270 Mary Barber 



Associated changes. The penicillin-destroying variant cul- 

 tures all grew less luxuriantly than the parent strains and 

 yielded small opaque colonies on nutrient agar. The variants 

 isolated from the Oxford staphylococcus showed little or no 

 coagulase or a-toxin activity and were not typable by bac- 

 teriophage. Those isolated from D3S had lost the capacity 

 to produce coagulase or a-toxin, but were weakly lysed by 

 some, but not all, of the phages lysing the parent strain. 



Appearance on ditch-plates. A comparison of these variants 

 with a typical penicillinase-producing strain of Staph, aureus 

 was made by examining penicillin ditch-plate cultures after 

 24 hours' and 7-8 days' incubation (Figs. 2-7). Fig. 2 shows 

 a plate with the Oxford staphylococcus on one side and an 

 active penicillinase-producing strain (D3R) on the other. 

 It will be seen that after 24 hours the Oxford staphylococcus 

 shows an almost straight line of inhibition, but growth is 

 semi-transparent and fades away as the level of penicillin 

 becomes bacteriostatic. After 7 days' incubation, the area of 

 semi-transparent growth is larger. With D3R, on the other 

 hand, penicillin sensitivity increases as the inoculum becomes 

 smaller by plating out, and growth becomes more luxuriant 

 as the concentration of penicillin increases. Thus the heaviest 

 growth is seen at the edge, just before the level of penicillin 

 is too high to have been neutralized by the penicillinase of 

 the culture and is, therefore, inhibitory. 



Fig. 3 shows a plate with on one side a penicillin-destroying 

 culture derived from the Oxford staphylococcus after 8 months 

 and on the other the parent strain. It will be seen that the 

 variant culture grows more closely to the ditch than the 

 parent, and shows an area of semi-transparent growth im- 

 mediately preceded by a ridge of increased growth. After 

 7 days of incubation, the ridge is very marked and a few large 

 opaque colonies are also seen at the edge of the semi-trans- 

 parent growth. This appearance was quite frequent during 

 these experiments and suggested the development of peni- 

 cillin-destroying variants at the site of the ridge, while the 

 semi-transparent growth consisted of cells showing only an 



