38 BULLETIN 621, U. S. DEPARTMENT OF AGRICULTURE. 
were obliged to string the top of our poultry range with binder twine (strands 
2 feet apart), the strands being stretched from division fence to division fence. 
In no other way were we able to stop the crows killing chickens. (1912.) 
As the majority of reports give evidence that the poultry-stealing 
habit is one of individuals, it appears that combative and protective 
measures may be employed with success. The shooting of the par- 
ticular marauders has in most cases put a stop to the trouble. Where 
losses to young poultry and eggs were at all severe, the use of ware 
netting to cover yards containing chicks and the provision of proper 
houses for nests have been effective, and the investment has soon 
paid for itself by the absolute protection afforded. The destruction 
of a near-by nest frequently prevents further loss to poultry, and 
crows may be retained throughout the general area to render good 
service to the farmer. 
For additional control measures see pages 73-80. 
SMALL MAMMALS. 
In the destruction of injurious mammals the crow does work seldom 
recognized by the average farmer. While individually it may not be 
so aggressive in its pursuit of a victim as some of the birds of prey, 
the crow^ makes up such deficiencies by its greater numbers. The 
persistence of a squad of the black fellows mobbing a frightened 
cottontail rabbit or patiently searching grass hummocks for nests of 
mice early in spring is distinctly corvine. Stomach analysis gives 
ample proof of this trait, and, while specific identifications can not be 
made in most cases, the fact that mammals have been eaten is 
readily proved. Difficulties arise, however, in the distinction be- 
tween mammal remains eaten as carrion and those secured alive. 
In identifying carrion among the remains of small mammals the 
writer relied entirely on the character of the associated insect remains. 
Necrophagous insects, as Silphidse among beetles and the larvse and 
puparia of blow flies, most frequently supplied this evidence. 
Mammals comprised 1.64 per cent of the yearly food of the adult 
crows examined and Avere eaten in every month. The greatest con- 
sumption was in April, when such food totaled 4.10 per cent of the 
diet. In May 3.24 per cent was recorded; in March, 2.77; and in 
July, 2.60. In all other months the percentage was less than 2, being 
practically nothing (0.05 percent) in September. Mammals occurred 
in 134 of the 1,340 stomachs, an average of 1 in every 10. Of these 
remains, injurious rodents formed the major portion. These pests 
were identified in 82 of the 134 stomachs in which mammals were 
present and doubtless occurred in many others where accurate de- 
termination could not be made. In bulk they formed approximately 
three-fourths of the mammal food, a yearly percentage of 1.22 being 
