Biological hypotheses framed in terms of ratios of measured variables run t
he risk of being untestable. Such avoidance of empirical verification resul
ts from a substantial widening of the sampling variation of ratios, compare
d to that of the original variables. Computer simulations show that in orde
r to be testable, models in which ratios are substituted for actual variabl
es would require either substantial magnitudes of the treatment effects, or
an unrealistic precision of measurements, or prohibitively large sample si
zes. Many theoretical models in biological sciences should be reformulated
to facilitate their verifiability by empirical research. Another problem wi
th the use of ratios occurs when the experimental treatment effects are exh
ibited not as shifts in means, but only in levels of variability of the var
iables. Focusing on average ratios, rather than allometric relationships be
tween variables, is likely to obscure important biological phenomena. In ge
neral, we advocate a multivariate approach, with explicit consideration of
the correlation patterns among variables.