984 
are however distributed arbitrarily over these two “isomerides” so 
that this distinction is quite worthless. 
A serious investigation as to the eventual identity of all these kinds 
of glutaconic acid has never been taken in hand; the investigators 
have always been content to pronounce this identity on account of the 
melting point of the different acids *) which was always fairly well 
the same. Still a similar investigation would be of importance; 
suppose the identity of all the acids was established, it would then 
be evident that there exists one form of glutaconic acid which 
exceeds over all other possible isomerides by a very great stability, 
for this variety was formed im all kinds of conditions. A. similar 
great difference in stability does not occur as far as I know with 
a series of cis-transisomerides. Hence, I have endeavoured in the 
first place to furnish the proof for this identity. 
It should be mentioned here that PerKiN and Tarrprsaut *) have 
tried in various ways to obtain a second isomerides of glutaconie 
acid, but always without success. They always obtained either the 
acid already known or the trans-trimethylenedicarboaylic acid. 
Il. The question whether to the known glutaconic acid apper- 
tains the cis- or the trans-formula has been decided by Bucnner *) in 
favour of the cis-formula and this on account of the ready formation 
of anhydride on boiling with acetyl chloride. This way of arguing 
is not satisfactory; it is generally known that fumaric acid on treat- 
ment with acetyl chloride *), phosphoruspentoxide °) or pentachloride °) 
is converted into maleinic anhydride. Moreover, the formation of 
anhydride proceeds anything but rapid. 
If to the glutaconie acid appertains the symmetric structure sug- 
gested by THorPE it. may be expected that this acid will possess 
neither all the characteristic properties of the cis- nor all those of 
a trans-acid, but will behave either as a cis or trans in reactions or 
definite properties, which can give exclusive evidence as to cis- or 
trans-positions of the carboxyl groups. 
I have been able to demonstrate the correctness of this view in 
the case of a few reactions and properties. 
1) Compare BucHner: Ber. 27. 881 (1894). 
2) Soc. 87. 361 (1905). 
3) loc. cit. 
4) AnscHutz: Ber. 14. 2792 (1881). Perkin: Ber. 15. 1073 (1882). 
5) TANATAR: Journ. Russ. phys. chem. Ges. 22. 312. 
6) VoLHARD: Ann 268. 255 (1892), 278. 31 (1893). 
