摘要："Altitudinal variation within a species is a common phenomenon in the mountains of southeast New Guinea. It occurs both in species inhabiting areas of grassland isolated at different altitudes by forest belts and in birds inhabiting the forest which is continuous from one altitude to another. In nine cases these differences are great enough to use in recognizing races; in nineteen cases these differences are not great enough or there is too great an area of intergradation or overlap to characterize races. In forty-one species in which I have enough material to arrive at a conclusion, there is no altitudinal variation. Increase in size, correlated with increased altitude and consequently lowered temperatures, is the most common type of variation (in eight races and fifteen cases of smaller variation); there was only one race which showed no size variation and one case in which the higher altitudinal population was smaller. Variation of pigmentation was found in seven races and in only three cases within subspecies. Increased pigmentation is usually associated with increased humidity so that only the Subtropical Zone forms would be expected to be darker. This is true in Sericornis nouhuysi, with a paler Upper Temperate Zone race, but is not the case with Ptiloprora guisei, though here other factors may come in. With the other races there does not seem to be a definite general tendency toward lightness or darkness with increase in altitude. Possibly when the climatic conditions in the mountains of southeast New Guinea are known this may help to explain the variation in pigmentation. Though I have few examples, the forest species which have evolved very distinct races at different levels in the continuous forest show a sudden change from one form to the other, as though the cumulative effect of the environment reached a threshold level, beyond which the change was sudden. The grassland races are isolated, so this criterion cannot be applied to them. The one species in which the races are slightly differentiated (Collocalia esculenta) shows a gradual change with increased altitude. In the forms with variations insufficient to recognize as races there is a gradual variation with altitude. The exact levels to which the grassland races are restricted in part may be due only indirectly to climate acting on the vegetation. Temperature appears to be the single factor which changes most constantly in reference to altitude, and however the result is produced the larger size of the higher altitudinal populations appears correlated with the lower temperatures. This appears to be true only for minor variations, the most extreme of which form well-marked races. Variation does not appear to be restricted to any one vertical level but may appear within almost any range of altitude, though in the present work it was not perceptible in less than 800 meters of altitude"--P. 13-14.
