in nature and that could be a new source of infection and tumors. 
On several occasions I raised this interpretation to my scientific 
colleagues on the RAC. I have not been satisfied with the responses of those 
who shrug off the Signer-King interpretation. The argument that these newly 
created hybrids are less infectious or less tumori genic than the free polyoma 
does not speak to the fact that the polyoma-phage- E . col i system may bring the 
polyoma to new niches. So we have less infectious or tumori genic polyoma in 
new places. 
(If I were a mouse, with intellects would surely be concerned about the 
possibility of spreading polyoma DNA in E.col i . ) 
There is yet another aspect to the polyoma experiment that has been troubl 
some to me. I have been unable to grasp its general izability. Can we 
expect to have the same result (lower infective capacity of intact virus 
in E.col i than free virus) if other viruses were used in the risk assessment 
experiment? Does the experiment set an upper bound on risk for this class 
of experiments? The clearest argument that was addressed to this point 
was by my RAC colleague David Baltimore at the last meeting. But I found 
his logic at the least confusing and at the most flawed. He maintained that 
the mouse polyoma experiment was among the most sensitive that could be 
designed for determining whether a DNA fragment could be released from its 
carrier molecule while in E.col i . If the polyoma experiment is indeed the 
most sensitive one that can be done for that purpose, then the probability 
of detecting the event is clearly greater than other similar experiments (with 
other DNA viruses and other animals). But the probability of detecting the 
event should be independent of the occurence of the event itself (except 
for some unusual cases in quantum mechanics). 
Since the outcome of the polyoma experiment has been used as one of 
the principal reasons for the exemption of E.col i K12 from the guidelines, 
I would like to see some reviewed scientific papers on these questions that 
meet the tests of time and criticism. And because I found the justifications 
for generalizing the polyoma experiment unsatisfactory, I gave the results 
much less value in the total risk assessment than other RAC members. 
Dr. Stuart Levy of Tufts University's School of Medicine is under con- 
tract to study the plasmid transfer and survival of certain strains of 
E.col i . Dr. Levy has reported that the survival of one strain was higher 
than previously expected. He emphasized the impertance of testing the 
survival of Chi 1776 with different plasmids since he finds that "plasmids 
may influence survival of a host bacterium in yet undefined ways." ( Recom - 
binant DNA Bulletin , July 1979). 
It is for the above reasons that I feel it is premature to exempt the 
vast majority of recombinant DNA experiments with E.coli K12. I hope that 
you defer the exemptions until all the risk assessment work is completed 
and until there has been appropriate discussion within the scientific com- 
munity of the significance of the mouse-polyoma experimental results. 
i Urban 
Social and Environmental Policy 
SK/pw 
[329] 
