Thanks for the reply, but I remain unconvinced. These are not reasons why someone would WANT to make a skeletal more schematic than necessary as much as they are excuses why people DO make them more schematic. As you said in your original post, a schematic reconstruction "emphasizes key features of the anatomy (sometimes literally "emphasizing" them) while perhaps demonstrating other key information, such as which bones are preserved." So for things like diagrams showing biomechanical calculations, we'd all agree they're fine. But just what is the supposed emphasis in Martinez and Alcober's Panphagia recon that makes its schematicity preferrable, to use your example? My guess is that they fully intended their reconstruction to be realistic and would not have positive answers if confronted about the various ways it is not.

Scott,

I think I would disagree that realistic should be used in lieu of schematic. I can think of several reasons to use schematic diagrams of bones, especially when discussing an element that cannot or should not correspond to a single organism (labelling the various fenestrae of the skull in anticipation of tetrapod evolution), or because the material being discussed is incomplete.

That our artistic depictions of skeletons and bones, barring photography, are to a degree necessarily schematic, we cannot even divorce ourselves of this. We already adapt the image of the material we see into ink and graphite or digital line when necessary, and this always (!) results in loss of possible data, especially when we make outlines. It should matter also that when we place skeletal diagrams in phylograms, to show the body forms in progression on a tree, we are being necessarily schematic, even if the skeletons (at 1.5 inches or less!) are so small that a simple silhouette would be preferable.

@Mickey - I'm not trying to convince you that schematics are superior, I'm trying to convince you to start first with encouraging authors to universally adopt proper labeling.

In the long run I think it's preferable for skeletals to be realistic, but it's not fair (or realistic) to expect authors to adopt that convention when there's no real guidelines for how to achieve it.

As for Martinez & Alcober's Panphagia - my guess is they simply never considered it at all, like most authors in the history of the field. Asking for authors to label their diagrams will raise awareness of the issue (i.e., if they want a realistic skeletal they need to put in the effort to get one) and reduce systemic errors that are created when others assumed an unlabeled skeletal is realistic.

@Jaimie - I agreed with much of what you wrote in the first post on this subject. But just because all illustrations are to some degree schematic is not at all the same as saying that there's no difference between a skeletal that is properly scaled and accurately reproduces the outlines of bones, and those which do not take care to attempt this. My biggest concern right now is that people label their intentions in professional articles, so as not to accidentally mislead readers.

The ultimate issue of all skeletals being illustrations is why I use the term "realistic" rather than "real" or "literal". Regardless, a realistic skeletal is a very different beast than a schematic one is, and they ought to be labeled, just like any salient data in a figure or diagram.

Seeing how much interpretation is involved in every stage of 'real' skeleton preparation - ie. going from excavation to description of a bone as characteristic of a specific species - I've become convinced that an abstraction of the results is desirable. Particularly in 3D, the representation of the skeleton needs to focus on the defining characteristics and not on the surface characters, which can be called up in the more realistic scanned meshes.

@David - Agreed - any representation of data is an abstraction; even photography is, with its ability to alter lightinng show off (or hide) details, exposure settings, etc.

When it comes to realistic vs schematic it's an issue of intent - if what is being abstracted are the proportions or the shape of the bones, then it's schematic and should be labeled so other workers aren't mislead.

To be fair, there probably are additional classes of portrayal here. For example in 3d you can simplify bone models without making their proportions inaccurate. In the case of peer review they probably deserve their own label ("simplified rendering of taxon X" or some such), but I was intentionally addressing the far ends of the spectrum here, which also happen to be the most common forms in professional papers, although that may change in coming years.