Mr. Kaye on Fossilifcrons Beds in Southern India. 231 



attributes the scratches. They follow one another in nearly equi- 

 distant rows of three scratches in a row, and at intervals of about 

 two inches from the point of each individual scratch to the points 

 of those next succeeding and preceding it. They are all slightly 

 convex outwards, three on each side of the median space, or supposed 

 place of the body of the fish. Each external scratch is about one inch 

 and a half in length ; the inner ones are about half an inch, and the 

 middle one about an inch long. These proportions are pretty constant 

 through a series of eight successive rows of triple impressions on the 

 slab from the Mostyn coal- pit. The impressions of the right and left 

 fin-ray are not quite symmetrically opposed to each other on a straight 

 line of progression ; but the path of the animal appears to have been 

 curvilinear, trending towards the right : each impression or scratch 

 is deepest on its supposed frontal side, and becomes more shallow 

 gradually backwards. All these conditions seem to agree with the 

 hypothesis of their having been made by three bony processes pro- 

 jecting from the anterior rays of the pectoral fin of a fish. They are 

 not consistent with conditions that would have accompanied the im- 

 pressions of claws proceeding from the feet of any reptile. 



Dr. Buckland refers to the structure of existing Siluroid and Lo- 

 phoid fishes, and of the climbing perch (Anabas scandens), and Has- 

 sar (Doras costata), as bearing him out in the conclusions he has 

 come to regarding those markings. He also refers to the observa- 

 tions of Prof. Deslonchamps, on the ambulatory movements under 

 water of the common Gurnard, as confirmatory of his views. He 

 has been informed of a slab of coal sandstone bearing similar mark- 

 ings in the museum of Sheffield ; and remarks, that there are several 

 fossil fishes of the carboniferous system approximating the characters 

 of Gurnards, and capable of making such markings as those described. 



" Observations on certain Fossiliferous beds in Southern India." 

 By C. T. Kaye, Esq., F.G.S., of the Madras Civil Service. 



The beds described in this paper are found at three localities ; viz. 

 Pondicherry, Verdachellum and Trinchinopoly. 



1. Pondicherry. — This town, like Madras, is situated on a very 

 recent formation of loose sand, which extends for a considerable di- 

 stance along the eastern coast of India, and which in many places 

 contains marine shells in such abundance that they are dug up and 

 burnt for lime. They are all species which now inhabit the Indian 

 seas, such as Pyrula vespertilio, Purpura carinifera, Cardita antiquata, 

 Area granosa and Area rhombea. The sand is usually bounded by 

 granite, which appears at the surface at Sadras, Madras and other 

 places. Immediately beyond the town of Pondicherry, however, the 

 recent beds rest upon some low hills of red sandstone. A bed of 

 limestone containing numerous fossils succeeds, and at the distance 

 of four miles due west the red sandstone is again met with and there 

 abounds with silicified wood. At about sixteen miles from the sea 

 the sandstone is bounded by hills of black granite. 



The surface of the country does not offer any section exhibiting 

 the relative positions of the limestone and sandstone. In the former, 

 numerous fossils in a high state of preservation were discovered by 



