Page 5 of 8 FirstFirst ... 34567 ... LastLast
Results 81 to 100 of 156
  1. #81
    Join Date
    May 2009
    Location
    Canterbry, UK
    Posts
    1,656

    Default Re: An Option Towards Developing Treatment-Free Bees

    Quote Originally Posted by WLC View Post
    REVIEW ARTICLE
    Standard methods for characterising subspecies and ecotypes of Apis mellifera
    Marina D Meixner1*, Maria Alice Pinto2, Maria Bouga3, Per Kryger4, Evgeniya Ivanova5 and Stefan Fuchs6

    https://bibliotecadigital.ipb.pt/bit..._etal_2013.pdf

    Yes, it can be challenging.
    There you go. Page 6:

    3. Available methods and markers
    3.1. Morphometry
    There is no morphological "key" to honey bee subspecies, no simple logical tree based on a sequence of single discriminating characters. Instead, measurable morphometric characters show gradual changes and their ranges mostly overlap between subspecies. Thus, subspecies often differ only slightly in the mean values of several body characters, and therefore advanced statistical methods are required for discrimination of groups. The concept of numerical taxonomy was introduced into honey bee taxonomy by DuPraw (1964, 1965) and further elaborated by Ruttner et al. (1978).
    [1]


    The table on page 7 shows 31 different markers, and with those "advanced statistical methods are required for discrimination of groups"

    I would imagine that the more markers you have, the less challenging is the math required to reach a (more probable) result. Given, however, that beekeepers have been selecting for particular wing pattrerns for some time, I would guess that one is close to useless!

    From the same report, the bit we should really take notice of:

    "Honey bees show considerable geographical variation, resulting in adaptation to regionally varying factors of climate and vegetation, but also to prevailing pests and pathogens. However, this natural heritage is increasingly subject to diffusion by human beekeeping efforts at a worrisome speed. The demand for high economic performance of bee colonies, combined with desirable behavioural characteristics, has led to considerable changes caused by systematic bee breeding. Thus, the original geographic distribution pattern is being dissolved EU-wide by mass importations and an increasing practice of queen trade and colony movements. These activities endanger regional races and ecotypes by promoting hybridisation (De la Rúa et al., 2009; Meixner et al., 2010), and by adding various breeder lines with distinct properties to the picture. Yet another dimension is added by the deliberate replacement of native subspecies in some regions by non-native bees with more desirable characters and greater commercial interest (for instance, the replacement of A. m. mellifera in northern and central Europe by A. m. carnica or A. m. ligustica) (Bouga et al., 2011).

    The downside of these economically driven processes is an increasing trend towards uniformity of honey bee populations across Europe, leading to a loss of both genetic diversity and specific adaptations to local conditions (reviewed in De la Rua et al., 2009; Meixner et al., 2010).

    Honey bees are particularly sensitive to inbreeding (Seeley and Tarpy, 2007 and references therein). Therefore, the loss of genetic diversity is of grave concern. It has been shown that colonies with reduced genetic diversity are less capable of controlling hive temperature (Jones et al., 2004) and more prone to develop diseases when challenged by parasites (e.g. Tarpy, 2003). This reduction in genetic diversity may also affect the capacity of honey bee populations to adapt to new threats, such as newly introduced parasites like varroa.

    Thus, there is a widely recognised need to encourage regional breeding efforts to preserve local adaptation, and to maintain local strains in isolated conservation apiaries."


    Back to where we started: what matters is that we help the local variants - of whatever ancestry - to adapt to varroa, and thus thrive and preserve their diversity. To say that another way, we stop treating and/or manipulating against varroa.

    Mike (UK)
    [1]
    Standard methods for characterising subspecies and ecotypes of Apis mellifera
    Marina D Meixner1*, Maria Alice Pinto2, Maria Bouga3, Per Kryger4, Evgeniya Ivanova5 and Stefan Fuchs6
    Anti-husbandry: Medication + Reproduction = Continuing Sickness
    http://www.suttonjoinery.co.uk/CCD/

  2. #82
    Join Date
    May 2009
    Location
    Canterbry, UK
    Posts
    1,656

    Default Re: An Option Towards Developing Treatment-Free Bees

    Quote Originally Posted by WLC View Post
    Here's an abstract from a paper in press that suggests that GWV can be both automated and done online:

    http://www.sciencedirect.com/science...74954113001222

    "Several studies have shown that features extracted from patterns of bee wings are good discriminatory elements to differentiate among species, and some have devoted efforts to automate this process. However, the automated identification of bee species is a particularly hard problem, because (i) individuals of a given species may vary hugely in morphology, and (ii) closely related species may be extremely similar to one another. This paper proposes a reference process for bee classification based on wing images to provide a complete understanding of the problem from the experts’ point of view, and a foundation to software systems development and integration using Internet services. "
    I don't think any of that says 'wing analysis alone tells you something about an individual's subspecies identification' - and if it did then it would be in direct conflict with Meixner et all (as just cited)

    It seems to me to be talking about automating a process - that may then be of use to to people who know what they're doing, and know not to read (far) too much into a single morphometric.

    Has anyone actually read the paper?

    Mike (UK)
    Anti-husbandry: Medication + Reproduction = Continuing Sickness
    http://www.suttonjoinery.co.uk/CCD/

  3. #83
    Join Date
    May 2009
    Location
    Canterbry, UK
    Posts
    1,656

    Default Re: An Option Towards Developing Treatment-Free Bees

    Interesting paper, thanks.

    It seems to me that a study of a particular population has to be done, to identify those characters that can be used as 'markers' for a subspecies or local ecotype. (This study concerned a local ecotype - late brood bees adapted - by beekeepers to some extent, and perhaps a lot) to a strong heather source.)

    "Discussion
    While no single diagnostic morphological or molecular
    character was found to distinguish the Landes ecotype from
    the broader A. m. mellifera population, the combined
    morphological and molecular analyses provided a powerful
    suite of characters for identification of the Landes population.
    Morphological analysis seemed to be more informative
    than molecular data for the characterization of the
    Landes population (both ecotypic and non-ecotypic colonies)
    from other populations of A. m. mellifera yet appears
    to have little utility in differentiating ecotypic and nonecotypic
    Landes colonies. Molecular data were quite useful
    to distinguish A. m. mellifera from imported subspecies in
    the study area but again provided little information to
    distinguish among ecotypic and non-ecotypic Landes colonies."


    "The similarity of morphometric results from Landes
    colonies in the present study and the results obtained
    by Cornuet et al. (1982) provide further evidence of the
    continued presence of the Landes ecotype (Strange et al.
    2007). By focusing morphological analysis on the characters
    identified as informative by discriminant analysis

    (wing venation and external abdominal characters) the
    process may be greatly simplified. The statistically significant
    differences in two morphological characters between
    the ecotypic and non-ecotypic colonies (width of tomentum
    light stripe and wing vein angle g18) must be viewed
    cautiously with regard to their utility for selecting ecotypic
    colonies. Moritz (1991) demonstrated that selection based
    upon a few well-differentiated morphological characters
    did not arrest significant hybridization between two sympatric
    subspecies when all morphological characters were
    analyzed. That is, the selection program produced hybrids
    that went undetected when assays were based on only a few
    morphological characters typical of the desired parental
    population. Thus, it is critical to use a broad array of
    informative characters in the selection of potential breeding
    stock."

    A can't see any indication here that selection based on a few markers would be likely to achieve anything other than to raise those markers in the population in any population unless that had been shown to be so by this sort of local close study.

    I remain of the view that its a game that sells microscopes.

    Mike (UK)
    Anti-husbandry: Medication + Reproduction = Continuing Sickness
    http://www.suttonjoinery.co.uk/CCD/

  4. #84
    Join Date
    May 2009
    Location
    Canterbry, UK
    Posts
    1,656

    Default Re: An Option Towards Developing Treatment-Free Bees

    Quote Originally Posted by WLC View Post
    "A high degree of consistency between wing morphometry and molecular information has been demonstrated by Miguel et al. (2010). Therefore, wing geometry is particularly suitable to track phylogenetic relationships between subspecies, where the full "classical" character set can be misleading."

    I wouldn't use the term 'footnote' to describe the treatment of morphometrics in that paper.
    OK, on the surface... However, its not clear exactly what is being said in that extract. The first sentence is not at issue - of course there is. What that 'therefore' is doing there however I don't know - what is written simply doesn't follow, and conflicts with what we've just seen. Can we read the paper?

    Mike (UK)
    Last edited by mike bispham; 01-23-2014 at 03:24 AM.
    Anti-husbandry: Medication + Reproduction = Continuing Sickness
    http://www.suttonjoinery.co.uk/CCD/

  5. #85
    Join Date
    May 2009
    Location
    Canterbry, UK
    Posts
    1,656

    Default Re: An Option Towards Developing Treatment-Free Bees

    Quote Originally Posted by Oldtimer View Post
    Which means it works, sometimes.
    Works at what? All physiological features correlate with dna coding (microsatellite data) - whether or not in any particular case that has been shown (i.e. the molecular features that correspond have been located.) That is where the 'design' originates - the dna is the 'blueprint', the instruction for making that pattern.

    What has been shown is that the position of the particlar dna molecules responsible (for wing vein pattern) in this case have been found. What that means is: 'we' don't have to photograph and analyse the wings themselves, we can just refer to dna analysis to find out if that feature is present.

    If it is... that doesn't tell you anything about the other 99.99% of the bee - unless.. other studies of that particular local population have indicated that there is a high level of corrolation between that marker and another feature, or group of features (like those typical of a particular subspecies) within that local ecotype (and/or more generally).

    For general application you need more markers, and then you need to be able to calculate the likelihood that all those markers would appear at once if the individual was not in fact in possession of a high proportion of the dna of a particular racial type - i.e. 'pure'. That's a complex undertaking. (see the first extract supplied in my post #81)

    What the math ('stats') will do is give you a really quite complicated idea of how likely it is that the individual is mostly (and how much 'mostly') of a particular subspecies.

    The question is, what does knowing a wing pattern/dna feature/s tell you about ancestry. This alone: either mum, or dad, or both, had that feature. (Someone will hopefully tell me why its more complicated/simple than that, and I'll be grateful for the detail)

    Does that feature entail the presence of any other features - that might collectively demonstrate a level of racial purity. Nope.

    The best reason I can come up with for the conflict between what this paper says and what the others we've seen say is that the authors of this paper are software developers who lack proper advice on the detail of what-means-what here.

    Mike (UK)
    Last edited by mike bispham; 01-23-2014 at 03:23 AM.
    Anti-husbandry: Medication + Reproduction = Continuing Sickness
    http://www.suttonjoinery.co.uk/CCD/

  6. #86
    Join Date
    Jul 2010
    Location
    Auckland,Auckland,New Zealand
    Posts
    5,999

    Default Re: An Option Towards Developing Treatment-Free Bees

    It's rare indeed Mike, but I agree with your post. And pretty much because as previously stated the alleles that influence wing vein pattern do not have to be linked to race, as time goes along the method will become less and less applicable.

    The conflict between which papers?
    44 years, been commercial, outfits up to 4000 hives, now 120 hives and 200 nucs as a hobby, selling bees. T (mostly).

  7. #87
    Join Date
    Feb 2010
    Location
    New York City, NY
    Posts
    4,317

    Default Re: An Option Towards Developing Treatment-Free Bees

    Fellas, I bought resistant stock from a breeder who open mates Italian/Buckfast queens in Texas where the 'hybrid swarm' is found.

    So, my own particular need would be to see, as quickly and cheaply as possible, where a sample of workers from those open mated queens 'cluster' in a plot so that I could compare them to other Honeybee types.

    GWV would also give me a chance to see if I've lost the original queen and hybrid workers.

    Both GWV and molecular methods require statistical analysis. However, I think that using DNA markers has it's own set of caveats above and beyond what can be found using GWV.

    In my opinion, it's six of one, and half a dozen of the other.

  8. #88
    Join Date
    Jul 2010
    Location
    Auckland,Auckland,New Zealand
    Posts
    5,999

    Default Re: An Option Towards Developing Treatment-Free Bees

    Quote Originally Posted by WLC View Post
    So, my own particular need would be to see, as quickly and cheaply as possible, where a sample of workers from those open mated queens 'cluster' in a plot so that I could compare them to other Honeybee types.
    Why?
    44 years, been commercial, outfits up to 4000 hives, now 120 hives and 200 nucs as a hobby, selling bees. T (mostly).

  9. #89
    Join Date
    Feb 2010
    Location
    New York City, NY
    Posts
    4,317

    Default Re: An Option Towards Developing Treatment-Free Bees

    I think that using GWV cluster analysis to monitor degrees of hybridization, and how that changes over time in relation to resistance, is useful information for a TF beekeeper.

    I paid for open mated hybrids. I want to be able to monitor that along with its relationship to resistance.

    Think of it as a way to see how your queens are doing.

  10. #90
    Join Date
    Jul 2010
    Location
    Auckland,Auckland,New Zealand
    Posts
    5,999

    Default Re: An Option Towards Developing Treatment-Free Bees

    I just find it surprising that of late you have been presenting yourself as if you are a genetics wiz kid, quite able to examine the DNA of your bees, but instead you opt for a debatable method based on wing design which may or may not be related to the breed.
    44 years, been commercial, outfits up to 4000 hives, now 120 hives and 200 nucs as a hobby, selling bees. T (mostly).

  11. #91
    Join Date
    Nov 2009
    Location
    Belfast, Ireland
    Posts
    396

    Default Re: An Option Towards Developing Treatment-Free Bees

    Quote Originally Posted by mike bispham View Post

    I remain of the view that its a game that sells microscopes.
    Mike (UK)
    No need for the cynicism about the plot to gain control of world microscope sales but it is definitely wise to be cautious.
    Does Monsanto do microscopes!
    Wing morphometry has some use, especially in populations where beekeepers have never used it as a selection criteria.
    It is useful for telling you what something is not but not so clear at proving an individual bee is of a specific subspecies.
    Picking up on some of your other points, I would argue strongly that it is important to preserve pure race subspecies where they are still free from hybridisation. When they are hybridised they cannot be unhybridised at a later date so the subspecies is lost for ever.
    In my case I believe strongly that A.m.m. is under threat due to the proliferation of traditional commercial bee types and as such it needs to be protected. Protection is not a static concept and there is no reason why A.m.m. cannot be a candidate for varroa tolerance - in the natural genetic variation within the subspecies.
    Mellifera is not native to the US so this argument does not apply but it would be a shame to lose the native bee from North West Europe due to uncontrolled hybridisation with other subspecies.
    In the South of England there is a complete mish-mash of different bee types so you might as well work away with your collection of swarms and cutouts.

  12. #92
    Join Date
    May 2009
    Location
    Canterbry, UK
    Posts
    1,656

    Default Re: An Option Towards Developing Treatment-Free Bees

    Quote Originally Posted by jonathan View Post
    No need for the cynicism about the plot to gain control of world microscope sales but it is definitely wise to be cautious.
    Does Monsanto do microscopes!
    No need for cynicism about my cynicism! Seriously, a few folks who find a nice little line in sales patter, selectively presenting matters in a light that offers something of value, can and do shift billions of quids worth of gear. Never drop your guard near salesmen! Talk of high tech gear, and technical terms is sufficient to bamboozle most non-scientists into thinking they're getting something of value. As the shampoo ad says " here comes the science". Haha.

    Quote Originally Posted by jonathan View Post
    Wing morphometry has some use, especially in populations where beekeepers have never used it as a selection criteria.
    It is useful for telling you what something is not but not so clear at proving an individual bee is of a specific subspecies.
    But Jonathan, what does it tell you except:

    a) this bee has genes that code for this vein pattern.

    b) (therefore) at least one of its parents had genes that code for this wing pattern.

    By looking a lots from the same hive you can deduce that probably the queen had two alleles for that pattern or that she mated with several drones with that allele, or even both. But without a number of other morphological traits to help reach a conclusion about likely make-up and purity of race/ecotype that offers no help at all.

    It doesn't tell you that a bee is not of a particular subspecies. You do realise that the longer this goes on, the more bees of all backgrounds will tend to have Amm wing patterns? The more people around you are doing it, the more likely it is that you are getting a false result?

    Quote Originally Posted by jonathan View Post
    Picking up on some of your other points, I would argue strongly that it is important to preserve pure race subspecies where they are still free from hybridisation. When they are hybridised they cannot be unhybridised at a later date so the subspecies is lost for ever.
    Well, yes. But I suspect that the proper way of doing that is, first, by excluding - at least minimising - foreign genetic input; second by allowing natural selection to run its course. As soon as you start to interfere - especially if doing so badly because unequipped with the proper information to make selections, you become part of the problem - one of the factors that make the job harder.

    I can't see that making selections based on just two or three features is going in the right directionway at all - its just making more bees with the features that will fool more amateurs further down the line. Unless.. it has been shown that in that place those features do tend to represent a high level of local native purity.

    Quote Originally Posted by jonathan View Post
    In my case I believe strongly that A.m.m. is under threat due to the proliferation of traditional commercial bee types and as such it needs to be protected. Protection is not a static concept and there is no reason why A.m.m. cannot be a candidate for varroa tolerance - in the natural genetic variation within the subspecies.
    I agree - although I might argue for stable thriving local ecotypes whatever their origins. Put it this way: if we disappeared tomorrow, the bees that would be around in fifty years would probably be as good as any that might result from exterminating all but high-purity native black bees, then disappearing.

    Quote Originally Posted by jonathan View Post
    Mellifera is not native to the US so this argument does not apply but it would be a shame to lose the native bee from North West Europe due to uncontrolled hybridisation with other subspecies.
    Again, I agree. I just don't think that having amateurs fiddling around with microscopes is assisting that aim, and it could well be making it harder. Amateurs who keep unadapted strains alive which then poison the genetically diverse natives around, on the ground that they exhibit native morphological features and so must be preserved at all costs, are worst of all. (Present company, if that applies, excluded of course!)

    Quote Originally Posted by jonathan View Post
    In the South of England there is a complete mish-mash of different bee types so you might as well work away with your collection of swarms and cutouts.
    Sure. Not only that but a constant input of bees from heaven knows where.

    Mike (UK)
    Anti-husbandry: Medication + Reproduction = Continuing Sickness
    http://www.suttonjoinery.co.uk/CCD/

  13. #93
    Join Date
    Nov 2009
    Location
    Belfast, Ireland
    Posts
    396

    Default Re: An Option Towards Developing Treatment-Free Bees

    You don't actually use a microscope for scanning bee wings.
    All you need is a decent flatbed scanner and the software is free to download at drawwing.org.
    I use the Epson scanner I use for general scanning of documents. Anything at 2400 dpi or over is sufficient.

    Wing morphometry easily distinguishes between pure race subspecies.
    The average cubital index of Carnica is about 4.0 whereas with A.m.m it is around 1.8.
    A.m.m. has a negative Discoidal shift whereas Carnica is positive.
    The problem is that it is of dubious use with regard to hybrids.
    If you cross a pure A.m.m. queen with pure carnica drones does it produce a CI half way between the two? Who knows.
    If you cross a mongrel queen with mongrel drones what do the wings tell you? Probably nothing

  14. #94
    Join Date
    Feb 2010
    Location
    New York City, NY
    Posts
    4,317

    Default Re: An Option Towards Developing Treatment-Free Bees

    By using PCA, De Jong was able to study those kinds of issues.

  15. #95
    Join Date
    May 2009
    Location
    Canterbry, UK
    Posts
    1,656

    Default Re: An Option Towards Developing Treatment-Free Bees

    Quote Originally Posted by jonathan View Post

    Wing morphometry easily distinguishes between pure race subspecies.
    The average cubital index of Carnica is about 4.0 whereas with A.m.m it is around 1.8.
    Jonathan,

    What that says is... one subspecies is associated with one pattern, one with another.

    If you have a hybrid - of the least sort - what it tells you is: the dna that coded this wing originally came from a particular subspecies.

    It doesn't tell you how long ago the allele was passed on from a pure individual - it might be hundreds or thousands of generations ago.

    That bee might be 99.99% Carnica, 0.001% Amm! It might be 99.99% Amm, 0.001% Carnica. And it might be anything inbetween, taking in several subspecies.

    Again:

    3. Available methods and markers
    3.1. Morphometry
    There is no morphological "key" to honey bee subspecies, no simple logical tree based on a sequence of single discriminating characters. Instead, measurable morphometric characters show gradual changes and their ranges mostly overlap between subspecies. Thus, subspecies often differ only slightly in the mean values of several body characters, and therefore advanced statistical methods are required for discrimination of groups. The concept of numerical taxonomy was introduced into honey bee taxonomy by DuPraw (1964, 1965) and further elaborated by Ruttner et al. (1978).
    Standard methods for characterising subspecies and ecotypes of Apis mellifera
    Marina D Meixner1*, Maria Alice Pinto2, Maria Bouga3, Per Kryger4, Evgeniya Ivanova5 and Stefan Fuchs6
    https://bibliotecadigital.ipb.pt/bit..._etal_2013.pdf

    Quote Originally Posted by jonathan View Post
    The problem is that it is of dubious use with regard to hybrids.
    If you cross a pure A.m.m. queen with pure carnica drones does it produce a CI half way between the two? Who knows.
    It will draw genes/alleles from, yes, about half each. At random.

    Quote Originally Posted by jonathan View Post
    If you cross a mongrel queen with mongrel drones what do the wings tell you? Probably nothing
    That somewhere in its ancestry the bee has a particular subspecies.

    Bear with me while I try to get a grip on this with an analogy.

    Imagine we have three lego sets, all building the same house, but with red, blue, or yellow bricks and other parts.

    We mix them all up, and a colour-blind person separates the parts back into three sets, now of mixed colours.

    We build a house, and throw a cover over it. Someone comes along and says, if you take a single brick out of that house, I'll tell your whether the house belongs to the red, the blue, or the yellow set.

    Eh? we say? That isn't, can't be, a legitimate question. You can't answer a question that isn't legitimate.

    It isn't legitimate because that house is made from a mixture of bricks from 3 sets. It doesn't belong to any.

    So.. what you're saying is, you've found a village with 90% blue houses, all covered up, and you can, by making a little hole in the same corner, and looking at one particular corner-brick, tell if one house is more or less than than 90% blue.

    Bear in mind, each house is made from hundreds of thousands of bricks.

    There is, I agree, a relation: finding a yellow or red brick indicates the likihood of a lower blue component - which is to say more hybridisation. Is that the basis of the reasoning?

    (That doesn't btw equate, for a moment, with your claim "Wing morphometry easily distinguishes between pure race subspecies")

    And: answer me this: how many deliberate selections made by beekeepers on this basis will it take to negate that diagnosis due to artificially enhanced levels? Have we already passed that threshold? If not, when are we likely to pass it?

    At the moment I'm going to stick with the expert line against the amateur would-be pure-subspecies breeder:

    "There is no morphological "key" to honey bee subspecies, no simple logical tree based on a sequence of single discriminating characters."

    Mike (UK)
    Last edited by mike bispham; 01-23-2014 at 11:11 AM.
    Anti-husbandry: Medication + Reproduction = Continuing Sickness
    http://www.suttonjoinery.co.uk/CCD/

  16. #96
    Join Date
    May 2009
    Location
    Canterbry, UK
    Posts
    1,656

    Default Re: An Option Towards Developing Treatment-Free Bees

    Quote Originally Posted by WLC View Post
    By using PCA, De Jong was able to study those kinds of issues.
    Using technical acronyms tends to baffle rather than explain.

    Mike (UK)
    Anti-husbandry: Medication + Reproduction = Continuing Sickness
    http://www.suttonjoinery.co.uk/CCD/

  17. #97
    Join Date
    Nov 2009
    Location
    Belfast, Ireland
    Posts
    396

    Default Re: An Option Towards Developing Treatment-Free Bees

    Mike.
    Post 95 makes no sense from start to finish!
    Genetics is not a subject to be made up as you go along.
    Stick to what you know.
    We are largely in agreement about the limitations of wing morphometry.

    This plot shows wing samples from one of my A.m.m. colonies compared with wings from a New Zealand carnica queen colony in Scotland which I scanned. The CI difference alone makes it very easy to separate the two.
    Like I said, I have data from about 100 colonies I have scanned and I am much less confident about the utility of this technique than I used to be - when you are dealing with hybrids.

    carnica plot with AMM.jpg

    Incidentally, the queen from the A.m.m colony was superseded in her 4th year and the colony never tried to swarm - for those who believe A.m.m. is naturally swarmy!
    Last edited by jonathan; 01-23-2014 at 12:24 PM.

  18. #98
    Join Date
    Jul 2013
    Location
    Morro Bay, California, USA
    Posts
    863

    Default Re: An Option Towards Developing Treatment-Free Bees

    I haven't gone back to look at the fundamental papers establishing the wing morphometric discipline. I played around with wing scans (and posted these to a previous thread). I see two big issues with its implementation.

    First is auto-correlation. The wing measurements are cross linked -- a shift in one, by necessity affects the intersections of its neighbors. This means rather than 7, or 18 INDEPENDENT measures you have a set of data that reflects the same underlying condition. A movement in one vertex affects its neighbors -- this amplifies apparent differences and distinctions. The statistical technique of Principal Component Analysis obscures the issue of auto-correlation because the x-y coordinates of a PCA are *not* reducible to any single "real-world" parameter. The axes are synthetic.

    Second issue affects the "discoidal shift" -- which is the amateur-friendly measure of the position of a single vertex.

    Discoidal shift is the measure of the angle of a line (d-e) relative to the perpendicular of the line (a-d-b). The constructed vertex d is erected from the wing vein intersection "c". (or a point of tangency on a circular arc "acb").

    In my interpretation, "c" is a vague point (as the wing veins are broad at this juncture). The extremely short segment (c-d) defines the perpendicular, but is very short relative to the very long d-e ray. Inherent error in the Discoidal shift is enormous -- it is at least 3 degrees of freedom away from its definitional vertex's. Extending the baseline from the tangent point "c" which is very close to the chord line a-b magnifies its error in location. "a" and "b" are points of maximum rather than vertexes, and also have error associated with them. Their exact location can shift the tangent perpendicular.

    I work with principal component dataset all the time, and am in awe of their power *and* their ability to lead astray. Auto-correlation can be your friend (as it defines "neighborhoods" and linked parameters) or your worst enemy.

  19. #99
    Join Date
    Nov 2009
    Location
    Belfast, Ireland
    Posts
    396

    Default Re: An Option Towards Developing Treatment-Free Bees

    Quote Originally Posted by mike bispham View Post
    I don't think any of that says 'wing analysis alone tells you something about an individual's subspecies identification' - and if it did then it would be in direct conflict with Meixner et all (as just cited)

    It seems to me to be talking about automating a process - that may then be of use to to people who know what they're doing, and know not to read (far) too much into a single morphometric.

    Has anyone actually read the paper?

    Mike (UK)
    I have. And I cited it in an article I wrote last October for the December edition of the IBRA Bee World magazine!
    It's worth joining IBRA to get access to the magazine and the Journal of Apicultural research.

  20. #100
    Join Date
    Nov 2009
    Location
    Belfast, Ireland
    Posts
    396

    Default Re: An Option Towards Developing Treatment-Free Bees

    JWC.
    You need a fairly large sample size to get any meaningful data about wing venation, and it may not even be that meaningful!
    If a queen mates with say 15 drones and you want to sample a colony you need to take a reasonable sized sample to reflect this.
    A large sample should even out some of the measurement error as well.
    Drawwing places the points at the wing vein junctions automatically but you do need to correct some of the placements manually.
    I agree with the caveats you are making about measurement.
    Getting off topic for a treatment free thread.
    Last edited by jonathan; 01-23-2014 at 12:46 PM. Reason: sp

Page 5 of 8 FirstFirst ... 34567 ... LastLast

Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
Ads