292 Discussion 



Westergaard: Yes, and this is not surprising, since X-rays also initiate 

 radiochemical processes. It has been proved very beautifully by Giles 

 and de Serres that you can produce back mutations with X-rays. 



Alexander: From the point of view of chemical mechanism, one will 

 have to distinguish between merely spoiling a molecule, making it useless 

 to the cell (which could be a much less non-specific process — antibiotics 

 or anything could do it) and reaction which can actually change the 

 gene so as to alter its function. 



Westergaard: This is what we have been trying to do for years but we 

 have not been very successful in selling the idea. 



Pontecorvo: When you get down to the molecular level what is the 

 difference between a breakage and a rearrangement? There may be 

 structural rearrangements of any kind, or additions or removals of 

 side-chains. I don't see that you can have a distinction at the chemical 

 level. 



Alexander: There are two clear distinctions: you can take the mole- 

 cules which are responsible for gene function and render them useless to 

 the cell so that they can no longer fulfil their normal physiological 

 function and then you get a certain type of mutation, and I would expect 

 that type of mutation to be produced by many agents, including the 

 antibiotics. Then there are some agents which can actually change a 

 gene so that it is not just destroyed but that it can now do something 

 different. Can you explain all mutations by loss ? In that case evolution 

 ought to have been working backwards very quickly. 



Pontecorvo : Not by loss, but by structural rearrangement; and as soon 

 as the structural rearrangement is at a molecular level I don't see that 

 we can distinguish the two. 



Alexander: One can chemically envisage rather crude structural re- 

 arrangements which cannot conceivably give the molecule an altered 

 function but can merely render it useless to the cell. One can also 

 conceive a rather more refined chemical alteration which can allow the 

 molecule to fulfil a modified function. 



Rose: If there has been a chemical deletion the chance of putting that 

 unit back again is remote in the extreme. 



Lederberg: There are just one or two points of difficulty here. One is 

 that we do not know the chemical differences between normal and 

 mutated genes and, until we do, this is bound to be very ethereal. In 

 principle, there should be an emprical search for agents which would 

 be purely unidirectional, which could give one type of mutation and 

 not the reverse. Until now there has been no evidence that such agents 

 exist; and things like X-rays which have formerly been thought to be 

 the best test of such a notion have given changes in both directions at the 

 same loci. If Prof. Westergaard is correct in saying that there are agents 

 which are capable of inducing mutations for biochemical loss and not 

 their reverse, he will have established, I think, a new principle in genetic 

 methodology, and if that is so we shall have to incorporate it in our 

 theory. In principle, all these hypotheses have to be considered but until 

 now there has been no critical evidence that would necessitate a distinc- 

 tion between those types of rearrangement that you call a loss and those 



