155 Corona Avenue
Pelham 65, New York

March 2, 1951

Dear Joshua,

I don't expect to do any more bacterial genetics in the immediate
future, if ever, and shall be happy to see somebody going on with
my exprpriments or anything related to them. I think Proline™,B 7
are not strict enough requirements (in wy stocks) to be ideally
suited to the study of complementary segregation, but their crprss-
over frequencies with TL and BM are the best of several I tested.
I tried this cross in at least 2 different ways and had the best
results supplementing with (1) MTL and (2) PB,. You would probably
want Biotin in (1) too. I got good results using either pour plates
or surface inoculation, but you remember I did find a higher yield
and a significant difference incomposition of the MTL prototrophs
when I added small am supplements of MTL. The PB, prototrophs have
to be picked from pretty heavy backround and carefully purified.

I learned to streak out 10 strains on one plate and get good isola-
tions. I also recommend spot-testing in rows of ten: 40 or 50 to
a plate, for nutritional requirements and fermentations; for testing
phage I used two rows of 10 streaks each per plate. I felt that

by reducing the number of plates and doing everything in multiples
of ten 1 could mee reduce scoring errors, a real problem when t=st-
ing 200 strains for a dozen jifferent markers, I also picked the
colonies by tens, using a twelve-depression black porcelain plate,
eliminating the handling of tubes altogether.

Several attempts to split up MTLPB] in
other ways were fruitless, but I didn't
try it with the most prolific strains,
58-161(Proline”) and W677. If I were
doing it over ag&an, I would make this
cross on 4 media simulganeously: PB], BMTL,

BMB, and PTL. Other strains either produced relatively few proto-
trophs or the prototrophs were aminority among mutants and back-
ground; and I'm talking about derived strains; I don't Know why
it should be so,

 

As you guessed, the cross I mentioned in my second MGS note involved
BMP x TLB]. It -would be reasonable to repeat the experiments with
Biotin. Density of the inoculum didn't seem to have much relation
to thickness of background growth, but I used surface inocula of
only .02 ce of the mixture, which produced from 30 to 60 colonies
per plate, the lowest number practicable for picking a 150 - 200,
since many could not be lifted off satisfactorily. The frequency

of"complementaries"from these colonies was as high as from plates
With inocula 5 x as great and colony counts around 200. Some of
the dilute plates had macroscopic background growth in the form of
milky streaks and lines, and this is where i found the other re-
combinants.
I'll be on vacation in June and if we are not in New England
I may get out to Cold Spring Harbor for one or two of the meetings.
But the cost of the trip in time and money and my lack of close
contact with the rather complex subjects would make it hardly
wise for me. I have so much work to do in my field of major in-
terest } I'1l start my internship at the U.S. Marine Hospital
on Staten Island July 1 if a I get my commission in the Public
Health Service, That will solve a difficult financial problem and
of course thedraft, and will give me all the clinical teaching |
I want. SBesides, I may be able to do my human genetics reasearch
Within the Public Health Service; I'm beginning to get some con-
crete aid, I think, original ideas for research,

My wife had a baby girl January 17 and all is going well

except our sleep !

Give my regards to Ester.

Sincerely yours,

lion.

Gordon Allen