Whitney, on the Metamorphosis of the Tadpole. 49 
cannot obtain a clear view of the latter. W ith much care 
and pains and many trials, I cannot boast of having accom- 
plished this task perfectly more than three or four times. On 
these fortunate occasions, however, the vessels were com- 
pletely unveiled and uninjured, and the entire plan of their 
distribution and connections clearly displayed. 
Each internal gill (by which expression I mean the entire 
branchial organ on either side) may be said to consist of car- 
tilaginous arches (fig. 13, PL IY, a), with a piece of additional 
framework (b ) of a solidly triangular form, stretching beyond 
the arches, and composed of semitransparent, gelatinous look- 
ing material. These parts, forming the framework of the 
organ, support upon their upper surface the three rows of 
crests with their vascular network, and the main arterial and 
venous trunks which lie parallel with and between them. 
The vascular system displayed by laying bare (in the manner 
already mentioned) the internal gill, is seen in fig. 13. The 
three systemic arteries (a, b, c) arising, right and left, from 
the truncus arteriosus, enter each gill on its cardiac side, and 
then follow the course of the crests, lying in close proximity 
to them. The upper of these branchial arteries runs along 
on the outside of the upper crest ; and if the operator has 
succeeded in clearly exposing the vessel without injury, he 
will detect a branch (c) leaving the trunk and passing into 
the network of the crest, whence a returning vessel ( d ) may 
be traced carrying back the blood across the branchial 
artery, and conveying it to another vessel lying close to and 
taking the same course as the artery itself. Carrying the eye 
along the latter vessel we find, at a short distance from the 
first of these crest branches, a second (e), which leaves the 
main trunk and enters the crest, whence a corresponding re- 
turning vessel conveys the blood across the arterial trunk 
into the vessel lying beside it, as in the former instance. A 
succession of these branches (each taking a similar course) 
may be traced from one end of the crest to the other. Put it 
is now to be observed that the trunk from which these arte- 
rial branches spring diminishes in size as it proceeds in its 
course (like the gill artery in fishes), while the vessel run- 
ning parallel to it and receiving the stream as it returns from 
the crest enlarges in the same degree.* Thus, the artery or 
afferent vessel which brings the blood to the gill is large at 
its entrance, but gradually diminishes and dwindles to a 
# The latter corresponds to the branchial vein in the gill of the fish ; only 
the arrangement of vessels differs in the two cases, inasmuch as in the 
tadpole the arterial and venous currents flow in the same direction, while in 
the fish they travel in opposite directions. 
