( 467 ) 
a tube connected with a mercury barometer and also with a mer- 
cury airpump, which caused a fairly rapid decomposition ; the pressure 
rose beyond 1 atmosphere. Even at 100°, decomposition takes place 
if we only take care to continually evacuate; the mass first becomes 
partially liquid owing to the valerie acid formed and now we can 
plainly see the evolution of carbon dioxide from the solid particles 
of methylethylmalonie acid still suspended in the liquid. We cannot, 
therefore, speak of a definite decomposition point of acids with two 
carboxyl groups attached to one carbon atom. The statement that 
these substances lose carbon dioxide by heating above their melting 
point is consequently not only very vague but to some extent also 
incorrect as methylaethylmalonie acid already loses CO, when still 
in the solid condition. 
Whilst, however, it takes days before the methylethylmalonic acid 
is decomposed at 100° at the ordinary pressure, this process is 
finished in a few hours if we continually evacuate. This would he 
most readily explained by assuming that this decomposition is a 
dissociation phenomenon. At each temperature, there would then 
exist a definite dissociation tension and if now by a continual 
evacuation care is taken that one of the decomposition products is 
always being removed, it is obvious that finally all must be decom- 
posed. The only difference between this phenomenon and the classical 
example of Ca CQO, is this that one has never succeeded in obtaining 
an acid with two carboxyl groups by heating an acid, containing 
one carboxyl group, in carbon dioxide. This may be explained either 
by assuming false equilibria, or by supposing that the velocity of 
reunion of the decomposition products is exceedingly small. I intend 
to further investigate this point. 
As it had now been proved that the temperature of decomposition 
of acids with two carboxylgroups to one carbon atom could be greatly 
decreased by diminution of pressure, it was obvious that the synthesis 
of |-valerie acid might also be improved by allowing the CO,-elimination 
to take place in vacuo at least if Marekwarp’s idea was correct 
that the bad yield of active material was due to atomic shiftings'). 
I have now heated the acid brucine salt of methylethylmalonic 
acid with continual evacuation at 120°, therefore far below its melting 
1) It is easy to understand that a decrease of the temperature at which carbon 
dioxide is expelled is in itself not capable of improving the synthesis. The velocity 
of the atomic shiftings would no doubt have much diminished but then also the 
velocity of the carbon dioxide elimination, and the complete decomposition of the 
substance would take a much longer time. The evacuation, therefore, merely serves 
to accelerate the decomposition process. 
