18 On Some Interactions of Organisms. 
Hence, as a general thing, the real limits of a species 
are not set by its organic environment, but by the inor- 
ganic; and the removal of the organic checks upon a 
species would not finally diminish its average numbers. 
Among the external checks upon the oscillations of spe- 
cies of insects, the most important are those predaceous 
insects and insectivorous birds which eat a varied food, 
using most freely those elements of their dietary which 
are, for the time being, most abundant. 
When we compare the results of the primitive natural 
order with the interests of man, we see that, with much 
coincidence, there is also considerable conflict. While the 
natural order is directed to the mere maintenance of the 
species, the necessities of a man usually require much 
more. They require that .the plant or animal should be 
urged to excessive and superfluous growth and increase, 
and that all the surplus, variously and widely distributed 
in nature, should now be appropriated to the supply of 
human wants. From the consequent human interferences 
with the established system of things, numerous disturb- 
aiice^arise — many of them full of danger, others fruitful 
of positive evil. Oscillations of species appear, not less 
injurious to man than to the plants and animals more 
directly involved. Indeed, most of the serious insect in- 
juries, for example, are due to species whose injurious 
oscillations have resulted from changes of the organic 
balance initiated by man. 
To avoid or mitigate the evils likely to arise, and to 
adapt the life of his region more exactly to his purposes, 
man must study the natural order as a whole, and must 
understand the disturbances to which it has been subject. 
Especially he must know the forces which tend to the re- 
duction of these disturbances and those which tend to 
perpetuate or aggravate them, in order that he may rein- 
force the first and weaken or divert the second. 
The main lesson of conduct taught us by these facts 
and reasonings is that of conservative action and ex- 
haustive inquiry. Reasoning unwarranted by facts, and 
facts not correctly and sufficiently reasoned out, are 
equally worthless and dangerous for practical use. 
