No. 405.] DEVELOPMENT AMONG ANURA. 793 
air-breathers, were seen quite out of reach of the water, in 
small spaces which they had cleared in the ground. The legs 
were full-grown, the tail undiminished, the jaws toothless and 
cartilaginous and ‘some quite larval in form." 
Two species of Dendrobates resort to a device by which it is 
supposed that the young are transferred from one pool to 
another. The accounts are sometimes referred to as if they 
related to a single species. The species closely resemble 
each other, except in size; the larger, Dendrobates trivittatus, 
is found in Dutch Guiana, the smaller, Dendrobates braccatus,? 
occurs in western Brazil. These frogs and Phyllobates trini- 
tatus from Trinidad and Venezuela have the habit of carrying 
their tadpoles attached to the back, whether by suckers or by 
a viscid secretion from the parent, or by both, has not been 
determined. It has been observed in the case of D. trivittatus 
that the frog spawns in water, and that the free-swimming 
tadpoles attach themselves to the parent. They adhere firmly, 
since they were not brushed off when the parent was hotly 
pursued and crawled rapidly through the grass; the frog on 
this occasion was found far from water, and emerged from the 
grass and bushes freshly wet with rain. 
D. braccatus is found on a table-land where, even in the wet 
season, the water may dry off in two or three days. 
The tadpoles of both species possess the usual larval 
organs, including the horny mouth-parts. The intestine of the 
tadpole of D. trivittatus is said to be shorter and less coiled 
than in the ordinary tadpole, and to contain yolk. 
In the case of Phyllobates it is the male frog that trans- 
Ports the young, but the sex of the frog is not recorded in 
either case of Dendrobates. The tadpoles of Phyllobates are 
ranoid, and the tail is about twice as long as the body; the 
anus is dextral, and the spiraculum sinistral. 
! Dendrobates trivittatus has been wrongly called by one observer (Wyman, 
'59) Hylodes lineatus. 
? Smith, in describing Dendrobates braccatus, reports that the natives of Brazil 
brought him from the Santarem forest a similar, though larger, frog and affirmed 
that, like Dendrobates braccatus, it transports its young. This may have been 
Dendrobates trivittatus, which is described by Wyman and others, since the differ- 
ence between the two species is one of size. 
