340 KAMAEICHI KtSHINOUYE : 



develoj^jed, and is found in the precaiidal rt^iou as well in Katsmvonus and Evfhyn- 

 nus (figs. 57, 58) ; hut iu Auxis tlae foramen is found poorly develo^xKl and in 

 a few |X)sterior caudal A-ertebrae ouly. In this family a pair of s^iecial 

 protul^rauces appear iu a few anterior vertebrae. These protulierances lie just 

 tehiud the superior zA^gapoiibj-ses, and tbej serve to fmiiisb points of attach- 

 ment to a pail" of strong tendons of the lateral muscle. 



Iu the Plecostei the uem-al and haeiual spiues and other processes fi-om 

 vertebrae for the mutual aiiiculation ai'e well developed. The neiu-al spine of 

 certain anterior vertebrae is broad and rough for the insertion of muscles, 

 and in bonitos the neural arch of these vertebrae is perforated with numerous 

 pores of different sizes. The other neural spiues are long, slender, laterally 

 compressed, aud nearly vertical to the vertebral column at then origin. 

 The haemal spine is remarkably well develo^ied iu the precaudal region 

 in tmmies ; but iu bonitos the spiue is scarceh" develo^ied in tliis region. 

 However a median spiue of quite new origiu makes its apjiearance in the 

 Katsuwonidae. It was fii-st described by Staeks (69) luider the name of 

 pedicle ; but I proj^wse to name it the epihaeuuil pi'oc«ss. The spine is 

 developed Ijetweeu the centrum and the haemal process or the haemal arch, 

 and is best developed in the ix>sterior part of the precaudal region. The anterior 

 superior zygapophj'ses of anterior precaudal vertebrae are long, more or less 

 lieut inward at the lower margin in the Thunnidae ; l)ut they are more or less 

 triangiüai- pya'amidal in the Katsuwonidae, and there is an accessory zyga- 

 jwphyses as in the Scombridae. The anterosui3erior zygapophyses in the 

 posterior jwrtion of the vertebral column are elongated and flat, both iu the 

 Thunnidae and Katsuwonidae ; but in the latter family the accessory zyga- 

 jx)physes are formed lieneath the ordiuaiy zyga^Mphyses to clasp the pasterior 

 sui^erior zyga^xjphyses between these two zygaixiphyses. In the Thiumidae the 

 inferior zygaixjphyses are short and jwinted at the end, more or less diverging 

 fi'oiu the middle of a veiiebra, and they do not come iuto close contact 

 with those of the next vertebra, as iu the Cybiidae. In the KatsmNonidae, 

 however, the inferior zj'gapophj'ses of a vertebra are long aud in close contact 

 with those of the next vertebrae. 



