268 BOTANICAL GAZETTE [OCTOBER 
Components Coal gas Water gas Oil gas 
MN oe, Poe et 5.0 16.6 45.0 
Heating gases : 
Ne et a 34.5 19.8 35-8 
Hy pt ee ay Saas 49.0 42.5 14.6 
codices to cag EE een 7.2 26.1 sees 
Imp 
vo OSIRIS El yt cen ce aa 2.4 1.1 
Seer OU ln Tix 3.0 vee 
Rae OG 17.5 25.0 65.0 
These are average numbers, and considerable variations from these propor- 
tions are often met with. The illuminants are unsaturated hydrocarbons, such 
as ethylene and acetylene, and the value of the gas for illuminating purposes 
depends on the amount of these particular components. 
_ The illuminating gas used in our experiments was water gas of the 
People’s Gas Light and Coke Company, drawn from a tap in the 
Botanical Laboratory. In numerous analyses of samples of this gas 
(see HEMPEL, p. 282) absolute alcohol absorbed 0. 2-0.6 per cent., 
and fuming sulfuric acid 1114 per cent. Absolute alcohol absorbs 
the so-called hydrocarbon vapors (mostly benzene); and fuming 
sulfuric acid the heavy hydrocarbons, including acetylene, ethylene, 
and their higher homologues. Bromin is often used as an absor 
of ethylene. Besides ethylene, however, it absorbs several other con- 
stituents of illuminating gas. In a number of analyses this reagent 
absorbed 9-13 per cent. A more definite determination of ethylene 
will be given in the experimental portion. 
At first the illuminating gas used was washed through 33 per cent 
potassium hydrate to absorb any traces of sulfur dioxid and hy drogen 
sulfid it might contain. This was found not to modify the toxicity, 
and hence the unwashed gas was used thereafter. The methods g 
deriving and purifying the other constituents (of illuminating gas) 
worked with will be described in the later paper, which gives ther 
effects. 
3. Experimental 
ILLUMINATING GAS Se 
As a later paper will deal fully with the effects and toxic - 
of the constituents other than ethylene, we need make only 2 gen©""" 
statement concerning them here. A number of experiments were ™ 
