Fig. 4. Statistics of variation in weight in the Nashville Warbler and Mourning Warbler.
Fig. 5. Statistics of variation in weight in the Yellow-throat and Dickcissel.
Fig. 6. Statistics of variation in weight in the Lincoln Sparrow.
These comments on weight suggest an additional factor which may play a part in rate of migration and which some day may be profitably studied. Suppose that in some species adults and immatures are nearly the same in weight but that immatures have shorter wings. In such a species the immatures are relatively shorter-winged for their weight than adults. In aerodynamic terms, they would have a higher "wing-loading." (Wing-loading is the result obtained by dividing area of effective wing by total weight; it is here assumed that in a single species wing area is directly proportional to wing length.) This being the case, immatures with higher "wing-loading" would require more energy (derived from fat) to fly the same distance as adults, or with the same amount of fat they would fly a shorter distance. Thus they might tend to be outstripped in migration by adults starting at the same time. The reverse, of course, would also be theoretically true, if adults possessed a higher wing-loading than immatures. Physical factors such as these rather than the differential "virility" postulated by Meinertzhagen (1930:56) might account for the arrival of certain classes of some species on the wintering grounds in advance of others. There are, of course, many other factors which must be taken into account before the effect, if any, of the wing-loading factor can be evaluated. Data for illuminating calculations will become available, however, with the accumulation of abundant information on weights, measurements, and migration patterns.
Computations of Longevity and Survival
Tanner (ms., and letter, April 21, 1955) recently devoted considerable ingenuity to computing by actuarial methods the longevity of the Oven-bird, using the adult-immature ratio in samples killed at a ceilometer at Knoxville, Tennessee. Tanner's computations were based on the simple assumption that