Table 1.

A summary of the predicted form of SADs when measured as biomass and numerical abundance for different sampling regimes.

sampling regime	log biomass–log number relationship	SAD using biomass	SAD using numerical abundance
sampling restricted to a single group of organisms which all belong to the same spatial guild	biomass and numerical abundance correlated; no triangular distribution apparent	the SAD relationship cannot be inferred	the SAD relationship cannot be inferred
all organisms over a wide size range sampled with similar efficiency	biomass–abundance plot distribution forms an approximately right-angled triangle if maximum biomass is independent of numerical abundance (figure 3a)	number of species–log biomass plot reaches a maximum at the highest biomass category	number of species–log abundance plot reaches a maximum at the lowest abundance category
large organisms under-sampled	biomass–abundance plot distribution forms an obtuse triangle (figures 1 and 3b)	SAD peaks at an intermediate biomass category; it may approximate a lognormal distribution	number of species–log abundance plot reaches a maximum at the lowest abundance category
small organisms under-sampled	biomass–abundance plot distribution forms polygon ideally like an approximately right-angled triangle with the lower corner cut-off (figure 3c)	number of species–log biomass plot reaches a maximum at the highest biomass category	SAD with a maximum at the lower abundance classes; may be similar to a lognormal with veil line
both large and small organisms under-sampled	biomass–abundance plot distribution forms polygon ideally approximately kite-shaped (figure 3d)	SAD peaks at an intermediate biomass category; it may approximate a lognormal distribution	SAD peaks at an intermediate abundance category; it may approximate a lognormal distribution