On the Steam Calorimeter. 



231 



effect of the jacket, increasing radiation. With everything else as in 

 (1) \_ante~\, observations every 5 minutes gave : — 



3-4 4-0 3-6 3-0 3-8 3-2. 

 Mean, 3*5 milligrams per 5 minutes. 



(7.) The lamp-black was now removed and the inside cleaned. 

 With the same order of procedure as before exactly, and everything 

 the same except for the absence of the lamp-black, the results 

 are : — 



07 07 0-8 1-1 1-1 0-8 1-0, 

 or nearly 0"9 milligram per 5 minutes. 



The effect of the water deposited between the walls of the jacket is 

 probably seen in the difference between this experiment and the results 

 of (1). The difference existing between the results of (6) and (7) is 

 conspicuous and so far as I can see can only be explained in the hypo- 

 thesis that the increase is due to radiation. The inference is strong 

 that the ordinary effect is also in some degree, if not entirely, of the 

 same nature. 



When in special cases it is thought necessary to allow for this effect 

 of radiation (as I will call it), I would suggest making an observation 

 on the amount of the increment, say over 10 minutes, subsequent to 

 the weighing being completed, and deducting proportionately to the 

 time occupied in weighing — not proportionately to the time since the 

 first admission of steam into the calorimeter, as I do not think it is 

 warranted to assume that the same effect obtains during the period in 

 which the substance is rising in temperature. It is to be remembered, 

 in fact, that (as before observed) an effect of the opposite sign as 

 affecting the observed condensation on the body may then have 

 obtained. It is to be observed that some of the increment will be 

 due to the carrier. This would have amounted to rather more than 

 one half in such a case as (2), considerably more in (1). The ordinary 

 deduction from the total condensation for the calorific capacity of 

 the carrier will eliminate approximately this portion of the radiation 

 increment. 



But it may be asked: how is it to be determined at what precise 

 momea, the true precipitation has ceased ? when has the substance 

 attained the temperature of the steam ? 



In considering this question it is necessary to realise the nature of 

 the phenomena observed in the course of an experiment. Suppose 

 the 37 grams of cryolite, previously referred to, was being dealt with. 

 Steam is admitted. For three minutes, about, we try in vain to 

 equilibrate the balance. Equilibrium is impossible for the reason 

 that condensation is progressing so rapidly upon the cryolite that so 

 soon as an approximation to equilibrium is obtained this is again 



