Assessing performance

[B+.]

I have often heard it said that colony A is better than colony B, or colony D is better than colony E. These are all subjective statements and have no value in bee breeding.
What do we mean, when we make statements like this? Do we mean one colony is better than another for honey production, disease resistance, swarming, docility, stability on the comb, etc? Well, we can only begin to make statements like this if we have records to back them up, but, one piece of information taken by itself is not sufficient. Let me show you what I mean...

In my 2018 test group, I have the following information on honey production (gross weight as at 15/8/2018):

DE-6-1-0456-2017 98.25
DE-6-1-0512-2017 133.65
DE-6-1-0535-2017 98.50
DE-6-1-0872-2017 116.10
DE-6-1-0884-2017 33.55
DE-6-1-1022-2017 93.75

Which would you prioritize? Most people would look at the highest weight and use this queen as a single "breeder" queen, but, is this the right approach?

Well, if I add information about the queens maternal heritage (pedigree) you can begin to look at this information in a more informed way.

0456 & 1022 are daughters of DE-6-1-95-2015
0512 & 0535 are daughters of DE-6-1-469-2015
0872 & 0884 are daughters of DE-6-1-82-2015

So, you can begin to look at average (mean) weight, as follows:

0456 (98.25 Kg) + 1022 (93.75 Kg) = 192.00 Kg (mean 96.00 Kg)
0512 (133.65 Kg) +0535 (98.50 Kg) = 232.15 Kg (mean 116.075 Kg)
0872 (116.10 Kg) + 0884 (33.55 Kg) = 149.65 Kg (mean 74.825 Kg)

Obviously, the more data points you have, the better. However, you can begin to see that daughters of DE-6-1-469-2015 performed better on honey production. These were all control mated to the same drones on an island (Neuwerk).

My point is: don't look at a single queens performance. Look at the entire sister group (if you can, look at the performance of queens tested in other areas too).

 

[Eyeman]

For the last 5 years mine have overwintered on their own stores without the need for any autumn feeding. I would take into account how much feed they needed when averaging my honey crop. At the moment zero.
Reasons for me to requeen include:
1. Bad Temperament & behaviour on comb runny,shaky, flighty and drippy
2. Disease: sac/bald/chalk brood
3. Poor Brood pattern
4. Poor Early season buildup
5. Filling brood box with stores when empty supers are in place
6. Mites levels: Just geting enought data from regular alcohol washes to make some meaningful conclusions
7. Honey production

All they attributes are recorded in some way at each inspection.

 

[B+.]

For the last 5 years mine have overwintered on their own stores without the need for any autumn feeding. I would take into account how much feed they needed when averaging my honey crop. At the moment zero.

It's a basic requirement in any scientific work that you interfere with the subject as little as possible. So, you don't feed a colony once the performance test starts. That is a given.
The alcohol wash, by itself, is insufficient to assess the varroa population of a colony. You're only counting phoretic mites with this and ignoring the reproducing mites inside the cells.
The idea was to show how anyone could begin to test their colonies for traits such as honey production. I've used real data from some of my test colonies but wanted to show how, the pedigree plays an important role in assessing groups of sister queens within the same apiary.

 

[B+.]

the pedigree plays an important role in assessing groups of sister queens within the same apiary.

Imagine if 0512 & 0884 had been sisters. How would this affect the results?

0512 (133.65 Kg) + 0884 (33.55 Kg) = 167.20 (mean = 83.60)

This would leave:
0456 (98.25 Kg) + 1022 (93.75 Kg) = 192.00 Kg (mean 96.00 Kg)
0872 (116.10 Kg) +0535 (98.50 Kg) = 214.60Kg (mean 107.30 Kg)

so, 0512/0884 now has the lowest mean performance and queens 0872/0535 has the highest. So, daughters of a different queen would be chosen

A simple change in the pedigree can affect the ranking. You can see how getting the pedigree right is so important now.

 

[For want of a better word]

Without any disagreement on your incredibly meticulous almost zealous record keeping or your choice of race of bee.. Dave Cushman gave a more generous ideology on how to access your bees...http://www.dave-cushman.net/bee/notbest.html
Yeghes da

 

[B+.]

Without any disagreement on your incredibly meticulous almost zealous record keeping or your choice of race of bee.. Dave Cushman gave a more generous ideology on how to access your bees...http://www.dave-cushman.net/bee/notbest.html
Yeghes da

I knew Dave Cushman very well. We were both at IINGRID and BIIG a long time ago.
I can honestly say; the BeeBreed protocols as described on Coloss (http://www.coloss.org/standard-methods-for-rearing-and-selection-of-apis-mellifera-queens/#3.) and in the German book "Selektion bei der Honigbiene" are much more comprehensive.

 

[DanBee]

Dave Cushman gave a more generous ideology on how to access your bees...http://www.dave-cushman.net/bee/notbest.html
Yeghes da

That's Dave Cushman talking rubbish, not common but not unknown either. I remember his contributions to the Irish beekeeping list - DC fell into some of the classic BIBBA pseudo-science traps, and recites several on this one page, including the 'single brood box' fallacy.

He is confusing general mongrelisation with hybrid vigour from crossing two unrelated lines, and is in essence stating that you can't improve performance by selective breeding because it's the hybrid vigour what done it. Oddly, he mixes this up with trying to say that you can only improve by selectively breeding once you know you have a fairly closely related population, via morphometry, which is of course deeply flawed when used as both a selection criteria and a measure of breed purity.

By proposing that the major selection criteria should be temper followed by calmness on the comb, he acknowledges the two main criticisms of the ease of management of Amm that we are not allowed to openly discuss

 

[B+.]

That's Dave Cushman talking rubbish, not common but not unknown either.

I think it's fair to remember that Dave's comments were written a long time ago. The "state of the art" has moved on considerably since then. If Dave had still been with us today, I'm sure he'd have adapted some of his views.

 

[Beefriendly]

.

0456 & 1022 are daughters of DE-6-1-95-2015
0512 & 0535 are daughters of DE-6-1-469-2015
0872 & 0884 are daughters of DE-6-1-82-2015

Using only 2 daughters to assess potential is not good statistically. I'm presuming you made the assessments from at least 10 and are just dumbing down the data to make it more easily understood?

 

[B+.]

Using only 2 daughters to assess potential is not good statistically. I'm presuming you made the assessments from at least 10 and are just dumbing down the data to make it more easily understood?

The recommendation is for family groups of 8. The graph (Fig 40, p 89, Selektion bei der honigbiene) shows there is very little improvement between a sample size of 8 and 50 (i.e. sister queens mated to the same drones)
Obviously, this is only a "snippet" of my data. Sometimes, it's necessary to "dumb down" the number of points to illustrate certain facts (my wife, who is a teacher, calls this "scaffolding")

I regularly see people refer to single queens they have bought in as "breeder" queens. I'm just trying to illustrate that testing is necessary to determine which queens deserve propagation. It doesn't matter where these queens have been bought from. Unless they are tested, there is no justification for propagating them

 

[Beefriendly]

Unless they are tested, there is no justification for propagating them

When you have spend a few £100's on a few breeder queens (isolated mated) and previous crosses of these lines are already known to produce good offspring then you have every justification in propagating from them. Of course there will be natural population variations and common sense comes into play. If she is laying badly you will think twice.
Currently I can't think of a single failure among the queens I've deliberately bred. All have been superb giving honey yields similar or greater to those you quote. But as I can only breed one generation before returning to breeder stock for their replacement, assessment of their potential for future breeding is non existent in my apiaries. F1 and then die.

I presume you are testing 8 daughters? 10 is the usual minimum number for any statistical test.

 

[B+.]

When you have spend a few £100's on a few breeder queens (isolated mated) and previous crosses of these lines are already known to produce good offspring then you have every justification in propagating from them.

Ok...You may be spending silly money on queens...I'm not. I just bought 2 II queens that are part of the Arista VSH programme for 75 euros each (plus 45 for postage)... ~£170 in total. It's completely unnecessary to spend more.
These are control mated...I only use island mated or II queens with proven ancestry.
If you breed from queens without testing them, that's up to you. I wouldn't!

 

[Beefriendly]

Ok...You may be spending silly money on queens...I'm not. I just bought 2 II queens that are part of the Arista VSH programme for 75 euros each (plus 45 for postage)... ~£170 in total. It's completely unnecessary to spend more.
These are control mated...I only use island mated or II queens with proven ancestry.
If you breed from queens without testing them, that's up to you. I wouldn't!

B+ sometimes you are really obtuse.
The difference between me getting 2 x 95 euro queens (Island mated from proven lines AND postage included)..... I was rounding up at £100.
Now let me see.... you bought two queens costing 75 euros plus 45 euro postage ...actually means your two queens cost 5 Euros more than mine.
And you tell me I am spending silly money on queens and you aren't....

 

[B+.]

The difference between me getting 2 x 95 euro queens (Island mated from proven lines AND postage included)..... I was rounding up at £100.

Good for you.
I was reacting to this

When you have spend a few £100's on a few breeder queens (isolated mated) and previous crosses of these lines are already known to produce good offspring then you have every justification in propagating from them.

I read that as a few hundred pounds per breeder queen, which I think is excessive. That's more like a reasonable price for II but still steep for an island mated queen.
In BeeBreed/AGT, there is a standard price of 75 euros for an II queen and 65 for an island mated one. These all come with pedigrees and performance stats going back through the generations.

 

[Beefriendly]

When you have spend a few £100's on a few breeder queens (isolated mated) and previous crosses of these lines are already known to produce good offspring then you have every justification in propagating from them..

Good for you.
I was reacting to this

No b+ you just misread it.

In BeeBreed/AGT, there is a standard price of 75 euros for an II queen and 65 for an island mated one. These all come with pedigrees and performance stats going back through the generations.

Good for all you little funny handshake members of BeeBreed/AGT.
Most of us are not members of this breeding organization and are happy to pay the going rate for an isolated mated queen. Seems we get themslightly cheaper than you do So YAH BOO to you

 

[B+.]

Seems we get themslightly cheaper than you do So YAH BOO to you

Not really. You said 95 for an island mated queen? I pay 65

 

[Beefriendly]

Not really. You said 95 for an island mated queen? I pay 65

And you need to add the extra 45 euros for postage....for a pair.

Two Island mated queens at 95 euros plus free postage = 190 euros.
You quoted 75 euros for II plus 45 euros postage.

I just bought 2 II queens that are part of the Arista VSH programme for 75 euros each (plus 45 for postage)...

Now I make that 195 euros. Perhaps you get your isolated mated queens for 10 euros cheaper from your organisation that non of us have access to...good for you.
Now stop being a silly boy and bragging about how yours are 15 euros cheaper than mine.
It's meaningless, 3 jars of honey. Small beer.

 

[For want of a better word]

I think it's fair to remember that Dave's comments were written a long time ago. The "state of the art" has moved on considerably since then. If Dave had still been with us today, I'm sure he'd have adapted some of his views.

******

****DNA techniques were in their infancy... and the latest state of the art Ion Torrent devices have given molecular biologists a fast and inexpensive new tool to better understand and define what is actually happening with our bees.... just waiting to see what the latest research into behavioral traits and DNA codes for them will show up!

At the moment we are no further forward than Elia and his biblical cronies producing spotty cattle!!!!!

Nos da

 

[Beefriendly]

. and the latest state of the art Ion Torrent devices have given molecular biologists a fast and inexpensive new tool to better understand and define what is actually happening with our bees.... just waiting to see what the latest research into behavioral traits and DNA codes for them will show up!

Hoppy they are just a faster DNA sequencer. Understanding what the DNA sequences do and how the genes they code for interact into behavioral traits is the complex slow bit.

 

[B+.]

You know, I tried to give a simple explanation so that beginners would get an appreciation of assessment. It's getting so that we can't pass on even simple things anymore.