The Amur tiger (Panthera tigris altaica) population in China, once widespread, is now reduced to an estimated 20 individuals widely dispersed over a large area. The Chinese government is making concerted efforts to restore this population from the contiguous Russian population. However, they face a challenge in finding an effective monitoring technique. We report on the development of a robust, non-invasive and cost-effective technique to identify the sex of Amur tigers from snow footprints. Between December 2011 and December 2012, we collected 523 digital images of left-hind footprints from 40 known captive Amur tigers (19 F, 21 M), of age range 3–13 years (F mean age = 8.07 ± 0.18, M mean age = 8.36 ± 0.19; F = 1.18, P > 0.05). Images were captured with compact digital cameras according to a standardized photographic protocol (Alibhai et al. 2008). Using JMP software from the SAS Institute, 128 measurements were taken from each footprint according to the protocol developed by Alibhai et al. (2008), and were subjected to a stepwise selection. With just 10 variables, and testing with both Jackknifing and 50% holdout methods, the resulting algorithm for sex determination gave 98% accuracy for individual footprints. The algorithm derived from captive tiger footprints of known sex was then used to identify the sex of 83 footprints from 8 trails collected from unknown free-ranging Amur tigers in the winter from the end of 2011 to the beginning of 2012. The algorithm predicted 5 trails from females and 3 from males. This technique is a potentially valuable tool for monitoring the recovery of Amur tiger populations at the landscape scale in northeastern China. © 2014 The Wildlife Society.