SUMMARYWhen Arthrobacter globiformis, A. simplex, A. crystallopoietes and A . atrocyaneus were grown in a chemostat, their morphology was related to growth rate. A transition from rod to coccus occurred at growth rates characteristic for each species; the higher the rate at which this transition occurred, the higher was the maximum specific growth rate. Organism size, concentration and viability were determined for A. globformis at a range of dilution rates between 0.01 h-l and washout. Yield remained constant except at low dilution rates, although the number of organisms varied because of the changes in their size and shape. The percentage of viable organisms in the chemostat remained almost constant at dilution rates above 0.1 h-l (92 %) and was only slightly lowered even at dilution rates as low as 0.01 h-?Rods grown in the chemostat and then placed in nutrient-free solutions divided and produced cocci which survived for about 56 days. Cocci survived for about 70 days. The respiration rate of freshly harvested rods was higher than that of cocci but within two days both fell to low levels.

I N T R O D U C T I O NThe genus Arthrobacter was erected by Conn & Dimmick (I 947) to contain pleomorphic bacteria with nonsporing, Gram-positive rods. In batch culture with appropriate media, Arthrobacter spp. have a life cycle in which coccoid organisms change to rods and back to cocci (Mulder et al. 1966). The rod stage is only present for a relatively short period of a few hours during which the environment and physiology of the organism is in a dynamic state, making studies on the rods difficult.Ensign & Wolfe (I 964) produced rod and coccus forms of Arthrobacter crystallopoietes by varying the composition of the growth medium. Whereas cocci were produced on a glucose salts medium, rods were formed by enriching the medium with additional carbon sources, e.g. succinate. The effect of using an enriched medium on morphology suggests that the transformation from coccus to rod is either induced by specific chemicals or is related to changes in growth rate. The aim of this experiment was to determine whether morphology was related to growth rate, and this was achieved by growing a strain of Arthrobacter in a chemostat under carbon-limited conditions and observing changes in morphology that occurred with changes in dilution rate. METHODSA strain of1 Arthrobacter (Cluster 3 NCIB 10683) described by Lowe & Gray (1972) and isolated from an acid forest soil was selected. It was grown in achemostat with a working volume of I 50 ml. and vortex aeration, using the basal simple salts medium of Owens & Keddie (I 969), with biotin (I pg/l.), and glucose (0.2 g./l.) as the limiting carbon source. Carbon-limited populations were grown at IO", 15", 2 0 ' and 25' at dilution rates between 0.01 h.-l and 0.4 h.-l. Samples were taken and observed under phase contrast ( x 800) when a steady state had been obtained at any temperature and dilution rate. Bacterial density was determined by direct counts using a Helber slide and biomass was measured by weighing washed bacteria, dried to constant weight at 110'. RESULTSBy varying the dilution rate, the morphology of the organisms in the chemostat could be changed. At 25" rods were produced at dilution rates above 0.25 h.-l, at 20" at rates above 0-1 I h.-l, at 15" at rates above 0.05 h.-l and at 10' at rates above 0.02 h.-l. At lower dilution rates, the bacteria were always coccoid. Both types could be maintained indefinitely by holding the dilution rate at an appropriate level.The transformation from cocci to rods at 25" was accompanied by a reduction in numbers from 6-19 to 4-47 x ro8 per ml. However, the biomass remained constant at all M I C 69 28

No abstract

No abstract