Summary:The vitamin requirements ofRhodotorula gracilisare satisfied by inclusion in the growth medium of pantothenic acid and thiamine. For maximum growth in the particular medium and with the size of inoculum used in our experiments a concentration of about 10 μg of pantothenate/1 was sufficient but for rapid early growth a higher concentration was beneficial.Pantothenic acid can be replaced by β‐alanine but maximum growth response then occurs only in the presence of asparagine. Either ammonium sulphate or urea provides a suitable source of nitrogen for growth of the organism.Shake flask experiments with a synthetic medium showed that a solution containing only glucose and phosphate was adequate for fattening, that wide limits in pH value (e.g. 4–8) were permissible and that the rate and efficiency of fat production were not affected by large differences in the concentration of sugar.Provided sugar was still present in the medium a progressive decrease in the percentage of protein was accompanied by a corresponding increase in the percentage of fat in the yeast. When aeration was prolonged beyond the point at which all the sugar was exhausted, this trend was reversed.The fat produced during fattening had a lower iodine value than that produced during normal growth. Only a small proportion of the fat inRh. gracilisis in the free condition and extractable by direct treatment with a simple solvent. Preliminary hydrolysis with hot HCl is necessary for quantitative release of the fat.Refinery cane molasses provides a suitable source of sugar both for growth and for fattening ofRh. gracilis. Growth in a fermentor with incremental feeding, under conditions of high aeration and rapid stirring, gave a yield of dry yeast equivalent to 52–54% of the sugar consumed; the doubling time during the logarithmic growth phase was 3 h.Aeration ofRh. gracilisin a phosphate‐molasses medium resulted in a rapid consumption of sugar during the whole period of aeration. This was accompanied by an increase in the weight of yeast, accounted for by an increase in all constituents—fat, carbohydrate, mineral matter and a small amount of protein formed from the residual nitrogen carried over from the growth medium. Since much more fat than protein was produced the percentage of fat in the yeast rose and the percentage of protein fell. The proportion of carbohydrate in the cell was subject to some fluctuation. Under certain conditions it appears that the reserves of carbohydrate may be drawn upon to provide energy for fat formation, resulting in an actual destruction of cellular carbohydrate.The proportion of sugar converted to fat rose to a maximum betweenc.14–23 h and then rapidly declined. For a short period a fat coefficient of over 40% was attained in one experiment. Thereafter conversion was relatively inefficient, a small quantity of fat being produced from the consumption of a large quan