Whitney, on the Metamorphosis of the Tadpole. 47 
cannot, of course, exhibit their vascularity. Fig. 6 represents 
them with the digit-like processes larger and longer, while 
the outer gills are shrinking and shortening. 
Soon, however, we are able to find some specimens trans- 
parent enough to show these tufts in what may be considered 
their second stage of development. Fig. 12, PL IV, represents 
the internal gill, with the c irculation in the tufts, as seen at this 
more advanced period. Each digit contains a single blood- 
vessel, which travels to the extremity on one side, loops at 
the end, and returns along the other. With a half-inch glass 
the blood-current in these loops is distinctly visible. 
Tadpoles well fed and well supplied with fresh water, 
plenty of light and air, and kept in a warm temperature, will 
at this stage grow rapidly, while the skin now attains its 
highest degree of transparency. At this middle period of the 
metamorphosis the internal gills undergo a further develop- 
ment, enlarge, and present, under the microscope, one of the 
most brilliant and dazzling specimens of living vascularity 
that can be seen. The change from the second stage of de- 
velopment (fig. 12) to this, the perfect condition of the gill, is 
highly interesting and beautiful. To trace and define dis- 
tinctly the anatomical nature of this change is not easy ; but, 
when discovered, our admiration is challenged by the beauti- 
ful simplicity of the means employed. 
Each cluster or tuft of the digit-like processes, seen in the 
incipient condition of the inner gills (fig. 4) is divisible into 
three double rows. In the second stage of their development 
the three double rows are not only clearly apart, but have 
between them red lines (fig. 12) indicating the trunks of blood- 
vessels, which in this stage are in process of enlargement. 
But on examining one of these gills a few days later, when 
it has reached the third or period of complete development, 
the eye is dazzled with a brilliant but confused display of 
dancing globules ; and a maze of rapid crimson currents, 
running in various directions, is seen in place of the cluster 
or tuft of single loops of blood-vessel perceptible a day or two 
before. It is clear that the gill has undergone a remarkable 
change ; and if we get rid of the dazzle and confusion caused 
by the maze of currents, by examining a dead tadpole (having 
first removed the skin that covers the gill — or what is better 
as a preliminary, having immersed the tadpole in chromic 
acid according to Mr. Archer’s plan), we shall find that the 
three double rows of digit-like processes with their simple 
loops of blood-vessel, have become elaborated into three rows 
of crests, each with a cauliflower-looking surface, as seen in 
fig. 7, e. Thus we arrive at the form, but to understand the 
