ON ORGANIC COLOURING MATTERS. 489 



Before proceeding to detail the second series of researches, I have thought it 

 desirable to offer some observations on the processes employed for drying gases. 



V. On the Methods applicable to the Drying of Gases. 



The methods at present in use for drying gases cannot be considered as yield- 

 ing more than an approximation to absolute dryness in the case of any elastic 

 fluid. The processes employed are inherently defective, both mechanically and 

 chemically. When chloride of calcium and pumice-stone, steeped in oil of vitriol, 

 are employed as the desiccating agents, they cannot be made use of except in fi-ag- 

 ments of considerable size, otherwise the containing tubes become choked, and 

 the gas does not pass. Interstices, accordingly, comparatively speaking large, 

 occur between the separate fragments of the drying agent ; and the gas, in moving 

 along, has a certain portion of its mass not in physical contact with the hygro- 

 metrics, or directly exposed to their desiccating action. In like manner, when a 

 gas is sent through a column of oil of vitriol, only the surface of each bubble 

 is in contact with the liquid, and the gas-bells rise very rapidly through so dense 

 a fluid, so that they can be dried only imperfectly during their ascent. 



Those defects admit only of partial remedy, by extending the surface of chlo- 

 ride of calcium or pumice-stone, or by multiplying the columns of oil of vitriol 

 through which the gas shall pass. A practical limit is set to such devices by the 

 obstruction which they offer to the passage of elastic fluids. This can be over- 

 come only by increasing the pressure at which the gas is delivered, and it is not 

 easy to regulate this, so that the gas shall not flow in too swift a current, and so 

 neutralise, by the rapidity of its passage, the benefit which would otherwise result 

 from its coming in contact with an extended hygrometric surface. 



The imperfections just alluded to are not, perhaps, beyond the reach of suit- 

 able mechanical contrivances ; but even if they were aU remedied, the important 

 question still remains, will the protracted and complete contact of a gas contain- 

 ing water-vapour with the most powerful hygrometrics, suflice entirely to de- 

 prive the gas of moisture .'' With a view to determine this point, I shut up mu- 

 riatic acid gas, previously passed through a freezing mixture, and over chloride 

 calcium, within a glass tube containing fragments of the same salt. It was left 

 for a week in contact with the chloride, and then allowed to meet carefully dried 

 blue litmus-paper, enclosed along with it at the commencement of the experi- 

 ment, in a small sealed bulb of thin glass, which was readily broken by shaking 

 the tube. The paper began to change tint as soon as it met the gas ; and if this 

 alteration in colour be accepted as an evidence of moisture being present in the 

 muriatic acid, then the latter was not dry. A similar experiment, with a like 

 result, has already been related in reference to ammonia. 



These results, however, are not decisive of the point whether the gas was an- 

 hydrous or not, for the change in tint of the litmus may possibly be accounted 



