122 RESEARCHES UPON ATOMIC WEIGHTS. 



acid lost the amount of water theoretically necessary for the formation of the 

 pentoxide, and further heating to 300° produced only a negligible diminution 

 in weight. On one later occasion, however, when a comparatively crude speci- 

 men of acid was being dehydrated, the second phase unexpectedly appeared, 

 and in all subsequent experiments the same result was obtained, no matter 

 what the source of the iodic acid, or how carefully the apparatus was cleaned 

 before use. This point is additional proof that the second phase is a definite 

 chemical compound. The obvious explanation of the phenomenon is that the 

 laboratory became inoculated with "germs" of the second phase so that its for- 

 mation was thereafter always induced. 



The iodic acid was dehydrated in a platinum boat contained in a hard-glass 

 tube heated by a solid aluminum oven (fig. 4, page 78). When heated with a 

 Bunsen flame, this bath could be regulated in temperature to within one or 

 two degrees without the least difficulty, even without the jacket of asbestos 

 paper which was usually employed. Thermometers placed inside the hard- 

 glass tube and in a cavity in the blocks registered the same temperature within 

 a degree even when the oven was heated to 300°. 



At first the air was purified and dried by passing over hot copper oxide in a 

 hard-glass tube, then through two towers containing beads moistened with 

 silver nitrate and caustic potash solutions, respectively, and finally through 

 four towers containing beads moistened with concentrated sulphuric acid to 

 v/hich a small quantity of potassium dichromate had been added in order to pre- 

 vent any possibility of reduction. The apparatus was constructed wholly of 

 glass with ground or fused joints, rubber and grease being carefully avoided. 

 When heated in a current of air thus purified and dried, even the purest iodine 

 pentoxide became somewhat brown owing possibly to liberation of iodine. 

 The use of electrolytic oxygen in place of air failed to prevent this phenomenon. 

 Finally it was found that if the air was dried more thoroughly, by means of re- 

 sublimed phosphorus pentoxide, the difficulty could be avoided. The cause of 

 this marked catalytic effect from the minute quantity of moisture which is not 

 absorbed by sulphuric acid was not determined. 



The hard-glass tube in which the heating was conducted formed part of a 

 bottling apparatus by means of which the boat could be transferred to a weigh- 

 ing-bottle without exposure to moist air (see fig. i, page 8). Although pre- 

 cautions were taken to prevent exposure of the substance to moisture before 

 weighing, little danger was to be feared from this source, for in air ordinarily 

 moist the pentoxide absorbs water very slowly. For instance 4.4 gm. of pent- 

 oxide, when exposed in the platiniun boat to the air of the laboratory for 2 

 hours, gained only 0.0008 gm. 



Even when dried under the most favorable conditions the above-mentioned 

 darkening of the pentoxide took place if the temperature rose much over 250°. 

 Accordingly the temperature was not allowed to pass this point in drying the 

 substance for analysis. Furthermore, since it was by no means certain that 



