18 . 
- Scant precedent exists in universities and other institutions for 
matters of safety to be decided in a pluralistic manner with the 
participation of all those who incur risks. 
In addition, we do have precedents that point to the very real difficulties 
of organizing a molecular biological community around safety issues especially 
on a voluntary basis. Dr. Andrew Lewis, for example, had an apparently dis- 
appointing experience in relying on voluntary compliance by investigators 
interested in studying the SV40-adeno virus hybrids that he isolated. Major labs 
did not support, or broke their support, or simply ignored a Memorandum of 
Understand i ng and Agreement to restrict distribution oF these viral strains. 
Furthermore , the casual handling, in molecular biology labs, or reagents that 
are known to be carcinogens is already a source of concern in many quarters. 
The handling of organisms of yet unknown properties that have the capacity to 
multiply can only increase this concern in many of us. Finally, disappointing 
precedent comes from work involving entirely known biohazards. The number of 
reported acquired infections in laboratories with special containment facilities 
has been around 1650 in the last 30 years (Science, 1 87 , 931 , 1975). 
In the light of the above considerations, we would like to see the 
following elements in a review system: 
1. - Biohazard decisions should be a matter of public record including 
the arguments for and against the decision. 
2. - Grant applications should include a Biohazard Impact Statement. 
This would describe not only local hazards to lab personnel but 
possible danger to the general public including possible long term 
effects. It would also serve as a means of self-education for the 
investigator. The impact Statements should be readily accessible 
to people in the lab to encourage discussion of safety issues. 
[ 369 ] 
