14 
obtained from culture fluids of Penzcilliwm glaucum in the manner 
above described. The crude #-catalase from animal organs contains 
more proteins as impurities than that from sweated tobacco. 
While £-catalase* behaves like an albumose, a-catalase behaves like 
a nucleo-proteid, since it is not soluble in water alone. It can be 
brought into solution by dilute alkaline liquids and precipitated again 
therefrom unchanged on the addition of acetic acid.” Sweated tobacco 
was finely pulverized and repeatedly extracted witha large amount of 
cold water, to which a few drops of ether had been added in order to 
orevent bacterial growth, and then treated for fifteen hours with a 0.2 
per cent solution of caustic soda. - This alkaline extract, after filter- 
ing, was cautiously acidulated with acetic acid (mere neutralization does 
not yield a precipitate), and the precipitate filtered off after standing 
one hour. This precipitate possessed, in a considerable degree, the 
power of decomposing hydrogen peroxid, while the liquid filtered 
from it did not show this power in the least. This behavior was also 
observed with green tobacco leaves. Six hundred square centimeters 
of healthy tobacco leaves from the greenhouse were crushed to a fine 
pulp with sand and left for three hours in contact with a 0.2 per cent 
solution of ammonium carbonate. The opalescent filtrate obtained 
had ina high degree the power of catalyzing hydrogen peroxid. Upon 
adding dilute acetic acid until a slight acid reaction was reached, a 
fine flocculent precipitate containing the active principle was formed, 
the filtrate from this precipitate being devoid of the catalyzing 
property. 
The a-catalase is, however, not very readily extracted by alkaline 
liquids. Evena 1 per cent solution of sodium carbonate dissolves it 
but slowly. The solution thus obtained from sweated tobacco that 
had been wel] exhausted with water before the treatment with sodium 
carbonate was begun does not give the characteristic blue reactions 
for oxidase and peroxidase (a previous slight acidulating of the alka- 
line liquid is necessary for these tests), hence a@-catalase also has no 
close relation to these two oxidases. In these tests 10 per cent alco- 
hol was added to the extracting liquids to prevent bacterial growth 
during the time of extraction. 
Connecticut tobacco, cured in 1899, was repeatedly extracted with 
water saturated with chloroform. After the removal of the soluble 
compounds this tobacco was left for two days with a 0.5 per cent solu- 
tion of sodium carbonate, just enough to cover the soaked tobacco 
with the solution. Then the liquid was separated from the tobacco 
and 20 ce. of the brown-colored filtrate mixed with 5 cc. hydrogen 
peroxide, and in fifteen minutes 7.5 ce. of oxygen were obtained. A 
1 Special tests have shown that it is not a coagulable albumin. 
? This militates against the supposition that a-catalase might be merely /-catalase 
in a state of absorption, the same way that colored matters are absorbed by 
charcoal. 
