The endonuclease of Bacillus subtilis specific for single-stranded deoxyribonucleic acid is absent in spores, appears during germination only after the start of deoxyribonucleic acid synthesis, and is located almost exclusively in the periplasm.The main DNase of Bacillus subtilis that degrades single-stranded DNA can be fractionated into two forms which account for most of the total activity: a very high-molecular-weight endonuclease (greater than 5 x 106) and a lowmolecular-weight endonuclease (36,000) (3). The higher-molecular-weight form can be converted almost quantitatively and irreversibly into the small enzyme by mild proteolytic treatment (4). A role for this sort of enzyme in physiological processes can be inferred from the observation that a recombination-deficient mutant of B. subtilis (rec3O) (6) is reduced in this activity to approximately 13% of the wild-type level. The study of the cellular site of this enzyme is therefore relevant to its possible function in the handling of the incoming DNA, whether in transformation or in transduction. We have therefore studied whether protoplast formation releases the DNase in question into the medium, and we measured the level of this activity in the spore and at different times after spore activation.Cells (1.3 g) of B. subtilis SB202 hisB2 tyrAl trpC2 aroB2 were treated with lysozyme to destroy the cell wall and remove the soluble periplasmic fraction. In this way 80% of the total activity as measured by the acid solubilization assay was released (see Table 1). The remainder was extracted by subsequent osmotic shock and sonication of protoplasts. Scher and Dubnau also described a B. subtilis endonuclease activity localized in the protoplast supernatant, but this activity is specific for double-stranded DNA (7). The DNase specific for single-stranded DNA released by lysozyme (and therefore probably located in the periplasm) is fractionable into two molecular weight species, like the enzyme activity extracted by less specific treatment (whether alumina grinding or sonication) of whole cells. DNA polymerase I activity was monitored to show that internal enzymes are not released by the procedure used to obtain protoplasts of vegetative B. subtilis cells.This observation raises the question whether the enzyme is present in the cell at a time when the cell wall is not yet formed, such as in the germinating spore.We have therefore measured the level of this activity in the spore and at different times after spore activation (Fig. 1). The spore is practically devoid of single-stranded DNAse activity (less than 1% of the log-phase value, expressed as units per genome). The level of the DNase begins to increase only after DNA synthesis has started, when the vegetative cell begins to emerge from the spore. The DNase level, corrected for DNA content, rises exponentially thereafter to reach a value of approximately 20 TABLE 1.