Looking back on why we look forward: A special research series by Adam Glover, Part 4

One of the weaknesses of the Visual Predation hypothesis is that when these primates do prey on insects, they typically do so using their sense of smell or hearing rather than vision; in fact, Lorisinae- which Cartmill pointed out as the culmination of the trend towards orbital convergence in primates- detect 85-95% of their prey using a highly developed sense of smell (Charles-Dominique, 1977). A recent investigation into the sensory basis of food detection in gray mouse lemurs (Microcebusmurinus) (Siemers etal 2007) also supports this claim. In this study, two separate experiments were designed such that the lemurs would be offered two similar plastic dishes. In the first experiment, which sought to test whether visual cues are sufficient for finding arthropodan prey, both dishes were covered with transparent lids and inside one was placed a plastic arthropodan dummy. In the second experiment, which sought to test whether the lemurs could make use of other types of sensory information to detect prey, both dishes were covered with opaque lids and inside one was placed a live insect that would provide both auditory (i.e., walking sounds) and olfactory cues.

Results obtained in the first experiment (Fig. 2a) showed that the ability of all 6 lemurs to visually detect prey and open the dish with the dummy first did not significantly deviate from chance level (50%). In the second experiment however, two lemurs opened the dishes holding the live insect with a frequency significantly higher than would be predicted by chance, and two others showed the same tendency (albeit below the level of statistical significance) (Fig. 2b). Thus the null hypothesis that there would be no significant difference between the number of correct (open prey dish first) and wrong decisions (open empty dish first) made by the mouse lemurs was accepted for all mouse lemurs the first experiment, and it was rejected in two mouse lemurs for the second, calling into question the viability of visual predation as the primary selection pressure for orbital convergence.

Additionally, while smaller primates do eat insects to some degree, the vast majority (over 95%) have been found to be omnivorous (Harding, 1981). The confluence of these confounds led to a challenge of this hypothesis as well; this challenge, which came to be known as the Angiosperm or Omnivore hypothesis, also suggested that the first primates were omnivores, primarily eating fruits and other plant material while taking insects more opportunistically (Sussman, 1991). Specifically, it hypothesized that the unique adaptations of the first primates (including their characteristic degree of orbital convergence) were a product of “diffuse coevolutionary interactions” with angiosperms which offered primates a variety of food items (fruits, flowers, etc., as well as insects) and angiosperms a means of dispersal for their fruit and seeds. According to Sussman, the unique visual adaptations of primates would increase the animals’ ability to feed on and manipulate these items of very small size (e.g. fruits, flowers, and insects) at very close range by way of increased power of discrimination and precise coordination. [Overlap between the fields of vision of the two eyes provided by orbital convergence increases the probability of capturing light within that region of overlap by a factor of approximately 1.25-2 (Warrant, 2008).]

This hypothesis has been weakened, however, by findings which suggest that diet makes little difference in the way of orbital convergence; wooly possums (Caluromys) from South America, for example, have been found to eat both fruits and insects at the tips of fine, terminal branches but show little orbital convergence (Rasmussen, 2002). But if diet makes little difference in the way of orbital convergence, then what type of selection pressure favored anteriorly directed eye orbits in ancestral primates? One hypothesis is that this trait is a protective adaptation- as in one that would protect primates from being preyed upon (Isbell, 2009). Specifically, this hypothesis- known as the Snake Detection hypothesis- proposes that the unique visual adaptations of primates evolved in part to assist primates in detecting and avoiding snakes.

Fig.2. Exploring the role of visual, olfactory, and auditory information in mouse lemurs ability to detect arthropods (Microcebusmurinus) (Siemers et al 2007). a) Two similar plastic dishes were covered with transparent lids and presented to the lemurs, one of which contained a plastic arthropodan dummy providing strictly visual cues for detection. b) Two similar plastic dishes with opaque lids were presented to the lemurs, one of which contained a live insect that would provide both auditory and olfactory cues for detection. Numbers of correct (open prey dish first) and wrong (open empty dish first) decisions were recorded. K indicates lemurs from Kirindy in the western dry forest of Madagascar. Asterisks denote level of significance as indicated by a 1-sided binomial test (***<0.001), whereas n.s. denotes no significance. A 1-sided test was chosen because it was not expected that the number of correct decisions (open prey dish first) made by the mouse lemurs (n=6, 5) would be significantly lower than the number of wrong decisions given the provided sensory information.