Introduction: Variation. 
63 
The moult. — In the temperate and cold regions of the northern hemisphere 
it is generally admitted that a complete moult takes place in birds in autumn 
after the breeding season ; many species moult again in spring, and some a third 
time in summer. The principal time for moulting in Celebes, Sangi, 
and Talaut seems to be from July to December, when the birds probably 
undergo a complete post-nuptial moult. It is questionable whether signs of a spring 
moult can be found; on the other hand some species may be found moulting 
during most months' of the year. For instance, specimens in the Dresden 
Museum of Heteractitis hrevipes are moulting in January, April, July, August, 
November; of Actitis hypoleucos in January, March, July, November. 
Some of the Waders, autumn visitors from the North to Celebes [Aegialitis 
geoffroyi., Heteractitis)., seem to moult first on the under parts, then the remiges, 
and finally on the upper surface. No regular order in moulting is pursued in 
the Black Sunbirds [Hermotimia), among which the transition from young to adult 
male dress can be particularly well observed. The characteristic metallic sub- 
gular stripe of the Celebesian species makes its appearance first, but the rest 
of the plumage is developed without any such regular sequence, and there is a 
specimen of Hermotimia talautensis in the Dresden Museum (C 15 377) in almost 
complete adult dress except on the forehead and crown, while a second of the 
same species (C 13 847) has the plumage of the adult on the forehead and most 
of the crown, but the young dress on most of the other parts. This proves 
that the transition from the young to the adult dress does not take place in 
perfect phylogenetic order; that is, the adult characters are not necessarily 
developed in the young male bird in exactly the same sequence as 
that in which they were acquired in the evolution of the race (see, 
also, immature male, pp. 469, 471). 
Besides their feathers some birds are known to shed certain corneous 
appendages or coverings on their bills; for instance, the white Pelican of 
America has a horny knob on the culmen during the breeding season, but 
which falls off when that period is over; and the Puffin {Fratercula) moults the 
horny sheath of its bill and the outgrowths over the eyes (Newton, D. B., 
pp. 599, 600). It is possible that a similar moult of the ribbed plates at the 
base of the bill of the Celebesian Hornbill, Cranorrhinus cassidix^ takes place. 
It is believed by the natives to add one rib-plate each year; and, though this 
notion is certainly wrong, it is possible that a shedding of the plates has been 
observed. 
Change of coloration without a moult. — In a recent number of the “Auk” 
(1897, April) Dr. Chadbourne has furnished what seems to be the first really 
conclusive evidence that a change of colour may take place in the perfect 
feather, this being caused by a redistribution of the pigments already present 
in the shaft and barbs. The observations were made on the male Bobolink, 
Dolichonyx oryzivorus (L.), but there can be now no doubt that the principle is 
