We posted last week about the 95 million year preservation in amber of plants and insects in recently discovered samples of ancient amber in Africa. This showed the early distribution of recognized types of organism very long ago in Africa, affecting reconstructions of paleoecology, since at least some had not formerly been known to be there.
That was interesting, and we raised another broader point that these specimens showed. It has to do with the conservation of form over evolutionary time. It's one thing to look like your granny, and quite another to look like your 95-million-times-great granny!
The evolutionary problem (and it's not a new one nor one we invented) is to explain the very ancient existence of species that look strikingly, perhaps entirely, those alive today. If (as we recently discussed) even a few hundred thousand years is enough for anthropologists to insist on naming a fossil as a new species--or even genus!--then how can something 95My old look so strikingly like what we see today?
One might think that insects, because they have a smaller genome than vertebrates, have less genetic 'room' for variation, and are more constrained by selection so that organisms like flies have no variation left--no ability to viably change form. Bacteria have maintained their form since literally the most ancient of times (fossils called stromatolites, ~3Bya old). But even bacteria have multiple genes contributing to traits, and insects and plants certainly to, too. Selection studies in these types of organisms, and mapping studies on even such exotic traits as startle reflexes, wing venation, bristle count, and sleep habits in flies show genetic causal complexity due to existing variation in their natural populations.
Thus, with many contributing genes, each susceptible to variation, one might expect plenty of room to vary, ever so slightly, over millions of years. Furthermore, these groups have continued to evolve diverse species so that there wasn't just one way, say, to be a fly that, once installed, was so tightly fenced in by natural selection as to be preserved for time immemorial. In genetic terms this would generally called an adaptive peak on the genetic fitness landscape, that the species simply couldn't get off. But this is a post-hoc argument that seems just too pat, without some further justification (speculation about this has certainly been offered, but is difficult to prove).
Anyway, the conservation of form and function, as well as its adaptive evolution present important problems that are not yet fully solved. You look a lot like your granny, and we have a clear understanding of why that is. But you don't look exactly like her, and for a fly or fern to look like its countlessly-great-great grannies is harder to explain.