'-^52 MR. E. P. ALLIS ON THE OTIC REGION OF 



both laterally and mesially by portions of the aJisphenoid bone,, 

 the groove thus splitting the dorsal efl.2e of that bone and being- 

 lodged between its two parts. In Albula the trunk-miiscles fill 

 the entire temporal groove. Whether or not this is also true of' 

 Elops I cannot tell from the remnants that I have of the skull of 

 this fish, but a, re-examination of these I'emnants shows that the 

 antei'ior semicircular canal is enclosed in the prootic, and not, as 

 stated in my work on the Mail-cheeked Fishes, in thealisphenoid. 



In Catosiomus and Moxostoma, the temporal groove is said by 

 Sagemehl ('91, p. 550) to be nearly transverse in position, a,nd to 

 open on to the lateral instead of the posterior surface of the 

 chondrocranium. The ej^iotic is said to form the posterior 

 boundarj^ of the groove, and the pterotic its anterior boundary ; 

 and the mesial portion of the groove is said to be roofed by the 

 epiotic, ptei'otic, and parietal. It is not said what muscle, or 

 muscles, occupy the groove in these particular fishes, but as no 

 muscle is anywhere mentioned in relation to the groove, in any 

 of the several species of the Cyprinidse described, excepting only 

 an anterior extension of the trunk-muscles, it is natural to 

 assume that this muscle was considered by Sagemehl to have 

 here invaded it. In Catosiomus, Sagemehl found a greatly 

 reducefl opisthotic, but in Moxostoma succetta he could find no 

 trace of this bone. 



In a specimen of Moxostoma aureolnm, I find the temporal 

 groove directed postero-mesially, practically as shown in Sage- 

 mehl's figures of Catosiomus teres, but its anterior wa,ll is formed 

 by a tall and thin ridge, instead of by a broad surface, as is 

 apparently shown in Sagemehl's figures of Catosiomus. The 

 antero-mesial end of the groove is roofed by the postero-lateral 

 portion of the parieta.l, that bone here Avirlely separating the 

 superficial portions of the epiotic and pterotic. A deep V-shaped 

 incisure, which lies in the axis of the temporal groove, cuts into 

 this roof-forming part of the parietal, the two limbs of the V 

 forming the a,ntero-mesial portions of the anterior and posterior 

 bounding edges of the groove. The reraainrler of the anterior 

 bounding edge of the groove is formed by a rirlge on the pterotic, 

 the corresponding part of the posterior bounding edge being- 

 formed, in its mesial portion, by a ridge on the epiotic, and, in 

 its lateral portion, by the doisal edge of the opisthotic. 'i'he 

 epiotic portion of the edge corresponds to the mesial edge of tlie 

 groove of Scorpcena, and to that edge of the deeper posterior 

 portion of the groove of Scomber, this portion of the e<lge of the 

 groove of Moxostoma ending abruptly in a perpendicular portion 

 which corresponds to the mesial edge of the posterior opening of 

 the groove of the other two fishes. The opisthotic portion of 

 the edge begins at the base of this perpendicular portion, and 

 corresponds to the floor and the lateral edge of the posterior 

 opening of the gi-oove of Scorpcena and Scomher. There is 

 .accordingly, in this edge of the groove of Moxostomcc, a large 

 angular incisure, and it opens directly into the dorsal portion of 



