46 Mr. D. Thoday. Experimental Researches on [June 11, 



ajjparatus. To avoid contact with the outer air between drying and weighing, the 

 mateiial was weighed in a tube which fitted into the open end of the wide drying tube 

 (D) ; the joint was made fast by a short length of wide rubber tubing (E), and air, .dried 

 by passing through tubes containing calciuni chloride and sulphuric acid, was drawn 

 through the whole by means of a water pump with a water resistance to keep the suction 

 force steady. After drying, the material was tilted back into the weighing tube, which 

 was then removed, closed at both ends, and left to cool in a desiccator. When cool, the 

 tube was opened for a moment to equalise the pressure within and without, and weighed 

 closed, and again weighed after removing the material. 



H 



Fig. 8.— a, spiral tube in which air is heated in passing to B, the wide tube in which leaf 

 material is placed to be dried. A and B pass through same cork, C. 1), weighing 

 tube, fitting into wider open end of B. E, short length of wide rubber tubing, 

 making D fast to B. ¥, glass stopper for small end of D ; the other end is closed by 

 a rubber stopper. C), cylindrical steam l)ath. H, glass tube functioning as condenser. 



By this method O'l to 0"3 gramme of loaf material was often dried and weighed 

 repeatedly without the successive weights obtained varying by more than one-fifth of a 

 milligramme. 



After the material had been di ied roughly in an oidiiiary water oven, it was fomid 

 sufficient for most purposes, for instance in investigating the asymmetry of leaves, to 

 jKiss a moderate current of air for 15 to 30 minutes before weighing ; but the material 

 analysed by combustion was always very carefully dried, usually for an hour at first, 

 and then two or three times for jjeriods of half an hour to ensure absolute constancy of 

 weight. 



A really satisfactory method of drying having l)ecn devised, attention was turned to 

 the metho<lrt wliich were used by Sachs and others. . 



Some conifKirisons have l)een mack; of tlie results obtained by drying in an ordinaiy 

 water oven and weighing in glass weighing ))ottles, with the results found after drying at 

 100" C. in a current of dry air. In making tliem it was found to be impossible to dry to 

 ii constant weight in the oven, and the ])ortions of material from two halves of a leaf 

 were, instead, dried together for the same length of time and treated as nearly as possible 

 in the same way in every respect. 



