We tested the hypothesis that Haemonchus contortus and Trichostrongylus colubriformis would adapt to long-term exposure to sheep that were either genetically resistant or susceptible to H. contortus. Sheep genotypes were from lines with 10 years prior selection for low (resistant, R) or high (susceptible, S) faecal worm egg count (WEC) following H. contortus infection. Long-term exposure of H. contortus and T.colubriformis to R or S genotypes was achieved using serial passage for up to 30 nematode generations. Thus, we generated four nematode strains; one strain of each species solely exposed to R sheep and one strain of each species solely exposed to S sheep. Considerable host genotype differences in mean WEC during serial passage confirmed adequate nematode selection pressure for both H. contortus (R 4900 eggs per gram (epg), S 19,900 epg) and T. colubriformis (R 5300 epg, S 13,500 epg). Adaptation of nematode strain to host genotype was tested using seven cross-classified tests for H. contortus, and two cross-classified and one outbred genotype test for T. colubriformis. In the cross-classified design, where each strain infects groups of R, S or randomly bred control sheep, parasite adaptation would be indicated by a significant host genotype by nematode strain interaction for traits indicating parasite reproductive success; specifically WEC and, for H. contortus strains, packed cell volume. We found no significant evidence of parasite adaptation to host genotype (P>0.05) for either the H. contortus or T. colubriformis strains. Therefore, we argue that nematodes will not adapt quickly to sheep bred for nematode resistance, where selection is based on low WEC, although selecting sheep using a subset of immune functions may increase adaptation risk. Our results support the hypothesis that nematode resistance is determined by many genes each with relatively small effect. In conclusion, selection of sheep for nematode resistance using WEC should be sustainable in the medium to long-term.