Sugars can be determined very accurately.  Fructose, glucose and sucrose could easy enough be determined to a precision of +/=  of 1% of the relative concentrations by routine lab tests.  The same is true of other tiny amounts of other sugars that can be naturally in honey.  This would tell you relatively little.  There is going to be plant to plant differences in these concentrations.  Few honeys are actually mono floral and mixes are going to be intermediate.  And, many plants all produce roughly the same ratios in nectar.  Most nectars are high in sucrose relative to fructose and glucose and that sucrose is inverted to fructose and glucose by the bees during ripening.

The three best techniques to go after floral sources would be DNA sequence analysis,  gas chromatography - mass spec analysis of the volatile identity and amount and pollen analyses.  

I have no idea how much DNA is in honey and if all plants put some of their DNA in the final product.  I also have no idea if bees during ripening alter the DNA present.  So, this method would require a lot of development.  But with some of the new micro pore sequencing technology running a raw honey sample should be pretty straight forward and it does not take much DNA.  With the micro pore methods you only need one strand of DNA in the sample so amounts are likely not an issue if some DNA is present.  By some I mean even ppb or ppt.

Gas chromotography - mass spec analyses are quite routine and also sensitive to ppb levels.  How well this would work would depend on different honeys having different essential oils.  We know there are differences or honeys from different sources would not taste different.  But, yuo will have problems with cultivar differences in the essential oils.  Two different oranges might, or might not differ quite a bit for instance.  The advantage of this method is it is fast and easy once the method is set up taking little human time per sample.  The problem would be finger printing all the honeys and finding enough differences in essential oils to give solid results.  It would probably be easy to miss sources less that were less than 10% of the total honey and in some cases a lot higher than that.

Pollen analysis is tedious and going to be fairly expensive per sample and depends a lot on the skill of the person doing the analysis to recognize pollens from each other.  You might be able to automate this method and use AI to do the identifications but such an effort would require running a lot of samples to get the per sample cost down to an affordable level to cover the method development costs of such automation.  I think it likely that pollen analysis would often not spot floral sources that contributed less than 10% to the honey.  And, if some flowers put very little pollen in honey could miss much larger amounts of such flowers.

There is a saying among analytical chemists.  You can pick any two of fast, easy or accurate, but not all three.  The better you want on your third pick the higher the price.

Dick

             ***********************************************
The BEE-L mailing list is powered by L-Soft's renowned
LISTSERV(R) list management software.  For more information, go to:
http://www.lsoft.com/LISTSERV-powered.html