;380 



Alfred J. E'ivart : 



a comprehensive view of the different agencies by which such 

 synthesis could be brought about, and the relationship between 

 them. In the first instance the action was tested of alkaline 

 metallic hydroxides and carbonates, and then of non-metallic 

 alkalies. 



Alkaline nitlalllc hydror/Ults. 



Caustic Soda. If stroing caustic soda is run into loiling 

 40 % formaldehyde until the 'liquid ceases to smell of the latter, 

 sodium formate is produced, but no reducing sugar. If ^0 c.c. of 

 40 % CH2O and 14 c.c. of '33 % NaHO are boiled to dryness in a 

 flask with a narrow outlet, the residue consists of nearly pure 

 .sodium formate, only a trace of matter is removed by alcohol, and 

 this contains no reducing sugar. 



If, however, while the evaporation is going on, hot water is added 

 several times, the residue contains a small trace of reducing sugar 

 removed by washing with 96 to 98 % alcohol. 



If 30 c.c. of 40 % CHoO are added to 500 c.c. of water and 

 40 c.c. of 10 % NaHO, and the bulk (vf the liquid dripped slowly 

 into a boiling portion, the li(|uid continues to smell of CH^O until 

 nearly dry, and the pale yellow residue contains an appreciable 

 amount of reducing sugar but consists mainly of sodium formate. 



If more soda is added during boiling all the CH^O disappears, 

 and the amount of reducing sugar is increased, but any excess of 

 soda turns the liquid dark brown. 



Further experiments showed that with increasing dilution the 

 proportion of sugar to sodium formate increased, but that the 

 latter was always formed in excess. 



Linhf-irafer. — 100 c.c. of Lime-water were added to 25 c.c. of 40 % 

 formaldehyde, and while boiling in a flask fitted with a condensing 

 arrangement, lime-water was run in slowly until all the CH^O was 

 used up. Approximately, 1 litr-e of lime-water was retjuired. Tbe 

 pale yellow syrup yielded calcium formate when evaporated, methyl 

 alcohol escaped, and the gunmiy residue contained a high 

 proportion of reducing sugar, largely pentose and giving the 

 phloroglucin reaction readily. The amount of sugar was consider- 

 ably less than with the joint action of sodium liydrate and calcium 

 hydrate as previously described. 



Barivm and Strontium hydratesi. — With Barium hvdrate, to 

 complete the reaction an excess of the alkali is required and pro- 

 longed boiling. Strontium hydrate is more active, only a slight 

 excess is necessary, and moderately prolonged boiling (2-4 hours 



