The Food of Birds. 95 



mates of the ratios each element bears to the other, so 

 that the average significance of the food can be discovered. 

 Practically, this is indispensal)le. Whatever method fails 

 of this, while its resnlts may be interesting, and may have 

 a certain general value, can never afford a basis for any- 

 thing better than indefinite opinion. It can never settle 

 the case for or against the l)irds. 



This method, while by far the best of the three, has its 

 slight disadvantages. Some tilings eaten liy liirds leave no 

 api^reciable trace in the stomach. For example, it is diffi- 

 cult, by this method, to determine with certainty those 

 birds which greatly injure grapes by breaking the skin of 

 the fruit and sipping the juice. This difficulty applies 

 only to liquid food. Other errors may arise from the 

 shorter or longer periods for which different kinds of food 

 will last in the stomach ; but of this we have no proof. I 

 have depended almost wholly on this third method of in- 

 vestigation, because it is evidently the most profitable and 

 reliable, and because the method of cursory observation 

 having l)een resorted to heretofore, most of the recorded 

 facts are due to it. So far as one method could correct the 

 deficiencies of the other, it was desirable that this more 

 tedious and laborious but more fruitful one should be giv- 

 en greater j)rominence. 



The stomachs of birds shot at all times of the year and 

 in all parts of the state, have been preserved in alcohol, 

 each labeled with name, date and locality. The contents 

 of these stomachs were afterwards transferred, for perma- 

 nent preservation, to separate vials, bearing copies of the 

 original labels. They were then examined, bit by bit, 

 with the microscope, with whatever powers were necessary 

 to the fullest possible understanding of each fragment. It 

 has been no uncommon thing to spend half a day over a 

 single Inrd. Full notes of the materials found in each 

 stomach were made on separate slips, and after this care- 

 ful examination an estimate was made and recorded of the 

 ratios of the dilferent elements to the whole mass of the 

 food of each individual. Objects which I was not able to 

 identify have usually been sent to some more experienced 



