SummaryComparisons were made of the flavour, free fatty acids and bacterial flora of commercial cheese made at different factories and experimental cheese made under aseptic conditions: (i) with δ-gluconic acid lactone instead of starter, (ii) with starter only, (iii) with starter and added floras derived from the curd of the commercial cheeses (reference flora cheeses).Comparison of the bacterial flora of commercial and reference flora cheeses showed that replication of organisms was better with some reference floras than with others. In all the cheeses the lactobacilli increased in numbers during maturation, whilst other groups of organisms died out.The amount of acetic acid present was influenced by the starter and by the lactobacilli. Single-strain starters produced some acetic acid, most of which was lost in the whey; commercial starters produced considerably more, due to the presence in them of Streptococcus diacetilactis. Later in maturation lactobacilli increased the acetic acid content, a greater increase being observed with homo-than with heterofermentative strains.The initial levels of butyric and higher fatty acids in the milk varied with source of the milk and with the season, summer milk having higher levels than winter milk. During cheese-making a slight increase of these acids occurred in every cheese made with starter and a further small increase occurred during ripening. However, there was no increase in the content of these acids in the cheese made with δ-gluconic acid lactone, indicating that lactic acid bacteria were weakly hydrolysing the milk fat.Flavour trials showed that Cheddar flavour was present not only in the reference flora and commercial cheese, but also in the cheese made with starter only. Different starters produced different intensities of flavour; one strain produced an intense fruity off-flavour. Cheeses made with δ-gluconic acid lactone were devoid of cheese flavour.