Beadle • What is a Gene? 



remains a key problem. In the same 

 sense, the precise way in which RNA 

 directs the synthesis of specific pro- 

 teins remains to be discovered. One 

 possibility that has several times been 

 suggested is that segments of RNA 

 serve as templates against which pro- 

 tein molecules are constructed from 

 their constituent amino acids. 



Using the Zinder-Lederberg tech- 

 nique of transduction in Salmonella 

 (transfer of small chromosome seg- 

 ments from one bacterial cell to 

 another by means of a virus capable 

 of being carried in a latent form in the 

 recipient cell), Demerec and his asso- 

 ciates (Proc. Nat. Acad. Sci. 41:359, 

 1955) have developed what appears to 

 be a powerful method for investigating 

 the fine structure of the gene. 



Defining functional units of the 

 genetic material of this bacterium both 

 in terms of their control of single 

 chemical reactions and by their geneti- 

 cally determined positions, it is found 

 that for each such unit mutant changes 

 are possible that can be shown to the 

 nonidentical (often complementary) 

 in cross-transduction tests. In terms of 

 the Watson-Crick structure these are 

 presumed to consist in substitutions, 

 deletions or rearrangements of base- 

 pairs at various positions along the 

 functional unit. 



Using a special technique for de- 

 tecting rare recombinations, Benzer 

 (Proc. Nat. Acad. Sci. 41: 344, 1955) 

 has presented evidence for the existence 

 of similar functional units in bacterial 

 viruses. Again mutational changes of 

 independent origins within a single 

 such unit are often recombinable in 

 crosses. 



It is not yet clear what the relation 

 of these results is to the so-called 

 pseudoalleles of higher forms (Lewis, 

 Amer. Nat. 89: 73, 1955). Neither can 

 it be said with certainty how they are 

 related to the remarkable phenomenon 

 reported by M. B. Mitchell in Neuro- 



239 



spora (Proc. Nat. Acad. Sci. 41: 215, 

 1955) in which there appears to be rare 

 transfer of a limited amount of genetic 

 material from one chromosome to its 

 homolog without conventional crossing 

 over. 



In higher plants and animals, in 

 which the chromosomes are several 

 orders of magnitude larger in cross- 

 sectional area than are the DNA 

 strands of bacterial viruses, Mazia 

 (Proc. Nat. Acad. Sci. 40: 521, 1954) 

 presents evidence indicating that the 

 chromosomes are made up of DNA- 

 protein segments about 4000 A long 

 bound together end-to-end by divalent 

 calcium and magnesium ion bridges. 

 It is possible that these units consist of 

 bundles of identical DNA segments, 

 somehow combined with protein, that 

 correspond to the functional units of 

 genetics. If the breaks of crossing over 

 occur only between such segments, they 

 would also correspond to the genetic 

 units of recombination. But such a 

 structure does not exclude the possi- 

 bility that crossing over is sometimes or 

 even always intragenic. 



Despite these manv important un- 

 answered questions, the main lines of 

 evidence suggest, or are consistent with, 

 the following hypothesis: 



Segments of DNA of perhaps sev- 

 eral hundred nucleotides make up the 

 primary functional genetic units. In 

 bacterial viruses these appear to be 

 transferred to the host without accom- 

 panying protein. In higher forms, on 

 the other hand, there is no convincing 

 evidence that thev ever become sepa- 

 rated from protein. Genetic specificity' 

 (information) is dependent on propor- 

 tions and sequences of base-pairs in 

 these DNA segments. In gene mutation 

 base-pair proportions and sequences 

 are altered by substitution, deletion or 

 rearrangement. Replication of genes is 

 accomplished by separation of com- 

 plementan' nucleotide chains followed 

 by synthesis of complementary daugh- 



