BACKGROUND: Information on the variation in water-extractable arabinopolymers (WEAP) in durum wheat and their impact on pasta quality is limited. A survey of the content of WEAP in selected bread wheat varieties and durum wheat was conducted and the influence of water-extractable arabinoxylans (WEAX) on dough and spaghetti quality was examined by fortifying low-WEAP durum semolina with purified WEAX.

In typical bakery products, wheat flour is the principal ingredient (1) , and wheat gluten has been associated with negative health consequences in individuals with gluten-related disorders (2) . Wholegrain spelt flour (WSF) appears to offer a suitable alternative to wheat for some gluten-sensitive individuals. In order to be accepted by consumers, any 'healthy alternative' version of a food product must be of comparable sensory quality to the standard product (3) . To date, new product-development (NPD) efforts involving WSF have mainly been focused on breads, with little NPD work in the area of other baked goods such as cakes (1) . The aim of this study was to develop WSF versions of sponge cake and butter cake -both popular in Singapore -, with similar baking quality and product acceptability to those of the standard wheat-based versions. In a series of controlled bakery trials, baking performance (specific volume: cm 3 /g) of WSF-and standard wheat-based versions was compared (seed-displacement method). During a five-day post-baking period, crumb firmness (g) and springiness (%) were compared (TX -XT Plusä Texture Analyzer). All assays were performed in triplicate. Analysis of the resulting data indicated an absence of significant difference (p > 0·05) between WGS-and standard wheat-based sponge cakes and butter cakes, in terms of specific volume, firmness and springiness (Mann-Whitney/Kruskal-Wallis tests used).In consumer testing, the perceptions of 50 mainly Singaporean consumers (aged 19-37 years) regarding the wheat-and WGS-based sponge cake and butter cake products were assessed using a 9-point hedonic scale (4) . At the 5% significance level, the wheat-based butter cake scored significantly better than the WGS-based butter cake in terms of overall appearance, overall aroma and texturesoftness, while the wheat-based sponge cake outscored the WGS version in terms of overall appearance and overall aroma. With the remaining attributes ('flavour' and 'overall acceptance', and additionally 'texture-softness' in the case of the sponge cake), no significant differences (p > 0·05) were noted between the wheat-based and WGS-based cakes (Mann-Whitney U test used).The results of the study demonstrate the potential of WGS as a potential substitute for wheat flour in some cake-type products. Although no significant differences were found in terms of specific volume, firmness and springiness between the WGS-and the standard wheat-based cakes, nonetheless the consumer-derived data indicate something of a "preference gap" in favour of the standard wheat-based products. If WGS-based cake products are to be fully successful, it appears that NPD efforts will be required to focus on improving sensory quality.