The analysis of the aroma of a food product requires the extraction of
its volatile compounds. As the composition of the extract obtained de
pends on the extraction method, it is important to verify that the odo
ur actually extracted is representative of the product. This verificat
ion is particularly relevant in the case of cooked cured ham, as its o
dour is not very strong as compared to other cured products (dry cured
ham or sausage). Extracts were obtained by four different vacuum dist
illation processes. A sensory comparison of the odour of these extract
s with the reference product was performed by ten trained panelists. T
he odour similarity of the extracts with the reference was assessed in
order to select the most representative one, upon defined descriptors
. Direct vacuum distillation of ground ham suspended in water was sele
cted as the most representative. Gas chromatography (GC)-olfactometry,
a technique of choice to identify potent odorants in food products, w
as then planned. As one of the key point of this analysis is to correl
ate the Flame Ionization Detector (FID) detection to the sniffing one,
a problem very often underestimated by authors, a GC-olfactometric st
udy of a model mixture was first performed with three selected people.
FID and sniffing detection of this model mixture were compared in ter
ms of retention time reproducibility. Standard deviation of retention
indices observed for sniffing and FID detection were found in the same
range. Retention indices calculated with both detections differed wit
hin a confidence interval of +/- 3 index values. FID and odour detecti
on were found to be correlated within this interval. (C) 1997 Elsevier
Science Ltd.