95 
DR. BOCK: I heard several statements that there was progress in that 
direction — one, the use of performance specifications for certain of the 
safety cabinets, and second, the description that, under measurement, that 
K-12 and the weakened strains gave certain particular behaviors under certain 
tests . 
I think that using that kind of objective criteria would be preferred 
to saying "equivalent to K-12." 
DR. FREDRICKSON: Mr. Helms. 
MR. HELMS: Are you saying such objective criteria as one survivor 
in 10^? Is that the kind of thing you are saying? 
DR. BOCK: Yes, or under a particular feeding protocol of live organisms 
to body weight of a test animal. 
MR. HELMS: Or with the safety cabinets that less than one ounce escapes 
in a year, or whatever it is? 
DR. BOCK: That is right. I believe that there are good indicator 
organisms that can be used for detection, introduced as aerosols, and you 
can put up a worst-case test very safely, by the way. You can use very safe 
organisms for the test. 
MR. HELMS: Dr. Fredrickson, could we have somebody on the Recombinant 
Committee suggest what difficulties there might be with that, because it is 
obviously an excellent suggestion in the sense that those of us who are 
unable to imagine all the things that have not yet been done, it would be 
easier to have some objective criteria; but then again, there are perhaps 
some problems with that. 
DR. FREDRICKSON: Dr. Gottesman, I think, is ready to respond. 
DR. GOTTESMAN: The instructions to investigators that I referred to 
when I spoke, I think, come the closest to meeting what you are talking 
about. What they are are a series of tests specifically designed to test, 
for instance, survival and transmission for EK2 organisms, and I think in 
developing those we have started to have a much clearer idea of the kinds of 
objective tests that we could set up for such things. So there does exist 
such a document, for instance, for testing E. col i plasmid and phage vector 
systems, which is not part of the Guidelines, but is an official something 
that has been voted on by the Recombinant Advisory Committee, and that Bill 
Gartland presumably sends out to people who ask for it. 
I think that it is hard to come up with very general izable tests. I 
think we will certainly use these tests as a basis for looking at other 
host systems, and we are starting to put together the same kinds of tests 
and numbers for EK3 testing. But to some extent it is hard to make the 
test until you see the experimental data and you discover that you really 
[ 299 ] 
