Vou. 111] VAN DENBURGH—REPTILES—CHINA, JAPAN, FORMOSA 247 
species. This difference is not quite constant, but nevertheless 
it is of considerable aid in separating specimens of the two 
kinds. 
Both species occur in western Formosa from the northern 
part of the island southward at least to Tainan. It certainly 
is most unusual to find in the same area two species so closely 
related yet so constantly distinct* ; and I suspect that it will be 
found that their local distribution is different, either as regards 
altitude or the character of the country. This is indicated by 
the fact that the two were not collected on the same dates even 
where, as at Tainan, both are labeled as from the same locality. 
Otherwise, it is difficult to understand how the two species could 
remain distinct, unless they breed at different seasons. 
Takydromus formosanus always (178 specimens) has three 
postmentals on each side. The ventral rows are never less than 
eight, and may be ten when one of the lateral rows is more than 
usually enlarged. They are strongly keeled. There normally 
are three rows of enlarged laterals, of which the upper corre- 
sponds to the lateral row in T. smaragadinus. The large dor- 
sals always are in three rows on each side. The small rows 
between these often are two throughout, but frequently are 
reduced to one row posteriorly. Rarely there is only one small 
_ row anteriorly, and one or none posteriorly. The posterior 
reduction is much less constant than in the one-pored species. 
The rostral is separated from the internasal in one hundred and 
fifty-eight specimens, and in contact with this plate in twenty. 
The large preanal is a single smooth plate in one hundred and 
fifty-seven specimens, a single plate with two keels in five, two 
keeled scales in four, and two smooth scales in twelve. The 
supralabials normally are six, but may be five or seven. The 
loreal is, of course, separated from the large anterior supra- 
ocular. 
The collection includes specimens from Keelung, Jenshiko, 
Polisia, Kanshirei and Tainan, Formosa. 
Takydromus smaragdinus Boulenger 
This lizard was first described from specimens labeled merely, 
Loo Choo Islands. It has since been definitely recorded from 
* The overlapping of characters found seems to be pure individual variation. 
