The objective of this study was to assess the relative importance of lamb birth weight (BW) and variation between BW of littermates (VAR) on lamb performance. Records from lambs born into twin (n = 4160) or triplet (n = 1279) litters from 1988 to 2005 were analyzed. Data collected for each lamb included dam, sire, sex, BW, birth date, lamb fate, and weaning weight and date. Data were analyzed with REML with BW and VAR grouped into classes (BW (kg): ≤2, >2–3, >3–3.5, >3.5–4, >4–5, >5; VAR (kg): 0–0.1, >0.1–0.35, >0.35–0.5, >0.5–0.75, >0.75–1.00, > –1.3, >1.3). The model included BW, VAR, sex and year of birth as fixed effects with sire of lamb included as a random effect. Overall, 85.3% of twins and 74.7% of triplet lambs survived to weaning. For both twin and triplet lambs, BW was associated with survival (P < 0.001) with lambs < 2 kg having very low survival rates (<40%). For triplets and the larger of twin lambs, VAR was not associated with survival. However, VAR did affect survival of the smallest lamb in twin litters (P < 0.001) with lambs with VAR > 1.3 kg being less apt to survive (67.8%) compared with those from litters of smaller VAR (range 80.4–86.2% survival). Growth rate of twin and triplet lambs was affected by BW (P < 0.001). As BW increased, growth rate also increased. For triplet born and raised lambs, VAR did not affect growth of the lamb. Surprisingly, twin lambs from litters with high VAR had slightly greater growth rate than those from lambs of low VAR (P < 0.05). Given the strong effect of BW on lamb performance, the large data set of twins was examined to determine whether variation in ovulation rate or pattern influenced BW. While year born, lamb birth date and sex and ewe age and weight before mating affected BW of the lamb (P < 0.001), neither ovulation rate nor pattern (i.e., both ova released from one ovary or one from each) were associated with BW of the lambs. In summary, the BW was strongly associated with survival and growth of the lamb. The litter VAR had little or no effect on survival and did not negatively affect growth of the lamb. Thus, birth weight, but not variability in birth weight within a litter, appears to be a major driver of lamb performance.