Fossil record shows whales evolved from land to sea


Whales and dolphins are among the most fascinating animals on the planet. That they descended from four-legged land animals makes their story compelling.

That story has become easier to tell given the excellent fossil record accrued in the past 50 years.

Being entirely aquatic it seems odd that whales are mammals. Other mammals such as seals, sea lions and walrusess spend a good part of their lives in the ocean, but they are clearly tetrapods (four-legged). But whales don’t have any leg-like appendages. So what gives?

First let’s be certain about why whales are mammals and not fish, and what unique characteristics they have.

By definition, mammals are warm blooded vertebrates. They give birth to live young that nurse from their mother. Mammals typically have significant body hair, and though adult whales have none, their fetuses do, and some adults have sensory bristles around their mouth.

Unlike fish, whales have thicker, more rounded bodies. They propel themselves through the water with horizontal flukes and their dorsal fins are not supported by bony structures.

Whale skeletons have numerous unique features; several are noteworthy. Some whale species have vestigial pelvises. Amazingly modern whales have the dormant genes required to develop legs. In rare instances these genes will get expressed and a whale develops miniature legs.

The structure of whale vertebrae and the spaces between them facilitate up and down rather than side to side motion. The up-down spinal flexion is characteristic of mammals while the side to side is characteristic of fish.

Whale skulls are different from most mammals in several ways. They have long heads with their lower jaw extending forward while the upper jaw also extends rearward. This moves the nasal opening (blow hole) to the top of the skull, to a more rearward location. It also places the brain and auditory apparatus at the rear of their heads.

The brief descriptions given below demonstrate a transition in body structure from creatures living a primarily terrestrial to ever more aquatic existence. They illustrate a portion of the taxonomic tree of life with many branches. Though some branches were undoubtedly dead ends, others are representative of the lineage of modern whales.

Whale’s closest living relative is the hippopotamus. That is to say, there was an extinct common ancestor of hippos and whales some 60 million years ago (mya). As is the case for all creatures, neither modern whales nor hippos emerged fully formed from their ancient common ancestor. Modern hippos don’t even appear until about 15 mya.

The first creature we can identify with the branch that eventually yields whales dates from about 60 mya and is called sinonyx. The structure of this wolf-sized animal’s skull shows it to have been a fish hunter. As a consequence, its skull shows the beginnings of the elongation that becomes enhanced in later descendants.

Fossils of pakicetus appear in geologic strata dated at about 52 mya. Though still a tetrapod, its elongated skull has many of the features of cetaceans, the order in which whales are classified. Oxygen isotope analysis indicates these animals roamed freshwater shallows, and like crocodiles they fed on fish and land animals coming to drink. The structure of its middle ear indicates it could not dive to great depths.

Ambulocetus fossils from around 50 mya are dramatically more aquatic in appearance. The structure of its hind and fore limbs indicates it waddled about on land somewhat like modern sea lions. Its hide limbs are oriented for forward thrusts to propel it in the water. Its skull is further elongated and has many transitional characteristics showing adaptation to an aquatic life.

Rodhocetus, at about 47 mya, had a spine demonstrating it had musculature to create powerful thrusts with a tail that clearly was more fluke-like. However, its spine and small pelvis indicates it could, with difficulty, still get around on land. Again, its skull has many more transitional characteristics, including nostrils located in a more rearward position.

Basilosaurus is representative of the next phase. Its fossils are dated between 35 and 45 mya. Its pelvis is no longer attached to the spine so that it could not bear weight. Its hind limbs were still external to its body, but they were ineffective, being merely 2 feet long on an animal 50 feet long. As one would expect of a mammal living in a marine environment, its single nostril is further back on the skull.

Dorudon lived at the same time as basilosaurus but was only about 15 feet in length. Its anatomy is more consistent with being in the direct line of descent on modern whales than is basilosaurus. It too had very short hindlimbs barely projecting beyond the body.

Space does not permit a more thorough description. But, in these examples I’ve highlighted a few changes in the skull, the size and position of the hindlimbs, and anatomy of the tail and spine which enabled swimming at the cost of land mobility.

I could elaborate on any number of features of the skull such as those relating to its teeth and hearing. For instance, the skulls of squalodons, which first appear about 33 mya, show the first evidence for echolocation. As with hindlimbs, the forelimbs transitioned from appendages accommodating life on land, to shallow water and eventually exclusively aquatic locomotion. There will undoubtedly be more discoveries. But, looking at what exist today is impressive. For some, there will never be enough. Their cynicism stems more from misunderstandings and preconceived ideas than the reality of the fossil record.

Steve Luckstead is a medical physicist of Walla Walla. Contact him at


DavidDalan 1 year, 4 months ago

The Burke Museum at the UW has a substantial collection of primitive whale fossils, worth a peek to the curious. Many of the fossils were collected right here in WA!


susnalmf 1 year, 4 months ago

I thoroughly enjoyed this article for two reasons. One is the typical explanation provided by Mr. Luckstead that assumes "sameness" means decent. The other was the explanation on the graphic provided in the paper-article that explains that NONE of the animals shown in the graphic descended from each other, but in fact each is it's own branch on the evolutionary tree (or bush).

I have been looking for any intermediate on any tree anywhere in any book for over 30 years now. The only one I have found is some "proto-bird" with a possible picture of what it may look like some day if we ever find it.

And yet...and YET...evolution is a fact? Amazing!


DavidDalan 1 year, 4 months ago

There are some transitional forms, but the lack of abundant transitional forms is kind of a red herring. They are not needed.

To demonstrate change in organisms over time, all you need are fossils that provide snapshots of the changes. Horse evolution is a great example of this. While there are gaps in the fossil record, there are many fossils that cover the ground from Eohippus to modern horses. The examples we do have provide enough data points to clearly show a trend in the morphology of horses, even if some data points are missing.


namvet60 1 year, 4 months ago

Curious - how far back does the data points indicate the horse morphology began? Thanks


DavidDalan 1 year, 4 months ago

As far as I know about 50M years.

This pretty much covers it. If you look at the diagram of tooth, hoof and body changes, the oldest animal is hyracotherium which appears to be one of the names ascribed to Eohippus.

and here are the Whales...


susnalmf 1 year, 4 months ago

I have seen the horse sequence. It is about the only sequence ever used in any biology book. Sometimes they use the "elephant sequence". However, these sequences come with the same proviso - there is NO DATA that any of these came from each other. They too are separate branches on a supposed tree with no known common ancestor.

There is no reason to hide this. The fossil record shows that each animal in the "horse sequence" shows up suddenly, remains unchanged for millions of years, and then dies off suddenly. Their supposed relationship is purely an assumption.



DavidDalan 1 year, 4 months ago

Nope. Genome sequencing has shown horse lineage (based on genes, not just fossils -- read the horse wiki I linked above) showing common ancestry back to 4.5M years ago. You can sequence genes of living animals and gather a lot of information about their relationships to other branches in their family tree. No need to depend solely on morphological analysis of fossils.


susnalmf 1 year, 4 months ago

Genome sequencing does not save the fossil record as many hoped it would. Changing the topic to genes will only complicate some simple facts that I believe we can both agree on.

The fossil record is a beautiful study. Although evolutionary theory expected to find much "change" in the fossils that we find within the history of life, we do NOT find change. We find "stasis" (quote from Stephen J. Gould - Wonderful Life).

We definitely find variety! But, every fossil ever found in the fossil record "shows up suddenly, remains static throughout it's history, and dies off suddenly" (another quote from Gould). There are no exceptions. It is the "remains static throughout its history" that is the surprise.

Thus...there has been no example ever in the fossil record that records a "changing" animal from one form into another. The only "change" we see - is what most of us would call "variety", not change.

And...thus, we find that there exists no honest "evolutionary tree" that ever shows any animal as intermediate to any other animal. Just a whole variety of animals on separate branches - tied to a possible ancestor that has yet to found. And this is a truth that cannot be denied.



DavidDalan 1 year, 4 months ago

Again, nope.

You seem to misunderstand Dr. Gould. He was actually an early proponent of Punctuated Equillibrium (he may have originated this idea). Make no mistake, Gould knew life changed over time. Rather he held there had to be a change in conditions for a species to express new mutations beneficially. Species do not differentiate at an even rate over time, the change is "punctuated" in the midst of otherwise "equilibrium" conditions.

Some life forms changer very little (crocodilians, Sturgeon, Latimeria, etc.) and some change a lot such as birds. Some changes are physically dramatic (whales are a good example) and some are not even reflected in the skeleton, but rather are soft tissue or behavioral.

Punctuated Equilibrium and Cladistics are a worthy read on Wikipedia. I pondered detailing Gould's ideas, but I suspect the Wiki does a good enough job.

As an aside, Gould did indeed reject the old "tree of life." In fact "the tree" is a VERY antiquated idea in evolutionary biology.

When I was in college, (Geology, 98 WSU), Darwin and the tree of life were historical references. Ideological frameworks to explain how the modern ideas of evolutionary biology arose. Our roots as it were. Not current thinking then, and that was 16 years ago.

Our understanding of evolutionary biology has changed much in the last 100 years, and will change again.

Biological Evolution is scientific fact. There is no argument in academia about it, and I have found (without fail) those that claim it is arguable either 1. have a poor understanding of the topic or 2. have distinctly non-scientific motivations for questioning its validity.


susnalmf 1 year, 4 months ago

I think if you read my last post carefully, you will agree that everything I stated about the fossil record is absolutely true.

Gould and Eldridge's Punctuated Equilibrium Theory was an attempt to match evolutionary theory (neo-Darwinian Synthesis) with the fossil record. His theory was developed as he re-classified Walcott's poor classification of the Burgess Shale (read "Wonderful Life by Gould - a very good book).

Punctuated Equilibrium does explain the fossil record well. Unfortunately, it throws out the mechanism of evolution (small changes over time eventuate into big changes over long periods of time). And thus, there was quite a battle throughout the 80's, with Gould (the popularizer of punk eek), and Dawkins (the popularizer of Neo-Darwin-synthesis).

Could be that they are both wrong. That is my bet.

I have read several of Gould's books, and several of Dawkins books. But the best book I have yet read on the issue was Stephan Mayer's "Darwin's Doubt".

There is no doubt that there has been variety and change throughout the history of life. But only the strict materialists seem to be stuck on mantra "evolution is a fact".

The rest of us see the obvious problems with the theory. So much so, that I would claim that IF a materialistic explanation is ever found for the history of life, evolution will play a very small roll.


DavidDalan 1 year, 4 months ago

I have to admit I love to argue. But I think we are getting nowhere. But futility never stopped me before....

First a little house cleaning. Stephen Mayer should not be discussed along with Gould, Dawkins or any other actual scientist. Mayer was published exactly one time in a peer reviewed journal (an accident), and it was rescinded because there is no valid science supporting Intelligent Design.

Mayer is a pseudo-scientist. He has a conclusion, is not open to the possibility of being wrong, and interprets all data in a manner to support his preconceived conclusion. That is the definition of Pseudoscience.

While Gould wrote many popular books, I would suggest something like the 2002 book "The Structure of Evolutionary Theory" as a better read on Gould's thoughts on evolution, as opposed to his opinions on baseball.

As an aside, "The Mismeasure of Man" is a good book of his as well. Dry as bark, but very enlightening.

Dawkins and Gould argued the mechanics, processes of evolution. And, as good scientists, asked many challenging questions. Unless one were to take passing quotes grossly out of context, I doubt anyone can make the case that either of the two authors cited doubt/doubted evolution happens. How it happens? You bet.

And in any event, they are but two of many. And even if they both said evolution does not happen, without convincing evidence...those statements would be meaningless. Opinions matter little in science. Actually they do not matter at all.


susnalmf 1 year, 4 months ago

Opinions matter little in science - unless the data is inconclusive and one's philosophy dominates (maybe even blinds) the interpretation. The difficulty arises when opinions are mistaken for the very one's opining.

I have read enough of Gould, Dawkins, George Sims Johnston, Robert Pennock, Francis Crick. They have convinced me that evolution is a theory in crisis.

Because I have also read Denton, Phillip Johnson, Behe, Mayer, Dean Kenyon...

I tried reading the young-earth authors, but never found them convincing in anyway.

The materialist cannot debate the data, and thus they attack the authors. Your definition of "pseudo-science" remarkably defines the materialistic atheist. The hardest part of discussing anything with the materialist is that they are convinced that everyone else is bias - and they are not. Trying to pry into that prideful mind is a daunting task.

Thanks for the debate.


Sign in to comment

Click here to sign in