Donkey Milk Study: Soy and Flaxseed Oil in Feed (Ren 2026)

28 donkeys received soybean and linseed oil in four mixing ratios for four weeks. 1167 lipids and 406 volatile compounds were identified in the milk; 64 lipids and 89 volatile compounds differed between the groups.

The study in brief

• Original title: Impact of dietary soybean and linseed oils on the lipid and volatile compound profiles of donkey milk
• Authors: Ren W., Sun L., Du X. et al.
• Journal: Food Science of Animal Resources, 46(1), 27 (2026)
• Country: China · Study type: Animal study

How it was investigated

28 lactating three-year-old donkeys from a farm in Shandong Province, China, were randomly assigned to four groups of seven animals each. The ration consisted of 80% wheat straw plus corn, soybean, mineral and vitamin premix; 1% of the ration was vegetable oil, mixed with soybean and linseed oil in ratios of 4:1 (M1), 3:2 (M2), 2:3 (M3), and 1:4 (M4). They were fed twice daily for four weeks in a semi-open housing system.

Key figures

• 28 – lactating donkeys in four groups of 7 animals
• 4 weeks – feeding period, oil content 1% of the ration in each case
• 7.43–11.18% – α-linolenic acid (C18:3n-3) in the feed fat from M1 to M4
• 1167 – lipids identified in the milk from 20 subclasses
• 638 – of which triglycerides (according to the text 55.09%)
• 64 – lipids with a significant difference between the groups

Results

Lactose, dry matter and fat-free dry matter were significantly lower in group M3 (soybean to linseed oil 2:3) than in the other groups (lactose 6.15% compared to 6.57–6.80%), while milk urea (15.20 compared to 12.68–13.22) and polyunsaturated fatty acids (0.30% compared to 0.20–0.25%) were higher. Fat (0.12–0.19%) and protein (1.28–1.52%) did not differ significantly.

1167 lipids from 20 subclasses were identified in the milk, including 638 triglycerides, 129 diglycerides, 93 phosphatidylcholines and 82 phosphatidylethanolamines. In terms of relative quantity, triglycerides made up around 94% in all groups (93.88–94.48%, without a significant difference).

64 lipids differed between the groups, predominantly triglycerides (47). In the OPLS-DA model, M1 separated itself from the other groups above all else. Among the lipid subclasses, only diacylhexosyl lipids (AcHex) differed significantly, with the highest value in M3 (0.089% compared to 0.046% in M1).

Limitations of the study

• Small groups, short period: Seven animals from one farm per group were sampled once after four weeks; baseline values and a group without oil addition are missing. The oil content was only 0.2–0.8% per oil type, and the fatty acids of the feed differed only partially stepwise between the groups.
• Text contradicts tables: The discussion describes a significant increase in lactose and dry matter in M4 and unchanged urea; according to Table 2, lactose and dry matter were lower in M3 and urea higher in M3 (p = 0.044), while M4 did not differ from M1 and M2. Likewise, the text states AcHex higher in M1 and M2, Table 3 shows the highest value in M3.
• Milk fatty acids not shown: The abstract mentions an increase in C18:3 with increasing linseed oil content. However, individual fatty acids are only reported for the feed; for the milk, only the sum of polyunsaturated fatty acids is available, and this did not increase stepwise (M3 0.30%, M4 0.25%).
• Counts and units: The broken-down classes of the 64 different lipids result in 63, and for the 89 volatile compounds 87. The stated triglyceride content of 55.09% does not exactly match 638 out of 1167 lipids (54.67%). Units are missing for urea and freezing point depression.
• Evaluation not derived: The conclusion that the 2:3 ratio improves lipid quality the most is not justified based on defined criteria; at the same time, M3 had the lowest lactose and dry matter values. Flavor notes are based on database assignments, not on a tasting. The study uses the same ethics approval number as other works by the group with a different experimental design.

Level of evidence: Animal study – describes measured values in the studied animals and their milk, no statement on effects in humans.

Our assessment

The study shows that even small oil contents in the feed of donkeys can reflect in the lipid and aroma profiles of the milk. Due to small groups, the lack of a control group and several contradictions between the text and tables, the results should be understood as indications; statements about taste, effect or individual commercial products cannot be derived from this.

Source: Ren W., Sun L., Du X., Meng L., Khan M.Z., Wang C., Li M. (2026). Impact of dietary soybean and linseed oils on the lipid and volatile compound profiles of donkey milk. Food Science of Animal Resources. DOI: 10.1007/s44463-025-00009-w · Summarized by the DON DONKEY'S editorial team, as of October 2026.

End of the study summary. The study was conducted independently by DON DONKEY'S® and does not refer to our products. Our donkey milk is a food.