Panderichthys is a widely recognized transitional form in tetrapod evolution (you know, one of those transitional fossils we're so often told don't exist). A description of a specimen with a well-preserved pelvic girdle has just been described in Nature, and it tells us some more about the history of tetrapod locomotion.

Panderichthys is an interesting animal—it definitely looks more like a fish than a salamander, but its fins are stout and bony, and other characteristics of its skeleton clearly ally it with the tetrapods. In the shift from an aquatic to a fully terrestrial life, the limbs and their supporting pectoral and pelvic girdles had to undergo major changes. In fish, the pectoral girdles are coupled to the skull, while the pelvic girdles are small and 'floating' in the musculature. To bear the animal's weight, the pectoral girdles lost their connection to the skull, and both became thicker, stronger, and more closely bound to the axial skeleton. The fins themselves had to change from a fan of slender fin-rays to more solid load-bearing digits. In Panderichthys, we see a mixture of these changes in process.

The fossil isn't the prettiest specimen you'll see, but it's virtue is that poorly described parts of the animals are in relatively good shape and have been exposed and described. Panderichthys contained a mix of primitive and derived characters, the kind of intermediate hodge-podge we'd expect in a transitional form.

a, Outline of the body of Panderichthys. Grey shading indicates preserved portions of Panderichthys rhombolepis specimen GIT 434-1. Redrawn from ref. 14. b, Panderichthys rhombolepis specimen GIT 434-1 with head (h) and body (b) outlined. The pelvic girdle and fin are shaded in orange. c, Pelvic girdle and fin. The matrix is distinguished from the fossil by an overlay of grey shading. d, Specimen drawing. F, femur; Fi, fibula; Fre, fibulare; Int, intermedium (proximal end of the); Pel, pelvic girdle; T, tibia. Vertical hatching indicates broken bone; grey shading indicates matrix; circles indicate thin dermal bone covering. e, Reconstruction of the pelvic fin. Thick outline indicates preserved margin, thin outline indicates inferred margin, dotted lines indicate uncertain margin. Solid black scale bars, 10 mm.

In particular, the hindlimb has been found to be significantly more primitive than the forelimb—while the forelimb is large and robust and capable of propping the animal up and pushing it along on the substrate, the hindlimb is relatively small and unimpressive, a glorified and lumpy fish fin. In the figure below, the fore- and hind-limbs of an even more primitive fish, Eusthenopteron, are compared with Panderichthys (in the middle) and Acanthostega. While the Panderichthys limbs have a more robust central axis than the branchy, fan-like Eusthenopteron fins, they aren't as obviously leg-like as those of Acanthostega.

Pectoral (a, c, e) and pelvic fins (b, d, f) of Eusthenopteron (a, b), Panderichthys (c, d) and Acanthostega (e, f) all in ventral view. F, femur; Fi, fibula; Fre, fibulare; H, humerus; Int, intermedium; R, radius; T, tibia; U, ulna; Ure, ulnare. Thick outline indicates preserved margin; thin outline indicates inferred margin; dotted lines indicate uncertain margin. Scale bars, 10 mm.

What's it all mean? In our adaptation to terrestrial life, it suggests it was done forelimbs first. The earliest tetrapod ancestors humped their way along, using the hindlimbs to anchor themselves, arching and extending their backs forward, then hauling themselves further forward with their front legs. Hindlimbs expanded later to contribute more to movement.

Because the paired fin morphology of Panderichthys is defined substantially by a combination of primitive characters shared with osteolepiforms (mostly in the pelvic fin) and derived characters shared with tetrapods (many pectoral fin characteristics) rather than autapomorphies, at least part of the tetrapod stem lineage around the Panderichthys node must have displayed a combination of tetrapod-like pectoral fins with less limb-like pelvic fins. This suggests that the general locomotory pattern of Panderichthys characterized part of the tetrapod stem lineage between osteolepiforms and tetrapods. The evolution of tetrapod locomotion therefore seems to have passed through a 'front-wheel drive' stage powered by body undulation and pelvic fins as anchors, demonstrated by Panderichthys, before shifting to a 'rear-wheel drive' leg-powered walk in the interval between Panderichthys and Acanthostega.

In case you have trouble keeping the names of all those antique tetrapods straight—they aren't quite as popular as dinosaurs, unfortunately—here's an illustration by Carl Dennis Buell. I wish every cladogram were this well done; this is the kind of first rate scientific illustration that Edward Tufte ought to highlight. You can not only see the lineage relationships, but you also see the temporal extent of the known fossils and illustrations of their form, making it easy to see how the transformation of fish into amphibians is represented in the fossil record.

Illustration by Carl Buell, from Carl Zimmer's At the Water's Edge ^{(amzn/b&n/abe/pwll)}

Boisvert C (2005) The pelvic fin and girdle of Panderichthys and the origin of tetrapod locomotion. Nature 438(7071):1145-1148.