¡®What¡¯s the best evidence that pre-carboniferous plants and animals colonized the land in partnership? Coprolites: petrified feces, great caches of crap which exist in the United Kingdom and eastern Europe. More than anything else in the sedimentary record these tell us about the nature of the animals in the late Silurian, and about the plants they ate.

Most of the droppings contain undigested land plant spores and a mass of transport tubes and cuticle. They come mostly from flood plain silt, which may mean the animals that shed them were not land-based themselves; the cuticles of aquatic scorpions sometimes do show in coprolites.

A clue about what the animals were like is traces of plant responses to injury. Psilophyton and other trimerophyte fossils show wounds borne of mandibles, those of arthropods, perhaps; the change in cell layers below the bites shows that the plants were alive at the time they were eaten, and had not merely washed like compost into the aquatic environment.

How the predators ate¡ªthat is, by chewing or sucking¡ªsuggests how they looked: they had mouth parts or suckers. But what about the harder issue of what plant parts they chose to eat? Large numbers of spores, and of one or two types principally, account for most of the bulk of extant coprolites. As an early form of herbivory pollen feeding makes intuitive sense, and the nutrient value of spores and sporangia would surely have been greater than vegetative tissue, especially in nitrates. In living Onoclea sensibilis l. are huge amounts of lipids and proteins identical to the storage materials of higher plants. The animals seem to have swallowed these particles whole and passed them unchewed into their feces. Living pollen feeders do sometimes chew their food, but others extract nutrients enzymatically. Perhaps this is what our ancient land animals did.

Was it enzymes or simple diffusion by which Silurian guts profited by a diet of spores? The walls of Onoclea spores at least seem not to be able to pass molecules larger than glycerol without help. This makes work difficult even for an enzyme. So the extraction of nutrients by Silurian animals may have been brought about by the ingestion of other substances such as perispore or locular fluids.

Did they always eat from the vine, so to speak, or were they also litter-feeders? This latter could explain the high concentration of spores in coprolites, to be sure. Less durable foods like fungi and parenchyma might have disappeared from the feces, of course. However the spores themselves show little or no sign of bacterial or fungal damage, suggesting that they were devoured from the living plant. Or perhaps there are other coprolites that did not survive, not being jammed with durable structures like spores, as suggested by recent tests with millipedes. Are fungal remains likely to have survived? In fact the reproductive structures do survive in aerial axes of Psilophyton dawsonii.

On the evidence of the non-animal fossils, how big were the animals themselves, and what may they have looked like? The size of the coprolites is too large for mites to have extruded, and too small for earthworms. Fossils of undeniably herbivorous animals do not exist from the Silurian. Modern embryophyte-spore feeders like ants and caterpillars are not good indices since they come from lineages not developed in the Silurian. Whatever they looked like, it is clear that the fortunes of animals colonizing land were more bound up with plants than anyone has thought until now. The plants themselves furnish the proof for this.

This all squares with the direct fossil record. We know of plenty of land animals from the Silurian. We are ignorant only of whether they ate plants or each other or something else altogether. A new cache of cuticle has been discovered from Upper Silurian formations in Britain which contains land-living centipedes and an arachnid. Another site in Gilboa, New York, is even richer, containing arachnids, eurypterids, chilopods, and possibly insects. The idea gains strength then, that plants and animals colonized the land in very close partnership indeed.¡¯