The influence of generation (under field conditions) and photoperiod (under laboratory conditions) on Lobesia botrana larvae development was studied. Some larvae were collected during three annual generations in two grape-growing areas of northeastern Italy, and others were individually reared in the laboratory from egg to pupa on an artificial diet under two different photoperiod conditions (respectively, daylight 16 h/d [long day {LD}] and 14 h/d [short day {SD}]). The mandible lengths of collected larvae were measured and the data analyzed morphometrically to determine the number of larval instars. In the laboratory study, the number of larval moultings, the mandible length of each instar, the development time from hatching larva to pupa, and the pupal weight were considered. The measurement of mandible lengths of larvae collected in the field indicated the existence of five larval instars in all three annual generations, but the size of the two oldest larval instars was significantly higher for third-generation larvae than for the previous generations. Under laboratory conditions, the larvae usually exhibited five instars, but the mandible lengths of larvae and the pupa size were greater for individuals reared under SD. These also took a greater number of days to develop from hatching larvae to pupae. Because a larger size of the final larval instar occurs in individuals that produce diapausing pupae under SD in both the laboratory and the field, a positive association between larval size and the probability of surviving the winter can be inferred.