**4. The protein limitation hypothesis**

32 Zoology

leaves from mature leaves, rather than the ability to distinguish ripe fruits, that provided the selective force behind routine trichromacy. Measuring the luminance of young leaves and fruit consumed by catarrhines in Kibale, Uganda they found that while fruits could be reliably distinguished by the yellow-blue colour channels alone, the detection of young, proteinaceous leaves required the red-green colour channel of the trichromat. They additionally demonstrate that red-greenness of young leaves is positively correlated with the ratio of protein content to toughness. These observations formed the basis for the hypothesis that the evolution of both the routine trichromacy of catarrhines and the polymorphic trichromacy of platyrrhines results from the ability to detect fallback foods

Dominy et al. (2003) hypothesize that the evolution of routine trichromacy in catarrhines is the result of selection for colour vision that allows for the detection of fallback foods that sustain catarrhine populations during periods of low overall fruit availability1. In the case of catarrhines, trichromacy thus evolved as a means of detecting young, red leaves as fallback foods. Fallback foods have been operationally defined as "items assumed to be of relatively poor nutritional quality and high abundance, eaten particularly during periods when preferred foods are scarce" (Marshall & Wrangham, 2007, p. 1220). Dominy et al. (2003) posit that climatic cooling at the end of the Eocene resulted in the local extinction or decimation of figs and palms which led catarrhines to "fall back" on leaf resources in periods of low fruit availability. In the neotropics, where figs and palms remain abundant, platyrrhines have evolved mixed capabilities for the detection of these cryptic fallback

Young leaves in the Old World are often red (Dominy et al., 2002). An estimated 50-62% of Old World species display red young leaf flush, compared to only 18-36% of New World species (Dominy & Lucas, 2001). The redness of young leaves results from delayed greening wherein plants postpone chloroplast function until full leaf expansion (Dominy & Lucas, 2001), coupled with the presence of anthocyanin pigment (Lee et al. 1987) which has several potential selective advantages to plants. Advantages to plants include fungicidal properties, photoprotection against UV damage of new leaves, the prevention of photoinhibition, and crypsis to protect against dichromatic herbivores (Coley & Aide, 1989; Gould et al., 1995; Stone, 1979). In a review of the evidence for and against these hypotheses, Dominy et al. (2003) conclude that crypsis is the most plausible, and that young leaf reddening is thus a plant strategy to reduce new leaf damage by herbivores. Providing evidence from the colour space occupied by young leaves, Dominy et al. (2002) conclude that to a dichromatic herbivore a young, red leaf would appear dark, dead, and not worth consuming. In Old World forests where young leaf flush coincides with low fruit availability, the ability to detect these important fallback foods would provide a clear advantage favoring trichromacy

1 Though Dominy et al. (2003) use the term "keystone resources" to refer to young leaves and figs, they define these as "resources (that) sustain frugivore populations during crucial periods" and that "consistently provide food during community-wide periods of fruit dearth" (p. 27). Keystone resources are more commonly defined as preferred resources (Terborgh, 1986). Dominy et al.'s (2003) definition is more commonly applied to foods that are considered "fallback foods" (Marshall & Wrangham, 2007)

(Dominy et al., 2003).

resources.

(Dominy & Lucas, 2001).

and as such we use that term here.

Trichromacy is useful for visually detecting long-wavelength colours of food targets against a green mature leafy background, whether the food targets are young red leaves or ripe fruits. The debate regarding whether frugivory or folivory has favored trichromacy continues. Both sides are supported and perhaps both the frugivory and the folivory hypotheses are partially correct. Here we present a new hypothesis that builds on both the frugivory and the young leaf hypotheses. We suggest that trichromacy is adaptive for finding a limiting resource critical to primates – protein. We further suggest that selective pressures may vary between catarrhine and platyrrhine primates based on food availability and body size constraints. Specifically, while protein sources for Old World primates are predominantly red leaves, among New World primates, they are red figs. Below we outline the basis for our unifying protein limitation hypothesis by reviewing the evidence that protein is a crucial limiting resource and the evidence that red figs provide a consistent and favored source of protein for New World primates and that trichromats have a foraging advantage on these fruits.
