Diapause decision in the small tortoiseshell butterfly,Aglais urticae

Insects in temperate areas spend the inhospitable winter conditions in a resting stage known as diapause. In species that diapause in the larval or pupal stage, the decision whether to diapause or develop directly is customarily taken during the late instars, with long days (i.e., long light phases) and high temperatures promoting direct development. Among butterflies that overwinter as adults, data are rare and variable, but imply that the larval daylength conditions can affect the pathway decision. We studied the small tortoiseshell, Aglais urticae L. (Lepidoptera: Nymphalidae, Nymphalini), which is partially bivoltine from Central Scandinavia and southwards, and tested whether the pathway decision is taken in the larval or adult stage. We reared larvae under long‐day (L22:D2) or short‐day (L12:D12) photoperiods, and recorded the pathway taken by the eclosing adults by scoring their propensity to mate and produce eggs. We also tested whether the larval photoperiod influenced adult ability to diapause by assessing adult survival. The results clearly indicate that (1) there is no detectable effect of larval photoperiod treatment on the pathway decision taken by adults whether to enter diapause or to develop directly, (2) some individuals are obligately univoltine and insensitive to photoperiod during adulthood, whereas (3) other individuals can facultatively enter diapause or direct development, depending on the photoperiod experienced after adult eclosion.     

Repeated larval diapause and diapause-free development in geographic strains of the burnet moth Zygaena trifolii Esp. (Insecta,Lepidoptera)

Summary Zygaena trifolii is a long-day insect with temperature-dependent photoperiodic responses. All larval instars are sensitive to photoperiod; however, diapause may occur at the third larval stage or any subsequent larval instars. There were quantitative differences within populations in the threshold photoperiod for diapause induction. The diapause response was polymorphic, so that larvae might enter diapause at different instars under the same culture conditions. Furthermore, decreasing photoperiods below a critical daylength shifted the diapausing instar towards earlier stages. Geographic strains of Z. trifolii showed discontinuous clinal variation. Near the northern edge of the distribution [Cologne (Köln), FRG], there is first an obligatory diapause, mainly during early instars, and additional facultative (repeat) diapauses during later larval instars in subsequent years. In the southern part of its distribution, this burnet moth is partially bivoltine in the field with a facultative first developmental arrest and a decreased capacity for repeated diapause (Valencia, Spain; Marseille, France). Further experiments indicated that the photoperiodically controlled diapause reaction is also influenced by the number of photoperiodic cycles experienced during the period spent in each larval instar, which depends on temperature. The adaptive significance of obligatory and facultative repeated diapause, varying even among the offspring of a single female, may be to buffer the populations against the more extreme and, from year to year, unpredictable fluctuations in climatic conditions at the northern edge of the distribution.Abbreviations L₃ feeding 3rd larval instar - L₄D diapausing 4th larval instar - L₅D2 repeat-diapausing larval instar with second diapause at the 5th larval stage - LD light-dark cycle - KT shortday conditions (e.g. LD 8:16) - LT long-day conditions (e.g. LD 16:8)     

Effect of photoperiod on winter and summer diapause of the soybean pod borer Leguminivora glycinivorella (Lepidoptera: Tortricidae)

To understand the geographical differences between diapause systems and synchronization of adult occurrence in the soybean pod borer Leguminivora glycinivorella (Lepidoptera: Tortricidae), we examined the timing of winter diapause termination and intensity of summer diapause using univoltine and potentially bivoltine individuals in Iwate, Japan. In laboratory rearing experiments of mature larvae maintained at constant temperature (20 °C), winter diapause intensity weakened by January without photoperiodic responses. Meanwhile, summer diapause was maintained by the long day length and presumably terminated with the photoperiodic transition from long to short day length. The intensity of summer diapause was stronger for cocoons that transitioned from a 16 h light to 8 h dark (LD 16:8) to a LD 15:9 photoperiod than for those that transitioned from LD 15:9 to LD 14:10. These results suggest that populations distributed in relatively low-latitude areas, with partly or potentially bivoltine individuals, would have a weaker summer diapause or none at all. Moreover, sexual differences in the number of days to emergence were not detected when individuals experienced a photoperiodic transition from long to short day length, suggesting that the summer diapause system may function to synchronize the emergence of males and females in the population examined.     

Diapause and diapause dynamics of Colias alexandra (Lepidoptera: Pieridae)

Summary The mechanisms and adaptations involved in the obligate third instar larval diapause of Colias alexandra, a native Rocky Mountain and intermountain region butterfly, are examined. Generally univoltine throughout its distribution, scattered, isolated bivoltine populations occur. Factors influencing continuous development were investigated under laboratory conditions on a single generation of offspring from a univoltine population each year from 1976–1979. If exposed to mean temperatures >24°C during the second instar, a significant number of larvae fail to diapause. These results are interpreted in the context of the actual environmental conditions experienced by the univoltine source population and a bivoltine population. A high degree of individual variability in response to continuous development conditions is shown. Variation in certainty of diapause, within an originally univoltine population, could be the evolutionary starting point for evolving a genuine multivoltine cueing system and provide the mechanism for ecological range expansion.     

The dependence of development and fecundity of Samea multiplicalis on early larval nitrogen intake

Larvae of the moth Samea multiplicalis (Lepidoptera: Pyralidae), developed more rapidly when their food plant, Salvinia molesta was richer in nitrogen. Larvae that fed on plants with less than 1.35% nitrogen during their first instar completed the larval stage in 18.0–24.2 days, after passing through six instars. In contrast, larvae that had fed on food richer in nitrogen during their first instar completed the larval stage in 14.1–17.4 days, and passed through five instars. Experience of nitrogen in later instars had little additional effect on total larval development time. Oöcyte complements of 1-day old adult females was correlated with the mean food nitrogen content over the larval period. Food nitrogen content experienced by larvae in earlier instars was as important as that in the final instars, in determining number of mature oöcytes, which suggests early conditioning of the nitrogen use effciencies of later instars.     

Effects of photoperiod and temperature on diapause induction in Conogethes punctiferalis (Lepidoptera: Pyralidae)

The yellow peach moth, Conogethes punctiferalis (Guenée), a multivoltine species that overwinters as diapausing larvae, is one of the most serious insect pests on maize in China. Effect of photoperiod and temperature on larval diapause was examined under empirical laboratory conditions. Short‐day treatments caused larval diapause at 25°C, and the critical photoperiod was between 12 and 13 h (or 12 h 51 min) light per day. No sensitive instar was identified for diapause induction under alternated short‐ (L : D 11 : 13 h) and long‐day (L : D 14 : 10 h) treatments at different larval stages. However, accumulative treatment of three instars and 10 d under short‐day treatment was required for the induction of 50% larval diapause. All larvae entered diapause at 20°C, whereas less than 3% did so at 30°C, irrespective of the long‐ or short‐day treatment. Furthermore, under the short‐day treatment, more than 90% of larvae went into diapause with temperatures ≤ 25°C, but less than 17% did so at 28°C. In contrast, under the long‐day treatment, less than 19% of larvae went into diapause with temperatures ≥ 23°C. The forward shift (5°C) of critical temperature under the long‐day regime demonstrated the compensatory effect of temperature and photoperiod on diapause induction. In conclusion, C. punctiferalis had a temperature‐dependent type I photoperiodic diapause response; there was no sensitive instar for diapause determination, but the photoperiodic accumulation time countermeasures both of the short‐day cycles and the number of instars exposed, and the photoperiodic diapause response, was a temperature‐compensated phenomenon.     

Photoperiodism and control of summer diapause in the Mediterranean tiger moth Cymbalophora pudica

Photoperiodic responses to both constant and changing photoperiods were studied in the Mediterranean tiger moth Cymbalophora pudica. Embryos, larval instars and prepupae were all sensitive to photoperiod, and the responses of larvae and prepupae to changing photoperiods were similar. At 23+/-2 degrees C, constant 24-h photoperiods with short photophases (11, 12h) induced a long diapause (mean 88days) whereas long photophases (14, 16h) induced a short diapause (mean 52days). A change to a longer or shorter photophase during larval development or during diapause caused a significant extension (up to a maximum of 138days) or shortening (down to a minimum of 10days) of diapause, respectively, but only when at least one of the photophases was longer than 14h. Thus, shortening and prolongation of photophase had an opposite effect than constant short and long photophases, respectively. Changes within the range of photophases of 10-14h did not cause a significant change in diapause duration.Experimental results enabled us to outline the mechanisms regulating voltinism and the duration of summer diapause. For the monovoltine cycle, cold autumn/winter temperatures slow down larval development, and prepupal aestivation starts relatively late (March, April). Prepupae are then kept in diapause by the increasing daylength (>14h after late April). Pupation is synchronized by decreasing daylength after summer solstice, and imagoes emerge in September/October. For the bivoltine cycle, when the autumn/winter temperatures are relatively warm, a certain portion of the population (depending on the individual rate of growth) may be diverted to a bivoltine life-cycle. In such a case, larval development is fast and short enough to allow an early start of diapause (prior to or during February). The duration of such early diapause is not influenced by changes in daylength (<14h); pupation occurs very early (April/May), and spring generation imagoes fly and oviposit in May/June. Summer larvae and prepupae live under decreasing daylength (>14h), which shortens their diapause to 20-30days. Imagoes of the autumnal generation thus occur in September/October, together with the univoltine portion of the population.     

The seasonal development cycle of <Emphasis Type="Italic">Aphantopus hyperantus</Emphasis> (L.) (Lepidoptera,Nymphalidae: Satyrinae) in Leningrad Province

The simultaneous influence of the photoperiodic and temperature conditions on pre-diapause and postdiapause larval development of the ringlet Aphantopus hyperantus was studied. At the short day (12 h of light a day) all larvae reached the III instar and entered diapause at all the temperatures tested (18–24°C). At the long day (22 h) the number of diapausing larvae increased with decreasing temperature, and the larvae diapaused at the III and IV instars. The I and II instar larvae (before winter) developed and grew faster under short-day, and the hibernated larvae, under long-day conditions. At the short day the growth rate of the I and II instar larvae did not depend on the temperature, and at the long day it decreased as the temperature increased. After hibernation the growth rate of the IV instar larvae increased with the temperature under long-day conditions, and did not depend on the temperature under short-day ones. Thus, acceleration or deceleration of development, depending on the photoperiodic and temperature conditions, regulated the timing of diapause in the overwintering stage (the III instar larva) and maintained the univoltine seasonal cycle. The males of A. hyperantus developed and grew faster than females while their weight at the larval and pupal stages was smaller. The thermal reaction norms for development (the lower temperature threshold and the coefficient of thermal sensitivity of development) were calculated for the eggs, IV and V instar larvae, and pupae. The coefficient of thermal sensitivity of development for the IV and V instar larvae taken together was lower, and the threshold was higher than the corresponding values for eggs and pupae. The thermal reaction norms for development of the latter stages were the same. The larvae of A. hyperantus retained their sensitivity to the photoperiod after hibernation. Resumption of development after cold reactivation occurred only under long-day conditions, whereas under short-day ones the overwintered larvae entered a repeated diapause. The possibility of photoperiodic reactivation and the absence of spontaneous resumption of development in these larvae were demonstrated.     

Photoperiodic responses during larval development and diapause of two geographic ecotypes of the rice stem maggot, Chlorops oryzae

Chlorops oryzae is bivoltine in northern Japan but trivoltine in the southern part of the country. In the bivoltine strain, both the egg and larval stages were found to be sensitive to photoperiod. When the egg stage was exposed to a long-day photoperiod (16L:8D), larval development showed a short-day type response, and mature third-instar larvae entered a summer diapause under a long-day photoperiod (15L:9D). When eggs experienced short days, the first-instar larvae entered a winter diapause under short-day conditions, and the critical photoperiod in the larval stage ranged from about 14L:10D to about 12L:12D as the photoperiod experienced by the eggs increased from 12L:12D to 14L:10D. However, the development of the larvae after overwintering was not influenced by the photoperiod. In the trivoltine strain, larval development was retarded under a 14L:10D photoperiod but not under either shorter or longer photoperiods, when larvae had spent the egg stage under a 16L:8D photoperiod. The critical photoperiod of the larval stage for the induction of a winter diapause in the first instar was about 12L:12D, though it varied to some extent with the photoperiod during the egg stage. Thus, Chlorops oryzae was able to adapt itself to the local climatic conditions by the development of variable and complicated photoperiodic responses.     

Larval aestivation and direct development as alternative strategies in the speckled wood butterfly, Pararge aegeria, in Sweden

ABSTRACT.

• 1 Sweden has two disjunct populations of the speckled wood butterfly, Pararge aegeria L. The southern population has two generations per year but the central Swedish population is univoltine. When rearing larvae from central Sweden under normal photoperiodic conditions but at temperatures slightly above the ambient, 42% of the larvae developed directly and produced a second generation of adults the same summer. The egg—larval development time of the directly developing individuals was about 40 days, whereas that of the individuals developing along the univoltine pathway was about 100 days.
• 2 Larvae of the central Swedish population normally aestivate during part of the summer even though abundant food is available. In the closely related Lasiommata petropolitana F., which is the only Swedish satyrid that overwinters in the pupal stage besides P.aegeria, larvae do not aestivate, indicating that there does not seem to be any obligatory association between pupal hibernation and larval aestivation.
• 3 Development rates of aestivating and directly developing P.aegeria are equal up to the third larval instar. During the third and fourth instars, however, the development rate of aestivating individuals is retarded and females also have an additional fifth instar.
• 4 Since the central Swedish P.aegeria have the capacity to develop directly, and the southern Swedish ones have the capacity to aestivate, the evidence indicates that the outcome of the cost/benefit balance of univoltine versus bivoltine development differs between the two areas.

     

Life‐history traits related to diapause in univoltine and bivoltine populations of Ypthima multistriata (Lepidoptera: Satyridae) inhabiting similar latitudes

In most temperate insects, diapause strategies and voltinism generally exhibit latitudinal clines, supporting the concept that they represent adaptations to climate. In contrast, in the satyrine butterfly Ypthima multistriata Butler, local populations with different voltinism patterns are geographically intermingled, suggesting that life‐history traits related to diapause may differ even between geographically and phylogenetically close populations. In this study, we experimentally examined the critical photoperiod for diapause induction and the larval developmental period in two univoltine and two bivoltine populations of Y. multistriata, all of which inhabit virtually the same latitude (34.652–34.750°N). We found that the critical photoperiod for diapause induction was longer in the univoltine populations than in the bivoltine populations. Moreover, the larval period under the long day length treatment was different among populations in both sexes, although significant differences were also detected between populations with the same voltinism. These results indicate that in Y. multistriata, life‐history traits related to diapause can not be attributed merely to climatic conditions such as temperature or day length, which depend largely on latitude. Therefore, we suggest that biotic elements, such as leaf toughness, as well as abiotic elements should be taken into account in attempts to explain the enigmatic pattern of geographic variation in the diapause strategies of Y. multistriata.     

Diapause induction in the oblique-banded leafroller Choristoneura rosaceana (Lepidoptera: Tortricidae): Role of photoperiod and temperature

Induction of diapause in the larval stage of the oblique-banded leafroller, Choristoneura rosaceana (Harris), was found to be dependent on both photoperiod and temperature. At constant temperatures of 24, 20 and 16°C, short photoperiods induced diapause. The critical photoperiod was between 14–15 h of light per day at 20 and 16°C. At 14 h light: 10 h dark, all larvae expressed diapause. Temperature had a modifying effect, and slightly shifted the larval response to diapause-inducing photoperiods. High constant temperatures of 28°C and above induced diapause in some individuals (< 20%), while fluctuating temperatures of 32 and 16°C in a 12-h cycle resulted in 67% diapause induction, suggesting that diapause could also be induced by fluctuating temperatures, particularly if the higher temperature exceeds 25°C.The first- and the second-instar larvae were the only two stages sensitive to diapause induction. Exposure of adult, egg and third, fourth, and fifth-larval instars to diapause-inducing conditions did not produce diapause. Although diapause was induced in the first or the second instars, it was always expressed in the third or fourth instar.     

Relationship between an increase of juvenile hormone titer in early instars and the induction of diapause in fully grown larvae of Sesamia nonagrioides

The larvae of Sesamia nonagrioides (Lepidoptera: Noctuidae) grown at 25 degrees C and long photoperiod (16:8h light:dark) pupate in the 5th or 6th (mostly) larval instar, while the larvae reared under a short photoperiod (12:12h) enter diapause during which they consume some food and undergo up to 12 (usually 3-4) stationary larval molts. Diapause programming includes an increase of juvenile hormone (JH) titer in the hemolymph from about 20 to 50 nM in the 4th and 5th instar larvae (titer in earlier instars was not measured). JH I, II, and III are present in approximate ratio 1-2:10:1. The JH titer drops to zero before pupation but remains around 20 nM during diapause. Perfect extra larval molts associated with a body weight increase can be induced in the non-diapausing larvae with a JH analogue (JHA). The weight rise is due to accumulation of reserves and not to a general body growth. The timing of extra molts is similar to the molting pattern of the diapausing larvae only when JHA is present since early larval instars. In the diapausing larvae, JHA application affects neither molting periodicity nor the body weight. It is concluded that (1) Increased JH titer in early larval instars is a part of diapause programming; (2) The extension of larval stage in the diapausing larvae, but not the timing pattern of extra molts, is due to continuously high JH titer; (3) The diapause program includes low food intake, maintenance of a certain body weight, and periodic larval molts.     

Photoperiodic induction of larval diapause and temperature-dependent termination in a wild multivoltine bruchid,Kytorhinus sharpianus

A wild bean weevil,Kytorhinus sharpianus Bridwell (Coleoptera: Bruchidae), has a multivoltine life cycle and enters a hibernal larval diapause at the fourth instar under a short daylength (Shimada & Ishihara, 1991). Here, we investigated their diapause incidence under different photoperiods at 24°C and 27°C. The critical photoperiods for diapause induction were 14.5 h at 24°C and 14 h at 27°C. The stages susceptible to diapause-inducing stimuli were estimated by transferring larvae of various instars from long days to short days and vice versa. Then we investigated the incidence of larval diapause. The sensitive stage was estimated to be from the third to early fourth instar. Though larval diapause, which was induced under a short daylength, was terminated only by increasing the daylength, the termination was more synchronized by an exposure to a low temperature followed by increasing temperature, irrespective of photoperiod.     

Pupal diapause of Helicoverpa armigera (Lepidoptera: Noctuidae): sensitive stage for thermal induction in the Okayama (western Japan) population

Helicoverpa armigera (Hübner) exhibits a facultative pupal diapause, which depends on temperature and photoperiod. Pupal diapause is induced at 20 degrees C by short photoperiods and inhibited by long photoperiods during the larval stage. However, in some pupae (35% of males and 57% of females) of a non-selected field population from Okayama Prefecture (34.6 degrees N), diapause is not induced by short photoperiods. In the present experiment, the importance of temperature for diapause induction was studied in the non-diapausing strain, which was selected from such individuals reared at 20 degrees C under a short photoperiod of 10L:14D. Furthermore, the sensitive stage for thermal determination of pupal diapause was determined by transferring larvae of various instars and pupae between 20 degrees C and 15 degrees C. Diapause was induced by 15 degrees C without respect to photoperiod. When larvae or pupae reared from eggs at 20 degrees C under a short or a long photoperiod were transferred to 15 degrees C in the periods of the middle fifth instar to the first three days after pupation, the diapause induction rate was significantly reduced in both males and females, especially in females. In contrast, when larvae or pupae reared at 15 degrees C were transferred to 20 degrees C in the same periods, diapause was induced in males, but not in females. However, the diapause induction rate of pupae transferred to 20 degrees C on the fourth day after pupation was significantly increased in females. The results show that temperature is the major diapause cue in the photoperiod-insensitive strain and the periods of middle fifth larval instar to early pupal stage are the thermal sensitive stages for pupal diapause induction with some different responses to temperatures between males and females in H. armigera.     

Diapause induction as an interplay between seasonal token stimuli,and modifying and directly limiting factors: hibernation in Chymomyza costata

Larvae of wild type (WT) strain of Chymomyza costata Zetterstedt (Diptera: Drosophilidae) enter diapause (stop developing) in response to short‐day signal at a constant 18 °C, whereas larvae of a non‐photoperiodic‐diapause (NPD) strain do not respond to photoperiodic signalling and continue in larval development irrespective of daylength. The present study shows that WT larvae also respond reliably to thermoperiodic signalling (daily cycles of temperature) under constant darkness, whereas the NPD larvae do not, suggesting that the pathways transducing the environmental token stimuli (photoperiod and thermoperiod) onto the diapause developmental programme might merge functionally in the central biological clock system known to be mutated in NPD strain. Temperature and larval population density modify the output of token stimuli signalling. High temperatures (>24 °C) tend to avert, whereas low temperatures (<18 °C), especially in combination with constant darkness, stimulate diapause induction in WT strain. Overcrowding (>200 larvae per 5 g of larval diet) lengthens the duration of larval development and induces a ‘diapause‐like’ developmental arrest of relatively weak intensity in up to 60% of larvae of both strains. At high temperatures (>30 °C), all WT larvae continue direct development but subsequently die during the pupal stage. Low temperature exposure (<12 °C) causes quiescence in the majority of the larvae of both strains. Starvation blocks development and causes mortality when applied in larvae younger than day 3 of the third instar. Older larvae survive starvation and their photoperiodically‐induced developmental pre‐programming is not affected. Collectively, the results show that diapause induction in C. costata is a result of various interacting effects of multiple environmental factors.     

Induction of pupal diapause and photoperiodic sensitivity during early development of Sarcophaga peregrina larvae

Effects of long day and short day treatments during the embryonic and larval stages on induction of pupal diapause were studied on a diapausing race of Sarcophaga peregrina. Two long day (15L 9D) cycles during 2 days before or after the larviposition completely stopped the induction of pupal diapause on larvae which grow in short day condition before and after the long day treatment. The sensitivity appeared to decrease during the early stage of the third instar and to increase again to some extent in the prepupal stage.     

Ground crickets singing in volcanic warm “islets” in snowy winter: Their seasonal life cycles,photoperiodic responses and origin

The ground cricket Dianemobius nigrofasciatus overwinters as an egg in Japan, being univoltine in Hokkaido and northern Honshu and bivoltine farther south. In Hokkaido, however, this cricket is heard singing in winter in several fumarolic fields covered with moss and grasses locally known as “bokke”. In such warm “islets” the adult density was high in early summer and again in autumn, indicating that the cricket is bivoltine in contrast to the univoltine life cycle outside the bokke habitats in Hokkaido. Eggs laid by females collected at regular intervals from a bokke habitat showed a clear seasonal cycle of diapause incidence. At 26°C, the bokke strains produced non‐diapause eggs under long days and diapause eggs under short days as in the southern bivoltine populations, although the critical day‐length was longer than in the south. Several strains derived from non‐bokke habitats in Hokkaido and northern Honshu produced high percentages of diapause eggs under long days as well as short days as expected for the univoltine life cycle. Winter adults singing in bokke habitats could be either survivors of the autumn generation or individuals derived from eggs laid in autumn and then matured in response to the high soil temperature. In the laboratory, the proportion of egg diapause in short days was decreased by selection only for several generations. Phylogenetic trees of bokke and non‐bokke populations, based on both the nucleotide sequence of the mitochondrial COI gene and four allozyme loci, suggest that bokke populations have not been isolated from non‐bokke populations for an evolutionarily significant time.     

Effects of temperature,photoperiod and soil humidity on induction of pseudopupal diapause in the bean blister beetle Epicauta gorhami

Larvae of the bean blister beetle Epicauta gorhami Marseul (Coleoptera: Meloidae) feed on grasshopper eggs in soil and undergo hypermetamorphosis. This beetle undergoes larval diapause in the fifth instar as a pseudopupa, a form characteristic of hypermetamorphosis in meloid beetles. The effects of temperature, photoperiod and soil humidity on larval development of E. gorhami are examined in a population in Miyazaki, Japan, using egg pods of Locusta migratoria L. as food. At lower temperatures (20 and 22.5 °C), all larvae become pseudopupae, regardless of the photoperiod. By contrast, at higher temperatures (27.5 and 30 °C), almost all larvae pupate at the end of the fourth instar, again regardless of the photoperiod. A long‐day photoperiodic response occurs only at an intermediate temperature (25 °C): under an LD 12 : 12 h photocycle, all larvae enter diapause, although the diapause incidence tends to decrease as the day length becomes longer. Pseudopupae are immobile and remain in diapause for ≥120 days when they are kept under the same conditions, except that diapause terminates within a relatively short time at 30 °C. Although lower soil humidity retards post‐feeding development, soil humidity has no effect on the diapause incidence. On the basis of the short developmental period and diapause avoidance under summer conditions, it is suggested that this beetle partially produces two generations a year in southwestern Japan.     

Photoperiodic control of larval development in the semivoltine cockroachPeriplaneta japonica (Blattidae: Dictyoptera)

Periplaneta japonica is semivoltine, entering early diapause in any (except the first) larval instar before the last, and late diapause in the last instar. Early diapause was induced under a short day of 13 h or less at 28°C, and under both short and long daylength (12–16 h) at 20°C. The shorter the daylength and the lower the temperature, the younger the instar was entering early diapause. Early diapause was terminated by a long day (16 h) or a high temperature (28°C), after which larvae grew faster in short days than in long days until the last instar, when they again entered diapause, always in short days and frequently in long days. This late diapause was terminated also by an increase in daylength and was always followed by adult emergence. In this case, 13 and 14 h daylengths after exposure to 12 h daylength were as effective as 16 h daylength. Ourdoor samples collected in late autumn, winter and early spring at Hirosaki (40.5°N) comprised two distinct size groups, corresponding with the early and late diapause instars. Based on these results, the seasonal development strategy and intriguing aspects of the photoperiodic response in this cockroach are discussed.