During the course of an experimental study of bacteriuria, specimens of urine thought to contain Bacillus prodigiosus which had been injected into the circulation became contaminated with Bacillus subtilis. In an effort to rid them of this contamination, the effect on the two organisms of a number of common dyes was studied and a striking selective power possessed by gentian violet was brought to light. It is the purpose of this paper to record the original observations and to report an extension of the study to other bacteria.In my first experiments the dye was applied directly to the organisms, and its effect was noted after transplanting the stained bacteria to agar. Five drops of a saturated aqueous solution of gentian violet 1 were added to a forty-eight hour broth culture of the organism to be tested. The mixture was shaken well and allowed to stand for one hour. Transplants were then made to agar, a control inoculation of unstained organisms being made at the same time upon agar in a second tube. After twenty-four hours' incubation at 37 ° C. a striking selective property of the dye had asserted itself. When Bacillus prodigiosus was the organism used, no difference was observed between the tu~be inoculated with the stained, and that inoculated with the unstained organisms; growth was equally luxuriant in ,both (figure I ). In the case of Bacillus subtilis, however, the results were quite different, for the stained subtilis failed to grow (figure 2).

AND J. CHURCHMAN, B.Sc.THE passage of amino-acids across the placenta has been studied for many years. Morel and Mouriquand (1 9 13), using the Van Slyke (1 9 12) method of amino-acid estimation, found no difference between human foetal and maternal whole blood amino-acid levels, and were thus able to postulate simple diffusion of amino-acids across the placenta. Rabinovitch (1914), using the same method, found the foetal to exceed the maternal levels.Morse (1917) studied 17 pregnant women, and found a higher level of presumed amino-acid nitrogen in plasma from the umbilical cord than in the maternal plasma. His mean values, using the gasometric nitrous acid method of Van Slyke (1912) which estimates all primary aliphatic amines, were 6 . 4 mg./100 ml. in the maternal and 8 -7 mg./100 ml. in the foetal plasma. Hellmuth (1924) and Plass and Matthew (1925), using the Fohn (1922) method, also found higher foetal than maternal plasma levels, suggesting either a diffusion gradient from foetus to mother, or an active process involved in the transfer of amino-acids from mother to foetus. Luck and Engle (1929) disputed this evidence, and revived the simple diffusion theory.Settlement of this question awaited the development by Hamilton and Van Slyke (1943) of an analytical method which was more specific for amino-acids. Christensen and Streicher (1948), using this manometric ninhydrin method, showed that the amino-acid level of foetal plasma was considerably higher than that of the mother, foetal/maternal ratios being 5/1 for the J . biol. Chem., 149, 255. Chem., 150, 231.C. (1951): Human Biol., 23, 61. 88,230.Paris, 75, 643. Brit. Emp., 58, 701. Hopk. Hosp., 36, 393. Gynaec.

If a small amount of aqueous solution of acriviolet, gentian violet, or acrittavine be added to a thick aqueous suspension of a young culture of B. anthracis it will be fouud that after a longer or shorter interval of time a large proportion, if not all, of the organisms are changed by the exposure from sharply Gram-positive to sharply Gram-negative. The speed with which this change takes place varies with the strain of B. anthracis examined. In one strain studied, Gram reversal began in a few minutes and was complete within 2 hours. In other strains 19 hours were required to produce the change and even at the end of this period a few Gram-positive individuals were sometimes still found in the smears. The Gram-negative forms of Bacillus anthracis produced by exposure to these dyes are notably smaller in calibre than the Gram-positive forms, the difference in diameter as measured by means of a filar micrometer being in the neighborhood of 40 per cent.That such changes could be produced in this organism by exposure to aniline dyes was first observed about a year ago. The present communication is concerned with studies made, since that time, of the reversal of Gram reaction, and of the change in calibre, which the dyes caused; and with attempts to interpret the significance of the two phenomena.