Flender/Flender Gear Units/Bevel-helical gearboxes B3
onof chemicals) or particular storage conditions, and still others may be due to the activity of microorganisms in packaged commodities.

Because there may be several hundred natu- rally occurring volatiles for certain commodities, and usually 3 or more major volatilesinvolved

in aromas, these were not tabulated. Sufce it to say that naturally occurringvolatiles of plant products do overlap to certain

extent with those related to microbialand fungal metabolisms. 4 DETECTION OF DISEASES BY VOLATILE MONITORING 4.1 MicroorganismsCharacterization of bacteria by

chromatography began by separation of cellular compo- nents by two-dimensional (2-) paper chromatography (3,, but these methods werereported to be poor in differentiating between genera and species. Some early GC work 5 Alvo et al. was done to distinguish bacterial strains on the basis of their pyrolysis products (3, or on the basis of their fatty acid methyl esters (3,. The rst attempts to distinguish bacteria on the basis of their native volatile metabo- lites were made in the 1s (. Volatiles from ether extracts of several strainsof seven Bacillus species and from Escherichia coli ,Aerobacter aerogenes , and Pseudo- monas aeruginosa grown in the same volatile-poor nutrient solution were analyzed by using ame ionization detection (FID) and electron capture detection (ECD) (. Aver-age peak areas for four replicates were ordered from largest to smallest, and differencesin size tested statistically. These signatures were tabulated against genus and strain forboth detection systems and led the identication of specic characters to distinguish be-tween strains. The strains of the different Bacillus species were easy to distinguish because large number of peaks were available for comparison, and in general, the genera that produced acetoin ( . aerogenes andBacillus spp.) produced greater number and total quantity of volatiles than the others. . coli andP. aeruginosa produced few or no peaks, despite good gr