The cricket tentatively identified as Pteronemobius taprobanensis shows a stepwise pattern of latitudinal variation in ovipositor length. Abrupt elongation of ovipositor at about 28° N marks the replacement of the subtropical form by the temperate form. The latter maintains almost a constant ovipositor length up to about 35° N within the bivoltine area. Further north in the univoltine area, an ascending cline extends to about 39° N, beyond which no further increase occurs. The ovipositor length adjusted for body size shows a northeastward increase in each of the univoltine and bivoltine areas, though this tendency is less clear in the latter. The optimum length of ovipositor would vary with the relative amounts of gain (due to the protection of eggs in the soil) and loss (due to the metabolic cost and the difficulty to emerge from the soil) in fitness. If so, it may be predicted that a longer ovipositor would be selected for, when the metabolic cost and the risk at hatching are smaller, the environmental pressure in the egg stage is stronger, and the egg stage lasts longer. The last parameter seems to be mainly responsible for the observed association of the type of life cycle with the geographic pattern of variation in ovipositor length.