Morningness-eveningness preference by the self-rated Morningness-Eveningness Questionnaire (MEQ) has earlier been shown to be associated with the subjects’ season of birth. Here, we obtain this result for a new sample of 2,125 university students and for the sample obtained by pooling the data with the earlier study, yielding totally 3,709 Italian and Spanish subjects. An nonlinear regression of MEQ as a cosine curve according to the month of birth, adjusting for age and gender, gave a maximum (morningness) around the transition between the birth months December and January, and a minimum (eveningness) around the transition between the birth months June and July. Multiple logistic regressions showed that for females as well as for males, the group born during the half-year April to September containing summer had a significantly lower proportion of morning types as compared with the group born during the half-year October to March containing winter. This was more pronounced for males. Moreover, a significantly higher proportion of morning types among females compared with males was found only in the group born during April to September, but not in the group born during October to March. There was a weak but statistically significant positive correlation between MEQ and age in the sample’s limited age range of 17–30 years. We discuss the results in terms of the mutually inhibitory systems of melatonin and dopamine, and find further support for a hypothesis that it is the variation in the length of photoperiod during the gestational or perinatal period that contributes significantly to the season of birth variation found in the morningness-eveningness preference among adults.