(914) 
above the critical temperature of carbonic acid, and also the second 
condition mentioned by var per Waats, is fulfilled. 
Our carbonic acid was prepared from commercial carbonic acid 
by a process of drying by means of P,0,, of condensing by liquid 
air, and of removing the volatile components (air) by blowing off at 
low temperature. For the urethane we used commercial urethane ; 
the quantity of it being so small compared with that of the carbonic 
acid, it seemed superfluous to purify it further. 
As it is of the highest importance in these experiments, as will 
appear, to regulate the concentration of the mixture very accurately, 
we must give a description of the way in which our Cailletet tubes 
were filled. A quantity of urethane shaped into a rod by melted 
urethane being sucked up in a capillary tube, was weighed off (once 
we used about 15 mg., the other times about 5 mg.). The rod was 
put in a Cailletet tube and spread over the wall by being melted ; 
after a stirrer had been applied the tube was placed in a pressure 
cylinder as usually; then the mercury was pumped up in the tube so as 
fill it half, and then the cock whieh led to the pump, was closed 
so that the mercury got a constant position. The Cailletet tube passed 
into a narrow capillary at the upper end, to which a steel capillary 
was fastened by means of sealing-wax, which led to a steel high pres- 
sure cock of a model specially prepared with a view to quantitative 
conveyance of measured gas quantities for the Amsterdam Laboratory 
by the firm ScHArrer and bupenperc. The two requirements (apart 
from the ordinary requirements which are in connection with perfect 
closure) which such a cock must fulfil, are: 1. Absence of all 
recesses or places where gas might be left behind. 2. Absence of 
all that might contaminate the passing gas. The model represented 
in fig. 1 meets both requirements, as all packing material has been 
avoided for the joints of the leading tubes A and B. A steel cone 
C, which is soldered *) to the leading capillary tube is tightly pressed 
against the conical wall of the cock by the nut D, and thus forms 
a perfectly tight closure of steel on steel. As the borings of the 
cock are made perpendicular to each other, in which care is taken that 
when the horizontal boring is being made, the borer does not pene- 
trate into the wall of the vertical one, a cock is obtained, in 
which the gas finds nowhere an opportunity of entering a recess 
5 To prevent the solder from being attacked by the mercury a raised rim is 
left when the steel conus C is pierced (see fig.), against which the capillary is 
tightly pressed. So the solder which is poured between the capillary and the cone 
cannot come into contact with the mercury. 
