92 
The Food of Birds. 
be incurred where the ultimate good is a certain and lib- 
eral compensation. 
Again, it is becoming evident that carnivorous insects 
and insectivorous birds all have their food-preferences. 
Probably no one species — certainly no one family — of 
birds or insects would quite take the place of another. 
Supposing, then, that some birds eat predaceous insects, 
in part, as well as phytophagous ones — eat the former, 
perhaps, in undue ratio — still, as the chances are practi- 
cally infinite that the predaceous insects it eats would not, 
if allowed to live, eat precisely the same amount and kind 
of injurious insects as the bird itself, by destroying the 
bird we should merely liberate a second cause of numer- 
ous oscillations Those species neglected by the carnivo- 
rous insects would increase beyond their bounds, and 
those eaten by them would be unduly diminished. It fol- 
lows from the foregoing reasoning that, as a general rule, 
ci bird should not be discredited for the regular and 
established habit of destroying predaceous or parasitic 
insects, unless it can be shown that those insects would, 
if left to themselves, check the fluctuations of some in- 
jurious species, or afford a better safeguard against the 
possible fluctuations of others. It must also be shown 
that this prospective good will not be overbalanced by 
some greater evil. In short, the whole burden of proof 
is on the side of those who would disturb the fixed order 
of Nature.* 
The most important question respecting the relations 
of birds to insects is, therefore, the determination of 
those species of birds which serve the most useful pur- 
pose as a constant check upon those insects which are 
either injurious or capable of becoming so if they appear 
in largely increased numbers. Fortunately, whatever 
oscillations or irregularities may arise, and whatever 
may be their cause, the general tendency of things is to- 
wards their correction. In course of time, if new dis- 
turbances do not continually unsettle even the newest ar- 
rangements, they will usually right themselves more or 
* For a discussion of the general subject, see Herbert Spencer’s Prin- 
ciples of Biology, Vol. 2, Pt. VI, Chap. TI, p. 397 ; and the preceding papery 
“On Some Interactions of Organisms.” j ' t ,-v>~r-£w'v»u 
