Widely, if inaccurately, known as the American chameleon, Anolis carolinensis is renowned for its ability to change color from a sparkling emerald to a deep brown. Surprisingly, we don’t really know what factors determine whether a particular lizard chooses to be green or brown at a particular time.
Here’s what I had to say about it in Lizards in an Evolutionary Tree (pp. 279-281; I’ve omitted most references here):
“In theory, we might expect green anoles to match their background, turning green when in vegetation and brown when against a woody surface. Although widely believed, this idea is not strongly supported (reviewed in Jenssen et al., Herp. Monographs, 1995). In one study, male A. carolinensis mismatched the surface upon which they sat (green on brown substrate or vice versa) more often than would be expected by chance (Jenssen et al., 1995; but see Medvin [Animal Behaviour, 1990] for an opposite result). Indeed, males of green species often adopt a bright green coloration when in the survey posture, although a darker appearance would almost surely be more cryptic against a woody background; this tendency suggests the possibility that skin color is being used to make the lizards more, rather than less, conspicuous (e.g., Macedonia’s work on A. conspersus and Trivers’ on mating A. garmani).”
You’d think we’d have a better idea of what’s going on with color in as common and obvious a species as A. carolinensis, but even Jenssen et al.’s very detailed field behavioral studies only began to suggest some ideas. Now, a recent study of A. carolinensis on the Japanese island of Chichi-Jima has tried to take this further. By noting the color of 169 anoles encountered in the field, Yabuta and Suzuki-Watanable tried to look for correlates of color. As with previous work, they found no evidence that the anoles were matching their background (incidentally, this is true of real chameleons as well—they apparently don’t change color to blend in). Other data, however, suggested a possible role for green as a social signal. First, adult males were more often green than other animals and, second, a weak association existed between perch height and proportion of adult males that were green. By contrast, no association was found between air temperature and color, thus working against the hypothesis that dark color was used in thermoregulation.
Clearly, there is a lot yet to be learned about color change in A. carolinensis, not to mention the many other—as yet unstudied—green anoles. One wouldn’t think such work would be all that difficult, at least the field component as an initial starting point. This study is a nice example of how one might start such a study, simply by going out and seeing under what conditions anoles are green versus brown, and how that varies between sexes and size classes.
One last quote for LIAET, because it’s tangentially relevant:
“Color change is under hormonal control in anoles and often occurs in social encounters (reviewed in Greenberg [2002, 2003]; the physical mechanism by which color change is produced is reviewed in Cooper and Greenberg ). For example, almost half of the instances in which A. carolinensis males changed from green to brown occurred in the context of aggressive encounters (see also Trivers  on A. garmani). Many of these occurred as males were approaching the boundary of their territory, but before an opponent was visible (e.g., the male which owned the adjacent territory was on the other side of a tree trunk), which suggests that the male in some sense anticipated an agonistic encounter (Jenssen et al., 1995). In general, dark color is a response to heightened stress, although a variety of other factors—including predation attempts, temperature, and light levels—also affect color in A. carolinensis (reviewed in Jenssen et al., 1995; Greenberg, 2003). During male-male interactions, lizards will change color frequently; by the end of the encounter, the winner is usually green and the loser brown (Greenberg, 2003).”