The curious tale of the Dimetrodon sail

The predatory pelycosaur Dimetrodon has always been a favorite of mine. Though not a dinosaur it has an appearance as bizarre as any dinosaur you care to name, and the function of the huge sail on its back is remains an enigma. What could such an impressive structure be used for?

Slight differences in the bony back struts of a similar animal that inhabited the same habitat as Dimetrodon provided paleontologist E.D. Cope with a clue. Where Dimetrodon was a large apex predator "Naosaurus" appeared to be a more peaceful creature. Today we know that "Naosaurus" was a chimera, the head of a Dimetrodon attached to the body of the sail-backed herbivore Edaphosaurus, but this was not initially apparent. The bodies of the two animals were similar but since "Naosaurus" lacked a skull it was difficult to be sure what its habits were. In 1908 paleontologist E.C. Case speculated on the habits of "Naosaurus" (attributing to it a more Edaphosaurus-like skull) in Popular Science;

In a recent restoration of Naosaurus it has been given the skull of the fiercely carnivorous Dimetrodon, the general similarity of the forms seemed to warrant this, but recent discoveries have made it probable that Naosaurus was not an eater of flesh, but a peaceful, sluggish eater of shell fish and perhaps of vegetation. This animal has perhaps the most wonderful dentition of any known animal the incisor teeth are sharp and chisel-shaped, such as might be useful in cutting strong vegetation; behind these are five sharp triangular cutting teeth, not unlike the sectorial teeth of such flesh eaters as the tiger and lion; behind these simple cones, such as would be useful in holding a struggling victim.

This would make Dimetrodon the dominating carnivore of the time and "Naosaurus" an omnivore, but this only made the puzzle of the sails more complex. Why should animals with different habits share the same structure? Multiple explanations had been proposed. The most abused notion was Cope's suggestion that the term "sail" was especially apt, i.e. that these animals were using these structures just like a boat sail.

In 1886 Cope wrote a short note on "The Long-Spined Theromorpha of the Permian Epoch" for the American Naturalist. Of the sail on Dimetrodon Cope wrote;

The utility [of the sail] is difficult to imagine. Unless the animal had aquatic habits and swam on its back, the crest or fin must have been in the way of active movements.

The sail of "Naosaurus", however, had an important difference;

There the spines are not quite so elevated as in the [Dimetrodon], but they are more robust, and have transverse processes or branches which resemble the yardarms of a ship's mast. ... The animal must have presented an extraordinary appearance. Perhaps its dorsal armature resembled the branches of shrubs then, as they do now, and served to conceal them in a brushy or wooded region. Or, more probably, the yardarms were connected by membrane with the neural spine or mast, thus serving the animal as a sail with which he navigated the waters of the Permian lakes.

This remained a popular notion (it was featured in a 1897 article on "Strange Creatures of the Past" in The Century, among some other ideas) but it seems to have been a target of more ridicule than careful consideration. The idea that the sails provided some kind of camouflage was more reasonable, but it seemed strange that a billboard should serve as a feature that would conceal already-large animals.

Cope's hypotheses had failed, but perhaps there was another explanation. Case wondered if the sails could be the result of a once-useful structure outgrowing its usefulness, the product of some kind of "evolutionary inertia";

Such a development seems to have occurred once and again in the history of the world, and the most bizarre types of life owe something of their condition, at least, to this principle. The late Professor Beecher, of Yale, has shown that there is a decided tendency, both in plants and in animals, for a species that is nearing the point of its extinction to develop a spiny or horny habit, covering itself with all sorts of excrescences, seemingly in an unregulated effort to find some condition which will prolong its existence. It is certain that these reptiles, dominant as they were, were rapidly completing their allotted span and as the end approached the spines grew ever heavier and heavier, until it seems plausible to suggest that they became at least a great drain on the animal's powers of nutrition and hastened in no slight degree the end.

It may be unusual to read this sort of language used in considerations of evolution, but Case was working at a time when non-Darwinian evolutionary mechanisms were popular in paleontology. (See yesterday's post about orthogenesis and horses.) It was not strange to think that not only might evolution be directional, but lineages might have a kind of natural timer that determined "their allotted span." In this way lineages were treated like individuals, having a "life" from birth (origin of a new group) to death (extinction). It was as if Dimetrodon and its kind were desperately trying to find some new adaptation to stave off the end.

(Another example of this idea was the saber-toothed cat Smilodon, which some researchers thought became extinct due to overdevelopment of their canines that prevented them from closing their mouths. I can recall hearing this idea in a children's documentary on ancient mammals as late as the 1980's.)

These sorts of notions were cast out when the modern evolutionary synthesis coalesced in the mid-20th century, but the sails of Dimetrodon, Edaphosaurus, and similarly-endowed creatures from other groups still puzzle paleontologists. Thermoregulation and sexual selection are most often considered to have played a role in the origin of these structures but since we cannot directly observe the physiology or behavior of these animals it is difficult to be sure. It is not a mystery that should be shelved, however. Ornate sails have independently evolved so many times in the course of vertebrate evolution that questions about their origins and functions are still worthy of consideration.