Context. Animal-borne devices can affect animal survival, reproduction, and behaviour through both the addition of weight and bulk and the direct effects of initial and subsequent capture. Researchers commonly employ a general rule of thumb that weight of the device must be less than 5% of bodyweight for terrestrial animals; however, this threshold has little empirical basis.

Aims. We evaluated the effects of environmental variables, repeated capture, and weight of animal-borne devices on bodyweight in free-ranging feral cats.

Methods. We recaptured feral cats at varying frequencies, wearing GPS and/or VHF collars that ranged from 0.29% to 4.88% of bodyweight, and recorded change in cat weight over time.

Key results. Collar weight as a percentage of bodyweight was not a significant predictor of feral cat weight change. Rather, change in bodyweight was best described by a negative relationship with an increasing temperature and number of captures, and a positive relationship with time since collar attachment.

Conclusions. Capture had a significant influence on feral cat weight but collar weights up to 5% of bodyweight did not significantly contribute to weight loss. However, the absence of control cats without collars hindered definitive conclusions on the effect of collar weight on cat weight change.

Implications. Researchers should space capture and handling events more than 30 days apart to reduce effects of weight loss from capture and handling. Researchers should also consider increasing collar weight and reducing frequency of capture (where collars are less than 5% of bodyweight), particularly if cat bodyweight is a parameter of interest.