Sijn thesis of Sugar ISS- 



of reducing sugar, but tlie reaction is imperfect, and is difficult 

 to comi^lete without overheating portion of the mixture, or volati- 

 lising much of the formaldehyde. 



It is the presence of water of crystallisation in the crystalline- 

 barium hydrate which enables its reaction with solid polymers of 

 formaldehyde to progress, and be completed throughout the whole 

 mixture. Some of this water of crystallisation is liberated by mere- 

 contact with solid formaldehyde. 



Summary of Results. 



Cold dilute solutions of formaldehyde yield with alkalies, format.<^s- 

 and methyl alcohol, sugar polymerisation being inappreciable. 

 The reaction is incomplete after months in contact. 



Maximum sugar production is given with boiling liquids, and 

 the reaction is completed almost instantaneously. 



The maximum sugar polymerisation varies from 68% (Ca), 60%. 

 Ba, 65% 8r. to 58% Mg. for divalent alkalies, and from 14% (K) 

 to 8% (Na) for monovalent alkalies. The high polymerisation value- 

 (68%) estimated from the amount of alkali required for neutralisa- 

 tion indicates either that tlie pentoses and hexoses are produced 

 directly, or that glycollic aldehyde is produced by the reaction 

 yielding formates, an^l 3 or 4 of formaldehyde added to it without 

 further decomposition of formaldeliyde. Tliis is also indicated by 

 the appearance of pentoses. Polymerisation of biose Avould yield 

 tetroses or hexoses, but not pentoses. 



The greatly increased polymerisation when soda is run into a 

 solution of formaldehyde containing calcium' formate is merely due 

 to the fact that calcium hydrate becomes the polymerising agent,, 

 and as a divalent metal exercises a greater polymerising action. 

 There is no evidence of any katalytic action, as was formerly sup- 

 posed. 



In terms of glucose, the reducing sugars produced represented a 

 polymerisation of 29% for calcium, and 2% for sodium. Appar- 

 ently the sugars have half the reducing power of glucose. 



The solid polymers of formaldehyde yield sugar readily, and 

 abundantly, Avhen in contact with solid alkalies. The most com- 

 plete reaction its shown with the powdered crystalline hydrates of 

 Ba and Sr. A feeble endothermic reaction precedes the violent 

 exothermic one. The production of sugar is greater than with solu- 

 tions, the polymerisation equivalent in terms of glucose being 

 30-32%. 



In all cases the sugar is a by-prodiict in a reaction, yielding for- 



