GEOLOGICAL SURVEY IN SILURTAN ROCKS OF SOUTH OF SCOTLAND. 839 



it is an opening at all. Immediately behind this the rest of the head is covered 

 by an area of little soutes, which have their long axes directed downwards and 

 backwards. 



Now a narrow band-like prolongation of this last-mentioned area pass.es obliquely 

 upwards and backwards along the posterior margin of the head as far as the first row 

 of body-scutes. In this band we see, in every specimen, eight small round openings 

 or their impressions. Are these branchial openings ? That is indeed the only inter- 

 pretation which I am able to put upon them. 



In no specimen can any trace of mouth, of jaws, or of teeth be found, nor is there 

 any certain evidence of eye-orbits. 



Passing now to the body, we find that it is covered on each side with five longitudinal 

 rows of scutes, which are much larger than those on the head, and of a narrow, elongated 

 oblong shape, pointed at each end. 



Commencing above, the Jirst row extends from the back of the head to within a short 

 distance of the dorsal fin. It consists of a succession of narrow parallel plates, whose 

 long axes are directed from above downwards and backwards, that is, in a direction 

 contrary to that of most of the scutes of which the other rows consist. This row is in 

 contact in the middle line of the back with the corresponding series of the other side. 

 The second row proceeds below the first as far as the dorsal fin, in fact passes into that 

 appendage ; the direction of its parallel scutes is downwards and forwards. So it is 

 with the scutes of the third row until they pass the dorsal fin, when, from that point 

 to the termination of the series on the tail pedicle, their direction is suddenly altered, 

 and their long axes point downwards and backwards. The fourth series, which extends 

 right back from the head to the tail pedicle, has its scutes directed downwards and 

 forwards from beginning to end without interruption. The fifth row, which runs along 

 the ventral margin from the head also to the tail pedicle, consists of scutes which in the 

 front half of the fish point downwards and backwards, but exactly in the middle of the 

 ventral curve the direction is suddenly reversed and the scutes come to have their long 

 axes directed downwards and forwards until the band ends at the caudal fin. 



In the middle line of the back, and placed exactly between the termination of the 

 first row of body- scutes and the dorsal fin, is a single narrow oblong azygous plate. 



Then on the opposite aspect of the body we have on the front half of the ventral 

 curve five marginal scutes, of which the first four are narrow and spurless, while the 

 fifth rises into a backwardly directed sharp conical elevation. So far as I can judge, 

 these scutes seem to be placed on one side of the middle line, leaving us therefore to 

 suppose that they are paired structures, but this I cannot prove, as in no specimen arc 

 their fellows of the opposite side to be seen, and analogy with Lasanius would lead us 

 on the other hand to suppose that they formed one continuous series with the remaining 

 five marginal scutes behind them, which undoubtedly are placed in the middle line, and 

 are remarkable for the prominent thorn-like spine borne by each. The first one of these 

 is apparently formed of two closely fused together, the two halves — anterior and pos- 



