18 ACIDITY AND GAS INTERCHANGE IN CACTI. 



was drawn into the bulb from the flask. By raising and lowering the mercury 

 reservoir this was rapidly repeated several times, to stir up the air in the flask. 

 This mixing of the gas was commonly aided by shaking the flask gently before 

 attaching it to the sampler, and, when considered complete, some gas was 

 drawn into the bulb and the stop-cock on the receiver was closed. The mer- 

 cury reservoir was then raised and the stop-cock of the delivery tube opened, a 

 vial carefully filled with mercury being held to receive it. The amount of the 

 gas sample taken was usually about 2 c.c. The vial was stoppered and stored 

 away as previously described. 



It is well to mention here a precaution that was observed in every case, 

 namely, that in withdrawing the sample only a very slight depression in pres- 

 sure, not more than 4 or 5 mm., was allowed at any time, so that the error that 

 might arise from internal gases withdrawn from the tissues could be neglected. 

 This is especially important in dealing with massive tissues like those of the 

 cactus joints, which must, without question, hold gases, especially carbon 

 dioxide, dissolved in their juices or held in the intercellular spaces. It is not 

 clear to the writer whether this precaution has always been observed by other 

 experimenters, but it is certainly an important one. It is needless to add that 

 the mercury used was clean and dry as possible, as these samples were to be 

 kept for some time before analysis. The steps herein described are not as 

 complicated as they sound and in a very short time the operator can learn to 

 carry through the whole procedure in not more than 5 minutes. As a matter of 

 practice the writer often set up series of four experiments at 5-minute intervals 

 and when the time for sampling came there was no undue hurry. 



For the purpose of determining the capacity of the receiver with its charge 

 of specimens, another much larger bulb, of the capacity of about 75 c.c., was 

 used. This was tightly wired to the outlet tube as before. The stop-cock on 

 the outlet tube was turned to allow mercury to flow out of the nipple and the 

 whole bulb and adjoining tube were filled with mercury, after which the stop- 

 cock was turned to connect with the receiver. Then the mercury was drawn 

 off to a definite mark in the lower stem, to which the volume had been accu- 

 rately calibrated. Simultaneous reading of the depression of the manometer 

 and of the barometer was taken, giving the necessary data for determining the 

 capacity of the flask. Before making the calculations there were added to the 

 calibrated value of the bulb the capacity of the short end of the small-bored 

 outlet tube of the receiver which projected beyond the stop-cock and that of the 

 bore of the stop-cock itself, all of which formed a part of the system of the 

 exhaust bulb. Possibly this correction was not material, for the actual cubic 

 capacity of these parts was very small, but after it had once been determined 

 it was a simple matter to add this figure to the capacity of the bulb. 



The general method of procedure is the one first used by Bonnier and 

 Magnin and one that has been successfully employed by many experimenters. 

 The gas samples were analyzed by means of the Bonnier and Magnin micro- 

 gas-analysis apparatus," which is too well known to need description. While 

 it is by no means a perfect apparatus, it gives very consistent results when 

 used by a careful experimentalist. All of the samples were analyzed at least 



Aubert, E. Nouvel appareil pour 1'analyse des gaz. Rev., Gen. Bot., vol. 3, p. 97, 1891. 

 Thoday, D. On the capillary eudometric apparatus of Bonnier and Magnin, etc. Annals of 

 Botany, vol. 27, p. 565, 1913. 



