in resistance is twofold or lower, on the average, 

 at every transfer. Resistance is relatively stable 

 on repeated subculture in the absence of Chloro- 

 mycetin. 



The smoothness of increase of resistance might 

 constitute a prima facie case for non-genetical 

 adaptation ; if, on the other hand, adaptation is 

 genetical, a relatively large number of genes should 

 be found affecting Chloromycetin resistance, assuming 

 that it is to be accounted for by known genetical 

 mechanisms. 



A cross was made between a W 677 strain 

 (T-L-B^Lac.-V/ Mal x - Gal x - Xyl- Ara- Mil-) 

 which had been grown with 640 |i.gm./ml. Chloro- 

 mycetin in broth, and a fully sensitive 58-161 strain 

 (B-M-). Crosses were made with the usual technique 

 introduced by Lederberg and Tatum 2 . By this 

 technique, recombinants are selected out of a mix- 

 ture of parental cells using 'fixed' recombination 

 markers, in this case capacity of growth in absence 

 of given growth factors. Strain 58-161 can grow in 

 the absence of threonine, leucine, vitamin 3 lt which 

 are needed by W 677 for growth ; strain W 677 can 

 grow in absence of biotin and methionine which are 

 needed for growth by 58-161. Plating the mixture 

 of the two strains on minimal medium, where such 

 growth factors are not supplied, parental cells will 

 not develop, while recombinants of the right type 

 can develop into visible colonies. The two strains 

 differ by a number of other markers, lactose fermenta- 

 tion (absent in W 677, present in 58-161), reaction 

 to virus Ti ( VS indicating resistance in W 677, while 

 58-161 is sensitive), etc., as indicated by the other 

 symbols. 



Such differences can be used as 'free' recombination 

 markers in that they are not directly selected for 

 or against in the minimal agar plates, and their 

 segregation in the recombined colonies is used to 

 build a 'chromosome map' 7 . The relative order of 

 the genes which have been 'mapped' satisfactorily, 

 and which are relevant for this and other crosses 

 conducted during the present work, is as follows : 

 fi 1 -( J B,il/)-F 8 -Lac 1 -F 1 -^z-(T,L), where V, refers to 

 virus T t resistance, Az to azide resistance. 



Recombinants between the chloromycetin-resistant 

 W 677 and sensitive 58-161 showed a scatter from 

 full sensitivity to nearly full resistance, the resistance 

 of each recombinant being roughly proportional, on 

 the average, to the length of the chromosome segment 

 which — as could be inferred from recombination 

 markers — given recombinants got from the resistant 

 parent. A closer analysis shows one and possibly 

 two loci for resistance in the segment between BM 

 and Lac ; one locus at least between Lac and V x — 



9 2 



