A humbling book, for me at least. It nicely illustrates the Red Queen’s Principle as applied to scientific understanding - you have to run as fast as you can just to stay in one place, and if you want to get somewhere you have to run even faster.


The urge to classify things must have evolved early on. It’s obviously a survival trait: grouping plants into those that are good to eat and those that are not; animals into those that are dangerous and those that are not, and people into those that are friends and those that are not. Classification is so ingrained that things which don’t fit neatly into a category - “neither fish nor fowl” - are regarded with some distrust. Creationists often think a taxonomist’s inability to assign some organism or another to a particular species was a refutation of evolution; of course, it’s the other way around. The basis of Linnaean taxonomy was that species and all higher taxonomic units had distinct boundaries, so that every organism could be put in its own pigeonhole. When evolutionary theory came along, Linnaean taxonomy was pounded into the round hole, even though it was obvious that it was a square peg; the whole point of evolution is that there will be transitional forms that won’t fit nicely into one category or another.


My personal introduction to tetrapod evolution came in the 1970s and the text I used was Vertebrate Paleontology, by Alfred Sherwood Romer. This is a magnificent “coffee-table” book, with hundreds of meticulous line drawings of skulls, skeletons, and miscellaneous odd bits. The organization is still Linnaean; there’s a chapter on “bony fish” followed by a chapter on “amphibians”; “amphibians” were held to be descended from rhipidistian crossopterygians and the closest living relative of amphibians was coelacanths. After long and careful perusal of Vertebrate Paleontology, I could make “rhipidistian crossopterygian” roll trippingly off my tongue. This would snow the average audience at parties (and insure I probably wouldn’t be invited again).


So it was with a certain smugness that I picked up Jennifer Clack’s Gaining Ground. Even though I hadn’t really paid much attention to tetrapod evolution in the 30 or so years since I’d last studied it, I assumed it wouldn’t be too hard to pick up where I’d left off - maybe a couple interesting new fossils here and there, but they would just build on my previous experience.


Gaining Ground starts off with a relatively basic introduction to geological time, the mechanics of fossilization, and what it means (in Linnaean terms, at least) to be tetrapod. Then things get interesting. Clack dives right into skull and skeletal anatomy, so the reader can tell what’s going on in subsequent chapters. While the postcranial skeleton isn’t that hard, fish and early tetrapods have a much more complicated skull than mammals. That meant I had to go back and try and refresh my memory on a whole bunch of anatomy that I was never even good at back when I was supposed to know it; the premaxilla connects to the maxilla, the maxilla connects to the lacrimal, the lacrimal connects to the quadratojugal, oh hear the word of the anatomist. To make matters worse, there’s a more or less standard set of abbreviations for all these so that you can fit them into a drawing. I found myself going back and forth to the drawings of generic lobe-fin fish and early tetrapod skulls all through the book to keep track of what (for example) “ectopt” meant in an illustration (ectopterygoid, in case you’re curious).



The next stage of my humiliation involved the sledgehammer-to-the-forehead realization that Linnaean taxonomy just doesn’t work anymore and it’s all cladistic taxonomy now. Of course I knew that on an intellectual level but it hadn’t really reached me practically, just like you might know that relativistic physics has replaced classical mechanics but don’t really consider that while trying to parallel park (for example). The catch here is that understanding the way cladistic taxonomy works very important to understanding tetrapod evolution; it’s as if somebody came out with a new model car that you couldn’t drive without a basic knowledge of curved space-time. Thus I discovered that old familiar groups like crossopterygians and rhipidistians and even amphibians aren’t valid any more, and that lungfish are now thought to be more closely related to tetrapods than coelacanths are. Fortunately, Clack explains why the names aren’t valid; otherwise I would have been hopelessly lost trying to figure out where they went. (As an aside, if you don’t know anything about cladistic taxonomy you’re going to be in trouble, but if you know a little bit about it, like I did, Gaining Ground has a lot of examples that put theory into practice).


As an example of this, Clack considers the use of the term “amphibian”. The Linnaean definition is nice and discriminatory; they’re air breathing animals that have to return to the water to spawn, and that (hot tub parties aside) separates them from everything else. The problem Clack points out is that we know next to nothing about the reproduction of early tetrapods or their ancestors; extant coelacanths are ovoviviparous and there’s no reason to assume that aquatic spawning in extant amphibians is the ancestral rather than a derived condition. Of course, this has been true all along; when Romer was classifying fossil forms as “amphibians” he didn’t know anything about their reproduction either, but based classification on other characters shared with extant forms. Clack prefers the term “tetrapod” but acknowledges that’s not perfect either; the problem is there’s a bunch of changes going from something that’s clearly a fish to something that’s clearly a tetrapod. She devotes a whole section to “How to turn a fish to a tetrapod”; legs are the obvious thing, but tetrapods have quite different skulls from fish (the elimination of gill arches allows the development of a neck and the need to breathe air requires changes in the palate) and quite different backbones (in most fish the backbone is used for locomotion; in most vertebrates it’s used to support the body). There are subtler changes that don’t show up in the fossil record - the inner ear is a lot more complicated in tetrapods (a fish needs to know its body orientation, but it can’t “lose its balance”) and the kidney function changes (in fish it’s concerned with maintaining salt balance and nitrogen waste is disposed of by the gills; in tetrapods the kidneys perform both functions). What happens with these considerations is the bane of Creationists; there are lots of transitional forms that have some “tetrapod” characteristics. Clack uses a group “Tetrapodiforms” that includes some things (like the middle Devonian Eusthenopteron) that would look just fine over the mantelpiece with a lure hanging out of its mouth but which has some tetrapod skull characters and others (like the late Devonian Icthyostega that have reasonable-seeming legs but still possess a very fish-like backbone.



After the fairly heavy introductory chapters, the middle part of the book is more like a traditional paleontological treatise. There’s a nice discussion of Devonian paleoecology, both general and as it relates to what we know about the specific environment that the earliest known tetrapods lived in. There’s lot of interesting things here. For one thing, the Devonian atmosphere had somewhat less than the modern oxygen levels (about 17-18%, but rising rapidly) and a lot more carbon dioxide (about 3-4%). (In fact, the modern atmosphere has the lowest carbon dioxide concentration in Earth history, with the possible exception of the late Carboniferous/early Permian). The day’s a little shorter, the Moon’s closer, tides are higher, the continents are bunched together, there are repeated glaciations in the high southern latitudes that cause repeated sea level changes, land plants are going strong, and land arthropods have made an appearance. In this context, Clack tries to figure out exactly why growing legs and breathing air was a good idea and has some interesting suggestions (although probably not the traditional “to get from one drying pool to another”). The stars here are Ichthyostega and Acanthostega, both known from multiple specimens in East Greenland. The limbs on these animals came as a surprise to vertebrate anatomists - it was always thought that five digits on each limb was the ancestral state in tetrapods (as evidence of this, when people or other modern vertebrates have extra digits, they’re always a duplicate of an existing digit, not an atavistic long-lost one). However, both Ichthyostega and Acanthostega have eight. Clack points out that these animals still have very fish-like backbones, including space for a notochord. She suggests that land-based locomotion was accomplished by flexing the entire body and using the legs as props to keep the animal from falling over and to gain traction against the ground. Interestingly, there’s Devonian track ways that seem to show exactly this: imprints of feet where the digit impressions point perpendicular to the centerline of the track, rather than forwards.

Continuing into the Carboniferous, Clack laments that just when things are getting interesting the bottom falls out of the fossil record. There’s next to nothing but some disarticulated bones for about 30 million years. By the time tetrapods start showing up again, several forms, after going to all the trouble to grow legs and breathe air, lose their legs and go back to the water. There are at least two limbless groups, aïstopods and adelogyrinids, and a whole bunch of aquatic types that have the limbs greatly reduced. (Part of this may be a preservation artifact - you usually have to end up dead in the water to get fossilized, so aquatic or marine forms of all organisms tend to be emphasized at the expense of terrestrial types). By the end of the Carboniferous, atmospheric oxygen was almost 35%; maybe they went back in the water to avoid bursting into flame. (I don’t recall reading about any special fire-retardant characteristics in Carboniferous plants; I’ll have to look into that).


Clack goes on to the Permian; by then there are lots of tetrapod groups, including “reptiles” (Clack prefers “amniotes”, although it strikes me this has the same problem as “amphibian”; how do we know if these organisms had an amniote egg or not?). The final chapters recapitulate the transition to the tetrapod form, with more detail about internal, non-fossilized changes that took place.


Overall, this is a very good book. For the nonspecialist it’s really easy to get put off by the first few chapters with their heavy emphasis on anatomical terms. If you persevere, you’ll get some useful information; and even if it’s too much for you the subsequent chapters can be read profitably as a summary of late Paleozoic ecology.
( )