Jumping data To test jump gerbil performance, we built a vertical tunnel (10x10x47-cm) using clear 1/16-mm polycarbonate sheets, including a ~0.5-cm gap at the bottom, and designed to not exert weight on the ATI mini40 force plate. We placed gerbils in the tunnel to acclimate for 30 seconds and set the force plate to record data at 1,000 samples per second using Igor (v7, Wavemetrics, Lake Oswego, OR, USA). Each trial began when gerbils stood on four feet and were sprayed with compressed air to simulate surprise by ambush predators. We recorded videos at 700 fps using two Phantom MIRO cameras and Phantom Camera Control software. To increase chances of measuring maximum jump force, we completed 2-4 trials for 20 gerbils (N = 12 males, 8 females); we stopped after 1 trial for 8 gerbils that exhibited stress (N = 2 males, 6 females). We scored each of 57 trials involving 28 individuals as a: (i) 'jump' in which the gerbil pushed off the ground vertically using its hindfeet, (ii) 'hop' on all four feet, or (iii) any 'other' response. We obtained time-matched morphological measurements for 15/17 of the gerbils who jumped (N = 11 males, 4 females), and the trials for which we recorded jumps are included in the attached dataset. We calculated maximum resultant force from the vector sum of forces in the x, y, and z directions and expressed this value in proportion to the gerbil's mass. We low-pass filtered force plate data at 70 Hz to account for vibrations. Resultant force outputs were given in body weight units (BWU); a BWU of one is equivalent to a jump force of 1x body weight. We used the maximum force recorded for each gerbil for analysis. Maneuverability data To quantify gerbil maneuverability, we conducted corner-turning trials. We built polycarbonate corridors of increasingly difficult turn angles (135°, 90°, 45°) with a 120-cm runway into the turn and 80-grip black traction tape to line floors and visually indicate turns. Gerbils acclimated by exploring the corner and back at least three times before we sprayed them with compressed air to begin the trial. We recorded runs at 240 fps using GoPro Hero 10 Black cameras (one above the arena; one lengthwise along the runway). We ended up using the top camera view for analysis, and the tracking data from that camera are included in these datasets. We tested each gerbil up to three times on the three angles, allowing >48 hours between tests and randomizing the order of angles. We categorized each of 226 trials across 29 individuals (14 male; 15 female) as a: (i) 'turn' if the gerbil's body was perpendicular to the lane after the turn (47.8%) (ii) 'no turn' if it stopped before its body was perpendicular (50.9%) or (iii) 'fail' if the gerbil ran into a wall (1.3%). We excluded failed trials from analyses, along with 15 trials for which a camera malfunctioned and 3 for which we did not have timely trait measurements. The final dataset included 192 trials across 28 individuals (N = 63 for 135°, 65 for 90°, 64 for 45°). To calculate turn speed for each maneuverability trial, we tracked the tip of each gerbil's nose through the trial using DLTdv8 in MATLAB R2022a to calculate instantaneous velocity using trajr and used filter order 3 with a filter length of 21 to smooth the trajectory. We calculated the average speed around the corner over 0.1-s intervals. The fastest turn speed was selected for each gerbil and angle, yielding 52 top-performance trials across 24 individuals (11 males, 13 females).

Theoretically, animals with longer hindlimbs are better jumpers, while those with shorter hindlimbs are better maneuverers. Yet experimental evidence of this intuitive relationship is lacking. We experimentally compared jump force and maneuverability in a lab colony of Mongolian gerbils (Meriones unguiculatus). We hypothesized that gerbils with long legs (ankle to knee) and thighs (knee to hip) would produce the greatest jump forces, while gerbils with short legs and thighs would be able to run most rapidly around turns. Consistent with these hypotheses, gerbils with longer legs produced greater jump forces after accounting for sex and body mass: a 1-mm greater leg length provided 1 body-weight-unit greater jump force on average. Furthermore, gerbils with shorter thighs were more maneuverable: each 1-mm greater thigh length reduced turn speed by 5%. Rather than a trade-off, however, there was no significant correlation between jump force and turn speed. The experiments revealed how distinct hindlimb segments contributed in different ways to each performance measure: legs to jumping and thighs to maneuvering. If variation in jumping and maneuvering influences survival during predator encounters, then hindlimb segment lengths may be subject to strong natural selection.

Funding provided by: Brown UniversityROR ID: https://ror.org/05gq02987Award Number: Funding provided by: Brown UniversityROR ID: https://ror.org/05gq02987Award Number: Funding provided by: Society for Integrative and Comparative BiologyROR ID: https://ror.org/006qmn341Award Number: