ls;»S-l>)(t>. No. 33,] UPPER DEVONIAN FISH REMAINS. 31 



that it is |)(i>sible to use them for comparison. In PI. IV, fij;. 2 I have 

 i;iven a i;i-rally enlarged photogiviph of the sculpture of one of the large 

 median body plates. In this picture we can clearly see that the points 

 of the denticles are situated peripherally, and that therefore their ribs 

 are longer and more powerful on one side than on the other. Traquair's 

 description of these denticles simply as "stellate tubercles' does not 

 therefore give the right idea of their form. We further see that the 

 denticles are rather distant and placed somewhat irregularly. However, 

 I assume that the denticles are considerably closer to each other than 

 appears from the illustration. Between the typical ribbed denticles we 

 see round dark spots and these I consider to be denticles that have 

 been torn off by the violent mechanical preparation. A.s is known. 

 these fossils are brushed out of the schists by the aid of a metal brush, 

 which on the whole cannot be particularly favourable for the preserva- 

 tion of the finer degrees of sculpture. 



I have given so thorough a description of the sculpture on the large 

 median body plates because it is of importance for the comparison! with 

 Psammosteiis remains. It would take up too much space to go through 

 all the variations that the sculpture of Drepanaspis undergoes on the 

 various body plates, scales and fulcra. 



Traquair's description of the dermal plates, their arrangement, the 

 caudal part, and the general form of the body, is so exhaustive that 1 

 cannot add anything of special importance. There is however one point, 

 in which I am cjuite at variance with him, and that is his views of 

 the dorsal and ventral sides. In this respect I agree with Bash- 

 ford Dean^ who in his review of Traquair'.s work on Drepanaspis 

 puts forward the opinion that Traquair had confused the dorsal and 

 ventral sides. As far as I can understand, Dea^^'s argument has not 

 been disproved by the new observations put forward by Traquair in 

 his supplemental work.- 



I believe this to be apparent from the following. 



In his main work on Drepanaspis, Traquair summarises his 

 views of the .systematic position of this foi-m in the following way : — 



"These charactei's assign to Drepanaspis a place in the Osira- 

 codermi, and its indubitable resemblance to Pteraspis leads us to class 

 it in the Heterostracous subdivision, although evidence from microscopic 



1 Science N. S. Vol. XIX, No. 471, 1904-. 



2 Supplement to the Lower Devonian Fishes of Gemilnden. (Transactions Roy. 

 Soe. Edinburgh, Vol. XLI, 1905, y. 469). 



