368 THE AMERICAN NATURALIST. [VoL. XXXII. 
in plants for some weeks (three to ten or more in Mr. Zinsser’s 
experiments), and frequently make a feeble growth, especially when 
injected in large numbers. Mr. Zinsser’s experiments include ten 
bacteria and were made on no less than fifty of the higher plants, 
Leguminosz included, unquestionably involving an enormous 
amount of hard work. Pure cultures were always used. ‘These 
were diluted with water and injected by means of a Pravaz syringe. 
In all cases the growth, if any, was restricted to the immediate 
vicinity of the puncture. On examination, bacteria were found only 
in the injured cells and in the intercellular spaces, never in the unin- 
jured living cells as a result of their own activities. The length of 
time the bacteria were able to live in the tissues varied with the 
plants, and was different for different micro-organisms. Even the 
most resistant spores (those of B. subtilis, B. prodigiosus, B. megate- 
rium, and B. anthracis) died in the plant tissues inside of eighty-six 
days. 
6. The author did not succeed in growing the genuine tubercle 
bacilli on agar or gelatin prepared according to Beyerinck’s direc- 
tion and Gonnermann’s. On such media numerous germs were 
obtained from the tubercles after sterilizing their surface, but they 
were non-infectious, and must be considered as associate forms or 
secondary growths (see Beyerinck’s statements). The tubercles 
were sterilized by soaking ten minutes in 1 : rooo sublimate solu- 
tion, or by washing thoroughly in sterile water, soaking for a few 
minutes in alcohol and then burning this off. After numerous fail- 
ures, the true germ was finally isolated on Winogradsky’s silicate 
jelly. In Petri dishes on this medium at the end of eight days the 
colonies were small and white, grew well in the juice of the host 
plants, and produced tubercles on their roots. Living colonies were 
found in anærobic cultures at the end of three weeks. The microbe 
is about 1 p long, and actively motile. Neither spores nor flagella 
could be demonstrated. The colonies obtained from the tubercles 
of different plants looked just alike, and addition of plant juices to 
the silicate jelly did not in any way change the appearance of 
these colonies. The organisms were also morphologically indis- 
tinguishable. 
7. Flasks of water, sugar, magnesium sulphate, and potassium 
phosphate in the following proportions, 100, 5.0, 0.1, 0.1, were 
inoculated, put in a dark place, and aérated with air from which by 
passage through potash water and then through sulphuric acid all 
nitrogen compounds were removed. After forty-nine’ days these 
