242 



BULLETIN 58, UNITED STATES NATIONAL MUSEUM. 



another to record. The carination of the ventrals, gulars, and tem- 

 porals is quite variable, but seems to be strongest in the younger speci- 

 mens; in the one from Sapporo, Yezo (No. 23437), a very large male, 

 even the gulars are smooth. This specimen also has one or two of the 

 central temporals greatly enlarged and smooth. The number of lon- 

 gitudinal rows of dorsals is also subject to considerable variation, the 

 common arrangement being three enlarged series on each side with a 

 single or sometimes double series of very small keeled scales between 

 the two median rows ; these scales in case of a smgle series alternating 

 on the right and left side of the middle line ; in a number of specimens, 

 however, there are only two series of very large scales on each side 

 with a double series of smaller ones between them, but in this case the 

 median scales are considerably larger than where there are three large 

 series on each side. There is also considerable variation in the way 

 the rostral, fronto-nasal, and nasals meet (though in most cases the 



Figs. 209-212.— Takydromus tachydromoides. 2 x nat. size. 209, top of head; 210, middle of 

 back; (No. 23437, U.S.N. M.) 211, top of head; (No. 13753, U.S.N.M.) 212, chin-shields. No. 

 23539, U.S.N.M. 



rostral and fronto-nasal are broadly in contact), as well as in the rela- 

 tion between the fronto-nasal and the frontal, there being sometimes 

 a small unpaired squarish shield interpolated between these two 

 shields and the prefrontals. Finally we may note that the granules 

 between the supraoculars and the superciliaries are subject to con- 

 siderable variation. AVhile in most specimens the anterior elongate 

 superciliary is followed by a continuous series of granules which sepa- 

 rate the second supraocular from the superciliaries, this series is often 

 broken. In one specimen (No. 13753, fig. 211) it is reduced to such 

 an extent that the second supraocular is entirely in contact with the 

 superciliaries, the two granules left only separating the posterior end 

 of the first supraocular from the latter, and in another (No. 34143, 

 one out of a lot of 20 from Mount Fuji) both supraoculars are in con- 

 tact throughout with the superciliaries, being a perfect T. hoJsti in this 

 respect. On the other hand, in one specimen (No. 23437) the one 



