Bovine collagen comes mainly from cattle hide and is mostly type I with some type III; marine collagen comes from fish and is mainly type I. Both share collagen’s repeating amino acid pattern, and peptide size is set mainly by how far the protein is cut, not by the animal. What a buyer must act on is the allergen: fish is one of the nine major food allergens a US label must name.
Marine collagen is marketed mostly on cosmetic promises. This site makes no cosmetic claim for collagen from any animal, and the evidence behind that decision is set out in does collagen actually work. What is left is a comparison of two raw materials, which is a narrower question and an answerable one: what each is made from, how it is cut, what reaches the blood, and what a buyer with a fish allergy needs printed on the tub.
What is the difference between bovine and marine collagen?
The animal, the tissue and the collagen types. Bovine collagen is made mainly from cattle hide, which is mostly type I with a minority of type III: about 13% type III in one analysis of insoluble cattle hide collagen (Scott and Veis, Connect Tissue Res 1976). Marine collagen is made from fish scales, bones and the outer layer left after filleting, and is mainly type I. Both are then cut by enzymes into peptides of roughly 3,000 to 6,000 daltons (a 2019 review in Molecules).
The fish typing comes from a 2000 allergy study that prepared its fish gelatins from that same outer layer and identified them as type I collagen. The type numbers name tissues rather than grades: no study cited here links the share of type III in a powder to anything measured in a person. Where each type sits in the body is set out in collagen types explained.
| Property | Bovine (cattle hide) | Marine (fish) |
|---|---|---|
| Raw material | Cattle hide | Scales, bones and the outer layer left after filleting |
| Collagen types | Mostly I, some III (about 13% III in one analysis) | Mainly I |
| Amino acid pattern | Glycine every third residue; proline and hydroxyproline high | The same pattern |
| As gelatin | Melts and sets below body temperature | Melts and sets across a much wider range, by species |
| What sets peptide size | Hydrolysis: how far it is cut | Hydrolysis: how far it is cut |
| Blood after 10 g (one study, n = 6) | Comparable unbound hydroxyproline | Comparable unbound hydroxyproline; 3 of 5 measured fragments higher |
| Major food allergen on a US label | No | Yes, with the species named |
| Supplements in the 8 tendon trials of the 2026 systematic review | Listed as hydrolysed collagen or collagen peptides; animal not recorded | None listed as marine |
Two rows in that table change what a buyer should do: the allergen row, and for anyone comparing labels, the peptide-size row. The others describe raw materials that end up, after hydrolysis, as very similar powders.
Is marine collagen better than bovine collagen?
Not on any measure that has been compared. The one study we found that put fish, pig and cattle collagen side by side in the same people gave six volunteers 10 g of each, a week apart, and unbound hydroxyproline in their blood rose 6- to 10-fold with no significant difference between sources (Virgilio and colleagues, Front Nutr 2024). The company that made all four powders funded it. We found no trial comparing the two on any tendon, bone or other tissue measurement.
The detail underneath that finding is more mixed than its summary, and both belong on the page. Single fragments did differ. Hydroxyproline-glycine, a two-amino-acid fragment, came out 2.3-fold higher after the fish product than after the cattle product cut to the same size, and two other fragments ran 1.7- and 1.9-fold higher; proline-hydroxyproline, the most abundant fragment in the blood, was 1.4-fold higher after the pig product than the cattle one. The authors still concluded that, despite those differences, absorption of the three sources is highly comparable.
- Six people. Three women and three men, average age 25, one 10 g dose of each powder with a week between doses.
- Blood, not tissue. The study measured what reached the plasma over six hours. Whether a fragment that runs 2.3-fold higher in blood does anything different in a tendon was not examined, and the authors write that larger studies are needed.
- One manufacturer, studying its own range. All four powders came from the company that funded the work, and five of the nine authors were its employees. That does not make the figures wrong. It makes them one company’s comparison of its own products.
Is marine collagen absorbed better because its peptides are smaller?
Peptide size is a processing decision, not a property of fish. In the 2024 comparison the same manufacturer cut fish, pig and cattle collagen to about 2,000 daltons and also cut cattle hide to 5,000. The two cattle products gave highly comparable fragment levels in blood, although total hydroxyproline ran 1.2-fold higher for the smaller one (Virgilio and colleagues, Front Nutr 2024). A small molecular-weight figure on a marine label describes its hydrolysis, which a cattle product can match.
For scale, intact collagen is a molecule of about 300,000 daltons and hydrolysed collagen is usually 3,000 to 6,000, according to the 2019 review in Molecules — a cut of fifty-fold or more, made with enzymes and conditions the manufacturer chooses. Next to that, the animal is the smaller variable. What a molecular-weight figure on a label can and cannot tell a buyer is its own question, answered in bioactive collagen peptides vs collagen peptides.
Can you take marine collagen if you are allergic to fish?
That is a question for an allergist, and the label exists to make it answerable. Fish is one of the nine major food allergens, and FDA guidance says the labelling rules cover “all packaged foods, including dietary supplements” and require the fish species to be named. In one study of children with fish allergy, 3 of 10 carried IgE antibodies to fish gelatin, which is type I collagen (Sakaguchi and colleagues, J Allergy Clin Immunol 2000).
Two further findings belong beside that one. The same study found little cross-reactivity between fish and cattle gelatin: antibodies to one mostly did not recognise the other. And collagen is not a trace fish allergen that heat strips out. It is recognised as the second fish allergen after parvalbumin, and in a 2016 study, heating fish meat at 100 degrees Celsius for 320 minutes, long enough to break its collagen down, left the patients’ antibody reactivity to the extracts unchanged (Kobayashi and colleagues, Allergol Int 2016).
The label has to be right for any of that to help. A 2024 laboratory analysis of collagen supplements, published in Marine Drugs, found the collagen content adequate and reported that “some discrepancies were found regarding the labeling and origin of the collagen, with possible safety implications.” A buyer avoiding fish is relying on the origin line being true, which is one more reason to read it first rather than last.
Do bovine and marine collagen have the same amino acids?
Largely, yes. The major collagens, types I, II and III, are built on a three-residue repeat in which every third amino acid is glycine, and they are held together by a high content of proline and hydroxyproline (Krane, Amino Acids 2008). Fish collagens follow the same pattern but vary more from species to species: their gelatins melt and set across a much wider range of temperatures than cattle or pig gelatin (Boran and Regenstein, Adv Food Nutr Res 2010).
For a buyer, that variation is a kitchen property rather than a nutritional one. A gelatin that sets at a lower temperature behaves differently in a dessert; once the protein has been cut into peptides and stirred into a drink, nothing in the studies cited here shows the species changing what it does. Either powder is also a poor protein for the job most protein powders are bought for, which is why collagen against whey ends by recommending whey.
Where each one comes from
Gelatin is made mostly from pig hide and from cattle hides and bones, and fish has drawn manufacturers partly because it meets religious dietary concerns, shared by Jewish and Muslim buyers, that mammalian gelatin can raise (Boran and Regenstein, 2010). The peptides in the 2024 absorption study were made by cutting gelatin further with enzymes, and its authors describe cattle-hide collagen as the predominant product in much of the world, particularly the West.
Two practical points follow from the raw material. Fish species differ more than cattle do, so “marine” on a label is a family name rather than a specification; the species line, which an allergen declaration has to carry anyway, says more. And a cattle source is not a quality claim on its own: hide is hide, and the words that sit on top of it, grass-fed included, describe the animal rather than the powder.
What the research does not show
- No head-to-head on any tissue. We found no trial comparing a fish collagen with a cattle collagen on a tendon, bone or other tissue measurement. The absorption comparison above is the closest thing, and it measured blood for six hours in six people.
- No tendon evidence for fish specifically. The 2026 systematic review of collagen and tendon in J Funct Morphol Kinesiol covered eight randomised trials and 257 people, 246 of them men, and lists every supplement as hydrolysed collagen or collagen peptides; it records no animal source, and none is described as marine. That literature is contested in its own right. The trial most dose advice traces to, Shaw and colleagues, Am J Clin Nutr 2017, gave gelatin with vitamin C to eight healthy men and measured a blood marker, not a tendon. Balshaw and colleagues, Med Sci Sports Exerc 2023, randomised 39 men to 15 g of collagen peptides or placebo a day for 15 weeks of training, imaged the tendon by MRI, and found no between-group difference.
- No outcome tied to species. The differences in amino acid balance and gel behaviour between fish species are real chemistry. None has been linked to anything measured in a person.
- One funder. The only side-by-side absorption study was funded and co-written by the manufacturer of every powder in it.
Where Under Load sits
Under Load is bovine collagen peptides, Type I and III — one ingredient, unflavoured, 350 g. The cattle are grass-fed and the source country is Brazil, both confirmed in writing by the supplier; grass-fed hide is the category standard, so it is stated here as a description, not an advantage. The certificate behind the tub was run by ViaCore, the manufacturer’s own laboratory, on lot 121125450, and what it does and does not cover is transcribed on Testing & Certification.
More in this cluster: collagen peptides vs gelatin, collagen vs bone broth, how to choose a collagen supplement, the collagen research, summarised, the supplement label glossary and every supplement form compared.
Sources
- Scott PG, Veis A. Connect Tissue Res 1976;4(2):117-129. PubMed 131673
- Leon-Lopez A, Morales-Penaloza A, Martinez-Juarez VM, et al. Molecules 2019;24(22):4031. PubMed 31703345
- Sakaguchi M, Toda M, Ebihara T, et al. J Allergy Clin Immunol 2000;106(3):579-584. PubMed 10984381
- Krane SM. Amino Acids 2008;35(4):703-710. PubMed 18431533
- Virgilio N, et al. Front Nutr 2024;11:1416643. PubMed 39149544
- US Food and Drug Administration. Guidance for Industry: Questions and Answers Regarding Food Allergen Labeling (Edition 5). January 2025. fda.gov
- Kobayashi Y, Kuriyama T, Nakagawara R, et al. Allergol Int 2016;65(4):450-458. PubMed 27236375
- Krzisnik N, Kurent E, Roskar R. Mar Drugs 2024;22(10):435. PubMed 39452843
- Boran G, Regenstein JM. Adv Food Nutr Res 2010;60:119-143. PubMed 20691955
- Buchalski A, Jeanfavre M, Altorelli C, Leff G. J Funct Morphol Kinesiol 2026;11(1):130. PubMed 41900537
- Balshaw TG, Funnell MP, McDermott EJ, et al. Med Sci Sports Exerc 2023;55(11):2083-2095. PubMed 37436929
- Shaw G, Lee-Barthel A, Ross ML, et al. Am J Clin Nutr 2017;105(1):136-143. PubMed 27852613