oS 
Introduction : V ariation. 
in a certain direction (as a complete ceasing from all variation, even under 
unaltered conditions, is inconceivable in the course of the propagation of organic 
forms), or they may be caused by local influences. For some cases the former 
assumption appears unavoidable; for other cases there is satisfactory evidence 
of the effect of local influences, though the exact nature of these latter is almost 
always uncertain; as a rule, probably, both causes operate together, but it very 
rarely happens that an opinion either way is permissible at present. In illu- 
stration of independent development in the same direction it may be mentioned 
that the genus Loriculiis has produced two species with very similar red crowns 
in Ceylon and in the Southern Philippines; and a corresponding distribution of 
markings is seen in the plumage in the two forms of Celebes [L. stigmatus) and 
Sooloo [L. honapartei), which are not closely related to one another (see p. 163 
et seq.). Local influences are sufficiently indicated when a number of sjrecies 
are found to vary in the same place in a corresponding manner; for instance, 
two phases of modification have been detected by Mr. Allen among North 
American birds and are recapitulated by Professor Newton (Diet. B. 1S96, 
p. 1005) as follows: there is “a general increase of intensity toward the south 
and development of dark markings at the expense of the light intervening 
spaces, so that of brightly-coloured species southern individuals are the most 
brightly coloured, and some tints, which to the northward cannot be called 
brilliant, become vivid in a lower latitude. In respect of longitude Variation 
occurs with like regularity, the differences appearing to hold a direct relationship 
to the humidity of the climate. Thus on the dry plains of the middle and 
western parts of the continent birds have a pallid complexion, while on the 
Pacifle slope they resume nearly the tints of the eastern form, though further 
to the northward, in the rainy belt that extends along the coast of British 
Columbia, they acquire a depth of colour far in excess of that which they 
display on the Atlantic border.”^) In such cases the direct influence of climate 
upon the colours appears to be proved. 
The following instances of correlated geographical variation are noticeable 
among the birds of the Celebesian Province. 
Increase of size in the Sangi and Talaut Islands. — The local species or races 
of Sangi and Talaut having their nearest affinities with species on the mainland 
of Celebes are: 
1 ) This quotation, which we have reprinted, is only from a resume of Mr. Allen’s original memoir: 
“On the Mammals and Winter Birds of East Florida” (Bull. Mus. Comp. Zool. Cambridge, 1870 — 1, n, pp. 161 
— 450, plates IV — VIH) which should be consulted here, chiefly pj). 239 sq., where he says: “Causes of 
Climatic Varation. — ... The southward increase in depth of color and in iridescence in birds specifically 
identical coincides also with the general increase in brilliancy of color in birds, taken as a whole, in the lower 
latitudes (as well as in insects and animals generally), the maximum being reached in the tropies. — The 
longitudinal Variation, or the westward increase in color, seems to be also coincident with the increased humidity 
to the w'estward, the darker representatives of any species occurring where the annual rainfall is greatest and 
the palest where it is least . . .” 
