THE ROCKEFELLER INSTITUTE
FOR MEDICAL RESEARCH

66TH STREET AND YORK AVENUE

NEW YORK 21, N, ¥.

Febe 7, 1956
Dear Joshua:

Thanks for the letter. I will answer it in detail
in the near future. asxs-a@mNeed some time to think
about it. The precis was just that,written more for me
than anyone else as + was trying out ideas ete .

Uniy sent it out on impulse, The paper when and if it
gets written will fill in the details.

Just a few points. Am afraid of uv/uv as there
are such things as mutations and the order of
magnitude of effect is 1 gather much lower than
those I have been dealing with. Can differentiate
strains A and C without this.
The law of maximun unhappiness has set in (you
know things simple in concept but difficult in detail
a la coli recombination).while linkage group twe has
settlei dovn so it is only necessary (at the moment)
to postulate two alternative compatability factors,
linkage group one may require three if not more ,.
Think I can lick it though.
The thing I should like to clarify most is that
recombination between
‘latent gonophage" and vegetative phage <samgmo
depends on the degree of leakiness between the
compatability factors and this is as yet unpredictable.
Thus P4 or P5 both breed true on strains C and D but
not on Ae I am sure this is only a partial truth but
if most of the phage breeds true the finding of the 2smrkx
mutants host range and others becomes an impossible
taske One has to have just the right degree of
leakiness and I can only take what I get here.
Thus LT22 delysogenized may breed all of the phages
true as does strain C. this is why + prefer
heterologous immunity to sensitivity. The heterology
refers here to compatability factors mak: specifically.
Homologous factors give complete danunity, heterologous,
sensitivity at all levels.

[o7) ne Pood

4
fF &. On tha avd dhe o cl,

oP Tae Che hen hd ot
-u,