214 University of California Publications in Zoology (Tol. 27 
of time geologically speaking, and the spread of an amphibian, even 
one as restricted ecologically as is Hyla arenicolor, could be accounted 
for readily. 
The habit of arenicolor of occupying the surfaces of rocks adjacent 
to and over the actual course of the stream probably permits the 
animals to ‘keep in touch,’ so to speak, with the stream. Were the 
animals to take shelter in crevices in rocks, as Hyla regilla sometimes 
does, a few days of aestivation in such a location during the season 
when the stream was in retreat might mean their separation from 
water to a distance too great to be covered before the toad perished 
from desiccation. 
Presuming for the time being that the spread and reinvasion of 
the species operates in the manner indicated it would seem that the 
deposition of solitary eggs would be of further survival benefit to 
the species. Such eggs would in a sudden increase of water be less 
likely to be damaged than if they were in a large mass and subject to 
the buffeting of a sudden torrent. If it is a regular circumstance for 
the eggs to become attached to dead leaves in the water there would be 
more likelihood of their being carried on the surface of the current 
and hence less chance of their being damaged than if they were 
attached to small rocks in the stream bed. 
The evidence at hand suggests a late date of spawning and an 
early season for transformation, in other w T ords, a short period in the 
critical and to the individual rather dangerous tadpole stage. No 
exact counts of eggs laid by single females are at hand, but the num- 
ber of ovarian eggs seen in several separate specimens would suggest a 
complement of several hundred eggs. This, in conjunction with the 
rather large population of breeding adults which is found along the 
foothill creeks in the spring (and which is not a ‘concentrated’ 
population from over a wide feeding area as in Hyla regilla, but the 
regular population of the creek), would suggest that the mortality is 
high at one or more stages in the life-history. 
It seems reasonable to expect that the seasonal program of Hyla 
arenicolor in the mountain and foothill canons of southern California 
is to a considerable degree controlled by water conditions. The same 
factors which seem to be operative in the case of Rana boy Hi boylii 
(see p. 259) would seem to be of importance here. Freezing of 
the water practically never occurs in the southern California streams, 
but the water supply may lessen so that the creeks retreat up into the 
mountains. The emergency which Hyla arenicolor has to meet then is 
that of drought instead of freezing. Temperature undoubtedly plays 
