48 IOWA ACADEMY OF SCIENCE 



slow. It was practically insoluble in cold dilute mineral acids 

 and only slightly soluble in hot acids. It was practically insoluble 

 in alcohol, ether, petroleum ether, methyl alcohol, acetic acid, 

 benzene, carbon disulphide and turpentine. In ammonia, pyridine 

 and caustic alkalies it dissolves with ease, giving at first a green- 

 ish blue solution which gradually loses its color. From the 

 alkaline solution it can be recovered by neutralizing with acid. 

 The filtrates from both the furfuralmalonylurea and the furfural- 

 malonylthiourea had a very slight tinge of yellow. Furfural- 

 malonylguanidine, on the other hand, is a very dark green, floc- 

 culent precipitate, appreciably soluble in hydrochloric acid. The 

 filtrate is an intense greenish brown. 



It was early noted that unless the thiobarbituric acid was care- 

 fully purified, the precipitation of furfural was not complete, 

 only 90 or 95 per cent of the latter being recovered, and the fil- 

 trate possessed a red color or sometimes a green color. In one 

 set of determinations the difficulty was traced with reasonable 

 certainty to the presence of cyanacetic ester in the malonic ester 

 from which the thiobarbituric acid was made. In preparing 

 malonic ester from chloracetic acid in the usual way, some 

 cyanacetic ester is apt to remain unless precautions are taken to 

 carry the saponification to completion. This is difficult to sepa- 

 rate from the malonic ester because the boiling points of the two 

 substances lie only a few degrees apart. The cyanacetic ester in 

 all probability reacts with the thiourea, forming a dicyanacetyl- 

 thiourea. On fractional crystallization of one of the impure 

 preparations of thiobarbituric acid, white needle-shaped crystals 

 were obtained, which on analysis yielded 26.66 per cent nitrogen ; 

 calculated for dicyandiacetylthiourea, 26.65 per cent nitrogen. 

 These crystals when dissolved in 12 per cent hydrochloric acid 

 gave an intensely green precipitate with furfural, just as did the 

 thiobarbituric acid before purification. For the preparation of 

 thiobarbituric acid it is, therefore, recommended that the malonic 

 ester be subjected to a repetition of the simultaneous saponifica- 

 tion and esterification before condensation with thiourea, and that 

 the thiobarbituric acid be purified by one or two crystallizations 

 of its sodium salt. 



DISCUSSION. 



Our experiments, quoted above, show that thiobarbituric acid 

 condenses readily with furfural in the presence of 12 per cent 

 hydrochloric acid. The reaction is quantitative, giving a volum- 



