prefrontal frontal pterosphenoid supraoccipital 

 frontal 



parasphenoid 



maxillary palatine 

 premaxillary 



dentary 



epiottc 

 pterotic 



basioccipital 



suboperculum 



mesopterygoid 

 ectopterygoid metapterygoid I preoperculum 

 quadrate interoperculum symplectic 



Monte Bolca, Italy. The holotype and only known speci- 

 men, IGUP 6890-91, 18.2 mm SL, in counterpart, is small 

 and only moderately well preserved. It is a lateral im- 

 pression and the medial region of the upper and lower 

 jaws is not exposed. Thus, it cannot be told at present 

 how the dentaries and premaxillaries articulate to their 

 opposite members. However, the body is covered by very 

 fine small prickles and the abdominal vertebrae (either 9 

 or 10) are fewer than the caudal vertebrae (12), while the 

 rodlike postcleithrum is in two pieces, all these features 

 being more characteristic of tetraodontids than diodon- 

 tids. There are about 10 or 11 dorsal and anal fin rays, 

 and the caudal fin seems to have 12 rays, one more than 

 in any Recent species of tetraodontid, but the same num- 

 ber of principal rays as in the ancestral gymnodont 

 triodontids and scleroderm eoplectins. Moreover, there is 

 a strong indication that there are slender ribs present on 

 many of the abdominal vertebrae, while ribs are absent 

 in all Recent gymnodonts except the basal triodontids 

 (and one of the two species of scleroderm eoplectins also 

 seems to have ribs). The above are the only features of in- 

 terest that I can distinguish in the specimen, which is 

 sufficiently distinct from all Recent species to merit 

 separate generic recognition, and Eotetraodon is pro- 

 posed herein for pygmaeus Zigno (type-species by mono- 

 typy and original designation). 



The only other tetraodontid fossil materials are jaw 

 fragments found from the Miocene onward, and usually 

 referred to as Tetraodon lecointrae Leriche and T. 

 lawleyi Carraroli. They indicate only that the individual 

 dental units incorporated into the biting edge of the jaws 

 were elongate rods such as in the Recent species. 



Figiire 204.— Canthigaster 

 roBtrata: lateral view 

 of head, composite based 

 on several specimens, 

 55.2-96.5 mm SL, Texas. 



The three external characters that have been 

 prominently used in conjunction with internal features in 

 subdividing the Recent tetraodontids (the number of 

 dorsal and anal fin rays, the form and number of the 

 lateral lines, and the form of the nasal apparatus) will be 

 discussed here first, followed by an analysis of the osteo- 

 logical diversity of the family. In the two most recent 

 revisions of the group (Fraser-Brunner 1943 and Le 

 Danois 1959, 1961a) five families have been recog- 

 nized in what is here considered a single family with two 

 subfamilies, and the term Tetraodontidae as used here is 

 senso lato, although closely corresponding to that in 

 general use in the contemporary ichthyological 

 literature. 



Chonerhinos and Xenopterus have the dorsal and anal 

 fins longer based and with more rays than in other tetra- 

 odontids. In Chonerhinos there are about 23 to 29 dorsal 

 rays and in Xenopterus about 32 to 38, with the anal fin 

 in both cases having several less rays than the dorsal fin. 

 In other tetraodontids the number of dorsal fin rays 

 ranges modally from about 7 to 18, with the anal fin hav- 

 ing either the same number or one to several rays less 

 than the dorsal fin. 



Sanzo (1930) was the first to describe in detail the 

 lateral line of tetraodontids, and Fraser-Brurmer (1943) 

 the first to call attention to its systematic importance. 



