I've thought about doing that and even played around a bit, but like all things it's just a matter of time and priorities.

Copying the plates is one thing, but the legends are on adjacent pages. Ideally they would be searchable text rather than just an image of the text, but who has time to enter all that, plus trying to update the taxonomy as appropriate. I tried some OCR software one night to see how well it would do at it, but even after identifying the font, it did a terrible job.

One of these days I'll get a flatbed scanner, then maybe I'll at least copy the plates. We can always refer to them from the guide pages. There's quite a few useful B+W illustrations in there as well.

I've got a flatbed scanner and would be willing to try it, but I'd be apprehensive about damaging a 100 year old book!

It might be possible to get the Distributed Proofreader project involved. I played around with it a bit, and it involves lots of humans proofing and fixing the OCR results. They put stuff into Project Gutenberg.

Someone might be working on it already in some capacity, this page indicates that someone cleared the copyright with Project Gutenberg before working on it. But it was over a year ago and they offer no contact info.

It's definitely a job, but it would be great to "unlock" that info somehow.