only consider the effects of producing a toxic substance through 
recombinant techniques but also of concentrating a naturally 
produced substance or increasing a rate of an exchange that 
might happen only rarely in nature. Until we can anticipate the 
results of altering an established ecological balance, such 
experiments should not be exempted from the Guidelines. 
This same consideration applies to self-cloning experiments 
exempted by the Guidelines. The Environmental Assessment section 
acknowledges that hormones and enzymes are among the proteins 
that might prove harmful to other organisms (p. 33103). 
Amplifying a naturally occurring protein within an organism 
should not a priori be accepted as a safe experiment to be 
exempted from regulation. We need to know much more about its 
interaction in a natural setting before we discount its risk. 
The Proposed Revised Guidelines exempt recombinant 
DNA experiments using naked DNA on the grounds that it is 
"extremely unlikely to be hazardous" (p. 33107). In addition, 
containment experiments involving animal cell transf orming 
viruses have been substantially reduced based on the supposed 
inefficiency of infection by naked DNA. Both these decisions 
do not take into account the results of the Rowe-Martin risk 
assessment experiments with polyoma virus which were reported at 
the Falmouth Conference. This experiment showed that naked 
polyoma DNA, when injected into the bloodstream or tube-fed into 
-4- 
[ 343 ] 
