[Discussion transcription]

Date:
[c. 1956]
Reference:
PP/CRI/H/2/15
Catalogue details

Licence: In copyright

Credit: [Discussion transcription]. Source: Wellcome Collection.

    3/48
    in, and, if we call the base X, the reasonable guess is that the dinucleotide going in is X-T. You have a certain amount of thymine, and it's going in in the places where you have a requirement for TT. If there isn't much T you will get the analogue going in in place of the T, and possibly it has a break there (the nucleotide doesn't seem to be very happy). Of course, it can be tested by looking to see if you get this sort of behavior. The fortunate thing about this idea is that it's not too difficult to test it. DR. CHARGAFF: I think that before long you will retreat from this untenable posi tion. You will notice that you avoid all the problems of protein synthesis. You see, you would have had to have a Maxwellian demon in each cell which would really direct the methyl cytosine to go next to a guanine before you can incorporate it. In a paper where I discussed it, the methyl cytidylic was unable to lie next to guanylic. This again we could explain if we assumed - and it won't fit in that hypothesis - that the neighboring and not the opposite base directed the incorporation. Then you would end up with an infinite number of determinants in the directing template. Now, maybe that could be so. If that were so, at the same time I am afraid you would erase your beautiful hypothesis. DR. CRICK; It seems you just haven't understood the idea. The idea is this, that you can incorporate, say, dxnucleotid.es so that you would have more than a number of dinucle oti de s. You would have G-A, C-M, and A-M. I am just making up a certain proportion here. Then what if the G-A nucleotides appeared in the highest proportion? Now, what one naturally assumes is that an enzyme makes these, and then you have to assume that the enzyme couldn't take in G-M but takes in G-A. That's the result. DR. CHARGAFF: The point is that this enzyme can't carry in the other pairs. I am willing to explain facts by a devious route, but I don't see any particular reason to explain away facts in order to fit the hypothesis. It is quite possible that the hypothesis will eventually be modified to adnit of such differences, but at the moment one thing is just there3 the other doesn't
    digitised image 1
    digitised image 2
    digitised image 3