amount of solar heat received by the
earth is diminished in the ratio of 0-905 to 1-000, the temperature
due to the sun should be diminished in about the same ratio, or
from 310° to 280°-5, say by 30 degrees.

This estimate makes, of course, not the slightest pretension to
accuracy; but it is probably nearly enough true to give an idea of
the scale of the effect. In one way it is much too high. There
would not be time to produce the result: the remoteness of the
sun when at aphelion distance would not have so great an effect
on climate as if his distance were permanently increased. But on
the other hand, we have every reason to believe that Herschel’s and
Pouillet’s estimate of the temperature of space is much too high;
and the greater the difference between the temperature of the
earth’s surface and that of space, the greater will be the effect of
any variation in the sun’s distance.

The hemisphere which has its summer in aphelion has its winter
in perihelion ; and the winter temperature will be raised by the
nearness of the sun about as much as the summer temperature
is lowered by his remoteness; so that the annual range of tem-
perature will be greatly diminished, without necessarily altering
the mean temperature at all. The mean annual temperature will
no doubt be lowered, from causes to be stated further on; but
for the present we will suppose it unchanged. If, then, the mid-
summer temperature is lowered by 30° and that of midwinter raised
by as much, the effect on the annual range will be 60°; and this will
be enough to destroy the present difference between summer and
winter in all except the most extreme climates. According to Mr.
Keith Johnston, jun. (‘Proceedings of the Royal Society of Kdin-
burgh,’ 1868-69), there are only three regions of the earth where
the range is more than 60°: these may be roughly defined as Siberia
with Central Asia, part of North America with Baffin’s Bay and
part of Greenland, and Lapland.

In the following tabular statement I assume the eccentricity to
be at its maximum, and the perihelion to occur at the midwinter of

2¥2

402 JOSEPH JOHN MURPHY ON THE GLACIAL

the northern hemisphere, which under those circumstances I believe
to be the glaciated one. The parallel columns describe the climates
of the world at the two solstices :—

When the Earth ts in , When the Earth is m
Perihelion. Aphelion.
The northern hemisphere .The northern hemisphere
has a mild winter. has a cool summer.
The equatorial region The equatorial region
has its hot season. has its cool season.
The southern hemisphere The southern hemisphere
has a hot summer. has a cold winter.

It is thus seen that the southern hemisphere will have an ex-
treme annual range of temperature, while that of the northern is
very small; and not only so, but the effect of the equatorial hot
season will spread far into the northern hemisphere, making the
winter of at least the lower latitudes of that hemisphere warmer
than the summer. This reversal of the seasons will probably in
no case extend to the pole; in the immediately circumpolar regions
the summer will always be warmer than the winter; but the line
of no annual range, where the temperatures of midsummer and of
midwinter are the same, instead of being, as now, near the equator’,
will be perhaps near the Arctic Circle. If this is the case, the
eccentricity is a little too great for the maximum of glaciation.
The condition most favourable to glaciation will probably be that
where the line of no annual range is at or near the margin of the
polar ice-cap. But if the maximum of eccentricity is too great for
the maximum of glaciation, the degree of eccentricity which will
produce the maximum of glaciation will be attained at some time
while the eccentricity is approaching, and again while it is receding
from, its maximum.

The statements in the above tabular form would probably be ac-
cepted by every one as self-evident, if they applied to a globe having
its surface all land, so that the subject of climate was not compli-
cated by the thermal effects due to evaporation, condensation,
freezing, and melting ; and Mr. Croll in that case would not have
propounded his strange paradox, that the mean temperature of the
whole earth at maximum eccentricity is higher when in aphelion
than when in perihelion. The principles stated in the above table,
however, are not physical, but purely astronomical; and we have
now to trace their physical results.

Round the north pole there is a considerable area, ineluding
Greenland and great part of the Asiatic and American continents,
where the mean temperature of the year is below the freezing-

* See the paper by Mr. Keith Johnston, jun., already referred to.

CLIMATE AND THE POLAR ICE-CAP, 403

point, and where consequently the ground at some depth is always
frozen. Yet so far is this frozen area from being covered with
perpetual ice, that there is no continental ice on any part of Asia
or America; and though the interior of Greenland is covered with
a true continental ice-sheet, yet this descends to the sea only at
particular places, forming glaciers resembling, in all but magnitude,
those which descend into the Alpine valleys. In few if any parts
.of the northern hemisphere does the ordinary line of permanent
snow descend to the level of the sea. The reason of this is, that
the short summer is warm enough to melt away the winter’s snow.
So effectually is this done, that in Siberia trees grow and crops of
rye are harvested over a permanently frozen subsoil.

But let us suppose, what has repeatedly occurred, that during a
period of maximum eccentricity, the precession of the equinoxes gra-
dually brings the midsummer of the northern hemisphere round until
it coincides with the aphelion. The total amount of heat received
from the sun at any given latitude will remain unchanged, but, as
shown above, the range of temperature will be greatly lessened.
Suppose that at the border of the frozen circumpolar area (that
is to say, along the isothermal of 32°) the annual range of tempera-
ture is reduced to nothing, so that the temperature is always freez-
ing, the result will be that over the frozen area the ice and snow
will never melt, and a polar ice-cap will be formed.

In one way this is a little overstated; for in no region is the
climate quite invariable, and even where the mean temperature
of the warmest month dves not rise above 32°, there will no doubt
be some melting. This, however, will have no effect, except slightly
to diminish the extent of the ice-cap at its margin; and it will
probably be much more than counteracted by the spreading of the
ice-cap, from the same causes that make a glacier descend. More-
over the ice-cap will tend to extend itself, in consequence of the
effect of masses of ice in preventing the temperature in their
neighbourhood from rising much above the freezing point. For this
reason the accumulation of ice will depress the temperature, espe-
cially the summer temperature, along the margin of the ice-cap ;
and this chilling effect will be spread. into lower. latitudes by means
of cold currents and icebergs.

The ice-cap, as Mr. Croll has elaborately shown, will, by displacing
the earth’s centre of gravity, draw a greater share of ocean-water to
the glaciated hemisphere; and this will promote glaciation by
diminishing the range of temperature; for the range is always
least in oceanic climates. Meantime the non-glaciated hemisphere
will have a climate of opposite character—a climate of extremes.
This is partly because of the withdrawal of ocean water from it,
which will increase the area of land and make the climate more
continental; but chiefly because of the nearness of the sun in sum-
mer and his remoteness in winter. The instances of Siberia and
North America show that such a climate may produce a vigorous
forest vegetation, which appears to be injured by no severity of

404 JOSEPH JOHN MURPHY ON THE GLACIAL

winter cold; and such was probably the climate of Greenland when
it was clothed with forests.

It is true that a long and cold winter will be favourable to the
formation of ice, by the freezing of water. But no great accumula-
tion of ice, like that of the glacial period, can have been due to this
cause, because the freezing of water is such a slow process that
the thickest ice thus formed does not approach the thickness of a
moderately thick glacier.

Thus the great heat of the perihelion summer will rapidly melt
away the snow which has fallen during the aphelion winter. The
question of the effect of summer heat on glaciation is, practically, to
what height the temperature of the hottest month is able to clear
the mountains of snow. This height is the height of the snow-
line; and I have shown in my former paper, by an appeal to the
facts of physical geography, that this depends chiefly on the tem-
perature of the hottest month.

Humboldt, in his ‘ Cosmos,’ makes the interesting remark, that if
the mountains of the world were high enough, we should see an
upper as-well as a lower limit to the region of perpetual snow. That
is to say, at a very great height the snowfall would be so small that
the snow would disappear by eraporation under the summer sun.
It seems not unlikely that this actually took place during the peri-
helion summer at maximum eccentricity, when the amount of heat
received by the earth in perihelion exceeded what it is now in the
ratio of 1:154 to 1-035, equal to 1:115 to 1:000, or about 10 to 9.

There is one fact of physical geography, which seems at first sight
to support Mr. Croll’s theory that the glacial climate was due to
an aphelion winter, and not, as I maintain, to an aphelion summer.
In the Antarctic regions there is a glacial climate now ; the entire
Antarctic continent is covered with perpetual snow down to the
water’s edge; yet the Antarctic summer is in perihelion. Mr. Croll
thinks this is the normal state of things, and in support of his theory
he states (p. 77) that

‘‘]. The mean temperature of the southern hemisphere is less
than that of the northern.

“2. The winters of the southern hemisphere are colder than those
of the northern.

“3. The summers, though occurring in perihelion, are also
comparatively cold.

‘‘4, The mean temperature of the whole earth is greater in June,
when it is in aphelion, than in December, when it is in perihelion.”

I believe it may be confidently asserted that the first two of
these statements are erroneous. The mean temperatures of the
two hemispheres appear to be very nearly the same: what makes
the great difference in their climates is difference of range of tem-
perature. As compared with the northern hemisphere, the range in
the southern is less ; the summers are cooler, and the winters warmer.
The following tabular statement of the climates of the two hemi-
spheres is from Mr. Hopkins’ s paper on “Changes of Climate” in
the ‘ Quarterly Journal ’ of this Society, vol. viii. (p. 72) :—

CLIMATE AND THE POLAR ICE-CAP. 405°

Lat, 20° N, Lat, 20° S.
Mean io Mean 745
AMOS se se 14 Range 2... 4% 9

Lat. 30° N. Lat. 30° S.
eam os a. a's 68 Mean 69
0 a a 20 Ranpe rs yeig 12

Lat. 40° N. Lat. 40° §

RU GBIIE. cma tiatiers 54:5 Mies’ ele ns ti 53°5
Bante ox. tack. 33 Range eee 9

Lat. 50° N, Lat. 50° S. (Falkland Islands).
Mead oii. oes 39 Mean ect c i 43
SRADED ek atin cee 46 Range 18

We know nothing of the winter temperatures of the Antarctic con-
tinent; but it seems in no way improbable that there also the
range is comparatively small, so that the mean temperatures are
not lower than those of corresponding Arctic latitudes, though the
summer temperatures no doubt are lower. From the fact that
the southern summers are cooler than the northern, it follows that
the mean temperature of the whole earth is lower in the southern
than in the northern summer. All this is generally and satisfactorily
referred to the law that maritime climates are less extreme than
continental ones, in consequence of water taking longer than land
to become heated and to become cooled; and the climate of the
southern hemisphere is on the whole maritime, and that of the
northern continental. But if the relative extent and distribution
of land and sea were nearly the same in the two hemispheres, I
cannot doubt that, contrary to Mr. Croll’s opinion, the southern
hemisphere, having a perihelion summer and an aphelion winter,
would have a warmer summer and a colder winter than the
northern.

If, however, it is hereafter shown that the mean temperature of
the Arctie regions is much higher than that of the Antarctic, this
will be amply accounted for by the influence of the warm currents
of the Atlantic. Mr. Croll has shown what an enormous quantity
of heat they carry northward ; but there are no corresponding cur-
rents in the southern hemisphere. The currents that carry heat into
the North Atlantic and the Arctic Ocean receive their direction, first
from the coast of South America, which diverts the equatorial current
to the north-west, and afterwards that of Europe, which diverts the
Atlantic currents to the north-east*. Mr. Croll is probably right in
supposing that the glaciation of the northern hemisphere would cause
the north trade-wind to blow further south than at present, and that
this would diminish the volume of the Gulf-stream, and so tend still
further to lower the northern temperatures. But this cause would

* See ‘Climate and Time,’ map facing p. 212.

406 ON THE GLACIAL CLIMATE AND THE POLAR ICE-CAP.

not act during the glaciation of the southern hemisphere, because,
in consequence of the totally different disposition of the land, there
are not in that hemisphere any great poleward currents.

My theory of the glacial climate, however, does not come into
collision with Mr. Croll’s on the subject of ocean currents, because
the effect on climate of any change in the currents will be the same
in kind, whether glaciation is caused by an aphelion winter or an
aphelion summer. Such a change in the currents as Mr. Croll
supposes cannot begin to act on climate until a change of climate
has first begun to divert the currents. I think, however, that the
cold of an aphelion summer at maximum eccentricity will amply
account for the glacial climate by its direct effect, without any agency
of ocean currents being needed.

ON THE ROCKS SURROUNDING THE LAND’S-END MASS OF GRANITE. 407

43. On the Metamorputc Rocks surrounding the Lanv’s-Enp Mass of
Granite. By S. Axxport, Esq., F.G.S. (Read June 21, 1876.)

[Prats XXIIT.]

In the following notes I propose to lay before the Society the results
of a microscopical examination of certain metamorphic rocks sur-
rounding the Land’s-End granite. In doing so my principal object
will be to point out the structural and mineralogical changes pro-
duced in clay-slates and certain igneous rocks by the intrusion of a
mass of granite, and also to compare the phenomena of contact-
metamorphism with those produced by other agencies in rocks of
similar character at a distance from the granite.

On referring to the map of the Geological Survey of Cornwall, it
will be seen that in the vicinity of Penzance the granite is sur-
rounded by a belt of altered Devonian slates extending from St. Ive’s
Bay on the north coast to a point a little beyond Mount’s Bay on
the south; they also occur along the coast north-east of Cape Corn-
wall, and reappear in the same direction at Porthmear Cove and
near Zennor. The so-called greenstones occur in four distinct
groups, also round the outskirts of the same mass of granite, viz.
around Penzance and in the other localities just mentioned. In all
cases they are represented as occurring in the altered slates either at
a short distance from the granite or in actual contact with it.

A microscopical examination of both these groups shows at once
that they have all been highly metamorphosed, and that the
alteration varies greatly in character and extent according to the
nature of the rock and its distance from the granite. The change
observable in the so-called greenstones, for example, differs essen-
tially from that produced by the intrusion of the granite among the
clay-slates. In specimens of the former, taken at some distance
from the granite, the alteration consists principally in the formation
of pseudomorphs quite similar to some described by me elsewhere in
the Carboniferous dolerites of Scotland; while, nearer the granite,
similar rocks have been completely decomposed, and hornblende has
largely replaced the original constituents.

The action of the granite on the slates in immediate contact with
it, however, has been far more direct and energetic, the result being
the frequent introduction of new minerals, and the conversion of an
ordinary clay-slate into a crystalline foliated rock.

On approaching either of the granite masses of Cornwall, one
cannot fail to recognize the fact that the clay-slates gradually
assume a different character: they become more and more indurated
and are traversed in all directions by numerous quartz-veins ; they
frequently become more or less micaceous, schorl begins to make its
appearance, and at the junction of the two rocks their slaty cha-
racter has in many cases been completely obliterated. In the
neighbourhood of Penzance the petrological relations of the granite
to the adjacent sedimentary deposits may be readily observed in

408 S. ALLPORT ON THE METAMORPHIC ROCKS

many places. At Mousehole, St. Michael’s Mount, and Cape Corn-
wall, where good junctions are exposed on the shore, there is not
the slightest indication of a gradation from one to the other; the
mass of granite cuts sharply through the slates, and has thrown out
numerous veins, both large and small, which have penetrated them
in various directions. In all such cases the slates have evidently been
greatly altered along the line of junction; and fragments of them
have not unfrequently been torn cff, and are now enclosed in the
eranite. There can be no question, therefore, that the granite here
presents all the characters of an intrusive igneous rock; and it will
presently be seen that its action on the surrounding strata is chiefly
characterized by the development in them of some of the minerals
which constitute its own mass, and that occasionally the original
lamination of the fine sedimentary matter has been replaced by a
very distinct foliated texture. It should be stated, however, that a
decided foliation is restricted to the immediate vicinity of the
granite.

MicroscopicaAL EXAMINATION.

I. Metamorphic Rocks of Sedimentary Origin.

Slates in contact with Granite.

Hand-specimens showing the junction of the two rocks may be
obtained with a little trouble ; and, as was observed in the case of the
larger masses, the altered slates never pass gradually into granite,
but present a distinct line of separation when examined by the naked
eye or pocket-lens. This is well seen in Pl. XXIII. fig. 1, drawn
half the natural size from a polished slab. The dark streaks on
each side of the granite-vein represent folia composed of tourmaline
with a little mica.

Tourmaline-Schist.—Under the microscope a thin slice exhibits,
however, a kind of junction rather difficult to describe in words.
Both rocks then appear to be so intimately associated together as to
form one continuous siliceous mass; they are, in fact, so completely
blended that the line of junction can only be recognized by the pre-
sence of innumerable grains and crystals of mica or tourmaline
which have been developed in the slaty portion of the slice, while
the opposite half presents the usual crystalline granular texture of
granite. This is the appearance presented when the crystals of
mica and tourmaline are moderately large and not too numerous ;
when, however, they occur in very small grains, and thickly crowded
together, the line of junction becomes far more sharply defined.
Pl. XXIII. fig. 2 represents a junction with slate from Mouse-
hole, as seen in a thin slice cut from the same specimen as fig. 1.
The granite has here cut through the slate obliquely to the plane of
bedding ; and this coincides in direction with certain lines of folia-
tion, in which small grains of mica and tourmaline are arranged in
irregular streaks alternating with narrow bands and lenticular
patches of quartz. The folia containing the larger grains consist

SURROUNDING THE LAND’S-END MASS OF GRANITE. 409

almost exclusively of brown tourmaline (dark shade in figure),
together with good examples of its micaceous or chloritic pseudo-
morphs; and the intervening layers are composed of finely granular
quartz, sometimes clear and colourless, but more frequently stained
by a brown colouring matter, also disposed in parallel bands. We
have here, therefore, an example of the conversion of a clay-slate
into a somewhat micaceous tourmaline-schist. Another specimen,
also taken close to the granite, and within twenty yards from the
one just described, is a still more characteristic example of a tour-
maline-schist ; it consists wholly of folia of crystalline granular
tourmaline alternating with others of quartz, as seen in Pl. XXIII.
fig. 3. The entire section is drawn twice the natural size, and
represents accurately the relative proportions of the two minerals.
The tourmaline does not occur in distinct crystals, but appears to
have run in among the grains of quartz, and to be moulded on to
them, as shown in fig. 4, which is a small portion of the same sec-
tion magnified 50 times.

Mica-schist.—In the same locality, and at a distance of a few yards
only, a different bed of slate has been converted into mica-schist.

The mica is a reddish-brown lepidolite, the erystals of which occa-
sionally contain lamine of a pale green colour ; a little schorl also
occurs here and there along the line of junction; and in one slice
there are three small veins of well-crystallized tourmaline showing
the prisms arranged nearly at right angles to the sides.

In a very fine-grained compact schist, also in contact with the
granite, the mass of the rock is crowded with innumerable minute
scales of lepidolite arranged in undulating lines interspersed with
narrow stripes of clear quartz.

The junction of the two rocks is here very sharp and well defined.

A section of the fine-grained schist, at about 15 feet from the
granite, exhibits a most remarkable arrangement of both quartz and
mica, as shown in Plate XXIII. fig. 5. The general appearance is
that of an irregular foliation produced by the linear grouping of nume-
rous red spots, each surrounded by a clear ring, and alternating with
narrow bands of red mica and strings of quartz. In the figure, the
clear spaces are quartz containing either extremely minute specks of
mica, or none whatever; the granular parts consist of numerous
larger flakes, thickly crowded together, with small grains of mag-
netite disseminated here and there among them. In this case the
two constituents are not merely foliated in the usual way, but have
also undergone a further process of segregation into spheroidal and
elliptical nodules. This structure is well brought out in polarized
light ; for the clear crystalline quartz forming the oval rings then
exhibits colours differing from that of the surrounding parts, and
the nodules thus appear to be distinctly separated from them.
Concurrently with this process of rearrangement and partial crys-
tallization of the silica, some of the flakes of mica became crowded
together in clusters towards the centre, while others were at the
same time pressed outwards so as to occupy the angular spaces and
lines between the numerous adjacent nodules. Moreover all the

410 S. ALLPORT ON THE METAMORPHIC ROCKS

nodules are elongated in the direction of the irregular planes of
foliation,—a clear indication, I think, that in this instance meta-
morphism was effected under an amount of pressure sufficient to
interfere with the process of crystallization ; or possibly the nodules
were originally spherical, and were forced into their present form
while the mass was in a somewhat plastic condition.

A singularly interesting example of metamorphism is afforded by
a specimen of altered slate from Botallack, collected close to the
granite, and represented in Pl. XXIII. fig. 6. On examining a
thin slice, the first thing that strikes one is the occurrence of innu-
merable minute black grains of magnetite arranged in rather wavy
lines of foliation, which traverse the section from side to side, and
occasionally bend round small patches of quartz. Then there are
numerous long flat blades of nearly colourless tremolite, many of
which are arranged in radiating groups, while others lie at all
angles to each other; and thickly scattered through the quartzose
matrix there are crowds of small flakes of lepidolite, none of which
are to be found in the tremolite. The most remarkable and inter-
esting point is, that the undulating parallel limes of black granules
pass uninterruptedly from the base through and across the blades of
tremolite, several of which are thus traversed by two or three rows
of granules; and as the flakes of mica also contain them in consi-
derable numbers, it is evident that the black grains must have been
arranged in lines prior to the crystallization of either tremolite or mica.
We have here, I think, the distinct record of two great changes in
the structure of the rock—first its conversion into foliated schist,
and subsequently the development in it of the tremolite. But,
during the process of crystallization of this mineral, the structural
character of the mass became greatly changed ; and the final result
has been the production of a highly crystalline rock in which the
bands of black granules constitute the only vestige of its previous
foliated texture.

There is no doubt as to the hornblendic character of the bladed
prisms ; for, by cutting slices at right angles to each other, there are
seen several excellent transverse sections of the four-sided prism with
opposite angles of 124° 3!.

Cape Cornwall_—Specimens containing junctions of granite and
slate from Cape Cornwall present the same general characters as
those from Mousehole, but also occasionally afford some interesting
varieties of texture. In one thin slice the granite contains many
small crystals of tourmaline irregularly scattered through it, while
parallel with the line of junction, yet still well within the granite,
they form a narrow band composed of numerous prisms, and associ-
ated with them are many plates of white mica; then there is a well-
defined sharp junction with a fine-grained micaceous schist con-
sisting of green, white, and red mica, together with quartz and
tourmaline. Close to a large granite-vein seen on the shore, there
is an imperfectly foliated schist of very similar character to the one
previously described from Botallack: it contains similar lines of
black granules and red mica; but instead of tremolite, there is a clear

SURROUNDING THE LAND’S-END MASS OF GRANITE. 411

mineral which may be andalusite; it gives moderately bright colours
in polarized light; but as it only occurs in granular patches and
irregular plates closely aggregated, I have not been able to make a
satisfactory determination of it.

St. Michael’s Mount.—At the well-known junction of the granite
and slate on the south side of the Mount, a porphyritic granite
occurs in contact with a fine-grained mica-schist of dark-purple
colour, the junction being very sharp and well defined. The granite
contains a few small prisms of tourmaline; the mica is white, while
that in the altered slate is deep red.

A thin slice showing both rocks contains a narrow vein extending
from the granite through the slate at right angles to the line of
junction ; it is filled with white mica and a little quartz, and along
both sides numerous flakes of red mica are thickly crowded together,
being of larger size than in other parts of the slice. Another speci-
men from the Mount is a fine-grained white granite in contact with
a contorted mica-schist, the diminutive crumpled folds consisting of
red mica and small grains of magnetite. The crumpling was pro-
bably not caused by the intrusion of the granite, as the pressure by
which it was effected evidently acted at right angles to the line of
junction. ‘This is the best example I have hitherto observed among
Cornish rocks of contortions on a microscopic scale. The structure,
however, is far more fully developed in other localities; and I will
conclude this portion of the subject with a brief description of two
specimens selected from several, for which I am indebted to Prof.
Hull and Mr. Wylie ; they are from the granitic district of the south-
east of Ireland, where the granite is intrusive in Lower Silurian
slates.

Plate XXIII. fig. 11 represents the metamorphosed portion of
a thin slice showing the junction of slate and granite near Ennis-
corthy, county Wexford. The granite is of medium texture, with
black mica disseminated through the mass. The altered slate is very
fine-grained and of a dark colour; but with a lens there may be seen
numerous small specks of light colour, which impart to it an appear-
ance of foliation.

The black mica of the granite is strongly dichroic, changing from
deep reddish brown to yellow ; the plates are often broken, ragged at
the ends, and the lamine partially separated from each other. The
altered slate is an extremely interesting object. It is an imperfectly
foliated gneiss containing numerous small well-formed crystals of
orthoclase surrounded by innumerable minute fiakes of brown mica
which stream round them in flowing lines; and these two constitu-
ents are enclosed in a quartzo-felspathic matrix. This specimen,
and others in my possession, show that some of the Irish Lower
Silurian slates are still more highly altered than those previously
described. Fig. 12 is a highly contorted mica-schist, also close to the
Wexford granite. The micaceous layers and alternating clear spaces
frequently contain Belonites, which always lie with their long axes
in the ever-varying planes of the contortions. I have omitted them
in the figure, as it is quite sufficiently complicated.

412 S. ALLPORT ON THE METAMORPHIC ROCKS

In this instance, the lateral pressure has reduced the dimensions
of the rock in one direction to about two thirds of its original
volume.

As previously stated, the extreme metamorphism of the Cornish
slates extends to a very short distance only from the granite; indurated
slates traversed by quartz-veins then occupy a much wider zone; and
the rocks at a still greater distance are ordinary “ killas” or clay-
slate, having a more or less distinct slaty cleavage, but differing
greatly in texture and mineral constitution from those just de-
scribed.

As is well known, nearly all the Cornish granite is schorlaceous ;
and it also very frequently contains two varieties of mica, the prin-
cipal mass consisting as usual of quartz and felspar. Now, of these
constituents, felspar is the only one absent from the altered slates ;
it is stated, however, by De la Beche that in other Cornish localities
rocks of gneissic character are found in contact with granite.

The facts observed appear therefore to warrant the conclusion
that the intrusion of the granite has not only altered the structure
of the slates, but has also developed in them some of its own con-
stituents. De la Beche was evidently inclined to adopt this view of
the introduction of mineral matter from the adjoining granite; and
he concludes his observations on the subject with the following
remark :—‘“ It is difficult to conceive that nothing should have
escaped from the hot fluid granite in the shape of vapour into the
pores and cracks of the adjoining disrupted beds, or that, when the
mass was sufficiently cool to allow water to permeate among both
rocks at a high temperature, some modifying combinations were not
then produced”’*,

Summary.

It appears from the preceding investigations that the alteration of
the slates in contact with the granite has been of two kinds, minera-
logical and structural. 1. The minute particles of an ordinary clay-
slate have been replaced by crystalline quartz, tourmaline, and three
distinct varieties of mica, to which tremolite, magnetite (and anda-
lusite ?) have occasionally been added, and in other localities felspar
also occurs.

2. The texture produced by the grouping of these minerals is
various, the two most remarkable being :—A, foliation more or less
perfect, with every gradation from nearly straight parallel lines to
the most complicated contortions ; and B, concretionary, showing a
decided tendency to segregation of both quartz and mica, the result
being a spotted schist.

Minerals produced by Metamorphic action.

The minerals by which the foliated schists are characterized are
quartz, tourmaline, and mica; each of them presents special points

* Report on the Geology of Cornwall, p. 270.

SURROUNDING THE LAND’S-END MASS OF GRANITE. 413

of interest, which it may be convenient to consider apart from the
descriptions of the rocks themselves.

Quartz.—The strata near the granite are far more highly silicated
than those at a distance from it; and there can be no doubt that
much of the quartz has been derived directly from the intruded
rock.

In thin sections, showing the junction of the two rocks, there are
many microscopic quartz-veins extending from the granite across the
slate; they are not fissures traversing both rocks, but commence at
the sharp line of junction, and are seen in the slate only: they are
continuous with the quartz of the granite, and sometimes contain
small quantities of felspar or mica.

A great part of the quartz may nevertheless have been produced
by a more or less complete decomposition of the constituents of the
original sedimentary deposit; the aluminous and other portions
would then be removed, leaving the silica free to rearrange itself
and form the separate folia. That movements of this kind have
taken place, and that the mass was at one time in a plastic state, is
evident from the fact that the quartz was often injected into fine
fissures, and very frequently has enclosed in it erystals of tourmaline
and mica.

One of the most interesting points connected with this mineral is
the number of minute fluid-cavities with which it is so frequently
crowded.

Fluwid-cavities—Fluid-cavities are abundant in the quartz of the
large masses of granite, in that forming the veins, in the schorl rock,
and in the altered siates.

I have examined very many of these cavities with excellent objec-
tives, and have satisfied myself beyond a doubt that there is very
great irregularity in the relative sizes of the cayities and vacuities, °
not only in different parts of the same mass of rock, but even in
single crystals or grains of quartz.

A remarkable example occurs in the granite in contact with
altered slate at Mousehole; in a single grain of quartz there are
several fluid-cavities containing active bubbles, three of which ex-
hibit widely different proportions to the size of the cavities.

There are also not a few cavities which appear to be quite filled
with a fluid of similar refractive power to that containing bubbles.
There are also very many cavities containing cubic crystals in addi-
tion to liquid and vacuity; these are shown by Mr. Sorby to be
either sodic or potassic chloride, and are no doubt preserved in a
saturated solution ; the fluid is probably therefore of a different den-
sity from that in which there are no such crystals. That-the den-
sity of the liquid varies considerably in different cavities is quite evi-
dent, however, from direct observation; for on rotating the stage of
the microscope in order to cause the bubble to move from one end
of a cavity to another, the motion is comparatively rapid in some,
while it is so sluggish in others as to be more like the rise of a
bubble in oil than in water.

Considering these difficulties, and the impossibility of obtaining

414 8. ALLPORT ON THE METAMORPHIC ROCKS

accurate measurements, I find myself compelled to differ from the
views maintained by Mr. Sorby in his well-known and valuable
paper on the “ Microscopic Structure of Crystals,” and which have
been adopted by Mr. Ward in his memoir on the granitic rocks of
the Lake-district, recently communicated to this Society *.

A reference to Mr. Sorby’s paper will show that his calculations
are founded on the following assumptions :—

1. That the cavities were exactly filled with the fluid at the time
the crystals were formed.

2. That the vacuity was produced solely by contraction on cool-
ing; and

3. That the quartz of the various igneous rocks crystallized at
about 360° C., a temperature assumed to be that at which the tra-
chyte of Ponza might probably have been formed under a pressure
of about 4000 feet of rock.

If the first two assumptions were true, it would of course follow
that the relative size of the cavity and bubble ought to be the same
in all quartz crystals found in any small portion of a rock, as,
according to the hypothesis, they must necessarily have been formed
at the same time and under the same conditions. This, however, is
so far from being the case, that it is difficult to find any speci-
men in which such uniformity exists. This difficulty appears,
in fact, to have presented itself to Mr. Sorby, who admits (p. 487)
that “there is sometimes a passage from fluid- to vapour-cavities
as if there had been an alternation of liquid and vapour or gases ;
both of which circumstances would be likely to occur ;” and again,
when describing certain vapour-cavities in the quartz of granite, he
says (p. 488) that they “ gradually interfere with and pass into fluid-
cavities.”

Now the extremely probable occurrence of an intermixture of
fluid and gases ought surely to be regarded as sufficient to invalidate
any argument founded on the presence of a fluid only. It may afford
a satisfactory explanation of the great irregularity which certainly
does exist in the relative dimensions of the bubble to the liquid; but,
as it is quite impossible to decide in what particular cases vapour of
some kind may have been enclosed together with the fluid, it is
difficult to see the value of any calculations based on such data.

An escape from the difficulty, however, is suggested by Mr. Sorby
who says(p. 473):—“‘ In determining the relative size of the vacuities
in fluid-cavities, of course care must be taken not to make use of
such as have caught up bubbles of gas along with the fluid, which is
more likely to have happened with large cavities than with small.
There is also a greater risk of the large coming across flaws in the
crystal, so as to lose fluid. The very minute should, however, be
avoided. It is therefore better to select those of moderate size,
which have vacuities of very uniform relative magnitude in parts
where vapour or gas-cayvities do not occur and the crystal is very
solid.” Now, as to the directions here given, I would observe that
it certainly would be better to avoid such cavities as have caught

* Quart. Journ. Geol. Soe. vol. xxxi. p. O71.

SURROUNDING THE LAND’S-END MASS OF GRANITE. 415

up bubbles of gas, if one only knew how to recognize them; there
appears, however, to be no reason why gases should not have been
mingled with water in various proportions, and have become enclosed
in the cavities, so that the latter were never exactly filled with fluid
only; nor is it by any means clear why large cavities should be
more likely to contain both than the small ones*. As regards flaws
by which fluid may have been lost, it is not likely that an experi-
enced observer would be misled by them or by a partial drainage
caused by grinding down the sections, as suggested by Mr. Ward.
Whatever may be the cause of the discrepancy, it is quite certain,
from an examination of a great number of Cornish rocks, that the
ratio of the bubble to the cavities is not constant in the same rock,
or even in the same crystals, whether the cavities be large or small.
This is a fact which may be readily ascertained without making a
single measurement by any one moderately experienced in micro-
scopic work, and is even apparent in the Tables given by Mr. Ward,
and in the figures in his plate xxx.; it is moreover a fact of pri-
mary importance in these investigations,.and appears to me abso-
lutely fatal to even an approximate estimation of either temperature
or pressure. In conclusion, it can scarcely be necessary to point
out that if it be permitted to select such cavities as have a very
uniform relative magnitude, and also to assume a standard tempera-
ture for the process of crystallization, there.can be no difficulty in
obtaining concordant results as to the amount of pressure under
which the rocks were formed.

Moreover, a mean value obtained from a selected number of
measurements or by rejecting others that appear to be extreme
cases, can obviously be of no value whatever as a basis for exact
calculation.

The supposed exceptions are, in fact, far too numerous to be
ignored; and it will be admitted that no theory as to the origin of
these cavities can be satisfactory unless it includes all the facts
observed.

Tourmaline.—The black tourmaline (or schorl) is usually of a rich
brown colour as seen in thin slices; but yellow, grey, and blue
varieties are not uncommon, and in one specimen of a schorl rock
from the Land’s End blue crystals (indicolite) are far more abundant
than the brown.

The phenomena of dichroism and absorption are strongly marked,
except in the comparatively rare case when a prism happens to be
cut at right angles to the principal axis. On rotating the polarizer
(the analyzer being removed) the colour changes from a rich brown
to pale yellow, and in the blue varieties from blue to a nearly
colourless grey.

Many crystals exhibit a fine blue in one part and brown in
another; and not unfrequently the arrangement of the two colours
appears to be closely connected with the crystalline form and the
mode of growth. This is well seen in Pl. XXIII. fig. 7, in which a
_* It must be remembered that all these cavities are of microscopic dimen-
sions.

Q.J.G.8. No. 128. 26

416 S. ALLPORT ON THE METAMORPHIC ROCKS

transverse section of a polygonal prism exhibits alternating bands of
blue and brown.

Of this arrangement I have observed several examples in “ schorl
rock” from Cape Cornwall; and there are excellent examples of blue
and brown tourmaline in one of the schorlaceous granite-veins near
Mousehole: the central portion consists of the usual schorl rock,
gradually passing into fine-grained granite on each side, the width
of the vein being about 14 inches.

The mode of occurrence and the various crystalline forms of this
mineral present several points of interest. The well-known “ schorl
rock,” which occurs so frequently in the Cornish granite, consists of
a mass of crystalline granular quartz, in which distinct crystals of
tourmaline are often enclosed; frequently, however, the latter has
no definite crystalline form, and then occupies the interstices be-
tween the grains and crystals of quartz, presenting the appearance
of having run in among them in the form of long straggling bands.
In such cases the tourmaline is not always in a granular state; for
some of the long branching arms prove to be portions of a single
crystalline mass, the action of which on polarized light is uniform
over the whole surface. Although the assumption of a regular crys-
talline form may thus be prevented by the presence of other rigid
bodies in sufficient quantity, yet, on consolidation, the molecular
grouping of the elements is precisely the same as though no such
interference had taken place; for, as we have seen, an irregular
shapeless mass may possess the same optical properties as a perfect
crystal. In specimens like the one here mentioned the tourmaline
occurs in comparatively small quantity, while in others it consti-
tutes a very considerable proportion of the mass, and then forms
more or less perfect crystals, in which quartz grains are enclosed.
Between these extremes there is every possible gradation.

It is evident, therefore, that in some cases the quartz solidified
before the tourmaline; for it has impressed its form on the latter ;
in other specimens, however, long prisms of schorl may be seen to
traverse one, or even several grains of quartz, which had evidently
not solidified when the prisms were enclosed. It appears, therefore,
from these facts, that when the schorl: rock separated from the
granite the entire mass must have been im a plastic state, and that,
in accordance with varying conditions, either of the two minerals
might be the first to crystallize.

In the granite itself the tourmaline and quartz present precisely
the same relations as in the schorl rock: sometimes both are inter-
stitial and appear to have crystallized together ; yet not unfrequently
the schorl must have been the first to consolidate.

Small well-formed crystals frequently occur in the quartz of the
granite, of the schorl rock, and of the altered slates, and in many
cases are built up of minute acicular prisms, which project from one
end of the crystal, while the other is perfectly terminated, as seen in
Pl. XXITI. figs. 8,9, & 10. The specimen in which they occur
was broken off close to the granite, and consists of fragments of slate
and granite intermingled with quartz containing tourmaline and

SURROUNDING THE LAND’S-END MASS OF GRANITE. 417

chlorite; the clear quartz is crowded with fine crystals of both
minerals ; and, in addition to individual crystals scattered through
the mass, numbers of small tourmalines are collected together and
form most beautiful microscopic groups of radiating prisms. Being
ne clear colourless quartz, they are exquisite objects in polarized
ight.

Alteration.—Many of the rocks here described contain ex¢ellent
examples of the alteration of tourmaline, and the process may be
observed in various stages.

Chloritic pseudomorphs are formed by the gradual conversion of
a crystal into a green substance which exhibits no bright colours,
but shows the distinctive characters of aggregate polarization, and
under a magnifying power of 200 is seen to consist of a mass of
confused fibrous bundles. In other cases the fibrous structure is
indistinct, and there are vermicular crystals in various stages of
formation.

Other pseudomorphs appear to be micaceous; some of the plates
exhibit distinct cleavage-lines, and they are slightly dichroic, chan-
ging from a light yellowish-green to a darker shade; the absorption
of light, however, is considerable.

These pale green substances show very clearly how greatly the
absorptive property of minerals depends on the dissemination through
them of minute quantities of colouring-matter: many of them con-
tain extremely minute black grains, which form the nuclei of indi-
stinct spots of a barely perceptibly darker shade than the light yellow
parts; the shade deepens, however, as the polarizer is rotated, and
they gradually become dark opaque spots, devoid of any well-
defined edge, but strikingly different from the surrounding parts.

Inthia Mica.—In the altered slates and granite there are three
varieties of mica, which may be readily distinguished from each other
under the microscope ; they occur, however, in such small flakes that
it is impossible to obtain sufficient for chemical analysis or optical
examination in the usual way. 1. A dark brown mica, resembling
biotite in its action on polarized light, changing from light yellowish-
brown to very dark brown, and nearly opaque. 2. Red mica, which
also exhibits dichroism and absorption, but to a far less degree than
the preceding. 3. White mica, in minute scales, having a pearly
lustre. I have examined the three varieties by the following me-
thod :—A few small flakes were carefully collected, moistened with
fluoric acid, and placed on a platinum wire, then exposed to the
flame of a Bunsen’s burner and examined with the spectroscope.
The red lithium-line was quite distinct in all three, but flashed out
more brightly and remained longer in the case of the white mica
than in the others, clearly showing a larger proportion of lithium.
The white mica is probably a typical lepidolite; and I think that the
difference in the absorptive power of the others is due merely to the
presence of a little more ferric oxide in one of them.

2 G2

418 8. ALLPORT ON THE METAMORPHIC ROCKS

Il. Metamorphic Rocks of Igneous Origii.
ALTERED DoLzErtites.

The rocks remaining to be described are the “ Greenstones” of the
Geological-Survey map; a few of them would doubtless be regarded
as diabase or melaphyr by some authors, and sinee the publication
of the geological map they have even been described as nothing but
altered slates. They appear to have been correctly mapped, and for
the most part rightly named, by the officers of the survey, according
to their use of the term greenstone. These rocks vary in colour
from a dark bluish green to dark brownish green, and afford several
varieties of texture. Some specimens from Tolcarn, Penlee Point, and
elsewhere are rather coarsely crystalline; they are not fissile in any
direction, and yield with difficulty to repeated blows of the hammer ;
while other portions of the same masses are either fine-grained or
compact, and possess an imperfect slaty cleavage. The coarsely
crystalline rocks are, as I shall presently show, altered dolerites ;
while some, if not all of the more compact. slaty varieties were
originally fine-grained basaltic portions of the same rocks, but have
been in most cases so highly metamorphosed as to be almost inca-
pable of recognition. These dark-coloured rocks are locally known
as blue or grey elvans, and differ greatly in appearance from the
‘‘killas” in which they occur; their stratigraphical relations are,
however, not quite clear. All the rocks of the district have been
greatly disturbed; and it would require more time than I had at my
disposal to determine whether these masses of trap are contempo-
raneous and interbedded, or intrusive sheets. Probably both occur,
as frequently happens in volcanic districts, and may thus, in either
case, belong to the same period as a whole; but, however this may
be, they are certainly older than the granite. )

A. good example of the coarsely crystalline variety occurs in an
old quarry at Tolcarn, between Penzance and Newlyn, where it may
be seen to overlie the slate; and on ascending the hill, slates are
again found above it. On the opposite side of the road it forms a
bold rock, and also appears in the bottom of the valley, close to the
new church. Thin slices of the least altered portions of the rock
exhibit under the microscope numerous. plates and crystals of augite
in various stages of alteration, many comparatively large crystals of
felspar (for the most part highly metamorphosed), grains of mag-
netite, and many long hexagonal needles of apatite.

These original constituents are enclosed in a more or less clear
felspathic matrix, thickly crowded with minute green and pale brown
belonites, together with flakes of hornblende, which give the general
colour to the mass.

The augite occurs in large grains of irregular shape, and also im
distinct crystals. The central parts are frequently quite unaltered,.
but are usually surrounded by a border of fibrous green hornblende :
sometimes there is merely a central grain of augite remaining; and-
there are numerous examples of the completion of the process by its:

SURROUNDING THE LAND’S-END MASS OF GRANITE. 419

entire conversion into hornblende, Frequently, however, the pro-
duct of alteration is a dull green substance, which is very slightly or
not at all dichroic. In the first sections examined, the rather large
irregular plates suggested the idea that the pyroxenic mineral might
be diallage; but the absence of regular cleavage-lines and the occur-
rence of characteristic forms in some of the smaller crystals render
it far more likely that it is common augite.

A comparison with the beautiful gabbro of Crousa Down (“ dial-
lage rock”) on the ene hand, and with several Cornish dolerites on
the other, also tends to confirm this opinion; I would, however, by
no means venture to assert that diallage is never present in the more
coarsely crystalline parts of the rocks. Although the felspar has
usually suffered an extreme amount of alteration, there is neverthe-
less unquestionable evidence of its presence in the form of rather
large crystals.

As they occasionally penetrate the augite, their original form is
perfectly preserved, and in two specimens of the Tolcarn rock there
are several prisms so slightly altered that they still exhibit a beau-
tiful striation; there can be no doubt, therefore, as to the true
character of the rock.

Besides the minute green belonites and pseudomorphs after augite,
there are numerous larger blades and flakes of dull green hornblende
disseminated through the base, and also filling cavities and veins.

Like all hornblende it exhibits colours in polarized light, and is
dichroic; but the crystals differ in character from those occurring in
diorites and other igneous rocks, nor are they the same as those
forming the true hornblende-schist of the Lizard district for ex-
ample ; they have often the character of actinolite, and are fre-
quently aggregated in radiating groups composed of flat blades of
a bluish-green colour, and not very translucent; occasionally, how-
ever, there are crystals quite similar to those in ordinary hornblende-
schist. .

In one specimen of the Tolcarn rock they fill small veins and
spaces which were evidently cavities in the originally felspathic or,
possibly, glassy matrix.

In many places the rock has been much fractured, and fine cracks
have broken across the various constituents, and have been subse-
quently filled by actinolite, with here and there a little pyrites. One
vein has been filled with clear quartz, in which actinolite has also
been formed.

A careful examination of the mode of occurrence here described
shows conclusively that all the hornblende of these rocks is a pro-
duct of alteration ; and it appears highly probable that a continuation
or modification of the process would produce a typical hornblende-
schist.

Certain portions of these metamorphosed dolerites are, in fact,
half-formed hornblende-schists, which only require the addition of
foliation to render the transformation complete.

In the Tolcarn rock and several others there are many patches of
a brown colour, and only dimly translucent; they have a distinct

420 S. ALLPORT ON THE METAMORPHIC ROCKS

angular outline, and several of them present the characteristic forms
of aggregations of magnetite. Many have the central parts occupied
by granular patches of that substance, which appear to be corroded
round the edges, and pass gradually into the brown matter in such a
way as to indicate that the latter is the product of their alteration.
They must, I think, be regarded as pseudomorphs after magnetite ;
but I have not been able to determine their composition.

Their mode of occurrence is precisely that of magnetite, being
found in the crystalline constituents and in the matrix; in the latter
they are surrounded by innumerable minute flakes of brown horn-
blende or mica, while that at a short distance is usually green—a
fact clearly indicating the diffusion of ferric oxide.

In the rocks just described some of the original constituents are of
considerable size, and have partially preserved their former character
and structure; but in more compact varieties the minute consti-
tuents have been so completely decomposed and obliterated as to be
quite unrecognizable in their present condition.

A specimen of this kind occurred in the mass of rock near Tolearn,
one part of a large block being coarsely crystalline, the other fine-
grained—a circumstance of common occurrence in rocks of this
character.

In a thin slice of the compact portion the most conspicuous con-
stituent is actinolite, with its radiating blades and flakes either fill-
ing veins or thickly disseminated through the mass in small patches,
many of which were evidently cavities. These are everywhere sur-
rounded by an extremely fine granular matrix containing many
comparatively clear spaces. Augite, in small grains, was abundant,
and has been converted into a brown granular substance.

The metamorphism is here so complete that there are no distinct
traces of any crystalline forms; it is, however, quite possible that
such forms may never have existed; for there are compact basaltic
rocks of various geological periods having a minute granular or
crypto-crystalline texture; and as regards the felspar, a rock from
Clicker Tor (to be presently described) and other examples show
clearly that the crystalline form is frequently quite obliterated.

An examination of this specimen only would not have enabled me
to determine positively its igneous origin; I had seen, however, that
it formed part of a crystalline mass of unquestionable character ; and
the following examples not only afford a striking illustration of the
various degrees of metamorphism to which these rocks have been
subjected, but also exhibit every gradation from an obscure speci-
men like the one just described, to others readily capable of recog-
nition.

A specimen taken from an old quarry in Rose Hill, near Castle
Horneck, is a highly altered dolerite intermediate in texture between
the coarse and compact varieties just mentioned. The remains of
the felspar prisms are still traceable ; and there is the same minutely
granular texture; there are many grains of magnetite and of altered
granular augite, together with much flocculent green matter, but
very little clearly recognizable hornblende. The slice is traversed

SURROUNDING THE LAND’S-END MASS OF GRANITE. 421

by a vein partly filled with this green substance, chlorite, and a little
imperfectly crystallized blue tourmaline.

This specimen was originally rather vesicular, as it contains many
well-defined cavities filled with the green substance.

At a short distance from the place just mentioned there is a new
quarry in the same mass; but the rock is in a different stage of
alteration: the flocculent green substance is here replaced by the
usual dull green hornblende ; and the slice is traversed by a quartz-
vein containing numerous blades of actinolite, with several blue and
brown cry stals of tourmaline,

Altered Dolerites from other localities.

It will now be interesting to compare these Penzance rocks with
others having similar petrological relations; and a reference to the
map will show that such may be found on the east and west sides of
the Dartmoor granite.

Tavistock.—At Peter Tavy the rock forming Smear Ridge and its
prolongation to Cock’s Tor, near the granite, is an altered gabbro or
dolerite, some specimens having a coarsely crystalline texture, while
others are fine-grained ; both varieties are rather less altered than
the Penzance rocks, and are therefore important for comparison.

In the coarsely crystalline variety the alteration is of precisely
the same character as that described in the Tolearn rock: the dial-
lage has been partially converted into hornblende, and this substance
also fills veins and former cavities; the felspar, though highly
altered, may be readily recognized; and there are the same pseudo-
morphs after magnetite.

In a fine-grained specimen of the same mass the pyroxene (appa-
rently augite) is very well preserved, but the felspar is completely
decomposed, being represented by pale green pseudomorphs; the
same substance also fills cavities, several of which contain groups of
radiating blades of tremolite.

Brentor, four miles north of Tavistock, presents many of the
features of a volcanic mass, being chiefly composed of purple bedded
ash, together with scoriaceous and compact trap of a greenish-grey
colour.

The latter is an altered basalt, in which the augite occurs in small
well-formed crystals; it has suffered little or no alteration, while
the felspar is converted into pale green pseudomorphs.

The mass of ‘‘ Greenstone,” half a mile north-east of Brentor, is
chiefly ash of the same character as that of Brentor itself. On the
east side of the Dartmoor granite there are also several masses of
the so-called greenstone, of which the following may suffice as an
example.

Hennock, N.E. of Bovey Tracey—This mass and the two adjacent
bands of trap are altered dolerites. Near Hennock, at the end
furthest from the granite, some specimens of the rock have a well-
marked porphyritic texture, many rather large crystals of dull white
felspar being scattered through the mass. Under the microscope

422 Ss. ALLPORT ON THE METAMORPHIC ROCKS

this mineral is seen to be much decomposed, although the original
striation is here and there quite distinct; the augite is but slightly
altered, and many of the crystals are well developed.

The Botter Rock, forming part of the same mass, but nearer the
granite, is rather coarsely crystalline, dark greyish-black in colour,
and without felspar porphyritically enclosed. A thin slice exhibits
quite similar characters to the last, except that the augite is rather
more altered, and there is much more flocculent green matter dis-
seminated through the mass: there are also numerous rather large
crystals of apatite, which appear either as long prisms or perfect
hexagons.

St. Austell_—_aAt a place called the “ Sanctuaries” near this town,
and just outside the band of altered slate surrounding the granite,
the largest mass of ‘‘ greenstone” is also an altered dolerite, some
parts of which are highly instructive. Part of the augite is un-
altered; some of the crystals have assumed a brown granular
appearance, while others have been converted into the hornblendie
substance previously described. Apatite is abundant; felspar cry-
stals are distinct, though much altered; and there is a considerable
quantity of actinolite, filling cavities and the angular spaces be-
tween the larger crystals.

Helston.—The ‘“ greenstone” to the north of Helston occupies
precisely the same relative position to the granite as the rock just
mentioned. The specimens examined, however, are far more highly
metamorphosed, and might almost be described as hornblende-schist;
the original character of the rock is nevertheless apparent from the
remains of felspar and other crystalline forms.

Having described several examples of dolerites occurring among
the altered slates close to the granite, it will now be well to examine
some of those found in the ordinary “ killas,” or clay-slate, at a dis-
tance from the altered band.

Clicker Tor.—This hill is close to the Menheniot Railway Sta-
tion, about three miles south-east of Liskeard; it forms the mass
coloured serpentine on the map of the Geological Survey. It is an
elongated ridge having a nearly east and west direction, and rising
rather abruptly from the surrounding undulating country; its gene-
ral aspect is quite that of an intrusive mass; and an examination of
its relations to the adjacent deposits shows clearly that such is its
true character. At the east end of the hill all the rocks are well
exposed in the railway-cutting ; and in the section exhibited close to
the station Devonian slates are seen on both sides of the central
mass. On the south side the line of junction is nearly vertical,
and the ends of the highly inclined and contorted slates lie directly
against the trap. Under the microscope, a thin slice of the in-
trusive rock exhibits a variegated mass of pale green serpentine
and a nearly colourless substance intimately blended together; |
imbedded in this matrix there are numerous pseudomorphs after
olivine, and irregular plates of unaltered augite together with minute
grains of magnetite scattered here and there through the mass.
The pseudomorphs after olivine are of two kinds, consisting either

SURROUNDING THE LAND’S-END MASS OF GRANITE. 493

of serpentine or the white substance just mentioned; both are
highly characteristic; the crystalline forms are perfectly preserved,
and they are traversed by veins representing the original cracks so
generally formed in this mineral.

In one slice the augite is also greatly altered ; and it is interesting
to compare the two pseudomorphs side by side.

Although the ground-mass of the rock exhibits a confused
amorphous appearance both in ordinary and polarized light, we are
fortunately not left in doubt as to the nature of at least one of its
original constituents. The augite frequently encloses highly cha-
racteristic pseudomorphs after felspar ; some are completely enclosed,
while others are only partially imbedded in it; the mode of occur-
rence is, in fact, precisely the same as that observed in the altered
gabbro of Corstorphine Hill, near Edinburgh, and in several Scotch
dolerites described by me on a previous occasion*. In many cases
the unaltered augite has preserved in the most perfect manner the
sharp edges and angles of the felspar prisms; and whenever the
latter project from the augite, it may be readily seen that both the
enclosed and outlying portions have been converted into precisely
the same serpentinous substance as that forming the ground-mass.
It should also be noted that, in the case of partially enclosed prisms,
only those sides or ends which have impressed their shape on the
augite exhibit a crystalline form, the outstanding portion being quite
undistinguishable from the surrounding mass.

As it would evidently be absurd to suppose that originally there
was no felspar save that enclosed in the augite, there can be no
doubt that the original felspathic matrix has been completely meta-
morphosed, and that we have in the Menheniot rock a highly inter-
esting and instructive example of the conversion of an intrusive
olivine-dolerite into a mass of imperfectly formed serpentine.

Near Torquay in Devonshire there are two masses of eruptive
rock exposed on the shore—namely, in Babbacombe Bay, and on the
south side of Anstis Cove; they are intrusive in Devonian slates and
limestone, and like other rocks of the district are altered dolerites.
The mass in Anstis Cove is in the best state of preservation ; it is
rather coarsely crystalline, of a dark green colour, and consists of
augite, small prisms of triclinic felspar, apatite, and magnetite.
Some of the augite is but slightly altered ; there is a little pale-green
serpentine disseminated through the mass, and a few crystals of
epidote. One part of the rock is traversed by a vein which contains
numerous small crystals of this mineral enclosed in quartz; the
clear quartz crowded with minute perfectly formed crystals of yellow-
green epidote is an exquisite microscopic object. Other portions
of the rock are so highly altered that there is nothing left but a
mass of pseudomorphs.

In Babbacombe Bay there is an excellent section of the rocks,
where the intrusive character of the trap may be well seen. There
are here two principal varieties of dolerite :—one rather fine-grained
and of a uniform grey colour; the other of a lighter shade and por-

* Quart. Journ. Geol. Soe. vol. xxx. p. 557.

424 S. ALLPORT ON THE METAMORPHIC ROCKS

phyritic texture, having conspicuous crystals of felspar scattered
through it. Both varieties are highly altered: the secondary pro-
ducts are serpentine and calcite; and in some specimens there is a
little quartz ; characteristic pseudomorphs after augite are abundant,
and, as not unfrequently happens, the triclinic felspar has suffered
less than either of the other constituents.

Summary and Conclusion.

Microscopical examination shows clearly that the pyroxenic mine-
ral, whether augite or diallage, has frequently been converted into
a hornblendic substance, and that the variety actinolite is found
filling cavities and fissures precisely in the same manner as other
products of alteration.

That the more compact varieties of these metamorphosed dolerites
or basalts should exhibit an imperfect cleavage which is entirely
absent from the coarsely crystalline portions, is quite in accordance
with the facts observed in typical slates. In such rocks it is found
that even a slight change of texture in the beds is frequently accom-
panied by an alteration in the direction of the cleavage-planes; and
it is no unusual circumstance to meet with coarse uncleaved gritty
bands lying between fine slates, and abruptly terminating a plane of
cleavage, which is as suddenly resumed on the other side of the
band.

All the metamorphosed dolerites which occur in the altered Devo-
nian slates have been collected at a greater distance from the granite
than the schists previously described; the metamorphism is of a
different kind, and there is but little evidence of any direct action
on the part of the intruded rock. The alteration that has taken
place appears to be the result of internal rather than of external
action; in other words, it must have been caused by a more or less
complete decomposition and rearrangement of mineral substance in
situ, and not to any great extent by the introduction of new material
from without.

Some of the facts observed in the course of these investigations
are singularly interesting and important from a petrological point of
view.

By an examination of a sufficient number of specimens the pro-~
cess of alteration may be followed step by step from a slight external
change in a single mineral to the production of a perfect pseudo-
morph, and ultimately to the complete transformation of the entire
mass of constituents, the result being the formation of a metamor-
phic rock whose original composition and structure could only be
detected by a study of a good series of the intermediate forms. In
a previous communication to this Society*, I have described a number
of pseudomorphs after augite and olivine, and the frequent diffusion
throughout the surrounding mass of various chloritic, serpentinous,
or other green-coloured products of their decomposition ; but in the
more highly altered rocks of Cornwall, it becomes easy to trace new

* Quart. Journ. Geol. Soc. vol. xxx. pp. 535-542.

SURROUNDING THE LAND’S-END MASS OF GRANITE. 495

combinations of these secondary products, and to follow the gradual
assumption of definite crystalline forms. In the rock from Rose
Hill, for example, one specimen contains a quantity of flocculent
green matter destitute of any definite character, except in one or
two places, where there is evidently an approach to crystallization,
whereas in another piece nearly the whole of the green substance
exhibits the well-marked characteristics of hornblende.

It appears, moreover, that two rocks of similar origin and compo-
sition may follow different lines of metamorphism, and thus become
converted into two widely different substances, the ultimate result
depending upon the nature of the active agent and the duration
of its operation. The Clicker-Tor dolerite, for example, situated
at a distance from the granite, has been converted by aqueous
agencies into a serpentinous rock; while the Penzance dolerites, in
closer proximity to the granite, have been transformed into horn-
blendic rocks differing from each other in texture and state of altera-
tion in accordance with the coarseness or fineness of their original
erystallization.

There remains one important point to be noted. Many of the
metamorphic rocks described in the first part of this paper have
undergone a second series of changes, in which chemical forces have
evidently played the principal part, this subsequent alteration being
indicated by the occurrence of micaceous and chloritic pseudomorphs
after tourmaline, and an alteration (probably hydration) of the mica.

The granite has also suffered similar changes.

The origin of the metamorphic crystalline schists has hitherto been
inyolved in the greatest obscurity ; and it is with the hope of gaining
some insight into the nature of the processes concerned in their
formation, that I have commenced the inquiry with an attempt to
ascertain the nature and extent of metamorphism produced by at
least one powerful agent which must necessarily have brought into
play enormous chemical and mechanical forces.

The results already obtained from a very limited number of ob-
servations afford sufficient encouragement to continue the investiga-
tion ; for it appears not only that clay-slate may be transformed
into mica-schist, gneiss, and tourmaline-schist, but that, among the
more basic rocks, hornblende-schists may be metamorphosed igneous
rocks, some being derived from dolerites or gabbros, while others
are very probably foliated diorites.

In concluding these notes I will refer very briefly to one other
point of interest, namely the origin of the granite. In Cornwall,
neither examination in the field, nor a microscopical study of the
rocks, lends the slightest support to the notion that granite is a
metamorphic rock in any proper sense of the term.

On the contrary, there is the clearest evidence of the former exist-
ence of deep-seated volcanic action; for not only have the slates
been subjected to the violent disturbance and alteration previously
pointed out, but the surrounding country is seamed for miles by vast
numbers of granitic and felsitic dykes (“‘ elvans”); and considering
the enormous amount of denudation to which the entire district has

496 S. ALLPORT ON THE METAMORPHIC ROCKS

certainly been exposed, it must be admitted that we have a sufficient
explanation of the absence of lavas or other superficial products, and
that only the more deeply seated masses of igneous rocks could
remain as evidences of former volcanic action.

As regards the principal masses of granite in other parts of these
islands, it appears to me that the evidence is equally strong in favour
of their eruptive origin. Mr. Judd has shown conclusively that the
phenomena presented by the great granitic masses of the Grampian
Mountains are the exact counterpart of those observed in the lower
portions of the Tertiary volcanic masses of the Hebrides ; and, lastly,
with respect to the granites and syenites of the Lake-district, I
cannot but think that all the favts so clearly described by Mr. Ward
in his recent valuable memoir*, are quite consistent with the vol-
canic origin of these masses. There is evidence to show that the
Cornish slates existed as metamorphic rocks (cleaved and sometimes
contorted) long before the intrusion of the granite. The contact
metamorphism produced thereby extends to a short distance only,
and may generally be distinguished from the other by microscopical
examination, although there will of course be a gradual transition
from the one to the other. Whenever clear indications of this con-
tact metamorphism occur among stratified rocks at a considerable
distance from granitic masses on the surface, it becomes extremely
probable that such masses exist at a very short distance below. This
is known to be frequently the case in Cornwall; and I should be
strongly inclined to infer the existence of similar relations between
the granite and slates of Skiddaw Forest.

Although there is an absence of direct evidence in the Lake-dis-
trict as to the geological period at which the granites and syenites
were erupted, there is apparently nothing inconsistent with the idea
that they may belong to the great series of volcanic outbursts
which occurred during the newer Paleozoic epoch; and it seems
to me highly probable that they may be approximately of the same
age as the granitic masses of Cornwall, Devonshire, and the High-
lands of Scotland.

EXPLANATION OF PLATE XXIII.

Fig. 1. Polished slice showing granite-vein in altered slate, collected close to
junction with the main mass near Mousehole. The slate has been
converted into a tourmaline-schist. Drawn half natural size.

Fig. 2. Portion of thin slice from same specimen as fig.1, showing the junction
of graniteand slate. Theclear spaces are quartz ; and in the granitic
part the light shade represents the felspar; the dark, tourmaline or
mica. In the slate the two shades are tourmaline and mica. See
p. 408. Magnified 10 times.

Fig. 3. Foliated tourmaline-schist close to the granite, Mousehole. See p. 409.
Magnified 2 diameters.

Fig. 4. Small portion of fig. 3, showing the relation of the tourmaline to the
quartz grains. Magnified 60 times.

Fig. 5. Spotted and foliated mica-schist containing a band of granular quartz
(a). Seep. 409. Magnified 10 times.

* Quart. Journ: Geol. Soc. vol. xxxii.

Quart JournGeol.Soc. Vol. XXXIL Pl XXIV).

i Yy |
. -a%
*

}
ed
we

Petia

a
ab
hy i mages pa PS —
*: -githe: Se es

porewehye” ee RY

—~
pete N . s. 4.8

aga
Lee oT:
WS Ce th Fh ER

ain: seepage a BERS om =~

dss

ee

:

> * +4 ae
4 i DB - ,

ee vino

> a 4% eet “f
> . . a a a
* be ts eh “3 v ;
wee’ © . ahd ' j
. We ae 2a a

bi AIRC ot
(Gee. +

) maeer cards Tih.
eee cel METAMORPHIC ROCKS OF CORNWALL. poe

SURROUNDING THE LAND’S-END MASS OF GRANITE. 427

Fig. 6. Altered mica-schist containing radiating crystals of tremolite and bands
of magnetite. See p. 410. Magnified 20 times.

Fig. 7. Transverse section of polygonal prism of tourmaline composed of alter-
nate bands of blue and brown colour. See p.415. Magnified
60 times.

Figs. 8, 9, & 10. Small crystals of tourmaline with one end regularly terminated,
the other formed of slender prisms. See p. 416.

Fig. 11. Gneiss (altered Silurian slate) in contact with granite, consisting of
orthoclase, quartz, and mica, Ballynamuddagh, co. Wexford. See
p- 411. Magnified 10 times.

Fig. 12. Contorted mica-schist (altered Silurian slate), co. Wexford. See
p. 411. The rock has been bent into numerous sharp contortions,
and compressed in the direction of the pressure to about two thirds
its original dimensions.

All the figures, with the exception of fig. 1, have been carefully drawn under
the microscope by the author, with the aid of the camera lucida.

428 H. GOVIER SEELEY ON THE OCCURRENCE OF

44. Norics of the OccuRRENcE of Rematns of a Brrrisn Fossit Zev-
atopon (Z. Wanklyni, Seeley) im the Barton Cray of the
Hampsnire Coast. By Harry Govier Srerry, Esq., F.L.S.,
F.G.8., &c., Professor of Geography in King’s College, London.
(Read June 21, 1876.)

In 1872 my late college friend Dr. Arthur Wanklyn, who had for a
long time devoted himself to a study of the Barton Clay, was so for-
tunate as to obtain, from the Barton Cliff, nearly the whole of the
skull of a zeuglodont of moderate size. The skull was extracted
entire; but the local collectors, in carrying it up the cliff, had the
misfortune to reduce it to fragments. The day before Dr. Wanklyn
was leaving London for professional duties, he desired me to draw up
some notice of the specimen for publication; but the time at my
disposal was too brief for me to attempt a full description of the
whole of the remains, and I contented myself with some memoranda
on the maxillary bones, teeth, and roof of the brain-case. These I
have hitherto kept for myself in the hope that Dr. Wanklyn would
be able to bring the fossil to London, in a more perfectly restored
condition, for fuller study. But the specimen never came; and
since Dr. Wanklyn’s death I have been unable to get any tidings
concerning it. It has therefore seemed desirable to offer to the
Society the notice which I made four years ago; and I do this the
more gladly that it enables me to associate with the species the
name of its discoverer, of whose enthusiasm for science this may
serve as a slight memorial.

Mawillary Bones.

The maxillary bones are imperfectly preserved posteriorly.
They show a length of about 8 inches of the bone on the left
side, and less than 7 inches on the right side. The anterior termi-
nation of the bones is perfect and rounded convexly; and there
they measure transversely across the palate 23 inches from side to
side. Each bone in its alveolar surface is about 2 inch thick; and
the palatal space between the bones 1s at present filled with matrix.
The external lateral surface of each bone is slightly convex from
above downward; but in length the sides of the bones are concave,
because they diverge posteriorly. The alveolar border of the bone
preserves for some distance backward approximately the same thick-
ness from side to side.

The fragment shows indications of five teeth on the left side.
Between the second pair of tooth-sockets the jaw measures 23 inches
from side to side; at the third pair of teeth it is 33} inches
from side to side; and then it widens more rapidly behind. There
is a strong oblique inner alveolar thickening of the jaw behind the
posterior denticulated teeth. These processes are prolonged inward,

REMAINS OF A BRITISH FOSSIL ZEUGLODON. 429

and meet in the mesial line so as to form a flattened slightly convex
palate. On the superior surface these bones form a concave chan-
nel, which widens behind. The median suture [of the palate] is
overlain, for 4 inches, by a fragment of a thin bone #4 inch thick,
which appears to have been 12 inch wide in the middle, and to have
narrowed in front and behind. Posteriorly it extends beyond the
maxillaries. This bone is presumably the vomer. External to it
laterally are the impressions of the premaxillaries, which were
about 2 inch wide where they rested on the maxillaries; but behind
the third tooth the impressions of these bones widen rapidly and
converge towards the median line, as though they met posteriorly.

The Teeth.

The first two teeth in the maxillary bone were presumably simple
and conical crowns, with asingle fang ; from the sockets for these the
teeth are gone. The first socket extends to within less than j inch
of the anterior termination of the jaw ; it is ¢ inch long and 2 inch
broad, is ovate in section, and placed evenly between the inner and
outer sides of the jaw. Between this and the second socket is an
interspace of ~®, inch. This interspace is hollowed out on each max-
illary bone into a shallow hemispherical pit outwardly placed on the
alveolar margin, caused apparently by absorption due to the pressure
from a tooth in the lowerjaw. The second socket is more than an inch
long, and less than 2 inch wide; so that it is more elongated than the
other. On the internal alveolar margin, behind these teeth, runs a
narrow elevated marginal ridge of bone.

The next interspace is about ;®, inch long; towards the third
tooth it is excavated as though by pressure from a tooth in the
lower jaw. The third tooth is 12 inch long, is compressed from
side to side, less than half an inch thick in the middle, and attenu-
ated to a sharp margin in such a way that the angle at which the
inner and outer sides of the tooth meet is less than a right angle.
The anterior and posterior parts of the crown are moderately serrated,
so as to form four small denticles on each side of the larger median den-
ticle which divides them. The denticles are much smaller in front than
behind. The cutting-margin of the denticles of the tooth is faintly
crenulated. Around the base of the crown runs a narrow elevated
cinguloid ridge of enamel, which does not extend mesialiy above the
fangs. The inner surface of the crown above these ridges is marked
with vertical striz on the enamel. The portion of the unenamelled
fangs which projects from the sockets is less than zinch deep. The
height of the crown is less than an inch. The interspace between
the third tooth and the fourth is j inch. The fourth tooth is also
13 inch long. Its crown, though equally high, has a different
aspect ; for the denticles are larger, longer, thicker, and more deeply
divided from each other than in the preceding tooth, so that the
outline of the crests of the denticles of each side is convex from back
to front. As in the third tooth, the posterior denticles are the
larger; and the denticles are similarly crenulated. The cingular

430 H. GOVIER SEELEY ON THE OCCURRENCE OF

ridge is much more developed; the anterior and posterior halves
of it do not meet. Above this tooth the outer bone of the jaw
thickens into a fold.

The fifth tooth is imperfectly preserved ; it appears to be shorter
than the others, somewhat thicker, with the denticles a little better
marked.

The teeth were inclined towards each other from both sides of the
snout, and locked between those of the lower jaw.

There was found with the specimen a remarkable single-fanged
tooth, which may be one of those missing from the empty sockets.
This tooth resembles the canine tooth of a Carnivore. Itislongitu-
dinally ovate in section, and in length recurved, with a tapering
crown and a fang which continues to enlarge for some distance below
the crown and then contracts somewhat towards the base. The
extreme end of the fang is broken away, while its whole lower part
has been greatly fractured and somewhat crushed. The extreme
length of the portion of the tooth which is preserved is 23 inches, of
which the crown covered with enamel constitutes 1 inch. The base
of the crown from back to front, where the enamel terminates,
measures 2 inch, and from side to side in the same line it measures
less than rf inch ; the crown is more compressed from side to side
in front than behind ; and while the anterior margin, in common
with the outline of the whole tooth, is convex, the posterior margin
in the same way is concave, but with a curve which belongs to a
much larger circle than that of the anterior margin. Along these
anterior and posterior margins on the crown runs an elevated ridge
of enamel. ‘The ridges do not extend over the point of the crown,
which is blunt, rounded, and unworn. On the sides of the crown
and towards its base the enamel, which nowhere has a burnished
smoothness, but an exceedingly fine subgranular shining texture,
becomes wrinkled into sharp short vertical folds, which are more
numerous on the back than on the front aspect, and may number
about a dozen on each side of the tooth. They are fine, irregular,
not straight; and sometimes two converge and unite into one in
passing up the crown. The fang where widest measures 3 inch
from front to back. Like the crown, it is more compressed from
side to side on its convex than on its concave side. It is marked
throughout with exceedingly fine longitudinal strie ; and at intervals
among these are faint ridges which are prolongations downward of
the chief ridges on the crown. The pulp-cavity is large.

This form of tooth would yield one of = best distinctive charac-
ters for the species.

The Parietal Bone.

The parietal bone, which is single, has the form usual in Zeuglo-
donts, as far as can be judged from the fragment preserved, though
various small pieces collected with it probably indicate that the
parietal region was longer in this animal than in the American
Zeuglodonts. The anterior suture, with the adjacent part of the

REMAINS OF A BRITISH FOSSIL ZEUGLODON. 431

frontal bone, is well seen, and on the inferior margins are two small
bones which are in the position of alisphenoids ; posteriorly the bone
is very imperfect; on the under side it is traversed by a groove
which widens posteriorly and becomes part of the cerebral cavity.
The parietal rests upon and encases the frontal bone, much in the
way seen among Ichthyosaurs; so that while in the fragment mea-
suring 63 inches in length 43 inches are occupied externally by the
parietal, on the cerebral surface only 24 inches are occupied by that
bone. The sides of the parietal bone are nearly vertical, and nearly
parallel, measuring 2 inches from side to side in front, where they are
about 17inch high. Above this height the sides become rounded, and
converge towards the middle line into an elevated mesial keel, which
rises higher the further it is prolonged backward; so that while the
bone as it stands on its sutural base is but little over 2 inches high
in front, it becomes, where fractured behind, nearly 33 inches high.
In front the ridge is rounded; but the last 1} inch of it preserved
becomes flattened horizontally and widens posteriorly, being at the
fracture 2 inch wide. Concomitantly with the formation of this
ridge the bone appears to widen out from side to side behind, and
the lateral inclined halves of the upper surface exchange their
rounded outlines for a sloping flattened surface. In front the ridge
dies away just behind the suture. The suture is deep and well
marked ; it is somewhat irregular, and penetrates back into the
parietal in the form of an inverted W with long outer arms,
which on the shoulder of the side of the bone contribute to form a
similar uninverted figure, the outer arm of which, in an irregular
line, is prolonged downward and backward at an angle of about 45°
with the basal outline of the bone; so that the parietal extends a
less distance along the side than along the superior surface, where
its termination is bifid.

The Frontal Bone.

The sutural end of the frontal necessarily corresponds closely
with the parietal ; but as the sides of the bone are similarly flat and.
vertical, and the superior surface is horizontal, it results that the
section of the bone at the fracture is nearly quadrate, being rather.
higher than wide. The frontal bone is also notched on the under-
side with a nearly quadrate olfactory canal about half an inch in
section. The angles between the lateral surfaces and superior sur-
face become elevated; and where it is fractured anteriorly the bone
appears to be widening outward from side to side. The horizontal
surface of the frontal continues forward the anterior depression in
height of the parietal, and at its anterior termination the frontal is
but 14 inch high.

Thus the portion of the frontal preserved has extremely thick
walls. The walls of the skull scarcely become thinner in the pari-
etal region ; for the groove which traverses the bones only slowly
widens and deepens. Its surfaces are nearly flat, and, except that
it is relatively deeper, it corresponds closely in form with the external

Q. J. G: 8: Ne. 128. 2H

432 ON THE REMAINS OF A BRITISH FOSSIL ZEUGLODON.

ridge already described; and under the region where the ridge
becomes flattened, the middle of the enlarging cerebral cavity be-
comes flattened, though there are concave excavations on each side.
The suture between the parietal and frontal is deeply impressed on
the cerebral surface. Behind its inferior termination is, on each
side, another suture 13 inch long and about 3 inch above the flat
irregular basal surface of the specimen. The small bone thus indi-
cated is not quite perfect behind, where it is shown by a fracture to
be imbedded between the outer and inner walls of the parietal.
These bones have none of the usual characteristics of the alisphe-
noids, and are probably to be regarded as orbito-sphenoids. They
are about 4 inch wide; and the film of parietal xternal to them is
about 3 inch thick. In front of these bones and internal to them
are concave grooves, which widen in passing forward to more than
% inch, which is the width of the frontal bone on each side of the
groove.

Almost the whole of the skull could be reconstructed from the ma-
terials preserved by Dr. Wanklyn. Among other parts I noticed the
two egg-shaped tympanic bones, remarkable for their regular form.

In no respect does it approach Squalodon, which not only has the
parietal region constructed on a different pattern, but also has the
teeth of different form.

It differs from all known species of Zeuglodon in the shortness of
the interspaces between the teeth, and apparently in the characters
of the premolar teeth, as well as in the shorter form of the skull.
But in the absence of the specimen any detailed comparison with
other species must be deferred.

The parietal and frontal fragment is about the same size as in
Zeuglodon brachyspondylus of Miller, which similarly has the frontal
region flattened, with a sharp crest along the parietal region—which,
however, does not become flattened posteriorly into a narrow table
as in the species described ; nor has the parietal in the foreign species
the folded sutural junction with the frontal of our English specimen.

I would express my thanks to Mr. W. Davis for assistance in
examining the Zeuglodonts in the British Museum, and a hope that
Dr. Wanklyn’s Zeuglodon may eventually be deposited in one of
our National collections.

ON VERTEBRZ OF POLYPTYCHODON. 433

45. On an associated Srrtes “of Cervican and Dorsat VERTEBR= of
Poryprycnopon, from the Campriper Upper GREENSAND, in the
Woodwardian Museum of the University of Cambridge. By
Harry Govier Seretezy, Esq., F.L.S., F.G.S., &c., Professor of
Geography in King’s College, London. (Read June 21, 1876.)

Tre rarity of vertebrae of Polyptychodon in the Cambridge Green-
sand is out of all proportion to the relative abundance of teeth. In
the Woodwardian Museum alone more than a hundred teeth are
arranged in illustration of the dentition of the genus, about half of
which (probably the remains of one animal) were found at a Phos-
phatite washing on the Huntingdon Road. Many hundreds of
teeth have probably been collected, all in a state of black minerali-
zation ; and these can only be referred to several individuals. The
vertebree collected do not appear to be the remains of more than two
individuals ; and these probably represent two species. They consist
of the bones so admirably described by Professor Owen in the
publications of the Palzontographical Society for 1860, which were
found near Haslingfield, and the associated series of remains from
the Huntingdon Road, belonging to a somewhat smaller animal,
catalogued at p. 45 of my ‘ Index to Aves, Ornithosauria, and Rep-
tilia,’ which latter are the subject of the present notice. All the
vertebre: are in a condition of somewhat pale mineralization. I
mention this fact, because it is within my own knowledge that dif-
ferent parts of the same specimen are sometimes differently minera-
lized, and the condition has no importance as indicating a native or
derivative origin of the fossils.

The Atlas and Avwis.

Professor Owen’s figure represents the atlas and axis as being
more excellently preserved than they are, and as showing a clear
separation between the atlas and axis, which is not very definitely
marked in the fossil.

The atlas and axis are preserved in the associated series, and show
no indication of consisting of separate bones—so perfectly are the
osseous elements blended—either where the external unarticular
surface is preserved or where the internal tissue is exposed by
abrasion.

The extreme transverse width of the specimen is 5 inches; its
depth from the middle of the neural canal is not more than 32
inches; so that it is rather broader than the first specimen, and not
so deep. It shows no visible marks of compression, though it may
perhaps have been a little depressed. There are no traces preserved
of the neural arches. :

The anterior surface of the specimen is concavely cupped for the
hemispherical basioccipital; but the cup is considerably abraded,

what remains of it being a circle of not more than 3 inches diameter.
i 2u 2

434 H. GOVIER SEELEY ON AN ASSOCIATED SERIES OF

The superior lateral angles of the atlantal cup show some indica-
tions of large flattened surfaces, which look obliquely forward and
upward and somewhat outward, and encroach on the cup; from
these the neural arch has come away. The inferior and outer ~
corners of the cup are greatly abraded ; but they show no indications
of haying been formed by distinct subvertebral ossifications.

The posterior articular face appears to be smaller. It is flattened,
slightly concave from above downward, and more concave from side
to side. As preserved, the axis-face measures 37 inches in depth
and 4 inches in transverse diameter.

The base of the specimen is slightly convex in vertical outline,
and less than 4 inches in length between the facets on the axis for
the cervical ribs.

The side of the fossil is formed of two areas :—first, a superior
surface 27 inches deep, on each side of the vertebre; these con-
verge somewhat upward towards the neural arch; and secondly,
a smaller inferior impressed surface on each side, 1} inch in
diameter, which looks backward and downward, and gave attach-
ment to the first cervical rib. This area was defined and separated
by ridges from the unarticular side of the centrum above, and from
its’ base beneath.

The length of the specimen from front to back was greater at the
base of the centrum than at the neural arch. Its greatest antero-
posterior measurement is nearly 2? inches, which is more than in
the species called by Professor Owen Polyptychodon interruptus.

The least antero-posterior measurement through the centres of
the articular cups is 12 inch; the antero-posterior measurement
below the neural canal as preserved is rather over 2 inches.

The basal or visceral surface of the atlas and axis is slightly con-
vex in length, slightly concave from front to back, and shows
rounded nutritive foramina at a distance of 2 inches from each
other, and 1 inch from the postarticular border.

Other Cervical Vertebre.

I have arranged the five succeeding vertebre in natural sequence.
They are remarkable for a Pliosauroid antero-posterior shortness,
and a Pliosauroid flatness of the articular surfaces; but they differ
from the cervical vertebre of Pliosaurs hitherto described in haying
a single articular facet on the centrum for the cervical rib, instead
of two facets; and this character is the only one so far as I am
aware by which a short-necked member of the Plesiosaur family
can be distinguished from a short-necked Pliosaur when only
centrums of vertebre are available for comparison. The facets for
the ribs on the side of the centrum gradually ascend the side of the
centrum from the neck backwards, after the manner of Plesiosaurs,
the ribs evidently becoming supported on the transverse processes
of the neural arch in the dorsal region ; so that Polyptychodon is
essentially a big-headed Plesiosaur with no characters at present
discovered to separate it from other Plesiosaurs, except the not very

CERVICAL AND DORSAL VERTEBR OF POLYPTYCHODON. 435

easily defined ones of a big head and the short cervical centrums
described.

It is probable that several cervical vertebre are missing from the
series; and although for convenience I shall number the vertebrx
3rd, 4th, 5th, 6th, and 7th, these numbers must not be taken to indi-
cate the sequence of the vertebre in the neck, for several may be
missing between the axis and No. 3.

The third vertebra has its anterior face in the form of a pentagon,
the median angle being in the middle of the base of the vertebra.
It is nearly 4 inches deep, and 41 in transverse diameter in the middle
of the vertebra. It is apparently much more cupped than the cor-
responding face of the axis. It measures rather more than 12 inch
from back to front in the middle of the centrum ; at the base, as pre-
served, it measures 27 inches from back to front, while at the neural
canal the antero-posterior measurement is 13 inch. The neural canal,
as in the axis, is somewhat impressed; it is not more than 3 inch
wide. On each side of the canal are the facets for the neural arch,
each more than an inch wide. Below these facets are the unarticular
sides of the centrum fully 14 inch deep. Then, below the middle
of the vertebra succeed the impressed facets for the ribs ; they have
a diameter of 12 inch, are more deeply impressed behind than in
front, and extend across nearly the whole antero-posterior length of
the vertebra. Below these surfaces is the external unarticular base
of the centrum; it is 32 inches long, terminates on each side in a
sharp ridge, and is prolonged forward in the middle as though a
subvertebral wedge-bone were blended with its anterior margin.
The posterior surface presents a corresponding concavity on its
basal border. These outlines are repeated in the succeeding neck-
vertebra. The posterior articular surface is more elliptical, smaller,
and flatter, rather more than 4 inches wide, and rather more than
33 inches deep.

The fourth vertebra is badly preserved, but appears to have
resembled the third.

The fifth vertebra increases a little in antero-posterior measurement ;
and the facet for the cervical rib becomes larger, and ascends a little
up the side of the centrum. ‘The articular faces remain flattened ;
but there is now on each a low, flattened, broad, central elevation,
between which and the margin is a concave area. The back-to-
front measurement is everywhere about 2 inches, though at the
base it is a trifle less. The facets for the neurapophyses are placed
a little further forward, so as not to reach within 7 inch of the
posterior articulation; each is an inch in diameter. The sides of
the centrum are worn. The articular facets for cervical ribs, 12
inch in diameter, are so placed that their superior margin is above
the middle of the centrum. The base measures 4 inches in trans-
verse diameter, in which. direction it has become a little more con-
vex, while from front to back it has become rather less concave.

In the sixth vertebra the larger articulation for the rib is so
placed that half is above and half below the middle of the centrum.
The width of the vertebra across these facets is 5 inches ; and though

436 ON VERTEBRZ OF POLYPTYCHODON.

the width of the anterior articular face is 44 inches, its depth is
33 inches. The articular faces become a trifle flatter, and the
eentral elevation rises a trifle higher. The posterior articular face
is 7-sided. The articulations for the ribs are now on very slightly
elevated pedicles, separated by a narrow margin from the anterior
face of the centrum. ‘The facets appear to be flat, nearly circular,
and 12 inch in diameter. The transverse convexity of the base of
the centrum increases, while the antero-posterior concavity becomes
flatter.

The antero-posterior measurement of the seventh vertebra is
23 inches. The facets forthe ribs ascend so high on the sides of the
centrum that they must touch the neurapophyses ; hence this would
be the last cervical. It is 32 inches deep, and as preserved
measures 43 inches from side to side over the facets for the ribs.

The remainder of the 16 vertebre are dorsal ; but they are not well
preserved. In these vertebre the centrum steadily increases in
length. Several have an antero-posterior measurement of 27 inches ;
and one, probably from the middle of the back, is 3 inches long and
4 inches deep. The articular faces, though 33 inches wide and 33
inches deep, look vertically elongated, owing to the centrum being
a little compressed from side to side below the neural arch. The
margins of the articulations are sharp, and the central boss still
remains. The upper part of the centrum leans a little forward as
in Phosaurus.

I entertain little doubt of the specific distinctness of this fossil;
but since the teeth of animals of this kind rarely furnish specific
characters, it will be difficult to prove the separation of the species
from Polyptychodon interruptus till better materials of that species
are discovered; and the remains described from the Cambridge
Greensand by Professor Owen were not so satisfactory as to yield
good characters for comparison.

The British Museum contains a tooth of Polyptychodon from the
Gault. The remainsin the British Museum from the Lower Green-
sand named Polyptychodon present no characters by which I can
recognize their claim to a place in that genus.

H. GOVIER SEELEY ON CROCODILUS ICENICUS. 437

46. On Crocopiius icentcus (Seeley), a second and larger SPECIES
of CrocopitE from the CamBringe Upper GREENSAND, con-
tained in the Woodwardian Museum of the University of Cam-
bridge. By Harry Govier Szetey, Esq., F.LS., F.G.S., &.,
Professor of Geography in King’s College, London. (Read
June 21, 1876.)

Ir is a curious coincidence that the evidence of the fossil now
described, like that of Crocodilus cantabrigiensis, consists of one
cervical and one dorsal vertebra. The vertebre in this species,
however, are associated ; and their striking resemblance to existing
Crocodiles had, I believe, already been recognized by Mr. Walter
Keeping.

The centrum of the cervical vertebra is 2} inches long, and is
divided from the neural arch by a rather deep suture, more than
usually well marked by the neural arch projecting laterally a little
beyond the centrum. In front the procelous articulation is as deeply
cupped as in any recent species. he cup is circular, about 13 inch
in diameter; it was surrounded by a narrow border, now somewhat
worn, so that the transverse diameter of the centrum, which was
about 13 inch, cannot be exactly determined. Its depth appears to
have been as great—though, as the small anterior hypapophysis is
broken away, the depth as preserved is only 13 inch.

The neurapophyses are very strong. As they rise from the cen-
trum they are directed outward and upward, constricted and rounded
anteriorly above the centrum for the passage of the cervical nerve.
The neurapophyses are constricted between the parapophyses and
diapophyses, as well as between the diapophyses and zygapo-
physes. The least measurement of the neural arch from front to
back is rather more than one inch.

The anterior zygapophysial facets are inclined at an angle of about
45° looking upward and inward; they measure about ? inch from
front to back, and about 1 inch from below upward. Only that
on the left side is preserved: itis smooth, its anterior margin hardly
projects in front of the anterior face of ‘the centrum, and its pos-
terior margin is hardly behind the anterior margin of the diapo-
physis. The width between the anterior zygapophyses was about
23 inches. The inner margin of the facets of the zygapophyses
descends below the top of the neural canal, which in front is sub-
triangular. As usual, there is an interspace between the facets, due
to the neural spine being placed further backward; and from the
middle of this area the vertical slightly elevated anterior margin of
the neural spine arises. On the side of the neural arch the zyga-
pophyses laterally curve downward and backward to the diapophyses,
which project prominently. Above the diapophyses and behind the
anterior zygapophyses the sides of the neural arch are considerably
compressed, measuring 13 inch transversely at the shoulders of the

433 H. GOVIER SEELEY ON CROCODILUS ICENICUS.

sides, where a concave area is terminated posteriorly by a ridge
arising from the diapophysis; which runs upward and backward,
presumably to the faces of the posterior zygapophyses, which are
broken away.

The neural canal somewhat excavates the middle of the neural
surface of the centrum. The posterior outlet of the neural canal
appears formed of two pairs of sides—a short pair belt converging
to the middle of the centrum, and a long pair, 1 inch in length,
converging upward and prolonged in the notch between the zygapo-
physes. The posterior aspect of the neural arch below the zygapo-
physes is flattened. The flattened surfaces are 4 inch broad, and look
obliquely outward and backward ; they converge somewhat upward,
and terminate inward.in the rounded margin of the neural canal,
upward in the posterior zygapophysial ridge, and outward and for-
ward in the diapophyses. The length of the neural suture is 12
inch.

The neural spine appears to have been compressed, but is entirely
broken away.

The diapophyses are obliquely ovate, } inch long and 2 inch

deep ; they are situate rather nearer the front than the middle of
the side of the centrum. The transverse measurement through the
diapophyses, as preserved, would be about 3 inches.
. The lateral unarticular part of the centrum is about 14 inch
long. Posterior to this region the centrum terminates in a large
rounded articular ball, 12 inch wide, not quite so deep, and extend-
ing backward for fully 3 inch. The greatest width of the centrum
behind the posterior articulation is 12 inch.

The side of the centrum carries on its upper half, at less than
half an inch behind the anterior articular margin and extending
upward to the aepeal arch, the reniacally ovate parapophysis, which
as preserved is 7%; inch high and 56 inch broad. Its lateral projec-
tion is cae but slightly worn, The width of the vertebra
through the parapophyses is 13 inch.

The sides of the centrum below the parapophyses are slightly con-
cave in length, somewhat flattened ; and towards the middle of the
base of the centrum they form a well rounded median surface, con-
cave in length, which in the anterior half inch is prolonged down-
ward into a small hypapophysis, now broken away.

The measurement from the rounded base of the centrum to the
upper margin of the anterior zygapophyses is nearly 3 inches.

This vertebra is probably the last cervical. It differs from that
of existing Crocodiles in the large size of the parapophyses, in the
distinct anterior notch in the neural arch for the vertebral nerve,
and the perfect convexity of the articular ball.

The dorsal vertebra has the centrum elongated, compressed from
side to side, and well rounded on the under surface. It is 24
inches long, the lateral unarticular part being 13 inch, while the
remainder is occupied by the well rounded, convex, posterior arti-
cular ball. The anterior cup is apparently circular, and measures 14
inch from side to side. The ball is 1} inch from side to side, and

H. GOVIER SEELEY ON CROCODILUS ICENICUS. 439

has an aspect as though placed slightly below its proper position
on the centrum; its external surface is rather concave behind the
ball. The transverse diameter of the centrum at one inch from the
posterior extremity is 1} inch. The neural suture is straight, 13
inch in length. The neural arch is remarkable for its length
and for the small extent to which it is notched in front and behind
for the vertebral nerves, its least length being fully 13 inch. It
rises vertically and continuously with the side of the centrum ; and
at a height of about 12 inch from the base of the centrum gives off
the compressed transverse processes, which are almost entirely broken
away. This process is continuous in front with the anterior zygapo-
physis, the articular facet of which is not horizontal, but looks a little
inward as well as upward; the facets appear scarcely to project
forward in front of the anterior articular margin. The transverse
processes at the base, as preserved, have an antero-posterior measure-
ment of 12 inch. Behind the process is a concave notch, posterior
to which the postzygapophysis is produced; across the notches the
measurement is 1 inch. There is a slight depression below the
notch, bounded in front by a ridge directed outward and forward.
As preserved, the transverse processes, which were horizontal,
measure 27 inches from side to side. The neural arch was greatly
compressed above the transverse platform. The neural spine is frac-
tured at a height of less than 3 inches from the base of the centrum ;
it measures 17 inch from front to back. The neural canal (which
is obscured in front) appears to haye been remarkably small; from
side to side behind, the neural arch measured 1 inch.

This vertebra is the 6th or 7th dorsal. The depression of the
posterior articular ball, and its perfect convexity, are the chief points
in which it differs from existing species.

The remains indicate an animal about 16 feet long.

I have seen no other vertebrate fossil from a secondary stratum
in which the bones preserved so closely resemble those of an exist-
ing type.

_ 440 H. GOVIER SEELEY ON MACRUROSAURUS SEMNUS.

47. On MacrvrosavRts sEMNtvs (Seeley), a Lone rarLep ANIMAL with
Procaxous Verresre from the CamBripge Urrrr GReEEnsanp,
preserved in the Woodwardian Museum of the University of
Cambridge. By Harry Govirr Sretey, Esq., F.LS., F.G.S.,
&c., Professor of Geography in King’s College, London. (Read
June 21, 1876.)

Axout twelve years ago the Woodwardian Museum acquired from
Mr. W. Farren a series of some 25 associated and successive caudal
vertebre, found at one of the deeper phosphatite washings on Cold-
ham Common, Barnweli. At the same date, the Rev. W. Stokes
Shaw, M.A., Caius College, obtained from a similar working at
Barton, a locality a few miles westward, another associated series of
15 smaller vertebre showing identical characters, and of such size as
to exactly join on to the first series and compléte the tail. These
latter vertebrae, not improbably part of the same individual, being
presented to the Museum, I arranged both sets in a continuous
series. Very few appear to be missing in any part of the sequence,
though the extremity of the tail is probably not preserved, and
there are no means of estimating how many vertebra may have
intervened between the last of the sacral region and the earliest
caudal which is preserved. The tail probably included 50 ver-
tebra, and may have reached a length of 15 feet, which would
have amounted to one half the length of the animal if the propor-
tions of modern crocodiles obtained. A few isolated vertebra
have also been collected; but no distinctive portions of the skeleton
have come under my notice. The affinities of the animal are at
present somewhat obscure; for the only available data from which
a determination could be made are the following facts :—The
articulation of the earlier vertebre is proccelous; this character
gradually changes till the articulations of the centrums are
nearly flat; then they become biconcave, and towards the end of
the tail are irregular. There-are no chevron bones; and the
centrum becomes elongated and rounded like a dice-box, after the
pattern of Cetiosaurus and Lelaps. The neural arch in the greater
and earlier part of the tail was supported on pedicles rising from
the centrum ; it was depressed, and devoid of neural spine.

The proccelous character in the caudal region has never before
been recorded, so far as I remember, in combination with an ~
absence of chevron bones in an animal of this size; and though the
tail as a whole is more in harmony with the Lacertian type than
with any other order of true Reptiles, yet we must look to future
discoveries for evidence of the systematic position of the animal to
’ which it belonged. In my ‘ Index to the Secondary Reptiles,’ &e. *,
I classed the animal doubtfully with the Dinosauria. If it is allowed

* Pp. xvii, 45.

H. GOVIER SEELEY ON MACRUROSAURUS SEMNUS. 44]

to remain there, I do not see my way to placing it in any one
of the subdivisions of that group.

The vertebre are in different states of mineralization—some show-
ing no indications of phosphatic infiltration, while in others this
process has gone on to a considerable extent.

The preservation of the specimens is occasionally such that I can-
not feel certain that all are arranged in exactly their true order of
succession. ‘The first vertebra preserved is an early caudal much
decomposed on one side, and considerably abraded, with only the
base of the neural arch preserved, and an indication of the antero-
posterior extent of the fractured transverse process. Its imperfect
preservation is due to the fact that the bone is open and cellular,
and but slightly mineralized with phosphate of lime.

The articular surface of the centrum is deeply cupped anteriorly,
and is more expanded in front than behind, where it forms a large
hemispherical ball. The sides of the centrum are compressed, and
converge towards the ventral surface, where they form a median
rounded ridge. The length of the centrum is 8 inches, and of this
the unarticular side of the centrum measures 53 inches; the depth
from the neural canal to the base of the hinder part of the centrum,
as preserved, is fully 53 inches, and in front was probably more.
The side is smooth, with some longitudinal vascular impressions,
gently concave from front to back. ‘The basal keel is well rounded
from side to side, and gently concave from front to back.

The transverse process appears to have been 37 inches in antero-
posterior extent at its base, where fractured. It was strong.

The neural canal was smooth and narrow.

What are probably pedicles for the neural arch extend to the
anterior border of the cup. This process on the right side is com-
pressed, less than half an inch thick, with an antero-posterior ex-
tent of 23 inches.

The worn fragment of the next vertebra is 74 inches long. On
the third the transverse processes have disappeared. The anterior
cup of this vertebra, as preserved, is 5g inches wide; and the cen-
trum does not exceed that length; but the ball and the margin of
the cup are both broken. The neural canal is more than an inch
wide. On each side of it the centrum gives off strong compressed
pedicles 2 inches in antero-posterior length, half an inch thick,
inclined a little towards each other, and approaching to within half
an inch of the anterior border, as preserved. These pedicles are not
much more than half an inch high, and were probably separated
from the neural arch by a horizontal suture.

The centrums now become rapidly smaller in diameter and flat-
tened on the visceral surface. The articular cup remains as deeply
marked ; but the ball appears to have a depressed, flattened margin
an inch wide around the elevated central boss, as in existing croco-
diles. That numbered 8 is 63 inches long and has the cup 43 inches
deep. No. 11 is 5; inches long, has the cup much less deep, but the
ball is not at all preserved. The centrum has now acquired a dice-
box form. No. 12 is as long, but the centrum ‘is smaller, the cup

442 H. GOVIER SEELEY ON MACRUROSAURUS SEMNUS.

is flatter ; and there is scarcely any trace of a ball. The greatest
diameter of the constricted part of the centrum is 23 inches;
the neural pedicles become more elongated, measuring 2% inches.
After the 13th the centrums get rapidly smaller. The 15th is
distinctly biconcave; the centrum is somewhat compressed later-
ally ; the neural canal is narrower, with a concave channel in the
centrum, margined by shorter pedicles. The 16th is 43 inches long.
The 20th is 4 inches long, has the centrum 23 inches wide pos-
teriorly, and, as preserved, is 23 inches deep. The least diameter of
the centrum where most constricted is 12 inch. The articular ends
are greatly flattened, but slightly concave, as in many Plesiosaurs.
The pedicles for the neural arch remain at one inch from the anterior
margin; the extreme external width across the pedicles is 13 inch ;
the width of the neural canal is } inch; the antero-posterior ex-
tent of the pedicles is 13 inch. Between No. 23, the last of the
Barnwell series, and 24, the first of the Barton series, a few are
probably lost. The Barton portion of the animal is in rather
better preservation, though a few of the vertebre, which have
been washed with the phosphatic nodules in the mill, show curi-
ously how the circumstances under which fossils are collected may
modify their appearance. In No. 24 the centrum is 34 inches long,
and 2 inches deep in front; the anterior articulation is deeply cupped ;
and the posterior articulation somewhat approximates toa ball. The
least diameter of the middle of the centrum is less than 13 inch. The
pedicles are now placed nearly in the middle of the length of the cen-
trum. No, 25 is 32 inches long, with the articular ends 23 inches in
diameter ; they are deeply cupped with a central deeper depression.
The next vertebra has the articular ends much flatter, with a trans-
verse depression which does not appear to result from pressure. No.
28 is 3 inches long, and has pits in the neural canal like foramina
for blood-vessels. No. 30 has the centrum anteriorly deeply con-
cave; posteriorly it is subconvex with a transverse groove. No. 35
is 23 inches long. The posterior articulation is convex with a slight
central depression ; as preserved it is 14 inch wide. The centrum is
compressed from side to side, measuring 2 inch in least diameter ;
the anterior articulation is very irregular. The neural canal is
about Zinch wide. The pedicles are compressed, } inch wide and 1
inch long.

A few neural arches are preserved. They are remarkable for
great antero-posterior extent, compression from side to side, and
absence of a neural spine, the superior margin being concave from
front to back, and only rising two inches above the top of the
neural canal in the deepest specimen. In that example the posterior
zygapophysial facets are preserved. They are 7 inch in diameter,
and are raised like wafers on the inferior margin of the specimen so
as to look outward and downward. The median posterior portion
of the arch is prolonged for some distance behind the facets; ante-
riorly the arch is forked. Further back in the tail, where the arch
is more depressed, the articular facets are lost; but the posterior
process, ovate in section, is still directed for some distance upward

H. GOVIER SEELEY ON MACRUROSAURUS SEMNUS. 443

and backward, and terminates in a rounded end. The length of
the neural surface on a large neural arch is 2¢ inches.

The irregularity of the articulation of the centrums seems
characteristic ; for in a second series of three large vertebrae, two
are proceelous and the third is biconcave.

Fig. 1.—Side view of caudal vertebra of Macrurosaurus semnus,
probably about the 35th, (Natural size.)

Fig. 2.— Anterior view of caudal vertebra of Macrurosaurus semnus.
(Natural size.)

Two or three.small vertebra were also found about 1859 at Barn-
well. One of these (figs. 1 & 2) retains the neural arch, and shows
an indication of the separation between the neural arch and centrum.
Tt is 25 inches long at the superior part of the centrum, and a little
less at the base. The anterior articulation (fig. 2) 1s subconcave and

444 ‘H. GOVIER SEELEY ON MACRUROSAURUS 8EMNUS.

irregular; the posterior is subconvex, with a transverse impressed
groove (fig. 1). The articular margin is somewhat worn ; but on the
base the centrum is somewhat flattened, and on one side posteriorly
there is a faint slight ridge such as might indicate a chevron bone,
had there been any other reason for suspecting such a structure. The
neural arch seen from above (where it is worn) is wedge-shaped, 1
inch wide in front, with the straight sides converging posteriorly
in a distance of 14 inch to $ an inch. The superior surface of the
arch is flattened, and rounds into the sides; it is straight and in-
clined forward ; but the extremities of the processes are broken both
before and behind. In front the height from the base of the cen-
trum is 1Z inch; 14 inch further back the height is 22 inches.
The antero-posterior extent of the pedicles of the neural arch be-
tween the concave notches in front and behind itis 1 inch. The
centrum is 17 inch deep at the posterior articulation, while in the
middle of the neuro-central suture it is 1 inch deep.

In the ‘Annals of Natural History’ for November 1871, I de-
scribed and figured under the name of Acanthopholis platypus the
metapodium of a large animal. As the middle bone is 6 inches
long, and the bones measure 9 inches over their proximal ends
from side to side, and there is no other evidence of bones of Acan-
thopholis reaching a corresponding size, I am inclined to speculate
on the probability of those bones being a part of the foot of
Macrurosaurus, probably the metacarpal bones. If the remains
both belong to the same genus, then Macrurosaurus would probably
indicate a gigantic modification of the Crocodilian type of Dinosaurs.

H. GOVIER SEELEY ON EMYS HORDWELLENSIS. 445

48. On Remains of Emys norpwetiensts (Seeley) from the Lowzr
Horpwett Bens in the Horpwett Curr, contained in the
Woodwardian Museum of the University of Cambridge. By
Harry. Govier Sretzy, Esq., F.L.S., F.G.8., &c., Professor of
Geography in King’s College, London. (Read June 21, 1876.)

By the intervention of Mr. Henry Keeping the Woodwardian
Museum acquired in 1868 some fine Chelonian fragments mine-
ralized nearly black (now arranged on shelf d of case 109)—which
after a little effort I found to reunite themselves into a plastron
from which the xiphoid bones are lost, and a large connected part
of the carapace which comprises the nuchal plate and the two
adjacent marginal plates, the first six neural plates, and portions
more or less perfect of the first five pairs of costal plates. The
marginal plates are all or nearly all lost, haying probably been
washed away by the sea while the specimen was lying on the shore.
Two disconnected marginal bones were collected with the other
remains. .

Mr. Keeping tells me that the horizon of the fossil is about 20
feet below the bed which yields the chief remains of Crocodilus
Hastingie, and about 10 feet above the brackish-water Upper
Bagshot beds, which are seen in the cliff rising westward at an angle
of 3° at Mead End; so that the position of the specimen is low
down in the Lower Hordwell series.

The fragment of carapace as preserved is 9 inches long and 6
inches broad; so that when perfect it probably measured about 12
inches in length and nearly 10 inches in breadth. In length it is
gently inflated, so that in the portion preserved (nine inches) the ~
highest part of the curve rises more than an inch above a base-line
drawn from the ends of the specimen. As is usual, the transverse
section is more inflated ; and in the width of 6 inches the highest
part of the curve rises 1} inch above a base-line drawn from the
two sides. The carapace is impressed with a small subtriangular
nuchal scute, the first and part of the second marginal scute on
each side, the first, second, third, and part of the fourth vertebral
scutes, and parts of the first, second, and third pairs of costal scutes.
I will first give the characters of the scutes, and then describe the
forms of the skeletal osseous plates.

The Scutes of the Carapace.

The nuchal scute (fig. 1, nw) is small, has its margins sinuous, is
2 inch in length, measures finch in breadth behind, and is 7 inch
wide on the anterior margin of the carapace. The first vertebral
scute (v 1) is six-sided and subpentagonal, three of the sides being
in front of the scute, there being a median side behind the nuchal

scute, and two lateral margins in which it joins the first marginal

~~

446 H. GOVIER SEELEY ON EMYS HORDWELLENSIS.

scute on each side of the nuchal scute. It is 24 inches long in the
median line, 24 inches broad at the anterior ier angle (where it
meets the first marginal and first costal scutes), and 12 inch broad
along the sinuous transverse posterior line (in which it meets the
second neural scute). Like all the other scutal areas, both of the
carapace and plastron, it is marked with subparallel impressed con-
centric lines indicative of the intermittent growths of the scutes.
These lines are least distinct on the posterior border.

The first vertebral scute extends transversely beyond the nuchal
plate (which is 23 inches broad) so as to impress the angles of the

Fig. 1.—Carapace of Emys hordwellensis. (One third natural size.)

nu. Nuchal seute. v1, v2, v3, v4. First, second, third, and fourth vertebral
scutes. ¢1, c2. First and second costal scutes.

marginal plates. The scutal area is convex in length as well as
in width; its posterior border impresses the first neural plate at
less than half an inch in advance of its posterior sutural border.

The second vertebral scute (v 2) is nearly square, being about 27
inches long and nearly 27 inches broad. The two pairs of sides,
which are somewhat sinuous, are subparallel; and the area is con-
vex in length as well asin width. Its anterior and posterior borders
cross the first and third neural plates, while its lateral borders cross
the first, second, and third pairs of costal plates.

H. GOVIER SEELEY ON EMYS HORDWELLENSIS. 447

The third vertebral secute (v 3) has subparallel sides, but is longer
than broad, measuring fully 2 inches in breadth and 24 inches in
length. Its anterior and posterior outlines cross the third and
fifth neural plates, behind the middle of the plate in both cases;
and the lateral outlines cross the third, fourth, and fifth pairs of
costal plates at about half an inch from the neural plate adjacent to
each.

The fourth vertebral scute (v 4) is imperfectly indicated by the
bones preserved, but was rather broader than the third.

The first costal scute (¢ 1) in antero-posterior length measures 3
inches; it stretches from the first marginal plate to behind the
middle of the second costal plate. Its breadth is uncertain, as is
that of all the other costal scutes, for every costal plate is imperfect;
but it appears to have been subtriangular and as broad as long.
The second costal scute (¢ 2) measures 24 inches from front to
back. The third pair of costal scutes is imperfectly indicated. The
boundaries between these scutes are all sharply incised rather than
deeply impressed.

The Osseous Elements of the Carapace.

The nuchal plate is six-sided, concave along its front margin,
which is about 12inch wide. It is similarly concave behind, where
it receives the convex anterior margin of the first neural plate, and
is one inch wide. Of the two lateral sutural surfaces on each side,
those which join the marginal plates converge in front, and are each
13 inch long; those which join the first costal plates converge
behind more rapidly, and are each about 17 inch long. The length
of the plate in the median line is 11 inch; its greatest width from
side to side is about 24 inches.

The first neural plate is subovate in outline, being convex in
front and behind and convex at the two sides. Its greatest width
is 13 inch, and its greatest length 1,9; inch. It articulates with
the nuchal plate in front, with the first pair of costal plates at the
sides, and with the second neural plate behind. It is depressed in
its hinder third, as well as impressed by the transverse scutal
margin.

The second neural plate, which is short and broad and gently
convex, has the usual shape. It is concave in front, and an inch
wide where it meets the first neural plate. Here there are lateral
shoulders a quarter of an inch long, which meet the first pair of
costal plates, and widen the bone to 17 inch. The posteriorly
converging lateral surfaces which meet the second pair of costal
plates, measure 13 inch in length. The bone in the median line
is 14 inch long, and is 43 inch wide posteriorly, where it joins
the third neural plate.

The third neural plate has the same general form as the second,
but is longer; and it is depressed behind the middle, but not so
deeply as the first neural plate. It is 13 inch long in the median

Q.J.G.8. No. 128. 21

448 H. GOVIER SEELEY ON EMYS HORDWELLENSIS.

line ; it is 13 inch broad across the shoulders for the second pair
of costal plates, measures nearly 17 inch along the posteriorly
converging lateral surfaces for the third pair of costal plates, and
is }2 inch wide behind.

The fourth neural plate is more like the second, but is eaaly
wider behind. It is smooth and convex, and measures 1,3, inch
in length, is 1,3, inch broad over the shoulders for the third pair
of costal plates, and measures 12 inch along the posterior sides for the
fourth pair of costal plates. ‘These sides do not converge so rapidly
as those of the plates in front; the bone measures { inch from side
to side behind.

The fifth neural plate is subquadrate; it is crossed transversely
behind the middle by the scutal impression. It meas 1 inch
long, 14 inch broad over the shoulders in front, 42 inch along
the lateral border, and 12 inch wide from side to side behind.

The sixth neural plate is transversely oblong, ? inch long, 13
inch broad, and fully an inch broad behind.

These neural plates are less than 7 inch thick; and to each of them
is ankylosed the neural arch of a vertebra. The neural arches are
greatly compressed from side to side, and are expanded a little in
the middle, where two centrums met and gave attachment to the rib.
The neural arches become successively wider from before backward :
this character was probably also seen in the centrums. ‘The fifth
neural arch has the anterior half of the corresponding centrum pre-
served. Itis remarkable for compression from above downwards,
expansion opposite the rib, and constriction in the middle of the
centrum. The rib articulates both with the centrum and with the
neural arch.

The nuchal plate is greatly thickened on the visceral surface, in
a line which corresponds to the transverse scutal impression. In
front of this line the marginal scutes are prolonged on the under-
side of the carapace; behind it the plate is smoothly excavated as
if to form a recess for theanimal’s head. The first epipleural plates
are greatly thickened where they join the marginal plates anteriorly,
and have blended with them a pair of strong greatly elevated ribs
which curve forward. But the second, third, and fourth epipleural
plates show no sign of the confluent rib, except where it rises near
the centrum at a sharp angle as a narrow pedicle compressed from
above downward. ‘The transverse space between the bases of the
first pair of ribs is 13 inch; but the distance lessens from before back-
ward; between the fifth pair it has diminished to 14 inch. This
pair of epipleural elements has the rib slightly elevated.

The costal or epipleural plates are all imperfect. The first pair is
by far the largest. Each is of irregular form ; and, owing to the fact
that it meets the nuchal plate and three marginal plates, its outlines
converge towards both ends. It is 2 inches long in a line with
the suture between the nuchal plate and the first marginal.

The second costal plate widens outward from 12 inch near the
neural plate to 14 inch where fractured. It is ‘marked with a
Y-shaped scutal i impression.

H. GOVIER SEELEY ON EMYS HORDWELLENSIS. 449

Third pair of costal plates hammer-headed at their junction with
the neural plates, where they measure 13 inch in length; but
they become narrower, and where fractured measure only an inch in
length.

The fourth costal plate where it joins the neural plate is
curved backward; it is about 13 inch long, and gets narrower
from within outward. ;

The fifth costal plate has the same form, and is an inch long.

The Plastron.

The plastron (fig. 2) is made up of the usual bones. A some-
what heart-shaped five-sided interclayicle (or sternum, fig. 2, 2.c) is
enclosed in front by two clavicles (c), and behind by two hyosternal
bones (hy), which are followed by two hyposternal elements (hp).
The plastron is more than usually convex from front to back.

Fig. 2.—Plastron of Emys hordwellensis. (One third natural size.)

e. Clavicle. i.c. Interclavicle. hy. Hyosternal. Ap. Hyposternal.

The gular scutes are small, and become narrower from the median
line outward; each measures 1} inch from the median line trans-
versely, and 2 inch from before backward. The postgular or humeral

scutes are large, and have the posterior margin convex, though the
212

450 H. GOVIER SEELEY ON EMYS HORDWELLENSIS.

border is sinuous; they are entirely in the hyosternal bones. They
measure 13 inch in length in the median line; and each scute is
24 inches broad; in the entosternal or interclavicular region the
plates are concave.

The pectoral scutes have subparallel borders; they are 12 inch
long; and each is 32 inches broad. Their posterior border is
entirely in the hyosternal bones.

The abdominal scutes are 3 inches in length ; and each is nearly
square. Their posterior border is 1} inch from the hinder border
of the hyposternal bones. The femoral plates are imperfectly in-
dicated.

The hyposternal plates are 33 inches long; and each is 28 inches
broad to the femoral notch, and 3 an inch thick where it joins the
xiphisternal bones behind.

The hyosternal plates measure 4 inches in extreme length from the
margins where they join the clavicles to the hyposternal suture; but
they measure only 13 inch in the median line. Lach is 23 inches
broad at the humeral notch, and 33 inches broad at the lateral
suture.

The interclavicle is 24 inches broad and 13 inch in length. It
is rounded in front, and has two straight anterior surfaces for union
with the clavicles ; posteriorly the bone is convex; and this margin
is received into a concavity formed by the hyosternal bones.

In front the clavicles form a concave border 27 inches long; this
border terminates laterally in a sharp angle, behind which the bone
extends laterally on each side for 1? inch. Every scutal area is
marked with concentric lines of growth.

The only Emydian remains from Hordwell hitherto noticed are
a few isolated plates figured by Prof. Owen in 1849, in the Palzon-
tographical Society’s Monograph of the fossil Tertiary Reptilia
(pl. xxiv.), and the hyosternal and the hyposternal bones of a large
Chelonian plastron (pl. xxvil.), named Hmys crassus. These latter
remains are briefly noticed on the last page of the monograph, and
are said to be remarkable for their large size and great thickness.
The hyosternal bone is in the British Museum. And the hyosternal
bone in the species now described differs from it not only in these
points, but in the form of the plate, which is more rounded anteriorly,
and in the positions of the scutal impressions.

The most distinctive characters by which this species is separated
from all others, recent or fossil, are:—the broad, short, gular scutes
with sinuous sutures; the subtriangular nuchal scute; the sub-
pentagonal first vertebral scute, broader than the succeeding
quadrate vertebral scutes; and the concentric ornamentation left on
the carapace and plastron by all the scutes,

ON BEDS OF DRIFTED COAL NEAR CORWEN, NORTH WALES, 45]

49. On the Move of OccurrENcE and Dertvation of Beds of Drirrep
Coat near Corwen, Norra Wares. By D. Macxrytosu, Esq.,
F.G.S.* (Read June 21st, 1876.)

Asovr four years ago the late Mr. Jones, of Angharad, directed my
attention to the remarkable fact that beds of drifted coal are to be
found in sand and gravel around Corwen, at a great distance from
any known coal in situ. The belief that this coal was of local deri-
vation had led to unsuccessful mining operations in the Wenlock
rocks of the neighbourhood.

Positions occupied by the Drifted Coal.—Around Corwen the lowest
drift consists of yellowish (occasionally bluish) clay, irregularly
alternating with beds of coarse gravel. Many large boulders are to
be found at or near its base; and still larger boulders occur at or
near its summit. The former are often rounded and local ; the latter
subangular, or angular, and erratic (chiefly from the Arenig hills to
the west). Above this drift there is a deposit of clean sand and
fine gravel, which, in other parts of the Dee valley, often graduates
into very coarse gravel with boulders. In many places around
Corwen this deposit is covered by an irregular kind of brick clay.
The coal is generally found in the sand and fine gravel. Imme-
diately to the east of the mountain-limestone quarry (about a mile
and a half to the west of Corwen), a roadside section shows fine
gravel and sand with streaks of decomposed coal. In the Dee
valley, some distance east of Corwen, and not far from Carrog station,
considerable quantities of coal may be seer imbedded in sand in the
railway-cutting. I was assured that coal may be found in many
other places around Corwen; but the principal instance occurs to
the north of the town, and near to where the Denbigh railway
crosses the river Dee. In the deep railway-cutting on the north
side of the river there are several beds of coal in forms varying
from large lumps to fine dust. At the time when the cutting was
made the coal kept a temporary smithy going for about a month.
The outcrop of the coal-beds may still be seen after removing a
facing of sand. The fine gravel associated with these beds, so far
as I could see, is almost entirely made up of local micaceous Silurian
grit ; and none of the small stones show any decided traces of glacial
action, in this respect differing from the boulders in the underlying
drift, which are often intensely striated. Northward from this
cutting, in the direction of Denbigh, I could find no indications of
coal in the sand and gravel; and this seemed to suggest that the
coal could not have been transported from the Vale of Clwyd.

* This paper is intended as a supplement to one entitled “Additional
Remarks on Boulders, &.” which appeared in ‘ Quart. Journ. Geol. Soe.’ for
December 1874, vol. xxx. p. 711. Since that paper was written, I have traced
the dispersion of felspathic boulders from the Great Arenig, in a south-easterly
direction, across Bala Lake, and have likewise found Arenig boulders around
Denbigh ; so that they must have radiated over the fourth of a circle.

452 D. MACKINTOSH ON BEDS OF DRIFTED

Derwation of the Drifted Coal.—In order to arrive at a satisfactory
conclusion on this subject, it may be necessary to take into account
the direction of the glaciation in the neighbourhood of Corwen.
On the north slope of the hill to the south of the town, after care-
fully eliminating a number of structural furrows on rock-surfaces,
which could be traced to where they ran wnder overlying strata, I
found true glacial striz coinciding with the rounded sides of rocky
projections pointing to about W.S8.W., or approximately in the
direction from which the boulders of the district were transported.
This at first led me to suppose that the coal had been drifted from
W.S.W.; but this involved the supposition that a portion of the coal
formation in situ must have been faulted down * on the top of the
remarkable outlier of Mountain Limestone which occurs about a mile
and a half west of Corwen, and that, as the limestone strata dip
between E.N.E. and N.E. at an angle of about 45°, the coal em situ
must have remained in a depression (now covered with drift) on the
east or dip side of the limestone until the glacial period, if not
until the present time. After corresponding with Mr. Aveline (who
formerly surveyed this part of Wales, and who informed me that he
regarded the above theory as very improbable), and after reconsidering
the subject, I became convinced that the drifted coal around Corwen
must have come from the Cefn and Ruabon district, about 12 miles
east of Corwen in a straight line, but at least 20 miles following the
sinuosities of the valley of the Dee. The height above the sea of
the drifted coal in the cutting north of Corwen is about 550 feet,
which is not high enough to preclude the idea of its having found its
way from the above district; and this idea is corroborated by the
similarity of the Corwen to the Ruabon coal +.

Mode of Transportation of the Corwen Coal.—As no boulders,
striated stones, or other traces of glacial action are to be found
immediately associated with the drifted coal under consideration, it
would seem that it must have been transported during a compara-
tively temperate interglacial period when very little ice floated on
the surface of the sea, but still sufficient ice to carry lumps of coal
(though not large fragments of rock) as far as Corwen. The limi-
tation of the Corwen coal to a low level suggests the idea that at
the time of its transportation the land was not submerged beyond
a few hundred feet, and that the coal was floated along the sinuosi-
ties of the valley of the Dee, until, arriving at shallow water in the
Corwen area, it stranded, and became imbedded in sand by sea-
currents or waves. It is not necessary to suppose that there was
any thing more than a tidal current flowing up the Dee valley at the
time when the coal was floated.

Principal Bearing of the subject on Glacial Greology.—tIf the fore-
going be a correct explanation of the derivation of the Corwen coal,
it follows that during the glacial (including interglacial) period, the

* The great Bala fault, which divides the whole of North Wales into twa
parts, runsthrough the Vale of Corwen.

+ In the drift around Ruabon and Wrexham there is a considerable quantity
of coal débris, showing that at least a local drifting of coal must have occurred.

COAL NEAR CORWEN, NORTH WALES. 453

transportation of débris may at times have taken place in a direction
almost diametrically opposite to that of the glaciation of a district.
But it would not be difficult to bring forward many instances in the
north-west of England and Wales, of the interweaving of drifts from
points of the compass ranging from 90° to even 180°. This inter-
weaving, however, is generally of very limited extent.
Postscript.—Since this paper was written I have unexpectedly
had occasion to visit the Corwen district a second time. I reexa-
mined the coal-beds in the railway north of Corwen, and in the
fine gravelly matrix of one of them found a considerable variety of
rock-specimens, in this respect differing from the coarser gravel I
had previously observed. ‘The gravel, in all probability, was partly
worked up out of the underlying or adjacent Boulder-drift : but such
could not have been the case with the angular lumps of coal ; for,
while the waves would have rounded them or reduced them to
powder, no trace of coal (so far as I am aware) has yet been found
in the Boulder-drift. A little north of this spot, Mr. Jones, toll-
house keeper, has lately found coal in still greater abundance,

454. R. ETHERIDGE, JUN., ON A LOWER-CARBONIFEROUS PRODUCTUS.

50. On an ADHERENT Form of Propuctus and a SMart SprRiFERINA
from the LowrrR Carzonirerous Limestone Grove of the East
of Scottanp *, By R. Erueriver, Esq., Jun., F.G.S. (Read
June 21, 1876.)

[Puares XXIV. & XXV.]

I. On an ADHERENT Form or PrRopvctvs.

1. Introduction.—The habits and mode of life of the Product
have been a fruitful source of speculation to the various authors
who have written on the genus. Some writers contended that the
various species lived a free and independent life; others, on the
contrary, conceived that they were attached by muscular fibres
passing from the interior of the shell, and in fact acting the part of
a byssus, whilst yet others believed that their position during life
was at least aided by the long tubes or spines which usually orna-
ment their surface.

I have now much pleasure in announcing the discovery (by Mr.
James Bennie, of the Geological Survey of Scotland) of an adherent
form of Productus, which, although not an entirely new fact in con-
nexion with this genus and its allies, has not, so far as I am aware,
been observed in so complete a manner before. Notwithstanding
that the facts now brought to light will not warrant us in applying
the principle to the whole of the Producti, they yet conclusively
prove that at least one form spent a very considerable portion if
not the whole of its existence adhering firmly by its spines to ex-
traneous bodies.

The specimens were met with by Mr. Bennie in the Lower
Carboniferous Limestone group at two localities in East Lothian, at
one of which the form occurs rather abundantly, as the collections
of the Geological Survey of Scotland and Museum of Practical
Geology now collectively contain at least one hundred and forty
individuals.

2. Bibliography.—Before proceeding to describe the specimens, it
will be as well to give a brief digest of the principal views held by
previous writers on the mode of life of Productus and the use of the
spines scattered over the valves of many of its species.

1828. The Rev. Dr. Fleming considered it “ probable that all the
Producti and Pentameri were free shells” }.

1831. Baron von Buch, under the generic name of Leptena,
described a shell resembling a Productus, with annulated tubes along
the hinge-line on each side of the umbo, and communicating by holes
placed under the hinge-line with the interior of the shell, and

* Communicated with the permission of the Director-General of the Geolo-
gical Survey of the United Kingdom.
+ British Animals, p. 380,

R. ETHERIDGE, JUN., ON A LOWER-CARBONIFEROUS PkODUCTUS. 455

through which, he considered, passed small tendons of attach-
ment *.

1831. M.G. P. Deshayes, in giving the characters which separate
the genera Terebratula and Productus, drew attention to the absence
of any foramen in the beak of the latter, the resulting want of a
peduncle, and probable free condition of the shell +.

1832. The same author, in his continuation of Bruguiére’s ‘ En-
eyclopédie Méthodique,’ stated that Productus showed no trace of
attachment, and lived a free life after the manner of a large number
of the Acephalous Mollusca t.

1835. The Rev. A. Sedgwick and Sir R. I. (then Mr.) Murchison,
in their paper “ On the Geological Relations of the Secondary Strata
in the Island of Arran ”’§, describe Producta scotica, Sow. (=P.
semireticulatus, Mart.) as occurring in the Carboniferous Limestone
of the Corry Caverns so abundantly “as entirely to form the lower
layer of many of the beds, being arranged very symmetrically in the
exact position of the living shell [misprinted shale] with their con-
vex valves downwards.”

1836. M. G. P. Deshayes, in the second edition of Lamarck’s
famous work ||, states that from the absence of spines in some Pro-
ducti he was disposed to abandon his opinion regarding the use of
the tubes or spines as a passage for tendinous appendages (appendices
tendineux) of attachment, replacing the peduncle of Terebratula 4.

1838. M. von Buch, in his ‘ Essai d’une Classification et d’une
Description dés Térébratules ’**, divides the Brachiopoda into two
groups. In the non-perforate division of the second section of the
first group, or those with the point of attachment from the edge of
the upper valve below the hinge-margin, M. von Buch placed the
genera Calccola and Leptena, including in the latter Productus and
Strophomena.

The views previously enunciated by M. von Buch as to the nature
of the tubes in Productus, appear to have been abandoned by that
eminent naturalist in his memoir “ UeberProductus oder Leptena” Tf.
He there considers it very doubtful if the tubes or spines were used
for fastening the shell; for he found that in some species, instead of
being confined to the hinge-line, they were scattered over the shell,
as in Productus aculeatus.

* Abh. d.k. Akad. Wiss. z. Berlin, aus dem Jahre 1828 (pub. 1831), pp. 53-56.

t Descr. des Coq. caractéristiques des Terr. p. 112.

+ Encyclopédie Méthodique: Hist. Nat. des Vers, par Bruguiére et de La-
marck, continuée par M. G. P. Deshayes, iii. p. 846.

§ Trans. Geol. Soc. 2nd ser. iii. p. 30.

|| Hist. Nat. des Anim. sans Verteéb., par J. B. P. A.de Lamarck. Deuxiéme
édition, par MM. G. P. Deshayes et Milne-Edwards, vil. pp. 378, 379.

¢ According to Prof. de Koninck, the division by M. Deshayes of the Tere-
bratuliform shells into two genera, Zerelratula and Productus, on the presence
or absence of any opening in the beak, was an erroneous one. Deshayes placed
in Productus all those with an articulated hinge in which the opening is quite
obliterated, and completely neglected the presence and form of the area &c.
(Anim. Foss. Terr. Carb. Belg. p. 150, and Mon. Productus et Chonetes, p. 8).

** Mém., Soc. Géol. de France, iii. p. 126.

++ Abh, d. k. Akad. Wiss. z. Berlin, aus dem Jahre 1841 (pub. 1843), pp. 53-56.

456 Rk. ETHERIDGE, JUN., ON A LOWER-CARBONIFEKOUS PRODUCTUS.

1842. M. Bouchard-Chantereaux described a Devonian Productus
from the neighbourhood of Boulogne possessing a rudimentary area
and a median triangular cleft under the beak, through which the
author supposed a peduncle for its attachment passed *.

1844 (1842-). Prof. L. de Koninck. It is impossible to reconcile,
says this eminent author, the presence of the long and delicate tubes
of the Producti with the idea that they could withstand the tossing
waves of the sea and the buffeting with a large number of rough
bodies, which evidently would have destroyed them. It is impro-
bable, he adds, that attachment took place by means of ligaments
issuing from these tubes or spines, according to the view formerly
held by M. von Buch ; neither does he think that there is sufficient
evidence to admit that the Producti were fixed by fibres issuing
along the hinge-margin of the valves. Prof. de Koninck supposes
that muscular fibres issued from between the free edges of the valves,
and so gave attachment to the shell, and in support of his view cites
the case of P. proboscideus, De Vern., in which he thinks the tubular
prolongation served to protect the fibres of attachment f.

1845. M. de Verneuil supported the later opinion of M. von Buch,
that the cardinal spines did not serve to attach the shells to submarine
bodies, but, he added, were probably used as channels for the introduc-
tion of water into the interior of the shell, necessary for its life and
respiration t+. M. de Verneuil considers that the Product, with the
exception of those species possessing a rudimentary area and a median
triangular cleft (P. horrescens, P. subaculeatus, and P. productoides),
have the cardinal margins so tightly pressed and united along the
whole length of the hinge that there is neither room for an area nor
for a triangular cleft. With regard to those species in which the latter
did exist he had considerable difficulty in conceiving that it gave
exit to a muscle of attachment §. Of the opinions expressed by M.
Deshayes and Prof. de Koninck, on the one hand that the Producti
were free and non-adherent, and on the other that they held fast by
muscular fibres passing from between the free edges of the valves,
De Verneuil conceived the former to be the more probable of the
two. Notwithstanding the peculiar prolongation of the shell in P.
proboscideus, which lends colour to the latter theory, the valves are
so united and tightly pressed along their margins that it is difficult
to imagine how they could have been separated by fibres during the
life of the mollusk ||. If the Producti were, as a rule, free shells,
adds M. de Verneuil,-some, however, were attached by the apex of
the beak of the convex valve, as for instance P. horrescens, De Vern.
In this species the part in question is obliquely truncated ; and it is
probable that it adhered by the apex of the beak or by very short
fibres which issued from the base of the median aperture or cleft 7.
Finally, in a note** to the general description of Productus, De

* Ann. d. Sc. Nat. 2° sér., Zoologie, xviii. p. 160.

t Anim. Foss. Terr. Carb. Belg. pp. 153 & 204.

{ Géol. de la Russie d’Europe, &c., par R. I. Murchison E. de Verneuil, et
le Comte A. de Keyserling, ii. p. 249.

§ Ibid. pp. 249, 250. || Ibid. p. 250. 9 Ibid. p. 286. ** Thid. p. 249.

R. ETHERIDGE, JUN., ON A LOWER-CARBONIFEROUS PRODUCTUS. 457

Verneuil adds that the tubes or spines could not in the least have
served to attach the shell.

1846. Count A. von Keyserling appears to have believed that the
Producti lived with their convex or ventral valves downwards, a
position which would be assumed by an example if thrown into
water, divested of matrix, and filled with a material having the same
specific weight as water *.

1847. Prof. L. de Koninck, in criticising M. de Verneuil’s theory
as to the admission of water through the tubes or spines, thinks that,
had they been intended for so important an office, it is probable that
their distribution over the valves of the shell would not have been
left so much to chance—in other words, that the distribution would
have been on a more defined plant. The views expressed by M. de
Verneuil on the theory of M. Bouchard-Chantereaux are acquiesced
in; and Prof. de Koninck further believes that the latter writer in-
cluded in his Productus certain forms which should have been placed
in Leptena. The learned author is unable to reconcile a free and
unattached habit (as supposed by M. Deshayes and others) with the
possession of long and delicate tubes by the majority of the species.
He believes that in the Producti, with the exception of P. probosci-
deus and its allies P. Nystianus and P. genuinus, the fibres of attach-
ment issued from between the free margins of the valves. The
truncated apex of the beak in P. horrescens is regarded, not as caused
by the attachment of the shell directly to submarine bodies by the
beak, but by friction of the same part against the rock to which the
animal was attached by its fibres f.

1848. Prof. H. B. Geinitz figures an example of Orthotrix Gold-
fussi, Miinster (= Strophalosia, King), from the Lower Zechstein of
Milbitz near Gera, which is distinctly represented with some of its
hinge-spines encircling and clinging to a spine of Productus hor-
ridus§. On the succeeding plate he gives a restoration of P. hor-
ridus, Sow., suspended by -its fibres of attachment passing from
between the valves ||.

1850. M. A. d’Orbigny supposed that the Producti rested on sea-
bottoms of fine sediment with their convex or ventral valves down-
wards, held in that position by the long spines, which he compared
to those of Spondylus striatus and other shells. The mode of
attachment of P. horridus he regarded as exceptional, and the pro-
boscis-like prolongation of P. proboscideus as a deformity due to an
accidentally constrained position of the shell in holes or crevices of
rock during life, the edges of the mantle being prolonged upwards
so as to reach the surface of the ground 4.

* Reise in das Petschora-Land im Jahre 1843, pp. 198, 199.
+ Monographie du genre Productus, Mém. Soc. Royale des Sc. de Liége, iv. *
pp. 104, 112-114; Rech. sur les Anim. Foss., Mon. des Genres Productus et

Chonetes, p. 15.
+ Ibid. pp. 21, 22. is
§ Die Versteinerungen des deutschen Zechsteingebirges, pl. 2. fig. 27, 8, ¢.
(a ae
« Ann, des Se. Nat. sér. 3, xiii. (Zoologie) p. 314.

458 R. ETHERIDGE, JUN., ON A LOWER-CARBONIFEROUS PRODUCTUS.

1850. Prof. W. King figured and described a specimen of Stro-
phalosia parva (King)* adhering to the interior of the shell of
Productus horridus by means of its umbone and the spines of that
part of the shell. Prof. King considered this to be the mode of
attachment of all the species of Strophalosia, at least in the young
state T.

1851. Dr. H. G. Bronn+ doubts if the spines were used for the
conveyance of water into the shell at all; for had they done so their
distribution over the shell would have been regular and strictly
determined.

1853. Mr. T. Davidson, in his magnificent ‘Monograph of the
British Fossil Brachiopoda,’ passed in review the various speculations
which had been advanced on the habits of Productus, and adopted
D’Orbigny’s view of the function of the proboscis of P. pro-
boscideus §.

1854. The late Dr. 8. P. Woodward considered that the Product
may have been attached when young, and that a few species were
perhaps permanently fixed. The large spines on the ears of the
ventral valve may have served to moor the shell ||.

1857. Mr. T. Davidson, in a note appended to the description of
Strophalosia Goldfussi, Minster (= S. parva, King?), quotes a
remark of Prof. de Koninck’s that ‘‘ sometimes the spines while inter-
lacing each other surround foreign objects which may lie within
their reach, and that those situated on the beak appear to have pos-
sessed that faculty in particular.” Specimens, says Mr. Davidson,
are in the possession of Prof. de Koninck in which the spines envelop
a fragment of the spine of P. horridus ¥.

1860. Chevalier d’Hichwald considers that the use of the spines
or tubes is unknown; they are probably simply ornamental.
According to his view P. proboscideus was fixed to rocks by its long
tube; and from this he argues that probably the other species of Pro-
ductus were sunilarly provided with a tube, which was membranous
and easily detached from the shell. It is extremely probable that
the numerous species of this genus were in general fixed by
muscular fibres which issued from the interior of the shell at the
lower margin **,

1862. Dr. J. C. Chenu describes Productus as a free shell t+.

1868. In the second edition of Dr. 8. P. Woodward’s valuable
work on the Mollusca, by Mr. R. Tate, similar views are expressed
to those previously noticed as given in the first edition of the
work.

The evidence contained in the foregoing quotations regarding the

* Considered by Mr. Davidson to be a synonym of S. Goldfussi, Minster.
t Mon. Permian Foss. England, p. 95, t. 12. f. 33.

t Lethza Geogn. 2nd ed. 1. p. 376.

§ Mon. Brit. Foss. Brach., Introduction, p. 119.

|| Manual of the Mollusca, Ist ed. pt. 2, p. 284.

“| Mon. Brit. Permian & Carb. Brach. p. 40, note 1.

** Lethza Rossica, i. pp. 885, 886.

++ Manuel de Conchyliologie et de Pal. Conchyl. ii. p. 226.

R. ETHERIDGE, JUN., ON A LOWER-CARBONIFEROUS PRODUCTUS. 459

habits of the Producti, and use of their spines, briefly summed up
under those two headings, is as follows :-—

A. Habits.

That the Producti were regarded as :—

1. Free shells, by Fleming, Deshayes, De Verneuil (partly), and
Chenu.

2. Living with the ventral or convex valve downwards, by
Sedgwick and Murchison, Keyserling and D’Orbigny.

3. Attached by tendons passing through tubes arranged along the
hinge-line, under the general name of Leptena, by Von Buch (an
opinion afterwards abandonod).

4, Attached by means of a peduncle passing through a central
triangular foramen, by Bouchard-Chantereaux.

5, Attached by fibres passing between the free edges of the shell,
by De Koninck, Geinitz, and Eichwald.

6. Probably attached when young, but a few permanently
attached, by S. P. Woodward.

7. Attached by spines (in the case of the allied genus Strophalosia),
by King and De Koninck (fide Davidson).

B. Spines.

That the spines were considered—

1. To be for the passage of tendons of attachment, by Von Buch.

2. To be for the introduction of water into the interior of the
shell, by De Verneuil.

3. To be for simply retaining the shell in position in fine sedi-
ments, by D’Orbigny.

4. To be without essential function, by Bronn.

5. To be of unknown function, by Eichwald.

6. Not to serve as a means of attachment to the shell, by De
Koninck and De Verneuil.

7. Not to act as a conduit to the interior of the shell, by De
Koninck and Bronn.

3. Description of the Specimens.—With the foregoing brief outline
of the views of some of the more eminent authorities, I pass on to a
description of the specimens.

We possess specimens of this adherent Productus in four condi-
tions :—I1st, attached by the spines of the ventral valve to foreign
bodies (figs. 1-5, & 11); 2nd, a few detached conjoined ventral and
dorsal valves; 3rd, detached dorsal valves (figs. 18-22); and, 4th,
fragments of detached ventral valves with the remains of spines
(figs. 16, 17). The first and fourth conditions are by far the most
common, and with regard to one another in about equal proportions.
In size the individuals vary from less than a line in diameter(fig. 13)
up to a little more than two lines (fig. 24), and are attached to
small Encrinite stems and fragments of Polyzoa, but generally to

460 Rk. ETHERIDGE, JUN., ON A LOWER-CARBONIFEROUS PRODUCTUS.

the former. The spines of the ventral valve are wound round the
bodies of attachment in the most complete manner (fig. 1 &c.), one
individual holding on by as few as two spines and a portion of a
third (fig. 11), another by six spines, and again another by seven,
and so on, sometimes singly (figs. 1, 2, 3, 5, &c), at other times in
clusters (fig. 4). When the organism to which the Productus is
attached is of larger size than the latter, the whole of the ventral
valve is applied to it, the spines spreading out and round on each
side (figs. 2,4, 5); but when the foreign body is of less diameter
than the Productus, as is usually the case with fragments of Polyzoa,
several of the spines are wound tightly round, especially near the
beak, and the remainder of the valve remains free (figs. 1, 3, 11).
Attachment took place during the life of the Crinoid ; for, in nearly
every case where the Productus remains adhering, we find that its
rate of growth was less than that of the Crinoid, the result being
that the substance of the latter surrounded or enclosed its parasite,
first the encircling spines disappearing and gradually the shell
(figs. 6-10). We have specimens showing this remarkably well in
all stages of the process, from the mere absorption of the spines by
the substance of the Crinoid, up to the total disappearance of the
Productus itself, when the Crinoid stem assumes a swollen or dis-
torted appearance (fig. 10). From a consideration of this gradual
absorption by the Crinoid stem there arise two questions :—Did the
Productus when once attached lose the power to free itself? or, Did
the absorption by the Crinoid, contrary to the view indicated pre-
viously, commence only after the death of the Productus? To show
how closely these little shells congregated I have given a figure of
an Encrinite stem on which may be seen the grooves caused by the
_ removal of a number of the encircling spines (fig. 12).

In the young state the characters of this hugging Productus are
obscure ; but in the mature (or what I take to be the mature) form
they become well marked and constant. In marginal outline the
shell somewhat resembles the P. aculeatus group, with the hinge-line
scarcely as long as the shell. The auricular expansions are well
marked but flattened (figs. 13, 14,19), and the cardinal angles
sharp (figs. 13, 14). The ventral valve is slightly convex when the
individuals are partially free, and without sinus (figs. 13-15). The
dorsal is flat or a little concave (figs. 18, 19, 22, 24), and granular
on the interior surface. The cardinal process is bifid and broadly
V-shaped (figs. 18, 21, 24), with strong lateral ridges proceeding
from it to the cardinal angles (a, figs. 21, 23). The mesial ridge in
the interior of the dorsal valve, immediately under the cardinal pro-
cess, is divided into two portions (6, figs. 21, 23), which unite at
about one third from the hinge-line, enclosing between them a
rather deep pit or space (c, figs. 21, 23), which partially represents
the convex visceral region of the shell externally. The mesial ridge
becomes stronger and more pronounced towards the front, and
divides the shell symmetrically. Both valves are covered with a
variable number of sharp, undulating, concentric wrinkles or corru-
gations, coarse for the size of the shell, and occasionally irregular

R. ETHERIDGE, JUN., ON A LOWER-CARBONIFEROUS PRODUCTUS. 461

(figs. 13-20, 22, 24), sometimes almost becoming geniculate. In
the dorsal valve they are perhaps finer and more numerous than in
the ventral valve ; and in the visceral region of the shell the circular
concentric character gives place to an almost square concentric form
of wrinkle or corrugation (figs. 18, 24). As the shells increased in
age the corrugations became broken up in a roughly reticulate manner
(figs. 22, 24). In the young state the ventral valve only is partially
covered with spines (figs. 13, 14), the dorsal being devoid of them
(fig. 19); but as growth went on they also appeared on the latter,
but apparently without attaining to the same size. The spines are
irregularly scattered over the surface of the valves ; those of the yen-
tral valve long, thin, tapering (figs. 16, 17), and encircling (fig. 11),
and when broken leaving the shell covered with a series of small warts
(figs. 13-17). A row of prominent strong spines occurs along and
immediately below the hinge-line of the ventral valve (a, figs. 16,
17), but in the two figured examples broken off. In a few instances
one or more of the spines may be seen pressed down against the sur-
face of the ventral valve, as if it had so grown between the shell
and the object to which it was attached, without encircling the
latter. The umbo of the ventral valve is small, sharp, plain, and
slightly incurved over the hinge-line. The visceral region of the
same valve is often grooved vertically or indented, the impression of
the body to which the individual was attached, such grooving eleva-
ting the beak (figs. 13, 14) and obliterating the corrugations on
other portions of the valve more than would otherwise be the case.
The adductor muscular scars are only preserved distinctly in one
specimen. They appear to be elongated vertically, and are appa-
rently lobed transversely. |

4. Considerations as to Specific Identity. The question which
most prominently presents itself is:—Is this adherent Productus the
young of some well-known species, or is it to be regarded asa minute
and undescribed form? Let us examine the evidence that we possess
to settle this question. Of the unattached examples fully two thirds
are dorsal valves. The predominance of these over the ventral valves
would appear to indicate that it was at least an attached species up
to that point of its existence represented by the largest of our speci-
mens, which are dorsal valves, and which may, I think, be taken as
representing maturity, unless very strong evidence to the contrary
can be produced. It is only reasonable to suppose that had the
Productus at a certain point of its life disunited itself from the
Crinoid stems, upon which it had hitherto lived, we should be in
possession of relatively as many large, and what I take to be mature,
ventral as dorsal valves. But we are not; for out of a large number
of specimens now in the possession of the Geological Survey of Scot-
land and Museum of Practical Geology nearly all the largest examples
are dorsal valves.

When Mr. Bennie was fortunate enough to light upon these
interesting fossils, I at once submitted characteristic examples to
Mr. T. Davidson, F.R.S., who, in his usually kind manner, gave
me the benefit of his valued opinion. He pointed out that the

462 R. ETHERIDGE, JUN., ON A LOWER-CARBONIFEROUS PRODUCTUS.

present fossils might either be the young of P. longispinus, Sow., or
P. Wrightii, Dav., and recommended careful observation as to the
other species of Productus found in company with the present form.
Mr. Bennie, when collecting the disintegrated shale from which the
specimens were obtained, observed that the only other Productus
present in any thing like appreciable quantity was P. longispinus,
Sow., with this peculiarity, however, that at one of the localities
(Kast Barns) P. longispinus was very plentiful and the adherent
form correspondingly rare, whilst at another locality (Skateraw) the
proportions were just reversed. The characters in which the adhe-
rent form differs from P. longispinus, Sow., are the following :—The
ornamentation of the shell is concentric and not longitudinal; there
is no trace of a sinus in the ventral valve, but a strong mesial ridge
in the interior of the dorsal valve extending to the front, which does
not appear to be so well marked in P. longispinus; the upper por-
tion of this mesial ridge splits into two, leaving a space between,
which is not the case in P. longispinus. The adductor impressions
of the latter are situated much higher up under the cardinal process
than in the present instance (fig. 23). Lastly, there is a good deal
of difference in the distribution of the spines, and in the relative
convexity of the valves.

The division which takes place in the mesial ridge of the dorsal
valve (6, fig. 21) appears to resemble that figured by Mr. Davidson
in Productus scabriculus, Martin*; but I think it is unnecessary in
this case to carry the comparison further.

The concentric wrinkles or corrugations are sharper and more
widely separated than those of Productus undatus, Defrance; and
there is no trace in the adherent form of the fine longitudinal striae
of that species.

The affinities with P. Wrightvi, Dav., are much closer than in any
of the foregoing cases. In both shells the ornamentation is con-
centric, although apparently more regular in P. Wrightit. The
size of the two forms is much more approximate, especially one of
Mr. Davidson’s figures ft; and the spines in the enlarged figure of
the same author have much the appearance of those of the adherent
form, although fewer in number. At the same time there do not
appear to be any along the hinge-line of P. Wright; and, so far as
my own observation has gone, there is no trace of the marginal frill
of the latter in the present instance.

I do not feel justified in immediately publishing this form as a
distinct and undescribed species, but would rather await the opinion
of those more intimately acquainted with the structure and variations
of the genus Productus than I can pretend to be. Atthe same time,
should further researches prove these little shells to be worthy of
separate specific recognition, I do not think any name could be more
appropriate than one which would express the peculiar habit for
which they are remarkable—such for instance as Productus complec-
tens, the embracing or encircling Productus.

* Mon. Perm. & Carb. Brach. pl. 42. f. 8a. t Ibid. pl. 33. f. 6.

R. ETHERIDGE, JUN., ON A LOWER-CARBONIFEROUS SPIRIFERINA. 463

In concluding this part of my communication, it may be well to
shortly recapitulate the chief peculiarities of this remarkable form
of Productus. First, its parasitic habit, the indication of which can
be seen in some form or other throughout nearly the whole suite of
specimens, either as a groove in the ventral valve from the beak
forwards, or as a distortion, or by a curving and twisting of the
spines. Secondly, the characteristic and well-defined nature of the
concentric ornamentation, and, when preserved, the elongated
clasping spines. Thirdly, the well-marked mesial ridge in the interior
of the dorsal valve, and its separation into two portions, as in P.
scabriculus, Martin. Fourthly, the absence of a sinus in the ventral
valve. Fifthly, the well-marked peculiarities of character continuously
traceable from the smallest to the largest specimen, and no apparent
gradation onwards to some well-known species.

5. Locality and Horizon.—Skateraw Harbour, near Dunbar, plenti-
ful; East Barns Quarry, near Dunbar, not common; Carlops Quarry,
near Carlops, Peeblesshire, very rare; shale over the No. 2 Lime-
stone (=Hosie Limestone) of the Lower Carboniferous Limestone
group of the Midlothian series.

Collected by Mr. James Bennie.

When collecting fossils from the Calderwood series in the East
Kilbride district, Lanarkshire, a year or so since, Mr. Bennie noticed
bodies with radiating processes adhering to Encrinite stems, which
he thinks may be this Productus.

II, On a SMALL SPIRIFERINA.

I take this opportunity of figuring and briefly describing a small
shell which may be new, in the hope that the attention of collectors
may be called to it. As we at present only possess one speci-
men, I shall content myself with merely placing it on record, and
await the discovery of other specimens, when we may be better able
to judge of its specific value.

SPIRIFERINA, Sp.

Specific Characters.—Small, transversely oval, wider than long ;
ventral valve the more convex of the two. Hinge-line less than the
width of the shell; cardinal angles rounded; beak incurved and
moderately produced above the area, which is concave and trian-
gular; fissure wide. Ventral valve with a simple shallow sinus, and
three rounded ribs on each side, the outer ones indistinctly marked.
In the dorsal valve the mesial fold is composed of a single rib, larger
than either of the others, two ou each side. No laminz of growth.
Shell-structure perforated by minute tubuli giving to the surface a
dotted appearance.

Obs.—This little shell differs from Spiriferina cristata, Schl., var.
octoplicata, Sow., by the hinge-line being less than the width, by its
very rounded cardinal angles, perfect simplicity of the mesial fold

Q.J.G.8. No. 128. OR

464 KR. ETHERIDGE, JUN., ON A LOWER-CARBONIFEROUS

and sinus, much less number of ribs, total absence of any lamine of
growth, spinous asperities, or granules on the surface.

From Spiriferina insculpta, Phill., it is distinguished by the same
character of the hinge-line as in the former instance, greater con-
vexity of the ventral as compared with the dorsal valve, absence of
concentric lamine of growth, and smaller area. On the other hand,
it agrees with S. insculpta in the number of the ribs when that
species is found in its simplest form; but these again are decidedly
obtuse and not angular. These points of difference can only be
regarded as provisional, as the discovery of further specimens may
bring to light varieties intermediate between the shell as here
figured and the typical S. insculpta, to which, of the two species
mentioned, it appears to be most nearly allied. In outline, simple
mesial fold, and sinus it is not unlike some forms of Spirifera
triradialis, Phill.; but the beak in the ventral valye is much less
elevated, and the shell-substance punctate.

Locality and Position.—Fullarton Quarry, near Temple, Edin-
burghshire ; in shale between bands of the No. 2 Limestone (= Hosie
Limestone) of the Lower Carboniferous Limestone group of the
Midlothian series.

In conclusion I have to acknowledge the obligations I am under
to my colleagues Messrs. G. Sharman and R. L. Jack, the former
for supplying me with a series of beautifully executed drawings,
and the latter for assistance in translating certain foreign memoirs,
&e.

DESCRIPTION OF PLATES XXIV. & XXV.

Propvuctvs, sp. (Figs. 1-24.)

Fig. 1. Portion of Encrinite stem with adherent Productus, showing the

spines tightly clasping it. Back view, ventral valves,

. A similar specimen, side view; three spines shown.

. A larger and more mature specimen, front view.

. Three individuals clustered together; front view, ventral valves.

A larger specimen, adhering by seven spines, front view.

. Interior of the ventral valve, showing the margins gradually enclosed

by the substance of the Encrinite stem.

7. The same process earried a little further.

8. The same process carried still further.

9. Almost total disappearance of the Productus within the substance
of the Encrinite stem.

10. Total absorption of the Productus, and distortion of the Encrinite
stem.

11. Exterior of a ventral valve adhering by two and a half spines to
the end of a segment of an Encrinite stem. Broken spines are
shown round the margins of the valves.

12. Encrinite stems with the remains of three young individuals and
the markings of numerous encircling spines.

13 Exterior of the ventral valve, with the elevated umbo and groove
caused by the object to which it was attached. Two broken spines
are seen.

14. A similar specimen. Both these figures show the concentric
wrinkles in a marked manner.

O Or yp 69 bo

Quart Journ Geol.Soc.Vol XXXI.P] XXIV

Lea oe 4 cS

\y
)

"
————

GSharmen del. R Etheridge Jun™ dirent Hanhart imp

ee Se ee ee EE EES

-
P
——
. = f
‘
‘
'
*
.
2
.
'
val
-
i
bs ?
.

~

y. «rum
|

Feat be Mak sy Net pe ES) AE t
WU ay JOULN GeO, OGE Veal XXXII PZ NY

1arman del. REtheridge Jun" dirextt. Hanhart imp

DID AMTTA mato aprmmomrre

A

ay

Fig. 15.
16,
18,
20.

Fig. 25.

PRODUCTUS AND A SMALL SPIRIFERINA. 465

A more transversely elongated ventral valve, with spines, exterior
view.

17. Portions of the exterior of two ventral valves, showing the
hinge-spines, 0.

19. Exterior of two dorsal valves, showing the cardinal process and
concentric wrinkles.

Hinge-line, and portions of both dorsal and ventral valves united,
with one of the large hinge-spines in position.

. Interior of the dorsal valve: aa, ridges from the cardinal process to

the cardinal angles; 0, 6, divisions of the mesial ridge ; c, enclosed
space.

. Exterior of the dorsal valve, with the cardinal process, concentric

wrinkles, and traces of the reticulate character assumed by the
latter.

. Interior of a portion of the dorsal valve: a & ¢ similar to fig. 21 ;

d, d, adductor impressions.

. Similar to fig. 22.

SPIRIFERINA, sp.

Dorsal valve, showing the mesial fold.

26. Ventral valve, showing the sinus,

Fig. 28.

. Hinge and fissure Kc.

SrropHALosiA (OrTHOTRIX) Gouprussi, Minster.

After Geinitz, showing the form clinging by its umbonal spines to
one of the large spines of a Productus horridus. (Die Verstein. d.
deuts. Zechst. pl. 2. f. 27.)

28a. The same, enlarged.

(N.B. 'I'ne natural size of the fossils is shown, as nearly as possible, by the in -
dicators placed near the figures. )

466 C. T. CLOUGH ON THE SECTION AT

51. The Szction at the Hie Forcz, Trxsparz. By C. T. Crovex,
Esq., B.A., F.G.S., Scholar of St. John’s College, Cambridge, of
the Geol. Survey of England and Wales. (Read June 21, 1876.)

(Communicated by permission of the Director-General of the Geological Survey
. of Great Britain.)

Ir has been my duty, in the course of Survey work, to examine the
ground near the High Force. In doing this, several facts have come
to light which will, perhaps, be not uninteresting, as they refer to
a somewhat classical district, and correct ideas which have been
hitherto erroneously held on the authority of Phillips’s section at the
High Force. These ideas have lately acquired a special importance
from their bearing on the subject of “selective metamorphism ;” and in
correcting them I may be helping, in some small degree, to clear the ©
way for the formation of a theory to satisfactorily account for those
instances of selective metamorphism which stand undoubted.

In Phillips’s ‘ Geology of Yorkshire’ (1836), part ii. pp. 78, 79, the
following passage occurs in connexion with the section at the High
Force :—‘‘ Shale or plate is so much altered at the High Force in the
relation of the joints that most persons mistake a part of the pris-
matic masses, really composed of metamorphic shale, for trap, and
suppose the latter to rest on limestone. The true series is as follows,
proceeding downwards :—

“a, Basalt, rudely prismatic, grey with lichen.
6. Thin plate, not very much indurated.
c. Bed of plate, subprismatic.
d, Beds of plate, laminated.
e. Thin limestone bed, with a superficial layer of pyrites.
f. Bed of hard pyritous limestone.
g. Several beds of common dark limestone, with white shells and corals.”

(Among the Dalesmen the word “plate” is equivalent to “shale;”
and the basalt referred to is part of the great “‘ whin sill” of Tees-
dale, &c.)

On reading over this section it naturally strikes one at once as a
very odd thing that there should be a subprismatic bed of plate, c,
below the thin plate not very much indurated, 6—~Jin other
words, that the plate which lies nearest to the altering cause
should be less altered than the plate which is further from it. Here,
indeed, if the beds 6 and ¢ had really been of the character described
above, we should have a most striking instance of selective meta-
morphism. I think I shall be able to show, however, that the above
description is not correct— that the bed ¢ is not altered plate,
but basalt, that an examination of the general rock-structure of
the beds a and c proves conclusively that the only differences
between them are such as should naturally be looked for in beds of
such different thicknesses, and that this conclusion is fully borne

=

THE HIGH FORCE, TEESDALE. 467

out by what can be seen of the physical disposition of these beds in
the grounds immediately adjoining the High Force.

What a very great difference there is between 6 and c and between
c and d is evident at once from a first look at the Force—b and d
being conspicuously crossed by lines of lamination rudely parallel
to their upper surfaces, while c is completely without any such lines,
and is rudely prismatic, like the basalt at the top of the Force. But
these differences are seen to be, if possible, still greater when we
come to examine the Force, and the minuter structure of the beds:
6 and d have no trace of crystalline structure whatever, while a and
c both have, and the structure of ¢ appears to agree exactly with
that of the lower part of a. Both ¢ and the lower part of a are much
more finely crystalline than the more centrally situated parts of a;
but this is only what one would expect on considering the different
conditions under which their cooling would probably take place. The
outside of the mass would naturally cool much more quickly than
the inside; and the underflow, c, is only about 6 feet thick, and
therefore the whole of this would probably cool down comparatively
quickly.

IT think, then, that there can be-no doubt that the bed ¢ is really
basalt (microscopic examination would prove this), here a sort of
underflow to the main mass. For Phillips’s section I would substi-
the following:

Fig. 1.—Section at the High Force.

a. Basalt, 24 feet. d. Shale, altered, 12 feet.
b. Shale, altered, 14 foot (average). e, f, g. Limestone, 32 feet.
c. Basalt, 6 feet (average).

_ The thicknesses given are only approximate.

468 C. T. CLOUGH ON THE SECTION AT

It must be understood that the above section refers to the Force
itself, the actual place where the water comes over. The sections
on either bank of the Tees a few dozen yards below the Force differ
from that at the Force itself, and at the same time differ inter se.

On the north bank, at about 20 yards below the Force, it can be
distinctly seen that there is no separate bed c, and that the space
between the top of a and the bottom of cis one uniform mass of basalt.
We have, then, the following as the section :—

Basalt, 82 feet (+ any that may be above the level of the top of the
basalt at the Force).

Shale, 12 feet.

Limestone, 32 feet.

We will calJl this section the section at X.
On the south bank the section, at a point about 30 ante below
tho Force, is:—

Basalt, 24 feet (+ any that may be above the level of the top of the
basalt at the Force).

Shale, 20 feet.

Limestone, 32 feet.

We will call this section the section at Y.

Owing to the banks being, in places, inconveniently steep, and
bearing a thick covering of hanging trees &c., the entire’ passage
between the above three sections cannot be quite so well seen as one
would wish; enough, however, is shown to enable one to come to a
safe conclusion as to the true state of the case. On looking at the
section at the Force more carefully, it is distinctly seen that the bed
of shale, 6, is thicker on the south side than on the north side of the
Force—that from 14 foot it dwindles down to one foot, and from one
foot it dwindles more and more as we go towards the section at X,
until itis only doubtfully represented by a thin line marked by ferns
and grass, and that this thin line then ceases, and we get the state
of things at X—that, on the other hand, as we go from the Force
towards the section at Y, the bed of shale, 6, gets thicker and thicker
(from 13 foot it gets to 2 feet, and so on), until, at a point about
20 yards below the Force, it is 3 feet thick, and in 10 or 12 yards
more it has thickened out and formed one mass with the bed of
shale, d; and now there is no underflow of basalt.

It appears, then, quite safe to conclude that the underflow of
basalt, which has its origin from the main mass at X, gradually thins
out as we go south-west and south from this point, until at Y it has
ceased altogether.

If we examine the bank a few yards beyond X, we find the sec-
tion change again, and again we have an underflow of basalt separ-
ated from the main mass by a thin bed of altered shale, thus:—

Basalt.

Sandy shale, about 2 feet.
Basalt.

The top of the underflow of basalt here is roughly on the same

THE HIGH FORCE, TEESDALE, 469

level as that of the underflow at the Force, and the shale parting,
which at first makes its appearance somewhat abruptly, can be seen
for a dozen yards or so gradually thickening as we follow the bank
of the river in a north-easterly direction. The section then be-
comes lost, owing to the covering of tumbled whin, &c., and we have
to go on for some 70 yards before the beds are well shewn again.
Here the section is much the same as that at Y, viz.:—

Basalt.
Shale, about 20 feet.
Limestone.

We have, then, good evidence that the section at X changes to-
wards the north-east in the same way as it does towards the south-
west and south.

And so, finally, we see that near X a small part of the lower por-
tion of basalt becomes separated off from the main mass, and that
this lower portion rapidly thins out towards both the south-west and
south, and the north-east.

There are other places, both in the immediate neighbourhood of
the High Force and elsewhere, where similar undoubted underflows
exist. Sometimes these underflows can be seen distinctly in the act
of being gradually separated off from the main basalt mass, altered
beds coming in between; and sometimes these altered beds consist
of altered sandstone, or of altered limestone, instead of altered shale.
And, as there are underflows, so also there are overflows; parts of
the top of the basalt gradually get separated off from the main mass,
and then thin out.

In some cases we can see beds of altered shale and sandstone
apparently surrounded by basalt on every side: there is basalt above ;
and there is basalt below; and the altered beds can be seen to end
in basalt as you follow the section along on either hand. There is
a very good dnstance of this in the Slate Scar, a little more than half a
mile above the High Force. The greatest dimensions of these beds
are seen to be generally in the direction of their planes of original
deposition ; their thicknesses are, in fact, generally strikingly less
than their length. For instance, with an average thickness of 2
feet, we may get a length of 60 yards. This is, I think, strongly
suggestive of some stable connexion between these and the other beds
among which the basalt has been intruded; they appear to have
kept something of their original position very much better than one
would suppose at all likely if they had been completely isolated in the
great basalt mass. I would suppose, then, that, certainly in most
cases, these beds are not surrounded by basalt on every side—have
not, in fact, been simply “‘ caught up” in it—but that, if they could
be traced in directions both away from the observer and towards
him, they would, in one of these directions, be found to join on with
the great mass of the sedimentary rocks of the country. It is eyi-
dent that, in any ordinary section, we have only four sides visible
out of the six wherewith any geographical point may be supposed

470 C. T, CLOUGH ON THE SECTION AT

to be bounded; and in either of the two unseen the state of things
may be very different from that seen in the others.

These subordinate beds of basalt have been spoken of as under-
flows and overflows; in every case observed, however, they thin
out so very rapidly that it might be better to call them simply
wedges. They appear to belong in every way to the main mass of
basalt; and probably one of the ordinary methods of the thinning
out &c. of the basalt is the giving off of these separation masses, and
their then dying out. Thus the thinning out &c. of the basalt as a
whole might probably be best represented, diagrammatically, some-
what as below :—

Fig 2.—Thinning-out of the Basalt at the High Force.

a. Sedimentary rocks. &. Basalt. c. Sedimentary rocks.

I do not wish to imply by the above diagram that there is one
point from which the basalt can be made out to be thinning away
in every direction, but simply that, where the basalt is thinning
out, the thinning-out probably takes place in some such way as re-
presented.

Prof. Sedgwick, in his paper on the Geology of High Teesdale
(Cambr. Phil. Trans. 1824), brought forward evidence which was
quite conclusive as to the truly intrusive character of the whin sill.
Prof. Phillips, at the time he wrote his ‘ Geology of Yorkshire,’ does
not seem to have loyally accepted this conclusion, but to have been
more impressed with its supposed ‘‘ general conformity over a vast
area to the beds above and below.” It is this partial conformity
which still causes it to be an article in the miner’s creed that the
great whin mass is a “sill,” a bed regularly interbedded with the
other beds or “sills.” On close examination of the ground, however,
it is seen that this conformity is not nearly so general as is supposed,
and that even in cases where the whin, as a whole, does keep on the
same geological horizon, there are numerous instances of small
subordinate beds separating off from the main mass in the manner
already described, and conclusively proving its intrusive character.
The High Force is one of the four localities specially mentioned by
Phillips (‘Geology of Yorkshire,’ part ii. p. 77) as showing the con-
formity between the surfaces of the whin sill and the surrounding
beds. We have seen that, instead of showing this, it affords striking
evidence to the contrary. The bed of the Tees near Winch Bridge
is another of the localities mentioned. This, too, shows good eyi-

THE HIGH FORCE, TEESDALE. 471

dence to the contrary ; and I have very little doubt that the more
the country is worked over the more and more evident will the in-
trusive character of the whin sill become.

P.S. Since the above was written, Mr. F. Rutley has kindly cut,
and microscopically examined, sections of the rocks a, 6, and c¢ of the
High-Force section (vide supra); and the results obtained entirely
confirm my statements: a and c are basalts; and d is evidently an
altered sedimentary rock. The examination of the rocks in the field
really left no doubt on this head,

473 R. MALLET ON THE MECHANISM OF

52. On the Mecuanism of Propuction of Votcantc Dyxzs, and on
those of Montz Somma. By R. Matter, Esq., C.E., F.RBS.,
F.G.S. (Read June 21, 1876.)

In the year 1864 I was enabled to employ some time in the
study of some of the chief volcanic phenomena presented by the
cone of Etna, devoting my attention principally to the laws which
govern the flow of lava currents, the formation of the parasitic
cones which abound upon certain. portions of the surface of the great
mountain, of which Monte Rosso and Monte Peleri in the neigh-
bourhood of Nicolosi are amongst the largest examples, and in
examining the Val del Bove and the many so-called dykes which
intersect its surrounding escarpment in so many places. These last
objects were examined by me with the able memoir of Sir Charles
Lyell, published in the Philosophical Transaction for 1858, in hand,
the title of which is “* on Lavas of Mount Etna,” &e.

In the second part of that memoir the author rests part of the
evidence upon which he concludes that an ancient great vent existed
in the Val del Bove at Trifoglietto, upon the convergence at about
that point of the prolonged lines of direction of thirteen dykes
existing in the surrounding escarpment, in accordance with the
views previously promulgated by Von Waltershausen (Phil. Trans.,
Part ii. 1858, page 703). I have no intention of casting any doubt
here upon any part of the above able memoir of Sir Charles Lyell,
in all the main conclusions of which, indeed, I concur, my pre-
sent object being limited to remarking upon some of the conditions
which affect the formation of volcanic so-called dykes of injection,
such as those existing in the escarpment of the Val del Bove, and
of that of Monte Somma, and to pointing out the extent to which
the orientation or direction of such dykes can be safely employed as
a means for determining, with any thing approaching to certainty,
the existence or position of a central chimney or crater, from which
such dykes may be supposed to have emanated, by means of the in-
tersection of lines or planes prolonged in the direction of such dykes.
On going round nearly the entire escarpment of the Val del Bove,
with Von Waltershausen’s map in hand, the dykes, which are almost
innumerable, when examined at many points both from above and
below the escarpment, presented phenomena in many other respects
as well as in direction, so perplexing as to raise in my mind much
doubt as to whether they could be at all accounted for upon the
commonly accepted theory of their production, and suggested the
inquiry whether that theory which assigns their production to the
injection of fissures in crater walls by the liquid lava within the
crater might not need considerable modification, and if so, whether,
in its imperfect state, the orientation of dykes could, by intersec-
tion, be safely employed at all for determining the position of craters
which have disappeared, as in the case of that of Monte Somma

PRODUCTION OF VOLCANIC DYKES. 473

and of those supposed to have existed in the Val del Bove. In the
latter region the orientations or lines of horizontal direction of the
numberless dykes are so various and discordant as to resemble
in very many places nothing more than the chance directions which
might be assumed by a number of short straws or sticks cast at
random upon the ground. Lines drawn through them might be
made to create great numbers of points of intersection; and yet
great uncertainty would attend attempts to infer from any of them the
former existence of a crater there. These perplexed phenomena in
the Val del Bove are upon too large a scale for easy disentangle-
ment, and are far less suitable for the study of the laws of production
of injected dykes than are those found so abundantly in the escarpment
of Monte Somma, where the whole scale is smaller, the dykes gene-
rally closer together round the amphitheatre, the phenomena less
perplexed than in the Val del Bove by the circumstances which pro-
duced that immense depression, and where there exists the main
chimney or axis of the cone of Vesuvius, the position of which is
not presumed to have materially changed from that before occupied
by the crater while Somma was in existence, and to which therefore
it might be presumed that the directions of all the existing dykes of
Somma should show a generai convergence. Reserving therefore
the formation of any conclusion as to the facts I had observed upon
Etna, I resolved upon my return to Naples to make a careful study of
the dykes of Somma, and to endeavour to decipher therefrom the laws
of production of these and other similar volcanic dykes. For though
these dykes of Somma have been again and again described in their
general aspects, and though almost all writers on volcanoes have as-
sumed without question that their directions do really converge at
the axis of the existing cone of Vesuvius, I could not find that any
careful determination by actual survey had ever been made of the
directions of these dykes; nor had sufficiently exact observations been
made as to many points in the dykes themselves tending to elucidate
the exact nature of their origination. Accordingly, in November
1864 I made arrangements for determining with instrumental ac-
curacy the positions and directions, both horizontal and vertical, of a
considerable number of these dykes of Somma. Having, through
the kind assistance of Professor Palmieri, been permitted to lodge at
the Observatory above the Hermitage upon Vesuvius, I ran a line
of trigonometric triangles from the south-east corner of the Obser-
vatory, passed the wooden cross existing at that time, and known
as La Croce del Salvatore, and into about the centre of the Atrio del
Cavallo, whence I was enabled to fix the position of a point chosen
for that of observations upon one of the slight elevations at the
northern base of the cone of Vesuvius, which commanded an unin-
terrupted view of a very large portion of the entire escarpment of
Somma, extending from the Canale del Inferno, or thereabouts, on
the eastward, and passing round. the amphitheatre westward to a
point nearly opposite La Croce, and a little to the west of north with
respect to the same. I was obliged to choose the quoin of the Ob-
servatory as an initial station-mark, although the building cannot be

A474 R. MALLET ON THE MECHANISM OF

seen from the floor of the Atrio, nor the latter from it, because no
station-mark of a permanent character, or offering a reasonable
chance of not disappearing in perhaps the very next eruption, could
be found in or nearer the Atrio del Cavallo. At this point, very
near the head of the Fossa di Vetrano, the first prominent dyke of
the escarpment of Somma occurs, which I called No. 1; and having
perambulated the entire base of the escarpment, thence going
eastward and choosing out for geodetic determination, from the in-
numerable dykes of every size which intersect it in almost every
direction, twenty-seven of those which seemed largest, highest, and
generally most characteristic, I numbered the whole of these with
red oil paint in figures five or six feet long upon the projecting
edge or face of each near the base or level at which these dykes
generally disappear below the floor of the Atrio del Cavallo, taking
care that all these figures should be discernible by the telescope of
my theodolite from the point of observation chosen at the base of
Vesuvius as already mentioned. I however examined carefully as to
their lithological circumstances a number of other dykes intercalated
at various points between those numbered, amounting with the
preceding to thirty-four in all.

Having in connexion with the line of triangles by which the
observing-point was fixed, measured a short base of 374 feet (Eng-
lish) upon a level part of the Atrio del Cavallo extending in a 8.E.
direction, the N.W. extremity being nearly in a line between the
Punta del Naso and the centre of the Vesuvian cone, I then
extended my triangulation so as to fix, in reference to my observing-
point, and to the then existing cone and crater of Vesuvius, the general
line of the escarpment of Somma from west to east, extending along
the curve of the escarpment to a distance of rather more than 24
English miles. This line has in reality much less of the form of a
regular circular curved amphitheatre than it appears to the eye as
viewed from the northern side of the cone of Vesuvius; and I found
it was by no means exactly laid down upon the engraved sheets
which contain Vesuvius and Somma, issued from the topographical
office at Naples by the former Government of the King of the Two
Sicilies. I was thus in a position to fix trigonometrically the
orientation of the twenty-seven numbered dykes jn reference to my
observing-point, and to the axis of the cone of Vesuvius ; and then,
again perambulating the whole line of dykes from west to east, I
observed and noted the characteristics of each, the lithological
character of its mass, the horizontal direction in which the plane of
the dyke penetrated the bank of the escarpment, the “hade,” or
dip, or departure in a vertical direction from perpendicularity in
the plate of rock, its curvature, if any, and other particulars to be
hereafter noticed. These operations and observations occupied me,
along with a staff of men hired at Naples and Resina, for about a
fortnight in November, during several days of which there was
continuous rain and so boisterous a wind as rendered work with a
theodolite difficult. My chief assistant was Antonio d’Antonini,
who had been my faithful and intelligent attendant in exploring

PRODUCTION OF VOLCANIC DYKES. 475

the results of the Neapolitan earthquake of 1857, and who is, I
believe, still in Naples.

After my return to England my survey was to a great extent
plotted ; and the results showed themselves as concordant as
could have been expected from geodetic work carried on upon
excessively uneven and elevated ground, and under the disad-
vantages of weather described. The necessities of professional
engagements and other circumstances, however, obliged my laying
aside for a considerable time the completion of my work, until
ultimately failure of sight and the impossibility of my note-books
and sketches being deciphered by another have compelled me to
lay it aside altogether. Unwilling, however, that the results of so
much labour and care should be wholly lost, especially as some
of the facts ascertained are new and appear not devoid of im-
portance, I have thought it best to note down some of the more
salient results, which others more fortunate than myself may per-
haps accept as a basis for further research. My inability to pro-
duce the whole of my work, together with the triangulation, in a
fit state for publication is now the more regretted by me since
most of the features, the surface-levels &c. of the Atrio del
Cavallo, as they existed in 1864 have been to an immense extent
altered by the great eructations of lava that poured into it during
the great eruption of 1872, so well described by Signor Palmieri.
It is also probable that the scorching heat then produced in the
Atrio del Cavallo, and extending westward far beyond the Ob-
servatory, may have partially, if not wholly, obliterated most of
the red numbers painted by me upon the dykes, and perhaps even
consumed by fire the wooden cross of the Salvatore, which was one
of my marks.

On examining with care, and aided by the telescope, the entire
surface of the escarpment of Somma, it is seen to consist of a pre-
cipitous but nowhere actually vertical bank, with a highly irregular
or serrated sky-line rising at some points 800 or more feet above
the floor of the Atrio. Its exposed face consists to a preponderant
extent of beds of volcanic conglomerate, composed of fragments,
subangular or more or less perfectly rounded, of various sizes,
from masses occasionally of a cubic yard or more in volume to
nodules of the size of a human head, or of an orange, mixed
with pebbles of various smaller sizes and with gravel or lapili. Of
the latter there are also great beds. Intercalated with all these are
many beds of lava, which, though now for the most part cross-
fractured and twisted, had flowed in a state of fusion from the
ancient crater of Somma and consolidated upon its slopes. All
these beds, though wildly confused in many places, yet preserve a
general horizontality round the amphitheatre. Crossing all these
in very various directions, forming different angles with the
horizon, and scattered at very irregular distances from each other,
existing sometimes singly but often several in close proximity, and
sometimes inosculating with each other, and with their edges highly
irregular and broken, projecting more or less from the abrupt face of

476 R, MALLET ON THE MECHANISM OF

the escarpment, whose slope they very nearly follow, are the
dykes so often referred to by authors. These sometimes run up
the entire height of the escarpment, sometimes stop far short of
this, sometimes give the observer the idea that they are the edges
of straight planes, but often show themselves twisted or curved at
the first glance. Many are broken across into separate fragments,
which are separated more or less from each other by conglomerate
or lapilli. (The general appearance of a large portion of the escarp-
ment, as seen from the north flank of the Vesuvian cone, is
shown in the panoramic photograph exhibited, for which I am
indebted to the photographer who executed it shortly before my
survey.) On examining in detail the face of the escarpment, it is
observable that the beds of conglomerate or of lapilli present in
many places evidences of severe compression in both lateral and
vertical directions, and are, over considerable areas, more or less
perfectly glued or cemented together by the percolation of water
having acted upon the partially soluble silicates of which the dis-
integrated material consists, and which have subsequently indurated
and formed the cementing material. This mode of forming a com-
pact mass, which is not an uncommon volcanic phenomenon, has been
observed by myself as taking place in large beds of broken-up slags
deposited in the neighbourhood of the Barron Iron Works in Cum-
berland. The large quantity of lime employed as flux for smelting
the hematite ores operated upon in these vast works produces
slags the silicates in which are sufficiently soluble to form a com-
pact mass, requiring the pick to break it up at a depth of only 8 or
10 feet from the surface, and within a period of about as many
years from the time when the slags were in liquid fusion, the water
acting upon them being derived from rain only. The bank of the
escarpment of Somma thus composed of masses of very variable or
of no coherence, is acted upon by rain falling upon its abruptly
sloping face; and in several places deep recesses or barrancos are
cut back into it, some of which are of great size, and have their
floors at the level of the Atrio covered more or less with a steep
sloping talus of detritus swept down by rain torrents from the lofty
bank above.

On ascending at the western extremity to the crest and following
along more or less nearly the edge of the escarpment upon the
outer slope of Somma, which falls towards the plain at angles
generally of from 25° to 32°, we find it intersected by numberless
rain-channels generally tending down the slope; these in many
instances attain the magnitude of large “nullahs,” or even of
gorges, and, from the loose nature of many of their materials, are
not crossed even in dry weather without some danger and difficulty ;
indeed the walk along the edge from west to east is extremely la-
borious and difficult. It is, however, instructive in several ways,
and especially as shdwing that not very many of the superior or
northern terminations of these dykes reached as far as even the
existing surface, and were far from ever reaching the surface of the
northern slope of Somma, while that mountain still existed as a vol-

PRODUCTION OF VOLCANIC DYKES, 477

cano; for a great depth of the loose material which mainly consti-
tuted the original cone of Somma, a depth which may in some places
have reached perhaps 200 or 300 feet, has been removed and swept
down towards the plain by the action through long ages of rain and of
pluvial torrents. In passing, it may be remarked that the rapidity
with which the incoherent matter deposited upon the flanks of vol-
canic mountains is thus swept away by rain is well exemplified upon
the flank of Etna by the condition of the old oak trees which con-
stitute the Bosco della Casa, which is crossed, three or four miles
above Nicolosi, by the usual mule-path towards the summit. The
slope of the mountain here, the inclination of which generally is very
moderate, not exceeding perhaps 10°, is composed almost wholly of
loose black pyroxenic sand, in which and probably in old tufaceous
beds beneath the oak trees grow.

On examining the trunks of these, it is at once apparent that
the level of the surface during their earlier life was from two or
three to (in several places and over large areas) five feet above what
itis at present, the lowermost part of the bole proper, or trunk, of the
tree, with its rough bark and spreading-out base, marking the original
ground-line, while the tree now stands, as it were, upon its toes,
the great tap-roots, one or several, forming its only connexion with
the present ground (1864), which they penetrate so as to support
the tree. None of these oaks are probably more than one hundred
or one hundred and fifty years old; within that time therefore
a depth of from three to five feet of material has been denuded from
from this part of the mountain over a very large area, and this not-
withstanding that relays of like detritus have during the interval
been brought down by rain from still higher parts of the slope.

The results of my survey may be classed into :— 1st, Observations
on the directions in space and in reference to the material of the es-
carpment of the dykes. 2nd. The lithological and other circum-
stances of the dykes themselves and of the matrix adjacent to them,
which throw some light upon the nature of their production. 3rd.
Conclusions to which these data appear to point.

The apparent horizontal direction in which a dyke observed at the
level of the floor of the Atrio plunged into the bank of the escarpment,
I have called the strike. This direction, when referred to the axis
of the cone of Vesuvius, I have called the orientation of the dyke.
The divergence, if any, from verticality of the plane of the dyke I
have called its dip or hade. Curvature refers to bending round some
imaginary line other than vertical, while twisteng means torsion round
some imaginary line exterior to the dyke and not far from vertical.
The dykes, at their exposed edges, rise to very various heights, some
to the top of the escarpment, generally thinning out as they ascend ;
others only rise sixty or seventy feet above the floor of the Atrio,
and taper rapidly, coming to an edge from a thickness of perhaps
five or six feet at the visible base, and without any observable fissure
extending beyond the summit into the matrix. Some are so curved,
twisted, and often dislocated into fragments of greater or less length
that no deduction as to orientation is possible; but many others,

478 R. MALLET ON THE MECHANISM OF

though not free from twist or curvature or even from cross fractures,
still admit, within moderate limits, of the deduction of their orienta-
tion as well as of their general dip. One source of possible error
which may affect some of my observations I must not omit to men-
tion. From the smallness of the exposed surface and the necessity
of observing each dyke at but a few fect from its base, the only practi-
cable method of determining the strike was by means of the compass.
But as many of these dykes are, like most volcanic products, mag-
netic, and sometimes with polarity, the indications as to horizontal
direction afforded by compass are somewhat uncertain; and the
amount of deviation from this cause, if it exist, is not easily determi-
nable. I do not think, however, that any very large error due to
these causes exists in my observations; for on comparing the directions
shown by two different compasses, one carrying a heavy and power-
ful pivoted bar-magnet, the other a very light needle, both success-
ively placed in the same position and within a few feet of the base of
some of the largest of these dykes, and on also comparing the direction
of both compasses when removed a hundred yards or more from the
base of the same dykes and still in the direction of the plane of the
dyke, I did not find a notable difference of indication, though some-
times reaching a degree or two. The observations of strike ‘thus
made, when combined with the position at the escarpment and level
of the floor of the Atrio, of the terminal edge of each dyke, com-
pleted the elements necessary for the determination of its orienta-
tion or line passing horizontally through its general plane when pro-
duced through the axis of the existing Vesuvian cone.

The orientation of five or six of the dykes thus obtained is such
as to pass, though only approximately, through the axis of the
cone—all the others observed presenting wide disparities in direc-
tion of orientation, which in most cases passed outside the confines
of the cone itself at the level of the Atrio. Indeed in some cases
two dykes of different ages cross one another; and while the orien-
tation of one roughly approximates to the Vesuvian cone, that of the
other passes wide away from it. The divergence from axial orien-
tation increases much as we approach nearer to the eastern end of
the escarpment ; and as we begin to descend from the level of the
floor of the Atrio along the prolongation of the escarpment towards
where it is lost in the eastern slope of the mountain, we find a large
number of dykes whose orientation is seen at a glance to be wide
away from any part of the existing cone of Vesuvius, and whose dip,
or hade, is towards the west and north, and rapidly increases as we
descend, so that several of the most easterly of these dykes present to
the eye the character of old beds of lava inclined at angles steeper
than the slope of the mountain, and dipping into the latter, in which
they are partially buried, rather than of dykes at all. I did not de-
termine the exact strike of any of these dykes or beds eastward
of my number 27, partly by reason of the facts just given, but mainly
because their bases could not be seen from my observing-station at
the base of the cone and could only be approached over a bed of
hugely scoriaceous lava which had overflowed from the Atrio at this

PRODUCTION OF VOLCANIC DYKES. 479

eastern outlet, and descended by the Canale del Inferno. Of the whole
of the numbered dykes, ten present as their main feature an approxi-
mately vertical plane, all the rest having a very sensible dip or hade.
This dip, from No. 1 to No. 13, is towards the west and north at
various angles, none exceeding 38 degrees with the vertical. Be-
tween Nos. 13 and 27 the dip trends towards the east and north, never
exceeding from 30 to 40 degrees, and the average being not more
than about 20°.

I did not observe any case in which a dyke was intersected by a
coherent bed of lava; but in one instance the highest part of one
dyke (No. 1) is stopped at its upper extremity by the underside of
an overlying and nearly horizontal bed of lava. Not only do many
dykes bifurcate or send off branch dykes at different heights, but two
dykes frequently intersect each other and often at considerable
angles, so that in one instance (No. 10) the general plane of one is
nearly orthogonal to that of the other.

These and many other circumstances prove these dykes to have
been produced at very different and successive ages. Many of the
curved, twisted, or broken dykes present proofs of having been sub-
jected to severe and long-continued pressures in both vertical and
horizontal directions, which have resulted in great displacements
from their original positions. Thus the third dyke examined, between
No. 13 and No. 14, has been crushed by descending moyements of
the surrounding matrix, as roughly indicated in fig. 1. In others
lateral displacement has attended also upon vertical movement, as in
the dyke No. 9, which has been cross-fractured with lateral displace-
ment and descent of one segment of the dyke, as indicated in fig. 2.
An ideal figure of this form of displacement has been given by

Fig. 1.—Dyke crushed Fig. 2.—Dyke frac- —
by descending move- _— tured with lateral _— Fig. 8.— Curved
ment of matrix. displacement. Dyke.

VK 4%
Te

vt

Me
py

2.
a

J

4.

4,
AN

4
y

4
%

\
SS AS

na?

pt AD,
Ye Tile
Ue 1
See
SE Soni wl ‘SS

SILLLEL

"

a% XY” " <r

Nr

dykes in that work leaves much to be desired. A third form of dis-
placement is seen in the first dykes examined, between No. 22 and

Q.J.G.8. No, 128. 21

480

R. MALLET ON THE MECHANISM OF

No. 23, which have been abruptly curved in contrary directions and
broken into numerous voussoir-like fragments with but little dis-
placement from each other by pressure while gripped by the matrix,

as indicated by fig. 3.

In the dyke No. 7 there is intersection of one dyke with

another, the surfaces at the intersec-
tion being so altered and dislocated
that it is impossible to say which of the
two is the intruder; both dykes are
here violently distorted and broken, as
roughly shown in fig. 4. In many in-
stances, as in the dyke No. 5, one or
more adjacent dykes have been broken
across into short fragments, twisted,
curved, or straight, which have been
forced away from each other for dis-
tances of many feet or yards, and now
occupy the most irregular positions in
the matrix, as remarkably evidenced in
the dyke No. 16, of which the annexed
figure 5 is a rough representation.
These transpositions and dislocations
are of much interest and importance.

Fig. 4.—Jntersecting Dykes.

We DI hat,
SANT Ay Ah 97%

N
Ps
NI > Wy

~\
Sov

They indicate the vast extent as well as the force of the internal
movements, due principally to gravity, which are constantly taking

Fig. 5.—Dykes fractured and displaced.

re Ws 6S
Qos WS
ws

) ae a
Ss Wi 34

s wyonlls BY ee
q as UU WU

ZEAL

place in the mass of volcanic cones.

Sw

\
<-
MN,

yh
\

a)

10 fin *

WLIPEIYTTII]D-’

Material in active volca noes

PRODUCTION OF VOLCANIC DYKES, 481

being constantly ejected about the upper parts of the cone, and with
a weight in reference to surface increasing with proximity to the
crater, the tendency would be to increase the angular slope of every
volcano as we approach nearer the top. But as the great mass of
ejected material is fragmentary and incoherent, and can therefore
only repose permanently within a certain slope (or angle of repose),
so the tendency to increase the angle of the cone by the deposit of
new material is more or less met by irregular creeping downwards,
and sometimes by slipping of parts of the mass downwards as in
landslips, the discontinuous materials, when thus forced downwards,
often carrying along with them and dislocating or distorting such
connected beds of lava or dykes as may be involved in the mass. An
inspection of the general surface of the escarpment of Somma for its
entire length and height at once shows the observer how small, pro-
portionally, is the mass of lava-beds or dykes included in the enor-
mous surrounding mass of volcanic conglomerate, lapilli, and sand,
and appears to the writer fully to sustain the estimate made by him
in his paper ‘‘On the Nature and Origin® of Volcanic Heat and
Energy ” (Phil. Trans. 1873), that the incoherent material, heated
but not fused, ejected during volcanic eruptions, exceeds by twenty-
fold the mass of the material which is poured forth in fusion.

The relatively attenuated ribs or plates of consolidated lava, present
extremely small resistance to the pressures produced by the enor-
mous weight of the discontinuous material around them, and which,
by reason of its discontinuity, acts somewhat as if it were an imper-
fect liquid or a mere heap of dry sand, in transmitting the internal
movements which are continually taking place within it under the
joint influence of slope and gravitation.

Although the rounding, more or less complete, of the fragments
which constitute beds of volcanic pebbles, large or small, and of the
constituents of conglomerates compacted together after deposition, is,
as has been stated by several authors, produced by mutual attrition
prior to their ejection, there can be little doubt that further com-
minution and rounding is produced by these powerful though little-
noticed movements of descent after their deposit in the mass of vol-
canic mountains. The same phenomena, though upon a minuter
scale, are to be seen in the rounding which is known to take place
to considerable depths in the macadamized stratum of roads which
have been long traversed. In these the originally angular fragments
of broken road-metal are found rounded when the massis broken up
and exposed to view, to far greater depths than the impacts or pres-
sures upon the surface would seem to have been transmitted, intes-
tine movements, extending deep into the mass though not noticeable
at the surface, being sufficient to rub together and round the once
angular fragments to an extent less as their position is deeper from
the surface.

It is obvious, from what has been stated as to the movements of
dykes at periods subsequent to their production, that the present
positions in which they may be found do not necessarily infer any
thing as to the position, whether in orientation or e op, in which

L

482 R. MALLET ON THE MECHANISM OF

they may have been originally deposited. Any such inference as to
a particular case might be grossly in error, and would probably be
so in the case of curved or twisted though more or less continuous
dykes. ‘The error may be greatly less, or perhaps no error may exist,
where the inference is drawn from nearly flat dykes, whose planes
present neither considerable twist, curvature, nor fracture. But as
we never can be certain that any dyke may not have had its position,
especially its orientation, more or less changed or shifted subsequently
to its production by injection, and as we also never can be certain
that any two dykes, whether distant or adjacent, and of which we can
only see some one section of a part of the dyke, which may be chiefly
a vertical section as in Somma, or may be more commonly horizontal
sections as well as vertical ones, such as are found surrounding the
Val del Bove, so we never can be certain that any two, though
now apparently converging, must have emanated from a common
volcanic axis or crater. The evidence as to such common origin
can never be more than probable, though the probability may cer-
tainly be strengthened in some ratio to the number of dyke-planes,
or planes of orientation, which we may observe (after sufficiently
exact geodetic operation) all to appear to converge to a single axis.
The divergences from even approximation to any such single axis,
found by me in the dykes of Somma, at one time raised doubts in
my mind as to whether they were injected dykes at all, or might
not be ancient lava-beds which had flowed down the slope of
the mountain at various epochs, and had, by some unexplained
causes, been broken up, more or less turned on edge in planes
in a line with, or transverse to, the slope, or even occasionally
inverted altogether. This view I was subsequently compelled to
abandon, though some facts observed during the careful lithological
survey which I made of these beds seemed for a time to support it ;
and I should add that one class of facts observed and about to be
referred to, which seemed to support this hypothesis strongly, are
to me still without an explanation and remain for the more com-
plete investigations of others. J may remark here that the descent
of large surfaces of lava-beds, along with the sand and detritus
which, from the causes already referred to, and aided by rains, are
in many instances observed to have descended more or less upon
volcanic flanks, and the turning up of such beds more or less
on edge, either in a direction transverse or approaching to their
line of descent, are by no means uncommon phenomena in volcanic
countries. Ifthe large plate of lava thus descend upon the detrital
slope of an uninterrupted volcanic cone, the lower or forward end
of the huge slab, like a flat coulter, tends constantly to increase
the volume of a ridge of detritus heaped up before its foremost
edge; and a time may arrive when, from this cause and the
slowly digging deeper into the detrital bed of the advancing edge,
the whole huge plate is brought to rest. If sufficiently large,
the detrital matter of the bed is interrupted in its descent by the
upper or higher edge of the plate, while it is washed out more or less
from the lower edge by the percolation of rain, which makes its exit

PRODUCTION OF VOLCANIC DYKES. 483

in rain-channels towards the lower edge. The angular tilt of the
plate therefore continues to increase slowly with time, and to exceed
that of the slope, until at last the lava-plate stands on edge, its
lower part being buried in the detritus, and the upper part in free
air, the plane of the plate in this position being more or less nearly
transverse to the line of slope; and this process, if continued, may
even cause the inversion of the bed, so that what was its upper
surface as it flowed from the crater over the upper part of the slope,
and consolidated there or further down, becomes the lower surface,
upon which it now reposes, and what was the bottom surface would
-then face the sky. But the surface of no volcanic mountain is a
simple cone; its sides, at least so far as they are deeply covered
with detrital material, are deeply furrowed and broken into small
valleys and torrent-beds, having banks of increasing steepness at
either side as they get nearer the bottom, and the general direction
being often with but little sinuosity up and down the slope, as
frequently seen on the flanks of Monte Vulture. Now, if a plate
of lava descending under the mechanism just described should reach
the brow of one of these rain-gullies, its course of descent directly
down the mountain-slope may be gradually changed, and it may
begin to descend by quite the same mechanism along the banks of
the gully in a direction diagonally transverse thereto, and may
finish its course by being turned up on edge as before described,
but now with its plane in a direction more or less nearly orthogonal
to that just described, and in an approximately vertical direction,
its plane being directed more or less nearly towards the axis of
the mountain. If such a plate be subsequently buried in detrital
matter, it will present very much the characters of a dyke of injec-
tion, with this difference, that the two sides of the plate will present
those marked differences always seen in the top and bottom surfaces
of consolidated plates of lava. I have observed several instances of
these remarkable changes of position; thus, for example, in Au-
vergne, on the left and within a short distance of the carriage-road
from Clermont Ferrand and about five miles from that city, passing
over the great volcanic plateau on the direct road to Pont Gibaud,

on the slope of a hill of volcanic detritus, there exists a thick bed of
lava with many included boulders tilted ‘thus and standing several
feet in the air. I observed also several such instances in 1858
upon the flanks of Monte Vulture, upon the side towards Melfi;

and I have been informed by a friend who resided in Mexico that
like facts are not unknown there.

In making my lithological survey of these dykes I was peepee
to notice particularly whether the two opposite surfaces of the same
dyke were alike or different, whether the parts more or less adjacent
to each surface were the same as those of the interior of the dyke,
or differed in character or mineral composition, compactness, frac-
ture, &c., from the material of the latter, or whether any pyromor-
phic changes were observable at or near the surfaces of contact
where a dyke intersected another dyke or a lava bed, whether the
mass of the dyke was amorphous, and if jointed, in what direction

484 R. MALLET ON THE MECHANISM OF

the joints lay, and whether the structure differed at different parts,
being amorphous and jointed into great blocks or having a more or
less slaty or shivered structure conferred upon it.

I gave particular attention to the prevalent directions in which
elongated air-bubbles were found imbedded in each dyke.

It is a fact now familiar to most geologists that the longest
axis of air-bubbles occurring in lava or other igneous rocks is
turned more or less exactly in the direction in which movement or
flow has taken place in the mass while in fusion, just as in most
igneous granites which contain large crystals of felspar or others of
a flattened form, the larger dimensions of these crystals indicate
the line of flow of the granite while in fusion. This is prominently
seen in the granite of Dartmoor in those quarries from which a
large portion of the pavements of the city of Exeter have been ob-
tained. It is obvious, therefore, that in a dyke the longest axes of
these air-bubbles should indicate the direction in which its molten
material has been forced into the cleft. If the injection had been
from below the lowest point of the dyke, and the flow into the cleft
had been therefore approximately vertical, as was perhaps the case
in many of the subaerial dykes of geologists, and, so far as I can
comprehend the not very clearly expressed views of Mr. Scrope,
also in volcanic dykes (Volcanoes, 1862), then it is obvious that
in these dykes of Somma the prevalent direction of the longest
axes of these bubbles should be approximately vertical. If, on the
other hand, the injection of these dykes has been lateral, and
though perhaps inclined upwards, still, on the whole, horizontal, as
being derived from the molten matter, more or less filling a crater,
from which the dykes diverged, then the longest axes of the air-
bubbles should be found nearly coincident with the plane of the
dyke, and if not horizontal, not very divergent by inclination
slightly upwards, or, with much less probability, still more slightly
downwards. Should the prevalent direction of the longest axes
prove transverse to the plane of the dyke, it would cast considerable
doubt upon the fact of the dyke having been filled by injected matter
at all—yjust asin lava streams which have flowed and consolidated on
gentle slopes, although, owing to intestine movements of the still
viscous mass, bubbles with their longest axes vertical are occasionally
found, the greatest preponderance of these axes is in the direction of
the flow, and were we to find the preponderant direction to be
vertical, it would cast much doubt upon whether there had been an
approximately horizontal flow at all.

I also observed the characteristics of each dyke as indicated by
its included minerals. On this last point I may at once remark
that although it is generally and correctly stated by authors that the
distinguishing characteristic between the ancient lavas of Somma,
and the modern ones of Vesuvius is the great prevalence of leucite
in the former, the statement that the dykes of Somma are all of
leucitic lava is by no means correct. Leucite is, no doubt, the charac-
teristic mineral included in the larger number of these dykes; but
pyroxene is also found in most of them, and in some is very preva-

PRODUCTION OF VOLCANIC DYKES. 485

lent, though I am not aware that any one dyke is to be found quite free
from leucite in large or small crystals. One dyke of those examined
by me (namely, 11 4) contains crystals of idocrase in notable pro-
portion. As respects the lithological character of the entire mass of
each dyke, the whole may be divided into three classes :—1. those of
dark-coloured compact and generally subcrystalline lava, which con-
stitute a large proportion of the whole; 2. those of a much lighter-
coloured, softer, and very imperfectly subcrystalline lava of that
sort which Lyell has denominated ‘“ Graystone,” of which material
immense dykes also occur in the escarpment of the Val del Bove
(a few of the larger of these dykes in Somma show, though ob-
scurely, a tendency to a cross columnar structure, one which, in the
case of some of the very largest of these dykes of the Val del
Bove, is magnificently developed, the hexagonal prisms being, in the
case of one very thick dyke examined by me upon the west side,
nearly horizontal and perfectly square to the sides of the dyke,
and in prisms perfectly detachable from each other, and from
ten to fifteen inches in diameter); 3. the third class on Somma
consists of thin dykes of highly compact material, dark-coloured,
and either gray or green, very hard, and with but little appearance
of even minute crystalline structure.

The substance of these may be called phonolite in the grey, and
greenstone in the other dykes. In one of the triple dykes exa-
mined by me, viz. no. 11, a, 6, and c, a and b are in absolute con-
tact with a mere joint of separation, and are nearly vertical, while
¢ appears to meet the others at a point a few feet below the Atrio,
diverging from them at an angle of about 45°, the intervening
wedge being filled in with volcanic conglomerate and pebbles. All
these three adjacent dykes are of different sorts of lava: a is largely
leucitic, and contains very small crystals of olivine, or perhaps of
sphene; 6 is leucitic with crystals of pyroxene, and containing also
idocrase; while ¢ is an extremely hard and absolutely compact
pyroxenic lava, like a greenstone, with a well-marked cross sub-
columnar structure.

These three dykes would therefore appear to be either of different
ages, or derived from different sources of crater-contained lava, differ-
ing in composition and perhaps injected from very different depths.
While the jointing in all these dykes is generally more or less trans-
verse to their plane, evidences of a distinct laminated or slaty cleay-
age, sometimes with complete separation of the flakes, are perceptible
in some of the dykes which have been subjected to severe lateral
pressures, producing curvatures and twisting. The planes of lami-
nation are very generally inclined to the general plane of the dyke,
the strike of both more or less coinciding ; and this structure usu-
ally becomes evanescent at a few yards in length of the dyke, which
then returns to the normal transverse jointing. The dykes num-
bered 5, 6, 7, 10, 12, 13, 15 bes, 16, 22, and 25, all present more
or less distinct evidences of this structure, the crushing forces
producing which do not seem to have been always in a direction
normal to the lamination, but to have shattered the mass by a

486 R. MALLET ON THE MECHANISM OF

diagonal or “wiping” local thrust. These facts present striking
evidence of the immense power of the internal movements which
take place within the mass of volcanic cones at periods long sub-
sequent to original deposition.

Referring now to the disposition of the air-bubbles, these are
comparatively rare in the thin and more compact dykes. They are
not uniformly diffused, generally existing in greater numbers near
the sides of the dyke, although more or less parsemés throughout; in ©
some instances they thickly congregate in zones towards the centre
of the mass, running nearly parallel with its sides. Some large
and generally longish irregular cavities of considerable size, with
very rough internal surfaces, and often a foot or more in length,
occasionally occur, the direction of their longest axes being very
indeterminate. But these must not be confounded with the air-
bubbles, as they are obviously rents produced in the viscous lava as
it was forced onwards. The long axes of the air-bubbles are found
on the whole to be nearly horizontal, or pointing at moderate angles
upwards, and in directions very nearly parallel with the plane of the
dyke at the place where they occur. This is well observable in the
dykes Nos. 1, 5, 6, 9, 11,17, 22, and 27. In Nos.11, 17, and 27
the rents or cavities above spoken of are also to be found.

From this great preponderant horizontality or slight slope up-
wards and coincidence with the planes of the dykes of the long axes
of these air-bubbles, I consider the conclusion is fully warranted that
these dykes have been filled by injection, not from below, in a
nearly vertical and upright direction, but horizontally, as from
some source or sources of liquid lava exercising statical pressure
upon the containing walls, as in the case of volcanic craters filled
more or less with liquid lava. It does not seem possible to assign
any physical reasons why the two surfaces of a dyke thus injected
should present any characteristic differences, in the material
of the dyke differmg near one or other of its surfaces from
the interior of its mass, but not so differing at the other sur-
face, nor any reason why the two opposite surfaces of the same
dyke should present any obvious differences, both being derived
alike from the impress given to the viscous and cooling mass by the
opposite surfaces of the cleft or fissure into which that was injected.
That the parts adjacent to both surfaces of the dyke should
present certain differences in texture may be expected; for if the
walls of the fissure be at a considerably lower temperature than the
viscous mass forced into it, the surfaces of the latter will become,
when cold, more or less vitreous, while the interior of the mass
remains stony or suberystalline, this being in accordance with all
the well-observed facts as to the rapid or gradual cooling of molten
stony masses; but that one surface of a dyke should present this
vitreous character and the other present none, is not easily ex-
plained; and equally difficult is it to explain why one surface
should have its irregularities mainly concave to the general surface
and contain imbedded pebbles picked up and partially involved,
and being of the same sort as those pebbles contained in the matrix

PRODUCTION OF VOLCANIC DYKES. 487

adjacent, while the opposite side of the dyke has its irregularities
mainly convex to the general surface, and sometimes running into
wrinklings such as do not occur on the first-mentioned side.

This difference always characterizes and distinguishes the bottom
from the top surface of lava beds which have flowed over slight
slopes, and which have not become superficially broken up into lumps
of scoria: the lower surface takes concave markings from protuberant
or irregular portions of the surface over which it has run, and
contains stones and pebbles imbedded in the mass which have been
picked up from that same surface; the upper surface of the lava
presents, on the contrary, irregularities which are in the main
convex, protuberant, or bombées.

This diverse character at opposite sides of the same dyke, how-
ever, 1s obscurely presented by some of the dykes examined by
me, in which pebbles are found imbedded in one side and none in
the other, although the matrix at both sides appeared the same. In
others one side only shows the superficial change into vitreous
matter, while the other shows no such change,—the more general
fact, however, being that both surfaces of the dyke show a vitreous
erust, which is often but a fraction of an inch in thickness. The
following dykes show in a marked manner these disparities of the
two opposite surfaces :—In No. 6 A, 3}, feet thick, a thin vitreous crust
at the south-west side, at the opposite side a thick vitreous crust,
with pebbles imbedded in the surface. In dyke No. 8, 8 feet thick,
there are large imbedded but not adherent pebbles on ‘the east face,
none on the other, which is covered with a wrinkled crust. In dyke
No. 9, thickness 33 feet, large and small imbedded pebbles on the
north-east side, with adherent blackish scoriz; opposite side no im-
bedded or adherent matter, and irregularities of the surface convex or
bombées. Dyke No. 20, 6 to 7 feet thick, many imbedded pebbles,
some as large as a man’s head in east face; opposite sides wrinkly
and scoriaceous without pebbles. Dyke No. 22, 3 to 33 feet thick,
east side hollow, with some pebbles and cavities, and wrinkled ; west
side bombé, and contains many air-bubbles. Dyke No. 24, 8 to 9
feet thick, north side bombé irregularly, no crust, pebbles, or
cavities ; south side with deep large and small pits or cavities, and
with adherent crust and pebbles. It will be noticed that these dif-
ferences in opposite surfaces are most observable where the dykes
themselves are thick, indeed are not seen at all in the thin dykes.
The vitreous external crusts generally present the mineral cha-
racters of obsidian passing into a less vitreous material, having the
external character of basanite or Lydian stone, and always very
dark-coloured, the fracture vitreous and subconchoidal. I can only
vouch for the above particulars as affecting such a height in each
dyke above the floor of the Atrio as I was enabled to climb to and
examine. It is remarkable that no pyromorphic change or action
is noticeable in the material lying in contact with or adjacent to any
of these dykes. The disparity in the characters of opposite faces
here described appears to be very difficult of explanation, and
presents a proper subject for further observation here and in other

488 R. MALLET ON THE MECHANISM OF

localities ; I venture to commend it to the attention of volcanic ex-
plorers. Some light may be thrown on the subject by the remarks
about to be made on the mode of production and of injection of the
fissures of these dykes, a subject which has engaged but very slight
and insufficient attention hitherto.

So far as my reading extends, I do not find that any writer on
geology or volcanoes has attempted to frame a physical theory profess-
ing to account for the mode of production of the fissures which, when
filled, form volcanic dykes, nor for the positions they are found to
occupy, 2. ¢ their apparent general fan-like arrangement when
viewed horizontally. Nearly all that Dr. Daubeny has to say upon
the subject is comprised in the following sentence :—‘* Whatever
theory we may adopt with respect to the formation of the beds in
which these dykes occur, it is probable that we shall all concur in
considering them produced by the injection of a leucitic lava into
cracks which previously existed in the containing rock ” (Daubeny,
‘ Volcanoes,’ 2nd edit. 1843, p. 232).

The phrase “previously existed” is indistinct; no crack can be
filled until after it has been previously opened; but how opened, or
how long previously to filling, we are afforded no light by this
passage.

Most English and several foreign text-books make reference to
voleanic dykes in so casual and incomplete a manner that the
reader can gather little more than the general notion that they
appear to radiate from some central axis or axes; even this is very
often not stated, but left to be inferred from the description of the
dykes given. In the late Professor Phillips’s generally admirable
little book upon Vesuvius, an ideal diagram is given at page 1382 of ©
the aspect of a part, at least, of the dykes of Somma; but throughout
the work, where a most complete description and discussion of these
dykes and of the mode of their production might have been looked
for, we do not find a single paragraph which treats of them.

De la Beche, generally so exact as to questions of physical geology,
is little more explicit (Geol. Observer, 2nd edit. 1853, p. 378),
except that he distinctly affirms that fissures formed in the mass of
volcanic cones are filled by the liquid lava, which exercises hydro-
static pressure upon the interior walls of the crater.

Sir Charles Lyell, in his paper already referred to, ‘‘ On Lavas of
Mount Etna” (Phil. Trans. 1858), following Von Waltershausen, by
employing the apparent convergence of thirteen of the dykes of the
Val del Bove as the means of fixing the position of the second axis
or ancient crater which he assigns to Etna at Trifoglietto within the
Val del Bove, tacitly assumes that the clefts in which these dykes
were produced were so formed as all to emanate radially from
Trifoglietto as an axis. He however assigns no reason why this
should be so; and from the words (page 760) “ near the principal
crater, where earthquakes rend the mountain, and where lava is ever
ready to flow into rents,” he appears to be of opinion that the clefts
of these dykes were rent open by earthquakes—admitting which, it is
difficult to conceive why they should pass outwards from the

PRODUCTION OF VOLCANIC DYKES, 489

volcanic axis radially, or indeed why they should show any deter-
minate directions at all in reference to that axis. It seems, there-
fore, that while all authors have tacitly or expressly supposed the
radial arrngement of volcanic dykes, none have made the attempt
to explain the mechanism by which the fissures occupied by these
dykes have been produced, or appear to have discerned that the pro-
duction of a fissure and its filling with molten matter had been
simultaneous and due to the same cause, viz. the hydrostatic pres-
sure of the liquid lava more or less filling the crater. A right
understanding of the physical conditions of this mechanism leads to
some not unimportant conclusions.

A volcanic cone may be considered simply as an embankment
surrounding a central cavity or crater, approaching usually in plan
more or less to a circular form, an irregular cone whose base is the
brim of the crater. The embankment being formed, as Lyell
rightly insisted, merely by the continued projection of materials
from the crater, some poured forth as liquid lava, but by far the
larger proportion projected as dust, lapilli, and stones, must be
regarded as formed by loose material having little or no coherence.
It therefore resists the tendency to being burst open by hydro-
static pressure of lava filling up the crater, merely by the weight or
mass of incoherent matter forming the embankment, the general
vertical section of which anywhere round the axis is rudely trian-
gular, the slope of the interior side being comparatively steep and
constantly changing, while that of the exterior at the upper part of
the cone depends in a great degree upon the angle of repose of the
loose material composing it, while further down, by the degrading
action of rain and other erosive causes, the angle constantly de-
creases to the base, where the mountain rises from the plain.

A straight embankment, like one of those constructed of earth and
clay across a natural valley basin for the purpose of impounding water,
and having usually a triangular section, very much like that which
we have just assigned, the inner or water-slope being usually in the
ratio of 3 horizontal to 1 vertical, while the outer slope is generally
rather steeper (exactly the reverse of the case of the volcanic
embankment, in which the inner slope is small and may even be
evanescent, while the outer slope varies from perhaps 25° near the
top through an average of about 18° for the whole slope, becoming
- evanescent at the bottom or plane), resists the hydrostatic pressure
of the water which it impounds simply by its mass, which is too
great to permit its being caused to slide upon its base by the hori-
zontal component of the fluid pressure ; and so long as the mass of
earth and clay is kept free from the percolation or leakage of water
through it, its resistance is almost wholly independent of cohesion in
the material of the bank, and wholly so as regards cohesion in the
direction of its length. The sides of a volcanic mountain are not
straight, but form an embankment approximately circular in plan
about the cavity of the crater, which it surrounds; and hence
another condition of its resistance to fluid pressure within the crater
comes into play. The radial and horizontal pressure of the fluid

490 R. MALLET ON THE MECHANISM OF

within a ring-formed embankment whose plan is a closed curve, is
normal to the embankment at any point, and is resolved into tan-
gential forces tending to pull asunder or rupture the mass transversely
to its circumference, as well as to push every part of it outwards
upon its base. The degree of cohesion which the material of a
circular embankment possesses is thus an important element of its
resistance to fluid pressure within it; and such resistance in any
circular embankment of given section, material, and depth of cavity,
or of fluid pressure within, decreases with the increase of the
diameter or surface of the cavity within. There is, in fact, a close
analogy between the mechanical conditions of resistance of a circular
embankment and those which resist bursting in a water-pipe, the
cylinder of an hydraulic press, or the chase of a cannon. The
resistance in these depends almost wholly on cohesion of the mate-
rial; and it has been shown by mechanical authors that in a thick
tube, such as a cast-iron smooth-bore gun, the exterior portions of
the metal are greatly less strained by any expansive force from
within than are the interior portions next the cavity; and as soon
as the pressure within that, whether liquid or gaseous pressure,
reaches the rupturing-point, 7. e. exceeds the cohesion per unit
of section of the material at the interior surface, the thick
hollow cylinder begins to burst by fissures passing outwards from
the interior surface, extending in planes radial to and passing
through the axis of the gun or tube. After this point has been
reached, no increase of thickness can add to its strength or prevent
the formation of such fissures, because, as is obvious, the circum-
ferential strains at any distance from the inner surface are every-
where propartional to the normal pressures transmitted to the same
distance ; and as these pressures are only transmitted to any exterior
couche by the compression already sustained by those interior to it,
and as, in accordance with Hook’s law, the resistance in all cohesive
bodies is proportional to the rupturing force acting upon them, so
the exterior portions of the gun or pipe are exposed to a much
smaller strain than are those within, and the strain is greatest at
the interior surface, where, as has been said, the tube first com-
mences to burst.

The construction of modern ordnance in ringed structure, or in
two or more separate tubes, each grasping and therefore compressing
those within it with a certain amount of initial tension, is based
upon the conditions just stated. Were the cavity of a volcanic
crater an exact form of revolution, either a cylinder or a cone, and
were the materials composing the embankment around it perfectly
uniform, and the mass and slope such that the resistance was every-
where proportional to the depth of liquid lava filling it, fissures, if
formed by the liquid pressure, would be in true radial and vertical
planes, like those in which burst cannon are found to divide; and
this would be so whether the cohesive resistance of the material of
the cone were great or small. Like fissures would take place if the
material of the cone or circular embankment were absolutely devoid
of cohesion, as if it consisted of an incoherent mass of sand or earth ;

PRODUCTION OF VOLCANIC DYKES. 491

for all such material is more or less compressible, the resistance to
compression depending upon the size, form, closenes, and mutual
friction of the particles composing it, whether small or great. A
circular embankment composed of such compressible but incoherent
material, when exposed to liquid pressure from within, suffers a
radial compression or approximation of its particles, which is greatest
at the interior surface and becomes less and less as we approach the
exterior. but the radial or normal pressures which produce this
compression or packing up of the particles, also tend by resolution
of force, as has been already explained, to separate or draw off the
particles one from another in the circumferential direction ; and this
drawing asunder of particles, like the compression radially, will be
greatest the nearer we are to the interior surface. In other words
a circular embankment such as we have imagined that of a perfectly
regular volcanic cone composed of compressible though perfectly
incoherent or loose material, must under sufficient liquid pressure
from within begin to give way by the formation of fissures (unless,
indeed, the interior material were viscous or plastic for a consider-
able thickness, the exterior still being compressible, in which case
the interior circumference of the cone would enlarge by mere change
of dimensions without any fissures being formed) commencing at the
interior surface and passing outwards into the mass in planes
whose directions would be perfectly radial and vertical if the
material were perfectly uniform or homogeneous ; and if the thick-
ness of the embankment be sufficient in relation to the pressure
and the time of its continuance, the fissures may never reach the
outer surface.

The hydrostatic pressure of the liquid originates the fissure into
which the pressing liquid at the same time enters and fills it as fast
and as far as itis formed. ‘The fissure is in fact the inceptive of
bursting ; and if the liquid pressure and its continuance be sufficient
the fissure might be prolonged throughout the entire thickness. As
in nature, however, the thickness is generally enormous in propor-
tion to the cavity of the crater, fissures commenced from within are
seldom found to reach the exterior surface of a volcanic cone, but
generally stop short within its mass, partly by reason of the cooling
and solidifying of the molten matter forced into the fissure.

A fissure thus produced and at the same time filled with molten
matter will always be widest where nearest the crater ; and were the
material of the cone which it divides perfectly uniform, the general
form of the dyke produced would tend towards that of a wedge
whose outer edge would be bounded by a curve, the nature of which
would depend upon the vertical extent of the fissure, and the liquid
pressure at increasing depths, and, in nature, of course upon many
other conditions.

The material composing the cone of a volcano, however, is far
from uniform. Besides the material of various sorts and various
degrees of coherence, usually small, there are intercalated beds of
lava of various ages, thicknesses, and qualities, all sloping more or
less outwards and at different depths dividing like shelves the mass

492 R. MALLET ON THE MECHANISM OF

of little-coherent material lying between them. These lava beds
probably seldom if ever traverse the entire circumference. of the -
crater, but merely such portion of it as represented the width of the
overflow or lava stream at its brim. Were such plates or shelves
of intercalated lava free from joints and unbroken, they would, by
their cohesion, otfer great resistance to the opening of any fissure
through them; but from the effects of their cooling after outflow,
and from the immense and unequal pressures to which they have
been subjected by the superincumbent mass of the cone, they are all,
we must suppose (as, indeed, observation indicates), greatly broken.
Still, however, such intercalated plates of lava must greatly increase
the resistance of that portion of the crater-walls im which they
occur to becoming fissured in the way described.

From all these causes, and the concurrence of many other cir-
cumstances which will readily occur to those familiar with the
features of volcanic mountains, it will be obvious that fissures com-
menced at the interior and propagated into the mass ef volcanic
cones can seldom be uniformly distributed round the crater, being
chiefly formed at the weakest places, and can be but rarely pro-
duced in perfectly regular vertical planes having a truly radial
direction from the crater; and the disparity from such normal form
or direction, which must always exist more or less, will become
greater with the increased length of the fissure (that is, with the
distance from the crater), having at every point of its course been
forced through heterogeneous material and such as may have its
directions of easiest division in very different directions at different
points of the path. Thus, for example, the position assigned by
Lyell in the Val del Bove to the ancient crater of Trifoglietto is in
a right line more than a mile from the nearest points of the escarp-
ment in which the orientation of the dykes is observable, and by
the convergence of which the position of Trifoglietto was fixed.
Again, assuming, as 1s commonly done, that the axis of the ancient
Somma was the same in position as that of the present Vesuvius,
the distance of that point from which the dykes of Somma are
supposed to have emanated is more than a mile.

In traversing such a distance through material highly irregular
as regards hardness, cohesion, and direction of easiest fracture, and
remembering that all these may vary in any one dyke at every few
fathoms of its depth, we are forced to conclude that, even supposing
the fissures to have commenced perfectly normally at the respective
craters of Somma and Etna, they can scarcely but have diverged
largely in many cases, both in dip and strike, and in different
degrees at different parts of their height, from the directions in which
they were first started.

A fissure, for example, commencing vertically and, let us suppose,
in a direction due north in the interior of the ancient crater of
Somma, and being caused by the diversity as to material through
which it is opened to diverge gradually, so as to take a curved
direction towards the east or west, must have its orientation changed
to an extent proportionate to the distance it has traversed outwards ;

PRODUCTION OF VOLCANIC DYKES. 493

so that at a distance of a mile from its crater origin, the orientation
of the fragments, as well as the dip of the dyke now observable in
the escarpment, may be wide away from those at the point whence
it is supposed to have started. Further, this change of direction
may be repeated in opposite directions, producing planes of double
curvature such as were actually observed by Lyell and Hartung in a
dyke in Madeira, and as observed by me in some of the twisted
dykes of Somma.

Without disputing the fact that the horizontal coincidence at or
near a single point of the apparent orientation of a considerable
number of the fragments of dykes observed only at considerable
distances, may afford a presumptive probability, greater or less, in
proportion to the greater number of the coincidences and to the
generally straight and unshattered character of the fragments of the
dykes observable, | must yet come to the conclusion from what has
preceded that the concurrence or intersection at a single point of
the apparent orientation of dykes is too unsafe a guide to warrant
our fixing the position of an ancient crater upon such a basis alone,
and that such a mode of determining crater-position, though it may
attain to probability, can never amount to certainty.

In stating this conclusion I wish not to be misunderstood as
casting any doubt upon Sir C. Lyell’s views expressed in his paper
to which I have already referred, as to the ancient existence and
position of a second crater to Etna. The evidence which he has
adduced as to such a crater at Trifoglietto does not rest alone upon
the convergence there of the greenstone dykes observed by Von
Waltershausen and himself, but is also sustained by the synclinal
position, as adduced by him, of the lava and conglomerate beds
towards some line between the existing crater of Etna and that
concluded to have existed at Trifoglietto; and to this evidence as
supporting the former existence of a second crater there is no objec-
tion to offer. At the same time I must remark that the position
of the lowermost beds of any volcano upon the vast scale of Etna
must have greatly changed in the course of ages by the depression
beneath the central parts produced by the enormous superincumbent
weight, which is constantly being added to, of the cone itself, and
from the erosion constantly proceeding irregularly under the base
of the mountain.

If Etna, after having broken through the Tertiary strata, de-
posited its materials on a nearly level surface, that, as the mass of.
the mountain has been heaped up through long ages, must from the
causes mentioned have become depressed towards the central parts
into a shallow saucer-shaped dish; so that the lowermost beds with
the non-yolcanic strata upon them, could we now see them, might
be found to dip more or less from the circumference of the moun-
tain downwards towards its axis, the same phenomena here oc-
curring upon a colossal scale that are well known to happen when
heavy earthen embankments are formed over even very firm soil,
which is depressed beneath the weight, and sometimes more or less
raised at considerable distances from that, but often not sensibly

494 R. MALLET ON THE MECHANISM OF

raised at all. These views, however, tend rather to strengthen
Lyell’s inferences from the contrary dip of the Etnean beds observed
by him.

“lise desirous of doing full justice to the views of M. Elie de
Beaumont, admits that the intrusion of voleanic dykes may increase
the slope of volcanic mountains. It appears to me, however, that
the possible increase of slope thus conceived to be producible has
been much overestimated, the following conditions being taken
into account :—

1. The united or total thickness of all the dykes in the escarp-
ment of Somma, as examined by me, was a little under 150 feet;
and that is the thickness of the intruded matter in the amphitheatre
of 2 miles round the curve, which, regarded as circular, has a mean
radius from the axis of Vesuvius of about a mile. If we even
double this thickness of intrusive matter in order to allow for that
of unobserved broken up or obliterated dykes, intermixed with those
surveyed, and bear in mind that the north slope of Somma at the
level of the Atrio averages about 25°, it can be easily seen that the
increase of inclination, taking the entire original slope of Somma
from the summit to the plain at about three miles, would prove
very insignificant and might be estimated as still less in reference
to the much larger features ef Etna.

2. Although intrusive dykes may occur at all altitudes, they pre-
ponderate in numbers and in thickness about a zone extending
hypsometrically to a height very far short of that of the entire
mountain, few or none probably reaching any considerable depth
below the base, where the horizontal resistance to the fissuring pres-
sure is enormous, and but seldom reaching a height greater than
about two thirds the entire height of the mountain, because the
filling of the crater with liquid lava high above this level is pro-
bably not very common. Hence the effect of such intruded dykes
must rather be to produce gibbosity or some local increase of slope
about the middle zone of volcanic mountains than any general in-
crease in the inclination of the sides.

3. As these dykes are not produced at one time or radially all
round the circumference, but singly or a few at a time, and spora-
dically distributed where the weakest parts at different times permit,
so it seems difficult to assume any noticeable increase of general
slope or of inclination of the internal beds as arising from this
cause.

The length of this paper forbids my entering at all upon the
mechanism by which so many of the parasitic dust cones existing
upon Etna, of which Monte Rosso may be regarded as a type,
and found also in the numerous “ Puys” of Auvergne, have been
opened at one side and become horseshoe-shaped. ‘This has been
commonly attributed by authors to their having been burst by the
hydrostatic pressure of lava, fillmg more or less completely these
steep cones of almost completely incoherent material; and the
streams of lava still observable in some of them at the opened
side have been appealed to in proof of this, though the greater

PRODUCTION OF VOLCANIC DYKES. 495

number of these opened cones show no trace of ever having con-
tained lava. The true nature of the mechanism by which they have
been opened may, I believe, be inferred from the preceding pages.
The opening has been effected not by bursting but by the under-
mining of one part of the base of the crater-wall, either by a small
outburst of lava beneath or, in the case of those opened cones which
present no trace of lava, by the erosive leakage of rainwater filling
to some depth the cavity of the crater. Although these hollow dust
heaps are porous, the bottoms are more or less stanched to rain by a
sort of natural puddling with very fine material washed down from
their inner sides, in the same way as certain portions of the Atrio
del Cavallo have been made watertight and contain ponds of water
in the winter time.

Q.J.G.S. No. 128. Qu

496 HARRY GOVIER SEELEY ON THE

53. On the British -Fosstt Creraceous Brrpos. By Harry Govier
Szetey, Esq., F.L.S., F.G.S., Professor of Geography in King’s
College, London. (Read June 7, 1876.)

[Puates XXVI., XXVII.}
' History.

Tue oldest remains of British fossil birds are recorded by Sir Charles
Lyell as having been first found in 1858, in the Cambridge Upper
Greensand*. This stratum still continues the only member of the
British Secondary deposits in which the bones of birds have been
identified by morphological characters; for though the Rev. Mr.
Dennis had previously asserted the occurrence of birds in the
Stonesfield Slate t, on the evidence of the microscopic structure of
osseous tissue, it is safer, in the absence of recognizable bones, to
believe that the ornithic structure he detected was found in an
ornithosaurian rather than in a true bird. The discovery of bird-
bones in the Cambridge Greensand was made by Mr. Lucas Barrett,
F.G.8., then Assistant Naturalist to the late Professor Sedgwick in
the Woodwardian Museum of the University of Cambridge; but I
am not aware that Mr. Barrett ever published any account of his
discovery. The bones are mentioned by Lyell as “the remains of a
bird which was rather larger than the common Pigeon, and probably
of the order Natatores, and which, like most of the Gull tribe, had
well-developed wings. Portions of the metacarpus, metatarsus, tibia,
and femur have been detected ; and the determinations of Mr. Barrett
have been confirmed by Professor Owen.” What became of Mr. Bar-
rett’s specimens I was never able to find out. They were not in the
Woodwardian Museum when I succeeded to Mr. Barrett’s duties in
1859; and the whole of the remains to which I shall have to refer
were collected subsequent to that date. Professor Owen, in his
‘ Paleontology,’ remarks {:—‘‘ One of the evidences of birds from the
Cambridge Greensand, transmitted to the writer by their discoverer,
Mr. Barrett, is the lower half of the trifid metatarsal, showing the
outer toe-joint much higher than the other two, and projecting
backwards above the middle joint ; it indicates a bird about the size
of a Woodcock.”

In the ‘ Annals Nat. Hist.’ for August 1866 I mentioned, in a
Note on new Genera of fossil Birds, that I had collected or seen from
the Cambridge Greensand cervical, caudal, and dorsal vertebree,
proximal and distal ends of the tarsometatarsus, proximal ends of
tibie, proximal and distal ends of femora, humeri, metacarpal bones,
&ec., and suggested that the typical species should be called Pela-
gornis Barretti. In my ‘Index to Aves, Ornithosauria, and Reptilia,’

* Supplement, Elements of Geology, 1859, p. 40.
t Quart. Journ. Micr. Sci. vol. vy. (1857) p. 163.
t Second edition, 1861, p. 327. AE

BRITISH FOSSIL CRETACEOUS BIRDS. 497

&c., 1869, I abandoned the generic name Pelagornis (which had
been preoccupied by Lartet) for Enaliornis, and recognized a second
species. In that catalogue forty-seven fragments of bird-bones
are enumerated and briefly described. In 1870, in “ Remarks on
Dimorphodon”*, I drew attention to the fact (afterwards confirmed
in American specimens by Prof. Marsh) that some dorsal vertebra
of Cretaceous birds have flat or slightly concave articular ends; and
in other writings I considered the affinities of the birds to be with Oo-
lymbus t and the Penguins. Most of the Upper-Greensand bones were
found in the neighbourhood of Coldham Common near Cambridge, or
at Granchester. When I was residing at Cambridge, and endeavoured
to add to the University collection every portion of the bird’s skeleton
which came under my notice, I was never so fortunate as to secure
a perfect long bone; and the only specimen which I have seen with
the shaft unbroken is a femur in the collection of Mr. Jesson, F.G.S.
Many of the fragments are worn; but most are broken without being
much abraded. And it is quite possible that the specimens may
have been more perfect in the deposit than the state in which they
are collected would imply; but allowance, in judging of this, must
be made for the accidents of fossilization, and the carelessness and
ignorance of the phosphatite diggers, whose interest in these remains
is usually too small to ensure the preservation of inconspicuous bones
which are only met with at long intervals of time; and thus por-
tions of one skeleton, or even of one bone, sometimes get distributed
to several collectors. In the remains of other animals I have seen
several cases in which the fragments of fractured shafts of bones
have lain in juxtaposition in the deposit, and oysters have grown on
the fractured surfaces—showing, I think, that though the nodules
may have been formed and mineralized in shallow water, they were
rolled, with such of the bones as are unassociated, into a somewhat
tranquil depth of sea before they were covered up by the Upper
Greensand sediment.

There is no evidence of more than one bird-bone being found at a
time; so that every fragment of bone may have belonged to a sepa-
rate individual bird.

The Skull. (Pl. XXVI. figs. 1-4.)

A suspicion often occurred to me that the hinder part of the cra-
nium (drawn in pl. 11. figs. 3-6 of my book on the Ornithosauria)
might perhaps be the skull of a bird. In describing it seven years
ago as Ornithosaurian I was influenced chiefly by probabilities and
considerations such as these:—The specimen closely resembles the
back part of the cranium in several Pterodactyles from Solenhofen ;
its Avian characters are in harmony with the characters of other

* Annals Nat. Hist., Aug. 1870, p. 129. + Ibid. Nov. 1871, p. 305.

t May 2, 1864, I communicated a note to the Cambridge Philosophical
Society ‘‘ On the Fossil Birds of the Upper Greensand (Palgocolymbus Barrettt
and Pelagornis Sedgwicki) ;” but nothing beyond the title of the paper was

printed.
2m 2

498 HARRY GOVIER SEELEY ON THE

parts of the skeleton of Ornithocheirus from the Cambridge Green-
sand, and it closely resembles the specimen figured on pl. 11. figs. 1
& 2, the Ornithosaurian nature of which I see no reason to doubt ;
while it seemed improbable that so many specimens (as were known
to me) of birds’ skulls should have been found, and so few of Ptero-
dactyles—the bones of birds being so rare, and those of Ornithosaurs,
especially jaws, being relatively abundant. Confessedly it is ex-
tremely difficult, if not impossible, to distinguish the hinder part of
the cranium in some Natatorial birds from the corresponding region
in some Ornithosaurs merely by the form of the head, the arrange-
ment of its regions, or the proportions of its several osseous elements.
When first describing the specimen referred to, I pointed out* the
difficulties in the way of distinguishing the fossil from the cranium
of a bird. The withdrawal of the specimen from the Ornithosaurian
list would leave the question of the Avian nature of Ornithosaurs
unaffected, as in most respects some of the admitted crania of Ptero-
dactyles are almost as bird-like as this specimen.

I propose now to regard the skull as that of a true bird :—first,
because, on careful reconsideration, I find no characters by which it
can claim more than a generic distinction from living birds; secondly,
it is more like a bird’s skull than is the skull of any British Ptero-
dactyle ; thirdly, the cranial bones are comparatively thin and dense,
instead of having the cellular structure characteristic of Ornithosaurs
from the Cambridge Greensand, and in this character the skull agrees
with the other portions of the bird’s skeleton. It is about the size
of the Red-throated Diver’s cranium, and, in common with all the
other bird-bones found, presents a marked resemblance to that Avian
type. ‘
I have already described elsewhere? the essential features of this
specimen, and now offer a few measurements of its several regions,
which were then omitted.

Transverse measurement over the lateral processes which form
the articulation for the quadrate bones, 13 inch.

Transverse measurement of the constricted squamosal region of
the skull just in advance of the preceding measurement, ? inch.

Transverse measurement of anterior termination of the squamosal
region 42 inch.

The parietal region rapidly rises as it extends forward from the
occiput.

The parietal bones, which met in the median line in a slight
ridge, are oblong, ;2; inch long, though, owing to the anterior mar-
gin being concave, each becomes much wider (;% inch) where the
parietal joins the squamosal bone in a nearly straight suture, ;% inch
below the median cranial crest. There is a bend in the upper third
of the parietal bone; the superior portion looks upward and out-
ward; the lower portion is more inflated than in the Diver, but is
similarly directed outward a little to meet the expanded frontal.
The parietal bone (; inch thick) had a smooth union with the

* Ornithosauria, pp. 80-83.
t+ Ornithosauria, p. 80, pl. 11. figs. 3-6.

BRITISH FOSSIL CRETACEOUS BIRDS. 499

frontal, which it obliquely overlapped. The parietal appears to
have been prolonged further forward in the line of the median crest
than is evidenced by the preservation of the bones. The bones are
transversely arched.

The squamosal bone differs from that element in the Divers, Cor-
morants, and Grebes (which in general form of cranium approximate
to the fossil) in being folded so as to form a narrow inferior surface
more than } inch wide and about 2 inchlong, and a superior tri-
angular, somewhat V-shaped area fully j inch deep and as wide.
In this folded character there is some approximation to the condition
of the squamosal bone in the Goose tribe.

The occipital crest seems to have been well elevated. It is sinuous
and curves forward laterally, as in the Diver, but is more elevated
inferiorly than in that type, and has the articulation for the qua-
drate bone placed entirely behind it, which is not the case in the
Diver. The occipital aspect of the fossil is wonderfully like that of
Colymbus, and differs chiefly in the fossil being somewhat broader,
owing to the two sides of the occipital crest making a greater angle
with each other. The basioccipital articular condyle in the fossil is
probably removed by fracture. The Diver has the bone in front of
the condyle vertically compressed ; but in the fossil the basioccipital
was compressed even more, since the fracture has left but a narrow
rim to the base of the foramen magnum. The foramen magnum is
vertically ovate, and relatively rather larger than in the Diver; it
is 3 inch high, and ;5; inch broad ; its upper border considerably
overhangs the base.

The tympanic area is large, internal to the articulation for the
quadrate bone, and is best exposed in lateral view. It is limited
by four irregular sides, and has an irregular surface in which are
several perforations. The basitemporal bones, which are missing,
were shaped as in birds. The remarkable short-triangular form of the
base of the sphenoid may result from abrasion. It is broader than
in water-birds, but is essentially a modification of the sphenoid of
the Diver. As preserved, it is ; inch long and ;°, inch wide at the
base. This specimen I regard as the skull of Enalornis Barretti.

Another specimen, in the collection of Mr. Reed, at York, probably
belongs to another species.

A fragment which I regarded in 1871 as a premaxillary of a
Toothless Pterodactyle has perhaps an equally good claim to be con-
sidered a portion of a premaxillary bone of a bird. It has already
been figured by Prof. Owen as the proximal end of the metacarpal
of the wing-finger of an Ornithosaur*. I have suggested for
it (Ann. Nat. Hist. January 1871, p. 35) the name Ornithostoma ;
for, if a bird-bone, it has nothing in common with Enaliornis ; and if
an Ornithosaur, the name will be appropriate. It has the base flattened,
concave from side to side, marked with blood-vessels, and terminating
laterally in well-defined ridges gently rounded, which are concave
in outline from front to back, so as to make the back wider than the
front. If the specimen should hereafter prove to be a jaw, this

* Pal, Soc. 1859, pl. 4. figs. 4 5.

500 HARRY GOVIER SEELEY ON THE

would be the palatal surface. It is 27 inches long, ? inch wide in
front, and 1, inch wide behind. The sides are smooth, finely
wrinkled, concave in length, convex from above downward, with
small oblique vascular foramina near the border, which may be
alveolar ; there are also foramina on the middle of the side. Supe-
riorly the sides converge in a rounded ridge. It is 13 inch deep in
front, and 13 inch deep behind. ‘The great difficulty i in the way of
regarding it as an Avian premaxillary is that, although it widens
behind somewhat rapidly, it shows no indication of a nasal groove
or cavity, though the Diver has no lateral groove on the beak. And
the recent discovery of Toothless Pterodactyles by Prof. Marsh in
the Cretaceous beds of America lends some support to my original
determination of the bone as Ornithosaurian.

I would suggest that since the birds from the American Greensand
having teeth in their jaws show affinities with Colymbus, it is pro-
bable that Hnaliornis, which has vertebre resembling those of
Ichthyornis in biconcave character, may also have had teeth in its
jaws.

The Vertebral Column.

Cervical vertebra have rarely been found. Some twelve years ago
I detected what seemed to be a lower cervical in the collection of
W. Reed, Esq., F.G.8., but have not noticed any other specimens.
The yvertebree most commonly found, and best known, are dorsal.
One in Mr. Reed’s collection has the visceral surface of the centrum
terminated by a sharp ridge, while all the other specimens (including
four in the Woodwardian Museum) have this surface convex. Mr.
Reed’s two vertebree probably belong to the two species of birds to
which I refer the bulk of the remains, though this could only be defi-
nitively determined by a careful study of the somewhat abraded fossils.
Of the sacrum, the Woodwardian Museum has three important speci-
mens demonstrating the essential points of structure ; another frag-
ment of the postfemoral part of a sacrum has been obtained by James
Carter, Esq., of Cambridge. These specimens are all from the an-
terior and middle regions of the sacrum, and are such as might all
belong to one genus. The Woodwardian Museum also contains two
other vertebra, apparently proccelous, which I am disposed to re-
gard as bird-bones, and refer to the tail. As they have no trans-
verse processes, they are probably elements which, in existing birds,
are united in the ploughshare bone.

Lower Cervical Vertebra. (Pl. XXVI. figs. 5, 6.)

The centrum measures © inch in bleh over the anterior zygapo-
physes ; the length of the Bet slags is ;& of an inch; the width of
the centrum in ‘front is nearly ~ ink "where the anterior articu-
lation is concave irom side to side, convex from above downward ;
it is more than 54 inch deep ; and at its base a short tubercle is
developed. ‘The base is concave, long, narrow, flattened, and con-
cave from side to side. It is margined by sharp concave borders.

BRITISH FOSSIL CRETACEOUS BIRDS. 501

The lateral areas are concave; the posterior articulation is slightly
concave in depth, and more convex in breadth: it is nearly square,
but broader above than below. The neurapophyses rise obliquely
from above the upper part of the centrum behind, widening supe-
riorly to above the neural canalin front. The depth in front of the
centrum and neural arch is more than 2 inch. The upper surface
is obscured, and the neural spine is not preserved; it does not rise
from quite the front of the arch, which is nearly horizontal above.
The facets for the zygapophyses look upward and inward.

This specimen, in the collection of W. Reed, Esq., is not unlikely
to belong to Hnaliornis Barretiv.

Dorsal Vertebra. (P1. XXYVI. figs. 7-13.)

The four dorsal vertebre in the Woodwardian Museum, like all
the other specimens which have come under my notice, except Mr.
Reed’s, agree in their characters, though they differ a little in size.
They are small in proportion to the size of the head, and show no
resemblance to any vertebrae of the Diver. Two of the centrums
of Enaliornis Sedgwicki (fig. 12a) display to some extent indications of
the peculiar Ichthyoid biconcave articular condition already referred
to; but, as it is wanting from the largest specimen, it is possible
that the smaller specimens though adult are imperfectly ossified ;
many natatorial birds have the lower dorsal centrums nearly flat, .
while very young birds have a small central notochordal depression
on the articular face. The centrums are much less compressed from
side to side than is usual in the dorsal vertebra of water-birds.

The largest specimen, Hnaliornis Barretti, has the centrum rather
more than 2 inch long, expanded at both articular ends, somewhat
flattened at the base and at the sides, which round into each other
(fig. 9), and are each concave from back to elas The anterior ar-
ticulation (fig. 11a) is 2 inch broad and j inch deep. It is re-

markable for having the sides of the centrum prolonged forward for
a considerable distance, so that the surface is markedly concave from
side to side, while vertically the convexity, if it exists, is so slight
as to be scarcely detected. Posteriorly the articulation (fig. 106) is
necessarily narrower; it is more than j inch wide and less than
z inch deep. It is convex from side to side, flat from above down-
ward, with a central concavity, is subquadrate, emarginate in the
upper part of the side, with a slight impression dividing the basal
margin. Laterally (fig. 7¢), at the side of the anterior border of
the cent arch is an oblique oval facet slightly elevated, 3 inch
long and concave, which has exactly the form and position of the
diapophysis in the vertebrie of existing birds like the Gannet. The
whole centrum closely resembles that of the Solan Goose, differing
chiefly in being much more depressed, and not much more than half
the size. The form of the neural arch, however, much more
closely resembles that of Colymbus. A narrow, compressed, hori-
zontal, transverse platform is given off (fig. 8d). It is emarginate
anteriorly and posteriorly, and directed outward and somewhat back-

502 - HARRY GOVIER SEELEY ON THE

ward; itis imperfectly preserved. Small zygapophysial facets (fig. 8g)
are preserved or indicated both in front and behind. The neural
spine (fig. 8¢), as preserved, is thicker posteriorly than in front,
does not extend quite to the anterior border of the neural canal,
and posteriorly is cleft vertically. Specimens of Enahornis Sedgwicki
(fig. 13) show the centrum to be considerably impressed laterally
below the transverse process *, and that the neural canal is much
wider than high; the neural spine is directed somewhat backward.
The shortest centrum is more than 7 inch long. The amount of the
concavity of the centrum varies.

The Sacrum. (Pl. XXVI. figs. 14-19.)

There are three fragments of the sacrum in the Woodwardian
Museum, and a fourth (from the posterior part of the sacrum) in the
collection of James Carter, Esq., M.R.C.S. The Woodwardian spe-
cimens appear all to belong to the same genus, are in sequence to
each other, and show the sacrum to have been constructed exactly
after the water-bird pattern. The vertebrae, however, are more
elongated, so that the sacrum is much longer in its several regions
than are the corresponding parts of the sacrum of the Diver. The
first fragment (fig. 14) comprises the first, second, and part of the third
sacral vertebree anchylosed, with the elevated margins of the con-
fluent centrums unusually well marked on the basal border. The
first centrum is depressed, with the articular surface flattened in
front, as in the Gulls and the Gannet, and somewhat concave cen-
trally. The anterior articulation is nearly 2 inch broad. The
centrums increase a little in length from before backwards; the
first is about =, inch long, with a depressed form of centrum, broad
and: convex from side to side, and slightly concave from back to
front. I am unable to detect any indication of the diapophysial
articulation on the anterior border of this centrum usually to be
seen in existing birds. A narrow transverse process, 3 inch wide, is
given off from the anterior part of the side of the centrum and
directed outward and somewhat upward ; its outline is emarginate
both anteriorly and posteriorly. In the sneceeding vertebre the
centrums become a little deeper, and rapidiy more compressed from
side to side, the third having the sides flattened, and base forming a
narrow rounded ridge. The third vertebra shows the transverse
processes, which, thick and strong, appear to have formed horizontal
neural tables, the bases of which, as preserved, have a transverse
measurement of ? inch. From the neural platform to the base
of the centrum is more than 2inch. No indication of the neural
spine is preserved.

The second specimen (figs. 15, 16), from the middle of the sacrum,
is in a pale state of mineralization, and’more free from matrix. It
was collected by M. R. Pryor, Esq., late Fellow of Trinity College. It

* The impression is never as deep as in Gulls, or even as in the Wealden
fossil Ornzthopsis, a genus so named from the resemblance of its centrum to
that of certain Gulls.

BRITISH FOSSIL CRETACEOUS BIRDS, 503

comprises portions of three vertebrae—one entire in the middle, and a
fractured vertebra at each end. In general form the vertebre rather
recall the middle sacral region of Penguins. The centrum is rather
more than 3 inch long, while the least width at the intervertebral fora-
menis;°;inch. The sides, concave in length, are compressed so as to
conyerge inferiorly in a rounded median keel. The vertebral fora-
mina are large, rounded, and placed behind the transverse process
just at the hindermost part of the centrum (fig. 16). The neural plat-
form is wide, horizontal, and imperfectly preserved; it is traversed
in the median line by a low neural spine about, inch high and as wide.
The depth from the neural spine to the base of the centrum is fully
7 inch. ‘The substance of the centrum is largely excavated for the
neural canal, which is cylindrical and j inch in diameter. The
position of the transverse processes is above the middle of the neural
canal. Both the specimens described are from the part of the sacrum
anterior to the femoral articulation in the pelvis. i

The third specimen (figs. 17, 18, 19) is postfemoral ; it has almost
exactly the form of the region immediately postfemoral in the Diver.
It comprises portions of four vertebre perfectly ankylosed together.
In transverse section the mass is triangular (fig. 17), having the base
flattened ; and the flattened sides converge dorsally to a sharp ridge,
with which the iliac bones were in contact. The specimen is 13 inch
long, ;& inch high as preserved anteriorly, and about 3 inch high
posteriorly. The base (fig. 18), which is channelled in length as in the
Diver, is margined oneach side byasubangular ridge, external to which
is a narrow flattened lateral area which looks downward and out-
ward. The width of a centrum in front, where narrowest between
the transverse processes, is less than 2 inch; posteriorly the sacrum
becomes steadily narrower. On the lower portion of the lateral
aspect (fig. 19) is a row of four large rounded intervertebral foramina,
less than 3 inch in diameter, between which are the eminences forming
the bases of the short transverse processes. The flat side above
these foramina and processes, =°; inch deep, is divided into rhomboid
areas by impressed lines running upward and backward at unequal
distances from each other. Posteriorly at the fracture the sacrum
is becoming greatly compressed ; but there is no means of estimating
the number of vertebre in the posterior missing portion. In the
Diver there are six yertebre posterior to the corresponding sacral
region; but if Hnaliornis had a long tail, the sacrum may have in-
cluded fewer postfemoral elements.

Caudal Vertebre. (Pl. XXVI. figs. 20, 21, 22.)

Two small vertebrae, apparently caudal and proccelous, I am in-
clined to assign to the caudal region of Enaliornis ; if so, they are
separate elements of the ploughshare bone, and would indicate a
peculiar condition of the termination of the tail. The centrum is
compressed from side to side; and the neural arch, which hangs for-
ward a good deal (fig. 20), shares in the compression. The centrum is
about 7 inch deep, and ;%, inch in width superiorly, where it is widest ;

O04 HARRY GOVIER SEELEY ON THE

and the base is rounded, with a narrow median furrow. The arti-
culations of the centrum are oblique, looking downward and for-
ward; the anterior one (fig. 22)is concave from above downwards, and
shows a less concavity from side to side. The posterior articulation
(fig. 21) is evenly convex, somewhat compressed from side to side,
broader above than below; it is inch deep, and 53, inch wide supe-
riorly. On the middle of the side of the neural arch, on a line with
the base of the neural canal, is a slight eminence indicative of arudi-
mentary transverse process. The zygapophyses are broken away, but
extended far forward, and were elevated to the upper border of the
neural canal. The neural arch in front is wider than the centrum ;
posteriorly it is compressed and narrower than the centrum. There
appears to have been a slight neural spine; but it is broken away.
As preserved the vertebra is ;4 inch high. The neural canal is
small, but is larger in front than behind.

The vertebral column, as a whole, is unlike that of any existing
bird ; but its affinities with members of the natatorial tribe admit of

no question.
The Pelvis. (Pl. XXYVI. fig. 23.)

Only one fragment has come under my notice ; it is in the Museum
of the Geological Survey, and had already been identified by Mr. E.
T. Newton, F.G.S8. I am indebted to Professor Huxley for the op-
portunity of studying this and other bird-bones in the Geological-
Survey Museum. The ilium is large and deep, the acetabulum
is perforated, and the small, but distinct, public and ischiace bones
are directed backward.

The specimen is a fragment of the left ilium (fig. 23 a) with the
ischium broken short, and an indication of the pubis. The acetabulum
is ;°- inch wide ; it is subcireular, and is perforated as in living birds,
the perforation being apparently nearly 2 inch wide. The external
surface of theilium is smooth ; but the posterior margin of the aceta-
bulum is elevated considerably, as in existing birds, the anterior
margin being similarly depressed. The pubis (fig. 23 6) is seen on the
internal side of the fragment not to be blended with the ilium, but to
be separated bya short straight suture directed upward and backward.
The pubis is directed backward, but forms the inferior anterior mar-
gin of the acetabulum; it is slender, 54, inch wide, subquadrate in
section, with the external and internal anterior sides converging in
a sharp anterior marginal ridge ; consequently there are also poste-
rior, external and internal ridges. As preserved, the pubis is =% inch
long; it has a large medullary cavity. The ischium (fig. 23¢) is
more slender and more circular in section; it forms the inferior
posterior part of the margin of the acetabulum; is flattened ante-
_ riorly, and directed backward at the same angle as the pubis, from
which it appears to be entirely separated: as preserved, it is 3 inch
long. The ilium is remarkably thin; it is compressed above and
behind the posterior acetabular thickening, above which (as pre-
served) it extends for 7 inch. The length of the specimen is about
1 inch.

BRITISH FOSSIL CRETACEOUS BIRDS. 505

The Femur. (Pl. XXVII. figs. 1-12.)

More specimens of femur have been found than of any other bone.
The Woodwardian Museum contains twenty-four, of which eight are
proximal ends. ‘This is nearly half the total number of specimens
of bird-bones in the Museum. In the collection of Mr. T. Jesson,
F.G.S., in which there are seventeen bird-bones, six are fragments
of femora—five of them distal ends, and one proximal. On the
lowest computation these remains alone must have pertained to at
least thirteen birds. The femora differ in size, form, and slender-
ness, but not to a marked degree; so that if the bones belonged
to three different species, as is probable, they all may have been
closely allied. The smallest femur is rather longer than that of the
Red-throated Diver, but has the articular ends less expanded. One
specimen in Mr. Jesson’s collection (figs. 1, 2, 3) is 14 inch long,
but has both articular ends worn and short of their true length.
It is bowed superiorly, much as in the Diver, being concave in length
inferiorly and convex superiorly. The shaft measures } inch from
within outward in the middle, and widens proximally to 2 inch as
preserved, and distally to ;% inch as preserved.

The external surface is flattened, somewhat pitted with muscular
attachments, rounded into the anterior and posterior surfaces, and is
concave from the proximal to the distal end. The antero-posterior
diameter of the shaft in the middle is more than ; inch, but becomes
somewhat less towards both ends. The internal surface is somewhat
compressed on the posterior margin, but is well rounded in the shaft
from back to front, though towards the extremities it becomes more
compressed. The proximal end is subtriangular when seen from
above, the superior antero-posterior external measurement being
much more than in Co/ymbus, while the articular head of the bone
is smaller; and the shaft is slightly compressed superiorly external
to the articular head, the impression being greatest in front. The
distal end is marked in front with a shallow broad channel, nearly
the width of the anterior surface, margined on the inner side by a
sharp ridge, which is the edge of a flattened, narrow, short inner
surface that. looks inward and backward. As usual in birds, the
proximal and distal articulations are both in the same plane, and
the outer distal articular surface is the larger of the two.

Another femur, imperfect proximally (fig. 8) (presented to the
Woodwardian Museum by Rev. T. G. Bonney, F.G.8.), appears to
have been shorter and stouter, approaching nearer to that of the Diver,
and the muscular attachments very like those of the Diver. As
preserved, the fragment is 17 inch long and =; inch wide in the
middle of the shaft; distally it is about 4 inch wide. At the
distal end there is wide pit rather than a groove, resembling that in
the Red-throated Diver, only not so deep.

A third femur, badly preserved distally, and showing no trace
proximally of the articular end, is 1? inch long; distally it is 4 inch
wide, while the larger condyle measures 2 inch from front to back.
This indicates an animal somewhat larger; but it is difficult, with

506 HARRY GOVIER SEELEY ON THE

such worn fragments, to estimate aright the differences of size and
stoutness.

The isolated ends exhibit corresponding differences of size. The
smallest measures ;% inch from within outwards over the proximal
articulation, while the largest specimens are ;% inch in that mea-
surement, and have a corresponding thickness in the shaft. In
all, the external proximal surface terminates in a sharp margin
on each side, is flattened and wide—;% inch wide in the larger spe-
cimens, and less than ;3, inch wide in the small bones. Usually
this area is somewhat concave below the articulation. ‘The articular
surface itself extends over the whole of the proximal end.

A medium-sized distal end of a left femur (Pl. XXVII. figs. 9, 10,
11) I refer to Enaliornis Sedqwicki. The fragment is about an
inch long, subovate where fractured proximally, being =; inch in the
greatest oblique measurement from before backward, and less from
side to side. The shaft becomes compressed on the inner and pos-
terior side towards the distal articulation, above which its section
would be subtriangular ; for the inner margin of the anterior aspect
is a somewhat sharp ridge more marked than in the Diver, which
is prolonged into the inner distal condyle. The outer condyle is
much the larger, and extends about 5% inch further distally than
the other. The intercondylar space in the fossil has not so great
an antero-posterior compression as in the Diver; and the inner con-
dyle is rather deeper than in that type. The width over the con-
dyles is 5% inch; the depth of the outer condyle is nearly ~4 inch.

A larger distal end of a right femur (Pl. XX VII. fig. 12), which I
regard as that of Hnaliornis Barretti, more than an inch long, is
nearly 2 inch wide over the articulation; and the whole bone is
proportionately larger. The condyle is relatively larger and deeper
than in the other species, and the anterior channel between the
condyles is relatively deeper. Another specimen, in better preserva-
tion, I noticed in the collection of W. Reed, Esq. I shall be quite
prepared to find that the femora which have passed through my hands
beiong to more species than I have indicated; but, in the absence
of other evidence, I do not see my way to giving them any useful
definition.

The Tibia. (Pl. XXVII. figs. 13-21; Pl. XXVI. figs. 24, 25.)

The remains of tibiz comprise two proximal ends of right tibiee in
the Woodwardian Museum, and one in that of Mr. Reed, and a left
one in the collection of Mr. Jesson, all of which have a moderate
patelloid process, which is more developed than in the Gannet, but
makes no approximation in length to that of the Diver. There are
also in Mr. Jesson’s collection two specimens of right and left proxi-
mal ends of the smaller of the two forms of tibise, which are imper-
fectly ossified and have no trace of a patella, but terminate in rough
cartilaginous surfaces. These bones seem to be young specimens of
Enaliornis Sedgwicki, though the remains are not inferior in size to

BRITISH FOSSIL CRETACEOUS BIRDS. 507

those in which ossification is complete. In the Woodwardian Mu-
seum is a similar larger proximal end of a tibia without a patella,
which I take to be that of Enalornis Barretti (figs. 20, 21). There
are also distal ends of tibie: the Woodwardian Museum is fortunate
in possessing an example of each species.

The large proximal ends of the bone in Enalornis Barretti are four
in number. Mr. Reed’s specimen is the finest. Twelve years since I
noticed that it was 1} inch long, with a subquadrate shaft and strong
compressed semiovate patelloid process fully 2 inch high, with flat-
tened subparallel surfaces. Both lateral margins of the patella
develop ridges which are directed outwards. The ridge on the
fibular side of the bone is the stronger. The shaft terminates upward
internal to the patella in a flattened horizontal slightly convex sur-
face wider than deep, though the depth is increased by projecting
backward over the posterior face of the shaft. This articulation for
the femur is nearly 4 inch wide, but is not so deep on the inner side
and is rather narrower towards the fibula; the surface is slightly
oblique, sloping from the front backward. In front the shaft is
channelled ; and posteriorly it is somewhat concave from side to side.
It is compressed externally and develops a fibular ridge on the pos-
terior external margin at 2 inch below the femoral articulation.
The internal surface of the shaft is flat. The proximal end figured
Pl. XXVII. fig. 20 is in the collection of T. Jesson, Esq., F.G.S.

The distal articular end of the right tibia of Knaliornis Barretti in
the Woodwardian Museum (Pl. XX VI. figs. 24, 25) has the shaft
much compressed from back to front, and sharp along the fibular
margin. The articulationis2inch broad. The(?) tarsal element is as
completely ankylosed to the tibia as in most Ornithosaurs and existing
birds. The condyles are broad and rounded, the internal one being
the larger of the two. The depression between the condyles appears
to be less deep than usual. There is a deep channel in front on
the fibular side; but it does not appear to have been arched over by a
bridge, but was defended by a strong process directed outward from
above the inner condyle. This is one of the most distinctive parts
of the skeleton of Hnaliorns.

The tibia of Enaliornis Sedqwicki, like all the other bones, is
smaller than the tibia of the species just described. The Wood-
wardian fragment of a right proximal end (figs. 13, 14, 15) is
about 13 inch long. The shaft, where fractured at the commence-
ment of the alae ridge, is triangular i in section, less than 2 inch
wide, and more than } inch deep on the slightly convex internal
aspect. The other surfaces of the shaft appear to be more flattened
than in £. Barretti, the vertical channel on the anterior surface
being conspicuously shallower. The proximal articulation for the
femur is subquadrate, more than #2 inch wide, and not quite so
deep. The articular surface consists of a flattened internal sur-
face for the condyle, and a mamillate process rising above that
level, which extends to the outer margin of the bodes showing
that the external condyle of the femur was supported entirely on
the fibula. The patelloid process is shaped much as in the other

508 HARRY GOVIER SEBLEY ON THE

species, and is 4 inch wide. It is ovately rounded proximally,
strong, compressed from front to back, and extends a little out-
ward and forward beyond the shaft, where it terminates in a sharp
ridge, which is directed forward and continued a short distance
distally down the shaft. On the inner side there is no ridge on the
patella; but a sharp angle extends from its base down the shaft.

The distal end of a right tibia of this species in the Wood-
wardian Museum is well preserved (figs. 16, 17,18). The shaft
is much compressed from front to back, and set on to the small
distal articulation so as to be flush with it behind. The shaft
measures ;°; inch from side to side at about 2 inch from the distal
end, and is nearly ;4, thick. The distal articulation widens to ;% inch;
posteriorly it extends up the shaft for 2 inch, forming a broad shal-
low channel margined laterally by two strong ridges which are con-
tinued round into the condyles (fig. 18), forming their external mar-
gins, and defining the lateral regions of the lower part of the shaft.
The convex base of the inner side of the shaft 1s more than 7 inch
wide, while the base of the inner side, which is rounded superiorly, is
=; inch wide. The anterior surface (fig. 16) of the shaft consists of
an external half, which is deeply excavated, and an internal part which
is somewhat thickened, and at its base, just above the middle of the
articulation, gave off a slight process which appears to have been
directed over the channel. The margins of the articulation are ele-
vated ; but the space between them is concave from side to side. The
internal condyle is rather the larger of the two.

Fibula. (Pl. XXVI. figs. 26, 27.)

This identification is less evident than are the others. But as no
bone of the fore limb, or scapular arch, or sternum, has hitherto
been recognized among the bird-remains from the Cambridge Green-
sand, the only bone left to be identified which could unite to a slen-
der shaft a moderately expanded proximal end, is the fibula. The
bone is, relatively to the tibia, much larger than in existing birds ;
but since Professor Marsh finds this to be a characteristic of the
fibula in fossil birds from the American Cretaceous deposits, it is
probable that Hnalornis had a community of structure with the
transatlantic genera in this particular as well as in general affinities.
Two specimens have been preserved in the Woodwardian Museum,
both probably from the left side. The longest is more than an inch
long, is compressed from side to side so as to be four-sided at the
distal fracture, where it is + inch wide and less than =, inch thick.
Proximally the shaft becomes wider on one of the narrow sides, and
compressed on the opposite side, so that in section it is triangular,
and wider in both dimensions than lower down. On what I take
to be the inner posterior margin is a slight sharp ridge, which may
be an indication of the ridge by which, lower down, the fibula united
with the tibia. The proximal end is expanded and articular, sub-
triangular, inclined a little inward, and somewhat rounded superiorly
from side to side. It measures ;4 inch from back to front, and

BRITISH FOSSIL CRETACEOUS BIRDS. 509

2 inch from within outward ‘on the posterior side, and less on the
anterior side.

Proximal end of Metatarsus. (Pl. XXVII. figs. 22, 23.)

Although many examples of the distal end of this bone have been
found, I have seen but one specimen of the proximal end. It has
been for many years in the Museum of Practical Geology; and I
would express my thanks to Professor Huxley for permission to study
its characters.

The bone is trifid, having lost the proximal ossifications which are
at present usually reckoned as the distal tarsal row. The frag-
ment is } inch long, and measures } inch in width over the articu-
lation, but is only two thirds as deep. At the fracture distally the
diameter of the almond- shaped section is 2 inch,

The middle bone is entirely ankylosed with the other two. All
are of equal length. In front the grooves between the bones are
not very deep, and seem at the fracture to be disappearing. The
bone which is probably external has the outer side flattened and the
proximal end twice as long as wide. ‘The middle bone is flattened
in front; its proximal end is rather shorter. The third is half a
cylinder, compressed a little behind. Its proximal end is put a little
behind the other bones; and it extends a little above them.

Distal end of Metatarsus. (Pl. XXVII. figs. 24, 25.)

The distal ends are usually much broken, and rarely show the
articulations perfect. Occasionally the want of definition on their
articular surfaces may be the result of imperfect ossification. The
specimens show some variation in size, probably a specific character ;
the large one figured is from the collection of T. Jesson, Esq., F.G. S.,
and may be referred to Hnalornis Barretti ; while a smaller aul
less perfect examples at Cambridge seem to have belonged chiefly
to Enaliornis Sedgqwicki. The bone presents a very close resem-
blance to the metatarsus of the Red-throated Diver, but is larger.
The several metatarsal elements in both types occupy the same
positions and terminate in similarly grooved pulley-like ends, rounded
from front to back.

The fragment is 12 inch long, and is deeply channelled in front
between the middle ‘and outer metacarpal elements, as in Colymbus ;
but I have not been able to determine whether the groove in Ena-
liornis is similarly prolonged so as to perforate the metatarsus, or
whether a deep cleft renders the perforation unnecessary. The
external lateral surface is rather more inflated than in Colymbus,
rather more enlarged at the distal articular end, and is thrown a
little further backward. The two outer bones measure from side to
side 3 inch ; they are nearly of the same length; but the outer one
is slightly the longer: at — fracture proximally the diameter of
these elements is less than j inch. The middle articulation mea-

510 HARRY GOVIER SEELEY ON THE

sures 5°; inch from front to back; and the outer one is nearly as
long. The inner articulation is placed about this height above the
middle one, and is more compressed from side to side than the others,
but measures as much from front to back; it terminates posteriorly
in a sharp ridge, which is prolonged up the bone. The inner ele-
ment of the metatarsus is thrown behind the middle bone, so as to
be very imperfectly seen from the front.

So far as I have seen, no digital bones have been collected.

These remains offer evidence-of many parts of the skeleton in at
least two species of birds, distinguished from each other by size and
minor osteological characters: the larger, here named Enaliornis
Barretti, is indicated by skull, vertebral column, pelvis, and all the
larger bones of the hind limb; the smaller species, Hnaliornis Sedg-
wicki, is at present only known by its dorsal vertebra, femur, tibia,
and metatarsus.

Although some of the bones and parts of the vertebral column in
both these species sometimes show evidence of remarkable persistence
of cartilaginous conditions of the articulations, especially in the region
of the dorsal vertebree, I do not find this character to suggest that
the animals should not, in classification, be placed along with existing
Natatorial birds, with which all their other osteological characters
closely associate them. Even if these Greensand fossil birds should
hereafter prove to have had teeth like Jchthyornis and Hesperornis,
I should be inclined to remember the variableness of dental and
other characters in the existing orders of Mammals and Reptiles as
a reason for more than hesitation before contemplating the removal
of these Cretaceous fossils from the Natatorial section of the class
Aves. When the affinities of the fossil type to. Colymbus are so per-
sistent, both in the English and American genera, the biconcave
condition of certain vertebra seems to be of no more value than
the opisthoccelous condition of dorsal vertebrze in the Penguins; and
the toothed condition of jaws I should estimate in classification by
our knowledge of dental variation observed in the jaws of Monotre-
mata and Edentata, and conclude that the character is only generic.

Besides thanking the gentlemen whose assistance has been already
mentioned, I would express my thanks to Professor T. M‘Kenny
Hughes, F.G.S., for the kindness with which he has permitted me
to make use of the resources of the Woodwardian Museum, and
figure the specimens referred to in this and other papers.

EXPLANATION OF THE PLATES.
Puatte XXVI.

Fig. 1. Right side of occipital and parietal regions of the cranium of Hnaliornis
Barretti in the Woodwardian Museum.
2. Anterior view of the same specimen, showing a section of the cerebral
cavity.
3. Posterior view, showing the foramen magnum and occipital bones.

Quart Journ. Geol.Soc.Vol. XXXII: P1.XXVI.

C.L.Gmesbach. Mintern Bros j

BRITISH. GCRETACEOUS: BIRDS

Quart. Journ. Geol. Soc. Vol. XXXIL. PL. XXVIL

=

ee = x
C.L.Griesbach. Naniex: Bios nom.

BiiGtSr SCRE TAGHONS BIRDS

‘ BRITISH FOSSIL CRETACEOUS BIRDS. 5

Vig. 4. Inferior aspect of occipital and sphenoid bones, showing the anterior
divergence of the squamosal bones: a, suture between parietal and
frontal bones; 0, brain-cavity ; ¢, foramen magnum; d, area from
which the basitemporal bones have come away ; ¢, articulation for ©
the quadrate bone ; jf, basisphenoid.

. Basal aspect of a cervical vertebra of Hnaliornis Barretti in the collec-
tion of W. Reed, Esq., F.G.S.

. Anterior aspect of the same specimen: a, anterior articulation; 0, pos-
terior articulation.

. Left side of dorsal vertebra of Enaliornis Barretti in the Woodwardian
Museum.

8. Neural aspect of the same specimen.

9. Visceral aspect.

10. Posterior aspect.

11. Anterior aspect: a, anterior articular surface ; 4, posterior articular

surface; c, tubercle for rib; d, transverse process; é, neural spine ;
f, neural canal; g, anterior zygapophysis.

12. Anterior aspect of dorsal vertebra of Hnaliornis Sedgwicki in the
Woodwardian Museum.

13. Lateral aspect of the same specimen. Lettering as in previous figure.
a, the concave anterior face of the centrum.

14, Visceral aspect of anterior portion of the sacrum of Hnaliornis Barretti
in the Woodwardian Museum, showing the nearly flat anterior ar-
ticulation of the first vertebra.

15. Visceral aspect of middle portion of sacrum of Hnaliornis Barretti in
the Woodwardian Museum.

16. Lateral aspect of the same specimen.

17. Anterior end of postfemoral portion of sacrum of Enaliornis Barretti
in the Woodwardian Museum.

18. Visceral aspect of the same specimen.

19. Lateral aspect of the same specimen.

20, ?Caudal vertebra of Enaliornis Barretti in the Woodwardian Museum.

21. Posterior aspect of same specimen.

22. Anterior aspect of same specimen.

23, Portion of pelvis around the acetabulum of Enaliornis Barretti in the
Museum of Practical Geology: a, ilium; 4, pubis; c, ischium.

24. Distal end of right tibia of Hnaliornis Barretti in the Woodwardian
Museum.

25, Distal view of the articulation of the same specimen.

26. Proximal view of the articulation of a fibula of Znaliornis Barretti in
the Woodwardian Museum.

27. Lateral view of proximal portion of another specimen of the same spe-
cies in the Woodwardian Museum,

or

i=)

Pruate XXVII.

Fig. 1. Anterior aspect of a left femur in the collection of T. Jesson, Esq.,

F.G.S.

2. Inner view of the same specimen.

3. Outer view of the same specimen.

4, Proximal end of a femur, probably of Knaliornis Barretti, in the Wood-
wardian Museum. -

5, Proximal articulation of the same specimen.

6, Proximal end of a small femur, possibly Enaliornis Sedgwicki, in the
Woodwardian Museum.

7. Proximal articulation of same specimen.

8. Anterior view of distal end of left femur of another species of bird in
the Woodwardian Museum.

9. Posterior aspect of distal end of left femur of Enaliornis Sedgwicki in
the Woodwardian Museum.

10. Anterior aspect of same specimen. %

Cr. J, G, S. No. 128. 9 N

512

Fig. 11.
12.

13.

ON BRITISH FOSSIL CRETACEOUS BIRDS.

Distal articulation of same specimen. ;

Distal end of right femur of Hnaliornis Barretti in the Woodwardian
Museum.

External view of proximal end of right tibia of Enaliornis Sedgwichi
in the Woodwardian Museum.

. Posterior aspect of same specimen.
. Fractured distal end of same specimen.
. Distal end of right tibia of Hnaliornis Sedgwicki in the Woodwardian

Museum.

. Fibular aspect of same specimen.
. Distal articulation of same specimen.
. Internal aspect of proximal end of left tibia of Hnalicrnis Barretti.

(Distal end, figs. 24 & 25 in previous Plate.)

. Proximal end of right tibia of Fnaliornis Barretti, in which the arti-

cular surface has remained cartilaginous, and has not been anky-
losed to the patella.

. Proximal cartilaginous surface of same specimen.
. Proximal end of metatarsus of Hnaliornis Barretti, seen from above,

showing imperfect ossification. In the Museum of Practical Geology.

. Anterior aspect of same specimen.
. Anterior aspect of distal end of left metatarsus of Hnaliornis Barretti

in the collection of T. Jesson, Esq. °

. Distal articulation of same specimen, showing the backward position

of the articulation for the inner digit.

A. IRVING ON SOME RECENT SECTIONS NEAR NOTTINGHAM. 513

54, Some Recenr Sxcrrons near Norrincnam. By the Rey. Av
Irvine, B.Sc., B.A.., F.G.S, (Read June 21, 1876.)

[Abridged.]

Tne author refers to papers on this district previously published *.
In his previous paper on the district, reasons had been given for
considering the Permian and the Triassic strata to be much more
closely related to one another than is generally stated in the text-
books, and for questioning the propriety of attempting to draw any
hard and fast line of demarcation between the two formations, at
least so far as they are represented in the north-eastern area. For
further proof of the reasonableness of the position taken by him
then, the author of the present paper refers to the careful and accu-.
rate paper of Mr. E. Wilson, F.G.S., of Nottingham, whose whole.
life has been passed in the district, and whose observations may be
thoroughly relied upon. The description of the sections described
by Mr. Wilson were included in the present paper, but are omitted in
this abstract ; and the reader is simply referred to the paper at p. 533
of the present Quarterly Journal for an account of a section which
must strike every one who examines it as one of the most valuable
pieces of evidence yet obtained of the relation that subsists between
the Permian and the underlying Coal Measures.

There is an interesting point, however, in connexion with this
particular geological horizon, which has not been touched upon by
the author of the paper last referred to, and which illustrates the ex-
treme difficulty which must have presented itself to the minds of .
the Geological-Survey officers in the earlier days of their work, when
the extension of railways and coal-mining had not supplied such an
ample store of data as we possess at present. Reference is here
made to the uppermost strata of the Coal Measures, and particularly
to a portion of them consisting of red sandstones and grits, with
interbedded shales, known locally to mining-engineers as_ the
** Rotherham Rock.” ‘This rock has not been penetrated near Not-
tingham, as it lies too high in the series to reach so far south as
that place, which is probably not far from the southern limits of
the synclinal basin in which the Coal Measures lie; but it was
pierced in the Shire-Oak Pit, near Worksop. It crops out some
miles to the west of that place. Prof. Hull, in the 1873 edition of ©
his work on the Coal-Fields of Great Britain, has included the
strata in question among the Upper Coal Measures, thus confirming
the judgment of local observers; and on referring to the earliest
edition (1852) of the map of the district issued by the Geological
Survey, we find it represented there also as a part of the Coal

* J, Memoirs of the Geological Survey, by Prof. Hull, FRS., and Mr. T.
Aveling, F.G.S.

2. ‘“ Notes on the Geology of the Nottingham District,” by A. Irving, F.G.S.,.
Proceedings of the Geologists’ Association, January, 1874.
2Nn 2

514 ~—s A. IRVING ON SOME RECENT SECTIONS NEAR NOTTINGHAM.

Measures. Yet, strange to say, in a later edition of the same map
(1867), the Rotherham Rock is coloured as Lower Permian. Of
course, if this were rightly so named, there would be no great diffi-
culty in pointing to an instance of close stratigraphical relationship
between the Permians and the Coal Measures. But will it hold?
Meanwhile, the anomaly which has been referred to is puzzling to
students, and tends to destroy their confidence in those whom they
ought to be able to look upon as authoritative guides in these matters ;
and it is to be hoped that another edition of the map may soon ap-
pear, by which the doubt may be removed. There will be much
good work to be done in the district in the next few years, on ac-
count of the rapid development of the coal-field and the local exten-
sion of railways. As the railway on which the section above
referred to is found proceeds eastward it exposes good sections of
the Bunter and Keuper. Between the river Leen and the turnpike
road which leads from Nottingham to Mansfield, the cuttings are
not deep, and, for a great part of the way, the rock exposed consists |
of the usual red and mottled soft sandstones of the Lower Bunter.
East of the road just mentioned the Pebble-beds of the Middle
Bunter are well shown. ‘They partake very much of the nature of
those which compose the scarped rock on which Nottingham Castle
stands, being massively bedded, of varying hardness (though
throughout much harder than the Lower Bunter), whitish, dis-
tinctly jointed, the more compact beds forming a true conglomerate
with sparsely scattered pebbles, and occasionally loose layers of
pebbles between them. Frequent alternations in the force of the
currents in the shallow arm of the sea in which these materials
were deposited are recorded by the above facts, and by the general
prevalence of false-bedding. The last-named character is even more
prevalent in the higher strata of the Bunter which are exposed
to the east. The characteristic pebble-beds of the Middle Bunter
disappear; and in their place one finds a series of thinly laminated,
micaceous, whitish, false-bedded sandstones, destitute of pebbles,
which may perhaps be regarded as representatives of the ‘“‘ Upper
Mottled Sandstones ” of the Cheshire area, described by Prof. Hull*.
The junction of the Bunter with the Keuper is not exposed, as it
hes in one of the numerous minor valleys of erosion so common in
this district. The exposures of the Keuper strata are extensive and
interesting, and include almost the whole range (with the exception
of a few feet at the base) of the ‘‘ Waterstones,” together with some
twenty or thirty feet of red marls above them, which may mark the
passage into the Upper Keuper ; but the greater part of the latter
has, at any rate, been long ago removed by denudation from above.
For more than two miles, in the direction of the village of Gedling
(which lies in a lateral valley opening into the main valley of the
Trent) the Waterstones are exposed, with a gentle easterly dip.
There are no indications of faulting, nor contortions of the strata.
The beds are a series of alternating red shaly marls and sandstones,

* Hull: ‘Permian and Triassic Rocks of the Midland Counties,’ one of the
Memoirs of the Geological Survey. ©

A, IRVING ON SOME RECENT SECTIONS NEAR NOTTINGHAM, 515

the latter varying in thickness from one to four and five feet. One
of these, with a colour approximating nearly to white, is charac-
terized by numerous cavities, without any definite form or arrange-
ment, which are probably to be accounted for by the dissolving away
of calcareous concretions previously existing in the sandstone, by
infiltrations of water charged with carbonic acid. About the middle
of the Waterstone Series, the sandstones and shales are replaced by
a kind of sandstone, in which the fine siliceous and argillaceous
materials are intimately intermingled. At an horizon somewhat
lower the sandstones are more purely siliceous, with numerous inter-
stratified bands and thin marl-partings. The surfaces of these have
received numerous ripple-marks, rain-pittings, and sun-cracks—
many hundreds of truck-loads of such marked slabs having been
brought up from the tunnel which penetrates underneath the
Mapperley ‘‘ Plains” to the north-east of Nottingham, at a depth
below their highest point of 180 feet.

Though no fault is intersected by the railways, there are never-
theless some rather large ones in the Keuper strata of the district,
the result, as I take it, of increased dislocation of the subjacent
paleozoic strata in post-Triassic times. One such fault is exposed in
the face of the cliff which overhangs the railways at Colwich, and
has a throw of not less than 100 fect. On the east side of it are
the upper marls of the Upper Keuper, with many bands of gypsum ;
on the other side of the fault are the sandstones of the Lower
Keuper, which exactly correspond with the lower portion of those
pierced by the Mapperley Tunnel. As this fault runs north-west,
but not precisely in the direction indicated on the Survey-map (the
effect of its throw being reversed by a second fault between it and
the railway above described), it is exposed in several places. At two
points, (1) at Blue-bell Hill, (2) at the top of Red Lane, where
this joins the Mapperley Road, the fault forms what must bave once
been an open gaping fissure: at the first-named place, its width is
11 yards; at the latter, 15 yards. In both instances the fissures
are filled with débris of Keuper strata broken up into a confused
rubbly mass by drift action.

516 . &.J. L. GUPPY ON THE MIOCENE FOSSILS OF HAITI.

65. On the Miocens Fosstts of Hartt. By R. J. Lecumnre Guppy,
Ksq., F.L.S., F.G.S., C.M.Z.8., President of the Scientific Asso-
ciation of Trinidad. (Read May 10, 1876.)

[Puatrs XXVIII. & XXIX.]

A veRY important memoir “On the Topography and Geology of
Santo Domingo” has recently been published by Professor Gabb*.
Whatever tends to elucidate the geology of the island of Haiti must
of course contribute to our knowledge of West-Indian and American
geology generally, and may also throw further ight upon some of
the problems connected therewith which have ‘not as yet been
solved.

- Santo Domingo is a republic occupying two thirds of the Island
of Haiti. A very large portion of the territory comprised within
its boundaries has been geologically surveyed and mapped out by
Prof. Gabb and his assistants. The formations examined by them
are classified as Postpliocene, Miocene, and Cretaceous. Hruptive
rocks are also developed to a large extent.

It is not my intention to speak of the geological part of Prof.
Gabb’s labours. That could only be done usefully by those ac-
quainted with the country and conversant with its structure. My
present business is with the paleontological portion of the work.

The Geological Society possesses the first regular collection of
fossil Tertiary shells and corals made in Haiti, and, indeed, I might
almost say, in the West Indies. This collection contains the types
of the species described by Sowerby for Mr. Carrick Moore, and by
Prof. Duncan. I have from time to time, when treating of the West-
Indian Tertiary rocks and fossils, alluded to the fact that several of
the species contained in that collection were unnamed.

Most of those species have now been described by Prof. Gabb ;
but it is to be regretted that his work is not accompanied by figures,
so that our determinations may sometimes be open to doubts which
the aid of pictorial illustration would enable us to dispel.

{n'the Proceedings of the Scientific Association of Trinidad for
December 1873, I described several new fossils from Jamaica ; and
deeming it desirable that those fossils should be more widely known
than they could be by means of that publication, whose circulation
is very limited, I republished the descriptions and figures in the
‘Geological Magazine’ for September and October 1874, together
with a revised list of the organic remains (exclusive of the corals
and other fossils) which had been found in the Tertiary deposits
of the Caribbean region. This list was a revision of one previously
published by me in the Proceedings of the Scientific Association of
. Trinidad for 1867. The first knowledge I had of Prof. Gabb’s work
* Trans. Amer. Phil. Soc. vol. xv. p. 49, with a geological map of Santo

Domingo. See also, for the new genera described, Proc. Acad. Nat. Sci. Phil.
1872, p. 270.

R. J. L. GUPPY ON THE MIOCENE FOSSILS OF HAITI. 517

was from the January (1875) number of the ‘ Journal de Conchylio-
logie,’ which, however. I did not receive until some time after the
date of its publication. In August last Prof. Gabb did me the
kindness to send me a copy of his publications relating to the
geology and paleontology of Haiti. I deemed the opportunity a
favourable one for reexamining the Haitian fossil mollusca, which
had been untouched since 1853. The present communication,
which embodies the results of my reexamination, is confined for
the most part to such species as I have identified in the collection
of the Geological Society, or have noticed in other collections of
West-Indian Tertiary fossils.

We have not escaped the tendency, almost inevitable in the case
of little-known forms, to describe mere varieties, some more or less
permanent, some merely individual, under distinct specific names.
Prof. Gabb has done good service by uniting together some of these
artificially separated forms; and in this labour he has had the
advantage of large suites of specimens collected by himself in Haiti.
While contributing my quota towards a reduction of our superfluous
nomenclature, I may observe that it is not always a disadvantage in
the beginning (particularly when the descriptions are accompanied
by good figures) to distinguish the various forms which are mis-
taken for distinct species. Owing apparently to the great exube-
rance of Molluscan life in the West-Indian Miocene, much variation
- occurred, especially in certain genera.

In all the corrections I have ventured on in respect of Prof.
Gabb’s determinations, I have not acted in any spirit of derogation of
his work, which I consider very valuable; and I could congratulate
myself if my own had been at all times as well done. ;

One of tke results stated by Mr. Carrick Moore has been brought
out more strongly by Prof. Gabb’s examination of the Haitian
fossils. It is the alliance of the West-Indian Miocene fauna to that
of the west coast of South America; and I think now that the
‘conviction can hardly be resisted that during some portion of the
Miocene period there was a free communication between the Pacific
and the Atlantic. But other alliances point to the west coast of
Africa; and there remain other alliances still more close with the
Eastern and Indian Seas. ;

In a letter to me, dated 18th September, 1871, the late Prof.
William Stimpson informed me that in the deep-sea explorations
off Florida he had discovered shells either identical with or yery
closely allied to some species of the West-Indian Miocene. Among
them he cited Conus planiliratus, Pleurotoma Barretti (= Pl. hattensis),
Phos elegans, and Corbula viminea.

For the sake of conciseness I have, in the following remarks, used
the abbreviation G. J. for the Quarterly Journal of the Geological
Society.

1. Denratium DissmILe, Guppy, G. J. vol. xxii. p. 292, pl. xvii.
fig. 4. . an
T am inclined to believe that Gabb’s D. ponderosum is a form of

518 RB. J. Le GUPPY ON THE MIOCENE FOSSILS OF HAITI.

this species having thick walls. Three other species are named by
Prof. Gabb ; but I cannot refer the specimens in the Geological So-
ciety’s collection to more than two species, one of which is named
below.

2. DENTALIUM MISSISSIPENSE, Conrad.

This is very like, if not identical with, D. Aickaii of the Belgian
Miocene.
3, BuLLa PAUPERCULA, Sow. G. J. vol. vi. p. 32.

This, and B. Vendryesiana, may be regarded as probably marked
forms of B. striata, Brug.

4, Buia eRanosa, Sow. G. J. vol. vi. p. 51, pl. x. fig. 10.

5. VotvuLa cytinprica, Gabb, Trans. Amer. Phil. Soc. vol. xv.
p. 246.

6, ToRNATINA COIX-LACRYMA, Guppy, Geol. Mag. 1867, p. 500, fig. 3.
T. recta, Gabb, J. c. p. 246 (as of D’Orb.).

The same species occurs in the Miocene of Trinidad and Jamaica.
7’. canaliculata, T. olivula and T. recta, D’Orb., which may all be
synonymous, are nearly related to our fossil.

7. Cyricuna Brpentara, D’Orb. Moll. Cuba, pl. iv. figs. 13-16.

This is erected into a new genus, Oylichnella, by Gabb; but I do
not concur in the necessity of that proceeding. An allied but larger
species (C. ovum-lacerti, Guppy) is found in the Eocene of Trinidad.
8. ToRNATELLA (AcT#oN) cUBENSIS, Gabb.

This may be identical with the shell I have described as 7. textilis
(Geol. Mag. 1874, p. 407, pl. xvi. fig. 4); but if so, it-is distinct from
1’. punctata, D’Orb., as respects the surface ornamentation, which in
7’. tectilis has more of a cancellate than a punctate character.

9. Riyercuna tRIpenrTATA, Guppy, Geol. Mag. 1874, p. 406.

The specimens identified by Gabb as R. semistriata, D’Orb., may
belong to this species.

10. Torzonirra tuRRIs, D’Orb. Moll. Cuba, vol. i. p. 219, pl. xvi,
figs. 10-13.

A very variable species, found also in the Miocene of Jamaica and
Pliocene of Trinidad. Other forms of the same have been described
by D’Orbigny as modesta, ornata, and pulchella; and I believe that
T. domimeensis, angusta, and pertenuis of Gabb may be placed in ©

the same category.
11. Narica canrena, Linn.

12, Natica sutcats, Born,

R. J. L. GUPPY ON THE MIOCENE FOSSILS OF HAITI. 519

13. NatrcA MAMMILLARIS, Lam.
Gabb considers NV. subclausa, Sow., to belong to this species.

14. Narica PHASIANELLOIDES, D’Orb.

Gabb redescribes this species under the name of Amaura Guppy,
but does not give his reason for disagreeing with my identification.

15. SIGARETUS EXCENTRIcUs, n. sp. (Pl. XXIX. fig. 11.)

Shell globose-depressed, adorned with many regular rounded
spiral ribs with much narrower threadlike interstices: spire de-
pressed ; whorls convex, the last much the largest; aperture oval,
somewhat angulated posteriorly.

A more globose shell with much more convex whorls than S. de-
pressus, Phil. (=S. haliotideus, D’Orb., =NS. Leachii, Sow., but not the
S. halotideus of Linné. See Hanley, Ips. Linn. Conch. p. 390).

16, Turriretta tornata, Guppy, G. J. vol. xxii. p. 580, pl. xxvi.
fig. 12,

17. TuRRITELLA PLANIGYRATA, Guppy, Proc. Scient. Assoc. Trinidad,
1867, p. 169; Geol. Mag. 1874, p. 408, pl. xviii. fig. 5.
Gabb has recognized this species among the Haitian fossils; and
his remarks upon it are appropriate. The figure in Geol. Mag. shows
only the general shape.

18. VERMETUS PAPULOSUS, Guppy, G. J. vol. xxi. p. 292, pl. xvii.
fig. 3.
Gabb regards this as the end of the tubes of Petaloconchus; and
in this he may be right, though I do not feel completely satisfied on
the point.

19. PrraLoconcHus scuLptuRAtus, Lea; P. pomincensts, Sow. G. J.
VOL vip. Ol, pl. x. ne. 'g.

20. CERITHIUM UNISERIALE, Sow. G. J. vol. vi. p. 51. (Pl. XXIX.
fig. 4.)

This is a remarkable species ; and Sowerby’s description applies
strictly to only one specimen. The species may be distinguished
generally by its peculiar ovate shape, also by the ascension of the
last whorl, near the aperture, halfway up to the suture, forming a
canal upon the body-whorl, between the angle of the aperture and
the callus of the inner lip, which terminates in a broad and blunt
tooth.

21. Crerrruium opEsum, Gabb, Trans. Amer. Phil. Soc. vol. xy. p. 237.
(Pl. XXIX. fig. 9.)

22. CeRITHIUM PLEBEIUM, Sow. G. J. vol. vi. p. 51; Guppy, G. J.
vol. xxii. p. 290, pl. xvi. fig. 9.

Near to C. (Vertagus) gemmata, Hinds, Voy. Sulph. pl. x. fig. 56.

O20 R. J. L. GUPPY ON THE MIOCENE FOSSILS OF HAITI.

23. SOLARIUM QUADRISERIATUM, Sow. G. J. vol. vi. p. 81, pl. x. fig. 8.

The Torinia rotundata of Gabb can scarcely be other than a form
of this species, differing, it would seem, in the greater elevation of
the spire and convexity of the whorls.

24, CANCELLARIA Barrerri, Guppy, G. J. vol. xxii. p. 289, pl. xvii.
ne. 4.

Gabb considers this identical with the recent C. reticulata. I am
not sure whether it is distinct from the species next named, which
Gabb has allowed to stand; butit is a pronounced form whose pro-
portions are more elongate than those of C. levescens.

25, CANCELLARIA LEVESCENS, Guppy, G. J. vol. xxii. p. 289, pl. xvii.
fig. 12,

26. CancettariaA Moorer, Guppy, G. J. vol. xxii. p. 289, pl. xvi.
fie. 7.

Gabb includes in his list the names of (. brevis and C. tessellata,
Sew. I am inclined to think that the forms so named by him are
young specimens of C. Mooret and C. levescens respectively.

The following species appears to me quite distinct. It does not
appear to have been mentioned by Gabb.

27. CANCELLARIA EPISTOMIFERA, n. sp. (Pl. XXVIII. fig. 9.)

Shell pyriform, scarcely rimate, adorned with stout spiral ridges
having longitudinal ones crossing them and rising into points on
the intersections.

Spire acuminate, having seven or eight whorls (of which the nu-
clear ones are smooth). Aperture oblong, canal short, outer lip
sinuous, carrying a spout-like protuberance.

J udging from the sculpture alone, this might be aiken fora form
of C. Barretti or C. levescens. It is, however, rather more coarsely
sculptured than either of those species, in the latter of which the
last whorl is smooth, or nearly so. The sinus of the outer lip is
carried in a sort of spout, instead of forming a depression as in the
outer lip of C. levescens.

Cancellaria scalatella, Guppy (Miocene, Jamaica), does not seem
to have been found in Haiti (see Geol. Mag. 1874, p. 408, pl. xvii.
fig. 4).

28. ORTHAULAX INoRNATA, Gabb. (Pl. XXVIII. fig. 8.)

O. inornatus, Gabb, Proc. Acad. N.S. Phil. 1872, p. 272, pl. ix.
f. 3, 4; Trans. Amer. Phil. Soc. vol. xv. p. 234.

To whatever genus this shell may be assigned, I have little doubt
of its near alliance to Rostellaria macroptera, Lam. (Strombus am-
plus, Brander), of the European Hocene (Sow. Min. Conch. vol. v.
p. 177, tab. cexcviil., ecxeix., and cce.). The description of that shell
by Sowerby might, with a few alterations, be taken for that of our
species, even to the remarks as to the state of preservation of the fossil.
The differences appear principally to be that in O, inornatus the

R. J, L. GUPPY ON THE MIOCENE FOSSILS OF HAITI. 521

spire is usually more concealed, while the wing is not so much ex-
tended ; the young shells of this species are also much less elongate
in their proportions. There may be other differences; but the spe-
cimens in the Geological Society’s collection do not enable me to
speak with certainty of them, as no example is perfect. One (in
fragments) attained the dimensions of at least six inches long and
four wide.

29. Stromsvs HAITENSIS, Sow. G. J. vol. vi. p. 48, pl. ix. fig. 7.
Gabb regards this as being S.. bituberculatus, Sow.

30. SrromBus proximus, Sow. G. J. vol. vi. p. 48, pl. ix. fig. 8.

Strombus pugilis, C. Moore.
pugiloides, Guppy.
ambiguus, Sow.

In his synonymy of S. pugilis, Gabb includes the following :—am-
higuus, Sow. ; bifrons, Sow.; and pugilis (=pugilordes, Guppy).

Whilst admitting much truth in Gabb’s rectification, I cannot go
so far as to regard S. bifrons as a synonym of either S. pugilis or
S. provimus.

In Geol. Mag. 1874, p. 483, I have given what I consider good
reasons for separating the fossils formerly regarded by Mr. Carrick
Moore and myself as S. pugilis under a distinct name, and I pro-
posed that of pugiloides. I am, however, after an examination of
the types in the Geological Society’s collection, prepared to agree
that S. prowimus and S. ambiguus are not separable from the Strom-
bus pugilis of Carrick Moore and myself. One of Sowerby’s names
(being prior) must therefore take the place of my name pugiloides ;
and the name prowimus may be allowed, to indicate the position of
the species so near to the recent S. pugilis.

31. SrromBus Birrons, Sow. G. J. vol. vi. p. 48, pl. ix. fig. 9.

I prefer to retain this name for the specimens which resemble
Strombus dilatatus and columba more than pugilis.

32. Murex pomineensts, Sow. G. J. vol. vi. p. 49, pl. x. fig. 5.

Gabb notes the relationship of the species to M. haustellum. To
my eye its nearest kindred is M. messorius, Sow., a West-Indian
recent species by no means very close to MW. haustellum.

I cannot decide if M. antillarum, Gabb, is distinct from MM,
domingensis, Sow.

33. MuREXx cornvRrectus, n. sp. (Pl. XXVIII. fig. 4.)

Ovate-turreted, with three varices, which are nearly continuous,
and stout revolving ridges accompanied by finer lines; two or
occasionally three variciform tubercles between each varix ; varices
fringed by subtubular spines, of which the one corresponding to the

*keel on the angle of the whorls is much the longest. Aperture oval,
the inner margin callous, the outer margin dentate, the dentations
_ running In pairs. Canal mcderately long and slightly curved.

O22 RB. J. L. GUPPY ON THE MIOCENE FOSSILS OF HAITI.

Nearly related to MZ. cornucervi, Mart. (= brevifrons = M. cal-
citrapa of some authors).

34. Murex Textitis, Gabb. (Pl. XXIX. fig. 1.)

Murex textilis, Gabb, Trans. Amer. Phil. Soc. vol. xv. p. 202.

This species is a member of the group Pteronotus, and is allied to
the following species, and more closely to the first named than either
of the others :—

M. festivus, Hinds, Voy. Sulph. pl. ii. figs. 13, 14.

M. pinniger, Brod.; Wood, I. TI’. Suppl. Murex 4.

M. tripterus, Born; Reeve, C. I. Murea 55.

M. macropteron, Desh.; Reeve, C. I. Murev 123.

M. gambiensis, Reeve.

M. cyclopterus, Millet, J. Conch. 1875, p. 147.

Under M. teatilis, Gabb, I include M. compactus, Gabb.

35. TROPHON DOMINICENSIS, Gabb, l.c. p. 202.

The description of this species seems not to be inapplicable to
Murex collatus, Guppy (Geol. Mag. 1874, p. 433, pl. xvi. fig. 8), and
may refer to the same species. There is a specimen of Z’rophon in
the Society’s collection upon which I cannot venture to decide, as it
seems different.

36. TypHis ALATUS, Sow. G. J. vol. vi. p. 48, pl. x. fig. 4.

I have recorded this species from the Miocene of Jamaica and the
Pliocene of Trinidad.

The form described by Gabb as 7’. obesus is represented in the
Society’s collection; and I have it also from Jamaica; it is not
specifically distinct from 7’. alatus.

37. RANeLxa crassa, Dillw. G. J. vol. xxii. pl. xviii. fig. 9.

38. Triton DominGENSIS, Gabb, pl. B. fig. 2.
Triton domingensis, Trans. Amer. Phil. Soc. vol. xv. p. 212.

39. Triron vARIEGATUS, Lam.
40. Triron FEMORALIS, Linn.

41, Triton cemmatvs, Reeve.

42, Prrsona stmitima, Sow. G. J. vol. vi. p. 48.

Persona simillima, Guppy, G. J. vol. xxii. pl. xvii: fig. 13.

Gabb remarks on the alliance of this species with P. constricta of
the west coast of America. It is, however, as near to the recent
Persona reticularis, Linn., found somewhat rarely in the West Indies,
and to P. clathrata, a closely related form from the east coast of
Africa (Madagascar) and Ceylon.

R. J. L. GUPPY ON THE MIOCENE FOSSILS OF HAITI. 523

43. TuRBINELLUS (LAtTrRUS) INFUNDIBULUM, Gmel.

Gabb describes as new species four forms. I entertain doubt
whether they are distinct from the above; but want of material
prevents me from speaking with certainty.

44, TURBINELLUS ZDIFICATUS, n. sp. (Pl. XXVIII. fig. 5.)

Shell solid, rimate, very shortly fusiform, spire high, composed of
seven or eight whorls adorned with strong longitudinal ribs each
terminating on the angle in a subtubular spine, and with numerous
close spiral ridges, which are crossed by fine squamose lines of growth.
Aperture narrow; inner lip covered with a thick callus bearing
about four plaits.

Nearest to 7’. capitellum, Lam. (Kien. Cog. Viv. Lurbinella, pl.
12). Indian Seas.

This species seems very different from any noticed hitherto. I
have endeavoured to identify it with one of Gabb’s descriptions, but
without success, and therefore assign it provisionally a new name.
I am not, however, without a suspicion that the specimens called
Vasum haitense by Gabb belong to this species, while his V. tubercu-
latum is really the 7’. haitensis of Sowerby.

45. TurBINELLvs VALIDUs, Sow. G. J. vol. vi. p- 50.
Allied to 7’. scolymus.

46, TuRBINELLUS ovorpEvS, Kiener.

47, TURBINELLUS HAITENSIS, Sow. (Pl. XXIX. fig. 3.)

Turbinellus haitensis, Sow. G. J. vol. vi. p. 50.
I think Vasum tuberculatwum, Gabb, may be referable to this
species.

48, FascroLaRIA SEMIsTRIATA, Sow. G. J. vol. vi. p. 49.

Fasciolaria semistriata, Guppy, G. J. vol. xxii. pl. xvi. fig. 12.

I can detect no difference between LF. intermedia and F. semi-
striata. The papillary apex noticed by Sowerby as a character of
the former is, as Gabb remarks, simply the single nuclear whorl.
It may have originally existed in the specimens assigned to /, semi-
striata. This species (including L’. intermedia) is allied to F. tulipa.
F. textilis, Guppy (Geol. Mag. 1874, p. 410, pl. xvi. fig. 5) is nearer
to F. filamentosa, but is more ovoid in shape, and is, indeed, almost
intermediate between the groups designated as Turbinellus and
Fasciolaria, which, together with Pyrula ( = Cassidulus of some
authors, but excluding Ficwla), may in a large sense be considered
as subgenera of Musus. Another species of Fusciolaria found in the
West-Indian Miocene, but not yet recorded from Haiti, is F. Tar-
belliana, Grat.

49, PyrvuLa MELoNGENA, Linn.

P. consors, Sow. G. J. vol. vi. p. 49.
Gabb considers that P. patula is conspecific. I have doubts

524 R. J. L. GUPPY ON THE MIOCENE FOSSILS OF HAITI,

whether they ought to be so regarded, though unquestionably nearly
allied. A reexamination of the Haitian fossil revealed to me some
difference in the surface characters of the Miocene and recent shells.

50. Fusus HEnrkent, Sow. (Pl. XXVIII. figs. 2 & 6.)

Fusus henekeni, Sow. G. J. vol. vi. p. 49.

An examination of Sowerby’s Ff’. haitensis proves that it is only
an individual form of /. henekent. Similar variations occur in the
recent analogue of the species, /. distans, Lam.

51. Cuma tectum, Kiener.
This shell is well represented in the Geological Society’s collection.

52. Paos Gurpyi, Gabb, l.c. p. 212.
Phos erectus, Guppy, Geol. Mag. 1874, p. 410, pl. xvi. fig. 1.

53. Pos evecans, Guppy, G. J. vol. xxii. p. 290, pl. xvi. fig. 13.

Gabb makes this a synonym of P. veraguensis, Hinds, which I am
not prepared at present to indorse; but I concur with him in re-
garding P. Moore as a form of P. elegans. I should go a step
further and unite, under the name of P. elegans, the forms described.
by Gabb as P. costatus and P. semicostatus. The Nassa solidula of
the West-Indian Miocene appears to be distinct; but it is not re-
corded from Haiti.

54. Nassa rncrassaTa, Mill.

55. Crea rrunoata, Gabb. (Pl. XXIX. fig. 6.)

Ectracheliza truncata, Gabb, Proc. Acad. N.S. Phil. 1872, p. 271,
pl. ix. fig. 2; Trans. Amer. Phil. Soc. vol. xv. p. 213.

The new genus Lctracheliza, described by Gabb, appears to me
identical with Clea, Adams, of which the type Olea nigricans, Adams,
was figured in Gen. Moll. vol. 11. p. 625, pl. exxxvii. fig. 8. The
two species agree in general form, in the truncation of the apex, and
even in the presence of an impressed more or less double spiral line
below the suture. The fossil C. truncata appears to differ from the
recent C. nigricans chiefly in the following respects :—The latter has
a more vertical aperture, the outer lip not being so prominent an-
teriorly ; and its columella is rather more strongly twisted.

The genus is with little doubt closely related to Planawxis, forming
one of the links of connexion between Oliva and the other members
of the Buccinidee. Another species which may be compared with
Clea truncata is the Quoyia decollata of Gray (Reeve, El. Conch.
vol. i. p. 63, pl. ii. fig. 18), which, besides other differences, has a
considerably higher spire.

56. CREPITACELLA CEPULA, Guppy.
Melanopsis cepula, Guppy, G. J. vol. xxii. p. 580, pl. xxvi.
. 14, ,

fig

R. J. L. GUPPY ON THE MIOCENE FOSSILS OF HAITI. 525

Crepitacella cepula, Guppy, Geol. Mag. 1867, p. 500.

Dolophanes melanoides, Gabb, Proc. Acad. N. 8. Phil. 1872,
p. 273, pl. xi. fig. 7; Trans. Amer. Phil. Soc, vol. xv. p. 234.

After a careful examination no doubt remains on my mind of the
correctness of the above synonymy, which is suggested by Prof. Gabb
himself. The points of difference which he notices are not constant,
as I have ascertained by the inspection of a series of specimens from
Jamaica,

57. TEREBRA SULCIFERA, Sow. G. J. vol. vi. p. 47. (Pl. XXIX.
fig. 8.)

T’. bipartiia, Sow.

7’. inequalis, Sow.

Gabb considers 7’, sulcifera to be identical with 7’. robusta, Hinds.
After an examination of the numerous specimens in the Society’s
collection, I can establish no constant differences between the indi-
viduals described under the three above-quoted names by Sowerby.

7’. flammea is included by error in our list of West-Indian Miocene
fossils.

58, Cassis sutcrrpra, Sow. G. J. vol. vi. p. 47, pl. x. fig. 1.

59. Cassis mMoniLirerRA, Guppy, G. J. vol. xxii. p. 287, pl. xvii.
fig. 8.

C. reclusa, Guppy, Geol. Mag. 1874, p. 434.

Gabb identifies this species with C. granulosa; and I find a very
close resemblance between them. But C. monilifera is distinguished
by its varices, which, however, are wanting in the form I have
called C. reclusa; and the latter is moreover devoid of the row of
tubercles on the angle. The species is as near to C. subulosa (Mio-
cene of Europe) as to C. granulosa.

60. CassrpariA tmvieata, Sow. G. J. vol. vi. p. 47, pl. x. fig. 2.
Cassidaria sublevigata, Guppy, G. J. vol. xxii. p. 287, pl. xvii.
fig. 10.

” While agreeing with Gabb that C. sublevigata, Guppy, is only a
variety, I may mention that it is a distinguishable form. Examples
of it were separated in the Geological Society’s collection by Mr.
Carrick Moore as C. levigata, var.

61. Oniscta pomtncEnsis, Sow. G. J. vol. vi. p. 47, pl. x. fig. 3.

62. Matra camura, Guppy, G. J. vol. xxii. p. 287, pl. xvii. fig. 9.
Gabb regards this as identical with MW. ringens, Swains. (latlabris,

Val.).
63. FicuLa MissIssIPrensis, Conr. Journ. Phil. Acad. 2nd ser. vol. i.
p. 117.

F. carbasea, Guppy, G. J. vol, xxii. p: 580, pl. xxvi. fig. 7.
Found also in the Miocene of Jamaica and Trinidad. ”

526 R. J. L. GUPPY ON THE MIOCENE FOSSILS OF HAITI.

64, ConuMBELLA veNuSTA, Sow. G. J. vol. vi. p. 46, pl. ix. f. 6.

For this species Gabb creates a new genus Metulella, in which he
includes another shell found by him in the Haitian Miocene and de-
scribed as Metulella fusiformis, Gabb, a figure of which is given by
him in Journ. Acad. N. 8. Phil. 1872, pl. xi. f. 3. |

65. CoLUMBELLA GRADATA, Guppy, G. J. vol. xxii. p. 288, pl. xvi.
£.. 20;

IT am inclined, upon a reexamination of the fossils, to consider
C. ambigua, Guppy, as only a marked variety of C. gradata; and I
should include under the same name the Strombina caribwa of Gabb.
The Cumana specimens of C. gradata exhibit all the characters as-
cribed by Gabb to his S. caribea, including the Ranella-like flatten-
ing, which is less pronounced in the Jamaican examples, and not found
at all in the form called C. ambigua. I further doubt if C. inflata,
Gabb, is separable from C. gradata.

66. CotuMBELLA HAITENSIS, Sow. G. J. vol. vi. p. 46.

From the description I should suppose that C. eailis, Gabb, may
be comprised under this name.

67. Oxrva cyLinprica, Sow. G. J. vol. vi. p. 45.

I cannot find any sufficient characters upon which to base a sepa-
ration into distinct species of the Olives found in the West-Indian
Miocene. Some of them approach more or less closely to O. reticu-
laris; but I am not sure of their specific identity. One of my spe-
cimens of O. cylindrica from Jamaica is more than 34 inches long,
and has lost about 3 inch of its spire. Gabb names five other species ;
but I can give no certain opinion as to their validity.

Gabb describes under the name of Plochclea crassilabrum (Proc.
Acad. N. 8. Phil. 1872, p. 971, pl. xi. f. 5) a form of Olive which I
have not met with.

68. ANCILLARIA GLANDIFORMIS, Lam.

I am not certain of the distinctness of my A. pinguwis (Jamaica)
from this species.

69. Prevroroma wArrEnsis, Sow. G. J. vol. vi. p. 50.

P. Barreiti, Guppy, G. J. vol. xxu. p. 290, pl. xvi. f. 6.

Gabb considers P. haitensis to be the same as P. virgo, and includes
in the synonymy (besides P. Barretti) P. Jelsku, Crosse, and P. an-
tillarum, Crosse.

70, PLEUROTOMA HENEKENI, Sow. G. J. vol. vi. p. 50, pl. x. f. 6.
P. jaquense, Sow. 1. ¢. p. 51.
I concur with Gabb in the fusion of the above two specific names ;

and I would add his P. longicaudata and P. hwmerosa to the syno-
nymy.

R. J. L. GUPPY ON THE MIOCENE FOSSILS OF HAITI. 527

71. Prevroroma venvsta, Sow. G. J. vol. vi. p. 50, pl. x. f. 7.

_ My P. jamaicensis may possibly be referred to this species as a
small and marked variety. The living analogue of this species ap-
pears to be P. gibbosa, Chemn.

72. Prevroroms consors, Sow. G. J. vol. vi. p. 50. (Pl. XXYIII.
fig. 7.)
I cannot undertake to say that this is identical with P. militaris,
Hinds, though Gabb considers it so. The likeness was remarked by
Sowerby. :

73. Prevroroma sqeuamosa, Gabb, Trans. Amer. Phil. Soc. vol. xv.
p. 208, (Pl. XXX, fig. 7.)

There is one example of this species in the Geological Society’s
collection. The sculpture is very remarkable. Several other forms
of Pleurotoma are named by Gabb; but I have seen none which may
not be referred to one or other of the above five species, which are
well marked and decidedly distinct.

74. Cravarura Laprata, Gabb, Trans. Amer. Phil. Soc. vol. xy. p. 209.
(Pl. XXVIII. fig. 3.)

Near to C. imperialis, Lam. West Africa.

75. Conus rEcoanirus, Guppy.

C. solidus, Sow. G. J. vol. vi. p. 45; Guppy, G. J. vol. xxii. pl.
xvi. fig. 1.

Gabb identifies this with C. pyriformis, Reeve. The C. solidus
of Sowerby (Thes. Conch. 580,=C. retifer, Menke) is a different
species, very unlike the Haitian shell.

76. Conus consosrinvs, Sow. G. J. vol. vi. p. 45; Guppy, Geol. Mag.
1874, pl. xvii. f. 3.

C. granozonatus, Guppy, G. J. vol. xxii. p. 287, pl. xvi. f. 5.

Prof. Gabb has so conscientiously worked out the Haitian Cones
that I accept this rectification, which otherwise would have appeared
to me difficult. The figure given by me in the Geological Magazine
is fairly representative of the usual form.

77. Conus catenatus, Sow. G. J. vol. vi. p. 45, pl. ix. f. 2.

C. interstinctus, Guppy, G. J. vol. xxii. p. 288, pl. xvi. f. 3.

C. stenostoma, Sow. G. J. vol. vi. p. 44; Guppy, G. J. vol. xxii.
p. 29 7,. pl. xvis £2.

Gabb records C. stenostomus as distinct from C. catenatus; but I
find great difficulty, on comparison of many specimens, in drawing
the line of demarcation.

78. Conus eRactiissrwus, Guppy, Geol. Journ. vol. xxii. p. 288,
pl. xvi. f. 4.

Gabb refers this to C. Orbignyi, Audouin (C. planicostatus, Sow.).
Q.J.G.8. No. 128. 20

528 R. J. L. GUPPY ON THE MIOCENE FOSSILS OF HAITI.

Tt bears much likeness to a shell recently described by Sowerby as
C. gracilis (Zool. Proc. 1875, p. 125, pl. xxiv. f. 6). 7

79. Conus maRernartus, Sow. G. J. vol. vi. p. 44. (PI. XXIX. fig. 5.)

I believe the specimens from Cumana, which I formerly attributed
to O. haitensis, really belong to C. marginatus—a mistake due to
want of fioures.

80. Convs pLanitrRratus, Sow. G. J. vol. vi. p. 44.

C. planiliratus, Guppy, G. J. vol. xxii. p. 287, pl. xvi. f. 7.
Gabb entices this with OC, Stearnsi, Conrad, Florida.

81. Conus marrensis, Sow. G. J. vol. vi. p. 44.

C. symmetricus, Sow. 1. c. p.. 44, pl. ix, f. 1.

C. domingensis, Sow. l. ¢. p. 45.

I adopt the above synonymy from Prof. Gab. The shell de-
scribed by me as C. prototypus may possibly be the same as C%
strombiformis, Gabb.

Of the other species of Conus enumerated by Prof. Gabb I have
no knowledge. All the West-Indian Miocene forms I have seen
may be assiened to one or other of the species named above—though,
owing to the great range of variation, some difficulty is sure to be
felt until the student has obtained a closer acquaintance with these
fossils.

82. Mirra Henexent, Sow, G. J. vol, vi, p. 46, pl. ix, f, 5,

T have recognized no other Mitre than this in the West-Indian
Miocene ; and J should be inclined to place under it most, if not all,
of the forms described by Gabb under different specific names—the
only exception being M. tortuosa, which Prof. Gabb states to belong
to the group Costellaria, Swains., and to be akin to WM. semifasciata,
Lam. The M. varicosa of Sowerby has not been identified either by
Prof. Gabb or myself. It is possible that MM. titan, Gabb, may be
a valid species.

83. VoLUTA PULCEELLA, Sow. G. J. vol. vi. p. 46, pl. 9. f. 4.

Gabb is right in considering V. soror, Sow., to belong to this spe-
cies. It has no characters by which even to separate it as a variety.
I have not met with Scapha striata, Gabb.

84, Marernerza contrormis, Sow. G. J. vol, vi. p.45; Guppy, G, J.
vol, xxii. p. 288, pl. xvii. f. 2,

85, Mareretta SowErsyi, Gabb, Trans. Amer: Phil. Soc. vol. xv.
p- 221. (Pl. XXVIII. fig. 1.)

86. CyprmA Henexent, Sow. G. J. vol. vi. p. 45, pl, ix, fig, 3,

87. Crprma Gapprana, n. sp. (Plate XXIX, fig. 10.)

The cowry for which I propose the above name has hitherto been
considered by me to be OC. pustulata, and has been identified by

R, J. L, GUPPY ON THE MIOCENE FOSSILS OF HAITI. 529

Gabb as C. nucleus, I think it may be regarded as intermediate
between those two species; and it presents, I think, some characters
which, combined with its distance in time and space from its nearest
congeners, may warrant a provisional specific name :—

Oval-elongate, rostrated at both ends, superiorly covered with
large shining tubercles, which are almost circular upon. the back,
but become elongate and have a tendency to run into ribs near the
thickened and regularly grooved lip, whose dentations are continuous
with the ribs on the outside. A dorsal groove separates the back
into two nearly equal halves.

The tubercles are larger than those of C. nucleus.

88. Cattiostoma contcum, Gabb, l. ¢. p. 243."

I possess a shell from the Miocene of Jamaica which may belong
to this species. It differs from Gabb’s description in having beaded
ribs, of which the one on the angle of the whorls is largest, The
shells described by me from the Pliocene of Trinidad under the
names T'rochus decipiens and plicomphalus are nearly allied.

A critical examination of the species of this section of Trochus
found in the West-Indian area is much needed. They are very rare ;
but a few are occasionally found. The present shell resembles some
of the European Miocene forms, e.g. J. laureatus, Mayer, and
T’. Pauluccie, Mayer.

I do not recognize the Margarita tricarinata nor the Adeorbis
carinata among the Haitian fossils. It has occurred to me that one
or both of them belong to the species described by me as Cyclo-
strema bicarinatum ; but I can speak with no certainty on this point.

89. Pyorts DELECIUS, n. sp. (Plate XXVIII. fig. 10.)

Shell conical, umbilicate, whorls sharply angulated below, form-
ing a very concave base, and bearing on the periphery a keel carry-
ing obtuse tubercles. Upper surface covered with close undulated
ridges, occasionally dichotomous or anastomosing, and running in a
more or less spiral direction, but directed rather towards the outer
margin. Base concave, covered with spiral rows of small grains.

I do not feel sure that this is the shell indicated by Prof. Gabb
under the name of Phorus agglutinans, Lam. It is certainly very
different from the Eocene P. agglutinans, and it is apparently suffi-
ciently distinct from the living Phorus of the West Indies (whose
name is properly P. conchyliophorus) to deserve a specific appella-
tion.

90. Trrepo FistuLA, Lea, ?=Kupnus 1ncRAssatus, Gabb.

91. CorpuLa viurnza, Guppy, G. J. vol. xxii. p. 293, pl. xvii.
fie. 11.
Bothrocorbula viminea, Gabb, Proc. Acad. N.8. Phil. 1872, p. 274,
pl. x Be. 3:
The new genus Bothrocorbula of Gabb is founded upon this species,
and may possibly pass as a subgenus or section.
202

530 R. J. L. GUPPY ON THE MIOCENE FOSSILS OF HAITI.

92, Corputa viera, Guppy, G. J. vol. xxii. p. 580, pl. xxvi. fig. 8.

According to Gabb this is C. desparils of D’Orbigny. I am not
prepared to assert the specific identity of the recent with the fossil
shell, although the latter is found in the Eocene, Miocene, and Plio-
cene of the West Indies.

Gabb records three other species of Corbula, of which one is de-
scribed as new.

93. Nezra- oRNATISsIMA, D’Orb.

It is possible that to this species may be referred the shell formerly
regarded by me as JV. costellata, Desh. Gabb (I think, rightly) re-
gards as synonymous the JV. alternata of D’Orbigny.

94, Tetrina Brericata, Conrad, Tertiary Fossils, p. 36, pl. xix.
fig. 4.
T’. Sagre, D’Orb. Pal. Cuba.
Allied to 7. interstrialis, T’. sobralensis, and T'. ephippium.
I have not identified with certainty any of the other Telline re-
corded by Gabb. This one is not mentioned by him.

95. TELLIDORA CRYSTALLINA, Chemn.

This is recorded by Gabb, who identifies a Haitian fossil with the
west-coast species of the above name. I mention it only to observe
that I have found a shell in the West Indies which is nearly allied,
if not identical, presenting some slight though constant differences.

96. Lucrva tTieRmNA, Linn.
97. LucrmnA PENNSYLVANICA, Linn.

98. CrRASSINELLA MABTINICENSIS, D’Orb.

Crassatella martinicensis, D’Orb. Moll. Cuba, vol. 1. p. 288, pl.
xxvii. figs. 21-23.

C. quadelupensis, D’Orb., ib. p. 289, pl. xxvii. figs. 24-26.

This species occurs throughout the Miocene and Pliocene beds of
the West Indies. I have no doubt of the two forms named by
D’Orbigny being referable to one species, which belongs to the genus
Gouldia of C. B. Adams. The name Gouldia having been used for
a genus of birds, I proposed in 1874 thename Crassinella to replace
it for these shells.

99. Venus paputa, Linn.
100. Venus Branprana, Guppy, Geol. Mag. 1874, p. 436, pl. xvii.
fig. 8.
Allied to V. rugosa.
101. Venvs Woopwarn1, Guppy, G. J. vol. xxii. p. 292, pl. xviii.
fie. 1.

I think it not improbable that Chione Guppyana, Gabb, belongs
to this species.

Rk. J. L. GUPPY ON THE MIOCENE FOSSILS OF HAITI. 531

Gabb also records V. magnifica, Hanley, the recent analogue of
which is an inhabitant of the Philippine Islands.

102. CrruerEa JuncEa, Guppy, G. J. vol. xxii. p. 582, pl. xxvi.
fig. 13.

Closely related to C. circinata, with which Gabb considers it
identical—a view which I am not prepared to controyvert. But
should we not add C. acuticostata and Tryoniana of Gabb, although
placed in a different section?

103. CyrHEREA cARBASEA, Guppy, G. J. vol. xxii. p. 292, pl. xviii.
fig. 13.

104. Cyrnerna PLANIvieta, Guppy, G. J. vol. xxii. p. 292, pl. xviii.
fig. 3.
The figure cited is taken from a slightly distorted specimen.

105, Carpium HAITENSE, Sow. G. J. vol. vi. p. 52, pl. x. fig. 11.

106. Carprum LivevaLznonis, Guppy, G. J. vol. xxii. p. 292, pl.
xviii. fig. 7.
Gabb regards this as identical with C. subelongatum, Sow.

107. Carprum rnconspricuum, Guppy, G. J. vol. xxii. p. 293, pl.
xviii. fig. 12.
Gabb describes another species under the name C. dominicense;
but although he gives it nearly 60 ribs, I am not sure that it is dif-
ferent from C. inconspicuwm.

108. Erycrva tensa, Guppy, G. J. vol. xxii. p. 582, pl. xxvi. fig. 6.

This shell was described by me from beds which are now regarded
as Eocene. Gabb records it from Haiti.

109. Carprra scaBricostaTa, Guppy, G. J. vol. xxii. p. 293, pl.
xvii. fig. 10.

110. Caama Arcrnetra, Lam.

111. Cama rnvotuta, Guppy, Geol. Mag. 1874, p. 436, pl. xvii.
figs. 5 a, b, ¢.

Some enormously thickened and heavy examples of Chama in the
collection appear to me to be very old specimens of this species.

112. Arca parriora, Sow. G. J. vol. vi. p. 52.
Undoubtedly near to A. grandis, with which Gabb identifies it.

113. Arca consoprina, Sow. G. J. vol. vi. p. 52, pl. x. fig. 12.
Gabb states that A. floridana, Conr., is a synonym.

114. Arca occrpENTALIS, Phil.
A. noe, Guppy, G. J. vol. xxii. p. 293.

I am not sure that I have the right name of this shell, which is
determined by Gabb to be A. embricata, Brug.

OL R. J. L. GUPPY ON THE MIOCENE FOSSILS OF HAITI.

Gabb describes three other Arks ; and we have also A. pexata re-
corded from the Haitian Miocene.
115. Precrvncvnts pectssatus, Linn.

This appears to be the prior name of the species formerly re-
corded by me as P. pennaceus, Lam. (Hanley, Ips. Linn. Conch. p. 96),

116. Prcrvnovtus acuticostatus, Sow. G. J. vol. vi. p. 58, pl. x.
fie, 3.

Of Leda and Nucula we have several species recorded from the
West-Indian Miocene; but I have not recognized any of them among
the Haitian fossils.

117. Pecren tHeripts, Sow. G. J. vol. vi. p. 52.

118. Prcten oxyeonvs, Sow. G. J. vol. vi. p. 52.

P. exasperatus, Guppy, G. J. vol. xxi. p. 294.
I agree with Gabb in the above rectification.

119. Precren 1nmevatis, Sow. G. J. vol. vi. p. 52.

Pecten inequalis, Guppy, G. J. vol. xxii. p. 294, pl. xvii, fig. 6.
120. Osrrea virernica, Gmel.
121. OstrEa Harrensis, Sow. G. J. vol. vi. p. 53.

Gabb has corrected the error into which I fell in regarding this
as identical with the preceding.

122. Dirrupa DENTALINUM, Guppy, Geol. Mag. 1874, pl. xvi. fig. 11,
and 1875, p. 41.

EXPLANATION OF THE PLATES:

Puiate XXVIII.

Fig. 1. Marginella Sowerbyi. Fig. 6. Fusus Henekeni, type.

2. Fusus Henekeni, var. hai- 7. Pleurotoma consors.
tensis. 8. Orthaulax inornata.
3. Clavatula labiata. 9. Cancellaria epistomifera.
4, Murex cornurectus. 10, Phorus delectus,
5. Turbinellus eedificatus.
Puate XXIX.

Fig. 1. Murex textilis. Fig. 7. Pleurotoma squamosa.
2. Triton domingensis. 8. Terebra sulcifera.
3. Turbinellus haitensis. 9. Cerithium obesum.
4, Cerithium uniseriale. 10. Cypraea Gabbiana.
5. Conus marginatus. 11. Sigaretus excentricus,
6. Clea truncata,

1. XXVII.

Journ. Geol .Soc . Vol. XXXII.

Quart :

Mintern Bros. imp :

C.L.Griesbach,

MIOCENE, FOSSILS.

HAITIAN

Quart. Journ .Geol.Soc .Vol. XXXII. Pl. XXIX.

C.L.Griesbach.

Mintern Bros. amp :

HAITIAN MIOCENE, FOSSILS.

.
.
:
i
|
|

Geological Section along the Great Northern

Hempshill, near Cinderhill.

Hucknall Lane.

——
H 7 vy pt
a oP areal i
2 aoe a OOP ee ce SL
= (Se ES a SE = —s rai
ti +] ae DD :
it: —= — : - ; ae 7 OT y ’ p 5 r
Y es! eee ( ene a 1 $ i
Sei ae ee may Soe a ce ged ee 44 oe —
eee oe a A a ST 7 a -Z
fon g , a a, a a EE

Colliery.

Fauit,
Pre-Permian.

d5, Coal-measur
é. Permian Bre
e1, Lower Permi
e. Magnesian L

ON THE PERMIANS OF THE NORTH-EAST OF ENGLAND, 533

fF

56. On the PermrAns of the Nortu-East of Enexann (at their
SourmprN Maren) and their Retations to the unpER- and
OVERLYING Formations. By EK. Witson, Hsq., F.G.S. (Read
June 21, 1876.)

Tup main object of the present communication is to bring under the
notice of the Society a capital section (see figure), exposed by the
Great Northern Company’s Derbyshire Extension line, the peculiar
interest of which is that it exhibits the whole series of the Permians
of this district, and their relations to the under- and over-lying for-
mations.

The section begins at the Hempshill cutting near Cinderhill, and
ends at Kimberley West station. It bears, roughly speaking, north-
west for a mile, and then gradually curves round and bears south-
west, and throughout corresponds approximately in direction with
the dip of the red rocks exposed.

Going west we get the following succession of strata :—

Characteristic mottled and obliquely laminated Lower Bunter
sandstone (1), about 30 feet exposed, including its lower brecciated
portion. The breccia, which is alternately sandy, marly, and calca-
reous, expands near fault No. 1, to 5 feet, including sandy inter-
stratifications, but locally disappears on the east, where it becomes
very difficult to separate the Lower Bunter from the next under-
lying deposit.

The breccia contains semiangular green, blue, and purple slates,
more or less rounded grits, quartzites, quartz-breccia, and numerous
white and discoloured slabs and nodular balls of fossiliferous Carbo-
niferous-Limestone chert. Seeing that the yellow chert-balls are
the chief centres of cementation, and are deeply indented by the ad-
hering pebbles, it appears probable that their conversion into chert
took place after being deposited in the breccia. Beneath the breccia
comes a series of comparatively hard red-and-yellow-mottled and
soft grey sandstones, becoming, after exposure, chocolate-coloured, in
beds 1 inch to 1 foot thick, obliquely bedded, and containing lenti-
cular purple marls, especially in their lower portions. The Hemps-
hill upeast shaft passed through 39 feet of strong red sandstone,
nearly all of which must represent these beds. Similar strata were
first noticed by myself, several years ago, in Cinderhill brickyard
close by, as something different from typical Lower Bunter.

The question arises, what are these beds? Are they Permian?
are they Bunter? or are they passage-beds between the two?
Little importance need, I think, be attached to the breccia as a line
of demarcation, seeing that at Kimberley Knoll, Annesley, and else-=
where precisely similar deposits are met with at higher levels in the
Lower Bunter. I would suggest that they form a connecting link
between the Permian and Bunter formations of this district, and are
the lateral equivalents of the lower portion of the Lower Bunter.

reologucal Section along the Great Northern Railway between Hempshill, near Cinderhill, and Kimberley. (Horizontal scale, 240 feet to 1 inch; vertical scale, 60 feet to 1 inch.) Bo aes nageaee:

Hempshill, near Cinderhill.

Fault No. 1.
Downthrow to west about 10ft. 6in.

TUNNEL 242 YARDS LONG

Hucknall Lane.
eae 3
oe TEE
B.W. by W.
Old Quarry face, Kimberley
Kimberley. Turnpike-road.
1 '
1

Colliery.

Fault,
Pre-Permian.

d>. Coal-measure Shales, Sandstones, Ironstones, &c. é. Upper Permian Marls and Calecareous Sandstones.
e. Permian Breccia. ef. Sandstone, hard, dull, laminated, with lenticular marls
el, Lower Permian Sandstones, Shales, Lignite, Limestones. (passage-beds from Permian to Bunter).

f. Lower Bunter mottled Sandstone, with breccia at base.

@. Magnesian Limestone.
(NB. The base-line is that of the Railway.)

id

o34 E. WILSON ON THE PERMIANS OF

At Bestwood, for instance, they replace all but 8 or 10 feet of that
rock.

Then follow ordinary Upper Permian marls (e’) with dolomitic
sandstones, &c.; thickness in Hempshill upcast 21 feet. One of
these sandstones is literally covered with annelid-tracks.

These are succeeded by Magnesian Limestone (e*), rising west at a
small angle, and forming a dip slope ; when these beds fall west, near
Kimberley, their dip also governs that of the ground: 33 feet are
exposed in the vertical, at the east end of the tunnel; and this must be
near their maximum hereabouts. I call particular attention to the
indications this deposit gives of the proximity of land on the south.
It is here (z. e. on the Great Northern Railway linc) a coarse-grained
thin-bedded yellowish and red arenaceous dolomite. Atvarious levels,
and sometimes through several feet, it is more or less gritty, and
even becomes finely brecciated and conglomeratic. (Similar gritty
bands are met with in the new Midland line west of Bulwell, but
disappear going north.) There are also partings of highly micaceous
laminated soft sand, running to 3 or 4 inches thick, and often at
small intervals apart. Oblique lamination on a small scale often
occurs. Generally in connexion with the gritty bands, the rock
shows numerous cavities and casts of bivalves (Schizodus and Myti-
lus) difficult to determine with accuracy.

Further south at Strelley and at Radford Old Engine Houses the
dying-out Magnesian Limestone consists of red flagstones, with odd
pebbles and red shales. At Wollaton Old Park Farm and at Bobbers
Mill (south of Basford) it reposes on the Lower Permian breccia ;
and at Bobbers Mill I noticed, while the Leen sewerage works
were in progress, that in the course of a couple of hundred yards
south from the turnpike gate the thinning-out Magnesian Limestone
passes from an ordinary granular dolomite through a fine to a coarse
brecciated rock, which seems to show that here, at any rate, we
have reached a definite point on the original south margin of the
Magnesian-Limestone sea.

Succeeding with perfect conformity the Magnesian Limestone,
comes a series of thin-bedded slate-coloured sandstones and shales (e').
Inclusive of a breccia at their base, they maintain along the Great-
Northern-Railway section a uniform thickness of 19 to 20 feet,
which (as shown at a point on the section) comprise no less than
75 different beds 7" to 8” thick. Several of the sandstones con-
tain a large amount of imperfect plant-remains, woody stems and
lignite permeated with iron pyrites; other bands are covered by
annelid markings; and at least one sandstone shows casts of Schi-
zodus. ‘This series represents the Marl Slate of Durham &e.

These beds, which with the overlying limestone are traversed by
numerous vertical joints, become along these, as also along some of
the beds, rusted or discoloured light yellow. Supposing the blue
colour to be due to finely disseminated carbonaceous matter inter-
mixed with the bisulphide of iron, the discoloration is most pro-
bably due to the production of the sulphate and its decomposition
into the hydrous peroxide, though the same result may have been

THE NORTH-EAST OF ENGLAND. Son)

brought about by the peroxidation of the carbonate by carbonic
anhydride, or by the conversion of the protoxide into the hydrated
peroxide by access of oxygen introduced in water finding its way
from above along the joints. In the three lowest feet of these beds
are (locally) incorporated three minor brecciated seams. At the
base of all comes a breccia varying in texture from a fine siliceous
sandstone or grit a few inches thick to a coarse and massive brecci-
ated rock 4 feet thick. Its upper surface frequently exhibits beauti-
ful gently swelling ripple-marks. Its contained fragments are
plainly seen to be largely derived from the fine ferruginous red and
yellow coal-shales, the sandstones, and ironstones of the neighbour-
hood; there are also angular or subangular fragments of slate,
quartz, quartzite, &c. The fragments are stuck-in confusedly at all
angles. The breccia, as also the overlying sandstones and lowest
Magnesian Limestones, contain many geodes lined with cale-spar
and iron pyrites.

It will be seen that these sandstones, which have a marl-slate
facies and horizon, have but locally this character. Going south, in
a couple of miles they thin out altogether, and the breccia comes up
against the base of the degraded and dying-out Magnesian Lime-
stone ; while, on the other hand, as we go north and north-east
to Watnall, Bestwood, Linby, Annesley, Kirkby, Skegby, &c., they
pass into a series of shales increasing in thickness, with subordinate
bands of fine-grained sandstone and compact limestone. These beds,
from their close resemblance in colour and texture, have in many
instances been taken for coal-measures, and mapped accordingly.

Returning to our main section, we find that the breccia rests in a
series of very gentle undulations on a planed-off surface of coal-
measure shales, sandstones, &c. (d’). These latter beds dip in a
north-easterly direction, at an angle of fully 15° (but, as the railway
section is there running about N.E. by E., the full dip is not exhibited
in the face of the cuttings). At one point a fault of unascertained
throw crosses the line, affecting coal-measures, but not Permians.

Hence it will be observed that the unconformability between the
Coal-measures and the Permian is most pronounced.

This is, I am aware, nothing new: it was a long while ago
pointed out by Sedgwick ; and colliery observations have rendered it
a matter of notoriety in the Derby-Notts-Yorkshire coal-field: but
never before has there been in this district such a splendid above-
ground verification of the view that an enormous lapse of time
intervened between the close of the Coal-Measure and advent of the
Permian epoch in England, accompanied by the elevation and fold-
ing of the strata, not only along east and west (¢. g. Pendle and
Cheshire anticlinals), but also north and south (¢. g. Pennine) axes,
and by the sketching-out of the great coal-basins by denudation.

The general tendency of the breccia (at base of Permian) to expand
westwards, and the occasional presence, according to Sedgwick, of
Mountain-Limestone pebbles therein, seem to indicate the existence
of a barrier of high ground, and perhaps also the exposure of the

bested

ef

536 E. WILSON ON THE PERMIANS OF

Carboniferous Limestone in that direction during its accumulation * ;
but surely the pre-Permian elevation and denudation of the Pennine
axis is satisfactorily demonstrated by the above-mentioned overlap of
Coal-measures, dipping east at so high an angle, by Permians prac-
tically horizontal. The total absence of that very durable rock the
Magnesian Limestone at any distance from its escarpment (e.g. west
of the Erewash) indicates, to my mind, that we have there passed
beyond its original margin in that direction, a margin that would
probably have been found as on the south, had it not happened to
lie in the course of a powerful denudation. The great dissimilarity
in mineral character and sequence of the Permian deposits on the
west from those on the east of the Pennine axis points to the same
general conclusion. It is not without considerable diffidence and
reluctance that I venture to differ in any way from so high an autho-
rity as Professor Hull; but, with all due deference, I must say that
I cannot help considering the evidence afforded by the Red-Rock and
Anticlinal faults near Stockport, and adyanced by him in proof of the
pre-Permian age of the Pennine axis, to be at best equivocal. For,
as I apprehend the matter, we have there two contiguous and parallel
faults. One of these, the Red-Rock fault (A), Professor Hull sup-
poses has had two moyements, the earlier of which was synchronous
with the only (local) moyement of the other or Anticlinal fault (B);
but if so the later throw of A was the greater throw of A; and yet
that movement did not affect Bin the slightest degree—though, being
presumably at that time dislocated, it might be expected thatif they
ever acted in unison they would do so then. But ag they did not
do so then, their parallelism does not prove that they did so before ;
and the old objection still remains that A faults Bunter &c. and B
passes under without affecting it. Hence there is no particular reason
for believing that the Anticlinal fault, or by consequence the Pennine
system to which it belongs, was post-Permian. That there was an
old east-and-west barrier, as enunciated by Jukes, and that this
barrier divided the Lancastrian and Salopian Permian hydrogra-
phical areas, as so ably argued by Hull (Q.J.G.8. 1869), I do not
doubt—as also that the easterly extension of that barrier may be
taken to explain the difference between the Permian deposits of Leices-
tershire and Notts, and the southerly degradation and attenuation
of the latter; but, so far as I have been able to trace them, I do not
find that the Permians of the north-east of England show any ten-
dency to assume a Lancastrian facies going west. There is indeed
a change in their character most decided and general; but that
change is a tendency to get thicker and (with local variations)
purer going north or north-east, indicating successive increasing
subsidences of the Permian north-east basin in such directions, As

* The before-mentioned prevalence of Mountain-Limestone chert in the
brecciated Lower Bunter clearly shows that prior to the accumulation thereof
the Pennine axis in Derbyshire had not only been formed, but had undergone
an enormous amount of denudation.

THE NORTH-EAST OF ENGLAND. fen

to the details of this process I hope on a future occasion to be able
to give more definite results*.

One word as regards the origin of the lower breccia. ‘This
striking rock band (the miners’ “‘ mingled” or “ plumcake ” rock)
maintains its average thickness and coarse texture over a consider-
able area, and must have been simultaneously deposited over several
square miles, and in water of variable depth and distance from land.
Though no striz have, I believe, been found, the angularity and con-
fused arrangement of the fragments, the fact that some of the largest
have travelled a long distance, the general absence of any attempt at
stratification, and the sudden transitions in thickness and texture of
the breccia, point possibly to its glacial origin, as droppings, say, from
the melting of icebergs or ice-floes. (Occurring, however, along a
plane of erosion, no such special theory may seem required to account
for its origin.) The same remarks may apply almost equally well to
the uppermost or Lower Bunter breccia.

To Richard Johnson, Esq., Engineer to the Great Northern Rail-
way Company, the author desires to express his acknowledgments.
The principal section exhibited is founded on the Company’s sec-
tions he has so liberally supplied ; and its completion is due to the
facilities given by him for exploring the railway-cuttings from
their commencement to their present advanced state.

* Tt will be noticed that along the line of section we have no representative
of the Rothliegende. This is only one of many instances in which I have met
with proofs of the non-existence of any such formation in Notts or Derbyshire.
My limited experience of the so-called Lower Red Sandstone of Durham inclines
me to concur with those who would relegate such rocks to the Carboniferous
formation, as has been recently done with similar strata in Yorkshire. Hence
it appears that the Marl Slate, or in its absence the Lower Magnesian Lime-
stone, forms the true base of the Permian series in the north-east of England,

iy

*% *s) | yf as
‘aa
ey

ce

SMITHSONIAN INSTITUTION LIBRARIES
“TATAIETANTAT EMT
3 9088 01350 1754