prokaryote genes. Any further experiments to be conducted with that 
HVl system would require explicit approval by the RAC. Similarly, 
E. coli host-vector systems (EK) are the only candidates for HV2 
and HV3 certification. For certification at the HV3 level the 
vector must be host dependent or defective in mobilization and, 
if a plasmid, not contain markers conferring resistance to antibiotics 
used clinically or in agriculture. Dr. Gottesman further noted 
that other HV2 and HV3 systems can be considered as they are developed; 
and that testing of such systems would in the process create better 
performance criteria for their final evaluation and certification. 
Similarly, the designation of a system as HV3 would not necessarily 
inhibit its risk assessment since each candidate could include 
a marker that would permit assay of its presence during in viv o 
testing . 
In the discussion that followed Dr. Gottesman' s presentation, 
Dr. Fredrickson indicated that other types of risk assessment experi- 
ments in addition to testing and evaluation of HV2 and HV3 systems 
could be specifically exempted from the provisions of the Guidelines. 
In a related comment. Dr. John Tooze ( EMBO) stated that the polyoma 
DNA risk experiment proposed by NIH was being already carried out 
in Europe. Polyoma CNA lambda hybrids had been constructed, were 
currently being characterized, and would be given to mice. Dr. Tooze 
pointed out that containment levels for cloning of viral DNA in 
Europe are less stringent than those in the United States. 
vii 
[ 193 ] 
