It has been found previously that a significant number of Micrococcus luteus cells starved in a prolonged stationary phase (up to 2 months) and then held on the bench at room temperature without agitation for periods of up to a further 2-7 months can be resuscitated in liquid media which contained (statistically) no initially-viable (colony-forming) cells but which were fortified with sterile supernatant from the late logarithmic phase of batch growth. Here it was found that such resuscitation can be done only within a defined time period after taking the first sample from such cultures, necessarily involving agitation of the cells. The duration of this period depends on the age of the starved culture: cells kept on the bench for 3 months possess a 2 month period of resuscitability while cells starved for 6 months can be resuscitated only within 10 days after the beginning of sampling. It is suggested that the input of oxygen to the starved cultures while they are agitated may exert a negative influence on the cells, since cultures stored in anaerobic conditions (under nitrogen) had a more prolonged 'survival' time. The cells which experienced between 10 and 60 days of starvation on the bench could be resuscitated, although the number of resuscitable cells depended strongly on the concentration of yeast extract in the resuscitation medium. This concentration for cells stored on the bench for more than 2 months was 0.05% while '1-month-old' cells displayed a maximum resuscitability in the presence of 0.01% of yeast extract. Application of the fluorescent probe propidium iodide revealed the formation of cells with a damaged permeability barrier if resuscitation was performed by using concentrations of yeast extract of 0.1% and above. Thus the successful resuscitation of bacterial cultures under laboratory conditions may need rather strictly defined parameters if it is to be successfully performed for the majority of cells in a population.