The young of 1948, first sampled after their emergence from their first hibernation in mid-April of 1949, were then somewhat intermediate in size as compared with those of 1949 and 1950 at the same times of year. Their subsequent development was rapid; by late May they had caught up and passed the stage reached by the 1949 young at the same time of year. The young of 1950 after having a late start, were further set back by cold weather in April 1951 delaying their emergence from hibernation. As a result they were still unusually small in late April and May. Even though they grew rapidly subsequently, they were consistently smaller than those of other broods on corresponding dates. Favorable fall weather prolonging the 1951 growing season into late September beyond the time of retirement in other years may have permitted many of them to attain adult size. [Illustration: FIG. 16. Records of immature individual skinks marked and recaptured within the same growing season, showing the trend of rapid growth, and differences in growth rate between individuals.] The varying fortunes of the several annual broods studied were closely correlated with weather trends, and suggest possible effects of slight changes in climate. An unfavorable sequence of weather might bring about drastic reduction of the population without causing any direct mortality. A late spring in two successive years would have cumulative effect in delaying emergence and breeding of adults the first year, and delaying in the second year emergence of the young, already retarded by the lateness of their hatching. If this sequence were followed by onset of unusually cool and dry weather in early September, or even in late August, the young might be "caught short," and forced to hibernate while still in the 50-60 mm. size class. Emerging the following spring, they might have failed to mature sexually, reducing by perhaps half the number of productive adults. At the northern extreme of the species' range, length of growing season may be more critical than extremes of temperature in limiting the numbers and distribution. Growing seasons that average long enough and warm enough to permit attainment of maturity by onset of the second hibernation period may be essential to the species. While no two annual broods of young in the same locality come under exactly the same weather influences, extremes of retardation or acceleration continuing throughout development are relatively rare. Retarding effects of unfavorable weather causing delayed breeding and hatching, may be offset by prolongation of warm weather in the fall thus delaying hibernation, or by warm spring weather hastening emergence from hibernation.