This article describes a geographic trend of size variation and its adaptive background. The Emma field cricket, Teleogryllus emma (Ohmachi et Matsuura) is one of the most conspicuous and widely distributed insects between 30 and 440 north latitude in the Japanese islands. This species produces only one generation a year. The nymph hatches in late spring or early summer and matures in autumn. The adult lays diapause eggs in the soil. This pattern of the life cycle is stable throughout the islands, posing a question about geographic adaptation. A conspicuous climatic gradient exists within the range of this species. The annual mean temperature is about 7 C in the northernmost habitat of Hokkaido, while it reaches as high as 19 C in the island of Tanegasima near the southern limit of distribution. Under such circumstances, local adjustments of development are required for stabilizing the univoltine cycle. Otherwise, the life cycle would tend to depart from the established pattern in different climates. A physiological cline has been found in the intensity of egg diapause of this cricket, which is closely correlated with the local climatic conditions (Masaki, 1965). The nymphs of a northern strain give rise to smaller adults after a shorter period of development than do the nymphs of a southern strain, and intercrossing between them produces hybrids of intermediate size (Masaki, 1963). Body size is one of the most variable and easily discernible external characters among the local populations of this field cricket, and yet the broad tendency of this geographic variation has remained to be investigated. Variation in size frequently occurs in many other insects and has been the concern of entomologists in the past. Previously, geographic regularities of such variations have been reported for several species (for example, Dobzhansky, 1951; Huxley, 1963). Nevertheless, the difficulty in determining their adaptive significance is particularly acute (Mayr, 1963), and no principle comparable to Bergmann's rule for warm-blooded animals has been established for insects. This is probably due to the difficulty in distinguishing between environmental and genetic variations of field material. Without this distinction, the real adaptive significance of size trends cannot be recognized and factors of natural selection cannot be inferred with certainty. The close examination of field specimens and the experimental study of local populations in the laboratory seem, in most cases, indispensable for elucidating the genetic and adaptive background of size trends. In the following account such an analysis is attempted.