From Fins to Fingers

Australia: The Land Where Time Began

From Fins to Fingers

The key innovation of converting fins to fingers was first made in the Devonian, about 370-360 Ma, described as the “fish-tetrapod transition”. The fossil evidence of this transition has increased almost exponentially in recent years in quantity and quality, and all this new material has stimulated new ideas and caused perspectives to change (Clack, J.A., 2002)). On page 90 of this issue (Shubin, N.H., 2004) it is shown how even fragmentary specimens can be used to make inferences about changes, their nature and sequence, that need to have occurred during evolution to terrestrial locomotion by tetrapods. A humerus of Devonian age that was discovered in Pennsylvania (Shubin, N.H., 2004) has been determined by its morphology to be from a tetrapod that was predicted to be a completely new, unusual form. Some of the anatomical and functional changes that have taken place in the tetrapod lineage during the Devonian have been inferred from this humerus, and some possible routes by which terrestriality was attained have in turn been suggested by these inferences.

The lobe-finned fish that are related to the early tetrapods had internal skeletons that were complex, and in their paired fins a suite of muscles, and it was these features that were eventually exploited to form limbs that were weight-bearing. The most proximal 3 elements of the limbs of the immediate relatives of tetrapods are readily identifiable as being related to the humerus, radius and ulna in the forelimbs and the femur, tibia and fibula of the hind limb. In the tetrapod-like fish such as Eusthenopteron and Panderichthys from the Devonian the humeri bear particular resemblance to those of the early tetrapods with several key features being common to all of them. Fins and limbs are, however, conspicuously different with respect to orientation, range of movement and function. Some clues to the timing and sequence in which these differences arose are provided by this new humerus, though yet more intriguing questions are posed by it.

The new humerus is flattened dorsoventrally, as is the case in other tetrapods from the Devonian and Panderichthys, a closely related form, and there appears to be a greatly restricted range of movement in the shoulder joint. However, as in other fish, the fin and shoulder joint of Panderichthys face posteriorly, whereas they are reoriented to face laterally in tetrapods. The attitude of the limb to the body is essentially horizontal rather than vertical, as a result of this reorientation; the operational space within which the limb acts is level with the shoulder joint instead of posterior to it; also, the direction in which the muscles pull is approximately at right angles to the body and not at an acute angle to it. It is suggested (Shubin et al., 2004) that it is shown that the orientation of a ridge on the ventral side of the new humerus that some of these changes had already occurred. A leg that was effective in the dorsoventrallplane was the net result of this change, and it could prop up the plated head that was relatively large in an early tetrapod, though its movement from front to rear, that was required for walking, was very limited.plane was the net result of this change, and it could prop up the plated head that was relatively large in an early tetrapod, though its movement from front to rear, that was required for walking, was very limited.

The distal end of this new humerus is, according to Clack¹, probably its most striking aspect. It is fairly similar to those of related forms proximally. The new humerus shows unique modifications distally, such as a short, very stout entepicondyle, an ectepicondylethat is robust and projecting, and articular facets that are relatively enormous for the radius and ulna. According to Clackthat is robust and projecting, and articular facets that are relatively enormous for the radius and ulna. According to Clack¹ the former 2 features suggest elaborate muscles to the ulna and radius, respectively, whereas the radial and ulna facets that are large and tucked beneath the humerus end, suggest short’ stout epipodials forming an elbow that was permanently flexed. A similar arrangement is proposed for Ichthyostega, a tetrapod from the Devonian of Greenland (Jarvik, e., 1996) that is more widely known, though in this case it is more exaggerated. The new humerus doesn’t appear to be of a type that would work as a walking limb, and it is suggested by recent work that Ichthyostega is unlikely to have “walked” in a conventional way (Clack, J.A., Blom, H. & Ahlberg, P.E.).

One question is about what was at the end of the limb from which the humerus came. In the lineage of tetrapods it is unclear when fin rays were replaced by digits; it is therefore a theoretical possibility that at some point there was a combination of digits and remnant fin rays. It has become clear, contrary to some previous development evidence, that an extensive fin endoskeleton is not incompatible with an extensive web that is ray-supported. The rhizodonts, a group of lobe-finned fish, specialised in just such a combined morphology, having large pectoral fins that have been suggested to have been used for station-holding, maintaining a steady position while waiting for prey to come close enough to ambush (Jeffery, J.E., 2001). This kind of function is suggested by Shubin et al., as a possible precursor to the evolution of limbs that are weight-supporting in tetrapods. The 3 tetrapod limbs from the Devonian that are more completely known were polydactylous, and given the phylogenetic position that has been proposed for the animal that this humerus came from, it seems likely that there were more than 5 digits, as well as an ulna that was shorter than the radius. Clack¹ suggests it is possible that digits arose prior to the evolution of wrists and ankles (Coates, M.I., 1996), and it is possible that an array of digits could have assisted to spread the load on appendages that were prop-like, before the evolution of stable load-bearing joints at wrists and ankles.

It has been shown by recent discoveries that tetrapods from the Devonian were distributed worldwide, and they had a range of specialisations that were previously not suspected, and the view of them as conservative, clumsy animals has changed completely. The new view of them is as having a range of morphologies and adaptations that is consistent with them being innovators that were exploiting niches that were previously unoccupied. One of these adaptations was the ears of Ichthyostega which were highly specialised to hearing underwater (Clack, J.A. et al., 2003). Though their walking was probably not efficient, their modes of locomotion varied widely, as their skeletons and sensory organs evolved to deal with the challenges inherent in emerging from the water onto the land.

The site in Pennsylvania where the new humerus was discovered has also produced 2 other new taxa of Devonian tetrapods. Clack¹ says that if this humerus proves to be from a third new form to be recovered from the site it would indicate that there was a wide diversity of tetrapods existing in close proximity, at a site that is proving to be one of the richest and most varied sites from which vertebrates from the Devonian have been recovered [a faunal and floral list is provided in (Daeschler, A.C., Fromes, A.C. & Mullison, 2003).

For vertebrates from the Late Devonian and Early Carboniferous, North America is relatively unexplored and these new discoveries will no doubt stimulate further exploration in an area of study that has the potential to be highly rewarding.

Sources & Further reading