362 : BOTANICAL GAZETTE [MAY 
and various smokes) in which the response is quite certainly due 
to one of these constituents. It also omits methane, for which 
it is not yet certain whether the response produced by the very 
high concentration given above is due to methane or to impurities. 
This uncertainty exists because methane derived by three different 
methods gave extremely great differences in magnitude of toxicity, 
though analyses showed the three sorts to contain approximately 
the same percentages of methane. At most, methane is very 
slightly toxic, if indeed further experiments do not prove it entirely 
harmless. 
PARTS PER MILLION OF ATMOSPHERE TO PRODUCE 
GAS USED fax ; - , 
nhibition o Seer orizontal nutation 
growth Declination and swelling 
Dok oi 5. es Ss fee 0.2 0.4 
PREV ONON Soi eS SG ee aes 5 100.0 250.0 500.0 
ne hs se 75.0 1000.0 1000.0 
Carbon monoxide. ...........: 5000.0 5000.0 10,000, 
Biethane.........-...------- 60,000. 0? 200,000 . 0? 500,000. 0? 
feiss i oe None None 
Hydrogen sulphide........... 500.0 None None 
POR iain pk eek 3000.0 None None 
The nature of the response (triple response) of the seedling to 
paper and tobacco smoke shows that it must be caused by one of 
the four carbon-bearing gases mentioned above or by homologues 
of some of them. Before going into the probability as to which of 
these determine the effective limit of the smoke, let us consider the 
concentrations of other constituents in unwashed tobacco smoke 
and the chances that they may play some part, at least, in the 
inhibition of growth, in experiments such as reported in this paper 
or those performed by Mo iscu. 
According to LEHMANN (18) 100 gm. of tobacco when smoked in 
a pipe or as a cigar produces about 16 cc. of H,S. In experiment 
II (3) in which the smoke from 0.0375 gm. of cigar is placed in a 
50-liter can, NELJUBOW’s third response appears. The H.S con- 
tent in this experiment, if it had not been washed out, would be 
I part in 8 million of atmosphere, or about 0.0002 sufficient to 
reduce growth. According to the same worker, 100 gm. of tobacco 
when thus smoked produces 0.935 mg. of NH;. In experiment II 
