Experimental and Applied 259 



progressive differentiations which are therefore, in the limited 

 sense of the term, * organically ' related. The differentiations are of 

 two kinds : (i) the reactive groups (CO, OH, &c.) may vary in 

 number and position ; (2) the substantive group (C n H m ) may vary 

 in configuration (aliphatic, cyclic, condensed, &c., hydrocarbons). 

 To trace the corresponding variations in function or reactivity is a 

 very important part of every investigation ; and the student should 

 diligently study the papers of original workers, and familiarise 

 himself with such general expositions of the science, from this 

 point of view, as he will find in the opening chapters of Beilstein's 

 great work, Handbuch der organischen Chemie ; and more espe- 

 cially, in regard to the carbohydrates, in Tollens' monograph, 

 Die Kohlenhydrate (Breslau, 1888). 



The student should endeavour to form a critical estimate of the 

 position of cellulose in relation to starch and others of the 

 'saccharo-colloids' (Tollens), to which it is most nearly allied. 

 Such an estimate must be based upon the comparative study of 

 reactions. 



Comparing cellulose with starch we find the former resistant 

 to hydrolysis and acetylation, but giving the highly characteristic 

 thiocarbonate reaction. These differences are differences of function 

 or reactivity of OH groups on the one hand, and of the linking 

 together of the unit groups on the other. 



We have then to consider whether these differences are 

 sufficient to constitute the cellulose group a special type of 

 constitution. We think they are. 



Assuming the unit group C 6 H 10 O 6 , and decomposing this, 

 on the evidence before us, into C 6 H 6 O.(OH) 4 , we have but few 

 alternatives. The conclusion we draw can only be expressed 

 within the limits of these alternatives, and therefore in general 

 terms, as follows : (i) The C atoms of the unit groups are com- 

 bined in a closed ring ; (2) the linking of the unit groups is not an 

 oxygen, but a carbon linking ; (3) the synthesis of the unit groups 

 together may be assumed to occur between the CO of one group 

 and the CH 2 of another, giving the alternative form CH C(OH). 



LIGNOCELLULOSES OF CEREALS. The student should 

 read the special papers on the subject, Berl. Ber. 1894, and J. 

 Chem. Soc. 1894. 



The cereal straws in which maybe included the envelopes of the 



