Check the internet.  I never heard of it.

Ian,

It's good idea to use SPME or if you have autosampler headpsace of GC would be more better.

Try GC-MS

(A job for Time-of-Fly MS?) Seriously, though, preconcentration onto fibrous absorbent seems required technology. (Takes a lekking and keeps on tech-ing?)

Methanol by SPME is not straightforward (ethanol is no problem), and definitely not straightforward by MS. I would tend more towards sealing the fly in a vial and then sweeping the heated headspace onto the head of a PLOT column as my concentrator. My experience is that FID is actually more sensitive for methanol than MS, but I don't have one of the newer MS systems (like a 5975).

Interesting project - definitely not your everyday analysis!

I agree that headspace is the way. For those tiny amounts you should not need a trapping column because the vapours can be trapped at the head of an ordinary wall-coated capillary.

I don't think purge and trap methods are needed here. For various reasons static headspace (which can sample a fair proportion of the contents) often gives equally good sensitivity, though you do need a solvent (water?). If you don't have an autosampler these alcohols  can be done by hand using a gas syringe at room temperature - insert a disposable hypodermic needle through the septum, and sample the headspace by inserting the syringe needle through it.

The difficulty I see is that, since you don't think the fles are losing all their volatiles in flight, you will need to crush or digest them somehow, probably avoiding extremes of pH. Also, you don't want artefactual formation (gut microbes...) or losses. You could cool the vial rapidly with a jet of liquid N2 or CO2 (which may have more cooling power), and then grind them with a glass rod or add some hard beads and use ultrasound or a vibratory grinder.

A remaining question is whether you need a denaturant, which could be an organic dissolution solvent or an inorganic compound dissolved in water. If you use water, it may help sensitivity to salt out the alcohols; not all of the usual salts work, but I can't remember on which Q&A thread that was discussed.

Dr. Lee brings up some interesting points. Grinding will definitely aid in the transfer of the alcohols to the gas phase. I'd make sure I held the vial above the boiling point of ethanol; this drives more of the alcohol into the headspace. 

Normally for aqueous samples we do salt out (sodium chloride works the best), but in this case I think you're better off not adding water to the ground sample as the alcohols are miscible with water. I agree that you'll have to grind them cryongenically in order to avoid analyte losses in the sample prep stage. 

I'd use something similar to a GasChrom Q to trap the alcohols at the head of the column. You, unfortunately, will probably need essentially a full transfer of the gas phase from the vial due to sensitivity issues, so I really don't see any way of you avoiding sweeping the headspace in the vial. If the vial is small (say 2 mL) then the swept volume is minimal and it will very efficiently trap at the head of the column, thus giving you all of the analyte on the column. You'll need to start at a relatively low temperature in order to prevent band broadening.

Thank you all for the suggestions! I will try to prepare some samples, and see if I am able to measure from single insects or small cohorts. I will try to update as I learn more! Please let me know if you have further suggestions as well. Cheers!