NATURAL HISTORY 



We have vast flocks of female chaffinches 20 all the 

 winter, with hardly any males among them. 



of a rich blue, all the under parts of a brilliant red, the wings, tail, and 

 upper part of the back, green, and the lower part of the back and the 

 rump of a changeable coppery red. When fed upon seed in confinement 

 it loses its brilliancy after the first moult ; the red of the under parts 

 degenerates to a dull pale yellow, the blue of the head becomes less 

 intense, and all the upper parts are of a dull green. Under the same 

 treatment these birds often moult with difficulty, and die. If, in addition 

 to their usual supply of seed, they have melting pears and elder berries 

 given to them, they will moult freely and their natural colours will 

 reappear, on the new feathers, in full brilliancy. Flies and other insects 

 are also essential to them occasionally. 



The linnet and redpole in confinement lose after the first moult their 

 red colour, and it does not return. Is this owing to the want of the pe- 

 culiar food they would take in the spring, if at liberty, or to their being 

 less exposed to the sunshine ? I once saw the English white water lily 

 blow of a pale rose colour after a week of unusual heat in July. 



Birds that change their colours at different seasons, usually put on their 

 bright garb in the warm season. I have repeatedly observed, in a splen- 

 did bird (Loxia Madagascar iensis, LINN.) which I possess, that, although 

 it moults partially twice in the year, the colour of the larger feathers on 

 the wings and back changes gradually from yellowish brown to scarlet, 

 and fades again at the approach of winter. In this bird, the change to 

 red is very clearly occasioned by the increase of temperature. I have 

 observed in the spring that the supervention of cold weather stops its 

 progress In the Whidah bird, the mutation of dress is rapid, accom- 

 panying the moult in June and July. The American blue bird pushes 

 brown feathers in its summer moult, which are very suddenly turned to 

 blue. There is a mystery in these mutations which we do not understand, 

 but they certainly depend in some degree upon temperature. The 

 Whidah bird acquires usually its long tail and fine colours at the vernal 

 moult and loses them in the autumn. It happened one year that the 

 months of August and September had been very cold, and the tem- 

 perature was unusually high in October and the beginning of November, 

 so that with the addition of a fire my room was much warmer at the 

 moment of the autumnal moult than it had been for some time before, and 

 the consequence was that the Whidah bird produced a long tail and 

 coloured plumage again at that season, and continued in beauty for the 

 space of a year and a half. Food has also appeared to me to affect the 

 brilliancy of the plumage, for the nonpareils which had had elder berries 

 or soft pears to peck acquired a deeper red on the breast. 



The Loxia Madagascariensis has been ten years in my room and is still 

 in perfect health. It belongs to a genus quite distinct from Loxia, to 

 which Lox. Oiyx (the Cape grenadier bird), Lox. Phillippina, and Lox. 

 pensilis belong also, as well as two splendid species which have been 



20 British Zoology, vol. ii. p. 306. 



