370 Mr Fenton, Oxidation in Presence of Iron. 
repeated in a similar way several times, allowing the mixture to 
stand for some hours between each saturation. The resulting 
product now gives with phenylhydrazine a bright orange-yellow 
precipitate, which when recrystallized from hot chloroform melts 
at 222 — 224° and is shewn to be identical with Nastvogel’s 
osazone CHN^HPh . CN 2 HPh . COOH which has been mentioned 
above. [The investigation of this product was in the first instance 
considerably retarded owing to the fact that all previous observers 
have given a value to the melting-point of this osazone which is 
far too low. A complete re-examination of the osazone has now 
proved that the true melting-point is as here stated, 222 — 224°.] 
By concentrating the liquid under reduced pressure and allow- 
ing the unaltered tartaric acid to crystallize out as far as possible, 
a syrupy liquid is obtained, which however still retains some 
tartaric acid, and which gives a very abundant yield of the above 
osazone. The oxidation-product is shewn to contain no chlorine, 
since it may be converted into a barium salt which is free from 
chlorine, and which when decomposed with dilute sulphuric acid 
gives the same osazone. Under these circumstances it is evident, 
from the established constitution of the osazone (Nastvogel, 
Annalen, 1888. 85 ; Will, Ber. 1891. 400, etc.), that the product 
must be either hydroxypyruvic acid (or dihydroxyacrylic acid), 
tartronic semi-aldehyde or mesoxalic semi-aldehyde. 
It gives no violet coloration with ferric chloride, so that it is 
not identical with the glyceric acid oxidation -product above 
referred to. 
In order further to decide the nature of the product the 
following experiments were tried, and the results of these, apart 
from other considerations, definitely prove its constitution. 
Action of Hy dr oxylamine. 
The tartaric acid-chlorine oxidation-product obtained as above 
described after removal of most of the unaltered tartaric acid was 
mixed with excess of hydroxylamine hydrochloride in aqueous 
solution and neutralized with sodium carbonate. After standing 
over night it was acidified with dilute sulphuric acid and extracted 
several times with ether. Most of the ether was distilled off and 
the remaining liquid allowed to remain in a vacuum desiccator. 
The resulting solid reddish mass was dissolved in the least 
possible ammonia solution and acidified with strong hydrochloric 
acid, the operations being carried out with careful cooling. Long 
colourless needles separated, which were purified by again dis- 
solving in ammonia and acidifying with dilute sulphuric acid. 
These melted at 178 — 180° and were found by analysis to have 
the formula C 3 H 4 N 2 0 4 . They are somewhat sparingly soluble in 
cold water, and the aqueous solution gives with ferrous sulphate 
