The origin of fingers and toes in basal tetrapods

If you ever wondered
how five fingers and toes came to be the ‘standard’ for reptiles (including mammals), we can turn to the large reptile tree (LRT, 1426 taxa; subset Fig. 1) to sort out this question.

With so many taxa
among basal tetrapods known only from skulls, the following is an exercise in phylogenetic bracketing.

Figure 1. Graphing the presence of fingers and toes in basal tetrapods, updated today with the addition of 4 digits in Panderichthys.

We start with lobefins
These are fish that have no fingers or toes. The most primitive bony fish, like Cheirolpis, had lobe fins and rays. Sarcopterygians emphasized the lobe part. Bony fish reduced the lobe part and emphasized the ray part. Within the lobe the humerus, radius, ulna and smaller parts appeared (one bone, two bones, many bones). Originally the radius was much longer than the ulna.

Dvinosauria
are the most primitive taxa in the LRT to have a sub equal radius and ulna (preserved in Laidleria) and a sub equal tibia and fibula (preserved in Gerrothorax). Gerrothorax is the most primitive taxa to preserve metacarpals. They were poorly ossified, but there were five in number.

Colosteus 
(Fig. 2) preserves four fingers (1-4) on a tiny forelimb. Only the front half of this taxon is known.

Pholidogaster
(Fig. 2) more or less preserves five toes. The manus was not preserved, but the radius and ulna were slender beneath a robust humerus.

Figure 6. Colosteus relatives according to the LRT scaled to a common skull length. Their sizes actually vary quite a bit, as noted by the different scale bars. Only Pholidogaster and Colosteus are taxa in common with traditional colosteid lists.

The vast majority of basal tetrapods
retained this digit pattern: four on the forelimbs, five on the hind limbs.

Exceptions include
Acanthostega (Fig. 3) with 8 fingers and 8 toes. Ichthyostega has 7 toes (manus unknown).

Acanthostega demonstrates a reversal:
The radius is twice as long as the ulna, as in lobefin fish. Apparently neotony produces this reversal as Acanthostega became sexually mature as a more fully aquatic ‘tadpole’, much smaller than its ancestor, Ossinodus (Fig. 2), for which only a few toe parts are known.

We looked at the convergently more aquatic Ichthyostega
earlier here. Both are Late Devonian taxa, appearing tens of millions of years later than the Middle Devonian trackmaker.

Figure 3. Ossinodus is the more primitive taxon in the LRT compared to the smaller Acanthostega, the tadpole of the two.

Proterogyrinus had five fingers and five toes,
but it appears to have developed the extra digits all alone and convergent with amniotes (= reptiles) and their kin (see below).

Cacops and kin (Dissorophidae, Lepospondyli)
also developed five fingers and five toes by convergence with reptiles. Other lepospondyls, like the frog, Rana, did not have more than four fingers.

The first taxon in our lineage with five fingers and five toes
is Utegenia, which gave rise to the clade Seymouriamorpha and the clade Reptilomorpha (by definition, taxa closer to reptiles than to salamanders: Eusauropleura, Gephyrostegus, their last common ancestor and all their descendants).

As early as the Late Devonian
the basal reptilomorph, Tulerpeton (Fig. 4), developed an exceptional and tiny sixth finger. Since no more taxa in this lineage had a sixth finger, this is not a reversal, but a novel digit. Originally this taxon was thought to have six toes (digit 5 had to be fully restored), but new reconstructions do not confirm this hypothesis.

Figure 4. Tulerpeton peps in situ and several reconstruction options using published images of the in situ fossil. The one at upper left most closely resembles sister taxa and has more complete PILs (parallel interphalangeal lines).

The above LRT fish-to-tetrapod transition
only partially replicates and confirms the traditional one provided by Clack 2009 (Fig. 5) with far fewer taxa.

Figure 5. The classic paradigm illustrating the fish-to-tetrapod transition from Clack 2009.

If anyone knows of taxa pertinent to this subject
please let me know and I will add them. At present very few taxa represent many more taxa (phylogenetic bracketing) since so many taxa in the above subset do not preserve extremities or they were overlooked and not collected or published.

References
Clack JA 2009. The fish-tetrapod transition: new fossils and interpretations. Evo Edu Outreach (2009) 2:213–223. DOI 10.1007/s12052-009-0119-2