1

Topic: Carabus cancellatus/granulatus

I posted this in the Species dictionary Forum yesterday, but Chris reckons it is a problem with the Import Wizard, so here it is again.  I would tend to agree with Chris, as there is no problem evident when you enter a species directly, nor apparently in the synonymy on the Gateway.

Chasing up a rogue record of C. cancellatus, today, which should have been granulatus, I find a problem that has affected a number of providers, I think.  The R6 import wizard, taking in Carabus granulatus, defaults without prompt or highlight to the 'sensu Stephens' version, which is returned as part of C. cancellatus.  That explains my single rogue, and will probably explain the mass of other cancellatus (which according to John Walters has not been seen since 1895) on the Gateway map.

I think this must have been OK in an older version of R6, as it only affected a record I imported last summer, and not those imported previously.

Murdo

2

Re: Carabus cancellatus/granulatus

To amplify what I have discovered about this, I have had confirmed by two other R6 users that they have had exactly the same problem - granulatus goes in, cancellatus comes out.

M.

3

Re: Carabus cancellatus/granulatus

I also have several records where the taxon common name is Carabus granulatus but the taxon latin name is Carabus (autocarabus) cancellatus. Imported over the last 2 years using the BEETLES checklist. Others imported using Recorder 3.3 checklist are correct

Gordon Barker
Biological Survey Data Manager
National Trust

4

Re: Carabus cancellatus/granulatus

This is just to nudge gently as the problem seems to have escaped significant notice.

M.

5

Re: Carabus cancellatus/granulatus

Same here with three records imported using the BEETLES checklist - now corrected.

Mark Pollitt
SWSEIC (formerly DGERC)

6

Re: Carabus cancellatus/granulatus

I can see why this is happening, but not do not have a solution. Will need to follow up with Chris to see if we can resolve this via the Dictionary or if we need to change the way the matching is done.   In the Bettles checklist Carabus granulatus occurs just the once so matching will assume that this is a  safe match. What it is ignoring is the fact that Carabus (Carabus) granulatus is also on the list.

Mike Weideli

7

Re: Carabus cancellatus/granulatus

Thanks for this, Mike.  The obvious question is 'How many other similar cases exist?'.  In some of the more fluid and difficult taxa (I don't include Carabus in that category) I would be surprised if there were not parallels - at least if subgenera are implemented.

I think your explanation suggests that doing away with the subgenus would solve the problem for all affected taxa at a stroke.  Without the subgenus, an attempt to import 'Carabus granulatus' would provoke a prompt that there is more than one match, and the user would be forced to make the choice.

M.

8

Re: Carabus cancellatus/granulatus

On the surface of it the fix is fairly easy in that all we need to do is to test for duplicates after removing the sub-genus from both the name being matched and the names in the Dictionary.  What is difficult to predict is the effect on performance of implementing this, but it is a change I can make, so we can give it a try.

Mike Weideli

9

Re: Carabus cancellatus/granulatus

Have now run some test on having made temporary changes  which removes the sun-genus before looking for duplicates. Tested using an import which contained every species on the Beetles list. This was  matched against the Beetles  list. The number of taxa which failed to find a match using the current approach was 221, while the number failing with the changed code  was 376.   The probably doesn't mean that 155 species are being incorrectly allocated, but potentially some could be.   Additional time taken on my machine was 20 seconds. We can probably get this change into R6 with the release in May, but if anyone would like to give it a test I can provide the SQL - Mangement Studio will be required to install the change and you need to be on R6 V6.18

Mike Weideli

10

Re: Carabus cancellatus/granulatus

I hope you were not up all night tackling this, Mike!  The potential problems will apply to all the taxa where subgenera are used (Hymenoptera and Heteroptera for example).  When the solution is implemented, would it be possible (I would say it would be essential) to provide a list of the misallocations that might have occurred so that users can check and change if need be?  I assume that if the dictionary lists are not modified, corrections will not happen automatically.

M.

11

Re: Carabus cancellatus/granulatus

I attach the list of possible problem taxa for  all the database. If required I can do an xml report which will identify where they have been used for determinations.

Post's attachments

QryPossibleMatchProblems.xls 203.5 kb, 12 downloads since 2013-03-09 

You don't have the permssions to download the attachments of this post.
Mike Weideli

12

Re: Carabus cancellatus/granulatus

MikeWeideli wrote:

If required I can do an xml report which will identify where they have been used for determinations.

Mike, that would be very useful

Charlie Barnes
Information Officer
Greater Lincolnshire Nature Partnership

13

Re: Carabus cancellatus/granulatus

Thanks, Mike.  Although that is a horrifyingly long list, the actual number of affected taxa will be much smaller as I think the Carabus instance will only be replicated if Genus (Subgenus) speciesA has an obsolete synonym identical to a valid current taxon Genus speciesB.  I have not noticed any problems with the aculeates, for example.

M.