July a, 195l.

Dear K-a-

Esther and I have bean working lately B bit on the old and thmxy
problm of ~pr+adap&tioan- uhethw phqpa and druga and ao forth
simply saleat pra-axtatent qpontanraotaa mutcmta, or uhether they w
dimutly induur reaiatmue. Our convlcrtiona have baa SrtrongcJr tlrirn
the evideca in f mar of the f bat Qpotheati ae acaount5ng for mot&
if not all esmplsa of inherited adaptive changcss.

In our approarsh to the probleq se have been using what we shall c&.1
a nrep3.ica plating method3t. Coloaies or filars of growth on the, mrfam
of an bM,al agar plate are transferred by mm of a abeet of fabric
(velvcrt or velevetaan) to EI amiss of fresh platee, to which various
wings aray be added. The v&vet may ba thought of aa a Wzbber-sta@ng~l
devius, or better aa a large asmnblq of amall inoeu3.a~ needles held
ti fLxed relative position. At any rata, this mthod haa worked unesq>errbdl~
reU. in preparing replhaa to different nisdia. It was idtidy dlrtiad
to econoti on the work needed in the isolation and olaasifitation of
mutanks tmd raroorrdPWtsp and has been wed in that connection for aeoring
auxotrophs, fsmantatfve dlffaranarss (on E3B agar), and resistance to
drugs or phages. "?4y main question to you is whether such a proaedurs haa
been demr&bad before. I found it r&her difficrult to look into the bitora-
tare for it, for Iauk of a auitabl@ head%ng. The oloa6at I could find ~88
the f%latsuhpr~parat fl, but aa far a8 I could M.l th%a w8a axhluaimly an
irrpre88iot+8w mathod for
geationu on this poiatl I aho%r~u;%8$!%?%%*&. haver atny -

The replioaa are used in Wo ways for the present problem. For examplq,
fiba of growth on plain Bgar mn ba proven to aontain $Loms of pha*
reaiafmta, for aerial replicaa to phaga-croated agar &on a large propor-
tiOA Of cOwFU0At, MQXhRpOSab;24, P@UiStEUit CdOAh. kit bsJWAd thi@, the
phw mutanta man be select& ind2.reMJ.y by taking fresh inooula front the
aitea on the first plate oorrsaponding to the congrucmma on aiatclr, phage-
Croat4 platea. With a fcm auucasaive mrirrhmntar in thiLs wi3y, realatant ibiutmta
have bearA isolated from cml3.a never dimatly (I
We are in the process of afmutsr excperhntu wi % acsd to the aelt#:tive agent,
                           8tX%ptOI3Q%bHWJ18tiMW0,
which ia more difficult owing to the ex%rae)sly low mutation rati.
                            Yours aincsrelg,

Joshua kderberg