Aste.

YALE UNIVERSITY

OSBORN BOTANICAL LABORATORY
NEW HAVEN, CONNECTICUT

February 8, 1947.
Dear Mather-

It was very gratifying indeed to receive your recent letter;
Your discussions with me have had a very beneficent effect in
clarifying what I. must do, and I hope your interest in this
problem will continue. The results so far have satisfied the
criteria a, b, and c which you mentioned. I had just completed
my own analysis of the data which I had sent you with substantially
the same results. In addition, I have estimated the homogeneity
of the various samples. Parts of Sets 2 and 4 give 'p's of
-04 and .06 respectively, for reasons which I am at a loss to
assign, so that it is if anything remarkable how well the data
for 'tcoupling' and repulsion phases fit; for sets 1-4 respectively
(with some new data) the 'p's are .005, 0.3, 0.18, .02, which are
commensurate with the homogeneity of the data.

While the 4 sets were chosen from the point of view of the
cycle: BM, BOC, TP, TLB,, I had been so pessimistic about the
possibility of an effect of both BéC/BM and TP/TLB, that I had not
gotten to sit down and analyse the data. The lack of effect of the
first substitution is surprising, but seems to be authenticated
by the data; thank you very much for pointing this out 0 pes ana
for showing how the frequehcy of the 4th rare recom ae class
can be used to estimate the absolute distance(although inefféciently)
On the basis of the comparison of the number of colonies which
appear. on minimal and and TL ager in the BMELB cross, I had come
to. the comelusion that the distance ‘BM-—L was. from 70-30 units.

try

 
Since distances of more than 50 units are cerahinly involved,
pd it is now of crucial importance ‘to determine whetker criteria for
a dav a 2- or 4-strand system can be elaborated; results so far are ambi--
guous.
It is syrprising that only one linkage group has been uncovered,
but that seems to be the situation. I think the possibility of
nt spurious linkages (like B.....M) in your notation has¢ been covered.
There is one interpretation with which I should like to take issue:
',.we must assume that TS is linked to L and P rather than to |
L..e. T..must be further from Ij, than L and P are.' If P,? and

hy L are in linear order (not necessarily this one), Y--P-1 would

 

wth |
com show a smaller recombination frequehcy than f[-TL, regardless of the

fond |
Vu hy order of T and L. In assense, P masks whatever is beyond it. On this
tol lor theore the map should be ? g- -~ — — -t

4 eo | % (ah ?

By ---BAt ----- Tao--~--- Ye pony, B, AvAurtp 6 Low
+ —= + r pas ~ ~~ fo

Fortunately, we have the mutant BP”, so that we can perform the

cross: B, +B-Lac+VSP-T+L+ X ..-t-r+--. The results of the segregations
of Lac and V in the prototroph, By and B™ classes should provide material
for a confirmation (or refutation) of our previous hypotheses.

The question of the relationship between T and L ah be best examined -
in the cross B-T-y7 xX L-B,-v* » Simply by studying the segregation of Vv .
into the prototrophs. I hope to have more definite information in a

few weeks, Till a with best regards,

\ ee, (te shel oye ot) : Ceabewbitedrrusihy Yours sincerely,

| A patataghe : 8S 4K of +9,03 -S, nti
2 kB, He sme, 9, > 6,7 peal

3 B > BY; ~-R 30 +R 50 48S -S 0S

Th deface Lawn O~ ad! dhanddle pets cuclarly steheng .