Vol. 44, 1958 BIOCHEMISTRY: BENZER AND F REESE 115 



of spontaneously appearing r particles is typically 2 X 10"''. Under the conditions 

 used for induction, on the other hand, r mutants appeared in a proportion of around 

 2 per cent in the progeny from a single growth cycle. To obtain the probability of 

 mutation per phage replication, these values must in each case be adjusted to ac- 

 count for the accumulation of mutants during the growth of the population. This 

 effect, which is roughly proportional to the logarithm of the ratio of final to initial 

 phage titer, should be several fold (perhaps four times) larger for a lysate than for 

 one cycle of growth. Therefore, the rate of induced over-all r mutations can be 

 estimated as several hundred times the spontaneous rate. 



rl region 



cistron A 



■^ 



r H88 



r 164 



cistron B 



r 638 



r A 105 



r 196 r 187 



segment I I | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | II | 



Fig. 1. — Genetic map of the rll region of phage T4, showing the division of the map into seg- 

 ments by a set of rll mutants having large alterations. Each of the two "cistrons' ' is a functional 

 unit defined by the cis-trans test. 



The relative positions of the alterations have been established by their overlapping relationships 

 with a much larger set of mutants. To locate the defect of a new mutant, it is crossed with each 

 member of the set, and standard-type recombinants are tested for. From the results of these 

 crosses the defect can be assigned to one of the segments. For example, if a mutant gives stand- 

 ard-type recombinants with all except r H88, r A105, and r 164, its defect must be located in seg- 

 ment 4. One which gives recombinants with all must have its defect in segment 1. Segments 

 7, 9, and 1 1 are not distinguished by this set of mutants. 



Genetic Analysis of the Mutants. — Spontaneous mutants: A series of 238 spon- 

 taneously arising mutants was isolated, numbered r 101 through r 338, and analyzed 

 as described above. Of the mutants, 132 belonged to the rll group. Of these, 6 

 were not readily mappable because of high reversion rates or leakiness (ability to 

 grow feebly on K). The results of mapping the remaining 126 mutants are shown 

 in Figure 2. 



The relative positions of the sites indicated within a map segment have no 

 significance. Any two mutants that yield standard type progeny when crossed are 

 sketched as horizontally separated, while those groups for which recombination 

 could not be detected (at a level of about 0.01 per cent recombinant progeny or 

 more) are assigned to the same or adjacent horizontal sites. Some mutants fail to 

 give recombinants with (any of two or more) other mutants known to give recom- 

 binants with each other; the alterations of such mutants are represented as hori- 

 zontal bars. A striking feature of the map is the existence of certain "hot spots," 

 where mutations recur with high probability. 



Induced mutants: A series of 170 r mutants, arising under the action of 5- 

 bromouracil in sulfanilamide medium, was numbered N 1 through N 170. Of these, 

 89 were of the rll type. Twelve of the latter were not readily mappable because 



223 



