146 MUTATIONS 



character (judged by plaque morphology). In this case the experi- 

 ment gave the opposite result, that the forward mutation frequency at 

 this site was much less in the revertant than in the original standard 

 type. In other words, by this trick of going forward and back, it is 

 possible to eradicate partly at least this hot spot. 



Lederherg: Assuming you had a specific suppressor of the very 

 frequent reversion at that site, right? 



Atwood: By two steps you arrived at this figure? 



Lederherg: You had a third state at that site? 



Benzer: This experiment does not prove that. It could be that the 

 false revertant is due to a specific suppressor at another site. In that 

 case, mutations at the original site could appear to be observed as a 

 result of reversion of the suppressor. 



Lederherg: No, that's the mutation you're measuring there. 



Benzer: It might be interpreted in more than one way. My point is 

 that the experiment doesn't always come out to give a demonstration of 

 true reversion. 



Deletions, as in the case of point mutations, do not occur in a random 

 way throughout the structure. 



Figure 23 shows some of the deletions; it does not show their 

 distribution, however. The deletions here have been chosen to be all 

 different and are the ones used to define segments of the map. It shows 

 at least the number of different end points which deletions can have. 

 There are many other deletions, several hundred more, that are avail- 

 able, but they are not shown here because they do not overlap in just 

 the right way to be used to define the sequence of the segments. These 

 are almost all from spontaneous mutations. A few have been obtained 

 by Freese (23), by the use of low pH and high temperature treat- 

 ment. 



This figure does not illustrate the frequencies of various end points, 

 which are in fact nonrandom. One of the striking things is that there is 

 a kind of polarity. Many cases are observed where "deletions" — I 

 use the word in quotation marks although we have evidence that they 

 are deletions — begin at one point, and proceed all the way to the end 

 of the B cistron. But there is no case observed, so far, which starts 

 at a point and proceeds all the way to the left. This could, of course, 

 be an artefact due to what there is outside the r II region, if there 

 were an expendable structure on one side and, on the other side, one 

 which were not expendable, so that one kind of deletion would be 

 lethal. Some end points recur wdth high enough frequency so that they 

 clearly represent hot spots for deletion. For instance, the joint be- 



