221 
1906-7.] On the Application of a Differential Densimeter. 
marked oiliness developed inside the flask almost immediately. Hence the 
rubber tubing was obviously the chief cause of the abnormal condition of 
the instrument. 
The rubber tube connections were now removed and treated with 
moist chlorine gas, then boiled with potash solution to which a little 
hydrogen peroxide had been added, and finally soaked for some hours in 
water, dried without heating, and replaced in their former positions. The 
densimeter then behaved quite normally for a considerable period, and then 
became a little unsatisfactory ; but after re-treating the rubber tubes no 
further trouble was experienced. 
It was found that the rubber tubes might be freed from the volatile 
matter by boiling them with potash and hydrogen peroxide alone, or even 
by soaking them for some weeks in distilled water. Experiments were 
also made with grey rubber pressure tubing; these showed that such 
tubing might be safely used without subjecting it to treatment. 
These observations are also of importance in connection with various 
forms of apparatus used for gas analysis. 
Portable Form of Densimeter. 
The densimeter already described is well adapted for laboratory use, 
but for work on board ship certain modifications are desirable. 
The somewhat rough usage to which the instrument would be exposed 
at sea might lead to injury. Again, the arrangement of the cisterns leaves 
something to be desired in the matter of convenience, though nothing in 
point of accuracy. Chiefly for these reasons a new apparatus, having the 
form indicated in fig. 2, was constructed. 
In this instrument the long glass tubes are not fused to the headpiece, 
but are attached by means of grey pressure tubing. The portions of the 
tubes above the taps t, p, have an internal diameter of 5 mm., and their 
ends are bent a little outwards, so as to enable an observer to scrub the 
interior of the wider portions of the tubes with a small brush. Below t, p, 
the tubes are quite narrow, the internal diameter being about 2 mm. only. 
The cisterns have the same size as those used in the other form of the 
apparatus, but side tubes are sealed into them at the lowest possible point, 
and these are joined to the vertical tubes by purified black rubber tubing. 
When in situ, the cisterns rest side by side upon one and the same levelled 
plate-glass shelf S. To prevent evaporation of the water, the mouths of the 
flasks are closed by loosely fitting glass stoppers with flat tops. 
This apparatus is both portable and convenient ; it possesses the 
