, Volume 96, Issue 6, pp 737–741

Optional strategies for reduced metabolism in gray mouse lemurs


    • Department of Experimental EcologyUniversity of Ulm
  • J. U. Ganzhorn
    • Animal Ecology and Conservation, Biozentrum Grindel, Martin-Luther-King-Platz 3University of Hamburg
Short Communication

DOI: 10.1007/s00114-009-0523-z

Cite this article as:
Schmid, J. & Ganzhorn, J.U. Naturwissenschaften (2009) 96: 737. doi:10.1007/s00114-009-0523-z


Among the order of primates, torpor has been described only for the small Malagasy cheirogaleids Microcebus and Cheirogaleus. The nocturnal, gray mouse lemur, Microcebus murinus (approx. 60 g), is capable of entering into and spontaneously arousing from apparently daily torpor during the dry season in response to reduced temperatures and low food and water sources. Mark–recapture studies indicated that this primate species might also hibernate for several weeks, although physiological evidence is lacking. In the present study, we investigated patterns of body temperature in two free-ranging M. murinus during the austral winter using temperature-sensitive data loggers implanted subdermally. One lemur hibernated and remained inactive for 4 weeks. During this time, body temperature followed the ambient temperature passively with a minimum body temperature of 11.5°C, interrupted by irregular arousals to normothermic levels. Under the same conditions, the second individual displayed only short bouts of torpor in the early morning hours but maintained stable normothermic body temperatures throughout its nocturnal activity. Reduction of body temperature was less pronounced in the mouse lemur that utilized short bouts of torpor with a minimum value of 27°C. Despite the small sample size, our findings provide the first physiological confirmation that free-ranging individuals of M. murinus from the humid evergreen littoral rain forest have the option to utilize short torpor bouts or hibernation under the same conditions as two alternative energy-conserving physiological solutions to environmental constraints.


Microcebus murinusDaily torporHibernationBody temperature


Heterothermy, which includes daily torpor and hibernation, is the most extreme adaptation among mammals to cope with seasonal resource shortages and low temperatures (Wang 1989; Geiser 1998). Daily and prolonged torpor is characterized by a controlled reduction of body temperature (Tb) and metabolic rate that can reduce energy expenditure and water flux rate (Wang 1989; Geiser 2004). Daily torpor is usually employed during the daily rest phase and lasts only for several hours, whereas hibernation is characterized by a sequence of prolonged torpor bouts with mean bout duration of approximately 2 weeks (French 1982; Geiser and Ruf 1995; Geiser 2007; Bieber and Ruf 2009). Although widespread among mammals and birds, torpor is reported among primates only for the cheirogaleid family (Microcebus, mouse lemurs, and Cheirogaleus, dwarf lemurs) restricted to the island of Madagascar (Dausmann et al. 2004; Dausmann 2008). Cheirogaleus spp. (body mass ranges between 140 and 500 g) are obligate hibernators and show continuous hibernation with suspended thermoregulation for up to 7 months during the austral winter (Wright and Martin 1995; Dausmann et al. 2004, 2005). In contrast, Microcebus spp. (body mass ranges between 30 and 70 g) enter torpor during the cool dry season, even though only on a daily basis (Ortmann et al. 1997; Perret et al. 1998; Schmid 2000, 2001; Schmid and Speakman 2000, 2009; Perret and Aujard 2001; Randrianambinina et al. 2003). Based on mark–recapture data and control of sleeping sites, it has been proposed that adult female gray mouse lemurs (Microcebus murinus) remain inactive and extend torpor bouts and hibernate for some portion of the dry season (Schmid and Kappeler 1998; Schmid 1999). However, at present, there are no physiological data from M. murinus to confirm that individuals of this species are capable of hibernation. The specific aim of this study was therefore to investigate the occurrence of heterothermy by free-ranging M. murinus in two littoral rain forest fragments in southeastern Madagascar.

Material and methods

We examined body temperature patterns of free-ranging gray mouse lemurs in the evergreen littoral rain forest of Mandena (24°56′S, 46°59′E, 5–20 m above sea level), 10 km northeast of Tolagnaro (Fort Dauphin) in southeastern Madagascar (for a detailed description of the study site, see Ramanamanjato and Ganzhorn 2001; Schad et al. 2005). Mandena is a 10–15-m-high evergreen littoral rain forest with a thick understorey. The average yearly rainfall in the Fort Dauphin region is approximately 1,600 mm/year, and there is no distinct dry season. The average temperatures per month range from 20°C to 27°C with an annual mean of 23°C (Ramanamanjato and Ganzhorn 2001). The coolest months are June, July, and August. In June 2001, two female M. murinus were captured in two forest remnants (M13 and M16). M13 is a highly degraded forest fragment of about 152 ha without large trees, and M16 is a slightly degraded forest fragment of 75 ha (Ganzhorn et al. 2007). The lemurs were implanted with DS1922L iButtons ® (Dallas Semiconductor; accuracy 1°C from −20°C to 70°C) subdermally under deep anesthesia (0.05 ml/100 g body mass Ketavet: 100 mg/ml, 0.01/100 g body mass Rompun: 2% solution). The iButtons were programmed to record body temperature with a resolution of 0.5°C every 60 min for 85 consecutive days (6 July to 29 September 2001). Mouse lemurs were released at their site of capture. In June 2002, the animals were recaptured and the iButtons were removed. The animals were released at their site of capture. Ambient temperatures (Ta) were measured with iButtons at the two study sites. Mouse lemurs were considered to be torpid when the Tb was equal to or lower than 30°C (Geiser and Baudinette 1988). The total length of each torpor bout was defined as the time between the torpor onset when Tb fell below 30°C and the end of the torpor bout when Tb reached normothermic levels again. Hibernation was defined as the period of consecutive days (i.e. >24 h) during which Tb was not endogenously raised to normothermic values at dusk, when nocturnal activity would usually commence. For both individuals, we determined daily minimum, mean, and maximum Tb during torpor. Mann–Whitney U test was performed using SPSS 13.0. Numeric values are given as mean ± SD for the number of recordings “n.” The Kolmogorov–Smirnov goodness-of-fit test was used to test the distribution of the data for normality.


Measures of two physiological parameters of heterothermy (body temperature and length of torpor bouts) of M. murinus under natural conditions revealed that torpor bout length varied between individuals to an extent previously unknown for heterothermic mammals. For the female 1830 (body mass: 66 g) from a slightly degraded forest fragment, 24 torpor bouts were observed during all 85 days of data collection (i.e., on 28% of animal days; Fig. 1a, b). The mean time of entry into short bouts of torpor was 06.05 h (range 04.20 to 11.20 h). The torpor episodes lasted on average for 1 h (±0.83 h) with a maximum length of 5 h. The lemur then remained normothermic and, in all probability, also active throughout almost the entire night. The mean minimal Tb of female 1830 during torpor was 29.8 ± 1.2°C (n = 24 where n was the number of torpid days) at a mean minimal Ta of 13.9 ± 3.1°. The lowest absolute Tbmin recorded was 27°C.
Fig. 1

a Daily patterns of body temperature (Tbthick solid line) of a free-ranging M. murinus (1830; fragment M16) measured over 14 consecutive days in August 2001 with short bouts of torpor when Tb dropped below 30°C. Corresponding ambient temperature (Ta, thin solid line) was measured in the forest fragment M16. b Temperature patterns during 3 days as indicated. The black bars indicate night-time

In contrast, female 795A from a highly degraded forest fragment (body mass: 89 g) showed evidence of a hibernation season lasting for 30 days (3rd August–1st September; Fig. 2a). This female displayed long bouts of torpor daily and had only very brief normothermic periods during early afternoon. Here, Tb was not endogenously raised to normothermic values at dusk, when the mouse lemurs would usually start being active. Instead, after occasionally attaining a peak daily Tb of above 30°C in the afternoon, Tb decreased again following the decreasing Ta and the mouse lemur remained inactive in its sleeping place and continued hibernation (Fig. 2a, b). During the entire 30 days of inactivity, the female was normothermic for only 17.1% out of 720 h, and torpid for the remaining hours. The maximum torpor bout duration (i.e., Tb was always below 30°C) of female 795A was 89 h. The mean temperature difference between Tb and Ta during hibernation was 5.7°C (±6.2°C), with a maximum value of 17°C. The mean minimal Tb of female 795A (21.8 ± 6.4°C, n = 63; Ta: 16.1 ± 2.3°C) was significantly lower than the mean minimal Tb of female 1830 during short bouts of torpor (Mann–Whitney U test: z = −5.45, P < 0.001). For hibernating female 759A, the lowest absolute Tbmin recorded was 11.5°C.
Fig. 2

a Daily patterns of body temperature (Tb, thick solid line) of a free-ranging M. murinus (795A; fragment M13) measured over 36 consecutive days in August and September 2001 during hibernation. Corresponding ambient temperature (Ta, thin solid line) was measured in the forest fragment M13. b Temperature patterns during 3 days (22–24 August). The black bars indicate night-time

Mean minimal Ta recorded during torpor episodes in the slightly degraded forest fragment of female 1830 was significantly lower the mean Ta measured in the heavily degraded fragment of 759A (Mann–Whitney U test: z = −3.15, P = 0.002; for mean values, see previous paragraphs). Thus, although the ambient temperatures were lower in the slightly degraded fragment, the mouse lemur individual did not hibernate but employed short bouts of torpor.


The present study represents the first report of Tb in free-ranging M. murinus, where an individual hibernated for several weeks. Thus, we could confirm indirect mark–recapture data, which pointed out that parts of the M. murinus population extend torpor bouts and hibernate for some portion of the dry season (Schmid and Kappeler 1998; Schmid 1999). Here, the records of Tb in mouse lemurs illustrate that M. murinus have the option to extend torpor bouts and to hibernate for some portion of the dry season. During this time, its Tb passively followed Ta during its diurnal raise and nocturnal drop. This pattern of Tb is similar to that of the fat-tailed dwarf lemur (C. medius) during hibernation (Dausmann et al. 2004, 2005). This physiological study on C. medius from the dry deciduous forest indicated that, with no exception, all individuals of this species continuously hibernate for several months. In contrast, free-ranging individuals of M. murinus from the littoral rain forest have the option to utilize short bouts of torpor or hibernation as two alternative energy-conserving physiological solutions to environmental constraints. Thus, M. murinus represents the first Microcebus species that could, but does not have to, hibernate.

Most heterothermic mammals and birds appear to use either daily torpor or hibernation as adaptation to seasonal fluctuations in climate or energetic shortfalls. Nevertheless, it is likely that not all species conform to these two patterns of torpor but display some intermediate pattern (Geiser 2004; Willis et al. 2006). Optional strategies of heterothermy exhibited by M. murinus are comparable with that measured in lesser hedgehog tenrecs, Echinops telfairi, under seminatural conditions in Madagascar (Lovegrove and Génin 2008). Tenrecs entered daily torpor for several hours during winter, and one individual remained inactive on two occasions for periods of 2–4 days. In addition, a study on daily torpor of stripe-faced dunnarts (Sminthopsis macroura) in the wild revealed that one female aroused daily to normothermic Tb values around midday, but did not forage and remained inactive for over five successive days (Körtner and Geiser 2009). To our knowledge, these few examples represent the only mammals that show an intermediate pattern of hypometabolic states of this kind.

Generally, heterothermy is associated with unpredictable environments with moderate seasonality and not necessarily with constant low temperatures during winter (Geiser and Ruf 1995; Lovegrove 2000; Nicol and Andersen 2002; Dausmann et al. 2005; Warnecke et al. 2008). Factors that initiate hibernation instead of short bouts of torpor and sporadic activity in M. murinus from the littoral rain forest, however, remain unknown. In a dry deciduous forest of western Madagascar (Forêt de Kirindy/CFPF), mark–recapture studies indicated that intrasexual differences in the use of hibernation are linked to early dry season differences in body fat accumulation, as well as sex differences, where females above 60 g body mass remain inactive and possibly hibernate while males do not, regardless of their body mass (Schmid and Kappeler 1998; Schmid 1999). In the present study in the littoral rain forest, the hibernating mouse lemur was, in fact, fatter than the individual displaying short bouts of torpor. But the body mass of both animals was above the threshold required for hibernation in the dry deciduous forest. In contrast to previous studies on the physiological capabilities of mouse lemurs, where only short torpor bouts were observed during periods of energy and water shortage, the present study illustrates the potential of M. murinus to even maintain hibernation (Schmid and Speakman 2000; Schmid 2001).

In summary, this study shows that the use of short bouts of torpor and hibernation in M. murinus differs in different individuals of the same population. Even though the present study is based on only two individuals, it appears that thermoregulatory strategies within populations of species are less uniform than originally thought. It is likely that such transitional patterns of torpor are more widely employed among mammalian species than assumed so far.


We thank the Direction des Eaux et Forêts and the Commission Tripartite for their authorization to carry out this work. The study has been conducted within the framework of biodiversity assessment studies of the littoral forest fragments initiated by QIT Madagascar Minerals (QMM). J.-B. Ramanamanjato, M. Vincelette, and their environmental and conservation team of QMM provided excellent support in the field. This paper is part of the Accord de Collaboration between the Université d'Anananarivo (Départements de Biologie Animale and d'Anthropologie et Biologie Evolutive), QMM, and Hamburg University. Financial support from the German Research Foundation (SCHM 1391/2-1, 2-3, 2-4) is gratefully acknowledged. We declare that the experiments complied with the current laws in Madagascar (N° 101—DGDRF/SCB).

Copyright information

© Springer-Verlag 2009