A. D. HERSHEY AND MARTHA CHASE 



plaques is tested by examining plaques originating from bacteria mixedly infected 

 with r mutant pairs, which shows that for map distances up to and including 10 

 units (r? X rl3) the mottling is negligible. 



The results of tests of this kind with six pairs of r mutants are shown m Table 4. 

 Three facts emerge. The frequency of heterozygosis with respect to five different 

 r loci is at least approximately the same. The effect of close linkage between the 

 loci r3 and r4 is to produce a sufficiently large class of doubly heterozygous phage 

 particles to cause a sharp decrease in the yield of mottling phage. No significant 

 effect is seen for crosses between markers separated by a distance of three or four 

 units or more. The method is evidently valid in principle, but too inaccurate to 

 yield detailed information. 



We return once more to crosses between phages carrying h and r markers to test 

 the effect of unequal multiplicity on the pattern of segregation of the resulting 

 heterozygotes (Table 5). The effect seen is to increase markedly the frequency of 



TABLE 5 PER CENT DISTRIBUTIONS OF r-r^ HETEROZYGOTES WITH RESPECT 

 TO SEGREGATION PATTERN: CROSSES WITH 5-FOLD EXCESS OF hr OVER 



WILD-TYPE 



hrl X wild hr7 X wild hrlS X wild 



(40) (20) (2) 



Segregants found 



•,j 2 4 71 



hr-wild ^ 



h-hr 78 71 7 



r-wUd 20 25 12 



hr ..0 



No. tested 117 55 161_ 



Per cent yields of recombinants for each cross are shown in parentheses. 

 the parent-recombinant class contaming the parental virus available in excess. 

 This effect is not visible when the markers involved are closely linked. 



Another experiment yields information about the sequence of events in the cell. 

 It is based on the work of Doermann (1948a), who has shown that infected bac- 

 teria artificially lysed at various times during the latent period of viral growth do 

 not yield any virus during the first half of the latent period, and that the yields 

 rise linearly from zero to a maximum during the second half of the latent period. 

 The current interpretation of this result is that the first half of the latent period is 

 devoted to the multiplication of non-infective virus, and the second half to the 

 conversion of non-infective into infective virus (Doermann and Dissosway, 1949; 

 Luria, 1950; Hershey, 1951). During the second half of the latent period, the 

 partial yields of virus can be obtained simply by adding cyanide to the cultures 

 Doermann (1948b) has also shown that very large yields of virus can be obtained 

 when lysis is delayed for several hours, as happens in cultures contammg high 

 concentrations of bacteria infected with r+ virus. 



We have compared viral yields obtained from samples of the same bacterial 

 suspension, infected with hr7 and wild-type, by adding cyanide ten minutes after 

 infection, by spontaneous lysis at the end of the normal latent period (21 to 40 

 minutes), and by spontaneous lysis delayed for five to six hours. The yields of 



188 



