Diurnality and cone photopigment polymorphism in strepsirrhines Examination of linkage in Lemur catta.код для вставкиСкачать
AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 122:66 –72 (2003) Diurnality and Cone Photopigment Polymorphism in Strepsirrhines: Examination of Linkage in Lemur catta Gerald H. Jacobs1* and Jess F. Deegan II2 1 Neuroscience Research Institute and Department of Psychology, University of California, Santa Barbara, California 93106 2 Department of Psychology, California State University, Bakersﬁeld, California 93311 KEY WORDS lemurs; cones; electroretinogram; opsins; primate evolution ABSTRACT Trichromatic color vision is routine among catarrhine primates, but occurs only as a variant form of color vision in some individuals in most platyrrhine genera. This arises from a fundamental difference in the organization of X-chromosome cone opsin genes in these two lineages: catarrhines have two opsin genes specifying middle- and long-wavelength-sensitive cone pigments, while platyrrhines have only a single gene. Some female platyrrhine monkeys achieve trichromacy because of a species polymorphism that allows the possibility of different opsin gene alleles on the two X-chromosomes. Recently, a similar opsin gene polymorphism was detected in some diurnal strepsirrhines, while at the same time appearing to be absent in any nocturnal genera. The aim of this study was to assess whether cone pigment polymorphism is inevitably linked to diurnality in strepsirrhines. Cone photopigments were measured in a species usually classiﬁed as diurnal, the ring-tailed lemur (Lemur catta), using electroretinogram ﬂicker photometry, a noninvasive electrophysiological procedure. Each of 12 animals studied was found to have the same middle-wavelength cone pigment, with peak sensitivity at about 547 nm. In conjunction with earlier results, this implies that cone pigment polymorphism is unlikely to exist in this species and that, accordingly, such variation is not a consistently predictable feature of vision in diurnal strepsirrhines. Am J Phys Anthropol 122:66 –72, 2003. © 2003 Wiley-Liss, Inc. Many extant strepsirrhines are nocturnal. In accordance with this lifestyle, the retinas of these animals are heavily rod-dominated and lack foveas (Rohen and Castenholtz, 1967), and their eyes also typically contain a reﬂective tapetum (Wolin and Massopust, 1970; Pariente, 1979). All these features are characteristic of the adaptations for nocturnal vision classically observed in many different mammals (Walls, 1942). A number of species from two strepsirrhine families (Lemuridae and Indridae) depart from this pattern in having activity rhythms that are described as either diurnal or crepuscular, or in some cases, as cathemeral, i.e., including both nocturnal and diurnal characteristics (Tattersall, 1987; van Schaik and Kappeler, 1996; Rowe, 1996; Curtis and Rasmussen, 2002). Despite striking differences in their circadian behaviors, the eyes of these latter species often (although not inevitably) share a number of adaptational features found in the eyes of nocturnal strepsirrhines including, for example, tapeta and afoveate retinas. A conventional interpretation of these facts is that primitive primates were nocturnal and that progression to the more diurnal lifestyles has occurred secondarily in a minority of strepsirrhine lines (Martin, 1990). Studies of mammalian cone photopigments and color vision conducted over the past 20 years seem generally consistent with this view. The retinas of most mammals either contain two classes of cone that set the stage for dichromatic color vision, or alternatively, in some species, a gene mutation has reduced the cone pigment complement from two to one, leaving such animals without color vision capacity (Jacobs, 1993, 1996). Among all eutherian mammals, only primates have the three classes of cone photopigment needed to support trichromatic color vision, and even within the order, there is considerable variation in the details of biology that allow trichromacy. Speciﬁcally, catarrhine monkeys, apes, and humans were found to have multiple Xchromosome opsin genes that specify spectrally distinct middle-wavelength-sensitive (M) and longwavelength-sensitive (L) cone pigments. In conjunction with an autosomal gene that expresses short-wavelength-sensitive (S) cone pigment, this arrangement provides the photopigment basis for © 2003 WILEY-LISS, INC. Grant sponsor: National Eye Institute; Grant number: EY02052. *Correspondence to: Gerald H. Jacobs, Neuroscience Research Institute, University of California, Santa Barbara, CA 93106. E-mail: firstname.lastname@example.org Received 17 December 2002; accepted 5 March 2003. DOI 10.1002/ajpa.10309 STREPSIRRHINE DIURNALITY/PHOTOPIGMENT POLYMORPHISM trichromatic color vision (Nathans et al., 1986). In striking contrast, most platyrrhine monkeys are polymorphic at a single X-chromosome opsin gene site. This polymorphism allows females heterozygous at this gene site to produce two types of M/L pigments and, with an S-cone photopigment that all monkeys share in common, to achieve trichromatic color vision (Jacobs, 1998a). Note that the nature of this arrangement renders homozygous females and all males as obligatory dichromats. Beyond this general pattern, across-species variations have also been found, both in the spectral absorption properties of the M/L photopigments and in the number of M/L gene alleles. Finally, members of two genera (Aotus and Alouatta) have distinctly different arrangements from those of any other New World monkeys. Aotus has only a single functional cone pigment and thus lacks color vision capacity (Jacobs et al., 1993), while members of Alouatta are like the catarrhines in routinely having three types of cone pigment and apparent trichromatic color vision (Jacobs et al., 1996b). The result of all this variation is rich mixture of color vision phenotypes distributed across the platyrrhines. The photopigments and color vision of strepsirrhines have not received nearly as much attention as that accorded the haplorrhines. Results from two studies that assessed the cone photopigment complements in brown lemurs (Eulemur fulvus), ringtailed lemurs (Lemur catta) (Jacobs and Deegan, 1993), and greater bushbabies (Otolemur crassicaudatus) (Deegan and Jacobs, 1996) were taken to suggest that color vision in strepsirrhines in general might be most similar to that of nonprimate mammals. This conclusion was based on the ﬁndings that each of these three species apparently had only a single X-chromosome opsin gene site, while one of them (the bushbaby) had additionally suffered loss of function of its S-cone opsin gene as a result of mutational changes (Jacobs et al., 1996c). Given this understanding, it was a surprise when a recent study involving a large-scale screening of M/L strepsirrhine opsin genes revealed the presence of polymorphic variation in 3 of 20 species examined: Coquerel’s sifaka (Propithecus verreauxi coquereli), the red-ruffed lemur (Varecia variegata rubra), and the greater dwarf lemur (Cheirogaleus major) (Tan and Li, 1999). In each of these three species, two M/L opsin gene alleles were detected, and further examination of the inferred amino-acid sequences of the allelic versions of the opsin gene led to the prediction that these speciﬁed two spectrally distinct M/L pigments. Subsequent analysis indicated that the claim for polymorphism in the greater dwarf lemur was probably based on a sample misidentiﬁcation and is therefore not correct (Y. Tan, personal communication). Measurements of photopigments documented the variable presence of M/L cone photopigments in the retinas of Coquerel’s sifaka and the black-andwhite-ruffed lemur (Varecia variegata variegata) (Jacobs et al., 2002; Jacobs and Deegan, 2003b). In 67 each species, individual females were found that had both classes of M/L pigment present in their retinas. By analogy to the results from studies of platyrrhine monkeys, this arrangement predicts the presence of trichromatic color vision in some individual strepsirrhines, in addition to variant forms of dichromatic color vision. Based on their observations of cone opsin polymorphism in some strepsirrhines and on the phylogenetic distribution of the M and L opsin gene alleles in this group, Tan and Li (1999) were led to suggest that, counter to the traditional view, the common ancestor of the strepsirrhines may have been trichromatic. Since vertebrate trichromacy seems inherently incompatible with nocturnality, this implied that these ancestral primates were also diurnal. This issue was recently analyzed by Heesy and Ross (2001), who reconstructed the activity patterns in fossil primates through an examination of the evolution of a number of features of visual systems. Their results reinforced the traditional view in implying that both nocturnality and dichromacy are primitive traits for primates, and that shifts from nocturnality have occurred only sporadically among the strepsirrhines. This conclusion is also supported by the results of a comparative analysis of orbit size in fossil and extant primates (Kay and Kirk, 2000). The three strepsirrhine species in which the biological basis for trichromatic color vision has so far been shown to exist are all diurnal. That linkage raises the question as to whether the acquisition of an M/L opsin gene polymorphisms is a characteristic of all diurnal strepsirrhines, or whether perhaps the coexistence of diurnality and M/L polymorphism is happenstance. A test case on this matter is offered by the ring-tailed lemur, an animal judged diurnal in most accounts (Sussman, 1999; Macdonald, 2001). The earlier pigment measurements made on four members of this species revealed only a single type of M/L photopigment (Jacobs and Deegan, 1993), and a later genetic examination of a single member of this species supports that view (Tan and Li, 1999). Although these results are consistent with the conclusion that ring-tailed lemurs lack an M/L cone pigment polymorphism, the number of animals so far studied is too few to yield much conﬁdence in that conclusion. To provide further insight into this issue, we used an electrophysiological technique, electroretinogram (ERG) ﬂicker photometry, to study the cone photopigment complements of additional animals. Our results add support to the contention that this strepsirrhine does not have an M/L cone photopigment polymorphism. METHODS ERG ﬂicker photometry is a noninvasive procedure in which electrical signals evoked in the outer retina by the presentation of stimulus lights are sensed from a contact-lens electrode placed on the corneal surface of the eye. We have frequently employed this technique to make spectral sensitivity 68 G.H. JACOBS AND J.F. DEEGAN measurements and so draw inferences about photopigments in a range of different species. The physical features of the recording setup and details of this procedure were fully described elsewhere (Jacobs et al., 1996a), so only a brief account is provided here. In this procedure, a train of temporally alternating ﬂashes of light derived from two beams of an optical system are imaged so that they fall in register onto the retina in Maxwellian view in the form of a 59° circle. One of these lights originates from a 50-W tungsten-halide lamp and serves as a reference stimulus; the other, the test stimulus, comes from a high-intensity grating monochromator (10-nm half-bandwidth). A third beam of the optical system, also originating from a tungsten-halide lamp, is used as required to provide accessory adaptation. The ERGs generated by the test and reference lights are electronically compared and, over presentations, the test light is adjusted in radiance until the ERG it elicits is the same as that produced by the reference light. Repeat equations made using this procedure seldom vary by an amount greater than 0.04 log units (Jacobs et al., 1996a). The intensity relationship between the two lights at the point of equation deﬁnes one point on a spectral sensitivity function. Repetition of this procedure for a range of different test wavelengths then permits the derivation of a complete spectral sensitivity function. The ERGs are differentially recorded, and the ampliﬁed signals are windowed by a sinusoid that is set to the frequency of the stimulus train. For each experiment, the window position is initially adjusted so as to maximize its correlation with the ERG signal. The comparison of the responses to test and reference lights is based on the averages of the last 50 of 70 responses elicited from each source. Two experiments were run on each animal. First, complete spectral sensitivity functions were derived. For this procedure, the stimulus rate was set to 31.25 Hz. The reference light was achromatic (2,450 K) and provided a retinal illuminance of 3.3 log td. Photometric equations were made between the reference light and monochromatic test lights that were successively varied in 10-nm steps from 450 to 640 nm. Two complete runs were made across this spectral span, and the resulting pair of equations was averaged. For any individual test wavelength where the two equation values differed by more than the 0.04 log unit, a third measurement was made, and these results also contributed to the ﬁnal mean value. A second experiment provided an explicit test of whether the eye contained more than one type of M/L cone pigment. It followed a procedure ﬁrst used and validated on platyrrhine monkeys (Jacobs and Neitz, 1987). For this test, ERG ﬂicker photometric equations were made between a 540-nm test light and a 630-nm reference light (pulse rate ⫽ 31.25 Hz). These equations were obtained during a period while the eye was being alternately adapted either to bright 540-nm or 630-nm light. As for the spectral sensitivity measurements, the photometric equations were made twice, and the resulting values were then averaged. The adaptation lights were ﬁrst adjusted in intensity so that each produced the same threshold elevation (0.5 log units) for a 560-nm test light presented at 31.25 Hz. In this experiment, the spectral properties of the reference and adaptation lights were controlled by placing different interference ﬁlters (half-energy passband ⫽ 10 nm) into the stimulus beams. Finally, a spectral sensitivity function was obtained for one animal under scotopic test conditions. To accomplish this, the animal was initially dark-adapted for 15 min. Subsequently, spectral sensitivity measurements were carried out in a darkened room, with the stimulus pulse rate reduced to 12.5 Hz and with the reference light dimmed to yield a retinal illuminance of 1.5 log td. In the fashion described above, wavelengths from 440 – 600 nm were then tested at 10-nm steps. Recordings were obtained from 12 (10 male, 2 female) adult (age range, 12–20 years) ring-tailed lemurs (Lemur catta). According to the American Zoo and Aquarian Association (AZA) Studbook records, two of these animals were siblings, but none of the other subjects had a shared parentage. Recordings were made during the course of routine physical examinations conducted on these animals at the Santa Ana Zoo (Santa Ana, CA). Each animal was initially anesthetized with isoﬂuorane (5%) delivered through a mask. Subsequently, the animals were intubated and maintained on 1.5–2% isoﬂuorane for the duration of the recording session. Heart rate, expired CO2, and body temperature were monitored throughout the experiments. The cornea of one eye was anesthetized by topical application of proparacaine hydrochloride (0.5%), and the pupil was dilated with mydriacyl (0.5% tropicamide). The ERG electrode was installed and the animal was positioned for recording with the aid of a specially designed head holder that had no pressure points. The recordings were conducted in a lighted room that provided about 150 lux of ambient illumination at the position of the test eye. All of the experimental procedures were conducted in accordance with the National Institutes of Health guidelines on the care and use of animals. RESULTS The ﬂicker rate used in these experiments exceeded the 30-Hz photopic standard deﬁned for human clinical ERG recording (Marmor and Zrenner, 1999), thus insuring that the signals recorded reﬂected cone activity. With these test conditions, highly reliable signals were recorded from each of the animals studied. The sensitivity functions obtained from members of this sample were remarkably consistent in their spectral shapes, as can be seen in Figure 1, where we plotted mean spectral sensitivity values (⫾1 SD) obtained from 12 ringtailed lemurs. That there are no obvious differences among these animals can be inferred from the very STREPSIRRHINE DIURNALITY/PHOTOPIGMENT POLYMORPHISM 69 Fig. 2. Results from a test of middle- to long-wavelength response univariance in two species of strepsirrhine. Plotted are differences (given in log units) in ERG photometric equations obtained for a 540-nm test light and a 630-nm reference light when eye was steadily exposed to either 540-nm or 630-nm adapting lights. Individual data points were for tests earlier run on Coquerel’s sifakas (Jacobs et al., 2002). Shown are results for animals having either one (triangles) or two (circles) types of M/L photopigments. Equation values obtained for 12 ring-tailed lemurs are shown above, where horizontal line encompasses the entire range of variation in equations measured for this group. Fig. 1. Spectral sensitivity function for ring-tailed lemur. Data points represent mean sensitivity values for 12 animals, as derived from ERG ﬂicker photometric measurements. Error bars show ⫾ 1 SD. Continuous line is best-ﬁt photopigment absorption curve (max ⫽ 547.9 nm). small variations in the sensitivity values recorded across the spectrum: for a total of 20 test points, the mean SD value was 0.036 log unit. The continuous line drawn through the data points represents the best-ﬁtting single photopigment absorption function. To derive this ﬁt, a visual pigment template (Govardovskii et al., 2000) was progressively shifted along the wavelength axis in steps of 0.1 nm to determine the spectral location providing the best least-squares ﬁt to the entire data array. The values of Figure 1 were not corrected for any possible intraocular ﬁltering, and they thus reﬂect corneally derived spectral sensitivity. With that restriction, the best-ﬁt template derived for this grouped data has a spectral peak value (max) of 547.9 nm. An additional indication of the small dispersion of spectral data across subjects is that individual spectral ﬁts made for the 12 subjects yielded a mean max of 546.2 nm (SD ⫽ 1.26). The second experiment was a test for spectral response univariance, and it provided an explicit means for assessing whether the retinas of these ring-tailed lemurs contained more then a single type of M/L photopigment. The assumption underlying this test is that if there is only a single M/L pigment, then the photometric match made between 540-nm and 630-nm lights will remain invariant, irrespective of the nature of any chromatic adaptation. On the other hand, if more than one type of M/L pigment is present, then the equation will change as a function of the character of the chromatic adaptation, so that relatively more 540-nm light will be required when the eye is exposed to 540-nm adap- tation, and relatively less 540-nm light will be required when the adaptation is shifted to 630 nm. Figure 2 illustrates results obtained from such tests. Plotted for context are results of the identical test obtained from an earlier experiment run on Propithecus verrauxi coquereli, a strepsirrhine that, as noted above, has an M/L cone polymorphism (Jacobs et al., 2002). Shown are the differences (expressed in log units) in the 540/630-nm equations obtained for the two adaptation conditions. The retinas of 7 of these animals (triangles in Fig. 2) contained only a single type of M/L pigment having a peak sensitivity at either 545 nm or 558 nm. Note that there is no signiﬁcant change in the equation value recorded for the two test conditions. The retinas of two other female members of this species contained both of these types of cone pigment and, as can be seen, they (solid circles in Fig. 2) showed clear shifts in the equation values obtained for the two chromatic adaptation conditions. As indicated by the full range of equation values plotted in Figure 2, none of the ring-tailed lemurs showed any hint of signiﬁcant differential adaptation, and so we conclude that none of these animals had more than a single type of M/L cone pigment. In the earlier study of photopigments in lemurs, a spectral sensitivity function was obtained from one subject under test conditions expected to index rod activity (Jacobs and Deegan, 1993). The resulting function had a spectral peak slightly longer than might seem typical for a mammalian rhodopsin. To see if that result could be replicated, we measured a spectral sensitivity function for one of the subjects of this experiment under similar scotopic test conditions. The resulting spectral sensitivity function is given in Figure 3 where, as in the fashion described above, the data array (solid circles) was best ﬁt to a photopigment absorption template. The max for the best ﬁt curve of Figure 3 is 509.5 nm. 70 G.H. JACOBS AND J.F. DEEGAN Fig. 3. Scotopic spectral sensitivity function obtained from ring-tailed lemurs, using ERG ﬂicker photometric measurements. Circles represent values obtained from an animal examined in current experiments; those for another animal (triangles) are from earlier published measurements (Jacobs and Deegan, 1993). Fitted photopigment absorption curve has a peak value of 509.5 nm. DISCUSSION Each of the 12 ring-tailed lemurs examined was found to have only a single type of M/L cone photopigment, and analysis of the measured spectra indicates (Fig. 1) that pigment was shared in common by each animal. These results are thus indistinguishable from what was found in 4 animals previously studied with ERG ﬂicker photometry (Jacobs and Deegan, 1993) and in a single subject whose M/L cone opsin genes were recently examined (Tan and Li, 1999). Of the total of 17 animals that have been examined in the current study and the two previous ones, 3 were female. Since these opsin genes are X-chromosome-linked, this means that the 20 Xchromosomes sampled from ring-tailed lemurs were all found to have the same M/L opsin gene. To examine the implications of this result, two-sided conﬁdence intervals were computed based on this single proportion (Newcombe, 1998). The results of this test indicate that the true proportion of ring-tailed lemur X-chromosomes harboring this same gene lies somewhere in the range between 83.9 –100% (P ⫽ 0.05). Our survey thus suggests that if a second M/L pigment exists in ring-tailed lemurs, it must be present only at quite a low proportion (⬍16%). Obviously, the study of a very much larger sample of animals would be required to achieve reasonable certainty that Lemur catta formally lacks an M/L cone pigment polymorphism. In the absence of such an expanded sample, it is nevertheless useful to ask how likely it is that a polymorphism exists at a frequency as low as or lower than that implied by the above analysis. One line of evidence that could bear on the matter comes from studies of M/L cone polymorphisms conducted in other nonhuman primates. Most of these studies necessarily involved only relatively small numbers of subjects, but in cases where the samples were more sizable, there are reasons to believe that the polymorphic genes in these species are represented at approximately equal frequency. The evidence for this conclusion is drawn most directly from the proportion of heterozygous females detected in these polymorphic species. For example, the polymorphism of squirrel monkeys (Saimiri) features three versions of the M/L pigment, and around two thirds of all females are found to be heterozygous (Jacobs, 1998b), while in spider monkeys (Ateles), where there appear to be only two M/L alleles, roughly 50% of the females are heterozygous (Jacobs and Deegan, 2001). In both of these cases, the incidence of female heterozygosity (and thus predicted trichromacy) is close to what would be expected if the allelic versions of the M/L genes are equally frequent in the population. One explanation for this correlation is that the polymorphism in these species is maintained by heterozygous advantage, thus ensuring relatively equal representation of the polymorphic genes. Making the reasonable assumption that there may be similar advantages for presence of trichromatic color vision in other primates would in turn lead to the expectation that in strepsirrhines there should also be nearequal representation of any polymorphic M/L opsin genes. Since the strepsirrhines known to be polymorphic have only two versions of the M/L opsin gene (Tan and Li, 1999), this suggests that these two versions should be about equally frequent in the population. Given the number of animals tested to date (see above), a calculation based on the binomial sampling distribution shows the probability of there being two equally frequent M/L alleles in L. catta to be vanishingly low (9.5 ⫻ 10⫺7). Accordingly, we suggest that ring-tailed lemurs probably do not have some (as yet undetected) low-frequency M/L cone polymorphism; rather, they probably lack such variation entirely. The current results can also provide further insight into strepsirrhine photopigments. Our earlier measurements of the spectral sensitivity of the M cone of the ring-tailed lemur yielded an estimated peak value of 546 nm (Jacobs and Deegan, 1993). That location is extremely close to the one derived from the current measurements (from Fig. 1, a value of 547.9 nm) and, indeed, most of the residual difference between the results of the two experiments reﬂects the fact that the photopigment template in current use routinely predicts peak values slightly longer than the metric we used a decade ago (Jacobs and Deegan, 2003a). Although very consistent across animals, these values do not necessarily give an accurate account of the photopigment absorption spectra. To properly infer those estimates requires STREPSIRRHINE DIURNALITY/PHOTOPIGMENT POLYMORPHISM knowledge about any spectral ﬁltering within the eye and of the pigment density in the cones of ringtailed lemurs. ERG spectral sensitivity measurements made at shorter wavelengths than those here reported (unpublished) provided an indication that, as in the eyes of many other diurnal animals (Douglas and Marshall, 1999), the lens of the ring-tailed lemur provides some preferential absorption of short-wavelength light. This could be particularly relevant for the present sample, where many of the subjects were quite old. One strategy that can be used to minimize such inﬂuence is to truncate the spectral sensitivity function so that it is based only on those values for test locations of 520 nm and longer, a portion of the spectrum that is relatively immune to inﬂuence from any primate preretinal ﬁlters (Sharpe et al., 1998). Following that procedure, and making the further assumption that the cones of ring-tailed lemurs might have a pigment optical density of 0.15, we reﬁtted the averaged data obtained from the 12 subjects. The result is an estimated M-cone peak value of 545.8 nm. Spectral sensitivity measured under scotopic test conditions for a single ring-tailed lemur yielded a function (Fig. 3) quite similar to the one previously obtained for this species (Jacobs and Deegan, 1993). A template best ﬁt to the averaged data from both animals has a peak value of 509.4 nm. This scotopic curve is thus shifted somewhat toward long wavelengths, relative to what might be expected for typical rod-based primate vision. Adding an assumption about cone pigment density will shift the peak of this function somewhat toward the shorter wavelengths (e.g., using a pigment density value of 0.3 moves the peak to 507 nm). Even this is long-wavelength-shifted relative to the peak value of 501 nm reported for direct microspectrophotometric measurements made on rods in the retina of another lemur, the black lemur (Eulemur macaco; Bowmaker, 1991). One possible explanation for this residual deviation is that tapetal reﬂection may have a relatively greater impact on scotopic spectral sensitivity than it does on measurements made under photopic conditions. Following effectively identical procedures, we measured M-cone spectral sensitivity in representatives of three genera of diurnal strepsirrhines (Lemur, Propithecus, and Varecia). If one uses the analysis assumptions described above, the spectral properties of M-cone photopigments of these three are quite similar, with derived M-cone peaks as follows: Lemur, 545.8 nm, n ⫽ 12; Propithecus, 545.3 nm, n ⫽ 3; and Varecia, 547.4 nm, n ⫽ 1. None of these measurements take into account any potential inﬂuence that tapetal reﬂection may have on spectral sensitivity measured in the intact eye. We suggest that any such effects are likely to be slight for two reasons: (1) none of the spectral sensitivity functions show obvious departures from the pigment templates, as one might expect if tapetal reﬂection were signiﬁcantly inﬂuencing spectral sensitivity; 71 and (2) there are striking differences in the prominence of the tapetal reﬂex among these species (for instance, on gross examination, tapetal reﬂection is much more obvious in Propithecus than in Varecia; indeed, Wolin and Massopust (1970) suggest that Varecia does not have a tapetum), but despite that, the shapes of the M-cone spectra are very similar for the three genera. Within measurement error, it seems likely that these strepsirrhines share in common their M-cone pigments. The available evidence suggests that M/L cone pigment polymorphism (and thus potential withinspecies variations in color vision) is infrequent among the strepsirrhines. The only genera in which polymorphism has been clearly documented are diurnal. But while diurnality seems likely to be a necessary condition for M/L pigment polymorphism to emerge as an adaptive trait, the survey results from Lemur catta suggest that diurnality, by itself, is not sufﬁcient to predict M/L pigment polymorphism. The absence of a compelling linkage between diurnality and M/L cone pigment polymorphism could reﬂect nothing other than the stochastic nature of the alterations in opsin genes required to shift the absorption spectra of the cone pigments they specify, and thus initiate a polymorphic condition. In that view, Lemur catta may just be unlucky. On the other hand, it may be worth remembering that simply describing different animals as diurnal does not mean that they have identical visual adaptations. For instance, although we are not aware of any modern examinations of photoreceptor distribution and density in the ring-tailed lemur, cone density in the central retina of Propithecus is twice that seen in an equivalent region in the retina of the brown lemur (Eulemur fulvus), and there are corresponding differences in overall cone/rod ratios of these animals (Peichl et al., 2001). This difference is potentially important, since an increased cone density seems likely to enhance the likelihood that signals from two different cone types get processed in the neural networks of the retina in a fashion needed to support color vision (Dacey, 1999), and so one could argue that the retina of Propithecus provides a more favorable substrate for the emergence of cone pigment polymorphism as an adaptive trait than does the retina of the brown lemur. Of course, the activity patterns of the brown lemur are typically judged to be cathemeral, so the reader may ﬁnd tenuous any implications of this argument to the ring-tailed lemur. Against that, we note two things: ﬁrst, that ﬁeld observations often suggest that ring-tailed lemurs are less stringently diurnal than, say, Propithecus (Jolly, 1966), and an analysis of several social and morphological features of the gregarious lemurs concluded that Lemur catta is “mainly diurnal” and that they, along with Propithecus, are not yet “fully adapted to a diurnal life style” (van Schaik and Kappeler, 1996). A second observation is that in the course of doing parallel ERG recordings on Pro- 72 G.H. JACOBS AND J.F. DEEGAN pithecus, Varecia, and Lemur, we gained the impression that cone-based signals are typically larger in the former two than in the latter. That fact could have a number of interpretations, including the possibility that the retinal cone mosaic of these species, all often considered diurnal, varies in the manner just suggested. Whether that is true or not, our examination of the cone pigments in Lemur catta suggests that it is unwise to automatically assume that all strepsirrhines with diurnal traits have M/L cone pigment polymorphisms and thus the prospect for within-species color vision variations, including potential female trichromacy. ACKNOWLEDGMENTS We are indebted to the staff at the Santa Ana Zoo (Santa Ana, CA) for their cooperation and enthusiastic help during the course of these experiments. LITERATURE CITED Bowmaker JK. 1991 Visual pigments and colour vision in primates. In: Valberg A, Lee BB, editors. From pigments to peception. New York: Plenum. p 1–9. Curtis DJ, Rasmussen MA. 2002. Cathemerality in lemurs. Evol Anthropol [Suppl] 11:83– 86. Dacey DM. 1999. Primate retina: cell types, circuits and color opponency. Prog Ret Eye Res 18:737–763. Deegan JF II, Jacobs GH. 1996. Spectral sensitivity and photopigments of a nocturnal prosimian, the bushbaby (Otolemur crassicaudatus). Am J Primatol 40:55– 66. Douglas RH, Marshall NJ. 1999. A review of vertebrate and invertebrate ocular ﬁlters. In: Archer SN, Djamgoz MBA, Loew ER, Partridge JC, Vallerga S, editors Adaptive mechanisms in the ecology of vision. Dordrecht: Kluwer Academic Publishers. p 95–162. Govardovskii VI, Fyhrquist N, Reuter T, Kuzmin DG, Donner K. 2000. In search of the visual pigment template. Vis Neurosci 17:509 –528. Heesy CP, Ross CF. 2001. Evolution of activity patterns and chromatic vision in primates: morphometrics, genetics, and cladistics. J Hum Evol 40:111–149. Jacobs GH. 1993. The distribution and nature of colour vision among the mammals. Biol Rev 68:413– 471. Jacobs GH. 1996. Primate photopigments and primate color vision. Proc Natl Acad Sci USA 93:577–581. Jacobs GH. 1998a. A perspective on color vision in platyrrhine monkeys. Vis Res 38:3307–3313. Jacobs GH. 1998b. Photopigments and seeing—lessons from natural experiments. Invest Ophthalmol Vis Sci 39:2205–2216. Jacobs GH, Deegan JF II. 1993. Photopigments underlying color vision in ringtail lemurs (Lemur catta) and brown lemurs (Eulemur fulvus). Am J Primatol 30:243–256. Jacobs GH, Deegan JF II. 2001. Photopigments and colour vision in New World monkeys from the family Atelidae. Proc R Soc Lond [Biol] 268:695–702. Jacobs GH, Deegan JF II. 2003a. Photopigment polymorphism in prosimians and the origins of primate trichromacy. In: Mollon JD, Pokorny J, Knoblauch K, editors. Normal and defective colour vision. Oxford: Oxford University Press. p 14 –20. Jacobs GH, Deegan JF II. 2003b. Cone pigment variations in four genera of New World monkeys. Vis Res 43:227–236. Jacobs GH, Neitz J. 1987. Inheritance of color vision in a New World monkey (Saimiri sciureus). Proc Natl Acad Sci USA 84:2545–2549. Jacobs GH, Deegan JF II, Neitz JA, Crognale MA, Neitz M. 1993. Photopigments and color vision in the nocturnal monkey, Aotus. Vis Res 33:1773–1783. Jacobs GH, Neitz J, Krogh K. 1996a. Electroretinogram ﬂicker photometry and its applications. J Opt Soc Am [A] 13:641– 648. Jacobs GH, Neitz M, Deegan JF, Neitz J. 1996b. Trichromatic colour vision in New World monkeys. Nature 382:156 –158. Jacobs GH, Neitz M, Neitz J. 1996c. Mutations in S-cone pigment genes and the absence of colour vision in two species of nocturnal primate. Proc R Soc Lond [Biol] 263:705–710. Jacobs GH, Deegan JF II, Tan Y, Li W-H. 2002. Opsin gene and photopigment polymorphism in a prosimian primate. Vis Res 42:11–18. Jolly A. 1966. Lemur behavior: a Madagascar ﬁeld study. Chicago: University of Chicago Press. Kay RF, Kirk EC. 2000. Osteological evidence for the evolution of activity pattern and visual acuity in primates. Am J Phys Anthropol 112:235–262. Macdonald D, editor. 2001. The new encyclopedia of mammals. Oxford: Oxford University Press. Marmor MF, Zrenner E. 1999. Standard for clinical electroretinography (1999 update). Doc Ophthalmol 97:143–156. Martin RD. 1990. Primate origins and evolution. Princeton: Princeton University Press. Nathans J, Thomas D, Hogness DS. 1986. Molecular genetics of human color vision: the genes encoding blue, green and red pigments. Science 232:193–202. Newcombe RG. 1998. Two-sided conﬁdence intervals for the single proportion: comparison of seven methods. Stat Med 17:857– 872. Pariente GF. 1979. The role of vision in prosimian behavior. In: Doyle GA, Martin RD, editors. The study of prosimian behavior. New York: Academic Press. p 411– 459. Peichl L, Rakotondraparany F, Kappeler P. 2001. Photoreceptor types and distributions in nocturnal and diurnal Malagasy primates. Invest Ophthalmol Vis Sci 42:48. Rohen JW, Castenholtz A. 1967. Uber die Zentralisation der Retina bei Primaten. Folia Primatol (Basel) 5:92–147. Rowe N. 1996. The pictorial guide to the living primates. East Hampton, NY: Pogonias Press. Sharpe LT, Stockman A, Jagle H, Knau H, Klausen G, Reitner A, Nathans J. 1998. Red, green, and red-green hybrid pigments in the human retina: correlations between deduced protein sequences and psychophysically measured spectral sensitivities. J Neurosci 18:10053–10069. Sussman RW. 1999. Primate ecology and social structure. Needham Heights, MA: Pearson Custom Publishing. Tan Y, Li W-H. 1999. Trichromatic vision in prosimians. Nature 402:36. Tattersall I. 1987. Cathemeral activity in primates: a deﬁnition. Folia Primatol (Basel) 49:200 –202. van Schaik CP, Kappeler PM. 1996. The social systems of gregarious lemurs: lack of convergence with anthropoids due to evolutionary disequilibrium? Ethology 102:915–941. Walls GL. 1942. The vertebrate eye and its adaptive radiation. Bloomﬁeld Hills, MI: Cranbrook Institute of Science. Wolin LR, Massopust LC. 1970. Morphology of the primate retina. In: Noback CR, Montagna W, editors. The primate brain. New York: Appleton-Century-Crofts. p 1–27.