Finally my time to shine haha ! Professional paleontologist here, the answer is “yes, but”
We do undersample EVERYTHING solely due to how rare and exceptional the fossilization process is. We especially undersample anything other than hard mineralized tissues, so yes, a lot of metamorphosis is forever lost, but that’s a drop of water compared to the universe of soft bodied creatures that we will never get to know…
However, we are not that clueless, for several reasons. First, using a kind of uniformitarian thinking, we can speculate that ecological communities were similarly structured to those we know today, and as such estimate what we’re missing, particularly how much we’re missing. But also, and perhaps the most exciting, we do discover new things everyday, and I don’t think that people grasp how much our discipline has evolved and been changed by new technologies in the last 20 or so years. Scans and 3D imaging techniques allow us to examine the most minute details of fossils previously hard or impossible to study, and methods such as Synchrotron powered X ray fluorescence allow us to unravel unprecedented levels of details and information on soft tissues. I had the opportunity during a research conference to discuss the results of a young researcher working on amphibian metamorphosis on specimens about 300 million years old, using these technologies. He was able to study very early stages of development, and observe structures such as intestines, eyes, and even the various layers of skin !
In conclusion, yes we definitely undersample metamorphosis, but we probably don’t underestimate it that much today, because we can estimate how much we’re missing. Furthermore, new technologies allow us to study unprecedented levels of detail, including unseen before soft tissues.
Oh thank god. This is about professional interpretation of a very vast and complex dataset. If you didn’t answer it there was going to be a lot of trash and guessing in the replies.
As one of people I can tell you that they dont. (But this might also be a small sample size)
I still can’t quite wrap my head around it. I remember from my childhood that it was a big deal when they found samples of dinosaur skin and eggshells. So I wonder, is this information simply “hidden” in the rocks surrounding the fosil bones, and can now be examined with these new techniques? Or does it take an exceptionally well-preserved sample that can now be analyzed in much greater detail, revealing more information?
It’s both, as in not all fossils preserve things other than hard tissues, but new methods allow both deeper examination and understanding of known exceptional preservation, and discovery of previously unknown soft tissue fossilisation. And anyways, finding any fossil is still as cool as it ever was ;)
This is a legitimate concern and has been addressed to some degree in some areas. Unfortunately we don’t have a perfect way of knowing that a specific specimen is from a specific species. Two very similar skeletons could be from the same or closely related species. The same goes for development over the life history of a specific organism. Adult humans have a different skull to height ratio to babies, but the ratio between toddlers and young chimps is very similar.
Fortunately we have many different aged animals of the same species in the same context to compare. We can see the infant, child, adolescent, adult, and aged forms for many species and this acts similarly to transitional fossils, they help close the gap. We can be more sure with more hints like sharing a space, being buried in the same context, having the same nitrogen isotope ratios in teeth, and eating the same prey. Lots of other things can act as clues to the relationships and make us more or less certain of a given relationship.
That said, fossilisation is rare. Not all that many individuals will be fossilised. Different types of tissue fossilise to different degrees and in some cases not at all. If an animal is mostly spongy material they may degrade too fast to fossilise and preserve structure. Other examples may only leave their imprint as a hollow or pressing of one material into another. I think the record is very sparse and will remain so, but adding more example allows more connection and conclusions to be made.