176 THE AMERICAN NATURALIST. 
investigated by Steindachner, Garman, Eigenmann, and the 
present writer, but with all that has been done there remains a 
residue rich in undescribed species. 
Although the study of recent fishes was only a minor incident 
in Agassiz’s multifarious activities, it was a branch in which 
he felt the deepest interest. His keen and broad insight helped 
him to bring its chaos into order, and his name deserves a large 
place in the history of systematic ichthyology. 
With all his acumen, Agassiz was never able to correctly 
interpret the facts of the succession of the fishes. He said to 
me in 1873: “At one time (about 1842) I was on the verge of 
anticipating Darwinism, but I made up my mind at last that 
progress by transmutation of species could not be, because we 
had our highest fishes first.” 
Of course this is true in a sense, because the ganoid fishes 
are beyond a doubt nearer to the higher vertebrates than the 
recent fishes are, and the early sharks are in some regards 
(brain, reproductive system, alimentary canal, and teeth) more 
highly developed than is the case with the true or modern 
fishes. But neither of these facts, as we now understand, bears 
on the real question. The ganoid fishes constitute a synthetic 
or generalized type, from which at least two great lines of 
descent have sprung. From the double-breathing or amphibious 
ganoids we have descended the Batrachia, who pass through 
the fish-like or water-breathing stage as a phase of youth. From 
these come the Reptilia, who pass through their fish-like stage in 
embryo. And from Reptilia come the warm-blooded mammals 
and birds. The other great line of descent remains aquatic and 
fish-like. Its line of specialization has been to make its members 
more and more intensely fish. The purely aquatic life demands 
not higher development, but adaptive or “ fish-like ” structure. 
The typical recent fish loses its amphibian possibilities very 
early in life, and its development is along the lines of the 
demands its fish life is to make. In the abstract a modern fish 
is not “higher ’* than its ancient ganoid ancestors; nor is it 
properly in most respects lower. It has diverged and is become 
specialized and adapted to its condition in life. Adaptation, 
not progress, is the meaning of organic evolution, 
