4 
scientists see the Levy results as increasing rather than de- 
creasing the grounds for concern. Curtiss, for example, has con 
eluded that the results and other recent observations on the sur- 
vival of E.coli in mice, humans, and in sewage "would indicate 
that tne cumulative likelihoods for transmission of recombinant 
DNA from EK1 and EK2 host-vector systems are considerably higher 
than previously believed."* 
ii) The "Rowe-Martin polyoma" experiments. (M. A. Israel et al . , 
"Molecular Cloning of Polyoma Virus DNA in Escherichia coli: Plasmid 
Vector System," Science 203 (2 March 1979), 883-892; "Molecular 
Cloning of Polyoma Virus DNA in Escherichia coli: Lambda Phage 
Vector System," Science 203 (2 March 1979), 887-892; "Molecular 
Cloning of Polyoma Virus DNA in Escherichia coli: Oncogenicity 
Testing in Hamsters," Science 205 (14 September 1979), 1140-42.) 
These experiments were designed to test the possibility that 
E.coli bacteria carrying the genes of a cancer virus might become 
vehicles for the spread of cancer genes to human and animal cells. 
The reactions of scientific community are documented in the attached 
article from Environment . As indicated in this article, considerable 
controversy surrounds the interpretation of these results . While 
Rowe and Martin claim that their results have "negated" the cancer 
scenario, other scientists have arrived at a diametrically opposed 
conclusion. ** Furthermore, these experiments have been conducted 
only with the weakened strain of E . coli , 9( 1776 . No experiments 
have been conducted with the "EK1" strains contemplated for use 
in the proposal under consideration. 
11. What is the probability of recombinant DNA experimentation 
leading to aut immune disease? 
No experiments have been conducted to measure this probability 
and scientific opinion is divided as to its significance. Resolution 
of these differences can only be achieved through an empirical 
assessment as proposed in the NIH risk assessment plan ( Federal 
Register 44 (13 September 1979), 53412f.). 
12. Lack of demonstrated hazard to date. 
This argument was used to justify the 1978 revisions of the 
guidelines. In effect, the argument reduces to the claim that 
because no harm has occurred for work conducted under the guidelines, 
no harm will occur when these controls are removed. As was noted 
in 1978, this line of reasoning contains a logical error.*** 
* R. Curtiss to W.Gartland, 11 May 1979, p.4. (Letter in section 3 of 
the "Background Documents.") 
** See also the analysis given in the letter of S.Krimsky to D. 
Fredrickson, 24 September 1979, in section 16 of the "Background 
Documents . " 
*** Testimony of J.King at a public hearing on the proposed revised 
guidelines on recombinant DNA research, Department of Health , Education 
and Welfare in DHEW, Recombinant DNA Research (December 1978), IV, 88. 
[ 634 ] 
