42 
in the application of these safeguards and procedures, for informing 
personnel of both the potential hazards and the basis on which these 
hazards have been estimated, and for maintaining these practices on a 
continuing basis. 
Appropriate monitoring of personnel (including, where appropriate, 
serological monitoring and checking of intestinal flora) should be con- 
ducted frequently. Evidence presented to this committee has suggested 
that E_. coli chromosomal markers such as nalidixic acid resistance ( nal r ) 
and the inability to ferment arabinose ( ara ~) or mannitol ( mtl ") are 
useful for monitoring the presence of a host organism in the laboratory 
environment and in laboratory personnel. In addition, we recommend that 
NIH designate a data collection group to which principal investigators 
must report any laboratory incident involving a failure of P3 or P4 
containment or any serious or extended illness of a worker. 
The experimental guidelines given here are to help the principal 
investigator determine the nature of the safeguards that should be 
implemented for experiments with different types of recombinant DNAs . 
Because the complexity of types is great, these guidelines are bound to 
be incomplete in some regards. Hence they are not meant to substitute 
for the investigator's own evaluation of the containment conditions 
required for each experiment. Whenever this evaluation calls for an 
increase in containment over that indicated in the guidelines, the 
investigator has the responsibility for instituting such an increase. 
By contrast, the containment conditions should not be decreased over 
[ 113 ] 
