A2 milk is cow's milk whose beta-casein, one of its main proteins, is only the A2 type instead of a mix of A1 and A2. Small human trials suggest some people report milder digestive symptoms on it, but the evidence is limited, in several cases linked to the A2 milk industry, and does not settle any wider health question.
This article is not medical advice. If you have a digestive condition, a milk allergy or a diagnosis of lactose intolerance, follow your doctor's advice rather than anything on a carton or in an article.
At a glance
- A2 milk is cow's milk whose beta-casein is only the A2 type; A1 and A2 differ by one amino acid at position 67, proline in A2 and histidine in A1.
- Digesting A1 can release beta-casomorphin-7; EFSA's 2009 review found no established link between it and any disease, and no study measuring how much forms during real digestion.
- In a 2023 study of 1,000 Indian milk samples, the A2 gene made up about 91 per cent of copies in seven indigenous breeds against 31 per cent in Jersey and Holstein-Friesian cows; Badri was not tested.
- Small trials of 16 to 45 people found milder digestive symptoms on A2 milk on some measures and mixed results on others; a 2019 systematic review rated that evidence moderate and everything else low.
- A2 milk still contains lactose, and the distinction barely reaches ghee, which is at least 99.5 per cent fat; FSSAI's 2024 advisory against A1 and A2 claims was withdrawn five days later.
What is the difference between A1 and A2 milk?
The difference is a single building block in a single protein. Beta-casein makes up roughly 30 per cent of the protein in cow's milk. In the A2 form, position 67 of the protein chain carries the amino acid proline. In the A1 form, it carries histidine. Like any cow's milk, A2 milk still contains lactose.
That one swap changes how the protein can be cut up during digestion. EFSA's 2009 review explains that when proline sits at position 67, the enzymes that would cut the chain at that point do so very slowly or not at all. In the A1 form they can, releasing a short peptide called beta-casomorphin-7, or BCM-7. BCM-7 belongs to a family of peptides that can act on the body's opioid receptors, and it has been suggested as a contributor to several diseases.
Three points from the same review keep this in proportion:
- Fresh raw milk from healthy cows does not appear to contain BCM-7. The question is what forms during digestion.
- No study the review found reported how much BCM-7 forms during real, in vivo digestion of dairy products.
- Compared with medicinal opioids, BCM-7 from cow's milk does not seem to be a very potent one, and in most if not all animal studies its opioid effects were seen only after injection, not after eating.
Which cow breeds give A2 milk?
It depends on the breed, and on the individual cow, because the variant is inherited. In a 2023 study of 1,000 milk samples from Indian cattle, the A2 version of the gene made up about 91 per cent of copies in seven indigenous breeds, against about 31 per cent in Jersey and Holstein-Friesian cows.
The indigenous breeds were Kosali, Tharparkar, Gangatiri, Sahiwal, Gir, Khariar and Motu. None of the indigenous cows carried two A1 copies, but some carried one A1 and one A2 copy. Gir had the highest share of A2-only samples, at 93 per cent; Gangatiri had the lowest. In the two exotic breeds, only 3 per cent of samples were A2-only. Two of the five authors are affiliated with a company called A2 Milk Research Corporation, and Badri cattle were not among the breeds tested.
Among Western breeds, EFSA's review of decades of breed surveys found Guernsey cattle almost exclusively A2, Jersey cattle relatively low in A1, and Holstein and Friesian herds varying by region, some high in A1 and some high in A2. So breed is a strong hint, not a guarantee. For more on Indian breeds, see our comparison of Badri, Gir and Sahiwal cattle.
What does the research say about A2 milk and digestion?
A handful of small human trials, most in people who already felt unwell after drinking milk, found milder digestive symptoms on A2-only milk on some measures. Others found mixed results or no difference at all. A 2019 systematic review rated the digestive evidence as of moderate certainty, and the 2017 review calls for more research in humans.
Industry links below are given only where the abstract states them.
| Trial | Who took part | What it found | Industry link stated |
|---|---|---|---|
| Ho et al., 2014, Australia | 41 adults; 2 weeks on each milk (pilot crossover) | Looser stools (higher Bristol scale scores) on A1 milk | One author from A2 Dairy Products Australia |
| Jianqin et al., 2016, China | 45 adults with self-reported lactose intolerance; 14 days on each milk | More symptoms, slower gut transit and higher inflammation markers on milk with A1 | One author from The a2 Milk Company |
| Milan et al., 2020, New Zealand | 40 women; single 750 mL servings | Lactose-intolerant women had less nausea and urgency on A2 milk; other intolerant women had the same symptoms whatever the milk | Not stated |
| Choi et al., 2024, Korea | 40 adults with discomfort after milk; 2 weeks on each | Mixed: more bloating and loose stools on A2 milk on one scale, less pain and urgency on another | None disclosed |
| Ramakrishnan et al., 2024, USA | 16 lactose maldigesters; 2 weeks on each | More faecal urgency on milk with A1; day-to-day symptoms did not differ | Authors declare none |
| Greenway et al., 2026, USA | Adults with self-reported milk intolerance (pilot) | No significant differences in symptoms, stools or gut transit | One author funded by The a2 Milk Company to give a talk |
Two systematic reviews pull this together. The first, in 2017, covered 39 studies, many in rodents or the laboratory. In humans it found evidence from a limited number of studies: one clinical study of slower transit and two of looser stools. Three of its four authors disclosed consulting work, an honorarium or grant support from The a2 Milk Company or the A2 Corporation. The second, in 2019, by a team based mainly at the University of Freiburg, graded studies published before October 2017. It rated the evidence for worse digestive effects with A1 than A2 as of moderate certainty, and the evidence for every other health outcome as low or very low.
How strong is the evidence on A2 milk?
Moderate at best, and only for digestive symptoms. The trials are small and short, and most enrolled people who already reported trouble with milk. Several had authors linked to the A2 milk industry, and the most recent pilot trial found no difference. For wider diseases, the evidence is weak.
That last point has a long history. Ecological studies once linked BCM-7 or A1 intake across countries with heart disease and type 1 diabetes. EFSA's 2009 review examined those links and concluded that a cause-and-effect relationship between eating BCM-7 or related peptides and any of the suggested non-communicable diseases could not be established. It noted that ecological studies cannot establish cause or adjust for confounding factors. The review predates every trial in the table above, the earliest of which is from 2014.
The limits, in brief:
- Size. The trials whose abstracts give a number enrolled between 16 and 45 people each.
- Duration. Most tested each milk for about two weeks, and one used single servings.
- Consistency. The Korean trial found some symptoms worse on A2 milk, and the 2026 pilot found no difference.
Is A2 milk the same as lactose-free milk?
No. A2 milk still contains lactose, the milk sugar. In the New Zealand trial, both the A2 milk and the conventional milk contained lactose, and a lactose-free conventional milk was tested separately as a third option. If lactose is what troubles you, an A2 label does not take it out.
In the New Zealand study, women diagnosed as lactose intolerant had some milder symptoms on A2 milk, while women who felt unwell after milk but digested lactose normally had the same rapid symptoms whatever they drank. In the American trial of lactose maldigesters, two weeks of drinking conventional milk did not bring its tolerance up to that of A2 milk, and breath hydrogen, a measure of undigested lactose, did not differ between the two. None of this replaces a diagnosis, which is a job for your doctor. Ghee raises a separate lactose question, covered in is ghee lactose free.
Does the A1 or A2 label matter for ghee?
Much less than for milk. The A1 and A2 difference lies in a protein, and ghee is almost all fat. FSSAI requires ghee to be at least 99.5 per cent milk fat, made by processes that remove almost all the water and milk solids-not-fat. That leaves at most 0.5 per cent for everything else, protein included.
The digestion research above is about milk, not ghee. We make A2 bilona ghee from Badri cattle, and we would rather you knew this than paid for a protein that can make up only a fraction of one per cent of the jar. What genuinely separates one ghee from another, from the animal's diet to the method, is set out in A2 ghee vs regular ghee.
How does India regulate A1 and A2 claims?
The position is unsettled. In August 2024, FSSAI told food businesses to remove A1 and A2 claims from milk and milk products, saying its standards do not recognise the distinction. Five days after that advisory, it withdrew it "for further consultation and engagement with the stakeholders", which let businesses carry on using the claims.
The first advisory was reported by All India Radio's News on AIR; it named milk, ghee, butter and curd sold with these claims. The withdrawal of the 21 August advisory, on 26 August, was reported by PTI. In practice, an A2 claim on a pack is the seller's statement, not a category Indian food standards define.
The short version
A2 milk is cow's milk whose beta-casein is only the A2 type, a difference of one amino acid at position 67. Indian indigenous breeds mostly carry the A2 gene, and many Western herds carry A1, though individual cows vary. Small trials, several with industry links, suggest some people report milder digestive symptoms on A2 milk, while others found mixed results or none, and EFSA found no established link to wider disease. The distinction matters far less in ghee, which is almost pure fat. If milk troubles you, see your doctor and follow their advice.
Sources
Further reading behind the figures above. Himdhara's own numbers — yields, batch results — come from our production records and the batch reports on the lab reports page.
- European Food Safety Authority (2009). Review of the potential health impact of β-casomorphins and related peptides. EFSA Scientific Report 231.
- Brooke-Taylor S, Dwyer K, Woodford K, Kost N (2017). Systematic Review of the Gastrointestinal Effects of A1 Compared with A2 β-Casein. Advances in Nutrition.
- Küllenberg de Gaudry D et al. (2019). Milk A1 β-casein and health-related outcomes in humans: a systematic review. Nutrition Reviews.
- Khan R, De S, Dewangan R, Tamboli R, Gupta R (2023). Potential status of A1 and A2 variants of bovine beta-casein gene in milk samples of Indian cattle breeds. Animal Biotechnology.
- Food Safety and Standards Authority of India. Food Safety and Standards (Food Products Standards and Food Additives) Regulations, 2011: Chapter 2.1, Dairy products and analogues (2.1.8 Standard for Milk Fat Products). FSSAI.