CARNEGIE INSTITUTION OF WASHINGTON

DEPARTMENT OF GENETICS

COLD SPRING HARBOR, LONG ISLAND, N. ¥,.
*

23 October, 1947

‘Dear Joshua,

Very exciting about your new job! I had heard about
it only vaguely, and still don't really know quite what you're
doing. I gather its teaching and research, which is much better
than straight research as far as I'm concerned, if you have a light
load and good students. I miss teaching a lot, and, find that
ideas don't crystallize well without a forum. .
‘ As you can see, I'm enclosing my last battered copy
of my Symposium typescript. Forgive the hasty sketches which were
the best I could do for figures, as I had no copies of those. The
paper seems very outdated to me now, in the light of our new results.
Nothing in it has been contradicted, but I would certainly write it
very differently now! In case you might want to mention some of
the recent work in the discussion, I can give you an idea of what
has been happening. First of all, a number of controls have been
added to the technique to take care of possible loopholes that
occurred to us in the course of the experiments. One example:

We worried about the possibility that some of the chemicals might
cause delayed infection with phage (by coating the surface, or some
such thing),thus permitting some division on the plate before lysis,
and consequently production of spontaneous mutants. To check this,

we now determine the rate of infection with phage of bacteria treeted
with the chemicals as compared with controls. Sofar, all are 0.K.
Desoxycholate-treated bugs seem to be more rapidly infected than cons
trols, which was better than we hoped for. Our experiments to test
for differential sensitivity to the chemicals of mutants and nonmu-
tants is now much improved. Also, we are developing a few other
mutations, completely independent as far as we know of the phage
system, to use along with phage resistance. I was never happy about
having just one class of mutants to work with -- too much possibility
of being fooled by peculiar effects on the adsorption and lysis system.
we now have several strains of B, each deficient for one known growth
factor, which undergo reverse mutation to prototrophy (or do you pre-
fer prototrophism?) at a very convenient spontaneous rate. All we
need do to detect them is plate a large sample (109 or thereabouts) on
minimal agar, and count colonies. They meet the Luria and Delbruck
variance test beautifully, and are exceedingly easy to work with. we
are just at the point of adding two or three of them to the routine
battery of experiments, along with resistance to T,. I'll be a lot
happis2 about the whole thing if they work. Other things have been
added to the technique, but it would take too long to go into them al
Suffice it to say that it's a lot better than it was.

As for new results with chemicals, the early indications
that practically everything works are being borne outs, pith the

&enicity,
qualification that voxicity is a cine que non of y
Latest positive tests have been obtained with colchicine, caffeine, -
formaldehyde, urethane (Bryson) and the most potent,of all, believe
it or not, is sodium chloride. We regularly get 10° mutants per 10
survivors with good eld NaCcl. All these things are positive ohly

mt when bacteria are killed, and I have 4 few fancy theories about
the relation of killing to mutagenic action. 8m

There is one correction to be made in the script, and
that is the statement that acriflavine is negative in Drosophila.
Demerec has repeated acriflavine with an improved technique ( much
longer exposure ) and has obtained clearly positive resuts. sSofar,
no known descrepancies between bacteria and Drosophila, tho only
a few of my compounds have been tried on the flies.

I am very sorry that I never answered your last letter,
in which you raised a question about induction of radiation-resistance
by ultraviolet. I don't know if you remember your point after all
this time. In case you've forgotten, you wondered if selection
during irradiation could have accounted for apparent induction of
B/r. pS (B/r) at highest dosage shown on curve (1800 ergs) was 3.5
You asked if killing curve could break sharply enough between 1800 er
and 3800 (dosage used in induction. expt.) to give pS ( d 3800, B/r) of
3.8, which it would have to be.if selection is the answer. This infor
mation should have been included in the paper, wt it is a legitimate
heckle in view of the fact that it wasn't. Actuelly, pS B (d 3800
B/r) is about 6, which I think should answer the question. Sozrry for
the long, long delay. .

I hope you will be able to return the typescript as @
soon as you're through with it, as I have no otier copy, and the
proofs have not yet come. Plesee write when you huve time, arn
tell me how things are with you, and with sex.

Very best regards,