Proximate and ultimate determinants of cathemeral activity in brown lemurs Giuseppe Donati a, * , Nicoletta Baldi b, 1 , Valentina Morelli b,1 , Jo ¨rg U. Ganzhorn c, 2 , Silvana M. Borgognini-Tarli b,1 a Nocturnal Primate Research Group, Department of Anthropology and Geography, Oxford Brookes University b Department of Biology, Unit of Anthropology, University of Pisa c Biozentrum Grindel, Department of Animal Ecology and Conservation, University of Hamburg article info Article history: Received 20 April 2008 Initial acceptance 2 July 2008 Final acceptance 27 September 2008 Published online 22 November 2008 MS. number: 08-00262 Keywords: cathemerality diel activity dietary needs Eulemur collaris Eulemur fulvus rufus lemur Madagascar masking factor Among primates, only two lemur genera, Eulemur and Hapalemur, and a population of the platyrrhine Aotus azarai are known to be cathemeral (i.e. show a mixture of daytime and night-time activity). Given the rarity of this phenomenon in primates except for Malagasy lemurs, its proximate and ultimate factors are still debated. The adaptive reasons for this apparently odd lifestyle are mainly based on four not mutually exclusive hypotheses: thermoregulatory benefits, antipredator strategy, competition avoidance and metabolic dietary-related needs. However, little effort has been made to tease apart proximate cues from potential ultimate factors in the multivariate context of activity correlates. To investigate these hypotheses, we compared the year-round cathemeral activity of two groups of Eulemur collaris and one group of hybrids E. collaris  Eulemur fulvus rufus in a humid littoral and in a dry gallery forest of southern Madagascar. Data were collected using a 5 min instantaneous method, with sampling equally distributed between day and night. We weighted the different effects of proximate and ultimate factors via the ANCOVA analysis using as the dependent variable the ratio between diurnal and nocturnal activity. Photoperiodic changes and nocturnal luminosity were the two proximate factors that accounted for most of the variability at the two sites. Diet quality was the only ultimate factor that had a significant effect on the diurnality index of the two lemur populations, suggesting a role of metabolic dietary-related needs in determining cathemeral activity in these lemurs. Ó 2008 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved. The consequences of being active during the day or at night have been pervasive at the beginning of each animal radiation, given the opposite sensorial worlds represented by the two segments of the 24 h cycle (Charles-Dominique 1975; Aschoff et al. 1982; Halle 2006). However, some mammals routinely show the ability to shift activity over the 24 h cycle despite the challenge of compromising between opposite adaptations (Halle & Stenseth 2000; Curtis & Rasmussen 2006). So far, the adaptive reasons for diel activity patterns in mammals are mainly based on four nonmutually exclusive explanations: thermoregulatory benefits (Beltran & Delibes 1994; Chiarello 1998; Zielinsky 2000; Rezende et al. 2003), predator–prey strategies (Zielinsky 1988; Lode 1995), avoidance of competition (Ziv et al. 1993; Fenn & Macdonald 1995; Kronfeld- Schor & Dayan 1999; Jones et al. 2001) and metabolic dietary- related needs (Halle 2000; Merritt & Vessey 2000). Two decades ago, a diel activity pattern (termed ‘cathemerality’) was described in primates, which were traditionally viewed as strictly nocturnal or strictly diurnal (Tattersall 1987). This lifestyle seems to be rather rare in this order compared to other mammalian groups (Curtis & Rasmussen 2006). So far, only two lemur genera, Eulemur and Hapalemur , and a platyrrhine population, Aotus azarai, are known to show a mixture of daytime and night-time activity (Curtis et al. 2006). Given the rarity of cathemeral activity in primates, there is an intense debate as to which proximate cues and ultimate factors are related to this flexible biorhythm. Two proxi- mate determinants, photoperiodic changes and nocturnal lumi- nosity, have been shown to influence the activity profile of cathemeral primates. As in many other taxa, photoperiodic changes seem mainly to entrain the activity rhythm, that is, they act as a Zeitgeber (Curtis et al. 1999; Donati & Borgognini-Tarli 2006; Fernandez-Duque & Erkert 2006), while low nocturnal luminosity seems to inhibit activity via a negative masking effect (Aschoff 1988; Erkert 1989; Kappeler & Erkert 2003; Donati & Borgognini- Tarli 2006; Fernandez-Duque & Erkert 2006). As for the ultimate determinants, the hypotheses proposed so far are in accordance with the selection pressures demonstrated for other mammals. Thermoregulatory advantages to avoid heat stress or cold stress (Aotus azarai: Fernandez-Duque 2003; Hapalemur griseus alao- trensis: Mutschler 1999; Eulemur mongoz: Curtis et al. 1999) and * Correspondence: G. Donati, Nocturnal Primate Research Group, Department of Anthropology and Geography, Oxford Brookes University, Oxford OX3 0BP, U.K. E-mail address: gdonati@brookes.ac.uk (G. Donati). 1 N. Baldi, V. Morelli & S. M. Borgognini-Tarli are at the Department of Biology, Unit of Anthropology, University of Pisa, Via Derna 1, 56126 Pisa, Italy. 2 J. U. Ganzhorn is at the Biozentrum Grindel, Department of Animal Ecology and Conservation, University of Hamburg, 20146 Hamburg, Germany. Contents lists available at ScienceDirect Animal Behaviour journal homepage: www.elsevier.com/locate/yanbe 0003-3472/$38.00 Ó 2008 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.anbehav.2008.09.033 Animal Behaviour 77 (2009) 317–325