266 BOTANICAL GAZETTE [OCTOBER 
up the experiment the cork is placed on the stem of the carnation by 
opening the cork at the split and inserting the stem. The flower or 
bud and the long tube are put into the nozzle of the bottle and the 
cork forced in until the whole apparatus is air-tight. The free end of 
the calcium chlorid tube is placed into a dish (h) of water or (with 
gases highly soluble in water) mercury; both pinchcocks (e and g) 
are opened; and suction applied to the tube / until the liquid rises 
to the small portion of the calcium chlorid tube, at which time pinch- 
cock g is closed. The desired quantity of gas which is now poured 
into the wide end of the calcium chlorid tube rises to the top of the 
liquid. A one-holed rubber stopper, furnished. with a tube and 
attached to a column of the same liquid as is contained in the dish, 
is now inserted into the free end of the calcium chlorid tube (d), and 
the pressure of the column allowed to force the liquid to the inner end 
of the long tube. This forces the gas into the end of the bottle 
farthest from the flower and allows a gradual distribution by diffu- 
sion. For ethylene and illuminating gas water was always used as 
the forcing liquid. 
In determining the toxic limits of illuminating gas and ethylene, 
20-liter carboys were used; while smaller bottles were employed 
some of the earlier experiments with these gases, as well as with all 
the determinations of the least toxic gases. The question of ae 
effect of corking a bud or flower in a closed chamber of this kind 
naturally arises, and suggests a criticism upon the method. It was 
found that flowers opened without any apparent injury when corked 
in flasks of only one liter capacity. In all checks and in all eS 
where the concentration of the injurious gas was below the toxic limit, 
the flowers bloomed normally while yet in the bottles. To avo 
undue rise of temperature within the chambers basket-covered CaF 
boys were used. The experiments were carried on in the labor atory 
greenhouse during the months of May to September. The tempe* 
ture in the experiments reported varied from 20°-28°, and within this 
Tange no noticeable variation in toxicity appeared. 
Two varieties of carnations were used—the Boston Market and 
the pink Lawson. The two varieties vary so little in their sensitive 
ness and reaction to ethylene and illuminating gas that a description 
of the responses of one applies equally well to the other. 
