17. Genetic Diversity of Primate Color Vision
271
Color vision is apparently of no use to a nocturnal animal. Indeed, all the nocturnal primates (the owl monkeys and several prosimians) that have been studied
were found to have lost color vision - although their green\red opsin gene is functional and has been well conserved, their blue opsin gene has become nonfunctional. (Note that color vision requires at least two color photopigments.) In contrast, all diurnal pro simians studied were found to have two functional opsin genes
(the blue and green\red opsin genes). Thus, for diurnal primates dichromacy is
evidently more advantageous than monochromacy.
As noted above, the single-origin hypothesis for the triallelic system appears to
be far more plausible than the multi-origin hypothesis. Note that the single-origin
hypothesis implies that the triallelic system has persisted in these NW monkeys for
more than 20 million years, because the divergence of the howler monkey lineage
and the squirrel monkey-marmoset lineage has been estimated to be about 20 million years (Schneider et al. 1993, 1996). The antiquity of the system strongly suggests balancing selection for the maintenance of the system, because without balancing selection one or two of the three alleles would have become lost in a relatively short time (Kimura and Ohta 1969; Takahata and Nei 1990). Note that the
existence of three rather than two polymorphic alleles at the X-linked opsin locus
increases the chance of being heterozygous and thus of being trichromatic for a
female NW monkey. On the other hand, if the triallelic system has multiple origins,
then the repeated occurrences of the system and the numerous parallel amino acid
substitutions required to explain the sequence differences within and between species (see Fig. 3) also suggest positive Darwinian selection. Actually, even the most
parsimonious tree (Fig. 4) requires several parallel substitutions at some critical
sites; e.g., at least 4 F-Y changes at site 277. Thus, regardless of which of the two
hypotheses is true, the triallelic system probably has been maintained by natural
selection.
8 Concluding Remarks
It is clear that much progress has been made in the evolutionary genetics of primate
color vision. However, several issues remain unresolved. For example, although
the single-origin hypothesis of the triallelic system in NWMs is more plausible
than the multiorigin hypothesis, it remains to be substantiated by further data. The
major difficulty is that gene conversion has drastically distorted the evolutionary
history of genes and alleles. One way to overcome this problem is to have sequence
data from many NWM species, so that the effect of gene conversion can be better
understood and incorporated into phylogenetic analysis. As another example, the
evolutionary genetics of pro simian color vision has not been well studied. Our preliminary data suggest that there is polymorphism at the X-linked locus in white
sifakas (diurnal), but the extent of polymorphism is not clear and it is not known
whether there is also polymorphism in other prosimian species, especially noctur-
271
Color vision is apparently of no use to a nocturnal animal. Indeed, all the nocturnal primates (the owl monkeys and several prosimians) that have been studied
were found to have lost color vision - although their green\red opsin gene is functional and has been well conserved, their blue opsin gene has become nonfunctional. (Note that color vision requires at least two color photopigments.) In contrast, all diurnal pro simians studied were found to have two functional opsin genes
(the blue and green\red opsin genes). Thus, for diurnal primates dichromacy is
evidently more advantageous than monochromacy.
As noted above, the single-origin hypothesis for the triallelic system appears to
be far more plausible than the multi-origin hypothesis. Note that the single-origin
hypothesis implies that the triallelic system has persisted in these NW monkeys for
more than 20 million years, because the divergence of the howler monkey lineage
and the squirrel monkey-marmoset lineage has been estimated to be about 20 million years (Schneider et al. 1993, 1996). The antiquity of the system strongly suggests balancing selection for the maintenance of the system, because without balancing selection one or two of the three alleles would have become lost in a relatively short time (Kimura and Ohta 1969; Takahata and Nei 1990). Note that the
existence of three rather than two polymorphic alleles at the X-linked opsin locus
increases the chance of being heterozygous and thus of being trichromatic for a
female NW monkey. On the other hand, if the triallelic system has multiple origins,
then the repeated occurrences of the system and the numerous parallel amino acid
substitutions required to explain the sequence differences within and between species (see Fig. 3) also suggest positive Darwinian selection. Actually, even the most
parsimonious tree (Fig. 4) requires several parallel substitutions at some critical
sites; e.g., at least 4 F-Y changes at site 277. Thus, regardless of which of the two
hypotheses is true, the triallelic system probably has been maintained by natural
selection.
8 Concluding Remarks
It is clear that much progress has been made in the evolutionary genetics of primate
color vision. However, several issues remain unresolved. For example, although
the single-origin hypothesis of the triallelic system in NWMs is more plausible
than the multiorigin hypothesis, it remains to be substantiated by further data. The
major difficulty is that gene conversion has drastically distorted the evolutionary
history of genes and alleles. One way to overcome this problem is to have sequence
data from many NWM species, so that the effect of gene conversion can be better
understood and incorporated into phylogenetic analysis. As another example, the
evolutionary genetics of pro simian color vision has not been well studied. Our preliminary data suggest that there is polymorphism at the X-linked locus in white
sifakas (diurnal), but the extent of polymorphism is not clear and it is not known
whether there is also polymorphism in other prosimian species, especially noctur-