THE HEREDITARY ORDER: GENETIC INFORMATION 



end, ho\\- do the amino acids know which template to read? If there 

 is no overlapping, the process can start in A or in C or in B, and 

 the end results will be determined by the point of departure. 

 Moreover, there are 64 permutations. If there is a possible over- 

 lapping, the difficulties are the same. Crick, Griffith, and Orgel have 

 worked out a code possessing the following properties: 



1. A given amino acid can be coded only on one triplet. 



2. A given triplet corresponds to only one amino acid, the total 

 number of "good triplets" being twenty. 



This means that 44 triplets out of 64 do not make sense. The good 

 triplets — which make sense — are represented in Figure 8. All the 



A 

 B 

 C 

 D 



Figure 8. Crick's Tentative Solution of the Coding Problem. 



Any combination ABA, ABB, ACB, BCA, etc., is valid. Any other does not 

 make sense. 



others are bad triplets. No good triplets can be obtained by com- 

 bining the adjacent bases of two good triplets. Thus no overlapping 

 is possible. A given sequence of bases necessarily gives rise to a 

 given sequence of amino acids. In this hypothesis, the template is a 

 single chain. The possibility^ that a protein may be synthesized di- 

 rectly on DNA is not excluded. But the double DNA helix is 

 built from two complementary chains. If one is composed exclu- 

 sively of "good" triplets, the other chain would necessarily have a 

 large fraction of "bad" triplets and would therefore not make 

 sense. If proteins were synthesized on a DNA template, it would 

 therefore be, according to the discussed code, on one of the chains 

 only. 



[25] 



