730 Dft R. H. TRAQUAIR ON 



JPteraspis, — only in Drepanaspis the plate and its round opening, though close to the 

 margin, are nevertheless on the ventral surface of the carapace ; a circumstance which 

 rather militates against the idea of an eye having had its place there during life. But 

 as these Hunsriick-slate specimens are all crushed absolutely flat, it is by no means 

 certain that in the original uncompressed condition the opening did not look out to the 

 side. It is, however, perfectly clear that a sense-organ of some sort is here indicated,, 

 and so we may safely apply to this plate and its round opening the term sensory. 



Immediately behind this "sensory" plate is another and larger one (a. v. I.) of an 

 approximately triangular form, a long irregularly-scolloped side internally, a gently- 

 convex outer one, and an acutely-pointed posterior angle. The outer margin of this 

 plate, which we may call anterior ventro-lateral, fits on below the anterior pointed 

 extremity of the great postero-lateral plate (p.l.) so extensively seen on the dorsal 

 surface, but, as already explained, appearing on the ventral aspect only as a thickened 

 margin (see PI. III. and text-figs. 2 and 3). These lateral elements on the ventral 

 surface are on each side separated from the mental and median ventral plates by a 

 series of small polygonal ones, as seen in figs. 2 and 3, but, just behind, there is an 

 ovate-oblong one of considerable size (p.v.L), which may be called posterior ventro- 

 lateral. The space between this and the posterior external angle of the plate p.l. (left 

 empty in the figures) seems in one specimen to be covered by another smaller one ; any 

 way, I think that in this region the branchial aperture must have been placed, though 

 its position is as yet not exactly determined. 



The tail, springing from the middle of the posterior margin of the carapace, is compara- 

 tively short, and terminates in a heterocercal though scarcely bilobed caudal fin, but there 

 is no trace of any other fins or appendages, paired or unpaired. The mode of origin of 

 the tail, as seen from above, is well shown in PI. II. Here we see that the small 

 polygonal plates behind the median dorsal pass into angular imbricating scales, the 

 outer surface of which is covered by small sharp tubercles, which are longer than they 

 are broad, and are arranged in concentric lines which are parallel with the free margins 

 of the scale. But in the middle line, shortly behind the median dorsal plate, there 

 developes a series of elongated median, acutely-pointed and imbricating fulcral scales, 

 which, becoming in succession longer and more acute in contour, form a row along 

 the dorsal margin of the tail. These dorsal fulcra are seen also in PL I. fig. 1 , PI. IV. 

 right-hand side, PI. V. figs. 1 and 2, but more especially in the latter. 



On the ventral aspect the relations of the corresponding inferior row of fulcra are 

 best seen in a specimen belonging to the Prussian Geological Survey, and represented in 

 PI. V. fig. 1. In this specimen, which I have used in the restoration of the region of 

 the body here concerned (see text-fig. 3), the posterior notch of the median ventral 

 plate seems to form, along with an elevated median scale just behind it, a narrow 

 opening, which I take to be the orifice of the cloaca. This is succeeded in the back- 

 ward direction by four narrow elevated scales, which pass into the median fulcra of 

 the ventral margin of the tail and caudal fin. It will be seen, on inspecting PI. V. 



