824 The American Naturalist. [October, 
by many generations of physiologists, each adding some new 
fact or correcting some misconception; and I trust that this 
brief sketch has recalled to your minds the salient facts in our 
knowledge of respiration, and that it will give a just perspec- 
tive and enable me, if I may be permitted, to briefly describe 
what I believe to be my own contribution to the ever accumu- 
lating knowledge of this subject. 
In 1876-1877 Prof. Wilder (6-7), who may be said to have 
inherited his interest in the ganoid fishes directly from his 
friend and teacher, Agassiz, who first recognized and named 
the group, was investigating the respiration of the forms Amia 
and Lepidosteus, common in the great lakes and the western 
rivers. As his assistant it was my privilege to aid in the 
researches and to acquire the spirit and methods, as in no 
other way is it so readily possible, by following out, from 
the beginning to its close, an investigation carried on by a 
master. The results of that investigation were reported to 
this section in 1876, and form a part of the proceedings of the 
Association for that year. From that time until the present the 
problems of respiration in the living world have had an ever- 
increasing fascination for me, and no opportunity has been 
lost to investigate the subject. The interest was greatly 
increased by the discovery that a reptile—the soft-shelled tur- 
tle—did not conform to the generalizations in all the treatises 
and compendiums of zoology, which state with the greatest 
definiteness that all reptiles, without exception, are purely air 
breathing, and throughout their whole life obtain their oxy- 
gen from the air and never from the water. The American 
soft-shelled turtles (Amyda and Aspidonectes), at least, do not 
conform to this generalization, but on the contrary naturally 
and regularly breathe in the water like a fish, as well asin the 
air like an ordinary reptile, bird or mammal (8). 
In carrying on the investigation of the respiration of the 
turtle there appeared for solution the general problem, which, 
briefly stated, is as follows: In case an animal breathes in both 
air and water, or, more accurately, has both an aerial and an 
aquatic respiration like the ganoid fishes Amia and Lepidosteus, 
like the soft-shelled turtles, the tadpoles and many other forms, 
