686 
SUMMARY OF CURRENT RESEARCHES RELATING TO 
buddings in one or more places along the rod, the angle at which the 
branches came off being very variable. Some of these branches were 
quite as long as the original rod. Since then several observers have 
described the branching of tubercle bacilli. In the summer of 1892 the 
author found this form of the bacillus in the liver of the green jay of 
Mexico, XantJioura luxosa. 
The author inclines to the view that the branched form represents 
an involution life-cycle of this germ, chiefly because young bacilli when 
introduced into the animal tissues produce tuberculosis, while older 
cultures gradually lose their virulence, probably owing to their inability 
to reproduce themselves. 
Spread of Leguminosse-Bacteria in the Soil.* — It had been fre- 
quently remarked by Herren F. Nobbe, E. Schmid, L. Hiltner, and 
E. Hotter, that in top-inoculation only in the uppermost layers of the 
soil did root-tubercles develope, the rootlets extending more deeply 
being free. This fact might receive its explanation either in the 
deficiency of air in the deeper layers of the soil, or in that the bacteria, 
retained by the upper layers and roots, were withdrawn from the dis- 
seminating influence of the irrigation-water. Experiments were made 
with marrow-fats (Laxton’s Prolific) to determine which of these alter- 
natives was the correct one, and also to ascertain if the plants were 
only liable to infection in the youthful condition. 
Inoculations were made (May 16) fourteen days after sowing the 
plants in sterilized non-azotized soil enriched with mineral manure by 
introducing in a sterilized glass tube an emulsion (5 ccm. per plant) of 
a pure cultivation of Leguminosae-bacteria 20 cm. below the surface. 
On October 2 the plants were examined, and the washed roots showed 
tubercles only at the deeper parts, the upper parts being quite free. 
Another experiment, where the inoculation was made in the middle of 
the pot (about 12 cm. deep), showed that the tubercles were confined 
to the neighbourhood of the inoculation site. If two inoculations were 
made, one at the surface and one 1/2 cm. below, there were two tuber- 
culous zones separated by a non-tuberculous stratum. Assuming that 
root-formation is normal, the age of the plant is not in itself deter- 
minative of tubercle formation; young rootlets are only capable of 
infection so long as they possess sensitive hairs. 
Bacteriology of Artificial Mineral Waters.j — Mr. C. Slater, who 
has made a bacteriological investigation of various artificial mineral 
waters, finds that they contain a considerable number of micro-organisms ; 
the number present in the finished waters gives no clue to the original 
purity of the source of supply, or to the efficacy of the methods of purifi- 
cation. When these waters are kept there is a decided decrease in the 
number of micro-organisms present ; and this decrease is most noticeable 
after about three weeks. The decrease is due to the presence of carbon 
dioxide which arrests the development of the bacteria, but does not 
always kill them. 
Aeration undoubtedly increases the safety of water ; it offers a 
practically absolute safeguard, against cholera, since Spirillum cholerse 
* Mittkeil. Pflanz. Phys. Versuchestat. Tharand. Die Landwirtks. Versucksstat. 
xli. (1892) p. 137. See Centralbl. f. Bakteriol. u. Parasitenk., xiii. (1893) pp. 191-5. 
f Journal of Patliol. and Bacteriol., i. (1893) pp. 468-88. 
