METHODS OF STUDYIXd THE FOOD OF BIRDS. 11 



perhaps, from one position, an occupation not nearly so 

 irksome as it looks on paper. 



Wherever an abundance of some particular kind of food 

 occurs, it is a good plan to sit down where you can see with- 

 out being seen and wait for visitors. In this case your notes 

 will take on a different look. Instead of having a bird's name 

 at the head and a list of food items beneath, you will have a 

 food name at the top and the names of birds that partake in 

 the columns below. Thus, you may sit on the shore and see 

 what birds live on fish and what on mollusks. You may 

 stroll across the fields at haying time and discover the birds 

 that feed on the myriads of leaf-hoppers, grasshoppers, and 

 " millers" that take to wing at every step. So may you learn 

 what birds are addicted to any seed or fruit that you may 

 bring under observation. It is well to note in passing that 

 birds are excellent judges of quality in fruits, for which reason 

 it is well to see " which way the birds fly" before selecting a 

 site for operations. 



In the laboratory birds may be kept alive and tested as to 

 their preferences for different kinds of food, though such ex- 

 periments are not likely to be very satisfactory, for the reason 

 that birds in captivity quickly learn to relish things they would 

 never taste in the wild state. 



A knowledge of the amount of food eaten by caged birds 

 is of value, however, 'as whatever difference there may be 

 between the quantity consumed in the wild and in the captive 

 state is on the safe side. A prisoner cannot dispose of so 

 much as the activity of a free bird demands. 



The determination of bird food by dissection requires an 

 extensive outfit, if it is thoroughly done. There must be at 

 hand good collections of botanical specimens, including seeds ; 

 of insects, mollusks, fish, frogs, reptiles, birds, and small mam- 

 mals, everything, in short, likely to be eaten by a bird, in 

 order to name correctly the visceral contents. Even the 

 bones of the smaller vertebrates will be necessary for identi- 



