We determined the seasonal nutrient and secondary compound intake, in vivo diet digestibility (DD) and diet species composition of goats in different physiological states in a desert rangeland of Argentina. During the wet and the dry seasons, 24 mature female Creole goats (six non‐lactating, non‐pregnant and six lactating in each season) were used in a 2 × 2 factorial design. Diet botanical composition, DD and dry‐matter intake (DMI) were estimated by the combined use of visual observation, microhistological analyses of faecal samples and n‐alkane technique. Browse species dominated the goats’ diets (>80%) in both seasons and physiological states, with a twofold increase in the proportion of forbs during the wet season (p < 0.10). Lactating goats showed greater DMI than non‐lactating animals, and they also displayed greater intakes of some tanniferous (Mimosa ephedroides) and high‐protein (Atriplex lampa) shrubs (p < 0.10). Lactating and non‐lactating goats selected diets with adequate crude protein (CP) and fibre content but insufficient energy, although lactating goats tended to increase energy and tannin content in their diets (p < 0.20). Physiological state did not influence DD, but digestibility was greater (p < 0.10) in the wet season, attributed to the greater protein content and lower tannin content of the diet. These results have important implications for managerial interventions on traditional goat production systems in the study area aiming at enhancing goat nutrition and welfare, as well as the sustainability of rangeland resources in the region.

The effect of three different vertical structures of a perennial ryegrass‐dominated sward, defined by pseudostem heights (cm) — 1·3 low (L), 2·5 medium (M) and 3·5 high (H) — on the bite depth and selection of plant parts within the grazed horizon of the sward by sheep and guanacos was studied. The bite depth (cm) was similar between sheep and guanacos across the different swards (L, 1·5; M, 3·6; H, 3·8) and was related to pre‐grazing sward surface height rather than to pseudostem height. There were differences in diet composition between species that were related to differences in selection for plant parts. Sheep had a higher proportion of green leaf in the diet than did guanacos in L (0·84 vs. 0·71, P < 0·05) and M (0·75 vs. 0·59, P < 0·05) swards, and possibly selected this plant part in H swards in which the proportion of green leaf in the grazed horizon was low (0·45–0·50). Guanacos had a higher proportion of dead leaf and sheath in the diet than did sheep in L (0·23 vs. 0·09, P < 0·05) and M (0·30 vs. 0·18, P < 0·05) swards, and possibly selected this plant part on these swards, in which the proportion of green leaf in the grazed horizon was high (0·70–0·95). The proportion of pseudostem in the diet of sheep and guanacos was similar across all swards (0·03) and was generally much lower than that in the grazed horizon (0·01–0·18). Guanacos had a higher proportion of dead stem in the diet than did sheep (0·06 vs. 0·02, P < 0·05) across all swards, but the proportion was similar to that in the grazed horizon of each sward.