Parallel and divergent morphological adaptations underlying the evolution of jumping ability in ants

Jumping is a rapid locomotory mode widespread in terrestrial organisms. However, it is a rare specialization in ants. Forward jumping has been reported within four distantly related ant genera:Gigantiops,Harpegnathos,Myrmecia, andOdontomachus. The temporal engagement of legs/body parts during jump, however, varies across these genera. It is unknown what morphological adaptations underlie such behaviors, and whether jumping in ants is solely driven directly by muscle contraction or additionally relies on elastic recoil mechanism. We investigate the morphological adaptations for jumping behavior by comparing differences in the locomotory musculature between jumping and non-jumping relatives using x-ray micro- CT and 3D morphometrics. We found that the size-specific volumes of the trochanter depressor muscle (scm6) of the middle and hind legs are 3-5 times larger in jumping ants, and that one coxal remotor muscle (scm2) is reduced in volume in the middle and/or hind legs. Notably, the enlargement in the volume of other muscle groups is directly linked to the legs or body parts engaged during the jump. Furthermore, a direct comparison of the muscle architecture revealed two significant differences between in jumping versus non-jumping ants: First, the relative Physiological Cross-Sectional Area (PCSA) of the trochanter depressor muscles of all three legs were larger in jumping ants, except in the front legs ofO. rixosusandM. nigrocincta; second, the relative muscle fiber length was shorter in jumping ants compared to non-jumping counterparts, except in the front legs ofO. rixosusandM. nigrocincta. This suggests that the difference in relative muscle volume in jumping ants is largely invested in the area (PCSA), and not in fiber length. There was no clear difference in the pennation angle between jumping and non-jumping ants. However, the length of hind legs relative to body length was longer in jumping ants. Based on direct comparison of the observed vs. possible work and power output during jumps, we surmise that direct muscle contractions suffice to explain jumping performance, in two species, but elastic recoil is likely important in one. We suggest that increased investment in jumping-relevant musculature is a primary morphological adaptation that separates jumping from non-jumping ants. These results elucidate the common and idiosyncratic morphological changes underlying this rare adaptation in ants.