Numerical simulations were used to investigate the power of methods for detecting changes (between years) in pup growth rates, the loss of power associated with obtaining weight samples on only three dates in a season, and potential means of increasing the power of the CCAMLR Ecosystem Monitoring Program Standard Method C1 (B). Using estimates of variance in pup weights obtained at Seal Island (60°59.5'S, 55°24.5'W), the simulations suggest that following the protocol of the Standard Method (50 pups of each sex, weighed on four dates at 28-day intervals) would result in the ability to detect a 17% change in the growth rate with a 10% chance of committing Type 1 or Type 2 errors. If only three samples(50 pups, 28-day intervals) are obtained, the detectable change increases to about 34% under the same conditions. Slight gains in power can be obtained in some cases, without increasing the total number of pups weighed in a season,by decreasing the time between weighings.