UNIVERSITY OF CAMBRIDGE

DEPARTMENT OF PHYSICS

TELEPHONE:
CAMBRIDGE 55178

MEDICAL RESEAL9CH COUNCIL UNIT

CAVENDISH LABORATORY
  FREE SCHOOL LANE
     CAMBRIDGE

8th December 1954.

Dr. Rosalind Franklin,
Birkbeck College Crystallography Laboratory,
21 Torrington Square,
London, Y.C. 1.

Deas Rosalind,

Many thanks for your letter and the draft paper. I had

heard that Max had had a copy, but as I expect you know, he is
away sick,  and so I haven't been able to discuss it with him.

A few comments occur to ne.

                        On page 4: the reason for
meridional reflection being absent is strictly that the electron
density projected onto the axis is srliooth. For the case you give,
- all the 3n + 1 and 3n + 2 layer lines would be weak, which is perhaps
true up to a point.

Experience in the past has shown that it is rash to include

a drawing with speculation features. It turns up for years and
years, and ones reservations get lost in the process. I should be
very cautious about your figure 3.

At the bottom of page 5: the real amplitudes of' the

equatorial reflections comes simply from the screw axis, and not
of course, because of any pseudo-symmetry. (This is just a matter
of your wording.)

I think you should also be cautious about the X-ray value

for the molecular weight of the sub-unit. How possible are values
of n other than 121 The chemical values, on the face of them, seem
rather good. While we are on this point, I am puzzled about the
molecular weight and hydration of TMV. Eave you been over all the
data carefully? Jhat is wrong, for instance, with i'iilliams'
argument in the Cold Spring Harbour Symposium, 1953, page 1931

     I don't really like the bit on page 7 about  the extra
surface. The groove is,
How niuch surface, in term of , does the chemical work suggest?
If you find you can change the salt concentration of the medium
you may be able to get much better evidence for the groove.

all, not yet firmly established.

aft ei2

-2-

Finally, have yo

              1 considered that we have no partici
reason that the layer-lines are regular, i.e. why the 3n + 1

lar
and

3n + 2 are not staggered relative to the 3n ones; as they are,  for
example, in the a-helix. Do your layer-line measurements show
any sign of this? If not, how does one explain the fact that
the helix repeats exactly after 3 turns?

I hope you won't mind all the criticism. Otherwise the
                         You don't say

paper is fine, though I was surprised you omitted the method by
which you obtained the signs,@ the equatorials.
where you propose to publish I would have thought 13. et B.
Acta. Practically everybody interested reads it nowadays.

Beatrice is still waiting to hear about her job.

Sydney was here for a few days. A great treat. We

managed to cover a lot of ground.

Come and see us sometime.