The central points in cur proposal are the following: 
1. Since the potential hazardous consequences are extremely serious, 
and the probabilities of the hazards occurring are non-quanti f i abl e , 
worthwhile experiments should be conducted as safely as possible, p. 2 
2. Vie feel consensus is possible to defer performing 8 specific kinds 
of experiments. p. 3. 
3. Ecological considerations concerning procaryotic hosts are more 
important than any possibilities of genetic modification ("disarming" 
host-vectors ), as far as reducing human health hazard. pp. 4-6. 
4. He therefore urge the abandonment within 2 years of E. coli and 
development of safer procaryotic host-vector systems. pp. 6-7. 
5. The real levels of physical containment are P3 and P4. P3 needs 
precise description to be meaningful, pp. 8-10 
6. The recombinant ON A committee can at the December meeting specify 
the "EK2" host strain, the specific, individual "EK2" vectors, and 
insure their distribution to all investigators, p. 11. 
7. Experiments putting certain prokaryotic strain DMA into E. col_i_ are 
reasonably done with P1-P2 containment. Experiments putting 
. eukaryotic DMA into E_. col i are hazardous and suggest EK2, 
suppplrnented by P3 or P4. Several of these kinds of experiments are 
so hazardous as to be deferred. pp. 11-14. 
8. Experiments using eukaryotic host-vector systems are described. 
P3 and P4 containment, and the use only of well characterized cell 
culture lines, and genetically modified vectors are supported. Many 
unknowns concerning plant experiments are raised, and we question 
the view that lower eukaryotic host- vector systems should be 
regarded in a light similar to prokaryotic systems. pp. 15-16. 
9. Implementation of guidelines will face certain obstacles. Valuable 
procedures would include a public record of biohazard decisions, 
biohazard impact statements accompanying grant applications, 
epidemiological surveillance procedures, and local broad-based 
safety committees. pp. 17-20. 
The social benefits which may arise from DNA recombinant work will be of 
equal value whether they come in 20 versus 25 years, 50 versus 55 years, 
100 versus 105 years. They will endure, if valuable, for many centuries 
thereafter. For 5 or 10 years now , aslow, thoughtful, research-based 
approach to limiting hazards makes sense. 
[ 351 ] 
