Genetics Department, University of Wisconsin, Madison 6, Wis., Febr. 27, 1952

Dear Dr. Hayes:

Thank you for the reprints and your letter of the 20th. I am replying in
haste, for f£ am going out of town presently; also, I hoped this might reach
you in time for Cavalli's visit to London, Cavalli is, I think, very intimate-
ly acquainted with my views and, if he will, can speak for both of us on such
questions as publication. I have proposed to him that since we hava baen col-
laborating as closely as the distance will permit we should publish fairly
soon a full account of our work on self-ineompatibility under an authorship
such 2s Cavaiil, Lederberg, and Lederberg.

There ig very little doubt that we have been working on much the same thing,
although my own interpretations are somewhat more conservative. It will help
in the following Jiscussion to symbolize the "iniective" determinant of selfs
Compatibility aus F+. Such strains as W677 and the BY- infertile strains that
each of us seems to have picked up independently are then F-. This is referred
to as self-incompatibility since F- x F- appeass to be completely infertile.
The transmission of F+ is quite unique, especially in its efficiency-- Cavalli
can give you the details on this. I think your conciusion that recombination
of other markers may be equaiiy frequent is probabiy mistaken. In rather ex-
tensivs tests, vuhe cmiy exchanges that were detected between well-marked F+
and F- invoived F only,as far as could be detected without selective mg thods.
the freyusacy vor excuange of markers other than ¥ is some 10°? to LO? that
of Fr, or do you have some evidences to the contrary? (that is for the conditkm
tions unter which the transmission vi F has been studaied). I think that a care-
fui distinction should be made between the transmission of the F+ agent, what-
aver it is, anc tos transmission of “gametic' material.

4s I wrots previcusly, 1 was nut convincea that the aifferent response of
N-o77 and of 56hk51 to Sli was pertinent to the sexual process itself, but
thougiit 1t might be an ircelevant difference in sensitivity to streptomycin,

As socu 23 I sew your paper in nature (1/19 issue}, however, I tried the
foilowing experlcent which obliges me to wiblhdraw this reservation. ve were
qware, Like yourseli, that She f+ S° «x TLB,- F- 8” (S5d-161 x w-+1277) was
moaerately fertile on Gdeminimal agar, white F* S’ x F- S* was not. If the
difference ware slgnificunt (in ve sexuaiity) vhen one might expect that

F+ S* would be fertile with F+ S°, Since W-677 could readily acauire F+ by
tras duction fron K-l2, the experiaent coula be done witnout any reasonabie
doubt that differences in streptomycin response per se would interfere. This
peoved bo ve corrects W-677F+ was, in my first trial, fertile both with
58-LOLF +37, and with F-S" also. fxcama 1 am obliged to admit, therefore, that
your SM experiments have revealed a second function (may I caii it G), syeh
that a S° is effectively G- in the presence of SM. For a dross, then, one
parent must be G+, the other F+. I think it unneccassary to assume that an
F+G+ cell can act only weoptem uniquely as a "gene donor’ or acceptor. In a
general way, the situation around F and G is symmetrical, and we cannot unequi-
vocally assert the direction of transmission, if there is 4n fact a "direction"
such as anisogamy. From a genetic point of view, in various F+ x ¥- crosses,
the two parents are equivalent. I see no reason yet why we may not atill be
dealing with the union of morphologically equivalent esiements, although the
F-G setup does Boint to some degree of functionel differentiation. To assume

a microgamete when filtration experiments, or simple sedimentation, have

given no evidence for it at all even in sensitive tests, calls, 1 think for
Occam's ragcr, at least for public pronouncements. I say this More emohati-
cally because of our experience with Salmonella transduction where the

genetic and physical evidence point very much the cther way (Spicer has a
resume}. I think I mast not understand your hypothesis of the "self-repro-
ducing gamete". What meiogig is it a product--in the haploid cell? It surely
is not the F+ agent, which dces not carry any other markers when transduced.

I see nothing more infective about recombination in bacteria (i.e., in E. coli)
than in the "parasitism" of any cell by its nucleus. F+ is a unique deter-—
minant for, as far as we know, a single trait: self-compatibility. Whether
this. can be inteppreted in terms of the capacity to produce a microgumete
time will teli. Eut tne genete itself cen no more be identified with the

-F+ agent than with lambda (as. you seemed.te imply in ths Nature article).
Yours gincerely, = 4 «i ss i s, Lesderborg