prefrontal cartilage pterosphenoid sphenotic epiotic 



ethmoid-vomer 

 palatine 

 premaxillary 



supraoccipital 



^- exoccipital 

 pterotic 



basioccipital 



hyomandibular 



interoperculun^ quadrate metapterygoid parasphenoid 



is some indication that the hypurals were not as fully fus- 

 ed together as in Recent species, with it being possible 

 that there were two more or less separate plates of about 

 equal size, one above and one below, but, again, it is only 

 clear in this regard that there was some sort of groove or 

 less ossified narrow horizontal region in the middle of the 

 caudal plate posterior to the region of the last vertebral 

 centrum. 



Four additional specimens (see Material Examined) 

 that are identified as P. tenuispinus in European collec- 

 tions add little to what is known on the basis of the type- 

 specimens. In MCSNV B.18, 31.5 mm SL, the only items 

 of interest to the present discussion are that, as in the 

 type-specimens, there seem to be 11 caudal fin rays and 

 the trituration teeth are paired, at least one transversely 

 elongate tooth being placed to each side of the midline, 

 and that the spines just behind the pectoral fin base (up 

 to 1.1 mm long) are a little more than twice the length of 

 the spines elsewhere (0.3-0.4 mm long), while the basal 

 plate of the spine has up to six radiations, these being 

 relatively small, about one-half the length of the pro- 

 jecting spine. The scales are similar to this in MCSNV 

 B.19, 40.5 mm SL. 



In the type-specimen of D. erinaceus some of the 

 scales, especially posteriorly on the body, do seem to be 

 three rooted and thus unerectile, although the roots are 

 relatively short and do not approach the massiveness of 

 those found in Chilomycterus. It is possible, however, 

 that the larger size of the roots in the type of P. erina- 

 ceus (77.5 mm SL) in comparison to those of the types of 

 P. tenuispinus (12.6-46.5 mm SL) is at least partially ac- 

 counted for by the differential specimen sizes. The spines 

 just behind the region of the pectoral fin (which is not 

 clear) base are longer (3.3 mm) than those elsewhere (up 



Figure 282.— Diodon holocanthus: 

 lateral view of head, composite based c 

 several specimens, 12. .3-124 mm SL, 

 Gulf of Mexico and Caribbean. 



to L6 mm) on the body of the type of P. erinaceus and 

 may have been two rooted and erectile, although their 

 impressions are so poor that this is difficult to tell for 

 sure. 



Other specimens labeled P. erinaceus in European col- 

 lections (see Material Examined) variously show the 

 transversely elongate trituration teeth in a single series of 

 one or two units to each side of the midline, just as in P. 

 tenuispinus, and spines slightly longer behind the pec- 

 toral fin base than elsewhere and of a small size similar 

 to that of P. tenuispinus, some seeming to have short 

 multiradiate bases and others only biradiate bases, and 

 thus probably some combination of small erectile and 

 nonerectile spines. The number of caudal fin rays and 

 the structure of the last few caudal vertebrae are not 

 clear in any of the specimens assigned to P. erinaceus, 

 but in one of them (IGUP 8670-71) there is a strong indi- 

 cation that the first branchiostegal ray is an enlarged 

 horizontal plate just as in Recent species. 



Since there is the possibility that the Eocene diodon- 

 tids assigned to P. tenuispinus, and by implication to the 

 at least closely related P. erinaceus, differ from the Re- 

 cent species by having 11 caudal fin rays and the 

 hypurals not fully fused to one another, the generic term 

 Prodiodon is retained for them, especially in light of the 

 fact that the numerous diodontid jaws known from the 

 Eocene onward have been given generic nomenclatural 

 status separate from that of the Recent species, as ex- 

 plained below. 



353 



