3J: The Wilson Bulletin — Xo. 98 



the larger numbers of " firsts " and in doing- so it becomes 

 very obvious that in the data studied we are at the same time 

 dealing with records showing important bulk movements. 



The data of " firsts " shows a great lack of uniformity of 

 distribution throughout the season. The average season has 

 had 89 days and the number of " firsts " has averaged 120 

 but these have been so bunched that 61 of them have been 

 recorded on 9 days. The extremes are found in the season 

 of 1907 w^hen it took 14 days to record one half of the " firsts," 

 and in 1912 when one half of them were bunched on 5 days. 

 This lack of uniformity in distribution is still more apparent 

 when we examine the records of the last 30 days of each sea- 

 son, during which three fifths of all of the arrivals make their 

 appearance. On the average, 74 " firsts " are recorded in 

 these 30 days and 39 of them are bunched on 4 or 5 days. 

 Extremes occurred in the years 1915 in which it took 8 days 

 to record one half of the " firsts " of the last 30 days, and in 

 the years 1909 and 1916 in each of which it took but 2 days. 

 One half of all the " firsts " of those 14 seasons of 30 days 

 each were recorded on a total of 63 days and we have now 

 to examine the weather conditions which existed on those 

 days. 



For such comparison weather maps are of course desirable 

 and preferably a series of the Washington maps which ap- 

 pear daily including Sundays and holidays. 



An examination of the weather maps for the 63 days on 

 which are recorded a half of all the " firsts " of the last 30 

 days of each of the 14 seasons shows that on 54 of those days 

 there were approaching areas of low pressure, with south 

 winds which had been effective during the preceding night. 

 On 5 days there were southerly winds, or had been at points 

 further south in the state during the preceding night, although 

 an approaching " low " was not well defined. On 3 days the 

 winds were light and either due East or West. On one night, 

 April 30, 1907, there was a rather light northerly wind and 

 yet 5 " firsts " were recorded on the morning of May 1st 

 where none of them had been found on the preceding day. 



