Ses. ap ere c; ? Ezek ; ° ? ? > » a > fem G06 “ Setanataneae od ae ‘ . pets — a — par eae Ame on - - orth nehr-e=- Ba ie tnepaart - ’ a. nal . “ nt-hebeboge ew - ‘ . po we A. . ; san ON term enn ey te 4 Doberpettei deride teed ng ; ue ~ ‘ ath Wand tem rhe Cr oe 5 Tabs > te b> 2 nit mine e nob te rahe tchanennd ste Berbeany Sastuiee tN ean beter tit te ateaee dan tee h pricien Seaes rember eee ese = Z ee ~~ 4h ~ ‘ ‘ +t ie paar e Ce en ia Fa : ; ~ ty dat " ee ee =e © tert Se se ee ae ee rile . : : ae ene ah — ‘ ot eae Sa RAST a “> Beds ne rr San eandene “ ee . . .“ re ee) PMA Pee ce we a ee E FS ¥ i ne lee ee i i lee at! Je ook ashe We re as ot 30 7 ~% bi 7 rT 7 \ s A 2 Lipaa sh a re, OL : Aas ta ‘ ai = -—% ==> ¢ ie bi ™ ARN we mere alt axe tecoes Sa Pattie Wiaded = I iiecace ts MESA MEH , we] Ope naldyess YG : ‘ PROS 4 Seciriey Selligh, Tearing. sen) See eran as I -- 71 “_~ , f » CHE THIRTY: Nie oi é i Bary, WUE RAOTS 7a, ih a Se, aia anf TEN SO0sS TY, of oc Sale B50. 642 a ce ee ae é 4eo/. cde saci di ie ; THE QUARTERLY JOURNAL GEOLOGICAL SOCIETY OF LONDON. EDITED BY THE ASSISTANT-SECRETARY OF THE GEOLOGICAL SOCIETY. Quod si cui mortalium cordi et curz 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-NINTH. PN 1883. [: \ LONDON: LONGMANS, GREEN, READER, AND DYER. PARIS: FRIED, KLINCKSIECK, 11 RUE DE LILLE; F. SAVY, 24 RUE HAUTEFEUILLE. sade LEIPZIG: T. 0. WEIGEL. SOLD ALSO AT THE APARTMENTS OF THE SOCIETY, MDCCCLXXXIIL, List OF THE OFFICERS GEOLOGICAL SOCIETY OF LONDON. Elected February 16, 1888. PrestVent, J. W. Hulke, Esq., F.R.S. Wice-PrestVents. Prof. P. M. Duncan, M.B., F.R.S. J. Gwyn Jeffreys, Esq., LL.D., F.R.S. R. Etheridge, Esq., F.R.S. Prof. J. Prestwich, M.A., F.R.S. Secretartes. Prof. T. G. Bonney, M.A., F.R.S. | Prof. J. W. Judd, F.R.S. Foreign Secretary. Warington W. Smyth, Esq., M.A., F.R.S. Creasurer. Prof. T. Wiltshire, M.A., F.L.S. COUNECIL. H. Bauerman, Esq. J. Gwyn Jeffreys, Esq., LL.D., F.R.S. W. T. Blanford, Esq., F.R.S. Prof. T. Rupert Jones, F.R.S. Prof. T. G. Bonney, M.A., F.R.S. Prof. J. W. Judd, F.R.S. W. Carruthers, Esq., F.R.S. S. R. Pattison, Esq. Prof. P. M. Duncan, M.B., F.R.S. J. A. Phillips, Esq., F.R.S. R. Etheridge, Esq., F.R.S. Prof. J. Prestwich, M.A., F.R.S. John Evans, Esq., D.C.L., LL.D., F.R.S. F. W. Rudler, Esq. A. Geikie, Esq., J.L.D., F.R.S. Prof. H. G. Seeley, F.R.S., F.L.S. Rev. Edwin Hill, M.A. Warington W. Smyth, Esq., M.A., F.R.S. G. J. Hinde, Esq., Ph.D. W. Topley, Esq. Prof. T. M‘Kenny Hughes, M.A. Prof. T. Wiltshire, M.A., F.L.S. J. W. Hulke, Esq., F.R.S. Assistant-Secretary, Librarian, anv Curator. W. S. Dailas, Esq., F.L.S. Clerk. Mr. W. W. Leighton. Library and Museum Assistant. Mr. W. Rupert Jones. TABLE OF CONTENTS. Page Bonney, Prof. T. G. The Hornblendic and other Schists of the Lizard District, with some Additional Notes on the Serpentine. “2 E bo) UB AEA AEM IS igen CA ae A a apie ieee re Notes on the Lithological Characters of some Rocks SEPM COSe ANE I VEEMESS SUITES. \e/¢ u's = «oye sins wie chm ecco e one 159 . Additional Notes on Boulders of Hornblende-Picrite near gue Western Coast of Anolesey’ ono ctan od ese wi slswie eres 254 Notes on a Series of Rocks from the North-west Highlands, gallecied. by, ©. Callaway, (Msgs: ira iroch wuebtxteinclete ails le » dejo’ +) 414 On a Section recently exposed in Baron-Hill Park, near ee AERIS ogc a PO yt at el arate Soa aah ea oatele «.da,aee oe, de 470 On the Rocks between the Quartz-felsite and the Cambrian Series in the Neighbourhood of Bangor ................000. 478 Catitaway, Dr. C. The Age of the Newer Gneissic Rocks of the Northern Highlands.— With Notes on the Lithology, by Prof. PG DONNY foie fae shai end oe BEF See tclee cee te perce es 355 CLAYPOLE, Prof. E. W. On Helicopora,a new Spiral Genus (with three species) of North-American Fenestellids. (Plate IV.) .. 30 Daviv, J. W. Encewortn, Esq. On the Evidence of Glacial Action in South Brecknockshire and East Glamorganshire .... 39 Dawkins, Prof. W. Boyp. On the Alleged Existence of Ovibos moschatus in the Forest-bed, and on its Range in Space and EEG Ras PAA Ae es ee cis meade «see soe n ses o sider: 575 Ditter, J.8., Esq. Notes on the Geology of the Troad. A brief Summary of the Results derived from the Observations made in connexion with the Assos (U. 8.) Expedition. With an meppond ix, by: We PORiny) Hedin ysis atcha op tte alae dade ones 627 GARDNER, J. S., Esg. On the Lower Eocene Section between Reculvers and Herne Bay, and on some Modifications in the Classification of the Lower London Tertiaries .............. 197 GEIKIE, ARCHIBALD, Esq. On the supposed Precambrian Rocks of pieomaes (Ch iates VIELE M Ao ceviche res anene cca’ 261 iv _ TABLE OF CONTENTS. Page Gray, THomas, Esq. On Gray and Milne’s Seismographic Apparatus 218 ———, and Prof. J. Minx on the Elasticity and Strength-constants of Japanese TOCKS 2.0.10. ai. ns de cee OR ERs. 2 eee 1389 HAvsier, Dr. R. Notes on some Upper Jurassic Astrorhizide and ituolides. (Plates TL. sealll.). 0. 00. Ree oe. Hicks, Dr. H. On the Metamorphic and Overlying Rocks in parts of Ross and Inverness Shires. With notes,on the Microscopic Structure of some of the Rocks, by Prof. T. G. Bonney. (Plate Rat) oo eenene tel ba bik cae Cone elem sip.6 eles «as le ietere 141 Hupueston, W. H., Esq. Notes on a Collection of Fossils and of Rock-specimens from West Australia, north of the Gascoyne River. (Plate RAIL) ya ehh es ooh cca cca ek eee ee 582 Irvine, Rev. A. On the Mechanics of Glaciers, with especial Reference to their supposed Power of Excavation . On the Origin of Valley-Lakes, with especial Reference to the Lakes of the Northern Alps ......5....... <0 Jupp, Prof. J. W. On the Basalt-glass (Tachylyte) of the Western Isles of Scotland. ‘(Plates XT. & XIV.) ...... +... cen Att Juxres-Browne, A. J., Esq. On the Relative Ages of certain Rrver-valleys in Lincolnshird © ....%4... +s <2. eo 596 Keepine, H., Esq., and E. B. Tawney, Esq. On the Section at Hordwell Cliffs from the top of the Lower Headon to the base of the Upper, Bagshot Sands... . . .. PI. Six, f. VS occa swsecscas ae Page 537 537 536 FOSSILS FIGURED AND DESCRIBED. 1x’ Membranipora arethusa. PI. xii. PM tncie sce naciae=avremanas snes. lusoria, var. coarctata. PI. | Name of Species. Formation. Locality. | Page. PoLyzoa. Catenicella alata. PI. xii. f. 15, 16.. ) (; 428 ARENA eciaicina sisleiesislsjerinics sicaiewmein 429 BIRECUMCINONG «5 00.000 0e's gsi scien 432 —— Harveyi. PI. xii. f.5 ......... | 431 BASGATA s.0c0s.000500 BPRS Shar < : J) 431 | sunsigden, Geae yepe f Tertiary ...,..0. Australia......... | 430 —— levigata. Pl. xii. f.1 ......... 432 longicollis. Pl. xii. f. 2-4...... 432 SAMOA U.meect inser dcieensrscses | | 431 MEMUNICONA ..-.00. 0. te---... Pinna claviformis. Pl. xvi. f. 11... dundriensis. Pl. xvi. f. 10... |. Placuna Rupertina. Pl. xv. f.16.. sagittalis, Pl. xv. f.17 ...... Plicatula Sollasii. Pl. xv. f. 21, 22 subserrata Pl. xvi. f. 4,5... Sphera crassicosta. Pl, xvi. f. 16. —— fimbriata. Pl. xviii. f. 10, 11 Thracia leguminosa. PI. xviii. f. 23 —— Studeri. Pl. xix. f. 6 ......... ) ( Cephalopoda.) Ancyloceras Waltoni. PI. xix. f. 1 “| Toxoceras Orbignyi. PI. xix. f. 2... Inferior Oolite... Locality. England eereee x1 | Page. eS ES SS Sees ee Seen ents NS EE EE SE a ———-_——, (1 530 530 518 519 522 523 522 529 491 490 490 492 492 493 497 498 498 500 499 500 501 501 502 502 534 532 533 533 534 518 518 496 497 515 515 526 525 531 531 488 488 xil FOSSILS FIGURED AND DESCRIBED. Name of Species. Formation. Locality. Page. | VERTEBRATA. ( Reptilia.) Megalosaurus Bredat...........0..000.00 Cretaceous ...... Maestricht ...... 246 bucklandi.” (Pl. xi 22s. Inferior Oolite...|. Dorsetshire ...... 334 Orthomertis DallOt «.c.cbe.oc sais connec Cretaceous ...... Maestricht ......| 248 Sphenospondylus ....... bie desist See Wealden: <.... 2s. Isle of Wight ...| 55 (Mammaiia.) Ovibos moschatus .........+. mie cekenat | Forest-bed ...| Norfolk.......... sn DAS EXPLANATION OF THE PLATES. PLATE PaGR r APPARENT CURRENT-BEDDING in Lizard Rocks, to illustrate i Prof. T. G. Bonney’s paper on the Hornblendic and other Senisiniet the: Demmed Wistriet. soi sv.2.0 cet eaasceeenacukedaweve 24 II fh Swiss Jurassic ForRAMINIFERA, to illustrate Dr. R. Hausler’s lr paper on some Upper Jurassic Astrorhizide and Lituo- PERNT I ae Gk RCC EI SNM Na ps Seee see nae Re ed 28 Iv, 4 Hericopora, to illustrate Prof. E. W. Claypole’s paper on is t (SLATES FECTS ae al a ee ee ee ee ee ee 38 Mar or tHe Basin of the Mersey and of part of the Dee, to Vv illustrate Mr. T. Mellard Reade’s paper on the Drift-beds of the North-west of England and of North Wales ......... 132 Map or Parts or Ross- AnD INVERNEsS-SHIRES, to illustrate VI. Dr. Hicks’s paper on the Metamorphic and Overlying Rocks Mi vGISRLELD so... ve sobeslse mast sad speapeae snes tiecioesiods sdenateccas 166 Grear-Ooute CoraLs, to illustrate Mr. R. F. Tomes’s paper VIL on the Fossil Madreporaria of the Great Oolite of Glou- Gesiers AMG OXtOLOSAIEES, <.2.\ sap aceeelenaccenosas on tdasicadioveoeean 195 Sxetcu-Map of the St. David’s district to illustrate Mr. A. VIII. Geikie’s paper on the supposed Pre-Cambrian. Rocks of Diese pee RS eee o6e IX. { Rocx-sections, to illustrate Mr. A. Geikie’s paper on the X. supposed Pre-Cambrian Rocks of St. David’s ............... 325 XI Mecarosavrvs Buckuanpt, to illustrate Prof. Owen’s paper ; ONL HAS SUMO MCG OIOSHU TUS «cio co's ssigwed salto tipwors's odin cian on 346 XII VicrortaAn CninostomAtTous Bryozoa, to illustrate Mr. A. Wi. Weaters's paper om those fossils) oo) ic.0ckcesseetWices ee 443 XIII. f Basaur-cuass, to illustrate Messrs. Judd and Cole’s paper XIV. on the Basalt-glass of the Western Isles of Scotland ...... 463 Xl1vV EXPLANATION OF PLATES. PLATE PAGE XY: XVI. ’ : Inrer1or-Oouite Fossius, to illustrate Rev. G. F. Whid- XVI bOrMe's! Paper a. sk... ned iinet. anode ee Benee «ot «ses ee eee rr 538 XIX. Iyrertor-Oo1itE Sroness, to illustrate Prof. Sollas’s paper on War! ‘ a ~ Se Q= —— D. A rounded knoll, probably a large roche moutonnée. Intruded Boulder-clay is indicated by a light horizontal tint. Ps rv which are soft and easily disjointed, . Palio ah are more often shattered and con- + 1 oe 9G torted than striated. This is the case in the Old Red Sandstone area south of the Beacons, and with many of the finely laminated sandstones of the Coal-measures and the Triassic and Liassic rocks of the Ely valley. : This shattering and contorting of rock-surfaces is much more common as an evidence of glacia- tion in the coal-basin than the grooving and striating of the sur- face-rock. It is to be met with everywhere, on the north, north- western, and north-eastern slopes of the Carboniferous-sandstone hills. In the Ely valley this rock- shattering is particularly well marked in the rocks belonging to the Lower Lias formation. At Ty- fry, near St. Fagans, there are seve- ral roches moutonnées, which illus- trate the contorting and crushing power of ice and the intrusion of Boulder-clay. At Ty-fry, where a quarry has been opened in two of these roches moutonnées, the rock has been so much crushed as to form a fine angular breccia, - : “No iC 6 C4 = \ age 6 > = q IAD. (tales 2 ka q Vv ay SS B, ©. Rounded blocks of Lias scratched all over by ice. . BRECKNOCKSHIRE AND EAST GLAMORGANSHIRE. 49 well adapted for gravelling drives and footpaths, for which purpose it is actually employed. Boulder-clay has been forced into the shattered strata to a depth of 12 feet at least. Ice-scratched stones are often to be found mingled with this débris, as shown in the accompanying section (fig. 3). 4. Grooved and striated Rock-surfaces are exhibited, under favour- able circumstances, by three kinds of rock in this area—the Carbo- niferous Limestone, the Millstone Grit, and the Carboniferous sand- stone, especially the Pennant rock. The Millstone Grit retains glacial markings on its surface even after long exposure to the weather ; the Carboniferous sandstone shows grooves or striae, if its surface has not been long exposed, or is sheltered by a thin covering of Boulder-clay ; the Carboniferous Limestone quickly loses all trace of glaciation on exposure. The furthest point east at which the author has seen striz is at Craig-y-gaer, 5 miles west of Abergavenny, somewhat beyond the prescribed area of which this paper treats. The mamillated sur- face of the Millstone grit is here very faintly striated, the trend of the striz being about 70° E. of N. and W. of 8. The author was unable to determine from which end the strize started ; if from the east end, they may have been produced by the Scandinavian ice-sheet ; if from the west, they may be due to a local glacier nearly conforming to the trend of the Clydach valley. Going westward, the next point at which the author has seen striz is at Twyn-cilog, three miles north of Rhymney. Here their trend is about 37° H. of 8., the ice which caused them haying come from the north-west ; the surface of the Millstone Grit here is mamillated; the height of the striz above the sea-level is about the same as that of those at Twynau Gwynion, 14 mile west, viz. 1840 feet ; surface of Millstone Grit at Twynau Gwynion mamillated, trend of strie 36° H. of 8., pomting in the direction of the Brecknockshire Beacons. This is the highest point above the sea-level at which striz are known to the author. They are 7 miles distant from the Beacons, 2910 feet high, giving a fall of 153 feet per mile. Proceeding westward, from Gwern-cefn-y-garreg, the strie gradu- ally swing round so as nearly to converge on the top of the Beacons. At the E. “sile of Cefn-cil-sannus their trend is 12° E. of S.; the sur- face of Millstone Grit is crushed, mamillated, grooved, striated, and polished; greatest height above the sea-level of striee, as taken by aneroid, 1278 feet; fall from top of the Beacons 233 feet per mile. On the west side of Cefn-cil-sannus, the trend of the striz is 8° EH. of 8.; atthe east end of Mynydd Penmailard, first 3° E. of S., then, one mile further west, N.and 8S. ‘The respective heights of these two last points above the sea-level are 1095 feet and 1199 feet, as taken by aneroid; fall from the Beacons of highest point 228 feet per mile. Between these two points there are several angular slabs of Old Red Sandstone, the largest measuring 4 ft.x2 ft. 6 inches x 2 ft. 6 inches. There are a great number of angular and rounded blocks of Old Red Sandstone on the eastern slope of this mountain. A mile and three quarters further west, on the same mountain, QO) J.G.o: No. lds. BE 50 J. W. E. DAVID—GLACIAL ACTION IN SOUTH the trend of the strie is 30° W. of S.; at Cefn Cadlan they run 38° W. of S., and at Gwern-cefn-y-garreg 42° W. of 8. All these striz are on mamillated surfaces of Millstone Grit. At Gwern-cefn-y- garreg, their greatest height above the sea-level, as taken by aneroid, is 1444 feet. The surface of the Millstone Grit here has been moulded into " elongated domes and hummocks (fig. 4), showing a bare bright surface where they are not covered by peat or powdered rock. Wherever Fig. 4.— Roches moutonnées, Gwern-cefn-y-garreg ; Millstone Grit. its surface has escaped being crushed, it has been well striated, grooved, and polished, the striz pointing direct to the two tabular tops of the Beacons, distant 6 miles, giving a fall of 244 feet per mile. A mile. S.W. of Gwern-cefn-y-garreg, on the right bank of the Mellte, 500 yards below Ystrad-fellte church, a recent slip of Boulder-clay has exposed a well-glaciated surface of Carboniferous Limestone. Where it has not been completely shattered, this rock has taken a fine polish, is firmly striated, and slightly grooved. This surface forms a steep slope in a bend of the river, which the ice must have struck obliquely at an angle of 30°. Allits projections have been striated in a determinate direction, while the sheltered ledges show strize running in various directions, often at right angles to the true lines of glaciation. The. author thinks that all these striae were formed contemporaneously. The strike of the Carboni- ferous-limestone beds here being nearly horizontal, and the bedding even and definite; the stones, which the ice was forcing obliquely up the slope would be arrested at the points of junction between the beds, and would have a tendency to slip along these lines of weakness, and scoop out stepsin the slope, as has actually been the case. The direction of the grooves and coarse strie 1s less persistent than that of the fine scratches, the former often describing a curve, which increases with the steepness of the slope. These strie run down to the brink of the river Mellte, and were evidently continued below the present level of the valley-bottom. Their height above the sea-level here is 740 feet, giving’ a fall from the top of the Beacons, 7 miles distant, of 310 feet per mile. The furthest point west to which the author has traced these strize BRECKNOCKSHIRE AND EAST GLAMORGANSHIRE. 5 on the Millstone Grit, is Pant-llwyd, half a mile 8.W. of the last point. Their trend here is about 38° W. of 8. Traces of them are probably to be found much further westward, as the strie at this last point look very fresh and are firmly cut. The total extent of this striated surface of Millstone Grit from Twyn-cilog on the east to Pant-llwyd on the west is 113 miles; and the striz at either end make with one another an angleof 75°. This extent of convergence seems clearly to imply that the ice which produced these strize radi- ated from the Beacons. Going south, down the Neath valley, towards the steep scarp of the Coal-measures, a striated surface of Carboniferous sandstone is to be seen at the top of the cutting at the east end of the trend on the Neath-valley Railway, between Hirwain and Glyn Neath. This striated surface, which has been exposed from under a sheet of Boulder-clay from 12 to 15 feet thick, is fairly flat; and the trend of the strie is about from 80° to 85° E. of N. and W. of S. The author was unable to determine the end from which the strize started ; if it be at the east end, as seems most probable, they may have been caused by glaciers from the Beacons deflected into the Neath valley by the steep northern scarp of the Coal-measures, or may be connected with the striz at Craig-y-gaer, and be due to the Scandinavian ice-sheet; if they. begin at the western end, they may be deflected striz belonging to a great glaciation from the north-west, of which there are possible traces at many points in the area of the Coal-measures. The author has not ascertained the height of these striz above the sea-level; but they are probably not much over 700 feet, whereas Carn Mosyn, the highest point in the escarpment, 24 miles south of this point, is 1971 feet high. 13 mile east of Carn Mosyn a mamillated surface of Carboniferous sandstone is grooved and striated in a direction about 58° E. of S. Height above the sea-level, as taken by aneroid, 1530 feet, distance from Beacons 12 miles, fall 115 feet per mile. 12? mile further east some faint scrapes (?) on a shattered surface of Cockshot point in about the same direction ; their height above the sea-level is 1417 feet, as taken by aneroid; distance from Beacons 114 miles; fall about 130 feet per mile. Neither of these striated surfaces can be connected with the glaciers of the Beacons. At Abernant, near Aberdare, 4 miles east- wards, the trend of the strie, on a mamillated surface of Carboni- ferous sandstone, is 18° E. of 8S. Their height above the sea-level is 1130 feet, as taken by aneroid; distance from the Beacons 114 miles ; fall about 170 feet per mile. At Navigation House, 7 miles distant from the last point, and bearing 35° E. of S., the trend of the strie, on a mamillated surface of Carboniferous sandstone, is 20° E. of 8.; height above the sea-level, as taken by aneroid, 350 feet ; fall from the Beacons, 17 miles distant, about 150 feet per mile. Bearing 38° H. of N. from last point, 32 miles distant, there are striz, on a smoothed surface of Carboniferous sandstone, on Gelligaer Common, about 300 yards N.N.W. from the ruins of Capel Gwladis. EZ 52 J. W. E. DAVID—GLACIAL ACTION IN SOUTH Their trend is 47° E. of 8.; height above the sea-level, as taken by aneroid, about 1320 feet; fall from the Beacons, 15 miles distant, 106 feet per mile. Bearing nearly due south from Gelligaer Common, 53 miles distant, there is a faintly grooved flat surface of Carboni- ferous sandstone, near Cwm Sarn, on the south-east spur of the Eglwysilan mountain. Their trend is about 9° E. of 8.; height above the sea-level 980 feet ; fall from the Beacons, 293 miles distant, about 94 feet per mile. Bearing nearly due west from Cwm Sarn, 34 miles distant, there are well-marked grooves on a mamillated surface of Carboniferous sandstone, near the ‘“ Rocking-stone,” Pont-y-pridd. The impact- surface (‘‘Stossseite”) of this hill, as of most of the hills in the coal- basin, 1s too much shattered to show striz. The trend of the grooves here is about 7°E. of §., or nearly parailel to the grooves at Cwm Sarn ; height above the sea-level, as taken by aneroid, 360 feet; fall from the Beacons, 194 miles distant, about 131 feet per mile. Passing now to the Rhondda valley, west of the Taff valley, faint striz are discernible on a faintly ground surface of Carboniferous sandstone, near Pen-y-graig, in the Rhondda-fawr valley. Their direction is from 30° to 40° E. of §., following the trend of the valley. Their height above the present level of the valley-bottom is about 40 feet. They bear 68° W. of N. from the striz near the “ Rocking-stone,” 54 miles distant; distance from the Becaons 18 miles. Bearing nearly due east of the Pen-y-graig striz, 44 miles distant, between Cefn and Lan farmhouses, a mamillated surface of Car- boniferous sandstone has been exposed from under some rubble in making a road. This is grooved and striated in a direction about 33° E. of 8. Its height above the sea-level, as taken by aneroid, is 555 feet; distance from the Beacons 183 miles ; fall about 127 feet per mile. Bearing 17° W. of 8. from the “ Rocking-stone” striz, 3 mile distant, are some faint strize on a rounded surface of Carboniferous sandstone, running 26° E. of 8.; height above the sea-level, as taken by aneroid, 482 feet; distance from Beacons a little over 20 miles; fall about 128 feet per mile. Bearing 8° W. of S. from this last point, one mile distant, an extensive striated surface of Carboni- ferous sandstone has recently been exposed from under a covering of Boulder-clay from 5 to 6 feet thick, at Maendu quarry, near Treforest. Trend of striz nearly due N. and 8.; surface flat, height above the sea-level 705 feet. This is the southernmost point in the whole area at which the author has seen a striated rock-surface. The distance from the Beacon is 21 miles; fall 105 feet per mile. Summary.—It would be premature for the author, in the present imperfect state of his knowledge, to attempt to systematize all these glacial phenomena; as, however, a few points seem already esta- blished, it may be well to state them. I. The erratics in the Eglwysilan and Caerau group were pro- bably transported by floating ice. This is implied (1) by the BRECKNOCKSHIRE AND EAST GLAMORGANSHIRE. 53 smallness of the fall between the levels of the resting-places of these erratics and that of their parent rocks, (2) at Cwm Sarn by the want of conformity between the lie of the longest axes of the erratics and the trend of the grooves on the neighbouring rock- surface. ‘This point, however, needs to be more fully worked out before it be accepted as true. (3) The contact in this group of perfectly angular and rounded boulders can be most easily explained on the hypothesis of an ice-foot, transporting a mixed load of water- worn or glaciated boulders from shore-lines and angular rock-frag- ments from overhanging cliffs. No definite conclusion can as yet be arrived at with reference to the rest of the boulders, with the exception perhaps of those composed of Old Red Sandstone, in South Brecknockshire, the frequency with which these erratics occur angular, and the certainty of land-ice having once radiated from the Beacons, justifymg the inference that many of them are the relics of old moraines. II. The Boulder-clay of South Brecknockshire is chiefly the pro- duct of land-ice; for (1) it is unstratified ; (2) its included stones are intensely glaciated; (3) angular stones greatly predominate over rounded ; (4) in one case the largest of the included stones were observed to lie with their lengths parallel to the lines of glaciation on the underlying rock-surface. Til. The striated rock-surfaces of South Brecknockshire have been formed by land-ice, descending existing valleys, from the Beacons, at least as far south as the edge of the Coal-measures. South of this line the striated rock-surfaces present a more complicated problem. Those at the Craig-y-gaer and at the top of the Rhondda- fawr valley can hardly be attributed to the glaciers of the Beacons ; and yet the former were certainly, and the latter probably, caused by land-ice. . In the coal-basin the striz having a S.E. trend and occurring at high levels may be due to the glaciation of an ice-sheet coming from the N.W., or the grounding of icebergs; the mamillated sur- faces, however, point to land-ice as the agent. The striz which are found at lower levels in the Taff and Rhondda valleys may be due to (1) this N.W. ice-sheet, (2) the glaciers of the Beacons, (3) glaciers having their origin in the coal-basin. The difference of direction in which the low-level striz run as compared with the high-level, ¢. g. near Pont-y-pridd, may be due to the land-ice conforming more and more to the trend of the valleys as it became localized, or to their belonging to a glaciation altogether more recent. The author has received much valuable advice from Professor Prestwich. He has also to acknowledge the important help given him by Dr. C. T. Vachell, M.D., Mr. W. Adams, F.G.S., and many other members of the Cardiff Naturalists’ Society. 54 GLACIAL ACTION IN 8. BRECKNOCKSHIRE AND E, GLAMORGANSHIRE. Discussion. The Cuarruan (Dr. J. Gwyn Jeffreys) said that the district was well known to him, and that he had examined parts of Glamorgan- shire with reference to the present question, and he believed that there was evidence there of land-glaciation only, the remains of moraines being occasionally discernible—differing in this respect from the northern counties of Wales and England, which give unmis- takable indications of marine action, and sometimes at a present height of over 1300 feet. In the southern counties of England there was a similar absence of so-called glacial shells, and no evidence of elevation. He invited discussion on this interesting communication. Prof. Prestwicn expressed his regret at the absence of the author, and his sense of the value of the paper. It formed an excellent sup- plement to Mr. Mellard Reade’s important paper, to which the Society had already listened. Thesouth of Wales had hitherto received but little notice as compared with North Wales. One distinction to be noticed in this district was that the erratics were local, while in North Wales erratics were present from districts far to the north. It was singular that south of the Bristol Channel the indications of ice-action should be so obscure and uncertain, when they were so clear to the north of it. ON A NEW DINOSAUR FROM THE ISLE OF WIGHT. 55 5. On the Dorsat Rueton of the VERTEBRAL CoLumN of a@ new Drvosaur (indicating a new Genus, SPHENOSPONDYLUS), from the Weatpen of Brook wm the Iste or Wieut, preserved wm the WoopwarvdiaAn Mouszvum of the University of Cambriper. By Prof. H. G. Szetzy, F.R.S., F.G.S., &c., Professor of Geography in King’s College, London. (Read June 21, 1882.) Tuts small series of six vertebral bones is remarkable for the great lateral compression of the centrum and the depressed form of the neural arch, and, as exhibiting the characters of the dorsal region in a new generic type, seems to me worthy of some notice. Fig. 1— Dorsal Vertebra of Sphenospondylus, right side of type specumen in Woodwardian Museum. (One half nat. size.) AK, Ca WAN QQ . BN QQ a Adnness * REG, Le WS (Ad \ 5 AY Ss tye LL i a UT) 6. Facet for head of rib. tp. Transverse process. The centrums of the vertebre have an average length of 9 cen- tim. each; the transverse width of the articular faces is about 7 cen- tim., with a vertical depth of at least 8 centim. ; but the lower part’ of the centrum is much compressed from side to side, so as to have a wedge form, and terminates inferiorly in a sharp longitudinal ridge. The articular margins of the centrums are moderately elevated. The transverse processes of the neural arch are at first directed back- 56 - PROF. H. G. SEELEY ON A NEW ward, but soon become directed transversely outward, and retain their upward direction. The facet for the head of the rib is at first large, and at the base of the transverse process, and bounded poste- riorly by the sharp ridge which runs below the transverse process to the hinder margin of the neural arch; but after a time the rib- head rises higher, so as to be chiefly above the zygapophysial facet ; and then it becomes smaller, the ridge behind it more or less dis- appears or rounds away, and the transverse process, which was at first triangular in section, becomes vertically compresed and thin. The interest of the series is in exhibiting the gradation of cha- racters as the bones extend backward, though, as the surfaces are invested with a thin argillaceous layer, and the bones have lost the neural spines, there is still something to be desired in their con- dition. In the lst vertebra the anterior face of the centrum is flattened, but somewhat concave, except towards the margin, which is convex. The basal outline of the face is rounded, the sides are subparallel, and the superior corners are rounded, while the neural canal is concave. The median vertical measurement is 7°4 centim., the trans- verse measurement nearly 7 centim. ‘The measurements of the pos- terior surface are slightly less; but the articular face is more con- cave, and its convex outer border is narrower. The sides of the centrum are concave from front to back, and moderately convex from above downward, becoming closely approxi- mate towards the base, so as to form a blunt basal ridge or keel, which is most developed towards the anterior end: it is concave in length. The transverse measurement in the middle of the centrum below the neural arch, is 5°3 centim. The neural arch encloses a rather small neural canal, winel is not higher than wide. The zygapophysial facets are inclined at a right angle, are 27 centim. broad, and parted below by an interspace about 1 centim. wide. The external surfaces of the anterior zyga- pophyses are oblique, extending, with a slight concavity, back to the elevated border of the facet for the head of the rib, and extend- ing concavely downward to form the pedicle, which joins the cen- trum by an imperceptible suture. The facet for the head of the rib is large, vertically oval, less than 4 centim. deep, and 34 centim. wide; itis deeply concave, in advance of the middle of the side, 8 millim. below the anterior border of the transverse process, which extends behind it, and above its posterior half. These facets are not quite vertical, the transverse measure- ment over their upper parts being more thau the measurement at the base. Below the base of this rib-facet there is a slight con- yexity on the line of the neurocentral suture. The transverse processes are strong and directed upward and outward, almost at the same angle as the zygapophysial facets, but they are also directed a little backward. The superior surface is flattened, smooth, looks inward and upward, 7 centim. wide at the base, with the margins compressed and gently concave in length, though the concavity 1s deeper on the posterior side, where it DINOSAUR FROM THE ISLE OF WIGHT. 57 terminates above -the posterior zygapophyses. At the terminal facet for the tubercle of the rib the width of the process is about 3 centim. Inferiorly the transverse process is compressed into a strong ridge, the base of which is rounded; this ridge de- scends posteriorly behind the facet, for the head of the rib to form the anterior margin of the canal for the intervertebral nerve, while at the free end of the process it expands to assist in forming the transversely ovate tubercular articular facet. The anterior and posterior inferior surfaces of the process are concave channels, which deepen as they descend, the posterior enlarging into a con- siderable excavation between the vertical buttress and the posterior zygapophysis. The transverse width over the transverse processes, as preserved, 1s about 17 or 18 centim. The base of the neural spine is about 9 centim. long. It is broken away, but was compressed, widening posteriorly, where it is about 1 centim. thick at the fracture. It there descends, widen- ing in an A shape, with the lateral part concave; and the flattened posterior surface extends backward, to terminate inferiorly in the ovate posterior zygapophysial facets, which are divided behind by a slight concavity, which becomes narrower as it extends between them inferiorly. The transverse width over the two facets is 5 centim. As is usual in dorsal vertebrae, the superior border of the centrum is a little longer than the inferior border, indicating a slight arching of the back. The 2nd vertebra differs chiefly in having the anterior articular face flatter, and in having a more elevated ridge margining the anterior border of the articulation for the head of the rib, while the ridge below the transverse process is directed less backwards, and more obviously curved. In the 4th the inferior ridge of the transverse process is nearly transverse, and more obviously forms the posterior border for the articular facet for the rib. In the 5th this ridge is inclined backward, and the posterior border of the neural ets is rounded. The 6th and last vertebra of the series is a little crushed, so that the inferior basal ridge appears to be either obliterated or less deve- loped. The base of the neural arch below the transverse process has now become a somewhat compressed area, rounding anteriorly into the pedicle of the neural arch, slightly concave in the middle, and margined posteriorly by the broad flattened rounded ridge descending from the transverse process. Behind this ridge is a deep impression like a thumb-mark on plastic substance. The tubercle for the rib has now ascended so as to be quite on a level with, or rather above the transverse process, and immediately in front of it; and it has become smaller. It is wider than deep; and the measurements are ‘between two and three centim. Its superior surface is convex ; and the convexity extends inward towards the neural spine. The greater part of it rises above the level of the zygapophysial facet, 58 PROF. H. G. SEELEY ON A NEW external to which it projects laterally more than a centimetre. The facet is still concave, with a central pit. (Fox Fig. 2.—Sphenospondylus, left side of a Dorsal Vertebra. Collection, Brit. Mus. = One half nat. size. i 1 1 1 \ / ry a i 1 i] 1 t 1 i t ! Ms Pan § ' é Mh i 2 i 4 1 t . i i i i ! { ! i ' ’ ? ' b) it a i] ee ‘ \ ~© \ 6. Facet for head of rib. tp, Tranverse process. z. Postzygapophysis. | DINGSAUR FROM THE ISLE OF WIGHT. 59 The anterior zygapophyses do not reach so far forward as to be quite level with the anterior face of the centrum. I have long been seeking for specimens with which this series might be compared; and although the old collection of the British Museum contains but one centrum of this type, Mr. Davies, of the British Museum, on seeing these vertebre, at once drew my atten- tion to a series of more than a dozen in the Fox Collection, as which belong to the same species (fig. 2). I am inclined to think that many of them may have been part of the same individual with the Fig. 3.—Posterior Aspect of Dorsal Vertebre of Sphenospondylus. (Fox Collection, Brit. Mus. =) One half nat. size. 60 PROF. H. G. SEELEY ON A NEW Cambridge bones. The length of the centrum is the same, and its wedge-like character identical. The surface behind the transverse processes, which is not clearly seen in the Cambridge bones, is found to form a wedge with lateral excavations, which extend forward under the posterior expansions of the transverse processes. In several vertebre there are slight pits at the base of the anterior margin of the neural spine, and well above the large deep depres- sion bordered by the zygapophysial facets, beween which the sharp anterior margin of the neural spine is prolonged. The notch be- tween the zygapophyses only extends for their anterior third. A few vertebre have the neural spine preserved, but it is not perfect. In one (fig. 2) itis 164 centim. high, 9 centim. wide at the base ; it contracts a little in the middle, chiefly by concavity of the posterior border, widening a little above, so as to make both margins concave. All the British-Museum specimens show the facet for the head of the rib, though it varies much in character. At first it is small (6, fig. 2, p. 58) vertical, and low in position. Gradually ascending, it widens and becomes more circular. And when the transverse processes become horizontal, the superior and anterior margins be- come greatly elevated. In the hinder part of the series the facet becomes small and transversely oval. I have seen neither cervical nor caudal vertebrae which present this type. The Cambridge specimens were obtained by Mr. Keeping in 1866, and brought to the Museum with the coracoid lately described ; but he has no recollection of their having been asso- ciated with that bone. But in view of the hkelihood of these ver- tebree pertaining to the ‘“ Jguanodon Seelyi,” I have refrained for the present from giving a specific name to the specimens. The characters on which I establish the genus Sphenospondylus for these remains are the laterally compressed form of the base of the dorsal centrum, the depressed form and character of the neural arch, the upward inclination of the transverse processes, and the condition of the facet for the head of the rib in rising so as to be placed be- tween the transverse process and the anterior zygapophysis. ‘These characters clearly differentiate it from [yuanodon, which is the only genus with which it can be compared, supposing ‘that we take Iguanodon Mantelli as the type (figs. 4 & 5). But the genus has since been enlarged to include such types as Iguanodon Prestwiche and J. Seelyi, both of which differ from the type in well-marked and varied characters. What the significance of those differences is may, I fancy, be determined by comparing together existing genera of animals, and noticing the nature of the characters by which they differ. Judged in this way, I think it possible that both these species might be referred to new genera; and from such a point of view I conceive of these vertebre as indicating a new genus. But if we take the older conception of a genus, which is anatomical and not zoological, and more a matter of paleontological convenience than a step in evolutionary history, we may rank all these forms under the one name Jguanodon. It is a matter on which there is DINOSAUR FROM THE ISLE OF WIGHT. 61 Figs. 4 & 5.—Posterior and left lateral view of Dorsal Vertebra of Iguanodon Mantelli. (From the specimen figured by Prof. Owen, in the ‘ Wealden Reptiles,’ suppl. il. pl. vil. figs. 4 & 5.) One eighth nat. size. Fig. 4. Fig. 5. Se eo o|-= Se ag Ny Fig. 4. Posterior view, for comparison with fig. 3. Fig. 5. Lateral view, for comparison with figs. 1 & 2. likely to be difference of opinion for some time to come. And if I lean towards defining genera so as to make them small, it is because I believe that characters are in this way better appreciated, and because the groups of fossils become better comparable with existing natural- history types. I have compared the vertebre here described with those of Iguanodon Prestwich and other available named Iguanodons, and believe that the character of the neural arch justifies my distinction, though the types have some characters in common. Discussion. The Presrpent said that he had been long familiar with vertebre of this type, and had had a strong suspicion that they would prove to belong to Iguanodon. 62 : REV. A. IRVING ON THE 6. On the Mxcrantes of Guacrurs, with especial Reference to ther supposed PowsER of Excavation. By Rey. A. Irvine, B.A., B.Sc., F.G.8. (Read December 6, 1882.) Opinions as to the kind and extent of work done by glaciers upon the rocks over which they move seem still to be very divided in the geological world. On the one hand it is still maintained that these agents are capable of excavating basin-like hollows, such as those which are now (or have been) filled with lakes; on the other hand we find such high authorities as Profs. Bonney * and Credner rejecting the hypothesis of excavation, while they fully recognize (as every Alpine observer must do) the scouring, grooving, striating and polishing work done by glaciers upon the floors and sides of valleys previously formed by ordinary valley-erosion, as well as their indirect action in contributing to the formation of lakes by the dams which their moraines form (e.g. at the southern end of Lake Garda) in some cases across valleys. Amid this diversity of opinion I may be pardoned for attempting to add something to the discussion of this interesting subject. The whole discussion would seem to narrow itself, theoretically, to the answer to be given to the question, Can a glacier dig or ea- cavate basin-like hollows? Those who answer in the affirmative seem in their arguments to assume that the ice of the glacier moves as a rigid mass. If it did so, its scooping-out power would be enormous ; but that it does not has been shown by Prof. Tyndall, in his little work ‘ Forms of Water’ and elsewhere, and demonstrated experimentally by himself and Prof. Helmholtz ¢ of Berlin. The writings also of Forbes and others on this subject are no doubt familiar to geologists. The snow of the upper névé becomes gradually transformed into the solid ice of the glacier in two ways :—(1) By pressure from above the crystalline particles are partly melted, the liquefied portions finding their way between those which still remain solid; (2) the heat of the sun melts the surface-particles, the water at 0°C. thus formed trickling into the snow. In both cases the water is again transformed into ice, its latent heat being taken up by the snow, which at these high altitudes is at temperatures below 0°C. Rege- lation occurs, as it does behind the shearing-wire in a well-known experiment. Liquefaction by pressure of portions of the ice-mass, and regelation at points where the pressure is relieved, are not con- fined to the névé; it goes on continuously throughout the mass of the glacier, though more in some parts than in others, and goes a long way to account for the “ plasticity ” of the ice-mass—its power, that 1s to say, of adapting itself to the form of the trough or hollow in which it hes. * See Quart. Journ. Geol. Soc. vol. xxx. p. 479. ; tT ‘ Elemente der Geologie,’ p. 245. t Vide Lecture, ‘ His und Gletscher.’ MECHANICS OF GLACIERS. 63 Prof. J. Thomson has deduced from the mechanical theory of heat, and Sir W. Thomson has verified by experiment, the law that the freezing-point of water is lowered by pressure; and Helmholtz * has shown how it follows, as a corollary to this important law, that the temperature of ice is lowered when it is subjected to pressure within a confined space, the ratio of the liquid water to ice being at the same time increased. The thermal energy which is generated by pressure becomes latent in the newly liquefied ice, and so is not available to affect the temperature of the mass. This liquefaction must take place most in the lower layers of the glacier; and owing to the great latent heat of water (=nearly 80 thermal units at a pressure of one atmosphere) the pressure, though great, melts only a small proportion of the ice-mass, which is very large. Helmholtz? has pointed out the bearing of this principle upon glacier-work. The ice being pressed, and a small portion of it melted, the water is free to escape. The temperature of the pressed ice is lowered, but not that of the water, which, being free to escape, does not suffer any lowering of temperature. ‘‘So we have, under these circumstances, ice colder than 0°C. in contact with water at 0°C. The consequence of this will be, that water is continually frozen around the pressed ice and forms new ice, whilea portion of the pressed ice is melted.” Owing partly to imperfect homogeneity of the ice of a glacier, partly to the inequalities of its bed, pressure acts more upon some points and in some directions than in others. Severalresults may follow. (1) Ifthe pressure is applied continuously and rapidly enough, and the tempera- ture of the ice is below that required for liquefaction under the given pressure, the ice cracks; work is done in overcoming cohesion. (2) Some ice is melted, mechanical force is transformed into heat, which becomes latent in the melted ice, the water is squeezed out and regelates in contact with the colder ice, its latent heat being given up to the colder ice in contact with it, raising the temperature of this ice, until it and the regelated film have acquired again a uniform temperature. (3) Friction follows, if, as in the glacier, the force continues to be applied, by the faces of the cracks sliding over one another. Heat is generated; portions of the ice surfaces are lique-’ ‘fied, the water trickling out as before and becoming regelated in contact with colder ice, the thermal energy given up by it in the act of regelation being diffused by slow conduction, as before. (4) As the heat given up by the water in the act of regelation to the contiguous ice is transmitted by slow conduction through the neighbouring ice, it causes expansion, or a tendency to expand, which can only be prevented by increase of resistance. (5) If ex- pansion occurs, as in (4), or water trickles out, as in (2) and (38), its tendency will be, under the influence of gravitation, rather down- wards than in any other direction; and so the centre of gravity of the whole mass is lowered, though the mass may not be moving as . a rigid body. This I conceive to be the outline of the history of the expenditure of that portion of the potential energy of the weight of * © Kis und Gletscher.’ t Loe. cit. 64 REY. A. IRVING ON THE a glacier-mass which is used up within the glacier. Now for the bearing of this upon the question of excavation. Exactly so far as the glacier-mass possesses this yielding property due to the transformation within it of the mechanical energy due to its weight, is ats digging- or excavating-power diminished. Further, whatever theory we adopt to account for the yielding property of ice (its ‘‘ plasticity ” or ‘ Nachgiebigkeit”), it is plain that the forward and downward thrust is not wholly expended in propelling the glacier mass as a rigid whole; hence the absence of any traces of excavating-action where glaciers have receded in recent years. As it is, the forward thrust is to a very large extent resolved into an indefinite number of smaller forces, which are expended, either directly or indirectly (if first transformed into thermal energy), in overcoming cohesion. It follows at once from this, as a simple de- duction from the law of the conservation of energy, that the residuum of energy available for any swpposcd excavating action of a glacier is comparatively small. And this deduction would seem equally sound whether (following Tyndall and Helmholtz, as I have done above) we adopt the regelation-theory, or the ‘‘ viscous theory” which was propounded by Forbes, to explain the ‘flow ” of the glacier. The essential point is, that the greater part of the forward thrust of the glacier mass is expended in overcoming cohesion and in causing movements among the parts of a glacier relatively to one another. Such relative movements of the parts of a glacier are, since the numerous observations of Tyndall and his fellow glacialists, too well known to need further description here; the relative rates (1) of the middle and the sides, (2) of the top and bottom, having been made matters of exact measurement *. Tyndall and Helmholtz have both also given experimental demonstrations of them. The above reasoning applies of course to such portions of the glacier as form a continuous whole. There is yet another way in which some part of the potential energy due to weight is expended ; that is, in the formation of crevasses. Ice is not viscous, and there- fore does not preserve its continuity under the influence of tensile strain. So small is its power to resist tensile force, that the slight bending of its mass which is caused (according to Helmholtz) by an: increase of gradient in its bed of from 2° to 4° is enough to form transverse crevasses in its upper surface. Such crevasses penetrate further into the ice in proportion as the increase of gradient is greater. Here then is an expenditure of a portion of the weight of the ice-mass immediately below each crevasse, which is quite un- available for purposes of erosion. Again, in the formation of the well-known Bergschrund, the ice below it having torn itself away from the névé above it, the weight of the latter is no longer capable of cooperating with the weight of the ice below it. Marginal cre- vasses result also from the same absence of ductility in ice. The movement forwards and downwards of the central parts of a glacier being greater than that of the lateral parts, which are retarded by friction against the sides of the valley, a strain and tear result ; * Vide ‘ Forms of Water,’ by Prof. J. Tyndall. MECHANICS OF GLACIERS. 69 the ice parts asunder in a number of planes, forming an equal num- ber of crevasses, which proceed with gradually diminishing width from the extreme lateral limits of the glacier towards the centre— not, however, in a strictly transverse direction, but in that of the tangent to the direction of the strain, tending upwards therefore to- wards the source of the glacier. Finally longitudinal crevasses are formed when a glacier has to force itself through a narrow gorge. As it emerges from the gorge, the central portions move on faster than the lateral portions, which are retarded by the sides ; and that portion of energy (even here where*the action against the rocky sides is at a maximum) which is expended in parting the middle portions from, and producing friction of them within the gorge against, the lateral portions, cannot be expended at the same time upon the work of erosion. Generally, we may say that the origin of all these varieties of crevasse is the same property of glacier-ice which makes it unable to yield to tensile force; and the consequence is, in each case, a breaking-up, more or less, of the glacier-mass, and the consequent distribution of its force as a moving body. The whole weight of any given mass of the glacier may be re- solved into two forces, the one acting parallel, the other at right angles, to the inclined plane on which the glacier lies*. The former, which will vary with the sine of the angle of inclination, and will therefore be nil when the glacier rests on a horizontal bed, is, as has been shown, partly used up within the glacier; and the portion thus used up, whatever it may be, is not exerted against the rocky floor, and therefore can do no work in the way of erosion. The ice moves on this floor, if it be inclined at a sufficient angle; but it moves with less velocity at its bottom than the centre of gravity moves. * The relation which subsists between the angle of inclination of the slope on which a given mass of a glacier lies, and the pressure and shoving force due to the weight of the given mass, will be made clearer by the following simple mathematical reasoning. Suppose a given g glacial mass to lie on a slope, repre- sented in the accompanying diagram by a line AA’, which makes an angle @ with the hori- zontal ; and let us suppose the whole weight of the given mass to be represented by one resul- tant force W, acting vertically through its centre of gravity. upon the point P, as indicated by the arrow. By a simple “triangle of forces” it is easy to see what parts of the weight W are represented by the forces acting (1) as pressure in the direction of the normal Pg upon the surface AA’ at P, (2) as a shoving force parallelto AA’. For the first we have Wiad = W xcos8; for the second, Wx =Wxsin 0. Pr Ons, G.SoNo. (53: F 66 REY. A. IRVING ON THE So far as we have now proceeded in the argument, this difference of velocities would appear to represent the work done within the glacier; and we might reason in a similar way as to the difference of velocities of the median and marginal portions. In the above reasoning I have assumed only the action of a part of the shoving force due to gravity in making the glacier slide upon its bed. As a matter of fact, other causes may promote sliding by diminishing friction, such as (1) the non-contact of the ice with the rocks in places where the glacier streams are flowing, (2) the ther- mal effect of the heat flowing: from the earth’s crust by conduction from below, (3) comparatively warm water rising in places from deep-seated springs, (4) the partial liquefaction of the ice by pres- sure against the rocks. All these, by diminishing friction, dispense with some portion of the shoving force due to gravity; so that the proportion of that force used up within the mass of the glacier is even much greater than the estimate from the difference of velocities alone would give. So far as any supposed excavating action is concerned which could form rock-basins, the differential movement of the upper portion of the glacier, as compared with its base, is the most im- portant point. Measurements taken by Prof. Tyndall in the case of the Glacier du Géant, at the foot of the Tacul, showed a move- ment of the portions near the surface more than double that of the base. A fortiord, this differential movement must be greater, owing to the greater retardation of the base of the glacier, when on a horizontal bed—so much so, that the greater pressure acting at right angles to that bed (which varies ceteris paribus with the cosine of the angle of inclination of the bed) would seem to avail nothing, since the movement of the base of a glacier lying upon a horizontal bed would be nl. The only propelling force to which it could be subjected would be the shoving force acting against it from the weight of the glacier lying upon an inclined slope immediately above. Prof. Tyndall* has shown us how this would act. When the glacier passes from a steeper to a less steep gradient, the crevasses close up, the yielding-property of the ice comes into play, the ice at the surface is thrown into a series of transverse terraces or huge wrinkles, the differential motion is increased so much that stones of the medial moraine, which have fallen into the crevasses, are brought again to the surface. From all which it would appear that the movement of the base of the glacier upon a horizontal bed is nal; and therefore here, where a theory of excavation most requires it, its erosive action is almost nzl. This reasoning seems further confirmed by observations made on the Morteratsch ‘7. Some distance up the glacier the movement, at its maximum, was found to be 14 inches per diem; yet at the snout, which lies on a nearly horizontal bed, even without any ice in front to offer any resistance to its motion, the movement forward was only 2 inches ina day. It is no reply to this argument to say that, higher up, the erosive power must be greater. The ordinary law of valley-contour, * Forms of Water, p. 18U. t Ibid. pp. 96, 97. MECHANICS OF GLACIERS. 67 the steepness of the valley as a rule increasing as we approach its head, is well known; and it follows from this that the biting-power of the glacier upon the rocks (which diminishes with the steepness in proportion to the cosine of the angle of inclination) is less as we ascend into the steeper slopes of the glacier-region. Moreover it is self-evident that it is not on such steeper parts of the valley that the advocates of the excavation-theory would call its action into requisition. In the work of erosion, such as is imputed to glaciers, it is of course well known that the work is mainly done by the stones and sand which have found their way, either through crevasses or between the glacier and the rocky sides of its channel, to the base. These form the teeth of the file. We must recollect however (1) that these and the rocks upon which they act are of about equal hardness ; only at most, therefore, one half of the finer detritus which comes away in the glacier-stream is produced by the grinding away by these stones of the rocks themselves, the wearing of the stones furnishing an equal amount of it ; (2) that the stones held in the ice are only passive instruments, and can only do work upon the rocks when they movye—that is, when the resistance to their motion against the rock is less than the yielding-power of the ice which holds them. Moreover, much of the detritus which comes away in the glacier-stream may well be derived directly from the finer portions of the moraines which have fallen to the bed of the glacier. It has been suggested that the freezing of water within the crevices and pores of the rocky bed of the glacier must by its ex- pansion break up the rock-surface, and thus furnish detritus for the glacier to carry away, as the loosened materials are caught up by the ice. This primd facie seems a sound argument in favour of excavation ; we must therefore examine it. We must recollect (1) that the water contained in this way within the rock is exposed to subterranean heat passing up by conduction from below, and that, if this is slow, owing to the low conductivity of the rock-materials, the cooling effect of the ice of the glacier is, @ fortiorz, equally slow. (2) The actual surface of the rock at any given point is in contact with either (a) the water of the glacier-stream, which is not below 0°C., and therefore cannot freeze the water within the rock, or (6) in contact with the ice (or a stone stuck in the ice and at the same temperature as the ice), in which case the ice may be either at or below 0°C., according to the pressure at the point of contact, as Helmholtz’s reasoning shows. Ice at 0°C. has no power to freeze water at 0°C., since with equality of temperature there can be no exchange of heat between the bodies; and if the ice be below 0° C.. it can only be so at a pressure proportionately greater. This very pressure must be exerted upon the rock, and so counteract the ex- pansive force of the water within the rock. The hypothesis is thus shown to be wholly inadmissible. Further, the actual appearance of glaciated rocks shows that they have not been thus broken up by freezing water while the glacier covered them. It is conceivable that some of the surface-portions of the glacier F2 68 REV. A. IRVING ON THE which are melted by the sun’s heat or by a warm current of air may, in penetrating the glacier, come into contact with ice which is locally colder than 0° C., and in this way undergo refrigeration. If this took place anywhere around a surface of contact of the ice and the rocky bed, the surface of contact being increased (as Helmholtz has shown in the valuable notes which he has appended to his lecture, ‘Kis und Gletscher’), the pressure on the previous plane or point of contact would be distributed over a larger space, with the thermal effect which can easily be deduced from the foregoing principles. In this way as the glacier-mass is diminished at the surface, a partially compensating formation of ice may go on at the base. So far as it acted, there would be a transfer of materials to some extent away from the bed. A little further reflection will show us, however, that this could only happen where the water was unconfined, and consequently only where the glacier was moving down a slope. In such a case the result would seem to be rather adverse to erosion. When however the glacier lies on a level or hollow surface, such a process must soon bring itself to anend; for, the glacier being motionless at the base, all spaces would soon be filled with ice and cold water kept there by gravitation, and the whole pressure would bear directly upon the rock, tending, not to break it up, but to compress it. A word or two is needed on the influence of terrestrial heat. The facts connected with the flow of heat by slow conduction from the interior to the exterior of the earth’s mass can be learnt from any good text-book of physics. The point with which we are here concerned is this :—Since the rock in contact with the ice remains at the same temperature as the ice, it can only doso by parting with its heat to the glacier as it receives it from below. This heat must do work. What is that work? Clearly the heat must be expended in overcoming the cohesion of ice-particles in contact with the rock. And this it would do whether the ice were at or below 0° C., since in the latter case the heat received from the crust of the earth cooperates with the pressure which is the necessary condition of a temperature below 0°C. The bearing of. this fact as tending to diminish friction, and therefore erosion, has been pointed out above*. In connexion with glaciers there is, in addition to the polishing, striating, and grooving work, observable everywhere on glaciated rocks, a still more extensive work of erosion going on, by the action of the glacier-streams which flow beneath them. Such streams, while they intervene in places between the ice-mass and the trough in which the glacier moves, are capable of doing much more work than the ice itself, by virtue of the greater velocity with which they carry stones and sand along. Since, however, this action depends entirely upon the movement of the water, it is clear that a descent is necessary for this movement; and as water does not flow up-hill, * The conduction downwards of absorbed solar heat from the sides of the valley to below the surface of the glacier produces in the summer a want of contact, for some feet down, between the ice and rock—a fact familiar to every observer of glaciers, MECHANICS OF GLACIERS. 09 each glacier-stream must maintain an open channel for itself; for this reason the same objection applies to it as to a river flowing in an open valley, as an agent of excavation of basin-like hollows. Here and there, where we see the convergent surface-streams of the glacier rushing down the shaft of a moulin, and carrying from time to time earth and stones from the surface-moraines, a certain slight work of excavation is no doubt accomplished, such as we see now exposed to the light of day in the well-known “glacier-garden” at Lucerne; but this would hardly meet the requirements needed for the excavation of lake-basins. ‘The main glacier-stream, just like any other stream, acts of course as an erosive agent and deepens the trough in which the glacier moves, an excellent ex- ample of which I observed only last summer at the end of the Hochjoch Glacier, at the head of the Rofen Thal, where the ice forms an arch resting upon the sides of the gorge as its buttresses. Again it is undeniable that an advancing glacier may do a certain amount of “ploughing” work, such as Prof. Tyndall has described im connexion with the Gorner Glacier *. This, however, is only evidence of the inability of the glacier-ice to move over the super- ficial obstacles which it encounters in its path; it would seem to tell rather against the notion of the ice being driven up-hill out of a basin, as is assumed by some writers ; and it would be interesting to inquire whether such phenomena have ever been observed where a glacier was not descending a slope. All that has been put forward with reference to the distribution of the potential energy due to gravitation in the procession of the glacier down a valley must tell equally against the propulsion of it up-hill. On the other hand, we may, I think, draw a distinction between the “‘ice plough” and what I may be allowed to call the “ ice chisel.” In cases where successive portions of a glacier descend a vertical, or nearly vertical, precipice, a different set of mechanical conditions is presented to us. Impact may in this case perhaps do the work of excavation to such an extent as is represented in the excavation of many small reck-basins, such as some of those which lie upon the Bernina Pass, or at the foot of the precipices of Snowdon. But when all this is admitted we have no right to reason from “ the hundreds of tarns that are found in all glaciated mountain-countries ” to the formation of lake-basins which form long depressions in narrow valleys. Many tarns, however, do not occupy rock-basins at all; some are formed by moraine-heaps left by the retreated glaciers, between which and the mountain-side they may be frequently observed to lie in the Alps; other tarns simply fill depressions formed by earth-movements on the mountain-slope, where the clayey materials produced by the disintegration of horn- blendic, augitic, and felspathic débris get loosened by water and move unequally downwards, as clay often does on a smaller scale in a railway-cutting. Numerous instances of such could be pointed to in the Alps and in other mountain regions. It is not without regret personally that I find myself driven to * Forms of Water. 70 | REV. A. IRVING ON THE conclusions adverse to the theory of excavation which has been advocated for some twenty years by Sir A. Ramsay, to wines geological writings we are all so greatly deus oer General Conclusions. It will be seen from what has been advanced that I do not question’ the power of glaciers to do a good deal of surface-erosion in grinding, polishing, grooving, and striating the rocks; no one who has seen any thing of glaciers or of glaciation could do this for a moment. My contention, from the consideration of mechanical and physical principles, is that far too much work has been ascribed to them by some writers in the way of erosion, and that the notion of actual excavation of lake-basins is inadmissible, except under very special circumstances such as those under which some tarns may have been formed. The causes of the differential movement of glaciers would appear to be three :— (1) Cracking and partial melting in places under pressure and strain, followed by regelation, as propounded by Tyndall and accepted (after independent experimental investigation of the phenomena) by Helmholtz. This is probably the principal regular cause. (2) Friction generating thermal energy, and so producing lique- faction, which is followed by regelation. These two causes, it will be seen, are in constant operation, and @ fortiora must have operated still more powerfully when the glaciers _ were of much greater dimensions. (3) There remains to be accounted for a secondary differential motion, which has, it appears, not yet received a satisfactory expla- nation, though some recent writers have attempted it *: the move- ment is greater (a) by day than by night, (6) in summer than in winter. ‘This was very nearly explained some years ago by Canon Moseley, when he maintained that somehow or other ‘radiant heat’ must enter the ice. Had he known those principles of physics which are illustrated by the action of Crookes’s radiometer, there is little doubt that he would have seen his way to the right explanation. The theory which I venture here to advance is based upon a series of experiments with ice subjected to different sources of radiant energy, in which I have been engaged, an account of which I hope to publish elsewherey. For the present purpose we must consider heat to mean energy capable of melting ice or tending to melt it. Whatever notion we may attach to the term ‘‘radiant heat,” it is clear to me from my experiments that heat, qua heat, cannot enter the ice and be afterwards expended in the work of liquefaction—that is to say, in overcoming cohesion and so promoting differential movements of parts of the glacier. All such heat must become latent in the liquefaction of ice at the sur- face. It is in the transformation of energy that the clue is to be * See Croll, ‘Climate and Time,’ chap. xxxi. + See Nature, No, 698. MECHANICS OF GLACIERS. Th found. To say that ice is transparent is to say that luminous radiant energy can freely traverse it. It does not do so however equally at all times. Obviously, more luminous energy enters the glacier from the sun by day than by night—more during the more numerous hours of daylight, the higher ascension of the sun, and the greater freedom from the diffusive action of the snow at the sur- face in summer than in winter. A certain amount of the luminous portion of a beam of solar radiation is absorbed by even clean ice, more especially by granular ice, otherwise the beautiful silvery blue colour which is perceived in the ice overhead when one enters an ice-cavern would be wanting. But it is chiefly by opaque and semi- opaque bodies within the glacier (stones, earth, organic germs, &c.) that the luminous solar radiation which enters the ice is absorbed ; and the radiation thus absorbed is, in accordance with the law of conservation of energy, converted into heat, just in the same way asit is in an ordinary greenhouse. Heat thus developed within the mass of the glacier, during the hours of daylight, and most so during the summer, must promote the differential movement of the glacier. Of course, if the glacier is clean enough and thin enough, some of the luminous energy may penetrate to the rocks beneath ; but the transformation of energy would in that case be the same, the heat developed promoting the descent of the glacier as a whole, instead of its differential motion. Discussion. Mr. Cattarp stated that in 1878 he had had the opportunity of studying the base of the Rhone glacier, which glacier has been slowly receding for a length of time. Beneath the glacier, as seen in the terminal ice-cave, the ground was not ploughed but only some- what smoothed. Prof. SzELEy pointed out that before we could accept the author’s views it would be necessary for him to explain the origin of the quantity of detritus carried out: from the end of the glacier and the smoothed surfaces on the bottom and sides of the glacier-bed. He thought that the effect of the continual melting and freezing of water at the bottom of the glacier on the rock masses below had been overlooked by the author, and that these effects must be very striking. The late Mr. Clifton Ward had shown that some lake- basins occur at the point where tributary glaciers join the principal ones. Mr. Branrorp referred to the @ priort argument, based on the fact that rock-bound lake-basins abound in districts which have been glaciated, and are rare in other regions. He pointed out that the author seemed to have lost sight of the fact that the erosion was performed not by ice but by stones &c. held in the ice. Lieut.-Col. H. H. Gopwry-Avsren said that the great glaciers of the Himalayas, which are so much larger than those of the Alps, are advancing, breaking away the ground along their sides, and pushing forward their older moraines. It must also be remem- bered that glaciers of the great thickness of the older glaciers in Switzerland must have produced very different effects on rocks of 72 REY. A. IRVING ON THE different degrees of hardness, and that thus rock-basins might have been produced. Mr. Watrer Browne agreed with previous speakers that the fact that the rivers flowing from the ends of glaciers contain so much mud is proof positive that they erode their beds. He had seen the Gorner glacier pushing the turf of fields before it like a plough. He called attention to considerations laid before the Royal Society in his recent paper, especially that the Greenland glaciers move, in the depth of winter, at temperatures far below 0° C.; and he thought this fatal to the movement of glaciers being due to rege- lation. According to the author’s theory, glaciers would cease to move when they reach a horizontal surface. Dr. Hicxs thought that the author had proved his case, which was, that ice, by itself, cannot erode rock-surfaces. Dr. Woopwarp referred to the observations of Dr. Hector, in New Zealand, and said that no glaciers are absolutely clean, but all contain rock fragments, consequently all glaciers are capable of erosive action. Mr. Baverman objected to the author’s comparison of ice and glass. Ice is perfectly crystalline, as shown when examined by polarized light, the principal axis of the crystals being perpendicular to the planes of cooling. Apart from this point, he was inclined to agree with the conclusions of the author. He did not think there were any experiments which could be quoted in support of the view that glaciers can erode rock-basins. He thought that what is usually called glacier-erosion would be found to be due to the erosion of water in confined channels beneath the glacier. The Rev. E. Hitt said that the argument that the thrust was mainly expended in overcoming cohesion rather told against the author’s views than supported them, as the friction of the sides of the glacier is necessary to produce a resistance to the thrust. The Rev. E. 8. Dnwick agreed with Mr. Bauerman’s views on the subject. The AvtHor said that, in the remarks objected to by Mr. Hill, he had been merely arguing against the view that the glacier moves as a solid mass. He fully recognized the “ filing” action of the stones held in the ice; but since these stones and the rocks they grind against are of equal hardness, only at most one half of the pulverized material brought away by the glacier-stream comes from the rocks over which the glacier moves. He agreed with Mr. Bauerman as to the work done by streams beneath the glacier, and had given an illustration of it in the paper. He argued, not against the erosive power of glaciers, but against their power of cutting rock-basins such as those occupied by lakes. He had already met many of the objec- tions raised to his views in a paper to be read in a fortnight’s time. He had pointed out one cause of the motion of glaciers which is quite independent of climatal conditions, namely the conduction of heat to the glacier from below. The surface-conditions in a region of excessive radiation like Siberia, which Mr. Browne had referred to, were not comparable with the conditions found beneath the glaciers in Greenland. ORIGIN OF VALLEY-LAKES. Vo 7. On the Oricin of VaLLEY-LAKEs, with especial reference to the Laxzs of the Norraern Aups. By the Rev. A. Irvine, B.A., B.Se., F.G.S. (Read, December 20, 1882.) THis paper is intended as nothing more than a supplement to a paper by the author on the Mechanics of Glaciers. The main strength of the position taken’ up by the advocates of the glacier- excavation theory is derived from arguments which labour to show that it is the only feasible hypethesis*. I have endeavoured to show in my previous paper that this hypothesis is inadmissible on mechanical and physical grounds; all I attempt to do here is to show that we are not shut up to the hypothesis; for many agencies can be pointed to which may have cooperated to form lake-basins, some of which have hardly been considered with sufficient care, while others would appear to have been ignored. In dealing more especially with the lakes of the Northern Alps, I have simply selected a region which abounds in lakes occupying expansions of river-valleys which I know pretty intimately from direct obser- vation, including a district already treated more in detail by Prof. Bonney. Nor do I pretend either that all which is here advanced is new to geology, or that I am able to give an exhaustive account of all the agencies which have helped to form lake-basins past and present. Many lakes also, for example such as those which occupy the craters of extinct voleanoes in the Hifel and elsewhere, are excluded from consideration here by the title of the paper. I shall first draw attention to what appear to me defects in the arguments of the excavationists, as propounded by Sir A. Ramsay himself, by which they would shut us up to the “ excavation hypothesis.” 1. With reference to these Alpine lakes, the position of some of them is very difficult to. account for if they were excavated by the great glaciers which, during the Glacial Epoch, moved mainly from the central chain of the Alps towards the low-lying country. The position, for example, of such lakes as Neuchatel, the Bodensee, the Wallensee, the main portion of the Vierwaldstattersee, and the lakes lying in the Traun valley above Aussee would be very difficult to explain in this way, especially if we take into account also the con- tours of their basins. Seeing the thing in three dimensions, as in nature, gives a truer idea than the representation of it in two dimen- sions only, as on a map. 2. The argument against lakes lying in valleys formed by synclinal folds of the strata, so far as it goes, seems to overlook altogether the frequent occurrence of valleys lying along the axes of * Vide Prof. Ramsay’s paper, Quart. Journ. Geol. Soc. 1862. The arguments put forward in this paper have been recently summarized by Sir A. Ramsay in his ‘ Physical Geology and Geography of Great Britain,’ 1878. t Quart. Journ. Geol. Soc. vol. xxix. p. 382. 74 REV. A. IRVING ON THE anticlinal flexures. Such valleys occur very frequently in the Alps and elsewhere. Von Hauer* has given us an idealized section across the North-eastern Alps, representing the repeated occurrence of this phenomenon among the Secondary (chiefly Triassic) strata ; and I could furnish instance after instance from my own note-book in verification of his generalization. A capital instance of a line of valley thus placed with reference to a great line of anticlinal fracture, though quite subordinated to the main anticlinal of the whole Alpine chain, has been described by me elsewhere t. All such valleys have of course been widened and deepened by the erosive action of run- ning water; but it is difficult to conceive of their initiation in any other way than by open fissures, since otherwise the flow of water must have taken place along*an elevated ridge, which is absurd. Fracture, therefore, has played a part in the production of some valleys; and if so, it may have had something to do with the for- mation of any lake-basins which occur in such valleys. 3. The argument drawn from the fact (which every one will be ready to admit) that lines of dislocation produced in highly meta- morphosed strata must have appeared as closed lines when such strata were first laid bare by denudation, does not apply universally. It fails utterly when applied to the region with which we are espe- cially dealing; these valley-lakes do not occur among such strata, but among well-stratified rocks of Secondary and Tertiary age. Nor is it fair to reason from the close faults which occur among the comparatively undisturbed strata of Secondary age in England, where lakes do not occur, as to what can or cannot occur among the highly disturbed and contorted strata of the Alps, where valley- lakes abound. ‘The force of this argument will be fully appreciated only by those who know pretty intimately those portions of the Alps to which I now refer; seeing with one’s own eyes carries more conviction on this and many other points in geology than mere book- knowledge can do. In some cases disturbance and dislocation of © strata have gone to such an extent as to invert the original order in which the strata were deposited. A conspicuous instance of this sort of thing occurs in North Tyrol, which was described some years ago by Richthofen, and has since been more fully described by Gumbelt. Between Weissenbach and Vils the strata have been thrown into a sharp anticlinal, accompanied by inversion, so that strata of Jurassic age are found now underneath strata of the age of the lowest Trias. In the immediate proximity was once formed an extensive lake, studded with numerous islets, perhaps once the most beautiful lake ever produced among the Alps, though now filled up by the alluvial detritus of the river Lech and its affluents. On this stands the thriving modern town of Reutte. Again, in- version of a similar nature has taken place about the Rigi, in close proximity to the Lake of Lucerne, as Sir. A. Ramsay pointed out long ago§. May not both these lakes be connected with fractures * Die Geologie, fig. 251. +t Vide Geol. Mag. Nov. 1882. t{ Vide Von Hauer’s ‘ Die Geologie, pp. 366, 377, and fig. 254. § Quart. Journ. Geol. Soc. 1862. ORIGIN OF VALLEY-LAKES. ¥i5 and hollows formed: by those movements which produced the in- versions here alluded to? 4. The theory of subsidence has had but scant justice done to it. Alpine geology opens one’s eyes to the importance of this in a way that English geology does not; it is not enough therefore to dismiss this subject on the negative grounds that in England we have nothing more than mere pools formed in this way, where the solution and removal of the salt-beds beneath has led to subsidence. Nor do I see any force in the argument against the theory from the mere number of instances. No one would maintain that all the lakes even of the Northern Alps are due to subsidences ; the point is that here we do find conditions favourable to subsidence eatending over hundreds of miles in a district abounding in lakes. The materials of which the mountains are composed are for the most part calcareous and dolomitic in character, the strata being often almost pure limestone or dolomite. There is no necessity for explaining here how underground erosion is carried on by water in such strata; the fact is well known, asin the case the famous caverns of Castleton and Adelsberg, to mention on «wo most con- spicuous instances. Again, in this very region beds of rock-salt and gypsum abound, some of them of great dimensions. The lateral outflow from the Konigsee through a hole in the mountain, the water passing for miles underground and issuing again as a river in the direction of Hallein, serves very well to illustrate the way in which the waters which have done the work of underground erosion can find their way from these upland valleys, in which most of the lakes we are considering occur, down to lower levels. In some in- stances the actual amount of material carried from the interior of the mountains has been found to be astonishingly large. I shall mention here two examples given by Credner * :—(a) At Neusalz- werk, in Westphalia, it has been found that the brine-springs of that district bring to the surface 376 cubic metres of carbonate of lime annually ; (6) In the Visp-Thal, in Canton Wallis, a series of ‘ Erdfalle’ accompanied by small earthquakes continued for a whole month in the year 1855, due to the undermining of the valley by the removal -of gypsum in solution. The district contains some 20 gypsiferous springs, one of which alone brings to the surface over 200 cubic metres of gypsum per annum. Such facts enable us to realize more vividly than any thing which English geology presents to us the importance of underground erosion as a geological factor. It is the mere enunciation of an axiom to point out that in all such cases the materials cannot at the same time be removed from the interior of the mountains and remain inside them. ‘Time is the most important factor in such cases. The prevalence of salt and of brine-springs in those northern Alpine regions where the lakes are most numerous is recorded in the frequent occurrence of local names indicative of their presence (e. g. Salzburg, Hallein, Hallstadt); and an exploration of any of the salt-mines of the district opens one’s eyes to the great extent to which underground excavation has been * Elemente der Geologie, pp. 207, 210. ? 76 . REV. A. IRVING ON THE carried by mere solution within a few centuries only. It is some- what incomprehensible to me that the significance of the fact that so many of the valley-lakes of the northern Alps lie among strata where conditions specially favourable to subsidence are found, has been so much overlooked by writers in this country. The subject will be found more fully treated in the excellent text-book of Credner, who (p. 211) mentions the Konigsee as a special instance of a lake most likely formed by subsidence. The evidence is indirect but strong. Here we are led to the recognition of another agent which, so far as I know, has not been before specified as promoting subsi- dence. It has been pointed out by Credner that the conversion of anhydrite into gypsum by taking up water of crystallization, breaks up a valley-floor under which the more massive beds may have been removed by solution, and thus promotes subsidence. But I am not aware that it hasever been before pointed out that, where underground erosion from either of the causes mentioned had taken place in pre- glacial times beneath the floor of a valley (the natural operations which go on now having gone on then), the dead weight of the enor- mous glaciers * which filled valleys thus undermined may have crushed in their floors. For such reasons I think that the suggestion of Playfair that the bed of the Lake of Geneva had sunk owing to underground erosion, deserves more attention than it has received. The connexion between the frequent occurrence of lakes and limestone regions becomes more important when we consider the relative numbers of lakes in such regions and among the crystalline rocks. Something like 100 lakes may be counted (large and small) among the stratified deposits of the Alps, while I do not think a | dozen (if we exclude mere tarns) could be pointed out as lying among the crystalline rocks. And if we extend this examination to other parts of the earth’s surface, we find a similar rule holds good in Ireland (where tho lakes occur for the most part in the Car- boniferous Limestone), in the Apennines, in the region of the Middle Danube, in the Balkan peninsula, and in Asia Minor. Again, it is possible, reasoning from what we know of solution of the chalk strata at the surface in England by rain-water holding CO, in solution, to see how, in some cases, lakes once formed among limestone mountains may have been deepened by a process of che- mical solution. The long exposure of the snow which, on melting, feeds many of these lakes, and the mechanical division of the water during its descent into them from the mountains, are both circum- stances favourable to the saturation of the water which enters the lakes with the gases of the atmosphere. So far as this cause operates, it tends to deepen a lake, though it could only do so effec- tually in a case where the quantity of detritus carried into the lake was exceptionally small. Other Lake-forming Agencies.—(1.) Itis clear that any cause which leads to a change in the contour of a valley of erosion by distur- * Such as that of the Etsch-Thal, which filled the valley up to about 1500 metres (Credner, zbid. p, 663). ORIGIN OF VALLEY-LAKES. Te bance of the strata of the region may produce such alterations in the relative levels of different parts of the valley as to bring the floor of the valley at some points to a higher level than the floor of the valley nearer to its head. It is no drawing upon the imagination to say that such movements have occurred on a large scale in the Alps since the older lines of valley were eroded by water-action. To mention one fact only, the changes of level of the European con- tinent which brought the shores of the continent so far south as to allow the southerly drifting icebergs from Scandinavia to deposit their erratic blocks as far inland as Bonn, Westphalia, Thuringia, Saxony, and Moscow* could hardly have happened during Quater- nary times, and been reversed, without some considerable squeezing and consequent disturbances of the strata of the Alps, such as we observe today on such a gigantic scale. I take it, then, that many of the lines of flexure which have been before referred to as abounding in the stratified Alpine deposits are of comparatively recent date. The effect of such lines of flexure occurring transversely to the older lines of valley-erosion would be to form hollows, which must get filled with water from the surface- drainage of the valley. This has been already pointed out by Prof. Bonney+; and I may perhaps illustrate the principle more fully than he has done from the example of the Lake of Hallstadt. This beautiful lake lies in a north-to-south valley, and is one of some half a dozen lakes which le in the line of the river Traun. The stratigraphy of the Hallstadt basin may be understood readily from the accompanying sketch (p. 78), which was made on the spot, not from a cursory glance, but during a stay of more than a fortnight in the vicinity of Hallstadt. It of course represents the face of the mountain on the western side of the lake. The strata of the Sarstein massif on the eastern side show a corresponding synclinal arrangement. The line of anticlinal flexure, which is recorded in the dip of the strata from the Gosau Schlucht, having been formed in comparatively recent times, has produced so much of the Hallstadt lake as lies in a true rock-basin. The lake owes, however, a considerable portion of its depth to the diluvial detritus which has been brought into the Traun valley from a valley on the eastern side, and has dammed up the valley from Steg in the direction towards Ischl. The section represents a thickness of strata of over 4000 feet, entirely of Triassic age. About midway between the two gorges the strata above the pine- forests are highly contorted; and beyond them lie the contorted upper Dachstein strata. It is worthy of remark that such changes of level as are here re- ferred to are not so distinctly recorded as are many similar changes which are known to have occurred in mountain-regions contiguous to the sea (e.g. in Norway, in South America, and in New Zealand), where the mean sea-level serves as a datum-line from which to esti- mate their extent t. * Credner, El. der Geol. p. 651. + Quart. Journ. Geol. Soc. vol. xxix. t Vide Credner, zbid. pp. 169, 170. Section on the West Side of the Lake of Hallstadt. ——— mee me me ee PLY Present level of Lake. REY. A. IRVING ON THE (2) Again, it may well have hap- pened that in some of these later movements of the mountains, lateral thrusts may have acted vertically, or nearly so, to older lines of valley. The effect of such forces would be, if the resultant force was not too deeply seated, to partly close in the valleys upon which they were brought to bear, until the sides of the valleys in their narrowest parts came together, leaving the wider portions to form true rock-basins, which, getting filled with water from the surface-drainage, would, of course, form lakes. The direction of the principal lines of flexure of the Jura chain affords a strong pre- sumption in favour of the hypo- thesis that in this way the valley of the Rhone below Geneva may have been partly closed in. This alone would account for so much of that lake as lies in a true rock-basin. Part of its depth, however, it owes to the delta of the Arve, which has been brought down from the Mont- Blane region*. (3) There i is a phenbinatie well known to mining-engineers as a “creep.” In old coal-workings it is found that the vertical pressure of the superincumbent strata, acting upon the strata immediately below the galleries where the coal-seams have been removed, causes the floor of the mine to rise, so as in time to completely fill up the disused work- ings. Such a resolution of vertical pressure is possible where more mas- sive strata, as often occurs in the lake-regions of the Alps, rest upon strata of a more yielding nature on’ opposite sides of a gorge-like valley. In such cases the erosion of the valley being continued down to the lower strata, these, being at the * Vide ‘Les Causes actuelles en Géolo- gie,’ by Meunier, pp. 203, 204. t Vide Lyell, ‘Student’s Elements,’ p. 56. ORIGIN OF VALLEY-LAKES. 79 same time made more accessible to the surface-water and con- sequently rendered more yielding, might very well be elevated in places where the mechanical action of the valley-stream was too sluggish to enable it to wholly counteract any upthrust by the deepening of its own channel. In this way I conceive that some of the smaller valley-lakes may have been formed. (4) Some valleys occupy lines of faulted dislocation—such, for example, as that which runs northward from the Inn Thal, and coin- cides with the gorge of Jenbach and the Achen Thal. The evidence, which I recorded on the spot during three ascents of the surround- ing mountains, and which has been given by me elsewhere’, is confirmed by a reference to Von Hauer’s map of Tyrol. In this line, at nearly its highest part, lies the Achensee, the greatest depth of which (2500 feet) is said to be just off the eastern shore. Here is a connexion of a lake with a fault. (5) The function of moraines in forming dams across valleys is too well known to require that I should occupy much space here with the consideration of it. As, however, one’s acquaintance with the Alps increases, one is greatly impressed with the frequency of such occurrences, notable examples of which occur in the Oetz-Thal, in the cases of the Obersee and Toplitsee, and many others which might be mentioned, where a terminal moraine has blocked up the valley. In some cases, as with L. Garda‘ and the L. of Llanberis, such moraine-material would appear to have been redistributed, and so partly stratified by marine action. Crednert has already pointed to the connexion of the outlying lakes of the Alpine region in 8. Bavaria—the Ammersee, Starnbergersee, and Chiemsee—with moraines. Huge lateral moraines, left by the glaciers on the sides of valleys, such as may be observed at the head of the Rofen Thal below the Hospiz, and at Carthaus in the Schnalser Thal, may descend and block up a valley if once permeated by water, which increases its weight and at the same time diminishes the internal friction of its materials as well as their friction against the moun- tain-side. Such a movement of a great moraine is now actually . going on at Fetar, in the Unter-Engadin §. (6) The last point leads to the consideration of Bergstiirze, of — which there are some notable instances, e.g. the formation of L. d’Alleghe and L. Derborence during the last century only, and the catastrophes of Goldau and Elm within the present century. The extent of the part which these have played in Alpine physiography ean only be faintly understood from what is observable in such worn-down stumps of more ancient mountains as we meet with in Wales, Cumberland, and Scotland. ‘We find,” says Heim, “no Alpine valley without such heaps of mountain débris and tradi- tions haying reference to them. A still older series belongs to a time extending far back beyond the range even of tradition. The largest and, perhaps, the oldest we can recognize is that of * Geol. Mag. Nov. 1882. “The Triassic Deposits of the Alps.” + Cf. Mr. John Ball, F.R.S., Phil. Mag. 1863, vol. xxvi. p. 499. { Elem. d. Geol. p. 663. § Vide Heim, ‘ Ueber Bergstiirze.’ 80: "REY. A. IRVING ON THE Flims, in Graubindten. It forms a mountain 600 metres high, and extends from near Hanz to Reichenau, in the Vorder-Rhein Thal. Upon its surface eight small lakes are found. The Rhine and its tributary streams have sawn it out into a number of hills which abut as spurs upon the valley ; and at the present time above Ilanz there are still visible traces of the ancient lake which was formed by the waters of the Rhine as the valley was blocked up by the Fels- sturz of Flims. Upon the back of this huge mountain of débris there are traces of glacier-moraines and enormous blocks from the Puntaiglas Thal. This Bergsturz is therefore older than the period at which the moraine-materials were spread upon it—that is, older than the glacial period.” A few smaller lakes formed by Bergstiirze were observed by me last summer near the Firn Pass in Tyrol; but I am not aware that they have been yet described anywhere. (7) Lastly, there is the action of diluevwm in blocking up valleys. This fact must be so familiar to all Alpine travellers as to need no proof here. In many cases the materials have been more widely distributed, so as to form alluvial deltas, of which I have noted a score or two of instances, at points where two valleys converge. In other eases the diluvial detritus remains piled up in huge chaotic masses at the mouth of a gorge opening laterally into a valley. I saw such a recent accumulation only last summer, near Solden, in the Oetz Thal; and Credner* relates an instance which happened in the year 1818, in the Bainen Thal, in which a mass of débris 100 metres high was accumulated at the mouth of the gorge, some of the transported blocks of granite measuring 40 cubic metres. Sudden and great downpours of rain in Alpine regions, or the rapid melting of snow by the warm dry wind known as the Fohn, may furnish the rushing waters. Collected from an extensive and steeply sloping basin high in the mountains, and driven with enormous velocity downwards, as through the narrow neck of a funnel, the transport- ing power of water becomes enormous, while the rock-materials, being only generally of a specific gravity of 2-00 to 2°90, lose nearly one half of their weight in water. As soon as such a powerful cur- rent escapes from the confines of its narrow channel, its velocity of forward movement is again diminished, and it deposits its burden near the mouth of the gorge as chaotic masses of stones or as mud- streams. In time all the smaller stones and earth get spread out as deltas, which in numerous cases are seen protruding into lakes already formed, and in some instances have actually divided a lake into two, as is the case with the Plansee, and with the more notable instance of Lakes Brienz and Thun. Some instances of enormous work done in this way, as the result of a single storm, have come under my own observation. The side valleys which pour their water and rock-débris in this way into another valley are often of younger date than the principal valley: they may be seen in all stages of recession from older valleys in the Alps. Difference of composition and the strike of the strata may also help to forward disintegration along some lines rather than along others. The one * Klem. d. Geol. p. 222. ORIGIN OF VALLEY-LAKES. 81 essential point in connexion with diluvial accumulations is the fact, which can be verified over and over again, that the lakes at the outflow of which such accumulations are found, lie above the point of convergence of two valleys. It is for the advocates of the theory of glacial excavation to show why the work of excavation was not done where the union of two glaciers should have been capable of doing the greatest amount of work. We may define a lake as a hollow in the surface of the earth, which is so placed in relation both to the superficial and underground drainage of the district in which it is situated, that it rs filled with water, either to the lowest point or points of overflow, or to such a mean level as corresponds with the excess of the mean annual precipi- tation over the mean annual evaporation of the district in which rt lies. This is about as much as can be asserted generally of lakes ; and it is impossible to give any general rule as to the formation of their basins. Each basin must be worked out as it is, in relation to all the geological, cosmical, hydrographical, and other conditions which bear upon its existence. The hypothesis of the glacial exca- vation of lake-basins is not only irreconcilable with the known physical properties of ice; it is, besides, unnecessary, since lake- basins can be accounted for without it. Discussion. Mr. W. R. Browne suggested that a lake would be emptied by a slight tilt in the general slope of the country ; and a corresponding diminution in the slope would form the lake. Wastwater would require as much as one in twelve. This change, which would only be one in fifty for the Lake of Como, and one in ninety for the Lake of Geneva, might well have taken place in the various alterations which have happened in times since denudation had commenced. The elevation of the central part of the chain would tend also to increase the glaciers. Prof. Brake stated that there were many causes for lakes, and that those mentioned by Mr. Irving were familiar to geologists. The main question is, Can a glacier ever excavate a lake-basin? If that could be made clear, some progress would be made. He did not think it was proved that glaciers could not excavate lakes, From the shape of the sides of glacier-valleys, there must be most erosion at the bottom. If the medial moraines are brought up from the bottom, so might the excavated material be. The lakes of Neu- chatel and Constance were difficult to explain as eroded by glaciers. It was singular that, as the Lake of Geneva had been filled with ice, Miocene fishes should still be found in it. Mr. A. Tytor had proved by experiment that ice has erosive power. The velocity increases as the cube-root of the mass; and the erosive power augments in a high ratio with the velocity. Multi- plying the present mass of a glacier by 64 would increase its velo- city by 4, and its erosive power by at least 200. This he had Q. J.G.S8. No. 153. @ 82 ON THE ORIGIN OF VALLEY-LAKES, pointed out in the ‘ Geological Magazine’ for 1875, In lake-glaciers motion is chiefly derived from the expansion of ice in the act of. freezing. The water produced by the heat developed by friction during the progress of the glacier freezes in the glacier and causes motion (in the direction of the least resistance) towards the outlet of the lake where the glacier is thinnest. Mr. Tratt said the burden of proof must rest with those who asserted that a glacier could excavate. The differential motion of a glacier was a most important point. If the ice were moving up- hill it would be stopped by friction and gravitation; hence it was difficult to understand how a basin could be excavated, and very strong proof was needed. Prof. Bonney criticised Mr. Tylor’s section and idea, and stated that, while agreeing with Mr. Irving in his general principles, he differed as to the details. He thought that Mr. Irving had not allowed for the fact that many of the large lakes lay in true basins, | so that he could not explain them by moraines or débris. Still he thought that rock-movements accounted for the larger basins. Mr. W. Maruews referred to the recession of the Swiss glaciers during the last few years, and asked if on any rock thus exposed a basin had been seen. The AvtHor, in reply to Prof. Blake, referred to the last discussion and his last paper. He entirely dissented from Mr. Tylor’s view of the possibility of the ice doing erosive work by freezing at the base of the glacier; it would have to expand under great pressure, and so would compress and solidify rather than erode the rocky floor. The question of how much of a lake was true rock-basin and how much formed by moraine or detritus must be worked out in each case separately. ON TEE DRIFT-BEDS OF THE N.W. OF ENGLAND AND N. WALES. 83 8. The Drirr-pups of the Norru-Westr of Enetanp and Norte Wares.—Part Il. Their Nature, STRATIGRAPHY, and DistRI- purion. By T. Metiarp Reape, Esq., C.E., F.G.S. (Read November 15, 1882.) [Prats V.|] ConrTENTs. INTRODUCTION. RIVER-BASINS. Drifts of the Basin of the River Mersey. ” 99 99 Dee. » Ribble. Coast from Blackpool to St. Bees. re North Coast and Mountain-district of North Wales. CLASSIFICATION AND CONCLUSIONS. Pre-drift Surface of the Land. Red Sand and Rubble débris of the Trias. Gully-gravels. Low-level Boulder-clay and Sands. Mountain- and Hill-Drift. Conclusion. 33 InTRODUCTION. In Part I. of this paper, devoted to the ‘‘ Shells of the Lancashire and Cheshire Low-level Boulder-clay and Sands ’*, read before this Society in 1873, I discussed in outline some of the physical problems involved in a correct interpretation of the phenomena of the Drift as it is exhibited in Lancashire and Cheshire. It is now my intention, by the aid of numerous sections I have been more or less continuously recording before and since that time, together with observations in other parts of England and in Ireland, Scotland, and Wales, to redraw my picture with afirmer hand. This delineation of the Glacial period in the north-west of England will rest almost entirely upon my personal observations ; but, at the same time, I shall not neglect those important additions to our knowledge contained in the Geological Survey memoir by Mr. C. EH. De Rance, F.G.8.7, or the valuable papers by that untiring observer Mr. Mackintosh, F.G.8.¢, by Mr. Shone, F.G.8.§, the late Mr. Binney, F.R.S., and others by Messrs. Tiddeman, Goodchild, and the late Rev. Clifton Ward, published at various times in the Quarterly Journal of this Society. Neither will the more general papers by * Quart. Journ. Geol. Soc. February 1874, vol. xxx. pp. 27-87. t+ Superficial Geology of South-west Lancashire, { “ High-level Shelly Deposits of the Eastern Slopes of the Welsh Mountains,” Quart. Journ. Geol. Soc. vol. xxxvii. 1881, pp. 351-369. ‘Dispersion of Erratics of West of England and Hast of Wales,” Quart. Journ. Geol. Soc. vol. xxxv. 1879, pp. 425-453. § “Glacial Deposits of West Cheshire,” Quart. Journ. Geol. Soc. vol xxxiv. p. 383. Q. J.G.8. No. 154. sf 84 T, M. READE ON THE DRIFT-BEDS OF THE Mr. Searles Wood, jun., F.G.S.*, nor the work of Dr. James Geikie, F.R.S.7, ke neglected where they bear upon the subject in hand and lend any help towards unravelling this most complicated and difficult problem. RIVER-BASINS. To describe in a manner that can be realized the distribution of the Drift-deposits, it will be necessary to divide the country treated of into local areas. The most natural division that suggests itself to me is that of river-basins; for, as will be seen further on, the nature of the rocks occurring in these drainage-areas profoundly affects the character of the Drift distributed in them. . DRiFts oF THE BASIN oF THE River Mersry. In a communication to the Liverpool Geological Society in January 1873, entitled “‘The buried Valley of the Mersey,” I showed by a careful comparison of a number of borings, well-sinkings, and exca- vations between Warrington and Liverpool that, buried and obscured by drift, there exists a rocky valley, probably Preglacial, the bed of which at Warrington Bridge is about on a level with the present bed of the river at Wallasey Pool, about 26 miles further down. This gorge does not pass Runcorn 1n the line of the present river, but cuts across the low-lying land upon which the town of Widnes is now built. At this point its greatest depth is 76 feet below the present bed of the river at Wallasey Pool. From data detailed in that paper JT also inferred that a gorge most probably exists in the bed of the river between Liverpool and Birkenhead of a still greater depth. Since the above paper was written I have had opportunities of examining the excavations at the North Docks; and here, again, a system of lateral rocky gullies buried in drift was disclosed, one of which, filled with shingle and gravel under a coating of Boulder-clay, was excava- ted 40 feet below ordnance datum and further tested with an iron rod to a depth of fifty-two fect below ordnance datum without being bot- tomed +. And, further, I see from the Report in 1880 of Vice-Admiral Spratt, the acting Conservator of the river, that borings undertaken by Mr. George Hill, C.E., for the Upper-Mersey Navigation Com- missioners, show that between Weston Point and Hale Head there exists a deep depression. In Mr. Hill’s words, ‘“‘ The most unex- pected feature revealed by the borings is the existence of a deep depression in the underlying rock about the centre of the river, the sides of which appear in some parts to be precipitous ;” ‘“‘ the rock in the depression is from 10 to 20 feet below the rock adjoining on either side, or from 20 to 30 feet below the Old-Dock sill’’S. This depression is hardly deep enough to have formed the main channel; it may have been one of the tributaries; and such an * «The Newer Pliocene Period in England,’ Quart. Journ. Geol. Soe. vol. xxxvi. pp. 457-527. t The Great Ice Age. 1 See article by me in the ‘Builder’ on the Mersey Tunnel, February 4th 1882. § The Old-Dock sill is 4:25 feet below ordnance datum. NORTH-WEST OF ENGLAND AND NORTH WALES. 85 occurrence is what’ my previous investigations would have led me to expect. A boring for water at Burscough-Bridge Station shows the rock to be there 228 feet below ordnance datum, and to have 278 feet of drift, mostly sand and gravel, with some Boulder-clay beds, lying upon it. ‘hus there is progressive declination in the rock-level from Warrington towards the mouth of the Ribble. Drift in the neighbourhood of Liverpool and on the Lancashire side of the River Mersey. The Boulder-clay in the neighbourhood of Liverpool, though largely used for brick-making, is not a brick-clay in the same sense as the Clyde laminated clays. It contains many stones and in some places a good deal of limestone, which is apt to burst the brick when exposed to the weather after being calcined by the burning. In some cases the clay is passed through a machine to crush the stones. In places there are considerable depths of a fine unctuous clay, which has hitherto been considered unfit for brick-making; but the New Ferry Brick and Tile Company have proved this to be an old-fashioned delusion, as they produce much more perfect bricks than the ordinary eee shire or Cheshire make, and even tiles, with a very fine clay belonging to this series. I have touched upon the economical aspect of the question, as it enables a stranger much better to realize the geological nature of the clays. In this clay and underlying it, as I shall presently describe, are beds of sand Ae gravel. They rarely show at the surface in the district delineated; but sand beds are common in some other areas, and replace the Boulder- clay as a suriace-deposit. In my descriptions | do not propose to use the terms “ Upper and “ Lower” Boulder-clay and.“ Middle Sands and Gravels” as I consider these titles beg the question which it 1s my purpose to discuss. Sections exposed in the North Docks at Bootle (fig. 1).—Excava- tions for the Huskisson Branch Dock in 1872 showed the old shore-line of the river, consisting of silty gravels (with recent shells) from 1 ft. Gin. to 2 feet thick, buried below the spoil filled in to form what was the Dock quay. ‘The shore-line had been cut in the Boulder-clay, which was 24 feet thick at the deepest part, and consisted of an extremely homogeneous plastic clay containing occasional rounded pebbles and afew of the usual erratics. Beneath the clay was a bed of gravel from 2 ft. 6 in. to 3 ft. 6 in. thick, resting on red sand which covered the rock. The rock-surface would therefore be about fam 12 to 14 feet below ordnance datum. A few irregular patches of sand occurred in the clay ; but they were devoid of form or stratification. In June 1874 I examined the excayations for the new Alexandra HZ 9 86 T, M. READE ON THE DRIFT-BEDS OF THE Fig. 1.—Plan of Atlantic Docks, Liverpool, 1874-81. SSS, US TANGTON=D0CKSS0 ZN SS u =RV-ER = SS ——— = a soo 0 100 Sau ST eae ee ae Ya ee REBT Branch Dock No. 3; and here the Boulder-clay was again exposed. In the trench for the dock-wall parallel with the river-wall the surface of the Boulder-clay was 16 feet below ordnance datum. The section disclosed 9 feet of Boulder-clay, 3 feet of sand occasionally developing into gravel, and then from 3 to 5 feet of Boulder-clay resting on red sand and red ‘‘roach” rock. A Postglacial gully excavated in the Boulder-clay intersected the dock; and in this was the trunk of an oak tree, about 3 feet in diameter, lying longitudinally in the gully, as if it had been washed down it, at a level of about 22 feet below high water ; a layer of recent silt covered the whole. In May 1876 further excavations showed below the silt a con- siderable thickness of peaty matter full of drift timber, principally oak and pine. In one place only, at the base of the peat, I observed the stump of a tree rooted into sand resting upon the Boulder-clay. At various times between 1876 and 1878 I carefully examined the excavations of the series of docks and graving-docks, extending over an area of about 80 acres, and made record sections of the beds disclosed: these I reproduce (figs. 2-5, p. 88), as they will serve to explain the drift-geology better than any written description*. It must be understood that the beds are so irregular that another observer coming at a time when a different face was exposed would no doubt give a different representation of it. The sections I give I considered typical; and I vouch for their accuracy. Speaking generally, the same succession prevails as occurred in the sections already described. In ascending order we have the red rock of the Trias with the red sand and rubble covering it; upon * T must here express my indebtedness to Mr. G. F. Lyster, the engineer to the Mersey Docks and Harbour Board, for the facilities of inspection afforded me and for much information given. NORTH-WEST OF ENGLAND AND NORTH WALES. 87 this lies the shingle or gravel; but sometimes the Boulder-clay lies directly on the rock, and in other cases the shingle is divided by patches of Boulder-clay. At another point the gravel is enterel by a sand seam which divides the Boulder-clay where the deposit becomes thicker in consequence of depressions in the surface of the rock on which it rests. The lower part of the clay may be described as containing more waterworn boulders than the upper, thus partaking of the nature of the shingle it is in contact with. The upper part of the clay is more plastic, like that of the Huskisson Branch Dock, but contains, usually, large and numerous glaciated erratics of granite, dolerite, diorite, limestone &c., some of which must weigh several tons. In November 1881 I made a very careful inspection of the exca- vations then going on in Dock F, the most northern of the series : in this I was assisted by Mr. John Dickson, the contractors’ engineer, who has had charge of all the excavations from the commencement. The sections are shown on p. 88. A-—B (fig. 2) shows the face of the cutting of the south side of the dock, C-D (fig. 3) that on the north side ; they are seen in reverse directions. This dock over its whole area showed a much more decided and persistent arrangement of the beds than did any of the others. Commencing at the base, in section A-B, we have, in ascending order, No. 1, a bed of gravel and sand not bottomed, 5 feet; No. 2, a bed of short hard clay containing many waterworn drift stones. This clay contains shell-fragments, and in excavating requires to be first loosened with the pick. No.3,a bed of sand varying in thickness from 2 or 3 feet to a mere line. It is remarkably persistent all over the dock; it appears to be a line of erosion running nearly in one plane with a slight dip towards the river, excepting in places where it curves into hummocks, as shown in section C—D (fig. 3), where it in one place develops into gravel, and at another is split up by a tongue of clay. No. 4, strong plastic clay which has to be got with the grafting- spade, consequently costing more labour in the getting. It contains many erratic boulders, principally of diorite, the largest measuring 4 ft. 6 in. by 4 ft. 6 im. by 4 ft. 6 in. I counted thirteen lying at the 8.E. corner of the dock. It is about 13 feet thick at the east end, but thins down to about 7 feet at the west end of the dock. Bricks are made from this clay for use in the dock-works. The elay No. 2 is unfitted for brickmaking. No 5is a Postglacial silt, in which many bones of Cetaceans have been found, also red deers’ antlers, skulls and bones of a small variety of horse &c. They usually lie near the base, which rests on the Boulder-clay. As these last sections were decidedly the most markedly stratified, it was of importance to find out if any distinctions occurred in the organic remains which they contained. This could not very well be done through evidence of the Mollusca, their remains, though present, being scattered and fragmentary, and no heaps of clay being left from which weathering would separate them, as happens in brick-fields. I therefore submitted, for microscopic examination 88 T. M. READE ON THE DRIFT-BEDS OF THE “OAT JUESAId JO oUT]-24048 PIO “” ‘(sv1ay,) oor pee “ZL ‘guys dn-perny ‘9 ‘IMS “Gg ‘apuds oy} 4914 409 ‘Avyo-aopmnog “z ‘a[sulYs pur ‘puvs ‘aatag ‘¢ ‘Avjo-aeptnog Luojg °Z ‘e[SUIYS pus ‘pues TOAtAL) TT — SS ‘(ued uo H-D) TON yoog younug we woyoag—e "SLT ‘(ued uo G-9) q yoog fo apis Y,LONT UO UOLV0AG7—"G “BIA D AA. ‘(avid uo qT) g ‘ON yoog ‘PUL 1 04 400} OFS OTCOS [e}WoZIA077) 7 = L aT YOUnIT UL WON — FP "SIT *({ sy ‘ued uo q—y) 7 yoo fo apis YING UO UW0L009—¢ “BLT (4993 OT 03 yout 4 o7vos peory20a , "ooduanyT ‘syoog oyunny oyr bugnaroxa ur posodia SU0LIIIgGI—"G-ZG *SSI,q NORTH-WEST OF ENGLAND AND NORTH WALES. 89 specimens of the several beds, weighing about 8 Ib. each, to my friend Mr. David Robertson, F.L.S., F.G.S., of Glasgow, who kindly made a very full Report, which I annex in an appendix. It will be observed that bed No. 1 contains a few freshwater Ostracoda among numerous fragments of marine shells. As I believe from other evidences that this is the nearest approach to a littoral deposit we find in our Drift, it might well happen, if a side stream had entered the Mersey at this point, that freshwater Ostracoda would get mixed with marine remains. No. 2, the reddish-brown clay, consists of 60 parts of mud, 32 of sand, and 8 of gravel, about one half of the sand and gravel being more or less angular, the other half well rounded, and one piece finely striated. No. 3, red muddy sand, much waterworn, contained very small chips of shells and a fragment of a starfish and one Foraminifer. No. 4, red clay, consists of 58 parts fine mud, 20 of sand, and 22 of gravel, mostly small and waterworn. It is unnecessary for me to give more particulars, as they will be found in Mr. Robertson’s very careful report; but it is important to notice the effect of the differing proportions of the three constituents, viz. mud, sand, and gravel, on the nature of theclay. It will be ob- served that bed No. 4, which I describe as plastic clay from which bricks are made, contains 2 per cent. less sand than clay No. 2; but the sand is replaced by a larger proportion of gravel, which does not tend to make the clay “short” as sand does. The gravel is not very noticeable to the eye; but on attempting to cut the clay with a knife, its presence is very soon discovered; at the same time I was hardly prepared for so large a proportion of gravel as 22 per cent. I have dwelt much on these sections, as it is seldom one has an opportunity of examining excavations on so large a scale, and they are instructive as showing how the beds of which the drift is com- posed vary as new faces are exposed. I have given an exact and literal description of the appearance of the beds, leaving for discussion afterwards whether it is possible geologically to separate them. The New Garston Dock.—These excavations I have already de- scribed in a communication to the Liverpool Geological Society * therefore it will suffice to say that the Boulder-clay here rests upon a very regular shelf of rock covered with the usual red sand and rubble, varying from a few inches to three feet thick. In some places current-bedded sand and gravel, and mixed clay and gravel, occur between the red sand and the Boulder-clay. Garston is about 5 miles ae Liverpool, on the same bank of the _ Mersey. Widnes.—The deposits here are described in the Le on the “ Buried valley of the Mersey” already alluded toy. The deepest * “The Glacial and Postglacial Deposits of Garston and the surrounding District, with remarks on the structure of the Boulder-clay,’ Proc. ef the Liverpool Geol. Soc., Session 1874-75. t Proce. of Geol. Soc. of Liverpool, Session 1872-73. 90 T. M. READE ON THE DRIFT-BEDS OF THE bering, No. 5, at Lambert’s copper-works, showed the rock to be 163 ft. below the surface. The bulk of the clay penetrated by the various bore-holes was of a very fine plastic nature, like that used by the New Ferry Brick and Tile Company. ‘There did not appear to be many stones in the _ clay ; but they were said to increase near the bottom both in number and size. There were erratic pebbles mixed with the red sand. A piece of cannel and some pieces of bituminous coal were found in one of the wells, 49 ft. from the surface. Sankey Bridges.—The Boulder-clay was here penetrated to 100 feet below the surface, the last 20 feet showed 5 feet of sand mixed with coal-dust, and 15 feet of clay with bands of gravel. A sand- and-gravel seam 2 feet thick was passed through about 53 feet from the surface. Cheshire Lines, Liverpool Extension Railway.—Between Garston and Warrington there were some excellent sections of the Drift disclosed from the beginning of 1872 until the middle of 1873. One continuous section commencing at Garston was 3,4, miles long. The whole of the features hitherto described as distinguishing the drift were to be found in one place or the other in this section. The diagram (fig. 6) exhibits, as accurately as I could depict it, the varying nature of the drift. Speaking generally, the Boulder-clay was divided by very persis- tent seams of sand as at A, though in places these thinned out and the upper and lower beds coalesced and became one with no ob- servable division as at B B, or the gradation from clay to sand was undefinable, as at C C. The sand beds were often distinctly strati- fied,as at A. The beds of sand were occasionally arched and lami- nated, as at E. On some of the sand beds the clay immediately resting upon them was in book-leaf laminations which showed very plainly by weathering, as at F'; or the bed of sand actually termi- nated in laminations as at D. The usual red sand and rubble, G, occurred at the base of the clay when the rock H was reached. At ‘one point a gravel-bed occurred between the clay and the rock. In places the lower part of the clay was decidedly harder and stonier than that above, as at L; but this quality was not always persistent; it also contained beds of gravel, as at M. ‘The upper part of the clay, as is common, was split by shrinkage, and broke, with blue faces or irregular joints to the sand-seams below. Irregular sporadic patches of sand occur in the clay, as at J. The usual glacial erratics and boulders were met with. In one place thin beds of clay, A’A’, were interstratified with the sand; I saw no boulders in the sand-seams ; but I cannot affirm that none were met with. Shell-fragments were to be picked up; but they were not very frequent; they are given in my list, Part I. Hitherto I have made no remarks on the shell-fragments; but they are to be found more or less in all the clays I have described, occasionally in the sand-seams. Some sand appears utterly barren ; but the clay always contains them. | The section at Farnworth (fig. 7) was just over a milelong; and [To face p. 90. Fic. 6 _jond Garston Station. ob: Road Halwood ie to Speke. peke Yo) Y Life Ld Vy Manes: LY Li LA Soe) La. YO: Z) LOAN EY a0 OG/ CIN, MR TOM, LU Dy, WW, Fs iyi EE = ZRET IE TILA RLY ASD Z 7s 2 Miles L. & N. W. Railway, 8 Roady Junction with line to near Allerton Station. uiverpool. Brunswick Station. or 88 ‘ : [Zo face p. 90. Vig. 6.—Cheshire Lines, Liverpool Extension Railway : Section from beyond Hunt's Cross Station to beyond Garston Station. (Vertical scale + inch to 10 feet.) Road Halwood to Speke. Rond frjpin Woolton to}Speke. en ° Ry SS OO YW. Yl] YY Ws Vi. ip SS DN ARG Fe tee = * stall de Hunts) Cross 0 Dies A/a, 5/5 a | L. & N. W. Railway, near Allerton Station. St. Mary’s Road, Garston to Liverpool. ai) ins'88 ay [ee Ae SO LLY aMilewn Junction with line to Brunswick Station. yy, Wa Ye Vz Note.—The figures on thi il Ws Eyze Wo le surface-line represent the depth of the cutting. SDiiles Ye ant eee a NORTH-WEST OF ENGLAND AND NORTH WALES. ! here for ? of a mile the Boulder- clay was split in two bya persistent seam of sand, N, not more than 6 inches thick; the surface of the ° clay and the imbedded seam of sand followed pretty closely the contour of the rock, O, below. The remainder of the cutting showed a seam of sand, P, also; but it was not so defined, and appeared to thin out to nothing at either end or lose itself in indefinable divi- sions or ramifications, P’. Shell-fragments were in the sand-seam. The base of theclay rested on the usual red sand, Q, degraded from the rock below. Nearer Warrington there was a eutting in the Boulder-clay; but it presented no features to indivi- dualize it. Wigan Junction Railway. — These sections are described in my paper “On a Section through Glazebrook Moss ” (Quart. Journ. Geol. Soc. vol. xxxiv. pp. 808- 810). The same feature of a divi- sional seam of sand shows in the cutting; but a bed of “ book- leaf” clay 18 inches thick, 8 feet from the surface and a quarter of a mile long, distinguishes the section. Cheshire Lines from Hunt’s Cross to Aintree.—I examined the sec- Nay): tions along this railway while it ¥I\\\\\\AGISSS-------- was being made in 1876-7. The “i same peculiarities occur at the base of the clay where it is seen resting on the rock, viz. red sand and oc- casional grayel-beds. There were a few seams of sand in the clay itself; but there was nothing pe- culiar to mark these sections from the last. About Liverpool.-—Excavations in the City of Liverpool show the Boulder-clay resting on the red sand ; and I have seen very large erratics taken out of them. It WA ) < 8/5 JaUler2 iG . 4 = [ Ws AS % Vi; iy, yp, Ne) — ‘hompny wuorsuaug poodwaavy ‘sauvy auysayg—) *S1q ‘qouelg 8 UoTeT 49 BATION “M “N29 “'T "CI-GLEL “YplOnrUlDyT 20 wWorzoag “ABMOSNED asnoy poy (-q00F OT 0} Your { ofeos Tworj19 4 ) 92 T. M. READE ON THE DRIFT-BEDS OF THE is very striking to see these records of a Glacial age extracted from under the busy offices and warehouses of Liverpool. Brick-pits are studded all round the town. In the excavations of the Granby- Street Board Schools, Toxteth Park, of which I was the architect, the Boulder-clay was covered, first with a soft blue clay, evidently the washings of the Boulder-clay, probably subaerial, about 2 feet thick, then ‘‘ Washed drift sand,” covered by a peat-bed with remains of firs therein, and then surface soil. There is nothing special to notice about the clay of the brick-pits. When the Boulder-clay is absent, granitic and other glaciated erratics are often met with resting directly on the rock. In an excavation on the side of the lake in Sefton Park the surface of the rock below thin beds of sand and gravel was worn smooth as if by a stream of water. A section of the cutting at Bootle-Lane Station is given in Part I. of this paper, Feb. 1874, p. 28; the bed of Boulder-clay No. 6 resembles the bulk of the clay disclosed by the Widnes boreholes. Rwwer-cliffs—One description will do for these Boulder-clay cliffs. At Egremont there is a divisional seam of sand in the Boulder- clay; and Mr. Mackintosh* argues that below this the clay is “ Lower Boulder-clay,” and above it the “ Upper.” Between the Dingle and Garston a similar divisional seam of gravel occurs. I will reserve the discussion of the relations of these beds until my description is finished. The cliff south of Garston also shows a divisional plane with a few pebbles marking it. Boulder-clay cliffs are to be seen near Kastham ; but to describe them would be only reiterating what I have already said. Drift on the Cheshire side of the Mersey Basin. That portion of the peninsula of Wirral draining into the Mersey is generally, excepting over the more prominent elevations, covered with a Boulder-clay answering to that on the Lancashire side ; where the base is exposed the same phenomenon of red sand and rubble and yellow sand and rubble, according to the nature of the underlying rock, is to be seen. This I observed at the excava- tions for the new Birkenhead Station, in excavations for houses at Oxton, and in sand-pits in the Happy Valley, Tranmere yt. Striated rock-surfaces have been recorded also from time to time by various observers. From an examination J made in 1877, before reporting to the Wallasey Board on a site for the proposed cemetery, I came to the conclusion that on the coast just beyond the “ Red Noses ” there exists a sea-cliff buried in Boulder-clay, which is again covered with blown sand. Borings at Leasowe Castle by Mr. Cunningham, reported to the British Association in 1854, also showed that there was a con- siderable and undetermined thickness of Boulder-clay near the shore. The phenomenon of deep rock-gullies filled in with drift also * Quart. Journ. Geol. Soc. vol. xxxiii. 1877, p. 732. t Described by Dr. Ricketts in Proceedings of Liverpool Geol. Soc., Session 1876-77, p. 254. NORTH-WEST OF ENGLAND AND NORTH WALES. va occurs at Tranmere Pool and at Hooton, thus bearing out the in- ferences drawn from the Lancashire side *. At Upton, in May 1873, I examined with Mr. Shone a sandhole which showed a section of current-bedded sand and gravel with rolled pebbles of Red Sandstone mixed with erratic pebbles (fig. 8); in it the usual shell-fragments occurred. The whole was overlain by a thin bed of Boulder-clay of irregular thickness, containing only a small proportion of stones. JI have very seldom found rolled pebbles of Triassic rocks, except in some of the clay of the dock- excavations on the Lancashire side; and this is worth bearingin mind. The same section, excepting that the sand. is of a flesh- or reddish colour generally, was to be seen at the Backford sandhole. Here the valley is about 50 feet deep, and must formerly have been filled with sand, which has been denuded. Fig. 8.—Seciion im Sandpit, Upton. (ge ' a. Sands and gravels. b. Bed of gravel containing many rounded pebbles of red sandstone. ec. Current-bedded sands containing beds of gravel, in which are many fragments of shells, especially in the gravel. d. Boulder-clay with few stones. Helsby and Frodsham.—Ascending the Mersey, the Boulder-clay skirts the foot of the Helsby and Frodsham hills which overlook the upper marshes of the Mersey. When, however, we follow the valleys which ramify among these hills, we find they are largely filled with a sandy drift not unlike that at Backford. A section disclosed by a well I sunk at the late Mr. George Eastee’s house, about a mile from Helsby, showed 20 feet of yellow flesh-coloured and red sand, evidently arranged by water, with a seam of clay about the middle of it, the whole resting on soft rock. On the northern face of Helsby Hill, about halfway up, a cutting at the side of a road showed the following section (fig. 9), the rela- tion of the Drift to the escarpment of Helsby Hill being shown in fig. 10. Valley of the Weaver.—From Frodsham up to the Grand Junction Viaduct the river Weaver runs through a deep valley cut in the * “ Buried valley of the Mersey,” Proc. of Geol Soc. of Liverpool, Session 1872-73. 94 T. M. READE ON THE DRIFI-BEDS OF THE Fig. 9.—Section on side of road, northern face of Helsby Hill. : Surface of road. 1. Hard flesh-coloured Keuper sandstone. 2. Fine yellow laminated sand with interlaminations of reddish clayey sand, very regular and distinct, with a slight slope up the hill; occasional waterworn travelled pebbles occur, principally near the top. 3. Sand blackened with iron oxide (“ fox-bed ”). 4, Subaerial detritus, or “ talus,” from the cliffs above, with white sand filling the interstices between the rubble, and large irregular blocks which were freely strewn on the slope. I could find no shells. At several points along this road are to be seen very large granite boulders lying in the sand-drift. Fig. 10.—Section™ showing the position of the Stratified Drift m relation to the escarpment of Helsby Hills. 2 reper t a, b. Talus of angular blocks of Red Sandstone, lying upon and covering stratified yellow sand (c) containing occasional rounded drift-pebbles. Keuper marls. At one point there is a remarkable basin-shaped sweep from the cliffs on the right bank of the river, looking very like the effect of a large landslip afterwards modified by rain. There is a remarkable absence of drift in this valley, which is worth noting. On the plateau on each side drift can be found. About Aston there is generally sand covered with sandy clay and then with yellow sand; and near Aston Grange the Boulder-clay is to be seen in the farmyard. Erratic boulders are also to be met with. Stockport.—In April 1874 I examined and drew a section (fig. 11) of the drift disclosed in the London and North-Western Railway Company’s ballast-pit at Stockport. At the base was very light yellow sand, 1, in all sorts of curved contortions, looking at a distance like NORIH-WEST OF ENGLAND AND NORTH WALES. 95 rock. Resting upon an uneven surface of this bed was a bed of current-bedded sand and gravel, 2. Above this occurred lenticular patches of clay, 3, almost forming a continuous bed, and of irregular thickness. The whole was capped by stratified sands and gravels, 4 and 5, and surface soil, 6. The river Mersey runs in a valley much below the level of this drift; and no doubt there has been considerable river-denudation. Fig. 11.—Section in Ballast-pit, Stockport. = Sy) 7M SMOG 1. Light yellow sand, contorted. 2. Obliquely stratified sands and gravels. 3. Clay. 4, Stratified sand. 5. Gravel. 6. Soil. Hazel Grove.-—A. somewhat similar series of sand and gravel beds on a small scale occurs at Hazel Grove; and there are many Triassic and Carboniferous pebbles in it. Poynton.—Boulder-clay is to be seen in a brick-pit at Poynton. The erratic stones are distinguished by being very much weathered. The nature of the clay is evidently affected by the neighbourhood . of the hills of Carboniferous sandstones and shales, being of an arenaceous nature and of a yellowish hue. It appears to rest upon a sand like that of the Macclesfield-Cemetery beds. Macclesfield.—About the same date I also examined the Maccles- field-Cemetery beds described by Mr. R. D. Darbishire, F.G.S., in 1865*. The section exhibited a considerable thickness of current- bedded yellow sands and gravels with shell-fragments. Mr. Darbi- shire estimates these beds at 70 feet thick, and states that they rest on Boulder-clay. He enumerates 49 species of marine shells from them. The elevation above the sea is stated at from 500 to 600 feet. Beyond the bridge over the Macclesfield canal, at about 550 feet above ordnance datum, was a brick-yard consisting of purple-red clay covered with yellowish arenaceous clay. ‘The contained stones were more rounded and weathered than those about Liverpool. I found several flints, one being about 23 inches in diameter. In a stream by the “Setter Dog” was to be seen the drift- gravel first described by Mr.-Prestwich. The gravel is very pecu- liar, being full of loamy matter; and along with well-worn travelled gravel are angular, subangular, and flat flagey pieces of Carboni- ferous sandstone. It is very singular to find this isolated patch of marine gravel full of shell-fragments among these hills. The neigh- bouring hills have only a thin covering of subaerial detritus; but erratics of Buttermere syenite and greenstone are to be found scat- tered over them as far as the “ Bow Stones.” Isaw no evidences of * Memoirs of the Literary and Philosophical Society of Manchester, vol, ii. series 11. pp. 56-66. 96 T, M. READE ON THE DRIFI-BEDS OF THE striation or planing on any of the numerous erratics | examined. There is at Styperson, covering the rock, a red clay with erratic stones; but all over these hills, as far as Whaleybridge, drift is generally absent. Drift of the Rwington Hills. . Perhaps the most interesting section of the Drift occurring among the hills was that shown by the excavation for the puddle- Fig. 12.—Turner’s Trench, Yarrow Reservoir, 1872-75. (Horizontal and vertical scales 220 feet to 1 inch.) Top wecekL LiLezz7 THT TT Z SINS Za iff KY, = dDPMbMpMC QI ys Yipee Ye S J = a AWN Ao ARGON A, A. Bottom, as excavated. G. Gravel. B, B. Shale excavated. H. Good blue clay, containing a few C. Gravel and sand. ° stones, sand-veins, and some gravel. D, D. Loam. I. Peat and clay. E. Boulders and gravel. J. Soil. FB. Gravel and sand. wall of the Yarrow Reservoir at Rivington*. The site of the reservoir is to a large extent in the coal-shales, as proved by the tunnels and trenches. The river Yarrow runs from Allan’s Bridge into the Anglezark Reservoir; and the valley dammed up by the embankments thrown across at two points to form the Yarrow Reservoir is a side valley, which the excavations proved was the Preglacial valley of the Yarrow. This had been sufficiently filled up with drift-deposits to divert the river up to the valley of the Anglezark Reservoir. The excavations for what was called Turner’s Trench (fig. 12) disclosed a gully in the solid rock 140 feet deep from the surface, which, from the winding of its course, was intersected by the same trench in two places, Gand H. ‘The bottom was in Mill- stone Grit, which had a surface highly waterworn. The lower part of the gully, Mr. Martin, the resident engineer, informed me, was filled up with gravel, sand, and large stones, the largest being boulders of Millstone Grit. The upper part was of blue Boulder- clay; and throughout there were waterworn boulders and pebbles of limestone, granite, porphyry, and greenstone. The excavation for. the puddle-wall of the Yarrow embankment intersected the same gully at a higher level, and was filled up with loam, sand, veins of gravel, and boulders (fig. 13). * The Yarrow is a tributary of the river Douglas between the Mersey and the Ribble. Not having devoted a special division to the basin of the Douglas, the description of the Drift is most naturally in place here. NORTH-WEST OF ENGLAND AND NORTH WALES. 97 The sections of the two trenches are from diagrams and descrip- tions kindly given me by Mr. Martin, the resident engineer. Fig. 18.—Section in Yarrow Trench, Yarrow Reservoir, 1872-75. (Horizontal and vertical scales 220 feet to 1 inch.) op tveater A, A. Bottom, as excavated. EK. Soft shale. B. Hard grit rock. F. Loam, sand, and veins of gravel. C, C. Black shale. G. Gravel. D. Soft broken shale and clay. H. Soil. Fig. 14.—Section of Turner’s Side Trench, Rwington. (Scale as in fig. 12). 1 in TMT TESTIS TTI) Sa et ea 1. Brown clay with boulders. 2. Sand and gravel current-bedded, together and mixed with clay. 3. Rubble débris of coal-shale in large angular blocks. 4. Coal shales. At a the clay was not. bottomed at 20 feet deep. 6 is supposed to be another section of the gully. Fig. 15.—Section in Cutting on the west side of Yarrow Reservoir. Zz = YUL YT; LE Uf YY YY YI PUT iW MD LI ] 1. Yellowish clay, 3 feet. 2. Brown clay with angular fragments of coal shale. 3. Blue clay (Till) very hard, but falls when wet—full of pebbles, and containing many Millstone-grit boulders (up to 8 tons) mostly waterworn, but none glaciated. 98 T. M., READE ON THE DRIFT-BEDS OF THE I examined these excavations in May 1872, and again in January 1875. Turner’s Side Trench, a bend in Turner’s Trench along the side of the hill, showed a section of coal-shales, with an angular rubble débris resting on the upturned edges of the shale (fig. 14). Upon this lay a patch of sand and gravel, current-bedded, the whole being overlain by brown clay with boulders. At another point in the reservoir the side of the hill was being excavated for clay for the puddle walls, and for material to form the embankments (see fig. 15). The working face was 200 feet long and 16 feet deep. The lower stratum was of Boulder-clay, blue-grey in colour, unlaminated, and very hard, but falling with water. The clay was very full of pebbles and Millstone-grit boulders up to 8 tons in weight, which had to be blasted. I have since seen the Scotch Till, and find this Yarrow Boulder-clay to be a deposit apparently identical with that found on the Carboniferous areas of the Clyde and Forth *. Upon this ‘‘ Till” lay a brown clay containing angular fragments of Coal-measure shales; and the whole was capped by a yellowish clay about 3 feet thick. There were no signs of shells or shell- fragments about; nor had any been found. The deposit was evi- dently unfossiliferous. In 1875 I saw a similar excavation, also on the west side of the valley, in the Till, which was of a grey colour and full of limestone, granite, syenite, and greenstone boulders, all rounded and slightly scratched indefinitely over the whole surface. The limestone showed the scratches plainly. The deposit was hard and compact, requiring a pick to disturb it; and, as I noted at the time, it answered to the description of the Scotch Till, which I have since verified by examination. The level of the reservoir is about the 500 feet contour. At Ad- lington, at a lower level, I found a clay used for making bricks similar to our marine Boulder-clay, but containing more pieces of shale, and in it one small fragment of a bivalve. So far I have confined my description entirely to personal obser- vations. Over so large an area as the Mersey-basin, containing some 1748 square miles, there are necessarily numerous places left entirely undescribed. Still I think sufficient has been given to form a fair general view of the distribution of the Drift in it. It would be only wearisome iteration to go on further describing section after section. Before proceeding to describe the deposits in other river-basins, it will be well, while the facts are in our minds, to try if we can discover any general principle which will connect all these detached observations. Observations on the preceding Sections. A glance at the large map I have prepared (PI. V.), showing the position of the preceding drift-sections in relation to the geological formations and drainage-lines of the country, shows at once that there is an intifmate connexion between them and the nature and distribution of the Drift. The greatest depth of Boulder-clay is at * « The Glacial-beds ot the Clyde and Forth,” Liverpool Geol. Soc., Session 1879-80, pp. 139-153. NORTH-WEST OF ENGLAND AND NORTH WALES. 99 Widnes, near the mouth of the river Weaver, which, with its tribu- taries, is almost confined to the Keuper marls. The extremely fine unctuous nature of the clay down to a depth of 140 feet from the surface, together with the position of the deposit, points most conclusively to the river Weaver as the origin of it. None the less clear to my mind is it that the whole of the Boulder-clay, as seen in the cliffs on either side of the river Mersey, from Runcorn to its mouth, and the fine clay in the North Docks, have a similar origin. The plateau of Keuper marls also appears to have provided the matrix of the Boulder-clay along the valley of the Mersey up to Manchester. When we get to Stockport we are in the region of sands and gravels, as is also the case at Hazel Grove. Here the drainage is from the steep hills of Carboniferous sandstone; there- fore it is anatural and fair inference that the character of the Drift has changed from this cause. At Macclesfield we again come onto the Boulder-clay and over- lying sands and gravels ; and as that town is situated at the foot of steep hills of Carboniferous sandstones on t!.e one hand, and the Keuper-marl plateau on the other, we are led to conclude that the sands have come from the one, the clays from the other. The northern side of the Mersey is occupied with Triassic rocks; and beyond lie the Coal-measures. May not these have supplied the material for the included beds of sand found in the Boulder-clay ? At Upton, in Cheshire, the drift again changes; but here we are getting very close to the drainage of the river Dee, which is largely fed from a sandstone area; so now I will proceed to examine the drift found in the basin of the river Dee. DRIFT OF THE BAstIn OF THE River DEx. From West Kirby to Parkgate, on the east bank ef the estuary of the Dee, is a cliff of Boulder-clay which has attracted some atten- tion. The section at Dawpool has been described by Mr. Mackin- tosh in the Quarterly Journal of the Geological Society, vol. xxviii. pp. 888-392, and vol. xxxiil. pp. 731,732. Nearer to Hoylake the cliff is lower, and overlain by blown sand. Between the blown sand and the Boulder-clay is a bed of Postglacial blue silt or clay, 1 foot 9 inches thick and 4 feet 6 inches above high-water level; further up the estuary this bed rises to 9 feet above the same level, with a thin bed of soil under it, and then sand lying on the Boulder- clay. The highest level of these Postglacial beds noticed by me is from 12 to 14 feet above high-water. They consist of blue clay con- taining a few pebbles underlain by yellow sand resting upon the Boulder-clay. The Boulder-clay is of a deep purple colour. As you ‘approach Dawpool the cliff becomes loftier, and has in it a thin bed of intercalated sand and gravel. Further on, again, the cliff was of homogeneous Boulder-clay. I have examined these cliffs at various times; but the sections Q.J.G.8. No. 154. ; 100 T. M. READE ON THE DRIFT-BEDS OF THE differ more or Jess from time to time. In December 1872 I made sections at several points (figs. 16-19). There are in the Boulder-clay sand and gravel seams more or less persistent, ‘Ss sometimes two, occasionally a third, and sometimes only one, or, it may be, ~ ss aseam of finely laminated clay, the clay in oo above the sand seams or seam having EB & 2 blue facings. The lower part of the Sidiay —* ae ee <4 << 2 Si\veass cs : hes ELLY HZ DEAE AR WO ACRE Ww ane AWN A(X nn LE ee a SS ETE SI Aas TE NIA Fig. 26.—Section in St. Bees Cliffs, 1872. (Horizontal scale 1120 feet to 1 inch. Vertical seale 112 feet to 1 inch.) Sea Cote Hotel. Footbridge. Sea Wall. NORTH-WEST OF ENGLAND AND NORTH WALES. TS Fes = 2s SEA RR KR WYK —— coh WS Pa “ai YN a. Soe wars 110 T. M. READE ON THE DRIFI-BEDS OF THE again, changes southwards into sand of a red colour stratified in an imperfectly arched form, P ; lying upon this “ haunch ” of the arch is a large lenticular patch or poeket aE gravel and boulders, Q. A great mass of gravel, with boulders, R, at all angles, now sets in, having a very irregular or contorted surface ; this is succeeded by sand with curved laminated bands,S. The southward termination of this cliff, T, at the stream by Sea Mill is a confused mass of huge erratics and boulders thrown about at all angles. Over the whole of the deposits described lies a mass of irregularly stratified sand, U, containing patches of gravel. The highest part of the cliff is about 70 feet above ete ae mark. Southward of the stream another cliff commences; the basement- bed is here along its whole length a red clay, V, with few stones, evidently the same clay as that at the north end of the section, Lying upon the irregularly arched surface of the clay is an irregular bed composed of sand and gravel, W, full of large blocks _ of sandstone and other rocks; and overlying this and partially incorporated with it is the bed of irregularly stratified sand with gravel bed, U. The total length of these sections is about 2600 yards, or nearly one mile and a half. Remarks on the Coast-Sections. The section at Blackpool has long been considered a typical one by those who think they see evidences of a division of the Drift into three important beds each representing a distinct condition of climate. My own section (fig. 25) certainly shows a clay generally of a harder nature than the upper bed, and containing more stones, with a greater proportion of them striated, and separated from the upper by a great development of sands and gravels, which, however, northwards, die into the Upper Clay in a fork-like manner. If we could see a section further inland, it is highly probable that these two clays would coalesce and shade into each other in places, as, indeed, is shown to be the case in Mr. Binney’s section made twenty years earlier. Mr. Binney himself says that his bed No. 2, ‘‘a brownish-coloured clay containing stones and so many pieces of limestone as to render it unfit for the purpose of making bricks,” is oiten replaced by stratified beds of sand and gravel. If we confine ourselves to Blackpool, it may be perfectly natural to speak of a threefold division; but the moment we attempt to apply the same classification elsewhere we are met by insurmountable difficulties. If we are to classify by superposition, the clay at St. Bees will be Lower Boulder-clay ; but, unfortunately, it does not correspond in any other way with that of Blackpool, being nearly stoneless. Again, at Rampside, we have two sections side by side on the same horizon, one being full of glaciated stones and answering to the usual description of ‘‘ Lower Boulder-clay,” the other corresponding more with what is called “* Upper Boulder-clay ;” but in it are the arched beds of sand. There is no evidence whatever that one clay wraps over the NORTH-WEST OF ENGLAND AND NORTH WALES. PF other. The “‘ Middle Sands and Gravels” are supposed by some to represent “ Interglacial,” and by others “ milder ” conditions, in consequence of the stones being all rounded and unglaciated. If this be the case, what do the sands and gravels of St. Bees represent ? They are even a more distinct deposit, and overlie a lower clay with an eroded surface: but here the conditions are reversed ; for the sands and gravels contain in places more and larger blocks than the ‘“‘ Lower Boulder-clay” of Blackpool, and they are full of contortions and evidences of violent action of some sort. It is noticeable that the Blackpool Drift hes upon an extensive district of the Keuper marls, and Mr. Binney specially notices as “an interesting fact” the ‘quantity of both granular and fibrous gypsum and waterstone from the upper red marls of the Trias ” occurring in bed 4 (the lower clay). I have no doubt that the greater abundance of sands and gravels in the Drift here is due, the first to the abundance of sand-producing rocks in the uplands, and the second, together-with the preponderance of stones in the lower clay, to proximity to the mountains of Westmoreland and Cumber- land. It is also noticeable that in the neighbourhood of the Carboni- ferous hills the Boulder-clay possesses a much greater proportion of rocks that have come from these hills. At Carnforth the shingle and large blocks partake of the nature of the rocks of the valley in which they are situated. At Grange we have another striking instance of the effect of drainage-lines on the nature of the Boulder-clay ; for here it is richly calcareous, being largely made up from the Mountain-lime- stone rocks of the valley of the Kent. At Rampside the Boulder-clay overlies Permian rocks; and it is also far from improbable that the Keuper marls may extend from Fleetwood across Morecambe Bay in that direction. At St. Bees the lower clay has that purple hematitic colour which would lead to the belief that its origin has been local ; and the sands and gravels evidently have been formed from the detritus of the Permian and perhaps Carboniferous sandstones mixed with that from the Silurian mountains of Cumberland. Drirt oF THE Norra Coast anp Movuntarn-Disrricr or Norte . WALES. At Colwyn is a ballast-pit used by the Chester and Holyhead Railway : a very great amount of material has been taken out of it. T examined it in 1872 and again in 1874. In descending order is a bed of yellowish-white clay and shingle, from 2 to 3 feet thick ; red prick-clay with blue facings and containing only a few stones, about 12 feet; gravel and shingle-ballast, about 25 feet ; yellow stratified sand not bottomed, tested to 20 feet deep. These dimensions will not by any means hold good over the whole area, aS by a comparison with my notes in 1872 it appears that the clay lies in lenticular patches interbedded or wedged in with other strata. Thestones of the ballast are all rolled; and among them are granites, most of the stones being of local Welsh rocks. bg BY T. M. READE ON THE DRIFT-BEDS OF THE At Llanfairfechan, on the banks of the river, at about 100 feet above high-water mark, there is a section of Till, containing large angular, subangular, and rounded stones imbedded in a matrix of cream-coloured clay; some of the stones, I noticed, were faintly striated, and others ground flat. They appear to be all local felstones, porphyries, and greenstones. The river, at times, is very furious, having a rapid descent, and carries down large blocks of stone. A few years before 1874 it piled up a mass of stones against the railway-bridge and broke one of the cast-iron girders. On the top of the Till was a patch of gravel and shingle; considerably above this and on the western extension of a moraine-like mound is a gravel-pit (0, fig. 27) showing current-bedded and contorted lamina- tions like the gravel-pit which will be described at Bangor. The mound is very steep on the north and west sides. Crossing this gravel mound we came upon the river again, which has cut ciiffs (EH, fig. 27) I should judge to be about 80 feet high out of a deposit of current-bedded sand, gravel, and shingle which fills the valley. Higher up the river this deposit appears to rest upon the Till; and the same appears to be the case lower down (as at F, fig. 28). There iS an immense deposit of sand, gravel, and shingle in this valley. I could find no shells in any of these deposits. Fig. 27.—Section above Llanfairfechan. ffi 1a Yj" At Penmaenmawr, in an excavation made to repair the sea-retain- ing wall of the railway, I noticed, lying upon the slate rock, a brown clay in which was imbedded a boulder of felspar porphyry; over this was a thin bed of green clay, then a bed of yellow clay. ; Ascending the valley at Aber, we find that the gorge where two streams meet is swept pretty clear of Drift; but above this there is a great deposit of Drift, which rises rapidly, apparently in a series of benches, towards the cascade, where the Drift becomes very stony, being mostly shingle and boulders of local rocks—porphyries, green- stones and felstones. All the way to the head of the valley the rocks are Silurian shaly slates. There are a great many greenstones and other boulders of great size lying on the sides of the valley, usually where it is denuded of Drift; but none that I saw were NORTH-WEST OF ENGLAND AND NORTH WALES. ES glaciated. It is a curious fact that so few stones are glaciated in these upper valleys, and so many in the lowlands. The slaty rock of the hills on either side protrudes in great splinters and pinnacles, showing that if it ever had been smoothed down by ice, great denuda- has since taken place. The Drift in the gorge consists of clay and stones. It is evident that a tremendous amount of Drift has been removed by denudation. At Bangor, west of the station, the railway-cutting shows lamin- ated current-hedded and contorted gravels and sand lying upon the upturned edges of the Silurian rocks; in places at the base is packed angular débris, and there are some beds of laminated clayey loam in the gravel. These laminated clays are also to be seen in the ballast- pit below the station. At Ogwen Bank in the bed of the Ogwen are magnificent glaciated corrugations in the direction of the stream ; but they are disconnected, as if the stream had roughly quarried out some of the rock between. Further up the stream is an inclined flank of slate rock splendidly polished, with perfectly straight and parallel striations running across it diagonally ; and lying against this rock is a laminated con- torted gravel-drift containing large blocks at the base. On the opposite side of the stream this gravel-drift is overlain by a grey Drift containing angular pieces of slate and also rounded boulders, and over all is a cover of alluvium. East of Beaumaris, near the Menai Straits, is a cliff of Boulder-clay, of a brownish grey colour generally, containing large boulders, and capped with a reddish clay with pebbles from a great variety of rocks. The Boulder-clay contains a great many large Mountain- limestone blocks, some being smoothed and glaciated. Shell-frag- ments are also to be found, but sparsely. Beyond this section, towards Puffin Island, are several lower cliffs of Boulder-clay. On the railway by Menai Bridge station, in a cutting for a siding, in 1872, lying upon a shaly rock of the Carboniferous series was a black laminated band, then a yellow one, and over this a grey-coloured laminated band, all composed of fine scales of shale beautifully arranged in lamine. Capping it all was a Boulder-clay, varying in colour from brown to yellowish white and blue, evidently from the Carboniferous shales and clays. Near Menai Bridge, and on the level of the last section, was a brown clay containing large angular blocks of a coarse gritty sandstone (Carboniferous). The railway from Caernarvon to Pen-y-Croes, soon after leaving Caernarvon, is cut througha boulder-drift entirely composed of stones. At Dinas there is a cutting, about 24 feet deep, through Boulder-clay very full of stones; and from this place to Pen-y-Croes are frequent cuttings in Drift composed wholly of boulders and gravel. AS Pen-y- Croes are mounds of fine sandy gravel ; and a further inspection shows that there is a broad expanse of gravel cut up by streams into undu- lating outlines and mounds, filling up and spreading round the valley- mouth. This gravel continues across Caernarvonshire to Afon Wen. There must have been a vast denudation to supply all this material together with that which has been washed away. 114 T. M. READE ON THE DRIFT-BEDS OF THE At Brynker station a deep bed of this gravel (a, fig. 29)1s seen to rest Fig. 29,—Section at Brynker Station. ee ; + : Qs peter YH 07 Zi a. Gravel. saat upon a clay (6), which is to all appearance true Till. It is quite evi- dent that the esker-like mounds of gravel have been cut out of a vast plateau of Drift. Moel Tryfaen, so celebrated for the marine shells first found by Mr. Trimmer on its summit, is a rounded mound-like plateau, forming a spur of Snowdon. It is evident that the Tryfaen gravels are but an upward extension of those on the valley. Lying on the slate rock of the Dorothea Quarries, Nantley valley, in 1872, I noticed the following section (fig. 30)—purple slaty drift (2) lying on the slate rock (1) in a thin patch, and over it laminated current-hedded eravels overlain by a coarse gravelly drift (3), the total thickness in the deepest part being about 45 feet. Fig. 30.—Section at Dorothea Quarries. 1. Slate rock. 2. Purple slaty drift. 3. Coarse gravelly drift. Ascending from Pen-y-Orredd, you pass over the shoulder of Tryfaen ; many sinkings for new quarries were being made, which showed a great thickness of drift. There are no terraces on Tryfaen. At the Alexandra quarry, just on the summit, is a bed of current- bedded laminated sand and gravel containing large boulders. At about 12 feet from the surface I took out a glaciated stone ; the rolled gravel contains travelled granite and even flints. The larger stones appear to be local. The Drift has been described by Lyell: it lies upon the eroded edges of the slate rock; and from it Mr. R. D. Darbishire, F.G.S., collected the shells a list of which is given in the table, Part I. of this paper. The level is roughly 1400 feet above the sea. NORTH-WEST OF ENGLAND AND NORTH WALES. 115 In the cuttings of the railway from Bettws up the Dolwydellan valley, which I inspected in February 1876, while under construc- tion, small patches of stratified gravel are here and there intersected ; but, as a rule, very little drift was met with. About three quarters of a mile from Bettws a cutting in the slate rock disclosed a singularly smooth surface which had been protected by a cover of drift ; it was not planed down, but was in undulating dimples and hollows, smoothed as if ground with polishing paste. This was very instructive, as where the rock was unprotected by drift on the same shoulder, close by, without any transition, the rock was-jagged and irregular. On a rock near to the mouth of the tunnel, but not on the line of railway, towards the summit-level of the railway, some singular markings were to be seen. ‘They appeared as if gouged out at all angles ; I have never seen any thing like this before or since. Fig. 31.—Section on the Rivals Mountain. a. Slate rock. 6. Drift. In inspecting some excavations made for a sett quarry near the Rivals mountain, I was struck with the jagged nature of the rock underlying the drift (a, fig. 31), the very antithesis of that I have described near Bettws. The drift 6 was compact and solid argilla- ceous matter or Till, full of rough stones intermixed with angular and subangular ones: some of the stones were erratics ; but most con- sisted of the local felstone porphyry. 1 must bring this account of the Welsh drift to a close by a description of the section disclosed by the excavation of the puddle-wall of the Rhyl Reservoir® (fig. 32). Fig. 382. Rhyl Reservoir, Nant Uwydd ; section of puddle trench. (Horizontal and vertical scales 220 feet to 1 inch.) Ww Intended water vine Here we have :— A. Bed rock, Denbigh Shales, getting more slaty in structure towards y. The floor isquite level ; but the beds dip up the valley at an angle of about 12°. The sides are stepped. B. Fine waterworn gravel, angular and larger near the bed-rock. * Tam indebted to Mr. H. C. Beloe, the engineer of the works, for this in formation. Q.J.G.8. No. 154. K 116 T, M. READE ON THE DRIFT-BEDS OF THE C. Bookleaf clay, greyish brown in colour, containing a level bed of sand (shown by dotted line). The beds are perfectly level, containing no stones or gravel. The interlaminations were as if dusted with sand as fine as emery powder. D. Brown clay of very fine texture ; it contained some few glaciated blocks of white limestone. The top of the bed is irregular. E. Blue clay, powdery, full of slate fragments, some long and polished like a hone, most of them much striated along their longer axis. No limestone was found in this bed. The junction with the bed below is obscure; one shades into the other. Nodules were found in this bed. A patch of white clay with gravel occurred between the brown and blue clay at h. F, Alluvium composed of the washings of all the other beds. A cutting in the direction of the valley to drain the brick-croft, showed a rude bedding of pebbles a few feet below the surface. The brown clay in the brick-croft is about 7 feet below the surface; but it varies much in depth, and, I was informed, rested upon gravel. In the alluvium a good-sized oak was met with. Vyrnwy Water-works, Oct. 2, 1882. Since the preceding was written, I have had the opportunity of examining with Mr. G. F. Deacon, joint engineer with Mr. Hawksley, the trench excavated for the construction of the masonry dam of the Vyrnwy reservoir which is to supply Liverpool with water. It is not exactly in the district I am describing, as the watershed really contributes to the supply of the river Severn; but I give a description of it as helping to throw light upon the subject of the mountain-drift. The trench, when I saw it (Sept. 28 and 29), was opened out across the valley through the drift to the rock about a distance of 700 feet, or the whole way across the bottom of the valley (fig. 33). The trench was over 120 feet wide at the bottom—an ‘open cutting with- out timbering; and such an opportunity of examining a valley- section may never occur to me again. . : The river Vyrnwy is one of the affluents of the Severn; and the gathering-ground above the dam, about 7 miles S.E. of Bala Lake, contains 23,500 acres. The valley above the dam is remarkably flat- bottomed; for with 80 feet depth of water at the dam a lake of 1115 acres area and 5 miles long will be formed. The bottom of the valley opposite the Hirnant tunnel, which is to tap the intended lake, is 760 feet; and the top water of the lake will be 825 feet above ordnance datum. But one of the most interesting facts made evident by the works is that the alluvium and Drift forming the plain of the valley lie in a true rock-basin ; for borings through these deposits proved that the rock inside what will ultimately be the lake or reservoir, is con- siderably below any part of the rock in the bottom of the trench. iid FF NORTH-WEST OF ENGLAND AND NORTH WALES, Fig. 33.—Section of Trench for Masonry Hinbankment, Vyrnwy, 1882. Level of road across embankment, Top Water Lever Spas eS : 300 oF EEEY Oross Section at A. Cross Section at B. Surface ground. trench Surface of ground. N.W. a 5 Scale of Cross Sections, horizontal and vertical, 100 feet to 1 inch. — ' es —— eee wa = 118 T, M. READE ON THE DRIFT-BEDS OF THE Description of Sectron (fig. 33). Down valley side of trench. 1. Alluvial gravel and shingle. 2. Gravel. 3. Interlaminations of clay, sand, and gravel. 4. Gravel cemented together with iron oxide. 5. Shingle and gravel confusedly bedded. 6. Shingle and gravel, and Tilly matter intermixed with angular, rounded, and well waterworn boulders of considerable size. In places it contains large masses of slaty rock up to 100 tons weight, torn from the valley-bed. 6a. Similar material at side of valley. 7. Grey Till. 8. Blue hard Till containing angular and subangular boulders, mostly of the slaty rock of the valley. It is much more perfectly developed on the sides of the valley than in the bottom of the trench. There are, I believe, in this Till, as well as in Nos. 7 and 6, grits that have come from the mountains at the head of the valley. Some of the stones are scratched and striated, but not nearly so distinctly as those found in the Low-level marine Boulder-clays. The opposite side of the trench shows a somewhat different ar- rangement of the preceding beds, due apparently to currents, some of the arched beds of sand and gravel sweeping from near the bottom of the trench to the top. No. 9. Bed rock (summit of Llandeilo) ; a blue slaty rock with no proper cleavage; dip, measured on a plane of bedding 20 feet square, 38° N.W. The beds cross the bottom of the trench diagonally. Huge blocks have been displaced from their beds, and pushed up the plane of the bed below, leaving a cavity at the joint through which a man might walk. The mode in which the rocks have been forced out gives a jagged appearance to the bottom of the trench ; but the angles and asperities have been worn off, and there is a roche moutonnée in the middle. The striations are not very marked; they run magnetic N. & 8S. Mr. Deacon informs me that in cutting the trench at the side of the valley for the temporary diversion of the river a mass of rock of 300 tons weight was met with resting upon the rock below and imbedded in drift. yi Cross sections A and B (fig. 33) exhibit accurately the form of the rock surface at the bottom of the trench. Their direction, of course, corresponds with that of the valley, the beds dipping the op- posite way to the slope of the valley. Remarks on the Welsh Sections. — The effect of the nature of the rocks in each drainage-basin on the character of the drift in the same basin is quite as conspicuous in the examples I have quoted in Wales as it is elsewhere. The sands and clays of Colwyn are evidently derived from the Triassic rocks of the Vale of Clwyd. At Llanfairfechan and Aber the drifts, entirely different from that at Colwyn, are no less clearly traceable to the rocks of the valleys in which they respectively lie, the ar- rangement and size of the boulders and the character of the drift being also affected by the steeper gradients of the mountain-streams and slopes. Near the coast-level the Drift usually partakes more of the character recognized as marine. Distribution of the varius kinds of Rock-fragments throughout the Drift, and the Light they throw wpon tt. With the kind assistance of Professor Bonney I have been enabled NORTH-WEST OF ENGLAND AND NORTH WALES. 119 to trace with a high degree of probability several varieties of rock- fragments back to their places of origin. The Eskdale granite * is found in almost every position and at almost every level, from the Low-level Boulder-clay of the plains of Lancashire to the summit of the Moel Tryfaen in North Wales. It occurs both in large blocks and in rounded pebbles. It is found on the Macclesfield Hills and in the whole of the deposits skirting the coasts from Liverpool to Ulverstone, over a large part of Cheshire, and in the drifts skirting the coast of Wales. The syenite from Scale Force, Buttermere, is also of pretty general distribution ; I have found it in most of the localities about Liverpool, and in considerable blocks on the hills above Poynton and Maccles- field, up to 1200 feet above the sea-level at the Bow Stones. The Carrock-Fell syenite is a very probable identification. Ihave found it in several localities about Liverpool. There are also porphyritic felstones, probably from dykes by Scawfell. There is a large residuum of rocks of the Silurian series, which, though not individually traceable, assuredly largely came from the Lake district; and a remaining series of rocks, granites, felstones, diorites, and Old Red Sandstone, that probably haye come from the south of Scotland. Although it is thus seen that the individual identifications are only of different degrees of probability, the igneous and sedimentary rocks below the Carbo- niferous series, as a whole, are preponderatingly from the Lake district. The Carboniferous rocks, looking at the proportions in which the various members occur, and comparing them with those found in the Drift in the Ribble valley and the valley of the Kent, are unmistakably from the Carboniferous rocks forming the great Penine chain to the north and north-east of the actual localities in which the drift-specimens are found. Mr. Patrick Dudgeon of Cargen, Dumfries, has kindly gone over the rock-specimens which both Professor Bonney and I suspect came from the south of Scotland. He has identified eight of the speci- mens of grey granite as certainly from Criffel. This granite con- tains crystals of sphene and allanite; and he is not aware that these minerals occur in any other granite nearer than Aberdeenshire and Sutherland. The Dalbeattie granite is very similar in composition, as regards the proportions of hornblende, mica, and felspar it con- tains ; sphene crystals are also found in it, but not allanite. One of the specimens is of a granite found at Kirkconnel, about seven miles south of Dumfries. Five other specimens he considers probably to be Criffel. Three specimens probably came from granite-veins found in the district. Some of the Old Red Sandstones and Silurian rocks he considers very probably came from the district about Dum- fries; but they are more difficult to identify yt. — * Mr. Mackintosh was the first to trace the Eskdale and Criffel granites over a large area. + Mr. Goodchild mentions the occurrence of 2s from Dumfriesshire and Kireudbrightshire in the Eden valley (Q. J. G.S. 1875, pp. 66-7) ; also of Carrock-Fell and Buttermere syenite (ibid. p. 81). 120 T. M. READE ON THE DRIFT-BEDS OF THE The Triassic sandstone, Keuper marls, and fibrous gypsum, as a rule, cannot have travelled very far; but I have one or two sand- stones that may have come from St. Bees. I have found one or two pieces of what appears to be hard chalk ; and there are occasional flints found in the drift, and even on the top of Moel Tryfaen. One flint pebble I took from the Boulder-clay in Sefton Park is reduced to a plane at one end; but whether this is the effect of grinding I cannot tell. It appears most probable that these have come from Antrim. If these identifications are correct, it will be seen that all the stones are confined to the basin of the Irish Sea and the river-basins flowing into it, excepting some stray stones that may have come from the Highlands of Scotland. It seems curious that such littie patches of granite should have yielded such a harvest of blocks ; but it is noticeable that the largest patch, viz. the Eskdale, has fur- nished the greatest quantity. Probably the reason why the granites, syenites, and other igneous rocks occur in a larger proportion in the Drift than would seem to be due to the area they cover wm situ, is that they naturally break out in larger blocks, and, moreover, they are generally found at a high level. ; No Shap-Fell granite has ever been found by me; nor have I ever heard of its being found on the west side of the Penine chain—another fact proving that the erratic rocks of the area under consideration are confined to the drainage-basin of the Irish Sea. This fact seems to me fatal to the idea of an ice-sheet overriding the great water- sheds, and points to a system of glaciers radiating from mountain- nuclei. The distribution of the erratics, as described, seems unac- countable on any theory excepting that of their being sea-borne. CLASSIFICATION AND CONCLUSIONS. It now remains to consider the bearings of the facts I have de- tailed, and to attempt to account for the varied and complicated phenomena they present. The first part of this paper was published in February 1874; so I think it will be readily conceded that whatever errors may attach to my views are not due to the haste with which they have been adopted. As my desire throughout has been to form my opinions from the facis as seen by myself, while studying the many very valuable con- tributions that have appeared from other observers, I have purposely adopted a system of natural description rather than of geological labelling, so that I may place my readers as nearly as possible in my own position. PRE-pDRIFT SURFACE OF THE LAND. I must distinguish the Pre-drift surface of the land from the Pre- glacial: the former is the surface form of the rocks if the drift- covering were removed; the latter is to a great extent a matter of supposition. NORTH-WEST OF ENGLAND AND NORTH WALES. {21 In every case, without an exception that I can remember, other than in the gullies presently to be described, wherever in the neigh- bourhood of Liverpool the covering of Boulder-clay has been re- moved, and the underlying rock has been of a nature capable of receiving and retaining such impressions, I have found it planed and striated. The striations are more or less in the direction of north-west, ranging usually from N. 40° W. to N. 15° W.* = It cer- tainly seems as if an immense mass of ice had moved continuously over the country ; and though the theory of an ice-sheet radiating from the mountain-districts of Cumberland presents the most feasible explanation, yet it is not without its difficulties; so that some local geologists are inclined to revert to the iceberg and field-ice theory for an explanation. The portions of the rocks that have no cover are weathered, eroded, and worn down to a very considerable extent. But the most interesting feature of which we have proof is that the surface- form of the Drift does not always give an indication of the surface- form of the rock beneath. I have already described the Preglacial valley of the Mersey; and it is difficult to resist the conclusion that, if we could lay bare the beds below the Irish Sea, we should find a system of river-beds ramifying and uniting into one great river dis- charging into the Atlantic; and it is quite possible that this may have discharged northwards, between Scotland and Ireland, as there is off the Wigtonshire coast, in the words of Captain Beechy, “‘a remarkable ditch, upwards of 20 miles long by about a mile only in width, in which the depth is from 400 to 600 feet greater than the general level of the bottom about it.” These facts are quite in con- sonance with the relations of the British Isles to the continental area of Europe. We are on the edge of a great plateau; and every valley and mountain on the western coast of Scotland, where they touch the sea, gives evidence by its outline that the country has been submerged. If we pass over to Norway the same holds good ; every thing points to a former and Preglacial or Glacial greater elevation of the country; for the gullies which I have described cut out of the solid Triassic rocks could have been worn down by naught save subaerial river-action ; and to get this a very considerably greater elevation must have optained. It will have been observed that the bottom of the buried gullies, where they have been ex- amined, as at the Rivington Reservoir, on the Yarrow, in the. North Docks and the Rhyl Reservoir, all show the action of running water, not that of ice. When I took the Liverpool Geological Society over the North- Dock excavations in 1876, Mr. Lyster, the engineer, drew my atten- tion to the remarkable evidence of water-action on some of the rocks at the bottom sloping towards and flanking the river. It has long appeared to me, from the facts I have observed, that the period of greatest glaciation was one of considerable elevation. * A list of glacial stria found in 8.W. Lancashire and in Cheshire is given by Mr. G. H. Morton, F.G.8., in the Proc. of Liverpool Geol. Soc. 1876- 77, pp. 292, 293. 122 T. M. READE ON THE DRIFI-BEDS OF THE The existence of these riverine gorges in the solid rock so far below the present level of the sea is as certain a proof of former elevation as the marine beds on the top of Tryfaen are of depression. All the surrounding facts lend force to and consolidate that conclu— sion. If this be so, it is the more easy to understand the existence of a snowfield on what are now plains little above the level of the sea. If there ever existed this extension of ice from the mountain-centres (for I have observed no facts to warrant in the slightest degree the idea of a Scandinavian ice-sheet extending over our island), it must have melted back and separated into local glaciers before the drift-deposits I have described were laid down. During the progress of the great submergence it is as clear to my mind as any thing can be in so difficult a geological subject, that no ice-sheet existed in any part of Lancashire that I have examined*. I have studiously confined these speculations well within the boundary of personally ascertained facts. Rep Sanp AND RUBBLE DEBRIS OF THE TRIAS. When the underlying rock is not planed and striated, it is usually covered with broken rock and packed rubble, graduating into red or yellow sand, according to the colour of the rock beneath. In no case that I have seen (and I have devoted much attention to this point), where the sand is undisturbed, does it contain any erratic pebbles or stones, nothing more than a half-imbedded. boulder. In some cases a considerable amount of this sand is stratified and evidently rearranged by sea-action; and it then may contain both shell-frag- ments and erratic stones. In the description of the section at Bootle- Lane Station (Part I. p. 27), I have called this sand ‘“ Ground-mo- raine equivalent of the Scotch Till.” At the time it was written I had not seen the Scotch Till, but have since had the good fortune to see a section in the new Cartsdyke Dock, Greenock, of the “ Till” resting upon red sandstone of the Calciferous series, in which the red sand and rubble is interposed between the rock and the Till in a manner precisely similar to what occurs between our marine Boulder- clay and the Triassic rocks*. That this formation is the product of land-ice, I think is the most feasible supposition, though this explanation is not without its diffi- culties, as the same sort of beds sometimes arise from subaerial decay. But I cannot conceive how icebergs or field-ice alone could have so universally polished the rocks of our Lancashire plains, seeing that, if so, it must have been done on a sea-bottom covered with Boulder-clay, sands, or gravel. Nor have I in one single instance with the most careful search found any disturbance in these red * The late Mr. Clifton Ward has expressed views on the glaciation of the Lake district, with the moderate character of which I am disposed to agree. (See Q. J. G. S. vol. xxix. p, 422 and vol. xxxi. p. 152.) t This is described in my paper “On the glacial Beds of the Clyde and Forth ” (Proce. of Liverpoc] Geol. Soc., Session 1879-80). NORTH-WEST OF ENGLAND AND NORTH WALES, 123 sands traceable to any thing similar to the grounding of ice in any form, though they have rested on rock in immediate proximity to a polished surface. As a rule, the Boulder-clay rests upon the sand as if deposited in the most quiet manner. GuLLy-GRAVELS. The ramifications of gullies below low-water mark which I have described as existing in the substratum of rock beneath the Drift are usually filled with rolled gravels. This is not invariably true, as, if deep, they sometimes contain gravels aud sands, or, as in the case of the valley of, the Mersey at Widnes, clay. Tosome extent their presence may be dependent upon the rate of inclination of the bed of the gully. The gravels are of rocks similar to the stones found in the Drift. It is also not an un- common occurrence to find gravel and shingle resting upon the rock under the Boulder-clay*, or, more rarely, in patches within the lower clay. It is readily conceivable that, if there was any tendency to the accumulation of gravel on the sea-bottom, it is just at these gullies kept clear by the rush of fresh water down them that we should expect to find gravel and shingle. The invariably rolled character of these stones seems to point to their accumulation at or near ‘‘between tides.” I have, however, never seen any arrangement of the gravels that could be distinctly described as littoral; probably increasing submergence brought them under the influence of other currents which disturbed what was originally a littoral arrangement. It is difficult to draw a distinct line of demarcation between these gravels and the Drift above, as in some sections they alternate with beds of sand and clay. At Hooton Station, Cheshire, there was a great depth of them, and also at Fidlers Ferry, near Warrington. And lately a boring at Halewood, about 2 of a mile from Hunts Cross station, and not far from my line of section of the Liverpool Extension Railway (fig. 6, p. 90), showed a depth of 137 feet of Drift. The surface-level was the 100 feet contour, the bottom bed being gravel 8 feet thick resting on Keuper Marl unbottomed at 417 feet. They underlie the Boulder-clay in the bed of the Mersey in many cases where it has been exposed by dock-excayations, and often show distinct signs of current-bedding. I see no reason to infer that these gravels have been derived from any preexisting Boulder-clay. If they had been, or if the Boulder-clays were remaniés, some vestiges of the original Till would surely have been before this discovered. I have searched in vain for it. Low-LEvEL BovuLDER-cLAY AND SANDs. These represent by far the greatest bulk of the drift-deposits of the north-west of England. . It was to these beds that my attention was at first more particularly directed; andin Part I. I have described * The excavations of the Atlantic Docks proved that these gravels lie in stream- like patches on the rock (see plan of Docks, p. 86). 124 T, M. READE ON THE DRIFT-BEDS OF THE a typical section at Bootle Lane, as well as given a list of the shells and shell-fragments found in them by me, amounting to 44 species ; this list has been increased by Mr. Shone by his finds at Newton, Cheshire, to 56 species ; 16 species of Foraminifera were also deter- mined by Mr. D. Robertson of Glasgow, out of one sample of clay I sent him. Ostracoda also occur, and other minute reliquie of the sea*. My object in these preliminary investigations was to ascertain ifthere were any organic remains by which the drift-beds might be separated into geological horizons—because,if, as some maintain, two glacial and one interglacial period are represented in these beds, there ought to exist, a priori, some decided distinction in the molluscous fauna. I utterly failed to detect any; nor isit maintained that any subsequent observations have succeeded, though the observers have been many and zealous, and only anxious to make the discovery. The facies of the fauna is of a more northern character than that of the existing Mollusca of the British seas, the typical shell being Astarte borealis, which I found in almost all the localities, though Turritella terebra is the most abundant. I do not place much reliance upon the paleontological evidence in this case, either as determining age or climate. The Clyde laminated brick-clays contain an assemblage perhaps more distinctly northern than that of my list; yet these laminated clays can scarcely be consi- dered other than Postglacial, and are certainly of a date long posterior to the beds I am now describing. I have therefore ceased to attaca much value to this kind of evidence unless backed up by facts of another and more complete kind +. The most distinct and reliable signs which justify us in calling these beds Glacial are the numerous striated and planed erratic blocks, boulders, and pebbles they contain. I have made a large collection of these included rocks from various localities in the neighbourhood of Liverpool ¢. ‘ These various rocks occur promiscuously through all the beds of the Low-level Boulder-clay, their proportions: changing in different localities ; so that I have found it impossible to separate the beds by the included stones. In the Drift of the upper part of the Ribble valley an order of arrangement vertically may be made out, as I have already described, but never in the true sheet of Low-level Boulder-clays and sands. The larger blocks (say, weighing upwards of a ton) are usually fluted in the direction of their longer axes, and are convex in cross section. * And in 1882 in samples of clay sands and gravels from the North Docks and Atlantic Docks, described by Mr. Robertson in the Appendix. Tt Mr. Searles Wood has brought forward considerable evidence to prove that the east-coast beds are older than the west-coast deposits; but my faith in the inferences drawn from the contained shells is much shaken. The east-coast glacial deposits are much more disturbed than most of those described in this paper. On the other hand the erratics of the west coast are more distinctly planed and striated. + Much valuable information on the subject is to be found in Mr. Mackin- tosh’s paper “ On the Limits of Dispersion of the Hrratic Blocks of the West of England,” Quart. Journ. Geol. Soc., 1879, pp. 425-455. =< ~ NORTH-WEST OF ENGLAND AND NORTH WALES. 125 Smaller boulders. are often ground to a perfectly true plane, and sometimes without striations. In others the stone presents several distinct planes or facets with striations meeting in a herring-bone fashion ; others are irregular in form and indefinitely scratched all over as if bumped about, or rolled over, under ice. Sometimes boul- ders occur aggregated in nests or pockets in the clay. It is quite evident that these stones have never been disturbed since they were dropped in the mud of a glacial sea *. The proportion of the contained stones to the bulk of the deposit varies very considerably. Some of the clays are decidedly stony; and others consist of a great mass of unctuous clay with very few stones, but with a large percentage of fine gravel intimately intermixed with it. Probably half of the stones taken out of the clays in the neigh- bourhood of Liverpool used for brick-making are more less scratched ; and it is worthy of remark that by far the larger proportion of the ex- amples I have collected of specially well glaciated stones are Silurian grits or other old rocks from the mountain-districts. Butin beds of gravel and sand the stones are usuallyrounded. This, however, can hardly be an evidence of an interglacial climate; for we findin many cases, as I have detailed, thatthe lower and harder bedsofthe clay often, nay, generally contain more rolled stones than the upper or middle. Such is the case at Dawpool; and itwas so in the Atlantic Docks, Liver- pool. At Blackpool, on the contrary, according to Binney, the lower and harder beds contain the greatest number of striated stones. It is therefore clear that these distinctions cannot indicate climatic differences. And if this be admitted, can we on these grounds consider the lower bed of Boulder-clay as a geological subdivision? The true ex- planation seems to be that these stones have been rolled on the beach before being finally deposited on the muddy bottom of the Glacial sea. When beds consist solely of gravel, boulders, and shingle, that, in itself, is a proof that either tidal currents or shore-conditions have prevailed in the places where they have. been laid down, or that they have been at a depth no greater than where the wind waves can act upon and move them. ; The absence of stones in some of the sand, as in the section de- scribed by Mr. Shone, near Chester, it appears to me, can be accounted for in this manner. Beds of clay must necessarily be very slow accumulations. Bedsofsand, on the contrary, within reach of the tide are constantly moving ; as an illustration, the banks and chanuels in the estuary of the Mersey are constantly being surveyed and the chan- ges marked by buoys by the marine surveyors. Hundreds of thousands of tons will be shifted by a single gale. If there were floating ice conveying stones and depositing these in the estuary now, what would be the effect? They would be sorted by the currents, we should have shingle in one place, sand only inanother, * Dr. James Geikie has expressed the opinion that these deposits are sea- bottoms ploughed up by land-ice (“The Intercrossing of Erratics,” reprint from the ‘Scottish Naturalist’). 126 T, M, READE ON THE DRIFT-BEDS OF THE and in some sands and gravels. It must not be lost sight of that, during the submergence and emergence of the land, every river- valley at every point has been at one time or other an estuary. To expect under these circumstances all beds to be similar would be to expect that which could never happen. Sand is found in pockets in the Boulder-clay as well as in horizontal beds ; and these often assume very curious shapes ; they also seldom contain stones, though the clay surrounding them does. The sands and gravels at St. Bees which are in great force are distinguished by contorted bedding, confused aggregation, and the great number of included blocks and boulders. The shingle, gravel, and sands are intruded or folded into each other. The underlying clay is peculiarly free from stones. If we found geological subdivi- sions on such grounds, what are we to do with these beds? No one has even suggested an explanation. As assisting to explain these Low-level Boulder-clays and sand beds so puzzling to observers, I may here state that I have not seen in the whole series one single example of well-defined shore-conditions. No continuous section of any length ever displayed the horizontality distinguishing most littoral deposits. They are all more or less arched, as my sections show ; this is especially remarkable where continuous thin beds of sand like that near Farnworth extend for a long distance. I infer from this circumstance, as well as from certain collateral pecu- liarities already described, that the beds from base to summit sub- stantially represent a sea-bottom, the conditions of depth altering as the land slowly subsided or reemerged. It is the circumstance that the beds represent conditions that are not within our ken, as littoral deposits are, that has, I believe, led to so much confusion. In a paper “ On Tidal Action as a Geological cause” (Proc. of Liverpool Geol. Soc., 1873-74), a collateral investigation I felt it incumbent on me to make before I could properly attack the problem presented by the Glacial beds, I have shown that the tide-wave, unlike the wind-wave, acts at the greatest depths ; and among other examples I have quoted that of the excavation or keeping open of the “ditch” before referred to opposite the coast of Kircudbrightshire. Other examples of almost equal force could be given. My conclusions are that the long continuous beds, to some extent arched, are the effects of tidal arrangement below low water; the horizontal lami- nations of the clay occasionally met with I attribute to the same cause, and the short arched book-leaf laminations also. I have shown in Part I. that the shell-fragments found distributed through the clay are, as a rule, only of such a size as a moderate current could convey. The much-contorted and cross-laminated beds of which the St. Bees sands and gravels are the best examples, | am of opinion were laid down partly above and partly below low water, like our estuary sandbanks within reach of the wind-waves. It is also quite evident that as the depth of the sea changed the direction of the tidal currents would change also. Thus what was a hollow in clay might become filled with sand, which, again, might NORTH-WEST OF ENGLAND AND NORTH WALES. 17 have clay laid upon it as the bottom further subsided. Or in the direc- tion of the greatest rush of the tide shingle might form, as is the case now in places in the bed of the Irish Sea and English Channel*. Bearing these facts in view, let us see if we can apply them to an explanation of the peculiarities of the deposits I have detailed. The first-formed deposits were undoubtedly the gully gravels and sands ; the next were the beds of gravel, current-bedded, lying at the base of the Boulder-clay. As the land further subsided the lower clay began to form, and with it became mixed the red sand washed from the previously degraded Triassic rocks—the mixture of materials being calculated to form a hard, short, red clay. On still greater subsidence these clays would be brought within the reach of cross tidal currents, creating those lines of erosion often mis- taken for lines of unconformity representing geological subdivisions ; and on this a greater or less bed of sand might or might not be deposited. The deposition of more plastic clays would then set in; but there might be intermediate beds of sand still laid down. Ata certain depth of iminersion the deposits would reach their maximum, one element being the extent and nature of the land still unsub- merged. But while this submergence was still going on, part of the deposits would be washed in by the sea landwards, to be again redistributed as they came successively under the influence of the waves and tide. When the submergence reached its extreme limit the amount of sediment contributed by the land to the ocean would reach a minimum. As the land rose again out of the waters, something of a similar kind would happen in inverse order, only the rivers would reexcavate the drift which had been deposited in them. I have already pointed out that the character of each basin is influenced most profoundly by the nature of the rocks within it, so that in different areas or basins diverse beds will mark the same stages of submergence. If I have succeeded in establishing this point, it follows as anatural consequence that the bulk of the drift- deposits of these low-level plains have been formed when the seas were comparatively shallow, probably at depths of from 100 to 300 feet. The shallower the sea the more local the deposits ; for as the water-partings of the basins became submerged the nature of the deposits would to a larger extent be due to the mechanical distribution of the tides and waves. It would occupy too much time to attempt to explain each bed on these principles; but I am fully satisfied they are susceptible of it. If my explanation be correct, it is evident that the successions of beds are local phenomena, and that no geological subdivisions can be founded upon them. MovunNTAIN- AND Hiti-Drirr. I have before remarked that the distinguishing feature of our * «Tidal Action as a Geological Cause,” Proc. of Liverpool Geol. Soc., Session 1873-74. 128 T. M. READE ON THE DRIFT-REDS OF THE mountain scenery, as relates to our present subject, is, as compared with the plains, the absence of Drift. This, perhaps, seems what we might naturally expect from the Postglacial denudation which it has since undergone. But in the region of lakes we find lakes still; and one would expect that if much material had ever existed in the drainage-basins in the form of Drift, it would in travelling down streams be caught in the lakes; and hence we might expect them to be now filled up. There are no means of estimating the actual amount of materialthat has thus been entrapped ; a series of borings would be required to do this. But I would point out that the whole of the material brought down is not deposited in the lake ; on the contrary, we find the shores of lakes pretty generally formed of shingle and gravel. The wind-waves must move this ; and doubtless in most cases there is a general progression of materials along the shore, and eventually down stream from the outlet. This may account for some of the material; and the generally limited area of the drainage-basins in mountain-districts will also tend to explain the absence of great quantities of Drift. It will be seen, and it is in accordance with the theory here put forth, that the larger the drainage-basin in which the Drift lies the greater is its quantity. In this way it is that the plains of Lancashire are so drift-covered, while the hilly tracts are to a large extent driftless. The Drift met with in the mountain-valleys I have described is usually either true Till or boulder- and gravel-drift, when found together, the latter usually overlying the former. The Till most probably has been formed under or in front of the local glaciers during their recession. That the ice lingered longer about the mountains than elsewhere is proved hy the iceberg-borne erratics of the Lancashire Boulder- clay. Consequently, while the Drift of the plains was being formed, the greater part, if not the whole, of the materials denuded from the mountains and valleys was pushed forwards by the glaciers and protruded into the sea. I have shown how at the Atlantic Docks the majority of the large erratic stones, deeply grooved and worn, are in the upper bed of plastic clay. The boulders in the lower beds are more frequently rounded. This fact corresponds with what would happen under the sequence of events the theory assumes. The larger stones would not be conveyed away until there was depth of water sufficient for the flotation of the iceberg. But I have shown that an enormous amount of material has come down from the Welsh mountains in the form of boulders and shingle, which is distributed on the plains into which these valleys debouch ; and it is extremely probable that much of it has been moved to its present position during and since the emergence of the land from the sea. ConcLUsIon I fear that ‘my views will not appeal to the imagination in so lively a fashion as do some at present in vogue; but of this I am sure, I have been animated by the sole desire of seeing things as NORTH-WEST OF ENGLAND AND NORTH WALES. 129 they are. Complicated and inexplicable as the features of glacial geology are when viewed separately, I have attempted to show that there are recognizable features in common, running through all, that point to geological agents in the past not very dissimilar to those we now witness. Then as now, but from a more limited area pro- portionally to the amount of submergence, the Thames, the Mersey, and the Severn, and innumerable other streams and rivers, brought down their quota of material to the beds then forming at the bottom of the sea. APPENDIX. Report on the Sands and Gravels'and Boulder-clays and the top Silt, at the Dock F of the Atlantic Docks, Liverpool. By Davin Rozertson, Hsq., LS. PGS. No. 1. Sand- and gravel-bed at the bottom of the section is the most puzzling of the whole, as it contains a few freshwater Ostra- coda and numerous fragments of what appear to be marine shells, but so small and waterworn that no character is left by which they can be satisfactorily identified ; yet, from various peculiarities of the fragments, I have little or no doubt that they are marine. It is, however, singular that no Foraminifera were detected in it, or- ganisms that are in most cases present in marine deposits, even where no other animal remains are to be seen. The material consists, in round numbers, of 70 per cent. of fine sand, with very little mud, and 30 per cent. of gravel. Many of the pieces are distinctly striated and well rounded and polished. No. 2. The reddish-brown clay of this parcel consists of 60 per cent. of fine mud, 32 sand, and 8 of gravel; about one half of both sand and gravel is more or less angular; the other half is well rounded, and one piece in particular finely striated. Shell-frag- merits are numerous, but all so small and imperfect that they cannot be identified. The following are the Foraminifera :— Biloculina ringens, Lamk. Lagena melo, D’ Orb. Triloculina oblonga, Mon‘. Polymorphina communis, D’ Ord. Quingueloculina seminulum, Linn. Bulimina pupoides, D’ Orb, Lagena sulcata, W. § J. Cassidulina crassa, D’ Ord. marginata, W. & J. Truncatulina lobatula, Walker. —— globosa, Mont. Polystomella striato-punctata. —— squamosa, Mout. F.§ M. hexagona, Will. Nonionina depressula, W. & J. It is unusual to find so many Foraminifera and no Ostracoda or other marine animal remains. No. 3. Red muddy sand much waterworn. The organic remains are represented by very small chips of shells, a fragment of a star- fish, and one Foraminifer (Nonionma depressula). No. 4. Red clay, consisting of 58 per cent. of fine mud, 20 of sand, and 22 of gravel. No satisfactory striations were detected on any of the pieces, which were mostly small and waterworn. Shell- fragments moderately common. It is rather remarkable that these 130 T. M. READE ON THE DRIFT-BEDS OF THE shell-fragments are all angular, having no appearance of rolling or rubbing, while the gravel has suffered considerably from abrasion. FORAMINIFERA. Cornuspira foliacea, Phil. Bulimina pupoides, D’ Orb. Quinqueloculina seminulum, Zin. marginata, D’ Orb. subrotunda, Mont. Discorbina rosacea, D’ Orb. Lagena globosa, Mont. globularis, D’ Ord. sulcata, W. & J. Planorbulina mediterranensis, — Williamsonii, Alcock. D Orb. gracillima, Seguenza. ’ Truncatulina lobatula, Walker. squamosa, Mont. Rotalia Beccarii, Linn. Polymorphina gibba, D’ Ord. Polystomella striato-punctata, lactea, W.§- J. F.g M. — oblonga, Will. Nonionina asterizans, Ff M. Textularia pygmeza, D’ Orb. depressula, W. g J. ZOoPHyta. Eudendrium ramosum., SPATANGIDA. Spines and plates. Mo.uusca. Shell-fragments. No. 5. Silt (Postglacial) consisting of fine sandy mud free from gravel. Fragments and valves of shells common. The following is a list of the various organisms found in the bed :— PLANTs. A few seeds. FoRAMINIFERA. Cornuspira foliacea, Phil. | Nodosaria pyrula, D’ Ord. Biloculina depressula, D’ Orb. Polymorphina oblonga, Wail. carinata, D’ Ord. — communis, Will. Triloculina oblonga, Mont. Globigerina bulloides, D’ Orb, Quinqueloculina Candeiana, D’'Orb. | Bulimina pupoides, D’Ord. seminulum, L722. Discorbina rosacea, D’ Ord. bicornis, W. & J. Planorbulina mediterranensis, subrotunda, Mont. D Orb. Trochammina inflata, Mont. Truncatulina lobatula, Walker. Lagena levis, Mont. refulgens, Mont. gracillima, Sequenza. Rotalia Beccarii, Linn. suleata, W. & J. nitida, Will. interrupta, Wil. Polystomella crispa, Linn. globosa, Mont. striato-punctata, Ff. f M. lucida, Will. Nonionina asterizans, F’, ¢ hexagona, Wail. stelligera, D’ Orb, melo, D’ Orb. ECHINODERMATA. Ophiuroidea. Spines, sma]l and imperfect. Echinoidea. Spines, fragmentary. Pedicellaria, fragment. NORTH-WEST OF ENGLAND AND NORTH WALES. Tat Spatangide. Spines common. Thyonide. Spicules rare. ANNELIDA. Pectinaria, sp. ?, fragments. CRUSTACEA. Ostracoda. . Cythere lutea, Mii/ler. Loxoconcha multifora, Norman. albo-maculata, Baird. pusilla, Brady & Robertson. pellucida, Baird. Cytherura cellulosa, Norman. porcellanea, Baird. clathrata, G. O. Sars, castanea, G. O. Sars. —— striata, G. O. Sars. tenera, Brady. angulata, Brady. Robertsoni, Brady. — cuneata, Brady, —— pulchella, Brady. Cytheropteron latissimum, Norman. tuberculata, G. O. Sars. nodosum, Brady. —— villosa, G. O. Sars. Cytherideis subulata, Brady. concinna, Jones. Selerochilus contortus, Norman. badia, Norman. Paradoxostoma variabile, Baird. Cytheridea elongata, Brady. flexuosum, Brady. Loxoconcha impressa, Baird. arcuatum, Brady. CrrRIPEDIA. Balanus crenatus, Brug. | Balanus balanoides, Linn. Ponyzoa. Crisia eburnea, Linn. : Scrupocellaria scabra, Van Ben,, var. elongata, Smith. INSECTA. - Elytra. Mo.uuvsca. Conchifera. Anomia ephippium, Linn. Donax vittatus, Da Costa, mode- Mytilus edulis, Linn., fry and rately rare. small pearl. Venus gallina, Linn. Nucula tenuis, Mont. Tellina balthica, Linn. Montacuta bidentata, Mont. Mactra subtruncata, Da Costa, Cyamium minutum, Fabr. small valves. Cardium edule, Linn. Corbula gibba, Oliv, a valve. Pholas sp., a small fragment. Gasteropoda. Hydrobia ulve, Penn., var. sub- Aclis supranitida, S. Wood. umbilicata, Jeffreys, common, Utriculus obtusus, Monz.,one example. Seeing that the species that prevail most in No. 5, Silt-bed, Hydrobia ulve and Tellina balthica, are generally found most abundantly in muddy estuarine flats, there can be little doubt that the deposit partakes much of that character. Tre other Invertebrates in the bed do not disagree with such Q.J.G.S. No. 154. ii 132 oN THE DRIFI-BEDS OF THE N.W. OF ENGLAND AND N. WALES. conclusions. Although not peculiar to such conditions, they are frequently met with in situations more or less brackish *. The Boulder-clays differ from our Scottish Boulder-clay (Till) by being fossiliferous and intercalated with fossiliferous beds of sand and gravel. A fossiliferous section of Boulder-clay that I examined at Knock- burn, near Belfast, was much like that at Bootle, being fossiliferous and intercalated with beds of various composition. EXPLANATION OF PLATE V. Map of the Basin of the River Mersey, and of part of that of the Dee, to show the relation between the Drift and the Rock-structure of the Basins. Discussion. The Cuarrman (Dr. J. Gwyn Jeffreys) said that having examined probably all the Post-tertiary shells which had been recorded from the extensive district under consideration, he had come to the con- clusion that none of them were Arctic, but that nearly all were local, with a remarkable admixture of both northern and southern forms. The most peculiarly northern species is Astarte borealis, which is not now found living further south than Kiel Bay. The Foraminifera noticed in the paper are all local. Prof. Prestwicw asked how river-action could have taken place during the great submergence supposed by the author. He re- marked on the great value of the evidence which the sections in this paper afforded of the excavation of many river-valleys in pyeglacial times, and thus of the higher position of the land. Mr. Baverman remarked on the value of the work done by Mr. Mellard Reade in watching excavations going on during many years —excavations which would soon be filled in or obscured. He re- marked that the fragment of chalk exhibited was more like the Antrim than the English chalk. . Prof. Bory Dawxins said that fragments of chalk undistinguish- able from that of Antrim are found in Lancashire and as far to the south-east as Ironbridge on the banks of the Severn. The AurHor thought that the non-arctic character of the boulder- clay shells was accounted for by the fact that the severest cold had passed away when they were deposited. At Widnes the river-valley was excavated 140 feet below the present sea-level. * This silt is probably the equivalent of the beds underlying the great peat- bed, which the author has named the Formby and Leasowe marine bed. See ‘ Post- glacial Geology of Lancashire,” Proc. of Liverpool Geol. Soc., Session 1871-72. MAP OF TH R MERSEY A SHOW THE REL! D THE ROCK ST — ae dcale Ys | pa : ¥ i24 ‘ = } v Bi “s 1 t ’ Quart: Journ: Geol: Soc: Vol re ee INDEX. Blown Sand MAP OF THE BASIN OF THE RIVER MERSEY AND PART OF THE DEE, “TO SHOW THE RELATION BETWEEN THE DRIFT AND THE ROCK STRUCTURE OF THE BASIN. Alluvian New Red Mart Triassic Sandstone. SOUTH ee Wien Scale 1s Inch to 7 Mile. % RESERVOIR Permian . Coal Measures. Millstone Grit. Voredale Rocks — Carboraterous Limestone. Upper Silariam. Mm Position of Sections. apy) Cape ee os Y , Og YZ, A Gy Wily : Gis Al Gh SSO Ge SOL Via OD rr, Lae 4) Lee Kh AAAI C7, 7 iF, Ye GQ LIM iy y pocks @ ROO COULDA f ZA4ONAY LEG /, 4ALINT SILOS CLE WONG OS bs a, 2 E Z SRI: Wel es 7p Z S SS SES SS SS SSSssss = @N GENERIC CHARACTERS IN THE ORDER SAUROPTERYGIA. 133 9. On Guneric Cuaracters in the Orper Savropreryeis. By Prof. Owen, O.B., F.B.S., F.G.S., &. (Read December 20, 1882.) Tue progress of knowledge of the species of Reptilia associated by De la Beche under the collective name Enaliosauria, led to the sub- division of that maritime group into the orders Ichthyopterygia and Sauropterygia*, these terms being significative of their characters of resemblance respectively to Fishes and to Saurians. Large accessions of species have since been made known in both orders. With regard to the first, I have not deemed the modifica- tions in the dentition, in the shape and structure of the sterno- coraco-scapular frame, in those of the fins, or the gradations of general bulk, sufficient, as satisfactory characters, for generic sub- division. In the Sauropterygia, besides gradations of size, ranging, for ex- ample, from Plesiosaurus Hawkinsti to Ples. Cramptont, there is a difference in the proportional length of the neck and number of its vertebre relating to the size of the head it supports. In Plesio- saurus homalospondylus, for example, the cervical vertebrae are thirty-eight in number; in Ples. rostratus they are twenty-four This character alone would not have obtained a generic separation : but a shortening of the neck, due not only to decreasing number but to altered proportions of the cervical vertebre, when associated with a well-marked modification of the teeth, of the sterno-coraco- scapular frame, and of the paddle-bones, called for a separation of the Sauropterygia into Plescosaurus proper and a distinct genus, for which the name Pliosawrus was proposed‘, indicative of the nearer approach which its species made to a generalized Saurian type. In the Crocodilia, for example, .a common character of the teeth is to have the usually simple conical crown, whether: finely ridged or not, provided with a pair of enamel-ridges stronger than the rest, and placed on opposite sides of the crown. In the genus Plesiosaurus the coronal ridges of the teeth are uni- form or subequal, and the transverse section of the crown is circular . or subcircular?. In the genus Pliosaurus, besides the shortness of the neck con- comitant with hugeness of the head, a step towards the fresh-water Saurians is made by the presence of a pair of coronal ridges, longer and rather stronger than the rest, and rendered the more distinct by the characters of the parts of the tooth-crown so defined; the shorter enamel-ridges being limited to one division, and this portion moreover * Owen’s ‘Paleontology,’ 8vo, 1860, pp. 198, 209. +t “Report on British Fossil Reptilia,” part ii. in ‘ Reports of the British Association for the Advancement of Science,’ 1841. t “Monograph on the Sauropterygia,” volume of the Palzontographical Society, issued 1865, pl, ix. figs. 3-9. L2 134 PROF. OWEN ON GENERIC CHARACTERS being strongly convex; while the opposite portion is but slightly convex and is unridged. A single tooth, like some other fossilizable parts, serves unequivocally to indicate its belonging to a Pliosaur*. I will not dwell on the distinctive characters manifested by the fin-skeletony, but proceed to detail those shown by the sterno- Fig. 1.—Diagram of Sterno-coraco-scapular mass in Plesiosaurus. WAN mee \ r x \ e SQV i A NAA (MW 527 coraco-scapular frame, premising a more detailed description of the characters of that part of the skeleton of Plesiosawrus than has else- where, so far as I can find, been given. In Sauropterygia the place and function of a sternum are mainly supplied by the pair of coracoids (figs. 1 & 2, 52) which meet and * Id. ib. issued 1861, pl. vii., and 1862, pl. xii. t Id. ib. issued 1862, pp. 9-11, pl. iv. IN THE ORDER SAUROPTERYGIA. 135 join by a longitudinally extended suture (7b. 2b., s,s) below the thoracic part of the abdominal cavity. Posteriorly to (‘ sacrad of’) ’ this mesial suture, the coracoids diverge and terminate freely by a broad margin, each with an angle (7. 7b. 52’) inclining “ laterad.” Anteriorly (“ atlantally ”) the sutural portions slightly diverge (2. Fig. 2.—Diagram of Sterno-coraco-scapular mass in Pliosaurus. ib. 52') and expose the hinder (“sacral”) end (7b. ib., ms) of the mesial plate, 59, representing an episternum. Laterally, each cora- coid contracts in length, becomes thickened, and presents two syn- desmotic or roughened articular surfaces: the hinder one (2d. 2b. h’) contributes the corresponding portion of the articular cavity for the humerus, the fore one joins the scapula (7, ib., 51) by the suture, ch, laterad of which the scapula contributes the fore portion, h, of Cy 136 PROF. OWEN ON GENERIC CHARACTERS the glenoid cavity. In Plesicsaurus the hinder end of the scapula, which is the thickest part of the bone, is thus divided pretty equally between its coracoidal (ch) and humeral (fh) articular surfaces, both being rough or “‘ syndesmosal.” In advance of the surfaces h, ch, the scapula thins and contracts, chiefly by a strong margino-mesial concavity, contributing the outer border of the “ coraco-seapular vacuity ” (2b. 7b., cs). The outer or lateral and thicker border of the scapula is nearly straight ; and the bone extending forward and slightly mesiad, expands to unite with the episternum, 59, by the suture, sh. The episternum, 59, presents anteriorly a mesial notch, from each angle of which the bone extends outward and backward to its sutural union, sh, with the fore end of the scapula. At this union the episternum contracts, and is continued backward to join the coracoids, passing a short way internal to (“‘ centrad” of ) them, and appearing outwardly as a terminally pointed portion, ms, at the fore part of the narrow mesial interspace of the coracoids, 52*, which interspace interrupts anteriorly their extensive mesial suture with each other. Thus the sterno-coraco-scapular frame, or mass, presents an anterior and a posterior emargination and a pair of subcircular vacuities. The above-defined characters of this portion of the skeleton, save that of the scapular element, are common moreover to both the generic groups of the Sauropterygia. The chief and suggestive modification of the mass in the Plio- saurlan genus is the retention of a typical character of the scapula which is lost in the more modified or specialized Plesiosaurian forms, viz. the production of the part of the blade-bone (fig. 2, 51*), laterad and dorsad, where it terminates freely 7. This portion re- presents the main body of the scapula in the higher Vertebrates, but, as in the “ Allantoic group” (Reptilia and Aves)t, without expanding. The portion of the scapula, 51, common to both genera, which contributes its share (figs. 1 & 2, h) to the glenoid cavity, is separated in Plidsaurus from the free portion, 51*, by the notch, n. In ad- vance of this the Pliosaurian differs from the Plesiosaurian scapula by its greater relative breadth, extending its sutural border, sh, mesiad, so as to touch or join the fore end of the coracoid, 52*. The coracoids retain their large proportional size, but have a less even or flattened outer surface ; mesially they bulge to their common suture, s, giving more room to the ventral or visceral cavity ; and, at the transverse margin parallel with the hind border which they con- tribute to the vacuities, cs, cs, they bend dorsad, suddenly contract, + This character I added to the generalized illustration of the Sauroptery- gian skeleton in my ‘ Palzontology,’ p. 227, fig. 71. t “In their generation and development modern Batrachians differ from other cold-blooded air-breathers and agree with fishes. Birds, by genetic and developmental characters, as well as by the general plan of their organization, are more intimately and naturally allied to the oviparous Saurians than to the viviparous Mammals.”—Anaz. of Vertebrates, vol. i. 1866, pp. 6 and 7. IN THE ORDER SAUROPTERYGIA. LS7 52, but contribute, as in Pleszosaurus, the mesial border of those vacuities, and articulate, underlapping it, with the hinder end of the episternum, 59. The proportional characters of this element are given in figs. 1 & 2. In thus determining the homologies of the constituents of the complex bony buckler in Sauropterygia, 1 have exhausted every subject of comparison at my command, derivable from fossil remains of the group and from other Reptilian forms both fossil and recent, and in the latter have had recourse to modes and stages of develop- ment of the constituents of the answerable part cf the frame. The degree in which the abdominal surface is defended by bone in Sauropterygia resembles that in Chelonia. But the homology of the defensive parts can be safely predicated of but a small pro- portion only of the elements of the plastron. The episternum (figs. 1 & 2, 59) may answer to three of the Chelonian elements, viz. to the pair of bones so named, and marked es in fig. 3, and to the mesial piece, s, continued backward in a pointed form, and called “ entosternum.” But such constituents have coalesced into one bone in Sauropterygia, and I have no evidence, as in Chelonia*, of its development from several centres. One might be tempted by the size and shape of the parial elements of the plastron, As (hyosternals), in the immature tortoise (fig. 3), Fig. 3.—Development of Plastron, young Tortoise. to regard the broad coracoids, 52, in figs. 1 & Z%, as homologues, espe- cially in the Chelonian half-developed state, when the fore and outer angle is produced in direction and degree like the scapular process, 51*, in Pliosawrus. But the process in the plastron of the Tortoise expands as it grows, and ultimately articulates with the dermo-marginal pieces of the carapace ; it is, like them, a ‘‘ dermal _™* “On the Development and Homologies of the Carapace and Plastron of the Chelonia ” (Philosophical Transactions, 1849). 138 ON GENERIC CHARACTERS IN THE ORDER SAUROPTERYGIA. bone.” The true homology of the constituents of the sterno-coraco- scapular frame is yielded by the endo-skeleton of Chelonia, in which the true coracoid is the largest and broadest of the elements of the shoulder-arch, the mesial margins almost meeting beneath the fore part of the thoracic-abdominal cavity. Recailing the impression made on the mind of Cuvier * by the first account of the Plesiosaurus dolichodeirus, we can appreciate the advantage and reward of continued researches, in the blotting- out of seeming anomalies, and in the addition of features of affinity linking on the strange extinct form (brought to light by Conybeare) to the general Reptilian type, and diminishing the intervals which seemed to exist in the series. The few and short cervical vertebre in Pliosaurus manifest, with these Ichthyosaurian proportions, by their amphiccelian, almost flattened, articular surfaces, characters of contemporary Crocodilia, of which Reptiles we ure also reminded by the short and thick neck, the large head, and the powerful jaws of the later-found Sauropterygian genus, The teeth, moreover, are now implanted in distinct sockets ; and the blade-bone resumes its normal character. It is interesting to note that the species tending to diminish the interval that seemed to separate Conybeare’s Plesiosaurus from the typical Reptilia have been hitherto obtained from Mesozoic deposits less ancient than the Lias. All my evidences of Pliosaurus have been derived from Kimmeridgian and Portlandian beds. A third generic modification of the Sauropterygia is indicated by teeth and a portion of the skull from a part of the Cretaceous series ; but I wait for further acquisition of its remains before submitting to the Society the differential characters of the genus Polyptychodon. Discussion. The PresrpEnt stated that great differences of opinion existed as to the form of the scapular arch in Plesiosaurus. He himself believed that there is evidence that in Pliosaurus there was a dorsal pro- longation of the scapula similar to that found in the higher ver- tebrates. * “ Cest par cet habitant de l’ancien monde, peut-étre le plus héteroclite, et celui de tous qui paroit le plus mériter le nom de monstre, que je terminerai cette histoire des animaux perdus.”—Ossem. Foss. tom. v. pt. 2, p. 476, 1824. T. GRAY AND J. MILNE ON JAPANESE ROCKS, 139 10. On the Exasriciry and SrreneTH-constants of JAPANESE Rocks. By Tuomas Gray, Esq., B.Sc., F.R.S.E., and Prof. Jonn Mite, F.G.S. (Read November 2, 1881.) (Abridged.) Tu1s paper contains a description of experiments made for the purpose of determining the different moduli of elasticity, and the strength against rupture and crushing of some of the more common Japanese rocks. One main object of the experiments was to obtain data for the calculation of the theoretical velocity of propagation of _ earthquake-waves, and by comparison of this velocity with the results of observation, to gain some idea of the degree of continuity in the strata. The experiments on the moduli and the rupturing strength were performed on round columns of the rock about 4 centimetres in diameter and 60 centimetres long; those on strength against crushing were for the most part made on round columns 4 centi- metres in diameter and 12 centimetres long. The Young’s modulus of elasticity was calculated from the results of experiments on cross bending, the columns being supported at both ends. The modulus of rigidity was calculated from the results of experiments on the torsion produced by the application of measured twisting motives. The bulk-modulus was calculated from the known relation between it and the Young’s and rigidity-moduli. The modulus of rupture was found by observing the load required to rupture the column by cross bending. . The experiments on crushing were performed by means of a Bramah’s press, the pressure being estimated by means of a Bourdon’s gauge. The following Table (p. 140) gives the results of the experiments ; the headings of the different columns sufficiently explain their contents. Discussion. Rev. E. Hirt thought that the results obtained might have another application. In studying faults and joints, the elasticity and capability of resisting crushing cf different rocks must neces- sarily often be taken into consideration ; and for such a purpose the determinations made by the authors might be of great service. 140 T. GRAY AND J. MILNE ON JAPANESE ROCKS, 2°63 2°70 2:28 2°68 2°74 *yood Jo Ayisuaq "Sa.}ouII}Ua0 UL SUIYsNAO Jo Sn]Npow WWSUET 87:4 x 10? 70 4x 103 . 50:0 x 10? 57°5 x 10° 114 1063 ‘aajeul}ueo atenbs Jod sawiureis UL SUTYsNao Jo sn[NPOP, 190 x 10? 114 x 108 154 x 10° 313 x 10? "So.1jauI}U90 UL eanjdnat o10foq UOTPAPep UNUTIXVTT ‘070 | 230x103 ‘O75 ‘079 ‘069 079 ‘aajoutqueo earends ded sommmers ut oingdna Jo snjupoyy L710? 137 x 163 131 x 10° "QABM IS1OASURI} JO AjTOOTOA 04 TeULIOU JO AZLOOTeA JO OTe 1:37 | 252~x 10° 58-417 < 108 1:80 1:83 1°36 ‘puooes Jod sarjoutt}t00 UT DAVM OSLIASTIVA} JO AJLOOTOA PoJe[MOTRO 219 x 10° 208 x 10? 209 x 103 254 x 10° 286 x 10° UI @ABM [BULIOU JO APLOOTIA poye[NoTBO 395 x 10% 381 x 10? 285 x 108 348 x 103 451 x 10° "gaqJOUII] 199 UL SN[NpOW-ApIpIsia ayy JO Y}SueT . [ 48:9 x 108 44'1 x 106 44-7 x 106 66:0 x 106 36+ 106 8 "sa.joullyme0 UI SNTNPOUL s,sunoZ ogy Jo yyoueT 159 x 106 148 x 10° 82°9 x 106 ] 23 x 108 250 x 106 ‘aqjoumtyued eenbs Jod sommes UL sn[npoU-AqIpIsny 128 x 106 119 x 106 102 x 106 177 x 10° 229 x 108 ‘oqjoulyUed erenbs _ dod saummeas UI sn[upow ssunox 416 x 106 400 x 106 189 x 106 329 x 106 686 x 10° | | | | | | | | ai Granite . Marble Clay Rock . Tuff Slate APO so PEt |: METAMORPHIC AND OVERLYING ROCKS OF ROSS AND INVERNESS. 141 11. On the Meramorputc and overlying Rocxs im parts of Ross and Inverness Surres. By Henry Hicxs, M.D., F.G.8S. With Notes on the Microscopie Structure of some of the Rocks by Pro- fessor T. G. Bonney, M.A., F.R.S., Sec.G.8. (Read February 7, 1883. (Puate VI.) Contents. . Introduction. Upper part of Glen Logan. . Neighbourhood of Achnasheen. Ben Hay and Loch Clare to Glen Carron. Achnashellach, Loch Doule and Strathcarron. Loch Kisborn to Loch Carron. Attadale, Loch Carron. Strome Ferry and Loch-Alsh Promontory. Loch Shiel to Caledonian Canal. . Conclusions. SO AI GUA go bo — 1. Introduction. Sryce I communicated my paper to the Geological Society in 1878, on the rocks in the neighbourhood of Loch Maree, Ross-shire, I have revisited the north-west Highlands on two occasions, viz. in the spring of 1880, and in the spring of last year. During these visits I devoted my time to the examination, in various areas, of sections which seemed likely to furnish evidence confirmatory or otherwise of the views which I had ventured to lay before the Society. In this paper I propose to give the general results obtained from these exatninations, and to treat more fully those sections which appear to offer any conclusive evidence. In each of the areas examined I made large collections of rock-specimens ; and numerous thin sections for examination with the microscope have been made from these rocks, and submitted for special examination to Prof. T.G. Bonney, and Mr. T. Davies. The former has kindly furnished the notes on these sections which aré appended to this paper; and to both, on this as on so many former occasions, I am indebted for most valuable assistance in the petrological work. The notes by Mr. Davies in the paper I published in the Geological Magazine in 1880, on the Gairloch and Ben Fyn districts, have also so important a bearing on the questions discussed in this paper that I shall find it necessary to refer frequently to them. The notes in that paper on the Gairloch specimens are particularly important, as they prove clearly the presence in that western area of rocks which cannot be differentiated from many rocks in the eastern area of Ben Fyn, included by Murchison, Geikie, and others in the so-called newer gneiss of Silurian age. This group of Ben Fyn is in this paper taken as the type to which is referred the newer series of metamorphic rocks of Pre-Cambrian age in these areas as distin- guished from an older, or Loch-Shiel series, and the supposed still older Loch-Maree series. 142 DR. H. HICKS ON THE METAMORPHIC AND In a former paper* I placed the axis of the oldest rocks in the direction of Loch Maree, with newer beds to the south west at Gair- loch and to the north-east about Ben Fyn. In doing this I did not necessarily mean to correlate the series at Gairloch with those of Ben Fyn, but believed them to be so closely allied that they could be easily included in one great group. The more hornblendic varieties along the shores of Loch Maree and the granitoid rocks south of Poolewe were, in my opinion, of older types; but whether separated from the former by actual unconformity or not I was unable to say. Without referring again to the several views maintained in regard to the metamorphic rocks in the eastern areas, with which I have chiefly to deal, it is clear, as specially bearing upon those views, that the following points have to be kept constantly in mind in these inquiries. (1) Is the stratigraphical evidence as to a continuous up- ward succession, maintained by Murchison and Geikie, of that conclu- sive character that it must overthrow all petrological evidence which may seem to render this, if not impossible, yet highly improbable ? (2) Has this upward succession from Torridon Sandstone through the quartzite and limestone series anywhere been observed to gra- duate into gneiss rocks of the type such as we classify under the name Ben-Fyn series. (3) Are not the flaggy micaceous rocks, such as those which we have described from the east side of Glen Logan, which probably overlie the limestone series (though there undoubtedly separated from the latter by a fault), more intimately allied in their microsco- pical and general characters to the Torridon series than to those of the Ben-Fyn type, which they are supposed to immediately underlie? (4) Are there evidences of the disturbance of the strata by faults and inversions along these lines of such a kind as would be likely to greatly interfere with and to complicate the order of succession ? (5) Do the eastern metamorphic rocks show at different horizons and places that variability in the amount of alteration which is usually noticed in rocks subjected to local influences only ? or do they not rather everywhere show an identical state of alteration, as if the result of some wide-spread cause, rather than due to local mechanical disturbances ? Before proceeding to describe the various sections examined since my paper was read in 1878, I may here take the opportunity of referring to a point in connexion with the section published in that paper which needs some explanation (though I have to a certain extent done so in my paper in the Geological Magazine, 1880). In that section the beds were accidentally placed at too high an angle in the part east of Glen Logan, and the floor made to appear to continue too uninterruptedly eastward. The published section, however, as mentioned in the discussion, was intended merely as a diagram ; the views that I entertained were fully explained in the text t. Yet as it might tend to give a misleading idea as to the condition of the floor at this part, I think this explanation necessary. The presence * Geol. Mag. Dee. ii. vol. vii. 1880. t Quart. Journ. Geol. Soc. vol. xxxiv. p. 816, OVERLYING ROCKS OF ROSS AND INVERNESS. 143 of the floor in the entrance of Glen Docherty, as nearly as possible in the position indicated in that section, I clearly recognized again during my last visit. It is also seen as we leave the upper end of the glen; but between these points the ground has been so much broken up by faults passing along the glen and at right angles to it, that it is impossible to trace it continuously. The presence of the old floor in Glen Logan as far as the entrance into Glen Docherty (shown also in Prof. Bonney’s diagrams and recognized clearly by him *, and referred to also by Mr. Hudleston in his paper in the Proceedings of the Geologists’ Association, 1879) is a fact of so much importance in these inquiries, in consequence of its being found east- ward of the quartzite and limestone series, that the question of its presence in Glen Docherty may be almost dispensed with in the dis- cussion. If the oldest rocks can be seen to rise up anywhere eastward of the limestone series, then the probability of the eastern true metamorphic rocks being parts also of the old floor, as maintained in my former papers, is rendered still stronger, especially if, as I hope to show in this paper, there is abundant evidence tending towards that conclusion. I may here also refer briefly to the prevailing faults found in these areas. ‘There can be no doubt, as shown by the directions of the lochs and valleys, that the main faults trend from N.E. to 8.W. and from N.W. to S.E. Branching off from these, however, are many minor faults ; but though these are continually met with, they do not usually greatly interfere with the succession, though they occasion- ally cut off considerable thicknesses of strata. Comparatively a few only of the faults are indicated on the map (P1.VI.), and those mainly which have a bearing on the questions considered. Some of the local names are taken from the recently published Ordnance map; but the majority are from that published by Black, as they agree with those used in my former paper. At that time the one-inch Ordnance map of these areas had not been published. The numbers on the map refer to the positions of the rock-speci- mens described, and correspond with the notes by Prof. Bonney. All the rocks described are from points not referred to in my former papers, and they include all the varieties which appeared at all typical in the traverses made. One section only was examined to the north of the area previously described, the others being all to the south and east. The one to the north may be looked upon as a continuation to the N.H. of the former Glen-Logan sections. 2. Glen Logan (or Laggan), Upper part. The gneiss, note no. 1, has a well-marked strike from N.W. to S.E., and in its general appearance seems typical of the older or what we call the Loch-Maree series. About the cottages near the entrance to the glen on the west side leading to Loch Nafatt are exhibited well-rounded surfaces, upon * Quart. Journ. Geol. Soc. vol. xxxvi. p. 93. 144 DR. H. HICKS ON THE METAMORPHIC AND which the strike of the foliation is beautifully shown: here the gneiss is seen at a height of from 300 to 400 feet above the datum- line ; but it can be traced to a considerably greater height. A fault seems to extend along the glen above referred to; and the ground is broken up about the entrance. Though the rocks beyond this fault to the N.E. of the glen appear to differ in some important par- ticulars from those described on the south side, it is clear, from the notes 2, 3, and 4, that they are also a highly metamorphic series. The fault has probably cut out a considerable thickness of the strata, being, as it is, nearly parallel with the bedding; therefore there is a rather abrupt passage from the one group to the other. On the south side of the glen, as already stated, the rocks are of the Loch-Maree type ; but those to the north agree better in their petro- logical and physical characters with those of the Ben-Fyn type. I traced the section along the north side as far as Leckie, and found the beds to dip generally at a high angle to the north-east. In some places they are slightly contorted, but are seen quite evenly bedded and in regular succession in the gorge, where the bridge crosses the branch of the river flowing from a glen to join the main stream below Leckie. In the face of the hill on the south side of the main stream of the river and of the valley the beds dip to the S.E. These are much less altered than those described from the north side, and are evidently to be correlated with the flaggy micaceous series in the valley of Glen Docherty to the south. They compose the main portion of the mountain called Craig Roy. At one point they may be seen slightly bent towards the north-east, asif dragged down towards the fault extending along this valley towards Loch Fannich ; but as they are entirely discordant in strike to those on the north side, and dip at a lower angle, it is clear that they are a much newer © series, and in their petrological characters may be classed with the flaggy beds which are supposed to be seen in so many areas resting upon the limestone series. After a careful examination of the beds along Glen Logan and its branches, I came to the conclusion that these flaggy beds rest unconformably upon the Loch-Maree and Ben-Fyn ‘series. The faulting has not, in my opinion, been of a character which would so completely baffle the order of succession here as to make this an illusory appearance only; and the result of the faulting can be calculated without much difficulty. As these beds of Craig Roy approach the western shoulder of Ben Fyn, they appear to dip slightly towards that mountain; but this also is evidently the result of a fault; for immediately we pass this point eastward, an entirely distinct group of rocks, with a strike from N.W. to S.E., at a very high angle, is met with; and the beds continue to retain this high dip, with here and there a fold over, through the whole range of mountains east of this point and to the north of Achnasheen. 3. Neighbourhood of Achnasheen. The characteristic gneisses and mica-schists found in these mountains have been described in my former papers, and in that by OVERLYING ROCKS OF ROSS AND INVERNESS. 145 Prof. Bonney *. They include coarse and fine-grained granitoid gneisses containing garnets and sphene, a well-marked augen-gneiss, dull-coloured and bright silvery mica-schists containing abundance | of garnets, and micaceous gneisses with bands of white and black mica, with a moderate amount of quartz, felspar, &c. On the south side of Loch Roshk the same characteristic gneisses and mica-schists appear, and with a similar strike. No. 5 is typical of the series in the shoulder of the hill between the east end of Loch Roshk and the valley to the south towards Glen Carron, which the Dingwall and Skye Railway traverses. To the east of the railway the rocks are also highly typical of the Ben- Fyn series. They he at a high angle with an easterly dip. Here and there contortions are recognizable; but the only very definite change in the dip observed was in the mountains south of Loch Luichart f. In this area, as in the Ben-Fyn group of mountains, the degree of alteration is equal throughout from the base of the series to the top; and that the metamorphism cannot be due to any local mechanical disturbance is perfectly clear from the evenly bedded character of the majority of the rocks. In every specimen examined an intimately crystalline condition was observable, such as is usually considered characteristic of the true gneisses and schists; and gene- rally also the minerals sphene and garnet, which could not have been originally present in the sediments, occur in abundance throughout the rocks. 4, Ben Eay and Loch Clare to Glen Carron (fig. 1). _ This section is taken in a line nearly due south of one described in my previous paper. It illustrates the general order of succession to the south of Loch Maree, whilst the former indicated the g. 1.—Section from Ben Eay by Lochs Clare and Coulan to the S.E. of Glen Carron. (Seale 3 inch to 1 mile.) S.E. Loch Coulan. Glen Carron. ¢,¢. Gairloch and Ben-Fyn series. d. Torridon Sandstone. d'. Quartz rock. . e. Limestone series. f. Glen-Docherty series. order on the north side. As previously mentioned, the fault dlong Loch Maree has thrown down the floor and the overlying beds on the south side to a much lower level; and in consequence the old floor is lost further westward than on the north side. The diffi- * Quart. Journ. Geol. Soe. vol. xxxvi. p. 93. + It is also mentioned by Murchison as occurring in Ben Eigen 146 DR. H. HICKS ON THE METAMORPHIC AND: culty of tracing the floor in the area now to be described is also greatly increased in consequence of the very highly faulted condition of much of the ground. If we begin the section with Ben Hay, we shall find that the Torridon Sandstone, which forms the base of this picturesque mountain, rests towards the north-west on a floor of the old gneiss, but at a much lower horizon than it does at Ben Slioch on the north side of the lake. ; The Torridon Sandstone in both cases is alike in its general characters. It 1s occasionally brecciated, though much less so than in the mountains still further west, about Gairloch. For the most part it shows well-rounded fragments along this line and as far as the shores of Loch Torridon. Here and there shaly bands are found ; and on the north shore of Loch Torridon I found. these occurring almost to the base of the series. The colour of the sand- stone varies from a bright pinkish and reddish colour to a dark green. The former is due to the felspar present, and to a reddish coating of the quartz; and the latter to a fine chloritic material disseminated through the rock, and also coating the quartz grains. Indications are abundantly present in the sandstones and breccias of the kinds of rocks which must have yielded the materials ; and it is abundantly clear that they must have differed considerably at the time in their degree of crystallization. For instance, at Gairloch the fragments may be readily matched with the rocks on the upturned edges of which they now repose. In the Torridon area the evidences are clear that similar rocks were being denuded, as well as some chloritic rocks and some less altered slaty beds. That fragments of these less altered rocks occur here in combination with those of the true gneiss rocks is clear proof that there were some Pre- Cambrian rocks in these areas, as in Wales, which had not under- gone a great change before the Cambrian rocks were deposited upon them. These comparatively unaltered rocks in Wales have not, as shown by the fragments in the conglomerates, undergone much change since that time, though they have been depressed to great depths and have suffered repeated crushings and other effects of local disturbances. The finding in these areas, as in Wales, of frag- ments of highly crystalline rocks with those so much less altered is most interesting, and should lead us to be careful in placing any of the even moderately altered rocks in the newer series, unless the evidence of their geological position is beyond the possibility of doubt. We see therefore from the fragments in the conglomerates and sandstones in these areas that some comparatively but slightly altered rocks were being denuded by the Cambrian seas ; hence we must be prepared to meet with these somewhere. Possibly this fact may have amore important bearing on the supposed evidence of progressive metamorphism in these areas than we are able at present to realize. In this area there seems to be clear evidence that the quartzite series was deposited unconformably on the Torridon Sandstone. In the quartzite series are some subordinate bands of a yellowish calea- reous shale. These are theso-called fucoidal bands, found here and OVERLYING ROCKS OF ROSS AND INVERNESS. 147 at many points north of this area. This series is well exposed capping Ben Eay and the other mountains towards Loch Torridon. The dip is towards the east ; but the series is only found in a very broken condition in the low ground towards Loch Clare. A por- tion only of the next or Limestone series also is exposed here, in consequence of the very faulted condition of the ground. Were this section alone to be taken as showing the order of suc- cession in this area, the evidence as to upward passage into the next series would be almost valueless, as the faults are amply sufficient to bring up the old floor to appear, to rest upon the Limestone series ; and the advocates of that view could have justly claimed that the comparatively unaltered condition of the next series would only tend to show that the beds were the newer Pre-Cambrian rocks, which yielded the comparatively unaltered fragments found in the conglomerate here and at Loch Torridon. As there seems, however, much evidence tending to the belief that these beds, which belong to the Glen-Docherty series, do rest conformably on beds of the Limestone series further north, it will be better to take for granted here that these are newer beds, and that they are thrown down against the Limestone series. They are . to be traced continuously from Glen Docherty to the shores of Loch Clare and Loch Coulan. They dip to the 8.E. with a gentle incli- nation, and throughout are a flaggy micaceous group, in some places, however, almost a clay-slate; at other points fine-grained quartzose rocks may be found. Superficially they put on sometimes a more altered look than is seen either in the sandstones and clays of the Limestone series immediately below, or in the Quartzite and Torridon group; but on microscopical examination the fragmentary character of the series is always well marked, and after a very sught examina- tion no one would be likely to be led to the mistake of associating them with the true gneisses and mica-schists of either the eastern or the western area. Beyond and to the north-east of Loch Coulan they are entirely cut off by a fault from the next rocks to be described. To the south also they are cut off from the Torridon- Sandstone and Quartzite mountains, which extend eastward at a greater elevation parallel with these supposed much newer rocks. The presence of great mountains of Torridon Sandstone capped with quartzite to the south of Loch Clare and Loch Coulan, and ex- tending as far as Loch Carron, proves beyond doubt that this region has suffered enormously from various kinds of disturbances. The faults, it is certain, must in this area have had a marked influence on the succession, since it is perfectly clear that the same rocks as those met with thrown down in the valley to the north-west of Loch Clare are found capping mountain after mountain far away to the south, whilst the uppermost flaggy beds may almost be said here to dip towards and even below them. That the Sandstone-and- Quartzite of this area is nothing more than the same series that was met with at Ben Hay and in the mountains south and west of that point, carried further eastward, there cannot be the shadow of a doubt to the mind of any one who thoroughly explores it; therefore Q.J.G.8. No. 154. M 148 DR. H. HICKS ON THE METAMORPHIC AND any supposition of their being higher beds of sandstones and quartz- ites reposing upon a lower quartzite-and-limestone series must be entirely cast aside. These rocks will be further referred to when describing the areas of Achnashellach and Strathearron. Continuing the section eastward of Loch Coulan, still in some- what broken ground, we come upon a series of rocks entirely unlike those referred to as the Glen-Docherty series. As the road (private road belonging to Lord Wolverton) leading to Achnashellach crosses the ridge the mica-schists described in the Appendix, note no. 6, are found. The general dip is to the N.W.; but here they are consider- ably contorted, and a dip to the 8.E. is soon afterwards recognized for a short distance in the same rocks. In descending towards Achnashellach the rocks no. 7 are met with, and apparently as belonging to the same series as the above, though the evidence of this is not quite certain. As we approach towards the station, undoubted old rocks of the type described under no. 9, and a reddish felspathic group not unlike the old rocks of the Logan valley, appear. On the roadside immediately to the west of the station, Torridon Sandstone is found faulted against this old rock. 5. Achnashellach and Strathcarron (fig. 2). In the gorge of the river in the private grounds of Achnashellach Lodge an excellent section of the quartzite series is exposed. Towards Loch Doule the beds dip at an angle of about 45° to the S.E., evidently as the effect of the main fault in the Loch-Doule valley. A moderately high dip prevails also for a considerable distance up the valley leading to Loch Corry Lair, and the quartzite Fig. 2.—Section across the Carron Valley, between Loch Doule and Loch Carron. (Scale % inch to 1 mile.) W.N.W. 2 §.8.E aes B h of Fourell Bem ranch o More Bas Attadale River. t 1 ! 1 t { | AWN SW Oren \ RIAA ad @e c c ec. Gairloch and Ben-Fyn Series. d. Torridon Sandstone. d'. Quartz Rock. e. Limestone Series. series is seen as if resting almost conformably upon the Torridon Sandstone. ‘The latter is passed over at the highest point between Achnashellach and Loch Corry Lair, and the succession is to be clearly made out in the mountains to the N. andS. of the pass. The moun- tains to the west of this lake also seem wholly made up of Torridon Sandstone capped with quartzite. That these are true Torridon Sandstones, and not subordinate bands in the quartzite series, is perfectly clear to any one who has seen the succession on the Torridon shores. The thickness also is very great, evidently several OVERLYING ROCKS OF ROSS AND INVERNESS. 149 thousands of feet. Conglomerates are seen alternating with the sandstones ; and in these are found occasionally large fragments of greenish schistose rocks, very like some of the rocks found in the range of mountains to the north-east, supposed by Murchison and Geikie to be newer than these groups, but which I maintain belong to a much older series. How it can possibly be supposed that the rocks forming the mountain Cairn a Grubie to the N.E. of Achnashellach repose upon the quartzite series of this area, J am quite unable to make out. As already shown, the dip is in some cases in exactly the opposite direction ; and even when it is reversed to the S.E., the strike would be directly against, and not over, the Torridon and Quartzite series. Altogether the evidence here is most decidedly opposed to any view of an upward succession into the gneissic and chloritic series. In travelling along the Loch-Carron road we meet with the quartzite as far as Coolagin. In the bed of a river coming down from the mountains on the N.W., just beyond this point, another clear section is exposed, and the uppermost beds of the quartzite series full of so-called Annelid-tubes are found. These are succeeded in the low ground towards the main river by the Limestone series. The limestone here is undoubtedly, from its general appearance, identical with that found in the western area about Kishorn, to be described further on, and seems in many respects closely allied to the Durness and Assynt limestone of the north. The position of the quartz rock, with Annelid-tubes, rela- tively to the limestone seems also to lend further weight to this supposition. The Torridon Sandstone is met with higher up the stream, and is seen there to underlie the quartzite series as in the Achnasheliach valley. The succession is therefore, on the whole, more perfect also; for we meet with Torridon Sandstone to the N.W., and in descending towards the Loch-Carron valley the Quartzite series resting upon it, and afterwards the Limestone series in its proper position. These are thrown down together at a high angle towards the main fault in the Loch-Carron valley, and evidently immediately against the true gneiss rocks of the Ben- Fyn type which are found in the rising ground on the east side of the valley. The section, fig. 2, is intended to explain the succes- sion along this line as far as the fault, and it is then carried across the mountains immediately to the west of Strathcarron Station, which consist entirely of rocks of the Ben-Fyn type, such as are described in the notes (Nos. 23-27) by Mr. Davies to a former paper *, and of some allied rocks described by Prof. Bonney in the notes 18-21 in the Appendix to this paper, to be referred to further on in the description of the section in the Attadale valley. 6. Loch Kishorn to Loch Carron. In the area between these two lochs I examined several very interesting and important sections in connexion with some of the questions considered in this paper. I also ascended the highest * Geol. Mag. dee. 2, vol. vii. 1880. M2. 510) _ DR. H. HICKS ON THE METAMORPHIC AND mountain in the area, Glas Bheinn, to see whether the interpretation furnished by the lower ground was equally applicable to the highest points. To the north of this area we have the mountains of Torridon Sandstone, capped with quartzite, already referred to, and to the west the great mountains of Torridon Sandstone in the Applecross district. It seems to be bounded more or less by faults in all direc- tions, and some of these must be faults of considerable magnitude. With the exception of the broken patch of the Limestone series on ‘the shore and to the north of Loch Kishorn, I have indicated the whole as belonging to the Ben-Fyn series (or stage of crystalliza- tion); but this is to some extent only a provisional arrangement, as it is quite possible that a large proportion of the rocks in this area may prove to belong to a distinct group not represented in the Ben- Fyn district. That they are for the most part equally altered with that group there cannot, however, be any doubt; hence instead of classifying them under a new name, I have thought it best at present to associate them together. Sir R. Murchison has described the rocks in this area as being of Silurian age, the limestone of Loch Kishorn being at the base, with an ascending succession towards Loch Carron. In his joint paper with Prof. Geikie * it is stated that the Kishorn limestone may possibly ‘‘ be the same as that of Loch Carron, Loch Coulan, and Glen Cruchalie—that is, the limestone zone between the lower quartz-rock and the upper quartzoge flaggy series ; or, like the lower limestone of Ben Eay, it may be a local deposit occurring in the lower quartz-rock.” Itis also stated by them to be “underlain by white quartz rock, which, coalescing with that above the limestone, forms one series, below which lie the Cambrian sandstones swelling up into the great mountains of Applecross.” Prof. Nicol says that the “limestone rests on the quartzite, which in one place dips at 15°, to $. 40° H. The limestone is, as usual, more broken and irregular, but near the bridge to Apple- cross it dips at 64°, to EH. 8° N. The tale-slates on the east have a dip of 20°, to EK. 30° N.; and, on the whole, lower angles than those given in my former paper seem to prevail in these beds. Granulite and hornblende-rocks, however, abound near the line of junction ; and I was still unsuccessful in finding any point where the talc- or mica-slates overlap the limestone or quartzite. I have now no doubt, from the facts seen at the junction in other places, that the limestone and talc-slate are divided by a line of fault. The occurrence of the limestone in this position, though quite analogous to what is seen in Assynt, is very important. It lies in a low valley at the foot of the red sandstone hills of Applecross, more than 2000 ft. high, and, asits regular position is above the quartzite, it must have been thrown down fully 3000 ft.” After carefully examining the section at this point, I felt satisfied that the description given by Prof. Nicol was the correct one, that the Quartz-rock and Limestone series have been thrust in among the old rocks, by faults, and that the evidence is altogether opposed to the view that they underlie in conformable order the highly metamorphic series to the east. I examined the latter * Quart. Journ. Geol. Soc. vol. xvii. p. 197. t Ibid: pitty, OVERLYING ROCKS OF ROSS AND INVERNESS. 1 at many points and found none that could be considered even partially typical of the Glen-Docherty series. ‘The limestone of Loch Kishorn (note 32) is exactly like that which I have already described near Strathcarron. It is frequently brecciated and generally of a bluish or greyish colour. It is traversed also by cherty layers; and though fossils have not as yet been found in it, it seems altogether so like that in which fossils occur at Durness that I believe they may yet be discovered. The sandstones in association with the limestone are also so exactly like those found in other areas that they need not be referred to. They are in no part here more altered than are the ordinary fine-grained Torridon Sandstones of the typical areas. Besides the schists to the east of the limestone mentioned by Prof. Nicol, I found, in the gorge through which the road to Jeantown passes, a series of red augen-gneisses of rather a peculiar character, in association with greenish-looking hornblendic schists. A specimen of this augen-gneiss is described in note no. 12 by Prof. Bonney. In ascending the hill eastward from this point, horn— blendic schists with red felspathic lines and rather massive-looking rocks of a green colour freely permeated by epidotic veins are the prevailing types. At the crest of the hill and in descending to Jeantown, reddish quartzose and other gneisses more approaching the Ben-Fyn types are found, and the beds become on the whole thinner and more contorted, a distinct reversed dip being found to the south of Jeantown. Directly to the north of Jeantown, on the shore of Loch Carron, the dark micaceous schist (note no. 11) is met with. These schists and the quartzose gneisses are found extending along the shore to Loch Carron Kirk, from which point I ascended the mountain Glas Bheinn, and their dip is generally eastward. In ascending Glas Bheinn until a height of about 600 feet is reached the usual Ben-F yn types of gneisses are found; but beyond this, and reaching quite to the top of the mountain, the hornblendic gneisses described in note 11 are the prevailing types. Indeed the central portions of the mountain and its shoulder to the west seem chiefly to consist of these rocks. Numerous bands of a reddish fei- spathic rock and segregation veins of felspar and quartz are also abun- dantly present in the series. The whole aspect of these rocks to the very top of the mountain calls to mind rather the older Hebridean series than the Gairloch or Ben-Fyn types. Yet as the latter seem to repose upon these on the east side they are for the present grouped together. ‘The strike of the beds is usually from N.W. to 8.E. or from that to N. and’.; and the dip, generally high, is in some places almost vertical. The presence of a group of gneisses of so old-looking a character, and with a crystalline condition, as shown in microscopical sections, not to be distinguished from the oldest gneisses of the Loch-Maree type, reaching to the crest of a mountain of over 2300 feet in height in an area regarded as.containing the so-called newer Silurian metamorphic rocks only, and east of the limestone series, is a fact of enormous importance, especially as we are told by Murchison and Geikie that the rocks found in this area are newer than the Lime- 152 DR. H. HICKS ON THE METAMORPHIC AND stone series of Loch Kishorn, and that they repose conformably upon the latter. Not only does it seem incomprehensible from their pecu- lar mineral characters that these can be altered Silurian rocks, but the evidence afforded by the strike, which is directly opposed to that of the unaltered Torridon-Sandstone and Quartzite series forming the mountains directly to the north, and the fact, moreover, that here and there we meet with rocks in the latter area peculiarly like those of the Glas-Bheinn type peeping out from below the Torridon Sandstone, would seem to indicate that the conclusions which have been arrived at as to there being evidence of a continuous upward succession in this area are erroneous. 7. Attadale, Loch Carron. In the Attadale valley, which runs in a direction nearly east and west from the east side of Loch Carron towards its upper end, an unusually good exposure of the gneiss rocks and mica-schists of the Ben-Fyn type may be examined. In the railway-cutting (as we approach the railway station from Strathcarron) rather massive beds are exposed, dipping at an angle of about 45° to a little north of east. These gneisses contain a con- siderable amount of pinkish felspar in association with brown mica and quartz. Veins are occasionally found also traversing these beds. Between Strathcarron and this point the ends of the beds are fre- quently exposed, as if sharply cut off by the fault of the Loch-Carron valley, and deeper beds are found here than in the hills directly east of Strathcarron Station. In ascending the Attadale valley along the north side, the next series to be noticed, after those characteristic of the entrance, No. 18, are some dark mica-schists, No.19. These are considerably contorted and are well exposed near a farm-house on the left-hand side of the road. The general dip here, though there are some minor folds, is still to the east. Beyond this point augen-gneisses identical with those found at Ben Fyn are met with; these are succeeded by dark grey gneisses and mica-schists, and towards the upper end of the valley by the more quartzose gneisses, Nos. 20 and 21. These last are evidently the same gneisses as those found so well exposed in the mountains directly to the east of the Strathcarron Station, where they are seen dipping at a high angle to the east. After traversing these mountains in various directions, I met with no rocks that could in any way be looked upon as otherwise than typical of a highly metamorphic series. Bed by bed they may be examined, and in sections showing © thicknesses of several thousands of feet. An upward sequence is readily made out, and the several minor series show everywhere, as _ nearly as can be conceived possible, an identical state of alteration. After a time the observer cannot fail to realize in these areas that he is meeting continually with series having a very wide distribu- tion, many of them also attaining to great thicknesses, but repeated © here and there in great folds. The amount of contortion is not great, except in the more micaceous beds ; and there is not apparently OVERLYING ROCKS OF ROSS AND INVERNESS. ya any very marked difference in the crystalline condition where they are locally disturbed. ‘The succession, as made out in these areas, would indicate that rocks of the Glas-Bheinn type (greenish hornblendic gneisses and pinkish felspathic gneisses) are the lowest ; that the next are the greenish micaceous schists and the augen-gneisses with black mica; and that these are followed by the grey gneisses and silvery mica-schists and the very quartzose varieties found towards the upper end of the Attadale valley. Hast of the last-mentioned points the beds present a more decidedly reversed dip to the N.W., showing indications of a great synclinal fold. 8. Strome Ferry and Loch-Alsh Promontory (fig. 3). The rocks exposed along the east side of Loch Carron, in travelling from Attadale to Strome Ferry, may be grouped for some distance with those found towards the entrance of the Attadale valley. They are thrown somewhat back by the fault, but otherwise retain a similar strike to those further north. About midway between the two points the beds appear to be repeated in one or more folds. As Fig. 3.—Section in Loch-Alsh Promontory. (Scale 4 inch to 1 mile.) W. E. Durinish. ‘Faults. Craig More. Ling River. (HY .,J. ix C ec. Gairloch and Ben-Fyn Series. d. Torridon Sandstone. we approach Strome Ferry dark-green schistose rocks prevail. At and immediately to the south-west of Strome Ferry the rocks described under Nos. 13-17 may be said to be the chief types. Augen- gneisses and hornblendic schists, like those found between Jeantown and Kishorn, occur for some distance to the west of Strome Ferry, and may be examined on the shore. The hornblendic rocks are freely traversed by segregation-veins of pinkish felspar and quartz, and thin lines of a dull-coloured felspar are also frequently met with. ‘Ihese rocks have altogether an old look; but some of the thinner schistose rocks to the east do not show an equally crystal- line condition, though evidently true schists. It is probable that a secondary change has taken place in some of these, and that their apparent want of crystallization is due to a kind of decomposition. The reddish augen-gneisses are found about a mile below Strome Ferry, on the road to Duncraig; but I was unable to trace them in travelling westward of that point. I believe that the floor is here dropped by a fault which seems to pass in a §.8.H. direction across the promontory ; and this fault also tends to make the newer beds dip, as it were, towards and under the older rocks (see fig. 3). About this point also there are evidences of some minor faults. Together these faults seem to have dropped the whole of the Limestone and nearly all of the Quartzite series ; and the beds which are found west of this point should, I think, be grouped altogether 154 DR. H. HICKS ON THE METAMORPHIC AND with the Torridon series. Mountains of considerable height appear to be completely made up of thick beds of sandstone interstratified with more flaggy beds. Nos. 28 and 29 may be looked upon as characteristic of those found in this area. On reaching the coast of Loch Alsh, near Balmacarra Hotel, reddish sandstones like No. 30 are found. The promontory west of the line of fault already men- tioned consists therefore mainly of Torridon Sandstone, with little or no signs of important alteration, but with slaty and flagey bands showing indications of cleavage as the result of pressure. The rocks beyond the fault towards the upper end of Loch Alsh are almost identical in character with those that have been described as occurring about Strome Ferry. Last of this pomt, along the shores of Loch Duich, gneisses of the Ben-Fyn type are found, and associated with these are some bands of highly crystalline limestone. These are separated, according to Murchison*, by “ talcose, actino- litic, and micaceous schists, often serpentinous like the limestones themselves ; red felspar-porphyry and syenite also occur in bosses, dykes, and veins.” Though I was unable to visit the sections along the shores of Loch Duich, I feel satisfied from the descriptions that have been given of the rocks by Murchison and Geikie and by Prof. Nicol, and from an examination which I have made of the specimens deposited from these areas in the Geological Museum, Jermyn Street, that they must be older than the Torridon Sandstone. They are much like specimens which I collected along the shores of Loch Eil, from rocks underlying sandstones which must, I believe, also belong either to the Torridon or Quartzite series. These Loch- Duich and Loch-Eil types are probably the newest rocks of the Ben- Fyn type found on both sides of the axis of the older rocks as shown in the map (Pl. VI.). The main object I had in view in my explora- tion in these areas was to endeavour to trace the actual conditions at those points where it had been supposed there was clear evidence of a gradual passage from unaltered fossiliferous rocks to those but partially changed, and afterwards from the latter to the highly erys- talline schists found in the more central areas. I did not therefore think it necessary to examine many sections south of the point last described, as all who have written on those areas state clearly that metamorphic series only occur there; and the specimens I have examined, which have been collected from those areas, prove this very conclusively. The only other section therefore that I need refer to in this paper is one I examined carefully to see the connexion between the very highly crystalline series of the more central portions of this district and the apparently less highly altered rocks directly west of the Caledonian Canal and to the east of the axis. Several sections across, from the west coast to the line of the canal, have been carefully described by Murchison and Geikie, and in these sections they invariably show that the more crystalline rocks are found in a line from the neighbourhood of Loch Shiel (by Loch Quoich and Glen Shiel) to Loch Affrick ; and to the N.N.E. I have recognized the same types also as far north as the neighbourhood of Loch Luichart. * Quart. Journ. Geol. Soc. vol. xvii. p. 198. OVERLYING ROCKS OF ROSS AND INVERNESS. 9. Loch Shiel to Caledonian Canal (fig. 4). In figure 4 I show the general arrange- ment of the rocks between Glen Finnan, at the head of Loch Shiel, along the north shore of Loch Eil to the Caledonian Canal. The more highly crystalline rocks are well seen in the mountains about and directly to the east of Glen Finnan, as well as along the roadside. I ascended one of these moun- tains, called Ben Nan Tom, and found it to be composed entirely of massive-looking gneisses, frequently hornblendic, and con- taining lenticular segregations and bands of hornblende, of a reddish and light-coloured felspar mixed with quartz, and of black mica. The hornblendic bands are described in note 22, and the gneisses from this point in note 23 of the Appendix. The rocks are beautifully contorted in places, and thestrike in the folia- tion is from N.W.to8.E., orvarying from that to N.andS. In many respects these rocks are much like those found between Poolewe and Gairloch, on the west coast; and it seems impossible to conceive that they can be, as suggested by Murchison and Geikie, only Silurian beds in an altered and crumpled condition. The interpretation as given by them occurs in the following passage :— ‘¢ We have shown that the quartz-rocks and limestones of Sutherland range south-west- wards through Ross-shire into the Isle of Skye,—that they are covered by a vast series of micaceous flaggy or gneissose schists ; that these are disposed. as a great synclinal trough, the centre of which traverses the head of Glen Shiel, the middle of Loch Quoich, and the watershed of Glen Finnan, —and that, by the curving of this trough, the quartzose beds which form its outer or lower edge along the western coast at Arisaig are brought up again along the line of the Great Glen”*. The highest beds are therefore placed by them at the point where we meet with the most highly altered rocks; and the least altered mica- ceous flaggy beds and the unaltered rocks are placed at the base of the trough to reach the surface along the line already described * Quart. Journ. Geol. Soe. vol. xvii. p. 207. ‘soltog uA q-tlog pu qoopatey ‘oa ‘soldeg [EIYg-Yyoory *g “OUOJSPURG uop LOT, ‘Pp Soy = 5 B £5 2 e =| a Be oe BP da a QR iw) S S S oa > on S St SU EEE war iae =r ® Ea as eo o = R = SS So isn} | 2 Pe Dd om of by S a hee Pare. i (pte 8 Q ~~ D Sg, S —_ S, = P~ 2 = ~ SS (o[fur [ 0} your F oyeog) Banavyie. Caledonian bd Canal. » S 156 DR. H. HICKS ON THE METAMORPHIC AND to the N.W. on the one hand and along the line of the Caledonian Canal on the other. Such a condition of things would seem pecu-— liarly anomalous if true; but fortunately the evidence, when care- fully examined, does not tend to bear out this strange interpretation, and a more natural one, and one more in accordance with recent views, is found to be the true one. Instead of a great synclinal with the most altered rocks held up by the less altered, the interpretation, as read by me, is that we have here a great, but much broken, anti- clinal fold, and that these highly metamorphosed rocks formed an axis which threw off the newer beds on either side, or that these rocks are, if conformable with the series to the N.W. and S.E., the oldest rocks and at the base of the whole succession in this area. These older rocks also have been exposed by the denudation of the newer rocks, and the latter are dropped on either side by faults, in more or less broken synclinals. There is ample evidence along the N.W. to show that the newer or Ben-Fyn series are repeated in broken folds, and that they dip as they approach the axis, freyuently away from it. They do this clearly also as seen in the section on the 8.E. side; but here the N.E. and 8.W. fault has somewhat interfered with the order where they actually meet. In tracing the section eastward from Glen Finnan, rocks more nearly allied to the Gairloch and Ben-I’yn types (no, 24) are met with. In the line of Glen Fionn, at the head of Loch Kil, a granitic-looking rock is seen, not unlike that found in Glen Logan (the so-called Logan rock), and its association here with these old gneisses is interesting. Crossing the fault, which is a most marked one at this point, we come rather suddenly on newer-looking rocks, still a metamorphic series, but evidently of a newer type altogether than those found in the area between Glen Fionn and Glen Finnan. These are the rocks supposed by Murchison and Geikie to dip under the latter ; but the evidence of a fault here is most marked, and, though the beds appear to dip towards the axis, the effect is clearly due to the fault. There is also some difference in the strike, as the newer rocks dip decidedly to the N.W. and afterwards to the 8.E. ' Grey micaceous gneisses, corrugated mica-schists, and strongly bedded quartzose gneisses are the prevailing types inthis area. Near Fassfern rather thick sandstone-beds are found dipping at a very low angle to the 8.E. These are clearly but little altered, and must, I think, be classed either with the quartzite series of the west or with the Torridon Sandstone. They repose upon the mica-schist series, and appear to have no direct relationship with the latter. They occur here in a faulted synclinal of the schistose series ; so their actual position, whether as resting unconformably upon the schists or dropped amongst them by faults, is not quite clear. About Kal- mallie the schistose series is again very well exposed, here dipping to the N.W. The rocks at this point consist of highly micaceous gneisses and mica-schists (no. 25), also some talcose and serpentinous schists. These are a truly metamorphic series, and cannot be differ- entiated from the Ben-Fyn types, but possibly should be classed with the newer portions of that series. The rock no. 26, evidently an OVERLYING ROCKS OF ROSS AND INVERNESS. V7 igneous one, is found on the Loch side of the road. The schists can be traced along the road towards Banavie; but before we reach the latter place a granitic-looking rock is met with, which, however, shows rather a gneissose appearance in places; this is described in the Ap- pendix, note 27. It has an old look, and if not of a gneissose cha- racter, it must, I think, be an igneous rock of Pre-Cambrian age which has suffered a considerable amount of crushing and some change. In the area west of Glen Finnan, as along the lines of that district directly to the N., the gneisses and schists are of the true Ben-Fyn type. A collection of these may be examined in the museum in Jermyn Street. Other specimens in that museum show clearly that the old axis referred to above extends further south than the neighbourhood of Loch Shiel, and that rocks of the Glen- Finnan type are exposed at Strontian, almost directly to the south, in Argyleshire. 2 10. Conclusions. The facts derived from the careful examination of the various sections referred to in this paper do not appear to me to lend sup- port in any way to the view propounded by Murchison and Geikie that the crystalline schists of their eastern areas repose conformably upon unaltered rocks containing Lower Silurian fossils. The supposed passage from unaltered to highly crystalline rocks has proved in each case, on examination, to be a deceptive appearance due either to a faulted junction or to some other accidental cause. The strati- graphical evidence therefore on the strength of which the whole theory depends for support fails entirely. To the east of Loch Maree it has been shown that the fossiliferous sandstones are suc- ceeded by flaggy beds at a low angle, which dip away from them to the south-east. ‘These flaggy rocks show, under the microscope, a slight amount of alteration; but the individual fragments out of which they have been built up are always easily recognizable, and there is no indication of that intimate crystallization of the felspar and quartz which is so characteristic of the true schists. The flaggy rocks meet the crystalline schists, which we recognize as belonging to the Ben-Fyn type, quite abruptly along a line to the east, and they are undoubtedly newer rocks than the latter. There is also generally a marked discordance in the strike of the two groups. At Achnashellach and Loch Doule, to the south of the lines last men- tioned, the quartz rocks with Annelid-tubes abut against the same crystalline schists of the Ben-Fyn type, the limestone and flaggy series recognized in the other areas being entirely absent. Further south, however, near the head of Loch Carron, the limestone reposing on quartz rocks with Annelid-tubes in abundance is brought against the crystalline schists along its eastern margin by a fault. The eastward dip which prevails in the unaltered rocks aswell as in the schists along this area has led to the belief that there was here a conformable upward succession between the series. This appearance, however, has been entirely produced by the fault; and the apparent conformability does not prove to be any thing like so marked as has 158 DR. H. HICKS ON THE METAMORPHIC AND been supposed, the strike in the crystalline schists being about due north and south, whilst in the unaltered rocks it is from north- east to south-west. At Loch Kishorn, much further to the west, the same limestone is thrust in among the schistose series by faults; whilst in the Loch-Alsh promontory the whole of the limestone and mostof the quartz rocks have been cut off by faults, and beds belonging to the Torridon Sandstone series have been brought against and apparently made to dip under schists identical with those immediately in contact with the Loch-Kishorn limestone. The evidence there- fore of great dislocations of the strata along these lines is most marked, and it is along these broken lines that the so-called gradual passage from unaltered to highly crystalline strata has been supposed to be seen. With the exception of the fiat-bedded series found on either side of Glen Docherty there is scarcely any evidence whatever in these areas of the presence of rocks which can be classified as newer than the Limestone series. The least altered of those which are included in this paper in groups older than the Torridon Sandstone are more highly crystalline than are the majority of the Pebidian rocks of Wales, or of the Huronian rocks of Canada, each of these being undoubtedly of Archean age. And by very far the largest proportion are equally crystalline with those found in the western series where overlain py Torridon Sandstone and also with the rocks characteristic of the groups found in the Hebrides. Were we also to exclude from consideration those which do not show an intimately crystalline condition, the chief conclusions arrived at would be still the same. ‘There are clear indications in the so-called eastern as well as in the western area of several well-marked series in the schists. The main groups recognizable along the north-west coasts can be made out with equal clearness in the central areas. The same types are found to succeed one another ; the same segregation-veins of hornblende, of yuartz and pink felspar, and of black mica, and the same disseminated minerals which are in any way found to be cha- racteristic of groups along the west coast are found equally abundantly in the schists of the so-called eastern areas. The most massive and most highly crystalline of the so-called eastern rocks are found towards the base of the series and in the most central portions of the area, and the more evenly bedded ones thrown off in broken folds towards the east and west. This interpretation, as explained in describing fig. 4, entirely reverses the order given by Sir R. Murchison and Prof. Geikie, who have maintained that the most highly crystalline rocks occur in a great synclinal trough supported by the fossiliferous quartzose series. ‘I'he idea that the so-called eastern rocks retain a more regularly bedded appearance throughout than is usual in Archean rocks is not borne out by examination. That a very regular stratification is well marked in a considerable proportion of these gneisses is perfectly true; but that the same may be said of a large proportion of the gneisses in the so-called western areas is equally true: similar evenly bedded gneisses, it is well known, are abuudantly present also in the Laurentian rocks of Canada. Such a high state of crystallization as is found throughout in the evenly OVERLYING ROCKS OF ROSS AND INVERNESS. 159 bedded gneisses of the central areas of Ben Fyn, Mulart, &c. is itself one of the strongest arguments which can be adduced in proof of the high antiquity of these rocks. The advocates of progressive metamorphism in this region have strenuously maintained that so long as the beds retain their evenly bedded character foliation remains feeble, and that they only become highly metamorphosed when they undergo rapid plications and foldings. These rocks of Ben Fyn therefore should, according to that view, show but a partial change, whilst, as will be readily seen from the notes by Prof. Bonney and Mr. T. Davies, it is proved that they are in a highly crystalline condition. Other so-called eastern rocks, however, such as those referred to in the neighbourhood of Loch Shiel, are as greatly plicated as are any in the western areas. I have divided the Archean rocks on the map into three groups, viz. the Loch- Maree, Loch-Shiel, and Ben-Fyn series ; these may be unconformable to one another, though at present the evidence of this is not conclusive. A fourth group, less altered probably than any of these, and some- what in the condition of the Pebidian rocks of Wales, may also pos- sibly be partially represented in some of these areas, as some frag- ments which occur frequently in the Torridon sandstones and breccias are more nearly allied to such a group than to any of the others specially referred to in this paper. Fragments, however, which can be identified with rocks belonging to each of the groups men- tioned are found in the conglomerates and breccias along the west coast ; and these show clearly that little or no alteration has taken place in the crystalline condition of the underlying rocks since these fragments were derived from them. The whole of the evidence obtained from these examinations tends therefore to confirm the views maintained in my former paper, that the crystalline schists of these areas must all be of Pre-Cambrian age, and that they are not the equivalents of the fossiliferous Silurian rocks of the southern Highlands and of Wales. APPENDIX. Note ov the Lirnotocican Cuaracters of a SERIES of Scotce Rocks collected by Dr. H. Hicks, F.G.S. By Prof. T. G. Bonnay, M.A., F.R.S., Sec. G.S. In the following remarks it is not my intention to attempt an ex- haustive analysis of the microscopic characters of the rock-specimens which Dr. Hicks has been good enough to entrust to me for descrip- tion. This seems to me hardly necessary after the full descriptions of rocks, in many respects similar, which have already been published by Mr. T. Davies* and by myself. Hence, in order to avoid burden- ing the pages of our Journal, I have, as far as possible, noticed in * Geol. Mag. dec. ii. vol. vii. pp. 103 &e, t Quart. Journ. Geol. Soc. vol. xxxvi. p. 93. 160 PROF, T. G. BONNEY ON THE LITHOLOGICAL detail only the points in which these specimens differed from those already familiar to me. In examining these Highland rocks (and I might say others also) T have observed three rather well-marked types, indicating stages of metamorphism. In the first it is obvious that many of the con- stituents noticed in the slide, especially those of larger size, with most of, if not all, the felspar, are original. Still many of the quartz-grains present somewhat irregular peripheries, and appear clotted or agglutinated together, as in the most highly altered quartzites, instead of presenting the definite outlines of ordinary fragments ; sometimes the smaller grains have even a chalcedonic aspect, as if of secondary origin. It is often doubtful whether the larger flakes of mica are endogenous or not, and it is quite certain that a minute flaky to fibrous mineral has been produced subse- quently to the deposition of the rock. This is probably, in part at least, a hydrous potash- or soda-mica (very like that which has been recognized as sericite), and in part perhaps is nearer to actinolite. Epidote and occasionally small garnets of secondary origin are present. Of this stage of metamorphism the rocks forming the northern escarpment of the “‘ Newer Gneiss,” in the neighbourhood of the head of Loch Maree, furnish excellent examples. In the second stage of metamorphism, while, when we regard the rock in the field, we can have no doubt of its sedimentary origin, bedding being often well marked and foliation distinct, yet, under the microscope, it is extremely difficult to identify any of the con- stituents, in their present condition, as of clastic origin. The . quartz presents the appearances above described yet more markedly. One might have hoped to have recognized, as has been done in quartzites, original grains as nuclei; but in this I have not yet succeeded, though I have examined a fair number of specimens. The mica, garnets, &c. are almost certainly endogenous; and this appears also to be the case (though on this point I speak hesitatingly) with the felspar where present. Of this stage of metamorphism, which, so far as one can tell, is as complete as can be, the rocks of the sonthern face of Ben Fyn furnish excellent examples. In Eng- land I may cite as instances the schists of the Lizard peninsula, and a considerable number of those in Anglesey, though some which have been sent to me from that island belong rather to the former type. Perce these types intermediate instances will of course be found. The nature of the constituents and the mode in which the agents of metamorphism have operated must bring about varieties of results, and it would be extremely rash to attempt in every case a classification by microscopic evidence alone; cases there are and will be where “noscitur a soctis” will be true, and we must then decide mainly by evidence obtainable by field-work. I do not mean by this to say that the two methods of investigation will pro- duce contradictory results, but that the microscopist, while certain that he has before him a distinctly metamorphic rock, will not venture to say to what extent alteration has taken place. I have ‘CHARACTERS OF A SERIES OF SCOTCH ROCKS. 161 found myself in this difficulty with a few of the specimens described in this paper. They may be rather exceptional instances of the one or the other type, and with that conclusion J must leave Dr. Hicks to classify them. They will neither contradict nor confirm strongly any stratigraphical theory which he may have formed. The above difficulty may be partly due to the fact that in certain cases the High- land rocks have undergone great local crushing, I believe, always in the vicinity of faults. They thus present a fragmental structure ; and it is by no means easy when some mineral changes have sub- sequently taken place to decide whether we are dealing with a rock of clastic origin, in the ordinary sense of the term, z. ¢. resulting from the denudation of one of the older gneisses, or with one of the latter, which has been crushed in sctw and recemented. No doubt the point could in every case be settled by making a larger collection of specimens, or by having several slides cut from different parts of the same specimen; but I cannot venture always to decide it from the examination of a piece no bigger than a shilling, and it hardly seems worth while, where a crucial point is not involved, to go to considerable expense. Nowhere have I seen such numerous and marked instances of this local crushing as in the Highlands; but I have observed nothing to countenance Mr. Mallet’s theory of vul- canicity. It is by no means impossible that the apparently close resemblance occasionally observed between the above-named two groups of rocks may be due to the fact that the older has in many cases supplied the materials of the newer. Under these circumstances, if the latter had been exposed to much pressure and some chemical change, it would be by no means easy to separate the one from the other, even under the microscope. Now the Torridon sandstone and quartz- ite in the Loch-Maree region certainly derive the bulk of their materials from the old gneiss rocks described below as the third type; but the flagey ‘‘ newer gneisses,” such as those in the escarp- ment near Loch Maree, seem to be made up of the débris from a schistose series, like that of Ben Fyn. For instance, the quartz in the older gneiss is very full of minute enclosures, many of them resembling irregular empty cavities, though occasionally very small bubbles may be detected, especially in the more minute and regular in form; so is that in the Torridon and quartzite (as a rule). The quartz in rocks of the Ben-Fyn type has comparatively few of these enclosures, so has that in the newest series. Mica also is common in the latter two, rarer in the former two rocks. The third type, while agreeing with those described under the second head, as being metamorphosed to the highest degree, appears to differ in respects which can hardly be due to a mere prolongation of the metamorphic action. The bedding of these rocks is ill marked ; they are coarsely crystalline and often granitoid in aspect, being then difficult to distinguish from rocks of igneous origin ; and the same is true of their microscopic structures. In such cases, in the present stage of our knowledge (though I do not think it will be so always), we must be content to be sometimes uncertain whether we have 162 PROF. T. G. BONNEY ON THE LITHOLOGICAL — before us a granite or a gneiss. Examples of this class are the coarse gneisses of the Hebridean series, which underlie the Tor- ridon Sandstone and many of the Malvernian rocks of England. At the same time it must be remembered that now and then beds more distinctly foliated also occur in this series. Naturally we should expect that as a rule the above distinctions would have a certain chronological value, and thus we are justified in using them, in default of other evidence and with due caution, for purposes of classification. In the following notes I have, for convenience of reference, followed the numbering which Dr. Hicks had placed upon the specimens, but have added L., IL., or III. ac- cording to the type of which the specimen reminded me, inserting a qualification where needed. 1 (Glen Logan). Principal minerals, quartz, felspar a good deal decomposed, but a considerable amount of plagioclase still discernible, generally extinguishing at small angles with the twin plane. In parts of the slide is a mineral which, at first sight, has exactly the aspect of grains of olivine in process of conversion into a greenish serpentinous mineral. Further examination, however, shows that the former mineral is garnet. The latter appears in part isotropic, in part shows filmy streaks of pale bluish light. I do not, however, observe here either the transverse microcrystalline structure of the “strings” or the opacite clotting so common in serpentine when formed from olivine. Granules of an earthy-looking mineral are probably formed by the aluminous constituent of the garnet, which, as these are not abundant, was probably Bredbergite rather than pyrope. So far as my experience goes, this replacement of garnet by a serpentinous mineral is very rare* (III.). 2 (north side of Glen Logan, eastern branch). Quartz, mica (both white and brown, the latter partly altered into a green micaceous mineral), some felspar (rather decomposed), and a good many garnets almost colourless, but with many microlithic enclosures (I1.). 3 (same locality). Generally similar, but with numerous small granules of a mineral which I take to be epidote. 4 (same locality). Macroscopically appears to be a compact gneissic rock; microscopically it exhibits a ground-mass, consisting mainly of quartz granules in which are scattered flakes of mica exhibiting a certain foliation, and grains, generally somewhat irregular in outline, of a rather decomposed felspar. The last look asif they might bear record of original constituents ; but on the whole the rock appears to me to agree best with those in Il. and to be more highly altered than is usual in L., the outlines of the quartz grains in the eround-mass being very irregular. 5 (south side of Loch Roshk). Macroscopically a not very fissile mica-schist ; microscopically quartz, white and brown mica, with one or two small garnets; very typical example of IT. 6 (top of hill-road to L. Coulan from Achnashellach). A rather fissile mica-schist, rich in a dark lead-coloured mica, very distinctly foliated. Two micas, both, I suspect, more or less hydrous, are present, * See Rosenbusch, ‘Mikroskopische Physiographie,’ vol. i. p. 163. CHARACTERS OF A SERIES OF SCOTCH ROCKS. 163 with a little chlorite in parts of the slide, a fair amount of epidote, and probably a little cyanite. Probably IJ.; but I think there has been some crushing which has given rise to difficulties. 7 (on side of hill, north of Achnashellach). A rather compressed micaceous schist, difficult to classify macroscopically. Microscopi- cally it consists mainly of quartz granules and flakes of pale green mica like that often described, with epidote, a little felspar, iron- oxide, &c. There are certainly indications of an original frag- mental structure, and the rock generally resembles No. I. series ; but there has been a good deal of metamorphism. 8 (the same locality). The microscopic differences from the last one are only varietal. 9 (towards base of hill north of Achnashellach). A compact reddish and dull greenish gneissic rock, seemingly highly altered and even microporphyritic ; but seen under the microscope a frag- mental structure is at-the first glance as conspicuous as in any of the No. I. series. At the same time, a more careful study gives rise to doubts as to whether this resemblance may not be illusory as regards its origin, and whether we have not here an instance of one of the older series locally crushed. At first I inclined to regard it as a member of the newer series; but repeated examination has. produced considerable doubt on this point. Probably the question could be settled by cutting a series of slices from different parts of the block; but as I learn this is not a matter of great importance, I have thought the expense needless, and must leave the matter in uncertainty. 10 (shore of Loch Carron, north of Jeantown). A dark, rather heavy, moderately fissile mica-schist. Microscopically it exhibits brown mica in addition to that described above, contains little fel- spar, but some garnets, and appears to have undergone considerable compression at right angles to the planes of foliation. It has a good deal in common with the last three, but seems yet more highly altered. It reminds me much of a mica-schist which I have described and figured from the head of Glen Docherty. 11 (Glas Bheinn). This consists of a dull green hornblendic or chlo- ritic mineral, parted by irregular cherty layers of variable thickness. The green mineral proves to be a well-crystallized hornblende, with characteristic cleavage. The cherty layers consist of grains of decomposed felspar crowded with micaceous films and granules of andalusite (?) and of quartz. The rock is much metamorphosed : but there has evidently been local crushing, infiltration, and secon- dary change to such an extent that it is difficult to say whether it is more typical of II. or III. 12 (valley between Jeantown and Kishorn, near Bridge). A handsome pale reddish augen-gneiss, the ‘“‘ eyes ” being of felspar, and the ground-mass rather compact-looking. Macroscepically one would not hesitate to refer it to series II.; but microscopically the struc- ture differs from what one expects: in a ground-mass of quartz and a very pale brown mica, together with a few felspar granules, occur irregular grains of felspar; the edges of these are, indeed, as it Q.J.G.8. No. 154. N 164 PROF. T. G. BONNEY ON THE LITHOLOGICAL were, fused with the ground-mass, and the granules of it one with another. Here, indeed, we seem to have a record of an original structure; and it is noteworthy that the felspar closely resembles that occurring, certainly in derivative fragments, in group I. Still I think the rock, on the whole, is nearer to I1.* 13 (a mile south of Strome, on road to Duncraig). Like 11, this has undergone so much crushing and recementation that it is difficult to come to a conclusion about it. The felspar is crowded with mica films and secondary minerals, some possibly fibrolite. 14 (shore of Loch Carron, south of Strome Ferry). A compact rather flagey dull green and reddish schist, consisting microscopically of quartz and the usual mica with microliths of hornblende, and occasional larger grains of felspar and quartz. Structurally this rather resembles series I.; but the alteration is considerable, so that field evidence must decide whether it be grouped with it or with II. 15 (at Strome Ferry). A dull green slightly schistose rock rich in hornblende. Microscopically it consists of quartz and hornblende— the latter well cleaved, but irregular in external form, with a fair amount of sphene, a little felspar and numerous secondary micro- liths, aluminous and magnesian. On the whole I think this most resembles II. 16 (the same locality). A dull green slightly schistose rock, streaked with pale red and yellowish green. Microscopically it consists mainly of quartz, hornblende, and epidote, often ratherimpure. For mineral condition cf. No. 15. The rock has been crushed, which at any rate partly, perhaps wholly, accounts for its fragmental aspect. 17 (the same locality). A claret-red schistose rock, with little whitish specks. The general microscopic structure resembles No. 4; but there is less quartz, a good deal of ferrite, and considerable evidence of decomposition. At Dr. Hicks’s request I have paid particular attention to the structure of these rocks (13-17). They present similar difficulties to No. 12; but on the whole I think they group best with IL., though here and there one seems able to identify original constituents more easily than is usual in this division. 18 (hill north of Attadale Station). A reddish, rather compact gneiss, poor in mica. Its microscopic structure is thoroughly cha- racteristic of series IJ.; it contains two kinds of mica (neither abundant), a very few garnets, and a little epidote. 19 (north side of Attadale Valley, near Farm). Macroscopically and microscopically very like No. 10, but perhaps a little more altered, so that we may with more confidence group this with II. A little epidote is present. 20 (roadside, east of above). A moderately fine-grained gneiss, * Since writing the above, I have had the opportunity of examining another rock from this group, a rather compact-looking pinkish gneiss. Here, too, the rock, while macroscopically and, in some respects, microscopically, resembling series IT., has certainly a fragmental aspect. Though the evidence is not decisive, I strongly incline to consider this aspect illusory and indicative only of subse- quent crushing, and to refer both the rocks to the older series. . CHARACTERS OF A SERIES OF SCOTCH ROCKS. 165 slightly porphyritic, of series Il. ; almost identical with rocks I have described from Ben Fyn. 21 (towards upper end of Attadale Valley, north side). A finer, less micaceous gneiss ; but microscopically it has only varietal diffe- rences from No. 20, so II. 22 (Ben Nan Tom, head of Loch Shiel). A heavy black horn- biende rock, with a parallel structure resembling one extremely metamorphosed. Under the microscope it exhibits numerous dark- ereen hornblende crystals, well cleaved and sometimes affording definite external faces; with these are brown mica, sphene, and grains of quartz, and of a plagioclase felspar, which has a rather large extinction-angle. The structure of this rock reminds me strongly (except that garnets are absent) of one described from Ben Fyn which was certainly intrusive; and I cannot help suspecting it, notwithstanding its gneissic aspect, to be really a diorite. 23 (Ben Nan Tom). A handsome, rather coarse, pinkish and dull greenish gneiss. Consists of quartz, felspar and dark mica, with a little epidote, apatite, and garnet. The mica is partly replaced by a dull-green chloritic mineral ; and between its cleavage-planes are occasional plates of a clear felspathic mineral. The curious ver- micular or micrographic structure often noted in highly altered gneisses is present in parts of the slide. JI should unhesitatingly class this with III. 24 (Druim na Saille, head of Loch Hil). A fine-grained gneiss of the same type as 20, and, like it, a good example of II.; contains sphene, epidote (?), and, I think, a little apatite. 25 (Kilmallie, Loch Hil). A brownish silvery mica-schist. Ex- amined microscopically, it shows so much felspar (rather decomposed) as to make the term gneiss more correct; both brown and white mica are present; the former a little altered. The micas are some- times interlaminated, and the brown exhibits the inclusions noted in No. 23; some apatite is present. Decidedly I. 26 (shore of Loch Kil, west of Kilmallie). A rather compact dull green rock. Consists microscopically of a plagioclastic felspar, with large extinction-angles, often much altered, and rather pale hornblende. There has been a little brown mica ; ; and there is a fair quantity of sphene. I have no doubt the rock is an igneous one, and, as I regard the hornblende as a secondary product, name it a hornblende- diabase. 27 (Banavie). A reddish felspathic granitoid rock. Microscopi- cally consists of quartz, felspar (orthoclase, microcline, and a plagio- clase), with a little rather decomposed brown mica. The structure is somewhat exceptional ; and I have had doubts whether this may not be a granitoid gneiss, but on a single slide, and without field- knowledge, will not venture to express a decided opinion. 28 (road to Duncraig, from Strome). A flaggy hard mudstone, looking little, if at all, metamorphosed. Under the microscope clastic quartz, felspar, and mica are very distinct; but there is a considerable quantity of very minute “ sericite,” which gives to the rock a microscopic foliation ; this implies a certain amount of meta- N 2 166 PROF. T. G. BONNEY ON THE LITHOLOGICAL morphism and may possibly suffice to enable us to group it with I. Its materials, however, are undoubtedly derived from the same source as those which have already been described in this group. 29 (near Duncraig). A darker and less flaggy rock, also appearing but little altered. Microscopically seems to be composed of the same materials, but with less sericite. The smaller quartz-granules are agelutinated ; and the rock may with less hesitation than in the last case be called métamorphic and grouped with series I. 30 (near Balmacarra Hotel). A dull red mudstone, seemingly little altered. With the microscope a certain amount of meta- morphism is perceived, as in the last case; so that it, too, may be grouped with series I. There is rather more “sericite” and ferrite: 31 (Glen Logan). A black compact limestone. This, under the microscope, is more completely crystalline than I should have ex- pected. It is somewhat dolomitic, exhibits no trace of organisms, and contaius a fair amount of opacite (graphite?) in scattered granules and streaky clots. I should expect to find it associated with rocks of the series I. type. 32 (Loch Kishorn). A dull-coloured compact limestone, traversed by numerous, irregular, thin, cherty veins. Under the microscope it appears to be an intimate mixture of calcite or dolomite with a microlithic mineral, probably quartz. The cherty veins are chalce- donic quartz. I should suggest that this also belongs to series I. group. . In conclusion I may perhaps be allowed to state that I wrote the rough draft of these notes in ignorance of the bearing of the specimens upon any theoretic views entertained by Dr. Hicks ; that, upon sending it to him, my remarks upon the amount of meta- morphism accorded, in the great majority of cases, with what his stratigraphical arrangement demanded, and differed only in a few instances where I myself was very doubtful, and had left an opening for further consideration. I know very little of the district ; so that these notes are strictly. lithological, and -I have thus offered no opinion as to the geological age of the different series. As to the vexed question of the correlation of the metamorphic rocks of Scot- land,,I would, at present, rather say no more than this—that having regard to the teaching of Wales, Cornwall, and the Alps, very clear evidence will be needed before we can accept the dominant rocks of the Central Highlands as of Lower Silurian age. EXPLANATION OF PLATE VI. Map showing the Archean and overlying rocks of parts of Ross- and Inverness- shires.—Scale 4 miles to 1 inch. ‘SS3 NY JAN | a NV ssoy 4 O SLu Vd 40 dV e CHE, DTUG i 73 YRT pyoypsale ¢: COUGm y yf a a 1 all li | i ; ii att alti vedi “apn t ane ANH RANA ANARDD : Masi o§ IP 36 Convexastreea Waltoni......... PE Pe || cl en ieee? Montlivaltia Slatteri ......... sh Mes Ns ci ol eel Mie al aan Ma fairfordensis. ......225<<: cb Nate 2 le ea a earyophyllata ............ ean eee rae He Se ok i OE Calamophyllia radiata ...... Je 3) hl oak i Thecosmilia Slatteri............ ee eames ctl RO | o.oo Tae 174 R. F. TOMES ON THE GREAT-OOLITE MADREPORARIA. LS De Seale) OE a Cladophyilia Babeana ......... creel Rae abe ad ieee teal Confusastrexa magnifica ...... seni eesiat weceithy seth peel cee Tsastraea limitata ..............- a ere Bl ae Ne Saag ee PIMOS s cgadeee cusseteeens Sorel Recall. | 0GES NPL Cen tee eae me microphylla) -seeeceesees.ee sha lA ae alte se LEYS IF ae yen aetna ctl settee tle A aces | ake * explamulatia ......cec-as0s aid pcan cfNI EAR Sec Alka * Latimeandra lotharinga ...... si bests Se ia % Chorisastreea obtusa............ soa MN Cigale ay Oe % Astroccenia Phillipsi............ vet. '| one healt ee BIE Anabacia complanata ......... Pmt econ % Tricycloseris limax .......... silt duis iliaseciyill samen slice) cee fein EE Thamnastrea mammosa ...... sees | een) wee ol Ace fase! See liyelivOR CS Scnerce ces otecce bok reve tl” seen Pee * — Waltoni .................. cael, el MER ee * —— microphylla... ......2:0: eiey (i reeiee ite tally oieoaly aa Oroseris Slatterie «....sacc-eter. 4 * Comoseris vermicularis ... .. Sse a Microsolena excelsa ............ cat | eee | ORE SI \ae: S| ese? «een Ui ——=- repularis® .....2...ceeeceds CPE ec a a | iran err ce | Column no. 1 includes the species found at Glympton, no. 2 those from Epwell, no. 3 those found in the Rollright cutting, no. 4 the Stonesfield species, no. 5 the Burford ones, no. 6 those from Ayl- worth, and the 7th and last column is devoted to the Fairford corals. On looking over the foregoing lists of species it will very readily be observed that there is a strong general resemblance existing between those from all the localities given. This remark will apply to both genera and species. One species, Jsastreea limitata, is recorded as occurring in six out of the seven localities; while Microsolena excelsa and Thamnastrea Lyell, have been observed in four of the seven; and Cladophyllia Babeana and Thamnastrea mammosa have been taken from three localities. Bearing in mind the great general similarity of the species in these lists, and remembering also that they occur in beds occupying very different stratigraphical positions in the Great Oolite, we are un- able to conclude that these coral-beds are any thing more than the repetition of each other, and that no satisfactory division of the Great Oolite could be made by means of the coral-fauna. In this respect it differs much from the Inferior -Oolite, the coral-beds in which, as I have elsewhere shown, contain each a own assortment. of species. It may perhaps be well to observe that as we guraneed from Fair- ford to Burford, and thence to Stonesfield and Rollright (that is, in a more or less northerly direction), these coralline deposits occur in -regular gradations lower in the series of the Great Oolite. To this, however, very little importance must be attached, as we cannot as- sume that deposits of corals may not have existed above those of Rollright and Stonesfield, or that they do not now exist, hidden in the strata, below those of Burford and Fairford. R. F. TOMES ON THE GREAT-OOLITE MADREPORABIA. ETS -ZOANTHARIA APOROSA. Family OCULINIDZ. Genus EnaLtonetta, d’Orbigny. M.de Fromentel, in his general work on Fossil Corals (Introduction 4 Etude des Polyp. Fossiles), makes an important addition to the definition of this genus by MM. Milne-Edwards and Haime, as given in their ‘Histoire Naturelle des Coralliaires,’ in the following words :— *‘Cloisons subenticres et présentant des lobes paliformes prés de la columelle.” The species I have now to describe appears to fall under this definition of the genus pretty accurately ; but I have failed to notice this peculiarity of the septa in specimens of either Hnallohelia compressa or H. elegans from the Corallian of Nattheim. Nor has M. de Fromentel himself, in the figure of EH. minima*, given the least indication of a paliform lobe near the columella. ENALLOHELIA cLAvATA,n. sp. Plate VII. figs. 12-14. The branches decrease in size as they ascend, but very gradually. They are smooth, but are regularly furnished over the whole of their surface with regular and delicate papille ; and the mural costs are only observable near the margin of the calices, where they correspond with the septa. The calices (fig. 14) are irregularly alternate in their position; they have a diameter equal to that of the branches, and are prominent. They are round and rather deep. The columella is small, irregular, and has little prominence. The septa are in six systems; and there are three cycles. Those of the primary cycle are of nearly equal thickness throughout, and approach very near to the columella, where they terminate in a pillar, which, when seen from above, looks like a rounded knob, and gives to the upper margin of the septa the appearance of a club. Their sides are ornamented with very distinct vertical ridges, which resemble the ridges seen on the septa of a great many of the Astrwide ; and their margins are subentire. The secondary and tertiary septa are nearly of a length, which is about half that of the primary ones. Neither the secondary nor the tertiary septa have the club- shaped termination observed in the primary ones. Diameter of the branches 13 line to 3 lines. This species bears a little resemblance to #. minima, Fromentel, in having six primary septa; but it has three cycles instead of two. From H. minima it differs by the presence of a paliform tooth or club-shaped termination of the septa. Z. crassa and &. elongata of Fromentel have only two cycles of septa, while the present new species has three. 4. crassa, however, has the septa denticulated near the columella. In four instances only has this species been met with at Fairford, where it was obtained by Miss Slatter. All are fragments. But at Broughton, near Banbury, in a quarry in the Great Oolite, * Polyp. Cor. des environs de Gray, pl. viii. fig. 7. 176 R. F. TOMES ON THE GREAT-OOLITE MADREPORARIA. I met with a specimen as long ago as in 1859, which, though too crystalline to admit of internal examination, yet has the external form well preserved, and shows that it was a thick and bushy species, and probably attained to a considerable size. Family ASTRAIDA. Subfamily HusMILIN&”. Genus Barayca@nia, n. g. The corallum is composite, compact, turbinate, and attached; and ' the corallites are intimately united by their walls. There is a common investing wall, which is costulated and some- times has bands of rudimentary epitheca. The calicular surface is superior and convex. The calices are more or less pentagonal or rounded, and deep. The septa are entire, thin, and project but little into the calice ; and when they meet those of other calices at the top of the wall they rise. into obtuse points. The primary ones meet in the bottom of the calices and form a rugged columella. At the angles where two calices meet, the walls are elevated into a kind of obtuse peak. The increase is by gemma- tion, which takes place only at the obtuse points just mentioned. There is considerable resemblance between this genus and some species of the genus Stylocenia, as S. emarciata; but it is wholly unlike other representatives of the genus, such as S. monticularza. BaTHYC@NIA SLATTeRI, n. sp. Plate VII. figs. 1, 8. The corallum has a depressed turbinate form, and was attached by a small surface, which in some instances was slightly pedun- cular. The common wall is thick, and has broad and slightly prominent costee, with occasional and rudimentary bands of epitheca. The calicular surface is convex ; and the calices (fig. 8) are penta- gonal or hexagonal, but are much rounded by the septa filling up the corners near the top of the wall. They are as deep as wide, and have rather thin and nearly vertical walls. The septa are smooth, and project very little from the walls of the calice. At the top of the wall, where they meet the septa of con- tiguous calices, they are thickened and rise into obtuse points ; and this is more especially observable of those septa which meet at the point of contact of three calices. All the septa of the outer calices are continuous with the mural coste of the common investing wall. There are six primary septa; they are thick and prominent at their outer or upper end; but they rapidly become thin and pass down the inside of the calice like a thread, and again enlarging as they pass across the bottom, unite in the centre and form a distinct but rugged columella. The septa of the second cycle also pass like threads down the R. F. TOMES ON THE GREAT-OOLITE MADREPORABIA. VRE inside of the calice, but are lost halfway down. Those of the third and only remaining cycle extend but a very little way down, and then thin off to nothing; but they, as well as all those of the other cycles, are thick at their upper and outer ends. Gemmation takes place in the angles where the calices meet; and the young calices have thick subcristiform septa (fig. 8). In some examples the common wall is horizontal; and then this coral bears considerable general resemblance to Jsastrwa moneta from the Cornbrash of Wast near Boulogne. The height of the corallum is from nine to twelve lines, and the diameter from twelve to eighteen lines. The diameter of the calices is from two to two and a half lines. At present it has been met with only in the Fairford coral-bed, and is by no means common there. BaTHYC@NIA sonia, n.sp. Plate VII. figs. 9, 10. The specimen consists of a portion only of a corallum, which, when complete, had a diameter of several inches, and appears to have been somewhat globular. The calices are evenly scattered over the upper and convex sur- face ; they are small, and nearly circular, but with a tendency to a hexagonal form, are rather deep, and regularly cup-shaped inside. The septa are thick and subcristiform where they meet with those of other calices at the top of the wall; but they speedily become thin, thread-like, and straight, and have very little projec- tion into the calice. There are three cycles ; and the septa forming the first, which are six in number, pass over the bottom of the calice and, joining in the centre, form asmall but very irregular and spurious columella. Some few septa of the second cycle do the same; but others graduate away and are lost before reaching the columella. The septa of the third cycle are very short, and appear as little more than short cristiform ridges across the top of the wall. The calices have a diameter of one line to one line and a half. I know of only one occurrence of this species. It was found by Mr. J. Windowes, of Chipping Norton, in the railway-cutting near Rollright, Oxfordshire, and very obligingly given by him to me. Genus ConvExAsTR#A, d’Orb. The existence of this genus in England was for a long time to me a matter of some doubt, no satisfactory confirmation of its occurrence (as stated by MM. Milne-Edwards and Haime) haying appeared. The doubt was due entirely to the particular condition or state of pre- servation of the specimen from which the magnified representation was taken*. A comparison of that figure with the figures of Convexastrea sexradiata given by Goldfuss , or of C. regularis of * Brit. Foss. Cor. p. 209, tab. xxiii. figs. 5, 6 (1881). t Petrefact. Germ. vol. i. p. 71, pl. xxiv. 178 R. F. TOMES ON THE GREAT-OOLITE MADREPORARIA. Klipstein*, or with the representations of casts of C. portlandica t and C. dendroideat figured by M. de Fromentel, will sufficiently explain what is here meant. There is no genus of corals with which I am acquainted which presents a greater diversity of appearance, according to its state of preservation, than Oonvewastrea; and the only representation I have yet seen which gives a correct idea of it in a perfect state is the one given by Klipstein of the St.-Cassian species, to which I have just referred. All the others have been taken either from casts or specimens which have lost their cristiform septal cost or were otherwise damaged. These peculiar septal coste are very characteristic of the genus, and distinguish it from Cryptoceenia, to which it bears considerable resemblance, but from which it also differs in not having the summits of the corallites prominent, in having the walls hidden, and in having the distal ends of the septal cost passing in between the distal extremities of those of contiguous calices. In their peculiar form, as well as in their connexion with the septa, the costze more nearly resemble those of Holocystis than those of any other genus; but the two genera are not otherwise similar. ConvexastR@A Wartont, M.-Edw. and Haime, Brit. Foss. Cor. p. 109, pl. xxiii. figs. 5, 6. It occurs, but is not common, in the Great-Oolite quarry near Burford, and in the railway-cutting near Rollright, from which localities I have collected specimens; and I have a specimen picked up from the surface of a field between Bourton-on-the Water and Northleach, Gloucestershire, not far from the latter place, I have seen corals, supposed to be of this species, which have been obtained from the neighbourhood of Stonesfield. These were nothing more than much-worn examples of the very common Thamnastrea Lyeth. Genus Cryproce@nta, d’Orb. The peculiar nature of the coenenchyma of the corals of this genus is visible in all the species I have yet seen, whether from the Oolite or Cretaceous formations. In all of them it forms an important _ and conspicuous part of the corallum, filling up completely and * symmetrically the intervals between the corallites; and in struc- ture it must be regarded as not merely porous matter to fill up with, but as a tabulated tissue from the tabule of which the young corallites spring. In another place$ I have explained my reasons for following M. de Fromentel, and regarding the genus Cryptocenia as distinct from Cyathophora; of this I shall speak more fully when I come to the latter genus. * Beit. zur geol. topogr. Kenntn. ostl. Alpen, p. 293, tab. xx. fig. 11. tT Polyp. Jurass. Szp. pl. iii. fig. 7. ¢ Polyp. Cor. des Environs de Gray. pl. xiv. fig. 4. § Proceed. Geol. Assoc. vol. vi. no. 4. R. F. TOMES ON THE GREAT-OOLITE MADREPORARIA. 179 CRrYProc@niA TUBEROSA, Dunc., sp.* Cyathophora tuberosa, Dunc. Supp. Brit. Foss. Cor. pt. ii. p. 15, pl. iu. figs. 15-18 (1872). A considerable number of specimens of this coral were obtained by Miss Slatter at Fairford; and the greater part of them are in a beautiful state of preservation. I think it almost certain that the specimen figured by Prof. Duncan was one of those collected at Fairford, for reasons which I have already mentioned, and because of its abundance there and its recorded absence from the coralline deposits in the Great Oolite. CryprocenraA Pratrrr, M.-Edw. and Haime, sp. Cyathophora Pratti, Edw. and Haime, Brit. Foss. Cor. p. 108, pl. xxi. fig. 3 (1851). It would seem that this is far from being an abundant coral; for MM. Milne-Edwards and Haime speak of having seen only three specimens, two of which came from the Great Oolite of Combe Down, near Bath, and the other one has no locality assigned to it. I have met with three instances only of its occurrence, two specimens having been found by me in the railway-cutting near Stonesfield, and the third was taken from the surface of a field at Glympton, near Woodstock, by the late Mr. Charles Faulkner, of Deddington. It was associated with a great number of examples of Jsastraa linutata. The corals referred by me to this species, in my paper in the sixth volume of the ‘ Proceedings of the Geologists’ Association,’ I have now reason to believe were incorrectly assigned to it. CRYPTOCENIA MICROPHYLLA, n. sp. Plate VII. fig. 2. The corallum is massive, more or less expanding, and has the upper surface gibbous, and in some specimens rising into dome- shaped prominences. The calices are rather thinly placed, very small, prominent, and have very thick walls. | There are two cycles of septa; the primary ones are six in number, and extend to about two thirds of the distance to the centre of the calice. ‘The secondary septa are merely rudimentary. The intercalicular coste are stout, prominent, and correspond with the cycles of septa. They are of equal size ; and when opposite those from other calices, they do not blend with them; and when placed alternately, they might almost be said to interlace with each other. The diameter of the calices is about three fourths of a line, and the spaces between them about twice that measurement. The present species differs from all the Oolite ones I have seen in having the calices smaller, the walls thicker and more prominent, the second cycle of septa rudimentary, and the intercalicular cost of equal size, and not continuous with those from other calices. * This species bears so close a resemblance to C. duciensis that it may very probably prove to be identical with it. Q.J.G.8. No. 154. 0) 180 R. F. TOMES ON THE GREAT-OOLITE MADREPORARIA. In the size of the calices it corresponds more nearly with a species from the Great Oolite of Bréqueréque, near Boulogne; but the latter differs wholly in having numerous thin and continuous septal costa. From the C. tuberosa of Prof. Duncan it differs in its much smaller calices and rudimentary second cycle of septa. It occurs and is common, though perhaps not abundant, in the Fairford coral-bed. Genus Styria, Lam. Sryrina soripa, M.-Edw. and Haime, Brit. Foss. Cor. p. 105, tab. xxi. fig. 3. Stylophora solida, M‘Coy, Ann. Mag. Nat. Hist. 2nd ser. vol. ii. p- 339 (1848). A few small specimens of this species have been obtained from the Fairford coral-bed, and are in Mr. Slatter’s collection. They appear to be somewhat worn, but nevertheless show conclusivelythat the calices in the figure given by MM. Milne-Edwards and Haime of this species are, as stated by them, a great deal too prominent. They are a little larger, too, than they are represented in the figure, and have rather thicker septa; otherwise they agree pretty exactly with the description and figure of the above-mentioned celebrated zoophytologists. STYLINA CONIFERA, M.-Edw. and Haime, Brit. Foss. Cor. p. 105, tap, xox ie.) 2. I have procured this coral in a few instances from the quarry near Burford, but do not think it 1s common there, as it is not seen in the weathered surfaces of the rugged stones of which the surrounding walls are built, and in which so many corals are observable. Subfamily astTR HIN &, Genus Monrrtvattra, Lamx. It is rather remarkable that the genus Montlhvaltia, which is usually so rich in species in the Jurassic formation, should be so poorly represented in the coral-fauna of the Great Oolite. MM. Milne-Edwards and Haime, in their great work on the fossil corals of this country, gave two species only. This very meagre list of Montli- valtice was not extended by Prof. Duncan, in his Supplement to the above work, by a single additional species of this genus, though several well-marked compound corals were added. To the two species (IM. Smithi and M. Waterhousei) made known by MM. Milne- Edwards and Haime I can now, however, add three others. One of these is the old and well-known M. caryophyllata, the species on which the genus was established by M. Lamouroux, and the remain- ing two are new. MonTLIVALTIA CARYOPHYLLATA, Lamx. Expos. méthod. des gen. des Pol. p. 76, pl. lxxix. figs. 8,.9, and 10. Plate VIL. fie de Several specimens which I have examined differ from typical BR, F. TOMES ON THE GREAT-OOLITE MADREPORARIA. 181 examples of WV. caryophyllata only in having the epitheca less regu- larly developed, and extending a little higher up the corallum. Most of the examples differ also in being a little more upright in their growth. There is an obvious error in the description given of this coral by MM. Milne-Edwards and Haime*. The number of cycles of septa given by them is five and part of a sixth. This statement, how- ever, has been corrected by M. de Fromentel, who says that there are in a large calice as many as 162 visible septa. In the calices of those I have examined there are about 108 septa. Monriivattia SLATTERI, n. sp. Plate VII. fig. 20. The corallum is broadly attached and low, and much resembles that of M. Smithi, but differs entirely from it in having the fossula linear. There is sometimes, but not always, a constriction around the corallum, about its middle, as in M. Smthi. The epitheca is strongly developed, and deeply marked with concentric wrinkles. It does not extend more than two thirds of the height of the corallum. The calice is open, a little ovoid; and the linear fossula corre- sponds with the greatest diameter. The septa are exsert, and rise from the upper margin of the epitheca vertically; and forming an angle which is less than a right angle, but which is rounded, they pass in a nearly straight, inward and downward line to the fossula. They are all extremely thin and delicate; and their margins have small and irregular points, which are not thickly placed, and become very indistinct towards the fossula. Their sides are orna- mented by a number of spots of very irregular form, which have very little prominence, and are placed in ill-defined vertical rows. These markings on the sides of the septa (fig. 20) are quite sufficient to distinguish the present species. There are as many as 130 to 135 septa, 18 of which extend quite to the fossula. The height of the corallum is from 8 lines to | inch 3 lines, and the greatest diameter of the calice 1 inch 5 lines. The length of the fossula is from 5 to 8 lines. It occurs at Fairford, but appears to be rare. MonTLIVALTIA FAIRFORDENSIS, n. sp. Plate VII. fig. 21. The corallum, when not rendered irregular by rejuvenesence, has very much the general form of an Awosmilia, but has a considerably elongated calice, and a fossula which is so much extended as to occupy fully one half of the calice. The epitheca is well developed, but thin and almost without con- centric markings. It does not extend to the edge of the calice ; and its superior margin is well defined and regular. The septa are exsert ; but the calice is deep, and the fossula well defined and open. The edges of the septa (fig. 21) are thin and serrated; and the teeth are formed by the upper termination of a row of closely placed tubercles, * Hist. Nat. Corall. t. ii. p. 303. t Paléont. Frang. Terr. Jurass. Zooph. p. 202. 02 182 R. F. TOMES ON THE GREAT-OOLITE MADREPORARIA. of an elongated and somewhat crescentic form, which are placed vertically just below the upper margin of the septum. Below this row of tubereles no markings of any kind appear, the sides of the septa being quite smooth. There are about 104 septa in a large calice. . When the upper edges of the septa of this coral have been broken off, and the broken edges worn smooth, as is often the case, the genus to which it should be referred becomes doubtful; it then resembles an Hpismilza rather than a Montliwaltia. A very few specimens only have been obtained from the Fairford coral-bed by Miss Slatter. Genus CALAMOPHYLLIA. CALAMOPHYLLIA RADIATA, M.-Edw. and Haime. It occurs in the Great Oolite of the railway-cutting near Rollright, and at Epwell, between Banbury and Brailes. At both these localities it 1s common. Genus CrapopuyriiA, M.-Edw. and Haime. It is probable that the genus Thecosmilia will have to be sub- mitted to division, the very different manner in which fissiparity takes place in the bushy and capitate forms (operating in con- formity with their variation in form) being of sufficient importance for generic distinction. If this division were made, the bush-shaped species, such as 7. Martini and TZ. Slattert, would approximate so nearly to the Cladophyllie that they would be removed to that genus, and there would be no necessity for the creation of a new one for | their reception. CLADOPHYLLIA Baprana, M.-Edw. and Haime, Pol. Foss. terr. Pal. p- 81 (1851); Brit. Foss. Cor. p. 113, tab. xxii. fig. 2 (1851). It occurs in the Great Oolite at Epwell, near Banbury, near Burford, and at Aylworth on the Cheltenham and Bourton-on-the- Water Railway. Genus THeEcosmitia, M.-Edw. and Haime. THECOSMILIA SLATTERI, ND. sp. This is a small and well-marked species, having much the aspect — of Cladophyllia Babeana. The corallites are free and branching, but crowded and forming a short and close bush. They increase by fissiparity quite rapidly, sometimes dividing three or four times on nearly the same level ; and as some, on the contrary, do so only occasionally, the corallum is crowded in some parts, while in others it is quite open. Notwith- standing the frequency of fissiparous division, the corallites main- tain their cylindrical form, and none of the calices become much elongated preparatory to division, as in such species as Thecosmilia RB. F. TOMES ON THE GREAT-OOLITE MADREPORARIA. 183 trichotoma and T. annularis. In the present species they retain their circular outline until two principal septa have met in the centre of the calice and divided it in half, or perhaps, as is occa- sionally the case, three septa have equally divided it into three triangular spaces. Externally the corallites are rather rugose, and have a well-de- veloped epitheca, marked with irregular circular constrictions, which are not, however, the effect of reyuvenescence. The calices are rounded and rather shallow. The septa are irre gular, owing to the different ages of the calices in near approxi- mation to each other, and owing also to the primary ones of some systems uniting in the centre of the calice while in others both primary and secondary septa meet and run together. They are thick, rude, and sometimes decrease in size inwards, while at other times they become thicker towards the centre, where a considerable number blend together and form a false columella. In some of the systems there are three cycles and the rudiments of a fourth; but quite as freqently there are only three. Tt appears to be not uncommon at Fairford, and is also met with near Burford. I entertain little doubt that the present species was figured by Prof. Duncan, in his ‘Supplement to the British Fossil Corals, as a . variety of Cladophyllia Babeana*. . Genus Favia, M.-Edw. and Haime. Although no revord has at present appeared uf the vccurrence of the genus favia in the English Oolites, I have for a long time been in possession of a specimen from the Stonesfield Slate. More recently, by the kindness of Mr. J. Windowes, of Chipping Norton, I have become possessed of a species of this genus which he found in the Inferior Oolite of the railway-cutting near Hook Norton, Oxfordshire, from which place I have also subsequently taken specimens. But before this, I had myself taken a Fuvia from the Inferior Oolite, in the excavation made -for iron-ore at Fawler, near Charlbury, Oxfordshire. Subsequently, that is during the present year (1882), I met with another example in the débris washed tu the bottom of a gully on the south side of the valley at Crickley Hill, near Chelten- ham. ‘These several examples of the genus have not yet received that close examination which is essential for their proper determin- ation, excepting the one from the Stonesfield Slate; but they are here mentioned as corroborative testimony of the existence of the genus in the Jurassic deposits of this country. Favia PepuNcuLarA, n.sp. Plate VII. figs. 16, 17. The corallum is small and pedunculate and has a subglobular and overhanging head. ‘The peduncle tapers downwards to an obtuse point, by which it was attached, and is furnished with regular and delicate coste. There is no epitheca. * Plate iii. figs. 1,2, and 3. 184 R. F. TOMES ON THE GREAT-OOLITE MADREPORARIA,. The calices are well defined by prominent intercalicular costa, which are thick, short, regular, and closely placed. The calices themselves are nearly circular, but become more or less oblong and irregular in form before fissiparity takes place. The septa are rather thin, but fully maintain their size as they approach the centre of the visceral cavity. There are six systems and three-complete cycles. The first cycle joins into the columella; those of the second are two thirds the length of the first; and those of the third are only a little shorter than the second, towards which they curve. Height of the corallum six lines, diameter of the calices from one: to two lines. The only specimen I have met with came from the Stonesfield Slate of Sevenhampton, and isnow in my collection. As fissiparity is oniy observable in one calice, and the growth therefore presumably slow, there is nothing to support the supposition that, as the small size would seem to indicate, it is a young individual. Genus Conrusastrma, M.-Edw. and Haime. ConFUSASTRHA BURGUNDIZ, M.-Edw. and Haime, Pol. Foss. des terr. Pal. p. 98 (1851 Astrea burgundie, Mich. Icon, p. 106, pl. 24. fig. 4, 1843. A specimen in my own cabinet which agrees with the figure given by’ Michelin of C. burgundie, excepting that the septa are somewhat stouter, was purchased of a dealer with other Great Oolite corals, and was said to have been received from Cirencester. I am unable to give further information respecting it, except to add that it has very much the appearance of a Fairford coral. ConFUSASTRHA MAGNIFICA, n.sp. Plate VII. figs. 15 & 22. One specimen only of this new and fine species was obtained at Fairford; and I describe it as follows :— The corallum is globular, and has a short and thick pedestal. Nearly the whole of the globular part is calicular. The calices are round or oval, and deep; and the septa are exsert outwardly, where they form a circular prominence. The space between the calices is flat and depressed. Thesepta are rather thin, and of nearly equal thickness throughout, and have their margins regularly denticulated, each prominence being the termination of a rib, which is vertically placed on the sides of Pe septa (fig. 22). The fossula is scarcely observable, but is a little elongated ; and the septa meet, but do not unite, in the centre of the visceral cavity. The septal coste are continuous with the septa, and are of the same thickness, and connect the septa of one calice with those of another. Their margins are furnished with small tubercular denta- tions like those of the septa. The first and second cycles of septa are of equal length, and meet R. F. TOMES ON THE GREAT-OOLITE MADREPORARIA. 185 in the centre; the third is half the length of the first.and second ; and the fourth is two thirds the length of the third. Height of the corallum 33 inches, greatest diameter 32 inches, diameter of the calices 2 inch. This species differs from C. burgundie in its globular form and in having larger calices, which are much more prominent, and with their centres much deeper, and the septa of the first and second cycles of equal length. A second example of this species was found by me on the surface of a field east of the Duke of Marlborough’s iron-works at Fawler, near Charlbury, by the side of the road leading from Fawler to Stonesfield. It is of smaller size than the Fairford specimen, and the septa are thinner. Otherwise the two are very similar. Genus Isastrma, M.-Edw. and Haime. IsastR@A LimitatTa, Lamx. in Mich. This is an abundant species at Fairford, Stonesfield, and Rollright, and has been found also in equal abundance at Steeple Barton and Glympton, but seems to be less plentiful at Burford, where the Isastrea caplanulata occurs in great numbers. All the specimens in Mr. Slatter’s collection have an expanded form with a thin margin and a more or less gibbous upper surface, and are provided with a rugged base surrounded by a strong and wrinkled epitheca. but the calices do not differ from those of the dendroid or foliaceous varieties which are so common in the Great Oolite of Oxfordshire. IsasTRHA MICROPHYLLA, Tomes, Proc. Geol. Assoc. vol. vi. p. 158. Since the publication of the description of this well-marked species, I have examined, by the kindness of Mr. J. Windowes, of Chipping Norton, a specimen from the railway-cutting at Rollright. This isina good state of preservation, and confirms the characters already assigned to the species. - IsastrmA Bexstuy1, Tomes, Proc. Geol. Assoc. vol. vi. p. 159. Besides occurring in the Rollright coral-bed, from which the type specimens were obtained, I can now record the existence of this species in considerable numbers in Mr. Slatter’s collection, and add Fairford as a locality where it has been found. IsastRHA GIBBOsA, Duncan, Supp. Brit. Foss. Cor. pt. ii. p. 15, pl. ii. , fies. 10, T1. In the Stonesfield cutting this species occurs, though not abund- antly, and is there associated with the allied species Jsastrwa limitata. The short, thick and strongly ribbed septa, shown in Prof. Duncan’s figure, are characteristic of this species. In all the specimens I have seen, the outer ends of the septa alternate very regularly with those of contiguous calices, and are never continuous with them. : 186 R. F. TOMES ON THE GREAT-OOLITE MADREPORARIA. This species possesses characters between Isastrea limitata and Isasteea explanulata. It also occurs and is common in the Fairford coral-bed. IsasTRHA EXPLANULATA, M.-Edw. and Haime, Brit. Foss. Corr. p.115, tab. xxiv. fig. 3. Astrea explanulata, M‘Coy, Ann. Nat. Hist. ser. 2, vol. 11. p. 400 1848). As , the case with Jsastrea limitata, the present coral occurs both as a massive and a dendroid form. But the branching variety may be regarded as the ultimate, and therefore the most typical growth. The calicves on the branches are even more superficial than those of the massive variety, and often have no depression whatever, but present the appearance of hexgonal patches of crowded septa, raised just their own thickness from the surface of the corallum. Their outward ends pass in between the outer ends of those of the con- tiguous calices. This interlapping of the septa is what I have not seen in any other of the Astreeide ; and it led meat one time to place this species in another genus. But as it is little observable on the more massive parts of the corallum, but is mostly confined to the newer growth, of course it 1s not a valuable character. This coral occurs and is common at Fairford and at Burford; and I have received, by the kindness of M. Rigaux, a specimen from the Great Oolite near Boulogne. Genus Latimmanpra, d’Orb. It is probable that the Lattmeandrea of the Secondary formations need a much more searching investigation during the earlier periods of their growth than they have as yet been subjected to, and to have their relationship with Chorisastraea, Heterogyra, and Phyllo- gyva more clearly made out. LATIMHANDRA LOTHARINGA, E. de From. Cat. Polyp. de l’Yonne, 1856 ; Tomes, Proc. Geol. Assoc. vol. vi. p. 160. Meandrina lotharinga, Mich. Icon. Zooph. p. 100, pl. xxii. fig. 2. Since the appearance of my paper in the sixth volume of the ‘ Proceedings’ of the Geologists’ Association, I have met with this species in the Great Oolite near Burford, and have examined speci-: mens from Fairford in Mr. Slatter’s collection. From. the locality before mentioned by me (the railway-cutting near Rollright, Oxfordshire) I have also obtained, a further supply of specimens, and have taken a single example from the surface of a field contiguous to the Stonesfield railway-cutting. Genus CuorisastRma, Ex. de From. CHORISASTREA OBTUSA, d’Orb. sp. Amblophyllia obtusa, d’Orb. Prodr. Paléont. t. i. p. 285 (1850). Thecosmilia ? obtusa, M.-Edw. and Haime, Pol. Foss. terr. Pal. p. 78 (1851). R. F. TOMES ON THE GREAT-OOLITE MADREPORARIA, 187 Thecosmilia obtusa, Dunc. Supp. Brit. Foss. Cor. pt. iii. p. 14, pl. 1. figs. 1-4. The coral figured by Prof. Duncan in his Supplement to the . British Fossil Corals of MM. Milne-Edwards and Haime, under the name of Thecosmilia obtusa, was received by him from Mr. Brown of Cirencester, and, I have no doubt, was one of the Fairford corals. There are many such in the collection made by Miss Slatter at that place. These I have examined with great care, and am fully satisfied that they are not referable to the genus Thecosmilia. In its general form this species is tall and turbinate, and attached by a rather narrow base. With upward growth the calice becomes lobular, and increase takes place by gemmation in the ends of the lobes. A number of corallites is the result, which may either remain attached to each other or become free. They all spring from nearly one level; and the greatest number I have seen is four. More frequently there are two or three. Genus AstrocaniA, M.-Edw. and Haime. Only one undoubted Oolitic species of the genus is given by MM. Milne-Edwards and Haime in their General History of Corals*. It is the Astrocenia tubercsa of d’Orbigny, and was met with in the In- ferior Oolite of Luc. Four doubtful ones are also given, namely :—the Prionasirea microcoma of d’Orbigny, from the Middle Oolite of Neu- vizi; the Astrea sancti-inihieli of Michelin, from the Middle Oolite of Saint-Mihiel; the Astrea crasso-ramosa of the same author, from the same formation and place; and the Astrwa pentagonalis of Goldfuss, from the Middle Oolite of Wiirttemberg. The last-named species, however, has been doubtfully referred by Becker and Milas- chewitsch to the genus Stephanoceniay. M. de Fromentel has described and figured a well-marked species from the Portlandian beds of Mantoche under the name of Astrocenia triangularist. In his ‘ Introduction’ §, only that species is men- tioned asan unquestionable Oolitic Astrocenia, all the others (including Astroceenia tuberosa, which was evidently regarded by Milne-Hd- wards and Haime as a true Astroewnia) being included in the list of doubtful species. Nothing new respecting the genus Astrocenia appeared in the subsequently published work by the same author on the fossil corals of the environs of Gray. From this we may conclude that the genus has been, up to the present time, represented by very few, if by more than one, undoubted Oolitic. species; and the interest attending its present introduction into the English list is not lessened by its comparative rarity in the Jurassic formations of-other countries of Europe ||. * Hist. Nat. Corall. vol. ii. p. 259. + Palzontographica, vol. xxi. t Bull. de la Soc. Géol. de France, 2me ser. t. xiii. p. 859, 1856. § Introduction a l’Etude des Polyp. Foss. p. 233. || I take the present opportunity of observing that I regard some of the so- called Astrocenieg from the South-Wales Lias as clearly referable to another genus. Three Astroceni¢ from the Lias of France, however, have been described by M. de Fromentel and by MM. Terquem and Piette. 188 R. F. TOMES ON THE GREAT-OOLITE MADREPORARIA. ASTROCENIA PHILLIPSI, n.sp. Plate VII. figs. 6, 7. The corallum is not very large; it 1s incrusting, but becomes elevated and somewhat gibbous by subsequent growth. The calices are irregular in size and form, but are generally | hexagonal. They are moderately deep, and the walls dividing them are thick and prominent. There are from eighteen to thirty-one septa; of the latter number nine are principal; they are smaller about the middle and are joined to the columella; nine others are about two thirds of the length of the first; and the remainder are short. They appear to represent nine systems, and have three cycles, of which the third is incomplete. All the septa are thick and mount onto the top of the wall, where they have a prominence corresponding to the cycle to which they belong. When unworn, their margins are regularly tuberculated; but the tubercles are elongated across the septa, and hence have more the appearance of transverse ribs than tubercles. The columella is not very prominent, but is well defined. The corallum is from 6 to 18 lines in height; and the calices have a diameter of from 1 to 2 lines. I have met with four examples of this coral, three of which were found attached to oysters in the Stonesfield railway-cutting, and the other taken from the surface of the adjoining field. Compared with the Asirocenia decaphylla from the Cretaceous formation of Gosau, it has much the same subgibbous shape, but has larger calices, septa which are divided into a different number of systems, and the edges of which have transverse papille or tubercles. With the so-called Astrocenia of the Glamorganshire Lias it has little affinity. ZOANTHARIA PERFORATA. Family PORITIDAZ. Genus THamNnastRmA, Le Sauvage. THamNnasTRm@A Lyrerir, M.-Edw. and Haime, Brit. Foss. Cor. pt. 11. p- 118, tab. xxi. fig. 4. This species is very abundant at Stonesfield, from which place the specimens were obtained which were described and figured by MM. Milne-Edwards and Haime. But it is rarely met with except in fragments; and this renders it difficult to make an approximate esti- mate of the height to which it attains. From the great difference, however, which exists in the diameter of the fragments, as well as their nearly cylindrical form, it may be assumed that they tapered upwards very slowly, and that it was a tall species. It occurs also at Fairford, Rollright, and Epwell, and appears to be common at all those places. THAMNASTRHA MICROPHYLLA, 0. sp. The general form of the corallum is that of a tall bush, perhaps about a foot in height, sections of the branches of which show that R. F. TOMES ON THE GREAT-OOLITE MADREPORARIA. 189 they are either rounded or ovoid, while at the same time they are rather strongly nodulated. The calices are very small, and evenly distributed. They are very superficial, excepting on the thinner parts of the branches, where they have greater prominence. They are round and have a small pim- ple-shaped columella. The septa are about twenty-four in number, and have their margins delicately but distinctly papillated. The six principal ones maintain their thickness quite up to the columella; the next six pass nearly to the columella, and become considerably thinner as they pass inwards; the remainder are short and small; but all are of nearly the same thickness at their outer ends. The septal costa generally form an angle where they join those from contiguous calices; but there is very little disposition to the parallel arrangement observable in so many Thamnastree, and none of the costs anastomose. Diameter of the calices about half a line, distance from the centre of one calice to the centre of the adjoining one three quarters of a line. This species may be briefly described as a miniature Thamnas- trea Lyelli, so far as the conformation and size of the calices are concerned; but the corallum is probably quite as large as'in that species. In the sizeof the calices the present species somewhat resembles 7’. mammosa; but the septal coste in the latter species anastomose considerably, whereas in 7’. mzcrophylla they never do so. Moreover the papillze of the septa in 7. microphylla are very much smaller and more delicate than in 7. mammosa. Only one specimen has been observed ; and it was found completely filling a large mass of stone in the quarry near Burford, by Mr. T. J. Slatter, to whom I am indebted for a portion for my use in preparing this paper. THAMNASTRZA Watton1, M.-Edw. & Haime, Brit. Foss. Cor. p. 120, tab. xxv. fig. 4; not Duncan, Suppl. Brit. Foss. Cor. pt. ii. pl. i. figs. 6-9. The fragment from which MM. Milne-Edwards and Haime drew their description and figure of this species, although possessing dis- tinctive specific characters, was not sufficiently complete to afford a good account of the species. Many well-preserved specimens are in Mr. Slatter’s Fairford collection; and while certain parts of them accord satisfactorily with the above-mentioned description and figures, other parts are so different that they might easily be mistaken for another species. It may be described as a digitate rather than a dendroid form, some individuals having at first a more or less irre- gular discoid base, from which arises a central dome-shaped promi- nence terminating in two or three finger-lke processes. It was probably one of these finger-like processes that furnished material for the original description. The base is sometimes concave and covered with a thick concentrically wrinkled epitheca, completely hiding the basal wall and coste. The newer calices (that is, those on the upper parts of the corallum) resemble those of the specimen which was figured by the original describers; but the older ones, 190 R. F. TOMES ON THE GREAT-OOLITE MADREPORARIA. especially those near the base of the corallum, are quite different. They are far apart, small, round, very clearly defined, and rather prominent. The septa are short, thick, and of equal size at their outer and inner ends. ‘The fossula is small and well defined. The septal coste are long, very uniform in size, closely placed together, and have a parallel arrangement. They are not so stout as the septa ; and a distinct notch, which often divides them from the septa, gives to the latter the aspect of a ring of pali. Both the septa and their costal continuations are very delicately and finely papillated; they might almost be said to be granulated. . Distance from centre to centre of the calices 14 line, diameter of the calices 1 line. The figures given by Prof. Duncan as of this species*, but unac- companied by letterpress, must, in my opinion, be referred to some other species which has the septa more strongly geniculated. At present I have only seen specimens of this species from Fair- ford ; but it is very common there. THaMNASTRa#A MAMMOSA, M.-Edw. & Haime, Brit. Foss. Cor. p. 119. From the surface of a ploughed field on the Stonesfield side of the Great Western Railway two examples have been obtained, which have all the calicular characteristics of the species as given by MM. Milne-Edwards and Haime, but differ considerably from their figure in the general form of the corallum. Although they possess the same nodular and gibbous upper surface, they are much more expanded and depressed. In a bed immediately overlying the Stonesfield Slate at Aylworth, on the Bourton-on-the-Water Railway, specimens of this coral have been collected by me which more nearly approach the upright form represented in their figure. It appears at that place to be a com- mon species. A few small specimens have also been found at Fair- ford; but it is probably there, as well as at Stonesfield, rather rare. Genus Microsorena, Lamx. MicRosoLENA ExcELsA, M.-Edw. & Haime, Brit. Foss. Cor. p. 124, tab. xxv, fie.) 5,,.1851. Tt was from specimens in Mr. Walton’s collection that the figures and description of this species which appear in the great work on ‘British Fossil Corals’ by MM. Milne-Edwards and Haime were taken; and I have had the advantage of specimens from the same collection for comparison when identifying the species mentioned in the present communication. By their assistance I have determined with certainty examples from Fairford, Stonesfield, and Roliright, at all which places it is a common species. In the early periods of its growth, WM. excelsa is very peculiar. It then presents the appearance of a cone attached by a broad base, the summit of which is obtuse, and has a single large calice, while around the sides there is a circle of smaller calices, just as in Genabacia. * Supp. Brit. Foss. Cor. pt. ii. pl. ii. figs. 6-9. R. F. TOMES ON THE GREAT-OOLITE MADREPORARIA. 191 All such examples, so far as I have observed, are attached to speci- mens of Thamnastrea Lyell. MicrosoLENA REGULARIS, M.-Edw. & Haime, Brit. Foss. Cor. p. 122, tab. xxv. fig. 6. Alveopora microsolena, M‘Coy, Ann. & Mag. Nat. Hist., ser. 2, vol. 11. p. 419. At present I have only met with this coral, as a Great-Oolite species, at Fairford and Aylworth; and it appears to be much less abundant at those places than in the Lower Trigonia-grit of the Inferior Oolite near Cheltenham. _ Genus TricycLoseris, Tomes. This genns was proposed by me for a coral obtained by myself from the Margaritatus-zone of the Middle Lias at Charmouth, and described in the‘ Quarterly Journal ofthe Geological Society’ in 1878*. It was there described as a compound Cyclolite, having an elongated and lobular form. Two specimens of a coral, having obviously the same generic peculiarities, were procured by Miss Slatter from Fairford; and these, while they confirm the genus, render some modification in its definition necessary. I now define it as follows :-— The corallum is oblong; and there is a flat basal plate furnished with an epitheca; the upper surface is in the form of a rounded ridge; and the calices are small and near together, and in a line along the ridge. The central one is the largest; and the others have been produced by gemmation from it on either side, somewhat as in Dimorpharcea. From this calicular ridge the long septal costz slope off all round; and the outer boundary is in the form of a thick rounded edge. The septa and endotheca are perforate and like those of Microsolena. TRICYCLOSERIS LiMAx, n. sp. Plate VIT. figs. 18, 19. The corallum is irregularly oblong; and there is a tendency to push out into lobes or corners at the ends. The under surface is concave, and consists of an imperforate basal wall, which does not extend all over the corallum and is furnished with an epitheca. The upper surface is shaped like a rounded ridge which slopes off on all sides to the thick and rounded outer margin of the corallum. The calices are four in number in both specimens. One, which is larger than the other, holds a nearly central position, and is round and well defined and has a small but deep fossula. The others, which are near to it and are in the same line with it, are small and irregular. About thirty septa enter into and compose the middle calice. They, as well as the septal coste, are very distinctly moniliform ; and they closely resemble the same parts in specimens of Genabacia stellifera, from the Cornbrash of Wast, near Boulogne, but are not quite so closely placed together. These septal costz are *. Vol. xxxiv. p. 190, pl. ix. fig. 1. 192 R. F. TOMES ON THE GREAT-OOLITE MADREPORARIA. long, thin, sometimes straight, but on some parts of the corallum curved or flexuous; and they often anastomose, especially near the outer margin of the corallum, over the rounded edge of which they pass. The synapticule, viewed outwardly, are cuneiform; that is to say they consist of thin horizontal perforate lamine projecting from the sides of the septa. ‘They do not spring from the same level on each septum, but, meeting those from the nextseptum, become oblique in their position across the loculus. Length of the corallum 1 inch 6 lines, breadth of the corallum 11 lines, length of the line of calices 9 lines, height of the corallum 6 lines. It might at first sight appear that this species is nothing more than a half developed form, perhaps of a Microsolena; but this is rendered very improbable by the nearness of all the calices to each other, and by the great length of the septal costae. Moreover the young forms of the genera Thamnastrea and Microsolena, which are common and well known, in no way resemble the present coral. Genus Comosmris, d’Orb. CoMOSERIS VERMICULARIS, M.-Edw. and Haime, Brit. Foss. Cor. p. 122, tab. xxiv. fig. 1. Meandrina vermicularis, M‘Coy, Ann. & Mag. Nat. Hist. ser. 2, vol. 11. p. 402 (1848). Only one instance of the occurrence of this coral in the Great Oolite has come to my knowledge. A specimen was obtained with the other species mentioned in this paper at Fairford, and is in Mr. Slatter’s collection. It appears to be a rare species. Only two examples were seen by MM. Milne-Edwards and Haime. I have already made known its occurrence in the Inferior Oolite at Crickley*; and the mention of the Fairford specimen adds another locality in the Great Oolite. Genus Orosrris, M.-Edw. and Haime. OrosERiIs Siatrert, n. sp. Plate VII. fig. 5. One specimen broken in half, but otherwise well preserved, of a species of Oroseris is all that represents the genus in the Fairford collection. The corallum, when perfect, had a massive and some- what lobed outline, with a greatly elevated helmet-shaped middle part, and an undersurface which had a corresponding and deep concavity. A strongly wrinkled epitheca covers the whole of the undersurface, which has lines and furrows concentrically arranged. The furrows of the calicular surface have a somewhat radiate direc- tion from the highest part downwards and outwards, sometimes running into one another; but they curve as they approach the lower and outer edge, where they follow somewhat the line of the boundary of the corallum. They are deep, well defined, and are about equal distances apart; and there is no part of the corallum on which the calices are scattered, as in Thamnastrea. * Quart. Journ. Geol. Soc. vol. xxxviii. p. 448. R. F. TOMES ON THE GREAT-OOLITE MADREPORARIA. 193 The calices are as far apart in the rows as the rows are distant from each other. They are round; and twelve septa enter into their composition. These are equal in length, and approach near to the centre of the calice; but there is a small round and well-marked fossula. They, as well as the septal costz, are strongly moniliform ; and the beaded prominences on their edges are very distinct. The septal coste are distinct, of equal size throughout, straight and cristiform where they pass over the prominent but rounded ridges, and none of them anastomose. The diameter of the corallum is 4 inches, the height of the same 3 inches, breadth of the ridges between the furrows 2 lines, distance of the calices apart in the furrows 2 lines. ANABACIA COMPLANATA, M.-Edw. and Haime, Hist. Nat. Corall. t. 111. p- 31 (1860). Fungia complanata, Defr. Dict. Sc. Nat. t. xvii. p. 217 (1820). Anabacia orbulites, M.-Kdw. and Haime, Brit. Foss. Cor. p. 120, tab. xxix. fig. 2 (1851). It is extremely doubtful whether this genus embraces more than one species. Every intermediate form may be met with, between the ordinary lenticular specimens of A. complanata and the globular one designated by MM. Milne-Edwards and Haime A. hemispherica so that it becomes impossible to separate the two. Again, some of the examples from the Trigonia-grit of the Inferior Oolite have a very symmetrical form, and agree so exactly in this respect, and in the delicacy of their septa, with the A. Bouchard: from the French Oolite that they cannot be distinguished. Leaving them for the present as merely varieties of one species, I may mention a well-marked variety from the Stonesfield railway- cutting. It is of large size, having a diameter as well as height of more than three quarters of an inch. In form, as well as in size, it corresponds very closely with Genabacia stcllulata, being pyramidal rather than globular superiorly, and having a deep circular depression beneath, which only occupies the centre of the base. The septa and septal costee are relatively thick, and they anastomose much less than do those of the ordinary examples. This species occurs so abundantly in so many localities in the Great Oolite as to render any mention of them needless; but it is worthy of remark that up to this time only a single example has been taken from the very rich coralline deposit at Fairford. ZOANTHARIA TABULATA. Family THECOSTEGITIDZ, de From. Genus Crarnopnora, Michelin. After great pains and the examination of a great many specimens of one species of Cyathophora, obviously referable to the same genus as the one figured by Michelin, I still adhere to the recognition 194 R. F. TOMES ON THE GREAT-OOLITE MADREPORARIA. of the two genera Cyathophora and Cryptocenia as proposed by d’Orbigny and afterwards adopted by M. de Fromentel. The defini- tion, however, of the two genera requires, in my opinion, some modifi- cation, which I give as follows :— Cyathophora. Coenenchyma small in quantity and dense. Gemma- tion proceeding from it in close proximity to the walls of the corallites, if not actually from the walls themselves. Septa feebly developed, and the cycles not traceable. Calices generally much crowded, appearing at many heights, often oblique, oval, or even polygonal. Cryptocenia. Ccenenchyma abundant, and of a loose nature, composed of a great many dissepimental tabule, from which gemma- tion takes place quite distinct from the walls of the corallites. Septa well developed, and their cycles distinct. The calices not crowded, always round, and oz the same level. CyatHorHora Boureveri, Defr. sp. Plate VII. figs. 3, 4. Astrea Bourgueti, Defr. Dict. des Sci. Nat. t. xlii. p. 380 (1826). Cyathophora Richardi, Mich. Icon. p. 104, pl. 26. fig. 1 (1848). Cyathophora solida, Phill. Geol. Oxf. and Thames Valley, pl. ii. fic. i, This coral, which is identical with the one I introduced into the British list in the sixth volume of the Proceedings of the Geologists’ Association, from the examination of a single specimen supposed to have come from Garsington, Oxfordshire, occurs in great abundance in the railway-cutting near Stonesfield, but is there confined to the lower parts of the coralliferous layer, and is not, so far as I have been able to observe, associated with any other coral. I now entertain but little doubt, from the appearance of the supposed Garsington specimen, that it really came from Stonesfield. It is no doubt also identical with the species mentioned by Professor Phillips, in his work on the Geology of Oxford and the Thames Valley, under the name Cyathophora sokda. In young examples the corallites are directed so many ways as to suggest that they grow out of each other; but in the larger individuals they are often packed closely side by side, and the calices squeezed into an elongated or polygonal form. In the latter case there is hardly any coenenchyma; but before they have attained to a considerable growth the coenenchyma surrounds the corallites, and encases them with a layer which has somewhat the appearance of epitheca, and is of a sienna-brown colour, with a glossy fracture. Besides Stonesfield, I have also met with this species in a bed quite at the top of the stone beds of the Hook-Norton railway- cutting. The layers there exposed are Inferior Oolite ; but the one in which these corals occur, and where they are unassociated with any other species, is called the “rifted bed,” and has been doubtfully regarded as Great Oolite, overlying all the others. Most likely this coralliferous layer corresponds with the lower part of the coral-bed of Stonesfield. Becker and Milaschewitsch quote Cyathophora Bourguett as occuring with another species, OC. marginestellata, in the corallian of RB. F. TOMES ON THE GREAT-OOLITE MADREPORARIA. 195 Nattheim ; but the figure given of the first of these shows a greater development of the septa than is seen either in Michelin’s figure or in any of the numerous specimens which have come under my ex- amination. It is with very great doubt that I give this species a place in the Zoantharia Tabulata, not having by any means satisfied myself as to its real affinities. If I am correct in referring the present species to the one on which Michelin established the genus Cyathophora, the genus Cryptocoenia will have to be retained for such species as C. Pratti, and C. luciensis. But, on the other hand, if it should eventually prove to differ from Cyathophora generically, then the genus Cryptocema must be dropped, and a new genus formed . to receive the present species: for this, Depaphyllum would not be an inappropriate name. EXPLANATION OF PLATE VII. . Bathycenia Slatteri, the corallum, natural size. . Cryptocenia microphylla, two calices, magnified. . Cyathophora Bourgueti, a young corallum, magnified, showing the position of the corallites m relation to each other. , a vertical section of a young corallum, showing the small quantity of dense coenenchyma and the tabule inside the corallites. The peculiarity in the septa is also shown in the figure. . Oroseris Slattert, a corallum, the natural size. . Astrocenia Phillipsi, the corallum, natural size. , some calices, magnified. . Bathycenia Slatteri, some calices, magnified, showing the smooth cristi- form septa and the mode of gemmation. 9. Bathyceniasolida, some calices, magnified, showing the smooth cristiform septa. 10. , some calices, seen from above, magnified. 11. Montlivaitia caryophyllata, a septum, showing the lateral ornamenta- tion. 12, 13. Enallohelia clavata, natural size. : , a calice, magnified. 15. Confusastrea magnifica, the corallum, natural size. 16. Pavia pedunculata, the corallum, natural size, 17. —— , two calices, which have been recently formed by the fissi- parous division of a larger one, magnified. 18. Tricycloseris imax, the corallum, natural size, seen from above. , the corallum, natural size, seen from below. 20. Montlivaltia Slattert, a septum, showing the lateral ornamentation. 21. Montlivaltia fairfordensis, a septum, showing the lateral ornamenta- tion. 22. Confusastrea magnifica, a septum, showing the lateral ornamentation. ae" » whe QO ATS ON Discussion. Tre Coarrman (Dr. Gwyn Jeffreys) expressed his sense of the value of the paper. He observed that most of these corals were com- pound, and some of them especially peculiar to reefs, although compound Madreporaria were found living as deep as 750 fathoms. They, therefore, did not seem to very much elucidate the question of the depth of the Mesozoic sea. Simple or solitary corals cer- Q.J.G.8. No. 154. P 196 R. F. TOMES ON THE GREAT-OOLITE MADREPORABIA. tainly did not throw more light upon the question, because they oc- curred from shallow water to very great depths, even to 3000 fathoms. ' Prof. Prestwicu said that Mr. Brown’s collection, mentioned by the author, came not from two horizons, but all from one, at a spot about 2 miles W. of Cirencester, in a zone about 6-18 inches thick, near the top of the Great Oolite. Prof. P. M. Duncan confirmed the statement of Prof. Prestwich about the horizon from which Mr. Brown’s collection was made. These corals, described by Mr. Tomes, were from lenticular coral- beds, not from reefs. They could hardly be very deep-sea formations, from the oolite contained in them, which seemed at the present time to be a shore-formation. It was a mistake to suppose that live reef-building corals ever occurred below about 25 fathoms. It was to be regretted that a good writer such as the author did not come more frequently among his fellow workers; for he would then have learnt that many of the statements made by him about calicular gemmation and fissiparity were already in print, and had been so from the days of Milne-Edwards. Fissiparity and gemmation were quite distinct things. Some corals keep the figure of 8 described by the author; some depart from it during subsequent growth. Unfortunately M. de Fromentel, referred to by the author, was not a student of recent corals. Thecosmilian forms had been found exhi- biting fissiparity ; these had been actually renamed by Mr. Tomes, though the speaker had already assigned them to an existing genus. He felt doubts as to the validity of some of the genera proposed by Mr. Tomes. The coral could not be named Confusastrea without a section; it presented some characters allied to FMavia. He called attention to the so-called Cyathophore, which had lost their-septa and all their internal characters. Sections, he would observe, were absolutely necessary for the study of fossil corals. rie yy am Bey T i . Quart Journ.Geol Soc Vol.XXXIX PLVI : CPLVI imp Hanhart CORALS OOLITE GREAT CBerjeau ith ’ " r ‘ > 4 La 4 . f Chae) i ) fi . 4 ree 4 ‘ f nl A rr 5 « i ee » \ ' " . ’ ’ | ocky } y i 4 ‘ ' } : J. S. GARDNER ON THE LOWER LONDON TERTIARIES. 197 13. On the Lowzr Kocrene Section between Recutvers and HERne Bay, and on some Modifications in the Classification of the LowER Lonpon Trrtrarizs. By J. 8. Garpner, Esq., F.G.S. (Read January 10, 1833.) Tur Lower London Tertiaries were defined by Prestwich as con- sisting of three divisions. ‘Two of these were relatively homogeneous in composition ; but the third was made up of very varied materials. The extreme care and accuracy with which these divisions were traced out over the whole of the Eocene area in England, and sub- sequently correlated with those of the French area, led to their speedy and universal recognition. No modifications in this classification were even suggested until 1866 *, when Mr. Whitaker, while unreservedly adopting Prestwich’s divisions of ‘‘ Thanet Beds ” and ‘‘ Woolwich and Reading Beds,’ making, indeed, copious use ofhis observations, separated portions of his “ basement-bed” of the London clay, where this was assigned any considerable thickness, together with a small portion of his Woolwich and Reading beds, as ‘‘ Oldhaven Beds,” and thus almost restricted the “‘ basement-bed” to the inconsiderable thickness of coarser material which nearly everywhere forms the base of the London Clay. Almost the only criticism that can be urged against Prestwich’s classification is that he places the ‘‘ basement-bed ” of the London Clay in a different group of the Hocenes from the London Clay itself, his nomenclature implying a closer relationship than he admits’. But the Survey, on the other hand, have unfortunately adopted a name (“‘ Oldhaven”’) which not only does not exist on the maps, but is scarcely known at the locality, an inquiry for “ Oldhaven Gap,” where the beds are typically developed, being useless even at the gap itself, which is known and mapped as “ Bishopstone Gap.” Stratigraphically, however, and perhaps as measures of time, these divisions leave little to be-desired; and it is only from certain other points of view that this proposal to modify them may be justified. Before entering into the details of the subject, it is advisable at least to endeavour to realize as far as possible the conditions under which the Lower London Tertiaries were produced. Although this must always to a great extent be a matter of theory and mere.con- jecture, a large amount of inference may be safely drawn from our actual present knowledge. Of all facts patent to those who study the Eocenes in our own or in adjacent areas, the near proximity of land and fresh water through- out the whole series is the most obvious. Next, speaking of marine formations only, itis observed that the faunas which succeed each other stratigraphically are by no means the most closely related to each other palzontologically, but that beds separated by great: vertical thicknesses contain faunas far more nearly allied than do those in * Quart. Journ. Geol. Soe. vol. xxii. p. 412. + Whitaker, /. c. p. 413. P2 198 J. S. GARDNER ON THE LOWER LONDON TERTIARIES. juxtaposition. It has come to be recognized that these interlacing faunas can be referred to two distinct types—the one perhaps best known as the fauna of the “‘ Calcaire grossier,” and the other most readily definable as that of the London Clay. Further, it is admitted by nearly every writer on the Kocene that the latter fauna is of a more temperate or northern type than the former ; and the conclu- sion has been repeatedly drawn that it must have belonged to a more northerly sea, completely shut off from direct communication with that sea in which the “‘ Calcaire-grossier”’ fauna lived. The “Calcaire- grossier”’ fauna is, in England, peculiarly and wholly distinctive of the Bracklesham period, and also seems to appear first in France at about the same time; it was therefore for a long time held to be cha- racteristic of the Middle Kocene and of nothing else. The fallacy of such views was signally demonstrated by the discovery of the “‘Montian system ” in Belgium*, containing at the very base of the Eocene a fauna allied to it in the closest possible manner. Since the abandonment of the belief in special creations it is perfectly obvious that slight but persistent modifications in groups of genera must be the ultimate test of the relative age of strata over the whole earth, the incoming of new types being only of value when limited areas are compared. Correlation by zones over wide areas, although extremely valuable, is misleading if perfect contemporaneity for the whole zone is thought to be implied, as zones can only represent the migrations of species following the successive spread or shifting of favouring conditions. Much of the difficulty in correlating the formations in England and America, for instance, has arisen from a disregard of these considerations. Starting with the Calcaire de Mons at the base of the Eocene, we | find the southern sea with its distinctive fauna occupying parts of - Belgium, while the next fact of which we have cognizance is an extension of the northern sea, with its distinctive fauna, occupying the same ground and spreading south into France. It deposited the ‘‘ Heersian,” “* Landenian,” “‘ Sables de Bracheux,” &c., whose faunas are all intimately related to those of the Thanet Sands. There is no trace of the ‘‘ Montian”’ in England; but the rest of the Eocenes are more or less represented here, and it is with them in England only that I am able to deal. The deposits of the northern sea occupied first a limited portion of the east of the London basin only, and then, during the London Clay, extended to the utmost confines of the Hampshire basin. It is important to notice, as corroborative of the complete separation of the two seas, that though a great increase in temperature took place at some time between the Thanet Sands andthe London Clay, none of the warmer “ Calcaire-grossier ” species found their way into the London-Clay sea, but an immigration instead took place of quite other species of Cowries, Volutes, Vautili, and heat- loving genera. Next, omitting freshwater formations, the southern * The appearance of this fauna, “anticipating its normal epoch of appari- tion,” is called by M. Barrandea ‘‘colony,” by M. Marcou a “centre d’' apparition détres précurseurs.” “Tt constitutes an exception to the laws of paleontology ” (Géol. Belgique: Mourlon, 1880, vol. i. p. 193).- J. S. GARDNER ON THE LOWER LONDON TERTIARIES. 199 sea, recognizable in the Brackleshams, occupied the Hampshire basin and covered over the London Clay to within 20 or 30 miles of London. That these formations were deposited from opposite directions is apparent from their relative easterly and westerly developments. ‘lhe thinning of the London Clay in Hampshire is not due to denudation by the Bracklesham sea, as it was everywhere protected by thick intervening freshwater beds, and its original thickness is preserved entire. The Barton beds show for the first time an admixture of the two faunas, though only the less tropical of the Bracklesham species remain. The Brockenhurst fauna shows an increased preponderance of southern forms, while the Hempstead fauna, if it can be said to show any thing, recalls the northern types. Now marine faunas could only be kept distinct in adjacent areas, whose conditions of depth and sea-bottom were so similar, either by very sharply detined cold and warm currents, or by intervening land. These were clearly not separated by currents, since each formation is limited by a shore-line, and freshwater strata intervene in every case, showing the area to have become land after the deposition of each ; and there is no supposition open that will explain the facts, except the continued existence of an isthmus, connecting England with the mainland, throughout the Eocene, until at least the Barton period. This isthmus was not stationary, however, but undulated from north to south and south to north without being broken through. Thus during the Thanet-Sand time it must have stretched from Dieppe along the Weald; but in the London-Clay period it could only have joined France to the south of the Isle of Wight and Purbeck; whilst in Bracklesham times it probably stretched from Belgium across the Weald to Hertfordshire. These various positions of this isthmus are not purely conjectural, except where is now sea; for its northern shore is distinctly traceable in the London Clay, and its southern shore in the Bracklesham beds. Further, the vast Hocene river, whose presence is felt in every deposit, had its estuary in the direction of the Thames valley throughout all the Lower EKocenes, but had its course diverted to the south by the change in the position of the isthmus which caused the London-Clay sea to recede; and its estuary remained in Hampshire as long as any further record of it is preserved. The detritus of the Kocene river is mainly quartzose and felspathic, and is such as to show that it probably drained a paleozoic area, while its bulk and the enormous variety of the forest vegetation imbedded in its silts leave no alter- native but the belief, supported in many other ways which cannot be entered into here, that it drained a vast continent with an in- definite westerly extension, even connected in some mysterious way with America. The breadth of this river in its purely freshwater reaches is actually seen, in Hampshire and Dorsetshire, to have been at least 17 or 18 miles* ; and the extent of homogeneous or similar fluviatile and estuarine deposits, stretching as they do from England over France on this area and over no other in Hocene time, shows * Dollfus compares its bulk in France to that of the Amazons. 200 J. 8. GARDNER ON THE LOWER LONDON TERTIARIES, its bulk to have been hardly inferior to that of the largest existing rivers. If we admit the preponderating action of this stupendous river in the formation of all Eocenes in England and France, their strati- graphy becomes comparatively simple ; but if we try to explain them by any other means, we are forced to suppose conditions which have no parallel at the present day. With the probable physical features of our area during the Eocene period before us, we are better able to appreciate the relative values of the divisions of the Lower London Tertiaries. The Lower London Tertiaries are divided by Prestwich and the Survey into the purely marine Thanet Beds, the fluviatile, estuarine, and marine Woolwich-and-Reading Beds, and the marine Oldhaven Beds. The object of the present communication is more especially to question whether more than one Eocene sea encroached upon our area prior to the sea which deposited the London Clay. The Thanet Beds. The limits of the area over which the Thanet Beds were deposited were pretty accurately traced by Prof. Prestwich; and it is quite unlikely that they ever had any very considerable extension beyond their present limits as mapped by the Survey. They are almost wholly unfossiliferous west of Rainham* in Kent, and, indeed, present no features of interest outside the Canterbury and, Isle-of-Thanet districts. The following diagram from the Survey Memoir of 1876, p. 56, explains their distribution. Fig. 1.—Diagram illustrating the Distribution of the Thanet Beds. Ww. E. Epsom, Rochester. Sittingbourne. Canterbury, f \ 1 { i 1 t : H { I t eé. Fine sand with occasional layers of sandstone, with fossils, sometimes silici- fied, thins westward, from 40 feet in thickness, and passes down into d. Bluish-grey sandy marl, with green grains and fossils, thins westward, and is almost confined to Hast Kent. ce. Fine sand, without fossils (“at least but a very few remains have been found”); it is 60 feet thick in West Kent, where it forms nearly the whole ot me Thanet Beds, thinning out westward in Surrey and eastward in Hast ent. 6, Alternations of brown clay and loam, without fossils, thin and local (in part of East Kent). a, The “‘ base-bed,” clayey greensand with unworn green-coated flints resting on the Chalk, thin (rarely over 5 feet), but constant. The division ¢ is typically exposed in railway-cuttings, especially about Swanley and St. Mary Cray on the London, Chatham, and * Otterham Quay, E. of Rainham, is the spot furthest west, except Upnor, from which fossils are recorded by the Survey and by Prestwich. J. S. GARDNER ON THE LOWER LONDON TERTIARIES, Dover Railway, and appears to be light- coloured sea-sand, with considerable cohe- sion, about 60 feet thick and destitute of fossils. The few localities in Kast Kent described by Prestwich and the Survey in which fossils have been found do not seem to differ in any important respect from the section at Herne Bay and Pegwell Bay, and need not, therefore, be further alluded to here. The fossiliferous divisions of the formation can in fact be perfectly studied along the coast, which presents by far the most perfect and accessible, as well as the most typical sections (fig. 2). The Herne-Bay section terminates at the Reculvers, and, as restricted by Prestwich and the Survey, only exposes some 20 feet of Thanet Beds—Pegwell Bay, however, for- tunately supplying a continuous section through the remaining thickness to the Chalk (fig. 5). The position of the lowest Reculver bed can be accurately determined paleontologically in the Pegwell section ; but to make it quite certain, the measure of its exact height above the Chalk, which is about 70 feet, is given in a well sunk at Reculver. The Thanet Beds rest conformably, as it is called*, on the Chalk at Pegwell Bay, with at first a dip of about 5° S. The base rests on a great tabular layer of flint, and is full of the usual unworn green- coated flints. Itis now generally recognized that these flints have been dissolved out of the Chalk by solvent action, like that which produced the piping in chalk subsequent to its upheaval 7 ; and the present upper surface of the Chalk is therefore not actually the same as that on which the first Eocene mud * Even the highest Chalk must have been greatly denuded, as it could hardly have risen to the surface before it was compressed into a solid rock without being covered with something capable of resisting wave-action. There is at least no instance of any such deposit having been preserved as a surface deposit in any Atlantic isles that have been up- heaved. The Thanet Sands are as little conform- able, and had as little to do with the Chalk as the Goodwin Sands. t Whitaker, Quart. Journ. Geol. Soc. /.¢. pp. 406, 407. 10 ‘2 ®[Q WopuoT ‘vp ‘spogq uoavyp “k ‘SUIPBOA MOU ‘sorTag JouLTT, { __ ‘Spog joueqy, *p SPog YTATOoM *0 -099 "9 TOT} *F WOT} -009 ‘dep WoAeypTO ‘g UoKoog 201 ‘suarnoary wna 02 hog ausazy fo "gq wot wonjv0g9—Z "SIT (‘oTIu T MoT49es Jo Y48uNTT) 202 J. 8S. GARDNER ON THE LOWER LONDON TERTIARIES. was deposited. The waves probably acted then on a chalk coast precisely as they do now. As they undermine a cliff, the face tumbles away ; and the water abrades and dissolves the chalk, leaving the flint to be rolled and drifted into bays, or onto shingle banks outtosea. The flint in any case ceaselessly travels, and is eventually ground into mud or preserved as pebbles. Below the line of shingle beach we generally have between tides a great stretch of waterworn uneven-surfaced chalk covered and pro- tected by fucoids and Laminarie ; and this extends as far out below low-water as the depth permits us to see. No chalk-ooze seems ever to be reconstructed; and the carbonate of lime must therefore be carried away in solution. A great deal of the chalk bottom would apparently remain bare, as it is over large areas of the Channel, ifno other sediment were brought toit ; but when sediment is brought, where the sea is shallow enough to permit the growth of sea-weed, it must be originally deposited on such a seaweed-covered surface. The Thanet-beds fauna shows that they must have been deposited within the depth to which the Laminarian zone extends; and this imbedded sea-weed may well account for the peculiar character of the “ blackish green mud-hke sediment in which the green-coated flints are imbedded.” Prestwich* states that in burning it gives off ammonia in abundance, an evidence, he considers, of the presence in it of animal matter. This bottom-bed might of course belong to a very much older period than the rest of the Thanet Beds. There is not the slightest evidence, however, that any elevated chalk coast-line ever came into contact with the waves that deposited the Thanet Bedsin England. There are no beaches throughout the Eocene with a proportion of angular flints such as we invariably see when flint beaches are immediately derived from the chalk; nor are there any indications of old chalk cliff-lmes. On the other hand, the relatively small size and the completely rounded character of all the pebbles show that they must have travelled a long way from their source, and that they may have formed part of. still older deposits. The fluviatile deposits show that no chalk was being cut through by the river, and the marine deposits that no chal was being “planed down on this area by the sea. ‘The chalk cliffs of the evens coast-line must therefore at all our British Eocene periods have been far distant. Above this “ base-bed” we have about 19 feet 6 in. of yellowish drab sandy clay, which, though of nearly equal consistence, weathers out in bands, and on this some 16 feet of bluish slightly mottled sandy clay with decaying pyrites. The beds here become fossiliferous, and continue to be davkien sandy clay for another 34 feet, when we reach the first distinct line of drifted shells. The shell-beds range through a thickness of about 6 feet, and enclose a band of somewhat calcareous sandy concretions; 4 feet higher there is a band of scattered black pebbles; and the section clones with some 12 feet ofsand. So far the section lies in Pegwell Bay. The whole of the Reculver Thanet Beds, as limited by Prest- * Prestwich, Quart. Journ. Geol. Soe. vol. viii. p. 246. —, J. S. GARDNER ON THE LOWER LONDON TERTIARIES. 203 wich, are comprised in this section; but, owing to extensive tracts being sometimes laid bare at low water at Reculvers, the upper beds can usually be more profitably studied there. The line of concre- tions is a marked feature in the cliffs as they rise at Reculvers. Below them about 12 feet of compact yellowish slightly argillaceous sand, darkening when exposed to the sea and highly fossiliferous, forms the base of the cliff. The cliffs at first trend W. by N., and the strata are horizontal, but gradually dip about 2? W. Beyond the point to Bishopstone Gap the cliffs are in a more south-westerly direction, and the dip is maintained at from 2° to 3°; but up the chines the strata show horizontal where the face rakes 8.W. by S. From the bottom of the bay the cliffs trend slightly W. by N. and the dip is again 1° to 2° W. These sands are quartzose, somewhat micaceous and argillaceous, with green grains ; ferruginous concretions are common in them, in one of which Mr. Prestwich found the cast of a most interesting pine-cone. Pieces of unrolled silicified wood are not uncommon; and one of these proved to be the magnificently preserved stem of Osmunda Dowkeri so accurately described and figured by Mr. Car- ruthers. The Mollusca are abundant, but of few species, almost wholly bivalves. By far the most abundant form is Thracia oblata, imbedded flat,with the two valves closed tightly. The next most abundant forms are Pholadomya Koninckvi, which always occurs in an upright position, and Cyprina Morrisiz. (Among the rarer forms from the Thanet Beds are several species of Nucula, some of which occur only at Pegwell Bay.) The only univalves that my brother, ‘Mr. E. T. Gardner, F.L.S., who has diligently searched, and myself have met with, are a Natica and an Aporrhais with a scalaria-like spire, and a Murex, though some others are recorded by the Survey. These were evidently deposited in water at a depth beyond the reach of the waves, as there are few broken or even separate valves of shells, and the wood is unrolled, while this and the fir-cone and fern-stem show that they were not wholly beyond the influence of estuary water. ‘They contain hardly any vertebrate or crustacean remains, so far as is yet known. ‘Their deposition was probably not very dissimilar to that of some of the sediments forming on the chalk on the same coast at the present day. At the point the concretions are 6 feet above the beach, and they and the sand below them are crowded with Cyprina ; 200 yards east the concretions are overlain by some 12 feet of looser pale grey mottled and piped sand, on which rests a bed with silicified Corbula regulbvensis, a bed traceable from this point to Bishopstone Gap, on the west side of which it appears on the beach below high-water mark. It is valuable as the only land-mark by which the division between the so-called Thanet and Woolwich-and-Reading Beds can be readily picked up in the cliffs. 70 yards west asmall chine occurs (fig. 6); and the concretions disappear at the base of the cliffs, but form a well-marked line of rocks cutting diagonally across the flats at low water, in a westerly direction. At another 250 yards west the beds at the base of the cliff are mottled with bright orange ; 204 J. S. GARDNER ON THE LOWER LONDON TERTIARIES, 290 yards further west the Corbula-bed is in grey sand near the baseof the cliff, and Dentalia occur ; and 150 yards west of this, at the Gap, it ison a level with the shingle: 22 yards from the west corner of the Gap, the Corbula-bed appears on the beach below high-water mark and 10 feet below the base of the cliff, and soon after dis- appears out to sea. A very rich zone of fossils occurs at this spot, just below the Corbula-bed and between it and the Gap. The great Oyprina, 53 inches in diameter, occurs here singly or in pairs of valves, and nowhere else. Astarte tenera very common and per- fect, Cucullea decussata, Pectunculus terebratularis, Sanguinolaria Edwards, and Dentalium are other shells whose range in the Thanet Beds seems almost limited to this zone, and which can only be col- lected at this one spot in Herne Bay. It will be observed that the fauna is much richer here than lower down; the unrolled silicified wood is replaced by rolled and bored lignitized wood ; and a few very small quartz pebbles occur. None of the shells are in a living po- sition, though the valves are very generally closed; a few valves are broken, and they seem to have been more under the action of moving water, and therefore deposited at a less depth than towards the base of the Thanet Beds. Seventy feet of muddy sand had in fact accu- mulated in what was originally shallow water; and unless this was wholly counterbalanced by subsidence, the bottom would naturally be raised nearer the surface. Prof. Prestwich includes this zone with its rich group of fossils in the Woolwich-and-Reading Beds. He, as well as Mr. Whitaker, ac- knowledge the extreme difficulty of separating them from the under- lying beds, both here and at Richborough, where the Corbula-zone also occurs. The former says:—‘‘ This want of clear divisional sur- faces, and the occurrence of several of the same species of shells in the two series, might be considered an objection to their being thus separated. The fossils, however, taken as a group, are different from those of the Thanet Sands, whilst the sands are more siliceous and contain a larger proportion of green sand and some disseminated flint-pebbles—two mineral characters deriving some importance from their breadth and constancy” *. These distinctions are in themselves very slight as bases of separation between two marine formations ; but even such as they are, they have since been materially lessened. Taking the marine bivalve mollusca, the only group at all perfectly known from these formations, we find, according to Prestwich, twelve species occurring in the Thanet Sand and not in the Woolwich-and-Reading Beds. On the other hand, fourteen species from the latter do not occur in the Thanet Beds. In fact six species only are cited as common to the two formations. The Survey list, however‘, gives very different results; and as all but two of the species have been also found at Pegwell Bay, and only admittedly Thanet Beds are present there, no errors as to the * Prestwich, Quart. Journ. Geol. Soc. vol. x. p. 112 (1854); see also Whitaker, Quart. Journ. Geol. Soe. vol. xxii. pp. 409, 410, 419. t Mem. Geol. Surv. vol. iv. 1876. | ) lerne | [To face p. 204. Bay and Pegwell Bay. Fig. 5.—Peqwell Bay. s Drift-loam. oS ae? 2 ao) 2g Buff clayey sand. aa < a3 4a ===| Black pebbles (not mentioned by >) 3) SS ae SS"! Tabular concretions and shells. Saxe Shell-band. ( ia — > 2 —————— a ———___—— — Dark sandy clay or marl weathered 5 : =| _ buff, with fossils, wi === — | Thracia, Cyprina, and Pholadomya, Ad =| as at Reculvers. 4 & es Soh as bo) So ~ © | mB me 1S ° SS ek === >| Dark bluish sandy clay with decayed { 15't = pyrites; unfossiliferous. . (4 Sree fa 2 ‘5 peewesmeers—| Orange and drab banded sandy clay. Ss, 3 yn) The bands only distinct when wea- bale & ———————| thered; unfossiliferous. S¢|2 peaeerme | SO |Green flints. (A Spee Tabular flint. Total 89 ft. igs. 3-6.—Comparative Vertical Sections of the Lower Tertiaries of Herne Bay and Pegwell Bay. Fig. 3.— Hast Corner of Oldhaven Gap, facing the sea, Greenish-grey clayey sand, mottled pale yellow 12! -- Harder clayey sand with ferruginons stains, 4/ 6”, | Do. giving strong olive streak. 7/. Mottled 2 shades grey with few small clay galls, 4’, Rusty brown, yellow and green sand, Corbula-beds, 3/, Total 70 ft. London Clay 9, — Oldhayen Beds 33', Only casts of shells here. and whitish, Fig. 6.—Huast side of Small Chine, a little west of 16/ 5! 33 6! Jig. +, showing pebble-bed. Yellow loam. Black pebbles in loose sand. Mottled greenish-grey clayey sand streaked with orange. {Deep orange mottled, becoming grey. === Pale clayey sand with sbotin cotanctl Partings, becoming mottled with orange at base. Whity-grey sand, slightlymottled, piped. Td., slightly more argillaceous, with casts of shells, and crowded with Cyprina at base. (Concretions a little lower.) Fig. 4.—d little east of Small Chine, near east corner of Herne Bay. Q” Vor 1 Mottled pale green and Pay f ovange clayey sand. ue Pure orange sand. a Grey sand, mottled. 3r 9” (re grey clayey sand, a0 mottled irony. Corbula-band. Pale grey sand mottled darker, and piped; the 14’ 3” base more compact, colour, and fos- Tous. Line of concretions. About 11! Compact fawn-coloured exposed! sand. here. Total 32 ff. 9 in, by Survey. —asa ——_ Apparently classed as drift Vertical. A. 40’ thick in Survey. ‘The Jnst 15/ at a slope of 50°. Only 10! in Survey. _— —~ Nearly vertical. Total 39 ft. Oe L40 7dce p, 204, © —Pegwell Bay. Drift-loam. uff clayey sand. Jack pebbles (nob mentioned by Prestwich or the Survey). Shell-band. ‘Tabular concretions and shells. Shell-band, ark sandy clay or marl weathered buff, with fossile, Thracia, Cyprina, and Pholadomya, as at Reculvers. Dark bluish sandy clay with decayed pyrites; unfossiliferous.. Orange and drab banded sandy clay. The bands only distinct when wea- thered; unfossiliferous. Green flints. Tabular flint. * 2 \ 4 * . Ld Ls fe e a] y . fv na ; f A ; s Meek! in 7 whl ‘ b] [ae +A, F ? La 2 “ « . ,' Pa t 2 ae | ‘ ‘ 2 » F Pa ae y | Pine ~ ds Aw . - Ace GMs zi Py Owe oe DR eee: ida Mew th ome J. S. GARDNER ON THE LOWER LONDON TERTIARIES. 205 horizon can have crept in. Of 57 specifically determined bivalves of the Thanet Sands, 13 pass into the so-called marine division of the Woolwich-and-Reading Beds, while of 12 bivalves from the latter, only two, Cardiwm Laytonz, Morr., and Teredo antenaute, Sby., do not pass down into the Thanet Beds. If the Survey list is accurate, there is thus no marked paleontological break in the marine series. There is equally no perceptible difference at Reculvers in the quan- tity of green grains in the two sets of beds; while at Pegwell Bay a layer of black flint pebbles occurs in what is acknowledged to be Thanet Sands*. On these grounds I think the supposed break at this point, the obscurity and difficulty in detecting which has been remarked upon by all authors 7, should be ignored, and the succeeding 20 feet of beds be regarded as an integral part of the same marine formation. Going back to where the Corbula-bed first rises, we see it ovel- lain by about 8 feet of pale clayey sand mottled rust-colour, then about 10 feet of grey mottled sand, 6 feet of fine orange-coloured sand, 3 feet pale greenish sand mottled with orange, and a foot or so of darker sand immediately under the Oldhaven pebble-bed. The arrangement is slightly varied at the small chine already mentioned (fig. 6), while at Oldhaven Gap (fig. 3) the succession is as follows :— 12 feet pale greenish-grey sand mottled with yellow and white; 4 feet 6 in. slightly more clayey sand with ferruginous stains ; 7 inches mottled grey and greenish sand giving an olive streak when cut; 4 feet greenish sandy clay mottled in two shades, and with a few small clay galls; and 3 feet rusty brown sand mottled with greenish yellow resting on the Corbula-bed. These beds are of comparatively little interest, as the few fossils they contain are not of a distinctive character. Iron pyrites, small lignitic fragments, and a few sharks’ teeth are found in them. They have apparently been mud-banks in shallow water; and the pyrites may be due to decomposed sea-weed. A well-marked break occurs here, separating the next series. The Oldhaven Beds. These beds are rather sharply separated from the underlying beds by a layer of black pebbles, sometimes 18 inches in thickness. They are almost wholly composed of fine quartzose sand, with occa- sional small lenticular patches of brown clay. Their distinctive cha- racter is doubtless due to the fact that whilst the lower beds were deposited under water, these were deposited between high- and low- water marks by surf. They are seen first at the top of the cliffs west of the point, and at the small chine (fig. 6) are about 36 feet above the beach. The pebble-bed is here very distinct, some of the pebbles being of large size, and all except the smallest perfectly rounded. The top line is level as if smoothed by the waves ; but the lower line is uneven. The pebbles are imbedded in loose sand, overlain by two * The Pegwell deposits seem to have occurred nearer shore, as indicated by the presence of pebbles and broken-shell layers. t £.9. Whitaker, Quart. Journ. Geol. Soc. vol. xxii. pp. 409, 410, 419. 206 J. S. GARDNER ON THE LOWER LONDON TERTIARIES. feet of deep yellow slightly loamy sand, and then by quartzose sand — without green grains. At Oldhaven Gap the full thickness of the beds, 26 feet, is seen, the upper surface being very uneven and the fossils casts only. The sands vary considerably in hardness, but at no par- ticular levels. 250 yards west there are several small clay patches, the thickest a foot deep. These are sharply defined above, but below mix gradually with the sand, contain much selenite, and are cemented into sandy concretions which display beautiful ripple- marks and false-bedding. ‘They are marked with horizontal borings from one to one and a half inch wide. The clay patches frequently rest on the pebbles, but are sometimes separated by drifts of broken shells; angular pieces of clay are occasionally included, and some- times an indurated patch of lemon-coloured sand with iron-stains. 140 yards west the beds are 21 feet thick and very fossiliferous, with green grains towards the base. A lenticular patch of clay with sandy lamine, larger than usual, here encloses a Sarsen stone; and on the same horizon a few yards east are irony concretions crowded with shells. 160 yards west the shells approach within 2 feet of the Lon- don Clay, and there is much false-bedding ; and 90 yards further west again a few concretionary blocks are seen just above high water-mark. The bedding of the sand is lenticular; and the shells are almost com- minuted, and also drifted into lenticular form, both the sand and the shell-drifts being cut through again and again. A good section is seen 500 feet before the beds dip under the London Clay (fig. 7). Fig. 7.—Section showing Junction of London Clay and Oldhaven Serves. a. London Clay. 6. Basement-bed. ¢. Sections across the ancient furrows of the shore. d. Broken shells deposited in furrows. The Oldhaven Beds sink under the London Clay at about a mile east of Herne Bay. Their upper surface is eroded and filled in by the basement-bed of the London Clay, here only from 12 to 18 inches thick. The basement-bed is of a warm drab colour mottled with J. S. GARDNER ON THE LOWER LONDON TERTIARIES. 207 whitish, weathering a rusty brown, giving a green streak to the tool when struck, and containing much woody matter and large green grains*. It is a compact bed with an uneven upper surface very distinct from the overlying blue London Clay. When last seen the Oldhaven Beds appear comparatively indurated and of the ordinary buff colour of sand when dry, and greenish grey when wet. Fossils last appear about 7 feet below the junction, and 200 feet east of the disappearance of the beds, drifted into pockets, though a little east of the latter irregular lines of fossils approach within 5 feet of the junction. These Oldhaven Beds can be traced a long way inland without any change in character, and are immediately recognizable by their base of black pebbles. Their fauna is very rich, far richer than the Survey list of 1872 would lead us to suppose, and approximating more closely to that of the London Clay than has been admitted; but among masses of comminuted shells only one here and there is perfect. I have not observed any very local distribution among them, except in the case of a large Aporrhais, all the specimens of which were found close together. By far the commonest shells are Cytherea and Protocardium. The tidal and surf-action is everywhere apparent, and their origin is unmistakable ; but whether they were formed, as I believe, on a coast-line, or on banks away from shore, is less ap- parent, the absence of unworn flints showing at least that no chalk was present in the immediate vicinity. The greater variety of the Mollusca may be accounted for by dead shells being heaped together from various depths, as we see at the present day on coasts; but the presence of a few turtle bones, taken with evidence gathered else- where, must also indicate that the increase of temperature had already commenced. We see by the Survey list of fossils that the fauna was intermediate between those of the London Clay and the Thanet Beds; for only two species are now not known to range beyond it; 7 of the species range downward only, 12 range both up and down, and 13 range upward only ; so that it is united by 75 per cent. of its species with the London Clay, and by 50 per cent. with the Thanet Beds. It is thus certain that there is but one break in the marine deposits of the Lower Eocene in East Kent; and I cannot help thinking that it would be far more convenient to retain the name Thanet Beds alone for the lower division. A great interval may have elapsed between these and the Oldhaven Beds, which belong paleeontologically to the London-Clay series. The Oldhayen Beds might fairly be con- sidered, as I shall show, to be a lower member of the latter series, as, indeed, was implied by Prestwich’s name for them. The upper portion of the marine Thanet Beds of this part of Kent may have been, and probably was, to a sight extent contemporaneous with the fluviatile Woolwich-and-Reading Beds of furtherwest. The incoming of these cannot be traced on this coast ; and there is no section appa- rently revealing them until the neighbourhood of Chatham is reached, _ * The lower beds of London Clay are here tenacious and blue even when weathered. - 208 J. 8. GARDNER ON THE LOWER LONDON TERTIARIES. when at Upnor the shelly clays with Cyrena make their appearance. The sharp sands in this section (Mem. Geol. Surv. vol. iv. part 1, p. 144) I should feel inclined to place with the Thanet Beds, thus in- creasing their thickness there by some 30 feet. At Erith the plastic mottled clay of the Woolwich-and-Reading Beds first appears, and is some 9 feet thick. At Charlton it is absent; but at Loam Pit Hill I detected it a few inches thick* in its proper position below all the fossiliferous Woolwich Beds. I should most decidedly place the base of the Woolwich-and-Reading Beds at this horizon, and consider all the marine shingles and sands below the plastic clay an integral part of the marine Thanet formation. In the magnificent section recently exposed in the new railway works at Croydon, the division is very clearly seen, and the whole Woolwich-and-Reading series is exposed. Below the plastic clay and above the admittedly Thanet beds is a series of clayey sands with green grains, of marine origin and with occasional fossils and flint pebbles. There is not the re- motest reason, that I can discern, for separating these from the underlying series and making them a marine member of the over- lying fluviatile series. As the section will probably be described by Prof. Morris and Mr. Klaassen, it is unnecessary to go into details here; but if my classification is admitted there would simply be a Lower Reading series with a very distinctive flora, and an Upper Woolwich series with a very distinctive flora and fauna—the former purely fluviatile, and the latter becoming estuarine towards its close and therefore marking a subsidence with its inevitable accompani- ment, an advance of salt-water conditions up the estuary. To meet beforehand one criticism that may possibly be advanced against this view, I may at once state that I feel no doubt that the marine Dorsetshire oyster- and pebble-beds hitherto referred to the Woolwich beds, are of London-Clay age, and mark the western shore of that sea, and not of a previous sea. The Reading Beds at Alum Bay are purely fluviatile mottled clay, and the London Clay rests directly on them; but further west the London Clay is replaced by oyster-beds, pebble-beds, and pyritous sandy clays resting on the same mottled clays on which the London Clay rests at Alum Bay. _ It would be rather inexplicable if the Woolwich-and-Reading series, after becoming more and more distinctly fluviatile from east to west, should again become marine in the furthest west; while on the other hand the London-Clay sea, which was of considerable depth as near ° as Alum Bay, must undoubtedly have possessed a shore to the west of that point. The Thanet Beds were probably deposited, as I have shown, by a rough sea in an area out of the immediate estuary of the river, but within its influence. The area became silted up until it finally rose above the surface and became covered over with shingles and sand. The Thanet Beds close with a period of elevation, during which the Reading Beds were formed ; and these were followed by a subsidence during the Woolwich period, which finally ushered in the Oldhaven and London-Clay deposits. The Oldhaven Beds may * Bed 9, section Loam Pit Hill, p. 127, Geol. Survey, vol. iv. pt. 1. J J. 8. GARDNER ON THE LOWER LONDON TERTIARIES. 209 represent some such action (but under a rougher sea) as that which now forms the beach at Shellness, not far off; and it is conceivable that if this were an area of gradual depression, as it was at the commencement of the London-Clay period, the beaches there would advance steadily over the flat area of Sheppey on the one hand, while on the other, as they sank out of reach of disturbing waters, they would become covered up by the silt of the Thames, just as the Oldha- ven beaches were covered by the London-Clay silt of the Hocene river. The depression was maintained for an enormous period, the salt- water estuary gradually extending up the river-valley as far as the western limits of the Hampshire basin, and deepening at Sheppey until nearly 500 feet of silt was deposited. The Thames mud at the same spot is already nearly 100 feet thick, and this on the present shore and consequently out of the main channel. The advance of the London-Clay sea was distinctly not due to any planing action, but to gradual subsidence; for the London Clay always rests on other Eocene formations, though elsewhere and far off the Chalk must have been incessantly attacked, as it is now, by the waves. Beaches such as the Oldhaven deposits could only be formed near the mouth of the river; and we accordingly see that they diminish and disappear in Hampshire and to the west. Beaches may equally have been formed along the shore of the Thanet- Sand sea, and left stranded when it retired ; and it is not always clear to which agency many of the vast aggregations of shingle and sand between the Thanet Beds and the London Clay may belong. They are, however, an integral portion of one or other formation, and should not be recognized as a separate formation at all approaching other divisions of the Eocene in value. The same may be said of the so-called marine Lower Bagshots, which mark the retreat of the London-Clay sea. That any vast lapse of time occurred between the Thanet-Sand and London-Clay period I do not at present believe, seeing that the enormous change of temperature that almost suddenly took place between them, a change which drove the indigenous flora northward to Greenland, would amply account for the difference in their faunas. The Woolwich Beds at least, however, were formed during the interval, and the European Tertiaries may furnish data as to its duration. I submit this as a simpler explanation of the formation of the Lower London Tertiaries, and a more definite classification of their complicated changes, than that hitherto prevailing. If accepted, it can by no means diminish the value of any work previously accom- plished ; and, while involving no new terms, it renders the recogni- tion of the divisions in the field far easier. Discussion. Prof. Prestwick remarked upon the value and beauty of the collections made by Mr. Gardner. At the time he had written his own paper the exact horizon of many forms had not been deter- mined, and the careful collecting of Mr. Gardner would clear up many of thesé doubtful points. He did not, however, think that the theoretical views propounded in the paper were of equal novelty 210 J. 8. GARDNER ON THE LOWER LONDON TERTIARIES. and interest. Every one admitted the great gap between the Chalk — and Thanet Sands, and that the Thanet Sands, Woolwich Series, and Basement Beds were, like the ‘“‘ Landenian System” of Belgium and the ‘‘ Sables Inférieurs” of France, merely minor subdivisions of one continuous formation, with variations caused by local subsi- dences and elevations. He had recently found marine shells ag far west as Reading. The fluviatile Woolwich Beds pass at Herne Bay into marine beds with a poor fauna. In Belgium and France these beds contain a much larger marine fauna than our own. He thought the actual facts did not support the views of the author. Mr. Wuirtaxer stated that the separation of the Oldhaven Beds on the Geological Survey maps was due to the fact that they were clearly recognizable, and capable of being followed and mapped. He found no reference in Mr. Gardner’s paper to the fact that the Oldhaven Beds are not always marine, but sometimes fluviatile and fluvio-marine. He agreed that the apparent conformity of the Chalk - and Tertiary in England does not prove that there is no gap. He had clearly recognized long ago that in Hastern Kent there is no marked separation between the Thanet and the Woolwich Beds. The three members of the Lower London Tertiaries are really very small and insignificant subdivisions. He thought that the perfectly rolled condition of the flint pebbles of the Oldhaven Beds pointed to the conclusion that they were formed, not on a beach, but a little way out at sea. The Woolwich Beds sometimes lie on eroded surfaces of the Thanet Beds, which is opposed to the author’s views. So also. is the alternation of plastic clays with shelly Woolwich Beds, which makes it impossible to divide the Reading from the Woolwich Beds. He remarked on the local character and distribution of the Oldhaven Beds as compared with the Woolwich- » and-Reading Beds; in some places the former rest unconformably on the latter, on the Thanet Beds, and on the Chalk. . Pror. SeeLry remarked upon the important fact pointed out by Mr. Sorby that the Thanet Sands were mainly derived from granitic rocks. He thought the existing classification was a convenient one, especially for teaching-purposes. He doubted the accuracy of the general view supported by the author, namely that the materials of these strata were derived from the west. He himself thought that both paleontological and physical evidence pointed to the conclusion that the materials were derived from the east. He doubted not only the author’s conclusions with regard to the climates of these different periods, but all conclusions of the same kind. The Autor stated that he had found a very characteristic flora under the mottled clays of the Reading Beds, a flora also found at Newhaven and in Greenland. He believed the mottled clays to be purely freshwater deposits. These beds are traceable at Lewisham and Blackheath under the Woolwich Beds. He maintained that there is no means of comparing the faunas of the two beds; but the floras are dissimilar and show a marked climatal change. He did not agree with Mr. Whitaker and the Survey that the Bromley leaf-beds belong to an Oldhaven freshwater series, but thought them to be on the same horizon as the Woolwich flora. LOWER CARBONIFEROUS ROCKS OF THE FOREST OF DEAN. 214 1 14. On the Lower Carsonirerovs Rocks of the Forust of Dzan, as represented in ‘TypicaL Sxcrions at Drysroox. By LE. Weruerep, Hsq., F.G.S., F.C.S. (Read February 7, 1883.) { Abridged. ] In this paper I shall first refer to a series of arenaceous and calcareous shales which lie between the Old Red Conglomerate and Lower Limestone Shales, and which represent the true basis of the Carboniferous rocks in Gloucestershire. (2) I shall refer to some beds in the Lower Limestone Shales which may serve as horizons for correlation with other coalfields. (8) I shall consider some features in the Millstone Grit. 1. The Old Red Conglomerate at Drybrook dips to the south-east at an angle of 38°. It is made up of ‘vein-stone” quartz * pebbles, grains of ordinary quartz, measuring about ‘021 of an inch in dia- meter, plates of mica, and a little felspar. Of the quartz grains about fourteen per cent. can be recognized as crystals; but the rest are waterworn. The grains are coated with oxide of iron; but on this being removed by acid, cavities are exposed to view, some of which contain bubbles. In addition to the deposition of oxide of iron on the grains, silica is often found attached to them in minute well- defined and clear crystals. The Old Red Conglomerate passes conformably into a series of loose sandy beds and light and greenish-coloured calcareous shales, dark shales being absent. A section of these beds, carefully measured by the late Mr. John Jones and Mr. W. C. Lucy, F.G.S., showing 247 feet and 108 divisions, appeared in the Proceedings of the Cotteswold Club for 1867. They are there termed “ Transition Beds.” Sir H. De la Beche has given sections of the junction of the Lower Limestone Shale with the Old Red Sandstone rocks T, and has noted strata similar to the beds to which I am referring; but they appear to have been much less developed in the localities where his sections were made than is the case at Drybrook. He has, however, referred them either to the Old Red Sandstone or to the Lower . Limestone Shales. The sandy beds and calcareous shales are characterized by the rapid succession of beds and by the variety of colour. In order to examine them minutely, five beds were selected as typical, and specimens taken from them ; the first was from near the base, and the others from various parts proceeding upwards. No. 1 was a light-coloured bed with but slight cohesion between the grains composing it. It contained a considerable quantity of mica * This term, vein-stone quartz, is used by Sir H. De la Beche, Memoirs Geol. Surv. vol. i. p. 64. t+ Memoirs of the Geol. Surv. vol. i. ‘‘ Formation of Rocks in the South-west of England,” p. 142, and in other parts of the monograph. Q.J.G.8. No. 154. Q 912 _ E. WETHERED. ON THE LOWER CARBONIFEROUS of white and greenish-brown colour. elspar was present in fairly well-defined crystals. The quartz contained cavities similar to those found in granite. The minerals occurred in clusters, and appeared to have been derived from granitic rocks. The larger grains of quartz averaged about -007 inch in diameter. No. 2. Colour light brown; more compact than the last. Same minerals present, but clusters less numerous. The grains, as a whole, more waterworn, the larger ones averaging about -008 inch in diameter. Calcareous fragments present. No. 3. From about the centre of the series. A calcareous shale of light colour with a pale tinge of pink, very micaceous. It gave the following chemical composition on analysis :— Saltese iexy. ces set eds and ret ota eS: re. oe ees 76°60 PURE SI EOC E: Nee Reeeeeen Meh ee En Pest Rare NeoN SN «REO 8 SS 8°30 COGS WOE ATOR Soto eckececicicc dace tousiecens eee ee ne ace eee 2-713 Mabe is deE ae Ronee oe coee iat tea e acer tee eee eee 5:00 Cxide Of Man PaNese: 2 sei soc .osseseeoee sehboeme eae eeee "13 Gar bionic ACiaMte te: MES os Mee ee Se) ee 5:00 Mineinesia Vig. £3655 sdcovecl oe caatwe deantispeiess nee sneer eee 1:85 AUoiliea oS 3. eos etek ce ee Cee one ene WN ciwcianae Seen 1:12 EGOSEOM: TOMNGION °c Macwen <0 vasa es bole gua ch deeenearccenen "39 100°48 No. 4. Contained small “vein-stone”’ quartz pebbles, similar, except in size, to those in the Old Red Conglomerate. Minerals present the same as previously noticed. Quartz grains more water- worn than before observed; the larger ones averaged about -012 of an inch in diameter, It gaye the following chemical composition on analysis :— SUTRAS. «sae ated Pvranictere ec cal aise be ete. Meme tiee oe erg a 92-66 ATi tasies! 3: ohn suse e cidt Sebi Nick gk Ree be eee be Eee 2-46 RD SICLES NOE PHETATO ee on igsae th cireck sh dtianrkss pee de heen SRR Tide ARE ih 3) C1 77 SN aR CU REE a em RI Ry 2 Oot 86 Cigar Gl MANMAMERS |. vcs somense orawee toes voriteee eee oe oe riley Warhanie aed sorte sss eee ee ee eee 45 Wiritesiey 6. '¥5 Gi). SEG UE eG ee RULE eee Mone ite trace. Alkalieses. cciat oes eee Sree Mees See, ay Pe aed 1:00 HORS OTE AOMGIOND, «oo athe mane passin ia dar escinerek eee 1:20 99°89 No. 5. Colour yellow; could not be distinguished, lithologi- cally, from some beds of Millstone Grit in the neighbourhood, except that the grains of quartz in the latter are more waterworn. Same minerals present as before noticed, but a greater preponderance of quartz; the larger grains measured about ‘Oli of an inch in diameter. 2. The beds to which I have just referred are succeeded by the Lower Limestone Shales, the nature of which is shown in the following section :— ROCKS OF THE FOREST OF DEAN. 213 No, I.—Section of a portion of the Lower Limestone Shales exposed in Quarry near Drybrook, Forest of Dean. Petrology. ft. in. Chief Fossils. Not exposed. ; eBeiittes HINESLON Gy cose vou casiea Seale 8 CRE 2 30 “47 GARG TEOUE Cook ake on cn ocionncanetspaetieeon 50 80 LETTE, Atha an ae a5 pleat nD Rik Rn 33 “44 Abrm DORGAN ee See oaks dedewcasetacaeretes 30 39 Ye 1) RGN TGS 8 i ag IRR Rc traces!) GORENES Loss on ignition (carbon) .................. 20 A ae RPGISGNING etre eee ease a te csacedecaeer deseen 22, Reviewing the evidence of sequence, the Old Red Conglomerate, the many-coloured sandy beds and calcareous shales which imme- diately follow, and the Millstone Grit, all seem to have been formed by materials derived from the same source, namely ancient granitic rocks ; but the sediment and pebbles have been subjected to various degrees of mechanical attrition: the evidence of this is fur- nished by the occurrence of the same minerals in these rocks, and especially by the “ vein-stone” quartz pebbles. There is no uncon- formability ; and therefore there is no reason for supposing that one set of beds can have originated from the denudation of the other: this being so, there is no alternative but to accept the view which I have expressed. As to the correlation, I will only say that the sandy beds and calcareous shales, between the Old Red Conglomerate and Lower Limestone Shales, have no Old Red affinities, and occupy the position of the Calciferous Sandstone group in Scotland; the lower beds of * JT have since come across the same bed in another place regularly interstra- tified with the Millstone Grit. 216 E, WETHERED ON THE LOWER CARBONIFEROUS Millstone Grit, together with the upper portion of the Carboniferous Limestone, occupy the horizon of the Yoredale rocks of the North of England. This matter I shall treat of in a future paper. In conclusion, I desire to return my thanks to Mr. W. C. Lucy, F.G.S., for his kindness in drawing my attention to the rocks referred to in this paper, and for his assistance in working them out. Mr. Lucy is not, however, in any way committed to the views which I have expressed. J also tender my thanks to Mr. F. D. Longe, F.G.S., for the ready way in which he placed his specimens from the Clifton Polyzoa-bed (Stoddart) at my disposal. The whole of the strata mentioned are shown in the following - section :— No. 3.— Vertical Section of the Carboniferous Strata in the Forest of Dean. Approximate thickness. Description. tt. Characteristic Fossils. Coal-measures (do not come in at 3243 Plant-remains. Drybrook). Millstone Grit, made up of well- 870 None yet found. rounded grains of quartz. Lower Millstone Grit, flagstones 100 Casts of Mollusca, plant-remains. and sandstones. Upper Limestone. 116 Corals &e. Carboniferous Limestone; hema- 3860 Producti, Spirifere, &e. tite in upper portion. Lower Limestone Shales. ............ 1380 Encrinites, Rhynchonella pleuro- don, Athyris Royssi, Polyzoa, Annelid-jaws. Many-coloured sandstones and cal- 247 No fossils identified. careous shales. Old Red Conglomerate. DIscussIon. Mr. Erueriver differed from the author as to two points in his paper. (1) He thought the term Calciferous Sandstone was inap- plicable to the series described -by him; for neither lithologically nor palontologically did ‘they correspond with the Tuedian or Cement-beds of the north. ‘Two fossils came from the latter and were always present, an Athyris and a Nautilus: these were wanting in the beds described by Mr. Wethered. (2) As regards the Upper Limestone Shales, there were Upper Shales in the Bristol district ; but these differed paleeontologically from the Yoredale series of the north, and it required better proof in order to identify them. As regards the Polyzoa-bed; there was no doubt of its position ; but it differed much from that described by Stoddart at Bristol. He ex- plained the reason why so large an area was occupied by Millstone Grit. The Yoredale beds could hardly come in there, as the lime- stone was a deep-sea deposit. Mr. Torey said the paper showed great. care and accuracy ; but -_ ROCKS OF THE FOREST OF DEAN. 217 in the question of correlation there were not only the difficulties mentioned by the last speaker, but further the Calciferous Sand- stone of Scotland represented a large part of the true Carboniferous Limestone of England. The Yoredale series also represented a part of the Carboniferous Limestone series. Mr. De Rance also objected to the correlation. As regards the Yoredale beds, though present in Derbyshire they are absent in North Wales and around the Shropshire coalfield. The Millstone Grit, when traced 8.S.W., thins out, and is missing in North Wales in one area. Going further south, the Millstone Grit comes in again ; but the Yoredale rocks do not. The name Yoredale was to be regretted ; for the Yoredale limestones of the typical valley of the Yore are really true Carboniferous Limestone, and correspond paleontologically and physically with it. Mr. WerHEzED, in reply, said that, if the beds in the Forest of Dean occupied the position of the Calciferous Sandstones of Scotland, the probability was that they corresponded with them in point of time. The paleontological argument also seemed contradictory ; for Mr. Etheridge said the Calciferous Sandstone had only two fossils, but Mr. Topley asserted there were plenty. .As for Mr. De Rance’s arguments, he did not see that lithological similarity could be ex- pected in the northern Yoredales and in these southern rocks, Even in the latter region the same series underwent a change in a comparatively short distance. All he said was that they differed from the Millstone Grit and were in the position of the northern Yoredales. 218 T. GRAY ON A SEISMOGRAPHIC APPARATUS. 15. On Gray and Mitnr’s SrismocrapHic APPARATUS. By Tuomas Gray, Esq., B.Sc., F.R.S.E. (Read March 7, 1883.) THE apparatus a description of which forms the chief subject of the present paper, is intended to give acomplete record of the nature, magnitude, direction, and frequency of the motions of a point of the earth’s surface during an earthquake, and at the same time to record the time of occurrence and the duration of the earthquake. The apparatus has been made by Mr. James White, Glasgow, and is to Fig. 1—Gray and Milne’s Seismographic Apparatus as set up. (Scale about one twelfth.) ! Ms ji is mt a iy, ( ! | ty’ lin My if be used by Prof. Milne in his investigations as one of the committee appointed by the British Association for the investigation of the earth- quake phenomena of Japan. ee eee T, GRAY ON A SEISMOGRAPHTC APPARATUS, 219 The instrument is suited for recording motions of the earth which are under one centimetre on either side of its normal position, while the period of oscillation is less than one second. The instrument may be adjusted to register motions having a longer period; but it is not expected to be called upon to do so. The actual motion of the earth is to be derived from a record of three separate components of the motion—two horizontal and at right angles to each other, and the other vertical. The horizontal components are recorded by a pair of conical pendulum-seismographs (see Phil. Mag. Sept. 1881), one of which is shown at P, figs. 1 & 2. These instruments are con- siderably simpler in form, and can be more readily adjusted, than the instruments described in the paper above referred to. They consist Fig: 2.—Diagram of Pendulum-Seismograph recording horizontal movements. 2 S S 7 NG NS ZN QUEL SASITE TELS QOL: MMH GHZ; 4 LE, LZ GA Waren Peay restceaircear cme (| | = TIAA Z Li = typ GY ty Gui Liles 544 i = ee COLLE, WHHHA tp he SoZ VIo Looe: ce y “ft SSS y WH: WOW AS NI VSS SLL. NN as =~ of cylinders of lead, P, supported from their centre of gravity by means of a strong silk thread, which is suspended from a point which can be moved by a simple screw arrangement in either direction, horizontally 220 T. GRAY ON A SEISMOGRAPHIC APPARATUS, or vertically. The pendulum, P, is kept deflected by means of a light brass tube, which passes through its centre in a direction at right angles to its axis, and which carries at one end a vertical knife-edge, and at the other end a very light index of aluminium foil. This thin tube rests with its knife-edge in a flat V-groove, which is fixed to the supporting column, C; and by sliding it more or less through the tube, a greater or less deflection can be given to the pendulum. The other pendulum is exactly similar to P, with the exception that the aluminium index is fixed at right angles to the tube and close to the knife-edge. By this arrangement the two pointers can be ad- justed to lie in the same horizontal plane, and in parallel directions. When so adjusted, the planes of deflection of the pendulums are at right angles to each other. The point of suspension is adjusted to be a little in front of, but very nearly vertically above the knife- edge, so that the period of free vibration of the pendulum round a vertical axis is long, compared with the period of vibration of the earth. The vertical component is recorded by means of a compen- sated spring seismograph (see Phil. Mag. Sept. 1881), which consists of a mass M (figs. 1 & 3) attached near one end of a horizontal lever J, Fig. 3.—Diagram of Spring-Seismograph recording vertical movements. which turns freely round a knife-edge at /, and is supported by the flat springs 8, which are attached to the lever by links working on knife-edges. To the free end of the lever one end of a small box, T. GRAY ON A SEISMOGRAPHIC APPARATUS. pA | B, is hung, the other end of the box being fixed to a horizontal axis. This box is partly filled with mercury for the purpose of compen- sating by its negative stability the positive stability of the spring. The mode of carrying the mass on the end of a lever is adopted in these instruments because the period of free vibration can by this means be made moderately long with a short spring, thus render- ing the instrument more compact and easily compensated. To the front of the box B, a very light vertical index is attached. This index is made of a very thin tube of aluminium supported by angle ties of silk thread so as to give stiffness. The threads are kept tight by a spring which presses against the top of the tube. The three indexes of the instruments just described rest with their outer ends against the surface of the drum D, covered with smoked paper, the points being arranged in the same straight line parallel to the axis of the drum. When the earth, to which the heavy cast- iron sole plate and supporting pillars are rigidly attached, moves in a direction at right angles to the line joining the knife-edge and the centre of inertia of the masses P and M of the respective instru- ments, they remain behind in virtue of their inertia, and thus the indices are made to move across the drum in a direction at right angles to that in which it is kept moving by the train of clockwork W. The distance moved by the points of the indices, combined with the known multiplication of the instrument, is sufficient to indicate the direction of the motion, while the shape of the curves traced on the paper and the distances between them indicate the nature and period of the motion, the rate of motion of the drum being known. The drum D is driven by a clockwork train of such length that a fall of the driving weight of one foot is sufficient to keep the drum going at the rate of one turn in two minutes for twenty-four hours. This train of clockwork is actuated by two separate driving wheels, one on each side of the first pinion, so as to avoid at the same time ex- cessive pressure on the bearings of the driving wheels and any great pressure on the bearings of the pinion wheel. It is governed by a continuous-motion governer, which will be understood from fig. 4, in which 6 is a cylindrical brass box mounted on a vertical axis of hard steel running at its lower and upper ends in jew- elled bearings. The box is connected to the clockwork through the pinion p and the crown wheel w. A small quantity of liquid, J, is poured into the box; and this, when the box is made to revolve rapidly, forms into a paraboloid of revolution. When the liquid reaches a certain height dependent on the speed required, it comes into contact with a fixed vane v, which checks the motion of the box and keeps it going at a uniform rate. It is intended to keep the clockwork continuously in motion, so that when an earthquake takes place itis ready to receive the record. The duration of the shock is then to be reckoned from the length of the record. The time of occurrence of the shock is determined by a modified form of the apparatus described in the ‘ Philosophical Magazine’ for Nov. 1881, p. 363, and called a ‘“time-taker.” The earthquake causes a very delicate circuit-closing apparatus (see ‘ Philosophical 22> T. GRAY ON A SEISMOGRAPHIC APPARATUS. Fig. 4.—Diagram of Continuous-motion Governor. 4 SAAAAAAASSAA $ SSS Magazine’ Nov. 1881, p. 358) to close an electric circuit which, ac- tuating electromagnets, causes a mark to be made on the drum D, and at the same time the dial of the time-keeper T to move suddenly forward against ink-pads fixed to the ends of the hands. In this way the hour, minute, and second at which the mark was made on the drum D is recorded ; and this, combined with the record of the motion which is being written on the drum, is sufficient to determine the time at which any particularly prominent feature of the shock took place. The instruments just described have been made to suit the earth- quakes which are commonly experienced in Japan, and are conse- quently not adapted for use where the displacement of the earth is great. A similar set of apparatus could, however, be made to record earthquakes of considerable magnitude. The utility of having such records of earthquakes will be readily admitted; and I should call special attention to the great value of the combination of the record of the time of the occurrence with that of the motions. When the time of occurrence of any of the chief features of the shock is known for a number of places, it becomes a comparatively easy problem to determine the origin of the shock and its rate of propagation. It has been the usual practice in earthquake-observations, to take an approx- imate measurement of the magnitude and direction of the movement at any place and to deduce angles of emergence and direction of pro- pagation. Now it is hardly necessary to point out that the direction of movement at any point is very much influenced by the presence or absence of surface undulations, and of refractions and reflections from strata of different elastic moduli. I am of opinion that the T, GRAY ON A SEISMOGRAPHIC APPARATUS. O25 direction of motion at any one point cannot be taken as giving at all reliable evidence of the direction of propagation of the shock. Discussion. The Rev. E. Hitt remarked that it was desirable that observa- tions of this kind should be made in this and other countries in which only small earthquake-movements take place. He called attention to the necessity of distinguishing the motions directly due to the shock from the subsequent swinging of the pendulum. Prof, Jupp asked if arrangements were made in this instrument for recording several earthquakes in succession. Prof. Hueuxs asked whether the instrument had been tried along- side of others differently constructed, and whether the results ob- tained were the same. Itseemed more complicated than the instru- ments used in the Vesuvius Observatory ; and he would be glad if Mr. Gray would explain the advantages of his arrangement for dipping the metal point in the mercury over that adopted by Palmieri, in which the weight carrying the point is simply suspended by a spiral spring attached to the solid framework. The Avtuor, in reply, stated that the swing of the pendulum was of such a long period, as compared with that due to the earthquake, that they could not be confounded with one sr DgHTete The instru- ment would record any movements between 54, of an inch and 3 of an inch; the multiplication might be varied “from 4to7. No pro- vision was made for the occurrence of more than one earthquake in the same night; but such repetitions of earthquake-shocks were rare in Japan. The instrument had not yet been employed anywhere in its present combination. In reply to Mr. Evans he stated that an instrument of this kind would cost about £45. 224 E. A. WALFORD ON THE “ NORTHAMPTON SAND ” 16. On the Retatton of the so-called “‘ Nortaampron Sanp” of Norra Oxon to the Cryprus-Grit.. By Epwiy A, WALFORD, Esq., F.G.S. (Read February 21, 1883.) Tae diverse lithological conditions of the deposits in the three great sea-basins of the Bajocian period, the Anglo-Norman, the Cotteswold, and that of the north-eastern counties, have for a long time engaged the attention of geologists in the correlation of their various subdivisions. The Anglo-Norman basin, however, with the rich molluscan fauna of its marly limestones, willnot come within the scope of the present paper. It is to that border-land where the estuarine character of the deposits of the northern area becomes merged and lost in the marine strata of the Cotteswold type that I wish more particularly to direct your attention. The greater part of this section of the Midlands, North Oxfordshire, is embraced in sheet 45 N.W. of the Geological Survey. It may perhaps be worth while to pause for a few moments to review the present subdivisions of the Inferior Oolite. Resting upon the sands for which Mr. Witchell has appropriately suggested the name of the Cotteswold Sands, and which, throughout the Cottes- wold area, cover the Upper Lias Clay, are the well-known pisolitic beds of the Murchisone-zone. They are exceedingly local in their range, and are soon overlapped by the succeeding Lower Freestone beds. These find their equivalents, according to Prof. Judd and Mr. Sharp, in the lower part of the sandy beds of the Midlands known as the Northampton Sands, and northwards in the equally well-known Dogger of the Yorkshire coast. The succeeding Cottes- wold stage, the Oolite Marl of the Sowerbyz-zone, passes from its marly character in the north to that of a Nerinwan and coralline limestone southwards, and, much reduced in thickness, courses through Dorsetshire to the coast. gg ge cae biia o> 5 aN ' 2. Sandy oolitic limestone rubble...............,..... 3a 6 3. Oolitic limestone with black shining con- Pog E1001 te oe een ners Ao Je A OO Va 3 10 A Sea thy Wal Bis ois cae ben 5055 elements eae ke at 13 to 5. Oolitic limestone with black concretions and dimiesbone pebules... .<..o.:.cte-aseseeee see dancss 2 2, 6. Friable oolitic very shelly marl, with masses of worn corals, sponges, (?) Lima pectiniformis, Terebratula submaxillata, Serpula socialis, &e. 7 7. Shelly oolitic limestone; base not shown ...... : ee Professor Judd, in the admirable introductory essay to his ‘ Geology of Rutland’ has briefly described the beds, and has given a list of fossils from the collections of Mr. T. Beesley, F.C.S., and the Survey. This list I have fortunately been able to supplement with a few species, see Table A (p. 239). Amongst the common fossils are Natica cincta (leckhamptonensis), Terebratula submaxillata, Serpula socialis, and Ostrea Marshiv. Amongst the Polyzoa occur species of Apseudesia, Diastopora, Tubulipora, and Stomatopora, whilst the Millepore Spiropora stra- minea isnot uncommon. In the ‘Geological Magazine’ for May 1882, Mr. Hudleston has shown WNatica cincta to be a characteristic shell of the zone of Ammonites Sowerbyi, not only in the Millepore- beds of Yorkshire, but also in the Oolite Marl of the Cotteswolds. The occurrence in Oxfordshire of this shell with the Polyzoa so fre- quently found in this horizon adds to its right to be so considered ; and hence we may infer that we have a local development of a zone hitherto unrecognized in our area. Perhaps, however, Professor Judd may have included this phase as a subzone of Ammonites Murchisone. The Ammonites are unfortunately so badly preserved as to furnish little additional evidence. The numerous corals, the bulk of which are rolled and waterworn, show the coralline conditions of this oolitie sea. The characteristic Brachiopoda, with the exception of one doubt- ful specimen of Terebratula fimbria in Mr. Beesley’s collection, are absent. Terebratula submawillata, however, is abundant*. Banbury and Hook-Norton type.—In theimmediate neighbourhood of Banbury a few sand-capped hills and an outlier preserved by a fault alone testify to the former extension of beds of the Bajocian series over the locality. Nine miles north-west of the town a small patch of sand occurs on the crest of one of the picturesque hills of Burton Dassett, marking, so far as we can ascertain, the limit in that direction ; whilst north-eastwards the ridges of Byfield, Red Hill, and Thorpe Mandeville border the more extended deposits of North- amptonshire. * By the kind courtesy of Prof. Prestwich, F.R.S., I have been enabled to examine collections from Blockley, made by Mr. 8S. Stutterd and now in the Oxford University Museum, A fauna almost identical with that of Coombe Hill, including even Natica cincta, is shown, associated with characteristic forms of the Oolite Marl, such as Terebratula fimbria, Waldheimia carinata, and Rhynchonella subosoleta. Q.J.G.8. No. 154. R 228 E. A. WALFORD ON THE ‘‘ NORTHAMPTON SAND ” At the Constitution-Hill section near Banbury all that remains ~ of the Inferior Oolite is from twelve to twenty feet of white and fawn-coloured sands with occasionally bands of stone containing such fossils as are here quoted, associated with numerous plant-remains. Between it and the Great Oolite above, both of which have been let down by a fault, a thin stratum of black clay intervenes. Pycnodus. Pecten articulatus. Trigonia v-costata. | Avicula braamburiensis. conjungens. | Ostrea gregaria. Tancredia axiniformis. , Sp. Corbicella bathonica. Diastopora. Astarte, sp. Serpula filaria. Car dium, sp. Montlivaltia lens. Gresslya abducta. Annulated Algee. The type of deposits of the district N. and E. of Banbury is very similar—sands, often ferruginous, with occasional courses of shelly limestone towards the base. These rest upon and pass into the clays of the Upper Lias, without, so far as I have seen, any appearance of unconformability other than in the marked change of life. South- westward the same conditions prevail. Over the high lands from Tadmarton Camp to Hook Norton stretches a series of beds of sand and sandy limestones, below which follow courses of brownish crystalline limestone containing Montlvaltia lens, Pecten demissus, P. personatus and P. lens, and rolled stones covered with Serpule and Polyzoa. From the sandy beds were obtained annulated stems of Alge; and by the Rey. P. B. Brodie a pretty new species of Trigonia, described by Dr. Lycett, in his monograph of the British Trigoniw, under the name of T'rigonia Brodiei, was got from the limestones which appear to represent the base of the Inferior Oolite. ~ The sequence of the beds may be noted in passing up the Mil- combe Hill towards the Camp, a fine specimen of a Romano- British earthwork. The road-cutting exposes the Upper Lias clays ; and further on a small limestone-quarry yields, though sparingly, the fossils mentioned. The overlying sands are shown best in the sand-pits on Tadmarton Heath at the top of the hill. . Other, though sparingly fossiliferous, sections, where courses of hard limestone come in between the sandy beds, may be seen near the Gate Inn, Hook Norton, by the roadside south of Sibford Ferris, and near the Temple Mill. The whole of these beds, hitherto termed ‘‘ Northamp- ton Sand,” are included under that phase of the Inferior Oolite marked in the Survey map 5'g 7’; and this same 5’ g 7’ aiso includes the lower zone of the Great Oolite of this district. The new railway-cutting about half a mile south of Hook Norton has fortunately supplied a section (fig. 1) so complete that we may take it for our type in attempting the correlation of other beds of the dis- trict. The hill has been but partially worked through, the cutting on its north side being the longer and deeper; that on the south side near Duckpool Farm, though shallower, still shows the fossili- ferous beds of the Inferior Oolite resting upon the Upper Lias, against which are faulted aseries of the Lower Limestones and clays OF NORTH OXFORDSHIRE AND THE CLYPEUS-GRIT. 229 Fig. 1.—Vertical Diagram Section, Hook Norton. (Scale 1 : 144.) Great Ooteiw. on } | = HK. Sands and flaggy and andy Baredi ston limestones. ky Trigonia. Ostrea. Plant-bed. D. Marly and sandy limestone Quenstedtia oblita. Ainm. Parkinsont. Rhynchonella spinosa. Astarte minima. Trigonia producta. Corals. C. Shelly limestone and sand .. a——_ Oo NT + CO Or''V sandy limmestOwe: wv.tace-sscee---5-n| == Amm. leviusculus. SS | Lima punctuta. A. Blue limestone ................+. L Mipper Wits) Clay s.,.s2ccadvoscentene Fah? Gea Amm. fibulatus. of the Great Oolite. A trench at the end of the base of the northern embankment shows the Middle Lias rock-bed of the zone of Am- monites spinatus, with the characteristic Rhynchonella tetraédra and Terebratula punctata and Edwardsii. Upon entering the cutting, the Inferior Oolite beds are seen to be flexured and broken in their dip with the slope of the hill; and it is not until the upper part of the cutting is reached that the strata are sufficiently undisturbed to allow of measurement with any degree of accuracy. That the ten or twelve feet of blue clay, with numerous hard claystone nodules is that of the superior division of the Upper Lias, the Leda ovwn beds, is evidenced by the following fossils :— Ammonites fibulatus. Leda ovum. fonticulus. Thracia glabra. subplanatus. Pleuromya rotundata. bifrons. Inoceramus dubius. sp. Monotis substriata. Belemnites subaduncatus, Discina reflexa. vulgaris. Eryon. Cucullza elegans. Scales and bones of fish. The following is the section in the Oolites measured with the kind help of Mr. Windoes :— bo R ft. i _ ie pe Errata hs. caesar ge oe se Sas OR RRR ee 2 Blagey, white oolitic Limestone... oo. ic. ..ncew sess anos ee ences yee eee ee 2 Pisand weathering. White %..scuciscsesegper doccenh cece er 3 Hard cream-coloured limestone, bored in places at the base by MATMELIAS $.. iol. cs Rane teeeeenreee ceca Ge. eee eon t Brown sand \ with Ostrea calceola 1 Sandy limestone weathering rusty and Lima PUSS (cn cee cone Se See oh ences MENMM OMB a Mo Sandy limestone becoming loose at bottom .................2c0s00s 2 Damd. | waRaes lav. cst Recpe eeeeee eeeeeemas tet eae ene Sandy Limestome aed 135s. sae crant CoRR EC EE Ses nies od sk il . Coarse grittiy Sang. ....<.sceeusec ws er meebanemeree cs lees thistwss cco ee Sandy lamestoue. <0... savancenwsscamsesemneenatoe we nethe tad g acces eae sl Samide. .cscccabencos cnas-s05 teeemamehSweets Renee coer eRe eee er oe ee Violet-coloured limestone with plant-remains.....................008 i . Purplish tenacious clay, SaREY..c 2204005. 4.0.0... eae nde ee . Shelly Mimeshone .: sisi eerekcciae- nee eetuen aes vewsss ence eee . Bandy clay 5 . 2:0. -cca0cd tesabeastnumbansuscneeyessh sack. seca | 18. Hard limestone ..............csesceeeeseecessesseeeeeeeseceteereesessnnens 1 19. Brown sandy limestone, plant-bed: .......2..........000.000es eee | 20.. Purplish marly clay 3.602 va sessacacen casita} acer *K * ees limitata, Lamzx. Thamnastrea Defranciana, Mich. ca i Sps cd Giadophyllin sp. * Montlivaltia trochoides, E. ¢ H. Delabechii, E. & H. Stylina? Spiropora straminea, Phill. Diastopora, ¢f. margo-punctata, Waagen. oolitica, Vine. Tubulipora, sp. Apseudesia clypeata, Haime. Proboscina, cf. prorepens, Waagen. S Toone sp Stomatopora, cf. Waltoni, Haime. * From Prof. Judd’s list. Taste B.—Fossils from the Inferior Oolite of Hook Norton. SNA BOW! 65.740/c2i0.020 cee Probably young &. Lylli &e. [depressa, Like a flattened form of A. eel e - isk . |Casts. OF NORTH OXFORDSHIRE AND THE CLYPEUS-GRIT. 241 TasLE B (continued). Quenstedtia oblita, Pil. ......... Pholadomya Murchisonz, Sow. . faicular: Sow.) 1.5) tive wasted , large sp. Tancredia axiniformis, Phil....... donaciformis, Lyc. ......... Sowerbya elongata, Lyc. ......... Ss PRIGATCIA | 22..dscqccmesanenncars: cf. nuculiformis, Rém....... Lithodomus Porteri, Lyc.......... Ayicula clathrata, Lyc. ..........-. Miinsteri, Goldf. ..........+. inzequivalvis, Sow......:....+ braamburiensis, Sow. ...... Lima pectiniformis, Sow. ..... sha cardiiformis, M@. & L. ...... ovalis, Sow. bye) NAD le / Alt, al D7 ae ee ee duplicata, Sow. ............... punctata, Sow. .............6- Ostrea flabelloides, Lam. ......... rugosa, Goldf. CORTALA SGU sons dace cent crista-galli, Quenst. ......... Gj SOUSCURAL SOU! .. 2b... 5s cf. pyxiformis, Wright ...... Pecten articulatus, Schl. ......... TE SRDISO Maras. ah dotes aundee genuissus, PLU, o2322..0heced. clathratus;, O77. | i .6ci5..c0 personatus, Miinst. ......... Hinnites velatus, Goldf. ......... Gervillia acuta, Sow. ............ LOREUIOSH eaten Lo 8) ke Ps Mytilus Sowerbyanus, d@’Orb. .. IBN: sis ee cso RTE, eae Terebratula submaxillata, Mor.... maxillata, Sow. ..:.......: aun perovaliss Sai: 2v.. neta — plicata, Buckm.............64. ofp teilineata, Y. g B. ....2: Waldheimia carinata, var. Man- delslohi, Oppel: ss...0féM iv. c0s anelieas Oppel. 0 ccies.tse eve Rhynchonella quadriplicata, Zzet. —— subtetraédra, Dav. ......... — concinna?, Sow. ............ —— angulata, Sow. .......: wants — spinosa, Schlot. ............04 eynocephala, Mich. ......... DA SEO OEAN yet. v ot da duc dee vwasheuag Serpula socialis, Goldf............. puledtees Som sees eet convoltasGold fens. .s048 plicatilis, Goldf. EGNEACFINMUS. 1c .ucdnoter ss. case ees Isastreea serialis, EL. §& H. ......... Clausastrza Conybeari, EL. f H.... a A A A A - A A A bd bi: og: : bob QQQaQ Sieger ie Mel te) Weole (aevaelelolera: De ou (@\ Mi@iekee. a a Go Bo.8'C: Vee 7e D - jacutangula or bathonica. - |? C. nuciformis. . |Very large species. 242 . E, A. WALFORD ON THE ‘‘ NORTHAMPTON SAND ” TaBLE B (continued). Thamnastrea Defranciana, Mich. 5 SPDR Sos SAAC eo ee Tsastreea limitata, Lame. ......... Richardsoni, /. f A. ...... oO (2 eee | Wrighti. sp. approaching I. ——, sp. Montliyaltia, SPaaweces heceene meee: sch Bere Reris ele ier ET. oo sengeaeees A |B Spines of Echinoderms, 2 or BMSHIS RES. c~ ovulum, Ag. Slatter). Zieteni, Aq. ovoides, Sow. Heraulti, Ag. * Berenicea, sp., on Terebratula. , Sp. nov., large. Heteropora conifera, Blainv. Gresslya latirostris, Ag. Isastraa Richardsoni, Z. & H. peregrina, Phil. — serialis, LE. & H. Ceromya bajociana, d’ Orb. * Montlivaltia trochoides, H. & H. Cardium striatulum, Phill. , &. lens, E. & H. Cypricardia. Thamnastrea Terquemi, Z. g& H. Isocardia cordata, Buckm. * Thecosmilia gregaria, M‘Coy. * Cucullea oblonga, Sow. Clausastreea Conybeari, L. & H. * Lucina Wrightii, Oppel. Thamnastrea Defranciana, Mich. Trigonia costata, Sow. Serpula. ocd signata, Aq. Wood. —— pullus, Sow. Strophodus, producta, Lye. Astarte minima, Phill. Brodiei, Lye. Mytilus Sowerbyanus, @’ Orb. striata, Sow. Trochotoma, sp. Astarte, cf. depressa, Goldf. Trigonia angulata, Sow. Modiola Lonsdalei, M. & L. a ise Leckenbyi, MZ. & L. Gresslya abducta, Phill. Pinna cuneata, Phil. Astarte, sp. * , Sp. Myacites Goldfussi, Mor, Lima pectiniformis, Sow. Avicula, sp. . punctata, Sow. Serpula convoluta, Goldf. cardiiformis, WM. & L. Chemnitzia lineata, Sow. ; cast. * Gervillia Hartmanni, Goldf. Tancredia, cf. donaciformis, Lyc. * lata, Phil. Turbo hamptonensis, M. & L. Hinnites abjectus, Phil., sp. Monodonta. Those marked * are from Prof. Judd’s lists. OF NORTH OXFORDSHIRE AND THE CLYPEUS-GRIT. 243 Notes and Corrections (14th April, 1883). Some little confusion has been caused by the variable conditions affecting the lower beds A and B of the Hook-Norton series. Whilst some new exposures show them to rest unconformably upon the Upper Lias, attaining at the south end of the cutting a thickness of 5 feet 6 inches, and passing in places into a compact blue sandy limestone, at the north end, and in the adjacent Duckpool-Farm cutting, they are absent altogether. Both at Hook Norton and at Otley Hill Rhynchonella cynocephala is found at the top of A, together with Trigonia striata, T. Brodie, Terebratula trilineata, Pholadomya fidicula, Montlivaltia cf. lens, &e. It is probably the superior bed B which has yielded Ammonites Murchisone, its variety corrugata, and another form like Amm. variabilis. It would seem, therefore, as if we had in the lower part of the series an equivalent of the ‘“‘cynocephala” stage of Dr. Lycett, thus confirming my surmise as to the existence of lower Inferior-Oolite beds over the region, and perhaps, in part, representing also Professor Judd’s zone of Ammonites Murchisone. Reposing upon the series mentioned, and just now admirably shown at Otley Hill (vide Section, fig. 4, p.244), 1s a band of hard, com- pact crystalline limestone, curiously waterworn both above and below, and measuring but 1 foot 4 inches in thickness. The great erosion it has suffered prior to the deposition of the overlying marls is at once apparent; and it seems to be the dividing line between the lower and upper Inferior-Oolite divisions. The coral-bed at Hook _ Norton, which I had previously, in error, placed at the top of A, is its equivalent, and is not only filled with concretionary or derived fragments, but is bored and worn at the top. Large masses of Isastrea serialis occur in it, accompanied by Clausastrea Conybeari, Terebratula perovalis, &c., with Trigonia producta and T. angulata on its upper surface. Discussion. Mr. Hupieston bore testimony to the value of the work done by collectors in the neighbourhood of Banbury, both in the Lias and Inferior Oolite. The latter was especially difficult to interpret in North Oxfordshire, which was the border land between the South- western or Gloucestershire types and the North-eastern or Nor- thamptonshire types. He remarked on the confusion which had been produced by the use of the term ‘“‘ Northampton Sand” as ap- plied to beds in this district, since the real Northampton Sand re- presents the zones of Amm. opalinus and Amm. Murchisone, whilst at Chipping Norton its Oxfordshire namesake overlies the Clypeus- grit, which is in the zone of Amm. Parkinsoni. Apart from the question of names, the relations of the Clypeus-grit to the general mixture of sandy limestones was a puzzle to which Mr. Walford’s careful observations afforded a clue of great importance. Prof. Jurp stated that the intricacy in the geology of the district was the result of the rapid thinning-out in passing north-eastward Q.J.G.8. No. 154. S 244 ‘TTIH AWILO "alrysplofxgQ YpLOAT fo yyog soruafuy oy) fo spag wamory fo uoynpnuag anuysnyr 02 “UT hayQ pwUpn wor40Ay YOOH 20 SU01I0g—'H ‘Oy OF NORTH OXFORDSHIRE AND THE CLYPEUS-GRIT. 245 of different members of the Lower Oolites, and of the different conditions. which prevailed in closely adjoining areas. The great sandy series, which had unfortunately been correlated as a whole with the Northampton Sand of the Midlands, appeared to include representatives of both the Great and the Inferior Oolites. Mr. Erneriper thought it very difficult to correlate these beds of the Chipping-Norton area with those of the Cotteswold Hills. He remarked upon the remarkable attenuation of the Inferior Oolite when traced into the neighbourhood of Bridport. He considered the working-out of the minute geology of local areas would furnish a key to many difficult problems. Mr. Baverman stated that he had himself drawn the lines in the district represented in the author’s map, the work in the remainder of the sheet having been carried on by Mr. Polwhele. He had then assigned the beds to the Inferior Oolite. Mr. Watrorp stated that his map and the whole of his work had been based on the publications of the Geological Survey. Some artificial openings he had recently made confirmed his views on the succession of beds in the area. Even if some of the beds were equi- valent to the Fuller’s Earth, these are, as Professor Tate showed, referable to the Inferior rather than to the Great Oolite. 246 H. G. SEELEY ON THE DINOSAURS 17. On the Drvosavrs from the Maasrricnt Buns. By H. G. Szrtry, Esq., F.R.S., F.G.S., &c., Professor of Geography in King’s College, London. (Read March 7, 1883.) | In 1871 the Geological Department of the British Museum acquired. the celebrated collection formed by Professor van Breda at Haarlem. It was especially rich in the remains of fossil reptiles from Maastricht ; and among the bones of Mosasaurus were arranged five other speci- mens, which Mr. William Davies, F.G.S., soon recognized as Dino- saurian. So far as is at present known, these are the most recent evidences of the Dinosaurian order in geological time ; and in view of this fact, I am happy in having the permission of Dr. Henry Woodward, F.R.S., the Keeper of the Department, to offer the Geological Society some account of the structure of the last known survivors of the group. I avail myself the more readily of this permission, since I. do not remember to have seen in any of the continental museums other specimens exhibited which would add materially to the British- Museum evidence or modify my conclusions. | | These five bones belong to two types. One femur is Megalosaurian ; and although it is quite possible that other parts of the skeleton may enable their discoverer to refer the animal to a new genus, I have not felt justified in differentiating the genus from Megalosawrus on the evidence of one bone, imperfect distally, and with the proxi- malend worn. The other specimens are Iguanodont. I have re- ferred them to an Iguanodont genus Orthomerus; and I have no) doubt that the remainder of the skeleton will eventually show them to belong to a new generic type. For more certain reference, I give the British-Museum numbers on the specimens. Mreatosaurus Brepat, Seeley. No. 42997. This right femur is of Megalosaurian type. The bone is very imperfect ; the distal end has been sawn away (fig. 1, 4), so that the bone shows no indication of the distal articula- tion, though some changes in the form of the shaft suggest that no large portion is lost (fig. 1,8). The proximal end is a good deal worn and broken away (fig. 1,8); but enough remains to demon- strate its more remarkable characters. The femur, as a whole, is remarkable for its slender form, its superior bow-shape curvature, the lateral compression of the proximal articulation (fig. 1, B, 2), and the extent to which it is directed inward, for the proximal trochanter, which is separated from the proximal end of the bone in front (fig. 1, 4, 7), and for the proximal position and small size of the lateral trochanter (fig. 1, B and c, Jt). The fragment, as preserved, is nearly 29 cm.(11#in.)long. The shaft of the bone is unusually compressed from side to side, so as to make the vertical thickness (fig. 1, B) as much as or more than the width (fig. 1, 4); and while the surfaces of the bone generally are rounded, the superior or anterior aspect is marked by a blunt ridge OF THE MAASTRICHT BEDS. . 247 Fig. 1.—Right femur of Megalosaurus Bredai. (7 nat. size.) pie iy A. Anterior aspect. B. Inner aspect. C. Posterior aspect. h, articular head ; 7, proximal trochanter restored ; / 2, lateral trochanter. which becomes more angular as it ascends towards the proximal tro- chanter; but the ridge does not diverge outward much from the median line. At 7cm. (22 in.) from the proximal end the bone is 41 mm. (12 in.) thick and 39 mm.(14in.) wide. At 13 cm. (5{ in.) from the proximal end the thickness is 85 mm. (1? in.), and the width 32 mm.(13zin.). At the distal end the bone becomes more flattened, and widens a little; the thickness is 28 mm. (13in.), and the width 38 mm. (14 in.) ; the increase in distal width is chiefly a widening on the inner side, which becomes vertical. The external margin, though slightly concave, is nearly straight ; viewed from the out- side it has an aspect proximally of broad inflation, which becomes much reduced towards the distal end, chiefly owing to the increasing convexity of the inferior or posterior surface. But there is a slight elevation at about 8 cm. (3+in.) from the proximal end; and the external surface curves inward from this point as it extends proxi- mally, forming a sharp angle with the distal part of the external margin. The posterior surface in its distal half has a median longitudinal rounded ridge which fades away distally, and inclines a little towards the inner side of the bone. DAB H. G. SEELEY ON THE DINOSAURS The lateral trochanter (1 ¢, figs. B and c) is rather less than 5 cm. (2 in.) long, tapers proximally and distally, is compressed from above and below, and shows on the anterior side a small muscular scar, which deepens towards the distal border. The trochanter extends within about 74 cm. (3 in.) of the proximal end; it is directed a little inward and backward. At the upper limit of the trochanter the shaft is approximately triangular, being flattened below, while the inner and outer sides converge to the median anterior ridge. The head of the bone (A, fig. B) has the aspect of being bent inward, the external outline being strikingly convex, and the inner border con- cave, so that the transverse width of the head, as preserved, is not more than 63 cm. (22 in.). Externally and superiorly there was a trochanter, which formed an angular ridge; it was divided from the head of the bone by a vertical eroove, but must have been short, since the bone could not have extended. more than from 1 to 2 cm. proximally beyond the base of the groove which proves its existence. Internal to this trochanteric ridge the bone is compressed and con- cave. As preserved, the articular head of the bone is 2 cm. (+ in.) thick, and less than 3 cm. (11 in.) deep; but it is abraded, and there is no trace left of articular surface, or of the proximal cartilaginous tissue of the bone, which from within outward did not measure more than 6 cm. (2,4, in.), and now measures rather less; the pos- terior surface behind the articular head is somewhat inflated, so as to make the transverse section semicircular. Seen from the inner side the head of the bone is inclined towards the inner trochanter, so that it has a slightly oblique appearance as to its vertical direction (fig. B); but itis almost at right angles to the external surface of the shaft. The differences of this bone from the femur of Megalosawrus Bucklandi extend to almost every detail: first, the distal end of the bone is compressed from front to back, while in Buckland? it thickens ; the lateral internal trochanter is much more proximal in position ; the external trochanter is much much closer to the head of the bone in this specimen, and more proximal in position ; the curvature of the shaft is relatively greater, and its antero-posterior thickness is greater. OrtHomeRvs Dottot, Seeley. No. 42955. A large femur (fig. 2), which is slightly imperfect at its articular ends, measures 494 cm. (193 in.) in length. The shaft is remarkably straight and strong. The bone is subtriangular at the proximal end, is subquadrate but wider than thick in the lower part of the shaft, and has the lateral trochanter in the middle of the shaft, with the proximal and distal ends modified on the [guanodont plan. The distal end of the femur is fractured in front (fig. 2, B), where the condylar portion was probably a little expanded anteriorly, as in a second and smaller specimen (fig. 2, p). Across the condyles of the type the measurement is fully 10 cm. (4in.). The posterior inner condyle is considerably the larger; and the two are divided by a deep concave channel (fig. 2, 4). The external or anterior distal con- dyles were similarly divided by a concavity, so that the thickness of OF THE MAASTRICHT BEDS. 249 Fig. 2.—Right-femur of Orthomerus Dolloi. (4 nat. size.) Lj Swe eo” A. Posterior aspect : /Z, lateral trochanter. B. Inner lateral aspect: p ¢, proximal trochanter. C. Outline of proximal end, reversed. D. Outline of distal end of another specimen (B. M. No. 42957). | a, anterior, p, posterior condyles. bonedividing the anterior and posterior surfaces on the distal face of the articulation is about 3 cm. (11 in.).. External to the outer of the posterior condyles is a slight ridge, which is rounded and situate behind the middle of the outer side of the bone, so as to make the shorter posterior area markedly concave and to form a slight con- cavity anteriorly. This modification, which is limited to the condylar region, has the effect of giving the outer posterior condyle a compressed aspect, and makes the bone compressed posteriorly. Distally there is a moderate concavity between the condyles from within outward ; but the articular surface is imperfectly preserved, though the outer condyle appears to have had the greater distal extension. The depth of the inner condyle is about 6 cm. (2;4in.). The width across the condylar region, as preserved, is 10 cm. (4in.). The antero-posterior measurement cannot be given. Above the condyles the distal end of 250 H. G. SEELEY ON THE DINOSAURS the shaft is concave from side to side posteriorly ; the concavity, diminishing in amount, extends proximally towards the region of the lateral trochanter. The width of the shaft just above the con- dyles is under 9 cm. (34in.) ; and the median thickness of the shaft is under 5 cm. (2in.). The two sides of the shaft converge a little towards the base of the lateral trochanter; and the sides converge upward towards the anterior surface so as to give the front of the bone a convex or subcylindrical aspect in its middle third. Dis- tally, towards the condyles the front of the bone is gently concave ; but the concavity narrows and deepens rapidly to descend between the anterior expansions of the condyles, which are broken away. The lateral trochanter and muscular ridge (fig. 2, 4, B, 7 ¢) ex- tends to within 19 cm. (72in.) of the distal end, 1s nearly 14 em. (54 in.) long, and extends to within about 15 em. (6in.) of the proximal end. Itisa compressed curved process which is directed mainly backward and a little inward, and is much more developed in its distal half than in the proximal part. It owes its existence to two powerful muscular attachments, which are on the inner side of the bone; they partly overlap each other, so that the proximal scar descends partly in front of the distal impression. The proximal scar is about 8 cm. (3+ in.) long, and 3 cm. (14 in.) wide; less than half of its width is attached to the trochanteric process. The distal scar is quite as long and as wide, but is pointed proximally, rounded dis- tally, is much deeper, and is chiefly attached to the trochanter (fig. 2, Beet). ? Lee pear edge of the trochanter, which is inclined ob- liquely backward (fig. 2, ce is nearly parallel to the anterior borders of the muscular scars. The width of the shaft just below the tro- chanter is 6 cm. (2;41n.); its thickness in the same position is 5 em. (2 in.). Proximally, above the trochanter the form of the shaft alters, becoming compressed and well rounded on the inner surface, and greatly widened on the external border, so that the transverse section is subtriangular (fig. 2, c); the anterior surface is broadly concave, with the concavity increasing as the proximal trochanteric ridge is developed externally ; the posterior surface is flattened, with a moderate median longitudinal concavity ; and the external surface is flattened along its whole extent, but is a little convex from above dowuward, and has a broad shallow concavity behind the lateral tro- chanter. : No. 42957. A second specimen is smaller, and worth describing because it shows the form of the distal end (fig. 2, D). ' This fragment consists of the shaft and distal end of a Dinogaurian femur of moderate size. The fragment measures 30 cm. (114 in.) in length, and extends for about 3 cm. (i+ in.) beyond the internal lateral trochanter. The shaft is more quadrate in section than in the larger specimen, is more concave on the inner margin, has a slight convexity in length on the external border, and exhibits various minor details of str ucture. : The width of the shaft just below the lateral trochanter at 13 cm. (53 in.) from the distal end is 35 cm. (121n.); the thickness in the same position 1s Just over 4 cm. (14 4 in.). OF THE MAASTRICHT BEDS. 251 The extreme length of the lateral trochanter is 10} em. (43 1in.); it is directed backward only. It is narrow, being compressed from side to side, and is most elevated in the middle, while in the larger specimen the greatest elevation is below the middle. One muscular attachment extends along the whole of its inner border, tapering above and below; while proximally there is a vertically ovate im- pression, nearly 5 cm. (2 in.) long, which runs side by side with the proximal part of this impression. The sides of the shaft are remarkably parallel, flattened behind and on the external surface, rounded in front and on the internal surface, though the convexity decreases distally. On the external surface is a longitudinal median muscular scar about 5 cm. (2 in.) long; it is rugose in the middle; and extends to within about 114 cm. (43 in.) of the distal end. The outline of the distal end is like the letter H, owing to the way in which the anterior and posterior channels between the condyles cut into the bone (fig. 2, p). The inner condyle, as usual, is much the larger posteriorly (fig. 2, D, p), measuring 9 cm. (34in.) from front to back, while the outer condyle is only 8 cm. (3+ in.) from front to back, and it is much more compressed from side to side, especially proximally. The transverse measurement over the condyles is 74 cm. (3in.). Anteriorly the condyles are deeply channelled by a nearly circular canal (fig. 2, p, a) which descends obliquely downwards and back- ward and expands on the distal surface, so as to be broader than the posterior channel. The extreme width of the proximal expansion of the bone on the external surface, as preserved at the base of the trochanter, is under 9 cm.(34in.). The trochanteric process is subtriangular ; it is broken away, but its base is defined by a narrow groove extending backward. The proximal articular end is entirely broken away, though slight traces of its deep median concavity remain on the posterior border. On the external lateral aspect is a large rough surface, which is an ill-defined very shallow muscular attachment. It is about 9 em. (34in.) long, nearly as broad as the lateral surface, is more distal in posi- tion than the lateral trochanter on the opposite inner side of the bone, and extends to within about 17 cm. (6,4, in.) of the distal articular surface. Tn the main characters this form of femur closely resembles Zqgua- nodon; in nearly all points in which it differs, it approximates to Hadrosaurus. No. 42954. Left Tibia (fig. 3). This is along slender bone, which exhibits the distinctive characteristics of the tibia, although the arti- cular surfaces are gone from both ends, and the cnemial crest is entirely sawn away (fig. 3, A), so as to give the specimen somewhat the aspect ofa slender humerus. The bone, as preserved, is 27 cm. (102 in.) long. The proximal end is at right angles to the distal end (fig. 3c). On the whole the specimen shows the nearest resemblance to Iguanodon, but is much more slender, and shows some difference in form. The distal end (fig. 3, D), as preserved, is under 9 cm. (34 in.) wide ; Q.J.G.8. No. 154. T P52 H. G. SEELEY ON THE DINOSAURS Fig. 3.—Left nibia ai Orthomerus Dolloi. (4 nat. size.) iN Ci \ i i Hy W \y sl wi A D Ze A. Outer lateral aspect : ¢, cnemial crest. B. Outline of proximal end, reversed. C. Posterior aspect. D. Outline of distal end. the shaft widens distally in a wedge shape (fig. 3, c), is flattened in front much more than in Jguanodon, with a moderate median concavity for the ascending process of the astragalus, which was more than 2 cm. (4in.) wide. On the outer border the bone appears to have formed a slight ridge in the distal 5 cm. (2 in.), though the ridge is abraded and. lost. The external fibular area, which in Dinosaurs is usually well de- fined by a sharp angle, is here ill defined, about 2 cm. (4 in.) wide, and almost in the same plane with the remainder of the distal end of the bone. ‘The fibular margin is sharp, sharper than in Iguanodon, though, from the abrasion of the posterior surface, it appears to be sharper than it really is. The fibular side appears to extend out- ward more rapidly than the inner side, as usual; but the inner side does not widen rapidly, as in Iguanodon. At 8 em (31 in.) from the distal end the width of the bone is 6 em. (2;4 in.); at 10 cm. (4 in.) the width is 5 em. (2 in.), at 12 em. (4,4) a. is 4 cm. (12 im.). At 15 cm. (6in.) from the distal end the bone is 3 cm. (11 in.) from back to front and 3 cm. wide. The posterior aspect of the bone is marked by a rounded median ridge (fig. 3, c), which becomes narrower and less elevated distally ; two thirds of the bone lie on its fibular side, which is flattened and OF THE MAASTRICHT BEDS. 2538 compressed, and one third on the inner side, which is necessarily more oblique and rounded. Since the head of the bone is nearly at right angles to the distal end, it necessarily happens that there is no appreciable increase in width as the bone extends proximally. Buton the external surface the bone curves a little outward and forms a compressed area, convex from front to back, which terminates proximally in two small convex condylar surfaces (fig. 3,8). The internal aspect of the proximal surface is decayed, so that no account of it can be given. The cnemial crest (fig. 3,4, c) was evidently’ developed on the Iguanodont plan ; but its proximal portion has been sawn away. The anterior outline of the bone, as preserved, is concave ; the posterior outline is sigmoid (fig. 3, 4). The posterior surface shows at 12 em. (4,4 in.) from the proximal end a large vascular perfora- tion in the bone, which, as it rises proximally, becomes a groove i tlie. A), As compared with Jguanodon this specimen differs chiefly in being more slender and in some details of conformation of the distal end. But although the differences are suggestive of generic distinction, the condition of preservation does not admit of the enunciation of generic characters. On the other hand the form closely approxi- mates to Hadrosaurus, and is certainly intermediate between Jqua- nodon and that type*; and this combined with the characters of the femur indicate a divergence from Jguanodon in the same direction as in that bone, which justifies the association of the tibia, femur, and metatarsal bone. The metatarsal bone is too imperfect for descrip- tion, but it differs in form from any similar bone that has been figured. Discussion. The PresipENT remarked upon the interest attaching to these latest known of the Dinosaurs. Dr. Woopwarp referred to an Iguanodont vertebra in the British Museum which was dredged from the Dogger Bank. Possibly this was derived from the Maastricht beds, and Prof. Seeley might be able to associate it with the bones described in the paper. The Avrnor stated that many Iguanodont remains were found in the Crag and the Drift; but it would be unsafe to infer any possible relations to one another. * Tt may be also compared with the tibia referred to Megalosaurus by Prof. Owen (Rept. Weald, part iii., 1856), but will be seen to be of distinct type. Q,J3.G.8. No. 155. U 954 PROF, T, G. BONNEY ON HORNBLENDE-PICRITE 18, ApprrronaL Note on BovLprrs of HornBLENDE PicritTE near the Western Coast of AneLtEsEY. By T. G. Bonney, M.A., F.R.S., Sec. G.S., Professor of Geology in University College, London, and Fellow of St. John’s College, Cambridge. (Read April 25, 1883.) In a short communication printed in this Journal (vol. xxxvii. p. 137) I described a large boulder of hornblende picrite which I found near Pen-y-Carnisiog in the autumn of 1880. Last summer I had an opportunity of spending an afternoon upon the western coast of Anglesey, and made use of it to examine the interesting section to the south of Porth Nobla. I did not, however, neglect to look out for boulders, as I thought it not impossible that I might meet with some more picrite. In this I succeeded beyond my expectations, and now lay the results before the Society, together with some remarks on the microscopic structure of the specimens collected. After quitting the railway at Ty Croes Station, I walked a short distance along the road to the south-west, and then turned up a field-way leading past a small farm called Bryn Gwyn. A short quarter of a mile from it a boulder (No. I.), perhaps roughly trimmed, has been utilized as the capstone of a gate-post. It measures 23 x 2 x 1} feet*. A small fragment projected, which I was able to detach for examination without j injury to the stone. The next (No. IT.) was a well-rounded boulder lying on the sandy shore at Porth Nobla. This had a rather greener matrix than those — which I had previously seen: and the porphyritic hornblende or augite crystals were not quite so large. It measured about 2 feet each way, and rose about 1 foot above the sand. Near the shore, in the little cove Pen-y-Cnwe, is a boulder measuring about 43x 3x 2 feet (No. III.), and within a short distance a smaller one about 2 feet in longest diameter. At the cromlech Barclodiad-y- gawras, further south, two of the supporting stones on the western side are, I have no doubt, picrite of the ordinary type; but of course I did not touch these with my hammer. Three others are dark augitic (or hornblendic) gabbro-like rock, common in boulders in this part of Anglesey. On my return from Langwyfen to Ty Croes, by Frondwl, I noticed a small picrite boulder built into a wall by the road-side; and then (a little over a mile from the station, and perhaps 300 yards from a chapel) I found no less than seven fragments of picrite, five of them being built into a rough wall, and one lying on either side of it (No. IV.) They ranged from about 2 to 4 feet in longest dia- meter ; but the time at my disposal did not allow me to go into details. I should suppose they were fragments of one boulder ; if so, it must have been even larger than that which I saw at Pen-y-Carnisiog. * Measurements throughout are only approximate. NEAR THE WESTERN COAST OF ANGLESEY. 255 I have examined microscopically specimens from four of the above localities. Numbers III. and IV. are, macroscopically, almost undis- tinguishable from the Pen-y-Carnisiog specimens ; nor is there any material difference under the microscope. The larger hornblende cry- stals in III. are commonly a light brown, but occasionally a pale green. The smaller crystals are more commonly green. Small portions of some of the larger crystals, sometimes external, sometimes internal, are almost colourless, and have a rather more milky aspect than the rest of the crystal. With crossed nicols they exhibit a different tint, but extinguish either at or as nearly as possible at the same angle. Still,:even then, on introducing a quartz plate, a marked difference of tint 1s perceptible. Whether this change denotes an hydration of the hornblende or a paragenesis of two slightly dif- ferent varieties, I cannot say. Cracks in the crystals are often filled, and the edges bordered, with a minutely granular mineral, giving light specks of colour with crossing nicols. In parts of the slide are many small grains and clusters of a mineral now consisting of aggregates of this secondary mineral and of earthy-looking dust. These may possibly have been a rather aluminous augite; this mineral, however, as in the case of the Pen-y-Carnisiog rock, cannot be certainly identified in the slide. There are numerous grains of a dark brown mineral, in some cases feebly translucent, in others including granules of a clear light-brown mineral ; a few appear to be sections of octahedra. They present a resemblance to chromite ; but, as no chromium has been detected on analysis, this mineral can hardly be present ; possibly they are spinel. For the various pseudo- morphic products occupying the rounded grains in the larger crystals, and the general ground-mass of the slide, I may refer to my former paper. I do not identify mica. It is just possible that the slide may have contained a crystal or two of felspar. No. LY. differs so little from the last that a separate description is needless; there can, I think, be no doubt that some of the serpen- tinous aggregates replace olivine. One exhibits an approach to aggregate polarization. There is a little apatite, and a few scales of brown mica. No. II. presents, macroscopically, some slight varietal differences : the ground-mass is greener; and the imbedded hornblende crystals are not quite so abundant or large as in the other cases, being commonly from 0:2 to 0:3 inch in diameter. But the microscopic structure has a close general resemblance to that above described, though it contains a few grains of a serpentinous mineral which I have not observed in the others. This has arather irregular outline, is nearly colourless, is of a somewhat silky structure, with rather in- frequent cleavage-planes parallel with the fibres of this structure ; it extinguishes when they coincide with the vibration-plane of either of the crossed nicols. Numerous minute belonites, slightly browner in colour, occur in the grains, lying often so as to cross one another at angles of about 120°, and to be very nearly bisected by the cleavage- planes. Ihave already noticed this microlithic structure, sometimes with minute rods of opacite, in one of the minerals of various serpen- 256 PROF. T. G. BONNEY ON HORNBLENDE-PICRITE tinous rocks, and think that probably we have here an altered ensta- tite*. The slide, however, does not exhibit in any part the peculiar brassy lustre of bastite. No. I., though in many respects agreeing with the first described, has one or two varietal differences. There is undoubtedly a little plagioclastic felspar, rather decomposed, with a considerable amount of an almost colourless mineral, containing fine earthy granules, acting rather feebly on polarized light, and without a very charac- teristic cleavage, which I am disposed to regard as akin to augite, Some of the brown hornblende crystals seem to pass through a narrow green border into a closely cleaved colourless mineral lke a diallage, with a larger extinction-angle than that of hornblende. Indeed, more than one specimen in the above-described rocks, though strongly dichroic, more resembles diallage in aspect, and has an extinction-angle too large for hornblende. The Schriesheim rock, said to contain diallage, gives a similar discrepancy. I am disposed to explain the apparently contradictory results afforded by these rocks by supposing that the mineral originally predominating in the picrite was a pyroxene, and that we find it now in various stages of conversion into hornblende. I am indebted to the kindness of my friend Mr. J. A. Phillips, F.R.S., for a duplicate analysis of the rock collected on the road to Ty Croes station (No. IV.), which I subjoin :— I, II. hygrometric ... Je. 0°60 0°53 Watety (aplitined |) Bane 2-84 2-90 alias 1 Si eRe Oe es one i les 42°94. 42:79 PAINS Chet ee ap ete Conte een c 10°87 10:98 Carbonic anhydride .......... 2°65 2°76 Phosphoric anhydride ........ trace trace WEG ORIG: Waste. tee tt ome 347 3°40 MerrousiOxetde. Go ine ees ee a te 10°14 10°13 Titanic oxide.) .o.). coe oe trace trace Manganous oxide 2)... 53.%: trace trace MepHIE -ehre aoe wees c Pee ea ee ee 9:07 9a Mine S ta ae eye eee ee 16°32 16°22 PGtASaE ee ee ee, eos nee Oa 0-10 ro 10 Sa ge ME TW ee Hates 0:93 99°95 99-89 SPCC CT AVIEN < a+ aii wile en see 2°88 2°88 Nickel and chromium were sought for but not found. For comparison, I append a series of analyses,of picrite. The first five are given by Tschermak (Sitzungsb. Akad. der Wis- sensch. Wien, vol. liii. pt. 1, p. 260). The next two are ‘ paleopicrite’ * A similar structure, but with the rods more nearly at right angles, is figured as occurring in magma-basalts by Boricky, ‘ Basaltgesteine Bohmens,’ pl. i. Be rte | 257 NEAR THE WESTERN COAST OF ANGLESEY. SiO,. Sohle, near Neutitschein ...............] 388°9 Freiberg and Gumbelberg ............| 40°79 Nichia imeces: percdtcs satecaaceeterswedlh “OO kat Sohle (looking rather decomposed) ..| 42°85 IDVSUL VO netics Overivagsea¥ss cdvereantouel TOOWL Schwarzenstein, N.H. of Hof .........| 37°12 Holler, near Lichtenberg...............| 41°48 Ottenschlag, S. Austria ...............] 45°93 Schriesheim an der Bergstrasse ......| 41:44 Fe,0,. | FeO. 4-9 7-0 3:52 | 6:39 6:30 | 6-14 627 | 6:86 275 | 7-62 8:92 | 762 626 | 8-46 1:87 | 11-45 13:87 | 6:30 CaO. 6:0 8:48 10°37 11°84 13°61 6:14 6°55 8°92 7:20 MgO. 23'6 23°34 18°59 9:01 7°28 26°92 25°27 14°82 18°42 K,O. |Na,0.| H,0. 0:8 0°71 1:57 161 181 0-49 0:09 0°22 0:93 1:3 171 1:50 1:65 0°59 0:40 0:29 1:93 0:24 * Including TiO0,=0'4, MnO=0'4, PO,=0'10, and trace of chromium oxide. t Including Ti0,=0'18, MnO =0'37, PO, =0:17, and trace of chromium oxide, CO,,. eereee Total. 99°1 99°39 100:27 99:09 98°70 98:60 * 97°47 t 100°81 100°63 | 258 PROF. T. G, BONNEY ON HORNBLENDE-PICRITE from Gumbel, ‘Geognostiche Beschreibung des Fichtelgebirges,’ p. 152; then ‘palexopicrite’ analyzed by Alois Gamroth, ‘ Jahrbuch k.-k. Reichsanst.’ 1877, Min. Mitth. p. 278; the last, the Schriesheim rock, analyzed by Fuchs, ‘ N. Jahrbuch,’ 1864, p. 326. By a comparison of these analyses with those made by Mr. J. A. Phillips, we see that the latter have shghtly more silica, considerably more alumina, and less magnesia than the two rocks described by Prof. Giimbel, the composition of which more nearly corresponds (except that there is less water) with that of a normal serpentine, from which they differ chiefly in a lower percentage of magnesia and a higher one of lime. Even these, however, differ in the ratio of the silica to the magnesia, which in the Schwarzenstein rock is about 13:10, in the Holler 16:10, while in such a rock as the serpentine of the Lizard, Cornwall, it is roughly 11:10. In the first rock from Sohle it is about 15:10. In the Welsh rock it is rather more than 26:10; but then this ratio is exceeded by two of Prof. Tschermak’s specimens, while in the Schriesheim rock it is 22°5: 10. This last has also been called a ‘“ Schillerfels” and an olivine- diallage rock; but the conspicuous mineral is not bastite, and my specimen contains little, if any, normal diallage. For comparison I append one or two analyses of serpentines and of olivine rock. I II 1M) LV SOs carieeatre cues ebanwencee ones 38°86 38°58 38°90 43°84 INS O sim srs lehs ent uascianeenss tee 2°95 3°06 1-02 1:14 HOMO aon och cntaae ns namntws 1:86 1:95 4:66- | (ae TREY C ARM acetal crea: (Abit. aN 5:04 5:10 Sol 8°76 iO) ee a eke trace trace dal Seca 0:12 iN ir 6 JM AEE aL ofa no es aS Sg 0:28 0:30 0:59 0°51 WO Oe Scag Sroheen wares eeepene: 0:08 OLO8: Ae es 0-42 EE 0 Meee Cann dae Mw A AC CAR trace trace 1:97 ea INGO een cate ue memes 34-61 34°32 36°40 44°33 TGC Beers aes 0°33 0:30 1-37 Na Oi ceed scp tede delec star. coe O77 O76 of oes 2 ELS) ain eee cee ee cnet 15°52 15:52 13-35 1:06 GS RR ete Ne Oe ain en oe ele tae Vee MeN a 0-41 100:30 99°97 100-58 101°89 I. & II. Duplicate analyses of a Lizard serpentine by Mr. J. A. Phillips (Phil. Mag. Feb. 1871). III. Analysis of a serpentine from near Cadgwith by Mr. Hudleston (Q. J. G. 8. vol. xxxiii. p. 925). IV. Olivine om rene Paul’s Island (Renard, ‘Challenger Voyage,’ Narrative, volva pet). In short, an examination of the former analyses shows that the chemical composition of the rock picrite is rather variable, and that, while it lies closer to the normal peridotites than to any other, it is, to NEAR THE WESTERN COAST OF ANGLESEY. 259 some extent, intermediate between them and the olivine-diabases*, that it is, in fact, (or has been) an augite- (or hornblende-) olivine rock, rather than an olivine-augite (with or without enstatitet), and that felspar is a common accessory mineral. Consideration, however, of these analyses, and of the structure of the Anglesey rock, both macroscopic and microscopic, fully justifies us in retaining it among the picrites or paleopicrites, if the latter term be preferred. It is evident from the above observations that this rock is hkely to be commonly met with in Anglesey ; and I have described it with minuteness in the hope that geologists will be on the watch for its occurrence. J have not yet been able to hear of its being found in situ in the Welsh area; but at the end of last year my friend Mr. Teall called my attention to a rock which he had collected at Little Knot, on the east side of Bassenthwaite Lake, which bore, macro- scopically, a considerable resemblance to my specimens. This rock is described as a diorite by the late Mr. Clifton Ward, in the Memoirs of the Geological Survey (Lake District), and a figure of its microscopic structure is given. He remarks on the abundance of hornblende, and the “very little felspar,” but does not seem to have suspected that olivine had been present, although, from the figure, I have little doubt that such was the case. A slide cut from Mr. Teall’s specimen exhibits well-marked crystals of hornblende, both green and brown, with rounded serpentinous interruptions, and the same pale edging as is seen in some of the Welsh specimens; there is much fibrous actinolite in the body of the slide, and various patches with the peculiar belonites crossing one another, as in some bastite. In many of them are minute rounded specks of granulated aspect, dustier at the edges, and giving with crossed nicols a pale whitish colour. There is a little apatite, a decomposed iron oxide, probably ilmenite, and one or two crystals of felspar, so much decomposed as hardly to be recognizable. The specific gravity of this rock is 2°93. A partial analysis (duplicate), kindly made for Mr. Teall by Mr. EK. K. Stock, gives $10,=46°37 and 46-42, MgO=15°73 and 15:30%. These percentages are not in very exact agreement with -those in the Welsh specimen (Si0,=42:94 or 42°79, and MgO=16°32 or 16°22); but the rock is evi- dently a variable one, and in mineral composition, macroscopi- cally and microscopically, there is a very near resemblance. It * These have, as a rule. a higher percentage of SiO, and still more of Al,O,. The analyses of troktolite, especially of that from Volpersdorf (see Houghton, Geol. Mag. dec. ii. vol. vi. p. 504), have a closer correspondence with that of this Welsh picrite; but troktolite is at once distinguished by its povertyin a pyroxenic or hornblendic mineral. In it also felspar is conspicuous, while in the picrite it is rare. + In some of these, 7. e. the peridotites (and the serpentines resulting from their alteration), the percentage of magnesia somewhat exceeds that of silica ; this would be the case where olivine was far the most abundant mineral. An inerease in the amount of the enstatite would correspond with a rise in the SiO, : an increase in the augite would be indicated by a marked percentage of CaO. t 1 have since received, through Mr. Teall, other specimens from Mr. A. Bloomfield, of Keswick. He states that there are seven or eight outcrops of the rock, one of which is quarried. The rock is evidently very variable in character. 260 PROF. T. G. BONNEY ON HORNBLENDE-PICRITE. is therefore possible that these Anglesey boulders have not been derived from that island or North Wales, as I had anticipated, but from the Bassenthwaite district. In admitting the possibility of this, I should not feel bound to regard them as proofs of the former extension of Cumberland glaciers to the Welsh area, but should con- sider them, like the Criffel and other northern boulders in North Wales and the Midlands, to have been transported by floating ice during the last great submergence. Discussion. Mr. Hupzreston thanked the author for his definition of the exact nature of picrite, and remarked on the high alumina per- centage of the analysis as being inconsistent with the composition of picrite. He thought the original home of these boulders was not the Lake District, but Anglesey, as suggested by the author in his first paper. Mr. Txatt said that the Inchcolm picrite contains a very variable quantity of felspar. He thought the presence of felspar in greater proportion could therefore scarcely be regarded as sufficient to differ- entiate the Lake-District rock from that of Anglesey. He cited a case of the occurrence of a boulder of faulted slate, evidently from the Lake District, in North Wales. Mr. Dz Rance cited examples of this rock from several localities in Flint, Denbigh, and Caernarvonshire, in boulder-clays, and asso- ciated with characteristic Lake-District rocks. The faulted-slate rocks of the Lake District were perfectly distinct in colour and other characters from those of North Wales. The Avurnor, in reply to Mr. Hudleston, stated that picrite was a very variable rock. He had difficulty in accounting for the high percentage of alumina; forasa rule he had failed to detect felspar. The Little-Knot rock is quite an exceptional one, as stated by Mr. Clifton Ward. He knew of no example of rocks of the kind in Anglesey. ON THE SUPPOSED PRE-CAMDRIAN ROCKS OF ST, DAVID’S. 261 19. On the Supposep Pre-Camprian Rocks of Sr. Davin’s. By ARCHIBALD GEIKIE, Esq., F.R.S., F.G.8., Director-General of the Geological Survey. (Read March 21 and April 11, 1883.) [Puates VITI.-X.] ConTENTs. page. Mae MRO RIE ee a se cel, cole c dade Be eC 261 Part I em STYEC TEENS sc ScR ce Ras wiiciaia gis agi on UAC aR EO . 267 ME DAR RNAI 0) coe pec y as cove cmcadueiceakes: noc conecateneed PER nee 280 Se ECT hi ria Ope yah Meant ee BAS gee a Da ey Oa 283 4. Relation of “ Pebidian” to Cambrian Rocks................eececees 286 3, (QUIET) ieee aoe RE eer an ee ee eee Ok Per ESE ate OP 291 Part II. faOrder of Succession“ of the) Rocks) sic. .scvckh asobadens laedosss@aaeen 294 me Geolocical: Structure-of the District: .1.j.s.:.s00cecnesectdescasstene 308 peotne Bolabion Of the: Disbrich. 2.25. oSence LieseScuccdcSecdssnevceoanee 310 4, The Granite, Quartz-porphyry, and accompanying Metamor- (DULSTI sea. onc dReeBnsen nae eH Me enae ine eee eee mah! ime aanpiiter ecmmeey 3138 5. The Diabase Dykes and Intrusive Sheets ..............scscseseeeees 323 MME PUSHOER, 5555 te samo a reise te Sea ce oto es Se Ae A ce ham 324 INTRODUCTION. Some explanation is required of the grounds on which another contribution is added to the already voluminous literature connected with the rocks of St. David’s. The circumstances under which I have reluctantly found myself compelled to make this contribution are briefly these. A new edition of the Rock Catalogue of the Museum of Practical Geology, Jermyn Street, being required, it became necessary to ascertain how far the stratigraphical order followed in previous editions required to be modified by the results of more recent research. In particular, the question of the existence of Pre-Cam- brian rocks in Wales, which has emerged since the last issue of the Catalogue, appeared to demand full recognition in any republication of the work. I felt it to be due to those authors who have written so largely on this subject that ample acknowledgment of the results of their labours should be given. On the other hand, I was equally . desirous that in admitting corrections of the views expressed upon the maps and sections of the Geological Survey I should do so from an actual inspection of the ground, which would enable me to judge how far and in what manner the required alterations should be made. It was obvious that in maps of districts surveyed forty years ago some important structures might have escaped notice, positive errors might have been committed, and petrographical details could not be expected to have been treated in a more satisfactory manner than in other English geological works of the same date. The existence of Pre-Cambrian rocks in South Wales had been Q.J.G.8. No. 155 . 262 A. GEIKIE ON THE SUPPOSED proclaimed so loudly and persistently that, m spite of the protests of my predecessor, Sir Andrew C. Ramsay, who will not admit the presence of such rocks in any part of the Principality, I had gradually been led to believe that they really must exist, though probably not to the extent that had been claimed for them. In visiting Wales, therefore, I went with no prejudice in favour of the views expressed by the Geological Survey. On the contrary, I had a conviction that these views must be, in some measure, at least, erroneous, and that this admission ought to be frankly made. I chose the St. David’s district as being that about which most had been written, and which had, in a measure, been taken as a typical area for the ‘“‘ Pre-Cambrian” rocks of Wales. It is desirable at the outset that it should be clearly understood that the conclusions to which I have come refer solely to that district, and that, in the meantime, I offer no opinion regarding other so-called Pre-Cambrian areas in the Principality. That my examination of the ground might be made in greater detail, I requested my colleague Mr. B. N. Peach, of the Geological Survey of Scotland, to accompany me. His long experience among crystalline rocks of many kinds has given him great practical - insight into the structure of these formations in the field. Like myself, he went prepossessed in favour of ‘‘ Pre-Cambrian ” ideas. We visited all the sections together, and came to complete agreement in our interpretation of them. The following pages give an account of our joint research in the field, and of my own subse- quent petrographical investigation of the rocks collected by us*. The earliest published account of the rocks of St. David’s appears. to have been that given by Dr. Kidd, in vol. ii. of the First Series of the Society’s ‘Transactions.’ ‘This author speaks of the rocks as being in some instances “ stratified,” in others ‘* unstratified ;” the hills, or rocky summits, consisting of materials that “bear no marks of regular stratification,” but “‘ appear as so many nuclei, about which is arranged a very curiously diversified series of highly-inclined strata of a kind of slate”. The next notice is that of Sir Henry De la Beche, in a paper read to this Society in 1823. He separates the “trap” and “ grey- wacke ‘rocks of St. David’s, and is inclined to regard the trap as ‘‘having been forcibly intruded amongst the other rocks at a period subsequent to their consolidation”t. ‘The areas respectively * Since this was written I have made a second visit to St. David’s, accom- panied by my colleague Mr. W. Topley, of the Geological Survey of England and Wales, with the object of collecting additional material for the second part of the present paper. I have thus been able fully to confirm the conclusions arrived at on the first occasion, and to obtain some additional evidence in the same direction. But nearly the whole of the data bearing upon the question of the existence of Pre-Cambrian rocks were collected in company with Mr. Peach; and the first part of the paper had been prepared before my second excursion. +t Trans. Geol. Soc. ser. 1, vol. ii. p. 79 (1814). } Trans. Geol. Soe. ser. 2, vol. ii. p. 2 (note). PRE-CAMBRIAN ROCKS OF ST. DAVID’S. 963 occupied by the two kinds of rock are depicted on the geological map of South Pembrokeshire accompanying this memoir. In the ‘ Silurian System ’ (1839) Sir Roderick Murchison inserted a brief description of the trap-rocks of Pembrokeshire, in which he spoke of them as consisting of two classes, “ viz., stratified masses alternating with sedimentary deposits, and amorphous masses which have burst through the strata.” Among the older trap-rocks he distinguished “ thick-bedded coarse felspathic conglomerates, contain- ing fragments of schist and slate, which range from north-east to south-west in allinement with the other ridges of amorphous trap.” These words would seem to imply that he had recognized the presence of true tuffs or fragmental igneous rocks in that region. He also noticed proofs of the intrusive character of some of the amorphous masses ; for in the district of Roche Castle, Trefgarn, and Ambleston he found that ‘‘ the intrusion of this [igneous] rock has produced a powerful effect upon the adjacent strata, particularly on those masses inclosed between the forks of trap in the gorge of the river at Trafgarn, where the red and green sandstones are converted into a brittle, siliceous substance resembling the ribbon jasper of mineralogists ” (pp. 401, 402)*. In the first volume of the ‘ Memoirs of the Geological Survey of Great Britain,’ De la Beche makes further reference to the geology of Pembrokeshire, and in particular to some of the rocks which have been the subject of more recent discussion. He refers to the rock of Roche and adjoining districts as illustrating the remarkable varieties of texture assumed by the same mass of igneous rock; and he cites the granite between St. Lawrence and Brawdy, a few miles to the east of St. David’s, as presenting along its margin a fine-grained variety, like a Cornish elvan, owing to more rapid cooling, and as “certainly seeming to have altered the stratified rocks in contact with it in many places’. The Geological Survey of the St. David’s district was begun as far back as the year 1842, by Sir A. C. Ramsay, soon after he joined the service. The Map was published in 1845, and was fol- lowed by a sheet of Horizontal Sections across the district t. In these publications all the igneous rocks are included in one colour (green). But in the Horizontal Section the belt of country so coloured, extending (on the map) from the sea at Porth-lisky to beyond Lianhowell, is stated to be composed of “ trap, in its strike of various structure,—syenitic greenstone and felspathic volcanic ash”$. * The intrusive nature of the Trefgarn rocks was shown by Murchison in 1836 (Proc. Geol. Soc. vol. ii. p. 229). These rocks are included by Dr. Hicks in his “ Arvonian” group. t+ Mem. Geol. Survey. vol. i. p. 230 (1846). The belt of rocks here referred to has been mapped by Dr. Hicks as belonging mainly to his ‘‘ Dimetian,” but partly to his ‘‘ Arvonian” group. Quart. Journ. Geol. Soe. vol. xxxv. p. 287 (1879). { Sheet 40 of the Geological Survey Map of England and Wales, and Sheet 1 of the Horizontal Sections. § Sir A. C. Ramsay’s field-maps, preserved among the official records of the Survey, show that he not only recognized marked differences among the igneous rocks, but that he mapped out the more important varieties. A MS. report Me - 264 A. GEIKIE ON THE SUPPOSED - A second (revised) edition of the Map was published in 1857, and was soon followed by a second edition of the Sections. The primary object in the revision of the work was the tracing of a line for the base of the Lingula-flags; but the opportunity was also used for separating some of the more important varieties among the igneous rocks. These were now classed as ‘“‘syenite” and “ greenstone,” all mention of the “ volcanic ash” of the previous edition being omitted from the St. David’s area. A considerable tract to the west of St. David's, which had been mapped by Ramsay as igneous rock, and which was regarded by him as composed mainly of volcanic ash, was now coloured as “‘ altered Cambrian”. As thus revised, the Map showed a long strip of syenite and fel- stone, stretching from the sea through St. David’s in a north-easterly direction for about seven miles, flanked on the south-east by Cam- brian strata, and on the north-west by “ altered Cambrian ” rocks through which long parallel sheets of greenstone had been erupted. The existence of abundant contemporaneous igneous rocks further north and east is clearly shown on the map; but in the immediate drawn up by him at the time, but never published, has fortunately been also preserved. From this interesting document a few extracts may here be appro- priate :—‘The igneous rocks of North Pembrokeshire are both intrusive and con- temporaneous ; the latter, however, occupy by far the greater area. The greatest intrusive mass is that which from Ramsay Sound stretches in a north-easterly direction nearly eleven miles. The best evidence of this trap being intrusive may be seen on the coast near St. David's, at Ogof-llesugn, where it cuts through the strata...at right angles to the line of strike. “The apparent composition of this large mass is very various, resulting, no doubt, in many instances from the different circumstances under which it cooled and consolidated. Thus, on the east side of Porth Lisky, and on both sides of the Allan at St. David’s, it is a large-grained, coarsely crystallized syenite, hornblende being comparatively sparingly diffused throughout. At Porth Lisky granular crystals of quartz are largely developed. In other places the same development may be seen. “The coast of Porth Lisky is almost entirely composed of volcanic ash, which, being of a softer texture than the traps on the H. and W. sides of the bay, has yielded to the action of the waves, and thus formed a little harbour. ... From hence, in the direction of Ramsay Sound, the rocks are composed of hard sili- ceous trap [and] various greenstones. ... These frequently alternate with partial layers of volcanic ash. These varieties constitute the rest of the mass to the eastward, without any very apparent order in the manner of their distribution. ...Near Trelethin (St. David’s) and elsewhere there is a little volcanic conglo- merate. With the exception already mentioned, the strike of the strata generally more or less conforms to the general run of the intrusive mass. ‘“‘ A glance at the contemporaneous traps shows that voleanic agencies had been in operation for long successive periods..... From the intermixture of voleanic ash and greenstone at Porth Lisky, it would appear that the forma- tion of this mass was in some measure the work of successive eruptions.” * Professor Ramsay, as Local Director of the Survey; agreed to this change, which was made on the ground by Mr. Aveline. In many respects the map was a marked improvement upon the first edition ; but the suppression of the refer- ence to fragmental volcanic rocks and the introduction of the term “altered Cambrian” were unfortunate changes, one effect of which has been to obscure’ the fact that to Ramsay belongs the merit of having first clearly recognized the presence of truly contemporaneous fragmental volcanic rocks in a formation of such high antiquity. p Sih PRE-CAMBRIAN ROCKS OF ST. DAVID’S. 265 vicinity of St. David’s all the rocks of igneous origin are repre- sented as intrusive. | Until the year 1864 this interpretation of the structure of the district appears to have remained unchallenged. At the meeting of the British Association in that year, however, a suggestion was thrown out by the late J. W. Salter that the syenitic belt of the Survey was a portion of Pre-Cambrian land*. The only grounds given for this suggestion were that the rock is syenitic, and that it does not penetrate the overlying Cambrian strata. Next year Mr. Salter, acknowledging himself to have been mistaken, stated that Dr. Hicks had found portions of schist entangled in the syenite, as well as altered strata on the south side of the latter.rock?. In the year 1871 appeared a joint paper by the late Professor Harkness*and Dr. Hicks on the “ Ancient Rocks of St. David’s Promontory ’Z. A foot-note in this paper contains an announcement by Dr. Hicks that he had subsequently found on the ridge of St. David’s evidence of bedding in its component rocks, and that, as the strike is discordant to that of the Cambrian strata, there must be here a more ancient group of rocks than the Cambrian, occupying a position equivalent to that of the Laurentian group of Canada. ' In 1875 Dr. Hicks asserts more confidently the Pre-Cambrian age of the rocks of this ridge, denies that these rocks are syenite as coloured by the Geological Survey, but maintains that they are bedded rocks—quartz-conglomerates and dark-green shales, partly meta- morphosed,—and affirms that they are covered unconformably by the Cambrian series §. Returning to the subject two years later, Dr. Hicks showed a still wider divergence from his original opinion. He now states that he call recognize two distinct series of Pre-Cambrian rocks at St. David’s, giving the name ‘“ Dimetian” to what he supposes to be the older, and “ Pebidian” to the younger series. Discarding the identification of any part of these rocks with syenite, he describes the ‘ Dimetian” as composed chiefly of compact quartz-schists, chloritic schists, and indurated shales, and the ‘“‘ Pebidian ”’ as con- sisting mainly of indurated shales, often porcellanitic in character. He regards the “‘ Pebidian” as resting unconformably upon and partly derived from the waste of the “ Dimetian” rocks. The Cambrian beds are stated to lie unconformably on both these series and to contain abundant fragments of them||. A little later Dr. * Rep. Brit. Assoc. for 1864, Sections, p. 64; Geol. Mag. vol. i. p. 289. The Rev. W. S. Symonds, however, claims to have first suggested to Salter a Pre-Cambrian origin for the St. David’s rock. See his ‘ Records of the Rocks,’ . 31, 1872. + Geol. Mag. vol. ii. p. 430. ¢ Quart. Journ. Geol. Soc. vol. xxvii. p. 384. For the introduction of bedding into the crystalline rocks of the ridge, as expressed in this paper, Professor Harkness does not appear to have been directly responsible. See footnote on p. 387 above referred to. § Quart. Journ. Geol. Soc. vol. xxxi. p. 167. || Quart. Journ, Geol. Soc. vol. xxxiii. p. 229 (1877). 266 . A. GEIKIE ON THE SUPPOSED ‘Hicks recognizes that his ‘‘ Pebidian”’ series is in great part made up of volcanic tufts and agglomerates*. r In 1878 the late Mr. E. B. Tawney, who, in company with Professor Hughes and Mr. Hudleston, went over the St. David’s district under Dr. Hicks’s personal guidance, gave an interesting and valuable description of the rocks, to which I shall have occasion to make repeated reference in the sequel. In this paper he accepts generally Dr. Hicks’s conclusions, but, though classing the crystalline rock of the axis as metamorphic, confesses that “over a portion of the area, at any rate, it did not show evident bedding enough to prevent our classing if as a massive crystalline rock”. In the same year, Dr. Hicks announces the discovery of what he terms a new group of Pre-Cambrian rocks, named by him ‘“ Arvo- nian ”t, consisting of “breccias, halleflintas, and quartz-felsites,” previously included by him in his “ Dimetian and Pebidian,”’ but now regarded as intermediate between them, and unconformable ‘with both. In later papers he summarizes the results of his re- searches§. During the progress of Dr. Hicks’s researches, the rocks of St. David’s have been referred to by other writers, sometimes on his authority, sometimes from personal observation under his guidance. It is not necessary to augment this bibliographical outline by citing all such references. Some of them are quoted in subsequent pages, where also several of Dr. Hicks’s own papers are dealt with more in detail. The object of the present communication is twofold,—first to discuss the evidence for the assertion that Pre-Cambrian rocks exist at St. David’s, and secondly to lay before the Society an outline of . what appears to me to be the true structure and geological history of that district. Parr I. The first part of the paper is unavoidably controversial. I pro- pose to examine the evidence for the alleged presence of Pre-Cam- brian rocks at St. David’s, and to state the facts which, when brought to the notice of geologists, will, I think, be admitted completely to disprove the existence of any such rocks at that locality. Disliking controversy so much as I do, it is with extreme reluctance that I now enter upon it. But Iam sure that, in the interest of truth, Dr. Hicks himself, whose published views I must oppose, will be glad that these views should be subjected to the most searching criticism. It is due to him, no less than to my colleagues on the Geological Survey, whose opinions he has controverted, that I should enter * Op. cit. vol. xxxiy. p. 153 (1878). [Their volcanic origin appears to have been first pointed out to him by Mr. Hudleston.] t Proce. Bristol Nat. Soc. new ser. vol. ii. part ii. p. 121. - t Rep. Brit. Assoc. for 1878, p. 536; Quart. Jou. Geol. Soc. vol. xxxy. p. 285 (1879). § Popular Science Review, N.S. vol. v. p. 289 (1881); Proc. Geol. Assoc. vol. vii. pt. i. p. 59 (1881). PRE-CAMBRIAN ROCKS OF ST. DAVID’S. 267 into the fullest details. In the discussion of the subject I am mainly desirous to get at the truth; and I feel confident that my natural and, I hope, laudable pride in the work done by my predecessors and colleagues on the Geological Survey will not lead me for a moment to forget the signal-services rendered to the history of the Cambrian rocks by Dr. Hicks, which no one can more cordially recognize than I do. At the outset I may allude to a characteristic feature in ‘the literature of the subject. From the brief summary above given of the various papers which have appeared, it will be clear that the views at present entertained regarding the Pre-Cambrian age and metamorphic character of the rocks of St. David’s are the result of a process of development during a course of years. Dr. Hicks at first put forward the idea of Pre-Cambrianism somewhat vaguely and timidly ; but each successive communication from him has shown increasing boldness in the enunciation and extension of his doctrine. Though interesting in itself, this evolution of opinion has been at- . tended with the disadvantage that so many of the statements and views expressed in the earlier papers have since been tacitly modi- fied or abandoned in the later ones, that it is difficult to know how far these earlier publications are available for citation as expressing Dr. Hicks’s ultimate opinions. Indeed they appear to possess little more than an historical value, as records of the successive stages through which their author’s present convictions have been reached. I shall only cite them where their observations are not positively contradicted in his later memoirs. It will be most convenient to discuss seriatim each of the three alleged Pre-Cambrian groups at St. David’s, beginning with the oldest. This, moreover, is nearly the order in which my investi- gations in the field were conducted. 1. “ Dimerran.” Immediately to the south of St. David’s a gentle ridge, roughened here and there with rocky prominences, stretches in a south-west- erly direction for rather less than two miles. Its component rock is seen in scattered knobs, but in no continuous section, until, at its seaward termination, from the bay of Porth-lisky eastwards for about half a mile, it forms a rocky shore. This coast-section is the only continuous exposure of the rock in the district. But laying that rock bare as 1t does, both in horizontal ledges and vertical cliffs, and revealing its contact with the adjacent strata, this coast-section affords the geologist every facility for determining the structure and. stratigraphical relations of the rock that forms the ridge. In the original work of Dela Beche, and in the subsequent investi- gations of the Geological Survey, the rock in question was regarded as eruptive, and as later in date than the Cambrian strata, through which it was believed to have been intruded. At the time when these observations were made, the study of petrography was in a sadly neglected state in this country. We must not, therefore, ex- A. GETKIE ON THE SUPPOSED 268 ‘Ss TILA °%%ld ‘dey 008 s104407 pue stoquunu Fo uoryeuedxe 10), ~ a ¢ ~~ “SUD}9-YINO 07 12980) Wau Wott WOLQIIG—T “SLT pug ag fo hibopoay oy) Buymsnyyr suonoag synununsborg “AA NTAL Quart. Journ. Geol. Soc. Vol. XXXIX. Pl. VIII. [To face p. 268. * WLUU LS FY DIL IN X 3 © DINIIZNS ATHIIH La on TPLOUAS JINITING beste Ate NA ud "ONw vrr7e SN 8 "Feveswa7102V |" os? S : an ‘ On 3 SLSIMS AWSIT HA YO N N > 2 WOTINO? PL(zL4YNe a 8 GS “SR TUNS NOINIIHD XN ‘S27uNS O27" of "sare araund VZ z “SIIVMS ONY T4149 NUNS *SPLAO NE » ‘ P asvavig } VV \ $ MARR ERE BAAS Wan weed eo IN Oo] "NWIOITTA NIA AMAIA DAW IL ——— UST Ve a! eal Spe stan Gee } Fe ee me et‘ s (ee Vi errdsus, = : Cie etre hain as u Moe (ae Lue me Pe cep See . V7 .' 4S 14+ RE MoAE ee ie ri AtIwWoags "SQCIAWG -uiS 3 O dVW IwWwWIIDO1TOS3 D WeOLA NICS N one ora a y NINSI77 SM iy ey B22 SM 1 NATIWM & LAIy Te ~ "NOLIYAWYS 7 & C27) Ne SYSNS SD bs \/- i st MLS OS inlog PRE-CAMBRIAN ROCKS OF sT. DAVID’S. 269 pect precision or accuracy in the determination of rocks then made, either in the Survey or out of it. The rock of St. David’s was marked down as “‘Syenite” on the Survey Map and Section; and so it remains up to this hour. But in the present discussion the main fact to be noticed in regard to the Survey mapping is that the rock in question was declared to be an eruptive rock, intrusive in, and therefore later than, Cambrian strata. In his latest papers Dr. Hicks thus describes the rock in question. ‘The rocks included under this name [Dimetian]. . . . everywhere show more or less distinct lines of bedding, striking from about N.W. toS.E. They vary also, to some extent, in their mineralogical char- acters, if examined at different points, and have apparently an order of succession in which these changes occur at recognisable horizons. The prevailing rocks in this group are of a granitoid character, usually of rather a massive, but sometimes of a schistose nature. ‘Sometimes almost pure quartzites are found; but by far the largest proportion contain an admixture of felspar of a white or pinkish colour. Specks of viridite usually occur more or less through- out, and sometimes give quite a tinge to the rock. Mica occurs sparingly, for the most part; but there are occasionally micaceous, chloritic, impure limestone and serpentinous bands. At some places also, thin, compact, white bands of a more highly felspathic char- acter occur. Some of the beds assume a brecciated appearance, the masses being generally angular or subangular, and in composition much like the associated rocks. “ Speaking generally, the majority of the rocks comprising this group are highly quartzose, of a granitoid or rather massively gneissic nature, and usually easily recognisable by these characters; their strike is about from N.W. to S.E.”*. “This formation [Dimetian] consists chiefly of highly crystalline gneissic rocks, the prevailing types, however, being the so-called granitoid rocks, made up largely of quartz with some pinkish or white felspar.. Hornblende is much less abundant than in the Lewisian rocks; but mica is more frequently met with. It may be said to consist chiefly of acid types of rocks, whilst the former is made up mainly of basic types. Bands of limestone, hornblende, chlorite and micaceous schists occur occasionally in this formation” f. These rocks are regarded by Dr. Hicks as a great, bedded, meta- morphic Pre-Cambrian series, later in age than the ancient gneiss of the Hebrides. They are subdivided by him into two groups—a lower, consisting of “the massive granitoid and gneissose rocks of Bryn-y-Garn, St. Davids,” and au upper, composed of “ the so-called quartz-schists of Porth-lisky ” +. * Proc. Geol. Assoc. vol. vii. no. i. p. 61 (1881). On the next page the “ Di- metian” rocks are asserted to be “ chiefly of clastic origin.” + Popular Science Review, N. S. vol. v. p. 291 (1881). It is to be noted that this paragraph relates to “Dimetian” rocks generally, and is not intended to apply specially to those of St. David’s. { Popular Science Review, Joc. cit. 270 -A, G@EIKIE ON THE SUPPOSED In reading Dr. Hicks’s papers I have been unable to find descrip- tions of, or references to, the numerous natural sections where the relations of the crystalline rock of the ridge to the surrounding mineral masses are displayed. He states, indeed, that the rock is un- conformably overlain by all later formations, but he does not men- tion any localities where he has observed this unconformability. In one passage he speaks of the Cambrian conglomerates resting imme- diately on the “‘ Dimetian” rocks, at the bend in Porth-clais valley, and of higher Cambrian beds in a similar position in the harbour*. It would be more correct to say that the “Dimetian” rock rests there on the conglomerate, as I shall afterwards point out. In another paper he states that “the junction of the Arvonian with the Dimetian may be seen at St. David’s, about a quarter of a mile to the south of the Cathedral, and near Rock House’. But he immediately adds that a slight depression probably marks a fault at that locality. Mr. Peach and I found on examination no evi- dence of any fault, nor of any line of demarcation between two formations. I shall have occasion to refer to this locality in a later part of the present paper. Dr. Hicks, in his various memoirs, introduces many lines of fault, of which, after diligent search, we could discover no trace on the ground, and which, for a clear understanding of the structure of the district, are not requiredt. An unconformability is so important a fact in the geological history of a region, that the most convincing proof of it ought to be demanded. We are entitled to expect that, unless where it is too clear to be mistaken, every available fragment of evidence regarding it should be produced. Still more must this expectation be fulfilled where the rocks in question have been greatly disturbed. No one who has not practically tried it can realize the difficulty of the problem to determine whether or not an unconformability exists be- tween two groups of rocks both of which have been intensely pli- cated or fractured. But to this difficulty no allusion is to be found in Dr. Hicks’s papers. Mr Peach and I began our work by an examination of the ridge of which Bryn-y-Garn is the crest. We were unable to detect any- where a trace of a structure which had the remotest resemblance to the foliation of gneiss or schist. Nor could we discover in the mass any alternations of other rocks. On the contrary, it everywhere retained the same general aspect, and presented all the familiar external characters of a massive eruptive rock. The presence of eneissic structure and intercalations of schistose bands, however, had * Quart. Journ. Geol. Soc. vol. xxxili, p. 231 (1877). Tt Quart. Journ. Geol. Soc. vol. xxxv. p. 289 (1879). t In his map published in 1877 (Quart. Journ. Geol. Soc. vol. xxxiii. pl. x.) Dr. Hicks represents a fault at every locality where the junction of the “Dimetian” with the other rocks is actually visible. These supposed faults have been introduced owing to a mistaken notion of the structure of the ground, and are not required even on the theory to support which they have been in- voked. PRE-CAMBRIAN ROCKS OF ST. DAVID’S. 271 been so confidently and constantly affirmed that, for a time, we were inclined to believe we had missed the proper exposures. It was not until we had diligently hammered every knob and boss of rock in the whole district, without discovering any other structure than that of an eruptive mass, that we were driven to abandon as entirely imagi- nary the idea of a bedded structure and metamorphic origin for the central rock of the ridge. The comparatively limited and dis- connected sections of the interior amply suffice to make this quite clear. But the admirable continuous sections of the coast-line reveal the structure of the rock so completely that one could not but ask oneself the question many times a day, how such a rock ‘could ever, by any possible stretch of the imagination, be credited with a bedded structure and a metamorphic origin. Cut at all angles by the sinu- osities of the coast-line, it can be studied foot by foot across its entire breadth. Did it, therefore, possess foliation of any kind, or were it made up of parallel bands of different lithological characters, such a structure could not possibly escape notice. After the most careful search, however, neither my companion nor myself could discover any thing of the sort. But for the published statements regarding the rock, we should never have thought of making any stich search ; for the first few exposures would have sufficed to mark it out as unquestionably an eruptive mass. The petrographical characters of this rock will be given in Part II. of this paper (p. 313). To the naked eye it appears everywhere to be thoroughly crystalline and granitic in structure, like a granite of medium grain, perfectly amorphous, without any trace of ground- mass or any approach to foliation. It can readily be seen to be composed mainly of a granular crystalline ageregate of quartz and felspar with abundant ‘minute black or dark-green specks, which, by their decomposition, give rise to a diffused greenish discoloration. These dark specks were regarded by the Geological Survey as horn- blende; and hence, according to the old nomenclature, the rock was termed a syenite. On the other hand, were these dark green specks shown to be a mica, there could be no hesitation in classing the rock as a variety of granite. Whether examined in mass, in hand- specimens, or under the microscope, it presents the ordinary struc- ture of a granite. I shall therefore speak of it simply as a granite, and leave its peculiarities of composition to be discussed in the sequel. I may here mention, in passing, that I have examined microscopi- cally a large series of slices of the rocks of St. David’s, and that the result of this examination will be given in Part II. (as previously stated). The numerous rocky bosses upon the ridge south-west from St. David’s, and still more the long coast-section from Porth-lisky eastward, everywhere present a massive rock entirely destitute of definite structure, but traversed by irregular joints, which divide it into blocks, as in any ordinary granite. Here and there by a dominant set of joints it is separated into rudely parallel beds, or even thin slabs, as at Porth-lisky and eastward, where one series of joints, running from N.N.E. to §.8.W. with a high inclination, 272 A, GEIKIE ON THE SUPPOSED gives, at a distance, a deceptive resemblance to bedding. This resemblance, however, disappears on examination. Numerous other systems of similar joints cut through the mass, precisely as they do through any eruptive crystalline rock. But nowhere have they the character of the divisional planes of a foliated rock, nor do they correspond with any internal arrangement of the component materials in parallel folia*. Dr. Hicks lays stress on the fact that, owing to its tendency to split, the rock cannot be dressed for building- or paving-purposes. He proceeds to generalize this fact into a kind of test ‘‘in distin- guishing many of the metamorphic rocks from those of igneous origin’ +. But surely there is no more familiar structure among the eruptive rocks than their tendency towards multiplied jointing in certain directions. Even in a massive homogeneous granite, where a practised geologist could not detect the least trace of any divis- ional planes, the quarrymen will at once show him what they call the ‘“‘reed”’ of the rock, along which it will break easily, but across which its fracture is less reliable and definite. From this condition every gradation may be traced, especially among the weathered parts, until the rock splits into slabs and might at first be mistaken for a bedded mass. A tendency to split in a given direction is therefore no necessary indication of bedding, and need have no connexion with foliation. Had the rock of St. David’s been one which might be classed with the gneisses and schists, it would certainly have revealed abundant proofs of foliation—that is, of a crystalline arrangement of its component minerals parallel with the general divisional planes of the rock. Dr. Hicks asserts that “traces of foliation are abun- dant” +. I can only meet this assertion by the statement that my companion and I searched most carefully every exposure of the rock we could find on the ground, and that I have since examined microscopically a series of specimens taken from ‘all parts of the ridge, without detecting either on the large or the small scale any, even the most distant, approach to a foliated structure. Many eruptive granites exhibit perfect foliation along certain pegmatite veins ; but even this structure we failed to detect §. Between the walls of the joints various decomposition-products * Tt may not be out of place to quote here, in confirmation of our observa- tions, those of Mr. Tawney. He recognized the tendency of the rock at Porth- lisky to split into flaggy and rhomboidal pieces, owing to concealed laminz coated with a thin chloritic lining, and was disposed to look on this structure as bedding; but he states that ‘‘ elsewhere it is difficult to say which divisional planes are dominant or less irregular than the others.” (Proc. Bristol Nat. Soc. N. S. vol. ii. pt. ii. p. 117.) Tt Geol. Mag. dee. 2, vol. viii. p. 148 (1881). { Geol. Mag. dec. 2, vol. viii. p. 142. § Dr. Callaway also could find no trace of foliated structure in the crys- talline rock of St. David’s, though he searched for it in the principal localities named in Dr. Hicks’s papers (Geol. Mag., dec. 2, vol. viii. pp. 94, 237, 1881), and, as he adds, it is not mentioned as existing in any of the microscopic descriptions that have been published of the rock. PRE-CAMBRIAN ROCKS OF ST. DAVID’S. 273 have been introduced. Most of these are greenish in colour and more or less earthy in texture. Where the opposite walls have been displaced, slickensided surfaces may be seen upon them and upon the substances interposed between them. But these appear- ances present no.features differing from what are universally found among massive jointed rocks. They cannot be confounded with any original structure of the mass. Occasionally veins of a paler colour and finer texture ramify through the rock. They vary in width from an inch or less up to more than a foot. very one familiar with a large mass of granite will recoguize such veins at once as characteristic features of it. These also are referred to more in detail in Part II. The only distinct species of rock which we could discover in the mass is a dull-greenish, more or less decomposing, diabase or wacke, occurring in the form of abundant dykes and veins. These vary from a few inches to several feet in breadth, and traverse the granite irregularly in all directions. Where several run parallel at a short distance from each other, and have a slight hade in the same direction, they produce a deceptive resemblance to an alter- nation of beds. That these dark-green rocks, however, are all eruptive, intrusive, and of later date than the granite, may be confidently inferred on the following grounds :—Ist. They have precisely the ordinary external forms of eruptive dykes and veins, ramifying in different direc- tions, coalescing and reuniting. 2nd. They present the usual micro- - scopic characters of dykes of diabase or ancient basalt-rocks, with which they agree in crystalline structure, in the presence and linear arrangement of amygdules, in the existence of a rude prismatic structure transverse to the walls, and in their tendency to sphe- roidal weathering. 3rd. In microscopic structure they unmistakably belong to the Vasalt family. 4th. They not only traverse the granite (being most abundant in it) but are found cutting through the Cambrian beds on many different horizons. The true character of these eruptive rocks appears to have been first detected by Prof. Judd *,—a conclusion confirmed by Mr. Tawney 7. Dr. Hicks draws a distinction between some of them which he admits to be intrusive and others which he seems disposed to regard as bedded in the “ Dimetian” mass¢. But for this distinction I was unable to discover any ground whatever. ‘There can be no question as to their universally intrusive character and late Cambrian or Pos Cambrian date. In some of his earlier papers Dr. Hicks refers to the occurrence of abundant shales among the more crystalline rocks of the ridge. In his more recent summaries no reference is made to such inter- * Quart. Journ. Geol. Soe. vol. xxxiii. p. 235 (1877), and xxxiv. p. 156 (1878). t Proc. Brist. Nat. Soc., new ser. vol. 11. pt. 2, pp. 113 e¢ seq. ¢ Quart. Journ. Geol. Soc. vol, xxxiv. p. 156. In his more recent papers he omits mention of these rocks as integral parts of his “ Dimetian ” series, uniess they are included in the “hornblende and chloritie schist,” which he states to be also a portion of the same series. 274 A, GEIKIE ON THE SUPPOSED calations, though the occasional occurrence of micaceous and chloritic schists is referred to. I have been unable to determine what portions of the mass of rock at St. David’s could have been regarded by him as stratified or foliated intercalations of any kind. I believe him to have been deceived sometimes by the greenish decayed material filling up the partially opened joints, sometimes by the diabase dykes and veins. He appears also to have included in his “‘ Dimetian ” group portions of the undoubtedly bedded rocks (quartz- schists, quartzites, shales, &c.) which flank the massive rock of the ridge, as will be shown in the subsequent description of the coast- section at Porth-lisky. I repeat in the most emphatic manner that, after an exhaustive search over the whole ridge in question, neither Mr. Peack nor [ could find the shghtest trace of any shale, schist, quartzite, gneiss, or other stratified rock, bedded with that composing the ridge between Bryn-y-garn and the headlands south of Porth-lisky—nor of bedding or definite structure of any kind, save the joints universally present in similar massive rocks. We cannot even conjecture on what grounds the assertion has been so often made that the central part of the ridge is bedded, and that its bedding has an invariable strike from N.W. to S.E. We could see absolutely nothing in the rock to afford any basis for such a statement *. Did no other than petrographical characters exist to guide us, these are so clear in their concurrent testimony that there could be no doubt as to the propriety of placing the rock in question among the granites. It has the usual typical features of a granite, and none of those of a schistose rock. But further evidence is abundantly available. That this rock is not only a granite but one which has been erupted through the Cambrian strata, and must therefore be younger than they, is admirably demonstrated by the way in which tt behaves to the rocks that surround it. A field-geologist naturally turns at once to the line of junction between two rock-masses to ascertain their mutual relations. Unfortunately, in most cases such a line is so much obscured by superficial deposits that the actual contact of the rocks can, at the best, be seen only to a limited extent and in few . places. At St. David’s, however, the coast-section and the trans- verse valley cut by the river Allan permit the actual junction of the granite with the surrounding rocks to be seen at several localities and on both sides. I have searched in vain among the published papers for any account of these localities. It is difficult to believe that they can have been actually seen by any one who could after- wards maintain the rock to be Pre-Cambrian in age and meta- morphic in origin tf. They show the granite to be unmistakably * On the occasion of my second visit to St. David’s I again sought, with Mr. Topley, for any trace of foliation or bedding in the crystalline rock of the ridge, but equally without success. t It appearsthat Dr. Hicks started with the idea expressed on the Geological Survey map that the crystalline rock of the St. David's ridge is intrusive. He afterwards wrote that “on further examination it seemed clear that the syenite PRE-CAMBRIAN ROCKS OF ST. DAVID’S., es eruptive; for the strata adjacent to it present examples of the induration and silicification so commonly, though not universally, observable along the borders of a granite boss. In describing the sections that exhibit the actual contact of the eruptive and sedimentary rocks, I would first allude to a fact of some importance which hitherto appears to have escaped notice. In the course of my examination my colleague and I observed that as it crosses the valley of the Allan above Porth-lisky, the granite sends out a tongue-like projection across the river at the ford, and that this projection is separated, by an intervening mass of Cambrian shales and sandstones, from another projecting tongue of granite lying further north. This northern portion may cross the river as a narrow belt and thereby connect the main mass of Bryn-y-garn with the lesser area that extends to Porth-lisky. As it contracts, however, to a breadth of not more than eighty yards on the west side of the Allan, there seems to be hardly any room for it to pass across the valley. Though, no doubt, continuous underneath, the granite mass is probably divided at the present surface into two separate areas by intervening Cambrian strata. (See Map, Plate VIII. p. 268.) The Bryn-y-garn granite mass projects for a few yards into the Cambrian grits and shales on the right bank of the Allan. I had several yards of the actual contact of the rocks laid bare at the foot of the hill, and found that the granite distinctly overlies the grit (fig. 3.), Fig. 3.—Section of Junction of Granite with Cambrian Strata. fight bank of Allan Rwwer, Porth-claas. the line of junction being a wavy surface inclined at an angle of about 55° *. The grits are much indurated; and their bedding is did not penetrate any of these beds” (Quart. Journ: Geol. Soc. yol. xxxiii. p. 229, 1877). I cannot conceive in what direction this further examination was carried, nor how the very clear proofs of intrusion could have been missed or misunderstood. * This junction was bared, with Mr. Topley’s assistance, on the occasion of my second visit. Some portions of this grit are so coarse as to pass into a quartzose conglomerate, which may be the conglomerate band above the vol- canic group. This is about the place where that band should come in, next the green and red beds seen at Porth-clais. Its position is suggested in Section fig. 1, p. 268. 276 A. GEIKIE ON THE SUPPOSED obliterated, though lines of pebbles can be traced which appear to indicate that the strata are nearly vertical. On the opposite side of the river another junction of the granite with the Cambrian beds can be seen. The latter consist of greenish shales and sandstones dipping N. 20 W. at 55°., and are here again distinctly overlain by the granite, which cuts across the edges of the strata that dip beneath it. At this point the line of junction has served as a channel for percolating water; and the rocks on either side are so decomposed that no satisfactory observations of their internal characters can be made. It is deserving of remark that, in its course across the valley, the projecting tongue of granite now described traverses obliquely a considerable thickness of strata. In particular, it can be seen to have cut out nearly the whole of the thick bed of grit above referred to, no portion of which-appears on the opposite side (figs. 4 and 5). Fig. 4.—Plan of Junction of Granite with Cambrian Strata, Porth-clais. conanent nas 9% . ites Ngee ae 5 aS a ry] . The eastward prolongation of the Porth-lisky granite mass likewise protrudes as a tongue into the Allan valley. This tongue has a breadth of about eighty yards ; but it seems to be narrowing eastwards, so that, as already stated, it probably does not cross the valley. On both its northern and southern borders its junction with the Cambrian rocks can be seen. On the south side greenish sandstones, shales, and silky hydro-mica schists, like some of those to be afterwards referred to as occurring at Porth-lisky, abut against the granite; but the rocks along the line of contact have been decomposed into clay by the rise of water. On the north side the junction is more satisfactorily shown ina quarry on the left side of the road from Porth-clais to Rhoscribed. Here the conglomerate, in highly inclined beds, is overlain by the granite, which leans against it. The conglomerate is indurated; but at the actual contact both rocks have been much decomposed by percolating water. Some of the details of these junctions of the granite with the stratified rocks are reserved for the second part of this paper (p. 317). Before leaving the relations of the granite to the Cambrian strata in the Allan valley, I must allude to the fact that this is the locality cited by Dr. Hicks as showing the Cambrian conglomerates and PRE-CAMBRIAN ROCKS OF ST. DAVID’S. A i higher beds resting on his “ Dimetian” ridge*. At every section, . instead of lying upon the granite unconformably, they plunge beneath it. The general disposition of the rocks in this part of the valley is expressed in the accompanying diagram (fig. 5). It will Fig. 5.—Sketch Plan of the Disposition of the Rocks in the Allan Valley at Porth-clais. be evident from this map and from the foregoing description that the relative positions of the two rocks cannot be accounted for by faulting +. At Porth-clais the actual terminal curve of the granite projection can be traced across the bed of grit through which it has risen (fig. 4). These junctions are characteristically those of an eruptive mass. But the most important junction of the granite and Cambrian beds is that which has been cut by the sea in the range of cliffs between Porth-lisky and Porth-clais at the little inlet of Ogof- llesugn (fig. 6). The granite, which extends continuously eastward from Porth-lsky, abruptly ends off, and is succeeded at once by vertical sandstones and shales, which are truncated by it nearly at a right anglet. On the seaward face of the cliff the granite has * Quart. Journ. Geol. Soe. vol. xxxiii. p. 231 (1877). t+ That there may have been some local slipping along the boundary-line between the granite and the rocks it has invaded is not unlikely. In the Allan valley the faults would need to be reversed ones, and to wind about so as precisely to counterfeit the boundary-line of an eruptive rock. This subject is further referred to in Part IT. (p. 310). This locality is referred to by Dr. Hicks as a line of fault; indeed, in his map (Quart. Journ. Geol. Soc. vol. xxxi. pl. vill.), as already stated, he makes the boundary-line between the two rocks everywhere a fault. I have admitted that, along the flanks of the granite, occasional local slips may have taken place ; but the visible sections prove that no continuous or important faults occur there. Possibly some slight displacement may have taken place at Ogof-lesugn ; but the mass of conglomerate is imbedded in the granite. It should be noted here that, the section described in the text is the same as that already referred to in the citation from Sir A. C. Ramsay’s early MS. report as affording the best evidence of the intrusive nature of the igneous rock. Q.J.G.S. No. 155. XY 278 A. GEIKIE ON THE SUPPOSED torn off a mass of conglomerate and associated tuffs. These rocks have been so intensely indurated and silicified that the quartz pebbles . are hardly traceable on a fresh fracture, though they project more evidently from a weathered surface. It is even difficult in places to say precisely where the line between granite and conglomerate should be drawn, so intimately are they welded together. The former rock, still presenting its normal petrographical characters, may be seen both underlying and overlying the involved portions of conglomerate, red shale, and fine tuff, the latter being altered into a kind of hornfels or porcellanite. Veins of granite penetrate these altered rocks. Great numbers of diabase dykes traverse the granite ; and some of them cut the Cambrian strata also. Fig. 6.—Plan of Junction of Granite with Cambrian Strata, Ogof-llesugn, St. David’s. The importance of this section in any discussion as to the nature of the crystalline rock of the ridge south of St. David’s cannot be exaggerated. It is rather difficult of access, which may possibly . account for the absence of any description of it in the papers hitherto published ; but it completes the demonstration that the rock, which can be traced from St, David’s to the coast south of Porth-lisky, is an eruptive mass that has been intruded into the Cambrian strata. The quartz conglomerate, here altered by the granite, is unquestion- ably the same band which can be traced along the coast for fully two miles eastward, and the greenish and reddish tuffaceous shales are recognizably those that everywhere accompany the conglomerate. On the opposite side of this portion of the granite ridge the eruptive mass comes into contact with the stratified rocks in the Bay of Porth-lisky. Unfortunately, however, the actual junction is obscured, on the cliffs by the decomposition of the schists that abut on the granite*, and on the beach by the quantity of fallen blocks. The condition of the beach doubtless varies from time to time ; but neither on the occasion of my first visit last September, nor on the second examination five months later, could I trace the actual contact of the two kinds of rock, though I followed them to within a few feet of each other. The schists are, in some bands, much indurated, passing into a kind of quartzite or quartz schist. The crystalline rock of the ridge, as exposed along the cliff, presents * There may have been a shift at the line of junction here. PRE-CAMBRIAN ROCKS OF ST. DAVID’S. 279 features not unfrequently observable along the edge of the granite, being fine-grained and of a diffused greenish tint. On the beach it becomes in places very quartzose and much impregnated with calcite, some portions weathering with a nodular surface not unlike that of a conglomerate. Possibly there may be some dislocation along the line of junction, and the calcareous portions may be due to infiltra- tion along the lines of fracture*. Though I searched the locality very carefully with Mr. Peach, and subsequently with Mr. Topley, I could not trace bedding in the granite such as has been described by Dr. Hicks. As aiready mentioned, the granite is much jointed here, and the joints are in some places close and rudely parallel ; but they are mere joints, readily distinguishable from any original structure of the rock. It is worthy of remark that the schists which abut on the granite, and extend across the Bay of Porth-lisky, dip at high angles towards N.N.W. They strike at the granite, so that, apparently, lower beds come out as they are followed seawards. Ishall afterwards show that all these strata are inverted, and that, consequently, the most easterly beds at Porth-lisky are stratigraphically higher than those immedi- ately to the west of them. Mr. Peach and I observed these peculiar schists at Ramsey Sound lying below the conglomerate ; indeed, if the rocks of Porth-lisky could be traced for a quarter of a mile further out to sea, the conglomerate would doubtless make its appearance‘. If now we turn to the Map (PI. VIIL., p. 268), it will be seen that the eranite, in its course from St. David’s to the sea, cuts across succes- sive horizons of Cambrian beds, penetrating deepest into them on the north and east, and reaching its highest platform on the south. The way in which it has broken through and pushed aside the con- glomerate is peculiarly striking. That band of rock has been assumed by Dr. Hicks to be the base of his Cambrian system ; we find, how- ever, that the granite not only invades it, but ascends across the over- lying shales and sandstones. One further statement calls for notice here. Dr. Hicks, having satisfied himself that the granite of St. David’s is a bedded metamor- phic rock, has ventured upon estimates of its thickness; in his paper of May 1878, he remarks that the thickness previously claimed by him for his “ Dimetian” group, viz. 15,000 feet, is not an over- estimatet. But, as will be afterwards pointed out, the only bedded rocks that occur between St. David’s and Porth-lisky lie on the flanks of the granite, and strike with the ridge instead of across it, * Mr. Tawney, who had the advantage of being guided over the locality by Dr. Hicks himself, regarded the calcareous bands as “due to the decomposition caused by water filtering down joints, removing alkaline silicates, and depositing carbonates of lime and magnesia” (Proc. Bristol Nat. Soc. N. 8. vol. ii. pt. 2, p-116). I cannot doubt that this is the true explanation of the limestone and dolomite bands described by Dr. Hicks from this locality. Tt In connexion with this southward prolongation of the conglomerate, we must look on the mass at Ogof-llesugn as having been torn off from the main body, which must lie somewhere beneath the granite, { Quart. Journ. Geol. Soc. vol. xxxiv. p. 154; see also vol. xxiii. p. 230 (1877). XZ 280 - A, GEIKIE ON THE SUPPOSED so that they must be measured at right angles to the line taken by Dr. Hicks. I am not aware of any method by which we can measure the thickness of a square mile of granite. To the phenomena of contact-metamorphosis I shall return in the second part of this paper (p. 317). The evidence now brought forward is, | submit, amply sufficient to prove that, whether studied in hand- specimens, in microscopic slices, or in the numerous natural sections which show its geotectonic relations, the “ Dimetian” group of Dr. Hicks, instead of being a ridge of Pre-Cambrian metamorphic rock, is really a boss of eruptive granite, later in date than the Cambrian strata through which it has been intruded, and that the term ‘‘Dimetian”, so far at least as regards its original locality, must be abandoned. 9. ** ARVONIAN,.” The rocks grouped under this name by Dr. Hicks at St. David’s are thus described by him. ‘The rocks now included in this group I originally associated with the Dimetian ; but in the year 1878 * I separated them from the latter, under the above name. ‘‘On the Survey Maps they are coloured generally as felstones and porphyries, usually intrusive amongst Cambrian or Lower Silurian rocks. They consistin reality of flows of rhyolitic lavas, alternating with felsitic breccias and halleflintas. The strike is from N. to §., and, hence, discordant to those newer rocks with which they are usually surrounded, as also to the underlying Dimetians. Like the Dimetian, this is a highly acid group, being mainly made up of the types of rocks known as the quartzo-felspathic. But, instead of | being like these, chiefly of clastic origin, we have here a great series of acid lavas mixed up with a comparatively small proportion only of rocks of a clastic nature. -In colour these lavas vary from being very dark (almost black) to a light grey, and from deep red or violet to fiesh-colour. The flow-structure is usually well marked, and in many cases the spherulitic structure also. A large number are por- phyritic, from the minute crystals of felspar or quartz. The hille- flintas are more siliceous-looking than the rhyolites, and havea horny-looking texture and fracture. Under the microscope they are still more easily distinguishable. Their chief peculiarity, perhaps, consists in the manner in which some of the quartz becomes separated away into nests, so as to give the rock a curious pseudo-porphyritic appearance; whilst the intervening parts exhibit the appearance of a micro-crystalline mass of quartz grains, with intervening felsite. The breccias usually consist of fragments of lavas and halleflintas, like those in association with them, and the pieces angular. “This group, therefore, #s characterized by being for the most part made up of acid lavas, breccias, and compact siliceous rocks of the halleflinta type, and as usually having the strike in a direction from N. to 8.”f. * (Quart. Journ. Geol. Soe. vol. xxxv. p. 285 (1879). ] t Proc. Geol. Assoc. vol. vii. no. i. p. 62 (1881). PRE-CAMBRIAN ROCKS OF ST. DAVID’S. 281 In the definition of the lithological characters here ascribed to the rocks in question I am disposed generally to agree. These mineral masses are partly eruptive quartz-porphyries, and partly highly sili- ceous strata of sedimentary origin to which the names halleflinta, hornfels, porcellanite, chert, Kieselschiefer, adinole, &c. might in dif- ferent places be applied. But here my agreement ends. Instead of finding evidence that these rocks lie with a discordant strike uncon- formably against the so-called “ Dimetian” below, and are covered unconformably by Cambrian or “ Pebidian” beds above, Mr. Peach and I discovered that Dr. Hicks had really created a separate stra- tigraphical “group” out of the zone of quartz-porphyry bosses and dykes with the accompanying indurated sedimentary rocks that surround the central core of granite. I shall discuss the phenomena of intrusion and metamorphism in the second part of this paper. There are only two questions that need be considered here. In the first place, Dr. Hicks asserts that his “ Arvonian” rocks usually present a north and south strike, and are unconformable to his ‘‘ Dimetian” group. This assertion has been virtually disproved by the evidence which I have now advanced as to the true nature of what he calls ‘‘ Dimetian.” But, for the sake of precision, I may here state that my colleague and I made careful observations of the strike of the rocks all round the granite, and found the dominant trend to be parallel with the granite ridge—that is, generally in a north-easterly and south-westerly direction. Where the average strike changes, it is rather towards east and west than towards north and south, as is more especially noticeable on the coast between Ogof-llesugn and Caerbwdy. Nowhere could we detect a prevalent north-and-south strike, nor any general tendency in the rocks to strike at the granite. The quartz porphyries, which appear to constitute a great part of the so-called “ Arvonian group,” show no strike. They are really as devoid of any semblance of bedding as an eruptive rock can well be. Dr. Hicks remarks that the junction of the ‘ Arvonian” and “‘ Dimetian” rocks is to be seen at St. David’s; but he immediately adds that there is a line of fault at the locality between the two groups. He describes the “‘Arvonian” rocks as “striking up towards the ridge” and “ the lowest beds ” as being visible near the Deanery. The rocks exposed over that area are of a type not un- frequently observable round the edge of the granite and doubtless connected with it, to which I shall refer more in detail in Part II. They are entirely amorphous, eruptive masses, without the least trace of any kind of bedding. On the road-side between the Deanery and Rock House, among the numerous joints there is one set that runs in a north-and-south direction, the joint-planes being inclined at high angles towards the west. This was the only structure that I could discover which might have suggested the idea of bedding. In the second place, Dr. Hicks alludes to the view expressed on the Survey Maps that the igneous rocks in question are intrusive, but only to dismiss it without further notice and to substitute for it the statement that “they consist in reality of flows of rhyolitic 282 A. GEIKIE ON THE SUPPOSED lavas, alternating with felsitic breccias and halleflintas”*. In another paper he writes that ‘they were marked on the Geological Survey Maps as intrusive felstones; but a very cursory examination proved that they were not of that nature, and that they were in reality bedded sedimentary rocks which had undergone metamorphic change” t. Passing over the discordance between these two emen- dations of the Survey Maps, I would observe that the author, in dis- missing the view taken by the officers of the Survey and substituting for it another of his own, offers no observations of any kind in support of his emendations. He simply declares the rocks to be rhyolitie lavas (meaning evidently, streams of lava that have flowed out at the surface), but mentions no character by which they are to be distinguished from intrusive masses. In a previous paper, indeed, he had admitted that they were “ possibly intrusive ’’t, though at the same time he regarded them as “appearing distinctly to lie in the line of bedding of their associated quartz rocks.” He would seem to have been led to regard them finally as lavas, from a remark made to him by Professor Bonney that they most resemble a lava- flow§. I presume it was the presence of fluxion-structure in them that suggested this identification ; but I shall subsequently show how fallacious this presumed test is for the purpose of distinguishing the superficial from the more deep-seated manifestations of volcanic matter. To go no further than the region of St. David’s, I find spherulitic structure and fluxion structure in the most obviously intrusive dykes. - In every example in this district where the actual contact of the porphyries with the surrounding stratified rocks can be seen, the porphyries are distinctly intrusive. In the quarries north of the | Church Schools the fine tuffs and schists or shales, which are un- doubtedly a portion of Dr. Hicks’s “ Pebidian” group, are much indurated close to the porphyry, which traverses them obliquely to their bedding. But this alteration insensibly dies away as the strata are followed northward; and at a distance of about sixty or seventy yards they assume their usual characters of fine tuff. The actual intrusion of one of the quartz-porphyries as a dyke or elvan through the strata, however, may be seen in the noble section among the cliffs south of Nun’s Chapel. Other examples occur further west, near Treginnis. The behaviour and structure of these rocks will be discussed in Part II., in connexion with the metamor- phism of the district. Dr. Hicks associates certain breccias with his rhyolites as con- temporaneous components of the ‘“‘ Arvonian group.” But there can be no doubt that they are portions of the volcanic (or what he terms his “ Pebidian”) group which have been invaded by the por- phyries and have been much indurated. They can be seen north of * Proc. Geol. Assoc. vol. vii. pt. 1, p. 62 (1881). Tt Quart. Journ. Geol. Soc. vol. xxxv. p. 286 (1878). { Quart. Journ, Geol. Soc. vol. xxxiv. p. 153 (1878). § Ibid. p. 154. '. PRE-CAMBRIAN ROCKS OF ST. DAVID’S. 283 the Church Schools -forming part of that group, also on the south side of the ridge near Nun’s Chapel. So far, therefore, as regards the evidence to be obtained at St. Dayid’s itself, there is no foundation whatever for the institution of a separate group under the name of “ Arvonian.” The rocks so called by Dr. Hicks are portions of his “‘ Pebidian”’ group invaded and altered by a central core of granite and abundant dykes or bosses of quartz porphyry*. What the “ Pebidian” group really is must now be considered. 3. “ PEBIDIAN. ” The general characters of this group of rocks at St. David’s are thus summarized by Dr. Hicks :— “Most of the rocks in this group differ from those already de- scribed, though occasionally there is a certain amount of resemblance remaining. Instead of the acid types prevailing, as in the pre- viously named groups, we find the basic types more largely developed. Basic lavas and breccias now predominate over the rhyolites ; and the clastic rocks are more micaceous, chloritic, and talcose. On theSurvey Maps these rocks are coloured as altered Cambrian, and partially as intrusive greenstones. On more careful examination the so-called ereenstones turn out to be bands of indurated volcanic ashes, and con- temporaneous basic lava-flows. Agglomerates and breccias occur in great thicknesses in the group; and the fragments are chiefly, except in the lowest beds, of a basic character. Chloritic, talcose, felspathic and micaceous schistose rocks occur also at various horizons, and occa- sionally purple and green slates. Serpentinous bands are also some- times found, as well as veins of jasper, epidote, and asbestos. Some of the finer and more quartzose beds assume a gneissose appearance, and others are porcellanitic. “The strike in this group is from about N.E. to 8.W., and hence nearly in accordance with that in the overlying Cambrian rocks. That this group, however, must have been in much the same con- dition in which it is found, before a grain of the Cambrian rocks was deposited, is perfectly clear from the fact that the conglomerates at the base of the latter are very largely made up of rolled pebbles and rounded fragments identical with the rocks below. An actual unconformity between the two groups is also seen at several points. ““This group consists of a far more varied series than the two former, and doubtless would exhibit a still greater diversity if fully exposed; for it is perfectly clear that, in consequence of the rapid * The intrusive character of the quartz porphyry south of Nun’s Chapel is admitted by Dr. Hicks in his paper of 1877 (Quart. Journ. Geol. Soe. vol. xxxiii. p. 236). He gives a section showing it cutting through the rocks (his “ Pebidian”), and says that it does not penetrate the Cambrian beds above. But at its western end, where it descends to the beach, it approaches the conglomerate, and would probably be found piercing it if the beach could be cleared of the fallen débris. T shall show that this rock is precisely similar in petrographical character to the so-called ‘‘ Arvonian” porphyries of St. David’s (p. 315). 284 A. GEIKIE ON THE SUPPOSED overlapping of the sections by the Cambrian rocks, much is hidden from view” *, I accept generally the lithological descriptions here given, but with important modifications to be afterwards stated. There can be no doubt that the group is almost entirely of volcanic origin— formed principally of various tuffs with bands of olivine diabase and occasional intrusive masses of quartz porphyry. I have already shown that the voleanic nature of these rocks was clearly recog- nized by my predecessor, Sir A. C. Ramsay, in his original map and section, and in his early MS. report on the St. David’s area, and that he afterwards allowed this view to be set aside in favour of the opinion that the peculiar bedded rocks on the west side of the granite ridge are altered Cambrian strata through which intrusive “ green- stones ” have been injected. It is this view which is expressed upon the second and latest edition of the Survey Map and Section. I at once acknowledge that in this respect the present Map and Section are seriously in error, and that Dr. Hicks deserves the thanks of geologists for having, as it were, rediscovered probably the oldest group of paleozoic voleanic masses yet known in this country. Reserving for the second part of this paper what I have to add to the published descriptions of these rocks, I proceed at once to consider the evidence for their forming a distinct Pre-Cambrian group lying unconformably on the groups below, and covered unconforma- bly by the Cambrian strata above, as has been so repeatedly asserted by Dr. Hicks. In this instance, again, I have been unable to discover in his pub- lished papers references to any sections where the proof of the alleged unconformability between the so-called “ Pebidian” and “‘ Arvonian” rocks can be seen. ‘The unconformability, if it existed, might be proved (1) by actual sections showing the line of junction, (2) by detailed mapping of the ground and the detection of proofs of overlap and discordance, or (3) by the evidence of in- cluded fragments. Dr. Hicks asserts that “resting unconformably upon the whole [Arvonian Group] are the great “agglomerates of Clegyr Hill, the base-beds of the Pebidian, which are made up of masses of Dimetian rocks, of quartz felsites, spherulitic felstones, and hialleflintas, and all in the condition in which they are now found composing the underlying ridges. From this evidence it is tolerably clear that the position of the Arvonian or halleflinta group is intermediate between the Dimetian and the Pebidian, and that there is, at least in this area, very clear proof of unconformity and hence of lapse of time having intervened” f. In this passage he enumerates two of the three kinds of proof just referred to as indicative of the discordance in question. Again my companion and I sought diligently for any trace of the alleged evidence, but completely without success. Like the other assertions with which I have been dealing, its groundlessness became more apparent at every step of the investigation. There is not only * Proc. Geol. Assoc. vol. vii. p. 63. t Quart. Journ. Geol. Soc. vol. xxxv. pp. 289, 290 (1879). PRE-CAMBRIAN ROCKS OF ST. DAVID’S. 285 no vestige of an unconformability, but the volcanic groups which Dr. Hicks has included in his “ Pebidian” can be seen in many places graduating insensibly into the altered shales which form a great part of his so-called “ Arvonian.’”’ In fact the latter group, as above stated, consists of portions of the volcanic breccias and tufts (that is, the ‘‘ Pebidian ” strata) where these are invaded by quartz porphyry. This is well seen in the series of quarries north from the Church Schools, where, as already remarked, a perfect gradua- tion can be traced from highly altered shales and tuffs, next the intrusive quartz porphyries, northward into the normal condition of these strata in the district. The assert. a that the Pebidian strata are made up chiefly of “‘ Dimetian” fragments must, from the evidence already adduced as to the late date and intrusive nature of the “‘ Dimetian ” mass, be founded on error of observation. I need hardly say that, after the most patient search, neither Mr. Peach nor my- self could detect anywhere in these tuffs and breccias the smallest fragment which, by the utmost stretch of fancy, could be referred to the granite of the “ Dimetian” ridge. There occur indeed, abundant lapilli of felsite, as I shall more fully describe in the sequel; but these fragments can readily be discriminated from the material form- ing the eruptive porphyry dykes and bosses. But the most extraordinary statement in the passage just cited is that in which the writer asserts that the great agglomerates of Clegyr Hill rest unconformably upon his ‘“ Arvonian” group. The agglomerates in question are well seen on the road-side east of Clegyr Bridge, where they dip towards 8.8.E. at 65°, and are inter- banded with siliceous layers (hilleflintas). These siliceous bands are precisely the same as those seen near the quartz porphyries north of the Board-Schools, on the shore at Nun’s Chapel, and elsewhere. They are characteristic of the volcanic group where it. is traversed by intrusive siliceous eruptive rocks, and particularly of a zone that lies not far below the conglomerate to be referred to in a later part of this paper. The presence of these bands in what Dr. Hicks himself cites as typical ‘“‘ Pebidian ” agglomerates is an important fact, in still further proving that, at St. David’s itself, “‘ Pebidian ” and ‘‘ Arvonian” are only different names for the same series of rocks. On Clegyr Hill the same agglomerate, interbedded with fine tuff and bands of siliceous schist, and traversed by a dyke or boss of spherulitic quartz porphyry, appears to dip towards the N.W. at 40°-50°—that is, actually towards the rocks which it is said to overlie. Nothing is seen at the surface for a distance of nearly a quarter of a mile, when the observer finds that a few square yards of quartz porphyry have been laid bare in a quarry to the south of Trepewit, while, about thirty yards distant, fine tuff, nearly vertical, but preserving the normal E.N.E. strike, is seen on the road-side. Yet Dr. Hicks boldly asserts not only that the rocks of Clegyr Hill overlie the porphyry, but that they do so unconformably ! 986 A, GEIKIE ON THE SUPPOSED : ; 4, Renatton oF ** PEBIDIAN” TO CAMBRIAN ROCKS. The unconformability between the top of the “ Pebidian ” rocks and the conglomerate which Dr. Hicks assumes to be the base of the Cambrian system likewise disappears on examination. As before, he rests upon presumed discordance in the dip and strike of the rocks, and on the alleged presence of fragments of the older rocks in the younger. But both these tests fail him. In the first place it can be conclusively shown that in dip and strike the volcanic group and the overlying conglomerate, sandstone, and shales are perfectly conformable throughout. Mr. Peach and I proved this by numerous measurements all over the district. We observed that, at the locality on Ramsey Sound, near Castell, men- tioned by Dr. Hicks as showing the unconformability of the con- glomerate, there has been a slight local disturbance of the strata. The beds below the conglomerate have been bent up, and the con- elomerate itself has been pushed over them. Seen from the top of the cliff the conglomerate lies in part on their edges. But when examined on tke spot the unconformability disappears, and the strata below the conglomerate are found at a little distance from the disturbance to be perfectly conformable with it. The same complete conformability is well exposed on the face of the next projecting cliff southward. At another locality, on the coast south of Caer-fai, also referred to by Dr. Hicks, there is an apparent discordance between the con- glomerate and the beds below it. But here again the seeming un- conformability at once disappears on examination. It is an instance of the familiar phenomenon of what has been called ‘“‘ contempo- raneous erosion.” Every field-geologist knows that this structure constantly occurs among pebbly strata, from the most recent valley- gravels to the most ancient sedimentary rocks yet known. Were it to be used as indicative of serious unconformability, we might have half a dozen discordant formations in a single grayel-pit. But the section at Caer-fai, as shown in fig. 7, might as well be taken to prove unconformability above the conglomerate as below it. In reality these are merely common local accidents of sedimentation, and, but for their relation to the question now under discussion, would not be worthy of special notice. Dr. Hicks, in one of his papers, remarks that “the line of strike of the Cambrian rocks appears at first sight to be nearly identical with that of the underlying Pebidian beds; but when examined carefully it will be seen that in no case is it truly so, but that the conglomerates overlap the beds irregularly and at different points in the succession”*. In another paper he states that the Cambrian conglomerate overlaps the “ Pebidian” group altogether, so as to rest upon the “‘ Dimetian” rocks‘. I was not surprised by these statements when I found that he had wholly missed the structure of the ground between Ramsey Sound and the granite ridge. At first an observer traversing * Quart. Journ. Geol. Soe. vol. xxxiv. p. 159 (1878). t Ibid. vol. xxxiii. p. 230 (1877). PRE-CAMBRIAN ROCKS OF ST. DAVID’S. 287 Fig. 7. Contemporaneous Erosion accompanying Cambrian Conglomerates. Caer-fai, St. David’s*. SSS Fes a = wore as oe. POD e BORA 5 ~ the coast-section, or looking at the occasional exposures inland, and finding that all the tuffs, breccias, and diabase sheets dip steadily in a north-westerly direction, would infer that he is crossing a continuous succession of beds, the highest being at the north-west end and the lowest at the south-east end of the section. This natural inference has been drawn by Dr. Hicks, and may partly account for some of the errors into which he has fallen. Further comparison, however, would have shown him that the strata are here isoclinally folded; that is, they have been thrown into an anticline, which has been bent over to the south-east, so that the strata in the south-eastern half of the fold are inverted (figs. 1 & 2, p. 268). That thisis the case, was proved by Mr. Peach and myself in the identification of the same beds on the two sides of the arch. In particular, the peculiar group of shales or schists immediately below the conglomerate on Ramsey Sound reappears at Porth-lisky. The conglomerate accompanies them ; but at the latter locality it has been cut out by the granite. It appears, however, a short way inland in the Allan valley, and on the east side of the granite at Ogof-llesugn. The reversed dip continues along the coast-line; but the beds are eventually seen to right themselves, and they appear in normal order to the east of Caer-fai. I shall return to this interesting structure in the second part of this paper (p. 309)*. * Dr. Hicks figures this junction as an unconformability of the Cambrian conglomerates on the Pebidian Beds. But he reverses the visible dip, making the rocks inclined towards the sea instead of towards the land (Quart. Journ. Geol. Soc. vol. xxxiii. p. 236). This subject is again referred to in the text. + It may be proper to notice here that the structure above described proves that Dr. Hicks’s estimate of the visible thickness of his “ Pebidian ” group is greatly exaggerated. He makes the thickness at least 8000 feet (Quart. Journ. Geol. Soc. vol. xxxiv. p. 159). Were the beds absolutely vertical all the way, they could not be more than 4000 feet ; for they extend across a belt which, to 288 A, GEIKIE ON THE SUPPOSED In the second place, Dr. Hicks has stated more than once that the Cambrian conglomerates are largely made up of the underlying “ Pre-Cambrian ” rocks*. As the result of a most careful examina- tion of the conglomerate belt along both sides of the fold, I feel myself warranted in stating confidently that it contains not a single pebble of the characteristic granite of the St. David’s ridge. The actual composition of the conglomerates will be best understood from the percentages taken by Mr. Peach on the west side, and by myself on the east side of the fold. Percentage of Stones in the Cambrian Conglomerates. West side of Isocline. East side of Isocline. Quartzite (often red)...... Bie) Quartzite (generally red)...... 59 Quaattig cones. See eee 25 Quarta: \.os... 5 ncccccs cece 35 IDE 0) o Try Ss S $2 - &2 |B. Yellow siliceous sands with a few 908 So green grains and casts of shells, co) N23 few and ill preserved. Beds A o and B are 226 feet or more thick. C. Pebble bed, more or less regular and of variable thickness, in a greenish or ferruginous sand. 10 to 18 inches. D. Yellow and greenish sands with ferru- ginous layers and light-coloured foliated clays. 10 to 20 feet. K. Very fine green sand, with subordi- Shells of nate dark clay and lignite. Fossils} Bracklesham abundant. Average about 20 feet. species. F. Laminated clays, often black or liver- coloured, with beds of impure green sand, lignite, and plant-remains. 15 to 20 feet. AVAING [B9150{00H Jo Speq MLvYysoTyoRlg ‘IF.09 °F OF “TIMysorg Jojoysseg orpprmt | 9 _ BEDS OF THE LONDON BASIN. 349 The railway-cuttings at Goldsworthy Hill, at Ascot, and at Wellington College afford good type sections in the lower part of these beds, and show that the above subdivisions, though varying in thickness, are very persistent. The Goldsworthy-Hill section was described by Prof. Prestwich in 1847 (Quart. Journ. Geol. Soc. vol, ii. p. 382). The subdivision C in the above table is there represented by a “ coarse greenish sand with a few flint pebbles ” two feet thick ; the subdivision D by Prestwich’s bed 2, foliated clays eleven feet; E by his beds 3, 4, 5, eighteen feet; and F by his beds 6 and 7, fifteen feet thick. A very similar section was opened a few years ago on the branch of the South-Western Railway between Ascot and Bagshot, 73 miles north-west of Goldsworthy Hill. Starting from Ascot station, the line passes over Lower Bagshot Sand for about three quarters of a mile; the middle Bagshot clays then come in, and are well shown in a brick-field close to the railway. ‘This brick-field was described in the Memoirs of the Geological Survey (vol. iv. p. 332); but sub- sequent excavation has greatly improved the section. The overlying fossiliferous beds are exhibited in an adjoining cutting on the railway, which was measured by Mr. William Herries and myself in 1879 (Geol. Mag. iii. p. 171); it is now much over- grown. Section on the South- Western Railway, near Ascot. Mate ish yellow ard. fe 2. Yellow sand with layers of darker-coloured iron-sand ............ 2410 C 3. Pebble-bed, with rolled flint pebbles in iron-sand matrix often PAGES grea ent cs iewe ieee tas oaenani talenequionaia. Jee 0 10 D 4. Yellow sand with layers of iron-sand, passing into a finely foliated sandy clay, with patches of yellow and greenish sand ...about 10 0 MAY EEO) Hii EO eRe: Sth. vans sea satee vee tains cacebe RaaN. deeclehs des stad 0 2 E 6. Yellow and liver-coloured foliated sandy clay .................20000- a ©6«0 7. Green sand with a little dark clay, casts of shells abundant in SAVER OL Vel wish SAMO 550.2 aces aeons ecice catches eee en's one Sor 7/6 Fusus longevus, Lam.; Fusus,? sp.; Voluta,? sp., like V. cithara, Lam.; Pleurotoma,? sp.; Natica sp.; Turriteila suleifera, Lam. ; Phorus aggluti- nans, Lam.; Ostrea flabellulum, Lam.; Pecten corneus, Lam.; Carditacarinata, ?Sow.; C. planicosta, Lam.; Cardiwim porulosum, Brand. ; Protocardium semistriatum, Desh. ; Cytherea nitidula, Lam.; C. suberycinoides, Desh. ; Corbula gallica, Lam.; C. striata, Desh. ; Gastrochena corallium, Sow. ; Serpula, sp.; and wood. F 8. Clays and clayey green sand, shown in a brick-field near the railway, where the beds exposed are :-— Liver-coloured and yellow clay laminated with white and green sand. A wedge-shaped bed of green sand. Nearly black foliated clay with iron pyrites, vegetable impressions, and wood, 18 to 20 ft. Fine waite Lower Bacsuor Sanp. At Hagthorn Hill, 2 mile to the north-west of the above brick- field, there is a pebble-bed 2 feet 6 inches thick, apparently in nearly the same relative position as that numbered 3 in the railway- * The letters in this and the following table refer to the table, p. 348. 350 H. W. MONCKTON ON THE BAGSHOT cutting section. The mass of stones is here such as to be worth quarrying. An almost precisely similar succession is shown in the series of shallow cuttings on the South-Eastern Railway, near Wellington College Station, which are shown in the annexed cut (p. 351). The following are the details :— Section on the South-Hastern Railway, near Wellington College. Gravel. B. Light yellow sand, with small patches of green sand, casts of shells, numerous but very imperfect, including species of Fusus, Natica, Phorus, Turritella, VOluiea, WOOA CEC. x cok cuneh easecaee ie hae eet aaa eee 50 ft. or more. C. A greenish sand, with two irregular lines of flint pebbles, and a few pebbles Of old: noek: bac ssdeedegs Sten cen eyes eat td Le eee ate eee 1 ft. 6 in. D. Yellowish sand, iron-sand concretions, and layers, a few casts of Turritella, ANC WOOK. :..c5 caten meet aerate ewan oa tere Thickness varies, 2 to 4 ft. Yellowish, reddish, or greenish sand, with numerons thin lamine of light- colombed: Clay | oie ea tisetnGeea Gusee asad. cna eeioe ee te ee ae About 15 ft. [This bed was until recently worked for bricks at Wellington College. | Dark-coloured laminated clay, with an irregular line of flint pebhiee 7: to 4 ft. pimeen sand. omy sccaccereceaeee oceaath canteen au eee Thickness not shown. . Dark laminated clay, with a little impure greensand. (This bed is shown in a pit at Upwick or Wick Hill near Finchampstead.) Lower Bagshot Sands, yellow clayey sand, with beds of stiff laminated clay, which are worked for brick-making at California near Finchampstead, and are there wrongly marked ‘“ Middle Bagshot” on the Geological Map. Fy The above examples establish, I think, with sufficient certainty, the succession of the strata in this district; and the well-sections at Wellington College and the Albert Asylum, Bagshot (Mem. Geol. Surv. iv. pp. 425, 537), are very similar. In alla pebble-bed appears at the top of the clayey beds, and affords good evidence of a break in the series at this point. With the exception of the Albert-Asylum well, there is no section through the whole of the upper sandy beds lettered A and B in the table of strata; and it is therefore not easy to ascertain their greatest thickness. That well gives 226 feet above the pebbles ; and I doubt whether this is exceeded in other places. In the lower beds of these upper sands numerous green grains occur, either in patches or disseminated through the sand, and casts of shells are to be found in several pits and cuttings. JI may men- tion Sandhurst cutting on the South-Western Railway, near Wel- lington College, Cexesar’s Camp, Easthampstead, a road-cutting about a mile from the Royal Military College, on the Windsor Ride, and the cutting through the spur of the Fox Hills (8.W. R.), from which Prof. Prestwich obtained several fossils. There are two very fine sections in the higher beds—the first at Crawley Hill, near Camberley, and the second at Tunnel Hill, on Pirbright Common. The Crawley-Hill cutting, on the South-Western Railway, near Camberley Station, is unfortunately overgrown ; but casts of shells were formerly very abundant there, though for the most part ill- 351 BEDS OF THE LONDON BASIN. Section on South-Eastern Railway at Wellington College. (Horizontal scale 4 inches to 1 mile. Vertical scale 150 feet to 1 in ch.) ep B te Se 28 é Bee 3 pegs ap 3 ep ads Be SO Be "20 s Si ae N. D Fa E ‘ Fa e . = : Peas i i Sh tt a. Upper Bagshot Sands (of Prestwich). f. The pebble-bed which marks the base of the Upper Bagshot. ; b. Light-coloured sands and laminated clays. g. Irregular line of pebbles in dark clay at the top of the green Se ome d with subordinate dark cl . : Bagshot. e. Green sand with subordinate dark clays. sand, _d. Clays, mostly dark-coloured. h, Hmbankments. e. Lower Bagshot Sands In the above cut the dip of the beds to the north of Wellington-College Station is too high ; they are, in fact, nearly horizontal, and are brought into the cutting by the fall of the line, which is here 1 in 154. Pebbles 1ike those in the pebble-beds cover the surface of the ground for some distance north of the station. I have omitted the gravels and surface-earth. 352 H. W. MONCKTON ON THE BAGSHOT preserved. They are apparently of nearly the same species as those at Pirbright. At Tunnel Hill, on the Woking-Aldershot branch of the South- Western Railway, the casts are well preserved, and impressions of the exterior of the shells are often to be found. The section is as follows :-— 1. Darkish yellow sand, passing into ochre sand, with casts of shells. 20 ft. 6 in. 2; Aline of bright yellow samd.seeycecce ont, muecterse'q.0saeedss ste eee e Ree eeee 8 in. 3. White sand, with numerous casts, passing into white sand, with patches of yellow sand, irregular iron-sand layers and concretions, numerous casts and impressions of shells, and a few flint pebbles..................... 27 ft. 4, Variously tinted sands, with very few casts of shells ......... 39 ft. exposed. Fossils from the Upper-Bagshot Sand, Tunnel-Hill Cutting, Pirbright, Surrey. ve Rostellaria rimosa, Sow. ve Pecten reconditus, Sod. c Terebellum fusiforme, Lam. Avicula media, Sow. ? Cancellaria (2 species). Pectunculus sp. Ancillaria canalifera, Lam. Nucula similis, Sow. Voluta, sp. (rare here, but very | vce Cardita sulcata, Sod. (variety with common at Crawley Hill). tuberculate ribs). Volvaria acutiuscula, Sow. c Crassatella sulcata ?, Sol. c Natica ambulacrum, Sow. ve Lucina mitis, Sow. sp., perhaps N. conoidea, | c divaricata, Linn. (Rigaulti- Sow. ana, Desh.). c labellata?, Lam. Diplodonta sp. ? fo) patula, Lam. Cardium porulosum, Brand. Sigaretus canaliculatus, Sow. c Protocardium parile, Desh. Cerithium sp.? turgidum ?, Sol. ve Turritella imbricataria, Lam. Cytherea, apparently 4 species. Turritella, or Niso, sp. ? ve Tellina scalaroides, Lam. © Littorina sulcata, Pik. Corbula gallica?, Lam. Solarium bistriatum ?, Sow. c Corbula ficus ?, Brand. plicatum ?, Lam. c pisum, Sow. Phorus (2 species). striata ?, Lam. Dentalium striatum, Sow. Clavagella coronata, Desh. (small, smooth species). Serpula extensa, Brand. Actzeon sp. ? sp. Bulla attenuata, Sow. Serpulorbis Marshii. —-— sp. Two or more species of corals. ve Ostrea flabellulum, Lam. Wood. Correlation of the Bagshot Beds with the Hampshire Series. A glance at the table of strata (p. 348) will show that there are two beds in which the occurrence of well-preserved fossils gives an opportunity for comparison between the Bagshot and Hampshire series. The first of these is the green-sand bed, EH; and the second the higher beds of the Upper Bagshot Sands, marked A in the table. Now, there can, I think, be little doubt that the green sand, H, must be correlated with some part of the Middle Bracklesham of the Hampshire basin ; and its great resemblance to the bed numbered BEDS OF THE LONDON BASIN. 353 11 in Prestwich’s Whitecliff-Bay section leads me to the conclusion that it should be placed in that portion of the Bracklesham termed Group C by the Rev. Osmond Fisher (Quart. Journ. Geol. Soc. vol, xviii. p. 65). The Bagshot bed resembles Prestwich’s bed 11 at Whitecliff Bay in colour and texture, in the abundance of Cardita planicosta and large Turritelle, and in the occurrence of Nummulites levigatus. Passing now to the upper fossil-bed at Bagshot, marked A, and best seen at Pirbright, I venture to submit that the shells clearly prove it to be of Barton age. The occurrence of such shells as Volvaria acutiuscula, Littorina sulcata, Dentalium striatum, Lucina divaricata (or Rigaultiana), Tel- lina scalaroides, and Clavagella coronata, all of which are well- marked types and not easily mistaken, taken together with the ab- sence of Cardita planicosta, Pecten corneus, &c., which are so abundant in the underlying green sand, is, I think, quite sufficient to prove that we are here in Barton beds. On the other hand, if we compare the list of shells from Pirbright Common with that from Long Mead End, Hordwell, published by Mr. Tawney (Proc. Cambr. Phil. Soc. iv, p. 150), we can, I.think, feel no doubt that the Long-Mead-End Upper Bagshot Sand, with Cerithium pleurotomoides, cannot be correlated with the Upper Bagshot Sand of Bagshot Heath. The upper beds of the Bagshot Sand of Bagshot Heath must therefore be correlated with the Barton beds of Hampshire and the Isle of Wight. It might be objected to this correlation that it does not ac- count for the absence of the Upper-Bracklesham beds in the London basin; but I think that, in the first place, the distance be- tween the two basins is enough to account for almost any amount of thinning-out or change in the nature of the strata; and in the second place, the clear evidence of a break in the London-basin series, Which is furnished by the remarkable pebble-bed marked C in the table of strata, is sufficient to account for the absence of the Upper Bracklesham in that basin. The only question remaining to be considered is the point at which the division between the Barton and Bracklesham should be placed in the Bagshot area. On the whole, it appears to me to be most convenient to place it at the pebble-bed marked C in the table of strata (this was suggested by the Rev. A. Irving [ Proc. Geol. Assoc. iv. pp. 334, 330 |, and is in accordance with Prof. Prestwich’s Golds- worthy section [Q. J. G.S. 111. p. 382 ]), giving the Barton beds a thick- ness of 226 feet at least, and the Bracklesham an average thickness of about 45 or 50 feet. In the Geological Survey Map some of the overlying sands are included in the Bracklesham, on the ground that they contain green grains. This, however, does not appear to me altogether satisfactory, a few green grains being no proof that a bed is not of Barton age ; and I think it better to take the pebbles as a division than the very uncertain line proposed by the Survey. 354. ON THE BAGSHOT BEDS OF THE LONDON BASIN. Discussion. Prof. Prestwich said it was forty years since the opening of a railway-cutting had first given us some idea of the position of these sands. The paper was a very careful record of observations. The fossils were so imperfect that comparison with the representatives of this series in Hampshire was very difficult. He himself had found but few. He thought that the so-called Upper Bagshots of that region were on the whole more probably synchronous with the Brackles- ham. The occurrence of a little green sand was not of much impor- tance, except that it sometimes was associated with occurrence of fossils. He asked how many species Mr. Monckton had found. Mr. J. S. Garpner said that he thought Mr. Monckton had made a sufficiently large collection to show that the beds were really equivalents of the Barton Beds. As for the lower beds, they were probably freshwater, but might perhaps rather belong to the lower part of the Middle Bagshot than to the Lower Bagshot. Prof. Jupp thought that the author had brought valuable evidence as to the age of the ‘‘ Upper Bagshots ” of the London basin, showing that they might be correlated with some parts of the Barton series. It was important to have shown that these beds did not agree with the Hordwell Beds or Headon-Hill Sands. There was, indeed, no reason for correlating the two series in the Hampshire and Bagshot areas. For himself he thought it was unsafe to attempt to draw exact parallels between beds sixty miles apart, so far as the minor members were concerned. Prof. Prestwich mentioned that he had found near Cooper’s Hill traces of casts of marine shells in Lower Bagshot. Mr. Moncxrton acknowledged the favourable way in which his paper had been received. He had obtained from the Upper Bagshot 28 species, and from the green-sand bed 18 species. It was easier to enumerate than to name the species. NEWER GNEISSIC ROCKS OF THE NORTHERN HIGHLANDS. 855 22. The Acs of the NEwER Gnetsstc Rocxs of the Nortuern Hien- tanps. By C. Cattaway, Esq., D.Sc., M.A., F.G.S. With Notes on the Lirmotoey, by Prof. T. G. Bonney, M.A., F.R.S., Sec. G.S. (Read May 9, 1883.) ConrTENTs. Introduction. I. The Formations concerned in the result (p. 357). a. Hebridean. 6. Caledonian. c. Assynt series. 1. Probable age. 2. Subdivisions. c,. Torridon Sandstone and Ben More Grit. > Quartzite. ec, . Seamy. ¢,gu. Annelidian. c,. Brown flags. cy. Salterella-grit and quartzite. cs. Dolomite. esl. Dark. ¢,u. White. II. The Relations between these formations (p. 363). Loch Broom (p. 363). . Section along the shore of the loch, from Ullapool to the S.S.E. . Section along the road, from Ullapool to the 8.H. . Section from Ullapool to the H.S.H. . Section from Ullapool over the high ground to the E.N.H. . Section along the south side of the Ullapool river, W. to E. . Section along the north side of the same, W. to EH. Assynt (p. 367). a. The “ Upper” Quartzite non-existent (p. 367). Section on the Burn of Calda. Scarp between Calda Burn and Poulan-drein. Section on Poulan-drein. G@uartzite of Cnoc-an-drein. “ Logan Rock” of Glasven. Section from the west end of Ben Uarran to the high road south of Inchnadamf. Section in the Balloch under Coniveall. Section up Glen Coul. B. The “Logan Rock” not separable from the Hebridean (p. 374). (a) Objections to the Hebridean age of the “ Logan Rock” considered. (1) Absence of accessory minerals in the “ Logan Rock.” (2) Absence of bedding in the same. (3) Foliation in the “ Logan Rock” not coincident with bedding. (4) Difference of weathering. (0) Author’s objections to the contemporaneity of the “ Logan Rock” with the Assynt series. (1) The “ Logan Rock” not conformable to the rocks below it. (2) Or to the rocks above it. (3) The strata below it not metamorphosed. (4) No beds of passage between it and the rocks above and below. (5) Alteration and crushing at its junctions with other rocks, (6) Strikes of the “Logan Rock” and Hebridean often concordant. OOP OO Ne 356 C. CALLAWAY ON THE NEWER GNEISSIC o. Direct proof of the overthrow of the Hebridean (p. 379). Section up the cascade from Dhuloch More. Section up the cascade from Dhuloch Beg. Section at the north end of Scounan More. Sestion across Coniveall. Section up Glen Coul. Ground between Coniveall and Glen Coul. Section on Camaloch. p. The “ Upper Limestone ” non-existent (p. 386). (a) The dolomite repeated. (6) Or a part of the Caledonian. gE. The Caledonian brought over the Assynt series by a reversed fault (p. 389). The Knockan section. Sections on the north and south sides of Glen Coul, r. Structure of Assynt (p. 392). Scounan, Brebag, Ben More, Ben Uarran, Cnoc-an-drein, Glasven, Ben Uie, Ben-na-Creisag, Ben-an-Uarran. The Stronchrubie basin. Loch Erriboll (p. 896). a. The “ Granulite” of Nicol a lower division of the Caledonian (p. 396). B. The Assynt series folded back upon itself (p. 397). Structure of Druim-an-tenigh. Ground between Druim-an-tenigh and the sections on Camas-an-duin. Sections (three) on Camas-an-duin, Section in the ravine above the Free Church. Section above Erriboll House. Section at Craig-na-faolin. Ground between Ben Arnaboll and Ben Heilem. Ground between Hope Ferry and Whitten Head. c. The Caledonian Gneiss brought over the inverted Assynt rocks by earth- movements (p. 403). Structure of Ben Arnaboll. Junction south of the Arnaboll valley. Junction north of Hope Ferry. Ground between Druim-an-tenigh and Ben Arnaboll. p. Outliers of the Assynt series on the Caledonian (p. 407). n. Granite not intrusive in the Assynt series (p. 408). III. Igneous rocks (p. 408). 1. Loch Ailsh group. 2. Igneous rock of the Quarizite. 3. Granite of Loch Erriboll and Durness. IV. Summary of results (p. 410). V. General considerations (p. 411). VI. Reply to objections (p. 412). Appendix. By Prof. T. G. Bonney (p. 414). INTRODUCTION. In the summer of 1880 I commenced an investigation into the relations between the fossiliferous limestone of Durness and the great gneissic series which was alleged to overlie it. A portion of my first results I communicated to the Society in a paper* in which * Quart. Journ. Geol. Soc. 1881, p. 239. ROCKS OF THE NORTHERN HIGHLANDS. SOL I endeavoured to show that the sections at Durness and in Assynt did not display a true ascending series, the limestone in both loca- lities being separated by a fault from the rock which was sup- posed to conformably succeed it. On the relation of the limestone to the quartzite I formed no opinion; but in the two sections de- scribed I found them brought together by faults. As the quartzite passed with apparent conformity below the eastern gneiss on the east of Loch Erriboll, I hinted, in accordance with the views of Murchison, which I had not then seen reason to abandon, that the quartzite might belong to the metamorphic series. From these limited observations I carefully abstained from drawing a general conclusion ; and I was conyinced that no satisfactory result could be achieved without a detailed survey. Accordingly I devoted a fortnight in 1881, and two months in 1882, to the districts which appeared most promising ; and | have now the honour to submit the conclusions to which I have been led. The published theories on the relation between the Durness lime- stone and the eastern gneiss are the following :— 1. That the eastern gneiss conformably overlies the limestone, and is of “ Lower Silurian” age (Murchison, Geikie, and most authors). 2. That the eastern gneiss is merely the Hebridean brought up east of the limestone along a line of dislocation and inversion (Nicol). 3. That the identity of the limestone of Assynt and Erriboll with that of Durness has not been demonstrated, and therefore that the eastern gneiss, though it conformably overlies the former, has not been proved to be of “‘ Lower Silurian” age (Heddle). The view which I have here to submit differs from all of the above, but approximates,most nearly to that of Nicol. I maintain, with that author, that the junction of the limestone with the eastern gneiss is a line of faulting and inversion; but I shall attempt to prove that this gneiss is a distinct series, newer than, and resting unconformably on, the Hebridean, that Nicol’s ‘‘igneous rock” overlying the limestone is usually a true gneiss, and that both the older and younger gneissic systems have been brought up over the limestone by great earth-movements. The eastern gneiss I propose provisionally to name the “‘ Caledonian.” I. Tue ForMATIONS CONCERNED IN THE RESULT. The groups whose relations are to be discussed are the Hebridean or Lewisian, the Caledonian, and the quartzo-dolomitic group which, in view of the doubt which has been cast upon its age by the re- searches of Prof. Heddle, I propose to call the Assynt series. a. Hebridean. In addition to the masses usually designated by this name, I in- clude under it a large proportion of the “igneous rocks” of authors. Recently the latter have been described by Dr. Heddle as a new Q.J.G.S. No. 155. 2p 308 C. CALLAWAY ON THE NEWER GNEISSIC group, conformably overlying the Dolomite; and he names them ““ Logan Rock,” from Glen Logan or Laggan, where they have been identified by Prof. Bonney as Hebridean. I shall submit evidence to the effect that by faulting, accompanied or followed by powerful lateral thrust, this rock has been thrown over on to the Assynt series to a maximum breadth of about a mile, and along an outcrop of over 15 miles in Sutherland alone. ‘This overlie often produces the appearance of SCHL between the Assynt series and the Hebridean. b. Caledonian. This system is divisible into two well-marked groups. The lower I have recognized only in the Erriboll area; and from its clear de- velopment in Ben Arnaboll, I have called it the Arnabol) series (6). It consists of grey, granitoid, very felspathic gneiss, overlain by dark, striped hornblende and mica gneisses, passing up through beds of an intermediate character into the ordinary flaggy gneiss which, from its exposure on Loch Hope and in the lofty peak of Ben Hope (3040 feet), I have designated the Hope series (b,). The latter is very uniform in its litholog gy wherever I have seen it between Loch Erriboll in the north, and Loch Broom, in Cromartyshire, in the south, and from this line eastwards to Lairg,in Sutherland, and Ben Wyvis, in Ross. It is normally a thin-bedded, highly quartzose eneiss, passing into a felspathic variety on the one hand, and into quartz schist on the other. Within the extensive area described, I have not been able to detect any evidence of the reappearance of the Hebridean. Near Lairg, it is true, the Shiness limestone, which is highly crystalline, is associated with bands of very massive horn- blende, together with bands and nests of mica, steatite, and several other minerals, all highly crystalline. From the general relations of these beds, I am disposed to regard them rather as a local varia- tion of the Caledonian, perhaps due in part to the presence of the limestone, than as an upthrust of the Hebridean. c. Assynt Series. 1. Probable Age.—Two chief objections are urged by Prof. Heddle against identifying the limestone of Assynt and Erriboll with that of Durness. (1) The absence of the Durness fossils in the Assynt group. Sir R. I. Murchison alleges that “an orthoceratite” was found in quartzite on Loch Erriboll, and ‘‘ orthoceratites” in the limestone near Inchnadamff. The Erriboll fossil was described by Salter, and com- pared by him with Orthoceras (Cameroceras) Brongnartit (Troost ?). It was given to Murchison by Mr. Clark, of Erriboll House, who fixed its locality. The Assynt fossils were discovered by Peach, and handed to Murchison on the spot. Mr. Peach has recently (1882), in a letter to ‘ Nature,’ positively reaffirmed the genuineness of the discovery. Unfortunately, none of these specimens can be produced. Many observers, including myself, have since diligently searched for fossils in the limestone, but without success. ROCKS OF THE NORTHERN HIGHLANDS. 359 (2) Difference of chemical composition. Dr. Heddle quotes ana- lyses of the Assynt and Erriboll limestone by Dr. Thomas Anderson, published in the ‘ Transactions of the Highland Society,’ which show that the rock is a “typical dolomite,” while the Durness limestone is a “fairly pure” carbonate of lime. I confess myself not fully satisfied with Dr. Heddle’s objection. The chemical argument would certainly be of force if it stood alone; but it would have little weight against the merest fragment of fossil proof. The evi- dence of the orthoceratites, is certainly not decisive; but at the same time I hesitate to reject the testimony of such competent ob- servers as Murchison, Peach, and Salter. That the specimen de- scribed by Salter was truly an orthoceratite no one who is acquainted with the work of that accomplished palzontologist can reasonably doubt; and that Mr. Clark palmed a deception upon Murchison is not easy of belief, both from the want of motive and from the diffi- culty of obtaining the specimen. But,even waiving the fossil testimony just discussed, there are other facts which militate against the Archean age of the Assynt series. Foremost amongst these I would place the comparatively unaltered. state of the rocks. The quartzite bears but little upon the point, since we have in the British Islands undoubted quartzites as young as the Llandovery and as old as the Dimetian. But the beds above the quartzite display very slight alteration. In the Brown Flags are shales and slates as unchanged as any of Lower Palzozoic age (Nos. 103, 104, 107, 108, p. 418). In Assynt the Flags contain abundant vegetable impressions, the carbonaceous matter being as distinct and apparently as unaltered as in coal-shales. The unme- tamorphosed state of these beds cannot be accounted for on any principle of selective metamorphism. The rocks furnish all the materials for the production of schists. Magnesia is provided by dolomitic flags and limestones. Iron is so abundant that it gives the flags their distinctive colour. Alumina and alkalies are sup- plied by argillaceous, felspathic, and micaceous constituents. Yet with all the ingredients at hand, not a seam of true metamorphic rock has been formed. It is not here contended that the absence of metamorphism is absolute disproof of great antiquity; but as all Archean rocks hitherto studied are more or less metamorphosed, the presumption that an unaltered rock is not Archean is very strong indeed. The abundance of fossil remains is also unfavourable to the Archean hypothesis. The upper part of the quartzite is full of the well-known “‘ worm-holes.” In the same beds are numerous conical bodies, with a cone-in-cone structure. In transverse section, as they appear on the surface of the beds, the fossils are like three concentric rings, the central one being in diameter about % inch, and the outer ring 1 inch. These peculiarities are hardly such as we should expect in Annelids. Besides the above, there are the “ Fucoids” of the Brown Flags. Some of these are described by Nicol as ‘‘ straight, cylindrical fragments, like stems or branches of trees .... Some are marked with obscure scars, as of leaves and 2D 2 360 C. CALLAWAY ON THE NEWER GNEISSIC branches. Round or oval markings, like those on the Stiqmaria of the coal, are also not uncommon.” This description appears to me strictly correct; and, in addition, I have observed in well-preserved specimens that the stems, which are hollow in the centre, present the appearance of several distinct concentric coats. Whatever this structure may mean, it appears clear that the fossils were true plants, and that they were of by no means the lowest type. I can hardly think but that woody tissue was present in these stems. Salterella (Serpulites) Maccullochit .will complete, so far as I know, the list of fossils. Salter describes these minute forms as “short, subconical, and curved tubes, of thick substance, and with but a slender central perforation;” and he compares them to Ditrupa. I have, however, obtained specimens in which a distinct cone-in-cone structure is visible, as in Piloceras of the Durness lime- stone. Transverse sections show the edges of three concentric septa (?), as in Salter’s plate* (fig. 21), and as in the conical fossils described above. These puzzling little things appear to be more like Cephalopods with conical septa than Annelids or Pteropods; but their small size, under 1 inch, suggests a doubt.. As we know of no undisputed fossil remains in undoubted Archean rocks, the facts just enumerated seem to bear against the Archean age of the Assynt series. On the whole, I see no sufficient reason to deny the received view of the correlation of this group ; but, since a doubt remains, I have preferred to use a local desig- nation. 2. Subdiviscons.—As I shall have to submit evidence to prove the frequent inversion of the Assynt series, it is of the first importance that we should ascertain the order of its subdivisions in their true position. I must here assume, what I shall hereafter attempt to prove, the non-existence of the so-called “‘ Upper Quartzite” and “Upper Limestone” of Murchison. The chief horizons which I have recognized are the following :— c,. Torridon Sandstone and Ben More Grit—The ordinary Tor- ridon is too well known to need further description ; but the variety which I have called “‘ Ben More Grit” requires special notice. A little below the top of Coniveall (3234 feet), the western peak of Ben More, is a band of grit and conglomerate, hitherto grouped with the “Upper Quartzite,” of which it forms the base. The lowest beds of this series, ag seen in a precipice 600 or 700 feet below the summit in the spur which projects to the south, are greenish con- glomerates. The included fragments, which are well rounded, are mainly of quartz and gneiss, the former sometimes reaching a diame- ter of 6 or 7 inches, and are imbedded in a chloritic matrix. These basement conglomerates pass up into a green grit (No. 94, p. 417), alternating with finer green sediments. Mr. G. H. Bailey, B.Sc., who accompanied me in 1881, descended from the peak by a route which afforded him a clearer section of the higher beds than the line followed by myself in 1882 ; and he states that he ‘‘ came upon Tor- ridon, somewhat changed in aspect, after about 300 feet of descent. * Quart. Journ. Geol. Soc. 1858, pl. xiii. ROCKS OF THE NORTHERN HIGHLANDS. 361 It was made up of a series of beds of the usual coarse sandstone with intercalated thin bands of shale, the grit-beds being generally about 6-10 feet, and the shale from 6 inches to 1 foot.’’ Mr. Bailey esti- mated the Torridon at about 300 feet. This would make the base of the series about 2600 feet above the sea, which agrees nearly enough with my estimate of 2500 feet, taken by aneroid at a time of considerable atmospheric disturbance. Judging by a large number of blocks of red grit scattered over the gneiss slopes below the con- elomerate, there can be no doubt that beds of this rock occur in the precipices above. The grit and conglomerate are continued down the Oykel valley for three or four miles; but the thickness I found not greatly to exceed 100 feet. The red variety is typically a quartzo-felspathic grit (Nos. 92, 93, p. 417). It occurs abundantly in the localities on the Oykel. It is important to observe that these rocks are truly fragmental. They are composed of material which might have been derived from the Hebridean. In Ben More the conglomerate and grit rest in horizontal beds upon the nearly vertical gneiss which forms the nucleus of the mountain, and are overlain by the Ben-More quartzite, which I regard as merely a faulted repetition of the quartzite under the Dolomite. It would therefore appear that the grit represents the Torridon sandstone. There are, indeed, slight differences, the Ben- More grit being frequently green in colour, its particles being less often rounded, and its general aspect more strongly suggesting deri- vation from gneissic rocks. Whether-or not this group is exactly contemporaneous with any part of the Torridon is immaterial. It is sufficient for my purpose that it lies above the Hebridean and below the quartzite. c,. Quartzite.—This series is tolerably homogeneous from top to bottom; but there are certain differences which render possible a separation into two divisions, a lower (c, 1.) and an upper (¢, u.). Seamy Quartzite, c,1.—The base of the quartzite is characterized by the occurrence ‘of, thin seams of shale or grit and bits of red felspar, and, in Assynt, by the intrusion of beds or masses of felsite or diorite. The felspar grains are seen throughout Assynt at this horizon; and the seamy appearance is, so far as I have seen, per- sistent along the whole line as far as the North Sea. Indeed the uniform character of the quartzite is remarkable. About 10 feet from the base is a thin seam of quartzose grit, which appears at the same horizon at points over 30 miles distant on the strike. The basement of the series in Assynt is often a thin band of conglo- merate of quartz in a brownish matrix ; and a similar seam occupies the same position on Loch Erriboll. Towards Whitten Head the partings expand, so as sometimes to reach a thickness of several feet. Annelhidian Quartzite, ¢, u. (Pipe-rock).—The abundance of vertical burrows has been noticed by most writers. I have been unable to detect them in the lower series; but they characterize the upper part wherever I have studied the group—that is, from Loch Broom to Loch More, a distance of 35 miles, and on Loch Erriboll, 55 miles 362 C. CALLAWAY ON THE NEWER GNEISSIC from Loch Broom. To these facts I have been able to discover no exception in sections in which the rocks were undoubtedly unin- verted. Throughout this paper I avoid the use of the terms “‘ Lower ” and ‘‘ Upper” in reference to the quartzite, in order to prevent any confusion which might arise from the Murchisonian meaning of the words. T have seen no reason to believe that the quartzite ever largely exceeds 300 feet. The seemingly great thickening in certain loca- lities is, in my opinion, due to repetition by folding or faulting. c,. Brown Flags (Fucoid beds).—The rocks of this zone are very varied. The predominant type is a fine-grained arenaceous flag, which is sometimes argillaceous and sometimes dolomitic. There are also soft, thin-bedded, argillaceous shales, dark slates, and thin beds of dolomite and quartzite. The series, whatever its composi- tion, is distinguished by the abundance of iron, which, by its per- oxidation on weathered surfaces, gives rise to the characteristic rusty-brown colour. I have found this band persistent from Loch Broom to near Whitten Head. On Loch Broom the thickness is about 30 feet; but in Assynt it has expanded to 100 feet, retaining about the same thickness, or rather less, on Loch Erriboll. In the last locality it becomes more arenaceous, its upper beds passing into a kind of im- perfect quartzite, in which siliceous seams, weathering out sharply from vertical surfaces, alternate with softer material. This variety graduates upwards into the next band. c,. Salterella-Grit* and Quartzite—This zone is distinguished by the abundance of specimens of Salterella Maccullochii. On Loch Erriboll, where alone the fossils have been found, the beds are about 15 feet thick. Some of the strata are an iron-stained quartzose grit; but quartzite proper also occurs. The Salterella abounds in both kinds of rock, but is in better preservation in the grit. In Assynt 10 feet of quartzite holds a corresponding position between the flags and the dolomite; and on Loch Broom thc quartzite has considerably expanded, the flags being much thinner; so that the ageregate thickness of flags and Salterella-quartzite is about the sameé as in Assynt. c;. Dolomite——Two years ago, before any doubt of the identity of the Durness and Erriboll limestones had been expressed, I was struck with the fact that the latter was but slightly susceptible to the action of hydrochloric acid; so that I was quite prepared for Dr. Heddle’s announcement of their chemical difference. I have re- cently applied the acid test to sixteen specimens, taken from widely separated localities between Ullapool and Erriboll. Fifteen of them exhibit but very slight effervescence, except along joints, where de- composition may be supposed to have supervened. ‘The sixteenth, from Loch Broom, where it was in contact with Hebridean gneiss, effervesces more freely. But in this case chemical reactions have * Since this paper was written, an article has appeared in the ‘ Geological pane by Prof. Lapworth, F.G.S., in which the same name is proposed 1S ZONE, : ROCKS OF THE NORTHERN HIGHLANDS. 363 apparently taken place. The gneiss is chloritic, its silica having probably taken up a part of the magnesia of the dolomite; and the dolomite, thus partially decomposed, displays imperfectly the reaction of calcic carbonate. On the other hand, the Durness limestone, so far as I have seen, always effervesces freely with acids. The chemical distinction between the Durness and Erriboll lime- stones, pointed out by Dr. Anderson and reasserted by Prof. Heddle, thus appears to me to be firmly established. In Assynt I observed that the lower half of the Dolomite was of a dark grey colour, sometimes weathering almost black, while the upper part was nearly white. The same differeuce is equally well marked on Loch Erriboll. This separation into a Lower (c,1.) and an Upper (c, u.) Dolomite is useful as an aid in determining whether or not inversion has taken place. The maximum thickness of the dolomite is about 300 feet, or rather less. This estimate does not materially vary anywhere on the line of strike between Ullapool and Erriboll, the local changes being due, I believe, not to thinning but to faulting. II. Tae RELATIONS BETWEEN THESE FoRMATIONS. I describe the districts examined from 8. to N., since the evidence I have to adduce grows progressively stronger in that direction. Loca Broom. As this is one of the districts in which “a gradually descending series” has been affirmed, I devoted to it in 1881 a very close ex- amination. If there is no break here, we ought to find the same succession whereyer we run our sections; for, within so limited an area, variations of thickness along the strike cannot be considerable ; and it is of course impossible that this cause, though it may account for the absence of a group, should interfere with the true order. Of the sections examined I will describe six. These I take from S. to N. 1. Section along the shore of the Loch from Ullapool to the S.S.E. At about half a mile from the hotel we come to sandstone and conglomerate, which occupy the shore for a considerable distance ; but in the rounded promontory which terminates in Corry Point a gneissic rock suddenly comes in. ‘This is the ‘‘ serpentine or felspar porphyry ” of Nicol; but in the paper of Sir R. I. Murchison and Prof. Geikie* it is described as “ serpentinous and felspathic rock containing pebbles of jasper,” and it is regarded as “a highly meta- morphosed band of felspathic grit.” It is the “‘ Logan Rock” of Dr. Heddle. As it plays a very important part in our inquiry, it is necessary to examine its true nature very carefully. I first made the acquaintance of the “ Logan Rock” on the high ground about half a mile due east of the hotel. It was a greyish * Quart. Journ, Geol, Soc. 1861, p. 185. 364 C. CALLAWAY ON THE NEWER GNEISSIC ageregate of felspar and quartz, without foliation ; and, not suspect- ing any thing wrong, I took it to be “ porphyry,” though there was something about it which suggested the granitoidite with which I was familiar in Wales and Shr opshire. Soon after I found distinet foliation, dark crystalline bands striking to the north-west. This was puzzling ; but as, at a short distance off, the rock was seen clearly to overlie the dolomite, I clung to the “igneous” theory. Coming down to the present section new light began to dawn. This formation occupies the shore of the loch for about a quarter of a mile, filling in the entire break between the sandstone and the Caledonian, no trace of quartzite or dolomite being visible, though bare rock is exposed almost the whole distance. The Caledonian crops up on the flat beach in contorted beds, dipping on the whole easterly, and ending abruptly on the west. Fifty yards from it we come to a green felspathic rock, without definite structure, but looking like partially decomposed gneiss. This passes towards the north-west into a true gneiss (No. 74, p. 416), composed mainly of reddish felspar, quartz, and a greenish mineral, the folia either dip- ping E.N.K. at a low angle, or rising nearly to the vertical, and stri- king to N.N.E. The resemblance of this gneiss to some of the Mal- vern and Wrekin types was very marked. Its appearance is widely contrasted with that ofthe Caledonian. The foliation is more mas- sive ; the crystallization is much coarser; the rock is quite destitute of that fissile structure which causes ordinary Caledonian to split up into flaggy pieces, and it is much tougher under the hammer. Its structure and behaviour was altogether that of the Hebridean, to which I referred it; andthe opinion then formed has been confirmed by a long study of the rock in more northerly localities during two successive summers, as will hereafter appear. Continuing along the shore to the N.N.W., we pass abruptly from the gneiss tothe conglomerate. By Murchison and Geikie it is stated that there is a gradual transition from the one into the other ; and hence it is inferred that the gneissic rock is a ‘‘ metamorphosed band of felspathic grit.” The section described by these authors 1s undoubtedly the one studied by me; but, after careful examination, I could not detect the slightest evidence of a gradation. ‘The line of junction is perfectly sharp. The gneiss (No. 63, p. 416), which mainly consists of felspar and a little quartz, and is coloured green (apparently by chloritic products of decomposition, as is commonly the case at faulted junctions), is in actual ‘contact with conglomerate and grit. The gneiss and the conglomerate are plastered together by the chloritic mineral, but are readily separable by a slight blow of the hammer. The plane of junction, which displays slicken- sides, is nearly vertical, the gneiss slightly overhanging the conglo- merate. ‘he bedding in both rocks is very obscure. In the same locality, a little way up the slope, the gneiss rests on quartzite; and close by the conglomerate rests on quartzite. My interpretation of these facts is, that by faulting, supplemented by lateral squeeze, older deposits, Hebridean and Torridon, are brought up and thrown over onto the quartzite. ROCKS OF THE NORTHERN HIGHLANDS. 365 2. Section along the road from Ullapool to the S.E. Near the hotel the Torridon is succeeded by the Quartzite, dipping E.N.E. at 15° and conformably overlain by Brown Flags and the quartzite which holds the place of the Salterella-zone. At the bridge and fall, red sandstone, undistinguishable from the Torridon, suddenly comes in; and close by is a small exposure of dolomite, with a dip of 10-15° to E.N.E. As the place of the latter is on the high ground, 200 or 300 feet above, there must here be both a faulting-up of the Torridon and a faulting-down of the dolomite. ‘Torridon then re- appears, overhanging the road for some distance, and is overlain by quartzite, the relations between the two being obscure, owing to the absence of clear bedding. The quartzite is apparently overlain by the Caledonian. This section differs materially from the preceding, though they are drawn along lines which diverge from each other not more than 200 or 300 yards at the maximum. ‘The level of the present section, however, is about 100 feet higher. The Hebridean, though con- spicuous on the shore close at hand, does not appear ; and there are no signs of the inversion of the Torridon on the Quartzite. The Dolomite (which, on the received hy pothesis, should intervene be- tween the Quartzite and the Caledonian) is also wanting in its proper place. 3. Section from Ullapool to the E.S.E. (fig. 1). This section starts from the same point as the last, but keeps the top of the slope above the road. Passing over the Torridon, Quartzite, Brown Flags, and Salterella-quartzite, we come to the Dolomite, which rises in dip, and to the east is suddenly contorted for a breadth of about twenty yards into a regular series of sharp folds, having evidently been crumpled up against the Hebridean, which appears near at hand. Beyond the Hebridean we soon reach grey thin-bedded gneiss (Caledonian) with undulating and unde- cided dips. Fig. 1.—Section from Ullapool to the E.SE. W.N.w. f fi <= . BRYOZOA FROM AUSTRALIA. 433 This only differs from Catenicella pulchella, Maplestone* (after- wards described by MacGillivray f as Catenicella concinna), in having a large pore below the aperture instead of a notch in the aperture. But we often find the suboral pore close up to the aperture; and I believe that we shall find that the notch is represented by this suboral pore; I therefore had much hesitation in giving a new name, which I anticipate will not be permanent. Loc. Fossil: Muddy Creek, Bird Rock, and Waurn Ponds. 14. CELLARIA MALVINENSIS, Busk. 15. CELLARIA OVICELLosA, Stol. 16. CELLARIA ANGUSTILOBA, Busk. Melicerita angustiloba, Busk, Quart. Journ. Geol. Soc. vol. xvi. p. 261. Cellaria angustiloba, Waters, Foss. Chil. Bry. from Mt. Gambier, Q. J. G.8. vol. xxxvii. p. 260, pl. ix. figs. 28, 29, 30. Loc. Fossil: Mount Gambier (Woods & Wat.), Orakei Bay (Stol.), Bairnsdale (W.), Muddy Creek, and second specimen from Muddy Creek or Bird Rock. 17. CELLARIA PERAMPLA, Waters. Cellaria perampla, W aters, Foss. Chil. Bry. from Mt. Gambier, p. 260. Loc. Mount Gambier, Waurn Ponds. 18. Canpa Fossitis, Waters. 19. ScRUPOCELLARIA SCABRA, Van Beneden. 20. MEmMBRANIPORA MACROSTOMA, Rss. For synonyms see ‘§.W. Victoria,” p. 323. This is closely allied to Membranipora roborata, Hincks; but the zoecia of I. macrostoma are twice the size of those of VM. roborata. 21. MremBRanrpora RoBORATA, Hincks. Membranipora robsrata, Hincks, Gen. Hist. of Mar. Poly., Ann. & Mag. Nat. Hist. ser. 5, vol. viil. p. 69, pl. 11. fig. 3. The fossil fragment from Waurn Ponds is very small and occurs with only one layer, or, as we may say, in the Hemeschara form; and a recent specimen in my possession from New Zealand, which was sent me by Miss EH. C. Jelly as MW. roborata, has also only one layer, and, like the fossil, only one avicularium above the zocecium, not two, as described by Mr. Hincks. In the recent WM. roborata a radicle tube from the base of each external zocecium unites with those of the neighbouring zocecia to form “the thickened rib along the margin,” as described by Mr. Hincks. Jn the fossil there are four zocecia in a longitudinal row. Opesia 0°18 millim. long. Loc. Living in bilaminate condition off Curtis Island (H.); in uni- laminate condition, New Zealand. Fossil: Waurn Ponds. * “New Species of Polyzoa,” Journ. Micr. Soc. of Victoria, vol. i. nos. 2, 3, p- 64, pl. v. fig. 4. Tt “On some new Species of Catenicella,’ Roy. Soc. Vict. 1880. 434 A. W. WATERS ON FOSSIL CHILOSTOMATOUS 22. MEMBRANIPORA LUSORIA, var. coARcTATA. Plate XII. fig. 20. Membranipora lusoria, Waters, Foss. Chil. Bry. from 8.W. Vic- toria, p. 324, pl. xiv. fig. 14, pl. xvii. fig. 82. The specimen from Waurn Ponds has two large bosses nearly meeting across the opesia (aperture). The figure is by an accident drawn slightly too large. This is allied to Cellaria cactiforms, d’Orb. (Pal. Fr. pl. 651. figs. 1-4). Loc. Fossil: Waurn Ponds. 23. MeMBRANIPORA ARTICULATA, Waters. Membranipora articulata, Waters, Foss. Chil. Bry. from Mt. Gam- bier, p. 264, pl. vil. figs. Lb, 16. 24, MemBRANnrpora ocuLata, Busk, Plate XII. fig. 22. Nellia oculata, Busk, Cat. Mar. Pol. p. 18, pl. Ixiv. fig. 6, pl. Ixv. fig. 4. ”"Nellia oculata, Smitt, Floridan Bryozoa, p. 3, pl.i. figs. 58, 54. Nellia oculata, MacGillivray, Nat. Hist. Vict. decade v. p. 51, pl. 49. fig. 5 Tn the fossil the avicularia are larger and more raised than in any recent specimen which I have seen, and there are four spines above the aperture of each zocecium, on which account it should perhaps be called var. spinosa. Length of opesia 0°15 millim, Loc. Living: Torres Straits, 9 fathoms (B.); Bass’s Straits (L’homs.) ; Florida, 13- 138 fathoms (Sm.) ; Queenscliff, parasitic on alge and zoophytes (MacG.); Holborn Island, Queensland (Haswell) ; Cape Grenville, N.E. Australia, 20 fathoms (4. W. W. coll.); Piper Islands, 9 fathoms (A. W. W. coll.). Fossil: Waurn Ponds. 25. MeMBRANIPORA ARETHUSA, dOrb. Plate XII. fig. 19. Eschara arethusa, @’Orb. Pal. Frang. p. 127, pl. 666. figs. 4—6. Eschara actea, d’Orb. loc. cit. p. 116, pl. 662. fig. 17. Eschara allica, d’Orb. loc. cit. p. 125, pl. 665. figs. 8, 10. The specimen from Muddy Creek has 12 longitudinal rows of zocecia in a compressed branch in the Hschara-form. The zocecia are of the same size as those in a specimen I collected from the Cre- taceous beds of Royan (France), and the aperture (opesia) is also of the same size and in a similar position; the sides of the zocecia, however, are somewhat straighter and not so much contracted below as in the Chalk specimen. In the Royan specimen there is a zocecial avicularrum (onychocellaire), while I do not find one in the Australian fossil. This is also allied to Semieschara disparilis, d’Orb., and many other species described by d’Orbigny. Width of the opesia 0°16 millim., length of opesia 0°25 millim. Loc. Cretaceous of France. Muddy Creek. ~ BRYOZOA FROM AUSTRALIA. 435 26. MicropoRA ORDINATA, Sp. nov. Zoarium in Eschara-form (flat piece about 3 inch square). Zocecia flat, separated by a wide margin, depressed, especially at the distal end, oblong, arched above, with few large granulations and few pores. Oral aperture close up to the distal end, rounded above, straight below, with a raised lip. It would seem to be a Micropora and not a Steganoporella, but the two genera are difficult to distinguish in the fossil state. This seems to be somewhat similar to Eschara verrucosa, T. Woods (non Peach), “On some Tert. Austral. Polyzoa,” Tr. Roy. Soc. N. 8S. W. 1876, p. 2, fig. viii. ; but it differs in not having the aperture about the middle of the zocecium as figured, and also in the bsence of marginal pores. Aperture 0°3 millim. wide. Loc. Waurn Ponds. 27. MicRopora CAVATA, Sp. Nov. Zoarium in Hschara-form foliaceous. Zocecia hexagonal, the area round the aperture very much depressed. The aperture is about a third of the length of a zocecium from the distal end, semi- circular, with avery distinct lip directed vertically upwards, streng- thened at each end by a thickening. Above the aperture the cell- wall is often much thickened, forming a hood to the aperture. Width of aperture 0°25 millim. Loc. Waurn Ponds. 28. MoNnopoRELLA SEXANGULARIS, Goldf. Eschara sexangularis, Hagenow, Maestrichter Kreide, p. 81, pl. x. figs. 3, 4, 5. Eschara Clarke, T. Woods, ‘“‘ On some Tertiary Australian Poly- zoa,” Tr. Roy. Soc. N. 8. Wales, 1876, p. 2, figs. iv.—vii. This seems to be the same species as one I have from Maestricht which I have called sewangularis, though with a little doubt, as I can- not find any zoccial avicularia. The zoccia from Bird Rock are slightly larger than those from Maestricht ; but the aperture is of the same size. Ido not understand what Mr. Woods means by “ pore for the avicularium upon the summit,” as there is nothing of the _ kind in my specimens and he does not show it in his figures. This would seem to be allied to Semzeschara disparilis, d’Orb. (Pal. Fr. pl. 709. figs. 9-12). In a specimen from Bird Rock a small mdentation on each side of the base of the aperture shows us that we are dealing with the oral aperture and not with an opesial opening; and it is therefore placed with Monoporella. Aperture 0°37 millim. wide. Loc. Muddy Creek (Woods, and Wilson’s coll.). Bird Rock and Waurn Ponds. 29. MonoporELLA CRASSATINA, Waters. Monoporella crassatina, Waters, On Foss. Chil. Bry. from Mt. Gambier, Quart. Journ. Geol. Soc. vol. xxxviii. p. 270, pl. vii. fig. 8. Loc. Mt. Gambier (W.), Napier, New Zealand (in Miss Jelly’s collection as recent Tertiary). Waurn Ponds. 436 A. W. WATERS ON FOSSIL CHILOSTOMATOUS 30. STEGANOPORELLA MAGNILABRIS, Busk. 31. MEMBRANIPORELLA NITIDA, Johnst. Membraniporella nitida, Hincks, Brit. Mar. Pol. p.200, pl. xxvii. figs. 1-8. Lepralia eximia, Seguenza, Le Form. Terz., R. Accad. dei Lincei, anno CCLXxviI. 1880, p. 203, pl. xiv. fig. 23. In the fossil from Waurn Ponds the zocecia are inclined to be hexagonal. The mesial line of the zocecium is carinate, with about 9 ribs on each side. Width of the oral aperture 0:14 millim. This 1s elosely allied to Plhiophlea sagena, Gabb & Horn, from the Cretaceous of New Jersey. Loc. Living: Britain, Northern Seas, Roscoft, New Zealand (Hutton) ; Capri from 225 fathoms (4. W. W. coll.). Fossil: Zan- elean of Calabria (Seg.), Waurn Ponds. 32. CRIBRILINA TERMINATA, Waters. Plate XII. fig. 17. Cribrilina terminata, Waters, Bry. from S8.W. Victoria, p. 326, pl. xvii. fig. 68, and Chil. Bry. from Bairnsdale, p. 507, pl. xxii. fig. 6. In one specimen from Muddy Creek there are several large zoce- cial avicularia directed downwards. By the side of one zocecium there are also narrow, almost acicular avicularia directed laterally. Loc. Fossil: S.W. Victoria, Bairnsdale and Muddy Creek. 33. CRIBRILINA TUBULIFERA, Hincks. Oribilina tubulifera, Hincks, ‘“‘ Gen. Hist. Mar. Polyz.,” Ann. & Mag. Nat. Hist. ser. 5, vol. viii. p. 56, pl. 1. fig. 7 (1881). The fossil has an aperture 0°12 millim. wide, grows incrusting shells, and seems to be distinct from my C. suggerens. Loc. Living: Bass’s Straits (H.). Fossil: Muddy Creek. 34, MucroneLLA MUCRONATA, Sm. 35. MicropormLLA VIOLACEA, Johnst., var. rissa, Hincks. Microporella violacea, var. fissa, Waters, Bry. from S.W. Vict. p. 329, pl xy. fie) 26, pl. xvii die: 7a: 36, MicRoPoRELLA ELEVATA, T. Woods. 37. MiIcROPORELLA YARRAENSIS, Waters. 38. MicroporELLA coscrnopora, Rss., var. ARMATA, Waters, loc. cit. p. 331, pl. xv. fig. 25. 39. MicRoPoRELLA SYMMETRICA, Waters. Microporella symmetrica, Waters, Foss. Chil. Bry. from 8.W. Victoria, p. 332, pl. xviii. fig. 83. From Muddy Creek there are several pieces of straight compressed branches about 1°5 millim. broad. In one case there is a large avicularium directed upwards at the side of the zoarium. BRYOZOA FROM AUSTRALIA. 437 40. MicroporELita mALusi, Aud. Only two zocecia from Bird Rock. Loc. Living : European Seas, 8. America, New Zealand, Australia. Fossil: English Crag and Pliocene of Italy. 41. MicrRopoRELLA cELLULOSA, MacG., form ADEONA. Dictyopora cellulosa, MacGillivray, Trans. Roy. Soc. Vict. 1868. Adeona cellulosa, Kirchenpauer, ‘‘ Ueber die Bryozoen-Gattung Adeona,” Journ. Mus. Godeffroy, 1879, p. 10. Dictyopora cellulosa, MacGillivray, Nat. Hist. Vict. decade v p. 37, pl. 47. fig. 1, & decade vii. pl. Ixvi. fig. Le. From the examination of some recent specimens, I am convinced that we should place this with Muzcroporella. Although I was unable to prepare out any opercula, yet sections showed me that the proximal edge of the oral aperture was straight, and there are two contractions near the base of this aperture. The oral aperture is at a slight distance from the surface, and what has been described as a round aperture is really a peristome. In recent specimens of WM. cellulosa and albida the suboral pore is round; but in the fossil, though most pores seem to be round, yet there are a few slightly elongated and denticulated. The interior of the shell is hollowed out round the region of the oral pore. M. cellulosa is closely related to M. violacea, var. fissa, a species which shows great variation both in the size of the avicularium and of the elevation of the front of the cell. In small pieces, such as the fossils in question, it is very difficult to distinguish between V. cellulosa, M. grisea, and M. albida. In M. albida, form Adeona, there are large zoccial avicularia near the borders of the fenestre, which have the same form as the smaller avicularia on the front of the zocecium ; and we may ask if this is not a case of the avicularium having enormously developed at the expense of the polypide. Perhaps this may be the way in which many zoccial avicularia have originated. Those who dredge these species should examine to see whether the frequency of these large avicularia depends upon any special condition, such as depth or the nature of the sea-bed. Loc. Living: Queenscliff. Fossil: Muddy Creek. 42, Portna coronata, Rss. This occurs from Muddy Creek in what I call the 6 form (Bry. S.W. Vict. p. 334) and from Bird Rock in the ¢ form. 43. LepraLia PERTUSA, Esper. 44, PorELLA MARsUPIUM, MacG. Lepralia marsupium, MacG. Trans. Roy. Soe. Vict. 1868, and Nat. Hist. of Victoria, decade iv. p. 22, pl. 35. fig. 4. Porella marsupium, Hincks, Contr. towards Gen. Hist. of Mar. Pol., Ann. & Mag. Nat. Hist. Aug. 1881, pl. i. fig. 6. Porella marsupium, MacGillivray, “Descriptions of new or little- known Polyzoa, pt. 11.,” Tr. Roy. Soc. Vict., 1882, extra page. ®. J. 6G. 8;: Nov 15, Qt 438 A. W. WATERS ON FOSSIL CHILOSTOMATOUS Mr. Ridley, in his description of a zoological collection from the Straits of Magellan, describes a species as Schizoporella marsupium, MacG.; but this Mr. MacGillivray has since called S. Ridleyz. Width of oral aperture of the fossil 0-07 millim. Loc. Living: Victoria (MacG.); Bass’s Straits (H.). Fossil Waurn Ponds. 45. Suirria Tarri, T. Woods. 46. Surrrra RETICULATA, MacG. 47. Surrtra Napreri, nov. Plate XII. fig. 14. Zoarium incrusting. Zocecia small, ovate, with very slight peri- stome, large punctures round the edge; oral aperture nearly flat on the lower edge; an opening (probably avicularian) just below the aperture almost into the peristome. Large acute avicularia, di- rected outwards, placed nearly halfway down the zoccia. Ovicell raised, globose. I have a specimen, fossil, sent me by Miss Jelly, from “ recent Tertiary,” Napier, New Zealand, and as it is better preserved than the one from Waurn Ponds, I have figured it. Loc. Fossil: Napier, New Zealand; Waurn Ponds. 48, Suirrra cottaRts, Norm., var. Plate XII. fig. 10. Thavealways found the greatest difficulty in distinguishing between Phylactella and Smittia, and have already expressed my doubts as to the advisability of using the shape of the peristome as a generic character ; and the present form, which is closely allied to, if not identical with, Phylactella collaris, Norm., has decided me to only . use the name Smitiia for what are looked upon as belonging to these two genera. The peristome is variable, and is thicker and higher at both sides, thus forming a depression in its proximal part; the distal part of the peristome is also separated, and rises in a tongue-shaped form. The peristome of a recent species, which I collected in Rapallo, N. Italy, and which I consider the same as Leprala obeliscus, Manz., and also perhaps as L. Gotriana, Rss., has a similar tongue-shaped distal portion. The surface of the fossil is granular. Aperture 0-17 millim. wide. Occasionally the peristome is entire, when there is little difference from that of the British P. collaris; the peristome rising on each side reminds us of Phylactella eximia, Hincks; but in P. collaris the sides are sometimes raised above the rest of the peristome. Loc. Fossil: Waurn Ponds and Waurn Quarry. 49, Smirrra TuRRITA, Sm. Lepralia turrita, Smitt, Floridan Bryozoa, p. 65, pl. xi. figs. 226— 228. The specimen, which was collected either from Batesford or from Muddy Creek, is in the Cellepora forma, the zoarium consisting of BRYOZOA FROM AUSTRALIA. 489 ‘many layers, and the zoccia are heaped together and irregularly disposed. There are usually only two “stout marginal spines,” not four as described by Smitt in the specimen from Florida, and in the Australian fossil these are very irregularly placed. ‘The oral aperture is nearly round, and of the same size as in the Floridan specimen, viz. 0°15 millim. wide. In the aperture there are three denticles, similar to those in Smittia reticulata, and therefore I unite it with that genus, though not without much doubt as to where it should be placed. Loc. Living: Florida, 26—44 fath. 50. ScHIZOPORELLA AUSTRALIS, T. Woods. Tetraplaria australis, T. Woods, Tr. Roy. Soc. N. 8. Wales, 1878, p. 5. fig. 4. Schizoporella australis, Waters, Bry. from 8.W. Victoria, p. 341, pl. xiv. fig. 15. A specimen from Bird Rock has fine granulations and fine pores, as described in my former paper, p. 341. Loc. Muddy Creek (Woods); 8.W. Victoria (A. W.); Bird Rock. 51. ScHIzoPoRELLA scHizostoma, MacG. ‘ Lepralia schizostoma, MacGillivray, Nat. Hist. Vict. deca deiv. p. 33, pl. 38. fig. 6. The zocecia are small, ovate, with a small oral aperture and large globose ovicell. Immediately below the oral aperture there is an avicularian prominence, which is not described by MacGillivray, but is figured about the centre of the front of the zocecium. Width of oral aperture 0-05 millim. Loc. Living: Williamstown and Queenscliff. Fossil: Waurn Pond Quarry, Suffolk Crag (A. W. W. coll.). 52. ScHIZOPORELLA SUBMERSA, Waters. Schizoporella submersa, Waters, Bry. S.W. Vict. p. 340, pl. xviii. fig. 5. Fossil : Curdies Creek (W.), Muddy Creek. 53. RETEPORA MARSUPIATA, var. MucRoNATA. Plate XII. fig. 13, 21. This differs from most specimens of R. marsupiata in having a very large oval avicularium below the aperture; but we see in Smitt’s fig. 248 (Floridan Bry.) that there is sometimes such an avicula- rium ; above this the peristome rises into a mucro. The opening on the front of the ovicell is wide, pointed above and rounded below. The dorsal surface is divided into nearly equal areas by an irregular line running down the middle, from which lines branch off; at the base of each dorsal division there is a narrow avicularium directed diagonally downwards (see fig. 36, pl. xv. Bry. fr. 8.W. Vict.). Loc. Fossil: Muddy Creek. 54, Rerepora Branrana, King. Retepora Beaniana, King, Ann. Nat. Hist. vol. xvii. 1846, p. 237 ; 212 440 A. W. WATERS ON FOSSIL CHILOSTOMATOUS Hincks, Brit. Mar. Polyzoa, p. 391, pl. lui. figs. 1-5. (See Hincks for synonyms.) In the fossil from Muddy Creek the central zocecia have a rounded avicularium, directed downwards, within the aperture; but the outer zocecia have a subspatulate avicularium directed laterally. It is somewhat doubtful if the species described by Stoliczka is really R. Beanana. Loc. Fossil: Muddy Creek and English Crag. 55. CELLEPORA Fossa, Haswell. 06, CELLEPORA ALBICANS, Hincks. 57. CELLEPORA GRANUM, var. Plate XII. fig. 18. Cellepora granwm, Hincks, Gen. Hist. of Mar. Polyzoa, p. 68, pl. i. fig. 8; Ann. & Mag. Nat. Hist. Aug. 1881. The fossil varies in having the peristome closed in, which thus © forms an elongate tube. In Cellepora Costazw, Aud., from the Mediterranean, a similar variation is found; for occasionally the peristome is entirely closed. Loc. Fossil: Batesford or Muddy Creek. 58. SeLenaRiA PunctatTA, T. Woods. Selenaria punctata, T. Woods, ‘‘ On some Recent and Fossil Spe- cies of Selenariade,” Trans. Phil. Soc. Adelaide, 1879, p. 9, pl. ii. fig. 8. The large pores below the aperture are denticulated, as are also the rosette plates. The vibracular areas are broken ; but they are | so similar in character to those of Selenarza maculata that I should expect that in perfect specimens they will be found covered with a cribriform calcareous expansion. Aperture 0°21 millim. wide. This species is very closely allied to Caleschara denticulata, MacG. (Nat. Hist. Victoria, decade v. p. 45, pl. xlviii. fig. 8). Loc. Living: Cape Three Points, Australia (Woods), 71 fath. Fossil: Muddy Creek. 59. Sevenarra MacuLata, Busk. Plate XII. figs. 7, 9, 12. Selenaria maculata, Busk, Cat. Mar. Pol. p. 101, pl. exvii. The shape of the opesia in the fossil is slightly different from that of recent specimens in my possession from Holborn Island; and I therefore give figures of both for comparison. In the fossil the zocecia and aperture are somewhat wider than in the recent specimens, and the lip is nearly as broad as the opesia, whereas in the recent specimens this lip is narrow and almost spicular. Some authorities may think that on this account the fossil ought to be called a va- riety ; others perhaps would attach specific value to this difference. The way in which the Selenariade begin to grow is a subject worthy of complete investigation; but in the meantime the few specimens in my possession throw some light upon it. The fossil has grown upon a small stone, probably crystalline; and upon BRYOZOA FROM AUSTRALIA. 441 breaking up recent specimens I found, placed as a nucleus, either a grain of sand or a minute Foraminifer, as shown semidiagramma- tically in fig. 12. The first cells are much smaller than the later ones and yery shallow ; and the calcareous growth of the Selenaria will completely surround the grain of sand &ce. I have Cupularia stellata, B., living, from near Capri, which there grows upon small shells and stones, and sometimes only covers the shell ; in other cases, where the stone or shell is small, it spreads over and grows free, thus assuming the form of the Selenariade. We thus see that at the commencement the mode of growth of the group is truly Membraniporidan. In Cupularia umbellata from the Antwerp Crag, and also from the Pliocene of N. Italy, I have found a nuclear grain of sand upon which the colony has grown. Stoliczka, when speaking of Lunulites latdorfensis, says (Olig. Bry. von Latdorf, p. 94) that the point of attachment is at the top, and points out that this is curious, as it is usually below; but surely the fact must be that the upper part has been broken away, and the shell upon which the young colony has grown is thus exposed. In my paper on Bryozoa from 8.W. Victoria, page 345, I sug- gested that it was perhaps by an error of the lithographer that Mr. Woods’s figures of the Selenariade were drawn upside down ; but as they have been figured and described thus by a large number of our leading authorities, it is as well to point out that the distal end of the zocecium must of course always be that which is nearest the growing end of the colony ; and this we are always able to decide by means of the operculum, as the hinged end will be the proximal, while the free end is the distal or upper end. I would ask a question which only those who dredge in Australia can auswer. It is whether, since such a form as Seélenaria macu- lata grows upon a grain of sand, we do not thus, by its presence, obtain an indication as to the sea-bottom of the locality where it grows ? . Selenaria alata, T. Woods, seems very closely allied to this. Width of opesia, at the widest part, 0°19 millim.; in the middle 0-13 millim. ‘The width of the opercular aperture in a recent spe- cimen is 0°12 millim. Loc. Living: Bass’s Strait (B.); Holborn Island (Haswell). Fossil: Muddy Creek, Bird Rock. 60. SELENARIA PARVICELLA, T. V/oods. Selenaria parvicella, T. Woods, “ On some Recent and Fossil Species of Australian Selenariade,” Trans. Phil. Soc. Adelaide, 1880, p. LO, pl. u. fie. 10. The specimen from Bird Rock is but badly preserved. A calca- reous expansion, sloping inwards, about half covers the front of the zocecium. ‘The opesia is thus nearly semicircular. Average width of opesia 0°19 millim. Loc. Fossil: Muddy Creek (Woods), Bird Rock. 442 A. W. WATERS ON FOSSIL CHILOSTOMATOUS 61. Lunvunires marta, sp. nov. Plate XII. fig. 8. Zoarium subconical, small, 1 millim. in diameter. Zocecia oval, surface granulated, concave, separated by a raised border. Aperture rounded above, with a real or apparent sinus below. A few conical cells, with three large pores between the zoccia and at the edge of the zoarium. Base of zoarium showing each zocecium raised and radiating from the centre. Loc. Waurn Ponds. 62. Lunuiires PeTALorpEs, d’Orb. Plate XII. fig. 11, a, 6, c. Lunulites petaloides, d’Orb. Pal. Frang. p. 353, pl. 705. figs. 6, 9. Lunulites androsaces (of Allioni), Manzoni, Bri. Phioe. AL cont. la, p. 12, pl. ii. fig. 18, and Bri. Foss. del. Mioc. d’Aust. ed Uneh. p- 25 (723), pl. xwii. eg Innulites distans, Gabb & Horn, Monogr. Foss. Pol. of Second. and Tert. Form. of N. Amer., Ac. Nat. Sc. Philad. vol. v. pt. ii. p. 119, pl. xix. fig. 4. . Oligotresium vicksburgensis, Gabb & Horn, loc. cit. p. 139, pl. xix. fig. 22. Innulites cupola, T. Woods, ‘On some Recent and Fossil Species of Selenariade,” Trans. Phil. Soc. Adelaide, 1880, p. 8, pl. 1. fig. 5a, c. Zoarium conical. Zocecia oval to hexagonal, area depressed, slightly raised round the aperture, surface granular. Aperture (oral or opesial) straight below, rounded above. Vibracula usually forming a row between the radial zocecia, but sometimes fewer, and irregularly spread over the zoarium ; opening of vibracula elongate, with a process projecting from each side, which are sometimes equal, but in other cases the projection on the left is much larger and toothed. Central zocecia smaller, frequently closed with a granular calcareous mass. This commences inside the distal end of the aperture, and is subtriangular, hanging down inside the aperture like the human tonsil in the throat. It becomes gradually wider until the whole aperture is closed. In one specimen the vibracula are placed quite regularly in radial rows ; in two others, in which both zocecia and vibracula are rather larger, there are fewer vibracula, and these are not regularly placed. This is a species which lived through the Cretaceous, Miocene, and Pliocene; and it and its allies were widely distributed in Cretaceous and Miocene times. The list of synonyms ought to be very large; but as the description, in some cases, leaves it doubtful, I only mention, as belonging to the group, if not to this species :— Inmulites Goldfussi, Hag.; L. Hagenowi, Bosq.; L. quadrata, Rss. ; L. hemisphericus, Rom.; L.microporus, Rom.; Discoescharites ma- millatus, Rom.; L. latdorfensis, Stol.; L. conicus, Busk; L. apertus, T.Woods ; L. exiqguus,T. Woods ; L. Bourgeoisiz, d’Orb.; L. regularis, d’Orb., L. cretaceus, Defr. If a new species is to be made each time that the shape of the zoarium varies, we can then make an enormous number of species of Lunulites: and for those who are fond of Quart. Journ.Geol. Soc .Vol. XXXIX.. PLAT. A W.Waters del’, A.T. Holliick. hth . Mintern. Bros. imp VICTORIAN CHILOSTOMATOUS BRYOZOA BRYOZOA FROM AUSTRALIA. 443 making genera, I would call attention to amethod which Messrs. Gabb and Horn have unconsciously hit upon, as upon one plate the same species is figured right way up and upside down, thus giving two good genera. I am not sure that this ought to be separated from S. marginata, Woods. Hig. 1. EXPLANATION OF PLATE XII. Catenicella levigata, sp. nov., X 25. 3, 4. longicollis, sp. nov., X 25. — Harvey, Thoms., x 25. ponderosa, Goldst., recent, showing the protecting plate, x 25. . Selenaria maculata, Busk, fossil, x 25. initia, sp. nov., X 25, maculata, Busk, recent, X 25. . Smittia collaris, var., X 25. . Lunulites petaloides. Apertures of central zocecia, closed by subsequent growth, x 89. 2. Selenaria maculata, Busk. Diagrammatic, showing grain of sand on which it grew. . Retepora marsupiata, Sm. Dorsal surface. . Smuttia Napierit, sp. nov., X 25. , 16. Catenicella alata, Thoms., x 25. . Cribrilina terminata, Waters, X 25. . Cellepora granum, Hincks, var., X 25. . Membranipora arethusa, @Orb., xX 25. . —— lusoria, var. coarctata, X 25. . Retepora marsupiata, Sm, X 25. / . Membraniporg oculata, Busk, X 295. 444 PROF. J. W. JUDD AND G. A. J. COLE ON THE 24. On the Basatt-Guass (Tacuytyte) of the WESTERN Isis of Scortanp. By Prof. Joun W. Jupp, F.R.S., Sec. G.S., and GRENVILLE A. J. Cotz, Esq., F.G.S. (Read May 23, 1883.) [Puatzes XIII. & XIV. In a previous paper * it has been pointed out that the Tertiary vol- canic rocks of the Western Isles of Scotland offer beautiful examples of materials of every variety of composition, from the most acid to the most basic, and of every type of structure, from the holocrystal- line to the vitreous. The detailed description of these varieties of volcanic rocks was reserved for a future occasion ; and in the present paper we propose to give the first of such a series of descriptions. As the more acid vitreous rocks have during recent years been dis- cussed in considerable detail in numerous papers read before this Society, it may not be inappropriate to direct attention to the rare but equally interesting glasses of basic composition, which have, up to the present time, received far less notice in this place. The studies on which this paper is based have been carried on in the Geological Laboratory of the Normal School of Science and Royal School of Mines. 1. History of Previous Opinion on the Subject. By the older writers on petrography rocks of the kind of which we now treat appear to have been classed as “ pitchstones.” Jame- son and the other followers of Werner, who endeavoured to intro- duce the precision of nomenclature and the exact methods of their master into the study of British rocks—though they recognized these materials, not as minerals but as rocks—do not seem to have dis- cerned the difference between the acid and basic varieties. Mac- culloch, who may be justly regarded as the father of British petro- graphy, as early as 1819 pointed out that, though basalt occasionally passes into glass, yet examples of such a transition are exceedingly rarey. He particularly records two such cases in the Western Isles of Scotland. One of these instances of a transition from basalt into “ pitchstone ” is given as occurring in the Isle of Lamlash (Holy Isle), near Arran, and is illustrated by a detailed description and drawing {. The other example is at Garbsbeinn, in Skye; but the specimen described was not found 7m sitw§. In 1827 Sedgwick and Murchison found a basalt dyke at the Beal near Portree, in Skye, the sides of which are seen passing into * Quart. Journ. Geol. Soc. vol. xxx. p. 233. t Western Islands of Scotland, vol. i. -p. 402. { Ibid. vol, ii. p. 437. § Ibid. vol. i. p. 402. BASALT-GLASS OF THE WESTERN ISLES OF SCOTLAND. 445 “ pitchstone;” and the fact is recorded by the latter author in the ” Transactions of the Geological Society*. In 1829 Bouét + referred to the two cases of basalt passing into . glass already fescrieh by Macculloch; and it was probably on specimens carried to Paris by Boué that Delesse made the interest- ing observations to which we shall refer in the sequel. On the Continent, the first clear recognition of the distinction between the basic and acid glasses appears to be due to Breithaupt in 1826. Klaprothf, it is true, had as long ago as 1807 described a “black fossil” from Guiliana in. Sicily as “slagey augite ;” and this substance is now generally regarded as belonging to the class of basalt-glass ; but to Breithaupt § we owe the precise indication of those characters by which the basic glasses are distinguished from similar materials of different composition. Breithaupt regarded the substance which he obtained from the Sasebuhl between Gottingen and Dransfeld as anew mineral species, and indicated its distinctive properties with great precision. He proposed for it the name of “ tachylyte,” signifying the rapidity with which it undergoes fusion || before the blowpipe. It is worthy of remark that the author of the species and his followers on the Continent down to quite recent times regarded the substances of this class as minerals and not as rocks. In 1840 Gmelin § described a somewhat similar substance from the Vogelsgebirge (probably from Bobenhausen); and in 1841 Klipstein ** gave more precise information concerning the mode of occurrence of this material. In 1844 Hausmann Tr gave a new description of the substance from the Sasebihl, accompanied by an analysis by Schnedermann ; and he connects with it a tachylyte- like ‘“ fossil” from the Wetter au, which Gmelintt, four years before, had analyzed and referred to augite. In 1847 Hausmann$§ sepa- rated the Bobenhausen material from tachylyte as another mineral species under the name of “hyalomelane,” though the grounds of this division do not very clearly appear. * Trans. Geol. Soc. vol. ii. p. 359. The veins of Carsaig in Mull, with which the Beal dyke is erroneously compared, are formed ofa black pitchstone which is of acid composition, containing 68 per cent. of silica. A similar comparison is made by Lyell and Murchison, Edin. Phil. Journ., July 1829. tT Essai géologique sur 1|’Ecosse, p. 284. + Beitrage zur chemischen Kenntniss der Mineralkorper, vol. iv. p. 190. § Kastner’s Archiv fiir die gesammte Naturlehre, vol. vii. p. 112 (1826). || A great amount of confusion has taken place with respect to this name “tachylyte.” It has been frequently written ‘‘tachylite,” “ tachylith,” and even ‘‘trachalite.” Naumann and others state that the rock is so called from the ease with which it is decomposed by acids. A reference to Breithaupt’s original memoir shows that the spelling and origin of the name are as stated in the text. The name ‘ bottleite,”’ said by Kinahan to have been locally given to it, does not appear to have come into any general use. q ’Poggendorff’s Annalen, vol. xlix. p. 233. ** Neues Jahrbuch fiir Min. &c. 1841, p. 696. tt Studien d. géttingischen Vereins bergmann, Freunde, vol. v. p. 91. tt Neues Jahrbuch fir Min. &c. 1840, p. 549. §§ Handbuch der Mineralogie, zweiter Theil, vol. i. p. 545. 446 PROF. J. W. JUDD AND G. A. J. COLE ON THE In 1853 Sartorius von Waltershausen * described similar sub- stances from Iceland under the name of “ sideromelane,” noticing -that they formed the anhydrous kernels of a material identical with his ‘* palagonite.” In 1868 Petersen + described a glassy variety of the nepheline- basalt of the Rossberg, to which he gave the name of ‘ hydro- tachylyte.” Zirkel, Vogelsang, Méhl, Boricky, Sandberger, and others describe different varieties of basalt-glass under the name of “ tachylyte” (often misspelt “tachylite”); and by this same designation the Scottish examples were referred to by one of us in 1875 in the paper already cited. Mr. Rutley, in 1877, applied the same name to a similar rock from County Down, Ireland. But in 1872 Prof. Rosenbuscht argued in favour of restricting Breithaupt’s name of tachylyte to those varieties which are easily decomposed by hydrochloric acid, and of giving Hausmann’s name of hyalomelane to such as are only partially or not at all decom- posed by the same agent; and this suggestion was adopted by Cohen§, Mohl||, and other writers. In 1877, however, Prof. Rosenbusch §[ gave up the distinction altogether, having apparently become impressed by the difficulty and uncertainty of the test of solubility in acid as applied to rocks. He was, moreover, influenced in this by the fact that, following the system already adopted by Mohl and Zirkel, he now classified these glassy products not as mineral species, but as rocks—the vitreous form of basalt. Penck**, in 1879, argued with much force in favour of adopting the same course; and the distinction between tachylyte and hyalo- melane, as well as their inclusion among mineral species, may now be regarded as being by universal consent abandoned. Various names have been proposed by different authors for substances of the class, such as “ glassy basalt,” “vitreous basalt,” ‘‘basalt-vitrophyre,” “ basalt-glass,” and “ basalt-obsidian.” In his most recently published classification of rocksf?, in 1882, Prof. Rosenbusch recognizes only two divisions among the anhydrous basic glasses—basalt-glass (including tachylyte, hyalomelane, side- romelane, and slagey augite), and hydrotachylyte or nepheline- basalt-elass. 2. Distribution and Mode of Occurrence. Basalt-glass, though a widely distributed rock, is one which can- not be considered as by any means of common occurrence. It does * Uber die vulkanischen Gesteine in Sicilien und Island, p. 202. t Neues Jahrbuch fir Min, &c. 1869, p. 32. + Neues Jahrbuch fiir Min. &c., 1872, p. 148. § Ibid. 1876, p. 746. || Zusammenstellung und Beschreibung einer Sammlung typischer Basalte. €" Mikroskopische Physiographie, vol. 11. p. 445. i ** Zeitschrift der deutschen geologischen Gesellschaft, vol. xxxi. pp. 521, 568, &e. tt Neues Jahrbuch fir Min. &c. 1882, vol. ii. p. 16., BASALT-GLASS OF THE WESTERN ISLES OF SCOTLAND. 447 not appear to constitute great rock-masses, but is usually found as a local variation of certain types of dolerite and basalt. An excep- tion to this rule exists in the case of the well-known glassy lavas of - Hawaii; but this exception we shall discuss in the sequel. The most general mode of occurrence of basalt-glass is as a sel- vage (Saalband) to basaltic dykes; and this is the only condition under which we have found it to occur in the Western Isles of Scotland. Under similar circumstances it has been observed in Ice- land*, Bohemiay, and many other districts. Other modes of the occurrence of basalt-glass which have been recorded are as follows :—Lyell and Murchison found the basaltic lava of Thueyts in the Vivarais, where flowing over gneiss, to be coated on its under surface with a band of glassy materialt. Many authors have described the surfaces of lavas and ejected blocks as covered with a glassy crust. Zirkel notices that the lava-stalactites in a cave in a basaltic lava-stream in Iceland have a surface-film of basalt-glass$ ; and a similar film is stated to surround the air-cavities in a basalt of Hellegrund, near Miinden||. At Bobenhausen in the Vogelsgebirge and elsewhere basalt-glass is recorded as occurring in basalt in nests from the size of a walnut to that of a child’s head]. Sartorius von Waltershausen and others have described kernels of basalt-glass (sideromelane) as being abundant in the basaltic tuffs of Iceland, Sicily, and other districts; and in these cases the outside portions of the kernels are often found to graduate into palagonite**, Some authors, indeed, regard all palagonite as the result of the hydra- tion and alteration of particles of basalt-glass. From a discussion of all the known occurrences of basalt-glass, Mohl ty? has argued that it has in every case been formed by the rapid cooling of portions of basaltic lava. Besides the two localities at which basalt-glass has been already noticed in the Western Isles of Scotland, we have only succeeded in finding it at three other points, namely Sorne Point and Gribun in the Isle of Mull and Screpidale in the Isle of Raasay. The condi- tions under which the rock occurs at these five known localities are as follows :— At the Beal, near Portree, in the Isle of Skye, a basaltic dyke has glassy selvages about two inches thick. Murchison, it is true, gives the width as four inches; but we have never found it exceeding the amount stated. This is by far the most striking occurrence of ‘basalt-glass with which we are acquainted. Its vitreous character * Zirkel, Lehrbuch der Petrographie, vol. ii. p. 303. Tt Boricky, Sitzungsberichte der k. boluaiteltin | Gesellschaft der Wissenchaften in Prag, 1873, p. 8; “also Petrogr. Studien an den Basaltgesteimen Bohmens, p- 182. { Edinburgh New Philosophical Journal, April to October, 1829, p. 29. § Lehrbuch der Petrographie, vol. ii. p. 304. | Ibid. {| Neues Jahrb. fiir Min. &c. 1841, p. 696. -** Sartorius v. Waltershausen, ‘ Ueber die vulkanischen Gesteine in Sicilien und Island,’ 1853, p. 202. tt Die Gesteine der Sababurg in Hessen (Cassel, 1871). A438 PROF. J. W. JUDD AND G. A. J. COLE ON THE gradually diminishes as we trace it towards the interior of the dyke, the glassy rock merging insensibly into the basalt. We shall point out later tue peculiar columnar and other joint-structures which are exhibited at this locality. At Lamlash (Holy Isle), Arran, the mode of occurrence has been - very clearly described and illustrated by Macculloch, who gives a sec- tion at the spot*. The glassy selvage of the dyke never exceeds one inch in width; it is not so perfectly vitreous as that of the Beal near Portree, its lustre inclining more to resinous, and it merges insensibly into the basalt of the body of the dyke. At Sorne Point, on the north-west coast-line of the Isle of Mull, a very vitreous selvage is found in a dyke that intersects the lava- streams composing the great plateau of Mishnish ; it seldom exceeds one fourth of an inch in width, and passes into the basalt of the dyke by the most insensible gradations. At Gribun, in the west of Mull, a dyke of basalt is seen by the road that leads from the village of that name to Kilfinichen, just where it emerges on the basalt-plateau; and the sides of this dyke are formed by glassy selvages never more than half an inch in thickness. Lastly, at Screpidale, in the Isle of Raasay, one of the dykes of basalt passing through the grand precipices of Jurassic strata has also glassy sides. The actually vitreous portions of this dyke, how- ever, are very thin indeed, forming little more than surface-films. The only other case which we have found in the Western Isles of Scotland at all comparable to these, is the occurrence of a basic glassy rock among the later-formed products of the volcano of Beinn Shiant in Ardnamurchan. Unfortunately the specimens were not found in situ; and their exact mode of occurrence is therefore unknown. Prof. Jameson t and Dr. A. Geikie t have described the occurrence of a “pitchstone” at Eskdale which appears to be similar in com- position to the rock of Beinn Shiant. Both are probably augite- andesite glasses. From the description given of the pitchstone of Eskdale it would appear as though a columnar dyke had opened along its central plane of weakness, where the two sets of columns meet, and permitted the extrusion of material which consolidated as a “‘ pitchstone.” - In the north of Ireland some of the basaltic dykes have been described as having vitreous selvages§. The microscopic characters of one of the basaltic glasses, that of Slievenalargy, co. Down, have been described by Mr. Rutley; and it has been analyzed by Dr. Haughton). 3. Specific Gravity. As Delesse so long ago pointed out, the specific gravity of a rock * ‘Western Islands of Scotland, vol. ii. p. 487. t Mineralogical Description of Dumfriesshire, p. 115. { Proceedings of the Royal Physical Society of Edinburgh, vol. v. p. 29. § Kinahan, Geol. Mag. decade ii. p. 426. || Journal of the Royal Geological Society of Ireland, vol. iv. p. 227. BASALT-GLASS OF THE WESTERN ISLES OF SCOTLAND. 449 of crystalline structure is always higher than that of a rock of the same chemical composition, but having a more or less vitreous or colloid structure. In the same way the glassy products formed by the fusion of crystalline rocks are always of less density than the rocks themselves. Delesse fused a basalt having the specific gravity of 2°85, and found the glass thus produced to have a density of 2°77. The well-known melaphyre (altered basalt) called Rowley Rag has a specific gravity of 2°84 according to our deter- minations; the black glass formed by Messrs. Chance, of Birming- ham, from the fusion and rapid cooling of the same, we found to have a density of 2°75; the same material when slowly cooled, yielding a product full of crystals of felspar and skeleton-crystals of magnetite, has a density of 2°77. But although the rule that the crystalline forms of rocks are more dense than the glassy is very generally true, there is, as Delesse himself pointed out *, a remarkable exception to it in the case of a basalt-glass of the Western Isles of Scotland. Delesse found that a portion of the basalt-vein of Lamlash has a density of 2-649, while the glassy basalt forming the selvage of the dyke is denser, having a specific gravity of 2°714. This anomalous circum- stance has been confirmed by a series of careful determinations which Mr. Thomas Davies, acting on our suggestion, has made on a specimen of this dyke in the British-Museum collection. He found that the basalt of the dyke had a specific gravity of 2°67, the part of the glassy selvage adjoining and graduating into basalt had a specific gravity of 2°72, while nearer to the side it was 2-74, and on the extreme edge 2°78. Delesse endeavoured to explain this anomalous rise of density in the more vitreous portions of the rock by a slight difference of composition in the two parts of the mass. It is true that his analysis shows the basalt to contain a little less silica and rather .more water than the glass; but the difference is so slight as to be . almost within the limits of errors of analysis, and perhaps does not sufficiently explain this curious difference of density. The micro- scopic study of this rock shows that the basalt of the dyke is a magma-basalt, in which only incipient crystallization has taken place. It is possible that the basalt has suffered greater alteration than the more compact glass of the selvages of the dyke. The average density of basalt-glass may probably be taken as about 2:7. Von Lasaulx gives the range as from 2°51 to 2-567; put this is certainly below the truth. The basalt-glass (sidero- melane) of Iceland has, according to Sartorius von Waltershausen ae a specific gravity of 2°53; the “ tachylyte” of the Sasebiihl has a specific gravity of 2°58$; and the “slaggy augite” of Sicily, * Annales des Mines, vol. xiii. pp. 369, 370. + Elemente der Petrographie, p. 225 (1875). ¢ Vulkan. Gesteine in Sicilien und Island, p. 203 § Hausmann, ‘Neues Jabrb. fir Min.’ &c. 1844, p.70. Breithaupt (Kastner's Archiv fiir die gesammte Naturl. vol. viii. p. 112) gives specific gravity 2°50 to 2°54. 450 PROF. J. W. JUDD AND G. A. J. COLE ON THE acccording to Klaproth, of 2°67*. Mohl states that the basalt-glass of Sababurg has in its most vitreous condition a density of 2-68, and in its less vitreous of 2°76. The density of the basalt-glass of Bobenhausen is, according to-the same author, 2°71 +, and that of Ostheim 2:74§. Cohen|| records the specific gravity of six varieties from the Sandwich Islands, the average of which is 2:71. The basalt-glass of the Western Isles of Scotland appears to be generally distinguished by its very high specific gravity. That of the Beal, near Portree, is 2°72; that of Lamlash, 2°78; that of: Gribun, in Mull, 2°82; that of Screpidale, in Raasay, 2°84; while the basalt-glass of Sorne has a density of no less than 2°89. This difference appears to be connected with peculiarities of chemical composition to be hereafter noticed. Of somewhat analogous substances we may observe that the tephrite-glass of Klein Priesen has a density, according to Boricky 4, of 2:65, the tephrite itself giving 2-696. The augite-andesite glass of Beinn Shiant, Ardnamurchan, gives 2°62, and the probably similar rock of Eskdale 2-7 **, 4, Other Physical Properties. The colour of the freshly broken surfaces of the basalt-glass of the Hebrides varies from velvet-black to pitch-black. The joint- planes and all exposed surfaces, however, are covered with a film of a grey or greenish-grey colour, occasionally passing into brown. This surface-film, which may be due either to chemical change or to molecular alteration of the exposed faces, appears to be analogous to what is seen in the pitchstones of Ponza, which, when freshly broken, are perfectly vitreous in appearance, but in a few seconds become coated with a delicate film which impairs their lustre +?. The beautiful iridescence seen on some surfaces of the basalt-glass of the Beal, in Skye, is probably due to interference ' produced by a thin film of the same kind. The lustre of the rock at the extreme edge of the dykes at the Beal, Gribun, and Sorne is perfectly vitreous; but it graduates to resinous, horny, and glimmering in passing inwards into basalt. The lustre of the Lamlash and Screpidale basalt-glass is never more than resinous. The hardness of basalt-glass appears to be not very different from that of orthoclase, varying from 5:5 to 6:5. Though probably, as a rule, softer than obsidian, the difference is not sufficient to distinguish it readily from that rock. But, on the other hand, the hardness of basalt-glass is so much greater than that of the pala- * Beitrage zur chemischen Kenntniss der Mineralkorper, vol. iv. p. 190. t Die Gesteine der Sababurg in Hessen, (Cassel) p. 40. t Ibid. § Ibid. || Neues Jahrbuch fiir Min. &e. 1880, vol. 11. p. 41. _ { Petrographische Studien an den Basaltgesteinen Bohmens, pp. 182 and 210. ** A. Geikie, Proceedings of the Roy. Phys. Soc. of Edinburgh, vol. vy. p. 253, tt Geol. Mag. dec. 2, vol. ii. (1875) p. 3808. BASALT-GLASS OF THE WESTERN ISLES OF SCOTLAND. 451 -gonites and other hydrated materials of that class that it serves as an easy means of distinction between the altered and unaltered forms. The whole of the basalt-glasses are strikingly magnetic. The powder of all of them is easily attracted by the magnet, those of Sorne, Screpidale, and Gribun exhibiting this character in a specially marked degree. Basalt-glass is at once distinguished from vitreous substances of more acid composition by its easy fusibility, the product being an opaque black-brown or black bead. The fusibility of the varieties from the Western Isles appears to be little above 2 of von Kobell’s scale, while that of typical obsidian is no less than 4°5 of the same scale, the product of fusion being a white enamel. The basalt-glass from the Hebridean localities appears to be always traversed by numerous joint-planes. The most pronounced of these are parallel to the side of the dyke, and enable the glassy material to be easily separated from the basalt in thin layers. Besides these principal joints, there are usually one or two other sets at right angles to one another and to the sides of the dyke. Hence the surfaces of the dyke, where in contact with the rock which it traverses, are broken up by these joints into small rect- angular areas. This character is strikingly exemplified by the basalt-glass of the Beal. The same dyke also exhibits in some portions of its glassy selvage a remarkable columnar structure. The columns are very minute, often as thin as fine needles, about 12 inch in length, and some- times beautifully curved. Some specimens present a miniature reproduction of the celebrated Clam-sheli Cave in Staffa. Occa- sionally the columns are as much as 3 inch in diameter (Pl. XIII. hee): This finely columnar structure affords an admirable illustration of the fact that the diameter of the columns in a rock is in part dependent on the coarseness of the materials of which it is com- posed. While the highly crystalline rocks of the Shiant Isles and Ailsa Craig form columns with a diameter of 10 or 12 feet, these vitreous rocks exhibit columns of almost microscopic dimensions. Besides the rectangular and columnar jointing, a minutely spheroidal structure is sometimes found developed in the basalt- glass. Hence, under the microscope, it frequently exhibits traces of the well-known “ perlitic” structure. This last-mentioned fact has been noticed by Zirkel in the case of the basalt-glass of Maros- tica (Monte Glosso) in the EKuganean Halls, by Mr. Rutley in the ‘“‘tachylyte” of Slievenalargy in the north of Ireland, and by Fouqué and Lévy in the lava of Graham’s Isle. 5. Solubility m Acids. So much importance has been attached by some authors to this character, as presented by rocks of this class, that it is perhaps desirable to discuss in some detail the behaviour of the rocks now being described when subjected to the action of acid. pay PROF. J. W. JUDD AND G. A. J. COLE ON THE _ As we have already seen, many writers have classified the easily soluble forms as “ tachylytes,” and those only slightly acted on by acids as “‘hyalomelanes.” It says little, however, for the value of this distinction, that different authors have placed the same rock in different classes when judged by this standard. Thus Mohl’s con- clusions concerning the solubility of the different basalt-glasses appear to be very different from those of Rosenbusch and other writers. This solubility in acids appears to depend on so many conditions, such as the more or less altered condition of the speci- men, the fineness of the powder operated upon, the degree of con- centration of the acid employed, the time and temperature of digestion, &c., that it is not difficult to account for these striking discrepancies. Penck, as we have already seen, has argued with much force in favour of abandoning the distinction based on solu- bility in acids*; and in his later works Rosenbusch f himself appears to have entirely abandoned it. In order to compare the behaviour of the several rocks here de- scribed under the action of acids, we submitted weighed portions of their powder to the action of concentrated hydrochloric acid for a period of ten days, boiling them each day for some time. The basalt-glass of the Beal left a residue of 81°6 per cent.; that of Sorne of 70°79; that of Gribun 83°40; and that of Lamlash 83°37 t; while the rock of Screpidale, which was even attacked by the cold acid, left a residue of only 57°82 per cent. In all cases flocculent silica was separated, but in some in much greater quan- tity than others. In order to compare these results with those given by other rocks of the same class, we submitted a glassy lava of Hawaii and the artificially fused rock of Rowley Regis to the same test. The former left a residue of 50-57, and the latter of 58:7. Treated for the same time, but with only one boiling, Mr. Grant Wilson § found that the Eskdale pitchstone left a residue of 83-2 per cent. We have determined that in the apparently similar rock of Beinn Shiant the insoluble portion is, after daily boilings during the same period, 81-17 per cent. Krukenberg || found the proportions dissolved from different varieties of Hawaiian lava, which were digested in hydrochloric acid from ten to sixteen hours, to vary from 49 to 62 per cent. ; while B. Silliman, jun. 4], after similar experiments, obtained as the insoluble proportion from 50 to 57 per cent. Cohen **, after a * Zeitschrift der deutschen geologischen Gesellschaft, vol. xxxi. 1879, p. 521, &e. t Mikroskopische Physiographie, vol. ii. p. 445 (1877), and Neues Jahrbuch fir Min. &c., 1882, vol. ii. p. 16. { Delesse found that, after boiling with hydrochloric acid, the glassy rock of the Lamlash dyke gave a residue of 87-7 per cent., and the basaltic central portion of 82 per cent. § Proc. Roy. Phys. Soc. of Edinburgh, vol. v. p. 253. || Mikrographie der Glasbasalte von Hawaii. Tibingen: 1877, p. 3. “| Memoirs of the Boston Soc. of Natural History, vol. i. p. 460. ** Neues Jahrb. fur Min. &e. 1876, p. 746. BASALT-GLASS OF THE WESTERN ISLES OF SCOTLAND. 453 digestion of a specimen in concentrated hydrochloric acid for thirty hours, found a residue of 39°62 per cent., and, with Rosenbusch, classes these Hawaiian lavas as hyalomelanes. Krukenberg, how- ever, regards it as doubtful whether they should be placed with tachylyte or with hyalomelane. Boricky * found the tephrite-glass of Klein Priesen to be decom- posed partially and with difficulty by hydrochloric acid, some floccu- lent silica being separated. Many of the German basalt-glasses, on the other hand, appear to be entirely decomposed by boiling hydro- chloric acid, that of the Sasebthl, according to Monhl +, after two days’ digestion, and some even more readily than this. Classified according to the test of solubility in acids, the rock of Screpidale might perhaps be placed with the tachylytes, and the Beal, Sorne, and Gribun rocks with the hyalomelanes; but our own results, like those of Rosenbusch and Penck, tend to the conclusion that it is advisable to abandon altogether a distinction founded on such an uncertain character. 6. Chemical Composition. If, as there seems reason to believe, basalt-glass is merely a rapidly cooled portion of a basalt-lava, we may expect the two rocks to have the same chemical composition. _ Delesse’s analysis of the dyke of Lamlash and of its vitreous selvage appears to indicate that this is really the case. The analyses are as follows + :— Basalt of Basalt-glass centre of side of of dyke. dyke LdIT 19) \a dele a Ae REA aire REC REEL 55°20 56°05 JOG TUT N00 eae a rae a a aE 16°98 17-13 Oxideroiirom wes ge Tk 11:00 10°30 , otmanganese: 220". 42. traces. traces GTS GPP Fe eo Ws das) 5 6°80 6°66 Niaeidestay 02 2h P.O. at. 22 8 0°52 1°52 Badan Ai wee it. 3 wy 5°65 (by 3°29 TAG Chg DO A Sea difference). } 0:98 Water and volatile matter.... 3°85 3°50 100-00 99°43 The differences in these two results, judging by their character, are such as may be fairly supposed to fall within the limits of error in analysis. It has frequently been pointed out, however, that the average composition of the basalt-glasses differs from that of the basalts, * Petrographische Studien an den Basaltgesteinen Bohmens, p. 182. t Die Gesteine der Sababurg, (Cassel, 1871) p. 41. ¢ Annales des Mines, vol. xiii. p. 369 (1851). In the analysis of the basalt- glass as given by Delesse, the total (99°53) does not agree with the figures of the column, which amount, as above, to 99°43. A misprint may have occurred in the column itself. Q.J.G.8. No. 155. 2 454 PROF. J. W. JUDD AND G. A. J. COLE ON THE the former being of more acid composition than the latter. It is conceivable that in a dyke the crystals brought up from below may be made to congregate in the central part of the mass, while its sides contain more than their due share of the fused liquid material in which these crystals are entangled. In this way, since we now. know, from the researches of Rosenbusch and others, that the glassy magma of a rock is often of far more acid composition than the crystals imbedded in it, the selvages of some dykes may have a higher silica-percentage than the rock in their central portions. But that this is ever the case we have as yet no definite proof; and, as we shall now proceed to indicate, there is another and a more probable mode of accounting for the more acid composition of most basalt-glasses. It is a well-recognized fact that the acid rocks more readily pass into a glassy condition than those of basic composition. Obsidians or rhyolite-glasses are common and widely distributed rocks ; trachyte-, andesite-, and phonolite-glasses are less common; and basalt-glasses, as we have seen, are comparatively rare. This fact is probably accounted for by the circumstance, which has been recog- nized by chemists *, that mixtures of silicates in which silicates of the alkalies abound more readily assume the vitreous condition than those in which silicates of lime, magnesia, iron, and alumina pre- dominate; and the acid rocks usually contain a large proportion of the silicates of the alkalies. Now, such being the case, we might expect that among basic rocks, like basalts, those varieties would be more likely to assume a glassy condition in which the silica-percentage is high, especially if the proportion of the alkalies at the same time were excessive. A comparison of the various analyses of basalt-glass, such as those brought together by Mohly, Zirkelt, and others, shows this to be the case, the percentage of silica in these rocks being seen to vary from 50 to more than 56. The proportion of the alkalies is in most cases considerably in excess of what is commonly found in basalts, varying from 3 or 4 to 10 or 12 per cent. The quantity of water in basalt-glass appears to be small and variable ; its presence may probably be regarded as accidental. In this respect basalt- glass strikingly differs from the palagonites, which contain from 11 to 25 per cent. of water. The rock known as hydrotachylyte, which is regarded by most petrographers as a nepheline-basalt glass, contains, it is true, nearly 13 per cent. of water, according to the analysis of Petersen$. Possibly, however, this rock must be considered to some extent a product of alteration, as was believed by Mohl ||. We may perhaps conclude that as a general rule the basalts which show a tendency to assume the vitreous condition are those s * Miller, Elements of Chemistry, part ii., Sth edition, p. 416 (1874). Tt Die Gesteine der Sababurg in Hessen, (Cassel) p. 41. Lehrbuch der Petrographie, vol. ii. p. 305. § Neues Jahrbuch fir Min. &. 1869, p. 32. || Die siidwestlichen Auslaufer des Vogelsgebirges, Theil 1, p. 21. BASALT-GLASS OF THE WESTERN ISLES OF SCOTLAND. 455 with a high percentage of silica and an unusually large proportion of the alkalies. The basalt-glass of the Beal near Portree seems to be an illustra- tion of this rule, as does the similar rock of Lamlash, Delesse’s analysis of which has been already cited. An analysis of the Beal rock, made in Dr. Frankland’s chemical laboratory, under the super- intendence of Dr. Hodgkinson, F.I.C. &c., was as follows :— Silica Mit ao ett e0 9d: 52°59 Adammmnaigetod xcthi eid 4 1733 PerricsOxidey 2: ).6.802 24 11°14 Manganous Oxide ...... 0-66 Tiga? BAS8R sic teases Sve. sR 6:47 Maomesian 7 chit ies 2°62 Dodaelies. wwe car bain 4-24 POLAR uty, WEL Ye 2°40 Loss on ignition ........ 3:27 100°72 Traces of copper and barium were also detected in this rock. The high percentage of silica and the large proportion of the alkalies are especially noteworthy in this analysis. But certain of the varieties of basalt-glass found in the Western Isles of Scotland are especially interesting to the petrographer, as showing that even the basalts with a low percentage of silica do sometimes, though rarely, pass into the vitreous condition. We have already pointed out that the varieties of basalt-glass of Gribun in Mull, Screpidale in Raasay, and Sorne in Mull, differ from those of the Beal in Skye and Lamlash near Arran by their high specific gravity. The density of the three first-named rocks is 2°82, 2°84, and 2°89 respectively, and that of the two last-named 2:72 and 2°78. All the foreign basalt-glasses recorded have densities ranging from 2°5 to 2:7. It became therefore a matter of interest for us to determine whether these exceptionally dense basalt-glasses of the Western Isles of Scotland were not vitreous forms of more basic basalts than had yet been found in this con- dition. Mr. J. J. H. Teall, F.G.S., has kindly undertaken the partial analysis of these interesting materials; and his results are quite in accordance with our anticipations. Mr. Teall’s analyses show that the basalt-glass of Gribun in Mull contains 50°51 per cent. of silica and 10:05 of metallic iron, the basalt-glass of Screpidale in Raasay contains 46°68 per cent. of silica and 10°80 of metallic iron, and that of Sorne in Mull 47:46 of silica and 12°47 of iron. It is evident that the exceptionally high specific gravity of the last- mentioned rock is due to the large quantity of the oxides of iron which it contains. : The only basalt-glass which has been hitherto described with a composition approaching that of the Scottish rocks is the siderome- 2K 2 456 PROF. J. W. JUDD AND G. A. J. COLE ON THE lane of Iceland, which has 49 per cent. of silica. Its specific gravity, however, is given by Sartorius von Waltershausen* as 2-531. This low density is the more remarkable when we find that the proportion of oxide of iron is above 20 per cent., and that there is little or no water in the rock. If this determination of the specific gravity be correct, it would appear that this sideromelane differs in a very marked manner from the ordinary basalt-olasses. While we have proof in these very basic glasses of Scotland that even basalts with a low percentage of silica sometimes assume the vitreous form, yet it is at the same time made evident that the tendency to do so diminishes as the silica-percentage declines. The basic glasses of Sorne, Screpidale, and Gribun form much narrower selvages to the dykes than the more acid ones of the Beal and Lamlash. The vitreous rock of Beinn Shiant, which contains 58-67 per cent. of silica, is, judging from its composition, probably like that of Hskdale, not a true basalt-glass, but an augite-andesite-glass. Similar vitreous conditions of augite-andesite have been described by Fouqué and Lévy 7 from Santorin, and by Cohen t from between New Britain and New Ireland, this latter being a pumiceous variety. Cohen also refers a Sandwich-Island rock, showing remarkable effects under polarized light due to internal strain, to the pumiceous condition of augite-andesite-glass §. In the Sandwich Islands, as we have already pointed out, we find a striking exception to the rule that basalt-lavas only assume the glassy condition locally and as the result of rapid cooling. In those islands the whole mass of lava-streams appears to frequently consist of basalt-glass. If the older analyses of J. C. Jackson, B. Silliman, jun., and other American chemists were to be relied upon, the exceptional cha- racters of these Sandwich-Island rocks would seem to be sufficiently explained by their peculiar composition. According to these analyses, the Hawaiian lavas contain a proportion of silica varying from 39 to 52 per cent.; the proportion of alumina is small, and that of iron very great, sometimes from 17 to 33 per cent.; while the percentage of the alkalies is often very high indeed, occasionally over 20. But the more recent investigations of Cohen || have suggested grave doubts as to the reliability of these older analyses. The analyses of six glassy lavas from the Sandwich Islands have given Cohen, Wagner, and van Werveke closely concordant results, the average composition of these rocks being, according to their analyses, as follows :— * Vulkan. Gesteine in Sicilien und Island, p. 203. + Minéralogie Micrographique, pl. xxxv. fig. 3. t Neues Jahrbuch fir Min. &c. 1880, vol. ii. p. 37. § Ibid. p. 38. || Neues Jahrb. fiir Min. &e. 1880, vol. ii. p. 41. BASALT-GLASS OF THE WESTERN ISLES OF SCOTLAND. 457 RUbCaIete thr 0, GTS Mao eT J. GU 0710 a 11°44 Oxides of iron. .. 94.4258 12°62 Himenrs, eat ok RS ee HO275 IM CIOSIAS. .)6'5\sis'0) ob) T'59 Ipokasiags 2. 203 '2 ioe 0°67 SOG Ee Sa Oe a a 347 NER mise oes ee eae 0-98 This result is in such close agreement with the average composition of common or felspar basalts that we cannot hesitate to regard these lavas of the Sandwich Islands as referable to ‘“ basalt-glass.” Another very interesting discovery of Cohen’s is the fact that a lava of Niuafou in the Friendly Islands has a glassy structure, and a composition aimost identical with that of the Sandwich-Island layas *. While, then, basalt-glass is, on the European continent and in North America, confined to small and local occurrences, as the selyages of dykes, the small fragments thrown out of volcanic vents and cooled rapidly in passing through the air, &c., we have in the Sandwich Islands, and perhaps also in the Friendly Islands, ex- amples of glass of the same composition as basalt constituting the whole mass of lava-streams. The basalt-glass of the Pacific Islands differs, however, from that of Europe by being clear and transparent. In this respect it agrees with the fused Rowley Rag and similar artificial products, in which the iron is united with the silica, and is not separated in the form of magnetite, the reverse being so very commonly the case in the locally-developed basalt-glass of Europe. The lavas of the Sandwich Islands appear to be no less peculiar and exceptional in their behaviour during ejection. Their extreme fluidity permits of their being thrown into fountains 400 or 500 feet in height (a circumstance nowhere witnessed in the case of ordinary basalts), and of their being drawn out by the wind into the delicate threads known as Péle’s hair. All these facts point to the conclusion that the Hawaiian lavas are ejected at a much higher temperature than that at which the basalts of Europe and North America issued. The only mineral, indeed, in the Hawaiian glassy lavas which is not fused appears to be olivine. In-this peculiarity of their lavas the great Pacific volcanos present another and most interesting exceptional feature in addition to many which have been already pointed out. 7. Microscopical Characters. The microscopical characters of many varieties of basalt-glass have been described by Zirkel?, Vogelsang t, Mohl§, and other * Neues Jahrb. fiir Min. &c. 1880, vol. ii. pp. 36 and 41. t+ Untersuchungen iiber die mikroskopische Zusammensetzung der Basalt- gesteine, p. 182. t Die Krystalliten, p. 111. § Die Gesteine der Sababurg in Hessen, (Cassel) 1871. 458 PROF. J. W. JUDD AND G. A. J. COLE ON THE writers. These rocks are all found to consist of a glass, usually of a brownish colour, but occasionally colourless or greenish, in which various crystallites are distributed. Basalt-glass, even in its most vitreous varieties, is very rarely clear and transparent. Usually opaque inclusions are so abundant as to render the rock non-translucent. In most cases it is only by the most careful grinding of slices of the rock that it is possible to obtain sections sufficiently thin to exhibit its internal structure, and it is necessary to employ the most powerful sub-stage illumination to transmit light through them at all. In their great opacity the natural varieties of basalt-glass differ very strikingly from the artificially fused basalts, such as Rowley Rag. These artificial glasses are clear and of a rich yellowish- brown colour by transmitted light; and only faint traces of a ‘ olobulitic ” structure can be made out in them with the highest powers of the microscope. Some of the German varieties of basalt-glass, for example those of Bobenhausen and the Sababurg, have spaces of a similar clear brown glass, the crystallites being collected into skeleton crystals or spherulites ; but in all the British varieties we have studied such spaces of clear glass are rare, and all these rocks are characterized by their great opacity. Indeed, in the most perfect specimens of our natural basalt-glasses, devitrification appears to have gone so far as to have resulted in the separation of the whole or nearly the whole of the magnetite, the minute crystallites of which, scattered through the rock, render it perfectly opaque, even in the thinnest slices which it is possible to prepare. Fortunately this dust of magnetite is occasionally collected into nebulous masses (cumulites), and in the more devyitrified varieties into skeleton crystals ; in such cases the nature of the intervening glass spaces can be made out. The structures found in these basic glasses appear to be quite similar to those which have become so familiar to geologists from the study of obsidians, pitchstones, and the glassy varieties of the acid class of rocks. In some cases the crystallites and microliths exhibit the parallel arrangement characteristic of the banded and fluidal structures ; in other cases the crystallites are united to form various kinds of skeleton crystals, or the globular and often con- centric concretions known as spherulites. Sometimes the glass is traversed by numerous fine cracks, some of which are curved and concentric, giving rise to the well-known perlitic structure. Per- litic basalt-glass has been already described by Prof. Zirkel from Marostica in the Euganean Hills *, by Mr. Rutley from Slieven- alargy, county Down‘, and by Fouqué and Lévy from Graham’s Isle £. The basalt-glass of the Western Isles of Scotland is usually, to a * Zeitschr. d. deutschen geol. Gesellschaft, vol. xix. p. 776; also Rosen- busch, Neues Jahrb. fiir Min. &c. 1872, p. 141. t Journ. Roy. Geol. Soc. of Ireland, vol. iv. p. 227. t Comptes Rendus, 1878, March 25, BASALT-GLASS OF THE WESTERN ISLES OF SCOTLAND. 459 greater or less extent, porphyritic in structure. Basalt-lavas con- sist, as is well known, of a number of perfectly formed crystals of augite, olivine, felspar, and magnetite, entangled in a mass of un- erystallized material. These included crystals have probably been formed at great depths below the surface. After the extrusion of the lava, and as it slowly cools, crystals of the same minerals become more or less completely developed in the entangling magma. Careful study of such rocks, as Lévy and Fouqué have so well shown, enables us, in most cases, to clearly distinguish those crystals which have been formed at considerable depths from such as have separated from the magma, near or at the surface, during the cooling of the lava. The crystals formed at great depths in the earth are usually larger and more perfectly developed than those separating from the glassy magma during its cooling, and they often contain liquid- cavities and enclose other crystals. Moreover they are frequently broken and rounded at their edges, and have suffered great corrosion by being partially melted-up and having their substance absorbed into the glassy magma. This glassy magma often sends prolongations of its substance into such crystals, which are, indeed, sometimes completely honey-combed by these extensions of the glassy mass*. The facts described point to the conclusion that crystals formed under great pressure in the midst of a fused magma may, on the relief of that pressure, be attacked and dissolved by the magma in which they were originally developed. Many of the basalt-glasses of the Western Isles of Scotland furnish examples of porphyritically imbedded crystals, these being, in all cases, of the same kind as are found in the basalts with which the glasses are connected. Olivine and-magnetite are the most abundant of these porphyritically imbedded crystals; but augite and sometimes felspar also occur (Pl. XIII. fig. 2). The basalt-glass of the Beal, near Portree in Skye, exhibits very interesting examples of olivine and felspar crystals, the latter much corroded and eaten into by the glassy magma (Pl. XIII. fig.3). The rock of Lamlash contains large and well-developed crystals of both felspar and augite, much broken and rounded (Pl. XIII. fig. 6). The other varieties in the Scotch area appear to contain only por- phyritic crystals of olivine, much decomposed; and in some cases these are apparently rare. All the basalts which have vitreous selvyages appear to be of the class which contains a large amount of glassy residuum between the erystals ; and some of them may in fact be classed as magma basalts. In certain cases, indeed, they appear to have little or no in- _ dividualized felspar, and may be grouped with the Limburgites of Rosenbusch. The most perfectly vitreous type of these rocks, which is ex- emplified by the extreme edges of the selvages in the Beal dykes, exhibits only the merest embryonic crystallites scattered through * Similar instances of corrosion are familiar among the porphyritic crystals occurring in acid rocks such as rhyolites, pitchstones, and obsidians. 460 PROF. J. W. JUDD AND G. A. J. COLE ON THE the mass of brown glass. These crystallites appear, with the highest power of the microscope, as excessively minute globular bodies (globulites), which in some parts of the mass are crowded together into cumulites, leaving other parts comparatively clear and transparent (Pl. XIV. fig. 1). Sometimes an approach to a linear arrangement can be detected in these globulites. The cloudy masses exhibit in some cases the parallel grouping characteristic of the banded or fiuidal structure. In the basalt-glass of Lamlash this linear arrangement of the globulites is much more strikingly exhibited than in that of the Beal in Skye. Here we sometimes find the resemblance to strings of pearls which has led to such objects being called “ mar- garites.” Some of the globulites are seen to be transparent bodies ; and this is beautifully shown by the Lamlash rock (Pl. XIV. fig. 2). The basalt-glass of Sorne, Screpidale, and Gribun shows a further development of the embryo crystals. The fine globulites are seen to be collected into skeleton crystals quite similar in form to those found in so many slags, but usually of much smaller dimensions. In addition to these, we find in the Screpidale rock abundant transparent colourless rod-like bodies (belonites), and in that of Gribun curious examples of forms intermediate between skeleton crystals and spherulites, which occur in the Hawaiian lavas, and have received from Krukenberg* the name of ‘“‘ chiasmoliths.” As to the nature of the crystals which are thus found in an embryonic and partially developed condition in these glasses, much discussion has taken place. The minute opaque forms, which seem to have the symmetry of the cubic system, can scarcely be other than magnetite; but the larger and more transparent forms, such as are seen, for example, in the chiasmoliths, appear to be augite, while the belonites are probably for the most part felspar. The margarites in the Lamlash rock may not improbably be felspar crystals in their earliest stage of development. Spherulites occur in the rock of the Beal, and also in the some- what similar rock (augite-andesite-glass) of Beinn Shiant, Ardna- murchan. In some cases, as in that of the Beal near Portree, the change from an ordinary basalt which occupies the centre of the dyke to the glass at its sides can be traced step by step. The porphyritic or first-formed crystals are alike in both the basalt and the glass, and consist mainly of olivine and felspar, the latter much rounded and corroded by the action of the glass uponit. In the glass at the extreme edge of the dyke the globulitic dust, occasionally col- lected into cumulites, fills the whole mass; nearer the centre this glass has the globulites united into skeleton crystals of magnetite. The basalt near the glass contains a very large quantity of unin- dividualized glass, in which transparent microliths of felspar and augite can be made out under a high power. In the basalt of the centre of the dyke the felspar, augite, and magnetite are well * Mikrographie der Giasbasalte von Hawaii, p. 8 and fig. 30. BASALT-GLASS OF THE WESTERN ISLES OF SCOTLAND. 461 crystallized, but a large amount of glass of a brownish or greenish colour is seen between the crystals (Pl. XIII. fig. 7; compare fig. 8 and Pl. XIV. fig. 1). In the Lamlash basalt-glass the crystallites of magnetite, and probably of some other minerals, are numerous, and the basalt is only slightly more devitrified than the glass. This basalt is a true magma-basalt or limburgite, with porphyritic crystals of augite, olivine, and plagioclase felspar. 8. Summary of Results. From the foregoing descriptions it will appear that in the Western Isles of Scotland we have examples of a somewhat rare class of materials of considerable interest to the petrographer. These materials, though often classed as mineral species, are really rocks, and, indeed, constitute merely a local condition of certain types of basaltic lava. The names tachylyte and hyalo- melane, as applied to mineral species, ought therefore to be aban- doned altogether. The supposed distinction between tachylyte and hyalomelane, founded on their behaviour with acids, altogether fails in practice as a means of discrimination between the different varieties. Of the several names proposed for the rocks of this class, that of “‘ basalt-glass” appears to us to be the most convenient and least open to objection. ‘This name indicates its mode of origin and its relation to basalt, and the only possible source of error which we can anticipate from its use is its confusion with the glassy magma, or uncrystallized residue, found in many basalts—a substance which may be of totally different composition. We would advocate in the same way calling the glassy varieties of other rocks by similar names, as rhyolite-glass, trachyte-glass, andesite-glass, phonolite- glass, &c., the names of obsidian and pitchstone or retinite being still used for the types with vitreous and resinous lustre respec- tively, and the terms spherulite-rock, perlite, pumice, &c. being applied to varieties exhibiting special modifications of structure. Tachylyte may, in the same manner, be a useful alternative name for the basic glasses, to be employed in contradistinction to obsi- dian or acid glass; but there seems to be no reason for the reten- tion of the term hyalomelane. From the glasses of more acid composition basalt-glass is at once distinguished by its higher specific gravity. While ordinary obsi- dians (rhyolite- and trachyte-glass) have a density varying from 2°3 to 2°5, the average being 2:4 or under, the density of basalt- glass varies from 2°5 to 2:9, the average being 2°7. The basalt- glass of several Scotch localities is of exceptionally high density, between 2°8 and 2:9. The glass, when unaltered, is probably in all cases of less density than the same material in a more crystalline condition. The striking magnetic properties of basalt-glass enable us to dis- tinguish it from other vitreous rocks, as does also its remarkable opacity even in the thinnest splinters. Perhaps the most noticeable 462 PROF, J. W. JUDD AND G. A. J, COLE ON THE and distinctive of all its characters, however, is its easy fusibility and the nature of the product resulting from its fusion. The hardness of basalt-glass is perhaps generally rather less than that of the obsidians ; but the difference is not. sufficiently great to afford a ready means of distinction between these two types of rocks, though it may serve to distinguish basalt-glass from the hydrated substances and altered forms known as palagonite, which are some- times confounded with it. In chemical composition basalt-glass agrees with the rock, of which it is, in all cases, merely a local variation. The proportion of silica varies from 45 to 55 per cent., just as in the basalts ; but those forms of basalt with the higher proportion of silica appear most frequently and most readily to assume the vitreous condition. The varieties of basalt containing an exceptionally large quantity of silicates of the alkalies seem also to pass more easily into glass than any other. In their microscopic character the basalt-glasses appear to be generally distinguished by their great opacity. When cut into sec- tions sufficiently thin to be transparent, the abundance of crystallites and skeleton crystals of magnetite serves at once to distinguish them from the obsidians. Like other vitreous rocks, they frequently exhibit the porphyritic, the pumiceous, the banded, the fluidal, the spherulitic, and the perlitic structures. In the Western Isles of Scotland basalt-glass has only been found as a selvage to dykes of basalt. In other districts, however, it has been observed in various situations where rapid cooling has taken place, as in fragments ejected from volcanic vents, and the surfaces of basaltic lava-streams. [ Note, July 12, 1883.—During the discussion on the foregoing paper, and subsequently, our attention has been directed to several examples of similar materials occurring in different parts of the Western Isles of Scotland. Professor T. G. Bonney, F.R.S., has very kindly placed in our hands, for the purpose of study, a glassy mate- rial forming a selvage less than half an inch in width to a dyke which is seen near the stable in the Castle grounds at Brodick in the island of Arran. This appears to be a true basalt-glass; it has a specific gravity of 2°33, and a silica percentage of 53:96. In its microscopical characters the rock very closely resembles the basalt- glass of the Beal in Skye. Treated in the same way as the other specimens, 83°69 per cent. of the rock was dissolved in hydrochloric acid. It is very fusible, and its powder is strongly magnetic. At the time when our paper was written we had not seen Pro- fessor Heddle’s analysis of and notes ona Tachylyte from the Quiraing in Skye (Min. Mag. vol. v. p. 8). In an erratum to this article, published with Part 23 of the Min. Mag., the same author calls attention to a very interesting note on the basalt-glass of the Beal in Skye, published by Necker in 1840 (Edinb. Phil. Journ. 2nd ser. vol. xxix.), wherein the true nature and properties of the material are very clearly defined. Quart Journ .Geol.Soc Vol XXXIX.PL XIII, K75 G.AJ Cole del. MP Parker hth. Hanhart imp. i BAS Alii Goi. BASALT-GLASS OF THE WESTERN ISLES OF SCOTLAND. 463 EXPLANATION OF THE PLATES. Prate XIII. Fig. 1. Portion of selvage to the basalt-dyke of the Beal, near Portree, Skye, natural size, showing the finely columnar structure exhibited by the basalt-glass of this locality. The most glassy portion of the rock, forming the outside of the dyke, lies to the left of the figure. Fig. 2. Slice of the basalt-glass of Lamlash, Arran, magnified two diameters, and exhibiting the marked porphyritic character of the rock. The glass is perfectly black and opaque, as is always the case except in sections of extreme thinness; and the enclosed crystals are those of felspar, augite, and olivine, the latter much decomposed. [Figs. 3, 4, 5, 6 illustrate the corroded, rounded, and fractured condition of the crystals contained in the basalt-glasses and the magma basalts which they accompany in the Western Isles of Scotland. ] Fig. 3 is a felspar crystal from the basalt-glass of the Beal in Skye. X15 diameters. Fig. 4. An augite crystal in the basalt of the dyke to which the last-men- tioned rock forms a selyage. X 15 diameters. Hig. 5. A greatly corroded crystal of felspar from the same rock as fig. 4. x 15 diameters. Fig. 6. A group of felspar crystals from the basalt-glass at Lamlash, show- ing rounding and fracturing. x 10 diameters. [Figs. 7 and 8 illustrate the characters of the magma-basalt dykes which pass into basalt-glass in the Western Isles of Scotland. | Fig. 7 is taken from the centre of the dyke at the Beal, near Portree, in Skye, and is an ordinary basalt with a large proportion of glassy. base. Fig. 8, taken from near the outside of the same dyke, contains much more glassy material, the substance of the felspar being almost wholly unerystallized. The passage from the rock shown in fig. 7 to that represented in fig. 8, and again into the glass illustrated in Plate XIV. fig. 1, is of the most insensible character. Puate XIV. [On this Plate are placed for comparison drawings made from exceedingly thin sections of five of the varieties of basalt-glass of the Western Isles of Scot- land, viewed with a magnifying-power of about 500 diameters. Beside them is placed, for comparison, an example of the clear brown glass of Hawaii, viewed with the same objective. ] Fig. 1 is the basalt-glass of the Beal, in Skye. In it the magnetite dust is simply collected into cloudy patches (cumulites), leaving clearer spaces of the dark brown glass between them, Fig. 2 is the basalt-glass of Lamlash, Arran. In this rock the minute crystallites are collected into linear series forming beaded rods similar to the structures which have been called “ margarites.” Fig. 3, the basalt-glass of Screpidale, in Raasay, exhibits a more perfect separation of the crystallites of magnetite, the skeleton crystals thus formed resembling, except in their much smaller size, the forms found in many iron slags. Around each skeleton crystal an area of colourless glass is produced by the abstraction of the iron oxides. Fig. 4. In this basalt-glass, from Sorne in the Isle of Mull, the separation of the magnetite in the form of skeleton crystals is more complete, and the forms of some of the transparent crystals are beginning to appear. 464 PROF. J. W. JUDD AND G. A. J. COLE ON THE Fig. 5. In this rock (the basalt-glass of Gribun, in Mull) we find that, in addition to the perfect separation of magnetite in skeleton crystals, the formation of transparent crystallites (belonites) has gone on to a considerable extent. Fig. 6. A glassy lava from Hawaii, consisting of a clear brown glass with a little cloudy material in parts, and a few crystallites. Some of these latter resemble the central portions of the structures to which Krukenberg gave the name of ‘ chiasmoliths,” Discussion. Professor Bonnry expressed his sense of the value of the paper. Students of British petrology would be much indebted to the authors for this exhaustive notice of so interesting and rarea rock. Asa proof of its rarity, he said that he had examined numerous basalt dykes in Scotland and elsewhere before he found a satisfactory spe- cimen of tachylyte ; and that, by a fortunate accident, appeared to be in another locality. It was by the road to Goatfell from Brodick, in Arran, at the back of the outbuildings of the Castle. Two veins of compact basalt, about 1 foot thick, intrusive in an older decom- posed basalt, had an edging of tachylyte from 4 inch to 1 inch thick. As regards the Sandwich-Island lavas, he believed that the masses of tachylyte were especially connected with Kilauea (for ordinary basalts were common in the islands); might this be due to the ex- ceptional condition of the lava in the crater of Kilauea, such an enormous mass of molten material? Mr. Tratt stated that the comparative rarity of basalt-glass in nature appeared somewhat remarkable when the readiness with which basalt can be fused and made to assume the condition of glass by rapid cooling is taken into consideration. He suggested that the apparent anomaly might be due to the lower fusibility of the basic as compared with the acidic material. The lower fusibility necessarily involved a greater prolongation of the time during which the conditions were favourable to crystal-development. The forma- tion of basalt-glass on the large scale in Hawaii might be due to the fact that the material before being ejected is cooled throughout its mass by convection-currents almost to the point of consolidation before being erupted as lava. According to this view the average temperature of emission of the glassy lavas of Hawaii would be less than that of a lava cooling to normal basalt. With regard to the crystals referred to by the author he thought that the honeycomb-structure might in part, at any rate, be due to the mode of crystal-growth. Mr. Bavrerman expressed the satisfaction which the paper had given him. The analysis of the Beal rock indicated a remarkably large proportion of alkalies. He had recently investigated a case of the formation of crystalline silicates and aluminates in a slag by the action of blast-furnace gas containing alkalies and oxides of zinc and manganese upon the firebricks of a hot-blast stove. The high density of the basalt-glass was probably due to separated magnetic oxide of iron. The structure, when seen under a high t Journ.Geol. mal i Qua x 500 conga e x 500 5 lanhart imp G.AJ.Cole del. M.-P Parker hth. BASALT-GLASS. BASALT-GLASS OF THE WESTERN ISLES OF SCOTLAND. 465 magnifying power, was analogous to that of certain iron-furnace slags, which also vitrified at the edges. He thought the imperfections of the crystals due to enclosures of glass in their formation rather than to subsequent corrosion. Mr. W. Mure spoke of the rarity of ieee and said that the dykes could not have run for a long time; if the glass occurred, it would be a sign of rapid cooling. Prof. SreLey asked how those basalt-glasses of Scotland differed from those of the basalts of Europe which differed chemically and mineralogically from them. The Sandwich-island glasses were very rich in iron. Might the greater abundance of the glass be attri- buted to the iron or to the rate of motion? He called attention to the fact that the basalt-glass of the Sandwich Islands adhered to the branches of trees without scorching them. Mr. Koc said that he had found a tachylyte in Skye, and near the leaf-beds in Mull; it occurred in bean-shaped masses and strings, generally of great thinness. He had made experiments in Siemens’s furnaces. He found that slag was partly affected by the amount of moisture contained; and this was confirmed by experi- ments with slag. If run into moist sand, the lower part of the slag was glassy. He therefore attributed the glassy condition to the action of steam. If one slag was poured on another, a glass would be produced at the point of junction. At the Sandwich Islands he thought that the basalt was at a very high temperature. The opaque glass was formed at the lower temperature in experiments. In experiments with crystals dropped into molten slag the edges were fused. Prof. Jupp said that similar glassy lavas flowed from Mauna Loa and Kilauea; so there did not seem to be any special reason for assigning the glassy state of the lavas to the peculiar conditions at the latter crater. He thought it more probable that the forms in the crystals were due to corrosion than to inclusion. A high per- centage of alkali was common to several of the basalt-glasses of Germany. ‘The question of the composition of the basalt was discussed in the paper. Cohen, however, had proved that there were glassy lavas identical in composition with true basalts. The rock Mr. Koch had found had not been proved to be a true basalt- glass. The quantity of moisture varied much in analysis, and truly vitreous rocks, like obsidian, seemed to be less rich in water than subvitreous ones, such as pitchstones. 466 C. J. WOODWARD ON A GROUP OF 25. On a Group of Minerats from Littmsnart, Sartor. By C. J. Woopwarp, Hsq., B.Sc., F.G.S., Lecturer on Chemistry and Physics, Birmingham and Midland Institute, Birmingham. (Read May 9, 1883.) Tue Carboniferous Limestone at Lilleshall has for very many years been worked, partly as a flux for iron-smelting, partly as a source of lime for agricultural purposes. Various workings in the neighbourhood have, from influx of water, been abandoned ; and there remains now but one mine, known as Stump Leasow*. It is situated by the canal, near a spot marked ‘Old Farm ” on the one-inch Survey map. There are two distinct beds of limestone, known as White and Grey, lying one above the other, separated by a distance of 40 yards. The upper bands of the lower bed, or grey, are full of verticai joints, which in many cases, speaking from memory, run from six inches to a foot in width. It is in these joints that the minerals occur. I have met with the followimg mineral species, put approximately in the order of frequency or abundance, commencing with the least frequent : -— | Quartz. | Hematite. Hrubescite. Barytes. Copper pyrites. Calcite. Tron pyrites. Dolomite. Quartz is extremely rare. I have, in fact, met with but one minute crystal. It showed under the microscope both prismatic and pyramidal faces. Erubescite. Only a few minute patches in the massive form have been met with. Copper Pyrites is met with occasionally. It occurs in more or less defined spheroids. One specimen has a peculiar radiated structure. Tron pyrites is not by any means common. Itoccurs in radiating masses of from $ to 13 inch diameter, showing faces of the cube, octahedron, and other forms. I have one specimen of the decom- posed iron pyrites, which in old mineralogical books is called “ hepatic pyrites.” This hepatic pyrites is in cubes and octahedra. Hematite. Occurs in minute chocolate-coloured semiglobular masses, made up of extremely minute crystals, also in an almost continuous film, coating the calc-spar and other crystals with a coppery sheen. In some cases the cavities containing minerals have the hematite much more thickly deposited on the surface * Thave just learned that this mine is now closed in consequence of the expiration of the Lilleshall Company's lease. MINERALS FROM LILLESHALL, SALOP. A467 lying lowest in the stone ; but I have not satisiied myself that this is general, Barytes.—Occurs in pink, lamellar, somewhat radiating masses, which cleave easily parallel to the 001 and 110 planes of Miller. At their free surfaces transparent crystals are developed, having the forms 001,110,and 101. Besides this mode the barytes also occurs in small transparent crystals (4-2 inch long). The prisms are chisel-ended, being terminated by the planes 110 and 110 of Miller, while the sides of the prisms are contained by the forms 001 and 012. Calcite.—In the uppermost bed, called by the miners “ cockles,” this mineral occurs in clustered groups of crystals, each crystal being a steep scalenohedron with a rhombohedral summit. The grouping commonly takes the form of a steep three-faced pyramid. The faces of the rhombohedral summits are striated in the direction of a line bisecting the angles formed by the edges of the rhombohedron. Some attempts were made to determine these faces; but the striations prevented any satisfactory measurement. The clustered crystals of calcite have a beautiful ice-like appearance. Some of the calcite in large masses has a pink colour, due to manganese. A portion of this pink calcite yielded Manganous oxide (MnO) ........ 1:20 per cent. Ferrous PME O) 77h uk SSO. vag The calcite also occurs in more or less spherical cavities in the lowest bed of stone; and these invariably contain the mineral in pointed scalenohedrons (Dog-tooth spar). Dolomite.—Is the most interesting mineral of the group. It occurs in nodules, which are apparently made up of a succession of lamine of varying diameters. These lamine crystallize at their edges in minute rhombs with curved faces, resembling pearl-spar. The substance of the nodule is cream-colour, but the crystalline surface is often brown from oxidation. The minute rhombs do not admit of measurement. Two samples of the typical nodules freed from calcite were analyzed. 13, at. | Sp. gr. 2°95. Sp. gr. 2°92. Warbou @dioxade (CO, ) ..22. 2008. 6.5 0-0-2 | 42°29 44-00 Ferrous oxide (FeO) .................. 21°49 12°51 Manganous oxide (MnO) ............ 1:02 2:08 PAHS (OBO) ooo 2.2.2 cer ade yh tebe sos tote 29°06 34°35 Mapnesia,( MeO) 2. 2... kates scones 5°96 7:64 99°82 100°58 x 1 contained ‘07 per cent. silica not included. + 2 is the mean of two analyses of the same sample, in which the following were the differences from the mean :—CO, 0:18, FeO 0°17, MnO 0:06, CaO 0°14, MgO 0:00. 468 _C. J. WOODWARD ON A GROUP OF Calculated into carbonates, these give i a. 2 | Calcic carbonate (CaCO,) ...........- 51-89 47°59 61°34 | Ferrous carbonate (FeCO,) ......... 34°62 34°74 20°15 _ Manganous carbonate (MnOO,) ... 165 213 3°37 Magenesic carbonate (MgCO3) ...... 12°52 13°73 16:04 100°68 98°19 100-90 No. 1 approximates in composition to a specimen of ankerite from Admont, Styria, given in Rammelsberg’s ‘ Handbuch,’ and put by Boricky as Ankerit a. The analysis of the Admont specimen is given in column a; and had I made but one analysis of the Lilles- hall mineral, I should perhaps have felt justified in calling it ankerite. Analysis 2, however, was made from a specimen undis- tinguishable in appearance from specimen 1, yet the relative pro- portions of carbonates are widely different. Analysis 2 cannot be made to coincide with either of Boricky’s values for ankerite*. The specimen, however, approximates to the formula 3 CaCO, +(FeMn) CO,+MgCO,, which requires CaCO, 60°00 per cent., FeCO, 23:20 per cent., and MgCO, 16°80 per cent., or a difference of 1°34 per cent. in the CaCO,, 0°32 per cent. in the (FeMn) CO, and 0°76 per cent. in the MgCO., and would properly be described as a ferriferous dolomite. - * Boricky (Mineralogische Mittheilungen von T'schermak, 1876, Heft i. p. 47) gives the general formula for ankerite as (CaCO, +FeCO,) +2(CaCO,+ Mg00,), where r=}, l, 4, %, $, 2, 3, 4, 5, 10; so that the possible ratios of carbonates are CaCO, =100 («+1), KeCO, =116, MgC0,=842. Putting x as 3, 1, $, &c., and calculating percentages, we have the following possible values CaCO.,. FeCO,. MgCoO,. =i 48°70 37°66 13°63 c—! 50:00 29:00 21:00 c—_ 50°54 25°16 24°30 ep 50°81 23°57 25°62 cs 51:05 22°18 26°77 c—2Z 51:37 19°87 28°77 c=? 52°09 15:10 32°81 x2=A4 52°52 12:18 35°30 L=) 52°81 10°21 36°97 x=10 53°50 5°64 40°85 MINERALS FROM LILLESHALL, SALOP. 469 Discussron. Mr. Hvpiesron remarked that since iron and magnesia were freely interchangeable elements, it seemed hardly necessary to base mineralogical distinctions on variations in their respective amounts, The Avrnor accepted the views expressed by Mr. Hudleston, Q.J.G.8. No. 155. 21 A470 “PROT. T. G. BONNEY ON A SECTION 26. On a Secrion recently exposed in Baron Hit Park, near Brav- maris. By T. G. Bonnry, M.A., F.R.S., Sec. G.S., Professor of Geology in University College, London, and Fellow of St. John’s College, Cambridge. (Read May 23, 1883.) In the autumn of 1880 I observed by the roadside leading up the hill from the gates of the Beaumaris Cemetery towards Llandegfan some peculiar grits which reminded me much of certain beds ex- posed in the district near Perfeddgoed House and other places in ~ the neighbourhood of Bangor. The rock, which crops out here and there by the roadside (perhaps about one third of a mile from the road to Menai Bridge) varies from a green “ bastard” slate to a grit containing small grains of reddish felsite. These beds are suc- — ceeded, after a very short interval, by the normal greenish micaceous or chloritic schists of the region. Although convinced that this indicated an extension of the group of beds which in the Bangor area underlie the conglomerate now generally taken as the base of the Cambrian, such specimens as I could obtain might not have convinced others, and I waited for a more favourable opportunity — of examining the district. Last summer, while on a visit to some friends in the neighbourhood, I was informed by them that a new drive was in process of construction through the woods on the face of the hill mentioned above ; and I applied, as the grounds of the Baron Hill estate are strictly private, to the owner, Sir R. Williams Bulkeley, for permission to examine the sections. This was at once accorded to me most courteously, and the results proved even more interesting than I had anticipated. The new drive is carried along the face of the craggy scarp which overlooks the Menai Straits, and is obviously an old line of cliffs, the base, as we approach Beaumaris, being now separated from the water by a considerable tract of nearly level ground. The drive ascends gradually along the face of the scarp, and crosses by means of a bridge the lane already mentioned close to the outcrop of the aforesaid grits. On entering the grounds from the main road, some little distance north-east of the Ferry from Garth, we find greenish and purplish schists, overlain by a compact chloritic schist, which dip at a rather low angle, roughly 8.E. Then come other greenish and purplish schists with thinner bands of the more compact rock, followed, yet higher, by a fissile green micaceous schist. This is succeeded by a compact purple variety, traversed by quartz-veins and looking as if silicified. So compact is it in structure that it might readily be mistaken for a flinty argillite. These schists may also be seen, though in a condition less favour- able for examination, cropping out by the roadside below, and the last named is identical with a rock exposed in a knoll by the turn- pike, which had always been a great puzzle to me. ‘The dip here- RECENTLY EXPOSED NEAR BEAUMARIS. 471 abouts appears to be about E., or even a little N. of E., but it bends back again to the S.E. The last schist exposed is of a greenish colour and rather compact structure, with some epidote and quartz- veins, which dips roughly 20° E.S.E.* Here we are on the higher ground, a little west of Gallows Point, overlooking the strip of flat land already mentioned. These schists are cut by two or three basalt dykes f. I have had slides cut from the principal varieties of the schist for mi- croscopic examination. The purplish schist (about 150 yards from the entrance to the drive) consists of quartz, a flaky green mineral, and an iron-oxide (probably hematite) associated with a little manganese. The quartz is in irregular granules of chalcedonic aspect, I believe developed zn situ. The other two minerals are to some extent scat- tered over the slide, but are also associated in wavy bands. In part the former may be a chlorite; but I am disposed to regard most of it as a hydrous mica akin to biotite. JI have examined these mine- rals with considerable care, but cannot say that I feel much confi- dence in determinations of the optical characters of these small crystallites, as there are so many obvious sources of error. In the compact chloritic rock, near the entrance-gate, there is more of the green mineral, some of which more closely resembles a chlorite, and less quartz and ferrite; calcite also occurs in a vein. The highest schist exposed in the drive has a general resemblance to the first * An interesting section is exposed in a pit near a house called, I believe, Pen-y-Parc, a short mile from the cemetery-gate. Here is a compact greenish- white quartzite overlain by a dull greenish-grey schist. The upper surface of the former is rather uneven; but I incline to consider this the result of a bend- ing of the beds, and not to indicate an unconformity. The former consists of quartz-grains of variable size, subangular to rounded, the larger being the most rounded, imbedded in a minutely granular quartzose matrix, the whole being occasionally cracked and recemented by very minutely erystalline quartz, in which are occasional flakes of a pale greenish micaceous mineral. The quartz- grains are generally crowded with small cavities, in very many of which are minute moving bubbles, occasional small prisms of a pale green mineral, and fine “‘nairs” like rutile. The schist is less altered than one would expect ; it consists of small subangular grains of quariz in a matrix of quartz and the usual filmy green mineral, with an occasional larger flake of pale mica. The whole has evidently been much compressed. Some twenty feet or so of the quartzite are exposed, and a little lower down the hill schists resembling those already described crop out. Macroscopically one would correlate these beds with the other schists; but the dip, which is about N.N.E., does not agree very well, the quartzite is not seen on the scarp below, and the retention of a fragmental condition, probably original, in the upper schist is a good reason for hesitating. The quartzite has a considerable resemblance to that on Holyhead Mountain, though in the latter bubbles seem to be much less frequent. Both, however, may very well have obtained their materials from such rocks as the granitoid series near Ty Croes. t The last seen of these consists of a plagioclastic felspar well preserved, with rather large angles (up to 30°) between the extinctions of successive lamelle. In these are many colourless belonites. The augite has lost its characteristic aspect, consisting of a mass of granules partly dusky, partly bright-coloured with crossing nicols; there are many scattered rods and wel:.-defined sharply angular crystals of hematite or perhaps, in some cases, ilmenite. The minerals appear to have consolidated in the inyerse order to that in which they are described. by 472 PROF. T. G. BONNEY ON A SECTION described, but has less ferrite and a greater proportion of the mica- ceous constituent in rather large folia. The compact purple rock, resembling a flinty argillite, shows a clear matrix, studded with belonitic crystallites of a pale yellowish-green colour and with inuu- merable granules of the above-mentioned ferruginous constituent, arranged in wavy paraliel bands. The matrix is siliceous, partly chalcedonic quartz and partly, I think, opal. Though the consti- tuents are very minute, the rock appears to be rightly considered a true schist; and I suspect its abnormal aspect is partly due to a subsequent infiltration of silica. I have compared these slides with specimens from near Menai Bridge, which much resemble the first and third (except that they have a hydrous white mica), and several specimens from the north-west of Anglesey, including those described in my Appendix to Dr. Callaway’s paper (Quart. Journ. Geol. Soc. vol, xxxvii. p. 234, Group B). All these have a strong general re- semblance one to another. | Beyond the last-named schist is an interval of about 60 yards, where every thing is concealed beneath loose soil and vegetation ; and then comes an outcrop of a massive grit, hard, jointed, weather- ing to a brownish colour, and at the first glance rather like one of the intrusive masses of basalt further down the hill. It is at first fine in texture and dark in colour; then it becomes a little coarser, and is full of rolled grains, up to about the size of a small pea, of felspar and felsite, which weather white. More angular fragments of a very compact black rock, up to nearly 2 inches in diameter, now and then occur. These grits are exposed by the roadside for about forty yards ; their dip is not easily ascertained, seemingly it is to the east. There is then an interval of 18 yards covered by vegetation, followed by an exposure of a similar grit for about 16 yards. Another interval of 60 yards succeeds, and we then find a slightly finer grit, which passes into hard blackish and brownish argillite, traversed by small quartz-veins, and looking rather crushed. The dip here ap- pears to be about 25° EH. Grits and argillite alternate for about 100 yards ; and then, after another interval ef about 36 yards, we find a hard argillite, in the middle of which is a greenish grit dip- ping towards the E. and exposed for about 130 yards. . Rather more than a hundred yards from the last exposure of the above-described rock, the lane already mentioned is crossed by a bridge; and in the park on the other side we find a considerable area of rounded surfaces of rock, seemingly ice-worn, partly masked by earth. The first reached is a grit (generally similar to the one last seen on the other side of the bridge), banded with greenish and grey- ish argillite; this soon becomes coarser, and in it may be noted frag- ments of the black rock already described, of a green argillite, and pebbles of the purplish felsite so well known on the other side of the Straits. The dips here appear to be high, and to have changed their direction to the N.W.; but in deposits of this nature it is not easy to be sure. ‘Then succeeds, sometimes by a very abrupt transition, a coarse purplish grit or conglomerate (the fragments weathering nearly white), full of bits of the felsite. In one spot I noted a lenticular patch where the materials sud- RECENTLY EXPOSED NEAR BEAUMARIS. denly became extremely coarse, the felsite fragments being some 9 inches in diameter; the gene- ral appearance suggests that this may be an agglomerate rather than a conglomerate. There is also a dyke of a compact dark rock with scattered crystals of whitish fel- spar often about half an inch long, resembling very closely a dyke in a quarry above the Bangor Ceme- tery *. Beyond this spot, for some distance the drive crosses an open field, and the slope is over- grown with vegetation of long standing ; so, as there was no pro- bability of any more exposures that would be of interest, I did not follow it further. On the crest of the ridge, about 150 yards to the N.W. or W. of the above rocks, is schist of the ordinary type. _ I have examined five slides cut to illustrate this series of grits :— one, from the mass nearest the schists,to represent the normal rock of this part; another, from near the same, to examine more especially the included fragments; a third from the greenish grit associated with argillite, near the bridge on the western side; and a fourth to illustrate the red felsite grit east of the bridge, cut from a_ part where the fragments are often about the size of small peas. We find, in these, grains of quartz, fel- spar, and several varieties of tra- _ * T have not examined this microsco- pically, as I could not have got a speci- men worth cutting without much de- facement of the ice-worn surfaces. The “Cemetery” dyke has in its ground- mass a considerable family likeness to the one already described, but is a little coarser, and the augite is better pre- served. The larger felspar crystals ap- pear to be of the same species as those in the ground-mass, but are more de- composed, and partly replaced by secon- dary microliths. “PSITOS ‘soyAp qyesvq Juosordea spuvq yoryq ou, pur 419 (your T 09 spre QOg ynoqe opeog + ‘poyeroSSexo yon st sdoroyno oy} JO JYSTOY [eor4I0A OTT) 473 YT Ir wowwg wr anug mau ayn fq pasodxa spag ayy fo UorssavINng Jouwuay oy Huryo.ysngjr WONIIY YOIIAS | 474 PROF. T. G. BONNEY ON A SECTION chytic lava and lapilli, associated with finer materials, granules of iron-oxide and some viridite, most of this matrix being doubtless decomposed volcanic dust or felspathic mud. The quartzes vary from subangular to rather rounded, the smaller grains generally being the more angular. With a low power they appear rather clear; but with a 3-inch objective a considerable number of cavities are detected, generally very small and often associated, mostly con- taining very minute bubbles, which usually move freely. In a grain in the fourth of the above-named specimens the fluid appears to be stained a reddish-brown colour. The felspar is usually a little more rounded externally. It is rather decomposed, but I recognize in some cases orthoclase, and think this predominates, as I obtain but rarely indications of the characteristic twinning of plagioclase. Of the fragments of igneous rock there are several varieties, even in the same slide. Some are slaggy, and in great part completely opaque with opacite; others are crowded with elongated crystallites of felspar, exactly as in slides from modern trachytic lavas in my collection ; others show a cryptocrystalline structure ; others, brown- banded with ferrite, show a fluidal structure; some are micro- porphyritic, exhibiting grains of quartz or erystals of felspar. One fragment in the first-named slide, with a rather irregular rounded exterior bordered with black, appears to be ‘‘ micro-amygdaloidal,” the numerous little vesicles being filled near the exterior with opa- cite, and within with celadonite (or a green serpentinons mineral) and a clear mineral, probably a zeolite. The whole appear to me to indicate the presence of materials truly voleanic—lapilli and frag- ments of trachytic lava. Some, at least, of the materials are water- worn; and I should suspect that, as a whole, they have been trans- ported to their present position (though the cones from which they must have been derived were probably at no great distance), rather than have been showered down as they now lie from a volcanic orifice. The fragments mentioned above, in the second slide, were two, an inch or so in diameter, which lay almost in contact in the first-named grit. Both were a very compact dark rock, but one had weathered to a paler colour. The latter is a rhyolite, with scattered angular grains of quartz and felspar in a matrix crowded with minute gra- nules of ferrite and (apparently) with extremely minute crystallites. There seem to be traces of a glassy base, but this is uncertain. Not so, however, in the other fragment; this exhibits admirably a fluidal structure, wavy bands, almost opaque with granular ferrite, appearing in a clearer base. As the field, except for some scattered crystallites, remains dark with crossed nicols during a rotation of the stage, and does not sensibly vary its tint when a quartz plate is inserted, I consider that we have here a fragment which has retained its original glassy condition. I have compared these slides with a rather numerous collection from the district near Bangor which I described in a former paper*. The resemblance to some of these is very close. The fine grit macro- scopically most resembles that from a pit near Hendrewen; but * Quart. Journ. Geol. Soe. vol. xxxv. p. 309. = RECENTLY EXPOSED NEAR BEAUMARIS. 479 microscopically the similarity is less marked. These Anglesey spe- cimens correspond most nearly with the slides obtained from various parts of the series which extends from the coarse breccia at Tair- ffynnon Quarry up to near the horizon of the Cae Seri rock. The argillite associated with these grits does not resemble any of the specimens which I have examined on the other side of the Straits; but there I chiefly directed my attention to the coarser beds as more important for classificatory purposes, so that I lay no stress on this difference. Under the microscope it appears, at first glance, not unlike a volcanic glass, consisting of a fairly clear mate- rial irregularly tinged with extremely minute granules of ferrite with a slightly wavy banding. In this are scattered rather angular erystalline particles and some patches of a dusky granular mineral. With crossed nicols the ground-mass remains black, except for the presence of many minute crystallites of variable form and often frag- mental aspect. The exact nature of these I cannot determine: some look like quartz, others may be felspar or more probably a zeolitic mineral; the larger patches are granular in structure, in some cases concretionary, in others resembling pseudomorphs after a erystal of felspar. With ordinary light they are a pale dull grey, like the colour of very diluted ink; with polarized light they prove to be very dichroic, becoming almost black in one position ; with crossed nicols they change from a sort of buff tint to black ; and on selecting one of the most definite forms, I find that extinction takes place when its longer axis makes an angle of 45° with the vibration-plane of the nicols. I am unable to identify this mineral. The same mineral occurs occasionally in some of the volcanic fragments in the other slides. I should regard a volcanic mud as the most probable material of this rock. It does not at all resemble the ordinary argillites or porcellanites, and, if a little more altered, would, I think, be a very typical hialleflinta, 7. e. a rock of sedimen- tary origin by no means easy to distinguish from a compact felsite. The relations of these two groups of rocks are, I think, of impor- tance in reference to some matters of rather recent controversy. Their junction is most probably a faulted one, and no one accus- tomed to the study of rocks can doubt that the felsitic-grit series is much newer than the schists. The former rocks are but little metamorphosed ; the alteration in them is not greater than we find in such beds as those at or just below the base of the Cambrian in Caernarvonshire or Pembrokeshire, at Charnwood, or at the Wrekin ; the latter have undergone very considerable metamorphism. As a rule, no original constituent in them can be identified with certainty. I have examined several specimens from this great series of schists, which extends from the west of the Menai Bridge to the neighbour- hood of Beaumaris, and only in the case of the quartzite and schist of Pen-y-Parc have I detected original constituents. Once, doubt- less, mudstones of variable chemical composition, they are now true foliated rocks, their chemical constituents having entered into new mineral combinations. Changes of this kind may doubtless some- times occur under exceptional circumstances in a series generally 476 PROF, T, G. BONNEY ON A SECTION unaltered, as when the temperature of one part has been locally elevated by the proximity of large masses of igneous rock, or when it has been exposed to very exceptional pressure im the presence of water at a slightly higher temperature than usual; but here there is no reason for suspecting any agent of change to have been more intense than usual. Further, at Garth Ferry we have Ordovician rocks * quite unaltered ; and on the Anglesey shore, on either side of the landing-place, is a quartz grit, faulted down against the schists, entangled with which may be seen some remnants of a slaty rock similar to that on the maimland. This appears to me to be rightly mapped by the Survey as Ordovician, and I should correlate it with the quartz grit and conglomerate near the opening of the sewer at Garth, which, as it seems to me, occurs there in association with the dark slaty beds, and not with any part of the Cambrian conglomerate. As, then, in the immediate neighbourhood, we find Ordovician rocks and even strata probably rather below the Cambrian practically unaltered, we cannot hesitate to consider these Menai schists Archean, and should not be disposed to refer them to the newest period in this great group. But these Menai schists have a general resemblance to the bedded schists which abound on Holyhead Island (as already described by myself and others) and in the intervening districts of Anglesey. I think no petrologist could hesitate, in the absence of any evidence of special local metamorphism, to refer all the above roughly to one group. It would therefore seem vain to speculate on any “gnarled rocks,” identical in constitution with these, being by any possibility of Ordovieian or even of Cambrian age. It might perhaps happen that a small patch of one of the older Paleo- zoics, locally nipped and tremendously squeezed, should exhibit a ‘‘universally slickensided” appearance, which would cause some macroscopic similarity to these true schists; but I doubt whether, even then, there would be much real correspondence when they were examined microscopically. Here, at Baron Hill, we have almost side by side true schists and rock which all must admit to be at least older than the greater part of the Cambrian; yet, as we have seen, the latter is almost unaltered. We may, then, I think, safely regard these micaceous and chloritic schists of Anglesey as Archean. By some authors they have been claimed as Pebidian. Here, however, the question arises, What is Pebidian? If we take as its type the series on which the name was first conferred, the group beneath the Cambrian conglomerate at St. David’s, we may at once repudiate the identification f. That, like the rocks of Charnwood Forest, is a group largely consisting of volcanic materials and comparatively slightly altered. To such a group we may assign the series in Caernarvonshire between the Cam- brian conglomerate and the great rhyolitic masses (which, indeed, I * Probably of about Llandeilo age (see Mem. Geol. Survey, vol. iii. ch. xxii.). + This paper was written prior tothe reading of Dr. A. Geikie’s paper “ On the supposed Pre-Cambrian Rocks of St. David’s;” but, notwithstanding the efforts of the Director General to efface the limit between Cambrian and Pebi- dian, the author sees no reason to alter any thing that he has written here. RECENTLY EXPOSED NEAR BEAUMABRIS. 477 should include with the overlying beds). With some part of this the felsitic grits at Baron Hill probably correspond. The term Pebidian, then, should designate a comparatively un- altered series (hypometamorphic, as it has been called. by Dr. Callaway) which does not appear to be much more sharply marked off from the Cambrian than the Ordovician from the Silurian. It is especially characterizéd by an abundance of volcanic material, chiefly ef an acid type. So far as we can conjecture, it appears to have commenced (I speak, of course, only of Britain) by an epoch of vol- canic activity, when, from orifices opened probably on an old land surface of the more ancient Archean rocks, flows of glassy lava and great masses of trachytic scoria were discharged—and to have been followed by gradual subsidence, which probably became still more general in the Cambrian period. [Note, June 26, 1883.—Since this paper was written I have received, through the kindness of Dr. J. W. Dawson, F.R.S., a very interesting series of small specimens of Huronian rocks from Canada. I have not yet had time to examine the microscopic structure of these; but macroscopically the resemblance of not a few of them to certain of the Pebidians of St. David’s and some of the rocks from Charnwood is most remarkable. They are, how- ever, wholly different from the true schists described in the above paper. | } These Anglesey schists (which present considerable resemblance to certain rocks recently described by myself from the Lizard district, and could perhaps be paralleled by others from the older part of the Pietra Verde group of the Alps) must be regarded as distinctly older than the typical Pebidian. ‘The correlation of them with the last- named group was due, I think, to two misconceptions—an erroneous identification of an Arvonian series in this part of North Wales, and an overestimate of the amount of metamorphism in the St. David’s Pebidians. As regards this correlation I see no escape from the following dilemma. If there is Arvonian at Ty Croes (Anglesey), then the schists of Holyhead Island and the Menai must either be much older than the Pebidian or the term must cease to have any classificatory value, as it would include rocks so very dissimilar in their amount of metamorphism. In the above remarks I have not attempted to correlate precisely the Baron-Hill grit with the beds on the mainland. All that I maintain is, that it cannot be newer than the Cambrian Conglome- rate of Professor Hughes or older than the great flows of rhyolite. Any thing more than this would involve a digression into the strati- graphy of the beds near Bangor, which I reserve for a separate paper. JI may, however, observe that the Baron-Hill beds differ considerably both from the conglomerate which, for some distance, fringes the opposite side of the Menai Strait and runs inland to the east entrance of the tunnel west of Bangor Station, and from that on the top of the hill above the eastern tunnel, both of which are referred by Professor Hughes to the base of the Cambrian series. (For the Discussion on this paper, sce p. 485.) 478 PROF. T. G. BONNEY ON ROCKS IN 27. On the Rocks between the Quarrz-FELsItE and the CAMBRIAN Serres in the Neighbourhood of Bancor. By T. G. Bonney, M.A., F.R.S., Sec. G.S., Professor of Geology in University College, London, and Fellow of St. John’s College, Cambridge. (Read May 23, 1883.) Tus district has already been the subject of papers published in the Quarterly Journal of the Society *. The authors of these agree in accepting the coarse conglomerate exposed (for instance) on the upper part of the hill pierced by the eastern tunnel of the railway near Bangor, as the base of the Cambrian; but there is still some diversity of opinion as to the relations and extent of the sub- jacent series. The views on the latter point will be found respec- tively expressed in papers by myself and Professor Hughes in vol. xxxv.; and since the publication of the latter I have four times re- visited the district, leaving, I believe, but few exposures of the rock unexamined (except such as may occur unknown to me in private grounds). As I adhere to my original opinion as to the sequence of the beds, I now ask leave to state as briefly as possible the addi- tional evidence which I have succeeded in gathering. At the same time I am glad to take this opportunity of frankly admitting that in the part of his paper which treated of the relations of the con- glomerate and the granitoid rock at Twt Hill I was wrong, and Prof. Hughes was substantially right. Repeated examination of the district and of specimens under the microscope has convinced me that, notwithstanding the apparent passage of the one into the other, the conglomerate is much later in date than the granitoid series. Whether it be of the same date as the Cambrian conglo- merate in the Bangor area is, I think, at present by no means certain. I can hardly regret the mistake, for in the correction of it (as re- lated in Geological Magazine, decade 2, vol. ix. p. 18) I have learned much; and that I made it at all was, I believe, due to the last lingering impressions of the instruction which I had found in the writings of those whose teaching was regarded some dozen years ago as authoritative. Had I then commenced my petrological studies in a spirit of absolute scepticism, I should have been saved much time and some errors. The main point, however, at issue between myself and Professor Hughes with regard to the Bangor district is the following :—I had described in the district south of the fault, which runs nearly along the line of the road from Bangor to Caernarvon, a series of rocks in which occurred at least four well-marked beds of breccia or con- glomerate, which appeared to me to be so distinct that I could only conclude that they indicated horizons, and that the succession as a - whole was an ascending one. ‘Three of these, unless I misunder- * Vol. xxxiv. pp. 187 & 147, and vol. xxxv. pp. 309 & 682. THE NEIGHBOURHUOD OF BANGOR. 479 stand him—all, in short, but the greenish breccia which occurs on Bangor mountain, Professor Hughes considers to be only the Cam- _ brian conglomerate repeated by faulting, so that the series sepa- rating the quartz felsites from the conglomerate at the base of the Cambrian is reduced to the above-named green breccias and the gritty slates in association with them—that is to say, rather less than one half of the thickness which I should assign to it. Of the faults which are needed for this repetition I have not been able to find satisfactory evidence in the field. The principal argument for their existence appears to be the hypothesis that one conglomerate is equivalent to another, an argument which it is obvious stands or falls with the hypothesis. I commence, then, my revision of the district with the eastern edge of the great mass of quartz felsite near Brithdir Farm. In apparent succession to this (I could not discover any evidence for the intermediate band of slate inserted on Prof. Hughes’s map) we find (on both sides of the lane which mounts in a §.8.W. direction from the main road along the bed of the valley) a grit mainly (if: not wholly) composed of quartz, felspar, and felsite fragments, occa- sionally becoming conglomeratic and containing pebbles of felsite, which can be traced pretty continuously round the brow of the hill near Wern, until, on its northern scarp, we again come to the main mass of felsite. We pick it up by the high road in the valley below (on the other side of the fault), still in apparent sequence with the felsite ; we follow it by Taffarn newydd to Beulah Chapel, where it is still in sequence, and find it thus in the field-way leading to a farm opposite to the lodge of Gorphwysfa, and it is seen for the last time just above the west entrance of the western tunnel. This felsite grit, the constituents of which certainly seem to have been derived mainly, if not wholly, from the adjacent felsite, appears to be succeeded near the turning to Wern Farm by greenish grits and a little purplish slate, over which comes a rock rather like that at Tairffynnon quarry, varying from a green grit with reddish spots to a coarsish breccia or conglomerate more like that inthe above quarry. It is also worth notice that about 80 yards to the east of the above-named tunnel-entrance is a small outcrop of a similar rock, and many fragments are to be seen in a spoil- bank consisting of materials which have doubtless been obtained in making the tunnel. It seems, then, not unreasonable to suppose that this grit and conglomerate, which, for purposes of reference, I will call A, thus seen at intervals between two points about 12 mile apart in a straight line, is in true sequence with the felsite, and is not brought by a fault into its present position. The next breccia which I have described (we will call it B)is well exposed in a pit in the wood at Tairffynnon, and some overlying beds of a more slaty character are seen in the lane. As mentioned in my paper in the ‘ Geological Magazine,’ I have since traced it on the upper parts of the ridge to the N.N.W. (roughly) of that pit; and in 1880 I found that a new pit opened by the Caernarvon road 480 PROF. T. G. BONNEY. ON ROCKS IN had exposed a similar rock almost exactly at the spot where I had anticipated it would occur. At this place the dip appears to be rather to the N. of E., perhaps E.N.E., while the dips about Tair- ffynnon are generally about E.S.E., a twist towards the north- ward of which we have other indications near this fault. This breccia, then, retaining its characteristic aspect, has now been traced for above half a mile. The peculiar breecia (C), containing elongated fragments of purple slate with smaller bits of felsite, had been traced prior to 1880 from near Tyddn Dreiniog to Cae Seri; and I had stated that I expected it would occur in the neighbourhood of the Poorhouse. In that year I observed a pit (perhaps 200 yards to the south-east of the building, which had been overlooked at my former visit) in which I found a rock impossible to distinguish from the Cae-Seri rock. Here also the dip (difficult to ascertain) seemed to be about N.E., a direction which is confirmed by other observations in greenish grits and slates in the neighbourkood. During a brief visit in 1882 I found traces of a similar rock in the interval between this pit) and Cae Seri. This peculiar breccia (C) has now been traced for a distance of full a mile as the crow flies. In succession to this comes, I believe, the greenish-grey felsitic grit of the Hendrewen quarry, in which are occasional well-rolled pebbles of similar-coloured felsite. Under the microscope it is seen to consist of fragments (generally rounded) of quartz, felspar, and an acid lava. The last commonly show more or less of a fluidal structure, and, examined with polarized light, are crypto- crystalline ; but two or three are crowded with elongated crystallites of felspar, as may be seen in many trachytes. The materials are rather more decomposed than is usual in the other breccias. ‘There is also one rounded fragment of a quartz grit. The grit described from the quarry by the white cottages west of Minffordd (I think the spot called Nant Gwtherin on the map) has some resemblance to this from Hendrewen, and is probably higher up in the same series. So far as I can gather from Professor Hughes’s map and paper (vol. xxxv. p. 682), he regards the quarries at Hendrewen and at the back of the Poorhouse as both in Cambrian conglomerate (pp. 690, 692), but the one near Nant Gwtherin and the rock of Cae Seri as part of the Bangor beds. The district between B and C is one not easy to unravel. I have zigzageed over it, and found most of the outcrops to be greenish gritty rocks, sometimes pink-spotted; and near Brynllwyd Farm are conglomerates, one of which contains some fragments of purple slate like that in the Cae-Seri breccia. At this place (e.g. near Careg Hwfa) Professor Hughes again brings in the Cambrian con- glomerate. I cannot, however, say that the rock closely resembles any of the admitted exposures of the latter, though it has something in common with both Band C. Hence I think it more natural to regard these conglomeratic beds as merely local intercalations 1 in the intervening zone. THE NEIGHBOURHOOD OF BANGOR. - 481 Fig. 1.—Sketch Section to illustrate the General Succession of the Beds as maintained, by the Author. (Scale 2 inches to 1 mile.) 8.8.W N.N.E.; 5. VV. = im D mM iA J i t 1 I - ' ( | ' ae : ' ( i H 1 i | { 1 ' 1 SES == ! <= ee eet Qe =——S= \ pS = SS Sa aaa eee eY °* SS SS SS SS SS = CS eg ee EL For explanation, see Map, fig. 2. Fig. 2.—Sketch’ Map of Bangor district, founded on the Geological Survey Map, from which the Faults (except the hypothetical one N.W. of the Caernarvon Road) are taken. (Scale 1 inch to 1 mile.) EeeH Grits. KG Cambrian ? Tairffynnon Breccia (B). Es Cambrian conglomerate (D). Slates and pink grit. roe == Bangor series. “&3 Felsite Grit (A). ‘Sy Cae-Seri Breccia (C). = Felsite. The black lines indicate faults. 482 PROF. T. @. BONNEY ON ROCKS IN The Cambrian conglomerate (D), above the admitted Bangor Series on the hill pierced by the tunnel east of the station, is as different as it well can be from B and C, but somewhat resembles A ; from which, however, it differs in containing numerous pebbles of quartz and quartzite (exceeding in number those of felsite), and in possessing a more sandy matrix. If, then, A, B, C, and D are the same rock, we must suppose :-— (1) That whenever there is an exposure along the edge of the quartz felsite for more than 14 mile, either the whole Bangor series is overlapped by the Cambrian conglomerate, or the latter by faulting is exactly fitted onto the igneous rock, and is composed of mate- rials wholly, or almost wholly, derived from it. (2) That in the space of about 1+ mile, measured at an angle of about 45° with the general strike, the Cambrian conglomerate changes from a rock consisting almost wholly of felsite to the mixed but different and more brecciated types of B and C,.and then returns to the more similar bed D, with its well-rolled pebbles. The annexed section (fig. 1) represents diagrammatically the result of the observations described in the above remarks ; and when we regard the general agreement in dips and the lithological character of the rocks, it seems more natural to consider the whole an ascend- ing series, and to extend the Bangor beds of Professor Hughes down to the quartz felsites. On the map (fig. 2) I have endeavoured to record the same observations as well as can be done on one of so small a scale. As regards the district between the Menai Strait and the fault running §.W. from Bangor down the valley, which has frequently been mentioned above, I have little to add, notwithstanding re- peated examination, to what I have incorporated in the fore- going remarks. Above the conglomeratic grits therein mentioned, and the great conglomerate extending from the shore by Gored Gith to the eastern entrance of the west tunnel, we find nothing but greenish or purplish slaty beds, occasionally slightly gritty, which perhaps generally agree better lithologically with the normal Cambrian than with these underlying beds; but the prevalent dip appears to carry them under the conglomerate. This same con- glomerate, so well exposed for a considerable distance, as traced by Professor Hughes (over half a mile), is a little perplexing. Felsite forms a much larger proportion of its materials than is the case in the conglomerate pierced by the eastern tunnel; and it seems intermediate between this and the one near Gorphwysfa*. Still, as it retains its character so uniformly for so great a distance, and ter- minates within less than half a mile of the Poorhouse quarry, we can hardly, I think, regard it as on the same horizon as the breccia (C), or, what would be simpler, identify it with that in the Hen- * This resemblance, and the occurrence of a patch of conglomerate between the two, caused the perplexity about the position of that near Gorphwysfa, men- tioned in Geol. Mag. decade 2, vol. vii. p. 302, which was not removed till 1880. THE NEIGHBOURHOOD OF BANGOR. 483 drewen pit *. It does not, however, appear to me that the quartz and jasper grit and conglomerate west of the pier at Garth can be associated with the above. The following extract from my note- book explains the sequence of the rocks on this part of the shore :— ‘ W. of the new slate-pier are dark beds, resembling in the upper part those on the other side of the pier mapped as Arenig, and in the lower becoming gritty and containing bands of small pebbles only a few inches thick, the general dip being 10°-20° S.E. Then we come to green and purple bastard slates, faulted down, and in places much disturbed in consequence, the prevalent dip (not easy to ascer- tain) being apparently 40° 8.W. Just west of the old slate-pier is a bank of screes of bastard slate and fine grit, among which I found bits of green breccia, reminding me of those seen above Bryniau, though in a much finer condition. About 150 yards east of the bathing-place green-banded argillites dip at about 40° S.E., and slaty beds then continue to the bathing-house, when the conglo- merate sets in, which, so far as one can ascertain, has an easterly dip.” If these observations are correct, it is evident the whole district is completely smashed up by faults. I have, however, found a rock, seemingly just on the north side of the main fault in the Caernarvon-road valley (in a quarry north of the crossing of the road from Maesmawr to Brithdir), bearing some resemblance macroscopically and microscopically to that at Tair- ffynnon; and in a pit at the back of Caemabadden Farm is a greenish grit, which, in its coarser parts, contains bits of green slate and red felsite, and may belong to the upper part of the zone between B and C. If I am correct in the position I assign to the bed A, then the whole series north of the fault has a strike rather to the east of N.N.E.; so that all the upper part of the Bangor series would be cut out by the faults which bring down the very Cambrian-looking beds above the cross roads at Pen-y-chwintain. Until the outcrops can be laid down on a map on the 6-inch scale, it will be hopeless to. come to any satisfactory conclusion in this complicated district north of the fault, and not easy to bring into complete order that on the southern side. I trust, however, that I have succeeded in showing that my original reading is the more simple explanation of the facts observed both in the field and with the microscope—namely, that there is a general ascending suc- ‘cession in this district, and that under the name of the Bangor Beds we must include not only the green gritty slates and breccias assigned to them by Professor Hughes, but also a large lower group extending from A to a little above C. This group has probably derived much of its material from the denudation of the great masses of rhyolitic lava to the south-west, and the lapilli which are often present may have been derived from its associated cones. At the same time, seeing that these lavas appear generally to rest upon old gneissic and granitoid rock, the slaty fragments, often very * Which would be lithologically the least difficult, were not this grit ap- parently so completely “ sandwiched” between the Bangor grits and breccias and the Cae-Seri breccia. 484 PROF. T. G. BONNEY ON ROCKS IN angular, would not seldom be more simply explained by supposing that sporadic volcanic action still continued, and that indurated fragments of the finer sediments formed from the denudation of the somewhat older lavas and ash were ejected together with lapilli. I see no reason for insisting on any great interval of time between the rhyolitic lavas and the base of the Bangor beds. An uncon- formity there seems to be; but that in a volcanic series is of little moment ; and [ am of opinion that on the whole it would be better to include these quartz porphyries or quartz felsites (old rhyolitic lavas) with the Bangor group. This belongs to the disturbed episode anterior to the quiet subsidence which marks the non-volcanic sedi- mentary series everywhere recognized as Cambrian. From the latter we seem justified lithologically and physically in separating these more or less volcanic beds, and in including them for con- venience in the Pebidian group of Dr. Hicks; but the interval in time need not have been a very enormous one. Below the rhyo- lites, as it seems to me, is the great gap in the record. They appa- rently broke forth, as. the older basalts in Auvergne, upon a plateau of crystalline rocks which belong to some of the earlier, as these do to the last, chapters in the Archzan volume of the Geological History. i Note.—As I believe that no analysis has been published of any specimen from the great masses of quartz-felsite, which occupy so considerable an area in this part of North Wales and contribute so largely to the rocks immediately overlying them, I take this oppor- tunity of giving one, for which I have to thank my friend Mr. J. J. H. Teall, F.G.S. It was made from a very typical specimen of the common purplish variety, collected from the crag near Brithdir Farm (No. I.). For comparison, I place beside it an analysis of the “ devitrified pitchstone” of the Wrekin district (No. I1.), of a felstone (of Arenig age) from Aran Mowddwy (No. III.), and of a felsite of Bala age from the Lledr valley (No. IV.). The last, an analysis of a specimen of the ground-mass of the peculiar spherulitic rock described on p. 290 of vol. xxxviil. of the Quarterly Journal, was kindly undertaken forme by Mr. F. H. Hatch of University College, London, when I was engaged on my paper ‘“‘ On some Nodu- lar Felsites in the Bala Group of North Wales;” but the result, owing to an accidental delay, did not reach me in time for publication. The close correspondence between I. and II. is most remarkable. The Ordovician felsites (III. and IV.) agree fairly together, and differ from the other pair in a larger percentage of SiO,, a less per- centage of alumina and alkalies, and in an excess of Na,O over K,O 2 ° THE NEIGHBOURHOOD OF BANGOR. 485 No. I. No. II. No. III. No. IV. SO). | ccaaamanadenseen 12°57 72°18 83°802 79°72 PAO) co. ave eemetn ec 13:64 14-46 7686 9°65 LIGK OSS Gaerne 2:28 1-78 tt 5°69 LOGO Ae) ae a set ‘91 “408 ase a ee eet owans owe | slight trace oe ais Bas AO os ae tassd esses 1:00 0:92 "896 v MO ies acne ssink's

Pak Be a. i, S Lie ee oy ie Ny \ ict 9 Bi to mii 4% ry © ) Di ‘ . \ 1\ Argillaceous as 1 Schist 1 ‘ gene oe 3 ey el Kennedy, ) 2 @ Mt 3" RA SS Ain, ‘ re = % 5 Range ai \s hi sYames 10) 2 oo as .% ‘ KG D 4 BS = : 8 ees 4” 5 hed Zz G@aartzite & TY pe ARSE Gain OO a ee ee te Ne ae ai inant Cet ieee aS 4 7 ose $k Sis Liver ub) Oy Zon} oy a ' \Schist Roc igMt.Clere ea 2, GS . 5 On iw Nic Grey' grit to OO, B \ ait: ° we YP *4.. Scher i, Quartzite y S ao The numbers indicate the positions where the specimens were obtained. The dotted line shows the approximate position of Mr. Gregory’s section. 27 Q.J.G.8. No. 156. 584 WwW. H. HUDLESTON ON WEST-AUSTRALIAN the geology of West Australia as far north as the Gascoyne river in latitude 25°S. Speaking of the Darling Range and the country to the eastward, Mr. Gregory says that the principal portion of West Australia consists of an undulating table-land of syenitic granite, which has a western face rising abruptly from a plain of small elevation to a height of from 800 to 1200 feet above the level of the sea, gradually ascending, for 200 miles to the eastwards, to 1400 or 1600 feet, and even as high as 2000 feet. Two diagrammatic sections were given, one, about 40 miles in length, in latitude 32° 8., where the edge of the syenitic granite is about 20 miles from the coast. The erystal- line rock, which in this latitude forms the western edge of the Darling range, is represented as being penetrated by numerous ‘‘ dykes ” of ‘ serpentine,” porphyry, and quartz. The second seetion is about 200 miles in length from west to cast, and may be said to include the entire valley of the Gascoyne river, in latitude 25° S., from its sources to the sea. From this section we learn that the outcrop of the crystalline rocks in this more northern region is about 90 miles from the coast, instead of 20 miles, as was the case on the more southern parallel, and furthermore that there is no such abrupt rise as occurs immediately to the east of the Swan river, in the higher latitude. On the contrary we learn that about the confluence of the Lyons and Gascoyne rivers, a very consider- able thickness of palzeozoic rocks reposes, as it were, on the flanks of the crystalline group, and thus serves to moderate the sharpness of the declivity. Annexed is a copy of Mr. Gregory’s second section (fig. 2, p. 585). The paper and sections by Mr. Gregory must, then, be regarded . as having laid the foundation of West Australian geology south of the parallel of the Gascoyne river, and, indeed, fully up to that river, whilst the collecting of Mr. Forrest, the present surveyor, has been limited to the regions upon, and to the north of, the Gascoyne river; as indicated in the map which accompanies this report. Beyond the fact that he has discovered a range, or, more properly speaking perhaps, a sort of continuous outcrop, trending N.N.W. for nearly 150 miles, which has yielded an interesting suite of Carbo- niferous fossils, there does not seem to have been any noteworthy discovery *. This fact, however, is in itself one of considerable im- portance, as it places the existence of a large sweep of Carboni- ferous rocks beyond the possibility of a doubt; whilst, owing to the poverty of the collection exhibited by Mr. Gregory, which only con- tained one coral (referred to Cyathophyllum), two or three species of Spirifer and Productus, and a few Encrinital stems brought from the Irwin river along with coal, the age of the coal-bearing beds of that river had even been questioned. Moreover, as will be seen subsequently, the Forrest collection is extremely interesting from a * Tn his letter to Sir Charles Nicholson, dated Perth, Nov. 4, 1882, he speaks of having collected many fossils, some of which he had forwarded to Hngland. 585 FOSSILS AND ROCK-SPECIMENS. Fig. 2.— Diagrammatic Section across a part of Western Austraha, in lat, 26° 15' S. (Length about 200 miles.) From Mr. Gregory’s paper in Quart. Journ. Geol. Soc. vol. xvii. p. 477. W. , EK. ° ~ H ae o oo by lee o oS o i a 9 : pp eta st Clive » 2 0) 2) Ores ee) Q |. Q¢q 30 a oo q Oo mars Ook 23 oS i= (e) [e) iQ) 2o ic ss 38 B ob 2 id ry Grey Plains. aa he pa 4 O's a's O's ' i : eg H a. Sand. | g. Carboniferous (?). 6 & c. Red drift-sand, 50 feet. h. Devonian (?) d. Sandstone. _| @ Metamorphic rock. | e. Cretaceous (?). In this group, which comprises both a white k. Granite, gneiss, and slate, with superficial deposits of granitic chalk-like (but non-caleareous) rock and ferruginous sand- breccia, ferruginous clays, thin beds of saliterous sandstones, __ stones, occur Ammonites, Trigonie, and Pecten. gypsum, calcareous tufa, and beds of nodular magnesian lime- f. Permian (?). | stone, 272 586 W. H. HUDLESTON ON WEST-AUSTRALIAN paleontological point of view, and affords a further testimony to the extent and importance of the Carboniferous formation on the Austra- han continent. Rock-specomens.—A few small rock-specimens accompany Mr. Forrest’s collection ; and as he has been careful to mark the localities on the accompanying map, it would have been possible to make a sort of guess as to the nature of some of the formations even without the aid of Mr. Gregory’s section. The specimens collected by Mr. Forrest, being for the most part from the basin of the Gascoyne river, enable us to test, as it were, Mr. Gregory’s section, which, on the whole, they seem to confirm. Subjoined is a brief description of the more important rock-speci- mens, with references to the localities whose petrology they are intended to illustrate. The character of No. 1 may be gathered from the numerous fossils as well as from the small rock-specimen so labelled. We thus arrive at the conclusion that the ‘‘ Range containing fossils” is, in the main, composed, of a limestone-grit varied by sandstones and flagey micaceous grits. Some of the fossil casts occur as a ferruginous fine-grained sandstone without lime, and sometimes as a dark hornstone or chert. Some of the Fenestelle occur in a flagey calcitic limestone which is tolerably pure. The corals also are calcareous, and the interior of the tubes filled with calcite. No. 2. There are three specimens with this label. One is the cast of an Orthis in a ferruginous fine-grained sandstone. The others are flaggy, fine-grained, and somewhat micaceous sandstones without a trace of carbonates. On turning to the map, we perceive that these specimens come from the 8.E. side of the Kennedy Range, facing the junction of the Lyons and Gascoyne rivers. Thus it is not improbable that here also is a sandstone formation homotaxially Carboniferous. The higher portions of this range are marked as possibly Cretaceous in Mr. Gregory’s section. No. 7. This lot comes next according to topographical arrange- ment. The specimens are derived from the opposite side of the Lyons river; between it, in fact, and the southern prolongation of the ‘‘ Range containing Fossils.” One is a soft, grey, micaceous, - flagey sandstone, and the other a dark grey, micaceous shale, with markings which may be fucoidal. I think that No. 7 represents the detached summit, immediately east of the Lyons river, marked e in Mr. Gregory’s section, and according to that section a Cretaceous outlier. No. 3 is from the east side of the “ Range containing Fossils,” between it and the northerly bend of the Gascoyne River, about the position where the crystalline rocks are first shown as coming to the surface in Mr. Gregory’ssection. Accordingly we find this specimen to be a coarsely crystalline aggregate of silvery mica with quartz, the rock being stained pinkish in places from oxidation of iron in the mica. Itis the most highly crystalline of all the rock-speci- mens. Those under the label Wo. 4 are varieties of quartz, some FOSSILS AND ROCK-SPECIMENS. 587 with pyrites. This exposure would seem to represent the granitic rise of the more southern parallel to which allusion has (already been made. Nos. 5 & 6. On either side of the next great bend of the Gascoyne river, and about 40 miles to the eastward of the last-noted exposure, are two hills, one on either side of the stream, known as Mt. Steere and Mt. James; the latter is marked in Mr. Gregory’s section as about 2000 feet high, and as composed of metamorphic rock resting on ‘‘granite” pierced by dykes. It is represented by specimens No. 6 of Mr. Forrest’s collection. One of these is a whitish quartzite or quartzose grit with a little mica and many specks and small crystals of magnetite. The second specimen is a very quartzose micaceous schist, or gneiss, similarly speckled with magnetite. These would seem to belong to Mr. Gregory’s metamorphic rock indicated in his section by the symbol z. On the other hand the rock of Mt. Steere, Vo. 5 of the collection, is a schistose mixture of quartz and kaolin (?), and may be regarded as belonging to the more highly crystalline series. Vo. 8, from Mt. Packford, is simply concre- tionary carbonate of lime. Sixty miles further east, and beyond the highest sources of the Gascoyne, is a hill marked Mt. Clere. No. 9, from this place, is a close-grained flaggy quartz-grit of a dun colour, which must be regarded as forming part of a sedimentary series but little altered in the direction of crystallization. Forty miles due north of this, at the head waters of the Lyons river, occurs a rock, No. 11 a, which may be described as a sort of yellowish grey phyllade. No. 10, from this district, is an opaque white chalcedony rust-coloured on the exterior, and No. 11} is opal. Quite in another direction, Vo. 12 is a kind of white flint. The beds from which this is derived may possibly be in the line of the prolongation of Mt. Kennedy Range. Palcontology.—There are afew forms of doubtful nature to which no further allusion need be made. The Actinozoa are very fairly represented. Indeed, considering that only one doubtful Cyatho- phyllum was known previously from West Australia, the Forrest collection may be regarded as rather rich in this respect, and the specimens are free from matrix and nicely weathered, so that the characters can be made out fairly well. Of the rugose corals (Zoan- tharia rugosa) there are several specimens of what is, in all pro- bability, a new species of Amplewus; one specimen of an Amplexus which is probably British, but new to Australia; and one specimen of a Zaphrentis, which may be new. Amongst the group of corals, if, indeed, they are corals, which used to be called “tabulate,” are several specimens of a branching form of Stenopora, which is probably identical with Stenopora tas- maniensis, Lonsdale. All the above fossils are Carboniferous or closely allied to Car- boniferous forms; but there is Just one specimen of the Favositide, which has a Devonian look about it. Indeed this specimen might 588 W. H.. HUDLESTON ON WEST-AUSTRALIAN almost be taken for the well-known Favosites polymorpha, which now figures as Pachypora cervicorms in correct lists of fossils. The above species has been quoted as occurring in the Lower Devonian of the Macleary river. This is the only trace of a thoroughly Devo- — nian fossil in the whole collection. Ifreally collected from the same beds as the others, its presence is somewhat singular. I can scarcely believe it to have been remanié from lower beds. Portions of crinoidal stems are numerous in the collection ; these probably belong to Poterrocrinus and Cyathocrinus. Along with these are many single “jomts” and smaller fragments occurring together with pieces of Polyzoa, &c., in the matrix of the larger fossils, forcibly reminding one of the contents of Carboniferous rocks at home. The Polyzoa also are well represented ; and besides forms hitherto recognized as abundant in Australia, are others whose allies must be sought in America. There are two species of the very curious genus Hvactinopora, only known hitherto, so far as I am aware, in the Lower Carboniferous of the Mississippi valley. Besides these are several specimens of Fenestella plebera, common in the Carboni- ferous Limestone of Ireland, and quoted from nearly every Carboni- ferous locality in Australia. One of the many varieties of Protore- tapora (Lenestella) ampla is also met with in the collection ; of this there are two specimens. The improvement in the list of Brachiopoda is not so great as in the lower forms of life just quoted ; but Mr. Gregory’s list is confirmed and strengthened. In Mr. Forrest’s collection there occur one species of Athyris, four species of Spirifer, of which two belong to the alate group, and two species of Productus, one of which may be a Strophalosia. ‘These are all from the “ Range containing fossils,” and there is a cast of an Orthis? from station No. 2, in all eight species of Brachiopoda. Here, again, the whole facies is strongly Carboniferous, as will be seen on referring to the table of fossils, though one or two species, which have a ercatt range both in time and space, are common to the Carboniferous and Devonian. The Conchifera are represented solely by Aviculopecten, of which there are two species. The specimen of A. dlawarensis is very fine. This fossil has a considerable amount of adherent matrix, consisting of a coarse red marly quartz-grit full of crinoidal fragments and of Polyzoa. Two fragmental casts of A. lomeformis occur in a brown, ferruginous, fine-grained sandstone. of the coal-bearing beds on the Irwin river, in lat. 29° 8, and on the Fitzgerald river, in lat. 34° 8, which were regarded by Mr. Etheridge as of Mesozoic age*, it is at least satisfactory to know that a thoroughly Carboniferous fauna occurs in the “ Range con- * See “ Description of the Palaozoic and Mesozoic Fossils of Queensland. Quart. Journ. Geol. Soe. vol. xxviii. p. 320. FOSSILS AND ROCK-SPECIMENS. 589 taining Fossils,” which extends for so many miles towards the tropic, north of the Gascoyne river. No outcrops of coal appear to have been discovered; otherwise the Surveyor would surely have found room for a specimen in his collection. But where there is such an extensive range bearing Carboniferous fossils, coal-seams of that age may reasonably be expected somewhere along its flanks. It may be worth noting that the Carboniferous beds of Queensland commence in lat. 26° §, and extend with some interruption almost to lat. 20° 8, on the Bowen river. Thus the latitudes on the east side of the Australian continent, corresponding with the position of the *‘ Range containing Fossils,” are just those which possess the great development of Carboniferous beds described by the late Mr. Dain- tree*, who tells us that “ whilst the affinities of the southern coal- field of Queensland are Mesozoic, a northern field of even larger extent has a distinct fauna, more resembling the Paleozoic Car- boniferous of Kurope.” He further states that in the lower strata Producti, Spiriferw, &ec., of true Carboniferous age, are associated with imperfect forms of plants resembling the G'lossopteris, Peco- pteris, &c., of the upper portion of the series. Numerous outcrops of coal had even then been noted in the group; but up to 1872 no commercial use had been made of them, owing to the difficulties of carriage. Having drawn certain inferences from what there is in Mr. Forrest’s collection, it may be permitted, though with more hesita- tion, to draw certain inferences from what there is not. It should be specially noted that no specimens of granite have been brought, so that the crystalline rocks are represented by schists such as num- bers 3 and 5. If the small rock-specimens are a fair sample of the country, it is evident that quartzose matter largely preponderates in the crystalline, subcrystalline, and plain sedimentary groups, whilst both opal and chalcedonic silica are not scarce. Notwithstand- ing the number of fossils limestone does not seem to be very cha- racteristic of the district, and volcanic rocks are entirely absent. With regard to the paleontological evidence, the only trace of a truly Devonian fossil is the Pachypora, which so. much resembles P. cervicornis (Favosites polymorpha). On the other hand, although it is the fashion to speak of the Australian Carboniferous as Permo- Carboniferous, it may be well to remember that the representatives and nearest relatives of the species occurring in the Fossil Range, are found in the Lower rather than in the Upper Carboniferous of other countries. That the rocks of the Fossil Range are homotax- ialiy Carboniferous there can be no doubt, whatever may be their place in time; and it does not seem necessary to suppose such a development of Permian beds, distinct from the Carboniferous, as is shown in Mr. Gregory's section. * Loc. cit. swpra. 590 W. H. HUDLESTON ON WEST-AUSTRALIAN APPENDIX. Subjoined is a list of the fossils, with remarks on some of the species. Inst of determinable Fossils from the Forrest Collection *. ACTINOZOA. A | B Zoantharia rugosa. 1. Amplexus pustulosus, sp. 1 ............008 Be 2. P MOMUlOSIS AP HHGDS) | secciss. «cent 1 3. . Zap leeaits, S)..4: ter see area ceeeheaene 1 ‘ Tabulata” (Favositide). Aw (Pa chy Ord Pas paler nes Bere cin cose l t5. Stenopora tasmaniensis, Lonsdale ...... oe Ne EcHINODERMATA. 6. Poteriocrinus, sp.| fragments of f{ 7. Cyathocrinus, sp.{ stemsonly | Saale POLYZOA. 8. Evactinopora crucialis, sp. n. ..........-. i a: Gendroidea, Sp. Mic cicciscs canes: oe Pl in : in a very cal- t10. Fenestella plebeia, MeCoy ...........02.. 3 { carecteeeen tll. Protoretipora ampla, Lonsdale ......... ee ee Mouuusca. Brachiopoda. T12. Athyris Royssti,-Levedlé ..............066 3 fl. Spirier striatus, Martin ..0:..s005..00cees ZA 14, ,» ef: crassus, KOninGh:..cx.osssces.000 I 1 115. ——- vespertilio, G. Sowerby ............ 1 TAG: , ef. convolutus, Phillips ............ 1 +17. Productus, cf. brachytherus, G. Sowerby| 1 LS) === or Strophalosiags pe. cx. lsc ..es8secu 1 : 1 k- LO MOrthisscmecies.(ast)iu. ote... koenceeeiee cece lees Ve aon ccs Lamellibranchiata (Monomyaria) +20. Aviculopecten illawarensis, Morris ...... i 21. Timavetornnis, MO7zTIS: yo. .co- +000, «olga i 28 | 17 |1 N.B. In the accompanying remarks on the apparently new forms, it should be distinctly understood that I regard the specific names chosen as merely provisional. It became necessary to do something * Column “ A” denotes the number of specimens of each species marked “ 1,” and believed to be derived from “ Fossil Range” north of the Lyons river- Column “ B” denotes the number of specimens of each species now marxed, but stated to come from “Fossil Range” near the junction of the Gascoyne and Lyons rivers, lat. 25° S. iP pete thus marked are quoted in Mr. Etheridge’s Catalogue of Australian fossils. FOSSILS AND ROCK-SPECIMENS. - 591 with these new fossils, and accordingly I have done the best I could under the circumstances. Paleozoic Corals and Polyzoa can only be adequately described by those who have made the subject their especial study. 1, AMPLEXUS PUSTULOsUS, sp. n. Plate XXIII. figs. la, 16, le. There are five specimens of a somewhat rugged Ampleaus with broad septa, which cannot be referred to either of the two species of this genus hitherto recognized as Australian. Of these species Amplexus arundinaceus was first described by Lonsdale from a specimen in black limestone ; but this is im such an imperfect state as to make a very bad type * for comparison. However, the septa of A. arundina- ceus are finer and more numerous, and it is pretty evident that there is no reason for De Koninck’s suggestion f that it approaches Za- phrentis cylindrica, Scouler. Certainly the transverse section of the British-Museum specimen, bad as it 1s, would not lead one to sup- pose that the septa were continuous to the centre. Amplexus pustulosus has more affinity with A. Selwyn, De Ko- ninck ft, both in external form and in the size and number of the septa. In De Koninck’s figure no epitheca is shown; but in the text the author speaks of it as being very fine; and no mention is made of any excrescences, such as form one of the distinguishing features of the species now under consideration. Moreover, the septa are probably rather more numerous in De Koninck’s fossil. Similar processes have been noted, though rarely, in other species of Amplexus. For instance De Koninck describes A. iacrymosus § from the Carboniferous Limestone of the neighbourhood of Dinant, which has pustules like tear-drops. Curiously enough, he remarks that it has analogy with A. arundinaceus, “from which it differs principally by its spiniform appendages.” The corallum of Amplexus pustulosus is moderately large and slightly curved. The epitheca is of varying thickness in the different specimens, according to the state of preservation, and is furnished at rare intervals with processes which probably supported spines, but which in their present condition, greatly resemble pustules||. In a specimen about one inch in diameter, the number of septa is 42 ; they are slightly less in width than the intercostal spaces, and ad- vance well towards the axis. All the three specimens figured have been more or less squeezed out of shape. 2. AmpLexus, cf. NopuLosus. Phillips, Pal. Foss. of Cornwall, p. 8; De Koninck, Nouv. Rech. Terr: Carb. Belg. p. 74, pl. vi. fig. 5. * The type is in the British Museum, and no better specimen has been seen by me. 4 Foss. Pal. Nouv. Galles du Sud, 1877, pt. 3, p. 149. { Op. cit. p. 73, Atlas, t. 2. fig. 2. § Nouv. Rech. Terr. Carb. Belg. p. 76, pl. vi. fig. 7. | See figure 1 a. Lest any one should be disposed to find a mare’s-nest, it may be as well to draw attention to the resemblance which certain crinoidal fragments adhering to the epitheca and the matrix present to these pustules. Both are shown upon the figure. 592 W. H. HUDLESTON ON WEST-AUSTRALIAN Genus Srewopora, Lonsdale, 1844. In a very able article which has recently appeared in the ‘Annals of Natural History’ *, on Paleozoic Corals from Northern Queensland, Messrs Nicholson and Etheridge contend that Lonsdale’s genus is a good one and worthy of being retained. They admit that, in trans- verse section, across a branch, for instance, the axial corallites are seen to differ in no essential feature of their structure from those of Monticulipora or Favosites. The tubes in this portion of the corallum are regularly polygonal, and are certainly, as a rule, in close contact. But in what they call tangential sections, taken a little below the surface, the characteristic feature of Stenopora 1s made manifest. Here it is possible to observe the periodical thick- ening which produces the annulations of the tubes in their outer portions 7. In Count Strzelecki’s work Mr. Lonsdale supplements his dia- gnosis, and describes four species which are figured in plate viii. Two of these are branching forms, viz., Stenopora tasmaniensis, and Stenopora ovata§. In the latter species those portions of the tubes tangential to the axis are very closely annulated, as is shown in the enlargement (fig. 3a of pl. viii. in Strzelecki’s work). The original specimen is in the British Museum, and testifies to the accurate drawing of Mr. J. de C. Sowerby. It is just weathered enough to display the internal structure, and in this way the close annulations of the “ horizontal” portions are admirably displayed. 3. STENOPORA TASMANIENSIS, Lonsdale, 1844; Darwin’s Geol. Obs. Vole. Islands, p. 161 (1844); Strzelecki’s Phys. Dese. N.S. Wales, p. 268, t. 8. fig. 2 (1845). There are four fragments in Mr. Forrest’s collection belonging to a branching coral; the branches are subcylindrical, and were pro- bably variously inclined or contorted ; tubes more or less divergent ; mouths slightly oval, indications of successive narrowing in each tube uncertain. This agrees only moderately well with Lonsdale’s description of S. tasmaniensis as given by Strzelecki, though on the whole these specimens have a considerable degree of resemblance to the figure in the plate. Since there is no specimen of S. tasmaniensis in the © British Museum, actual comparison has not been practicable. I have failed to detect any accumulation or successive thickening either in the natural sections, or in those that have been cut, though a more practised eye might be able to do so. It should be observed that Lonsdale himself merely says that * Ser. 5, vol. iv. pp. 265 e¢ seq. + Lonsdale’s diagnosis of Stenopora concludes as follows :—“ Corallites poly- gonal, thin-walled, and more or less completely in contact in the centre of the branches; but in the outer curved portion of their course, more or less cylin- drical, and annulated by periodical ring-shaped thickenings” &c. ¢{ Physical Description of New South Wales, pp. 262 ef seq. § De Koninck made these species synonyms of Chetetes twmidus, and sub- sequently, having shown the existence of mural pores or perforations in 8. ovata, he has referred it to the genus Favosites. Nich. and Hth. loc, cit. FOSSILS AND ROCK-SPECIMENS. 593 “several casts of a racemose Stenopora were noticed in the collection [Count Strzelecki’s] examined, but that they did not admit of com- plete identification.” It is highly probable that Mr. Forrest’s speci- mens belong to the species to which allusion is thus made. Genus Evactrrnopora, Meek and Worthen, 1865 *. This genus was instituted for the reception of certain peculiar forms, which are, perhaps, more nearly polyzoan than actinozvan, but which are not very easy to understand. The following is the generic diagnosis :— “‘ Polyzoum free ? consisting of afew large, more or less thickened and solid calcareous plates or lamine, radiating from an imaginary vertical axis, so as to present, in transverse section, a star-shaped or cruciform outline. Rays thickest and most dense on the under and outer edges; thinner and penetrated on either side by the pores within ; each apparently divided along the middle by a thin lamina separating the inner ends of the pores of the opposite sides ; substance showing in transverse sections a more or less laminated structure, the lamine being arranged parallel to the planes of the rays. Pores small, regularly arranged in quincunx, and separated by spaces equalling or exceeding their breadth.” Three species are described and figured from the Lower Carbo- niferous of Illinois and Iowa. The authors allude to the possibility of Hvactinopora being the same as Conodictywm of Munster’. 4, EVACTINOPORA CRUCIALIS, sp. 0. Plate XXIII. figs. 2a, 26, 2c. Somewhat similar to Meek and Worthen’s Evactinopora grandis (op. cit. p. 503, pl. 15. fig. 2, 2a, 26), this one differs principally in the arrangement of the pores on the flanks of the rays. As may be seen in the enlargement of one of the rays (fig. 2c), the pores or tubulated terminations leave quite a blank space or callus. This peculiarity completely disturbs the symmetry of the quincuncial arrangement, whilst the pores themselves, as a sort of compensation for being thrust out, are larger in immediate contiguity with these blank spaces. The rays are four in number, as in “vactinopora grandis, and ‘arranged in the form of a cross; the entire polyzoarium is very much smaller, but this may be merely a question of age. The laminated structure, mentioned in the diagnosis of the genus as being parallel to the rays, is remarkably well shown by the artist both in the cross section presented by the presumed base of the compound organism (fig. 2 a), and still better in the fracture-face of the fourth ray, en- larged (fig. 2c). It is very evident that weathering developes and perhaps exaggerates this structure. The specimen figured is the only one in the Forrest collection that can, with any certainty, be referred to this species. * Proc. Acad. Nat. Sci. Philad. 1865, p. 165. Meek & Worthen, Geol. Sur- vey, Illinois, vol. iii. p. 501. + Goldf. i. 108, pl. xxxvii. fig. 1; cf also D’Archiac, Mém. Soc. Géol. France, vol. v. p. 369, pl. 25. fig. 1 (Conipera eladiformis). 594 W. H. HUDLESTON ON WEST-AUSTRALIAN 5. EVAcTINOPORA DENDROIDEA, sp.n. Plate XXIII. figs. 3a, 30, 36, dd. . In this species the development is dendroid rather than cruciform, the rays or branches being cylindrical to ovate, and even flattened ; whilst the arrangement is possibly triradiate, if we are to regard the specimen fig. 3a as an original centre or nucleus. The other specimen figured (figs. 36, 3c, 3d) is believed to be a branch from a similar nucleus, and is in avery good state of preservation. All the external peculiarities with respect to the pores and the blank spaces are common to this and the previous species, and a very large series might possibly show connecting links. Still this form would seem to be triradiate rather than quadriradiate, the rays usually, though not always, having a tendency to be cylindrical. A weathered fracture in the second specimen (fig. 36) gives a good axial section, which has been enlarged (fig. 3 ¢), but not enough, perhaps, to show the very complex tubular structure, though the lamination mentioned with reference to the previous species is very conspicuous. Indeed the lamination is so brusque that it very much interferes with a correct understanding of the relations between the very fine axial tubes and the larger lateral ones. A study of this specimen seems to indicate that the laminated portion of the general structure is confined to the region of the large lateral tubes, as the central portion is free from lamination and in a somewhat different mineral condition. Owing to unequal development, this ray is very unsymmetrical in this portion of its course, and thus where there is a fresh manifestation of the fine vertical tubing, the absence of lamination again coincides. Whether this has any structural sig- nificance it is impossible to say. The transverse section (fig. 3d) of this same branch, or ray, shows us the tubular systems from a different point of view; whilst the laminar structure is much less apparent. Here we obtain a good view of the division along the middle by the “thin lamina separating the inner ends of the pores of the opposite sides,” the central axis, as it were, of the ray or branch. The relation of the axial or vertical tubes to the lateral ones is never to be made out in these transverse sections; but we learn that the larger tubes are sparingly tabulate*, and we also perceive unequally developed zones of the small tubes at intervals in the more external portions, which are, I suppose, connected in some way with the growth of the stock. There are two other specimens in the collection which may belong here?. * Dr. G. J. Hinde has drawn my attention to the circumstance that in Nicholson’s work on the Monticuliporide forms are depicted which, as regards their internal structure, are not unlike Hvactincpora. At page 88, Heterodictya, an undoubted polyzoan, is shown to have well-developed tabulz. + In the Mém. Soe. Géol. France, 2 sér. t. 2, pt. 1, there is a paper by M. D’ Archiac on fossils from the Nummuline beds of the neighbourhood of Bayonne, in which he describes, under the name of Gwettardia Thiolati (p. 197, pl. v. fig. 15, and pl. viii. figs. 5, 6, 7), an organism somewhat resemblirg this. He re- fers it to Michelin’s genus Gwettardia, which is generally regarded as a sponge. IL. Quart. Journ.Geol. Soc Vol. XXXIX.P1. XX] S. imp. Munxtern Bro A.S.Foord lith. WEST AUSTRALIAN FOSSILS. FOSSILS AND ROCK-SPECIMENS. 595 6. FENESTELLA PLEBEIA, M‘Coy, 1844. Fenestella plebeca, M‘Coy, Synop. Carb. Foss. Ireland, p. 203, t. 29. fig. 3 (1844). Fenestella fossula, Lonsdale, Darwin’s Geol. Obs. Vole. Isl. p-. 166 (1844). fenestella fossula, Lonsd. in Strzelecki, p. 269, t. 9. fig. 1 (1845). Besides three well-preserved fragments of this world-wide species, there are numerous indications of it in connexion with other fossils. Genus Prororetipora, De Koninck *, 1877. 7. Prororerrpora (FENESTELLA) ampLA, Lonsdale, 1844. Fenestella ampla, Lonsdale in Darwin’s Vole. Isl. p. 163 (1844). Fenestella ampla, Lonsd. in Strzelecki, p. 268, t. 9. figs. 3a-d (1845). There are two specimens in the Forrest collection. Mr. Etheridge, junior, remarks that the group is subject to great variation, being very abundant in what he calls the Permo-Carbo- niferous of Australia, and fining down through infinite gradations to a smaller mesh. EXPLANATION OF PLATE XXIII. Fig. la. Compressed specimen of Amplexus pustulosus, n. sp., showing the tubercles on the epitheca. Nat. size. 1. A smaller specimen, also compressed, having the epitheca partly removed. Nat. size. lc. Transverse section of a third fragment, also pressed out of shape. 2a. Under side of Hvactinopora crucialis, sp.n. Nat. size. 2 6. Upper side of same specimen. Nat. size. 2c. Do., flank view, magnified 23 times. 3a. Triradiate nucleus of Hvactinopora dendroidea, sp.n. Nat. size. 36. Branch of another specimen of H. dendroidea. Nat. size. 3c. Portion of ditto, showing an axial section, xX 2 diam. 3d. Transverse section of ditto (slice on glass), x 3 diam. Discussion. Prof. Bory Dawxtns pointed out the similarity of the sequence in Western Australia with that long ago established as occurring in Eastern Australia. Mr. J. C. Crawrorp pointed out the fact that, though Carbo- niferous rocks occur in West Australia, there is no coal yet known there; and referred to the difficulty of deciding as to the age of the coal of Australia. The Avrnor, in reply, stated that the coals of Australia appeared to be of two different ages, Paleozoic and Mesozoic. There is no proof of the presence of Devonian strata in Western Australia. * Foss. Pal. Nouv.-Galles du Sud. pt. 3, p, 178. 596 A. J. JUKES-BROWNE ON THE RELATIVE AGES OF 34. On the Retative AcxEs of Curtain River-vattnys in Lincoin- soire. By A. J. Juxes-Browne, Esq., B.A., F.G.8. Com- municated by permission of the Director-General of the Geolo- gical Survey. (Read June 20, 1883.) INTRODUCTION. In a country which is traversed by a series of escarpments or hill-ranges, the valleys by which its drainage is effected are usually separable into two sets or systems, one parallel to the strike of the ridges, and the other more or less at right angles to the same. The origin of these longitudinal and transverse valleys, and the process by which escarpments have been intersected by river-valleys, were first explained by Mr. Jukes*. He showed also that in the case of a river cutting through a ridge or escarpment, and receiving tributaries from the longitudinal valleys which are parallel to this ridge, the primary or first-formed stream is that opposite to the breach in the escarpment, and that the longitudinal branches, though often of much greater length than this primary stream, are really of secondary or subsequent origin. Stated in general terms, his theory amounts to this, that the original direction of all rivers which cut through ridges was de- termined by the general slope of the ancient surface over which they began to run, and that when the slope was transverse to the strike of the beds, the channels cut by the earliest rivers necessarily had a similar transverse direction, while the channels in the longi- tudinal valleys were formed subsequently and concurrently with the development of the ridges or escarpments. A further corollary to this may, I think, be considered as generally true, viz., that those portions of a river-valley which intersect the same ridge date from the same epoch of time; and that rivers flowing between the same two parallel ridges came into existence at the same time. It does not follow, however, that the channels of all the primary transverse streams were permanently maintained ; the extension of a longitudinal tributary may intercept the drainage of minor transverse streams. Thus in fig. 1, A, B, ©, are three transverse streams, each of which originally ran across the ridge DR as in- dicated by the dotted continuations of B and C; but the extension of the tributary T has intercepted the waters of B and C. Jukes, writing of the river-valleys in the south of Ireland, ex- presses this as followst:—“ The longitudinal valley may even be worn back across the course of many minor transverse streams, and deflect their waters down its course.” He also observes that ‘ to * Quart. Journ. Geol. Soe. vol. xviii. p. 878 (1862). t Manual of Geology, 3rd ed. p. 456 (1872). CERTAIN RIVER-VALLEYS IN LINCOLNSHIRE 597 whatever extent these longitudinal valleys might proceed, none of the waters coming down them could ever cross the lowest original transverse valley that was formed by the little primary river ; ” and he points out that the river Blackwater has never crossed the primary transverse channel which runs southward from Cappoquin in Waterford, and never could do so unless something happened to cut a channel lower down the remainder of the longitudinal valley to Dungarvan Bay, and deeper than the channel already cut down to Youghal Bay. Fig. 1.—Diagram of a Stream intercepting earlier Transverse Streams. X Y NIG NE rs ~~. ypvonaygni“©) Zany c B A Ss 5 T { uy, + lilly ‘suite A pn @iSd as Z2\35 p ZS = => To suppose the possibility of such a channel being formed, in any case was entirely outside the scope of Jukes’s argument ; butthe sequel will show that it is worth while considering this possibility, and such a channel. calculating the results which would follow from the developinent of Fig. 2.—Diagram of probable changes in the Valley of the Blackwater. Vy IG Hygn ZT Suppose, therefore, that a small stream ran into Dungarvan Bay as at S in fig. 2, and that while the valley of the Blackwater was ex- é 5 tending itself westward from Cappoquin, the valley of the stream S was also extending itself westward from the coast It is clear that 598 A. J. JUKES-BROWNE ON THE RELATIVE AGES OF the base-level to which the stream would cut down at its mouth, 8, would be much lower than the bottom of the Blackwater valley at C; and just as the Blackwater may have intercepted and diverted certain transverse streams to the westward, so might the stream § have intercepted the waters of the main river at C. In this process the detritive action of rain in lowering the retreating watershed (W) would be as much concerned as the action of the stream in eroding its channel and transporting detritus; and the whole process would form part of the gradual development of the longitudinal valley in the manner explained by Jukes. Eventually the watershed W might be so lowered that the difference of level between the Blackwater at C and the upper tributaries of the stream § might be very slight ; and during a time of flood the waters of the main river might force a way into the channels of these tributaries. When once this diversion was effected, the river would not be likely to regain its former course through the narrow gorge at V, which would be ultimately converted into a dry pass or gap through the ridge ; and the channel of the Blackwater would be confined to the longitudinal valley north of the Drum ridge, and would appear as a river running from west to east and emptying itself into the sea at Dungarvan Bay. The final result, therefore, of the whole process might have been this, that the course of all the transverse streams which originally crossed the Drum ridge from north to south, might have been di- verted into the longitudinal valley, and so have been converted into one continuous river flowing from beginning to end in this longi- tudinal valley from west to east. At the same time the volume of water in the original valley of the Blackwater south of the Drum ridge would be very greatly diminished, owing to the great con- traction of its drainage-area by the diversion of its upper tributaries. I have no doubt that a very good reason can be found why this last diversion did not happen in the case of the Blackwater; but what local conditions prevented from happening in that case may have happened in other cases, and I have only used the instance of the Blackwater in order to make my meaning clear. I will now, therefore, state in general terms the special thesis which I hope to establish in the following pages. Wherever a succession of escarpments and intervening longitudinal valleys has been developed out of a surface of marine denudation, and a river crosses any one of the longitudinal valleys which happens to stretch to the sea-coast, then this transverse river is liable to interception and diversion by the backward extension of a stream flowing down the longitudinal valley into the sea or into a tidal estuary. In considering the possibility of such an occurrence it must be remembered that most escarpments have an inclination in the direction of their strike, and that their base-line often has a decided slope from a given point inland towards their termination at the sea ; in such cases the intervening valleys have a general slope in the same direction. It is only under such conditions that the inter- cepting stream could come into existence, and that it could extend itself up the longitudinal valley far encugh to intercept the inland CERTAIN RIVER-VALLEYS IN LINCOLNSHIRE. 599 and previously established course of a transverse stream. I now proceed to describe two cases in which I believe such an inter- ception has taken place. Tur Catcesy AND StTeEEPING Brcxs In LINCOLNSHIRE. Relative Position of their Valicys.—The tract of country which is drained by these two streams lies at the southern extremity of the Lincolnshire Wolds, which here consist of two nearly parallel ranges of hills, the eastern range being that of the Chalk Wold, the western consisting of Neocomian sandstones and clays capped by Boulder-clay. Between these two ranges lies the broad valley of the Steeping, which has been cut down through the Neocomian strata into the underlying Kimmeridge Clay ; while the chalk range is completely severed into two portions or masses by a deep and narrow valley which gives passage to the Calceby Beck. The latter is therefore a transverse valley, and the former a longitudinal one ; but their relation to one another is not that which ordinarily exists between transverse and longitudinal valleys. The stream in the longitudinal valley is not a tributary of the other, but flows away from it, and the upper part of its valley lies at a lower level than the adjoining part of the transverse valley. The transverse valley is thus abruptly truncated and cut off from receiving the brooks rising on the high ground to the westward, which would naturally drain into it, but for the interposition of the longitudinal valley of the Steeping. It is difficult, therefore, to understand how the transverse valley could have originated, if the former configuration of the district was at all similar to that which it now presents. Both streams now rise in the neighbourhood of Tetford, a village about eight miles south of Louth. The Calceby Beck has its sources among the hills east of that village, the numerous springs which issue from the base of the Chalk forming several small brooks which unite below the small hamlet of Calceby, and thence flow north-east- ward to South Thoresby. Near Calceby it also receives the water of a tributary from the N.W. flowing in a longitudinal valley, which has a much greater length than any of those which unite at the head of the main valley. Another tributary comes in near South Thoresby, and springs swell its volume near Belleau, where the stream emerges into the broad undulating plain of Boulder-clay which intervenes between the chalk hills and the marshes along the coast. Through this plain the river pursues its north-easterly course in a shallow valley past Claythorpe and Withern, where it passes into the marshland, and is carried northward between raised banks to the sea-coast at Saltfleet. If we now turn to trace the course of the Steeping river, we find that its head waters are formed by a brook which rises near Belch- ford, and flows westerly through Tetford. The valley of this brook is continuous with the Ormsby valley, through which flows a tributary of the Calceby Beck; but the Tetford brook instead of Q.J.G.8. No. 156. 20 —— eS =——— a ———— ———— eee SS ———— eS em Sha aS ee =-<-- = ——s = — a Pa Neer WW Chalky Boulder-clay. WW St = Great Steeping. er | \ Pariney: CERTAIN RIVER-VALLEYS IN LINCOLNSHIRE. 601 pursuing what appears to be its natural course, turns southward at right angles, and passing through a narrow gorge, issues into the broad valley of the Steeping at Somersby. Three other tributaries descend from the high ridge above described on the western side of this valley; and another from the eastern side, which, like the Tetford brook, looks as if it had once flowed N.E. into the Calceby valley, but now turns southward through Harrington Carr into that of the Steeping. The united waters of these brooks flow south-eastwards by Saus- thorpe, Partney, and Ashby, receiving two more tributaries from the north, viz. the Langton and Skendleby Becks. At Ashby near Spilsby the stream turns southward, and passing between the villages of Great and Little Steeping, it enters the broad plain of the Fenland, and is conducted by a series of dykes to the outfall near Wainfleet. It is worthy of remark here that the entrance to the Steeping valley between Spilsby and Partney is very narrow compared with its breadth higher up and nearer its source. The peculiar relations of the Calceby and Steeping valleys at once suggest that the first has been excavated by streams flowing east- ward from the Neocomian hills, before the upper part of the Steeping valley had its present extension ; and that these streams, which were originally the head waters of the Calceby Beck, were, by the subse- quent formation of the Steeping valley, intercepted and diverted into that valley. Disposition of the Glacial Deposits.—This view of the relative ages of the two valleys receives strong confirmation from the dis- position of the glacial beds, and the extent to which they enter the two valleys. The eastern flank of the Chalk Weldsis bordered by a great mass of Boulder-clay, including beds of sand and gravel, and belonging to the series known as the Purple and Hessle Clays. These deposits are in many places from 60 to 80 feet thick, and are banked up against the chalk hills, sometimes sweeping over and resting on their tops. These clays and gravels enter many of the valleys which open eastward, but are totally absent from others which drain in the same direction, the natural inference being that the Boulder-clays are posterior to the excavation of some of the valleys and anterior to the erosion of the others. Now the valley of the Calceby Beck is largely occupied by these glacial accumulations; at its mouth between Bellean and South Thoresby the ancient bed of the valley is probably 50 or 60 feet below the present surface; and the glacial beds can be traced con- tinuously to the very head waters of the present stream, and far up the valleys of the two tributaries which come in from the N. W. A great mass of clay and gravel blocks up the space Where the ancient outlets of these two tributaries into the main valley appear to have been, so that the streams have been obliged to excavate deep and narrow channels through the solid Chalk on either side of this massive obstruction. Up the Ormsby valley Boulder-clay can be traced along one side of the park, and then gives place to a wide 20 2 602 A. J. JUKES-BROWNE ON THE RELATIVE AGES OF spread of gravel, sand and Joam, which forms an open plain between Ormsby Park and Tetford, the glacial age of the whole being proved by an intercalated patch of Boulder-clay near the western end of the deposit and beyond the present sources of the Calceby Beck. It appears therefore that the Calceby and Ormsby valleys are choked up with glacial deposits, and that the present beck has not been able to excavate its channel down to the bed of the ancient river by which the valley was originally formed. If we now turn to the Steeping valley we find that the Boulder- clays sweep round the southern end of the Chalk Wolds into the bay-like entrance of this valley, and extend in a narrow tongue far up the tributary valley of the Skendleby Beck, which comes in from the northward. Beyond the village of Partney, however, about a mile above the point where the Skendleby Beck enters the main valley, not a trace of glacial clay or gravel is to be found; the Boulder-clay terminates abruptly at this point, and not the smallest remnant exists further up the main valley or in any of its other tributaries to indicate that it ever had any further extension. This limit is only about two miles distant from a line drawn across the mouth of the valley from Halton to Candlesby. The ancient outlet of the Skendleby Beck is buried beneath the glacial deposits *, and is doubtless at a much lower level than the present bed of the Steeping river. This stream has excavated a channel through the western edge of the Boulder-clay, and down to the underlying Kimmeridge Clay, being evidently turned aside by the great mass of the Boulder-clay which fills up the bay-like entrance to the valley. It is worthy of remark in this connexion that the gravels here intercalated between the upper and lower sheets of Boulder-clay have yielded an abundance of mammalian remains. From the termination of the Boulder-clay near Partney to the furthest point at which the Kimmeridge Clay is exposed, viz. near Salmonby, is a distance of seven miles; it would appear therefore that the Lower Neocomian sandstone has been stripped off the Kimmeridge Clay throughout the whole of this distance during the time which has elapsed since the formation of these Boulder-clays ; that is to say, the greater part of the Steeping valley is entirely of Postglacial origin. On the other hand the whole of the Calceby valley was formed before the accumulation of the Purple and Hessle Clays f. ia Mode in which the Steeping Valley was formed.—l have already mentioned that the valley of the Tetford Beck is really continuous with that of the Ormsby Beck, the intermediate portion being a * See Quart. Journ. Geol. Soe. vol. xxxv. p. 403. t Mr. 8. V. Wood makes a great distinction between these clays, and even excludes the so-called Hessle Clay from the Glacial series altogether. I have not been able to detect any sign of unconformity between them, but on the con- trary believe that they are parts of one continuous series, though I am still in doubt as to the exact relations of this red and purple series to the Chalky Boulder-clay. CERTAIN RIVER-VALLEYS IN LINCOLNSHIRE. 603 broad plain between two of the steep hill-ridges into which the chalk escarpment is here cut up. This plain now forms the water- shed between the Calceby and Steeping valleys ; its centre is occupied by a broad strip of peat and alluvium, the western end of which drains into the Tetford Beck and the eastern end into the Ormsby Beck. This plain only exists because the ancient valley is choked up with glacial sand and gravel; and it is perfectly clear that if these accumulations were removed, the Tetford brook would continue its easterly course into the Ormsby brook, and so into the Calceby valley. Indeed, the strip of alluvium indicates that this was the course of the brook up to a very recent period (geologically speaking), and that it was only deserted when an easier exit was found by the present channel. The cause of this diversion now remains to be considered, but is not far to seek. he broad ridge of Neocomian sandstone which lies to the south of Tetford is traversed by a deep and narrow trench, and through this the brook now escapes from the upper plain, and descends to the lower level of the Steeping valley. It could not have done this until the spring-heads of the Steeping had receded to their present position. It is must probable that this trench was originally formed by a small tributary of the Tetford brook, draining the district to the southward and running northward to join it below Tetford. The drainage-basin of this tributary was gradually invaded aud sapped by the recession of the spring-heads on its southern border, until it ceased to convey any water into the Tetford valley and its northern portion would remain as a dry trench. As the combined action of rain and springs carried the head of the Steeping valley further and further back, they worked down to a lower base-line than thst of the Tetford valley, and the Kimmeridge Clay was gradually bared along the course of this dry trench. The strong springs which now issue from the base of the sandstone in this trench near Somersby, show how the work was done. Eventually when the country had assumed its present configuration, and probably when the Tettord brook happened to be in flood, the waters overflowed from the Tet- ‘ford valley along this trench into that of the Steeping; and when this communication was once established it would be maintained, because, being cut down to a lower base-line, the fall along the new channel is much greater than that along the old one. If this was the manner in which the Tetford brook became part of the Steeping-river system, it is very likely that other streams may originally have drained into the Calceby valley, and have been diverted in the same manner. The breadth of the valley in which Brinkhill stands, its occupa- tion by sand and gravel which is continuous with the drift of the Calceby valley, and its abrupt termination, are all facts suggestive of its once having had a longer extension to the south-west. In all probability the beck which rises near Warden Hill, and now flows southward through Harrington Carr, was originally a tributary of a stream flowing N. E. down the Brinkhill valley into the Calceby 604 A. J. JUKES-BROWNE ON THE RELATIVE AGES OF Beck. Its diversion into the Steeping valley may have been effected in exactly the same way as that of the Tetford brook. Before the diversion of the Tetford brook the extreme sources of the Steeping river would be at Salmonby; and before the diversion of the last-mentioned brook, the sources of the Steeping were pro- bably near Stainsby ; and atsome still earlier period, all the streams draining the country north of the ridge on which Hagworthingham stands, ran north-eastward into the Calceby valley. The mapping of the district by myself and Mr. Strahan, leads us to conclude that before the oldest Boulder-clay was laid down, the chalk escarpment occupied a more westerly position, and the line of the Steeping valley was occupied by the broad outcrop of the Middle Neocomian clay, with valleys opening south-westwards and drained by brooks running in that direction. On this surface the so-called Chalky Boulder-clay was deposited, and was probably banked up to the then edge of the Wold escarpment, as is the case further north. When detritive agencies began to operate on this district the rain which ran southwards would find a line of weakness along the junction of the Boulder-clay with the Wold scarp. A valley opening southward, or south-eastward, would in process of time be excavated along this line, just as the upper valley of the Bain has been formed under similar conditions. Its bottom for a considerable distance would be formed by the Middle Neocomian clay, and it would be some time before the Lower Neocomian sand- stone would be bared to any great extent above the position of Partney. It is not unlikely that the beck coming down from Lang- ton and Sutterby, and joining the Steeping at Partney, may indicate the course of this line of drainage. The Skendleby Beck was clearly also in existence at this time, and the united streams would doubt- less flow south-east to the coast-line wherever that was. As, however, the work of pluvial detrition went on, and a larger and larger area of the Lower Sandstone became exposed, a great proportion of the rainfall woulda be absorbed by the sandstone, and thrown out from its base in the form of springs. This would necessarily introduce a new element among the agencies of erosion, and the combined action of rain and springs would cause the valley’ to be extended in a north-westerly direction along the strike of the sandstone, far more rapidly than in any other direction. Under these circumstances there is nothing improbable in the hypothesis that the whole of the present valley from Partney to Salmonby has been excavated in Postglacial times, and that, being eut back along a lower base-line, the Steeping river has gradually intercepted the drainage of a district which was once a part of the Calceby-beck system. This therefore appears to be a case in which the extension of.a longitudinal valley has intercepted and diverted the course of certain streams which originally flowed imto a transverse valley, and has entirely altered the drainage-system of a considerable district. The conclusions to which I have been ied by a prolonged study oi this part of Lincolnshire are opposed to those of Mr. 8. V. Wood CERTAIN RIVER-VALLEYS IN LINCOLNSHIRE. 605 on almost every point which he has raised in his recent memoir*. Relinguishing his older opinions regarding the formation of Boulder- clays, he has adopted the theory of their terrestrial origin, and finds evidence of a theoretical ice-cap where other eyes can only see the results of ordinary rain- and river-erosion. He even goes so far as to doubt the capacity of certain rivers to make their own valleys, and does not believe that the rivers of Lincolnshire or Kast Anglia “‘ had anything to do with the excavation” of the valleys in which they flow. He prefers to attribute the excavation of these valleys to the rush of waters produced by the melting of a mass of ice on the top of the chalk escarpment in Lincoln and Norfolk. To any one familiar with the work of rain and rivers such an idea must seem highly improbabie, and to any one who can visit the valley of the Steeping, and will note the wonderful series of dales, combes, and hollows which have been fretted out of the edge of the Lower Neocomian sandstone, and which are so evidently due to the action of rain and springs, Mr. Wood’s hypothesis must seem extremely unlikely and utterly unnecessary. A mere inspection of the Geological Survey map, in fact, will afford sufficient grounds for deciding between the two explanations. Mr. 8. V. Wood confuses the two distinct valleys of the Bain and Steeping with the general trough-line in which they lie, and which he calls the Bain-Steeping trough. Now there is nothing remarkable about this trough ; it is merely the re-excavation of the Preglacial scarp-foot, and has been widened into a trough by the recession of the chalk escarpment on the one side and the edge of the Boulder-clay on the other side, this widening being due to the agency of rain and springs, and having been in progress ever since the formation of the Boulder-clay. In that portion of the trough which lies between the valleys of the Bain and Steeping there is a floor of Middle Neocomian clay stretching continuously from side to side; but the Steeping valley is a hollow within the trough, and is cut down through the Lower Neocomian sandstone, and into the Kimmeridge Clay. Mr. Searles Wood ‘has noticed the limitation of the so-called Hessle Clay to the entrance of the Steeping valley, but does not offer any explanation of this peculiar portion, merely remarking that if the clay were an aqueous accumulation it should stretch up the valley to the levels which it reaches elsewhere+. I am still of opinion that the Hessle Clay is a marine accumulation, and offer the simple explanation that at the time when the clay was formed, the upper part of the valley in question had no existence. Accor- ding to my view the formation of the Hessle Clay took place at au early epoch in the history of the valley, and when its excavation had not proceeded farther than the first stage described on p. 604, its entrance as far as Partney being an open bay, into which the Langton and Skendleby Becks emptied themselves. If Mr. Wood’s views regarding the age and origin of the Steeping * Quart. Journ. Geol. Soc. vol. xxxiv. p. 457, and vol. xxxviii. p. 667 (1882). Tt Quart. Journ. Geol. Soc. vol. xxxviii. p. 715. 606 A. J. JUKES-BROWNE ON THE RELATIVE AGES OF valley be admitted, the absence of the Hessle beds is apparently an inexplicable fact; for, whether these beds have been accumulated under aqueous or terrestrial conditions, it 1s equally surprising that no trace of them should occur in this valley (if it was then in existence) when they extend so far up the adjoining Calceby valley. What therefore is such a puzzle on Mr. 8. V. Wood’s hypothesis, is but a necessay consequence of my postulate that the longitudinal valley is of later date than the transverse valley. I may therefore reasonably regard the position of the Hessle Clay in the Steeping valley as a confirmation of my belief that the whole of that valley above Partney has been excavated in Postglacial times, 7. ¢. since the formation of the Hessle Clay, which I regard as the uppermost member of the Glacial series. TRENT AND WiITHAM VALLEYS. Modern Course of the Trent.—This river is formed by the union of several streams, of which the most important are the Trent, the Dove, and the Derwent, flowing from the north-west off the Derbyshire and North Staffordshire watershed, the Tame and the Soar flowing from the south through the counties of Warwick and Leicester. These streams converge towards a point about ten miles 8.8.E. of Derby, their united waters being known as the river Trent, which flows onward in a north-easterly direction through a well-marked valley as far as Newark. Here, however, the river bends to the northward, keeping to the west side of the low Rhetic escarpment, as if it had not been able to cross that comparatively slight obstruction. This northerly course it maintains till it reaches the estuary of the Humber. Now the course of the Trent as far as Newark favours the supposition * that it was determined by the westerly slope of a plane of marine denudation across the edges of the Lower Jurassic strata; but if so, and if it had ever flowed over a surface of Oolitic rocks, why did it not continue this course so as to run in a transverse valley through the Oolitic escarpment and into the Wash instead of into the Humber? It is a significant fact that if the general course of the Trent, south- west of Newark, be prolonged to the N.E., it points to the great gap in the Oolitic escarpment at Lincoln, through which the river Witham now flows. If this transverse gap or gorge be considered only in relation to the small river which runs through it, the manner of its origin is altogether inexplicable ; but if good grounds can be shown for supposing that the river Trent formerly . passed through it, the existence of such a gap is satisfactorily accounted for. Moreover the anomaly in the present course of the Trent is likewise explained, if it can be shown that the northerly bend of that river is a subsequent diversion and not its original course. * Ramsay, Phys. Geogr. and Geol. of Gt. Britain, ed. v. p. 518. _ CERTAIN RIVER-VALLEYS IN LINCOLNSHIRE. 607 Ancient Gravels of the Trent.—The width of the valley within which the Trent flows from Nottingham to Newark is remarkably even and uniform, the gravels and alluvium taken together forming a long straight tract with a breadth of from 14 to2 miles. Throughout this portion of its valley the river has a tendency to impinge upon its right bank, so that the higher and older gravels occur chiefly along the north-western side of the valley, and appear to lie entirely within the bounding slope of the valley on that side. By the Lowdham and Thurgarton stations on the Nottingham and Lincoln railway these gravels form a continuous terrace, which slopes gently down to the modern alluvium of the Trent. Near Newark, however, the relations between the older gravels and the modern valleys begin to change; they are found on the eastern bank and stretch away from the river south of Newark and Beacon Hill, forming a continuous tract of gravel-covered country, which extends in a north-easterly direction between the modern valleys of the Trent and Witham. From Coddington, near Newark, the tract passes by Stapleford and Thurlby Moors to Swinderby, and thence it is continued on either side of the Midland Railway as far as Boultham near Lincoln, and only a mile from the entrance to the transverse valley through the Oolitic escarpment. To Mr. Penning, who began to map these gravels from the Lincoln end in 1878, belongs the credit of suggesting that they represent the ancient course of the Trent ; and he would doubtless have brought the subject before the notice of the Society, had not ill health obliged him to leave England and afterwards to resign his post on the Geological Survey. It is also interesting to note that the Trent was joined by a powerful tributary from the south just before it reached the Oolitic escarpment. The proof of this statement is to be found in a clearly defined tract of gravel which stretches for many miles over the Lias plain below the Oolitic escarpment. This stream appears to have had its source among the hills near Belvoir Castle, where the small river Devon now rises; we may therefore consider this as the course of the ancient Devon, which then ran northward to join the Trent, just as the modern Brant joins the Witham near Aubourn. The Brant, indeed, may be regarded as the attenuated represen- tative of this ancient Devon. This line of gravel is traversed by the modern channel of the Witham between Hougham and Westborough, about six miles N.W. of Grantham, but continues northward along the ground between the present valleys of the Witham and the Brant as far as Aubourn, where it is again cut through by the Witham. Small patches, how- ever, still remain near South Hykeham, on the northern side of the Witham valley, and lead on to the wide spread of sand and gravel by North Hykeham and Boultham, already mentioned as part of the Trent gravels. This long tract of river-gravel is in every respect comparable to those which I have elsewhere described as existing in Cambridge- shire, and which so clearly connect themselves with the ancient 608 A. J. JUKES-BROWNE ON THE RELATIVE AGES OF lines of drainage proceeding from the chalk escarpment. Mr. 8. V. Wood has recently accounted for these in a very different way, and imputes to me the extraordinary postulate that I suppose a river to be capable of converting the centre of its bed into a hill*. My sections are simply illustrative of the fact that the general surface of the country has been so lowered by the action of rain and rivers since the time when the drainage was diverted from these old channels, that the gravels of the older rivers now form ridges with slopes on both sides; but if one of these ridges is traced towards the hill country, it gradually comes to occupy the position of a terrace lying on one slope of a valley side. In the case of these old Devon gravels, the Witham on the one side and the Brant on the other have cut down to lower levels, and left the gravels of the ancient Devon on the top of the intervening ridge. Strong confirmatory evidence, almost amounting to proof, of the hypothesis that the Trent once flowed through the Lincoln gap, is furnished by the composition of the gravels which occur on the western side of the strip of Fenland that forms the modern continua- tion of the valley. These gravels will be described in the Geological Survey Memoir on Sheet 83; but the point of interest in connexion with my present subject is that they are largely made up of rounded pebbles of quartzite, hornstone, and other old rocks, which have evidently been derived from the Triassic pebble-beds of the west. The presence of these pebbles in such large quantities on the eastern side of the Oolitic escarpment, and in gravels which border the Witham valley, seems inexplicable except on the hypothesis of their having been brought by the Trent through the gap at Lincoln. There is, therefore, every reason to believe that in early Post- glacial times the Trent flowed along the course above indicated from Newark to Lincoln, and from Lincoln south-eastward to the Fenland, which was then probably an open bay. It remains then to indicate the causes which seem to have operated in diverting the river from the ancient course to its present channel. Diversion of the Trent.—The Humber flows in a transverse valley, which has been cut down to a lower base-line than the Witham valley at Lincoln ; and the rivers which flow into it have always been able to keep the passage open. Consequently the longitudinal valley formed along the tract of soft Keuper marls has continually ex- tended itself southward, and the river in this valley must also have shifted its channel continually eastward, as the escarpment of the Rheetics receded, and the Humber valley was cut down to lower and lower levels. The river running northward along this valley would be supplied by the brooks flowing from the west and draining the tract of Bunter sandstones, and has ultimately developed into the river Idle, which now rises near Mansfield, and pursues a north-easterly course parallel to that of the Trent. I assume, therefore, that when the Trent ran vid the Lincoln * Quart. Journ, Geol. Soc. vol. xxxviii. p. 673. — CERTAIN RIVER-VALLEYS IN LINCOLNSHIRE. 609 valley to the Wash, the Idle ran vid the present Trent valley north of Gainsborough to the Humber. The watershed between the Trent and the Idle was then continuous, but was of course very low where it crossed the Keuper marls by Tuxford, Marnham, South Clifton, and Harby. The rills draining the northern slopes of this low watershed would naturally be collected into a stream running northwards along the valley subsequently usurped by the Trent, and falling into the Idle somewhere north of Gainsborough. In these early Postglacial times, the elevation of the land above the sea was probably higher by 40 or 50 feet than it is now: the rivers consequently had a greater fall,and erosion went on much more rapidly over the whole surface of the country. The sources of this tributary of the Idle would be carried back further and further southward, and the longitudinal valley would be gradually extended and lowered until merely a low col separated its head-waters from the slopes of the valley in which the Trent was then running north-eastward to Lincoln. At this epoch in the history of the Trent, floods must have been of frequent occurrence in this particular portion of its valley, owing to the narrowness of the outlet at Lincoln; and the large extent of | ground covered by its ancient gravels between Newark and Lincoln testifies to the width of the area over which its flood-waters were able to spread. Under such conditions, therefore, and on some occasion when a large head of water was accumulated in the Trent valley, it is not unlikely that an overflow would take place by way of the low col leading into the aforesaid tributary of the Idle. Every flood would then deepen the passage until the full stream of the river found it easier to take this course than to maintain its former channel through the Lincoln gap. The exact situation of this col is of course difficult to fix, though some evidence may be obtained from the disposition of the gravels, when these are completely mapped. There is also some reason to believe that there has been a slight elevation of the high ground occupied by the outcrop of the Lower Oolites east of the great escarpment in Postglacial times. The valley in which Ancaster lies is a transverse cut through this high ground, and is paved with deposits of sand and gravel, which are continuous with those of the Witham valley north of Grantham ; and it appears probable that the Witham passed through this valley at the time when the Trent flowed through the Lincoln valley. Now, however, a low watershed crosses the Ancaster valley, just west of the town, throwing off the Honington Beck to the west, and the Sleaford Beck to the east; and this watershed occurs in the midst of a broad tract of stratified gravel, which must have been deposited there before the watershed was formed. Further, the section in the railway-cutting west of Ancaster station shows that the Lower Oolites (Lincolnshire Limestone and Lower Estuarine series) are here bent into a low and broad anti- clinal curve, the axis of which seems to strike nearly due north and south. 610 ON THE AGES OF RIVER-VALLEYS IN LINCOLNSHIRE. It would appear, therefore, that there has been an uplift across this ancient line of drainage in Postglacial lines, causing the Witham to turn westward and take its present course; and the in- fluence of the same uplift, if continued further north, may have contributed to divert the Trent from its original eastward course wid the Lincoln valley. Whether this, however, was the case or not, the diversion of the Trent could not have been effected unless a course had been already prepared for it by the action of streams in lowering the longitudinal valley south of Gainsborough. In comparing the two instances [ described of the diversion of an older stream by the extension of a longitudinal river-valley, the chief difference appears to be this:—In the first case, that of the Steeping and Calceby Becks, it was the upper tributaries of the latter which were diverted into the longi- tudinal valley ; while in the second case it was the main stream of the Trent which was so intercepted and diverted. I believe that the principles above enunciated and exemplified will explain the courses of certain other English rivers, and purpose recurring to the subject in a future paper. Discussion. Mr, Evans instanced the case of the Waveney and the Little Ouse once flowing from different sources, but now from one near Lopham Ford, the deposit of glacial clay in this instance also forming an element in the case. [ To face p. 611. Q. J.G.S. vol. xxxix.] 4, 3. 2. ‘LeyeMospiagy ‘qQ1oMgun yy *“LoyVMID pL “Yq4.10M4un A "s[oAarT IepyeMosPLIgT “"‘|reur ~04.401 9190409) ON THE ESTUARIES OF THE SEVERN AND ITS TRIBUTARIES. 611. 35. The Estuaries of the SEVERN and its TRIBUTARIES; an INQUIRY into the Natur and Ortern of their Tipat Sepiment and ALLU- viaL Frats. By Professor W.J.Sorzas, M.A., F.R.S.E., F.G.S., Fellow of St. John’s College, Cambridge. (Read June 6, 1883.) Te tidal channel of the Severn is notorious for its mud. At high tide it is filled with a sea of turbid water, thick and opaque with tawny-coloured sediment; as the tide ebbs a broad expanse of shining mud flats is revealed fringing the coast; but so like is the water to the mud that, seen from a distance, it is often hard to tell where the sea ends and the shore begins. It is the same with its tributaries, the Wye, the Usk, Ely, and Rhymney on the Welsh side, the Avon, Yeo, Parrot, and others on the English coast. The source of this mud has been made a subject of much dispute. That it is chiefly supplied by the rivers themselves to their re- spective estuaries might sound to geologists like an obvious truth ; but such is certainly not the opinion of those who have most closely inquired into the matter. Engineers like Mr. C. Richardson and Mr. Howard have long been of opinion that the sediment of the tidal Avon is furnished to it by the Severn; the like is asserted of the Parrot, and I do not think one stands in any fear of contradiction when stating as a general truth that all the estuaries opening into the Severn derive their mud at least immediately from the main channel. This being so, whence then has the Severn obtained it? The answers given to this inquiry by engineers are various: some attribute it to the sea, meaning, it is to be supposed, the mouth of the Bristol Channel ; some to the mud shoals of the estuary ; some to its bordering cliffs; and others to the fresh water of its tributary rivers. There is, no doubt, truth in all these opinions, and the only mistake lies in regarding them as mutually exclusive, or in assigning to any one source a larger share than its due. With regard to the fluctuating mud-banks in the channel, they have been deposited by the tidal water, and will in time be washed away again, and redeposited, and so on again and again. However obviously a source of mud, they are certainly a long way from being an ultimate source, and nothing is to be gained from their further consideration. With regard to the relative share contributed by the remaining agents, the view which geologists would take on general grounds is no doubt correct ; the rivers which discharge into the Severn estuary, draining, as they do, a catchment basin of 9193 square miles, are the chief sources of supply ; but that much is produced by the waves which wash the shores of the estuary, assisted, as they are, by sub- aerial agents, is also clear, and to this the cliffs of Penarth, Aust, and Portishead bear striking testimony. That the distant sea has contri- buted anything at all is not an idea likely at first sight to find much Q. J.G.S. vol. xxxix.] [Zo face p. 611, 3. 4. 5. 6 7 8. 9 10. if 25 ak as Seebank. Pale eke a rd = ‘ FI —H.W., Aug. 16,1866, 5 EE EE 2 ae : De ae E% & 2 53 ae iI che g= a EE s EE zis) S'5 8 of Be oa Hea l=) z = ={s) Aa AO ine in. = ft. in in. High: SS ft. in. . = ft. in. (Made a SS {rates 6 eround. @ spring gh = water 0 0 40 spring. 0 = 6 Hehe e water . 6 neap. 0 Half 0 * tide. Low- water neap. 3 §Zow- § % Bf 9 water 0 0 JSG CG) Uepring, Hee BS cccoensees SEqui- : noctial [5 Springs. . = Blue Olay. Peat. ca Sand and gravel. im Trias. 612 PROF. W. J. SOLLAS ON THE ESTUARIES OF favour ; yet I shall hope to show that it is one, at all events, supported by evidence of considerable weight. | When, however, all these sources have been admitted as genuine, there still remains one difficulty which has much exercised the minds of many painstaking observers. The Severn and its tributaries are not, except when flooded, very muddy rivers; the wash of the cliffs is not, as a rule, excessive ; the sea, if it furnishes anything, certainly cannot furnish much; and yet the vast body of estuarine water which extends from Weston to Portishead is never otherwise than a sea of more or less diluted mud*. The explanation of this hes in the fact that the water in the tidal portion of the Severn channel flows up and down twice daily at the rate of from 6 to 12 miles an hour, a velocity much greater than that required to move along large boulders of rocks. Water moving at this rate is far more likely to denude than to deposit material ; and indeed in certain parts of the Severn it is, by scouring along great masses of boulders, deepening the channel; and to its past action in this way the deep water known as the “shoots” is attributed by Mr. Richardson. In this rapidly moving body of water, the direction of which is reversed twice daily, the mud discharged by the rivers and washed from the beach accumulates, and from it sediment is supplied to all the tributary estuaries during flood tide; a sufficient diminution in the velocity of the current will of course be marked by the subsi- dence of sediment, and when the velocity sinks to zero sedimentation is copious. Such a cessation of movement appears to take place in the Avon during ebb tide, as Mr. W. R. Browne has well shown by a series of experiments made with an ingenious current-meter devised by my colleague Prof. H. 8. Shaw. I give Mr. Browne’s results in his own words ~ :—‘ In ordinary tidal channels, such as the Avon below Bristol, the course of events during an ebbseemsto beas follows. At first the slope of the surface is exceedingly small (in the Avon it was about 14 foot in 7; miles), and, while the velocity at the surface is considerable, it diminishes rapidly from thence downwards, and at some distance from the bottom becomes nil. This continues for about two thirds of the ebb, the surface-velocity increasing up to a certain point, and then becoming nearly constant. During all this time not only is no scour going on at the bottom, but, if the waters be muddy, an actual deposition of silt is taking: place. At this time, after about two thirds of the ebb, the water has fallen about three quarters of its total height, the slope of the strface has con- siderably increased, and the conditions approximate to those of an ordinary river. The bottom layers of the water then spring suddenly into motion, the surface-velocity diminishes steadily as the tidal waters disappear, until it assumes the normal rate of the low-water flow. * This expression is doubtless somewhat too strong. Actual experiments made in 1837, showed the presence of 5;%5 part, by weight, of sediment in the tidal water opposite Avonmouth, and st part on the opposite coast (B. A. Rep. Trans. of Sections, p. 76). Fresh determinations are doubtless needed. ft Proc. Inst. Civil Engineers, vol. lxvi. p. 1. To this valuable paper is appended the report of a no less valuable discussion. THE SEVERN AND ITS TRIBUTARIES. 613 During this period a scour of the bottom is of course going on ; but, as this velocity is not much higher than in the subsequent period of low-water flow, the rate of scour will not be much greater ; and the actual scour will be insufficient to compensate for the amount of deposit from the tidal waters which has taken place, not only during the period of high water, but also during the first two thirds of the ebb.. It must follow therefore that the scouring effect of the tide is little or nothing, and the observed incapacity of tidal flows to sweep away the silt they have deposited is amply and satisfactorily explained.” Though some of the silt in the tidal water may, as thus explained, stay behind in the estuaries themselves, yet the greater portion is carried seaward ; for, in addition to the oscillating movement of the tidal water, there is of course a discharge into the sea of as much water as the rivers bring down into the estuary, and this is probably accompanied by a transference to the sea of a corresponding quantity of suspended mud, so that the final resting-place of the sediment of the Severn is situated some distance out to sea. But between this quiet spot and the margin where the Severn meets the tide the sedi- ment is carried up and down, far and frequently, and it is not till many journeys are accomplished that it comes permanently to rest. Thus in the waters of the Severn estuary there is a storage of suspended sediment, the accumulation of as many days, or weeks, or months as are occupied in its wanderings to and fro. The accu- mulation is always being diminished by withdrawals seaward, and as constantly renewed by fresh accessions provided by the denuda- tion of the land. This is the whole explanation of the remarkable turbidity of the estuarine Severn. Microscopical Examination of the Tidal Mud.—With a view to throwing some additional light on the sources of the Severn silt, I have examined under the microscope specimens from a large number of localities on both sides of the Severn, including its tributaries ; thus Weston-super-mare, Penarth, Portishead, Avonmouth, and Gloucester are sufficiently far apart for the sampling of the Severn itself; Bridgewater served for the Parrot, Rhymney Bridge (near Cardiff) for the Rhymney, Newport, Mon., for the Usk, Chepstow for the Wye, and Rownham Ferry, near Bristol, for the Avon. The character of the mud from all these places is so similar that a description of one would serve for all the rest. The ingredients may be classed as mineral and organic; the former consist of :—a variable quantity of fine argillaceous granules, smal] angular fragments of colourless transparent quartz containing numerous minute included cavities, a few similar fragments of flint, siliceous fragments of a glauconitic green colour, minute crystals of quartz of the ordinary form, minute prisms of tourmaline, highly dichroic and similar in form to macroscopic prisms of schorl, and minute rhombohedra of calcite. The organic constituents are siliceous and calcareous, the latter include :—coccoliths and rarely coccospheres, both of the ordinary 614 PROF. W. J. SOLLAS ON THE ESTUARIES OF cyatholith type so common in adjacent seas and in the Atlantic ooze; Foraminifera of various species, such as Miliola, which are usually small and scarcely if at all distinguishable from the young of Miliola obesa figured by Max Schultze (Org. der Polythalamien, plate ii.), Tevtularia, probably 7. variabilis, but more than one species is present, Vonionina crassula, Polystomella umbilicata, Rota- ‘lia, sp., Spirillina, sp., and others, including some finely arenaceous forms; spicules of Alcyonaria rarely; fragments of HEchinoderm skeletons and minute spines ; and triradiate spicules of Calcisponges, probably derived from Sycandra ciliata and S. compressa. Most of the Foraminifera are quite empty, glassy and transparent; but some contain a brownish soft granular material; and in one instance a small Rotaline form was observed partially replaced by pyrites. The siliceous constituents are chiefly sponge-spicules, very rarely Radiolaria, and a variable quantity of Diatoms. The sponge-spicules are of very various forms and sizes. They include :—simple acerates, some smooth like those of Amorphina panicea, others entirely microspined, like those of Hymedesmia inflata; simple acuates; acuates with a pin-like head, some of which may have been derivedirom Cliona, and others from Suberitis ficus and other Suberites ; small acuates entirely microspined, similar to the echinating spicules of some species of Dictyocylindrus, and others similar to the smaller acuates of Microciona armata; large pin-headed acuates with the head only spined, similar to the large spicules of Aicrociona armata ; large trifid spicules with simple projecting rays, probably derived from a Geodine sponge; and others with expanded bifid rays somewhat similar to those of Hecionema ponderosa; Geodine globates and stellates of Tethya lyncuriwm ; in one instance a bihamate, such as might have come from Halichondria nornatus. Though many of the spicules are entire, the majority are mere fragments, little rod-like cylinders of very various lengths and thickness, perforated by an enlarged axial canal. From the inorganic constituents but little is to be learned, least of all from the mud; the fragments of quartz are more interesting, since precisely similar fragments abound in the tributary rivers of the estuary, viz. the Severn, Stratford Avon, and Bristol Avon; similar quartz grains, however, are common in the cliffs of Aust. By far the most remarkable constituents are the remains of or- ganisms; for these are all marine, and yet occur on the banks of rivers at a great distance from a truly marine area. Past experience has shown me that rivers sometimes bear to the sea considerable quantities of undissolved caleareous matter derived from the formations through which they flow; thus coecoliths and Foraminifera derived from the denudation of the chalk are always to be found floating in the water of the river Cam near Cambridge, and have in past times been deposited along with other sediment in its gravels, as, for instance, near Barnwell, a fact which has led me to suggest * that the cornstones of the Old Red Sandstone have * Quart. Journ. Geol. Soc. vol. xxxv. p. 492. all THE SEVERN AND ITS TRIBUTARIES. 615 probably been formed from mechanical sediments derived from the denudation of Cambrian limestones, such as those of Bala and Hirnant. Hence in discussing the source of these constituents of the Severn ooze, three possibilities present themselves for investigation. The marine organisms may have been derived (i) from the older for- mations through which the Severn and its tributaries flow ; or (11) from the alluvial flats of its estuary ; or finally (111) from the coast of the Bristol Channel, where under trué marine conditions organisms which could furnish such structures as we have described are known to flourish abundantly. To commence with the first suggestion. Although it may seem improbable that the older formations should have furnished any considerable portion of the organic remains, yet to set the matter at rest, I visited the Severn and two of its tributaries, the Bristol and Stratford Avons (which seemed most likely sources, since they flow through Secondary formations), and obtained from them samples of mud at points above the limit of tidal influence. The locality selected on the Severn was about two miles above Worcester; on the Stratford Avon at Defford; and on the Bristol Avon, at Saltford and Keynsham. Im every sample sponge-spicules occuried pretty freely, but a close examination proved to a certainty that these were all fluviatile, and not marine; they not only agreed in size and shape with spicules of Spongilla fluviatilis, but were sometimes associated with the characteristic spicules of the stato- blasts, or even with the entire statoblast itself. This observation was interesting, as showing not only the entire absence of marine forms, but also a wider distribution of freshwater sponge-spicules than J had previously supposed. Not only were marine spicules absent, but there were also no traces of Foraminifera, coccoliths, or any other marine remains, such as are present in the tidal silt. Thus the rivers, as a possible source of these remains, are eliminated. The chief constituents of the silt at Worcester are fragments of quartz with some little flint ; at Defford similar fragments, together with rhombohedra of calcite, both separate and aggregated into masses like sugar candy, as well as Diatoms and spicules of Spongilla flmiatilis ; at Saltford also fragments of quartz containing minute cavities, minute crystals of quartz, some flint, occasional flakes of mica, many spicules of Spongilla fluviatilis, and Diatums. We have next to inquire into the second possibility, z.¢., the chance of derivation from the ancient alluvium of the Severn ; this is not so readily dismissed. The blue silt of the Severn alluvium is strikingly similar in all its characters to the modern ooze; it consists of a similar admixture of mud and angular siliceous fragments ; while marine sponge-spicules, Foraminifera, coccoliths, and other marine re- mains similar to those of the modern silt, are universally disseminated throughout its mass. ‘They occur in its upper portion where it forms the shore of the estuary, and on the surface of fields where it is tilled ; and they are just as plentiful deeper down, 15 or 20 feet below the surface, as in the new cutting for the railway to the Severn tuunel, Q..d.6. 8. No, Los. 2x 616 PROF. W. J. SOLLAS ON THE ESTUARIES OF aww * m4 2.Crelaceous, ? Paleozoic. = ?Eocene,. PARTLY RYSTFALLIN -J. S. DILLER—-NOTES ON THE 39 30 2 i=) Oo o a i=) ~ ye} as WO - o 1 Ss *) 2 ~ ‘~ . i) yy nd a) N § & A Pen WNIT = , Ugps Gi PA ri wi x S = Lif —_ Tl WS alla} AYA HI ss ot 5 {Qi jt % Ou. (ht i y I< | bar Wie OS—_= Se mht x o ” MOD Miri . je} ~ u 1 ero. ( (is os fe She ee m0 JE = SK fea A) VD6—=s == ey = == —_ lV as Fo V, ‘ S gs S a 9 = ‘ in bo =9 2 5: —=——=KARA at the Rhyl Reservoir, 115; at Vyrnwy water-works, 117; from Ben Eay, by Lochs Clare and Cou- lan to the south-east of Glen Car- ron, 145; across the Carron Valley, between Loch Doule and Loch Carron, 148; in Loch-Ailsh pro- montory, 153; from Glen Finnan, Tioch Sheil, to Caledonian Canal, 155; from east of Herne Bay to near Reculvers, 201; at east corner of Oldhaven Gap, 204; near east corner of Herne Bay, 204; at Peg- well Bay, 204; showing junction of London Clay and Oldbaven series, 206 ; showing a portion of the upper part of the Carboniferous Limestone at Drybrook, 214 ; of Carboniferous strata in the Forest of Dean, 216; vertical diagram, Hook Norton, 229 ; vertical diagram, Sharpshill, 233; general vertical, Chipping Norton, 235; at Ottley Hill, Hook Norton, showing denudation of lower beds of Inferior Oolite, 244 ; from near Castell to Porth-Clais, 268; from Porth-Clais to cliffs near Nun’s Chapel, 268; of junction of Granite with Cambrian strata, right bank of Allan river, Porth-Clais, 275; showing interstratification of tuff and conglomerate above Lower Mills, St. David’s, 290; on South- eastern Railway near Wellington College, 351; from Ullapool to the E.S.E., 865; on the Burn of Calda, 368; on Poulan-drein, 370; from the west end of Ben Uarran to the road south of Inchnadamff, 371; from Loch-Maolack Corry to Dhuloch More, 380; across Coni- veall, 383 ; up Loch Glen Coul and Glen Coul, 391 ; on Camas an Duin, 399 ; of north side of Ben Arnaboll, 403; of western escarpment of Ben Arnaboll, 404; at Whitten Head, 406; in Baron-Hill Park, 473; in the neighbourhood of Bangor, 481 ; across a part of Western Australia, 585; of the alluvial flats of the Severn, 611, 620. Seeley, Prof. H. G., on the dorsal re- gion of the vertebral column of a new Dinosaur (indicating a new genus, Sphenospondylus) from the Wealden of Brook, in the Isle of Wight, 55. GENERAL INDEX, Seeley, Prof. H. G., on the Dinosaurs from the Maastricht beds, 246. Seismographic apparatus, Mr. T. Gray on Gray and Milne’s, 218. Selenaria maculata, 440. parvicella, 441. punctata, 440. Serpentine of the Lizard, 21. of the Troad, 632. Serpentines, analyses of, 258. Settle and Moughton, quarries between, sections in, 104, 105. Severn, Prof. J. W. Sollas om the estuaries of the, and its tributaries, 611. Shales, Lower Cambrian, of St.David’s, 301; green and red, of the Lower Cambrian of St. David’s, 307. Sharpshill, vertical diagram section at, 233. | Shoulder-girdle of Ichthyosaurus, Proc. 45; of Plestosaurus, 46; of Chelone mydas, 47; lacertilian, 48; of Pipa dorsigera, 49; of Calamites cyaneus, 49; of Plestosaurus didu- chus, 58; of Plesiosaurus dolicho- deirus, §9; of Pliosawrus, 60; of Iguanodon} 82, 83. Shropshire, Mr. ©. J. Woodward ona group of minerals from Lilleshall, 466. ' Silt of the Atlantic Docks, Liverpool, Mr. D. Robertson on the, 129. Smittia collaris, 438. Napierii, 438. ——- turrita, 438. Sollas, Prof. W. J., on fossil sponges from the Inferior Oolite, with a notice of some from the Great Oolite, 541. on the estuaries of the Severn and its tributaries: an inquiry into the nature and origin of their tidal sediment and alluvial flats, 611. South Africa, diamond - fields of, Procns: Sphera crassicosta, 526. jimbriata, 525. Sphenospondylus, a new genus of Dino- saurs, Prof. Seeley on, 55, Spherulites, 315. Sponges, Inferior-Oolite, mineral cha- racters of, 550. Sponges, Prof. W. J. Sollas on fossil, from the Great and Inferior Oolites, 541. Steeping Beck, Lincolnshire, 599. Valley, mode of formation of the, 602. Stockport, dvift-deposits of, 94; sec- tion in a ballast-pit at, 95. GENERAL INDEX, Stonesfield, section of Great Oolite near, 171. Strathcarron, metamorphic rocks of, 148. Strength-constants of Japanese rocks, 139. Strome Ferry, 153. Stronchrubie Basin, 395. Stylina conifera, 180. solida, 180. Subsidence, as producing lake-valleys, 79. : Surface, pre-drift, of the land, 120. Switzerland, Jurassic Astrorhizide and Lituolidz from, 25. Synclinal folds, as producing valley- lakes, 73. Tachylyte of the Western Isles of Seotland, 444. Teniopteris Beyrichii, Proc. 3. Tawney, E. B., Esq., and H. Keeping, Esq., on the section at Hordwell Cliffs from the top of the Lower Headon to the base of the Upper Bagshot Sands, 566. -Terebratula Tawneyi, 536. Tertiaries, Lower London, Mr. J. 8. Gardner on some modifications in the classification of the, 197. —— of the Troad, 630. Thamnastrea arachnoides, 558. concinna, 559. —— Lyelli, 188. — mammosa, 190. — microphylla, 188. Waltoni, 189. Thamnonema pisiforme, 549. Thanet beds, 200. Thecosmilia Slatteri, 182. Thracia leqguminosa, 531. - Studeri, 531. Thrust, lateral, its effects in producing lake-valleys, 78. : Thurammina papillata, 27. hemispherica, 28. Tidal mud of the Severn, microscopical examination of the, 613. sediment of the estuaries of the Severn and its tributaries, Prof. W. J. Sollas on the nature and origin of the, 611. Tomes, R. F., Esq., on the fossil Ma- dreporaria of the Great Oolite of the counties of Gloucester and Ox- ford, 168. ——, on some new or imperfectly known Madreporaria from the Coral Rag and Portland Oolite of the counties of Wilts, Oxford, Cam- bridge, and York, 555, 649 Topley, W., Esq., Appendix to Mr. J. S. Diller’s Notes on the Geology of the Troad, 633. Torridon Sandstone, 360. Tortoise, young, development of plas- tron in, 137. Toxoveras Orbignyi, 488. Trent, modern course of the, 606 ; ancient gravels of the, 607 ; diver- sion of the, 608. Trent Valley, 606. Trias, red sand and rubble débris of Tricycloseris limax, 191. Troad, Mr. J. 8. Diller on the Geology of the, 627. , map of the, 628. Tuff and conglomerate, interstratifiva- tion of, St. David’s, 290. Tuts, Lower Cambrian, of St. David’s, 295, 307; analyses of, 296, 297. . Tunnel-Hill cutting, fossils from the Upper Bagshot Sand in, 352. Ty-try, St. Fagans, Cardiff, section of quarry in Lower Lias at, 48. Ullapool, section from, to 8.S.E., 363 ; to S.E., 365; .to. E.S.E., 365; to the E.N.E., 366. river, section along south side of, ey section along the north side of, 366. Upper Limestone of Murchison non- existent, 386. Upton, drift-deposits of, 93; section in sand-pit at, 93. . Valley-lakes, Rey. A. Irving on the origin of, 73. Vegetable remains im amber, Proc. 66. Victoria, Mr. A. W. Waters on fossil Chilostomatous Bryozoa from, 423. Vine, G. R., Esq., on the Corals and Bryozoans of the Wenlock Shales, Proc. 69. Volcanic group of Lower Cambrian, St. David’s, 294. Vyrnwy Water-works, drift-deposits at, IEG: Wales, North, Mr. D. Mackintosh on the positions of boulders in, Pree. 67 , Mr. T. M. Reade on the Drift- beds of the North-west of England and, 83. Walford, E. A., Esq., on the relation of the so-called Northampton Sand of North Oxon to the Clypeus-grit, 224, 650 Waters, A. W., Esq., on fossil Chilo- stomatous Bryozoa from Muddy Creek, Victoria, &c. 423. | Wealden, of Brook in the Isle of Wight, Prof. Seeley on the dorsal, region of the vertebral column of a new Dinosaur (Sphenospondylus) from the, 5d. Wealden fern, new to Britain, Proc. 3. Weaver, drifts of the valley of the, 93. Wellington College, section on the South-Eastern Railway near, 350, 351. Wenlock Shales, Mr. G. R. Vine on the Corals and Bryozoans of the, Proe. 69. Western Australia, section across a part of, 585; rock-specimens from, 586 ; fossils from, 590. Western Isles of Scotland, Prof. J. W.Juddand Mr.G. A. Cole on the Basalt-glass of the, 444. Wethered, E., Esq.. on the Lower Carboniferous rocks of the Forest of Dean, as represented in typical sec- tions at Drybrook, 211. Whidborne, Rey. G. F., on some fos- GENERAL INDEX. sils, chiefly Mollusca, from the Inferior Oolite, 487. Whitten Head, ground between Hope Ferry and, 402; section at, 406. Widnes, drift-deposits at, 89. Wigan Junction Railway, drifts on the, 91. Wiltshire, Madreporaria from, 555. , Wirral, drift-deposits of, 92. Witham Valley, 606. Wollaston Donation Fund, award of, to Prof. J. Milne, Proc. 30. Gold Medal, award of, to W. T. Blanford, Esq., Proc. 29. Woodward, C. J., Esq., on a group of minerals from Lilleshall, Salop, 466. Yarrow Reservoir, drift-deposits at, 96 ; sections at, 96, 97. Yorkshire, Oorallian Madreporaria from, 555. ——, North-west, Mr. D. Mackintosh on the positions of bouldersin, Proc. 6 7. Young, John, Esq., award of the Murchison Fund to, Proc. 31. ——, on Cone-in-cone structure, Proc. 75° END OF VOL. XXXIX, Piinted by TayLor and Francis, Red Lion Court, Fleet Street. PROCEEDINGS OF THE GEOLOGICAL SOCIETY OF LONDON. SESSION 1882-83. November 1, 1882. J. W. Hurxs, Esq., F.R.S., President, in the Chair. Prof. Louis Lartet, of Toulouse, was elected a Foreign Corre- spondent of the Society. The List cf Donations to the Library was read. Specimens of Rocks from Costa Rica (illustrating his paper read on the 24th May, 1882) were presented to the Museum by G. Att- wood, Esq., F.G.S. The following communications were read :— 1. **The Hornblendic and other Schists of the Lizard District, with some Additional Notes on the Serpentine.” By Prof. T. G. Bonney, M.A., F.R.S., Sec. G.S. 2. “Notes on some Upper Jurassic Astrorhizide and Lituolide.” By Dr. Rudolf Hausler, F.G.S. The following specimens were exhibited :— Rocks and Rock-sections, exhibited by Prof. T. G. Bonney in illustration of his paper. Specimens exhibited by Dr. Hiiusler in illustration of his paper. VOL. XXXIX. a 2 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. November 15, 1882. Dr. J. Gwyn Jerrreys, F.R.S., Vice-President, in the Chair. John Edmund Thomas, Esq., Dorset House, Alfred Place, Aber- ystwyth, and Joseph Williams, Esq., Pant-gwyn House, Holywell, Flintshire, were elected Fellows of the Society. The List of Donations to the Library was read. The following communications were read :— 1. “The Drift-beds of the North-west of England and North Wales.—Part 2. Their Nature, Stratigraphy, and Distribution.” By T. Mellard Reade, Esq., C.E., F.G.8. 2. * On the Evidences of Glacial Action in South Brecknockshire and Kast Glamorganshire.” By T. W. Edgeworth David, Esq. Communicated by Professor J. Prestwich, F.R.S., F.G.S. The following specimens were exhibited :— Rock-specimens exhibited by Messrs. Reade and David, in illus- tration of their papers. December 6, 1882. J. W. Hurxs, Esq., F.R.S., President, in the Chair. Charles Bird, Esq., B.A., Mathematical School, Rochester ; Enoch Cartwright, Esq., Park View, Wood Green, Wednesbury ; Henry Eunson, Esq., C.E., 20 Giles Street, Northampton ; William John- stone, Esq., F.I.C., F.C.S., Athenseum Chambers, Lynn; Henry Liversidge, Esq., Jun., Portington Hall, near Howden; Henry George Lyons, Esq., Royal Military Academy, Woolwich; Joseph Mawson, Esq., St. Ives, Acol Road, West Hampstead, N.W.; Horace W. Monckton, Esq., Hare Court, Temple, E.C., and Junior Carlton Club, W.; Henry Alexander Miers, Esq., B.A., Eden Cottage, Beckenham, Kent; John Postlethwaite, Esq., Eskin Place, Keswick, Cumberland; and Thomas Viccars, Esq., The Public College, Tor- quay, were elected Fellows of the Society. The List of Donations to the Library was read. A case of Rocks, Fossils, &c. from the Department of Salto, PROCEEDINGS OF THE GEOLOGICAL SOCIETY. > Uruguay, was presented to the Museum by A. K. Mackinnon, Esq., F.G.S8.; and four specimens of Mexican Building-stones by F. Newman, Esq., F.G.S. The following communications were read :— 1. “ Note on a Wealden Fern, Oleandridium (Teniopteris) Bey- rihit, Schenk, new to Britain.” By John E. H. Peyton, Esq., F.G.S. This fern, figured by Schenk in the ‘ Paleeontographica’ (vol. xix. plate xxix. figs. 6, 7), was discovered near Minden, in the North- west German Wealden-beds, and appears to have been hitherto unknown in England. It was first discovered in the Wadhurst Clay (‘‘ Tilgate stone” of Mantell) of the cliffs east of Hastings, by Mr. Charles Dawson, of Warrior Terrace, St. Leonards, who has a fine collection of Wealden fossils, and was brought to my notice by Professor Augusto de Linares, of the Valladolid University, who has lately discovered the Wealden in the north of Spain. This specimen*, which I have much pleasure in presenting to the Society for their Museum, I found about a fortnight ago, also in our local “ blue-stone” from the Wadhurst Clay of the Hastings cliffs. In connexion with the flora of the Wealden, I may perhaps mention that, besides the ordinary ferns recorded by Mantell, Fitton, Topley, and others, viz. Lonchopteris Mantelli, Sphenopteris gracilis, S. Mantelli, S. Phillipsii, S. Sillimani, &c., I have been fortunate enough to discover the following North-German forms :— Pecopteris Geinitzit, Pecopteris Murchisoni, Pierophyllum schaumburgense (Dunker), and an undetermined one, which I think is Sphenopteris Gepperti. They all occur in the beds of stone in the Wadhurst Clay, which are locally used for building and road-metal. 2. “On the Mechanics of Glaciers, more especially with relation to their Supposed Power of Excavation.” By the Rev. A. Irving, M.A., F.G.S. A specimen of Oleandridium Beyrichii, Schenk, was exhibited, in illustration of Mr. J. E. H. Peyton’s communication. * It varies slightly from the one figured by Schenk in the nervures; and the midrib is ‘“‘ herring-boned.” It bears a strong resemblance to Teniopteris vittata (Brongn.) of the Trias (Geikie’s ‘Text-Book of Geology,’ fig. 358) ; compare also TY. scitamineefolia (Sternberg), from the Stonesfield beds (Phillips’s ‘ Geology of Oxford,’ Diagram xxx. fig. 8). 4 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. December 20, 1882. J. W. Horxs, Esq., F.R.S., President, in the Chair. Percival Fowler, Esq., C.E., 2 Queen Square Place, Westminster, S.W.; Alfred Eley Preston, Hsq., C.E., Belle-Vue, Manningham, Bradford, and Exchange, Bradford ; and Robert Blake White, Esq., C.E., 23 Anerley Park, S.E., and Medella, U.S. of Colombia, South America, were elected Fellows of the Society. The List of Donations to the Library was read. A specimen of Oleandridium Beyrichii, Schenk, from the Wad- hurst Clay of the Hastings cliffs, was presented to the Museum by J. E. H. Peyton, Esq., F.G.S. The following communications were read :— 1. “On Generic Characters in the Order Sauropterygia.” By Prof. Owen, C.B., F.R.S., F.G.S., &e. 2. “On the Origin of Valley-Lakes, mainly with reference to the Lakes of the Northern Alps.” By the Rev. A. Irving, B.A., B.Sc., EGS. January 10, 1883. J. W. Huxxs, Esq., F.R.S., President, in the Chair. T. W. Edgeworth David, Esq., St. Fagans, Cardiff, and Sydney, Australia ; the Earl of Dysart, 29 Chesham Place, 8.W.; John James Hamilton, Esq., Villa Clara, St. Mark’s Road, Notting Hill, W.; Francis Alfred Lucas, Esq., 39 Gloucester Square, Hyde Park, W.; and Meaburn Staniland, Esq., Jun., Harrington Hall, Spilsby, were elected Fellows, and Dr. Otto Torell, F.C.G.S., of Stockholm, a Foreign Member of the Society. The List of Donations to the Library was read. Three specimens of Rocks from the Sierra Buttes, and a spherical stone from a Pot-hole, North Yuba river, Sierra Nevada mountains, California, were presented to the Museum by F. Tendron, Esq., F.G.8. PROCEEDINGS OF THE GEOLOGICAL SOCIETY. 5 The following name of a Fellow of the Society was read out for the first time in conformity with the Bye-Laws (Sect. VI. B, Art. 6), in consequence of the non-payment of the arrears of his contri- bution :—J. T. Dawes, Esq. The following communications were read :— 1. “On the Lower Eocene section between Reculvers and Herne Bay, and on some modifications in the classification of the Lower London Tertiaries.” By J. 8. Gardner, Esq., F.G.S. 2. “On Mr. Dunn’s Notes on the Diamond-fields of South Africa, 1880.” By Francis Oats, Esq., F.G.S. The author referred to the hypothesis put forward in 1880 by Mr. Dunn (Quart. Journ. Geol. Soc. vol. xxxvii. p. 609), that the carbon for the production of the South-African diamonds was furnished by the black carbonaceous shales found throughout the district, and the conclusion drawn by him therefrom that therefore diamonds would not be found below the level of these shales. The author stated that the shales, so far as he knows, do not occur below 270 feet, whilst the ground is successfully worked for diamonds at a depth of 350 feet. He maintained that the carbonaceous shales have nothing to do with the origin of the diamonds, and stated that the “ craters” containing the diamantiferous rock erupted quite different material at an earlier date; and he instanced the occurrence in the Kimberley mine of a mass of “dolerite” between the diamantiferous ground and the surrounding shales. The following specimens were exhibited :— Specimens of quartz with gold, from the Sierra Buttes Mine, Cali- fornia, exhibited, by F. Tendron, Esq., F.G.8. A Perna and other fossils from the lowest division of the Wool- wich and Reading Bottom Bed, in the Croydon section of the Wood- side and East Grinstead Railway, exhibited by H. M. Kiaassen, Esq.., A specimen from the Oxford Clay, exhibited by T. J. George, Esq., F.G.8.; and A series of fossils, exhibited by J. 8. Gardmer, Esq., F.G.8., in illustration of his paper. VOL. XXXIX. h 6 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. January 24, 1883. J. Gwyn Jerrreys, LL.D., F.R.S., Vice-President, in the Chair. Walter Raleigh Browne, Esq., M. Inst. C.H., 38 Belgrave Road, S.W.; Thomas Charles Maggs, Esq., Yeovil; Lieut.-Col. William Alexander Ross, R.A., Acton House, Acton, W.: and Cecil Carus Wilson, Esq., Mayland Vicarage, Maldon, were elected Fellows of the Society. The List of Donations to the Library was read. The following name of a Fellow of the Society was read out for the second time in conformity with the Bye-Laws (Sec. VI. B, Art. 6), in consequence of the non-payment of the arrears of his contri- bution :—J. T. Dawes, Hisq. The following communications were read :— 1. “On Streptelasma Reemeri, sp. nov., from the Wenlock Shale.” By Prof. P. Martin Duncan, M.B. (Lond.), F.R.S., V.P.G.8. 2. “On Cyathophyllun Fletcheri, Edw. & H., sp.” By Prof. P. Martin Duncan, M.B. (Lond.), F.R.S., V.P.G.S. 3. “On the Fossil Madreporaria of the Great Oolite of the Counties of Gloucester and Oxford.” By Robert F. Tomes, Hsq., F.G.S. Specimens of fossil corals were exhibited by Prof. P. M. Duncan, F.R.S., V.P.G.S8., and R. F. Tomes, Esq., F.G.S., in illustration of their papers. February 7, 1883. J. W. Hurxs, Esq., F.R.S., President, in the Chair. G. HE. D’Arcy Adams, Esq,, M.D., 1 Clifton Gardens, W.; Prof. Ferdinand Moritz Krausé, School of Mines, Ballaarat, Australia ; and the Rev. Alfred William Rowe, M.A., Felstead School, Essex, were elected Fellows, and Dr. Karl A. Zittel, of Munich, a Foreign “orrespondent of the Society. PROCEEDINGS OF THE GEOLOGICAL SOCIETY. 7 The List of Donations to the Library was read. The following communications were read :— 1. “On the Metamorphic and Overlying Rocks in parts of Ross and Inverness shires.” By Henry Hicks, M.D., F.G.S. With Notes on the Microscopic structure of some of the Rocks, by Prof. T. G. Bonney, M.A., F.R.S., Sec.G.S. 2. “On the Lower Carboniferous Rocks in the Forest of Dean, as represented in typical sections at Drybrook.” By E. Wethered, Bsa:, 2 :G.8.. F.C8, Specimens were exhibited by Dr. Hicks and EH. Wethered, Esq., in illustration of their papers. 62 8 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. ANNUAL GENERAL MEETING, February 16, 1883. J. W. Hux, Esq., F.R.S., President, in the Chair. REPORT OF THE Councin FoR 1882. In presenting their Report for the year 1882, the Council of the Geological Society regret that they are unable to announce to the Fellows any marked general improvement in the condition of the Society’s affairs, as compared with that shown in their last year’s Report. In one particular, indeed, there is, unfortunately, a falling off, the number of Fellows of the Society having suffered an actual, although very small, decrease. This continued depression is evi- dently due to the same causes that were indicated last year, and therefore, it is to be hoped, of a temporary nature; and it is some satisfaction to find that the Income of the Society during the past year showed an increase upon that of the year 1881. The number of Fellows elected during the year is 45, of whom 36 paid their fees before the end of the year, making, with 7 pre- viously elected Fellows who paid their fees in 1882, a total accession during the year of only 43 Fellows. Against this we have to set the loss by death of 27 Fellows, and by resignation of 13 Fellows, while 4 Fellows were removed from the list for non-payment of contributions, making a total loss of 44 Fellows. On the year, therefore, there is a loss of 1 Fellow. But of the 27 Fellows deceased, 5 were compounders, and 10 non-contributing Fellows, while 1 non-contributing Fellow became resident; and thus the number of contributing Fellows is actually increased by 9, being now 800. The total number of Fellows and Foreign Members and Corre- spondents was 1443 at the end of the year 1881, and 1441 at the end of 1882. At the end of the year 1881 there was one vacancy in the list of Foreign Members; and during 1882 intelligence was received of the death of 2 Foreign Members. ‘Two of these vacancies in the list of Foreign Members were filled up during the year. The death of a Foreign Correspondent in 1882, and the filling-up of the above- mentioned vacancies in the list of Foreign Members produced three vacancies in the list of Foreign Correspondents, two of which were filled up during the year. ‘Thus at the close of 1882 there was one ANNUAL REPORT. 9 vacancy in each of the Foreign lists, both of which have since been filled. The total Receipts on account of Income for the year 1832 were £2663 16s. 2d., being £9 2s. 4d. less than the estimated Income for the year. The total Expenditure, on the other hand, amounted to £2644 19s. 5d., or £118 3s. 7d. more than the estimated Expen- diture of the year, the excess being chiefly incurred in connexion with the production of the Quarterly Journal. At the desire of the President, who kindly bore half the expense of the evening, a Conversazione was held in the Society’s Rooms on the 24th November last. The attendance was hardly so numerous as might have been expected ; but in all other respects the entertain- ment was highly successful. The Council have to announce that Mr. G. W. Ormerod has kindly prepared a Second Supplement to his valuable Classified Index to the publications of the Society, including the volumes of the Quarterly Journal published from 1876 to 1882 inclusive (Vols. XXXII. to XXXVIII.). The thanks of the Society are due to Mr. Ormerod for the labour and care he has devoted to the task of keeping this useful guide to the contents of the Society’s Publications up to date. The manuscript of this Second Supplement has been furnished by Mr. Ormerod, and is now in the printer’s hands. It will be of about the same size as the First Supplement (published in 1876), and will be issued to Fellows at the same price, namely one shilling. The Council have to announce the completion of Vol. XX XVIII. and the commencement of Vol. XX XIX. of the Society’s Quarterly Journal. The Council have awarded the Wollaston Medal to W. T. Blanford, Esq., F.R.S., F.G.S., in recognition of services rendered by him to Geology in Abyssinia and Eastern Persia, and during his long-con- tinued labours in connexion with the Geological Survey of India. The Murchison Medal, with the sum of Ten Guineas from the proceeds of the Fund, has been awarded to Professor H. R. Goppert, F.M.G.S., in testimony of appreciation of his valuable researches in connexion with Paleozoic Botany, which have extended over a period of 50 years. The Lyell Medal, with a sum of Twenty-five Pounds from the proceeds of the Fund, has been awarded to Dr. W. B. Carpenter, C.B., F.R.S., F.G.S., in recognition of the value of his investigations into the Microscopic Structure of Fossils, especially the Foraminifera, and of his labours in connexion with the exploration of the deeper parts of the Ocean. The Bigsby Medal has been awarded to Henry Hicks, M.D., F.G.S., as a token of appreciation of the importance of his labours among the Oldest Fossiliferous and Archzean Rocks of the British Islands. The balance of the proceeds of the Wollaston Donation Fund has been awarded to Professor John Milne, F.G.S., in recognition of the interest attaching to his valuable investigations into the Earthquake- phenomena of Japan, and to assist him in further seismic observa- tions. se) PROCEEDINGS OF THE GEOLOGICAL SOCIETY. The balance of the proceeds of the Murchison Donation Fund has been awarded to John Young, Esq., F.G.S., Curator of the Hunterian Museum in the University of Glasgow, in recognition of his long- continued researches among the Polyzoa and other minute fossil organisms of the Carboniferous strata of the west of Scotland, and to assist him in further investigations of a like kind. The balance of the proceeds of the Lyell Donation Fund has been awarded, in equal parts, to P. H. Carpenter, Esq., M.A., as a testimony to the importance of his investigations into the structure and relationships of the fossil Crinoidea, and to assist him in extend- ing his researches to the order Blastoidea; and to M. Rigaux, of Boulogne, in recognition of the value of his investigations upon the fossils of the Devonian and Jurassic series of the Boulonnais, and to aid him in the further prosecution of his researches. Report oF THE LrpRaARY AND Musrum CoMMITTEE. Labrary, Since the last Anniversary Meeting a great number of valuable additions have been made to the Library, both by donation and by purchase. As Donations the Library has received about 107 volumes of sepa- rately published works and Survey Reports, and about 188 Pamphlets and separate impressions of Memoirs; also about 137 volumes and 102 detached parts of the publications of various Societies, and 16 volumes of independent Periodicals presented chiefly by their re- spective Editors, besides 22 volumes of Newspapers of various kinds. This will constitute a total addition to the Society’s Library, by donation, of about 310 volumes and 188 pamphlets. A considerable number of Maps, Plans, and Sections have been added to the Society’s collections by presentation from various Geo- logical Surveys, from the Ordnance Survey of Great Britain, and from the French Dépot de la Marine. They amount all together to 70 sheets, and include 225 sheets from the Ordnance Survey, and 26 sheets from the Dépdot de la Marine; the Geological Surveys from which Maps have been received are those of Belgium, Finland, Saxony, Sweden, and Switzerland. The Books and Maps just referred to have been received from 127 personal Donors, the Editors or Publishers of 15 Periodicals, and 140 Societies, Surveys, and other Public Bodies, making in all 283 Donors. By Purchase, on the recommendation of the Standing Library Committee, the Library has received the addition of 50 volumes of Books, and of 44 parts (making about 8 volumes) of Periodicals, ANNUAL REPORT. re besides 38 parts of various works published serially. Of the Geological Survey Map of France, 4 sheets and a sheet of sections have been obtained by purchase; and the Society has also purchased M. Ley- merie’s Geological Map of the Haute Garonne in 1 sheet, and a copy of the ‘ Carta Geologica d’ Italia,’ in 2 sheets. The cost of Books, Periodicals, and Maps during the year 1882 was £72 12s. 7d., and of Binding £36 14s. 10d. There was also expended on the mounting of Maps a sum of £13 13s. 4d., and on mounting the new Library Catalogue on writing-paper, in two volumes, £1 16s. 8d. The total expenditure on account of the Library was £124 17s. 5d. The Books in the Society’s Library are generally in good con- dition; and the bindings of many of the older volumes have been repaired during the past year. The Library continues to be much consulted by the Fellows of the Society. Museum. The Collections in the Museum remain in much the same con- dition as at the date of the last Report of the Committee. During the year 1882 several interesting Donations have been made to the Museum, including :—Rock-specimens from Costa Rica, presented by G. Attwood, Esq., F.G.S., and a specimen of Olean- dridium Beyrichii, Schenk, from the Wealden, near Hastings, pre- sented by J. H. H. Peyton, Esq., F.G.S., in illustration of communi- cations read before the Society ; a fine example of Platax altissumus, from Monte Bolea, presented by Lieut.-General Randolph; four specimens of building-stones in common use in Mexico, presented by F. Newman, Hsq., F.G.S.; and a case of fossils, rocks, and minerals from the Department of Salto, Uruguay, presented by A. Mackinnon, Hsq., F.G.S. I2 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. CoMPARATIVE STATEMENT OF THE NUMBER OF THE SOCIETY AT THE CLOSE OF THE YEARS 188] anp 1882. Dee. 3k, t8sl. Dec. 31, 1882. Compounders)..c. oo. see Po DAMM ee 312 Contributing Fellows... .. os dbs eee atenee 800 Non-contributing Fellows .. BO. pn eeaee 248 | 1361 1360 Honorary Members ...... Sk € er eee 3 Foreign Members ........ 8)” wh ad 39 Foreign Correspondents... . BOLT Mie Sater 39 1443 Ce General Statement eaplanatory of the Alterations in the Number of Fellows, Honorary Members, Sc. at the close of the years 1881 and 1882. Number of Compounders, Contributing and Non- 1361 contributing Fellows, December 31, 1881 .... Add Fellows elected during former year and paid 7 Tog Reto) ee arrests nee Seu alta Ramer Rarer ca <5 } Add Fellows elected and paid in 1882 ........ 36 1404 Deduct Compounders deceased .............. 5 Contributing Fellows deceased ........ 12 Non-contributing Fellows deceased .... 10 Contributing Fellows resigned ........ 13 Contributing Fellows removed ........ ey — 44 1360 Number of Honorary Members, Foreign Members, and Foreign Correspondents, 82 Decemiper ole MessiWrrice S505 os ace n «eee Deduct Foreign Members deceased .. ..... 2 Foreign Correspondent deceased.... 1 Foreign Correspondents elected Foreign) Members <:......4 a: } — 5 an Add Foreign Members elected ...... eae Foreign Correspondents elected ...... 2 — 4 — 81 1441 ANNUAL REPORT. 13 DECEASED FELLOWS. Compounders (5). Darwin, C. R., Esq. Stow, G. W., Esq. Parish, Sir Woodbine. White, J., Esq. Perry, Sir T. Erskine. Resident and other Contributing Fellows (12). Adams, Prof. A. Leith. Moggridge, M., Esq. Brown, R., Esq. Molyneux, W., Esq. Davies, Rev. E. Napier, C. G., Esq. Francis, H., Esq. Norman, G. W., Esq. Grimshaw, W. J., Esq. Sharp, S., Esq. Mitchell, R., Esq. Tawney, E. B., Esq. Non-contributing Fellows (10). Blanshard, H., Esq. | Moffat, Dr. T. Calvert, J., Esq. | Noyes, T. H., Esq. Falconer, T., Esq. Thomson, Sir Wyville. Hore, Rev. W. 8. Ward, Rev. J. MacLauchlan, H., Esq. Wood, Rev. H. H. Foreign Members (2). Desor, Prof. E. | Rogers, Prof. W. B. Foreign Correspondent. Kobell, Prof. F. Ritter von. Fellows Resigned (13). Brooke, Sir V. A., Bart. Marshall, J., Esq. Cherry, J. L., Esq. Pooley, E., Esq. - Etheridge, R., jun., Esq. Richardson, Dr. C. T. Griffiths, E. H., Esq. Shrubsole, W. H., Esq. Guppy, R. J. L., Esq. Slatter, G. W., Esq. Johnston, Dr. W. Vyse, G. W., Esq. Leach, Rev. C. Fellows Removed (4). Dyke, HE. G., Esq. Maury, M. F., jun., Esq. Lucas, Joseph, Esq. Williamson, E., Esq. I4 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. The following Personages were elected from the List of Foreign Cor- respondents to fill the vacancies in the Last of Foreign Members during the year 1882. Professor Sven Lovén of Stockholm. Professor Ludwig Ritimeyer of Basle. The following Personages were elected Foreign Correspondents during the year 1882. Professor Louis Lartet of Toulouse. Professor Alphonse Milne-~Edwards of Paris. After the Reports had been read, it was resolved :— That they be received and entered on the Minutes of the Meeting, and that such parts of them as the Council shall think fit be printed and distributed among the Fellows. It was afterwards resolved :— That the thanks of the Society be given to Professor N. 8. Maske- lyne and Professor J. Morris, retiring from the office of Vice- President. That the thanks of the Society be given to J.C. Hawkshaw, Esq., Sir J. Lubbock, Bart., Professor N.S. Maskelyne, Professor J. Morris, and Dr. H. Woodward, retiring from the Council. After the Balloting-glasses had been duly closed, and the Lists examined by the Scrutineers, the following gentlemen were declared to have been duly elected as the Officers and Council for the ensuing ~ year :— ANNUAL REPORT. T5 OFFICERS. PRESIDENT. J: W. Hulke, Esq.,.F.R.S. VICE-PRESIDENTS. Prof. P. M. Duncan, M.B., F.R.S. R. Etheridge, Esq., F.R.S. J. Gwyn Jeffreys, LL.D., F.R.S. Prof. J. Prestwich, M.A., F.R.S8. SHCRETARIES, Prof. T. G. Bonney, M.A., F.R.S. Prof. J. W. Judd, F.R.S. FOREIGN SECRETARY. W. W. Smyth, Esq., M.A., F.R.S. TREASURER. Prof. T. Wiltshire, M.A., F.L.S8. COUNCIL. H. Bauerman, Esq. W. T. Blanford, Esq., F.R.S. Prof. T. G. Bonney, M.A., F.R.S. W. Carruthers, Esq., F.R.S. Prof. P. M. Duncan, M.B., F.R.S. R. Etheridge, Esq., F.R.S. J, Hyans, D.C1., LUD: E.R:S. A. Geikie, LL.D., F.BS. Rey. Edwin Hill, M.A. G. J. Hinde, Ph.D. Prof. T. M*Kenny Hughes, M.A. J. W. Hulke, Esq., F.R.S. J. Gwyn Jeffreys, LL.D., F.R.S. Prof. T. Rupert Jones, F.R.S. Prof. J. W. Judd, F.R.S. S. R. Pattison, Esq. J. A. Phillips, Esq., F.R.S. Prof. J. Prestwich, M.A., F.R.S. F. W. Rudler, Esq. Prof. H. G. Seeley, F.R.S. Warington W.Smyth, Esq., M.A., F.R.S. | W. Topley, Esq. | Prof. T. Wiltshire, M.A., F.L.S. 16 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. LIST OF THE FOREIGN MEMBERS OF THE GEOLOGICAL SOCIETY OF LONDON, x 1882. Date of Hlection. 1827. Dr. H. von Dechen, Bonn. 1844, William Burton Rogers, Esq., Boston, U. S. (Deceased.) 1848. James Hall, Esq., Albany, State of New York. 1850. Professor Bernhard Studer, Berne. 1851. Professor James D. Dana, New Haven, Connecticut. 1853. Count Alexander von Keyserling, Raykiill, Russia. 1853. Professor L.G. de Koninck, Liége. 1854, M. Joachim Barrande, Prague. 1856. Professor Robert Bunsen, For. Mem. R.S., Hecdelberg. 1857, Professor H. R. Goppert, Breslau. 1857. Professor H. B. Geinitz, Dresden. 1857. Dr. Hermann Abich, Vienna. 1859. Dr. Ferdinand Romer, Breslau. 1860. Dr. H. Milne-Edwards, For. Mem. R.S., Paris. 1862. Professor Pierre Merian, Basle. 1864. M. Jules Desnoyers, Paris. 1866. Dr. Joseph Leidy, Philadelphia. 1867. Professor A. Daubrée, For. Mem. R.S., Paras. 1870. Professor Oswald Heer, Zurich. 1871. Dr. Sven Nilsson, Lund. 1871. Dy. Franz Ritter von Hauer, Vienna. 1874. Professor Alphonse Favre, Geneva. 1874. Professor E. Hébert, Paris. 1874. Professor Edouard Desor, Neuchatel. (Deceased.) 1874. Professor Albert Gaudry, Pars. 1875, Professor Fridolin Sandberger, Wiirzburg. 1875. Professor Theodor Kjerulf, Christiania. 1875. Professor F. August Quenstedt, Tiibingen. 1876. Professor E. Beyrich, Berlin. 1877. Dr. Carl Wilhelm Giimbel, Munich. 1877. Dr. Eduard Suess, Vienna. 1879. Dr. F. V. Hayden, Washington. 1879. Major-General N. von Kokscharow, St. Petersburg. 1879. M. Jules Marcou, Cambridge, U.S. 1879. Dr. J. J. 8. Steenstrup, For. Mem.R.8., Copenhagen. 1880. Professor Gustave Dewalque, Lrége. 1880. Baron Adolf Erik Nordenskiéld, Stockholm. 1880. Professor Ferdinand Zirkel, Leipzig. 1881. Il Commendatore Quintino Sella, Rome. 1882. Professor Sven Lovén, Stockholm. 1882. Professor Ludwig Riitimeyer, Basle. ANNUAL REPORT. LIST OF THE FOREIGN CORRESPONDENTS OF THE GEOLOGICAL SOCIETY OF LONDON, 1 18832. Date of Election. 1863. Dr. G. F. Jager, Stuttgart. 1863. Count A. G. Marschall, Vienna. 1865. Professor G. Meneghini, Pisa. 1863. Professor Giuseppe Ponzi, Rome. 1863. Dr. F. Senft, Eisenach. 1864. Dr. Charles Martins, Montpellier. 1866. Professor J. P. Lesley, Philadelphia. 1866. Professor Victor Raulin, Bordeaux. 1866. Baron Achille de Zigno, Padua. 1870. Professor Joseph Szabo, Pesth. 1870. Professor Otto Torell, Land. 1871. M. Henri Coquand, Marseilles. 1871. Professor Giovanni Capellini, Bologna. 1872. Herr Dionys Stur, Vienna. 1872. Professor J. D. Whitney, Cambridge, U. 8. 1874, Professor Igino Cocchi, florence. 1874. M. Gustave H. Cotteau, Auzerre. 1874. Professor G. Seguenza, Messina. 1874. Dr. J. 8. Newberry, New York. 1874. Dr. T. C. Winkler, Haarlem. 1875. Professor Gustay Tschermak, Vienna. 1876. Professor Jules Gosselet, Lille. 1877. Professor George J. Brush, New Haven. 1877. Professor A. L. O. Des Cloizeaux, For. Mem.R.S., Paris. 1877. Professor EK. Renevier, Lausanne. 1877. Count Gaston de Saporta, Aix-en-Provence. 1879. Professor Pierre J. van Beneden, For.Mem.R.S., Louvain. 1879. M. Edouard Dupont, Brussels. 1879. Professor Guglielmo Guiscardi, Naples. 1879. Professor Franz Ritter von Kobell, Munich. (Deceased.) 1879. Professor Gerhard Vom Rath, Bonn. 1879. Dr. Emile Sauvage, Paris. 1880. Professor Luigi Bellardi, Turin. 1880. Dr. Ferdinand von Hochstetter, Vienna. 1880. Professor Leo Lesquereux, Columbus. 1880. Dr. Melchior Neumayr, Vienna. 1880. M. Alphonse Renard, Brussels. 1881. Professor EK. D. Cope, Philadelphia. 1882. Professor Louis Lartet, Toulouse. 1882. Professor Alphonse Milne-Edwards, Paris. 18 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. AWARDS OF THE WOLLASTON MEDAL UNDER THE CONDITIONS OF THE ‘‘ DONATION FUND” ESTABLISHED BY WILLIAM HYDE WOLLASTON, M.D., E.R.S., F.G.S., &c. ‘To promote researches concerning the mineral structure of the earth, and to enable the Council of the Geological Society to reward those individuals of any country by whom such researches may hereafter be made,”—“ such individual not being a Member of the Council.”’ 1831. Mr. William Smith. 1858 ae Hermann von Meyer. 1885. Dr. G. A. Mantell. Mr. James Hall. 1836. M. L. Agassiz. 1859, Mr. Charles Darwin. 1837. ae T. P. Cautley. 1860. Mr. Searles V. Wood. Dr. H. Falconer. 1861. Professor Dr. H. G. Bronn. 1838. Professor R. Owen. 1862. Mr. R. A. C. Godwin-Austen. 1839. Professor C. G. Ehrenberg. 1865. Professor Gustav Bischof. 1840, Professor A. H. Dumont. 1864. Sir R. I. Murchison. 1841. M. Adolphe T. Brongniart. | 1865. Dr. Thomas Davidson. 1842. Baron L. von Buch. 1866. Sir Charles Lyell. 1343. a Elie de Beaumont. 1867. Mr. G. Poulett Scrope. M. P. A. Daufrénoy. 1868. Professor Carl F. Naumann. 1844. Rev. W. D. Conybeare. 1869. Dr. H. C. Sorby. 1845. Professor John Phillips. 1870. Professor G. P. Deshayes. 1846. Mr. William Lonsdale. 1871. Sir A. C. Ramsay. 1847. Dr. Ami Boué. 1872. Professor J. D. Dana. 1848. Rev. Dr. W. Buckland. 1873. Sir P. de M. Grey-Egerton. 1849, Professor Joseph Prestwich. | 1874. Professor Oswald Heer. 1850. Mr. William Hopkins. 1875. Professor L. G. de Koninck. 1851. Rev. Prof. A. Sedgwick. 1876. Professor T. H. Huxley. 1852. Dr. W. H. Fitton. 1877. Mr. Robert Mallet. 1853 he le Vicomte A. d’Archiac. | 1878. Dr. Thomas Wright. ’ |M. E. de Verneuil. 1879. Professor Bernhard Studer. 1854. Sir Richard Griffith. 1880. Professor Auguste Daubrée. 1855. Sir H. T. De la Beche. 1881. Professor P. Martin Duncan. 1856, Sir W. EK, Logan. 1882. Dr. Franz Ritter von Hauer. 1857. M. Joachim Barrande. 1883. Mr. W. T. Blanford. 1831. 1833. 1854. 1835. 1836. 1838. 1839. 1840. 1841. 1842. 1843. 1844. 1845. 1846. 1847. 1848. 1849. 1850. 1851. 1852. 1853. 1854, 1855. 1856. 1857. 1858. ANNUAL REPORT. ¢ AWARDS OF THE BALANCE OF THE PROCEEDS OF THE WOLLASTON “ DONATION-FUND.” Mr. William Smith. Mr. William Lonsdale. M. Louis Agassiz. Dr. G. A. Mantell. Professor G. P. Deshayes. Professor Richard Owen. Professor C. G. Ehrenberg. Mr, J. De Carle Sowerby. Professor Edward Forbes. Professor John Morris. Professor John Morris. Mr. William Lonsdale. Mr. Geddes Bain. Mr. William Lonsdale. M. Alcide d’Orbigny. Cape-of-Good-Hope Fossils. M. Alcide d’Orbigny. Mr. William Lonsdale. Professor John Morris. M. Joachim Barrande. Professor John Morris. Professor L. G. de Koninck. Mr. 8S. P. Woodward. Drs. G. and F. Sandberger. Professor G. P. Deshayes. Mr. S. P. Woodward. Mr. James Hall. 1859. 1860. 1861. 1862. 1865. 1864. 1865. 1866. 1867. 1868. 1869. 1870. 1871. 1872. 1875. 1874. 1875. 1876. 1877. 1878. 1879. 1880. 1881. 1882. 1885. =, Mr. Charles Peach. Professor T. Rupert Jones. ae W. K. Parker. Professor A. Daubrée. Professor Oswald Heer. Professor Ferdinand Senft. Professor G. P. Deshayes. Mr. J. W. Salter. Dr. Henry Woodward. Mr. W. H. Baily. M. J. Bosquet. Mr. W. Carruthers. M. Marie Rouault. Mr. R. Etheridge. Mr. James Croll. Professor J. W. Judd. Dr. Henri Nyst. Mr. L. C. Miall. Professor Giuseppe Seguenza. Mr. R. Etheridge, Jun. Mr. W. J. Sollas. Mr. 8. Allport. Mr. Thomas Davies. Dr. R. H. Traquair. Dr. G. J. Hinde. Professor J. Milne. AWARDS OF THE MURCHISON MEDAL AND OF THE PROCEEDS OF “THE MURCHISON GEOLOGICAL FUND,” ESTABLISHED UNDER THE WILL OF THE LATE SIR RODERICK IMPEY MURCHISON, Barr., F.R.S., F.GS. ‘To be applied in every consecutive year in such manner as the Council of the Society may deem most useful in advancing geological science, whether by granting sums of money to travellers in pursuit of know- ledge, to authors of memoirs, or to persons actually employed in any inquiries bearing upon the science of geology, or in rewarding any 290 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. . such travellers, authors, or other persons, and the Medal to be given to some person to whom such Council shall grant any sum of money or recompense in respect of geological science.” 1873. 1873. 1874. 1874. 1874. 1875. 1875. 1876. 1876. 1877. 1877. 1878. ESTABLISHED UNDER. THE WILL AND CODICIL Mr. William Davies. Medal. Professor Oswald Heer. Dr. J. J. Bigsby. Medal. Mr. Alfred Bell. Professor Ralph Tate. Mr. W. J. Henwood. Medal. Prof. H. G. Seeley. Mr. A. R.C. Selwyn. Medal. Mr. James Croll. Rev. W. B. Clarke. Medal. Rev. J. F. Blake. Dr. H. B. Geinitz. Medal. 1878. 1879. 1879. 1880. 1881. 1881. 1882. 1882. 1883. 1885. Mr. C. Lapworth. Professor F. M‘Coy. Medal. Mr. J. W. Kirkby. Mr. R. Etheridge. Medal. Professor A.Geikie. Medal. Mr. F. Rutley. Professor J.Gosselet. Medal. Professor 'T. Rupert Jones. Professor H. R. Gdppert. Medal. Mr. John Young. AWARDS OF THE LYELL MEDAL AND OF THE PROCEEDS OF THE “LYELL GEOLOGICAL FUND,” OF THE LATE SIR CHARLES LYELL, Barr... F.R.S., F.G.S8. The Medal ‘to be given annually” (or from time to time) “as a mark of honorary distinction as an expression on the part of the governing body of the Society that the Medallist has deserved well of the Science,”—‘‘ not less than one third of the annual interest [of the fund | to accompany the Medal, the remaining interest to be given in — one or more portions at the discretion of the Council for the encou- ragement of Geology or of any of the allied sciences by which they shall consider Geology to have been most materially advanced.” 1876. 1877. Sif. 1878. 1878. 1879. 1879. 1879. 1880. Professor John Morris. Medal. Dr. James Hector. Medal. Mr. W. Pengelly. Mr. G. Busk. Medal. Dr. W. Waagen. Professor Edmond Hébert. Medal. Professor H. A. Nicholson. Dr. Henry Woodward. Mr. John Evans. Medal. 1880. 1881. 1881. 1881. 1882. 1882. 1882. 1883. 1883. 1883 Professor F. Quenstedt. Principal J. W. Dawson. Medal. Dr. Anton Fritsch. Mr. G. R. Vine. Dr. J. Lycett. Medal. Rev. Norman Glass. Professor C. Lapworth. Dr. W. B. Carpenter. Meda. Mr. P. H. Carpenter. M. E. Rigaux. ANNUAL REPORT. 21 AWARDS OF THE BIGSBY MEDAL, FOUNDED BY Dr. J. J. BIGSBY, F-.R.S., F.G:S. To be awarded biennially “as an acknowledgment of eminent services in any department of Geology, irrespective of the receiver’s country ; but he must not be older than 45 years at his last birthday, thus probably not too old for further work, and not too young to have done much.” ; 1877. Professor O. C. Marsh. 1881. Dr. C. Barrois. 1879. Professor E. D. Cope. 1883. Dr. Henry Hicks. AWARDS OF THE PROCEEDS OF THE BARLOW- JAMESON FUND, ESTABLISHED UNDER THE WILL OF THE LATE Dr. H. C. BARLOW, F.GS. “The perpetual interest to be applied every two or three years, as may be approved by the Council, to or for the advancement of Geological Science.” 1880. Purchase of microscope. 1882. Baron C. von Ettingshausen. 1881. Purchase of microscope lamps. VOL. XXXIX., ; € 22 PROCEEDINGS OF THE GEOLOGICAL SOCIETY. Estimates for INCOME EXPECTED. £ s. d.. Lapse: Due for Subscriptions for Quarterly Journal .. 3 4 0 Due for Arrears of Annual Contributions ...... 190 0 0 Due for Arrears of Admission-fees ............ 50 8 0 2438 12 0 Estimated Ordinary Income for 1883 :— Annual Contributions from Resident Fellows, and Non- residents ‘of 1899. 10 186h 1. cis eet wiv ae bone 1420 0 0 Sh dhawstonctoes o<.- See ca eetame aes ak< coN eee 226 16 0 COMPOSIMONS to, em tin elena sadehote ee osok te eee 199 10 0 Annual Contributions in advance .......e.seeeeeees 21 0 O Dividends on Consols and Reduced 3 per Cents ........4. 233 0 11 Advertisements in Quarterly Journal............0c00 eee 810 0 Sale of Transactions, Library-catalocue, Orme- rod’s Index, Hochstetter’s New Zealand, and Naish or slellonys jen. fn ise. a akin ees as SA 10 05-0 Sale of Quarterly Journal, including Longman’s account Sale of Geological Map, including Stanford’s AOCO UE 5 sheydce a otek 5 plea ee blo 4s Haieiele leis ity « 20 0)0 £2607 8 11 THOMAS, WILTSHIRE, Treas. 7 Feb. 1883, FINANCIAL REPORT. 23 the Year 18838. EXPENDITURE ESTIMATED. sda rae 8, Gs House Expenditure : PANES itch JNSUTANCE: |. acaace sayencaedelwovsiemsacsess 36 3 9 ES Sah Ant Rein Aaa Be ge ea Abe RO RR emi ee 25 0-0 DE OCh eats oc oeiat itseiarcts yplniea sin aniie'acs'ca we aie 35 0 0 TENERTINGUES 5 28 fe Mets savior ch w scales ciieigols doe went eeves 157 0.0 House-repairs and Maintenance.................. 20 0 O PRTaNTUNL CLE ME oa paieam aainaves Mastriamee sian senteleee 207 Oese Washing and sundry small Expenses ......... 35 0 0 Darah Mectin a Sihts 5 Jan. s2<..0. 056s 202 140 a Reduced 3 per Cents ...... 30 61d ee Taylor & Francis: Advertisements in Journal, Vol.37.. 9 9 O Publications : Saleor Journal, Vole; 137 fic o.soecocoentee 166-"Sa00 a NOL OSt cccecutcurcs aeaceepe 91137 Sale of Library Catalogue ..................++- 7 13.90 Nilevor Geological Map: o.2 veaeteatsc-e-cncce- 3112 8 Salle:of Ormerod?s Index. i.22..0. 0. ccssaettocese 1S Sale of Hochstetter’s New Zealand .......... 0 6 0 Sale of Dist of Hellows®.2:.cis.c0<0s.ceete eens O10 298 17 9 *Due from Messrs. Longman, in addition to the above, on Journal, Vol. 38, &c.......... Rear ee: 63 12 4 Due from Stanford on account of Geological Map 1015 O (hep iho a: £2744 16 7 We have compared this statement with the Books and Accounts presented to us, and find them to agree. (Signed) H. BAUERMAN, ih W. H. HUDLESTON, ¢ 4"%rs 7 Feb. 1883. FINANCIAL REPORT. Year ending 31 December, 1882. EXPENDITURE. House Expenditure: ase SEED rte ds Salok yaa Sn eciasias stra ge dua ctee aoaaies £5 10 IPe-SASUT AEE osc ones od ne oe