Whitney, on the Metamorphosis of the Tadpole. 51 
accompanying fig. 14. I would first observe that in the 
post-mortem preparation (fig. 4) there is an obvious conti- 
nuity of structure between the bed of inner tufts and the 
outer gills. This observation prepares us for understanding 
the vascular connection between them. In accordance, then, 
with what soon becomes visible, we feel assured that the 
three systemic vessels a, b, c, in their course to the outer 
gills, run at the base of the three double rows into which the 
inner gill-tufts are divided (fig. 14, in which a few tufts 
in one row only are represented, for the sake of simplifi- 
cation). As soon as the arteries arrive at the projecting 
external gills, they become visible, are seen to run along on 
one side, and supply each division of the gill with a branch, 
which, having reached the extremity, forms a loop, by which 
the blood returns on the opposite side by a corresponding 
vein that travels parallel to the artery. With a frds power 
several small communicating branches may be seen between 
the vessels, perhaps as a proviso in case of accidental obstruc- 
tion to the main current. Now, it follows from the previous 
considerations that the branchial arteries, in their course to 
the external gills, become connected with the gill-tufts of the 
inner gills in the following manner : — Incipient twigs, given 
off from the main trunk, and represented by the dotted lines 
e,f, g, h, pass into the adjoining digit-like processes, each 
forming a loop by which the current returns, and then, leav- 
ing the digit, passes into the incipient upper branchial vein 
(i), which is thus filled by the aggregate of these tributaries. 
By close observation you may detect this vein in its early 
stage (before the disappearance of the outer gills) as a thin red 
line, taking the curved sweep of the future large vessel (the 
cephalic artery ) into which it becomes developed. In like 
manner the middle and lower branchial arteries must be con- 
nected with the second and third double rows of tufts, whence 
the blood returns to fill the incipient trunks which, at a later 
stage, supply the remainder of the body. 
Admitting this to be the correct explanation of the vascular 
connection between the incipient inner tufts and the full- 
grown outer gills, it follows that in proportion as the tufts 
develope, and derive to themselves a correspondingly larger 
quantity of blood, so the outer gills, being in the same degree 
deprived of the latter, will have their function superseded, 
while their delicate and feeble structure shrinks, and ulti- 
mately disappears under the gradual abstraction of the vital 
fluid. We may note the change of form with increase of 
size in the tadpole’s body that accompany the simultaneous 
