82 J. Bhaduri — Transformation of [No. 2, 



of experiments useless. As the solutioa of hypochlorite invariably 

 contained free alkali, chloride and traces of chlorate, the amount of 

 hypochlorite could not give results which might be compared. In order 

 that the experiments should be strictly comparable the same sample 

 of hypochlorite was diluted to requisite strengths, and then the tubes 

 were partially filled. As in most cases preliminary experiments were 

 made to know approximately the volume of oxygen disengaged from 

 solutions of similar strength, the volumes of the tubes were so regulated 

 that the internal pressures were very nearly the same in all. Duplicate 

 experiments were in almost all cases made. The operation of filling 

 was done as quickly as possible in the dark room, and the already narrow 

 end of the tube drawn to a capillary bore in the blowpipe flame. When 

 perfectly cold the end was sealed. In no case, however, did the time of 

 complete operation exceed ten minutes. When the required number 

 of tubes was ready they were at once introduced into the steam chamber 

 heated to 100° C. After the required number of hours they were re- 

 moved to the dark room and allowed to cool. The time required to 

 heat the tubes to lOO'' C was thus very nearly the same as the time, to 

 cool down to the original temperature. Caustic soda dissolves oxygen 

 to an appreciable extent, and the solubility, as with all gases, increases 

 with pressure. The gas thus dissolved escapes slowly when the pressure 

 is relieved, and as sufficient time could not be allowed for fear of decom- 

 position, a small quantity was generally lost. The loss had no effect 

 when the total volume of oxygen was considerable. Hence the volume 

 of oxygen and therefore its weight should be taken as approximate, and 

 not strictly accurate when the quantity was small. The ratio of the 

 chloratic oxygen to free oxygen should also be taken in the same light 

 under similar circumstances. The preceding table contains the results 

 of numerous analyses performed at different times with solutions of 

 different strength heated to different periods of time. 



Comparing the percentage decompositions of a number of solutions 

 in which relative quantities of free alkali, chloride, chlorate and hypo- 

 chlorite are the same, but with gradually diminished absolute quantities, 

 the following peculiarities are observed. The decomposition diminishes 

 with dilution up to a certain point when it is minimum. Further dimi- 

 nution in concentration instead of diminishing increases the decomposi- 

 tion. This peculiar deportment of the hypochlorite solution is seen 

 not only with similar solutions heated for different periods of time, but 

 also in all solutions which have been examined some of which only are 

 given. The ratios of hypochlorite to free alkali in the three solutions 

 mentioned in the table are approximately in (A) 2 : 5, in (B) 2 : 2, 

 and in (C)2 : I. The relation between concentration and percentage de- 

 composition is very strikingly shown in Plate IV. 



