August S, 1901] 



NA TURE 



567 



on the current, because the larger the current the more solid 

 carbon will the volatilisinp; surface cover, and the less, therefore, 

 will the specific resistances of the mist and vapour be lowered. 

 The way in which the core acts in each case is traced, and the 

 alterations in the specific resistances and cross-sections due to 

 the core are shown to bring about changes in the P. D. exactly 

 similar to those found by actual measurements of the P. D. 

 between the carbons. It is shown, for instance, how these 

 changes entirely account for the fact established by Prof. 

 Ayrton (Electrical Congress at Chicago, 1893) that, with a 

 constant length of arc, while the P. D. diminishes continuously 

 as the current increases, when both carbons are solid, it some- 

 times remains constant over a wide range of current, or even 

 increases again, after having diminished, when the positive 

 carbon is cored. 



The alterations in the value of 5V/5A introduced by the 

 cores are ne.\t discussed, and it is shown that the changes in 

 the resistance of the arcs that must follow the observed changes 

 in its cross-section, coupled with the alterations that must ensue 

 from the lowering of its specific resistance, would modify 5V,'5A 

 Justin the way that Messrs. Frith and Rodgers("The Resist- 

 ance of the Electric Arc," Phil. Mag. 1896, vol. xlii. p. 407) 

 found that it was modified by direct measurement. Thus all 

 the principal phenomena of the arc, with cored and with solid 

 carbons alike, may be attributable to such variations in the 

 specific resistances of the materials in the gap as it has been 

 shown must exist, together with the variations in the cross- 

 sections of the arc that have been observed to take place. 

 Hence it is superfluous to imagine either a large back E.M.F. 

 or a " negative resistance" 



" The Xature and Origin of the Poison Oi Lotus aralncus." 

 By Wyndham R. Dunstan, M.A., F. R.S., Director of the 

 Scientific and Technical Department of the Imperial Institute, 

 and T. A. Henry, B. Sc. , Salters' Company's Research Fellow 

 in the Laboratories of the Imperial Institute. 



The authors have already given a preliminary account 

 (Roy. Soc. Proc, vol. Ixvii. p. 224, 1900) of this investigation 

 and have shown that the poisonous property of this Egyptian 

 vetch is due to the prussic acid which is formed when the plant 

 is crushed with water, owing to the hydrolytic action of an 

 enzyme, lotase, on a glucoside, /olusiii, which is broken up into 

 hydrocyanic acid, dextrose and lotoflavin, a yellow colouring 

 matter. 



The authors have continued the investigation with the object 

 of ascertaining the properties and chemical constitution of 

 lotoflavin and of lotusin, and also of studying the properties of 

 lotase in relation to those of other hydrolytic enzymes. 



Lotusin. 



Lotusin can be separated Irom an alcoholic extract of the 

 plant by a tedious process giving a very small yield, about 0'025 

 per cent. 



Lotusin is a yellow crystalline glucoside, more soluble in 

 alcohol than in water. When heated it gradually decomposes 

 without exhibiting any fixed melting point. Combustions of 

 specially purified material gave numbers agreeing with those de- 

 duced from the formula CjsHjjNOig. 



In the preliminary notice the formula C2.,H]gNO)o was pro- 

 visionally assigned to lotusin on the assumption that one m.olecule 

 of dextrose is formed by its hydrolysis. The formula given above, 

 as the result of ultimate analysis, is confirmed by the observation 

 that two molecules of dextrose are produced by acid hydrolysis, 

 which is therefore represented by the equation — 



C,.H„NO„;-l-2HjO=:2C,H,.,Oe+ HCN -t- C,5H,„06. 

 Lotusin. Dextrose. Prussic Lotoflavin. " 



acid. 



When a solution 01 lotusin is warmed with dilute hydro- 

 chloric acid, hydrolysis readily occurs. The liquid acquires a 

 strong odour of hydrocyanic acid and a yellow crystalline pre- 

 cipitate of lotoflivin is thrown down, whilst the solution strongly 

 reduces Fehling's solution. Dilute sulphuric acid only very 

 slowly effects the hydrolysis of lotusin. 



When warmed with aqueous alkalis, lotusin is gradually de- 

 composed, ammonia being evolved and an acid formed to which 

 the name lotusinic acid has been given. 



Cj,H3,Oi6-f2H.,0 = C«H3As + NH3. 

 NO. 1658, VOL. 64] 



Lotusinic acid is a monobasic acid furnishing yellow crystal- 

 line salts. It is readily hydrolysed by dilute acids forming 

 latoflavin, dextrose and heplogluconic acid (dextrose-carboxylic 

 acid) : 



CaHjjOis -f 2H.O = C,,-,H,„0,; -I- C„Hi.,0,; -f C;H,iO,. 



Lotusinic Lotoflavin. Dextrose. HeptogIucon;c 



acid. acid. 



With the exception of amygdalin, lotusin is the only glucoside 

 definitely known which furnishes prussic acid as a decomposition 

 product. 



Lotoflavin. 



Lotoflavin is a yellow crystalline colouring matter readily dis- 

 solved by alcohol or by hot glacial acetic acid, and also by 

 aqueous alkalis forming bright yellow solutions. It is always 

 present to some extent in the plants, especially in old plants. 

 Ultimate analysis leads to the formula CiaHmOu. It is therefore 

 isomeric with luteolin, the yellow colouring matter of Reseda 

 luteola, and with fisetin, the yellow colouring from young fustic, 

 Rhus cotinus. Morin, from Mortis tinctoria, appears to be 

 hydroxylotoflavin. 



Lotoflavin does not form compounds with mineral acids. It 

 furnishes a tetracetyl derivative and two isomeric mutually con- 

 vertible trimethyl ethers which are capable of forming one and 

 the same acetyl-trimethyl-lotoflavin. By the action of fused 

 potash lotoflavin is converted into phloroglucin and 5-resorcylic 

 acid. 



Dextrose. 



The sugar resulting from hydrolysis has been found to corre- 

 spond in all properties with ordinary dextrose. 



Hydrocyanic acid. 



The amount of prussic acid given by plants at different stages 

 of growth has been ascertained. Mature plants bearing seed- 

 pods have furnished 0^345 per cent, of this acid, calculated on 

 the air-dried material which corresponds with 5 '23 per cent, of 

 lotusin. Younger plants bearing flower buds gave o'25 per 

 cent., whilst still smaller quantities were furnished by very 

 young plants and hardly any by quite old plants from which 

 the seeds had fallen. 



The formation of the poison, therefore, seems to reach its 

 maximum at about the seeding period, and after this period to 

 diminish rapidly. The Arabs are aware that the plant is safe to 

 use as a fodder when the seeds are quite ripe, but not before. 

 We have found that it is the lotusin which disappears during 

 the ripening of the seeds. Old plants contain sjme lotase and 

 lotoflavin, but little or no lotusin. 



Lotase. 



In its general properties lotase resembles other hydrolytic 

 enzymes, from which, however, it differs in several important 

 respects. It may be compared with emulsin, the enzyme of 

 bitter almonds. Emulsin, however, only attacks lotusin very 

 1 .slowly, whilst lotase has but a feeble action on amygdalin, the 

 glucoside of bitter almonds. Lotase is much more readily 

 injured and deprived of its hydrolytic power than emulsin. 

 On this account it is difficult to isolate in the solid state. Its 

 power is not only rapidly abolished by heat, but is also gradually 

 destroyed by contact with alcohol or glycerine. Besides lotase, 

 the plant contains an amylolytic and a proteolytic enzyme. 



Constitution of Lotoflavin and Lotusin. 

 Having regard to its reactions and especially to the produc- 

 tion, by the action of fused alkali, of /3-resorcylic acid and 

 phloroglucin, the authors conclude that lotoflavin should be 

 represented by the formula 



O OH 



oh/\/\,/'~\, 



OH CO 



OH, 



which is that o. a compound belonging to the same class, 

 of phenylated pheno-y-pyrones, as its isomerides luteolin and 

 fisetin. The peculiarity shown by lotoflavin of containing 

 four hydroxyl groups, but furnishing only a /'/methyl ether, i> 



