Fermentations were performed in a 3-L stirred-tank laboratory fer- mentor (Biostat A-DCU300; B. Braun Biotech International GmbH, Ger­many) with a working volume of 2 L. The bioreactor was equipped with a pH electrode and a polarographic oxygen electrode (Mettler-Toledo). To 1950 mL of sterilized (121°C, 20 min) growth medium containing a double set of nutrients, 50 mL of inoculum was added; thus, the volume of the inoculum was made up to 2.5% (v/v) of the total broth volume. Fermenta­tions were performed at 28°C with an agitation rate of 600 rpm and an aeration rate of 500 mL/min (0.25 vvm) at atmospheric pressure. A rather low airflow was used to avoid excessive foaming; however, the dissolved oxygen tension was always above 15% of the saturation value. The initial pH was adjusted to and further controlled at 5.0 by the automatic addition of 2 M H2SO4 and 2 M NaOH. Foaming was controlled by the manual addition of filter-sterilized antifoam agent (Sigma Antifoam 289). Fermen­tation was continued until the glucose was completely depleted, and then pulse addition of Solka-floc (10 g/L) was applied by adding a thick suspen­sion of 20 g of Solka-floc in a calculated volume of distilled water, thus filling the fermentor to the original volume of 2 L. Nutrients required to support the growth on the second batch of carbon source (i. e., Solka-floc) were supplied by including a double set of nutrients in the original batch in order to avoid any coincidences of response signals that may occur from adding any component along the Solka-floc. The outlet gas composition was continuously monitored using a gas analyzer (Tandem dual gas sen­sor; Adaptive Biosystems, Leagrave, England). Measurement values from the gas analyzer as well as electrode signals were logged to a computer every 10 min.