In the early 1940s, Luria began to a question how phage-resistant bacteria originated. Did exposure to the phage induce the
mutation to a resistant form, or was the mutation random? The answer came to him in February 1943 while observing a colleague
at Indiana University play a slot machine; Luria realized that random mutations would produce bacterial clusters in a pattern
analogous to slot-machine returns. After completing several preliminary experiments and noting the results, Luria wrote to
Delbruck at Vanderbilt University for comment. Delbruck responded that Luria had indeed stumbled on to something and put together
a theory to explain it. Among other things, their "fluctuation test" theory demonstrated that bacteria were ideal
subjects for genetic research. This letter from Delbruck is part of a series of exchanges which followed. This research resulted
in an article in "Genetics" in November 1943.
Number of Image Pages:
1 (101,978 Bytes)
1943-04-05 (April 5, 1943)
Luria, Salvador E.
Original Repository: American Philosophical Society. Library. Salvador Luria Papers
Reproduced with permission of Tobias Delbruck.
Genetics Lessons from Bacteriophage, 1938-1944
Mutations of Bacteria from Virus Sensitivity to Virus Resistance (November 1943)
sorry to hear that you are subculturing my strep in your throat, and that your doctoring is less effective on yourself than
on me. Hope you got a good rest from it anyhow.
The interference experiments on Shiga sound promising. It may well be that the interference is mutual, as with delta, in the
sense that most of the bacteria with mixed infection become neurotic and liberate neither phage. I dont see how the plating
eff. can come in as a disturbing factor, dont you use our old indicator strains A and C.
I have done some experiments with anti-gamma- serum. The serum is quite powerful. At adilution 1:5,000 90% of gamma is inactivated
in 10 minutes, and inactivation is exponential with time. I did one interference experiment, alpha and sensitized gamma,
alpha single, sensitized gamma multiple. Result no suppression of alpha growth.
I believe there is an error in your calculation of "a" from your experiment, and I believe the error is in the last,
third, calculation. This should give the same result as the second calculation. It seems that you forgot to drop the factor
"C" in the last calculation. If you lump all tubes together to give one total "N", your number of samples
must be taken as one. I believe the second calculation is the really justified one, except for the comparatively small Poisson
error due to sampling from the tubes. I think one should expect larger number of mutants when sampling from larger cultures.
In the limit, when one samples from infinitely large cultures one would expect simply aNt mutants, and only Poisson fluctuations.
I had at last word from the new man, Edward H. Anderson. He plans to be here July 1st. I wrote to him exploratively regarding
the state of the nation at Pacific Grove, and that you or both of us might want to go there in the summer. Will let you know
as soon as I here. Mean while it seems more probable every day that I will not be able to get away in the summer, but I told
Reinke that I would not even try to persuade you to come here in the summer unless I had positive
[NOTE: The rest of the letter is handwritten]
assurance from the Dean that my teaching load is light enough to give ample time for research. Reinke is writing a strong
letter in this sense to the Dean, and I will speak to him after he has received the letter.
Did you know that mail service to Switzerland has been suspended for months? I am just now getting all the letters returned
which I wrote since December.