STATE HORTICULTURAL SOCIETY. 14I 



APPENDIX. 



Details of the Food. — In presenting the data upon which the foregoing conclusions 

 are based, two methods were available — either to expand and write out the memoranda 

 for each specimen, printing the whole mass, and leaving each to gather such general 

 ideas as he could by reading ; or else to throw all the facts into tabular form, showing 

 all the details for each species, together with averages, totals and summaries of all 

 degrees of comprehensiveness, so arranged as to enable the reader to bring every state- 

 ment made into ready comparison with the facts upon which it rests. The latter method 

 was adopted, and the following tables are the result. 



It will be seen that a separate table has been given for each species, and that all 

 the totals for the family are collected in two additional tables — one comparing the 

 records of the species for the whole season, and the other comparing the records of the 

 whole family in different months. Although these tables are unnecessarily elaborate 

 for the matter here presented, it was thought best to adopt, in the beginning, a form 

 which could be used throughout the investigation, however far it may be carried. 



Explanation of the Tables. — At the left of each table for a species is a list of all 

 the elements of the food determined, arranged in the form of a synopsis of their scientific 

 classification. Against each of these names appear, in the columns to the right, under 

 the headings for each month, two numbers, the lower of which indicates the number of 

 birds in which the given food-element was found ; and the upper the ratio of the amount 

 of this element to the whole amount of food taken by all the birds of the species exam- 

 ined for that month. For example, in the exhibit of the food of the robin, we find in 

 the column for April, against the name I/arpalintT, the two numbers 8 and .23. The 

 lower of these indicates that of the eleven robins examined for this month (see head of 

 column) eight had eaten Harpalina. The upper figures (.23) express the ratio of the 

 mass of ihiSQ.Uarpalin^E to the whole mass of the food of the eleven robins; i. e., 

 counting the latter one hundred per cent., the former will be twenty-three per cent. 



These ratios were obtained in the following manner: After a minute examination 

 of the contents of each stomach, a careful estimate was made of the percentage of each 

 element to the whole of the food in the stomach, the material being usually reviewed for 

 the purpose. In summing up for the month the percentages of each element found in 

 all the stomachs were added, and then divided by the whole number of birds taken in 

 that month. The averages for the seasons were found by adding these monthly ratios 

 and dividing by the whole number of months represented by specimens. 



A more obvious method of computing the season averages would have been to add 

 the individual ratios for all the birds, and divide by the total number of the latter; and, 

 if the same number of birds had been examined for each month, the result would have 

 been the same as now; but, as many more birds were examined in some months than in 

 others, the effect would have been to give undue weight to the food-record of these 

 months. The column of totals and the column of ratios at the extreme right of the 

 tables relate, of course, to the season as a whole. 



The proximate accuracy of these estimates was frequently verified by repeating 

 them, the variations being too slight appreciably to affect conclusions. 



Although the birds studied were shot during several years, no account was taken of 

 this fact in discussing results. The number of specimens was too small and their 

 seasonal distribution too irregular to permit any comparison between the records of suc- 

 cessive years. For similar reasons, no attempt has been made to compare the food of 

 the same species in different localities. The figures given are consequently to be taken 

 as the average results for several years and for the northern half of the State in general. 

 All reasonable care has been taken to exclude from these tables everything but positive 

 identifications, and a much larger number of probable determinations might have been 

 added. Many specimens, not identifiable at sight, in their mutilated condition, have 

 been laid aside for want of time. The great mass of the matter is certainly not posi- 

 tively recognizable as to species, by any one. A long list of specific names would be, 

 in my opinion, sufficient ground for distrust of the honesty or competence of the inves- 

 tigator. Many additional names will undoubtedly eventually be added, especially 

 among the ants, caterpillars and curculios, few of which I have had time to attempt. 



