Abstract
Mare’s milk is recognized for its nutritional value, but information on the relationships of mare age and number of foalings with body conformation, udder morphology, milk yield, and milk composition in the indigenous Kushum breed remains limited. Forty-one clinically healthy lactating Kushum mares were evaluated in June and classified by age (4–7, 8–11, and 12–16 years) and number of foalings (1–3, 4–7, and 8–11). Body traits were measured once; udder morphometry was available for 30 mares. Milk yield was recorded at 08:00, 10:00, 12:00, 14:00, and 16:00 h on two consecutive days. On Day 2, a 100-mL aliquot was collected after each milking, and the five aliquots were pooled into one 500-mL composite sample per mare; physicochemical composition was available for 40 mares. Age and number-of-foalings effects were evaluated in separate models. Continuous morphometric and physicochemical traits were analyzed by one-way ANOVA with Tukey-adjusted comparisons, with Benjamini–Hochberg false discovery rate (FDR) correction across trait families; two-day milk yield was additionally evaluated with a repeated-measures mixed model. After FDR correction, cannon bone circumference, body weight, udder length, and udder width differed among both age and number-of-foalings groups. Teat shape also differed among groups. The two-day mean daily milk yield did not differ by age (p = 0.483) or number of foalings (p = 0.195), although significant group-by-day interactions were detected. In pooled Day-2 milk, SNF, lactose, glucose, freezing point, acidity, lactic acid, density, and urea differed among both classification factors after FDR correction (q < 0.05), whereas fat, protein, total solids, galactose, citric acid, and casein did not. PCA explained 60.9% of the variance in body/udder traits plus mean milk yield and 68.8% of the variance in physicochemical composition. These findings indicate selective associations of age and number of foalings with Kushum mare phenotype and milk composition, while emphasizing the limitations of small udder subgroups and equal-volume pooled sampling.