expanded in the middle about equally anteriorly and 

 posteriorly versus triangular in shape; 5) basihyal ab- 

 sent versus present; 6) pterosphenoids suturing to one 

 another in the midline of the posterior wall of the orbit 

 versus not in contact there; 7) spiny dorsal fin base much 

 shorter than soft dorsal fin base (except Protacanthodes, 

 the bases about equal) versus much longer (except 

 Eoplectus, the bases about equal); 8) epipleurals 

 relatively thick versus thin; 9) basal pterygiophores of 

 soft dorsal and anal fins with lateral flanges along their 

 lengths and sutured to one another distally versus 

 without flanges and not sutured; 10) a single free hypural 

 (probably more in Protacanthodes) versus three or more 

 free hypurals; 11) uppermost pectoral fin ray very short, 

 the lateral half smaller than the medial, versus 

 moderately short, the two halves of about the same size. 

 It is extremely unlikely that this great array of. 

 similarities between balistids and triacanthids in con- 



il spine of Ist vertebra 

 of 1st vertebra 

 centrum of 2nd vertebra 



Figure 77. — BatUtes capriscus: ventral view 

 of sltull, ca. 360 mm SL, Gulf of Mexico. 



tradistinction to triacanthodids has been arrived at in- 

 dependently. 



There are only three important ways in which balistids 

 are more similar to triacanthodids than to 

 triacanthids: 1) the parhypural is free; 2) the haemal 

 spine of the penultimate vertebra is autogenous; 3) and 

 the caudal peduncle is relatively shorter in balistids and 

 triacanthodids than in triacanthids, except 

 Protacanthodes. These specializations of the Recent 

 triacanthids are associated with their more rapid sus- 

 tained swimming in comparison to balistids and 

 triacanthodids, and it must be assumed that balistids 

 evolved from triacanthids at a level of organization less 

 specialized than that of the Recent species, probably at 

 about the level of Protacanthodes. 



Although balistids (and, by implication, the 

 monacanthids and ostracioids, which see) are clearly 

 derived from triacanthids, the number of ways in which 

 balistids differ from triacanthids is impressive. Most of 

 these differences are associated with the development in 

 balistids of: 1) a locking mechanism of the first dorsal 

 spine through the agency of the second spine; 2) an even 

 heavier skull structure for greater buttressing of non- 

 protractile jaws adapted to an even heavier crushing or 

 biting function of the teeth than found in triacanthids; 3) 

 a rudimentary but complex pelvic fin-ray element; 4) 

 heavier scales and less flexible body for less powerful sus- 

 tained swimming but with greater delicacy of move- 

 ment. 



The principal differences between balistids and 

 triacanthids, summarized and modified from Tyler 

 (1968), are as follows: There are usually six, and rarely 

 only five, dorsal fin spines and usually four, rarely three 

 or five, basal pterygiophores supporting them in 

 triacanthids, while in balistids there are only three 

 spines supported by two basal pterygiophores, although 

 the supraneural element of balistids is homologous to the 

 third basal pterygiophore of triacanthids, as discussed 

 subsequently. 



In triacanthids there is an autogenous mechanism of 

 locking the first dorsal spine in an erected position in- 

 volving the apposition of its roughened ventral end 

 against the similarly roughened dorsal surface of the first 

 basal pterygiophore below it, the first spine having an 

 anteroposterior canal through its base which rotates 

 around a high dorsal pronglike flange on the basal pteryg- 

 iophore, the locking mechanism not directly involving 

 the second spine. In balistids the locking of the first spine 

 is not autogenous, depending on a roughened area on the 

 slightly convex anteroventral surface of the second spine 

 fitting into a roughened medial concavity on the postero- 

 ventral surface of the first spine, while the first spine has 

 no anteroposterior canal through it and the basal pteryg- 

 iophore no pronglike flange. 



In triacanthids the first basal pterygiophore is of sim- 

 ple structure, being a stout vertical shaft articulated 

 between the lower rear of the skull and the neural spine 

 of the first vertebra, except in the Oligocene Crypto- 

 balistes in which it has only a short ventral shaft and ar- 

 ticulates high on the rear of the skull and apparently out 



132 



