June 27, 1889] 



NATURE 



215 



.i2H2oO,o)4, allhough the amylon and amylin groups in each 



ossess very different optical and other properties, and the pro- 

 "ducts of the hydrolysis with dilute acid are very different. It 

 was found impossible to apply Raoult's method to starch-paste ; 

 solutions of soluble starch produced so slight a depression that no 

 reliable results can be obtained ; a number of fairly concordant 

 results, however, pointed to a molecular weight of 20,000 to 

 30,000. In order to ascertain whether the failure in this case 

 was due to a high molecular weight, and not to the fact that the 

 method was inapplicable to colloid substances an arabinic 

 acid, having a rotatory power [al, = + 6i°-i6, was ex- 

 amined ; thi^ gave a molecular weight of 717, thus rendering 

 it probable that the small influence exercised by soluble starch 

 was due to its high molecular weight. Indirect evidence was 

 then sought f r by an examination of the dextrins. The authors 

 have previously shown that when starch is broken down by dias- 

 tase, a re-ting stage in the reaction is reached when the amount 

 of dextrin produced corresponds with one-fifth by weight of the 

 starch taken, and that the molecule of this stable dextrin is one- 

 fifth of the size of the starch-molecule from which it has been 

 derived. Determinations of the molecular weight of this low 

 dextrin pointed to the formula aoCjoHooOu, and consequently 

 the formula of soluble starch would be 5(CioH.,oO]o)2o. ^"^ i^s 

 molecular weight 32,400. The endeavour was also made to 

 apply Raoult's method to the determination of the question 



i^hether the dextrins are a series of polymers or whether they are 

 simply metameric. For this purpose a number of the higher 

 dextrins were prepared from starch-transformations which had 

 been stopped at an early stage of hydrolysis. All the numbers 

 obtained show that the freezing method affords no evidence of 

 there being any difference in molecular weight between the high 

 and low dextrins, the numbers being, in fact, almost identical. 

 From a consideration of the results obtained with soluble starch, 

 and with dextrins of varying position in the series, the authors 

 conclude that the evidence points to the conclusion that the 

 dextrins are metameric and not polymeric compounds. They 

 therefore abandon their former working hypothesis of the 

 hydrolysis of starch, and now suppose the starch-molecule to 

 consist of four complex amylin-groups, arranged round a fifth 

 similar group constituting a molecular nucleus. When 

 hydrolysis takes place this complex is broken up, four amylin- 

 groups being liberated, which in turn are capable of under- 

 going compleie hydrolysis into malto dextrins, and ulti- 

 mately into maltose, whilst the fifth amylin-group which 

 constituted the nucleus of the original molecule resists the 

 action of hydrolyzing agents, and forms the stable dextrin 

 of the No. 8 equation of the authors' previous papers (Chem. 

 Soc. Trans., 1879, 634 ; 1885, 539). Each amylin group of the 

 five has the formula (CjaHjoOio)*^ corresponding with a mole- 

 cular weight of 6480 ; the molecule of soluble starch being 

 represented .by 5(Cj2H2„Oi„).2o, corresponding with a molecular 



weight of 32,400.- Researches on silicon compounds. Part V., 

 'by Prof. J. E. Reynolds. — The isomerism of the alkyl-deriva- 

 tives of mixed diazoamido compounds, by Prof. R. Meldola and 

 Mr. F. W. Streatfield. A critical study of the mixed diazo- 



amide, NOj^ NNjHc^^ ') , has served to confirm the 



NO., 



conclusion arrived at by the authors in their previous com- 

 munications (Chem. Soc. Trans., 1886,624; 1887, 102, 434; 

 1888, 664), that this compound and its alkyl-derivatives are per- 

 fectly definite, and that they cannot be formed by crystallizing 

 together mixtures of the corresponding symmetrical di-meta- 

 and di-para compoimds. The authors restate their original 

 proposition, viz. that every pair of amines, X.NH, and Y.NH,, 

 can give rise to three isomeric alkyl derivatives — (l) by the 

 action of diazotized X.NHj on Y.NHR' ; (2) by the action of 

 diazotized Y.NH., on X.NHR' ; (3) by the direct alkylation of 

 X.N3H.Y. In support of the general truth of this proposition 

 large number of triplets have been prepared, all of which con- 

 form to the rule laid down. The isomerism of these triplets is 

 shown not only by their melting points, but also by their pro- 

 ducts of decomposition by cold hydrochloric acid. These pro- 

 ducts have in all cases been examined quantitatively, and the 

 general results are shown by the equations — (i) X.N.^.NR'.Y -F 

 HCl = X.N„.C1 + Y.NHR'; (2) Y.N.NR' X -f HCl = 

 Y.N2.CI + X'.NHR'; (3) 2X.N.,R'.Y -f 2HCI = X.NjCl -f 

 Y.N2.CI -h X.NHR' -I- Y.NH.R'. From this it appears that 

 the alkyl-derivatives of the mixed diazo-amides always split up 

 like the unalkylated products into a mixture of the two diazo- 

 chlorides and two alkylamines. The most feasible interpretation 



of these facts is that the mixed diazoamides have double the 

 molecular weight usually assigned to them, and in support of ihis 

 conclusion the authors have discovered that the nixed alkyl- 

 diazoamides can be synthesized by simply boilirg the alcoholic 

 solutions of the other two iscmerides of the triplet, thus — 

 X.N2.NR'.Y -f Y.N2.NR'.X = (X.NsR'.Y).;.— The atomic 

 weight of zinc, by Dr. J. H. Gladstone and Mr. W. Hibbert. 

 The authors have observed that when amalgamated zinc is used 

 as anode in a zinc sulphate voltameter, the metal dissolved 

 appears to be free from impurity. Results obtained in different 

 voltameters with the same current showed a very clos^e agree- 

 ment, and the authors therefore determined the atomic weight of 

 zinc by applying Faraday's law of electrolysis. For this purpose 

 a series of copper, silver, and zinc voltameters were arranged in 

 a simple circuit, and the quantity of zinc dissolved was compared 

 with the weights of deposited silver and copper. The silver and 

 copper voltameters were arranged according to the conditions 

 s^hown to be best by Lord Rayleigh, Gray, and others. The zinc 

 voltameters were almost of the same form as the silver voltameter 

 of Lord Rayleigh. The anodewasashetet of amalgamatedzinc sup- 

 ported so as to rest horizontally on the surface of the zinc sulphate 

 solution, which had a specific gravity varying from I'iSto i"2r. 

 The mean ratio of the equivalents of silver and zinc is 3*^98^ 

 O'CoooS. Taking the atomic weight of silver as 107 -93, this 

 ratio gives 65 '44 as the atomic weight of zinc. If silver is taken 

 as io7"66, zinc = 65*29. The copper sulphate voltameter is not 

 so accurate as the silver one, owing to the solvent action of the 

 solution on the copper, and the ratio Zn : Cu given by the ex- 

 periments (i"0322) is probably rather too high. Adopting 

 Shaw's value for the atomic weight of copper (63 '33), this 

 ratio gives 65*37 as the atomic weight of zinc. — The amount of 

 nitric acid in the rain-water at Rothamsted, with notes on the 

 analysis of rain water, by Mr. R. Warrington. The rain of 

 twenty months, analyzed by the copper-zinc method, contained' 

 an average of 0'I38 of nitrogen as nitric acid per million of 

 water ; this is a little higher than that found by Way in 1855-56 — 

 namely, o*i2 per million — but is almost identical with that found 

 by Frankland as a mean of his analyses of Rothamsted rain in 

 1869-70. In a whole year, 1888-89, with a rainfall of 29-27 

 inches, the quantity of nitric nitrogen in the rain was 0*917 lb. 

 per acre, and the nitrogen as ammonia 2823 lbs., or a total of 

 374 lbs. — The product of the action of sulphur on resin, by Dr. 

 G. H. Morris. — The vapour-pressures and specific volumes of 

 similar compounds of elements in relation to the position of 

 those elements in the periodic system, by Prof. S. Young. 

 Determinations have been made of the vapour-pressures and 

 specific volumes of the four haloid derivatives of benzene and 

 also of benzene itself, within very wide limits of temperature ; 

 benzene, fluobenzene, and chlorobenzene having been heated to 

 their critical points, 288°-5 and 286°-55 and 36o°-8 respectively. 

 It is shown that if the four haloid derivatives are compared at 

 such temperatures that their vapour-pressures are equal, the 

 (absolute) temperatures and also the specific volumes bear a 

 constant ratio to each other, whatever the common pressure. 

 But on comparing benzene with one of its haloid derivatives it 

 is found that these simple relations do not hold. The ratios of 

 the absolute temperatures of benzene and fluobenzene corre- 

 sponding to equal pressures are very exactly expressed by the 

 equation R' = R -f d, where R' is the ratio at a pressure for 

 which the corresponding Centigrade temperature of fluobenzene 

 is /■, R = 0-0838, and c — 0C000313.— The vapour-pressures of 

 uqinoline, by the same. 



Anthropological Institute, May 28.— Mr. F. Galton, 

 F.R.S., Vice-President, in the chair.— Lieut. -General Pitt-Rivers. 

 F.R S., exhibited some crania found during some recent 

 excavations at Hunsbury Camp, and the Roman villa at 

 Llantwit. — The Rev. H. G. Tomkins read a paper on the 

 Hyksos, or Shepherd-Kings, of Egypt.— In a paper on the 

 proprietorship of trees on the ground of others, Mr. Hyde 

 Clarke (Vice-President), showed that this was the case in Asia 

 Minor, Melanesia, Borneo (honey-trees), India, Chofo Nagpore 

 (moura), and was supposed to be so in parts of China. He 

 proposed the ownership of trees as a more probable origin in 

 jurisprudence of the rights of property than ownership of land, 

 which has no primitive, value. 



Paris. 

 Academy of Sciences, June 17.— M. Des Cloizeaux, Presi- 

 dent, in the chair.— Nun:erical results obtained in the study 

 of the vitreous and metallic reflection of the visible and ultra- 



