Abstract
The so-called “S” colonies of the human tubercle bacillus obtained by Petroff, 1 Reed and Rice, 2 and others, as described or shown on photographic plates, although more easily emulsified in saline and of greater virulence than R colonies, appear flat, wrinkled, dull, and irregular rather than convex, glistening, round, and smooth-edged as do S colonies of bovine and avian tubercle bacilli, and the S colonies of most other bacteria which have been dissociated. The flat colonies seem to be intermediates and not true S. This paper reports a successful attempt to dissociate human tubercle bacilli completely into S colonies resembling those of the bovine and avian types, and, moreover, to develop or modify a medium favoring the continued formation of such S colonies in pure culture by means of some substance other than anti-R serum.
The Bordet-Gengou 3 medium proved not only an excellent medium for growth, but lent itself well to experimentation. Unmodified Bordet-Gengou medium was used to maintain undissociated control strains. The human strains grown on Bordet-Gengou medium all exhibited morphology characteristic of human type colonies. 3 These appeared to be much nearer the S than the R. It was found that a small optimal amount of ferric chloride, 0.0004%, representing 2.5 × 10-4 mols of iron per liter, added to Bordet-Gengou medium stimulated the dissociation of 2 human strains studied into completely smooth, convex, glistening colonies. These S colonies were produced in pure culture by one strain in the first generation, and by the other strain in the third generation on this modified medium. The fourth and fifth generations continued to produce S colonies.
Avian R bacilli were dissociated to S on modified Bordet-Gengou medium. The optimal amount of ferric chloride in this case was 0.004%, representing 2.5 × 10-3 mols of iron per liter.
Get full access to this article
View all access options for this article.