I50 THE CARBOHYDRATES 



produced. The test is a very delicate one and will detect 

 quantities of formaldehyde varying from i part in 1,000,000 

 to I part in 100,000. Acetic aldehyde gives no colour with 

 this reagent. 



The following test, due to Deniges,* is sensitive for formal- 

 dehyde, even in presence of acetic aldehyde up to 2 per cent ; 

 5 c.c. of an aqueous solution of formaldehyde are mixed with 

 1-2 c.c. of pure sulphuric acid (sp. gr. i '66) and 5 c.c. of Schiff's 

 reagent. An intense violet colour having an absorption band 

 in the orange is produced. Schiff s reagent may be prepared 

 by adding a litre of o*oi per cent of solution of magenta to 

 20 c.c. of sodium hydrogen sulphite solution (sp. gr. 1-3), and 

 after five minutes adding 20 c.c. of hydrochloric acid (sp. gr. 

 i-i8). 



Kimpflinf tested for formaldehyde in the leaf of Agave 

 niexicana by injecting into it, by means of a capillary tube, a 

 concentrated solution of sodium hydrogen sulphite, contain- 

 ing an excess of /methylamino-;«cresol. The presence of 

 formaldehyde was indicated by the formation of a red pre- 

 cipitate on exposure to light. The precipitate is best seen 

 by examining a section of the leaf which has been dipped in 

 absolute alcohol. Formaldehyde is the only aldehyde giving 

 a stable red colour with the above reagent, but other aldehydes 

 give unstable green, yellow, or reddish-brown colours. 



In 1906, Usher and Priestly J stated that they obtained 

 formaldehyde by subjecting green leaves to steam distillation ; 

 the presence of formaldehyde in the distillate was proved by 

 evaporating it down with ammonia and adding bromine water 

 to the residue to convert the hexamethylene tetramine so 

 formed into its tetrabromo derivative. 



In the opinion of Curtius and Franzen, § the presence of 

 formaldehyde in plants has not been established with certainty 

 by previous workers, inasmuch as the majority of the above 

 tests for the substance in question are inconclusive, since they 

 are also given by other aldehydes. By working with 1500 

 kilos of hornbeam leaves these authors conclusively proved the 



* Deniges: " Compt. rend.," 1910, 150, 529. 



t Kimpflin: " Compt. rend.," 1907, 144, 148. 



+ Usher and Priestly: " Proc. Roy. Soc, Lond.," B., 1906, 77, 369. 



§ Curtius and Franzen : " Ber, deut, cliem. Gesells.," 1912, 45, 1715. 



