Rand sai Po 3 
In larger faunas the situation is quite otherwise, The islands are 
large, the predators relatively numerous. Body pattern is now less 
important, and dewlap color and pattern usually highly important, especially 
so in animals similar in size and shape and ecology (i.e. representatives of 
the same ecomorph). 
Here the mobility of the dewlap increases its utility as a cue to species 
identity: when social situations require, the animal may flaunt its identity; 
on other occasions, and particularly if predators appear, the dewlap is 
retracted and the animal may become relatively cryptic, 
Body pattern, size and shape may all be aids to species recognition. 
The new function of the dewlap in complex faunas adds a new signal for the 
same purpose, It adds redundancy, and as Rand and Williams have said 
previously, redundancy is itself selected because it is important in allowing 
species recognition in a variety of circumstances. 
Rand has some observations on the social aspects of thermal niche in 
Anolis oculatus. His conclusions are drawn from data collected first with 
Ross Kiester and later with Patricia Rand at the locality on the west coast 
of Dominica already discussed by Ruibal and Philobosian (1971) as showing 
the wide thermal niche width of a "solitary" anole. 
Kiester and Rand in November found a difference between the populations 
in open sunny coconut groves (mostly small adults and more males than 
females) and in adjacent denser stands of trees with heavier shade (large 
males, more females and many juveniles). 
In May at the beginning of the rainy season, Rand and Rand saw no 
juveniles but a shaded hedge now had a higher proportion of males to females 
and a larger average size of adult males than adjacent rows of coconuts. 
The differences were statistically significant. 
Rand suggests that in areas of high population density the preferable 
microhabitats - the shaded areas - are inhabited by the largest males. 
The smaller males are forced into adjacent tolerable but less preferable 
areas. He further suggests that in the Greater Antilles, with many species 
present, adjacent habitats not optimal for a given species would be occupied 
by other species, forcing younger males to wider dispersal in the search for 
suitable habitats. 
Rand therefore predicts (1) that in single species islands like Dominica 
the spread of a species through adjacent differing microhabitats will be 
greatest where the population densities of that species are greatest and 
least where they are least; (2) in the Greater Antilles such spread of a 
species through differing microhabitats will occur only in local habitats 
where one species is very abundant and others rare or absent, i.e. situations 
that locally approximate the conditions of one species islands, Rand, 
however, further predicts that where, as on single species islands, an 
appreciable portion of the population is forced to live in sub-optimal 
habitats selection will favor adaptations which fit the population for the 
wider niche, 
