Seasonal Variations in Circadian Rhythms of Plasma Melatonin in Ruin Lizards

Hormones and Behavior 41, 414 – 419 (2002) doi:10.1006/hbeh.2002.1781, available online at http://www.idealibrary.com on

Seasonal Variations in Circadian Rhythms of Plasma Melatonin in Ruin Lizards Cristiano Bertolucci,* Augusto Foa`,* and Thomas J. Van‘t Hof† ,1 *Dipartimento di Biologia, Universita` di Ferrara, via L. Borsari 46, 44100, Ferrara, Italy; and †Research Center for Ornithology of the Max Planck Society, Von-der-Tann Strasse 7, 82346, Andechs, Germany Received May 22, 2001; accepted December 11, 2001

We examined melatonin profiles of ruin lizards in different seasons (spring, summer, and autumn) under light: dark (LD) and concomitant responses when transferred to continuous darkness (DD) to determine the degree to which previously reported seasonally dependent effects of pinealectomy on locomotor behavior are related to melatonin secretion. The amplitude of the melatonin rhythm and the amount of melatonin produced over 24 h varied with season. In spring, the amount of melatonin produced was greatest and the amplitude 4 – 5 times that found in summer or autumn. The degree of self-sustainment of the melatonin rhythm when transferred to DD also varied with season. In DD, melatonin levels remained high but did not exhibit circadian variation in spring. In summer, the melatonin profile persisted virtually unchanged in DD, showing the existence of a circadian rhythm. Finally, in the fall there was no circadian variation in DD and levels remained low. These responses correspond closely to previously reported effects of pinealectomy on locomotor behavior where there is little or no effect of pinealectomy in spring or fall but a profound alteration of locomotor behavior in summer. These results suggest that the seasonally dependent effects of pinealectomy on locomotor behavior in ruin lizards are related to a seasonally mediated change in the degree of self-sustainment of some component of the circadian pace-making system of which melatonin plays some role. © 2002 Elsevier Science (USA) Key Words: melatonin; pineal; circadian rhythms; lizards; seasonality.

1

To whom correspondence and reprint requests should be addressed. Current address: Department of Biological Sciences, Wright State University, 3640 Colonel Glenn Hwy., Dayton, Ohio 45435-0001. Fax: (937) 775-3320. E-mail: thomas.vanthof@wright. edu.

414

INTRODUCTION In most vertebrates melatonin is synthesized mainly in the pineal gland and rapidly secreted into the blood. Whether measured in the pineal or the blood, melatonin oscillates with a daily rhythm with levels high at night and low during the day (Reiter, 1977; Arendt, 1987; Underwood, 1992). In some birds and lizards the isolated pineal cultured in vitro synthesizes melatonin with a circadian rhythm, which persists for several cycles in constant conditions, demonstrating the existence of circadian oscillators in the pineal that are coupled to melatonin synthesis (Binkley et al., 1978; Deguchi, 1979; Takahashi et al., 1980; Murakami et al., 1994; Tosini et al., 2000). Additional studies in the Iguanid lizard Anolis carolinensis have demonstrated that 24-h cycles of both light and temperature can entrain the pineal melatonin rhythm and that differences in length of daily photoperiod or thermoperiod affect the phase, amplitude and duration of this rhythm (Underwood, 1985; Underwood and Calaban, 1987). Hence, the current ambient lighting and temperature conditions are readily translated into an internal cue in the form of the pineal melatonin rhythm. This cue can be used to regulate both the daily and annual physiology of lizards (Underwood, 1985). In our model animal, the ruin lizard Podarcis sicula, seasonal differences in the locomotor activity pattern of intact animals kept in constant conditions were found to be associated with systematic differences in both the freerunning period (␶) of locomotor rhythms and the length of circadian activity (␣) (Foa` et al., 1994). The bimodal locomotor pattern expressed by ruin lizards in summer is typically associated with a short ␶ and a long ␣, whereas the unimodal pattern expressed in spring and autumn is typically associated with a long ␶ and short ␣ (Foa` et al., 1994). Melatonin

0018-506X/02 $35.00 © 2002 Elsevier Science (USA) All rights reserved.

415

Seasonal Melatonin in Lizards

and the pineal play a central role in this seasonal reorganization of the circadian system (Innocenti et al., 1994, 1996). Either pinealectomy or chronic administration of exogenous melatonin (implants) induces an immediate transition from the typical circadian locomotor pattern of summer (bimodal pattern, short ␶ and long ␣) to the typical circadian locomotor pattern of spring or autumn (unimodal pattern, long ␶ and short ␣) (Foa` et al., 1992b; Innocenti et al., 1994). Further experiments showed the existence of marked seasonal variations in the behavioral effects of pinealectomy in ruin lizards (Innocenti et al., 1996). In constant temperature and darkness (DD), changes in ␶ in response to pinealectomy were significantly greater in summer than in other seasons. Similarly, ␣ changed significantly in response to pinealectomy only in spring and summer. Again, while pinealectomy was effective in altering the locomotor rhythms of all lizards tested in summer, the same surgery left locomotor rhythmicity of many lizards tested in autumn and winter completely undisturbed. In summer, ruin lizards have robust circadian rhythms of plasma melatonin that become abolished in response to either pinealectomy or melatonin implants (Foa` et al., 1992a,b). The fact that in summer pinealectomy affects both locomotor behavior and suppresses plasma melatonin rhythms supports the hypothesis that the behavioral effects of pinealectomy are due to the withdrawal of rhythmic changes in plasma melatonin levels. If the hypothesis above is correct then the weak or absent behavioral effects of pinealectomy observed in winter, spring and autumn could be due to absence of a rhythmic production of melatonin during those seasons. The present investigation was aimed at determining if circadian rhythms of plasma melatonin are present in other seasons, and if there are seasonal variations in the self-sustaining properties of the melatonin rhythms when the lizards are placed in constant darkness. For this purpose, 24-h plasma melatonin profiles of ruin lizards maintained at constant temperature (29°C) in light:dark (LD) and DD were examined in different seasons.

MATERIALS AND METHODS Animals Ruin lizards (Podarcis sicula, De Betta 1857; adult male only, 6.5- to 8-cm snout vent length) were collected in April 1997 from the area of Ferrara. After

capture, lizards (n ⫽ 39) were transported to the lab, placed in groups of 4 –5 individuals inside plastic containers (43 ⫻ 26 ⫻ 15 cm) and kept in a vivarium under natural photoperiodic conditions. Approximately, 65 h before blood sampling, the containers were moved to environmental chambers kept at constant temperature (29 ⫾ 0.5°C) and subjected to either a light-dark (LD) cycle roughly corresponding to the natural photoperiod from the same season or to DD. In the chambers, lighting was provided by white fluorescent lamps with an intensity of 900 lx at the level of the head. Food (Tenebrio molitor larvae) and water were supplied twice a week. After each experiment the containers were brought back to the vivarium with natural lighting conditions until the next experiment. Blood Sampling Blood samples were taken from the lizards every two hours over a 24-h period. Each individual lizard was sampled at 6-hour intervals in a 24-h period and the median sample size per time point was 13 individuals. Each group of lizards was sampled at the same time of day during all tests. For the DD tests, the onset of the DD coincided with the light-to-dark transition of the LD cycle. Each test in DD started at 0800 on the third day after transfer to DD. In DD and during dark phase of the LD cycles, blood was collected with the aid of a weak red light (50 mW/m 2) from a fiber-optic light source. Sampling was accomplished by puncturing the infra-orbital sinus with the unpolished end of a 100 ␮l ammonium heparinized capillary tube (1.4 mm o.d., Drummond Scientific Company, Broomall, Pennsylvania, USA), and drawing 90 –260 ␮l of blood. Samples were immediately placed into a 1.5 ml microfuge tube and centrifuged for 12 min at 4000 rpm. Plasma was then aspirated and stored at ⫺ 80°C until assayed. Experimental Design The first sampling test in DD started in spring (May 21). Tests in DD were repeated in summer (August 06), and autumn (December 17) on the same lizards. Also control tests in LD were repeated at three different seasons: spring (May 07), summer (July 23), and autumn (December 03). In spring and summer, lizards were exposed to 14:10 h LD cycle, while in autumn lizards were exposed to 10:14 h LD cycle. These photoperiods were roughly comparable to natural photoperiods in the respective seasons.

416

Bertolucci, Foa`, and Van‘t Hof

RIA Analysis Melatonin was measured by radioimmunoassay after extraction and defattening following the procedure of Van’t Hof and Gwinner (1996, 1999). Briefly, 40 –125 ␮l of plasma was extracted twice in chloroform with 1 M NaOH, the chloroform fraction aspirated, dried under nitrogen, and redissolved overnight in assay buffer. The next morning the samples were washed with petroleum ether to remove lipids. The petroleum ether fraction was then aspirated and the buffer fraction incubated at 37°C for 10 min and then allowed to stand for at least one hour at 4°C to allow residual petroleum ether to evaporate before being aliquoted for assay. Intra- and interassay variations were 3.7 and 4.3%, respectively. To eliminate systematic error owing to interassay variability, samples were arranged so that representative samples from each time and season were extracted and assayed within each single assay. To validate the assay for use with ruin lizards, two methods were used. In the first, pooled nighttime plasma was serially diluted and extracted as above. In the second, daytime plasma was spiked with cold melatonin and serially diluted and extracted. In both cases, the plasma samples exhibited parallel displacement to the standard curve. The lower detection limit was 22 pg/ml. Statistical Analysis All results were expressed as mean ⫾ SEM. Oneway analysis of variance (ANOVA) was used to determinate significant differences among melatonin concentrations within experimental tests (e.g., spring DD, spring LD). If ANOVA showed an acceptable level of significance (P ⬍ 0.05), a Student–Newman– Keuls test was applied for post hoc comparison. To compare overall levels of melatonin in the different seasons, mean melatonin levels over a 24-h period were compared. All work presented here complies with current regulations covering animal experimentation in Italy.

RESULTS The results of the LD tests show the presence of robust daily rhythms of melatonin in the plasma in all seasons: melatonin levels were low during the light phase and high during the dark phase of the imposed cycles (Figs. 1A–1C). In all LD tests the maximum value of melatonin concentration (peak) occurred

FIG. 1. Seasonal variation of 24-h plasma melatonin profiles in ruin lizards in LD followed by the response in DD during spring, (A and D), summer (B and E), and autumn (C and F). Each point represents mean ⫾ SEM. White and black bars the duration of light and dark phases of the 24-h cycles. Sample sizes where: Spring LD n ⫽ 9, Spring DD n ⫽ 10, Summer LD n ⫽ 9, Summer DD n ⫽ 10, Autumn LD n ⫽ 9, and Autumn DD n ⫽ 9.

about 8 h after the light-to-dark transition. Melatonin peaked later, at 0400 in spring and summer, than in autumn when it peaked at 0200 (Student–Newman– Keuls test, P ⬍ 0.05). The spring peak was significantly higher (P ⬍ 0.05) than either the summer peak or the autumn peak. Summer and autumn peaks were not statistically different. Estimated melatonin production varied between the seasons. The mean melatonin level in spring lizards was significantly higher (P ⬍ 0.01) than in summer or in autumn (323.3 ⫾ 68.7 vs 77.6 ⫾ 14.2 and 86.4 ⫾ 11.6, respectively) and mean levels in summer and autumn lizards were similar (77.6 ⫾ 14.2 vs 86.4 ⫾ 11.6). There were no circadian rhythms in plasma melatonin after 60 – 63 h in DD in either spring or autumn (Figs. 1D, 1F). In contrast, in summer there was a clear circadian rhythm of plasma melatonin (Fig. 1E; ANOVA, F 11,96 ⫽ 3.43, P ⬍ 0.001) and melatonin reached a peak at 2400 (Fig. 1E; Student–Newman– Keuls test, P ⬍ 0.05). In summer, the peak of mela-

Seasonal Melatonin in Lizards

tonin in LD was not statistically different from the peak in DD. Mean plasma melatonin level over 24 h in summer was similar in DD and LD tests. In DD, the mean melatonin level was significantly greater (P ⬍ 0.01) in spring lizards than in summer or autumn lizards (421.7 ⫾ 53.2 vs 92.2 ⫾ 10.8 and 55.2 ⫾ 7.6, respectively). Furthermore, in DD summer lizards had significantly higher (P ⬍ 0.05) mean levels than in autumn lizards (92.2 ⫾ 10.8 vs 55.2 ⫾ 7.6).

DISCUSSION Several observations indicate that the circadian organization of ruin lizards changes dramatically with season. In the field, the daily pattern of locomotor activity changes from unimodal in spring to bimodal in summer and becomes unimodal again in autumn (Foa` et al., 1992; Tosini et al., 1992). In the laboratory at constant temperature (29°C) and DD, ruin lizards retain in their endogenous, freerunning locomotor rhythm the activity pattern typical of each season (bimodal/unimodal). In addition, the bimodal locomotor pattern expressed in summer is typically associated with a short freerunning period (␶) and a long circadian activity (␣), whereas the unimodal pattern expressed in the remaining months is typically associated with a long ␶ and short ␣ (Foa` et al., 1994). This demonstrates that the bimodal activity pattern is not merely a direct behavioral reaction of these ectothermic animals to the extremely high levels of soil temperatures around midday in summer, but reflects a seasondependent state of the circadian pacemaker that has evolved as an adaptation to high temperatures predictably occurring at that time of day (Foa` et al., 1994; Foa` and Bertolucci, in press). Melatonin and the pineal were previously shown to play a central role in the seasonal changes reported above, since in summer either pinealectomy or melatonin implants immediately abolishes the bimodal pattern, lengthens ␶ and shortens ␣, with the final effect of inducing a dramatic transition from the locomotor pattern of summer to the locomotor pattern of spring or autumn (Innocenti et al., 1994, 1996). Noteworthy, while pinealectomy is effective in altering circadian locomotor behavior of ruin lizards tested in summer, the same surgery has only weak or no behavioral effects in other seasons (Innocenti et al., 1996). Since in summer pinealectomy suppresses plasma melatonin rhythms, behavioral alterations in response to pinealectomy in summer were attributed to the abolition of the circadian melatonin signal following surgery. The present results confirm

417

the assumption above, as they show the disappearance of a circadian melatonin signal both in spring and autumn—seasons in which the behavioral effects of pinealectomy are weak or absent. Since ruin lizards show a bimodal locomotor activity pattern only in the presence of plasma melatonin rhythms, and are always unimodal in absence of these rhythms, the present results further support the view that circadian rhythms of plasma melatonin are involved in the induction and maintenance of the typical summer state of the circadian pacemaker, i.e., bimodal locomotor behavior (Foa` et al., 1992b, 1994; Innocenti et al., 1994, 1996). As the bimodal pattern is apparently a summer adaptation of the circadian system to high environmental temperatures, it is notable that the pineal is also involved in the regulation of the circadian rhythm of body temperature selection in ruin lizards (Innocenti et al., 1993). The fact that in summer pinealectomy actually abolishes the circadian rhythm of body temperature selection for several days suggests that melatonin is intimately involved in the regulation of behaviors leading to body temperature selection by ruin lizards during the most thermally critical season—summer, when risks of overheating are maximal (Tosini et al., 2001, for a review). Seasonal differences in daily amounts of plasma melatonin have been reported in a number of reptiles and birds (Underwood, 1992; Reierth et al., 1999; Brandsta¨ tter et al., 2001). In Australian Scincid lizards (Trachydosaurus rugosus), melatonin levels were lower in spring than in autumn (Firth et al., 1979). In the tortoise (Testudo hermanni) melatonin levels in the pineal where highest in summer and lowest in winter (Vivien-Roels et al., 1979) and in the red-sided garter snake (Thamnophis sirtalis parietalis) levels reached the highest amplitude in spring at emergence (Mendonic¸ a et al., 1995). In the ruin lizard, overall daily levels of plasma melatonin were greater in spring than in summer and autumn. These overall differences were also reflected in the levels of melatonin observed in DD. Thus, the striking seasonal differences in mean plasma melatonin level observed in ruin lizards could be associated with seasonal events such as reproduction or emergence from hibernation. The plasma melatonin rhythm expressed under LD cycles damps out completely in spring and autumn after transfer to DD and becomes arrythmic. This suggests that in both seasons, plasma melatonin production is not under circadian control (Figs. 1A, 1C, 1D, and 1F). In both spring and autumn the observed daily rhythm in LD may be due to direct melatonin suppression by light during light phase of the cycle or,

418

alternatively, to weakly coupled oscillators controlling melatonin production during those seasons. These weakly coupled oscillators still may be entrainable to LD cycles and produce daily melatonin rhythms. However, these oscillators, may go out of phase after three days in DD bringing about the observed arrhythmic melatonin profiles. In contrast, in summer, plasma melatonin profiles are clearly rhythmic after three days of DD, suggesting that in summer circadian oscillators controlling melatonin production are strongly coupled to each other (Figs. 1B and 1E). Although the pineal was shown to be the only source of summer circadian rhythms of plasma melatonin in ruin lizards, the presence of those rhythms does not demonstrate per se that the pineal gland of our summer lizards contains self-sustained oscillators coupled to melatonin synthesis (Foa` et al., 1992a). Rhythmic melatonin synthesis in summer could be driven by circadian oscillators located outside the pineal, as it occurs in mammals (Klein and Moore, 1979). A circadian clock coupled to melatonin synthesis/release was found in the cultured pineals of several Iguanid lizards, namely A. carolinensis, Sceloporus occidentalis, and Iguana iguana, but not in Dipsosaurus dorsalis, whose cultured pineals secrete melatonin in large— but arrhythmic— amounts (Menaker and Wisner, 1983; Menaker, 1985; Janik and Menaker, 1990; Tosini and Menaker, 1998). Overall, those Iguanid lizards offer an example of striking species-specific parallelism between effectiveness of pineal removal in altering overt circadian rhythmicity and presence of a circadian clock in the pineal. Removal of the pineal gland abolishes circadian rhythms of locomotor activity in A. carolinensis (pineal clock present), affects the period of the rhythm in S. occidentalis (pineal clock present), alters circadian rhythms of body temperature selection in I. iguana (pineal clock present), but leaves circadian rhythmicity completely undisturbed in D. dorsalis (pineal clock absent) (Underwood, 1981, 1983; Janik and Menaker, 1990; Tosini and Menaker, 1998). Interestingly, the situation we found in ruin lizards during spring or autumn where pinealectomy was not capable of altering circadian rhythmicity in the majority of the individuals suggests that the absence of influence of pinealectomy on locomotor rhythms of D. dorsalis may be a season-dependent phenomenon, instead of a species-specific feature (Janik and Menaker, 1990; Innocenti et al., 1996). Systematic studies of pinealectomized lizards in different seasons across several species are required to establish whether the differences found among Iguanid lizards are completely interspecific in nature or, depend in part, on the sea-

Bertolucci, Foa`, and Van‘t Hof

sons in which species were examined (for review: Tosini et al., 2001). So far, no in vitro culture studies on isolated pineals of ruin lizards have been carried out. However, the striking species-specific parallelism between involvement of the pineal in circadian organization and presence of a pineal circadian clock suggests that the summer plasma melatonin rhythm found in ruin lizards could be due to the appearance of a working pineal clock in summer (where pinealectomy affects the period of locomotor rhythms), and the arrhythmic plasma melatonin profiles observed in spring and autumn are due to the disappearance of a working pineal clock (in spring, autumn, and winter pinealectomy scarcely affects circadian locomotor rhythms). If this were true, then the pineal of ruin lizards may undergo a seasonally spontaneous “knock-out” of clock-driven melatonin synthesis, and thus provide a powerful model to shed light on the molecular basis of vertebrate clock mechanisms.

ACKNOWLEDGMENTS This work was supported by grants of the Italian Ministero dell’Universita` e della Ricerca Scientifica e Tecnologica (COFIN 1997, 1999), research grants of the Universita` di Ferrara (ex-60%), and the Research Center for Ornithology of the Max Planck Society. We wish to thank Prof. Eberhard Gwinner for general support of this work.

REFERENCES Arendt, J. (1995). Melatonin and the Mammalian Pineal Gland. Chapman and Hall, London. Binkley, S., Riebmann, J. B., and Reilly, K. B. (1978). The pineal gland: A biological clock in vitro. Science 202, 1198 –1202. Brandsta¨ tter, R., Kumar, V., Van’t Hof, T. J., and Gwinner, E. (2001). Seasonal variation of in vivo and in vitro melatonin production in the house sparrow. J. Pineal Res., in press. Deguchi, T. (1979). Ontogenesis and phylogenesis of circadian rhythm of serotonin N-acetyltransferase activity in the pineal gland. In M. Suda, O. Hayashi and H. Nakagawa (Eds.), Biological Rhythms and Their Central Mechanisms, pp. 158 –169. Elsevier, Amsterdam. Firth, B. T., Kennaway, D. J., and Rozenbilds, M. A. M. (1979). Plasma melatonin in the Scincid lizards, Trachydosaurus rugosus: Diel rhythms, seasonality, and the effect of constant light and constant darkness. Gen. Comp. Endocrinol. 37, 493–500. Foa`, A., and Bertolucci, C. (2001). Temperature cycles induce a bimodal activity pattern in ruin lizards: Masking or clock controlled event? A seasonal problem. J. Biol. Rhythms, in press. Foa`, A., Janik, D., and Minutini, L. (1992a). Circadian rhythms of plasma melatonin in the ruin lizard Podarcis sicula: Effects of pinealectomy. J. Pineal Res. 12, 109 –113.

Seasonal Melatonin in Lizards

Foa`, A., Minutini, L., and Innocenti, A. (1992b). Melatonin: A coupling device between oscillators in the circadian system of the ruin lizard Podarcis sicula. Comp. Biochem. Physiol. 103A(4), 719 – 723. Foa`, A., Monteforti, G., Minutini, L., Innocenti, A., Quaglieri, C., and Flamini, M. (1994). Seasonal changes of locomotor activity patterns in ruin lizards Podarcis sicula. I. Endogenous control by the circadian system. Behav. Ecol. Sociobiol. 34, 267–274. Innocenti, A., Minutini, L., and Foa`, A. (1994). Seasonal changes of locomotor activity patterns in ruin lizards Podarcis sicula. II. Involvement of the pineal. Behav. Ecol. Sociobiol. 35, 27–32. Innocenti, A., Bertolucci, C., Minutini, L., and Foa`, A. (1996). Seasonal variations of pineal involvement in the circadian organization of ruin lizards Podarcis sicula. J. Exp. Biol. 199, 1189 –1194. Janik, D. S., and Menaker, M. (1990). Circadian locomotor rhythms in the desert iguana. I. The role of the eyes and the pineal. J. Comp. Physiol. A 166, 803– 810. Klein, D. C., and Moore, R. Y. (1979). Pineal N-acetyltransferase and hydroxyindole-O-methyltransferase: Control by the retinohypothalamic tract and the suprachiasmatic nucleus. Brain Res. 174(2), 245–262. Menaker, M. (1985). Eyes—The second (and third) pineal glands? In D. Evered and S. Clarks (Eds.), Photoperiodism, Melatonin and the Pineal, pp. 39 –52. Pitman, London. Menaker, M., and Wisner, S. (1983). Temperature-compensated circadian clock in the pineal of Anolis. Proc. Natl. Acad. Sci. USA 80, 6119 – 6121. Mendonc¸ a, M. T., Tousignant, A. J., and Crews, D. (1995). Seasonal changes and annual variability in daily plasma melatonin in the red-sided garter snake (Thamnophis sirtalis parietalis). Gen. Comp. Endocrinol. 100, 226 –237. Murakami, N., Nakamura, H., Nishi, R., Marumoto, N., and Nasu, T. (1994). Comparison of circadian oscillation of melatonin release in pineal cells of house sparrow, pigeon and Japanese quail, using cell perfusion systems. Brain Res. 651(1–2), 209 –214. Reierth, E., Van’t Hof, T. J., and Stokkan, K-A. (1999). Seasonal and daily variations in plasma melaotnin in the high-arctic Svalbard

419

Ptarmigan (Lagopus mutus hyperboreus). J. Biol. Rhythms. 14, 314 – 319. Reiter, R. J. (1977). The Pineal. Eden Press, Montreal. Takahashi, J. S., and Menaker, M. (1982). Role of the suprachiasmatic nuclei in the circadian system of the house sparrow, Passer domesticus. J. Neurosci. 2, 815– 828. Tosini, G., and Menaker, M. (1998). Multioscillatory circadian organization in a vertebrate, Iguana iguana. J. Neurosci. 18, 1105–1114. Tosini, G., Bertolucci, C., and Foa`, A. (2001). The circadian system of reptiles: A true multioscillatory and multiphotoreceptive system. Behav. Physiol. 72(4), 461– 471. Underwood, H. (1981). Circadian organization in the lizard, Sceloporus occidentalis: The effects of blinding, pinealectomy and melatonin. J. Comp. Physiol. 141, 537–547. Underwood, H. (1983). Circadian organization in the lizard Anolis carolinensis: A multioscillatory system. J. Comp. Physiol. 152, 265– 274. Underwood, H. (1985). Circadian rhythms in lizards: Phase response curve for melatonin. J. Pineal Res. 3, 187–196. Underwood, H. (1992). Endogenous rhythms. In C. Gans (Ed.), Biology of Reptilia, Vol 18, pp. 229 –297. University of Chicago Press, Chicago. Underwood, H., and Calaban, M. (1987). Pineal melatonin rhythms in lizards Anolis carolinensis: I. Response to light and temperature cycles. J. Biol. Rhythms. 2, 179 –193. Van’t Hof, T. J., and Gwinner, E. (1996). Development of posthatching melatonin rhythms in zebra finches (Poephila guttata). Experientia. 52, 249 –252. Van’t Hof, T. J., and Gwinner, E. (1999). Influence of pinealectomy and pineal stalk deflection on circadian gastrointestinal tract melatonin rhythms in zebra finches (Taeniopygia guttata). J. Biol. Rhythms 14, 185–189. Vivien-Roels, B., Arendt, J., and Bradtke, J. (1979). Circadian and circannual fluctuations of pineal indoleamines (serotonin and melatonin) in Testudo hermanni Gmelin (Reptilia, Chelonia). Gen. Comp. Endocrinol. 37, 197–210.