Breeding of ruminant livestock for low CH4 emission is an attractive means of mitigating enteric CH4 emissions. However success requires that the mechanism responsible for among animal variation in emissions is repeatable and heritable and has a negligible negative impact on production and functional traits. This study was designed to estimate repeatability and heritability of the CH4 emission trait in sheep, and to determine whether the ranking of sheep based on their CH4 emissions is maintained over a range of contrasting diets. A flock of 105 ewe lambs (10 months old) of a progeny testing program were screened for their CH4 yields (i.e., emissions/unit feed dry matter (DM) intake) when a molasses containing grass silage was fed at restricted intake (1.3×maintenance metabolisable energy requirements). Methane emissions were measured on a single day in respiration chambers over 4 measurement periods with 13-15d between consecutive measurements (i.e., screening phase). Mean CH4 yield of lambs was 18.4±0.38g/kg DM intake during the screening phase, and estimates of repeatability and heritability for CH4 yield were 0.16 and 0.30, respectively. Methane yield in the screening phase was 7.9% higher for the high versus low ranked sheep (19.2±0.18 versus 17.8±0.26g/kg DM intake). The 10 lowest (low rank) and the 10 highest (high rank) CH4 yielding sheep were selected and retained for further study. Two repeated measurements of CH4 yield were conducted, the first measurement while sheep were fed fresh cut perennial ryegrass pasture (grass), the second with the same sheep fed a 400:600 concentrate:forage (wheat grain:lucerne hay; fresh basis) pelleted diet (pellet). Repeated measurements revealed that rankings were maintained among diets, but that there was a CH4 rank×diet interaction for CH4 yield. When fed the grass diet, the high ranked sheep had 13% higher CH4 yield than the low ranked sheep, but when fed the pelleted diet, the high ranked sheep had 36% higher CH4 yield than the low ranked sheep. Emissions of hydrogen were only measurable when sheep were fed the pelleted diet. This study is the first to report that ranking of sheep for CH4 emissions is consistent among diets, although the magnitude of difference among the rankings was affected by diet, suggesting that among animal variation in CH4 emission could be exploited to breed animals for low CH4 emission.This paper is part of the special issue entitled: Greenhouse Gases in Animal Agriculture - Finding a Balance between Food and Emissions, Guest Edited by T.A. McAllister, Section Guest Editors; K.A. Beauchemin, X. Hao, S. McGinn and Editor for Animal Feed Science and Technology, P.H. Robinson. © 2011 Elsevier B.V.