TT. Dames — On Squaloraia. 147 



Fossiles, tab. 42.), I found these cliaracters even more marked; 

 for in that specimen, as observed by Dr. Eiley, the upper pro- 

 cess does not lie upon the lower, and this is clearly represented 

 as being wholly covered by the skin. The upper is also shown 

 as having more of a bony or spine-like structure. In both examples 

 the position of the spine is the same ; each has the expanded proxi- 

 mal end, and the same general form. 



This close agreement in these most important points I thought 

 sufficient to prove the upper process to be a true spine, articulated 

 to the head by the expanded posterior end, and that the line of 

 separation is not a fracture, as Dr. Eiley considered it to be, but 

 a true articular end or basal joint. Also, that the spine is not (as 

 supposed by him) "thrust backwards," but occupies its normal 

 position upon the head. 



Nevertheless, other evidence was desirable to confirm and satis- 

 factorily establish these conclusions ; and this evidence was for- 

 tunately at hand. Among some unnamed and undetermined portions 

 of fish remains from the Lias of Lyme in the Museum were three 

 specimens which afforded the confirmation required. The most 

 important of these is an "inferior prolongation " or snout (PL IV., 

 Fig. 2). The specimen is 4J inches long, rather more than four- 

 fifths being covered with the dermal integument, the anterior 

 extremity is not present, the inferior surface is embedded in the 

 matrix, whilst the upper is exposed and well preserved.. At the 

 proximal end is a low transverse ridge, which, if entire, would have 

 been about three-fourths of an inch long, unfortunately only the 

 left half is preserved : the front is slightly rounded, forming a facet 

 to which the spine was articulated. This ridge is entire, and well 

 marked on the perfect example (PI. IV., Fig. 1 r.), its position 

 being on a line with the anterior edges of the orbits. Extending 

 from this ridge to the broad proximal end of the snout is a car- 

 tilaginous interspace of an inch in length, narrowest in the middle, 

 and expanding by a gentle curve towards each end ; it is bare, 

 and has a rather coarse fibrous structure probably for the attach- 

 ment of the muscular tissue which connected it with and served 

 to elevate the spine above (PL IV., Fig. 2 ^.). On the pos- 

 terior half of this interspace, which is the nasal continuation of the 

 cranial cartilage, is a short median crest at right angles to the 

 transverse ridge already mentioned, and in junction with it ; it 

 divides about the middle of the interspace, and ultimately forms the 

 two long cartilaginous processes which together form the snout, 

 similar to that of Bhinobatus. These are round on their upper sur- 

 face, but have a thin keel on their inner or opposing sides ; that they 

 are not connected by cartilage is shown where portions of the upper 

 skin have been displaced, for we there see that the dermal layers are 

 actually in contact. Anteriorly these processes are nearly unifoim 

 in height and thickness, until they approach the proximal end, where 

 they curve gently outwards, rising into two lateral ridges, which 

 again contract as they curve downwards and inwards, and terminate 

 on either side of the interspace (Fig. 2 e.). The skin which 



