14 



A slightly smaller number of species for the spring of 1884 give exactly 

 the same average speed over an average distance of 861 miles. Hence 

 it is probable that future observations will not materially change this 

 estimate. 



A study of the records for 1883 led to the statement that in spring 

 birds migrate more rapidly in the northern portion of their routes of 

 travel than in the southern. As this statement was based on the notes 

 of one year only, it became a matter of much interest to ascertain 

 whether the facts observed would hold good in future seasons and thus 

 admit of formulation as a general law, or whether they had been the 

 results of specially favorable conditions in the latter part of a single 

 season. Accordingly, iu the spring of 1881, twenty -five species of well- 

 known birds, concerning which we had full records, were selected for 

 careful study. The result bears out the foregoing statement. The dis- 

 tance traveled was divided as nearly as possible into two equal por- 

 tions and the speed was calculated for each. Some of the records do 

 not admit of division ; others show an equal speed throughout ; while 

 six show an increase of 77 per cent, in speed for the northern half, and 

 three show a decrease of 47 per cent. Thus it will bo seen that the 

 record is strongly in favor of the increase. The same result may be 

 reached by calculating the average speed of these twenty-five species 

 separately for each of the different months in which migration is per- 

 formed 5 tbe average speed for March is 19 miles, for April 23 miles, 

 and for May 26 miles, per day. The record for 1884 also confirms the 

 statement that tbe later a bird migrates the higher average speed it 

 will attain. This would naturally be inferred from the preceding re- 

 marks. 



These calculations are averages which give the rate of speed at which 

 the bird would travel provided it moved regularly each day. But we 

 know that many pauses occur, that on many days there is no advance ; 

 hence, on the days of movement the speed must be much higher than 

 that given. This is clearly seen in the case of the Purple Martin. From 

 latitude 38° 40' to 46° its average rate is but 13 miles a day ; but we 

 have good reason to believe that there was a pause from April 3 to 

 April 14, and another from April 18 to May 3. Taking out the first of 

 these pauses, the rate is raised between latitude 38° 40' and latitude 

 43° 43 / to 35 miles a day, and, not counting the second pause, the rate 

 for the rest of the distance is 28 miles. 



We must also take into consideration the fact that in all probability 

 the same bird seldom migrates for several nights in succession, but 

 stops to rest after a flight of a night or two, so that the birds migrating 

 one night are not the same individuals that were moving the night be- 

 fore. 



It has been stated above that the average rate for April is greater 

 than that for March, and is exceeded ])y that for May ; but it cannot 

 be said that the actual number of miles performed in a night's journey 



