Theory goes to water as fish finger found

The following article appeared the other day in Australia's national newspaper. My comments are bold.

Theory goes to water as fish finger found, The Australian, September 23, 2008. PARIS: Scientists have traced the origin of

[Above (click to enlarge): Reconstruction of Panderichthys: Wikipedia]

fingers and toes to fish-like creatures that roamed the seas 380 million years ago, a new study has found. Panderichthys was not "fish-like." It was a fish!

The findings, published yesterday in the science journal Nature, upend the prevailing theory on the evolution of digits. It had long been assumed that the first creatures to develop primitive fingers were tetrapods, air-breathing animals that crawled from sea to land about 10 to 20 million years later. So fingers arose under water, 10-20 million years (or more) before they were needed on land.

The need to adapt to swampy marshlands and terra firma, so the theory went, is what drove the gradual shift through natural selection from fish fins suitable only for swimming to weight-bearing limbs with articulated joints. This is based on the original Darwinian `just-so' story:

"Origin of tetrapods ... The Devonian, during which land adaptations originated, was seemingly a time of seasonal droughts when life in fresh waters must have been difficult. ... if the water dried up altogether, the amphibian had the better of it. The fish, incapable of land locomotion, must stay in the mud and, if the water did not soon return, must die. But the amphibian, with his short and clumsy but effective limbs, could crawl out of the pool and walk overland ... and reach the next pool where water still remained. Once this process had begun, it is easy to see how a land fauna might eventually have been built up." (Romer, A.S., 1945, "Vertebrate Paleontology," pp.140-141. My emphasis).

Which shows that this major transition had little, if anything to do with the Darwinian natural selection of random micromutations.

But the study reveals rudimentary fingers were already present inside the fins

[Right (click to enlarge): Computerized 3D reconstruction of the fin bones of Panderichthys, showing precursors of fingers: FOX News]

of the shallow-water metre-long Panderichthys, a transitional species that was more fish than tetrapod. Note that these "rudimentary fingers," complete with precursors of all the fore- and hind-leg bones:

"Panderichthys ... the internal, endochondral bones of the fin are closely comparable to those of terrestrial vertebrates. There is a single proximal humerus and more distal ulna and radius in the forelimb, and the femur, tibia and fibula in the hind limb. They are succeeded distally by bones that are homologous with proximal elements of the wrist (intermedium, ulnare, and centralia) and ankle (fibulare, intermedium, and possibly distal tarsals) of land vertebrates..." (Carroll, 1997, "Patterns and Processes of Vertebrate Evolution," pp.230-232. My emphasis).

were "present inside the fins" of this "species that was more fish than tetrapod" and therefore "they could not have functioned in the manner of these joints in terrestrial vertebrates":

"... but they could not have functioned in the manner of these joints in terrestrial vertebrates because they are extensively overlapped by the radius and the tibia. The entire endochondral skeleton is within a functionally continuous fin structure" (Carroll, 1997, Ibid, pp.230-232. My emphasis).

That is, they could not be `seen' by the environment and so were `invisible' to natural selection. So natural selection cannot, even in principle, explain these major features of the fish-tetrapod transition.

"What we have shown is that the hand and the foot emerge from pre-existing bits of the fin skeleton that were just reshaped, rather than being entirely new bits that were bolted onto the existing fin skeleton," said co-author Per Ahlberg, a researcher at Uppsala University in Sweden. Again, note that "the hand and foot" of all subsequent land vertebrates (amphibians, reptiles, mammals and birds) "were already present inside the fins of" a "fish"!

The discovery did not come from a new archeological find but from a re-examination of the existing fossils, he said. Previous research, it turns out, simply overlooked what was there. "The problem is that all good specimens of Panderichthys come from one location" -- a brick quarry in Latvia -- "where the clay is almost exactly the same colour as the bones," Dr Ahlberg said. "If you are interested in tiny, fragile bones at the outer end of the fin skeleton, it's nearly impossible to see what is going on." Clearly, these "tiny, fragile" finger and toe "bones at the outer end of the fin skeleton," conferred no selective advantage to:

"Panderichthys ... quite a large fish, with ... a total body length of over a meter" (Clack, 2002, "Gaining Ground," p.64. My emphasis).

So Dr Ahlberg and two colleagues ran a specimen, still embedded in clay, through a CT scanner at a hospital. The image shows stubby bones at the end of the fin skeleton clearly arrayed like four fingers, called distal radials. There are no joints, and the bones are quite short, but there could be no doubt as to what they were. That there were "no joints" in these "fingers" shows that they could not function as fingers, nor would they be rigid. Therefore they could have no selective advantage for Panderichthys but would be a selective advantage to its descendants millions of years in the future which had functional fingers with joints. But in that case it would be a "part of the structure of any one species had been formed for the exclusive good of another species" and so "would annihilate" Darwin's "theory" (Darwin, 1872, "Origin of Species," p.162)!

Primitive Fingers Found in Prehistoric Fish, FOX News, September 23, 2008, Jeanna Bryner. An ancient fish sported something like fingers that were the precursors to our own digits, according to an analysis of a new fossil skeleton. Since I accept Universal Common Ancestry (but not Evolution), I agree that these fish "fingers ... were the precursors to our own digits." And also that:

"... all tetrapods [and therefore all land vertebrates-amphibians, reptiles, birds and mammals] had a single common ancestor" (Clack, 2002, Ibid, p.66. My emphasis).

And therefore if that one fish "single common ancestor" (out of uncountable trillions of fish) had not by one, or a series of, `lucky' mutations which created precursors of: 1) both forelimbs and hindlimbs, complete with "humerus ... ulna and radius" and "femur, tibia and fibula," as well as "wrist (intermedium, ulnare, and centralia) and ankle (fibulare, intermedium, and ... tarsals)" and "our own digits"; and 2) "pectoral and pelvic girdles" attached to the spine for those limbs to articulate to, neither we, nor any land vertebrate, would be here. But according to my Progressive Mediate Creation general theory, God supernaturally intervened in the genome of that "single common ancestor" to create the:

"blueprint of terrestrial limb structure [in] the genome... of the ["single common ancestor"] osteolepiform fish" (Wilcox, 1990, "Created in Eternity, Unfolded in Time," pp.6:23-24).

"It's really the last piece of evidence to say fingers are not new. They were really present in fish," said lead researcher Catherine Boisvert, an evolutionary biologist at Uppsala University in Sweden. The fossilized skeleton belonged to Panderichthys, a predatory fish that spanned up to 4 feet (130 cm) and likely dwelled in shallow waters where it inched along the muddy bottom about 385 million years ago. The problem for Darwinian `blind watchmaker' evolution by the natural selection of random mutations is to explain:

"How can the world of an aquatic predator quickly select, collect and individuate the information for a highly coherent adaptive blueprint of terrestrial limb structure?" (Wilcox, Ibid, pp.6:23-24. My emphasis)

by inching "along the muddy bottom" of "shallow waters".

While the fossil was discovered in the 1990s by chance in a brick quarry in Latvia in northern Europe, scientists only recently analyzed the fins with computed tomography (CT) and found that the right paddle is tipped with four bony extensions. If you were to turn back the clocks to the Devonian period when Panderichthys lived and spied the fish, you would not have noticed its "fingers," Boisvert explained. Neither would natural selection "have noticed its `fingers'" being that tiny in proportion to Panderichthys' "4 feet (130 cm)" body.

The fan-like array of fingers, however, would have made Panderichthys' paddles broader at the ends. The broad fins would have made for stronger supports for the fish to lean on rather than for all-out swimming. This sounds like yet another Darwinian `just-so' story. But it is gross overkill for "a predatory fish that ... dwelled in shallow waters where it inched along the muddy bottom" to develop "inside" two pairs of fish fins all the fore- and hind-limb bones of "terrestrial vertebrates".

"It was probably using its front fins as supports to be able to look up, kind of doing push-ups at the bottom of the river looking outside with its eyes," Boisvert said, adding that the fish's eyes were on the top of its skull and thus probably good for looking above the mud for fish food. Though Panderichthys was not made for landlubbing, if the need to hop from the water arose, the fish had the means. "So if it was stuck in a pool and it was drying out, [the fish] would have been able to get itself out to the next water body," Boisvert told LiveScience. "It's doing push-ups on land with its big fins and then its pelvic fins (hind fins) are used for an anchor in the mud." That the proto- fore- and hind-limb bones of future terrestrial vertebrates, after they had appeared, may have had some use for Panderichthys (and originally "the single common ancestor") does not thereby explain why they appeared. To just assume that would commit the post hoc ergo propter hoc fallacy.

Basically, Panderichthys would have dragged its body along land. "It wouldn't have been pretty," she added. The way that Boisvert puts this indicates she has no evidence that Panderichthys (or "the single common ancestor") ever "dragged its body along land." The fossil finding, detailed in the Sept. 21 issue of the journal Nature, fills in a gap in the evolution of tetrapods, or four-legged animals. About 380 million years ago, our fishy ancestors crept onto land. And here we have a time problem, the time-frame of the fish-tetrapod transition was only 5-15 million years (my emphasis below):

"Panderichthys and Elpistostege flourished in the early Frasnian and are some of the nearest relatives of tetrapods. But tetrapods appear only about 5 to 10 million years later in the late Frasnian, by which time they were widely distributed and had evolved into several groups ... This suggests that the transition from fish to tetrapod occurred rapidly within this restricted time span." (Clack, Ibid, 2002, p.96).

"The transition from fish to crawling four-legged tetrapod occurred ... about 360 million years ago during a relatively short geological interval-no more than probably 15 or 20 million years ..." (Strickberger, 2000, "Evolution," p.410).

"At least 377 million years ago a lineage of lobefins arose that was more tetrapodlike than Eusthenopteron. One of these was an animal called Panderichthys... In 10 or 15 million years, however, relatives of Panderichthys reworked their bodies into tetrapod form." (Zimmer, 1998, "At the Waters Edge," pp.104-105).

Which is a chronospecies problem, i.e. each change had to be "locked up":

"morphological change may accumulate anywhere along the geological trajectory of a species. But unless that change be `locked up' by ... speciation ... it cannot persist ... and must be washed out ... among varying populations of a species. Thus, species ... provide the only mechanism for protecting change" (Gould & Eldredge, 1993, "Punctuated Equilibrium Comes of Age," pp.226-227)

in a sequence of separate species, "align[ed], end-to-end" i.e. a "chronospecies":

"If an average chronospecies lasts nearly a million years, or even longer, and we have at our disposal only ten million years, then we have only ten or fifteen chronospecies to align, end-to-end ..... This is clearly preposterous. Chronospecies, by definition, grade into each other, and each one encompasses very little change. A chain of ten or fifteen of these might move us from one ... form to a slightly different one..." (Stanley, 1981, "The New Evolutionary Timetable," pp.93-94. My emphasis).

But as with mammals, "A chain of ten or fifteen of these might move us from one ... form to a slightly different one."

Fossil evidence has continued to refine scientists' understanding of this transition, though they still have many questions regarding the fin-to-limb transition and development of other locomotion features. For instance, one such transitional fish called Tiktaalik roseae lived about 375 million years ago and showed signs of both water living and land trekking. However, Boisvert said, even though Tiktaalik is closer evolutionarily to tetrapods, its specimens lack the distinct finger precursors seen on Panderichthys. Therefore these "distinct finger precursors seen on Panderichthys" cannot be explained by natural selection for locomotion on land, including land underwater.

I agree with former atheist Antony Flew that:

"The only satisfactory explanation for the origin of ... life ... is an infinitely intelligent Mind" (Flew, 2007, "There Is a God," p.123ff).

But I see no reason why such an "infinitely intelligent Mind" (who I assume is the God of the Bible), would stop at the origin of life and not continue to supernaturally intervene at other strategic points in life's history, including the fish-to-tetrapod transition.

Stephen E. Jones, BSc. (Biology).
My other blogs: TheShroudofTurin & Jesus is Jehovah!

---

"Neither the fossil record nor study of development in modern genera yet provides a complete picture of how the paired limbs in tetrapods evolved ... The closest comparison between the paired fins of obligatorily aquatic fish and animals that were at least facultatively terrestrial is provided by the osteolepiform sarcopterygians Eusthenopteron and Panderichthys and the stem tetrapods Acanthostega and Ichthyostega. ... Superficially, the paired fins of the fish appear typical of strictly aquatic vertebrates. They are small relative to the body; they narrow at the base that articulated with the pectoral and pelvic girdles, but broaden distally to form an effective surface for locomotion or directional control in the water. ... In contrast, the internal, endochondral bones of the fin are closely comparable to those of terrestrial vertebrates. There is a single proximal humerus and more distal ulna and radius in the forelimb, and the femur, tibia and fibula in the hind limb. They are succeeded distally by bones that are homologous with proximal elements of the wrist (intermedium, ulnare, and centralia) and ankle (fibulare, intermedium, and possibly distal tarsals) of land vertebrates, but they could not have functioned in the manner of these joints in terrestrial vertebrates because they are extensively overlapped by the radius and the tibia. The entire endochondral skeleton is within a functionally continuous fin structure, as seen from its scaly covering. There is no trace of endochondral skeletal elements comparable with the distal carpals or digits of terrestrial vertebrates. ... In contrast with the clear homology of the more proximal limb bones in osteolepiform fish and early tetrapods, no obvious homologues of the digits is evident in any sarcopterygian. These bones appear de novo in the Upper Devonian tetrapods. How can this be explained?" (Carroll, R.L., 1997, "Patterns and Processes of Vertebrate Evolution," Cambridge University Press: Cambridge UK, pp.230-232).

"Panderichthys was quite a large fish, with a skull about 300 mm long, and a total body length of over a meter ... . Its body and skull were flattened and the snout rather pointed. The eyes were placed quite close together on the top of its head and were set beneath ridges, giving the impression of eyebrows and creating a subjectively tetrapodlike appearance. Other characters of the skull were also very tetrapodlike (Vorobyeva and Schultze 1991). Elpistostege is still relatively poorly known, and for understanding the story of the origin of tetrapods, Panderichthys will provide a satisfactory guide. By comparing its skull with that of a very early tetrapod such as Acanthostega, those tetrapodlike features that were present already in Panderichthys can be contrasted with those that had yet to evolve. This gives some ideas about the order and timing of the appearance of some tetrapod characters ..." (Clack, J.A., 2002, "Gaining Ground: The Origin and Evolution of Tetrapods," Indiana University Press: Bloomington IN, p.64).

"This review of the lobe-finned fish groups is not complete without the tetrapods, because this is where, evolutionarily speaking, they (and humans) belong. Modern tetrapods include on the one hand the amphibians-frogs, newts, caecilians, and their kin-and on the other the amniotes-mammals plus the `reptile' groups, including turtles, lizards and snakes, and crocodiles and their closest living relatives, the birds. It includes creatures that, although they do not have legs (limbs with digits) themselves, are descended from some that did. So bats and whales are tetrapods, as are birds and snakes. It also includes all the fossil forms such as dinosaurs, flying or swimming reptiles, and many other more bizarre and less well-known kinds, so long as they are descended from ancestors with legs ... Most current views maintain that tetrapods are a natural group, tied together by numerous unique characters that show that the group had a single common ancestor. Among the unique features that tetrapods share is the possession of limbs with digits ..." (Clack, 2002, p.66).

"Some things are clear, however. Both Elginerpeton and Obruchevichthys appear more closely related to tetrapods than was Panderichthys. They are also very closely related to each other, sharing some details that cause them to be placed in the same family (Ahlberg 1995). This family was widely distributed in the Frasnian. They were also different from the slightly later Devonian tetrapods, which will be described in the next chapter. They may represent an early and specialized offshoot from the tetrapod branch. Panderichthys and Elpistostege flourished in the early Frasnian and are some of the nearest relatives of tetrapods. But tetrapods appear only about 5 to 10 million years later in the late Frasnian, by which time they were widely distributed and had evolved into several groups, including the lineage leading to the tetrapods of the Famennian. This suggests that the transition from fish to tetrapod occurred rapidly within this restricted time span. Neither fishlike tetrapods nor tetrapodlike fish body fossils occur in the record before this (Clack 1997a). Indeed, the osteolepiforms as a whole are not found before the Middle Devonian. This lends weight to the suggestion that the tracks from the supposed Late Silurian or Early Devonian are not those of a tetrapod, and those from the Middle Devonian are unlikely to be so. Given our current understanding of phylogeny, tracks made by a terrestrial tetrapod are unlikely to be found before the late Frasnian, and the body fossil evidence conflicts with the interpretation of any pre-Famennian track as terrestrial." (Clack, 2002, p.96).

"Natural selection cannot possibly produce any modification in a species exclusively for the good of another species; though throughout nature one species incessantly takes advantage of and profits by, the structures of others. But natural selection can and does often produce structures for the direct injury of other animals, as we see in the fang of the adder, and in the ovipositor of the ichneumon, by which its eggs are deposited in the living bodies of other insects. If it could be proved that any part of the structure of any one species had been formed for the exclusive good of another species, it would annihilate my theory, for such could not have been produced through natural selection." (Darwin, C.R., 1872, "The Origin of Species By Means of Natural Selection," [1859], John Murray: London, Sixth Edition, Reprinted, 1882, p.162).

"But continuing unhappiness, justified this time, focuses upon claims that speciation causes significant morphological change, for no validation of such a position has emerged .. but why then? For the association of morphological change with speciation remains as a major pattern in the fossil record. We believe that the solution to this dilemma may be provided in a brilliant but neglected suggestion of Futuyma [Futuyma, D.J., "On the role of species in anagenesis," American Naturalist, Vol. 130, 1987, pp.465-473)] He holds that morphological change may accumulate anywhere along the geological trajectory of a species. But unless that change be `locked up' by acquisition of reproductive isolation (that is speciation), it cannot persist or accumulate and must be washed out during the complexity of interdigitation through time among varying populations of a species. Thus, species are not special because their origin permits a unique moment for instigating change, but because they provide the only mechanism for protecting change. Futuyma writes: `In the absence of reproductive isolation, differentiation is broken down by recombination. Given reproductive isolation, however, a species can retain its distinctive complex of characters as its spatial distribution changes along with that of its habitat or niche...Although speciation does not accelerate evolution within populations, it provides morphological changes with enough permanence to be registered in the fossil record. Thus, it is plausible to expect many evolutionary changes in the fossil record to be associated with speciation.' By an extension of the same argument, sequences of speciation are then required for trends: `Each step has had a more than ephemeral existence only because reproductive isolation prevented the slippage consequent on interbreeding other populations...Speciation may facilitate anagenesis by retaining, stepwise, the advances made in any one direction.' Futuyma's simple yet profound insight may help to heal the remaining rifts and integrate punctuated equilibrium into an evolutionary theory hierarchically enriched in its light" (Gould, S.J. & Eldredge, N., 1993, "Punctuated Equilibrium Comes of Age," Nature, 18 November, Vol 366, pp.223-227, pp.226-227. Ellipses original).

"When the mass media first reported the change in my view of the world, I was quoted us saying that biologists' investigation of DNA has shown, by the almost unbelievable complexity of the arrangements needed to produce life, that intelligence must have been involved. I had previously written that there was room for a new argument to design in explaining the first emergence of living from nonliving matter-especially where this first living matter already possessed the capacity to reproduce itself genetically. I maintained that there was no satisfactory naturalistic explanation for such a phenomenon. ... The philosophical question that has not been answered in origin-of-life studies is this: How can a universe of mindless matter produce beings with intrinsic ends, self-replication capabilities, and `coded chemistry'? ... The origin of self-reproduction is a second key problem. ... A third philosophical dimension to the origin of life relates to the origin of the coding and information processing that is central to all life-forms. ... So how do we account for the origin of life? The Nobel Prize-winning physiologist George Wald once famously argued that `we choose to believe the impossible: that life arose spontaneously by chance.' In later years, he concluded that a preexisting mind ... composed a physical universe that breeds life ... This, too, is my conclusion. The only satisfactory explanation for the origin of such `end-directed, self-replicating' life as we see on earth is an infinitely intelligent Mind." (Flew, A.G.N., 2007, "There Is a God: How the World's Most Notorious Atheist Changed His Mind," HarperCollins: New York NY, pp.123-125, 131-132).

"Origin of tetrapods.-The `why' of tetrapod origin has been often debated. Many of the earliest amphibians appear to have been fairly large forms of carnivorous habits, still spending a large portion of their time in fresh-water pools. Alongside them lived their close relatives, the crossopterygians, similar in food habits and in many structural features and differing markedly only in the lesser development of the paired limbs. Why did the amphibians leave the water? Not to breathe air, for that could be done by merely coming to the surface of the pool. Not because they were driven out in search of food-they were carnivores for whom there was little food on land. Not to escape enemies, for they were among the largest of vertebrates found in the fresh waters from which they came. Their appearance on land seems to have resulted as an adaptation for remaining in the water. The earliest-known amphibians lived much the same sort of life as the related contemporary crossopterygians. Both lived normally in the same streams and pools and both fed on the same fish food. As long as there was plenty of water, the crossopterygian probably was the better off of the two, for he was obviously the better swimmer-legs were in the way. The Devonian, during which land adaptations originated, was seemingly a time of seasonal droughts when life in fresh waters must have been difficult. Even then, if the water merely became stagnant and foul, the crossopterygian could come to the surface and breathe air as well as the amphibian. But if the water dried up altogether, the amphibian had the better of it. The fish, incapable of land locomotion, must stay in the mud and, if the water did not soon return, must die. But the amphibian, with his short and clumsy but effective limbs, could crawl out of the pool and walk overland (probably very slowly and painfully at first) and reach the next pool where water still remained. Once this process had begun, it is easy to see how a land fauna might eventually have been built up. Instead of seeking water immediately, the amphibian might linger on the banks and devour stranded fish. Some types might gradually take to eating insects (primitive ones resembling cockroaches and dragon flies were already abundant) and, finally, plant food. The larger carnivores might take to eating their smaller amphibian relatives. Thus a true terrestrial fauna might be established." (Romer, A.S., 1945, "Vertebrate Paleontology," [1933], University of Chicago Press: Chicago IL, Second edition, Fifth Impression, 1953, pp.140-141. Emphasis original).

"Darwin was spared a confrontation with the extraordinarily rapid origins of modern groups of mammals. He knew that the history of mammals extended back to the early part of the Mesozoic, but the record was not well enough studied in his day for him to recognize that the adaptive radiation of modern mammals did not commence until the start of the Cenozoic. Today, our more detailed knowledge of fossil mammals lays another knotty problem at the feet of gradualism. Given a simple little rondentlike animal as a starting point, what does it mean to form a bat in less than ten million years, or a whale in little more time? We can approach this question by measuring how long species of mammals have persisted in geological time. The results are striking; we can now show that fossil mammal populations assigned to a particular Cenozoic lineage typically span the better part of a million years without displaying sufficient net change to be recognized as a new species. The preceding observations permit us to engage in another thought experiment. Let us suppose that we wish, hypothetically, to form a bat or a whale without invoking change by rapid branching. In other words, we want to see what happens when we restrict evolution to the process of gradual transformation of established species. If an average chronospecies lasts nearly a million years, or even longer, and we have at our disposal only ten million years, then we have only ten or fifteen chronospecies to align, end-to-end, to form a continuous lineage connecting our primitive little mammal with a bat or a whale. This is clearly preposterous. Chronospecies, by definition, grade into each other, and each one encompasses very little change. A chain of ten or fifteen of these might move us from one small rodentlike form to a slightly different one, perhaps representing a new genus, but not to a bat or a whale!" (Stanley, S.M., 1981, "The New Evolutionary Timetable: Fossils, Genes, and the Origin of Species," Basic Books: New York NY, pp.93-94).

"Although the details are not yet fully known ... many paleontologists agree that land vertebrates, however they first evolved, were related to sarcopterygian lobe-finned fishes. The transition from fish to crawling four-legged tetrapod occurred by the end of the Devonian period, about 360 million years ago during a relatively short geological interval-no more than probably 15 or 20 million years-and encompassed perhaps three or more separate lineages (Carroll 1995)." (Strickberger, M.W., 2000, "Evolution," Jones & Bartlett: Sudbury MA, 1990, Third edition, p.410).

"All in all, Ichthyostega is a mystery of the first water. The theory that the direct application of environmental selection can `collect' the necessary morphological information, integrate it into individuated error checked blueprints, and thus create novel structures in organisms seems impossible to apply to Ichthyostega. How can the world of an aquatic predator quickly select, collect and individuate the information for a highly coherent adaptive blueprint of terrestrial limb structure? The theory that the blueprint already existed in some form in the genomes of the osteolepiform fish sounds more reasonable, but sorting it out under water is still difficult. However, if it were true, and if a life style of living on fish stranded on the edge of the swamp could provide a mild selective pressure, the previous encoding of a individuated blueprint could at least explain its tight coherence when it first appears." (Wilcox, D.L., 1990, "Created in Eternity, Unfolded in Time," Eastern College: St. Davids PA, Unpublished manuscript, Chapter 6, pp.23-24).

"Found in Latvia, this two-foot-long fish had a skull as flat as a coffee table and a smooth back that lacked the dorsal fins of other lobe-fins. Its shoulders-and the fins that attached to them-were so sturdy that it might have been able to move on them like crutches out of the water for short distances. Like coelacanths and lungfish, it probably could move with the left-right, left-right movements that would become our walk. Still, it would be impossible to mistake Panderichthys for a tetrapod. Its toeless limbs were buried inside a ring of fin rays, its braincase was still hinged, and instead of a stapes Panderichthys had a hyomandibular bone that was linked to its jaws and gills. In 10 or 15 million years, however, relatives of Panderichthys reworked their bodies into tetrapod form. Elginerpeton, the beast that Per Ahlberg found hiding in museum drawers, is not only the oldest tetrapod known but the most primitive as well. Its snout turned into a massive snapping trap, ligaments joined its pelvis to its spine. With only fragments of its limbs, it's impossible to know if there were toes yet, but Elginerpeton shows many signs of being an intermediate between lobe-fins like Panderichthys and later tetrapods. Its rear legs were twisted so much that its knees (if it had them) would have pointed to the ground, making the legs useless for walking but good for rowing. Judging from the fact that Elginerpeton was five feet long and hunted on river bottoms, one can assume that the first tetrapods must have been moderately successful at living like a lobe-fin. Within a few million years Elginerpeton was gone, but new kinds of tetrapods were evolving all around the world." (Zimmer, C., 1998, "At the Waters Edge: Fish with Fingers, Whales with Legs, and How Life Came Ashore but Then Went Back to Sea," Touchstone: New York NY, Reprinted, 1999, pp.104-105).