94 Incomplete Adaptation as illustrated [ February, 
wholly gratuitous and belongs, as already remarked, to the same 
class of ideas as that by which all changes were once explained 
as the results of great and sudden catastrophes. It is due to the 
kind of reasoning which denies change to everything which can 
not be seen to move—a kind of reasoning which leads the savage 
to deny that the great trees have ever been other than they are,’ 
while admitting growth in the herb and the sapling. In point of 
fact we find nearly all possible degrees of adaptation to the agency 
of insects. The mere existence of colored flowers must be re- 
garded as an initial step in this direction, and the greater part 
of all flowering plants exhibit in a more or less marked manner 
this evidence of the influence which insects have exerted upon 
them. But it is evident that an ordinary hermaphrodite flower, 
however showy or fragrant, if devoid of special appliances for 
preventing self- and securing cross-fertilization, represents a very . 
rudimentary and imperfect state of correlation to the insect world. 
This condition, which is now the predominant one, must therefore 
be regarded as constituting the first step of a long progressive 
series of morphological changes in the same direction, all tending 
to complete the degree of adaptation to insect life. The various 
specializations which a few species have already undergone mark 
so many additional steps taken by such species toward the same 
end and afford a faint idea of what the whole flora of the globe 
might become in the remote future, if wholly uninfluenced by 
man. 
In the great majority of plants, self-fertilization is doubtless still 
the rule, and cross-fertilization the exception, but this occasional | 
crossing, even though very rare, suffices to maintain the vigor of 
the stock, Such plants will appear to thrive as well when self- 
fertilized as when cross-fertilized, and this would probably be the 
case if the experiment were repeated a great number of times, for 
it is not once or a score of times, or a hundred times even, that 
count in these processes of nature, but vast periods and innumera- 
ble repetitions, each with its minute differential to add to or 
subtract from the general sum, When these facts are properly 
understood, therefore, the partial or total failure of all human 
experiments on cross-fertilization becomes nothing more than 
naturalists ought to expect. The really surprising fact in such 
1See an address by Maj. J. W. Powell, delivered before the American Geograph- 
ical Society, at Chickering Hall, New York, Dec, 29th, 1876. 
