(Storatll Wftiwttttg |f itag BOUGHT WITH THE INCOME FROM THE SAGE ENDOWMENT FUND THE GIFT OF iHcin u m. Sage 1891 to.(in.i8.flL flM).fc ' Cornell University Library QE 262.K55U87 1904 The geology of the country around Kings- 3 1924 004 552 034 The original of this book is in the Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924004552034 355 and 356. MEMOIRS OF THE GEOLOGICAL SURVEY. ENGLAND AND WALES. THE GEOLOGY OF - THE COUNTEY AROUND KINGSBRIDGE AND SALCGMBE. (Explanation of Sheets 355 & 356.) BY W. A. E. USSHER, F.G.S. PUBLISHED BY ORDER OF THE IOKBS COMMISSIONERS OF HIS MAJESTY'S TREASURY. LONDON : PRINTED FOR HIS MAJESTY'S STATIONERY OFFICE, By Wyman and Sons, Limited, Fetter Lane, E.C. And to be purchased from E. STANFORD, 18, 13, and 14, Long Acre, London ; JOHN MENZIES & Co., Rose Street, Edinburgh ; HODGES, FIGGIS & Co., Grafton Street, Dublin ; From any Agent for the sale of Ordnance Survey Maps ; or through any Bookseller from the Ordnance Surrey Office, Southampton. 1904. Price Is. 6d. LIST OF MAPS, SECTIONS, AND MEMOIRS OF THE GEOLOGICAL SURVEY OF ENGLAND AND WALES, AND MUSEUM OF PRACTICAL GEOLOGY. J.' J. H. Teall, M.A., F.R.S., Director of the Geological Survey and Museum, Jermyn Street, London, S.W. The Maps and Memoirs are now issued by the Ordnance Survey. They can be obtained from Agents or direct from the Ordnance Survey Office, Southampton. Museum Catalogues, Guides, dfcc., are sold at the Museum, '28, Jermyn Street, London. A Complete List of the Publications can be obtained from the Ordnance Survey Office, Southampton. Price 1». GENERAL MAP (one inch to 4 miles). ENGLAND AND WALES.— Sheet UTitle) ; 2 (Northumberland, &c); 3 (Index of Colours); 4 (L of Man); 5 (Lake District) ; 6 (E. Yorkshire) ; 7 (North Wales) ; 8 (Central England) ; 9 (Eastern Counties) ; 10 (South Wales and N. Devon) ; 11 (W. of England and S.E. Wales) ; 12 (London Basin and Weald) ; 13 (Cornwall, &c.) ; 14 (S rath Coast. Torquay to I. of Wight) ; 15 (S Coast, Havant to Hastings). Sheet 1, 2s. : sheets 2 to 15, 2s. ed. each. Printei in colours. MAPS (one-inch). Old Series. Nos. 1 to 110 in whole sheets and quarter-sheets, hand-coloured, in two editions, Solid and Drift — except 92 NE. , SE., 93 NW., SW., 97 SE„ 98 NW., SW, S.E., and 101 S.E., which are published Solid only. Prices, whole sheets, 4s. to 8s. ed. ; quarter-sheets, Is. ed. to 3s. Large map of London and environs, Solid, 22s. Sheet 7, Drift 18s. Bd. MAPS (one-inch). New Series. 1 to 73. These New Series maps are identical with the Quarter Sheets of the Old Series, Nos. 91 to 110, all of which are published with Drift, excepting Sheets 29, 38, 48, 49, 51, 61, 62, 69, and 70. LONDON in four Sheets (Colour printed) each . ISLE OF WIGHT in one Sheet (Colour printed) ■ ISLE OF MAN Sheets 86, 45, 46, 56 and 57 123. STOKE - UPON - TRENT (Colour printed) 156. ATHEHSTONE 156. LEICESTER (Drift ; colour printed) 187. HUNTINGDON 203. BEDFORD 281. MERTHYR TYDFIL (Colour printed) 282. ABERGAVENNY (Colour printed) . . 148. PONTYPRIDD (Colour printed) 249. NEWPORT (MON.), (Solid; colour printed) 261 and 262. BRIDGEND and COAST near KENFIG (Colour printed) 263. CARDIFF and WORLE, SOMER- SET (Colour printed) 267. NEWBURY 268. READING 282. DEVIZES B83. ANDOVER 284. BASINGSTOKE 298. SALISBURY<(Oolour printed) 299. WINCHESTER 300. NEW ALRESFORD 314. RINGWOOD (Colour printed) 315. SOUTHAMPTON .. Price. Solid. Drift. s. d. s. d. — 1 Price. Solid. Drift. 16 16 s. d. 316. HAVANT .. 317. CHICHESTER (Colour printed) .. 326. EXETER 328. DORCHESTER 329. BOURNEMOUTH, WIMBORNB MINSTpR 330. Parts of NEW FOREST and ISLE OF WIGHT (Drift; colour printed) 331. PORTSMOUTH and part of ISLE OF WIGHT (Drift; colour printed) 832. BOGNOR, SELSEA, LITTLE- HAMPTON 333. WORTHING, ROTTINGDEAN .. 884. NEWHAVEN, EASTBOURNE 389. NEWTON ABBOT 840. OTTERTON .. ..' .. 341. DORSET COAST, LANGTON HERRING 842. PORTLAND, WEYMOUTH, LUL- WORTH 848. SWANAGE, CORFE CASTLE .. 349. PLYMOUTH and IVYBRIDGE . . 860. TORQUAY 866. KINGSBRIDGE 866. START POINT s. d. 3 3 3 3 16 3 16 16 16 — 16 — 3 0. — 3 — 16 — 1 6 — 3 — 3 — 3 0. — 8 oi — 3 — 1 6 MAPS (SIX-INCH). The Coal-fields and other mineral districts are in part published on a scale of six inches to a mile, at 4s to 6» eacW coloured ; those of N. Staffordshire at Is. 6d., uncoloured. MS. Coloured Copies of other six-inch maps, not intended for publication, are deposited for reference in the Geological Survey Office, Jermyn Street, London, and copies can be suDDllei at tie cost of drawing and colouring the same. ft 5| HORIZONTAL SECTIONS. 1 to 140, 146 to 148, price 5s. each. VERTICAL SECTIONS. "il 1 to 86, price 8s. 6d. each. - : ' M ,* bi o tt} y— -. 'fe*^ ^£l & « n CO w £ II tcj a a s* •p-t s 5 s V N -3 1 m c "* +■» <» s o © £ s *o fe 855 and 356. MEMOIRS OF THE GEOLOGICAL SURVEY. ENGLAND AND WALES. THE GEOLOGY OF THE COUNTRY AROUND KINGSBRIDGE AND SALCOMBE. (Explanation of Sheets 355 & 356.) BY W. A. E. USSHER, F.G.S. [TBL1SHID BT OBDEB, OP THE LORDS COMMISSIOHBBS OP HIS MAJESTY'S TREASUBT LONDON : PRINTED FOR HIS MAJESTY'S STATIONERY OFFICE, By Wyman and Sons, Limited, Fetter Lane, E.C. And to be purchased from E. STANFORD, 12, 13, and 14, Long Acre, London -, JOHN MENZIES & Co., Rose Street, Edinburgh ; HODGES, FIGGIS & Co., Grafton Street, Dublin ; jTrom any Agent for the sale of Ordnance Survey Maps ; or through any Bookseller from the Ordnance Surrey Office, Southampton. 1904. Price Is. 6d. CONTENTS. Page. Preface by the Director * • "i Chapter I.— Introduction. General Geology and Table of Strata. Northern Area, Nature of Sediments, Chemical Changes, Effects of Lateral Pressure - ■ 1 Chapter II. — Distribution of the Kocks op the Northern Area. Dartmouth Slates, Meadf oot Group— Kingmore type — Torcross Type — Beeson Grits — Tinsey Head Slate Series 13 Chapter III.— Northern Area — continued. Igneous Rocks, Eastern Coast Section, Western Coast Section - 26 Chapter IV.— The Southern Area. General Notes, Relations of Groups, Nature of Disturbances, &c, Literature, Mica- schist Group, Green Schists 37 Chapter V. — Nature of the boundary between Northern and Southern Areas - 52 Chapter VI. — New Red Rocks or Permian 63- Chapter VII. — Pleistocene and Recent 66 Chapter .VIII. — Economics. Water Supply, Mines, Building materials, &c, Soils - 75 Appendix.— List of Principal Works on the Geology of the District ' 77 Index - 79 ILLUSTRATIONS. Frontispiece. Cliff between Outer and Inner Hope. Fig. 1. Diagram, explanatory of Strikes - - 6 „ 2. Gnarled fragments of Tinsey Head beds - - 6 „ 3-8. Gnarling by Kingsbridge Estuary Creeks and at Torcross - 8 „ 9. Corrugated quartz-veins, near Torcross - 9 „ 10. Contorted igneous band in slates, near Ayrmer Cove 9 „ 11. Gnarled grit-shales, behind Tea House, Westcombe Beach 10 „ 12. Gnarling, by Collapit Creek ... 10 „ 13. Contorted laminse, Bowcombe Creek - 10 „ 14. South of Biddlehead Point, Salcombe Estuary 38 „ 15. Coves under the Moult 38 „ 16. Southpool Creek, E. of Westerncombe 38 „ 17. Spirit of the Ocean Cove - 38 „ 18. Mica-quartz-schist of Bolt Head ■ - - - 39 „ 19. Gnarled quartz-mica-schist, Hall Sands Coast south of Hard Stone - 40 „ 20. Quartz-schist from the cliff on north side of Hall Sands Village - - 40 „ 21. Plication in quartz-schists near Southdown Farm - - 41 „ 22. Cliff at Greenstraight ... 5a „ 23. Southpool Creek Section, west side - - 55 „ 24. Faulted mica-schist, Kingsbridge Estuary 56 „ 25. Diagram of cliffs near Hope 58 „ 26. Mouthwell, north side -58 „ 27. Diagram showing the Beeson Grits as an outcropping series - 62 „ 28. Current-bedding near Slapton Bridge - - - 63 „ 29. Horswell House Quarry - g 4 „ 30. Cliff between Leas Foot Sand and Thurlestone Sand - 64 „ 31. The Pinnacles, near Matchcombe Sand ... 69 PREFACE. This Memoir treats of the area represented by Sheets 355 and 356 of the New Series of one-inch maps. Since the first edition of these maps was published in 1898, Mr. Ussher has established the fact that the Devonian rocks on the south of the Dartmouth slates are represented by the Meadfoot group which overlies them on the north. These slates, therefore, are the oidest member of the Devonian sequence in this area. As this is a point of great stratigraphical importance it has been deemed advisable to indicate upon the maps the approximate boundary of these two groups, and this has accordingly been done on a new edition which will shortly be published. The boundary is represented by a broken line, which may be followed from the coast near Ringmore (Sheet 355) to the north-east corner of the map. It reappears again near the north margin of Sheet 356 where, owing to folding, the relations of the two groups are more complicated than they are in the adjacent map. A striking feature of the area in question is the contrast between the intensely folded but comparatively unaltered Devonian rocks in the north and the metamorphic rocks on the south. This has given rise to a somewhat extensive controversial literature which is referred to in the following pages and in which many facts of great importance are described. The age of the schists is left undecided by the author, though more definite views on the subject have been expressed in favour of their antiquity by Professor Bonney and Miss Raisin, and in favour of their being metamorphosed Devonian rocks by Jukes and Holl, A. Somervail and A. R. Hunt. Although Mr. Ussher has not been able to settle the question of the age of the meta- morphic rocks by stratigraphical evidence, his detailed mapping; of the green schists and his careful description of the line which separates the altered from the unaltered rocks will be found of value to all future workers in the district. J. J. H. TEALL, Director. Geological Survey Office, 28, Jermyn Street, .London. 7th April, 1904. 7294. 500— Wt. 2304. 6/04 Wy. & S. 1449r. GEOLOGY OF THE COUNTRY AROUND KINGSBRIDGE AND SALCOMBE. CHAPTER I. INTRODUCTION. Sheets 355 and 356 of the Geological Survey Map show a land-area of about 78 square miles, of which 23 are included in Sheet 356. This, the most southerly district of Devonshire, lies to the south of a line drawn from the vicinity of Blackpool Sands in Start Bay westwards to the mouth of the Yealm. It is bounded on all sides except the north by the sea, and the coast line is breached by the estuary of the Avon and by the Salcombe creeks. Prawle Point is the most southerly headland ; to the east is Start Point and to the west Bolt Head and Bolt Tail. The area is hilly, the summit-levels being about 300 feet, except towards the north margin and in the district between the Start and Bolt Tail, where they attain to heights between 400 and 500 feet. The streams which fall into the Salcombe Estuary and its creeks drain an area of about 32 square miles, and, with the exception of the lower reaches of the Avon, the remaining portion of the area is drained by small streams which have direct outlets on the coast, the most considerable being the South Milton stream. The Salcombe creek is navigable at high water to Kingsbridge, but above Salcombe the scenery is tame in comparison to the seaward aspect. Salcombe may be regarded as the port of Kings- bridge, which is the only town in the district ; Bantham and more especially Torcross are places of summer resort. The coast scenery exhibits great diversity and many striking , features. The rugged grandeur of the cliffs between Bolt Head and Bolt Tail and near the Start is unsurpassed elsewhere on the Devon coasts. Further north, on either side of the promontory, bold stretches of cliff based by platforms of rock give place, here and there, to lower cliffs, flanked by the sand beaches of Thurlestone, Bantham, and Erme Mouth on the west, and by the shingle beaches of Hall Sands, Beesands, and of the celebrated Slapton Sands on the east. The inland scenery, although not devoid of interest and, where extensive views are obtain- able, of beauty, is mainly a succession of hills of greater or less 7204. A 2 GEOLOGY OF KINGSBKIDGE AND SALCOMBE. elevation, usually separated by narrow steep-sided valleys in places clothed with woods and copses. , , , V The geology of thedistrict was shown by Dela Beche on the old GeologLl Survey maps (parts of sheets 23 and 24) to consist of Devonian Kocks separated from mica and chlorite schists of doubtful age by a boundary runnW from Hope by Malborough eastward to Start Bay. Two patcles of New Bed Sandstone were indicated near Thurlestone rock and at Slapton, and a few lenticular strips of greenstone were mapped west of Torcross and on the Revelstoke coast. The new series of geological maps 355 and 356 present the results of a detailed survey of the area made on six-inch maps and reduced therefrom. In the Devonian area local developments of grit have been indicated, and many additional exposures of igneous rocks recorded. In the meta- morphic area the chlorite schists of De la Beche have been separ- ated from the mica-schists, with the result that the complicated folding of the rocks of this group is well expressed on the map. The district presents two great groups of rocks, sharply differentiated along a nearly straight line from coast to coast. On the south of this line is a series of highly metamorphosed rocks, on the north of it a series of comparatively unaltered Devonian rocks. These groups have been respectively included under the headings — Southern Area, Northern Area, priority ot description being given to the latter. The Devonian rocks con- sist mainly of slates, associated locally with fine-grained grit beds,silty slates and shales, and argillaceous and fine silty siliceous interlaminations. In these materials calcareous bands and basic igneous rocks occur sparingly on certain horizons, but are nowhere present in any connected mass. Owing to the general east and west strike the Northern Area may be divided into four geographical zones or belts characterised by certain generally recognizable peculiarities, but nowhere sharply defined. The Dartmouth slates are also present near Strete and west of Bigbury ; their southern boundary, which is a most important one, is, as. far as embraced in the Maps under consideration, exceedingly indefinite, but it was not drawn until some years after their first publication, and has since been added. On the south of these slates the rocks occupying successive belts Proceeding southward were named The Ringmore beds, The brcross slates, The Beeson Grits, The Tinsey Head Slate series, including the interlaminated beds of Tinsey Head. The chronological relations of the above zonal types depend on the nature of the outcrops of the Dartmouth slates and of the Beeson grit -group, which could not be made out during the survey of the area in a satisfactory manner. In the present year (1902) an opportunity was afforded of comparing the rocks of the coast sections of this area with those of the Looe, Tregantle and Ply- mouth coasts, and of bringing all the information acquired subsequent to the survey of this area to bear on them with the result that the seeming zones must now be regarded as carts of the same general group, viz., the Meadfoot Beds, in which rh P Rrngmore type corresponds to the fossiliferous Looe Beds INTRODUCTION. The Metamorphic rocks of the Southern Area consist of two groups, viz : — 1. Green Schists proved by penological investigation to con- stitute a series of highly altered basic igneous rocks. . 2. Mica and Quartz Schists, betraying in their mode of associa- tion and in less altered parts their aqueous origin as shales and fine silty shales often lnterlaminated, and with beds of grit. The Green Schists are as a mass above the Mica Schist group, although there are frequent signs of intercalation at or near the junction, and occasional traces of rock evidently of igneous origin are met with in the Mica Schists. The boundaries of the Green Schists can, as a rule, be traced without much difficulty. Their distribution is suggestive of a northerly attenuation. No separation can be made between mica and quartz schists, these being so intermingled .that unless specially stated the term Mica Schist covers all the associated quartz schists. The following is a table of Rocks and Deposits, etc. : — /Blown Sand. Recent. ( Sand and Shingle Beaches. I Alluvium. [ Submarine Forests. p r _.„„„„„„_ / Old alluvial tracts. l-LKI&TOCENE Head _ oM talus accllmu l a tions. \ Raised Beaches. New Red Sandstone (Permian.)— Breccia. Slapton, Thurlestone Sands. P o o t- Meadfoot and Looe Beds. Dartmouth Slates, Northern Area. I Tinsey Head slate series — dark grey slates in which are included the iriterlaminate beds of Tinsey Head. Beeson grits— slates with red- dish buff and grey grits of Warberry, Looe, and Mead- foot types. Torcross type — grey slates with siliceous and calcareous bands and lenticles, occa- sional grit beds and im- persistent bands of igneous rock, probably contempo- raneous. Ringmore type — lilac - red, greenish, red and buff and grey partly siliceous slates or shales, with occasional grit beds, and_ decomposed calcareous f ossilif erous seams. Glossy green, purple, lilac and buff, partly siliceous slates with occasional beds of hard grit and quartzite. 7294. Coast south of Beesands to coast north of Hope. . At intervals from Beeson to Beacon Point, north of Hope. Torcross and Sunnydale to the Thurlestone coast. Slapton to Avonmouth, Ringmore and Ayrmer (Arm- our) Cove. Strete Coast and northern margin of sheet 356. Revelstoke coast, west of Bigbury, sheet 355. A2 GEOLOGY OP KINGSBRIDGE AND SALCOMBE. Northern Area— continued. .Quartz-porphyry or rhyolite. Acid Igneous Bocks. Basic I Felsites. /Decomposed sheared rocks more or less altered, for the most part associated with rocks ol Torcross type, but locally found in Dartmouth slates ( and Kingmore beds. Typical ophitic dolerite. Diabase porphyrite. Associated -with Mica - trachyte at the Horswell House quarry. Near Black- aterry Point. South of Frog- more and West of Winslade South of East Charleton. Southern Area. Altered Sedimentary Series— Mica schists, Quartz schists, Inter- bedded Mica schists and Quartz schists. Green schists, locally red, buff and grey. Essentially composed of hornblende, Green Schists (uralitic or actinolitic), albite (water or clear usually without trace of cleavage Hornblende or twinning) and epidote. Where altered Epidote Schists, by surface agencies, hornblende more or less replaced by chlorite and calcite generally present. Altered Basic Bocks. As there is rib doubt concerning the Devonian age of the Palaeozoic rocks of the Northern area, whilst as regards to age of the altered rocks of the Southern area much difference of opinion exists, priority of description has been given to the Northern area. The Devonian sediments consist of clays and sands with a variable amount of calcareous matter at certain horizons. Traces of marine fossils are found throughout the series, and the fineness of the sediments, the entire absence of current- bedding, ripple-mark, or other indications of shallow water, tend to show that the beds were deposited in a moderately deep sea. The sedimentation is such as would take place near the zone where sand shades into mud. These materials have undergone considerable changes through chemical action and lateral pressure. Chemical Changes. Owing to this cause it is often difficult to distinguish the original character of the rocks. This is especially the case with regard to calcareous bands and seams of igneous rock. Throughout the Devonian rocks of North and South Devon where limestones are associated in slates in thin seams or fossiliferous lenticles, there are also bands, lenticles or patches INTRODUCTION. 5 of brown, yellow, or red friable material, due to the complete or partial replacement of carbonate of lime by ferrous carbonate and the subsequent decomposition of the latter accompanied by the oxidation of the iron. Beds of several inches in thickness both in Culm limestones and in the Middle Devonian limestones have been thus replaced. In the Lower Devonian calcareous bands the preservation of organic structure, owing to the com- pression of the friable residue, is very partial and imperfect. In the case of igneous bands, the replacement of the original minerals by secondary products has been in many cases entirely obliterated by decomposition rendering the rocks worthless for penological investigation, and only distinguishable from decomposed clastic materials by practice in the field. The local distribution of colouring matter affords a frequent obstacle to the ready recognition of rocks, otherwise similar, in the field. It is impossible to say in regard to calcareous beds, to what extent silicification may have taken place, as examples of the partial replacement of dark crinoidal limestone by quartz are observable in the coast section near Tinsey Head. The silicific- ation of fossils is also occasionally met with in the Torcross Section, while at Long Sands, south of Berry Head, (in Sheet 350), a fragment of Favositid coral was found to be entirely replaced by quartz. In the Torcross section cubes of iron-pyrites appear to have locally replaced organisms, such as crinoid ossicles. Effects of Lateral Pressure. It might be thought that, notwithstanding chemical alteration, to ascertain the relations of the rocks of this district with its extensive coast and creek sections would be a comparatively easy task, and were it not for the effects of lateral pressure, presently to be considered, such would be the case. In pre-Permian and post-Carboniferous times the rocks of the Northern and Southern areas were subject to long continued pressure from a north and south terrestrial con- traction by which they were bent into a series of east and west folds. In the Northern Area no traces of an anterior movement exist, and if such are present in the Southern area they have hitherto escaped detection. The appearance of zonal distribution of the rocks in the Northern area is due to the general east and west strike consequent on this movement. The problem of the district is, how far we are entitled to regard these belts as the outcrops of successive groups. This question involves the relation of the Lower Devonian rocks on the south of the Dartmouth slates to the Lower Devonian rocks on the north of that group in areas outside that under consideration, and will be referred to further on. 6 GEOLOGY OF KINGSISR1DGE AND SALCOMBE. Structural Phenomena. So far reference has only been made to the general east and west strike, but the whole district is affected by constant small plications, often very sharp, which obscure the larger curves. The distribution of the grits between Beesands and Beacon Point, north of Hope, may be due to the dying out of individual folds, and the same is the case in other parts of the area where the absence of sufficiently marked lithological horizons does not allow of their being shown by boundary lines. The distribution of the Green Schists, however, shows the rapid troughing and nosing out of synclinal and anticlinal curves in a marked manner. In consequence of this irregular plication local strikes are often at variance with the general strike, and, where this is the case, their prevalent direction is north-easterly and south- westerly ; as for example by Collapit Creek, off the Kingsbridge Estuary, and in parts of the coast between Hope and Thurle- stone. From Challaborough to the mouth of the Yealm the strikes on the foreshore with few exceptions range from E. 15° N. and W. 15° S., to N.E. and S.W. Throughout the district north-westerly and south-easterly strikes are so local and exceptional that they may practically be ignored. The accommodation of the divergent strikes to the general east and west strike of the rocks may be shown thus. Fig 1. — Diagram explanatory of Strikes. The gnarling in the Devonian Rocks may be drawn from hand specimens, as in the accompanying examples from Tinsey Head. Fig. 2.— Gnarled fragments of tinsey Head beds. tn I fi^ rla 1 minate 4 ar |i llaceous and silty bands, actual size to fine cleavage is often present in the argillaceous bands A tendency INTRODUCTION. 7 In interlaminated hard grits and shales the planes of schistosity are generally bedding planes ; in the argillaceous rocks they are generally due to cleavage with which bedding accidentally coin- cides ; while in argillaceous rocks with beds of grit, the former are usually cleaved whilst the latter are partially so or not at all. Across finely cleaved dark slates, gnarled bedding is locally shown by thin ferruginous bands, lines of fossils, or other marked indications. Gnarling, or a tendency to it, is more especially noticeable in hard grit-shales or interlaminated ' beds, and it is frequently present near small faults, or displacements through contortion- strain. . Further examples of these structures with relation to bedding and cleavage are given in the illustrations from Collapit Creek and in the succeeding sketches (see p. 8.) The third example from Collapit Creek (Fig. 5) may be merely the accidental distribution of coloring matter, but taken in con- nection with the second example (Fig. 4) it suggests the gnarling of soft bedding planes of a more or less ferruginous character. In this connection it is necessary to point out the very frequent representation of bedding at intervals in the dark shales and slates of the Torcross and Beesands, and the Hope and Thurle- stone coasts (and the same is the case in the Plymouth Sound section) by brown bands of colour only, as well as by hard red or brown ferruginous films. Hard bands or very thin beds, probably resulting from decomposition and oxidation of limestone films, occur locally, separated by comparatively wide interspaces of softer argillaceous matter, which, when the hard bands are gnarled, often shows, as in the sketch (Fig. 6), a fine cleavage in the direction of the axes of the gnarling. In Bowcombe Creek the instance of sharp contortion figured (Fig. 7), occurs in buff and yellowish finely interlaminated shales. In this case the shearing of sharp axes seems to terminate the contortions, as if cut off by fault on either side, and cleavage may have taken place. In the Torcross Curve (Fig. 8), also figured by Professor Bonney,* we have a very pretty effect of the gnarling of thin bands of harder material occurring •.in the slates. These have been split up and drawn out vertically along the cleavage so as to form bands of lenticles. Had the harder bands been able to resist dismemberment they would have been gnarled. Where the rather harder pale-coloured, silty materials pre- dominate in interlaminated beds, the lines of bedding are otter so broken and distorted in the direction of cleavage as to be blended throughout with the darker argillaceous sediment, causing the rock to present a curiously brecciated appearance with no dominant lines of lamination. In the coast sections south of Torcross and north of Hope, lenticles with fossils occur sparsely along the planes of schist- osity, but, as will be seen from the last example (Fig. 8), unless these occur in continuous alignment, it by no means follows that * Quart. Journ. Geol. Soc, vol. xl., Plate I, p. 26. 8 GEOLOGY OF KINGSBEIDGE AND SALCOMBE. Fig. 3. Collapit Creel. Fig. 4. Collar* Creek, W. of Gerston Q. Q, Shales (not cleaved) gnarled in vicinity of cracks filled by quartz, qq. Fig. 5. Collapit Greek. Distribution of browo concentric ferruginous coloring matter in soft slates or shales. <> 1 Grey shales with hard brown bed- films gnarled fora distanceof 4 yards; cleavage in direction of axes of gnar- ling, not breaking- the ferruginous bands. Fig. 6. Beach Reef near Gerston Lime Kiln. Fig. 7. Bowcombe Creek W. of Kingsbridge Cemetery^ Contorted shales. Fig. 8. Torcross Cliff. 50"\ \rv \w Contorted slates with igneous rocks (i). 80-1 V INTRODUCTION. the bedding coincides with the schistosity. In fact, taking every- thing into account, the schistosity of the argillaceous rocks must be taken as cleavage, unless accompanied by distinct signs of bedding. Hence the evidences of bedding afforded by gnarled bands become important, and here two more factors may occa- sionally be added, namely, quartz veins and volcanic seams. Reference has been already made to the partial replacement of calcareous matter by quartz, and the seams and films of lime- stone seem to have been more or less siliceous (often no more than silt impregnated partially with lime through embedded organisms) so that the occurrence of gnarled quartz bed-seams is not to be wondered at, and they are in places, seemingly, the only Fig. 9. Near Torcross. Corrugated quartz bed-veins. signs of bedding. Between Torcross and Sunnyvale there are two or three instances of this. In the case figured (Fig. 9) dark grey slates, cleavage inclining N., at 70°, snow corrugated quartz-veins crossing it m exact parallelism to gnarled indications of bedding. In the coast section by the southernmost house in Beesands irregular quartz veins, render the bedding Fig. 10. Near Ayrmer Cove* W\tn,\U sufficiently distinct to show several minor contortions. Quartz veins often accentuate or mark disturbances, but corrugated bed veins must be carefully distinguished from the infillings of irregular strain cracks 10 GEOLOGY OF KINGSBKIDGE AND SALCOMBE. Volcanic rocks, when their origin is clear, can sometimes be taken as evidence of bedding. In the case figured (Fig. 10), the igneous band seems to be an indication of the bedding of the lilac slates in which it occurs, with a displacement along a sharp anticlinal axis where bedding and cleavage are coincident. In hard shales, calcareous shales, grit or silty shales, and inter- laminations of these materials, as in the examples from Tinsey Head (Fig. 2), cleavage, if existent, is rather an undeveloped fissile tendency, more apparent in the softer laminae which often exhibit strain slip cleavage. Rocks of similar character to those of Tinsey Head form part of a beach reef at Cbckridge Point, near the mouth of the Eiver Avon. The grit shales of Westcombe Beach are beautifully gnarled behind Mr. Mildmay's Tea -House. At x in the sketch (Fig. 11), there is a quartz vein, but no cleavage is Fig. 11. Behind Tea House, Westcombe Beach. visibly developed. In Collapit Creek in one spot, near a sheared porphyritic rock, yellowish shales with hard red bands show partial gnarling unaccompanied by cleavage (Fig. 12). Fig. 12. B Collapit Greek. By Bowcombe Creek, N.E. of High House, dark, apparently partially calcareous, shales with traces of crinoids exhibit gnarling (Fig. 13). Fig. 13. Bowcombe Creek. On the Eevelstoke Coast, the strata consist of hard grit-shales, grits and mud-stones, greatly contorted and often cleaved ; they are repeated by innumerable folds the sharp INTRODUCTION. 11 edges of the reefs being frequently due to junction of cleavage and bedding planes. On this Coast very good examples of the partial nature, and variable angles of inclination, of cleavage are afforded by the slates and mudstones with interbedded grit. From the foregoing examples, it would seem that gnarling is indicative of the resistance of bedding planes under certain con- ditions to cleavage fission, or rather to the lateral compression which, in the soft rocks of Torcross Point, has produced cleavage planes, in the interlaminated gnarled Tinsey Head beds, a more or less slight tendency to cleavage, and in the examples from Erme Mouth, and Collapit, and Bowcombe Creeks, has apparently failed in producing it. In the grey slate sections, where local hard bands or seams are present, gnarled bedding is so frequently seen crossing cleavage, that in the absence of recognizable mdica- tions of bedding, the planes of schistosity must be regarded as cleavage planes, and their varying inclinations as indicative of irregular curvature or disturbance in bedding. In a specimen of grit-shale with thin argillaceous interlamina- tion from Stoke Fleming (referred to by Mr. Teall)* a partial cleavage is displayed; the schistose tendency (or A usweischungs- klivage) being divergent in synclinal and convergent in anticlinal axes. If this could be regarded as a principle, fan- structure in cleavage would denote synclinal, inverted fan-struc- ture anticlinal, beading curves. As, however, in this type of rock, the softer and thinner laminae were compressed between harder and thicker layers, their behaviour can scarcely correspond to that of nearly homogeneous materials under like conditions of compression. In the latter, we would rather expect the inclination of the cleavage to be influenced by irregularities in the axes of curvature — vertical in normal curves, inclined in inverted curves, more or less horizontal where vertically zig- zagged bedding prevails. The schistosity dips in the Northern area are, with local exceptions, at high angles ranging from 60° to vertical. The distribution of northerly and southerly dips along three tra- verses from the metamorphic boundary northward to the margin of the maps is as follows : — 1. From Mouthwell (Hope). Planes dipping north and vertical for a distance of 15 chains. Vertical, alternating with both northerly and southerly dips, for 60 chains. Vertical, northerly, and north-north-westerly dips, for 70 chains. Vertical and southerly, with occasional northerly dips, for 80 chains. Vertical, south-easterly and south-south-easterly dips for 120 chains to map margin. 2. From junction with the metamorphic rocks in Salcombe Estuary northward. Vertical with northerly dips, some as low as 30°, for 45 chains, Vertical, with northerly and southerly dips, irregularly distributed, for 160 chains. Vertical, with southerly and occasional northerly dips for 175 chains. 3. From Bickerton Valley northward. Vertical and northerly dips for 50 chains. No section for 50 chains (some southerly dips inland in the Beeson grits). Vertical, with northerly and occasional southerly dips for 70 chains. Northerly dips for 10 chains. Vertical with southerly dips for 93 chains. * See Mem. Geol. Survey, Geology of the Country around Torquay-p. 22-23. 12 GEOLOGY OF KINGS BRIDGE AND SALCOMBE. The resolution of these dips into curves does not give a workable structure corresponding to the distribution of the rocks, which is not to be wondered at owing to the presence of numerous small faults and thrusts in the sections, breaches in continuity of sections through -talus, vegetation, &c, and more especially to the local difficulty in distinguishing bedding from cleavage planes. However, notwithstanding these detractions, the constant variations in inclinations of the planes of schistosity may be taken as indicative of the presence of numerous shallow repeating plications, which is further borne out by the distribution of the rocks. DISTRIBUTION OF THE ROCKS OF "THE NORTHERN AREA. 13 CHAPTEE II. DISTKIBUTION OF THE BOCKS OF THE NOETHEEN AEEA. The distribution of the rocks will now be treated in detail proceeding southward, as some of the types are either concealed, or not satisfactorily exposed, in the coast sections which will be described later. Dartmouth Slates. This group, -structurally the most important factor in the Northern area, forms the southerly continuation of the Kings- wear and Dartmouth slates. The continuity of the group with the Polperro variegated slates (with Pteraspis remains) has been traced from the Eevelstoke Coast westward. The rocks consist of hard glossy slates, hard silty shales (with surfaces spangled with flakes of white mica) , and cleaved mudstones or argillaceous slates (often striated by friction). Hard grit or quartzite beds are irregularly distributed in the slates. These are well shown on the Eevelstoke Coast, where they are repeated by numerous contortions, and on the Strete Coast, whence they have been traced as a horizon for some distance inland, as shown on the map. Near Keaton Cove, on the south coast, a rather coarse rock resembling arkose was noticed. The characteristic tints of the slates are a blending of pink and pale green, mottled purple and greenish,* and pale green, but their prevalent pale lilac-red, greyish lilac, or buff, tints in inland exposures often prevent their being satisfactorily separated from rocks of the Eingmore type. Near Strete (on the south) and at Merrifield and near Dittis- combe (on the west) grey and greenish slates have been separated by broken boundary lines, but as in the southern part of sheet 350 there is no proof that such separations have any absolute stratigraphical value. At the junction of the maps under description the Dartmouth slates disappear, apparently, in a series of anticlinal plications; the continuity of their eastern (Dartmouth) and western (Eevelstoke) masses in the area on the north (sheets 349 and 350) is by no means certain, being represented by a doubtful band about 10 chains in width in the south-east corner of sheet 349 to the south of East Allington. * Near Erme Mouth, pale purple slates, mottled with dark purple splotches and greenish tints, are typically shown in the cliffs south of Meadowsfoot Beach. 14 GEOLOGY OF KINGSBRIDGE AND SALOOMBE. The most southerly exposure of characteristic Dartmouth slates is at the bend in the river Avon near its mouth, between Sharpland and Cockridge points, where purple glossy slates are succeeded by reddish slates with glossy surfaces in places mottled with occasional vivid green patches, and containing hard grit beds similar to the grits with Pteraspis remains at Portwrinkle, fee, on the south coast of Cornwall. This identification was recently confirmed by the discovery of the remains of Pteraspis in the slates at Cockridge Point. A coccostean dermal plate, kindly identified by Dr. A. Smith Woodward, had been found in a shaly grit at the same locality, during the survey of the district. This ihlier seems to nose out beneath ■ the river sand seaward, but its easterly extension is extremely doubtful. It is even possible that it may embrace the grits at Clannacombe, and the long band of igneous rock that terminates at Worthy Farm. At about 12 chains north west of Clanna- combe, very hard grits with quartz-albite veins are exposed. Similar veins are also found by the Avon near Cockridge Point. The banks of the Avon, at from a quarter to a half mile north of Bantham, display lilac, red, and greyish slates with hard beds of grey, greenish, and reddish compact grit, locally much con- torted. These must be included in the Dartmouth slate inlier, which through a curve in the strike may extend northwards on the east bank to Stiddicombe Creek. On the border of Aunemouth Plantation a red, glossy, sheared rock about a foot in thickness recalls the sheared igneous rocks of Brookhill near Kingswear. Through the constant repetition by minor curves it must be borne in mind that there may be anticlines of Dartmouth slates which have escaped detection, through absence of distinctive characters, both on the coast and in inland localities. For the same reason the main boundaries of the group can hardly ever be regarded as definite and satisfactory. Traces of limestone are very rare in the Dartmouth slates, and consequently the lack of fossil evidence is not remarkable. At Middle Cove, on the Strete Coast, a thin band or lenticle of reddish limestone presented obscure traces of small fossils suggestive of fish spines, etc. In the same cliff line, under Sea Cliff Cottage and Grounds, some markings on the surface 01 a hard red gritty shale were thought by Dr. A. Smith Wood- ward to be fish remains. On the south of Pilchard Cove, bands of crushed red-brown friable material and decomposed cubes of pyrites suggest the former presence of calcareous films of organic structure. Between Fernycombe and Fernycombe Beach* brown films and impersistent red laminae, probably calcareous, are present in finely laminated grit or silty shales. Between Piskey's Cove and Sandy Cove lenticular, partly calcareous, bands occur in hard purplish and green shales. These bands are formed by the inosculation of greenish argilla- ceous material with pale pink organic limestone containing the DISTRIBUTION OF THE ROCKS OF THE NORTHERN AREA. 15 honeycombe layers of plates of Pteraspis cornubicus. Their identification was kindly confirmed by Dr. Smith Woodward. Similar remains are found in rocks of the same character on the south-east coast of Cornwall, at Portwrinkle, Downderry, and Seaton. The beach reefs of Westcliff Cove, near St. Peter's Church, Revelstoke, are on the strike of the Piskey's Cove beds, and contain red-brown friable laminaj suggestive of decomposed calcareous seams. Brown friable bands were also noticed in fnarled lilac and purplish mudstone with hard grit in a quarry on [iddle Down, south of Worsewell. Red-brown powdery films occur in hard grey and lilac quartz- veined slates in Red Cove, at the mouth of the Erme, but proximity to sheared igneous rocks renders their origin doubtful. A specimen from the cliffs south of Strete (No. 3081) is thus described by Mr. Teall, " Fine grained purplish grit with laminae of micaceous sandy shale." Under the microscope — "Largely composed of quartz grains, about 1 mm. in diameter ; felspar and a few large flakes of mica are also present. The quartz grains sometimes interlock and are sometimes separated by cryptocrystalline matter containing minute scales of a micaceous mineral." Igneous rocks are very occasionally met" with in the Dart- mouth slates of this area. These, though mostly intrusive, may be in part contemporaneous. Igneous rocks occur at Sheval Rock and near Pittaford (at Pollard's Combe) in the Strete district. Near Marwell, north of Ringmore, near Westcombe beach, south-west of Ringmore ; in the cliffs on either side of Erme Mouth ; at Carsewell Cove, near South Battisborough. In Beacon Hill quarry, and at Gara Point, segregation quartz- albite veins are often present in the Strete and Revelstoke coast sections. The junction between the Dartmouth slates and the rocks of the Meadfoot group, except when the distinctive characters of each are pronounced, is very vague. They are separable, without the discovery of fossils, only in a general way. Where traces of crinoids, brachiopods or corals occur the strata are not included in the Dartmouth slates, whether distinguish- able from them by colour or not. No lithological characters have yet been discovered which would enable one without the assistance of fossils or of colouring to say positively where the Dartmouth slates end and the over- lying types of the lower part of the- Meadfoot series begin. MEADFOOT GROUP. Ringmore Type. These rocks consist of grey, greenish, reddish and lilac slates, partly siliceous or silty, and exhibiting here and there the deep red colour, mottled with yellow and greenish yellow, which locally characterise the fossiliferous slates, shales, and grits of Looe. Where grits are in distinguishable proportion, and where the mottling above referred to is pronounced, they are 16 GEOLOGY 01* KINGSBEIDGE AND SALCOMBE. indicated by darker tints on the map. There are dark grey slates splotched with red (oligiste) amongst the rocks of this type, as between Goveton and Buckland Tout Saints, very similar to dark red-splotched slates at Looe 'between the Dartmouth slates and the reddish fossiliferous Looe beds. The type corresponds in character and fossils to the reddish fossiliferous Looe beds, but, although it seems by plicated repetition to have a considerable extension, it evidently consti- tutes but a small and irregularly developed part of the group, which, as is shewn in the western coast section, consists to a great extent of grey or reddish slate and interlaminated beds, not distinguishable from beds in the Torcross slates and the Tinsey Head beds. Seams of brown friable fossiliferous materialhave been en- countered in many places throughout the extension of the Kingmore type, but these residua after calcareous bands have been so crushed that it is exceptional to find generically identifi- able fossils in them. Actual limestone is seldom discoverable in inland localities, an impersistent bed or two has been detected at Holditch, near Bowringsleigh (west of Kingsbridge). Crinoidal limestone films and lenticles, often red-stained, are met with in the river Avon section opposite Aunemouth Plantation and between Cockridge and Sharpland Point, near the mouth of the river. Films of grey and red crinoidal limestone are of frequent occurrence in the slates of Ayrmer Cove. The fossiliferous films, where apparent, give the rock the appearance of shale, but this may be due to coincidences of cleavage and bedding in a slate group, and in some cases to the distortion of bedding along cleavage planes. In the following notes on fossiliferous localities in the Kingmore group, proceeding from east to west, the term shale is used always subject to this doubt. On the south side of Slapton Wood crushed traces of Spirifera and crinoids are met with in shales with irregular shaly grit. Between Slapton and • Pool hard reddish glossy slates contain crushed brown fossil casts, amongst which Baron von Keinach recognised Hexacrinus echinatus, Sandb. and a Rhynchonella. South of Slapton, near France Wood, traces of crinoids, &c. were noticed in occasional brown films in reddish (partly siliceous) shales. Near Lower Coltscombe, on the north-east and west, there are quarries in hard lilac-red shales and grit shales, interleaved with some crushed fossiliferous material, containing traces of crinoids. A large distorted impression resembling Streptorhynchus (Stro- phiodonta) gigas and a fragment of Pleurodictyum. By the road from Darnacombe to Harleston, at a point south-west of Lower Coltscombe, numerous crushed fossil impressions, including crinoids, Zaphrentis, and possibly Homalonotus and Pleurodictyum, occur in lilac-red shales. On the hill, facing the above on the south, lilac and red-brown shales, with shaly grits and red-brown friable matter, recall the section at Smugglers Cove, Torquay. Near this lilac shalv grits DISTRIBUTION OF THE ROCKS OF THE NORTHERN AREA. 17 with compact grit and red-brown arenaceous beds are exposed. Near Darnacombe, on tbe north, a rock, resembling a tuff with brown friable matter and cavities suggestive of dissolved fossils, was first recognised as a probable type of decomposed contemporaneojy>4grretJus rock. It is mentioned here because of references to the occurrence of similar rock in the following notes. At Harleston crushed fossil impressions, suggestive of Spirifera speciosa and Pleuroddcty'um, were met with in nearly vertical red-brown and pale lilac shales. At Stancombe cottages red and grey shales, dipping south, contain fossils. At Stancombe in crushed fossiliferous residues, in reddish shales dipping north, some casts resemble Spirifera primceva, and Baron von Eeinach recognised in others resem- blances to Strophomena and to Rhynchonella danenbergensis, Kayser. A quarry by the valley, north of Stancombe, exhibits grey papery shales with very fossiliferous crushed brown powdery films interlaminated. The dip is south at 70°, and the rock is quarried in large tabular blocks owing to the cohesion of the laminae. A similar rock is quarried at Combepark (about half a mile north-west of Ohillington) , where the brown films show crinoid stems, and probably traces of Homalonotus. North of Instart (west of Colehanger) hard bluish and pale greenish glossy shales (in places mottled red on surface) dip south at 65°, and contain numerous crushed impressions of fossils. The above localities are to be found in sheet 356 : the following are in sheet 355. South of Malston, near Sherford Down, on the east, lilac shales and grit shales contain brown fossiliferous matter. Near Malston Cross, on the west, lilac-red shale with arenaceous matter is associated with softer very fossiliferous lilac shale containing a lamellibranch, Zaphrentis, etc. : this locality is promising. South of Malston Mill hard grey fossiliferous shales are exposed in a quarry. In Buckland quarry hard grey shales, in places softer and fossiliferous, dip south at 70° to 80° ; they resemble the beds in the Sunnydale slate quarry. Between Buckland Quarry and Flear Mill a small quarry exposes fossiliferous lilac shales dipping south at 63°, cut off by grits intersected by quartz- albite veins, dipping south at 50°; the junction is either a fault or the fossil matter and crinoid traces have been flattened out along cleavage. Near the north margin of the map, north-east of Buckland Quarry, crushed fossils are visible in the wood in the west side of the valley west of Cross. At Bearscombe, east of Kingsbridge, soft red-brown fossiliferous beds occur in lilac shales and shaly grits and apparently contain Pterinea, Zaphrentis ami Crinoids. The shales and shaly grits south of Crimps Cross, near the above, resemble rocks of the Beeson type. North-west of Crimps Cross, north of Addlehole, hard pale lilac-red grit and slaty shales with crushed 7294. B 18 GEOLOGY OF KINGSBEIDGE AND SALCOMBE. red-brown fossiliferous powdery films are exposed in quarries giving southerly dips of 70° to 80°, and of 50° ; a felspathic grit was noticed ; and near this and east of Cross, albite appears to be present in the quartz veins. At Goveton, on the west, fossiliferous greenish shales are exposed, dipping northward. Near the footbridge across the stream, south of Goveton, traces of crinoids were observed in lilac, grey, and greenish slates. Just south of Higher Sigdon a small quarry exposes fossiliferous grey slates containing Zaphrentis. At about 14 chains South of Whitepost Cross very fossiliferous hard reddish shale was turned up in excavating a hole for a tele- graph post; amongst the crushed fossil casts were forms resembling Rhynchonella daleidensis and Pterinea. In the North of Kings- bridge, near Prospect House, red fossiliferous shales dip north at 60 p . North of Kingsbridge and west of Croft, pale greenish- grey, very fossiliferous shales, of exactly the same type as the fossiliferous shales of Ringmore; are exposed in the Railway Cutting ; they contain crushed and distorted examples of Spirifera primoeva and other brachiopods and of Zaphrentis. At Holditch, near Bowringsleigh, west of Kingsbridge, lilac and red shales and grit-shales contain shaly red crinoidal lime- stone and calcareous lenticles, mostly decomposed to red-brown powder, with casts of much-distorted brachiopods. In Bowring- sleigh Wood, near Pier's Cottage, some very fossiliferous lilac-red glossy shales contain abundant traces of crushed and broken fossils, amongst which a part of the flattened cast of a valve of a large brachiopod (probably a Spirifera) was found ; judging from the size of the impression, the vertical depth from the hinge must have been about an inch and a half. In a quarry between Norton and Kingsbridge Workhouse, beds of hard lilac grit, from 6 in. to 18 in. thick, are exposed in lilac shales. North of Sorley, east of Rake, near the northern margin of the map (355), the surface stones of lilac brown and grey sandy grit and shale resemble types of Staddon grit. East-nof th-east of Churchstow (near the head of the Lordswood valley) south of Leigh, lilac shimmering shales are interlamin- ated with red- brown fossiliferous friable seams showing numerous distorted casts of brachiopods, including Spirifera, and of crinoids, etc. On the opposite side of the valley, in lilac shales with irregular shaly grit, red-brown fossiliferous friable matter is also present. On the slopes of the next valley on the west, between Churchstow and Hatch, there are several exposures of fossiliferous shales ; one of these on the west side of the valley, at about a quarter of a mile from Hatch, shows hard grey- lilac shales and grit-shales with red-brown friable material and numerous crushed casts of brachiopods, including Spirifera speeioda, Pterinea spinosa, and traces of crinoids and of Fmestella ? South -of this, on the same side of the valley, brown and red-brown grit fragments are found on the surface, and contain Plewodicfcyum probl&maticum, Streptorhynchus persarrnentosus ? and a Favositid coral {Siriatopora ? ). At Merrifield, near the bead of the next valley on the west, brown DISTRIBUTION OF THE ROCKS OF THE NORTHERN AREA. 1 9 crushed fossil films are visible in hard lilac and grey shales and grit shales. A quarry by the valley between Nuckwell and Laver's Tenement exposes hard grey shales with red (oligiste) mottled surfaces and with brown fossilferous interfilmings; similar shales are exposed in a quarry by the high road to Aveton Gifford, between Merrifield and Creacombe ; fossils in either case being crushed, apparently beyond generic recognition. By the alluvium of the Avon near South Efford, very fossili- ferous red-brown shales contain Spirifera, Zaphrentis, and Streptorhynchus. In a quarry near this, very fossiliferous greenish-grey shales are associated with lilac and grey shales and shaly grit, and contain Spirifera, Zaphrentis ? and crinoids. South of the above, near Lower Stadbury, near Reynold's Park, and between Whitley and Bantham, the surface evidence, in places, justifies the supposition that grit beds of Beeson and Warberry types are locally associated with the shales and grit- shales. This association is more apparent on the north and west of Clannacombe. Quartz-albite veins occur, in beds presumably belonging to this type, on the west bank of the River Avon at 30 chains above Stiddicombe creek, and in a quarry east of Thurlestone. The banks of the Avon from Stokes Hill to Bantham afford inter- esting exposures, in which are encountered grey and lilac shales and grit-shales exhibiting gnarling in places, occasionally with beds of friable grit of Beeson types, occasionally with hard compact grits; with thin seams of sheared and decomposed volcanic rocks in places (of the Darnacombe type), very rarely exceeding two feet in thickness. The shales contain powdery fossiliferous seams in several places and are identical in charac- ter with the fossiliferous shales of Coltscombe, etc. Following the west bank of the river : — At 15 chains down stream from Stokes Hill, grey (oligiste) red- mottled shales occur ; at from 30 to 34 chains lower down Spiri- fera is recognisable among the crushed fossils in the brown films which, at from 10 to 15 chains further south, in places still retain a calcareous character. At 35 chains down stream from the last mentioned, grey shales with lenticles ot crinoidal limestone are visible near lilac shales mottled with green patches, and further on grey red (oligiste) mottled shales and lilac red shales. At about 13 chains from the limestone lenticles, red crinoidal lime stone films with Zaphrentis occur in reddish shales with crushed fossils. From this point southward, and round the bend in the river, the exposures chiefly consist of lilac and grey shales with compact grit beds, much contorted, which probably belong to the Dartmouth slates, being in strike with the rocks of that series at Cockridge Point. The bend in the coast, south of Mount Folly, nearly coincides with the strike of red and purplish slates with red shaly crinoi- dal limestones containing Zaphrentis, and a red-brown igneous rock with fretted surface. This limestone,- judging by interven- ing dips, which show strike curvature, seems to be on the 7294. » 2 20 GEOLOGY OF KINGSBRIDGE AND SALCOMBE. horizon of the red crinoidal limestone films with Zaphrentis previously mentioned. The coast round Sharpland Point to Challaborough (including Borough Island) consists of lilac-red and dark grey slates, which are distributed in a manner suggesting much plicated repetition. There are several points of interest on the east bank of the Eiver Avon above Bantham ; lilac-red fossiliferous shales are visible by Stadbury Wood and by a wood further up, being probably the same band, prolonged from the opposite side of the river ; friable grit beds are evidenced in their vicinity Lilac and grey shales mterlaminated with gritty material show gnarling a few chains north of Stiddicombe Creek. At the mouth of Stiddicombe Creek there are green sheared igneous rocks in dark grey and lilac slates or shales, also a dark green-grey rock (fine tuff? ) with numerous impressions of Bhynchonella (R. daleidensis ?) The crinoidal limestone films on the opposite side were not observed on this, side of the river, but it is quite possible that, if persistent, their continuation may be concealed by the alluvium of Stiddi- combe Creek Between Bigbury and Ringmore fossiliferous grey shales are exposed in a quarry near the junction of streams; one of these flows southward from Houghton, a little way up it fossili- ferous greyish shales containing Zaphrentis are exposed. In Ringmore, opposite the church, a quarry at the road junction shews 2 feet of rock of the Darnacombe type in pale greenish- grey and lilac glossy shales and grey shales with red (oligiste) mottling. Numerous casts of fossils were found, including various flattened impressions resembling Spirifera primceva, Streptorhynchus gigas, or Orthis hipparionyx (Orthotetes personata) ; also Zaphrentis, a branching coral (Striatopora ?) and a large eye of trilobite, pronounced by Dr. H. Woodward to be apparently "an eye of Phacops latifrons or of Ph. Fernandi of enormous size." Between Ringmore and the coast, gnarled fossiliferous grey and lilac shales and grit shales are exposed near Manor Farm, and near Challaborough Coastguard Station lilac shaly grit and hard shales with brown fossiliferous bands. The coast section from Westcombe Beach to Challaborough commences in Dartmouth slates succeeded by dark shales con- taining thin, harder interbanding in places and very similar in many respects to those of Torcross and Tinsey Head. The dark grey shales contain crinoidal limestone films, andareassociated with lilac green-mottled and hardened shales with red limestone films, containing Zaphrentis and crinoids. In one place glossy grey papery shales with decomposed fossil films of the Ringmore type are visible. They may be in a sharp syncline or intercalated in the shales with crinoidal films. The number of inland localities in which traces of brachiopods are found contrasts with the extreme paucity of such remains on the coast where colour changes from red to grey are frequent and on the whole of very little if any stratigraphical significance. Between Ayrmer Cove and Sedge- well Cove, the rocks seem to be for the most part plicated DISTRIBUTION OF THE ROCKS OF THE NORTHERN AREA. 21 repetitions of those of the Torcross type. Large blocks of red crinoidal limestone have been used for gate posts at half a mile south of Bigbury ; they were no doubt obtained from the neighbourhood. Bocks of the Darnacombe type occur here and there throughout the area occupied by the Bingmore type as mentioned in the foregoing notes; but no volcanic horizon can be traced. There may be many more seams of igneous rock than can be detected by surface evidence. The beds of the Kingmore type are represented at Scabbacombe Sands and on the south of Crabrock Point in the coast section in sheet 350. Torcross Type. These rocks consist of dark grey slates weathering pale grey and brown, but north of a line between Stokenham and Thurle- stone Cottage they are locally red. Interlaminated silty films, lenticles, and occasional beds and bands of hard grit occur in them. Numerous thin bands of igneous rock, generally much decom- posed are associated with these slates. Although in many cases the igneous rocks seem to be contemporaneous, from impersistence or plication, this cannot be proved, nor can they be persistently traced beyond the limits indicated on the map. The beds of this type form a belt more than a mile in breadth, running from Torcross Cliffs to Thurlestone and Yarmouth sands. It is uncer- tain whether a series of interlaminated argillaceous and silty beds (or grit-shales) with occasional bands of grey or brown grit, which are in plicated association with grey slates on the north of Chillington, Winslade, Charleton, Park Farm, Baston, Auton, South Milton, and Thurlestone Sands, belong to this belt or to the Bingmore type. Zaphrentis and crinoids occur here and there in the slates of the Torcross type. Small organisms about the size and shape of canary seed have been found in the slates near Sunnyvale (Sunnydale). Orthoceras was found in a quarry north-west of Molescombe. Near the mouth of the Avon, north of Long Stone, in beach reefs a large Plewrodictywm was observed in grey shales with a northerly schistosity dip of 85°. From the west side of Ayrmer Cove to the mouth of the Biver Avon the coast section consists chiefly of beds of the Torcross type which may belong to this zone, in plicated association with beds of the Bingmore type. Igneous rocks similar to those in the Torcross section occur in eight or nine places. From the Avon mouth to Thurlestone Sands the Torcross type is well exposed in the cliffs and, with the exception "of the promontories on the north side of Long Stone and at the north end of Broad Sand and of a narrow band near Warren Point (on the north side of Thurlestone Sands) where rocks referable to the Bingmore or Beeson types occur, everywhere prevails. Igneous rocks have been noticed in six places on the north of the Long Stone promontory and in about twelve places further south. _ None of the igneous bands are sufficiently thick to be traced inland, but their recurrence in beds identical in character with those 22 GEOLOGY OF KINGSBBIDGE AND SALOOMBE- in the Torcross cliffs points to a constant repetition of volcanic horizons by plication. The Torcross slates do not appear to contain igneous rocks in the vicinity of the Beeson grits, none having been encountered south of Thurlestone sands on the west, and south of Sunnyvale Quarry on the east, coast, or in the intervening strip of country by North Pool, Ham Point, Blanksmill Bridge, and South Huish. In a specimen (No. 3078) of lead-coloured slate with traces of fossils, from the Torcross coast, Mr. Teall detected under the microscope — "Idiomorphic rhombs of a ferruginous carbonate, scales of mica, chlorite, quartz, &c." The rocks of the Torcross type are well represented in the coast section in sheet 350 in many places between Scabbacombe Sands and Southdown Cliff. Beeson Grits. To the south of the rocks of Torcross type beds of quartzose and compact grit occur in very variable proportion in the slates. Where possible the developments of grit have been shown on the map, in the series of patches between Beesands and BeaconPoint,but the boundaries of these patches are necessarily indefinite. The hard beds of grit occurring in the cliffs south of Beesands village, south of Huckham, and in the Southpool Creek section are not included in them, although they belong to the group, as it is impossible to trace them in the slate area inland where there is a dearth of exposures. The name of the group is taken from the village of Beeson, where the grits were first encountered in mapping the district. Both in their distribution and variable characters these grits have their counterparts in the Looe district, and exhibit types also met with in the Warberry (Staddon) beds of Torquay. The obvious irregularity of their occurrence as shown by the patches on the map suggests synclinal or anticlinal structures entailing the correlation of the Torcross slates with those south of Tinsey Head. On the other hand the same appearance ' would be produced by the plicated outcrop of a more or less impersistent series of grits in the slates, in which case the Tinsey Head beds would be above, and the Torcross group below them. The greatest development of the grits occurs in the Beeson and Malborough (About a half-mile north of Malborough between Ilton Castle and Beacon Point) patches, and across these there are very few exposures. Proceeding from east to west, the character of the rocks may be gathered from the following notes. By a wood on south-west of Beeson irregular lilac and buff shaly grits are exposed in quarries, presumably forming an anticline near their junction with the grey slates on the north. South of Huckham a quarry shows a syncline in thick-bedded lilac and grey grits. DISTRIBUTION OF THE ROCKS OF THE NORTHERN AREA. 23 Near this, on the opposite side of the intervening valley, there is a quarry in grey shales with compact grey grit irregularly contorted. Crinoids and obscure brachiopods have been found in the lilac red-speckled grits of the-Beeson patches. North-west of Ford there are two quarries on opposite sides of a valley. The eastern quarry consists of slates or shales with irregular compact grey grit. The western quarry is to the north of the strike of the beds in the eastern, it shows pale buff brittle grits and brown shaly grit, or sandstone, with numerous crushed and distorted casts of brachiopods, &c, amongst which Chcmetes is found. On the hill north of South Pool yellow and whitey-brown, brittle, irregular, more or less shaly grits occur, occasionally containing crushed and distorted casts of brachiopods and crinoidal joints. In the Southpool Creek section, whitish grit beds, irregular lenticular grit bands and seams, and beds of compact grit partly replaced by quartz, occur here and there in plicated association with grey slates. West of Harwell House, the grey slates of the Wallpark coast are associated with beds of brown brittle or friable grit and red shaly grit, containing in places casts of crushed brachiopods, amongst which Spirifera was recognisable, and crinoid remains. Small cubes of pyrites were noticed lining one of the fossil cavities. On the north coast of Harwell Wood a shaly grit in grey slates furnished a crushed organism, (Orthoceras '?). For about twelve chains south of Hal well Point shaly grits are very occasionally fwesent in the slates. At a mile south from Charleton Church a ight brown decomposed rock, with impersistent glossy shale films, contains brown friable material in cavities, in which crinoids seemed to be the only recognisable organisms. The rock may have been originally a fossiliferous or calcareous grit. The Ilton Castle Plantation grit patch, shown on the map, consists of grey slates associated with whitish, yellowish or brown grit-shales and shaly grits. The brown grits occasionally contain obscure traces of fossils, and present curiously corrugated surfaces, possibly of organic origin. A similar phenomenon is presented t>y beds of grit in the slates on the Thurlestone coast. Near Lincombe a grey grit band, eighteen inches thick, occurs in the grey slates on the south of the patch. In the red grits north of Malborough fossils have, been detected — near Alston, crushed Plev/rodictya (?) ; near Rugwell, cast of part of a branching coral and traces of crinoids. In their prolongation to Beacon Point the grits are represented by friable or brittle yellow-brown grit bands, similar to those constituting the patches shown on the map between the Malborough and Beeson developments. Casts of brachiopods, including Chcmetes, and crinoids were found between Beacon Point and Galmpton. Igneous rocks are not present in the area occupied by the grits, or in the slates on the south of them. 24 GEOLOGY OF KINGSBKIDGE AND SALCOMBE. Petrological Notes by Mr. Teall : — A specimen (No. 3082) from the hill on the north of South Pool. Fine- grained grit traversed by a thin quartz vein, resembling specimen No. 3081 from the cliffs south of Strete, but with less ferric oxide and other impurity. A specimen (No. 3083) from Beacon Point. Fine-grained purplish grit, composed of quartz, mica, felspar (?) and red patches of ferric oxide, some- times showing rhombic outlines, and thus clearly arising from the decompo- sition and disintegration of ferriferous carbonates. Mr. Hunt* notes : — A slice of the Beesands grit (4 chains south of Beesands) contains microscopic tourmaline, of which one minute crystal has been dislocated and recemented by quartz, thus indicating both a severe squeeze and the incipient formation of secondary minerals in the grits. In a slice of fine grit north of Tinsey Head a minute grain of tourmaline is slightly, albeit distinctly, affected by solution, with indications of the recrystallisation of secondary tourmaline. In view of the uncertainty as to the position of the Beeson grits, viz., whether an outcrop above the beds of the Torcross type or an anticlinal prolongation of the rocks of the Ringmore type, it would be idle to correlate them with any particular horizon in the coast section in Sheet 350. They do not resemble the Long Sands grits, respecting the position of which in the series a similar doubt exists ; yet they must be represented either in Southdown Cliffs or to the south of them. The same uncer- tainty must naturally attach to the rocks now to be described, the Tinsey Head slate series. Tinsey Head Slate Series. The Tinsey Head slate series begins with the dying out of the grits of the Beeson group southward. Its northern boundary is therefore very vague, but its southern limit is well defined by the brown rocks which border the Hornblende Epidote Schists, and is more or less easily determinable west of the Kingsbridge' Estuary, where the slates are in contact with Mica Schists. The provisional name applied to this series involves the inclusion with the slates of the interlaminated argillaceous calcareous and siliceous beds of Tinsey Head,' which in the Report of the visit paid by the Geologists' Association to the area (Easter, 1901) were grouped with the Beeson beds. There being no hard and fast boundaries between the zonal types, and as numerous faults and thrusts occur which cannot be traced from the coast sections inland, the inclusion of these beds with the slates and not with the grits is adopted, here for convenience. The interbanded, argillaceous and siliceous rocks of Tinsey Head contain seams of crinoidal limestone. Although they seem to be separated from the grits. in the slates near Beesands, by slates which (as far as imperfect exposure enables one to judge) do not contain grit beds, the Beesands grit bands are associated with inter- laminated beds similar to those of Tinsey Head, and at Beacon and Woolman Points (north of Hope) similar beds accompany Proc. Geologists? Assoc, Vol. xvii., Part III., p. 127, 3rd July, 1901. DISTRIBUTION OF THE ROCKS OF THE NORTHERN AREA. 25 the Beeson grit horizon. Again, interlaminated beds occur near Cockridge Point by the River Avon, and are associated with slates and hard grit beds which, though included in the Ringmore group, are comparable to the grit beds near Beesands, and give colour to the hypothesis that the Beeson grits may be anticlinal appearances of part of the Ringmore beds. The relation of these interlaminated beds to the slates on the south seems to be subordinate, so that it appears as if we had a generally descending sequence from the metamorphic boundary to the Beeson grits. To the south of Tinsey Head and Beacon Point the coast sections display dark slates with occasional hard lenticles and thin red or brown bands. The latter occur at intervals as if indicative of bedding, and also in veins ; they may be due to the decomposi- tion and oxidation of calcareous material. Very occasionally beds of dark, partly crinoidal, limestone are met with south of Tinsey Head, and a similar occurrence was noted at Beacon and Woolman Points. Rather small Zaphrentid corals are often visible in the hard lenticles in the slates. In the Hope and Beacon Point coast section veins and bands, or beds, of a hard whitish carbonate are met with, and in the section south of Tinsey Head hard carbonates occur in the brown, beds of Greenstraight (north side of the Bickerton Valley). In the Southpool and Kingsbridge Creeks the slates are softer and much weathered, a trace of crinoidal limestone occurs in them near Tosnos Point, and Zaphrentis has been found in the South- pool section. In point of character these slates are comparable to those in the Torcross group, but no igneous rocks have been found in them, and this seems to be the main argument against their correlation with the slates of the Torcross section. The following notes on specimens from Tinsey Head, Beacon Point, and Woolman Point are by Mr. Teall : — No. 3079. Tinsey Head.— Fragment of- a much puckered rock which must originally nave consisted of interlaminationa- _ of fine-grained arenaceous and micaceous shale, Slight effervescence with dilute hydro- chloric acid. Under the microscope, the gritty bands are composed of minute angular grains of quartz or quartz and felspar, carbonate, minute scales of sericitic mica, and larger flakes of clastic mica. The intervening laminae are much richer in mica, and often show a very perfect strain-slip- cleavage (Ausweischunqsklivage) which is not present in the more arenaceous parts. The mechanical structures in this rock are precisely similar to those seen in the quartzose mica schists. No. 3095. Beacon Point. — A hard grey fine-grained rock which effervesces slightly with weak acids. Under the microscope composed of quartz, carbonates and mica with zircon, rutile and tourmaline as unimportant accessories. The principal interest of this rock lies in its structure. The individuals of quartz and carbonate are authimorphic and probably in a great measure authigenic, so that the rock has many of the characteristics of a crystalline schist. There can be no doubt that it was originally a fine-grained calcareous sandstone as the accessory minerals are those characteristic of such a rock. The present structure of the rock is, however, largely the result of metamorphic changes since it was formed. The rock is traversed by narrow veins of quartz and carbonate. No. 3096. Woolman Point. — A brecciated quartz-carbonate rock. 26 GEOLOGY OF KINGSBRIDGE AND SALGOMBE. CHAPTEB III. NOBTHEBN ABE A— continued. The Igneous Bocks of the Nobtheen Aeea. The absence of igneous rocks in the Beeson grit and Tinsey Head slate groups renders it improbable that the numerous bands in the Torcross group as well as those locally present in the Bingmore group and in the Dartmouth slates east of Bevelstoke are all intrusive pipes, dykes, or sills . Unfortunately many typical varieties such as the Darnacombe tuff (?), and the sheared igneous bands by Collapit Creek, south of Gerston (which may be decomposed representatives of rocks in the Torcross section), are too decomposed for penological examina- tion, so that the microscope neither confutes nor confirms the similarities and differences recognised by the eye. In the actual exposures the conformity or discordance of the igneous bands with the schistosity of the slates affords- no proof of their real nature, as the stresses to which the rocks have been subject; would naturally produce different effects on different materials. Having regard to the behaviour of the igneous rocks of Stoke Fleming, etc., in the Dartmouth slates, and those of Old Mill Creek, etc., in the Meadfoot beds, north of Dart- mouth,* it seems reasonable to regard the igneous rocks of the area as evidences of intermittent vulcanicity represented by tuffs and lavas as well as by pipes and sills. The ophitic dolerite of the quarry west of Winslade (slide No. 1785) and the massive rock in East Charleton Quarry (slide No. 1786) may be intrusive ; but decomposed rocks resembling tuffs and lavas occur in their vicinity. A strip of igneous rock (thought by Mr. Teall to be an altered syenite) is shown on the map following a south-westerly direction from Worthy Farmf (between Churcnstow and Thurles- tone) along the dominant planes of schistosity of lilac slates, or shales, with occasional traces of grit. It varies from 1 to 5 feet in thickness, and may be a sill (slide No. 1869). Between Challaborough and Ayrmer Cove (slide No. 1866)and south of Leas Foot Sand (slide No. 1862) on the Thurlestone coast there are one or two bands mainly composed of carbonates. The original character of these bands is doubtful ; they may have been cal- careous sediments or calcareous tuffs. At Stiddicombe Creek Lime Kiln, in the Kiver Avon section, a rock resembling a schistose * Geology of the country around Torquay. Mem. Geol. Survey, v. 38-41. t A trace of igneous rock in line with this band was observed at Bantham Cross, near Churchstow, but there is no evidence for connecting them. NORTHERN AREA. 27 Juartzo-felspathic grit (slide No. 1874) may be of volcanic origin, n the Dartmouth slate area a decomposed igneous rock is visible at Sheval Rock, near Strete, and another small patch, ?robably of the same character, occurs due west of it at Pollards lombe, near Pittaford. Both these rocks are amygdaloidal, and may form anticlinal axes of contemporaneous volcanic material. The numerous bands and patches of igneous rock at and near the mouth of the River Erme belong to the same general group as the igneous focks of Stoke Fleming and of the Kingswear coast, and are met with more or less frequently along a line of country, presumbly corresponding to the general strike, between the mouth of the Erme and Stoke Fleming. The hard diabase of Muxham Point is associated with amygdaloidal rock (slide 1845) which seems to overlie it, and gives place irregularly to breccia or sheared tuff. These are intercalated with grey slates for a thickness of from 20 to 30 feet. St. Anchorites rock is an aphanite which, with the mass at Cars well Cove, may be intrusive. At Carsewell Cove a band of felsitic rock occurs in lilac slates with quartz-albite veins, and also two parallel bands of diabase. Near this, on either side of Blackaterry Point, felsite is exposed ; it forms a mass extending north from Windham Beach, and is exposed in a quarry on the summit, whence it extends irregularly eastward, apparently ramifying through rocks in the partly masked cliffs above Ryder's Hole. This intrusive mass has not been examined petrologically ; it may have a subterranean connection with the spherulitic grano- phyre of Whympston House, both are indicated as a continuous Elvan on the old geological map, Sheet 24. At Gara Point, opposite the Mewstone, a mass of green sheared igneous rock occurs in the Dartmouth Slates. At Aveton Gifford an igneous rock, too insignificant to show on the one-inch scale, exhibits some points in common with the crystalline schists (slide No. 1788). Traces of igneous rock have been met with, in the Torcross and Ringmore groups and occasionally in the Dartmouth slates, which could not be indicated on the one-inch map, and of those shown the majority are necessarily exaggerated, being from a few inches to a foot or two in thickness. Mica trachyte is exposed in a quarry on the west of Horswell House, south-east of Thurlestone. On the west side of the quarry, near its northern end, a rubbly brecciated rock, appar- ently New Red, rests on mica-trachyte. On the opposite side of the quarry red-stained Devonian slates make a very irregular junction with decomposed rock, resembling an agglomerate, in which mica-trachyte is distinguishable. In the inner or southern face of the quarry the mica-trachyte is associated irregularly with decomposed rock which may be a quartz-porphyry. On the visit of the Geologists' Association to the quarry, Easter, 1901, Mr. F. P. Mennell collected specimens which he submitted to Mr. F. Chapman, who considered the felsitic rock to be a rhyolite (possibly a tuff). Mr. Mennell also informs me that "fragments occurring in "the agglomerate show, the highly characteristic 28 GEOLOGY OF KINGSBRIDGE AND SALCOMBE. cloudy banding of the acid lavas." He considers from their petrographic characters that, if not lavas, the rocks "consolidated very near the surface." From the north and south trend of this small patch of igneous rock, and from the absence of any signs of interbedding with the Devonian, there seems to be little doubt that it is intrusive in that formation, and its occur-, rence in touch with a New Red outlier is suggestive of an eruption in Permian times. The rock may be the upper part of a neck or pipe from which the effusive materials have been denuded, if it is not the relic of these materials in part. A fault letting down New Red on mica-trachyte (lava ?) on the west, against the mdange of mica-trachyte, rhyolite and agglomerate (considered as the upper part of the material in the vent ?) is highly probable. The Torcross igneous type, described in the following peno- logical notes by Mr. Teall, does not indicate any special horizon, as similar rocks are seen in association with the Eifelian slates in the coast section south of Plymouth. No. 1777. Shore south of Torcross. — A pale greenish grey fine-grained, slightly schistose rock. Small brownish lenticles project from the smooth surface. The main mass is a confused aggregate of carbonates, chlorite and minute scales of what is in all probability a white mica ; colourless quartz and felspar occur in small grains. Scattered through the above are large irregular plates of opaque leucoxene-pseudomorphs after ilmenite or titano-magnetite. The rock is probably a modified diabase, but the structure and composition of the main mass has been entirely changed. 1 have often remarked that the plates of leucoxene remain comparatively unchanged in form after the structure and composition of the rock in which they occur have been com- pletely altered. 1785. Quarry west of Winslade. — A dark green, medium-grained massiye rock. Ophitic augite more or less changed to chlorites, lath-shaped sections of felspar (decomposed) carbonates and iron-ores partly changed to leucoxene. A typical ophitic diabase. 1786. Quarry south of East Charleton. — Dark greenish massive rock containing some large and more or less idiomorphic felspars as porphyritic constituents. Viewed under the microscope similar to the last in mineralogical composition Diabase-porphyrite. 1869. Fear Buckland Park (and Worthy Farm).— Decomposed light grey rock containing pinkish spots. Chlorite after biotite in a matrix of felspar taiore or less decomposed. White mica in small scales probably secondary. Apparently an altered syenite. 1862. South of Leas Foot Sand, Thurlestone.— Pale green slaty rock studded with minute crystals of pyrite. Mainly composed of carbonates which do not effervesce freely. 1866. Between Challaborough and Ayrmer Cove. Pale grey slaty rock. Microscopically similar to the above, with the addition of small opaque grains and streaks which appear light brown by reflected light, and thus remind one of leucoxene, pseudomorphs after pyrite). 1872. River Avon, south of the Stakes.— Microscopically composed of felspar, chlorite, iron-ores. The structure of this rock is not like that of the normal diabase. 1868. Stiddicombe Creek, River Avon.— Pale greenish grey schistose rock with lenticles and streaks which probably represent porphyritic felspars. Composed of broken felspars still showing twin striation, chlorite, iron-ores changed to leucoxene. The cataclastic phenomena as seen in the felspars are very striking. Diabase (sheared). 1874. Stiddicombe Creek Lime Kiln.— Similar to a rock in Holbeton Wood, by River Erme, but more slaty and finer in grain. Detached grains NORTHERN AREA. 29 of quartz and felspar in a cryptocrystalline matrix largely composed of sericitic mica. Much brown staining. Slight signs of a transverse cleavage. Probably but not certainly of sedimentary origin. 1844. Near Muxham Point, Erme Mouth. — Greenish grey, fine-grained amygdaloidal rock. Amygdaloids of yellowish green colour due to epidote. Ground-mass composed of lath-shaped felspars, chlorite, epidote : amygdaloids of epidote, quartz and chlorite. Fluxion structure well marked. Diabase. 1845. Same locality. — Grey rock slightly schistose. Large plates of ophitic augite, lath-shaped felspars often crushed and altered, chlorite, leucoxene. Cataclastic structure strongly marked. 1788. Aveton Gifford. — Pale green rock, slightly schistose. Quartz-felspar aggregate with chlorite and colourless or pale green mica in ragged scales. Under a low power the rock resembles a fine-grained grit, but with higher magnification the outlines of the individuals forming the quartz felspar aggregate are seen to be extremely irregular and to resemble the outlines of the grains in the corresponding aggregates of the crystalline schists. Specimens from quartz-albite veins in fine grained felspathic grit-shales in a quarry east of Thurlestone and from the coast at the mouth of the Avon toward Borough Island (where the albite is pink) have been examined. The albite has been determined by Szabo's method and by the examination of flakes produced by crushing. The Coast Sections of the Northern Area. In the foregoing notes reference has frequently been made to the coast sections. We will now confine our attention to them, without going into the consideration of the nature of the boundary of the metamorphic rocks, or touching on the coast sections of the Southern area, both of which subjects will be specially referred to further on. The coast section in Sheet 356 from Strete Gate northward, consists, where unmasked by Head, of Dartmouth slates — pale-red or lilac-red glossy slates with hard grey silty intercalation, and reddish slates of a less glossy appearance, with beds of hard grit or quartzite. The grits show contortions in places, and the planes of schistosity in the slates seem to be more or less coincident with bedding, an appearance largely due no doubt to the snapping of the axes of sharp plications, producing a sliding between the limbs of the folds and obliterating the appearance of reduplication in the beds, through nearly parallel rearrangement. Flakes of mica are most prevalent on the surface of the silty bands and grits. Between Sheval Rock and Pilchard Cove the slates are greenish or grey, an amygdaloidal igneous rock, the amygdules filled with carbonate of lime, seems to form an anticline at the Waterfall, as already mentioned. Although search was made in likely places remains of Pteraspis were not found. From Strete Gate to Torcross there is practically no section, the low cliff banks of Slapton Ley display reddish slates and silty beds of the Ringmore type, passing into dark grey slates of the Torcross type near Stokeley Farm. At the commencement of the cliffs near the Torcross Hotel, compact grey grit beds occur in dark grey slates, partly inter- laminated with hard silty bands ; bands of greenish igneous rock 30 GEOLOGY OF K[NGSBRIDGE AND SALCtMBE. are also present. Igneous rock overlies the slates in the section figured by "Professor Bonney, the junction being more or less parallel with the bedding shown by the alignment of hard lenticles, distorted by cleavage, in the slates. (See Fig. 8, p. 8.). On passing the lawn of a house by the beach, veins of quartz dipping at 60° cross the schistosity which dips north at 80°. Further on bedding is indicated by bands of brown colour, as is the case in the Eifelian slates of the Plymouth coast near their junction with the Staddon grits. The brown banding in one place with a northerly dip of 45° crosses cleavage planes dipping north at 70°. Near this, cleavage, at a low angle, traverses a couple of brown argillo-arenaceous beds separating them into nearly horizontal flakes, and recalling the brown beds on the south side of Bovisand Bay in the Plymouth section. Approaching the, headland with volcanic rocks (Dun Point) we encounter irregular developments of quartz in places, probably denoting local disturbances in the slates, which give northerly cleavage dips of from 50° to 80°, the bedding being only locally shown to cross them. Silty, partially calcareous, lenticles and interbandings of pale grey contrast with the dark argillaceous materials in places, being especially noticeable near the larger mass of igneous rock in Dun Point. Crinoids and Zaphrentis are occasionally discovered in the slates and hard lenticles. The corrugated quartz bed- veins already referred to (Fig. 9) occur at this part of the section. From this to the end of the cliff the dark grey slates, so well exposed in Sunnydale (or Sunnyvale) Quarry, prevail. From Sunnydale to Beesands there is no section for about 55 chains. At Beesands, behind the houses, the cliff begins in red mate- rials washed from the Beeson grits above. Absence of exposures leaves us in doubt whether the Beeson grits terminate as shown on the map near the summit of their features, or strike to the shore. The section recommences immediately south of Beesands in a cliff of grey slates with irregular beds of brown compact grit or quartzite showing much disturbance. These are associated with interlaminated beds, the silty bands being more frequent and pronounced than in the same type in the Torcross section. For about 70 yards, slates with ferruginous bands seem to prevail in the receding cliff, which is mostly overgrown. Near Tinsey Head, these become interbanded with silty sediments. On the north side of Tinsey Head, there is a fault crack filled with black crushed rock. The interlaminated Tinsey Head beds are beautifully puckered in places. Films of siliceous limestone with numerous joints of large crinoids occur in them in several places. On the north side of Tinsey Head, the silty interlaminations are thicker than in the Headland and are associated with beds of grit. In the headland, the thinner interlamination is very distinct through alternations of darker and lighter tints. Interlaminated beds in decreasing amount continue for a chain south from the Head, being succeeded by dark slates with much quartz, very irregularly distributed in NORTHERN AREA. 31 places, and with thin ferruginous bands at intervals. The inter- laminated beds reappear in a small projection of the cliff at 5 chains from Tinsey Head, and as they appear to be confined to the lower part of the cliff, there can be little doubt that they form an anticline. From this point for a quarter of a mile southward the section consists of dark slates, with cleav- age planes inclining north at high angles, and occasional thin ferruginous gnarled bedding seams. In three places, between 5 and 12 chains, south of Tinsey Head traces of impersistent beds, or films of crinoidal limestone, were noticed. Calcareous lenticles also occur in one of these spots, and crushed Zaphrentid corals are obtainable. The brown beds of Greenstraight will be mentioned in a subse- quent chapter. This coast section is disappointing, as it affords no clue as to the relations of the Dartmouth slates to the Meadfoot beds (of Kingmore type), and is even more unsatisfactory as regards the relations of the Beeson grits. Apart from the absence of igneous rocks between Beesands and Greenstraight, we have no proof of the occurrence of hard grits on the north of the Beeson grits, in the slates corresponding to those just south of Beesands ; and this, having regard to the impersistence of similar grits to those of Beeson in the Looe area, renders the correlation of the Tinsey Head slates with the Torcross slates very doubtful ; whilst at the same time there can be no question that they form parts of the same series. The general nature of the sediments in the Dartmouth slates and overlying strata is so similar in character, that apart from the discovery of fossils, and the localization of general colouring distinctions, it is often extremely difficult to effect even the most general separation between them, consequently we are led to attach perhaps undue importance to the Beeson grit developments, and to the frequent association of igneous rocks in the Torcross slates. The chief interest in the coast section in sheet 355 attaches to the eastern part between Hoist Point and Hope. To the west of Hoist Point the salient features have already been referred to, and no proofs have been obtained as to the troughing in of higher beds than the Dartmouth slates. (Quartz-albite veins occur in the greenish and reddish glossy Dartmouth slates of Hoist Point and Westcombe Beach (at the mouth _ ot the adjacent valley on the east). The even planes of schistosity are bant at intervals in parts of the section. Proceeding toward Ayrmer Cove from Westcombe Beach, the reddish glossy Dart- mouth slates give place to dark grey slates beautifully inter- banded with sifty material, in places, and containing occasional beds of hard grit. Although there are some signs of disturbance in the dark grey beds at their junction with the Dartmouth slates it is not certain whether the colour change is accompanied by fault, as further on the dark slates become in part pale red in colour. We next encounter an interbanded series resem- bling that of Tinsey Head, and containing some hard grit bands comparable to those on the south of Beesands. The dark grey 32 GEOLOGY OF KINGSBRIDGE AND SALCOMBE. colour of this series changes to red; it contains many films of crinoidal limestone with large crinoid ossicles, and irregular white organic patches (Monticuliporoid ?) which point to an absolute correlation with the limestone of Looe and Millendrea.th, of the Plymouth section south of Crownhill Bay, and of the coast section in Sheet 350 south of Crabrock Point and at the south end of Scabbacombe Sands. In this series, apparently, on the strike of the fossiliferous beds of Bingmore Quarry, a small thickness of red and dull greenish papery shales, interleaved with decomposed friable fossiliferous matter containing brachio- pods and Zaphrentid corals, is either intercalated or synclinally folded. Farther on the red crinoidal series is traversed by a fault, beyond which dark grey slates containing Zaphrentis and occasional films of crinoidal limestone come on. These resemble the slates of Crownhill Bay in the Plymouth section, and like them contain organic structures resembling Plewodictya. Bands of greenish igneous rock, with vesicles filled with calc- spar, occur in them as in the Torcross type. In the reefs of dark slate on the west side of Ayrmer Cove, not far from crinoidal and Zaphrentid remains, Mr. Brook-Fox discovered traces of Pteraspis in one of the hard dark patches, which cccur here, as in the slates on either side of the Long Sands grit in the coast section in Sheet 350 * It is remarkable that this resemblance should be further accentuated by the discovery of Pteraspis on the north side of the Long Sands grit. In both cases the probability of faults must be admitted. A north and south fault is visible near the northern angle of Ayrmer Cove. On either side of Ayrmer Cove the "dark grey slates exhibit mottling or variegation Avith straw coloured tints, a phenomenon often locally prevalent in the lower beds of the Meadfoot series, but still more pronounced in the variegated (Falmouth) slates of Southern Cornwall.! On the east of Ayrmer Cove hard dark (structureless) nodules or lenticles are met with in the slates, and gnarled quartz veins perhaps denoting bedding are observable. On rounding Toby's Point the dark grey gives place irregularly to reddish tints— the hard grey lenticles being still more pro- nounced — traces of structure resembling Fucoids were noticed here. Further on dark grey slates, paruy associated with silty interlaminations, form the' base of the cliff and beach reefs, the red colouring making a kind of anticline over them. We next encounter bands of greenish igneous rock in dark slates with hard lenticles giving place to red slates with occasional changes to grey, containing hard lenticles which very occasionally resemble Orthoceras. The projecting headland south of Toby's Point is chiefly composed of hard reddish silty shales, with a tendency to gnarling and apparently in an anticlinal undulation. Bands of brown-weathered igneous rock occur here, also films of red crinoidal limestone, and the colour changes once more to grey. The suggestion that these beds * Geology of the Country around Torquay.— Mem. Geol. Survey, pp. 16-18. •J Summary of Progress of the Geological Survey, &c, for 1898, p. 98. NOETIIEItN A1SEA. s 33 are an anticlinal reappearance of the rocks with crinoidal films between Westcombe Beach and Ayrmer Cove is forced upon one. Approaching Challaborough the rocks are for the most part identical with those of Torcross, having siliceous inter- bandings here and there, and containing several pale buff or greenish impersistent bands of igneous rock, sometimes much sheared and never exceeding three feet in thickness — many of the seemingly separate bands may be repetitions of the same horizons by plication and shiftings through small faults, or strain slides. In colour the section differs from Torcross, as grey beds in beach reefs strike into red beds in the cliff, and, besides this, there are several irregular changes throughout, from dark grey to red. Traces of fossils are mostly replaced by pyrites, which is developed in brown interlaminations near Challaborough. Veins of quartz and carbonate are also met with. A north-west and south-east fault, marked by a channel in the beach reefs, crosses the promontory at the north end of Challaborough Sands, where dark grey slates mottled with red (oligiste) splotches are cut off against lilac-red slates with gritty bands of the Ringmore type. From Challaborough to Sedgewell, near Borough Island, the rocks are very much the same as those at and near Toby's Point ; — Dark grey slates with greenish variegation, which is sometimes the prevalent tint, and red slates, with hard grey len- ticular patches and fucoid-like markings. The distribution of the colours is irregular, the red being in some cases distributed at the base of the cliffs and running up in irregular anticlines into the grey. Some thin igneous seams occur in grey slates near the beginning of the cliff ; near this, for 15 yards, red slates present an anticlinal appearance, then for 24 yards grey slates prevail, and lenticular patches resembling Ortkoceras were noticed ; then, after passing 34 yards in red slate, we encounter bands of sheared igneous rock, partly porphyritic, at a small promontory ; beyond this the upward change from red to grey takes place on each side of a fault. Dark grey and greenish mottled slates form Warren Point. On rounding the point red slates reappear in the beach reefs and pass up into dark grey greenish mottled slates in the cliffs. These colour changes are quite irrespective of lithological character. A band of decom- posed igneous rock is visible in red slates under the Fish Cellars. Borough Island is composed of dark grey, and grey and greenish variegated slates, with a band of reddish slates crossing the isthmus which connects it with Little Island, and a mass of similar colour near its south-eastern end. Igneous bands have been noted in two places, and also signs of fossils. Between Sedgewell Cove and Sharpland Point, in part of the section the slates are curiously mottled. The mottling is accom- panied by a bending at intervals in the planes of schistosity, But as these indications do not interfere with the strike of brown decomposed fossiliferous seams the bedding is more or less coin- 7294. C 34 GEOLOGY OF KINGSBRIDGE AND SALCOMBE. cident with the schistosity. These rocks are mottled purple-red in parts, otherwise grey. They belong to the Ringmore type, and exhibit the coherent lamination which, in rocks of that type near Kingsbridge and north of Chillington, permits of its being dressed and used as building stone. Beyond Sharpland Point we encounter grit beds in the slates, and an igneous rock with a weather-fretted surface. The slates also exhibit inter- laminations of silty material and red limestone films, with the large crinoid ossicles and white irregular organic patches (Monti- cnhporoid ?), already noted between Westcombe Beach and Ayrmer Cove. As Dartmouth slates come on between this and Cockridge Point, as mentioned in tbe notes on that group, we have here a repetition of the crinoidal beds west of Ayrmer Cove, which are also represented by the crinoidal limestone bands noticed in the River Avon section. The anticline of Dartmouth slates does not appear to extend across the river mouth, unless certain red rocks at the Old Ferry, which may equally be regarded as belonging to the Ringmore group, are included in it. At Ham's End reddish slates occur, but elsewhere on Bantham Sand, in the reefs and cliff, the slates are dark grey or pale greenish and grey, occasionally mottled with red (oligiste) splotches. A fault m this section may be prolonged by the Old Ferry across the bend in the river in an east-north-east direction. This may have some effect in terminating the Dartmouth slate anticline on the east. From the south end of Bantham Sands to the Long Stone Promontory slates, generally dark grey in the beach reefs and often variegated buff and greenish in the cliffs, present similar characteristics to those of the Torcross section. A large Plewro- dictyum was found in them near the low-water mouth of the river, and even silty interlaminations, occasional hard grit beds and decomposed bands and films of crinoidal limestone are present near the promontory. Masses of grey, pale buff- weathered vesicular igneous rock occur at intervals, also thin green shaly bands, apparently of igneous origin. Long Stone Promontory is composed of red slates, beds of quartz-veined grit and gritty or silty slates, apparently cut off on the north by a fault of no great magnitude. On the south side of the promontory an impersistent band of yellow-green purple-mottled igneous rock occurs in the red beds, and near this the red. colour passes irregularly into grey. The slates contain silty interlaminations as on the north of the headland. Grey and green igneous bands, dying out upwards along the planes of schistosity, occur in them. Farther on, the strikes of grey rocks in the cliff and shoreward reefs are continued by red beds of identical character in the seaward reefs. Then, for a chain, the red tints prevail in the cliff. At the north end of Broad Sand igneous rock occurs in the red beds. Farther south the grey beds of the cliff, which contain bands of igneous rock as in the Torcross section, become locally red, or mottled with red tints, in the reefs. One of these bands of igneous rock occurs in grey slates with lenticular silty interlamination. Crinoid stems and traces of brachiopods, Spirifera (?), are visible on some of their surfaces. NORTHERN AREA. 35 Between Thurlestone Sands and Leas Foot Sand, the rocks which support the New Red outlier consist of grey slates with silty interlaminations, more or less lenticular, and of brownish banded slates with greenish igneous rock three feet in thickness. These are associated with beds or bands of breccia or coarse volcanic tuff. Farther on the slates change in colour from grey to pale red, or lilac-red banded with deep red tints : in texture they resemble the Long Sands grit on the coast south of Brix- ham, but are soft and not arenaceous. The reefs south of Thurlestone Sand are composed of similar rather soft slates, buff or pale lilac-red in colour, with greyish banding; one or two bands of peroxidated igneous rock were found in them. Softish slates of this kind are locally prevalent in the Torcross type as near Gerston, by Collapit Creek off the Kingsbridge Estuary. There can be no doubt that the rocks from Avon Mouth to Thurlestone Cottage (at the south end of Thurlestone Sands) mainly belong to the Torcross type, and are on the same stratigra- phical horizon, and that the same type, if not the same series, is repeated by plication from Avon Mouth northward between Fish Cellars and Challaborough, and at Ayrmer Cove and Toby's Point. From Thurlestone Cottage to Beacon Point the rocks are chiefly dark grey slates (reddish near the commencement of the cliff) with hard lenticles and occasional traces of foss\\s,Zaphrentis, &c. At Great Ledge a colour change takes place to red and buff tints ; brown films and hard silty lenticles are present in dark grey slates between this and Beacon Point. Beacon Point consists of hard red and buff silty slates associated with grey slates with brown films, hard lenticles, and occasional fossils and with some thick beds of grit. The junction of these materials with the grey slates on the north is irregular and very obscure ; on the south, they are overlain by interlam- inated beds of the Tinsey Head type, and evidently on the same stratigraphical horizon as the rocks immediately south of Beesands. The distortion of the siliceous lenticles in the direc- tion of cleavage is well shown here, and a rather thick bed of very siliceous crinoidal limestone gives a plicated southerly dip bringing the interlaminated beds under a series of dark grey slates which form the cliffs between Beacon and Woolman Points ; the Tinsey Head beds being brought up at their base between faults in one place near the former. In these grey slates there are hard lenticles with ossicles of crinoids and Zaphrentid corals which, although tilted with the cleavage, in their general distribution show that the bedding crosses it. At Woolman Point the silty slates and Tinsey Head beds asain appear, probably brought up by fault. Here, as at Beacon Point, there are bands of hard carbonate* and many quartz veins along the strike of the schistosity and across it. Between this and Mouthwell at Hope Cove, the beach reefs consist of dark grey slates with much quartz, in strings and pellets, and thin hard red * Pelleks and lenticles of hard carbonate also occur in the section between Ayrmer Cove and Challaborough. 7-294. c 2 36 GEOLOGY OF KINGSBRIDGE AND SALCOMBE bands at intervals. In the cliffs, as is often the case, the slates are of a pale grey colour ; gnarling is locally shown by thin harder bands, and occasionally there are traces of limestone films. The reefs terminate at Mouthwell with yellowish-buff hard silty slates resembling those of Beacon Point in texture. From this point southward the section will be described in discussing the boundary between the rocks of the Northern and Southern areas. The section from Beacon Point to Mouthwell is the westerly strike prolongation of that between Beesands and Greenstraight, and in both cases the Tinsey Head interlaminated beds are to all appearance below the grey slates which separate them from the doubtful junction rocks of Hope Headland and Greenstraight, so that, notwithstanding minor curves and disturbances, there is a wonderful uniformity of strike. There is also no doubt as to the Beeson grit horizon being represented at Beacon Point at least partially. Farther north to Thurlestone Cottage, we have a series of slates devoid of igneous rocks which are the westerly con- tinuation of the slates of Sunnydale Quarry, in which no igneous rocks have been found. As the Torcross beds with igneous rocks are repeated over and over again by plication between Thurlestone and the mouth of the Avon and probably still farther north to Ayrmer Cove, the nature of their junction with the Beeson grit horizons is the vital question as regards the structure of the area. If the beds of Beacon Point form an anticlinal, then the Beeson grits are a series of anticlines, probably repeated at Long Stone and cropping out in the rocks of Ringmore type, thus bringing the Tinsey Head series into direct relation with the Torcross group. This view accords with the evident constant repetition of the beds, it is diagramatically represented in the Proceedings of the Geologists' Association.* The only permissible alternative hypothesis is to regard the Beeson grits as a more or less impersistent development underlying the Tinsey Head beds and overlying the Torcross beds, which toward their base blend with the rocks of the Ring- more type.f This view is the safer one to adopt, owing to the impersistence of grit developments in the Meadfoot Beds and their local attenuation in the Staddon group. Neither of these alternatives affects the statement that whether they form four general stratigraphical horizons or two, the rocks of the Northern area which overlie the Dartmouth slates, are parts of the Mead- foot series and correspond to the types met with in that series in the Looe, Plymouth, and Long Sands (South of Brixham) Coast Sections. In either case there has been much repetition through plication, but the first hypothesis postulates a lesser thickness for the strata than the second alternative which is here adopted. "■ Proc. Geologists' Assoc, Vol. xvii., Part 3, p. 124, t See Diagram, Fig. 27, p. 62. SOUTHERN AREA. 37 CHAPTER IV. SOUTHERN AREA. The rocks of this area present two well-marked stratigraphical groups — the Green schists and the Mica-schists. Of these groups the Green schists are, as far as can be inferred from numerous coast and creek sections., the uppermost. The nature of the evidence on this point may be gathered from the examples selected for illustrations (Figs. 14, 15, 16, 17, p. 38.) from Salcombe Estuary, Southpool Creek, and the southern coast of Start Promontory. Similar evidence is afforded by the cliff between Outer and Inner Hope. In "Spirit of the Ocean Cove" (Fig. 17), an indentation ot Foxhole Cove on the south coast of the Start Promontory, a dark green, red (calc) veined, variety of the Hornblende-Epidote schists apparently passing by intercalation into Mica-schists forms an isolated mass in the cliff, being evidently the axis of an inverted syncline. This is borne out by the axial shape of tumbled blocks. The Mica-schists undoubtedly represent a series of sediments of the same general nature as those found in the Devonian area ; and the Green schists may unhesitatingly be regarded as an altered series of basic igneous rocks allied to the diabases in composition and possibly consisting in part of altered tuffs. The rocks of the Mica-schist group occupy the greater portion of the district. They form the coast line of the Start and Bolt districts, and the cliff sections offer splendid opportunities for the study of the series. East of the Salcombe estuary the Mica-schist area is bounded on the north and on the south by two well marked zones of Green schist which gradually approach each other as they are followed to the westward and finally coalesce to the south of Malborough. The southern band forms the Prawle Coast, and may also be studied in the cliff sections for a mile from the mouth of the Salcombe estuary eastward. The northern band is well ex- posed in the Southpool, Kingsbridge, and Batson Creek sections.* The single zone formed by the coalescence of these bands runs westward to the Bolt Tail, being bounded on both sides by Mica- schists. From the above it will be apparent that from Bolt Tail to a point south of Malborough the Green schists form a syn- cline, and that to the east of this point their separation in two bands is due to anticlinal structure in the intervening Mica- schists. The compound nature of these larger structures and the * The further splitting up of the northern band which, owing to irregular interplication with Mica-schists and the absence of exposures west of Batson Creek, cannot be satisfactorily shown on the map, is described in the Chapter on the Relations between the rocks of the Northern and Southern area. 38 GEOLOGY OV KINGSBRIDGE AND SALCOMBE. intense folding to which the metamorphic rocks have been sub- jected is brought out in a striking manner by the detailed mapping of the Green schists. Fig. 14. — South of Biddlehead Point, Salcombe Estuary. Horizontal scale, 1 inch. = 60 yds. The dark portions represent the Mica-schists. Fig. 15. — Coves under the Moult. The Mica-schists (m) seem to be partly interbedded with igneous bands. The altered igneous rocks (h) are of green and brown colours. Fig. 16. — Southpool Creek, Bast of Westerncombe m. Mica-schists with brown ferruginous films at mi. lFiG.l7.—"Spiritofthe Ocean' Cove, near Foxhole Cove. 'th.- Head. x\\h\\\^ Hariblende-EjaddoteSehistB, ilfflll M ca- Sdksbs -witliquartz. Index to Figs. 14 to 17. From this point of view the part of the map between the Moult and Malborough is of special interest. The general SOUTHERN AREA. 39 strike of the Green schist zone is here north-west and south- east, but it is broken up into a number of boat-shaped or lenti- cular patches in consequence of minor folds, the axes of which run in a general east and west direction. Similar phenomena occur in the Northern Area, but the com- paratively slight and impersistent evidences of vulcanicity and the indefinite nature of the boundaries of the grit patches, through association with shales or slates, do not allow of their being traced with anything like the same certainty and minute- ness in the mapping. As far as our observations extend, the rocks of the Southern Area have been affected by the same set of movements as those experienced by the rocks of the Northern Area, and exhibit no traces of pre-Devonian disturbance.* The effect of these move- ments has been precisely the same in kind in both areas, but very different in degree. This is shown by the sharpness of the minor curves as well as by the much greater frequency and intensity of the gnarling. As in the corresponding interlaminated beds and grit shales of the Northern Area, the interlaminations of mica-schist with quartz- schist afford the best examples of gnarling in the Southern Area. Fig. 18 is from a photograph of interlaminated mica-schist from the Bolt Head, taken by Mr. A. R. Hunt, and reproduced by his Fig. 18. Mica-quartz-schist of Bolt Head. *Mr. Marcel Bertrand has shown that movements in the same direction may take place at widely different epochs in the same area, the subsequent movement following the lines of the previous one, so the fact of coincidences in the strike and folding are not sufficient in themselves to prove the absence of pre-Devonian movements. 40 GfcoLOGY OF KlNGSBftlDGE AN1) SALCOMEE. kind permission as it affords a beautiful illustration of gnarling. In gnarled quartz-schist with thin interlaminations of mica- schist the softer material has been sheared out oyer the facets of fractures produced by contortion strain, so as to give to specimens (such as those here figured) the appearance of a surrounding Fig. 19.— Hall Sands Coast, south of Hard Stone. Specimens of gnarled quartz-mica-schist. Natural size. film of mica-schist across the laminae of quartz-schist. This tendency to irregular fracture, accompanied by miniature slicken- sides, makes it very difficult to obtain good hand specimens showing intense gnarling. As a rule the gnarling can seldom be satisfactorily traced by the eye, through the nipping out of distinctive bands and through the irregularities consequent on intense pressure. Fig. 20. — From the cliff on north side of Hall Sands village. Quartz-schist, forming a small anticlinal axis. Natural size. In intervals of less continuous pressure, or where the quartz- schists have been able to resist it, the gnarling, as in the accompanying sketch (Fig. 20) of a specimen from Hall Sands Cliff, has been comparatively slight. This miuute puckering is not universal in the area, there are places in the mica-schist and quartz-schist districts in which it -seems not to have taken place. SOUTflEEN AM. 41 Near Southdown Farm in the Bolt district mica-schists and quartz-schists present nearly horizontal bed surfaces in places, and on the whole the major divisional planes have Fig. 21. — Near Southdown Farm. Plication in quartz-schists. J of natural size. rather the appearance of gentle undulations than of sharp plications. This structure, no doubt, explains the Tor-like weathering of the crags in the vicinity, just as in the Valley of Eocks in North Devon, where the low dips of the Lynton Beds account for the castellated forms of the Tors. In the mica and quartz-schists it is, however, very probable that these planes are lines of fracture along axes of horizontal plication and indicative of vertical rather than horizontal bed- ding on the whole. In the Bolt and Start districts the axes of the gnarling and minor curves give strikes ranging from east and west to E. 20° N., andW. 20." S., but north-east and south-west strikes are occasionally met with. Strain-slip cleavage is shown to have been developed in the mica-schists ; but apart from this there are no reliable indica- tions of the development of true cleavage, although cracks along the axes of contortion of the minor folds which are well shown at Sharpitor, Bolt Head, causing the rock to weather in great prisms, or joint-bounded masses, may be evidences of a rude cleavage, or strain-slip cleavage on a large scale. These cleavage joints represent the directions of the folding of the gnarled bedding, and are inclined, vertical, or horizontal, ac- cording to the position of the axes of the folds, and in rough parallelism to them. Gnarling is frequently shown in the Green schists where the alignment of mineral constituents renders it visible. The parallel planes by which the Green schists are intersected give the rock an even-bedded appear- ance, but the folia denoting bedding are seen to be gnarled vertically where the planes are nearly horizontal and vice-versa. These planes are strictly analogous to the planes intersecting the Middle Devonian slates and mudstones in East Cornwall. Quartz-albite veins occur here and there in the Mica-quartz- schists throughout the district. 42 GEOLOGY Off KiNGsBRlbGE Atffi SALCJOMBE. Literature on the Relations of the Rocks of the Northern and Southern Areas. Owing to the long standing uncertainty regarding the relative ages of the older rocks of the Northern and Southern areas, it will be convenient here to refer to the opinions of previous writers on this vexed question, preparatory to a general description of the rocks, followed by a somewhat detailed description of the nature of the boundary as investi- gated during the geological survey of the district in the year 1891. De la Beche approaches this subject with characteristic caution. On the ground of the absence of any apparent passage of the one series into the other and in view of the com- paratively marked line of separation between them ; . after disposing of the theory that a part of the same group should be subjected to conditions under superincumbent pressure suffi- cient to render them crystalline without the other part " having been subjected to like conditions ; " he concludes : " In the absence of any contradictory evidence, it would appear fair to infer that these rocks belong to that class which is more ancient than the grauwacke series; and though the mica and chloritic slates of the district may be associated with rocks possessing an arenaceous aspect, even supposing they may be detrital, that they form no part of that series which adjoins them on the north."* v From this it would appear that De la Beche advocated the pre-Devonian age of the rocks of the Southern area ; yet on the old Geological Survey map the boundary is a faint dotted line, and the rocks of the Southern area under the symbol K are shown as an altered part of the series to the north. Sedgwick and Murchison in 1840 speak of the rocks of the Southern area as " a crystalline group of micaceous and chloritic slate, unlike any other formation in North or South Devon probably of metamorphic structure."f The boundary between the rocks of the Northern and Southern areas on the old geological maps is drawn at the south of Beesands village so far north as to include a considerable part of the Devonian section on the east of the Kingsbridge Estuary, whilst from the Kingsbridge Estuary to Hope it is shown with sufficient accuracy to lead one to infer that it was traced on the ground in the latter case and conjectured in the former, unless, as Mr. HuntJ suggests, the error arose in process of transference. In harmony with this suggestion, it may be pointed out that a discrepancy exists between the published geological map sheets 23 and 24. On sheet 24 the boundary is shown by a continuous line from Hope to the southern end of South Pool village, but at this point two continuous lines are * Keport on the Geology of Devon, Cornwall, etc., pp. 35, 36. 1839- t Trans. Geol. Soc, Ser. 2, Vol. v., p. 661. I Paper on Examination of evidence in support of Archaean age, etc., printed by Stephen Austen and Sons, Hertford, 1893. SOUTHERN AftEA. 43 shown, the northernmost running out of the normal direction and continued by the dotted boundary in sheet 23, whilst the southern line is not prolonged into that sheet.* In 1868, Dr. H. B. Holrj- regarded the slates leading up to the metamorphic rocks as " high up in the Devonian system, ' com- menting on the absence of clear evidence of fault between them and the mica slate and chloritic rocks. He considered that the effectsof metamorphismincreasedsouthward.and that, though the metamorphic rocks may have been thrust up with granite, more probably they are altered Devonian rocks mantling over a granite mass. From this it would appear that Holl considered the alteration due to contact metamorphism in which the rocks of the Northern area did not share being outside the contact zone. The following notes are by Mr. Teall : The next paper we have to notice is one by Prof. Bonney " On the Geology of the South Devon Coast from Torcross to Hope Cove." J In this paper the microscope was for the first time applied to the solution of the vexed question as to the relation of the metamorphic series to the comparatively unaltered Devonian rocks. After describing the general characters of the coast section from Torcross to Hope by the Start Point, Prawle, Bolt Head and Bolt Tail, the author concludes '(1) that the foli- ated and slaty series are quite distinct rocks, the one much older than the other, so that very possibly the foliation of the former was anterior to the deposition or the latter, and their junction is a faulted one ; (2) that both have subsequently been subjected to tremendous lateral pressure which has bent them into huge folds, cleaving the newer and producing marked effects on the older rock." He then describes the microscopic characters of the " foliated " and " slaty" series, and especially emphasizes the fact that the later movements have affected the authigenic minerals, of which the former are composed ; he maintains that these movements in the Southern area have operated upon rocks already in the condition of crystalline schists, whereas in the Northern area they have operated upon ordinary sedimentary deposits. Prof. Bonney further describes certain curious structures pro- duced by pressure in the Devonian rocks of Torcross, and concludes with some interesting speculations as to the relations of the rocks in question to a vamshed mountain range of the Alpine type which may formerly have occupied the position of the English Channel. Three years after Professor Bonney's paper there appeared in the same journal another communication on the meta- morphic rocks of South Devon, by Miss Baisin.§ * If it were prolonged in the direction shown after bifurcation it would touch the coast at 30 chains south of the dotted line and 22 chains north of the Bickerton Valley. + Quart. Journ. Geol. Soc., Vol. xxiv., pp. 438, 439. J Quart Journ. Geol. Soc., Vol xl., pp. 1 26. § Notes on the metamorphic rocks of South Devon, by Catherine Raisin, Quart. Journ. Geol. Soc. Vol. xliii., pp. 715-733. 44 GEOLOGY OF KttfGSBRlDGE AND SALCOMBE. The first part of this paper deals with the boundary of the slaty and metamorphic series. At Hope the junction is placed just north of the small rocky headland. " While the mass of the headland is of mica-schist, fragments of what are evidently beds of the slaty series are found coating its northernmost portion. They agree with that series in direction and amount of dip ; and lithologically are very like beds occurring at a junction near Souf;h Pool, to be described immediately. They consist of impure limestone, very much indurated and cry- stalline in character, and of brittle bands of blackish material having a very crushed and slickensided look." The junction section by Southpool Creek is then described. The junction at Hall Sands is admitted by the authoress to be somewhat difficult to fix. She says : " Any simulations of pbyllite south of Hall Sands, and any simulations of schist north of Hall Sands, are in each case local and very inextensive; a small flake of the southern cliffs might be mistaken for phyllite, and a veined fragment of the northern rocks might be thought approximating to schist, but any longer examination, even in the field, would show the distinctness." Both Miss Raisin and Prof. Bonney attribute the phyllitic aspect of some of the mica-schists near the junction to the intense post- Devonian earth movement to which the original mica-schist has been subjected. The authoress summarizes her observations along the junction line (p. 719). The petrographical part of the paper contains descriptions ol the microscopic characters of the chloritic series, of the mica- schists and of a group of rocks intermediate between the two for which the term micaceo-chloritic i s proposed . The concluding part of the paper contains a description of the stratigraphical relations of some of the metamorphic rocks. In the same year as that in which Miss Raisin's paper appeared, Mr. Somervail commenced a series of communications "On metamorphic and associated rocks of the extreme south of Devonshire."* In these communications he strongly opposes the view that the two series are separated by a fault or, in general, by a sharp line of demarcation. He calls attention to the similarity in composition of the rocks on both sides of the junction line, and maintains that we have here a case of progressive metamorphism. In the course of his investigations, Mr. Somervail discovered an outcrop of the Green schists on the north side of the Bickerton valley, and thus proved that if the two groups are separated by a fault, it cannot be in the position assigned to it by Miss Raisin. Professor Bonney briefly noticed Mr. Somervail's papers, but declined to consider his arguments in detail, as they did not take into consideration the evidence furnished by the microscope. * Trans. Devon. Assoc. Part I., Vol. xix. (1887) pp. 419-437 ; Part II. Vol. xx. (1888), pp. 215-224 ; Part III., Vol. xxi. pp. 452-459. SOUTHERN AREA. 45 In 1892, Mr. A. R Hunt entered the lists with an important communication "On certain affinities between the Devonian rocks of South Devon and the metamorphic schists."* As a result of microscopic examination of a number of slices of different types of rock from the Devonian and metamorphic areas he snows that there are many points of resemblance, and therefore greatly strengthens one of the arguments employed by Mr. Somervail. It may perhaps be as well to state here that the work of the Survey has tended still further to emphasize this point, which would probably not be disputed by those who maintain the dis- tinctness of the two series. Whatever the age of the rocks in the metamorphic area may be, it is practically certain that they represent a series of arenaceous and argillaceous deposits with which a considerable amount of basic igneous material was associated, and that, so far as original composition is concerned, they can be matched by rocks occurring in the Devonian area though the igneous and sedimentary types do not occur in the same relative proportions. [J. J. H. T.] In the Proceedings of the Geologists Associationf for July, 1901, the author endeavoured to epitomize the arguments on the stratigraphical side for and against the pre-Devonian age of the rocks ot the Southern area. It will be seen that the first of the two diagrammatic sections given in that communica- tion connects the Torcross volcanic horizons with the Green schists, on the assumption that the Beeson grits form an anticlinal. Eecent investigations have disproved the other hypotheses, except the last given on p. 123, which is retained in a modified form, the Tinsey Head slates (zone 5) being taken as a higher horizon in the Meadfoot beds instead of Eifelian. It will be seen that this alternative structure does not point definitely to the correlation of the Green schists with igneous rocks in the Northern area embraced in these maps. (See Diagram, Fig. 27, p. 62). Mica-schist group. Tracts of pure mica-schists, entirely devoid of quartz-schist interlamination, occupy a comparatively small part of the Southern area, and owing to the frequency and sharpness of the plications cannot be traced as a separate group from mica- schists with very fine quartz-schist interbanding, and quartz- schists with mica-schist intercalations or interfilmmgs ; although there is every reason to conclude that such differences in sedimentation originally formed more or less distinct groups. When the silicification of limestone films and bands, rendering them undistinguishable from quartz-schist films and bands which were orginally grits, is taken into account, the impossibility of distinguishing a calcareous interbanded series from an inter- banded grit series will be apparent, and any observations tending * Geological Magazine, 1892, pp. 1-20. t Vol. xvii., part 3, p. 125. 46 GEOLOGY OF KINGSBRIDGE AND SALCOMBE. to suggest sequence of deposit would be hypothetical, and necessarily based on the correlation of the quartz-mica-schists with the Devonian series which cannot be proved to demon- stration. In parts of the district, as to west of Hall Sands, it would be difficult to distinguish the least altered examples of mica-schist, finely interlaminated mica-quartz-schists and quartz- schists from examples of the corresponding Devonian sediments. Professor Bonney, who has drawn attention to this fact, ascribes it to the deformation of true schistose character in the former case and to the simulation of schistose character in the latter. All remarks as to chemical alteration of calcareous bands in the Northern area might be reiterated here, as the association of brown or red brown decomposed materials is not infrequent in parts of the area of the Mica-schist group (at Splat Cove, near the Bolt, for example), and ferriferous carbonate rocks at Long Cove, south of Hall Sands, and elsewhere suggest the former presence of limestones or calcareous grits, similar to those of Beacon Point and Tinsey Head, in the series. In the introductory Notes to the Southern area reference has been made to the general relations of the Mica-schist group to the Green schists, and both there and in the subsequent descrip- tions of junction sections examples have been given in illus- tration which show that seams of mica-schists occur in places in the Green schists near the junction, and that in places the rapid alternation of mica-schist with bands of the green schist (often of brown color) may be due to interbedding. Besides these however, there are local examples of a kind of passage, as at the Moult, where the representatives of the green schist are inter- calated with thin bands of quartz-schist, the interbanding showing gnarling in a very marked manner. These tend to prove the intimate connection between the two groups. Further evidence is afforded by the interesting beach reefs and cliff sections at Langerstone Point on the east, and in Hamstone Cove and Moor Sand on the west, of the Prawle mass, and further west at Seacombe Sands and Rickham Sands. Clearly faulted junctions are exceptional, the best examples are fur- nished by two faults on the south coast. One of these, at Deckler'slsland, in a direction slightly east of north, cuts off a mass of Green schists against Mica-quartz-schists on the east ; the other, at the western end of Rickham Sands, cuts off Mica- schists in plicated association with Green schists in the beach reefs against the mass of Green schists on the west. Neither of these faults are actually exposed in section, but they are very marked in tracing the boundaries of the groups northward. The Rickham Sands fault, prolonged in a direction north by west, cuts off the brown rocks at Rickham Farm mentioned on page 49. The nearly horizontal planes in the Green schists con- trast so strongly with the contorted Mica-schists on the other side of the concealed fault as to give an appearance of uncon- formity between the two groups at the base of the cliff. The most interesting coast section of the.rocks of the Mica- schist group is that on the east coast, from the mouth of the SOUTHERN AREA. 47 Bickerton Valley southwards to Start Point, as it crosses the strike more abruptly than other parts of the coast, is much more continuously accessible than the Bolt coast line, and exhibits most of the typical varieties of the group, as well as the occa- sional intercalation of altered igneous bands or seams, and, especially between the Bickerton Valley and Hall Sands village, shows numerous faults and thrusts, as well as gnarling and minor curves. The old beach platform reefs between Peartree Point and the Green schist reefs of Horseley Cove afford excellent opportunities of studying the rocks in plan, and good examples of corrugated quartz and quartz-albite veins may be seen. In parts of this coast the appearance of the rocks suggests a comparison with the Strete cliffs as to original sedimentation. Here and there the members of the Mica-schist group break through the steep slopes in rugged crags, as between the Start point and East Prawle and between Eew and South Sands. Occasionally they crop out on the summits in sharp arretes, as in the case of the quartz-schists west of Start Light House, and east of Gullet Point to the north of Waterhead : but in these and similar cases the rocks are sharply contorted or highly inclined. Where the beds or planes of contortion are inclined at low angles, as near Southdown in the Bolt district between Higher Sewer Coastguard Station and Bolberry Down, the mica-schists and quartz-schists weather in flat- topped masses and tors. There are few exposures, however, in the more central flattish summit levels of the Mica-schist districts. In the interlaminated quartz-schists of the East Prawle neighbourhood, and in many other places in a less conspicuous manner, there occur small holes or pits in the quartz bands often containing ferruginous matter which at first sight suggest casts of crinoid ossicles, etc., but no traces of organic structure have anywhere been detected in the rocks of the Southern area, and these and similar cavities are probably due to the decomposition of cubes of iron pyrites, which are often met with in the rocks of the Mica-schist group. When the crushed and tortured condition of these rocks is taken into account, no arguments based on the absence of fossils in them can be legitimately advanced as indicative of originally azoic sedimendation. The distribution of the Green schists has been shown in the introductory notes on the Southern area to indicate an anticlinal structure in the mass of the Mica-schist group in the Portlemouth and Start area lying between their outcrops. The Mica-schists which lie to the north of the convergent bands of Green schist have, in discussing the relations between the Northern and Southern areas, been proved also to partake of an anticlinal structure, thus accounting for the splitting up of the northern band on the west of Kingsbridge Estuary and the dying out of the Green schists along the metamorphic boundary, either by the troughing out of their lower horizons by plication or through 48 GEOLOGY OF KINGSBRIDGE AND SALCOMBE. the northerly attenuation of the igneous materials. There are not the same facilities for deducing the general structure of the Mica-schist group -in the Bolt area, but the absence of traces of the Green schists, and, so far as I have been able to observe, the absence of local bands or seams of igneous rock forbid the occurrence of any large synclinal structures. Some specimens of types in the Mica-schist group are described by Mr. Teall as follows : — No. 3091. Waterhead Creek, Salcombe Estuary. A mica schist (?) of the thonglimmerschiefer type with puckered quartoze folia. The quartzose folia consist of micro- or crypto-crystalhne quartz. The micaceous portion is a confused aggregate _ of mica flakes with which some micro-crystalline material is associated. The planes of lamination, or foliation, have been much puckered and the mica flakes have consequently been bent and broken. Dark' streaks probably due to carbonaceous matter occur between the micaceous laminae. Chlorite is present. The original characters of the rock have been much obscured by dynamic action. No. 1778. Long Cove, South of Hall Sands. — A purplish fine-grained schistose rock containing numerous crystals or crystalline grains of a ferriferous carbonate. Under the microscope the principal constituents are seen to be the above-mentioned carbonate, white mica and quartz. Ferric oxide and rutile (in yellowish prisms and crystalline groups) occur as im- portant accessories. » No. 1779. Same locality — This rock is similar to the last. The car- bonates occur as large crystalline individuals without form and also as small idiomorphic crystals (rhombs). Entile is present as in 1778, but there is less ferric oxide. No. 1780. Start Point, near the Lighthouse. A quartz-schist with micaceous partings. Planes of schistosity well striated. The main mass of this rock is formed of irregular individuals of quartz which vary considerably in size and form, but which interlock with each other in such a way as to constitute a kind of ground-mass. The other constituents are chlorite, white mica, tourmaline, rutile and opaque grains. These lie in the quartzose aggregate as in a matrix. The alignment of the micas is very perfect. No. 1781. Same locality — This specimen is more micaceous and shows a marked puckering. Aggregate of quartz- grains (felspar probably also present in this aggregate) together with white mica, chlorite and some opaque particles— gnarled mica-schist. The following note is by Mr. Hunt : — * " South of Start Farm. In a slice of one of the quartz schist bands, two minute crystals of tourmaline occur ; one a longitudinal, the other a basal section. In both cases there is a re-crystallisation of the tourma- line, and in both cases also a slight shearing movement has detached and removed to a short distance portions of the newer growth. At this spot only has the quartz schist revealed any traces of its original sedimentary components in grains of quartz sand and of tourmaline. The quartz schist bands containing tourmaline have much resemblance to the grit bands in the slates, both in their relation to the mica schists and in their fineness of grain. If there is no connection between the two sets of rocks, the conditions of sedimentation at the deposition of the schists must have been exactly repeated in Devonian times. Green Schists. Wherever these rocks occur in mass, as in the Prawle district, on Rickham Common, between Kellaton and the Kingsbridge * Proc. Geol. Assoc, Vol. xvii., part 3, p. 129. SOUTHERN AREA. 49 Estuary, at Salcombe, and in the Bolt Tail district, they are dis- tinguished by an almost universal green colour, presenting various hues from pale yellow to deep green; but, where thin masses detached by plication occur in the Mica-schists, and often at and near the borders of the main masses, oxidation has ob- literated the greenish tints and changed them to red and brown. If the brown tints were confined to amorphous junction beds between the typical Green schists in the northern band and the Devonian slates, as in Hope Headland, they might be regarded as rotted Devonian volcanic rocks, on the north ofthe metamorphic boundary, but as already shown CKingsbridge Creek, &c.) the green colors are undoubtedly replaced by brown south of the j unction. Well within the metamorphic area brown tints are met with in the Prawle and Rickham Common Green schist masses. In the Waterhead Creek section the Green schists are very poikilitic*. In the plicated area between the Moult and Malborough, the replacement of the green by brown colour is frequent. A band of brown representatives ofthe Green schists is cut off by fault on the east at Rickham Farm. The peroxida- tion of parts of the Green schist section at Hope has been re- ferred to. The local peroxidation ofthe Green schists in irregu- lar banded and plicated association with peroxidated Mica-schists between Bickerton and Chivelstone, deprives the surface evidence in the absence of sections, of distinctive characters, and leaves the true association of the rocks exceedingly indefinite. Pale grey-green and straw-coloured tints are exhibited by varieties of the Green schists, and are more prevalent in the northern than in the southern band. Miss Raisin's micaceo-chloritic series would embrace some of these types. In Willow Cove, near Prawle Point, banded micaceous shaly rocks make a distinctive zone in the Green schists, and can be readily traced in the beach reefs. Evidences of vulcanicity in the Southern area are not wholly confined to the Green schist bands and the masses detached from them by plication. From the Hope cliff' section and the Start Bay coast it would appear that occasional volcanic bands and seams were intercalated here and there in the upper part of the Mica-schist group. In the Hope cliff, as before stated, several bands and seams of brown and yellowish rock, evidently of igneous origin, occur in the dark mica-schists, and masses of hard rock occur at the base of the section. In the Start Bay section, commencing at Hall Sands Lime Kiln, we find a volcanic band a foot in thickness, and very near it yellowish and greenish rock intercalated in the Mica-schists, which are here interfilmed with whitish bands and much gnarled. Further on, near Hall Sands village, pale yellow, whitish, brown, and buff schists are associated with dark grey mica-schists, neither showing much signs of gnarling. The next evidence is afforded by tumbled * In this section yellow, purple, red and brown tints and mottled masses contrast vividly with the green, and in the brown masses carbonates are present. 7294. D 50. GEOLOGY OF KINGSBEIDGE AND SALCOMBE. blocks of altered igneous rock at Long Cove, which (as in parts of the Devonian section) cannot be traced in the cliff through the undergrowth and talus obscuring it. Further south? at Little Broad Cove, decomposed igneous rock is associated with the Mica-schist group in two places. In the best example the rock is pale green and, in plicated association with dark grey and purplish mica-schists with quartz-schist interlaminations, is dis- turbed by a small thrust. Under the microscope all that can be said is " Apparently a green schist much crushed and altered."* At Freshwater Bay very thin impersistent greenish bands, apparently volcanic, are associated in the Mica-schists. On the north side of Start Cove red and greenish rock, apparently vol- canic, is interplicated with the Mica-schists, and in Start Cove, just under the Lighthouse, there appear to be two bands of decomposed igneous rock in the Mica-schists. A small mass of Green schist, forming the axis of an inverted syncline, occurs in the cliff of a small cove locally named " Spirit of the Ocean Cove," on the south of Foxhole Cove. Notwithstanding its isolated position, this rock seems rather to be the relic of an overlying mass of the Green schists than a mere local band, and having regard to the south-westerly strike of the Mica schists between Start Point and Peartree Point it suggests the former extension of the Green schists from the Prawle to the Start. South of Lannacombe Beach at Woodcombe Sand there are impersistent traces of green schist in the Mica-schist. On the south of Grants rock an igneous band is traceable in the beach reefs. On the south of Stinking Cove yellow-green, with inter- calated grey-banded, rocks form a zone about thirty yards wide clearly manifest in the beach reefs. This band, perhaps shifted by fault near the coast, seems to continue westward to East Prawle. The alternations of Mica-schist and Green schists on the foreshore between this and Langerstone- Point seem to be due to plicated repetition rather than intercalation. Many more instances of isolated igneous bands may have escaped detection through the impossibility of following round the Bolt coast line at the foot of the cliffs. The only volcanic development in the Devonian Area com- parable to the Green schists is furnished by the Ashprington series. The correlation- between these groups, should the Devonian age of the Metamorphic rocks be definitely established, is, however, less probable than the connection of the Green schists with the igneous rocks of Stoke Fleming or with those in the Torcross group. Petrological notes — by Mr. Teall : — No. 1784— From the quarry east of Bickerton. A fine-grained greenish rock with no well marked schistosity. Under the microscope it is seen to consist of epidote, chlorite, hornblende, felspar, ores and carbonates. Bickerton— Mr. . Hunt notes :t " A slice from this locality displays the felspars arranged in roughly parallel bands (miscroscopic) but still as granules." * Quotation from Mr. Teall's Notes. t Proc. Geologists' Assoc, Vol. xvii., Part 3, pagel28, SOUTHERN AREA. 51 No. 3086. — From coast south of Start Fm. (Signal House on old 1-inch map), at " Spirit of the Ocean " cove. Pale greenish schistose rock. Microscopically found to consist of hornblende, epidote, and a colourless microcrystalline mosaic. Portions of fairly compact hornblende are present iii this specimen. The rock is evidently one of the Green schist group. Mr. Hunt furnishes the following note:* " Spirit of the Ocean Cove. The water-clear felspar has lost its isolatedgranular character. It is distributed in roughly parallel streaks. Mr. Harker describes a slice as " a mass of minerals newly crystallised in situ and having a fresh appearance so characteristic of the crystalline schists. Original structures are completely obliterated. . . . The minerals described by Mr. Harker are compact and fibrous amphibole, felspar (albite or andesine) ripidolite, epidote, and zoisite." No. 1783. — Hamstone Cove. Prawle Coast. A greenish or yellowish green schistose rock. As seen under the microscope, a confused aggre- gate of epidote, chlorite, hornblende, felspar, calcite, quartz (scarce). No. 3090. — Seacombe Sands. A fine-grained greenish grey schist. Microscopically, albite (abundant), quartz (scarce), chlorite, epidote, white mica, garnet (scarce), sphene and iron-ores. This rock was originally an albite-schist, but it has been much affected by crushing. No. 1782. — Shore south of Portlemouth, west of Seacombe Sand. A pale green fine-grained schistose rock. Felspar crystals measuring sometimes one-sixteenth of an inch are scattered abundantly through certain bands. Microscopically this rock is essentially composed of albite, epidote and hornblende. The albite occurs for the most part as an aggregate of colour- less, untwinned grains, and might easily be mistaken for quartz. One or two of the larger individuals were isolated. Flakes showing the oblique emergence of a positive bisectrix (M. flakes) gave an extinction of 19° as referred to what was doubtless the trace of the basal cleavage, and small fragments tested by Szabo's method showed the flame-colouration and fusi- bility of albite. No. 1787. — Quarry by road to Combe, west of the Moult. Greenish grey fine-grained rock with a marked parallel structure. Microscopically, granulitic colourless felspar, epidote and pale green hornblende mostly in slender prisms. Hornblende-schist. * Proc. Geologist*? Assoc, Vol. xvii., Part 3, page 128. 729' T> 2, 52 GEOLOGY OF KINGSBKIDGE AND SALCOMRE. CHAPTER V. NATURE OF THE BOUNDARY BETWEEN NORTHERN AND SOUTHERN AREAS. On the whole the boundary of the metamorphic area can be located with tolerable certainty owing to the comparatively unaltered condition of the slates and shales on the north side. The rapid change from one group to the other is equally evident where the Green schist series is in contact with slates, and between Malborough and Hope, where Mica-schists are in contact with them. As the importance of this boundary cannot be overrated, the evidence on which it has been drawn will be given, in some detail. To begin with the east coast. Here the boundary was taken near Tinsey Head, south of Beesands, on the old Geological Survey Map, Sheet 23. Professor Bonney subsequently pointed out the erroneous position of this (dotted) line, maintain- ing that, though the rocks north of the Bickerton Valley locally " assume the macroscopic habit of schists," those to the south of the valley belonged, without doubt, to a more ancient metamorphic series. In his sketch map of the district Professor Bonney places an east and west fault at Hall Sands, but in his description leaves its exact position somewhat doubtful. Miss Raisin gives a more detailed map, and runs the supposed fault seaward from under the alluvium of the Bickerton Valley. The impossibility of accepting this view has been pointed out by Mr. Somervail* as the Green schists are exposed in a quarry on the north side of the valley. Mr. Teall describes a specimen from this quarry (E. 3087) as — " A dark green, slightly schistose rock, composed of albite, epidote, chlorite, hornblende, carbonates, and iron ores. Undoubtedly a rock belong- ing to the green schist group, and not to the normal Devonian diabases." Mr. Hunt observes : " In my two slices from this quarry water- clear granules of felspar are scattered evenly throughout the green matrix."f The strike of the Green schists in this quarry, if prolonged, would touch the coast at about a quarter of a mile south of Tinsey Head, and nine chains north of the commencement of the cliff at Greenstraight. The Hall Sands cliff being Mica-schist, and the Green schists of Bickerton either not continuous to the coast or concealed by alluvium, the cliff section north from Green- straight must be described for some chains northward in detail. The letters refer to parts of the accompanying illustration (Fig. 22). For thirty yards from its commencement the section con- sists of (a) intercalations of brown shaly beds containing hard, * Trans. Devon. Assoc, for 1887, Vol. xix., p. 354. + Proe. Geologists' Assoc, Vol. xvii., Part 3, p. 128. NORTHERN AND SOUTHERN AREAS. 53 brown-weathered irregular masses of siliceous carbonate rock with dark grey shales. These are overlain by (b) yellow and red shales, resembling beds associated with the Green schists at Bickerton, and probably tufts. The succeeding rocks (c) consist of brown and yellowish shales or slates with hard lenticles, probably of siliceous carbonate, separated by grey slates or shales with hard siliceous (?) filaments from decomposed yellow and red rocks Fig. 22 — Cliff at Greemtraig/u. Horizontal scale, 1 inch= 60 yards. Vertical scale, 1 inch=120 feet. associated with masses of siliceous carbonate, as in (a). At this point, three chains from the commencement, the section is obscured for twelve yards, and then exhibits (d) yellow-brown decomposed shaly or slaty beds, which continue for thirty-four yards. With the exception of the grey slates in (c), the rocks in (c) and (d) are probably of volcanic origin. The dips of both are northward at 60°. For ten yards further (e)grey slates or shales are exposed and then cut off by a thrust fault from (/) yellow decomposed material underlying hard grey, brown-weathered rock of the same appearance as the siliceous carbonate masses in (a). This rock is succeeded at ten yards further by grey and buff slates or shales, possibly in part fine volcanic tuffs, which are exposed by a path to the summit of the cliff over an irregular disturbed-looking grass-grown projection, probably an extensive outward slip of the cliff. In this ground grey and greenish-grey even slates or shales are exposed at the northern end of the sec- tion, and give a southerly dipof about 10°, which is, from the nature of the ground, unreliable. Beyond this point, which is ten chains from its commencement, it is unnecessary to trace the section. A specimen (No. 3098) obtained in this section, at about 5 chains from Greenstraight, is described by Mr. Teall as—" a grey, cherty carbonate, apparently interstratified with a brown rock of doubtful origin, possibly a tuff. The two rocks are separated by a narrow zone, extremely rich in pyrite." It will be seen that there is no break in this section, which could be taken either as a fault boundary, or satisfactory junc- tion, between altered and unaltered rocks. On the other hand, siliceous carbonates and decomposed brown rocks, not distinguish- able from the brown junction rocks, are met with in the Devonian area, both in the Eifelian (lower part of Middle Devonian) rocks, and in the Lower Devonian rocks of the Meadfoot series. More- over, in sheets 355, 356 this type, as we shall see further on, is frequently present where the Green schists are in immediate 54 GEOLOGY OP KINGSBRIDGE AND SALCOMBE. proximity to the unaltered Devonian, and forms as it were_ an intermediate series. If these rocks are included in the Devonian, the boundary would have to be drawn in the position assigned-to it by Professor Bonney and Miss Eaisin at the coast, but stepped byfault or deflected northward to include the Green schists pointed out by Mr. Somervail on the north side of the Bickerton valley. Kellaton Junction. — For a mile from the coast until the village of Kellaton is reached no actual junction sections are obtainable. The uppermost beds in the Green schist quarry on the north side of the Bickerton Valley are soft banded shales which would probably escape notice in a Devonian section. At Kellaton unaltered grey slates of the Devonian group become gradually intercalated with yellowish brown beds and harder rotten brown stony bands, giving place to soft peroxidated and yellow-brown materials with seams of soft reddish and dark grey shale or schist with highly micaceous surfaces and whitish interfilming. The section appears to be a more or less rapid passage from unaltered sedimentary to decomposed volcanic material, which cannot be separated from the Green schists. The oxidation of both Mica- schists and Green schists between Kellaton and Bickerton, and the decomposed character of the latter, coupled with rapid alter- nations by interbedding or plication, renders it impossible to prolong the lines with any approximation to certainty. The above remark is also applicable to a part of the district between Kellaton, South Allington, and Lannacombe corn mill ; and to the band of country between the Salcombe Gas Works, coast ot Batson Creek, and Malborough. Between Kellaton and Southpool Creek, the junction beds are partially exposed between Northern Town and Cousins Cross, where they appear to be similar to those at Kellaton. For the remainder of the distance the line had to be drawn in the absence of exposures, but a well-marked band of Mica-schist is visible, apparently separated from the Devonian slates by a narrow band of Green schist. A specimen, near the junction, on the north of a quarry, in the Green schists (No. 3093), to the north of Chivelstone, is described by Mr. Teall, as " alternating lenticular quartz-felspathic and micaceo-chloritic folia. The felspar is albite. The quartz has been more affected by the movements than the albite. The rock must be classed with the albite-schists." Southpool Creek Junction. — The Southpool Creek section* is best shown on the west bank ; here, the Devonian rocks consist of grey slates with brown ferruginous films and a local abundance of quartz veins. They are intersected by several faults. The schistosity dips are mostly northerly at angles of from 50° to 70°. In spite of the effects of surface weathering in the low cliff, there can be no doubt as to the absolute correlation of the rocks with the slaty series south of Tinsey Head on the Bickerton coast, and between Beacon Point and Mouthwell on the Hope coast. These beds are separated by a thrust with a northerly inclination * See Proc. Geologists' Assoc, Vol. xvii, Part 3, pp. 129, 130. July, 1901. NORTHERN AND SOTTTHERN AREAS. "5 o -« &3 =c O and the inner angle of Mouthwell Cove types) with the Mica- schists, The brown rocks of the Batson Creek sections have not furnished the. hard white siliceous carbon- ate found at Greenstraight, Kingsbridge Estuary (east side), and Hope, and are not seen in continuous exposure to contact with unaltered slates. The Hope section exhibits the east coast and Kingsbridge Estuary brown rocks in plicated conjunction with rocks ascribed to the Devonian slates (although rather more micaceous in, and on, the north side of Hope Headland) and with Mica-schists on the south side of it. The tendency of these sections taken together is to show that a series of de- composed brown volcanic rocks associated with argillaceous beds, locally calcareous and containing limestone in places, form an intermediate group between the altered and unaltered rocks, and where associated with Tocks sufficiently fresh to exhibit grades of metamorphism seem to be as intimately connected with the one series as with the other. Two alternative structures were given (see p. 36) of the relations of the Devonian groups in the Northern area. One of these sections, which connects the Torcross and Tinsey Head slates over an axis of Beeson Grit, naturally connects the Beeson Grits with the Ringmore Beds. As already stated, this section has been figured dia- grammatically.* The correlation of the Torcross and Tinsey Head slates naturally suggests the representation of the Torcross igneous rocks by the Green schists. The alternative hypothesis is sketched in the accompanying diagram (Fig. 27); in it the Beeson grits are taken as an out- cropping, more or less impersistent series. Unless these grits and overlying interlaminated beds can be proved to represent the Staddon grits, which would bring the Green schists into direct connection with the Ashprington volcanic series, there is no definite Devonian vol- canic series with which to connect the Green schists. 2£ &> * Proc. Geologists' Assoc, Vol. xvii., Part 3, p. 124. NEW BED ROCKS OR PERMIAN. 63 CHAPTER VI. NEW EED ROCKS OE PEEMIAN. Beyond the well-known outliers of Slapton and Thurlestone there is no evidence of the New Red rocks, either on the seaboard or inland, in Sheets 355 and 356. It cannot, therefore, be in- ferred that these rocks once extended across the promontory, from shore to. shore. However the positions of these outliers taken in connection with the traces of New Red on Berry Head on the one hand, and with evidences of its presence in con- nection with the Cawsand felsite on the other, point to the in- ference that the present coast line had been in no small degree determined by their ancient margin. The Slapton outlier consists of sand and breccia ; the former, as near the shed on the north side of Slapton Bridge, showing true and false laminations, or current bedding ; the latter Fig. 28. — Current-bedding near Slapton Bridge. v?^,^#...r?h. a -v~.-*,r-C'i"r derived from the local slates, grit-shales and grits, and from the quartz veins occurring in them, with some igneous and occasional limestone fragments*. A connection between the larger mass and the* small adjacent patch on the west is possible> but very doubtful, and the northern boundary, on the east of Slapton, is somewhat obscure. The Thurle- stone outliers consist of (1) the stack called' Thurlestone Eock, which is composed of thick weathered beds of breccia tunnelled by the sea into an arch, the pillars of which rest upon a flat reef of softish lilac, red and grey banded, Devonian slates or shales (type of shales in Collapit Creek, &c.) ; (2) the cliff outlier and (3) the patch near Horswell House. The two latter may, however, be connected under the alluvium of the intervening valley. The patch near Horswell House is exposed in the north side of the quarry (Fig. 29) in mica trachyte, where it seems to overlie and to abut against the mica trachyte (m. a.) ; an appearance which is probably due to fault (f.)- The chief point of interest in this patch is its being in touch with the mica trachyte. The latter is shown * See Peugelly — On Triassic Outliers. Trans. Devon. Assoc, for 1866, p. 53, &c. 61 GEOLOGY OF KINGSBRIDGE AND SALCOMBE. (a.) in very irregular junction with red, baked-looking, Devonian shales (p.), very like "Thonstein" fragments in Crediton and Teignmouth breccias. The mica trachyte also recalls types of fragments contained in these rocks. A rhyolitic rock, much decomposed, is blended with the trachyte in one part of the quarry (m. a. and E.). Fig. 29. — Horswdl House Quarry. The New Red on the coast is of a purplish grey colour, and forms a breccia resembling in assortment varieties of the rock between Teignmouth and Dawlish. It is composed of fragments of the soft buff Devonian shales of the beach reefs, with fragments of quartz, and apparently of quartz- albite, stones of quartzite and fine grit and of green, very micaceous, spangled schists in a comminuted matrix of similar materials. The schist fragments consist of " quartz, felspar (not abundant) and green mica " (Specimen No. 1877).* The line of junction in the cliff is irregular, the beds of breccia, dipping south-south-west (in the direction of Thurlestone Rock) at an angle of about 20°, are successively overlapped on the red and lilac soft Devonian shales or slates supporting them, which give inclinations of 50° to 60° in a direction north 10° west (Fig. 30). The section shows a con- siderable likeness to Torcross further on, and contains the Fig. 30. — Cliff between Leas Foot Sand and Thurlestone Sand. Horizontal scale, 2 Feet to the Mile. Vertical scale, 1 Inch=160 Feet. band south of Leas Foot Sand, described in the Devonian Chapter, p. 28. The occurrence of traces of New Red on the Brixham limestone plateau, near Waddeton, and in the Torquay penin- sula suggest the explanation of the limestone plateaux in Sheet * Quotation from Mr. Teall's Notes. NEW RED BOCKS OR PERMIAtf. 65 350, as fragments of an old plane of Permian (?) denudation, but' if such is the case the ancient shore line has been long since obliterated, and in sheets 355 and 356, although there are lands in the vicinity of the New Red outliers, which, in relative level, would bring them into relation with the limestone plateau, there is no indication of a contour to justify the belief that, between the heights on the south and those on the north of the districts under consideration, such a plane actually extended in Permian (?) times between the Thurlestone and Slapton coasts. 7294. E 66 GEOLOGY Of KINGSBRlDGE AND SALCOMBE. CHAPTER VII. PLEISTOCENE AND RECENT DEPOSITS. No ancient river gravels have been found in the area, the oldest signs of Pleistocene changes being furnished by shelves or platforms of rock, forming reefs round the coast in many places and supporting cliffs of Head, or ancient talus, at the base of which traces of Raised-beach are occasionally noticeable. The best example of Raised-beach is probably that at Splat Cove, near the mouth of the Salcombe Estuary. The old beach plat- forms, as Pengelly very clearly illustrated,* have a seaward slope, therefore the evidence to be derived from them as to the amount of elevation and depression is never absolutely reliable. For example, a platform covered by cliff Head at five feet above high water might be found in some parts of the coast between the Start and Prawle to slope upward beneath the old talus to a height of 20 or 30 feet above high water, if it were possible to remove the accumulation masking it. The coast line' of the Raised- beach period was cut a little further back between Prawle Point and Peartree Point, west of the Start, than elsewhere on the sea- board, hence the thickness of Head, in places from 40 to 50 feet, m the cliffs, and the nearly continuous strip of flatfish or gently - sloping ground flanking the steep slopes of the schists. Between the Prawle Point and Salcombe Harbour the sea has only spared occasional strips of this cliff-fringing Head plateau, and it is not again apparent in the Bolt coast or from thence to Erme Mouth. Beyond Erme Mouth the Revelstoke Coast, however, affords very food examples of it — and both here and on the coast east of rawle Point, the occurrence of pinnacles, or patches, of the Head on the rock platform, insulated by recent' marine denudation from the neighbouring talus in the cliffs, proves the antiquity and importance of this accumulation, which, in the words of Godwin- Austen, " marks a time when the degradation of the sur- face proceeded much more rapidly than by the mere effects of decomposition, and when fragments of rock far exceeding the motive power of any rainfall were conveyed down slopes along which the minutest particles of matter only are now carried." Raised Beach and Head. Revelstoke Coast. — At Netton Island the Raised-beach platform has been breached along the strike of the rocks, insulating two small ridges of Head from the Head in the main cliff; the base of the Head here is reddish-yellow, finer in texture than above, and contains a very few stones. Another insulated patch at Dunny's Cove may be mixed with Raised-beach material. * Trans. Devon. Assoc for 1866, p. 106. PLEISTOCENE AN'D RECENT DEPOSITS. 67 East of Carswell Cove traces of Raised-beach were met with, at from 8 to 10 feet above high water, on the rock platform, the Head adjacent talus contains wing cases of Beetles. The most continuous strip of Head on this coast is near Revel- stoke, it exceeds 20 feet in thickness. Ivy Island is capped by a ridge pinnacle of Head. At Revelstoke, as also in places between Prawle and Peartree Point, the coast-fringing Head runs up valleys inland — and in the strips of alluvial ddbris which terminate on cliffs, at Carswell Cove and two or three other places on the Erme Mouth coast, in the Lower Sewer Farm valley and the Middle Sewer valley (Starehole bottom) opening on the Bolt coast, in the East Portlemouth valley, the Seacombe Sands valley, and in the little valley near Sheval Rock on the Strete Coast, we have truncated valley deposits, or accumulations of the age of the coast Head. In some places there are signs of these old valley infillings having been denuded in the formation of the present alluvia. God win- Austen * ascribes the local unworn and unarranged gravel in which the Mouth of the Erme has been excavated to a period of long continued subaerial waste during which the West of England stood at a greater elevation. In the bend of the River Avon, near Cockridge Point, a cliff of Head flanks recent Blown Sand. On the opposite side (Bantham Ham), at 6 feet above Bantham Sands beach, 1 foot of brown stony loam is overlain by a bed of sand rock, probably recently consolidated Blown Sand. The Thurlestone cliff New Red outlier is partly masked by what seems to have been an accumu- lation marking the course of the adjacent valley in former times. Bolt Coast.--— The Lower Sewer valley accumulation gives the following section at its mouth, in the cliff flanking Sewer Mill Sand, in descending order : — Feet. Head of medium sized fragments of mica-quartz-schist in loam 5tol0 Large stones of quartz and mica-schist with impersistent seams of smaller fragments - - 5 Fragments smaller than above in brown loam ... 5 Base on mica-schist, at about 6 feet above high water. In Splat Cove, in the promontory bounding South Sands, the cliff is capped by Head, forming a flattish tract, apparently continuous along the lower slope of the South Sands valley for some distance. In the cliff on the south side of the Cove, the Head is about 10 feet thick and rests on buff-brown sand, about 5 feet _ thick, overlying about 12 feet of gravel of local fragments, occasionally well worn, but mostly subangular, in a drab-brown sand matrix which also shows impersistent bedding courses. At base this deposit is about 15 feet above high water, but further east, where * Quart. Jmtrn. Geol. Soc, Vol. vi., p. 94. 7294. B 2 68 GEOLOGY OF tflNGSBRlDGE AND SALCOMBE. the platform is about 3 feet lower, it is capped by a rugged con- solidated sand, made up of comminuted shells which seem to thin out upward on the platform slope under the overlying gravel. In this deposit, whilst the consolidated sand certainly must be regarded as a portion of Raised-beach, it is not unlikely that the overlying gravel and sand may be a subsequent, or rather succeeding, valley gravel or estuarine deposit partly mixed up with the beach materials. Prawle Coasts. — At Venerick's Cove on the Prawle Coast, east of Seacombe Sands, a pinnacle of Head caps the rock platform at a distance of 50 yards from the cliffs, between which and the pinnacle the sea has made a clean breach in the rock platform. Where the platform joins the cliff, at about 60 yards from the pinnacle, there are two patches of partially consolidated brown sand; the smaller being more consolidated and containing pebbles at its base ; the larger rising to 40 feet above high water mark and exhibiting bedding with a seaward dip. Neither of these patches is overlain by Head, which seems to be absent from the cliffs in the immediate vicinity. Either these sand patches are recent, partly blown from beach sand below, with pebbles cast up by storm waves, or they are the landward term- ination of the old Raised-beach (or of Blown Sand and Raised- beach) which is by far the most probable explanation. To account for the isolation of the pinnacle, we must assume that it formed a part of a mass of Head banked against the cliffs and overlying the old beach sand, and, that, in the denudation of the cliff and beach platform, the cliff part of the Head was washed away and the traces of the old beach preserved in the angle between the platform slope and the cliff at its termination. The old beach platforms range from high water mark to 10 or 15 feet above it, on this coast, 5 feet being about the average height where the Head plateaux rest on the platform, between Prawle and Peartree points. At the Prawle end of the Head plateau strip the accumulation forms a cliff of yellow-brown stony loam or sandy earth, occa- sionally presenting a stratified appearance through distribution of material; in one spot, where the cliff is about 24 feet high, the section consists of 20 feet of Head, grey in colour, and with many large fragments in the upper half, yellow-brown and with fewer fragments in the lower— upon 3 feet of red loam with small subangular stones, upon 1 foot of subangular stones and a few quartz pebbles, at base about 5 feet above nigh water. On the slate platform, at the base of the Head cliff, traces of consolidated old beach sand, brown and blackish, with small pebbles of quartz and schist, were noticed adhering to interstices in the platform. At about 15 chains from Lannacombe Mill, in a Cove near this, on the west, a similar occurrence was ob- served at about 8 feet above high water mark ; the Head here also rests, partly on sand, in part directly on the platform ; the present beach consists of quartz pebbles. On the old. beach plat- form near the Head cliff, near Malcombe Point, on the 6 inch PLEISTOCENE AND RECENT DEPOSITS. 69 map two boulders* were seen; one of fine whitey grey rock 3£ it. x If ft. x 1 \ ft., the other of coarse, grey rock with dark elvan veins, 3 ft. x 2£ ft. x \\ ft. These boulders may have been incorporated in old beach materials and remained after the accumulation of the Head, and the subsequent removal of both, in virtue of their size and weight, but it is also possible that they might have been stranded ballast from a wreck. At any rate on the French Coast and in Croyde Bay (North Devon) we have exceptional instances of granite boulders of large size in the Raised-beaches, suggesting ice flotation. East of Lannacombe Mill yellowish-brown sandy loam, visible here and there, at and near the base of the Head cliff, suggests Fig. 31. — The Pinnacles, near Matchcombe Sand. admixture with old beach material. The Head cliff is 40 ft. to 50 ft. high in places, near Matchcombe Sand.and isolated pinnacles of Head, from 15 to 20 feet in height, rest on the beach platform at about 5 feet above high water (Fig. 31). The cliff Head, through local absence of fragments in the red sandy loam, presents an horizontally stratified appearance. The pinnacles are in line from the Coast, the nearest being insulated by about 20 yards from the Head in the cliff. At the further end of Matchcombe Sand (mouth of Start Farm valley) the reddish- brown stony loam rests on a bed of fine reddish sand, containing small pebbles, blackish and consolidated in places, and enclosing numerous subangular stones and pebbles of mica-schist and quartz. In some places this bed is 2 feet thick and rests un- evenly on the beach platform, at about 5 feet above high water. Between the above and Peartree Point one part of the section has been particularly mentioned by Pengelly.f Here the section consists of : — (1) Head of reddish loam studded with large and small angular schist and quartz stones, with bands of stoneless loam here and there (2) Coarse yellowish-brown quartzose sand, reddish in the upper part, visible for a few feet downward, and then concealed where the cliff slopes downward more gently to the old beach platform by (3) Feet. 15 to 20 4 to 5 * Proc. Geologists' Assoc, Vol. xvii., Part 3, p. 130. Pronounced by Mr. Harker to be " A Granite Gneiss and Quartz Diorite.' ; + Trans. Devon. Assoc, for 1866, "On Eaised Beaches." 70 GEOLOGY OF KINGSBRIDGE AND SALCOMBE. (3) Head, of the same character as the Head (1) above, " Feet. resting on tne rock platform at about 5 feet above high water - - - 15 The obvious reading of the section is— to consider the lower Head (3), a recent talus from (1), the Head above, masking the downward continuation of the old beach sand or old blown sand (2) to the rock platform. In favour of this, in an adjacent spot the sand seems to continue downward to the platform, and in its vicinity a trace of blackish consolidated sand adheres to the surface of a narrow reef at 5 feet above high water. From this we can only admit another interpretation than that given by assuming that the sand has slipped downward and not the Head (3). This, appears to have been Pengelly's view, viz. :— That the platform after receiving a covering of talus (3) was let down by further subsidence, the talus cut back, and a sandy beach formed, which, during subsequent elevation, was overlain by the Head (1). The cliffs toward Peartree Point show in one place 30 feet of Head containing very large fragments here and there. At its base from 8 to 10 feet above high water, the surface of the rock platform is strewn with large angular schist boulders which are here and there cemented by coarse black consolidated sand, made up of angular and subangular slate grains and con- taining quartz pebbles. This occurrence and the traces of Head on the rock platforms at the Start and between the Start and Hall Sands, at 10 to 15 feet above high water, being close to the landward termination of the old talus, occupy too little space to be indicated on the one-inch map. Coast North of Start Point. — Between Long Cove and Broad Cove, in the southern angle of Storepot Cove, brown sand con- taining large subangular fragments of quartz and flat bevilled fragments of schist and .slate, often large, and also small pebbles of quartz, grit, and chalk-flint, the whole about 2J feet in thickness, rests on the rock platform at about 14 feet above high water. At some chains to the south a large quartz boulder was noticed on an inaccessible reef. At Hall Sands village the roof of a cleft, used as a boat house, is formed by a consolidated sandy accumulation containing subangular and rounded stones in horizontal lines, from 5 to 6 feet.in thickness and at base about 10 feet above high water. At Beesands the Head is made up of debris from the Beeson grits and slates above, and neither here nor at Strete Gate, or in the intervening district, does it deserve any special notice, the only point of interest being the occurrence by Slapton Ley, on north side of Slapton Bridge, of gravel, opposite to which " Eaised Beach " is engraved on the old geological map. The deposit is partially shown on the present geological map, and is superficially very insignificant. Proceeding from south to north the following sections were observed :— Ft. In. Brown earthy accumulation with unworn slate and a few quartz stone? - -40 PLEISTOCENE AND RECENT DEPOSITS. 71 Coarse earthy reddish sand of comminuted slate with a Ft. In. few small slate, quartz, and flint pebbles - - - 1 Brown gravel of compact grit quartz, slate, and a few flint pebbles and subangular stones ; some of the quartz pebbles are large - - The base of the_ accumulation rests on the upturned edges of the Devonian slates or shales at, perhaps, about 8 feet above high water Near the above, the gravel, containing flint pebbles of small size, is shown resting on the Devonian at about 10 feet above the water-level of the Ley. Further north, a breccia of slate or shale fragments containing - pebbles of quartz, grit and flint here and there, shows sign of false bedding ; but this observation in view of the presence of New Red breccia is unreliable, as a partial reassortment of it might lead to the incorporation of traces of more recent gravels. if the Slapton gravel patch is a fragment of RaisecUbeach, the Higher Ley, if it then existed, was completely severed from the Lower, at the time of its formation. The general inferences to be deduced from the occurrences above-mentioned are : — That the rock platforms on the coast, coupled with the Raised-beach relics, denote a submergence of at least 20 feet. The Head denotes an elevation of the Raised- beaches sufficiently high and of sufficiently long duration to {>ermit of the shedding of extensive talus fans, from the high ands above, over the Raised-beach cliff, and extending for con- siderable distances outward from it, on the old beach platform ; in the process of its accumulation the marine deposits were partially swept away, partially incorporated in the base of the talus. The occurrence of flints in the Slapton gravel and in the sand beach in Storepot Cove seem to point to an opposite drift of shingle to the present, namely, down Channel. Ad- mitting such a drift in the Raised-beach period, the origin of Slapton Ley becomes intelligible, viz., the southerly, deflection of the Gara (the northern tributary to the Ley) by a bank of shingle driven southward across its mouth, from which material would naturally find its way to the opposite bank, and account for the gravel patch on the north of Slapton Bridge as a fluviatile deposit. There can be no doubt as to the existence of the Salcombe Estuary during the Raised-beach formation. Submarine Forests. The submerged forest of Blackpool in Sheet 350, is a very short distance north of the Strete cliffs in sheet 356 ; yet there are no records of the occurrence of submerged vegetation from this spot round the Start Coast until we reach Salcombe Estuary, and beyond the Bolt Head the only recorded instance is on " the south of Thurlestone Sands in Bigbury Bay. Pengelly* described the submerged forest in Bigbury Bay, but only mentions the occurrence of similar phenomena at * J'rans. Devon. Assoc, for 1866, p. 77,