To continue this theme, the fundamental problem here and that is that we are losing taxonomic expertise through retirements and deaths, and not training new. This problem has been building for decades but its reaching critical proportions. Some insect groups benefit from having a healthy population of amateur taxonomists, but others that require specialized collecting and optical equipment do not have this backup.

I don't know what the solution is, but with grant funding going almost exclusively to molecular or ecological projects and not supporting alpha taxonomy no student is going to train in taxonomy if they can't get jobs with this training. Certainly universities aren't hiring taxonomists per se.

Digitizing everything is great but at what point does it become garbage in, garbage out. More emphasis needs to go to actually identifying specimens so that we can harvest data on environmental change and all the other kinds of data represented by insect collections.

Probably no one outside our community understands the issue, which is why we are the ones who are going to have to make a case for increased funding and training.

Lynn Kimsey
Bohart Museum of Entomology

From: Entomological Collections Network Listserve <[log in to unmask]> On Behalf Of Neil Stanley Cobb
Sent: Tuesday, June 23, 2020 6:39 AM
To: [log in to unmask]
Subject: Re: The bee challenge

There are many challenges concerning identification but in my opinion, it is just wrong to only digitize specimens that have been identified to species by an "authority". If you are not committed to having every specimen in your collection identified to the best resolution possible then it is just taking up space. If there is any hope of achieving Quentin Wheeler's challenge of identifying every species on the planet by 2050 then we need to start thinking outside the box. Nobody in 2020 should be sending out loans for identification without digitizing the specimens. It should be unacceptable to not link an identification back to a specific specimen. The cost of digitizing is small compared to have specimens sit in a drawer for decades taking up space. Digitizing does increase our ability to have more specimens properly identified faster than letting them sit in a unit tray. Digitization of material should make it easier for people to have revisionary work funded.  Digitization projects should typically prioritize, but anything that leads to checkerboarding in the digitization process should be avoided.

From: Entomological Collections Network Listserve <[log in to unmask]<mailto:[log in to unmask]>> On Behalf Of Doug Yanega
Sent: Monday, June 22, 2020 2:50 PM
To: [log in to unmask]<mailto:[log in to unmask]>
Subject: Re: The bee challenge

On 6/22/20 12:55 PM, Leblanc, Luc ([log in to unmask]<mailto:[log in to unmask]>) wrote:
Greetings friends,

As most of you are aware of, a NSF bee digitization proposal is being developed, and many of us are involved. We hope that we can extend achievements for our museums beyond just digitizing specimens. That should include assignment to species of part of our unidentified material. Many of us will have to rely on our limited experience and on locally available expertise to achieve determinations. The realistic goal would be to give priority to genera that can be more easily sorted to species. To establish a scale of priority, I would appreciate some advice from experienced bee identifiers in ranking by level of difficulty (in identifying to species level) each of the genera represented by large numbers of unidentified specimens (based on the Barr Museum). I propose ranking each genus from 1 to 4. I can safely assign level 1 for Bombus and level 4 for Lasioglossum and Nomada, but I am unsure about the other genera. Your insight would be greatly appreciated, and potentially help us all establish achievable goals for bee identification. I attached a spreadsheet with the list of the following genera:

Andrenidae: Andrena, Panurginus, Perdita

Apidae: Bombus, Ceratina, Diadasia, Eucera, Melissodes, Nomada, Triepeolus, Xylocopa

Colletidae: Colletes, Hylaeus

Halictidae: Agapostemon, Dufourea, Halictus, Lasioglossum, Sphecodes

Megachilidae: Ashmeadiella, Hoplitis, Megachile, Osmia, Stelis

If we set aside the one or two or three people alive that are actual experts in these respective genera, it seems to me that everything is going to be either a 1 or a 4, based on the standards you give. Only Bombus, Xylocopa, Agapostemon, and Halictus (and maybe Diadasia) can be *reliably* and *consistently* IDed by non-experts without recourse to exhaustive synoptic collections. The remaining genera either have no recent keys or no keys at all, and even if they have keys, those keys generally require you to have specimens of every single species in the key in order to be *certain* you have arrived at a correct ID. You can run any Colletes or Melissodes or Perdita or Andrena through a key, but the odds of arriving at the *correct* ID is extremely small unless you have authoritative reference material at hand.

Even this crude dichotomy has exceptions; it is impossible to *reliably* ID the "subspecies" of some of the Xylocopa, for example, while it is actually fairly easy to ID members of the subgenus Lasioglossum, because McGinley's revision contains such an excellent illustrated key, and Gibbs' recent work has made at least Eastern "Dialictus" almost tractable. The list is also missing some notable genera such as Anthidium (which might merit a 1) and Anthophora (which would be a 4).

As I was co-PI with John Ascher of the first large multi-institutional bee digitization project, I can personally attest that having trustworthy IDs is essential, and also not trivial. It did in fact greatly affect the prioritization criteria, such that only taxa and specimens for which we had ironclad, definitive expert IDs were databased. The problem we faced is the same problem you face: the digitization grant does not provide funding for the creation of expert revisionary work, and that means that some genera are going to be nearly impossible to digitize, because for most collections, only legacy material of certain genera will have expert IDs. Without better keys, there are collectively hundreds upon hundreds of species in the genera you list that cannot be reliably and consistently IDed right now, so there's an inherent "cart before the horse" issue to contend with, and it's not at all simple.



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)


  "There are some enterprises in which a careful disorderliness

        is the true method" - Herman Melville, Moby Dick, Chap. 82