a 
HI antibody to be~ a highly sensitive, unambiguous' and convenient test 
for the occurrence of a productive polyoma infection in mice. It is 
especially appealing in the context of the proposed studies which involve 
the use of an E^. coll cloning system because it is a continuing test 
(i.e. a single group of mice, colonized by an £. coll plasmid or phage 
system containing a polyoma component, can be tested repeatedly for 
whether productive virus infection has occurred at any time during the 
prolonged period of observation). 
In the following paragraphs we outline a- series of experiments 
which will utilize polyoma virus to answer some of the questions which 
have been raised about potential biohazards of recombinant DNA research, 
Specifically, we intend to generate recombinant molecules consisting of 
polyoma virus and bacteriophage lambda DNAs or polyoma virus and plasmid 
(Col El or a derivative thereof) DNAs, and clone such molecules in a 
suitable E. coll host. Progeny phage, as well as plasmid-containing E. 
coll will be fed to, and inoculated into mice, in the final step of the 
proposed experiments and the animals tested for evidence of virus in- 
fection. .Although, at first sight, this seems to be a simple, straight- 
forward experiment which could be done cheaply and expeditiously, this is 
not the case. In particular, since we are very concerned that a negative 
result will be misinterpreted as indicating that there is little biohazard 
in DNA recombinant work, we feel that a number of variables must be 
evaluated, and that the experiments should be designed to maximize the 
chances of a positive result. Also, if negative results are obtained with 
a given system, we feel that the experiments should be continued with 
modified systems until we are reasonably convinced that a positive result is 
just not going to occur. Variables to be considered, in roughly decreasing 
order of importance, are: the strain (s) of phage and type(s) of plasmid; 
[Appendix C — 79] 
