22 



could hardly be expected, even in the presence of sun- 

 light. On the other hand in the experiments of Herrera 

 it is possible for the formaldehyde CHoO to polymerize 

 into formose CoHioOe, which differs from starch CeHioOg 

 only by an extra molecule of water. 



If proteins were actually produced in the experiments 

 of Francis, then the carbon of such proteins must have 

 come from the carbon dioxide by partial reduction to 

 formaldehyde at the surface of the iron or ferrous hy- 

 droxide. The amino nitrogen could have come either 

 from the ammonium ions or from the nitrate ions. 

 Francis may have obtained the spiral polypeptide struc- 

 ture, but it is doubtful whether there was any reduction 

 of the sulphate to sulphides, with the formation of cys- 

 tine or cysteine. 



In the experiments of Herrera it is possible that sub- 

 stances similar to proteins may have been produced, 

 even though they were not detected by chemical tests. 

 The polypeptide spiral as pictured hereinabove may be 

 only one of several different structures that can be 

 formed by spiral polymerization. There are other 

 cyclic structures besides the diketopiperazine ring which 

 have diametrically opposite acidic groups and also dia- 

 metrically opposite basic groups so as to be capable of 

 polymerization by superposition of the rings upon one 

 another. 



It has been found by Schmerda (Z. Angew. Chem., 30, 

 176, 1917) that ammonium thiocyanate and formaldehyde 

 in aqueous solution will combine, without any liberation 

 of carbon dioxide, to form a soft yellow resin having no 

 definite melting point, which is insoluble in all ordinary 

 solvents, and which readily splits off formaldehyde when 

 heated. Schmerda has shown that this resin contains 

 twice as many nitrogen atoms as sulphur atoms, but ap- 

 parently he did not make any carbon and hydrogen de- 



