Breeding lines of Romney sheep, selected as lambs for consistently low or high faecal nematode egg count (FEC) following periods of natural challenge, have been maintained at Wallaceville for some years. In order to determine the extent to which FECs in low and high genotypes reflected their ability to resist the establishment of gastro-intestinal nematode burdens, we investigated the infection status and immune responses in 8- to 9-month-old progeny of selected rams from low and high FEC breeding lines following a period of grazing without anthelmintic treatment in autumn/early winter. In each of the 2 years of the study, outcross male progeny of the two lowest FEC (LFEC) (i.e. most ‘resistant’) and two highest FEC (HFEC) (i.e. most ‘susceptible’) rams from the divergent lines were slaughtered shortly after autumn/early winter FECs had been analysed. Post-mortem worm counts and examination of intestinal histology were then undertaken. Blood samples collected before slaughter in the second year of the study were assayed to measure serum levels of Trichostrongylus colubriformis-specific antibody and immunoglobulins (IgG 1 and IgM), and numbers of circulating eosinophils. Overall, correlations between pre-slaughter FEC and total trichostrongyle burdens in the lambs proved to be very high (0.91 and 0.85, respectively, for the 2 years studied). In the first year, LFEC lambs, which were shedding only 28.6% as many strongyle eggs as their HFEC counterparts at slaughter, were found to harbour 37.6% as many adult trichostrongyle worms, while in the second year, LFEC lambs, which were shedding 16.1% as many strongyle eggs as their HFEC counterparts at slaughter, were found to harbour 33.5% as many adult trichostrongyle worms. Results, particularly in the second year, confirmed that significantly fewer worms of most of the important abomasal and small intestinal nematode species which infest lambs in New Zealand (i.e. Haemonchus contortus, Ostertagia circumcincta, Cooperia curticei, Nematodirus spathiger, T. colubriformis, and Trichostrongylus vitrinus) had established in the LFEC genotypes than in their HFEC counterparts. In addition, in utero egg counts of female intestinal Trichostrongylus spp. were significantly lower in LFEC lambs than in their HFEC counterparts, indicating a reduction in fecundity of those worms which did establish. There was also some evidence of an effect of host response on the developmental composition of burdens in the case of some worm species. In relation to host responses, numbers of globule leucocytes/mucosal mast cells in the intestinal mucosa were significantly higher ( P < 0.01) in LFEC lambs than in HFEC lambs in both years of the study. Numbers of connective tissue type mast cells and eosinophils in the intestinal mucosa were also significantly higher in LFEC lambs but only in the second year of the study ( P < 0.01 and P < 0.05, respectively). Numbers of circulating eosinophils did not differ significantly between the genotypes. T. colubriformis-specific antibodies, IgG 1 and IgM to both L 3 and adult worm antigens were all significantly higher ( P < 0.01 or P < 0.05) in LFEC lambs than in HFEC lambs.