252 



s. WATASE 



ventral edge slightly furrowed by a median groove. Fig. 8, PI. XIX, 

 shows a tail of this kind. The greater part of it is vertical and median, 

 but its lower portion (C) occurs double. When viewed from the 

 ventral side, a shallow groove extends from the point of attachment 

 down to the extremity of the tail, as represented in Fig. 9(B). 



In some cases, this groove extends further upward, almost as far 

 as the middle line of the height of the tail. In other instances, the 

 groove is still deeper and extends above for more than half the 

 height of the tail. The furrow may reach the dorsal edge of the tail 

 and thus divide it into two halves which then expand right and left. 

 These halves may or may not be connected at the median line, at the 

 dorsal edge. In the former case, the tail is represented by a more 

 or less horizontally expanded single piece. In the latter it is dis- 

 tinctly paired. I shall soon return to these various forms of tails 

 when considering their skeletal elements. 



The remark may not be out of place here that it is only the 

 ventral lobe of the caudal fin (F. L., Fig. 7) that is liable to become 

 split into lateral halves. The dorsal lobe, i. e. that part lying dorsal 

 to the notochord (D. L., Fig. 7), has as yet never been met with in a 

 paired condition. 



It will be necessary first of all to get acquainted with the bony 

 structures of a normal tail fin. A close study has shown that the 

 vertical and median forms of the caudal fin of gold-fishes have the 

 bones arranged identically as in other Cyprinoides, such as a carp or 

 common Carassius. 



I shall give here a brief description of the caudal skeleton of a 

 carp (Fig. 6), the large size of which affords a greater facility for 

 study than that of a gold-fish. 



The few hindermost vertebrœ, as is well known, differ some- 

 what in form. Thus, the centrum of the autepenultimate vertebra 



