Previous page


next page >

has led to a much better understanding of the origins of cocoa flavour characteristics in terms of the effects of planting materials and methods of primary processing. A key development has been the use of analytical and expert sensory evaluation.

Davies et al. (24) studied cocoas from Brazil, Malaysia and West Africa tasted as plain chocolates by panels of up to 20 trained assessors using a profiling method to characterize the flavours. Maps of the flavours (see Fig. 2.9) were

Principal components analysis of flavour characterization data. (a) PC1 versus PC2

and (b) PC1 versus PC3. Source: Davies et al. (24) quoted by Clapperton (23). Copyright Cadbury Limited.

Principal components analysis of flavour characterization data. (a) PC1 versus PC2

and (b) PC1 versus PC3. Source: Davies et al. (24) quoted by Clapperton (23). Copyright Cadbury Limited.

made using the results of principal component analysis of the characterization data. The first component (PC1) used to make both maps separated cocoas with well-developed cocoa flavour from others with either bitter and astringent tastes, typical of under-fermentation, or acid tastes. Fruity flavours were associated with acid taste. The second component (PC2) separated cocoas with acidic and fruity flavours from those with either bitter and astringent tastes or hammy flavours. The third component (PC3) contrasted bitter and astringent tastes with hammy or spoiled flavours caused by over-fermentation or smoke contamination.

Cocoas with known processing histories were used to interpret the flavour differences. The four samples of West African cocoa which were plotted among the Malaysian samples were prepared using Malaysian conditions of post-harvest processing, i.e. deep box fermentation with a four-step cascade, followed by fast drying at 60°C. Corresponding samples from the same farm, which were fermented and dried according to normal West African practices, were centred in the group of West African cocoas.

The two samples of Malaysian cocoa that are close to the main group of West African cocoas were from trials carried out by Lewis and Lee (25) in which Malaysian cocoa was processed using pod storage for 710 days followed by shallow box fermentation for 5 days with a single turn at the end of the second day. After fermentation these cocoas were sun-dried. The flavours of West African and Malaysian cocoas could therefore be moved in either direction depending on the conditions of post-harvest processing. Flavour appeared, therefore, to be process driven rather than geographical in origin.

Pulp reduction by pressing prior to fermentation, which had been applied to a number of the cocoas from Malaysia, was less effective in changing the flavour. Cocoas which were known to have received that treatment could not be separated from the rest of the Malaysian samples, other than those already discussed.

Studies have shown that flavours developed from different planting materials are clearly distinguishable and differ in cocoa flavour intensity, acid taste, bitterness, astringency and fruity/floral notes (26, 27). The planting materials developed for commercial production in Malaysia are by no means inferior to those grown in Ghana, but they are different and produce distinctly different flavours. The flavour differences between planting materials reside in the different compositions of the cotyledons and are unaffected by gross differences in pulp composition; effects of pod storage on flavour differ markedly between genotypes.

Storing pods for up to 12 days before breaking them for fermentation has for some time been part of post-harvest treatments recommended to improve the flavour of Malaysian cocoa. Shorter periods of post-harvest storage followed by partial removal of pulp by spreading the wet beans to dry in the sun, or by air-blasting, have also been recommended. Organic acids formed during pulp fermentation are clearly the source of the acidity. When beans are dried slowly after

0 0

Post a comment