21 
creation of, infection of personnel by, or escape from the labor- 
atory of a harmful organism containing recomyj»inant DNA could have 
a world-v/ide and irreversD^le effect. Radioactivity is diluted 
in the environment; an organism capable of survival and repro- 
duction could multiply and spread. 
Is it to be assur>ed that a lab worker will never use a 
culture of an EK-1 strain of E. col i K-12, when the Guidelines 
specify an EK-2 strain? Or is it guaranteed that a culture 
flask of cells containing recombinant DTIA will never be spilled 
or dropped? Will tubes break in the ultra-centrifuge? Or an 
autcclave fail and not kill potentially hazardous organisms 
before disposal? Similar errors and mishaps occur under the 
best of conditions with the most competent and best-trained 
personnel. There is no reason to assume that they will not 
happen with recombinant DIjA experim.ents . The impact statement 
should properly have considered the consequences of human and 
mechanical failure. This it has not done. Instead, conclusions 
are drawTi on the basis of qualified statements expressing the 
ideal: "...under standardized procedures employing stable, 
v;ell-trained and well-disciplined personnel" (pg. 62) . To 
assume this level of perfection in all cases is unrealistic. 
Therefore, to conclude, as the impact statement does, that 
adherence to the Guidelines w'ill essentially guarantee public 
safety, must also be considered unrealistic. 
Appendix K — 75 
