Predicting how many individuals of a predatory species can survive within a given area, based on prey biomass, is important in understanding whether or not significant threats are related to prey availability. The study of ungulate populations and, particularly, reliable estimates of ungulate biomass are necessary for effective conservation and management of the wild Amur tiger. Dense forests, mountainous landscape and low detection probabilities preclude the use of direct count techniques in estimating ungulate populations, while direct methods are expensive and time-consuming. Indirect sampling, by counting footprints or snow tracking, is a widely use, reliable and inexpensive way of estimating ungulate populations and calculating biomass. Amur tiger-prey relationships are so close that data on prey (ungulate) availability may be used to reliably predict the tiger population and determine any threats or risks to the endangered Amur tiger. Little is known about the relationship between the Amur tiger population and ungulate biomass in the easte Wanda Mountains of Heilongjaing Province, China. Therefore, we collected data on the population size of three ungulate species (wild boar Sus scrofa, red deer Cervus elaphus and roe deer Capreolus capreolus) by establishing 240 line transects within 48 random sampling sites during late winter 2008 to early spring 2009. The results show that prey biomass, represented by the three ungulate species, was 1 85 849.00 205 335.00 kg, including 74 767.50 87 825.00 kg from wild boar (502 606 adults and 209 210 sub-adults), 79 744.50 85 984.50 kg from red deer (331 357 adults and 67 72 sub-adults) and 31 337.00 31 525.50 kg from roe deer (810 815 adults and 202 203 sub-adults). In the study area, the estimated total biomass of all prey species was 209 619.89 231 598.24 kg. The prey biomass, represented by three ungulate species, could support 5.22 6.92 Amur tiger individuals and the estimated total biomass of all prey species could support 5.89 7.81 tigers, assuming 8% biomass as the food demand of Amur tigers in the eastern Wanda Mountains. Finally, to evaluate the accuracy and precision of the line transect surveys, we analyzed the relationship between footprint frequency and sampling effort by bootstrap analysis, The differences in the coefficients of variation for footprint frequency were as follows: wild boar: 37.98% for 1 120 line transects and 2.74% for 121 240 line transects; red deer: 17.41% for 1 150 line transects and 2.86% for 151 240 line transects; roe deer: 39.72% for 1 115 line transects and 3.84% for 116 240. Trend analysis indicated that population sizes could reasonably be established from 120 line transects for wild boar (sampling distance: 600 km), 150 line transects for red deer (sampling distance: 750 km) and 115 line transects for roe deer (sampling distance: 675 km). Sampling effort, both in terms of the number of footprints and intensity of sampling, had a marked effect on the accuracy and precision of the survey results. These findings will provide scientific guidelines for the estimation of prey (ungulate) population size and conservation of the Amur tiger.