1909.] Presence of Carbon Monoxide in Normal Blood, etc. 523 



rate of evolution of iodine per hour became approximately constant. Different 

 specimens of iodine pentoxide, and even different tubes made from the same 

 sample, were found to differ considerably in this respect. These differences 

 may in part have been clue to traces of hnpority in the iodine pentoxide, 

 i or more probably to different states of physical aggregation. A specimen 

 obtained from a French source kindly recommended to us by Prof. Mcloux, 

 which qualitative tests showed to be pure, after prolonged heating at 

 157° C. evolved iodine equivalent to 1 c.c. of N/1000 sodium thiosulphate in 

 two hours, but we never reduced the amount below this value. We hoped at 

 first to be able to determine for each tube the constant loss of iodine per 

 gramme per hour at the temperature employed in our particular experiments, 

 but experience soon showed that, in addition to the difficulty of keeping all 

 conditions sufficiently constant, when chloroform vapour had been passed 

 through a tube the " constant " underwent change. It was therefore found 

 more satisfactory to make a blank determination of the iodine liberated in a 

 given time by heat alone before each experiment and sometimes also after. 

 The temperatures we employed were 100° C, the boiling point of xylene 

 (137° C), the boiling point of bromobenzene (157° C). 



The chloroform used in our experiments was made from acetone and 

 carefully purified for us by Dr. Wade. It boiled at 61° - 1I to 61° - 15 C. at 

 760 mm. and had a specific gravity T5008 15/15. 



Experiment I. — - 0588 gramme of chloroform, weighed in a thin bulb, was 

 placed in the vessel (3) and the bulb broken. The chloroform vapour was 

 swept over the iodic anhydride in a current of oxygen at the rate of 1 to 

 1J litres per hour. The current was maintained for 81 minutes. Iodine 

 equivalent to 102 - 9 c.c. N/1000 sodium thiosulphate was liberated. In a 

 control experiment which had previously been made under similar con- 

 ditions in 104 minutes, iodine was liberated equivalent to 1*5 c.c. N/1000 

 sodium thiosulphate. Thus each gramme-molecule of chloroform liberated 

 2 - 6 grammes of iodine, i.e. about 1/100 of a gramme-molecule of iodine. 



In the following experiments, mixtures of air and chloroform vapour 

 (amounting to 255 c.c. in each case), the concentration of which was deter- 

 mined by Waller's densimetric method, were passed through the apparatus 

 at different temperatures. A few of the results are given in the following 

 table (Table II). 



A very large number of experiments were made, and it was found that 

 the quantities of iodine liberated varied markedly with different tubes and 

 with alteration of the conditions of the experiment. This was particularly 

 the case in experiments conducted at 100° C. 



