Birds of Celebes: Falconidae. 
69 
latus should be considered, d'hat the plumage of birds sometimes betrays their 
descent is about as certain as any thing zoological can he. For instance, the 
Kingfishers of the genus Pelargopsw have blue rumps, but the Celehesian species 
has the rump huff; we have received the latter (one specimen) when not yet adult, 
and the rump is strongly washed with blue. The greatest caution is, however, 
required in applying this line of argument, for the plumage of the young seems 
also to be affected by all sorts of other causes not at all well understood. 
There is reason to supj)Ose that the plumage displayed by the young of Pernis 
celebensis and Sp. lanceolatus — a white under surface with dark streaks — not 
only represents that of an ancestral stock common to both these forms, but 
also more or less accurately that of a vast number of other birds of prey. Al- 
most all Falconidae, which acquire a transverse-barred pattern of under plu- 
mage when adult {Astur, Accipiter, Falco peregrinus etc.), are in the first dress, 
assumed on leaving the nest, brown above and white with dark longitudinal 
streaks below. The opposite case — of a cross-barred young one developing 
into a streaked adult bird — would, we believe, be an unheard-of phenomenon, 
though a temporary reversion of the kind sometimes api^arently takes place, in 
Spilornis (see p. 3); as a rule striped birds of prey, like the falcons, have 
young much resembling the females. If these facts have any meaning at all, 
they must signify that all these birds are descended from streaked races of Hawks 
of past ages, similar in coloration, laerhaps, to the Buzzards and Falcons of the 
present day. Consequently there is, at all events, no reason to suppose that 
the Pernes and Spizaeti have been developed from a more remote stock of Fal- 
conidae than that which is now represented by their first plumage. 
We take it that the young plumage, shown on our plate, is very ancient 
in type, though how closely similar it may be to the plumage of the ancestor 
of Pernis and Spizaetus, or how much changed by later influences, we have no 
opinion. From this common ancestor the present full dress of the Pernis and 
Spizaetus of Celebes may have been evolved under one of three conditions: 
1. the structural differences were first developed and then the evolution of the 
plumage proceeded similarly in both and the birds grew alike, or 2. the evo- 
lution of the plumage proceeded on similar lines while the structural changes 
were going on; or 3. the present adult plumage is ancient and had been acquired 
before the structural differentiation commenced. 
It is difficult to form an opinion as to which of these three ex])lanations 
is the more probable; the first may be true or false; there seem to be no me- 
thods for testing it, proving, or disproving it. If we look at Hate III of the 
young of today, and imagine w'e are viewing the birds as they were when adult 
at some period of the past, after the structural change had taken place, we maj 
well wonder why these two birds, structurally so different, should undergo a 
great but identical change of coloration, cannot imagine that two birds 
mav have undergone such a structural change and have assumed such different 
