166 
Psyche 
[Vol. 91 
As is evident from Table 2, low-temperature lethality is a rare 
event in this part of California. Although diapause may confer a 
degree of frost-tolerance, its principal benefit in P. zelicaon and P. 
rapae in north-central lowland California seems to be to delay the 
onset of adult development until the bulk of the rainy season has 
passed; thus, adult eclosion coincides with sunny, warm weather 
suitable for flight and hence for reproduction. Pupae accumulate 
enough chilling by roughly mid-January to come out of diapause; 
subsequent development is timed by the weather, which thus 
determines not only the date of first flight but the degree of 
synchronization of the spring brood (Shapiro, unpublished data). 
The unusual 1983-84 winter larvae pupated at about the same time 
as normal diapausing pupae would have resumed development; 
hence it is not surprising that their subsequent development was 
synchronized with the wild population. (On the other hand, dia- 
paused pupae of Pieris napi L. ssp. will develop to the pharate adult 
at a constant temperature of 3°C, while non-diapause ones will not; 
hence in that species, diapausers might be expected to eclose first; 
Shapiro, unpublished.) 
The 3 diapausing zelicaon presumably cannot accumulate enough 
chilling to break diapause in spring 1984 and will thus lay over until 
early 1985. This would expose them to additional risks of mortality, 
and in terms of contribution to the rate of population increase, place 
them at a great selective disadvantage relative to nondiapausing 
members of their 1983-84 cohort (a delay of 4 generations). 
Multiple-year diapause occurs in most populations of P. zelicaon 
even under normal circumstances, although it is rare in the multi- 
voltine populations; it is especially common in univoltine foothill 
races which face unusually unpredictable and stressful climates and 
which would be expected to engage in “risk-spreading” (Sims, 1980). 
On the other hand, although pupae of P. rapae will remain viable 
for two years under constant refrigeration, no multiple-year dia- 
pause has ever been observed in that species under field conditions, 
and there is no indication that it has any physiological ability to 
diapause in summer (Shapiro 1975b, 1980b). Thus any diapause 
pupae produced by winter larvae of this species would presumably 
be doomed. 
In a Mediterranean climate, then, a Type III diapause-induction 
curve permits stragglers at the end of the season to complete 
development and enter the reproductive pool the following spring at 
