emnernee rey

Joshua Lederberg SEP 8 1970

~ —~ “

Sam: I really cannot let this fargument) go
unchallenged, having heard it several times
originally from you.

For one thing, it would imply that
there are no gradations in carcinogenic acti-
vity. For another, mere random fluctuations‘
in a large scale survey will give some obsers
vations a "significantly" high estimate.
(Self-evidently, 1/120 of the shtunkodxNexexx
PREEXCENXLAXKX Bionetics
group should be expected to give a spurious
positive result with apparent significance

p< .O1. ) Whateconfidence do you have that

a replication of such a series would give
precisely the same list of offenders?

In particular, it is certain that
some mild carcinogens are exculpated merely
for having been tested on too small a scale.

So this mxxxgesx kind of study is really
not responsive to the question kY@).
what you should say is that seme compounds
are evidently much more carcinogenic than
others, and that only some of these will be
detected by the tests customarm@ly enforced
at the present time.

If we insist on a pure black/white ap-
proach, we will have to condemn all meat,
milk and fish, if not all food completely,
which is hound to have at least one molecule
of DDL per mouthful for at least another
century.