a د‎ IUS 5555 Se S NE SE NEN ( Y Nose SUN xS 1 if 1 РЛ "a m wa ta r Ў £ i MEMOIRS OF THE GEOLOGICAL SURVEY OF INDIA, m y $ MEMOIRS OF THE GEOLOGICAL SURVEY OF INDIA. VOL. XVI. PUBLISHED BY ORDER OF HIS EXCELLENCY THE GOVERNOR GENERAL OF INDIA IN COUNCIL, CALCUTTA: PRINTED FOR THE GOVERNMENT OF INDIA. SOLD AT THE GEOLOGICAL SURVEY OFFICE, OFFICE OF SUPERINTENDENT OF GOVERNMENT PRINTING, AND BY ALL BOOKSELLERS, LONDON: TRÜBNER & CO, MDOCCL ххх. - CALCUTTA : OFFICE OF THE SUPERINTENDENT OF GOVERNMENT PRINTING. 1880. CONTENTS. Авт. l.— Om the Geological Structure of the Bastern Coast from Lati- tude 15° northward to Masulipatam, by R. Bruce Foorz, F.G.S., Geological Survey of India. PAGE. CHApTER = Introductory. у za i II.—The Gneissic or Metamorphic Series DLE DOR d : ; 7 1.—The Schistose areas 5 - : : 5 а ЕТИ 2.— Тһе Granitoid areas . -. é 5 аЬ oT 3.—Intrusive rocks in the Gneissic area . i : 25-48) a.—Trap dykes and intrusions . : ; : d b.—Granite veins . с E ; ? 5 5 49 c.—Felsite veins -. Б Е F ; : ЛА d.—Quartz veins . : : : 3 : . 44 » III.—The Kadapa Series . : 5 5 2 P os : o. 46 » IV.—The Upper Gondwána Series . ; : ; ó 6 - 749 i V.—Cuddalore (Rajahmundry) Sandstones . : : 5 = 847 % VI.—The Lateritic Rocks 5 ; ‘ З : ; о o e T VII.—The Alluvial Deposits . ; 5 5 А 2 ; s 98) » WIII.—Soils and Sub-aérial Deposits . р F i д IT a IX.— Economic Geology . Я 5 : 3 ; о 5 . 103 Amr. 2.—The Gneiss and Transition Rocks, and other Formations of the Nellore portion. of the Carnatic, by Wittiam Kine, B.A., Deputy Superintendent. (Madras), Geological Survey of India. PAGE. PART I. CHAPTER I.— General Description Е ; } : el - . 109 II.—Physical Geology . ; : : : : ن‎ : o dl vi CONTENTS. PART. LI. CHAPTER IIT.— The Gneissie Series The Schistose Gneisses - IV.—The Transition Series ue V.—Granitic and Trappean Rocks . 4 F VI.— Other Formations 1.—Rajmahal Plant Beds 2.—Cuddalore Sandstones 3.-—Lateritic Deposits 4.— Recent Deposits » VIi —Nellore Copper Workings 33 PAGE. 125 133 144 164 171 171 175 179 180 185 Art. 3.—The Upper Gondwänas and other formations of the Coastal Region of the Godávart District, by Wırnıam Kine, B.A., Deputy Superintendent (Madras), Geological Survey of India. CHAPTER — L—General Description II.—Gneiss and Lower Gondwäna Rocks T III.—Upper Gondwánas Я 6 : . 5 IV.—Deccan Trap Series V.—Cuddalore Sandstones ». VI.—Economie Geology . PAGE. 195 206 211 231 248 252 ILLUSTRATIONS. Fig. l.— Distribution of the varieties of gneiss — . . » 2.— Section from near Sitarampuram to Pamur, 24 miles . » 9.— Well-section south of Mangamur В : о » 4.—Diagram-section across Vemávaram Ridge › 5.—ldeal-section from Budaváda to Pávulur Portrait of sloping surface of trap dyke . : Е : 166 MEMOIRS OF THE GEOLOGICAL SURVEY ОЕ INDIA. N Ы) 4 CNN MEMOIRS OF THE GEOLOGICAL SURVEY ОЕ. INDIA. VOL. XVI, Pr. 1. PUBLISHED BY ORDER OF HIS EXCELLENCY THE GOVERNOR GENERAL OF INDIA IN COUNCIL. CALCUTTA: PRINTED FOR THE GOYERNMENT OF INDIA. SOLD AT THE GEOLOGICAL SURVEY OFFICE, OFFICE OF SUPERINTENDENT OF GOVERNMENT PRINTING, AND BY ALL BOOKSELLERS. LONDON: TRÜBNER & CO. MDCCCLXXIX. CALCUTTA : OFFICE OF THE SUPERINTENDENT OF GOVERNMENT PRINTING. 1879. ^ Y Foote. GEOLOGICAL SURVEY OF INDIA. Memoirs. Vol: XVI. < v Bezwada U y GUNTUR b Chebrolu en 2 Y SAPATLAN (Meese Chendalur [LL ка & Karnul Rocks Gneiss 22 l. Singersyaconda, EES Suhistoce | Dc Gr wniteid | qRami ap atn am Ox Stone by Aminullah. FIG 1. GENERAL DISTRIBUTION , OF THE GRANITOID AND SCHISTOSE VARIETIES OF GNEISS. MEMOIRS OF THE | GEOLOGICAL SURVEY OF INDIA. On THE GEOLOGICAL STRUCTURE OF THE EASTERN Coast FROM LATITUDE 15° NORTHWARD TO MASULIPATAM,! dy R. Bruce Еооте, F.G.S., Geo- logical Survey of India. CONTENTS. CHAPTER I.—Introductory. CHAPTER IV.—The Upper Gondwana Series. т IL—The Gneissie or Metamorphic » V.— The Cuddalore (Rájahmundry) Series— sandstones. 1. The Schistose areas. E VI.—The Lateritic Rocks. 2. The Granitold areas. 5 VII.— The Alluvial Deposits. 3. Intrusive rocks in the » ҮІШІ.--бойв and Subaérial Deposits, Gneiss area. E IX.—Economic Geology. III.—The Kadapa Series. 2» CHAPTER I.—INTRODUCTORY. The region to be described in these pages is included in four sheets of the Atlas of India, wz., Nos. 75, 76, 95, and the extreme south- west corner of sheet 94. The topographical structure of the area dealt with is extremely simple. It is a mere inclined plane sloping gently 1 Masulipatam is in latitude 16° 10', twenty miles east of the Kistna: of the area under description that portion between the Kistna and the Godavari, on sheet 95 of the Indian Atlas, contains only deltaic alluvium, and has been omitted in the accompanying sketch (1) map. Memoirs of the Geological Survey of India, Vol, XVI, Art, 1, 9 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. eastward from the foot of the Eastern Gháts to the Bay of Bengal, and diversified only by seattered hills, or groups of hills, of no real import- ance geographically considered. The drainage is effected in the northern part by the Kistna river, in the southern and central parts by a number of small rivers running mostly in shallow valleys. Fiscally the area is divided between the Kistna! and Nellore distriets, the northern part belonging to the former, the southern to the latter collectorate. The geological structure of this area is almost as simple as the topo- QUT eT graphical. The western part of the area is occupied by a broad band or zone of ancient crystalline rocks belonging to the gneissic series, the eastern part is formed of the marine and fluviatile alluvia, while the intermediate part (speaking roughly) is occupied by a band of patches of sedimentary rocks of two ages, both older than the alluvial formations. Of these three bands the gneissie and the alluvial are by far the most Rocks form three great portant in point of extent, the former having onen; its greatest width in the southern, the latter in the northern part of the area. 'The aggregate area of the other two series of rocks above alluded to, which represent the Lateritie and the Upper Gondwana series,.is small when compared with the gneissie and alluvial areas. | | The geology of the Eastern Gháts will not be touched upon in this memoir, as they have already been described in sufficient detail in Mr. King's memoir “on the Kadapa and Karnul rocks,"? of which the several ranges of mountains and hills forming the eastern mountain barrier in this part of the peninsula are composed. 1 The Kistna district, as now constituted, includes the old collestorate of Guntür and the southern part of the old Masulipatam collectorate, the northern talugs of which were added to the Godävari district. | 2 Memoirs of the Geological Survey of India, Vol. VIII. In this memoir Mr. King has embodied the results of the examination of the Kadapa and Karnuljrocks by himself, the late Mr. Charles ZE. Oldham, and Mr. В. B. Foote. Cia) INTRODUCTORY, 3 Of the detached mountains and hills which occur scattered at inter- i ` -— vals over the inclined plane below the Gháts, the De s following are most important, taking them from south-west to north-east :— 1. Тһе Byrawudi Корда capped by an outlier of the u quartzites ; from 2,000’ to 2,200’ high, . The Gogulapalli hills. . The Picherla Konda, 1,283’. . The Kanagiri (Kunnigherry) Drug hills, 1,500’ to 1,600’. . The Podile (Poudellah) hills, 2,183’. . The Chendalur hill, 1482’. . The Chimakurti hills, 2,091’. . The Kondavidu hills, 1,680’ (1,749). Of less importance geographically are the Chundi (Soondy) hills, in the сото Ov gi» wh south-eastern corner of the gneissic area, but they are of far greater geological interest than most of the loftier masses. Kotappa Konda, south of Narasaraopett, a noted place of pilgrimage, is a bold mountain- ous mass of granitoid gneiss, 1,588’ high. Lastly may be mentioned the Bellam Konda, another granite gneiss hill 1,500 or 1,600 feet high, remarkable for its fine shape, and crowned by an old native fort of great strength. A large number of hills, including the Kanagiri Drug hills and several others of considerable height, have been entirely omitted from the Atlas Sheet 76, which thus represents the Kanagin taluq as flat, instead of very hilly. Of the rivers there is little to be said; their headwaters rise mostly in the east side of the Vellikondas. This is the case with the southern rivers, the Man-eru? (Mun Air), the Pal-eru (Pall Air), and the Musi-eru (Mooshee), and also with the Chilkalurpadu river in the northern part of the area. The Gundla- kamma, which drains the central part of the area, rises beyond that range, in the Nalla-malla? (Black-mountains); while the Kistna touches Hydrography. the area only with its very lowest reaches. ! Konda is the Telugu word for mountain or hill. ? Eru (air) is Telugu for river. 3 The western range of the Eastern Ghàt barrier, the Vellikondas being the eastern range. (9) 4 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. Except the Kistna all these rivers receive their principal supply from the north-east monsoon rains ; but the south-west monsoon rains appear to be heavier than in the latitude of Madras. The north-east monsoon is, on the contrary, lighter in this part of the peninsula than further to the south. Owing to the greater width of the peninsula, the westerly winds acquire a considerably higher temperature than at Madras, and begin to blow earlier in the season. The several geological formations occurring in the area under con- à ; sideration may be conveniently grouped as follows, Geological formations. , f in descending order :— 7. Soils and subaërial deposits. Recent ... < 6. Alluvia, marine and fluviatile. 5. Lateritic, sand and gravels, Tertiary... 4. Cuddalore (Rajahmundry) sandstones and conglomerates. Pavulur and Chebrole sandstones. Mesozoic ... 3. Upper Gondwana à 2 "Vemáva hales. SESS. Vemävaram shales Budaváda sandstones. 2. Kadapa Series. Azoic .. 41. Gneissic, or Metamorphie Series, and intrusive rocks included in 1%, In describing these several formations, it will be best to take them accord- ing to their respective ages in ascending order. Before passing to their detailed description, it will, however, be well to give a few notes on the geological information that had been collected and published prior to the Previous observers. geological survey. The earliest known geological notes on our area are those published by da man D Heyne! in 1814, in the maps aecompanying which he attempts to show roughly and partially the “geological situation” of the country. His geographical basis 1 Tracts, historical and statistical, on India, by Benjamin Heyne, M.D., F.L.S., &c., ќе. . Naturalist on the Establishment of Fort St. George, London 1814. Of the two maps given, the first is far the more correct in its delineation of the rivers and mountains along the, eastern coast. (77450) INTRODUCTORY, 5 being a very imperfect one detracts greatly from the value of his obser- vations on the eastern coast. The greater part of the country he shows to be covered by cotton-soil through which protrude hill ranges of syenite, basalt, gneiss, mica slate and clay slate. He failed to recognize the geological distinction of age between the | gneiss and overlying rocks. The notes which have most reference to our area are those in Tract XIV, which gives “a short description of Buggelconda hill, near Inna- conda, in the Guntur Cirear," a hill which then and since has by many been held to be an extinet voleanie cone, in some way connected with the frequent earthquake shocks which then and now stil agitate the Ongole country. He describes the highly hornblendie rock as a basalt, but in all other respects his deseription of the hill, in which he could trace no sign of voleanie action, is good. The earthquake shocks seem to have been both more frequent and more severe at the beginning of the century than of late years. He disposes also of theidea that Innaconda (Vinukonda) hill 1s of voleanie origin. In Tract IV he talks about minor ranges given off by the Eastern Gháts, “ which with very little interruption or variation of their consti- tuents approach the eastern ocean not far from Ongole;” “to these ranges belong several remarkable hills, as the Chicola,! celebrated for the frequent earthquakes originating at it and spreading over the country; and the Ongole hill, which seems to influence the compass in an uncommon degree." This statement is perfectly wrong as to the extension of the ranges eastward. No such ranges exist; the different hill ridges which are dotted about are quite unconnected, and in every case have a more or less northerly strike, as shown 1n the accompanying map ; and owing to this the changes in petrological character are marked and frequent. Dr. Heyne's error is accountable for only by reason of his rapid rate of 1 2 Chimakurti mountain, QT) 6 TOOTE: GEOLOGICAL STRUCTÜRE OF THE EASTERN COAST. travelling and by his having been misled by the very incorrect maps which alone then existed. Of the several localities he mentions where copper ores exist, none appear to lie within the limits of this memoir: Agnigundala, or Agni- condalah (as he calls it), lies among the Kadapa rocks; Ganypittah, and probably also Terrapally in the Pannur (? Pámur) pergunnah, are among the small copper mines (now worked out) lying within the limits of sheet 77. These are described in the first of a series of papers on the mineral resources of Southern India published by Captain Newbold in the seventh volume of the Royal Asiatie Society’s Journal, p. 151. Benza's “ Notes, chiefly geological, of a journey through the Northern 2 Cirears, " commence at Masulipatam beyond the Vip a Kistna, and treat entirely of the country to the northward. Newbold’s “ Notes, principally geological, across the Penin- sula of Southern India, from Kistnapatam to a len Honawar,"? just skirt the southern edge of sheet 76 of the Indian Atlas; whilst the same author's notes on the road from Masulipatam to Goa ? treat of the country between the coast and Hyder- abad, north of the Kistna river. The next reference is to be found in a paper by the author, on the distribution of stone implements in Southern Mr. R. B. Foote. : India, read before the Geological Society of London in 1868.4 The occurrence of implement-bearing gravels at considerable elevations in the neighbourhood of Pamur and at Nandanavanam, | at foot of the northern end of the Vellakonda range, was mentioned, and the situation of the beds illustrated by a seetion comparing their position with that of laterite beds in the Madras area held to be of the same age. Some of these observations were republished in the author's memoir of the Geology of the Madras area.* 1 Mad. Jour. Lit. Sci., V, p. 43. 2 Journ. As. Soc., Bengal, 1845, XIV, р. 398. 3 Id. 1844, XIII, p. 984. 4 Quart. Journ., Geol. Soc., London, Vol. XXIV, 1868, p. 484, * Mem. Geol, Survey of India, Vol. X, p. 52. (26). 22 GNEISSIC OR METAMORPHIC SERIES: geese 7 In the annual report of the Geological Survey of India for 1868,1 M Dr. Oldham mentioned that Mr. C. Ж. Oldham, Deputy Superintendent, Geological Survey of India, had in that season recognized the occurrence of sandstones and shales of supposedly Räjmahäl (Jurassic) age at Tangellamudi (Tunglamoo- dy) near Bezwada, and at Inkolu and Razpudi on the Guntur-Ongole road. The striking lithological resemblance of these rocks to the Sriper- matur beds near Madras was specially pointed out. The latest publication containing geological information touching : the area under notice is the Nellore District Mr. J. A. Boyle. ; Manual? the geological chapter of which, drawn up by the late Mr. J. A. Boyle, w.c.s. (in part from notes supplied by the late Mr. C. Ж. Oldham) refers, but only very briefly, to the geological formations in the northern part of the district. The shales and sandstones observed by Mr. C. Ж. Oldham at Razpudi, on the Guntur-Ongole road, and supposed by him to be of Rájmahál age (a supposition since confirmed by the discovery of numerous fossils), are mentioned. CHAPTER II.—THE GNEISSIC OR METAMORPHIC SERIES, The disposition of the various members of this oldest series of rocks in Southern India is, on the whole, less regular, and shows evidences of greater local disturbances within our area than in some of the more southerly districts, e. g., the Madras (Chingleput), North and South Arcot, Salem, and Trichinopoly districts. In Madras and North and South Arcot especially there is a remarkable degree of parallelism between the strike of the great bands of gneiss and the coast line as at present con- stituted. In the southern half of our present area this parallelism may be detected, but only on a much smaller scale and over short distances, 1 Published in the Records, Geol. Survey of India, Vol. II, 1869. 2 Written in part and edited by the late Mr. Boswell, w.c.s., Collector of Nellore. (ET 8 FOOTE : GEOLOGICAL STRUCTURE OF THE EASTERN COAST. while in the northern it is entirely wanting, and the prevalent strike of the rocks is more nearly vertical to the present coast line than parallel toit. Petrologically there are many important differences between the Differences from south- Sneissic rocks of our present area and those of the ern gneissic rocks. gneissic districts above alluded to. Тһе facies of the northern gneissics is newer than that of the southern rocks; they have, on the whole, been less metamorphosed, but more disturbed by flexures, or else the way in which the disturbing forces affected them has left more distinct traces. The most striking petrological difference, and that which especially ҚЫСА quos gives rise to the newer-looking facies of the northern gneiss. northern gneissics, is the presence of numerous important beds of quartzite interstratified with the older-looking highly crystalline varieties. These quartzites are so singularly like the typical beds of Kadapa quartzites, occurring in the overlying Kadapa system in the Vellakondas and as outliers capping gneiss hills, such as the Bairawudi Konda and (outside the limits) the Udayaghiri and Korisi Konda, that they immediately suggested the idea of their being really members of the newer series, let down into their present positions by a series of remarkable parallel longitudinal faults. It was only after long-continued exploration of their relations that the conviction forced itself upon me that they were truly members of the gneissic system. My reasons for arriving at this conclusion will be stated further on. سسس‎ Another point of difference 1s the greater development, in proportion to the general area, of highly micaceous and chloritie schists in the north- ern area, which for convenience I will call the Kistna-Nellore area. A third point of difference may be observed in the large develop- ment of syenite-gneiss, of much more intensely hornblendie form than is common anywhere in the more southern parts alluded to. Fourthly, may be mentioned the occurrence of considerable beds of mica schist highly charged with staurolite and kyanite crystals, a rock (782) GNEISSIC OR METAMORPHIC SERIES, 9 as yet quite unknown among the southern gneissies. These miner- als occur together in some of the beds, in others they are found separately. Much as the sub-division of this vast series of rocks would be ! UNT desirable, if it could be carried out safely, the Stratigraphical sub- . А STO division of gneissic series results of the survey would not in this case justify not eit it at present, as the general stratigraphical rela- tions of the principal groups of strata were not ascertained with sufti- cient certainty, owing partly to the disturbed and broken character of the strata, partly to the great extent and thickness of superficial depo- sits, and partly also the limited time which could be devoted to the work. In the absence of organic remains the sub-division of a great series of rocks must be based on well determined relations whether of infra or supra position of the several members of such series,. and where such relations cannot be ascertained, as in the case in question, no sub-division should be attempted. A much closer survey than that now carried out, and laid down on better maps of much larger scale, would very likely supply many of the data now wanting to correlate the different rock groups (to use this term in a merely geographical sense). At present as the relative positions of the principal groups remain to a great extent undetermined, it makes but little difference in what order they are considered, but for convenience in describing it may be as well to take them in a geographical order from south-west to north-east. ) | Although the gneissie rocks are not safely divisible into groups EE V ut based on well determined stratigraphical positions, sion into schistose and they can easily be classed in two divisions, PEN RES O bands. marked by great petrological differences. These two groups are the Schistose and the Granitoid. They occur distributed in several bands, of which four are principal ones and equally divided be- tween the two groups. The bands which are shown in the annexed sketch map, fig. 1, follow, on the whole, a north-east-by-north to south-west-by- south direction ; but the axes of the granitoid bands are by no means 90 10 TOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. everywhere coincident with the strike of the bedding, and appear to be axes of areas of extra metamorphism. The larger of the two prineipal granite gneiss bands occupies the centre of the gneissic area, and rhe E extends from a little south of the Paleru in Nellore areas, central and western. istrict up to and beyond the Kistna. The smaller or western band extends from the Kistna south-westward to the Gundlakamma river close to Vinukonda, its western margin being overlaid by the Kadapa rocks; but it is exposed in various inliers which will be separately referred to further on. Two other areas of granitoid gneiss of lesser but still considerable size occur, the one in the extreme south-west corner of the gneiss region deseribed in these pages, the other pretty nearly in the centre of this region. The former may be conveniently called d ee the Chandra Sekharapuram! area, after the large BEN village situated near its centre, and the latter the Chimakurti area, after the Chimakurti mountain which forms it. Be- sides these larger granitoid areas there are several quite small ones scattered here and there through the schistose areas, but they are not very well defined and of no special importance ; only a few of these will have to be mentioned separately. The schistose rock-area is divided, except at its southern end, into The schistose area and two great bands, by the main band or area of the bands, granitoid gneiss. The eastern of these bands, which extends northward from 15^ 15' to 16^ north latitude, consists mainly of hornblendie varieties of gmeiss; the western, on the con- trary, consists mainly of micaceous beds and associated quartzites, but includes some hornblendie beds in its southern part beyond the Gundlakamma river. The eastern schistose band contains also a few quartzite beds, and in several detached groups a number of beds of mag- netie iron, some of which are of considerable richness and possibly of future importance economically. 1 Chendra Shakrapoorum of sheet 76. (. 10 |) SCHISTOSE AREAS, 11 Тһе boundaries between the two rock varieties are in every case Boundaries of areas Obscure; where seen in close proximity, they pass ill defined. into each other by hardly perceptible gradation, as might be expected of rocks of which some parts have from various and as yet unexplained causes undergone more extreme metamorphism than others. Within the granitoid areas, different beds have under- gone varying degrees of metamorphism ; and in some, indeed in many cases, this would appear to have been influenced by the texture of the rock, for it would seem very probable that the metamorphism has advanced directly in proportion to the coarseness of the original | materials of which the old sedimentary rock was composed. "This sug- gests theidea that these areas of extra, or apparently extra, metamorphism may have been caused in part, if not entirely, by the peeuliar distribution of the coarser sedimentary materials of which these rocks were originally formed. The peculiarity of such distribution must have depended on the strength and direction of the currents then in action, and on the nature of the materials yielded by the yet older rocks then undergoing denudation. The hypothesis that the granitoid areas coincide with areas of deposition of coarse sediments, helps to explain the difficulties (which would otherwise be hard to deal with) in accounting for the occurrence of small isolated patches of granitoid rocks among the true schists, and similarly of schistose beds within the granitoid areas. 1.--ТнЕ SCHISTOSE AREAS. Beginning, as before proposed, in the south-west corner of our area, we find at foot of the Vellakonda mountains, in latitude 15° north, a very wild rugged traet of country made up of mica schists, with a few beds of quartzite. To the west this miea schist appears to be underlaid by the newer The great Vellakonda Kadapa rocks; but this appearance is due to a Eu. great line of fault by which the younger rocks have been thrown down, so that they abut against the gneissie beds. (u) 12 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. "This fault, however, does not everywhere form the boundary between | the gneiss and the Kadapa rocks in this corner, as after following the fault for about 25 miles northward, the boundary line diverges froni it and follows the eastern side of a patch of the basement beds of the Kadapa series, which here rests on the gneissie mica schist." Further north, however, the faulted boundary recurs and extends with only one break up to the Kistna. The most westerly set of шіса schist beds, which may be called the Gotlabailu (Gotlabyloo) series, forms several low but very rough jungle-covered ridges which coincide with the strike of the bedding. The quartzite beds above referred to run in a line nearly north and south through the centre Gotlabailu mica schists. an ran а of the mica schist area. The quartzite becomes ites. in places so micaceous as to pass into true mica schists. These beds may be conveniently called the Sitárámpuram (Sheetarampoorum) quartzites. The second set of mica schists forms the base of the Bairawudi Ti KOM Konda, a considerable mountain mass capped by the schist series. largest and most important outlier of the Kadapa rocks eastward of the Vellakonda mountains. The mica schists are well seen along its flanks, and the unconformability of the two rock series is very conspicuously displayed in the great ravine at the south end of the mountain, after which the series may be suitably named the Bairawudi Konda schist series. Тһе prevalent colours of these schists are dark greenish-grey and a dull dirty yellowish- brown. Quartz veins of small but irregular thickness are extremely numerous here asin most mica schists, and the surface is extensively and thickly covered with quartz debris, which frequently hides the rocks over a large extent of ground. 1 The fault is apparently continuous along the base of the Vellakondas, the western side of the patch of basement beds being itself faulted unconformably against younger i beds of the same series. (12) SCHISTOSE AREAS. 13 A couple of miles eastward of the Bairawudi Konda the mica schists CHennampalle (асык dip under а narrow band of quartzites (the Chen- ites. nampalle quartzites) separating them from the most westerly band of granitoid gneiss which I have above called the Chandra Sekharapuram band. Eastward of this gneiss again lies another set of quartzites which have a westerly dip, and appear to form with the former a synclinal basin, including the granitoid rocks. East of this second quartzite band comes in an extensive series of schistose beds, mainly micaceous, which may reasonably be regarded as the representatives of the great mica schist series forming the base of the Bairawudi Konda. These eastern mica schists occupy a wide valley Mica schists of Irur 22 which lies the village of Irur (Eroor), after | valley. which this valley and the schist series may be ^ conveniently called. The surface of the schists is greatly hidden by large spreads of cotton-soil in various parts of the valley. To the Gogulápalle quartzite east this valley is bounded by an important band ridge. of quartzites, which may be traced some 20 miles northward from the southern boundary of our area, and which rises into ridges of considerable height (500'-600') near Irur (Eroor) and near Goguläpalle. This quartzite ridge, which I will designate as the Goguläpalle (Gogoolpully) ridges, represents very probably the Sitärämpuram quartzite ridge first referred to, towards which it dips. It is in its Pémur schistose series turn underlaid by another schistose series which and area. includes locally several other sets of quartzite beds. To this schistose series the name of Pämur series may be appropriately given from the important village of that name lying some four miles north-east of Irur. The stratigraphical relation of the several quartzite ridges and schist valleys just named seems clear, within the region occupied by the head waters of the Man-eru; but northward of Gogulápalle and eastward of Pámur, the extension of these several rock series cannot be followed with any certainty. The continuity of the formations is interrupted, mainly by great (as) 14 E.byN. Batrcavudı Konda W. by S. FOOTE ; Vella Konda Me GEOLOGICAL STRUCTURE OF THE EASTERN COAST. Fig. 2. Section from near Sitarampuram to Pamur; 24 miles, accumulations of soil and vein quartz debris | furnished by the mica schists. Great and sudden changes of strike appear to indicate important faults, though none were actually traced, because the rocks are so insufficiently exposed. The annexed diagrammatic section illustrates the stratigraphy of the gneissic rocks just described be- Pámur Section. i tween Pámur and the foot of the Vellakonda range. Eastward of Pámur hornblendie schists and true gneiss become very common, and mica schists less and less common, as you proceed eastwards, and the strike of the beds varies greatly owing to great bends and contortions. The contortions are specially well seen in the quartzite ridges among the Chundi (Soondy) hills and the Picherla Konda. North of the Chundi hills the strike of the beds is generally pretty con- stant and not interrupted by many con- tortions. The prevalent direction is north- easterly. The micaceous schists in this region are, as a rule, much less mieaceous than those west of Pämur, and they often pass into true gneiss or into quartzite. The most important set of quartzites in Picherla Konda quartz- this region are those ites. forming the Picherla Konda, a bold ridge of hills rising on the crest of the water-shed between the Man-eru and the Pal-eru. "The main ridge forms the SCHISTOSE AREAS. 15 east side of an acute synelinal fold, the terminal eurve of which takes place at the southern end of the ridge, close to the trigonometrical station which crowns the highest summit. The synclinal opens to the north, and the ridges die away and disappear under superficial accumulations near the southern limit of the eastern granitoid gneiss band, which the quartz- ite beds appear to underlie. No connection was traced between these Picherla Konda beds and any of the other quartzites. Their position relatively to the great eastern granitoid gneiss band, which they have every appearance of un- derlying, suggests the possibility (if not the probability) of their repre- senting the quartzites underlying the Chandra Sekharapuram granitoid band. Of the quartzites in the Chundi hills only one set requires special R uS notice because of their extent. "These form the Chundi trigonometrical station hill, immediately north of the village of Chundi. Here as well as in the Picherla Konda the beds are many hundred feet in thickness. Another set of quartzites shows in the flat jungly country south- | > west of the Chundi hill group. They form two Ianekotai quartzites. e i 1 small but striking hills, one at Ianekotai (Iana- cotah), the other at Iawarpalle. They are of much smaller thickness than the Chundi hill beds, but like them rest upon a great series of highly hornblendie schists, and it is very probable they both occupy the same horizon in the general succession of the gneissic rocks. Overlying the Chundi quartzites are mica schists which in many Staurolite beds of parts are very thickly crowded with prisms of lb Eon, oe staurolite (staurotide), in others again with crys- talline masses of pale blue kyanite (disthene), or with both minerals mixed up confusedly. "The mica schist charged with both these minerals is best seen on the Mala Konda, the south-western extremity of the Chundi hills. It there rises into a considerable hill with a craggy sum- mit, crowned with a small Vishnu temple, one of the most frequented (1) 16 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. shrines in that region. It is one of the most pleturesque spots in the northern half of the Nellore district, some of the great masses of mica schist having fallen over and lying about in,wild confusion inter- spersed with trees and buildings. The great numbers of crystals which crowd the rock give it a very rough surface. The staurolite erystals are generally of large size, 2 to 3 inches long by 12 to 12 in width, but both larger and smaller sizes are met with. The large erystals are almost invariably of eoarse texture and much covered with a film of miea which conceals the true colour. Some of the smaller crystals have a fair amount of lustre, and are of dull blackish- red or reddish-brown colour. Twin crystals forming true crosses and diagonal (St. Andrew’s) crosses are common, especially at the Mala- konda, but not often of good shape. Along the west side of the Malakonda ridge and also at Kotapalle, two and a half miles north-west-by-north of Chundi, the sur- face is thickly strewn with the prisms weathered out. In various damp Of Kotapalle. spots the weathered crystals get covered with a shining ferruginous coat- ing and pass gradually into a species of lateritic concretion, as do also some richly ferruginous garnets in other parts of the gneiss country. The kyanite was seen only in the Malakonda beds, in which the crystal- line masses are nearly as numerous as the staurolite prisms. The staurolite crystals were met with in two other localities, one in a ay the ridge forming the extreme south point of the Chundi hills westward of Lingasamudram village, the other, a low hill, about two miles south-west Near Picherla Konda. ; of Picherlakonda Trigonometrical station. In the former case, the crystals are rather small and not numerous; in the latter they are large and very numerous, and the general appear- ance of the rock agrees with that occurring at Kottapalle near Chundi, except in the absence of kyanite. Several of the quartzitic beds in the Chundi hills are locally rich in tourmaline crystals. In a thin bed on the crest of the ridge west of Chundi, the rather Tourmaline in quartzite. Ко SCHISTOSE AREAS. 17 micaceous quartzite is crammed with prisms, mostly minute, of black. tourmaline. This is well seen where the ridge is crossed by the footpath leading direct from Chundi to Ramzlingapuram. A similar bed occurs on the east flank of the hill west of Lingasamudram, close to the staurolite bed already referred to above. Close by another bed contains very large coarse crystals of equally black tourmaline, several large lumps of which were brought to me by a native under the impression it was coal. Among the most noteworthy members of the schistose group are Magnetic iron and the ferruginous beds; these are of two kinds, the hematite schist beds. magnetic iron beds and the hematite schist beds. The former, though not so rich as the great magnetite beds in Salem district, &c., are yet deserving of much attention, both stratigraphically and industrially ; but the remarks to be made on their industrial import- ance will be reserved for the chapter treating on the economically im- portant geological features. Of the hematite schists only one example was met with which will be referred to more fully further on. All the more important magnetic iron beds of the gneissic series belong to the eastern schistose band; but several are found occurring in outliers away from the main band, the intervening space being occupied by the Two groups of mag- ewer sedimentary formations. They occur in two ES n е pon principal groups which I propose to call the Ongole kamma groups. and the Gundlakamma groups respectively, the former occurring near the town of Ongole, and the latter lying entirely in thelower part of the valley of the Gundlakamma river. Only one smal] bed was met with, which from its position could not be assigned to these groups; this occurs in a small inher of gneiss on the left bank of the Man-eru, 3 miles south-west of Singaraza Konda travellers? bungalow on the great north trunk road. "This magnetite bed forms ) . alow rocky ridge rising out of the alluvium and Shinampudi bed. Nn i ! lateritie shingle, and known locally as the Shinam- pudi Tippa. The bed is of considerable thickness, about three-quarters of a mile long, and moderately rich 1n iron. қ (7 18 FOOTE : GEOLOGICAL STRUCTURE OF THE EASTERN COAST. Fifteen miles to the north of Shinampudi (Sanampoody) Tippa he Ongolegroup of iron the most southerly members of the Ongole group beds. of magnetic iron beds, consisting of four beds forming a strong anticlinal curve round and through the mass of the Konijedu bed. Konijedu (Conjadoo) hills. The other members are the magnetite beds which form the main mass of the Ongole hills, and those forming the Parnametta hill, 4 miles north-west-by-west of Ongole. A glance at the map will show that, although separated from each other by considerable distances, the position of these three sets, or sub-groups, of magnetite beds, relatively to each other, is such as inevitably to suggest a direct connection between two of them, if not between all three. The strong petrological resemblance of the beds themselves and of the intermediate gneissie beds, as well as the near correspondence in number of beds, leads me to regard the Parnametta and Ongole beds as merely the extensions of the great Konijedu curve, the Parnametta beds corresponding with the northern, the Ongole with the north-eastern arm of the great anticlinal curve. A very trifling bend in the strike of the two arms of the curve would cause the extension of the beds, now masked by overlying lateritic and alluvial deposits, to coincide with the Parnametta and Ongole beds. It is, however, possible that the Ongole and Parnametta beds may form the arms of another curve (parallel to the Konijedu curve), the apex of which is hidden by the newer formations. I regard the first assumption as the more probable, and the more so as the Par- nametta beds resemble the Konijedu beds more closely than the Ongole beds. The Konijedu group consists of four prineipal beds, of which the two upper ones are the richest. They are fairly rieh as compared with the most important of the great magnetite beds in Salem district. The same may be said of the Ongole and Parnametta beds; they do not, however, appear to have been worked for iron ores, though largely used as rough build- ing stone. (221527) Konijedu iron beds. SCHISTOSE AREAS. 19 The number of beds at Ongole is five principal ones, though seven might be counted if two which contain thin Ongole iron beds. AN : gneissie partings be regarded as four several beds. The dip of the beds is from 40" to 55° south-south-easterly. Ав at Konijedu, the southern beds are the richest ; they form the low ridge abutting on the high road close to the shrine and Math of some Pir of great local sanctity. The northern beds, which form the Trigonometrical station hill, are much less rich in iron. The Math hill beds are fairly typical magnetic iron beds, in which the magnetite occurs in discontinuous granular Јат, interstratified between granular quartz lamine, which are generally stained of a reddish or brownish colour where weathering has taken place to any extent. The unweathered rock is generally of a purplish or blackish-grey colour. The non-ferruginous bands separating the different magnetite beds both here and at Konijedu consist of а quartzo-felspathic granite gneiss, in some places coarsely granitoid in texture, but showing in others a texture resembling woody fibre, which might for brevity be described as “ xyloid.” The Parnametta beds are four or five in number, have a westerly | dip of about 60^, and show signs of having under- Parnametta iron beds. 2 gone much greater disturbance than the other sets of beds to the east and south. А number of rude cleavage planes render the interpretation of the stratification by no means easy or satisfactory, The Gundlakamma group of magnetite beds consists of four sub- llama pam лор; separated from each other, except in one of iron beds. . case, by considerable distances, and showing no features of special resemblance on which to base any comparison between them. Still their general position, with reference to the adjoining members of the gneissie series and to the general strike of the bedding in that traet of country, indicates that they may very probably belong to one and the same horizon, though now broken into detached portions; it is quite probable too that they may be representatives of the Ongole group. ( 19:; 20 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. Of the four sub-groups, the southernmost, which lies close to the village of Burapalle, is the most important by far. Burapalle beds. А SUN Two thick beds, separated by a rather granitoid gneiss, here form a great part of a considerable hill. They appear to be connected with the magnetite beds in the Tammaveram hill, 2 miles to the north-north-west; but thick soil and talus accumulations eover the outerops of the beds for about a mile intermediately. Three beds are here seen. They are violently contorted, and appear to be cut off by a fault to the east; to the west and north they disappear under the alluvial deposits of the river valley. About a mile and a half to the north-north-west is another magnetic l iron bed which forms a small but very rocky hill Manikesvaram beds. د‎ Я Ў just west of the village of Manikesvaram (Mau- nikeswarum). This bed, which forms a very acute bend at the north end of the hill, is exposed for little more than a mile 1n length, its ends disappearing under the soil. It is underlaid by massive banded horn- blendie gneiss, and apparently overlaid by pink granitoid gneiss, but the contact with the latter is not seen, there being a considerable space between them in which no rock is exposed. This forms the second sub- group of the Gundlakamma group. The third sub-group consists of five‘ beds, which form the main rdges of two hills a couple of miles north of Addanki. Three of these, which are moderately rich in parts only, form the Singirikonda, a hill rising about 400 feet on the east side of the great north trunk road. The beds dip east at a high angle. In the lower hill west of the road are apparently two beds, but Singirikonda beds. their lie 1s very obseure, and they may possibly represent but one bed crumpled into an aeute synclinal trough. About 4 miles north-west-by-north of the Singirikonda group lies cpm another small group of four rather unimportant beds ; their relations to each other are not clear, the two northern beds running at right angles to the strike of the two southern ones. If, as the map suggests, they are parts of a great curve, ( 20 ) SCHISTOSE AREAS. 21 ` he two northern ones have probably been inverted by faulting of the strata. These beds form the Vemparala group. To the eastward of the Gundlakamma magnetite beds, the band of schistose gneiss is greatly eneroached upon by the granitoid gneiss baud, leaving but a very narrow belt of the former between the granitoid band and the younger sedimentary rocks fringing the coast. The schists occupying this narrow belt are mainly hornblendic, but are of no special interest. Two or three rather poor beds of magnetie iron occur within the limits of the granite gneiss band west of the Gundlakamma, to the east of Purimetla (Pooreemetta) tank, about 4 miles west of the last-men- tioned set of beds near Vemparala. But for their exceptional position, these beds offer nothing of any interest. It was mentioned above (page 17) that one instance had been noted Hematite schist near Of the occurrence of hzmatitie schist in the gneissic Cridi. region. This was seen in the Chundi hills, west of the village of Polenane Cheruvu, where several hematitic beds occur associated with a great thickness of micaceous clay and flaggy schists. One or two of the hæmatitic beds are rich, the remainder poor and of no importance. Their northern and southern extensions are lost sight of in the extensive low jungle which covers all the ridges in that quarter. Before concluding our references to the eastern schistose band, we ge ES must consider three other sets of quartzite beds which belong to it. Two of these are associated with the Chimakurti outlier of trappoid, or intensely hornblendic, granite gneiss above referred to (page 10), and lap round it as if under- lying it conformably and forming the walls of a surrounding synclinal ellipse. Of the real infra-position of these quartzites to the hornblendie beds, there can be no doubt on the western side of the Chimakurti hills, but on the south-eastern and northern sides the quartzites are not seen, but are either wanting or obseured by talus deposits. They show well for about 4 miles on the eastern side of the ellipse, but are inverted at ( 8) 22 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. the southern extremity of the hilly ridge they there form. If continuous, these beds would have to be considered as of one and the same set. The third bed is seen in a tiny inlier in the alluvium of the Man-eru, 5 miles south of Kandukur. It forms but a small rocky hillock abutting on the bund of the great Massávaram tank, but is worthy of special notiee, as being perfectly different from the ordinary quartzites of this region. The rock is a very glassy large-grained quartzite of delicate pale aquamarine-green, banded Massávaram quartzites. with deep purple. The purple bands oceur in discontinuous laminz, looking much like so many stains. If cut and polished it would be a stone of quite extraordinary beauty for decorative purposes. "The hill is formed by a boss on the outerop of a bed which has a steep dip to the east-by-south. Another mass of quartzite of glassy texture and pale aquamarine-green was noted in the jungle near Balla Venkatapur, on the west side of the Pieherlakonda quartzite curve. Among the less common forms of schistose gneiss to be mentioned are Taleose and chloritie the taleose and chloritie beds, which occur in various (eru parts of the schistose area, but especially in the western schistose band. Nowhere, however, do they form an important feature. The greatest development of the taleose rocks is to be seen near the head waters of the Mushi river, immediately east of the boundary of the Kadapa rocks, and at the villages of Gazulapalle (Gazoolapully), Chinna Managundum, and Soutapalem. They occur in considerable quantity also to the north of the Musi river, between Chendalur hill and the Kadapa boundary. Taleose rocks were also observed to the west of Vinnukonda, near the villages of Bharatapuram and Naddigudda. The rocks are generally taleose schists. A coarse gritty variety of talcose gneiss occurs along the boundary of the Kadapa basin near Chittapuram to the northward of Vinnukonda. Chloritie schists are found chiefly in the southern part of the schistose area, near the villages of Narapareddypully and Malareddypully, 8 or 9 miles south-west-by-south of Kanigiri. They here occupy the place that should be filled by the extension of the western branch of the S) SCHISTOSE AREAS. E great quartzite series forming the Picherlakonda (page 14). The quartzites have disappeared apparently by change of mineral character, a Rocks masked by de- phenomenon they often show, but in this case the [xs Gh wenn quain. probable connection with the chloritie schists cannot be traced owing to the enormous amount of quartz debris which covers the entire surface of the country hereabout. The immense quan- tity of this quartz debris, which is derived from the small quartz veins traversing the various schists, shows how great the denudation the schists have undergone in recent times. The chloritie rocks are exposed only in a few well-sections. Crystalline limestones are developed to a very small extent in the TE Nellore-Kistna gneissic area, and were seen but in Crystalline limestones. a few places. Of those noted, the most important occurs at a little distance south-west of Chundi. It occurs on the western slope of the ridge west of Chundi valley, where crossed by the path to Ramalingapuram. Тһе limestone, which is of grey colour speckled with small seales of mica, and of saccharoid texture, forms a bed 4 to 5 feet in thickness. The outcrop was not traced to any great distance, owing to its being covered by scrub jungle and talus of the higher lying quartz- ite beds. This limestone contains slight traces of copper in the shape of copper pyrites. Numerous fragments of crystalline limestone of pale grey and pink Near Pedda Allavala- Colours were observed on the high ground about 3 pad. miles north-east of Pedda Allavalapad, but the bed whence they were derived was not seen. Probably it is hidden by the extensive spread of cotton soil which there covers the face of the country. This limestone contains numerous crystals of black tourmaline and some small irregular inclusions of dirty white weathered felspar. Precisely similar tourmaline erystals are very common in numerous small irregular granite veins which traverse the hornblendic granite gneiss near by. A bed of sub-erystalline whitish limestone occurs east of the village of Pedda Arikatla, 8 miles north-by-east of Kani- gin. But little is seen of his bed owing to a В) Near Pedda Arikatla. 24 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. thick deposit of red soil which overlies it. Traces of a similar limestone were also noted to the south-west and east of Irslagundum, a village about 2 miles to the west-by-north. Both these beds belong to the | western schistose band. A small but very remarkable bed of impure crystalline limestone occurs at Punugodu (Poonoogodoo), on the right At Punugodu. 2 k bank of the small Makeru river 5 miles east-by- north of Kanigir. The bed forms an antielinal semicircular curve about 200 yards in length, the dip trending from west to nearly south-east, and the bed being from 6 to perhaps 10 feet in thiekness. In the purer parts of the bed the limestone is of bluish or greenish- white colour, but with it are interbedded many thick chert-like laminæ of a reddish mineral much resembling calderite (massive garnet). In parts of the bed these laminæ greatly exceed the limestone in quantity, With cleavelandite and aNd in others, especially near the base, the calca- epidote. reous laminæ disappear altogether. At the base the bed is very epidotie—epidote forming to a great extent the laminæ between the highly silicious limestone. Rather higher up the chert-like laminæ consist of a mixture of epidote (approaching to pistacite) and calderite. Near the southern end of the curve some of the partings have a perfectly granitoid texture, and look much like small granite veins injected between the planes of deposition; probably they are merely results of metamorphism. The two ends of the curve terminate abruptly, as if faulted against the adjoining granitoid gneiss ; but the junction is too obscure to speak with certainty as to the faults. No eastward exten- sion of the beds can be made out either in the bed of the river or on its left bank. A couple of miles down the river I observed several large | blocks of epidotie rock, doubtfully in situ, in the bed of the river close to the right bank opposite Patha Garlapetta. In texture and lustre the epidotie mass strongly resembled a green felsite. "They also resembled, though not very closely, the epidotie base of the Punugodu bed. І have seen no limestone elsewhere which resembles the Punugodu bed, but calderite was found in connection with crystalline lime- ( 94 ) SCHISTOSE AREAS. 25 stone in the case of a small bed of that rock at Kalpatti, in Trichinopoly district !. A graphitic gneiss was noted in one place within our gneissic area. This place is situated at the northern end of the rocky ridge south-east of Amravati (Umarawutty), close to the village of Peddamadur and a mile south of the Kistna river. The gneiss which contains the graphite in form of spangles is a decomposed quartzose rock. Graphite. The last variety of gneissie rock which demands our special attention is one occurring only in the extreme north-eastern corner of our gneissic area; but it there occurs in considerable beds, and forms the main mass of the group of hills on the right bank of the Kistna, opposite Bezwada. This variety of gneiss is a rather fine-grained quartzo-micaceous felspathie schist, containing several accessory minerals, chief of which are garnets of small size, but occurring in great numbers, and a reddish-brown felspar, forming small quasi-nodular aggregations, is also of common occurrence in the schist. The schist weathers considerably at the surface and becomes there of a rather powdery texture, or else coats itself with a close highly polished surface very like serpentine in appearance. This serpentinous mineral often shows various colours, and is then of considerable beauty. It is well seen on top of the westernmost ridge, south of the famous old Buddhist vihara at Undavilli,? also on the Mangalagiri hill. Bezwada group. The beds forming the Sitanagram and Undavilli ridges are continued across the river and form the Bezwada hills, in which they are also 1 Mem. Geol. Survey of India, Vol. IV, p. 275. 2 This very interesting relic of the Buddhist period is cut into this singular schistose rock at the north-western side of the extremity of the high ridge, close to the edge of the alluvium, and about half a mile from the right bank of the Kistna. The excavation was made without any reference to the strike of the rock, which it cuts diagonally ; owing to this the weather action on the detached pillars, &c., has been much greater than had they been eut square to the planes of bedding. ; The basement story of the vihara has lately been exposed by the very successful excavations proposed to, and carried out for, Government by Mr. Robert Sewell, M. C. S. An extensive rubbish heap had previously hidden great part of the front of the vihara, ( 25 ) 26 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. НЕ, displayed. They are so very distinet petrologically from any other members of the gneissic series known. to the geological surveyors in Southern India, that they should be recognised as a well-marked sub-group, to which Mr. King, Deputy Superintendent, Geological Survey of India, and I, have given the name of Bezwada Series, after the small but important town of that name where the Kistna is dammed back by the anicut, or weir, which forces the water to flow into the series of great irrigation canals traversing the Delta. The anicut is built of the stone quarried at the south and north ends of the Bezwada and Sitanagaram ridges respectively, between which the anicut is situated. At a cursory glance much of the stone used might be mistaken for a coarse and much weathered quartzo-felspathie grit, but closer inspec- tion shows its truly metamorphic character, and moderately weathered masses show the micaceous ingredient of the schist quite distinctly. In the southern half of Sitanagaram ridge the beds show a dip of 65° close to the bank of thecanal leading to Kommamur. In the Unda- valli ridge the beds appear to have a rather higher dip, which increases to as much as from 70° to 80° in the eastern spur of the Mangalagiri. a hill, in which the beds make an acute curve, from MET haue north by 5°-6° east, to east-by-north. The eastern limb of this great curve disappears after a course of about 14 miles under the alluvium of the delta. The greatest elevation of the Bezwada beds south of the Kistna is attained in the Mangalagiri hill, the tri- gonometrical station on the summit having an elevation of 889 feet above sea level Ав seen from this point the several ridges and hills forming the highest points of the Bezwada series show a very remark- ably level outline, as if the flat tops, speaking approximately, were remains of a former great plain of marine denudation. That such Plain of marine de. 2 plain did once exist appears quite certain from nudation: Mr. King’s examination of the country north of the Kistna, where he found this feature largely developed. The remark- ably level character of the ridge tops, and their near approximation in ( 26 ) SCHISTOSE AREAS. 27 height, are very noticeable from the summits of the granite gneiss hills, further west. _ Despite this uniformity of shape, the combination of hills with the large spreads of water in the Kistna renders the neighbourhood of Bezwada decidedly picturesque ; especially so are the views from the top of Mangalagiri and the Undavilli ridge. The connection between the Bezwada series and the adjoining granite gniess to the west and south-west has not been made out, no section having been found showing them in contact, or even in moderate prox- imity. As seen on the right bank of the Kistna, they appear to overlie the granite gneiss, and they certainly present a newer facies, as if they had undergone a lesser degree of metamorphism. Considerable beds of quartzite are associated with the Bezwada beds in the main ridge north of the Kistna, but not a trace of them was noted on the south of the river, and I think the quartzite must have passed into the micaceo-felspathie schist, as it so often passes, in the south of our metamorphic area, into true mica schist. We must now return to the western band of the schistose gneisses, The western schistose Which includes the northern extensions of the gneisses. several micaceous, hornblendic, and quartzose bands described above as forming the various ridges and valleys lying between Pámur and the eastern base of the Vellakondas. The band of granitoid gneiss to which I gave the name of the Chendra Sekharapuram band, disappears to the northward of the village of Immedesheruvu, under the alluvium of the Pal-eru and the thick spread of red sandy soil, stretching away to the north-east. "The rocks which then come into sight north of the red soil plain all belong to the schistose group, which then occupies the whole area between the lofty granite gneiss masses of the Kanigiri (Kun- nigherry) and Podile (Poudellah) hills, and the northern extension of the Vellakondas. Mica schist is the predomi- nant form of schist in this region, and next to it quartzite, which forms many long and important hil ridges. Hornblendie schists are ( 27 ) West of the Podile hills. 95 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. but rarely seen here, though so common in the eastern schistose band. The mica schist gives rise to only three hill groups of any im- portance ; they are, the group of hills south of Mogalur (Mogulloor), the Kodnikonda south-west of Narrava Gopalpur, and the great hill north- east of Nandana Marrila (Nundana Morrala). Except where hilly, very little rock is seen in this quarter, the greater part of the country being covered with red sandy soil of great thickness. It is only in well- sections that the softer schistose beds are, as a rule, seen exposed, for the nullahs do not cut deeply enough into the surface to expose the underlying rock. As already stated above (page 13), no connection could be traced with any certainty between the schistose rocks of the western band northward of Gogulapalle and those southward of that place ; there would, therefore, be no use in making mere speculations as to any possible equivalency of several similar beds, even of the most remarkable of the quartzites. Very little can therefore be said of theses chistose rocks, for they present no special features of interest, either geologically or economically. The several quartzite beds may be traced often for many miles, showing little change in mineral condition; but they then disappear, the ridges dying down gently under the surface soil, owing to the thinning out of the hard bands, or the graduation of the whole bed from hard quartzite into a soft micaceous rock. It will be seen by the map that an important band of quartzite extends from the valley of the Pal-eru up to and beyond the Musi (Mooshee) river; they are then lost sight of ; but ‘about 4 miles to the north- east another very remarkable set of quartzites forms the mass of the great Chendalur hill. They in their turn are followed a couple of miles further north by other quartzites in the Ubiahpalliam hills. These two last groups of quartzites are remarkable for being the most northerly development of this variety of rock in this region, and for the peculiar positions they occupy with reference to the other adjacent members of the gneissie system. (9007) SCHISTOSE AREAS. 29 The quartzite beds of Chendalur hill are so strikingly like those Gloss kill Gps of the newer Kadapa rocks, that, but for the ites. existence of the great band of unquestionably gneissic quartzites above referred to, I should have unhesitatingly mapped Chendalur hill as an outlier of Kadapa rocks. As it is, I look upon the Chendalur quartzites as belonging to the gneiss, but with a lingering feeling: of doubt, due, perhaps, to my having, when I first visited Chenda- lur hill, been strongly prepossessed in favour of its Kadapa age from its appearance as seen from the west and south-west. The quartzites are well exposed in a grand cliff at the south-western end of the ridge,! and the cliff recalls, though on a smaller seale, many of the grand precipices, scarping the Nagari mountains and the southern parts of the Vella- konda range. The Chendalur quartzites are seen to rest, at the northern end of the ridge, on granite gneiss, to which they appear conformable ; to the west they appear to dip conformably under a bed of iron-grey quartzose gneiss; while at the south-western end of the ridge the quartzites dip, also in apparent conformity, under the hornblendie beds close to the village of Bundevaleganla. Viewed as a whole, the Chendalur hill appears to be the southern extremity of a narrow and much contorted anticlinal ellipse. Тһе beds have a quaquaversal outward dip on all sides but the north. I failed in tracing any signs of faulting, which must exist supposing the anticlinal ellipse to be an outlier of the Kadapa rocks. Interealated between the quartzites on the back of the anticlinal is a thick bed of slaty argillaceous schist, which being much softer than the quartzites has been deeply eroded, and has thus given rise to the formation of a small but deep valley in part of the hill mass. 1 As seen from a distance from the west and east, the outlier of the southern end of Chendalur ridge presents a most striking likeness to a gigantic hippopotamus standing half immersed in water and looking south. This resemblance to an animal does not appear to have struck the natives, owing doubtless to their non-acquaintance with the genus hippo- potamus. à ( 29 ) 90 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. The quartzite beds in the Ubiahpolliam hill form likewise an Ubiahpolliam hill quart: anticlinal ellipse; they are overlaid conformably zites. by hornblendie sehists. No connection could be traced between these and the Chendalur quartzites; and, unless the respective positions of the two sets has been affected by faults hidden by the intervening thick spread of red soil, the Ubiahpolliam beds underlie the others at a considerable depth. The predominantly micaceous character of the schists noted to the Chanpe of character of southward of the Musi-eru is not continued north rocks. of that river: numerous hornblendic beds now come in, and by the time the Gundlakamma valley is reached, to the south of Vinukondas, the two varieties seem to be equally common. Northward of the Gundlakamma river the schistose band widens greatly, and makes a trend to the north-east, thereby maintaining a general degree of parallelism with the boundary of the Kadapa rocks. To the north and east of Narsaraopett (Atlur Narasaraopetta) the continu- ation of the schistose band is greatly obscured by the extensive and unbroken spreads of cotton soil which cover the face of the country general ; and ere it reaches the banks of the Kistna to the west of Amravati, it has diminished to a strip of only a couple of miles in width. Two oceurrences of gneiss containing epidote in the form of pistacite were noticed near Narasaropett; the one at Epidotic gneiss. BEN RR Pamidipadu, 4 or 5 miles north of the town, the other at Ravipadu, 3 miles to the north-west. Epidote is but rarely found in this region, though by no means an uncommon mineral in more southerly parts, in South Arcot for example. Only one bed of magnetic iron was observed in the western schistose band; it occurs at Yerraguntlapadu, 4 miles Magnetic bed. : g south-east of Sattenapalle. Very little is seen of the bed, which crops out a little to the north-west of the village, and is by no means a rich one, ( 30 ) GRANITOID AREAS. 81 9.—'TuE GRANITOID AREAS. To the granitoid group belong all the highest elevations in the gneissic area, e.g., the Kanigiri Drág hills, the Podile and Chimakurti hills; the Bogala Konda, the Yenema Konda and Yerra Konda (near ; Kommalapadu), the Kotappa Konda, the Kondavidu hills, and lastly the Bellam Konda. A great number of smaller hills also consist of granit- oid gneiss; but of these only four groups need be named, the Vinu Konda hills, the Darisi hills, the group around Rámákur to the north- east of Addanki, and the small hills around Nadella (Nadendla) to the south-west of the Kondavidu hills. Three severalforms of ternary granitoid gneiss were observed, the Varieties of granite hornblendie, the micaceous, and the epidotic, to Ешон. state them in the order of their relative importance and frequency of occurrence. Very frequently, however, the rock appears to be binary in its composition, only quartz and felspar being visible, and the closest search failing to show the third ingredient, unless a freshly fractured surface extending far beyond the weathered surface of the rock is obtainable. This is especially the case in but slightly micaceous varieties. Owing to this difficulty the exact nature of the rock was not obtainable in a large number of cases in which no quarries existed to afford really fresh sections, it being quite beyond the power of an ordinary geological hammer to break up the great rounded weathered masses sufficiently. In some few instances the granitoid rock contains both hornblende and mica, but then one or other of the two is to be looked upon as a mere accessory mineral. Though in very many cases beds of the two principal ternary varieties, the hornblendic and the micaceous, are met with together in alternate Areas of predominance Strata, one or the other generally predominates of hornblendie rocks. over a certain area ; such areas for the hornblendic variety are the environs of Pedda Allavalapad and Cherlupally, the (3) 32 FOOTE : GEOLOGICAL STRUCTURE OF THE EASTERN COAST. country east of Kanigiri, the Chimakurti mountain, the great spread of hornblendie, mostly trappoid, gneiss around Pothakamur (Pooth- kamoor), a very similar trappoid spread forming the Bogalakonda and adjoining hills, the Kotappakonda and the Ramakur hills. Another area of intensely hornblendie beds is found to the north of Guntür, in the Lam, Tadikonda, and Nirukonda hills. The micaceous variety occurs Areas of predominance principally in the Kanigiri and Podile hills, also in of micaceous rocks. the group of low hills on the border of the Darisi and Vinukonda talugs, and in the Vinukonda hills. In the Kondavidu hills and in those lying between Guntur and Amravati the two varieties occur together in about equal proportions. The epidotie variety forms the mass of the Bellamkonda in Satten- Ж 3 f apalle talug, and occurs also in the lower hills ex- ое tending northward towards the Kistna. Epidotie gneiss occurs also in the inlier north of Vinukonda, at a place half-way between Sarikondapalem and Vaddagunta. The most charaeteristie display of the granite gneiss in its scenic features takes place in the Kanigiri and Podile hills, and in the granitoid area around the Bellamkonda, in the extreme north of our area. In these localities are numerous ex- amples of “ blocky ” structure, whole hills appearing built up of loose Scenery. masses with slightly rounded angles. Perhaps the finest example of this is furnished by the Kanigiri Drág hill. The granitoid characters are rather less pronounced in the Kondavidu EIN hills, the bedding having on the whole been less The Kondavidu hills. ў 1 Bo obliterated than in the granitoid tracts last men- tioned. They form the most picturesque group of hills throughout the Guntür-Ongole region, especially as seen from the north-west or south- west. They must, however, yield the palm to the Kondapilly hills, 1m- mediately north of the Kistna, which appear to be formed of an extension of the same serles of beds. "The Kondavidu hills consist of two principal rdges, whieh do not correspond with the great lines of bedding, but ( 98 ) GRANITOID AREAS. 35 seem due to a system of jointing! ; near the middle of the western ridge, which is rather the lower, stands the well known old drág, or hill fort, now an extensive but most pieturesque ruin well worthy of a long visit by every lover of grand rock scenery. The western slopes of both ridges are much more precipitous than the eastern ones. The highest summit, called Sheikh Adam ka Pahar, after a Mussulman saint whose tomb stands close to the summit, attains the height of 1,680 feet above sea-level.” In some of the hollows near the summit there remain a few patches of thick jungle, showing that the present barrenness of the hills is due to human agency, and not to any barrenness of the soil. It is noteworthy that the three prineipal hill fastnesses in this quarter have been built on the granitoid hills; they are Kondavidu, just described, Bellam Konda, and Rani- giri. Of the three, Bellam Konda Drág was much the strongest by posi- tion, ав 16 occupies the whole summit of the mountain, whose flanks are Bellam Konda Drüg. very precipitous. Like Kondavidu it is very picturesque, and commands a most interesting view over the singularly faulted and broken eastern boundary of the Kadapa rocks. If the great northern peak of the Bellamkonda could be made accessible so as to get an unbroken view of the Kadapa boundary up to the Kistna, it would afford to the stratigraphi- cal geologist a panorama worthy of a long pilgrimage. Kanigiri Drüg is formed by the fortification of the crests of two high ridges enclosing a narrow valley running north-east i its northern, and south-east in its southern half. Very little remains of the old buildings except a couple of gates at the northern and south-western ends, but in olden times it was a place of great note, and the object of a good deal ! The time at my disposal was unfortunately so much curtailed by a serious illness, that I was obliged to forego the pleasure of climbing to the highest points on the hills, from which alone the real relation of the bedding and jointing on a large scale could be made out. | ? 'This is the height according to the measurement made by the officers of the Revenue Survey Department. The old Trigonometrical Survey measurement gave 1,749 feet. с (1189!) 34 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. of ies The upper part of the central valley hes in the strike of the bedding. Тһе extent to which the southern peak! has been eut up by jointing is quite extraordinary, and the natives say the peak 1s inaccessible from its extreme ruggedness. This statement is probably based more on their fears of some leopards which inhabit the caves on the hill, than upon the real inaccessibility. The most important and conspicuous mass of the hornblendie variety . is the Chimakurti mountain. As already men- E EIE tioned, the hornblendie rock here forms a great elliptical area, 8 miles in length and nearly 5 across its greatest width. The southern half of this ellipse is occupied by the mountain and its roeky spurs, which extend some distance into the northern half. At the extreme north end are two considerable and very rugged hills formed of the same intensely black rock. The whole mountain is very bare, and but little soil exists for vegetation to be possible, and the usual native recklessness has denuded the slopes of what little wood might grow there if any attempt at conservation were made. Despite the bareness of the mountain, the bedding of the rocks 1s not easy to make out, it being very obscure to begin with, and also much concealed by the great masses of loose and confusedly tumbled blocks which cover much of the summits and slopes. It is best seen by descending some distance | on the northern side of the summit, to where a good view is obtained of the great northern spurs. The southern elliptical curve is distinctly traceable here, though very obscure on the southern slopes. The rock is a coarse hornblendo-felspathie, and apparently quartzless, compound of dark greyish-black colour weathering to absolute black. The north-east faces of the various summits are all coated with grey and white lichens, due doubtless to the effect of the north-east monsoon. The highest summit, on which is a Trigonometrical Station, attains the height of 1 Į was unfortunately unable to get a guide to show the way to the summit, which must command a noble panorama, and had not the time to seek a path for myself through the bewildering chaos of large blocks amongst which no track can be made out from below. I think the summit would certainly be reachable from the north side, ( 84 ) GRANITO!D AREAS. 33 2,097 feet above sea-level, and commands a very extensive and fine view over both sea and land.! . The bedding of the rock in the northern hills is obscure in the extreme, but there can be no doubt as to the position of the underlying band of quartzites so well displayed in the low ridge south of Rámá- chandrapuram. The decomposition of this intensely hornblendie rock gives rise to the formation of much gravelly kankar (coneretionary tufa), a great thickness of which is to be seen near the bund of the upper of the two large tanks standing within the northern apex of the synclinal basin. The Chimakurti mountain is a great solid-looking mass, and possesses but few elements of the pieturesque as seen from a distance. It looks best from the south near Chillamkur. From the north or north-east the concentrie arrangement of the dark central mass of hornblendie rock and the surrounding quartzite beds is very distinctly seen. The soil covering the small area of approximately level ground close to the summit is of dark, nearly black, colour, a true humus in fact, and its existence proves that the mountain was formerly very much better wooded. than now. Tors, or isolated blocks formed by the action of atmospherie agencies ы оп rocks having a large spheroidal structure, or ors. e œ . much cut up by systems of jointing, are common enough throughout the granitoid areas above referred to, but none of very remarkable size or boldness of outline were noted, and in these respects they are greatly inferior to the tors of other granitoid regions, e.g., the neighbourhood of Adoni, in Bellary district, or the granitoid 1 If the Government scheme of creating a new Collectorate, with Ongole as its chief town, be carried out, the Chimakurti mountain will probably before long be built upon and usedas a sanitarium. "Though not of great height, dwellers on the summit, where there is plenty of room for several houses at an elevation of 2,000 feet, would pro- bably be out of reach of the terribly hot land wind. The absence of all jungle and the complete isolation of the mountain will also most likely prevent the formation of all malaria. The only real difficulty at first will be about the water-supply. ) 35 ) 36 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. country around and south of Gingi fort, in South Arcot, &e. Only one eroup of tors in the Nellore- Kistna country has impressed itself on my memory, and that through its grotesque shape. This group, which when seen from the south by morning light has a most striking resemblance to a huge elephant charging a gigantic tortoise from behind, lies a few score yards off the path leading from Nakri Kallu, on the great trunk road, to Gundlapalle, 4 miles to the north-west, and about half-way between the two places. It belongs to the western granitoid band. Of the small outliers of granitoid gneiss which, as before mentioned, Porphyritic rock at occur here and there within the limits of the Yikuru. schistose gneiss bands, only one needs special men- tion; it forms a low rocky hill close to Yikuru, 3 miles west-south-west of Narasaraopeth. The rock here is a very typical porphyritie variety of rich purplish-grey colour, which might be quarried and converted into a very handsome building stone. The enclosed crystals are prisms of grey felspar imbedded in a hornblendo-felspathie matrix of darker colour. The dark black variety of hornblendie granite gneiss which I have “Trappoidth” granite described as trappoid, occurs mostly in the central a part of our area, and chiefly in two patches north of the Chimakurti mountain, the larger lying to the east of Potha- kamur, in Daria taluq, the smaller some miles to the north, on the left bank of the Gundlakamma, and running up to and including the Bogala Konda, the reputed centre of the numerous slight earth- quake shocks that are experienced in the Ongole country. The typical trappoid rock shows nothing but hornblende and felspar, in a crystalline mass of varying degrees of coarseness. Quartz is very rarely seen in it. The colour ranges from dark blackish-grey to almost absolute black. Weathered surfaces are often absolutely black, and the rock, when seen in detached masses not large enough to show the bed- ding, is not distinguishable from bedded hornblendie trap. This ıs markedly the case at the eastern end of the Pothakamur trappold area above named, especially in the masses of rock seen about a mile to the north of At Pothakamur. ( 96 ) GRANITOID AREAS. 97 Thalur. Further west, on the higher ground between, Pothakamur and Mullamur, where huge surfaces of rock are exposed, the bedding becomes apparent, and a few rather less granitoid beds are interstratified with the ultra-erystalline beds. No section was met with showing this trappoid rock in actual contact with other varieties of granitoid gneiss, or with the schistose rock ; but nothing was seen suggestive of want of conformability with the closely adjoining beds. The true character of the < trappoidth " rock is best seen in the Bogala f Konda, or “Charcoal hill,” so called from its intensely black colour. This intense blackness, added to its nearly conical shape, and its supposed association with the slight earthquakes so fre- quent in the region around Ongole, has given rise to the idea that it is a voleanie cone. The hill itself presents, however, no signs of volcanic The Bogala Konda. action. The summit is divided by an irregular saddle into three unequal divisions, the western being between 100 and 200 feet the higher. The summit consists of a huge chaotie accumulation of blocks roughly rounded at the edges by weathering, from among which spring a few stunted trees of the fig tribe. The bedding of the great band of hornblendie rock out of which the. Bogala Konda rises can be very distinetly seen from the top, and traced by the eye for several miles southward. Petrologically these beds are identical in appearance with the Pothakamur beds, and they are very probably an extension of the same, though the actual connection was not traced. On the eastern side of the Bogala Konda are considerable “ serees,” to use a term familiar in the English lake district; the fallen blocks appear to form as it were streams down the sides of the hill. Many of the blocks are so loosely perched that a very small impulse, such as the slightest shock of an earthquake, would suffice to overthrow them; and to the frequency of earthquakes in this region may safely be ascribed the extreme confusion of the blocks on the summit, between the present position of the vast majority of which and the direction of the great joint planes to which they primarily owed their existence, no con- nection can now be traced. This cause has probably also affected the | (ey 38 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. loose blocks on the summits of the Chimakurti mountain in a nearly equal degree, the majority of the blocks being of very moderate size. The great confusion existing among the fallen blocks on the flanks of the south peak of the Kanigiri hills already referred to (page 34) is doubt- less also in a measure due to this earthquake action ; but from the vastly greater average size of the blocks, they would only be affected by violent shocks, which are of much rarer occurrence. The dip of the bedding of the Bogala Konda “ тарро” beds appears to be easterly, but is very obscure. A bed of ordinary hornblendie granite gneiss close to the western base of the hill has however a very distinet south-easterly dip.! I only experienced one earthquake shock while working in the Guntür-Ongole country, and that certainly did not emanate from the Bogala Konda centre. It was on the morning of the 11th March 1867, about 7 miles west of Vinukonda and 17 miles west of the Bogala Konda. The shock caused a distinet tremor of the ground, travelling from north- east to south-west, and was accompanied by a loud rumbling noise, lasting several seconds. Frequent slight shocks have been noticed by the officials and other residents at Ongole. Beds of equally trappoid hornblendic gneiss occur at a few other places as inliers within the schistose area, e.g., between Maddalur and Yanekepad, 12 miles west of Ongole; and again 8 or 10 miles further south-west, near Zarlapalem, on the path from Peddakandla Gunta to Tangella, and near the latter place. These latter beds are intercalated with beds of quartzose gneiss rolling in small anticlinals and synelinals. Similar trappoid beds, strongly simulating contemporary trapflows in their appearance, occur also north and west of Tangella and east of Byana- palem. Two other examples of intensely trappoid hornblendie rocks are worth mentioning. Both occur in the valley of the Man-eru, the larger ние and more important forming the Enemerla hill, 9 miles east of Pámur, the smaller forming a Bogala Konda is probably between 1,200 and 1,300 feet high, and extremely steep on all sides, I went up the south-west side and found it a very stiff climb, especially over the scree near the summit. The panorama from the top was very disappointing. ( 58 ) GRANITOID AREAS. 39 low rocky hill 63 miles further to the east-by-north. Enemerla hill is formed of bare black rocks rising to a height of from 300 to 400 feet above the surrounding plain. The quartzo-hornblendie rock is coarsely crystalline in texture, and towards the western end of the mass rather porphyritic; no bedding is seen, and its relations to the surrounding schistose gneiss is not clear, no contact of the two rocks being exposed. The Enemerla rock is not more trappoid than beds of crystalline horn- blende rock unquestionably forming part of the gneissie series, and but for its position miles away from similar extra-metamorphosed rocks, I should not have any hesitation in regarding it as such a rock ; but, as it is, I feel doubtful whether it may not be intrusive and trappean. The same remark applies to the smaller trappoid mass south-west of lanakotay (Tanacota), but in lesser degree, as it is unquestionably associated with a set of highly hornblendie gneiss beds underlying the rather remarkable Ayawarpalle- lanakotay quartzite band. The hill consists of great masses of rock, much weathered, rounded, but not sufficiently isolated and detached to be called tors. The hill is almost bare of vegetation, and of inky blackness. Ianakotay hill. It is not possible to examine the geological map of the Guntür-Ongole у region without being struck by the remarkable Parallelism of folding ‘ S t of gneiss and Kadapa parallelism subsisting between the great foldings BEN of the gneissie rocks and those of the Kadapa rocks in the Vellakondss and Nallamallas. The natural inference from this is that the gneissic series was affected by at least two great periods of (roughly speaking) east to west compression, and the Kadapa series by one such period, which was the second of the two. That the gneissic series had been compressed into huge folds at a period long anterior to the commencement of the Kadapa, is abundantly clear from the fact that such folds had undergone great denudation before the begin- ning of the deposition of the newer series. On the completion of the 1 My examination of Enemerla hill in 1866 was not as full as I could have wished, owing to a very heavy burst of the monsoon, which came up and drove me away, No subsequent opportunity occurred of my revisiting the locality. 9 40 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. Kadapa series came the second period of great east-to-west pressure, which erumpled the eastern half of the Kadapa basin into huge synclinals and anticlinals, several of which are locally inverted. These upheaved and eontorted strata were in their turn exposed to denuding agencies, and underwent considerable waste before the series of rocks, called by Mr. King! the Karnul series, began to be deposited. The completion of this series was followed by another period of disturbance and up- heaval, during which most probably a great fracture of the earth's crust took place a little to the east of the line of greatest contortion of the Kadapa rocks. This fracture formed the set of faults now seen to exist along the greater parts of the eastern boundary of the Kadapa basin, and coinciding nearly everywhere with the eastern foot of the Vellakonda range. This fracture was accompanied by great displacement of the rocks on either. side, and those on the eastern side were greatly upheaved. Sub- sequently to this, denuding agencies attacked the upraised area with intense energy and removed the superincumbent Kadapa rocks almost entirely, leaving in our area only a few outliers, namely, the Baira- wudi Konda quartzites and slates, and the same set of beds im the unconformable patch east of the great fault at foot of the Gali Konda (Gauly Conda) in the south, and the Biravallipaya and Atchammapett “ faulted domes 72 in the north-east. To the south of our area are the very striking outliers forming the Udayagiri Drüg and the Korise Konda. The Yerra Konda and Durgamma Konda, still further to the south, are doubtfully of Kadapa age. None of these gives any clue as to the limit of the former eastward extension of the Kadapa basin over the crystalline rocks. 1 The names of Kadapa and Karnul series were given by Mr. King of the Geological Survey of India to the two great series of submetamorphie rocks occurring in the Madras Presidency. * The northern end of the eastern boundary of the Kadapa basin is characterised by a singular series of elliptical anticlinal domes, six in number, extending from Vinukonda nearly up to the Kistna, Тһе two northern ones are true outliers faulted into the gneiss on all sides. (40) GRANITOID AREAS. 41 The members of the gneissic series are generally too coarse in texture Әрі” oem in 000 show true slaty cleavage well; moreover, it EROS mels generally coincides with the edges of the planes of deposition, and is therefore easily overlooked. For these reasons but very few instanees of it were noted, the chief of these being among the mica schists 7 miles to the.north of Pämur, and the slaty schists on the back of the Chendalur anticlinal (see page 29); but neither of e EG EI these are of any special interest. A case of some age in quartzite. interest, of cleavage in a quartzite bed, was observed in the western arm of the synclinal curve formed by the beds which make up the Picherla Konda (see page 14). Im this case one of the lowest of these beds (a httle to the south of the village of Balla Venkatapur), which has an east-to-west strike with northerly dip, is at the point where it curves eastward an ordinary quartzite, showing no special features; but as it extends westward it becomes cut up by vertical cleavage planes!, which become more and more numerous westward, and are lined with a film of greyish mica; the quantity of mica increasing with the number of cleavage planes, till, just.as the spur sinks down under the local alluvium of the adjoining nulla, the rock is almost an absolute mica schist. The intermediate gradations were instruetive, showing the progressive changes dependent on a very pecu- liar form of metamorphism. The half-way gradation had a strong general resemblance to a coarse “blotchy” porphyritic gneiss, which no one who had not seen the gradual change would be inclined to regard as the possible outcome of extra-metamorphic action on a true quartzite. ii A great show of cleavage of clayey mica schists may be seen in the Nandana Marrila hills north of Kanigiri. In this case the cleavage coincides actually, or very nearly, with the lamination of the true bedding. : 1 Тһеу might be mistaken for jointing where they first begin to show, but a little . further west they become far too numerous to be regarded as anything but cleavage planes. (4) 43 FOOTE : GEOLOGICAL STRUCTURE OF THE EASTERN COAST. 3.—INTRUSIVE ROCKS IN THE GNEISSIC AREA. All the intrusive rocks occurring within the gneissic area being, as far as at present known, of greater age than the overlying Kadapa rocks which they are nowhere seen to penetrate, it will be convenient to con- sider them before proceeding further. The intrusive rocks seen are referable to the following four groups: (a) Trap dykes and intrusions ; (0) Granite veins; (c) Felsite veins; (d) Quartz veins. a.— Trap dykes and intrusions. As compared with many other parts of the great gneissic region Weis off Kemper of Southern India, the Guntür-Ongole area is vocks. remarkable for the extremely small number of trappean intrusions that have taken place in it. Not only is the number of such intrusions small, but they are mostly of very small size, and in every way of extremely small importance geologieally. One exception may perhaps be made, but in this case the really trappean character of the rock is very doubtful; it relates to the black hornblendie mass of the Enemerla hill east of Pámur, already described above (see page 38). From its isolated position this mass suggests the idea that it 1s mtruded among the highly schistose beds which surround it; but unfortunately the contact between the two could not be traced. The petrological characters of the mass are so extremely like some of the other highly hornblendie metamorphie beds, that they do not afford, without special chemical or microscopical analysis, sufficient evidence safely to decide their exact nature. "The same remarks apply, though 1n lesser degree, to the equally doubtful hornblendie rocks of the Ianakotai hill occurring 6 miles to the east-by-north, near the village of Ayawarpalle. All the other unquestionably trappean intrusions occur in the form of true dykes. The greatest number of these oceur in the granitoid gneiss region around the Bellam Konda, in the Kistna valley. A few occur in the lower part of the Gundlakamma valley, and a small group 1n the upper valley of the (2407) Dykes. INTRUSIVE ROCKS IN THE GNEISSIC AREA. 48 Musi river; the remaining few are scattered about here and there at very great distances from each other. The dykes are all dioritie in character, and many of them, especially in the northern group, distinetly porphyritie, showing numerous whitish felspar crystals. Тһе diorite is mostly blueish or greenish black in colour. Exceptions to this rule are the dykes occurring at Ongole (on the Trigonometrieal Station hill), and Neputlapadu, 3 miles east of the southern end of Chimakurti mountain. The former is of greenish-whitish grey colour, the latter of a purplish-blackish grey. The majority of the dykes may be referred to two systems, depend- SESS trolleys: ing upon the direction of their strike, one of deus. these systems having a course from north-15°-west to south-15°-east, the other running north-east-by-east to south-west- by-west. All these dykes were intruded prior to the deposition of the Kadapa rocks. 0.— Granite veins. ~ Granite veins, except of very small size, are not at all common in the gneissic area, and none were met with of any importance either from size or special geological 1nterest. Some small veins of quartzo-micaceous granite traversing the schistose gneissies to the south-west of Kambaldinna, in the central part of the valley of the Man-eru (river), contain small garnets and prisms of tourmaline. Tourmaline occurs also in granite veins south-east of Petlur (18 miles south-west-by-west of Ongole). The tourmaline occurs here in some quantity, and good prisms are obtainable. Hemihedral erystals are not common, and the colour of the tourmaline is always black. A granite vein containing unusually large crystals of orthoclase felspar was observed in the bottom of the great tank west of Kondapy, at its upper end. Well shaped prisms of very pale flesh colour, 4 to 6 or 8 inches long and proportionately thick, were noticed, but were mostly too much cleaved to be extricable without breaking up into small fragments. (372517) 44 FOOTE : GEOLOGICAL STRUCTURE OF THE EASTERN COAST. Small irregular granite veins of pale pink colour and coarsely erystal- line texture are common in the micaceous and hornblendie schists south- east of the Chimakurti mountains, but they offer no points of special interest. = «—Felsite veins. A small series of hard vein-like bands of light coloured rock, having considerable resemblance externally to quartzite, traverse the hornblendie eneiss south-west of Pothakamur (Poothkamoor of sheet 76). These veins—for such they undoubtedly are—as they cut across the beds of gneiss at an angle of 45°, consist of compact felsite of pale bluish-grey colour, weathering a very pale cream colour. The freshly broken surfaces of unweathered parts of the rock show here and there sections of flesh-coloured crystals of felspar. The veins are small, being only 3 to 4 feet thick, and exposed only in short lengths. They occur over a tract about 2 miles long by a mile in width, which they traverse in a north-east to south-west direction. Though tolerably numerous, the veins form an inconspicuous feature in the landscape, and might be easily passed by unnoticed. No other occurrence of felsite rocks was observed in the Guntür- Ongole region. d.— Quartz veins. Quartz veins are numerous only in parts of the gneissic area, but, like the granite veins and trap dykes of this region, they are, with but few exceptions, of very small importance and size. The most remarkable exception is a considerable mass of quartz on the south side of the faulted southern boundary of the anticlinal ellipse west of Nakarikallu, at a spot close to the intersec- tion of the meridian of 80° east longitude by a parallel drawn in latitude 16° 20’ north. Though covering a large surface, the relations of the quartz to the fault are obscured by debris and soil, and it does not show any characteristics of being a fault rock, as might very well be inferred from its position. | E 9) At Nakarikallu. KADAPA SERIES. 45 The other quartz veins to be noticed occur mostly a little south of Vinukonda, near the villages of Ayanavolu, Payi- dipadu, and Ravaran. At the two latter villages the quartz veins run due north and south. "They are traceable only for South of Vinukonda. short distances. In the extreme south of the gneissic area numerous large north and south veins occur in the schistose gneiss of Dukanur hill, about 4 miles east-south-east of Pámur. Minute veins, such as characterize many mica schists and kindred rocks, occur in immense number, especially in the southerly parts of the gneissic area, and often give rise to prodigious accumulations of debris, by which the whole surface of considerable tracts of country is almost perfectly masked. Such is notably the case in the tract between Pámur and the Paleru valley. No quartz veins were seen containing sulphides of iron, &c., or any other indications of the presence of gold. Some small veins of milky-white quartz traversing the garnetiferous hornblendie and micaceous schists east of Bianapalle (on the banks and in the bed of the little Makeru river, which rises among the southern spurs of the Podile mountain) are wonderfully charged with minute dodecahedrons of brownish-red garnet. CHAPTER IIL—THE KADAPA SERIES. Assuming the numerous quartzite beds occurring within the gneissic area to be really members of that older metamorphic series, the newer or Kadapa series is but very slightly represented within our present limits, and only in the shape of a few patches, mostly outliers, four of which only require special notice, having already been partially dealt with in my notes! included in and appended to Mr. King’s memoir on the Kadapa and Karnul series. ! Memoirs of the Geological Survey of India, Vol. VIII, pp. 218 & 293. (457) 46 FOOTE : GEOLOGICAL STRUCTURE OF THE EASTERN COAST. The first outlier to be noticed is that capping the Bairawudi Konda, Ber out in the extreme south west corner of our gneiss lier. area, where it forms a very conspicuous object, as it rises into a considerable mountain, surrounded in most places by а cliffy scarp. Тһе outlier forms a small synclinal basin, out of the middle of which rises a great mass of coarse micaceous and chloritie schists, capped by a higher set of quartzites, forming the highest summit. The lower quartzites rest with great unconformity on the upturned edges of a great series of mica schist beds. They are doubtless a northern extension of the great beds capping the Udayaghiri, a few miles to the south; but there has been an appreciable thinning out of the quartzites between the two mountains, and the cliffs of the northern mountain are much inferior in height and beauty to those which rendered Udayaghiri such a famous stronghold in former ages. The relations of the Bairawudi Konda beds to those exposed in the main mass of the Vellakonda range have been illustrated in a section given in Mr. King’s memoir (7. e., page 222). Тһе lower quartzites of the outlier correspond to the Cheyair group of the series into which Mr. King divided the Kadapa formation of that region.! The lower quartzites occur in thick beds of whitish or buffy colour. At the northern end of the mountain they are much contorted, and the synclinal fold they form is beautifully shown in a fine vertical chff more than 100 feet high, at the southern end of a deep and very pieturesque ravine opening northward towards the village of Kothapalle. During the rainy season a small stream falls over this cliff ?, above which comes a considerable thickness of chloritie ! Memoirs of the Geological Survey of India, Vol. VILI, p. 126. ? Several small rock temples, probably of Buddhist origin, have been cut in the mica- schist on the slope of the spur on the western side of the ravine, and a few small niches on the face of the quartzite cliff; they are now sacred to Konabairuru deva, the deity of the waterfall. 00) KADAPA SERIES. 44 quartzite, and above this are chloritie and micaceous schists of dark green colour and coarse texture, which form the central mass of the mountain rising out of the rudely elliptical synclinal basin, and surrounded by a broad lip or margin of the lower quartzites, as shown in the section, fig. 2, page 14. The summit of the mountain is formed by a capping of younger quartzite. A small, but very conspicuous, outlier, consisting of the lower beds of the basement quartzites, caps the high hill rising. about a mile south of the mountain. A northern extension of the older quartzites occurs on the eastern eu Tug dah sue of the Tungur (Toongoodoor) pass, about 10 miles to the northward!. Here a remnant of the (local) basement quartzites has been left, abutting with its left or western side against the quartzites and slates here forming the mass of the Vella Kondas. There can be no doubt that, as already pointed out (page ul), the great line of fault, by which the Kadapa basin is abruptly bounded along the eastern flank of the Vella Kondas, passes along the axis of the valley lying between the lofty spur formed by the eastern edge of the patch of basement quartzites and the slope of the mountains locally distinguished as the Gali Konda (Gauly Conda). The basement quartzite dips westward, while the quartzites of the main range dip eastward, and thus appear to form an ordinary synclinal ; in reality, however, the quartzites of the western slope occupy a consider- ably higher position in the Kadapa series than does the basement quartzite, and the synclinal valley is, therefore, not an ordinary one, but one with unconformable sides? The basement beds thus constitute an outlying patch de jure, to which I will give the name of the Gali Konda pateh. 1 By some oversight this outlier of basement quartzites has not been shown on the map accompanying Mr. King's memoir with the same colour as the Udayaghiri and Baira- wudi Konda outliers, though unquestionably of the same age. 2 I could not examine this valley nearly as closely as I could have wished, owing to the extremely rugged nature of the country at foot of the mountains, which compelled me to camp at a great distance from the gneissic boundary. (4) 48 TOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. The next outlier of the Kadapas requiring mention in these pages occurs 85 miles to the north-east-by-north of the ln last-named one. It is one possessing considerable interest on account of its remarkable stratigraphical position, the pecu- liarity of which consists in the mass being an elliptical anticlinal dome let down among the gneissic beds by a series of faults, by which it has been eut into an elongated rather irregular hexagonal area, whose major axis extends about 7 miles north-east-by-north, the minor measuring about 5 miles from north-west-by-west to south-east- by-east. | The top of the dome has been much denuded, and the quartzites and other roeks eut away so much in the centre of the ellipse, that the underlying granitoid gneiss has been exposed 1n a narrow longitudinal valley, in which stands the hamlet of Biravallipaya. The dome is made up of four principal quartzite beds, which are separated from each other by three bands of slate. Here, as elsewhere, wherever the base- ment of the Kadapa rocks is seen, it is formed by a quartzite. As seen from the south-west, the slope of the hills is characterized by the bare surface of one of the quartzite beds dipping south-west at an angle of 30°, and which presents very much the appearance of a glacis leading up to the walls of a great fort. Тһе dip of the beds on the south-east and east side of the dome is from 45° to 50°, showing the anticlinal to be an unsymmetrical curve. The highest remaining part of the dome, which lies near the northern end, attains the elevation of: 1,379 feet over sea-level, and is crowned by a trigonometrical station. None of the beds exposed in this Biravallipaya dome could be identified with the beds forming the eastern part of the Nakarikallu elliptical anti- clinal, though they are separated by so small a distance. A similarly faulted dome forms the outlier west of Atchammapetta, Medo ths te M gum about 5 miles south-west-by-south of Chin-_ tapilly, on the Kistna river. In this case the out-lier has a rudely trapezoidal figure in plan, the greatest length. of side being about 13 miles. It forms a low broad-backed hill, of 190) UPPER GONDWÁNA SERIES. 49 which the greater part of the surface is bare quartzite, of brown and drab colours, with a quaquaversal dip. The apex of the dome lies consider- ably westward of the true centre. Though cut into by a deep ravine on the north side, the arch of the dome is not cut through, and the under- lying gneiss is not seen. No point of actual contact with the gneissic rocks is seen, but to the north of the dome is a large dyke of dioritic trap of the gneissic series of dyke which is older than the Kadapa system. North of this dyke is another outlier, also a domoid anticlinal, but of much smaller size, being only about 4 mile long by 4 wide, but consist- ing of bluish-drab and grey quartzites, unlike any of the beds seen in the greater dome. These two faulted anticlinal outliers complete the list of outliers of Kadapa age south of the Kistna. CHAPTER IV. —THE UPPER GONDWANA SERIES. The oM belonging to this very interesting and important division of the mesozoie rocks occur in numerous patches, mostly of small size, lying, with a few exceptions, along the eastern boundary of the gneissie area and dipping under the younger lateritie and alluvial formations. The exceptions are a few inliers situated within the area of the younger formations. "These, with the exception of two inliers on the western border of the Kistna delta near Guntür, are of trifling importance. Counting large and small patches, they number twenty-four, of which two possibly may be considered rather doubtful, being referred to this series solely because of their position. These will be referred to again further on. Besides the patches which were of sufficient size to be mapped, there are many other small exposures of these Gondwana rocks in wells and water-courses, showing that they are really much more extensively developed than they appear to be on the map. If the alluvium could be removed, the Gondwána beds would show a surface several hundred square miles greater in extent than that which they now D ( 49 ) 50 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. exhibit. For convenience in describing the several patches, they may be taken in four groups :— 1. The Kandukur (Cundacoor of sheet 76) group. Distributed into four groups 9. The Ongole group. SR 3. The Vemávaram-Budaváda group. 4. The Guntür group. Of these, the Vemávaram-Budaváda group is by far the most extensive, the most diversified in petrologieal features, and the richest in organie remains. In the Kandukur group Yinclude the various patches in the valleys of the Vupput-eru and Man-eru, and those occurring within or border- ing the Kandukur laterite area; also a small outlying patch, a mile east of Tlavara (Yellavurra of sheet 76), 11 miles north-north-west of Kandukur. | The Ongole group consists of two patches and several small exposures of shales about 6 miles west of Ongole town. The Vemdvaram-Budavada group includes all the patches lying between the Gundlakamma and the Perachur river. Of six patches, four are of importance; these are the Vemävaram, Budavàda, Idupula- padu, and Pámur patches, called after the geologically most important villages situated on them. The Guntür group consists of three patches, two of which are of con- siderable size; the Tangellamudi and Chebrolu (Bebbralu of sheet 95) patches lie in the alluvium of the Kistna delta to the south-east of Guntür. The third, of small extent, lies within a laterite area, 15 | miles south-west of Guntur. Besides these, there are a number of expo- sures through the lateritie deposits just alluded to, and also through that oceurring at Guntür itself. The most southerly of these four groups, the Kandukur group, is vibe ee, nd of a considerable тоо of small and widely scattered patches, all occurring, as already stated, in the valleys of the Man-eru and Vupput-eru, or bordering the (80-7) UPPER GONDWANA SERIES. 51 Kandukur laterite area, with exception of a small patch which lies 11 miles north-north-west of Kandukur, and a little east of the village of Tlavara (Yellavurra of sheet 76). Тһе group consists of thirteen separate patches, all but three being very limited in extent. Of these three the most southerly lies on the 15th parallel of north latitude, close to the village of Chinna Latarapy,on the right bank of the Vupput-eru. It covers only about 3 square miles of surface, but is doubtless connected below the alluvium and lateritie beds with other smaller patches which occur at small distances to the north, east, and west. The well-sections south of Chinna Latarapy show brown mieaceous sandstones, sometimes very thin-bedded and pebbly near their base, which rests on the gneiss. The beds have a south-easterly dip, and the plant remains they contain cannot be deter- mined. Chinna Latarapy patch. Traces of gritty or conglomeratie sandstone beds, in the shape of patches too small to be mapped, are found dotted here and there over the gneiss inlier east of Chinna Latarapy, and show the contact between the two rock-series very distinctly. Many of the patches are only a few inches across. The sandstone is composed of gneissie debris, chiefly quartzose in character. At the western extremity of the western bank at Chinna Latarapy, thin-bedded friable drab mieaceous sandstones are exposed, which show a dip of 2° to 3° to the north-west-by-north. South-west of Chinna Latarapy, at the village of Ramnaikenpett, drab sandstones with intercalated thin shaly beds Ramnaikenpet section. : are to be seen; in one well-section they are exposed to a depth of 16 feet, and show a dip of 7° to 8° to north-north- west.or north-by-west. In the nullah feeding the Ramnaikenpett tank, the following suecession of beds was noted :— Sandstone, pinkish-brown. Ditto, greyish-purple. A break Sandstone, very coarse, pale reddish-brown. Ditto, coarse, micaceous, pink and purple. (51 ) 53 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. Sandstone, whitish-grey. Ditto, grey. Ditto, shaly, yellow-brown. Ditto, ditto, purple. Ditto, purple. Ditto, micaceous, pinkish-brown. Grit, shaly, dark purple, ferruginous. Gneiss. These beds roll about a good deal, and are locally often obscure, besides showing false-bedding, but the general dip is westerly at a low angle. A large mass of silicified wood was observed in one of the beds at the top of the series, but no other fossils were found in this section. A couple of miles south-west-by-west from Ramnaickenpett, at Pedda Pamidi sec. (һе village of Pedda Pamidi (Pedda Powenny), tion. drab friable sandstones are exposed to a depth of 15-16 feet in well-sections, under several feet of kankar and lateritie gravel. Separated from these sections by about half a mile of alluvium, is a fair section exposed in the right bank of the Vupput-eru, at its junction with the Mutialpad nulla. The succession of beds is not very clear owing to the broken nature of the bank and to the want of definition of the several beds, which have a tendency to graduate one into the other. 'The beds, which have a general north-westerly dip, never exceeding 8° and generally much less, succeed each other in the following order :— Sandstone, shaly, Indian red and drab ... seca d Ditto, drab ves sae ru d Ditto, shaly, drab and Indian red .. 20. 0282 (о 02.302 Ditto, ditto in parts, mottled red and brown .. 2% to 3 Ditto, ditto, Indian red and drab ... ДР БҮ to 4 Some 6 miles to eastward of this last section is а low rocky bluff, EN er consisting ei coarse gritty conglomerate and sand- stones, rising out of a jungly plain and lapped _ round by laterite gravel. ` This bluff is formed by a small remaining patch of Rájmahál rocks resting on gneiss, which shows a very little distance to the south : the conglomerate and sandstones are of no great thickness, ( 92 ) | UPPER GONDWANA SERIES. 53 probably not more than, if as much as, 50 feet. Resting on the surface of the sandstones are concretionary fragments of purple clayey sandstone, containing plant remains of Rájmahál age, now worn into pebbles, and belonging to the lateritic gravels. Further north, 3 or 4 miles east of Chinna Latarapy, similar washed-up purple plant sandstones form a considerable proportion of the lateritic gravel resting on the gneiss. From similar purple grit and hematitic sandstone pebbles in the laterite gravel, unquestionable Rájmahál plant remains were obtained to the south-east of Razpalem, 3 or 4 miles north of the Peddavaram bluff just described. Among the plant remains here found was part of a frond of Ptilophyllum (Paleozamia). To the northward of the Chinna Latarapy patch of Räjmahäls are several sections of plant beds, chiefly sandstone, to be seen in the Mogallur nulla, and in the bed of the Man-eru east of Mogallur. In the latter. case, the friable brown micaceous sandstone yielded parts of broad Ptilophyllum fronds. This plant bed is overlaid apparently by the various sandstone and shale beds seen in the principal section in the Mogallur nulla which present the following series in descending order, the beds having a general dip of from 7° to 10° to the south-south-east :— Mogallur sections. Shale, sandy, ferruginous d ane Ов" Sandstone, brown, friable SUN ЕЖ A e CAE Kankar parting . Br bab Has ОА Бо Sandstone, brown, friable Ser 55) ИРА БЫ Kankar parting is bod Sod RUE to 3” Sandstone, brown, with grey and red shaly lamine, much false-bedded Js ae Dos Ee d Sandstone, brown ui Ad КЕН ХА Ditto, shaly, grey and red Kr j^ Овоо" Kankar parting sis She a u 1090587 Sandstone ... Bia 555 TR катет акты Ditto, shaly, grey and red ... VUE S00. 4A" to 14" Kankar parting ss ie aac see Oe 37 Sandstone, brown, micaceous friable n 6: TOTAL about n. 20 mamta ann 54 FOOTE: GEOLOGICAL STRUCTURE OF THE ESATERN COAST. Three miles to the north-east of the river-side section, plant beds are again seen in well-sections at the south end of the village of Gudlur. The beds here consist of soft drab micaceous shaly sandstone, with numerous fragmentary plant Gudlur sections. remains in horizontal beds divided by kankar partings. Only one identifiable fossil was obtained here, a small fragment of a Dictyozamites. A well-seetion north of two small ruined pagodas shows similar sandstone, with shaly partings of chocolate colour containing obseure fragments of plants in great abundance. The beds dip north-by-east at from 13° to 27. From 10 to 20 feet of the plant beds are visible above the water in the several wells in which they are exposed. Тһе surface is obscured by a thick deposit of kankar and lateritie gravel. Gritty mottled sandstones Ape EE are found under the laterite to the south-west of Virapalle (Veerapully) ; they are probably of Ráj- mahálage. This completes the enumeration of exposures of Rájmahál rocks south of the Man-eru. There is a great degree of resemblance in petrological and lithological characters between these beds and those seen in the Alieur and Pyanur areas of the Rájmáhal series west of Madras, the predominant feature in both cases being the soft and unconsolidated condition of the greater part of the constituent deposits. To the northward of this river the first inler of plant beds is met North of Mánceru. with at Kalamalla (Calamulla), but it is a very e ila sea small and unimportant one. Тһе beds seen in the north bank and bed of the river consist of drab and greyish-brown sandstones of moderate hardness, occupying a nearly horizontal posi- tion. The dip, if there be any, is slightly southerly. No fossils were found, but the lithologieal resemblance to the more southern beds and the geographical position justifies the conclusion that the Kalamulla beds are equally of Rájmahál age. Friable micaceous sandstones of reddish-brown and grey-drab colours Pues S form a miniature cliff 5’ 7" high on the south bank of the large nulla at Polenanepalem, 7 miles 1 See Memoirs of the Geological Survey of India, Vol, X. (692.0) UPPER GONDWANA SERIES. 55 north-west of Kalamulla. This also is a very small inlier of Rájmahál beds among lateritic sands. We now come to the Chautapalem patch, the largest of the Kandu- Chad E kur group, but en 5. its ee offers but little of interest, as it contains not a single good or instructive section, only a few poor well-sections, and no good fossils were obtained from the soft, often shaly, drab or brown sandstones they expose. | The Ponnalur patch, which is only separated from the foregoing by _ about amileof overlying lateritic sands, shows soft Ponnalur patch. shales and shaly friable buffy-brown sandstones underlying the lateritie beds at its northern extremity, close to the village after which it is called. Soft drab sandy beds are exposed in a well-section about a mile to the south-west. Plant remains are scarce and very fragmentary, but enough were found to determine the age of the beds beyond doubt. Between 6 or 7 miles to the eastward lies the village of Kovur, UE which stands on a narrow strip of Räjmahäl Kovur section. : rocks lying between the northern: boundary of the Kandukur laterite area, and the alluvium of the Paleru, and extending rather more than 4 miles from east to west. A few good well-sections, south of, and close to, Kovur village, give some insight into the nature of the plant beds here occurring. The rock exposed is drab micaceous sandstone, shaly in part, having a north-easterly dip. Some of the more shaly lamine, as exposed in the waste heaps of material dug out in sinking the wells, show recognisable plant remains, though but small and fragmentary parts of the plants are preserved. Among the specimens collected were fragments of three species of Ptilophylium, and with them impressions of minute bivalve shells which were remitted to Calcutta for determination. "The eastern part of the patch is greatly obscured by a remarkable bed of massive kankar-like limestone of undetermined age, which is especially developed at Kunda-Kandukur (Conda Cundacoor). Сва) Fossiliferous shales. 56 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. Unfortunately no section could be found throwing any light on the rela- tion of this kankar bed to either the under or overlying deposits. The nature of the small shells collected, whether marine or fresh water, has yet to be determined by comparison; they were of too small а size and inconspicuous a character to be safely determined off-hand. Some minute shells of precisely similar character were found on.a well heap in the field about three- fourths of a mile north-west of Kandukur, and a few yards off the path leading to Kovur. These were associated with fragmentary plant Shales at Kandukur. remains. The beds seen in various well-sections, or their debris exhibited in the waste-heaps, shows the prevalent character of the rock to be drab or brownish-buffy sandy shales, or shaly sandstones with reddish laminæ. These rocks, whatever their character, are largely composed of scales of miea, and mostly very friable. Westward of Kandukur the Räjmahäl rocks are seen in well-sections, Anantasagaram and And a few very shallow surface sections at several up sections. places at and around the villages of Anantasagaram and Yedlurpad. Shaly micaceous sandstone from a well just east of the former village showed just distinguishable traces of Pirlophyllum. This sandstone contains but few plant remains, and those few are very fragmentary. A well-section at Yedlurpad shows about 20 feet in thickness of drab shaly sandstone containing numerous concretionary surfaces, white and porcellanoid in appearance, and striated like a fine slickenside. ‘Some concretionary masses also had their whole exterior covered with this glistening surface. This quasi-porcellanoid surface is due to the presence of a film, as a rule of extreme thinness, of calcareo- argillaceous material. Though porcellanic in appearance, the bright surface is really soft and very easily scratched. Eastward of Kandukur, chocolate-coloured shales were observed in the ditch by the side of the new high road leading from Kandukur to the great north trunk road at Singarayakonda, at a distance of about 3 miles from the former (56) East of Kandukur. UPPER GONDWANA SERIES. 57. place; and in a well-section close to south of the road which leads from the trunk road to the sea at Voolapalem, a section of purple and white mottled gritty shales, some 12 feet thick, may be seen. No distinct traces of plants could be made out in either locality; but the shales have a very strong resemblance to other unmistakably Rájmahál beds, as, for example, those of which traces in the lateritic gravel, at and to the south of the village of Razpalem, were referred to above, page 53. The one member of the Kandukur group of patches of Rajmahal beds lying northward of the Paleru occurs a mile east- llavara patch. д ward of Ilavara (Yellavurra), 10 miles north-north- west of Kandukur. Неге, in the bed of the large nullah flowing east- ward into the Musi river, and in several small gullies opening into 16 from the south, are to be seen beds of micaceous sandstone overlaid by grits and shales with clayey bands, all dipping southward at angles varying from 5° to 10°, or else rolling about. Another similar series of sandstones and shales, apparently underlying the above-mentioned beds, occur on the northern bank of the nullah, and on the banks of a small tributary from the north. One bed of sandstone low down in the series, exposed in the gully immediately west of the little hamlet of Netivaripolem (not shown in the map), contains a few rather large but much weathered boulders of gneiss, reminding one of the boulders so common in some of the beds at the base of the plant-bearing series in Trichinopoly district. In the bed of the large nullah a little east of the boulder bed the sand- stones are very coarse and gritty; they roll about a good deal, but the general dip is low south-south-westerly. The shaly beds in the more westerly gullies yielded a very few plant remains of the most frag- mentary kind, but only after long-continued search. The most recog-. nisable specimen seemed to be part of the mid-rib of a Ptilophyllum frond. This small patch of plant beds appears to.owe its continuance to the fact that it is sunk in a depression in the gneiss, and has therefore escaped the full force of the erosive agencies which have so greatly affected the Bäjmahäl beds over a large part of this particular region. (507) 58 FOOTE? GEOLOGICAL STRUCTURE OF THE EASTERN COAST. The second or -Ongole group of patches of the plant-bearing beds ШАЛЫ ра consists of only two patches, the southernmost and patches. smaller of which lies 11 miles north-east of Ilavara and 6 miles west-south-west of Ongole. This patch is greatly obscured by the overlying thick cotton soil; the only tolerable section found is ое man. “seen in! a small well on the east side of the road gamur. leading from Konijadu (Conjadoo) to Santa- nuthatapad. The western side of this well shows the section illustrated Fig. 3. Well-section south of Matigamur. in the accompanying diagrammatic sketch. The base of the Räjmahäl rocks is here formed by a gritty red sandstone (4), resting on coarse, highly felspathie granite (1). This sandstone is seen to pass suddenly into a singular rock (3), partly conglomerate, partly breccia, with a calca- reous matrix enclosing quartz pebbles and fragments. On the opposite or north-east corner of the well the bed consists of a coarse sandstone uo | UPPER GONDWANA SERIES. 59 conglomerate, with some enclosed white quartz pebbles of large size, The calcareous rock thickens southward. The red sandstone is overlaid by whitish shales (5) much mixed up with kankar, which may be traced with difficulty for a few hundred yards northward, and are then com- pletely hidden by cotton soil, which extends up to and far beyond the village of Mangamur (Mungamoor). In some well-sections south of the village, buffy brown sandy shales have been cut through under a bed of kankarry lateritic gravel. These shales establish the connection with some plant-bearing sandy shaly clays of drab-buff colour which are Section at Kamapatte. ©Xposed near the head of a small stream a little to nn, the north-west of the small hamlet called Kama- patte-varipalem (Kaumayputty-vareepully). The plant remains are unfortunately very fragmentary ; but amongst them I recognised a small part of a Ptlophyllum frond; underlying the shaly clays, of which a thickness of 4 or 5 feet is seen, is a coarse friable micaceous sandstone of buff colour. Rather more than a mile to the north of this section very similar buff and brown friable shales and sandstones show meee in the road drains and in various ballast pits, and in a few wells off the Ongole-Kambam high road, both on the slopes westward to the Santa Nuthalapad valley, and on the high ground to the eastward. The surface of the shales is thickly covered by yellow kankar gravel, mixed with lateritic pebbles. This yellow kankar descends Yee overly: deeply into the greatly weathered surface of the ing shales. shale beds, not only here, but also in many other sections, so much so that the yellow colour of the kankar always led me to look out for underlying plant shales. The friable shale and sandstone beds extend north-eastward and northward from the high road down to the village of Yendlur, where they disappear under the great alluvial flat formed by the Mudigondi and Gundlakamma rivers. The high ground south of the high road was, at the time of my visit, covered by a very singular concretionary Singular sandstone. > RR 2 calcareo-ferruginous sandstone, often of jaspi- ( 59 ) 60 FOOTE : GEOLOGICAL STRUCTURE OF THE EASTERN COAST. deous texture, the reddish yellow mass being largely permeated by minute silicious veinlets and threads of lighter colour. No organic remains could be traced in this strange bed, nor could its relations with the typical plant beds near-by be seen, owing to the great development of cotton soil around it. All that can be said of it is that it lay unconformably on the gneiss. When I first visited it in May 1875, it was being largely worked as revetment material for sundry tanks and wells; and by December of the same year I found that but very little remained to be earried away. A spread of alluvium, of 7 miles in width, lies between the The Vemdvaram Buda. Southernmost point of the Yendlur patch of the pada DIT ee Ongole group of plant beds and the nearest ex- kamma. posure of equivalent rocks belonging to the Vemávaram group which occurs on the north side of the Gundlakamma river at Gazulapadu, about a mile north-east of the travellers’ bungalow at Velampalle (Valumpully). The Gazulapadu section, which can only be seen when the river is quite low, shows very sandy kankarry clay, underlaid by orange and brown rather friable sandstone and grits, pebbly grits, and coarse conglo- merate, all apparently part of one large variable bed. The conglomerate contains fragments of an old laterite; and a number of boulders of an identical rock are seen in one place resting on the clay bed just mentioned. The boulders are from 18 to 20 inches in diameter, and are themselves surrounded by the overlying alluvium, which here forms a considerable cliff. | The sandstones and grit bed can be traced all along the base of the alluvial cliffs from Gazulapadu to Kirtipadu (Keerteepaudoo), where it is hidden by a sandbank; but it re-appears again in the next reach of the river in the northern bank, and can be followed till nearly opposite Nandipadu (Nundeepaudoo), when it disappears. Grey or brown sandy shales—true plant shales— are shown, by means of well-sections both east and west of Nidamanur, to oceur below the thick cotton soil which there forms the surface. "They (60) Vemávaram patch. UPPER GONDWANA SERIES. 61 have also been turned up from a deep well sunk at Kavurupalem (Cavoororpalliam), 2 miles north-east of Nidamanur. Both these places ` lie within limits of the Vemávaram patch, the most important and most; interesting exposure of the Räjmahäl rocks in the Guntür-Ongole country; but before proceeding to the full description of the rocks here met with, it will be better to refer briefly to three small and important patches lying at but small distances. These are, firstly, the Bolaveram patch lying to the west, and the. Small patches at Bola- Nag alupalapadu (Nag ooloopalapaudoo) and Cha- veram, Chadulavadu, and dulavadu patches lying respectively to the sonth- Nagalupalapadu. east and south-east-by-south. The surface of the Bolaveram patch is entirely covered by the Thome teristic yellow kankar referred to as accompaying the plant shales in the Yendlur and other more southerly patches. The Chadulvada and Nagalupalapadu patches are exposed only in solitary sections, the former in the bed of a tank a mile and a half east of the village, the latter in the great square dowry, or tank-well, east of the village. In both cases the rocks exposed consist of shales. These two patches of shale are separated from each other by a great unbroken spread of cotton soil underlaid by a low ridge of gneiss ; but the cotton soil is so thick and continuous, that it is impossible to draw any boundary lines between the underlying rocks. Of all the representatives of the Rájmahál rocks south of the Kistna, the beds exposed at Vemávaram (Womayaveram) . Vemávaram sections. x р are the most interesting and noteworthy, as they contain a larger number of both animal and vegetable remains, mostly in beautiful preservation, than any other beds met with elsewhere. The presence among the animal remains of Cephalopoda and Echinodermata establish the marine nature of the deposit beyond controversy. The village of Vemävaram, close to which these plant-bearing marine beds are exposed, lies 14 miles north-east-by-north of Ongole, and 8 miles inland from the coast, on the eastern slope of a low bare (oU 6 9 FOOTE : GEOLOGICAL STRUCTURE OF THE EASTERN COAST. Figure 4. Diagram-section across Vemávaram ridge. ridge of hard shale about three-quarters of a mile in length. The strike of the ridge is north-by- east to south-by-west. To the north-east and south it dies down under the great spread of cotton soil which locally covers the coast allu- vium. To the south-west it sinks into a low neck of ground by which it is connected with a low Ue ып broad-backed hill close to the.village of Uppu Gundur | (Ooppoo Goondoor), which also consists of beds of hard shale, which are nearly everywhere bare and devoid of soil and vegetation. Тһе two hills form a peninsula jutting into the great alluvial plain. То the west the ridge is joined by а low neck with the western half of the Vemávaram patch. -The section given below shows the position of the beds seen in the central part of the Vemáva- ram ridge. The beds shown in this section, which coin- А ; cides with a low depression Details of section. : И crossing the shale ridge а httle to the north-west of the village, follow in descending order as below :— 8. Shales, purplish. 7. Do, buffy. 6. Do. softish, brown, white, and purple. 9. Do. hard, with red and brown sandstone partings. 4. Do, do., partly flaggy, with sandstone partings, variegated. 3. По. thin, flaggy, rather hard, buff and white, “fish bed.” 2, Do. sandy, mottled. . Sandstone, shaly, buffy. ке UPPER GONDWANA SERIES. 68 Both the top and the base of the section are obscured by cotton soil, but rather more than 2 miles to the west, softish gritty sandstones appear from below the black soil. The beds included in the section have a general low dip to the east- ward, but they show also several small local rollings. The angle of dip ranges from 2° to 5° on the western side of the ridge, and 12° to 15° on the eastern side, close to the village. The section represents a distance of about 350 yards in length. The colouring of the beds on the ridge varies frequently and within very short distances. Unfortunately, the shales are of no use economi- cally, because of their breaking up into small pieces owing to the great number of joints traversing the beds; hence they are nowhere quarried, though other hard shales, which break into great flaggy pieces, are largely quarried at a little distance to the south. These latter unfortunately are poor in fossil remains of all kinds. Fossils oceur in all the shaly beds and are numerous in most, plants being rather less numerous than animal remains. The two classes of organisms constantly occur together in the same hand specimens. The most remarkable and about the best preserved fossils were obtained from the thin flaggy hard shale (No. 3 of the section) which, for brevity, I call pe Ai die s une bed.” sms came doubtless the very interesting Eryon which I obtained on the occasion of my second visit in 1876. This macrurous crustacean, the first crusta- cean found in the Rájmahál rocks, which has since Eryon comp. Barroven- ў ң sis not obtained at Sri- been specifically determined and figured by Dr. O. Веде | Feistmantel! as Eryon comp. Barrovensis, McCoy, is of special interest and importance as being the first case of identifi- cation with a European species of any of the animal remains as yet dis- covered in the marine beds of the Rájmahál series on the eastern coast of the peninsula. ! Records of the Geological Survey of India, Vol. X, p. 4. Note on * Eryon comp. Barrovensis," McCoy, &c., by Ottokar Feistmantel, M. D. (0950) 64. FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. It was the discovery of this Bryon on a loose piece of shale that led me to look for the peculiar and, till then, unobserved bed it must’ have come from; by excavating I procured many of the finest and most im- portant specimens collected at Vemavaram, including nearly all the fish remains obtained there. The missing parts of the Eryon could unfortu- nately not be found, though most carefully searched for. By some mistake or other, Dr. Feistmantel has unfortunately des- cribed the Eryon as having been obtained from the Sripermatur group west of Madras, a series of beds which probably represent a rather higher horizon than at the Vemávaram shales. The error of locality is consider- able, as the two places are rather more than 200 miles apart. The remarkable lithological similarity between the typical Sripermatur and Vemávaram shales must have contributed not a little to the possibility of such a mistake having been made. The highest known member of the Vemavaram group is a hard, coarse, white shale, breaking into large flaggy masses (which, as already mentioned, are largely quarried to the south of the ma village). They contain a few, mostly ill-preserved, fossils, consisting of undeterminable stalks of plants, of bivalve shells allied to Leda, and of thin-shelled Ammonites, too much flattened by pressure for satisfactory determination of their specific characters. These coarse shales occur to the south-east of the Vemävaram section first described, and would, if that section were extended sufficiently to the east, be seen to lie above, and probably immediately upon, the purplish shales No. 8, though it is possible that a small thickness of finer grained whitish shales might intervene. Very similar coarse flaggy shales form a thick bed exposed in numer- ous quarries close around the village of Razpudi, a Flaggy beds at Raz- ER little more than a mile north of Vemávaram ridge. They are very likely a continuation of the Vemá- varam flags, the only difference they show being in colour, which is not pure white, but white variegated with delicate bands of red, pink, mauve, ( 64 ) UPPER GONDWANA SERIES. 65 purple, and occasionally orange, arranged concentrically, and presenting surfaces of no little beauty of colour. "Phe form of "Their colour. f those patterns on the shale vary considerably, as they are generally confined between the limits of a set of joints to which the concentrie curves show a clear relation. Some examples were noticed several feet in diameter, others only a few inches, but these latter were generally independent of any visible joints, No Ammonites were found in the Razpudi bed, but the other Fossils found at Razpudi. : Д 5 fossils agreed with those from the Vemávaram flags, and, like them, are ill-preserved, many being blurred by an incrustation of extremely minute quartz crystals. Fragments of Dic- tyozamites and Ptilophyllum were the only recognizable plants met with at Razpudi. In their mode of preservation and condition of occurrence, the Vemá- varam fossils agreed very closely with those obtaimed at and near Sriper- matur. In the great majority of specimens the impressions or casts are stained of some colour different to that of the enclosing shale. The colour, which is generally some shade of red or purple, more rarely Fossils at Vemávaram. Plack or brown, is in most cases confined to the Colour, &c. organism whieh, for that reason, contrasts strongly with the matrix. The plant remains are all fragmentary, many of them very much so, showing that they must have been washed out to sea as torn-off leaves or fronds, but speedily embedded, as they show no ICON BE pre eoa. signs of decay. The shells are frequently crushed tion and its indications. by pressure supervening after they were buried in the mud, but very few show signs of previous injury, and in a very large percentage the two valves of the pelecypodous shells remain in opposition, showing that the living animal had not been brought hither from any great distance, but was rather an inhabitant of the place where entombed, or of its immediate vicinity, As in the Sripermatur beds, the character of the most common shells indicates that the shales were deposited in a rather shallow tranquil sea, E (Оа) е 66 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. The following list of the organie remains, both vegetable and animal, shows their nature, but only in a pre- EEE EN liminary manner, as complete lists can only be drawn up after close examination of the entire collection now in the Museum at Caleutta. The list of animal remains, indeed, must be accepted merely as a rough field list, as no opportunity offered for any close examination or comparison before the collection was shipped to Caleutta. | Further examination and excavation of the shales will doubtless add very considerably to the number of fossils now known to occur in the Vemávaram beds. PLANTS. ANIMALS. Fucoida ? 2 sp. ? Rib, impression of, ? mammalian. Sphenopteris, sp. Fish, 2 or 3 species? Dicksonia, sp. Do. scales, cycloid. Cyclopteris ? sp. Eryon comp. Barrovensis.! Alethopteris indica. ` Ammonites. Angiopteridium spathulatum. Belemnoteuthis, sp. conotheca, crushed. Macroteniopteris ovata. Gasteropodous shell, impression of. Do. Sp. Inoceramus. Pterophyllum distans. Pecten. Do. sp. Exogyra. Do. jissum. Leda, 2 or 3 species ? Zamites proximus. | Tellina, do. P Ptilophyllum acutifolium. Yoldia, do. P Do. cutchense. Lithophagus, sp., burrows of. Otozamites, sp. P Sp. a broad shell quite Dictyozamites indicus. crushed. Palissya indica. Terebratula 2 . Chirolepis ? sp. Eschara. Echinostrobus rajmahalensis. Ophiura, 2 species or varieties. Araucarites, sp. (? macropterus). Cunninghamites dubiosus ? Coniferous leaves undetermined. Stalks, seed-vessels, &c., &c. 1 Since determined specifically by Dr. India, 1877. (6677) Feistmantel. Records, Geol. Surv. of UPPER GONDWANA SERIES. 67 Тһе” ef cetera includes at least three more undetermined plants of whose botanical affinities I was in doubt, but could obtain no solution from the books of reference at my command. The stalk-hke remains also probably include several species yet to be determined, as they show considerable variety of marking and branching. Though similar in general appearance to the Vemávaram beds, the ` Uppu Gundur beds differ in being generally very Uppu Gundur beds. р ^ i | poor in fossils. The most important fossils found here were a few Ammonites of small size, but unfortunately they are too much crushed for specific determination. Owing to the great sheet of cotton-soil between the two hillocks, the real relationship between the several beds composing them cannot be determined positively, but the probability is, the Uppu Gundur beds overhDe the Vemávaram shales. As before mentioned (page 64), the coarse flamgy ammonitiferous shale exposed in small quarries between the two hillocks must be looked upon as a southern extension of the Razpudi flagey beds. The beds on the hill are in parts horizontal; to the north-east and east they dip corres- pondingly, and they also make a slight roll to the west. "They are more massive than the beds on Vemávaram ridge. To the west of Vemávaram ridge, cótton-sol covers everything thickly, till the ground rises again beyond the old Madras-Guntár road. Barely perceptible traces of the plant-beds are seen eastward of Kallagumta. A little gritty sandstone is seen at Shales at Pyindipad. | UT Kistarazpalem, but at Pyindipad sandy shales and friable sandstones are cut through in several well-sections. No fossils were seen in these beds, which are underlaid to the north-westward by brown gritty sandstones, which show close to the boundary of the gneiss, on which they doubtless rest. South of Pyindipad tank is a band of hard gritty sandstone extend- Nelken) ing from the east end of the Annavallawarpalem зола они иа tank north-eastward nearly to Kistarazpalem. No fossils were seen in this sandstone, which is of varying colours, ranging BIO 68 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. from dark blackish-brown to grey or drab. It rests upon the granitoid gneiss close to Kallagumta village. South of the hard sandstone, and probably overlying them, come soft friable gritty sandstones and shaly clays. The latter are of drab colour, the former orange to reddish or Plant-remains in shaly Arab, and full of impressions, in bright red, of clays. fragmentary plant-rémains, amongst which the pinnule of a small variety of Dictyozamites indicus are most commonly recognizable, Ptelophyllum acutifolium being also present, but much less common. The dip of these beds is westerly from 3° to 5°. A little further south still are coppery-red mottled sandy clays, with orange to reddish-brown gritty sandstone, lying close to, or actually on, the gneiss. A shallow alluvial valley, nearly 2 miles across, divides the Vemá- Soc Lu varam patch from the next northern patch, which lies around the village of Nucherlapalle (Noocherla- pally). The beds here seen in well-sections are white and mottled shales, very like the Vemávaram ridge beds ; but, unlike these, they are very poor in fossils, a long search only yielding a single specimen,—a thin, broad, bivalve shell, erushed flat by pressure, of a species very common at Vemá- varam. Тһе beds show а dip of from 10° to 15°, to east-by-south. Sandy shales are seen in a small field well, about a mile to the westward, near to the gneissie boundary. | The next patch of Rájmahál rocks, proceeding northward, is the URN Budaváda (Boodhawadah), the southern point of which lies 8 miles to the north-north-east, near Gangävaram, where shaly buff sandstone is seen in well-sections west and south-east of the village. Similar shaly sandstones and sandy | shales begin to appear in small field sections as the village of Budaväda is approached from the south, and in the bottom of a tank lately con- Shale bed east of Structed about half a mile east-by-south of the Ro. main village. Неге а few score square yards of shales were to be seen when the tank bottom was dry; they are under- laid by, or intercalated with, friable soft sandstone, also of whitish to ( 68 ) UPPER GONDWANA SERIES. 69 yellowish-drab colour. Both contain fragmentary plants and small sea- shells. "The plants and shells found were as follow :— PLANTS. ANIMALS. Dictyozamites indicus. Ammonites sp., crushed. Pterophyllum sp. Pecten 2 sp. Leda ? Ptilophyllum cutchense. Do. acutifolium ? Avicula. Palissya indica ? Fish scales, cycloid, very thin. These beds represent No. 6 of the section given below. Underlying these soft sandstones and shales is apparently a set of sandstones on which stands the village of Buda- бейнеле section. väda, and overlying them is another set of sand- stones on which the village of Pävulur has been built. The relation of the several beds is best explained by the accompanying ideal section from Budaváda to Pávulur. Pavulur Budavada iy a E = N >= Sn _-s —— IT SS >= 78 Gneiss Fig. 5. Ideal section from Budaváda to Pávulur. The section begins half a mile west of Budaváda, at the gneiss boundary, runs due east for the first half of its length, and then turns north-east up to the village of Pávulur, the total length being two and 2, half miles. The beds here seen fall naturally into a triple grouping, thus :— 10. Lateritic gravel. 9. Sandstones, friable, coarse, reddish-brown. 8. Sandstones, hard, greenish or bluish-black, calcareous, Pévulur group __... Я is ? " slightly shaly, weather grey or brown. 7. Sandstones, friable, drab pale brown. 6. Shales, various, hard and soft, mottled in parts, gene- rally whitish or light grey in colour. ( 69 ) Vemävaram group |... 70 FOOTE : GEOLOGICAL STRUCTURE OF THE EASTERN COAST. ( 5. Sandstones, gritty, calcareous, full of shells, rather hard and tough when fresh. | 4. Sandstones, shaly, friable, dark-buffy. 3. Sandstones, hard, brown, alternating several times with thin shaly beds. Sandstones, massive, hard, brown. Budaváda group 1 m ( 1. Sandstones, pebbly outerop, much weathered. Gneiss. The general dip of these beds is easterly, at varying and mostly low angles. Beds Nos. 4, 5, 6, and 8 contain marine shells; Nos. 5 and 6 contain plants as well, and No. 9 plants only. The shales No. 6 represent, I believe, the Vemávaram shales, from - parts of which they do not at all differ, while they have a strong re- semblance to many of the shales in the Utatür patch of plant-beds in Trichinopoly district. They are best seen in well-sections to the west and south-west of Nakkalapalem and north-west of Pávulur. The shelly caleareous sandstone No. 5 is quite unlike any known member of the Räjmahäl series throughout India. Unfortunately, what little remains of the bed is very badly exposed, and its relations to the under and overlying beds are consequently obscure. Most of the bed had been quarried away when I first visited Budaváda ; what little remains is to be seen at the western side of the village, just south of the end of the main street, and in a few wells south of the village close to some old indigo works. This remarkable bed was probably a drifted accumulation of shells deposited in discontinuous lenticular patches of limited extent. The matrix is generally gritty, but here and there clayey. The Budavada section seems to take in the whole of the Rájmahál series in this region, but owing to the great and continuous spread of cotton soil, which covers the face of the country generally, parts of the section are by no means clear, and I offer it with some hesitation, as other observers might draw different conclusions from the data available. It is constructed from the examination chiefly of a series of well-sections supplemented by a few poor outcrops and small quarries. (oe) UPPER GONDWANA SERIES. 71 The shelly sandstone No. 5 is extremely rich in fossils ; indeed in Hue n many parts the whole mass almost consists of shells, broken or entire as the case may be. The plant remains it contains are very rare, and only one specimen of a Bryozoon was found. No other fossils were found. The following list of the specimens collected must be taken as purely preliminary, a rough field list in fact: PLANTS. Ptilophyllum acutifolium. Angiopteridium spathulatum. Otozamites sp. Undetermined leaflets and small woody Dictyozamites indicus. | fragments. ANIMALS. Ammonites sp. ? 2 sp. Pecten, 9 sp.? Buccinum вр. ? Lima sp. Natica sp. - Plicatula Sp. Cerithium ap. ? Avicula sp- Turritella. sp. ? Modiola sp. Eucyclus sp. ? Yoldia sp. ? Turbo sp. ? Trigonia sp. ? Trochus sp. ? Cardium sp. ? Patella sp. large, broad. Astarte sp. ? Do. sp. 2 high. Mactra sp. ? Do. sp., small, radiately ribbed. Myacites sp. Rhynchonella 2 от З вр. Gresslya sp. ? Terebratula ?2 sp.? A bryozoon. Ostrea 3-4 sp. ? Serpula Sp. The plant remains, which, as already mentioned, are of rare occurrence, Mn d cs were, with one or two exceptions, all obtained from one mass of the stone which was unusually clayey in charaeter. "They are sufficiently well preserved to be easily determin- able. Some of the best specimens, especially of the large Patellas, were broken out of masses that had been used in the walls of the furnace under the boilers in the old indigo factory, and show strong indications of fire action, approaching caleination, being much reddened or blackened, and having become much more brittle than the unburnt specimens, (CURIAM) 12 FOOTE : GEOLOGICAL STRUCTURE OF THE BASTERN COAST. The most numerously developed species are the Ostreide, then the Numerical ratio of Lerebratule and Rhyuchonellide. The Ammonites жесе. are fairly numerous, but mostly fragmentary. Of the Gastropoda, the Patellide are the most remarkable, both for number and great size, but unfortunately most of the specimens are in poor preservation. Three, if not four, species are recognisable among them. Many of the shells were imbedded in a broken condition. No determinable fossils were found in the Pávulur sandstones 20 situ ; those seen were extremely minute fragments en of shells, but little larger than the grains of sand they are associated with. Fossils of recognisable character were, however, Caleareous sandstone found in a number of blocks of coarse calcareous Eee E sandstone forming part of the revetment of the bund of a little field tank about half a mile north-east-by-north of Budaväda village. . Unfortunately, these blocks could not be traced to their original sites. None of the contained fossils could be identified with true Budaváda Species, and the blocks themselves bear no resemblanee to either the Budaväda or Pávulur rocks, According to the villagers they had been brought from the Pávulur quarries, but no similar stone could be traced by me; their origin must, therefore, remain uncertain. The caleareous sandstone forming these blocks is very hard and ex- ceedingly tough, and the fossils very hard to extract. The following list is a rough determination of the few that I Its fossils, succeeded in getting after much severe labour :— Belemnites sp., small Cerithium sp. Turbo sp. P Patella sp. Iehynehonella sp. Cardium Spe Ostrea sp. Lithodomus bores in Ostrea. Small bivalves undetermined, several. UPPER GONDWANA SERIES, 73 These Belemnites are ће only South Indian representatives of the genus that have been collected outside of the cretaceous areas in Trichi- nopoly and south-west districts. Not the faintest trace of plant-remains accompanied the shells. Unconnected with any of the rocks named in the list descriptive of ШАҚ the Budaváda section 1s part of a large Nautilus that I picked up loose m the fields about three-quarters of a mile east-by-south of Budaváda village. This specimen, which shows part of the body chamber of a good-sized individual, cannot be referred to any of the known rocks of that locality, and is unfortunately too fragmentary to admit of satisfactory specific determination. A long and careful search failed to throw any light on the origin of this interest- ing specimen, the only representative as yet of the genus Nautilus in the Rájmahál beds of the Peninsula. The Pavulur sandstones Nos. 7, 8, and 9 of the section form, as there Е shown, a small Bau approximately circular in shape and about a mile and a quarter in diameter. The thickness of this Pávulur group is small, but all the members not occurring together in any one section, the total thickness is doubtful. The beds roll about a little at low angles, but are here and there quite horizontal. The hard dark-coloured slightly shelly bed No. 8 is from 2 to 33 feet thick. The underlying drab sandstone was not pierced by any of the pits open at the time of my visit, but is probably not more than 5 or 6 feet thick, according to the quarrymen and the owner of the ground, which agrees with the estimate I formed independ- ently. Some of the blocks of the dark sandstone, of irregular shape, and covered externally with a brown weathering crust, have at the first glance a singularly trappean look, and I was for a moment startled by the idea that I had come upon a bedded trap. The drab sandstone No. 7 contains some intercalated flatly lenticular masses of hard dark sandstone, similar to that in the overlying bed. The latter dips gently eastward near the village of Pávulur, and is here apparently overlaid by the friable reddish-brown sandstone No. 9, which (ha) 14 FOOTE : GEOLOGICAL STRUCTURE OF THE EASTERN COAST. contains numerous obscure fragments of plants of red colour very similar to those observed at Annavallawarpalem in the Vemávaram patch. Obscure vegetable remains occur also occasionally in the lowest mem- ber of the Pávulur group, the drab sandstones No. 7. In the eastern part of the Budaváda patch no natural exposure of ` Eastern partof Buda. (һе Rájmahál beds was met with, and they are an revealed only in a few well-sections to the east of Pävulur, at Inkolu (Yinkolu) and a little distance to the south-west of Hanumazipalem. In all these the rock shown consists of soft sandy shales of buffy colour containing few or no traces of plant-remains. In proceeding northward from the Budaváda patch of Räjmahäl beds, we find it to be separated from the Idupulapadu Idupulapadu patch. З ; patch by a flat alluvial valley a little more than a mile in width. The plant-beds form a broad low whale-backed spur a couple of miles wide from north to south, and about 6 miles long from west-by-north to east-by-south. Here also nearly the entire surface is covered by cotton soil so thickly that but very few exposures of the Räjmahäl rock can be found. Only two genuine exposures were seen, the one a little to the west of Idupulapadu village, the other just north of Ambatampudi hamlet on the southern boundary of the patch close to where it is crossed by the path leading from the above-named village to Nakkalapalem and Budaváda. Both these exposures are too slight in depth to afford a section of any value, and no other information was procurable than from well-sections in the fields. In many of these, too, only the dug out material was available, the sides of the wells being inaccessible. Enough was seen to ascertain that the whole patch is made up of grey and white and rusty-brown mottled shales, which are doubtless an extension of the shale beds No. 6 north and north-east of Pävulur, which I regard as in their turn an extension of the Vemävaram beds. Fossils were found іп two or three of the well-sections west of Fossils at Idupulapadu ldupulapadu, and in one section in the fields қа Dron E belonging to Dronädula (Deranadula), about a mile and a half south-east of the village. The fossils found, which ( WM) Oe UPPER GONDWANA SERIES, 15 agree in type and mode of preservation with those from the Vemávaram beds, were as follows :— From IpUPULAPADU. FRow DRONADULA. Plants— Plants— Dictyozamites indicus. Dictyozamites indicus. Pterophyllum sp. Ptilophyllum cutchense ? Alethopteris indica. Do. acutifolium. Equisetum ? Angiopteridium spathulatum. Otozamites ? Araucarites macropterus ? Shells— Shells— Pecten sp. . Ammonites sp., fragt. Pleurotomaria sp. ? Astarte sp. ? Dentalium sp. ? Yoldia sp. ? and several obscure ill-preserved bivalves. In several cases the plant-remains and shells were found side by side in the same pieces of shale. АП the plant-remains are very fragmentary, but otherwise they are well preserved and easily distinguishable. The beds roll about to some extent at various but generally low angles. Yellowish buffy sandy shale underlies the cotton soil and lateritic gravel near the eastern extremity of the Idulapadu patch, but is only exposed in a well-section, which lies about three-quarters of a mile east-south-east of Duggubadu. A very small patch of Rájmahál beds separated from the Idupulapadu patch by the alluvium of the Dronädula nullah occurs immediately north-east of the latter village. It is separated to the north by another strip of alluvium from the large Punur patch of Rajmahäl rocks. -- As shown by the map, the Punur patch is of very irregular trilobed shape, but of considerable size. Like the other patches, it rises but little over the general level of the alluvium, and is almost everywhere thickly covered by cotton soil, The Punur patch. and the sub-rock is exposed only in artificial sections. The largest exposure of the rocks is seen in the bottom of the tank south of the village Cm) 76 FOOTE: GEOLOGICAL STRUCTURE ОЕ THE EASTERN COAST. of Punur, where mottled shales, precisely similar to those occurring to - the south-east of Dronádula and west of Idupula- Punur tank section. 3 padu, have been cut into about a couple of feet, and used in strengthening the tank mud. A cursory examination of these shales, which are moderately rich in organic remains, yielded— PLANTS. ANIMALS. Dictyozamites indicus. Astarte ? ; Ptilophyllum acutifolium. jg 2 AY u. Pterophyllum sp. West of the village the shales become more sandy in character and appear to pass into, or are replaced (underlaid) by Shales replaced by : : sandstones west of friable sandstones which must rest upon the X gneiss. Buffy friable sandstones are exposed, but very badly, in well-sections in the northern lobe of the Punur patch, at the villages of Gannavaram and Yanamaduh. The Gannavaram sandstone contains a great quantity of highly de- composed granite gneiss, and is very clayey. In the north-eastern lobes of the patch, buffy friable shaly sandstones are exposed under thick cotton soil, and some lateritic gravel in several well-sections between the villages of Zagariamudi and Tanubudivaripalem. The Punur shales extend fully a mile and a half north-eastward of the | exposure in the tank bottom just referred to, and have been turned up in the bund of a small tank by the side of the road to Polur. White stiff mottled fine-grained sandy shales have also been cut into in making the bunds of the tanks at Vankoelupadu and Nutalapadu in the eastern lobe of the Punur patch. Northward of the Punur pateh comes a spread of alluvium which hides all the Rájmahál rocks for a distance of 11 or 12 miles. The first patch of the juras- sic beds to be found here occurs at Kopparu, which stands at the northern end of a peninsula-like spur of slightly raised ground stretching south from the southern end of the Kondavidu hills. (doe) The Guntur group. Kopparu patch. UPPER GONDWANA SERIES. 77 Tke beds here exposed in various wells are sandy shales and thick- bedded friable buffy sandstones, apparently containing no organie remains. Similar friable sandstones oceur (also in well-sections) under lateritic gravel a little to the east and north-east of Karchola (Kartibola). That these beds and others associated with them extend further eastward under the alluvium is proved by the fact that similar beds, typical sandy shales, and white and mostly mottled clayey shales are penetrated by the deeper wells sunk through the overlying thick cotton soil and lateritic and kankarıy beds at Gorizagoluguntapalem and at A Ravipadu, lying respectively 22 and 44 miles Shales at Ravipadu, &c. à AUN east-north-east of Kopparu. These sections indicate the junetion of these beds with similar beds in the two Räjmahäl inliers of Chebrolu and Tangellamudi, on the western border of the Kistna delta. Returning to the westward we shallfind pale buffy shaly sandstone and sandy shales underlying lateritie gravel at Chinna Kondrapadu, 10 miles south-west of Guntür. The next sections of Rajmahal beds to be noticed occur at Gunttir d tte itself, where са 8 gritty, pebbly sand- stones are exposed in various well-sections in the western part of the town. These gritty beds consist entirely of gneiss debris. "Тһе best section is afforded by an old unwalled bowrie (square Sections at west end Well) close to the north gate of the compound pu belonging to the Judge’s house. About 12 feet of friable gritty sandstone of extreme coarseness, locally quite a conglo- merate, and penetrated by large numbers of kankar veins running in. various sets, is here seen. One chief set of these kankar veins appears to indicate the true dip of the otherwise unstratified bed. Тһе dip, if it be thus really indieated as supposed, is south-easterly at a low angle. No signs of any organism could be seen. A similar but much deeper and totally inaccessible section is to be seen close by in a large well sunk by Mr. John Rohde, c.s., when Judge of Guntür. QU 78 FOOTE : GEOLOGICAL STRUCTURE OF THE EASTERN COAST. Similar gritty sandstone was turned out in small quantity from the bottom of the municipal filtermg tank at the north end of the great irrigation reservoir known as the Black Tank. Separated from the Guntür sections by a band of alluvium varying from 54 to 94 miles in width are the two Rájmahál inliers, already Chebrolu and Tangel. referred to several times, of Chebrolu and Tan- Nail allen, А gellamudi. These two inliers form a low but well-marked ridge running 144 miles through the alluvial flat in a very nearly due north-east-to-south-west direction, with only one break through which fiows the Guntür nullah, which falls into the “Old Tungabhadra,” the most westerly branch of the Kistna in the delta. The ridge has an average width of about 2 miles, roughly speaking. It is the more conspicuous from being generally covered with red sandy soil, which contrasts strongly with the dark black regur which covers the surrounding alluvium flat. Here, as elsewhere in this region, no satisfactory section is to be No good sections tobe found showing the relation of the various rocks god, composing the inliers, and considerable uncertainty exists about their stratigraphical position. A friable gritty sandstone, in many respects a good deal like the Guntür beds, is exposed in a new well to the south- west of Kolakalur, at the extreme north end of the Tangellamudi inler. Тһе relation of this to the reddish and purplish, hard, gritty sandstones at Tangellamudi, and the soft and purple mottled Tangellamudi inlier. sandstones at Kazipett (Khabampeta), is quite problematie, and there are no intermediate sections to throw light on this point, for, in the three- quarter mile distance between Kazipett and Kolakalur and the 24 miles between the latter place and Tangellamudi quarries there is room for a variety of changes of position. Similarly, in the Chebrolu inlier it is quite doubtful whether the hard sandstones forming the northern half of the inlier lie under or over the soft shaly beds forming the southern half. . In this ease, however, I was able to form the conclusion that the hard GS) UPPER GONDWANA SERIES. 19 beds are the newer, as they are at the Pávulur end of the Budaváda section (see ante, page 69), but with regard to the Kolakalur soft gritty bed, I feel very doubtful what to think, though inclined to regard them as newer than the hard beds. The Tangellamudi sandstones have been quarried for centuries pro- bably, and great part of the bed removed in consequence. The sand- stones are best seen south-west of Tangellamudi and west of Shekur, where they can still be traced as a continuous bed, but much eut up by jointing, especially where the purple colour prevails. The beds roll about considerably, chiefly to the north-east and north-west. From 6 to 7 feet in thiekness of rock 1s exposed, but the underlying beds are not seen anywhere. The prevalent colours of the sandstone are reddish and purplish-brown, ranging to positive purple with yellowish-buff stains, Much quarrying is now going on, there being a large demand for the stone for many purposes. In the Chebrolu inlier the sandstones are best seen in the pits near E the old ld station, 3 miles north by east of Chebrolu, and in those immediately north of the village. In the former from 9 to 10 feet of rock only is seen 20 situ, and it is mostly of purple.colour, with white spots ; but buff and pur- ple mottled and purely buff colours are also to be found. The surface of some beds 1s formed by a thin layer of small pebbles. The sandstone underlying the lateritie gravel close to Chebrolu is a fairly massive compact gritty rock of purple, brown or reddish colour, tolerably thick bedded, and the beds generally pretty level. From the lie of the ground Iinfer that these hard sandstones overlie the friable mottled purple (or red) and white sandstones exposed to the south of Gundavaram village 13 miles to the north-west of Chebrolu. To the south-west of Chebrolu, where the ridge begins to sink, it is covered by cotton soil instead of red soil, and every thing is obseured till the village of Mutlur ! The Tangellamudi quarries appear to have been used by the Jains, for beautifully carved specimens of the sandstones, evidently taken from an old Jain temple, have been built into some of the walls, e. g., the northern gate, of the great hill fort of Kondavidu. CD Sections at Mutlur. 80 FOOTE : GEOLOGICAL STRUCTURE OF THE EASTERN COAST. (Mutnur) is approached, where a number of well-sections show the nature of the underlying rock. In a well east of the road to Chebrolu and a little to the north of Mutlur the following section is seen :— A. Alluvium. 3. Sandstone, friable, buff a et, 2. Sandy clay 1. Clay white (slightly ы yellow, red and dark-grey). The white clay is ehiefly below the water-level. Тһе beds appear to dip east-by-north, at 5°, On the highest ground more to the north of the village, several wells show rather friable gritty sandstones of pale mottled buff pink and pur- plish colours ; lateritic gravel 2/-5' thick rests upon these. To the west and south-west of the village the sandstones have a southerly dip, and the thickness of the lateritic gravel overlymg them in- creases up to 8 or 10 feet. None of the beds at Mutlur yielded any fossils, though they were carefully examined. The last case of rocks that might possibly be referred to the Räj- mahal series must now be mentioned. It occurs Ippatam conglomerate. k at Ippatam, 43 miles south-by-east of the southern end of the Bezwada anieut, and 3 miles east-by-north of Mangalagiri, where a very small pateh of coarse conglomerate, resting on the gneiss, disappears eastward under the alluvium of the delta. This conglomerate consists of pebbles of quartz, some of gneiss and a few of quartzite, united by a lateritic matrix showing true vermicular cavities, where the included pebbles are not very numerous. The bed dips at from 3? to 5° to east, and about 30 feet in thickness of 16 is exposed. No fossils were seen in it, so its age can only be inferred on other grounds. On petro- logical grounds it is not assignable to the Rajmahal beds, though its stratigraphical position is suitable. Petrologically it differs greatly also from the ordinary lateritic formations of the surrounding country, nor is there any stratigraphical connection with closely adjoining lateritie СШ) UPPER GONDWANA SERIES. 81 gravel which contains chipped stone implements. The great coarseness of this conglomerate and its stratigraphieal position both unite in sug- gesting that it may be of the age of the immensely coarse Rájahmundry conglomerates, recognized by Mr. King as equivalents of the Cuddalore sandstones. The only difficulty in regarding the Ippatam conglo- merate as representative of the Rájahmundry beds, lies in the con- siderable distance between the two formations without any intermediate link. It will have been seen in the foregoing pages that the beds which a T may, with good reason, be regarded as extensiong tür group with the Vem- of the Vemávaram group can be traced northward ávaram Budaváda group. 7 i from Vemávaram fora distance of more than 20 miles, and the evidence of the fossils obtained at Budaváda, Dronádula, and Punur quite confirms the deduction made on stratigraphical and petrological grounds. Beyond this point petrological resemblance is the only guide in correlating the different groups of Rájmahál rocks that have been met with in the Guntür country. "The petrological resemblances to the southern rocks found in these northerly patches justify the assumption that the triple sub-division found to hold good in the south may reasonably be extended to the north. The sandstones of Karti- chola and Kopparu, and the gritty beds of Guntür, may be fairly classed with the sandstones of Annavallawarpolliam, Budaváda, Punur, and Gannávaram ; while the shales of Ravipadu, Gorizagoluguntapalem, and Mutlur represent the Vemávaram shales, and the purple sandstones of Chebrolu and Tangellamudi represent the dark sandstones of the Pávulur group. The Upper Gondwána beds occurring north of the Kistna have been a. OR similarly divided by Mr. King (Records Geological Ongole Rájmaháls with Survey of India, Vol. X, page 56; 1877) into Godävari Rájmaháls. 86 > 4 3 three sub-divisions called respectively the— Tripetty sandstones. Ragavapuram shales. Gollapilli sandstones. eo FOOTE : GEOLOGICAL STRUCTURE OF THE EASTERN COAST. Whether this triple grouping will fit in with the triple division made ЕА out by me in the Upper Gondwánas south of the Kistna, remains to be seen when all the fossils from both regions have been fully worked out; but I thinkit is very doubtful. From the preliminary examination of the fossils, Mr. King and Dr. Feistmantel incline to consider the Ragavapuram shales as the equivalents of both the Vemávaram group of the Guntür-Ongole region and the Sripermatur group of the Madras region. Dr. Feistmantel regards the Ragavapuram beds as equivalents also of the Utatár plant beds of the Triehinipoly district. | My present view on this subject is, that the Vemávaram shales are not exactly the equivalents of the Sripermatur group, and therefore also not exactly equivalent to the Ragavapuram shales, but are of rather greater age, and represent the Golapilli group, and may therefore be regarded as true marine equivalents of the typical Rajmahdl beds of Bengal. Dr. Feistmantel has, in his description of the flora of the Gola- pilli beds, in the Paleontologia Indica,’ stated his belief that the Utatür, Sripermatur, and Ragavapuram plant-beds are younger than the true Raj- mahal beds. He groups the Vemávaram shales in the same category ; but at the time he stated this, he was, owing to an unlucky confusion of the collections from Sripermatur and Vemévaram, under the impression, strongly supported by the similarity of the two sets of shales, that both came from the same region. Judging the case by the evidence of the fossil plant-remains, it appears to me that the Vemavaram shales show a facies distinctly older than that of the Sripermatur group, and contain a greater number of characteristically Rájmahál plants than do any of the other plant-beds in the peninsula, Mr. King’s Golapilli sandstones excepted. This view is Qo on GP Veni- borne out by a comparison of the Golapilli plants varam and Golapilli fos- as compared with those obtained from the Vemá- ils. pr varam shales. The list of Golapill plants I take 1 Ser. II, 3, p. 3. UPPER GONDWANA SERIES. 83 from Dr. Feistmantel’s Notes on the Fossil Floras of India, No. II!; that of the Vemávaram fossils from my Notes on the representatives of the Upper Gondwána series.? GOLAPILLI. Alethopteris indica. Asplenites macrocarpus. Gleichenites bindrabunensis. Angiopteridium spathulatum. Ditto ensis. Pterophyllum morrisianum. Ditto carterianum.. Ditto comp. distans. Ptilophyllum acutifolium. Ditto cutchense. Dictyozamites indicus. Williamsoma sp. Palissya pectinea. Ditto oldhami. Eehinostrobus sp. Araucarites sp. VEMÁVARAM. Sphenopteris sp. Dicksonia sp. Cyclopteris ? sp. Alethopteris indica. Angiopteridium spathulatum. Macroteniopteris ovata. Ditto sp. Pierophyllum distans ? Ditto Sp. Ditto Jissum. Zamites proximus. Ptilophyllum acutifolium. Ditto eutchense. Otozamites sp. Dictyozamites indicus. Palissya indica. Cheirolepis ? sp. -Echinostrobus rajmahalensis. Araucarites macropterus ? Cunnighamites dubiosus ? Coniferous leaves, undetermined. It will be seen from these lists that, out of the sixteen forms described from Golapilli, eight, 1f not nine, are also found in the Vemávaram beds ; and of the eight common to both series, five occur numerously at, and are eharacteristic of, the Vemavaram series. These are Pterophyllum dis- tans, Ptilophyllum acutifolium and. cutchense, Dictyozamites indicus and Echinostrobus raymahalensıs. In a private letter I received on this subject some little time since from Dr. Feistmantel, he dwells on the absence from the Vemávaram beds of the broad fronded Macroteniop- teride : this is true; but it is an objection which applies equally to the ! Records Geol. Surv. India, Vol. IX, p. 39, 1876, and Pal, Ind, Ser. II, 3. 2 Records Geol. Surv. India, Vol. ХІ, p. 259; 1878. (28И) 94 FOOTE : GEOLOGICAL STRUCTURE OF THE EASTERN COAST. Golapilli beds. If the large Macroteniopteride are wanting at Vemá- varam, per contra we have a far larger development of the pre-eminently Rájmahál genus Dictyozamites than at Golapilli or anywhere else out of the Rájmahál hills. The Vemávaram Dictyozamites may have to yield the palm in point of mere size to some of the specimens from Amrapura, in the Rájmahál hills, figured by Oldham and Morris in plate XXIV of Series II of the Paleontologia Indica; but several of the fronds I collected (at Vemávaram) were in every other respect far more perfect, and quite as beautifully preserved. Although I cannot accept Dr. Feistmantel’s and Mr. King's views, that the Vemávaram shales are the exact equivalents of the Ragavapuram and Sripermatur shales, I quite agree to their idea of the equivalency of the two last named groups, and regard them as younger than both the Golapilli and Vemá- varam beds. The most characteristic plants of the Sripermatur group are P/;/o- phyllum acutifolium and cutchense, and a conifer that agrees with none as yet figured from the typical Rájmahál beds. Angiopteridimm spathu- latum is also a by no means uncommon plant at Sripermatur, and it can- not be reasonably contended that the Sripermatur and Vemávaram groups agree in the facies of the fossil plants they contain, but rather that they differ considerably, more so, indeed, than do the Vemavaram and the Utattr plant-beds. No division of the members of the Rájmahál beds in the Ongole and Kandukur groups was found practicable. CHAPTER V.—CUDDALORE (RAJAHMUNDRY) SANDSTONES. The rocks to which the above title is applicable, if they occur at all within the Guntür-Ongole region, are developed Ippatam conglomerate. close to the Kistna. This one spot lies near the village of Ippatam, 4$ miles south-by-east of the Sitanagaram end of the great anicut. Here oecurs a small patch of immensely coarse conglomerate, with (75477) but toa very small extent, and only in one spot, LATERITIC ROCKS. 85 а lateritic matrix, which conglomerate has already been referred to and described when treating of the Upper Gondwána rocks (see page 80). As there stated, nothing but speculation can be advanced as to the real position of the small patch of conglomerate which differs so greatly from all. the other formations within the area of this memoir, no clear evidence, whether paleontological or stratigraphical, being avail- able. The petrologieal evidence amounts to a mere probability, but a very reasonable one, that this patch of conglomerate may be an outlier of the once far more widely extended Rajahmundry conglomerates. These conglomerates are regarded by Mr. King as northern equivalents of the group established in the south by Mr. H. F. Blanford, as the Cuddalore Sandstone.! CHAPTER VI.—THE LATERITIC ROCKS.’ The formations that have to be grouped under the above heading cover a not inconsiderable space in the Guntür-Ongole region, but they are mostly thin superficial deposits, and really of small importance, but that in parts they give character to the general surface of the country. As already mentioned, they form part of the band of sedi- mentary rock which generally lies between the old gneissic rocks and the recent coast alluvium. In addition to the areas shown in the map, this group must include sundry patches and remnants of patches of shingle, mixed or unmixed with lateritie particles, which are scattered in various parts over the surface of the older rocks, especially the gneiss. From their exceedingly irregular (ragged) shape and want of definition of boundaries, these shingle patches could in hardly any cases be satisfactorily mapped, but their presence has been indicated on the Atlas maps, either by writing the words “ shingle" or “gravel,” or by a special mode of marking ; on the reduced skeleton map attached to this paper, most of these small patches had to be omitted. 1 Memoirs Geol. Sury. India, Vol. IV, p. 165, (ou 86 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. In their geological age all these deposits belong to the recent period, Lateritie rocks of те. having been formed since man’s advent upon earth, genuas as proved in many cases by their containing traces of man’s handiwork, in the shape of chipped stone implements of paleolithie types. Wherever such implements were found, beds of ovavel were also found, or traces of their having once shingle and g existed there, or at no great distance. By far the greatest development of lateritie deposits and implementi- ferous gravels oecurs in the southern part of our a area. Out of five prineipal areas of lateritie rock, four lie south of the Gundlakamma river; the fifth lies nearly 50 miles further northward, and forms an appendage as it were to the south- eastern flank of the Kondavidu hill range. These principal lateritie areas may be conveniently called after the chief places standing on or near to them, and beginning in the south, we find :—1, the Rámapatam (Ramiaputnum) area; 2, the Kandukur (Cundacoor) area ; 3, the Devagudur (Davagoodoor) area ; 4, the Ongole area; and 5, the Kopparu area. Besides these larger areas there are sundry small detached patches or outliers, and a considerable number of fringes surrounding the older rocks, especially the various patches of the Rajmahal rocks. The shapes of the larger areas as well as of the fringing patches are so varied and complicated that it would be useless to waste time in a hopeless attempt to describe what can only really be learnt by inspection of the map. A brief description of the petrological features of the principal areas and patches mapped and unmapped must suffice. 1. The Ейтарайат area lies between the sea at Rämapatam and the alluvial valley of the Man-eru and its lower tributaries. A little to the north-west of the town the laterite approaches within a mile of the sea. It forms a low undulating plateau largely covered with thorny scrub jungle. The lateritic formations are generally very ferruginous, whether they The Ramapatam area. be sands, gravels, or conglomerates. The gravelly form covers by far (0) LATERITIC ROCKS. 87 the larger area, and is generally only 3 or 4 feet thick, and fre- quently pierced by rugged masses of the underlying gneiss. Only a few of the larger protrusions of gneiss could be shown on the map. The soil is almost everywhere of an intensely red colour, and where the water-supply is sufficient, of very considerable fertility. The lively contrast between the bright red soil and green jungle or fields gives this tract of country an eminently cheerful appearance. In the south-western parts of the area, much of the lateritic gravel has been formed of washed-up purple sandstone and grit of Rájmahál age. Both rocks were highly ferruginous (hzmatitioe), and in some fragments are to be found remains of the typical fossil plants. This is particularly the ease between Rázpalem (Razpolliam), Chinna Lätarapy, and Peddávaram (Peddawarrum), where the lateritic deposits often rest upon Rájmahál rocks. Near Rämapatam the laterite, which is there a typical eenglomerate, shows an easterly dip of from 2? to 3°. It contains no foreign substances, but a few quartz fragments; but further west, between Potelur and Gudlur (Goodloor), it includes many angular quartz fragments, some of quartzite (broken pebbles) and a number of rather Chipped stone imple. Poorly made chipped implements and flakes, also ments. of quartzite, Though generally thin, the lateritic beds here and there show considerable thickness ; as at Allampuram, east of Chinna Latarapy, where from 12 to 15 feet of intensely ferruginous gravel is cut through by a well-section. The conglomeratic form was specially noted near Lutchmepuram in the south of the area, and near Virapalle (Veerapully) and Bhimavaram (Beemawaram) in the north. The most striking and typical development of massive laterite conglo- merate in this region is to be seen in a low hill immediately east of the great northern road and 8 miles north-west of Rámapatam. It is more clayey and less ferruginous than much of the conglomerate seen else- where. 2. The Kandukur area.—The lateritic formations within this area are principally sands and gravels. Conglomerates MI sane are rare, the only really striking examples being (5) 88 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. the spreads which cap the low gneiss hills at and around Singáraya Konda, a noted temple just off the great north road a couple of miles north of the ford over the Man-eru. The sandy forms of the lateritie formations are specially well seen to the west and south-west of Kandukur and around Pokur, 6 miles fur- ther south-west. The sands, especially near the latter place, are intensely red in colour and highly charged with minute pisolitie grains of 1mpure red or brown hematitic clay, highly indurated and coated with an external Shingle bed at Mala- glaze. Gravel and shingle beds, often of extreme palle, &e. coarseness, are very commonly met with, e.g., near Malapalle (Malapally), west of Singáraya Konda, near Anantásagaram, west and north-west of Kandukur, at Kondasamudram (Condasamo- drum), Sakávaram (Sacawarum), and Kálamalla (Calamulla), villages from 8 to 10 miles south-westward of Kandukur. At some of these places as Anantäsagaram and Kondasamudram the shingle is partly conglomerated by a lateritie cement, but at the others, and especially at Malapalle, the shingle is quite loose. At Malapalle the shingle bed is very remarkable for its extent, which is between 3 and 4 square miles, and for its great coarseness. It consists mainly of rolled vein quartz, but contains also numerous large pebbles of gneiss and quartzite. The thickness of the bed may be estimated safely at from 20 to 30 feet. In many parts the name of boulders would be almost a better definition of the water-worn masses than the term shingle. The ferruginous element is entirely wanting here, and the prevalent colour is white. Much of the surface of the Kandukur area is covered by thick cotton soil. The westernmost extension of the lateritics up the valley of the Pal- eru shows mainly coarse shingle beds with little or no ferruginous matter. 3. The Devagudur area.— Here the deposits, which are almost entirely of gravel and shingle, are greatly obscured by cotton The Devagudur area. a Я ; ў soil, and no sections worth mention were met with. The ferruginous element is almost entirely wanting throughout this area. "The shingle beds are most exposed along the northern side of the area between Venkanapalem and Ilavarra (Yellavurra). On the south (25227) LATERITIC ROCKS. 39 side of the area, at Kattambadipalem (a hamlet not shown in the map), 3 miles west of Devagudur (Davagoodoor), the shingle bed appears to rest upon a bed of saline clay which in that case must be reckoned as of lateritie age, for it certainly does not belong to the Rájmahál series. The section is, however, obscure, and the grey clays may really be part of the river alluvium and the overlying shingle merely a washed-up bed. 4. The Ongole area—The characteristic ferruginous element of typieallateritie beds re-appears again in this area, The Ongole area. ! i у owing doubtless to the presence of rich magnetic iron beds in the adjoining gneiss rocks of the Konijedu, Parnametta, and Ongole hills. The western and northern half of thearea is entirely covered with thick cotton soil, but the southern half is occupied by bright red soil resting on lateritic gravel of very ferruginous character contain- ing chips and pebbles of quartzite. The laterite in the western part of the area, where seen, is always gravelly in character; but the coarse shingle so common in the Deva- gudur and Kandakur areas is here found no more. 5. The Kopparu area.—Only the gravelly form of laterite is here seen in a few well-sections, the general surface of Kopparu area. à 2 the country being thickly covered by cotton soil. The laterite, which rests upon Rájmahál beds, extends eastward under the alluvium as far as Ravipadu and doubtless beyond, but the alluvium, which is pure cotton soil, then becomes too thick to be piereed by the village wells, or rather the water level of the country is reached before the cotton soil 1s traversed. Of the smaller areas, patches, or outliers of lateritie rock, and of the fringing beds, but little of special interest can be recorded. The few remarks required will be given while taking up the several patehes seria- tim in geographical sequence from south to north. The first in order is the Parlapalle (Parlapully) patch, lying south Parlapalle patch. of the Vupput-eru and under the parallel of 15° north latitude. The rock here consists of mottled ferruginous clay, overlaid by kankarry clay and very red sandy soil, which becomes (48915) 90 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. covered with cotton soil on the higher ground to the south. Both this patch and the Rámapatam area are shown on the map as extending south beyond the limits of sheet 76; this is correct, but the extension is for a very short distance only, as the gneissic rocks appear rapidly with the rise of the ground. Proceeding north to the valley of the Musi (Mooshee) river, the three small patches near Annakarlapudi (Unnakarlapoody), Maddalur (Muddaloor), and Chilikipad consist of coarse shingle, but are much obscured by cotton soil. The Amanabrolu (Ammanabroloo) laterite area, 9 miles north-east- : ward of Ongole, forms a low flat down, which Amanabrolu patch. d f У ? is quite bare of vegetation opposite Sowtapalem. It consists generally of lateritie gravel, but a true conglomerate with included quartzite and gneiss pebbles, and numerous fragments of the Rájmahál rocks, occurs in the southern slopes. To the west and north everything gets soon covered up by thick cotton soul. The fringing beds around the Räjmahäl patches of Budaváda, Idupu- lapadu, and Pánur, and the small outlier at Parachur, on the old Madras- Guntur road, all consist of lateritic gravel much mixed with and obscured by cotton soil. 16 is only here and there over small surfaces that the beds are conglomeratie and consolidated. The laterite at Guntür is partly gravelly, partly conglomeratie, the l latter variety being best seen to the south-east of Guntür laterite. j the town, close to the alluvial boundary. In the town the laterite rests on soft gritty sandstone of Rájmahál age, but to the west and north-west 16 overlaps on to the gneiss. The patch of gneiss which runs up from near Guntur to the bank of the Kistna is fringed by gravelly laterite more or less all along its eastern side. The two Räjmahäl inliers in the alluvial spread adjoining the Kistna delta to the east of Guntür both show extensive fringes of lateritie gravel. Тһе northern or Tangellamudi inlier is fringed nearly all round. Тһе southern or Chebrolu inler is fringed simi- larly in its northern half and at its extreme southern point. The (2507 LATERITIC ROCKS. ol intermediate part is very probably equally fringed with the lateritic gravel, but its whole surface is completely masked by thick cotton soil. The occurrence of stone implements of -palolithic type in the Paleolithic stone im. laterite at Potelur in the Rémapatam area has plements. already been mentioned (page 87). From the Kandukur area several very fine implements were obtained, to the east and west of Kandukur itself and from Kondasamudram. Others were found in the northernmost part of our area, in the fringing laterite at Ippatam, close to the Kistna. A much larger number, however, was obtained from some of the outlying unmapped shingle patches, the more important of which will now be enumerated. They are especially numerous in the valley of the Man-eru and its tributaries, where they occur at intervals all up the course of the river, at levels above the recent alluvium, e.g., at Nàkanampetta, near Maregunta, around Velegunla (Valegunlah), at Sallawarpalle, and at Kambaldinna (Cumbaldinna), also at Kattakindäpalle (Cuttakindapully), Mupád (Mopaud), Chintalpalem, and Kothápalle, all on the south bank of the river; nearly all these beds yielded implements. Of the places on the north bank, two especially, Lingasamudram and Comarapallem (Bompeadopaud of map), yielded many good impiements, though but little remained of the shingle beds they had been preserved in. Further west still, implements were found associated with the shingle beds near Irur (Eroor) and Iskapalle, and in the shingle bed which lies hike a talus at the eastern foot of the Vaimpad quartzite hills south of Pamur. In parts these beds are so immensely coarse as quite to deserve the appellation of “boulder gravels.” They are generally more or less ferruginous. A remarkable shingle deposit, in which, however, no implements were found, occurs on the banks of the Chundi nullah, about 31 miles east-by-north of Chundi village. The shingle, which is of gneissic origin, is far too extensively developed to be attributed to the formative power of the local stream. The ferruginous element is greatly wanting in this case. И) 93 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. Considerable spreads of coarse gravel and shingle, the ruins as 16 were of once existing beds, remain scattered over the gneiss to the north and west of Maddalur (Muddaloor), in the valley of the Musi-eru. Тһе most westerly point north of latitude 15° at which I found im- plements in this region is Ramiapalle (Ramiahpully), 23 miles north- west-by-north of Pámur, where considerable traces remain of a former extensive development of lateritie beds in the shape of lateritic gravels and dark red sandy clays. The surface of this tract, and indeed of the whole area drained by the headwaters of the Pal-eru, is much covered and obscured by drifts of bright red blown sand, which will be again referred to further on. . The last shingle bed to be mentioned occurs far north of al the others at the village of Angulur Agranaram, 9 miles north of Vinu- konda. Неге a highly kankarry shingle of gneiss and quartzite contains many rude, and some good, paleolithic implements. A strong petrological resemblance exists between many of the non-ferruginous shingles and gravels just described, Resemblanee of non- і ; ferruginous shingles to and the equally non-ferruginous Conjeveram gra- са саа MO vels of the Madras агеа but they (the northern beds) are not restricted to one particular tract, as are the southern ones, and there is no reason to look upon them as in any way distinct from the ferruginous beds, their mineral character only excepted. CHAPTER VIL—THE ALLUVIAL DEPOSITS. A very large area of alluvium lies within the limits of the tract described in this memoir, as it includes the major part of the delta of the great Kistna river. A minimum of time was available for the alluvial deposits, so the examination they underwent was necessarily quite eursory, but that was enough to give an idea of their general 1 See Memoirs Geol. Surv. India, Vol. X, p. 41. ( 99 ) ALLUVIAL DEPOSITS. 93 character. They must be grouped as of two great classes, marine and freshwater, the marine including the brackish water estuarine formations. Marine and astua. But few data have been collected to show how far rine alluvia. inland from the present coast line marine alluvia exist. 'The most recent information on this subject was obtained dur- ing the extension of the Kommamur canal, which starts from the southern end of the Bezwada anicut to meet the East Coast canal now known as the Buckingham canal. The beds cut through in excavating this сапа! in the southern part of the Bapatla taluq were largely marine, as shown by numerous subfossil shells exposed. A small collec- tion of these’ shows the following species :— Murex, sp: Turritella. Eburna, sp. Placuna placenta. Oliva, sp. Arca granosa. Voluta, sp. Cyprina, sp. Natica, sp. . Cytherea castanea. Terebralia telescopium. These were obtained at Santaravur, in dark grey clay associated with considerable quantities of selenite crystals in large complicated macles, which were found a great source of discomfort to the coolies, by cutting their bare feet. The shells are mostly in excellent preservation. The furthest point inland at which marine beds are positively known Marine fossils found tO occur is at, Golabapilli in the Gudivada taluq, шы a place a little outside the northern limits of the map illustrating this Memoir, and lying at a distance of about 90 miles from the nearest point on the coast. Fossils were here obtained by Mr. Peters, C. E., D. P. W. Among them were a good-sized crab, a shell of the genus Mactra (?) and a branch of Nullipore. These fossils also were found in a bluish-grey clay ; unfortunately no detailed information concerning them had been noted, and the section from which they were obtained had been filled up again. 1 I am indebted for this collection, as well as for much courteous and kind assistance in other ways, to Major (now Lieut.-Colonel) Hasted, R.E., who was Superintending Engineer of the Kistna and Godávari ranges during the time I was surveying in that neighbourhood. (За 94. FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. From the materials thrown up into a dyke near the light-house at Subfosil shells from Masulipatam, a considerable number of subfossil ln shells were colleeted by me. The following is an incomplete list of the genera obtained from the sandy clay :— Pyrazus. Arca, 3 sp. Potamides. Cultellus. Nassa. Solen. Nerita. Tellina. Valvata. Mactra. Paludina. Pholas. Lymnea. Ostrea, З or 4 sp. The swampy flats near the coast abound in shell-fish ; those within tidal influence show brackish-water forms, as Potamzdes, and those fur- ther inland purely fresh-water forms, as Paludina and Planorbis, some- times in enormous number. The eastern part of the delta has a sandy surfaee; the western part is covered with dark black washed up regur (cotton soil). The boundary line between the two runs through Weeyoor in a north-easterly direction. My attention was drawn to this by Lieut.-Colonel Hasted, R.E., who also informed me that this boundary is distinctly marked for a long distance, the western edge of the sands forming a slightly elevated ridge across country. It would be very interesting to ascertain whether this dis- position of the sands depends in any way on wind action, or can be connected with the great tidal waves which have at long intervals devastated the delta, as in the terrible eyleone of 1864, when the sea water rushed inland for a distance of over 15 miles. Considerable changes have taken place in the form of the coast line within the memory of living men. There is a marked difference between the lines laid down by the surveyors from whose survey sheet 95 was compiled and the limes ascertained a few years since by the officers of the Madras Revenue Survey. As might be expected, the delta is advancing on the sea, and would do so much faster but for the very powerful coast currents which flow:up or down according to whether ( 94 ) \ ALLUVIAL DEPOSITS. 95 the south-west or north-east monsoon шау be prevailing. The mouth of the Masulipatam river is now quite different from what is shown in Atlas-sheet 95, edition of 1828. A long spit has been formed which has thrown the bar much further to the north. In the region round the mouth of the Gundlakamma river, traces of marine beds were noted at considerable distances inland at three places: to the east of Uppu Gundur ; also at Vinnawaddaroydipalem, on the north bank of the river, where subfossil marine shells are found in a clay bed about 4 miles inland; and at Biramgunta, on the south side of the river, where similar subfossils occur at a distance of 8 miles from the sea. The sandy character of the sea board holds good generally down the coast as far south as Ramapatam. To turn to the fresh-water alluvia: they are all of river origin in this part of India, and represent largely the char- acter of the soil prevalent in the several hydrolo- gical basins. Thus the alluvium of the Kistna, which flows through enor- Fresh water alluvia. mous tracts of cotton soil inthe Deccan, consists Characters of alluvia А : of different rivers: ofthe mainly of washed-up regur. Gritty sand and sands ee are comparatively rare. The Gundlakamma allu- vium is chiefly sandy, and the alluvia of the other small rivers to the south of it are variable, according to the diversified tracts they have drained. The alluvial basin of the Gundlakamma is the most interesting of the smaller ones. The river has changed its course near its mouth, and flows mainly through the southern arm which flows past Gundiapalem, and was an unimportant branch when the trigonometrical survey was made. The northern or Pedda Devarampadu branch, then the main вітра 1s now nearly aban- doned and fast silting up. A singular mistake occurs in the old Atlas-sheet 76 in connection with the Gundlakamma valley, a couple of hills being shown close to the river, near the crossing of the road from Ongole to Ammanabrolu. These hills have no real existence; their site is a perfect flat, covered with reddish-sandy loam. Of the Gundlakamma. (3595 05 96 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. Further up the river the banks become cliffs, and continue so gener- ally up to Venkateshwarapuram, where a reef of gneiss crosses the Human bones at great ‘iver. About half a mile east of the village of depth in alluvial cliffs. — Velampalle (Valumpully), where the great north trunk road crosses the Gundlakamma, the gritty alluvial sand cliffs were found to contain human bones imbedded at very considerable depths below the present surface—depths so great as quite to preclude the idea that the bones might be connected with any recent, or quasi- recent, burial. The first bone found was a lower jaw imbedded in an undisturbed bed of loamy sand 18 or 20 feet below the present surface. This find set me looking for more, and I succeeded in finding others im- bedded in equally undisturbed sands at depths from 16 to 18 feet below the present surface. The other bones consisted of a scapula, femur, tibia, fibula, humerus, ulna, radius, and a few doubtful fragments belonging to more than one, and probably to three individuals. "The bones present a rather recent aspect, no infiltration of mineral matter having taken place. "They owed their position in the alluvium doubtless to flood action at a period when the Gundlakamma was forming the great flat it now cuts deeply into, and flowed at a level of from 30 to 35 feet higher than at present, or was subject to floods of vastly greater magnitude than those now occurring. Thin beds of gritty silt are intercalated with the sandy loam both below and above the sites of the bones, and in them are numerous Unios and Melanias of the same species as now live in the river. The alluvium of the small rivers draining the country: between the - Alluvia of the smaller Kistna delta and the Gundlakamma consists mainly an and воша, of washed-up cotton soil, though sand, sandy loam, | and silty grits are occasionally met with. Gravels are rarely seen. The streams draining the great cotton soil tract lying north and west of the Kondavidu hills form remarkably distinet and striking alluvial flats along their courses, and, as might be expected, these flats show hardly anything but washed-up cotton soil. The river side cliffs frequently show beds of the dark reddish-brown clay underlying the top bed of washed up regur. This may be seen in. (240) ALLUVIAL DEPOSITS. 97 all the rivers. Unfortunately no organic remains were found in any of these lower clays. i The alluvium of the Man-eru (Mun Air) is by far the most sandy Sandy alluvium of the Of all the smaller rivers, a fact quite in keeping E with the general character of the area drained by it, which show less regur than do the basins of the other rivers. Much of the red sand covering the flats near the junction of the Man-eru and its principal branch, the Vupput-eru, is washed-up lateritie sand, and so red in colour as to make it very difficult in many places to distinguish the boundary between it and the true lateritic sands. At Gundlapalem, as at various other places along the lower reaches of the river, the banks are high, and cut into small cliffs showing red-brown loamy sands 20 to 30 feet in thickness. In the upper part of its valley the Man-eru cuts through two large flats ub of a very kankarry gute вао, quite Bats the reddish loamy alluvium the river is now forming on a small scale in a few spots above local barriers of gneiss. These flats are mentioned here as the shingles might, from their situation, very easily be taken for old river alluvia, but the probability is they are really of lateritie age, and implements were found on the surface of the upper flat. These flats, which have already been referred to (page 91), are situated to the north and east of Pámur, at Chintalapalem, near Kothapalle and Mupad respectively. | The alluvium at the junction of the valleys of the Pal-eru and Musi- eru is largely sandy, but higher up the valleys of both rivers cotton soi] predominates markedly. CHAPTER VIIL—SOILS AND SUB-AERIAL DEPOSITS. Three classes of soils are met with in the Nellore-Kistna country—the black, red, and white; but of these only the two first are of any import- Distribution ofredand ance. The black soil, cotton soil, or regur, covers the Dur: largest surface, and predominates in the northern G (OED 98 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. and eastern parts of our areas; the.red soil, or lál, prevails over the southern and western parts, but is also to be seen almost everywhere around the bases of hills, or where the ground is very broken. It is the direct produet of decomposition of ferruginous rocks. Cotton soil, on the contrary, is an indirect product, as it contains a much larger admixture of organic matter. Its great development over such large areas may in all probability be attributed to the former existence of large and thick forests, and to the former prevalence of a moister elimate than now exists. Cotton soil an old The eotton soil was the humus formed in such Durus forests. It overlies all the formations of these regions indiseriminately, but yet shows no signs of aqueous deposition except in the manifestly washed up beds in the alluvial basins. Much of the regur, however, which lies on the surface of the river and coast- alluvia was formed 7% situ as a humus; at least this appears to be the only explanation of its occurrence over such extensive surfaces in the absence of all traces of any transportation by water and deposition as a true sediment. The sub-aérial deposits met with ean all be referred to two classes, the results of chemical and of mechanical action,—the former including all the tufas (kankar, &e.), the latter represented by the blown sands, The tufas, including all the numerous forms of kankar, are met with Sub-aérial deposits. e pretty well everywhere and in every formation as foreign bodies introduced as products of decompo- sition or by infiltration. The commonest form of kankar is the gravelly б. or nodular, which forms so large an ingredient at Varieties of kankar. ў the base of nearly every cotton soil throughout the country, and which plays so large a part in so many of the less ferru- ginous gravels of lateritie ages as already pointed out. It is needless to specify any examples of a formation of such exceedingly wide distribu- tion and common occurrence. Infiltration kankar which fills the joint clefts, and all possible eracks and crannies in the older rocks, and which in innumerable cases cements ( 98 ) SOILS AND SUB-AERIAL DEPOSITS. 99 together loose materials of all ages and descriptions, forming conglomerates and breccias of all characters, is the next in importance, and probably also in point of quantity. Lastly come the massive tufas usually forming superficial deposits, which are occasionally of some little importance. These appear to be in many, if not in all, cases to be really only a fuller consolidation, or a completion of the process to the partial action of which the other two forms of tufa are due. Examples of this are also very common, but a few of the more striking examples Massive kankar near deserve to have attention drawn to them. An Tamur. extensive pavement-like spread of massive tufa occurs a couple of miles west-by-north of Pámur, under the bund of the Dupuguntla (Doopoogoontla) irrigation tank. Several small springs issue from below this mass, which rests upon gneiss. Another 2146; remarkable and extensive spread of tufa occurs at At Kanigiri. о the Kalingula (waste weir) of the large tank at Kanigiri, and extends for a considerable distance westward under the red soil. Another great pavement of kankar is to be seen at the north-eastern end of the Kotappa Konda, and another of the same character at Yallamanda, a mile or so to the north- east and 4 miles south of Narasaraopett. The surface of the tufa here shows a remarkable coralloid sculpturing unconnected with any internal structure, but due apparently to the action of rain. Near Kotappa Konda. There is a great development of hard sub-crystalline kankar tufa north of Naganla, and about 33 miles north of Budavada. Неге the junction of the gneiss and: overlying Räjmahäl beds is completely concealed by the thick bed of very large quasi-stalaemitie masses. At Naganla. The last example of massive tufa is perhaps the most remarkable of all. It occurs along the brow of the high ground south of the Pal- eru valley, З miles northward of Kandukur. It is best seen at Konda- kandukur, where it forms sheets as it were in the bed of the large tank, (9) 100 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. The sheets of this бла, which is very hard, close-grained, and of yellow- um ish-white colour, show a lentieularly concretionary structure, which gives rise to holes and depressions in the surface ; no trace of any organism could be found init. At Kovur (Covoor), 2 miles to the west, it is seen to rest upon shales of Räjmahäl age, but the section is too poor to show the real relation of the beds. It is the densest and most massive tufa I am acquainted with, and but for its concretionary structure, I should regard it as a sedimentary limestone, in which case it would doubtless belong to the Rájmahál series. BLOWN SANDS. These /Eolian accumulations are more commonly developed in the southern: part of our area than in any other equally large distriet that I am acquainted with. But though so commonly met with, they are mostly of very moderate dimensions, and, except in a few cases, of no particular interest. Three sets of blown sands may be recognized: Istly, the coast dunes and flats; 2ndly, the river-side dunes; and 3rd/y, the inland dunes uncon- nected with any river beds. Blown Sands. Of the coast dunes but very little need be said; those lying be- tween Rámapatam and the southernmost part of the Kistna Collectorate attain to no height, though the spreads of loose sand are often more than a mile across. Of those in the Kistna district only a very small part in the extreme south, at Pedda and Chinna Ganjam, and those immediately to the north-east of Masulipatam, were visited. Тһе remainder were for the present left · untouched, but they had already been mapped by the officers of the Madras Revenue Survey, whose representation of them has been adopted and shown in the map accompanying this Memoir. It was considered unnecessary to devote to geologieal features of such very minor im- portance the considerable period of time that would have been requisite to : travel over such a difficult country as the sandy and swampy sea edge of the delta of the Kistna. 'The sand hills at Chinna Ganjam and near © 100) Coast dunes. SOILS AND SUB-AERIAL DEPOSITS. 101 Masulipatam attain in parts a height of from 30 to nearly 50 feet, and where overgrown with pandanus bushes, cashew-nut trees, and other sand- loving plants, give rise to picturesque peeps one would hardly expect among sand dunes. A belt of Palmyra palms (Borassus flabelliformis) accompanies the coast sands with hardly a break. The moving of the sands can easily be checked by planting casuarinas, or the trees named above. Small sand-hills are often to be seen at the bends of the different | smaller rivers, where long reaches of sand, dry during the greater part of the year, are exposed to the strong land winds. Several of the larger are shown on the map, but many were seen too small to be mapped. Some of these river-side dunes are troublesome, as the sand advances from them over the fields and covers them injuriously. No attempts appear to be made to plant these small dunes to fix the sand. The self-sown Ipomea (bindweed) and wild grass are practically insufficient to stay the advance. River-side dunes. The inland dunes are a feature which is rarely seen in other parts Inland dunes of red Of the country. Those to be seen in this region spin are found in the upper part of the Pal-eru valley, ‚east and south-east of Nandananam (Nundanawonum). That they are not river-side dunes is clear from their positions, and from the fact that the sand in the river beds is drab or greyish, while the sand hills are bright red, almost scarlet when seen at a little distance in strong sunlight. The best-defined and most striking is that of Narrava Gopalpur, at the northeru extremity of the Kodni Konda ridge. This sandhill is 4$ miles long, and generally about a quarter of a Narrava Gopalpursand- Mile wide. The sand contains a proportion of Zus clay sufficient in parts to make the whole bind into a tolerably compact mass,—not so compact, though, but that it crumbles down under one's feet. Another well-defined red dune occurs at Kondareddipalle (Condareddypully), about 5 miles to the north-east. Several other accumulations, of equally red colour, occur to the west of Hanumantapad and near Chinna Gollapalle, and others also to the west of the Kodni Konda, near Nandananam, Ramapalle, and Pondhova. To (ТОЛА) 102 FOOTE : GEOLOGICAL STRUCTURE OF THE EASTERN COAST. the east of Hanumantapad are two or three vividly red hillocks piled up against the side of pale quartzite hills, and making a striking contrast of colours. As seen from the top of the quartzite ridge west of Kondareddipalle, the whole of the great flat valley of Nandananam in the corner between the Kodni Konda and the bend of the Vellakonda range, is seen to be covered by soil of the most intense red colour—the reddest tract of country I can recall to mind. The derivation of the red ingredient is a problem to be solved; the great richness in iron of the local soil does not agree with the character Nandananam valley. of the surrounding older rocks so far as they are seen. Both the mica schists of the gneissie series and the quartzites and slate of the Kadapa rocks are locally very poor in iron. This suggests that the ferruginous material came from elsewhere, and I believe it will be found that there was once a large development of lateritic beds in this old bay, remnants of which remain in the lateritic gravels exposed here and there under the sandy red soil of Nandananam at Ramapalle and Pondhovah, the red sands themselves being detritus of parts of the formerly more extensive beds. This redness of the soil is confined to the basin of the Pal-eru, and is not seen on, or north of, the water-shed between it and the Musi-eru. A solitary pateh of similar bright red sand occurs piled against i o TU the side of the Iskapily hill 30 miles to the south. These red dunes are due to the action of the local winds (generally strongest from the south-west), tearing up the surface of the red soil during the dry weather. Оп a very small scale they resemble the “terais” of Tinnevelly district, but are brighter in colour, and contain a larger admixture of clayey particles. ECONOMIC GEOLOGY. 108 CHAPTER IX.—ECONOMIC GEOLOGY. The circumstances of the case unfortunately compel the admission that the mineral resources of the area described in the foregoing pages are very small as regards all the really valuable and important minerals. It is only in the matter of building and road materials that the country is well off; and even in this matter it is no better off than the average districts further to the south. Of metallie ores, there is certainly an abundance of good iron; but as unfortunately no eoal or other mineral fuel accompanies it, it must for the present remain unused and uncared for. Whether this state of things will remain so or Influence of the new 190 remains to be seen. The opening of the new E a possible iron east coast, or Buckingham Canal, may possibly render the magnetic iron of the Ongole beds valu- able enough to be carried to Madras, or to some other point where sea-borne coal might be used to smelt it. Canal carriage being extremely cheap as compared with all other methods, it may be worth the while of those interested in the matter to look into it. The Napier Iron Foundry in Madras is said to use indigenous iron ore from Nellore district, but there is no magnetic iron in the southern part of the dis- trict, so they are probably using a lateritie ore greatly inferior in wealth and quality of iron. Should the Ongole iron beds be made available to supply ore, it might lead, later on, to the working of some of the other beds that are close to the Gundlakamma or the Man-eru rivers, as they could be placed in direct communication with the canal for a few days in the rainy seasons. No iron smelting industry seems to be followed by the natives at present in the parts described in this Memoir, though it formerly existed, and is mentioned by Dr. Heyne. The only other metallie ore met with was copper, but merely in very minute traces, as malachite, or green carbo- Copper ores. А : : оло nate, oecurring in tiny cavities, and those very scarce indeed. This was found at the northern end of the Gogulapalle quartzite ridge (see page 13). Тһе indications were not at all promising (UB N 104 FOOTE : GEOLOGICAL STRUCTURE OF THE EASTERN COAST. of the discovery of the further occurrence of the ore in any really valuable quantity; and the past experience of copper-mining in Nellore district is so extremely unfavourable, that it is hard to anticipate that any valu- able lode or pocket is likely to be found. Several natives were said to be prospecting in that neighbourhood at the time of my visit (in 1875). I fear they would be most grievously disappointed. Garnet sand is collected by the natives and sold in the bazaars as a substitute for emery, but they seem unacquainted Garnet sand and stau- з | ; ( rolite substitutes for with the fact that staurolite, of which an immense ET quantity is to be had in and near the Chundr hills for the mere trouble of picking it up, is a material of superior hard- ness to common garnet. Even in the European markets the value of staurolite in that respeet appears to be unknown, probably because hitherto untried. The supply of lime is abundant, most of it being obtained from EN burnt kankar, except on the coast, where recent and sub-fossil shells are collected : the natives do not appear to have had recourse as yet to any of the crystalline limestones. The selenite occurring in the marine beds along the new canal might Se be worth collecting, as there is an increasing о um. E . JUN. B demand for plaster-of-paris in Madras, and the specimens of it sent me by Lieutenant-Colonel Hasted, R.E., are cer- tainly far larger and purer than those dug from corresponding marine clays in the neighbourhood of Madras. The architectural inclinations of the Telugu people being very small as compared with those of the Tamil people, there - Building stones. ona с ee are very few fine buildings of any kind by which to judge of the beauty or adaptability of the various rock varieties for building or decorative purposes. Most of the stone buildings to be seen are coarse and rude in their construction : this remark applies emphatically to nearly every modern structure. Good carving Nate cutus at Am. ANd elegance of design is to be seen in the few гатай. remains of Jain or Buddhist architecture to be ( 104 ) ECONOMIC GEOLOGY. 105 found here and there. Foremost among these are a few carved slabs derived from the old Buddhist * Tope? at Amravati. The materials these slabs are carved out of is Palnád limestone, a sub-erystalline or erystal- line rock of the age of the Karnul limestones, and therefore strictly speaking not belonging to this area. Carved fragments of the red Tangelamudi sandstone, derived from some Jain temple, are to be seen in Kondavidu Kondavidu. Bu Drug, built into the walls of the north-eastern gate, and contrasting strongly with the remainder of the gneiss-built walls. The sandstone, some pieces of which bear inscriptions, has worn well, and shows but little weather action. Two gems of carved and polished stonework are to be seen at Amin- bodu, north of Kondavidu, and. 10 miles west Aminbodu. of Guntür. Both are small temples of Jain origin ; the northern of the two has once upon a time been converted into a Mussulman building by the construction of mauresque arches of rough stone between the round and highly polished stone pillars. "The other is a low Hindu temple standing imposingly on a bold mass of granite gneiss. The southern doorway is surmounted by a very elegantly carved portico, while to the east another very elegant portico shelters a recessed shrine beautifully cut out of dark green stone, a true diorite apparently, to which a very high polish has been given. The pillars in the smaller temple appear to be of the same stone, and are equally meri- torious specimens of stone-cutting and polishing. The only special branch of industry connected with stone that is now Stone cart-wheel in. Pursued in the Nellore-Kistna country, is the dustry. manufacture of stone wheels for agricultural carts of different sorts and shapes. This industry is followed at several villages near the boundary of the schistose area, east of the Podile hill. The stone used for this, to European ideas so singular form of industry, is all quarried in the. Kuchupudi (Koochoopoody) hill, or Andrakonda, a bold preeipitous mass of granite gneiss, 800 to 900 feet high, lying 3 H | ( 105 ) 106 FOOTE: GEOLOGICAL STRUCTURE OF THE EASTERN COAST. miles east of the south end of the Podile mountain. The rock is a grey or pinkish-grey homogeneously ` bedded form of granite gneiss, the felspathic ingredient predominating a little over the quartz, and both very much over the mica. The masses to be converted into wheels are broken out with wedges in rude squares, which are first dressed into octagons or hexagons, these are then raised and set on end, and kept up while worked till complete. The wheels are generally perfect discs, in the plane of bedding, all but the box, whieh is kept nearly three times as thick as the peripheral parts. In a few cases I have elsewhere seen lensiform wheels. The hole for the axle is drilled from both sides, till a thin diaphragm only remains, which is then carefully broken out. Occasionally the sides of the wheels are ornamented with elegant scroll patterns. i The cost of the wheels increases very largely in proportion to their size, which is measured in spans and finger-breadths from the centre to the circumference. A pair of wheels of three spans and four finger- breadths semi-diameter will cost 8 rupees, a pair of four spans semi- diameter 10 rupees, and a pair of five spans semi-diameter not less than 20, owing to the increased difficulty of getting the large-sized blocks. A large pair of wheels is two months’ work for one stone-cutter. These wheels are said to be very durable, unless exposed to sudden collision with rocks, and their durability is said to improve with time: a pair of large wheels equal to a burden of one candy in the first year, will bear two candies in the second. Тһе Kuchupudi quarries occupy 30 stone-cutters, and turn out about 100 pair of wheelsin the year, and neighbouring villages supply a smaller number, in addition to gneiss troughs, curry-stones, 62. : Тһе stone-wheel industry exists also, according to the Kuchupudi people, at Dekerekonda near Darisi, where a similar but rather paler granite gneiss is worked. I had no opportunity of examining this latter quarry. But little use is or has been made, except as rough building-stone, of the many very handsome varieties of granite gneiss that might be quarried at very many places in the granitoid areas, Kondavidu especially. The ( 106 ) ECONOMIC GEOLOGY. 107 largest rough stone buildings of olden times are the old hill forts of Kondavidu and Bellamkonda ; of modern buildings the great anicut,'or weir, over the Kistna is pre-eminent. The Rájmahál sandstones of Tangellamudi, Chebrolu, and Pávulur supply the Publie Works Department with considerable quantities of building material for the various works—bridges and sluices—connected with the irrigation channels in the western delta. "The two former localities were also largely resorted to at earlier times for their red and purple sandstones, which, from their easy workability and rich colour, were in great demand for temples and for gravestones by various castes of Hindus. | The new field demarcation accompanying ће revenue settlement has also created a local demand for rough boundary stones in many places where they were not previously quarried, and in many of the alluvial tracts such stones have been carted a long way. Some fine examples of old carved and polished stones, which must have been carried long distances, are the carved bulls and lingams remain- ing among the ruins of several temples at Kanupati, on the coast, a little north of the mouth of thé Gundlakamma. These ruins do not stand, as stated in the Nellore District Manual (page 431), upon a reef of hard rocks, but upon the pure alluvium, in which not a scrap of stone is to be traced except the ruins of the temples themselves. Large quantities of salt are made at numerous salt-pan stations along the coast, but they offer no peculiarity requiring to be mentioned here, i ( 2070) ^i heal DENN N) f ? ME GEOLOGICAL SKETCH of the EAST COAST SOUTH or тне KISTNA vo LAT: 15. d A la BAT TENATALLI| 7 E тен Scare S Mines — | Inch MEMOIRS OF THE GEOLOGICAL SURVEY OF INDIA. [3 $t x. 5” It wor 1 ж- MEMOIRS OF THE io, S GEOLOGICAL SURVEY OF INDIA. VOL. XVI, Pr. 2. PUBLISHED BY ORDER OF HIS EXCELLENCY THE GOVERNOR GENERAL OF INDIA IN COUNCIL, CALCUTTA: PRINTED FOR THE GOVERNMENT OF INDIA. SOLD AT THE GEOLOGICAL SURVEY OFFICE, OFFICE OF SUPERINTENDENT OF GOVERNMENT PRINTING, AND BY ALL BOOKSELLERS, LONDON: TRUBNER & CO, * оыгссосһххх. CALCUTTA : OFFICE OF THE SUPERINTENDENT OF GOVERNMENT PRINTING. 1880. CONTENTS. PART I. CHAPTER I.—GENERAL DESCRIPTION. Introductory, 1.—Area of country defined, 2.—Towns and communications, 2.—Climate, 3.—Eertility, 8.—The people, 4.—Mineral resources, 4.— Previous literature, 6.—Work of the survey, 7 . : 4 . 1—7 Page, CHAPTER II.—PHYSICAL GEOLOGY. Table of rock formations, 7.—Marked features of low country and ghát step, | 7.—How formed, 8.—Approximate age of, 9.—Minor orographie fea. tures, 9.—The low plateaus of the coast, 10.—Age of, 10.—Shore line of sandhills, 11.—Intermediate alluvial flat, 11.— Drainage system, 11.— The passage of the Penner at the ghäts, 12.—A pproximate age of river valley, 13.— The delta of the Penner, 13.—The Pulicat Lake and its origin, 14.—The belt of coastal deposits may have been much wider, 14. — Correspondence of physical features with arrangement of rock series, 15.—Less regularly arranged series, 16 . 6 à о . 7—16 PART II. CHAPTER IIL—THE GNEISSIC SERIES. The Nellore gneisses are generally foliated, 17.—The massive and schistose gneisses, 17.—Succession of these, 18.— Тһе mountain gneiss of Southern India not represented, 18.—Classification of the Nellore gneisses, 18.— Correlation with other gneisses of India, 19.—With Scottish gneisses, 19.—The massive gneiss of the Swarnamukhi, 20.—Granitic and trappean intrusions, 21.— The grey gneiss of Kálahásti, 22.—A certain likeness to the massive gneiss of Southern India, 22.—More varied than the red gneiss, 23.—Porphyritoid gneiss of Kálahásti, 23.—Breaks in continuity of strike, 24.—The porphyritoid belt changes about Raptr, 24.—Inter- ruptions of transition rocks, and faulting, 24.—THE SCHISTOSE GNEISSES, 25.—Change in the dip, 25.—Lithology, 25.—An eminently schistose 11 CONTENTS. Page. band, 25.— Less schistose band, 26.—Garnetiferous band, eupriferous, 26.—Ridges of massive gneiss, 26.—Style of the gneiss north of the Penner, 27.— The subordinate quartz-schists, 27.— Difficulty of distin- guishing these from the Cuddapah quartzites, 28.—Lithology and distribution, 28.— The Narasimhakonda quartz-rock, 28.—Finer and compacter varieties about Ingoort, 29.—Micaceous, 30.—Saccharine, 80.— Epidotiferous or pistacitic quartz-rock, 80.— Waxy quartz-schists, rip- pled, 31.—Close similarity to the Cuddapah beds, 31.—Conglomeratoid eneiss, 31.—A gradation in the series evident, 32.—The pistacite quartz- schists, 33.—Ironstone quartz-schists, 34.—Great development of quartz reefs, 35 . à : : E 2 А ; . 17—35 CHAPTER IV.—THE TRANSITION SERIES, CUDDAPAH FORMATION. The Cuddapah groups in this area, 36.—Age of the series, 37.— Distribution, 37.—Relation to the gneiss, 38.—Obscure members of the series, 38.— In the neighbourhood of Kálahásti, 39.— The Pillaméru outcrop, 39.— The Kandra area, 40.—Remnants of a fold involved among and associated with traps, 42.—No cases of association with gneiss, 43.— Faulted to some extent, 43.—Southern end of Cuddapah field faulted, 44.—Gelacapad- Káluváya area, 44.—Relation to the gneiss, 45.—Appearance of uncon- formity with the gneiss, 46.—Faulted eastern boundary, 46.—Crushing and alteration, 46.— Trap flows, intrusive, 46.— The Káluváya ridge well separated from the Veligonda beds, 47.— Continuation of the outerop across the Pennér, unconformable on the gneiss, 47.—The Cuddapahs in the Veligonda range, 48.—Faulted against the gneiss, 48.— The rocks of the Veligondas, 49.—C. Æ. Oldham's notes on the rocks north of the Pennér, 50 B : Д ; 5 А 2 . 86—50 CHAPTER V.—GRANITIC AND TRAPPEAN ROCKS. The Ojili and Güdür granite, 56.—Trappean rocks in definite and ill- defined outbursts, 57.—Dykes in the gneiss, 57.—Relation between spheroidal weathering and jointing, 58.—Dykes in the transition rocks, 59.—Intrusive traps of Paremkonda, 60.—Irregular outburst of trap, 60.—The Kándra outburst, 60 . : , 4 5 . 56-60 CHAPTER VI.—OTHER FORMATIONS. (1, Патмланат, Prant BEps, 63.—Mere traces of Upper Gondwánas in this area, 63.—(2) CUDDALORE SANDSTONES, 67.—Diflleulty of distinguishing these from the lateritic deposits, 67.—Resemblance to, in weathered rocks, 68.— The Nellore plateau, 69.—More sandy to north of Nellore, 70.— CONTENTS. 111 Page, (3) Плтевітіс DEPOSITS, 71.--баһ-абгізі, 71.—A source of the ferru- ginous element in the ironstone beds of Rasanur, 72.—(4) RECENT DE- POSITS, 72.— Back-waters and coastal alluviums, 73.—Sub-fossils, 73.— River alluviums, 74.—Blown sands, 75 . ur. ; с . 63—75 CHAPTER VII.—NELLORE COPPER WORKINGS. On northern edge of field, 77.— Results not very encouraging, 77.—Occur- rence of the ore at Garimanipenta, 78.—Latest explorations of Mr. Lavelle, 79.— Ores obtained, 83.—Prinsep’s assay of the ores in 1836, 83,— History of the copper workings, 84 : А : > 77—84 MEMOIRS OF THE GEOLOGICAL SURVEY OF INDIA. THE GNEISS AND TRANSITION Rocks, AND OTHER FORMATIONS OF THE NELLORE PORTION OF THE Carnatic, Jy WILLIAM Kine, B.A., Deputy Superintendent (Madras), Geological Survey of India. PART T. "CHAPTER I.—GENERAL DESCRIPTION. The selection of a fitting title for this memoir has been beset with much difficulty, for the country, the geology of which is about to be described, is not defined by any political or natural boundaries, except on two sides, while it includes portions of three districts in the Madras Presidency ; neither is there any special formation or series typically displayed in the area, sufficiently so at least to be thoroughly worthy of separate treatment. Still, as the most prevalent Introductory. and conspicuous rocks are of the crystalline series, iud the relations between this and the transition series are more fre- quently displayed within this part of the Nellore District than in any other portion of the Carnatie, these formations are selected as the more particular theme of this work. The modus operandi of the Geological Survey is likewise largely aecountable for the presentation of such an undefined area, for at the time when 1t was examined, the grouping of the various rock systems of India was still in abeyance, and the areas for survey were chosen partly according to the sheets of the Indian Atlas. It isin this way that other portions of the Carnatic have been already (7109) Memoirs of the Geological Survey of India, Vol. XVI, Art. 2. 2 KING: NELLORE PORTION OF THE CARNATIC. reported on, such as the Madras and Kistna regions,! which tie on to the northern and southern edges of the present area. The examination of this ground was originally taken up by the late Mr. Charles Æ. Oldham and myself in 1861 while following out the transition rocks of the Cuddapah distriet; but the publication of the geology of more important areas, and the death of our colleague, who would most likely have written this memoir, have delayed the description of this country until now. Dr. Oldham deferred any memoir until there might be an opportunity of revisiting that portion of the district in which the complicated and obscure relations of the Cuddapah rocks and the gneiss are displayed, but as there has been no chance of this with the reduced number of the survey party in this presidency, and this condition being likely to continue, I have been requested to prepare this treatise from the observations and notes then recorded. The country so introduced to the reader is a compact more or less rectangular strip of the Carnatic, lying between about the 13° 20’ and 14° 59’ parallels of north lati- tude, with the Bay of Bengal on its eastern side and that portion of the Eastern Gháts called the Veligondas? as its western edge: it includes rather more than the southern half of the Nellore district and portions of the northern edges of those of Madras and North Arcot. There is only one large town of importance, namely, the zillah station Area of country defined. of Nellore, but two other smaller and purely native towns, residences of the Chiefs or great zemindars Venkatagiri and Kálahásti (Calastry), mark respectively the western and south-western parts of the field. It is traversed by two important rivers, the Pennér and Swarnamukhi, the Towns and communica. Proper basins of which lie, however, to the west- tions. ward of the Eastern Gháts. Communication is effected by the great northern trunk road, and this is connected at Nellore and Güdür with the western distriets of Cuddapah and North Arcot by two less important roads crossing the Eastern Gháts at the 1 Mem. Geol. Surv. of India, Vols. X and XVI, pt. 1. 2 Also Vellacondas, Yellacondas, and Vellicondas. ( 330 ) GENERAL DESCRIPTION. 3 Doranal and Rápár passes, and by a third along the valley of the Swarnamukhi. A great canal connecting Madras with Cocanada, a distance of nearly 400 miles, traverses the country by a system of salt and fresh water lagoons extending from the Pulicat lake nearly parallel to the coast. The sea-board, though quite unsheltered, and dangerous at ‚its southern end, where it is marked by a small light-house opposite the great Armogamor shoal, offers anchorage to occasional vessels at Ramia- patam and Kristnapatnam at the proper seasons. | The climate is a hot and dry one for most parts of the year, the effects of the south-west monsoon being only felt shghtly on this side of the Peninsula, while the north-east rains do not generally last more than a month in any force, though the wind blows strongly for a longer time. During the former monsoon the two main rivers are down in flood, especially the Pennér, whose waters are, however, retarded by a dam or anicut near Nellore ; and thus indirectly, and in the one case artificially, a part of the country is enabled to derive great benefit from the rains falling far to the west- ward. At this season also, it often chances that the sand banks barring many of the sea-outlets of the smaller rivers are eroded, when a refresher is given to wastes of water which are too often liable to become Climate. stagnant. With such a elimate and because the soils are generally poor, the country is not natural a very fertile one, the wilder parts in the neighbourhood of the western hill-ranges being covered merely with low tree and scrub jungle, while the eleared ground, except near the numerous artificial tanks or alongside the rivers, is only favourable to dry grain crops, and those not very rich ones. The soil is often largely impregnated with soda, and the waters of most of the streams, the Pennér excepted, are also much charged with this mineral, for whieh reason much of the alluvial deposits bordering the coasts are very poor. The soils of the Penner valley and delta are, on the other hand, formed of deposits brought down from distant fields of rock and cotton-soil, giving much more fertile sediment. Тһе fringe of ( Ell ) Fertility. + KING: NELLORE PORTION OF THE CARNATIC. cocoanut and palmyra palms, so characteristic of the Coromandel, is here much diminished in breadth and luxuriance, becoming very narrow and poor at the northern end of the area,—a feature which is perhaps attributable to some extent to the diminishing width of the sandy deposits of the coast. At all events, the palmyra is in much greater force towards the Pulicat lake, where also the belt of sand-hills is widest. The people are principally Teloogoos, in fact the southern edge of the area А may be roughly considered as ће boundary e the Teloogoo and Tamil-speaking races. There is · also a remnant of an aboriginal race, sometimes called Yanédis, dwelling in and about the island of Sriharikota, where there is a good deal of waste jungle-land. Indeed, itis a strange feature of this part of the Carnatie that one may here meet, within such close proximity to Madras, a tribe of people which still retains some of the manners and customs of a very primitive section of the human race, to such an extent that an iron arrow-head is treasured by them as a rare thing, the points of their wooden arrows are hardened by charring, and they are in the habit of obtaining fire by friction of wood, although fire is proeurable at villages close by. Mineral resources are few and far between, but some of these, such as the copper ores, have not as yet undergone all the investigation that their indieations entitle them to. Building-stone from the gneiss is often conveniently situated and of first-rate quality, but that most easily and cheaply worked is the laterite, which eame in well for the trunk road, and again in the irriga- Mineral resources. tion works on the Pennér near Nellore. Some crystalline limestone occurs associated with micaeeous and quartzose schists, but it is not suitable for the manufacture of lime, the common kankar or ealeareous travertine being the general source for mortar. There is, however, a special store of this material in the seams of sub-fossil shells to be found a short distance inland from the coast back-waters and at a small depth from the surface. ( 142 ) GENERAL DESCRIPTION. 5 The frequent occurrence of the products of felspathic decomposition offers sources for the extraction of salt, soda, and saltpetre, and for the manufacture of bangle glass; but these operations are really only spasmodie, natives, as a rule, seldom manufacturing more than is sufficient for local use. The extraction of salt from the lagoon waters is, on the other hand, an extensive and profitable industry, and this is nearly altogether in the ‚hands of Government. There are good indications of fairly rich copper ores on the northern edge of the field, but in spite of many costly attempts which have been made within the last fifty years to work them, the results have hitherto been ruinous. Iron has never been extensively smelted, only a few furnaces being worked at allregularly, though there is a strong run of iron beds in the southern part of the field. Diamonds are said to have been found, but reliable information could never be obtained on this point, neither were any traces of old workings ever met with. There is, however, a primd facie expectation that those gems should be found near the mouth of the Pennér gorge, this being the only funnel through which any debris of the diamond-bearing deposits could be carried from the Cuddapah or Kurnool district, where diamonds are worked, or were formerly found at no great distance up the river. Тһе conclusion is, that if few stones are or were obtained so much higher up the rivers, then fewer, or none at all, could be expected to occur below the gorge. However, there are old diamond workings in the Kistna district, in the north-east prolonga- tion of the Eastern Ghäts, which appear to have been in true Cuddapah strata, and not merely in beds of the newer Kurnool rocks, as is more generally the case; and as the Veligondas are of Cuddapah beds, it may after all be that diamonds have been found to the seaward of these mountains. Other minerals, sometimes of use to the jeweller or lapidary, such as garnets, rock crystal, amethyst, and the pretty blue kyanite, are not unfrequent, but these are never good enough to be of any particular value. Mica is sometimes obtainable in moderately-sized plates large enough to be used for small wall-lights and other decorative purposes. 1 See Mem. Geol. Surv, of India, Vol, VIII. Cue j) 6 KING: NELLORE PORTION OF THE CARNATIC. The previous literature touching on any geological points of interest in this part of the Carnatie is very slight, and indeed this is only what might be expected, there being little to attract the attention of the mere amateur or official Previous literature, traveller whose bent may be scientific. That once indefatigable explorer, Captain Newbold, has, however, left the traces of his footsteps here as elsewhere over Southern India in notes of a traverse across the Veligondas, and of an examination of the site of the old copper mines of the Nellore district ; but these are very short, and do not enter into much detail. Dr. Benjamin Heyne (1800) also refers to the copper mines in his Tracts on India, but not at any length, at least as regards the old mines of this area. In 1836, Mr. James Prinsep, the then Assay Master of the Calcutta Mint, examined specimens of the Nellore copper ores, and a report of his results is given in the Madras Journal of Literature and Science. In that year also, Colonel Mentieth, of the Madras Engineers, published an account of his trip to the Kambák Droog,’ but there is nothing strictly geological in this paper. A further visit (1839) was | made by Captain J. A. Smith to the same range, but more for the purpose of ascertaining its height, there being at that time an enquiry as to the availability of this little hill-range as a sanitarium for Madras.? In 1857 the question of the existence of coal, which used to trouble Madras for many years, was revived by Mr. G. Powell, Assistant Engi- neer, who reported on the appearance of coal at Nellore.* Тһе mineral so brought to notice was schort; and since then up to within the last few years, schorl has done duty for coal several times, not only here, but also in the Kistna and Godavari districts. Mr. J. А. C. Boswell’s ex- haustive Manual of the Nellore district appeared in 1873, and it com- pletes the history of the hterature so far. "The geological chapter in it was supplied in part by Charles ZE. Oldham, and in the mineralogical chapter there is a full history of the copper workings, as well as much information on the little iron ore obtained in the distriet. ! Vol. III, Old Series, p. 154. ? Madras Journ. Lit. and Sci., Vol. IV, O. S., p. 134. 3 Madras Journ. Lit. and Sci., Vol. IX, O. S., p. 311. 4 Madras Journ. Lit. and Sci., Vol, XVIII, O. S., p. 291. COELO) PHYSICAL GEOLOGY. 7 The geological survey, as already stated, was conducted for the most part by the late Charles Æ. Oldham, then Deputy Superintendent, and by the writer, but certain parts of the country were revisited in later years by Mr. R. Bruce Foote mainly in connection with his examination of the extension of the Work of survey. Rajmahal plant shales. CHAPTER II.—PHYSICAL GEOLOGY. The form of the country, within the limits laid down, is very simple, Table of rock forma. being а wide uneven plain, having the sea along Mons: one of its edges and a mountain wall on the other ; but it is so intimately connected in its origin with the physical structure of the adjacent country, that important questions regarding the southern portion of peninsular India may to a certain extent be touched on in describing it, and for the readier comprehension of this treatment of the physical history, the following list of the geological formations or groups is now given, even though there must necessarily be some repetition in the more detailed discussion of them in their proper ehapter :— LOCAL GROUPS. INDIAN EQUIVALENTS. EUROPEAN EQUIVALENTS. g. Blown sands. tee TM f. Alluvial deposits (River | Recent and sub-recent. and Backwater), 2... e. Lateritic conglomerates and Lateritie deposits (К. B. POST-TERTIARY. breccias, Gravels, Talus Foote). debris. d. Laterite and Ferruginous Cuddalore sandstones. .. — ?* Middle Eocene. sandstones. : е с. Plant shales. оо Rajmahal beds. s JURASSIC. b. Quartzites and Clay-slates. CUDDAPAH SERIES. „e ? LOWER PALXOZOIC. а. Gneisses of various kinds. GNEISS. Granites and traps assos ciated with (а) and (b). The southern portion of the Peninsula is distinguished by a great Marked featuresof low and well-known physical feature, namely, the country and ghat step, arrangement of the land surface as a coastal plain or the low courtry, with a more or less distinct step or ghát edging an ERU 5 8 KING: NELLORE PORTION OF THE CARNATIC. upland ; and these are well marked in the present area by the wide beit of plains and the western barrier of the Veligonda mountains, which are at the same time geologically distinct, the former being of gneiss, covered up in a scattered way (more perfectly towards the coast; by later forma- tions, while the western hill wall is of the hardest rocks of the Transition or Cuddapah series. The plains are, however, much diversified by low hills and ridges in their middle portion, and by outliers of. the mountain wall and other lower ranges lying to the south-east, they themselves hav- ing at the same time an easy slope up to the base, or rather up to a great talus of debris collected at the base of the western hills. The low country proper seldom rises more than 150 feet above the sea, but the talus, which is often more than 4 miles in width, sometimes attains a height of 250 feet. The Veligondas have an average height of about 2,000 feet, the peaks of Penchalakonda and Nagwaram being at least over 2,500 feet. The immediate outlier of Udayagiri is said to be 3,030 feet high, and the larger one of Kambák Droog’, on the southern edge of the field, is about 1,800 feet. The Veligondas present a generally steep face to the sea, but this is very much scored by deep valleys and ravines, and is even cut right through by the Pennér river at the Somisilla gorge. There are some very fine cliffy headlands, especially to the north of this gorge, and the southern outlying Kambak Droog and the smaller Kálahásti range are even more precipitous. To the north of the Pennér the Durgama Konda outlier differs from the rest in having a long sloping back to the seaward, while its much steeper face is to the westward; and Udayegiri Droog is a conspicuous plateau with lofty and nearly perpendicular edges. The carving out of the great plain aud step of mountain wall was Бе акч evidently in the greatest measure the work of marine denudation, though subsequent atmospheric and river degradation, following on possibly more than one elevation above the sea, and later levellings up by the deposition of newer forma- tions, have almost entirely obliterated all trace of this. The marked distinetion between plain and wall is at the same time in part due to the (116 ) PHYSICAL GEOLOGY. 9 very different constitution and durability of the rocks composing each, while erosion must have been considerably facilitated by a system of faults oceurring along the boundary ; and had it not been for the very unassail- able constitution of the Velingonda rocks, it is quite possible that this part of the Carnatie might have been nearly as broad as it is in the Madras and North Areot distriets, over which the transition series does not appear to have ever extended much further than the Naggery and Narnaveram hills. As it is, the belt of hard strata withstood the slow wearing of the sea, and it now remains as perhaps the most clearly marked and abrupt step between the lowland and upland on the eastern side of the Peninsula. | The process of carving out having gone on in both the transition rocks (E AME i: E. of the step and the gneiss of the plains, it must - have taken place after the period of the former and before the deposition of any later series on the plain below the gháts. The only evidence bearing on the latter point in this area is, that the Jurassie plant shales far out on the plains and near the coast show beds containing what I take to be pebbles of transition or Cuddapah quart- zites. Тһе upper Gondwäna beds do, however, lie on the gneiss and close up under the Kambäk Droog outlier of the gháts in the Madras district : hence the step and plain must have been in existence at the latest in early mesozole times. ; As regards minor orographie features, the northern half of the Minor orographie fea. Country is marked by many ridges and some roche- tures - moutonnée-like masses of small elevation, such as the Narasimhakonda ridge to the west of Nellore and the Buchireddi- . palem hill north of the Pennér, the southern half having only low and broad hill masses. "The ridges which are the most characteristie of the smaller elevations are attributable to the greater frequeney of bands of crystalline quartzites among the gneiss series of the middle ground, the intervening softer schistose strata having been worn down to the lower general level of the plains; and the more rounded and less conspicuous detached hills owe their form to some large masses of trap associated with hornblendie rocks. 10 KING: NELLORE PORTION OF THE CARNATIC. A much less conspicuous elevation as compared with those described, The Tow plateaus of but, still remarkable and worthy of attention as E OE being continued with tolerable distinctness along nearly the whole of the Coromandel, is a narrow and very low plateau ridge of sandstones, with a lateritie covering running generally north and south at from 10 to 20 miles inland from the sea shore, and which core responds with or is an extension of the Red Hills of Madras, Pondicherry, Cuddalore, and Samulcottah, in the Godävari distriet. Here, however, itis perhaps better defined than in any other equal length of the coast line, while it preserves a tolerably uniform level surface seldom varying much between 40 and 70 feet above the sea on its higher or western side. The edges are well defined, particularly on the western side, by a slight steep slope, or else a low scarp, and sometimes even by fair head- lands giving a clear look-out over the intervening low country towards the mountain wall, whereas the eastern side is distinguished by the flat alluvial deposits shoring up along the very gentle seaward slope. This plateau ridge marks what may be considered the last permanent upheaval of the Coromandel; but the views of the Survey differ as to the period when this may have taken place. Mr. Foote is inclined to consider that it took place Age of. during the human period, since he has found stone implements of human manufacture which he considers were embedded in the laterite capping the southern portion of this ridge in the Madras district, having been dropped by their owners in the waters then covering the country. I too found the same kinds of implements embedded in a lateritoid deposit, but it certainly appeared to me then, and still does so, safer to look on these as having been cemented with other debris by ferruginous waters under ordinary atmospheric exposure, and that these weapons were dropped on dry land or in fresh waters, that land being possibly more covered with jungle and water much after thestyle of the southern coast of Ceylon, where an analogous rock to laterite, locally called “cabook,” is still being formed. The sandstones of this ridge being of presumably early tertiary age, the last elevation of the Carnatic must be of later times, but I think it is yet doubtful whether we can fix it so late as the human period. CTE) PHYSICAL GEOLOGY. 11 The remaining lower and newest ridge of all is that of the sand-dunes р í on the seashore, which is entirely subaérial and con- Shore line of sandhills. 4 5 j sequently of ever-varying height according to the force of the winds. This belt is scarcely noticeable to the north of the Pennér delta, while it only obtains any marked width between the outlet of the Gridtir river (Kandléru) at Kristnapatnam and the Pulicat Lake, on the northern shore of which some of the dunes are 30 to 40 feet in height. Between this and the chain of sandstone plateaus lies the compara- Intermediate alluvial tively level belt of fluviatile estuarine alluvium, flat. а good part of which is really at a lower level than the sea. Flooding is prevented by the banks of shore sands and dunes, and by the raised banks of the rivers, though there are still extensive flooded portions in the numerous large back-waters and lagoons,! of which the Pulicat Lake is the most striking example. Most of the lagoons, and especially the Pulicat Lake, are being gradu- ally silted up, while others to the north of the Pennér delta are evidently very much smaller now than they used to be, the deposits from the smaller rivers being of steady accumulation and the blown-sands lending their feeble assistance. Fitful or permanent communication with the sea | necessarily delays the silting up for a time, but the levelling up of hol- қ lows lower than the sea is just as certain а ргосевв as the denudation of land that once rises above it. Beds and banks of sub-fossilized marine and estuarine shells occur at shallow depths in this alluvial flat, but not at any decided elevation over sea-level, and these point to a further slight elevation of the country, though 1t is quite possible that they may have been brought up with the elevation of the Nellore plateaus. The drainage system is very simple, there being only two large rivers, neither of which, however, rises in this field, and as Drainage system. m Д c о this 15 not very wide, only some 57 miles, the rivers ! There are also a few fresh-water lakes, which, however, may possibly have been re- duced in number by the recent extension of the coast canal; but at the time of my survey in 1869 the long stretches of water at KÁpür and Pogaradinna were quite fresh, and the villagers stated that such had been the case for many years. (UN s 19 KING: NELLORE PORTION OF THE CARNATIC. J actually belonging to the area are not of much importance. The Pennér receives only the waters of a few small streams, prior to which it comes suddenly on the field with a nearly completed course through the clean- cut pass! in the Veligondas, thence flowing nearly due east to the sea in the shortest line it could take past Nellore. The Swarnamukhi, though much smaller, is still a large river as com- pared with others in the district, and it brings down a large volume of water, many of its best feeders being in the neighbourhood of the Chit- tür hills. It is also from this side of the country that some of the best affluents of the Pennér fetch its waters. The Swarnamukhi follows a much more irregular course to the sea, and does not show that decided general easterly trend as it approaches the coast, which is a marked feature in the Pennér and the other large rivers of Southern India. It also enters the country by a wide open valley, and only meets its later alluvial banks at a few miles from the sea. The other stream worthy of note is the Kandléru, the larger affluent of which rises in the Veligondas behind the Venkatagiri and Räpür country and meets 16 in the alluvial flat of coastal deposits near Güdür, whence it flows to, and enters the sea by, the large back-water at Kristapatnam. To the north of the Penner, there are only two small rivers flowing direct to the sea, which at flood times break the sand barrier to the eastward of Allur at Thulipallem, and again near Juviladinne. The noticeable feature in the course of the Penner is that it passes The passage of the Fight through or across the western mountain wall? Penner at the ghats. having worn its way downwards into the transi- tion rocks covering the greater part of the Cuddapah district to the 1 So suddenly and with such a gathering do the waters of this large river rise at the gorge, that the flood waters have occasionally been reported as reaching the anicut at Nellore before the runners can signal that preparation should be made at Nellore to meet the flood. ?]t is as well to notice the mistaken, though oft-repeated, view that this and other great rivers, such as the Godávari, further north, have absolutely cut through what are now, in point of fact, walls of mountains, some 1,000 or 3,000 feet in height, whereas the wall or basin is merely one of the results of the river denudation of the area of varied and softer strata behind the less destructible band of mountain rocks. ( 120 ) PHYSICAL GEOLOGY. 13 depth of at least 3,000 feet, thus leaving a deep, though wide, upward sloping V-shaped pass or gorge of about 2 miles in length through the Eastern Gháts. The late Dr. Oldham long ago communicated the view that the great Approximate age of drainage basins of India were оп the large scale SURGE valley marked out and existed as drainage basins at the enormously distant period which marked the commencement of the depo- sition of what we now call the Gondwána formation, and he considered the Penner valley as one of these larger basins, though no traces now exist, and perhaps never did, in its valley of any Gondwana strata. There are, it is true, the patches of Rajmahal (jurassic) shales occurring out on the plains towards the coast, to the north and south of the river, but these do not necessarily bear on the age of its basin. I have, however, already shown how the plains and ghát step of the Coromandel probably existed 1n early mesozoie times, and this, taken in connection with the occurrence of the plant-remains, seems to favour the view that the Pennér valley was then in existence also. This is, I think, all that can be safely said of the age of the Pennér valley in reference to Dr. Oldham’s generalization, though as regards the marking out of that valley in the Cuddapah dis- trict and in the eastern portion of the Bellary district covered by the same transition rocks, it is possible to conceive a much earlier time, fol- lowing almost immediately on the elevation of these supposedly very early paleozoic rocks. The Pennér has a narrow strip of alluvium along either bank as far ~ 7Ţ7 backas the Somasilla pass or gorge, but the first The delta of the Pennér. с T З h spreading out of its deltaic deposits occurs at San- gam. There is, however, really very little of a fair delta at the mouth of this river, for ıt does not sub-divide into two or more streams until within about 10 miles of the seashore, a feature which seems to me to indicate that the alluvial deposits must after all be greatly eroded, and that we have here only a portion of the original delta. The Swarna- mukhiand the Kaudleru have nearly as large deposits of alluvium spread out around their mouths, and they join with the Penner and the other : (ЗІМІ) 14 KING: NELLORE PORTION OF THE CARNATIC. small streams in having formed the great unbroken streteh of alluvium here lying along the whole coast. The courses of these rivers in their alluvial flats all trend to the north-east, or with the prevalent currents along the coast. The most remarkable feature of the coast connected with the alluvial The Pulicat Lake and deposits is the extensive, but very shallow, hollow its origin. of the Pulicat Lake, which has only one small stream flowing into it near Sultirpet, while the Narnaveram river at the extreme south end keeps it open to the sea. One way of accounting for the existence of lagoons such as the present, is that a hollow must have been left in the contiguous alluviums of two rivers which was eventually lapped round by their outspreading and rising deposits, though still kept open by minor streams flowing into it. Such is apparently the condition of things which went on in the formation of the Colair Lake between the Godávari and Kistna deltas. But in the case of the Pulicat Lake there is no such evident leaving of a hollow between the great rivers, though the suggestion that this is an analogous case to that of the Colair Lake is more valid if we suppose that a considerable area of alluvial land has possibly been removed from the eastern side, and that the Swarna- mukhi and Palár (in the Madras district) may have been the great allu- vial distributors to the north and south. The small extent and shortness of the delta of the Pennér when considered with regard to the length and drainage area of this large river and the generally even width of the alluvial belt along the coast, seem to point also to the former existence of alluvial land considerably to the eastward of the present coast line. It is to be noted that though six great rivers, viz., the Cauvery, Pennar, Palár, Pennér, Kistna, and Godávari, flow out on the Coromandel, yet only three of these have decided and extensive deltas; the next largest, the Pennér, having only a very smallone as proportioned to its importance, in fact, as I suppose, a reduced delta, while seaward shoals are prevalent along the adjacent coast, especially at Armogamor! and opposite Ramiapatam and Ongole further to the north. ! The “shoal is about 10 miles long, the shallowest part is 12 fathoms; and it lies from 33 to 53 miles east-by-north of the light-house,"— Manual of the Nellore District. (19% 4 PHYSICAL GEOLOGY. 15 'The partially parallel arrangement of the land features in broad or Correspondence of narrow strips of hill and plain is, as already ne with incidentally noticed, dependent on a more or less series, banded distribution of the later formations over the gneisses which give the most important and extensive outcrop, having a width of some 40 miles between the low plateau ridge of Nellore, &c., and the Veligondas. The oldest covering rocks or the Cuddapah series, consisting of detri- tal quartzites and slaty beds, are very clearly displayed and defined in the western mountain belt, though there are some important outliers in the Kambák and Kálahásti ranges to the south-south-east and again in the Udayagiri Droog and the Dargadevi Konda to the north of the Pennér. | | Тһе next later formation is merely represented by the barest traces of shales belonging to the Rajmahal group of the upper Gondwána system, which are here exposed in well-sections at only a few spots under- neath the lateritie band of Nellore. At the same time, it is quite possible that these small patches are part of a hidden belt connecting similar strata in the Madras and Kistna districts. Over these and covering them comes the broken strip of sandstones and laterite forming the narrow plateau ridges running parallel with the coast, on one of which the town of Nellore is situated. These sand- stones are rangeable with the Cuddalore samdstones, distinguished by Mr. H. F. Blanford as a group overlying the eretaceous rocks near Cuddalore, or with my Rajahmundry sandstones in the Godávari dis- triet, which overlie rocks whose fossil “relations appear to be rather with the upper cretaceous rocks of South India,”? and are therefore presumably of tertiary age. Of recent deposits, there is the belt of alluvium bordering the eastern edge of the sandstone plateaus just noticed; and of yet newer age comes the thin strip of blown sands on the sea-shore. ! Rec. Geol, Surv. of India, Vol. X, p. 56. ? Manual of the Geology of India, p. 319, ( 104 ) 16 KING: NELLORE PORTION OF THE CARNATIC. So far, the. distribution and succession of the different rock groups Less regularly arranged. ME clear; but there are other irregular distribu- series. tions of doubtful strata, the most important of which have really so long delayed the publishing of this memoir, as Dr. Oldham was never quite satisfied whether in these we had not either ` a newer group among the gneisses or whether they were really only more highly altered Culdapahs. These strata are still quartzites like those of the Cuddapah beds, but they have, as shown in the map, à somewhat irregular diagonal lie across the gneiss field, and are involved among schistose gneisses. Hence it was surmised that these schistose gneisses themselves might be after all only altered Cuddapahs. Of much less importance is a series of patches of ferruginous or lateritie conglomerates and breccias scattered at wide intervals all over the field, and forming a skin on the laterite and sandstone belt of Nellore. These belong to Mr. Foote's comprehensive group of * Lateritie deposits.” A well-defined belt of subaérial rocks lies immediately long the eastern base of the Veligondas, forming a series of great heaps of talus debris derived. from the destruction of the strata forming the mountain wall. Manifestly much of this band of angular and sub-angular debris must be as recent as any deposits in the field for its formation is still going on; but in many places it must be considered as coeval with the lateritie deposits or ferruginous conglomerate and breccia patches men- tioned in the last paragraph. So much as is seen of it must be essen- tially recent, ranging at any rate from the lateritie deposits up to the present time; but it would be an interesting problem to decide for the whole length, breadth, and thickness of this talus, reihen it may not be to a great extent a beach deposit which was even in process of formation during much of the mesozoie period. 1 Mem, Geol, Surv, of India, Vol, X, pt. 1; Vol. XII, pt. 1; Vol. XVI, pt. 1. THE GNEISSIC SERIES. 17 SP ARE LI: CHAPTER III.—THE GNEISSIC SERIES. Generally over Southern India, the gneiss is rather a crystalline The Nellore gneisses aggregate of minerals than a foliated or schistose gel YEE rock; and so manifest is this absence of foliation, that the rock is still more usually spoken of as granite. However, there is perhaps in this part of the Carnatie a greater display of the more schistose varieties of the rock, or even the more evident laminated results of sedimentation, than in many other parts of the Madras Presidency. While many classifications or groupings of the different gneisses are The massive and schis- Possible, looking at their differences of constitution, O ses. structure, and modes of occurrence, there is one very broad distinction apparent in Southern India in the massive and the schistose gneisses, the latter being necessarily foliated, while the former are only partially so ; a very important and extensive section being scarcely distinguishable as a foliated rock at all, it being only admissible intothe great family of gneisses on account of its banded or apparently bedded character displayed over very large areas, in contradistinction? to the intrusive or exotic condition attached to the recognition of granites. ! In the present immature condition of the survey of the crystallines in South India, a geographical classification of the gneisses is not without its advantages if only as a means of separating and recognising the many gneisses already included under such general groups as granitic, syenitoid, or hornblendie, and in this view I have long held that the massive granitoid red gneiss might be distinguished as the Bellary or Mysore, or better as the Bäla Ghát gneiss, since it is remarkably prevalent in the upland of Southern India, its eastern confines from the Palár to the Kistna being almost continuous with the edge of the gháts. It is typically developed, as far as my own knowledge goes, in the western parts of North Arcot, in the Cuddapah sub-division, in the eastern part of the Bellary district, in the Kurnool district, and thence all over the eastern portion of the Hyderabad "Territory up to the higher reaches of the Godavari river. In the same way, such a term as the Carnatie gneiss might be a convenient synonym for the schistose gneisses of South India, and a further term of Nilgiri or mountain gneiss for a very typical intermediate and less foliated rock composing the higher ranges of the Peninsula and Ceylon. There are, of course, grave objections to such a classification, but it is put forward as a convenient one, and in any case the fact still remains that there are such marked gneiss groups, and that they are geographically separable, B © 5 18 KING: NELLORE PORTION OF THE CARNATIC. This general sub-division of the crystallines is quite clear in the northern part of the Carnatic: in the south-west corner of the present field, that is, in the Venkatagiri and Kálahásti country, the rock.is a hard massive granular crystalline gneiss, of pale-flesh and grey colours, with little or no foliation, while on the other hand schistose gneisses are in great force all over the middle of the field, in eastern Kálahásti and the taluqs of Güdür, Nellore, Rápár, Kávali, and Udayagiri. These gneisses also appear to be associated in a fairly serial order, though, as is usually the case in South India, Succession of these. 5 =, no good boundaries are recognisable, the change from one to the other group or sub-division, often within a narrow width, being nearly always gradual. Such want of definition would necessarily oceur between the morehighly granitoid or irregular erystalline aggre- gates and the less distinctly foliated masses, the metamorphie action having been presumably more intensein these rocks: but as the more and more distinetly foliated gneisses are met with, ashere and further north in the Kistna and Godavari districts, it seems to me that boundaries are becoming more evident, and will ultimately be recognisable down south when more time can be devoted to the examination of these rocks. A distinguishing feature of the Nellore gneisses is that they do not number among them a good representative of The mountain gneiss of Southern India not the massive grey syenitoid or hornblendie rock represented. which is so persistent in the mountain ranges of South India, and which shows in diminishing force nearly as far north as Madras, but which, by its absence here, may perhaps be indicative of a much more important separation of the gneisses of Venkatagiri and Gúdúr than is evident in the field. The Nellore foliated rocks may then be considered under the great A classification of the Sub-divisions of the Massive and the Schistose Nellore gneisses. Gneisses :— 4. Micaceous, taleose, and hornblendie schists, with few quartz-schists or quartz-rocks : : o 3. Foliated gneisses, with frequent quartz-schists or SCHISTOSE GNEISSES. quartz-rocks . . : Grey gneiss (sometimes porphyritoid) , , Red granitoid gneiss . . - ( 126 ) = bo i | MASSIVE GNEISSES. THE GNEISSIC SERIES. 19 This serial order is, however, to a certain extent, rather arbitrarily made out in accordance with the received scale of metamorphism, the most erystalline rocks being assumed to be the older. "The real condition of affairs is that though the schistose groups are to all appearance over- lying the red and grey massive gneisses, yet the much more schistose group (4) is situated between the latter and the group (3) apparently overlying both. There seems, besides, no reason, in this field, to question the suecession deduced from varied stages of metamorphism, since this assoclation of the groups is explainable on a faulted position of the younger strata (4) against the massive and older gneisses, or a former condition of extreme unconformity by overlap between the latter and the schistose gneisses. Our observations of the erystallines are as yet too few and scattered Correlation with the to admit of any good attempt being made at а cor- guess D! шоа, relation of these Nellore gneisses with others in the Peninsula, though it would appear on the whole that they belong to the newer of the two series of the peninsular gneisses, namely, that of the main or eastern area.! At the same time there are many features about the massive gneiss of the Swarnamukhi valley or the granitoid gneiss (1) which point to a comparison with the late Dr, Stoliezka’s central gneiss of the Himalayas. Subsequent to my survey of this field, I had an opportunity. when M d euet at home on furlough, of visiting ine neighbour- hood of Lochmaree, in the western highlands of Scotland, which has been reudered famous by the controversy between Sir Roderick Murchison and Professor Nicol on the differentiation of the Laurentian and paleozoic gneisses occurring there ; and as far as observa- tions on therocks of such immensely separated countries can go, it ap- peared to me that the similarity between the eneiss (Laurentian) of Loch- maree and that of Venkatagiri and Kálahásti is remarkably close, while the schistose group of the Nellore district is comparable with the Silurian ! See Manual of the Geology of India, pp. xviii, xix, and p. 17 e£ seg., Part I. QU WENT y 20 KING: NELLORE PORTION OF THE CARNATIC. gneiss. This is, however, all that can be safely said at present, for unless the observer is aequainted by very long and careful surveying with the eneiss of both regions, his comparison of their rocks can only be very general. The опе great characteristic or feature of the older Scottish and Lewisian gneiss is its persistent transverse strike to that of the super- incumbent gneiss, the most absolute condition of unconformity ; but this cannot be cited as existing among the Madras erystalline series ; for, as far as we yet know, there is no marked deviation of strike in any of the gneisses, and the Cuddapah rocks, however unconformable they may be to their subjacent erystallines, have always very much the same strike. Hence in this one particular point the comparison of the Madras gneisses with those of Seotland loses much of its significance. In the following more detailed deseription of the rocks, that of the country north of the Pennér is taken from the progress reports left by Charles ZE. Oldham, while that of the country south of the river is mainly from my own work, though here also a good part was. surveyed by my colleague. The red massive or granitoid gnueiss.— The massive red gneiss, or The massive gneiss of granitoid form, occupies but a small portion of the nn south-west corner of the field along the valley of the Swarnamukhi and northwards past Venkatagiri, whence it is traceable westward right into the upland of North Arcot and Cuddapah. Its east- ern edge is ill defined, but it may be considered to run from below Batanaik Konda in a curving north-north-west line, a short distance to the west of Kälahästi and Venkatagiri, after which it passes under, or becomes the floor of, the Transition Series, only showing again very slightly to the north of the Penner. This gneiss is generally a close-grained aggregate of quartz and felspar (orthoclase, ? oligoclase and a little albite), hornblende being often scarcely discernible, while mica is even more rare. It is thus very like a binary granite, only it is perhaps more generally smooth and rounded on its weathered surfaces than the granite occasionally traversing it, the latter being very often more distinctly and largely crystallized (felspar easily C€ A THE GNEISSIC SERIES, 2] recognisable), or again very fine-grained and then sharp and smooth in its outlines and surfaces. The quartz is of white colour, in two forms, dull and amorphous or glassy, while the felspar is generally of pale flesh-colour, or occasionally of a dull white, the rock itself being essentially of a pale reddish colour. Itis very hard and compact, and requires most laborious chiselling when being dressed, but is easily hewn into large blocks by wedge splitting, a practice which is or was very common, for wherever there is a mound or hill of this gneiss, there are often old trenches whose sides show the holes originally cut for the reception of the large iron wedges. It is this variety of gneiss which oftenest gives the roche-moutonnée-like form to many of the hills, and in this feature! it answers somewhat to the “dome gneiss” of the northern area of the Bengal gneiss. Foliation is very. faint, but what can be made out in the Venkatagiri country is gener- ally north-south in its direction and vertical: here also signs of bed- ding are rare. Perhaps along the southern edge of the field, on the right bank of the Swarnamukhi, the rock would be considered more like gneiss, for here there 1s occasionally distinet and wavy foliation. Jointing is developed at wide intervals in the rock generally in north-north-east to south-south-west and east-by-north to west-by-south directions, the latter having the lower dip 60°—75° southward. Granitie and trappean intrusions are not common, or rather they кн GEE DOL often visible here owing to the area being u іе. in part a river valley and much covered up by superficial deposits. Dykes of compact and porphyritic greenstone are however often visible in the Swarnamukhi valley, particularly on the left bank at Kirkambádi and thence eastward and westward on the right side of the valley. Eastward of Püdi (on the railway just south of map) there are again many east-west dykes of greenstone, this locality being just on the eastern edge of a very remarkable and extensive development of trappean intrusions in that part of North Arcot. In the banks of the large nala about 3 or 4 miles eastward of Kirkam- ! Manual of the Geology of India, p. 20. (129) 99 KING: NELLORE PORTION OF THE CARNATIC. padi there are frequent outcrops of granitoid gneiss much eut up by numerous easterly and westerly small granular veins of white quartz with pale yellow-green pistaeite. Irregular strings of white quartz, . sometimes tolerably straight but oftener curved and branching, are com- mon in this gneiss. The grey massive gneiss.—To the eastward of the boundary indicat- Te ons of Ші striking a short distance westward of Kála- Kálahásti. hásti and Venkatagiri, hornblendie strings and nests begin to show in the red gneiss, and hornblendie foliated rocks crop out of the wide-spread superficial coverings of this part of the coun- try, after which the prevalent rock is a grey gneiss. This is in point of fact all that can be said of the change from one variety to the other of ihe massive gneisses, though there is no doubt that to the right or left of this vague line, or rather extremely narrow interval, the rocks are decidedly different in structure and constitution. The dip of the two is eastward, their lie being very often vertical: so that in serial. order, the grey gneiss is apparently above or younger than the red gneiss. The grey gneiss forms a band of from 10 to 20 miles in width, alongside the red granitoid rocks as far north as Rápür, after which its course is broken by the intervention of transition rocks, while there are sudden alterations in the rocks of the band itself. "The general north- north-west run of the strike at the same time carries this band. under the strata of the Veligondas, and the outerop north of the Penner is only very narrow. While working at this division I was fain to consider it a syenitoid A certain likeness to gneiss in contradistinetion to the granitoid rock the massive gneiss of 2 2 Southern India. from its being occasionally like the syenitoid rock of the Nilgiris and other mountain ranges of Southern India ; but it is here more generally a foliated rock, and thus perhaps rather to be ranged with the foliated gneiss of the low country of Coimbatore and Salem. ( 130 ) 1 THE GNEISSIC SERIES. 28 There is more variety in the rocks of this band than in the red More varied than the gmeisses, but the more prevalent form is a massive red gneiss. rather rough gneiss of quartz felspar and horn- blende, arranged more or less distinctly in a laminar manner, the laminæ or irregular folia being seldom continuous for any extent, unless they are thick, and then they are mainly of quartz or felspar. Foliation of this kind can generally be distinguished over a tolerably large surface of the rock, but with difficulty, and it is occasionally so obscure that the syenitoid character is then very striking. This form is never so essentially syeni- toid, that is, hornblendie, as the mountain gneisses of Southern India, nor is it so well laminated or schisted as is often the case in the low country between these mountain masses. In the neighbourhood of Kálahásti, or on the western edge of the belt, there are frequent outcrops of a more laminated or foliated rock, with fair hornblendie strings and nests, and towards Venkatagiri and to the north of that place this begins to show in greater force until about Räpür. In the southern part of this band the syenitoid variety is often more compaet, rather pasty and full of small rounded or bean-shaped masses of pale grey or white felspar, thus assuming rather a porphyritie character. In the neighbourhood of Kálahásti and northwards from the Swarnamukhi, it becomes more erystalline and coarse, the minerals occur- ring in larger masses. The long bay of gneiss running southwards between the Kálahásti Porphyritoid gneiss of ridges and the larger Kambák range shows abun- Cad dant outerops and masses of this compact por- phyritoid rock, with kernels of white felspar ranging through it in north- north-west to south-south-east laminz, in the long hog-backed hillocks and ridges in the neighbourhood of Royagunta or in the flatter-rounded exposures among the villages 4 or 5 miles westof Kälahästi. At that town itself, on the right bank of the river, a more granitoid form of this gneiss shows under the small hill to the north. The porphyritoid variety also forms a short width of the low country on the eastern side of the Kambák range. (ISL) 94 KING : NELLORR PORTION OF THE CARNATIC. North of the Swarnamukhi there is a very marked belt of the same EL n оао porphyritoid and, now, rather pasty gneiss running of strike. north-west of Káteheroo to the Kyoor tank for a breadth of nearly 4 miles, in which position, however, and with which dimensions it cannot be said to be quite continuous with the wider spread of outcrops in the Kálahásti and Kambák area; neither are there any fair indications partly owing to the wide covering deposits of the river—of any curving round in the strike of the foliation of the Kálahásti outerops as towards a contraction into the Kátcheroo width. Contrac- tion and even displacement must, however, I think, have taken place, for there certainly are evidences of such in the Kálahásti ridge and in the curious narrow strip of detrital quartzites of Pillaméru to the north which has evidently been wedged in here in a most abnormal way among the gneisses. Beyond, or to the north of the Räpür tributary of the Saidapuram The porphyritoid belt river, the Kätcheroo-Kyoor belt of grey gneiss | changes about Rápür. either dies out altogether or becomes so much altered or changed, or replaced by igneous-looking rocks, as not to be fairly recognisable. Тһе mass of the rock is a pasty-looking aggregate of hornblende, with a little quartz and felspar, and is more of a porphyri- tie syenite, but towards Bodanapali it becomes even more compact and pasty, and then looks more like a grey bedded trap, though 1t 1s still full of httle masses of grey felspar, and such rock is found extending into the bay of quartzite hills between Gelacapäd and Bodanapali. For a few miles further north there are only slight traces of a some- Interruptions of transi. What similar rock, the country being for the most OOS Banas part of abnormally placed quartzites of the Cud- dapahs among micaceous and hornblendie schists much traversed by a west-north-west to east-south-east system of trap dykes; but beyond this point, though there is still great similarity of constitution in further outcrops in the ridges extending from Sudamulla past Panumurti, I do iet feel justified in carrying this member of the grey gneisses. There is every evidence that a strong fault runs across this part of the country east-north-east from Räpür, by which the continuity of the Venkatagiri- (519250) | THE GNEISSIC SERIES. 95 Saidapuram strata is broken, for to the north of this hne the grey gneisses are found more to the westward running close up alongside the talus of quartzite debris below the Veligondas and so filling up most of the narrow valley extending north from Rápür to the Pennér, while the schistose belt of gneisses next to be described has also been thrown more to the westward. | THE SCHISTOSE GNEISSES. A tolerably close boundary can be drawn between this group and the massive gneisses from the southern edge of the area up to Rápár, but thence to the Penner the junction is covered up or ob- secured, the transition series likewise coming into juxta-position with it, while both are so crushed and metamorphosed that it is extremely diffi- cult, if not quite impossible, to say exactly to what series they belong, and it is clear that both have been contemporaneously altered. At this boundary the lie of the schistose gneisses is with that of the Massive Series, namely, with a high dip to the Change in the dip. ; eastward and having a general north-north-west to south-south-east strike; but after a short distance eastward the dip is soon reversed, and nearly all over the rest of the crystalline area the beds are dipping westward at generally high angles or are vertical, while the strike becomes more due north and south and is only in very rare cases ever to the east of north. This sub-division consists in great part of hornblendic, micaceous, taleose, or chloritie schists, or well-foliated or Lithology. i Р : о . laminated and more massive gneisses with varying proportions of quartz, felspar, hornblende, and mica; while there are very many subordinate beds of laminated quartz-rock or quartz-schist approaching very closely to detrital quartzites. A fairly distinct band of more eminently schistose rock occupies the An eminently schistose Western edge of the field, the schists being talcose, pande chloritic, and micaceous, with frequent intercala- tions of hornblendic bands, throughout which is disseminated a tremen- C eo) 26 KING: NELLORE PORTION OF THE CARNATIC. dous deal of quartz in the form of irregular segregations and very in- distinet veins. The micaceous and talcose forms occur in greatest force on the right bank of the Penner, between Käluvaya and Thälagapur, whence they run to the south-east as a narrowing but tolerably distinet set of strata as far as Saidapuram. About halfway down the southern course of these, and on their western side, come hornblendie schists, oc- casionally massive (coarsely crystalline with radiated assemblages of short needles of hornblende), but mainly schistose and acicular, and these gradually inerease in width and persistence down to the Swarna- mukhi. To the north of the Penner the hornblendie and more highly schistose bands are not quite so distinet, though they are traceable as such for some short distance on towards Udayagiri. Eastward of a line drawn from the latter village past Yarabah on the Penner through Tummulataluptru to Ojili Less schistose band. Е on ће Swarnamukhi, the rocks become gradually coarser and more distinctly bedded, though there are still frequent interealations of fine mica schists with a more quartzose constitution, | with separate beds of quartz-rock or quartz-schist. About Saidapuram and Tummulatulapüru there is a strong run of na нуі garnetiferous strata, and this is traceable at inter- опаш ор vals right up to the northern edge of the sheet, where also are associated cupriferous schists well known as the copper beds of the Nellore district. The garnets in these beds are sometimes quite remarkable in their size, beauty of crystallization, and number. Towards the line of Nellore and Gudir, a few thin outcrops of more Ridges of massive Massive grey gneisses of quartz felspar and horn- SUCI blende occur, and these are, if anything, in greater force to the northward on the other side of the Pennér near Buchireddi- palem, where this rock has been quarried for the new pagoda at Nellore. With this exception, however, there are no strong or distinct representa- tives here of the more massive gneisses of the Madras, South Arcot, and Triehinopoli distriets, which are more apparently represented, however poorly, by the Kálahásti outerops of the massive crystallines. ( ізі ) THE GNEISSIC SERIES. 27 To the north of the Penner, the succession of the foliated gneisses Ie ae mes MEL OES eastward of the more eminently schist- north of the Pennér. ose band is in tolerable accord with that just described for the middle area, but here the series has widened out con- siderably until it occupies almost the whole northern edge of the field, the strike of the foliation or bedding being, however, more regular and northerly. This great increase in width is in part accounted for by the more frequent intercalation and greater thickness of the subordinate quartz-rock or quartz-schist beds which here become quite a marked feature in the district, and give rise to many and distinct hilly ridges. The remarkable and unique feature in the schistose erystallines as The subordinate quartz. 9 Series, is the presence of these subordinate beds schists. of quartz-rock or quartz-schist, and these attain an additional interest from the fact that they, in the middle area, that is, between the Pennér and the Swarnamukhi, are to all appear- ance a perfect graduating set of beds, from manifest coarse quartz-rock to fine compact quartzites, very similar to those in the Cuddapah or Transition Series. To the north of the Pennér also, the beds as they are crossed westward become finer and finer-grained until they are compact waxy quartzites on the western edge of the foliated gneisses ; but there do not appear to be any fair cases of conglomerates or pebbly beds, or even of ripplings, though the resemblance to the Cuddapah quartzites is otherwise most remarkable. ! Captain Newbold also noted the character of these quartz-schists, writing of the neighbourhood of Sangam on the Pennér, andthe rocky ridge there running down into the river: “It is composed * * of a massive quartz rock in indistinct stratification, cleft occasionally, like the laterite, by intersecting partings and vertical fissures which divide the rock into parallelograms. * * This quartz rock passes from opaque and granular to compact, translucent chert of various shades of red, brown, green, and white. It contains disseminated scales of mica of a golden colour, which glitter like those in avanturine, and nests of brown iron ore. “If Ше marly (? nearly) horizontal partings are really the planes of stratification, it may be inferred from its conformability that this quartz rock does not belong to the hypogene series which is seen in highly inclined beds near its base, penetrated by veins of granite (as seen at Pollium, a village between Doroor and Sangam), but that it is an altered outlier of the sandstone mural crests which are seen from this on the western horizon capping the granite and hypogene schists of the Eastern Ghauts."—Jowr. As. Soc., Bengal, Vol. XIV, 1845, p. 898 et seq. ( 185 ) 28 KING: NELLORE -PORTION OF THE CARNATIC. The difficulty indeed was at last to decide what we should consider > as quartzites of the one series or the other, for . Difficulty of distin- , enishing these from the it was soou found out that along the outer edge Cuddapah gua алы of the transition rocks recognisable, Cuddapah strata are associated with the gneisses in a most obseure way, even to theirhaving apparently been wedged or let in as thin strips among the gneisses. Mr. C. Æ. Oldham appears to have relied principally on the micaceous constitution and the cleaved character of the beds in the gneiss, though such characters are common enough among the mani- festly detrital quartzites well into the interior of the Cuddapah area. I myself was rather guided by the untoward positions of fairly rippled and pebbly beds in my recognition of them as transition strata, and by the continuance of strike with the gneiss for tolerable distances, or as- sociation with hornblendie beds, as distinguishing features of the gneiss quartzites; though even with these guides I was continually thrown out by the knowledge that the different quartzites could be quite easily thrown together by faulting among the hornblendie rocks, as in fact is often the case with them between Káluvaya and Paremkonda. The succession or changes in the subordinate quartzites may, however, Lithology and distri. be perhaps best shown by the following details patiga; taken in the country westward of the town of Nellore; and the same style of beds is traceable to the north of the Pennér and in the Swarnamukhi region, though in this latter direction they are much, and eventually completely, covered up by superficial deposits. To the west of Nellore, in the rather pieturesque ridges of Narasimha- IG sudo hs konda, there is а strong development of thick beds quartz-rock. of quartz-rock among schistose gneisses. Here the rock is a massive pale-brown or white, occasionally ferruginous, usually very coarse granular crystalline aggregate of quartz, sometimes not at al unlike a vein quartz. Still it 1s not so massive but that there are small ragged and irregular crevices between the glassy and semi-translucent partieles of quartz which were once in part oceupied ( 1867) THE GNEISSIC SERIES, 29 by mica. There are coarser and finer beds, but no approach to a dense fine-grained quartzite, and in the less coarse beds lamination is recognisable; though, on the whole, the sedimenfary origin is only manifest in the distinetly and "for long distances bedded character of the rock among the other gneisses. The debris of these beds scattered over the flats below the ridges is quite as peculiar as the rock itself, being a sort of sharp gravel of glassy quartz. This Narasimhakonda quartz-rock is easily recognisable all over the distriet. To the west of these ridges are further beds of this rough quartz-rock, which show better to the north-west at Tärndipali as they drop down to the Penner alluvium. The beds here are still of the same character as those of Narasimhakonda, very coarse, crystalline, glassy, and laminated. Still going west, a strong band of frequent out-crops and ridges of quartz- rock runs from Suripalem on the right bank of the Pennér, south and south-east through Nandiwai, Lingumpilly, and Davanavamur to the Kandleru valley, which still possess generally the character of the Nara- simhakonda rock, but have frequent intercalations of finer and more compact strata. This band is the eastern arm of a fan-like or divergent strike of the foliated gneisses which commences at and spreads north- ward from Ingoort, and it will be seen further on how the rocks vary in the western arm. In this the Ingoort neighbourhood, the first ridge west of Lin- Finer and compacter gumpilly! is of various styles of quartz-rock, from varieties about Ingoort. tho very coarsely crystalline to the more compact granular, among which are lenticular seams of schistose hornblendie beds. There are compact grey and yellowish quartzites, looking like strings of vein quartz, running in the bedding, and, again, coarsely crystalline granular rock like that of Narasimhakonda. About a mile further west the main or highest ridge forming the western side of the Ingoort nucleus is almost entirely made up of thick beds of pale grey and sreenish quartz-rock or quartzite, a quartzite which more and more 1 Among the ridges west of Lingumpilly there is a poor outcrop of beds of grey crystalline limestone striking north-north-west with a vertical dip. (| 9059) 30 KING: NELLORE PORTION OF THE CARNATIC, resembles the dense and fine granular rock of the Transition Series. At the highest point of the ridge these are nearly vertical, and show faint traces of ripplings. To the west of Ingoort, frequent beds of i quartz-rock, with micaceous partings and lamin, ae occur among garnetiferous hornblendie schists. The northern end of the Lingumpilly ridge is composed of beds of al- most pure white or bluish-white coarse erystalline granular quartz- rock with scales of silvery mica running through it. In a low ridge about 2 miles north-east of Lingumpilly, there Saccharine. Қ d 2 are beds of a beautiful bluish-green semi-trans- lucent saccharine quartz-roek. he western branch of the Ingoort diver- gence runs up and meets the Penner about Pátapád, where at the north end of Ulavapali tank are beds of granular-erystalhne grey and buff quartz-rock, less coarse than that of Narsimhakonda, but not so compaet as ordinary Cuddapah quartzites. In the same neighbourhood, Pálaeole ridge consists of beds of slightly micaceous quartzites, much contorted in the strike of the bedding. Near the bottom of the ridge on the eastern side, there are some thick beds of compact dark-bluish- grey quartzite, and the same beds appear, but riddled with strings of white quartz, at the foot of the Mämadüru ridge. This last ridge runs right up to the Pennér alluvium, and is made up of successive beds of flaggy mivaceous quartzose gneiss, micaceous and taleose schist, and quartz-rock. As the rocks are now examined westward, the quartz-rocks, or rather more generally quartzites, become much more frequent in thinner bands with thinner beds or even flags; while the proper gneisses are be- coming more finely schistose and less quartzose as the more decided band of fine schists is approached. At the same time, there 1s a greater Epidotiferous or pis. frequency of trappean and trappoid outerops ac- (CU RELL: companied by a marked occurrence of epidote or pistacite through the strata, the quartzites occasionally being quite green-coloured from the amount of this mineral distributed through them. Southward of Dásür there is a narrow and rugged little ridge ( 188.) THE GNEISSIC SERIES. 31 of very vitreous-looking quartzites much split up by jointing and con- torted. In the south-west side of these beds is an outerop of jaspery quartz-rock with epidote. In the stream between Däsür and Purvati- puram there are sehistose quartzites. Further south-south-west and to the south of Kumarigunta, these seams or beds of quartzite are intercalated with taleose and chloritic schists. Between this and Tummulatulapáru very little rock is seen owing to the superficial coverings, except in the beds of the larger Ingoort nala and that south of the first village, where occasional interesting sections are exposed of the clear association of these quartzites with the schistose gneisses. In the nala south of Tummulatulapüru beds of quartzites Waxy quartz-schists, 20d gneissoid rock are continued past and due rippled, west of the village, and close by hereis a band of pale-blue and grey waxy quartzite, which is slightly rippled. As this belt is followed south-east, we again pass from the more compact and quartzite forms of this rock to varieties approaching the Narasim- hakonda type. To the westward of Tûramulla there are two rocky ridges composed Close similarity to the Of beds of white grey and greenish waxy-look- Cuddapah beds. ing quartzites, associated quite conformably with the schistose gneisses. These areas like quartzites of the Cuddapah series as possible, the only difference being that they are, if anything, more crystallized and sugary-looking. They are distinctly laminated and oeeasionally false-bedded, but there are no signs of conglomer- ates or pebble beds. The continuation of these same beds shows east of Tummulatulapüru, when they are still exceedingly like trans- ition beds, while they are furrowed slightly on the bed surfaces either as ripplings, or as crumplings from lateral squeezing, the furrows being long and parallel to each other, though they also run into each other after the manner of ripplings. Lastly, at the southern end of this band, there is a remarkable case of variation in the style of these quartz-schists in a very obscure case of an apparent conglomerate, At C 10 Conglomeratoid gneiss, 33 KING: NELLORE PORTION OF THE CARNATIC. - Shamudtha, about 4 miles due south of Saidapuram and at the north end of the tank, there is a ridge of highly micaceous quartz-rock, the bedding striking north-north-east to south-south-west, and the dip being an undulating or rolling one. The massive quartz-rock is occasionally weathered into what appears to have been a coarse conglomerate of big quartz pebbles with their longer axes in the direction of the strike. "These lentieular rounded masses are thickly eoated with silvery mica, which is stringed out as it were along the lamina ; close by and east of the quartz- rock there is a schist in which the miea preponderates over the quartz to an enormous extent, and beyond these are dark-green chlorite schists. I think this ean hardly be considered a conglomerate, particularly as to the north, in the same line, compact quartzites occur of a clear conglomeratie constitution. Itis more probably a caseof micaceous quartz-rock showing a very exaggerated form of segregation of the quartz in rude spindle- shaped masses—just as the quartz in ordinary coarse micaceous schists is often found assembled in small lentil-shaped masses—around which the mica is gathered and runs thence in the laminz of the rock. The original condition of the quartz-rock or quartz-schist beds is often shown in the weathered outcrops, a very good display occurring in the bed of the nala running past Vádanapurti some 6 miles north of Däsur, where the section gives a series of schists and shales of pale-green, grey- ish-green, and dark-green colours, with grey, yellow, and yellowish-red sandy shales, the latter being merely weathered quartz-schists and exactly like recent sandy shales. Thus far, I think, it is fairly clear that there is a gradation in the IV GERENS quartz-schists from the excessively coarse beds of series evident. Narasimhakonda through more and more compact beds to the sugary and then the waxy form of quartzite, and even to fairly rippled and pebbly beds, though coarse granular crystalline forms are. still met with here and there with the denser strata. However, this gradation runs closer in the neighbourhood of the boundary between the erystallines and the transition series; and here I must allow the great difficulty there 1s in distinguishing the quartzites (0 | THE GNEISSIC SERIES. 33 of one or the other series, while to some extent my discrimination has been more guided by the experience gained while working over such rocks in the ` field. It is a well-known fact, as may be fully exemplified in the ex- perience of some of my colleagues, over the Gondwána groups for in- stance, that one can very often tellto which members certain very similar- looking rocks belong, without being able to explain why he so recognises them; and it was thus with both Mr. Oldham and myself after some time among the rocks of this field. There are certain outerops of quartz- ites which we were not able to diseriminate, and which possibly never can be relegated with certainty to one or other series ; but these are either on or very close to my approximate boundary, and they are so smallas to have little or no effect on the main division. A very curious band of quartzose rocks, already slightly noticed, The pistacite quartz. ОССЕ just east of the boundary, among the Bea. hornblende schists, which are highly charged with epidote, either in a minute way so as to give them a green colour: ог with the mineral scattered through them in the lamination : or irregularly, like the garnets in а garnetiferous schist, when also the beds sometimes assume the appearance of a vesicular igneous rock, the epidote having weathered out leaving small cavities. These quartzites occur as a narrow band among the schistose acicular hornblende rocks and traps of the more finely schistose erystallines running through Dastir nearly to Támoy, and again—possibly by a throw—further to the eastward of the latter village from Thocapalem south-eastward. This form of quartzite is often so hard and compact that it looks almost like a jaspideous rock, and in my notes I find it often referred to as jaspery quartzite. It occurs frequently in the schistose gneisses between Yarabali and the boundary at Kaluvaya, and thence southwards as far as the Kandléru river, always associated with hornblende schists and occasionally with what I cannot consider as other than strong intrusive sheets of diorite ; but they are in greatest force in some low ridges extending from Dásúr past Tümoy, where also there isa great deal of extravasated trap. Again to the south of Thocapalem, on the Täramulla stream about Chinapalem and westward C (БЛА) 34 KING: NELLORE PORTION OF THE CARNATIC. of that village, pistacite quartzites are also common, but always associated with schistose and massive hornblende strata, while there is also a further development of the trappean intrusions, this part of the field being just at or near the south-western extremity of the greater trappean development yet to be detailed as extending diagonally across the Swarnamukhi-Kand- léru country. Certainly the pistacitie character of the quartzite band seem- ed to me to be intimately connected with the stronger development of in- trusive trap along this particular line of country; but there are features against this view, such as that the quartzites of both series alongside or on the trappean development of the Swarnamukhi-Kandléru country are not charged with epidote, nor are those above and below the traps with intermediate pistacitic band extending south from Käluvaya. The more probable view is that we have here a set of quartzites charged with an accessory mineralin a similar way to that of the garnetiferous quartzites and schists a httle further to the east, the association of them with the traps being merely a coincidence. To the south of, or on the right bank of the Swarnamukhi, the schist- à ose band becomes industrially important through Iron-stone quartz-schists. the occurrence of a rather strong development of quartzose iron beds in the ridges extending from Tresulmare to Ircolah, when they are in association with, apparently above, a thick series of hornblendie schists. These are traceable again on the left bank of the Swarnamukhi and run up alongside the great trappean outburst, and are probably, though now broken across and displaced; a continuation of the Däsür series. The proper iron-stone band is on the west side of the ridges extending from Tresulmare, where, how- ever, the country isflat and the outerop of the beds is only seen in the flat ground below the tank of that village. The rocks on the western side of the iron-stone band are all hidden by alluvial and superficial deposits, but on the eastern side there is a broad belt of the hornblendie schists touching on the great northern high road, and the more acieular form of the hornblende rock is again to the east of this band about Rásanur and Coonrum. (13127) THE GNEISSIC SERIES. 85 The beds are thick and numerous, consisting of lamine of quartz and grey iron-ore weathering into the brown or red hematites—more usually the brown peroxide—, the lamine being often thick and very distinet, with occasionally a brecciated character, angular frag- ments of quartz being cemented together in a ferruginous matrix. At the Ircolah end of the outcrop the lamine of iron ore are well weathered out,leaving the surface of the rock curiously roughened by outstanding and wavy little ridges of quartz. Small smelting furnaces are set up at Räsanür and at the neighbouring villages of Con- noor and Mahulpadi, the latter, though some 8 miles to the north- east, obtaining its ore from Räsanür. I did not see that any ex- cavations had been made in the beds of ore; the people, asin most other iron distriets, only work at the debris and more weathered por- tions of the out-erops, where they can get fragments of ore with the smallest cores of the hard grey oxide, and from which the siliceous particles have been well weathered or loosened. The neighbourhood of the boundary between the two gneisses, and Great development of again between the newer gneiss and the transition E series, is marked! by an extraodinary developmen of small reefs or veins of quartz which nearly always run for short distances apparently with the strike of the foliation or bedding, though their generally real lie is with an ill-defined north-north-west to south- south-east cleavage, having a low dip to east-north-east. These reefs are more frequent on the schistose side of the gneiss boundary, and they are developed to a most remarkable extent on either side of the line between the foliated gneisses and the transition rocks for a width of 7 or 8 miles. It seems quite evident that this development of quartz followed on or took place during the period of movement or crushing up and fracture of the transition series, though it has not followed - the present lie of these rocks. The probability is that the line of disturb- ance was more nearly with the boundary between the gneisses thus strik- ing from Gelacapad to Kälahästi, while the present fainter occurrence of quartz in the gneisses is only the dying out of a southern extension of ( в ) 36 KING: NELLORE PORTION OF THE CARNATIC. the northern development which has been denuded from the Vekatagiri interval. The material of these reefs is nearly always white quartz, much cleaved and fissured, giving very sharp and acute-angled debris in long fragments and splinters. There is often a good deal of schistose tale and chlorite fragments distributed through the veins; and small strings or minute specks of micaceous iron ore are common. I only saw the faintest traces of green carbonate of copper, or of copper pyrites, in the quartz associated with the massive traps between Yarabali and Kaluvaya ; but, as will be seen further on, it is rather in the traps themselves that the ores of copper oceur in this district. CHAPTER IV.—THE TRANSITION SERIES. CUDDAPAH FORMATION. This series oceupies only a very small portion of the country, as com- pared with the gneiss area, while it can be defined very simply as a set of quartzites and slaty beds belonging to three or more groups of the Cuddapah formation. Full details of this Madras member of the tran- sition rocks are given in the Memoir! devoted to its description, the classification or grouping therein laid down being as follows, in descending order :— 4. Kistna group. CORDED ATC ORMATION 3. Nullamallay group. 2. Cheyair group. 1. Paupugnee group. The lowest of these groups is not, as far as is kuown, represented in The Cuddapah groups the present area, but the strata of Kambák Droog, in this area. the Kálahásti ranges, and the Veligondas as far north as the parallel of Venkatagiri, are of the Cheyair group. A further reach of the Veligondas, as far as Räpür, is made up of succeeding beds of the Nullamallay quartzites and slates, and the rest of. this range represents the Kistna group. The eastward outliers of this range and 1 Mem. Geol. Sury, India, Vol. VIII, p. 1. (2 3347.) THE TRANSITION SERIES, 37 some small outerops in the low country are of undecided age, but they may be of the Cheyair and Nullamallay groups. No fossils have as yet been discovered in these rocks, or in their representatives in other parts of India, so that the little that 1s definitely known of their possible age is only ascertainable from their stratigraphical relations. In this field itself they are not directly associated with any other formations, but in the Cuddapah distriet they are unconformably overlaid by the Kurnool series, which is again unfossiliferous. In the adjacent Madras district they are, I think, directly overlaid by the upper Gondwäna beds; and in this area, these beds, or the plant shales, contain pebbles of the Cuddapah rocks. Іп the Godavari district, the representatives of the Kurnool and Cuddapah series are each overlaid by the oldest fossiliferous rocks of Pen- insular India, namely, the Talchirs, which are supposed to be of upper paleozoie age; hence the Cuddapah series must be very low palzozoic rocks, if not much older. In our Indian classification, the CUDDAPAHS, Gwatiors of Central India, and the Karapars of the interior Deccan, are rovisionally ranged together as upper transition rocks. y 5 8 p Age of series. In this region the rocks of the series are either quartzites or clay- EAM es, the former polus; the more prevalent, und giving the grand cliffs and scarps so characteristic of the above hill ranges. Kambäk Droog may be said to consist almost entirely of quartzites, though in its southern portion its base is of gneiss, capped, however, by a good thickness of quartzite sandstones and con- glomerates. The Kálahásti range, though mainly of thick beds of quartz- ites, still shows many bands of coarse clay-slates. The Veligondas, up to Venkatagiri, consist of some very decided and thick beds of clay- slates among still preponderating quartzites, the range of hills being thus broken up into many long valleys and some conspicuous ridges and outstanding cliffy masses, such as Venkatagiri Droog and Koyamon Корда. Тһе strike of these beds runs north-north-west, and, as the range of hills bends more northerly, these with their eastward dip gradually become hidden under, or are succeeded by, the further higher groups of (15 | 38 KING : NELLORE PORTION OF THE CARNATIC. quartzites in the rest of the Veligonda range. From Räpür northwards, the range runs with the strike of the beds, and is almost entirely made up of tremendous thicknesses of quartzite strata, giving the more regular vertically furrowed wall of this part of the mountains of which Penchala- konda is one of the highest points. To the north of the Penner the Veli- gondas are still mainly of quartzite strata, but with many bands of slaty beds. Except in the case of. the southern end of the Kambák Droog, these masses of transition strata, however lofty they may be, are always cut or denuded down to the average level of the gneiss floor, but to the north of the Penner, as in the Udayagiri Yerakonda and Dargadevi Konda, the quartzites are capping the gneiss either at a lofty elevation (3,000 feet), or on the long slope of the hills, and so forming a back on the gneiss mass. In fact, for this part of the field at least, there are, in these hills, unmistakable bottom beds at fully 2,000 feet above the level of the junction of the gneiss and sub-metamorphie rocks a few miles to the westward. The Udayagiri outlier is the only perfectly natural one in the district, lying, as it does, with clean eut cliffy edges, on a denuded floor of gneiss ; and the Kambák Droog is the next most perfect. In all other cases the boundaries of the transition rocks are ill-defined with the strata dipping at high angles or faulted against the adjacent rocks. So far there is no doubt as to the rocks being of the Cuddapah form- Obscure members of tion, and their lie presents little difficulty to the Sus series: observer, but other patches and outlying strips of quartzites are met with, which are so altered in their character and appearance, and so placed in relation to the older crystallines, or so un- usually situated themselves, as to have rendered the ranging of them in their proper series a matter of considerable perplexity and difficulty. These may be best considered under the headings of the Pillaméru, Kandra, and Gelacapäd-Käluväya areas, and also in this order starting from the Kambák and Kälahästi ranges, from which the first is only separated by a short interval of river deposits. In Kambák Droog the lie of the strata is in accordance with its ( 146 ) Relation to the gneiss. THE TRANSITION SERIES. 39 plateau-form, the dip being inward from the edges of the range, In the neighbourhood though there is much undulation in places. The of Kälahästi. rocks are more generally grey and buff thick- bedded compact quartzite sandstones and conglomerates. In the Kala- hásti ridge and the neighbouring Batanaik Konda the lie of the strata is not so easy, the undulations in the latter being intensified to the north until there are fair indications of a squeezed and then faulted arrange- ‘ment of the beds at Kálahásti, where in the small hill (250 feet high) to the north of the town, false bedded altered conglomerates and sand- stones are faulted against the gneiss. To the north of this hill no rock is seen for some 7 miles until the north bank of the Swarnamukhi is reached, when there crops up a narrow somewhat curved strip of quart- zites which I cannot but look on asa wedged-in outlier of the Cuddapahs. It is to be noted that the river is here diverted from its ordinary course in a decided north and south run of nearly 8 miles in length, which may be attributed to the existence of a now-denuded and hidden prolongation of the hard quartzites of Kalahasti, or to a line of fracture in this direction. This outcrop, or the Pillaméru strip of quartzites, forms only a very The Pillaméru out- low ridgy rise of rocky ground, much strewn on crop: its slopes with its own debris, and around which no exact boundary could be drawn. It is not a continuous outcrop, though it appears to be so owing to covering soil and lateritic breccias and conglomerates, and I am inclined to consider that it is really broken, as it follows a more or less angular line not always parallel with the strike of the adjacent gneiss. The two northern outcrops run with this strike, while the southern ones are rather across it. A set of trap dykes on the western side also seems to point to a fractured lie: that striking in a north-easterly direction from Sheearum would, if it be continued, cut across and between the two southern ridges ; a second, to the south-east of Carsearum, runs at the northern end of the proper Pillaméru ridge; and a third to the north strikes at the middle outcrop, ‘These quartzites cannot be ranged with those of the gneiss series ; they are undoubted sandstones and conglomerates, and are in every way like (ЛИЯ) 40 KING : NELLORE PORTION OF THE CARNATIC. the beds of the Cuddapah formation. "They are, however, much altered and squeezed, being cleaved and jointed into crude slates and flags often having a rudely fibrous strueture resembling that of fossil wood, while the pebbles and shingle of the conglomerates are apparently elongated in the direction of the strike. This strip is evidently isolated among the gneisses and lying with them for part of the outcrop, besides being in line with the Kálahásti beds. This position and their crushed-up character would therefore appear to point to their being simply a wedged-in extremity of one of the folds of the Kálahásti beds, and thus of the Cuddapah series. Indeed, it would scarcely appear necessary to enter on any discussion of the relations of these Pillaméru beds, so evidently are they, on the face of the country, a portion of the Kálahásti strata, were it not that they are possibly allied to the more peculiarly-situated rocks of the next area, About 10 miles east-north-east of Pillaméru, isolated masses of EUN quartzite un and conglomerates occur in. most abnormal positions in and on a great out- burst of trap rocks forming the group of low hills near the village of Kandra. The following notes will show the extraordinarily confused association of strata, and further details will be found in the chapter on igneous rocks. On one of the low hills to the west of Chillamanchen there is a frag- mental mass of quartzites, evidently of the Cuddapah series, consisting of blue and grey conglomerates and breccias, the pebbles of which are all of quartz. On the north side of the outcrop the beds have been much squeezed, the longer axes of the pebbles being with the strata. The main mass of the hill is of massive trap without any definite form or lie. The outerop of quartzites is a broken curve on the top of the conical hill, eiving rather the appearance of the lip of a crater, but the strike of the beds is not with the curve. The beds are dipping west-north-west into the trap, those on the top of the eastern slope at 10? or so, and those on the western side at about 50°. This curious curved fragment of Cudda- pahs presents the appearance of lying on and being sunk into the trap, ( 148 ) THE TRANSITION SERIES. 41 as though the mass had been torn up by and floated on the igneous rock. 15:18 not a remnant of deposition on the trap, though it at first sight looks like this, but is cut off across the strike by the intruded rock. Buff and blue waxy quartzites occur again in and on the long ridge to the west-north-west of Bonagudapolliam, striking north-north-west and dipping vertically into the trap of the south-west side of the ridge. The main backbone of the ridge is of these quartzites, but they are crossed by very large trap dykes in west-by-north or west-north-west directions. The south-west end of the ridge is nearly altogether of compact, occa- sionally flaggy, buff and grey waxy quartzites striking west-by-north or west-north-west with a dip of 70° or 80° south wards, lying on a great mass of trap below which are hornblendie and quartzose schists of the gneiss series striking north-north-west and dipping irregularly. The southern end of the low ridge due east of Kandra is of massive trap, twisted beds of quartzite similar to those of Chillamanchen being on itsnorth side. In passing from the trap to the quartzites I crossed a heterogeneous mass of outerops of hornblendie schists (or schistose trap) and quartzites without any appearance of parallelism between them, Тһе idea given is rather that of broken outerops of the quartzites with the intermediate gaps filled in by a schistose trap rock. A low ridge lying between Kandra and Thimmasamüdrum is partly made up of coarse grey quartzites traversed by or lying among dykes of massive compact trap, but the relations of the rocks is much obscured by the enormous amount of quartzite debris, for which reason also the direction of the dykes cannot be made out. The false-bedded quartzite sandstones are contorted, but have a general east-west strike. At the eastern end of the ridge about north-north-west of Kandra the beds are wholly eut off by a great dyke of aphanite running west-by-south to east-by-north. The southern end of the Kanakandroyan ridge, consisting of two detached hills, is a great mass or plexus of trap dykes, with an east-west band of vertical quartzite strata. "The smaller hill south-west of Con- dagünta is part of a great dyke, having on its south side a broadish band ( 149 ) 43 KING: NELLORE PORTION OF THE CARNATIC. of false-bedded quartzites striking east-west. Quartzites are lying on the top of the ridge in among the trap. At the eastern end of the ridge, about due west of Thimmasamüdrum tank, trap shows all round the abrupt slope, the quartzites being distinctly situated on the top. Further down the slope, however, there is a wide spread of greenstone, jointed in lozenges, north-west to south-east, and north-east to south-west, enclos- ing or embracing, as it were, bands of quartzite which appear burnt into the igneous rock. Mr. Oldham writes of this region :—“ The hill of Kanakandroyan consists almost exclusively of trap, hard close-grained, rather amorphous, having much the appearance of being bedded, striking north-west to south-east and dipping generally at a very low angle to south-west, with a tendency to break up pentagonally. Along the ridge, a little quartzite is seen, nearly continuous, only the highest point being free from it, It caps the ridge, apparently lying on the trap which in some large pieces may be seen united to it, the trap below and the quartzite above, both . somewhat altered, the trap being close to the contact, more earthy, and the quartzite more flinty." Proceeding northwards from this last ridge across the valley of the Venkatagiri river, trap ridges are again met with on which are two further long strips of Cuddapah quartzites. In allthese cases the quartzites are intimately associated with the basie igneous rocks and entirely separated from any gneiss. The resem- blance to strata of the Cuddapah transitions is perfect, while they are not like any quartzites of the erystallines. In the latter, it is true, the approach to ordinary compact and waxy beds of the newer series is very close, but there are no such distinet conglomerates, and fair obliquely laminated beds are rare. The difficulty was to account for the present beds being in such unusual position, and the only conclusion we could RHP e ucome to at the time of our survey was that they volvedamongand associ- are the remains of an outlying strip or portion of ated with traps. 0.00 с a compound anticlinal which formerly extended in this direction from the area next to be noticed, that of Gelaeapad and 505) THE TRANSITION SERIES. Үг. 48 Káluváya, and which possibly may have been connected with the Kála- hásti and Kambák masses. The lie of the Cuddapahs in their proper field is after all in a series of sharp undulations, so strong at times as to be foldings and even reduplications, having a north-north-west to south- south-east strike, the maximum of crushing being in and to the westward of the Veligonda range. The thickness, too, in this range, along its abrupt eastern boundary, is so great that the series must, prior to the faulting, have extended much further to the east, still perhaps in undulations which might thus give faulted or squeezed-in strips of strata among the erystallines, such as that of Pillaméru, or even broken and disrupted portions of such strips in association with a violent igneous outburst like that of Kándra. The noticeable feature, however, about the Kándra outerops is that No cases of association 14 they are disrupted, it is strange that none of the with gneiss. fundamental rocks are found in contact with them. In all cases the masses of quartzite are in contact only with trap. The conclusion is that these quartzites were deposited directly on the igneous rocks; or, preferably, that they are disassociated from the gneiss floor, and from each other in some cases, by contemporaneous or extravasated flows. A remarkable feature of this and the Pillaméru outlier is, that they end abruptly, the latter to the north and this to the south, their extremities being in a line parallel with the direction of some of the strong trap dykes in their immediate neighbourhood. I could not, however, carry such a line into the gneiss on either side, the different bands of the latter being continued on in the country between Pillaméru and Kandra, though these bands do appear to be broken in the Swarnamukhi valley on a line having somewhat the same direction. The northern end of the Pillaméru ridge may be cut off merely by faulting, along the strike, running up from Kálahásti. The Kándra area, however, has an abrupt end of some 3 miles in width, which is hardly explainable, except by some disturbance across the line of strike of the crystallines. Faulted to some extent. Caa) 41 KING : NELLORE PORTION OF THE CARNATIC. The southern extremity of the main area of the Cuddapahs, in the ОЕ Veligondas, is distinctly faulted, the intensity of dapah field faulted. the displacement having increased eastwards from Kirkambádi, while the upthrow to the south is shown in the Kálahásti hills Batanaik Konda and Kambak Droog in the present area, and Naggery Nose Nagwaram hill and others in the Madras area, most of which have scarped cappings of bottom quartzites looking to the south, at elevations of over 1,000 feet. The true direction of this Kırkambädi line or lines of disturbance could not, however, be ascertained ; all that could be determined being that the throw ranged more or less east and west, if anything rather to the north of east, that is, very much in line with the northern end of the Kálahásti ridge and the course of the river. Hence it is possible that minor disturbances in the same direc- tion, with others along the strike of the strata, combined with the extra- ordinary outburst of igneous rocks, were instrumental in bringing about the untoward position of these outliers and the highly altered condition of their rocks. Following these Kándra ridges of trap and hornblendie rocks by Gelaeapad-Káluváya — Potagünta, a broken line of quartzite outerops leads areas on to the Gelacapad-Käluväya area, which differs from that of Kándra in the associated igneous rocks, being dykes and interbedded intrusions instead of generally indefinable masses. Quartzites are still prevalent, but there are now some clay-slates, and the whole are mueh more highly altered, often to such an extent that they are hardly distinguishable from beds of the schistose gneiss. The Gelacapad portion of the area 1s an irregular group of quartzite : hills in three arms stretching from the village, west to Bodanapali, south- east to Räspolliam, and northward past Tamoy. Smaller ridges rise up on either side among schistose gneisses, but the bay to the south between Ráspolliam and Bodanapah is of trappoid rocks, apparently bedded north- by-west with an eastward dip, among pasty-looking syenitoid gneiss. It was utterly impossible in such a jungle-covered country as this is to make out clearly how the quartzites are lying with regard to the (31595 THE TRANSITION SERIES. 45 schists and traps alongside of them, but they are certainly not seen to i : lie naturally on the gneisses from and at which Relation to the gneiss. ESAE о : : they are dipping as it were along lines of faulting, thoagh the general outlines of the boundaries seem to be curved. The southern end is cut off in great part by the east-north-east line of fault striking out from the Veligondas at Rápür. To the west of the patch the hornblendie schistose rocks of the plain and the small ridges of quart- zites are much traversed by west-by-north to east-by-south dykes of ereenstone, and the main ridge itself 1s underlaid in part and cut by the same rock. The east side is again bordered, with an interval of grani- toid and schistose rocks, by a strong ridge of quartzite breccias and con- glomerates towards Govindapully, beyond which are schistose gneisses with many interealated small flows of trap and bands of quartz-schist hardiy to be distinguished from Cuddapah quartzite. These quartz-schists are, however, essentially micaceous, and are much charged with epidote in little separate assemblages of crystals, or in minute particles when the rock is of a green colour. "These quartzites are dipping at vari- ous angles to the eastward, and strike with the curve of the two southern ridges, making a bifureation round the bay of gneiss and traps to the south. The Tümoy arm is not continued across the Tummalatulapüru stream, but the smaller quartzite ridges to the west of it are traceable north- wards very nearly to the next strong outerop of transition beds extend- ing as a narrow curved ridge from Paremkonda to Káluváya, while these smaller ridges and the southern end of the Paremkonda ridge are flanked on the western side by a strong band of gneiss and granitoid rock. The smaller ridges are, however, in many cases so doubtfully either of crys- tallines or transition strata, that I have thought it best to rank them as of the former series, thus leaving a gap between the Gelacapad and Paremkonda outerops. From the latter place to Käluväya the outcrop is very decided, the beds of the ridge being good quartzite sandstones and conglomerates of unmistakable Cuddapah facies, lying with an eastward dip against (7 1587” 46 KING: NELLORE PORTION OF THE .CARNATIC. schists and granitoid rocks which are also dipping eastward on their sea- ward edge. To all appearance, this ridge might be one of a set Appearance of uncon- of quartzites in the schistose gneisses were it formity with the gneiss. not that I think there is some sign of general un- conformity of the beds on the schists to the westward, and that the schists on the eastern side do not follow the regular and marked curve of the ridge beds. It will also be seen directly how, in the prolongation of these beds beyond the Pennér, good evidence of unconformity was noted by Charles Oldham. There must be a curved line of fault running along the eastern edge Faulted eastern boun- OÉ this ridge, though all direct evidence of such dary. is concealed beneath superficial deposits. The erushed-up curve in the middle of the ridge length, the altered condition of the beds themselves, the enormous quantity of quartz collected in lamination and cleavage fissures all over and in the neighbourhood of the ridge, are all points in favour of this view. About the middle length of the ridge, where there is a sharp bend 1n і the strike of the beds, the quartzites are compaet Crushing and alteration. À А ; flint-like and much traversed by strings of white quartz silvered with mica and tale. The lower beds forming the back of the ridge are coarser ; and with them are some thin outcrops of trap. Further north towards the great tank, flint-like or jaspideous green quartzites are frequent, and the intrusive traps are tremendously deve- loped in a main sheet cropping up down the valley between the now widened and double ridge and in two other smaller ones. These extra- vasated traps continue northwards to the Penner, but not always between TET the same beds, for, though I could never find them Trap flows, intrusive. А ў breaking through, they disappear and re-appear above or below well-marked outerops, the large one of the valley appear- ing to have passed up over the highest quartzites as it is seen under the village of Káluváya, which is outside or to the eastward of the quartzite outerops. At the same time, there are many similar outerops run- ning with the schists on the eastern side of the ridge. Some of the ( 154 ) THE TRANSITION SERIES. 47 quartzites associated with the traps in the ridge аге epidotiferous, but these are not necessarily in contact with the trap. In the field,and for long after the survey, I was much inclined to Pu ETE consider that the rocks between the northern half uet p E of the Káluváya ridge and the Veligondas were Veligonda beds also of the same series, but on a thorough review of all my notes, I am compelled to give this point up and accept the suggestion often urged by Charles Oldham that the gneiss series is really continued in this direction. The rocks are certainly more schists than slates, the latter being essentially characteristic of the Cuddapah clayey strata, and they are associated with hornblende schists which must be considered, in this field at any rate, as belonging to the gneiss series. The most important point, however, is, that while the Kaluvaya ridges are continued on the north bank of the Pennér, the apparent Cuddapah facies of the rocks between Kaluväya and the Veligondas is not at all so marked in that direction. c At the same time I do not put forward the boundaries of this part of the field as hard and fast lines, for there is no doubt that many thin and long outerops of compact waxy quartzites and even some conglome- rates do occur alongside the Gelacapud and Käluväya areas, which I could not but look on as true Cuddapah strata, and which, if so settled, would narrow the belt of gneiss between the Veligondas and the outly- ing ridge of transitions very considerably. On the north bank of the Pennér, the low ridge of Varagüntapad is Continuation of the ОҒ Cuddapah quartzites, which are carried on to OS ee and form the long eastward-sloping back of the gneiss. Bommavaram hil or Dargadevi Konda, whence they are continued, with intervals of plain and covered-up ground, in the Yerakonda ridges. These beds are now beginning to cap the hills, giving grand scarps on nearly all sides, but particularly to the westward, and they are more evidently unconformable on the schists. At the same time, the eastern boundary, which is probably a faulted one, is hidden under the talus of (215527 48 KING: NELLORE PORTION OF THE CARNATIC. debris on this side of the hills. "Their strike now trends in to the Veh- gondas, from which, however, they must be cut off by faults, there being no outerop of bottom beds corresponding to them in the range. The Udayagiri plateau may be a portion of the same band, but even so, it is detached from it by a wide interval and by tremendous displacements, being, as it is, a denuded cap of nearly flat strata at an elevation of about 9,500 feet over the level of any quartzite boundary to the south or west. Mr. Oldham wrote of these ridges and hills north of the Pennér, and some of his notes are partieularly interesting as giving instances of un- conformity of the quartzites on the eneiss, even in the Dargadevi Konda, the rocks of which are undoubtedly continued in the Káluváya ridge. He also shows that these qnartzites are overlying not only the schistose eneiss, but that they extended on to the massive gneiss, a narrow strip of which crops out here. The Veligondas themselves consist for the most part of quartzites The Cuddapahs in the (conglomerates and sandstones of all kinds and | Mehgondu uso: colours) 1n great thicknesaes associated with fewer and thinner bands of clay-slates, micaceous and taleose slates, and still fewer schistose beds, all of which are dipping generally to the eastward, but with many undulations and some reduplieation. "The eastern edge of the range is fringed for the greater part of its length by a talus of debris Faulted against tho OF good width, which conceals the main eastern gneiss. boundary of the Cuddapah formation, though the newer and older rocks are at times traceable to within very close proximity. In all such cases, however, the indieations are that the boundary must be an abrupt one and faulted to a great extent. The beds dip constantly at high angles at the gneiss and have a crushed appearance, and the serial order of the rock groups seems to justify the conclusion that there must be great thicknesses of quartzites and slates faulted immediately west of and below the level of the adjacent gneiss. The boundaries striking westward and northward from Yärapet, at the southern end of the range, are certainly faulted, the downthrow inside these lines being at least 1,000 feet at the village, whence it decreased ( 156 ) THE TRANSITION SERIES. 49 westwards towards Kirkambádi. The great eastward break must run north-north-west from Yárapet to about the parallel of Kossi Konda, when it turns more northerly and continues still faulted, for it is running at а good angle across the well-marked strike of the beds to the parallel of Nagwaram hill. i Hence, but now with the strike, there’ must still be a faulted line running north- by-west to the parallel of Карі, еге being too narrow a space between the quartzites and the gneiss to allow of any sudden thinning out of such a thick series of rocks as 18 displayed in the sections tothe westward. At Räpür there is a wide and ver y abrupt abutment of the ends of the transition strata against the gneiss at right angles to the more general lines of fracture, on either side of which the older/and newer series approach very closely. From the same town, the boundary again runs north-north-west, but not always with the straightness assumed to be peculiar to a faulted edge, up to the Penner. Northwards from this river, the boundary is more curved, the eneiss and quartzites still, however, running so close that a natural boundary seems out of the question, and this is only more clearly indi- cated by the features already described in the account of the outliers to the east of this part of the range. | The rocks of the Veligondas have been deseribed in the memoir on -The rocks of thé Veli: the Cuddapah formation, and as they are of little gondas. . - i interest petrologically, being merely recurring and succeeding, from south to north, bands of quartzites and slates, it would be mere repetition to refer at greater length to them in this paper than has been already done. The quartzites are everywhere seen to be true sedimentary beds, great spreads of rippled sandstones being frequent all over this range, and their sandstone or conglomeratie character, if not seen at once, as is often the case in the wonderfully compact and dense rocks, is soon evident where they are weathered. Тһе clayey beds, on the other hand, have assumed more of a schistose character than is usual in the rest of the Cuddapah area, strong bands of rusty-brown and dark- green taleose and chloritie schists being common in the portion of the range south of Räpür, which are traceable to the westward into easy- D (С Та) 50 KING: NELLORE PORTION OF THE CARNATIC. lying clay-slates; and their altered. condition is plainly attributable to the great crushing and folding displayed in nearly every eross gully or gorge in the mountain wall. Throughout this deseription of the metamorphie and sub-metamor- C. JE. Oldham’s notes phic rocks, I have availed myself largely of the p ofthe notes and field maps of the late Charles Oldham, referring to the southern half of the field ; but as I myself only touched very slightly on the country north of the Pennér, I think it best to give his remarks on that area in extenso :— “The rocks to be noticed are throughout of the metamorphic class, with exceptions to be noticed hereafter, and may be generally described as alternating bands of hornblende schists, micaceous taleose schists, gneiss (proper), and quartz-rocks—an intensely quartzose gneiss. This quartz-rock forms a very marked feature of the country—many, indeed the great majority of the small ridges and hills, consisting largely or exclusively of it—owing, doubtless, to its great hardness and consequent power of resisting denudation and atmospheric action. * There are, however, as noticed hereafter, some instances near the ghats of a quartz-rock of another series, overlying and unconformably overlying these older metamorphic rocks. “In going across the country westward from the coast to the line of the ghats, we first rise from the alluvial deposits and sands over a scarped ridge of laterite nearly continuous from the river to the north of the sheet, and descending on the west, frequently over well-marked scarps and bluffs of 20 to 40 feet, find ourselves in the region of metamorphic rocks, which extend thence in uninterrupted succession to the gháts in a series of rolling beds. “ Loeally granite and quartz veins are numerous, but small. There is considerable variety in these metamorphie rocks. Typical gneiss, of compact substance, is comparatively poorly represented, though it occurs in several places, and is in some of these largely quarried for building purposes, as, for instance, near Boochareddypalliam (a large village about 12 miles north-west of Nellore), where on the north of the village a close-grained eompaet granitoid gneiss occurs in considerable queat ( 158 ) THE TRANSITION SERIES. 51 and has been worked extensively for buildings, both in the immediate neighbourhood and at some distance, I believe almost all the materials ofthe new pagoda at Nellore were quarried here. Much further west also, near the gháts in the neighbourhood of Bijjampalle, there is a con- siderable extent of granitoid gneiss (well adapted for building stone) forming hills of some size. Here, however, and also further south near Govindapully, where it also oceurs, but in smaller quantity, it has been but little utilised. ** More schistose varieties of the gneissic rocks are much more large- ly developed, and hornblende schists, mica schists, and taleose schists form the greatly preponderating portion of the rocks of this part of the district, garnetiferous hornblende schists being largely abundant. From these, however, I did not succeed in obtaining any crystals of garnets at all so good as some from the more southerly parts of the district, e.g., near Chittaloor, Thooroomulla, &c., from which places we procured numerous excellent crystals of various sizes from 1 inch and more in diameter to } inch. In the north-western parts of the area under notice micaceous and talcose schists prevail largely, frequently much contorted and often very slaty and earthy in character, and sometimes closely re- sembling some taleose slates in the * Cuddapah series.’ Indeed, in places the resemblance is so great that hand specimens of each could rarely be distinguished one from the other, and in some cases, where the more slaty beds of this Cuddapah series overlie these similar beds of the older metamorphic rocks, it becomes difficult to fix the line of boundary between them, specially when, asis sometimes the case, the strike is almost or quite identieal, and owing to the contortions (of the lowest beds specially) the dip becomes so locally ; and the similarity is great, not only between the talcose schists or slates of either series, but also between the quartz-rock or intensely quartzose gneiss of the one or the quartzite of the other. A case in point oceurs in the hill east of Bomaram (Bom- maveram), the lower part of which consists of hornblende schists, gneiss, and quartz-rock in alternating bands, striking about north-north-west and having a general dip to east-north-east at various angles, the beds yolling much, while the hill is capped by quartzite of the Cuddapah WEE. | 59 KING: NELLORE PORTION OF THE CARNATIC. series with nearly or quite the same strike and in some parts almost coin- cident in dip. And it is only by observing that in parts of the hill these Cuddapah rocks overlie the metamorphie series at a very different angle, that their general unconformability can be satisfactorily established. “А somewhat similar ease is seen in Oodagherry (Udayagiri) hill further north, where the low dip of the beds of quartzite which cap the hill, sometimes almost exactly coincides with that of the rolling contorted beds of taleose and mieaceous schists and slates, which form the mass of the hill, while this quartzite, although as I think generally, even in small speeimens, distinguishable, frequently so closely resembles quartz- rock oceurring in bands in the lower metamorphie rocks, that it renders the distinetion between the two at first sight difficult and doubtful. * Generally, however, as above noticed, the quartz-rock of the lower series, or metamorphie rocks proper, contains minute grains or scales of mica, which in some specimens are very clearly seen and appear to form extremely thin layers in the rock (sometimes well seen on a slightly weather-worn surface), but in other specimens are во minute as to be hardly observable. “ Here also, as in the other instance cited near Bomaram, though apparent conformability exists locally, the general unconformability is tolerably distinct. In both these cases, the capping quartzites form a fine precipitous scarp, which in Oodagherry hill is of about 200 feet in height, and almost vertical and wall-like. In Bomaram hill there is a much greater thickness of these beds, probably 400 feet, appearing on the southern face of the hill, and forming a remarkably fine precipitous bluff, a most conspicuous object as the traveller approaches from the westward. * In smaller hills north-east of Chunchulur I found a thin capping of quartzite lying over micaceous and hornblendic schists, the strike of both locally exactly the same; the upper beds dipping at lower angles and in places the schists underneath vertical. “Tn a small ridge south of Govindapully, nearly due west of Musta- fapuram, I note micaceous and quartzo-micaceous schists dipping at Т9? to 75° to east-by-north, capped by quartzites which dip in the same (ie) THE TRANSITION SERIES. 58 direction, but at much lower angles (25° to 30°). The lowest bed of these quartzites is here a coarse conglomerate, containing nümerous peb- bles of quartz of considerable size, 6 inches and more in length, of a flattened oval form, the longer axis always being in the direction of the strike of the beds. Above this is a hard grey quartzite, and locally, lying rather in broken disconnected patches, not forming any continuous bed, are some pieces of slates very like some of those occurring in the * Cuddapah series.’ j * Proceeding further north, outside the line of the Gháts, a low ridge extends from south-east to north of the village of Chabolu, composed of quartzites capping the schistose beds. Close to Chabolu are seen quartzo- felspathie gneiss and micaceous and talcose schists and some hornblende schists, striking north 5° to 15° west, twisted and rolling, but dipping generally at a rather high angle to east 10° to 15° north. The quartzites above dip at lower angles, but in the same general direction. The lowest bed is a conglomerate, a siliceous and siliceo-taleose matrix, including pebbles of quartz of considerable size. I have noted this bed as very similar to that seen near Govindapully, but the unconformability not so well marked. The ridge stretehes almost exactly in the direction of the strike. “There is also a thin capping of quartzite, partly conglomeratie, seen lying on the small hills north-west of Kothapulla and south of Annumpulla. Here there is very little seen, only a few feet covering the top and eastern slopes of the hill. The rock beneath here is a rather massive typical gneiss. <“ As regards the character of the metamorphic rocks generally in this part of our area, there is little to be said of much interest or novelty. As remarked above, typical massive gneiss is but feebly represented. A band of some considerable breadth, generally massive and granitoid, with indistinet bedding or lamination, extends along the east of the Ghäts from the Pennér to the northern edge of the sheet, with a breadth of 2 or 3 and even 4 miles, varying somewhat in character. Іп the southern portion of this, near Govindapully, &e., hornblendie and mica- ( 161 54 KING: NELLORE PORTION OF THE CARNATIC. ceous bands occasionally appear, but the mass of the gneiss is much less schistose and more massive, and has much more the character of typical eneiss than the rocks further east, and it is almost entirely without the numerous bands of quartz-roek which form so marked a feature through- out the remainder of this north-east quarter of the sheet. "The northern portion of this band is even more massive and granitoid, and from west of Beejumpulla to near the north edge of the sheet there are from 3 to 4 miles in breadth of a very massive granitoid gneiss, forming hills of considerable: size, the lamination or bedding of which is frequently very indistinet, and indeed, except in the mass, or where considerable surfaces are exposed, often not reeognisable at all. It is not, however, very largely or markedly crystalline. It is almost entirely devoid of schistose beds. Only in one or two places did I notice some micaceous sehistose bands in it. “In several other places, indeed almost everywhere throughout this _ area, oecasional, but generally very narrow bands of more typical gneiss, alternate with the hornblendie and mieaceous schists and quartz-roek | which form the main portion of the rocks, but none of them are of suffi- cient importance to require special notice. “The remainder of this portion of the sheet (north-east quarter) is made up of a series of alternating bands of hornblendie, mica, and taleose schists, and quartz-rock. The bands of the last-mentiened, as above observed, form a very marked feature throughout this part of the country. A large proportion of the small ridges and hills which vary the otherwise rather monotonous level ground consist either prineipally or exclusively of this rock, which is a highly erystalline quartz-rock, commonly grey or whitish in colour, but often weathering somewhat brown or reddish. It bears a very close resemblance to many of the very hard and crystalline quart- zites of the Cuddapah series, perhaps more specially to some of the bands capping the hills along the Eastern Gháts, e.g., in the Budvail Taluq, It may, however, be generally distinguished from these by some char- acteristie peculiarities. The presence of minute scales of mica can generally be detected, although they readily escape notice, and are com- monly clearly seen only on the surfaces of the lamine or thinner beds. C NGA) THE TRANSITION SERIES. ; 55 The surfaces moreover, though frequently well exposed in large flat sheets, are never, as far as my observations go, ripple marked, whereas the presence of ripple marks is extremely common in the hard quartzites of the Cuddapah series. "This quartz-rock is seen alternating in beds of very various thickness, with the hornblendie and mieaceous schists. The bands are sometimes only an inch or two in thiekness, and very numer- ous within short spaces, but frequently they are of very much greater thickness, and form ridges of no inconsiderable size. “These bands of quartz-rock with a similar character occur also abundantly south of the Penner, and there also form ridges as here, but they appear to increase in frequency north of the river, and spread out northwards in a fan-like manner, with their alternating schistose bands from about 90 miles south of the Penner to the northern edge of the sheet. 'The general strike of,the beds is about north-north-west, vary- ing, however, considerably, trending more to north-west in the westerly parts of this. quarter-sheet, and approaching more nearly to a north- south strike in the eastern parts. The quartzose bands are somewhat less numerous in the western portions, which consist more largely of mieaceous and talcose schists, with some hornblendie and quartzo-horn- blendie bands, and a few narrow bands of quartzo-micaceous and quartzo-felspathie gneiss. | | “ ''aleose bands occur also apparently continuous with some of these south of the Pennér, where I notieed them in the neighbourhood of Navoor, Yatoor, and north-west of Saidapuram, where near Jogipali, a highly taleose gneiss or tale-rock is quarried to a small extent and worked into small pots, figures, &e. - “In many bands, chiefly of hornblendie and quartzo-hornblendic schists, garnets are largely abundant, specially so in a series of horn- blendie schists extending in a north-north-west direction by Jummaw- drum, Kotapela near Kanigiri, Woopaloor, &c. A similar (probably the same) band of rocks continues south of the river near Ayergarpilly, Lingumpully, Thooroomulla, and near the last-named place we procured many excellent erystals, very perfect and of various sizes, from more than CIGA) 56 KING: NELLORE PORTION OF THE CARNATIC, an inch in diameter to less than ith of an inch. I noticed in one or two places in quartzo-micaeeous schists, mica appearing to replace garnets, having at least assumed their crystalline form." CHAPTER V.—GRANITIC AND TRAPPEAN ROCKS. The massive metamorphie rocks of the Kálahásti and Venkatagiri regions are only slightly marked by seams of hard binary granite of pink and white felspar and dull grey quartz running with and across the foliation ; but the gneiss is so frequently granitoid itself, that it is difficult to decide whether these are really intrusive or that they are mere segregations. The most noticeable development of granite is, however, among the The Ogili and Gádár ‘foliated gneisses, and here again the reefs are ganie nearly always running with the foliation, so that they must, if they are igneous, be considered extravasated veins, though they also appeared to me to be the result of segregation. These occur on or alongside the main northern road between Ogili and Gúdúr as a very crowded system’ running generally in a north-north-west to south-south- east direction among hornblendic, micaceous, and quartzose schists. From Güdür northwards granite is frequent at intervals all along this belt of gneiss right up to the northern edge of the field. "The rock is mainly a very coarse binary aggregate of quartz and felspar (orthoclase), the ordinary ternary form with mica (muscovite) being, however, frequent at places. At the Calingula or weir of the Karevádu tank, the micaceous and hornblendie gneiss is intruded on by veins, 3 or 4 feet wide, of- coarsely crystallized granite which run between the beds and along the east-west joints. The rock in this neighbourhood is of flesh-coloured 1 I found it difficult to represent the reefs of Gádár granite on the map as they are so numerous and yet not large enough to allow of their being separately mapped on the 4-inch scale. I have therefore delineated bands in one colour, which are to be considered as areas of stronger development. In the same way I have represented areas of trappean outburst in one colour, though the real condition of things is often a confused plexus of dykes and irap masses among hornblende schists. ( 164 ) GRANITIC AND TRAPPEAN ROCKS, 51 felspar bluish-white quartz and light-brown silvery mica, largely cıys- talised and charged with garnets. It often assumes the form of graphie granite. The granite veins do not often contain garnets, or indeed any foreign minerals, but the association with garnetiferous. schists is remarkable. In one instance about a mile and half east of Turamulla (on the southern tributary of the Kándléru) there is a string of tourmaline and garnet crystals occurring in one of the larger granite reefs, the garnets of the adjacent schists being large and perfect dodeca- hedrons. Every now and then along the road from Ogili to Güdür there are frequent outcrops of this granite, the road metal having been obtained from decomposed masses on either side, and sometimes the mica is so strong that the road and the adjacent fields shine out in the sun light. At the junction of the Madras and Dugarazupatam roads, the plates of mica are 3 or 4 inches in diameter. The commonest igneous rocks in the distriet are greenstones or diorites, which are, however, more particularly de- Trappean rocks in de- 3 í Қ finite and ill-defined out- veloped in the Kándra and Gelacapád areas already Шаға specialised оп aecount of their quartzite outerops. These trap rocks may be very conveniently classified in the present connection as definite and ill-defined outbursts ; the former being, of course, the well-marked dykes and intrusive sheets, while the latter are irregular masses of obscure origin, often presenting the characteristies of both dykes and sheets, but of great extent as compared with the true and obvious igneous intrusions. | A few large dykes occur in the south-west corner of the field, which are the dyings-out of a tremendous development in the Dykes in the gneiss. eee : South Arcot country. They are striking either west-by-south to east-by-north, or about east-west, and others run north- west or again nearly with the general strike of the foliation. The rock is usually a heavy massive somewhat coarse-granular diorite of dark-green or nearly black colours, and it is occasionally porphyritie with large erystals of pale-green felspar. In the Swarnamukhi valley, to the east- south-east of Kirkumbádi fort, there 1s a large west-by-south to east-by- ( 165 ) 58 KING: NELLORE PORTION OF THE CARNATIC. north dyke of compact porphyritic greenstone, with numerous separate crystals of yeilowish-green felspar. Many traces of dykes occur in this neighbourhood, some of which are evidently porphyritic at times and oftener compact and homogeneous. An interesting and instructive example of the spheroidal weathering KURE DA SE of passe traps occurs in the Kirkumbadi dyke, roidal weathering and where it crosses a stream about 4 miles east- Janine south-east of the fort, in so far as 1t illustrates the important part taken by jointing in the production of these rounded forms. The trap shows apparent bedding due to a plane of jointing dipping north at 15?, and it is also cut by north-west-by-west, north- by-east, and east-west nearly vertical joints. It will be seen from the accompanying portrait, of one of the flat joint surfaces, that a prismatic Portrait of a sloping joint-surface of a trap dyke. ( 166 ) GRANITIC AND TRAPPEAN ROCKS. 59 mass with a triangular base has been separated by the three main joints, but that other less persistent joints, or, preferably variations of the main joints, have given the core of the prism a polygonal and finally a rudely circular base, while the nearly flat joint system with its slight variations assists in the ultimate rounding of the block. From the Swarnamukhi northward to Rápür the number of dykes is small, none of them being of importance as to size or length, and they are all of ordinary compact diorite; but as the latter village is ap- proached, they become frequent and are remarkably persistent up the valley between the Veligondas and the Gelacapad-Käluväya hills. At Bodanapali, many wells, 20 to 30 feet deep, are sunk through grey schists, showing numerous small trap dykes, 3 or 4 feet wide, running in all directions and at all angles. They run' with the bedding, or across it, to the north and to the east, Between Rapur and Gelacapad, all the rocks are very igneous-looking ; a grey pasty porphyritie rock, which І, however, suppose to be the same as that of the grey gneiss series towards the Swarnamukhi valley, fills up the bay between the quartzite ridges and is traversed by strong dykes of aphanite trap in a north-by-west direction with an eastward dip. Further north, and eastward of Ulla- puram, the country is much intersected by trap dykes striking east-north- east, the rock being often a dense black aphanite. The low hill north- east of Poerapali is a perfect network of dykes, as 1s also the ridge to Ње south. The cross dykes are also associated with the transition rocks, for the Dykes in the transition Most part obscurely, but still at times as undoubted rocks. Intrusions. To the south-south-east of Tümoy, there 1s a large dyke at the north end of a low hill of quartzites, and this runs right at and under the ridges in an east-west direction, showing every now and then in the small valleys between, while a small branch of this dyke traverses the beds very clearly. Some of the dykes are very broad and long, that of Tümoy and another parallel one to the north being traceable for 3 to 5 miles with a breadth of 100 feet, while a third runs north-north-west past Panumurti for a length of about 7 miles, with a breadth sometimes of 150 yards. (467 9 66 KING: NELLORE PORTION OF THE CARNATIC. The Paremkonda-Káluváya ridges are also intruded on by sheets Intrusive traps of Ра- Of diorite which for a good part of their length Bu appear to be contemporaneous. However, on fol- lowing them out they are seen to vary much in thickness, swelling out every now and then, while they die out and are succeeded in the strike too suddenly by other similar outerops above and below the adjacent quartzites to be contemporaneous flows. "They commence thinly in the sharp eurve north-west of Paremkonda, and then rapidly increase in thickness towards and under Káluváya. Itis of course diffieult to re- cognise in generally altered strata any effects produced by such intrusions, but the quartzites in their vicinity are certainly jaspery or flinty, an uncommon condition of these beds, and they and the traps themselves are epidotiferous. To the south of Gelacapäd, there is again an unmistakable region Irregular outbursts of Of flows and ill-defined outbursts of basic rocks, trap. which, though probably part and parcel of one. great igneous development, is certainly disconnected from that of Parem- konda, either by an area of milder action, or, as I prefer to think, by an actual break along the Rápür line of faulting. I have represented this: southern area as continuous by the Potagünta ridges! on to Kándra in the map, and there is every reason to consider that it is so in fact, but a good part of it is covered up by the superficial deposits of the Venkata- girl river. At first sight, the Kándra or Kana-Kondroyan hills seem to be Wc d Meet wholly of trap, like the remains of a great pro- truded mass of voleanie rock ; but on examination they are found to be really of massive hornblendie and chlorito-horn- blendie rock, extensively traversed by dykes and irregular masses of trap, running mainly in north-west to south-east, east-west, and east-by-north to west-by-soath directions, the whole occupying a wide belt thinning out ! The Potagunta ridges were examined by Charies Oldham, and hie seems to have con- sidered that they, on the whole, consisted more of massive and schistose hornblendic strata, with fewer trap dykes and dioritic masses. (251650 GRANITIC AND TRAPPEAN ROCKS. 61 to the north- west and cut off abruptly to the south-east near Chillaman- chen. At the north-east end the traps have more of a bedded look, and from this point the inter-bedded or extravasated character of the out- crops becomes more evident as the Paremkonda ridges are neared. The hills immediately west of Kándra show more frequently the dyke form of intrusion, while the Chillamanchen hills are so devoid of any dyke-like development, except on their flanks, that they look more like a great mass or centre of outburst. Immediately west of the latter village, the small group of dark-coloured moderately-rounded hills consists 1n the main of a dense blue-black aphanitie greenstone, which weathers of a dark red- brown colour, occasionally into large rounded masses, or into an earthy rock speckled with littlé rounded masses of dull carbonate of lime. On the west side of the main hill there is a broad dyke striking east-by- north to west-by-south through the rest of the trap and across a narrow band of quartz-rock and tale schist, which 1s also 1n among the traps and has a north-west strike (the strike of the Cuddapah quartzites on the top of the cone is nearly north-south). Armenpadi, about 3 miles north-north-west, is a compact fine-grained blue-black aphanite, the ridge to the north-west being also composed of coarser trap weathering into great rounded masses. Kándra village is on trap which can be traced, associated with hornblendie schists, thence to Vendodu ; and northwards from the latter village the path to Vadacherla passes over a series of hornblendic schists traversed by, and interbedded as it were with traps in the same way as in the Kandra valley. The schists are con- torted, but strike generaily north-north-west. _ The dykes intersect each other both here and in the Raptr country, but without giving any indications of different periods of intrusion: both those across and with the strike of the country rocks appeared to belong to the same outbursts. The obscure association of the traps with the country rocks is to be seen on the western edgeof the Kándra area. On the Vizinagram side, highly contorted beds of micaceous and hornblendic schists, with frequent small seams of granite and small trap dykes, border the trap mass of the hill. The strike of the schists is north-north-west with various dips, (7169 ) 62 KING : NELLORE PORTION OF THE CARNATIC. but close on the trap the contorted beds are strong, with a west-by-south to east-by-north strike. The western side of the ridge is a great mass of east-by-south to west-by-north dykes traversing massive hornblendie taleose and chloritic rocks and micaceous schists, having a north-west to south-east strike, the traps being, however, greatly in the ascendant. The dykes are generally of a massive coarse dark-brownish-green diorite, which weathers into a coarse earthy ferruginous rock, while others again are more of an aphanite. The highest part of the ridge is entirely of massive and coarsely weathered trap, extending over many hundred square yards. Along the south side of the ridge the character and appearance of the rocks vary much. First, the trap forming the summit is a massive well-erystallized rock of a dark-grey nearly black colour of hornblende and felspar, through which run large, probably segregated, strings of dense blue-black rock. Next, the rock becomes more felspathie with the hornblende, showing in large crystals, or oftener in small assemblages of radiating needles, giving the weathered surface a starred appearance. This syenitoid character of the rock extends up to the Vendodu tank. Vendodu itself is on coarsely erystallized granitoid gneiss, perhaps rather a granite, a great band of which passes along this side of the range past Vizinagram and Armenpádu. Near these villages a band of coarse quartzo-felspathie rock, having a somewhat laminated structure, and including angular fragments of hornblendie schist, extends in a north-north-west to south-south-east direction. The rock is very coarsely crystallized and is riddled with numerous strings of quartz. To the eastward of this there is a band of massive acicular hornblendie rock, and east of this again is more granitic rock, after which come the schists with granitie strings and small trap dykes above noticed as bordering the trappean ridge. There are no cases of actual contact, the base of the hills being always covered with debris of the rocks above, or else by the soils which run close up to the traps of the hill masses. (7370) OTHER FORMATIONS. 63 CHAPTER VL—OTHER FORMATIONS. (1).—Rasmauat PLANT Bzps. The merest traces of some plant shales of this series were discovered by Charles Oldham on the northern edge of this area, which he afterwards carried on into the Kistna District, where they have since been thoroughly worked out and reported on by Mr. Foote." Indeed, it was after all by Mere traces of Upper the merest chance that the first indications of these Gondwanas in this area. Rajmahal plant shales were found in the debris thrown out of some wells which had pierced the laterite and sandstone plateaus of Nellore and Kávali. I myself found pale-yellow and buff clayey and sandy shales, very similar to those of the plant-beds of the Trichinopoly district, but unfossiliferous, at three localities—Chümuru- gunta, about 6 miles south-south-west of Nellore; Kálavakonda, 7 miles east-south-east of Güdür, and at Shengapetta, 17 miles north-north-east of Nellore, all in well-excavations. About the same time I heard from Oldham of his having found fossihferous shales of the same kind, and thus it became evident that the Trichinopoly and Madras belt of Rajmahal rocks extended nearly all along the Carnatie. Oldham subsequently found many little patches, so small that they can only be represented very general in the map attached to this memoir, and concerning which he furnished the following notes in 1862-63 :— * In the north-east quarter sheet, in the Kävali Taluq of the Nellore district, I have discovered and secured some specimens of plant remains from sandstones which from their similarity to, or identity jwith, plants discovered elsewhere, possess considerable interest. < The sandstone and shales in which they occur are confined, as far as I could determine them within the limits of sheet 77, to a few square miles west and north-west of the town of Kávali. I have noticed them close to the villages of Timmasamtdrum, Kún Korepolliam, Ramanja- puram, Comy Mutumarpolliam, and Sodawaram. Beds of similar charac- 1 Mem. Geol. Surv. of India, Vol. XVI, part 1. ie) 64 KING: NELLORE PORTION OF THE CARNATIC. ter appear in two or three other places, from which, however, I did not succeed in proeuring any specimens. « T first noticed these beds close to Ramanjapuram, where a very poor section is exposed in a bowry (or well). Here I noticed about 3 feet of a eoarse unconsolidated felspathie sandstone, with numerous small scales of mica, lying horizontally under about 2 feet of pisolitic kunker ; a red lateritie soil covers this to a depth of 3 feet, with a thin layer of vegetable mould on the surface. Here I could detect only traces of stems. For some distance east of Ramanjapooram, frag- ments of similar sandstones appear thrown up from bowries and in tank bunds, but the beds are certainly not continuous, as about 1 mile west of Sodawaram a very hard quartzose gneiss appears in situ, striking north-north-west with a high dip to west-south-west. The sand- stones re-appear on the west of the village of Sodawaram, where about 10 feet of them are seen in a well. The lower part a greenish-yellow felspathie grit, with scales of miea, with some pink and purplish bands . running through them ; the upper portion is finer and more compact, - slightly ferruginous. In this locality again I found only indefinite stem markings, and could secure no good specimens. This also is apparently only a small patch of these sandstones, as immediately on entering the village of Sodawaram, quartzo-micaceous gneiss appears,and on the east of the village quartzo-hornblendic gneiss and hornblendie sehists; and the sandstones, as far as І could find, do not re-appear in this direction. | « North-west of Ramanjapooram, in small stream courses and bowries close to Comy, similar sandstones appear; I find I have noted them as coarse sandstone, greenish-yellow, with pink bands, in flat-rolling beds. Kunkur of no great thickness overlying them, no rock seen unde but gneiss and hornblendie schists thrown up rom bowries close pd no. definite remains. In the stream north-east of Comy a very little similar sandstone is seen. From this eastward the stream runs through banks, of soil of some thickness, no rock appearing; but close to Mootoomor- polliam, in three or four bowries, from 15 to 20 feet of sandstone appear in level beds, quartzose and felspathie with scales of mica. Some finer EN) OTHER FORMATIONS. 65 and more shale-like beds oceur through them, chiefly in the upper part, and some of these are slighty ferruginous. “Tn several of these bowries I searched in vain for anything more than very indefinite stem-markings, but was at last rewarded by finding Some well-defined plant remains of Zamias, two or three species, identical as far as I could then determine them, with some of those obtained from the Rajmahal beds of Bengal. “ Passing eastward from this, the sandstones do not appear, and fragments of other rocks only (gneiss) are seen in the red soil which covers the ground ; but close to Mootealpaud north-north-east of Moo- toomorpolliam and just outside this sheet (77), I came upon them again in wells. There they are chiefly of the coarser and more gritty kind, > rather soft, variously coloured, greenish-yellow, pink, and in well-marked horizontal beds, nearly 20 feet of them being seen in one bowry and no rock appearing underneath, only indistinct stem-markings found. They have probably been cut through by the stream north of Moote- alpud, as they do not appear in the banks which are of brown sandy loam. <“ South from this and south-west to Kún Korepolliam red lateritie soil covers the whole surface of the ground, but on the south and south- west of the village the sandstones re-appear in horizontal beds in bowries, no other section being visible. About 7 or 8 feet are here seen, and I found specimens of Paleozamia and some stem-markings, but could secure no speeimens that would bear carriage, the rock being very soft and friable. * From this southwards towards Timmasamoodrum laterite soil with fragments of ferruginous sandstones conceals all rock, but close to Tim- masamoodrum the plant-bearing sandstones again appear, similar in character to those at Mootoomorpolliam, but of less thickness, about 8 feet only being seen, with kunkur (24 feet) above, and red soil on top. There I secured a few tolerable specimens of Palzozamia and? Teniopteris, and south of this the beds seem to die out, and I could not find them anywhere else to the southwards, quartzo-hornblendie and highly quartzose gneiss (quartz rock ?) coming in with some hornblendie schists. “Тһе next locality I have noted is north-east of Guttugully, about E Na) 66 KING: NELLORE PORTION OF THE CARNATIC. 1 mile, where I observed a coarsish ferruginous sandstone, with some fine yellowish bands, very little of them being seen, chiefly from small bowries. In some of these finer bands I found traces of plants which I considered sufficient to identify the beds as being the same as the plant bearing sandstones seen elsewhere. TEE ar vA 50) | | 50 | E | a hene ] as] AA / \ = cá y ATTN T» — | Го a LF тұман J \ ) ay a 4 N «| af ұшы X 49. [ 1 \ E ) E. ^ So NU ІШІ таласы \ | iM x as ы ر‎ еу; j 14 M 30 30 er TU. LJ e tna | ا‎ e. N E A = = i xj c N lau рт TA ЕЗ ТТТ ‘se ones ee 5 | i ri re alis as = GEOLOGICAL MAP а AS d N Ят E = E Joyipete ofthe ا‎ d | NELLORE PORTION. ее . j OF THE 2 ( 2 : г P E i CARNATIC. = 5 Е Р : Ш База Ера i > > = ms š З | Scale 8 Miles - lInch. pog N : wes > WG 25 14) w = E! E E WERTET \ N 3 Dd F m а МЕ С Б = / 1 DE br INDEX Ё \ 55 55! Blown Sand 7 E ee Men apadem hause) EU a so’ Be] Tats Debros бо! | Cuddalore Sansletorer БШ Najmakal Plant. Beds 45. gs REFERO APT СА aa Sehir tosa Gneunrer 1 Marsıve nesses P 0ء‎ к LL! Trap Dyker & Areas of Trappaan. outburst : L| ЕС m 554 EM (0: 15 alo’ als 7930 als dv als: Ar On Stone by S Khndem Hursein pea, i MEMOIRS OF THE GEOLOGICAL SURVEY OF INDIA. MEMOIRS OF THE GEOLOGICAL SURVEY OF INDIA. VOL. XVI, Pr. 3. PUBLISHED BY ORDER OF HIS EXCELLENCY THE GOVERNOR GENERAL OF INDIA IN COUNCIL, CALCUTTA: PRINTED FOR THE GOVERNMENT OF INDIA, SOLD AT THE GEOLOGICAL SURVEY OFFICE, OFFICE OF SUPERINTENDENT OF GOVERNMENT PRINTING, AND BY ALL BOOKSELLERS, LONDON: TRÜBNER & CO, MDOCCH ххх. CALCUTTA : OFFICE OF THE SUPERINTENDENT OF GOVERNMENT PRINTING. 1880. CONTENTS. CHAPTER IL—GENERAL DESCRIPTION. Introductory, 1.— Literature, 2.—Topography, 4.—Physical geography, 4.— Delta lands, 4.—Low country and hills, 5.— Rivers, 5.— The Godavari gorge, 6.—Its history, 7.—Delta branches of Godavari, 8.—System of irrigation, 8—The Kolér lake, 9.— Traces of an old marine floor, 10.— Its age, 11.—List of formations, 11 — . - CHAPTER IL. —GNEISS AND LOWER GONDWÁNA ROCKS. Tue Brzvapa GNEISS, 12.—Previous knowledge of, 12.—Extent and rela- tion to other gneiss, 12.—Lithology, 13.—KAMTHI SANDSTONES, 14.— Traced down from Central Provinces, 15.—Lie and association with the gneiss, 15.— Fossils and their localities, 16.— Association of fossils, 16 CHAPTER III.—UPPER GONDWÁNAS. Grouping, 17.— Extent and association, 17.— Fossils, 18.—GOLLAPILI SAND- STONES, 18.— Paleontology, 18.—Fossil localities and type area, 20.— Thickness, 20.—Persistent form of bottom beds, 21.—Lie, 21.—Suc- cession of strata, 22.—Thinning out and denudation, 23.— Unconform- able to Kämthis, 23.—RAGAVAPURAM SHALES, 24.—Lithology, 24.— A lenticular band between other groups, 25.— Local and recent disturb- ance in outcrops, 25.—Marine organisms, 25.— Fossil plants, 25.—Order of strata, 27.—Thick bands of shales, 27.—Clay-ironstone band, 28.— Section below Tripati scarps, 28.—Relation to Gollapili group, 29.— Very different rocks, 30.—Indication of separation of groups, 30.— TRIPATI SANDSTONES, 30.—Thickness, 31.—Lie, 31.—Lithology and succession, 32.—Variations in members of group, 32.—Tundkalpudi beds, 33.—Peddavegi beds, 33.—Cut off at the Tammiler, 33.— Represented in the Vizagapatam country, 34.—Fossiliferous beds of Ayaparáz-Kotapili, 34.— Marine fossils, 35.—General conclusions, 35 CHAPTER IV.—DECCAN TRAP SERIES. Trap scarps of Rajahmundry, 37.—Long known, 37.— Worked out by His- lop, 38.—Grouping, 38.— Fossils, 39.—INFRA-TRAPPEANS, 40.—Thick- ness, 40.— Lithology, 41.— Turritella zone, 41.— Fossils, 42.—Apparently Pıan. 1—11 19—17 17—86 vi CONTENTS. PAGE. unaffected by superincumbent trap, 42, —Relation to Deccan trap series, 42.— Cretaceous affinities, 42.—Resemblance to Lametas, 42.— Views of W. T. Blanford, 43.—' TnAPS AND INFRA-TRAPPEANS, 44.—Overlap the infratrappeans, 44.— Thickness and lithology, 44.—Fossils, 45.—Katéru outcrop, 46.— Two outcrops, doubtfully of the same band, 47.—THE TRAP, 47,—THE INTER-TRAPPEAN BAND, 48.— Моге or less crystalline, 49. —Thinning out of beds, and denudation, 49.— Fossiliferous seam, 49.— Apparent alteration of beds from below upwards, 49.—Case against alteration, 50.—Influence of weathering, 50.—Explanation, 50.—Ap- parent alteration of beds in Central India, 51.— Relation of this outburst to that of the Deccan, 52.— Асе of inter-trappeans, 53.—Hislop’s view, 53.—Blanford’s view, 53 5 с : 5 t : à e 37—54 CHAPTER V.—CUDDALORE SANDSTONES. Rájahmundry beds, 54.—Extent and mode of occurrence, 55.—Succession and lithology, 55 . 2 5 Я : : à : 5 6 . 54--58 CHAPTER VI.—ECONOMIC GEOLOGY. Building stones, etc., 58.—Diamond workings, 59.—Iron-smelting, 61.—The Pentlam area, 62.—Dr. Heyne’s account, 64.— Other minerals, 69.— Coal, 70 ; 5 : : : қ Д : : 4 А . 58—70 "41105 9m wos IH VAYVOOS sg Јо 39409 TEAM ; i : Ponts сты енды eei etia s St ee re Pea erect e Е пем SIS rer сар MR nos RRO ario) a St ‘TL Td ТАХ Лол Saro uray YENI oO АНАМДЫ —"L wo 6E OM 0 8 9 MEMOIRS OF THE GEOLOGICAL SURVEY OF INDIA. Tug UPPER GONDWÁNAS AND OTHER FORMATIONS OF THE COASTAL REGION or THE GODAVARI District. By Wıruıam Kine, B.A., Deputy Superintendent (Madras), Geological Survey of India. CHAPTER I.—GENERAL DESCRIPTION. Introductory.—The great tract of Gondwána rocks occupying the lower half of the Godávari valley is very conveniently and advantageously divisible into two areas, which are separable, physically, by that portion of the Eastern Gháts sometimes called the Golgonda range of mountains, and, geologically, in that the upper division of this formation is in great part of marine origin to the south of the hills while there are only river and lacustrine members of the series in the upper part of the valley. ' The portion below the Ghäts, or, in other words, the coastal region, also corresponds to what is called, in local parlance, the lower division of the Godávari distriet; and this is the area treated of in the present Memoir. The prineipal formation to be described is thus a part, though in many respects a very important part, of its development on the Coro- mandel or Eastern Coast in the patches already examined or described! 1 See Parts 1 and 2 of this Volume, and Vols. IV, Parts 1 and 2, X, Part 1, Memoirs Geol Surv. of India; also Records, Geol, Surv. of India, Vols. III, p. 11, XI, p. 249, XII, p. 187. ( 195 ) Memoirs of the Geological Survey of India, Vol. XVI, Art. 3. 3 KING: COASTAL REGION OF THE GODAVARI DISTRICT. by the Survey, as extending thus far northwards from the Trichinopoly distriet, through Sriperumbudar (Sripermatur), the Nellore area, and so up to the Kistna. The other formations to be noticed were nearly all known, more or less, long before the examination of the country was taken up by the Survey; this region having been rendered classie ground by several of the pioneers of Indian geology, more especially through the researches of the late Revd. Mr. Hislop into the relations of the inter-trap- pean fossil beds in the neighbourhood of Räjahmundry. It was, how- ever, reserved for my colleague, Mr. W. T. Blanford, to trace the Gondwána series down the valley of the great river from the Central Provinces in the year 1871; and in 1873 I had the good fortune to discriminate some groups of the upper division of this formation, as also some infra-trappean beds, to increase the roll already made out by unoffi- cial observers. Literature.—The work of these latter explorers commences as far back as 1814, when the ‘ Tracts, Historical and Statistical, on India’ of Dr. Benjamin Heyne appeared. These contain many references to this part of the peninsula which are extremely interesting, and they shall be duly noticed in the following pages, as they may happen to bear on the rocks or formations under description] In 1830, Н. Н. Voysey, Surgeon and Geologist to the Great Trigonometrical Survey, travelled over the western edge of this area, and his reports were subsequently collected and published in the Journal of the Bengal Asiatic Society. He refers to the diamond mines at Muléli, near Ellore, which even as far back as that time were not being worked. He also noted that the sandstone tract, now known as of Gondwana age, was traceable all down the Pránhita-Godávanri valley into the Ellore country. In 1835, when the 1 Tract III, p. 92, e£ seg.—Account of the diamond mines of India. Tract XII, р. 212—Account of the method of smelting iron in the Northern Circars. Tract XIII, p. 224—Account of the iron works at Ramanakapetta. Tract XV, p. 230—Cursory observations made during a tour from Bezwáda to Timmericotta. Tract XVII, p. 247 — Observations made on а tour from Samulcotah to Hyderabad. Tract XVIII, p. 280—A brief account of the Circars on the Coast of Orissa. ? Vol. III, pp. 298 and 392 (1833). ( 196 ) e GENERAT, DESCRIPTION. э) then Governor of Madras was making a tour in the Northern Cirears the medical officer, Dr. P. M. Benza, took the opportunity of looking up points of geological interest which were published in the Madras Journal of Literature and Science; and in these is, I think, the first pub- lished notice of the now well-known occurrence of traps with inter- calated and subjacent fossiliferous beds in the neighbourhood of Rájah- mundry. The peculiar character and constitution of the gneiss in the neighbourhood of Bezváda, and thence along the west to Vizagapatam, were first noticed by the same writer, In 1837, Mr. J. G. Maleolmson read a paper before the Geological Society of London,’ wherein he refers, incidentally, to some of the rocks of the lower Godavari region. Captain Newbold passed along the western edge of this area, some four or five years later, in one of his tours across the peninsula, when he made some further observations on the rocks of Bezväda and the old diamond mines at Malél, which were embodied in one of his many “notes.” Mr. Walter Elliott, M.C.S., then stationed at Rájahmun- dry, worked at the inter-trappean beds of Kátéru in 1850; and in 1854 he sent a collection of the fossils obtained from that place, with notes, to the Asiatie Society of Bengal. In 1855, these Kátéru beds and those of Pungadi on the right bank of the Godavari were again brought to notice by the Revd. Messrs. S. Hislop and R. Hunter, and again, in 1860, a further paper? was contributed by the previous gentleman which deals more closely with the same rocks. Mr. Hislop never person- ally visited these localities, so that he had to depend on the old ob- servations of Dr. benza, and later information obtained for him by Lieutenant Stoddard. Colonel Stoddard has since informed me that 1 Notes, chiefly geological, of a journey through the Northern Circars in the year 1835, Vol. V, 1837, pp. 43—70. ? On the fossils of the eastern portion of the Great Basaltie District of India: Trans. 2nd Ser., V, p. 837. 3 Journ. Ав. Soc., Bengal, Vol. XIII, 1844, рр. 984—1004. ^ Journ. As. Soc., Bengal, Vol. XXIII, p. 397. 5 On the geology and fossils of the neighbourhood of Nágpur, Central India—Quar. Journ. Geol. Soc., Lond., Vol. XI, p. 365. $ On the tertiary deposits associated with trap rock in the East Indies—Journ. Geol, Soc, Lond., Vol, XVI, 1860, pp. 154—166. (О ТИ) 4 KING : COASTAL REGION OF THE GODAVARI DISTRICT. he had no technical knowledge which could have enabled him to give reliable details, and Benza’s observations were in some respects inaccurate : thus the relations of the beds are not correctly stated, and some erroneous conclusions were drawn therefrom which shall be noticed in their proper place. Topography.—The country under description has somewhat a lozenge- shaped outline, with its longer diagonal striking east-north-east from the neighbourhood of Bezväda, on the Kistna, to a point on the sea-coast, about 45 miles south-west of Vizagapatam. From this point the shore forms one edge, running still south-west to about the 16° 20' parallel of north latitude, which parallel, again, is the southern edge of the field. The two other sides, namely, those on the north and north- west, meet in that part of the Eastern Ghats culminating at over 2,400 feet in the Kaurkonda and Pápakonda range, which is here crossed by the grand gorge of the Godavari. The larger and central town is Rájahmundry;'next comes Coconáda! the sea-port and zillah station, and the minor towns are Bezváda and Ellore. Physical geography.—The country is mainly low-lying and alluvial, more than half of it consisting of the deltaic deposits of the Godavari and a portion of those of the Kistna, and this is to the north-east joined on to a wide stretch of the alluviums of the minor streams on the edge of the Vizagapatam district. From this flat there is then a gra- dual rise which is broken by many small hills and groups of these, more especially in the Rampa country, until the greater masses of the Kaur- konda-Pápakonda range are reached. The uniform level of the great alluvial plain is broken by a large out- lier of gently rising ground, an island as it were in lands. : : 1 1 Deltulands the alluvium, at either end of which lie the towns 1 T have accented the name here, though it is not marked so in the official list, the proper name being Kékindda (the place of crows). Of course, it must be often rather diffi- cult for strangers to tell how these Indian names should be pronounced ; and a story to this effect is told of a child of a former collector of the place, who was rather taken aback on entering school in England at finding himself corrected in his pronunciation of the name of the town in which he was born, which the master improved into Cócánada, as though it were related to our possession in America. ( 198 ) GENERAL DESCRIPTION. 5 of Rajahmundry and Samalköt (Samuleottah); and again, but in an ex- tremely smaller way, there are a few elevations in the north-east plain, as near Pithápuram and Ayaparáz-Kotapili, or still nearer to the rising land, the very pieturesque group of tolerably high gneiss bosses in the alluvial bay of Lingumpurra. At the Bezváda end there are also a few similar alluvium surrounded hilly masses, two of which, on either bank of the Kistna, have been joined together by th» great dam there built across the river and giving a head to the splendid system of eanals distributed over its delta. Indeed, at the remote period when this great alluvial plain was occupied by the sea, the then Godavari coast must have been eminently varied in its contour, from the Bezváda islands in the Kistna bay past the large island on which the pleasant bungalows above Dowlaishweram now stand, and so by the many curving-bays of Kirlum- pádi, Túni, &e., right up to the Dolphin’s Nose and numerous other fine hilly masses in the neighbourhood of Vizagapatam : a very different style of scenery to that now exhibited along the low and uninteresting shores of the Coromandel. From this old coast line the country in the middle area rises very ; gradually by long-backed sandstone hills which Low country and hills. Ern 4 drop down again in low parallel escarpments facing north-westward, and beyond these there 1s further low-lying but broken ground, marked more partieularly by the Chintalpüdi hills. In the Bezvada area the country is still very flat, but there are several low hills and ridges rising out of it such as the groups around Augerpali and the larger masses of Vellatür and Kondapili. Over the north-eastern country, low hillocks rise immediately from the edge of the alluvial plain having a long and quite even slope to the south-east and a much steeper fall to the north-west; and these are succeeded by rather flat- topped ridges, rapidly increasing in number and height towards the more hilly Rampa country. The important local rivers are the Tammilér, flowing past Ellore into the Kolérlake; the Yera Kalwa, draining much of the country below the Kaurkonda or Bison ( 199 ) Rivers. 6 KING: COASTAL REGION OF THE GODAVARI DISTRICT. Range ; and the Eléru, Golgonda, and Pandáru rivers in the Rampa and Vizagapatam area; but the country is eut right across by the waters of the great Godávari river which have been collected behind and far to the north-westward of the Eastern Gháts through which the famous gorge or defile passes (See plates 1 and 2). The mountain passage of this river is a deep and tortuous trench, emer with a V-shaped section of about 129 miles e Godávari gorge. д : i б 3 Pi in length, of which 4 miles in the middle may be said to be the proper gorge, the river being here about 300 yards wide,! while the mountain spurs rise more abruptly to a height of over 2,000 feet. Here the river is more like a loch or fiord lymg among lofty mountains, and it is often difficult, as one is floating along, to guess where any outlet can be, specially when the lofty flat-topped * Bison НІШ 2 is in front (See Plate 2). But the wonder of this river- pass is, as to how and why it should have been cut down through this 2,000 feet high range of crystalline rocks, when to all appearance the river might have pursued its course through the more easily-worn sandstone of the low water shed? to the south-west, near Ashwaraopet where the great gap (erossed by the Kistna) in the continuity of the Eastern Gháts commences. 1 Mr, Vanstaveren, lately resident engineer on the Upper Gódávari, informs me that the narrowest part of the gorge is 832 feet across; the deepest sounding taken by him was 121 feet, with a bottom of soft blackish clay, the shallowest parts being 63 feet. The bottom towards the sides is very uneven and rocky, but the mid-channel is pretty even. 2 Another name for the Kaurkonda hill, overlooking the right bank of the river, said to be over 2,520 feet in height, and one of the few spots in this part of the Presidency still frequented by bison. An attempt was made several years ago to utilise this plateau as a sanitarium for the men employed on the works at Dowlaishweram, but it was found to be feverish, and the supply of water was poor. 3 I have not been able to obtain the level of this water-shed above the sea, but it can scarcely be more than 200 feet. The country is wonderfully flat at the shed and away down on either side of it; in fact, it appeared to me that the height can hardly be so much as this. At any rate, there is a tradition (told me by Mr. Vanstaveren of Damagüdem on the Godávari, formerly Executive Engineer at that place) among the people that the Godávari was very many years ago so ponded up behind its gorge in a great flood, that its waters actually poured over this water-shed! This appeared so amazing a phenomenon that we could hardly believe it. However, subsequently when working at the Beddadanol coal-field ( 200 ) E 3 OL Q ETC A x G Wi VE Ww Q Im IMN DUA omoırs Vol: XVL P1:2. EEE IAE i Schaumburg, L ithd Narrows of GODAVARI GORGE - (Locking South). 3 mr VET N 45; nt GENERAL DESCRIPTION. “7 Of course, the obvious view of the case is, that the Ghäts and the great valley of Bhadrächalam behind them have all grown with the denudation of this part of the river, which commenced to eat its way into the earth as the upper Its history. surface (or thereabouts) of the Päpakonda range rose out of the sea; but still there is the great diffieulty that the field of harder rocks should have been attacked rather than the sandstone area to the south-west. A reference to any good map of India will show that the Godávari for some 30 miles before reaching the gorge, and in the neighbour- hood of its passage from the sandstones to the gneiss area, follows a remarkable bend nearly at right angles to the generally south-south-east course it has pursued for over 100 miles between Sironcha and Bhadrá- chalam. At the same time, this area, or the mountain region of Pápa- konda, and the lowering hills between it and the alluvial flat of the coast, exhibit plateau and bevelled surfaces which are the remains of an old on the south side of the shed, Mr. Vanstaveren made further enquiries among the oldest in- habitants of the neighbourhood, when he obtained confirmatory evidence. It appears that there have been three well-ascertained great floodings of the Godávari, namely, in the years 1818, 1849, and 1861, the earliest of which was tremendously high, and 20,000 people are said to have been carried off by it on the Nizam's side of the country. Mr. Vanstaveren asked a man incidentally if he could point out how far the waters of 1861 approached within the neighbourhood of the water-shed; the mam pointed out a village not far from the northern edge of the divide. On this, Mr. Vanstaveren is inclined to think that it is quite possible that the greater flood of 1818 must have really overflowed and carried the Godávari water into the valley of the Yerakalwa. Mr. Vanstaveren, who has perhaps had more practical experience of the Godavari river than most men, says that the water rises in the gorge at least 100 feet in heavy flood time. Colonel Beatty, the District Engineer in 1878, in answer to some enquiries I made regarding the river, says, “the depth of the river in the gorge has never, so far as I know, been carefully recorded. In summer, when there is no perceptible current, the depth is supposed to be about 80 feet, and in maximum floods it is supposed the water rises 100 feet over summer level, so that there would then be about 180 feet at the gorge." Hence on my supposition of the height of the Ashwaraopet divide, the water would have to rise in the gorge at least 70 feet higher before it could possibly flow into the Yerakalwa valley, However, this is little more than guess-work, and therefore the tradition must be let stand for what it is worth until the height of the water-shed is obtained. The people are doubtless given to exaggeration in many things; for instance, they might have a tradition of a great flood that overwhelmed a countless number of people, but they would hardly take the trouble to invent a story of such a phenomenon as that those waters flowed over a particular water-shed. ( 201 ) 8 KING : COASTAL REGION OF THE GODAVARI DISTRICT. marine plain now having a gentle slope of two or three degrees to the south-east, the elevation of which plain above the sea would appear to have commenced at the end of the jurassie period. I would venture to suggest that this trending of the river so mueh more to the eastward may be attributed to an elevation of the land with an initial slope such as is presented by the present surface configuration of the hilly country; while concurrently with this elevation, the denud- ing power of the river would be gradually directed to and kept working in the harder crystalline rocks. This abrupt swerve in the river's course was more particularly brought to my notice by the late Dr. Oldham in connection with corres- ponding deviations from their general or average courses in the Kistna near Kurnool, the Penner near Cuddapah, and the Cauvery at Kerür, which he seemed inclined to think indieated a period of physieal change of considerable importance in the geological history of the peninsula. Further study of these rivers and of the rocks traversed by them may help to develop this generalisation ; but so far the course of the Goda- - vari below Bhadrächalam does appear to be one of the marks of a middle mesozoie period of great change in the eastern coast of India. The alluvial banks on either side of the gorge between the hill Delta branches of Go. Spurs, are left as a well-marked terrace fully 70 or davari. 80 feet over the level of the dry-weather waters. Below it, or near Polawaram, the alluvium begins to spread out widely on either side, the flood waters being now kept back by artificial banks all the way down to the commencement of the delta at Dowlaishweram, a few miles below which town the river bifureates, the Gowtámi branch going off towards the northern mouth at Hope Island, while the Vasista branch flows in the direction of Bendamürlanka and again bifurcates, giving one outlet near the latter place and the other and larger near Narsapur. The delta waters have now for many years been distributed by a magnificent system of navigable and irrigation System of irrigation. ; : canals over the great alluvial plain, and are even GENERAL DESCRIPTION. 9 in eommunieation with a further great system on the Kistna delta, the junetion being at Ellore, which again within the last year or two has been connected with the coastal eanal from Madras. The Godävari system has been effected by a great dam or ‘anicut’ which is built across the river at Dowlaishweram, at either end of which and in the middle are locks leading the waters into what are called the eastern, western, and central deltas, the latter being between the two branches of the river. At flood-time the anieut is completely covered, the only evidence of its existence being a broad wave across the river's course. The height of a maximum flood at Dowlaishweram is about 53 feet above mean sea level, and at the gorge it is supposed to be about 150 feet; so that there is a fall of about 97 feet from here to the anicut, that is, in a distance of about 45 miles. The current of the river at that time through the hills is so great that the most powerful steamers of the Godavari works fleet cannot make any head against 16.1 The great alluvial plain of the coast is broken by a large surface l of inland brackish water called the Kolér lake, The Kolér lake. : é ор с А : into which the Tammiler river empties itself. This lake 1s very shallow, and is situated about midway between the Godávari and the Kistna; and it seems to be obviously a hollow which has been left between the gradually growing deltas of the two rivers which has not as yet been filled up by the deposits of the smaller river, though this has undoubtedly spread out a delta of its own in the bay between the long low uplands behind Ellore and thus helped to give the even north-west shore line. The lake і in tidal communication? with the sea in the dry weather by the Upputéru, a stream at its eastern end 1Mr. Vanstaveren, who has occasionally made the voyage on the river at these flood times, tells me that in the gorge the surface of the river has then a very hollow section and that the voyager races away as it were in a trench of water; and that just below the pass there is all the sensation of going down a slope, or working up one as the case may be. The great flat-bottomed stern-paddle steamers are often tried up this visible slope before the current diminishes sufficiently to allow them to proceed. 2 Colonel Beatty writes: “ When I visited the head of the outlet in May 1876 it was open, and the water in the mouth (which was some 100 or 150 feet broad) rose and feli about 3 feet with every tide. ( 203 ) 10 KING : COASTAL REGION OF THE GODAVARI DISTRICT. near Akíd ; but when the Tammilér is in flood, there is then a current out- wards. As the dry weather comes on this outlet is closed and as much water as possible is retained for irrigation, &e., along the shores. Occa- sionally, as in the year 1875, the Tammilér comes down with great force and floods the country all round Ellore. In that year the banks of the canal, whieh is here below surface level, had to be cut to allow of the water being carried off partly by it, the river waterway under the aqueduet not being sufficient for the off-flow. Such are the more evident physieal features of the country, but a further remarkable one, and not so patent to the casual observer, is a certain bevelling off, or truncation, of many of the hills and ridges into pseudo-plateaux from the smaller elevations along the edge of the. alluvial plain even up to the loftier masses of the Eastern Gháts at the river gorge; all which bevelled surfaces appear to have been portions of one great and ancient plain which has since been denuded Traces of an ola and cut into, in fact, the remnants of a deeply marine floor. denuded old marine floor. In the Rájahmundry- . , Vizagapatam area, the hills have all very much the same gentle and even sloping surfaces hading up out of the alluvial flats, while the hills and ridges a little further inland have their summits bevelled off in what appears to be the same plane. This is a marked feature of the hills about Kirlumpudi and Bendapudi, and I noted it at the time of my survey as a strange one. Subsequently, in working over the country to the north of Rájahmundry, about Korakonda and Nägumpalı, I was again struck by the curious planed-off aspect of most of the hills, their tops all appearing to lie in a plane of gentle south-south-east slope which should meet the flat-topped Bison hill in the Ghäts. The same feature is seen, though less vividly, in the Yernagüdem and Ellore eountry, from the low hills near Gütálla to the higher group near Chintalpüdı. Again, towards Bezváda, in the numerous hills about Nüzvid, Augerpali, Nüna Stalum, and Bezváda itself, there is the same lie of the hill and ridge tops in a generally south-easterly sloping plane, displayed in a remarkably clear manner. ( 204 ) GENERAL DESCRIPTION. 11 This old marine floor is, as shall be shown later on, made up of HAS gneiss, and, in one portion of the field, lower Gondwána rocks (presumably ranging up into triassic times); while upper Gondwana strata (of ascertained jurassic age) are lying evenly over both of these. It is therefore clear that this floor is of pre-jurassie age; while the history of this portion of the Eastern Gháts and the great river defile through them can only be con- sidered to have commenced after this period. The different groups of rocks occurring in this field are given in the following list, in which they are ranged as closely as possible with the groups of the survey classification, and approximately so with the like sub-divisions of the European formations. The crystalline rocks, of which there is a large area, and the sandstones of the lower Gondwäna series have not yet been examined in sufficient detail to admit of final deseription ; hence List of formations. they are only treated of cursorily in this Memoir, and as the floor-rocks on which the other formations more partieularly described are laid down. Я Groups of Survey Approximate position in Local groups. classification. European classification. Alluvium, Recent. Rajahmundry sandstones. Cuddalore sandstones. ? Middle Eocene. Pangadi and Katéru Traps and ) œ ? Lower Deccan Traps. Upper Cretaceous. Inter-trappean Limestone (es- E 1 : tuarine). = 8 88 Düdkür Infra-trappean Beds, \ 3 ^ (marine). а ? Lametas. Middle Cretaceous. Tripati sandstones (marine). 4 ( Umia Beds of Cuteh. — Uppermost Jurassic. = Ши 2 2 Е m~a J? Chari Beds of Cutch, Upper Jurassic (Ox- fgacapunom etes e ? Jabalpur: fordian or Callovia). zi А Gollapili sandstones. 5 Rajmahal. Middle Jurassic (Ba- thonian). E : ER 1 Chintalpüdi sandstones. E E Kämthi, Damida. ? UPPER PALXOZOIC, Ha о D Bezváda gneiss. Gneisses of the main or AZOIC, CRYSTALLINE, eastern region of the or METAMORPHIC peninsular area. BERIES. ( 205 ) 18 KING : COASTAL REGION OF THE GODAVARI DISTRICT. CHAPTER IL—GNEISS AND LOWER GONDWANA ROCKS. The Bezváda gneiss.—Notwithstanding the more rapid and general survey of the area of crystalline rocks, enough was observed of them to justify the recognition of a partieular variety or sub-division which other observers besides myself had noticed, more especially in the neighbourhood of Bezvada and on the Vizagapatam coast. It might be called murchisonite gneiss from the characteristic form of felspar entering so largely into its composition, which mineral was, however, eventually recognised by my colleagne, F. R. Mallet, in the specimens which were sent up to him for determination at Caleutta. Dr. Heyne noticed the garnetiferous and felspathie constitution of Previous ESO 12156 the rock and its tendency to weather of a dark of— or even nearly black colour ; writing of it: “ Veins of felspar often run through this rock in oblique or horizontal direction. Such veins are much harder than the felspar, which enters as a constitu- ent into the rock. * * * Thefelspar is white, foliated, and appears, when in large pieces, transversely striated. It is uncommonly soft, and is entirely disintegrated when long exposed to the air.”! Dr. P. M. Benza describes, but not in much detail, the garnetic gneiss of Bezväda, and he saw it again at Tüni and other places in the Vizagapatam district ; the felspar is specialised as albite or cleavlandite, and its beautiful scarlet red colour is noticed. He also refers to the easily-weathered character of the rock. Captain Newbold wrote—“ The gneiss comprising the ridge of Bezwáda is garnetiferous, cleavlandite often replaces the common . felspar, and renders the gneiss liable to decay. It contains large veins. of quartz and is intersected by green-stone dykes, the presence of which may serve to account for the distortion observable in its strata.” A decided band of this garnetiferous felspathie gneiss edges the Extent and relation to alluvium and older aqueous rocks right across the other gneiss. field, from the Kistna district into that of Vizaga- 1 Op. cit. p. Tract XV, p. 230. ? Op. cit. p. 45. ( 206 ) e GNEISS. 13 patam. The Bezväda ridges and outlying hills are almost entirely composed of it, the only other strata being a few bands of quartz-schists and caleareous rocks, and thence to the north-east, in the Nina Stalum country, it has a width of some 16 miles and so continues occasionally narrowing or widening out into the Golgonda country. Alongside of 1t, on the north, there is the less schistose, or even more massive band of gneisses, stretching from, the Kondapilli hills, which answers more or less to those of the Nellore and Kistna areas.! Indeed, so distinguishable is this form of gneiss in the area pointed out, that I gradually came to call it by the name here adopted ; and the area occupied by it is so great that it seems quite worthy of being considered as a fair group among the gneisses of the main or eastern area of the peninsula. І am as yet unable to draw a clear line of demareation between it and the gneiss on its northern side, but the tentative boundary given in the map shows its general limits. The dip of the foliation, or, as it is to all appearance. lamination, is generally high and to the south-east, though folding and contortion are frequent; and so far its later age over that of the Kondapilli gneiss, as evidenced by its less highly metamorphosed condition, 1s borne out. With the exception of the thin subordinate bands of quartz-schist and й quartzose gneiss, the usual rock of this band Lithology. A 5 1s generally of а dark brownish-red colour composed mainly of a bright, lustrous, well-cleaved, and occasionally foliated red felspar.? Itis rough and granular, but well foliated or more or less schistose 1 See Parts 1 and 2 of this Volume of the Memoirs. ? Mallet writes—“ Your mineral is murchisonite, a variety of orthoclase, the distin- guishing character of which is the presence of an abnormal cleavage, making (in the original mineral from Dawlish, in Devonshire) an angle of 106" 50’ with the basal cleavage and 90° with the clino-diagonal cleavage. These anglesin your mineral are about 104° and 90°, or very near 90°. The lustre of Murchisonite on this abnormal cleavage is pearly, as in your specimens. The latter in fusibility, &c., have the characters of orthoclase. The hardness of the original murchisonite is rather less than that of felspar; that of your mineral is about 5:0. This low degree of hardness and the presence of free peroxide of iron led me to think that the mineral was in an altered state, but its translucency when looked nt parallel to the pearly cleavage does not support this. It may have caught up the oxide when originally formed." ( 207 ) 14 KING: COASTAL REGION OF THE GODAVARI DISTRICT. or even fibrous and then somewhat silky, though it is never quite a schist, or again tolerably massive. Sometimes the felspar predominates to such an extent that there are seams and even thick beds of what might be called a felspar rock, the murchisonite being then massive and granular. At other times, the rock is more like a granite with the felspar in largish crystalline masses; but usually when granitoid, it is a coarse granular aggregate of felspar, less quartz, and a little mica. Garnets are very frequently distributed through it, often to such an extent that it may be called a garnetiferous gneiss, as at Bezváda where the rock is often crowded with small crystals of bright red and purple colours which are only wanting in size to render them beautiful and valuable stones. Here also, and in the Augurpali country, there is a good deal of graphite thinly scattered through the rock, giving at times graphite schists or massive graphitic rock with the graphite in minute scales. The felspar is generally reddish or a pale salmon colour weathering lighter, but it is frequently of a decided red, even rosy-red, and then, on well worn and smoothed surfaces it has somewhat the look of rhodonite . while it has nearly always a fine pearly silvery or bright bronze sheen. When weathered, the gneiss often presents, particularly in the Vizaga- patam country, the most startling imitations of ferruginous sandstone, the garnets being so erowded together that there isa difficulty in recog- nising them as separate masses after they are decomposed. Even long before I took up work in this district, and at the time I was re-arranging the specimens in the Madras museum, my attention had been drawn to specimens from Bezwáda which were all labelled *sandstones, and yet were obviously part and parcel of other fragments from the same quarries which were unweathered garnetiferous gneiss. Kämthi sandstones. —The remaining rocks of the floor are brown ferru- ginous, and variegated felspathic sandstones of this group of the lower Gondwana formation, which are strongly developed in the hill country to the east of Chintalpádi whence they stretched away northwards into the eastern portion of the Nizam’s dominions bordering the right bank of the Godavari river. ( 208 ) LOWER GONDWANAS. 15 Mr. W. T. Blanford first identified these beds 1n this district, having Traced down from traced. them down from their typical area at Central Provinces. K4mthi in the Central Provinces, and he describes them as “ frequently variegated in a peculiar and characteristic manner. They are associated with numerous hard bands of ferruginous grit and compact red and yellow shale. In one instance sandstone was found with a peculiar semi-vitreous texture, which is very characteristic of some beds in Chanda and Berár. All these characters lead unmistakably to the conclusion that these rocks are the representatives of the Kámthi beds of Nagpur and Chanda? At the same time, nearly half a century before either he or I worked at these beds, it had been pointed out by Voysey? that sandstones were traceable all the way down the Godavari valley from the Central Provinees into the coastal region of Ellore. These sandstones occupy an old hollow in the gneiss, and are occa- Lie and association with sionally much more irregular and undulating in the gneiss. their lie than the beds of the upper division of the series overlying them, though, along the northern outcrop of the latter rocks, they lie easier and appear аб times to be almost conformable with them. Like the rest of the gneiss country, the ground occupied by them is now at places very broken and irregular, as in the Chin- talpudi hills, where the curved and gradually assumed north-west strike, nigher and undulating dip, and great thickness of varied beds, give very different contours and slopes to those of the low long-backed ranges with north-western scarps of little height of the gently dipping strata of the upper Gondwana beds bordering the deltaic plains. It will be shown further on that the belt of these latter rocks lies evenly over these sandstones, and is so continued on to the gneiss to the east and west of them ; or, in other words, that they are lying on a tolerably even plane of both crystallines and Kamthi sandstones. The Chintal- padi hills do not, as far as I remember, show any of the unmistakable 1 Rec. Geol. Surv. of India IV, p. 49, e£ seg., 1871. 2 Jour. As. Soc., Bengal, Vol. II., 1833, p. 400. ( 20) >) 16 KING: COASTAL REGION OF THE GODAVARI DISTRICT. flatly-denuded surfaces so characteristic of many of the hills in the gneiss country; but these could hardly be expected to have been pre- served among such easily worn rocks, even if, which is barely possible, any of the old upper surface remain in these hills. The fossil plants determining the age of these beds were obtained by Fossils and their local. Blanford at Somavaram, south of Nuzvid, and at ne Kunlacheru, 16 miles north of Ellore; fronds of Glossopteris at the former locality and Vertebraria at the latter. I sub- sequently obtained Vertebraria also at Somávaram, and other vegetable remains from fine pale dove-coloured compact shales a short distance south-east of the village at the bottom of a low scarped hill below soft coarse felspathie sandstones. On the west side and to the south of the village there are many wells, the debris from which consist of soft white and yellow clayey sandstones very coarse-grained, and in these were many fine specimens of G/ossopteris, &c. The number of fossils from the Kunlacheru locality was increased by finer specimens of Vertebraria, Glossopteris, and Phylotheca. Frag- ments of the first plant occur close to the village in a thick bed of compact splintery clayey sandstone, but the proper fossil locality is at the base of the ‘ Gut,’ a small conical hill about 100 feet high to the west of the village. On the east side of this hill a tolerably fair section is exposed, all the rest of the slopes being covered with thick jungle. At the base, about 10 feet of very coarse open-textured felspathie У sandstones succeeded by 10 feet more of hard, very Association of fossils. =, slightly calcareous fine clayey sandstones with fossils, over which come 20 feet of coarse beds like those at the bottom. Next come 6 feet of splintery silicious beds, overlaid by 40 feet of the coarse felspathic beds: all topped by a capping of nearly 20 feet of hard semi-vitreous ferruginous purple and brown sandstones and pebble beds belonging to the Upper Gondwánas. The beds are dipping about 5° east- wards. The fossiliferous band near the base is thus: a. Glossopteris seam „ . . . . 3 . . . ` 2 feet b. Fertebraria ss 2 ; 5 : > о 1 5 : 5 2; 3 c. Splintery caleareous clayey bed . с o E : o c c 645 CI) UPPER GONDWANAS. 17 The Glossopteris seam is a fine sandy and clayey fawn-coloured rock, laminated and much easier broken up than the seam below. The Perte- órarie are matted together but not very thickly, in the upper portion of the seam (а) which is very hard and splintery. Indeed, I had to leave some very fine examples of branched Vertedrarie which could not be got out in any size without the rock splitting up into small fragments. The suggestion, I think, of Dr. Oldham, that the Vertebraria may be the root of some other plant, seems to gain some confirmation here through this association of them in this matted condition immediately below the Glossopteris bed. CHAPTER IIL—UPPER GONDWANAS. The representatives of this series form the greater width of the low rather flat ranges of hills, with north-westerly escarpments, to the north of Ellore, extending with an east-north-east strike from Maleli to the right bank of the Godavari, near Thállapádi. Beyond this break at the riverside no further traces of them are seen until the village of Jag- campet, on the high road from Rájahmundry to Vizagapatam, is reached, and then the series is only represented by a chain of outliers belonging to its higher group. In the Ellore area the series consists of three groups, two of sand- stones and an intermediate one of shales, which Е are exposed with more or less clearness on the slopes and scarps of the hills, the lower group forming a sort of foot or toe and the upper a low scarp and capping which slopes very gently down towards the alluvial plains. The groups are thus in tolerably conformable lie, with a low dip of from 5° to 10° or 15° south-eastward. The lower group is nearly continuous in its strike from Maléli, 16 miles west of Ellore, to within 5 miles of the sociation. Б с Extent and associati Со аа bank; the middle one has only about a B (rer) 18 KING: COASTAL REGION OF THE GODAVARI DISTRICT. third of this length of outerop in the middle of the field ; and this is again overlapped by the higher one, which extends from the Tammiler or Flore river to within a short distance of the Godavari. The latter is also, as far as сап be made out, continued on the Rájahmundry-Vizaga- patam side of the country, but without any association with other sand- stones, where it lies directly on the gneiss. The three sub-divisions are fossiliferous at rare intervals; the lower AI only showing vegetable remains, whereas the other two give marine animals, one of them also having a few associated plants. The Gollapili sandstones, with their typically Rájmahál flora, have given one of the most definite geologieal horizons on the Coromandel, while the Rägavapuram shales and Tripati sandstones are brought into correlation by their marine fossils, the former with the previously-known fossiliferous patches of shales further south in the Nellore-Kistna area and at Sriperumbüdür (Sripermatur) near Madras, and the latter with the Umia beds of Cutch. Gollapili Sandstones —About 12 miles west of Ellore, the large village of Gollapili stands on a series of dark-brown and yellowish- red sandstones grits and conglomerates which have yielded numerous plant remains of Rájmahál age. At the northern edge of the field, near Sómavaram, these sandstones are lying on, and overlap, other coarser but softer and more vari-coloured felspathie sandstones and sandy shales of Kamthi age, with the characteristic glossopteris and vertebraria, The fossils of these Rájmahál beds were found only at afew localities late in the progress of the survey ; and as there is Paleontology. often a very close resemblance between these sand- stones and those of the Kámthis, it was thus some time before I was able to separate this group from the larger area of rocks which W. T. Blanford had approximately sketched in as of the latter series during his prelimi- nary traverse between Nüzvid and Ellore. The complete collection was finally determined by Dr. Feistmantel, the paleontologist of the Survey, 1 Records, Geol. Surv. of India, Vol. IV, p. 49; V., p. 23. (‚0320 UPPER GONDWANAS. 19 and subsequently described by him in the Palzontologia Indica," from | which the following list of fossils is extracted :— FILICES. Alethopteris indica, O. M. Pecopteris macrocarpa, O. M. Angiopteridium ensis, Schimp. WE gr MOL qf Angiopteridium spathulatum, Schimp. CYCADEACEER. a. ZAMIR. .. Ptilophyllum acutifolium, Morr. Ditto cutchense, Morr. 5 6 7. Dictyozamites indicus, Fstm. 8. Pterophyllum morrisianum, Oldh. 9 5 Ditto carterianum, Oldh. 10. Ditto kingianum, , Е stm. 11. Ditto distans, Morr. 12. Williamsonia gigas, Carr. CONIFERZE. 13. Palissya conferta, О. M. sp. 14. Ditto indica, O. M. sp. 15. Cheirolepis comp. munsteri, Schimp. 16. Araucarites macropterus, Fstm. With regard to this flora, Feistmantel says: “ Taking the flora of the Rájmahál group in the Räjmahäl hills as it has been partly described by Messrs, Oldham and Morris, and lately concluded by myself (Pal. Indica, 1876, contin., Pl. 36-48, pp. 53-110) as typical of that group, we find the flora of Gollapili to be a pure representative of it on the south-east- ern coast of India." <“ This local flora contains * * * most of the typical Rájmahá plants, the manner of preservation only being different. While in the Rájmahál hills, the rock containing the fossils is almost throughou of very hard consistence, and mostly of pale colour, the plants nea 1 Flora of the Gondwäna System, Vol. I, p. 163 e£ seq. 20 KING: COASTAL REGION OF THE GODAVARI DISTRICT, Gollapili are preserved in a fine-grained sandstone of red-brown colour ; this rock is totally different from the former one, and yet belongs to the same group and includes the same fossils." “The Filices are not very frequent, but all that occur bear exactly the Rájmahál character.” * As well as in the Rájmahál group of the Räjmahäl hills, also here at Góllapili, the Cycadeacee are very well represented, except the genus Cycadites ; but we find frequent Рі орлу Шит and Pterophyllum; Dic- tyozamites, Oldh., also occurs," “Remains of coniferous plants were found also pretty abundantly near Gollapili, as is generally the case in the upper portion of the Gond- wana system.” The plant remains were nearly all found near Ravacherla and Bura- Fossil localities and Vancha, 6 miles west of Gollapili, in softish brown dye OR: and purple sandstones, there forming three low and south-eastward sloping terraces with low scarps to the north-west- ward. These beds are very near the bottom of the series—in fact just above a thin set of bottom beds, the village of Ravacherla itself being on the garnetiferous or Bezváda gneiss. Other, but very fragmentary, remains were found further north, around Müsunür, in a set of soft micaceous shales and flags of much finer texture than the Ravacherla beds, which are even lower in the series, there having been a thickening out of the strata towards Nüzvíd. Gollapili is on still higher beds, but it is the largest and most important village in the field; hence its name has been given to this group. The group is not of much thickness, seldom at its thickest more than 300 feet. At the Gollapili end, the beds are Thickness. : Я А . lying in a shallow basonal way, with a low rise to the north-westward ; and from this area they are continued to the east- north-east in a long outerop at the foot of the scarps, striking past Tand- kalpádi, Tripati, and Rágavapüram to Bimülá, some 5 miles from the right bank of the Godávari, where they die out altogether and are over- | lapped by the Tripati sandstones, C RIE у) UPPER GONDWANAS. 91 For nearly the whole outerop there is а tolerably persistent form of Persistent form of bottom beds, namely, a very hard, compact, some- bortom beds; times semi-vitreous or again jaspideous, red ferru- ginous clayey sandstone, grit, or conglomerate. The sandy and gritty particles are frequently of clear glassy quartz, and sometimes, when the rock is vitreous or jaspery, these are scattered through a kind of paste and give the rock a semi-igneous look like a dark-red quartz porphyry. This jaspery or vitreous form is well seen in the low scarps of the Nüzvíd area to the south of Sómavaram, or at Rämakapeta; or, again, 16 is common some 12 miles eastward of the Ellore area to the south of Kunlacheru there capping the two outlying hills, called ‘guts,’ and the scarps behind them. Further east of this, as below Tripati and the Rága- vapüram outlier, the group is represented by coarser dark-brown con- glomerates and grits, still hard and vitreous, but not so clayey or jaspery. At the Nüzvíd end, the western boundary between Ravacherla and i) Sómavaram is, in part, faulted or erushed against the gneiss; the beds being curved and sharply tilted up along this line, especially opposite Rämakapetta, while they are also much jaspidified and infiltrated with quartz. "The lie of the strata at Ravacherla is, however, quite easy, at least to within a short dis- tance of the gneiss, the contact being covered up by debris; butif there be no fault here, then the lower heavy conglomerates and hard jaspery clayey sandstones are overlapped by the higher beds containing the plant remains. The more reasonable supposition is, that faulting has extended along here, and that higher strata have thus been brought directly along- side the gneiss. At the same time, the boundary is not faulted for its whole length, some outcrops not being squeezed or altered in any way, for instance, it 1s quite natural just south of Nüzvíd, where the gneiss is covered by easy-lying beds. At a very short distance from the western edge, the strata become nearly flat again and dip easily under fine mica- Suecession of strata. E : j ceous sandstones, which occur in good force in the small group of flat-topped hills to the south of Akreddigádem. These (om cS 22 KING : COASTAL REGION OF THE GODAVARI DISTRICT. are from 80 to 100 feet high and are made up of thin, fine, soft mica- ceous sandstones and sandy flags of buff and purple colours, while about 20 feet at the summit is of thick-bedded, coarse, soft, micaceous, brown weathering sands. The Müsanár valley is altogether of the lower beds, which dip gently to the southward, under the Ravacherla beds, the lowest of which must correspond to those capping the hills just referred to. These are succeeded by about 50 feet of thinner, generally rather coarser and more variegated, sandstones, with a few yellow and red clays, never so micaceous as the Musanür beds, over which comes a thin band of hard brown sands, giving a low head-land and scarp near Pólsanpih. Then come some 30 feet of lighter-ccloured sandstones on which Gollapili is situated, succeeded by about 1(0 feet of massive thick-bedded, dark- brown micaceous sandstones in the long rounded slopes below the Düdü- gut scarps. The latter sandstones weather of a dark-brown purple colour, in smooth rounded humps which stand out in the jungle very like humps of gneiss. This succession may be put thus in tabular form :— „7. Massive brown micaceous beds below Düdügat scarp . . about 100 feet. e. Light-coloured variegated sandstones of Gollapili itself . о 55 GU) 5 d. Pólsanpili headland beds р H қ 5 5 ; o » ПО с. Softer and lighter-coloured sands ; and, lower, the brown fos- siliferous beds of Ravacherla Я ; ч 5 Д А 55 ЕЙ) 2; 6. Müsanür micaceous beds a $ А 2 А 4 : "E JB gg a. Bottom beds, indurated clayey sandstones, &c. 5 о 5 ер 20 ,, About 285 feet. These thicknesses vary a good deal; but the total here given is about the greatest thickness there can be of the whole group. This succession in the Nuzvid basin is more properly a local assem- blage of varied strata, for it does not extend, or is not represented, be- yond the Tammilér side, except by the bottom indurated clayey sand- stones (a), and by the massive micaceous beds (£), which are continued in the low jungle-covered hills north of Víjarai. (^ 7906: UPPER GONDWANAS. 98 These Víjarai hills show no eapping of other sandstones and conglo- Thinning out and de. merates as in the Dadtigut range, though to the nudation. south-east at Nayanapalem and near Tundkalpüdi there are again ` sandstones rising into very low scarps, between which and the jaspideous conglomerates and sandstones coming out from under the Víjarai strata is only a slight thickness of beds to repre- sent such a thickness as is in the Víjarai hills; hence it is quite clear that the Gollapili series must have thinned out, or have been much denuded, to the eastward of the Tammilér river. The bottom beds (а) hade out from under the sandstones of the Víjarai hills in the Kunlacheru valley, and there form a skin over the Kámthi sandstones of Kámarapükota, whence they continue to the east-north- eastward, essentially a set of dark-brown and reddish, hard, ferruginous sandstones, and heavy conglomerates and stony clay beds, now thickened out to some 50 or 60 feet, and quite unfossiliferous. It certainly seemed to me that there must have been great denuda- tion of the whole group, though no doubt there was also thinning out to the eastward, prior to the deposition of the succeeding group next to be described. The unconformity of the Gollapilis on the Kamthis is very evident Unconformable to Ou the whole, but i cannot point to any section Kamin. : showing this directly. Around Kunlacheru, the lie of both series is very flat, and the bedding of the very coarse friable felspathie sandstones of the older series is very obscure or much covered up by debris; still the hard, heavy, vitreous, and jaspery conglomerate, and claystones are lying evenly over a planed down floor of different sands, and not over a bed or band of beds of these. Towards Kámara- pákota this is more evident, and to the north and east the strike of the Kámthis begins to eurve round from north-east to the north-west run which it has in the Chintalpádi hills. A feature connected with the decided change from one to the other series along their boundaries is the remarkable difference in the quality of their debris. Of course, the detritus of both is for the most part sand ; but that of the Kämthis is (21 )) 94. KING: COASTAL REGION OF THE GODAVARI DISTRICT. remarkably harsh, coarse, and, if anything, lighter coloured, though there is a good deal of ferruginous matter in them also. І got so used to this difference in the feel of the sandy ground after passing and re-passing across the boundary for many days, that I eould note the change very soon as my feet fell on the harsher and coarser debris. The debris is also often gravelly, and much of the gravel is made up of small pieces of hard stony clay and buff cherty material which is not often met with in the upper Gondwánas. The ferruginous matter of these last is also I think, on the whole, more evenly distributed through the beds: in the Kámthis it is oftener distributed in seams, knotty and warty segregations, and surface infiltration. Rägavapııram shales —These and their fossils were first noticed in the side of a small flat-topped hill, a short distance east of Rágavapüram, about 28 miles west-by-north of Rajahmundry. They аге generally very fine-grained, rather unctuous, shaley clays and clayey shales of white buff and lilac colours, laminated, but not very easily split up in the planes of lamination, and breaking up easiest across the bedding in clunchy | sub-angular lumps having a rude eonchoidal fracture. The group does not, however, consist entirely of shales, but contains several seams, of more or less strength and per- Lithology. Н А sistence, of sandy beds at different levels, none of which are ever strong enough to take away its decidedly shaley facies. There are altogether about 100 feet of shales themselves at the thickest, the whole series never exceeding about 160 feet; white and buff towards the bottom, purple and buff in the upper half. Among the shales there are three or four thin beds of greenish-yellow sands, soft and friable but rather hard at the outerop, with brown ferruginous coating. These thin seams are each usually about 9 inches in thickness. The shales are much seamed with brown ferruginous matter in east-west joint planes, and in other minor fissures by infiltration. Some of the smaller fissures and the exposed surfaces of these are also coated with a bright sulphur-yellow ferruginous deposit. The outerop shows for about half the length of the run of the north- ( 918 ) UPPER GONDWANAS. 25 western slopes of the Tripati range of hills as a thin lenticular band Alenticular band be- for alength of some 16 miles, from the village tween the other groups. — of Dävanavärgüdem to within a short distance of Yádavol where it thins out, so that the over and underlying sandstone groups come into superposition. There has been a good deal of local slipping or faulting in the out- Local and recent dis. Crops near Rágavapürum, which, I think, are in TENG? dn OE great part of quite recent date. At first sight the succession of the beds here is rather obscure, especially in the small hill east of the village, where a well-marked but thin sandstone band occurs in several stepped outerops along the steep northern slope, but without any corresponding breaks in the coarse sandstones capping the plateau. It appeared to me that this series of step throws might be entirely due to occasional slips or slides of these same slopes which are very apt to give way owing to the clayey character of the rocks. At certain levels the series is crowded with marine shells of a few : ; genera, with which, however, are occasionally Marine organisms. н 2 associated rare vegetable remains. The commonest fossil is the cast of a Leda, which is nearly always much crushed and distorted ; 1t 1s specifically undeterminable. Dr. Feistmantel! has determined the plants and some of the animal remains ; and he writes—“ These beds overlie those of Gollapili, the flora of which has been also already described as the representative of the flora of the true Rájmahál group. The flora of the Rágavapüram shales is of a somewhat different Fossil plants. character, containing some forms common to the lower beds; and also some others, which are Jabalpur plants, or at least closely related to them." “The plants of the Rágavapüram shales are— Pecopteris reversa, u. sp. The same form was also found in the Sripermatur group (Sripermatur area). Angiopteridium (Teniopteris) spathulatum, Schimp. (McCl., sp.), Occurs in various ! Pal. Indica : Gondwana Flora, Vol. I, p. 191 et seq. ( 219 ) 26 KING: COASTAL REGION OF THE GODAVARI DISTRICT. forms in the Rájmahál group and in the Sripermatur group. Known also from the Gollapili beds. : Angiopteridium me’clellandi, Morr., sp. A Rájmahál plant. Pachypteris ellorensis, n. sp. Peculiar to the Rágavapüram shales. Pterophyllum, sp. ? Fragment of a leaflet. Podozamites lanceolatus, L. & H. Occurs in the Jabalpur group. Otozamites abbreviatus, Fstm. Is a Rájmahál plant. Ptilephyllum acutifolium, Morr. À common form in the Rájmahál group (also at Gollapili). Taxites tenerrimus, Fstm. A Jabalpur plant. Tawites planus, n. sp. Occurs in the Sripermatur and Nellore-Kistna areas, more numerous in the former. Gingko crassipes, Fstm. Occurs also in the Sripermatur group (Sripermatur area). Another species of this genus is known from the Jabalpur group.” « This list shows that the flora cannot be well taken as the representa- tive of the Rájmahál group proper, and as the shales of Ragavaptram overlie the Gollapili beds (true Rajmahal group), the flora of the former may be considered of somewhat younger date. | ` * Besides the plants there are some animal remains, amongst them a form of Ammonites, which is also found in the Sripermatur group, un in the shales from the Nellore- Kistna district ; then Leda, Mytilus, and some others, which are ill-preserved and ean hardly be determined speci- fically, and a specimen which can hardly be distinguished from Zrigonia interlavigata.” ! The other remains found were— Fish scales (cycloid) ? 2 sp. Ammonite (a further species). Solen. Tellina. Peeten. Besides these,—these are very common,—some strange, flat, straight, or sometimes slightly curved tubiform bodies from 1 to 2 or 3 inches 1 For Feistmantel’s references to Sripermatur and Nellore-Kistna, see Foote, Memoirs Geol. Surv. of India, Vol. X, pt. 1; Vol XVI, pt. 1; Records Geol. Surv. of India, Vol. XI, p. 156, e£ seg.; also Manual of the Geology of India, pt. 1, pp. 149, 247. (21990098) UPPER GONDWANAS. 27 in length and a quarter of an inch broad, tapering slightly from one end to the other, having very narrow closely-packed folds or рПса- tions transverse to their length, about which folds numerous foraminifera are impacted. There is no enclosing case to these bodies, and from their form they appear to me to be of the nature of exerementitious discharges of some animal. The succession of strata, exposed on the north side of the outlying hill near R4gavaptiram, is as follows :— At the bottom of the slope and some distance out from it, the under- lying dark-brown and nearly black ferruginous Order of strata. TM sandstones and conglomerates of the Gollapili group form a narrow terrace. Then all is obscured by the debris from the shales and sandstones of the hill for a good width ; indeed I do not think the junction is anywhere clearly exposed along the foot of the whole range of these low hills. Good measurements could not be made on such broken slopes, the lE ҮШІ loce slipping having been very Bison on pus hill- side, but the following approximate thicknesses are given. The lower half (about 60 feet) of the hill is of fine white and buff shales, which towards the top become coarser and more sandy. At 40 feet there 1s a fossiliferous zone about 6 feet thick, with greenish- yellow sandy seams containing the smaller of the two Ammonites and nearly all the other shells named. A tolerably persistent seam, about 1 foot thick, of hard sandstone succeeds the 60 feet of shales, and over this come reddish and brownish sandy shales, with Pizlophyllum, &e., Tellina, and the larger Ammonite, for about 3 feet. ; Next come about 53 feet of pale (brown weathering) very fine sandy shales, which again towards the top become coarser, more sandy and rough to the touch, with purple spots and blotches, in the middle of which are occasional Ledas, Pectens and the tubiform bodies with fora- minifera. ( 221 ) 98 KING: COASTAL REGION ОЕ THE GODAVARI DISTRICT. Over these are 30 feet of lighter-coloured (nearly white), purple- blotched ferruginous sandy shales, having 4 or 5 feet of soft white clayey shales at the top. On the top of these is a 3 feet band of ferruginous dark-brown р red yellow and purple sands and clays, with Clay-ironstone band. 3 x seams of flat oval clay-ironstone concretions and irregular segregations having very hard and ferruginous cores, which is again sueceeded by 10 feet of the soft white clayey shales. This clay-ironstone band is fairly persistent throughout the shale exposure. ; About 4 feet of softish irregularly vesieular lateritoid clayey sand- stones then succeed the soft white shales, and are in their turn capped by the semi-vitreous gravelly beds of the Tripati group. The dip is very low, seldom more than 5° to 8° to the south-east. This succession is seen for some distance round the slopes towards Unnamalanka and to the south of Régavaptiram. South-westwards, in the direction of Komera, the clay-ironstone band is strongly developed,. though the whole group is gradually becoming thinner, and the slopes are profusely covered with fragments of the concretions some of which are very large, 3 to 6 feet long. The shell of these is generally of innumerable fine concentrie lamine of brown, yellow, red and purple colours, the core being of a brilliant red colour, or, as often, dark purple; many fragments being also seamed with in- filtrations of gypsum. I failed to find any organism in them. The . ironstone band is continued to the west-south-westward, beyond the limit of the shale member of the group under the Tripati sandstone of Dávanavárgüdem. A further good exposure of this group, though the shales are not now Section below Tripati 50 well developed, occurs in the slopes below the u? scarps of Tripati, which are very much covered up by the debris from the sandstones and ironstone band above. The following section is made up from measurements taken not very distant from each other, commencing at a pot-hole in the small таѓо a short (1022299) UPPER GONDWANAS. 99 distance north-east of the village, where beds of the Gollapili group are exposed :— Tripati beds. Shales 45% PO 5% a about 20 Concretionary clay-ironstone band s Em Рб Fine clayey shales, grey; purple paper shales обо 2 84 Pale grey lilae shales, or soft unctuous shales with thin aka irregular seams of grey and yellow sandstones, ver y variable 2/ 4 to 3-inch flaggy beds of pale brown sandstone passing up into a 6-inch seam of thinner and more ferruginous brown and purple layers Sad is vei CONO Soft and purple reddish sands Er oot Sev Hard brown sandstone ES vs ... 0—6” Fine-grained hard brown sand Е 07—81” Thick and thin-bedded white laminated yellow a pale buff fine soft sandstone 224 080 den .. 9-6” (Blank, junction not well seen for a foot or so). Fine-grained soft purple sandstones mottled with round spots and kernels about the size of peas, of nearly black ferruginous matter 2 onc "E 59 el) 7-3” Grey and parple soft vesicular-looking or porous ferruginous sandstone... 45, 260 > l'or more. Gollapilis; coarse pebbly ferruginous sandstone, Bast 3 feet exposed. No fossils, except a few indistinct fragmentary plant remains, were found here, The lie of the Peli is rather flatter than at Régavaptiram, but there is at times a good deal of undulation, which, however, is in part due to slipping of the strata on the slopes. Here there is little correspondence with the Rágavapüram section, except in the thick band of shales, now much lessened, and in the stronger development of the ironstone band. The beds below the shales are very variable, and they are not constant for any distance, showing that they are, in fact, bottom beds of a series. j There is no large exposure of the mode of superposition of this shale Relation to Gollapili series on the Gollapili beds. Where the junction soU is seen, they are always lying on hard ferru- (50282) 30 KING: COASTAL REGION OF THE GODAVARI DISTRICT. ginous sandstones and conglomerates which it would be very difficult to distinguish from each other, though it did appear to me that below Tripati they are lying on an undulating surface of these—not neces- sarily a denuded one—which is overlapped by different beds of the lower part of the group, but still to such an extent that this could not be called a case of local false bedding. The group is, however, remarkable in its lithological constitution, and is thus eminently separable from the underlying Very different rocks. Я 4 х i sandstone series, irrespective of its much smaller extent; and this character or condition offers ground for the conclusion that a decided interval of time must have intervened between its depo- sition and the conclusion of the period of sandstone formation. The material deposited and the remains of animal life are altogether different to those of the lower group, though there are still two of the vegetable forms common to both. The interval may represent a period of gradual depression, during Indication of separa. Which the area became too distant from the OA OÈ EIS receding shore line to be within the range of coarse arenaceous deposition such as had been deposited in the Rájmahál period. Under these new conditions, the bottom was then more likely to be covered by the finer sediment only, containing shell remains with a pelagie facies generally making up the Rágavapüram strata, the few plant remains having been floated out to sea. Tripati sandstones. —The very low scarped edge crowning the south- western slopes of the Tripati range is made up of these beds, and their outerop is continued, but with a less constant lithologie facies, to the right and left, dropping down to the hollows of the streams crossing: the range, and rising up again in less elevated prominences until it finally shores up against the Gutalla gneiss ridges on the right bank of the Godávar or sinks down under the narrow alluvial strip of the Tammiler to the north-north-west of Ellore. There are many places where this group of sandstones is well and conspieuously developed, but a good and notable headland of them ( 224 ) UPPER GONDWANAS. ol ‘overhangs the well-known temple of Tripati,! some 23 miles north-east of Ellore and 28 west-by-south of Rájahmundry; and the name of the group is so taken. In no section is there a fair idea given of the thickness of the series, the scarp only showing some 40 or 50 feet at the most of the lowest strata. From the scarp there is a long slope down to the south-east, on which little can be learnt of the upper beds; but wells are sunk at two or three points, to a depth sometimes of nearly 70 feet, and these do not appear ever to have touched the lowest or scarp beds, so that at a rough caleulation the whole thickness cannot be taken at more than 150 feet, if indeed it can Thickness. be so much, and from personal observation I can only write off 190 feet of this. The lie of the beds is from 5° to 10° south-east ; perhaps they may, on the whole, dip slightly more to the eastward of this and may be flatter than the Rágavapüram beds. The lower portion, or the scarp beds, are often scarcely to be distin- Lithology and succes- guished from the bottom Gollapili beds; indeed, it pon may also be said that they are frequently just . as undistinguishable from the much newer Rájahmundry sandstones. Lie. They are essentially a set of dark-brown and reddish sandstones, gravel beds, and conglomerates, with bands of highly indurated or vitreous siliceo-argillaceous beds of the same kind and concretionary elay-iron- stones. They are rather softer and more varied in colour towards the bottom, becoming harder and more ferruginous higher up; and it is these harder beds which make up a good dealof the Nullacherla and Yernagüdem country leading down to the delta lands. Much of the hardness of the upland rocks is, however, due in great part to weather- ing, these heavy ferruginous sandstones and conglomerates having a wonderful tendency to assume a lateritoid character on exposed surfaces, 1 This is Chinna, or the ‘smaller’ Tripati and only inferior as a place of pilgrimage to the more famous and larger Tripati, in North Arcot, the most sacred temples of which are above a far grander scarp of quartzites of the upper transition series.—/$ee Memoirs Geol, Surv. of India, Vol. VIII, pp. 18, 177, 179. 32 KING: COASTAL REGION OF THE GODAVARI DISTRICT. Nevertheless, there are undoubted hard and vitreous bands and beds whieh owe their induration and jaspery condition to other than mere weathering forces, which are properly characteristic of the group. Such is the style and condition of the first 40 feet of the group in the greater part of the length of the outerop, that is, between Dávanawár- güdem and Yádavol. Beyond the latter place the beds are generally softer and freer ferruginous sandstones, still of the usual dark-brown and - reddish colours. To the south-west of Dávanawárgudem there is quite a change in the character of the beds, which will be referred to imme- diately. Above the scarp beds is a set of somewhat softer and more variegated freer sandstones which are only known by the few wells between Unnamalanka and Yernagudem. A few poor and unrecognisable vegetable remains, such as fossil wood, probably coniferous, were all that could be found; and these are always in the softer beds between Yádavol and Annadávarupád. To the south-west of Dávanawárgüdem, the hard ferruginous or Variations in members SCarp beds thin out under the low scarp to the сев south of Tundkalpüdi, and are very little separated from the heavy ferruginous beds of the Gollapilis, which are here locally the ‘ Gut’ beds of Sanáshi and Kunlacheru ; and they are more or less recognisable as far as the left bank of the Tammiléru. The upper softer variegated beds are replaced by a set of .strata in the scarp which are quite different to the usual style of beds in the three groups, with the exception of some of the Kunlacheru beds, the resem- blanee to which is such that W. T. Blanford! was very naturally in the preliminary survey led to suppose that they belonged to the Kamthi series. | These new beds begin to show in ће jungle to the west of Dávana- wárgüdem, south of Tripati. — Yellowish-brown laminated felspathie sandstone, the sand fine or not easily distinguishable, more properly a fine clayey sandstone with frequent seams of white clayey fragments 1 Rec. Geol. Surv. of India, Vols. IV, p. 49, V, p. 23. ( 336 ) | UPPER GONDWANAS. : 99 or galls; also highly ferruginous brown and purplish indurated clay and hard sandstone, with seams and patches of grey iron-ore granules. At Tundkalpádi scarp itself, the strata are hard, compact, rather fine-textured yellow sandstone and pebbly beds. | The thick beds of yellow sandstone are often во compact hard and vitreous that they are more like coarse jaspery beds. There must be about 100 feet of these yellow. beds, but they are very irregular in their separate thicknesses, swelling out considerably at times while they and the pebbly beds shade into each other. .. Under the yellow beds are more variegated and various sandstones, with a thin band of dark-brown ferruginous sandstones the particles Tundkalpüdi beds. of which are coarse and like rice grains of quartz. The yellow beds give a very favourite building stone in this part of the country, one thick bed of 10 feet having been opened up by two large quarries. Further south-east, the yellow beds of the Tundkalpüdi scarp have been denuded in the Náyanapalem stream, while the underlying variegated beds show in better force; and among these is a band of bright-red (often vermilion red with occasional white and yellow blotches) fine-grained sandstones which are quarried about a mile north-east of Peddavegi, and are well known all over the Ellore side of the country having been largely used in the eanal works there. Close to the left bank of the Tammilér, fine compact yellow sandstones are quarried at the village of Janampet which must belong to the same group, though I think they are not of the Tundkalpádi band. On the other side of the river, there is no further sign of any Cut of at the Tam. Sandstones answering to those of Tundkalpádi : miler. the change is abrupt and remarkable, to a series of dark-red, brown, and nearly black ferruginous conglomerates and sandstones which lie directly on the gneiss and then run up to the north- west as the long slopes of the low Düdugut range. The disappointing feature about these sandstones of the Tripati c ERN Peddavegi beds. 34 KING : COASTAL REGION OF THE GODAVARI DISTRICT. and Tundkalptidi scarps is that they have not as yet yielded any recog- Represented in the Bizable fossils, while their abrupt finish at the Шағала river banks limits their identifieation with other sandstones in the field. However, some 24 miles away to the north- east of Rájahmundry, at Jaggampet, coarse ferruginous sandstones and conglomerates are lying directly over a sloping floor of gneiss; and beyond this, smaller and smaller patches of these oceur near Kirlampudi, Ayaparáz-Kotapili, Paidikonda, and to the east of the Srirámpüram, one of whieh has yielded fossils of Umia (Survey classification) or upper jurassic age. These sandstones ате, in many respects, very like those of either the Gollapili, Tripati, or even the Rájahmundry zones but they are, on the whole, more like Tripati beds, and their upper jurassie age favours this correlation of them. At Ayaparáz-Kotapili (24 miles north-east of Coeanáda, and 4 miles Fossiliferous beds of east of Bendaptdi, on the Rájahmundry-Vizaga- Sup оар: patam road), a low ridge rises out of the alluvial flat on the south side of the village. "This is of gneiss overlaid by very coarse ferruginous indurated clays and conglomerates, which are suc- ceeded by a set of fine thick and thin-bedded grey and purple sands, with a few clay-ironstone concretions, containing a few fossil shells. The beds are dipping about 52-10% south-south-west, and the ridge may be about 40 feet high at the eastern end. The fossil bed is thin, a coarsish soft muddy sandstone of a purple color, ferruginous, full of fragments of shells, which, from their ferru- ginous constitution, break up very easily and fall to dust. It runs along about half-way up the ridge between pale yellow and buff, and pale purplish sands, soft and fine-grained, largely made up of concre- tionary masses with hard purple sandy cores. Occasionally, fossils are seen on a hard brown surface of some of the sandstone beds. The finer and more compact beds are overlaid by very coarse-grained hard ferruginous sandstones of little rounded particles of clear semi- translucent quartz, with occasional small pebbles of white quartz and clay. ( 22) UPPER GONDWANAS. 38 ` I obtained the following fossils, which were hastily determined by the late Dr. Stoliekza just before he left Calcutta on his ill-fated journey to Yarkand :— Marine fossils. Belemnite, ; | Imoceramus, Ammonites, | Pseudomonotis, Helicoceras, | Lima, Trigonia ventricosa (Krauss), | Pecten, M sıneei (Sow.), | : Fossil wood, ? coniferous. Dr. Stoliezka considered that these fossils showed their beds to be the equivalent of the Umia beds in Cutch, which are of uppermost jurassie age : and since, on phy- Uppermost jurassic age. sical а lithological grounds, it seems very probable that these beds are on the same horizon as the proper Tripati beds, it is for the present presumable that this group is of like age. We have thus, as representatives of the Upper Gondwána series on : this part of the Madras coast, three groups of General conclusions. at : | rocks clearly distinguishable from each other by their superposition, lithological constitution, and fossil remains, though they are not so clearly separable by undoubted denudation, the strati- graphieal breaks showing, rather, intervals of ordinary depression and elevation. | At the same time, had there been no fossil remains, I do think that the lithologieal differences and the different extent of the three groups would have attracted more notice than W. T. Blanford! is inclined to suppose, for we had already become acquainted to the south with the patches of shales at Sriperumbudur (Sripermatür), and in the Trichinopoly, Nellore, and Kistna distriets; while I do not myself recollect any case of so distinct a set of beds in all the vast area of Kámthis extending from Chintálpádi right up the valley of the Godávari to the Central Provinees. On lithologieal grounds alone, ! See Manual of the Geology of India, p. 149. ( 229 ) 86 KING: COASTAL REGION OF THE GODAVARI DISTRICT. 1 should have distinguished the present beds as three groups in the area between the Godavari and the Tammilér; and I should certainly have been inclined to look on the whole series as very possibly repre- sented by some of the Räjmahäl beds to the south. The want of evidence as to strong stratigraphieal breaks, though a great loss, is, as will be seen later on, common more or less to the whole series of formations in this area, from the Rájmaháls of Gollapili up to and including the Rájahmundry sandstones, which represent a period ranging from, say, middle mesozoie times up to middle eocene ; for there is no greater show of unconformity—except by overlap—between the Tripati sandstones, the infratrappeans, the traps and intertrappeans and the Rájahmundry beds, than there is between the groups now in question. There is, however, the lithological difference, there is also overlap, and, as I have endeavoured to show, there is some ground for looking on the Rágavapáram shales as overlying an unevenly worn surface of the Gollapili group—in fact, that the latter are perhaps more separable from the former than these are from the Tripati beds. : The fossil evidence, as far as it goes, is tolerably decisive as to the separation of the groups. Only two plants of the Göllapili beds occur іп the Rágavapáram shales; and there is the entirely new feature of these shales being distinguished by a marine fauna. Of course, it is quite possible that the Gollapili beds may have been deposited in salt- water, or rather, close to the shore; but considerable changes must have taken place in the area of that sea and the adjoining land before the shales and their pelagie remains could have been deposited. In the succeeding group there are no recognisable plant remains; but its repre- sentative at Ayaparäz-Kotapili is remarkably distinet in the facies of its fauna, which has two fossils, Trigonia ventricosa (very common) and T. smeei, which are common to the Umia beds of Kutch; and its beds, like those at the eastern extremity of the Tripati outcrop, are lying direetly on the gneiss. ( 230 ) DECCAN TRAP SERIES. 37 CHAPTER IV.—DECCAN TRAP SERIES. In the neighbourhood of Rájahmundry and on the opposite side of Trap scarps of Rájah. the river, further ranges of low hills, with searps ur CE to the north-north-west and long south-eastward backs, of sandstone and other rocks, form the first fringe of rising grounds edging the alluvial flats. The escarpments and their lower slopes have long been known for their trappean outerops which display en both sides of the river interbedded fossiliferous limestones, and on one side (the western) are underlying sets of beds.also containing numerous fossils. On January 30th, 1835, Dr. P. M. Benza, while travelling in this distriet, visited the western or Pungadi region in Long known. ( і company with Colonel Cullen who had told him that he would here meet with “shell limestone underlaying and alter- nating with basalt."! I have been unable to find any record of Colonel Cullen's observations on these beds, for he had evidently become ac- quainted with them in previous years; but he was at any rate right as to there being an infra-trappean set of beds, as well as an intertrappean outcrop. It is not at all clear from Benza's paper whether he did see these underlying rocks, but he does not distinetly mention them ; and he appears to have misunderstood the position of the Rájahmundry sand- stones which have, somehow or other even by subsequent writers, been considered as subjacent to the traps. Otherwise, Benza’s paper is an interesting record of the early knowledge of this so distant and isolated an outlier of the Deccan traps on the eastern coast. He does not offer any opinion as to the age of the fossiliferous beds. The locality near to Rajahmundry does not appear to have been brought to notice until 1854, when Mr. Walter Elliott, of the Civil Service, sent a collection of fossils from Katéru to the Asiatic Society of Bengal, with notes? thereon. Не shows that the Kátéru limestone crops up between flows of basalt, but offers no comment on the age of 1 Madras Journ. Lit. and Sci., 1837, Vol. V, pp. 50—53. ? Journ. As. Soc., Bengal, Vol. XXIII, p. 399. (2/9309 ) 58 KING : COASTAL REGION OF THE GODAVARI DISTRICT. the rocks, further than that the trap is of the same kind as that he had observed in the Deccan. | : However, this region had not long to wait after this for thorough ee ы scientifie treatment, for ın 1855—1860 the Rev. Messrs. Hislop and Hunter issued papers! treating incidentally of this region. Hislop’s last paper is mainly taken up with the traps and intertrap- peans in the neighbourhood of Nägpür ; but there is a good deseription of the Pungadi and Kátéru rocks, and nearly all the fossils which have been found up to this time in the intertrappean beds are described and figured. Heis strongly of opinion that the trap underlying the limestones is newer than that above,—in fact thatit 15 an intrusive sheet ; and I must say that, though he did not see the rocks here, thereis rather more apparent evidence about them for this view than what he offers for the Nagptir outcrops. Of the age of the Godavari rocks and their fossils, he says,—' From all these facts, I am disposed to deduce the inference that our intertrappean or subtrappean deposits belong to the lower eocene;" and this may be true for the intertrappean beds, at any rate, though, as will be seen further on, there is some evidence towards ranging them rather lower than this in the geological scale, even to the possibility of their being of intermediate age between the secondary and lertiary periods; while the infra-trappean beds appear to have upper cretaceous affinities. The succession of rocks in the whole series of traps and fossili- ferous beds in the Pungadi region, where it is Grouping. ENDE : also most perfect, is, in descending order:— | Basalts . occ Fossiliferrous limestones ... 500 ... Jnter-trappean. Basalts -Fossiliferous limestones and calcareous sandstones ... Infra-trappean. At Käteru, there are only the upper and lower traps, with an inter- mediate band of fossiliferous beds. 1 Quart. Journ. Geol. Soc., Lond., Vols. X, p. 1655; XI, p. 365; and XVI, pp. 154—166. ( 982 ) DECCAN TRAP SERIES. i 39 In addition to the specimens obtained by me, a large collection was sent up to the Calcutta Museum in 1876 by Mr. A. J. Stuart, then Sub-Collector at Rájahmundry ; and these, including the large number of fossils described by Hislop,! give the following list :— Fossils. INTER-TRAPPEAN, Pangadi and Käteru. GASTEROPODA. Fusus pygmaeus, Hislop. : Cerithium leithü, Hislop. Pseudoliva elegans, Hislop. S stoddardi, 5 55 sp. Vicarya fusiformis, т Natica stoddardi, Hislop. Turritella prelonga, ,, F 2 sp. Paludina, sp. Cerithium multiforme, Hislop. Physa prinsepi, Sow. > subeylindraceum, Hislop. Dentalium, sp. PELECYPODA. Ostrea pangadiensis, Hislop. Corbicula, 2 sp. Anomia kátéruensis, » Cardita variabilis, Hislop. ss 2 sp. = pusilla, > Lima, sp., Hislop. Cytherea orbicularis, ss Perna meleagrinoides, Hislop. 55 wilsoni, » Э ATY » wapsharet, ., Modiola, sp., Hislop. A rawest, m Arca striatula, ,, 5 jerdoni, 25 Nucula pusilla, ,, А elliptica, = Lucina parva, 5, 5 hunteri, ,, » (Kellia) nana, Hislop, Tellina woodwardi, ,, Corbis elliptica, Hislop. Psammobia jonesi, 2 » Asp. Corbula oldhami, Ен Oorbicula ingens, Hislop. $3 suleifera, 5; INFRA-TRAPPEAN. CRUSTACEA. Chele of crab. (These are very numerous, but they are the only remains of the crab.) CEPHALOPODA, Nautilus, sp. 1 Op cit. ( 958 7 40 KING: COASTAL REGION OF THE GODAVARI DISTRICT. GASTEROPODA, Rostellaria. Pseudoliva. Murex. Pleurotoma, 2 sp. Fasciolaria. Volutilithes. Latirus. Natica. Pyrula. Turritella, 2 sp. Fusus, 2 sp. Dentalium. Buccinum. CILIOPODA. Lunulites. PELECYPODA. Ostrea. Cardita variabilis, Hislop. Cucullea. » ? dsp. Pectunculus. Cytherea. Corbis. Out of all these, only one, Cardita variabilis, is common to the inter and infra-trappeans. The latter are clearly marine, in contradistinction to the Pungadi and Kátéru beds which are estuarine. | Infra-trappeans.—The lowest beds associated with the traps occur at the base of the northern slope and along the western half of the Pungadi range of hills, 12 miles to the west of the right bank of the Godävari, where by their position they succeed the Tripati sandstones of Yerna- etidem. ` They consist of a series of sandstones, calcareous towards the top, with an upper seam or crust of fossiliferous limestone. "These strata are more or less exposed at three points along their outcrop, namely, a little south of Gowripatnam, at their eastern end; again on the road from Pungadi to Düdkür, about a mile before reaching the latter village; at Dádkür itself, and a little south of it; and again at a point near Devarapili, at the western end of the outerop. From end to end the exposures indicate a length of about three and three-quarter miles. The greatest thickness ascertainable occurs at Düdkür, where there г are 48 feet of sandstones, &c., and the lowest Thickness. beds are not exposed, a good width of ground between these and the Yernagüdem sandstones of the Tripati group being made up of superficial deposits. ( 234 ) DECCAN TRAP SERIES. 41 The fossiliferous band is variable in its constitution and thickness: it is, south of Dádkür, a hard brown and greyish sandy limestone of about 6 inches, resting on good sandstone which is rather calcareous. At the crossing of the road (mentioned above), it is a hard dark-grey sandy and earthy limestone of above 2 feet in thickness. "Towards Gowripatnam the limestone Lithology. band is not so definite; it shades down into a coarse brown calcareous sandstone, full of fossils, 2 feet in thickness, and this is the character of the outerop still further on, past Gowripatnam. Near Devarapili, or about } a mile east of 16, and to the south of the high road, there is a small section exposed in a nala at a little cliff over which the small stream falls and wanders through the fields below. There are, uppermost, about 5 feet of soft friable weathered fine-grained dull dirty greenish trap. This trap is lying and undulating slightly over limestone (crowded with Turritelle) and yellowish and green friable sand- stones; that is, the trap is sometimes covering up solid hard parts of the Turritella limestone, and close by, within a few inches, it is lying on the friable sandstone which is properly the lower band of the fossiliferous seam. In this Turritella zone. short outcrop, the more calcareous band is very thin, or in massive blocks, or it has disappeared altogether, which feature may be due either to the seam of limestone having been originally deposited of a varying thickness, or, as I am more inclined to think, to its surface having been denuded prior to the flow of trap. The proper upper fossiliferous bed is not more than 4 or 5 inches to a foot in thickness; where the blocks or masses of bed are thickest (say 2 feet), there is ап upper part about 6 inches thick of hard, compact, sandy limestone, with fragments of Turri- - delle and other shells scattered through it; and below this is a more friable sandy clay limestone, crowded with fine and large specimens of these shells. The thick band of fossiliferous rock rests on yellowish and greenish fine soft sands, also containing a few fossils. The distinctive and commonest fossil is a Turritella, the rock being often quite crowded with it. The other fossils are generally found in (52555) 43 KING: COASTAL REGION OF THE GODAVARI DISTRICT. the more friable seam below. The greatest show of fossils was, how- ever, obtained from the Gowripatnam outerop, and it was here that the fragments of Nautilus and the chele of the crab were found. The large blocks seen in the small temple and chabutra at Düdkür were doubtless obtained from the out- Fossils. crops towards Gowripatnam, for I could see no thick enough beds for them anywhere near the village. The dip of the beds is at 5°—10° south-east, or east-south-east ; the more prevalent being about 10°. There is no sign of any alteration of these beds by the superincum- bent trap; they are good ordinary rough sandy Apparently unaffected 5 by superincumbent trap. limestones. Regarding the age of these beds, they are certainly, for this region, Relation to Deccan Older than the trap series, only it is not at all совр clear to what portion of this series these flows of the eastern coastal region belong ; and they appear to have been denuded before the trap was poured out: indeed, as far as the generally massive - condition of the trap goes, the surface over which they were poured may have been dry land; at any rate, it could hardly have been the bottom of the sea in which the shells now preserved in the rock lived. It will also be seen later on that the non-parallelism of these beds with the over- lying trap and their intermediate fossiliferous band, and the overlap of these, are against the two being very closely connected groups of a series. The majority of the fossils are such as are usually considered as of гж? tertiary age, particularly the Gasteropoda ; but the Cretaceous affinities. X prevalent Turritella appears to be very close to T. dispassa, Stol., a eretaceous species from the Ariyalár beds of the Trichinopoly district." The Volutilithes is very like Voluta torulosa, Desh., which is a Caleaire Grossier species. There is a strong resemblance, lithologically, between these beds and those of the Lameta group, as described by the late J. б. Medlicott for the typical locality on Resemblance to Lametas. 1 Cretaceous Fauna of Southern India, Vol. II. ( 236 ) DECCAN TRAP SERIES. 45 the Narbada river,! and in the accounts given of other localities of this group by subsequent writers.? The position of these beds, underneath the trap, which is, I think, not perfectly conformable to them, but suffi- ciently separated from them by overlap and by denudation, is also in favour of these Düdkür beds being a marine representative of this group. Mr. W. T. Blanford has already discussed? the relations and age of S D M LAE tbe beds; regarding which he says, after a brief description of them and their fossils: * the most abundant of which is a Turritella, apparently identical with 7, dispassa of the eretaceous Arialár group. If not identical, the two species are very closely allied. A Nautilus, about fifteen Gasteropoda, and eleven Lamellibranchrata accompany the Turritella ; but not a single species except Turritella dispassa, has been recognized as identical either with the cretaceous beds of Southern India or with the eocene fossils of the nummultie group. Тһе collections have not, however, been sufficiently compared to enable the species to be determined with any certainty. Only one single species, too, Cardıta variabilis, has been recognized as occurring also in the overlying intertrappean bed. Although the whole facies is tertiary, there is a remarkable absence of characteristic genera, and the chief distinetion from the cretaceous fauna of the upper beds in Southern India is simply the want of any marked cretaceous form. The fauna is distinctly marine. * [t 1s diffieult to say whether this bed should be referred to the Lameta group or not. The mineral character is similar, but all known Lameta outerops are so distant that the identification is somewhat doubtful. The distinctions between the fossils of the Bágh beds and those of the infratrappeans of Düdkír and Pungadi appear too great to be attributed solely to the existence of a land barrier between the two areas; it is difficult to suppose that the two formations can be of the same geological age, and the difficulty consequently arises that, if the 1 Mem. Geol. Surv. of India, Vol. II, pp. 196-199. ? Mem. Geol. Surv. of India, Vols. VI, p. 216; IX, p. 315; XIII, p. 87, and Rec. Geol. Surv. of India, Vol. V, pp. 88, 115. 3 Manual Geology of India, Part I, p. 316. (БІ) 44. KING: COASTAL REGION OF THE GODAVARI DISTRICT. Lameta beds represent the Bágh group, they are probably more ancient than the Pungadi infratrappeans. Still the balance of evidence is rather in favour of referring the latter to cretaceous times than to tertiary. They may be of intermediate age. Traps and intertrappeans.—The Düdkár beds are overlaid by coarse Overlap the infratrap- Compact blackish-green basalts, whieh attain a pou. thiekness of nearly 200 feet and extend far to the east and west, lapping on to the gneiss in the one case and the Tripati sandstones in the other. About half way up this thickness of traps, there is a thin band of fossiliferous limestones ; but this does not always lie in the middle of them, or parallel to the strike of the Düdkür beds whieh have the greatest thickness of trap on them at their eastern end, while there is again another thickening out of the traps at the western extremity of the field to the south-south-west of Devarapili. At Düd- kúr there are only some 27 feet of trap between the Turritella lime- stone and the intertrappean band; while at the road crossing further . east there must be a thickness of 40 feet at least below the seam of limestone as it passes round the spurs in the direction of Pungadi. In the valley south-south-west of Gowripatnam, the band of inter- trappean limestone is about 4 feet thick; and to the east of this, it thins down to 2 feet in the direction of Pungadi. "Towards Didktir it thickens out to 8 or 10 feet. The seam is generally of two or three or more beds Thickness and litho. the rock being very often a compact crystalline logy. jj: slightly magnesian limestone of white, pink and grey or greyish-green colours, but oftener grey ; at other times it is very coarsely crystallized and fibrous in bands and seams of alternating crys- tallized and fibrous structure with pearly lustre, or often, rudely nodular and concretionary with a radiating fibrous structure. Then again the rock is less crystallized, or dull compact, or soft and friable. АП these | different structures and conditions are of course on the weathered out- crop, or in the quarries where it is presumed that the more generally crystalline character of the rocks becomes changed by exposure; and they are so irregularly distributed through the thickness that it is quite ( 238 ) - - DECCAN TRAP SERIES. 45 impossible to say whether any portion has been more particularly affected by any possible alteration consequent on the eruption of the traps. Theimpression produced is, that the limestone band has certainly been altered in some way, and naturally by the trap which was found over it. The fossils are only seen in the less crystalline or clayey seams, which are those that are mainly quarried and picked out, at various points along the hill side, as when the high-road erosses the out- erop in descending from Pungadi towards Gowripatnam and in the bay south of that village. Here, the uppermost layer is an Ostrea (O. pan- gadiensis) bed, which in some places is chalky-white and crumbly, with nodules of hard compact limestone. The Ostreæ are also found scattered about on the ground, having weathered out of the rock. The other fossils are seen here and there throughout the middle of the seam in hard half-weathered grey limestone, the bivalves being particularly crowded together. Much of the seam in the valley south of Gowripatnam is of thick beds of coarsely and finely sub-erystalline yellowish rock with nests and strings of more lustrous brown and grey saccharine lime- stone and calc-spar. Тһе upper bed or layer is at times a pale-brown or yellowish (weathering of a darker colour) sub-crystalline limestone, the Ostre« standing out on the weathered surface but scarcely recogniz- able in the interior and unweathered portion of the rock. To the west of Dúdkúr, the band of limestone again becomes very thin and is eventually scarcely recognizable to the south of Devarapili. The dip of this seam to the south or south-east is very slight, in fact, iti is often almost horizontal. I obtained the following fossils from the Pun- gadi outcrop :-- Fossils. ? Nassa. Pseudoliva elegans, Hislop. Natica stoddardi, Hislop. P » 2 ер: Cerithium subeylindraceum, Hislop. Тһе massive beds are less easily quarried, and, as the principal use for the stone is for lime-making, the contractors prefer to collect it from weathered outcrops. ( 239 ) 46 KING: COASTAL REGION OF THE GODAVARI DISTRICT. Cerithium leithii, Hislop. op stoddardi, Hislop. Vicarya fusiformis, Hislop. Turritella prelonga, Hislop. Physa Prinsepii, Hislop. Ostrea pungadiensis, Hislop. Perna meleagrinoides, Hislop. Septifer, sp. Corbicula ingens, Hislop. 2 sp. Corbis elliptica, Hislop. Cardita variabilis, Hislop. . Cytherea, ? sp. These are all perfect specimens, and there is a fair number of each of them; Ostrea, Lima, Modiola, and Corbicula having only been pre- viously known from these beds. These specimens were principally got out of the quarried lumps of rock which had been carted from the line of quarries at the head of the shallow valley south of Gowripatnam and laid down near the travellers! bungalow at Pungadi: the rock had become somewhat weathered by being there left lying out in heaps, and thus the shells were exposed and easier knocked out. There is no doubt, however, that they are all from these intertrappean beds; the one thing not known is their relative position in these beds, though some clue is given to this in the account I have given of the few shells obtained in situ. On the Räjahmundry side, or left bank of the Godävari, and about 2 miles along the road to Korekonda, there is a low headland of basalt overlooking the alluvial flat to the north, at the base of which a long line of quarries has been sunk in beds of yellow and buff limestone. ‘These quarries are in the form of a long trench which has been unfortunately filled up in great part by the discarded debris of the calcareous rocks, so that the bottom of the outerop is always covered up, and I do not think a clear exposure of the junetion between the lowest beds and the underlying trap has ever been noted. However, in a small stream bed (xala) at the north- east end of the quarries, the underlying basalt is exposed at only a ( 940 ) Kátéru outcrop. DECCAN TRAP SERIES. 47 few feet lower down than the limestones. On the other hand, the contact of the upper beds and the overlying trap is often clearly exposed. Fossils are not very easily got out from the exposed surfaces, nor are many to be seen, except when the quarries are re-opened at the end of the rains and when the outerop and the debris thrown out during the previous working season have had some time to weather. Between this outerop and the village of Kátéru (half a mile to the Two ‘outcrops, doubt. West), a further trench has been opened up in fully of the same band. — ]imestones which appear to belong to another and lower band; but this is not sufficiently clear, as the dip is very low in both bands, while the second of these is rather to the south-west of the first, so that by the mere flatness of lie it is possible that the latter may be a continuation of the former. I think myself that there are two separate bands, because the rocks differ in some respects, and no fossils are known from the western outerop. The natives (and the views of quarrymen who have worked from their childhood at these beds are not to be hghtly thrown aside) say that the two outerops belong to one and the same band. One thing is clear,-—the western outcrop is also underlaid by trap. The limestones of the band nearer Kátéru are on the whole much more crystalline than the rocks in the fossiliferous band. The trap, both above and below the limestone bands, is a dark-green or greyish compact basalt, very much weathered p DA all over the outcrops and well in below the surface. It has a strong tendency to separate in rounded masses and blocks, with rudely concentric lamins surrounding cores of the solid unweathered rock; and so much is this the case that 1t has been found very difficult to obtain blocks large enough for the irrigation and eanal works in the district. There is an indistinct sort of lamination parallel with the strike of the intertrappean beds when vertical surfaces are exposed, as. ` near the quarries; but there is no thick enough exposure anywhere on the slopes, either here or in the Pungadi field, to show any such bedded lie as is developed among the traps of the Deccan proper. C AN) 48 KING: COASTAL REGION ОЕ THE GODAVARI DISTRICT. There is occasionally a good deal of silica segregated through both upper and lower bands, in the form of caleedony agate common opal and rock crystal in amygdala and geodes, and large masses of calc-spar are common ; but all these are more frequent in the Pungadi field, where the exposures are vastly larger. In the face of the Kátéru quarries, the trap, when least west. is a compact dark-brownish-green earthy rock separating in rude sub-angular masses with large sub-conchoidal faces. The bottom edge of the upper trap is weathered for a few inches of a yellow colour; itis also amygdaloidal with small kernels of dark olive-green clayey matter, the vesicularity being strong for 3 or 4 inches and then rapidly disappearing. Тһе contact of the limestones with the underlying trap is, as already observed, not exposed. There are about 100 feet of the upper traps, which are again overlaid by the Rájahmundry sandstones; the thickness of the subjacent band is unknown, but the stream channel at the north-east end of the quarries shows from 40 to 50 feet of them. The intermediate hand of limestones 1s, when best seen, from 12 to 14 feet in thickness ; but 1t thins out to the north- p c east and is not found at what ought to be its point of outerop on the Korekonda road. It consists of beds of varying thickness, none of which are constant for any length; but the following is the arrangement of beds generally met with, in descending order :— 1. Dark chocolate-brown and greenish clayey mud.—8 or 9 inches. 2. Dirty greenish fossiliferous caleareous mud, hardening somewhat on exposure, — A few inches, to a foot or more in thickness. 5 3. Earthy, clayey, brownish and pale-coloured limestone; with occasional fossils, —1 to 2 feet. 4. Compact waxy pinkish limestone.—1 feet. 5. Yellow ferruginous clayey limestone.—2 to 3 feet. 6. Compact waxy limestone, changing downwards into crystalline and fibrous pearly-grey, brown, and reddish dolomite.—6 to 8 feet. All these, except the fossiliferous mud, may be found more or less ( 242 ) DECCAN TRAP SERIES. | 49 crystalline and arranged in seams and large nodular masses with ver- e TON tical or DOE aa ur, and they go wavering about and tailing into one another, no beds but (5) and (6) being very constant. No. 1 seam has always an even upper surface, but it is not always ` Thinning out of beds, Present, the trap being sometimes in immediate aud samo contact with the fossiliferous seam No. 2;—indeed, it sometimes happens that this latter bed thins out, and then No. 3 is overlaid by the chocolate clay No. 1. It 15 to me a matter of considerable doubt as to whether this choco- late clay is not an accompaniment of the trappean outburst, rather than a final deposit of the intertrappean band. It might have been a very fine dust, and it is curious how its upper surface should be so worn. The upper surface of the layer No. 2, or the fossiliferous mud, is uneven and shows gaps here and there, as though it had been subjected to denuding forces prior to the overflow of the trap, or, at any rate, to the deposition of the chocolate clay. ; There are cases in which the greenish calcareous fossiliferous sand is absent altogether, when at times the rest of the band is more or less crystallized right through, The dirty-green calcareous mud is full of fossils, particularly Cerithi- um subeylindraceum, Vicarya fusiformis, Turritella prelonga, and afew small bivalves. Others, such as Cerithium stoddardi, C. leuthii, Nassa, Corbicula, and Cytherea, &e., are found in the more massive buff limestone below the mud seam. One specimen of Physa was obtained from the fossiliferous mud. The whole length of outcrop at Katéru is not more than about a quar- Fossiliferous seam. ter of а mile. There is no doubt that here in the Kátéru area there is apparent alteration of the intertrappean beds from below Apparent alteration of Р beds from below up. upwards, that is, they are generally more largely NO crystalline and fibrous towards the bottom, and as is the case with the infratrappean beds, the strata nearest to the D ( 248 ) 50 KING: COASTAL REGION OF THE GODAVARI DISTRICT. superincumbent trap do not appear to be altered at all; and the first idea that strikes one is that for which Hislop contends, namely, that the alteration must have been produced by the underlying trap, or (to put it more correctly) this is the view Hislop takes of the situation near Nägpür, and he indirectly extends it to the Rájahmundry area, though it must be remembered that he never saw the latter. A very important point here is, that if the traps between the Case against alters. ІШітабғарреап beds and the intertrappean zone tion, were intruded, they might be expected to have altered the contaet beds equally aboveand below; but there is no more sign of their having altered the subjacent beds than there isof the upper- most trap having altered the fossiliferous layers on which it was poured. Again, it seems to me that any direct alteration by the traps may Influence of weather- һауе been completely obliterated by weathering, ne. and this would account for the unaltered look of the infratrappeans, and of the upper portion of the intertrappean limestones. | The real facts of the case appear to me to be, that the condition of alteration from below upwards is only apparent, and due nearly altogether to the varied eonstitution of the beds, while the metamorphie action of the superincumbent traps was not very strong. The general constitu- ; tion of the beds in the fossiliferous outerop at Explanation. ? Е Kátéru is, that the lower beds are more. purely calcareous, while the upper beds are less and less pure, or are more and more clayey, until in the topmost beds they are caleareous muds. In the Pungadi outcrop the clayey bands are rarer, and there is, as far as I could see, no representative of the muddier beds; and as a consequence, the alteration or the degree of crystallization 1s more persistent throughout the whole thickness, there being still, however, a more purely calcareous constitution in the lower beds. At the same time, throughout the two outerops, there is a good deal of tailing in of the limestone beds among the more clayey ones, and thus, at places, we find the crystalline character waving irregularly up and down the series. ( 944 ) DECCAN TRAP SERIES. 51 Tke late Mr. J. б. Medlicott, long асо, discussed the view of Apparent alteration of Hislop as to the probability of the intertrappean EAS E Can) imere. beds of the Central Provinces having been invaded and altered by a subjacent sheet of trap; and in the area examined by him he shows very good evidence against this view, principally that <“ in every case within our experience, the sedimentary beds have been deposited tranquilly on the previously indurated and moreover previously denudated surface of the trap rock : and in several cases a large portion of the material constituting these sedimentary beds is derived from the debris of the trap flows themselves. Unfortunately, the sections at Pungadi or Katéru do not give such clear views of the condition of the subjacent trap; nor did I see a case of any inclusion of material derived from it. On the other hand, though Medlicott is quite as decided in his remarks as to the altered condition of the sedimentary beds between the two traps being from above downwards, still he gives a section illustrating the exceptional developments of the intertrappean rocks, that is, developments in which the alteration is variable, or is distributed through the beds in the opposite direction to that usually obtaining, or again, when there is no apparent alteration at all; which section, on comparison with the details above given, will show what a strong resem- blance there is between the conditions of the rocks in this exceptional case and those more general ones in the Kátéru and Pungadi outcrops. Concerning this section, he says :—“ It is exposed in one of the glens of the Gorchutta valley, a few miles from the village of Singwarra. “ In descending order— ** 30 to 40 feet of sub-columnar trap, showing well a concentric structure. * * 3 feet to 6 inches of dove-coloured grey earthy limestone, containing many shells. This bed, which varies (as stated) considerably in thickness, does not seem to have been even slightly influenced by the superincumbent basalt. It rests on— 4 6 to 7 feet of a mass which is made up of irregular lenticular patches dying out and replacing each other, and which differ from each other as follows :— a).—A grey limestone, somewhat like the bed above, but is more earthy, Т 1860, Mem. Geol. Surv. of India, Vol. II, p. 208. ? Mem. Geol. Surv. of India, Vol. II, pp. 203-204. ( 345 ) 5% KING: COASTAL REGION OF THE GODAVARI DISTRICT. and contains fewer shells. Though less calcareous on the whole, it is traversed by many veins of pure carbonate of lime, which do not occur in the upper bed. (0).—A fine-grained green loam with an incipient concretionary structure, containing a few threads of carbonate of lime, no shells. (c).—The last-named variety passes into this one; the veins of lime become larger, and highly crystalline, until the mass is a crystalline limestone, with a few flakes and strings of the green earth, and these at last disappear. The limestone is a faint yellow or drab colour, and very coarsely crystalline. The facets of the crystals have the pearly lustre of dolomite." * Beneath this comes trap, * ж V Such exceptional cases did not affect the general and more natural condition of affairs in Central India ; but here on the eastern coast, it has seemed necessary for me to state the apparent condition of the intertrappean beds, which I feel sure Hislop, had he seen 1t, would have seized on as helping to favour the far-fetched, though fascinating, expla- nation which one is sometimes apt to take when considering the possible action of intrusive trap sheets. I have gone over this Lower Godá- vari outerop of traps overand over again with Hislop's theory before me; but the intrusion of 40 or 50 feet (at the lowest caleulation) of traps for a length of what must have been at least 14 or 15 miles between the infratrappeans and an intertrappean band of 192 or 14 feet thick was so inconceivable a phenomenon to me that I was ultimately driven to hold by the simpler theory of a pouring out of the upper trap over a estuarine deposit which had been simply laid down on a pre- vious flow. This very small patch of traps and associated rocks is so isolated Relation of this ont. Pere on the eastern coast from the well-known burst to that of the De immensely gıeater development of traps in the Deccan that it is difficult to conceive how they can be connected either by previous continuity or by contemporaneity ; for though their distance (200 miles) from the traps is small as compared with the length or breadth of the Пессап area, still they are entirely isolated without any connecting outlier between them and it. ‘There is, ( 246 ) DECCAN TRAP SERIES. 58 however, the connection by their associated deposits and their fossils: and it seems almost certain from these, that they and the lowest division of the traps of the Deccan are contemporaneous to some extent. Mr. Blanford, in the Manual of the Geology of India and in other works, has treated very largely of the Deccan QU dE COE, traps, and ranges the lower member of them as preferably of upper cretaceous age ; or rather, that the volcanic outbursts began while the upper cretaceous rocks were being deposited on the south-eastern coast of India. The occurrence of the infratrappean beds of Düdkür, with fossils having what appears to be a cretaceous facies, and on which I think the traps are slightly unconformable, would imply a somewhat later age than this; but it is possible that when the collection of these fossils is thoroughly and competently examined, the correlation may be drawn closer. The intertrappean beds are clearly of the trappean period, and they are in some respects rangeable with Hislop’s Hislop's view, 2 2 intertrappeans of Central India; but the latter are essentially of laeustrine or fresh water origin, while the Pungadi and Kátéru fauna is estuarine. However, there are three shells common to the two, namely, Paludina normalis, Physa prinsepi and Lymnea subulata, and, asit appeared when Hislop wrote, there seemed very good reason for his conclusion that the rocks of the one locality were a fair estuarine representative of the lacustrine rocks of the other, and that they are of lower eocene age. I am unable myself to enlarge on or narrow Hislop's generalization further than to instance the fact that the traps do not appear to be disassociated from the infra- Blanford's view. trappean beds in this locality to such an extent of unconformity as the suppesedly upper eocene age of the intertrappean beds would require. W. T. Blanford has, however, entered on this question with reference to the character of the mollusea of these latter beds in the following extracts from the Manual of the Geology of India !:—“ The mollusca, however, 1 Part 1, р. 319. ( 247 ) 54 KING: COASTAL REGION OF THE GODAVARI DISTRICT. cannot be considered as very characteristic of age. They were compared by Mr. Hislop with the nummulitie fauna of Western India; but, as he points out, no forms appear to be identical, and although WNatzca dolium, Turritella affinis, and an unnamed Cerzthiwm found in the ter- tiaries of Sind and. Cutch, resemble N. stoddardi, T. prelonga, and. C. stod- dardi, the intertrappean forms are more closely allied to the cretaceous N. (Mammilla) carnatica, Т. elicita, and Cerithium vagans than to the eocene species mentioned!; and other forms might easily be shown to be affined to those occurring in the cretaceous rocks of Southern India. In the case of Turritella pralonga and Т. elicita, the affinity is very great. The shell called Vicarya fusiformis appears not to be really congeneric with F. verneuilli, the type of the genus?; and the latter has now been found to be miocene, not eocene. On the whole, it may be safely asserted that no tertiary allianees of any value have been detected amongst the intertrappean Rájahmundry fossils, and that their relations are rather with the upper eretaceous rocks of Southern India, although the con- nection is not strong." CHAPTER V.—CUDDALORE SANDSTONES. Rajahmundry Beds. —Both at Pungadi and close to Rajahmundry, the uppermost trap is overlaid by a series of reddish sandstones and conglo- merates, which in other parts of the field extend far over the gneissic, jurassic, and Rájmahál rocks. As far as is known, they are entirely unfossiliferous, and thus their age is only to be made out from the fact that they evidently are strongly unconformable by overlap on the trap- pean series. They bear a wonderful resemblance to the Cuddalore sandstones of the Carnatic, and in fact must be considered as representatives of these. 1 < When Mr. Hislop wrote, the South Indian cretaceous fossils had not been des- cribed.” 2 «This was pointed out by Mr. H. M. Jenkins, Q. J. б. S, 1864, p. 58. He also (p. 65) suggested that the Sind beds containing Vicarya were newer than eocene,—a view since confirmed." ( 848 ) CUDDALORE SANDSTONES. 55 They form four patches of slightly-elevated and somewhat hilly ground on the edge of the deltaie alluviums, which may be considered Extent and mode of under the names of the Rájahmundry-Sámalkot, M Ic the Pungadi, the Pentlum (further to the south- west), and the Düdügut (south of Ellore) patches being separated from one another by the Godävarı, Yera-kalwa, and Tammilér rivers. Each of these patches rises gently to the north in low plateau form, seldom attaining an elevation of more than 250 feet above the sea. In the larger area, or that of Rájahmundry-Sámalkot, there are a few small flat-topped hills in the interior, as to the north-west of the latter town, along the northern edge, and again at the western end where the sandstones face the Godávari in some low scarped hills and plateaus which' run down to the river bank in low spurs at Rájahmundry and Dowlaishweram. Тһе hills near the latter place have been long known for their stone, which was quarried and run down into the river for the great anicut or dam which here stretches from bank to bank (with an intermediate ‘lanka’ or island) for a length of more than 2 miles across the river. The general suecession of beds in the main areas is, lowest, exceed- ingly eoarse and harsh red and pinkish hard conglo- merates and sandstones, felspathie, such as may be seen in the river face of the town of Rájahmundry, and away north-east towards Kátéru or along the road to Vizagapatam. In this latter direc- tion, the ground is covered to such an extent with the coarse gravel and debris of these beds, that clearings had to be made for the few good riding grounds іп the neighbourhood. Several wells have been sunk through these sandstones, gravels, and heavy conglomerates, as at the Central Jail, on the northern edge of outcrop, the deepest of which is 91 feet, but without piercing the series ; so that there are at least 100 feet of these coarse beds over the traps of Katéru. Succession and lithology. Over these coarse beds come yellow-brown and reddish friable clayey sands with, uppermost, white (purple mottled) and red soapy clays, rather hard and much given to breaking up in irregular fragments. ( 249 ) 56 KING : COASTAL REGION OF THE GODAVARI DISTRICT. With these upper clays are seams of brown and pink ferruginous sand- stones. There may be 150 feet of these less compact sands and clays, as made out from the plateau hill south-east of Rájahmundry. These are succeeded by heavy and coarse dark-brown, nearly black, thick-bedded sandstones, occasionally quite vitreous and not unlike quartz- ites. The coarser beds are made up of fine angular grains of clear glassy quartz, dirty quartz, and jasper clay-ironstone, cemented by dark- brown peroxide of iron. With these are also seams of hard scabrous (lateritoid weathering) ferruginous sandy indurated clays, It is in this band that the Dowlaishweram quarries have been worked, and it forms the cappings of some of the small plateau hills east and south-east of Rajahmundry. The long south-east slopes of the quarry hill north- east of Dowlaishweram are made up of brown ferruginous sandy indu- rated clay beds. The thickness of the band is, at the greatest, about 30 or 40 feet. Over this, again, was a further good thickness of white and purple ‘mottled and red soapy clays, with brown and pink sand, which only remains now as outlying hills and plateaus east of Dowlaishweram, as at Rázálá hill, &c. On the top of this hill there are soft friable pale-brown coarse ferruginous sandstones. The long southerly sloping plateaus, with headlands to the north, in the direction of Kuddum, have, in addition to this capping of sandstones, another of lateritic sandstone or laterite, which may, however, be of somewhat later age. We have, thus for the Rajahmundry sandstones at their thickest, the following series in descending order :— ‘4. Upper white and red clays and sands, with capping of soft reddish and brown sandstones, about . ; - 100 feet. 3. Thick-bedded brown, occasionally quartzitic, ferruginous sandstones of Dowlaishweram hills, about 5 о ӨМ) 2. Intermediate soft soapy clays and thin-bedded friable sand- stones . . 150 5 1. Heavy Ы and sss fade E of Rajahmundry town А с 5 5 . 100 وو‎ 300 or 400 feet. ( 250 ) CUDDALORE SANDSTONES. 57° Very much the same series occurs at the Sämalkot end of this area, where the beds run up into rather elevated ground towards and beyond Peddäpüram. This high ground is, however, I think of the Dowlaishweram beds, and is largely quarried. The fort beyond Peddä- püram is built on the lower beds of the band, or on the upper part of the intermediate band of white and red clays; and the colour of these is so bright and rich, that the fort, ridges, and hills around, which are also much surrounded by trees, often present a most glowing picture in the morning or evening light. The Pungadi plateau is made up of the lower conglomerates and sands only, and here the bottom beds are well exposed along, the north- ern boundary, where they form at times a low scarp of a few feet over the upper trap slopes; and also a couple of small outliers on the trap headlands to the west-north-west of Pungadi. The lowest beds are coarse hard sandstones and heavy conglomerates of fragments and pebbles of clear quartz and quartzite, having sometimes a matrix of hard ferruginous clay though the usual cement is simply brown peroxide of iron. The higher beds, as in the rising ground to the south-west of Pungadi, are coarse yellow and buff felspathie sandstones, not at all unlike some Kámthi beds, or the Tripati beds above the Rägavapüram shales. The further Pentlum patch is also of the lowest beds; but this is avery low-lying area, hardly to be recognized, except here and there, as much higher than the alluvium along its northern edge. Some of the beds in this patch are highly ferruginous, and these are worked for iron, which is rudely smelted at two or three villages, principally near Marillamádi and Débcherla. The Dádügut area is again a rather elevated plateau-like range, the back and scarps of which consist of dark-brown and red, sometimes nearly black, hard, and often vitreous, sandstones and heavy conglo- merates, which are here capping sandstones of the Gollapili beds of the Upper Gondwánas. They are more like bottom beds of this latter group, as it is developed in the Kamerapükota country, or as often like the Al 58 KING : COASTAL REGION OF THE GODAVARI DISTRICT. Tripati sandstones. On the whole, however, they answer best, by their constitution and position, to the lower beds of the present group. The old diamond mines of Muléli on the south-west edge of this patch were partly worked in these sandstones. The lie of the Rájahmundry beds is at a very low angle to the south- south-east, at about 5° or so; occasionally they are nearly horizontal. At Dowlaishweram, there isa good deal of undulation and various dipping · on the quarry hill, accompanied by strong alteration of the beds almost to the condition of quartzites, evidently the result in great part of local squeezing and faulting. The dip is on the whole, that is, over the whole field, rather more to the southward of east than 1s the ease with the beds in the trappean series. The unconformability and overlap of this set of sandstones on the Deccan traps are after all much more decided than those existing between the ages of infra and intertrappean deposits, showing that it must be con- sidered as belonging to a series of much later age, or that it is undoubtedly of the tertiary system, and possibly of middle eocene age. Weare thus . able to give the Cuddalore sandstones, which were only known in their own field to be post-cretaceous, a rather more definite position, though we must still wait for that further evidence which shall probably be deduced from the occurrence of tertiary fossihferous strata known to - be associated with the laterite, or, possibly, sandstones of this present series in the neighbourhood of Quilon on the western coast.1 CHAPTER VI.—ECONOMIC GEOLOGY. Building stones, &c.—Beyond the occurrence of very good building stones, such as those obtained from the different sandstone groups and the trappean series, and the limestones associated with these, which have been very largely used for mortar, there is little of economie interest in this part of the Godavari district, as far as the development of other 1 See Manual of the Geology of India, Part I, рр. 337-338. ( 852 ) | | ECONOMIC GEOLOGY. ы 59 mineral resources is concerned. At the same time, the sandstones of Peddavegi, Tundkalptidi, and Jánampet (north-north-west of Ellore) are worthy of special notice. That of the first locality is a tolerably com- pact even-grained rock, in good thick beds, easily worked, and, like most red sandstones in India, given to hardening considerably on exposure, though in the latter process of change its usual brilliant red colour is given to fade or turn brown, and thus the stone is not so well adapted for external ornamentation. The finer and more brilliantly vermilion red variety is one of the handsomest stones I have seen in India, and it would be well worth carrying great distances, as by the canal which is close by, for interiors. The more useful stone is, however, the buff granular felspathie rock of the Tundkalpidi scarp, which appears to be of perfect durability : it certainly exceeds in this respect any gneiss of this part of the district. The old pillars of many small temples in the neighbourhood have become, in a sort of a way, porcelanized on their surfaces, the matrix or hard clayey medium enclosing the quartz granules having become glazed over, while the latter stand out a little over the surface giving a rough but still rounded somewhat saccaharoid surface to the stone. The old unused blocks lying about near the quarries at Jánampet are weathered and hardened in the same way, and they are said to have lain there from time immemorial, the pagoda for which they were intended never having been completed. Such are the specialities among the sandstones; but good stone is to be obtained in many other places all over the sandstone area. Thus, near Peddaptiram in the Sámalkot area, there is very good building material in what appears to be the same band of beds in the Räjah- mundry sandstones which has been nearly worked out at Dowlaishweram, Diamond workings.—A far greater interest attaches itself, however, to the sandstones of one particular part of the district, namely, those near Muleli (to the west of Ellore), which are reported to have yielded diamonds. There is no doubt that these sandstones of the low plateau above the village have been broken up and searched for diamonds, as have also the recent deposits in the valley below; but they are not ( 253 ) 60 KING : COASTAL REGION OF THE GODAVARI DISTRICT. worked now, nor are they thought of in this way; just as very many of the diamond-mining localities in the Cuddapah, Kurnool, and Kistna distriets are now quite deserted. Still, in all these other places of mining or washing with which I am acquainted, the works have been either on known diamond-bearing rocks or in alluvial deposits presum- ably derived from and in the neighbourhood of them. | At Muléli, е old workings are either in very pebbly sandstones of the Dádágut range, which are not known as diamond-bearing in any other part of India, or are in superficial deposits collected below the range and presumably consisting for the most part of the debris of these sand- stones. There is of course no reason why diamonds should not occur in these beds just as well as they occur in sandstones (quartzites) of the vastly more ancient Banaganpili group of the KURNOOL FORMATION! : indeed these very sandstones above Muléli are most likely in part made up of the debris of these Kurnool rocks; but the fitfulness of com- position of such reformed and derived rocks, and their difficulty of being worked as compared with alluvial patches, only make the chances of successful working poorer. At апу rate, whatever may have been the productiveness of the Muléli mines in old days, they are now in a - state of desertion, and have been so for at least half a century. I myself only saw, by the numerous old pits dug over the flat-bedded sandstones to the north of the village, both in the Rájahmundry sand- stones of the Didügut range and in the Góllapili beds underneath, and by the ruins of old washing troughs and sorting floors, that diamonds had been sought for and probably found. But I believe also that many of the pits were, on the other hand, dug for iron ore, such search being still in progress at the time of my visit. Dr. Heyne gives, in his third tract, an aecount of the diamond mines at Muléli (Mallavilly), which, however, enters into little detail, and is really subordinate to a description of those of Cuddapah and Kur- nool According to him, the Muléli workings seem to have been mainly ! See Mem. Geol. Surv. of India, Vol. V, pt. 1. ? Op. cit., p. 92. ( 254 ) ECONOMIC GEOLOGY. 61 alluvial, and these were carried on in the plain around the village, the ore or gravelly bed being some 14 feet beneath the surface, in the kunkury caleareous travertine) clays so common in the older alluvial deposits. Voysey, Benza, and Newbold also refer to these mines in their papers already quoted ; but only to the general effect that the diamonds appear to have been obtained from the sandstones of the plateau, or the alluvial grounds of the valley, but that no productive work appeared to be going on in their time. My colleague, W. T. Blanford, who passed over the ground before I did, refers to the region thus:—* The low rises south of Gollapili are covered with the remains of old diggings, said to have been diamond mines. I could not learn how long a time had elapsed since the works had been abandoned: an old man, at least 60 years of age, told me there had been no mining within his recollection, and the pits have all fallen in, the whole country being covered over with thick bush jungle. The diggings appear not to have been in the sandstone itself, but in the very gravelly laterite which rests upon the sandstone, but the sur- face is so much broken and altered by the pits that it is difficult to say. Тһе workings evidently cover а very considerable area, and are part of the old diamond mines of Golconda,! the ancient name of the hill range north of the Godávari and the adjoining country."? Iron-smelting.—The iron industry of this part of the country might be improved to a great extent, for there are large stores of ore distri- buted through the sandstones of Gollapili, Tripati, and Rájahmundry, but as yet only small workings are carried on in the usual desultory manner in а few localities. Тһе general source of the ore is naturally from that group of sandstones which is the most extensively exposed, namely, the Rájahmundry group; in the other groups there are only edge outerops for the most part, or much smaller surfaces; for it is to be remembered that the natives only work at the more easily-obtained 1 <“ See Voysey, Journ. As. Soc., Bengal, 1833, p. 403; Newbold, Journ. As. Soc., Bengal, VII, p. 232." 2 Rec. Geol. Surv., India, Vol. V, pt. 1, p. 27. ( 255 ) 62 KING : COASTAL REGION OF THE GODAVARI DISTRICT. ores, or such as are exposed to weathering or detrital influences, and these are the red and brown peroxides in manageable fragments, with often small cores of the grey or micaceous iron ore. The places where I have noticed iron furnaces are in the sandstone region between the Yera-Kalwa and Tammilér rivers, or in the Pentlam patch of Rájahmundry and other sandstones near Nullacherla, and further north at Komera in the Tripati scarps. There are also others in the Nüzvid area, at Rámakapéta and Somavaram; and Heyne describes another locality at Latchmipuram in the Pungadi patch of sandstones. The Pentlam area.—Some of the beds in this field are, as already noticed, highly ferruginous and much banded with clay-ironstone con- cretions, &e., and the debris of these, as well as the beds themselves, are worked and rudely smelted at Marillamádi, Jaganatgudem, and Dóbeherla. At the first place the ore used is a dark-brown and purple clay-ironstone, whieh oceurs in small lumps in a lateritoid gravel on. the south side of the: village at a depth of about 15 feet, whence they are dug up by the people from shafts, at the bottom of which they grub about for a few feet. The furnaces of this part of the country are somewhat different from those I have seen in Southern India, in so far as they are not complete truncate cones, but only half completed, so that one side of the interior chamber is left open when the furnace is not in use. At every smelting, this open side of the chamber is closed up: by a thin slightly-curved wall of clay plates or slabs luted together with wet clay; this side of the furnace being flat, while all round the bed of the chamber there is a backing of 2 to 3 feet. A conical chimney of about a foot 1n length is placed over the top of the chamber, and on this again is placed a broken half of a country chatty or earthen- ware pot, through which a hole has been broken, so as to act partly as a funnel for the readier throwing in of the ore and fuel. At one side of the furnace there is a hollow having a small communication with the floor of the furnace, and through this the slag or cinder 1s drawn off from time to time. After the smelting the wall is broken ( 250 ) ECONOMIC GEOLOGY. 63 down and the bloom removed. At this place there were two furnaces and forges at work, and only one at Jagandtetidem. © At Gopallapáram, a couple of miles to the south, the ore is obtained from purple and brown ferruginous sandstones occurring in 6-ineh beds. The Koméra ore is a red and brown, rather yellowish, clay-ironstone obtained from the sandstone beds towards Tripati. Here I had an opportunity of seeing one of the blooms drawn. The clay wall in the front part of the chamber was broken down and carried off in large pieces (to be used again), and the burning charcoal raked out, when a man pounded the pasty—but still, on the surface, in a state of slow ebullition—mass of metal at the bottom of the furnace for a few seconds with a heavy wooden mallet. Then a huge pair of iron tongs was brought forward and rammed, or rather chopped down, here and there on the mass of metal, during which time it hardened sufficiently to allow of one arm of the tongs being lowered down the side of the pit when the bloom was clasped, hauled out, and carried off to a small pit or hollow in the ground limed with charcoal dust. "The partially flaming mass was now heavily beaten over its upper surface (the old under-surface on the furnace floor) for some time with wooden mallets, after which it was left to cool, though a lot of old slag and einder was piled up around it, but not over it, while a red-coloured dross from the floor of the furnace was ladled over the upper surface. After about 10 minutes the bloom was then partially split up by wedge hammers, which left three-deep clefts in one side. The blooms are larger than those of Southern India, and are said to be about 2 maunds (Madras), or 80 Ibs.,in weight, and of the value of 4 rupees when handed over to the contractors. Two furnaces were at work here, and the measurements of each of these were:—height, without chimney, 4 feet ; "breadth across the front, 5 to 4 feet, tapering ; width from front to back, 4 to 3 feet, tapering, the front being nearly vertical. The chamber is about 13 feet in diameter widening to the floor which is about 2 feet; and the whole is surmounted by a tapering chimney with funnel top of about 2 feet high. (ais) 64 KING: COASTAL REGION OF THE GODAVARI DISTRICT. In the Nüzvíd region, there were only three furnaces at work at Rámakapéta, and two at Somavaram ; the ore being obtained from the debris of the bottom beds of the Gollapili sandstones. The old iron working at Lutehmipáram, where I was told smelting is still pursued in a small way, is so well described by Heyne in his ` tract on the subject, that I take the liberty of reproducing the greater part of it here, as it will add considerably to the few details I have been able to collect during my hurried visits to these iron villages, most of which I only came on by chance and while hastening to more important work. | <“ About the end of the month of June 1794, when the thermometer stood at 115, I set out from Vuppáda, for Dr. Heyne's account, ? 2 Letehemporam, a small village in the Polaveram district. This village lies about 14 miles south-west from Räjah- mundry. I found the people in this village extremely willing to show and explain to me everything concerning their iron works. “ The iron-smelters themselves are a poor set of people, and obliged | to plough the land for their subsistence during the wet season, and work as smelters only during the hottest part of the year. The finest and mildest season they employ in cutting wood in the hills, in burning char- coal, and, after these occupations are over, in recovering their health at home; for, besides their repeated experience that every one contracts the fever during his stay among the jungle, we have only to observe their sickly look and their whole appearance, to be convinced that their accounts are correct. To this cireumstance, together with the necessity they are under of cultivating the ground for a part of the year, we may ascribe the unproductiveness of their labour as manufacturers of iron. Yet the iron which they produce is considered as the finest in every respect for tools, razors, etc. Hence the demand for it is great, and the numbef of workmen miserably small, for the miners, smelters, wood-cutters, and labourers, all united together, do not exceed eight or nine men. <“ Stones containing iron ore in considerable quantity are found every 1 Op. cit., p. 218. 0 898 ) ECONOMIC GEOLOGY. 65 where near the village, from six to eight fathoms under ground, imbed- ded in a lithomarge, the discovery of which is a sure sign of being near a bed of iron-stone. “ These iron-stones lie in beds of a small extent, of irregular thick- ness, and various in their breadth from side to side, though this is never very great. * They at first mine downwards in a perpendicular direction, till they are obliged by the different directions of the beds to alter their course accordingly. Тһе breadth of their perpendicular shaft amounts to about 23 cubits, and small steps are cut out in the sides for the convenience of descending. When they have exhausted a bed of iron-stone, they abandon their mine without any further trial, and dig another in a different direction. This negligent mode of proceeding puts them to many inconveniences, and produces much unnecessary trouble, by oblig- ing them to dig holes almost every six yards, which they fill up again when the iron-stone 1s exhausted. “The ground in this place, and the ore itself, being of a very soft nature, no other instruments are required for working their mines than a pointed pickaxe. Тһе ore and extraneous stones are drawn up in baskets; the latter (consisting chiefly of clay) is separated, and the former broken by mallets to the size of a hazelnut. “This ore has much the appearance of a yellow and brown ochery clay. It appears also to contain a mixture, or rather coating, of calca- reous earth. When reduced to powder, it aequires a red colour, and ex- hibits many sparkling particles. “The miners prepare the charcoal which they require for smelting this ore, by burning the wood of the Sandra cciu (Mimosa sandra), which furnishes a solid, good charcoal; but as it is rather scarce in ‘the vicinity of this village, and as а conveyance of 24 or 30 miles, the nearest place where it may be had in abundance, is very ex- pensive, this, together with various other obvious circumstances, must render any attempt to establish a large iron-manufactory in this village a.very hazardous undertaking. No doubt, other kinds of wood might be Е ( 259 ) 66 KING: COASTAL REGION OF THE GODAVARI DISTRICT. found which would yield a charcoal that would answer the purpose suffi- ciently well; but unluckily all the jungle in the neighbourhood of the village consists of very small brushwood. <“ These smelting-works, however, notwithstanding their diminutive scale, attract the attention of every eurious observer, on account of the simplicity of every part of the process and the goodness of the iron obtained. * The furnace consists of a small semi-circular mud wall, very much resembling in shape the half of a hen's egg divided longitudinally, with the largest end uppermost. The wall is built of clay or mud. From the apex to the base is usually 4} feet, while its greatest breadth is 3 feet 9 inches. The external and convex surface has on one of its sides, at the bottom, an excavation serving to receive the scoriz, which are let out through a bole in the bottom. “The internal surface of this mud wall is plain, except a semi-cir- eular excavation throughout its middle part, commencing at the apex and terminating in a circular hole in the ground, which is 13 feet deep, | and as much in diameter. This part corresponds with the square cavity in European furnaces, in whieh the iron is collected. “The use of this semi-circular excavation will be understood by considering the temporary part which is destroyed every day after the smelting is finished. It is a thin, convex, semi-circular wall, and is to complete a cireular hole with the excavation in the permanent part of the furnace. It is constructed in the following manner :—At 5 o'clock in the evening, the hole in the ground is cleaned from the ashes and the remainder of the last smelting, and its bottom and sides coated with powdered charcoal moistened with a little water. At the bottom, to the right hand, is a small circular hole for letting off the scorie. This hole must also be cleaned, and then stopped up with some moistened clay. Charcoal is then thrown into the hole and placed in such a manner that the apex of the heap touches the margin of the hole opposite to the principal work, and another heap of pounded ore is so placed on the opposite side that the middle of the hole is left an empty space. These ( 269 ) i ECONOMIC GEOLOGY. 67 two heaps are distant from each other at the apex about a foot, at the bottom about an inch. This is done in order to rest on the charcoal a kind of funnel-formed channel for the admission of the stream of air produced by the constant action of the bellows. The external aperture of the funnel receiving the nozzle of the bellows is in breadth 5 or 6 inches. Clay is then put upon it, which serves both to fix it and to form the firstlayer of clay that constitutes the temporary part of the furnace. This part is not to be thicker than 2 inches, and it decreases in thiekness the higher it advances. The funnel itself is made of a mixture of clay and husks of rice; and previous to its application is hardened by fire, and then made firmer in its position by a coat of clay laid over 1t. “Тһе funnel being fixed in this manner, the wall is raised, becoming gradually thinner, so that when it arrives at the middle part, it does not exceed the thickness of an inch. Then a barnt stone of the same thickness, from 10 to 12 inches high and from 8 to 9 broad, is fixed upon it, so that it inclines to the opposite side, the circle becoming narrower the higher it rises. This stone is connected with the principal wall by means of mud. In this manner the circle is completed; some holes of 2 inches square being left, one or two on each side. On the stone itself is placed a second stone of the same kind and shape, but smaller, and fixed in the same manner. Its apex is on a level with the top of the opposite or principal part of the furnace, The top of the furnace now serves as the basis for a cone, the use of which is sufficiently obvious. | “This сопе в 12 inches long. Its under-aperture rests on the top of the furnace, where its breadth is 14 inches. At its upper part or apex its diameter is 7 inches. To facilitate the introduction of charcoal and ore into the furnace, the cone is crowned with a large cutcherie pot, the bottom of which is broken out, and thus serves not only to facilitate the introduction of fuel, ete., but is supposed of much consequence as the representation of a swamy. ^ [t has been already stated that some charcoal and ore had been (ol) 68 KING : COASTAL REGION OF THE GODAVARI DISTRICT. placed at the bottom of the furnace, and that the funnel for conveying the wind was placed upon this heap. Some lighted charcoal is put before the opening of the funnel, and the whole cavity is then filled with charcoal, and this is continued as the wall advances in height, the charcoal serving as a support to it, for it is so thin that it would not be able to support its own weight. Within the holes left on each side of the stones, which constitute the middle part of the furnace, some lighted charcoal is also placed. “The under part of the cone is also filled with charcoal. Then a small basket of ore is thrown upon it, and upon this likewise some lighted charcoal is placed. Finally, the whole cavity is filled up to the top with charcoal. | <“ Matters are allowed to remain in this state till 5 o'clock next morning, when two pair of bellows are applied to the aperture of the funnel, adapted for the insertion of the nozzles; each pair of bellows is worked by one man. The several vent-holes in the side of the furnace are stopped up with a mixture of clay and sand. The bellows are then worked without intermission, and an intense degree of heat is soon produced. « The ore is thrown in by small quantities at a time, in small baskets which do not hold above three or four pounds; and for every basketful of ore two basketfuls of charcoal are added. As the charcoal burns the ore gradually sinks downwards, and at last the melted iron and scorize make their way to the bottom. The great object cf the workmen is to supply the requisite quantity of charcoal and ore, and they continue their additions till within a little of the time when the reduced iron is taken out of the furnace. * A hole was left ready at the bottom to be opened occasionally in order to permit the scoriæ to run out. This is done regularly every second hour, or six times during the whole operation. They pierce the clay which stops the passage with a pointed iron, suffer the liquid scoriz to run out, and then secure the hole as before with clay. The cracks produced by the intense heat in the exterior thin wall they take care to CoD ECONOMIC GEOLOGY. : 69 stop up occasionally with moistened clay, and now and then they wet the whole with water in which clay is suspended. * At five o'clock 1n the evening the bellows are removed, and the exterior wall'of the furnace knocked down. The iron, which is found in a solid state, is taken out and beaten for about five minutes with wooden sticks, in order to separate as much of the scorie as possible. Finally, it is cut with axes into two pieces in order to show its internal quality. “Т do not know the precise quantity of iron-stone which is employed in one smelting process. In general, I believe twelve baskets of ore are required, containing each from 4 to 5 computed mercals. Тһе whole produce of one process is about 112 lbs. of iron, which are usually sold for about a rupee. «The iron, as thus produced, is of a very inferior quality, porous, and its pores filled with scorie, and in fact a little more than half | ` smelted, if such an expression may be used; for I am persuaded that ` the whole mass never has been fused, as in that case it would naturally have run out with the scorie through the hole at the bottom of the fireplace. The metallie partieles in the ore are probably at some dis- tance from each other. The fusion of the scorie lays them open to the action of the charcoal. They are reduced to the metallic state, tumble down in consequence of their weight, and, coming in contact with each other at a welding heat, are connected or agglutinated together, without having experienced actual fusion. “The шоп thus obtained is, indeed, of such an inferior quality, that none of the names, by which any of the various kinds of cast-iron are distinguished, can be applied to it. But if it be exposed to the heat produeed by urging a fire by a pair of common bellows, while it is quite covered with charcoal, and, when the scorie begins to melt, if it be taken out and hammered, it aequires the properties of steel, and may then be usefully employed in the making of instruments. ” Other Minerals —Graphite is rather common, but very sparely dis- tributed among the crystalline rocks along the north-western edge of the field, more particularly perhaps in the neighbourhood of Bezväda, (509098) 10 KING: COASTAL REGION OF THE GODAVARI DISTRICT. and again towards the Vizagapatam end of the field, but it is not known to oceur in masses of sufficient size or purity to make it worth working otherwise than in the small way required for local demand. Kondapilli, north of Bezváda, has been long known for its carbuncles or polished garnets; but these are not of much value. They are there washed or pieked out from the sands in the neighbourhood of the town, and their original matrix appears to be a band or bands of massive, garnetiferous, hornblendie gneiss, which occur in and alongside the great hill mass from whence this old and once important town takes its name; whence also they are continued to the north-north-east and then northwards into the Chundragunda taluk or division of this eastern portion of the Nizam’s dominions. Coal.—Just outside the north-western edge of the area under description, the field of lower Gondwánas in the neighbourhood of Ashwaraopet possesses an outcrop of coal-bearing strata, which gave some · promise of the long-dreamed-of chance of coal in the Madras Presi- dency ; but a series of borings which were put down at Beddadanole, though they struck many seams of more or less coaly matter, showed that none of these were worth working, while the borer appears to have penetrated so near the floor of crystalline rocks on which this patch of Barakar strata lies that there does not seem to be the slightest hope of finding deeper or better seams at this place.! ! For further particulars of this enquiry, see Rec. Geol. Surv. of India, Vols. V, p. 112, VL, p. 57; and Proceedings of the Madras Government, Publie Works Department, December 1873, No. 273; July 1876, &c. yer N a] ея f л | 9 4 Penamudam X \ GE 0) 90, U (6 E n GEOLOGICAL MAP. of the COASTAL REGION. OF THE GODAVARI DISTRICT. Scale 8 Miles — 1 Inch. Nuzyıd ® S UR VE Y // j ше. j 1 2 y үші, ) У. D 41: | Memmrs Vol XVI. РЫШ Index of Geological Colors ET [aes [ Rayahmundry Sandatowes . . Cuddalore Sandstens| RäteruTraps S Tater traypoens Deccan Trapa‏ ج || г... infra truppeane агала EE Tipat smistones Г SS ре» [ee Shules E N Z = E Gellapıli Sandatoner. | ПЕ замара Sendytenes. Laer; Gonbranen Жы” E ~.[Metamerghne Series Diy are shewn by arrow thos d 4 Nara а QS b 3 M: t Т f Шіру m 0 27; a s rane’ a SMIT HSONIAN INSTITUTION Т BRARIES ШІШІІ А