In the course of attempts to purify the antibacterial agent we had extracted from Spanish moss (Tillandsia usneosides) (1), by adsorption on charcoal, we found that a powerful antibacterial substance for certain organisms could be recovered from charcoal itself. Blood charcoal and previously acidified animal charcoal contained it, but norit did not. This antibacterial substance recovered from charcoal proved to be elemental sulfur?Although sulfur is one of the oldest remedies known to man surprisingly little experimental work has been carried out with it. Colloidal sulfur was first prepared in 1888 (2), and since then many preparations have been used in the treatment of different conditions with varying degrees of success. It has been administe.red orally and by injection in the treatment of mental disease 0-5), by injection in the treatment of arthritis (6-8), and topically, as a fungicide in the treatment of cutaneous infections (9, 10). In both the precipitated and the colloidal forms, sulfur has been used extensively with notable success in the treatment of certain plant diseases (11,12).It is generally assumed that the size of the sulfur particles is basic in the activity of a preparation, the smaller the particles the more active the preparation obtained. In the skin it is thought that the particles of sulfur are converted through the medium of certain cells in the epidermis to sulfur-containing compounds (for example sulfides or pentathionic acid) and that these compounds are responsible for the antimicrobial action (9, 10).We are aware of only two reports of experiments showing the in vitro antimicrobial activity of sulfur upon human pathogenic organisms. The first was of experiments conducted in 1934 by Lawson (13), who investigated the effect of precipitated sulfur incorporated into Corper's mashed-potato medium on the growth of tubercle bacilli. He found that the addition of as little as 3 rag. of sulfur to 100 cc. of medium completely inhibited the growth of this organism. He stated that sulfur appears to have no inhibitory effect upon the growth of some of the ordinary pathogenic bacteria, not specified by him.The second report was of experiments conducted in 1935 by Kingery (14) who'stated that colloidal sulfur was fungicidal and fungistatic for Trichoi~hyton interdigitale and tinea corporis. The fungicidal experiments were carried out by adding a broth suspension of the organism to be tested, to a 1 per cent or 5 per cent dilution of colloidal sulfur. The mixture was shaken for 2 minutes and then several loopfuls were streaked on a suitable agar medium. No growth occurred in the streaks from the preparation a We are indebted to Dr. E. G. Miller, Jr., for these analyses. 531

1. A method is described whereby toxic substances may be extracted from hemolytic streptococci with inactivated serum. 2. Such extracts contain large amounts of hemotoxin and leucocidin. 3. Their intravenous injection into mice causes marked hemoglobinuria, anemia, and death. 4. There is evidence that this anemia is not the only cause of death of these animals. 5. Incomplete work seems to indicate that the hemotoxin and the lethal poisons are not antigenic. 6. Certain biological properties of the extract are described.

1. The same Streptococcus hemolyticus organisms may be subjected to extraction six times in 2 days with untreated inactivated serum with no loss in potency of the later extracts when the organisms are kept frozen solid during the night between the extractions. 2. The serum extract toxins of hemolytic streptococci can be preserved without deterioration for at least 6 months if kept frozen solid. 3. No toxins stronger than those containing 10 units per cc. for mice have been prepared. Reasons for thinking that this is due to the saturation of the serum with the toxin at this point are given. 4. Half saturation with (NH4)2SO4 precipitates out practically all of the hemotoxin in a preparation. 5. Serum extracts were made from strains of hemolytic streptococci other than the Gay strain and attempts were made to correlate the virulence and toxin production from each strain. No such correlation could be established. 6. The principal pathologic finding in mice inoculated with the streptococcus serum extract toxin is a marked degeneration of the tubular epithelium of the kidney.

1. Sterile filtrates from certain hemotoxic strains of Staphylococcus aureus have several toxic properties, of which the most important are the hemotoxic, the necrotoxic, the leucocidic and the property of killing rapidly. 2. The necrotoxic action appears to be caused by a constituent in the filtrates different from either the hemotoxic or the leucocidic one.

No abstract