82 



SCIENCE 



[Vol. LVI, Xo. 1438 



trees. Since, according to the values obtained in 

 tliis series, tlie woods fall into two groups of 

 fairly uniform composition, depending upon 

 whether they are from broaS-leafed trees or froni 

 conifers, the suggestions are made that a more 

 detailed study of the splitting products, particular- 

 ly the cellulose and lignin, of the woods of these 

 two groups be made. 



Some observations on the determination of ecVu- 

 lose in woods. II: S. A. Mahood. In a previous 

 communication the author showed that the yields 

 of cellulose from woods obtained by Schorger 's 

 modification of the Cross and Bevan method of 

 cellulose determination are higher than those ob- 

 tained by Seiber and Walter's modification. It 

 was suggested that this difference might be due 

 ( 1 ) to a difference in concentration of the 

 chlorine in contact with the sample or (2) to a 

 difference in temperature at which the ehlorina- 

 tions are made. The present paper covers an 

 investigation of these two points. It has been 

 found that the yield of cellulose obtained by the 

 Seiber and Walter method of ehlorination can be 

 made to accord with that obtained by the Schorger 

 method by diluting the chlorine used with an 

 equal volume of air. There appears to be no 

 difference in the percentage of cellulose obtained 

 by the Seiber and Walter method when ehlorina- 

 tion takes place at zero and at room temperature. 

 The Willard crucible has been found to be more 

 suitable tlian the Gooch crucible with the fiber pad, 

 as proposed by Seiber and ^Walter, for making 

 chlorinations of cellulosie materials. The results 

 sh.ow that the lower yield of cellulose obtained 

 by the Seiber and Walter method of ehlorination 

 is due only to too concentrated a stream of chlo- 

 rine and not to too high a ehlorination tempera- 

 ture. 



Preparation of a "standard cellulose" : W. 0. 



MlTSCHERLlNG. 



Synthesis of derivatives relating to polysae- 

 charides: Harold Hibbert and Harold S. Hill. 

 Brojuaeetaldehyde has been condensed with 

 mannitol, glycerin, glycerin brom-hydrin and vari- 

 ous glycols. The bromine atom in the resulting 

 eylic derivatives may be replaced by hydroxyl by 

 boiling with dilute alkali. In the case of glycerin 

 a bicylic derivative is apparently formed at the 

 same time. The condensation products are being 

 subjected to a careful examination in view of their 

 relationship to the anhydro-sugars and polysac- 

 charides. 



Synthesis and properties of eylic acetal deriva- 

 tives: Harold Hibbert and John A. Timm. Par- 



tition experiments carried out on the condensation 

 of a mixture of one mol. each of a 1:2 and a 1 : 3 

 glycol with one mol. of an aldehyde show that in 

 the various cases investigated a much higher yield 

 is obtained of the six- than of the five-niembered 

 ring compound. Also, the stability of the latter 

 towards dilute acids is much less than that of the 

 former.- Similar experiments are being carried 

 out using a mixture of one mol. each of two alde- 

 Tiydes with one mol. of a glycol. The nature of 

 the products resulting from the condensation of 

 various aldehydes (1 mol) with glj'cerol (1 mol.) 

 is also under investigation as well as the part 

 played by the catalyst in bringing about such 

 condensations. 



Constitution of sedoheptose anhydride and its 

 relation to cellulose: Harold Hibbert. The con- 

 stitution of this anhydro-sugar first prepared by 

 Hudson and LaForge is probably represented by 

 a 5:7 dilactouyl-coufiguration for the follow- 

 ing reasons: It is stable towards mineral 

 acids so that the absence of an ethylene linking 

 may be assumed. The product is a mono-molecu- 

 lar, stable, crystalline derivative showing no ten- 

 dency towards polymerization, in this respect, 

 differing from the intra-molecular condensation 

 product of a yA di-hydroxy aldehydo-derivative, 

 .the outstanding feature of which, as found by the 

 writer, is its rapid ease of conversion into a poly- 

 morphic form. The evidence for this view could 

 be found in tlie oxidation of sedoheptose anhy- 

 dride to the acid, from which, by elimina- 

 tion of carbon dioxide, levoglucosan should 

 be formed. Experiments with this object in view 

 are in progress. The properties of y\ dihydroxy- 

 aldehyde derivatives are being examined, the first 

 under review, viz., 2 : keto--5 ; 6 dihydroxyhexane, 

 being found to yield a highly polymerized product 

 ^vhen submitted to the intra-molecular condensing 

 action of a small amount of a mineral acid. 



Sym,posium. The action of allcali and acid on 

 cellulose, wood and waste cellulosie material with 

 special reference to the production of cheap cattle 

 foods. 



Joint symposium with division of organic chem- 

 istry on recent loorh on the constitution of starch 

 and cellulose. 



SECTION OF petroleum CHEMISTRY 



T. G. Delbridge, chairman 



W. A. Gruse, secretary 

 Volume changes in petroleum products: A. P. 

 BjERREQAARD. This paper will deal with some 

 peculiarities exhibited by petroleum oils under 



