734 DEE. H. TRAQTJAIR ON 



Incert^e Sedis. 



Family GtEMUKndinid.e. 



Genus Gemiindina, Traquair. 

 Gemiindina Stiirtzi, Traquair. PI. VII. 



Gemiindina Stiirtzi, Traq., Proc. Brit. Assoc, Belfast, 1902, p. 263. 



This is without doubt a vertebrate organism, but its affinities are so problematical, 

 and the appearances present are so difficult, that it is not without much misgiving that 

 I enter into its description at all. 



The specimen is represented in PI. VII. fig. 1, being what I take to be the ventral 

 surface of the creature, while fig. 2, the same slab worked from the opposite side, shows 

 what in that case must be the dorsal aspect. 



The side shown in fig. 1, being the one first exposed and worked out, naturally 

 presents all that remains of the contour of the fossil, as, bearing in mind the thinness 

 and fragility of the remains, it was obviously impossible to lay bare the entire surface 

 on the reverse side. Seen from this presumable ventral aspect, the fish measures eight 

 inches from the tip of the snout to where the tail is cut off by the margin of the stone, 

 and, judging from the rate of attenuation of that part, I should say that at least three 

 inches more would be required to complete the original length. 



The general contour is somewhat like that of a ray, there being on the left side 

 (the right is imperfect) a lateral expanse like that of the pectoral fin in Torpedo, but 

 there is no trace of a ventral. The distance between the outer convex margin of this 

 expanse and the middle line of the fish is 2\ inches, so that were the contour perfect 

 on the left side the entire breadth would be 4^ inches. The snout is bluntly rounded ; 

 the tail gradually narrows till it reaches the edge of the stone, and has still a breadth 

 of § inch where there cut off. 



On examining the fossil (fig. 1) two sets of appearances are distinguishable : — firstly, 

 clear evidence of dermal hard parts covering the entire or nearly the entire surface ; 

 secondly, markings which seem to indicate the presence of endoskeletal structures below 

 the skin. We begin with the latter. 



Immediately behind the rounded snout we see what is certainly strongly suggestive 

 of the two rami of a lower jaw, the articular part of which would be 1^- inch from 

 the front. It is possible that this appearance may be deceptive, but whether or not 

 homologous with a true mandible, the parts exhibited seem, in my opinion, to support 

 the mouth. Behind these parts is now seen an area of a strongly convex-concave 

 crescentic form, the convex aspect being anterior, the concave back posterior, each limb 

 of the crescent ending in a right and left backwardly-directed pointed process. Just 

 on the posterior concave margin of this crescent, and somewhat to the left of the 



