202 Richards — Transition Temperature of Sodic Sulphate. 



points of a number of organic substances* showed that these 

 bodies are too much subject to contamination with clinging 

 impurities to serve as accurate standards. Organic compounds 

 are too plentiful in number to make the easy obtaining of any 

 one alone an easy matter, and among inorganic substances no 

 single suitable substance beside water seems to exist. 



There is no reason, however, why we should be confined to 

 the use of a single component in this search for fixed points. 

 Two components, requiring four fixed conditions, should answer 

 equally well. The only essential is that the substance involved 

 should form perfectly definite phases, and should be capable of 

 being obtained in the pure state. Nernst has suggested the use 

 of " cryohydric" points as a means of maintaining constant low 

 temperatures, but the possibility of utilizing higher nonvariant 

 points involving two components as a basis of thermometry and 

 a means of maintaining constant temperature does not seem to 

 have been generally realized. 



Of the many pairs of substances which might serve the pur- 

 pose in view, the pair, sodic sulphate and water, seems to be 

 the most suitable for several reasons. 



In the first place, the system (Na 2 S0 4 . + Na,SO 4 .10H,O -f 

 saturated solution + vapor) is in equilibrium at 32*5°f — a most 

 convenient point, less above ordinary temperatures than 0° is 

 below them, and within the field of even very large-bulbed 

 thermometers. This small elevation involves very little dis- 

 turbance in the tension of the glass, as well as very slight cor- 

 rection on account of the column projecting into the cooler 

 atmosphere of the room. Moreover 32*5° is near the tempera- 

 ture of greatest difference between the hydrogen and the mer- 

 cury thermometer. 



On the other hand, sodic sulphate does not " melt " so easily as 

 to cause any difficulty in keeping it unfused in a reasonably 

 warm place. A substance melting only a few degrees lower 

 would be- continually freezing into a solid lump, which could 

 be extracted from its bottle only by melting. The fact that it 

 is not deliquescent is also of value. 



A much more important advantage, as far as exactness is 

 concerned, is to be found in the great ease with which the sub- 

 stance may be obtained in a pure state. Its solubility is far 

 less at 10° than it is at 33°, hence its recrystallization by 

 cooling does not involve much loss. Moreover, it may also be 

 crystallized in the anhydrous state by simply melting the 

 aqueous crystals, and since this process necessarily brings into 

 play a wholly new set of isomorphous relations, it may be relied 

 upon to eliminate impurities which are not rejected by the 

 hyd rated crystals. 



*Zeitschr. phys. Chem , iv, 349 (1S89). 



f Lowenherz found 32-39° (Zeitschr. phys. Chem., xviii, 70). 



