[Nhcoll-l] VEROs

[Douglas Yanega]

On 8/13/21 12:44 PM, Samuel Bolton wrote:
>
> /Doug Yanega: /For example, almost any group of organisms that is 
> traditionally stored in liquid or on microscope slides is pretty 
> certainly going to be unsuitable for application of 3D imaging
>
> I must completely disagree with this, especially for freshly collected 
> specimens. High quality 3D models can be generated from slide mounted 
> specimens. Confocal microscopy is a good tool for this. Admittedly, 
> for older specimens it is trickier but certainly not impossible. And I 
> feel confident that advances will be made in this field. I am not sure 
> why you think that specimens stored in liquid cannot be used to 
> generate 3D models.
>
It's a matter of practicality (and probability). A fair proportion of 
extant invertebrate holotypes stored in liquid are too delicate to 
remove from their containers, and taking images of the specimens while 
still in their containers is difficult enough with regular photographic 
tools (dealing with image distortion, movement of fluid, lighting, and 
other complicating factors) - I doubt that applying 3D techniques will 
improve matters significantly under these circumstances. Many are badly 
deteriorated, or bloated, or shriveled. Also, you didn't respond 
directly to the basic point I made about comparability, perhaps because 
I wasn't quite as explicit as I could have been in the following way: if 
a type specimen is preserved in a manner different from specimens that 
you want to compare to it, or if it is in poor condition, then a 3D 
image of that type will neither make comparisons easier, nor will it 
improve your ability to use the specimen as a point of reference given 
its condition. This is *far more likely* to be true for specimens stored 
in liquid or on slides than it is for dry specimens. Just because you 
can MAKE a 3D image doesn't mean that image will be significantly more 
useful than a regular photo, and if it *isn't* more useful than an 
regular photo, then - for taxonomy - it isn't cost-effective unless 3D 
images are as cheap as (or cheaper than) regular photos. By extension, 
this applies to ANY species for which a conventional photo of the type 
is suitable for taxonomic work; if a regular photo will do, then you are 
arguing for something that isn't *needed*, nor cost-effective. Sure, if 
you have money to burn, then great, but much of taxonomy is a zero-sum 
game, where every dollar spent on X means a dollar less to do Y. Again, 
I agree that a 3D image of every holotype in existence would be great, 
but you haven't, I think, made a convincing case that 3D images offer a 
unique benefit that justifies the extra expense.

If we had wealthy patrons lining up to give millions of dollars for 
VEROs, then that money would arguably be better spent hiring people to 
process bulk sample backlog from biodiversity hotspots, and THAT would 
be the sales pitch we'd want to be making, because that's the biggest 
bottleneck in overall species discovery; not that images of types aren't 
*A* bottleneck, but that's not where the most serious impediments lie. 
Our biggest impediments are labor for processing, and expertise. If 
there are no new specimens to study, you don't need experts, and if you 
don't have experts, you don't need type images.

> /Doug Yanega: I won't deny that I have often longed for nice 3D images 
> of insects, but I've come to recognize that such images will probably 
> NEVER be a major tool for the practice of taxonomy./
>
> I don’t buy this argument. Advances in confocal z-stacking, 
> photogrammetry, micro-CT, etc. mean that high resolution models of 
> insects can now be generated. I admit that for insects, photogrammetry 
> is not always ideal, but advances are definitely occurring in this 
> field. I spend a considerable proportion of my time looking at 
> confocal 3D models precisely because 2D (based on DIC or phase 
> contrast) images just do not cut it.
>
I didn't say that it would never be useful, just that it would never be 
a *major tool*. SEMs were definitely a huge advance for certain 
disciplines of taxonomy, but it's not a major tool used by taxonomists, 
even now - most campuses I know of have an SEM facility, and the costs 
are not outrageous, but even though the technology is readily 
accessible, it's still used primarily in special circumstances or *when 
nothing else will do what is needed*. Likewise, I can't think of many 
situations where *only* a 3D image would serve a taxonomist's purpose. 
Something that is only used by a small number of taxonomists *could* 
still be important for those that use it, but it's still a minor part of 
the collective taxonomic toolkit.

> /Doug Yanega: Right now, with few exceptions, one institution can just 
> put a box of specimens in the mail to another institution, nearly 
> anywhere in the world, for little more than the cost of postage, and 
> no one bats an eye, including our administrators, even if it involves 
> thousands of specimens or potentially new species./
>
> But this is based on the understanding that specimens need to be moved 
> around, which is potentially risky when dealing with type specimens. 
> It seems better to produce 3D models and share them online.
>
"Better" isn't the same as "necessary". Insect specimens on pins are 
exceedingly fragile, and no one LIKES to put them in the mail, but - 
again - if that's the problem we are looking to solve, then you need to 
make the case that 3D images offer something that regular photos 
*can't*. Also, bear in mind that most loans are of unidentified 
material, or of synoptic material, and NOT of holotypes. You can't 
expect any entomology collection, when asked to send a loan of 20,000 
specimens of taxon X, to take 3D images of all 20,000 specimens. We are 
NEVER going to be free of the need to make loans.

Even if we limit the discussion to types: as it stands, only a moderate 
portion of the world's collections contain more than a relative handful 
of primary types, and of those, only a moderate portion have even been 
able to provide online access to ANY digital images of their holdings. 
Making that process *more expensive* and technically more challenging is 
not going to speed the process up. The insect collection I manage is the 
20th largest in the US (with some 4 million specimens), and it took us 
until 2016 to get money to hire a grad student to take digital images of 
our types using an automontage system belonging to their faculty 
adviser, who had bought the system on a grant. Failing that, we would 
STILL not have any type images online. I really do get where you're 
coming from, but when you're fighting an uphill battle, you look for 
things that will make the road easier, not harder.

Peace,

-- 
Doug Yanega      Dept. of Entomology       Entomology Research Museum
Univ. of California, Riverside, CA 92521-0314     skype: dyanega
phone: (951) 827-4315 (disclaimer: opinions are mine, not UCR's)
              https://faculty.ucr.edu/~heraty/yanega.html
   "There are some enterprises in which a careful disorderliness
         is the true method" - Herman Melville, Moby Dick, Chap. 82

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.yale.edu/pipermail/nhcoll-l/attachments/20210813/eddb1b10/attachment.html>