Records on 175 young Angus heifer and bull progeny from 46 sires, measured for methane production in respiration chambers, were used to evaluate the repeatability of methane measurement over short- and long-term periods. The traits assessed were dry matter intake (DMI), methane production rate (MPR), methane yield (MPR per unit DMI), and four residual methane (RMP) traits. The RMP traits were computed as actual MPR minus expected MPR, where the expected MPR for the first three RMP traits were calculated from three different published and widely used equations. The expected MPR for the fourth was computed by regressing MPR on DMI, using the data from the study. Animals underwent an initial (first) methane measurement test for 48 h, and one repeat methane measurement test up to 450 days after the first test. Repeat tests were classified into four different time periods: tested across consecutive days; re-tested within 60 days of first test; re-tested 61–120 days after first test; and re-tested 121–450 days after first test. Repeatabilities were calculated for all traits across all time periods, and phenotypic correlations for the same trait measured over time were obtained from bivariate analyses. Methane traits from tests conducted over consecutive days were highly repeatable (0.75–0.94) and highly phenotypically correlated (0.85–0.95). Repeatabilities from tests conducted within 60 days of the first test were moderate to high (0.59–0.91), whereas phenotypic correlations were, in general, moderate (0.30–0.44), with the exception of MPR (0.78). Results for both longer-term time periods (61–120 days and 121–450 days after the first test) were very similar, with low estimates of repeatabilities (0.16–0.27) and phenotypic correlations (0.12–0.27). Correlations between sire progeny means from the first and repeat methane test were moderate (0.46–0.77) for all traits except RMPR (0.19). Results from this study indicate that methane traits from tests conducted either on consecutive days or within a short-term time frame afterward (~60 days) are highly repeatable and highly phenotypically correlated. However, methane tests conducted over longer-term time frames are substantially, but consistently, less repeatable and are lowly phenotypically correlated, which indicates that multiple measures may be required to accurately record methane traits over the life time of an animal.