USE OF PHAGES IN EPIDEMIOLOGICAL STUDIES 4 1 7 



proved to be of some value for typing (Fisk and Mordvin, 1944; 

 McClure and Miller, 1946). Wilson and Atkinson (1945), 

 using Fisk's method of isolating phages, introduced a typing 

 scheme in which 21 types were defined with 18 phages. Williams 

 and Rippon (1952) have published a detailed account of the 

 method now in use, which is developed from that of Wilson and 

 Atkinson (1945). As in all the phage-typing schemes other than 

 that of .5'. typhi, phage-type diagnosis in the staphylococci is 

 based on pattern reactions. Twenty-one phages are employed 

 at present for routine work. Because of pattern variability, 

 even in epidemiologically related staphylococci, the custom has 

 been adopted of designating strains by recording the actual 

 spectrum of typing phages to which they are susceptible, and no 

 attempt is now made at type designation as it is employed in 

 salmonellas. 



Staphylococcal phage typing presents different problems from 

 the typing of salmonellas. To begin with, the staphylococcus 

 is much more widespread than the pathogenic Enterobacteria- 

 ceae. In contrast to the few hundred typhoid carriers known to 

 exist in England, there are probably at least 15,000,000 carriers 

 of staphylococci (Anderson and Williams, 1 956) . A much greater 

 variety of types might thus be expected to exist among staphylo- 

 cocci than in the typhoid bacillus. Practical experience seems 

 to confirm this expectation, for a random sample of 221 epi- 

 demiologically distinct strains, examined with the phages of 

 Williams and Rippon over a six months' period, yielded 169 

 different patterns (and therefore "types") which fell into 26 

 pattern groups (Williams, personal communication). With 

 such a variety, the routine use of type designations is clearly 

 impracticable. 



The variability in reactions that has been reported between 

 different cultures of apparently epidemiologically uniform groups 

 of staphylococci is greater than would be accepted as satis- 

 factory in a phage-typing scheme for salmonellas. In fact, of 

 course, the widely different ecology of the two groups of bacteria 

 makes the two typing methods valuable in different ways. 



