STATE HORTICULTURAL SOCIETY. I 27 



in any one State, I should only betray my ignorance of the magnitude 

 and complexity of the problem in hand, and my consequent unfitness to 

 solve it. I shall, therefore, offer the data I have collected only as a con- 

 tribution to the subject, anxious rather to underrate them than to 

 exaggerate their value. We must not forget that the actual facts I have to 

 give you relate only to the contents of the stomachs of forty-one robins, 

 and that the estimates of benefit and injury that I shall read simply 

 express my best judgment of the good or harm done by those forty-one 

 birds at a single meal each. How much of an inference as to the habits 

 of the species we may erect upon this foundation depends upon the 

 fixity or invariableness of the food-habits of this group. If the same 

 species will eat substantially the same food, year after year, in the same 

 situation, then, of course, a good deal may properly be inferred from 

 comparatively few data; but if the food varies widely, either arbitrarily 

 or under slight changes of condition, then we can infer but little. Upon 

 this fundamental question I have two suggestions to make. 



First, if several species allied in structure, occupying the same terri- 

 tory at the same time, living side by side, with the same sources of food- 

 supply open to them, are found, on the examination of a limited number 

 of stomachs, to present certain characteristic differences of food, so that 

 the investigator can point out definite peculiarities of the food of each 

 species, and finds these peculiarities reasonably constant, year after year, 

 then we may say unquestionably, without going farther, that there is a 

 fixity of food-habits in this group of birds which will allow us to reason 

 from the data observed. This is especially true if the species are not 

 mutually hostile, if they do not actually fight, so that a stronger or bolder 

 species may exclude the weaker or more timid from certain situations. 



Second, if there are any other habits of the species in which there 

 does not seem to be any greater reason for invariableness than in those 

 relating to the food, which are nevertheless found to be substantially 

 unvarying, then we may, with considerable force, argue the probability of 

 a like unvarying character in the habits of alimentation. 



Respecting the first of these tests, you will see, when I sum up the 

 food of the family now under consideration and bring the data respecting 

 the various species into comparison with each other, that I have made out 

 certain very well-marked specific differences of food, even among those 

 eating at the same table, that the different species of this group, while 

 agreeing in many particulars in food as they do in structure, present also 

 certain peculiarities, so marked that I can usually determine the species 

 by the contents of three or four stomachs. 



For the second test we may properly use the nesting habit. There 

 seems to be no more urgent reason why one species should select from 

 the same store-house different materials for its nest from those used by 

 another closely allied species of nearly the same size and similar general 

 habits, and building in the same locality, than why each should use a 

 similar fixed discrimination in selecting its food. Yet no expert, scarcely 

 a school-boy even, will hesitate a moment between the nest of a robin and 

 that of a cat-bird; and the descriptions of the two given in the books 



