458 McAtee, Contents of Bird Stomachs. [oct^ 



will exceed the former in bulk. If only the tibial plate of a grass- 

 hopper is present, a mosquito will far exceed it in size, and will get a 

 correspondingly greater percentage. Under the numerical system 

 we would write down one grasshopper and one mosquito, and none 

 could guess that the mosquito was a real component of the present 

 meal, and the grasshopper a mere trace of a meal gone by. A small 

 fl,nt may thus surpass in bulk the remains (accessory genital glands) 

 of several much larger moths, and so on. Long series of stomachs, 

 taken in many localities at all seasons, tend to smooth over the 

 irregularities due to differences in size. This happens because the 

 present meal being greatest in bulk always gets chief recognition, 

 and past meals represented by mere traces, receive little or no 

 percentage valuation. A large series of stomachs yields many 

 present meals of all the important food elements. Each of these 

 elements therefore is represented in the stomachs by numerous 

 freshly taken specimens which receive full percentage allowance, 

 as well as by residual traces which add little to the total propor- 

 tion. Approximate proportionate representation is thus assured. 



On the other hand if we use the numerical system, an insect or 

 other food item receives the same recognition if represented by a 

 mere trace as it would if entire. In summing up the food of a 

 number of individuals therefore, instead of getting a cross-section 

 as it were of the various typical meals, we get records of the more 

 durable elements of meals piled on each other, until an entirely 

 false idea of the food is obtained. 



Dr. Fisher justly observes that this is not the case with the 

 majority of the birds of prey, which disgorge the less digestible 

 remains of each meal, leaving the stomach entirely empty. It is 

 probable that a numerical system is better adapted to stating 

 the food of these birds than of any other group. 



In the writer's opinion the estimation of the percentage of food 

 items by bulk, logically rests on the firm foundation of a bird's 

 natural requirement of a certain average quantity of food per day. 

 A bird, just as a man, needs a certain food value, or number of 

 calories per day,'^ and in the long run, this is obtained from a 



1 Dieticians make their computations on the basis of weight, and one economic 

 ornitliologist — • Rorig of Germany — lias used dry weights in part of his worlt. 

 But tliese methods consume much time and are probably unnecessary for the 

 degree of accuracy now required in economic work. Estimation of bulk per- 

 centages comes nearest in scientific accuracy to weighing. 



