Fishy fragments tip the scales

by Henry Gee

The discovery of strange fragments of what may be the earliest fossil fish will spark a debate about how we recognise the most primitive fossil representatives of any living animal group. The minute, 510-million-year-old bone-like fragments come from Queensland in Australia, and are described in the 31 October 1996 Nature by Dr Gavin Young and colleagues.

The fragments appear to be part of the external skeleton of jawless fishes called ostracoderms, distantly related to modern lampreys, as well as to modern vertebrates with jaws (including all modern fish, and land vertebrates such as ourselves).

Ostracoderms were the earliest known true vertebrates and have been known since Darwin's day, but because the internal backbone of ostracoderms was made of cartilage and fossilizes poorly, the external skeleton is all we have. There have been debates about the status of the earliest undisputed ostracoderm, Anatolepis, known for twenty years from fragmentary fossils from North America, Greenland and Spitzbergen, in the Arctic Ocean. Anatolepis fossils come from rocks a few million years younger than those bearing the newly found Australian fossils. The fishy nature of Anatolepis has been much disputed, with some researchers suggesting that the fragments belonged to a crustacean, and not to a fish-like vertebrate. Earlier this year, new techniques in microscopy allowed researchers to resolve the issue, confirming that Anatolepis was indeed a vertebrate.

The vertebrate affinity of these minute shards can be established because of the distinctive nature of vertebrate hard tissues. The hard shells of invertebrates such as molluscs and crustaceans consist of a simple layered structure, mineralized with calcium carbonate. In vertebrates, the mineral of choice is calcium phosphate, and the structure of the hard tissues is more complicated. Vertebrate hard tissues consist of bone (which takes various forms), enamel, and a tissue called dentine, that falls somewhere in between.

Our teeth typify the archetypal three-layered structure of vertebrate hard tissue. A thin layer of tough enamel caps a body made of dentine, which is permeated by small tubules branching off from a central pulp cavity. The whole thing is securely fastened to a third tissue, bone, in the jaws. Sharks have teeth of a very similar structure, but they are found not only in the jaws, but clothing the entire body‹the rough texture of shark skin comes from thousands of embedded tooth-like scales called denticles. Dentine, in particular, is absolutely characteristic of vertebrates. The discovery that Anatolepis armour contained tooth-like dentine bodies, each capping a pulp cavity, settled its vertebrate nature.

It was the other aspects of Anatolepis that caused the debate. The dentine bodies are embedded in an otherwise unbroken sheet of hard tissue, arranged in thin layers, and occasionally perforated by pores. This is reminiscent of what one sees in invertebrate hard tissues, leading to the debate.

This is where the new Australian fossils come in. They consist of an unbroken, three-layered sheet of hard tissue, perforated by pores, but which do not (as far as we know) contain denticles, or any dentine-like structures. Although similar to vertebrate hard tissues in a general way, in detail they are quite unlike any vertebrate tissues ever seen before.

So, are they vertebrates, or aren't they? Claims of vertebrate affinity are risky, but given the history of discoveries in this area, it's a risk worth taking, at least according to Philippe Janvier of the Muséum National d'Histoire Naturelle in Paris. Ostracoderms in general display a wide and sometimes baffling range of bone tissues not seen today, so it was not unlikely that something would turn up that seemed, initially, to extend the range of vertebrate tissue types beyond existing knowledge.

This observation leads to a deeper justification for taking such a risk. That is, our recognition of the affinity of strange fossils depends very much on what is available for comparison among modern animals. But this approach may be mistaken.

For example, we observe that the range of hard tissues in modern vertebrates includes bone, dentine and enamel, usually found together in particular structures such as teeth. It is natural for us to assume that because vertebrates have these tissues and structure now, they must have had them for all time. Furthermore, because we observe these tissues and structures in all vertebrates, and not in any other animals, we use them as defining features of vertebrates as a group.

But these assumptions conceal a flaw. Consider that the range of form in modern vertebrates, or indeed modern animals, has been impoverished by extinctions over the course of hundreds of millions of years. Animals with bodies as bizarre as you like have passed away. So it may be unwise to impose the depauperate range of form seen in present-day animals on the incalculably larger range of form in extinct creatures. Extinct, primitive vertebrates may have had combinations of enamel, bone and dentine not seen today. And they may, of course, had kinds of hard tissue completely different from enamel, bone and dentine. But if our 'search image' of vertebrate hard tissues is restricted to those found in modern animals only, we would be hard put to recognising extinct varieties for what they are.

© Macmillan Magazines Ltd 1996 - NATURE NEWS SERVICE

Note: This item from the Nature News Service is mounted on this Web page by special permission of Nature.

[Return to Chapter 7]

[Return to UC Davis Geology Department Home Page]