[GregV]

Count me as one impressed.

To be sure, Russ, I only did some auto mechanic-level hacking (not the car design).
I will not pretend to understand the complete logic behind the algorithms.
No need - smart people have done the smart work already for us.

Will keep reporting.

 

[GregV]

This is a good enough picture for land-marking.
BS attachment actually lost some zooming ability - in actually it is better.

 

[GregV]

This is not a good picture for land-marking (cheap scanner output).

 

[Gray Goose]

Hi Greg,

Let Us know when we can Send you 30 bees and get a report...


looks like a fun project.

GG

 

[GregV]

BTW, per the documentation I read so far - takes 24 wings to accurately identify a makeup of a colony.
So, per the same doc recommendation, I go for 30 wings per a colony.
This gives some room for failed identifications and still produces good enough set to profile a colony.

Per the alternative methodology:

2.2. For a rough estimate of the family in the first year of life of the uterus (preliminary estimate), as a rule, requires 30 working bees. The exact assessment of the family on the basis of (full study) requires a collection of at least 50 bees, in special cases and 100 bees. For the assessment of the parent's family it is enough to study 50 drones.

I guess, the more the merrier.

 

[GregV]

Hi Greg,

Let Us know when we can Send you 30 bees and get a report...


looks like a fun project.

GG

Hopefully, we figure this out and I just share the tools and the documentation (IF/ONCE I get the permission from the tool designers).
Else, I will need to charge for my time!

 

[Gray Goose]

Hopefully, we figure this out and I just share the tools and the documentation (IF/ONCE I get the permission from the tool designers).
Else, I will need to charge for my time!

Well I did not presume it would be free. Was thinking of a side line for you
But sure if the camera is not too costly I can just do it here.

would be nice to know what works in case of a bear wipe out one could recreate.

GG

 

[GregV]

Well I did not presume it would be free. Was thinking of a side line for you
But sure if the camera is not too costly I can just do it here.

would be nice to know what works in case of a bear wipe out one could recreate.

GG

I hear you.
Side-line deal sounds good too.
Pretty much any camera with macro-setting can do this (mine is about 5-6 years old mid-range cam).

Once I figure this out I might advertise my services too - LOL!
There is a long winter to go yet.
No rush.

 

[GregV]

Speaking of IdentiFly....
I will not say this is a practical solution for a practical beek - just too much work.

So yesterday I re-did the land-marking for my set #1 (after some painstaking photo manipulations to get the Comp03 image stiched together from 10 separate camera pictures).

8 points x 30 wings = 240 points to be done (required for the Russian/Ukrainian programs).
This much one can tolerate - granted the output analysis is good enough for some practical utility. So that is where I will continue my hacking.

3 main indexes per a wing that will allow for "good enough" probability determination for the main groups of interest should be sufficient.
The main groups of interest IMO - Millifera, Caucasica, Sossimai, Ligustica, Carnica, and (ideally) Scutellata.

The same for the IdentiFly will take:
19 points x 30 wings = 570 points

(PS: though IdentiFly doc says "at least 10 workers" - unsure how sufficient that is;
just feels too low;
maybe the high # of the land-marks points per the wing is to make up the low wing #???
feels to me more wings/fewer points per a wing is a better way for non-academic uses).

240 points vs. 570 points to get a rough colony make up estimate.
AND IdentiFly requires following the special land-marking rules too.

The Rus/Ukr tools feel more like AK74-type weapons - less maintenance work required and fewer parts, but they will still shoot in the same general direction and still hit the target.

Not doing the IdentiFly. Not worth it to me. Too much work and overkill.
IdentiFly is for full-time academia, not for practical beeks and such.
That's my take.

 

[GregV]

I verified the MorphoXL tool from Ukraine is working as-is on my system and so should for everyone else too.

IF/ONCE I get permission from the owner of it - will just hang it out here.
The version I got is already pre-set to English and has a full PDF with the usage instructions (Google-translated, clearly - FYI). The tool is free for non-commercial use, but just making sure.

 

[GregV]

Here is the Ukranian tool (the public version):
Morphometry.rar

I will be working with the designer to help with the English documentation and the program improvements.
I already asked for a provision to include additional, user-defined bee races to the analysis (e.g. Scutellata could be added as a user-defined race to the standard races).
But this is a free product and up to the producer what he will do - he does not charge for it.
If any feedback, let me know.

 

[GregV]

So, guys, does anyone know values for A. M. Scutellata similar to these used by the MorphoXL?

The best I found so far is just CI from Ruttner - not good enough.
Still need DsA and Hi ranges from reputable sources

• A. m. scutellata 2.52±0.46 (Ruttner 1988, Tab. 12.2) [3]

 

[HaplozygousNut]

Of course, it is a hybrid.

However - a hybrid with certain average proportions of the original bees.
For example, one of the inputs of the Primorsky bees was the Ukrainian Steppe bee (Apis mellifera sossimai)
And of course, AMM, Ligustica, Caucasica (possibly Carpathica).

So if none of these together are detected in a sample (like I tested above) - these are not really Primorsky bees, seems to me.
However, if significant presence of AMM/AMS are detected (I have not a clue how much) - then we have something to think about.

And so, be great to obtain a morpho-analysis done directly in the Primorsky region and see the original source to get some idea what to look for. Not to mention, they have local sub-populations - upland bees/lowland bees/coastal bees.

Of course, this is a very rough tool and not a replacement for genetic analysis.
But hey - it is free, everyone can do it, and it gives some info (vs. no info).
Especially, when someone sells "Russian" bees those "Russians" should resemble the Primorsky bees. I mean to ask for samples of dead "Russian" bees on my local bee forum and run them through.
Will see what turns up.

GregV, I am excited that you are using the wing veining to try to identify the bee subspecies of your area in Wisconsin. About whether Russian bees would retain the same proportion of subspecies in its genes, now that they have been in the US for a while; even if the line of Russian bees have been kept pure without any outside introductions of other bee genetics, the Russian bees may change to become genetically a larger proportion of the more southern climate adapted subspecies of bees, (like the Italian bee), because the Russian Far East has very cold winters compared to most of the US here.

 

[GregV]

GregV, I am excited that you are using the wing veining to try to identify the bee subspecies of your area in Wisconsin. About whether Russian bees would retain the same proportion of subspecies in its genes, now that they have been in the US for a while; even if the line of Russian bees have been kept pure without any outside introductions of other bee genetics, the Russian bees may change to become genetically a larger proportion of the more southern climate adapted subspecies of bees, (like the Italian bee), because the Russian Far East has very cold winters compared to most of the US here.

Well, winter or not - less important.
The annual bee selling runs largely negate much of local adaptations here.

The Russian Bee Breeders Association should be maintaining the stock as close as possible to the original import.
If they do any good job, the breeding queens should be having the profiles suitably close to the originals.
And with that, the lineage should be trickling down to the end buyers (if the re-sellers are any good and honest).

Basically, why I would like to morpho-test any bees labeled or suspected - "Russian".
Then keep the better ones for myself.

 

[William Bagwell]

Here is the Ukranian tool (the public version):
Morphometry.rar
(snips)
If any feedback, let me know.

Thank you for finding and sharing this!

Be awhile before I have time to learn to use it. Also doubt my current "all in one" is as good as the flat bed scanner I used to use. Will say that Morphometry will run under Wine on Linux. A bit glitchy and do not know if every thing works yet. Will also test it with Crossover (paid version of Wine) once I decide whether to buy a new version or just re-download the last one I purchased 7 years ago.

 

[GregV]

Will say that Morphometry will run under Wine on Linux.

Thanks.
The program designer did say this program is supposed work on Mac OS too.
He asked me to test it, but I don't do Mac OS.
Do share your experience on Linux (for feedback to the designer).

 

[William Bagwell]

Do share your experience on Linux (for feedback to the designer).

Will do. Found my last two purchased Crossover versions (11 & 13) so only have to recover my license key. Will eventually buy the latest... (New computer needs to be updated first.)

Mageia 5 x86_64, Wine 1.7.30 Morphometry opens and will import pictures and mark points. Menus open as 'black boxes' but display as you scroll down. No bee wing scans yet so used a dragonfly wing as a substitute.

 

[Gray Goose]

Well, winter or not - less important.
The annual bee selling runs largely negate much of local adaptations here.

The Russian Bee Breeders Association should be maintaining the stock as close as possible to the original import.
If they do any good job, the breeding queens should be having the profiles suitably close to the originals.
And with that, the lineage should be trickling down to the end buyers (if the re-sellers are any good and honest).

Basically, why I would like to morpho-test any bees labeled or suspected - "Russian".
Then keep the better ones for myself.

Greg,

As I understand re: the breeders keeping the line "intact"
The Russian breeders need to send in bees from each of the hives they intend to use for breeders in the spring, By fall (before winter freeze up)
These are then "tested" to determine if the queens are pure, and "allowed" to be breeders in the spring.
The person I spoke with was a breeder, he stated he needed to send in 10-12 hives of bees to test ,to have 6 he could use in the spring, at some cost to himself. He would lose 1 or 2 to winter, and if the survivors would supersede or swarm before breeding also loose a couple that way.

I am not sure how these were tested, I could look into it, but likely it would be "trade secrets" and not sharable.

there "may be wing data somewhere used for this, unless they smashed and used a DNA test.

GG

 

[GregV]

The local bee importers started the annual spam.
Here is one message.

I will be offering splits from my bee yard. Nucs and package bees from California and Georgia. Queens, bee equipment, and advice. ........ Package bees start early March and into May. Splits and nucs will be mid May. ......

Another dump of almond bees is coming.

 

[HaplozygousNut]

The local bee importers started the annual spam.
Here is one message.



Another dump of almond bees is coming.

LOL. Non-local commercial bee genetics mixing is like spam in mail when you are trying to identify your local bee stocks!