University of Washington
Comparative biomechanics lab home
Comparative biomechanics lab
Comparative biomechanics lab Adam Summers Biomechanics Research comparative biomechanics peer reviewed papers Adam Summers writing for popular press comparative biomechanics in the news
Comparative biomechanics lab home
NavLab1a NavPpl1a NavResrch1a NavPub1a NavPopSci1a NavNews1a

amnhhealiz1Squeeze Play

Brobdingnagian earthmoving "worms" dig their tunnels with a hydraulic ram.

Story by Adam Summers - Illustrations by Roberto Osti

When I first saw a live caecilian, I was convinced that I was looking at an earthworm large enough to strike fear in the heart of an Alabama largemouth bass. The animal squirming through the sphagnum moss was Dermophis mexicanus, a Central American species of amphibian that reaches two feet in length and is as fat around as the most decadent Cuban cigar. Like common earthworms, caecilians’ brown-gray bodies sport closely spaced, circumferential grooves; the animals’ blunt heads bear a striking resemblance to their tails, their eyes are quite small, and they lack arms and legs. If you were to grasp one in your hand, it would squirm like a healthy night crawler trying to escape the hook.

But such a scene is about as likely as latching onto a fifty-pound bass. Caecilians so seldom have contact with people that most species have no common name. Although they are amphibians, caecilians are denizens of the terrestrial underworld. (One odd species, the atypically aquatic Typhlonectes natans, can be bought in pet stores, albeit under the misleading name “rubber eel.”) Anyone hoping to find one should bring a shovel to the world’s humid tropics.

As you dig, however, you’ll quickly be reminded that burrowing is tough. The short, stout arm bones of moles and armadillos reflect the extreme demands of tunnel excavation, as do the thick, reinforced skulls of other burrowing vertebrates, such as the caecilians. Those animals have abandoned limbs altogether in favor of slicing through the earth with their narrow bodies.


In this recreation of the work of James O'Reilly and his colleagues, a two-foot-long burrowing amphibian known as a caecilian (Dermophis mexicanus) has moved along a clear plastic tube and encountered soil. To move underground, the caecilian relies on two complementary groups of muscles, in three different ways. One group controls the simple battering action of its vertebral column. The second group is connected to spirals of tendons just under the skin. When the latter muscles contract, the animal becomes thinner; because the caecilian's volume is constant, the now squeezed animal must become longer. By anchoring itself with S-shaped kinks, the animal can apply this lengthening force in a forward direction. At the same time, the tendons (not shown), which are arranged much like the material in a "Chinese" finger trap, push on the skull, providing a third source of force. (The contracted, elongated state of the animal is outlined in red; its diameter, but not its length, is exaggerated here for clarity.)

Like digging, studying the mechanics of burrowing is also tough, because, well, it happens underground. Nevertheless, James C. O’Reilly, a biomechanist at the University of Miami in Florida, has managed the task, and in the process has discovered that caecilians such as D. mexicanus not only look like worms, they move like them.

A caecilian faces one primary constraint as it burrows through the ground: the hardness of the soil. So if you want to understand how fast and through what kinds of soil a caecilian can move, the critical factor to measure is how forcefully the animal can manage to ram the earth. To understand the mechanics of burrowing O’Reilly designed an experiment that took advantage of the species’ poor eyesight. Laboratory animals were fooled into “burrowing” into a clear acrylic tube with a ninety-degree bend. Beyond the bend, a second tube, filled with soil and connected to a sensitive force gauge, was set inside the first. When a caecilian encountered the soil-filled tube, the animal would push against the soil as hard as it could, seeking to escape the alien environment of the artificial burrow. And as hard as it could push, it turns out, was much harder than what O’Reilly had expected.

D. mexicanus burrows by straightening its vertebral column and ramming its head into the dirt. (The action is not unlike pushing a tent peg into the ground.) Large bundles of muscle that can move the vertebral column line both sides of the caecilian’s spine. The muscles obviously contribute to burrowing, but their cross-sectional area can account for only about a quarter of the pushing force. (As regular readers of this column may recall, the potential force a muscle can generate depends directly on its cross-sectional area. A muscle with a cross section of a square centimeter can exert about enough force to hold up a ten-pound weight.) The mismatch between force and cross-sectional area implied either that caecilians possess a different kind of muscle tissue than do other vertebrates, or that the animals possess another source of pushing power.

It turns out that caecilian muscle is much like yours and mine. The extra power comes, somewhat obliquely, from another group of muscles. Just under its skin lies a coiled layer of connective tissue that wraps its insides from head to tail. That tissue in turn surrounds and joins to several thin layers of muscle, laterally lining the animal’s body. When these muscles contract, they don’t directly push the head forward. But the contraction does increase the pressure in the caecilian’s body, which, now thinner, must become longer if its volume is to remain constant.

By anchoring the rear half of its body against the inner walls of the burrow, the animal can direct virtually all the force of the muscular compression toward the head, much like a hydraulic ram. The head shoots forward with the extra force measured during O’Reilly’s experiment.

The mechanism is known as hydrostatic motion. Once extended, the animal, kinking its body near its head against the burrow wall to provide friction, can then draw its tail forward by relaxing the same muscles and bringing up its spine.

The sequence is just like a worm’s squirm. But worms don’t have spinal cords, and caecilians do; the spine has to go somewhere when the animal is short, plump, and at rest. Unlike most vertebrates, caecilians can kink their vertebral column up inside their body, for which they possess a very lax set of connections between the skin and the spine. The spinal nerves, for instance, are set in S-bends at rest, leaving plenty of slack for the short-and-fat, then long-and-thin sequence during locomotion.

Borrowing technology from heart surgeons, O’Reilly and his colleagues, David Carrier of the University of Utah in Salt Lake City and Dale Ritter at Brown University in Providence, Rhode Island, implanted miniature pressure gauges, smaller than a grain of rice, into the body cavities of several caecilians. The pressure peaked, they discovered, at the same time as the forward force did, confirming their hydrostatic-motion hypothesis. Thus what a caecilian does while burrowing is more like driving a steam piston into the ground than pounding a tent stake. Furthermore, when the animal was prevented from sealing its single lung—thus preventing the pressure of the muscles from being transmitted throughout the rest of the body—the caecilian’s burrowing force dropped considerably.

Biomechanists have known for some time that the earthworm (a caecilian’s favorite meal) also advances by pressurizing its body and squeezing its head forward. So there is a certain symmetry to this story: the only known vertebrate to move by hydrostatic locomotion happens to prey on an invertebrate that relies on the same mechanism. What would it feel like to bait a hook with one of these animals, and reel in a fifty-pound largemouth?

University of Washinton Home

Friday Harbor Laboratories
Integrated Center for Marine Biomaterials and Ecomechanics


Popular Science


Biomechanics Columns


Film & Television