Page Nine 
September 13, 1978 
What does bother me is how well they will be implemented in 
the real world, and whether the internal checks and balances 
are truly adequate. 
I have said little thus far about either the Decision of 
the Director, or the Environmental Impact Assessment. The 
primary value of the latter is that it provides a discussion 
and justification for lowering containment levels which is 
genrally adequate and upon which I need not comment further. 
The decision statement is thorough and fairly treats the 
numerous points raised both in comments on the RAC proposed 
revisions and at the December, 1977 meeting. Those instances 
where I would disagree with you are covered in my comments 
on the revisions. 
There is only one argument which you used in your discussions 
of the current assessment of risk which I was disappointed to 
see included and which I would prefer to see omitted, particularly 
since it does nothing logically to strengthen your arguments 
for lowering certain containment levels. That is the observation 
repeated in at least three places, that no untoward event has 
been observed in five years of experiments with recombinant 
DNA. There are plenty of valid grounds upon which to justify 
the lowering of containment levels without resorting to a 
nearly irrelevant statement from which the inference the 
reader is presumed to make is that somehow this proves, or 
at least strongly suggests, that recombinant DNA activities, 
in general, are without risk. All it proves is that with 
the systems used, the risk is not great. But, then even 
the severest critics did not predict a high risk. Given 
that most of our experience has been with E. coli systems 
which most would regard as posing little liklihood of risk, 
and that even the worst hypothetical scenarios that one could 
imagine still involve the occurrence of a rare event or sequence 
of events, five years experience and even tens of thousands of 
safely performed individual experiments allow one very 
little basis upon which to predict the occurrence of future 
rare events. Moreover, if one considers the possibility of 
long latency period diseases (e.g. cancer) as a potential 
hazard, then a five year time point is of no value whatsoever. 
While I do not wish to suggest that recombinant DNA activities 
pose risks comparable to exposure to radiation or chemical 
carcinogens, one should bear in mind that it took considerably 
more than five years of living with these agents to learn 
that they posed a considerable hazard. Therefore, even if 
such observations may be reassuring to some, from a purely 
logical point of view, they have no place in evaluating risk. 
[A-136] 
