J. BRONFENBRENNER 



545 



the initial level. If, in place of susceptible bacteria, one employs a culture of a homol- 

 ogous resistant strain, or of heterologous bacteria, or a suspension of susceptible bac- 

 teria killed by heat, the viscosity of the mixture remains constant. Similarly, the 

 viscosity of the mixture remains unaffected if heated phage is employed in place of 

 the active preparation. As calculated on the basis of viscosity changes,' the entire 

 bacterial mass occupies, at the time of maximum swelling, from six to twelve (or 

 more) times the volume occupied by a similar number of normal bacteria. The fact 

 that, following visible clearing, the viscosity of the mixture reaches approximately its 



Fig. 5. — Section of a cinematographic record, representing three successive exposures taken at 

 2-5 seconds' intervals and showing disappearance of two slightly swollen cells. Magnification X 1,500. 



originallevel (or even below it), seems to indicate that the swollen bacteria must be 

 the ones which eventually disappear. 



This conclusion was confirmed by a cinematographic record of lysis of B. coli 

 on agar, at 37° C.^ As the illustrations show, swollen bacteria with perfect outline 

 in one photograph are completely gone in the next taken less than three seconds 

 later (Figs. 5 and 6). Usually in the space formerly occupied by a bacterium, there 

 remains nothing but the faintest trace of amorphous debris, which, in turn, disappears 

 in less than one minute, leaving only a few more or less refractile granules (Fig. 5). 

 Occasionally the debris left after the disappearance of the bacteria is less easily soluble, 

 and is recorded on the film for a longer period of time (Fig. 6). This happens usually 



' Kunitz, M.: /. General Physiol., 9, 715. 1926. 



= Bronfenbrenner, J., Muckenfuss,R.,andHetler, D.: Am. Jour, of Pathol. (In press.) The prog- 

 ress of lysis was photographed by ]Mr. Rosenberger by means of an automatic camera at a rate of 

 twenty-five exposures per minute. The record shows that, following a short period of lag, bacteria 

 began to multiply at a rate noticeably exceeding that of normal bacteria photographed under similar 

 conditions but in the absence of phage. (This stimulating effect of phage has been reported by a num- 

 ber of investigators [Schwartzmann, G. : loc. cil.; d'Herelle, F.: The Bacteriophage and Its Behavior, 

 p. 74. 1926; Meuli, FL: loc. cit.; Saldanha, A.: loc. cil.; Doerr, R.: loc. cil.; Hadley, P.: /. Infect. 

 Dis., 37, 35. 1925]). Owing to the rapidity of growth, many cells failed to complete division, and 

 gave rise to filaments having a length of from ten to twenty times that of the average length of a 

 normal cell (Fig. 3). By the end of the first hour of growth, occasional cells began to show signs of 

 swelling, and by the end of the third hour the great majority of cells in the field appeared more or 

 less swollen, a few among them reaching several times their normal dimensions. Filaments swelled as 

 well as ordinary bacteria (Fig. 4). The swelling continued slowly until about the fifth hour, when, one 

 by one, bacteria began to disappear suddenly and quickly. Some of the bacteria disappeared when 

 only moderately swollen. 



