126 Proceedings of the Royal Society of Edinburgh. [Sess. 
In fig. 3 the densimeter tubes A A are seen in section, and the shelf 
upon which the microscope stands is shown within the gimbals. 
During the voyage the tubes were washed out with nitric and chromic 
acids once a week ; and the rubber tubing RRR was frequently changed 
and kept in acidified bleaching-powder solution. The tubing was never 
allowed to remain dry for more than a few hours. The taps were lubricated 
with phosphoric acid or sea-water only. A close observance of these 
several precautions is, as Mr Manley has shown, absolutely necessary for 
ensuring success. 
Before the voyage was commenced, a large jar of sea- water was obtained 
and its density in terms of air-free distilled water at 18° C. determined. 
The density measurements were carried out with the aid of thin-walled 
pyknometers somewhat similar to those devised by Sprengel : these 
determinations were repeated at Colombo. As the means of the first and 
second series were in very close agreement, it was concluded that the 
standard sea- water used throughout the voyage had remained practically 
unchanged. 
In order to measure the density of any sea- water on hoard ship the 
following procedure was adopted : — 100 c.c. of the standard sea- water were 
placed in one flask of the densimeter and an equal volume of the water to 
be examined introduced into the other; the waters were then drawn up 
into the tubes by applying suction at H, and the upper tap t closed. 
Sufficient time was then allowed for the two columns to attain not only 
equilibrium, but also a common temperature. The two taps TT were then 
simultaneously closed, and the difference between the lengths of the two 
columns measured with the microscope : the necessary correction (066 mm.), 
mentioned above, was then applied and the density calculated. Here it 
may be stated that the correcting factor remained, so far as could be 
discovered, constant during the Whole of the voyage. 
When the ship is in motion and the taps TT open, the water always 
oscillates within the tubes, owing to a slight swinging of the instrument ; 
but with a little practice no difficulty is experienced in closing the taps 
during a moment when the tubes are vertical. Towards the end of the 
voyage it was found that the oscillations of the columns could be very 
considerably lessened by removing the bulbs KK : this modification of the 
original apparatus was accordingly introduced, and a distinct advantage 
gained. 
In his paper* Mr Manley has shown that the ratio of the densities of 
two samples of sea-water measured at any common temperature within 
* Proc. Roy. Soc. Edin ., vol. xxvii., Part III., p. 224. 
