21 
4. A research program consisting principally of standardized repetitive 
procedures is less hazardous than a program that requires comparatively 
frequent changes in technique and equipment. 
5. In the absence of effective vaccination, it is not possible to do 
basic research with a highly infective agent on the open bench top, nor with 
Class I cabinets that are accompanied by all the secondary barriers specified 
in P4, without laboratory infections. (Record during 1944-1950 and the record 
with tularemia in Buildings 525-560 during 1944-1969) 
6. If neither P3 nor P4 requires use of Class III gastight cabinet 
systems, it will not be possible to prevent laboratory infections if an agent 
is used that is as infective as Pasteurella tularensis or Coxiella burneti . 
A definition of P4 that consists of adding secondary barriers (air locks, 
| 
negative pressure environment, showers and change rooms) to P3 conditions 
£ will not result in a significant decrease in the number of clinical or sub- 
clinical laboratory-acquired infections. The addition will reduce the chance 
f 
of infection of unauthorized visitors, personnel in rooms close to the 
recombinant DNA research unit, and a wife at home. 
7. For research on recombinant DNA, the most effective safety measure 
to prevent infection of laboratory personnel is to utilize a microorganism that 
will not infect humans. Otherwise, the greater the required human infective 
dose, the safer it is. 
8. Microorganisms for which there is an effective vaccine, and also in 
some cases specific effective therapy, could be considered for use in 
research on recombinant DNA. 
[ 392 ] 
