Ik) S. HIROTA. 



Seen from tbe dorsal yide there exist at least two whorls of old 

 scales behind the level of the crotch between the two tails. They are 

 followed by a whorl or two of rather ambiguous nature, but taking into 

 consideration other facts to be directly mentioned, I think I am justified 

 in assuming that the regenerated portion of the main tail begins with 

 the third whorl behind the crotch. On the ventral side (fig. 2) the 

 matter is clearer. Here we see the rows of old ventral scales slightly 

 deviating from their proper course, undoubtedly as a consequence of the 

 origin of the accessory tail, and from the third whorl behind the level of 

 the crotch, there is a sudden decrea.se in the number of ventral rows. 

 This decrease, which evidently consists in the absence of the tvk'o rows 

 on either side of the median row in the old portion of the tail and 

 characterized by conspicuous black spots on every alternate scale, is only 

 explicable by assuming that here the regenerated portion begins. In 

 the first whorl behind the crotch, there is seen, on the right (as seen on 

 fig. 2) of the shghtly displaced median scale, the last spotted acale of 

 that side. On the left side, a corresponding scale is wanting ; probably 

 it had been lost, the position wliere it should be, being partially taken 

 up by the root of the accessory tail. The second whorl, like all scales 

 of the regenerated portion shows no spots ; but this fact does not in- 

 terfere in referring that whorl to the old tail portion, or in fact, to the 

 last that was left to the tail when the rest of the latter was severed 

 away, since a reference to more anterior parts at once shows that no 

 spotted scale is expected in this whorl. 



The above assumption finds corroboration internally in the fact 

 that the transition of the normal vertebral column into a sim])le 

 cartilaginous tube, known since Cuvier as the characteristic axis of the 

 regenerated tail in lizards, takes place at approximately the same level 

 as the end of the second whoi'l of scales behind the crotch, i. e. a short 

 distance behind the level maked (6) on fig. '2. 



In fig. 5, I have represented a cross-section passing through about 

 the middle of the origin of the accessory tail, i.e. at the level of the line 

 (5) in fig. 2. At this position, as will be seen from the figure, there is 



