Inconsistency in the expression of locomotor and ERG circadian rhythms in the German cockroach Blattella germanica (L.)код для вставкиСкачать
Archives of Insect Biochemistry and Physiology 48:155–166 (2001) Inconsistency in the Expression of Locomotor and ERG Circadian Rhythms in the German Cockroach, Blattella germanica (L.) Huan-Wen Chang and How-Jing Lee Department of Entomology, National Taiwan University, Taipei, Taiwan ERG recordings from German cockroaches showed that the amplitude of light-evoked responses have a circadian rhythmicity in adult males that coincided with the locomotor circadian rhythm. The peak of the response occurred during the subjective night, and the circadian period was less than 24 h under DD condition. In contrast, although the locomotor circadian rhythm was masked by the development of ovaries and pregnancy in females, their visual responses displayed circadian rhythmicity. This inconsistency in expression of locomotor and visual sensitivity circadian rhythms in females implied separate pacemakers for these two overt rhythms. After severing the optic nerves, changes in ERG amplitude of the operated cockroaches still displayed a circadian rhythm under DD condition, demonstrating that the visual sensitive pacemaker was located in the eye and independent from the locomotor pacemaker. Arch. Insect Biochem. Physiol. 48:155–166, 2001. © 2001 Wiley-Liss, Inc. Key words: electroretinogram; circadian rhythm; masking effect; pacemaker; ovarian factors INTRODUCTION Male adults of the German cockroach, Blattella germanica (L.), express robust circadian rhythmicity in their locomotor activity (Dreisig and Nielsen, 1971; Sommer, 1975; Leppla et al., 1989; Lin and Lee, 1996). In contrast, female adults do not display a locomotor circadian rhythm. Rather their cyclic changes of daily locomotor activity are masked by the effects of their ovarian development (Lee and Wu, 1994; Lin and Lee, 1996). Based on the finding that ovariectomized females express a locomotor circadian rhythm that has the same circadian period as male adults, the endogenous circadian clock for locomotor activity is present in adult females. However, it is normally masked by ovarian factors (Lin and Lee, 1996; Tsai and Lee, 2000). © 2001 Wiley-Liss, Inc. The locomotor pacemaker is located in optic lobes of the German cockroach (Wen and Lee, 2000) as well as in other cockroach species (Helfrich-Forster et al., 1998). Since the internal masking effect of the locomotor circadian rhythm is a unique characteristic of the German cockroach, identification of the site at which masking factors interact with the circadian system is an interesting project to pursue, assuming that the locomotor pacemaker is the master clock that conContract grant sponsor: National Science Council, ROC; Contract grant number: NSC 89-2313-B-002-030. *Correspondence to: Dr. How-Jing Lee, Dept. Entomology, National Taiwan University, Taipei 106, Taiwan. E-mail: email@example.com Received 3 November 2000; Accepted 11 June 2001 156 Chang and Lee trols other overt circadian rhythms. However, if another circadian rhythm can be identified in female adults expressing arrhythmic locomotion, we would have eliminated the possibility of the master pacemaker itself serving as an interaction site for the masking factors. The ERG (electroretinogram) has been proven to be particularly useful as an easy method for quantifying of visual responses. Using this technique, circadian rhythms of visual responses in a variety of arthropods including several species of insect have been documented (Koehler and Fleissner, 1978). Wills et al. (1985) suggested that the pacemaker that controls circadian rhythmicity of the ERG resides in the optic lobes and is anatomically close to, or identical with, the pacemaker that regulates the locomotor activity rhythm in the cockroach, Leucophaea maderae. The ERG in the German cockroach, therefore, was chosen for investigating the relationship between the locomotor activity and visual response circadian rhythms and between pacemakers of these rhythms. MATERIALS AND METHODS Insects All experiments were performed with adult German cockroaches, Blattella germanica (L.), collected from households in Taipei, Taiwan, and reared in an environmental chamber at 28°C and LD 16:8 h for several generations. Dog chow and water were provided ad libitum. Last-instar nymphs were separated daily and kept under identical rearing conditions. The adults that emerged within 24 h were defined as 0-day-old. Detailed rearing procedures were described in a previous report (Lee and Wu, 1994). ERG For recording the ERG (Electroretinogram), a cockroach that had been maintained in a lightdark cycle (16:8h) was immobilized by waxing the head capsule to the lucite stage. The cockroach’s legs were restrained with wax, and its antennae were removed. An active electrode was made of a glass micropipette filled with three M KCl. A small hole in the compound eyes was made with a sharp needle to allow access for the active electrode that penetrated just below the surface of the cornea of the eye. An indifferent electrode was made of silver wire and placed in the haemocoele through a small hole in the tergum of the thorax. The laceration was sealed with petroleum jelly. After the electrodes were in place, the cockroach was placed in a light-tight box, and kept in constant darkness throughout the recording period. During the recording period, the cockroach was given access to water. To stimulate the retina, a fiber-optic solidstate color wheel (WPI, Sarasota, FL) was used to generate a blue light stimulus (spectral maximum at 470 nm). Based on the spectral sensitivity of the German cockroach (Koehler et al., 1987), blue light stimulation produces maximal visual responses. Stimulation was controlled by a digital stimulator (Grass, Inc., Quincy, MA) that triggered a train of light pulses that consisted of a burst of 23 to 150 ms pulses presented at 5-s intervals for 2 min every 2 h. The light pulses were delivered through fiber optics to the vicinity of a compound eye where maximum responses of ERG amplitude could be recorded. The distance between the eye and light source was 5 mm. The differences in voltage recorded by the active and indifference electrodes were amplified by a preamplifier (S-7071A, WPI, Sarasota, FL) and digitized with an A/D converters (digidata 1200A , Axon, Foster City, CA), and filtered (gain 8×, bandpass 0.5–10 Hz). A computer program, Axoscope 1.1 (Axon, Foster City, CA), was used to analyze the data and transfer them to Origin 4.1 (Microcal Software, Inc., Northampton, MA) for graphic presentation. Changes of retinal sensitivity were indicated by changes in ERG amplitude based on an averaging 15 responses to light pulses every 2 h. The cyclic pattern of ERG was determined by eye-fitting sine wave course and the free-running period was calculated as the mean value of successive time laps between two amplitude peaks (Colwell and Page, 1989). Optic Nerve Severance Newly emerged male adults were anesthetized on crushed ice and their optic nerves were exposed by cutting the cuticle across the inner margin of the compound eye at the postgena with a microscalpel. The optic nerves between retina and lamina of optic lobes were cut by inserting Inconsistent Expression of Two Circadian Rhythms fine iridectomy scissors. The operated insects were placed in a humid environment without food for one day to recover from the surgery, after which they were examined individually for ERG responses from the operated compound eyes. Locomotor Activity Locomotion of individual German cockroaches was automatically recorded with an infrared motion detector system described in a previous report (Lin and Lee, 1998). All insects were tested at 28°C in constant darkness (DD). Although they were fed beforehand, adults being tested were kept in the motion detector boxes with water only. Locomotor activity was measured as the number of times in an hour that the infrared sensor was triggered by movement of the cockroach. The rhythmicity of daily locomotion and the circadian period (τ) were calculated and for data recorded during 10 consecutive days using a chi-square periodogram (Sokolove and Bushell, 1978) at 0.05 significant level. RESULTS ERG Waveform An ERG is an extracellular recording of a light-induced electrical activity in the retina of 157 compound eyes. A typical response of the ERG elicited by a moderate intensity (0.0317 Lum/sf) of blue light for 150 ms, showed a slow monophasic negative potential after 10–30 ms of illumination, then maintained itself at a steady level and returned to base line slowly when the light was extinguished (Fig. 1). The negative potential typically peaked at 150 ms after the beginning of the light stimulation. Although the retina immediately responded to the light pulse, the negative potential of ERG did not return to base line right after the light off. ERG waveforms from male and female adults were the same. The kinetics of the ERG at the most sensitive time (Zeitgeber time = 19 h) under constant darkness condition were analyzed using the following parameters: (1) The time required to initiate the ERG (latency): defined as time elapsed between the stimulation of light on and the initiation of the ERG; (2) The rate of rise of the ERG (initial slope): defined as the rate to reach 50% of the maximal response following the initiation of the response; (3) The recovery of the ERG (time to repolarization): defined as the time to return halfway to the resting potential before the stimulation following the termination of the stimulation (Fig. 2). The high light intensity caused not only a high amplitude response (Fig. 2A), but also Fig. 1. An example of electroretinal response to a light pulse recorded from the compound eyes of the German cockroach, B. germanica (L.) under constant darkness condition. The response was elicited with a 150 ms of light pulse (λmax 470 nm). 158 Chang and Lee Fig. 2. The kinetic analysis of ERG responses to various intensity of light pulses in the German cockroach B. germanica (L.). A: The amplitude of ERG. The amplitude was defined as the magnitude difference between the potentials before lights on and the peak of the ERG. B: The latency of the ERG. It was defined as time elapsed between the stimulation of light on and the initiation of the ERG. C: The rate of rise of the ERG response. The initial slope of the response was defined as the rate to reach 50% of the maximal response following the initiation of the response. D: The recovery of the ERG response. The time to repolarization was defined as the time to return half way to the resting potential following the termination of stimulation. The different letters (a, b, c, d) on the histogram indicates significant difference (Fisher PLSD, P < 0.01). The error bars indicate SEM. This typical kinetic analysis of ERG responses was a representative of 12 male cockroaches. rapid initiation of the ERG (Fig. 2B). On average, the retina responded to the light pulse within 10–30 ms. The initial slope of the ERG also increased as light intensity increased, producing a more rapid ERG response (Fig. 2C). However, the recovery of ERG following light off required 80 to 100 ms to reach halfway to the resting potential (Fig. 2D). Within the range of light intensities employed, there were no significant differences in time for recovery. From the kinetic analysis of ERG, latency and initial slope were the two key parameters to contribute to the changes in ERG regardless of circadian time. When circadian activities were examined, the amplitude changes of the ERG to the various tested intensities showed a free-running rhythm (Fig. 3A). The circadian pattern persisted throughout the recording period, and the rhythmic changes had a period close to 24 h. From the analysis of kinetic changes of the ERG, only one parameter, the rate of rise of the ERG, displayed circadian rhythmicity (Fig. 3C). The rate to reach 50% of maximal amplitude following the initiation of the response increased in the subjective night, and decreased in the subjective day. However, the other parameters, the latency and time for repolarization of Inconsistent Expression of Two Circadian Rhythms 159 Fig. 3. Plots of the kinetic changes for one out of five testing males of B. germanica (L.) in the ERG as the function of circadian time. The daily changes of amplitude (A), latency (B), speed (C), and recovery (D) of the ERG for 3 consecutive days. The German cockroach (B. germanica) was maintained in constant darkness except when they were stimulated every 4 h by a 150ms light pulse which triggered an ERG response. The legend is described in Figure 2. the ERG, did not display clear circadian oscillation (Fig. 3B and D). Intensity-response curves for the ERG measured during a free-running circadian cycle are presented in Figure 4. The amplitude changes of the ERG in the adult male cockroach were measured as the function of light intensity during one circadian cycle under constant darkness (Fig. 4A). In responding to various light intensities, the amplitude of ERG increased as the light intensity increased and it peaked during the subjective night (at zeitgeber time 15 and 19 h). Since the initial slope was the only parameter that displayed cir- cadian rhythmicity when we examined ERG kinetics, it also showed the same pattern of change as that for change in ERG amplitude (Fig. 4B). Circadian Rhythm of ERG Male German cockroaches were maintained in a light-dark cycle (LD 16:8 h) for one day before being placed in the ERG recording apparatus. The daily amplitude changes of the ERG of adult male cockroach exhibited a distinctive temporal cycles (Fig. 5A). The amplitude of ERG reached its maximum during the subjective night, before gradually decreasing in the subjective day. 160 Chang and Lee Fig. 4. Intensity-response curves for the ERG measured during a free-running circadian cycle of one male B. germanica (L.). A: Amplitude changes during one circadian cycle are plotted. B: Changes in the initial slope of the ERG during one circadian cycle are plotted. Each response was evoked by a 150-ms light pulse under darkness condition (re-plotted from the first day data of Fig. 3). This circadian cycle persisted for the entire recording period (at least 7 days). These rhythmic changes were similar to the locomotor circadian period in that the period was shorter than 24 h (Table 1). After a few cycles, the rhythm had drifted completely out of phase with the entrained lightdark cycle. In order to show the consistency of the ERG and locomotor circadian rhythm, the locomotion of starving male adults was monitored (Fig. 5B). The male showed a free-running locomotor rhythm with a period of 23.6 h under DD condition. Both the ERG and locomotor rhythms of male adults expressed circadian rhythmicity with similar duration of circadian periods (Table 1). Since male and female German cockroaches differ in their circadian locomotor activity (Lin and Lee, 1996), the ERG of starving females in various stages of reproductive activity were also monitored. ERG amplitude changes and locomotor patterns of 1-day-old starving adult females are shown in Figure 6. Both the ERG (Fig. 6A) and locomotor activities (Fig. 6B) displayed a freerunning rhythm under DD condition. Although the ERG data were not monitored over a long enough time period to calculate the circadian period by a statistical program, the cyclic pattern did show a < 24-h circadian period (by measur- ing the time between two peaks) that was similar to the circadian period of locomotion (Table 1). The peak amplitude of the ERG occurred during the subjective night at the same time as the locomotion displayed the peak activity. After 5 days of normal feeding, the ERG and locomotor activities of female adults were monitored under conditions of starvation and DD. The daily fluctuations of the ERG and locomotor activity of a 5-day-old female are shown in Figure 7. Although changes in ERG amplitude displayed a free-running rhythm with a <24-h period (Fig. 7A), a locomotor circadian rhythm was not expressed (Fig. 7B). The peak amplitude of the ERG occurred in the subjective night. The locomotor pattern was similar to that monitored from normal fed virgin females, in that their locomotor circadian rhythms were masked by the mature development of their ovaries (Table 1). Typical ERG fluctuations and locomotor activity patterns are shown in Figure 8 for starving pregnant females. Amplitude changes in the daily ERG displayed a free-running rhythm with a period of less than 24 h, and the peak amplitude occurring during the subjective night (Fig. 8A). However, the minimal locomotor activity of the starving pregnant female prevented the de- Inconsistent Expression of Two Circadian Rhythms 161 Fig. 5. A: Changes in the amplitude of the ERG as a function of circadian time. The 1-day-old male German cockroach (B. germanica) was provided with water only and maintained in constant darkness except every 2 h it was stimulated with a 150-ms light pulse (light intensity = 0.0317 Lum/sf) to trigger an ERG response. The amplitude of the ERG fluctuated cyclically and peaked during the subjective night. B: An actogram of a starving 1-day-old male cockroach under 28°C, constant darkness conditions. The bar on the top of the actogram represents the photoperiod conditions. The horizontal lines show successive 24-h periods of locomotor activity. The superimposed line across the actogram represents the predicted onset time for the locomotor activity based on the length of the male’s circadian period (τDD= 23.6h). A and B were the typical representative of 5 and 6 starving males, respectively. tection of circadian rhythmicity (Fig. 8B). Therefore, although the locomotor circadian rhythm was masked by pregnancy, a circadian rhythm for visual sensitivity was clearly expressed (Table 1). located in the optic lobes where the locomotor pacemakers are located, and therefore pacemakers for visual sensitivity and locomotion of the German cockroach are separated. The pacemaker for visual responses is located in the eye while the locomotor pacemaker is in the optic lobe. Pacemaker of ERG A classical way to demonstrate the possible site of a pacemaker was to separate the neural connection between the site and the brain. A freerunning ERG was produced by the compound eyes in a male whose optic nerve was severed, disconnecting the eye from the optic lobe (Fig. 9). Therefore, the pacemaker of visual sensitivity was demonstrated in the eye. This finding indicates that the pacemaker for visual sensitivity was not DISCUSSION ERG Waveform The responses of insect compound eyes to light stimuli can be grouped on the basis of their electrical activity and their visual behavior (Autrum, 1958). One category of responses shows a relatively simple negative potential when illu- 162 Chang and Lee TABLE 1. Effects of Age on Circadian Period of ERG and Locomotion in the German Cockroach, B. germanica, at 28°C and Constant Darkness Conditions Y X a 1 X a 5 X a 10 Y b dn ERG N Expression of circadian rhythm (%) Circadian period (τDD)c 5 100 22.8 ± 1.89 5 100 23.5 ± 1.00 5 60 22.5 ± 0.1 5 100 22.38 ± 0.54 3 100 21.75 ± 0.35 Locomotion N Expression of circadian rhythm (%) Circadian period (τDD)c 6 100 23.6 ± 0.23 10 100 23.7 ± 0.27 10 0 – 10 0 – – – – a The initial age (days) of the testing is indicated in the subscript. dn: the optic nerve of male was severed at the day of emergence. c The free-running period was calculated from measuring the time lapse between two amplitude peaks. Dashes (–) indicate data not available. b Y Fig. 6. A: The amplitude changes of ERG responses for an adult female German cockroach (B. germanica) after emergence. The animal was provided with water only and maintained in constant darkness except every 2 h when it was stimulated with a 150-ms light pulse (light intensity = 0.0317 Lum/sf) to trigger an ERG response. The amplitude of the ERG fluctuated cyclically and peaked during the subjective night. B: An actogram for a starving adult female German cockroach under 28°C, constant darkness conditions. The legend is described in Figure 5. A and B were the typical representative of 5 and 10 starving females, respectively. Inconsistent Expression of Two Circadian Rhythms 163 Fig. 7. A: The amplitude changes of ERG responses for an adult female German cockroach (B. germanica) after 5 days of normal feeding following eclosion. The animal was provided with water only and maintained in constant darkness except every 2 h when it was stimulated with a 150-ms light pulse (light intensity = 0.0317 Lum/sf) to trigger an ERG response. The amplitude of the ERG fluctuated cyclically and peaked during the subjective night. B: An actogram for a starving adult female German cockroach under 28°C, constant darkness conditions. The legend is described in Figure 5. A and B were the typical representative of 5 and 10 starving females, respectively. minated, while the other responses display a more complex waveform (Armington, 1974). Based on the waveform of its ERG (Fig.1), the German cockroach expresses the first type of response. The negative potential of the ERG will remain at a steady level while the light is on and returns to the base line when the light is extinguished. These types of ERG responses can be found in relatively slow jumping and crawling insects, such as most cockroaches and crickets (Armington, 1974). In contrast, fast flying insects, such as blowflies and bees, have more complex ERG waveforms. Although the cockroach Leucophaea maderae is a slow-moving insect, it is a strong flyer that is reflected in the complex waveform of its ERG (Colwell and Page, 1989). In contrast, the minimal flight capability of the German cockroach (Cornwell, 1968) is correlated with simpler waveform of its ERG. Circadian Rhythms of ERG An increase in the number of photons reaching the photoreceptors should increase the amplitude of the ERG response (Colwell and Page, 1989). Kinetic analysis of the ERG in the German cockroach indicates that changes in ERG amplitude are primarily due to the changes in the probability of photons captured, and that the rate of rise of the ERG response is the reflection of the rate of photons captured by the photoreceptors. Based on these reasons, the daily changes in the ERG of 164 Chang and Lee Fig. 8. A: The amplitude changes of ERG responses for a pregnant female (carrying ootheca) German cockroach (B. germanica) 10 days after emergence. The animal was provided with water only and maintained in constant darkness except every 2 h when it was stimulated with a 150-ms light pulse (light intensity = 0.0317 Lum/sf) to trigger an ERG response. The amplitude of the ERG fluctuated cyclically and peaked during the subjective nigh. B: An actogram for a starving 10-day-old pregnant female German cockroach under 28°C, constant darkness conditions. The legend is described in Figure 5. A and B were the typical representative of 5 and 10 starving females, respectively. the German cockroach may be caused by the movements of screening pigment in the photoreceptor cells of retina, similar to that found in the housefly (Pyza and Meinertzhagen, 1997). The ERG has proven to be a useful, easy measurement of visual sensitivity in cockroach (Colwell and Page, 1989). Based on the finding of the current investigation, the high visual sensitivity of the German cockroach at night coincided with high locomotor activity (Lin and Lee, 1996). This matching of physiological function with a behavioral pattern is important for the German cockroach since it is a nocturnal animal. In the German cockroach, changes in ERG amplitude that peak during the subjective night show circa- dian rhythmicity (Fig. 5). From the analysis of ERG kinetics, the speed of ERG response is the main contributor of the three parameters to cause circadian rhythm in the ERG amplitude changes. Similar behavior has been reported for recordings of ERG amplitude from the cockroach L. maderae, although only two (off-transient and sustain component) out of three components of ERG show circadian rhythmicity (Colwell and Page, 1989). Inconsistency of Locomotor and Visual Sensitivity Circadian Rhythm A unique characteristic of the circadian system in the German cockroach is the existence of a masking on the locomotor circadian rhythm as Inconsistent Expression of Two Circadian Rhythms 165 Fig. 9. Daily changes of ERG response in the amplitude of a male German cockroach whose optic nerve was severed unilaterally. The animal was provided with water only and maintained in constant darkness except every 2 h when it was stimulated with a 150-ms light pulse (light intensity = 0.0317 Lum/sf) to trigger an ERG response. The amplitude of the ERG fluctuated cyclically and peaked during the subjective night. This is a typical representative of three operated male cockroaches. a result of the development of the ovaries (Lin and Lee, 1996). This masking effect can be removed either by ovariectomy (Lin and Lee, 1996) or by preventing ovarian development by starving the animal (Fig. 6). Once the females are pregnant (carrying ootheca), their locomotor activity is very depressed, thereby masking the observation of a locomotor circadian rhythm (Tsai and Lee, 2000). The only way to remove this masking effect is to prevent the formation of ootheca. Female adults carrying ootheca express circadian rhythmicity in their visual sensitivity even though their locomotor circadian rhythm is masked (Table 1). This inconsistency in the expression of locomotor and visual sensitivity circadian rhythms indicated the possibility that separate pacemakers exist in the German cockroach. Some research findings suggest that the circadian pacemaking system in the optic lobes of cockroach regulate locomotor activity, modulate photoreceptor sensitivity, and influence central processing of visual information (Wills et al., 1985). In the horseshoe crab Limulus polyphemus, a circadian clock in the brain has an important role in peripheral visual system (Kass and Barlow, 1992). The clock can transmit neural activity to the eyes via efferent optic nerve fibers. Since the locomotor pacemaker of the German cockroach is located in the optic lobes (Wen and Lee, 2000), we severed the optic nerves to examine whether the optic lobes control expression of the circadian rhythm for visual responses. It is possible that locomotor pacemaker can mediate ERG rhythm through humoral signals, since we did not remove the pacemaker in the surgery. For the regulation of visual sensitivity and locomotion in the cockroach Leucophaea maderae, there was a neural regulation between them (Colwell and Page, 1989). In addition, when considering the nature of ERG rhythm (it showed a rapid amplitude change pattern), it implied neural regulation. Therefore, our result may show the existence of a visual sensitivity pacemaker in the retina (Fig. 9). In cockroach L. maderae, the pacemakers for locomotion and visual sensitivity are either identical or located close to each other in the optic lobes (Wills et al., 1985; Colwell and Page, 1989, 1990). Although the offtransient of the ERG disappeared when the nerve connection between the lamina and the medulla of the optic lobe was cut, the sustained component still persisted (Colwell and Page, 1989). Since Colwell and Page (1989) did not discuss the persistence of the circadian rhythm for the sustained component, we doubt that the visual response pacemaker of L. maderae is the 166 Chang and Lee same one as the locomotor pacemaker. Based on this reasoning, we conclude that the pacemaker for visual responses is located in the eye, but not in the optic lobes. This inconsistency in the expression of locomotor and visual response circadian rhythms is supported by our finding that the masking factors from developing ovaries or pregnancy only acts on the locomotor pacemaking system. However, we do suggest that the pacemakers for locomotion and visual responses are strongly coupled or mutually interact (Wills et al., 1985). Since under some artificial conditions that allow both locomotion and visual responses to display circadian rhythmicity, their circadian periods are the same (Table 1). In addition, our results showed that the masking factors do not act on the locomotor pacemaker itself, but on the output pathway of the circadian system. Helfrich-Forster C, Stengl M, Homberg U. 1998. Organization of the circadian system in insects. Chronobiol Int 15:567–594. Kass L, Barlow R Jr. 1992. A circadian clock in the Limulus brain transmits synchronous efferent signals to all eyes. Vis Neurosci 9:493–504. Koehler PG, Agee HR, Leppla NC, Patterson RS. 1987. Spectral sensitivity and behavioral response to light quality in the German cockroach (Dictyoptera: Blattellidae). Ann Entomol Soc Am 80:820–822. Koehler WK, Fleissner G. 1978. Internal desynchronization of bilaterally organized circadian oscillators in the visual system of insects. Nature 174:708–710. Lee HJ, Wu YL. 1994. Mating effects on the feeding and locomotion of the German cockroach, Blattella germanica. Physiol Entomol 19:39–45. Leppla NC, Koehler PG, Agee HR. 1989. Circadian rhythms of the German cockroach (Dictyoptera: Blattellidae): locomotion in response to different photoperiods and wavelengths of light. J Insect Physiol 35:63–66. ACKNOWLEDGMENTS We thank Dr. En-Chang Yang for his technique support and critical review on the original draft of the manuscript. Thanks are also due to Dr. Charles Page of the Rutgers University for his critical review and constructive suggestions. This study was supported in part by a grant (NSC 89-2313-B-002030) from the National Science Council, R.O.C. LITERATURE CITED Armington JC. 1974. The electroretinogram. New York: Academic Press, 478 p. Autrum H. 1958. Electrophysiological analysis of the visual systems in insects. Exp Cell Res 5:426–439. Colwell CS, Page TL. 1989. The electroretinogram of the cockroach Leucophaea maderae. Comp Biochem Physiol 92:117–123. Colwell CS, Page TL. 1990. A circadian rhythm in neural activity can be recorded from the central nervous system of the cockroach. J Comp Physiol 166:643–649. Lin TM, Lee HJ. 1996. The expression of locomotor circadian rhythm in female German cockroach, Blattella germanica (L.). Chronobiol Int 13:81–91. Lin TM, Lee HJ. 1998. Parallel control mechanisms underlying locomotor activity and sexual receptivity of the female German cockroach, Blattella germanica (L.). J Insect Physiol 44:1039–1051. Pyza E, Meinertzhagen IA. 1997. Circadian rhythms in screening pigment and invaginating organelles in photoreceptor terminals of the housefly’s first optical neuropile. J Neurobiol 32:517–529. Sokolove PG, Bushell WN. 1978. The chi square periodogram, its utility for analysis of circadian rhythms. J Theor Biol 72:131–160. Sommer SH. 1975. Experimental investigation of the circadian locomotor activity of Blattella germanica L. (Dictyoptera: Blattellidae). Biol Zentralbl 94:455–467. Tsai CW, Lee HJ. 2000. Circadian locomotor rhythm mask by the female reproduction cycle in cockroaches. Physiol Entomol 25:63–73. Cornwell PB. 1968. The cockroach, vol. 1. A laboratory insect and an industrial pest. London: Hutchison Press. 391 p. Wen HW, Lee HJ. 2000. Unequal coupling between locomotor pacemakers of the German cockroach, Blattella germanica (L.). J Insect Physiol 46:89–97. Dreisig H, Nielsen ET. 1971. Circadian rhythm of locomotion and its temperature dependence in Blattella germanica. J Exp Biol 54:187–198. Wills SA, Page TL, Colwell CS. 1985. Circadian rhythms in the electroretinogram of cockroach. J Biol Rhythms 1:35–37.