REIT ,
\ceocral” Oct 29,1952

Dear Joshua, :

I am answering your letter of Sept.18 with more than
one moth delay. I h:ve been busy at work on one strain,and hoped to
get more information out of it before writing you,<cs this might also
involve a change jn symbolism, However,the change is quite trivial
and you may upprove it even if the evidence is not final,or reject it
altogether.

You may remember that in our JGM paper a strain is mentio-
ned,which is a BM-F- I isolated early in Campgdfge.It has the curious
property of crossing with a frequency 25x smaller to F+ than other
F- strains.lt does not cyppss to F-. When I tried to infect it with F+
I did not succeed,even in conditions in which an ordinary F- strain,
in the same mixture,was infected in 100% of the cases. When crossed
to F+, there was a segregation into F+ and F- in the progeny, and, what
is funnier, a 1:1 segregation in 3 iudependankx crosses to 3 indepen-
dent F+ auxotrophs. Analysis of one of these showed no linkage with
any of 7 markers. I do not yet know enough of therproperties of the
prganxy progeny,in respect of F+infectability,segregation xakas pattern.
and F+:F- segregation osrecrossing to make a precise statement,but she
I-have another experiment (still fairly small) showing that this
F- strain is a strain resistant to the virus,and which I am therefore
provisionally calling FY . This experiment can be called one testing
abgorption of F+,although I still have not F4 in cell-free condition.
A suspension of F+ cells is incubated with an excess of F— cells for
45',in aergation, then a small amount of F- cells of a distinguishable
type are added and incubation continued for 30'%,.Finally,the two types.
of F- are separated,and the second addedz is tested for F+ . When,in

_ such an experiment, the first F- added is an ordinary one, all the

kakker cells argxuak of the F- added as second are not infected,althou-
gh they would be infected in controls. The first F- has absorbed all
the available virus which is only slowly formed. When,however, FT cells
are used as the first F- addition,and ordinary F- for the second, the
latter are infected,as if FF had not absorbed the virus. Incidentally,
Hfr behaves in this experiment as FY,

I have no good explanation for the low frequency of recombination,
except that there is also a change in segregation paptern (chromosome
mutation decreasigg exchange in TI-M region’?).As to the 13g" segrega-
tion, perhaps the easiest explanation is that we have hé¥e a locus for
mainéénance of the virus,and that this locusx ig on a chromosome not
containing loci TL, Lac, Gal, Xyl1,Mal,S,B,,Ara,M. I am now testing new
mutants for linkage. If this is true,then we have a situation similar
to kappa,sigma,lamda etc. It might then be better to call ? the virus

Re : . 2-4 re a 1 i . ra . nN an
2.

 

and F,f the alleles permitting and inhibting growth/fhe virug respec-
tively. The genotypes of ordinary F+ would then be F +f , / ordinary
F- would be F+ 0 , EK and of new FT would be f (+0) . T am now cross
sing Frm with non-lysogenic and especially with Hfr to see what hap-
pens. «hat @¢@ you think of altering F into Ff in the joint paper on JGM
I agree that the question is not an important one,but it would help
I take it,to use the new terminology. -

The above experiments have also some bearing on kineticak of
F¥ transfer. I am not planning detailed experiments on the physical
aspects of this problem,because I hope you and Nelson will be carrying
them out.I have only ageertained that dilution reduces infection abdve
a certain point (10° cells/ml is the smallest amount still giving al-
most full transduction).I am still confident that the ratio of Fe:F-
cells is the most important factor. I may continue kinetics under ano-
ther viewpoint, i.e. that of the physiology of the virus. The findings
reported above show that the F+ cell which has infected remains uninfec
citve for some time,as earlier suspected. I have some other data sho-
wing that when a mixture of F+ and BF is ineubated,and the F+ area
minority, there is no tendency to generaliwation of th T+ type,as would
be expected if infection chains were effective enough to do it.This, by
the way,suggests that exppriments of filtration may be insufficigntly

.sensitive for this aim.I have used in the past small amounts of ¢ellsn

added to F+ filtrates,but even if there were enough F+ particles in th
filtrate to infect them, dilution may have maffe the traneduction inef-
fective or inapparent. a

Concerning corrections to the JGM paper : I am preparing a list
of them, inclading all ‘yours,plus some that I have been ripening in
the meanwhile,and shall send a copy of it . The most important refer
to p.18 last PAR ,and to p.21 line 2 ff. ,which regard change of text.
eee

You are yuite right about the Circular flavor of the argument (3
at page 16. I should propose tae following alteration,which per-
haps obtains the sam@kx result a th one you suggest,without too
serious aiteration :

vi

"Hfr forms an apparent exception to the rule that,in K-12,
the presence of F (i.e. of the T¥ State) in one of the mates is
essential for recombination to oocur. This rule would mean etc,"
All the rest unchanged,

Concerning p.21, I am not satisfied about the hypothesis of the 4
different degree of ploidy.It seems very difficult to testxani
numerically; a very rough test indicates that it does not work,
80 that bent? 42 preferable to dispose of it altogether,and re-
main £3" Re ey a Peihaps one way would be to alter the test as
follows : " Paexvgxisxak At least one hypothesis, based on Mende-
(lian theory,can be put forward to account for them: the elimina-—
lagfead. tion of a specific segment of the chromosome contributed by the
F+ parent may take place regularly at every fertilization. There
Tro hyptryis at present no definite evidence to sugsest whether such elimi-
a nation might occur during formation of the Fy gametic cell, during
fertilisation,or at the ensuing reduction." The paper might perha-,
ps end thus, and we cauld delete all the rest. My objection to
your interpretation of elimination after fertilisation is that
homozygous diploids at one locus may originate through somatic
¢.0.in an unreduced diploid.
if you approve these corrections 1 shall incorporate them ;
directly in the proofs when they come,
Finally, I quite agree about your sugeestion,of not giving
Hfr to other workers for the time being.However,i find myself in |
some embarrassment on a specific case. Last summer [I saw Delbrtick |
at Royaumont,and asked him about results M.Vogt had obtained with
Hfr,which 1 had sent to hem from Cambridge two years ago. He said
he and she found it quite normal in recom ination behaviour, and
seemed to doubt the whole story. I therefore promised 1 would send
the strain again, I have not sent it,and think 1 prefer to wait,
in view of what you tell me, We might perhaps agree that 1 send it
later on,at a date convenient for both of us, whdn the major facts
we are interested in at the moment will be known, and reservitg
perhaps some special use of the strain if necessary. By the way,1.
have no progress on cytology.Have you ?
One technical point: I find a method of crossing which lL call
SAT1 rather useful, It is the usual streptomycin azide reinforced
 

by T, in'coupling with azide im view of the close linkage. sx

x Aghn are: incubated together overnight in aereation, then plated
(centrifugation may be useful for enrichment) on StazT,. The yield
is about 10 °. Controls m&xkhz are clear,which is, the only advan-
tage in respect of S Az crossings. Quregietaciee shant LE! .T

find this: useful when testing transductions,as it is-mich nore
expedite than-any other method.. I have prepared sera against F+

i:

and ‘peto see if the difference eould be detected with antiséra~
but have not yet started absopptions, | a
Yours — ot oe
‘ 4 ‘ ‘ 3 ‘
. dice -
4 4 4 5
”
4 ‘ ‘ ‘ ‘
yo : w ‘
\ + _— ‘
, t
4 - & oy ;
} 7
* -~