nelson: screening methods in microbiology 325 



between antibiotic production as an industrial process and produc- 

 tion of similar substances in a natural habitat. Even though no con- 

 vincing demonstration of production with chemical isolation of a 

 major commercially important antibiotic under natural conditions 

 has been reported, the suspicion remains that antibiotic production 

 is but a magnified and variant expression of a normal biochemical 

 function (27). Experiments to determine whether antibiotic produc- 

 tion occurs in the soil or to determine the effects of antibiotics in soil 

 have been performed, but none have suggested methods for the ready 

 recovery of new antibiotic-producing organisms. 



The premise that microorganisms produce antibiotics in the 

 soil as a mechanism conferring adaptive value in competition with 

 other soil organisms does not have to be accepted before speculating 

 on the function of antibiotic production. Antibiotic activity may not 

 be the prime property of these substances conferring an adaptive 

 advantage upon the producing organism. The principal groups of 

 antibiotic-producing microorganisms, the actinomycetes, possesses a 

 little but not a lot of morphological differentiation during the growth 

 cycle, both during colony formation on solid media and during sub- 

 merged growth in liquid media. Antibiotic production may represent 

 a sloughing off of structures in passing through stages of differenti- 

 ation (12), such as discarded or abnormal cell wall fragments or other 

 components of the vegetative cell or spore forming cells. Production 

 of bacitracin by Bacillus species, a group of microorganisms that may 

 lie on the evolutionary path of actinomycetes, occurs at sporulation 

 (10, 11). Antibiotic activity may be a secondary and accidentally 

 derived property of these substances. If this view is correct, then sub- 

 stances chemically similar to antibiotics but inactive should occur. 

 None have been found, but the present biological screen would not 

 be expected to uncover them. Chemical tests for antibiotics do not 

 depend upon biological activity but are too insensitive and non- 

 specific. Although antibiotics can be classed with major biochemical 

 groups, none have known roles in biochemistry even when related 

 structurally to normal metabolites. 



An approach to new antibiotics, alternative to random screening 

 of newly isolated cultures, is induction of antibiotic synthesis in 

 nonproducing organisms. Such proposals take one of two forms, viz., 

 (a) the potential to produce an antibiotic may be considered already 



