MEMORANDUM 
DEPARTMENT OF HEALTH, EDUCATION, AND WELFAR 
PUBLIC HEALTH SERVICE 
NATIONAL INSTITUTES OF HEALTH 
to : Dr. Donald Frederickson, 
Director, NIH 
DATE: 
February 13, 1976 
Maxine Singer 
from : Head, Nucleic Acid Enzymology Section 
Laboratory of Biochemistry, DCBD, NCI 
subject: Proposed Guidelines For Research Involving 
Recombinant DNA Molecules 
During the meeting o n February 9 and 10 I did net co nsider it 
appropriate for me to make comments on the substance of the proposed 
guidelines. It was my understanding that my primary responsibility 
to the proceedings was as an information source and indeed, I believe 
that to have been the most important contribution I could make. 
Nevertheless, you will not be surprised to learn that I do have 
some specific views concerning the decisions you must now make. 
A. The Experimental Guidelines 
The LaJolla version represents a cautious approach to the experiments. 
Overall, I believe the recommendations afford a high level of confidence 
that those experiments which appear to present very serious potential 
hazards will be satisfactorily contained, either because they are not 
to be carried out at all, at present, or because of the requirement for P4 
and/or EK3. Furthermore, given the evidence that expression of eukaryote 
genes in £. col i hosts is either unlikely or at best inefficient under 
current experimental conditions, it seems that nature itself is ensuring a 
"go slow" approach. 
In my view, the proposed guidelines do not seriously compromise the 
ability to undertake the bulk of the scientifically important problems. 
There is one substantive matter in the guidelines about which I have 
some reservations. A reasonable argument can be made for restricting use 
of SV40 as a vector solely to P4 conditions, when viral particles are 
produced. Indeed, a reasonable argument can be made for avoiding the use 
SV40 as a vector altogether, when viral particles are produced. Thus, 
while I find the present recommendation acceptable (primarily because 
ignorance does not permit positive decisions, even in the face of 
reasonable arguments) I would also find a more conservative guideline 
on this point acceptable. 
[ 454 ] 
