1909.] 



Vegetable Assimilation and Respiration. 



45 



Section VII. — Ox the Killing and Drying of Leaf Material. 



In view of Brown and Escombe's suggestion that the colloids of the leaf 

 may undergo an alteration in their power to retain water, it was essential 

 that special cave should be taken to ensure the complete drying of material 

 to lie analysed by combustion. 



Brown and Escombe dried their Catalpa leaves at 100° C. in a current of 

 dry air or dry hydrogen. As dry leaf material is extraordinarily hygroscopic, 

 I have used a modification of this method in which all contact with the 

 outer air is avoided until the material has been weighed. 



The pieces of leaf were first killed by steam as Sachs recommended,* then 

 dried roughly in a water oven ; and, finally, dried at 100° C. in a current of 

 dry air in a special glass apparatus enclosed in a steam bath.f 



The requirements for subsequent analysis made it desirable to use comparatively 

 small quantities of material ; therefore, to avoid the relatively great loss which 

 powdering would entail and the trouble of estimating it, the material was so treated 

 that it would easily fit the porcelain boat while remaining intact. 



This was done in two ways. In using the templet method the pieces of leaf, after 

 being cut out, were rolled up into cylindrical fonu, retained in this form by a spiral of 

 platinum wire, and killed by suspension in steam. In the process of killing, the roll 

 shrank in diameter and became flaccid, and could be removed with ease, without fear of 

 its becoming unrolled. It was then dried in a drying oven, on a curved metal support. 

 Care was taken to put the free edge below, and this precaution, aided by the curvature 

 of the support, prevented the roll from curling out of shape in di'ving. 



The other method of prepai-ation was connected with the use of the rotating punch. 

 The circular cutting edge had a diameter of about 1 cm., and the discs so cut from one 

 half-leaf were threaded on a thin weighed glass needle. After killing, by suspension in 

 steam, a number of such needles of material were put into the ordinary drying oven in 

 special supports made by sticking a number of short narrow pieces of glass tube radiallv 

 into a bung, the needles fitting into the radiating tubes. 



The apparatus in which the final drying took place (see fig. 8) was of such a shape and 

 size as would conveniently accommodate the porcelain boat in which the material was to 

 be burned, so that it could be dried fiually in the boat immediately before analysis. The 

 air was heated, before entering the wide tube (B) in which the leaf material was placed, 

 by passing through a narrow tube (A) sealed to the wide tube and spirally coiled round 

 it. Both tubes jiassed through the same cork (C), which supported them in a speciallv 

 constructed steam bath,(G). The whole formed a convenient and comjjact piece of 



* There is a theoretical objection to putting pieces of leaf straight into the oven in 

 which the preHminary drying is to take place, for the rate at which the material is heated 

 up is relatively slow, and respiration, which increases greatly with the temperature, must 

 entail some loss. The error is probably small, especially when the portions to be compared 

 receive identical treatment. On the other hand, the rapidity with which steam kills the 

 material, owing to its high latent heat, ensures the reduction of all such errors to a 

 minimum. This method has been adopted for all except a few early experiments, and 

 identical treatment given in every respect to portions which were to be compared. 



t Heating to 108" C. in a toluol bath drove otf no more water. 



