Visual ecology of true lemurs suggests a cathemeral origin for the primate cone opsin polymorphism

Functional Ecology 2016, 30, 932 942 doi: 10.1111/1365-2435.12575 Visual ecology of true lemurs suggests a cathemeral origin for the primate cone opsin polymorphism Kim Valenta*,1, Melissa Edwards 2, Radoniaina R. Rafaliarison 3, Steig E. Johnson 4, Sheila M. Holmes 4, Kevin A. Brown 5, Nathaniel J. Dominy 6, Shawn M. Lehman 1, Esteban J. Parra 2 and Amanda D. Melin 7 1 Department of Anthropology & Archaeology, University of Toronto, 19 Russell St, Toronto, ON, Canada; 2 Department of Anthropology, University of Toronto at Mississauga, 3359 Mississauga Rd. North, Mississauga, QC, Canada; 3 Department of Paleontology, University of Antananarivo, Tsiadana, Antananarivo, Madagascar; 4 Department of Anthropology & Archaeology, University of Calgary, 2500 University Dr. NW., Calgary, QC, Canada; 5 Dalla Lana School of Public Health, University of Toronto, 155 College St., Toronto, QC, Canada; 6 Department of Anthropology, Dartmouth College, Silsby Hall, Hanover, NH, USA; and 7 Department of Anthropology, Washington University, One Brookings Dr., St. Louis, MO, USA Summary 1. In contrast to the majority of primates, which exhibit dedicated diurnality or nocturnality, all species of Eulemur are cathemeral. Colour vision, in particular, is strongly affected by the spectral composition and intensity of ambient light, and the impact of activity period on the evolution of primate colour vision is actively debated. 2. We studied three groups of wild brown lemurs (Eulemur fulvus) in Ankarafantsika National Park, Madagascar, over a 1-year span. We also used non-invasive faecal DNA collection and analysis to study the opsin genes underlying the colour vision of 24 individuals. We quantified the colour and brightness of dietary fruits and modelled the chromaticity and discriminability of food objects to different visual phenotypes under daylight, twilight and moonlight conditions. 3. We found that E. fulvus possesses routine dichromacy, unlike its congener E. flavifrons, for which polymorphic trichromacy has been reported. Our models suggest that dichromacy is well suited to the foraging ecology of E. fulvus. The performance of modelled dichromats and trichromats is comparable under nocturnal illuminants, and the luminance values of most diet items are detectable across light conditions. The trichromatic phenotype demonstrates a modest advantage under daylight conditions. 4. Our results, taken together with reports of polymorphic trichromacy in E. flavifrons, suggest functional ecological variation in the visual system of the genus Eulemur. Interspecific phenotypic variation in the colour vision of a genus is both unexpected and instructive. 5. Ecological differences between species of Eulemur could reveal thresholds for the origins of polymorphic trichromacy, which preceded the evolution of routine trichromatic vision in humans and other primates. Key-words: cathemerality, colour vision, dichromacy, Eulemur fulvus, trichromacy, visual foraging Introduction Animals that concentrate their activities during the day or night face profoundly different light environments. *Correspondence author. Kim Valenta, Department of Anthropology, McGill University. 855 Sherbrooke St. W. Montreal, QC 514-398-1242, Canada. E-mail: valentakim@gmail.com Ambient light levels can span six orders of magnitude between a starlit night (1E+06 photons/cm 2 /s/nm) and midday sunlight (1E+12 photons/cm 2 /s/nm) (Warrant & Johnsen 2013), leading to suites of visual specializations associated with nocturnal or diurnal activities (Kirk 2006). The visual system of diurnal animals features anatomical adaptations that enable high-acuity vision, including small corneas relative to eye diameter (Kirk 2004) and 2015 The Authors. Functional Ecology 2015 British Ecological Society

Opsin variation and ecology of cathemeral lemurs 933 cone-dominated retinas (Jacobs 2008). Nocturnal animals possess relatively enlarged eyes, a high degree of retinal summation and a tapetum lucidum (the structure responsible for eyeshine ). These traits maximize sensitivity at the expense of visual acuity (reviewed in Hall, Kamilar & Kirk 2012). As a general rule, these diurnal and nocturnal character traits are viewed as mutually incompatible (Ankel-Simons & Rasmussen 2008), and opportunities for studying adaptive shifts between activity patterns are few (Melin et al. 2013; Moritz et al. 2014). In the 1980s, another kind of activity pattern cathemerality was defined to describe behaviour occurring across the diel cycle (Tattersall 1987, 2006). The eyes of cathemeral mammals have attracted attention due to the conflicting demands on the visual system. Relative to strictly diurnal or nocturnal mammals, cathemeral species demonstrate traits of the visual system that are intermediate (e.g. eye size; Kirk 2004, 2006), or of variable presence (e.g. tapetum lucidum; Table 1), a pattern that suggests evolutionary compromise. Although some diurnal primate species are occasionally active at night, for example Alouatta pigra (Dahl & Hemingway 1988), true cathemerality is rare among primates (Curtis & Rasmussen 2006; Kirk 2006). The genus Eulemur (Strepsirrhini; Lemuridae) is unique in that every species is unambiguously cathemeral (Table 1). Accordingly, the eyes of Eulemur exhibit intermediate ocular morphologies (e.g. absolute eye diameters; the ratio of mean corneal diameter to transverse and axial eye diameters) suggesting that cathemerality is an ancient trait that has been under stabilizing selection for ~9 13 million years (Tattersall 1987; Kirk 2006; Griffin, Matthews & Nunn 2012). Compatible with this view is a recent phylogenetic analysis indicating that cathemerality preceded diurnality in lemurs, and is of relatively ancient origin (Santini et al. 2015). An alternative hypothesis for cathemerality was advanced by van Schaik & Kappeler (1996), who suggested that cathemeral species were in a state of evolutionary disequilibrium due to predator release. They argued that most lemurs are adapted to nocturnality and that modern cathemeral and diurnal activities are a response to the recent extinction of so-called giant subfossil lemurs and large diurnal raptors. This view has been challenged (Wright 1999), but the concept of a visual system in flux could account for puzzling variation within Eulemur such as the irregular presence of a tapetum lucidum (Wolin & Massopust 1970; Kirk 2006). Setting this debate aside, a cathemeral visual system in stasis or transition is attractive for interpreting the adaptive origins of trichromatic colour vision, a trait thought to be exclusive to diurnal primates until it was reported in Eulemur flavifrons (formerly E. macaco flavifrons, Veilleux & Bolnick 2009). Trichromatic colour vision is the result of three retinal cone types; it evolved from a dichromatic ancestor and has been linked to twin shifts in diet and activity pattern, from nocturnal insectivory to diurnal frugivory (Mollon 1989; Heesy & Ross 2001), although the timing of a nocturnal diurnal transition is debated (e.g. Tan et al. 2005; Santini et al. 2015). This view is based in part on widespread variation among primates. Monochromatic vision, the presence of a single functional opsin gene, exists in some nocturnal genera (e.g. Aotus, Nycticebus, Phaner) (Jacobs et al. 1993; Tan & Li 1999; Veilleux, Louis & Bolnick 2013), whereas polymorphic trichromatic vision, which results from a genetic polymorphism at the single locus of the X-chromosomal gene that codes for middle- to long-wavelength-sensitive (M/LWS) opsin proteins, characterizes some diurnal lemurs (Tan & Li 1999; Jacobs et al. 2002; Jacobs & Deegan 2003; Tan et al. 2005) and most New World monkeys (Jacobs et al. 1996). Heterozygous females are trichromatic, whereas males can inherit only one M/LWS (OPN1LW) opsin gene and are therefore dichromatic (Mollon, Bowmaker & Jacobs 1984). Finally, routine dichromatic colour vision, as inferred by two invariant opsin genes, can exist among nocturnal and diurnal species (Tan & Li 1999; Kawamura & Kubotera 2004; Melin et al. 2012, 2013). Given the purported utility of dichromacy under both nocturnal and diurnal ambient light conditions (Melin et al. 2007, 2010, 2013; Perry, Martin & Verrelli 2007), and costs that colour vision may impose on achromatic vision, cathemeral primates might be predicted to have routine dichromatic vision. Yet the X-linked photopigment opsin polymorphism of Eulemur flavifrons (Veilleux & Bolnick 2009) and the potential utility of trichromacy under mesopic light conditions (e.g. dawn, twilight) (Melin et al. 2013) suggests that trichromatic colour vision could be advantageous to cathemeral species. The trichromatic vision of some Eulemur taxa also challenges another long-held view: that highly acute vision precedes the evolution of primate trichromacy (Mollon, Estevez & Cavonius 1990; Boycott & Wassle 1999; Dacey 2000). Trichromatic primates typically have high-acuity vision due to the dense packing of cone photoreceptors; indeed, the retinal declivity (fovea) of haplorhine primates is unique among mammals, and it has greater resolving power than any other kind of retinal specialization (Moore et al. 2012). The retina of Eulemur, however, is afoveate and reported to have relatively low densities of cones (Peichl, Rakotondraparany & Kappeler 2001). This puzzling combination of traits trichromatic vision coupled with low cone densities raises the possibility of an incipient form of trichromacy. Such supposition casts new light on the role of cathemeral behaviour as an adaptive pathway. Despite the great deal of attention devoted to the study of primate colour vision, the visual ecology of cathemeral primates is poorly known (Ankel-Simons & Rasmussen 2008), limiting our ability to assess the factors that might have impelled the origin of trichromatic colour vision in primates. Here we directly address this deficit by investigating the opsin genes and food colours consumed by a population of wild brown lemurs (Eulemur fulvus). We genotyped the M/LWS opsin gene in order to determine

934 K. Valenta et al. Table 1. Variation in the activity patterns and visual systems of strepsirrhine primates Family Phylogeny* Genus (species) Activity Pattern Tapetum lucidum Colour Vision Phenotype** Lorisidae Loris Nocturnal Y Monochromat Galagidae Galago Nocturnal Y Monochromat Otolemur Nocturnal Y Monochromat Daubentoniidae Daubentonia Nocturnal Y Dichromat Indriidae Propithecus Diurnal Y M/LWS polymorphism Avahi Nocturnal Y Dichromat Indri Diurnal Y Cheirogaleidae Lepilemur Nocturnal Y Dichromat Cheirogaleus (major) Nocturnal Y Dichromat Cheirogaleus (medius) Nocturnal Y Monochromat Allocebus Nocturnal Y Monochromat Mirza Nocturnal Y Dichromat Microcebus Nocturnal Y Dichromat Phaner Nocturnal Y Monochromat Lemuridae Varecia Diurnal/Cathemeral N M/LWS polymorphism Lemur (catta) Cathemeral Y Dichromat Prolemur Cathemeral Y Hapalemur (alaotrensis) Cathemeral C? Hapalemur (meridionalis) Cathemeral C? Eulemur rubriventer Cathemeral C? E. macaco Cathemeral C N E. flavifrons Cathemeral C Y/N M/LWS polymorphism E. mongoz Cathemeral A? E. fulvus Cathemeral B/C N Dichromat E. sanfordi Cathemeral C? E. rufifrons Cathemeral B/C? E. collaris Cathemeral A? *Based on Horvath et al. (2008), Rumpler et al. (2011) and Weisrock et al. (2012). Prevailing subclassifications of cathemerality: A seasonal shift from diurnal to nocturnal activity; B seasonal shift from diurnal activity to 24-hr activity; C 24-hr activity year-round. For Eulemur, type C is observed in all rain forest habitats, whereas types A and B are observed only in seasonally dry habitats (Curtis & Rasmussen 2002). Cathemeral assignments follow Wright (1999), Curtis & Rasmussen (2002), Schwitzer et al. (2007) and LaFleur et al. 2014. Cathemeral subtype (A, B or C) is undetermined. Tapetum lucidum data sources: Pariente (1970), Wolin & Massopust (1970), Noback (1975), Pariente (1976, 1979), Geissmann (2002), Jacobs et al. (2002), Schwitzer, Kaumanns & Zahner (2005), Tan et al. (2005); and personal communications from E. Ehmke, R. Schopler, F. Spector, M. Uhl and J. Dowling (Eulemur fulvus). Contradictory evidence for the presence or absence of a tapetum lucidum. **Colour vision phenotype data sources: Tan & Li (1999); Heesy & Ross (2001); Jacobs et al. (2002); Veilleux & Bolnick (2009); Carvalho et al. (2012); Kamilar, Heesy & Bradley (2013); Veilleux, Louis & Bolnick (2013); Veilleux et al. (2014). Published accounts uniformly report dichromatic colour vision, but at least one unpublished finding indicates the presence of polymorphic trichromatic vision in a single individual of Lemur catta. Published colour vision types are reported. Rain forest-inhabiting diurnal and cathemeral C species are predicted to express the M/ LWS cone opsin polymorphism; other species are predicted to be dichromatic. the colour vision phenotype of E. fulvus. In addition, we measured the spectral composition of foods in the diet of E. fulvus and modelled the utility of dichromatic and trichromatic colour vision for discriminating foods with a perception-based analysis premised on just noticeable difference (JND) of chromatic and achromatic (luminance) contrast. If a majority of important food items are modelled to be more conspicuous to trichromatic lemurs across lighting conditions, we predict natural selection to favour trichromatic colour vision opsin polymorphism in E. fulvus. If food items are of similar conspicuousness to dichromats and trichromats, or more conspicuous to

Opsin variation and ecology of cathemeral lemurs 935 dichromats, then we predict dichromacy and an invariant M/LWS opsin gene. Materials and methods STUDY SITE AND SPECIES From January to December 2012, two observers (KV and RR) followed three habituated groups of common brown lemurs (Eulemur fulvus) with home ranges adjacent to the Ampijoroa forestry station (1618717 S; 4648972 E) in the tropical dry forest of Ankarafantsika National Park, north-western Madagascar. Groups A and B were habituated during a pilot season from July to December 2011. Group C was habituated to observer presence in May of 2012. Group sizes ranged from 5 to 10 individuals at any one time, with several individuals disappearing from groups, and some new immigrants appearing over the course of the study period. All study individuals were identifiable based on sex, age, scars and variation in pelage. FRUIT COLOUR SAMPLING Observers conducted 5-min continuous focal animal follows during the wet season (January to May, 2012) and 10-min focal animal follows during the dry season (May to December, 2012) between 5:00 and 18:30 h (=13655 total contact hours). For each food resource consumed by the study groups, we recorded the local or scientific name of the species and collected botanical specimens. We collected ripe fruits and leaves from one individual tree of each tree species (N = 56) in which focal animals were observed feeding. All fruit and leaf samples were returned to the field laboratory and analysed within ~2 h of collection. The reflectance spectra of ripe fruits (targets) and upper leaf surfaces (backgrounds) were measured relative to a Spectralon white reflectance standard (Labsphere) in the field using a Jaz portable spectrometer and a PX-2 pulsed xenon lamp (Ocean Optics) emitting a D-65 light source. For each species of fruit, one individual ripe fruit and one individual leaf were selected for measurement. The fruit scanning angle was fixed at 45, and external light was blocked using thick black fabric. DNA COLLECTION AND ANALYSIS We collected faecal samples from individual lemurs during group follows under viewing conditions where individuals could be definitively identified. Faecal samples were collected and stored using the two-step procedure (Nsubuga et al. 2004); faeces (~5 g) were collected in 50-mL tubes containing 30 ml of 97% ethanol. Ethanol-stored samples were mixed by inversion, and, after 24 36 h, decanted before transferring the solid material to 50-mL tubes containing ~20 g of silica gel beads (Sigma S7625; Sigma, Oakville, Canada). We collected 136 faecal samples from 27 individuals (1 6 samples per individual). Whenever possible, a minimum of five faecal samples was collected from each individual. In five cases, lemurs disappeared from study groups before it was possible to obtain any faecal samples. DNA was extracted from lemur faecal samples using the QIAamp DNA stool mini kit (QIAGEN, Toronto, Canada). The quantity and quality of each extraction product was measured using a NanoDrop spectrophotometer (Thermo Fisher Scientific, Ottawa, Canada). In total, we were able to obtain DNA from 26 of the 27 sampled animals. Studies of strepsirrhine primates indicate a sole amino acid site responsible for colour-vision polymorphism (G/C) at position 285 on exon 5 of the M/LWS opsin gene (Jacobs & Deegan 1993; Yokoyama & Radlwimmer 1998; Tan & Li 1999; Veilleux & Bolnick 2009). Based on sequences of E. flavifrons available in GenBank (accession number FJ228726.1), we designed two primers in order to amplify the relevant region: gcatggtggtggtgatgat (forward) and tcagagggtggaaggcatag (reverse). We ran 30-lL PCRs that contained 1X CoralLoad PCR Buffer, 25 mm of MgCl2, 240 lm of dntps, 025 lm of both the forward and reverse primers and 075 units of HotStarTaq DNA Polymerase. As the M/LWS opsin gene is located on the X chromosome, on average samples recovered from female lemurs will have double the target DNA. To minimize this bias, 3 ll of extraction product from female lemurs were added to the PCRs, whereas 6 ll of extraction product from male samples were added. Whenever possible, at least two samples from each individual were analysed. The PCR temperature was initiated at 95 C for 5 min to denature the DNA, followed by 36 cycles of 95 C for 30 s, 65 C for 30 s and 72 C for 30 s. These steps were followed by a final 10-min annealing step at 72 C. PCR products were sent to The Centre for Applied Genomics (TCAG, The Hospital for Sick Children, University of Toronto) for Sanger sequencing. Sequencing was carried out on both the forward and reverse strands. The sequences were inspected for the presence of polymorphisms using the FinchTV viewer (http:// www.geospiza.com/ftvdlinfo.html). To be classified as a polymorphism, we required (i) the presence of two peaks of approximately similar size (for female heterozygotes) visible in both forward and reverse strands, or in different sequencing reads of a single strand or (ii) by the presence of different bases at the same tuning site of different individuals. COLOUR MODELLING Chromatic and luminance values of targets (fruits) and backgrounds (leaves), along with the chromatic contrast between these items for each species were calculated for Eulemur colour vision phenotypes following established methods (Hiramatsu et al. 2008; Supporting Information). We followed the dim illumination version of the JND model to estimate the visibility of each target (Osorio et al. 2004; Matsumoto et al. 2014). The JND approach (Osorio & Vorobyev 1996; Higham et al. 2010) is useful for comparing the relative chromatic performance of dichromats and trichromats for discriminating food from background objects. Developed, and most often used, for anthropoid primates, the extension of the JND model in the present study may not exactly predict the abilities of lemurs, which have, among other differences in their visual systems, a much lower density of cones (Peichl, Rakotondraparany & Kappeler 2001). To address this, we analyse our results using a range of JND cut-off values and discuss the alternate interpretations. The parameters of the JND model mirror those presented by Matsumoto et al. (2014) and are included in Supporting Information. To represent the varying ambient light conditions important to cathemeral lemurs, we used three different irradiance spectra: diurnal forest shade (Melin et al. 2009), downwelling moonlight in a large forest gap (Melin et al. 2012) and a Madagascar-specific spectrum of unobstructed twilight. The former two measurements were not made in Madagascar; however, the spectral compositions of down-welling light in similar habitats, here tropical forest shade and large forest gaps, are comparable across localities and not anticipated to bias our results (Endler 1993). The twilight spectrum was recorded at the Kianjavato Ahmanson Research Station, south-eastern Madagascar (21381285 S, 47897342 E) using a multichannel spectrometer with a highly sensitive photomultiplier detector and an integrating sphere to ensure a cosine angular response (OL-770VIS; Gooch & Housego). The spectrum was measured at 1-nm intervals from 380 to 700 nm on 27 June 2012 during civil twilight (17:20 h) and in the absence of moonlight.

936 K. Valenta et al. STATISTICAL ANALYSES We first compared the percentage of fruits with the standard JND cut-off value of >1 for trichromats vs. dichromats using the two proportion z-test, for each of the three illuminants. To assess the impact of shifting the JND threshold, we also ran similar tests but with the JND cut-off set to 05, 075 and 2 JND. To test whether fruits and leaves differ from one another chromatically or achromatically, we ran one-way ANOVAs of the blue-yellow chromaticity values of fruits vs. leaves, the red-green chromaticity values of fruits vs. leaves, and the luminance values of fruits vs. leaves. We ran these under all three illuminant conditions. To examine whether the visual channels expressed by fruit are associated with one another, we ran bivariate linear regression models between B-Y chromaticity and R-G chromaticity, B-Y chromaticity and luminance, and R-G chromaticity and luminance for fruit and leaves under each illuminant (3 9 2 9 3 = 18 models). In addition, we measured the B-Y chromatic, R-G chromatic and luminance distances between fruits and leaves for each species (i.e. the R-G chromaticity Leaves R-G chromaticity Fruit, etc.) and ran similar bivariate models under each of the three illuminants (3 9 3 = 9 models). For each bivariate model, we extracted the correlation coefficients r and the P-value corresponding to the regression F-statistic. For all tests, statistical significance was set to the two-tailed P < 005-level; all analyses were conducted in R version 312 (R Development Core Team 2014). Results SEQUENCING Sanger sequencing reads were successful for 24 of the 26 animals for which we obtained DNA. For 21 out of the 24 individuals, there were multiple sequences from independently collected faecal samples available, and for 20 of these 21 individuals, sequence data were successfully obtained from the forward and reverse strands. The analysis of the sequencing data indicated that: (i) E. fulvus shows two fixed differences with respect to the sequence reported for E. flavifrons, corresponding to positions 78 (C in E. flavifrons vs. T in E. fulvus) and 108 (C in E. flavifrons vs. T in E. fulvus) of GenBank sequence FJ228726.1 (Fig. 1a). However, both are synonymous changes that do not entail amino acid change (isoleucine for position 78 and tyrosine for position 108); (ii) contrary to what has been described for E. flavifrons, we did not find a polymorphism at site 285 of exon 5 in E. fulvus. All of the individuals sequenced in our study had a guanine base, corresponding to the middle wavelength (~545 nm) sensitive opsin (Fig. 1b); and (iii) no polymorphisms were identified at any other site of the sequenced region, which encompasses ~200 bases and includes site 277, which is a known tuning site in other primates (Fig. 1b). COLOUR MODELLING For trichromats under the diurnal illuminant, the chromaticity distribution of fruits did not significantly differ from leaves in either the blue-yellow colour channel (F = 109, N = 56 per group, P = 030; Fig. S1) or the luminance channel (F = 002, N = 56 per group, P = 089), but it did differ significantly in the red-green colour channel (F = 3472, N = 56 per group, P=<0001, Fig. 2). Similarly, for dichromats under the diurnal illuminant, the chromaticity distribution of fruits did not differ from leaves in either the blue-yellow colour channel (F = 0619, N = 56 per group, P = 043) or the luminance channel (F = 0004, N = 56 per group, P = 095; Fig. S2). These patterns remained consistent under twilight and moonlight illuminants (Figs S3 and S4). Under the diurnal illuminant, blue-yellow and red-green chromaticity values of fruits were significantly negatively correlated (N = 56, r = 054, slope = 56, P=<0001), as were blue-yellow chromaticity and luminance values (N = 56, r = 071, slope = 034, P=<0001). Luminance values were positively correlated with red-green chromaticity values (N = 56, r = 031, slope = 67, P = 002; Fig. 2). These patterns were consistent under twilight and moonlight (Fig. S2; Table S1). Under the daylight illuminant, blue-yellow chromatic contrast and luminance contrast between fruits and leaves were significantly positively correlated (N = 56, r = 031, slope = 018, P = 002), and the association between blue-yellow chromatic contrast and red-green chromatic contrast was negative but not significant (N = 56, r = 026, slope = 19, P = 005), while no correlation or trend between luminance and redgreen chromatic contrast was discernible (N = 56, r = 004, slope = 058, P = 074, Fig. 3). Results were qualitatively similar for the other illuminants (Figs S5 and S6; Table S2). We found that 80% of fruit species had chromatic JND values >1 for the modelled trichromatic phenotype (45/56) under the diurnal illuminant, signifying that fruits of these species are chromatically conspicuous against a background of leaves. For the E. fulvus dichromatic phenotype, 57% of fruit species (32/56) had chromatic JND values >1, and this difference between trichromats and dichromats was significant (z = 265, P=<0001). Under the twilight illuminant, 39% of fruit species had chromatic JND values >1 for the trichromatic phenotype (22/56), while 30% (17/ 56) of the fruit species had chromatic JND values >1 for the E. fulvus dichromatic phenotype; however, this difference was not significant (z = 099, P = 032). Under the moonlight illuminant one fruit species had a chromatic JND value >1 for the trichromatic but not for the dichromatic phenotype (Fig. 4a). Using a less-stringent chromatic JND cut-off (JND >075 or >05), we found the same patterns; however, the twilight comparison between dichromatic and trichromatic phenotypes became significant at the JND >05 cut-off (Table 2). Alternatively, by increasing the JND cut-off to >2, the trichromat advantage disappeared under each illuminant (Table 2). With regard to achromatic JND values, 95% of fruits (53/56) had luminance JND values >1 for both the trichromatic and the dichromatic phenotypes under the diurnal illuminant. Under the twilight illuminant, 88% of fruits (49/56) had JND values >1 for the trichromatic phenotype, while 86% of fruits (48/56) had JND values >1 for the

Opsin variation and ecology of cathemeral lemurs 937 (a) i ii (b) Fig. 1. (a) M/LWS opsin gene sequence. Sequences of the critical amino acid position for (i) Eulemur fulvus with the Alanine allele (codon GCC), and (ii) a polymorphic Eulemur flavifrons female. (b) Comparison of exon 5 between E. fulvus and E. flavifrons. R denotes a G/C polymorphism at position 285 of exon 5 in E. flavifrons. dichromatic phenotype. These differences between colour vision types were not significant. Similarly, under the moonlight illuminant, 70% (39/56) of fruits had JND values >1 for both the trichromatic and dichromatic phenotypes (Fig. 4b). Adjusting the cut-off values for achromatic JND did not change our conclusions about the relative performance of dichromats and trichromats (Table 2). Discussion As predicted, dichromatic colour vision has measurable value for the foraging ecology of cathemeral lemurs. Our findings also suggest that dichromatic vision is uniformly present in this wild population of E. fulvus. This result replicates the findings of earlier studies based on a handful of captive animals (Jacobs & Deegan 1993; Tan & Li 1999), suggesting that all members of the species are dichromatic. Yet the sum total of X chromosomes investigated (n = 1, Jacobs & Deegan 1993; n = 5, Tan & Li 1999; n = 36, present study; total = 42) is perhaps inadequate to be conclusive. To explore this premise, we calculated the prevalence estimate and lower and upper confidence intervals (CIs) of obtaining 42 identical alleles out of 42 sequences sampled, given a binomial distribution that assumes two alleles in equilibrium, and a confidence level of 095. The Clopper Pearson exact binomial method (estimated prevalence, lower CI, upper CI: 1, 0916, 1) and the less conservative Jeffreys method (estimated prevalence, lower CI, upper CI: 1, 0942, 1) of estimating prevalence and CIs from a single proportion (Brown, Cat & DasGupta 2001) both favour a prevalence of 1, suggesting that E. fulvus is routinely dichromatic, or, at the very least, that allelic trichromacy is exceedingly rare. Our models of achromatic and chromatic properties suggest that, overall, the conspicuousness of food items is comparable for both colour vision phenotypes (observed dichromatic; hypothetical trichromatic) across a range of illumination conditions. Furthermore, the positive correlation between blue-yellow and luminance contrast between fruits and leaves could act to reinforce the visual localization of fruit. Although we did not collect foraging data at night, we observed high overlap in the diurnal and nocturnal diets of E fulvus. This view is consistent with nocturnal and diurnal observations (825% dietary overlap on average) of E. collaris (Donati et al. 2007). Similarly, the nocturnal diet of E. macaco resembles its diurnal diet, although the latter is more diverse (Andrews & Birkinshaw 1998). Thus, we believe that our sample is representative of the diet. The achromatic conspicuity of foods was equivalent for dichromatic and trichromatic phenotypes across all illumination conditions; only under diurnal illuminations did we detect a trichromatic advantage on the basis of chromatic JND values >1. This advantage extended into twilight conditions when JND detection thresholds were lowered to 05, suggesting a modest advantage under mesopic conditions. However, the predicted advantages to trichromats disappeared under all illuminants when JND thresholds were raised above 2. Taken together, dichromatic vision of E. fulvus appears to be adequate for accomplishing most or all foraging tasks, or alternatively, the benefits of moonlight (Donati et al. 2009) luminance discrimination offset the marginal advantages of trichromatic vision (Morgan, Adam & Mollon 1992; Vorobyev 1997; Saito et al. 2005; Melin et al. 2007, 2010; Caine, Osorio & Mundy 2010; Kelber & Lind 2010). We acknowledge, however, that our model is based on the invariant absence of a tapetum lucidum, a topic of debate with respect to Eulemur. We never detected a tapetum lucidum during field observations of E. fulvus, an impression that is corroborated in captivity (R. Schopler, F. Spector, pers. comm.). Other species of Eulemur appear to be devoid of this sensitivity-enhancing structure, although the histological evidence is contradictory and uncertain (Castenholz 1965;

938 K. Valenta et al. (a) (b) (c) Blue Yellow [S/(L+M)] 0 1 0 2 0 3 0 4 0 5 0 6 Leaves Ripe fruits Luminance 3 0 3 2 3 4 3 6 3 8 4 0 Leaves Ripe fruits Blue Yellow [S/(L+M)] 0 1 0 2 0 3 0 4 0 5 0 6 Leaves Ripe fruits 3 0 3 2 3 4 3 6 3 8 4 0 Luminance 0 50 0 51 0 52 0 53 0 54 0 55 Red Green [L/(L+M)] 0 50 0 51 0 52 0 53 0 54 0 55 Red Green [L/(L+M)] Fig. 2. Scatterplots showing relationships between chromatic and luminance values of ripe fruits consumed by Eulemur fulvus, and leaves of the same species (N = 56) under a diurnal illuminant. Ripe fruits: (a) blue-yellow chromaticity values and luminance values are negatively correlated (r = 071, slope = 034, P =<0001), (b) luminance and red-green chromaticity values are positively correlated (r = 031, slope = 67, P = 002) and (c) blue-yellow and red-green chromaticity values are negatively correlated (r = 054, slope = 56, P =<0001). Leaves: (a) blue-yellow chromaticity values and luminance values are negatively correlated (r = 041, slope = 021, P = 0002), (b) luminance and red-green chromaticity values are positively correlated (r = 055, slope = 23, P < 0001), and (c) blue-yellow and red-green chromaticity values are negatively correlated (r = 074, slope = 16, P < 0001). (a) (b) (c) Blue Yellow contrast [S/(L+M)Fruit S/(L+M)Leaf] 0 0 0 1 0 2 0 3 0 4 Luminance contrast [LuminanceFruit LuminanceLeaf] 0 0 0 2 0 4 0 6 0 8 Blue Yellow contrast [S/(L+M)Fruit S/(L+M)Leaf] 0 0 0 1 0 2 0 3 0 4 0 0 0 2 0 4 0 6 0 8 Luminance contrast [LuminanceFruit LuminanceLeaf] 0 00 0 01 0 02 0 03 0 04 0 05 Red Green contrast [L/(L+M)Fruit L/(L+M)Leaf] 0 00 0 01 0 02 0 03 0 04 0 05 Red Green contrast [L/(L+M)Fruit L/(L+M)Leaf] Fig. 3. Scatterplots showing relationships between different axes of chromatic and achromatic distance between fruits and leaves of the same species (N = 56 in all cases), (a) blue-yellow chromatic contrast and luminance contrast (r = 031, slope=018, P = 002), (b) luminance contrast and red-green chromatic contrast (r = 004, slope=058, P = 074) and (c) blue-yellow chromatic contrast and red-green chromatic contrast (r = 026, slope = 19, P = 005), under a diurnal illuminant. Kirk 2006). Field reports variously describe the presence or absence of eyeshine in closely related species (e.g. E. (macaco) macaco absent, Colquhoun 1997; E (macaco) flavifrons present, Schwitzer et al. 2007; but see Fig. S7). Such incongruities suggest an underappreciated level of variation in the visual system of lemurs generally and Eulemur specifically (Table 1). This variation is unparalleled among mammalian taxa and could be associated with niche partitioning on the basis of mean light intensities (Yamashita et al. 2005); or, alternatively, senses other than colour discrimination may mitigate the importance of vision for determining foraging success (Valenta et al. 2013). Olfaction is a highly developed sense in Eulemur and, like most strepsirrhine primates, it retains extensive neuroanatomical structures associated with enhanced olfactory discrimination, including moist rhinaria, large olfactory bulbs, well-developed vomeronasal organs and accessory olfactory systems (Barton, Purvis & Harvey 1995; Siemers et al. 2007; Valenta et al. 2013). Further research on the integration of different sensory systems at molecular, anatomical and behavioural levels will help address these issues. THE ORIGINS OF TRICHROMATIC COLOUR VISION: INSIGHTS FROM CATHEMERAL PRIMATES Here we report routine dichromacy in a wild population of E. fulvus. The existence of allelic trichromacy in a congener, E. flavifrons, invites a functional explanation for variation in the genus. If E. flavifrons is active under higher light levels more often than E. fulvus, for example, this would suggest a positive relationship between trichromatic vision and increasing levels of photopic, or mesopic (twilight), activity. Interspecific comparisons in similar habitats, using standardized data collection, are essential to address this question (Donati & Borgognini-Tarli 2006). Recent work has also revealed cathemerality, with stronger diurnal biases, in Lemur and Hapalemur, exemplifying the large variation in activity pattern among cathemeral species (Donati et al. 2013; Eppley, Ganzhorn & Donati 2015). We suggest, as an area deserving of future investigation, that differences in the visual ecology of cathemeral lemurs, in association with median light levels, may span a critical threshold above which trichromatic vision is adaptive.

Opsin variation and ecology of cathemeral lemurs 939 (a) 60 % JND>1 % JND>1 50 40 30 20 10 0 (b) 60 50 40 30 20 10 0 Diumal Twilight Nocturnal Diumal Twilight Nocturnal Trichromat Dichromat Fig. 4. Percentage of fruits with a JND > 1 in the (a) the chromatic blue-yellow and red-green colour channels of trichromats and the blue-yellow colour channel of dichromats, and (b) the achromatic luminance channel, for dichromats and trichromats, under three illuminant conditions (diurnal, twilight, nocturnal). Differences in the environment and colouration of dietary items between species, including crucial fallback foods (Dominy, Svenning & Li 2003), could also contribute to the differing visual systems of these species. Habitat appears to play an important role in shaping activity pattern as populations of the same species in different habitats are characterized by different types of cathemeral patterns; for example, E. flavifrons is reported to be more diurnal in primary forests and to occupy the understorey (Tattersall 1987; Curtis & Rasmussen 2002). Such conditions have long been posited to favour trichromatic vision in primates (Mollon 1989). In secondary forests, E. flavifrons is more nocturnal, likely due to the relationship between diurnal predation risk from large raptors and canopy foliage density (Curtis et al. 1999; Rasmussen 2005; Colquhoun 2006; Schwitzer et al. 2007). Importantly, environmental variation has been reported to lead to intraspecific and interspecific variation in cathemeral strategies in primates and other mammals as well as variation in opsin polymorphisms (Fernandez-Duque 2003; Tattersall 2006; Prugh & Golden 2014) (Table 1). We cannot therefore rule out the possibility that examining the opsins of rain forest-dwelling E. fulvus may reveal variation not found in the seasonal dry forest population studied here. Alternatively, Eulemur may be in a state of evolutionary disequilibrium (van Schaik & Kappeler 1996) and visual traits may not be adapted to current ecological conditions. To the extent that current adaptations can be used to infer past selective pressures, we suggest that further study of Eulemur has the potential to reveal critical variables surrounding the origins of polymorphic trichromacy and high-acuity vision, tandem traits that distinguish anthropoid primates (monkeys, apes and humans) from other mammals. Table 2. Percentage of fruits with chromatic and achromatic JND values >1, >075, >050 and >2 for trichromats and dichromats under three illuminant conditions. Per cent of JND values at each JND threshold for each of the three illuminants were compared using a z-score proportion of difference analysis. Z-scores and P-values of each comparison are provided Moonlight, Dichromat z-score P-value Moonlight, Trichromat Twilight, Dichromat z-score P-value Twilight, Trichromat Daylight, Dichromat z-score P-value Daylight, Trichromat Chromatic JND > 1 80 57 265 >0001 39 30 099 032 2 2 000 100 JND > 075 86 66 243 002 61 50 114 025 11 9 032 075 JND >050 93 73 277 001 80 63 209 004 25 23 022 083 JND >2 38 23 164 010 5 5 000 100 0 0 000 100 Achromatic JND>1 95 95 000 100 88 86 028 078 70 70 000 100 JND>075 95 95 000 100 93 89 066 051 75 77 022 083 JND>050 96 98 059 056 93 93 000 100 86 86 000 100 JND >2 93 91 035 073 80 80 000 100 39 38 019 040

940 K. Valenta et al. We conclude by drawing attention to what is perhaps the most surprising aspect of trichromatic vision in lemurs: the relatively poor visual acuity of E. flavifrons and other species (51 7 c deg 1 ), which is much lower than the acuities of anthropoid primates (30 75 c deg 1 ; Veilleux & Kirk 2009). If visual acuity is an important prerequisite of trichromacy, then a threshold of at least 5 7c deg 1 may exist; available evidence suggests that the acuity of bats and treeshrews lies within the range of 1 4 c deg 1 (Souza, Gomes & Silveira 2011). Interestingly, the ancestor of tarsiers has also been recently reconstructed as potentially trichromatic (Melin et al. 2013), and the inferred acuity of extant tarsiers is slightly above this purported threshold, at 89 c deg 1 (Veilleux & Kirk 2009). Nevertheless, a critical point to stress is that acuity at lemur-like levels is far below that of monkeys and apes. Polymorphic M/LWS opsin genes in E. flavifrons therefore demonstrate the potential for trichromacy without visual acuity at haplorhine-like levels an unexpected combination given the prevailing belief that high-acuity vision (> 30 cpd) was preadaptive to primate trichromacy (Mollon, Estevez & Cavonius 1990; Boycott & Wassle 1999; Dacey 2000). This insight further demonstrates the utility of Eulemur for understanding colour vision evolution. Continued research on the diets, activity patterns, molecular ecology and retinal morphologies of Eulemur taxa and other cathemeral lemurs is anticipated to shed further light on selective pressures shaping primate visual systems. Acknowledgements This work was supported by the Natural Sciences and Engineering Council of Canada [grant number CGS-D3-3741752009 to K.V., and PDF- 420810-2012 to A.D.M], Sigma Xi, GM Women in Science [K.V.], the University of Toronto [K.V. and M.E.] and the University of Calgary [S.E.J. and S.M.H.]. We thank MICET and Madagascar National Parks for permission to conduct this research in Madagascar. We are grateful to three anonymous reviewers whose comments improved a previous version of our manuscript. We thank Rachel Jacobs, Tara Paton and Dr. Brenda Bradley for helpful advice on DNA analysis. For sharing E. macaco sequence data, we thank Dr. Carrie Veilleux. We are grateful to Mr. Paul Tsiveraza, Jean de-dieu, Mamy Razafitsalama, Razafindravelo Cressant and Mbana Ferdinah for contributions in the field, to Dr. E. Ehmke and Dr. R. Schopler of the Duke Lemur Center, Felicia Spector, Monica Uhl and Jamie Dowling of the Lemur Conservation Foundation, and Dr. Grainne McCabe of the Bristol Zoo for information on tapeta. This research adhered to the Laws of Madagascar governing primate research, the American Society of Primatologists principles for the ethical treatment of primates, and the University of Toronto (Animal Care Protocol #20DD9283). Author contributions K.V., S.M.L., A.D.M., E.J.P. and N.J.D. designed the study; K.V., R.R.R., S.M.H. and S.E.J. collected the data; K.V., A.D.M., M.E., K.A.B. and E.J.P. analysed the data; K.V, A.D.M, E.J.P. and N.J.D. wrote the manuscript. All authors contributed to editing and revisions. Data accessibility Modelling formulas are available as Supporting Information. Raw fruit trait data sets, both chromatic and luminance values for fruit and distances between fruit and leaves are archived in the Dryad Digital Repository (http://dx/doi.org/10.5061/dryad.57vt3) (Valenta et al. 2015). E. fulvus sequence data are available in Fig. 1b. References Ankel-Simons, F. & Rasmussen, D.T. (2008) Diurnality, nocturnality, and the evolution of primate visual systems. Yearbook of Physical Anthropology, 47, 100 117. Andrews, J. & Birkinshaw, C. (1998) A comparison between the daytime and night-time diet, activity and feeding height of the black lemur, Eulemur macaco (Primates: Lemuridae), in Lokobe Forest, Madagascar. Folia Primatologica, 69(suppl 1), 175 182. Barton, R.A., Purvis, A. & Harvey, P.H. (1995) Evolutionary radiation of visual and olfactory brain systems in primates, bats and insectivores. Philosophical Transactions of the Royal Society of London, Series B: Biological Sciences, 348, 381 392. Boycott, B.B. & Wassle, H. (1999) Parallel processing in the mammalian retina: the Proctor lecture. Investigations in Opthalmology and Visual Science, 40, 1313 1327. Brown, L.D., Cat, T.T. & DasGupta, A. (2001) Interval estimation for a proportion. Statistical Science, 16, 101 133. Caine, N.G., Osorio, D. & Mundy, N.I. (2010) A foraging advantage for dichromatic marmosets (Callithrix geoffroyi) at low light intensity. Biology Letters, 6, 36 38. Carvalho, L.S., Davies, W.L., Robinson, P.R. & Hunt, D.M. (2012) Spectral tuning and evolution of primate short-wavelength-sensitive visual pigments. Proceedings of the Royal Society of London, B, 279, 387 393. Castenholz, E. (1965) Uber die struktur der netzhautmitte beit primaten. Zeitschrift fur Zellforschung und Mikroskopische Anatomie, 65, 646 661. Colquhoun, I. (1997) A predictive socioecological study of the black lemur (Eulemur macaco macaco) in Northwestern Madagascar. Ph.D. dissertation, Washington University, St. Louis, MO. Colquhoun, I.C. (2006) Predation and cathemerality: comparing the impact of predators on the activity patterns of lemurids and ceboids. Folia Primatologica, 77, 143 165. Curtis, D.J. & Rasmussen, M.A. (2002) Cathemerality in lemurs. Evolutionary Anthropology, S1, 83 86. Curtis, D.J. & Rasmussen, M.A. (2006) The evolution of cathemerality in primates and other mammals: a comparative and chronoecological approach. Folia Primatologica, 77, 178 193. Dacey, D.M. (2000) Parallel pathways for spectral coding in primate retina. Annual Review of Neuroscience, 23, 747 775. Dahl, J.F. & Hemingway, C.A. (1988) An unusual activity pattern for the mantled howler monkey of Belize. American Journal of Physical Anthropology, 752, 201. Dominy, N.J., Svenning, J.C. & Li, W.H. (2003) Historical contingency in the evolution of primate color vision. Journal of Human Evolution, 44, 25 45. Donati, G. & Borgognini-Tarli, S.M. (2006) Influence of abiotic factors on cathemeral activity: the case of Eulemur fulvus collaris in the littoral forest of Madagascar. Folia Primatologica, 77, 104 122. Donati, G., Bollen, A., Borgognini-Tarli, S.M. & Ganzhorn, J.U. (2007) Feeding over the 24-h cycle: dietary flexibility of cathemeral collared lemurs (Eulemur collaris). Behavioral Ecology and Sociobiology, 61, 1237 1251. Donati, G., Baldi, N., Morelli, V., Ganzhorn, J.U. & Borgognini-Tarli, S.M. (2009) Proximate and ultimate determinants of cathemeral activity in brown lemurs. Animal Behaviour, 77, 317 325. Donati, G., Santini, L., Razafindramanana, J., Boitani, L. & Borgognini- Tarli, S.M. (2013) (Un-)expected nocturnal activity in the diurnal Lemur catta supports cathemerality as one of the key adaptations of the lemurid radiation. American Journal of Physical Anthropology, 150, 99 106. Endler, J.A. (1993) Some general comments on the evolution and design of animal communication systems. Philosophical Transactions of the Royal Society B: Biological Sciences, 340, 215 225. Eppley, T.M., Ganzhorn, J.U. & Donati, G. (2015) Cathemerality in a small, folivorous primate: proximate control of diel activity in Hapalemur meridionalis. Behavioral Ecology and Sociobiology, 69, 991 1002. Fernandez-Duque, E. (2003) Influences of moonlight, ambient temperature, and food availability on the diurnal and nocturnal activity of owl monkeys (Aotus azarai). Behavioral Ecology and Sociobiology, 54, 431 440.

Opsin variation and ecology of cathemeral lemurs 941 Geissmann, T. (2002) Vergleichende Primatologie. Springer-Verlag, Berlin. Griffin, R.H., Matthews, L.J. & Nunn, C.L. (2012) Evolutionary disequilibrium and activity period in primates: a Bayesian phylogenetic approach. American Journal of Physical Anthropology, 147, 409 416. Hall, M.I., Kamilar, J.M. & Kirk, E.C. (2012) Eye shape and the nocturnal bottleneck of mammals. Proceedings of the Royal Society B, 279, 4962 4968. Heesy, C.P. & Ross, C.F. (2001) Evolution of activity patterns and chromatic vision in primates: morphometrics, genetics and cladistics. Journal of Human Evolution, 40, 111 149. Higham, J.P., Brent, L.J., Dubuc, C., Accamando, A.K., Engelhardt, A., Gerald, M.S. et al. (2010) Color signal information content and the eye of the beholder: a case study in the rhesus macaque. Behavioral Ecology, 21, 739 746. Horvath, E., Weisrock, D.W., Embry, S.L., Fiorentino, I., Balhoff, J.P., Kappeler, P.M. et al. (2008) Development and application of a phylogenomic toolkit: resolving the evolutionary history of Madagascar s lemurs. Genome Research, 18, 489 499. Jacobs, G.H. (2008) Primate color vision: a comparative perspective. Visual Neuroscience, 25, 619 633. Jacobs, G.H. & Deegan, J.F. (1993) Photopigments underlying color vision in ringtail lemurs (Lemur catta) and brown lemurs (Eulemur fulvus). American Journal of Primatology, 30, 243 256. Jacobs, G.H., Deegan, J.F., Neitz, J. & Crognale, M. (1993) Photopigments and color vision in the nocturnal monkey, Aotus. Vision Research, 33, 1773 1783. Jacobs, G.H., Neitz, M., Deegan, J.F. & Neitz, J. (1996) Trichromatic colour vision in New World monkeys. Nature, 382, 156 158. Jacobs, G.H., Deegan, J.F., Tan, Y. & Li, W. (2002) Opsin gene and photopigment polymorphism in a prosimian primate. Vision Research, 42, 11 18. Jacobs, G.H. & Deegan, J.F. (2003) Diurnality and cone photopigment polymorphism in strepsirrhines: examination of linkage in Lemur catta. American Journal of Physical Anthropology, 122, 66 72. Kamilar, J.M., Heesy, C.P. & Bradley, B.J. (2013) Did trichromatic color vision and red hair color coevolve in primates? American Journal of Primatology, 75, 740 751. Kawamura, S. & Kubotera, N. (2004) Ancestral loss of short wavelength sensitive cone visual pigment in lorisiform prosimians, contrasting with its strict conservation in other prosimians. Journal of Molecular Evolution, 58, 314 321. Kelber, A. & Lind, O. (2010) Limits of colour vision in dim light. Ophthalmic and Physiological Optics, 30, 454 459. Kirk, E.C. (2004) Comparative morphology of the eye in primates. The Anatomical Record, Part A, 281, 1095 1103. Kirk, E.C. (2006) Eye morphology in cathemeral lemurids and other mammals. Folia Primatologica, 77, 27 49. LaFleur, M., Sauther, M., Cuozzo, F., Yamashita, N., Youssouf, I.A.J. & Bender, R. (2014) Cathemerality in wild ring-tailed lemurs (Lemur catta) in the spiny forest of Tsimanampetsotsa National Park: camera trap data and preliminary behavioral observations. Primates, 55, 207 217. Matsumoto, Y., Hiramatsu, C., Matsushita, Y., Ozawa, N., Ashino, R., Nakata, M. et al. (2014) Evolutionary renovation of L/M opsin polymorphism confers a fruit discrimination advantage to ateline New World monkeys. Molecular Ecology, 7, 1799 1812. Melin, A.D., Fedigan, L.M., Hiramatsu, C., Sendall, C.L. & Kawamura, S. (2007) Effects of colour vision phenotype on insect capture by a freeranging population of white-faced capuchins, Cebus capucinus. Animal Behaviour, 73, 205 214. Melin, A.D., Fedigan, L.M., Hiramatsu, C., Hiwatashi, T., Parr, N. & Kawamura, S. (2009) Fig foraging by dichromatic and trichromatic Cebus capucinus in a tropical dry forest. International Journal of Primatology, 30, 753 775. Melin, A.D., Fedigan, L.M., Young, H.C. & Kawamura, S. (2010) Can color vision variation explain sex differences in invertebrate foraging by capuchin monkeys? Current Zoology, 56, 300 312. Melin, A.D., Moritz, G.L., Fosbury, R.A., Kawamura, S. & Dominy, N.J. (2012) Why aye-ayes see blue. American Journal of Primatology, 74, 185 192. Melin, A.D., Matsushita, Y., Moritz, G.L., Dominy, N.J. & Kawamura, S. (2013) Inferred M/L cone opsin polymorphism of ancestral tarsiers sheds dim light on the origin of anthropoid primates. Proceedings of the Royal Society of London, B, 208, 1 7. Mollon, J.D. (1989) Tho she kneel d in that place where they grew. The uses and origins of primate colour vision. Journal of Experimental Biology, 146, 21 38. Mollon, J.D., Bowmaker, J.K. & Jacobs, G.H. (1984) Variations of colour vision in a New World primate can be explained by polymorphism of retinal photopigments. Proceedings of the Royal Society of London, Series B: Biological Sciences, 222, 373 399. Mollon, J.D., Estevez, O. & Cavonius, C.R. (1990) The two subsystems of colour vision and their roles in wavelength discrimination. Vision: Coding and Efficiency (ed. C. Blakemore), pp. 119 131. Cambridge University Press, Cambridge. Moore, B.A., Kamilar, J.M., Collin, S.P., Bininda-Emonds, O.R.P., Dominy, N.J., Hall, M.I. et al. (2012) A novel method for comparative analysis of retinal specialization traits from topographic maps. Journal of Vision, 12, 1 24. Morgan, M.J., Adam, A. & Mollon, J.D. (1992) Dichromats detect colour-camouflaged objects that are not detected by trichromats. Proceedings of the Royal Society of London, Series B: Biological Sciences, 284, 291 295. Moritz, G.L., Melin, A.D., Tuh Yit Yu, F., Bernard, H., Ong, P.S. & Dominy, N.J. (2014) Niche convergence suggests functionality of the nocturnal fovea. Frontiers in Integrative Neuroscience, 8. Noback, C.R. (1975) The visual system of primates in phylogenetic studies. Phylogeny of the Primates (eds W.P. Luckett & F.S. Szalay), pp. 199 218. Plenum Press, New York. Nsubuga, A.M., Robbins, M.M., Roeder, A.D., Morin, P.A., Boesch, C. & Vigilant, L. (2004) Factors affecting the amount of genomic DNA extracted from ape faeces and the identification of an improved sample storage method. Molecular Ecology, 13, 2089 2094. Osorio, D. & Vorobyev, M. (1996) Colour vision as an adaptation to frugivory in primates. Proceedings of the Royal Society of London, Series B: Biological Sciences, 263, 593 599. Osorio, D., Smith, A.C., Vorobyev, M. & Buchanan-Smith, H.M. (2004) Detection of fruit and the selection of primate visual pigments for color vision. American Naturalist, 164, 696 708. Pariente, G.F. (1970) Retinographie comparee des lemuriens malgasche. Comptes Rendus de l Academie des Sciences, 270, 1404 1407. Pariente, G.F. (1976) Les differents aspects de la limite du tapetum lucidum chez les prosimiens. Vision Research, 16, 387 391. Pariente, G. (1979) The role of vision in prosimian behavior. The Study of Prosimian Behavior (eds G.A. Doyle & R.D. Martin), pp. 411 459. Academic Press, New York. Peichl, L., Rakotondraparany, F. & Kappeler, P.M. (2001) Photoreceptor types and distributions in nocturnal and diurnal Malagasy primates. Investigations in Opthalmology and Visual Science, 42, 270. Perry, G.H., Martin, R.D. & Verrelli, B.C. (2007) Signatures of functional constraint at aye-aye opsin genes: the potential of adaptive color vision in a nocturnal primate. Molecular Biology and Evolution, 24, 1963 1970. Prugh, L.R. & Golden, C.D. (2014) Does moonlight increase predation risk? Meta-analysis reveals divergent responses of nocturnal mammals to lunar cycles. Journal of Animal Ecology, 83, 504 514. Rasmussen, M.A. (2005) Seasonality and predation risk: varying activity periods in lemurs and other primates. Seasonality in Primates: Implications for Human Evolution (eds DK Brockman & CP van Schaik), pp. 105 128. Cambridge University Press, Cambridge. R Development Core Team (2014) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. Available: http://www.r-project.org/ Rumpler, Y., Hauwy, M., Fausser, J., Roos, C., Zaramody, A., Andriaholinirina, N. et al. (2011) Comparing chromosomal and mitochondrial phylogenies of the Indriidae (Primates, Lemuriformes). Chromosome Research, 19, 209 224. Saito, A., Mikami, A., Kawamura, S., Ueno, Y., Hiramatsu, C., Widayati, K.A. et al. (2005) Advantage of dichromats over trichromats in discrimination of color-camouflaged stimuli in nonhuman primates. American Journal of Primatology, 67, 425 436. Santini, L., Rojas, D. & Donati, G. (2015) Evolving through day and night: origin and diversification of activity pattern in modern primates. Behavioral Ecology, 26, 789 796. van Schaik, C.P. & Kappeler, P.M. (1996) The social systems of gregarious lemurs: Lack of convergence with anthropoids due to evolutionary disequilibrium? Ethology, 102, 915 941. Schwitzer, N., Kaumanns, W. & Zahner, H.S.C. (2005) Cathemerality in blue-eyed black lemurs (Eulemur macaco flavifrons) on the Sahamalaza Peninsula, north-west Madagascar. Primate Reports, 72, 88 89. Schwitzer, N., Kaumanns, W., Seitz, P.C. & Schwitzer, C. (2007) Cathemeral activity patterns of the blue-eyed black lemur Eulemur macaco