30 BULLETIN 1165, U. S. DEPARTMENT OF AGRICULTURE. 



* * * This destruction of insect-eating birds over a large part of the State is a 

 serious matter in itself, but evidently similar conditions were maintained over a much 

 larger area through northern New England to New Brunswick, at least as far south 

 as Pennsylvania and as far west as Michigan. A few reports from other parts of the 

 country indicate that the decimation of birds in many other regions was quite as 

 serious as in Massachusetts. 



* * * No one can tell how far the destruction of bird life extended, but a glance 

 at the weather map seems to indicate that conditions more or less similar to those 

 in New England during May existed nearly all the way across the continent. 



* * * Nothing has been heard of any bird catastrophe in the South, and little 

 is known about the effect of the backward spring in Canada, where it may have been 

 very destructive to bird life. The great catastrophe, then, seems to have been con- 

 fined mainly to a large part of a tract about 1,000 miles in width, extending from 

 Pennsylvania and Michigan north into Canada: but just how far, no one knows. 



Because of the stormy weather and the backward season, com- 

 paratively few young birds were raised that year. Cold prevented 

 the hatching of many eggs, and lack of foliage left the eggs and 

 young exposed not only to storms but to their natural enemies. 

 Jays and crows, unable to secure insects, turned to the eggs and 

 young of the smaller birds for food. Heavy rains caused floods, in 

 places doubtless drowning out birds that nest in low bushes or on 

 the ground. 



It is the generally accepted belief that under normal conditions 

 the young birds of one season about equal the number killed from 

 all causes before the young of the next season are on the wing. It 

 becomes evident, therefore, that following such an unfavorable 

 breeding season as that of 1917, the birds affected would return to 

 their breeding grounds the next year in greatly depleted numbers. 



The birds that survived the migrations to and from their winter 

 home again encountered unfavorable conditions when they reached 

 the United States in the spring of 1918. Migration tables indicate 

 that these birds should have arrived in the United States from the 

 middle of March to the latter part of April. According to weather 

 reports for 1918, February and March throughout the Gulf States 

 were warm and dry, but April, except the first few days, was unusually 

 cold and rainy, with excessive precipitation in some places. On the 

 12th a killing frost extended as far south as northern Florida. Such 

 conditions doubtless further decimated the numbers of migrant birds. 



In the case of those species that winter in the United States, the I 

 unusually deep snows of the previous winter might well be considered 

 the principal cause of their scarcity during that summer. It is 

 known that in the vicinity of Washington, D. C, Carolina wrens 

 became very scarce after a heavy snowfall early in February, and 

 there seems good evidence that they perished rather than migrated, 

 since they did not return with the warm weather but continued to 

 be scarce for at least two years. ' After the storm, they were seen 

 searching in unusual places, indicating that food was difficult to find. 

 The same thing has been noted elsewhere in different years with 

 regard to this and other species. 



That this scarcity of birds in 1918 extended into Canada is f 

 indicated by a series of reports from Manitoba. Two tracts with 

 a total of 102 acres, on which counts have been made since 1914, 

 showed in 1918 a decrease in 13 species elsewhere noted as scarcer 

 that year; and a total decrease was noted of 20 pairs, or slightly 

 over 13 per cent. 





