6 



The Wilson Bulletin — No. 82. 



could see the comments that have been entered on their reports. 



But to return to the original subject. The average varia- 

 tions of 2.9 and o.-i days already mentioned are the average 

 for the whole spring migration period of March, April and 

 May, but birds are much more variable during the early part 

 of the season when March storms interfere with their uni- 

 form northward progress. The average variation from the 

 normal date of arrival at Lanesboro is for March, -i.O days ; 

 April, 2.7 days, and for May, 2.6 days. At Grinnell the same 

 averages are for ]\Iarch, 3.6 days ; April, 3.6, and May 2.9 

 days. The average of the two places is for March, 3.8 days ; 

 April, 3.1, and May 2.7 days. The average extremes, that is, 

 the difference between the earliest and the latest record for 

 each sp'Ccies is at Lanesboro', for March, 12 days; April, 9 

 days, and for May, 8 days. At Grinnell, Alarch. 10 days ; 

 xA.pril, 10 days, and May 9 days. Average of the two places 

 for March, 11 days; April, 9.5 days, and for May, 8.5 days. 



When the records of single species are considered of course 

 the variations are much larger. The earliest ducks and geese 

 migrating in the unsettled weather of the breaking up of 

 winter show the widest variations, as will be seen from the 

 followins: table of the movements of the Canada Goose. 



SPRING ARRIVAL OF THE CANADA GOOSE. 



S 



■Q S. .§ 



Nebraska, lat. 40°30'. 



Iowa 41° 



Iowa 42° 



Minnesota 43" 



North Dakota 47-'°. . . 

 Manitoba 50° 



Averasj-e 



14 I February 21 1 -hO to 

 9 I February 241 -1-9 to- 



15 I March 4|4-6to 

 8 I March lS|+2tO' 



11 March 27|-|-5to 



15 April (i|-f6to 





DAYS 1 



34 



43 1 



-39 



48 1 



-31 



37 



-23 



25 



-28 



33 



-23 



29 





3(i 1 



I DAY3 



10 

 14 

 12 

 10 

 10 

 9 



11 



