1886.] SQUALORAJA POLYSPONDYLA. 531 
the specimens numbered I. and III. reveal a few hitherto unnoted 
facts concerning the arrangement of the small prickly tubercles. 
As already indicated in Davies’s large figure, but still more satis- 
factorily shown in our Plate LV. fig. 1, a series of the tubercles with 
especially long recurved hooklets is arranged along either edge of 
the prenasal (intertrabecular) cartilage ; and these two rows are 
precisely parallelled in the snout of certain living species of Rhino- 
batus (e.g. R. granulatus). But immediately at the base of the 
rostrum, where the cartilage is particularly firm and expanded into 
two lateral elevations (Davies, fig. 2), the tubercles become densely 
clustered in a manner not observable in the existing form ; and this 
arrangement is in intimate relation with the overlying spine. The 
disposition of the tubercles along the trunk, evenif originally regular, 
is now no longer evident, and none but scattered examples are to be 
seen ; but the slender tail was provided on each side with a longi- 
tudinal row of comparatively large recurved hooklets, upon inconspi- 
cuous bases, as is very well shownin the female, no. III. (fig. 7). A 
small tuft of these dermal structures also occurs at the extremity of 
each clasper in no. I. (fig. 1, 24), aud there are distinct indications 
of a patch of very minute prickles upon the membranous portion of 
the (right) ventral fin in the same specimen. 
In regard to the rostral spine, Davies’s figures and descriptions 
leave little to be added. The conclusion as to its absence in certain 
individuals (females) is confirmed in an interesting manner by the 
fossil no. I1I., which has been so ‘‘ developed’’ on the dorsal aspect 
that there cannot remain the slightest doubt upon the subject. But 
a new specimen, from the Enniskillen Collection (no. V. fig. 5), still 
further demonstrates its prehensile character in the individuals that 
possess it ; for a number of blunt conical tubercles, without radiated 
bases, are clustered together upon its inferior aspect (A) to oppose 
the group of more slender hooklets already described at the base of 
the snout. When well preserved (as in no. I.), the surface of the 
spine exhibits the reticulate impressions of the vessels in a once 
enveloping integument *; and on each side there is a marked longi- 
tudinal groove (fig. 5, g), which gradually disappears on approaching 
the distal extremity. 
The peculiar form of the spine is also worthy of note, more parti- 
cularly as it is repeated in two other cartilaginous fishes whose 
remains have been found in the same geological formation ; it differs 
but little from that of the rostral appendage in the chimeroid 
Ischyodus*, and is still more similar to another Liassic spine which 
there is some reason for suspecting may belong to the remarkable 
Prognathodus*. ‘The peculiar shape, indeed, taken together with 
1 Mr. Boulenger has kindly helped me to determine that the corresponding 
appendage in the living Chimera monstrosa is likewise covered with skin. 
* Sir P. Egerton, “ On a new Chimzroid Fish from the Lias of Lyme Regis 
(Ischyodus orthorhinus, 3 ),’ Quart. Journ. Geol, Soe. vol. xxvii. 1871, pp. 275- 
278, pl. xiii. 
3 Sir P. Egerton, “ Prognathodus Giintheri (Egerton), a new Genus of Fossil 
Fish from the Lias of Lyme Regis,” Quart. Journ. Geol. Soc. vol. xxviii. 1872, 
pp. 233-237. 
35* 
