UNIVERSITY OF CAMBRIDGE

PROFESSOR R. A. FISHER, Sc.D., F.R.S. DEPARTMENT OF GENETICS WHITTINGEHAME LODGE
MISS M.F.1. SPEYER, B.Sc. 44 STOREY'S WAY
Research Assistant and Secretary 2 3r a Ty

« dune 199. CAMBRIDGE

Tel. 55822

Dear Dr. Lederberg,

Thank you for your letters of June 9th. and
pa 6 Vth. I hope to receive soon your strains, which will
certainly be all very useful to me.

My time has been devoted mainly, in the last two months, to
the new stock of which I wrote you in my last letter. Your
questions are all answered by the fact that the stock is a
derivative of 58-161, end has kept the same vhysiological and
biochemical characteristics. I agree as to the kinretical
difficulties you put forward. AS a matter of fact, the frequency
of prototrophs is increased only 2-300 times when the mixture
of 58-161 with high frequency of recombination (Hfr) and W 583
is plated in agar depth, in respect of the usvel cross 58-161
x W 583. But if the same mixture is plated on the surface of
agar, the increase is nearly sot times, and when few (1400-200)
cells of one of the two strains (asser of 50-161 Hfr or W 583)
are plated with an excess (10°-109) of the other strain, the
frequency of prototrophs in presence of aneurin varies between
10% and 100% according to conditions.

The Hfr stock was obteined through selection for nitrogen
mustard resistance, but probably the two phenomena, nitrogen
mustard resistance (ny®) and Hfr are uncorrelated, for (i)

further indenendent ny® mutants do not show the same behaviour,
(ii) after recombination the two characters ny® and Hfr seem
somewhat independent in segregation. Incidentally, resistance
to nitrogen mustard seems to be due to several genes with grossly
additive action.

The cross 58-161 Hfr x W 583 shows many discrenancies in the
gene frequencies observed in the prototrophs. With the exception
of Gal+, the frequency of which is very markedly reduced, the
frequencwyof Lac+, Mal+, Xyl+, v; and By are significantly
inereased. Thus I wondered whether the vrototrophs which arise
in this way might not be mixed, and show the vhenotype which
would dominate in’ the mixture. T have found that most
prototrophs are mixed with respect to one or two characters.
Diploid formation does not seem to play a role. Of the two
major explanations: a) more than one zvgote taking part in the
formation of the prototrophic colony b) 4-strand c.o. with
survival of sk meiotic products, I cannot vet favour any

Microscopnic examination has not yet given great help, though
serial photographs of a developing plate show that some bacteria
which are at the beginning clearly separated lie closer together
after some time. Unfortunately the examination is made
difficult by the fact that microcolonies up _to 100-200 bacteria

Sr AN A tant me

are often formed, probably due to traces of growth factors in

Difco agar.
I was very interested in your communication of branching
at the Maltose locus. I was realisine that something went

wrong with the Mal locus. In order to make a comparison with
Ve ,

B, (GA) gat tee WM (Th)

—T x 5 > ~ 1
we (4) Am ey (708 (iv?

 

i" ’ 4 s

ts
{ ‘ v) { wi) ny) v Wy tee lu) ‘y
Vv . v

Lac ~ £ 4 4 = = . +
Vy R S S R R S S R
Gal - 7 - - + + 4
With B 58 95 Wh 6 6 9 19 2 pe 234

4
Without B, h2 96 36 2 4 2 27 2 an
yAt

One might assume that the epistatic action of Gal+ on Lac has not
been fully eliminated by scoring on lactose at 3%, and that
therefore the 18 Lac- Gal+ observed are really gal+ lac+; but,
even so, one should assume too many triple crossovers to allow
for the segreration of vi. It is also to be noticed that the
Gal segregation is different from any other in the Hfr strain.
Unfortunately I found Ara~, which is closely linked to Gal and
might have helped to explain, very difficult to score in a
consistent way.

A disturbing phenomenon which I noticed is the apnearance
of mixed vrototroph colonies in about 2% of the cases, and
especially from uncrowded plates (supplemented with threonin,
or with no supplements). The mixture can be easily observed
from the fact that, on E”B plates with T,, some mixed colonies
made of Lae’v> and. Lae"v", for instance, give easily recognisable.

4 1
peculiar streaks.

I am very anxious to know more of your results. The
situation seems rather complex, but no doubt very interesting.
As concerns ENB, it works now quite properly, the aifficulty

lay in methylene blue.
the data of 58-161 Hfr, I have collected a certain number of
data on 58-161, end I have prepared a summary of them, should
they be of any use to you. A comparison with your data will,
at any rate, be of sreat help to me.

Xyl and Mal seem to be (in W 583) rather closely linked,
on the left of BM, at about 15% of the distance between BN and
the B, locus. Their mutual relationshins are, however, not

4
fully clear. These are the data:

Prototrophs in presence Prototrophs in absence
of vitamin B, of vitamin B,
xyl*Ma1* 2 15
+ oo 0 8 (
. y af
_ - had. . zw | Me a .
+ 5 6 pial
- 7 2314 179

It is possible that there are two chromosomes with a
reciprocal translocation. Another possible way of explaining
the results is, of course, that of assuming a higher average
number 6f chiasmata than apparent at first sight. Ne.

I was thinking of a "branch" in the case of Galactose, to
which it is very difficult to assign a locus on the main
chromosome on the theory of random segregation. Gal+ segregation
is not influenced by the presence of aneurin; Gal+ is linked ©
with BM and should therefore lie close to Lac, but even so the

data given below do not favour a linear order.
I shall be grateful if you will send me a repvrint of your

very interesting paper on heterozygous strains.

Yours very sincerely,

Lp Crk