When patterns of abundance of several species occupying the same habitat are
being quantified, different-sized sampling units can be used for the different species
simultaneously. Therefore, for intertidal and subtidal rocky shores in Australia, a
50 cm×50 cm quadrat has been shown to provide independent samples that
accurately quantify patterns of dispersion and abundance for several of the larger
gastropods and barnacles (Underwood, 1981). There are many smaller species
that show most variability at the scale of a few centimetres, e.g. littorinid snails
(Underwood, 1981). To count these in a 50 cm×50 cm quadrat is not only time-
consuming (because there may be many thousands in a quadrat of this size), but
would also not measure variation at the scale that is most important for these
small animals. It is more useful, therefore, to count these and similar small species
in replicated, smaller subquadrats, scattered representatively over the area of the
large quadrat. Similarly, there may be large animals, such as sea urchins or large
whelks that may be seldom sampled in a 50 cm×50 cm quadrat, unless a very
large number of quadrats is used. Therefore, these animals may be sampled in large
units, e.g. 3 m×3 m, within which one may sample some 50 cm×50 cm quadrats,
within each of which some 5 cm×5 cm subquadrats (Fig. 1.6) may be sampled.
Each of these scales is relevant to the species being measured and no single scale
is appropriate for all species.