226 Proceedings of the Koyal Society of Edinburgh. [Sess. 
position by means of tara and sliding weight. The balance is then 
arrested, the water is removed by means of a siphon, H, and a new sample 
suspended in water is poured into the vessel as quickly as possible. The 
balance is now released, and the chronograph started, after which the 
observer has simply to record the values of the counterweights added, as 
well as the moments when the Osram lamp lights up. 
Fig. 1. 
In order to avoid convection currents due to variations of the 
temperature, all the experiments were carried out in the room for constant 
temperature in the Chemical Institute at Uppsala, the aqueous suspensions 
being always allowed to assume the prevailing temperature of the room. 
The cylindrical vessel was kept within a non-conducting envelope, M, and 
covered with a lid which had a small aperture for the suspension-wire 
from the balance to the disc. Fig. 2 gives a photograph of the 
arrangement. 
From observations made with the apparatus just described we can plot 
