Did dinosaurs drink seawater?
Seawater is too salty for most animals to drink. The ones that manage it have a special gland that gets rid of the salt — and whether dinosaurs had one is an argument that is still going on, between named people, in print.
Real photograph · 228988681/Shutterstock.com · Shutterstock Standard (free image collection, API) License
What one paper reports
Wang and colleagues, writing in 2018, described a fossil bird with a hollow above the eye where a salt gland would sit, and then looked for the same hollow in other animals.
“We find no evidence of avian salt gland fossae in phylogenetically earlier stem birds or non-avialan dinosaurs, even in those argued to be predominantly marine or desert dwelling.”
Xia Wang, Jiandong Huang, Yuanchao Hu and others (2018). The earliest evidence for a supraorbital salt gland in dinosaurs in new Early Cretaceous ornithurines. Scientific Reports. doi:10.1038/s41598-018-22412-8. Licensed CC BY 4.0. Quotation unchanged.
Unpacked: they looked for the mark a salt gland leaves on the skull, in dinosaurs that were not birds, and did not find it — not even in the ones other researchers had argued lived by the sea or in deserts, where you would most expect it.
Read that sentence carefully, because it is careful. It says no evidence was found. It does not say dinosaurs could not drink seawater, and neither does this page. Not finding the mark of a gland is not the same as the gland never having been there, and it is a long way from an answer about drinking.
And a later paper disagrees
In 2026 a different group — Cau, Gostling, Lacerda, Falasca, Paterna, no author in common with the first — published this in Historical Biology:
“Avian-like salt glands in Spinosauridae”
Andrea Cau, Neil J. Gostling, Mauro B. S. Lacerda, Mario Falasca, Alessandro Paterna · Historical Biology, 2026 · doi:10.1080/08912963.2026.2669954
This site cannot show you a single word of it, and you should know why. The 2018 paper is published under a licence that lets anyone reproduce it, so its sentence is above. This one is not. Its title is a fact anyone can look up; its words belong to its publisher.
That is an uncomfortable thing to put on a page for children, and leaving it out would be worse. If this page quoted only what it is allowed to quote and said nothing about the rest, it would look like one settled answer — and the reason it looked that way would be a licence, not the science.
So what is the answer?
There is not one yet, and the honest version of that is more interesting than a guess. Two groups of researchers, with no one in common, have looked at the same question and reached different conclusions, eight years apart. That is not a failure of science. It is what science looks like from the inside while it is still happening.
Notice also how far the question still is from the title of this page. Even the paper that can be quoted here is about a hollow in a skull, not about drinking. Whether any particular dinosaur ever drank from the sea is a question neither paper asks, and this site is not going to answer it on their behalf.
What is not in doubt is that every one of them needed water. Whether it is the same water you drink has its own page, and the page on Spinosaurus refuses a related question for a related reason.
Where this comes from
- The papers
- Xia Wang, Jiandong Huang, Yuanchao Hu and others (2018). The earliest evidence for a supraorbital salt gland in dinosaurs in new Early Cretaceous ornithurines. Scientific Reports. doi:10.1038/s41598-018-22412-8. Licensed CC BY 4.0. Quotation unchanged. The quotation is checked against the article’s own published XML every time this site is built, and that file is pinned by its checksum. The second paper is named from its publisher’s own bibliographic record and is not reproduced.
- How sure is this?
- Not sure, and deliberately so. This is the first page on this site built on peer-reviewed papers rather than on a public dataset, and a paper is a claim by named people rather than a settled fact — which is why every sentence here says whose claim it is and when they made it. Nothing on this page rests on a reconstruction; the uncertainty is in the disagreement itself, and it is an estimate of nothing because neither paper offers a number.