The size–grain hypothesis: do macroarthropods see a fractal world?

1. In the size–grain hypothesis (a) long legs allow walking organisms to step over gaps and pores in substrate but prohibit them from entering those gaps; (b) the world is more rugose for small organisms; and (c) the relative cost of long legs increases as organisms grow smaller. The hypothesis predicts a positive allometry of leg length ( = mass b where b > 0.33 of isometry), a pattern that robustly holds for ants. 2. Toward testing for leg length allometries in other taxa, arthropods were extracted from the Panama leaf litter and measured. Three common taxa (spiders, diplopods, Coleoptera) yielded b s that exceeded 0.33 while three others (Acarina, Pseudoscorpiones, and Collembola) did not. The exponent b tended to increase ( P = 0.06, n = 7) with an arthropod taxon’s average body mass. 3. Since leg length in cursorial organisms tends toward isometry in very small and very large taxa (i.e. mammals) this suggests that the size–grain hypothesis may best apply at a transition zone of intermediate body mass: the macroarthropods. 4. Body length was a robust predictor of mass in all groups despite variation in shape.