Second Eon, the First Age of Life: 4 Gigayears
In terms of biological complexity, the First Age should be considered roughly equivalent to everything between Earth's Cambrian and Permian Periods. Since we want lots of land aminals and plant analogues, to begin with we'll look at constructing a biosphere in the later end of that timespan, although along the way we'll be coming up with primitive forms that can later be used to fill in the earlier periods.
The kingdoms of life we have to work with are Aminalia (animal analogues), Pseudoplantae (plant analogues), Spongimorphs (fungi with mineralized skeletons), Pseudofungi (fungus analogues), Rangeomorphs (ediacarans), and microbe groups (bacteria, archaea, algae).
For our plant analogues, we could concievably use any of the Pseudoplantae, Spongimorphs, Pseudofungi, or Rangeomorphs. Complex plants are out, period. However, complex is a very subjective word, so we can probably rationalize a lot. Mostly, we should be looking at cycadoids, to bring in an earthling comparison, especially for the plant analogues derived from the Pseudoplantae kingdom. However, it's quite possible that we come up with plant body plans utterly unlike anything that we find on earth, in which case we'll just have to wing it. The biggest markers of complexity that I know of, however, are seed production, flowers, and the associated animal symbiosis. We should probably be going mostly for spore producers, and a few gymnosperms for out plant analogues. Of course, if we use rangeomorphs/ediacarans as plant analgues, then we could come up with something else entirely- maybe plants that are semimobile at some point in their lifecycle and can reproduce like animals? Rangeomorph plant analogues are likely to provide quite a bit of room for designing carnivorous plants as well.
Concerning mobile, animal analogue organisms, we know that Antipode will be dominated by Rangeomorphs, with few if any Aminals, while the Mainland will be dominated by Aminals. However, the distribution of other kindoms is completely open. Do we put everything earthlike on the Mainland, and make Antipode completely alien, or mix it up a little, so that, for example, Mainland Aminals frolick in giant Spongiform land-reefs instead of tree-forests, while Antipodan mobile Rangeomorphs live with a mixture of immobile Rangeomorphs and Pseudoplantae in roughly Earth-like Rainforests (but with a larger number of plants that can eat you)?
Earthling animals and plants have a whole lot of body plans to choose from. There's the vertebrates (under which classification are four-limbed land creatures, swimming creatures with lungs, and swimming creatures with gills), annelids, mollusks (which are further divided between gastropods and cephalopods, each of which can have a shell or lack one), arthropods (which can have two or three segments, and have a very wide range of limb numbers to choose from), etc. And with plants, we can choose between woody plants (which can be gymnospermic or angiospermic), soft-stem plants, vines, ferns, etc. We have, however, very few body plans worked out for Epimetheus. Even if we don't get any complete creatures very soon, we should at least look at what sorts of alien body plans we have available to base those creatures on.
For the aminals, we have basically three body plans- septopods, pentapods, and pseudoannelids, which serve to replace vertebrates and annelids. And the pentapod is really just a variation on the septapod, so that probably doesn't count. While these can serve as a basis for a huge variety of animals (we can replace reptiles, amphibians, mammals, whales, sharks, and birds with just pentapods if we have to), its not as interesting as it could be and comes nowhere near mimicking the diversity that we find on Earth. Please, somebody come up with analogues for some other body plans! Or even better, completely new, non-analogous body plans. Note, however, we do have an extra mobile kingdom to work with (Rangeomorphs), so Aminalia don't have to fill in for everything.
Rangeomorphs are interesting. They are just alien analogues, so we don't have to stick entirely to the ancient Earthling ediacarans, but that is the basic idea. We could, for example, give some of them mineralized shells. I'm looking mostly at hydrostatic skeletons for supporting structures, though. That might be helped along by the development of extensile, rather than (or, more likely, in addition to) contractile muscle. Rangeomorphs are by definition extremely simple, modular creatures, and can be made to fit the roles of either animals or plants. Most will have very little in the way of internal organs, and cellular differentiation is at a minimum, so most everything will be built up out of self-similar subcomponents. This makes them perfect for developing community organisms, where multiple species combine to serve different functions in the complete organism. We can probably squeeze quite a lot of diversity and interesting body plans out of these babies. For example, I've been toying with the idea of using Rangeomorphs to replace mollusks, but with a structure more like a heavy-duty Portugese Man-of-War jellyfish, where one polyp species would build a shell or body cavity, another would form tentacles, another would provide a digestive tract, etc.
Well, that's enough of me talking. Have at it, fellow SciFi Writers!