12 ON THE MIGRATION OF BIRDS. 



supply of food for other climates, obliges them to quit those places 

 which fail at certain seasons to furnish them with the means of subsist- 

 ence. I think I have traced the separation of families, and their re- 

 union in bands, of ages more or less equal, to a very natural cause, 

 produced by the difference in the time of moulting in the old and 

 young; and this also appears to be the cause that the bands composed 

 of adults, migrate to a much greater distance, whether in autumn or 

 at their return in spring, than the bands composed of young ones, 

 which do not, in either season, extend their journey so far. The plu- 

 mage of these birds being still imperfect, and the colours not yet du- 

 rable, they are generally one or two years old before they are in a state 

 to breed ; they then choose those places where adults of their own 

 species do not build their nests, the latter always expelling them from 

 the districts which are to give birth to a new progeny. When the old 

 ones extend their journey to the arctic regions, those of one or two 

 years old are found in the middle countries of Europe ; and when the 

 old ones choose the temperate climates, the young ones remain at the 

 south, or at farthest do not pass the seas which separate Europe from 

 the northern parts of Africa ; countries in which the greater number of 

 the largest species of our erratic birds, that do not perfect their growth 

 within the first year, choose to reside in winter. It is from these coun- 

 tries, or the numerous islands of the Archipelago, and those of the 

 Mediterranean and the Gulf of Venice, that they set off on their return 

 in the spring; numerous flocks are then seen on all our southern 

 coasts, especially where the sea forms large gulfs, such as the Archi- 

 pelago, the Adriatic Gulf, and those of Genoa and Lyons. These 

 meetings continue eight, ten, or at most fifteen days ; in which time 

 the passage of those countries is completed. 



The routes taken by water fowl and birds which frequent marshes, 

 depend very much on the course of rivers, and the beds of the great 

 lakes ; the waters furnishing to each species its proper food, they seem 

 to be impelled, by a wonderful instinct, to choose for a rallying point 

 and place of departure, those spots whence the passage from the 

 great sea to the lakes and rivers is shortest and least occupied by 

 land. Thus the bands that assemble in the environs of Genoa 

 and Lyons, repair forthwith to the banks of the Po ; following 

 afterwards the passes of 'the great valleys of the Alps, which de- 

 scend into Piedmont, they rise above the mountains, where different 

 species of the birds in question are annually killed. From these points 

 they appear to direct their flight towards the great lakes of Switzer- 

 land, particularly that of Geneva, which all the water and fen birds of 



