Vol. 6$.^ CARBONIFEROUS ROCKS OF DEVON AND CORNWALL. 23 



and the diameter 5 to 6 mm. In the Hartland district, however, 

 they have been found as long as 25 mm., and with a diameter 

 of 14 mm. 



A number of microscope-sections have been made, of one of which 

 a photograph is reproduced in fig. 3 (p. 22). Unfortunately, all 

 have proved to be badly preserved, and consequently this discovery 

 has no botanical value. Little can be made out as to the structure, 

 beyond the fact that these bodies are of vegetable origin. In 

 some cases a few cells are preserved ; in others, there is an 

 indication of a central cylinder. 



These petrifactions are, however, interesting geologically from 

 the fact that they are rolled and water-worn, as their shape 

 clearly shows. They are thus obviously derived fossils, of different 

 origin, and possibly of earlier age, than the sandstones and shales 

 in which they occur. Their history appears to have been briefly 

 as follows, fragments of plants, deposited during the formation 

 of some pre-existing beds, became partly silicified, and later these 

 beds were denuded, and their petrifactions redeposited in the sands 

 and muds in which they are now found. There need be, however, 

 no great discrepancy between the ages of the two constituents of 

 these conglomerates. 



VI. The Marine and Freshwater Faunas. 



Evidence has been found of the existence of two distinct types of 

 fauna in the Devon rocks : the one consisting of freshwater 

 lamellibranchs, the other of marine fishes, cephalopoda, and 

 lamellibranchs. In this respect also, the Upper Carboniferous 

 deposits of Devon and Cornwall agree with the Coal-Measures of 

 Yorkshire, 1 Lancashire, 2 Durham, 3 and North Staffordshire. 4 

 Further, neither in Devon nor in Cornwall has any sign been 

 seen of an intermingling of these two faunas. Thus these beds 

 confirm the important conclusion on which special emphasis has 

 recently been laid by Mr. Stobbs and Dr. "Wheelton Hind, 5 that in 

 North Staffordshire, and elsewhere, there is a clear and distinct 

 separation of the marine mollusca from the freshwater forms. 



Although a special study of these faunas formed no part of my 

 programme of research in Devon and Cornwall, I may mention 

 briefly such discoveries as have been made incidentally to the main 

 object of the enquiry. I may add that the great majority of the 

 examples of both the marine and estuarine types obtained were 

 collected by Mr. Iukermann Rogers, to whose skill as a collector I 

 have already referred. 



The marine fauna is much the more abundant in Devon and 

 Cornwall. The fish-remains have been already noticed (p. 15). 



1 Phillips & Daubeny (45) pp. 589-90. 



2 Spencer (98) p. 305. 



3 Kirkby (60) p. 414. 



4 Walcot Gibson (05) p. 295, &c. 



5 Stobbs & Hind (05) p. 515 ; also Stobbs (06). 



