Thermoperiodic effects on diapause of Sesamia nonagrioides ( Lepidoptera : Noctuidae )

The role of the thermophotoperiod on larval diapause induction and development of Sesamia nonagrioides was investi­ gated under laboratory conditions. The incidences of diapause by TC 12 :1 2 thermoperiods under DD, in which the thermophase was 30°C and cryophases varied from 10 to 25°C, were compared to the effects of thermophotoperiods with the same configuration. Higher incidences of diapause were recorded when larvae were exposed to thermophotoperiods with cryophases of 15-25°C, than under thermoperiods under constant darkness. In addition, larvae exhibited significantly lower percentages of diapause when exposed to cryophases < 15°C under either thermoor thermophotoperiodic condition. Diapausing larvae when exposed to photope­ riods 16L : 8D h and different temperatures of 25-30°C proceeded to development in 14-16 days. Moreover time needed for the pupation of diapausing larvae after their transfer to 12L : 12D h, was reduced from approximately 49 to 12 days depending on tem­ peratures of exposures (25-30°C). Diapause termination was also compared under 12L : 12D h and constant temperature of 25°C or a thermoperiod of TC 12:12 (30°: 20°). Days needed for pupation at constant temperature or under the mentioned thermoperiodic regime, were approximately the same when the high temperature of the thermoperiod coincided with the scotophase, while with the co-occurrence of the high temperature of the thermoperiod and the photophase significantly additional time was needed for diapause development. Significantly shorter time was also needed for diapause development after exposure of diapausing larvae to continuous light, than to continuous darkness under 25°C. Thermoperiods of different amplitudes with the same mean temperatures proved that the higher temperature of scotophase led to higher rate of diapause development in spite of the relatively low temperature of the pho­ tophase tested. It is suggested that thermoperiod could hasten the time needed for diapause development, but it is not the crucial factor for the termination of diapause.

The need for extensive research into the thermoperiodic influences on seasonal behavior of insects may be impor tant from an ecological and phenological standpoint in temperate regions.One problem encountered in the study of a thermoperiod is its coincidence (or non-coincidence) with a concurrent photoperiod.This reveals further prob lems arising from the alteration of the mean temperature between different thermophotoperiodic regimes.For this reason, most such studies deal with photoperiods of 12L : 12D, or take place under continuous darkness (DD) or light (LL).
Photoperiod has been reported as the major factor con trolling the induction and termination of larval diapause under laboratory conditions, while temperature could influence the response to daylength (Hilal, 1977;Eizaguirre et al., 1994;Fantinou et al., 1995).According to Galichet (1982) this species exhibits an oligopause (see for definition Mueller, 1970) and a temperature of 25°C is effective in hastening diapause development under both long and short-days.
In a recent study, Fantinou & Kagou (2000) examined the effect of thermoperiod on diapause induction of this species and suggested that the photoperiodic induction of larval diapause was enhanced by superimposed thermope riods, particularly when the low temperature coincided Each symbol shows the mean value of 4 replicates of 25 larvae each.
Values labeled with the same letter are not significantly dif ferent at P = 0.05 by LSD test after ANOVA.
with the scotophase.Moreover, thermoperiod per se could induce diapause, whereas both the duration and the temperature of the cryophase were important factors con trolling the diapause response.
In the current study, we attempt to determine the inter acting effects of thermoperiod and photoperiod control ling induction, as well as the termination of diapause and the ecological significance of the different thermoperiods on the diapause development of this species.

MATERIALS AND METHODS
A laboratory colony of the stalk borer, originating from larvae field-collected in 1999 in Kopais, central Greece and kept on artificial diet (Tsitsipis, 1984) for 2 yr, was used in all our experiments.Maintenance and handling of the colony has been described (Fantinou et al., 1995).Larvae were considered diapausing when they did not reach at 25°C larval-pupal ecdysis by the age of 55 d from egg hatch (Fantinou et al., 1995).
Experiments were carried out in small incubators providing either constant or changing temperatures and any photoperiodic regime by daylight fluorescent lamps (8 W).Reported tempera tures were accurate to within ± 1°C, while the transition from one temperature to another was complete within ca 30 min after switching.Observations under DD were made using red light, provided by a combination of red and orange plastic filters allowing light transmission at a wavelength longer than 540 nm.Larvae were kept on artificial diet, which was changed weekly.Four replicates of 25 larvae were used per treatment.
To investigate the different effects of cryophase and cryosco tophase temperatures on the incidence of diapause in S. nonag rioides, neonate larvae less than 24 h old were exposed either to constant darkness and a thermoperiod (Thermophase: Cryo phase) TC 12:12 in which the thermophase was 30°C and cryophases varied from 10 to 25°C, or to a thermophotoperiod with the same configuration.
The effect of constant temperatures (25, 27.5 and 30°C) on diapause development was evaluated by rearing 55 d old dia-pausing larvae under different photoperiodic regimes (LD 10:14,12:12 or 16:8 h).The involvement of thermoperiod on diapause development was examined by measuring the time needed for the pupation of diapausing larvae after their transfer to different thermoperiodic treatments TC 12:12 (25° : 25°, 30° : 20° or CT 20° : 30°), and light conditions such as LD 12: 12 h, constant light (LL) or constant darkness (DD).Moreover, the role of temperature cycles of different amplitude was deter mined by comparing TC 12:12 (26° : 24°, 28° : 22°, 30 °: 20°, 32° : 18° and 34° : 16°), under a photoperiod of 12L : 12D h.Each regime therefore, had a mean temperature of 25°C and a symmetrical thermoperiod of two 12 h phases.Pupation was recorded daily in each treatment.Data were analyzed by using the procedures of a statistical programme provided by Statistical Graphics (1991).Data on diapause incidence (%) were trans formed using an arcsine square-root transformation before ANOVAs were conducted.The diapause response and days needed for pupation between treatments were analysed with analysis of variance (ANOVA); means were separated by least significant difference test (LSD) ( p < 0.05).

RESULTS
The effect of different cryophase and cryoscotophase temperatures on diapause incidence in S. nonagrioides is shown in Fig. 1.The presence of the photoperiod 12L : 12D (which was the difference between the two treat ments) revealed quite great differences in insect response, although there was a similarity in curves representing the two treatments.A negative correlation of the incidence of diapause with the increase of the temperature of the cryophase was also evident, as the temperature rises from 15 to 25°C, whereas lower percentages of diapause appeared at very low cryophase temperatures, such as 10°C, under both lighting treatments.Moreover, significantly higher percentages of diapause occurred under LD compared with constant darkness when cryophase temperatures were greater than 15°C (Fig. 1 and Table 1).
The effect of constant temperatures on different photo periodic regimes on diapause development was compared (Table 2).The progress of diapause development is expressed in days needed for the 50% of pupation of lar vae.Photoperiod had a definite effect on diapause termi nation in S. nonagrioides.Time needed for the pupation of larvae was considerably reduced when larvae were transferred to 16L : 8D or 12 : 12 h, compared with those remained under the diapause promoting photoperiod (10L : 14D h) at 25°C.The data also show that diapause devel opment was enhanced at 27.5°C in all photoperiodic regimes tested.A further great reduction of the time needed for pupation resulted from a relatively small increase in temperature (from 27.5 to 30°C), and no sig nificant differences in time occurred in the photoperiods tested (Table 1 and 2).
Table 3 shows the effect of a thermoperiod of TC 12:12 (30° : 20°) combined with a photoperiod of 12L : 12D h, continuous illumination LL or darkness DD on the termi nation of diapause.Thermoperiod did not have a decisive role in reducing the time needed for the diapause comple tion (Table 2 and 3).Under 12L : 12D h and a constant temperature of 25°C, time needed for the completion of diapause (48.5 d) was comparable with that under a ther-  4 Data on diapause incidence (%) of Fig. 1 were transformed using the arcsine square-root transformation before the ANOVAs were conducted.aAll P were <0,05 moperiod CT 12:12 (20° : 30°) (54 d), in which the high temperature of the thermoperiod coincided with the scotophase of the photoperiod.In contrast, significantly more time (77 d) was needed for the pupation of diapausing larvae when they have been transferred to 12L : 12D h and a thermoperiod of TC 12:12 (30°: 20°), where the cold phase of the temperature cycle coincided with the scotophase.Hence, it seems that diapause persisted when larvae were subjected to thermoperiods where the low temperature coincided with the scotophase.The data also show that under a constant temperature of 25°C, signifi cantly less time was required for diapause development in continuous illumination than in continuous darkness.However, the larvae maintained in constant darkness and a thermoperiod of TC 12:12 (30° : 20°) reached pupation in less time (54.5 d) than those at the constant tempera ture of25°C and DD (70d).The termination of diapause was also evaluated when diapausing larvae were exposed to thermophotoperiods of different temperature amplitudes with the same mean temperature of 25°C (Table 4).Days needed for pupation were significantly reduced with the decrease in the ampli tude of the thermoperiod (Table 1 and 4), i.e. the higher the amplitude the longer the delay before pupation.Fur thermore, the increase in temperature of the scotophase (from 18 to 20°C) resulted in a significant reduction in time for the completion of diapause (Table 1 and 4).

DISCUSSION
From the data described above, it is obvious that ther moperiod had a measurable effect on diapause induction in S. nonagrioides, but this effect was less pronounced than that of the thermophotoperiod (Fig. 1).Furthermore, low cryoscotophase temperatures greatly modified the photoperiodic response.Cryophase temperatures lower than 10o C need not be evaluated since the lower thermal limit of larvae of S. nonagrioides is 10.85°C (Thanopoulos & Tsitsipis, 1989).Fantinou & Kagou (2000) reported that larvae of this species were able to distinguish a long-day thermoperiod from a short-day thermoperiod inducing diapause.Conse quently, the absence of any photoperiodic stimuli could start the free-running rhythm of the circadian clock.The results of the present research indicate, however, that a weaker response (diapause incidence) was demonstrated when larvae were subjected to thermoperiods in absence of light (Fig. 1).The thermophase, therefore, does not seem to function as a simple direct substitute of the pho tophase of the photoperiod as the two response curves differed.Therefore, this difference must be accounted for Table 3. Days needed for pupation (mean ± sd) of 50% of diapausing larvae of S. nonagrioides after their transfer at the age of 55d from 25°C and LD 10:14 h to different thermoperi odic TC 12:12 (25° : 25°, 30°: 20°) or CT 12:12 (20° : 30°) and lighting conditions (LD, LL and DD). by the role of the photophase.Temperatures lower than 15°C provided lower incidences of diapause.It seems that very low temperatures may also fall outside the effective range of the temperature-compensation mechanism or maybe they slow the mechanism.Furthermore, it must be also investigated iflarvae really respond to a short or long thermoperiod and become induced to diapause, or simply accumulate temperature effects on growth and development (Beck, 1983).
The present study showed that photoperiod was the cru cial factor for the termination of diapause of the S. nonag rioides larvae, a finding supporting results from previous research (Fantinou et al., 1998).Temperatures higher than 27.5°C lessen the time required for diapause development in all photoperiodic regimes tested (Table 2).These data reinforce the photoperiod -temperature interaction dem onstrated in previous work (Eizaguirre et al., 1994;Fan tinou et al., 1995Fan tinou et al., , 1998)).The roles of temperature and photoperiod therefore, seem to be synergistic in enhancing diapause development.According to Hodek & Hodkova (1988) diapause in a large number of insects may be completed more quickly at higher temperatures even in diapause -promoting photoperiods.As the termi nation of diapause in S. nonagrioides could be achieved almost immediately by the reversal of the diapause inducing factor (photoperiod), this state of diapause is characterized an oligopause according to Mueller's classification (Mueller, 1970).
Thermoperiods and the corresponding constant tem peratures exerted comparable effects on the time required for pupation, although a significant difference was recorded (Table 1 and 3).Moreover, diapause develop ment was retarded when larvae were exposed to regimes in which the cryophase temperatures coincided with the scotophase, as reported for other insects (Chippendale et al., 1976;Rock, 1983).In S. nonagrioides diapause development was more rapid under LL or LD at tempera tures employed, than under DD, and this difference may be related to the role of the light on diapause termination.It seems that constant darkness evoked a response to dia pausing larvae treated as being under a short daylength, since the 24h of scotophase is obviously above the insect's intensity light threshold.The differential response of diapausing larvae under LD or DD has also been reported by Lopez et al. (1995) and needs more investiga tion.
Diapause development could be intensified under ther moperiodic regimes with the progressively higher cryo phase rearing temperatures even under the relatively low thermophase temperature (Table 1 and 4).These data sug gest that the scotophase temperature must exceed a tem perature level of ca 18°C to evoke developmental response.Therefore, under natural conditions, the increase in night temperatures in late winter and early spring could function as a signal in eliciting diapause development.This is ecologically meaningful because in temperate regions insects are exposed to daily photope riods and thermoperiods, in which the long nights coin cide with low temperatures.
In nature, thermoperiods are associated with and syn chronized to photoperiods.Insects exhibiting facultative diapause respond to seasonal changes in the environment to both photoperiodic and thermoperiodic cues.Therefore, the synergistic role of thermo-and photope riod on eliciting developmental responses may be extremely important as a stimulus for the preparation of insects for the winter.More studies with field experi ments on thermoperiodic stimuli in relation to post dia pause development and adult appearance will be neces sary to better understand the seasonal occurrence and the population dynamics of this species.

Fig. 1 .
Fig. 1.Diapause incidence (%) in S. nonagrioides larvae reared under thermoperiodic (TC 12 : 12) at DD and thermo photoperiodic (12L : 12D h) conditions by regimes with dif ferent cryophase and cryoscotophase temperatures and a thermophase of 30°C.Each symbol shows the mean value of 4 replicates of 25 larvae each.Values labeled with the same letter are not significantly dif ferent at P = 0.05 by LSD test after ANOVA.

Table 1 .
Analysis of variance results (ANOVA), (A) for diapause incidence (%) and (B-D) for days needed for pupation in S. nonagrioides larvae reared under various combinations of temperature, photoperiod, and thermoperiod.

Table 2 .
Days needed for pupation (mean ± sd) of 50% of diapausing larvae of S. nonagrioides after their transfer at the Values followed by the same letter are not significantly dif ferent at P = 0.05, by LSD test, after ANOVA.