Morphological traits influence the behaviors that organisms use to evade their predators. Animals escape from predators to ensure survival, but do so in ways that match the performance capabilities of their morphology. When species are sexually dimorphic in locomotion-related traits, escape locomotor performance and behavior might differ accordingly. However, this relationship remains poorly understood. Therefore, we studied one species of the arachnid group Opiliones (Leiobunum aldrichi: Sclerosomatidae) in the field. We hypothesized that morphological traits (body mass and limb length) are associated with sexual differences in locomotor performance and antipredator escape behaviors. We first quantified that males have a 39% smaller body size and 10% longer limb length than females. Then, when we experimentally simulated a situation in which individuals were grabbed and escaped from a predator, males and females exhibited similar locomotor performance (average velocity) during their initial escape bout. Males and females also showed similar patterns in three metrics of escape behaviors—movement time, freezing time, and distance traveled from the stimulus. Hence, variation in performance and escape behavior is independent of variation in morphological traits. Alternatively, the magnitude of morphological variation in these arachnids might not be enough to drive sex-specific differences in their performance capabilities or behavioral responses. Overall, our findings suggest that locomotor performance and escape behaviors might be shaped by multiple sources of selection.

Field site

We conducted research at Lake Avenue Woods East, a forest preserve site in Cook County, Illinois, USA (42°05'23"N, 87°53'06"W, 195m in elevation). We surveyed L. aldrichi between 0900 and 1700 hours in a ~5,500m² forest patch, in July and August 2025. The site is a riparian woodland with high canopy cover, dense leaf litter, and sparse vegetation that sustains high abundance and diversity of Sclerosomatidae (DeLong et al. submitted).

Locomotor performance

To measure sexual differences in locomotor performance, we first located L. aldrichi individuals and recorded the substrates where they were found: leaf litter, fallen logs, or tree trunks. We then conducted a predator simulation trial. We gently grabbed all animals by multiple legs to prevent limb loss, following a standard care and handling protocol (Escalante & O’Brien, 2024). Then, we released them < 60 s later, 150 cm away from where they were found, and at 84 cm height. We did this to simulate attack displacement and release height from the mouth of a bird or mammal predator (Pomini et al., 2010). We ensured that individuals landed in areas with low leaf litter coverage and clear of understory vegetation.

We video-recorded each individual's initial escape after reaching the ground, to quantify locomotor performance. Using an iPhone 16 Pro mounted on a tripod, we recorded videos in 1080p or 4K at 30 frames per second. Before each trial, we placed a 30 cm ruler on the ground to spatially calibrate the videos. Each individual only experienced one trial. Within a sampling day, we always moved the location of the releasing site, to minimize the potential influence of conspecific chemical cues on performance or antipredator escape behavior.

We explored the average velocity for each trial, as this metric is a relevant proxy for sustained velocity in Opiliones over periods of time of variable length (Domínguez et al., 2016; Escalante et al., 2019, 2020). For this, we analyzed videos using the Tracker software (version 6.3.2). We labeled the positions of individuals in each frame of their trial video, providing x and y coordinates for velocity calculation. Tracker software uses the following equations to calculate velocity for a given frame:

1.     In the x or y direction:

a.   v_i = (x[i + 1] - x[i - 1])/(2dt) or v_i = (y[i + 1] - y[i - 1])/(2dt)

   i.     v_i represents the velocity at a given frame

   ii.     i represents a frame/time point

   iii.     x or y represent positions along axes of the video frame in Tracker

   iv.     dt represents the time between each frame (0.033 s)

2.     Total velocity:

a.    v = sqrt(v²_x + v²_y)

We analyzed the average velocity during their initial escape bout. This bout included the time from when an individual reached the ground to when their velocity substantially decreased, or reached 0 cm/s. We qualitatively determined this cutoff for each individual by visually inspecting a plot of their velocity (y-axis) over time (x-axis). From the plots, we extracted frames with a consistent pattern of peaks and troughs indicating sustained locomotion (sensu Escalante et al., 2019).

Antipredator escape behavior trials

To measure sexual differences in antipredator escape behavior, we sequentially recorded three variables after releasing an individual from the tripod (Figure 1A): moving time, freezing time, and straight-line distance traveled. We defined moving time as the duration of active locomotion and leg II probing and tapping (Wijnhoven, 2013; Escalante et al., 2019, 2020). We consider leg tapping and probing as movement, even if locomotion does not occur simultaneously, as any movement might increase conspicuousness. For freezing time, we quantified the duration each individual remained motionless, up to a maximum of 240 seconds (freezing time rarely lasted longer in pilot trials). Finally, we measured the straight-line distance, to the nearest 0.5 cm, from where individuals reached the ground to the point where they froze. Several individuals (26%, N = 23) climbed trees or fallen logs during their escape. In these instances, we summed the distance to the substrate (horizontal) and the distance up that substrate (vertical).

After the trials, we measured morphological traits. For this, we first placed each individual in their own deli dish container (11.5 cm x 7.5 cm) supplied with a Petri dish of soaked paper towel, cubed cucumber, circular cutout of bond paper for substrate, and proper airflow. We then transported individuals to the University of Illinois - Chicago (20 km away). Animals spent less than eight hours in their individual containers, and all containers were always kept in the shade.

Morphological measurements

We measured body mass and limb length in the lab, as these traits influence locomotor performance (Grossi & Canals, 2015; Wrinn et al., 2025; Huie & Kawano, 2026; Kuchibhotla et al., 2026) and antipredator escape behavior (Fernandez-Juricic et al., 2006; Carlson et al., 2014; Mikula et al., 2017). We recorded mass to the nearest 0.0001 g by securing individuals in a 2 cm x 2 cm plastic cube that fit within the pan of a Mettler Toledo AE 50 analytical balance. We used digital calipers to measure the length of one leg IV to the nearest 0.1 mm as a proxy for limb length (Escalante et al., 2020). Leg length was the distance from the basal end of the coxa to the distal end of the tarsus. We also recorded whether each individual had all legs (intact) or did not (autotomized) to secondarily explore the potential influence of limb loss on locomotor performance and escape behavior (Escalante & O’Brien, 2024).

We returned all animals to the same collection site less than 24 hours later. After measuring and weighing them in the lab, we marked each individual with a green, non-toxic marker (Posca brand) on their rear end and placed them in a temporary mesocosm (large plastic bin with drill holes, 60 cm x 41.6 cm x 33.7 cm, with leaf litter and sticks from the field site, wet paper towels, cubed fruits or vegetables, and fish flakes). Individuals were temporarily housed in lab conditions. The following day, we returned them and avoided sampling marked animals on subsequent visits.