130 
DR. AVALTER AVAHL: PHYSICO-CHEMICAL DETERMINATION’S 
may occur in any part of the apparatus between the manometer and the investigation 
bomb, and, especially if the entire contents of this is solidified, the part of the 
substance observed may remain under a very different pressure from that indicated 
by the manometers. Evident enough as such a cause of error may appear, it has 
been one of the principal practical difficulties to get over in all high-pressure work, 
and is one of the chief causes of errors in earlier work of this kind. Tamma]s^n 
refers in several instances to this circumstance as a cause of erroneous experimental 
results, and he has had to renew several of his earlier series of measurements. He 
has, in consequence, paid much attention to the problem of bringing the substance 
into the pressure bomb under such conditions that it is really submitted to homo¬ 
geneous pressure during the entire investigation. In his most recent work Tammann 
has used a kind of bag or cylinder made out of parchment paper and coated with 
a varnish of collodium.* This cylindrical bag, when containing the substance and 
closed at both ends, is surrounded on all sides in the pressure bomb by some 
pressure-transmitting fluid, and as it is entirely soft the hydraulic pressure is evenly 
transmitted. The material for such a pressure-transmitting cover has of course to 
be chosen with regard to the nature both of the substance investigated and of the 
surrounding fluid, and in many instances it is not possible to And a material which is 
entirely insolulde, and which is not chemically acted upon by the substance to be 
investigated, especially when a large amount of substance is used, and the 
investigation is carried on at comparatively high pressures and temperatures. 
In the case of an optical investigation it is, of course, essential to make observa¬ 
tions upon a sufficiently thin layer of substance, so that it is possible to see the 
changes in the optical properties due to polymorphic transition. At first a small 
glass tube of about 2 mm. diameter was used as a vessel to hold the substance to be 
investigated. Both ends of the glass tube were bent over, approaching each other, 
and connected one to another by a small piece of rubber tube. The whole was filled 
with the substance, and the rubber tube part acted as pressure transmitter. Some 
melting-point determinations on dimethylether of oxalic acid were made in this way, 
but it was not easy to observe the optical characters of the crystallized body with 
this device, and subsequently the following arrangement has been used in most cases. 
The substance is melted between two circular glass plates, about f mm. thick and of 
10 mm. diameter, care being taken not to superheat the substance. When the 
substance has crystallized, a piece of black rubber tube, 5 mm. long and about 
5 mm. wide, is slipped over the joint edges of the two glass plates, thus forming a 
surrounding cap, which at the same time gently presses the two glasses against each 
other, the whole forming a slightly elastic extremely narrow vessel. The rubber 
piece used is previously kept for some time in the molten substance to be investigated, 
and as the substance during the experiments comes into contact with the rubber only 
* G. Tammann, ‘Zeitschr. f. Anorg. Chem.,’ 63, p. 285 (1909), and ‘ Zeitschr. f. Rhys. Chem.,’ 75, 
p. 77 (1910). 
