121 
this particular point. And last, the recombination in nature of plant or 
animal DNA and bacterial DNA is also underscored by the recent publication 
of uptake and propagation of eukaryotic DNA by bacterial cells utilizing 
the intracel lularly located restriction enzymes of these bacteria. Now, we 
can complain that Ms. Chang and Dr. Cohen have overstated their case and 
so on, but their results are believable, and they clearly are telling us 
something about the potential of exchange of DNA between diverse organisms 
such as eukaryotic cells and bacterial cells, and there is no getting away 
from that particular point. 
In conclusion, the results of a variety of recent experiments with E^_ 
coli K-12, its Chi-1776 derivative, and various plasmid and phage vectors 
have greatly strengthened our confidence in the EK1 and EK2 host-vector 
systems for recombinant DNA research. This increased confidence in these 
systems, coupled with other information relevant to a particular category 
of a recombinant DNA experiment has, in our judgment, amply supported the 
proposed one-step reductions in the containment levels of certain experi- 
ment s . 
Thank you. I don't know whether Dr. Fredrickson would like to hear 
from Dr. Rowe at the present time, or have questions. 
DR. FREDRICKSON: I think, Dr. Helinski, we will entertain a very few 
questions. You have gone over a great deal, you have presented much factual 
information, reviewed it very cursorily. I would call to the attention of 
the Committee that much of what Dr. Helinski has talked about appears in the 
Green Book. People might want to review that during dinnertime, because I 
would say that we will ask you to come back after we all return. But let us 
take one or two comments now, and then I would go on to the other two to sort 
of have the first reading of the ordinance before we return. 
Dr . Ahmed . 
DR. AHMED: I would like to ask a question of Dr. Helinski. You made 
two statements that I want to go back to. You said it is not how many 
things must go wrong, but how many things must go right. I would like to 
phrase a question in regard to what that really means; I want to get some 
clarification. Aren't there really two questions here in terms of looking 
at the probability? One is the purely biological question about how many 
things must go right from the point of view of genetic engineering so that 
some unknown properties can be either regained or sort of formed. The 
other is purely a containment probability question. This is what I think 
Schwartz and the others have been talking about. So when you talk about 
how many things can go right and wrong, it depends on where you start and 
where you go. That is one question. 
The other question is about R plasmids you mentioned — about the experi- 
ments that show, even with a conjugative R plasmid, that the transfer in the 
intestinal tract is not very high. But were these experiments done in an 
environmental setting, like in a soil situation, where if the organism 
[ 325 ] 
