84 THE RETURN OF THE SWALLOW. 



degree — and consequently it attracts attention by its movements 

 over water- surfaces. Why does it make these movements? 

 Because it finds there an abundance of the winged insects on 

 which it feeds. The swallow skimming over the water and the 

 trout leaping from it have the same end in view — to make 

 a living. The second fact is that in autumn, at sundown, swallows 

 have been frequently observed to gather and settle in large num- 

 bers in the vicinity of water. They are preparing for their descent 

 into the water, it has been said. The fact is that on the 

 branches of the willows and osiers which flourish near water 

 swallows find most convenient perches. On these branches they 

 can crowd together and so secure on a chilly autumn night a 

 measure of warmth which they could not obtain on separate 

 branches or in trees or shrubs of a different growth. 



These birds undoubtedly leave this country with the advance of 

 winter, because their means of subsistence fails and the tempera- 

 ture becomes killing to them; and the few authentic instances of 

 the appearance of the swallow in winter here can easily be 

 explained. They return in spring; and it is to the when, the how, 

 and the why of this return that I shall devote the remainder of 

 this paper. Gilbert White gives the earliest date of their first 

 appearance at Selborne as 26th March; latest, 20th April; usual 

 date he gives as about 15th April. Markwick's earliest is 7th 

 April; latest, 27th. Forster communicated a rather elaborate 

 table of the movements of the hiriindinidcs to the Linnsean 

 Society, and the swallow is reported in it thus: — Naples, 27th 

 February; Rome, 3rd March; Pisa, 5th March; Vienna, 25th 

 March; Bruges, 5th April; and London, 15th April. Mr. Grant 

 Allen says that the average date of their return to the south- 

 western counties of England is the second week of April. A 

 return compiled at the instance of Mr. J. E. Harting appeared in 

 the Field in 1872, and gives the first appearance as 2nd March, 

 four miles south of Glasgow. Mr. R. Gray stated that this was 

 the earliest record of an arrival in Scotland, and as six weeks 

 elapsed between the first and second appearance, I take it that 

 this bird was a restless and adventurous spirit, a very far advanced 

 pioneer of the army of migrants, which either perished miserably 

 or retreated southwards until a more convenient season. The next 

 record in this return is at Cromer, 31st March, and Great Cotes, 



