142 
fusca, but, unfortunately, as no comparable collections were made 
in the intermediate years, the evidence is not complete. Westcott's 
Maywood collections, alread}'' referred to, on the other hand, hint at 
a two-year -period for L. gibbosa. This species predominated over 
fusca in 1886, as shown by records made on five nights, from the 
26th of May to the ist of June, in a ratio of 3.6 to i, and again in 
1888, in a ratio, as already stated, of 25 to i, but was much less 
abundant than fusca in the intermediate year. 
Food and Feeding Habits oe the Species. 
The white-grubs and the May-beetles, larvae and adults of the 
same insect species, differ totally in their food and feeding habits, 
the grubs eating the roots of various kinds of plants and the beetles 
eating the leaves of trees and shrubs. In the absence of any means 
of distinguishing the species one from another in the grub stage, no 
evidence has been obtained of any special choice, by any of the 
species in this stage, among the various elements of the food of 
grubs in general. So far as known, all of them may take, with 
equal relish, all kinds of food which any one of them will eat. If 
the different species of grubs do indeed make definite and varied 
choice of food, the fact may be ascertained by breeding to the adult, 
for determination, pupae and full-grown grubs obtained in various 
situations near or among different kinds of food plants ; but there 
is at present too little recorded information on this point to permit 
us to infer any difference whatever in the choices of our more 
abundant species. 
The species of beetles themselves, on the other hand, differ con- 
siderably, although not sharply and completely, as a rule, in their 
choices of food where several kinds are equally accessible to them. 
Evidence of these preferences has been obtained by us in three dif- 
ferent ways : ( i ) by experimental feedings with beetles kept in 
confinement; (2) by parallel collections of beetles made from dif- 
ferent kinds of food plants; and (3) by the dissection of specimens 
of various species, made to determine, by an examination of the 
contents of their alimentary canals, what kind of vegetation the}' 
have been feeding on. The first method shows what the beetles may 
eat under stress of hunger, or perhaps with starvation as a penalty 
for refusing what is offered them; the second shows what the dif- 
ferent species actually choose w.hen a choice is open to them ; and 
the third enables us to determine with certainty what the beetles 
have eaten where direct observation is impracticable. The results 
of our feeding experiments have been so different from those ob- 
